Shaivism and The Tantric Traditions PDF

Śaivism and the Tantric Traditions
Dominic Goodall, Shaman Hatley, Harunaga Isaacson, and Srilata Raman - 978-90-04-43280-2
Downloaded from Brill.com08/09/2020 12:54:05AM
via free access
Gonda Indological Studies
Published Under the Auspices of the J. Gonda Foundation
Royal Netherlands Academy of Arts and Sciences
Edited by
Peter C. Bisschop (Leiden)
Editorial Board
Hans T. Bakker (Groningen)

Dominic D.S. Goodall (Paris/Pondicherry)
Hans Harder (Heidelberg)
Stephanie Jamison (Los Angeles)
Ellen M. Raven (Leiden)
Jonathan A. Silk (Leiden)
volume 22
The titles published in this series are listed at brill.com/gis
via free access
via free access
Alexis G.J.S. Sanderson
via free access
Śaivism and the Tantric Traditions
Essays in Honour of Alexis G.J.S. Sanderson
Edited by
Dominic Goodall
Shaman Hatley
Harunaga Isaacson
Srilata Raman
LEIDEN | BOSTON
via free access
This is an open access title distributed under the terms of the CC BY-NC 4.0 license,
which permits any non-commercial use, distribution, and reproduction in any medium,
provided the original author(s) and source are credited. Further information and the
complete license text can be found at https://creativecommons.org/licenses/by-nc/4.0/
The terms of the CC license apply only to the original material. The use of material from other sources
(indicated by a reference) such as diagrams, illustrations, photos and text samples may require further
permission from the respective copyright holder.
Cover illustration: Standing Shiva Mahadeva. Northern India, Kashmir, 8th century. Schist; overall: 53cm
(20 7/8in.). The Cleveland Museum of Art, Bequest of Mrs. Severance A. Millikin 1989.369
Library of Congress Cataloging-in-Publication Data
Names: Sanderson, Alexis, honouree. | Goodall, Dominic, editor. | Hatley, Shaman,

editor. | Isaacson, Harunaga, 1965- editor. | Raman, Srilata, editor.
Title: Śaivism and the tantric traditions : essays in honour of Alexis G.J.S.
Sanderson / edited by Dominic Goodall, Shaman Hatley, Harunaga Isaacson,
Srilata Raman.
Description: Leiden ; Boston : Brill, [2020] | Series: Gonda indological studies, 1382-
3442 ; volume 22 | Includes "Bibliography of the published works of Alexis G.J.S.
Sanderson (1983-2019)." | Includes bibliographical references and index.
Identifiers: LCCN 2020022036 (print) | LCCN 2020022037 (ebook) |
ISBN 9789004432666 (hardback) | ISBN 9789004432802 (ebook)
Subjects: LCSH: Śaivism–History. | Kashmir Śaivism–History. | Tantrism–History.
Classification: LCC BL1280.53 .S25 2020 (print) | LCC BL1280.53 (ebook) |
DDC 294.5/513–dc23
LC record available at https://lccn.loc.gov/2020022036
LC ebook record available at https://lccn.loc.gov/2020022037
Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill‑typeface.
ISSN 1382-3442
ISBN 978-90-04-43266-6 (hardback)
ISBN 978-90-04-43280-2 (e-book)
Copyright 2020 by the Authors. Published by Koninklijke Brill NV, Leiden, The Netherlands.
Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi,
Brill Sense, Hotei Publishing, mentis Verlag, Verlag Ferdinand Schöningh and Wilhelm Fink Verlag.
Koninklijke Brill NV reserves the right to protect this publication against unauthorized use.
This book is printed on acid-free paper and produced in a sustainable manner.
via free access
Contents
Preface xi
List of Figures and Tables xii
Notes on Contributors xiv
A Note on Alexis Sanderson and Indology xxv
Dominic Goodall and Harunaga Isaacson
Bibliography of the Published Works of Alexis G.J.S. Sanderson
(1983–2019) xxxi
Introduction 1
Dominic Goodall, Shaman Hatley, Harunaga Isaacson
and Srilata Raman
Part 1
Early Śaivism
1 From Mantramārga Back to Atimārga: Atimārga as a Self-referential

Term 15
Peter C. Bisschop
2 Why Are the Skull-Bearers (Kāpālikas) Called Soma? 33

Judit Törzsök
3 Dressing for Power: On vrata, caryā, and vidyāvrata in the Early

Mantramārga, and on the Structure of the Guhyasūtra of the
Niśvāsatattvasaṃhitā 47
Dominic Goodall
Part 2
Exegetical and Philosophical Traditions
4 Further Thoughts on Rāmakaṇṭha’s Relationship to Earlier Positions in

the Buddhist-Brāhmaṇical Ātman Debate 87
Alex Watson
via free access
viii contents
5 Some Hitherto Unknown Fragments of Utpaladeva’s Vivṛti (II): Against

the Existence of External Objects 106
Isabelle Ratié
6 Alchemical Metaphors for Spiritual Transformation in Abhinavagupta’s

Īśvarapratyabhijñāvimarśinī and Īśvarapratyabhijñāvivṛtivimarśinī 144
Christopher D. Wallis
7 On Vāgīśvarakīrti’s Influence in Kashmir and among the Khmer 170

Péter-Dániel Szántó
8 Reflections on the King of Ascetics (Yatirāja): Rāmānuja in the

Devotional Poetry of Vedānta Deśika 194
Srilata Raman
Part 3
Religion, the State, and Social History
9 Not to Worry, Vasiṣṭha Will Sort It Out: The Role of the Purohita in the
Raghuvaṃśa 217
Csaba Dezső
10 Buddhism, Kingship and the Protection of the State: The

Suvarṇaprabhāsottamasūtra and Dhāraṇī Literature 234
Gergely Hidas
11 Adapting Śaiva Tantric Initiation for Exoteric Circles: Lokadharmiṇī

Dīkṣā and Its History in Early Medieval Sources 249
Nina Mirnig
12 Innovation and Social Change in the Vale of Kashmir, circa

900–1250C.E. 283
John Nemec
13 Toward a History of the Navarātra, the Autumnal Festival of the

Goddess 321
Bihani Sarkar
via free access
contents ix
Part 4
Mantra, Ritual, and Yoga
14 Śārikā’s Mantra 349

Jürgen Hanneder
15 The Kāmasiddhistuti of King Vatsarāja 364

Diwakar Acharya
16 The Lotus Garland (padmamālā) and Cord of Power (śaktitantu): The

Brahmayāmala’s Integration of Inner and Outer Ritual 387
Shaman Hatley
17 The Amṛtasiddhi: Haṭhayoga’s Tantric Buddhist Source Text 409

James Mallinson
18 A Sexual Ritual with Māyā in Matsyendrasaṃhitā 40 426

Csaba Kiss
19 Haṭhayoga’s Floruit on the Eve of Colonialism 451

Jason Birch
Part 5
Art and Architecture
20 The Early Śaiva Maṭha: Form and Function 483

Libbie Mills
21 The Kriyāsaṃgrahapañjikā of Kuladatta and Its Parallels in the Śaiva

Pratiṣṭhātantras 507
Ryugen Tanemura
22 Mañjuśrī as Ādibuddha: The Identity of an Eight-Armed Form of

Mañjuśrī Found in Early Western Himalayan Buddhist Art in the Light of
Three Nāmasaṃgīti-Related Texts 539
Anthony Tribe
via free access
x contents
23 Life and Afterlife of Sādṛśya: Revisiting the Citrasūtra through the

Nationalism-Naturalism Debate in Indian Art History 569
Parul Dave-Mukherji
Index 589
via free access
Preface
This volume results from a symposium held at the University of Toronto in

honour of Alexis G.J.S. Sanderson. The symposium was convened in March
2015 in anticipation of his retirement as the Spalding Professor of Eastern Reli-
gions and Ethics at All Souls College, Oxford University. The event was con-
ceived by Srilata Raman, who worked tirelessly and resourcefully to make it
a success. In this she was aided by Shaman Hatley, co-convener of the sym-
posium, and a number of graduate students, especially Kalpesh Bhatt, Tamara
Cohen, Larissa Fardelos, Nika Kuchuk, and Eric Steinschneider, to whom we
offer our sincere thanks. It was immensely satisfying to have so many of Pro-
fessor Sanderson’s former doctoral students assemble from across the world
for the occasion, students whose graduate studies at Oxford spanned more
than three decades of Alexis Sanderson’s teaching career. The volume is based
mainly on papers presented in the symposium, with additional contributions
by several of his former pupils who had not been able to present their work at
that time (Parul Dave-Mukherji, Csaba Dezső, Csaba Kiss, Ryugen Tanemura,
and Anthony Tribe), as well as by Diwakar Acharya, his successor to the Spald-
ing Professorship. We would also like to extend our thanks and recognition to
those who enriched the symposium with excellent papers, but who for various
reasons could not include these in the present volume: Hans Bakker, Gudrun
Bühnemann, Shingo Einoo, Alexander von Rospatt, and Somadeva Vasudeva.
We would like to acknowledge the sponsors who made the symposium pos-
sible: All Souls College, Oxford University; the Centre for South Asian Studies,
University of Toronto; the Department of Religion of the University of Toronto,
and its Chair, John Kloppenborg; Brill Publishers; and Srilata Raman, who con-
tributed quietly and generously from her own research funds. We would also
like to thank University College of the University of Toronto, and John Mar-
shall, its Vice Principal, for making available the lovely Croft Chapter House, in
which the symposium was held.
The contributors to the volume and the publisher have endured a long wait
for this volume to come to fruition, and we would like to thank them for their
patience and cheerful support. Special thanks are due also to Anusha Sudin-
dra Rao, who proofread the volume carefully on short notice, and to Liwen Liu,
who prepared the index.
via free access
List of Figures and Tables
Figures
4.1 Self as essence 88

4.2 Self as substance 88
4.3 Self as agent 89
4.4 Rāmakaṇṭha’s view in the middle ground 91
4.5 Rāmakaṇṭha’s view as equivalent to Nyāya 94
4.6 Rāmakaṇṭha’s view as more extreme than Nyāya 96
4.7 Nyāya as equivalent to Rāmakaṇṭha’s view 99
5.1 Manuscript S3 of Abhinavagupta’s Īśvarapratyabhijñāvimarśinī (f. 43A), with
fragments 4 and 5 of Utpaladeva’s Vivṛti on vv. 1.5.8–9 in the top and right
margins (photograph: National Archives of India). 108
14.1 Śārikā stone. Photograph by Walter Slaje. 350
16.1 Garland of the Devīs and Dūtīs (padmamālā I) 396
16.2 Garlands of the Mātṛs and Yoginīs (padmamālās II–III) 397
16.3 The pure body of power (avadhūtatanu) 398
20.1 The 9×9 plan, Bṛhatkālottara, chapter 112 (the vāstuyāgapaṭala) 486
20.2 Deity, nakṣatra, and consequence of door position. Bṛhatkālottara,
prāsādalakṣaṇapaṭala 238c–243b. 487
20.3 The nandyāvarta/nandikāvarta set of nine chambers 490
22.1 Eight-armed Mañjuśrī, uppermost level, Sumtsek, Alchi. Photograph by Jaroslav
Poncar (JP93 26.4.05 WHAV) 540
22.2 Maṇḍala of eight-armed Mañjuśrī, uppermost level, Sumtsek, Alchi.
Photograph by Jaroslav Poncar (JP83 26.4.01 WHAV) 541
22.3 Eight-armed Mañjuśrī, two-armed Maitreya Chapel, Mangyu. Photograph by
Christiane Papa-Kalantari (CP02 44,39 WHAV) 542
22.4 Standing eight-armed Mañjuśrī, Village Stūpa, Mangyu. Photograph by
Christiane Papa-Kalantari (CP02 41,14 WHAV) 543
22.5 Maṇḍala of eight-armed Mañjuśrī, Serkhang, Lalung. Photograph by Jaroslav
Poncar (JP92 500,15 WHAV) 544
via free access
list of figures and tables xiii
Tables
6.1 Two types of Turyātīta in the ĪPVV 164

13.1 Developmental phases of the Navarātra 326
14.1 The bījas as defined in the Śārikāstotra 360
16.1 Deities of the Nine-Syllable Vidyā 393
19.1 A comparison of the number of verses in early and late texts on
Haṭhayoga 457
19.2 The Proliferation of post-fifteenth Haṭhayogic techniques 460
via free access
Notes on Contributors
Diwakar Acharya
is the Spalding Professor of Eastern Religions and Ethics at Oxford University,
a fellow of All Souls College. Before succeeding Professor Sanderson at Oxford,
he was a visiting lecturer and then associate professor of Indological Studies at
Kyoto University (2006–2016), before which he held positions at Hamburg Uni-
versity and Nepal Sanskrit University. His research covers a wide range of topics
in Indian religious and philosophical traditions, Upaniṣadic studies, epigraphy,
the early history of Nepal, ritual, and Sanskrit literature. Recent publications
include Early Tantric Vaiṣṇavism. Three Newly Discovered works of the Pañcar-
ātra: The Svāyambhuvapañcarātra, Devāmṛtapañcarātra and Aṣṭādaśavidhāna
(Pondicherry, 2015), and a number of articles, such as “ ‘This world, in the begin-
ning, was phenomenally non-existent’: Āruṇi’s Discourse on Cosmogony in the
Chāndogya Upaniṣad” ( Journal of Indian Philosophy 44.5, 2016). Acharya now
also serves as editor-in-chief of the Journal of Indian Philosophy.
Jason Birch
(SOAS University of London) completed a first class honours degree in San-
skrit and Hindi at the University of Sydney under Dr Peter Oldmeadow, and
was then awarded a Clarendon scholarship to undertake a DPhil in Orien-
tal Studies at Balliol College, University of Oxford, under the supervision of
Professor Sanderson. His dissertation (submitted 2013) focused on the earliest
known Rājayoga text called the Amanaska and included a critical edition and
annotated translation of this Sanskrit work, along with a monographic intro-
duction which examines the influence of earlier Śaiva tantric traditions on the
Amanaska, as well as the significance of the Amanaska in more recent yoga tra-
ditions. Birch is currently a post-doctoral research fellow at SOAS University of
London on the Haṭha Yoga Project, which has been funded for five years by the
ERC. His area of research is the history of physical yoga on the eve of colonial-
ism. He is editing and translating six texts on Haṭha and Rājayoga, and supervis-
ing the work of two research assistants at the École française d’ Extrême-Orient,
Pondicherry. Birch has taught courses at SOAS and Loyola Marymount Univer-
sity, and given seminars on the history of yoga at the Università Ca’ Foscari in
Venice, Italy and Won Kwang University in Iksan, South Korea. He also collab-
orates with Jacqueline Hargreaves on TheLuminescent.
via free access
notes on contributors xv
Peter Bisschop
is Professor of Sanskrit and Ancient Cultures of South Asia at Leiden Univer-
sity. In 2004, after finishing his PhD at the University of Groningen under Hans
Bakker, co-supervised by Harunaga Isaacson, he was offered the opportunity
of spending a year in Oxford as a Junior Research Fellow at Wolfson College.
During his spell at Oxford, he met on a weekly basis with Alexis Sanderson in
All Souls College to discuss his ongoing work on the Pāśupata tradition. In par-
ticular he was able to read with him a draft of his critical edition of chapter 1 of
Kauṇḍinya’s Pañcārthabhāṣya, including a previously lost passage of Kauṇḍi-
nya’s commentary on Pāśupatasūtra 1.37–39 on the basis of a newly identified
manuscript from Benares. An edition and translation of this passage was pub-
lished the year after in the Journal of Indian Philosophy, 33. In 2005 he was
appointed Lecturer in Sanskrit Studies at the University of Edinburgh, where
he remained until his move to Leiden in 2010 to take up the chair of Sanskrit. He
has published extensively on different aspects of early Śaivism, in particular the
Pāśupatas and associated lay traditions, from his monograph Early Śaivism and
the Skandapurāṇa: Sects and Centres (2004) to his contributions to the ongoing
critical edition of the Skandapurāṇa, as well as a new book entitled Univer-
sal Śaivism: The Appeasement of All Gods and Powers in the Śāntyadhyāya of the
Śivadharmaśāstra (Brill, 2018). He is also the editor-in-chief, with Jonathan Silk,
of the Indo-Iranian Journal, and general editor of the Gonda Indological Stud-
ies.
Parul Dave-Mukherji
is professor of Visual Studies and Art History at the School of Arts and Aes-
thetics, Jawaharlal Nehru University, New Delhi, India. She holds a DPhil from
Oxford University, where she worked with Alexis Sanderson on a critical edition
of the Citrasūtra of the Viṣṇudharmottara Purāṇa from 1988–1991. Introduced
to the rigour and intricacy of critically editing a text based on manuscripts, her
work vindicated Sanderson’s view that the earlier work by Priyabala Shah was
far from being a critical edition. In 2001, her critical edition, The Citrasūtra of
the Viṣṇudharmottara Purāṇa, was published by the Indira Gandhi National
Centre for the Arts and Motilal Banarsidass, New Delhi. The śilpaśāstras and
pre-modern Indian aesthetics remain important areas of research, along with
modern/contemporary Indian/Asian Art. Dave-Mukherji has held fellowships
at the Clark Art Institute, Williamstown, USA; South Asia Institute, Heidelberg,
Germany; British Academy fellowship, Goldsmiths’ College, London; and Kun-
sthistorisches Institut, Florence. Her recent publications include “Whither Art
History in a Globalizing World” (The Art Bulletin 96.2, 2014); Arts and Aesthetics
in a Globalizing World, co-edited with Ramindar Kaur (Bloomsbury, 2014); and
via free access
xvi notes on contributors
“Who is Afraid of Mimesis? Contesting the Common Sense of Indian Aesthetics

through the Theory of ‘Mimesis’ or Anukaraṇa Vāda” (in Indian Aesthetics and
the Philosophy of Art, ed. Arindam Chakrabarti; Bloomsbury, 2016). Currently,
she is co-editing, with Partha Mitter and Rakhee Balaram, a comprehensive his-
tory of modern and contemporary Indian art in a volume entitled 20th Century
Indian Art.
Csaba Dezső
studied Classical Philology, History and Indology at Eötvös Loránd University,
Budapest. After finishing his masters degrees (Latin Language and Literature
and Indology), he went to Oxford in 1998 to study for a PhD under the supervi-
sion of Professor Alexis Sanderson. He submitted his doctoral thesis in 2004, a
critical edition and annotated translation of Bhaṭṭa Jayanta’s Āgamaḍambara, a
satirical play about religious sects and their relations with the court in Kashmir
around 900 CE. He then returned to Budapest and has been teaching Sanskrit
since then at the Department of Indo-European Linguistics, Eötvös Loránd
University. He has published, among others, first editions of fragments of San-
skrit plays based on codices unici, as well as a new critical edition and English
verse translation of Dāmodaragupta’s Kuṭṭanīmata, “The Bawd’s Counsel,” in
collaboration with Dominic Goodall. Recently he has been working on the crit-
ical edition of Vallabhadeva’s commentary on the Raghuvaṃśa together with
Dominic Goodall, Harunaga Isaacson and Csaba Kiss.
Dominic Goodall
began studies in Classics and German at Pembroke College, Oxford, before fin-
ishing a BA in Sanskrit with Pali. After two years spent in Hamburg to learn
medieval Tamil with S.A. Srinivasan, he returned to Oxford, to Wolfson Col-
lege, where, under Alexis Sanderson’s guidance, he produced a critical edition
of the opening chapters of Bhaṭṭa Rāmakaṇṭha’s tenth-century commentary
on the Kiraṇatantra, which he submitted as a doctoral thesis in 1995 and pub-
lished from Pondicherry in 1998. He was attached to the French Institute of
Pondicherry as a junior researcher in 1996–1997 before returning to Oxford as
Wolfson Junior Research Fellow of Indology from 1998 to 2000. In 2000, he
became a member of the École française d’Extrême-Orient and was appointed
Head of its Pondicherry Centre in 2002. For his habilitation, he submitted
to Hamburg a first edition of the Parākhyatantra, which was later published
from Pondicherry in 2004. Posted in Paris from 2011 to 2015, he gave lectures
at the École pratique des hautes études (Religious Sciences Section), at the
invitation of Gerdi Gerschheimer, on Cambodian inscriptions in Sanskrit and
on Śaivism. Now back in Pondicherry, he continues to pursue his interests in
via free access
notes on contributors xvii
Sanskrit poetry (both Indian and Cambodian) and in the history of the Śaiva-
siddhānta. With Marion Rastelli, he co-edits the Viennese dictionary of tantric
terminology, the Tāntrikābhidhānakośa. In May 2016, he was elected membre
correspondent étranger de l’Académie des Inscriptions et Belles-Lettres.
Jürgen Hanneder
studied Indology, Tibetology and Comparative Religion in Munich, Bochum,
and Bonn, where he took his MA. His interest in the Śaiva traditions of Kashmir
led him to Oxford, where he studied under the supervision of Alexis Sanderson.
After completing his PhD in Indology in Marburg, and working as a research
assistant in Bonn, he joined the Mokṣopaya Research Group initiated by Wal-
ter Slaje in Halle. After some terms as substitute professor in Freiburg he fol-
lowed his former teacher Michael Hahn to the chair of Indology in Marburg in
2007. The main areas of his research interests are within classical and modern
Sanskrit literature, i.e. poetry, religious and philosophical literature, including
Indo-Tibetan studies and occasional excursions into neighboring fields, as for
instance the names of lotuses (“The Blue Lotus. Oriental Research between
Philology, Botany and Poetics?,” ZDMG 152.2 [2002]: 295–308), and a study of
Indian crucible steel, which has played an important role for the modern steel
industry (Der “Schwertgleiche Raum”. Zur Kulturgeschichte des indischen Stahls;
Stuttgart: Franz Steiner Verlag, 2005). Current larger projects include August
Wilhelm Schlegel als Indologe and The Minor Works of Sahib Kaul.
Shaman Hatley
completed an interdisciplinary liberal arts degree at Goddard College (1998),
and then studied Indology and Religious Studies at the University of Pennsyl-
vania. He completed his doctorate on the Brahmayāmala and Śaiva yoginī cults
in 2007 under the direction of Harunaga Isaacson. In 2003 and 2006, he had
the opportunity to read the Brahmayāmala with Professor Sanderson while a
visiting student at Oxford, a formative scholarly experience that was crucial to
his doctoral project. He taught at Concordia University, Montréal, from 2007 to
2015, and is now an associate professor of Asian Studies and Religious Studies
at the University of Massachusetts Boston. His research mainly concerns early
Tantric Śaivism, goddess cults, and yoga, and he regularly contributes to the
Tāntrikābhidhānakośa. Recent publications include The Brahmayāmalatantra
or Picumata, Volume I: Chapters 1–2, 39–40, & 83. Revelation, Ritual, and Material
Culture in an Early Śaiva Tantra (Pondicherry, 2018).
via free access
xviii notes on contributors
Gergely Hidas
started a DPhil in Oriental Studies under the supervision of Alexis Sander-
son at Balliol College, University of Oxford, in 2002, after earlier studies in
Budapest. The revised version of his doctoral thesis, on a principal scripture of
Buddhist dhāraṇī literature, Mahāpratisarā-Mahāvidyārājñī, The Great Amulet,
Great Queen of Spells, was published in New Delhi in 2012. Between 2007 and
2012 he held research and teaching positions at Eötvös Loránd University,
Budapest, and thereafter contributed to the Cambridge Sanskrit Manuscripts
Project, an Oxford medieval sources project, and the Vienna Viscom project,
and was also was awarded a research grant by the Hungarian Academy of Sci-
ences. In 2013–2014 he was appointed as Khyentse Fellow at the Centre for Bud-
dhist Studies, Eötvös Loránd University. Since 2014, he has had a postdoctoral
affiliation with the British Museum in the ERC Synergy project “Asia Beyond
Boundaries: Religion, Region, Language and the State,” where, among other
forthcoming publications, he is finalizing a book manuscript entitled Vajra-
tuṇḍasamayakalparāja, a Buddhist Ritual Manual on Agriculture.
Harunaga Isaacson
was born in Kuma, Japan, in 1965; he studied philosophy and Indology at the
University of Groningen, and was awarded a PhD in Sanskrit in 1995 by the
University of Leiden for a thesis on the early Vaiśeṣika school of philosophy
(1995). From 1995–2000 he was a Post-doctoral Research Fellow in Sanskrit at
the Oriental Institute, Oxford University. After holding teaching positions at
Hamburg University (2000–2002) and the University of Pennsylvania (2002–
2006), he was appointed Professor of Classical Indology at Hamburg University
in 2006. He has been a Visiting Fellow at All Souls College, Oxford, and is a
member of the Akademie der Wissenschaften in Hamburg. His main areas of
study are: tantric traditions in pre-13th-century South Asia, especially Vajrayāna
Buddhism; classical Sanskrit belles-lettres (kāvya); classical Indian philosophy;
and Purāṇic literature.
Csaba Kiss
began his doctoral studies at the University of Oxford under the supervision
of Professor Alexis Sanderson in 2003. After defending his thesis, a critical
edition of selected chapters of the Matsyendrasaṃhitā, he worked at ELTE Uni-
versity, Budapest, as a research assistant. From 2008 to 2010, he was member
of the Early Tantra Project, conducting research on the Brahmayāmala, since
published as The Brahmayāmalatantra or Picumata, Volume II. The Religious
Observances and Sexual Rituals of the Tāntric Practitioner: Chapters 3, 21, and
45 (Pondicherry, 2015). He has since been taking part in a number of research
via free access
notes on contributors xix
projects: writing entries for the Tāntrikābhidhānakośa, contributing to the edi-

tion of Vallabhadeva’s Raghupañcikā led by Dominic Goodall and Harunaga
Isaacson, and contributing to research on jātis, as well as to digitization of
Gupta-era inscriptions within the ERC Project “Beyond Boundaries.”
James Mallinson
met Alexis Sanderson at an open day for prospective undergraduates at the
Oriental Institute in Oxford in 1987. As a result of this meeting he changed his
choice of course to Sanskrit. As an undergraduate he had Professor Sanderson
as his essay tutor. Mallinson was not always the most diligent of students, so
was delighted when Professor Sanderson agreed to supervise his doctoral stud-
ies at Oxford, which he started in 1995. His doctoral thesis was a critical edition
and annotated translation of the Khecarīvidyā, an early text on haṭhayoga. After
receiving his doctorate, Mallinson worked as a principal translator for the Clay
Sanskrit Library for six years. In 2013 he became Lecturer in Sanskrit and Clas-
sical Indian Studies at SOAS, University of London. Since his doctorate he had
continued to work on unpublished materials on yoga, often reading his working
editions with Professor Sanderson, and in 2015 Mallinson was awarded an ERC
Consolidator Grant for a five-year project on the history of haṭhayoga. Among
the members of the project team is Jason Birch, another former doctoral stu-
dent of Professor Sanderson, and the team have continued to work closely with
him. Among Mallinson’s publications is Roots of Yoga (Penguin Classics, 2017),
an anthology of translations of texts on yoga, including several by Professor
Sanderson and his former students, together with a detailed analysis of the his-
tory of yoga and its practices.
Libbie Mills
teaches Sanskrit and Pali at the University of Toronto. She completed her doc-
torate under Alexis Sanderson’s supervision at Oxford University in 2011. Her
principal research interest is in South Asian architectural theory and practice.
Her work on the architectural instruction given in the early Śaiva installation
manuals (pratiṣṭhātantras) of North India features the first exposition of these
texts’ contents. By examining extant buildings in light of texts, her research
introduces new tools for dating, which are valuable for these as well as other
sources. Her textual study has since expanded to cover domestic and temple
building practices up to the modern period, in both South Asia and its dias-
pora. She is currently engaged in two research projects: “The Nāgara Tradition
of Temple Architecture: Continuity, Transformation, Renewal,” funded by the
Leverhulme Trust, and “Tamil Temple Towns: Conservation and Contestation,”
funded by the Arts and Humanities Research Council.
via free access
xx notes on contributors
Nina Mirnig
undertook her studies at the Oriental Institute at Oxford University. She first
met Alexis Sanderson during the second year of her undergraduate course
when joining an MPhil reading class in 2002. His inspiring teaching and the
insights he offered into the fascinating world of Śaiva Tantrism prompted her
to do her BA special paper in this field. She continued her post-graduate studies
under his supervision on the topic of the socio-religious history and develop-
ment of Śaiva tantric cremation rites (antyeṣṭi) and post-mortuary ancestor
worship (śrāddha). Upon completing her DPhil in 2010, she continued to work
on early Śaiva religious history in an NWO-funded project on the composition
and spread of the Skandapurāṇa under the direction of Hans Bakker. After a
Jan Gonda Fellowship at the International Institute for Asian Studies in Lei-
den and briefly joining an AHRC project on the manuscript collections at the
University Library in Cambridge, under the direction of Vincenzo Vergiani, she
moved to her current position at the Institute for the Cultural and Intellectual
History of Asia at the Austrian Academy of Sciences. In addition to the early
Śaiva tantric traditions, her research now concerns early Śaiva lay traditions, a
critical edition of Śivadharmaśāstra 1–5 and 9, and the cultural history of early
medieval Nepal, with special focus on the Sanskrit Licchavi inscriptions.
John Nemec
is Associate Professor of Indian Religions and South Asian Studies at the Uni-
versity of Virginia. He earned his PhD from the University of Pennsylvania
(2005), an MPhil degree in Indian Religions from Oxford University, an MA in
Religious Studies from the University of California at Santa Barbara, and a BA
in Religion and Classics from the University of Rochester. At Oxford, he worked
extensively with Alexis Sanderson in the course of completing his MPhil degree
and thesis, most notably in the form of weekly private tutorials on Śaiva litera-
ture in Sanskrit, which Alexis generously offered every academic term for two
full years. Nemec again profited from Alexis’s boundless generosity as a visiting
doctoral student, when he once more read Sanskrit with him at All Souls Col-
lege in the Trinity Term of 2002. His publications include The Ubiquitous Śiva:
Somānanda’s Śivadṛṣṭi and His Tantric Interlocutors (Oxford University Press,
2011).
Srilata Raman
is Associate Professor of Hinduism at the University of Toronto and works on
medieval South Asian/South Indian religion, devotionalism (bhakti), histori-
ography and hagiography, religious movements in early colonial India from
the South, as well as modern Tamil literature. Her areas of interest are Tamil
via free access
notes on contributors xxi
and Sanskrit intellectual formations from late medieval to early colonial peri-
ods, including the emergence of nineteenth-century socio-religious reform and
colonial sainthood. Her publications include Self-Surrender (Prapatti) to God in
Śrīvaiṣṇavism. Tamil Cats and Sanskrit Monkeys (Routledge, 2007).
Isabelle Ratié
is Professor of Sanskrit Language and Literatures at the Sorbonne Nouvelle Uni-
versity (Paris). She defended her doctoral thesis in 2009 at the École pratique
des hautes études after reading about two thirds of Abhinavagupta’s Īśvara-
pratyabhijñāvimarśīnī in Oxford under Alexis Sanderson’s guidance (2005–
2006). She has published several monographs on Śaiva and Buddhist philoso-
phies (Le Soi et l’Autre. Identité, différence et altérité dans la philosophie de la
Pratyabhijñā, Leiden: Brill, 2011, Weller Prize 2012; Une Critique bouddhique du
Soi selon la Mīmāṃsā, Vienna: Austrian Academy of Sciences, 2014; and with
Vincent Eltschinger, Self, No-Self, and Salvation. Dharmakīrti’s Critique of the
Notions of Self and Person, Vienna: Austrian Academy of Sciences, 2013). She
has also coedited with Eli Franco the collective volume Around Abhinavagupta.
Aspects of the Intellectual History of Kashmir from the Ninth to the Eleventh Cen-
tury (Berlin: LIT Verlag, 2016). She is currently editing and translating recently
discovered fragments of Utpaladeva’s lost Vivṛti on the Īśvarapratyabhijñā trea-
tise, and she is working with Vincent Eltschinger, Michael Torsten Much and
John Taber on a translation of Dharmakīrti’s Pramāṇavārttika 1 (the section on
apoha).
Bihani Sarkar
completed her BA (First Class) in English from St. Hilda’s College, Oxford, then
undertook an MPhil in Classical Indian Religions and a DPhil in Sanskrit (Ori-
ental Studies) from Wolfson College, Oxford, both under the supervision of
Alexis Sanderson. Her doctoral thesis, now the book Heroic Shaktism (Oxford
University Press, 2017), is a history of the rise and spread of the cult of Durgā
between the 3rd and the 11th centuries CE and its influence on heroic ideol-
ogy and the rising Indian kingdom. She was a Nachwuchsinitiative Postdoctoral
Fellow in Hamburg University from 2012–2014, and a British Academy Postdoc-
toral Fellow at Oxford University from 2014–2017. She is presently a Teaching
Fellow in South Asian Religions at Leeds University and Associate Member of
Christ Church College, Oxford. She is working on her second book, on the sub-
ject of grief and lamentation in kāvya.
via free access
xxii notes on contributors
began his studies at ELTE Budapest, where he received diplomas in Tibetology
in 2004 and in Indology in 2006. He first met Alexis Sanderson at his depart-
ment in 2002, where he held a week-long intensive reading of Abhinavagupta.
These sessions were so inspirational that Szántó decided to apply to Oxford,
where he was successful in joining in 2006 with the help of Csaba Dezső,
thus becoming both the śiṣya and praśiṣya of Sanderson. His doctoral thesis,
defended in 2012, was on the Catuṣpīṭha, an early Buddhist Yoginītantra. After
being a Junior Research Fellow at Merton College, Oxford, and then having a
ten-month stipend in Hamburg, Szántó returned to Oxford as a Post-doctoral
Research Fellow at All Souls College, thus having the enormous pleasure and
privilege of spending many splendid dinners with Sanderson. He is currently a
postdoctoral fellow at Leiden University in the Open Philology project. Most of
Szántó’s publications deal with the literature of esoteric or tantric Buddhism
in India, but he has also authored papers on poetics, epigraphy, and material
culture. His latest publication, “Mahāsukhavajra’s Padmāvatī Commentary on
the Sixth Chapter of the Caṇḍamahāroṣaṇatantra: The Sexual Practices of a
Tantric Buddhist Yogī and His Consort” ( Journal of Indian Philosophy 46), was
co-authored with Samuel Grimes. Szántó is currently working on the editio
princeps of the Sarvabuddhasamāyogaḍākinījālaśaṃvara, a project featuring
much input and inspiration from the man we celebrate in this volume.
Ryugen Tanemura
is an associate professor of Buddhist studies and classical Indology at Taisho
University, Tokyo, Japan. After having been educated at the University of Tokyo,
he went to Oxford in 1997, where he did his doctoral research on tantric Bud-
dhism under the supervision of Professor Sanderson for five years. He took
his DPhil in classical Indology at the University of Oxford (2003). After post-
doctoral research at the University of Tokyo and some other institutions, he
began his current position in 2014. His main research field is Indian tantric
Buddhism, and he has authored many works in this area, including Kuladatta’s
Kriyāsaṃgrahapañjikā: A Critical Edition and Annotated Translation of Selected
Sections (Groningen: Egbert Forsten, 2004) and Kriyāsaṃgraha of Kuladatta,
Chapter 7 (Tokyo: Sankibo, 1997). His recent publications include critical edi-
tions of Śūnyasamādhivajra’s Mṛtasugatiniyojana, a manual of the Indian Bud-
dhist tantric funeral, and chapters 1 (part), 13 (part), 19, and 22 (pratiṣṭhā sec-
tion) of the Padminī, a commentary on the Saṃvarodayatantra by Ratnarakṣita
(chapters 1 and 13, in collaboration with Kazuo Kano and Kenichi Kuranishi).
via free access
notes on contributors xxiii
Judit Törzsök
completed an MA in Indic Studies at ELTE University, Budapest, and then con-
tinued her studies at the University of Oxford in 1993, where she was funded by
the George Soros Foundation to do research on Abhinavagupta under Professor
Sanderson’s supervision. In 1994, having received the Domus Senior Scholar-
ship at Merton College, she started working on the Siddhayogeśvarīmata for her
DPhil, supervised by Sanderson. The years spent in Oxford under his guidance
determined the course of her research, which has focused on the early history
of yoginī cults ever since. After postdoctoral research fellowships at Emmanuel
College, Cambridge, and at the University of Groningen (supervised by Profes-
sor Hans Bakker), she was elected Associate Professor (maître de conférences)
in 2001 at the University Charles-de-Gaulle Lille III in France, and professor
(directeur d’études) at the École pratique des hautes études (EPHE) in 2018.
She defended her Habilitation in 2011, at the École pratique des hautes études
(Religious Studies Section), entitled The Yoginī Cult and Aspects of Śaivism in
Classical India, supervised by Professor Lyne Bansat-Boudon. She regularly con-
tributes to the dictionary of Hindu tantric terminology (Tāntrikābhidhānakośa,
Austrian Academy of Sciences, Vienna) and participates in the Skandapurāṇa
Project (Leiden-Kyoto). In addition to papers on various aspects of Śaivism and
the early yoginī cult, she has also published on epigraphy, Tamil Śaiva devo-
tional poetry, and classical Sanskrit literature.
Anthony Tribe
is an independent scholar working in the field of Indian tantric Buddhism, and
at present a senior fellow of the Center for Buddhist Studies at the University
of Arizona, Tucson. He received his doctorate in Indian Buddhism from Oxford
in 1995. Subsequently, he taught in the Asian Studies program at the Univer-
sity of Montana, Missoula, USA. In Oxford, his doctoral thesis on Vilāsavajra’s
Nāmamantrārthāvalokinī was supervised by Professor Sanderson. He remem-
bers with deep gratitude and much warmth the personal tutoring, encour-
agement and friendship he received from him during that time. Many after-
noons and evenings were spent in Professor Sanderson’s study at his home
in Eynsham: learning how to read manuscripts and produce critical editions;
exploring Vilāsavajra’s tantric Buddhism; and being fed before catching a late
bus back to Oxford. He remembers too the infectiousness of Professor Sander-
son’s enthusiasm and commitment, his humour, looking after his house and
cat, the clatter of the keyboard on his early, tiny-screened but magical, Apple
Macintosh computer, and the kindness of being given the use of it while he
was on sabbatical for a term lecturing in Paris. Dr. Tribe is also a fellow of the
Royal Asiatic Society, and his publications include Tantric Buddhist Practice in
via free access
xxiv notes on contributors
India: Vilāsavajra’s commentary on the Mańjuśrī-nāmasaṃgīti. A critical edition

and annotated translation of Chapters 1–5 with introductions, and (as co-author)
Buddhist Thought: a complete introduction to the Indian tradition, both pub-
lished by Routledge. At present he lives in Tucson, Arizona, with two cats and
too many books. He tries to keep cool in the summer.
holds a BA (magna cum laude) in Religion and Classics from the University of
Rochester, an MA in South Asian Studies from U.C. Berkeley, an MPhil in Classi-
cal Indian Religions from Oxford, and a PhD in South Asian Studies (Sanskrit)
from U.C. Berkeley. His doctoral dissertation of 2014 focuses on the role of reli-
gious experience in the traditions of Tantric Śaivism, and is entitled “To Enter,
to Be Entered, to Merge: The Role of Religious Experience in the Traditions of
Tantric Shaivism.” He has studied with Professor Sanderson formally and infor-
mally at Oxford, Leipzig, Kyoto, and Portland. He is currently a freelance scholar
lecturing internationally and a guest lecturer at Naropa University in Boulder,
Colorado.
Alex Watson
completed his BA in Philosophy and Psychology (Oxford), and MA in Hindi,
Hinduism and Indian Philosophy (SOAS, University of London), before return-
ing to Oxford to complete an MPhil in Classical Indian Religions. After this,
he began his DPhil under Alexis Sanderson’s supervision on the Śaiva thinker,
Bhaṭṭa Rāmakaṇṭha II, and his arguments against the Buddhist doctrine of no-
self. Following a postdoctoral fellowship at Wolfson College, he taught Sanskrit
at St James’ School, London, and then held short-term visiting appointments at
the University of Vienna and Kyushu University, Japan. He was associated with
the EFEO, Pondicherry, for a number of years, and was Preceptor in Sanskrit at
Harvard University before taking up his present position as Professor of Indian
Philosophy at Ashoka University. He is author of The Self’s Awareness of Itself
(2006) and, with Dominic Goodall and Anjaneya Sarma, An Enquiry Into the
Nature of Liberation (2013), as well as several articles in the Journal of Indian
Philosophy and Philosophy East and West.
via free access
A Note on Alexis Sanderson and Indology
Dominic Goodall and Harunaga Isaacson
On the occasion of the Symposium organised by Srilata Raman and Shaman

Hatley in Toronto in March 2015, Harunaga Isaacson was given the gratifying
but also daunting task of delivering a eulogy of Alexis Sanderson. This note is
only very slightly based on what we, Dominic Goodall and Harunaga Isaacson,
remember of the speech given on the occasion, since it was in large part extem-
porised from skeletal notes, and since it contained jokes and science-fictional
scenarios that worked well in the telling, but that proved hard to commit to
writing without losing their intended flavour.
Born in 1948, G.J.S. Sanderson later chose to be known as Alexis because he
liked the name and was known by it by friends in Greece. His early education,
at the Royal Masonic School for Boys in Bushey, a no-frills charitable boarding
school where bromide was said to be administered in the boys’ tea, was fol-
lowed by undergraduate years at Balliol College, Oxford, where he took degrees
in Classics (1969) and Sanskrit (1971). He then spent a large part of a six-year
period in Kashmir, studying with the scholar and Śaiva guru Swami Lakshman
Joo, during which time he was simultaneously Domus Senior Scholar at Merton
College (1971 to 1974) and then Platnauer Junior Research Fellow at Brasenose
College (1974 to 1977).
Despite not having taken a doctorate, he was appointed University Lecturer
in Sanskrit and Fellow at Wolfson College in 1977, where he remained until he
became Spalding Professor of Eastern Religions and Ethics at All Souls College
in 1992. He never got around to taking a doctoral degree, or indeed to finish-
ing any book-length publication in that period, partly because he was so busily
occupied with teaching all manner of Sanskrit texts to students of every level.
As a by-product of his projected thesis on the little-read and still unpublished
Yonigahvara, he had in fact produced a grammar of aiśa language, in other
words of the sorts of irregular Sanskrit encountered in the Nepalese palm-leaf
manuscripts that transmit many tantras. But that grammar has not seen the
light of day. Looked at in this light, his career is reminiscent of a 19th-century
tradition of scholarship, where recognition depended less on publications, cita-
tions and the acquisition of degrees.
A few months after the Toronto Symposium in Alexis’s honour, one of the
two authors of this preface, Dominic Goodall, and perhaps several others of
the contributors too, found himself in the strange position of being asked by an
American administrative authority to supply a letter of reference for his own
via free access
xxvi goodall and isaacson
tutor. The letter perhaps now gathers dust in some bureaucratic archive, but we
can now aptly quote its first two paragraphs in this note:
I had the great good fortune to begin my studies of Sanskrit at Oxford

under Alexis Sanderson in 1988. At that time, he had the post of Univer-
sity Lecturer in Sanskrit, a rare achievement because he had not taken a
doctorate, and had then published rather little: a couple of reviews (1985),
one ground-breaking article on “Purity and Power among the Brahmins
of Kashmir” (1985)—an article so compact that it seemed like a tightly
compressed book—and one article for “general readers”. This last had few
references, since it was intended as an overview, in an encyclopedia of
the world’s religions, of “Śaivism and the Tantric Traditions” (1988). Clear
and very readable, this article remains, twenty-seven years later, the best
overview there is of a huge subject, covering a broad range of largely still
unpublished early medieval literature for the first time, and providing,
again for the first time, a model of how the various parts of the vast and
complex corpus were related and hierarchized by followers of a major cur-
rent of Indian religion.
His “lectures” at that time were really more like intensive reading
classes, which sometimes seemed to take place round the clock, with stu-
dents often filing in to his room for one class just as others filed out from
another, and they were for me the most intellectually exciting events I
attended at Oxford. I had come up to study Greek, Latin and German in
the autumn of 1986, but decided to switch to Sanskrit after the first pub-
lic exam, in 1988, so I had had 5 terms of lectures and weekly tutorials on
Homer, Virgil, Ovid, Plato and the like behind me. From the settled cer-
tainties of centuries of classical scholarship on a relatively small corpus,
I had moved to a literary universe with few well-founded editions, few
published translations and annotations in European languages and seem-
ingly endless questions. Classes were therefore essential, and although
there were other learned teachers in Oxford at the time who were well-
read in certain genres, it was “Mr. Sanderson” whom everybody acknowl-
edged to have the broadest reach, and who therefore was called upon to
teach whatever was required in the genres of philosophy, courtly poetry,
exegesis of traditional Indian law or indeed any sort of technical com-
mentary.1 Because there was so often no time between classes, there was
presumably never much time to prepare; but preparation never seemed
1 In the realm of classical Sanskrit literature, a notable exception here was the technical liter-
ature of traditional grammar, for which we were fortunate to have the guidance of Dr. James
via free access
a note on alexis sanderson and indology xxvii
necessary. We, the students, would attempt to render a line of a given text,
and Mr. Sanderson would interrupt, constantly, with explanations to set
us right where we were going wrong. Nearly every word called for com-
ment or explanation of knowledge that needed to be taken into consider-
ation: details of manuscript-transmission, issues of text-criticism, seman-
tic flavours not recorded in dictionaries, particle-usage not recorded in
grammars, essential religious or historical context not described in pub-
lished secondary literature, and so forth. This might all sound rather dry,
but it was delivered with humour, verve, plenty of eye-contact, a rich and
well-chosen vocabulary and an evident delight in teaching. And it always
zipped by so fast, provoking further questions along the way, that it could
never all be noted down. In short, it was thrilling. So much so, that after
two years of post-graduate study in Hamburg, I decided in 1992 that there
was no alternative as interesting to me as returning to Oxford with a doc-
toral theme consciously chosen to be of potential interest to the same
teacher.
In the interim, Alexis had become Professor Sanderson, having acquired the
Spalding chair for Eastern Religions and Ethics at Oxford’s most prestigious
college. There was, in consequence, a marked change in teaching style. The
lectures were now magisterial, theme-oriented, weekly talks on aspects of his
chosen field: early medieval religion, focusing on the history of Śaivism, its rela-
tions with the state, and its influence on Buddhism and Vaiṣṇavism. And they
were well-attended events, taking place around a very long dining table in the
Wharton Room of All Souls College. Each week, there would be a substantial
and beautifully typeset hand-out giving passages of often unpublished materi-
als,2 and each week several of us gathered naturally together to discuss it after-
wards over lunch at Wolfson College, for it was there that several of the throng
of new doctoral students were enrolled, or over tea in the crypt of the Univer-
sity Church. It was in this period, because he was at last less rushed than he had
been as a lecturer, that Alexis Sanderson entered his first phase of prolific writ-
ing, to begin publishing his many discoveries. To date, his work has appeared
exclusively in articles, although several of them run into hundreds of pages and
are actually book-length studies, accepted nonetheless in journals and volumes
of essays because of their truly exceptional quality and importance.
Benson, who first introduced me to Sanskrit and painstakingly began to reveal the complex-
ities of the thought of Pāṇini. Rereading this letter, I am prompted to add that I am of course
grateful to him and to all of my other teachers too.
2 Some of these hand-outs are now available online for download from Alexis Sanderson’s
academia.edu page.
via free access
xxviii goodall and isaacson
His most celebrated piece is perhaps “The Śaiva Age—The Rise and Domi-
nance of Śaivism during the Early Medieval Period” (2009), an “article” of more
than 300 pages that bears in fact upon all the classical religions of India, and
not just upon Śaivism. Among his other outstanding articles we may mention
just two that might be said to have revolutionised different fields of study that
were not in fact at the centre of Alexis Sanderson’s scholarly interests. The first
is “The Śaiva Religion among the Khmers. Part I” of 2004, which covers fully
114 pages of the large-format Bulletin de l’École française d’Extrême-Orient. This
paper has much of importance to say about how Śaivism may be defined and
how it has manifested itself in different regions, but it is also essential read-
ing for historians of medieval Cambodia and other parts of South East Asia, by
whom it is much quoted. The second, in length a more conventional article of
just 18 pages, is entitled “Vajrayāna: Origin and Function” (1994); this proposes
a new paradigm for the understanding of Tantric Buddhism and has therefore
relaunched a vigorous debate among scholars of Buddhism about the relations
between Śaivism and Buddhism.
As stated above, these articles are in fact peripheral to Alexis Sanderson’s
abiding central focus of interest, the work of India’s most famous tantric
thinker, the prolific polymath Abhinavagupta, who lived in Kashmir at the
turn of the first millennium, where he produced a corpus of rich, difficult and
influential Sanskrit works on poetry, theatre, aesthetics, theology, ritual and sal-
vation. In 2015 Alexis Sanderson retired from the Spalding professorship and
since then has been able at long last to concentrate exclusively on the most
celebrated work of this seminal thinker, the vast and complex “Light on the
Tantras” (Tantrāloka), working on a critical edition of the text, with an anno-
tated English translation and a detailed commentary.3
In other words, the work that Alexis is currently engaged in is the culmi-
nation of a lifetime of research on Abhinavagupta’s place in Indian thought
and the diverse Śaiva and Śākta traditions that informed his Śaivism and are
in varying degrees subsumed within it. His other contributions to our under-
standing of Indian intellectual history, dazzling though they may be, are mostly
the offcuts and side-products of his preoccupation with this literary giant. Since
“retiring” he has now been able to write up his prodigious knowledge about
what has for him always been the “central story.”
3 In spite of its fame and in spite of its being the focus of numerous scholars’ work over the
last century, only one translation of the Tantrāloka into a European language has ever been
completed, that of Raniero Gnoli.
via free access
a note on alexis sanderson and indology xxix
We do not always find excellence in research combined with excellence

in teaching. But Alexis’s career as a teacher has been extraordinary too. Tes-
timony to the truly exceptional qualities of Alexis Sanderson as an inspiring
teacher may be found by looking around the universities of India, Europe,
North America and Japan where his students have been employed; they are not
clustered together in one academic fiefdom, but have spread widely abroad and
attained international recognition as scholars in a range of subjects from classi-
cal Indian theatre to the history of yoga. They include, for example, Jason Birch,
Parul Dave, Csaba Dezső, Paul Gerstmayr, Dominic Goodall, Jürgen Hanneder,
Gergely Hidas, Madhu Khanna, Csaba Kiss, Nina Mirnig, John Nemec, Srilata
Raman, Isabelle Ratié, Péter-Dániel Szánto, Judit Törzsök, Somadeva Vasudeva,
James Mallinson, Ryugen Tanemura, Joel Tatelman, Anthony Tribe, and Alex
Watson.
We have mentioned Alexis’s reading-classes and his impressively rich lec-
tures, but what many of his direct students may remember best are interactions
with him in tutorials. He tended to offer aspiring doctorands many hours of
extremely helpful criticism and coaching for the first year or so, and then, when
he judged them capable of working more independently, he would nudge the
doors of opportunity half-closed and so encourage them to get on with their
work by themselves. Once they were thus launched, they would be invited to
deliver a lecture in his graduate research seminar, an experience which many
will remember as both daunting and exhilarating, requiring the victims to give
of their very best before an audience of fellow students along with Alexis and
Harunaga Isaacson, typically seated to their right and left, at whom they would
be casting furtive glances to search for their reactions!
Alexis is something of a raconteur when the mood takes him, imitating
the accents and mannerisms of the cast that people his narrations, and so we
tended to learn unwritten snippets of history about other indologists from him.
One annual occasion was particularly propitious for this. Professor Gombrich
used to mark the end of the summer term, and so of the academic year, with
a lunch in his garden, after which several of us would walk to the churchyard
of St. Mary’s in Kidlington to visit the grave of another former Boden Professor
of Sanskrit, Thomas Burrow. This never failed to call forth a string of reminis-
cences of Alexis, beginning with something about Professor Burrow himself,
but leading often to Professor Brough and others.
The above paragraphs recall Alexis’s interactions with students who spent
years at Oxford. But there were also many others who came for only short
periods or whose interactions were only or largely epistolary and upon whose
work Alexis nevertheless had an important influence. Those who have received
Alexis’s immense letters, typically packed with quotations from unpublished
via free access
xxx goodall and isaacson
Sanskrit texts marshalled to demonstrate ideas, doubts and conclusions, will

know just how extraordinarily rich and useful they are. In some well-known
cases, they have provided invaluable evidence for the recipients’ books. Parts of
David Gordon White’s The Alchemical Body, for instance, or Frederick Smith’s
The Self Possessed, or François Grimal’s edition of Harihara’s commentary on
the Mālatīmādhava, are heavily indebted to lengthy letters from Alexis.
In a bygone age, it might have been appropriate to gather together in one
publication all Alexis’s Kleine Schriften, or all his published work, as was done
just over a century ago for another illustrious thinker whose work helped
shaped knowledge both of Indian and Cambodian history, namely Auguste
Barth, the first volume of whose complete Œuvres was published in 1914. But
unless and until the internet implodes, such an endeavour seems unneces-
sary: Indologists throughout the world have PDF copies of Alexis’s published
works, a list of which is appended to the end of this preface. A collection of
his many fascinating letters would be a boon, but gathering and editing them
seems impracticable, and we hope that Alexis will himself continue publishing
such discoveries as they document, as well as others, in the publications that
he continues to work on today.
What we hope and expect instead is that the papers gathered in this volume
will reveal some of the many ways in which Alexis has been influential, their
authors showing us in the mirrors of their own bright intellects some reflec-
tions of the radiance, the prakāśa, of Alexis Sanderson.
via free access
Bibliography of the Published Works of Alexis
G.J.S. Sanderson (1983–2019)
Books
2015 (in collaboration with Dominic Goodall, Harunaga Isaacson, Diwakar Acharya
et al.) The Niśvāsatattvasaṃhitā: The Earliest Surviving Śaiva Tantra. Collection
Indologie 128/Early Tantra Series 1. Pondicherry: Institut français de Pondich-
éry/École française d’Extrême-Orient/Asien-Afrika Institut, Universität Ham-
burg.
2013 (in collaboration with Jürgen Hanneder and Stanislav Jager) Ratnakaṇṭhas Sto-
tras: Sūryastutirahasya, Ratnaśataka und Śambhukṛpāmanoharastava. Indolog-
ica Marpurgensia 5. München: Kirchheim Verlag.
Articles
2019 “How Public was Śaivism?” In Tantric Communities in Context, edited by Nina
Mirnig, Marion Rastelli, and Vincent Eltschinger, pp. 1–46. Sitzungsberichte
der philosophisch-historischen Klasse 899; Beiträge zur Kultur- und Geistes-
geschichte Asiens 99. Vienna: Austrian Academy of Sciences Press, 2019.
2015 “Tolerance, Exclusivity, Inclusivity, and Persecution in Indian Religion During
the Early Mediaeval Period.” In Honoris Causa: Essays in Honour of Aveek Sarkar,
edited by John Makinson, pp. 155–224. London: Allen Lane.
2014 “The Śaiva Literature.” Journal of Indological Studies (Kyoto) 24 & 25 (2012–
2013), 2014: pp. 1–113.
2013 “The Impact of Inscriptions on the Interpretation of Early Śaiva Literature.”
Indo-Iranian Journal 56 (2013): pp. 211–244. https://doi.org/10.1163/15728536‑135
60308.
2010 “Ritual for Oneself and Ritual for Others.” In Ritual Dynamics and the Science
of Ritual, vol. II: Body, Performance, Agency, and Experience, edited by Angelos
Chaniotis et al., pp. 9–20. Wiesbaden: Harrassowitz Verlag, 2010.
2009 “The Śaiva Age—The Rise and Dominance of Śaivism during the Early Medieval
Period.” In Genesis and Development of Tantrism, edited by Shingo Einoo, pp. 41–
349. Tokyo: University of Tokyo, Institute of Oriental Culture.
2007 “Atharvavedins in Tantric Territory: The Āngirasakalpa Texts of the Oriya Paip-
palādins and their Connection with the Trika and the Kālīkula, with criti-
cal editions of the Parājapavidhi, the Parāmantravidhi, and the *Bhadrakālī-
via free access
xxxii bibliography of the published works of alexis sanderson
mantravidhiprakarana.” In The Atharvaveda and its Paippalāda Śākhā: Histor-

ical and Philological Papers on a Vedic Tradition, edited by Arlo Griffiths and
Annette Schmiedchen, pp. 195–311. Geisteskultur Indiens: Texte und Studien,
11, Indologica Halensis. Aachen: Shaker Verlag.
2007 “The Śaiva Exegesis of Kashmir.” In Mélanges tantriques à la mémoire d’Hélène
Brunner/Tantric Studies in Memory of Hélène Brunner, edited by Dominic
Goodall and André Padoux, pp. 231–442 and 551–582 (bibliography). Collec-
tion Indologie 106. Pondicherry: Institut Français d’Indologie/École française
d’Extrême-Orient.
2007 “Swami Lakshman Joo and His Place in the Kashmirian Śaiva Tradition.” In
Samvidullāsaḥ: Manifestation of Divine Consciousness. Swami Lakshman Joo,
Saint-Scholar of Kashmir Shaivism, edited by Bettina Bäumer and Sarla Kumar,
pp. 93–126. New Delhi: D.K. Printworld.
2006 “The Date of Sadyojyotis and Bṛhaspati.” Cracow Indological Studies 8 (2006):
pp. 39–91.
2006 “The Lākulas: New evidence of a system intermediate between Pāñcārthika
Pāśupatism and Āgamic Śaivism.” Ramalinga Reddy Memorial Lectures, 1997.
The Indian Philosophical Annual 24 (2006): pp. 143–217.
2005 “A Commentary on the Opening Verses of the Tantrasāra of Abhinavagupta.”
In Sāmarasya: Studies in Indian Arts, Philosophy, and Interreligious Dialogue
in Honour of Bettina Bäumer, edited by Sadananda Das and Ernst Fürlinger,
pp. 89–148. New Delhi: D.K. Printworld.
2004 “Religion and the State: Śaiva Officiants in the Territory of the Brahmanical
Royal Chaplain (with an appendix on the provenance and date of the Netratan-
tra).”Indo-Iranian Journal 47 (2004): pp. 229–300. https://doi.org/10.1007/s10783
‑005‑2927‑y
2004 “The Śaiva Religion Among the Khmers, Part I.” Bulletin de l’Ecole française
d’Extrême-Orient, 90–91 (2003–2004): pp. 349–463. https://doi.org/10.3406/bef
eo.2003.3617
2001 “History through Textual Criticism in the study of Śaivism, the Pañcarātra and
the Buddhist Yoginītantras.” In Les Sources et le temps. Sources and Time: A
Colloquium, Pondicherry, 11–13 January 1997, edited by François Grimal, pp. 1–
47. Collection Indologie 91. Pondicherry: Institut Français de Pondichéry/École
Française d’Extrême-Orient.
1995 “Meaning in Tantric Ritual.” In Essais sur le Rituel III: Colloque du Centenaire de
la Section des Sciences religieuses de l’École Pratique des Hautes Études, edited by
A.-M. Blondeau and K. Schipper, pp. 15–95. Bibliothèque de l’École des Hautes
Études, Sciences religieuses CII. Louvain-Paris: Peeters.
1995 “Vajrayāna: Origin and Function.” In Buddhism into the Year 2000. International
Conference Proceedings, pp. 89–102. Bangkok and Los Angeles: Dhammakāya
Foundation.
via free access
bibliography of the published works of alexis sanderson xxxiii
1995 “The Sarvāstivāda and its Critics: Anātmavāda and the Theory of Karma.” In
Buddhism into the Year 2000. International Conference Proceedings, pp. 33–48.
Bangkok and Los Angeles: Dhammakāya Foundation.
1992 “The Doctrine of the Mālinīvijayottaratantra.” In Ritual and Speculation in Early
Tantrism. Studies in Honour of André Padoux, edited by T. Goudriaan, pp. 281–
312. Albany: State University of New York Press.
1990 “The Visualization of the Deities of the Trika.” In L’image divine: Culte et médi-
tation dans l’hindouisme, edited by A. Padoux, pp. 31–88. Équipe de recherche
n° 249: “L’hindouisme: textes, doctrines, pratiques.” Paris: Éditions du Centre
National de la Recherche Scientifique.
1986 “Mandala and Āgamic Identity in the Trika of Kashmir.” In Mantras et Dia-
grammes Rituelles dans l’Hindouisme, edited by Andre Padoux, pp. 169–214.
Équipe de recherche n° 249: “L’hindouisme: textes, doctrines, pratiques.” Paris:
Éditions du Centre National de la Recherche Scientifique.
1985 “Purity and Power among the Brāhmans of Kashmir.” In The Category of the Per-
son: Anthropology, Philosophy, History, edited by M. Carrithers, S. Collins and
S. Lukes, pp. 190–216. Cambridge: Cambridge University Press.
Reference Articles
2015 “Śaiva Texts.” In Brill’s Encyclopedia of Hinduism, Volume VI: Index, edited by
Knut A. Jacobsen, pp. 10–42. Handbuch der Orientalistik. Zweite Abteilung,
Indien 22/6. Leiden and Boston: Brill.
2009 “The Hinduism of Kashmir,” published as the entry “Kashmir” in Brill’s Ency-
clopedia of Hinduism, Volume I: Regions, Pilgrimage, Deities, edited by Knut
A. Jacobsen, pp. 99–126. Handbuch der Orientalistik. Zweite Abteilung, Indien
22/1. Leiden and Boston: Brill.
1988 “Śaivism and the Tantric Traditions.” In The World’s Religions, edited by S. Su-
therland, L. Houlden, P. Clarke and F. Hardy, pp. 660–704. London: Routledge
and Kegan Paul, 1988. Reprinted in The World’s Religions: The Religions of Asia,
edited by F. Hardy, pp. 128–172. London: Routledge and Kegan Paul, 1990.
1987 “Śaivism in Kashmir.” In The Encyclopedia of Religion, vol. 13, edited by Mircea
Eliade, pp. 16–17. New York: Macmillan Publishing Company.
1987 “Trika Śaivism.” In The Encyclopedia of Religion, vol. 13, edited by Mircea Eliade,
pp. 15–16. New York: Macmillan Publishing Company.
1987 “Krama Śaivism.” In The Encyclopedia of Religion, vol. 13, edited by Mircea Eli-
ade, pp. 14–15. New York: Macmillan Publishing Company.
1987 “Abhinavagupta.” In The Encyclopedia of Religion, vol. 13, edited by Mircea Eli-
ade, pp. 8–9. New York: Macmillan Publishing Company.
via free access
xxxiv bibliography of the published works of alexis sanderson
Reviews
2002 “Remarks on the Text of the Kubjikāmatatantra.”Indo-Iranian Journal 45 (2002):

pp. 1–24. https://doi.org/10.1163/000000002124994513
1990 Review of: “Paul Eduardo Muller-Ortega, The Triadic Heart of Śiva: Kaula Tantri-
cism of Abhinavagupta in the Non-Dual Shaivism of Kashmir. SUNY Series in the
Shaiva Traditions of Kashmir. Albany, N.Y.: State University of New York Press,
1989.”Bulletin of the School of Oriental and African Studies 53 (1990): pp. 354–357.
https://doi.org/10.1017/S0041977X00026355.
1985 Review of: “N.R. Bhatt, Mataṅgapārameśvarāgama (Kriyāpāda, Yogapāda et
Caryāpāda), avec le commentaire de Bhatta Rāmakaṇṭha. Édition critique. Col-
lection Indologie 65. Pondichéry: Institut Français de I’Indologie, 1982; Idem:
Rauravottarāgama. Édition critique, introduction et notes. Collection Indologie
66. Pondichéry: Institut Français de I’Indologie, 1983.” Bulletin of the School
of Oriental and African Studies 48 (1985): pp. 564–568. https://doi.org/10.1017/
S0041977X00038714.
1983 Review of: “Lilian Silburn, Śivasūtra et Vimarśinī de Kṣemarāja. (Études sur le
Śivaϊsme du Cachemire, École Spanda.) Traduction et introduction. Publications
de I’Institut de Civilisation Indienne, Série in-8°, Fasc. 47. Paris: Institut de Civil-
isation Indienne, 1980. (Distribution: Diffusion E. De Boccard, 11 Rue de Médi-
cis, Paris VIe.).” Bulletin of the School of Oriental and African Studies 46 (1983):
pp. 160–161. https://doi.org/10.1017/S0041977X00077673.
via free access
Introduction
Dominic Goodall, Shaman Hatley, Harunaga Isaacson
and Srilata Raman
Academic study of Asia’s tantric traditions has blossomed in recent decades.

Once dismissed as marginal, or unworthy of serious attention, we now under-
stand the Śaiva, Buddhist, Vaiṣṇava, and Jaina tantric traditions as integral to
the religious and cultural landscapes of medieval South, Southeast, Central,
and East Asia. This shift, which is reshaping the historiography of medieval
India, is in no small measure due to the magisterial contributions of Alexis
G.J.S. Sanderson, Fellow of All Souls College and Spalding Professor of Eastern
Religion and Ethics at Oxford University, from 1992–2015, and now Emeritus
Fellow of All Souls College. The present book is a collection of essays in his hon-
our, written by specialists of the various fields he has influenced from around
the world, most of whom were his students at Oxford.
The twenty-three chapters of this volume span multiple fields of Indology.
Organized around the theme of “Śaivism and the Tantric Traditions,” the essays
are nonetheless diverse in method, historical context, and source material.
Hinting at Alexis Sanderson’s own scholarly breadth, the essays here assembled
span the history, ritual, and philosophies of Śaivism and Tantric Buddhism,
Vaiṣṇavism, religious art and architecture, and Sanskrit belles-lettres. Together,
they represent a significant contribution to our understanding of the cultural,
religious, political, and intellectual histories of premodern South and South-
east Asia. Most of the contributions are original studies of primary sources of
the tantric traditions, reflecting Sanderson’s relentless commitment to philol-
ogy and the discovery of new sources. The essays have been grouped into five
parts, within which they appear more or less chronologically, according to sub-
ject matter. Part 1 concerns early Śaiva traditions: the pre-tantric Śaivism of the
Atimārga, as well as the Niśvāsatattvasaṃhitā, perhaps the earliest surviving
Śaiva tantra. The essays of part 2 concern Śaiva and Buddhist exegetical and
philosophical traditions. Part 3 brings together studies on the topics of religion,
polity, and social history, while part 4 (“Mantra, Ritual, and Yoga”) concerns
religious practices. Part 5’s essays on art and architecture complete the vol-
ume. Naturally, the five parts of this book overlap somewhat, and several essays
would be at home in more than one.
The volume’s first chapter, by Peter Bisschop, bears the title “From Mantra-
mārga back to Atimārga: Atimārga as a Self-referential Term.” Following the
publication of Alexis Sanderson’s groundbreaking article “Śaivism and the Tan-
© dominic goodall et al., 2020 | doi:10.1163/9789004432802_002

This is an open access chapter distributed under the terms of the CC BY-NC 4.0 license.
via free access
2 goodall et al.
tric Traditions” (1988), the division of Śaivism into Atimārga and Mantramārga
has become commonplace among students of Śaivism. Atimārga in this classi-
fication refers to the ascetic path associated with the Pāśupatas and Lākulas,
while Mantramārga refers to the ‘higher’ tantric path with its various sub-
divisions. Bisschop’s paper first of all observes that this division represents a
purely Mantramārga perspective on Śaivism, for so-called ‘Atimārga sources’
seemingly do not use the term Atimārga. The main part of the paper then draws
attention to a passage from an unpublished ca. twelfth-century Māhātmya of
Vārāṇasī that does uniquely use the term in what may be called an Atimārga
context. The passage in question centers on Vārāṇasī’s cremation ground and
Bhairava’s teachings and activities there. The passage on the one hand attests
to the existence of a strong Atimārga community in Vārāṇasī around the time
of the text’s composition, but also to the transmission and knowledge of the
Svacchanda there. It also testifies to the fact that the views on what constituted
Śaivism in early-medieval India differed across different Śaiva traditions and
that much of our modern understanding derives from specific textual tradi-
tions that only represent one layer within a much broader spectrum of religion
oriented around the worship of Śiva.
Chapter 2, by Judit Törzsök, addresses the question, “Why are the Skull-
Bearers (Kāpālikas) Called Soma?” One of the alternative names by which
pre-tantric Kāpālikas or ‘Skull-Bearers’ were referred to in classical India was
“Those Who Profess the Soma Doctrine” or simply “Soma People” (somasi-
ddhāntavādin, somajana). The word Soma also appears regularly as the last
part of their initiation name. Törzsök’s chapter explores what this appella-
tion could have meant for the Kāpālikas according to period sources, includ-
ing inscriptions, purāṇic and dramatic literature, without arriving at a defini-
tive answer. First, it is argued that the later derivation sa-Umā (‘accompanied
by Umā’) is probably to be rejected, for female initiates also bore the name
-Somā. Second, the appellation is also unlikely to refer to the vedic Soma,
for Kāpālikas were not commonly known to perform vedic sacrifices. It is
possible that the name Soma may derive from the name of their legendary
founder, Somaśarman, but it may have had additional connotations. It may
also have referred to the moon and its whiteness (evoking the whiteness of the
ashes of the cremation ground or of human bones), the nectar of immortal-
ity (amṛta) the moon is supposed to contain (immortality being a main goal
of all power-seekers), or any nectar, such as alcohol, regularly used in Kāpā-
lika worship. Finally, given the polysemy of the word, it could also be under-
stood to mean “the best,” implying that the Kāpālikas, just as other Śaivas,
considered themselves to follow the best way that leads to power and final
release.
via free access
introduction 3
Chapter 3, by Dominic Goodall, is entitled “Dressing for Power: on vrata,

caryā, and vidyāvrata in the Early Mantramārga, and on the Structure of the
Guhyasūtra of the Niśvāsatattvasaṃhitā.” One of several meanings proposed
by Monier-Williams for the term vrata is “a religious vow or practice,” which
has led to the widespread tendency to translate vrata with “vow,” thus call-
ing to mind a web of partly alien ideas about promised religious undertakings
that culminate in offerings made ex voto suscepto, upon attainment of one’s
desired end. A better approximation is perhaps “timed religious observance.”
This paper attempts to address the question “What is a vrata?” by attempt-
ing to uncover how the notion is used and understood in early works of the
Mantramārga, in particular the sūtras of the Niśvāsatattvasaṃhitā. In doing so,
it touches upon the layered composition of the Guhyasūtra. The early history of
the term caryā (in the tetrad jñānā, kriyā, caryā and yoga) is also illuminated,
as well as the use of the expressions vidyāvrata, puraścaryā and pūrvasevā.
Part 2 contains five chapters concerned with Śaiva and Buddhist philosophi-
cal and exegetical traditions. Chapter 4, by Alex Watson, discusses where pre-
cisely the self-theory (ātmavāda) of Bhaṭṭa Rāmakaṇṭha II—the most prolific
and influential of the early Śaiva Siddhānta exegetes (c. 950–1000)—should
be placed in the nexus of other rival positions. Its relation to the self-theory of
Rāmakaṇṭha’s Buddhist and Naiyāyika interlocutors is considered, and so too in
passing to that of the Sāṅkhyas and the non-dualistic Śaivas. A previous article
(Watson 2014) places Rāmakaṇṭha’s Saiddhāntika view in the middle ground
between Nyāya and the momentariness theory (kṣaṇikavāda) of the Buddhists.
The present chapter adds a number of considerations that, while not invalidat-
ing the ‘middle ground thesis,’ show it to be one-sided and incomplete. Some of
these considerations weigh in favour of seeing it as just as ‘extreme’ as Nyāya;
others in favour of seeing it as more extreme than Nyāya. The conclusion con-
siders whether and how these varying perspectives can be integrated.
Chapter 5, by Isabelle Ratié, is entitled “Some Hitherto Unknown Fragments
of Utpaladeva’s Vivṛti (II): Against the Existence of External Objects.” As Ratié
highlights, Utpaladeva’s detailed commentary (the Vivṛti or Ṭīkā) on his own
Īśvarapratyabhijñā treatise was certainly the most innovative text of the Pra-
tyabhijñā corpus; unfortunately, however, to date we only have access to frag-
ments of this work, as first discovered by Raffaele Torella. This chapter is part of
a series of papers by Ratié devoted to the edition, translation and explanation of
shorter fragments of the Vivṛti found in the margins of manuscripts containing
Abhinavagupta’s commentaries on Utpaladeva’s treatise. The paper deals with
fragments of the Vivṛti on verses 1.5.6–9, which argue against the Sautrāntikas’
thesis that we must infer the existence of a reality external to consciousness in
via free access
4 goodall et al.
order to account for phenomenal variety. In these fragments Utpaladeva shows

not only that, as already emphasized by the Vijñānavādins, postulating the exis-
tence of an external world is of no use in the realm of everyday practice, and
that an external object must have contradictory properties whether it is under-
stood as having parts or not, but also that the very act of mentally producing
the concept (and therefore the inference) of an external object is in fact impos-
sible to perform, because an object by nature alien to consciousness is simply
unthinkable.
Chapter 6, by Christopher D. Wallis, is entitled “Alchemical Metaphors for
Spiritual Transformation in Abhinavagupta’s Īśvarapratyabhijñāvimarśinī and
Īśvarapratyabhijñāvivṛtivimarśinī.” In this essay, Wallis examines an alchemi-
cal metaphor for spiritual transformation found in Abhinavagupta’s two com-
mentaries on the Īśvarapratyabhijñākārikā of Utpaladeva. Analyzing this trope
provides insight into Abhinavagupta’s innovative usages of the key terms samā-
veśa, turya, and turyātīta. Additionally, the essay considers his homology of
the fivefold self—Void, prāṇa, the subtle body consisting of the mind and its
faculties (puryaṣṭaka), and physical body, plus the transindividual Power of
Awareness (citi-śakti)—with the five phases of lucidity: the states of waking,
dreaming, deep sleep, the transcendental ‘fourth’ state, and the state ‘beyond
the fourth’ (turyātīta). As Wallis shows, these passages in the two different
commentaries do not entirely agree, and both present textual problems. His
provisional conclusion is that Abhinavagupta seems to change and develop his
view in the time between the two commentaries: the Vimarśinī features a sim-
pler model of a gnostic transcendentalist turya succeeded by an ‘immanentist’
yogic turyātīta (the latter being marked by the transcendent element’s perva-
sion of all that was previously transcended), while the Vivṛtivimarśinī proposes
two distinct versions of both turya and turyātīta, gnostic and yogic, respectively
(yielding four categories in total), where the yogic is to be preferred despite be-
ing more gradual because in it the saṃskāras of dualistic experience are finally
dissolved. Wallis’ analysis of these problems gives us a deeper understanding
of Abhinavagupta’s thought, and points us in some intriguing directions.
Chapter 7, by Péter-Dániel Szántó, is entitled “On Vāgīśvarakīrti’s Influence
in Kashmir and Among the Khmer.” In this essay, Szántó seeks to elucidate
the role and importance of an early eleventh-century Buddhist scholar, Vāg-
īśvarakīrti, far from his homeland in Eastern India. The first part examines a
passage showing that, probably still during his lifetime, he was considered an
important opponent by a Kashmirian scholar, Ratnavajra. The debate in ques-
tion comcerns the validity of the so-called Fourth Initiation. The second part of
Szántó’s essay advances the hypothesis that, although not mentioned by name,
Vāgīśvarakīrti is referred to in a Khmer inscription from the same century.
via free access
introduction 5
Chapter 8, by Srilata Raman, is entitled “Reflections on the King of Ascetics

(Yatirāja): Rāmānuja in the Devotional Poetry of Vedānta Deśika.” This paper
is concerned with examining one specific hagiographical genre within the
Śrīvaiṣṇava tradition—the praise-poem addressed to the ācārya, in this case
Rāmānuja. It looks in detail at two of these poems, one in Tamil and the
other in Sanskrit. These are “The 100 Antāti Verses on Rāmānuja” (Irāmāṉuja
Nūṟṟantāti) of Tiruvaraṅkattamutaṉār, one of the earliest hagiographical/sto-
tra works we have at hand on Rāmānuja, and Vedānta Deśika’s “The Seventy
Verses on the King of Ascetics” (Yatirāja Saptatiḥ). Analysing the main motifs
of these poems as traceable to the Tamil devotional poetry of the Āḻvārs, the
paper also demonstrates that a central motif within the poems contributes to
a reconsideration of prapatti doctrine in the post-Rāmānuja period, leading to
the idea that “love for the ācārya” (ācāryābhimāna), and, in the most extreme
case, belief in Rāmānuja’s prapatti, is itself sufficient for salvation. The analysis
of the stotra literature on Rāmānuja here, by no means exhaustive but rather
illustrative of the formative phase of doctrine, also reinforces a central con-
tention of this paper: that devotional poetry composed not just by the āḻvārs
but also by later the ācāryas is central—as central as commentaries and inde-
pendent works—to the evolution of Śrīvaiṣṇava doctrine.
The essays of part 3 concern various aspects of religion, the state, and the
social history of premodern India. Chapter 9, by Csaba Dezső, is entitled “Not
to Worry, Vasiṣṭha Will Sort it Out: The Role of the Purohita in the Raghuvaṃśa.”
This essay examines the various tasks Vasiṣṭha fulfils in the Raghuvaṃśa as the
royal chaplain of the kings of the Sūryavaṃśa. As Dezső shows, these are in
harmony with the standards laid down in the Arthaśāstra, from officiating at
life-cycle ceremonies to empowering and defending the king and his army with
the help of Atharvavedic mantras. Vasiṣṭha also acts as the king’s mentor and
chief advisor who tries to reason against Aja’s overwhelming grief, placing the
interests of the dynasty before the king’s private emotions. These verses of the
Raghuvaṃśa invite comparison with a passage in Aśvaghoṣa’s Buddhacarita in
which the chaplain and the minister try to persuade the bodhisattva to return
to the palace and to carry out his role as the heir to the throne.
Chapter 10, by Gergely Hidas, is entitled “Buddhism, Kingship and the Pro-
tection of the State: The Suvarṇaprabhāsottamasūtra and Dhāraṇī Literature.”
Hidas’s essay focuses first on the ritual core of the Suvarṇaprabhāsottamasūtra,
which represents the ritual establishment of the state’s protection as an act
of mutual benefit to the Buddhist Sangha and the monarch. The essay then
explores how this theme appears in some examples of dhāraṇī literature from
the first half of the first millennium. It is shown that offering safeguard for
via free access
6 goodall et al.
rulers and their realms is a long-established practice in South Asian Buddhism,

one that perdures up to modern times, while there have been a variety of incan-
tation scriptures available for such purposes.
Chapter 11, by Nina Mirnig, is entitled “Adapting Śaiva Tantric Initiation
for Exoteric Circles: Lokadharmiṇī Dīkṣā and Its History in Early Medieval
Sources.” The article investigates the history and scope of usage of the term
lokadharmiṇī dīkṣā, one of the most accessible and mainstream-conforming
classes of Śaiva tantric initiation. In essence, this category denotes a form of
initiation that allows the practitioner to maintain his exoteric register of reli-
gious practice (the lokadharma), in this context the brahmanical mainstream.
As such, it is contrasted with the śivadharmiṇī dīkṣā, which operates on purely
Śaiva ritual and soteriological premises. The terminology of the lokadharmiṇī
dīkṣā was used in different initiation-classification schemes, reflecting differing
and evolving ways of negotiating the interface between initiatory and exoteric
practices among different Śaiva tantric groups throughout the early medieval
period. By tracing the shifting history of the lokadharmiṇī dīkṣā terminology in
pre-twelfth century Śaiva tantric sources, the article points to the complexities
of interpretation of terminology relating to initiatory categories.
In chapter 12, John Nemec investigates “Innovation and Social Change in
the Vale of Kashmir, circa 900–1250 C.E.” This essay addresses the nature of
social and religious change by examining the ways in which they are negotiated
in the writings of selected post-scriptural Śaiva authors of the Kashmir Val-
ley. Nemec argues that the writings of Somānanda (ca. 900–950), Utpaladeva
(ca. 925–975), Abhinavagupta (ca. 975–1025), and Jayaratha (early 13th C.E.)
evince a self-consciously constructed, emic theory of scriptural authority and
social conduct that exemplifies what the author contends should be taken as
a maxim in the study of South Asian religions and religion more generally—
namely, that change is not inimical to religion, even if particular religious agents
are not infrequently inimical to change. Envisioning a layered hierarchy of
authoritative scriptural sources, both Vedic and Tantric, these authors deemed
otherwise proscribed religious and social practices permissible in their particu-
lar contexts; yet, because they promoted novel practices only as modifications
to otherwise universally applicable social strictures, the changes they autho-
rized were necessary incremental in nature, which Nemec suggests is in fact
the normative pattern for social change in premodern South Asia. In doing so,
he argues that the complex model of scriptural authority exemplified in this
emic theory challenges Sheldon Pollock’s characterization of the relationship
of theory (śāstra) to practice (prayoga) as monolithic and simple in premodern
South Asia. The essay concludes by sketching the implications of this study for
our understanding of Indian religions, and for religion tout court.
via free access
introduction 7
Chapter 13, by Bihani Sarkar, is entitled “Toward a history of the Navarā-

tra, the autumnal festival of the Goddess.” This essay provides a chronological
chart of the development of the Navarātra, the Nine Nights festival of the God-
dess. Drawing on Sanderson’s work on the Orissan Mahānavamī traditions of
Bhadrakālī and ritual descriptions outlined in Sanskrit sources, Sarkar iden-
tifies four phases in the trajectory of the Navarātra, as it grew into the pre-
eminent political rite for authorizing and creating royal power. These were:
an early Vaiṣṇava rite in the monsoon, its incorporation of a pre-established
Brahmanical military tradition in Āśvina, its expansion into a ten day affair
and inclusion of tantric rituals for powers (siddhis) in East India, and the
growth later of the distinctive Southern and Western Navarātras. Tamil sources
of uncertain date, however, add complexity to this picture.
Part 4 of this book contains six essays on various aspects of religious praxis,
including yoga. In chapter 14, “Śārikā’s Mantra,” Jürgen Hanneder studies the
tantric deity Śārikā, who is worshipped in the form of a large stone on the
“Śārikā Peak” or Pradyumna Peak in Śrīnagar. Hanneder examines several rit-
ual texts that describe the iconography and worship of this goddess, includ-
ing her mantra. In the seventeenth century the Kashmirian author Sāhib Kaul
wrote a Stotra devoted to Śārikā, in which her mantra is given in the style of
a mantroddhāra, that is with code words, so that the “sounds” of the mantra
need not be explicitly uttered. This chapter contains an edition and trans-
lation of this text and an analysis, which shows that Sāhib Kaul’s version of
the mantra of Śārikā strangely fails to accord with most other sources of this
mantra.
Chapter 15, by Diwakar Acharya, is entitled “The Kāmasiddhistuti of King
Vatsarāja.” In this chapter, Acharya presents an edition and translation of the
previously unpublished Śaiva Kāmasiddhistuti attributed to Mahārājādhirāja
Vidyādharacakravartin Vatsarāja, who can perhaps be identified with King Vat-
sarāja of the Gurjara-Pratihāra dynasty (c. 775–805A.D.), the father of Mahārā-
jādhirāja Nāgabhaṭa II (805–833A.D.). As a pūjāstuti, this text guides its reciter
through the mental or actual worship of Goddess Nityā Sundarī, of whom
the poet is a devotee. He invokes the goddess as Maheśvarī and Gaurī, but
concedes that some call her Lakṣmī and Parā Prakṛti. As Acharya shows, the
poet appears unaware of the systems of nine, eleven, or sixteen Nityās, which
are worshipped in the traditions of the Nityākaula, Manthānabhairava, and
Vāmakeśvara Tantras, respectively. Rather, the author is aware of only one
Nityā, who is simply called Sundarī and is installed as Nityā Sundarī at the altar
of worship in the centre of the maṇḍala, without a consort, independent and
supreme.
via free access
8 goodall et al.
Chapter 16, by Shaman Hatley, is entitled “The Lotus Garland (padmamālā)

and Cord of Power (śaktitantu): The Brahmayāmala’s Integration of Inner and
Outer Ritual.” This essay examines the relationship between “ritual” and “yoga”
in the Brahmayāmala, a voluminous early tantra whose place in the history of
Śaivism was first identified by Professor Sanderson (1988). The early history of
Śaiva yoga remains inadequately studied, and foundational early sources such
as the Niśvāsatattvasaṃhitā and Brahmayāmala diverge widely in their tech-
nical vocabulary and conceptions of the body. The focus of this essay is the
explication of the Brahmayāmala’s manner of integrating inner and outer rit-
ual processes, both of which have their basis in a system of Nine Lotuses and
Nine Knots (granthi) strung together by the “cord of power” (śaktitantu, śak-
tisūtra). In analysing this unique system, the essay examines a number of key
issues, including the role of visualization in ritual, mantra-installation (nyāsa),
the body’s subtle channels (nāḍī) and knots (granthi), and shifting conceptions
of the relationship between knowledge ( jñāna) and ritual action (kriyā).
Chapter 17, by James Mallinson, is entitled “The Amṛtasiddhi: Haṭhayoga’s
Tantric Buddhist Source Text.” The unpublished circa eleventh-century Amṛta-
siddhi is the oldest text to teach any of the principles and practices that came
to distinguish the haṭha method of yoga practice taught in later Vaiṣṇava and
Śaiva manuals, such as the Dattātreyayogaśāstra and Haṭhapradīpikā. Many of
its central teachings have no precedents in earlier texts: the yogic body with the
moon situated at the top of the central channel dripping amṛta and the sun at
its bottom consuming it; the three physical techniques that make up the text’s
central practice (mahāmudrā, mahābandha and mahāvedha); the four stages of
the practice (ārambha, ghaṭa, paricaya and niṣpatti); the principle that bindu or
semen is the most important vital constituent and hence that its preservation
is paramount; and the principle that the mind, breath and bindu are connected,
so that controlling one controls the others. These are then repeated, often ver-
batim, in almost all subsequent haṭha texts. The Amṛtasiddhi has been the
subject of only one previous study, an article by Kurtis Schaeffer (2004), which
analyses the text as found in a bilingual (Sanskrit and Tibetan) manuscript that
probably dates to the twelfth century CE. Schaeffer, because of some seem-
ingly non-Buddhist teachings in the text, in particular those on jīvanmukti,
understands it to be a Śaiva work. This paper, however, shows that some of its
teachings are specifically Buddhist, and concludes that the text was composed
in a Buddhist milieu and that later Indian and Nepalese manuscripts of the text
either misunderstood its Buddhist features or deliberately removed or changed
them.
Chapter 18, by Csaba Kiss, is entitled “A Sexual Ritual with Māyā in Matsye-
ndrasaṃhitā 40.” The Matsyendrasaṃhitā, a 13th-century South Indian Kubji-
via free access
introduction 9
kā-Tripurā-oriented tantric yoga text of the Ṣaḍanvayaśāmbhava tradition, de-

scribes a unique sexual ritual in its 40th chapter. Kiss analyses sigificant ambi-
guities therein, in addition to providing an edition and annotated translation
of the relevant passages. The chapter recommends that the yogin have sex-
ual encounters with (human) yoginīs, while avoiding pāśavī (uninitiated?)
women, but devotes most of its attention to a ritual with Māyā, a rather ambigu-
ous female. Is she an imagined goddess (kuṇḍalinī?) or an uninitiated woman
of low birth? Is the sexual act visualized or ‘real’? Kiss argues that these ambi-
guities may be deliberate. The ambiguity between actual sex and visualization
reflects the tension between sexuality and asceticism manifest in the frame
story of the Matsyendrasaṃhitā, a unique version of the legend of Matsyen-
dra and Gorakṣa. While possibly echoing or quoting older tantric texts con-
taining descriptions of sexual rituals, the redactors of the Matsyendrasaṃhitā
were perhaps transitioning towards ascetic or brahmacarya-oriented teach-
ings. As a result, Kiss argues, they came up with an obscure variant on the
figure of the tantric yoginī: Māyā, first described as a phantom, resembling
a goddess visualized in worship, then also takes part in an actual sexual rit-
ual.
Chapter 19, by Jason Birch, is entitled “Haṭhayoga’s Floruit on the Eve of Colo-
nialism.” The aim of this article is to provide a framework for examining the
textual sources on Haṭhayoga that were composed from the sixteenth to eigh-
teenth centuries. After a brief introduction to the early history of Haṭha- and
Rājayoga, the main section of the article focuses on the salient features of the
late literature on Haṭhayoga by dividing the texts into two categories; ‘extended
works’ and ‘compendiums.’ The extended works expatiate on Haṭhayoga as it
was formulated in the Haṭhapradīpikā, whereas the compendiums integrate
teachings of Haṭhayoga within a discourse on yoga more generally conceived.
Both etic categories include scholarly and practical works which, when read
together in this way, reveal significant changes to the praxis and theory of
Haṭhayoga on the eve of colonialism. The article concludes with a brief dis-
cussion on the regional distribution of the literature of Haṭhayoga during
this period and how the codification of its praxis and theory appears to have
diverged in different regions.
The papers of part 5 concern religious art and architecture. Chapter 20, by Lib-
bie Mills, is entitled “The Early Śaiva Maṭha: Form and Function.” We are not
told a great deal in the early Śaiva textual record about the practicalities of
life inside the maṭha. In this chapter, Mills seeks to find a way into the topic
by looking at the physical structure of the buildings. Drawing on materials
that include instructions for building, the essay considers the designs given
via free access
10 goodall et al.
for the construction of the maṭha, and what those designs might tell us of
what took place inside. The paper aims to add to the exploration of the maṭha
treated to great effect by Tamara Sears.1
Chapter 21, by Ryugen Tanemura, concerns “The Kriyāsaṃgrahapañjikā of
Kuladatta and its Parallels in the Śaiva Pratiṣṭhātantras.” As Sanderson demon-
strates in “The Śaiva Age: The Rise and Dominance of Śaivism During the Early
Medieval Period” (2009), tantric Buddhism devised a number of ceremonies
in the domain of public religion following the Śaiva models, such as consecra-
tion (pratiṣṭhā) and funeral rites (antyeṣṭi). Tantric Buddhist manuals called
maṇḍalavidhis teach the details of these public social rituals. These manu-
als closely resemble the Śaiva Pratiṣṭhātantras and Paddhatis. Among these,
the Kriyāsaṃgrahapañjikā of Kuladatta, most probably written in the Kath-
mandu valley in the eleventh century C.E., is particularly rich in information,
as are also the Vajrāvalī of Abhayākaragupta and the Ācāryakriyāsamuccaya
of Jagaddarpaṇa or Darpaṇācārya. The purpose of this paper is to present var-
ious Śaiva parallels in the Kriyāsaṃgrahapañjikā, especially textual parallels
between the nimittokti section of the Kriyāsaṃgrahapañjikā and the Śalyo-
ddhārapaṭala of the Devyāmata, a Śaiva Pratiṣṭhātantra. The relevant sections
and chapters of these two texts concern the topic of how to find and remove
extraneous substances (śalya) underground during the rituals in order to avoid
the calamities which they may cause. Tanemura also presents as an appendix a
preliminary edition and translation of a section of the Ācāryakriyāsamuccaya
called Bhūśalyasūtrapātananimittavidhi; this also contains some parallels with
the Kriyāsaṃgrahapañjikā and the Devyāmata.
Chapter 22, by Anthony Tribe, concerns “Mañjuśrī as Ādibuddha: The Iden-
tity of an Eight-armed Form of Mañjuśrī Found in Early Western Himalayan
Buddhist Art in the Light of Three Nāmasaṃgīti-related Texts.” This chap-
ter examines the identity of an eight-armed form of Mañjuśrī found in early
western Himalayan Buddhist art (11th–13th centuries). This figure is found
most prominently in the Sumtsek (Gsum-brtsegs, “Three-Storeyed”) temple
at Achi, Ladakh where, as a mural, it is the central deity of a maṇḍala. Its
position on the top storey, suggesting that it is the figure of highest status in
the temple’s iconographic programme, has long puzzled scholars. Less well-
known are two other examples of this figure: a mural in the chapel of the
two-armed Maitreya at Mangyu, Ladakh, and a clay figure in the Golden Temple
or Serkhang (Gser-khang) at Lalung, Spiti. Based on descriptions of the fig-
1 Tamara I. Sears, Worldly Gurus and Spiritual Kings, Architecture and Asceticism in Medieval
India (New Haven and London: Yale University Press, 2014).
via free access
introduction 11
ure in three Nāmasaṃgīti-related texts—the Ākāśavimala of Mañjuśrīmitra,

the *Sādhanaupayika (Sgrub pa’i thabs) of Agrabodhi, and the Nāmamantrā-
rthāvalokinī of Vilāsavajra—Tribe argues that the figure should be identified
as (Mañjuśrī in the form of) the Ādibuddha and not Mañjuśrī as a bodhisattva.
The authors of these three texts comprise some of the earliest commentators
on the Nāmasaṃgīti (also known as the Mañjuśrīnāmasaṃgīti), and share a
concern to promote Mañjuśrī by placing him, as the Ādibuddha, at the heart
(both literally and metaphorically) of Mahāvairocana in an adaptation of the
yogatantra Vajradhātu-mahāmaṇḍala.
Chapter 23, by Parul Dave-Mukherji, is entitled “Life and Afterlife of Sādṛśya:
Revisiting the Citrasūtra through the Nationalism-Naturalism Debate in Indian
Art History.” This essay revisits the Citrasūtra, a seminal section on painting
from the Viṣṇudharmottarapurāṇa, in the light of key concerns around the cul-
tural politics of art historiography, the śāstra-prayoga debate, and the related
question of interpretative frames to study traditional Indian art. The last con-
cern has lately come to the forefront in the context of postcolonial studies
and global art history. If the former is critical of intellectual parasitism, the
latter pushes postcolonial thought to explore ‘native’ interpretative frames to
study pre-modern Indian art. This paper attempts to complicate the search
for alternative frameworks by underlining gaps and slippages that surround
the meaning of terms in the given text and their modern appropriation. To
this end, it traces a genealogy of a term, sādṛśya, from the śilpaśāstric lexicon
through its twentieth-century reception in art-historical discourse. How does a
term acquire an afterlife when it enters into the force field of reinterpretation
steeped in cultural nationalism? How would a newly ‘discovered’ Sanskrit text
function in such a space? Dave-Mukherji also addresses a larger question: what
is the genealogy of India’s cultural past, and specifically its “art,” as transcen-
dental/idealistic/spiritual, which has translated itself into a belief? And why
does this belief persist, although in different configurations? The essay then
turns to alternative interpretative frames for the study of Indian art, first by
critically examining ethnographic approaches to the study of texts, and then
by relating Coomaraswamy’s transcendentalism with David Shulman’s recent
discourse around the ‘more than real.’
via free access
via free access
part 1
Early Śaivism
via free access
via free access
chapter 1
From Mantramārga Back to Atimārga:

Atimārga as a Self-referential Term
Peter C. Bisschop
1 Introduction
The impact of Alexis Sanderson’s scholarship can be easily measured by the

subject of the present paper. Before 1988, when Sanderson published his
groundbreaking article “Śaivism and the Tantric Traditions,” the term Atimārga
was hardly used by anyone with the exception of a few specialists of Tantric
Śaivism, and it certainly was never addressed systematically.1 Thus, for exam-
ple, Minoru Hara, who completed his dissertation on the Pāśupatas at Harvard
in 1966 and published extensively on the Pāśupata tradition in the subsequent
decades, never once used the term.2 In the years to come, however, various
scholars started to use it with great confidence in increasing numbers and cur-
rently the Atimārga is widely regarded as one of the two major divisions of
Śaivism, alongside that of the Mantramārga. Quite influential in the dissem-
ination of the term has been Gavin Flood, who adopted it in his An Intro-
duction to Hinduism (1996). The chapters on the Śaiva and Śākta traditions in
this book are deeply dependent on Sanderson’s scholarship. As an illustration
of how commonplace and accepted its use has become, reference may also
be made to the entry on Atimārga in the popular A Dictionary of Hinduism
by W.J. Johnson (2009). The description clearly reflects Sanderson’s scholar-
ship:
One of the two main branches of Śaivism described in the Śaiva Āgamas
or Tantras (the other being the mantramārga, or “path of mantras”). The
atimārga, which is entered on solely in order to attain liberation, is open
only to ascetics. It has two divisions, the Pāśupata, and the Lākula, itself
1 An early reference may be found in Goudriaan and Gupta 1981, 35, 45 (referring to the Niśvā-
satattvasaṃhitā).
2 Although there are many entries starting with ati-, the index of Hara’s collected Pāśupata
Studies (2002) has no entry on Atimārga.
© peter c. bisschop, 2020 | doi:10.1163/9789004432802_003

via free access
16 bisschop
a development from within the Pāśupata tradition; both are concerned

with Śiva in his wild and terrible form of Rudra.3
In the present paper I would like to reconsider the meaning and use of the term
Atimārga, in the light of an intriguing passage from an unpublished Māhātmya
on the holy city of Vārāṇasī (sometimes referred to hereafter as “VM”). In par-
ticular, I want to take up the question whether the term Atimārga was ever
used by Pāśupatas or other groups that one would associate with the Atimārga
themselves. Was it, in other words, ever used self-referentially or does it only
represent a higher, tantric (“Mantramārgic”) perspective on the ascetic path of
Śaivism?
2 Atimārga and Mantramārga
For a start, here is the introduction to the Atimārga from “Śaivism and the
Tantric Traditions,” without a doubt the single most influential article on Śai-
vism of the twentieth century:
The Teaching of Śiva (śivaśāsana) which defines the Śaivas is divided

between two great branches or “streams” (srotas). These are termed the
Outer Path (Atimārga) and the Path of Mantras (Mantramārga). The first
is accessible only to ascetics, while the second is open both to ascetics
and to married home-dwellers (gṛhastha). There is also a difference of
goals. The Atimārga is entered for salvation alone, while the Mantramārga
promises both this, and for those that so wish, the attainment of super-
natural powers (siddhis) and the experience of supernatural pleasures in
the worlds of their choice (bhoga). The Atimārga’s Śaivism is sometimes
called Raudra rather than Śaiva. This is because it is attributed to and con-
cerned with Śiva in his archaic, Vedic form as Rudra (the “Terrible”), the
god of wild and protean powers outside the śrauta sacrifice. It has two
principal divisions, the Pāśupata and the Lākula. (Sanderson 1998, 664.)
An important aspect of the above definition is that it is written from the view-
point of the Mantramārga. Although it follows an emic division of Śaivism,4
it is one which appears to have been coined by the Mantramārga and not by
3 Quoted from the electronic version: http://www.oxfordreference.com/view/10.1093/acref/978

0198610250.001.0001/acref‑9780198610250
4 Cf. Sanderson 2006, 163: “The term Atimārga, which I suggest we use for the non-Āgamic
via free access
from mantramārga back to atimārga 17
the Atimārga itself. In other words, it reflects the higher Mantramārgic system-
atization of doctrines and practices. The Atimārga, in contrast, represents a
more archaic ascetic strand of Śaivism, predating the development of tantric
Śaivism. Initially it was not known as Atimārga, nor do we have evidence
of ascetics who refer to themselves as Atimārgins, “Followers of the Outer
Path.” The name Atimārga was well chosen, however, for the Pāśupata ascetics
adhered to what they called the Atyāśrama “Outer Discipline,” by which they
indicated that they were beyond the four disciplines (āśrama) that define
orthodox Brahmanism. Pāśupatasūtra 2.15–17 plays on the theme of being
“beyond” (ati-) customary practise:
atidattam atīṣṭam atitaptaṃ tapas tathā atyāgatiṃ gamayate
The extraordinary gift, the extraordinary offer, and the extraordinary

practise of asceticism leads to the extraordinary goal.
Whoever coined the term was therefore closely familiar with the tradition.
The division of Śaivism into Atimārga and Mantramārga appears for the first
time in the Niśvāsamukha, the introduction to the Niśvāsatattvasaṃhitā. The
Mantramārga is presented here as the fifth and highest stream (srotas) of reli-
gion, which has been revealed by Śiva’s fifth, upper face.5 The Atimārga, by
contrast, is said to have been revealed by Śiva’s fourth, eastern face. In terms of
hierarchy this indicates a lower position, on a par with that of the revelations
by Śiva’s three other faces: the Laukika or mundane religion taught by his west-
ern face, the Vaidika or brahmanical religion taught by his northern face, and
the Ādhyātmika or system of knowledge of the self taught by his southern face.
On the other hand, its connection with Śiva’s eastern face sets it apart from the
three other religious traditions and it is clear from the Niśvāsatattvasaṃhitā
itself that some of the text’s teachings are in fact deeply influenced by those
of the Atimārga.6 The Niśvāsamukha’s fivefold scheme itself appears to be an
expansion based on a passage from the Manusmṛti (2.117), where three forms
of knowledge are distinguished: Laukika, Vaidika, Ādhyātmika.7
Śaivism of the Pāśupatas and related systems, is extracted, then, from a stage of the tradition
which predates our famous commentators and perhaps even some of the Āgamas them-
selves.”
5 For a critical edition, with annotated translation and accompanying study, of the Niśvāsamu-
kha, see Kafle 2015.
6 See Sanderson 2006.
7 Manusmṛti 2.117 (Olivelle 2015):
via free access
18 bisschop
Following the Niśvāsamukha, we find the division in a range of Mantramārga

scriptures and in texts of authors with a Mantramārga affiliation, although it is
certainly not ubiquitous across tantric literature. As will be seen below, most
important for the present paper is the following passage from the Svacchanda-
tantra:
It is called Atimārga because it is beyond the mental dispositions. It is

taught as “Atimārga” because the doctrine is beyond the worlds. And the
lokas are designated “bound souls,” in the cycle of birth and death. They
who are established in the Atimārga, [that is to say] the followers of the
observance of the skull and the Pāśupatas, they are to be known as beyond
them. There is no rebirth for them and they abide in [the reality of] Īśvara,
in [the world of] Dhruva.8
The division into Atimārga and Mantramārga appears to be found only in texts
belonging to the Mantramārga.9 No Pāśupata or for that matter “Atimārga” text
refers to the concept of Atimārga, let alone to Mantramārga. The only excep-
tion that I am aware of is a medieval Māhātmya about Vārāṇasī, the subject of
this paper.
laukikaṃ vaidikaṃ vāpi tathādhyātmikam eva ca |

ādadīta yato jñānaṃ taṃ pūrvam abhivādayet ||
The first half of this verse is identical to two pādas in Niśvāsamukha 1.26–1.27ab, where the
fivefold division is introduced for the first time:
śṛṇvantu ṛṣayas sarve pañcadhā yat prakīrtitam |
laukikaṃ vaidikaṃ vāpi tathādhyātmikam eva ca ||
a[[timārgaṃ ca mantrākhyam]]—|
For a discussion of this incomplete passage, see Kafle 2015, 21; and pp. 49–53, for further evi-
dence of the influence of the Manusmṛti on the composition of the Niśvāsamukha.
8 Svacchanda 11.182–184:
atītaṃ buddhibhāvānām atimārgaṃ prakīrtitam |
lokātītaṃ tu taj jñānam atimārgam iti smṛtam ||182||
lokāś ca paśavaḥ proktāḥ sṛṣṭisaṃhāravartmani |
teṣām atītās te jñeyā ye ’timārge vyavasthitāḥ ||183||
kapālavratino ye ca tathā pāśupatāś ca ye |
sṛṣṭir na vidyate teṣām īśvare ca dhruve sthitāḥ ||184||
9 See Sanderson 2006 for references.
via free access
3 The Atimārga in the Vārāṇasīmāhātmya of the

Bhairavaprādurbhāva
For some years now I have been working on a manuscript containing a unique
collection of Vārāṇasīmāhātmyas.10 It concerns an old palm-leaf manuscript,
currently in the Kaiser library in Kathmandu (Acc. No. 66).11 The manuscript
may be dated on palaeographical grounds to the end of the twelfth century
CE. It was most probably penned down in Vārāṇasī itself, as suggested by
comparison of the old Nāgarī script with that of other manuscripts written
in twelfth-century Vārāṇasī, but now likewise surviving in the collections of
Nepal.12 It is an extensive but incomplete manuscript: 145 folios survive but
the text breaks off in the middle of a long quotation of the Skandapurāṇa.
The manuscript consists of Māhātmyas taken from and attributed to a range
of Purāṇas, including the Matsyapurāṇa, Nandipurāṇa, Brahmapurāṇa, Liṅga-
purāṇa, Śivapurāṇa and Skandapurāṇa. Several Māhātmyas can be identified
in the present editions of works bearing the same name (most importantly
the Māhātmyas of the published Matsyapurāṇa and the early Skandapurāṇa),
but quite a few of them are unknown from any other source. The manuscript
provides a unique glimpse into the production of Māhātmya literature in early-
medieval Vārāṇasī.
The first 13 chapters, covering the first 59 folios, about one third of the sur-
viving text, contain the complete text of a Māhātmya not known otherwise.
It is attributed in the colophons to the Bhairavaprādurbhāva of the Matsya-
purāṇa, but there are strong grounds to think that this attribution is incorrect
and that it was originally intended to belong to the Bhairavaprādurbhāva of
the Vāmanapurāṇa instead. My reasons for this assumption are the follow-
ing:
– The attributions of Māhātmyas to other Purāṇas are incorrect in several
other cases of the manuscript as well.
– There is no section called Bhairavaprādurbhāva in the surviving text of the
Matsyapurāṇa.
10 See Bisschop 2007 and 2013.

11 The manuscript has been microfilmed by the NGMPP on reel C 6/3. I also have access to
excellent colour photographs kindly provided to me by Harunaga Isaacson. There is also
a paper apograph in a private collection, microfilmed as NGMPP E 766/7.
12 Some comparable manuscripts produced in Vārāṇasī and dated to the twelfth century are:
Jayadrathayāmala (NGMPP A 996/3, A 997/1), Bṛhatkālottara (A 43/1), Harivaṃśa (A 27/5),
Manusmṛti (C 44/4). The scribe of our manuscript is, however, less neat. The text also
abounds in scribal errors.
via free access
20 bisschop
– Colophons in several manuscripts of the Vāmanapurāṇa, critically edited by

the Kashiraj Trust, reveal that various chapters from chapter 9 onwards are
attributed to the Bhairavaprādurbhāva of the Vāmanapurāṇa.13
– The narrators of the present Māhātmya and the Vāmanapurāṇa are the
same: Nārada and Pulastya.
– The “Origin of Bhairava(s)” (bhairavaprādurbhāva) is narrated in relation to
the destruction of Andhaka by Bhairava in Vāmanapurāṇa 44.30 ff.
– The eight Bhairavas (aṣṭabhairava) originating from the blood of Andhaka
when Śiva strikes him correspond with a set of eight Bhairavas mentioned
in our Māhātmya.14
I therefore conclude that the Māhātmya was originally meant to be included
in the Bhairavaprādurbhāva section of the Vāmanapurāṇa. In all likelihood
the text was composed by a local pandit in twelfth-century Vārāṇasī, for the
description of the town evinces clear connections to the layout of the town
around the time when the Gāhaḍavālas were ruling north India and had made
Vārāṇasī into their religious capital. It appears to be the work of a Śaiva author
who engages with the increasing presence of Vaiṣṇava worship in Vārāṇasī sup-
ported by the Gāhaḍavāla kings.15
The Atimārga is referred to towards the end of the first chapter of this Māhā-
tmya. After Pulastya has told Nārada about the abodes of the eight Bhairavas,
13 See also Bonazolli 1982.

14 Vāmanapurāṇa 44.23–38: Vidyārāja (east), Kālarāja (south), Kāmarāja (west), Somarāja
(north), Svacchandarāja, Lalitarāja and Vighnarāja. Bhairava himself should be added as
the eighth. In VM 1.53–54 seven similar Bhairavas are mentioned: Kālarāja (in Avimukta),
Kāmarāja, Saumya, Svacchanda (in Jayantika), Lalita, Vighnarāja (in Kālañjara), and Bhai-
rava (in Bhṛgutuṅga). The passage is corrupt and requires heavy emendation. I have ten-
tatively reconstructed the text as follows, but many readings remain doubtful:
viśiṣṭaṃ sarvasattvānāṃ tāraṇaṃ jagataḥ param |
kālarājasya ca kṣetraṃ dattaṃ caiva svayaṃbhuvā ||53||
kāmarājā tu saumyaś ca svacchandaś ca jayantike |
lalitaś ca kalau devi vighnarājā kaliñjare |
svayaṃ tu bhairavo devo bhṛgutuṅge vyavasthitaḥ ||54||
53a °sattvānāṃ] conj.; °sattānāṃ V1pcV2, °sattanāṃ V1ac (unmetr.) 53b jagataḥ] V1pcV2;
jagata V1ac (unmetr.) 54a kāmarājā tu] conj.; kālālayaṃ tu V1, kālālāyaṃ ca V2 54b
svacchandaś ca jayantike] conj.; svacchande ca jayantikā V1V2 54c lalitaś ca] conj.;
lalitasya V1V2 54d °rājā] conj.; °rājaṃ V1V2 ● kaliñjare] V1, kaliṃjale V2
15 Particularly relevant in this connection is chapter 7 of the Māhātmya. It deals with the
north end of the town, which was the centre of Gāhaḍavāla religious activity. This area,
referred to in the text as “Brahmapura,” is presented as an area of brahmanical author-
ity where gifts of gold, land, etc., are practised, as is indeed attested by the Gāhaḍavāla
inscriptions, and where Vedic recitation constantly takes place. See Bisschop 2011.
via free access
the most important of which is Kālarāja, who resides in Vārāṇasī and is also
known as Kālabhairava or Āmardaka,16 he introduces a number of important
tīrthas in the town. From VM 1.99 onwards Pulastya zooms in on the cremation
ground, the śmaśāna, also called ūṣara (saline ground), where, at the time of
destruction, all beings and worlds enter into Bhairava’s mouth. He tells Nārada
that it is because of this that the cremation ground grants release.17 He also
reports that there is a pond there called Kālodaka, which arose when Kālaru-
dra was playing on the cremation ground. At that time the Lord taught the
observance of the skull (kapālavrata). He chopped off the fifth head of Brahmā,
as a consequence of which there arose the holy Pāśupata pond.18 Mahādeva
subsequently carries Brahmā’s skull around the world and unites with the cre-
16 VM 1.37–38:
kālarājasya devena purī vārāṇasī śubhā |
nirdiṣṭā mokṣaṇārthāya pāpānāṃ caiva sarvadā ||37||
tasyāṃ caiva sthitaḥ sākṣād bhairavaḥ kālabhairavaḥ |
mardayan sarvapāpāni tena cāmardakaḥ smṛtaḥ ||38||
37c mokṣaṇārthāya] conj.; rakṣaṇārthāya V1V2

17 VM 1.99cd–101:
śmaśānasya śṛṇuṣvemāṃ kathāṃ pāpapraṇāśanīm ||99||
śmaśānaṃ ūṣaraṃ proktaṃ yatra khādyanti jantavaḥ |
lokāś caiva tu saṃhāre praviśya bhairavaṃ mukham ||100||
kālānalamahādīptaṃ kālarājasya bhairavam |
tena caiva śmaśānaṃ tu vārāṇasyāṃ tu mokṣadam ||101||
99c śmaśānasya] V1pcV2; śmaśāna V1ac (unmetr.) 100b khādyanti] conj.; kṣidyanti V1V2
100c lokāś caiva tu] V1pcV2; lokāś caiva tu lokāś caiva tu V1ac (unmetr.) 101b bhairavaṃ]
V2; bhaivaṃ V1 (unmetr.)
18 VM 1.102–105:
kālodakaṃ ca nirdiṣṭaṃ kālarājasya cāgrataḥ |
kālānalasamaprakhyaṃ tasminn eva mahat saraḥ |
saṃbhūtaṃ kālarudrasya śmaśāne krīḍataḥ purā ||102||
tataḥ krīḍānusaktena devadevena śambhunā |
kāpālavratam uddiṣṭaṃ tasmin kāle mahāmune ||103||
kāpālavratam āsthāya brahmaṇaś ca śiro mahat |
cakartta bhagavān kruddhaḥ pañcamaṃ ghoradarśanam ||104||
tasminn eva samutpannaṃ divyaṃ pāśupataṃ saraḥ |
tena tṛpyanti sakalāḥ kṣaṇamātraniyojitāḥ ||105||
103a krīḍānusaktena] em.; krīḍānuśaktena V1V2 103b devadevena] V1; devadena V2

(unmetr.) 104a kāpālavratam] V2; kāpālaṃ vratam V1 105b divyaṃ] V1pcV2; divyaṃ
divyaṃ V1ac (unmetr.) ● saraḥ] em.; śaraḥ V1V2 105d °niyojitāḥ] conj.; °nijojite V1pc,
°niyojitaṃ V1ac, °niyojite V2
via free access
22 bisschop
mation grounds.19 There he teaches that transmigration results from attach-

ment (sneha), that attachment comes from desire (rāga), and that attachment
is destroyed by indifference (vairāgya). When people attain indifference, their
karma is destroyed and they attain unequalled happiness (saukhya).20 He then
teaches the following:
The great tree of transmigration has arisen from the seed of desire. After
cutting the tree with the axe of indifference, whose sharp blade is disat-
tachment, they proceed on the Atimārga.21
Here the text introduces a term that we do not expect to come across in a
Māhātmya. The text continues in the same vein, however, attesting to the
appropriation of this significant terminology. The passage is worth quoting in
full:
Engaged in the path of the observance of the skull (Kāpālavratamārga),

the Lord wanders, free from attachment, displaying the Lokamārga and
the supreme Lokātīta. And the lokas are designated “bound souls,” includ-
19 VM 1.106:
sakapālaṃ mahādevo babhrāma sakalaṃ jagat |
krīḍamāṇas tadā kṣetrair uṣarair samapadyata ||106||
106a sakapālaṃ] conj.; taṃ kapālaṃ V1V2 106c tadā] V1; tada V2 106c kṣetrair] em.;
kṣetrer V1V2 106d samapadyata] V1ac; sampadyate V1pc (unmetr.), samupadyate V2
20 VM 1.112–113:
aho mūrkhā na jānanti saṃsāraṃ snehasaṃbhavam |
rāgāc ca jāyate sneho vairāgyāt snehasaṃkṣayaḥ ||112||
tad vairāgyaṃ yadā puṃsāṃ jāyate karmasaṃkṣayaḥ |
tadā saṃprāpyate saukhyam asādṛśyaṃ tu kasyacit ||113||
112b saṃsāraṃ snehasaṃbhavam] V1; saṃsārasnehasaṃbhavā V2 112d vairāgyāt]

V1pcV2; vairāgyā V1ac 113a vairāgyaṃ] V1; vairāgya V2 113b °saṃkṣayaḥ] conj.; °saṃ-
kṣaye V1V2 113c saukhyam] conj.; mokhyaṃ V1, mokṣam V2
21 VM 1.114:
rāgabījasamutpannaḥ saṃsāraviṭapo mahān |
vairāgyasya kuṭhāreṇa niḥsaṅgatīkṣṇa-m-aśmanā |
chittvā caiva tu taṃ vṛkṣam atimārgaṃ vrajanti te ||114||
114d °tīkṣṇa°] V1pcV2; °tikṣṇa° V1ac

The reading of pāda d is highly uncertain: it shows syncopation throughout and a mean-
ingless compound-breaking -m- is inserted. As Harunaga Isaacson has pointed out to me,
the image calls to mind Bhagavadgītā 15.3d (= Mahābhārata 6.37.3d): asaṃgaśastreṇa dṛ-
ḍhena chittvā.
via free access
ing gods, demons and men. No one realizes the supreme certainty with
respect to knowledge of the self. And except for Śarva, the supreme god,
there is no such behaviour of another [god]. No other god has certainty
of knowledge. There is no such behaviour anywhere in the world with
all its gods. The gods, beginning with Brahmā, also proceed along the
Laukikamārga. The God of gods, Virūpākṣa, who is established in the
Lokottaramārga, proceeds beyond [the institutes of] sacrifice ( yajña),
giving (dāna) and asceticism (tapas). But those sages who are on that
path, delighting in the knowledge of the self, also proceed along the
Lokottaramārga, abandoning their bodies. And there is no rebirth in this
world (iha) for those Pāśupata sages who follow the observance of the
skull, they who abide by the Atimārga. For the practitioners of the Ati-
mārga there is only (kevalam) indifference. Those who have set out on
the Atimārga only (kevalā) delight in indifference. Those who die on the
saline ground (uṣara) go along that path, but of all saline grounds Vārāṇasī
is the best, O sage. And there is no sprouting for those who die there.
The body abandoned on the cremation ground merges in the Lord of
Time.22
22 VM 1.116–125:
kāpālavratamārgastho niḥsaṅgo bhramate prabhuḥ |
darśayan lokamārgaṃ tu lokātītaṃ paraṃ ca yat ||116||
lokāś ca paśavaḥ proktāḥ sadevāsuramānuṣāḥ |
na kaścit paramaṃ vetti ātmajñāne tu niścayam ||117||
ṛte śarvān mahādevān nānyaceṣṭāsti cedṛśī |
jñānasya niścayo nāsti anyadevasya kasyacit ||118||
na ceṣṭā īdṛśī loke vidyate sāmare kvacit |
laukikenāpi mārgeṇa yānti brahmādayaḥ surāḥ ||119||
devadevo virūpākṣo mārge lokottare sthitaḥ |
atītya vartate devo yajñadānatapāṃsi ca ||120||
tasmin mārge tu ye viprā ātmajñānānurañjitāḥ |
te ’pi yānti tanuṃ tyaktvā mārge lokottareṇa tu ||121||
ye ca pāśupatā viprāḥ kāpālavratadhāriṇaḥ |
na teṣām udbhavo ’stīha atimārgeṇa ye sthitāḥ ||122||
vairāgyaṃ kevalaṃ tatra atimārganiṣeviṇām |
atimārgaprayātānāṃ vairāgye kevalā ratiḥ ||123||
tena te yānti mārgeṇa uṣare tu mṛtā hi ye |
sarveṣāṃ uṣarāṇāṃ tu śreṣṭhā vārāṇasī mune ||124||
tasyāṃ caiva mṛtānāṃ ca praroho naiva vidyate |
kālarājalayaṃ yāti śmaśāne tūjjhitā tanuḥ ||125||
116d ca] V1pc; om. V1acV2 (unmetr.) 117d ātmajñāne] V1; ājanmajñāne V2 (unmetr.) 118b
nānya°] V1pcV2; nānyā V1ac ● cedṛśī] V1; cedṛśaṃ V2 119b sāmare] V1; sāmaraḥ V2
120b lokottare] V1; lokottaraḥ V2 120d °dānatapāṃsi] V1; °dānaṃ tapāṃsi V2 121b
via free access
24 bisschop
This passage is revealing in several respects. First of all, it is the only known
non-Mantramārga source that uses the term Atimārga. Moreover, it does so
in a proper Atimārga context: its message is strict asceticism and there is no
mention of the Mantramārga at all, neither here nor in the following chap-
ters of the Māhātmya. As for the identity of the Atimārga in question, the text
brings together a number of key terms under one umbrella: Kāpālavratamārga,
Lokātīta, Pāśupata, and Atimārga. In appearance it is a form of Kāpālika asceti-
cism, involving cultivation of indifference (vairāgya) and aiming for death at
the cremation ground. This is an old theme in the Pāśupata tradition, harking
back to the Pāśupatasūtra itself.23
There are several hints in the above passage that indicate a relation with the
Svacchanda. The Svacchanda, as we have seen earlier, also refers to the division
of the Atimārga and, like our passage, stresses that it is Lokātīta (“Beyond the
Worlds”). The Vārāṇasīmāhātmya elaborates further on this theme, drawing a
clear distinction between the way of the Lord and that of the other gods. The
latter are merely followers of the Laukika path, or Lokamārga.
Moreover, we can identify two textual parallels:
1. lokāś ca paśavaḥ proktāḥ (Svacchanda 11.183a = VM 1.117a)
2. kapālavratino ye ca tathā pāśupatāś ca ye |
sṛṣṭir na vidyate teṣām (Svacchanda 11.184ac)
≈
ye ca pāśupatā viprāḥ kāpālavratadhāriṇaḥ |
na teṣām udbhavo ’stīha (VM 1.122ac)
Aside from these textual parallels it is noteworthy that some of the themes
that precede the verses on the Atimārga in the Svacchanda are taken up in
the passage of the Vārāṇasīmāhātmya. Thus, in Svacchanda 11.176 it is stated
that those who follow the hetuśāstra “science of reasoning” find no certainty
or conviction (niścaya) in matters of Dharma, Artha, Kāma or Mokṣa: dhar-
mārthakāmamokṣeṣu niścayo naiva jāyate (Svacchanda 11.176cd). The theme of
niścaya is taken up in the Vārāṇasīmāhātmya with respect to the “knowledge
of the self” (ātmajñāna), which is restricted to Śiva alone. No other god has
it.
°rañjitāḥ] V1; °rakṣitāḥ V2 121c yānti] V2; jānti V1 122a ca] V1; tu V2 123a vairāgyaṃ]
V1; vairāgya° V2 124c uṣarāṇāṃ] V1; uśvarāṇāṃ V2 124d śreṣṭḥā vārāṇasī mune] V1;
siddhā vārāṇasī ne V2 (unmetr.) 125d tūjjhitā] V1ac?; tyajatāṃ V1pc, tūktitā V2
The last sentence may be translated alternatively as: “he merges in the Lord of Time, but
his body is abandoned on the cremation ground.”
23 The Pāśupata ascetic was supposed to die on the cremation ground. See Pāśupatasūtra
5.30–40.
via free access
Moreover, the Svacchanda also introduces the concept of vairāgya, a key

term in the Vārāṇasīmāhātmya, just before its discussion of the Atimārga. The
Svacchanda distinguishes several religious traditions on the basis of their char-
acteristic mental dispositions (buddhibhāva):24
– Laukika is connected to dharma
– Pāñcarātra and Vaidika are connected to dharma and jñāna
– Bauddha and Ārhata are connected to vairāgya
– Sāṃkhya is connected to jñāna and vairāgya
– Yoga is connected to jñāna, vairāgya and aiśvarya
The Atimārga is said to be beyond these mental dispositions.25 Instead of this,
the Vārāṇasīmāhātmya passsage rather stresses the importance of vairāgya,
along with knowledge of the self (ātmajñāna), as the key to liberation. Al-
though we can thus trace some influence of the Svacchanda the teaching is
put in a new, proper Atimārga perspective.
The passage also involves some genuine Pāśupata ideals and concepts. The
statement that Śiva “proceeds beyond (atītya) [the institutes of] sacrifice
( yajña), giving (dāna) and asceticism (tapas),” instantly calls to mind Pāśupata-
sūtra 2.15–17, quoted above: atidattam atīṣṭam atitaptaṃ tapas tathā atyāgatiṃ
gamayate. Finally, the prominent presence of the word kevala in this passage,
which is employed to stress that for the follower of the Atimārga only vairāgya
(indifference) remains, can be connected to the technical use of Kevala in the
Pāśupata tradition. For the terms Kevalijñāna and Kevalārtha are used as syn-
onyms for the Pāśupata teaching in several sources.26
The Māhātmya continues to use the terminology of Atimārga and Loka-
mārga in the subsequent two chapters. In chapter 2 Śiva returns to Vārāṇasī
after he has wandered around with the skull for twelve years. He enters the
24 Svacchanda 11.179–181:
mohakāḥ sarvajantūnāṃ yatas te tāmasāḥ smṛtāḥ |
dharmeṇaikena deveśi baddhaṃ jñānaṃ hi laukikam ||179||
dharmajñānanibaddhaṃ tu pāñcarātraṃ ca vaidikam |
bauddhaṃ ārahataṃ caiva vairāgyeṇaiva suvrate ||180||
jñānavairāgyasambaddhaṃ sāṃkhyajñānaṃ hi pārvati |
jñānaṃ vairāgyam aiśvaryaṃ yogajñānapratiṣṭhitam ||181||
The list of mental dispositions calls to mind Sāṃkhyakārikā 23:
adhyavasāyo buddhir dharmo jñānaṃ virāga aiśvaryam |
sāttvikam etadrūpaṃ tāmasam asmād viparyastam ||
25 Svacchanda 11.182:
atītaṃ buddhibhāvānāṃ atimārgaṃ prakīrtitam |
lokātītaṃ tu taj jñānaṃ atimārgam iti smṛtam ||
26 See Sanderson * 2012, 9, n. 3.
via free access
26 bisschop
cremation ground and installs the skull to the north of Kālarāja. When he has
completed his observance, he bathes and satiates the sages, gods and ances-
tors. In this way, it is said, he displays the Lokamārga and sets an example for
the people to follow.27
The logic is illustrated by drawing on a phrase from the Bhagavadgītā:
Otherwise, if Deva would not display it, the path would be destroyed. The
people follow what he sets up as the standard.28
A tīrtha called Ṛṇamocana appears, provided with three liṅgas. The three
liṅgas release from the threefold debt to the gods, the sages and the ancestors.29
In this way the narrative integrates the teachings of the Lokātīta, or Atimārga,
and the Lokamārga. Śiva next continues to display more observances. He even
gives the Kāpālavrata to Kubera.30
27 VM 2.5–6:
evaṃ vratasamāptiṃ tu kṛtvā devo maheśvaraḥ |
snānaṃ kṛtvā tataḥ paścāt pareṇa vidhinā haraḥ ||5||
tarpayitvā ṛṣīn devān pitaraś ca yathāvidhi |
evaṃ ca lokamārgaṃ tu darśayāno jagatprabhuḥ ||6||
5d vidhinā] V1pcV2; vidhi V1ac (unmetr.) 6a ṛṣīn] V1; ṛṣin V2 6b pitaraś ca yathāvidhi]
V1; piraś ca yathāvidhiḥ V2 (unmetr.) 6d darśayāno] V1; darśayāmo V2
28 VM 2.7:
anyathā naśyate mārgo yadi devo na darśayet |
sa yat pramāṇaṃ kurute lokas tad anuvartate ||7||
7a naśyate] em.; tasya te V1V2 7c kurute] V1; kute V2 (unmetr.)

Cf. Bhagavadgītā 3.21 (= Mahābhārata 6.25.21):
yad yad ācarati śreṣṭḥas tat tad evetaro janaḥ |
sa yat pramāṇaṃ kurute lokas tad anuvartate ||
Also Kūrmapurāṇa 1.16.45:
evaṃ hi laukikaṃ mārgaṃ pradarśayati sa prabhuḥ |
sa yat pramāṇaṃ kurute lokas tad anuvartate ||
29 VM 2.8–9:
evaṃ jñātvā gato devas tarpayann ṛṇamocane |
tasmāt tatra mahātīrthaṃ saṃbhūtam ṛṇamocanam ||8||
trīṇi liṅgāni jātāni devadevasya tarpaṇe |
ekaṃ devamanuṣyāṇāṃ tṛtīyaṃ pitṛsaṃbhavam ||9||
8c tatra] V1pcV2; atra V1ac ● °tīrthaṃ] V1; °tīrtha V2 9a trīṇi] V2; triṇi V1 9c ekaṃ]
conj.; evaṃ V1V2 ● devamanuṣyāṇāṃ] V1; demanuṣyāṇāṃ V2 (unmetr.) 9d pitṛsaṃ-
bhavam] V1; tu trisaṃbhavaṃ V2
30 VM 2.20:
via free access
In chapter 3, the apparent contradiction in Śiva’s behaviour is articulated by

Brahmā, who says that those who have no faith see the path of the world, viz.
the Laukika and Apavargika, on the one hand, and the great path, the Lokottara,
on the other, as a contradiction:
ekataś ca jaganmārgaṃ laukikaṃ cāpavargikam |

lokottaraṃ mahāmārgaṃ viparītam aho tvayi |
dṛśyate bhuvanādhāra yatra śraddhā na vidyate ||6||
6a °mārgaṃ] V1ac; °mārge V1pc V2 6b cāpavargikam] V1; cāpavarggakaṃ V2 6c

mahāmārgaṃ] V1; mahāmārgga V2 6 f. vidyate] V1pc V2; khadyate V1ac
But in God the two paths are united. In him there is no contradiction. His ways
are inscrutable; only he can unite these contradictions and he does so because
he delights in play (krīḍā).31
4 Concluding Observations
I started this paper with the observation that the term Atimārga, although
certainly useful for referring to the ascetic strand of early Śaivism, appears
not to have been used by “Atimārgins” themselves. The term represents, by all
accounts, a Mantramārgic perspective on the formation of the Śaiva religion.32
The Vārāṇasīmāhātmya discussed in this paper is the only exception that I
know of in which we do find the term Atimārga—as well as the related term
Lokātīta—used outside of a Mantramārgic context. Several observations can
be made with reference to the passages of the Vārāṇasīmāhātmya discussed
here:
dhanadasya vrataṃ dattvā kāpālaṃ parameśvaraḥ |

so ’pi tatra vratī bhūtvā devam ārādhayan sthitaḥ ||20||
20a dhanadasya] V1; vanadasya V2

31 VM 3.7:
krīḍayā yāni deveśa karmāṇi kuruṣe prabho |
tāni lokeṣu dṛśyante paramārthapradāni tu ||
32 The same applies mutatis mutandis to terms such as Lokamārga or Laukika. Although
Mantramārga sources take the Laukika religion to refer to the merit-making rituals and
practices of the Śaiva laity taught in the Śivadharma, the term is not used by the Śiva-
dharma itself to refer to its own religious practice, which is rather referred to as “Śiva-
dharma.”
via free access
28 bisschop
– The VM provides important evidence on traditions relating to Bhairava, in

particular attesting to knowledge of the Svacchanda in Vārāṇasī around the
twelfth century.
– The VM integrates the divide between Lokamārga and Atimārga or Lokotta-
ramārga in a narrative of the origin of several sacred spots in Vārāṇasī and
the rituals to be performed there.
– The VM adopts the term Atimārga within a proper Atimārga context. The ide-
ology is clearly Atimārgic, representing a mixture of Pāśupata and Kāpālika
Śaivism.
– Finally, and most importantly, in appropriating the term, the Atimārga is pre-
sented here as the ultimate path of liberation.
It is striking that the author of the text does not seem to have felt the need
to contextualize the Mantramārga from which he has adopted the term in
the first place. Although it seems certain that the author had knowledge of
the Mantramārga—the influence of the Svacchandabhairava is quite clear—
he chose to neglect the Mantramārga entirely. Is this because he considered it
irrelevant to the content of the text, which is after all a Māhātmya of Vārāṇasī,
or because the Māhātmya represents a different perspective on what it means
to be a Śaiva? It certainly testifies to the fact that views on what constituted
Śaivism in early-medieval India differed across distinct Śaiva traditions. Much
of our understanding today derives from specific textual traditions that only
represent one layer of a much broader spectrum of religions oriented around
the worship of Śiva that sought to define themselves and claim their place.
5 Postscript
After the presentation of my paper at the symposium in Toronto, Professor

Sanderson kindly drew my attention to a veiled reference to the Atimārga
in the Halāyudhastotra. The text of this stotra is recorded, together with the
Mahimnastava and a Narmadāstotra, on an inscription of 1063AD (saṃvat
1120) in the Amareśvara temple at Oṃkāreśvar/Māndhātā.33 The inscription
33 For an overview of the inscriptions of the Amareśvara temple and the publication of
several more hitherto unpublished inscriptions from the temple, including the Narmadā-
stotra, see Neuss 2013 and 2015. The Halāyudhastotra was first published by P.P. Sub-
rahmanya Sastri, with an additional note containing the prose part of the inscription
by N.P. Chakravarti, in Epigraphia Indica 25 in 1939–1940 (appeared in 1948: Sastri 1948
and Chakravarti 1948). The text of the Halāyudhastotra was constituted on the basis of
the inscription and two manuscripts of the stotra from the Madras Government Oriental
Manuscripts Library. The inscription was subsequently republished by Mittal 1979, 322–
339, and Trivedi 1989, 604–611.
via free access
was written by a Pāśupata, paṃḍita Gāndhadhvaja, while the text itself is said
to have been composed by a brahmin named Halāyudha.34 The spiritual lin-
eage of the engraver of the inscription is recorded in a prose passage at the
end of the inscription and mentions the names of several Pāśupata teachers:
Bhāvavālmīka → Bhāvasamudra → Bhāvaviriṃci → Supūjitarāśi → Vivekarāśi →
Gāndhadhvaja.35
The Halāyudhastotra is an ornate poem in praise of Śiva. Phyllis Granoff
(1993) has introduced and translated several parts of the inscription in an arti-
cle on the experience of religion in medieval hymns and stories, but the verses
that concern us here have not been translated or discussed before. Verses 34–35
contain what appears to be an allusion to the Atimārga:
The visitation of the wives of the distinguished sages in the Pine Park, the
oblation with seed in Fire, the twilight dance: Your behaviour is not rep-
rehensible.36 O Three-eyed one! The doctrines of the world do not touch
those who have left worldly life, having passed far beyond the path of those
whose minds are afflicted by false knowledge.
The gods all wear gold and jewels as an ornament on their body. You
do not even wear gold the size of a berry on your ear or on your hand.
The one whose natural beauty, surpassing the path [of the world], flashes
on his own body, has no regard for the extraneous ornaments of ordinary
men.37
These verses poetically allude to the distinction between the Lokamārga and
Atimārga, although the terms themselves are not used. Significantly, the Halā-
yudhastotra also makes reference to the term Kevalajñāna, in the first pāda
34 Sastri (1948, 74) argues that he is identical with Halāyudha the tenth-century author of the
Abhidhānaratnamālā and the Kavirahasya.
35 Chakravarti 1948, 185, lines 51–55.
36 The verse refers to three key mythological events: Śiva’s visit as a naked ascetic to the
Devadāruvana, the emission of his seed into the mouth of Agni leading to the birth of
Skanda, and his performance of the twilight dance.
37 Halāyudhastotra 34–35:
dārūdyāne dvijavaravadhūpaplavo retasāgnau
homaḥ sandhyānaṭanam iti te ceṣṭitaṃ naiva duṣṭam |
[mithyājñānopa]hatamanasāṃ mārgam ullaṅghya dūraṃ
ye niṣkrāntās trinayana na tān lokavādāḥ spṛśanti ||34||
devāḥ sarve dadhati vapuṣā bhūṣaṇaṃ hemaratnaṃ
guñjāmātraṃ kanakam api te nāsti karṇe kare vā |
mārgātītaṃ sphurati sahajaṃ yasya saundaryam aṅge
tasyāhārye[ṣv itarajanavan nā]daraḥ syād guṇeṣu ||35||
via free access
30 bisschop
of verse 3: “Victorious is the One God, Śiva, the embodiment of the Kevala-
jñāna.”38 As mentioned above, Kevalajñāna may be used as a synonym for the
Pāśupata teaching. A Pāśupata background of the Halāyudhastotra is further-
more indicated by the prose part of the inscription, which records the names of
several Pāśupata teachers connected to the Amareśvara temple. The inscription
also makes reference to the liṅgas at five famous Pāśupata centres: Avimukta,
Kedāra, Oṃkāra, Amara (Amareśvara) and Mahākāla (Ujjain).39 All in all the
Amareśvara inscription merits further study as a testimony of the survival of
Pāśupata Śaivism in north India in the medieval period.
Abbreviations
conj. conjecture
em. emendation
NGMPP Nepal German Manuscript Preservation Project
unmetr. unmetrical
VM Vārāṇasīmāhātmya
References
Primary Sources
Kūrmapurāṇa. Anand Swarup Gupta, ed. Varanasi: All-India Kashiraj Trust, 1971.
Mahābhārata. V.S. Sukthankar et al., eds. The Mahābhārata. For the First Time Critically
Edited. 19 vols. Poona: Bhandarkar Oriental Research Institute, 1927–1959.
Pāśupatasūtra. R. Ananthakrishna Sastri, ed. Pasupata Sutras with Pancharthabhashya
of Kaundinya. Trivandrum Sanskrit Series, no. CXLIII. Trivandrum: The Oriental
Manuscripts Library of the University of Travancore, 1940.
Sāṃkhyakārikā. Har Dutt Sharma, ed., trans. The Sāṃkhyakārikā, Īśvara Kṛṣṇa’s Mem-
orable Verses on Sāṃkhya Philosophy with the Commentary of Gauḍapādācārya.
Poona: The Oriental Book Agency, 1933.
Svacchanda. Madhusūdan Kaul Shāstrī, ed. The Svacchanda-Tantra with commentary by
Kshema Rāja. KSTS, nos. 31, 38, 44, 48, 51, 53, and 56. Bombay: Nirnaya-Sagar Press,
1921–1935.
38 Halāyudhastotra 3a: eko devaḥ sa jayati śivaḥ kevalajñānamūrtir.

39 Line 50:
avimuktaś ca kedāra oṃkāraś cāmaras tathā |
pañcamaṃ tu mahākālaḥ paṃcaliṅgāḥ prakīrttaye ||
via free access
Vāmanapurāṇa. Anand Swarup Gupta, ed. Varanasi: All-India Kashiraj Trust, 1967.
Vārāṇasīmāhātmya. Peter Bisschop, ed. Draft edition of the Vāmanapurāṇa-Bhairava-
prādurbhāva-Vārāṇasīvarṇana on the basis of NGMPP C 6/3 (= V1) and its apograph
NGMPP E 766/7 (= V2)
Secondary Sources
Bisschop, Peter. 2007. “Notes on a Vārāṇasīmāhātmya Compendium.” Newsletter of the
NGMCP 5: 21–30.
Bisschop, Peter. 2011. Voer voor Filologen: Geüpgradede anonieme Sanskrit literatuur.
Oratie uitgesproken bij de aanvaarding van het ambt van hoogleraar op het gebied
van Sanskrit and Ancient Cultures of South Asia aan de Universiteit Leiden op
vrijdag 29 april 2011. Leiden: Leiden University. https://openaccess.leidenuniv.nl/
handle/1887/19582
Bisschop, Peter. 2015. “The Abode of the Pañcamudrās: A Yoginī Temple in Early Medie-
val Vārāṇasī.” In ‘Yoginī’ in South Asia. Interdisciplinary approaches, edited by István
Keul, 47–60. London and New York: Routledge.
Bonazolli, Giorgio. 1982. “The Colophons in the Critically Edited Purāṇa-s.” Purāṇa 24
(2): 353–383.
Chakravarti, N.P. 1948. “A Note on the Halayudha Stotra in the Amaresvara Temple.”
Epigraphia Indica 25 (1939–1940): 183–185.
Flood, Gavin. 1996. An Introduction to Hinduism. Cambridge: Cambridge University
Press.
Granoff, Phyllis. 1993. “Halāyudha’s Prism: The Experience of Religion in Medieval
Hymns and Stories.” In Gods, Guardians and Lovers: Temple Sculptures from North
India, A.D.700–1200, edited by Vishaka N. Desai and Darielle Mason, 66–93. New
York: the Asia Society Galleries.
Goudriaan, Teun, and Sanjukta Gupta. 1981. Hindu Tantric and Śākta Literature. A His-
tory of Indian Literature II.2. Wiesbaden: Otto Harrassowitz.
Hara, Minoru. 1966. Materials for the Study of Pāśupata Śaivism. PhD thesis, Harvard
University.
Hara, Minoru. 2002. Pāśupata Studies. Edited by J. Takashima. Publications of the De
Nobili Research Library XXX. Vienna: Institut für Südasien-, Tibet-, und Buddhis-
muskunde.
Johnson, W.J. 2009. A Dictionary of Hinduism. Oxford: Oxford University Press.
Kafle, Nirajan. 2015. The Niśvāsamukha, the Introductory Book of the Niśvāsatattvasaṃ-
hitā. Critical Edition, with an Introduction and Annotated Translation, Appended by
Śivadharmasaṃgraha 5–9. Ph.D. thesis, Leiden University.
Mittal, A.C. 1979. The Inscriptions of the Imperial Paramāras (800A.D. to 1320A.D.).
Ahmedabad: L.D. Institute of Indology.
Neuss, Jürgen. 2013. “Oṃkāreśvar-Māndhātā. Tracing the Forgotten History of a Popular
Place.” Berliner Indologische Studien 21: 115–172.
via free access
32 bisschop
Neuss, Jürgen. 2015. “Unpublished Inscriptions from the Amareśvara Temple, Mānd-
hātā.” Berliner Indologische Studien 22: 123–150.
Olivelle, Patrick. 2005. Manu’s Code of Law. A Critical Edition and Translation of the
Mānava-Dharmaśāstra. Oxford: Oxford University Press.
Sanderson, Alexis. 1988. “Śaivism and the Tantric Traditions.” In The World’s Religions,
edited by S. Sutherland et al., 660–704. London: Routledge.
Sanderson, Alexis. 2006. “The Lākulas: New Evidence of a System Intermediate Be-
tween Pāñcārthika Pāśupatism and Āgamic Śaivism.” Indian Philosophical Annual
26 (2003–2005): 143–217.
Sanderson, Alexis. *2012. “Śaivism and Brahmanism. All Souls Hilary Term 2012, Hand-
out 3, 8 February.” Unpublished handout made available by the author at https://
www.academia.edu/6388859
Sastri, P.P. Subrahmanya. 1948. “Halayudhastotra from the Amaresvara Temple.” Epi-
graphia Indica 25 (1939–1940): 173–182.
Trivedi, Harihar Vitthal. 1989. Inscriptions of the Paramāras, Chandēllas, Kachchha-
paghātas and Two Minor Dynasties. Corpus Inscriptionum Indicarum VII.3. New
Delhi: Archaeological Survey of India.
via free access
chapter 2
Why Are the Skull-Bearers (Kāpālikas) Called

Soma?
Judit Törzsök
1 The Question
The Kāpālikas or Skull-Bearers, who formed the third group of the Atimārga,
alongside the Pāśupatas and the Lākulas,1 were perhaps the most notorious
Śaiva ascetics of classical India. They were known for their cremation ground
rituals and for wandering around with a skull for an alms bowl. The skull
(kapāla), their most conspicuous attribute, also provided their name. But the
Kāpālikas are also designated as Somasiddhāntins, “Those of the Soma Doc-
trine,”2 or the “Soma People with the Skull.”3 These appellations seem to have
been of some importance because their initiation names also included or
ended in -soma in most cases (e.g. Satyasoma, Devasomā,4 Somibhaṭṭāraka5).
What was this Somasiddhānta, doctrine of Soma or teaching about (the) Soma?
1 Concerning these distinctions within the Atimārga, the term Atimārga itself and the Kāpā-
likas as an Atimārgic group, see Sanderson 1988 and 2006. Professor Sanderson has published
several ground-breaking papers focussing on the Kāpālikas. This paper, inspired by his discov-
eries, is dedicated to him.
2 This term figures in the Pauṣkaravṛtti of Jñānaprakāśācārya, IFP transcript no. 110, p. 591. They
are also called “Knowers of the Doctrine of Soma” (somasiddhāntavedinaḥ in Sarvajñānottara
14.4, edited by Goodall), which could be corrupt for somasiddhāntavādinaḥ, “Those who Pro-
fess the Doctrine of Soma.”
3 Or “Skull-Bearers Who are the Soma People,” somajanakāpālī in Jayadrathayāmala 3.35.33c.
There may be an attempt here to distinguish the skull-bearing Soma ascetics from other skull-
bearers, such as those who follow a Bhairava tantra or a Kaula tantra.
4 The names of the two Kāpālikas in the Mattavilāsaprahasana.
5 The name or title of a Kāpālika in the Kannada inscription of ancient Koḷḷipāke, Andhra,
in 1050 CE, cited by Lorenzen 1989, 233–234. This Kāpālika is said to be mukha-kamala-
vinirggata-Sōmasiddhāntābhiprāya- “devoted to the meaning of Somasiddhānta issued from
the lotus mouth [?of Śiva]” (Lorenzen’s translation). While this implies that the Somasi-
ddhānta or Soma teaching was ultimately considered Śaiva revelation (if we accept Loren-
zen’s suggestion of supplying Śiva), it does not tell us anything about its nature and content,
nor about the meaning of the word soma itself. The wording suggests, nevertheless, that it is
not Śiva who is called Soma.
© judit törzsök, 2020 | doi:10.1163/9789004432802_004

via free access
34 törzsök
In what way was it typical of Kāpālikas? Why did -soma figure in their initi-
ation names?
I am afraid I will not be able to offer conclusive answers to most of these
puzzling questions. However, I propose to look at a few passages about the
Kāpālikas which may shed more light on what the word or name Soma pos-
sibly meant for them.
Now I am not the first to ask this question. An ingenious answer can already
be found in commentaries on the Prabodhacandrodaya of Kṛṣṇamiśra (itself
dating from 1041–1073): commentators understand soma to mean sa-umā, i.e.
“with/accompanied by Umā,” with reference to the fact that a male Kāpālika
normally had a consort, just as Śiva is accompanied by Umā.6
This understanding seems rather forced. Female Kāpālikas or tantric con-
sorts are not normally called Umā and this interpretation does not seem to
figure at all in earlier sources. It also fails to explain how we are to understand
the element -soma in female initiation names (such as Devasomā), in which
it cannot mean “with Umā/with a female consort.” Nevertheless, the sa-umā
explanation of soma higlights an important trait of the Kāpālikas, namely that
they were exceptional in the Atimārga in that male and female initiates per-
formed rituals together7 and were obviously not required to maintain celibacy,
unlike (most probably) the ascetics of the Pāśupata and Lākula groups.
David Lorenzen has proposed a different hypothesis.8 He identified a Kāpā-
lika called Kāpāli-śarman in a (probably) sixth century inscription from Kar-
nataka. This Kāpāliśarman is said to have performed vedic Soma sacrifices.
Lorenzen therefore suggests that Kāpālikas were perhaps dedicated vedic Soma
sacrificers.
This is also rather unlikely, for at least two reasons. First, Kāpāliśarman may
not have been a Kāpālika in the strict sense, for his name does not include Soma
and does not appear to conform to other kinds of Kāpālika names either.9 Sec-
ond, nowhere else is it said that Kāpālikas performed vedic Soma sacrifices.
However, as we shall see they took interest in particular ‘essences’ other than
the vedic Soma, and in a metaphorical sense, perhaps did perform their own
kind of Soma ritual.
6 For this and some other references to Kāpālikas associated with some Soma doctrine, see
Lorenzen 1991, 83.
7 Two well-known literary examples are the Kāpālika man with his partner in the Matta-
vilāsaprahasana and the Kāpālika couple in the fifth act of Bhavabhūti’s Mālatīmādhava.
8 Lorenzen 1989, 235 citing K.V. Ramesh Inscriptions of the Western Gangas 70–74, no. 19, inscrip-
tion from Bangalore district, 6th cent.? (sic!).
9 Since Kapālin/Kāpālika can denote Śiva/Bhairava himself, the name could simply mean “Pro-
tected by Śiva/Bhairava” without being markedly Kāpālika.
via free access
why are the skull-bearers (kāpālikas) called soma? 35
2 Somaśarman and the Moon Image
We reach firmer ground when we turn to the often-cited Malhar or Junwani

copper plate inscription (647CE, see Bakker 2000 and 2015; Sanderson 2012),
which lists a lineage of Kāpālikas as identified by Sanderson. This mentions
Somaśarman and the “line of tradition starting with Soma” (continuing later
with Rudrasoma, Tejasoma, Bhīmasoma). It is in Somaśarman’s house that
Lakulīśa, founder of the Pāśupata order, is said to have been born as an incar-
nation (avatāra) of Śiva. Lakulīśa was then initiated into or through the mahā-
vrata, perhaps by Somaśarman himself:
[…] adhunā kali-kālam āsādya śrīmal-Lakulīśa-nātho ’vatīrya Somaśarmā-

khya-brāhmaṇa-kule jātaḥ mahāvrate[ te?]na dīkṣito jagad-indus tenāpi
Musalīśas tataḥ Somādi-pāramparya-krameṇa sthānaguru-śrī-Rudraso-
ma-praśiṣya-śrī-Tejasoma-śiṣyebhyaḥ śrīmad-Bhīmasoma-pādebhyaḥ […]
mahāvrate[ te]na ] conj. Isaacson; mahāvratena

Musalīśas ] conj. Sanderson; mugalīśas
sthānaguru ] conj. Majumdar; sthāne guru
[…] reaching the present Kali age, the venerable Lord Lakulīśa took up an
incarnation and was born in the family of a brahmin called Somaśarman.
He was initiated into the Great Observance by him (?) [and became] the
Moon of the World. Then by him, Musalīśa [was initiated], then, by the
unbroken tradition starting with Soma, the local Master Rudrasoma, his
disciple Tejasoma, whose pupil is the venerable Bhīmasoma […]
Before examining the question of Soma and related issues, I would like to
point out some details concerning the word mahāvrata or Great Observance.
Lakulīśa and others were most probably initiated into the mahāvrata (mahā-
vrate) and not with/by the mahāvrata (mahāvratena), for this observance is
not known to be used as a rite of initiation in any Śaiva system.10
By the beginning of the seventh century, the mahāvrata certainly referred
to the ritualized mimesis of Śiva’s expiation for cutting off Brahmā’s fifth head.
10 This follows the conjecture proposed by Isaacson, although the reading of mahāvratena
is of course grammatically acceptable. However, it is also possible that the instrumental
mahāvratena was understood to stand for the locative, and that no additional instrumen-
tal tena was intended. In this case, it is not expressed that Somaśarman initiates Lakulīśa,
although it may again be implied. Bakker (2015, 143) opts for the instrumental, but assumes
that this only implies a Pāśupata affiliation of Lakulīśa.
via free access
36 törzsök
According to this well-known story, Śiva must wander with a skull he uses as
an alms bowl, for he has committed the sin of killing a brahmin, i.e. Brahmā.
Wandering with a skull for twelve years is in fact the expiatory observance for
killing a brahmin as prescribed in the Dharmasūtras,11 but there it is not yet
called mahāvrata. Although most attestations of the mythological story come
from late puranic sources, the myth already figures in the (original) Skandapu-
rāṇa (chapters 5–7),12 dated around the end of the sixth and beginning of the
seventh century.
Moreover, the Kāpālika Satyasoma in the Mattavilāsaprahasana (600–
625CE) mentions that it was thanks to the practice of the mahāvrata that his
Lord bearing the crescent moon on his head was purified of his sin, which he
had committed by cutting Brahmā’s head.13 The verse clearly identifies Śiva’s
mahāvrata as the expiation rite for a brahmin slayer, and also shows that Kāpā-
likas were practising the mahāvrata in imitation of Śiva. In fact, the Skandapu-
rāṇa (6.5–6) also seems to associate this observance with sanguinary practices,
such as those of the Kāpālikas. For when Śiva-Nīlalohita starts looking for suit-
able alms, Viṣṇu tries to fill his kapāla-bowl with his own blood—a very odd,
distinctively Kāpālika notion of what alms should consist of.
Now returning to the question of the Soma lineage: Lakulīśa, whether he was
indeed initiated by Somaśarman or not, is said to have been born in Somaśar-
11 See e.g. Baudhāyanadharmasūtra 2.1.2–3: bhrūṇahā dvādaśa samāḥ kapālī khaṭvāṅgī ga-
rdabhacarmavāsā araṇyaniketanaḥ śmaśāne dhvajaṃ śavaśiraḥ kṛtvā kuṭīṃ kārayet | tām
āvaset | saptāgārāṇi bhaikṣaṃ caran svakarmācakṣāṇas tena prāṇān dhārayet |. “A man
who has killed a learned Brahmin should do the following for twelve years. He should
carry a skull and a post from a bed-frame; wear the skin of an ass; reside in the wilder-
ness; and, using the head of a corpse as his flag, get a hut built in a cemetery and live
in it. He should maintain himself by begging almsfood from seven houses while pro-
claiming his crime.” Translation by Olivelle (2000, 241). See Gautamadharmasūtra 22.4:
khaṭvāṅgakapālapāṇir vā dvādaśa saṃvatsarān brahmacārī bhaikṣāya grāmaṃ praviśet |
karmācakṣāṇaḥ. “Or else, for twelve years he should live a chaste life and, carrying the
post from a bed-frame and a skull, enter a village only to beg for food while proclaiming
his crime.” Translation by Olivelle (2000, 175), who remarks in the notes to this passage
that khaṭvāṅga must mean skull-staff (a staff topped with a skull) rather than the post of
a bed-frame. Brick 2012, however, shows convincingly that the post of a bed-frame was
probably intended at the time of the Dharmasūtras. See also Yājñavalkyasmṛti 3.243: śi-
raḥkapālī dhvajavān bhikṣāśī karma vedayan | brahmahā dvādaśābdāni mitabhuk śuddhim
āpnuyāt ||.
12 The version related here replaces the figure of Śiva with one of his ectypes, Nīlalohita. On
this narrative device, which is used here to distance the supreme deity from sanguinary
practices, see Granoff 2006.
13 17ab: āsthāya prayato mahāvratam idaṃ bālenducūḍāmaṇiḥ, svāmī no mumuce pitāma-
haśiraśchedodbhavād enasaḥ |.
via free access
man’s house. According to the inscription, the Kāpālikas belong to the spiritual
lineage starting with Soma, and their initiation names therefore seem to be
derived from the founder’s name. The name Soma can naturally be understood
as a short form of Somaśarman. Thus, Kāpālikas are the Soma people because
they follow the tradition started by Somaśarman.
Our investigation could stop here. For the name Soma seems to be suffi-
ciently explained in this way. However, several issues remain unexplained. It
is not clear whether Somaśarman was a historical person. If he did exist, it still
remains uncertain whether he was indeed the founder of the Kāpālika move-
ment or whether Kāpālikas claimed retrospectively that he was their founder.
Thus, we cannot take it for granted that the Soma name indeed derives from
him.
For this reason, I suggest we look at some other details more closely. Lakulīśa,
after his initiation in Somaśarman’s house, is called the Moon of the Word
( jagadindu) in the inscription. There are at least three interpretations of jaga-
dindu:
1. A natural understanding of the moon as having cool rays. Thus, “Moon
[whose cooling rays have calmed the fever] of the world” (translation by
Sanderson 2012).
2. Moon on the earth, i.e. having a lunar-white body (sitāṅga) on account
of the bathing with ashes (Bakker 2000; 2015, 153). This understanding is
backed up by the description of Lakulīśa in the Skandapurāṇa as being
white-bodied when covered with ashes.14
3. Without going against either of these interpretations, both of which are
plausible, I propose as a tentative hypothesis that, in addition, the expres-
sion jagadindu may indirectly allude to a Kāpālika affiliation if we under-
stand this in the sense of *Jagatsoma, a compound suggestive of a Kāpā-
lika initiation name.15 This may imply two things. First, it is possible that
14 It must also be noted that the Skandapurāṇa (180.10) calls the mere ash-bath a/the great
observance (mahāvrata). It also says that Somaśarman with his family received Lakulīśa’s
grace when he visited them in their house, and that they were given yogasiddhi (167.125ff.).
The Skandapurāṇa appears to represent an earlier(?)/pāśupata version of the story. (Cf.
also Bakker 2015, 143 ff.) Bakker 2015, 143–144 (note 442) also proposes that the Soma name
suggests a parallel with the Soma-vaṃśa dynastic affiliation of Mahāśivagupta. However,
the Soma name figures elsewhere, in seventh-century South India in the names of Kāpā-
likas of the Mattavilāsaprahasana, where no such parallel can be assumed; such implica-
tions therefore seem unlikely.
15 I cannot cite any initiate with this name, and I do not intend to suggest that Jagatsoma (or
Jagadindu) was necessarily an actual initiation name. The word may, nevertheless, be con-
strued in the same manner as Kāpālika initiation names such as Satyasoma, Devasoma,
via free access
38 törzsök
the Kāpālikas derive their name Soma/Moon from Lakulīśa, too, who is
considered the Moon of the World. Second, the sequence of events as
presented in the inscription may also suggest that Lakulīśa himself came
to be called the “Moon of the World” (suggestive of a Kāpālika initiation
name) because of Somaśarman. Whatever is the case, the lunar image is
emphatically present in the names or epithets of both alleged founders
as well as in the initiation names of Kāpālikas.
Now the naming of Lakulīśa as the “Moon,” in the manner of a Kāpālika, may be
more than a coincidence. It may well be understood as an attempt to present
Lakulīśa as a true Kāpālika, perhaps via his association with another “Moon
person,” Somaśarman. Or, from another point of view, by presenting Lakulīśa
as a Kāpālika initiate, the text may suggest the preeminence of the Kāpālikas
over the other two Atimārga groups, the Pāśupatas and the Lākulas.
But no matter how we understand the hidden agenda of the above inscrip-
tion (if there is one), it is undeniable that the Kāpālikas’ initiation name ending
in -soma is understood to recall both their founder(s)’ name and the image of
the moon.
3 The Moon and the Nectar of Immortality (amṛta) in the Skull(s)
The moon also forms an important element of Pāśupata yogic practices. As we

learn from the Skandapurāṇa (179.28ff.), as pointed out by Bakker (2015, 141),
their “accomplishment in yoga” ( yogasiddhi) comes about through a process
of withdrawing the senses until the practitioner can see a lunar disc (soma-
maṇḍala) in his heart. From the moonlight within his body, yogic powers,
omniscience and the like arise. These powers include being immune to disease
(vyādhayo nāviśanty enam) and possessing a divine body (divyaṃ vapuḥ).
Speaking of the moon and practices related to its visualization, the Kāpālikas
appear to share the pan-Indian idea that it also contains the nectar of immor-
tality.16 In one passage of Bhavabhūti’s Mālatīmādhava (5.23) an invocation is
addressed to the fierce goddess, Cāmuṇḍā, which describes her violent tāṇḍava
dance. The verse is uttered by the two Kāpālikas in the cremation ground. Dur-
ing this dance, the goddess inadvertently slashes the moon, from which the
etc., which can be interpreted as “Moon of Truth,” “Moon of the Gods,” etc. What I pro-
pose is that this parallelism of names ending with “-moon” seems too remarkable to be
accidental.
16 This idea perhaps also contributed to the spread of various visualization practices cen-
tered around the image of the moon.
via free access
amṛta flows downward and fills her garland of skulls. The skulls, thus resur-
rected, start emitting a loud and harsh laughter.17
This image is not particularly significant in itself. However, it seems that
Kāpālikas were particularily interested in a special sort of ambrosia. In their
quest for the amṛta, they probably joined a large range of ascetics or yogins of
the period who, in various ways and through different practices, all sought the
same magical essence.18 So what exactly was the amṛta of the Kāpālikas and
how did they expect to find or produce it?
4 What is the Kāpālikas’ Nectar (amṛta/soma)?
We now turn again to the Kāpālikas of the Mālatīmādhava.19 At one point in

the story (5.2), the female practitioner mentions that she can extract the so-
called “five nectars” (pañcāmṛta), which are five vital essences of the human
body. They have a powerful, invigorating effect (as one would expect from such
a nectar), so much so that the female Kāpālika can fly a great distance in a few
seconds.
17 pracalita-kari-kṛtti-paryanta-cañcan-nakha-āghāta-bhinnendu-niḥsyandamānāmṛta-
ścyota-jīvat-kapālāvalī-mukta-caṇḍāṭṭahāsa-trasad-bhūri-bhūta-pravṛtta-stuti śvasad-
asita-bhujaṃga-bhogāṅgada-granthi-niṣpīḍana-sphāra-phullat-phaṇā-pīṭha-niryad-viṣa-
jyotir-ujjṛmbhaṇoḍḍāmara-vyasta-vistāri-doḥ-khaṇḍa-paryāsita-kṣmādharam | jvalad-
anala-piśaṅga-netra-cchaṭācchanna-bhīmottamāṅga-bhrami-prastutālāta-cakra-kriyā-
syūta-dig-bhāgam uttuṅga-khaṭvāṅga-koṭi-dhvajoddhūti-vikṣipta-tārā-gaṇaṃ pramudita-
kaṭapūtanottāla-vetāla-tāla-sphuṭat-karṇa-saṃbhrānta-Gaurī-ghanāśleṣa-hṛṣyan-ma-
nas-Tryambakānandi vaḥ tāṇḍavam devi bhūyād abhīṣṭyai ca hṛṣṭyai ca naḥ |.
18 Obtaining the nectar of immortality and, thanks to it, an immortal physical body is the
main goal of the haṭhayogic and Nath yogic traditions; see Mallinson 2007 and Ondračka
2007. Mallinson (2015, 120 ff.) proposes that there may have been an early, nonsectarian
tradition of ascetics, the precursor of what is later known as haṭhayoga, for which he finds
traces already in the Pali Canon.
19 I understand, in the context of this paper, the Kāpālikas of this play to represent the
Kāpālikas of the Atimārga here under discussion. They could alternatively be considered
skull-bearing tantric practitioners, as Hatley (2007, 143ff.) argues on the basis of numerous
parallels with prescriptions found in the Brahmayāṃala and elsewhere. In fact, one could
interpret the evidence in two ways: either take the Kāpālikas of the play to belong to the
Atimārga, which has strong influence on later Śaiva tantras of the Vidyāpīṭḥa, or one can
take them to be tantric skull-bearers of the Vidyāpīṭha, who certainly inherit much from
the Kāpālikas of the Atimārga. In either case, the practices described in the play may well
reflect what is a ritual core common to both the Kāpālikas of the Atimārga and those of
the Vidyāpīṭha. For this reason, in the context of the present argument, I understand the
Mālatīmādhava’s Kāpālikas to be representative of Atimārga Kāpālikas or their practices,
even if this identity remains uncertain.
via free access
40 törzsök
The extraction of the five nectars (pañcāmṛtākarṣaṇa), as well as other,

Kāpālika-type cremation ground practices, also figure in the Brahmayāmala,
as Hatley (2007, 143ff.) points out. The five substances are not listed in a
systematic way, but they usually seem to include these four: semen (śukra),
blood (rakta), fat/marrow (medas) and sneha (see also the entry pañcām-
ṛta in Tāntrikābhidhānakośa, vol. III). Other sources also describe Kāpālikas
as making use of various parts of the human body. Kāpālikas use human
flesh (mahāmāṃsa), brain (mastiṣka), intestines (antra), fat (vasā) and blood
(kīlāla) in ritual, and drink alcohol (surā), according to Prabodhacandrodaya
3.13.
In addition to the extraction of the five nectars, the Brahmayāmala also
includes rituals which make particular use of human body parts and are to be
performed in the cremation ground. A notable series of chapters prescribing
such rites forms a small cluster around chapter 46. Since the practices pre-
scribed have close affinity to the kinds of ritual attributed to Kāpālikas, these
chapters could well be adaptations or assimilations of originally Kāpālika ritu-
als,20 although this remains a hypothesis in the absence of any surviving Kāpā-
lika scriptures. It is also notable that the Brahmayāmala describes possession
(āveśa) by Bhairava, stating that through possession one obtains Bhairavahood;
and possession was, according to numerous Śaiva sources about the subject, the
way in which Kāpālikas claimed to attain final liberation.21
Now in chapter 46 of the Brahmayāmala, much like the Kāpālikas, the prac-
titioner makes ritual use of human flesh, hair (keśa), bones (asthi), body fluids
(picu), particularly blood (rakta), and intestines (antra); moreover, he offers
and drinks alcohol (madirā).
It is in this chapter that the amṛta comes to fore again in ritual. For the
main subject here is amṛtamanthāna, the churning and drinking of the amṛta.
The Sādhaka is to make pots from clay obtained from the cremation ground,
contruct a sacrificial pavilion from bones, and place the ritual cauldron upon
a corpse. He is required to churn a mixture including mahāpicu (sexual flu-
ids or various other human fluids?) using a piece of bone as the churning
stick, with a rope made of human hair, intestines and skin (?). The cauldron
is identified with Aghorī and the churning stick with Bhairava. In this rite, the
Sādhaka re-enacts the cosmic churning of the ocean, and the same miracu-
lous objects emerge (the Kaustubha gem, etc.) as the gods brought forth dur-
ing the primordial churning. Then the terrifying goddess, Caṇḍikā, receives
20 For more arguments, see Törzsök 2011 and 2015.

21 On this, see e.g. Sanderson 2009, 133 note 311.
via free access
homage. She appears in the form of Aghorī, offering the Sādhaka a boon.
The practitioner chooses to drink the milk of Aghorī’s breasts. The chapter
ends by saying that having produced the amṛta and having drunk left and
right (probably meaning having drunk Aghorī’s milk from both breasts), one
becomes omniscient, Bhairava himself. Although Aghorī’s milk and the amṛta
that the Sādhaka prepares are not identified, they could well be the same
thing.22
22 The following working edition is based on Shaman Hatley’s transcription of the old palm-
leaf manuscript. Only the relevant verses are given here:
mahāpicususaṃpūrṇaṃ kuryāt sthāliṃ śavopari ||44||
āyatasya tu nalakaṃ mahā-m-asthi śavopari |
eṣa manthānako devi asmin tantre praśasyate ||45||
keśeṣu netrakaṃ kuryād antraiḥ karttṛvimiśritaiḥ | (karttṛ in the sense of kṛtti?)
navahastaṃ susaṃpūrṇaṃ vidyāmālāniyojitam ||46||
suviśuddhamahībhāgaṃ rajasaṃpātaśobhitam |
mahāsthālī tu pārśve tu evaṃ kṛtvā mahātape ||47||
Aghoryā sthālirūpāṃ tu dhyāyen mantrī suśobhanām |
śaktisthāṃ śaktirūpāṃ ca dhyāye somātmake sthitām ||48||
Manthānabhairavaṃ devaṃ śuddhasphaṭikanirmalam |
sahasrabhujaparyantaṃ cinten manthānarūpiṇam ||49||
…
mahāmanthāna kurvīta yaṃ sthitvā tu Śivo bhavet ||61||
…
namaskṛtvāsurīṃ divyāṃ tataḥ sādhanam ārabhet ||62||
…
evaṃ mālais tu tāṃ dīptāṃ dhyātvā manthānamandiram |
netrakaṃ ca tathaiveha cintayed Vāsukirūpiṇam ||66||
kṣīrodaṃ sthāpayet sthāli ātmā bhairavarūpiṇam |
pūjayitvā tu manthānaṃ prakṣipet sthālimadhyataḥ ||67||
…
kṣaṇamātraṃ mathed yāvac chaśāṅkottiṣṭhate priye ||92||
Kaustubhaṃ ca tato tiṣṭhe vimānaṃ Puṣpakaṃ tathā |
evam ādyāni siddhīni pūrvaśāstreṇa bhāṣitāṃ ||93||
uttiṣṭhati mahābhāge śataśo [’]tha sahaśraśaḥ | (mahābhāgo Ms.)
…
kṣaṇamātraṃ mathed yāva namaskṛtvā tu Caṇḍikāṃ |
tatrottiṣṭhati vai devi Aghorī siddhidāyikā ||107||
…
sādhakovāca ||
yadi tuṣṭāsi māṃ devi stanaṃ me dada Ambike ||114||
śrutvā vākyaṃ tato devyāṃ sādhakasya suśobhanaṃ |
ehi ehi mahāsattva stana me piba putraka{ḥ} ||115||
tvaṃ muktvā tu mahāsattva{ḥ} ko [’]nyo putratvam arhati |
pariṣvajya tato vīraṃ stanaṃ dadāmi sādhaka{ḥ} ||116||
…
via free access
42 törzsök
What emerges from this chapter is that the preparation of the nectar of
immortality was also an important way in which one could obtain Bhairava-
hood. Moreover, it involved worship of a fearsome goddess, Caṇḍikā or Aghorī,
which again recalls the Kāpālikas in the Mālatīmādhava, who worship Cā-
muṇḍā.
But was the bodily amṛta the only nectar the Kāpālikas collected and con-
sumed?
5 Alcohol, Immortality and Soma
The Kāpālikas were also known for their use of alcohol in ritual. The Prabodha-
candrodaya (3.20 and prose) presents a Kāpālika rite of initiation, in the course
of which the Kāpālika offers alcohol to the initiands and calls it amṛta, for, once
again, this releases someone from the bondages of this world and of the state
of being a bound soul. In this sense, amṛta is not just a simple metaphor denot-
ing a precious or delicious liquid. It is a genuine nectar of immortality, for it
actually makes one immortal by bestowing final release, mokṣa.23
In a more satirical way, the Kāpālikas’ alcohol is also treated as their equiva-
lent to the vedic Soma. The Kāpālika in the Mattavilāsaprahasana cries out as
follows when he sees a pub:
My darling, look. This pub resembles the vedic sacrificial ground. For its
signpost resembles the sacrificial pillar; in this case alcohol is the Soma,
drunkards are the sacrificial priests, the wine glasses are the special cups
for drinking Soma, the roasted meat and other appetizers are the fire obla-
tions, the drunken babblings are the sacrificial formulae, the songs are the
Sāman-hymns, the pitchers are the sacrificial ladles, thirst is the fire and
the owner of the pub is the patron of the sacrifice.24
evaṃ kṛtvāpi vai devi -m- amṛtaṃ sādhakottamaḥ |

savyāsavyaṃ tato pītvā sarvajño bhavate kṣaṇāt ||120|| (kṣaṇāṃ MS)
bhairavo [’]tha svayaṃ sākṣā guhyakānāṃ prabhu[r] bhavet | (prabhu MS unmetr.)
(Superfluous visargas are placed between curly braces. Square brackets indicate editorial
additions. Comments and variants are in parentheses. A hiatus-filling m is printed as -m-.)
23 Śraddhā: bhaavaṃ, sulāe pūlitaṃ bhāanaṃ [= bhagavan, surayā pūritaṃ bhājanaṃ].
Kāpālikaḥ (pītvā, śeṣaṃ bhikṣukṣapaṇakayor arpayati):
idaṃ pavitram amṛtaṃ pīyatāṃ bhavabheṣajam |
paśupāśasamucchedakāraṇaṃ Bhairavoditam ||20||.
24 Kapālī: priye! paśya paśya | eṣa surāpaṇo yajñavāṭavibhūtim anukaroti | atra hi dhva-
jastambho yūpaḥ, surā somaḥ, śauṇḍā ṛtvijaḥ, caṣakāś camasāḥ, śūlyamāṃsaprabhṛtaya
via free access
Nobody would assume from this passage that the Kāpālikas were Soma
sacrificers—the comic effect intended is readily evident. It is nevertheless
interesting that, once again, the Kāpālikas are presented as having a special
nectar of their own, whether it is called amṛta or Soma, and that the ritual
significance of this nectar may be, it seems, comparable to the Soma of vedic
ritual.
6 Inventors of a New Nectar (soma/amṛta)? Or Simply ‘the Best?’
From the passages looked at here, no firm conclusion can be drawn as to why
the Kāpālikas included the word Soma in their initiation names and what
exactly they meant by “the teaching of or about Soma” (Somasiddhānta). The
most readily explicable case is found in the Junwani copper plate inscrip-
tion, which associates the Soma name with the name of their alleged founder,
Somaśarman. It is, nevertheless, possible that Kāpālikas identified Somaśa-
rman as their founder only retrospectively and that this derivation of Soma
from Somaśarman is secondary.
Conveniently, Soma as a proper name is also one of Śiva’s names,25 although
it does not necessary imply that he is accompanied by Umā (sa-umā). Soma
is probably used metaphorically for Śiva, just as it is used for other gods such
as Viṣṇu or Kubera. In any case, somasiddhānta can accordingly simply mean
“Śiva’s doctrine.” However, as a rather generic appellation of the god’s teaching,
it seems unlikely to designate the Kāpālika doctrine in particular.
Soma, meaning “moon,” and more particularly the nectar of immortality the
moon is supposed to contain, is another possible explanation. Various kinds of
nectar (amṛta/Soma), whether alcohol or essences of the human body, appear
to be in the focus of attention in Kāpālika rituals. The vital essences in par-
ticular were considered to have an invigorating effect that provided Kāpālikas
with the magical power they were apparently famous for. Concoctions of the
vital essences were probably thought to bestow omniscience and Bhairava-
hood. Whether it was really this nectar or these nectars that were at the origin
of the name Soma is impossible to tell; but whatever the case may be, the bod-
ily nectar of the vital essences was most probably a crucial element of Kāpālika
doctrine and practice.26
upadaṃśā havirviśeṣāḥ, mattavacanāni yajūṃṣi, gītāni sāmāni, udaṅkāḥ sruvāḥ, tarṣo

’gniḥ, surāpaṇādhipatir yajamānaḥ |.
25 See e.g. Skandapurāṇa 2.3, 29.103, Kathāsaritsāgara 7.1.101.
26 This may not be sufficient to explain what was meant by their “doctrine of/about Soma,”
via free access
44 törzsök
Given this rather wide range of possibilities, it is possible that the Kāpālikas
themselves intended to make full use the natural polysemy of the word Soma,
although it is less likely that such polysemy was intended from the very begin-
ning.
Finally, to add one more possible interpretation: Soma at the end of a com-
pound can also mean “chief, principal, the best.”27 In this sense, one could
understand the Kāpālika names to imply that they considered themselves sim-
ply the best Śaivas around.
Acknowledgements
I would like to thank all the participants of the Symposium Śaivism and the
Tantric Traditions held in honour of Professor A.G.J.S. Sanderson for their ques-
tions, comments and criticism concerning the first version of this paper, in par-
ticular Professor Sanderson and Professor Bakker. I am also indebted to Csaba
Kiss and Shaman Hatley for comments on and corrections of the final draft. The
written version of the paper reproduces some aspects of the oral presentation
stylistically.
References
Primary Sources
Kathāsaritsāgara of Somadeva[bhaṭṭa]. Pandit Durgaprasad and Kashinath Pandurang
Parab, eds. Revised by Wasudev Laxman Shastri Panshikar. Bombay: Nirnay Sāgar
Press, 1930.
Gautamadharmasūtra. See Olivelle 2000.
Jayadrathayāmala. National Archives, Kathmandu ms. nos. 5–4650 (ṣaṭka 1 and 2); 5–
722 (ṣaṭka 3); 1–1468 (ṣaṭka 4, NGMP reel no. A151–116). Electronic text provided by
Olga Serbaeva. I am grateful to Dr. Serbaeva for making her transcription available
to me.
Pauṣkaravṛtti of Jñānaprakāśācārya. IFP T. no. 110. Electronic text of the Muktabodha
Indological Research Institute, https://etexts.muktabodha.org/digital_library.htm.
Prabodhacandrodaya of Kṛṣṇamiśra. Armelle Pédraglio, ed. and trans. Le Prabodhacan-
drodaya de Kṛṣṇamiśra Paris: de Boccard, 1974.
Baudhāyanadharmasūtra. See Olivelle 2000.
but such “doctrine” was not necessarily an abstract theory. The word siddhānta may be
used in the sense of “teaching” rather than “philosophy” or “philosophical conclusion.”
27 For this meaning, see Apte 1957, entries on soma and nṛsoma.
via free access
Brahmayāmala. National Archives, Kathmandu Ms. No. 3–370. Electronic text provided
by Shaman Hatley. I am grateful to Dr. Hatley for making his transcription available
to me.
Mattavilāsaprahasana of Mahendravikramavarma. T. Gaṇapati Śāstrī, ed. Trivandrum:
Rājakīya Mudraṇayantrālaya, 1917.
Mālatīmādhava of Bhavabhūti. M.R. Kale, ed. Mālatīmādhava of Bhavabhūti with the
Commentary of Jagaddhara. Delhi: Motilal Banarsidass, 1996. (1st ed. 1928.)
Sarvajñānottara. Dominic Goodall, ed. Draft edition based on IFP Ms. T.334, IFP Ms.
T.760 and NAK Ms. 1–1692 and 3 printed editions. I am grateful to Dominic Goodall
for making his edition available to me.
Yājñavalkyasmṛti of Yājñavalkya. Narayan Ram Acharya, ed. Yājñavalkyasmṛti of Yogīś-
vara Yājñavalkya, with the Commentary Mitākṣara of Vijñāneśvara. Bombay: Nir-
nayasagara Press, 1949.
Skandapurāṇa. R. Adriaensen, H.T. Bakker and H. Isaacson, eds. Skandapurāṇa, vol. I.
(Adhyāyas 1–25). Critically Edited with Prolegomena and English Synopsis. Supple-
ment to Groningen Oriental Studies. Groningen: Egbert Forsten, 1998.
Skandapurāṇa. Kṛṣṇaprasāda Bhaṭṭarāī, ed. Skandapurāṇasya Ambikākhaṇḍaḥ. Ma-
hendraratnagranthamālā, no. 2. Kathmandu: 1988.
Secondary Sources
Apte, V.S. 1957 [1998]. The Practical Sanskrit-English Dictionary. Revised and enlarged
edition. Reprint (3 vols. in 1), Kyoto: Rinsen Book Company.
Bakker, Hans T. 2000. “Somaśarman, Somavaṃśa and Somasiddhānta: A Pāśupata Tra-
dition in Seventh-Century Dakṣiṇa Kosala.” In Harānandalaharī: Volume in Honour
of Professor Minoru Hara on his Seventieth Birthday, edited by Ryutaro Tsuchida and
Albrecht Wezler, 1–19. Reinbek. Dr. Inge Wezler Verlag für Orientalistische Fachpub-
likationen.
Bakker, Hans T. 2015. The World of the Skandapurāṇa—Northern India in the Sixth and
Seventh Centuries. Leiden; Boston: Brill.
Brick, David J. 2012. “The Origin of the Khaṭvāṅga Staff.” Journal of the American Oriental
Society 132 (1): 31–39.
Granoff, Phillis. 2006. “Śiva and His Gaṇas: Techniques of Narrative Distancing in
Purāṇic Stories.” In Voice of the Orient (A Tribute to Prof. Upendranath Dhal), edited
by Raghunath Panda and Madhusudan Mishra, 77–103. Delhi: Eastern Book Link-
ers.
Hatley, Shaman. 2007. The Brahmayāmala and Early Śaiva Cult of Yoginīs. Ph.D. thesis,
University of Pennsylvania.
Lorenzen, David N. 1991 [1972]. The Kāpālikas and Kālamukhas: Two Lost Śaivaite Sects.
2nd edition. Delhi: Motilal Banarsidass.
Lorenzen, David N. 1990 [1989]. “New Data on the Kāpālikas.” In Criminal Gods and
via free access
46 törzsök
Demon Devotees: Essays on the Guardians of Popular Hinduism, edited by Alfred

Hiltebeitel, 231–238. Reprint, New Delhi: Manohar.
Mallinson, James. 2007. The Khecarīvidyā of Ādinātha. A Critical Edition and Annotated
Translation of an Early Text of Haṭhayoga. London; New York: Routledge.
Mallinson, James. 2015. “Śāktism and haṭhayoga.” In Goddess Traditions in Tantric Hin-
duism—History Practice and Doctrine, edited by Bjarne Wernicke-Olesen, 109–140.
London; New York: Routledge.
Olivelle, Patrick. 2000. Dharmasūtras. The Law Codes of Āpastamba, Gautama, Baud-
hāyana and Vasiṣṭha. Delhi: Motilal Banarsidass.
Ondračka, Lubomir. 2007. “Perfected Body, Divine Body and Other Bodies in the Nātha-
Siddha Sanskrit Texts.” Pre-print version of a paper presented at the 13th World San-
skrit Conference. Available on the author’s page at www.academia.edu.
Sanderson, Alexis G.J.S. 1988. “Śaivism and the Tantric Traditions.” In The World’s Reli-
gions, edited by S. Sutherland, L. Houlden, P. Clarke and F. Hardy, 660–704. London:
Routledge.
Sanderson, Alexis G.J.S. 2006. “The Lākulas: New Evidence of a System Intermediate
between Pāñcārthika Pāśupatism and Āgamic Śaivism.” Indian Philosphical Annual
24: 143–217.
Sanderson, Alexis G.J.S. 2009. “The Śaiva Age.” In Genesis and development of Tantrism,
edited by Shingo Einoo, 41–350. Tokyo: Institute of Oriental Culture, University of
Tokyo.
Sanderson, Alexis G.J.S. 2012. “Śaivism and Brahmanism. All Souls, Hilary Term 2012.”
Lecture handout available on the author’s page at www.academia.edu.
Tāntrikābhidhānakośa. Dictionnaire des terms techniques de la littérature hindoue tan-
trique. A Dictionary of Technical Terms from Hindu Tantric Literature. Vol. III. Dominic
Goodall and Marion Rastelli, eds. Beiträge zur Kultur- und Geistesgeschichte Asiens,
no. 76. Vienna: Austrian Academy of Sciences Press, 2013.
Törzsök, Judit. 2011. “Kāpālikas.” In Brill’s Encyclopedia of Hinduism, vol. III, edited by
Knut A. Jacobsen, et al., 355–361. Leiden: Brill.
Törzsök, Judit. 2015. “The (Un)Dreadful Goddess: Aghorī in early Śākta tantras.” In God-
dess Traditions in Tantric Hinduism: History, Practice and Doctrine, edited by Bjarne
Wernicke-Olesen, 33–50. London; New York: Routledge.
via free access
chapter 3
Dressing for Power: On vrata, caryā, and vidyāvrata

in the Early Mantramārga, and on the Structure of
the Guhyasūtra of the Niśvāsatattvasaṃhitā
Dominic Goodall
If, twenty years ago, you had read most of the literature published before the
1990s about the Śaivasiddhānta, you would probably have received the impres-
sion that this was primarily a South Indian movement, whose scriptures, called
āgamas, were divided into four sections, or pādas, devoted to ritual (kriyā),
doctrine ( jñāna), yoga and pious conduct (caryā). The first two of these four
sections, the kriyā-pāda and the jñāna-pāda, you would have learnt, were the
most important, the kriyāpāda being devoted to describing the rituals practised
in the Śaiva temples of the Tamil-speaking area, and the jñānapāda (or vidyā-
pāda) being devoted to teaching and defending a strictly dualist system that
presents an ontological ladder of thirty-six tattvas, but that recognises three
irreducible ontological categories: pati, paśu and pāśa. That is to say: the Lord
(pati), bound souls (paśu), and the bonds that bind them (pāśa), namely Mat-
ter, karman and an innate impurity called mala or āṇava-mala.
Each one of these pieces of received wisdom has been challenged by the dis-
coveries of the last two decades, so that we now know that none of the above
propositions actually holds true for the earliest strata of the religion to which
surviving primary literature can give us access. A great many of those discov-
eries are those of Alexis Sanderson and the students to whom for decades he
devoted much of his time and energy.
Of course it is wide reading of a very broad corpus of published and unpub-
lished sources that has gradually revealed to us quite a different picture of the
early phases of the religion. But if one were to single out any one text for its
importance in expanding our knowledge of the early history of the Mantra-
mārga, it would probably be the Niśvāsatattvasaṃhitā.
Ten years ago, hardly any aspect of the text had been explored in print, but,
thanks in part to the spotlight of the Franco-German ‘Early Tantra’ project,
which between 2008 and 2011 focussed the minds of many people present at
the Toronto symposium on the Niśvāsa and on its relation to other early tantric
literature, parts of the work have been commented upon in an array of publica-
tions. The first major arcticle actually predates the ‘Early Tantra’ project, and is,
© dominic goodall, 2020 | doi:10.1163/9789004432802_005

via free access
48 goodall
of course, by Professor Sanderson himself: it is his study of the Lākulas (2006).

Apart from the first volume of the collaborative edition and translation, cov-
ering the earliest three books of the Niśvāsa—the Mūlasūtra, Uttarasūtra and
Nayasūtra—there are now substantial articles on, for instance, the evolution of
the system of tattvas that can be traced as it gradually takes shape within the
Niśvāsa-corpus (Goodall 2016), and on the lengthy grimoires of magical rites
contained in the Guhyasūtra that are similar in style and content to those found
in Buddhist kriyātantra works, and most strikingly similar to the those in the
Mañjuśriyamūlakalpa (Goodall and Isaacson 2016).
Now that the earliest three books of the Niśvāsa are published at last, and
now that the introductory book, the Niśvāsamukhatattvasaṃhitā, has been
thoroughly examined in a doctoral thesis defended this year at the University
of Leiden by Nirajan Kafle (2015*), what remains is the largest book of them
all: the Guhyasūtra. The ninth-century manuscript of the corpus (NGMPP Reel
No. A 41/14) comprises 114 folios, and all of the hitherto edited works together
cover only the first 40 of those. The remaining 74 folios give us the text of the
Guhyasūtra. Here is a very brief outline of the structure of its eighteen chapters:
A. Sādhana
Ch. 1 personality-types of sādhakas and types of liṅgas that may be used
for siddhi.
Ch. 2 liṅgapratiṣṭhā.
Ch. 3 preparations for sādhana, prognosticatory rites, vratas, procedures
for attaining certain siddhis.
B. Cosmography
Chs. 4–7 a lengthy cosmography (prakriyājñāna).
Ch. 8 a variant form of dīkṣā in the form of worship of a series of maṇḍalas
peopled by deities of the different levels of the cosmos (prakriyāyāga).
C. Other Mantra-systems
Finally, the use, primarily for magical powers, of mantra-systems other
than those given in the earlier three sūtras, namely
a) Chs. 9–11 The vyomavyāpin.
b) Chs. 12–14 The five brahmamantras.
Ch. 15 Long forms of their aṅgamantras.
c) Chs. 16–18 A ten-syllable mantra called vidyā.
The Guhyasūtra is somewhat like a series of appendices to the earlier sūtras,

containing more detailed accounts of some topics that have already been cov-
via free access
dressing for power 49
ered (cosmography), but also entirely new subjects (new mantras) or treat-
ments of subjects that have hitherto only been alluded to, notably the acqui-
sition of siddhis. As I have tried to indicate with the overarching titles (A, B,
C) in the brief summary above, I think that it can be said that chapters 1 to
3 have a certain sort of unity because they cover the acquisition of magical
powers in much greater detail than we see in earlier layers of the text: the
first chapter gives information about sādhakas, then stresses the importance
of the liṅga for attaining siddhis, after which, in chapter 2, the installation of
liṅgas is covered, and then in the third chapter we return to the preparations
for sādhana and finally the procedures to be followed. Chapters 4–7 then give
us a very detailed account of the Śaiva cosmos, the higher reaches of which
have been further expanded and embroidered upon since the composition of
the earlier sūtras of the text.1 This is undertaken because dīkṣā involves purg-
ing the soul of the fruits of karman that would need to be experienced—and
thus expended—through every layer of the Śaiva universe. Using the same
cosmography, chapter 8 describes an alternative dīkṣā involving the worship
of maṇḍalas representing successive layers of the universe, and it then con-
tains a number of add-on discussions that suggest, it seems to me, that the
text once drew to a close at that point, as we shall see below. What follow, tak-
ing us up to the end of the Guhyasūtra, are three distinct textual layers each
devoted to introducing an extra mantra-system, namely 1) that of the 81-word
vyomavyāpin, 2) that of the brahmamantras, and 3) that of the ten-syllable
vidyā. To each of these is attached a grimoire of magical recipes (kalpa).
Turning to the conclusion of chapter 8, I think that we can see from the sum-
mary given below that it reads like a series of codas. Verse 105 gives a clear
statement of what we are supposed to have learnt from the preceding chap-
ters, and it is followed immediately by remarks about the persons to whom
these teachings may and may not be submitted, a typical closing device. Tagged
on to this, from verse 116 onwards, is a treatment of religious suicide, again a
theme suitable to the conclusion of a work of scripture. The final section, from
verse 125, is introduced by Devī’s question about the status of rival religious tra-
ditions. In answer, Śiva explains that He and Devī, as consonants and vowels,
are the source of all language, and that they are the source of all the universe
in that they are to be identified with the various tattvas from which all else
evolves.
1 A full examination of this embroidery will have to await the publication of the relevant parts
of the Guhyasūtra, but some idea of its extent and nature may be gained from the table on
pp. 290–293 of Goodall, Sanderson and Isaacson 2015 and from the surrounding annotation,
as well as from Goodall 2016.
via free access
50 goodall
Summary of the conclusion of Guhyasūtra 8:2
8.88–89 Devī asks how an initiate may foretell his own death.
8.90–98 Śiva recounts signs of death.
8.99–104 Activities that can be done under particular asterisms that
grant release [from death?].
8.105 Summary of teachings from chapter 4 up to this point in chapter 8.
8.106–110 Those to whom one should and should not transmit this
knowledge.
8.111–114b The 4 means of liberation: dīkṣā, jñāna, yoga, caryā.
8.114c–115 One should transmit this only to someone worthy.
8.116–117 Devī asks about religious suicide.

8.118–122 Śiva deprecates death in tīrthas for initiates; he teaches instead
5 varieties of a ‘razor’-mantra for suicide by japa.
8.123 Increasing length of life by yogic dhāraṇās.
8.124 The supreme Śaiva knowledge, without which one cannot be liber-
ated, has been taught!
8.125–127 Devī asks about the fate of those who follow rival religions.
8.128–133 Śiva explains that He and Devī, as the consonants and vowels
of the alphabet, are the source of all linguistic expression (vāṅ-
mayam) and of all that has evolved (vikārāḥ).
8.134–136 The mūla-mantra is a panacea (mṛtasañjīvanī).
8.137–138 Śiva and Pārvatī are parents of everything in that they are
respectively these tattvas: puruṣa and prakṛti; kāla and niyati; īśvara
and māyā+vidyā; sadāśiva and kalā.
8.139–140 They are also respectively [supreme] Śiva and His Will (icchā).
8.141 Those who do not know the navātman, who are devoid of dīkṣā
and jñāna, who do not know the mūla, do not attain the highest
state.
2 Much of this conclusion has, by the way, been borrowed and adapted into the Niśvāsakārikā,
which seems itself like another series of addenda that further modify and extend the teach-
ings of the sūtras of the Niśvāsatattvasaṃhitā. Thus Guhyasūtra 8.88–89 and 8.92–104 have
been reworked to produce chapter 21 of the jñānakāṇḍa of the Niśvāsakārikā (T. 17, pp. 131–
133; T. 127, pp. 36–38); and Guhyasūtra 8.125–136 have been reworked to form the beginning
of chapter 20 (23 in T. 127) of the jñānakāṇḍa of the Niśvāsakārikā (T. 17, pp. 122–124; T. 127,
pp. 231–233). The chapter continues, at least in T. 17, for a further 32 verses on rival notions of
liberation and methods for attaining it.
via free access
We may remark in passing that a noteworthy feature of this final passage is

that the text takes no clearly defined position on the debate between dualism,
which was to become a defining characteristic of classical mainstream Said-
dhāntika doctrine, and non-dualism: it seems as though this debate had not
yet caught the interest and attention of Śaiva thinkers. Here is the passage in
question:
ahaṃ pumāṃs tvaṃ prakṛti niyati puna[[r eva ca]]

niya---[2] kālarūpī maheśvaraḥ 8.137
tvam māyā ca tathā vidyā aham punas tatheśvaraḥ
sadāśiva ahan devi tvañ caturdhā kaleśvarī 8.138
īśitvāc ca vaśitvāc ca sarvajñatvāc ca nityaśaḥ
śāntatvān niṣkalatvāc ca samatvāc ca ahaṃ śivaḥ 8.139
mama icchā na hanyā tvaṃ tvaṃ hi śaktibalodayā
tvatsūtañ ca jagat sarvaṃ śivadā sadanugrahe 8.140
I am puruṣatattva and you are prakṛti and also niyati; … Maheśvara is

Time; you are Māyā and Vidyā, while I am Īśvara-tattva. I, O goddess,
am Sadāśiva [and] you are mistress of the 4 kalās. (137–138)
Because I rule, I control, I am omniscient, because I am permanently at
rest, without division and in equilibrium, I am Śiva. (139)
You are my Will, not to be crossed, for you are the one from whom the
power of the śaktis arises!
The whole universe has sprung from you; You bestow Śiva-nature, O you
of true compassion! (140)
Having proposed an identification of the principal layers of redaction that are

detectable in the Guhyasūtra, I should like now to skip back to the verses in
the conclusion of chapter 8 that speak about those to whom the teachings may
and may not be transmitted, since these verses bear both upon the themes that
structure the work and upon the subject of caryā. For, slipped into the middle
of that section, beginning in verse 8.111, is a short sequence of verses that make
the claim that the teachings of the text comprise four independently salvific
parts: dīkṣā, jñāna, yoga and caryā.
etad buddhvā na dātavyaṃ śivadevāmṛtam param 8.110

dīkṣājñānena yogena caryayā ca yathākramam
pratyekaśaḥ śivāvāptis tantre ’smin pārameśvare 8.111
dīkṣayā sukaraṃ mokṣaṃ yad gurus sādhayet sadā
jñānañ ca gurum āsādya labhyate tat[[prasādataḥ]] 8.112
via free access
52 goodall
---[f. 72r] te mX yo XX(?) ca gurupādataḥ3

ātmaśaktyā carec caryāṃ sarvasiddhipradāyikām 8.113
etac catuṣṭayam proktaṃ saṃsārabhayanāśanam
parasyaiva na deyan tu yadi ’cchet siddhim ātmanaḥ 8.114
Knowing this, one should not give [lightly] the supreme nectar of Lord
Śiva. (110)
According to this scripture of the Lord, one may attain Śiva by each
of the following [practised individually] (pratyekaśaḥ): initiation,
knowledge, yoga and caryā in due order. (111)
By initiation one attains liberation easily, since it is the guru who invari-
ably accomplishes it.4 And knowledge is obtained, once one finds a
guru, through his grace. (112)
… yo[ga] … from the feet of the guru; One must practise caryā, which
bestows all supernatural powers, using one’s own strength (ātmaśak-
tyā). (113)
This tetrad has been taught to destroy the dangers of saṃsāra. It should
not [lightly] be given to others if one desires supernatural power for
oneself. (114)
An innocent might here at first suppose that we find here what may be the ear-
liest allusion to the notion that each Śaiva scripture should be arranged in four
text-units called pādas, for it is not difficult to see that the kriyāpāda might
easily be referred to by the most significant ritual of all, namely dīkṣā. Now
Brunner (1992) and others5 have shown that most early scriptures are not in fact
divided into four such text-units, and the Niśvāsa certainly is not. Nonetheless,
one might reasonably suppose that the four topics to which some later scrip-
tures devote four text-sections called pādas are referred to here. But are they?
Plainly the first three, dīkṣā, jñāna and yoga, may be found treated at length
in the Niśvāsa; but is there anything that we might recognise as caryā? This is
a word we are rather used to seeing translated as “conduct” or “comportment,”
as for instance in the title of Brunner’s 1985 translation of the kriyāpāda and
caryāpāda of the Mṛgendratantra: Mṛgendrāgama[.] Section des rites et section
du comportement. When she characterises the content of the caryāpāda there,
she observes (p. xxxvii):
3 Perhaps N once read: yogañ ca gurupādataḥ?

4 At the beginning of the kriyāpāda of the Mataṅga (1.2), initiation is similarly presented as an
alternative route to salvation that is easier than taking the more difficult path of jñāna.
5 See Goodall 1998, lviii–lxv.
via free access
La presque totalité de l’exposé (śl 1–105) est consacrée à un sujet unique:

le comportement normal des différents groupes d’ initiés.6
This is a topic that we really do not find addressed in the Niśvāsatattvasaṃhitā,

and it is therefore clear that caryā probably does not mean “comportement nor-
mal” in this text. When speaking in Toronto, I was unaware that Christian Wede-
meyer, faced with similar difficulties of interpretation resulting from assuming
such a meaning, had already devoted a chapter of his work on Making Sense
of Tantric Buddhism (2013, 133–169) to discussing how caryā and related terms
should be understood in early Buddhist and Śaiva tantric works.7 An explo-
ration of this theme therefore now seems in some respects less pressing to me
than it once did. But there are still some issues that can usefully be commented
upon, and there are several early Śaiva attestations of the nexus of caryā-related
terms of which Wedemeyer was not aware and which serve to adjust, I think,
some of what he has said about this semantic field, and that go some way to
explaining a significant juncture in the semantic voyage of the term caryā that
led to its being commonly assumed in modern scholarship to mean something
like “comportement normal,” even in passages in which such a meaning does
not fit. Wedemeyer’s account does clarify a number of confusions, and he is
to be commended for taking into account several Śaiva passages, but a combi-
nation of a desire to show that it is the Śaivas who have borrowed from the
Buddhists rather more than the Buddhists have from the Śaivas here (2013,
136–137, 154) and of not having had access to the earliest known Śaiva mate-
rial (which I should like to have made widely available long ago, but editing
concurrently the Niśvāsa, the Kiraṇa and the Sarvajñānottara is proving a very
time-consuming project) have led him to some problematic assertions and
assumptions, some of which I hope to correct below.8 In what follows, I will
6 “Almost the entire exposition (verses 1–105) is devoted to a single subject: the regular com-
portment of different groups of initiates.”
7 I am grateful to Tim Cahill for bringing Wedemeyer 2013 to my attention by kindly giving me
a copy when he was visiting Pondicherry in 2015.
8 John Nemec too expresses some reserves in his generally positive review (2014, 272–273) and
encourages further investigation of the Śaiva understanding of vratas:
Even if we grant that Wedemeyer limits his argument to instances of the antinomian prac-
tices that were understood to lead to liberation through a nondualistic, epistemological,
or gnostic insight, as I think he wishes to do, there is nevertheless some work left to be
done, in my view, to prove that even this particular understanding of the rites in ques-
tion originated with tantric Buddhism (and the Guhyasamājatantra in particular [160–
162, 166]). What is needed is a more thorough effort to establish the relative chronology
of the relevant texts and, more importantly, a more detailed account of the Śaiva self-
understandings of the religious observances in question.
via free access
54 goodall
be expanding upon and shoring up what was advanced rather too tentatively
in a lengthy note on Mūlasūtra 4.17c–18 (Goodall, Sanderson, Isaacson et al.
2015, 284–287).
In fact, the basic difficulty with the central term caryā had arguably already
been resolved, in nuce, by Alexis Sanderson in his 2006 article on the Lākulas,
but in a somewhat laconic fashion. What he writes, just before presenting the
vratas in the ninth chapter of the caryāpāda of the Mataṅgapārameśvara, is
the following:
The Śaivas have conventionally divided the means of liberation taught in

the Āgamas, that is to say their subject matter, into the four categories,
ritual (kriyā), doctrine or gnosis ( jñānam, vidyā), meditation ( yogaḥ),
and ascetic observance and other rules governing the conduct of the
various classes and kinds of initiate (caryā). Continuities between the
Lākulas and the Śaivas have now been shown in the areas of the ritual
of initiation and in the doctrine of the path to liberation, […] Insuffi-
cient evidence exists to permit much of a comparison in the domain of
meditation. […] This leaves only the domain of ascetic observance (vrat-
acaryā).
It is clear, in other words, that caryā, in early Śaiva works, may refer specifi-
cally to ascetic observance, presumably indeed because it is a contraction of the
collocation vratacaryā/vratacaraṇa, “the performance (caryā/caraṇa) of timed
religious observances (vrata).” The verb car, “to move,” but also “to be engaged
in,” has indeed long been the natural idiomatic verb of choice for use with vrata,
and this accounts for the frequency of such bahuvrīhi expressions as cīrṇavrata
(“who has observed his observances”), both in non-Mantramārga works (e.g.
Bodhāyanagṛhyasūtra 4.12.2 on p. 118, Yājñavalkyasmṛti 3.298c, Mahābhārata
3.81.135c) and in works of the Mantramārga (e.g. Mālinīvijayottara 10.17c and
10.34c, Mohacūḍottara 1.14a, etc), as well as for the distinctively tantric bahu-
vrīhi expression cīrṇavidyāvrata (e.g. Siddhayogeśvarīmata 13.1a),9 to which we
shall return below.
9 One non-tantric instance has been pointed out to me by Harunaga Isaacson (email of 26.xii
.2015).
… I find one occurrence of cīrṇavidyāvrata in a non-tantric text and a non-tantric context.
It is in Vyomaśiva’s Vyomavatī, the oldest of the commentaries on the Padārthadhar-
masaṃgraha (perhaps early 10th century; might be even slightly earlier). Of course, even
though the context is here of orthodox Vedic/brahmanical practice, we can be pretty cer-
tain that Vyomaśiva, as his name already suggests, was familiar with the Siddhānta and
via free access
If we were concerned only with the meanings of the word caryā, then it
might seem that we could almost end our essay here: Wedemeyer has pointed
out that caryā conventionally refers to virtuous behaviour and conduct in
accordance with religious precepts in a number of early non-tantric Buddhist
texts (2013, 135), where he characterises it as “by far the most common generic
term for the spiritual undertakings of buddhas and bodhisattvas,” just as it
does in much later Śaiva works of the Mantramārga, such as the Mṛgendra;
Sanderson has alluded (in the passage just quoted) to the observed fact that
it may refer in Śaiva sources both to the prescribed “conduct of the various
classes and kinds of initiate” as well as to “ascetic observance,” and he has
pregnantly suggested that this second meaning is connected with the notion
of vratacaryā; finally, Wedemeyer has observed that caryā in Buddhist tantric
sources, and in some Śaiva ones, refers not to life-long virtuous conduct, but
rather to timed antinomian practices, in troubling places such as cremation
grounds and involving transgressive sexual and mortuary elements.
But, as my title indicates, there is in fact a nexus of terms to be exam-
ined here. Wedemeyer indeed points out that there are several other related
terms that seem to be used in places where caryā might have served instead,
caryāvrata and vratacaryā being apparently “used with identical meaning”
(2013, 136), to which he adds instances of these words “in compound with
qualifiers related to ideas of secrecy or madness,” such as guhyavrata, guhya-
caryā, prachannavrata, unmattavrata, and “a cluster of interrelated terms that
appear in the same contexts, and which seem to be largely synonymous,”
which he tabulates on p. 137. Among these, he singles out vidyāvrata, for which
he suggests the translations “knowledge observance, spell observance, and/or
consort observance” (2013, 136) as being “treated as essentially equivalent to
caryāvrata/vratacaryā in Buddhist and Śaiva sources.”
Now it may indeed be the case that several of these terms appear to be
used interchangeably, but a slightly broader and chronologically deeper slice
of Śaiva samples reveals, it seems to me, both how the terms in fact differ from
each other and also why it is that they may in some contexts appear to be inter-
changeable, while at other times they are not. This may seem hair-splittingly
tedious, but, as Wedemeyer points out, if we do not understand the words,
then we cannot understand what it is that they serve to express.10 The Viennese
quite likely other forms of Śaivism, so there is a chance that his terminology has been
unconsciously influenced by tantric usage.
In the passage in question, on p. 233 of the edition, Vyomaśiva is glossing Praśastapāda’s
use of the term vidyāvratasnātaka.
10 Wedemeyer does not put this truism into such simple words, for he is particularly con-
via free access
56 goodall
endeavour that has so far produced three out of five volumes of the Tāntrikāb-
hidhānakośa is a step towards a better understanding of technical terms and
of common terms used with technical senses in the literature of the Mantra-
mārga, even if it does not, alas, cover Tantric Buddhist literature.11
Turning to the Viennese dictionary for an understanding of caryā is, how-
ever, not yet particularly useful, for the entry under this word consists only of a
cross-reference to the term caryāpāda. But the account of that term does con-
tain what will one day be a useful cross-reference to a future article on the term
vrata, and it includes one useful pointer to a moment in the history of the term
caryā:
Note that vrata is substituted as a synonym for caryā in Kir[aṇa] 6.6c;

indeed it is conceivable that the term is an abbreviation of vratacaryā
(Kir[aṇa] 49.4).
The volume of the Tāntrikābhidhānakośa in question appeared back in 2004;

by the time the fifth volume appears, including the terms vrata and vidyāvrata,
this dictionary will be a still more useful resource for tracing out the shifting
semantics of this and many another nexus of tantric terms.
Let us follow up this reference to the Kiraṇatantra. Since its chapter, 49, on
vratacaryā is short, we may quote much of it below, omitting from the middle a
detailed treatment of the ideal kamaṇḍalu, and giving just the readings of the
Devakottai edition (ED) and the Nepalese manuscript of 924 ad (N, f. 70r):
garuḍa uvāca—
samayī putrakaś cāpi deśikaś ca maheśvara
eṣāṃ vṛttiḥ samākhyātā sādhakasya bravīhi me 49.1
bhagavān uvāca—
sādhakaḥ sātviko dhīraḥ sahiṣṇur mantradhīr varaḥ
apradhṛṣyo mahāprājñaḥ samaloṣṭāśmakāñcanaḥ 49.2
udyukto homaniṣṭhaś ca japadhyānarataḥ sadā
cerned with understanding the meanings of common words that are used with technical
senses (2013, 134): “Recognition of these terms as terms of art is, however, essential, inso-
far as failure in this regard creates and sustains broad and systemic misinterpretation of
Tantric literature and of the traditions that produced (and were, in turn, produced by)
these works.”
11 The desirability of covering Buddhist Tantric literature is alluded to in the preface to the
third volume (p. 11), but it is obvious that the project cannot be simply “tweaked” at this
late stage to incorporate a huge extra corpus only in volumes 4 and 5.
via free access
vighnaprotsāraṇe kalyo vrataniṣṭhaḥ samaḥ śuciḥ 49.3

sasakhāyo vanaṃ gatvā vratacaryāṃ samārabhet
asahāyo tadā tasmiṃ svasakhāyaḥ kamaṇḍaluḥ 49.4
1ab samayī putrakaś cāpi deśikaś ca] N; samayīsūtayoś cāpi deśikasya ED ● 1b maheśvara]
ED; maheśvaraḥ N ● 1c eṣāṃ vṛttiḥ] ED; eṣā vṛtti N ● 2a sādhakaḥ sātviko dhīraḥ] ED;
sādhaka sātviko dhīra N ● 2b varaḥ] ED; vara N ● 2c apradhṛṣyo] ED; apradṛṣyo N ●
3c °protsāraṇe] ED; °procchāraṇe N ● 3d śuciḥ] ED; śucit N ● 4a sasakhāyo] N; sasa-
hāyo ED ● 4b vratacaryāṃ samārabhet] ED; vratācaraṇam ārabhet N ● 4cd tadā tasmiṃ
svasakhāyaḥ] em.; tadā tasmiṃ svasakhāya° N; yadā tasmin susahāyaḥ ED
Garuḍa spoke:
You have taught me, O great Lord, the activities of the neophyte, the
putraka and the ācārya. Tell me those of the sādhaka. (1)
The Lord spoke:
The excellent (varaḥ) sādhaka [should be] full of sattva, firm, capable
of endurance, his mind fixed on [his] mantra, unassailable, of great
wisdom, looking impartially on mud, stones and gold, (2)
engaged, regular in [the performance of] oblations, always devoted to
recitation and meditation, dexterous in the dispelling of obstacles,
firm in [the practice of his] religious observance, calm, pure. (3)
Accompanied by his ritual assistant, he should go to the forest and
begin the practice of his religious observance (vratacaryāṃ). [If he
is] without a ritual assistant, then his spouted water-pot is his ritual
assistant in that [practice].12 (4)
[Description of spouted water-pot omitted.]
saśalyas tumbako vā syād evaṃ kṛtvā vratañ caret 49.13

jaṭāmakuṭasāṭopaṃ śūlakhaṭvāṅgalāñchitam
śuddhamuṇḍārdhasaṃyuktaṃ tṛlocanakṛtādaram 49.14
12 This idea that the sādhaka, when embarking on the pursuit of supernatural powers, must
be accompanied either by a ritual assistant or by his water-pot is expressed elsewhere too,
for instance in Sarvajñānottara 25.19:
susakhāyo yadā mantrī mantrasādhanam ārabhet
asakhāyo yogī siddhiṃ kamaṇḍalukaraḥ sadā 19
19a susakhāyo yadā] N; ⊔ sāyo yadā L ● 19c asakhāyo yogī siddhiṃ] conj.; asākhayo-
gasiddhiṃ N; ⊔ hāyo yogī siddhiṃ L ● 19d kamaṇḍalukaraḥ] N; kamaṇḍaludharas
L
via free access
58 goodall
vyāghracarmāmbaraṃ śāntaṃ raudraṃ vratam idaṃ śubham

suniṣṭhasya bhavet ṣaḍbhir mmāsaiḥ siddhir ihottamā 49.15
madhyā māsaiś caturbhiś ca kṣudrā māsais tribhir bhavet
vratānām pravaraṃ raudram tatsiddhau sakalo bhavet 49.16
kāryaṃ mantravrataṃ siddhyai sādhakenānurūpakam
13c saśalyas tumbako] em.; saśalyatumbako N; saśalāstambhako ED ● 14c śuddha°] N; śud-

dhaṃ ED ● 14d tṛlocanakṛtādaram] N; trilocanakṛtodaram ED ● 15b raudraṃ] N; raudra°
ED ● 15cd suniṣṭhasya bhavet ṣaḍbhir mmāsaiḥ siddhir ihottamā] N; kaniṣṭhasya bhavec
chuddhir māsaiḥ ṣaḍbhir ihottamā ED ● 16ab caturbhiś ca kṣudrā māsais] ED; caturbhi
syāt kṣudra māsai N ● 16c vratānām pravaraṃ] N; vratamapravaraṃ ED ● 16d sakalo
bhavet] N; sakaḷaṃ punaḥ ED ● 17a kāryaṃ mantravrataṃ] ED; kāryamantravrata N ● 17b
sādhakenānurūpakam] N; sādhakair nānurūpataḥ ED
Alternatively, [instead of a kamaṇḍalu,] it may be a gourd with a shaft.

Having made this [ready], he should practise his observance. (13cd)
This is the auspicious Raudra-vrata: imposing with a chignon of mat-
ted locks, marked by a trident and khaṭvāṅga, equipped with a clean
half skull, awe-inspiring with a third eye, clothed in the skin of a tiger,
peaceful. (14–15b)
For one firm [in this observance], the highest siddhi will arise in six
months; middling [powers] in four months; the lowest [powers] will
arise in three months. (15c–16b)
The highest of the observances is the Raudra[-vrata]. On accomplishing
that, one becomes [equal to] the Sakala [form of Śiva]. (16cd)
For attaining siddhi, the sādhaka should perform a mantra-observance
that is appropriate [to the mantra in question].13 (17ab)
What we see here, it seems to me, is a reflection of the old notion that caryā
refers to vratacaryā, “the performance of a vrata,” where vrata is a timed reli-
gious observance that typically involves adopting an unusual diet (not men-
tioned here), an unusual style of dress (often with accoutrements of the cre-
mation ground, in this case the khaṭvāṅga), and unusual behaviour (sexual
transgressions, mortuary obsessions, or, as here, ascetic detachment). This
observance is furthermore a preparation for the attainment of magical powers
through the use of a mantra.
This is, I think, in essence, the same as what is meant by the term vrata in all
early tantric literature. It explains therefore how vratacaryā, “the performance
13 Or perhaps “appropriate [to the desired siddhi].”
via free access
of such a vrata,” and therefore sometimes also caryā have come to be used
interchangeably with vrata.14 But the term caryā evidently began to expand
and then slip in meaning as the Mantramārga expanded to include not just
sādhakas (who seem to be the only audience of the Niśvāsatattvasaṃhitā15),
but also other categories of initiates. We can see that this slippage has in
fact already taken place by the time of the Kiraṇa, for that work begins its
thirty-first chapter with an announcement that the next topic to be taught
will be caryā, and yet, as we have just seen above, does not deal with the
vratacaryā of the Sādhaka until chapter 49, which follows eighteen chapters
later.
Here is the beginning of chapter 31, in which the topic called caryā is first
introduced in such a way as to suggest that the primary meaning has now
become something like regular enjoined “comportment.”
garuḍa uvāca—
samayisutayor deva kā vṛttis tu dine dine
evaṃ mayi samācakṣva caryā me noditā purā 31.1
1a samayisutayor] N (unmetrical); samayīputrayor ED ● 1b vṛttis tu] ED; vṛtyatra N ● 1c

mayi] conj.; mayā N; sarvaṃ ED
Garuḍa spoke:
What are the day-to-day activities of the neophyte and the putraka?
Tell me this. You have not taught me the [regular rules of] behaviour
(caryā) before. (1)
No such clear evidence can be found of this broadening of the meaning of the
word caryā in another post-Niśvāsa but pre-tenth-century Saiddhāntika scrip-
ture for which we have an early Nepalese palm-leaf witness, this time appar-
ently of the ninth-century, namely the Sarvajñānottaratantra.16 That work gives
14 As for caryāvrata, which, as we have seen above, Wedemeyer considers to be synonymous

with vratacaryā, I suspect that it rather means “one of the timed religious observances
belonging to [the body of activities that can collectively be called] [vrata-]caryā.” No
doubt there is, in certain contexts, little difference between saying this and saying “the
performance of timed religious observances” (vratacaryā).
15 For the absence of initiates who are not sādhakas from the religious teachings of the Niś-
vāsa, see Goodall, Sanderson, Isaacson et al. 2015, 47ff.
16 In case it should be supposed that the work’s structure is itself evidence of a shift in mean-
ing of the term caryā, I should mention that, although Aghoraśiva’s commentary divides it
via free access
60 goodall
us an account of another sort of vrata that will be useful to us in the discussion

below, in this case somewhat more detailed, but involving no transgression of
brahmanical rules about purity and sexual behaviour.
ataḥ paraṃ pravakṣyāmi vratānāṃ vratam uttamam17

śivavrateti vikhyātaṃ sadevāsurapūjitam18 18.1
viśuddhaṃ pāṇḍaraṃ bhasma19 śuklavāsonulepanam
śuklayajñopavītī20 ca jaṭāmakuṭamaṇḍitaḥ21 18.2
sarvābharaṇasampannaḥ22 śuklamālyavibhūṣitaḥ23
carubhug brahmacaryasthaḥ24 śivāgnigurupūjakaḥ25 18.3
mantramūrtiḥ26 śivasyaiva yathā rūpaṃ27 prakīrtitam
tathā vai sādhakendrāṇāṃ28 vrataṃ jñeyaṃ tadātmakam29 18.4
śuklakaupīnavāso vā uṣṇīṣākṣakamaṇḍaluḥ30
śivālaye vasen nityaṃ31 bhikṣābhakṣo32 jitendriyaḥ 18.5
japadhyānarato33 maunī śivāgnigurupūjakaḥ
saṃvatsare vyatikrānte śivatulyo bhaved asau34 18.635
into four pādas, including a caryāpāda, it was clearly not originally so divided: see Goodall
1998, lix–lxi.
17 vratam uttamam] NL; uttamaṃ vratam M
18 śivavrateti vikhyātaṃ sadevāsura°] N; śivavratam iti khyātaṃ sarvodāsura° M; śivavratam
iti khyātaṃ sarvadā sura° L
19 pāṇḍaraṃ bhasma] N; pāṇḍuraṃ bhasma M; pāṇḍaraṃ janma L
20 śukla°] NL; ⊔ M
21 °maṇḍitaḥ] ML; °maṇḍitam N
22 °sampannaḥ] M; °saṃpanna° NL
23 °mālyavibhūṣitaḥ] N; °mālāvibhūṣitaḥ M; °mālāvibhūṣitaiḥ L
24 carubhug brahmacaryasthaḥ] M; carabhug brahmacaryastho N; carubhūt brahmacar-
yasya L
25 °pūjakaḥ] NM; °pūjitaḥ L
26 °mūrtiḥ] M; °mūrti° NL
27 rūpaṃ] NM; pūrvaṃ L
28 sādhakendrāṇāṃ] NM; sādhakaindrāṇā L
29 vrataṃ jñeyaṃ tadātmakam] NM; vrata jñeyaṃ tathātmakam L
30 śuklakaupīnavāso vā uṣṇīṣākṣakamaṇḍaluḥ] NM; śuklakaubī(pī)navāsaṃ va uṣṇīṣākṣata-
(ka)maṇḍalum L
31 vasennityaṃ] N; vasannityaṃ ML
32 bhikṣā°] ML; bhikṣa° N
33 japa°] NM; śiva° L
34 bhaved asau] conj.; bhavedasauditi N; bhaved iti ML
35 Here there is a flourish marking a chapter-break in N, and in the Southern sources there is
a chapter-colophon: iti śrīmatsarvajñānottare śivavratapaṭalo ’ṣṭādaśaḥ M; iti sarvajñān-
ottare kriyāpāde śivavrataprakaraṇam L
via free access
ataḥ paraṃ pravakṣyāmi lakṣaṇaṃ36 tu śivālaye

sthāpanaṃ caiva liṅgasya yasmin sarvaṃ pratiṣṭhitam37 19.1
brahmādyā devatāḥ sarvā liṅgam āśritya saṃsthitāḥ38
tasmād vai sthāpayel liṅgaṃ39 śāstradṛṣṭena karmaṇā 19.2
cīrṇavidyāvrato40 yogī gurudevāgnipūjakaḥ41
Next, I shall teach the best observance among observances, which is

known as the Śiva-vrata and which is revered by asuras and gods
alike. (18.1)
Pure pale ash [should be used, and] white dress and unguents; he
should wear a white sacred thread and be adorned by a chignon of
matted locks. (18.2)
He should be equipped with all [suitable] ornaments, [and] adorned
with white garlands; he should consume [only the pure ritual gruel-
offering known as] caru; he should observe the chaste conduct of a
student; he should venerate Śiva, the fire and his guru. (18.3)
He should be mantra-bodied;42 the appearance (rūpam) of excellent
sādhakas [who follow this observance] is to be the same as that of
Śiva: the observance must be understood as consisting in this. (18.4)
Alternatively, he may wear [just] a white loin-cloth, [and bear] a turban,
rosary and spouted water-pot. (18.5ab)
He should dwell constantly in a temple of Śiva, eating alms, controlling
his senses, devoted to recitation and meditation, maintaining silence,
venerating Śiva, the fire and his guru. When a year has passed, he will
become equal to Śiva. (18.5c–6)
Next, I shall teach the characteristics of a temple of Śiva, as well as [how

to perform] the installation of the liṅga, in which the universe is
[itself] ‘installed.’ (19.1)
36 lakṣaṇaṃ] NMT; lakṣmaṇaṃ L

37 sarvaṃ pratiṣṭhitam] M; sarvvapratiṣṭhitam N; sarvapratiṣṭhitaḥ TL
38 devatāḥ sarvā liṅgam āśritya saṃsthitāḥ] em.; devatā sarvā liṅgam āśṛtya saṃsthitā N;
devatāḥ sarve liṅgam āśritya saṃshitaḥ M; devatā sarve liṅgam aśritya saṃsthitāḥ T;
devatās sarve liṅgam āśritya saṃsthitāḥ L
39 sthāpayel liṅgaṃ śāstra°] MTL; sthāpaye liṅgaṃ śastra° N
40 cīrṇavidyāvrato yogī] N; cīrṇo vidyāvrato yena M; ciṇyāṃ vidyāvratā yoga T; ciṇyāṃ vidyā-
vrato yogī L. Cf., e.g., Mohacūḍottara (NGMPP A 182/2, f. 2r): cīrṇavidyāvrato mantrī jñāna-
vān susamāhitaḥ.
41 °pūjakaḥ] NML; °pūrvakaḥ T
42 I assume that this means that the mantras of Śiva should be installed upon his body.
via free access
62 goodall
All the gods, beginning with Brahmā, reside in the liṅga; therefore a
yogin who venerates his guru, God and the fire and who has per-
formed his vidyāvrata should install the liṅga, following the proce-
dure taught in scripture. (19.2–3b)
We shall return below to the use of the term vidyāvrata, which I have not trans-
lated here. First we may observe that these passages of the Kiraṇa and Sarvaj-
ñānottara might appear to confirm Wedemeyer’s observation that the vratas in
early Śaiva works were observances in which the sādhaka imitated God (2013,
165).
The early Śaiva Tantric paradigm for the transgressive vrata, then, was
one of imitatio dei—mimicking the activity of the god in the interest
of eliding the (presumably mistaken) sense of a gulf between him and
the devotee. In none of these rites is there mention of transcendence of
conceptuality or attainment of any epistemic nonduality—the concern
seems entirely to be one of nonduality in the sense of union with the god
Śiva.
Imitation of forms of god, whether pure or transgressive, seems indeed to be

typical of vratas in early Śaiva sources, but it is not the invariable rule, as the
Niśvāsa demonstrates.43 Several vratas are described in the course of the work,
but there is one passage in which a set of nine is concisely described together,
namely in chapter 3 of the Guhyasūtra. A brief summary of the contents of that
chapter will help to contextualise that description, showing that it is part of a
chapter devoted to 1) preparations for magical pursuits, and 2) magical proce-
dures:
43 Nemec also (2014, 273) expresses doubt about this point of Wedemeyer’s:
I am, in a word, not convinced that the many transgressive practices in the “early
period” of Śaiva tantra, defined as “pre-tenth century” (165), involve a practice of imita-
tio dei, “of union with the god Śiva,” to the exclusion of “transcendence of conceptuality
or attainment of any epistemic nonduality” (ibid.).
As to what states of consciousness such non-imitative observances might or might not be
intended to achieve, the text gives us no direct information; we can only say that it does
elsewhere describe practices whose purpose is said to be transcending duality, for exam-
ple in yogic meditations described in Uttarasūtra 5.42–43 and Nayasūtra 4.55ff., and that
a non-dualist cosmogony is sketched out in Uttarasūtra 1.13.
via free access
Summary of Guhyasūtra 3
Preparations for magical pursuits
3.1–2 Having set up the God of gods in a suitable place, one may employ
a ritual assistant (uttarasādhaka) for attaining the highest siddhi.
3.3–6 Qualifications of the uttarasādhaka.
3.7–11 Construction of a special dwelling for the pursuit of siddhi.
3.12c–16 Alternative: a suitable cave or empty temple. One should live
from vegetables or begging or from roots, and perform fasts (cān-
drāyaṇa, etc.)
3.17–22 Prognostication of success in siddhi by consulting Svapna-
māṇavaka by calling him to appear in one’s sleep.
3.22–23 Prognostication by consulting Amoghamantrarāja.
3.24–27 Catoptromantic prognostication (prasīnā) using virginal chil-
dren and the mantra of Caṇḍī.
3.28–29 japa using the akṣamālā [in order to prepare it].
3.31–43b vratas.
Magical procedures
3.43c–112 sādhanas for attaining various siddhis.
Having underlined the context, we may now turn to the vratas given in this
chapter, followed, by way of example, by one magical procedure. We may note
that all of these observances transgress social norms to some degree, but that
none unambiguously involves imitatio dei:
siddhi-m-aiśvaryayogyas44 tu na ca hiṃsanti hinsakāḥ 3.29

siddhavidyāvratastho hi jape45 ca vratam ārabhet
go mātā ca pitā bhrātā atithir mitra brāhmaṇaḥ 3.30
hato me pāpa[[(cāre)]]ṇa46 caren (1) Mithyāvratam vratī
<<karasthena kapā>>lena47 khaṭvāṅgī bhasmaguṇṭhitaḥ 3.31
śmaśāne carate rātrau (2) Śmaśānavrata ucyate
44 °yogyas tu] NW; °yogye tu K

45 jape] NW; japaṃ K
46 pāpa[[(cāre)]]ṇa] K; yāpa XXṇa N; yāpa Xreṇa W; pāpakāreṇa conj. Sanderson (2006, 209)
47 karasthena kapālena] conj. Sanderson; ---lena NW; ⊔na K
via free access
64 goodall
nṛtyate gāyate caiva unmatto hasate bruvan48 3.32

bhasmāṅgī cīravāsaś ca (3) Gaṇavratam idaṃ smṛtam
japayukto bhaikṣabhujo loṣṭuśāyī jitendriyaḥ 3.33
dhyānasaṃyamayuktaś ca (4) Loṣṭukavratam ācaret
rikṣavyāghrasamā[[kīrṇe]]49 <<vane siṃ>>hasamākule 3.34
jitanidrāśano50 jāpī (5) Kāṣṭhavratam idañ caret
gāyate nṛtyate jāpī strīrūpī valayabhūṣitaḥ51 3.35
śūrppakandukaveṇībhiś (6) Citravratam idañ caret
śastrapāṇir dayāyukta-m-aṭe trāteva52 †jatavān† 3.36
japadhyānārccanirato (7) Vīravratam idañ caret
varṣaśītātapair ddehan tāpayed dhi su--- 3.3753
japadhyānarataś caiva (8) Mahāvratas sa ucyate
ratisaṃbhogakuśalāṃ rūpayauvanaśālinīm 3.38
īdṛśīṃ striyam āsādya niruddhendriyagocaraḥ
cumbanāliṅganaṅ kuryāl liṅgaṃ sthāpya bhagopari 3.39
japadhyānaparo bhūtvā (9) Asidhāravratañ caret
yadi kāmavaśaṃ gacchet patate54 narake dhruvam 3.40
navātmakañ japel lakṣaṃ ((tasya)) ---ddhaye55
abdaṃ ṣaṇmāsamātraṃ vā yaś cared vratam uttamam 3.41
tasya siddhiḥ prajāyeta adhamā madhyamottamā56
vratasthaḥ57 pañcalakṣāṇi punar japtvā tu siddhyate 3.42
sarve mantrāś ca siddhyante īpsitañ ca phalaṃ bhavet
[A spell for travelling great distances:]
oṃ namo vāyupathacāriṇe amitagatiparākramāya vimale kulu 258

ṭhaṭha 3.43
48 bruvan] NW; dhruvam K

49 °kīrṇe] conj.; °kīrṇa° K; °kā--- N; °kā ⊔ W
50 °drāśano] N; drāmano K; °drāsano W
51 valayabhūṣitaḥ] conj. (unmetrical); valabhūṣitaḥ NKW
52 m-aṭe trāteva] conj.; maṭhe trā--- N; maṭhatrā ⊔ ca K; maṭhe trā ⊔ W
53 °tāpayeddhi su] N; °tāpaye ⊔ K; °tāpayedvi ⊔ W
54 °vaśaṃ gacchet patate] conj.; °vaśaṃ cche patate N; °vrataṃ gacchet patate K; °vaśaṃ
chai Xpatate W
55 lakṣaṃ tasya ---ddhaye] em.; lakṣa---ddhaye N; lakṣa ⊔ dvaye K; lakṣaṃ tasya ⊔ dvaye W;
lakṣaṃ tasya mantrasya siddhaye conj. Diwakar Acharya; lakṣaṃ saṃvatsaradvaye conj.
Diwakar Acharya
56 °mottamā] K; °mottamāḥ NW
57 vratasthaḥ] K; vratastha NW
58 kulu 2] conj. Sanderson; kula---2 N; kulaṃ 2 K; kula 2 W
via free access
śilā suvarṇadhātuñ ca ((ka)) --- tam

varāhavaśasaṃpiṣṭaṃ59 sahasraparimantritam 3.44
navātmakam punar japtvā pādau caiva pralepaye[[t]]
gacchate so ’pariśrānto yojanānāṃ śatadvayam 3.45
[Once the rosary has been thus prepared, he becomes] ready for siddhis
and power. (29cd)
Dangerous creatures do not harm one who has [first] accomplished
an observance [that qualifies one] for [using] Spells: he should
begin an observance by means of recitation ( jape[= japena]).60
(30ab)
The one engaged in observance should practise the False Observance
(mithyāvrata) [by wandering about proclaiming]: “I have committed
bad deeds: I have killed a cow, mother, father, brother, a guest, friend,
brahmin!” (30c–31b)
[If] one wanders in the cremation-ground at night, with a skull in one’s
hand and a khaṭvāṅga, covered in ashes, that is called the cremation-
ground observance (śmaśānavrata). (31c–32b)
If one dances, sings, laughs and talks madly, with the body smeared in
ashes and wearing rags, this is called the Gaṇavrata. (32c–33b)
One performs the Clod-of-Earth Observance (loṣṭukavratam) by
being engaged in recitation, feeding on alms, sleeping on the earth,
with senses controlled, engaged in meditation and restraint. (33c–
34b)
One may perform the Block-of-Wood Observance (kāṣṭhavratam) in a
forest full of bears, tigers and lions, conquering the urges to sleep and
eat, [constantly] reciting. (34c–35b)
If one takes on the appearance of a woman and sings and dances,
adorned with bracelets, with a winnowing fan, ball and plait, one
observes the Colourful Observance (citravratam). (35c–36b)
With a weapon in hand, full of compassion, if one wanders like a saviour
of creatures (?)61 focussed upon recitation, meditation and worship,
one performs the Warrior Observance (vīravratam). (36c–37b)
59 varāhavaśasaṃpiṣṭaṃ] conj.; varāhava+sa+saṃpiṣṭa° N; varāhavamasaṃpiṣṭaṃ K; varā-

havaśasaṃpiṣṭa W
60 This replaces the translation, in which I no longer believe, of 30ab that we offered on p. 285
of Goodall, Sanderson, Isaacson et al. 2015.
61 This tentative translation assumes that jatavān is an error for jantavām, intended as a
genitive plural with the sense of jantūnām!
via free access
66 goodall
If one torments the body with rain, cold and heat, …, devoted to
recitation and meditation, this is called the Great Observance
(mahāvrataḥ). (37c–38b)
A woman skilled in the pleasures of love-making, endowed with beauty
and youth; such a woman one should procure, holding one’s senses
back from the objects of the senses, and one should kiss and embrace
[her], placing the penis upon her sex while remaining focussed upon
recitation and meditation—one performs [thus] the Sword-Blade
Observance (asidhārāvratam). If one should succumb to the control
of desire, then one certainly falls into hell. (38c–40)
One should recite the navātman one lakh times … for [si]ddhi: one
who [thus] observes such an excellent observance for a year or
just six months attains lowest, middling or best siddhi. But if, while
observing such a vrata, someone recites five lakh times, then [that
mantra] succeeds [for him] (siddhyate), and all mantras succeed for
him and he attains the fruits he desires. (41–43b)
[Using the mantra] oṃ namo vāyupathacāriṇe amitagati-
parākramāya vimale kulu kulu svāhā, [and taking] arsenic,
gold [and?] a mineral, …, ground up with pig fat/marrow, over which
one has recited [the navātman] 1000 times, he should smear [the
mixture] on his feet/legs, while once again reciting the navātman:
he will travel 200 yojanās unwearied! (43c–45)
Right at the beginning of the above-quoted passage, we find a further attesta-

tion of the term vidyāvrata, and this time, rather than prevaricating further, I
have proposed translating it as “an observance [that qualifies one] for [using]
Spells.” There are other passages that can and will be adduced in support of
this, but I think that it should already be becoming clear from this passage of
the Guhyasūtra and from the passage quoted just before from the Sarvajñān-
ottara that the different particular vratas that are performed serve to prepare
the performer for some subsequent religious activity that involves the use of
the mantra or vidyā. In the case of the Guhyasūtra, it is the pursuit of siddhi
for which the sādhaka is prepared; in the case of the Sarvajñānottara, the indi-
vidual is prepared for the performance of the installation (pratiṣṭhā) of a liṅga.
From the beginning of chapter 10 of the kriyāpāda, we learn that it is also an
essential to the consecration of an ācārya in the Mataṅgapārameśvara:
catuṣpādārthakuśalaṃ mahotsāhaṃ hy aninditam 10.2

ṣaṭpadārthapraṇītārthaṃ sarvabhūtahite ratam
gurus tam abhiṣiñcet tu cīrṇavidyāvrataṃ naram 10.3
via free access
Rāmakaṇṭha: atha kiṃ tad vidyāvrataṃ yat tena cīrṇam ity ucyate:
vidyāśaktir ihopāttā japtavyā prāk chivālaye

saṃniyamyendriyagrāmam abdam ekaṃ śuciṣmatā 10.4
nityaṃ carubhujā bhūmyāṃ kuśaprastaraśāyinā
pūjāgnibhavane yuktacetasā bhāvitātmanā 10.5
Rāmakaṇṭha: [vidyāśaktiḥ] vyomavyāpilakṣaṇā.62
The guru should consecrate [as an ācārya] a man who is skilled in what
is taught in all four pādas, who has great energy, who is beyond reproach,
who expounds the meaning of the teachings [encapsulated] in the six
topics [of this scripture], who is devoted to the welfare of all beings, who
has performed the observance for [the propitiation of his] mantra.
Rāmakaṇṭha: Now if you ask what this vidyāvrata is which he must have
observed, this is what the text teaches:
The power of the vidyā that is mentioned here [in this compound vidyā-
vrata] is first to be recited for a year in a temple to Śiva, while exercising
control of the senses, maintaining purity, eating daily [only the sacrificial
gruel known as] caru, sleeping on the ground in the room reserved for
pūjā and fire[-sacrifice] on a spread of kuśa-grass, with his mind engaged
[in meditation], focussed.
Rāmakaṇṭha: It [viz. the power of the vidyā] is the vyomavyāpin.
Once this preparation, taking the form of the observance of one among a vari-
ety of possible vratas (but ideally one suitable to the mantra to be put to use,
as seems already to be implied in Kiraṇa 49.17ab above, and as we shall see
confirmed below), is complete, the observer can be called cīrṇavidyāvrataḥ (as
here and in Sarvajñānottara 19.3a) or siddhavidyāvratasthaḥ (as in Guhyasūtra
3.30a).
If we make such an assumption, then we can see how the expressions vrat-
acaryā and vidyāvrata might be regarded as interchangeable in some con-
62 From Rāmakaṇṭha’s commentary, only the avataraṇikā and the commentary to 10.4–6 are
quoted here (not the remarks on 10.2c–3). The pratīka in square brackets is supplied by the
editor.
via free access
68 goodall
texts,63 even though they are not actually synonymous. It also becomes clear
how it is that vidyāvrata can be characterised as preparatory mantra-pro-
pitiation and therefore equivalent to what may also be called puraścaraṇa.
Sanderson characterises it in such a way when referring to the account of the
Bhairavācārya in Bāṇa’s Harṣacarita:
With this account of a pūrvasevā, also called puraścaryā or vidyāvratam,

i.e. the initial period of ascetic japaḥ etc. to be undertaken after one has
received a Mantra, whereby one becomes able to accomplish feats (kar-
māṇi) with that Mantra …64
So if vrata and vratacaryā and some other terms may seem to be used inter-
changeably in some works with vidyāvrata, it is because the principal purpose
of the vratas taught in the early Mantramārga seems to be to propitiate mantras
prior to further religious activities involving those mantras, rendering those
who complete such observances describable by such terms as cīrṇavidyāvrata,
siddhavidyāvratastha, vidyāvratasnāta and so forth.65 Various observances can,
in other words, be observed in order to become one “who has completed the
observance [required for the propitiation] of a vidyā.”
This discussion might seem to suggest that in finding the original meaning of
the expression vidyāvrata we believe that we have found its immutable seman-
tic core, but that is not really what I intend to say. Of course the term vidyāvrata
may have gone on to evolve and be used in contexts that suggest that one
might elsewhere also or instead render it as “knowledge observance” or “con-
sort observance” (Wedemeyer 2013, 136) or, as we shall see below, “observance
relating to a vidyāṅgamantra.”66 Furthermore, one might argue that we have in
63 Cf. Wedemeyer 2013, 159: “It is worth noting that the Tantrasadbhāva/Kubjikāmata and the
Siddhayogeśvarīmata clearly take the terms vidyāvrata and vratacaryā to be synonymous.”
64 Sanderson 2001, 13, note 11. Wedemeyer quotes from this definition (2013, 255, note 96),
but in a manner that suggests that he was oddly not convinced by it, or not convinced
that the same kind of vidyāvrata was being alluded to by Sanderson:
Sanderson (“History”, 13n11) also describes a very different rite [scil. from that referred
to in Vīṇāśikhatantra 180?] when he speaks of vidyāvrata as an “initial period of ascetic
japaḥ etc. to be undertaken after one has received a Mantra,” i.e., he takes it to be a kind
of pūrvasevā or puraścaryā.
65 Another purpose of performing vratas in the early Mantramārga is of course expiation:
see, for example, Guhyasūtra 9.10a, Siddhayogeśvarīmata 10.3c, both quoted below, and
Hṛdayaśiva’s Prāyaścittasamuccaya passim (appendix to Sathyanarayanan 2015).
66 We do not aim, however, to examine here all later passages in which the meaning of
vidyāvrata is arguably stretched. One such passage is a sequence of verses discussing the
term that has been borrowed from the Tantrasadbhāva into the Kubjikāmata: that dis-
via free access
any case not started from its point of origin, for the term has presumably been
drawn from or at least coloured by the brahmanical expression vidyāvratasnāta
that is common from the Gṛhyasūtras onwards and that we find, for instance,
in Manusmṛti 4.31:
vedavidyāvratasnātān śrotriyān gṛhamedhinaḥ

pūjayed dhavyakavyena viparītāṃś ca varjayet
Olivelle (2005, 125–126) translates:
At rites for gods and ancestors, he should honor individuals who have
bathed after completing the Vedas, vedic learning, or vedic vows, who are
vedic scholars, or who are householders, but avoid individuals different
from these.67
It seems to me very likely that the use of vidyāvrata in the Mantramārga—and

a fortiori of vidyāvratasnāta (Siddhayogeśvarīmata 10.20 and Svāyambhuvasū-
trasaṅgraha 21.35)—should have been influenced by earlier brahmanical usage
such as we see in the Gṛhyasūtras and in the Manusmṛti. Nonetheless, in the
cussion begins with Bhairava saying (Tantrasadbhāva 4.2ab = Kubjikāmata 25.30ab): śṛṇu
devi pravakṣyāmi vidyāyā vratam uttamam, “Listen, O goddess: I shall teach the excellent
observance of/for vidyā.”
67 Olivelle also adds a note that explains that there is doubt about the term (2005, 270):
[M]ost commentators take vedavidyāvrata as three separate categories. The first refer
to those who have only learned the Veda by heart; the second to those who have mas-
tered its meaning; and the third to those who have completed the vows associated with
vedic study, such as living with the teacher for a certain number of years, even if they
have not mastered the Veda.
This interpretation is not wholly consistent with what we find earlier in Gṛhyasūtra liter-
ature. In Jaiminigṛhyasūtra 1.19 (p. 18), for instance, we read:
trayaḥ snātakā bhavantīti ha smāhāruṇir gautamo vidyāsnātako vratasnātako vidyā-
vratasnātaka iti teṣām uttamaḥ śreṣṭhas tulyau pūrvau.
Caland (1922, 32) translates:
According to Âruṇi Gautama there are three kinds of Snātakas: the Snātaka by knowl-
edge, the Snātaka by the completion of his observances, and the Snātaka by knowledge
and by the completion of his observances. Of these the last ranks foremost, the first two
are equal (to each-other).
Cf. Pāraskaragṛhyasūtra (kāṇḍa 2, kaṇḍikā 5, sentences 32–35, p. 220):
trayaḥ snātakā bhavanti: vidyāsnātako vratasnātako vidyāvratasnātaka iti 32 samāpya
vedam asamāpya vrataṃ yaḥ samāvartate, sa vidyāsnātakaḥ 33 samāpya vratam as-
amāpya vedaṃ yaḥ samāvartate, sa vratasnātakaḥ 34 ubhayaṃ samāpya yaḥ samāvar-
tate, sa vidyāvratasnātaka iti 35
via free access
70 goodall
Niśvāsa certainly, and probably throughout the early Mantramārga, the use
of vidyāvrata to mean “observance for [the propitiation of] a mantra” seems
to be the norm. As Sanderson has observed in the note of his that we have
just quoted, vidyāvrata seems indeed to be used in the same way as pūrva-
sevā in the Niśvāsa. Many short paragraphs of prose in the grimoires (kalpa)
that we find in the Guhyasūtra sketch out the essential features of particular
observances, and these paragraphs are very often concluded with a succinct
statement of the magical powers that can be won by following them (the power
to fly, for example, or to disappear); but sometimes we find instead the assertion
that the observance fulfils the requirements of pūrvasevā (10.27, 10.99, 14.26)
or puraścaraṇa (14.24) or, as here in Guhyasūtra 10.91, the requirements of the
vidyāvrata:
devaṃ pūjyāgnau juhuyād audumbarasamidhānāṃ tryaktānāṃ sahas-

raṃ tṛsandhyaṃ. kṣīrāśī saptā dināni juhuyāt. cīrṇṇavidyāvrato bhavati.
Having worshipped the Lord, he should oblate into the fire at the three
junctures of the day a thousand pieces of Udumbara-wood smeared with
the three [sweet substances]. Consuming [only] milk, he should make
oblations [in this manner] for seven days. He will become one who has
accomplished the vidyāvrata.
Before I wrap up the discussion, it should be mentioned that a different hypoth-

esis as to the meaning of vidyāvrata was advanced some years ago, when many
of the above-cited passages had not yet come to light, by Judit Törzsök when
translating chapter 10 of the Siddhayogeśvarīmata. We quote here the begin-
ning of the chapter (without the apparatus) from Törzsök’s 1999 edition:
devy uvāca
mayā deva purā pṛṣṭaṃ vratayāgavivarjitam
siddhayogeśvarīṇāṃ tu mataṃ mantraprasādhakam 1
kiṃ tu deva pratijñātaṃ siddhir vidyāṅgasaṃsthitā
tasmāt teṣu samāsena vratacaryāṃ bravīhi me 2
bhairava uvāca
ādau tu sarvasiddhyarthaṃ sarvavighnavināśanam
sarvapāpāpanodārthaṃ vidyāvrataṃ samārabhet 3
sādhakaḥ sādhakī vātha mantratadgatacetasaḥ
yāgaṃ kṛtvā vidhānena vratacaryāṃ samācaret 4
bhasmalepitasarvāṅgo maunī śuklāmbaraḥ sudhīḥ
sitayajñopavītaś ca akāmo niyame sthitaḥ 5
via free access
Here is Törzsök’s translation (1999*, 143):
The Goddess spoke

I have previously asked you about the Doctrine of the Yoginīs (Siddha-
yogeśvarīmata), O God, which helps to make mantras effective
(mantraprasādhakam) without any observances or worship. (1)
However, you have asserted, O God, that success depends on the ancil-
lary mantras; therefore, tell me briefly about how to practise the
observance[s] associated with them (teṣu). (2)
Bhairava spoke
First [before any other practice to attain a specific supernatural power],
for all kinds of supernatural powers, [and] for expiatory purposes,
one has to start the observance of the [ancillary] mantras, which
destroys all obstacles. (3)
The male or female practitioner, with his/her mind focused on the
mantra, should perform worship according to prescriptions and then
undertake the vow (vratacaryāṃ). (4)
[In the first of these] all his limbs covered with ashes, the practitioner is
to observe silence and should wear a white garment; he should be of
good understanding. He must have a white sacred thread, he should
be free from desire and established in self-restraint. (5)
Now the reason that Törzsök translates vidyāvrata with “the observance of
the [ancillary] mantras” is that each of the vratas in the chapter is specific
to the cultivation of a particular auxiliary mantra—the first one, given above
in verse 5, must, by elimination, be an observance for the hṛdaya-mantra—
and those auxiliary mantras belong to a set known in this work and in others
as the vidyāṅgamantras.68 It was therefore reasonable for her to assume that
vidyāvrata was short for vidyāṅgavrata, for she had no evidence to suggest oth-
erwise, and she had parallel evidence that seemed to reinforce this hypothesis,
namely the testimony of Svāyambhuvasūtrasaṅgraha 21 (from which Törzsök
quotes, citing Alexis Sanderson’s collation, in her notes on p. 78).
That passage again gives a series of vratas, which are, by the way, again not
instances of imitatio dei, and which are again specific to the vidyāṅgamantras;
so it is wholly understandable that this seemed to Törzsök in 1999 to confirm
68 For these mantras and their individual names, see Brunner 1986 and, more recently,
Goodall 2004, 222–223.
via free access
72 goodall
the notion that the element vidyā in the collocation vidyāvrata must refer to
the vidyāṅgamantras. I think, however, that it will now be clear that chapter 21
of the Svāyambhuvasūtrasaṅgraha might in fact be interpreted equally well
internally if we assume that vidyāvrata is used instead to mean “observance
for [propitiation of] a mantra” and that, given the other attestations of the term
that we now know about in, for instance, the Guhyasūtra, the Mataṅga, and the
Sarvajñānottara, it actually makes better sense to assume this broader interpre-
tation in this passage too.
There is somewhat better evidence for pinpointing the place of the Svāyam-
bhuvasūtrasaṅgraha in a relative chronology of Saiddhāntika writings than
there is for most other pre-tenth-century Siddhāntatantras, for in terms of both
doctrinal and social developments, it seems later than the sūtras of the Niś-
vāsatattvasaṃhitā (see Goodall, Sanderson, Isaacson et al. 2015, 41–44, 47–50,
58), and yet it cannot be later than Sadyojyotiḥ, who has written a commen-
tary upon it and who, Sanderson argues (2006b, in particular p. 76), lived
between c. 675 and 725ad. The edition is not widely accessible, which may
be why Wedemeyer did not refer to this passage, and its text almost invari-
ably needs to be corrected against manuscripts, but this particular chapter has
just been published anew, in the form in which it appears when quoted by
Hṛdayaśiva in his Prāyaścittasamuccaya (see Sathyanarayanan 2015, Appendix
chapter 10).69
japtvā vrataiś ca vidhivat snātaḥ siddhyai japet manum 10.27

na vilambitam aspaṣṭan na cāsvīkṛtam adrutam
nāsaṃkhyaṃ na manobhrāntaṃ japaṃ kuryād vicakṣaṇaḥ 10.28
sitavāsāḥ sitoṣṇīṣī sitayajñopavīty api
sitānulepanasragvī cared vidyādhipavratam 10.29
raktāmbaradharo mantrī raktamālyānulepanaḥ
māsam ekam puroktañ ca cared brahmaśirovratam 10.30
pītavāsāś caren māsaṃ pītamālyānulepanaḥ
pītayajñopavītī ca rudrāṇyā vratam uttamam 10.31
vrataṃ puruṣṭutasyāpi māsam ekaṃ cared budhaḥ
sarvakṛṣṇopacāreṇa śivārcanarataḥ sadā 10.32
69 In the collation below, H is the reading of Sathyanarayanan’s transcription of the twelfth-

century manuscript transmitting Hṛdayaśiva’s work (where this chapter is the tenth); Ed.
is of course the Mysore edition of 1937 (where the chapter is the twenty-first), and N
marks the readings of the old Nepalese palm-leaf manuscript filmed by the NGMPP on
Reel A 30/6.
via free access
citrāmbaradharo dhīraś citramālyānulepanaḥ

sarvasiddhiparo mantrī caret pāśupataṃ vratam 10.33
vratavratasamāptau ca kalaśena śivāmbhasā
svamantraparijaptena svātmānam abhiṣecayet 10.34
evaṃ vidyāvratasnātaḥ sarvatrādhikṛto ’naghaḥ
japen mantram anudvignaḥ svakalpavidhinā tataḥ 10.35
27c vrataiś] H, Ed.; vrajaiś N ● 27d manum] H, Ed.; matam N ● 28a na vilambitam] H, Ed.;
avilambitam N ● 28c manobhrāntaṃ] H; mobhrāntaṃ N (unmetrical); manobhrānta°
Ed. ● 29ab sitavāsāḥ sitoṣṇīṣī sita°] em.; sitavāsā śitośnīṣī sita° N; śitavāsā śitoṣṇīṣī śita°
H; sitavāsāḥ sitoṣṇīṣaḥ sita° Ed. ● 29c sitānulepanasragvī] conj.; sitānulepanaḥ sragvī N,
Ed.; śitānulepanaḥ sragmī H ● 29d vidyādhipa°] Ed.; vidyādhipaṃ N, H ● 30a raktā°] H,
Ed.; rattā° N ● 31a pītavāsāś] N, Ed.; pītavāsā H ● 31c rudrāṇyā] N, Hac, Ed.; rudrāṇī° Hpc
● 32a puruṣṭutasyāpi] N, H; puruhūtasyāpi Ed.pc; purutasyāpi Ed.ac ● 33c sarvasiddhi°]
N, H; sarvasiddha° Ed. ● 33d pāśupataṃ] N, H; pāśupata° Ed. ● 34a vratavratasamāptau

ca] Ed.; vrataṃ vratasamāptau ca N; vratavrataṃ samāptau tu H ● 35a vidyāvrata°] N, H;
vidyāvrataḥ Ed. ● 35b sarvatrādhikṛto] N; sarvatrāvikṛto H; sarvato dhikṛto Ed.
Having recited [a particular mantra] along with [the practice of one of

the] observances in accordance with the rules, and having bathed [at
the end of the observance], one may recite that mantra for attaining
supernatural powers. (27cd)
The skilled practitioner should do his recitation not too slowly, not
indistinctly, not without taking [the meaning of what he recites]
in, not too fast, not without counting, and not with his thoughts in
confusion. (28)
Dressed in white, with a white turban and a white sacred thread and
white unguents and garland, he should perform the observance for
the vidyādhipa-mantra. (29)
Dressed in red garments and red garlands and unguents the Mantrin
should first perform for one month the stated observance for the
brahmaśiraḥ. (30)
Wearing yellow garments and yellow garlands and unguents and a yel-
low sacred thread he should perform the excellent observance of
rudrāṇī for a month. (31)
The competent ritualist (budhaḥ), constantly devoted to the worship of
Śiva, should perform the observance for puruṣṭuta for one month
with all accoutrements being black. (32)
The mantrin, intent on attaining all manner of special powers, should
perform the observance for the pāśupatāstra resolutely (dhīraḥ)
dressed in multi-coloured garments and with multicoloured garlands
and unguents. (33)
via free access
74 goodall
And upon the completion of one or another of these observances

(vratavratasamāptau),70 he should pour upon himself Śiva-water
that has been consecrated by recitation of his mantra over it from a
pot. (34)
Being thus bathed after the observance [in propitiation] of [his] mantra,
invested in the right to [pursue] all [manner of special powers], fault-
less, he should then recite [his chosen] mantra according to the rules
of his hand-book,71 without being afraid. (35)
In the above lines, 27d and 35 seem to make particularly plain that these vratas
are performed by way of pūrvasevā, also known as vidyāvrata, as a preliminary
to the pursuit of siddhi.
Csaba Kiss, following Judit Törzsök’s lead, has also alluded to the link
between vidyāṅgamantras and the name vidyāvrata, but I think it will be clear
from what he says below that the evidence that his new edition of parts of the
Brahmayāmala has recently brought to the discussion again supports rather
the broader interpretation, in which the element vidyā alludes to any mantra,
not just a vidyāṅgamantra. Below are his remarks (2015, 211) on Brahmayāmala
21.4–5b, which he constitutes and translates as follows:
ete nava vratā proktā vidyābhede vyavasthitā

eteṣāṃ tu yathānyāyaṃ yathā caryā bhave tv iha
kathayāmi mahādevi tan me nigadataḥ śṛṇu
These are the nine ascetic observances (vrata), corresponding to [the

syllables of] the Vidyā[, Caṇḍā Kāpālinī’s nine-syllable mantra]. I shall
now teach you how to perform them correctly, O Mahādevī. Listen to me
[while I] teach you.
… these observances are in fact called the vidyāvratas in 21.10c, 42d, 47a,
51b, 53b, 75d and 102d; the nine types of observances obviously corre-
spond to the nine syllables of the Navākṣaravidyā (oṃ caṇḍe kāpālini
svāhā), taught in BraYā 2; vidyāvratas may also serve, as seen in many
tantric texts, as preliminary propitiation, by the use of vidyāṅgamantras,
of a mantra to be applied later, or simply for the purpose of gaining
70 The text and interpretation are not certain here. A possible conjectural emendations that
suggests itself is vratī vratasamāptau.
71 For this sense, see Tāntrikābhidhānakośa, vol. 2, s.v. kalpa.
via free access
mastery over the Vidyā, similarly to the way it is taught in the Yoginīsaṃ-
cāra; as Sanderson (2009:134 n. 311) remarks: …
More closely parallel to these nine observances, as we can now see, are the nine
taught in chapter 3 of the Guhyasūtra, which are probably each for one of the
nine elements of the navātman. So perhaps the association with vidyāṅga-
mantras is simply a red herring.
Kiss’s reference there to Sanderson 2009, 134, note 311 proves to be another
passage in which Sanderson reveals that, although he did not spell out every
detail of his assumptions and the evidence upon which he based them, he had
in fact already assumed the interpretation for which we have been somewhat
long-windedly arguing here, both of vidyāvrata and of its relation to individual
named vratas:
The Yoginīsaṃcāra requires anyone who has gone through its initiation
ceremony and then received consecration (abhiṣekaḥ) to adopt one of
three forms of ascetic observance in order to gain mastery over the Vidyā
(vidyāvratam): the Bhairavavrata, the Cāmuṇḍāvrata, or the Triṣaṣṭikula-
vrata, the observance of the sixty-three families [of the Mothers], which
it also calls the Kāpālavrata, i.e. the Kāpālika.
So let me reiterate my conclusion: we should probably assume that, even if

the expression vidyāvrata was originally drawn from the common brahman-
ical expression vidyāvratasnāta that we saw in Manusmṛti 4.31, and even if
it may have been subsequently coloured in some contexts by other associa-
tions of the word vidyā (vidyāṅgamantra, “knowledge,” “consort”), the expres-
sion vidyāvrata appears throughout the early Mantramārga to be used with
the understanding that it refers primarily to an “observance for [the prelimi-
nary propitiation of a] mantra.” While some works (such as the Saiddhāntika
Kiraṇatantra, and the Sarvajñānottaratantra) appear to mention only one way
of fulfilling the requirements of the vidyāvrata, many others (Niśvāsa, Yoginī-
saṃcāra, Brahmayāmala, Svāyambhuvasūtrasaṅgraha, Siddhayogeśvarīmata,
Tantrasadbhāva, Kubjikāmata) teach several vratas, not all of which are equally
transgressive and not all of which involve imitatio dei, as alternative ways of
realising the vidyāvrata.
Finally, let me show how this assumption seems to me to throw light even
on passages where none of the terms that we have been discussing actually
occur. The first few times I read the beginning of chapter 9 of the Guhyasūtra, it
seemed to begin with a curious non sequitur: the goddess asks a question about
how the alphabet, treated as mantra, can be used to bring about supernatural
via free access
76 goodall
power and liberation; Śiva’s reply, however, first explains at some length how
someone should dress half his body as the goddess and the other half as Śiva.
How could this, I asked myself, be an answer to the goddess’ question? Was the
apparent incoherence of the text here an indication that it had become cor-
rupt?
devy uvāca—
mātṛkāyā bhavet siddhir mmokṣañ caiva maheśvara
mātṛkāsiddhim ākhyāhi mokṣañ caiva yathā bhavet 9.1
tatsamutthāś ca ye mantrāḥ kimarthaṅ kathitās tvayā
etat praśnavaraṃ brūhi ((bha))---phalaṃ hi me 9.2
īśvara u—
arddhastrīveśadhārī tu arddhena puruṣas tathā
arddhena alakaṃ kuryād arddhenaiva jaṭādharaḥ 9.3
tilakārddhena netrārddhe vālikā hy ekakarṇṇake
kuṇḍalaṃ hy ekakarṇṇe tu śūlan dakṣiṇahastataḥ 9.4
vāmapārśve72 stanaṅ kuryād vāmārdhe caiva mekhalām
valayaṃ vāmahaste tu vāmapāde tu nūpuraṃ 9.5
rucakaṃ dakṣiṇe pāde muñjamālāṃ tathā kaṭau73
kaupīnan dakṣiṇe kuryād vāme strīvastradhāritā 9.6
śūrppaṃ vāmakare gṛhṇed ardhanārīśvaravrate74
etad vrataṅ gṛhītvā tu bhikṣāśī tu jitendriyaḥ 9.7
japahomarato nityam pratigrahavivarjjitaḥ
triṣkālam75 arccayed devaṃ triṣkālaṃ snānam ācaret 9.8
śākayāvakabhikṣāśī skandamūlaphalāśinaḥ76
māsam e[[ka]]---samanvitaḥ 9.9
mucyate ’sau77 mahāpāpāt kṣudrasiddhiñ ca vindate
dvimāsān madhyamā siddhir abdārddhād uttamā bhavet 9.10
saṃvatsareṇa siddhis tu vidyāsiddhim avāpnuyāt
aṇimādyās tu jāyante siddhaiś ca saha modate 9.11
īpsitāṃ labhate kāmān akāmo mokṣam āpnuyāt
72 vāmapārśve] KW; vāmapārśe N

73 dakṣiṇe pāde muñjamālāṃ tathā kaṭau] conj.; dakṣiṇe pa—ñjamālāṃstathā kaṭe N; dak-
ṣiṇe ⊔ muñjamālāṃ tathā kaṭau K; dakṣiṇe ⊔ṇṭhamālāṃstathā kaṭe W
74 °nārīśvara°] conj.; °nārīśvaraṃ NW; °nārīśvare K
75 triṣkālaṃ] KW; triṣkāla° N
76 skanda°] NW; kanda° K
77 mucyate ’sau] em.; mucyate so NW; mucyate sa K
via free access
The Goddess spoke:

From the mātṛkā supernatural power and liberation can come about,
O Lord. Tell me [how to attain] supernatural power and liberation
through the mātṛkā. (1)
Why did you teach the mantras that arise from it? Tell me [the answer
to] this excellent question. … fruit to me. (2)
The Lord spoke:
Wearing half the dress of a woman and half [that of] a man, on one half,
he should place [feminine] tresses, on one half, he should wear mat-
ted locks. (3)
On one half, there should be a forehead mark; on one half a [forehead]
eye. A ring (vālikā) [should be] in one ear; a [pendant] ear-ornament
(kuṇḍalam) in one ear. He should put a trident (śūlam) in his right
hand and a breast on his left side, a girdle (mekhalām) on the left
half, a bangle on the left arm, a woman’s anklet on the left leg, a man’s
anklet on the right leg and a muñja-grass belt. At the hips, he should
put a loin-cloth on the right and wear a woman’s garment on the left.
(4–6)
In the left hand, he should hold a winnowing fan in the observance of
Ardhanārīśvara. Adopting this observance he should eat alms, keep
his senses under control, be devoted to regular obligatory recitation
and oblation, rejecting the receipt of gifts. (7–8b)
He should venerate God three times [a day] and perform ablutions
three times [a day]. Eating vegetables and barley-gruel, eating bulbs
(skanda° [= kanda°]), roots and fruits, for one month … (9)
He will be released from [the retributive force of] major transgressions;
and he will attain low siddhis after two months, middling siddhis after
half a year and high siddhis after a year; he will attain power over the
spell (vidyāsiddhim). (10–11b)
The ability to make himself atomic, along with the others [of the yogic
powers], will arise. He will take pleasure in the company of siddhas.
He will attain the wishes he desires; if he is without desires, he will
attain liberation. (11c–12b)
Although the words vrata, caryā and vidyāvrata are none of them to be found, it
is now clear to me that this passage makes implicit allusion to the structure now
familiar to us from numerous other passages: the sādhaka propitiates a given
mantra, here the mātṛkā, by performing a timed religious observance involv-
ing unusual dress and diet, the rules of which are ideally held to be in some way
appropriate to the mantra in question, and then becomes eligible for the pur-
via free access
78 goodall
suit of particular siddhis. In the case of the mātṛkā, adopting the appearance
of Ardhanārīśvara is particularly appropriate because the mātṛkā is made
up of feminine vowels and masculine consonants, which, as is explained else-
where in the Niśvāsa-corpus, are to be applied respectively to the left and right
halves of the sādhaka’s body before worship in a preliminary rite that prefigures
what came to be called sakalīkaraṇa.78 The imitatio dei that is such a promi-
nent feature of some observances, such as this one, now seems as if it should
be more precisely characterised as identification with the mantra-deity being
propitiated.
Conclusion
So what can be learned from the foregoing pages? In the beginning of this
paper, I tried to emphasise the layered structure not only of the Niśvāsa-
tattvasaṃhitā, but also of the Guhyasūtra itself. This incidentally means that
we should not only be, as always, cautious in proposals for dating both this and
related literature, but that we should perhaps also allow for a broad fourchette
for the composition of this work, broad enough to cover the periods of compo-
sition of other related works.79 We should also bear in mind, while attempting
to model the relative chronology of early Tantric literature, that it is the very
latest layers of the Guhyasūtra that provide the closest parallels with the gri-
moires of the Buddhist kriyātantras, in particular with the final chapter, 55, of
the Mañjuśriyamūlakalpa (see Goodall and Isaacson 2016, passim).
We have also learned, I believe, about a further early stage in the history
that precedes the familiar idea that Śaiva scriptures were divided into four sec-
tions treating knowledge, ritual, yoga and day-to-day behaviour ( jñāna, kriyā,
yoga, caryā). In the period of the redaction of the Niśvāsa, initiates were, de
facto, all sādhakas seeking to harness the power of mantras and their caryā
was not a matter of approved day-to-day behaviour, or “comportment,” but
rather of vrata-caryā, the performance of timed religious observances. Such
timed religious observances could be used, as in many other religious tradi-
tions, for expiation, but their primary use in the early Mantramārga was for
78 See Uttarasūtra 2.8 and annotation on pp. 351–352 of Goodall, Sanderson, Isaacson et al.
2015.
79 For the possibility that, for instance, the Rauravasūtrasaṅgraha might have influenced
the cosmographical chapters of the Guhyasūtra, see Goodall 2016, 89ff. For the most
recent discussion of the dating of the layered corpus that is the Niśvāsatattvasaṃhitā, see
Goodall, Sanderson, Isaacson et al. 2015, 30–73.
via free access
mantra-propitiation prior to other activities involving the mantras. Such pre-

liminary mantra-propitiation could also be referred to by such expressions as
vidyāvrata, puraścaryā and pūrvasevā. These notions about mantra-use and
mantra-propitiation may be found reflected in a wide range of post-Niśvāsa
pre-tenth-century writings. But once the intake of the religion had broadened
to include many who were not occupied with spell-mastery at all, the term
caryā in the tetrad of tantric topics also regained its (usually non-technical)
sense of day-to-day “comportment.” This shift in usage had taken place by the
time of the composition of the Kiraṇa, in other words by the beginning of the
ninth century at the latest.80 Finally, the early Śaiva evidence furnished here
suggests that this nexus of notions and labels is not such a fertile field as might
have been supposed for those searching for evidence of instances of Buddhist
influence upon the early Śaiva Mantramārga.
Acknowledgements
I am grateful to Harunaga Isaacson for the suggestions he made for improving

this article.
References
Primary Sources
Kiraṇatantra. (ED) Ti. Rā. Pañcāpageśaśivācārya and K.M. Subrahmaṇyaśāstrī, eds. śrī-
mat kiraṇāgamamahātantram. Śivāgamasiddhāntaparipālanasaṅgha, no. 16. Deva-
kōṭṭai, 1932.
Kiraṇatantra. National Archives, Kathmandu, MS 5–893, NGMPP Reel No. A 40/3, a
tenth-century Nepalese palm-leaf manuscript (most recently described by Goodall
1998, lxxxiv–lxxxv).
Kubjikāmata. Teun Goudriaan and Jan A. Schoterman, eds. The Kubjikāmatatantra.
Kulālikāmnāya Version. Critical Edition. Orientalia Rheno-Traiectina, no. 30. Leiden;
New York; København; Köln: Brill, 1988.
Jaiminigṛhyasūtra. W. Caland, ed. The Jaiminigṛhyasūtra belonging to the Sāmaveda with
extracts from the commentary edited with An introduction and translated for the first
time into English. The Punjab Sanskrit Series 2. Lahore: Moti Lal Banarsi Dass, 1922.
80 For the observation that the Kiraṇa belongs to a group of scriptures whose teachings can
be found paraphrased in the Haravijaya, which was composed in Kashmir around 830ad,
see Sanderson 2001, 5–6.
via free access
80 goodall
Tantrasadbhāva. Electronic edition of Mark Dyczkowski based on NGMPP Reel Nos.

A 188/22, A 44/1, and A 44/2.
Niśvāsakārikā. Institut Français de Pondichéry, T. 17, 127 and 150. Paper transcripts in
Devanāgarī script. (I am grateful to S.A.S. Sarma, R. Sathyanarayanan, Nibedita Rout
and Nirajan Kafle, all of the Pondicherry Centre of the EFEO, for their help produc-
ing electronic transcriptions of these witnesses.)
Niśvāsatattvasaṃhitā. For the Mūlasūtra, Uttarasūtra and Nayasūtra, see Goodall,
Sanderson, Isaacson et al. 2015.
For the Guhyasūtra: NAK MS 1–227, NGMPP Reel No. A 41/14. Palm-leaf, early Nepalese
‘Licchavi’ script. Two apographs have been consulted throughout, both in Devanā-
garī and on paper: NAK MS 5–2401, NGMPP Reel No. A 159/18, and Wellcome Insti-
tute for the History of Medicine, London, Sanskrit MS I.33. The verse and chapter
numeration used in our annotation is that of Goodall’s edition in progress. Diwakar
Acharya, Peter Bisschop and Nirajan Kafle helped Goodall to produce the first com-
plete transcription.
Niśvāsamukhatattvasaṃhitā. See Kafle 2015*.
Pāraskaragṛhyasūtra. Mahādeva Gangādhar Bākre, ed. Grihya-sūtra by Parashar with
five commentaries of Karka Upādhyāya, Jayarām, Harihar, Gadādhar and Vishvanāth
as well as appendices called Vāpyādi-pratisthā Kandikā with Kāmdeva Bhāshya
Showcha Sūtra, Snāna Sūtra with Harihar Bhāshya and Shradha Sūtra with three com-
mentaries by Karka, Gadādhara and Shrādha Kāshika by Krishnamishra and Bhojana
Sūtra. Reprint, New Delhi: Munshiram Manoharlal, 1982 [Bombay: Gujarati Printing
Press, 1917].
Bodhāyanagṛhyasūtra. R. Shama Sastri, ed. The Bodhâyana Grihyasutra. University of
Mysore Oriental Library Publications Sanskrit Series, no. 32/55. Mysore: University
of Mysore, 1920.
Mataṅga. N.R. Bhatt, ed. Mataṅgapārameśvarāgama (Kriyāpāda, Yogapāda et Caryā-
pāda) avec le commentaire de Bhaṭṭa Rāmakaṇṭha. Publications de l’Institut Français
d’Indologie, no. 65. Pondicherry: Institut Français d’Indologie, 1982.
Mohacūḍottara. National Archives, Kathmandu, MS 5–1977, NGMPP Reel No. A 182/2.
This is a 1926 paper apograph manuscript in Devanāgarī script of a twelfth-century
palm-leaf manuscript from Western India: National Archives, Kathmandu, MS 1–
1633, NGMPP Reel No. B 26/29 (for which see Tāntrikābhidhānakośa 2, s.v. gaṇa).
Vyomavatī. Gaurinath Sastri, ed. Vyomavatī of Vyomaśivācārya [Part Two]. M.M. Śiva-
kumāraśāstri-Granthamālā, no. 6. Varanasi: Sampurnanand Sanskrit University,
1984.
Sarvajñānottara. Edition in progress by Dominic Goodall, based on various sources
including IFP T. 334 [= L], Madras GOML R 16829 [= M], the early Nepalese manu-
script (NGMPP A 43/12 [= N]), and manuscripts transmitting the twelfth-century
commentary of Aghoraśiva.
via free access
Svāyambhuvasūtrasaṅgraha, Veṅkaṭasubrahmaṇyaśāstrī, ed. Svāyambhuvasūtrasaṅ-

graha. Mysore, 1937.
Svāyambhuvasūtrasaṅgraha. (N) National Archives, Kathmandu, paṃ 348, NGMPP
Reel No. A 30/6, a Nepalese palm-leaf manuscript. See also Appendix to Sathya-
narayanan 2015.
Secondary Sources
Brunner, Hélène. 1985. Mṛgendrāgama Section des rites et section du comportement avec
la vṛtti de Bhaṭṭanārāyaṇakaṇṭha. Publications de l’Institut Français d’Indologie,
no. 69. Pondicherry: Institut Français d’Indologie.
Brunner, Hélène. 1986. “Les membres de Śiva.” Asiatische Studien/Études Asiatiques XL
(2): 89–132.
Brunner, Hélène. 1992. “The Four Pādas of Śaivāgamas.” The Journal of Oriental Research,
Madras. Dr. S.S. Janaki Felicitation Volume vols. 56–72 (1986–1992): 260–278.
Caland, Willem. 1922. See Jaiminigṛhyasūtra.
Goodall, Dominic. 1998. Bhaṭṭa Rāmakaṇṭha’s Commentary on the Kiraṇatantra. Vol-
ume I: Chapters 1–6. Critical Edition and Annotated Translation. Publications du
département d’indologie 86.1. Pondicherry: Institut Français de Pondichéry/École
française d’Extrême-Orient.
Goodall, Dominic. 2004 The Parākhyatantra, a Scripture of the Śaiva Siddhānta. A Crit-
ical Edition and Annotated Translation. Collection Indologie, no. 98. Pondicherry:
Institut Français de Pondichéry/École française d’Extrême-Orient.
Goodall, Dominic. 2016. “How the Tattvas of Tantric Śaivism Came to Be 36: The Evi-
dence of the Niśvāsatattvasaṃhitā.” In Tantric Studies: Fruits of a Franco-German
Collaboration on Early Tantra, edited by Dominic Goodall and Harunaga Isaacson,
77–111. Collection Indologie, no. 131/Early Tantra Series, no. 4. Pondicherry: Insti-
tut Français de Pondichéry/École française d’Extrême-Orient/Asien-Afrika-Institut,
Universität Hamburg.
Goodall, Dominic, Alexis Sanderson and Harunaga Isaacson. 2015. The Niśvāsatatt-
vasaṃhitā. The Earliest Surviving Śaiva Tantra, Volume 1, A Critical Edition & Anno-
tated Translation of the Mūlasūtra, Uttarasūtra & Nayasūtra, edited by Dominic
Goodall, in collaboration with Alexis Sanderson & Harunaga Isaacson, with contri-
butions of Nirajan Kafle, Diwakar Acharya & others. Collection Indologie, no. 128/
Early Tantra Series, no. 1. Pondicherry: Institut Français de Pondichéry/École fran-
çaise d’Extrême-Orient/Asien-Afrika-Institut, Universität Hamburg.
Goodall, Dominic, and Harunaga Isaacson. 2016. “On the Shared ‘Ritual Syntax’ of the
Early Tantric Traditions.” In Tantric Studies: Fruits of a Franco-German Collabora-
tion on Early Tantra, edited by Dominic Goodall and Harunaga Isaacson, 1–76. Col-
lection Indologie, no. 131/Early Tantra Series, no. 4. Pondicherry: Institut Français
de Pondichéry/École française d’Extrême-Orient/Asien-Afrika-Institut, Universität
Hamburg.
via free access
82 goodall
Isaacson, Harunaga. 2015. See Goodall, Sanderson, Isaacson et al. 2015.

Isaacson, Harunaga. 2016. See Goodall and Isaacson 2016.
Kafle, Nirajan. 2015*. The Niśvāsamukha, the Introductory Book Prefacing the Niśvāsa-
tattvasaṃhitā, Critically Edited and Furnished with an Annotated Translation and an
Appendix Giving an Edition of Chapters 5–9 of the Śivadharmasaṅgraha. Ph.D. thesis,
the University of Leiden.
Kiss, Csaba. 2015. The Brahmayāmala Tantra or Picumata, Volume Two. The Religious
Observances and Sexual Rituals of the Tantric Practitioner: chapters 3, 21 and 45. Col-
lection Indologie, no. 130/ Early Tantra Series, no. 3. Pondicherry: Institut Français
de Pondichéry/École française d’ Extrême-Orient/Asien-Afrika-Institut, Universität
Hamburg.
Nemec, John. 2014. Review of Wedemeyer 2013, Journal of Religion 94 (2): 271–273.
Olivelle, Patrick. 2005. Manu’s Code of Law. A Critical Edition and Translation of the
Mānava-Dharmaśāstra … With the Editorial Assistance of Suman Olivelle. New York:
Oxford University Press.
Sanderson, Alexis. 2001. “History through Textual Criticism in the Study of Śaivism, the
Pañcarātra and the Buddhist Yoginītantras.” In Les sources et le temps. Sources and
Time. A colloquium. Pondicherry 11–13 January 1997, edited by François Grimal, 1–47.
Publications du département d’Indologie, no. 91. Pondicherry: Institut Français de
Pondichéry/École française d’Extrême-Orient.
Sanderson, Alexis. 2006. “The Lākulas: New Evidence of a System Intermediate
Between Pāñcārthika Pāśupatism and Āgamic Śaivism.”Indian Philosophical Annual
24 (2003–2005): 143–217.
Sanderson, Alexis. 2006b. “The Date of Sadyojyotis and Bṛhaspati.” Cracow Indological
Studies 8: 39–91.
Sanderson, Alexis. 2009. “The Śaiva Age.” In Genesis and Development of Tantrism,
edited by Shingo Einoo, 41–349. Institute of Oriental Culture Special Series, no. 23.
Tokyo: Institute of Oriental Culture, University of Tokyo.
Sanderson, Alexis. 2015. See Goodall, Sanderson, Isaacson et al. 2015.
Sathyanarayanan, R., and Dominic Goodall. 2015. Śaiva Rites of Expiation. A First Edi-
tion and Translation of Trilocanaśiva’s Twelfth-century Prāyaścittasamuccaya (with a
transcription of Hṛdayaśiva’s Prāyaścittasamuccaya), critically edited & translated by
R. Sathyanarayanan, with an introduction by Dominic Goodall. Collection Indologie,
no. 127. Pondicherry: Institut Français de Pondichéry /École française d’Extrême-
Orient.
Tāntrikābhidhānakośa. 2004. Tāntrikābhidhānakośa II. Dictionnaire des termes tech-
niques de la littérature hindoue tantrique. A Dictionary of Technical Terms from Hindu
Tantric Literature. Wörterbuch zur Terminologie hinduistischer Tantren, ed. Hélène
Brunner†, Gerhard Oberhammer, André Padoux. Sitzungsberichte der philoso-
phisch-historischen Klasse 714, Beiträge zur Kultur- und Geistesgeschichte Asiens
44. Vienna: Österreichische Akademie der Wissenschaften.
via free access
Tāntrikābhidhānakośa. 2013. Tāntrikābhidhānakośa III. Dictionnaire des termes tech-

niques de la littérature hindoue tantrique. A Dictionary of Technical Terms from Hindu
Tantric literature. Wörterbuch zur Terminologie hinduistischer Tantren, sous la direc-
tion de Dominic Goodall et Marion Rastelli. Österreichische Akademie der Wissen-
schaften, Philosophisch-historische Klasse, Sitzungsberichte, 839. Band. Beiträge
zur Kultur- und Geistesgeschichte Asiens 76. Vienna: Verlag der Österreichischen
Akademie der Wissenschaften.
Törzsök, Judit. 1999*. ‘The Doctrine of Magic Female Spirits’. A Critical Edition of Selected
Chapters of the Siddhayogeśvarīmata(tantra) with Annotated Translation and Anal-
ysis. Ph.D. thesis, the University of Oxford.
Wedemeyer, Christian K. 2013. Making Sense of Tantric Buddhism. History, Semiology, &
Transgression in the Indian Traditions. New York: Columbia University Press.
via free access
via free access
part 2
Exegetical and Philosophical Traditions
via free access
via free access
chapter 4
Further Thoughts on Rāmakaṇṭha’s Relationship to

Earlier Positions in the Buddhist-Brāhmaṇical
Ātman Debate
Alex Watson
The present article is a continuation of my previous work on where precisely

to place Rāmakaṇṭha’s self-theory (ātmavāda) in the nexus of other rival posi-
tions. I am delighted to have been included in this volume and in the con-
ference which led to it, in honour of my former DPhil supervisor, Profes-
sor Alexis Sanderson, with whom I spent many hours reading Rāmakaṇṭha’s
Nareśvaraparīkṣāprakāśa (as well as Kumārila’s Ślokavārttika, ātmavāda, and
Vasubandhu’s Abhidharmakośabhāṣya, pudgalaviniścaya)—and indeed at
whose suggestion I began working on Rāmakaṇṭha’s philosophical texts.
A previous article of mine (Watson 2014) places Rāmakaṇṭha in the middle
ground between Nyāya and Buddhism. What I would like to do here is present
some considerations that run counter to that. I do not think they invalidate my
earlier contentions, but they do reveal them to be one-sided and incomplete.
In section 1 I introduce key issues in the self debate between Nyāya and Bud-
dhism, in order to then be able to locate Rāmakaṇṭha in relation to these two.
In section 2 I briefly explain my “middle ground” idea that was put forward in
the 2014 article.1̀ In section 3 I present evidence for seeing Rāmakaṇṭha as just
as extreme as Nyāya. In section 4 I present evidence for seeing him as being
even more extreme than Nyāya, with Nyāya being the moderate position. In
section 5 I present evidence for Nyāya not being so moderate after all. In the
concluding section I ask where all of this leaves us.
1 The Naiyāyika-Buddhist Debate about the Existence of the Self
As a means of identifying what precisely separated Nyāya and Buddhism2 on

the question of the self (ātman), I like to use a triple distinction, one that I had
1 In sections 1 and 2 some material is reused from the 2014 article.

2 I use “Buddhist” and “Buddhism” in this article to refer specifically to the Buddhism that
was engaged with in debate by Naiyāyikas and Śaivas, that is to say the Abhidharma of
© alex watson, 2020 | doi:10.1163/9789004432802_006

via free access
88 watson
figure 4.1 Self as essence
figure 4.2 Self as substance
not seen in secondary literature prior to my 2014 article, but which I think arises
naturally from the primary sources. This is a distinction between three ways in
which the self was envisaged by the Naiyāyikas, and three corresponding ways
in which the self was argued against by the Buddhists. To elaborate the idea, I
will use the same set of diagrams that I used in my 2014 article.
Figure 4.1 is intended to illustrate the difference between the Buddhist doc-
trine of momentariness, according to which we are something different in
every single moment, and the Brāhmaṇical idea of the self as the unchanging
essence of a sentient being. The self is something that remains the same—
both numerically and qualitatively—over time; a Buddhist individual is both
numerically and qualitatively different in each moment.
Whereas in figure 4.1 we were focusing on the conception of the self as
unchanging essence, in figure 4.2 we are dealing with the self as substance, rep-
resented by the large circle on the left—a substance that is the owner of certain
qualities (guṇa), represented by the small circles. For Naiyāyikas qualities can-
not exist without some substance in which they inhere; hence we can infer the
existence of the self from the existence of qualities such as cognition, plea-
Vasubandhu’s Abhidharmakośabhāṣya, and the Yogācāra and Sautrāntika of Dignāga, Dhar-

makīrti and their followers. What is said of Buddhism in this article does not, therefore, hold
true of Madhyamaka, for example.
via free access
rāmakaṇṭha’s relationship to earlier positions 89
figure 4.3 Self as agent
sure and pain.3 For Buddhists cognitions, feelings and the other constituents
(skandha) of an individual exist bundled together, without belonging to some
greater whole.
In figure 4.3 we are dealing with the conception of the self as agent; and we
have to distinguish two senses of “agent”—the doer (kartṛ) of actions, and the
subject of cognitions ( jñātṛ). Actions and cognitions (the circles can represent
either) all share a common agent for Nyāya, namely the self (represented by
the line). For Buddhism they do not: the agent of one particular action or cog-
nition will have ceased to exist by the time another action or cognition comes
into being.
The Nyāya and Vaiśeṣika concept of the agent of physical actions was some-
times elaborated by comparison to a puppeteer: as a puppeteer brings into
being movements of the body of the puppet, so the self brings bodily move-
ments into being.4 And the Naiyāyikas and Vaiśeṣikas both borrowed the Gram-
marians’ (Vaiyākaraṇas) definition of the agent as that which is autonomous,
i.e. that cause of action which does not depend on anything else.5 Buddhism
opposes this concept of an autonomous agent standing over the psycho-phys-
ical stream of events, manipulating it from above. Any physical action will
depend on the previous moment of consciousness, this will depend on what
causes it, and the latter will depend on what causes it, etc. There is nothing
3 The argument involves three contentions, each of which had their own supporting argu-
ments: (1) Qualities cannot exist without substances to which they belong; (2) Cognition,
desire, aversion, pleasure, pain, and volition are qualities; (3) The self is the only possible
substance to which these qualities could belong. See Nyāyavārttika ad 1.1.10, p. 62,12–18 (that
is to say, p. 62, lines 12–18—a convention used throughout this essay), and Praśastapāda-
bhāṣya p. 16,3–7. For the second stage of the argument in particular, see Nyāyavārttika ad
3.2.18, Nyāyamañjarī Vol. 2, p. 278,14–15 and Candrānanda ad Vaiśeṣikasūtra 2.2.28. For the
third stage of the argument, see e.g. Nyāyamañjarī, vol. 2, pp. 284,6–293,2 and Nyāyasūtra
3.2.47 with the commentaries ad loc. See also Chakrabarti 1982; Oetke 1988, 255–256, 258–
260, 280, 286–300, 359–360, 464; Matilal 1989, 74, 77; Matilal 1994, 286; Preisendanz 1994, 187,
209, 278–281; Kano 2001; and Watson 2006, 174–184.
4 See for example Praśastapādabhāṣya, p. 15,12 and Candrānanda ad Vaiśeṣikasūtra 3.2.4,
p. 28,18–19.
5 Aṣṭādhyāyī 1.4.54.
via free access
90 watson
here that is autonomous; and if there were it would not be able to be a cause
as it would stand outside the causal chain. So for Buddhism the agent of an
action is simply its principal cause, and this will be an event within the psycho-
physical stream.6 Buddhism thus replaces a two-tier model with a one-tier
one.
If we are talking not of a physical action but specifically of a cognition, its
agent will not be a previous moment of consciousness but rather the cognition
itself. The agent of a perception, i.e. the thing doing the perceiving, is the per-
ception itself. The subject of any act of awareness, i.e. that which is aware, is
the stream of consciousness at that particular moment. So here too we have
a contrast between a Naiyāyika two-tier model, in which the cognizer ( jñātṛ),
i.e. the self, is ontologically quite distinct from the cognition ( jñāna), versus a
Buddhist one-tier model in which the cognizer is pluralized and dispersed into
the stream of cognitions.
2 Rāmakaṇṭha in the Middle Ground
Where does Rāmakaṇṭha fit in all of this? On two counts he falls with Nyāya
and on two counts he falls with Buddhism. He falls with Nyāya (1) in oppos-
ing the Buddhist doctrine of momentariness and in upholding the existence
of a self that is the unchanging essence of the individual; and (2) in maintain-
ing that there is an autonomous, unconditioned agent of our physical actions.
But he agrees with Buddhism (1) that cognition does not inhere in something
other than itself: there is no self-substance over and above cognition; and (2)
that there is no cognizer over and above cognition, no agent of awareness over
and above awareness. For him, as for Buddhism, that which does the cogniz-
ing or perceiving is just cognition/awareness/consciousness ( jñātṛ/grāhaka =
jñāna).
How does he manage to reconcile all of these positions? I.e., how does he
manage to preserve the existence of an unchanging, autonomous self when he
denies that there is anything—any substance or agent—over and above cogni-
tion/consciousness? The answer is that he equates the self and cognition/con-
sciousness ( jñāna/saṃvit/cit/caitanya). This means that he has to maintain
that cognition/consciousness is permanent and unchanging, not plural and
momentary as it is for both Buddhism and Nyāya. Thus the three views can
be laid out on a continuum.
6 Abhidharmakośabhāṣya, p. 1228,1–1229,1.
via free access
figure 4.4 Rāmakaṇṭha’s view in the middle ground
For Nyāya there is both cognition, and a self over and above that. For Bud-
dhism there is no self, there is just cognition. For Rāmakaṇṭha there is a self,
but it is just cognition. Rāmakaṇṭha falls in the middle in the sense that, like
Buddhism, he as it were crosses out the line under Nyāya—he reduces the self
to cognition; but he does not go all the way down the path of Buddhist argu-
mentation, as he joins up the dotted line into one unbroken one: he argues that
cognition is permanent.
I will not here dwell on how Rāmakaṇṭha further elaborates and defends his
idea of unchanging cognition, as I have done so elsewhere.7
3 Rāmakaṇṭha as Extreme as Nyāya
It is at this point that I would like to question and supplement my earlier line
of thinking. What precisely is this continuum intended to capture? What deter-
mines the particular location on it that a position will occupy?
One thing it might be intended to capture is the amount of change that a
position posits in the subject of experience. On the left we have the Naiyāyika
position according to which the self is eternally unchanging. On the right we
have the Buddhist position according to which the subject is changing, both
qualitatively and numerically, in every moment. But for Rāmakaṇṭha the self
is completely unmodifiable (avikārya), so on this measure his position should
fall with Nyāya at the left hand extreme.
One thing it was certainly intended to capture is heaviness or lightness of
postulation. On the right we have the Buddhist position that postulates no
more than all the disputants agree exists: a sequence of cognitions. On the left
we have the Naiyāyika position that postulates an extra entity over and above
that: an eternal self that is the owner and knower of those cognitions. On this
7 Watson 2006, 333–382; 2010; and 2014, 186.
via free access
92 watson
measure it looks as though the Naiyāyika position should indeed fall to the left
of Rāmakaṇṭha’s: the former seems to postulate two things where Rāmakaṇṭha
postulates only one.
But even here there is a consideration that seems to negate this difference.
Rāmakaṇṭha actually accepts two very different kinds of “cognition” ( jñāna).
In Rāmakaṇṭha’s discussions with Buddhism, a common objection put by the
Buddhist pūrvapakṣin is: if cognition ( jñāna) is unchanging and single,8 as
you Saiddhāntikas claim, how can we account for the difference between a
cognition of blue, a cognition of yellow, an awareness of pain, an awareness
of pleasure, etc.? Rāmakaṇṭha’s usual answer is to maintain that throughout
such a sequence cognition itself is indeed single and unchanging.9 All change
happens on the object side of the subject-object (grāhaka-grāhya) divide. The
subject/perceiver, i.e. cognition or awareness, is like a light shining out always
in the same form; the red, blue, pleasure and pain are different objects that are
illuminated in turn by the light, but they do not affect the nature of the light (=
cognition) at all.
But sometimes he gives a different answer, namely that there are two dif-
ferent kinds of cognition, i.e. that the word jñāna can be used in two differ-
ent senses. There is cognition proper, i.e. the self (ātman), the perceiver (grā-
haka/jñātṛ), and there is the cognition that is located not in the self but in
the intellect (buddhi). The first is termed pauruṣaṃ jñānam,10 the cognition
of the self, or grāhakātmasaṃvit,11 cognition whose nature is the perceiver; the
second is termed adhyavasāyātmakaṃ jñānam12 / adhyavasāyātmikā saṃvit,13
cognition that is of the nature of determination, or parāmarśātma jñānam,14
8 “Single” here means single in each individual; the Saiddhāntikas, unlike the non-dualistic
Śaivas, maintained that each being’s self/consciousness was eternally separate from every
other’s.
9 See for example Nareśvaraparīkṣāprakāśa ad 1.6ab, pp. 26,19ff., passage beginning kas
tarhi nīlaprakāśāt pītaprakāśasya bhedaḥ? na kaś cit … (translated and discussed in Wat-
son 2006, 335–348); Mataṅgavṛtti, vidyāpāda introducing 6:34c–35a, p. 172,7ff. (translated
and discussed in Watson 2006, 349–382); Nareśvaraparīkṣāprakāśa ad 1.5, p. 14,2–18 (trans-
lated and discussed in Watson 2006, 220ff.), which has parallel passages at Mataṅgavṛtti,
vidyāpāda pp. 158,5–10, 172,16–21, 173,1–7 and Paramokṣanirāsakārikāvṛtti ad 43, p. 168,50–
169,62 (following the line numbers as printed on those pages); Nareśvaraparīkṣāprakāśa
ad 1.6ab, p. 26,4–13 (translated and discussed at Watson 2010, 111–112 and 2014, 186, note 26).
10 Kiraṇavṛtti ad 2.25ab, p. 54,3.
11 Mataṅgavṛtti, vidyāpāda p. 174,1–2 and 174,8.
12 Kiraṇavṛtti ad 2.25ab, p. 54,2.
13 Mataṅgavṛtti, vidyāpāda p. 174,4.
14 Mataṅgavṛtti, vidyāpāda p. 175,4.
via free access
cognition that is of the nature of verbal determination.15 The second kind is

not unitary and unchanging, but divided into discrete instances, verbalizable
as, for example, “this is blue,” “this is yellow,” “this is a pain.” If these are discrete
instances of cognition ( jñāna), how is this answer compatible with the first
answer that all change occurs on the side of objects, not on the side of the sub-
ject? The answer is that these discrete instances of determination are actually
objects of cognition in relation to the perceiver.16 If we are using “cognition” to
refer to the perceiver, they are objects of cognition; but they themselves can be
referred to as “cognitions” if we understand “cognitions” to mean determina-
tions (adhyavasāya, parāmarśa). These products of the intellect (buddhi) are
transient, i.e. plural over time; the nature of the self is unchanging and single
over time.
This is no innovation of Rāmakaṇṭha’s: both the earlier exegetes in his Sai-
ddhāntika tradition, and the Saiddhāntika scriptures themselves, locate jñāna
at two different places on the scale of tattvas that is partially inherited from
Sāṅkhya: within the material world (i.e. that which evolves out of māyā) at
the level of buddhi-tattva, where jñana features as one of the eight properties
(referred to in Śaiva literature with such terms as bhāva, dharma, guṇa, aṅga)
of the buddhi,17 and as the immaterial nature of selves (which are qualitatively,
but not numerically, identical to Śiva).
Bearing in mind this distinction between two different kinds of cognition,
let us revisit the situation that is illustrated in figure 4.4. That diagram, since it
prints the line under Rāmakaṇṭha on the same level as the dotted line under
Nyāya and the dotted line under Buddhism, makes the following assumptions:
(1) Cognition for Rāmakaṇṭha refers to (more or less) the same thing as cogni-
tion for Nyāya. (2) Cognition for Rāmakaṇṭha refers to (more or less) the same
thing as cognition for Buddhism. These were not unreasonable assumptions.
They reflect the way that Rāmakaṇṭha himself presents the situation. When
arguing against Nyāya, he effectively says: unlike you Naiyāyikas, we accept
only cognition, not some extra self over and above that in which it inheres.18
15 For an elaboration of Rāmakaṇṭha’s distinction between these two kinds of cognition, see
Watson 2006, 360–382.
16 See Kiraṇavṛtti ad 1:15, p. 18,33–35, Parākhyatantra 4:93ab, Tattvasaṅgraha of Sadyojyotis
13ab, Mṛgendratantra, vidyāpāda 10:23, Boccio 2002, and Watson 2006, 376ff.
17 See for example Parākhyatantra 4:74 ff., Mataṅgapārameśvara 17.2, Mṛgendratantra, vidyā-
pāda 10:24 (cited at Goodall 2004, 254, note 383), Bhogakārikā 55, and Sadyojyotis’ com-
mentary on Svāyambhuvasūtrasaṅgraha 2:12. This feature of Śaivism is inherited from
Sāṅkhya; see Sāṅkhyakārikā 23.
18 In the Nareśvaraparīkṣāprakāśa ad 1.4cd, pp. 8,17–12,19 (on which see chapter 1 of Watson
2006, and Watson 2010, 86–89), Rāmakaṇṭha aligns himself with Buddhism against not
via free access
94 watson
figure 4.5 Rāmakaṇṭha’s view as equivalent to Nyāya
And when arguing against Buddhism he assures us that his self/cognition is

just the jñāna that Buddhists talk about, not some container of it, some self-
substance to which it belongs.19
But now that we have seen that Rāmakaṇṭha actually recognizes two kinds
of cognition, different possible assumptions, and hence different possibilities
for the drawing of the diagram, emerge. We could take Buddhist jñāna as cor-
responding more to Rāmakaṇṭha’s plural cognitions. And if we do, then the fact
that Rāmakaṇṭha adds an extra layer above that makes him very much akin to
Nyāya—at least on this point of how many things they are postulating.
Similarly, we could take Rāmakaṇṭha’s first kind of cognition as compara-
ble to Nyāya’s self, and the second kind as comparable to Nyāya’s cognitions
( jñāna). When we adopt this perspective, the difference between Nyāya and
Rāmakaṇṭha regarding the extent of their postulation disappears. We arrive at
the situation depicted in figure 4.5.
To what extent is it reasonable to take the first kind of cognition as equiva-
lent to Nyāya’s self? It is reasonable to the extent that we are aligning Nyāya’s
self and Rāmakaṇṭha’s self, though the former is not of the nature of jñāna
whereas the latter is. To what extent is it reasonable to take the second kind
of cognition as equivalent to Nyāya’s cognitions and Buddhism’s cognitions?
The fact that the second kind is plural makes it a more natural equivalent than
the unitary first kind; but the fact that the second kind of cognition is uncon-
scious (a product of primal matter) makes it less equivalent to Buddhist and
Naiyāyika cognitions, which are sparks of awareness, than the first kind, which
is equivalent to awareness.
As well as the amount of change in the subject, and the amount of postu-
lation, there is a third possibility for what the continuum could capture: the
amount of dynamism in the subject. On this measure does Rāmakaṇṭha’s self
only the Naiyāyikas, but also the Vaiśeṣikas and Sāṅkhyas, on the grounds that they all
accept a self over and above cognition, whereas he accepts only cognition.
19 See Nareśvaraparīkṣāprakāśa ad 1.5, pp. 13,5–14,2 (discussed and translated in Watson
2006, 213–220); and Kiraṇavṛtti ad 2:25ab, p. 53,4–8 (discussed and translated in Watson
2010, 87–89).
via free access
fall closer to a Buddhist stream of consciousness than a Naiyāyika self does? Yes.
Nyāya’s rock-like20 self is in its own nature devoid of consciousness and agency;
Rāmakaṇṭha’s is of the nature of an outpouring—the outpouring of the light
of consciousness—and during liberation it remains not just conscious, but also
an agent. But Rāmakaṇṭha’s self is not as dynamic as that of the non-dualistic
Śaivas. For them vimarśa, representative cognition, is a power (śakti) of the
self.21 Rāmakaṇṭha resists this move, since for him (as for them) there is no dif-
ference between a power and the thing that has the power (śakti and śaktimat),
so to accept that vimarśa, which is changing, is a power of the self would have
been to accept some change in the self. Vimarśa for him belongs in the buddhi,
not the self;22 so it does not affect the unchanging nature of the self. Rāmakaṇ-
ṭha was more concerned than the non-dualist Śaivas to protect the self from any
change, though his self does have more dynamism than a Naiyāyika one. There
is a tension in Rāmakaṇṭha’s self between its lack of change and its dynamism,
one to which we will return in the final paragraphs of the article.
4 Rāmakaṇṭha More Extreme Than Nyāya
We have seen that by two out of these three measures, Rāmakaṇṭha’s self
looks to be just as “extreme” as Nyāya’s. In fact there is one consideration that
arguably makes it more extreme. Rāmakaṇṭha’s self is perhaps even further
removed from change than Nyāya’s, given that for Nyāya changing cognitions
20 See Nyāyamañjarī vol. 2 p. 432,4: jaḍaḥ pāṣāṇanirviśeṣa eva tasyām avasthāyām ātmā
bhavet; “The self in that state [of liberation] would be unconscious, just like a rock.” These
words are put by Jayanta into the mouth of an opponent; but the Naiyāyika siddhāntin
does not deny the similarity of the liberated soul to a rock. I thank Harunaga Isaacson for
this reference.
21 See Sanderson 1992, 288–289.
22 See (1) Mataṅgavṛtti, vidyāpāda ad 6:35b–d, p. 175,4–5 (and Watson 2006, 366), where
Rāmakaṇtha asserts that parāmarśa, despite rising and passing away, does not indicate the
non-eternality of the self, as it occurs not in the self but in the buddhi (and the ahaṅkāra):
that vimarśa is synonymous with parāmarśa for Rāmakaṇṭha is suggested by Nareśvara-
parīkṣāprakāśa ad 1:17, pp. 41,5–44,3 (translated and discussed in Watson 2006, 313–332),
where they are used interchangeably; (2) Kiraṇavṛtti ad 2:25ab, pp. 53,10–54,17, where
Rāmakaṇṭha distinguishes adhyavasāya, which is transient and a property (dharma) of
the buddhi, from saṃvedanātmakaṃ jñānam, which belongs to the soul and is always
occurrent; that Rāmakaṇtha uses adhyavasāya and parāmarśa (and niścaya) as synonyms
is suggested by Mataṅgavṛtti, vidyāpāda ad 6:35b–d, p. 174,4–175,9; (3) Mataṅgavṛtti, vidyā-
pāda ad 17:2, p. 382,12–21, which distinguishes adhyavasāya from saṃvedana and identifies
the former as the svabhāva of the buddhi; (4) Nareśvaraparīkṣāprakāśa ad 1:6ab, p. 28,12–
19, where adhyavasāya is said to be a guṇa of the buddhi.
via free access
96 watson
figure 4.6 Rāmakaṇṭha’s view as more extreme than Nyāya
inhere in the self. For Rāmakaṇṭha, by contrast, changing cognitions, as we saw

above, are objects of awareness (grāhya) for the self; they occur outside of it,
in the intellect (buddhi). So we arrive at a different continuum, illustrated in
figure 4.6, where the large circle is the self and the dotted lines are cognitions.
The position that postulates the most amount of change in the subject is the
Buddhist position on the right; the position that postulates the least amount
of change in the subject is Rāmakaṇṭha’s position on the left; Nyāya falls in the
middle.
And what is asserted here of cognitions is also true of latent impressions
(saṃskāras), pleasures and pains. Whereas for Nyāya these all reside in the self,
Rāmakaṇṭha and his fellow Saiddhāntikas, arguing that this would entail the
unwanted consequence that the self is subject to change, locate latent impres-
sions, pleasures and pains outside of the self in the buddhi.23
The fact that we are dealing here with two different conceptions of the rela-
tionship between self and cognitions—for Nyāya inherence (samavāya), and
for Rāmakaṇṭha a subject-object relation (grāhyagrāhakabhāva)—means that
certain objections that are put to Rāmakaṇṭha are not so applicable to Nyāya.
First, a common objection that is articulated in Buddhist texts against a
Sāṅkhya self,24 and in Rāmakaṇṭha’s texts as a Buddhist objection to his own
view,25 is as follows. If the self were completely unmodifiable, it would not be
able to perceive objects, for to perceive a pot, say, followed by a cloth requires
being able to register the change in the object-sphere from the pot to the cloth.
And there is no way to register a change without being affected by the change.
The way in which a perceiver perceives an object is by being modified by that
23 See Mataṅgavṛtti, vidyāpāda ad 17:2, p. 383,1–8 and ad 6:34c–35a, pp. 173,11–174,1.

24 E.g. Tattvasaṅgraha of Śāntarakṣita 287 and 288.
25 See Mataṅgavṛtti, vidyāpāda introducing 6:34c–35a, p. 172,7–10; Nareśvaraparīkṣāprakāśa
introducing 1:6ab, p. 17,3–4; and Nareśvaraparīkṣāprakāśa ad 1.6ab, p. 26,19–20.
via free access
object in some way. If the perceiver is permanently unaffected and unmodified,

it is mysterious how it can be a perceiver.
Here the Naiyāyika will respond that the self is modified to the extent that
separate cognitions inhere in it, the first caused by the pot and the second
caused by the cloth.
Second, Rāmakaṇṭha combines the claim that the self is unmodifiable with
the claim that it is an agent. This lays him open to an objection that, unlike
the last one, is not applicable to Sāṅkhya: how can something that is beyond
all change be an agent? There are two aspects to this objection. (1) How can
something that does not itself “move” cause movement? Here Rāmakaṇṭha’s
response is to adduce the example of a magnet. A magnet causes movement in
iron filings without itself moving; so similarly, the self causes bodily movement
without itself moving.26 But even if it were conceded that the magnet example
renders plausible a self as “unmoving mover,” there is a related but different
problem. (2) If the self is beyond all change, how can we explain that at one
moment it brings about one action and at another moment it brings about a
completely different action? Surely an agent that is always in the same state
would have to either always be doing nothing or always be doing exactly the
same thing?
Here again, the objection is not so applicable to Naiyāyikas, for they accept
the existence of different impulses (prayatnas) that inhere in the self. A prior
bodily movement will be brought about by a particular impulse, and a subse-
quent, different kind of bodily movement will be brought about by a different
impulse.
5 Nyāya Not so Moderate after All
Section 3 gave reasons for judging Rāmakaṇṭha’s position not to fall in the mid-
dle ground between Nyāya and Buddhism, but to be just as extreme as Nyāya.
Section 4 gave reasons for judging Rāmakaṇṭha’s position to be more
extreme than Nyāya’s, with the latter as the moderate one, closer to Buddhism.
But this section introduces a consideration that reveals Nyāya to be not so mod-
erate after all.
It is true that Nyāya holds cognitions, impulses (prayatna), latent impres-
sions (saṃskāra), pleasures, pains, etc., to inhere in the self, whereas Rāmakaṇ-
26 Kiraṇavṛtti ad 3:1; Paramokṣanirāsakārikāvṛtti ad 3c, pp. 115,99–116,106; and Watson, Good-

all and Sarma 2013, 258–259, especially note 196.
via free access
98 watson
ṭha (in accord with his Sāṅkhya inheritance)27 locates all of these outside of
the self, indeed outside of the realm of the immaterial altogether, within the
world derived from māyā and prakṛti. We thus seem to be left with the picture
given in figure 4.6, where the small lines can represent any of the things just
listed: they fall within the self for Nyāya, but outside of it for Rāmakaṇṭha. But
the statement that they fall within the self for Nyāya has to be qualified. For
Nyāya actually regards them as part of the “not-self.” How can that be, given
that they are qualities of the self? Because of Nyāya’s view of the firm differ-
ence between a thing and its qualities (guṇaguṇibheda). Any substance for
Nyāya is a completely different thing from its qualities. Change in the quali-
ties of a substance will not necessarily result in any change or modification of
the substance. And there is even more distance between the self and its quali-
ties than there is between a physical substance and its qualities. For the self is
eternal, its qualities temporary; it is omnipresent, while they are restricted to
a particular location.28 This Naiyāyika idea that the self in particular, and sub-
stances in general, are unaffected by changes in their qualities was one of the
things that separated their view from that of Kumārila and the Bhāṭṭa Mīmāṃ-
sakas. For the latter held the relation between a thing and its qualities to be
difference-cum-nondifference (bhedābheda); they maintained that changes in
the qualities of a thing do affect the thing itself. The self, for them, although
never ceasing to exist, is subject to modification when its qualities change.29
The fact that the Naiyāyikas rejected this, protecting the self from any effect of
change in its qualities, means that their view cannot be so starkly differentiated
from Rāmakaṇṭha’s view as figure 4.6 suggests.
This impression is strengthened by Naiyāyika passages dealing with libera-
tion (apavarga, mokṣa). Here Naiyāyika authors assert that the self’s nature is,
and always has been, free of all its particular qualities (sakalaguṇāpoḍha).30
These qualities are thus irrelevant to its nature. They are described as extrinsic
to it, not innate (na naisargikaḥ).31 They are part of the “not-self” with which it
confuses itself while in saṃsāra, and which is to be abandoned (heya).32
This thoroughgoing difference and separateness of the self’s qualities from
the self—indicated by all of the considerations above—means that the Nyāya
27 For the Sāṅkhya stance on these matters and its difference from Nyāya, see Dasti 2013,
121–135.
28 For an account of the evolution of the increasing distance that developed between the
self and its qualities in Nyāya and Vaiśeṣika, see Frauwallner 1956, 91–104; 1984, 61–71.
29 Ślokavārttika, ātmavāda 21–23.
30 Nyāyamañjarī Vol. 2, p. 359,6: sakalaguṇāpoḍham evāsya rūpam.
31 Nyāyamañjarī Vol. 2, p. 359,5.
32 Nyāyamañjarī Vol. 2, p. 265,10–12 and p. 430,3–4. Nyāyabhāṣya p. 6,9–11.
via free access
figure 4.7 Nyāya as equivalent to Rāmakaṇṭha’s view
position can arguably be represented as it is in figure 4.7, and hence regarded

as just as extreme as Rāmakaṇṭha’s position.
6 Concluding Remarks
Where do all of these divergent analyses leave us? What is cancelled out by
what? It may be better to see each perspective as having its own validity. We
have four perspectives:
1. Rāmakaṇṭha in the middle ground (section 2 and Watson 2014).
2. Rāmakaṇṭha as just as extreme as Nyāya (section 3).
3. Rāmakaṇṭha as more extreme than Nyāya (section 4).
4. Rāmakaṇṭha as just as extreme as Nyāya (section 5).
Or rather, three perspectives, since 2 and 4 are the same. In what sense is 1
valid? The Śaiva equating of self and consciousness/cognition and its view of
the self as the shining forth of prakāśa, the light of consciousness, is indeed
something that differentiates Rāmakaṇṭha’s view from Nyāya in substantial
and significant ways. Related to this is the Śaiva rejection of the existence of
substances (dravya) over and above qualities (guṇa), or property-possessors
(dharmin) over and above properties (dharma). And just as, for Rāmakaṇṭha,
there is no self-substance over and above consciousness/cognition, so too there
is no self as agent (kartṛ) or knower ( jñātṛ/grāhaka) over and above conscious-
ness/cognition. From the point of view of this equating of self and conscious-
ness, perspective 1 is valid, and figure 4.4 captures a certain relationship that
Rāmakaṇṭha’s position bears to Nyāya and Buddhism. This perspective and this
figure also accord with the way that Rāmakaṇṭha’s self is more dynamic than
Nyāya’s.
But it is not valid to see things only in this way, because when we add the
consideration that Rāmakaṇṭha accepts two kinds of cognition, this allows for
via free access
100 watson
the possibility of aligning not his first kind, but his second kind (that which
is plural and is located in the buddhi) with Nyāya’s and Buddhism’s cognition.
When we do that, and we add Rāmakaṇṭha’s self (i.e. his first kind of cogni-
tion) above that, his position becomes equivalent to Nyāya’s (both in terms of
the number of things postulated and the amount of change in the self)—and
figure 4.5 becomes appropriate.
But this does not exhaust the number of ways of seeing the relationship
between Rāmakaṇṭha and Nyāya, because there is a further significant consid-
eration. While it is true that the selves of both Nyāya and Śaivasiddhānta are
eternally unchanging (which supports the idea that they are both as extreme
as each other: perspective 2), it is also the case that cognitions, impulses (pra-
yatna), latent impressions (saṃskāra), pleasures, pains, etc., inhere in the self
for Nyāya, but fall completely outside it for Rāmakaṇṭha. Emphasizing this
makes it appear that Rāmakaṇṭha’s self, like a Sāṅkhya self, is even more
removed from change than Nyāya’s. This perspective 3 (illustrated in figure 4.6)
carries some weight, but is no final resting place, for if we concentrate on those
places where Nyāya emphasizes that qualities are extrinsic to substances, and
that consciousness/cognitions, etc., are part of the not-self, not affecting its
nature in any way, then it appears that there is in fact no more change in a
Naiyāyika self than a Saiddhāntika one. So to adopt perspective 4 is to revert
to perspective 2, which this time can be illustrated with a slightly different dia-
gram: figure 4.7.
The explanation for the difference between perspectives 3 and 4 is a cer-
tain tension within Nyāya, one that was already noted by Frauwallner in 1956.33
Frauwallner argues that the Naiyāyika self is the product of two separate influ-
ences. On the one hand, it derives from the self as characterized in the old
philosophy of nature; on the other hand it was subject to continual attrac-
tion to the self of the Sāṅkhyas. According to the former, selves were of limited
size, and hence capable of movement; they were that which transmigrates from
body to body; and they were all qualitatively distinct from each other, as a result
of being characterized by their own qualities (svaguṇaviśiṣṭa). There was thus a
huge difference between these selves and those of Sāṅkhya: the latter were all-
pervading and hence incapable of movement; it was not they that transmigrate
(but rather a subtle body); and they were devoid of all qualities. But over time
the difference of the selves of the Naiyāyikas and Vaiśeṣikas from those of the
Sāṅkhyas decreased. The former became all-pervading and hence incapable
of movement; they ceased to be that which transmigrates from body to body
33 Frauwallner (1956, 91–104), (1984, 61–71).
via free access
(the manas took on this role); and their relationship to their qualities became
reassessed. They were no longer characterized by their own qualities (svagu-
ṇaviśiṣṭa); rather those qualities came to be regarded as extrinsic to them, as
part of the not-self, and as needing to be abandoned if the self is to rest in its
own nature. Previously selves were affected by changes in their qualities; subse-
quently they were not. Previously selves were qualitatively different from each
other; subsequently they were all qualitatively identical as a result of being, in
their true nature, devoid of particular qualities.
The residues in a Naiyāyika self of earlier philosophy of nature weigh in
favour of perspective 3; the influence of the Sāṅkhya model of a self pulls it
towards perspective 4.
Just as the difference between perspectives 3 and 4 results from focusing
on different aspects of Nyāya, so the difference between perspectives 1 and 2
results from focusing on different aspects of Śaivasiddhānta. Just as we have
offered an explanation of the first difference in terms of a tension within Nyāya,
is there also an explanation of the second difference in terms of a tension
within Śaivasiddhānta? In fact there is. Rāmakaṇṭha’s self is one whose nature
derives to a large extent from Sāṅkhya: passing cognitions do not inhere in it,
saṃskāras do not reside in it, pleasures and pains fall outside of it. It is elevated
above all change; it sits above and beyond the various tattvas that comprise
the psycho-physical universe. This is the kind of self that Rāmakaṇṭha inherits
from his scriptural tradition. But his concept of self is also the result of a sec-
ond influence, one which goes back to some of the Śaiva scriptures34 and which
had been increasingly making itself felt among the Śaivas of Kashmir (both
Saiddhāntikas such as Nārayaṇakaṇṭha and non-dualists such as Utpaladeva)
during the previous century. This second influence sees the self not as a static
Sāṅkhya-like entity, but as dynamic and as an outpouring of light (prakāśa)—
the light of cognition/consciousness. The contrast between the two influences
is strong. (1) According to the first influence it is a passive non-agent; according
to the second it is a fully-fledged agent. I.e., not only is it an agent while unlib-
erated, as in Nyāya, but it is also an agent in liberation; indeed at that time its
agency expands into omnipotence. (2) According to the first, jñāna falls com-
pletely outside of the self; according to the second, the self is of the nature of
jñāna.
Perspective 2 (which was presented in section 3) results from focusing on
the first influence, the Sāṅkhya inheritance. Perspective 1 (presented in section
2) results from focusing on the new Śaiva insights.
34 See Watson (2010, 80).
via free access
102 watson
Acknowledgements
I would like to thank Michael Allen for a conversation over lunch in Harvard
Square in 2013 that planted the seeds of many of the new directions of thought
in this article.
References
Primary Sources
Abhidharmakośabhāṣya. Swāmi Dwārikā Dās Śāstrī, ed. Abhidharmakośa and Bhāṣya
of Ācārya Vasubandhu with Sphuṭārthā Commentary of Ācārya Yaśomitra. Bauddha
Bhāratī Series, nos. 5–7, 9. Vārāṇasī: Bauddha Bhāratī, 1970–1973.
Aṣṭaprakaraṇa. Brajavallabha Dvivedī, ed. Tattvaprakāśa, Tattvasaṅgraha, Tattvatra-
yanirṇaya, Ratnatraya, Bhogakārikā, Nādakārikā, Mokṣakārikā, Paramokṣanirāsa-
kārikā. Yogatantra-granthamālā, no. 12. Vārāṇasī: Sampurnananda Sanskrit Univer-
sity, 1988.
Aṣṭādhyāyī. S.M. Katre, ed. Aṣṭādhyāyī of Pāṇini. Delhi: Motilal Banarsidass, 1989.
Kiraṇavṛtti. Dominic Goodall, ed. Kiraṇavṛttiḥ. Bhaṭṭa Rāmakaṇṭha’s Commentary on
the Kiraṇatantra. Volume 1: Chapters 1–6. Collection Indologie, no. 86.1. Pondichéry:
Institut français de Pondichéry/École française d’Extrême-Orient, 1998. (In refer-
ences to this edition, line numbers are not counted from the top of the page; they
rather follow the line numbers printed in the edition, which start counting from the
previous verse-segment.)
Tattvasaṃgraha of Śāntarakṣita. Dwarikadas Shastri, ed. Tattvasangraha of Ācārya
Shāntarakṣita with the commentary “Pañjikā” of Shri Kamalashīla. 2 vols. Bauddha
Bhāratī Series, nos. 1–2. Vārāṇasī: Bauddha Bhāratī, 1981–1982.
Tattvasaṃgraha of Sadyojyotis. See Aṣṭaprakaraṇa.
Nareśvaraparīkṣāprakāśa. Madhusūdan Kaul Śāstrī, ed. Nareśvaraparīkṣā of Sadyojy-
otis with the Prakāśa Commentary of Rāmakaṇṭha. Kashmir Series of Texts and Stud-
ies, no. 45. Reprint, Delhi, 1989 [Śrīnagar: Research Department Jammu and Kashmir
State, 1926].
Nyāyabhāṣya. Anantalal Thakur, ed. Gautamīyanyāyadarśana with Bhāṣya of Vātsyā-
yana. Nyāyacaturgranthikā, no. 1. New Delhi: Indian Council of Philosophical
Research, 1997.
Nyāyamañjarī. K.S. Varadacharya, ed. Nyāyamañjarī of Jayantabhaṭṭa, with Ṭippaṇi—
Nyāyasaurabha by the Editor. 2 vols. Oriental Research Institute Series nos. 116 and
139. Mysore: Oriental Research Institute, 1969 and 1983.
Nyāyavārttika. Anantalal Thakur, ed. Nyāyabhāṣyavārttika of Bhāradvāja Uddyotakara.
Nyāyacaturgranthikā, no. 2. New Delhi: Indian Council of Philosophical Research,
1997.
via free access
Nyāyasūtra. See Nyāyabhāṣya.

Paramokṣanirāsakārikāvṛtti. See Watson, Goodall and Sarma 2013.
Parākhyatantra. Dominic Goodall, ed. The Parākhyatantra. A Scripture of the Śaiva
Siddhānta. Collection Indologie, no. 98. Pondichéry: Institut Français de Pondich-
éry/École française d’Extrême-Orient, 2004.
Praśastapādabhāṣya. Johannes Bronkhorst and Yves Ramseier, ed. Word Index to the
Praśastapādabhāṣya: a Complete Word Index to the Printed Editions of the Praśas-
tapādabhāṣya. Delhi: Motilal Banarsidass, 1994.
Bhogakārikā of Sadyojyotis. See Aṣṭaprakaraṇa.
Mataṅgavṛtti. See Mataṅgapārameśvara.
Mataṅgapārameśvara. N.R. Bhatt, ed. Mataṅgapārameśvarāgama, Vidyāpāda, avec le
commentaire de Bhaṭṭa Rāmakaṇṭha. Collection Indologie, no. 56. Pondichéry: Insti-
tut Français d’Indologie, 1977.
Mṛgendratantra. Madhusūdan Kaul Śāstrī, ed. The Śrī Mṛgendra Tantram: Vidyāpāda &
Yogapāda, with the Commentary (-vṛtti) of Nārāyaṇakaṇṭha. Kashmir Series of Texts
and Studies, no. 50. Reprint New Delhi: Panini, 1982 [Bombay; Srinagar: Jammu and
Kashmir State Archaeological and Research Department, 1930.]
Yuktidīpikā. Albrecht Wezler and Shujun Motegi, eds. Yuktidīpikā, the Most Significant
Commentary on the Sāṃkhyakārikā. Alt- und Neu-Indische Studien, no. 44. Stuttgart:
Franz Steiner Verlag, 1998.
Vaiśeṣikasūtra. Muni Sri Jambuvijayaji. Vaiśeṣikasūtra of Kaṇāda with the Commen-
tary of Candrānanda. Gaekwad’s Oriental Series, no. 136. Baroda: Oriental Institute,
1961.
Ślokavārttika. Dvārikādāsa Śāstrī, ed. Ślokavārttika of Śrī Kumārila Bhaṭṭa with the Com-
mentary Nyāyaratnākara of Śrī Pārthasārathi Miśra. Prāchyabhārati Series, no. 10.
Varanasi: Tara Publications, 1978.
Sāṅkhyakārikā. See Yuktidīpikā (Appendix II).
Svāyambhuvasūtrasaṅgraha. Pierre-Sylvain Filliozat, ed. Svāyambhuvasūtrasaṅgraha:
The Tantra of Svayaṃbhū, Vidyāpāda, with the Commentary of Sadyojyoti. Kalāmū-
laśāstra Series, no. 13. Delhi: Indira Gandhi National Centre for the Arts/Motilal
Banarsidass, 1994.
Secondary Sources
Boccio, Fabio. 2002. “Die Konzeption der buddhi als ‘Genußobject’ in Sadyojyotis’
Bhogakārikā.” In Śikhisamuccayaḥ. Indian and Tibetan Studies (Collectanea Marpur-
gensia Indologica et Tibetica), edited by D. Dimitrov, U. Roesler and R. Steiner, 11–26.
Wiener Studien zur Tibetologie und Buddhismuskunde 53. Wien: Arbeitskreis für
Tibetische und Buddhistische Studien, Universität Wien.
Chakrabarti, Arindam. 1982. “The Nyāya Proofs for the Existence of the Soul.” Journal of
Indian Philosophy 10: 211–238.
via free access
104 watson
Dasti, Matthew. 2013. “Nyāya’s Self as Agent and Knower.” In Free Will, Agency, and Self-
hood in Indian Philosophy, edited by M. Dasti and E.F. Bryant, 112–136. New York:
Frauwallner, Erich. 1956. Geschichte der indischen Philosophie. Volume 2. Salzburg: Otto
Muller Verlag.
Frauwallner, Erich. 1984. History of Indian Philosophy. Volume 2. Translated from original
German into English by V.M. Bedekar. Delhi: Motilal Banarsidass.
Goodall, Dominic. 2004. The Parākhyatantra. A Scripture of the Śaiva Siddhānta. A Crit-
ical Edition and Annotated Translation. Collection Indologie, no. 98. Pondichéry:
Institut Français de Pondichéry/École française d’Extrême-Orient
Halbfass, Wilhelm. 1976. “Zum Begriff der Substanz (dravya) in Vaiśeṣika.” Wiener
Zeitschrift für die Kunde Südasiens 20: 141–166.
Hulin, Michel. 1980. Mṛgendrāgama: sections de la doctrine et du yoga avec la Vṛtti de
Bhaṭṭanārāyaṇakaṇṭha et la Dīpikā d’Aghoraśivācārya. Traduction, Introduction et
Notes. Collection Indologie, no. 63. Pondichéry: Institut français d’Indologie.
Kano, Kyo. 2001. “Pariśeṣa, Prasaṅga, and Kevalavyatirekin—the Logical Structure of
the Proof of Ātman.” Journal of Indian Philosophy 29: 405–422.
Matilal, Bimal Krishna. 1989. “Nyāya Critique of the Buddhist Doctrine of Non-soul.”
Journal of Indian Philosophy 17 (1): 61–79.
Matilal, Bimal Krishna. 1994. “The Perception of the Self in Indian Tradition.” In Self as
Person in Asian Theory and Practice, edited by R. Ames, et al., 279–295. New York:
State University of New York Press.
Oetke, Claus. 1988. “Ich” und das Ich. Alt- und Neu-Indische Studien 33. Stuttgart: Franz
Steiner Verlag.
Preisendanz, Karin. 1994. Studien zu Nyāyasūtra III.1 mit dem Nyāyatattvāloka Vācas-
pati Miśras II. Alt- und Neu-Indische Studien, no. 46. Stuttgart: Franz Steiner Ver-
lag.
Sanderson, Alexis. 1992. “The Doctrine of the Mālinīvijayottaratantra.” In Ritual and
Speculation in Early Tantrism. Studies in Honour of André Padoux, edited by
T. Goudriaan, 281–312. Albany: State University of New York Press.
Watson, Alex. 2006. The Self’s Awareness of Itself. Bhaṭṭa Rāmakaṇṭha’s Arguments
Against the Buddhist Doctrine of No-Self. Publications of the De Nobili Research
Library, no. 32. Wien: De Nobili Research Library.
Watson, Alex. 2010. “Rāmakaṇṭha’s Concept of Unchanging Cognition (nityajñāna):
Influence from Buddhism, Sāṃkhya, Vedānta.” In From Vasubandhu to Caitanya.
Studies in Indian Philosophy and Its Textual History. Papers of the 12th World Sanskrit
Conference. Volume 10.1, edited by J. Bronkhorst and K. Preisendanz, 79–120. Delhi:
Motilal Banarsidass.
Watson, Alex. 2014. “The Self as a Dynamic Constant. Rāmakaṇṭha’s Middle Ground
Between a Naiyāyika Eternal Self-Substance and a Buddhist Stream of Conscious-
via free access
ness-Moments.” Journal of Indian Philosophy 42 (1): 173–193. doi: 10.1007/s10781-013-

9215-2.
Watson, Alex, Dominic Goodall and S.L.P. Anjaneya Sarma. 2013. An Enquiry into the
Nature of Liberation. Bhaṭṭa Rāmakaṇṭha’s Paramokṣanirāsakārikāvṛtti, a commen-
tary on Sadyojyotiḥ’s refutation of twenty conceptions of the liberated state (mokṣa).
Collection Indologie, no. 122. Pondichéry: Institut Français de Pondichéry/École
via free access
chapter 5
Some Hitherto Unknown Fragments of

Utpaladeva’s Vivṛti (II): Against the Existence of
External Objects
Isabelle Ratié
1 Fragments of Utpaladeva’s Vivṛti in Marginal Annotations of

Manuscripts Belonging to the Pratyabhijñā Corpus
Raffaele Torella’s discovery and remarkable edition of the only (and very
incomplete) manuscript of the Vivṛti thus far known1 has enabled us to deter-
mine with certainty that some marginal annotations in manuscripts of Abhi-
navagupta’s Īśvarapratyabhijñāvimarśinī (henceforth ĪPV) and Īśvarapratya-
bhijñāvivṛtivimarśinī (henceforth ĪPVV) regularly quote Utpaladeva’s lost work.
Nine of the ĪPV manuscripts that I have been able to consult contain such
quotations;2 two ĪPVV manuscripts3 known to me bear several fragments, the
most important of which is by far the lengthiest Vivṛti fragment known to
date (including the one found in the codex unicus edited by Raffaele Torella),
covering three Vivṛti chapters;4 and two manuscripts of the Īśvarapratyabhi-
jñāvṛtti (henceforth Vṛtti) have been shown to contain some much shorter
Vivṛti quotations.5 The following pages are an attempt to edit and translate
1 The manuscript as it stands now only covers verses 1.3.6 to 1.5.3. See Torella 1988 (this first part
of the fragment was in fact edited from a transcript of the original manuscript D later found
by R. Torella in the National Archives of India and used for the edition of the remaining parts
of the fragment), and Torella 2007a, 2007b, 2007c and 2007d.
2 For a list of parallels between marginal annotations in D2, SOAS and S3 and the fragment
edited by R. Torella, see Ratié 2016. Five other ĪPV manuscripts—namely JR, S2, S5, S7, S9—
have been shown to contain such fragments in Ratié 2016b, and one more (S15) in Ratié
forthcoming (see also below).
3 S12 and J11.
4 On this discovery, see Ratié 2017 and Ratié forthcoming.
5 See Kawajiri 2016 and 2016b for a diplomatic edition of marginal annotations containing
Vivṛti quotations in a Vṛtti manuscript from Lucknow in Devanāgarī script. Note, however,
that S19 (a Śāradā manuscript used by Raffaele Torella for his edition of the Vṛtti) also con-
tains many Vivṛti fragments. It is described in Torella 2002, l–li (under siglum N) as “profusely
annotated by a different hand with single glosses or long passages (mostly drawn from the
ĪPV) between the lines and often entirely covering the margins.” Indeed, many of these quota-
© isabelle ratié, 2020 | doi:10.1163/9789004432802_007

via free access
fragments of utpaladeva’s vivṛti (ii) 107
some brief marginal annotations borrowed from Utpaladeva’s lost Vivṛti on

Īśvarapratyabhijñākārikā (henceforth ĪPK) 1.5.6 and 1.5.8–9.6
This is no easy task, because very little is known of the source(s), purpose(s)
and history of these annotations,7 and because most of the time their authors
do not specify whether they are quoting8—and if so, which text9—or rather
giving a comment of their own. We can quite safely assume that these anno-
tations are indeed genuine quotations from the Vivṛti whenever we can find
in them a sequence of words corresponding exactly to the pratīkas in Abhi-
navagupta’s ĪPVV. However, Abhinavagupta’s quotations of the Vivṛti are some-
times so brief that we cannot always reach such a high level of probability,
as can be seen from one problematic instance below.10 Besides, the marginal
tions are from the ĪPV; but some are Vivṛti fragments, as shown in Ratié forthcoming (see
also Ratié 2018b, Appendix I). S19 was not taken into account here, not because it would
be irrelevant (its margins bear several fragments on ĪPK 1.5.6–9, and interestingly, most of
them are not found in the ĪPV and ĪPVV manuscripts used below), but because unfortu-
nately, I made this discovery too late to incorporate the new findings in this article.
6 So far I have not found any Vivṛti quotation on ĪPK 1.5.7 in ĪPV or ĪPVV manuscripts.
It should also be noted that S12, to which I only recently had access, bears the frag-
ment edited in Ratié 2016, 239–240 from marginal annotations in S3. In S12 the frag-
ment appears in the margins of folios 119A–120B. The missing words in S3 (Ratié 2016,
239, note 35) appear as tathā carati in S12; the compound parajñānaviṣayabhāvagamanā-
bhisandhir (ibid., 239, § 1, last line) has the wrong reading parajñānaparajñānaviṣaya-
bhāvagamanābhisandhir; S12 shares with S3 the wrong readings °vyavahārādy° instead of
°vyāhārādy° (see ibid., 239, note 36 and 37), and occasionally reads °vyavahārādy° where
S3 has the right reading °vyāhārādy° (in tajjñānagatavyāhārādy°, ibid., 239, §2, line 5);
it reads yadi where S3 has yad iha (ibid., 239, § 2, line 6—I had conjectured yadi hi); it
also has ābhāsabhedahetutva° instead of ābhāsabhede hetutva° (ibid., 240, §5, line 5); it
shares with S3 the wrong readings yathāvabhāsānāṃ (see ibid., 240, note 41) and pramā-
tṛprābhāsaparyavasānaṃ (see ibid., 240, note 42); finally, it reads nāpi pratibodhitā ekaiva
instead of S3’s nādhipratitvocitaikaiva (and my conjecture nādhipatitvocitaikaiva).
7 See Formigatti 2011 and Ratié 2018b. Formigatti 2011 (a very interesting study of marginal
annotations in Sanskrit manuscripts from northern India and Nepal) bears in part on ĪPV
manuscripts; unfortunately it makes no mention of Vivṛti fragments, although at least one
of the manuscripts examined, Ś7 (= S9 in the present paper), contains quotations from
Utpaladeva’s detailed auto-commentary (see Ratié 2016b). The reason for this is that due
to the length of the ĪPV, Formigatti chose to confine his research to the annotations on the
first chapter, which are apparently devoid of such quotations.
8 This happens, however: see Ratié 2016, 228, note 10, and below, note 58.
9 The marginal annotations most often quote from Utpaladeva’s Vṛtti and Abhinavagupta’s
ĪPVV, but other works are occasionally cited, including Śaiva texts (such as the Śivadṛṣṭi) as
well as Buddhist works (e.g. Dharmakīrti’s Sambandhaparīkṣā) or Brahmanical treatises
(e.g. Jayanta’s Nyāyamañjarī): see Ratié 2017, 165.
10 See fragment no. 6? on ĪPK 1.5.8–9.
via free access
108 ratié
figure 5.1 Manuscript S3 of Abhinavagupta’s Īśvarapratyabhijñāvimarśinī (f. 43A), with frag-

ments 4 and 5 of Utpaladeva’s Vivṛti on vv. 1.5.8–9 in the top and right margins
Photograph: National Archives of India
annotations are sometimes barely legible, and Utpaladeva’s prose is difficult.

For all these reasons, both the edition and translation given here are only ten-
tative.
2 The Context and Gist of Fragments 1–3 on ĪPK 1.5.6
The overall context of the fragments presented here is a controversy over the
existence of a reality external to consciousness, and Utpaladeva’s main inter-
locutor at this point of the debate is a Buddhist Sautrāntika.11 The latter admits
that we can never have direct access to any reality existing beyond conscious-
ness, since the objects of our perceptions are nothing but internal aspects
(ākāra) of consciousness; yet he claims that an external reality must be inferred
so as to account for phenomenal variety: according to him, consciousness is in
itself an undifferentiated manifesting entity or light (prakāśamātra), so that
the variety of objects of which we are aware must have a cause outside of con-
sciousness, just as a variety of reflections in a mirror must have as its cause a
11 On the portrayal of this philosophical figure by the Śaiva nondualists and Utpaladeva’s
refutation of his thesis (both of which are only briefly outlined here), see Ratié 2011, 368–
402 and 442–473, and Ratié 2011b.
via free access
variety of forms existing outside of the mirror. In ĪPK 1.5.6, Utpaladeva endeav-
ours to criticize this thesis by showing that the sphere of ordinary human prac-
tice (vyavahāra) functions on the mere basis of phenomena (so that from this
practical point of view, at least, all inquiries into the existence of an external
reality are useless) and that further, an external object is not even rationally
possible, so that there is no point in assuming the existence of such an absurd
object.12
In fragment no. 1, Utpaladeva first points out that human activities pertain
only to phenomena and that no inquiry over a potential external object is
needed to ensure practical success in the everyday world: in the course of ordi-
nary existence we are content to deal with mere manifestations, because these
manifestations are endowed with an efficacy (arthakriyā) capable of fulfilling
our needs (when we are hungry, we do not ask ourselves whether mangoes
exist outside our consciousness: we look for the phenomenon of a mango and
are content with experiencing its perceptible effects, since no entity by nature
imperceptible can be consumed or utilized).13
To this the Sautrāntika could reply that assuming the existence of some
imperceptible entities is nonetheless a necessary requirement if we are to
make sense of our own perceptions; thus most Indian philosophical systems
acknowledge that our eyes, which are perceptible elements of our body, are
not enough to explain the phenomenon of vision (since blind people too may
possess them), so that we must consider them (as well as our ears, etc.) as mere
receptacles for the real sense organs (indriya), understood as instruments of
perception that are too subtle to be perceptible, but that must nonetheless
be inferred from the fact that we do perceive. Utpaladeva’s opponent argues
that just as we must assume the existence of the imperceptible indriyas so
as to account for our faculty of perceiving, in the same way, we must infer
the existence of an external reality so as to explain phenomenal variety. In
response to this argument, Utpaladeva points out in the fragment that con-
trary to what the Sautrāntika claims, the assumption of the indriyas’ existence
is not universally shared among theoreticians, and in this connection he men-
tions two theses that claim to explain perception without assuming that such
12 See ĪPK 1.5.6: syād etad avabhāseṣu teṣv evāvasite sati | vyavahāre kim anyena bāhyenānu-
papattinā || “Let [us admit] this: since ordinary human practice (vyavahāra) is determined
on [the basis of] these manifestations alone, what is the point of some external [object]
that would be distinct [from consciousness and] that is not [even] logically possible (anu-
papatti)?” On the two possible interpretations of syād etat according to Abhinavagupta,
see Torella 2002, 114, note 12, and Ratié 2011, 386, note 48.
13 Cf. ĪPV, vol. I, 178, quoted and translated in Ratié 2011, 386.
via free access
110 ratié
imperceptible sense organs must exist. According to Abhinavagupta, one of

these theses belongs to Bhartṛhari and was expounded in his now lost Ṣaḍ-
dhātusamīkṣā,14 while the other is that of a “follower of Bṛhaspati” (bārhas-
patya), i.e., a materialist (cārvāka).15 In fact Utpaladeva disagrees with both
theories since in ĪPK 1.5.8–9 he clearly admits that the inference of the indriyas
is valid; but contrary to the Sautrāntika, he believes that inferring the existence
of a purely imperceptible entity is illegitimate and even impossible (which is
why he claims that the indriyas are perceived at least in some respect).16 Most
importantly, it seems that his goal here is merely to show that there is no con-
sensus on the necessity of assuming the indriyas’ existence: the fragment only
points out that there are competing ways of explaining perception, some of
which do not involve the assumption of imperceptible organs of perception,
so that the Sautrāntika cannot invoke any universal agreement on the indriyas
in order to substantiate the claim that external objects too must be inferred.
Utpaladeva then goes one step further in this first fragment and asserts that
the investigation of the external object is not only useless but also doomed to
failure, because the external object is not rationally possible.
It is worth noting in this respect that in his commentaries, Abhinavagupta
points out that there are two different kinds of argument showing that the
external object is impossible from a rational point of view.17 The first one sim-
ply consists in stating that the external object cannot exist because it cannot
be manifest in any way, even as an object of inference.18 The second set of argu-
ments (which obviously paraphrases Vasubandhu’s Viṃśikā19 but may have
been borrowed, at least in part, from Śaṅkaranandana’s synthetic reformula-
tion of Vasubandhu’s arguments in his Prajñālaṅkāra20) consists in showing
that the external object cannot exist because there is no way of satisfactorily
14 On this work see Ratié 2018.

15 See below, notes 88–90.
16 See below, note 115.
17 See Ratié 2014.
18 See ĪPV, vol. I, 178: … yatra sādhakaṃ ca nāsti pramāṇaṃ bādhakaṃ ca prakāśād bhede
’numeyatayāpi prakāśanābhāva iti tāvan mukhyam. “And there is no argument proving
[the existence] of the [external object], and the main (mukhya) [argument] refuting
[its existence] amounts to this much: the fact that there [can] be no manifestation
(prakāśana) [of it] even as a [mere] object of inference if [this object] is distinct from
the manifesting consciousness (prakāśa).”
19 On the title Viṃśikā (rather than Viṃśatikā), see Kano 2008, 345.
20 See ĪPV, vol. I, 181: abhyuccayabādhakaṃ cedam iti nātrāsmābhir bharaḥ kṛtaḥ. vistareṇa
ca prajñālaṅkāre darśitam ācāryaśaṅkaranandanena. “And we have not taken the trou-
ble of [detailing] here what these additional refuting arguments are; and the master
Śaṅkaranandana has shown [this] at length in his Prajñālaṅkāra.”
via free access
explaining it, whether or not we consider it to be made of parts.21 Abhinav-

agupta calls the first argument “principal” (mukhya)22 and specifies that while
the other, additional arguments focus on the contradictions involved in this or
that particular property of the external object, the main argument functions
as a global refutation of the external object’s existence.23 This distinction is
not found in Utpaladeva’s Vṛtti, which merely mentions what Abhinavagupta
calls the “additional” arguments;24 nonetheless, the ĪPVV’s phrasing shows that
it was no innovation on Abhinavagupta’s part and that it must have been stated
21 See ĪPV, vol. I, 178: abhyuccayabādhakās tv avayavino vṛttyanupapattiḥ samavāyāsiddhiḥ

kampākampāvaraṇānāvaraṇaraktāraktadigbhāgabhedādiviruddhadharmayogaḥ … “As
for the additional arguments refuting [the existence of the external object], they are: the
impossibility of the existence of a whole (avayavin) [in its parts]; the fact that the inher-
ence (samavāya) [of the whole in its parts] is not established; the fact that the [external
object must] possess some contradictory properties, such as movement and the absence
of movement, being covered and being uncovered, being colored and being colourless,
being differentiated into parts according to [the six] directions (digbhāgabheda), etc.” On
Abhinavagupta’s explanation of these additional arguments and its Buddhist sources see
Ratié 2010, 446–452, and Ratié 2011, 390–403.
22 See above, note 18.
23 See ĪPVV, vol. II, 138: na ca kevalam idam eva bādhakaṃ yat pramāṇamukhena
pravṛttaṃ yāvat svato ’pīti svamukhena prameyasvarūpanirūpaṇapravṛttapratītidvāreṇāpi
pravṛttam, prameyasvarūpam evāsattvāviṣkaraṇenonmūlayaty* eva bāhyam iti yāvat. pūr-
vakaṃ hi bādhakaṃ prameyasvarūpam anapekṣayaiva sāṃśam anaṃśaṃ viruddham
aviruddhaṃ yad bhavatu tad bhavatu, prakāśād bhinnaṃ na prakāśata iti tu sarvagrāsena
pravṛttam. [*evāsattvāviṣkaraṇenonmūlayaty conj.; evāsattvāviṣkaraṇaivonmūlayaty T; eva
sattvāviṣkaraṇenonmūlayaty J10, ĪPVV; eva sattāviṣkaraṇenomūlyaty J11.] “And [against the
thesis of the existence of the external object,] there is not only this [aforementioned]
refuting argument (bādhaka) which functions through the means of [valid] knowledge
(pramāṇa) [lacking in the case of the external object]; [there is] also [a refuting argument]
which functions ‘by itself’ [according to the Vivṛti], that is to say, through the [external
object’s] own [nature, or more precisely,] through the awareness arising from the exam-
ination of the [contradictory] nature of the object of knowledge (prameya). [The latter
argument] completely eradicates the very nature of the object of knowledge—that is
to say, the external [object]—by showing that [this contradictory nature can]not exist.
For the first refuting argument functions while completely disregarding the nature of the
object of knowledge—[i.e.] whether it has parts or is devoid of parts, whether it is contra-
dicted or not [by this or that particular property]—rather, [it functions] through a global
refutation (sarvagrāsena [lit. ‘by devouring everything’]), thus: ‘[What is] distinct from the
manifesting consciousness (prakāśa) is not manifest (na prakāśate).’”
24 See Vṛtti on ĪPK 1.5.6, 20–21: … bāhyaś cārthaḥ pramāṇabādhitaḥ sāvayavo viruddhadhar-
mādhyāsāder niravayavaś ca dikṣaṭkayogāder bahuśaḥ. “Moreover, [the existence of] the
external object is refuted by a means of [valid] knowledge if it has parts, because of [the
necessity then] of attributing to it contradictory properties, etc.; [and it is contradicted] in
many ways if it has no parts, because [then] it must be in contact with the six directions,
etc.”
via free access
112 ratié
by Utpaladeva himself in his Vivṛti.25 This is confirmed by the fragments pre-

sented below. While fragment no. 3—which mentions the famous criticism
of atomism in Vasubandhu’s Viṃśikā 11–15—is obviously concerned with the
second set of arguments,26 in fragments no. 1 and 3 Utpaladeva clearly refers
to the first, “principal” argument against the existence of the external object;
thus the end of fragment no. 1 explains that the only possible objects for con-
sciousness are phenomena and that objects are nothing but the manifesting
consciousness (prakāśa), and fragment no. 2 further explains that even objects
known through inference must be manifested (avabhāta) in some way so as to
be objects for consciousness.
This latter point is crucial, because the idea that even conceptual objects
must be directly manifested in some way is one of the main divergences
between Buddhist and Śaiva nondualistic epistemologies, as Abhinavagupta
himself makes clear in his commentary.27 The Śaiva nondualists thus accuse
the Vijñānavādins of unwittingly granting some legitimacy to the external-
ism of their Sautrāntika opponents by admitting that contrary to a perceived
object, the object of a mere concept does not require any immediate mani-
festation at the time of its conceptualization. According to the Śaivas, such
a concession entails that there is nothing inherently wrong in trying to infer,
as the Sautrāntikas do, an external object by nature alien to manifestation,
whereas Utpaladeva and Abhinavagupta deny the mere possibility of even
conceptualizing—and therefore inferring—an entity fundamentally distinct
from consciousness.28
At the end of fragment no. 2, Utpaladeva states a new objection on the part of
his externalist opponent. According to the Sautrāntika, if we do not assume the
existence of an external reality, we cannot explain why we are aware of a phe-
nomenal variety that occurs in a specific sequence: there must be a reason for
the fact that some things are manifest to us and that these manifestations occur
successively rather than simultaneously. As Abhinavagupta puts it, “a manifes-
tation necessarily requires a cause as regards both [its] arising and [its] not
arising,”29 and if we do not attempt to determine that cause, the very notion
of causality as well as that of the relationship between the apprehending sub-
ject and the apprehended object (which, according to the Buddhists, depends
25 See ĪPVV, vol. II, 138, quoted above, note 23.

26 For an explanation of Abhinavagupta’s commentaries on this point see Ratié 2010, 451–
452 and Ratié 2011, 395–399.
28 On this divergence see Ratié 2011b, 496–498.
via free access
on the causality relationship) is lost. The Sautrāntika’s intention is obviously to

have his opponent admit that the only possible cause for this phenomenal vari-
ety endowed with a specific order is a variety of external objects possessed of a
corresponding specific order and projecting their forms onto consciousness.
Unfortunately, Utpaladeva’s answer to this objection is not to be found
in the margins of the manuscripts that I have consulted so far. It is possi-
ble, however, to retrieve the gist of Utpaladeva’s reply from Abhinavagupta’s
ĪPVV: Utpaladeva seems to have argued in the missing passage of the Vivṛti
that the Sautrāntika cannot rightfully look for an external cause explaining
the arising of manifestations as well as their lack of arising, because reality
(understood as a unique and all-encompassing consciousness) is always man-
ifest and always evident for any conscious entity, in the form of the immedi-
ate self-awareness inherent in any subjectivity.30 In fact temporal and spatial
sequences (kāladeśakrama) are nothing but ways for consciousness to manifest
its unitary nature as if it were differentiated, and this differentiated manifesta-
tion is the result of the fundamental dynamism of consciousness, conceived
as a constantly exerted power of manifesting itself in all possible forms. Thus
according to the Śaiva nondualists, consciousness is not a mirror passively
reflecting a world of objects existing independently of it, and phenomenal vari-
ety, far from being the mere reflection of such an external world, is an expres-
sion of consciousness’s freedom (svātantrya) to manifest itself in an infinite
variety of forms while playfully hiding its fundamental unity. The Sautrāntika
is therefore wrong to assume that phenomenal variety and its spatio-temporal
sequence can only be explained by postulating an external cause, since they
can be accounted for as the result of the freedom of consciousness31—a free-
30 See e.g. ĪPVV, vol. II, 133: paramabhūmir āgameṣu paramarahasyāpi satī naivānābhāsa-
mānā bhavaty api tv ābhāsamānaiveti tātparyam. etad ādisiddhasūtre ca vyākhyātam. “The
highest level [of reality], although it is concealed to the highest point within the [Śaiva
nondualistic] scriptures, is absolutely never unmanifest; rather, it is always [in the pro-
cess of] manifesting [itself]—this is the gist [of Utpaladeva’s answer]. And [Utpaladeva]
has explained this in the verse on [the Self being] always already established.” The verse
to which Abhinavagupta is alluding here is ĪPK 1.1.2: kartari jñātari svātmany ādisiddhe
maheśvare | ajaḍātmā niṣedhaṃ vā siddhiṃ vā vidadhīta kaḥ || “What conscious Self could
produce either a refutation or a demonstration [of the existence] of the agent, the know-
ing subject, the always already established (ādisiddha) Self, the Great Lord?”
31 On the attempt in the Pratyabhijñā treatise to eliminate all rival explanations for phenom-
enal variety (i.e. that of the Sautrāntikas, but also that of the Vijñānavādins, who account
for phenomenal variety by invoking an internal mechanism of latent traces or imprints,
vāsanā) in order to show that the freedom of consciousness is the only possible cause for
it, see e.g. ĪPV, vol. I, 163–164 (quoted and translated in Ratié 2011, 367–368); see also ĪPV,
vol. I, 184–185, and ĪPVV, vol. II, 80–81 (quoted and translated in Ratié 2010, 461–464).
via free access
114 ratié
dom which, contrary to an external object, is experienced by all conscious

beings, for example when they imagine.32
3 The Context and Gist of Fragments 4–6? on ĪPK 1.5.8–9
In ĪPK 1.5.8–9,33 Utpaladeva comes back to the Sautrāntika’s contention that

the external object can and must be inferred, and he endeavours to show that
this inference is simply impossible. In fragment no. 4, he first argues that infer-
ential knowledge is conceptual in nature and that the concept in which the
inference results arises thanks to the residual trace (saṃskāra) left by a previ-
ous experience. When we draw from the fact that a distant hill is smoky the
conclusion that there must be a fire there, our concept of fire arises thanks to
our previous experiences of fire and smoke; fire can be a conceptual object for
us only because we have already witnessed fires in the past, and these previous
experiences of fire have left in the conscious stream a trace that is reactivated
when we form the inferential concept of “fire.” It is this residual trace that guar-
antees the possibility for the conceptual object to be an object manifest for
consciousness, as well as the reliability of concepts as tools in the sphere of
human practice, because it enables an object directly perceived in the past but
no longer existing to be manifested again.
The consequence of this remark is that the external object cannot be
inferred precisely because it is by nature imperceptible: it has never been man-
ifest to any consciousness (if it had, it would not be external) so that we cannot
form any concept of it.34 Utpaladeva also points out that an inference is valid
insofar as it manifests the conceptual object as particularized by a specific time
and place rather than as a pure abstraction: our knowledge that a hill is on fire
32 See Ratié 2010b.

33 anumānam anābhātapūrve naiveṣṭam indriyam | ābhātam eva bījāder ābhāsād dhetuvas-
tunaḥ || ābhāsaḥ punar ābhāsād bāhyasyāsīt kathaṃcana | arthasya naiva tenāsya siddhir
nāpy anumānataḥ || “No inference can be admitted as regards that which has never been
manifested previously. [Before being inferred,] a sense organ has already been manifested
[at least in the general form of a cause] through the manifestation of a real entity that
[happens to be] a cause, such as a seed [which is the cause of a sprout]; but the mani-
festation of an object external to manifestation has never occurred in any way. Therefore
there can be no establishment of this [external object]—not even through an inference.”
34 Cf. ĪPVV, vol. II, 161: tad anupapannaṃ saṃvidbāhyasya svapne ’py anābhātasya vikalpya-
tvāyogād anumeyatānupapannā yataḥ. “This [inference of an object particularized by its
being external to consciousness] is impossible, since [an entity] external to conscious-
ness—[and therefore] unmanifested, even in a dream—cannot be an object of inference,
because [such an entity] cannot be the object of a concept.”
via free access
because it is smoky would be useless if it were not the knowledge that the par-
ticular hill currently seen is on fire—an idea shared with the Dharmakīrtians,
particularly Dharmottara35 (who might be targeted, here as in Abhinavagupta’s
ĪPVV, as the Sautrāntikas’ most prominent champion).36 Here too, Utpaladeva
does not explicitly draw the consequence of this affirmation, but his aim is
probably to show that the external object cannot be inferred because, as he
emphasizes later in the treatise, spatial and temporal relations, which are noth-
ing but a partial incompatibility between some determined phenomena, can
only belong to manifest entities,37 whereas the external object must remain
purely indeterminate as regards form, time and space.
To this argument the Sautrāntika could object—and obviously did in a now
missing passage of the Vivṛti—that some inferences do not rest on a pre-
vious perception and actually regard an entity that is by nature impercep-
tible. According to Abhinavagupta’s commentaries, here Utpaladeva’s oppo-
nent relies on the traditional distinction between the pratyakṣatodṛṣṭa and
sāmānyatodṛṣṭa types of inference,38 and he argues that in the case of the sense
organs, the indriyas are not inferred as some particular entity directly witnessed
in the past, but as a mere generality (sāmānya), i.e. as a “mere something” (kiñ-
cinmātra) that must be assumed as a purely indeterminate cause of perception.
This enables the Sautrāntika to claim that the external object can be inferred as
the cause in general of phenomenal variety, although just as the sense organs,
it has never been perceived.39
Fragment no. 5 is a response to the Sautrāntika’s objection. Utpaladeva
claims that the reason why we can infer the sense organs as a mere indetermi-
nate cause of perception is that in fact the sense organs have already been per-
ceived, at least in the general form of a cause. Utpaladeva can afford to defend
this paradoxical thesis because he has already shown that the perception of any
particular entity is in fact the synthetic apprehension of a number of elemen-
tary phenomena (ābhāsa) that, when put together, constitute the singularity of
the perceived entity, but that, when taken separately, function as generalities
(sāmānya).40 This means that according to him, the generality “causality” is in
fact directly perceived in all our experiences of entities endowed with a causal
35 See Torella 2002, 117, note 19.

36 On the probability that Abhinavagupta at least saw Dharmottara as defending the
Sautrāntika position, see McCrea and Patil 2010, 141–142, note 71, and McCrea 2016.
39 On Abhinavagupta’s explanation of this objection see Ratié 2011b, 486–488.
via free access
116 ratié
power (for instance, in our perception of a seed capable of producing a sprout,

or in that of clay as the material cause of a pot).41
Of course, Utpaladeva’s strategy here might seem very risky, since the
Sautrāntika could argue that just as the sense organs are inferred as a mere
general cause enabling us to account for the phenomenon of perception (a
general cause already perceived in countless previous experiences of partic-
ular causes), in the same way, the external object can be inferred as a mere
generality, given that the generality “externality” is constantly experienced in
the course of everyday life: we can draw the concept of an object external to
consciousness from our past experiences of objects external to houses, villages,
mirrors, etc. If Utpaladeva concedes that the sense organs have already been
perceived as a general cause and can legitimately be inferred as such, he should
accept the same kind of reasoning regarding the external object inferred as
a mere “external thing” in general. This objection is the content of fragment
no. 6?. As explained below,42 I am not sure whether this last passage really
is a fragment of the Vivṛti; but if it is not, it must at least be a paraphrase of
it, because Abhinavagupta’s ĪPVV43 shows that Utpaladeva had expressed this
very objection in the Vivṛti.
Unfortunately, once more Utpaladeva’s answer to this objection is missing
from the margins of the manuscripts that I was able to consult; but here too, the
gist of his answer is preserved in Abhinavagupta’s commentaries on ĪPK 1.5.9.44
Utpaladeva apparently argued first that in the case of the sense organs, what
is inferred as a cause is a mere generality, whereas in the case of the exter-
nal object, the Sautrāntika tries to infer a cause that is not a mere generality
since it is particularized by the peculiar property of being external to conscious-
ness. Now, contrary to the notion of causality, this property cannot have been
included in any previous perception; indeed, we do perceive entities that are
external to other entities, but the case of an entity external to consciousness
is radically different from that of the pot being external to a house or a mirror,
because in the latter examples externality only means a spatial relationship of
proximity between two forms occupying a particular place, whereas when the
Sautrāntika talks about an object external to conscious manifestations, he does
not have in mind an object that would stand next to consciousness (since con-
sciousness is not spatially determined): he rather means an object that is not
41 As noted by Torella (2002, 116, note 18), Utpaladeva thus “include[s] the concept of
sāmānyatodṛṣṭa within the pratyakṣatodṛṣṭa.”
42 See note 79.
44 See Ratié 2011b, 491–498 and below, note 119.
via free access
consciousness. Such an object cannot be conceptualized on the basis of our for-

mer perceptions involving externality, since there is only a “verbal similarity”
(śabdasāmya) between the externality (i.e. proximity) of a pot with respect to
a house, and the externality (i.e. absolute otherness) of an object with respect
to consciousness. And the latter type of externality constitutes the very limit of
our power of conceptualization, since however wild our imagination may be,
we are incapable of picturing anything that would be by nature alien to con-
scious manifestation. This experience of our limitation as conscious beings is
paradoxically what reveals the infinite power of consciousness: the very notion
of a reality existing beyond consciousness is unthinkable and therefore absurd,
and according to the Śaiva nondualists, this means that whatever exists only
exists as a manifestation of consciousness.
4 Fragments of Utpaladeva’s Vivṛti on ĪPK 1.5.6
The words quoted in Abhinavagupta’s ĪPVV are in bold type. I have taken the
liberty of adding punctuation marks and standardizing sandhi. The margins of
S3 are often damaged and difficult to read; illegible akṣaras are noted below
with the sign “[?].” The ĪPVV quotations given in footnotes below only mention
readings found in the three consulted ĪPVV manuscripts when the text of the
KSTS edition seems likely to be corrupt.
[1]45 vyavahartrapekṣayā tāvad bhāvavyavasthā kriyamāṇāvabhāseṣv eva

nirapāyā paryavasyatīty adhikatarādṛṣṭavastuparyālocanam idaṃ svacchan-
daceṣṭitam. yathaiva46 hīndriyavicāraṃ vinā na na kiñcid upapadyate,47 tathā
45 This passage is found in S3 (where it begins on the top margin of folio 40B), S15 (top mar-
gin of folio 47A), JR (folio 82B, top margin) and (up to the quotations mentioned below in
note 51) ĪPV, vol. I, 176–177, note 170. It does not appear in D2, S2, S5, S7, S9 or SOAS, but it
is found in ĪPVV manuscript S12 (folio 121B, right and bottom margins).
46 yathaiva ] conj.; yathā S3, S15, JR, note 170 ĪPV (vol. I, 176). Cf. ĪPVV, vol. II, 131: yathaivety
asya tathaivety anena sambandhaḥ.
47 na na kiñcid upapadyate ] conj.; na kiñcid upapadyate S3, S15, JR, S12, note 170 ĪPV (vol. I,
176). The conjecture is of course tentative but I cannot understand the text as it stands.
Vincent Eltschinger and Eli Franco both suggested eliminating the negation as a corrup-
tion instead of doubling it, and they might be right. However, I personally find the double
negation more in accordance with the writing habits of Utpaladeva and Abhinavagupta
(I am assuming that the former shared with the latter a propensity for double negations).
For his part Eli Franco considers that “nobody writes like this” and that “if a second nega-
tion should be added, one would expect it in the beginning ( yathā na hi …).” (Personal
written communication.) However, there is no doubt that at least Abhinavagupta writes
via free access
118 ratié
hi kecid indriyāpavarjanenaiva ṣaḍdhātuvādinaḥ, anye taddvayavādina iti

tathaiva pratibhāsātiriktātyantaparokṣārthānveṣaṇam antareṇaiva vyavahāra-
parisamāptir ity anaṅgaṃ tadvicāraḥ.
nāpi vicārasiddhatvam48 asya49 bāhyasyānupapatter ity āha syād etad ity
evaṃ cābhāsānām evārthatve prakāśo ’rthaḥ, artho ’pi ca prakāśa iti paras-
parātmakatvam50 anayor bhavati.
[…]51
like this (i.e. by simply juxtaposing two negations, whether the sentence comprises sub-
ordinate clauses or not), and in fact such occurrences of na na are quite often found in
his works, not only in those which are not directly related to the Pratyabhijñā (see e.g.
Tantrāloka 9.251cd and 12.5a; Parātrīśikāvivaraṇa, 190, line 11; Mālinīślokavārttika 1.123ab,
etc.) but also in the ĪPV (see e.g. vol. I, 216, line 5, or vol. II, 33, last line), and they are par-
ticularly numerous in the ĪPVV (see vol. I, 36, line 21; 38, line 16; 60, line 12; 109, line 17;
181, line 12; 287, line 17; 296, line 11; vol. II, 6, line 5; 57, line 22; 82, line 3; 325, line 6; 332,
line 9; 334, line 5; 344, line 7; 347, line 11; 350, line 16; 357, line 12; 415, line 23; 436, line 21;
vol. III, 34, line 19; 66, line 13; 107, line 12 and line 20; 218, line 6; 301, line 3; 344, line 9; 366,
line 12; 369, line 7; 382, line 24). These double negations often occur with verbs the mean-
ing of which is close to upapad- (such as yujyate/yuktam, vaktuṃ śakyate, etc.) and both
the expressions na nopapadyate and na nopapannā are found in the ĪPVV (see vol. II, 334,
line 5 and 357, line 12).
48 vicārasiddhatvam ] S3, S15, JR, S12; vicārasiddham note 170 ĪPV (vol. I, 176).
49 asya ] conj.; api S3, S15, JR, S12, note 170 ĪPV (vol. I, 176).
50 parasparātmakatvam ] S3, S12, note 170 ĪPV (vol. I, 176); parasparātmatvam S15, JR. There
is a discrepancy here between the text of the Vivṛti as found in the marginal anno-
tations (parasparātmakatvam) and Abhinavagupta’s pratīka in the ĪPVV (pratītiparam-
parātmaketi). Abhinavagupta’s explanation, however, seems to require rather parasparāt-
maketi since it makes no mention of a series or succession (paramparā) of cognitions
(pratīti) and explains that ultimately the vyavahāra requires the identity of the object
with the manifesting consciousness (see below, note 92), which fits with the text given
in the marginal annotations. So I assume that parasparātmaketi first got corrupted into
paramparātmaketi and that someone then added pratīti° to try and make sense of this
“succession,” but I might be entirely wrong (this pratīka might in fact belong to a passage
of the Vivṛti that is missing in the marginal annotations; see the following note).
51 At this point the marginal annotations in S3, S15, JR and S12, as well as note 170 in ĪPV,
vol. I, 176–177, give a series of quotations that do not seem to belong to the Vivṛti itself
(?). The ĪPVV, which makes no mention of these quotations, comments on a hi and a
tat not found in the annotations: ity abhipretyoktaṃ gītāsu* yo māṃ paśyati sarvatra sar-
vaṃ ca mayi* paśyatīti tathā sarvabhūtastham ātmānaṃ sarvabhūtāni cātmanīti tathā
mahāgurubhir* ātmaiva sarvabhāveṣu sphuran nirvṛtacid vibhuḥ | aniruddhecchāprasaraḥ
prasaraddṛkkriyaḥ* śivaḥ || iti śivadṛṣṭau*. [gītāsu S3, S15, JR, S12; śrīgītāsu note 170 ĪPV
(vol. I, 176). *mayi S3, JR, S12, note 170 ĪPV (vol. I, 176); maya S15. *mahāgurubhir S3, S15,
JR, S12; mahāgurubhiḥ śivadṛṣṭau note 170 ĪPV (vol. I, 177). *°kriyaḥ S3, JR, S12, note 170 ĪPV
(vol. I, 177); °kriyā S15. *iti śivadṛṣṭau S3, S15, JR, S12; iti note 170 ĪPV (vol. I, 177).] “It is with
this [very] intention that it has been said in [Bhagavad-]gītā [6.30]: ‘He who sees me and
sees everything in all circumstances in me …’; and similarly, [in 6.29,] ‘[He sees] himself
via free access
[2]52 yo ’py ānumāniko vyavahāraḥ so ’py avabhāsamānenaiva vahninā, vikal-

papratyaye ’py avabhāto bāhya eva vahnir avasīyate. tata eva bāhya eva pravṛttir
vahnāv eva ca pravṛttiḥ, atīndriyeṇāpīndriyādinā svargāpavargādinārthena53
vikalpapratibhāsinaiva54 vyavahāraḥ. evaṃ cābhāsamānarūpataivārthatā tā-
vanmātra eva ca prayojanam.
[as] residing in all entities and all entities [as residing] in himself’; [and] similarly, the
great master [Somānanda has said] in Śivadṛṣṭi [1.2]: ‘Śiva, who is [constantly] engaged in
flashing forth, whose consciousness is blissful, who is all-pervading, whose flow of will is
not obstructed [by anything], [and] whose perception and action are [constantly] flow-
ing, is none other than the Self of all entities.’” S3 then gives the Vṛtti (bottom left margin)
followed by what seems to be the Vivṛti again; in JR this passage, in the right margin, is
clearly separated from the Vṛtti, which is copied at the bottom of the page (and indicated
as such at the end: iti vṛttiḥ), and immediately followed by the same passage apparently
taken from the Vivṛti. The same happens in S15 (where the Vṛtti and the remainder of
the annotation are found in the left margin). Note that note 6 of the Śivadṛṣṭi edition,
4, which is apparently borrowed from a marginal annotation in manuscript D3 (the latter
being in all probability the manuscript “a” used by the KSTS editors), gives the same quota-
tions from the Bhagavadgītā while commenting on this verse, and introduces them with
a text almost identical to that found here: evaṃ cārthasya prakāśatve prakāśo ’py arthaḥ,
artho ’pi prakāśaḥ. “And [since it is] so, since the object is the manifesting consciousness,
the manifesting consciousness is the object, and conversely the object is the manifest-
ing consciousness.” My assumption that the Śivadṛṣṭi manuscript called “a” by the KSTS
editors is none other than D3 is based on the fact that the description of “a” (Śivadṛṣṭi,
I–II) closely corresponds to D3 (which unfortunately I was only able to examine very
quickly at the National Archives of India) as regards the number of folios, the average
number of lines per folio, and the average number of akṣaras per line, not to mention
that both manuscripts bear many marginal annotations (at least some of which are iden-
tical). Both, moreover, are bound together with manuscripts of the ĪPV, Parātrīśika and
Paramādvayadvādaśikā. It should also be noted that another Śivadṛṣṭi manuscript origi-
nally owned by the Śrinagar Research Library is now preserved in the National Archives
of India (D4); I could not consult it as it had been sent for repair at the time of my visit to
the Archives, but from the catalogue description I suspect that it might be the Devanāgarī
transcript called “b” by the KSTS editors. A thorough analysis of D3 and D4 would there-
fore certainly be a welcome addition to the examination of some Śivadṛṣṭi manuscripts
provided in Nemec 2011, 79–90.
52 This passage occurs in S3, folio 40B (third line from the bottom of the page), S15, folio 47A
(left margin) and JR, folio 82B (bottom margin, from line 2 onwards, then on the right mar-
gin). It is not found in D2, S2, S5, S9, S10, SOAS or in the footnotes of the KSTS edition of
the ĪPV, but it also occurs in the ĪPVV manuscript S12 (folio 121A, bottom margin).
53 svargāpavargādinārthena ] conj.; svargāpavargādinā nārthena S3, S15, JR; sargāpavargā-
dinā nārthena S12. Cf. ĪPVV, vol. II, 132, quoted below, note 95.
54 S3, S15, JR and S12 all bear this reading, and although the pratīka in ĪPVV, vol. II, 132 is given
as vikalpapratibhāsiteti, at least one ĪPVV manuscript has vikalpapratibhāsineti (see below,
note 95).
via free access
120 ratié
sā tv ābhāsāt pṛthagbhūtānām eveti ko ’yaṃ graha ābhāsaikatvena kā vyava-

hārakhaṇḍanā, tad āha te cābhāsātmakāḥ santv iti.
kevalam ūrdhvam ābhāsamānatāyāḥ pūrvaṃ ca55 yadi te na syus tadā-
bhāsamānataivākasmikī,56 kāryakāraṇabhāvo jñāpyajñāpakabhāvaś ca nopa-
padyata57 iti paryanuyojyam.
[…]
[3]58 atha bāhyārthāḥ59 paramāṇava eva niravayavāḥ sañcitās tathāpi
ghaṭo vitatarūpatvena prathamānaḥ pūrvāparādibhāgavān eva prathate, na
caivaṃ paramāṇumayatve ghaṭate. tathā hi yo ’sau pūrvābhimataḥ paramā-
ṇus tena dvitīyaḥ paramāṇuḥ saṃśliṣṭamānas tadekatām āpadyeta60 niravaya-
vayor hi saṃlagnatve kiyad asaṃlagnam avaśiṣyate, saṃlagnau cānyonyasvarū-
paparimagnāv ity ekaparamāṇvābhāsataiva,61 punar apy aparaparamāṇusaṃ-
sparśe62 tathaivety anantaparamāṇuyojane ’py ekaparamāṇumātraprakāśaḥ63
syāt, so ’pi vā na syād ekaikaparamāṇor64 atīndriyatvāt.65
55 ca S3, JR, S12; om. S15.

56 tadābhāsamānataivākasmikī ] S15, JR, S12; tadā[?]samānataivākasmikī S3.
57 ca nopapadyata ] conj.; copapadyata S3, S15, JR, S12.
58 This passage is again found in S3 (folio 41A, top margin, from line 3, continuing on the
left margin), S15 (from folio 47B, bottom margin, to folio 48A, top margin), JR (folio 83B,
bottom margin, immediately after a quotation from the ĪPVV corresponding to vol. II, 141,
and folio 84A, top margin) and S12 (folio 124B, top margin). In all of these manuscripts it is
preceded by etad evāha svavṛttivivṛtigranthakāra utpalācāryaḥ (S12 has tad āha instead of
etad evāha). It is also found in D2 (folio 108B, right margin), SOAS (folio 100B, top margin)
and note 192 in ĪPV, vol. I, 180, in all three cases from ghaṭo vitatarūpatvena prathamānaḥ
… (with a hi inserted after ghaṭo) to atīndriyatvāt. I could not find it in S2, S5, S7 and S9.
59 bāhyārthāḥ ] conj.; bāhya .. S3, S15, JR, S12.
60 āpadyeta ] S3, JR, D2, S12; āpadyete SOAS; āpadyate note 192 ĪPV (vol. I, 180). It is impossible
to decide whether S15 reads āpadyeta or āpadyate.
61 ekaparamāṇvābhāsataiva ] S3, S15, JR, D2, S12; ekaparamāṇur bhāsetaiva SOAS, note 192
ĪPV (vol. I, 180).
62 aparaparamāṇusaṃsparśe ] S3, S15, JR, D2, S12; aparaparimāṇusaṃsparśe SOAS; apara-
paramāṇusaṃsparśo note 192 ĪPV (vol. I, 180).
63 ekaparamāṇumātraprakāśaḥ ] S3, S15, JR, D2, S12, note 192 ĪPV (vol. I, 180); ekaparimāṇu-
mātraprakāśaḥ SOAS.
64 ekaikaparamāṇor ] S3, S15, JR, D2, S12, note 192 ĪPV (vol. I, 180); ekaikaparimāṇor SOAS.
65 From this point onwards, the marginal annotations in S3, S15, JR and S12 quote ĪPVV, vol. II,
141 (evaṃ kramasaṃśleṣe doṣaḥ, etc.) without mentioning this change (but the sentence
has been put in what looks like parentheses in S12). The annotation on the right margin
in S3 is also taken from the ĪPVV (see vol. II, 142: ṣaṭkoṇe devanākṣādau, etc.).
via free access
5 Fragments of Utpaladeva’s Vivṛti on ĪPK 1.5.8–9
[4]66 anumānaṃ hi vikalpaḥ, sa ca pūrvānubhavasaṃskārasamuttha iti tāvad

arthasya pūrvapratyakṣatāpekṣā,67 tasmiṃś cārthe vikalpapratītir avyabhicā-
rajanitānumānam,68 tac cānubhavasaṃskārād antaḥsthitatadartham idan-
taucityena yāvad vimṛśati tāvad vicchedena prathayati. na ca tāvatā pravṛt-
tiviṣayo ’sāv arthaḥ, tena prakāśito bhavati viśiṣṭadeśakālo viśiṣṭadeśakālasyaiva
svalakṣaṇasya69 prāpyatvenābhīṣṭārthakriyākāritvena70 ca pravṛttiviṣayatāyo-
gyatvād71 atas tathaiva yad vimarśanam, tad āha pūrvāvabhāteti72 nāntarīya-
kārtheti73 tattaddeśakālādīti.74
[…]75
[5]76 indriyam api pūrvānubhūtam eva, na hi tat prātisvikenākāreṇānu-
mīyate, api tu sāmānyena, tad āha vṛttau kiñcinmātraṃ nimittam iti. ayam
artho nimittatāsāmānyam, pūrvapratibhātam eva bījasyāṅkuranimittatājñāne
66 This fragment appears in marginal annotations in S3 (top of folio 43A, line 1, after a quo-
tation from the ĪPVV, then on the right margin—see figure 5.1), S15 (folio 50A, top and
left margins), JR (folio 87B, right margin) and S12 (folio 130A, left and bottom margins); I
could not find it in D2, S2, S5, S7, S9 and SOAS.
67 pūrvapratyakṣatāpekṣā ] S3, JR, S12; pūrvapratyatāpekṣā S15.
68 I assume that the reading avyabhicārīti in ĪPVV, vol. II, 161 is a corruption for avyabhicāreti
(see below, note 108).
69 viśiṣṭadeśakālo viśiṣṭadeśakālasyaiva svalakṣaṇasya ] conj.; viśiṣṭakālasyaiva svalakṣaṇa-
sya S15, JR; viśiṣṭadeśakālasvalakṣaṇasya S12; vi[??]deśakāla[?]svalakṣa[?]sya S3. This is
merely tentative, but from the meaning of the passage as it is summed up by Abhi-
navagupta (see below, note 110), I assume that something is missing in the text of the
manuscripts.
70 prāpyatvenābhīṣṭārthakriyākāritvena ] S15, JR, S12; prāpya[?]nābhīṣṭārthakriyākā[?]tvena
S3.
71 pravṛttiviṣayatāyogyatvād ] S15, JR, S12; pravṛttiviṣayatāyo[?]tvād S3.
72 pūrvāvabhāteti ] S15, JR, S12; pūrvāva[?]teti S3.
73 nāntarīyakārtheti ] S15, JR, S12; nāntarīya[?]rtheti S3.
74 tattaddeśakālādīti ] corr.; taddeśakālād iti S15, JR, S12; taddeśakālād[?] S3. Cf. Vṛtti, 21–22,
quoted below, note 111.
75 Here a part of the text is missing, as is obvious from Abhinavagupta’s commentary (ĪPVV,
vol. II, 162), which mentions the words eṣa punar and vyāptivimarśāt, and again
(ĪPVV, vol. II, 163) dharmapratyakṣa. The next fragment is commented in ĪPVV, vol. II,
163.
76 The following fragment appears in marginal annotations in S3 (folio 43A, right margin—
see figure 5.1), S15 (folio 50A, left margin), JR (folio 87B, right margin) and S12 (folio 130A,
bottom margin), in all cases immediately after the preceding Vivṛti fragment. The first sen-
tence up to sāmānyena is also found in note 236 in ĪPV, vol. I, 188, but I could not find it in
D2, S2, S5, S7, S9 and SOAS.
via free access
122 ratié
mṛdāder ghaṭādinimittatājñāne.77 na hi tatra viśiṣṭaṃ bījam eva nimittatvena

pratibhātam, mṛdo ’py ajñānaprasaṅgāt. nāpy uttaratra mṛd eva bījāder
ajñānaprasaṅgāt. prasaktataditaravyavacchedāyaiva viśeṣaṇaṃ prayujyate
cātrāpi, mṛd eva kāraṇaṃ ghaṭasya78 na bījam iti.
[…]
[6]?79nanu80cakṣurādiviśeṣaparihāreṇa yathā bījādyanubhavāt81 kāraṇatā-
sāmānyam82 anubhūtam evoktam, tathā śarīrād gṛhād grāmāder vā bāhyasyā-
nubhavād bāhyatvam api sāmānyenānubhūtam evājñānaviśeṣeṇa83 bāhyatvā-
nanubhavoktau vā cakṣurāder api kāraṇaviśeṣasyānanubhūtapūrvatvān na
syād anumānaṃ tat kathaṃ nānumānasya vyāpāra84 ity uktiḥ.
6 Translation of the Fragments on ĪPK 1.5.6
[1] To begin with, as far as agents of ordinary human practice (vyavahartṛ)

are concerned, it is on [the basis of] mere phenomena that [they] manage to
77 mṛdāder ghaṭādinimittatājñāne ] S3, JR, S12; om. S15.

78 kāraṇaṃ ghaṭasya ] S15, JR, S12; kāraṇaṃ gha[??] S3.
79 This annotation is not found in S3, S15, JR or S12, but it appears in both SOAS (folio 106B,
upper right corner) and D2 (folio 115B, upper and right margins); it is also found in note 248
in ĪPV, vol. I, 190. In all three cases this passage is preceded by a quotation from ĪPVV, vol. II,
165 (from bāhyatvaviśiṣṭo ’rthaḥ up to kramikābhāsavaicitryāt). This ĪPVV quotation is also
found in S9, S10 and followed in all cases by ity āśaṅkyāha grāmagṛhāder iti. But after
that point, while S9 and S10 add ayam āśayaḥ and then quote again the ĪPVV from nedaṃ
bāhyatvaṃ ghaṭasya onwards, SOAS, D2 and the ĪPV note give tathā cātra pūrvavyavasthā
sphuṭam eva dṛśyate yathā, and then the text quoted above. There is very little direct evi-
dence that this annotation is indeed a quotation from the Vivṛti: the only ĪPVV passage that
could comment on it apparently explains a sentence beginning with nanu cakṣurādi° (see
ĪPVV, vol. II, 165), whereas no nanu is to be found in the annotations and I had to conjec-
ture it. Nonetheless, it seems to me quite probable that the words preceding cakṣurādi°
are not part of the quoted text, as I cannot understand the sentence with the presence of
the word yathā twice (the KSTS edition has vā after the second yathā, although this vā is
found neither in D2 nor in SOAS; but even with this additional vā I fail to see the structure
of the sentence). Besides, as shown in Ratié 2011b, 491–499, this passage expresses a crucial
objection against Utpaladeva’s reasoning, and it is at least certain from Abhinavagupta’s
explanations in the ĪPVV (see below, note 119) that the Vivṛti stated this very objection in a
passage beginning with nanu cakṣurādi°. So it seems likely that this passage is a genuine
fragment from the Vivṛti, although this is by no means certain.
80 nanu ] conj.; om. D2, SOAS (see above, note 79).
81 yathā bījādyanubhavāt ] D2, SOAS; yathā vā bījādyanubhavāt note 248 ĪPV (vol. I, 190).
82 kāraṇatāsāmānyam ] conj.; kāraṇasāmānyam D2, SOAS, note 248 ĪPV (vol. I, 190).
83 evājñānaviśeṣeṇa ] conj.; eva jñānaviśeṣeṇa D2, SOAS, note 248 ĪPV (vol. I, 190).
84 vyāpāra ] D2, note 248 ĪPV (vol. I, 190); vyāpārā SOAS.
via free access
successful[ly] ascertain85 the existence (vyavasthā) of [this or that] entity—

an ascertainment in which [they necessarily] engage [in their ordinary activi-
ties]; so this pondering over an unperceived reality that is [supposedly] some-
thing more (adhikatara) [than phenomena and the consciousness manifesting
them]86 is a [purely] arbitrary endeavour!87 For it is not [universally acknowl-
edged that] no [perception] at all can be accounted for (upapadyate) without
speculating about the [imperceptible] sense organs;88 thus some89 propound
85 See ĪPVV, vol. II, 131: nirapāyety apāyaḥ pramāṇābhāvo dūṣaṇasaṃbhavaś ca, tadrahitā.
“[This ascertainment] is successful[ly performed] (nirapāya), i.e.,] it is devoid of the fail-
ure (apāya) [consisting in] the lack of a means of [valid] knowledge, and [it is devoid of]
the possibility [that the means of valid knowledge] may be refuted.”
86 The word adhikatara is the comparative form of adhika, which by itself means “some-
thing more.” According to Abhinavagupta, Utpaladeva uses it because if the manifesta-
tions of objects in consciousness are something more than the consciousness manifesting
them, the so-called external object must be considered as something more than these very
manifestations: the external object must be considered as existing over and above phe-
nomena and the consciousness taking their form—which is absurd. See ĪPVV, vol. II, 131:
adhikatarety ābhāsā eva darpaṇāt pratibimbānīva saṃvedanād adhikāni, yac ca tato ’py
adhikataraṃ na kvacid bhāvi tat tata eva* dṛṣṭaṃ tac ca vastu katham. [*tat tata eva conj.;
tata eva J10, J11, T, ĪPVV]. “ ‘adhikatara’ [means the following]: the [various] phenomena
are [something more (adhika)] than consciousness, just as reflections are something more
than a mirror [reflecting them]; and that which is something more than something more,
[i.e., that which is something more] than these very [phenomena,] can never be perceived
in any [circumstance] for the very [reason that it is distinct from phenomena]; and how
could that be a [real] entity (vastu)?”
87 Cf. ĪPVV, vol. II, 131: svacchandaceṣṭitam iti na pramāṇabalopanatam ity arthaḥ. “[It is] ‘a
[purely] arbitrary endeavour,’ i.e. it is not brought about by the force of [necessity inherent
in] a means of [valid] knowledge.”
88 Cf. ĪPVV, vol. II, 131: dṛṣṭaṃ cedaṃ prāmāṇikānāṃ yat saty api vastuny anupayogināṃ-
śena na prayasyanti vicāritena yathendriyādiṣu ṣaḍdhātusamīkṣākārādyā bārhaspatyā vā.
ekair* hi bhūtapañcakaṃ cetanā cetīyati vyavahārasamāptir aṅgīkṛtā tatraivānyasyen-
driyāder anupraveśāt. anyair api bhūtacatuṣṭaye vicitramelanoditasaṃvedanākhyavikāra-
viśeṣe ’nuditatadvikāre ca grāhakagrāhyavyavahārasamāptir upagatā. [*ṣaḍdhātusamī-
kṣākārādyā bārhaspatyā vā T; ṣaḍdhātusamīkṣākārādyo vārhatyo vā J10, J11; ṣaḍdhātusamī-
kṣākārādyo bā(vā)rhatyo vā ĪPVV. *ekair J10, J11, T, Ka (quoted in ĪPVV, vol. II, 131, note 1);
etair ĪPVV.] “And [we] observe the [following] among philosophers (prāmāṇika): even if
[something] exists [as] a real entity, they do not concern themselves with the aspect [of
it] that is useless [and] speculative (vicārita)—for exemple, regarding such [impercepti-
ble things] as the sense organs, [this is the case of philosophers] such as the author of the
Ṣaḍdhātusamīkṣā, or of followers of Bṛhaspati. For the former acknowledge that ordinary
human practice is accounted for if this much [is admitted]: the five elements and con-
sciousness, because such other [things as] the sense organs are included in these; whereas
the latter admit that the ordinary human practice [consisting in the relationship between]
an apprehending [subject] and an apprehended [object] is accounted for if a particular
transformation called ‘consciousness’ arises in the four elements from [some of their] var-
via free access
124 ratié
the theory of the six elements (ṣaḍdhātu) while not taking the sense organs
into account in any way, [and] others90 defend the theory of the two [sorts
ious combinations, and if this transformation does not arise [from other combinations of
the four elements].” On the author of the Ṣaḍdhātusamīkṣā and the followers of Bṛhaspati
see below, notes 89 and 90, and Ratié 2018.
89 According to Abhinavagupta’s commentary (see above, note 88), their “theory of the
six elements” (ṣaḍdhātuvāda) includes a set of five elements (bhūtapañcaka)—i.e. most
probably earth, water, fire, wind and ākāśa—to which consciousness (cetanā) is added. At
first sight one might assume that Utpaladeva and Abhinavagupta had a Buddhist author
in mind (on the Buddhist theory of the ṣaḍdhātu see e.g. Abhidharmakośabhāṣya 28
[on Abhidharmakośa 1.27]: ya ime tatra ṣaḍ dhātava uktāḥ pṛthivīdhātur abdhātus tejo-
dhātur vāyudhātur ākāśadhātur vijñānadhātuḥ …; cf. La Vallée Poussin 1923, 49, note 2 for
sources). But according to Abhinavagupta’s commentary (ibid.), here Utpaladeva is rather
hinting at the Ṣaḍdhātusamīkṣā—a lost work ascribed to Bhartṛhari. This treatise, alluded
to by Somānanda in Śivadṛṣṭi 2.73 as the Samīkṣā and mentioned by Utpaladeva himself
in Śivadṛṣṭivṛtti, 83–86, is also known as the Śabdadhātusamīkṣā (see e.g. Iyer 1969, 9–
10), but as already pointed out in Torella 2002, xxvi–xxvii, note 39 (cf. Torella 2014, 573),
the latter title, which appears in the KSTS edition of the Śivadṛṣṭivṛtti, is a corruption
(note that as seen above, note 88, the ĪPVV gives the title as Ṣaḍdhātusamīkṣā). On the
few known fragments of it and what might have been the overall goal of the work, see
Ratié 2018. As pointed out there, Abhinavagupta makes some interesting remarks regard-
ing this ṣaḍdhātuvāda while commenting on another part of the Pratyabhijñā treatise
(ĪPVV, vol. I, 93): sāṅkhyasya yady api pañcaviṃśatis tattvāni sphuranti, tathāpi pañcaiva
bhūtāni cetanā cety etāvanmātre tāvat sākṣātkārarūpo ’nubhavo nādhike. tata eva tatra-
bhavato dhātuṣaṭkanirūpaṇa eva viśvaṃ nirūpitaṃ bhavatītyāśayena tatsamīkṣodyamaḥ.
saṃbhāvanānumānāgamajaniteṣv api jñāneṣu tad eva bhāsate, kevalaṃ yojanāmātram
adhikam. yathā pṛthivy eva yā surabhyasurabhyādivicitragandhā tatraiva viśeṣatyāgena
saukṣmyeṇa ca gandhatanmātrarūpateti. “Even though for a [follower of] Sāṅkhya, the
twenty-five principles (tattva) are manifest [as the universe], to begin with, experience
(anubhava), that is, immediate perception, consists in nothing but this: the sole five ele-
ments (bhūta) and consciousness (cetanā)—and nothing more. This is why for the master
[Bhartṛhari] (tatrabhavant), the universe is [entirely] explained as soon as the six ele-
ments are explained—it is with this intention that he has undertaken their Examination
(Samīkṣā). It is also this [set of six elements] that is manifest in cognitions that arise
from hypothetical inferences or scripture [and not only in perceptions; and] anything
else is nothing but a mere combination [of these elements]. For example, [the prop-
erty of] consisting in the subtle sensory object (tanmātra) of smell lies in the [element
(dhātu) of earth]—that same earth that may possess various smells, some pleasant, oth-
ers unpleasant, etc.—insofar as [earth] is devoid of particularities and subtle.” So it seems
that Utpaladeva alluded to this theory at this point of the Vivṛti because according to the
author of the Ṣaḍdhātusamīkṣā, perception can be explained in its entirety as the result of
a mere combination of the six elements, without inferring the existence of imperceptible
indriyas that would be fundamentally distinct from these elements; or, as Abhinavagupta
puts it (see above, note 88), according to this theory the indriyas are already included
within the six elements.
90 If we follow Abhinavagupta’s interpretation of this passage (see above, note 88), these
via free access
of combinations of four elements—namely, the sort that produces conscious-

ness and the one that does not—without taking imperceptible sense organs
into account either]. And exactly in the same way, ordinary human practice
[can] be entirely accounted for (parisamāpti) without any investigation about
an absolutely imperceptible object [considered as] something more than phe-
nomena. Therefore speculating about this [absolutely imperceptible object] is
[nothing but] air.
Nor [can] the [external object] be established by any [such] speculation,
because of the impossibility of this external [object]—this is what [the verse]
“others” are “followers of Bṛhaspati” (bārhaspatya), i.e. some materialists (cārvāka) who
embrace the ideas expressed in the sūtras traditionally ascribed to Bṛhaspati (on the
known fragments of this work and the probability that it was written around the sixth
century, see Namai 1976, Bhattacharya 2002 and Franco 2011, 634–636). According to Abhi-
navagupta, they understand the relationship between the apprehending subject and the
apprehended object as the mere result of various combinations between four material
elements. Thus the fragments of the Bṛhaspatisūtras do not deny the very existence of
sense organs, but they refuse to see them as distinct entities: whereas the Nyāya and
Vaiśeṣika theories of vision for instance present the visual organ as an imperceptible ray
of light fundamentally different from the perceiving body and perceived object (see e.g.
Preisendanz 1989), the materialists see the indriyas as mere aggregates made of the same
matter constituting the subject and object. See fragments A2 to A4 in Namai 1976, 39 (cf.
Bhattacharya 2002, 603–604 and Franco 2011, 635): pṛthivy āpas tejo vāyur iti tattvāni || tat-
samudāye śarīrendriyaviṣayasañjñāḥ || tebhyaś caitanyam || “Earth, water, fire and wind are
the principles. [It is merely] their aggregates that the terms ‘body,’ ‘sense organ’ (indriya)
and ‘object’ designate. Consciousness [arises] from these.” Besides, according to the “fol-
lower of Bṛhaspati” whose views are mocked by Jayanta Bhaṭṭa (the term bārhaspatya
appears e.g. in Nyāyamañjarī, vol. I, 496 and Āgamaḍambara, 63), inferences are valid
only if they regard entities that have already been perceived, but not if they are meant
to establish the existence of entities that are by nature imperceptible, such as God and
the Self. See Nyāyamañjarī, vol. I, 326: dvividham anumānam, kiñcid utpannapratīti kiñcid
utpādyapratīti. īśvarādyanumānaṃ tūtpādyapratīti. tatra dhūmānumānādeḥ prāmāṇyaṃ
kena neṣyate | ato hi sādhyaṃ budhyante tārkikair akṣatā api || yat tv ātmeśvarasarvajña-
paralokādigocaram | anumānaṃ na tasyeṣṭaṃ prāmāṇyaṃ tattvadarśibhiḥ || “Inference is
of two sorts: one [concerns an object] the perception of which has [already] occurred [at
some point]; the other [concerns an object] the perception of which has [yet] to occur.
But the inference of [entities] such as God [concerns an object] the perception of which
has [yet] to occur. Among these [two types of inference,] who would not admit the valid-
ity of an inference such as that [of fire] from smoke? So [people] apprehend what is to
be established [by such an inference] even though they are not pestered by logicians. But
the validity of an inference regarding such [entities] as the Self, God, an omniscient or an
afterlife is not acknowledged by those who know reality.”
Whether the materialists of Jayanta’s time had also explicitly targeted the inference of
the indriyas or not, Jayanta criticizes their thesis by emphasizing that it leaves them with
no choice but to reject the inference of imperceptible indriyas as well. See Nyāyamañjarī,
vol. I, 501: śrotrādyanumāne ’pi yathodāhṛte śakyam evam abhidhātum … “And also with
via free access
126 ratié
says with “syād etat.”91 And [since it is] so, since only phenomena [can] be
objects, the manifesting consciousness (prakāśa) is the object, and conversely
the object is the manifesting consciousness; therefore they consist in each
other.92
[…]93
respect to the inference of the hearing organ and so on, according to the way in which it
is formulated, one can declare similarly, [as in the case of the inference of God or the Self,
that it is invalid] …” Cf. Āgamaḍambara, 68, where the Saiddhāntika arguing with a Cār-
vāka who denies the validity of the inference of īśvara points out that then the Cārvāka
should equally dismiss the inference of the indriyas. The Cārvāka replies: “Let [us admit
that] the [sense organs] too are not inferred; why should it be a problem for us?” (tad api
mānumāyi, kiṃ naś chinnam); the Saiddhāntika asks in turn: “But [then] how can you see
a visual form without any visual organ?” (nanu cakṣuṣā vinā kathaṃ rūpaṃ drakṣyasi). See
Dezső 2005, 168–169. On the possibility that the materialist portrayed by Jayanta might be
a caricature of a ninth-century Kashmirian materialist, poetician and grammarian named
Udbhaṭa, see Solomon 1977–1978, Bronkhorst 2008 and Bronner 2016. Cakradhara, while
commenting on the Nyāyamañjarī, explicitly identifies Jayanta’s materialist target as Udb-
haṭa, to whom he ascribes a commentary on the Bṛhaspatisūtras (see Solomon 1977–1978,
988–990), and it is not impossible that Utpaladeva and Abhinavagupta also knew this
work.
91 Utpaladeva has just argued that there is no universal acknowledgement of the necessity of
positing the existence of imperceptible entities (such as the indriyas) so as to account for
perception. He now adds that even granted that we admit this necessity (hence the “Let
us admit this,” syād etat, in the verse), such an argument is in fact invalid because, as he
is about to show, the existence of the external object is contradictory to reason. Cf. ĪPVV,
vol. II, 131: anupayogino ’pi vastuvicāracakṣuranveṣyatvāt* tattvam, iha tu bāhye ’rtha idam
api nāstīty āha nāpīti. [*vastuvicāracakṣuranveṣyatvāt conj.; vastuno vicāracañcuranviṣya-
tāt ĪPVV, vastuvicāracañcuranveṣyatāt T; vastuvicāracaṃcaranviṣyatāṃ J10, J11.] “[With]
‘Nor …’, [Utpaladeva] says that [something may be considered as] having a reality (tattva),
even though it is useless [in ordinary human practice], on the grounds that [the existence
of] the visual organ [for instance] is required when speculating on the real entity [that
contributes to the production of a visual perception]; but in the [case of the] external
object, there is not even [room for] such an [assumption].”
92 See ĪPVV, vol. II, 131: parasparātmaketi. krayavikrayāder api tathāprakāśanam eva prāṇāḥ.
yad evedaṃ yasyetthaṃ*prakāśate tad etenaiva mama prakāśamānasya prakāśamānāyām
evārthakriyāyām upayogaḥ. [*parasparātmaketi conj. (see above, note 50); pratītiparam-
parātmaketi ĪPVV, J10, T; pratītiparasparātmaketi J11. *yasyetthaṃ conj.; asyetthaṃ ĪPVV,
J10, J11, T.] “[They] ‘consist in each other’ [means the following]: even the essence of [ordi-
nary human practice] such as buying and selling, etc., is nothing but such a manifesting
consciousness. Only that which is manifest in such a way is for this very reason useful
[in providing] to me, for whom [this thing is manifest and while I am] being manifest
[myself], an efficacy which is [itself] necessarily manifest!” That is to say, as far as I under-
stand the passage: objects can be desired and aimed at by a subject in the sphere of human
practice only if they are forms taken on by the subject’s consciousness, and are ultimately
nothing but the manifesting consciousness itself.
via free access
[2] Even an ordinary human practice that is based on an inference[—such

as trying to reach a fire the presence of which is merely inferred from the per-
ception of smoke—can only occur] thanks to a fire that is necessarily being
manifest [at the very time of this endeavour]; even in a conceptual cognition,
fire is determined [as being] external [to consciousness only insofar as] it is
manifested.94 It is for this reason that an activity [can occur] with respect to
94 Cf. ĪPVV, vol. II, 131–132: bāhya eveti saugatais tāvad adhyavasāyabalāt prāmāṇyaṃ
vadadbhir anumānam api bāhyaviṣayam evety uktaṃ prāmāṇyaṃ vastuviṣayaṃ dvayor
apīti. asmanmate tu vikalpasyāvaśyaṃ vastuniṣṭhatvam evety uktaṃ bhrāntitve cāvasāya-
syetyādyantare. tata eva yad āha bāhyavādī vikalpasyāsatyaprakāśanatvam iti, tad asman-
mate niravatāraṇam eva. sākārajñānavādinā vikalpaḥ svātmani prakāśaḥ, arthe tv avasāya
ity ukte katham asatprakāśanam. tad āstāṃ tāvat. “[Here is what Utpaladeva means with
the words] bāhya eva. To begin with, the Buddhists, when saying that the validity of a
means of knowledge arises from the force of determination (adhyavasāya), admit that
even inference must have an object that is external [insofar as it is based on the determina-
tion of a previously perceived object. This is what Dharmakīrti has said in Pramāṇaviniś-
caya 2.7b]: ‘In both [inference and perception], the validity [of the means of knowledge]
concerns a real thing …’. But in our doctrine, the concept necessarily rests on the real thing,
[even at the time of conceptualization: Utpaladeva] has [already] stated this elsewhere,
for instance in [verse 1.3.5 beginning with] bhrāntitve cāvasāyasya. For this very reason,
in our system the [Buddhist] externalist’s claim that a concept involves no real manifes-
tation cannot be accepted at all: since the proponent of the theory that cognition has
aspects says that a concept is [immediately] manifest in itself [insofar as every cognition
is immediately aware of itself,] even though with respect to the object, [this concept] is
a [mere] determination, how could it have a nonexistent manifestation? So enough with
this.”
Here is the gist of Abhinavagupta’s reasoning as I understand it. The Buddhists see
conceptualization as bearing on a mere generality or a pseudo-object resulting from a
process of exclusion (apoha); yet they acknowledge (as Dharmakīrti in the Pramāṇaviniś-
caya) that a valid inference has as its object a real entity inasmuch as its object is capable
of efficacy (see Pramāṇaviniścaya 2, 48: … arthakriyāyogyaviṣayatvād vicārasya, “because
a [valid inferential] speculation has an object that is capable of efficacy”). According
to Dharmakīrti, this is the case because although the generalities aimed at in concepts
are no real entity (since reality is purely singular), they result from the determination
(adhyavasāya) as one single entity of many perceived singular entities that have a some-
what similar efficacy (thus the concept of “fire” results from our determining as one many
perceived fires that are all in fact different from each other but share the capacity to
burn, etc.). So even inferences are somehow based on at least one aspect of perceived
entities (namely arthakriyā), and this is why valid inferences enable us to obtain desired
entities and to shun undesirable objects. Nonetheless, the Buddhists consider that only
perceptions involve an immediate manifestation of the object, and the Śaivas see this
as a contradiction in the Dharmakīrtian system: according to Abhinavagupta, the Bud-
dhist theory of determination entails that conceptual objects too are directly manifested
at the very time of their conceptualization (and not only in some perception anterior to
the conceptual elaboration), at least as regards their efficacy (see e.g. ĪPVV, vol. II, 132:
via free access
128 ratié
[something determined as being] external, and that an activity [can occur]

with respect to [something determined as being] fire. Ordinary human practice
[can even occur] with an object such as the sense organs, or heaven and liber-
ation, although [these always remain] beyond the reach of the sense organs,
[but] only inasmuch as they are [somehow] manifest in the concept [repre-
senting them].95 And [since it is] so, being an object is nothing but having a
arthakriyāpi cābhāsamānatayaivābhilaṣaṇīyeti sāpy ābhāsaviśrānteti. “And even efficacy

can be desired only insofar as it is being manifest; therefore it too rests on manifestation.”).
Besides, the Buddhist epistemologists cannot deny that concepts involve a real, immediate
manifestation since they consider that all cognitions have two aspects (“an aspect [con-
sisting in] the object,”viṣayākāra, and “an aspect [consisting in] themselves,” svākāra), and
that even concepts, insofar as they possess self-awareness (svasaṃvedana, svasaṃvitti),
involve an immediate manifestation of themselves (on this idea and its appropriation by
the Śaivas see Ratié 2011, 44–45, note 20).
Abhinavagupta thus emphasizes in this passage the major point of contention
between the Śaiva nondualists and their Buddhist counterparts as regards concepts:
according to Utpaladeva, all concepts involve the immediate manifestation of a real thing
at the very moment when they arise, because there can be no conscious representation of
something that consciousness cannot present to itself in an immediate way; without such
a manifestation, concepts could not occur or would be similar to a state of unconscious-
ness. And this means that even objects conceptualized as being external are ultimately
internal aspects of consciousness manifesting itself as if it were external to itself. See e.g.
ĪPVV, vol. II, 129–130: anumito ’pi bāhyo ’rthaḥ prakāśamāna eva vaktavyaḥ, prakāśād bhede
hy aprakāśanaprasaṅgād anumitatvam api vastuno mūrchāprāyaṃ bhavet. “Even [if] an
external object [is] inferred[, it] can be talked about only insofar as it is being manifest,
for if [it] were distinct from the manifesting consciousness (prakāśa), since as a result it
would not be manifest, [the awareness of] the very fact that the entity is inferred would
amount to a state of stupor!”
95 Cf. ĪPVV, vol. II, 132: vikalpapratibhāsinety* upalabdhikriyākaraṇatayā sāmānyendriyaṃ,
niratiśayasukhatatsādhanaviṣayatayā svargaḥ, anāhṛtasahajānandaghanataikasāra-
paripūrṇatātmanāpavargaś* cābhāsabhuvam abhyety* eva, anyathābhilaṣaṇīyatvatatsā-
dhanānveṣaṇatadrūpadeśatadavadhāraṇatadanuṣṭhānādyayogāt. [*vikalpapratibhāsine-
ty J11; vikalpapratibhāsitety ĪPVV; vikalpapratibhāsanety T, J10. *°paripūrṇatātmanāpavar-
gaś T; °pūrṇatāpavargaś ĪPVV, J10, J11. *cābhāsabhuvam abhyety T; cābhāvabhuvam abhye-
ty ĪPVV; cābhāva .. satyety J10; cābhāvam abhyety eva J11.] “[The passage] ‘inasmuch as they
are [somehow] manifest in the concept [representing them’ means the following]. The
‘sense organ,’ [apprehended as] a generality (sāmānya) [by conceptual thought, i.e.,] as
the instrument of the action of perceiving, [as well as] ‘heaven,’ [apprehended] as the
object of unsurpassed pleasure and as [the means of] realizing it, and ‘liberation,’ [appre-
hended] as consisting of an absolute fullness the essence of which is nothing but the
plenitude of a bliss that is not brought about [because in fact it is] innate—[all these]
must belong to the realm of phenomena; otherwise such [things] as the fact that [they]
can be desired, the search for the realization of this [desire], their determination [as hav-
ing] this [particular] form and place, the practice in accordance with [this determination],
etc., would [all] be impossible.”
via free access
form that is [presently] being manifest, and the goal [of human practice] only
concerns what is merely such [and nothing beyond manifestation].
[If you reply:] “But this [property of being an object] can only belong to
[things] that are distinct from manifestation,” what apprehension [of these
objects] could there be [if they are distinct from manifestation]?96 [And] what
is this [so-called] annihilation of ordinary human practice [that must inex-
orably occur according to you] if [objects] are one with phenomena? This is
what [the Vṛtti] says in “let [us admit that] they consist in phenomena.”97
Only [the following] could [still] be objected:98 if these [objects] did not
exist after as well as before [their] being manifest, [then] the very fact that they
96 Here the opponent is arguing that by definition, an object is what we apprehend as being
distinct from—i.e. external to—us considered as conscious subjects, so that the ideal-
ist’s thesis is absurd because it contradicts our most ordinary experiences in the sphere
of human practice. He is therefore implicitly invoking the evidence of common knowl-
edge (prasiddhi) to show that there must be some external object. Utpaladeva answers
that this evidence, far from being contradictory to his idealism, is in fact an argument in
favour of it. Cf. ĪPVV, vol. II, 132: tataś ca prasiddhyaiva bāhyaḥ sidhyatīti pratyuta viparī-
tam etat … “And therefore, [to the objection:] ‘the external [object] is established through
mere common knowledge,’ [one should answer] that it is rather the contrary.” According
to this fragment, the reason why this is so is that apprehending an object means mak-
ing the object manifest to consciousness, and such a manifestation can only occur if the
object is nothing but consciousness taking the form of an object. This idea is justified at
length at the beginning of chapter 1.5 (see Ratié 2011, 309–366).
97 Cf. the beginning of Vṛtti on ĪPK 1.5.6, 20: ābhāsamānair evārthair vyavahāraḥ, te cā-
bhāsātmakāḥ santu, kā kṣatiḥ. “Ordinary human practice occurs thanks to objects pre-
cisely insofar as [they are] being manifested, and let [us admit that] they consist in phe-
nomena [and nothing else]—what harm [might ensue for human practice if it is the
case]?” See Torella 2002, 114.
98 See ĪPVV, vol. II, 132–133: ābhāsānām eva vastutām abhidhāya prāmāṇikatvāt svayaṃ sva-
pakṣe dūṣaṇam āśaṅkya darśayaty āśayaśuddhipradarśanena vītarāgatāṃ* vaktuṃ keva-
lam iti. paryanuyojyam idaṃ paryanuyogārham, kim, āha ābhāsamānatāyā iti. pūrvam
ūrdhvaṃ ceti … [*āśayaśuddhipradarśanena vītarāgatāṃ conj.; āśayaśuddhipradarśane-
nāvitārakatāṃ ĪPVV, J10, J11, T.] “Having explained that only phenomena are real entities
because [only they are] established by a means of [valid] knowledge, [and] anticipating
by himself the refutation of his own thesis, [Utpaladeva now] expounds [this refutation
with the passage beginning with] ‘only …’ by empasizing the purity of his intentions, in
order to state that [he] is free of bias. [According to him] this ‘could [still] be objected,’
[i.e.] it deserves the [following] objection. Which one? This is what [Utpaladeva says]
in ‘[if these objects did not exist] after as well as before [their] being manifest …’” My
tentative emendation above (vītarāgatām instead of avitārakatām) is based on passages
such as Nyāyamañjarī, vol. I, 25: “and it [i.e. tarka] can be used in order to show [one’s]
purity of intentions in a debate” (sa cāśayaśuddhim upadarśayituṃ vāde prayokṣyate).
Cakradhara (Granthibhaṅga, vol. I, 17) explains that it can be used “in order to make
clear ‘[one’s] purity of intentions,’ [i.e.] the fact that one is free of bias” (āśayaśuddhiṃ
vītarāgatvaṃ prakaṭayitum), and he adds: tarkakrameṇa svārthānumānakāle yathā prati-
via free access
130 ratié
are manifest would be causeless, and [under such conditions,] the relation of
cause and effect and the relation between the knowing subject and the object
of knowledge would not be possible.99
[…]
[3] If, on the other hand,100 external objects are only atoms that are part-
less [and] aggregated, even so, a pot, which appears in a [spatially] extended
pannaḥ svayam asāv arthas tathaiva parasya pratipādyate, vītarāgakathātvād vādasyeti.

“Through the method of tarka, something is demonstrated for the opponent exactly as
it was understood by oneself at the time of an inference for oneself, because a debate is
a discourse of [people] who are free of bias.” This is an allusion to the Naiyāyikas’ dis-
tinction between three kinds of discourse, namely debate (vāda), disputation ( jalpa) and
sophistry (vitaṇḍā), and the Śaivas are well acquainted with it. See e.g. Kṣemarāja’s Sva-
cchandatantroddyota, vol. VI, 98 (alluding to Nyāyasūtra 1.2.1, according to which “debate
comprehends both the thesis and the antithesis,” … pakṣapratipakṣaparigraho vādaḥ): …
pakṣapratipakṣaparigraho vītarāgakathārūpo vādaḥ. “Debate, which comprehends both
the thesis and the antithesis, [and] which is a discourse of [people] who are free of bias …”
On vāda as a vītarāgakathā see also e.g. Nyāyamañjarī, vol. I, 18 or Nyāyabhūṣaṇa, 328; cf.
Tantrālokaviveka on Tantrāloka 13.126, vol. VIII, 90. So as far as I understand the ĪPVV pas-
sage, Abhinavagupta simply means here that by raising himself (svayam) this objection to
his own thesis, Utpaladeva is showing to his opponent that he is an unbiased interlocutor.
99 Cf. ĪPVV, vol. II, 133, which explains the objection in the following way: … ābhāsanam
avaśyaṃ bhavane cābhavane ca nimittam apekṣate. tac cen nāsti, tan nirhetukatā. tataś
ca kāryakāraṇābhāvād ghaṭārthī na mṛdam āharet, na kumbhakārakulam abhigacchet, na
dhūmārthī hutabhujam ādadīta. kāryakāraṇatāmūlaś ca jñāpyajñāpakabhāvaḥ … “A man-
ifestation necessarily requires a cause as regards both [its] arising and [its] not arising.
And if there is no such [cause], then [this manifestation] is causeless. And since as a con-
sequence there is no relation of cause and effect, [someone] who wants a pot should not
get clay [and] should not go see a family of potters; [and someone] who wants smoke
should not get himself a fire. Moreover, the relation between the knowing subject and the
object of knowledge has as its root the relation of cause and effect …” As for the gist of
Utpaladeva’s reply to this objection according to Abhinavagupta, see above, note 30.
100 This fragment, which targets the Buddhist Vaibhāṣika theory according to which external
objects are nothing but aggregated atoms, must have followed a criticism of the Vaiśeṣika
thesis that the external object is a whole (avayavin) distinct from its parts (avayava), hence
the atha here. See ĪPVV, vol. II, 140: evaṃ kāṇādasaṃmataṃ bāhyaṃ dūṣayitvā vaibhāṣika-
paribhāṣitam api dūṣayaty atheti. “Having thus refuted the external [object as it is] under-
stood by the followers of Kaṇāda, [Utpaladeva now] refutes as well [the external object as
it is] explained by the Vaibhāṣikas [in the sentence beginning with] ‘If, on the other hand.’”
This line of argument is of course borrowed from the famous criticism in Viṃśikā 11–15 of
the various theories claiming to account for the external object. Vasubandhu shows there
that we cannot make sense of this external object whether we try to understand it as a
whole distinct from its parts, as atoms taken individually, or as aggregated atoms. On the
recent debate as to whether Vasubandhu’s reasoning was meant to show that the external
object cannot exist, or whether his goal was merely to point out that such an object cannot
be known, and for convincing arguments in favour of the first hypothesis, see Kellner and
via free access
form, necessarily appears as having [different] parts [respectively located in

the] east, west, etc.; and [this spatial extendedness]101 is not possible if [this
pot] is thus made of atoms[, since by definition an atom cannot have different
parts].102 To explain:103 a second atom that is connected with the atom con-
sidered as the first [one]104 must be one with this [first atom]; for if [these
atoms] devoid of parts are in contact, how much [of them could] remain that
Taber 2014; on the Śaiva nondualists’ unambiguous interpretation of the Viṃśikā along the
same lines, see Ratié 2014. On Vasubandhu’s refutation—summed up in this fragment—of
the Vaibhāṣikas’ atomism, see e.g. Kapstein 2001, 181–204; for Abhinavagupta’s explanation
of it, see Ratié 2010, 450–452, and Ratié 2011, 395–399.
101 Cf. ĪPVV, vol. II, 140: na caivaṃ paramāṇumayatva iti vaitatyaṃ digbhāgabhedavattvam.
“And [spatial] extendedness, which [consists in] being differentiated into parts according
to the directions, is ‘[not possible] if [this pot] is thus made of atoms.’”
102 The main idea is that if perceptible things, which are made of imperceptible atoms, are
spatially extended, the atoms too must have such an extendedness, but spatially extended
atoms must have parts, which is absurd since the atom is by definition partless. Cf. Viṃ-
śikā 14ab: digbhāgabhedo yasyāsti tasyaikatvaṃ na yujyate | “That which is differentiated
into parts according to the directions cannot be one.” See also Viṃśikāvṛtti, 7: anyo hi para-
māṇoḥ pūrvadigbhāgo yāvad adhodigbhāga iti digbhāgabhede sati kathaṃ tadātmakasya
paramāṇor ekatvaṃ yokṣyate. “For if an atom has one part in the direction of the east,
[and others in the directions of the south, west, north, above] and below, given that [the
atom] is differentiated into [various] parts according to the directions (digbhāgabheda),
how could the atom be one [whereas it] consists of these [different parts]?”
103 According to Abhinavagupta, here Utpaladeva has the externalist face a dilemma, the first
part of which remains implicit in the Vivṛti. See ĪPVV, vol. II, 140: tac ca bahūnāṃ para-
māṇūnāṃ bhinnadeśatve mūrtatvenānyonyarūpadeśākramaṇayogyatvābhāvāt* pratilab-
dhe kimiti na ghaṭata ity āśaṅkyāha tathā hīti. ayaṃ bhāvaḥ—yadi sāntarāḥ paramāṇavo
ghaṭaḥ, tad adṛśyatā … [*mūrtatvena° corr.; ’mūrtatvena° ĪPVV.] “[The opponent:] ‘And
why is [this spatial extendedness of the pot] not possible if [we admit that] the numerous
atoms get to have different places because, since they are of a material, [i.e. solid] nature
(mūrta), [they] cannot extend to the place of the others’ forms?’ Anticipating this [objec-
tion, Utpaladeva] says ‘To explain …’ Here is the implicit meaning [of this passage]: if the
pot is [nothing but] atoms with intervals [separating them from each other], then [the
pot] must be imperceptible …” In other words, the opponent is tempted to respond to the
objection just stated in the fragment by explaining that atoms can indeed occupy different
locations because although partless, they cannot penetrate each other so as to coexist in
the same spot. But this reply is unsound, first and foremost because then the macroscopic
pot should remain imperceptible, as it would merely consist of imperceptible atoms and
intervals between them. After enumerating a few more of the “countless defects” (śataśo
doṣāḥ) inherent in the thesis that atoms are separated by intervals, Abhinavagupta adds
(ibid.): nairantarye tu parasparasaṃśleṣaḥ. “But if [the atoms] are contiguous, they [must]
be in contact with one another.” It is the absurdity of the latter hypothesis that the frag-
ment is now going to point out.
104 Alternatively, one could understand: “a second atom that is connected with the atom con-
sidered as [located] in the east.”
via free access
132 ratié
might not be in contact? And [if they are thus entirely] in contact, their natures
must be immersed in each other, therefore [they] can only be manifest as one
[single] atom; and if [they are] in contact with yet another atom, the same
[consequence follows]—therefore even if an infinite number of atoms were
connected, they should be manifest as having the size of one [single] atom;105
or [rather], even this [manifestation] would not exist, because atom[s], [taken]
one by one, are beyond the realm of the sense organs.106
7 Translation of the Fragments on ĪPK 1.5.8–9
[4] For inference is a concept, and this [concept] arises thanks to the resid-
ual trace (saṃskāra) [left by] a previous experience; so to begin with, [it]
depends on the fact that the object was directly perceived [at some point] in the
past, and inference is a conceptual cognition that arises as an unfailing (avya-
bhicāra)107 [means of knowledge] with respect to this [previously perceived]
object.108 And insofar as this [inference]109 produces the realization (vimṛśati)
105 If atoms are in contact, they must all share the same place since they are partless, and as a
result their aggregate, however complex, cannot be larger than a single atom. See Viṃśikā
12cd: ṣaṇṇāṃ samānadeśatvāt piṇḍaḥ syād aṇumātrakaḥ | “If the six [atoms supposedly
surrounding the first one] share the same location [as the first one], [they] must [con-
stitute] a lump [of matter] that has the size of a [single] atom.” Cf. Viṃśikāvṛtti, 7: atha
ya evaikasya paramāṇor deśaḥ sa eva ṣaṇṇām. tena sarveṣāṃ samānadeśatvāt sarvaḥ piṇ-
ḍaḥ paramāṇumātraḥ syāt parasparāvyatirekāt. “If, on the other hand, the place of one
atom is also that of the six [atoms supposedly surrounding it], then since they all have
the same place, [they] must all [constitute] a lump [of matter] that has the size of a [sin-
gle] atom, since they are not distinct from each other.” See also ĪPVV, vol. II, 141: iti dvāv
api militau nādhikaṃ rūpaṃ samutthāpayeyātām. evam anyamelane ’pi vācyam. tad āha
ananteti. “Therefore two connected [atoms] cannot bring about a form larger [than one
single atom], and the same must be said if another, [third atom] gets in contact [with
them]—this is what [Utpaladeva explains] with ‘[even if] an infinite [number] …’”
106 If aggregated atoms are not larger than one atom, no material object should be percepti-
ble, since a single atom is imperceptible. Cf. Viṃśikāvṛtti, 7: iti na kaścit piṇḍo dṛśyaḥ syāt.
naiva hi paramāṇavaḥ saṃyujyante niravayavatvāt. “So no lump [of matter] at all should
be perceptible; for the atoms cannot be in contact [with each other] at all, since they have
no parts.”
107 Literally, “non-deviating.”
108 According to Abhinavagupta, here Utpaladeva specifies that inference is a “non-deviating”
means of knowledge (that is, according to the Buddhist epistemologists, a means of knowl-
edge that is valid inasmuch as it unfailingly enables us to reach a given object) so as to
point out that although conceptual, it is not a mere mental construct absolutely unre-
lated to perception, since it regards a previously perceived object. See ĪPVV, vol. I, 161:
pūrvābhāta evānumānam ity etat sādhayituṃ tāvad āha anumānaṃ hīti. yata evaṃ tas-
via free access
of this object—[which only] exists inside [consciousness at the time when

we infer]—thanks to the residual trace of the [past] experience, [and inso-
far as it produces this realization] as is appropriate [for an object, i.e.] in the
form “this” (idantā), it only manifests [this object] as being separated [from
the subject, the latter being expressed as “I”]. And this mere [realization that
the object is something separated from the subject] is not enough to transform
this object into something on which [human] activity may be exerted; there-
fore [this object] is [also] made manifest (prakāśita) as having a specific place
and time, because only a particular having a specific place and time can be
something on which [human] activity may be exerted, since [only such a par-
ticular] can be obtained (prāpya) and since [only such a particular] may have
the efficacy (arthakriyā) that [we] expect [from it]. So [a valid inference is] an
act of realization (vimarśana) that is exactly so, [i.e. it is an unfailing means
of knowledge with respect to a previously perceived object that it manifests
as having a specific place and time].110 This is what [the Vṛtti] says [with the
words] “manifested in the past” (pūrvāvabhāta), “an object invariably concomi-
tant [with this entity]” (nāntarīyakārtha) [and] “[due to the association of this
entity with] this or that [specific] place and time, etc.”111
mād yuktaḥ sūtrārthaḥ. nanu vikalpamātram eva kim anumānam. netyāha avyabhicāreti*.
[*avyabhicāreti conj.; avyabhicārīti J10, J11, T, ĪPVV.] “In order to demonstrate that infer-
ence only regards a previously manifested [object], first [Utpaladeva] states [the sentence
beginning with] ‘For inference …’ [And] since [inference] is so, the meaning of verse [1.5.8]
is justified. But is inference nothing but a mere concept? With [the word] avyabhicāra°,
[Utpaladeva] answers ‘no’ [to this question].”
109 Cf. ĪPVV, vol. II, 161: anumānaṃ kartṛ prathayatīti sambandhaḥ.
110 See ĪPVV, vol. II, 162: na ca tāvatety antaḥsthitasaṃskāraśeṣasya bahīrūpatāvabhāsanalak-
ṣaṇena vicchedamātreṇābhāsitenāpi viśiṣṭau deśakālau vinā kāyīyavyāpāraparyantapra-
vṛttijananam anumānena na kṛtaṃ bhaved iti viśiṣṭadeśakālāliṅgite ’rthe pravṛttiyogye yad
vikalpanaṃ tat pramāṇarūpam avisaṃvādakatayānumānam ucyate. “‘And this … is not
enough’ [means the following]. [Something] that [only] consists of the manifestation in
an external form, [i.e. in the form ‘this,’] of what remains [of the past experience in the
shape of] an internal residual trace; [that is to say, something] that is merely separated
[from the subject expressed as ‘I,’ but] that, although manifested, [appears] without any
specific place and time—[this is] not [enough] for inference to trigger any activity ending
in a bodily action. Therefore what [we] call an inference is [not just any] conceptualiza-
tion (vikalpana), [but only one] that consists in a means of [valid] knowledge since it is
reliable (avisaṃvādaka), [and one] that concerns an object possessed of a specific time
and place, [therefore being] fit to be [something on which] an activity [is exerted].”
111 Cf. Vṛtti on ĪPK 1.5.8, 21–22: pūrvāvabhātāntaḥsthita evārthe nāntarīyakārthadarśanavaśāt
tattaddeśakālādiyojanayā vimarśanam anumānam. “Inference is an act of realization (vi-
marśana) with respect to an object that exists internally [now in the form of a residual
trace because] it has [already] been manifested in the past. [This act occurs] thanks to the
perception of an object invariably concomitant [with this entity,] due to the association
via free access
134 ratié
[…].112
[5] Even a sense organ has already been experienced before [being inferred];
for this [sense organ] is not inferred in its own specific form, but rather, as
a [mere] generality (sāmānya).113 This is what [I] say in the Vṛtti [with the
words] “a cause that is a mere indeterminate thing” (kiñcinmātra).114 The object
[inferred in this inference of the sense organs] is a generality (sāmānya) [that
simply consists in] being a cause; [and this generality] was manifested before
[the inference] in the cognition that the seed is a cause of the sprout, [or]
in the cognition that clay for instance is a cause of the pot and [other such
objects].115 For in that [cognition that the seed is the cause of the sprout],116 it
is not [something] specific[, i.e.] only a seed [and nothing else], that is mani-
[of this entity] with this or that [specific] time and place, etc.” As noted in Torella 2002,
117, note 19, Utpaladeva’s definition specifies that inference “aims at proving the existence
of an object that is able to propose itself as the possible object of purposeful activity (i.e.
specified by a definite time and space […]).” And this amounts to saying that the external
object cannot be an inferential object, not only because there can be no previous experi-
ence of such an object, but also because according to Utpaladeva’s definitions of time and
space (on which see e.g. Ratié 2011, 197–201), spatial and temporal relationships can only
belong to manifested entities. See e.g. ĪPV, vol. II, 16: nanv evam ābhāsaviṣayābhyām eva
deśakālakramābhyāṃ bhavitavyam. “But [if it is] so, the spatial and temporal sequences
can only regard phenomena.”
113 On the Śaivas’ understanding of this type of inference called sāmānyatodṛṣṭa (which infers
the mere existence of some imperceptible and therefore indeterminate entity so as to
account for a phenomenon that would remain inexplicable otherwise) see e.g. Torella
2002, 117, note 20, and Ratié 2011b, 486–488.
114 Cf. Vṛtti, 22: indriyam apy anumīyate kiñcinmātraṃ nimittaṃ tac ca bījādyābhāsād ābhā-
sitam eva. “Even a sense organ is inferred [as] a cause that is a mere indeterminate thing;
and this [sense organ] has already been manifested due to the manifestation of a seed,
etc.” See Torella 2002, 117.
115 Utpaladeva considers every perceived object as a particular synthesis of general features
(including that of causality); see Torella 1992, 332–333, and Torella 2002, 89–90, note 3.
On Abhinavagupta’s explanations of this point see Ratié 2011b, 488–491. It is this theory
that enables Utpaladeva to claim here that although the sense organs are imperceptible
in the sense that they can never be apprehended as singular entities existing in a deter-
minate place and time, they can still be considered as having somehow been perceived
before being inferred (so that their inference is valid); this is so because the inference of
the sense organs does not concern a particular entity but a mere generality (that is, an
indeterminate cause that must be postulated on top of other conditions such as light so
as to account for the phenomenon of perception), and this general notion of causality is
part of countless perceptions of particular fires seen as causing smoke, etc., since the per-
ception of a particular is nothing but the apprehension of a synthetic unity of generalities,
including causality.
116 Cf. ĪPVV, vol. II, 163: na hi tatreti bījasya nimittatājñāne.
via free access
fested as a cause; [otherwise,] as a consequence [we] would not know that clay
too [is a cause]. Nor is clay only [cognized as a cause] in a [cognition occurring]
after [that of the seed being a cause], because [if it were so] there would follow
that [we] would not know that the seed and so on [are also causes]. And [one
should rather consider that] within this [notion of causality that is present in
the perception of the clay] as well as [in that of the seed], a [process of] partic-
ularization leads to an exclusion [that takes the form]: “it is the clay that is the
cause of the pot, not the seed”; [and this exclusion eliminates from the general
notion of cause] that which [the particular cause] is not but which is intrinsi-
cally linked (prasakta) [with the notion of cause in general].117
[…]
[6]?118 [—Objection from the Sautrāntika:] But just as [you] have said that
[in the case of the inference of the sense organs,] the generality “causality”
has already been experienced through the experience of [particular causes]
117 Utpaladeva is arguing here that within any act of perception, the apprehension of uni-
versals or generalities (such as causality) comes first, and it is only once these general
features are apprehended that we determine our perception as being that of a singu-
lar synthesis. This determination of the perceptual object as a singular entity takes the
form of the mental process of exclusion (vyavaccheda, apoha) which, according to Dhar-
makīrti, produces generalities (Utpaladeva’s system thus integrates the Dharmakīrtian
notion of apoha, but not without turning it upside down). In other words, upon see-
ing a pot being made out of clay, we first apprehend a number of generalities inherent
in the clay, and then we exclude from e.g. the generality “causality” whatever falls into
the general category of cause (or, as Utpaladeva says, is intrinsically linked with it) but
has a different efficacy—seeds for instance. According to Abhinavagupta, it is because
we thus apprehend the generality “causality” before the particularities of the perceived
cause that we are capable of using the word “cause” to describe various entities. See ĪPVV,
vol. II, 156: anyathaikatra bīje kāraṇaśabdaḥ saṅketito na mṛtpiṇḍādau saṅketito bhavet.
“Otherwise, the word ‘cause’ would be conventionally associated with one single [entity,
such as] the seed, [but] it could not be conventionally associated with a lump of clay,
etc.” It is also this theory that enables Utpaladeva to claim here that in a sense, all causes
(including imperceptible ones such as the sense organs) are perceived when we perceive
clay, although contrary to the singular clay perceived at that particular time and place,
all the other causes are only apprehended in the form of the general feature “causality.”
Cf. Abhinavagupta’s explanation in ĪPVV, vol. II, 163: nanu kumbhaṃ prati mṛdo nimit-
tatājñāne bījam api nimittabhāvena svīkṛtam iti kutaḥ. atrocyate prasakteti. prasaṅgaś ca
kāraṇatāsāmānyapratibhāsasamutthāpita iti bhāvaḥ. “To the [question:] ‘But how is it that
in the cognition that the clay is a cause with respect to the pot, [we] apprehend the seed
too as a cause?,’ [Utpaladeva] replies [with the expression] ‘that which is intrinsically
linked (prasakta) [with the notion of cause in general].’ And this intrinsic link arises from
the manifestation of the generality [consisting in] causality—this is what [Utpaladeva]
means.”
118 On my reasons for thinking that this passage might be a fragment from the Vivṛti, see
above, note 79.
via free access
136 ratié
such as a seed, without considering the particulars that are the visual organ
and so on, in the same way, [you must admit that] externality too has already
been experienced as a generality from the experience of [various objects that
are] external to the body, the house or the village, etc. Alternatively, if [you]
claim that [the external object cannot be inferred because] there is no expe-
rience of externality through a particular [entity characterized as] not being
consciousness, [then] since the visual organ and so on have not been previ-
ously experienced [as] a particular cause either, there can be no inference [of
the sense organs either]; so why don’t [you simply] admit that inference applies
[in both cases]?119
119 I now think that my understanding of the last part of this sentence was mistaken in
Ratié 2011, 466, note 221, and Ratié 2011b, 494, note 48. As for the general meaning of
the passage, see ĪPVV, vol. II, 165: grāmād darpaṇād gṛhād dehāt saṃvedanād bāhyam iti
bāhyatāsāmānyam* ekam, tac ca gṛhād bāhyam iti pratītau siddham. tataś ca samarpakaṃ
darpaṇād iva bāhyaṃ* saṃvedanād api setsyati kramikābhāsavaicitryād dhetor ity āśayena
pūrvapakṣayati nanu cakṣurādīti. [*bāhyatāsāmānyam conj. Ratié 2011b, found in J10, J11,
T, marginal annotations in SOAS, D2, and note 248 in ĪPV, vol. I, 190; bāhyataḥ sāmānyam
ĪPVV. *darpaṇād iva bāhyaṃ conj. Ratié 2011b, found in T; darpaṇādibāhyaṃ J10, J11, ĪPVV,
marginal annotation in SOAS, D2 and note 248 ĪPV (vol. I, 190).] “With [the sentence
beginning with] nanu cakṣurādi-, [Utpaladeva] states the prima facie thesis [that he is
about to refute] with the intention [of conveying the following objection]: It is one [and
the same] generality of externality [that is present in these various cognitions: ‘this is]
external to the village, to the mirror, to the house, to the body, to consciousness.’ And
this [generality of externality] is established in the [mere] experience: ‘[this is] exter-
nal to the house.’ And as a consequence, [something] may be established to be external
to consciousness as well and to project [its specific form onto consciousness], just as
[something is external] to a mirror [and projects its specific form onto the mirror, and
it may be established to be so] on the basis of the [logical] reason that is the variety
of successive phenomena [experienced by any conscious being].” As to how, according
to Abhinavagupta, Utpaladeva was answering this objection in the now missing Vivṛti
passage following this fragment, see ĪPVV, vol. II, 165: naivaṃ* bāhyatvaṃ ghaṭasya grā-
māc ca saṃvedanāc caikam, saṃvedanād bāhyaṃ hy asaṃvedanarūpam, na tu gṛhād
bāhyam agṛharūpam. evaṃ sati hi gṛhaikadeśaḥ kuḍyādir gṛhāntarvarty api ca ghaṭādir
gṛhabāhyaḥ syāt. na caivam, gṛhasannikr̥ ṣṭaṃ ca yadvad* bāhyaṃ gṛhān na tadvad eva
saṃvedanāt tasyāmūrtasya sannikarṣādideśavyavahāryatvābhāvāt. tataḥ śabdasāmyamā-
treṇedaṃ* sādhyam ekaṃ pratibhātīty abhiprāyeṇottarayaty atrocyata iti. [*naivaṃ T;
nedaṃ J10, J11, ĪPVV. *ca yadvad T, ĪPVV; yadvad J10, J11, Kha (ĪPVV, note 1). *śabdasāmya-
mātreṇedaṃ conj. Ratié 2011b, found in T; śabdasāmānyamātreṇedam J10, J11, ĪPVV.] “In
‘To this [objection we] reply …,’ [Utpaladeva] answers [the objection] with [the follow-
ing] in mind. The externality of the pot is not thus one [and the same whether it is
considered] with respect to the village or with respect to consciousness; for that which
is external to consciousness consists in that which is not consciousness, whereas that
which is external to the house does not consist in that which is not a house! For if that
were the case, a particular element of the house—such as a wall—or a pot, for example,
via free access
Acknowledgements
Although my doctoral research was undertaken and completed at the École

Pratique des Hautes Études (Paris) under Lyne Bansat-Boudon’s supervision, I
studied in Oxford under Alexis Sanderson’s guidance in 2005 and 2006 while
working on my PhD dissertation, and I am immensely indebted to him. He
was kind enough not only to let me attend his classes there but also to orga-
nize many extra reading sessions so we could cover more of Abhinavagupta’s
Īśvarapratyabhijñāvimarśinī, and we ended up reading about two thirds of this
lengthy and difficult text in less than two years. It was an extraordinary experi-
ence for me, not only owing to Alexis’s remarkable erudition and exceptional
philological acumen, but also because his generosity, humour and curiosity
have always created among his students a unique atmosphere of intellectual
freedom and enthusiastic research. I am proud to consider myself a pupil of
his, and it is a great honour and joy for me to be given this opportunity, by
contributing to the present volume, to express my profound admiration and
affection for this wonderful scholar and teacher.
Most of the research presented in this paper was carried out thanks to the
generous financial help of the Deutsche Forschungsgemeinschaft, as part of
project FR 2531/3–1. The National Mission for Manuscripts, the SOAS Library
and Mr. Chetan Pandey were kind enough to provide photographs of the manu-
scripts used for the edition of the fragments given below. Many thanks are also
due to Vincent Eltschinger and Eli Franco, who read a draft of this paper and
provided insightful remarks.
References
Primary Sources: Manuscripts and Sigla

D. Īśvarapratyabhijñā[-Vṛtti,] Vivṛti [and Vimarśinī]. Delhi: National Archives of India
(Manuscripts belonging to the Archaeology and Research Department, Jammu and
Kashmir Government, Srinagar), no. 30, vol. IX. Paper, Śāradā script.
although it is located inside the house, should be external to the house [since they are
not the house itself]; and it is not so. And whereas that which is external to the house is
next to the house, it is absolutely not the case as regards that which is [external] to con-
sciousness, because of the impossibility for [consciousness]—which is devoid of material
form (amūrta)—of having any spatial relation whatsoever such as proximity. Therefore
this [externality] that must be established appears to be one thanks to a mere similarity
(śabdasāmya).”
via free access
138 ratié
D2. Īśvarapratyabhijñāvimarśinī. Delhi: National Archives of India (Manuscripts

belonging to the Archeology and Research Department, Jammu and Kashmir Gov-
ernment, Srinagar), no. 5, vol. II. Paper, Śāradā script.
D3. Śivadṛṣṭi. Delhi: National Archives of India (Manuscripts belonging to the Archeol-
ogy and Research Department, Jammu and Kashmir Government, Srinagar), no. 7,
vol. II. Paper, Śāradā script.
D4. Śivadṛṣṭi. Delhi: National Archives of India (Manuscripts belonging to the Archeol-
ogy and Research Department, Jammu and Kashmir Government, Srinagar), no. 83,
vol. LXVI. Paper, Devanāgarī script.
J10. Īśvarapratyabhijñāvivṛtivimarśinī. Jammu: Sri Ranbir Institute, Raghunath Mandir
Library, no. 1475. Paper, Devanāgarī script.
J11. Īśvarapratyabhijñāvivṛtivimarśinī. Jammu: Sri Ranbir Institute, Raghunath Mandir
Library, no. 5077. Paper, Devanāgarī script.
JR. Īśvarapratyabhijñāvimarśinī. Jammu: Rashtriya Sanskrit Sansthan, no. 47 (formerly
no. 52A) and no. 70 (formerly no. 52B). Paper, Śāradā script, incomplete.
S2. Īśvarapratyabhijñāvimarśinī. Śrinagar: Oriental Research Library, no. 1035. Paper,
Śāradā script.
S3. Īśvarapratyabhijñāvimarśinī [catalogued as Īśvarapratyabhijñākaumudī]. Śrinagar:
Oriental Research Library, no. 838. Paper, Śāradā script.
S5. Īśvarapratyabhijñāvimarśinī [catalogued as Sāṃkhyatattvakaumudī by Vācaspati
Miśra; Abhinavagupta’s ĪPV is copied after the latter]. Śrinagar: Oriental Research
Library, no. 1212. Paper, Śāradā script.
Śāradā script.
Śāradā script.
S12. Īśvarapratyabhijñāvivṛtivimarśinī [catalogued as Īśvarapratyabhijñā(vimarśinī)].
Srinagar: Oriental Research Library, no. 2403. Paper, Śāradā script.
S15. Īśvarapratyabhijñāvimarśinī. Srinagar: Oriental Research Library, no. 787. Paper,
Śāradā script.
S19. Īśvarapratyabhijñāvṛtti [catalogued as Īśvarapratyabhijñākārikā]. Srinagar: Orien-
tal Research Library, no. 824. Paper, Śāradā script.
SOAS. Īśvarapratyabhijñāvimarśinī [catalogued as Pratyabhijñāsūtra with Abhinava-
gupta’s Sūtrārthavimarśinī]. London: School of Oriental and African Studies
Library, no. 44255/no. 207 in Dogra 1978. Paper, Śāradā script.
T. Īśvarapratyabhijñāvivṛtivimarśinī. Trivandrum: Oriental Research Institute and
Manuscripts Library, no. 50413 B and C. Palm-leaf, Grantha Malayalam script.
via free access
Primary Sources: Edited Texts

Abhidharmakośabhāṣya. Yasunori Ejima, ed. Abhidharmakośabhāṣya of Vasubandhu.
Chapter I: Dhātunirdeśa. Bibliotheca Indologica et Buddhologica, no. 1. Tokyo: The
Sankibo Press, 1989.
Āgamaḍambara. Csaba Dezső, ed. Unpublished critical edition and online supple-
mental materials available at http://www.claysanskritlibrary.org/extras.php#11. Clay
Sanskrit Library, 2004.
Granthibhaṅga. Gaurinath Sastri, ed. Nyāyamañjarī of Jayanta Bhaṭṭa, with the Com-
mentary of Granthibhaṅga by Cakradhara. 3 vols. Varanasi: Sampurnanand Sanskrit
Vishvavidyalaya, 1982.
Īśvarapratyabhijñākārikā (ĪPK). See Torella 2002.
Īśvarapratyabhijñāvimarśinī (ĪPV). Mukund Rām Shāstrī and Madhusudan Kaul Shās-
trī, eds. Īshvara-pratyabhijñā Vimarshinī of Utpaladeva [sic] With Commentary by
Abhinava-Gupta. 2 vols. Kashmir Series of Texts and Studies, nos. 22 and 33. Bom-
bay: Niraya Sagar Press, 1918–1921.
Īśvarapratyabhijñāvivṛtivimarśinī (ĪPVV). Madhusudan Kaul Shāstrī, ed. Īśvarapratya-
bhijñā Vivritivimarśinī of Abhinavagupta. 3 vols. Kashmir Series of Texts and Studies,
nos. 60, 62 and 65. Bombay: Nirnaya Sagar Press, 1938–1943.
Nyāyabhūṣaṇa. Svāmī Yogīndrānanda, ed. Śrīmadācāryabhāsarvajñapraṇītasya nyāya-
sārasya svopajñaṃ vyākhyānaṃ nyāyabhūṣaṇam. Varanasi: Ṣaḍdarśana Prakāśana
Pratiṣṭhānam, 1968.
Nyāyamañjarī. K.S. Varadacharya, ed. Nyāyamañjarī of Jayanta Bhaṭṭa, with Ṭippaṇī—
Nyāyasaurabha by the Editor. 2 vols. Oriental Research Institute Series, nos. 116 and
139. Mysore: Oriental Research Institute, 1969–1983.
Nyāyasūtra. Anantalal Thakur, ed. Gautamīyanyāyadarśana with Bhāṣya of Vātsyāyana.
Nyāyacaturgranthikā, no. 1. New Delhi: Indian Council of Philosophical Research,
1997.
Parātrīśikāvivaraṇa. Mukunda Rāma Shāstrī, ed. Parātrimshikā with Commentary, the
Latter by Abhinavagupta. Kashmir Series of Texts and Studies, no. 18. Srinagar: Tatva
Vivechaka Press, 1918.
Pramāṇaviniścaya. Ernst Steinkellner, ed. Dharmakīrti’s Pramāṇaviniścaya, Chapters 1
and 2. Beijing-Vienna: China Tibetology Publishing House/Austrian Academy of Sci-
ences Press, 2007.
Mālinīślokavārttika. Madhusudan Kaul Shastri, ed. Śrī Mālinīvijayavārttikam of Abhi-
navagupta. Kashmir Series of Texts and Studies, no. 32. Srinagar: Kashmir Pratap
Steam Press, 1921.
Śivadṛṣṭi. Madhusudan Kaul Shastri, ed. Śivadṛṣṭi of Śrīsomānandanātha with the Vṛtti
by Utpaladeva. Kashmir Series of Texts and Studies, no. 54. Srinagar: 1934.
Śivadṛṣṭivṛtti. See Śivadṛṣṭi.
Spandapradīpikā. Mark S.G. Dyczkowski, ed. The Spandapradīpikā: A Commentary on
the Spandakārikā by Bhagavadutpalācārya. Varanasi: D.K. Book Agencies, 1990.
via free access
140 ratié
Svacchandatantroddyota. Madhusudan Kaul Shāstrī, ed. The Svacchanda Tantram with

Commentary by Kṣemarāja. 6 vols. Kashmir Series of Texts and Studies, nos. 31, 38,
44, 48, 53 and 56. Bombay: Nirnaya Sagar Press, 1921–1935.
Tantrāloka. Madhusudan Kaul Shāstrī, ed. Tantrāloka of Abhinavagupta with commen-
tary by Rājānaka Jayaratha. 12 vols. Kashmir Series of Texts and Studies, nos. 23, 28,
29, 30, 35, 36, 41, 47, 52, 57, 58 and 59. Allahabad: Indian Press, 1918–1938.
Tantrālokaviveka. See Tantrāloka.
Viṃśikā. Silvain Lévi, ed. Vijñaptimātratāsiddhi. Deux traités de Vasubandhu. Viṃśatikā
(la vingtaine) accompagnée d’une explication en prose et Triṃśikā (la trentaine) avec
le commentaire de Sthiramati. Bibliothèque de l’École des Hautes Études 245. Paris:
Honoré Champion, 1925.
Viṃśikāvṛtti. See Viṃśikā.
Vṛtti (= Īśvarapratyabhijñāvṛtti). See Torella 2002.
Secondary Sources
Bhattacharya, Ramkrishna. 2002. “Cārvāka Fragments: a New Collection.” Journal of
Indian Philosophy 30 (6): 597–640.
Bronkhorst, Johannes. 2008. “Udbhaṭa, grammarian and Cārvāka.” In Linguistic Tradi-
tions of Kashmir. Essays in Memory of Paṇḍit Dinanath Yaksha, edited by Mrinal Kaul
and Ashok Aklujkar, 281–299. Delhi: DK Printworld.
Bronner, Yigal. 2016. “Understanding Udbhaṭa: the Invention of Kashmiri Poetics in the
Jayāpīḍa Moment.” In Around Abhinavagupta. Aspects of the Intellectual History of
Kashmir from the Ninth to the Eleventh Century, edited by Eli Franco and Isabelle
Ratié, 81–148. Berlin: Lit Verlag.
Dezső, Csaba. 2005. Much Ado about Religion by Bhaṭṭa Jayanta. New York: New York
University Press/JJC Foundation.
Dogra, R.C. 1978. A Handlist of the Manuscripts in South Asian Languages in the [SOAS]
Library. London: School of Oriental and African Studies.
Formigatti, Camillo. 2011. Sanskrit Annotated Manuscripts from Northern India and
Nepal. PhD Dissertation. Hamburg: University of Hamburg.
Franco, Eli. 2011. “Lokāyata.” In Brill’s Encyclopedia of Hinduism, vol. III, edited by Knut
A. Jacobsen, Helene Basu, Angelika Malinar and Vasudha Narayanan, 629–642. Lei-
den: Brill.
Iyer, K.A. Subramania. 1969. Bhartṛhari. A Study of the Vākyapadīya in the Light of the
Ancient Commentaries. Poona: Deccan College.
Kano, Kazuo. 2008. “Two short glosses on Yogācāra texts by Vairocanarakṣita: Viṃ-
śikāṭīkāvivṛti and *Dharmadharmatāvibhāgavivṛti.” In Manuscripta Buddhica 1. San-
skrit Texts from Giuseppe Tucci’s Collection, Part I, edited by Francesco Sferra, 343–
380. Roma: Istituto Italiano per l’Africa e l’Oriente.
Kapstein, Matthew T. 2001. Reason’s traces. Identity and Interpretation in Indian and
Tibetan Buddhist Thought. Boston: Wisdom Publications.
via free access
Kawajiri, Yohei. 2016. “New Fragments of the Īśvarapratyabhijñā-Ṭīkā.” In Utpaladeva,

Philosopher of Recognition, edited by Raffaele Torella and Bettina Bäumer, 77–101.
Delhi: DK Printworld.
Kawajiri, Yohei. 2016b. “New Fragments of the Īśvarapratyabhijñā-vivṛti (3).” Nagoya
Studies in Indian Culture and Buddhism (Saṃbhāṣā) 33: 17–46.
Kellner, Birgit and John Taber. 2014. “Studies in Yogācāra-Vijñānavāda idealism I: The
interpretation of Vasubandhu’s Viṃśikā.”Études Asiatiques/Asiatische Studien 68 (3):
709–756.
La Vallée Poussin, Louis de. 1923. L’Abhidharmakośa de Vasubandhu, traduit et annoté,
Premier et deuxième Chapitres. Paris: Paul Geuthner.
McCrea, Lawrence and Parimal Patil. 2010. Buddhist Philosophy of Language in India—
Jñānaśrīmitra on Exclusion. New York: Columbia University Press.
McCrea, Lawrence. 2016. “Abhinavagupta as Intellectual Historian of Buddhism.” In
Around Abhinavagupta. Aspects of the Intellectual History of Kashmir from the 9th to
the 11th Century, edited by Eli Franco and Isabelle Ratié, 263–286. Berlin: Lit Verlag.
Namai, Mamoru. 1976. “Kōki Bukkyōto ni yoru Bārhaspatya Hihan (I).” Indological
Review 2: 33–74.
Nemec, John. 2011. The Ubiquitous Śiva. Somānanda’s Śivadṛṣṭi and his Tantric Interlocu-
tors. Oxford; New York: Oxford University Press.
Preisendanz, Karin. 1989. “On ātmendriyamanorthasannikarṣa and the Nyāya-Vaiśeṣika
Theory of Vision.” Berliner Indologische Studien 4–5: 141–213.
Ratié, Isabelle. 2010. “The Dreamer and the Yogin—on the Relationship between Bud-
dhist and Śaiva Idealisms.” Bulletin of the School of Oriental and African Studies 73
(3): 437–478.
Ratié, Isabelle. 2010b. “‘A Five-trunked, Four-tusked Elephant is Running in the Sky’—
How Free is Imagination According to Utpaladeva and Abhinavagupta?” Asiatische
Studien/Études Asiatiques 64 (2): 341–385.
Ratié, Isabelle. 2011. Le Soi et l’Autre. Identité, différence et altérité dans la philosophie de
la Pratyabhijñā. Leiden; Boston: Brill.
Ratié, Isabelle. 2011b. “Can One Prove that Something Exists Beyond Consciousness? A
Śaiva Criticism of the Sautrāntika Inference of External Objects.” Journal of Indian
Philosophy 39 (4–5): 479–501.
Ratié, Isabelle. 2014. “On the Distinction between Buddhist Epistemic and Metaphysi-
cal Idealisms: a Śaiva Perspective.” Journal of Buddhist Philosophy 42: 353–375.
Ratié, Isabelle. 2016. “Some Hitherto Unknown Fragments of Utpaladeva’s Vivṛti (I):
on the Buddhist Controversy over the Existence of Other Conscious Streams.” In
Utpaladeva, Philosopher of Recognition, edited by Raffaele Torella and Bettina
Bäumer, 224–256. Delhi: DK Printworld.
Ratié, Isabelle. 2016b. “Some Hitherto Unknown Fragments of Utpaladeva’s Vivṛti (III):
on Memory and Error.” In Around Abhinavagupta. Aspects of the Intellectual History
via free access
142 ratié
of Kashmir from the 9th to the 11th Century, edited by Eli Franco and Isabelle Ratié,
375–400. Berlin: Lit Verlag.
Ratié, Isabelle. 2017. “In Search of Utpaladeva’s Lost Vivṛti on the Pratyabhijñā Treatise:
a Report on the Latest Discoveries (with the Vivṛti on the end of Chapter 1.8).” Jour-
nal of Indian Philosophy 45: 163–189.
Ratié, Isabelle. 2018. “On the Ṣaḍdhātusamīkṣā, a Lost Work Attributed to Bhartṛhari:
an Examination of Testimonies and a List of Fragments.” Journal of the American
Oriental Society 138 (4): 709–742.
Ratié, Isabelle. 2018b. “For an Indian Philology of Margins: The Case of Kashmirian
Sanskrit Manuscripts.” In L’espace du Sens: Approches de la philologie indienne. The
Space of Meaning: Approaches to Indian Philology, edited by Silvia d’Intino and Shel-
don Pollock, with the collaboration of Michaël Meyer, 305–354. Paris: Collège de
France/Diffusion De Boccard.
Ratié, Isabelle. Forthcoming. Utpaladeva on the Power of Action: a First Edition, Anno-
tated Translation and Study of Īśvarapratyabhijñāvivṛti, Chapter 2.1.
Solomon, Esther A. 1977–1978. “Bhaṭṭa Udbhaṭa.” Annals of the Bhandarkar Oriental
Research Institute 58–59: 985–992.
Torella, Raffaele. 1988. “A Fragment of Utpaladeva’s Īśvarapratyabhijñā-vivṛti.” East and
West 38: 137–174.
Torella, Raffaele. 1992. “The Pratyabhijñā and the Logical-Epistemological School of
Buddhism.” In Ritual and Speculation in Early Tantrism, Studies in Honor of André
Padoux, edited by Teun Goudriaan, 327–345. Albany: State University of New York
Press.
Torella, Raffaele. 2002. Īśvarapratyabhijñākārikā of Utpaladeva with the Author’s Vṛtti,
Critical Edition and Annotated Translation. Corrected edition, Delhi: Motilal Banar-
sidass [Roma: 1994].
Torella, Raffaele. 2007a. “Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part I: anu-
palabdhi and apoha in a Śaiva Garb.” In Expanding and Merging Horizons. Con-
tributions to South Asian and Cross-Cultural Studies in Commemoration of Wilhelm
Halbfass, edited by Karin Preisendanz, 473–490. Wien: Verlag der Österreichischen
Akademie der Wissenschaften.
Torella, Raffaele. 2007b. “Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part II:
What is memory?” In Indica et Tibetica. Festschrift für Michael Hahn zum 65. Geburts-
tag von Freunden und Schülern überreicht, edited by Konrad Klaus and Jens-Uwe
Hartmann, 539–563. Wien: Arbeitskreis für tibetische und buddhistische Studien
Universität Wien.
Torella, Raffaele. 2007c. “Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part III. Can
a Cognition Become the Object of Another Cognition?” In Mélanges tantriques à la
mémoire d’Hélène Brunner, edited by Dominic Goodall and André Padoux, 475–484.
Pondicherry: Institut Français de Pondichéry/École Française d’Extrême-Orient.
via free access
Torella, Raffaele. 2007d. “Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part IV.

Light of the Subject, Light of the Object.” In Pramāṇakīrtiḥ. Papers dedicated to Ernst
Steinkellner on the Occasion of his 70th Birthday, edited by Birgit Kellner, Helmut
Krasser and Helmut Tauscher, 925–940. Wien: Arbeitskreis für tibetische und bud-
dhistische Studien Universität Wien.
Torella, Raffaele. 2014. “Notes on the Śivadṛṣṭi by Somānanda and its Commentary.”
Journal of Indian Philosophy 42: 551–601.
via free access
chapter 6
Alchemical Metaphors for Spiritual Transformation

in Abhinavagupta’s Īśvarapratyabhijñāvimarśinī
and Īśvarapratyabhijñāvivṛtivimarśinī
1 Introduction
It is a great privilege to be able to present a paper in honour of Alexis G.J.S. San-

derson. I was fortunate enough to be his pupil for two years at Oxford, and to
study with him subsequently in Leipzig and Kyoto. In my view, if one were to
accord Professor Sanderson the praise that he in fact merits, it would sound
(to those who do not know him) like embarrassingly unrestrained hyperbole.
Suffice it to say here that his example formed my ideal of intellectual integrity,
an ideal which entails relentless pursuit of the truth as part of a community of
scholars engaged in the kind of longitudinal study that prioritizes the field as a
whole over personal glory. Professor Sanderson taught me the value of admit-
ting when I don’t know, of sacrificing my own agenda in deference to the truth,
and of striving to be as transparent a mediator as possible in the act of trans-
mitting the words and ideas of the ancient Sanskrit thinkers to students of the
present day. It is with enormous gratitude to his unstinting scholarly labours (I
estimate he has logged well over a hundred thousand hours of research so far)
that I offer this paper in his honour.
The oeuvre of the Kashmirian Tantric master Abhinavagupta (fl. c. 975–1015)
is one of the many areas of research Professor Sanderson has mastered, and it is
this author which the present paper treats. Specifically, we here focus on a trope
found in Abhinavagupta’s two commentaries on the Īśvara-pratyabhijñā-kārikā
(ĪPK) of Utpaladeva, viz., that of an alchemical metaphor for spiritual transfor-
mation. These passages provide no small number of difficulties, because the
text as we have it is not secure, and because some knowledge of Indian alchemy
(rasāyana, dhātu-śāstra) is needed in order to translate it correctly. While I
do not claim to have solved these problems, this paper may certainly con-
tribute to our understanding of how Abhinavagupta thought about the process
of spiritual transformation conferred by the uniquely potent insight ( jñāna)
and yoga offered by initiatory Śaivism. Specifically, we learn much about his
usage of the key terms samāveśa, turya, and turyātīta, and it is hoped that
© christopher d. wallis, 2020 | doi:10.1163/9789004432802_008

via free access
alchemical metaphors for spiritual transformation 145
this paper advances our understanding of these topics, which are significant
within Śaiva theology.*
2 The Pratyabhijñā Doctrine of the Fivefold Self
Some readers are no doubt aware that in Utpaladeva’s ĪPK we find a teaching
on the “layers” of the individuated self (see, e.g., III.1.8), parallel to the later
Vedāntic teaching of five kośas based on Taittirīya Upaniṣad 2.3–5.1 This teach-
ing, formed as Sanderson says on “slight scriptural precedent,” is adopted by
subsequent gurus of Utpala’s lineage; for example, it has a prominent place in
Kṣemarāja’s Pratyabhijñāhṛdaya.2 In this model, the self is said to be fourfold:
void (śūnya), life-force (prāṇa), the subtle body consisting of the mind and its
faculties (puryaṣṭaka, i.e. the antaḥkaraṇa plus tanmātras), and the physical
body (śarīra). It is fivefold with the transindividual Power of Awareness (cit,
saṃvit) that permeates the whole. In fact, it is not only cit that permeates the
other levels: Kṣemarāja tells us that “it is clear that the very essence of each of
these levels is the fact of its pervasion by all the loci of perception prior to it,”3
where “loci of perception” (pramātṛ) refers to these levels of embodiment as
those realities with which contracted souls identify, and “prior to” means “more
fundamental than.”
Abhinavagupta adds to this teaching a homology implied but not spelled out
in the ĪPK itself, one that assimilates these five levels to the five “phases of lucid-
ity,” as Vasudeva (2004) calls them: the states of waking, dreaming, deep sleep,
the transcendental “fourth” state, and the state “beyond the fourth” ( jāgrat,
svapna, suṣupta, turya, and turyātīta). We will come to understand the last two
terms as we proceed.
Our texts in this study are Abhinavagupta’s two commentaries on the ĪPK,
his -vimarśinī (hereafter ĪPV) and his -vivṛti-vimarśinī (ĪPVV). The former is
his commentary on the kārikās themselves, the latter is his commentary on
Utpaladeva’s lost Vivṛti or longer auto-commentary. For both texts, we will
* An earlier version of this work is found in my unpublished doctoral dissertation (Wallis 2014).
1 “The kośas, mediated through the Pañcīkaraṇa system ascribed to Śaṅkara, had become part
of the [Deccani] vernacular tradition by the end of the twelfth [century, and proceeded
from there into the pan-Indian Sanskrit tradition].” Jason Schwartz, personal communica-
tion, April 2018.
2 See, e.g., Pratyabhijñāhṛdaya, chapter 7: śūnya-prāṇa-puryaṣṭaka-śarīra-svabhāvatvāt catur-
ātmā.
3 Pratyabhijñāhṛdaya, chapter 8: … dehādiṣu bhūmiṣu pūrva-pūrva-pramātṛ-vyāpti-sāratā-pra-
thāyām.
via free access
146 wallis
use the Kashmir Series of Texts and Studies (KSTS) edition. First, though, we
will consider the two verses of the ĪPK that Abhinavagupta is commenting
on, using Torella’s critical edition (2002), and summarize Abhinava’s initial
remarks thereon. All translations are my own unless otherwise noted.
kalodbalitam etac ca cit-tattvaṃ kartṛtā-mayam |

acid-rūpasya śūnyāder mitaṃ guṇatayā sthitam ||
ĪPK III.2.11
And this Awareness-principle, consisting of [unlimited] Agency, [be-

comes] limited—[though] it is strengthened by partial agency (kalā)—
abiding as a [mere] attribute in a person whose [habitual] nature is
unconscious, [identifying as he does with] the void, [prāṇa, mind,] and
[body].
mukhyatvaṃ kartṛtāyās tu bodhasya ca cid-ātmanaḥ |

śūnyādau tad-guṇe jñānaṃ tat-samāveśa-lakṣaṇam ||
III.2.12
By contrast, the characteristic of “immersion into That” is realization of

the primacy of the Self-that-is-awareness as the [only] true Knower and
Agent, and [a concomitant] insight regarding [the other layers of individ-
uality,] the void, [prāṇa, mind] and [body], as mere attributes of it.4
Explaining the first of these verses in his ĪPV, Abhinava first describes how
consciousness—which in its real nature is primordial, a priori, unlimited and
free—comes to be in the degraded state we consider as normal. Through the
power of his māyā expressed as the three malas, Śiva contracts himself into a
limited form (aṇu, the individual soul), then equips himself with the five kañ-
cukas beginning with kalā (cf. ĪPK III.1.9), resulting in a being that identifies
itself with what is actually objective, that is, the body, mind, prāṇa, and void
4 While these verses have been translated a number of times (cf. Torella 2002, 202–203), they
are not easy to translate in such a way that the reader clearly understands what is being said.
Here I capitalize words that are equivalent, on this view, to the Deity. An unobtrusive but
important word here is tat-, which I have translated as That but could also have been ren-
dered Him. Assuming that it is to be taken as compounded with what follows, then it must
denote what one is immersing in. The use of a gender-neutral pronoun that could just as well
denote neuter tattvam as masculine Him (= Śiva) is exemplary of the decreased theism of the
Pratyabhijñā phase of the tradition.
via free access
(cf. ĪPK III.1.8).5 Identification with the void (śūnya) can be identification with
the state of deep dreamless sleep (as Abhinava states it here) but also, and more
importantly, the void is the considered the primary locus of the limited “I” (see
ĪPK III.2.13), which, being in reality empty (śūnya), vainly seeks to reify itself
through identification with the body, mind, and prāṇa. This identification per-
sists in all three states of ordinary consciousness (waking, dreaming, and deep
sleep).6
Note that the real “I” is not here the core of an individual being as in Sāṅkhya,
but the one transindividual Self of all beings. The individual soul (aṇu) only
exists as a particular phase of that transindividual Consciousness, specifically,
an expression of the contracted state of bondage. Thus, one may argue, the non-
dual Śaiva’s “I” is closer to the view of the Vijñānavāda Buddhists than it is to
the ātman of Vedānta. (Even the dualistic Śaivas, who did posit a separate and
eternal soul, distanced their view of the ātman from those of the brāhmanical
schools (Watson 2006).)
3 The Exegesis of the Īśvarapratyabhijñāvimarśinī
Now let us look more closely at our first text as Abhinavagupta charts the trajec-
tory from bondage towards liberation, commenting on ĪPK III.2.12 (KSTS vol. 33,
p. 230–231):
yadā tūkta-gurūpadeśādi-diśā tenaivāhaṃ-bhāvena svātantryātmanā vyā-

pakatva-nityatvādi-dharma-parāmarśam ātmani vidadhatā tataḥ śūnyā-
deḥ prameyād unmajjya iva āsyate tadā turyāvasthā7 |
But when, through realizing [that the divine] qualities such as all-per-
vasiveness and eternality apply to oneself, by having the experience of
the [real] “I” whose nature is [unqualified] freedom—[an experience]
5 For example, Abhinava writes idantāpanna-dehādi-śūnyānta-prameya-bhāga-nimagnatvāt

prameyam, yo gauro, yaḥ sukhī, yas tṛṣito, yaḥ sarva-rūpa-rahitaḥ so ’ham: “The levels from
body to the Void are objects of awareness, [but] because of the submerging of a portion of
that objectivity, [there arise the erroneous cognitions] ‘I am the one who is pale’ (body), ‘who
is happy’ (mind), ‘who is thirsty’ (prāṇa), ‘who was devoid of all appearances’ (void)” (KSTS
vol. 33, p. 230).
6 seyaṃ jāgrat-svapna-suṣupta-rūpā saṃsārāvasthā (KSTS vol. 33, p. 230). The illusion of sep-
arate individuality persists even in the deep sleep state because of the presence of the saṃ-
skāras (cf. ĪPK III.2.13).
7 turyāvasthā ] conj. em. Torella (email communication, July 2014); turyātītāvasthā Ed.
via free access
148 wallis
pointed out by the guru’s instruction and other methods that I have
explained—[and] having therefore emerged as it were from [identifica-
tion with] the objective knowables of the Void etc., and [as a result] abid-
ing [in one’s real nature], then that is the [transcendent] state [called] the
Fourth.
yadāpi parāmṛṣṭa-tathābhūta-vaibhava-nityatva-aiśvaryādi-dharma-
saṃbhedena8 eva ahaṃ-bhāvena śūnyādi-deha-dhātv-antaṃ siddharasa-
yogena vidhyate, tadāsyāṃ turyātīta9-daśāyāṃ tad api prameyatām ujjha-
tīva |
When further [the layers of the objective “self”] from the Void to the [very]
tissues of the body are transmuted10 by means of the “alchemical elixir,”
i.e. by the [fundamental] “I”-sense which is certainly conjoined with the
qualities of magnificent power (vaibhava), eternality, sovereignty, [and
8 -vaibhava-nityatva-aiśvaryādi-dharma-saṃbhedena ] conj. em.; vaibhava-nityaiśvaryādi-

dharma-saṃbhedena KSTS ed.; vaibhavādi-nityaiśvarya-saṃbhedena Iyer & Pandey ed.
9 turyātīta- ] conj. em. Torella (email communication, July 2014); turya- Ed.
10 Āyurveda scholar Dominik Wujastyk (of the University of Alberta) recommended this
translation of vidhyate as “transmuted” (over that of “penetrated”) based on his reading
of the rasāyana literature, especially the Rasa-ratna-samuccaya 8.94–95 and the Bodhinī
thereon (email communications, 7 and 9 July 2014). Ashok Aklujkar also contributed a
citation from the same text (5.11: vedhajaṃ suvarṇam—pārada-vedhena saṃjātaṃ suvar-
ṇam), which I believe verifies that vedha must mean transmutation (or similar), not
piercing or penetration, though he would wish to retain the latter translation (email,
7 July 2014). (See also n. 27 below.) The alchemical metaphor here (elaborated further
in the ĪPVV, infra) is of course not original to Abhinavagupta; we find it earlier in the
well-known eighth century Buddhist text, the Bodhicaryāvatāra, 1.10cd: rasajātam atīva
vedhanīyaṃ sudṛḍhaṃ gṛhṇata bodhicitta-saṃjñaṃ, which I translate as “Firmly take
hold of the alchemical elixir called Intent to Awaken (bodhicitta), which must be thor-
oughly transmuted.” Vesna Wallace (1997, 19) translates almost identically; in this verse,
it appears, rasajāta is unmodified mercury that must be properly transmuted to be safe
for consumption, implying that some refinement of the initial raw bodhicitta is neces-
sary. (However, Matthew Kapstein [email communication, 9 July 2014] points out that
both the Sanskrit commentator and the Tibetan translation do not take vedhanīyaṃ as
translated above, but rather in the active sense, “able to transform [this aśuci-pratimā to
a jina-ratna-pratimā, 1.10ab]”—e.g., Prajñākaramati glosses atyuccavedhakāritvād—atīva
vedhanīyam.) Our passage does support this latter reading, for here we certainly have
siddha-rasa denoting a mercury preparation that can transmute base metals into gold (or
more accurately, extract gold from base metals). Thus, the pure dynamic power of aware-
ness called cit is here compared to a chemical catalyst: it needs no refinement or alteration,
but can alter that which it contacts.
via free access
others] of such nature that are cognized [as aspects of that “I”], then
in this state [called] Beyond the Fourth they abandon (as it were) their
objectivity.
Having introduced the three states of waking, dreaming, and deep sleep, Abhi-
navagupta now discusses turya and turyātīta, which complete the set of the
“phases of lucidity.” Now, in this passage, I take Abhinava to be reconciling two
modes of realization: one gnostic, rapid, transcendent, and liberating, and the
other yogic, gradual, immanent, and siddhi-conferring. Here, the Fourth state is
the gnostic realization that one has wrongly taken objective realities to be the
self; it is waking up out of the trance of believing “I am the body,” etc. Such a real-
ization can be sudden because it requires no transformation, only a recognition
of what is already the case, including a reflective awareness (parāmarśa) of the
qualities (dharmas) of one’s real self. As Torella puts it, “the adept, after becom-
ing aware of the supreme nature of the I, becomes as though withdrawn from
the knowable which formed his fictitious identity” (2002, xxxiv). Turya is then
an exclusive kind of realization. By contrast, the process of turyātīta (“Beyond
the Fourth,” but not actually a fifth state)—here described in terms of pene-
trating the layers of that constructed identity with this deeper awareness or
transcendent I-sense—is inclusive and gradual, requiring yogic practice. In the
turyātīta experience, the objective layers of the limited self are seen as expres-
sions of the transindividual divine consciousness, and thus are recovered as
part of a greater “I” than the one they were excluded from in the previous turya
state. This process by which the cidātman penetrates the layers of body, etc.,
is likened to alchemical transformation, whereby the elixir called siddha-rasa
transforms a base metal into gold (or extracts the gold from the base metal).11
The use of the word iva (last word of the passage just cited) denotes that the
body etc. do not actually cease to be knowables when they come to be seen
as nothing but crystallizations of the dynamic “liquid” essence of conscious-
ness in the turyātīta state,12 just as the previous iva denoted that emerging from
11 Cf. Kulārṇava-tantra 14.89: rasendreṇa yathā viddham ayaḥ suvarṇatāṃ vrajet | dīkṣā-
viddhas tathā hy ātmā śivatvaṃ labhate priye ||, “Just as iron penetrated by mercury
becomes gold, even so a soul penetrated by initiation becomes divine.” Torella writes,
“[here] the various components of the levels of the limited subject are gradually pene-
trated by the elixir of the I, until they become, so to speak, transfigured, removed from
their nature of [being merely] knowable realities” (2002, xxxiv).
12 Kṣemarāja’s phrase, cidrasāśyānatā-prathanātmā samāveśaḥ, in the context of a parallel
discussion, in Pratyabhijñāhṛdaya, chapter 19.
via free access
150 wallis
identification with knowables does not mean completely leaving them behind
(which would entail physical death).
Now, the coherence of this passage only emerges after the emendation to the
edition suggested by Torella and adopted here, that of exchanging the words
turya and turyātīta. This may seem a dramatic emendation, but it would make
little sense for Abhinava to list the three states of ordinary consciousness, then
proceed to turyātīta when what is obviously called for is turya, “the Fourth.”
Further, it would make no sense to argue that turya is an extension of the tury-
ātīta state without completely ignoring the meaning of those two words; but
the other way around exactly matches the meaning of the words.
But what would occasion such a confusion in the edition? It may well be
that later scribes (for we do find the edition’s reading in the manuscripts), influ-
enced by the more transcendentalist mainstream Indian philosophies, simply
could not imagine that turya could denote the transcendent state while tury-
ātīta, which is obviously intended as the higher attainment, embraced imma-
nence. But this is precisely in line with Abhinavagupta’s Kaula view, for with
the text emended as Torella suggests, we have here a model that is central to
the Kaula Kālīkula, which Sanderson characterizes as “transcendence followed
by an expansion that causes the state of enlightenment to pervade the tran-
scended” (Sanderson 2007, 402–403). The Śivasūtra (well known to Abhinava)
inherits this model, teaching the “establishing of this realization first through
withdrawal into the heart of consciousness and then through its expansion
into the states that constitute the mundane awareness of the bound” (ibid.),
which precisely characterizes our ĪPV passage. For example, in the Śivasūtra
(1.7) we find the teaching that the Fourth state can spread to the ordinary
states of jāgrat, svapna, and suṣupta, imbuing them with awakened conscious-
ness, which the Śivasūtra calls turyābhoga but which is simply turyātīta under
another name.13 Of course, an examination of all the extant ĪPV manuscripts,
preferably after forming a critical stemma, would be necessary to make a final
ruling on the reading of the passage.
To return to our text, Abhinava concludes his ĪPV discussion of ĪPK III.2.12
by informing us that turya and turyātīta are forms of samāveśa, which, when it
becomes continuous and stable (āsyate), is itself liberation.
seyaṃ dvayy api jīvanmuktāvasthā samāveśa ity uktā śāstre, samyag-āve-

śanam eva hi tatra tatra pradhānam, tat-siddhaye tūpadeśāntarāṇi |
13 jāgrat-svapna-suṣupta-bhede turyābhoga-saṃbhavaḥ.
via free access
This twofold state of one who is liberated while living is called samāveśa
in the scriptures. For complete entering14 is itself primary in each of these;
other teachings are [only] for its attainment.
This is a surprising statement, perhaps, for I know of no scriptural passage in

which these two states are called samāveśa. What Abhinava wants us to under-
stand, I think, is that when the scriptures use the term samāveśa, they are
always referring to one of these two states. In turya, then, one fully and directly
penetrates into one’s true nature, while in turyātīta, one causes that nature to
fully and gradually penetrate the objective levels of one’s limited selfhood; for
this reason they can both be appropriately referred to with the word samāveśa
(from ā√viś, to penetrate). Abhinava continues:
dehapāte tu parameśvara evaikarasaḥ, iti kaḥ kutra kathaṃ samāviśet
But at the fall of the body, there is only one essence: the Supreme Lord.
Thus, who could enter (/immerse), where and how?
In other words, it is only meaningful to speak of samāveśa in the context of

embodiment, for only in that context are there apparently differentiated layers
of selfhood such that there can be an “entry” of the locus of identity (ahaṃ-
bhāva) from the body, etc., into cit, or an “entry” of cit into one of the layers
of limited selfhood (dehādi)—the former entry being turya and the latter tury-
ātīta.
4 The Parallel Passage in the ĪPVV
The corresponding ĪPVV passage (KSTS vol. 65, 327–331) is similar but sheds
more light on some important points while simultaneously greatly complicat-
ing the issue. Abhinava elaborates further on the alchemical metaphor briefly
introduced in the ĪPV; here, though, if we do not emend the published text, he
appears to have changed his view from that seen in the ĪPV. There the alchemi-
cal metaphor was reserved for the turyātīta state, while here we see two stages of
the alchemical metaphor, corresponding to both turya and turyātīta.15 Further-
more, it seems that he now posits two different modalities for attaining both
14 Abhinava is telling us that the sam- in samāveśa is in the sense of samyañc.

15 Though Torella proposes an emendation which would bring the ĪPVV in line with the ĪPV
(see below).
via free access
152 wallis
states, one gnostic and one yogic. (Here I differ from Torella’s 1994 hypothesis
that a single turya state bifurcates into two kinds of turyātīta.) Abhinava writes:
etad ajñāna-rūpa-mala-pratidvandvitayā samāveśa-lakṣaṇaṃ satya-sva-

rūpe samyag āsamantāt praveśa-lakṣaṇaṃ jñānaṃ, yal-lābhena jñānī,
yad-abhyāsena ca deha-prāṇādāv ananta-saṃvid-dharmātmaka-vibhava-
samāsādanāt yogī bhavati |
[Utpala teaches that] the “distinguishing mark of samāveśa” is “insight,”

since it is opposed to the Impurity that is ignorance, being characterized
by a perfect (samyag), that is to say complete (ā samantāt), entry into
one’s true nature,16 obtaining which one becomes a gnostic ( jñānī), and
practicing which, on the levels of body, prāṇa, etc., one becomes a yogī,
due to attaining the glory (vibhava) that is an intrinsic quality of infinite
Consciousness.
etad uktaṃ bhavati—yadā ahaṃbhāvaḥ svātantrya-diśaiva vyāpitva-nit-

yatvādi-parāmarśa-balāt śūnyādeḥ prameyīkṛtād unmajjya iva āste, tadā
turyatā;17 tadāpi ca śūnyādi-saṃskāro ’pi asti,—iti vyatireka18-turyātīta-
samatā eva |
This is said [already in the ĪPV]: when the [true] I-sense, due to the
power of the realization of its all-pervasiveness, eternality, etc., through
the [scriptural] indication of its [innate] autonomy, emerges as it were
from the objectified [levels of limited selfhood]—Void etc.—and abides
[in its real nature], then that is the state [called] the Fourth. Nevertheless
[in that state] the impressions of the Void, etc., still remain. Thus this has
exactly the same [nature] as [that which is called] the “separated tury-
ātīta.”
16 We have here an implicit analysis of the word samāveśa: samyag and/or āsamantāt +
praveśa = samāveśa.
17 turyatā ] conj. em. Torella (email communication, 16 July 2014); turyātītatā Ed. Without
this emendation, the following comment turyātīta-samatā eva makes little sense.
18 iti vyatireka ] conj. em. Torella; iti avyatireka Ed. Following this emendation (proposed in
an email, 15 July 2014) we can take vyatireka in the sense of kevala or kaivalya, i.e., a spir-
itual state which is separated from the saṃskāras but does not dissolve them. Even if we
do not emend, we can still argue for the same meaning: avyatireka- could indicate that
he is “unseparated” from his saṃskāras in the sense of still having them, though they are
now powerless to obscure his real nature. However, the emendation makes for a clearer
meaning.
via free access
Though I am not aware of another usage of the technical term vyatireka-

turyātīta, the meaning here is clear enough (after applying Torella’s suggested
emendation): the gnostic who does no yoga enters into a transcendental turya
state in which he is authentically immersed in his essence, but the impres-
sions of limited selfhood from which he has successfully separated (vyatireka)
himself from remain undissolved (thus his social self might exhibit little to no
change). Thus, Abhinava argues, the attainment of turyātīta of the vyatireka
variety is in fact no different from the turya state itself. This obviously sets
up the possibility of a higher attainment, an avyatireka-turyātīta in which one
dissolves those impressions through practice, allowing the practitioner to be
not-separated (avyatireka) from his body, mind, etc., yet still liberated; i.e. an
immanentist state of liberation.
It is hard to see what Abhinavagupta has gained here, because in the sim-
pler ĪPV model, turya was the transcendental state and turyātīta the immanent
(and therefore higher) attainment. Perhaps he simply wants to indicate here
that either state can be attained by either gnostic or yogic means. But there is
more evidence to examine before drawing conclusions.
Now we see the yogic version of the turya → turyātīta progression. In the fol-
lowing paragraph (continuing directly from the previous ĪPVV citation), note
that the first part closely parallels the ĪPV passage we have seen above (pp. 147–
149), while the second part is new data.
yadā tu parāmṛṣṭa-nityatva-vyāpitvādi-dharmakaiśvarya-ghanātmanā
ahambhāva-siddharasena śūnyādi-deha-dhātv-antaṃ19 vidhyate yena
prameyatvāt tat cyavata iva, tadā turya-daśā;
But when [all the layers of limited selfhood] from the Void to the tissues
of the body are penetrated by the “alchemical elixir” that is the [true]
I-sense—replete with the sovereignty in which the qualities of eternal-
ity, all-pervasiveness, etc., are cognized [as aspects of that “I”]—through
which [penetration] they abandon (as it were) their objectivity, then that
[too] is [called] the Fourth State.
yadāpi viddho ’sau prāṇadehādi-dhātuḥ saṃvid-rasena abhiniviṣṭo ’tyan-

taṃ kanaka-dhātur iva jīrṇaḥ kriyate yena sa druta-rasa iva ābhāti kevalaṃ
tat-saṃskāraḥ, tadāpi turyātīta-daśā sā bhavati |
19 Oriental Research Library manuscript no. 2403 has śūnyādi-deha-dhāturtvaṃ here.
via free access
154 wallis
When, further, these elements of prāṇa, body, etc., [already] penetrated

by the elixir of Awareness, are thoroughly permeated [by it], they are
[then] “digested” like the element of gold [is by mercury], by which [pro-
cess] their purifier, the “liquefied essence” [of Awareness] as it were, alone
remains—then that too is the state Beyond the Fourth.
Here we have a clear progression of turya → turyātīta without the necessity of

emendation. Or do we? According to the earlier ĪPV passage, in the Fourth state,
one simply transcends the objective layers of the self, rather than those lay-
ers losing their objectivity. Thus, either Abhinavagupta has changed his view
since writing the ĪPV, or an emendation is indeed necessary here. If the lat-
ter, we could either emend tat to sa (“one leaves behind their objectivity”) or
we could emend turya-daśā (in the first paragraph above) to turyātīta-daśā.
The latter solution, tentatively proposed by Torella in an email communica-
tion (July 2014), seems to me to ignore the grammar that suggests two stages
here (the first structured around the relative/correlative yadā tu … tadā, and the
second around yadāpi … tadāpi); or rather, more correctly, he sees the grammar
(after his emendation) as referring not to two successive stages but to two kinds
of turyātīta, the api in tadāpi informing us that “thus, this too is turyātīta.” How-
ever, then we have the problem that two apparently distinct stages of the same
process are denoted by the very same word, turyātīta. That they are distinct
stages is evidenced by the fact that in the first phase (the paragraph ending
with the compound turya-daśā), we see the verb √vidh (penetrate20), and in
the second (turyātīta) phase we have √vidh followed by abhini√viś, which here
denotes a further development of the same process (as also indicated by the
adverb atyantam, construing with abhiniviṣṭa). Furthermore, Abhinava’s reca-
pitulation of this discussion (KSTS vol. 65, 348) would seem to argue against
Torella’s conjecture here. On the other hand, if we did adopt the emendation,
it would allow us to preserve the notion seen in the ĪPV that turya is the gnos-
tic attainment and turyātīta the yogic one. It seems to me, however, that this
creates more problems than it solves.
In summary, by not adopting the emendation, we see here a yogic21 ver-
sion of the Fourth state that can be developed into the state Beyond the
Fourth, thus indicating a change in Abhinavagupta’s thought since the ĪPV.
20 Note that the verb is here being used in a more precise sense than in the ĪPV passage, in
which (I argue) it means “transmute.” See also the extended discussion on vedha-dīkṣā in
TĀ 29, translated in an appendix to my doctoral dissertation (Wallis 2014).
21 Besides the alchemical metaphor, Abhinava signals to us that this is a yogic process with
the word aiśvarya, which, like v(a)ibhava, often relates to yogic power (siddhi).
via free access
To explain in more detail my understanding of this rather difficult passage,

the process goes like this: having inundated/penetrated (viddha) the objective
layers of selfhood (body etc.) with the “elixir” of one’s ultimate nature (i.e.,
saṃvid-rasa, autonomous dynamic consciousness), the “gold” hidden within
them is extracted, i.e. their dependence on consciousness as their substrate is
revealed.22 When those layers have become completely permeated (abhiniviṣṭo
’tyantaṃ), through, one presumes, further spiritual practice,23 all trace of their
objectivity (and the saṃskāras thereof) is “worn away” or “digested” ( jīrṇa)
by the elixir of consciousness—as mercury eats up gold flakes—which thus
becomes a single unitary mass of awareness (prakāśa-ghanam eva saṃvid-
rūpam, cited infra).
Our understanding of Abhinava’s vision of this process depends in part on
grasping his use of an alchemical metaphor rooted in the complex and often
ambiguous rasāyana (alchemy, proto-chemistry) theories of medieval India.
In this matter I was fortunate to receive the helpful comments of Professors
Wujastyk and Houben (of the University of Alberta and the Sorbonne, respec-
tively), who clarified that jīrṇa here stands in for jāraṇa, one of the sixteen
rasa-saṃskāras (alchemical processes). Jāraṇa can mean digestion, assimila-
tion, or swallowing (in much the same sense that we speak of an acid “eating
away” at a metal). Here the alchemical elixir of the metaphor is of course pre-
pared mercury (siddha-rasa), which can indeed “digest” gold (the modern term
is amalgamate).24 That Abhinavagupta was aware of the basics of alchemy is
confirmed by his use of the compound druta-rasa, for according to Houben,
“initially the mercury remains as fluid as before it started to ‘eat’ the gold etc.
but at a certain point its viscosity increases significantly … [it] remains fluid
or druta [only] as long as it is not saturated.”25 Clearly, Abhinava wishes to
emphasize that here this saturation does not occur, that Consciousness must be
22 Cf. Sarvajñānottara 1.5: tāmrasyaiva tu hematvam antarlīnaṃ yathā sthitam | antarlīnaṃ

tathā jñeyaṃ śivatvaṃ pudgalasya tu (“Just as gold is hidden within copper, in the same
way the Divinity which a man seeks to know is hidden within [him].”).
23 In the present context, the nature of the yogic practice alluded to is very likely the proto-
kuṇḍalinīyoga that Utpaladeva outlines at ĪPK III.2.19–20 and which was presumably
elaborated in his Vivṛti. For Abhinava, such practice must be animated by bhāvanā (con-
templative insight) to be truly effective.
24 This can be clearly seen in a video made by Dr. Andrea Sella (Department of Chemistry,
University College London), which includes the following comment: “[In ancient times]
mercury was absolutely essential … in extracting gold and purifying it … what gold can do
is, it can actually dissolve in mercury” (https://www.youtube.com/watch?v=gKxCw889qck
&feature=youtu.be).
25 Email communication from Dr. Jan E.M. Houben, 7 July 2014.
via free access
156 wallis
understood purely as a catalyst (something that effects change but is not itself
affected): it remains as it is, a dynamic “fluid” essence (druta-rasa = cid-rasa).
First, then, in this alchemical vision, the mercury transmutes the base metal
into gold,26 then “digests” or absorbs it without a (perceptible) trace (as can
be seen in the chemistry video cited in note 24).27 That is to say, if we follow
the terms of the metaphor strictly, first the layers of body, etc., are experienced
as expressions of the dynamic essence of awareness, then all the saṃskāras
implanted in those layers through one’s earlier experience of them as other
than awareness are dissolved or “digested.”
To summarize, if we are constituting and interpreting the text correctly,
Abhinava has changed his view as follows: in the ĪPV, turya is an exclusive,
gnostic, transcendental state and turyātīta an inclusive, yogic, immanent one
(the progression from one to the other exemplifying the typical Kaula model of
transcendence followed by pervasion), with the alchemical metaphor denot-
ing only the turyātīta stage; whereas in the ĪPVV, there is a rapid gnostic version
of turya progressing to turyātīta (in which saṃskāras are not dissolved) and a
gradualist yogic version of the same (in which they are dissolved in the turyātīta
phase), both stages (of the latter) being described in terms of the alchemical
metaphor. In the second text, then, we have a fork in the road, giving us four
stages, only two of which a given practitioner is likely to traverse.
Before we move on to examine the last version of the alchemical metaphor,
we have one more problem with the present passage: how to interpret the final
compound of the phrase sa druta-rasa iva ābhāti kevalaṃ tat-saṃskāraḥ. Here
I differ from Torella (2002, 209 n. 35), who seems to interpret it to mean that
only the saṃskāras (impressions) of śūnyādi-dehāntam remain. However, that
case was already specified for the first, gnostic turyātīta (tadāpi ca śūnyādi-
saṃskāro ’py asti, above), and if that were intended here we would have noth-
ing to differentiate the two turyātītas described. Thus I take tat-saṃskāraḥ in
apposition to druta-rasaḥ, in the meaning “the purification (or refinement) of
that,” or, as translated above on page 10, as a bahuvrīhi meaning “their purifier,”
the antecedent of the neuter pronoun tat being śūnyādi-dehāntam in either
26 Even though professional alchemists must have known that mercury actually extracts gold
from a base metal, rather than magically transmutes that metal into gold, vedha is cer-
tainly used in the sense of transmute or transform—see the citations in note 10, in one of
which √vidh is glossed with pari√ṇam (Rasa-ratna-samuccaya-bodhinī ad 8.95). See also
the relevant statement in Roşu 1982, 366: “la transsubstantiation [alchimique] du corps
(deha-vedha) étant calquée sur la transmutation des métaux vils (loha-vedha) …” He cites
Rasārṇava 12.165–166 in support (ibid., note 21).
27 The reader who has German and wishes to know more about this arcane world of Indian
alchemy is referred to Hellwig 2009.
via free access
case. Torella argues (email communication, 10 July 2014) that the saṃskāras
of śūnyādi-dehāntam cannot be entirely dissolved here because then there
would be no possibility of samāveśa, since, consciousness having become a
single unitary mass, there would be nothing that could enter or be entered (cf.
p. 151 supra). However, while Abhinava unambiguously does say this with regard
to the after-death state (dehapāte tu eka-ghanā eva śivateti tadā samāveśādi-
vyavahāro na kaścid, KSTS vol. 65, 328), I am not at all sure that he thought
it impossible to go beyond samāveśa, as generally understood, before death;
after all, in a continuous nondual state of “complete immersion” (a new sense
of samāveśa starting with Utpala’s usage; see Wallis 2014) there will no longer
be any kind of “entry” or “penetration” (ā√viś) per se.
We need not speculate overmuch on this question, however, for we can find
evidence to suggest that Abhinava did regard such supervention of samāveśa
as possible. That evidence is found almost twenty pages further on in the ĪPVV
(KSTS vol. 65, 348, commentary on ĪPK III.2.19), where he recaps his earlier dis-
cussion (our most recent passage above) but also adds new information:
aham ity eka-rasena anuvedhe tu, yadā idantā ācchāditā bhavati, bhāvanā-
sātmyād īśvara-sadāśiva-saṃvidi iva turya-daśāyāṃ rasa-viddha-tāmra-
kanaka-nyāyena, yadā vā sarvathaiva pradhvaṃsitā vidrāvitā vā bhavati
turyātīta-daśāyāṃ śākta-saṃvidi iva tan-nija-rūpa-samyag-viddha-kana-
ka-rūpatātyanta-jaraṇāpādita-tat-saṃskāra-vaśa-pītatā28-avaśeṣa-vidru-
ta-rasa-nyāyena; tadā pūrṇa-svātantryollāsa eva deha eva sati api …
In the [process of] transmutation by the “one taste” that is [the funda-
mental] “I,” when
[a] objectivity is covered, i.e. in the Fourth state [that arises] due to
becoming habituated to meditative contemplation [on reality], in
which one possesses the consciousness of Īśvara or Sadāśiva as it
were, according to the maxim of gold [being extracted] from cop-
per due to being penetrated by mercury,
28 pītatā ] conj. em. Isabelle Ratié (email, July 2014); pītalatā Ed., though pītatā as “gold”
is problematic. Another possibility is to not emend the text, and take it instead to be
speaking of the digestion/dissolution of the brass or copper (pītala) that remains after
gold has been extracted from it. This has not been adopted on the assumption that the
present passage recaps the one on p. 153 supra. Also, I presume that Abhinavagupta, not
being himself an alchemist, viewed the process of vedha as one of transmutation more
than extraction—and if so, there would be no brass (or copper) left to digest. Further, the
metaphor of transmutation suits his purposes better.
via free access
158 wallis
or when
[b] [objectivity] is completely destroyed or “liquefied”—i.e. in the state
Beyond the Fourth, the level of Goddess-consciousness, as it were
[śākta-saṃvit, i.e. śakti-tattva]—according to the maxim of liquid
mercury thoroughly digesting the remaining “gold,” i.e. the power
of the impression(s) of that [objectivity], which [now] have the
appearance of gold [i.e., radiant and soft] due to having been thor-
oughly penetrated by the innate form of that [“I”/rasa],
then [in either case]
[c] there is simply the delightful blossoming of full autonomy, even
while the body exists.29
This passage features a piling on of parenthetical phrases that is rather easier

to understand in the Sanskrit than in literal English translation;30 I have illus-
trated the basic structure here as “when a) or b), then c).” Several things become
29 Some parallel passages: cf. TĀ 14.12, TĀ 5.151 (svayaṃbhāsātmanānena tādātmyaṃ yāty

ananyadhīḥ | śivena hematāṃ yadvat tāmraṃ sūtena vedhitam ||), and Yogarāja’s com-
mentary ad Paramārthasāra 96, a verse on anupāya and atitīvra-śaktipāta, the effect of
which is compared by the commentator to alchemical transformation by means of siddha-
rasa ( yathā tāmra-dravyaṃ siddharasa-pātāt suvarṇībhavati). Note that later in the same
passage the aspirant is referred to as anugraha-śakti-viddha-hṛdayasya, “one whose heart
has been penetrated/transmuted by the power of divine grace (i.e., śaktipāta).” We find
the same terminology used with reference to dīkṣā, e.g. in the Kulārṇava-tantra (14.89):
rasendreṇa yathā viddha-mayaḥ suvarṇatāṃ vrajet | dīkṣā-viddhas tathā hy ātmā śivatvaṃ
labhate priye ||, “Just as [a metal] penetrated by mercury becomes gold, even so a soul
penetrated by initiation becomes divine.” Cf. Goodall 2004, 402 note 904:
The conception that gold can be created out of copper with an alchemical preparation
is, as Prof. Isaacson has pointed out to me, commonly used in tantras as an image for the
irreversible transformation that takes place in dīkṣā. See, e.g., Kiraṇa 59.36c–38b … And
cf. Sarvajñānottara 1.6 (Devakoṭṭai ed.): rasa-viddhaṃ yathā tāmraṃ hematvaṃ pratipady-
ate | tathātmā jñāna-sambandhāt śivatvaṃ pratipadyate ||. Cf. also Haravijaya 6.137 … In
his commentary thereon Alaka cites the following verse: rasa-ghṛṣṭaṃ yathā tāmraṃ na
bhūyas tāmratāṃ vrajet | evaṃ yuktaḥ śivatvena na bhūyaḥ paśutām vrajet.
I render the last verse cited as: “Just as copper rubbed with mercury [becomes gold
and] does not again become copper, in the same way one united with Divinity does not
again become a bound soul.” Goodall informs me (email communication, 13 July 2014)
that the Sarvajñānottara verse cited in his footnote (viz., 1.6) is the most typical form of
the maxim; the verse immediately preceding it is cited above in note 22.
30 And I wonder if the largely redundant first part of the long compound (tan-nija-rūpa-
samyag-viddha-kanaka-rūpatā), which fits awkwardly, might have been a marginal anno-
tation in a manuscript that got incorporated into the main text. However, it may be more
likely that Abhinava is here incorporating language that Utpala used in his lost Vivṛti, as
he tends to repeat Utpala’s compounds with added glosses.
via free access
clear from this paragraph, despite its density. First, if my translation is correct,
it corroborates my reading of the previous alchemical passage. Second, it con-
firms that the saṃskāras are indeed “thoroughly digested” in the state Beyond
the Fourth (atyanta-jaraṇāpādita-tat-saṃskāra-), and that this can occur with
the body still existing (deha eva sati api). However, having said this, we must
note that “thoroughly digested” does not mean “entirely destroyed” if Abhi-
navagupta is holding strictly to the terms of his metaphor; for when mercury
absorbs gold leaf such that the gold is entirely dissolved and thus completely
invisible, it is in fact still present in the mercury and can be retrieved by evap-
orating the latter in a retort. We have no way of knowing if Abhinava knew
this, but if so, Torella could well be correct in arguing that a subtle trace of the
saṃskāras (which are themselves subtle traces) can remain in the turyātīta-
daśā. What certainly is entirely dispelled or dissolved (sarvathaiva pradhvaṃ-
sitā vidrāvitā vā) in that state is objectivity, which was only “covered” (ācchāditā)
by subjectivity in the turya state.
We find similar language in chapter three of the Pratyabhijñā-hṛdaya,
authored by Abhinavagupta’s disciple Kṣemarāja, where the Sadāśiva-tattva is
described in these terms: “[a level of consciousness] in which an implicit and
indistinct objectivity is covered by [the predominant] subjectivity [literally,
‘I-ness’]” (sadāśiva-tattve ahantācchāditāsphuṭedantā-mayaṃ).31 Thus, accord-
ing to the Pratyabhijñā schema, abiding in turya means achieving the īśvara-
or sadāśiva-tattva and abiding in turyātīta means reaching the śakti-tattva. In
either case the result is the “delightful blossoming of full autonomy” (pūrṇa-
svātantryollāsa), i.e. liberation. But if the yogic/alchemical turyātīta is equiv-
alent to reaching śakti-tattva, are we supposed to understand that the gnostic
turyātīta reaches śiva-tattva, despite the fact that the latter turyātīta does not
dissolve the saṃskāras and the former does? If so, does Abhinava mean to
imply subtly that Śakti is in reality higher than Śiva? This would contradict
the mainstream doctrines of Śaivism, but not of the Krama, the sect of Abhi-
nava’s first initiation. We know from the Mālinī-vijayottara-tantra, Abhinava’s
root-text, that the tattvas of Sadāśiva and Īśvara are indeed associated with
the Fourth state and that “Śiva and Śakti exist in the state Beyond the Fourth”
(2.28c–29b32). So we would expect that the two turyātītas are associated with
tattvas 1 and 2. But which is which?
31 However, Kṣemarāja loc. cit. posits īśvara-tattva as a level in which subjectivity and objec-
tivity are equal and opposite, whereas in our present passage Sadāśiva and Īśvara are not
differentiated, both being described as a level at which objectivity is “covered.”
32 See Vasudeva 2004, 209–210.
via free access
160 wallis
The reader will recall that we looked ahead eighteen pages in our primary
source to see how Abhinavagupta recapitulated his alchemical metaphor. Now
we return to our main passage (KSTS vol. 65, 330) to address the questions just
raised. First we see that the distinction we have posited between the gnostic
and yogic paths to liberation is not as clear-cut as it would seem:
yadā tu dehādeḥ kiṃ tattvam iti cintopakramaṃ prakāśa-ghanam eva

saṃvid-rūpam iti, tadā bodha-svarūpīkṛtaṃ tad-rasānuviddham iva33 śūn-
yādi-dehāntam avabhāti. iti abhyāsāt tasya saṃvid-dharmāḥ śakti-viśe-
ṣāḥ samyag āviśanto vibhūtīr utthāpayanti. anabhyāse ’pi tu tat-kṣaṇāve-
śa eva ānandodbhava-kampa-nidrā-vyāpti-rūpa-ghūrṇy-āvirbhāvana-kra-
meṇa jīvanmuktatā-lābhaḥ |
When one begins to contemplate “What is the reality of the body, etc.?”
[and subsequently realizes] “it is simply a form of awareness, replete with
the Light of Consciousness,” then those [levels] from the Void to the body
manifest as [they really are,] of one essence with Awareness, as if trans-
muted by its elixir. Thus, due to practicing [this insight], the qualities
of His consciousness, which are aspects of śakti, fully penetrate [those
various levels], causing the [various] powers (vibhūti) to arise. But even
without practice, in the [rare] case of an instantaneous immersion into
That, one obtains the state of liberation-in-life through the process of
the direct experience of [the Five Mystic States]: Bliss, Ascent, Trembling,
Sleep, and “Whirling,” which means Pervasion.
This passage serves as an explanation of tat-śakti-samāveśa in Utpala’s Vṛtti on

III.2.12. The passage confirms for us that Abhinava sees samā√viś (= atyantam
abhini√viś in the previous iteration of the alchemical metaphor) as denoting
the further development of the process first denoted by √vidh. More impor-
tantly, here we find crucial evidence that the distinction between the yogic
and gnostic in Abhinavagupta’s thought is not as clear-cut as we have been
led to expect: in this passage, a contemplation (cintā) on the nature of real-
ity leads to a realization that entails a spiritual transformation metaphorically
described as alchemical transmutation (= turya stage), which then may be sta-
bilized and enhanced with yogic practice such that the qualities of this deeper
awareness (e.g., svātantrya-śakti) come to fully penetrate or infuse (samā√viś)
33 iva ] conj. em.; eva Ed.
via free access
all the layers of limited selfhood (= turyātīta stage). Gnostic realization is here
inseparably wedded to the pañcāvasthāḥ or Five Mystic States that we see
repeatedly in the Kaula scriptures. This emphasis on direct experience (āvir-
bhāvana) demonstrates that Abhinava’s understanding of the path of the jñānī
is not one of intellectual or conceptual realization, but rather one of insights
into the nature of reality so powerful that they spontaneously bring on psy-
chophysical experiences.
Curiously, he uses the phrase “instantaneous immersion” or kṣaṇāveśa in
describing gnostic realization but then immediately follows it with the term
krama, denoting a sequential process of passing through the Five States. I
would hypothesize that Abhinava is saying that each of the Five States is (or
rather can be) an example of kṣaṇāveśa; even though there is a process, it may
unfold spontaneously and in sudden leaps, in connection with the jñānī’s deep
contemplation of the nature of reality.
At any rate, now our reading of two distinct tracks (or two distinct turyātī-
tas), one gnostic and one yogic, is problematized. Yet we cannot abandon it, for
on the very next page of our text (KSTS vol. 65, 331) we find the following:
turya iti etad-daśā-samāpatti-paryanta-rūpā api turyātītatā tatraiva uktā,

vyatirekeṇa tu śūnyāder avasthāpane bodhasya turyātītatā tatraiva uktā—
śuddhātmatā niṣkalatā śuddha-caitanyam iti sāmānya-darśaneṣu sarvot-
tīrṇaika-tadrūpa-vedakeṣu darśiteti sūcayati |
The state of turyātīta taught [above] with reference to that [blossoming of

insight34] is simply the [further] extension of the realization of the state
called turya. But that state of turyātīta was taught there as a state of aware-
ness in which Void etc. remain [as objective knowables], but is separated
(vyatireka) [from them].35 This is the state referred to as “the pure Self,”
“the Formless,” and “pure Consciousness” in the Saiddhāntika scriptures.
It is taught with reference to those who know the Deity36 solely as [being]
all-transcendent; so [Utpaladeva] indicates [in his Vivṛti].
34 Inferring the antecedent of tatra from the previous line: jñānollāsa iti ajñāna-vigamād
advaya-bodha-prasaraṇād ullāsa[ḥ], “ ‘the blossoming of insight is a blossoming that
results from the departure of ignorance and the [concomitant] spread of nondual aware-
ness.”
35 Exactly how to construe the grammar of this sentence is not clear to me, nor am I con-
vinced that the text is secure.
36 For Abhinava’s use of tadrūpa to mean the Deity, see, e.g., TĀ 1.173c–174b.
via free access
162 wallis
Again we are presented with the notion of a vyatireka-turyātīta, though here

its transcendental as opposed to its gnostic character is stressed. Since this state
is associated with the transcendent deity (sarvottīrṇaika-tadrūpa, a kenning for
Anāśrita-śiva, usually considered tattva 1), we are inclined to conclude that,
as suggested previously, Abhinava wishes us to understand that the turyātīta
which accesses the śakti-tattva is in fact the superior one. Perhaps this is not
stated explicitly because it is a subversive view relative to the mainstream of
the tradition.
Here Abhinava criticizes the exclusively transcendentalist view of the Said-
dhāntikas, since as a nondualist Śākta he wishes to privilege the Kaula “imma-
nentist” view. Likewise, his disciple Kṣemarāja argues that the defining feature
of their lineage (trikādi-darśana, meaning the view of the Trika-Krama synthe-
sis uniquely expounded by Abhinavagupta) is the view that the Divine Self is
simultaneously transcendent of all and immanent in all.37 Now, since the tran-
scendental turyātīta is identified with the teachings of the sāmānya-śāstra (i.e.,
right-current Śaiva Siddhānta), we would expect the yogic-cum-alchemical
turyātīta to be identified with the viśeṣa-śāstra (i.e., left-current Śākta streams),
and just such identification is found by Torella (2002, xxxiv note 52) on the
same page of the ĪPVV as the passage just quoted, but to me the text is not
so clear. Apparently glossing viśeṣa-darśaneṣu in Utpala’s lost Vivṛti, Abhinava
says (KSTS vol. 65, 331): “The specialized views are those which predominantly
teach the nonduality of Power [and the Power-holder, i.e. Consciousness]; they
are superior (viśeṣa) because they teach the Power-characteristics of the Self,
[and thereby] lead one to the direct experience [of that]” (viśeṣa-darśaneṣu
iti śāktādvaita-pradhāneṣu, viśeṣaḥ śakti-lakṣaṇa ātmano darśyate sākṣātkāraṃ
nīyate yeṣu). There is no explicit correlation with the second, yogic turyātīta,
but perhaps we can assume it because the power element of the self (vai-
bhavādi) is repeatedly mentioned in the passages we have examined in con-
nection with that turyātīta.
37 From chapter 8 of his Pratyabhijñāhṛdaya: “The Tāntrikas (= Saiddhāntikas and other rit-
ualists) hold that the reality of the Self is all-transcending. Those attached to traditions
such as the Kula say it is all-embodying. Those who hold [our] viewpoint of the Trika and
[the Krama] hold that it is [simultaneously] all-transcending and all-embodying” (viśvot-
tīrṇam ātma-tattvam iti tāntrikāḥ, viśva-mayam iti kulādy-āmnāya-niviṣṭāḥ, viśvottīrṇaṃ
viśvamayaṃ ca iti trikādi-darśana-vidaḥ).
via free access
5 Summary of the Data
Now to summarize what we have learned about these two states of immer-
sion.38 The primary distinction Abhinava wishes to make is that the first turya
→ turyātīta constitutes an “active” entry/immersion into one’s essence-nature
(one’s satya-svarūpa or ahaṃbhāva, which is cidānanda and prakāśa-ghana),
and the second turya → turyātīta denotes a “passive” process by which one is
entered; i.e., that in which the various layers of selfhood are permeated by that
ultimate I-sense (note that it is passive only in the grammatical sense, for the
yogic method involves considerably more work). This distinction is summed up
as āveśyāveśaka-bhāvaḥ (p. 331). Since the yogic process is a gradual one, differ-
ences are noted between turya and turyātīta, whereas the gnostic turyātīta is
said to be identical in nature (though presumably not in degree) to the turya
that precedes it (turyatā … turyātīta-samatā).
Table 6.1 summarizes the data in tabular form (items in parentheses are only
implicit in the primary source text).
6 Problems of Interpretation
To close, I will briefly discuss some interesting ambiguities and difficulties of

these sections of the ĪPV and ĪPVV, apart from those already discussed above
and in the notes. The first problem is the one raised (but not discussed) by
Torella (2002, xxxv note 52), who tells us that Abhinavagupta’s inclinations
lie towards the second turyātīta, the yogic-alchemical one. This seems evident
from the passages we have translated here, but on the other hand it is surpris-
ing, because in general Abhinava tends to privilege the gnostic over the yogic
in his work. Perhaps the context of a clearly liberation-focused yoga outlined
in ĪPK III.2.19–20 allowed him to endorse a term that otherwise so commonly
denotes the pursuit of siddhi and bhoga in the Śaiva literature.
But we have another problem with Torella’s hypothesis, one briefly discussed
already above: the implication in these passages is that the gnostic turyātīta—
(1) in Table 6.1—reaches the śiva-tattva, while (2) explicitly only reaches the
śakti-tattva. My reasons for concluding that turyātīta (1) reaches the śiva-tattva
are as follows: since the term turyātīta refers to the highest liberation where
only the absolute “I” remains, and in the tattva-system that attainment is explic-
itly identified with the top two tattvas, turyātīta (1) must correspond to the śiva-
38 sā iyaṃ dvayī api daśā samāveśo, KSTS vol. 65, 328.
via free access
164 wallis
table 6.1 Two types of Turyātīta in the ĪPVV
Turyātīta (1) Turyātīta (2)
gnostic (realization) yogic (practice)

sudden gradual
āveśaka āveśya
knowables transcended knowables transmuted
(objectivity transcended) objectivity dissolved
vyatireka (avyatireka)
(śiva-tattva) śakti-tattva
(īśvara- or sadāśiva-tattva in turyaa)
qualities of the Self: qualities of the Self:
vyāpaka, nitya vaibhava, aiśvarya
(emptiness) fullness
sāmānya-darśanas (viśeṣa-darśanas)
a Note that the ĪPK itself (III.2.20) lists the levels of turya-attainment as those of the Vi-
jñānākalas (= level of Mahāmāyā, just outside the śuddhādhvan and therefore not yet lib-
erated), Mantras (= śuddhavidyā-tattva, lowest level of liberation), and Īśvara; but Abhinava
takes mantreśa in that verse to refer to the Mantra-lords of īśvara-tattva, then reads -īśa a sec-
ond time, taking it to refer Lord Sadāśiva (Torella 2002, 208 note 33). Here he is making the
correlation correspond to what is found in Trika scripture, for the Mālinī-vijayottara-tantra
teaches that “the Mantras, Mantreśas and Mantramaheśvaras occupy the Fourth state” and
“Śakti and Śiva exist in the state Beyond the Fourth” (2.28c–29b, trans. Vasudeva 2004, 209–
210).
tattva, which also matches its transcendent nature (the Śiva of tattva 1 is often
called Anāśrita-śiva39). And we have seen that turyātīta (2) aligns with the śakti-
tattva. Now, it doesn’t seem altogether likely that Abhinavagupta would favor
an attainment that reached only to tattva 2. However, he may well have held the
view explicitly articulated by his successor Kṣemarāja, i.e. that there is no real-
ity to hierarchy with regard to Śiva/Śakti, they being two aspects of one reality,
one or the other of them being more prominent in the liberated experience at
any given moment (see Vasudeva 2004). In other words, Abhinava may have
held the view that since Śiva and Śakti are in fact inseparable except heuristi-
cally, to attain one is to attain the other. Or perhaps, as already posited above,
we are to understand that Abhinava is allusively suggesting that in this system,
39 E.g., in Pratyabhijñā-hṛdaya chapters 4 and 5.
via free access
Śakti is in fact tattva 1. Such a hidden doctrine would be in keeping with the
rest of his esoteric theology, which constitutes a Trika doctrine with a Krama
core.
The other problem of interpretation that arises in connection with these
passages is a philosophical one, not yet to my knowledge addressed in the
secondary literature. This issue centers on the question of who is the agent
of the various verbs used here, most especially √viś. For example, when it is
said that “there is an entry into one’s true nature,” who or what is the agent
of that entry, since the satya-svarūpa that is entered is the only real source
of both awareness and agency? This problem is not glaringly obvious because
of the nature of the Sanskrit language, in which passive voice is so common,
and nouns or pronouns denoting the agent can be omitted entirely, the verb
conjugation itself communicating a generic unspecified third-person agent.
When a first-person active verb is used, for example when Kṣemarāja glosses
namas or naumi as samāviśāmi (in his commentaries on stotras40), the prob-
lem is made evident: what exactly is the “I” that enters? Obviously it cannot be
mind, body, etc. (since they have no agency of their own), nor can it be cit, for
it would make no sense to say that consciousness, which is undivided, enters
into itself.
Two possible solutions occur. First, that what enters from the mind, etc., into
the Self is a kind of “locus of subjective identity” or ahaṃbhāva. But Abhinava
seems to use this term in the passages we have considered to mean the true I-
sense, the Self-that-is-awareness. The second and more likely solution is that
this language of entry is purely metaphorical, derived from the phenomenol-
ogy of the experience it denotes (i.e., what it feels like to have that experience),
and that in actuality there is no entry at any time: the true “I” simply realizes
itself or wakes up to itself, clearly apprehending itself. It may be the case that
Abhinava addresses this problem of agency in relation to √viś verbs somewhere
in his vast body of work, and I simply have not yet come across it.
7 Postscript: Samāveśa and Turya in Tantrāloka 10
Abhinavagupta also discusses turya and turyātīta briefly in Tantrāloka 10.264–

284. Though this passage is not directly a commentary on the ĪPK (being instead
a commentary on the phenomenological categories of Mālinī-vijayottara-tan-
tra, chapter 2 and passim), it clearly has in mind Utpaladeva’s phrase tat-śakti-
40 See Stainton 2013 or Stainton 2019.
via free access
166 wallis
samāveśa in the Vṛtti on ĪPK III.2.12 and very likely also alludes to his lost
Vivṛti on the same.41 For Tantrāloka 10.265 informs us that turya is in fact śakti-
samāveśa!42 This seems to imply that turyātīta is śiva-samāveśa, which could
match the ĪPV account but not the ĪPVV; again, perhaps Abhinava changed his
view between the two, the latter having been composed after the Tantrāloka.
Abhinava goes on (in TĀ 10) to subdivide this śakti-samāveśa into four stages,
corresponding to the four epistemological categories of knower, knowing,
known and the autonomous pure awareness (para-pramātṛ) which is the
source and ground of the previous three (Vasudeva 2004, 230).43 The four
stages are as follows. When immersion into this parā saṃvit is only proximate
(tat-samāveśa-naikaṭyāt), object-consciousness is dominant (TĀ 10.270d–271a).
When there is contact44 with this immersion, the process or faculties of know-
ing are dominant (tat-samāveśoparāgān mānatvam, 270cd). In full identifica-
tion with this immersion, the state of the knower becomes clear (tat-samāveśa-
tādātmye mātṛtvaṃ bhavati sphuṭam, 270ab). Beyond this, in the state of the
so-called pure awareness in which all three previous categories have perfectly
fused, in which there is perception but no sense of a separate perceiver or per-
ceived, the Light of Consciousness is self-manifest and we cannot speak of an
immersion, except metaphorically (10.269). This helps us understand that the
movement into the transcendent Fourth state can indeed happen in stages,
the first three of which (prameya, pramāṇa, pramātṛ) are an expression of
divine grace (trayaṃ tat tad-anugrahāt, 270b). We can infer that speaking of
grace is meaningless in the fourth stage, where there is no duality (though the
ĪPV and ĪPVV accounts tell us that there are still saṃskāras of duality at this
stage).
41 This supposition receives support from the fact that Abhinava explicitly comments on
ĪPK III.2.15–17 further on in the same chapter (viz., TĀ 10).
42 pūrṇatāgamanaunmukhyam audāsīnyāt paricyutiḥ | tat turyam ucyate śakti-samāveśo hy
asau mataḥ ||, paraphrased by Vasudeva (2004, 229) as “In the fourth state … knowable
entities appear as awareness on the verge of reaching plentitude because [the] indiffer-
ence [that characterized the third state of deep sleep] is abating. Abhinavagupta further
identifies this state as an immersion into Śakti.”
43 For the fourth category of pure awareness, see TĀ 10.269: pramātṛtā svatantratva-rūpā
seyaṃ prakāśate | saṃvit turīya-rūpaivaṃ prakāśātmā svayaṃ ca sā ||, and Jayaratha ad
loc.: parā saṃvid evam aṃśa-trayottīrṇā … svātantrya-mayī para-pramātṛtā … sā hi para-
pramātṛ-rūpā śuddhā saṃvit svayam prakāśate na tu paśyāmītyādi-vikalpollekha-bhūmiḥ.
That there are four epistemological categories, not three, is due to the influence of the
Krama, in which Kālī is identified with the fourth.
44 Uparāga seems a strange word to use here; its commonest use is “eclipse” or “affliction”—
perhaps we should emend to upayoga.
via free access
Abhinava then defines turyātīta at 10.278: “that [state] whose beautiful

nature is full and undivided, overflowing with joy, is called Beyond the Fourth;
that alone is the supreme state” ( yat tu pūrṇānavacchinna-vapur ānanda-nir-
bharam turyātītaṃ tu tat prāhus tad eva paramaṃ padam45). His discussion
goes on, but it reaches beyond the purview of the present work.
8 Conclusion
The present paper does not, of course, entirely solve the complex textual and
exegetical problems we discovered in the comparison of these passages of the
ĪPV and the ĪPVV, but perhaps it contributes towards an understanding of their
significance for the study of Tantric Śaiva theology. Provisionally, I propose that
we see Abhinavagupta changing, developing, and nuancing his view in the time
between the ĪPV and the ĪPVV (with Tantrāloka falling between the two). If
I am reading the texts correctly, the ĪPV features a simpler model of a gnos-
tic transcendentalist turya succeeded by a “immanentist” turyātīta (the latter
being marked by the transcendent element’s pervasion of all that was previ-
ously transcended), while the ĪPVV proposes two distinct versions of both turya
and turyātīta, gnostic and yogic respectively (giving us four categories in total),
where the yogic is to be preferred despite being more gradual because in it the
saṃskāras of dualistic experience are finally dissolved.
Abbreviations
conj. em. conjectural emendation

em. emendation
ĪPK Īśvarapratyabhijñākārikā of Utpaladeva
ĪPV Īśvarapratyabhijñāvimarśinī of Abhinavagupta
ĪPVV Īśvarapratyabhijñāvivṛtivimarśinī of Abhinavagupta
KSTS Kashmir Series of Texts and Studies
TĀ Tantrāloka of Abhinavagupta
TĀV Tantrālokaviveka of Jayaratha
45 Or we could take pūrṇānavacchinna-vapur as a bahuvrīhi meaning “in which every beau-

tiful embodiment is full and unlimited,” the implication being that every part is now
experienced as containing the whole (akhaṇḍa-maṇḍalākāram).
via free access
168 wallis
References
Primary Sources46
Īśvarapratyabhijñākārikā. Raffaele Torella, ed. The Īśvarapratyabhijñākārikā of Utpala-
deva with the Author’s Vṛtti: Critical Edition and Annotated Translation. Rome: IsMEO,
1994.
Īśvarapratyabhijñākārikāvṛtti. See Īśvarapratyabhijñākārikā.
Īśvarapratyabhijñāvimarśinī. Madhusudan Kaul Shāstrī, ed. The Īśvarapratyabhijñā of
Utpaladeva with the Vimarśinī by Abhinavagupta. KSTS, nos. 22 and 33. Pune: Aryab-
hushan Press, 1918–1921.
Īśvarapratyabhijñāvimarśinī. K.A.S. Iyer and K.C. Pandey, eds. Īśvara-pratyabhijñā-
vimarśinī of Abhinavagupta: Doctrine of Divine Recognition. Vol. 2. Reprint, Delhi:
Motilal Banarsidass, 1986 [1950].
Īśvarapratyabhijñāvivṛtivimarśinī. Madhusudan Kaul Shāstrī, ed. The Īśvarapratyabhi-
jñā Vivṛtivimarśinī by Abhinavagupta. 3 volumes. KSTS, nos. 60, 62 and 65. Pune:
Aryabhushan Press, 1938, 1941, and 1943.
Tantrāloka. Paṇḍit Mukund Rām Shāstrī (vol. 1) and Paṇḍit Madhusudan Kaul Shāstrī
(vols. 2–12), eds. The Tantrāloka of Abhinava Gupta, With Commentary by Rajānaka
Jayaratha. 12 vols. KSTS, nos. 23, 38, 30, 36, 35, 29, 41, 47, 59, 52, 57 and 58. Allahabad:
The Indian Press Ltd. (vols. 1, 5–6); Bombay: Nirnaya Sagar Press (vols. 7, 10–12),
Tatva-Vivechaka Press (vols. 3–4, 8–9), and Shri Venkateshvar Steam Press (vol. 2),
1921–1938.
Pratyabhijñāhṛdaya. Paṇḍit Mukund Rām Shāstrī, ed. KSTS, no. 3. Pune: Aryabhushan
Press, 1911.
Secondary Sources
Hellwig, Oliver. 2009. Wörterbuch der mittelalterlichen indischen Alchemie. Supple-
ments to eJIM, vol. 2 Eelde: Barkhuis & University of Groningen.
Roşu, Arion. 1982. “Yoga et alchimie.” Zeitschrift der Deutschen Morgenländischen
Gesellschaft (bd. 132): 363–379.
Sanderson, Alexis. 2007. “The Śaiva Exegesis of Kashmir.” In Mélanges tantriques à la
mémoire d’Hélène Brunner / Tantric Studies in Memory of Hélène Brunner, edited
by Dominic Goodall and André Padoux, 231–442. Collection Indologie, no. 106.
Pondicherry: Institut Français d’Indologie/École française d’Extrême-Orient.
Sella, Andrea. YouTube video, 6 April 2011. https://www.youtube.com/watch?v=gKxCw
889qck&feature=youtu.be
46 The text editions cited are available online at the Muktabodha Indological Research Insti-
tute’s Digital Library: http://www.muktabodha.org/digital_library.htm
via free access
Stainton, Hamsa. 2013. “Poetry and Prayer: Stotras in the Religious and Literary His-
tory of Kashmir.” Ph.D. dissertation, Columbia University. https://doi.org/10.7916/
D8VT209V.
Stainton, Hamsa. 2019. Poetry as Prayer in the Sanskrit Hymns of Kashmir. Oxford:
Torella, Raffaele. 2002 [1994]. The Īśvarapratyabhijñākārikā of Utpaladeva with the
Author’s Vṛtti: critical edition and annotated translation. 2nd ed. Delhi: Motilal Banar-
sidass.
Vasudeva, Somadeva. 2004. The Yoga of the Mālinīvijayottaratantra. Critical Edition,
Translation and Notes. Pondicherry: Institut Français de Pondichéry/École française
Wallace, Vesna A. and B. Alan Wallace, trans. 1997. A Guide to the Bodhisattva Way of
Life: Bodhicāryāvatāra. Ithaca, NY: Snow Lion Publications.
Wallis, Christopher. 2014. “To Enter, to be Entered, to Merge: The Role of Religious
Experience in the Traditions of Tantric Shaivism.” Ph.D. dissertation, University of
California at Berkeley. https://escholarship.org/uc/item/4p0666qj
Watson, Alex. 2006. The Self’s Awareness of Itself: Bhaṭṭa Rāmakaṇṭha's Arguments
against the Buddhist Doctrine of No-Self. Wien: De Nobili.
via free access
chapter 7
On Vāgīśvarakīrti’s Influence in Kashmir and

among the Khmer
1 Introduction
The aim of this article is to point out the far-reaching influence of an East
Indian tantric Buddhist scholar, Vāgīśvarakīrti ( floruit early 11th c.). In the first
part I will show that his views were considered important enough to be con-
tested sometime before 1057ce, probably still during his scholarly activity, in
Kashmir. In the second part I wish to propose the hypothesis that although
unnamed, he is a master alluded to with great reverence on the Sap Bāk inscrip-
tion from the Khmer Empire, dated 1067ce.*
“Our” Vāgīśvarakīrti should not be confused with his namesake, a Newar
scholar from Pharping, whence his epithet Pham mthiṅ ba (for what we can
gather about this person, see Lo Bue 1997, 643–652). Nor should we confuse
him with a rather nebulous person, whose name is re-Sanskritised as *Suvā-
gīśvarakīrti, author of a number of small works extant in Tibetan translation.
Lastly, there is no good reason to assume that he is the same as a commen-
tator of Daṇḍin’s Kāvyādarśa; this person’s name is often re-Sanskritised from
the Tibetan as *Vāgīśvara, but it is more likely that his name was Vācaspati or
Vāgīśa.
The writings of Vāgīśvarakīrti are fairly well known to scholars of esoteric
Buddhism. A significant portion of his oeuvre survives in the original Sanskrit.
* I have already discussed these two subjects in two separate lectures. The first subject was
tackled at the First Manuscripta Buddhica Workshop in Procida, Italy in May 2011, where I
received some extremely valuable feedback, especially from Professor Harunaga Isaacson,
with whom I also had the opportunity to briefly study the passage in question in Kathmandu
some months earlier. The second problem I have merely alluded to in a lecture at Kyoto Uni-
versity in February 2015; Professor Arlo Griffiths commented on an early draft of my notes
and kindly encouraged me to publish my findings (e-mail, December 4, 2014). A later draft
was read by Dr. Johannes Schneider, whose suggestions greatly improved some of my state-
ments and saved me from a couple of blunders. To all involved, I offer my sincerest thanks.
All remaining errors are mine.
© péter-dániel szántó, 2020 | doi:10.1163/9789004432802_009

via free access
on vāgīśvarakīrti’s influence 171
While some of the attributions in the Tibetan Canon are disputed, the follow-
ing major works may be assigned to him with confidence.
The Mṛtyuvañcanopadeśa is a learned anthology of rites to cheat death
once its signs have been perceived. This work, which survives in at least four
manuscripts,1 has been admirably dealt with recently by Johannes Schneider
(2010). His German translation supersedes Michael Walter’s earlier English
translation (2000). As Schneider conjectures (2010, 23), the Tibetan transla-
tion must have been completed in 1042/3ce, since this is the only time the
two scholars mentioned in the translators’ colophon, *Adhīśa (better known as
*Atīśa or *Atiśa) Dīpaṃkaraśrījñāna and Rin chen bzaṅ po, spent time together
at Tho liṅ. This date is also Schneider’s terminus ante quem for the text.
The Saṃkṣiptābhiṣekavidhi is a succinct initiation manual for the Guhya-
samāja system, which also contains a fascinating polemic passage (Onians
2002, 279–289). At present we may access only one manuscript; this has been
edited by Munenobu Sakurai, but is in dire need of being revisited. Another
witness, now probably in Lhasa, is mentioned in the catalogue KCDS, p. 139.
The Tattvaratnāvaloka (henceforth TaRaA), a short treatise in twenty-one
verses, and a largely prose auto-commentary thereof, the Tattvaratnāvalokavi-
varaṇa (henceforth TaRaAVi), are usually mentioned in the same breath and
are indeed transmitted together in the only known manuscript. These texts
have been edited by (presumably) Banarsi Lal. The Tibetan translations were
undertaken by ’Gos Lhas btsas (although only Tōh. 1890 / Ōta. 2754 is actually
signed by him), whose activity falls in the middle of the 11th century (Davidson
2005, 139).
The *Saptāṅga (henceforth SaA), another treatise, this time in mixed verse
and prose, is the only major work of Vāgīśvarakīrti which appears to be lost
in the original. One of its most important verses survives in quotation (Isaac-
son and Sferra 2014, 171, 271, passim). The Tibetan translation is the work of the
same ’Gos Lhas btsas.
I shall not discuss here Vāgīśvarakīrti’s other, minor works, or the fact that
some of his major works are present more than once in various recensions of
the Tibetan Canon, some of them even annotated.
We shall have the opportunity to study some of Vāgīśvarakīrti’s ideas later
on, although I cannot hope—nor do I propose—to be exhaustive here. The
two most important features to keep in mind for the time being are these: that
for the author, the most important cycle of tantric Buddhist teachings is the
1 A fragment missed by Schneider can be found in NAK 1–1697/vi. bauddhatantra 60 = NGMPP

B 31/19. Nearly twenty-six verses survive on this single leaf (1.63c–1.89b), which may in fact be
the earliest attestation of the original (Schneider’s earliest manuscript is from 1290ce).
via free access
172 szántó
Guhyasamāja, and that he was a proponent of the view that full initiation con-
sists of four consecrations, culminating in the so-called caturthābhiṣeka.
There is very little hard evidence for a prosopography of Vāgīśvarakīrti. All
modern authorities conjecture that he lived during the 10–11th centuries and
all seem to accept the statements of Tibetan hagiographies, namely that he was
active in Vikramaśīla in the rather nebulous capacity of “door-keeper.” The pri-
mary source for this information is Tāranātha’s famous historiography, the Rgya
gar chos ’byuṅ, which dedicates a long passage to Vāgīśvarakīrti, presenting him
as a scholar, an accomplished tantric practitioner, a miracle worker, and a pious
founder (Chimpa and Chattopadhyaya 1970 [2004], 296–299).
2 Vāgīśvarakīrti in Kashmir
The source I shall be using for starting the discussion here is found in an
unpublished and little-studied commentary of the Mañjuśrīnāmasaṃgīti, the
Gūḍhapadā of one Advayavajra (incorrectly spelt as Advayavakra or perhaps
Advayacakra in the colophon), which survives in a single manuscript. This is
a voluminous text, occupying 180 densely written palm-leaf folios; according
to the colophon, it measures 4,000 granthas. It has not been translated into
Tibetan. This Advayavajra is very likely not the same as the famous Advayava-
jra or Maitreyanātha (some good reasons against this identification are listed
in Isaacson and Sferra 2014, 74–75).
Since it was not translated into Tibetan and it survives in a single manuscript,
the Gūḍhapadā may nowadays be perceived as obscure. However, it was not an
unknown work, at least not in the 12th century. Raviśrījñāna, one of the most
famous exegetes of the Mañjuśrīnāmasaṃgīti, mentions it as one of the main
sources he relied upon. Oddly, the particular verse where he does just this is
not found in the published Amṛtakaṇikā, because the Sarnath editor did not
have access to or ignored the tradition that transmits it. One such witness is
Royal Asiatic Society London, Ms. Hodgson 35 (the so-called Vanaratna codex;
see Isaacson 2008), folio 40r1–2.2 There can be little doubt that this closing
verse is authorial: the Tibetan translation, although in a garbled way, mirrors
2 I give here a diplomatic transcript of the śārdūlavikrīḍita stanza in question: śrīvajrāṅkita-

pāṇigarbhabhagavallokeśaṭīkārthayā (°ānvayā?) ślāghyā gūḍhapadāśritādbhutabṛhatkāśmī-
rapañjīsakhā (?) | nānātantrarahasyavibhramavatī nānopadeśāśritā prītā ṭippaṇikā raviśriya
iyaṃ prīṇātu cetaḥ satāṃ || The first pāda alludes to three exegetes of the Kālacakra tradi-
tion, known as the bodhisattva commentators: Vajrapāṇi, Vajragarbha, and Puṇḍarīka (here
Bhagavallokeśa for metrical reasons).
via free access
it (D 96a5–6, P 115a8–115b1), and Vibhūticandra’s sub-commentary, the Amṛ-

takaṇikoddyota, has some of its words in lemmata (Ed., p. 216, ll. 13–16).3
Vibhūticandra’s sub-commentary gives us the upper limit for Raviśrījñāna.
The former first came to Tibet in 1204ce (Stearns 1996). We also know that the
Amṛtakaṇikā is one of Raviśrījñāna’s earlier works, because he refers to it in his
Guṇabharaṇī (Sferra 2000, 100). He cannot be much earlier than the late 12th
century, since one of his masters, Dharmākaraśānti, lived during the reign of
Gopāla (pace Sferra 2000, 47–48), in whose court he was a royal preceptor. This
Gopāla must be Gopāla IV (r. ca. 1128–1143 ce), because the other Gopālas are
too early for Raviśrījñāna to mention all three bodhisattva commentators of the
Kālacakra system (see here, note 2). Therefore the Gūḍhapadā must have been
in existence around these dates. The debate we are about to examine seems to
suggest a mid-11th-century environment, but we cannot be entirely certain.
The Gūḍhapadā has the following commentary to Mañjuśrīnāmasaṃgīti
8.41ab on folio 128r4–128v4. The text is first given in diplomatic transcript. The
line is here marked in bold; the manuscript highlights it in red. An akṣara added
in the upper margin in the second hand is marked thus ⟨ ⟩. A deleted anusvāra
is marked with ⟨| |⟩. I have added the line numbers in square brackets. I split the
passage into seven units—marked (a) to (g)—for the sake of easier reference.
(a) tasyā ekakṣaṇamahāprajñaḥ sarvadharmāvabodhadhṛg iti | araṇa-

śrī nāmataḥ | tatra ekaś cāsau kṣaṇaś ca eka[5]kṣaṇaḥ | mahāś cāsau prā-
jñaś ca mahāprājñaḥ sarvadharmavivekātmakas tataś cāyam arthaḥ | (b)
ekakṣaṇaś caturānandaikamūrttitvāt | sahajasaṃbodhikṣaṇaḥ | sa ca [6]
turyātītalakṣaṇaṃ | (c) tathā coktaṃ | śrīmaṅkāsmīrīyai sūkṣmāvarttabha-
ṭṭapādaiḥ | kasmīreṣu kathā nāsti caturthasyaiti niskṛpaḥ | asti deṣāntare
tāvac caturthaṃ samyaksevida [7] iti | (d) kutaḥ yad vāgīśvarakīrttino-
ccyate | dambhaulibījaśrutaddhautasuddhaḥ pāthojabhūtāṃkurabhūtaḥ
puṣṭi || turīyam asya paripākam eti | sphuṭaṃ caturthaṃ binduso [128v] pi
gūḍham iti | (e) ātmīyābodhāhaṃkāratvāt nāstipakṣyā bhilaṣatam iccha-
nti | evamādikṛta⟨|ṃ|⟩sya śrīmaratnavajrāṃghrim āha | bhrāntyā yatra
pravaramatayaḥ kīrttisāntyādayo [2] pi | idaṃ caturthālokakārakā pūrva-
deśapaṇḍitaiḥ | vāṭyamātraṃ na vijñātā tadgranthatodgataṃ | (f) asmadī-
yagurupādamatam āha | turyātītam avācyaṃ tu kṣaṇam ekam arū [3]pa-
3 The Sarnath edition lets us down once again here. For the pratīkas of pāda b we have this
printed: … dapadam āśritā | śrīnāropādapañjikāsandhī(m adhītya) | The only manuscript of
the Amṛtakaṇikoddyota I can consult for the time being is Tokyo University Library no. 18 (old
no. 348), last folio, l. 1 and this fairly clearly reads ślāghyā gūḍhapadām āśritā | śrīnāropāda-
pañjikāsaṅgī (?) |
via free access
174 szántó
kaṃ | sahajasaṃbodhinā me jñātājñeyau tu nirvitār iti | (g) evaṃ eka-

kṣaṇaiva kṣaṇena sampadyate mahāprajñatayā yathoktasarvadharmāva-
bodhanatayā ekakṣa[4]ṇamahāprājñaḥ sarvadharmāvabodhas tad dhā-
rayatīti ekakṣaṇamahāprā⟨ jña⟩ḥ sarvadharmāvabodhadhṛk |
The beginning (tasyā eka° up to °vivekātmakas) and end (evaṃ eka° up to

°āvabodhadhṛk) of the text—i.e. (a) and (g)—are of no concern to us here.
These sentences should nevertheless act as cautionary devices that the passage
is quite corrupt. Most of the content here is in any case an almost word-for-
word copy of Vilāsavajra’s Nāmamantrārthāvalokinī.4
Unit (b) explains the compound ekakṣaṇa (“a singular moment”) in terms
of post-Hevajra yoginītantra doctrine. According to this teaching, during ini-
tiation one experiences in sexual union the four “blisses” (ānanda), which are
linked to four “moments” (kṣaṇa). This experience is then cultivated in medi-
tation leading to enlightenment, a state sometimes referred to as “the innate”
(sahaja). The passage should therefore be interpreted something like this: “[It is
a] singular moment, because it embodies in a unitary manner the four blisses.
[And this is nothing else but] the moment of innate complete awakening,
which, in turn, is beyond the fourth [state of consciousness—the four being
wake, sleep, deep sleep, and the fourth].”
Unit (c) seems to take a turn. Apparently, we have a continuation of the pre-
vious topic, but in fact here we turn to the problem of initiation. The passage
is doubtless a quotation, introduced by tathā coktam. The author of the quoted
passage is referred to—once we emend the text slightly, śrīmatkāśmīrīyaiḥ for
śrīmaṅkāsmīrīyai5 and understand that the plural shows respect—as “the glo-
rious Kashmirian, the venerable Sūkṣmāvartabhaṭṭa.” The anuṣṭubh verse fol-
lowing may be restored thus:
kaśmīreṣu kathā nāsti caturthasyeti †niskṛpaḥ† |

asti deśāntare tāvac caturthaṃ samyak[ ]sevitam ||
4 For the sake of comparison, I give here Vilāsavajra’s text from Ms. Cambridge University
Library Add. 1708, folio 81v5–7: ekakṣaṇamahāprājñaḥ sarvadharmāvabodhadhṛg iti | ekaś
cāsau kṣaṇaś ca ekakṣaṇaḥ | mahāṃś cāsau prājñaś ca mahāprājñaḥ (em., °prājña Ms. post
corr., °prājñaś ca Ms. ante corr.) sarvadharmavivekātmakaḥ (Ms. post corr., °ātmākaḥ Ms. ante
corr.) | tataś cāyam arthaḥ sampadyate | ekenaiva kṣaṇena mahāprājñatayā yathoktasarva-
dharmāvabodhanatayā | ekakṣaṇamahāprājñasarvadharmāvabodhas (em., °āvabodha|s Ms.)
tad dhārayatīti ekakṣaṇamahāprājñasarvadharmāvabodhadhṛk ||
5 The first error ṅ for t in ligature with k is a simple orthographic error. The second is a banal
feature of East Indian scribal habits: sibilants are freely interchangeable. The third is a cus-
tomary loss of visarga before sibilants, which may reflect pronunciation.
via free access
The only real intervention here is sevitam for sevida[ḥ], which is nonsense,
whereas sevitam is both grammatical and yields good meaning. The emenda-
tion kaśmīreṣu for kasmīreṣu is rather banal, but I remain undecided whether
this should be emended further to kāśmīreṣu (“among Kashmiris” rather than
“in Kashmir”); the plural is otherwise often used with both toponyms and
inhabitants of a region. As we shall see, there must be an iti hiding in caturtha-
syaiti. Spelling ai for e is not uncommon in East Indian manuscripts, although
of course it is incorrect. We can safely dismiss the idea that the reading is cor-
rect and what we have here is the present third person singular of the root i, “to
go”; in that case we would expect an accusative, probably of an abstract noun,
but nothing of the sort can be conjectured. The corrupt niskṛpaḥ (or perhaps
niṣkṛpaḥ) unfortunately masks a crucial word. We shall return to it forthwith.
The name of the author and the first third of the verse can be traced in
Tibetan. The work in question is the *Caturthasadbhāvopadeśa, attributed in
the colophon (D 159b3–4) to “the great Kashmiri master … *Ratnavajra” (kha
che’i slob dpon chen po … dpal rin chen rdo rje). We see the same name in unit
(e). Sūkṣmāvarttabhaṭṭa and Ratnavajra are one and the same, as the last verse
of the work reveals (D 159b3):
| dpal kha che’i slob dpon rin chen rdo rje daṅ |
| mtshan gźan phra bar rtogs pa’i dpal źes bya’i |
It is immediately apparent that there is something wrong with the Tibetan

translation: the number of syllables per quarter is out of balance—the ini-
tial dpal could be superfluous—and the daṅ seems just a little bit off, unless
we think it is justified inasmuch as it links a name and an alternative name.
However, the underlying meaning is clear enough: the author identifies him-
self as a Kashmiri master called Ratnavajra, also known as *Sūkṣmāvartaśrī, as
āvartta—with a slight stretch—can be reconstructed from rtogs pa, whereas śrī
is perhaps a metrical equivalent of bhaṭṭa or an unusual rendering into Tibetan.
Whatever doubts we may have about having correctly traced the author, they
are dispelled once we look at the first line of the treatise (D 156b2–3):
| kha che ba la bźi pa’i gtam | | sṅar yaṅ yod par ma thos (em., thas) la |
| da ltar yod pa mthoṅ na yaṅ | | bźi pa rñed pa ma yin no |
Translated somewhat loosely, this means:
Previously, in Kashmir (or: among Kashmiris) not a [single] word was

heard of [this] “Fourth” [Initiation]; although nowadays we see it prac-
via free access
176 szántó
ticed, [I will show that in fact] the “Fourth” cannot be found [to have
authoritative sanction].
It is now clear that this is what we have echoed in our Sanskrit verse: kaśmīreṣu
kathā nāsti caturthasyeti. The Tibetan ba (or pa?) seems to suggest that we
should understand “among Kashmiris.”
But now we have a stylistic problem. Apparently, the introduction to the
quotation applies only up to iti. The corrupt †niskṛpaḥ† must be some sort
of dismissive statement, since the second line seems to contradict Ratnavajra:
true, Kashmir has not heard of the Fourth Initiation (caturtham), but this does
not mean anything, since it does exist (asti) and is correctly practiced (samyak
sevitam) in another land or other lands (deśāntare). One tentative solution for
niskṛpaḥ may thus be niṣkṛtam (“disregarded”, “dismissed”), but I must confess
that I still regard this as nothing more than a diagnostic conjecture.
It is conceivable that we are wrong to emend śrīmaṅkāsmīrīyai to śrīmatkā-
śmīrīyaiḥ and that we must boldly conjecture bṛhatkāśmīrīye. As we have seen
in the verse given here in note 3, Raviśrījñāna knew of such a work, since he
lists it as one of his sources of inspiration. Moreover, this is not the only time
he refers to it: in the body of Amṛtakaṇikā we find at least one reference (Ed.,
p. 25, l. 18), which is mirrored in the Amṛtakaṇikoddyota (Ed., p. 197, last line).
If I interpret Vibhūticandra’s commentary correctly, in the text given here in
note 4, he attributes this work to the famous Nāropāda. If Vibhūticandra is cor-
rect, we cannot take Ratnavajra/Sūkṣmāvarttabhaṭṭa to be the author of the
Bṛhatkāśmīrapañjikā, since there is nothing to suggest that he might be the
same as Nāropāda. (Of course, Vibhūticandra could be wrong, but then the
number of variables becomes too great to contemplate meaningfully.) If we fol-
low this idea, the introduction would give the source for the entire verse—i.e.
the Great Kashmiri Commentary—, in which Ratnavajra’s idea is embedded
as a prima facie view. However, the stylistic problem remains: it would be very
unnatural to give a title and then a name, which is not that of the author but
that of an interlocutor in it. Perhaps it is not impossible that Ratnavajra’s other
name was once a gloss meant to elucidate the ownership of the point to be
refuted, and that this gloss made it into the main text at some point during
transmission.
Unit (d) is somewhat easier to tackle. The quotation reinforces the existence
of the Fourth Initiation by quoting Vāgīśvarakīrti. The verse is very corrupt in
the form given here, but fortunately we have access to the source, which is
the TaRaA, verse 17. The TaRaAVi does not offer any explanation for the verse;
indeed, it shrouds it in secrecy, stating that the meaning should be obtained
from the oral teachings of a qualified guru (Ed., p. 100, l. 20: dambholītyādi|
via free access
etat sadgurūpadeśato jñeyam|). Vāgīśvarakīrti seems to have changed his mind

about this when he wrote his other major work, the SaA. There, he offers a very
long explanation (D 199b7–201a3, P 235b4–237a2), alluding to the fact that mis-
understandings of his position prompted him to do so. This verse is also quoted
by Raviśrījñāna in the Amṛtakaṇikā (Ed., p. 76, ll. 22–23).
Thus, with the help of the TaRaA manuscript (Ms.), the edition (Ed.), the
Tibetan translation (D and P agree in all the readings), the lemmata in the SaA
(SaA), and Raviśrījñāna’s testimony as edited (AKa) with the readings of the
unused Vanaratna codex (V, folio 29r10), we may restore unit (d) thus:
kutaḥ | yad Vāgīśvarakīrtinocyate–
dambholibījasrutidhautaśuddha-
pāthojabhūtāṅkurabhūtapuṣṭi |
turīyaśasyaṃ paripākam eti
sphuṭaṃ caturthaṃ viduṣo ’pi gūḍham || iti||6
The verse does not immediately lend itself to understanding, but in the present
context, as a cited authority, it must have been understood along these lines:
Cleansed by the oozing of the seed (i.e. semen) from the thunderbolt (i.e.
the officiant’s penis) growing as a sprout born from a purified lotus (i.e.
the consecrated vulva of the consort), the crop that is the fourth [state of
consciousness] comes to full bloom; [although] the Fourth [Initiation] is
manifest, it is hidden even from the wise.
The coded language expresses what happens in the three higher initiations
(guhyābhiṣeka, prajñājñānābhiṣeka, caturthābhiṣeka), the first two of which
are of a sexual nature. The SaA makes it clear that the first stage, where the seed
from the thunderbolt oozes and cleanses, alludes to the guhyābhiṣeka, where
in practice the officiating master copulates with a consort and the ejaculates
are placed in the mouth of the blindfolded initiand. Via this rite, the mind of
the initiand, which is similar to a field, is purified. The second stage, where a
6 Variants are provided only for the verse, naturally: dambholi°] Ms. Ed. AKa V, rdo rje’i D, rdo rje
SA • °sruti°] corr., °śruti° Ms. Ed. V, °śruta° AKa, ’bab pa D SA • °dhauta°] Ms. Ed. AKa, °dhota°
V, dri med dga’ (!?) D, dag pa SA • °śuddha°] Ms. Ed. V, °śuddhaṃ AKa, dag pa’i D, dag pa SA
• °pāthoja°] Ed. V, °pāthauja° Ms., °pāthojña° AKa, chu skyes D SA • °bhūta°] Ms. Ed. AKa V,
’byung D, ’byung ba SA • turīyaśasyaṃ] AKa V, tutīyaśasyaṃ Ms., tṛtīyaśasyaṃ Ed., bźi pa’i ’bru
D, bźi pa’i ’bras bu SA •paripākam eti] Ed. AKa V, paripākam eta Ms., yoṅs gsal smin ’gyur ba’i D,
yoṅs gsal smin par ’gyur ba yi SA • viduṣo] Ms. Ed. AKa V, mkhas pa rnams la D, no lemma in SA
via free access
178 szántó
sprout is said to be born and made to grow in a pure lotus, alludes to the prajñā-
jñānābhiṣeka, where it is the initiand who copulates with the/another consort.
The sprout represents his fledgling wisdom. The logic of the allegory demands
that it is in the caturthābhiṣeka where this sprout comes to full bloom, that is
to say, reaches the highest state, here called “the Fourth.” This is somewhat con-
fusing, since just above the author of the Gūḍhapadā seems to advocate a state
“beyond the fourth” as the highest. Also note that sphuṭam could be construed
in a different way, either as an adjective to paripākam or an adverb to eti.
Unit (e) is certainly the most challenging part of the passage. The first sen-
tence seems to condemn those who do not recognise (nāstipaḳsyā[ḥ]) the exis-
tence of the Fourth Initiation on account of their stupidity and selfishness.
What exactly the aim of their desire (doctrinal or otherwise) is, I cannot tell. We
do not fare any better with the next statement. The compound evamādikṛtasya
escapes me completely. The next statement again seems to introduce a quota-
tion, but the accusative case is puzzling. It is perhaps better to emend to a nomi-
native while also fixing the first honorific. Thus we get: śrīmadratnavajrāṃghrir
āha. The other honorific, aṃghri (lit. foot), is somewhat unusual. It doubtless
stands for the more common °pāda and it may do so for metrical reasons: if
we observe the metrical pattern of Ratnavajrāṃghrir āha, the words would fit
the last seven syllables of a Mandākrāntā line. However, in this case we must
give up on śrīmad°, since here we would require laghu-guru (short-long) and
not guru-guru. This idea must be considered, for what follows is indeed a per-
fect Mandākrāntā line (with some minor corrections applied): bhrāntā yatra
pravaramatayaḥ kīrtiśāntyādayo ’pi, i.e. “in which respect even those of the
choicest intellect, such as Kīrti and Śānti, are deluded.” We may safely assume
that yatra refers to the matter at hand, i.e. the veracity of the Fourth Initia-
tion, and we can reasonably suppose that these are Ratnavajra’s words, paying
respect to his opponents, but claiming that they are wrong. Kīrti no doubt refers
to Vāgīśvarakīrti, whereas Śānti is most likely shorthand for another great intel-
lect of early 11th-century Eastern India, Ratnākaraśānti. While Vāgīśvarakīrti’s
position on the Fourth Initiation is known, we know very little as to what Ratnā-
karaśānti thought of the matter.7
Although we seem to understand this particular passage, there is a slight
problem: it is not from the *Caturthasadbhāvopadeśa and it is not from any
other work attributed to Ratnavajra in the Tibetan Canon. The next sentence,
a corrupt anuṣṭubh, on the other hand can be traced in the *Caturthasadbhā-
7 The most likely place for addressing this would have been his Hevajra initiation manual (to
which he refers as the Hevajrābhyudayamaṇḍalopāyikā; see Muktāvalī, Ed., p. 215, ll. 15–16),
but this text is most unfortunately lost.
via free access
vopadeśa (D 157a7–157b1). I repeat the lines here along with the Tibetan trans-
lation, because they are quite crucial:
idaṃ caturthālokakārakā pūrvadeśapaṇḍitaiḥ |

vāṭyamātraṃ na vijñātā tadgranthatodgataṃ |
| bźi pa sgron (em., sgrol) ma mdzad pa yi |

| śar phyogs kyi ni mkhas pa yis |
| gtam tsam yaṅ ni mi śes so |
| de yi chos la bdag gis rtogs |
The pronoun idaṃ is puzzling and I am tempted to disregard it completely,

all the more so since the Tibetan does not mirror it. Alternatively, one could
emend it to ittham and consider it as an introductory statement. Emending
the Tibetan sgrol to sgron is warranted by the Sanskrit °āloka°, and this makes
good sense, because this is an attested title to which we will return shortly. If
this is a title, it would make sense to emend °kārakā to °kārakaiḥ following the
Tibetan mdzad pa yi understood in the sense of “author” to qualify the com-
pound in the next pāda, or to °kāraka° joining the line into one compound.
Unfortunately, this creates a serious metrical problem. I do not have a solution
at present, except suggesting that we do indeed need a word meaning “author”
for the broken part.
It is apparent that in the second verse quarter we are one syllable short. The
most straightforward solution is probably to read °deśīya° for °deśa°. Assuming
that the cruces hide a word meaning “author,” as I suggest, we would thus get
a line in harmony with the Tibetan expressing a logical subject: caturthāloka-
†kārakā†pūrvadeśīyapaṇḍitaiḥ, i.e. “the learned one from the Eastern lands,
the author of the Caturthāloka.” The plural is, again, for showing—here,
mock—respect. This pundit is none other than Vāgīśvarakīrti, since Caturthālo-
ka is an alternative title for his TaRaA. In fact, this alternative title seems to be
the one preferred by the author himself, because this is the way he refers to his
previous work in the SaA. He does so thrice (D 190b3, 199a2, 199b7 & P 225a7,
234b4, 235b4–5), calling the treatise Bźi pa snaṅ ba, and citing lemmata from
TaRaA, verses 8, 11, and 17 respectively.8
8 For the sake of clarity, here are the passages: 1) de dag la yaṅ Bźi pa snaṅ ba ñid las | ñid
mtshuṅs lha mos ’khyud daṅ źes bya ba la sogs (D, la sogs pa P) tshigs su bcad pa bdun […]
citing svābhāṅgaśleṣi; 2) ji ltar Bźi pa snaṅ ba las sku ni nam mkha’ daṅ mtshuṅs źes bya bas
[…] citing gaganasamaśarīraṃ; 3) da ni Bźi pa snaṅ ba las smos pa’i rdo rje źes bya ba’i tshigs
su bcad pas […] citing the beginning of the verse discussed above, dambholi°.
via free access
180 szántó
As for the next pāda, this is probably where we should find the logical pred-
icate, but for this we must emend vijñātā to vijñātam. For the first word, the
Tibetan seems to suggest vākya° (gtam). No doubt, vāṭya° (“fried barley,” “fig”
(?)) seems to be the more difficult reading, but I am not familiar with this
idiomatic expression, if indeed it is one. Let us accept the emendations and
read vākyamātraṃ na vijñātaṃ together with the first line, i.e. “not a single word
was understood by the learned one from the Eastern lands, the author of the
Caturthāloka [understand: Vāgīśvarakīrti].”
What exactly Vāgīśvarakīrti did not understand or how exactly it became
clear to the Kashmirian author that the Easterner is a fool is a mystery, since the
last pāda seems, at least to me, beyond redemption and the Tibetan is not very
clear either. One may conjecture with great hesitation after having harmonised
the two something like tadgranthato mayodgatam, in the sense “[this] I have
understood from his work.” But this would create a metrical issue, since a ra-
gaṇa is not permitted for syllables 2–4. A more serious intervention would yield
tadgranthāt/tadgranthe tan mayodgatam. Another problem is that udgatam is
not entirely elegant and does not make very good sense. However, we are prob-
ably not too far from a genuine solution. What Ratnavajra seems to be saying
then is that he had read Vāgīśvarakīrti’s work, the Caturthāloka, and realised
that the Eastern scholar is an ignoramus.
The meaning of unit (f) is somewhat clearer, but it is impossible to say who
the speaker is. Somebody—deutero-Advayavajra? Ratnavajra?—states the
viewpoint of his guru: the introductory clause asmadīyagurupādamatam āha
does not need any correction. The first line of the verse seems to be fine as
is, although the meaning is somewhat obscure: turyātītam avācyaṃ tu kṣaṇam
ekam arūpakam, i.e. “as for that ineffable [state of consciousness called]
‘beyond the fourth,’ it is a singular, formless moment.” The second line is prob-
lematic. Pāda c is unmetrical: both the second and the third syllables are short,
while the fifth, sixth, and seventh syllables form a ra-gaṇa, thus a ra-vipulā,
but there is no caesura after the fourth. Pāda d with the closing particle should
probably read jñātājñeyau tu nirvṛtāv iti or jñātṛjñeyau, etc., or jñānajñeyau,
etc. The first of the possible solutions, that is to say, leaving the compound
as transmitted, contains an irregular dvandva, perhaps inspired by the well-
known mātāpitṛ-. In spite of these serious irregularities, I have no reason to
think that the first half of the line is corrupt, especially since we already had
the collocation sahajasaṃbodhikṣaṇaḥ in unit (b). The line therefore proba-
bly meant, “due to complete awakening of the innate, for me [the duality] of
knower [or: knowledge] and objects of knowledge has been extinguished.” Let
us attribute the irregularities to the ecstatic power that must have overcome
the nebulous guru at the moment of enlightenment.
via free access
We will now return to Ratnavajra and his scathing attack on Vāgīśvarakīrti.

Much light would be thrown on this matter if the Sanskrit original of the
*Caturthasadbhāvopadeśa were to become available, or better said, accessible.
The work is reported to exist in the now famous—and notoriously inac-
cessible—Lhasa birch-bark manuscript. The existence of this unique codex,
once penned in Kashmir and until very recently kept on display at the Tibet
Museum, was first reported by Kazuhiro Kawasaki (2004). The Japanese scholar
was allowed to consult the table of contents and the colophon on the last folio
of this remarkable document. We know from his report that the manuscript is
a composite codex containing twenty-seven works. The sixth item listed in the
table of contents (Kawasaki 2004, 51/904) is none other than Caturthasadbhā-
vopadeśaḥ Śrīratnavajrakṛtaḥ.
A decade later, Kazuo Kanō (2014) provided a new reading of the colophon
and converted the date precisely. According to his calculation, the [Kali] year 29
corresponds to 1057ce, since the ruling king, Anantadeva, is also mentioned,
and his reign falls between 1028 and 1063ce (Kanō 2014, 62–63). This date is
extremely important, because it gives us a rather early upper limit for the cre-
ation of the Caturthasadbhāvopadeśa and thus Ratnavajra’s mature activity,
besides confirming his reading of Vāgīśvarakīrti’s work.
Ratnavajra was an authority not to be taken lightly. His influence and fame
can be gathered from other sources as well. For example, *Prajñāśrīgupta, in
his commentary on the Mahāmudrātilaka, now extant only in Tibetan (Tōh.
1201), eulogises him thus (D 154b3–4): “the great scholar, who has obtained ini-
tiation and the oral teachings, the king over overlords of kings, the best of gurus,
the teacher from Kashmir, the glorious Ratnavajra.” *Prajñāśrīgupta also claims
that Ratnavajra stood at the end of an uninterrupted lineage of masters stem-
ming from the famous Indrabhūti. He also quotes from at least one of his lost
works, probably simply called *Utpannakrama (D 155b3–4). In another work,
the *Ratnamañjarī (Tōh. 1217), a commentary on the *Tattvapradīpa, he again
eulogises Ratnavajra as his master (D 325a2).
We find several more references to Ratnavajra in the work of *Sambhogava-
jra, probably *Prajñāśrīgupta’s disciple or junior contemporary. In his *Raha-
syanalinī (Tōh. 1418), he claims the same spiritual descent, while adding the
equally prestigious name of *Avadhūtīpāda along the way. He is only slightly
less loquacious when it comes to praising the famous master (D 250b2–3):
“he who has crossed to the other shore of all Vedas, who has obtained the
accomplishment of the mantra, the Kashmiri master, the glorious Ratnavajra.”
*Sambhogavajra quotes the same passage (partially) from the lost *Utpanna-
krama (D 240b1) and a verse from another work entitled *Adhyātmapadma
(D 244b6). He too emphasises that Ratnavajra was a Kashmiri (D 246b5).
via free access
182 szántó
Ratnavajra’s name was probably exploited to make other works more pres-
tigious. One such pseudepigraphical case in my view is an initiation manual of
the Sarvabuddhasamāyogaḍākinījālaśaṃvara system, the *Sarvasattvasukho-
dayā (Tōh. 1679). Contrary to the opinion voiced in the *Caturthasadbhāvo-
padeśa, this text does teach the Fourth Initiation (D 294a1–4). A less likely
scenario is that Ratnavajra eventually changed his mind and accepted what
was, to him, initially a controversial abhiṣeka.
Ratnavajra, too, seems to have been particularly proud of his Kashmiri her-
itage. In his *Akṣobhyavajrasādhana (Tōh. 1884) he proudly claims to have sat
at the feet of Northern gurus (D 162b5), perhaps in yet another attempt to disso-
ciate himself from innovations in the East.9 This work—which, in spite of the
title, is a manual in the tradition of Jñānapāda—is most likely genuinely his,
since the qualifications of the practitioner mention having received all initia-
tions but the Fourth (D 144b1–2).
Returning to his Caturthasadbhāvopadeśa, here too Ratnavajra voices what
is perhaps a challenge to all non-Kashmiris. The third verse of the text says
(D 156b3–4):
| yul phyogs gźan na la la dag | gal te skal ldan gyur pa dṅos |

| yod pas gter ni mi nub ces | | bdag blo ṅes par dbugs dbyuṅ ṅo |
The verse is not entirely clear, but it probably means something along the fol-
lowing lines:
Should there be any fortunate ones (i.e. worthy Buddhists) in other coun-
tries, let them [come forward and] put my mind at ease so that [this]
treasure would not fall into oblivion.
The “treasure” (gter, which more correctly would be rin chen) he refers to is in
my view the Fourth Initiation, cf. an oft-quoted and later scripturalised pāda
from the Prajñopāyaviniścayasiddhi (3.38d): vācaiva dadyād abhiṣekaratnam.
Perhaps less likely is the possibility that he uses *ratna as a shorthand for
TaRaA, in which case we have here yet another reference to Vāgīśvarakīrti’s
work. In my reading, Ratnavajra was being ironic. He would not have found any
9 This is stated in the first quarter of the penultimate concluding verse: | de ltar byaṅ phyogs lam
pa’i mkhas pa’i źabs bsten nas | […] Byaṅ phyogs lam pa mirrors Sanskrit *uttarapatha/uttarā-
patha.
via free access
comfort in his being convinced of the genuineness of the Fourth Initiation. It

follows that the “treasure” he seems to be worried about should be read kākvā,
in a mocking tone.
Although Ratnavajra seems to refer to the Caturthāloka/TaRaA exclusively,
it is very likely that his reading of Vāgīśvarakīrti was much broader. After all,
verse 17 from that work does not reveal much about his position concerning
the Fourth Initiation. I think that there is a possibility that this is what he refers
to in the verse (D 157b1) immediately following the one we had attested in the
Gūḍhapadā:
| ’on kyaṅ gti mug bsal ba daṅ | | som ñi rab tu źi ba daṅ |

| phyin ci log ni bsal ba’i phyir | | mkhas pas bstan bcos rgyas par mdzad |
Now, in order to dispel confusion, to put doubts at ease, and to clear up

distortions, [this] learned man expanded his treatise.
In what follows, Ratnavajra echoes in his pūrvapakṣas many points brought up

by the TaRaAVi. If this conjecture holds, it would seem that some time elapsed
between Vāgīśvarakīrti’s writing his verses and the auto-commentary. It is also
not impossible that the SaA was written partly as an answer to Ratnavajra’s crit-
icism. However, for all this to be determined one would need access to the San-
skrit original of the Caturthasadbhāvopadeśa. Until that time, we must leave
the matter to rest.
In spite of the numerous corruptions we have encountered in the Gūḍha-
padā passage, let us recapitulate what may be gathered with certainty. At
some point before 1057ce, an influential Kashmiri master called Ratnavajra or
Sūkṣmāvarttabhaṭṭa published a scornful refutation of the idea that there is a
Fourth Initiation (caturthābhiṣeka). This position he seems to have attributed
first and foremost to “Eastern” scholars, singling out Vāgīśvarakīrti and his
Tattvaratnāvaloka or Caturthāloka. In spite of the vitriolic dismissal, it is evi-
dent that Vāgīśvarakīrti was too important to be ignored. The debate between
the two remained well known, as some of its salvos were preserved perhaps
already in the now lost Bṛhatkāśmīrapañjikā, and certainly in the Gūḍhapadā.
The latter was still an influential work before the end of the 12th century, since
the famous Raviśrījñāna used it as a source.
One could extrapolate a more general point from this debate, namely that
scholarly communication between East India (at this point in time under
Pāla sovereignty) and Kashmir was vigorous. Kashmiris seem to have been a
little more orthodox in their views, but innovations—for which the hotbed
was undoubtedly Pāla East India—did filter in. This exchange between the
via free access
184 szántó
two regions was certainly not a unique event. The famous satirist Kṣeme-
ndra describes Bengali students flocking to Kashmir around the same time
(Deśopadeśa, chapter 6 in 45 verses). He is even more scornful of Easterners,
describing them as illiterate, dim-witted, pretentious, sanctimonious, vulgar,
and ugly. I find it very likely that the poet’s bigoted diatribe met with many a
sympathetic ear in his local audience.
3 Vāgīśvarakīrti among the Khmer
The document known as the Sab Bāk inscription (K. 1158), consisting of fifteen
(idiosyncratic) Sanskrit verses and a few lines in Old Khmer, was discovered at
an unconfirmed location in what is now Nakhon Ratchasima province, Thai-
land. It is one of the most important sources testifying to the presence of the
Guhyasamāja system in Southeast Asia. The text of the inscription was first
edited by Prapanvidya (1990). Since then it has been noted and discussed in
a fairly large number of publications, the latest of which, at least to my knowl-
edge, is by Conti (2014). This article also features a new translation by Tadeusz
Skorupski.
The inscription, dated Śaka year 988, 7th of the waxing fortnight of Tapasya
(Friday, 23rd of February 1067ce), records the words of one Vraḥ Dhanus, given
the title ācārya in the Khmer portion, a devotee of the Guhyasamāja. The text
first eulogises three teachers of Vraḥ Dhanus, all indicated by toponymic sur-
names: the venerables of Cuṅ Vis, Campaka, and Dharaṇīndrapura. It then
describes the erection of an unspecified number of icons beginning with an
image of the Buddha. The Khmer portion refers to previous installations as well.
The document is fascinating and important, but still requires substantial
work. I cannot touch on these topics here; instead, I wish to concentrate on
a particular aspect, the identity of a master referred to in verses 3 and 4. The
most reliable edition of the text is that of Estève (2009, 557–558), which I have
checked against an estampage of the original (ÉFEO n. 1497); here I quote only
the relevant couplet:
śrīsamāje parā yasya bhak(t)iḥ śraddhā ca nirmmalā

tasya dāsasya dāso haṃ bhaveyaṃ sarvajanmasu || [3]
ity ājñā paramaguroḥ śrutvā stutyā namaskṛtā
anukathyā mayā bhaktyā śrīsamājan name sadā || [4]
These are the translations that have been published thus far. Prapanvidya (1990,
12) interpreted the text as follows:
via free access
In all my births, may I become the slave of that slave, who has great devo-
tion to and impeccable faith in the Śrīsamāja. ‘Having saluted with praise,
I must hear and repeat the teacher’s command devotedly:’ thus is the com-
mand of the supreme teacher. I constantly pay my obeisance to Śrīsamāja.
Estève’s French translation is perhaps a bit more accurate (2009, 561):
«Ce Śrī Samāja pour lequel j’ai une dévotion suprême et une foi pure,
que j’en sois le serviteur du serviteur dans toutes mes existences». Après
avoir entendu l’ ājñā du paramaguru, je dois lui rendre hommage avec des
louanges puis, avec dévotion, le répéter. Je rends hommage perpétuelle-
ment à Śrī Samāja.
Finally, Skorupski’s translation published in Conti (2014, 393) is quite similar to

that of Prapanvidya:
In all my existences, may I become a servant of the servant who has

supreme devotion and stainless faith in the glorious Samāja. Having thus
heard the command of the supreme guru, I respect it with praises, (and)
having repeated it with devotion, I always pay homage to the glorious
Samāja.
The first hemistich of the quoted portion is in my view an echo of Vāgīśvarakī-

rti’s words. The penultimate closing verse of his TaRaAVi is this:
śrīsamāje parā yasya bhaktir niṣṭhā ca nirmalā |

tasya vāgīśvarasyeyaṃ kṛtir vimatināśinī ||10
This is the work to dispel all opposed opinions of Vāgīśvara[kīrti], whose

dedication to the glorious [Guhya]samāja is supreme and whose devotion
is without blemish.
Moreover, this is the closing verse of his SaA (D 202b7–203a1, P 238b5–6):
| dpal ldan gsaṅ ba ’dus pa las | | dri med dad mchog mthar phyin pas |
| ṅag gi dbaṅ phyug de yis ’di | | byas pas blo ṅan (D, ldan P) ’jig gyur cig
10 There are two silent emendations by Sarnath editors; the Ms. reads bhaktiniṣṭhā and
vimatināsanī.
via free access
186 szántó
I find it almost certain that this is a clumsy Tibetan rendering of the same
verse. The small emendation las to la would fix the first pāda, whereas the sec-
ond would better read *dad mchog mthar phyin dri ma med, were it not the case
that mthar phyin pa (*niṣṭhāgata?) is a very bad choice for niṣṭhā. The third and
fourth pāda may be seen as a very loose rendering: “may this work written by
Vāgīśvarakīrti destroy wicked views!” However, here too the choice of words is
inapposite, as we would expect lta and not blo for mati.
I am unaware of any other texts that would use the same phrasing; it can be
said therefore that this is a ‘signature verse’ of Vāgīśvarakīrti.
The only difference between the hemistich of the Khmer inscription and
the verse transmitted in India, Nepal, and Tibet is a mere synonym, śraddhā
for niṣṭhā. Perhaps the Khmer author thought that the latter is a lesser-known
word for “devotion” and decided to replace it with a metrically and gender-wise
unproblematic, more current noun. Otherwise the echo is unmistakable.
In light of this discovery, the second line of the Khmer inscription would
mean that someone is wishing to become a devotee (dāsa) of that devoted one
(tasya dāsasya) in all subsequent rebirths—i.e., these are the words of a stu-
dent of Vāgīśvarakīrti.
In the first pāda of verse 4, these words seem to be described as “the com-
mand (ājñā) of the paramaguru.” The syntax here is quite incorrect, since
ājñā should also be construed with śrutvā, and we would therefore require an
accusative; however, namaskṛtā and anukathyā forces the author to leave it in
the nominative. This is not the only bizarre usage of Sanskrit in the text. To
note only the most glaring examples, in verse 6 we have a double sandhi, saiva
for sa eva, in verse 15 the enclitic cet stands at the beginning of the line, and
so on. Another oddity is that such pious exclamations are not called ājñā, but
praṇidhāna or praṇidhi, even in the tantric context (e.g. Hevajratantra 2.8.6–7
and prose before).
It should also be noted that the first quarter of verse 4 is a na-vipulā, with
the minor blemish that the fourth syllable is not long. This perhaps suggests
that the composer found it important to include the term paramaguru. This
does not only mean “supreme teacher,” but has a more technical sense, namely
one’s spiritual grandfather, i.e. one’s guru’s guru. I could not find Buddhist texts
that clearly have this usage (a possible exception is the Balinese Buddhaveda,
p. 75); however, it is prevalent in Śaiva literature. In Abhinavagupta’s Tantrasāra
(Ed., p. 156) we find the sequence guru, paramaguru, and parameṣṭhin, followed
by the collective pūrvācāryāḥ. In the glosses to the Īśānaśivagurudevapaddhati
(Ed., vol. 3, ad 13.58ab) we find this list stretched for five generations: guru,
paramaguru, parameṣṭhiguru, pūjyaguru, and mahāpūjyaguru. We sometimes
(e.g. Puraścaryārṇava, Ed., vol. 1, 3.578cd-579ab) find parāparaguru between
parama and parameṣṭhin.
via free access
Keeping in mind the hypothesis that paramaguru has a technical meaning

here and does not simply mean “supreme teacher,” two further likely hypothe-
ses emerge. The command, which Vraḥ Dhanus relates, is spoken by one of his
teachers, namely his guru’s guru, in which case Vāgīśvarakīrti was the guru of
this person, that is to say, Vraḥ Dhanus’parameṣṭhiguru or parāparaguru. How-
ever, given the loose phrasing seen elsewhere in the document, it might just be
the case that Vraḥ Dhanus is referring only to the first half of the verse, in which
case he is acknowledging Vāgīśvarakīrti as his paramaguru. The latter is gram-
matically speaking less likely, but more likely if we think about the number of
spiritual generations elapsed between Vāgīśvarakīrti, active in the earlier half
of the 11th century, and Vraḥ Dhanus, whose commissioned inscription is dated
1067ce. Whichever scenario we accept as the most plausible, what seems to
be certain is that by this date Vāgīśvarakīrti’s person and Guhyasamāja-related
teachings were known in the Khmer lands.
If Vāgīśvarakīrti was known in the Khmer lands, then so was Jñānapāda’s
school of thought. Although more attention should be dedicated to this mat-
ter, I feel confident in saying that Vāgīśvarakīrti was a follower of the Jñānapāda
exegesis. The strongest evidence for this claim is his brief overview of Jñāna-
pāda’s Samantabhadra or Caturaṅgasādhana as well as at least one unattrib-
uted quotation from the same work in his SaA (D 202a2 ff., P 238a1 ff.).
4 Epilogue
During the editorial process, I came across two further possible testimonies for
Vāgīśvarakīrti’s influence. I am grateful to the editors for allowing me the oppor-
tunity to include them here.
The first comes from what is now Burma/Myanmar, an inscription dated
1442ce celebrating the construction of a monastery and related donations by a
district governor called Thirizeyathu (Taw Sein Ko 1899, 37–47). The document
records a large number of books as part of the governor’s generous gift, includ-
ing the famous couple Mahākālacakka and Mahākālacakka ṭīkā, long accepted
as evidence for the presence of the Kālacakra system in Burma. The work listed
immediately before this (p. 47) is called the Mṛtyuvañcana. While this could
refer to any death-cheating ritual manual, the most celebrated such work was
that of Vāgīśvarakīrti. There is therefore a strong possibility that he was still
read in Burma as late as the first half of the 15th century. Naturally, I do not
claim this as conclusive evidence.
Another possible allusion to Vāgīśvarakīrti, or at the very least the debate
he was famous for, comes from certain recensions of Saraha’s Dohākośa. In a
via free access
188 szántó
verse criticising fellow Buddhists, the famous siddha (or the author posing as
the siddha) says that “[without having realised reality,] some are immersed in
explaining the Fourth.”
I read the verse in the following three sources: a) Niedersächsische Staats-
und Universitätsbibliothek Göttingen manuscript Xc 14/16, folio 2r: ko vi vak-
khāṇa caüṭṭhihi laggo; b) Tokyo University Library manuscript 517, folio 17v: ke
vi vakkhāṇa caüṭṭhahiṃ laggaü; and c) NAK 1–1607 = NGMPP A 49/18, recto of
83rd leaf: koï vakkhaṇa caüṭhṭhihi laggo. The verse with this line was not avail-
able for Śāstrī (1916, 85), Shahidullah (1928, 129—although the Tibetan given
here does mirror our reading very closely: kha cig bźi ba’i don ’chad pa la źugs)
or Sāṃkṛtyāyan (1957, 4); Bagchi (1938, 16) reconstructs the verse, and his read-
ing is followed by Jackson (2004, 58), who also suggests that one possibility for
interpretation is the “Fourth Initiation”; see also Schaeffer 2005, 136.
The single-folio NAK fragment is a part of the Sahajāmnāyapañjikā commen-
tary, a very precious (and unfortunately very corrupt) witness, as here Bagchi’s
manuscript has a lacuna. The relevant passage in Tibetan can be found in
D 184r7–185v2. Interestingly, here the target is identified as a monk, but caü-
ṭhṭhihi is glossed either as a cardinal number, in which case the four schools
are meant (Vaibhāṣika, Sautrāntika, Yogācāra, Madhyamaka), or as an ordinal
number, in which case the meaning is innate bliss (sahaja). This commentator
would therefore not think that the object of the (fruitless) explanation is the
Fourth Initiation.
It is of course possible, and perhaps even likely, that Saraha here refers to
the fourth and ultimate state of consciousness or the fourth and highest bliss
(ānanda), but it is not out of the question that what he has in mind is the (or a)
debate regarding the Fourth Initiation. However, again, this is hardly conclusive
evidence.
Abbreviations
Aka Amṛtakaṇikā.
D Tibetan text in the Canon’s Derge (Sde dge) print. Numbers accord-
ing to Tōh.
Ed. edition
KCDS 中国藏学研究中心收藏的梵文贝叶经（缩微胶卷）目录 [Zhong-
guo zangxue yanjiu zhongxin shouzangde fanwen beiye jing (Suowei
jiaojuan) mulu] Kruṅ go’i bod kyi śes rig źib ’jug lte gnas su ñar ba’i
ta la’i lo ma’i bstan bcos (sbyin śog ’dril ma’i par) kyi dkar chag mdor
gsal, n.a.
via free access
Ms. manuscript
NAK National Archives, Kathmandu
NGMPP Nepal German Manuscript Preservation Project
Ōta. Daisetz T. Suzuki (ed.), The Tibetan Tripitaka, Peking Edition—kept at
the Library of the Otani University, Kyoto—Reprinted under the Super-
vision of the Otani University, Kyoto. Catalogue & Index, Tokyo, 1962.
P Tibetan text in the Canon’s Peking (Pe ciṅ) print. Numbers accord-
ing to Ōta.
TaRaA Tattvaratnāvaloka.
TaRaAVi Tattvaratnāvalokavivaraṇa.
Tōh. Hakuju Ui, Munetada Suzuki, Yenshô Kanakura, Tôkan Tada (eds.),
A Complete Catalogue of the Tibetan Buddhist Canons (Bkaḥ-ḥgyur
and Bstan-ḥgyur), Tôhoku Imperial University, Sendai, 1934.
SaA Saptāṅga.
References
Primary Sources
Amṛtakaṇikā
(AKa). Banarsi Lal, ed. Āryamañjuśrīnāmasaṁgīti with Amṛtakaṇikā-ṭippaṇī by Bhikṣu
Raviśrījñāna and Amṛtakaṇikodyota-nibandha of Vibhūticandra. Bibliotheca Indo-
Tibetica XXX. Sarnath: Central Institute of Higher Tibetan Studies, 1994.
(V) Royal Asiatic Society, London, Ms. Hodgson no. 35. Palm leaf, 62 folios (of which
this work once occupied 1 to 40r), East Indian script (Maithilī/Bengali), undated
(ca. mid-15th century).
(D) Tōh. 1395, translated by *Maṇiśrījñāna, Ñi ma’i dbaṅ po’i ’od zer, Chag Chos rje dpal.
(P) Ōta. 2111, ditto.
Amṛtakaṇikoddyota
(Ed.) see Amṛtakaṇikā.
(Ms.) Tokyo University Library, no. 18 (old no. 348). Palm leaf, 90 folios (once complete
in 91 folios), Old Newar script, dated Nepālasamvat 420 = 1300ce. (same as Ed.’s Ms.
Ka)
Īśānaśivagurudevapaddhati
T. Gaṇapati Śāstrī, ed. Īśānaśivagurudevapaddhati, vol. 3. Trivandrum Sanskrit Series
77. Trivandrum, 1922.
via free access
190 szántó
Gūḍhapadā
(Ms.) Royal Asiatic Society, London, Ms. Hodgson no. 34. Palm leaf, 180 folios, hook-
topped Old Newar script, undated (ca. 12–13th century?).
Caturthasadbhāvopadeśa
(Ms.) not accessible.
(D) Tōh. 2475, translated by *Vidyābhadra and Tshul khrims bkra śis.
(P) Ø.
Tattvaratnāvaloka
(TaRaA) [a.k.a. Caturthāloka]. (Ms.) see Tattvaratnāvalokavivaraṇa.
(Ed.) see Tattvaratnāvalokavivaraṇa.
(D) Tōh. 1889, no translator given.
(P) Ōta. 2753, no translator given.
Tattvaratnāvalokavivaraṇa
(TaRaAVi). (Ms.) NAK 5–252 = NGMPP A 915/4.
(Ed.) Dhīḥ 21, 129–149, reprint (used here) Bauddhalaghugranthasamgraha (A Collec-
tion of Minor Buddhist Texts). Rare Buddhist Texts Series 14. Sarnath: Central Insti-
tute of Higher Tibetan Studies, 1997: 81–103.
(D) Tōh. 1890, translated by ’Gos Lhas btsas.
(P1) Ōta. 2754, ditto.
(P2) Ōta. 4793, no translator given.
Tantrasāra
Mukund Rām Shāstrī, ed. The Tantrasāra of Abhinava Gupta. Kashmir Series of Texts
and Studies no. XVII. Bombay: Nirnaya-Sagar Press, 1918.
Deśopadeśa
E.V.V. Rāghavāchārya and D.G. Padhye, eds. Minor Works of Kṣemendra. Hyderabad:
Osmania University, 1961: 273–306.
Dohākośa
See Śāstrī 1916, Shahidullah 1928, Bagchi 1938, Sāṃkṛtyāyan 1957, and Jackson 2004.
Puraścaryārṇava
Muralidhara Jhā, ed. Puraścaryārṇavaḥ, vol. 1. Benares: Prabhakari & Co., 1901.
via free access
Prajñopāyaviniścayasiddhi
Samdhong Rinpoche and Vrajvallabh Dwivedi., eds. Guhyādi-aṣṭasiddhi-saṅgraha.
Rare Buddhist Text Series 1. Sarnath: Central Institute of Higher Tibetan Studies,
1987: 63–87.
Buddhaveda
Sylvain Lévi, ed. Sanskrit Texts from Bali. Baroda: Oriental Institute, 1933: 72–84.
Muktāvalī
See Hevajratantra.
Saṃkṣiptābhiṣekavidhi
(Ed.) see Sakurai 1996, 412–421.
(Ms.) NAK 3–387 = NGMPP A 1156/24 and retake B 24/15.
(D) Tōh. 1887, translated by Sumatīkīrti, Klog skya Gźon nu ’bar, Mar pa Chos kyi dbaṅ
phyug.
Saptāṅga
(SaA). (D) Tōh. 1888, translated by ’Gos Lhas btsas.
(P) Ōta. 2752, ditto
Sahajāmnāyapañjikā
(Ed.) see Bagchi 1938.
(D) Tōh. 2256, translated by Vairocanavajra of Kosala.
Hevajratantra
Ram Shankar Tripathi and Thakur Sain Negi, eds. Hevajratantram with Muktāvalī Pañ-
jikā of Mahāpaṇḍitācārya Ratnākaraśānti. Bibliotheca Indo-Tibetica Series XLVIII.
Sarnath: Central Institute of Higher Tibetan Studies, 2001.
Secondary Literature
Bagchi, Prabodh Chandra. 1938. Dohākoṣa (Apabhraṁśa Texts of the Sahajayāna
School). Part I. (Texts and Commentaries). Calcutta: Metropolitan Printing and Pub-
lishing House.
Chimpa, Lama and Alaka Chattopadhyaya. [1970] 2004. Tāranātha’s History of Bud-
dhism in India. Reprint, Delhi: Motilal Banarsidass.
Conti, Pia. 2014. “Tantric Buddhism at Prasat Hin Phimai: A New Reading of Its Icono-
graphic Message.” In Before Siam. Essays in Art and Archaeology, edited by Nicolas
Revire and Stephen A. Murphy, 374–394. Bangkok: River Books and The Siam Society.
via free access
192 szántó
Davidson, Ronald. 2005. Tibetan Renaissance. Tantric Buddhism in the Rebirth of Tibetan
Culture. New York: Columbia University Press.
Estève, Julia. 2009. Étude critique des phénomèmes de syncrétisme religieux dans le Cam-
bodge angkorien. Ph.D. thesis, École Pratique des Hautes Études.
Isaacson, Harunaga. 2008. “Himalayan Encounter: The Teaching Lineage of the Marmo-
padeśa (Studies in the Vanaratna Codex 1).” Manuscript Cultures 1: 2–6.
Isaacson, Harunaga and Francesco Sferra. 2014. The Sekanirdeśa of Maitreyanātha
(Advayavajra) with the Sekanirdeśapañjikā of Rāmapāla. Critical Edition of the San-
skrit and Tibetan Texts with English Translation and Reproductions of the MSS. Manu-
scripta Buddhica 2. Napoli: Asien-Afrika-Institut, Universität Hamburg and Univer-
sità degli studi di Napoli “L’Orientale.”
Jackson, Roger. 2004. Tantric Treasures. Three Collections of Mystical Verse from Buddhist
India. Oxford: Oxford University Press.
Kanō, Kazuo. 2014. “「普賢成就法」の新出梵文資料について [Fugen jōjuhō no shin-
shutsu bonbun shiryō ni tsuite = ‘Newly Available Sanskrit Materials of Jñānapāda’s
Samantabhadrasādhana’].” 密教学研究 [Mikkyōgaku kenkyu] 46: 61–73.
Kawasaki, Kazuhiro. 2004. “On a Birch-bark Sanskrit Manuscript Preserved in the Tibet
Museum.” Journal of Indian and Buddhist Studies 52 (2): 50–52/905–903.
Lo Bue, Erberto. 1997. “The role of Newar scholars in transmitting the Indian Buddhist
heritage to Tibet (c. 750–c. 1200).” In Les habitants du toit du monde. Études recueil-
lies en hommage à Alexander W. Macdonald, edited by Samten Karmay and Philippe
Sagant, 629–658. Nanterre: Société d’ethnologie.
Onians, Isabelle. 2002. Tantric Buddhist Apologetics or Antinomianism as a Norm, D.Phil.
thesis. Oxford: University of Oxford.
Prapanvidya, Chirapat. 1990. “The Sab Bāk Inscription. Evidence of an Early Vajrayana
Buddhist Presence in Thailand.” Journal of Siam Society 78 (2): 11–14.
Sakurai, Munenobu. 1996. インド密教儀礼研究−後期インド密教の灌頂次第− [Indo
mikkyō girei kenkyu: kōki Indo mikkyō no kanchō shidai = A Study of Indian Esoteric
Ritual. Consecration in late Indian Tantric Buddhism]. Kyoto: Hozokan.
Sāṃkṛtyāyan, Rāhul. 1957. Dohā-koś [Hindī-chāyānuvād-sahit]. Paṭnā: Bihār-Rāṣṭrabhā-
ṣā-Pariṣad.
Śāstrī, Haraprasād. 1917. Hājār Bacharer Purāṇa Baṅgalā Bhāṣāy Bauddhagān o Dohā
(Caryācaryaviniścay, Sarojavajrer Dohākoṣ, Kāṇhapāder Dohākoṣ o Ḍākārṇav). Cal-
cutta: Vaṅgīya Sāhitya Pariṣad.
Schaeffer, Kurtis. 2005. Dreaming the Great Brahmin: Tibetan Traditions of the Buddhist
Poet-Saint Saraha. Oxford: Oxford University Press.
Schneider, Johannes. 2010. Vāgīśvarakīrtis Mṛtyuvañcanopadeśa, eine buddhistische
Lehrschrift zur Abwehr des Todes. Vienna: Austrian Academy of Sciences Press.
Sferra, Francesco. 2000. The Ṣaḍaṅgayoga by Anupamarakṣita with Raviśrījñāna’s Guṇa-
bharaṇīnāmaṣaḍaṅgayogaṭippaṇī. Text and Annotated Translation. Serie Orientale
Roma LXXXV. Rome: Istituto Italiano per l’Africa e l’Oriente.
via free access
Shahidullah, Muhammad. 1928. Les chants mystiques de Kāṇha et de Saraha. Les Dohā-
koṣa (en apabhraṃśa, avec les versions tibétaines) et les Caryā (en vieux-bengali) avec
introduction, vocabulaires et notes. Paris: Adrien-Maisonneuve.
Stearns, Cyrus. 1996. “The Life and Tibetan Legacy of the Indian Mahāpaṇḍita Vibhūti-
candra.” Journal of the International Association of Buddhist Studies 19 (1): 127–171.
Taw Sein Ko. 1899. Inscriptions of Pagan, Pinya and Ava. Translation, with Notes. Ran-
goon: Government Printing.
Walter, Michael. 2000. “Cheating Death.” In Tantra in Practice, edited by David Gordon
White, 605–623. Princeton; Oxford: Princeton University Press.
via free access
chapter 8
Reflections on the King of Ascetics (Yatirāja):

Rāmānuja in the Devotional Poetry of Vedānta
Deśika
Srilata Raman
The ocean of Śrīvaiṣṇava literature is vast and it is a humbling scholarly endeav-

or to realize that the more one works on it, the more there is to discover; thus,
any conclusions that one reaches on the intellectual history of the tradition can
only be tentative postulations which can and must be superseded by further
research.1 This being said, it has become increasingly clear that we are see-
ing a particularly fertile period between Rāmānuja (traditional dates: CE 1017–
1137CE) and Vedānta Deśika (traditional dates: CE 1268–1369 CE): a period when
doctrinal ideas are emerging from a wide spectrum of genres—from devotional
poetry and hagiographies to commentaries and kāvya literature. We see also
that, for instance, when it comes to the issue of the salvational means—the
upāya for mokṣa—and their definitions, there was in fact much variation and
a spectrum of views, without one single overarching version. Thus, to take one
example, even within what emerged as a consensus on the significance of pra-
patti as the more appropriate upāya for the Kali Yuga, as opposed to bhakti,
matters were by no means settled in the immediate post-Rāmānuja period as
to how to understand the qualifications (aṅgas) for prapatti, or who was qual-
ified for it.2
1 In this essay, the transliteration of maṇipravāḷa passages is uneven in that I have adhered to
the exact wording in the respective printed books, which is often not systematic, sometimes
giving the devanāgarī letters and sometimes not. All translations are my own.
2 This has been suggested in the tradition particularly of the writings of Meghanādārisūri, a
senior contemporary of Vedānta Deśika. Raghavan (1979), in his survey of Viśiṣṭādvaitic liter-
ature post-Rāmānuja, suggests that the principal work of Meghanādārisūri on prapatti, called
the Mumukṣūpāyasaṃgraha, is currently lost. I have been able to acquire a copy of a single
printed Telugu manuscript with this title which, in the Upodghātaḥ section, points to at least
one view of the author, cited as Meghanādārisūri, which is dramatically different from both
the mainstream Vaṭakalai and Teṅkalai ācāryas on prapatti. The relevant sentence is: asmin
granthe dvitryaṃśavyatirikteṣu bahvaṃśeṣu śrīmadvedāntācaryapakṣīyair upādeyārthā eva
pratipāditā iti bhāti. taditareṣv artheṣv ayam anyatamaḥ sāmānyaviśeṣalakṣaṇalakṣitapra-
pattijñāne tadarthānuṣṭhāne ca traivarṇikānām evādhikāraḥ na śūdrasya iti. śrīmadvedān-
tācāryapraṇīteṣu grantheṣu sapramāṇaṃ prapatteḥ sarvādhikāratā yā samarthitā sā tatraiva
© srilata raman, 2020 | doi:10.1163/9789004432802_010

via free access
reflections on the king of ascetics (yatirāja) 195
With the wisdom of hindsight, therefore, one could say that it is not at all
surprising that a tradition that grappled with such wide divergences would
find some of them encapsulated and formalized in the Teṅkalai and Vaṭakalai
aṣṭādaśabhedas by as late as the eighteenth century. The perception of Rāmā-
nuja, as well as ideas regarding his role in the salvation of his community,
correspondingly, also change with the other doctrinal divergences that emerge
in the Śrīvaiṣṇava tradition after him. This topic forms the core of this paper.
Here, I hope to show that there are many points of convergence in the
hagiographical understanding of Rāmānuja between what consolidated into
the later Vaṭakalai and Teṅkalai ācārya stotra traditions. The primary texts
of comparison here are the Irāmāṉuja Nūṟṟantāti (henceforth, IN) of Tiru-
varaṅkattamutaṉār, one of the earliest hagiographical/stotra works we have
at hand on Rāmānuja (see the next section for some tentative dating), and
Vedānta Deśika’s Yatirājasaptati (henceforth, YS). The paper suggests that we
do not see any major doctrinal divergence between the views of Vedānta
Deśika and those expressed by Tiruvaraṅkattamutaṉār. What we are therefore
seeing is a consolidated viewpoint regarding Rāmānuja’s stature in the tradi-
tion as a whole, which remains unchanged through the centuries. Further, I
would suggest that both these works see Rāmānuja not simply as any other
ācārya of the tradition, but unique in his special role as being central to the
salvation of every Śrīvaiṣṇava. Nevertheless, there are definite variations of
emphasis and hence of doctrinal inflection that can be traced and these, in
turn, the paper suggests, feed into eventual soteriological differences in sig-
nificant ways. Exploring these subtle inflections is also the purpose of this
paper.
The paper begins with a brief examination of the motifs regarding Rāmānuja
in the IN, contextualizing the poem within the history of earlier devotional
poetry addressed to the ācārya within the Śrīvaiṣṇava tradition. It then pro-
ceeds to consider two of the main motifs of the poem relating to Rāmānuja’s
divinity before turning to the YS of Vedānta Deśika. It then discusses the dif-
draṣṭavyā. (“In this work, other than with regard to two or three sections, in many sections,
only the meanings established by those who adhere to Vedānta Deśika’s position have, it
seems, been proven. In the case of meanings that differ, there is this one—that, with regard
to the knowledge of prapatti characterized by general and special features, as well as in the
accomplishment of the meaning of it, only those of the three varṇas are qualified, not the
śūdra. In the texts authored by Vedānta Deśika the establishment of [the position that] all
are qualified for prapatti, together with authoritative proofs, can be seen there [within those
texts themselves].”) I am currently in the process of having this text transcribed and will begin
a study to determine whether something definitive can be said about its authorship.
via free access
196 raman
ference in emphases of the two poems and concludes with some historical
observations about what this might mean for Teṅkalai and Vaṭakalai soterio-
logical doctrine, post-Rāmānuja.
1 Amutaṉār’s Pirapantam on Rāmānuja
The Śrīvaiṣṇava hagiographical tradition places the author of the IN, Tiruvara-
ṅkattamutaṉār or Amutaṉār, as he is also referred to, within Rāmānuja’s own
lifetime. In the hagiography, the Splendour of the Previous Ācāryas (Pūrvācārya
Vaibhavam), Prativāti Payaṅkaram Aṇṇaṅkarācāriyār (1955) states that he was
either the father or the grandfather of Piḷḷai Perumāḷ Aiyaṅkār, the author of
the Aṣṭaprabandham, who was a disciple of Parāśara Bhaṭṭar. Since Bhaṭṭar is
traditionally considered to date to the twelfth century, this would place Amu-
taṉār within Rāmānuja’s own traditional dates of 1017–1137CE. This is further
consolidated by details given in the hagiography of Amutaṉār. In it, Amutaṉār
is introduced as a recalcitrant employee of the Śrīraṅgam temple (kōyil kottil
uḷḷavar) who first resists but later accepts Rāmānuja’s chief disciple Kūrattāḻ-
vāṉ as his own teacher and, rewarded for this by being made the administrative
head of the temple (srīkāryam) by Rāmānuja, subsequently also composes the
IN.3
The IN is an extensive poem of 108 verses composed in the kalittuṟai viruttam
metre. In terms of its poetic type it falls within the “super-genre” called pirapan-
tam. Zvelebil (1974, 193) has pointed out that the great variety of poetic forms
that fall within this category have only this much in common: “the character
of a connected narrative with strong elements of description.” Traditionally,
pirapantams were listed as thirty-six in number but by the eighteenth century,
when Beschi wrote his Caturakarāti, the number had increased to ninety-six.4
The IN faithfully follows the logic of the genre of Nuṟṟantāti, being a poem of
technically one hundred verses (here we actually have eight, auspicious addi-
tional verses), in veṇpā and kallituṟai metres in an antāti arrangement, where
the last syllable or word of the preceding stanza is identical with the first syl-
3 Aṇṇaṅkarācāriyār (1955, 40): … irāmāṉuca nūṟṟantāti aruḷicceyta amutaṉār emperumāṉāru-

ṭaiya niyamaṉattiṉāl āḻvāṉai aṭipaṇintu tiruntiṉavar. A further Tamil work attributed to Amu-
taṉār is the Tiruppatikkōvai, a poem in Tamil of 40 verses in the kōvai genre of pirapantam
literature that is considered to list the 108 sacred places of the Śrīvaiṣṇavas.
4 In effect the pirapantam emerged as the most prolific genre of poetic composition in pre-
modern Tamil literature, outdoing the tanippāṭal, bhakti poetry and epic/narrative forms. On
this see Zvelebil 1974, 194.
via free access
lable or word of the succeeding one. It models itself consciously on the poetry
of the āḻvars, on the Tiviyappirapaṉtam, and its very specific model is the small
work attributed to Maturakavi āḻvār, the “Knotted, Fine, Small Rope” (Kaṇṇinuṉ-
ciṟuttāmpu), since the latter is the first work of the Tiviyappirapaṉtam where
the ācārya himself, rather than Viṣṇu-Nārāyaṇa, becomes the object of vener-
ation and worship.5 Hence, the Kaṇṇinuṉciṟuttāmpu functions as a precursor
to all subsequent hagiographical stotras on the ācāryas of the tradition. The
opening verse of this very short poem of ten verses is:
Sweet it will be, my tongue will fill with nectar

in saying, “Nampi of South Kurukūr”
after nearing him through my Lord,6
the Great Māyaṉ, who allowed himself to be tied
by the knotted, fine, small rope.7
The verse sets the tone for the entire poem, where the poet seeks out Nammāḻ-
vār, implicitly seen as the focus of direct devotion rather than Māyōṉ/Kṛṣṇa. In
his introduction (avatārikai) to the text, Periyavāccāṉ Piḷḷai (an older contem-
porary of Vedānta Deśika traditionally dated to the mid-thirteenth century)
makes it clear that Maturakavi takes refuge with Nammāḻvār himself as the
person who can most help him when it comes to reaching God, as the person
who can bring about the experience of the Blessed One for him (bhagavadanu-
bhavasahakārī) and who, due to helping him in various ways, is himself the
object to be aimed at (uddeśya).8 There are three motifs in the poem: first, that
Nammāḻvār himself is the direct object of refuge; second, that he has rid the
5 Tiruvaraṅkattamutaṉār (1999, 1): nālāyirativyaprapantattil “kaṇṇinuṇ ciruttāpu” nammāḻvār

viṣayamāka amaintatu pōla emperumāṉār viṣayamāka irāmāṉuca nūṟṟantāti eṉappaṭum ti-
vyaprapantam amaintuḷḷatu.
6 The commentators interpret the words “eṉṉappaṉil naṇṇi” in two different ways. They either
take the “il” to be a negative suffix, thereby suggesting that the poet moves away from
Māyaṉ towards Nammāḻvār, or they take it as an ablative suffix, thereby suggesting that he
approaches Nammāḻvār through or via God.
7 Kiruṣṇasvāmi Aiyyaṅkār (1993, 31):
kaṇṇinnuṇciṟuttāmpiṉāl kaṭṭuṇṇap
paṇṇiyaperumāyaṉ eṉṉappaṉil
naṇṇit teṉkurukūr nampiy eṉṟakkāl
aṇṇikkum amut ūṟum eṉ nāvukkē.
8 Kiruṣṇasvāmi Aiyyaṅkār (1993, 23): … prāptitacaiyil pakavatanupavasahakārikaḷāyum, ippaṭi
pahumukamāṉa upakārattālē sarvāvastaiyilum tatīyarē uttēcyareṉṟu atyavacittu, tamakku
vaiṣṇava viṣayīkārattukku muṉpē ajñātajñāpakamukattālē pakavatviṣayattilē mūṭṭiṉa mahō-
pakārakararāṉa āḻvār tammaiyē paṟṟukiṟār.
via free access
198 raman
poet of his bad karma/sins, and finally, that he has rendered the Vedas into
Tamil. As we proceed to look at the IN we will see that these three motifs as
well as the religious assumptions and the contents of the Kaṇṇinuṉciṟuttāmpu
both permeate and influence the poem.
The IN itself came to occupy a special place within quotidian Śrīvaiṣṇava
religiosity. Its significance can be gauged from the fact that, though it is not
composed by the āḻvārs, it is one of the few anomalous works to be included
within the corpus of the Tiviyappirapantam itself, as part of the concluding
section of the Iyaṟpā,9 and hence also a part of the daily cycle of prayers—
the nityānusaṃdhānam—for all Śrīvaiṣṇavas, both Teṅkalai and Vaṭakalai. A
further name attributed to it, and included in Amutaṉār’s traditional hagiogra-
phy, is Prapannagāyatrī—in other words, that like the Gayatrī mantra recited
by the twice-born male during the daily morning and evening prayers (saṃ-
dhyāvandanam), the IN must also be a part of the daily prayer rituals of the
one who has undertaken prapatti (prapanna).10 The hagiography of Amutaṉār
points out that the decision to include the IN within the Iyaṟpā as well as the
decision to make it part of the daily prayer cycle of Śrīvaiṣṇavas was taken by
Uṭayavar (Rāmānuja) himself, as recorded in the chronicles of the Śrīraṅgam
temple (Kōyil Oḻuku).11
In this brief consideration of this poem, with its verses of simple and elegant
beauty, I begin with two examples which show its debt both to the Tiviyappira-
pantam and to the stotra tradition of the ācāryas generally. In the first example,
towards the latter part of the poem, we have the following verse (76) where
there is the piling up of examples of all that which is precious and delectable,
culminating in the assertion that only Rāmānuja can offer the devotee that
which is most precious, which are his feet:
O Rāmānuja!
Enduring fame, expansive waters,
splendid Vēṅkaṭam with its golden summit,
9 The Tiviyappirapantam is traditionally divided into four books of which the Iyaṟpā, mean-
ing, “short metres,” forms the third book. For a detailed discussion of the possible rationale
of the arrangement of the poetry into this four-fold division and their internal contents,
see Hardy 1983, 247–256.
10 Tiruvaraṅkattamutaṉār (1999, 2).
11 Tiruvaraṅkattamutaṉār (1999, 2): uṭaiyavar, nūṟṟantātiyai iyaṟpāvōṭē—mutalāyirattil kaṇ-
ṇinuṇciṟuttāmpaip pōlē cērttu sakala śrīvaiṣṇavarkaḷukkum prapannakāyatrīyāka nityā-
nusantēyam eṉṟum; inta amutaṉārukkum tiruppaṇi ceyvārukkum maṭamum tirumāḷikai-
kaḷilē atyāpaka śrīvaiṣṇavarkaḷukkum muṉṉē pahumāṉamum tiṭṭam ceytu aruḷiṉār. ivai
(kōyiloḻuku) tarum ceytikaḷ.
via free access
the land of Vaikuṇṭha, the celebrated Milk Ocean—

How much pleasure all these give you!
That much pleasure your lotus feet also give me.
Then, grant them, graciously.12
Not only does this verse immediately echo for us the opening line of Nam-
māḻvār’s verse 68 of the Periyatiruvantāti, which begins, “mountains, the grove-
ringed Ocean and the heavenly kingdom of Vaikuṇṭha” (kallum kaṉaikkaṭalum
vaikunta vā nāṭum), but even while echoing the prior poem it is doing some-
thing entirely novel. In Nammāḻvār’s verse the poet is talking about how Kṛṣṇa,
the dark God, has entered his heart and will not be dislodged from it. Hence,
all God’s usual habitations—the sacred mountain, the Cosmic Ocean and even
Vaikuṇṭha itself appear abandoned. Here, in the IN, Amutaṉār seems to suggest
that these places where Kṛṣṇa normally dwells are the very same places which
are within Rāmānuja’s reach—and give him pleasure. For him, in contrast, he
would gladly forsake all these for Rāmānuja’s feet.
Verse 92 of the IN turns to another ubiquitous theme in Śrīvaiṣṇava prap-
atti stotras: the unworthiness of the person seeking refuge. Here, in the first
two lines Amutaṉār says, “I have not done the virtuous vows” (puṇṇiyanōṉpu
purintum ilēṉ), and “I have not spoken the subtle, rare and sacred words”
(nuṇaruḷkēḷvi nuvaṉṟum ilēṉ). These lines resonate for us with the words of
another verse of Nammāḻvār’s: Tiruvāymoḻi 7.5.1 (nōṟṟanōṉp illēṉ, nuṇṇaṟiv-
ilēṉ). We are further reminded of the poetry of another ācārya whose Sanskrit
work was also heavily influenced by the Tiviyappirapantam—Āḷavantār’s Sto-
traratna, verse 22: na dharmaniṣṭho’ smi na cātmavedī, na bhaktimāṃs tvaccara-
ṇāravinde.
The IN thus consciously positions itself within this lineage of devotional
poetry—showing us that it continues and builds upon doctrinal views regard-
ing God and the ācārya, who are seen as mirror images of each other, just as the
poetry of each previous ācārya becomes a template available for embedding, as
a literary echo, within the work of a successor. Indeed, the Kaṇṇinuṉciṟuttāmpu
itself becomes a direct source of further emulation in the Maṇavāḷamāmuṉikaḷ
Kaṇṇiṇuṉciṟuttāmpu—a 13-verse poem (circa fourteenth-century) by the lat-
12 IN, verse 76:

niṉṟa vaṇkīrtiyum nīḻpuṉalum niṟaivēṅkaṭap
poṟkuṉṟamum vaikuṇṭanāṭum kulaviyapāṟkaṭalum
uṉṟaṉakk ettaṉai iṉpantarum uṉ iṇaimalarttāḷ
eṉ taṉukkum atu irāmānucā ivai īṇṭaruḷē.
via free access
200 raman
ter’s disciple Kōyil Kantāṭai Aṇṇaṉ, which centres on Maṇavāḷamāmuṉi’s feet

as a refuge because he, in turn, had taken refuge with Nammāḻvār.
This brings me to one of the main motifs that appears again and again in the
poem: the indirect identification of Rāmānuja with Māyōn/Māyavaṉ/Kaṇṇaṉ
as the dark cloud who brings cool rain and alleviates suffering.
In this regard a few examples from the poem should suffice: “the dark
cloud that is Rāmānuja” (irāmānucaṉ eṉṉum kār), in verse 24; in verse 25, “O
Rāmānuja, benevolent as a dark cloud” (kārēy karuṇai irāmāṉucā ); verse 27
speaks of Rāmānuja as that capacious rain cloud that enters the heart of the
poet; verses 55 and 60 use the word koṇṭal for rain cloud, where Rāmānuja
is likened to a cloud pouring out the rain of the Vedas (55), or a rain cloud
endowed with good qualities (60); and verse 83 speaks of his generosity being
like a dark rain cloud (kār koṇṭa vaṇmai), as does verse 104, likening him to
a bountiful rain cloud (ceḻuṅkoṇṭal). Finally, verses 74 and 104 even suggest,
through a kind of poetic hyperbole similar to that which will surface later in
the Ācāryahṛdayam of Aḻakiya Maṇavāḷapperumāḷ Nāyaṉār,13 that Rāmānuja,
the dark cloud who convinces everyone through cool reason, might even be
more effective than Māyavaṉ with his fiery discus—or that even if one were
shown Kaṇṇaṉ like a gooseberry in the palm of one’s hand, one would want
nothing other than Rāmānuja’s greatness (cīr).14
13 The reference here is to Ācāryahṛdayam, sūtras 56ff., where the Tiviyappirapantam is com-
pared to a golden vessel and the Vedas to an earthen vessel.
14 IN, verse 74:
tērārmaṟaiyiṉ tiṟam eṉṟu māyavaṉ tīyavaraik
kūrāḻi koṇṭu kuṟaippatu koṇṭal aṉaiya vaṇṃai
ērārkuṇatt em irāmānucaṉ avv eḻilmaṟaiyil
cērātavarai citaippatu appōtu oru cintai ceytē
Māyavaṉ reduces with the sharp discus
those who don’t understand the path of the Vedas.
My Rāmānuja of great qualities
with the strength of a raincloud
destroys those who don’t accept the splendid Vedas
merely through a thought.
Verse 104:
kaiyiṟ kaṉiyeṉṉak kaṇṇaṉaik kāṭṭittarilum uṉṟaṉ
meyyiṟ piṟaṅkiya cīr aṉṟi vēṇṭilaṉ yāṉ nirayat
toyyil kiṭakkilum cōtiviṇ cērilum ivv aruḷ nī
ceyyil tarippaṉ irāmānucā eṉ ceḻuṅkoṇṭalē
Rāmānuja, my bounteous cloud!
Even if one were to show Kaṇṇaṉ as a fruit in the hand
other than the splendour emanating from your body
via free access
These multiple references to Rāmānuja as a dark, bountiful, compassionate

and cooling rain cloud cannot but take us back again to the Tiviyappirapantam
and, even prior to it, to the early history of Kṛṣṇa worship in the Tamil coun-
try. As Hardy in his magnum opus Viraha Bhakti has convincingly shown (1983,
149ff.), the earliest references to the God Māyōṉ come to us from specific verses
in the Puṟanāṉūru and Pattuppāṭṭu. The word itself, meaning “the dark com-
plexioned one,” clearly is the Tamil equivalent of Kṛṣṇa and the references from
the earliest poetry are to his dark colour, which is compared to the ocean. Hardy
masterfully traces the images of Māyōṉ from these early references through the
late Caṅkam poetry into the use of the akattiṉai by the āḻvārs. The IN clearly
situates itself with this lineage of poetic images and plays with the image of
Rāmānuja as both Māyōṉ, as better than Māyōṉ, or in a crucial verse I shall
come to later, as the incarnation of the weapons of Viṣṇu-Nārāyaṇa. There is
no doubt that this invocation of Rāmānuja as Kṛṣṇa is meant to set the stage
for the second ubiquitous motif in the poem—that Rāmānuja has cleansed the
poet of his sins by giving him his feet as a refuge.
2 Rāmānuja as Saviour from Sin and Granter of the Ultimate Goal
As mentioned previously, the motif of the ācārya saving the disciple appears in
the Kaṇṇinuṉciṟuttāmpu already. The relevant verse is number 7, where Mat-
urakavi āḻvār says Nammāḻvār removed and destroyed his ancient, bad deeds
(paṇṭaivalviṉai pāṟṟiy aruḷiṉān). It is this very claim that the IN makes with
regard to Rāmānuja. Let me give a few salient examples for this motif in the
poem: verse 4 states that Rāmānuja destroys karma, as in the lines, “After mak-
ing me a substantive thing in this world, he [Rāmānuja] first cut at the root, age-
old karma that secreted illusion …” (eṉṉaip puviyil oru poruḷākki maruḷ curanta
muṉṉaippaḻaviṉai vēr aṟuttu); verse 7 refers to Rāmānuja as he who enables
the poet to cross over sin (paḻiyaik kaṭattum irāmāṉucaṉ); in verse 26 the poet
reiterates that Rāmānuja has removed his true defects that are in the form of
his terrible sins/karma (eṉ cey viṉaiyām meyykuṟṟam nīkki); in verse 35 the poet
asks how karma could ever shroud him after he has in his memory the lotus feet
of Rāmānuja (aruviṉai eṉṉai evvāṟ iṉṟiy aṭarppatuvē); verse 41 plays upon the
idea of the accessibility of Rāmānuja by pointing out that even when Mādhava
I do not want.
Even if cast into the affliction of Hell,
or reach the Heaven of light,
this grace done by you sustains me.
via free access
202 raman
is born on earth within each body, as the Inner Dweller, those who not recog-
nize him are able to become his devotees when Rāmānuja incarnates and gives
them true knowledge; in verse 45 the poet emphatically declares that there is
no other greater desired object than Rāmānuja’s feet, and that no one can make
them attainable except the feet themselves (pēṟ oṉṟu maṟṟ illai niṉ caraṇ aṉṟi,
appēṟ aḷittaṟk āṟeṉṟum illai maṟṟaic caraṇ aṉṟi), etc. In verse 66 the poet sets
up an analogy that both compares and distinguishes between the generosity
of God and Rāmānuja. Mādhava, he says, gives the supreme abode (vāṉam) to
those who pine (naipavar) for him. But this same goal also becomes available
to those who have committed grave sins, because these have been removed by
Rāmānuja from their hearts (valviṉaiyēṉ maṉattil īṉam kaṭinta irāmānucaṉ).
These verses build up to a crescendo in verses 71 (“my previous karma was
removed due to your gracious conduct,” muṉ cey viṉai nī cey viṉayataṉāl pōn-
tatu) and 77 (“he burnt out my sins at their roots,” eṉ viṉaikaḷai vēr paṟiyak kāyn-
tanaṉ). In verse 94 the poet emphasizes the assurance that Rāmānuja grants
right conduct, the wealth of service to God, and the ability to execute this.
In a brief digression it must be added that we cannot fully understand the
theological developments on the Teṅkalai side, or the Tamil stotra tradition
on the āḻvārs and ācāryas in particular, without seeing very similar parallel
developments on the Tamil Śaiva side. We are speaking of a shared religious
and, most importantly, poetic landscape here where there was mutual influ-
ence and emulation. Hagiographical pirapantams were composed on the lead-
ing religious figures of both traditions. Thus, the Irupā Irupaḵtu, attributed to
Aruḷnaṉti Śivācāriyar (traditional dates: thirteenth century), who is considered
one of the foremost disciples of Meykaṇṭar whose work Civañāṉapōtam (Śiva-
jñānabodham) inaugurates the Tamil Śaiva Siddhānta canon, is a pirapantam
poem of 12 verses whose resemblance to the IN is striking. The Irupā Irupaḵtu
equates Meykaṇṭār with Śiva himself, his greatness as a guru being such that
he can rid devotees of primal impurity (mala). Such parallels between poetic
traditions must also be kept in mind when reflecting on the hagiographical
traditions linked to Tamil devotional poetry, considered across the sectarian
divides.
Careful reading of the poem shows that IN sets up a series of analogies:
between Kaṇṇaṉ/Māyōṉ and Rāmānuja, and between the salvation promised
by Kṛṣṇa in Bhagavadgītā 18.66, the Caramaśloka, according to which taking
refuge with him would remove all sins of the devotee, and the descent of
Rāmānuja in the Kali Yuga, which removes the sins of the poet. This analogy
is by no means the only one and I do not wish, hereby, to simplify the poem,
reducing it to this doctrinal message. Indeed, in other verses the poem suggests
that taking refuge at the feet of the devotees of Rāmānuja is equally important,
via free access
or that Rāmānuja is not so much Māyōṉ as the incarnation of Viṣṇu-Nārāyaṇa’s

weapons. But there is no doubt that many of its verses suggest that Rāmānuja
is connected with the promise of the Caramaśloka; and it is this motif, with
the direct analogies it draws between Kṛṣṇa-Vāsudeva, on the one hand, and
Rāmānuja, on other, that is central to the hagiographical impulse of the poem.
But before I expand on what this means for the representation of Rāmānuja
from the earliest hagiographical tradition on him, I would like to focus our
attention on Vedānta Deśika’s 70 verses on the King of Ascetics, the Yatirājasap-
tati, to see how it might converge or diverge in its representation of Rāmānuja.
3 The Yatirājasaptati
The YS is a poem of remarkable lyrical beauty consisting of a total of 74 verses.

Of a very different aesthetic feel than the IN, the poem begins with ten verses on
the guruparamparā, where the poet salutes his lineage of teachers, beginning
with Viṣṇu-Nārāyaṇa and concluding with Rāmānuja. It is carefully structured
to hold in a dialectical unity the greatness of Rāmānuja, his divine incarnation,
on the one hand, and the greatness of his works and his achievements, on the
other. The poem moves in circles, again and again, around these two themes,
with the additional theme of the greatness of his devotees inserted in some of
these circular reflections. Vedānta Deśika tells us, at the very end of the poem
why he has composed it (verse 73):
upaśamitakudṛṣṭiviplavānām upaniṣadām upacāradīpikeyam |

kabalitabhagavadvibhūtiyugmāṃ diśatu matiṃ yatirājasaptatir naḥ ||
May the Yatirājasaptati, the lamp that aids the Upaniṣads by which the
distortions that are wrong views are extinguished, give us the knowledge
that encompasses both the vibhūtis of the Blessed One.
Thus in this verse, which is a phala-śruti, Vedānta Deśika makes it clear that the
poem is as much about Rāmānuja’s feat in composing the Śrībhāṣya, thereby
giving the right interpretation of the Upaniṣads, as it is about the greatness of
Yatirāja himself. Indeed, Vedānta Deśika makes it clear that Rāmānuja’s great-
ness lies in his composition of the work on Vedānta.
The poem’s first ten verses create the context for the exaltation of Rāmānuja.
This context is the lineage of teachers (guruparaṃparā) who preceded him
and are listed in the first eight verses of the poem in the following order:
Nārāyaṇa, Śrī-Lakṣmī, Viṣvaksena, Nammāḻvār, Puṇḍarīkākṣa (Uyyaṅkoṇṭār),
via free access
204 raman
Śrīrāmamiśra (Maṇakkāl Nampi), Yāmuna (Āḷavantār) and Mahāpūrṇa (Periya

Nampi). In verse 11 Rāmānuja is addressed, for the first time, with the phrase
“Lord of the Ascetics” (patiṃ yatīnām).
It is relevant that this is not the first time in the hagiographical literature
that this epithet is used for him. Instead, as the modern commentator of the YS
Vātūla Nīlameghācārya points out, the word is already used for him, as the title
bestowed upon him by Lord Varadarāja at Kāñcīpuram when he took the vows
of an ascetic, according to the hagiographical literature composed contempo-
raneous to the poem. Here the commentator references the Yatirājabhairavam,
composed by Rāmānuja’s immediate disciple Āndhra Pūrṇa (Vaṭuka Nampi)
probably around the same time as the IN, where this episode is narrated.15
When we come to the poem’s understanding of the divinity of Rāmānuja we
find a wide spectrum of meanings. In verse 12, which agrees with verse 33 of the
IN, Rāmānuja is seen as the coming together of all the five weapons of Murāri
(apṛthakpratipanna yanmayatvaiḥ vavṛdhe pañcabhir āyudhair murāreḥ,
12cd). In the very next verse, 13, there is a playful simultaneous narration or
śleṣa on the victorious activities of both Lord Kṛṣṇa and Rāmānuja, where
the meanings converge in the word Rāmāvarajaḥ, referring both to Yatirāja
as well as Kṛṣṇa himself as the younger brother of Balarāma, thus establish-
ing identity between the two figures. Verse 27 lauds Rāmānuja as the aggre-
gation of the threefold splendour (saṃvalita-tridhāman) of Agni (śikhāvān),
the moon (auṣadhīśaḥ) and the sun (tāpanaḥ). In verse 32 Rāmānuja is seen
as having the same capacity to offer protection to the world as Viṣvaksena,
with the latter’s cane staff transformed into his ascetic’s rod.16 Verse 28 is par-
ticularly eloquent in describing and encapsulating all his nurturing and pro-
tecting qualities, which are compared to those present everywhere in nature
itself—as the mountain from which originate all the streams of knowledge
(sakalavidyāvāhinījanmaśailaḥ), the tree under which the weary traveller wan-
dering in saṃsāra takes rest ( janipathaparivṛttiśrāntaviśrāntiśākhī), the ris-
15 YS (2010, 20): parigṛhītaturyāśramasya śrīrāmānujasya bhagavatā śrīvaradarājena yatirāja

iti nāma kṛtam iti guruparaṃparāprabhāve varṇyate. Yatirājabhairavam, verse 50: patnīṃ
parityajya sa vītarāgaḥ śrīdevarājaṃ praṇipatya tasmāt | turyāśramaṃ svīkṛtavān dadau
sa devo ’pi tasmai yatirājanāma ||. The use of the epithet Yatirāja was not the preroga-
tive only of the Śrīvaiṣṇavas but of the hagiographical traditions of other sectarian groups
as well, seeking to exult their ascetic religious founders, as for instance the Madhvas of
Vyāsatīrtha. On this see Stoker 2016.
16 YS, verse 32cd:
viśvaṃ trātuṃ viṣayaniyataṃ vyañjitānugrahaḥ san
viṣvakseno yatipatir abhūt vetrasāras tridaṇḍaḥ ||
via free access
ing sun that keeps the illusionary darkness of those with distorted views at
bay (nikhilakumatimāyāśarvarībālasūryaḥ) and the full moon that brings to
high tide the ocean of the Vedas (nigamajaladhivelāpūrṇacandraḥ). Finally, in
verse 63 Rāmānuja is Viṣṇu himself in his form (mūrtiḥ) as Dattātreya, with his
yellow-ochre robes (pītavasanaḥ) and protective ascetic rod.17
When we consider the range of figures that Rāmānuja is identified with in
the YS we see that they converge overwhelmingly, with some exceptions, on
Viṣṇu-Nārāyaṇa. Thus, the references to him as Agni, the sun and the moon, as
the yellow-robed ascetic God, and then, secondarily, as Viṣvaksena or the col-
lective of the weapons of Viṣṇu, all draw upon images that have a long geneal-
ogy in Vedic, epic and Purāṇic literature on Viṣṇu-Nārāyaṇa. If the YS echoes
other Vaiṣṇavite devotional poetry at all, in addition to the Tiviyappirapantam,
we would do well to consider the images of Viṣṇu presented in the Paripāṭal,
in the context of the overall Vedicism of the early poetic work.18 Thus, despite
the commonality between the IN and the YS in stressing Rāmānuja’s divinity,
his being elevated to a level above that of the other ācāryas, we see a subtle
differentiation in the manner in which the IN foregrounds the significance of
the Bhagavadgītā in contrast to the Vedic, epic and Purānic representations of
Viṣṇu-Nārāyaṇa in the YS. This difference in inflection can also be understood if
we consider the overwhelming importance that the YS attributes to Rāmānuja’s
establishment of the “right” kind of Vedānta.
17 YS, verse 63ab:

kāṣāyeṇa gṛhītapītavasanā daṇḍais tribhir maṇḍitā
sā mūrtir muramardanasya jayati traiyyantasaṃrakṣiṇī ||
18 On this see Raman 2011, 661–662: “A brief comparison of the Tirumāl motifs of the Pari-
pāṭal with those relating to Viṣṇu in the Vedic and immediate post-Vedic Brāhmaṇa and
epic literature shows how thoroughly vedicized the Vaiṣṇavism of the text is. It would
not be unreasonable to speculate that this Vaiṣṇavism of the Paripāṭal may be drawn
from Sanskrit textual sources: the Vedas and Brāhmaṇas, the Mahābhārata (particularly
the Nārāyaṇīya section of the Śāntiparvan) and the Viṣṇupurāṇa. Here, one should draw
attention to the following motifs which already appear in these sources: […] the general
description of Tirumāl clad in yellow garments, with the Goddess and the jewel Kaus-
tubha on his chest and, finally and, most importantly, the identification of Tirumāl with
elements of the Vedic sacrifice in Paripāṭal 2.61–64. … There is also the repeated insistence
in the text that our only true source of knowledge for Tirumāl’s appearance, his deeds, his
prowess and his divine grace are the Vedas (called, variously, maṟai, mutumoḻi and vāy-
moḻi), guarded and transmitted by the Brahmins (antaṇar).”
via free access
206 raman
4 The Eyes of the Vedas: Rāmānuja’s Vedānta19
There are allusions to Rāmānuja’s “protection” of the Vedas, his defeat of those
who hold other Vedāntic views as well as the significance of his establishment
of the right interpretation of the Vedas in innumerable verses of the YS. Thus,
Rāmānuja’s words are a firmly established cage of logic to prevent the wan-
derings that are the Vedas (v. 14) and they also draw in the texts that are the
Vedānta (v. 26).20 They cause those who have touched the bed of the Ocean
that is the rules of the Vedānta to rejoice, and have them establish the state
of salvation.21 Rāmānuja, the poem states, is the tridaṇḍa-bearing ascetic who
sits at the base of the tree of the Vedas, removing the fear of deceitful people
from the minds of people (v. 22).22 Verse 31 captures in a lovely set of images the
nature of Rāmānuja’s works. They are wish-fulfilling trees for the imagination of
debaters (kathakajanamanīṣā-kalpanākalpavṛkṣāḥ), oozing with the nectar of
Hari’s feet (haripadamakarandasyandinaḥ), possessing many branches (anu-
gatabahuśākhāḥ) so that they can remove suffering/heat (āpam unmūlayanti),
and subduing (with their perfume) the stench of sins (śamitaduritagandhāḥ).
The repeated reference to the Vedic basis of Rāmānuja’s teachings is stressed
in several further phrases in verses 44, 47, 50, etc. In verse 57 Vedānta Deśika
interestingly historicizes the tradition, accepting that even if Rāmānuja’s doc-
trine (mata) is new (navīnaṃ) and others might have come before (prāk), this
does not matter. For Rāmānuja is within the lineage of those ancient commen-
tators such as Ṭaṅka, Dramiḍa and Guhadeva, who were fearless (nirātaṅkāḥ)
because of their unobscured vision (nijamatitiraskāravigamāt). This view of
Rāmānuja’s central role in the establishment of a new, Viśiṣṭādvaita Vedānta
is the second central motif in the poem. Thus, even while we also have verses
in the poem that, like in the IN, echo the sentiments of taking refuge at the
feet of Rāmānuja, such as verses Verse 18, where those who take refuge at the
feet of Yatipati become free of sin (anaghā) or Verse 20, where the feet are a
19 YS, verse 30: śrutinayanasanābhiḥ śobhate lakṣmaṇoktiḥ.

20 YS, verse 14:
abahuśrutasaṃbhavaṃ śrutīnāṃ jaratīnām ayathāyathapracāram |
vinivartayituṃ yatīśvaroktir vidadhe tāḥ sthiranītipañjarasthāḥ ||
Cf. verse 26a: ākarṣaṇāni nigamāntasarasvatīnāṃ
21 YS, verse 21:
śvasitāvadhūtaparavādivaibhavāḥ nigamāntanītijaladhes talaspṛśaḥ |
pratipādayanti gatim āpavargikīṃ yatisārvabhaumapadasātkṛtāśayāḥ ||.
22 YS, verse 22ab:
mūle niviśya mahatāṃ nigamadrumāṇāṃ
muṣṇan pratārakabhayaṃ dhṛtanaikadaṇḍaḥ |
via free access
refuge compared to medicine that stills the fluctuations of the mind (calacit-
tavṛttivinivartanauṣadhaṃ śaraṇaṃ yatīndracaraṇaṃ vṛṇīmahe), it becomes
clear that the framework of the poem does not allow for a predominant focus
on the analogy between Rāmānuja and Kṛṣṇa nor that his feet themselves are
the predominant upāya for mokṣa.
In summing up, we see a common emphasis in the representation of Rāmā-
nuja in the two poems, the one composed in Tamil and the other in Sanskrit
separated by centuries. This was a representation rooted in a strong hagio-
graphical, stotra tradition parallel to that of the guruparamparās that emerge
from the twelfth century onwards and the various commentaries with hagio-
graphical elements on the Tiviyappirapantam. It reaffirmed Rāmānuja’s divine
descent, similar to the divine descent of the āḻvārs. It was less uniformly codi-
fied, though, as to in what or in whom to locate Rāmānuja’s divinity—in Kṛṣṇa
himself, in Viṣvaksena or Dattātreya, in Viṣṇu-Nārāyaṇa or his weapons. It is
also equally clear that from early on the Śrīvaiṣṇava tradition was united in
seeing Rāmānuja as central to the salvific process, and in considering that, for
those who were his disciples, taking refuge in Rāmānuja would accelerate the
path to Vaikuṇṭha and the state of servitude (kaiṅkarya) to Viṣṇu-Nārāyaṇa.
Nevertheless, the frequent references to Rāmānuja as destroying one’s sins, a
reference directly calqued on the central promise of the Caramaśloka, is not
absent but does not function as the fulcrum of the YS as it does in the IN.
Here, we would do well to recollect Vedānta Deśika’s poetic tribute to other
figures of the guruparaṃparā in the kāvya style, such as his references to Nam-
māḻvār in the Pādukāsahasram, the mahākavya of 1008 stanzas that centres
around the Rāmāyaṇa episode where Bharata takes the sandals of Rāma on
his head and has these reign as the symbol of Rāma’s presence, in the latter’s
absence. Playing in certain verses on the double meaning of the pādukās—
one as the divine sandals and one as Nammāḻvār himself being the sandals,
adorning the feet of God as his ideal devotee—allows the poet to speak of Nam-
māḻvar’s subordination (śeṣatva) as well as his greatness in giving us the Tamil
Vedas.23 Similar to the treatment of Rāmānuja in the YS, the Pādukāsahasram
section on Nammāḻvār is careful to place him within the context of the entire
guruparaṃparā tradition as someone who performs an extraordinary function
in establishing the doctrinal foundations of the school, rather than as a unique
figure who towers entirely above all the others of the lineage. Such a perspec-
tive, which even while lauding the achievement of each respective ācārya also
23 For a brief analysis of the relevant verses (22–29) of the Pādukāsahasram, see Hardy 1979,
64–67.
via free access
208 raman
relativizes it within a historical understanding of the tradition, is very different

from the affective framework of the IN, which dwells on the emotional reso-
nance of a single figure.
5 Post-Prapatti
The IN’s greater emphasis on Rāmānuja in his role as the ācārya significant
for salvation anticipates Teṅkalai doctrinal development, as we see it evolv-
ing in the works of Piḷḷai Lokācārya (traditional dates: CE 1264–1327) and in the
commentaries on his works. In his works attention had decisively shifted to a
soteriological path that was considered even more appropriate than prapatti
for the seeker of salvation—both in terms of being easier as well as the most
appropriate expression of the humility of the soul. This new salvific means was
“love for the ācārya”—ācāryābhimāna.
In her thorough study of the writings of Maṇavāḷamāmuṉi in the immediate
post-Vedānta Deśika period, Mumme (1988) demonstrates the consolidation of
the doctrine of ācāryābhimāna as the preferred form of salvation in his com-
mentaries on Piḷḷai Lōkācārya’s works. Summarily put, the doctrine of ācāryā-
bhimāna mirrors, in a mimetic fashion, the reasons why prapatti was favoured
above bhaktiyoga in the immediate post-Rāmānuja stage of doctrinal forma-
tion. Ācāryābhimāna is not simply for the person incapable of prapatti; it is
also “not only a separate means, but the superior or ultimate means (caramo-
pāya), given the ultimate limits of the soul’s nature and destiny” (p. 243). The
pragmatics of it meant a total surrender of all one’s own responsibility for doing
anything for salvation by relying on the ācārya to do the needful. What, then,
exactly was the ācārya to do? As the commentary of Piḷḷai Lōkam Cīyar on Piḷḷai
Lokācārya’s Arthapañcakam explains, this meant, in effect, that one surren-
dered all responsibility for performing prapatti to the ācārya, who out of great
compassion and being a great devotee himself, does it for you in your stead
and advises you on all your future conduct, so that the prapatti he does for you
might prove efficacious.24 Thus, as Mumme (1988, 226–227) puts it succinctly,
24 Arthapañcakam, Sūtra 9 commentary: “The person who does this ācāryābhimāna is inca-
pable of doing anything else. With regard to him, reflecting on his lowliness and the
happiness of the Lord if he were to obtain him, like a mother, who when her child falls ill,
sees this as her own fault and gives it medicine, the ācārya is that greatly compassionate
one (paramadayālu) and great devotee (mahābhāgavata) who can undertake the means
of salvation for the soul. All that the souls has to do is to surrender to the love for him and,
via free access
The only means of salvation in practice today among the Teṅkalai is

ācāryābhimāna—the love of the ācārya. As they see it, Rāmānuja has
already done prapatti to the Lord for all future generations of his follow-
ers. Therefore, rather than surrendering to the Lord himself, one merely
has to take refuge with an ācārya of his lineage. For the Vaṭakalai, the
way of salvation is through the performance of prapatti to the Lord in
the manner prescribed by the śāstras under the guidance of a qualified
ācārya. Though the ācārya can perform prapatti specifically for a partic-
ular disciple, mere relationship with the ācārya is no substitute for the
performance of bharaṇyāsa [understood in the broadest sense as the lay-
ing down of all one’s agency] with it’s aṅgas.
Hence, she also goes on to point out that there is a widely held belief among
contemporary Teṅkalai Śrīvaiṣṇavas, not established in any of the writings of
Piḷḷai Lōkācārya, Maṇavāḷamāmuṉi or Vedānta Deśika, that “Rāmānuja is the
ācārya who has saved all future generations of Śrīvaiṣṇavas with his prapatti …”
There is a conspicuous textual exception to the absence of this doctrine in the
writings of the major ācāryas of both traditions in the thirteenth to fourteenth
centuries, though. This is the Caramopāyanirṇaya of Nāyaṉār Āccāṉ Piḷḷai (tra-
ditional dates: 1227–1327), the nephew and adopted son of Periyavāccāṉ Piḷḷai,
a work that clearly endorses the view that taking refuge with Rāmānuja himself
is central to salvation, thus making this a doctrinal position found well before
contemporary belief. Mumme’s arguments (1988, 87–89) for doubting that this
could be the work of Nāyaṉār Āccāṉ Piḷḷai is not dissimilar to the controversy
regarding Rāmānuja’s authorship of the Gadyatraya, which I have addressed in
some detail in my 2007 book: stylistically, she suggests it is different (too sim-
ple) compared to the same author’s other doctrinal works and ideologically it
is far too radical for its time, anticipating a doctrine of ācāryabhimāna centred
on Rāmānuja which only came much later.
The detailed look I have taken at the Caramopāyanirṇaya, to be dealt with in
another paper, shows a great deal of doctrinal convergence with the IN. Indeed,
both works in their emotional appeal to Rāmānuja within a salvific framework
[as in the saying], “if you were to give me the certain prize, then we are meant to obtain
it”, with regard to all his activities do and not do what he [the ācārya] commands.”
Arthapañcakam vyākhyāna, p. 68: ācāryāpimāṉamāvatu ivayoṉṟukkum caktaṉaṉrik-
kēyiruppāṉ oruvaṉaik kuṟittu, ivaṉuṭaiya iḻavaiyum, ivaṉaip peṟṟāl īcuvaraṉukk uṇṭāṉa prī-
tiyaiyum aṉusantittu staṉantaya prajaikku vyātiyuṇṭāṉāl atu taṉ kuṟaiyāka niṉaitta auṣata
sēvaiyaip paṇṇum mātāvaippōlē ivaṉukkāka tāṉ upāyānuṣṭāṉam paṇṇi rakṣikavalla para-
matayāluvāṉa makāpākavataṉ apimāṉattilē oṭuṅki, vallaparicu varuvipparēl atu kāṇṭum
eṉṟu collukiṟapaṭiyē sakalapravṛttinvṛttikaḷaiyum avaṉiṭṭavaḻakkākkukai.
via free access
210 raman
allow one to assume that Rāmānuja’s salvific qualities, stretching to a point

where they can extend over historical posterior time, had become accepted
doctrine in at least one important strand of the tradition, even within (if not
shortly after) his postulated dates. That Vedānta Deśika, for his part, was also
not unaware of these views centred around Rāmānuja becomes clear from a
particular passage of his major work on prapatti, the Rahasyatrayasāram. In
Chapter 8 of the text, where he is discussing the different categories of those
qualified for prapatti (adhikārivibhāga), Vedānta Deśika differentiates between
the uktiniṣṭhaṉ person, whose prapatti is based upon the utterance taught to
them by their ācārya, and the ācāryaniṣṭhaṉ person, whose prapatti is done for
them by their ācārya. Regarding this second category of persons, he has this to
say:
Among these [two categories], the ācāryaniṣṭhaṉ is himself included

within the ācārya’s laying down of his burden (bharasamarpaṇam)
with regard to him and his own …. For this ācāryaniṣṭhaṉ, according to
the axiom of “how much more, then” (kaumutika nyāya), there can be
no doubt as to the attainment of the fruit. Mutaliyāṇṭāṉ [Rāmānuja’s
nephew] taught the verse: like those creatures on the body of a lion
that leaps from one mountain to another, when Bhāṣyakāra [Rāmānuja]
jointly leaps [does prapatti], then, due to our bodily relationship with him
[i.e. being related to him due to kinship ties], we too have been elevated
[we get the same salvific benefits as he does].25
Vedānta Deśika is fully aware that these words attributed to Mutaliyāṇṭāṉ

are capable of being understood as implying that Rāmānuja’s act of prapatti
becomes the single soteriological event that saves all those who come after him
in the community. Precisely to firmly reject such an understanding he adds,
almost immediately, the following emphasis:
In the Nyāsatilakam [verse 21] we also said: “The blind man is able to move
about led by the one with sight, O Lord of Śrīraṅgam; the cripple, placed
within a boat, is taken across by a boatman; the children of a servant relish
[royal] food though they don’t know the king; thus will my compassion-
25 Vedānta Deśika (1980, 293–296): ivarkaḷil ācāryaniṣṭhaṉ …. ācāryaṉuṭaiya ātmātmīyabha-

rasamarpaṇattilē tāṉum antarbhūtaṉ. … ācāryaniṣṭhaṉukku kaumutikanyāyattālē pha-
lasiddhiyil sandeham illai. oru malayil niṉṟum oru malaiyilē tāvum simhacarīrattil jan-
tukkaļaippōlē bhāṣyakārar saṃyoga laṅghanam paṇṇa avarōṭu uṇṭāṉa kuṭaltuvakattālē
nāmum uttīrṇar āvuvōm eṉṟu mutaliyāṇṭāṉ aruḷicceyta pācuram.
via free access
ate teacher help me to reach you.” The Lord of All will not relent to give
the supreme goal without prapatti being done in some way, through some
person.26
The significant point here is that Deśika is citing poetry of his own to show
that the ācārya is the helper, even up to a point where he can do the prap-
atti for you. What this means that is that it is your current ācārya who can do
this for you, and this is what the Nyāsatilakam verse elaborates. This does not
mean that Rāmānuja’s prapatti at one historical moment absolves his entire
community from henceforth doing it. Thus, Vedānta Deśika is concerned to
subtly disagree with the statement of Mutaliyāṇṭāṉ’s or at least differs in his
interpretation from what it comes to mean later within the Teṅkalai tradition.
Ultimately, the doctrinal musings on Rāmānuja are central to a theme that
lies at the heart of the tradition: the issue of how to reconfigure the nature of
God’s and the religious canon’s own accessibility to the community of ordi-
nary devotees. Succinctly put, the literature of the Śrīvaiṣṇava tradition has
repeatedly reconfigured the motif of accessibility both to enable the incor-
poration of doctrinal change and to allow for canonical expansion. Thus, the
central argument in the equation of the Tiviyappirapantam and the Vedas is
one of the accessibility of Vedic revelation in a manner that overcomes lin-
guistic, caste and gender boundaries. We see this elaborated in detail for the
first time in Nañjīyar’s Āṟāyirappaṭi vyākhyānam and then more radically reit-
erated in the Ācāryahṛdayam of Aḻakiya Maṇavāḷapperumāḷ Nāyaṉār. The IN is
also drawing upon this theme of accessibility—suggesting that Rāmānuja’s feet
provide the same refuge in the Kali Yuga that Kṛṣṇa’s did in the Dvāpara Yuga.
While Vedānta Deśika is determined to pay homage to Rāmānuja’s significance
within the tradition and sees him as occupying a summit of his own, he is also
determined to place him within a framework that establishes the coherence of
the school of Viśiṣṭādvaita as a whole. The differences between the two poems
are reflective of the differences, broadly speaking, in the hermeneutical strate-
gies between the Teṅkalai and Vaṭakalai literature, which Mumme (1988) again
summarizes so well: the Teṅkalai literature, in general, leans towards hyperbole
and dramatic intensity, drawing heavily upon popular idioms and metaphors
and rooting itself in the Tamil, devotional poetry. In contrast, in Vedānta Deśika,
26 Vedānta Deśika (1980, 296–297): andho’ nandhagrahaṇavaśago yāti raṅgeśa yadvat, paṅ-
gur naukākuharanihito nīyate nāvikena | bhuṅkte bhogān aviditanṛpaḥ sevakasyārbhakā-
diḥ, tvatsaṃprāptau prabhavati tathā deśiko me dayāluh || eṇṛu nyāsatilakattilē coṉṉōm.
Ētēṉum oru parkaramākavumām, ārēṉum oruvar anuṣṭikkavumām prapattikk allatu sar-
veśvaraṉ puruṣārtham koṭukka iraṅkāṉ eṉṟatāyiṟṟu.
via free access
212 raman
particularly, we see the need to balance rhetorical flourishes with scholastic

precision. He wishes to be seen to be speaking for the coherence of the tradi-
tion as a whole, particularly in order that it withstand pan-Vedāntic scholas-
tic scrutiny. For him, homage to Rāmānuja must not lead, through rhetorical
excess, to a new and radical turn in the doctrine of prapatti, as it does in the
Teṅkalai case.
The analysis of the stotra literature on Rāmānuja here, by no means exhaus-
tive but rather exemplary of the formative phase of doctrine, also reinforces
for us what has already been established for Rāmānuja’s own writings, as in the
case of the Gadyatraya: that devotional poetry composed not just by the āḻvārs
but also the later ācāryas is as central as commentaries and independent works
to the evolution of Śrīvaiṣṇava doctrine.
Finally, it is no mere coincidence that both the Vaṭakalai and Teṅkalai nityā-
nusaṃdhānam texts begin with a pool of common single, independent verses
(taṉiyaṉ) of salutation to each figure in turn in one’s guruparaṃparā, which Śrī-
vaiṣṇavas are encouraged to recite on a daily basis. Both contain Kūrattāḻvāṉ’s
taṉiyan that ends with the words asmadguror bhagavato ’sya dayaikasindhoḥ,
rāmānujasya caraṇau śaraṇam prapadye. This convergence in Rāmānuja as the
object of veneration speaks for his centrality—as the central gemstone in the
necklace of the guruparamparā, as Vedānta Deśika says in his poem (verse 15):
a pendant that holds together, gives structure to and makes resplendent not
just the individual beads of the necklace but indeed, as the other ācāryas who
are themselves likened to the beads of the necklace would say, adorning and
forming the dazzling central gemstone of the entire tradition.
Abbreviations
IN Irāmāṉuja Nūṟṟantāti of Tiruvaraṅkattamutaṉār.

YS Yatirājasaptati of Vedānta Deśika.
Acknowledgements
My thanks go to Elisa Freschi and Marcus Schmücker for their careful reading
of the first draft and Harunaga Isaacson for his generous correction of errors
and felicitous reformulations of some textual passages.
via free access
References
Aṇṇaṅkarācāriyār, Prativāti Payaṅkaram. 1955. Pūrvācāriya Vaipavam. Kāñcīpuram:

Granthamala Office.
Hardy, Friedhelm W. 1979. “The Tamil Veda of a Śūdra Saint. (The Śrīvaiṣṇava Inter-
pretation of Nammāḻvār).” In Contributions to South Asian Studies, edited by Gopal
Krishna, 29–87. Delhi: Oxford University Press.
Hardy, Friedhelm W. 1983. Viraha-Bhakti. The Early History of Kṛṣṇa Devotion in South
India. New Delhi: Oxford University Press.
Kiruṣṇacuvāmi Aiyyaṅkār, S. 1993. Srīmaturakaviyāḻvār aruḷicceyta Kaṇṇinuṇ-ciṟuttām-
pu vyākyāṉaṅkaḷ. Tiruccirāpaḷḷi: Puttūr Akirakāram.
Mumme, Patricia Y. 1988. The Śrīvaiṣṇava Theological Dispute. Maṇavāḷamāmuni and
Vedānta Deśika. Madras: New Era Publications.
Mumme, Patricia Y. 1998. “The Evolution of the Teṅkalai Understanding of the Ācārya:
Teacher, Mediator and Saviour.” Journal of the Ananthacharya Indological Research
Institute 1: 75–98.
Piḷḷai Lōkācārya. 2011. Arthapañcakam. With the Commentary of Piḷḷai Lōkam Cīyar.
Tiruchi: Srīvaiṣṇavasrī.
Raghavan, V.K.S.N. 1979. History of Viśiṣṭādvaita Literature. Delhi: Ajanta Publications.
Raman, Srilata. 2007. Self-Surrender (Prapatti) to God in Śrīvaiṣṇavism. Tamil Cats and
Sanskrit Monkeys. Routledge Hindu Studies Series. Routledge: London and New
York.
Raman, Srilata. 2011. “Tamil, Vaiṣṇava, Vaidika: Kiruṣṇacuvāmi Aiyaṅkār, Irāmānuja Tāt-
ācāriyār and Modern Tamil Literary History.” Journal of Indian Philosophy 39 (6):
647–676.
Stoker, Valerie. 2016. Polemics and Patronage in the City of Victory: Vyasatirtha, Hindu
Sectarianism, and the Sixteenth-Century Vijayanagara Court. California: University
of California Press.
Tiruvaraṅkattamutaṉār. 1999. Irāmānucanūṟṟantāti. Periyacīyar Cīyar aruḷicceyta urai-
yum, Piḷḷai Lōkācārya Cīyar aruḷicceyta Vyākyānamum. Srīraṅgam: Srīvaiṣṇavasrī.
Vedānta Deśika. 2010. Yatirāja Saptatiḥ. Vaiyākaraṇamūrdhanyena Kūttapākkaṃ Vātsya
Nīlameghācāryeṇa kṛtā saṃskṛtavyākhyā. Srīraṅkam Srīmat Āṇṭavaṉ Āsramaveḷiy-
īṭu.
Vedānta Deśika. 1980. Srimad Vedanta Desika’s Srimad Rahasya Trayasara with Sara Vis-
tara (Commentary) by Sri Uttamur T. Viraraghavacharya. Madras: Ubhayavedānta
Granthamāla.
Zvelebil, Kamil V. 1974. A History of Indian Literature. Tamil Literature. Wiesbaden: Har-
rasowitz.
via free access
via free access
part 3
Religion, the State, and Social History
via free access
via free access
chapter 9
Not to Worry, Vasiṣṭha Will Sort It Out:

The Role of the Purohita in the Raghuvaṃśa
Csaba Dezső
It is my greatest pleasure to dedicate this paper to Professor Alexis Sanderson,

whose guidance I was fortunate to have during my doctoral studies at the Uni-
versity of Oxford. I couldn’t have wished for a better supervisor. His formidable
knowledge of the intellectual history of early-mediaeval India has been both
overwhelming and inspiring right from the first lectures I attended. The reason
why I hold that without the sound skills of philology it is pointless to undertake
the study of the past is to a large extent thanks to the mastery he has shown in
using those skills. The breadth of his learning and his keen interest in diverse
aspects of classical Indian culture prevented me from ever feeling uncomfort-
able for not doing research on Śaivism, his main field, but persisting in studying
kāvya. Our tutorials had a formative influence on my scholarship; they were
always stimulating and eminently enjoyable. I am grateful to have had the good
fortune of being his student.
Recounting the deeds of several kings of a royal lineage, Kālidāsa’s epic
does not describe a single hero’s rise to success (abhyudaya). One could regard
with Bonisoli-Alquati (2008, 105) the dynasty itself as the protagonist of the
Raghuvaṃśa. There is one character, however, whose timely interventions help
the continuance of the dynasty throughout the whole epic: this character is
Vasiṣṭha, the royal chaplain, purohita. But how is it possible that Vasiṣṭha was
the purohita and guru of the Sūryavaṃśa for so many generations? Is it the same
Vasiṣṭha?
A legend in the Taittirīya Saṃhitā tells us that among the sages it was Vasiṣṭha
alone who could see Indra. The god taught him the Stomabhāgas with the
charge that any king who had him as purohita would thereby flourish if Vasiṣṭha
did not tell the Stomabhāgas to other sages. “Therefore—teaches the text—one
should have a descendant of Vasiṣṭha (a Vāsiṣṭha) as one’s brahman priest.”1 The
1 Taittirīya Saṃhitā 3.5.2.6: tásmād Vāsiṣṭhó brahmā ́ kāryàḥ. In the later Śatapatha Brāhmaṇa
we read that formerly only a descendant of Vasiṣṭha could become brahman, but now anyone
who is suitable can (12.6.1.41, quoted in Minkowski 1991, 126).
© csaba dezső, 2020 | doi:10.1163/9789004432802_011

via free access
218 dezső
brahman was the priest who silently monitored the ritual. He was associated
with the Atharvaveda and with the office of the family priest, the purohita of
the patron of the sacrifice, the yajamāna.2 We may connect with these Vedic
passages the tradition that Vasiṣṭha or several Vasiṣṭhas were the purohita(s)
of the kings of Ayodhyā, the members of the Ikṣvāku- or Sūryavaṃśa. Pargiter
(1922, 203ff.) distinguished seven Vasiṣṭhas in the legends, but, as he pointed
out, these Vasiṣṭhas merged into one person. Thus we read in the Mahābhārata
(1.164.9cd–11):
Ikṣvākavo mahīpālā lebhire pṛthivīm imām.

purohitavaraṃ prāpya Vasiṣṭham ṛṣisattamam
ījire kratubhiś cāpi nṛpās te kurunandana.
sa hi tān yājayām āsa sarvān nṛpatisattamān
brahmarṣiḥ Pāṇḍavaśreṣṭha Bṛhaspatir ivāmarān.
The Ikṣvāku kings conquered this world.

Having obtained Vasiṣṭha, the best of sages, as their excellent purohita,
those kings performed sacrifices, O descendant of the Kurus.
For that brahmin sage officiated for all those great kings at their sacri-
fices,
O best of the Pāṇḍavas, as Bṛhaspati did for the gods.
In the Raghuvaṃśa, too, it is the same Vasiṣṭha who is the purohita of the
whole dynasty. As Dilīpa says to the priest (1.71cd): Ikṣvākūnāṃ durāpe ’rthe
tvadadhīnā hi siddhayaḥ,3 “For the successes of the Ikṣvākus depend on you in
goals that are hard to achieve.” When Rāma returns from Laṅkā he also greets
Vasiṣṭha as the guru of the lineage of Ikṣvāku (ikṣvākuvaṃśagurave praṇamya
…, 13.70).
Both Vedic and epic texts emphasise the purohita’s role by the king’s side
in achieving victories and averting dangers.4 As Gonda (1975, 322) has shown,
this protective function is already expressed by the name of the office: puro
hita means “placed in front,” that is in front of the king like a shield. In the
Aitareya Brāhmaṇa we read that “the gods do not eat the food of that king
2 Cf. Pañcaviṃśa Brāhmaṇa 12.8.6: annaṃ vai brahmaṇaḥ purodhā, “the office of the purohita
is the food of the brahman.”
3 I quote the verses of Raghuvaṃśa 1–6 from Goodall and Isaacson 2003, and the verses of can-
tos 7–19 from Nandargikar 1982, unless indicated otherwise.
4 Gonda 1975 passim, Gonda 1956, 150 ff.
via free access
not to worry, vasiṣṭha will sort it out 219
who has no purohita,”5 and “the purohita is half of the kṣatriya’s self,”6 he is
the protector of the kingdom, rāṣṭragopa.7 This close relationship between
king and chaplain and their co-operation in the protection of the state is
expressed in a verse of the Raghuvaṃśa in which Kālidāsa writes about king
Atithi (17.38):
Vasiṣṭhasya guror mantrāḥ sāyakās tasya dhanvinaḥ

kiṃ tat sādhyaṃ yad ubhaye sādhayeyur na saṅgatāḥ?
The mantras of Vasiṣṭha, the guru, and the arrows of that archer—
what is there to achieve that these two could not achieve when
united?
According to the Arthaśāstra the purohita had a central place among the king’s
closest advisers. The ninth chapter of its first book gives us some details about
the person of the royal chaplain (1.9.9–10):
purohitam uditoditakulaśīlaṃ sāṅge vede daive nimitte daṇḍanītyāṃ

cābhivinītam āpadām daivamānuṣīṇām atharvabhir upāyaiś ca
pratikartāraṃ kurvīta.
tam ācāryaṃ śiṣyaḥ pitaraṃ putro bhṛtyaḥ svāminam iva cānuvarteta.
He should appoint as chaplain a man who comes from a very distin-

guished family and has an equally distinguished character, who is
thoroughly trained in the Veda together with the limbs, in divine
omens, and in government, and who could counteract divine and
human adversities through Atharvan means.
He should follow him as a pupil his teacher, a son his father, and a ser-
vant his master.
trans. Olivelle 2013, 74
When Dilīpa visits Vasiṣṭha in his ashram he extolls the merits of the purohita
in preserving the welfare of the kingdom. Vasiṣṭha is the one who “counteracts
5 Aitareya Brāhmaṇa 8.24.2: na ha vā apurohitasya rājño devā annam adanti (quoted in Gonda
1975, 320).
6 Aitareya Brāhmaṇa 7.26.4: ardhātmo ha vā eṣa kṣatriyasya yat purohitaḥ (quoted in Gonda
1975, 320.)
7 Aitareya Brāhmaṇa 8.25.2 (quoted in Gonda 1975, 332).
via free access
220 dezső
both divine and human calamities.”8 The Arthaśāstra lists eight divine disas-
ters (4.3.1): fire, water, famine, rats, harmful animals, snakes and demons. In
another passage (8.4.1) it mentions fire, flood, disease, famine and epidemic.
The mantras and rituals of the Atharvaveda and thus the purohita who was
an expert in Atharvavedic rituals had major roles in averting these calamities.9
Vasiṣṭha’s “holy spells defeat the enemy even from a distance,”10 and “his offer-
ing thrown into the fire becomes rain for the crops parched by drought.”11 It
is the sacred power (brahmavarcasa) of the purohita thanks to which diseases
and other disasters avoid the people (1.63). The king attributes the welfare and
safety of his country to his guru (1.64):
tad evaṃ cintyamānasya guruṇā Brahmayoninā

sānubandhāḥ kathaṃ na syuḥ sampado me nirāpadaḥ?
Therefore when my guru, who was born from Brahmā, takes care of me
in this way,
how could my accomplishments not be continuous, free from calami-
ties?
Vasiṣṭha, who is well-versed in yoga, discovers with the help of concentration

what causes Dilīpa’s childlessness: once he failed to salute Surabhi, the sacred
cow, who cursed the king, at least in some versions of the texts. In another ver-
sion it is the king’s negligence in itself that made him childless.12 Luckily the
grandchild of Surabhi is in Vasiṣṭha’s ashram, so the king can wait on her to
obtain her blessing. As soon as her name has been uttered the holy cow appears:
a good omen according to the sage, who is also an expert in the interpretation
of signs, as a purohita should indeed be.
In the eleventh sarga Daśaratha is on his way back from the court of the
king of Mithilā, where they have celebrated the weddings of his sons, when he
encounters ominous signs: strong headwinds tear the flags of his army, a fright-
ening circle surrounds the sun, the points of the compass filled with red clouds
8 Raghuvaṃśa 1.60: daivīnāṃ mānuṣīṇām ca pratikartā tvam āpadām, cf. Ayyar 1925, 8.
9 Arthaśāstra 9.7.84 prescribes the rites of the Atharvaveda against too much or too little
rain, and against demons, 4.3.40 also against demons, while 4.3.35 against snakes. On the
association of the purohita with the Atharvavedic tradition see Sanderson 2007, 204f.,
notes 28 and 29.
10 Raghuvaṃśa 1.61: mantrair dūrāt saṃyamitāribhiḥ.
11 Raghuvaṃśa 1.62: havir āvarjitaṃ … vṛṣṭībhavati sasyānām avagrahaviśoṣitām.
12 On these variants see Dezső 2014, 163 f.
via free access
resemble a menstruating woman, and female jackals are howling.13 Daśaratha

asks for Vasiṣṭha’s help (11.61 in Vallabhadeva’s text):14
tat pratīpapavanādi vaikṛtaṃ kṣipraśāntam adhikṛtya kṛtyavit

anvayuṅkta gurum īśvaraḥ kṣiteḥ svantam ity alaghayat sa tad-
vyathām.
The king who knew what to do asked his guru about those portents like
the headwind etc.,
if they would be averted soon, and he removed his fears saying, “It will
end well.”
The southern commentators (Mallinātha, Aruṇagirinātha and Hemādri) read

the second pāda as prekṣya śāntim adhikṛtya (v. 62 in their version), that is
Daśaratha asks the purohita about the pacifying ritual that quells the ills.
Vasiṣṭha’s reply, “it will end well,” might mean in this case that there is no need
to perform such a ritual. The omens indicate the arrival of Paraśurāma, who is
defeated by Rāma, so the story does indeed end well.15
The purohita performed such important ceremonies for the royal family as
the rituals associated with childbirth. As we read in the Arthaśāstra (1.17.26):
prajātāyāḥ putrasaṃskāraṃ purohitaḥ kuryāt, “When she (i.e. the queen) has
delivered, the chaplain should perform the rite of passage for the son” (trans.
Olivelle 2013, 89). As Olivelle (2013, 486) remarks, this may also refer to the rites
of early childhood, possibly including the upanayana. In the Raghuvaṃśa we
see that Vasiṣṭha performs the priestly duties when Dilīpa’s son Raghu is born
(3.18):
sa jātakarmaṇy akhile tapasvinā tapovanād etya purodhasā kṛte

Dilīpasūnur maṇir ākarodgataḥ prayuktasaṃskāra ivādhikaṃ babhau.
When the complete birth ritual was done by the ascetic chaplain who
had come from the grove of ascetics,
Dilīpa’s son shone yet more, like a precious stone taken from a mine and
then polished.
13 Raghuvaṃśa 11.57–60 in Vallabhadeva’s commentary, 58–61 in Mallinātha’s.

14 Quoted from a draft edition prepared by Dominic Goodall, Harunaga Isaacson, Csaba Kiss,
and myself.
15 The source of the story is Rāmāyaṇa 1.73.7 ff.
via free access
222 dezső
In Kālidāsa’s poetry, as Ingalls (1976, 19) has observed, ritual is associated

with beauty. The purohita is like the jeweller who polishes the precious stone
and transforms it into a gem. He performs the life-cycle rituals that make the
individual a member of civilised society.
The purohita also had an important role in the marriage ceremonies of
the royal families. Aja and Indumatī were wed by the chaplain of Indumatī’s
brother (7.20):
tatrārcito Bhojapateḥ purodhā hutvāgnim ājyādibhir agnikalpaḥ

tam eva cādhāya vivāhasākṣye vadhūvarau saṃgamayāṃ cakāra.
There lord Bhoja’s venerable chaplain, who was like fire, offered clarified
butter and other things to the fire,
and having made the same [fire] witness to the marriage he wed the
bride and the groom.
At the end of his life Raghu entrusted the kingdom to his son, Aja, and lived
the life of a yogin, aiming for apavarga, liberation from the cycle of rebirths
(8.16). He chose the time of his death himself: “by yogic meditation he reached
the eternal Soul, beyond darkness” (tamasaḥ param āpad avyayaṃ puruṣaṃ
yogasamādhinā, 8.24). The verse describing Raghu’s funeral ceremonies is
transmitted in two versions. Vallabhadeva knew the following version (8.26):16
śrutadehavisarjanaḥ pituś ciram aśrūṇi visṛjya Rāghavaḥ

vitatāna samaṃ purodhasā kratum antyaṃ pṛthivīśatakratoḥ. (8.26)
When he heard that his father had laid off his body, Raghu’s son shed
tears for a long time,
and then performed the final sacrifice of that Indra of the earth together
with the chaplain.
The second half is transmitted in the southern commentaries as follows (8.25cd

in Mallinātha, Aruṇagirinātha and Hemādri):
vidadhe vidhim asya naiṣṭhikaṃ yatibhiḥ sārdham anagnim agnicit.
and myself.
via free access
… he, who had arranged the sacrificial fires, performed his last rites with-
out fire together with the ascetics.
According to this version Raghu’s body was not cremated but buried, as it was
the custom in the case of saṃnyāsins, ascetics striving for final liberation. The
purohita is not mentioned in this version; it was rather Raghu’s fellow yogins
who attended the ceremony. In the version known to Vallabhadeva it is not
made explicit that the ritual was performed without fire, and Vasiṣṭha, the puro-
hita was also present. If we suppose that this was the original version, the text
known to the later commentators might be the result of a correction or dis-
ambiguation: since Raghu had retired from secular life and was striving for
mokṣa as a renunciant, it might have seemed more correct to have his burial
performed by the community of ascetics.17 However, the question might be
more complicated, since in the variants of the verses describing Raghu’s final
years, as well as in the Keralan commentaries and that of Hemādri, we can
detect a tendency to weed out or explain away references to Raghu being a
saṃnyāsin. But in this verse, interestingly, it is the Southern commentators’ ver-
sion in which Raghu is clearly treated as a deceased yati. Raghu’s position as a
retired king was not without ambiguities: his son, Aja, did not allow him to
retire to the forest, so Raghu moved to a place outside the palace, bahiḥ kṣiti-
pālaveśmanaḥ, at least in Vallabhadeva’s version (8.14),18 because the Southern
commentators knew a text according to which the old king moved outside the
town, purād bahiḥ.19 Here we perhaps see the same effort that tries to make
Raghu’s position less ambiguous and to move him further away from the palace
and worldly life. But this effort is not completely successful since in the second
half of the same verse we read that Śrī, the goddess of wealth and royal majesty,
who was now enjoyed by his son, served Raghu like a daughter-in-law (samu-
pāsyata putrabhogyayā snuṣayevāvikṛtendriyaḥ śriyā), which means that Raghu
did not entirely live the life of an ascetic.
The king of the last canto of the epic, Agnivarṇa, lived a dissolute life and
died of a consuming disease. His death was kept secret (19.54):
taṃ gṛhopavana eva saṃgatāḥ paścimakratuvidā purodhasā

rogaśāntim apadiśya mantrinaḥ sambhṛte śikhini gūḍham ādadhuḥ.
17 Thus Goodall 2001, 121.

18 8.14ab: sa bahiḥ kṣitipālaveśmano nivasann āvasathe yativrataḥ … (Quoted from a draft
edition prepared by Dominic Goodall, Harunaga Isaacson, Csaba Kiss, and myself.)
19 8.14ab: sa kilāśramam antyam āśrito nivasann āvasathe purād bahiḥ.
via free access
224 dezső
The ministers joined by the chaplain who knew the last rites placed him
on the pyre in secret
in the palace garden, under the pretext of a ceremony that averts dis-
ease.
The ministers and the purohita concealed the king’s death because they wanted
to secure the succession to the throne. Since Agnivarṇa did not have a son, they
placed his pregnant widow on the throne, in accordance with the instructions
of the Arthaśāstra (5.6.36).20
The purohita had a central role in the ceremony of the royal consecration.
About Aja we read the following in the Raghuvaṃśa (8.3–4):
anubhūya Vasiṣṭhasaṃbhṛtaiḥ salilais tasya mahābhiṣecanam

viśadocchvasitena medinī kathayām āsa kṛtārthatām iva.
sa babhūva durāsadaḥ parair guruṇātharvavidā kṛtakriyaḥ
pavanāgnisamāgamo hy ayaṃ sahitaṃ brahma yad astratejasā.
Having experienced his great consecration with water gathered by

Vasiṣṭha,
the earth seemed to express her contentment with clear sighs.
When the ritual had been performed for him by the guru who knew the
Atharvaveda, he became unassailable by his enemies,
for when brahman is united with the power of weapons it is a union of
wind and fire.
Both Aruṇagirinātha and Mallinātha quote ad loc. the Manusmṛti which

teaches the inseparability of royal and priestly power (9.322):21
nābrahma kṣatram ṛdhnoti nākṣatraṃ brahma vardhate

brahma kṣatraṃ ca saṃpṛktam iha cāmutra vardhate.
The Kṣatriya does not flourish without the Brahmin, and the Brahmin
does not prosper without the Kṣatriya;
but when Brahmin and Kṣatriya are united, they prosper here and in the
hereafter.
20 See Dezső 2014, 161 f. Though many generations separate Raghu and Agnivarṇa, Vallab-
hadeva also identifies the purohita of the latter as Vasiṣṭha (commentary to 19.54).
21 Cf. Ayyar 1925, 8.
via free access
In the case of Atithi’s consecration Kālidāsa stresses the importance of

Atharvavedic mantras (17.13):
purohitapurogās taṃ jiṣṇuṃ jaitrair atharvabhiḥ

upacakramire pūrvam abhiṣektuṃ dvijātayaḥ.
The brahmins headed by the chaplain began to consecrate him who was
destined to victory
first with Atharvavedic mantras that lead to victory.
The rituals performed by the purohita made the king invincible in battle. When
Raghu was about to set off to defeat Kuvera, the god of riches, Vasiṣṭha empow-
ered his chariot (5.27):
Vasiṣṭhamantrokṣaṇajāt prabhāvād udanvadākāśamahīdhareṣu

marutsakhasyeva balāhakasya gatir vijaghne na hi tadrathasya.
Due to the power arisen from the sprinkling performed with mantras by
Vasiṣṭha
the course of his chariot was not blocked on the ocean, in the sky and in
the mountains, like that of a cloud helped by the wind.
The source of the mantras used by the purohita was again probably the Athar-
vaveda. The ritual manual of the Kauśikasūtra (15.11) contains several verses
prescribed for the consecration of the chariot.
In the Raghuvaṃśa we see the purohita at every important ceremonial occa-
sion taking place in the royal court, from the birth of the heir through the
marriage of the prince to the funeral of the king. Vasiṣṭha protected the king’s
person and made him invincible in battle with his Atharvavedic mantras. On
the other hand Vasiṣṭha was also the guru and adviser of the king who showed
him the path to be followed: we see him in this role in the eighth canto when
he comes to the support of the despairing Aja.
When a garland falling from the sky kills his wife, Aja laments over her with
some of the most moving verses of Sanskrit literature. The only reason he does
not throw himself on the funeral pyre is that people would speak badly of a
king who follows a woman in death. Vasiṣṭha learns in his ashram that Aja has
been paralysed by grief,22 and since he is engaged in a ritual and therefore can-
22 abhiṣaṅgajaḍam: 8.75 in Mallinātha, 76 in Hemādri, 73 in Aruṇagirinātha. Vallabhadeva

reads abhiṣaṅginam (8.76).
via free access
226 dezső
not visit the king personally he sends one of his disciples to convey Aja his
message. Vasiṣṭha knows everything that has happened, happens or will hap-
pen in heaven, on earth or in the netherworld, so he tells Aja in his message
that Indumatī was a cursed apsaras who had to live in a human body until she
was touched by a heavenly garland. Aja should stop grieving (8.84 in Vallabha-
deva):23
tad alaṃ tadapāyacintayā vipad utpattimatām avasthitā

vasudheyam avekṣyatāṃ tvayā vasumatyā hi nṛpāḥ kalatriṇaḥ.
So do not think about her death. Those who have been born will surely
die.
Have regard for this Earth, for the Earth is the true wife of kings.
Vasiṣṭha first expresses the truism we also find in the Bhagavadgītā, jātasya hi
dhruvo mṛtyuḥ, “death is certain for all that is born.”24 Then he reminds Aja of
his duty as a king: he must protect the earth, who, as the goddess Earth, is also
his wife. This idea has already appeared in the same canto: Raghu handed over
Aja the earth as a second Indumatī, and Aja took possession of the earth gently,
as one enjoys a newly wedded wife (8.7). His divine wife gave him lots of gems,
his human wife gave him a valorous son (8.28). But Aja’s lament makes it clear
whom he regards as his real wife (8.53 in Vallabhadeva):25
manasāpi na vipriyaṃ mayā kṛtapūrvaṃ nanu kiṃ jahāsi mām

vata śabdapatiḥ kṣiter ahaṃ tvayi me bhāvanibandhanaṃ manaḥ.
Surely I have not offended you even in my thoughts, why are you leaving
me?
Truly I am the earth’s husband only in name, my heart is bound with
feelings to you.
If we consider Aja’s words we might suspect that Vasiṣṭha’s admonition will not
have much effect on him.
and myself.
24 Bhagavadgītā 2.27, quoted by Vallabhadeva ad loc.
and myself.
via free access
The disciple continues Vasiṣṭha’s message as follows (8.85 in Vallabhadeva):26
udaye yad avāpyam ujjhatā śrutam āviṣkṛtam ātmanas tvayā

manasas tad upasthite jvare punar aklīvatayā prakāśyatām.
The learning that you showed when you avoided what could have been
achieved in the time of success—
demonstrate the same again like a man now that your heart is suffering.
“What could have been achieved” (avāpyam) and what Aja avoided was, ac-
cording to Vallabhadeva, ecstasy or being overjoyed (praharṣaṃ madaṃ vā). As
at the time of success he could stay away from an excessive mental state, in the
same way he should behave like a man and be composed at the time of grief.27
Aja could not find Indumatī even if he followed her in death, since those who
live in the other world go on paths that differ according to their karma (8.86 in
Vallabhadeva). Aja should stop crying because the tears of the relatives burn
the departed souls (a belief we also read about in texts on dharmaśāstra28).
Then Vasiṣṭha’s message becomes more philosophical:29
maraṇaṃ prakṛtiḥ śarīriṇāṃ vikṛtir jīvitam ucyate budhaiḥ.

kṣaṇam apy avatiṣṭhate śvasan yadi jantur nanu lābhavān asau. (8.88 in
Vallabhadeva)
The wise say that death is the natural state of embodied creatures and
life is a change in that state.
If a being remains breathing even for a moment it is surely fortunate.
avagacchati mūḍhacetanaḥ priyanāśaṃ hṛdi śalyam arpitam,

itaras tu tad eva manyate kuśaladvāratayā samuddhṛtam. (8.89 in Vallab-
hadeva)
and myself.
27 Instead of yad avāpyam Mallinātha (8.84) and Aruṇagirinātha (8.82) read madavācyam,
while Hemādri reads yad avācyam (8.85).
28 E.g. Yājñavalkyasmṛti 3, 11: śleṣmāśru bāndhavair muktaṃ preto bhuṅkte yato ’vaśaḥ | tas-
mān na roditavyaṃ hi kriyā kāryā prayatnataḥ ||, quoted by Vallabhadeva and Aruṇagiri-
nātha ad loc.
and myself.
via free access
228 dezső
The foolish man regards the loss of his dear one as a dart shot into his
heart.
Another man looks on the same as a dart that has been pulled out, for it
is a door to beatitude.
svaśarīraśarīriṇāv api smṛtasaṃyogaviparyayau yadā

virahaḥ kam ivānutāpayed vada bāhyair viṣayair vipaścitam? (8, 90 in
Vallabhadeva)
When we are taught that our own body and soul unite and then sepa-
rate,
tell me which wise person should be tormented by separation from the
external objects of the senses?
Death should be regarded as an unavoidable, natural event, which can even

be beneficial: as Vallabhadeva says, when attachment ceases, one attains the
imperishable state (snehakṣayāt kila nirapāyapadāvāptiḥ), the cutting of one’s
emotional bonds helps achieve the stopping of rebirth and redeath. Aja should
regard his dead wife as one of the sense objects he meets and is parted from
every day.
Finally Vasiṣṭha appeals to Aja’s pride:30
na pṛthagjanavac chuco vaśaṃ vaśinām uttama gantum arhasi.

drumasānumatoḥ kim antaraṃ yadi vāyau dvitaye pi te calāḥ? (8.91 in
Vallabhadeva)
Best of the self-controlled! You ought not to become subject to grief like
common people.
What would be the difference between a tree and a mountain if both
shook in the wind?
When Duṣyanta gives way to despair over the loss of Śakuntalā, the vidūṣaka
expresses his puzzlement with similar words: “How have things got this far?
How could the fortitude of good men be affected by grief? Surely mountains
remain unshaken even by a storm wind.”31 Duṣyanta is shaken up by Mātali,
and myself.
31 Abhijñānaśākuntala 6, 103 (p. 278): kiṃ ṇedaṃ īdisaṃ uvaṇadaṃ? kadā uṇa sappurisā
soabaddhadiyyā honti? ṇaṃ pavādeṇa vi girīo ṇippakampā. Trans. Vasudeva 2006, 279.
via free access
Indra’s charioteer, who pretends to kidnap the vidūṣaka and thus arouses in
the king his fighting spirit and his sense of duty to protect others.
Aja seemingly acquiesced in the words of his purohita, but, as Kālidāsa
writes, Vasiṣṭha’s counsel “could not find a place in his heart full of grief and
returned, as it were, to his guru” together with the disciple.32 The king took
into consideration that his son was not yet of age and reigned eight more years,
while his only solace was the likeness of his beloved (similarly to Duṣyanta who
found pleasure only in painting a portrait of Śakuntalā). When his heart was
completely shattered by grief, Aja starved himself to death and left his body
behind at the confluence of the Gaṅgā and the Sarayū. Then he met his beloved
again in Indra’s heaven in a form that surpassed her former beauty.
We do not know what Kālidāsa’s sources were for Aja’s tragic story, but there
is a work with which we can make interesting comparisons: Aśvaghoṣa’s Bud-
dhacarita.33 When prince Siddhārtha leaves the palace to live the life of a
wandering ascetic in search of the teaching that gives liberation, the king’s min-
ister and purohita undertake to find him and appeal to his better self. Though
they do not trust in the success of their undertaking, they set off nevertheless,
setting off a battle between the bodhisattva and the diverse rules of scrip-
ture.34
Like in the Raghuvaṃśa, we read about a message, but in the Buddhacarita it
is the purohita who conveys the words of the king to his son, a king whose heart
has been transfixed by the dart of grief.35 The king is inconsolable because Sid-
dhārtha has not waited until his old age, as is prescribed in the śāstras, to retire
to the forest in search of mokṣa, and so his father could not transfer the king-
dom to his son and retire himself (9.22). Only the coward needs such things
as the forest and the symbols of ascetics (vanaṃ ca liṅgaṃ ca hi bhīrucihnam,
9.18), for kings may also walk on the path leading to mokṣa (9.20). The purohita
tries to stir the bodhisattva’s feelings and paints in vivid colours the sorrow he
caused to his father, foster mother, wife, little son and the whole harem when
he left the palace behind (9.23ff.).
The bodhisattva, after some thought, replies that he had no other choice:
because of the fear of sickness, old age and death he had to forsake his fam-
32 8.92 in Vallabhadeva: tad alabdhapadaṃ hṛdi śokaghane pratiyātam ivāntikam asya guroḥ.
33 On the parallels between the works of Aśvaghoṣa and Kālidāsa see Gawronski 1914–1915;
Johnston 1984, lxxxi; Nandargikar 1982, 161–196; and Tubb 2014.
34 Buddhacarita 8.85: yadi tu nṛvara kārya eva yatnas tvaritam udāhara yāvad atra yāvaḥ |
bahuvidham iha yuddham astu tāvat tava tanayasya vidheś ca tasya tasya.
35 9.13: tvacchokaśalye hṛdayāvagāḍhe.
via free access
230 dezső
ily.36 Separation is unavoidable anyway, as in the case of travellers who meet

on the way, therefore his father’s clinging to him is pointless (9.33):
maddhetukaṃ yat tu narādhipasya śokaṃ bhavān āha, na tat priyaṃ me,

yat svapnabhūteṣu samāgameṣu saṃtapyate bhāvini viprayoge.
You spoke about the king’s grief on account of me; I am not pleased that
he is so distressed,
amidst associations as fleeting as dreams, when separation is bound to
take place.
Such ideas might be familiar from the Raghuvaṃśa where the purohita tries
to shake up Aja with similar words, but without much success: during the
remaining eight years of his life Aja consoles himself with the festivals of brief
encounters with his wife in his dreams.37 While Aja out of a sense of duty does
not throw away his life immediately and reigns until his son grows into a man,
nothing and nobody can convince the bodhisattva to break his vow, return to
the palace and take upon himself the burden of kingship, which he deplores.38
After the purohita the minister also tries to persuade the bodhisattva to
return to the palace. He argues that we have no certain knowledge about the
afterlife, so one should enjoy power and wealth as long as one can (9.53 ff.). Real
liberation is the clearing of the three debts: towards one’s ancestors by father-
ing offsprings, towards the gods by sacrificing and towards the sages by studying
the Vedas (9.65). But the bodhisattva cannot be discouraged: for him there is no
way back to the palace, and instead of a sceptical attitude he wants to acquire
the knowledge leading to liberation himself (9.73).
The mission of the purohita in the Buddhacarita does not succeed; the bodhi-
sattva is unwilling to leave the path he has chosen himself (9.78):
tad evam apy eva ravir mahīṃ pated, api sthiratvaṃ Himavān giris tyajet,
adṛṣṭasattvo viṣayonmukhendriyaḥ śrayeya na tv eva gṛhān pṛthagjanaḥ.
Therefore, although the sun may fall to earth, or Himālaya lose its fixity,
I’ll not return home like a common man,
36 9.31: vyādhijarāvipadbhyo bhītas tv agatyā svajanaṃ tyajāmi.

37 Raghuvaṃśa 8.93: svapneṣu kṣaṇikasamāgamotsavaiḥ.
38 Buddhacarita 9.40: mohāyatanaṃ nṛpatvam.
via free access
whose senses yearn for sensual things, and who has not perceived the
truth.
Siddhārtha does not want to live the life of a common man, a life into which
the purohita and the minister want to pull him back. For him separation from
his loved ones is indeed the door leading to liberation. Vasiṣṭha asks Aja to rise
above common people and not to let his feelings take hold of him. Aja should
accept what cannot be avoided; he should not grieve because of separation that
is bound to happen, but he is unable to do this. Siddhārtha has recognised the
impermanence of human relations and he is not shaken by appeals to family
bonds.
The bodhisattva is more than human; his firm resolution to break away from
the values of brahmanical society inspires awe. Aja remains very much human
in his grief. Vasiṣṭha demands of him detachment while remaining in the world:
a way of life that Vasiṣṭha himself, an ascetic sage who is at the same time a royal
chaplain, lives. Vasiṣṭha is truly the purohita of the dynasty and the dynasty
needs kings who rule in an exemplary fashion, beget offspring and transfer the
kingship to a suitable heir when the time has come. He saves the dynasty again
since the transfer of power between Aja and his son Daśaratha takes place
smoothly, but he fails to have any influence on Aja’s feelings. Here we touch
upon one of the key issues of Kālidāsa’s poetry: the difficulty of creating and
maintaining a harmony of duties, interests, and emotions, be they love or grief.
Aja’s example shows that there is a grief so deeply felt that no teaching about
detachment, no appeal to duty can assuage.
Acknowledgements
This research has been made possible by financial support from the European
Research Council synergy project “Beyond Boundaries: Religion, Region, Lan-
guage and the State,” grant agreement no. 609823.
References
Primary Sources
Abhijñānaśākuntala. Somadeva Vasudeva, ed., trans. The Recognition of Shakúntala by
Kalidasa. Clay Sanskrit Library. New York: NYU Press & JJC Foundation, 2006.
Arthaśāstra. R.P. Kangle, ed. The Kauṭilīya Arthaśāstra, Part I: A Critical Edition With
via free access
232 dezső
a Glossary. Delhi: Motilal Banarsidass, 1988 (reprint of the second edition of

1969).
Aitareya Brāhmaṇa. Theodor Aufrecht, ed. Das Aitareya Brāhmaṇa. Mit Auszügen aus
dem Commentare von Sāyaṇācārya und anderen Beilagen. Adolph Marcus: Bonn,
1879.
Kauśikasūtra. Bloomfield, M. 1890. “The Kāuçika-Sūtra of the Atharva Veda, with Ex-
tracts from the Commentaries of Dārila and Keçava.” Journal of the American Orien-
tal Society 14: i–lxviii, 1–424.
Taittirīya Saṃhitā. A. Weber, ed. Die Taittirîya-Saṃhitâ. Leipzig: Brockhaus, 1871–1872.
Pañcaviṃśa Brāhmaṇa. Ānandachandra Vedāntavāgīśa, ed. Tāṇḍya Mahābrāhmaṇa
with the Commentary of Sāyaṇa Āchārya. Kolkata: Asiatic Society of Bengal, 1870–
1874.
Buddhacarita. E.H. Johnston, ed., trans. Aśvaghoṣa’s Buddhacarita, or Acts of the Bud-
dha. Delhi: Motilal Banarsidass, 1984 (combined reprint of 1936)
Manusmṛti. Patrick Olivelle, ed., trans. Manu’s Code of Law. A Critical Edition and Trans-
lation of the Mānava-Dharmaśāstra. New York: Oxford University Press, 2005.
Mahābhārata. V.S. Sukthankar, ed. The Ādiparvan, Being the First Book of the Mahāb-
hārata, the Great Epic of India, For the First Time Critically Edited. Poona, Bhandarkar
Oriental Research Institute, 1933.
Raghuvaṃśa with Aruṇagirinātha’s commentary. K.A. Poduval, and C.K. Raman Nam-
biar, eds. Raghuvamsa by Mahakavi Kalidasa with Prakasika Commentary of Aruna-
girinatha & Padarthadeepika Commentary of Narayana Panditha. 3 vols. Tripuni-
thura: Sanskrit College Committee, 1959–1964.
Raghuvaṃśa with Mallinātha’s commentary. The Raghuvaṃśa of Kālidāsa With the Com-
mentary of Mallinātha Edited with A Literal English Translation, Copious Notes in
English Intermixed with Full Extracts, Elucidating the Text, from the Commen-
taries of Bhaṭṭa Hemādri, Cāritravardhana, Vallabha, Dinakaramiśra, Sumativijaya,
Vijayagaṇi, Vijayānandasūri’s Varacaraṇasevaka and Dharmameru, with Various
Readings etc. etc., edited by G.R. Nandargikar, 5th edition. Delhi: Motilal Banarsi-
dass, 1982.
Raghuvaṃśa with Vallabhadeva’s commentary, sargas 1–6. Dominic Goodall and Haru-
naga Isaacson, eds. The Raghupañcikā of Vallabhadeva, Being the Earliest Commen-
tary on the Raghuvaṃśa of Kālidāsa. Vol. I. Groningen: Egbert Forsten, 2003.
Raghuvaṃśa with Vallabhadeva’s commentary, sargas 7–19. Critical edition in progress
by Cs. Dezső, D. Goodall, H. Isaacson, and Cs. Kiss.
Raghuvaṃśa with Hemādri’s commentary. R.P. Dwivedī, ed. Raghuvaṃśadarpaṇa:
Raghuvaṃśa Commentary by Hemādri, Vol. 1. Patna: Kashiprasad Jayaswal Research
Institute, 1973.
via free access
Secondary Sources
Ayyar, K.B. 1925. “A Study of Kālidāsa in Relation to Political Science.” In Proceedings and
Transactions of the Third Oriental Conference, December 22nd to 24th, 1924, Madras,
1–16. Madras: The Law Printing House.
Bonisoli-Alquati, A. 2008. Kālidāsa’s Raghuvaṃśa: An Analysis. Dottorato di Ricerca in
Studi Indologici e Tibetologici Ciclo XXI. Universita degli Studi di Torino.
Dezső, Cs. 2014. “‘We do not fully understand the learned poet’s intention in not com-
posing a twentieth canto’: Addiction as a Structuring Theme in the Raghuvaṃśa.”
South Asian Studies 30 (2): 159–172.
Gawronski, A. 1914–1915. “Gleanings from Aśvaghoṣa’s Buddhacarita.”Rocznik Oryental-
istyczny 1: 1–42.
Gonda, J. 1956. “Ancient Indian Kingship from the Religious Point of View (Continued).”
Numen 3 (2): 122–155.
Gonda, J. 1975. “Purohita.” In Selected Studies. Vol. II. Sanskrit Word Studies, edited by
J. Gonda, 320–337. Leiden: Brill.
Goodall D. 2001. “bhūte ‘āha’ iti pramādāt: Firm Evidence for the Direction of Change
Where Certain Verses of the Raghuvaṃśa are Variously Transmitted.” Zeitschrift der
Deutschen Morgenländischen Gesellschaft 151: 103–124.
Ingalls, D.H.H. 1976. “Kālidāsa and the Attitudes of the Golden Age.” Journal of the Amer-
ican Oriental Society 96 (1): 15–26.
Minkowski, Ch.Z. 1991. Priesthood in Ancient India. A Study of the Maitrāvaruṇa Priest.
Sammlung de Nobili vol. XVIII. Vienna: Institut für Indologie der Universität Wien.
Olivelle, P. 2008. Life of the Buddha by Ashvaghosha. Clay Sanskrit Library. New York:
NYU Press & JJC Foundation.
Olivelle, P. 2013. King, Governance, and Law in Ancient India. Kauṭilya’s Arthaśāstra. New
York: Oxford University Press.
Pargiter, F.E. 1922. Ancient Indian Historical Tradition. London: Oxford University Press.
Sanderson, A. 2007. “Atharvavedins in Tantric Territory. The Āṅgirasakalpa Texts of
the Oriya Paippalādins and their Connection with the Trika and the Kālīkula.” In
The Atharvaveda and its Paippalāda Śākhā: Historical and Philological Papers on a
Vedic Tradition, edited by Arlo Griffiths and Annette Schmiedchen, 195–311. Aachen:
Shaker Verlag.
Tubb, G. 2014. “Baking Umā.” In Toward a History of Kāvya Literature. Innovations and
Turning Points, edited by Y. Bronner, D. Shulman, and G. Tubb, 71–85. New Delhi:
via free access
chapter 10
Buddhism, Kingship and the Protection of the

State: The Suvarṇaprabhāsottamasūtra and
Dhāraṇī Literature
Gergely Hidas
1 Introduction
This article focuses first on the ritual core of the Suvarṇaprabhāsottamasūtra,

which teaches the protection of the state for the mutual benefit of the Bud-
dhist Sangha and the monarch. The essay then explores the ways in which this
theme appears in dhāraṇī literature in the first half of the first millennium.1 It
is shown that offering safeguard to rulers and their regions is a long-established
practice in South Asian Buddhism, persisting into modern times, and that there
have been a variety of incantation scriptures available for accomplishing this
task.
2 The Suvarṇaprabhāsottamasūtra: The Sūtra of Golden Light
This scripture of Mahāyāna Buddhism survives in its oldest form in South Asia
in two palm-leaf manuscripts from Nepal dated to the 14th and 16th centuries.2
The remaining witnesses from the subcontinent are paper codices of later cen-
turies, but dozens of Central Asian Sanskrit fragments survive from the second
half of the first millennium. The Sanskrit text of the Suvarṇaprabhāsottama
has been edited a few times, most recently by Skjaervø (2004). This scripture
was remarkably influential across the Buddhist lands of Asia, and was trans-
lated into Chinese, Tibetan, Khotanese and several additional languages over
the course of many centuries. The Chinese translation by Dharmakṣema is the
earliest, from circa 417CE, which marks the terminus ante quem for the emer-
gence of this sūtra. Editions of Tibetan and Chinese recensions were published
1 For a survey of dhāraṇī literature, see Hidas 2015.

2 Cambridge Add. 2831, 1385 CE; Tōyō Bunkō 1979, 1581 CE. It appears that previous publications
have left the dated colophon of the former manuscript unnoticed (NS 505 [1385CE], written
as pañcādhikaḥ pañcaśataḥ; see cudl.lib.cam.ac.uk/view/MS-ADD-02831/96, lines 3–4).
© gergely hidas, 2020 | doi:10.1163/9789004432802_012

via free access
buddhism, kingship and the protection of the state 235
by Nobel in the 1940s and 1950s, while Skjaervø (2004) provides the most com-
prehensive treatment of the Khotanese version. Several translations of this
scripture are available in Western languages, with Emmerick 1970 being the
classical point of reference, based on the Sanskrit. An excellent overview of
the Suvarṇaprabhāsottama with a detailed account of previous research was
published by Gummer (2015).
3 The Ritual Core of the Suvarṇaprabhāsottamasūtra
The structure of this scripture shows evidence of considerable expansion over

the centuries. The most widespread Sanskrit version contains nineteen chap-
ters, but Nobel (1937) suggests that there may only have been fourteen pari-
vartas initially.3 He also proposed that chapter 3, the Deśanāparivarta, is likely
to represent the original core of the Suvarṇaprabhāsottama, containing verses
on the confession of sins, the resolution to attain buddhahood and praise of
the buddhas, the recitation or hearing of which brings about great benefit.4 If
we look, however, for chapters that actually have detailed ritual instructions,
it appears that the ritual core of this sūtra is contained in sections 6, 7 and
8: the Caturmahārājaparivarta, Sarasvatīparivarta and Śrīparivarta. The latter
two include more complex observances and dhāraṇī-spells.
Chapter 6, the Caturmahārājaparivarta or Chapter on the Four Great Kings,5
conveys an explicit message: those kings (manuṣyarāja) who venerate the
Suvarṇaprabhāsottama and support the Buddhist Sangha will be protected
from hostile armies and other dangers by the Four Great Kings, and their coun-
tries will exist in highest state of harmony. Simultaneously, those who ignore
this tradition will face decline. This sūtra directly and repeatedly refers to
monarchs, and throughout this chapter and the whole scripture it is obvious
that kings are envisaged as the principal target audience. At one point the
3 On the structure of this sūtra, see Skjaervø 2004, lii–liv. The earliest Chinese version contains
eighteen chapters, while later Chinese and Tibetan recensions consist of 21, 24, 29 or 31 sec-
tions.
4 Cf. Suzuki 2012.
5 These eminent yakṣas Dhṛtarāṣṭra, Virūḍhaka, Virūpākṣa and Vaiśravaṇa are celestial guard-
ians of the cardinal directions, along with their retinues in Jambudvīpa. From the Pali Āṭānā-
ṭiya-sutta (Dīgha-nikāya 32) to various Sanskrit rakṣā texts, they often take the role of chief
protectors. Note that the Four Great Kings guarded the cardinal points at the great stūpa of
Bhārhut, circa 100 BCE (Skilling 1992, 163). For a study of the Sanskrit Āṭānāṭīya/Āṭānāṭika-
sūtra, see Sander 2007.
via free access
236 hidas
scripture calls itself a rājaśāstra,6 a text for kings, and it is notable that chap-
ter 12 is entitled Devendrasamayarājaśāstraparivartaḥ, or Chapter of the Royal
Treatise called Conventions for Divine Kings, which expounds the ideal of the
Dharma-following monarch.7 On the basis of the general tone of the Caturma-
hārājaparivarta it is fairly evident that the main intended function of this sūtra
was proselytizing, that is, attracting leaders of various regions to Buddhism by
offering them mutually beneficial services. The Buddhist community appears
to have been in perceptible need of securing support from the highest places;
some passages seem to indicate a degree of despair, as reflected in their accen-
tuation of vital threats for those not prepared to follow this tradition.8 It is
also not unlikely that Buddhism was at times persecuted, as certain references
indicate.9 Occasionally the text becomes somewhat guarded or equivocal, for
example in its claim that the Four Great Kings and the summoned deities arrive
in the king’s palace with invisible bodies,10 which suggests an effort by the text’s
compilers to achieve their goals in the most secure possible ways. As for the
ritual instructions in this chapter,11 it is prescribed that the king should clean
the palace, sprinkle the premises with perfumes and scatter flowers. He should
prepare an ornamented Dharma-throne and a lower seat for himself. He should
listen to this sūtra recited by a dharmabhāṇaka monk and honour those of the
Sangha who present the Suvarṇaprabhāsottama. As a result, the whole universe
will be adorned, lit up by golden light. The Four Great Kings and other divine
beings will approach the palace to listen to the recitation and will protect the
sovereign and his realm.
In chapter 7, the Sarasvatīparivarta,12 the goddess Sarasvatī grants her sup-
port to the Dharma preacher through the gift of eloquence, and presents a
bathing ritual with enchanted herbs for him and his audience in order to
appease all disturbances. It is promised that, invoked by praise, Sarasvatī herself
6 Suvarṇaprabhāsottamasūtra 6.2.6 (Skjaervø 2004): yaḥ kaścid bhadanta bhagavan manu-

ṣya-rājo bhavet ya-m-anena Devendra-samayena rāja-śāstreṇa rājatvaṃ kārayet.
7 Cf. also chapter 13 on King Susambhava.
8 Skjaervø 2004, 6.5.14–6.5.32. Chapter 12 also includes a description of grave dangers for
kings who fail to follow the Dharma; see Skjaervø 2004, 12.17–12.61.
9 Suvarṇaprabhāsottamasūtra 6.1.35–36 (Skjaervø 2004): manuṣyarājā … teṣāṃ sūtrendra-
dhārakāṇāṃ bhikṣuṇāṃ sarva-pratyarthikebhya ārakṣāṃ kuryāt.
10 Suvarṇaprabhāsottamasūtra 6.2.17, 6.2.37 (Skjaervø 2004): te vayaṃ bhadanta bhagavaṃś
catvāro mahā-rājānaḥ sa-bala-parivārā anekair yakṣa-śata-sahasrair adṛśyair kāyātma-
bhāvais tenopasaṃkramiṣyāmaḥ. It seems that the redactors wished to avoid the risk that
the listener expects these divine beings to come in a perceptible form, and then perhaps
be disappointed.
11 Skjaervø 2004, 6.3–6.4.
12 Skjaervø 2004, 7.1–7.67. For a detailed study of this chapter see Ludvik 2007, 145–221.
via free access
will appear and remove all diseases and difficulties. The ritual instructions pre-
scribe that one should pound herbs and consecrate the powder with mantras
at the time of the Puṣya constellation. A maṇḍala should be drawn with cow-
dung, flowers should be scattered and gold and silver vessels filled. Four armed
men and four well-adorned maidens should be placed there holding pots in
their hands, thus accomplishing the protective sealing of the boundaries (sīmā-
bandha). One should use incense, music, umbrellas, flags, banners, mirrors,
arrows, spears and dhāraṇī-spells, and in due course bathe behind an image
of the Buddha.13
In chapter 8, the Śrīparivarta,14 the goddess Śrī offers support to the Dharma
preacher and good fortune to his audience. A rite which provides prosperity is
described, through which Śrī herself enters that place. According to the ritual
instructions, one’s home must be purified and one should bathe and wear clean
garments. One should offer worship (pūjā) with perfumes, flowers and incense,
then sprinkle juices (rasa) and utter the names of Śrī, Ratnakusuma Tathāgata15
and the Suvarṇaprabhāsa. One should then recite dhāraṇī-spells, draw a maṇ-
ḍala of cow-dung and offer perfumes, flowers and incense. Finally, a pure seat
should be provided where Śrī descends and stays.
As can be seen, chapters 7 and 8 contain ritual instructions primarily for
securing health and wealth. These seem to be ancillary rites which accompany
the ritual for the protection of the state taught in chapter 6. Interestingly, these
rituals are actually more complex than those in the Caturmahārājaparivarta,
where recitation is the main focus, potentially indicating a somewhat later date
of composition. It is worth noting that chapters 9 and 10, the Dṛḍhāparivarta
and Saṃjñāyaparivarta, also offer further support and protection but do not
include detailed ritual instructions.
4 Dhāraṇī Literature Presenting Rites for the Protection of the State
Perhaps contemporaneous with the Suvarṇaprabhāsottamasūtra, there are at

least three pieces of dhāraṇī literature proper which include detailed ritual
instructions for the protection of the state: the Dhvajāgrakeyūradhāraṇī, Nārā-
yaṇaparipṛcchā/Mahāmāyāvijayavāhinīdhāraṇī and the Mahāsāhasrapramar-
13 Skjaervø 2004, lvi notes that the treatise on magical herbs and the dhāraṇīs may be later
additions to this chapter.
14 Skjaervø 2004, 8.1–8.71. On Śrī Lakṣmī see Shaw 2006, 94–109.
15 Note that a Ratnakusumasaṃpuṣpitagātra Tathāgata is listed in the shorter Sukhāva-
tīvyūha.
via free access
238 hidas
danasūtra.16 The former two are centered on this theme, while in the latter the
defence of the state is included as one among various rituals. In all three cases
the monarch is directly referred to as the beneficiary of the rites.
The Dhvajāgrakeyūradhāraṇī17 was edited by Giunta (2008) based on an
undated palm-leaf manuscript, and compared with the Tibetan translation by
Jinamitra, Dānaśīla and Ye shes sde of circa 800 CE.18 The Chinese translation
by Dānapāla/Shi-hu dates to 982CE.19 The setting of this scripture is the celes-
tial Trāyastriṃśat abode where Śakra requests help from the Buddha after the
gods have been defeated by the asuras, led by Vemacitrin. The lord tells Śakra
to learn the invincible Dhvajāgrakeyūradhāraṇī, which he had come to know
from Aparājitadhvaja Tathāgata during a previous existence. Then the Bhaga-
vān reveals the dhāraṇī and teaches that its user becomes victorious in battles
and conflicts. The spell is to be fixed atop a banner (dhvajāgra) or around the
neck, and protects kings (manuṣyarāja) or heroic people (śūrapuruṣa). Mani-
festing as a divine female, it stands in the frontline, providing fearlessness and
protection, driving the enemy away as well as granting blessings and prosper-
ity.20 Following the main text, the manuscript incorporates the brief Dhvajā-
grakeyūrahṛdaya “spell-essence.”
The Nārāyaṇaparipṛcchā/Mahāmāyāvijayavāhinīdhāraṇī21 was edited,
along with the Tibetan translation22 by Bstan pa’i nyin byed,23 by Banerjee
(1941) on the basis of a palm-leaf manuscript dated to 1361CE. No Chinese
translation seems to be available. The setting of this scripture is Mount Svar-
16 While the Chinese and Tibetan translations of these scriptures are relatively late, on the
basis of language and terminology it seems that all three date to the first half of the
first millennium. On the approximate antiquity of the Mahāsāhasrapramardanasūtra see
Hidas 2013, 229.
17 Cf. the Dhajagga-sutta in Pali (Saṃyutta-nikāya I.11.3), where the Buddha teaches that, just
as Sakka encouraged the gods to behold his banner or that of other deities when experi-
encing fear in a battle against the asuras, monks experiencing dread should call to mind
the Buddha, the Dharma and the Sangha.
18 E.g. Derge Kanjur 612 = 923. Cf. also Lhan kar ma No. 376 (Herrmann-Pfandt 2008, 215) and
Dunhuang IOL TIB J 372 (Dalton and van Schaik 2006, 106).
19 Taishō Tripiṭaka 943. Cf. also Taishō Tripiṭaka 1363 from 988CE.
20 Dhvajāgrakeyūradhāraṇī (Giunta 2008, 190): dhvajāgre kaṇṭhe vā baddhvā dhārayitavyā
| manuṣyarājñā śūrapuruṣāṇāñ ca sarvveṣā rakṣā karoti strīrūpadhāriṇī bhūtvā purataḥ
tiṣṭhati abhayaṃ dadāti | rakṣā karoti parasainya vidrāpati māṅgalyaṃ pavitra śrīlakṣmī
saṃsthāpikā. Cf. the Śrīparivarta of the Suvarṇaprabhāsottamasūtra.
21 Note that there exists another text titled Nārāyaṇaparipṛcchā, quoted, for example, in Śān-
tideva’s Śikṣāsamuccaya.
22 E.g. Derge Kanjur 684.
23 Known more widely as Si tu Paṇ chen (1700–1775).
via free access
ṇaśṛṅga,24 the mansion of Vaiśravaṇa, where Nārāyaṇa requests help from the
Buddha upon defeat by the asuras, much like the Dhvajāgrakeyūradhāraṇī.
The lord tells him that earlier, during the reign of King Ratnaśrī of Maga-
dha, there lived Sarveśvara Tathāgata, from whom the Bhagavān learned the
Mahāmāyāvijayavāhinī spell. For hundreds of thousands of years that king
ruled righteously by the power of this dhāraṇī. In his next existence he was
born as Māndhātā, a bodhisattva and cakravartin king,25 who practiced char-
ity for sixty-four thousand kalpas and became a buddha. Then the lord tells
Nārāyaṇa that this spell should be learnt, recited and taught to others. The
dhāraṇī should be fixed upon five models of chariots and placed across the
battlefield. Then the personified Queen of Spells (i.e. Mahāmāyāvijayavāhinī)
should be visualized there as devouring the enemy.26 At midday the king should
write down this incantation with saffron, after which he is to conquer the hos-
tile army.27 By reciting this dhāraṇī three times daily one is freed from even
the five sins of immediate retribution, gains heaps of merit and shall be able
to remember former existences. Laypeople, monastics, kings (rāja), princes
(rājaputra), Brahmins and Dharma preachers are all promised to benefit from
this spell.
The Mahāsāhasrapramardanasūtra was edited in Iwamoto 1937 based pri-
marily on a paper manuscript from 1553 (CE28), while its rituals are studied
in Hidas 2013. The Tibetan translation by Śīlendrabodhi, Jñānasiddhi, Śākya-
prabha and Ye shes sde dates to circa 800CE, and the Chinese one was com-
pleted by Dānapāla/Shi-hu in 983CE. The setting of this long and complex
24 The Suvarṇaśṛṅga is listed as a King of Mountains (parvatarājā) in the Mahāmāyūrīvid-

yārājñī (Takubo 1972, 49).
25 For an overview of the classical Buddhist ideal of cakravartin kingship see Harvey 2007.
On the vidyādhara-cakravartin of tantric Buddhism see Davidson 2002, 330–334.
26 For references to consuming the enemy cf. the Dhvajāgrakeyūrahṛdaya: oṃ bhagavati
dhvajāgrakeyūre parasainyavidhvaṃsanakari | svasainyaparipālanakari | ulkāmukhi | kha
kha | khāhi khāhi | parasainyam anantamukhenānantabhujena prahara prahara | hūṃ hūṃ
phaṭ phaṭ svāhā.
27 Nārāyaṇaparipṛcchā (Banerjee 1941, 4): nārāyaṇa atha tasmin samparāye senayor ubhayor
madhye pañcasu sthāneṣu etad dhāraṇīcakraṃ rathapratikṛtau yuñjyāt | nārāyaṇa ubha-
yor madhye parasenāgre tasmin rathamadhye mahāmāyāvijayavāhiṇīṃ nāma vidyārājñīm
anekaśatasahasrarūpām anekaśatasahasrabhujāṃ trinetrāṃ lohitakṛṣṇavarṇāṃ dīptaiś
caturvaktraiḥ parasenāṃ bhakṣayantīm iva cintayet | svayam eva cakravartyākāraṃ kṛtvā
madhyāhne dhāraṇīṃ kuṅkumena likhitvā rājā parasya ca senāṃ svabhāvena māyāsa-
dṛśena cchādayan vividhena sāreṇa vijayī bhaviṣyati.
28 Iwamoto (1937) consulted two undated paper manuscripts as well. All three codices are
Pañcarakṣā collections kept in Japan. Many thanks to Dr. Kenichi Kuranishi for his help
with the Japanese introduction to the edition.
via free access
240 hidas
scripture is Rājagṛha, where the Buddha and the Four Great Kings offer pro-
tection from calamities following an earthquake in Vaiśālī. Towards the end,
this sūtra provides detailed instructions for several rituals, including one for
the protection of the state. This prescribes that the royal residence (rājadhānī)
should be cleaned and purified with flowers, incense and other offerings. Four
maidens should be placed in the four directions with swords in their hands.29
The dhāraṇī should be recited and written on strips of cloth, mounted on the
top of caityas, trees and banners. Recitation should continue for a fortnight,
thus saving the state.30
5 Dhāraṇī Literature Making General Reference to the Protection of

the State
Various examples of dhāraṇī literature, in the wider sense, include references

to the defense of the state, while not being centered on this theme and lacking
detailed ritual prescriptions. Schopen (1978: 363–367) lists the Bhaiṣajyagu-
rusūtra, Prajñāpāramitā, Ratnaketuparivarta, Sarvatathāgatādhiṣṭhānavyūha
and Śrīmahādevīvyākaraṇa as relevant texts surviving among the Gilgit man-
uscripts (6th–7th centuries). The Bhaiṣajyagurusūtra claims that all distur-
bances to an anointed kṣatriya king, including hostile armed forces, shall cease
by performing worship (pūjā). The Prajñāpāramitā teaches that those who
study this text shall be protected in the battlefield. The Ratnaketuparivarta
declares that those kings who commit this spell-text to writing shall not face
any danger from an enemy’s army. If they raise the Ratnaketudhāraṇī on a
banner they shall be victorious in battle. Paying reverence to the Sarvatathā-
gatādhiṣṭhānavyūha will cause victory for kings and queens in war and riots.
Those monarchs who preserve the Śrīmahādevīstotra will face no calamities
in their realm, and Śrī herself will reside in their homes.31 In addition to these
scriptures, the Amoghapāśahṛdaya states that to protect one’s land from the
enemy or calamities the ritually pure officiant should fill a vessel with vari-
ous substances, perform great pūjā and recitation; peace shall follow thus.32
The Amoghapāśakalparāja teaches a couple of methods for success in warfare.
29 Cf. the Sarasvatīparivarta of the Suvarṇaprabhāsottamasūtra.

30 For a detailed translation see Hidas 2013, 236–237. See also Gentry 2016.
31 Cf. the Śrīparivarta of the Suvarṇaprabhāsottamasūtra.
32 Amoghapāśahṛdaya (Meisezahl 1962, 325–326): paraviṣayarājyarāṣṭropadravarakṣāsu
pūrṇṇakalaśaṃ sthāpayitvā śucinā śucivastraprāvṛtena mahatīṃ pūjāṃ kṛtvā vācayita-
vyam | mahāśāntir bhavati. Cf. also Amoghapāśakalparāja 3b.
via free access
By the use of a magical noose the army of the enemy and its weapons shall
be bound.33 By encircling a sword the weapons of the enemy shall be broken
and disabled.34 The Mahāpratisarāmahāvidyārājñī promises the destruction
of hostile armies and victory in battle with the help of its dhāraṇī.35 This scrip-
ture includes a testimonial narrative in which King Brahmadatta overpowers
the army of the enemy by fixing the spell upon his body before entering the
battlefield.36 In another story it is Śakra who defeats the asuras with the help
of this incantation.37
6 The Characteristics of Buddhist Rituals for the Protection of the

State in the First Half of the First Millennium
As the passages above show, the protection of the state and acquiring safety
in battle were recurring topics in Buddhist ritual literature from the early cen-
turies of the common era onwards. The target audience was primarily monar-
chs.38 Besides defence, the texts occasionally promise additional rewards as
well, most commonly health and wealth. While these incantation scriptures
principally aim at worldly goals, it is worth noting that in a few cases they
promise a better future existence, such as rebirth as a bodhisattva. Even bud-
33 Amoghapāśakalparāja 18b (Mikkyō Seiten Kenkyūkai. 1998–2011): sainyamadhyotkṣipet

sainyā pāśabandhā bhavanti | saṃgrāmamadhye kṣipet sarvvaśastrapraharaṇadhanu-
śaraśakti˘asimusalamusuṇḍicakrakunta˘āyudhavarmakavacā sarvve pāsabandhā bha-
vanti ||. 21b: saṃgrāmamadhye kṣipeta yasya nāmā kṣipyate sa ca pāsabaddhā agratam
upatiṣṭhati |. 30a: saṃgrāme yudhyamāne krodharājaṃ japatā pāśaṃ saṃgrāmama-
dhye kṣeptavyam | mahāntaṃ āśīviṣaṃ prādurbhavati | vikaṭākṣaṃ lolajihvaṃ aṅgāradīpti-
nayanaṃ saha darśanamātrāṇi caturaṅgaṃ balakāya diśividiśāni prapalāyante |.
34 Amoghapāśakalparāja 29a (Mikkyō Seiten Kenkyūkai. 1998–2011): ato vidyādhareṇa krud-
dhena khaḍga bhrāmayaṃ sarvvavighnavināyakānāṃ duṣṭayakṣarākṣasānāṃ sarvvabhū-
tānāṃ chinnā bhaviṣyati | sakṛtasaṃgrāmamadhye bhrāmayet sarvvayodhinānā⟨ṃ⟩m
āyudhānā cchinnā bhaviṣyanti | stambhitāni bhaviṣyanti | sarvve daśavidiśāni prapatā-
yanti |.
35 Hidas 2012, 205, 241, 243, 249.
36 Hidas 2012, 216–217. Note the setting similar to the Dhvajāgrakeyūradhāraṇī.
37 Hidas 2012, 227.
38 Note that monarchs were also addressed in a highly elevated literary style by prominent
Buddhist personalities such as Mātṛceta and Nāgārjuna in the first centuries CE (cf. e.g.
Zimmermann 2006, 228–229, Sanderson 2009, 103–104, Bronkhorst 2011, 103–107). While
works such as the Ratnāvalī lack the ritual instructions, as might be expected, their coun-
sel of righteous governance, however idealized, sometimes resonates with certain parts of
the Suvarṇaprabhāsottamasūtra.
via free access
242 hidas
dhahood is promised.39 These texts of both mainstream and Mahāyāna Bud-

dhism employ various ritual means, such as the use of mantra, dhāraṇī, maṇ-
ḍala and sīmābandha. It is difficult to suggest a precise chronology for the
emergence of these scriptures and, for example, to estimate whether the grand
Suvarṇaprabhāsottamasūtra appeared first, with independent dhāraṇī texts
following afterwards, or the reverse. What can be stated more or less with cer-
tainty is that all the aforementioned scriptures originate from the first half of
the first millennium, but emerged over an uncertain interval of time.
As far as actual users or evidence for the protection of the state in South Asia
is concerned, only indirect information is available. Various Chinese sources
refer to the successful use of the Suvarṇaprabhāsottamasūtra in some Indian
kingdoms as early as the first century CE, although these may only be legendary
accounts for the promotion of Buddhism in China.40 A more telling piece of evi-
dence, however, from the middle of the first millennium, is the Gilgit collection
of manuscripts, where royal patronage and apotropaic literature are closely
linked, albeit without surviving exemplars of the Sūtra of Golden Light.41 In the
southern island of Sri Lanka, where monastics and worldly rulers were deeply
interdependent for most of the centuries, we have no proof that the Sangha
provided protection for the state or that the king used the above mentioned
texts. There are however accounts from at least the fourth century CE onwards
reporting the performance of paritta-recitals with the Ratanasutta in times of
national calamities.42 Simultaneously, in East Asia we find plenty of direct evi-
39 See the Nārāyaṇaparipṛcchā.

40 Ludvik 2007, 150–151.
41 See von Hinüber 2014 and 2018. It is notable that in Bāṇa’s Harṣacarita (early-7th-century),
the Mahāmāyūrī (surviving in Gilgit in numerous manuscript copies) is mentioned as
being recited for healing at the royal palace: “So amid the salutations of the chamberlains
he slowly entered the palace. There he found people bestowing all their goods in presents,
worshipping the family gods, engaged in cooking the ambrosial posset, performing the Six
Oblation sacrifice, offering tremulous Durvā leaves besmeared with clotted butter, chant-
ing the Mahā-Māyūrī hymn, purifying the household, completing the rites for keeping out
the spirits by offerings. Earnest Brahmans were occupied in muttering Vedic texts; Śiva’s
temple resounded with the murmur of the Hendecad to Rudra; Śaivas of great holiness
were bathing Virūpākṣa’s image with thousands of vessels of milk.” (Cowell and Thomas
1897, 137).
42 Gunawardana 1979, 226–227; Norman 1983, 174. On paritta texts, the Ratanasutta/Ratnasū-
tra tradition and its remodelling as the Mahāsāhasrapramardanasūtra see Skilling 1992.
Cf. also Hidas 2013. On modern Thai amulet cultures incorporating the Ratanasutta see
McDaniel 2014, 143–144. On Southeast Asian aspects and the recitation of the Mahādib-
bamanta before going to battle see Skilling 2007, 195, and for a detailed study of this text
Jaini 1965. Note that Pali canonical sources acknowledge the efficacy of spells, but do not
have a high opinion of them: in the Kevaṭṭasutta (Dīgha-nikāya 11) the Buddha teaches
via free access
dence for protective practices employing dhāraṇī texts. De Visser (1935), May
(1967), Sango (2015) and Gummer (2015) provide a detailed picture how the
Suvarṇaprabhāsottamasūtra was actually used for the defense of the emperor
and the state. From these studies it becomes clear that Japanese ritual proce-
dures follow what is prescribed in the original Sanskrit remarkably closely;43 it
thus appears that the ‘periphery’ preserves well what may have once been the
norm in the native center of this tradition.
7 The Continuity of Buddhist Rituals for the Protection of the State

into Later Times
From the second half of the first millennium CE, new types of Buddhist rit-
ual texts incorporated the theme of state protection in South Asia within the
tantric traditions. Sanderson (2009, 105–106) refers to passages in the Sarvavaj-
rodaya of Ānandagarbha and the Guhyasamājamaṇḍalavidhi of Dīpaṅkara-
bhadra (9th c.) being used for the protection of the monarch, in connection to
rites of initiation. Sanderson (2009, 125) also observes that the Mañjuśriyamū-
lakalpa and Sarvadurgatipariśodhanatantra both offer protection to the ruler
through royal consecration. In the latter text protection is accomplished by
Vajradhara and the Four Great Kings, reflecting continuity with earlier sources.
As for chronicled accounts of the protection of royal dynasties, Sanderson
(2004, 238; 2009, 93–94) refers to Tāranātha’s history of Indian Buddhism,
which reports that upon seeing omens of the future ruin of the Pāla dynasty,
the eminent master Buddhajñānapāda of Vikramaśīla persuaded Dharmapāla
(r. circa 775–812) to institute a regular fire-sacrifice at the monastery in order
to protect his dynasty. The rituals lasted for many years at immense cost. In
another reference to Tāranātha, Sanderson (2009, 107) notes that tantric rit-
uals were often performed to avert the enemy, especially Turuṣkas. Based on
passages from the Mañjuśriyamūlakalpa and a grant by the Rāṣṭrakūṭa king
Govinda III, dated to 805 CE, Sanderson (2003–2004, 433–434 fn. 308) writes
that “[t]he practice of going into battle with an image of one’s personal deity
and the belief that this will protect one’s troops and confound those of the
that monks who display supernatural powers in order to promote faith in the Bhagavān
among people would be thought by some to have used incantations like the Gandhārīvijjā
or Maṇikāvijjā to achieve these accomplishments, rather than the power of their contem-
plation.
43 The Sūtra of Golden Light reached Japan as early as the end of the 6th century (Skjaervø
2004, xxxii).
via free access
244 hidas
enemy, is well attested in Indian Buddhist sources and in Far-Eastern sources

derived from them.” A few texts of this period also pick up on the theme of pro-
tection in warfare. The Siddhaikavīratantra introduces mantric amulets worn
on the body which, with help from enchanted swords, ensure victory in bat-
tle.44 The Laghusaṃvaratantra claims that one who enters the battlefield after
reciting a certain mantra one thousand times towards the enemy cannot be
hurt by weapons and obtains an indestructible vajra-body.45 Finally, one of the
latest and most complex tantric scriptures in South Asia, the Kālacakratantra,
mentions a number of war-machines ( yantra), presumably integrated into this
religious text because of the imminent threat from Western ‘barbarian’ invaders
in North India around the 11th century.46
8 The Protection of the State in Modern Nepal
As for recent practices in the Kathmandu Valley, there is evidence for the per-
formance of a royal ritual for protection involving Pañcarakṣā recitation in
the era of Mahindra Vira Vikram Shah (r. 1955–1972). As the officiant, Ratnarāj
Vajrācārya of Patan, kindly provided information, in 1962 he performed a cere-
mony in the royal palace for the protection of the monarch and his realm using
this influential apotropaic collection.47 In this part of South Asia, few early
manuscripts of the Suvarṇaprabhāsottama survive, though this sūtra became
part of the renowned Navadharma collection of nine texts in medieval times.48
44 Siddhaikavīratantra (Mikkyō Seiten Kenkyūkai. 1995, 7): kuṅkumādilikhitaṃ kaṇṭhe bāhau

mantraṃ dhārayan saṃgrāme śastraṃ stambhayati; saṃgrāme ’nusmaran mantraṃ vija-
yaṃ labhate; khaḍgam abhimantrya saṃgrāme praviśaṃ jayam āsādayati.
45 Laghusaṃvaratantra 12.4 (Pandey 2002): parasainyābhimukhaṃ sahasrajaptaṃ kṛtvā
saṃgrāme vā prativiśati. śastraśatair hanyamānasya vyathā notpadyate. na ca śastrasa-
hasrair bhidyate. vajraśarīraṃ bhavati. On the formation of the concept of vajrakāya see
Radich 2011.
46 See Grönbold 1996. Many thanks to Dr. Péter-Dániel Szántó for calling my attention to this
article. On Buddhism and warfare in a pan-Asian context through the ages see Jerryson
and Juergensmeyer 2010.
47 Personal communication, April 2009. Many thanks to the late Min Bahadur Shakya for
arranging a meeting with Ratnarāj Vajrācārya.
48 On the Navadharma or Navagrantha collection, consisting of the Prajñāpāramitā, Pañcar-
akṣā, Nāmasaṃgīti, Gaṇḍavyūha, Daśabhūmi, Samādhirāja, Saddharmapuṇḍarīka, Lali-
tavistara and Laṅkāvatāra in an earlier set, with the Suvarṇaprabhāsa and Tathāgataguh-
yaka replacing the Pañcarakṣā and Nāmasaṃgīti in its present form, see Tuladhar–
Douglas 2006, 144–147 and von Rospatt 2015, 819–821. The latter remarks that “[t]hese
canonical works are not so much studied for their content as liturgically recited or put
to other ritual uses.”
via free access
Various other scriptures of dhāraṇī literature, however, such as the Pañcarakṣā,

are available in several old palm-leaf witnesses, and we have colophons from
the wider region which indicate their use at the court.49
Acknowledgements
This research was funded by ERC Synergy Project No. 609823. I am grateful
to Dr. Michael Willis and Dr. Sam van Schaik for their support and Professor
Harunaga Isaacson for useful suggestions.
References
Banerjee, Anukul Chandra. 1941. Nārāyaṇaparipṛcchā: Sanskrit and Tibetan Texts. Cal-
cutta: University of Calcutta.
Bond, Kevin. 2014. “Buddhism on the Battlefield. The Cult of the ‘Substitute Body’ Tal-
isman in Imperial Japan (1890–1945).” In Material Culture and Asian Religions. Text,
Image, Object, edited by Benjamin J. Fleming and Richard D. Mann, 120–134. New
York, London: Routledge.
Bronkhorst, Johannes. 2011. Buddhism in the Shadow of Brahmanism. Leiden: Brill.
Cowell, Edward B. and Frederick W. Thomas. 1897. The Harṣa-Carita of Bāṇa. London:
Royal Asiatic Society.
Dalton, Jacob and Sam van Schaik. 2006. Tibetan Tantric Manuscripts from Dunhuang.
A Descriptive Catalogue of the Stein Collection at the British Library. Leiden: Brill.
Davidson, Ronald M. 2002. Indian Esoteric Buddhism: A Social History of the Tantric
Movement. New York: Columbia University Press.
Emmerick, Ronald E. 1996 [1970]. The Sūtra of Golden Light. Third revised edition.
Oxford: The Pali Text Society.
Gentry, James. 2016. Destroyer of the Great Trichiliocosm: Mahāsāhasrapramardanī.
https://read.84000.co/translation/toh558.html
Giunta, Paolo. 2008. “The Āryadhvajāgrakeyūrā nāma dhāriṇī. Diplomatic Edition of
MS Tucci 3.2.16.” In Sanskrit Texts from Giuseppe Tucci’s Collection, edited by Fran-
cesco Sferra, 187–194. Roma: Istituto italiano per l’Africa e l’Oriente.
49 For the use of the Pañcarakṣā by a Pāla queen in the 11th century see Pal 1992 (cf. also
Hidas 2012, 84–85). For modern talismanic cults related to warfare in Japan see Bond
2014, and for the protective Jinapañjaragāthā used in Thailand see McDaniel 2011, 77–
120.
via free access
246 hidas
Grönbold, Günter. 1996. “Kriegsmaschinen in einem buddhistischen Tantra.” In Fest-

schrift Dieter Schlingloff, edited by Friedrich Wilhelm, 63–97. Reinbek: Verlag für
Orientalistische Fachpublikationen.
Gummer, Natalie. 2015. “Suvarṇabhāsottamasūtra.” In Brill’s Encyclopedia of Buddhism.
Vol. I. Literature and Languages, edited by Jonathan Silk, Oskar von Hinüber and
Vincent Eltschinger, 249–260. Leiden: Brill.
Gunawardana, R.A.L.H. 1979. Robe and Plough. Monasticism and Economic Interest in
Early Medieval Sri Lanka. Tucson: University of Arizona Press.
Harvey, Peter. 2007. “Buddha and Cakravartins.” In Encyclopedia of Buddhism, edited by
Damien Keown and Charles Prebish, 153–155. London, New York: Routledge.
Herrmann-Pfandt, Adelheid. 2008. Die lHan kar ma: ein früher Katalog der ins Tibetische
übersetzten buddhistischen Texte. Kritische Neuausgabe mit Einleitung und Materi-
alien. Vienna: Austrian Academy of Sciences Press.
Hidas, Gergely. 2012. Mahāpratisarā-Mahāvidyārājñī, The Great Amulet, Great Queen of
Spells: Introduction, Critical Editions and Annotated Translation. New Delhi: Interna-
tional Academy of Indian Culture and Aditya Prakashan.
Hidas, Gergely. 2013. “Rituals in the Mahāsāhasrapramardanasūtra.” In Puṣpikā: Trac-
ing Ancient India through Texts and Traditions edited by Nina Mirnig et al., 225–240.
Oxford: Oxbow.
Hidas, Gergely. 2015. “Dhāraṇī Sūtras.” In Brill’s Encyclopedia of Buddhism. Vol. I. Lit-
erature and Languages, edited by Jonathan Silk, Oskar von Hinüber and Vincent
Eltschinger, 129–137. Leiden: Brill.
Hinüber, Oskar von. 2014. “The Gilgit Manuscripts: An Ancient Buddhist Library in
Modern Research.” In From Birch Bark to Digital Data: Recent Advances in Buddhist
Manuscript Research. Papers Presented at the Conference Indic Buddhist Manuscripts:
The State of the Field, Stanford, June 15–19, 2009, edited by Paul Harrison and Jens-Uwe
Hartmann, 79–135. Vienna: Austrian Academy of Sciences Press.
Hinüber, Oskar von. 2018. “Magic Protection in the Palola Ṣāhi Kingdom: History and
Context of Rakṣā Texts and Dhāraṇīs in 7th Century Gilgit.” In Katā Me Rakkhā, Katā
Me Parittā. Protecting the Protective Texts and Manuscripts. Proceedings of the Sec-
ond International Pali Studies Week, Paris 2016, edited by Claudio Cicuzza, 217–237.
Bangkok and Lumbini: Fragile Palm Leaves Foundation and Lumbini Research Insti-
tute.
Iwamoto, Yutaka. 1937. Mahāsāhasrapramardanī. Kyoto.
Jaini, Padmanabh S. 1965. “Mahādibbamanta: A Paritta Manuscript from Cambodia.”
Bulletin of the School of Oriental and African Studies 28: 61–80.
Jerryson, Michael K. and Mark Juergensmeyer, eds. 2010. Buddhist Warfare. Oxford: Uni-
versity Press.
Ludvik, Catherine. 2007. Sarasvatī: Riverine Goddess of Knowledge. From the Manu-
script-Carrying Vīṇā-Player to the Weapon-Wielding Defender of the Dharma. Leiden:
Brill.
via free access
May, Jacques. 1967. “Chingokokka.” In Hōbōgirin IV. Paris; Tokyo: 322–327.

McDaniel, Justin. 2011. The Lovelorn Ghost and the Magical Monk: Practicing Buddhism
in Modern Thailand. New York: Columbia University Press.
McDaniel, Justin. 2014. “The Material Turn. An Introduction to Thai Sources for the
Study of Buddhist Amulets.” In Material Culture and Asian Religions. Text, Image,
Object, edited by Benjamin J. Fleming and Richard D. Mann, 135–148. New York, Lon-
don: Routledge.
Meisezahl, Richard O. 1962. “The Amoghapāśahṛdaya Dhāraṇī. The Early Sanskrit Man-
uscript of the Reiunji Critically Edited and Translated.” Monumenta Nipponica 17:
265–328.
Mikkyō Seiten Kenkyūkai. 1995. “Siddhaikavīratantra.” Taishō Daigaku Sōgō–Bukkyō-
Kenkyūsho-Kiyō 15: 366–349 and 16: 1–9.
Mikkyō Seiten Kenkyūkai. 1998–2011. “Transcribed Sanskrit Text of the Amoghapāśa-
kalparāja.” Taishō Daigaku Sōgō–Bukkyō-Kenkyūsho-Nenpō 20–23, 26, 32, 33.
Nobel, Johannes. 1937. Suvarṇabhāsottamasūtra. Das Goldglanz-Sūtra. Ein Sanskrittext
des Mahāyāna-Buddhismus. Leipzig: Otto Harrassowitz.
Norman, Kenneth Roy. 1983. Pāli Literature. Including the Canonical Literature in Prakrit
and Sanskrit of all the Hīnayāna Schools of Buddhism. Wiesbaden: Otto Harras-
sowitz.
Pal, Pratapaditya. 1992. “Queen Uḍḍākā’s Illuminated Manuscript.” In Indian Numismat-
ics, History, Art and Culture: Essays in Honour of Dr. Parameshwari Lal Gupta, edited
by David W. Macdowall, et al., 291–296. Delhi: Agam Kala Prakashan.
Pandey, J.S. 2002. Śrīherukābhidhānam Cakrasaṃvaratantram with the Commentary of
Bhavabhaṭṭa. Sarnath: Central Institute of Higher Tibetan Studies.
Radich, Michael. 2011. “Immortal Buddhas and Their Indestructible Embodiments: The
Advent of the Concept of Vajrakāya.” Journal of the International Association of Bud-
dhist Studies 34: 227–290.
Rospatt, Alexander von. 2015. “Local Literatures: Nepal.” In Brill’s Encyclopedia of Bud-
dhism. Vol. I. Literature and Languages, edited by Jonathan Silk, Oskar von Hinüber,
and Vincent Eltschinger, 819–830. Leiden: Brill.
Sander, Lore. 2007. “Preliminary Remarks on Two Versions of the Āṭānāṭīya (Āṭānāṭika)-
Sūtra in Sanskrit.” Journal of the International College for Advanced Buddhist Studies
11: 159–196.
Sanderson, Alexis. 2003–2004. “The Śaiva Religion Among the Khmers, Part I.” Bulletin
de l’École française d’Extrême-Orient 90–91: 349–463.
Sanderson, Alexis. 2004. “Religion and the State: Śaiva Officiants in the Territory of the
Brahmanical Royal Chaplain.” Indo-Iranian Journal 47: 229–300.
Sanderson, Alexis. 2009. “The Śaiva Age—The Rise and Dominance of Śaivism Dur-
ing the Early Medieval Period.” In Genesis and Development of Tantrism, edited by
Shingo Einoo, 41–350. Tokyo: Institute of Oriental Culture, University of Tokyo.
via free access
248 hidas
Sango, Asuka. 2015. The Halo of Golden Light. Imperial Authority and Buddhist Ritual in
Heian Japan. Honolulu: University of Hawaii Press.
Schopen, Gregory. 1978. The Bhaiṣajyaguru-sūtra and the Buddhism of Gilgit. Ph.D. the-
sis, Australian National University.
Shaw, Miranda. 2006. Buddhist Goddesses of India. Princeton, NJ: Pirnceton University
Press.
Skilling, Peter. 1992. “The Rakṣā Literature of the Śrāvakayāna.” Journal of the Pali Text
Society 16: 109–182.
Skilling, Peter. 2007. “King, Sangha and Brahmans: Ideology, Ritual and Power in Pre-
modern Siam.” In Buddhism, Power and Political Order, edited by Ian Harris, 182–215.
Abingdon; New York: Routledge.
Skjærvø, Prods Oktor. 2004. This Most Excellent Shine of Gold, King of Kings of Sūtras:
The Khotanese Suvarṇabhāsottamasūtra. Cambridge, MA: Department of Near East-
ern Languages and Civilizations, Harvard University.
Suzuki, Takayasu. 2012. “Who Is the One That Has to Make Confession under the
Instruction of the Suvarṇaprabhāsa?” Journal of Indian and Buddhist Studies 60:
1220–1228.
Takubo, Shuyo. 1972. Ārya-Mahā-Māyūrī Vidyā-Rājñī. Tokyo: Sankibo.
Tuladhar-Douglas, William. 2006. Remaking Buddhism for Medieval Nepal: The Fif-
teenth-Century Reformation of Newar Buddhism. Abingdon; New York: Routledge.
de Visser, Marinus Willem. 1935. Ancient Buddhism in Japan: Sūtras and Ceremonies in
Use in the Seventh and Eighth Centuries A.D. and Their History in Later Times. Leiden:
Brill.
Zimmermann, Michael. 2006. “Only a Fool Becomes a King: Buddhist Stances on Pun-
ishment.” In Buddhism and Violence, edited by Michael Zimmermann, 213–242.
Lumbini: Lumbini International Research Institute.
via free access
chapter 11
Adapting Śaiva Tantric Initiation for Exoteric

Circles: The Lokadharmiṇī Dīkṣā and Its History in
Early Medieval Sources
Nina Mirnig
1 Introduction
A consistent concern in the study of early Śaivism is to understand the scope

of the clientele texts were written for, and how representative this literature
was of practices on the ground at given times. The methodological limita-
tions we face in exploring this largely prescriptive body of literature in relation
to actual practice are notorious. Yet, in his seminal works on Śaiva religious
history, Alexis Sanderson developed groundbreaking insights into the socio-
religious and political landscape in which the surviving texts were produced,
reconstructing processes that led to the dominant position of Śaivism in the
medieval Indic world. Thanks to his pioneering research, and that of other
scholars who have taken up the subject in the wake of his work, our understand-
ing of the relative chronology of Śaiva scriptures and medieval authors has also
greatly improved.1 This increasingly enables us to investigate these materials
historically, and thus to pursue questions of a socio-historical nature, which
are at the heart of Sanderson’s work. In his own words (Sanderson 2005, 230),
[…] it is possible, I would say necessary, to read the literature and inscrip-
tions with the sort of questions in mind that a social historian would wish
to ask.
For this felicitation volume, the present article is intended as a small contri-
bution of this kind, applying a social-historical approach to a certain form of
Śaiva initiation, namely the lokadharmiṇī dīkṣā. In essence, this denotes a form
of initiation that is compatible with maintaining one’s socio-religious and rit-
ual obligations in society—the lokadharma.
1 See, e.g., Goodall 1997, xxxvi–lxxvi, and the preface and introduction to Goodall 2004, as well
as Sanderson 2001 and 2014.
© nina mirnig, 2020 | doi:10.1163/9789004432802_013

via free access
250 mirnig
Scholars familiar with tantric traditions know well that initiation (dīkṣā) is
one of the core tantric rituals. As Goodall explains, tantric initiation was novel
and different from earlier Indic forms of religious initiation, being not only “a
preparation for a particular religious undertaking, but … also a transformative
rite that purifies the soul.”2 This purification is achieved through an innova-
tive ritual technology, using the power of mantras to destroy an individual’s
karmas and connected rebirths at all reality levels (tattvas). This enables the
Śaiva Guru to free the bound soul of all ties, and to guide the candidate to
a cosmic level where he can either realize liberation through union with the
highest form of Śiva, if he is a liberation seeker (mumukṣu), or attain divine
powers, if he is a seeker of supernatural powers and enjoyments (bubhukṣu).
As such, tantric initiation confers spiritual benefits to the candidate a priori
to his practice. It is thus arguably the most powerful tantric rite, designed in
principle for practitioners who intend to devote their lives to the religion, as
was probably the case in its original setting of esoteric, probably ascetic, cir-
cles.
At the same time, we know from history that tantric Śaivism did not remain
confined to the margins of society, but emerged as a religious movement which
successfully interfaced with the mainstream, and gradually came to dominate
the religious and socio-political discourse of the early medieval Indic world.3
As one would expect, this development is reflected in ritual and in changes to
the constitution of the initiatory community, which by then encompassed not
only ascetic practitioners fully devoted to Śaiva religious practice, but also brah-
manical householders. The increasing engagement with the mainstream led to
the creation of exoteric forms of initiation, including versions that would con-
fer spiritual benefits without requiring the candidate to adopt a purely Śaiva
ritual lifestyle. The most well-known example is the nirbījā dīkṣā (“initiation
without the seed [of having to perform post-initiatory rites]”), which offers the
candidate the highest soteriological goal of ultimate liberation at the time of
death. However, according to the scriptures, such an initiation was only granted
in special cases, namely when the candidates were considered unable to per-
form any kind of post-initiatory practice for special reasons. The list of such
people classically comprises the king, but also the old, the young, the sick, and
women.4
2 See Goodall’s entry on dīkṣā in Tāntrikābhidhānakośa, vol. III.

3 See, e.g., Sanderson 2009.
4 See, e.g., Svacchanda, 4.88, as quoted by Goodall in his entry on nirbījadīkṣā in Tāntrikābhi-
dhānakośa, vol. III.
via free access
adapting śaiva tantric initiation 251
The other, lesser-known form of exoteric initiation is the lokadharmiṇī dīkṣā,

which forms the subject of this paper. This term translates as “the initiation
which contains/adheres to the lokadharma,” the lokadharma denoting the
sphere of observance that Śaiva sources consistently define as the exoteric
religion of śruti and smṛti, that is to say the domain of the brahmanical house-
holder.5 Unlike the nirbījā, access to the lokadharmiṇī initiation is not limited
to specific groups, and in its original context, the initiation is mainly benefit-
oriented: a lokadharmī initiate is said to pursue mundane spiritual goals before
attaining some kind of divine status. He does so by means of the merit procured
through lay worship, and not through Śaiva rites that require the propitiation
of mantras. This seemingly antithetical initiatory category is little discussed
by Śaiva authors, probably precisely because of the doctrinal implications of
such an initiation, which foregrounds a level of practice and spiritual goals
that Śaivas claim to surpass. In the course of time, the lokadharmiṇī initiation
underwent several shifts in application. These developments are deeply entan-
gled with the complex history of the system of initiatory categories, which was
subject to processes of division and reassignment reflecting local and histor-
ical circumstances. A historical investigation into this form of exoteric Śaiva
initiation therefore touches upon the larger issue of how tantric initiatory com-
munities related to the brahmanical mainstream, at various stages, after their
emergence from esoteric ascetic circles.
The first part of the paper investigates the lokadharmiṇī in its original set-
ting, as an initiation for Sādhakas, first by surveying the available sources and
then by contextualizing them in the socio-religious setting of tantric Śaivism’s
early stages. The second part will examine how this category of exoteric bubhu-
kṣu initiation is subsequently re-assigned to the sphere of the mumukṣu, thus
becoming integrated into the Śaiva soteriological path, especially in the Jaya-
drathayāmala. The third section looks into the usage of the lokadharmī cate-
gory in the ritual manuals of the 11th–13th centuries, particularly sources from
the South, tracing how in some works, notably the Jñānaratnāvalī, there is
another shift in usage: the lokadharmiṇī dīkṣā comes to denote a regular Śaiva
initiation for the householder, as opposed to the ascetic, who receives a śiva-
dharmiṇī dīkṣā. The final section reflects upon how the initiatory category of
the lokadharmiṇī fits within the larger context of tantric Śaivism’s development
in medieval India.
5 For an example, see Kṣemarāja ad Svacchanda 4.85: lokamārgaḥ śrutisṃrtyācāraḥ …, also

quoted in note 10.
via free access
252 mirnig
2 The Original lokadharmiṇī dīkṣā as a Form of Sādhaka Initiation:

Sources and Mechanics
The earliest sources that mention a lokadharmiṇī dīkṣā—always as the counter-

part of the śivadharmiṇī dīkṣā—feature it as a form of Sādhaka initiation6 in the
domain of the bubhukṣu, i.e. the seeker of power and enjoyment, as opposed
to the liberation-seeker (mumukṣu). The first extant text to feature the loka-
dharmiṇī is the Svacchanda. In its fourth chapter, dedicated in its entirety to
initiation, the text gives an overview of all the initiation categories, dividing the
initiatory community into mumukṣus and bubhukṣus. The former may receive
either a sabījā or nirbījā initiation, and the latter a śivadharmiṇī or lokadha-
rmiṇī initiation.7 Of these, the śivadharmī Sādhaka is the Śaiva Sādhaka proper,
who through initiation gains the ability to propitiate Śaiva mantras in order to
attain benefits and supernatural powers, while the lokadharmī Sādhaka is his
exoteric counterpart, seeking enjoyments and spiritual benefits through mun-
dane religious acts rather than mantras:
The Sādhaka is of two kinds. On the one hand, there is the śivadharmī,
for whom the cosmic path is purified by Śaiva mantras and who is yoked
to [particular] mantras that are to be mastered; he is knowledgeable,
consecrated [to office], and devoted to the propitiation of mantras. This
Śaiva Sādhaka is capable [of mastering] the threefold supernatural pow-
ers.8 The second [kind of Sādhaka] adheres to the mundane path and
6 Brunner, in her 1975 paper dedicated to the figure of the Sādhaka, also outlines many of the
general features of the Śaiva Sādhaka’s initiation and practice thereafter. She does, however,
not treat the lokadharmiṇī kind of initiation at length.
7 Svacchandatantra 4. 79b–81b: atha dīkṣādhvaśuddhyarthaṃ bhuktimuktiphalārthinām || vi-
dhānam ucyate sūkṣmaṃ pāśavicchittikārakam | guruḥ saṃpṛcchate śiṣyaṃ dvividhaṃ pha-
lakāṅkṣiṇam || phalam ākāṅkṣase yādṛk tādṛk sādhanam ārabhe |, “Next there is the initiation
for the purpose of the purification of the cosmic path (adhvan) for those who seek the fruit
of [either] enjoyment or liberation. The subtle method that causes the cutting of the bonds
is explained. The Guru asks the candidate seeking benefits [about] the two-fold [option].
Whatever fruit he desires, accordingly he should start the propitiation of Mantras.”
8 This may be a reference to the old classification of siddhis into three kinds, which goes back
to early sources such as the Niśvāsa and Buddhist Kriyātantras, as Goodall (2014, 16 and 80–
82) discusses. Thus, in the Niśvāsa Guhyasūtra three levels of siddhi can be attained after
preparing some potion. Which kind of siddhi has been obtained can be deduced from the
manifestation of heat, if the power is to cover great distances on foot, the manifestation of
smoke if he is to have the power to disappear, and the manifestation of flames if he is to have
the power to walk through the sky (Goodall 2013, 81). For more on siddhi-practices in Tantric
Śaivism, see Vasudeva 2012.
via free access
is devoted to the performance of good and meritorious works (iṣṭāpūr-

tavidhau); desiring the fruits produced by [his] karma, he abides solely
[devoted to] meritorious [karma], free of the unmeritorious. [The Guru]
should always perform the destruction of the unmeritorious portion [of
the candidate’s karma] with mantras.9
Kṣemarāja, in his commentary, further specifies that meritorious acts in the

case of the lokadharmī consist of bathing at sacred places, providing food and
building wells, tanks and monasteries.10 These are precisely the kinds of reli-
gious activity that we would expect of an ordinary lay worshipper. As we see
in this passage, what the lokadharmiṇī initiation does is to enhance the initi-
ate’s spiritual benefits by destroying bad karma connected to all reality levels,
leaving only good karma and its positive outcomes in place.
This function of the lokadharmiṇī dīkṣā is reiterated later in the chapter,
when the text explains how the two kinds of bubhukṣu initiations differ from
that of the mumukṣu. These differences concern two points:11 firstly, the portion
of karma that is purified and thus removed for the candidate; and second, the
cosmic level at which the candidate is united with the respective deity upon
initiation’s conclusion:
Next, he should then bring about destruction of the past and future karma
for the liberation-seeker (mumukṣu), because of his indifference [to the
world]. He should not purify the one [part of karma] that is the prāra-
bdha [karma], [which fuels his present existence]. But for the Sādhaka, he
should purify [only] one [part of the] past karma for the purpose of power,12
and having manifested the past and future karma together (ekastham),
he should initiate [the candidate]. This is the śivadharmiṇī dīkṣā. The
9 Svacchanda 4.83–86b: sādhako dvividhaḥ tatra śivadharmy ekataḥ sthitaḥ | śivamantra-

viśuddhādhvā sādhyamantraniyojitaḥ || jñānavāñ cābhiṣiktaś ca mantrārādhanatatparaḥ
| trividhāyāḥ tu siddher vai so ’trārhaḥ śivasādhakaḥ || dvitīyo lokamārgastha iṣṭāpūrtavi-
dhau rataḥ | karmakṛt phalam ākāṅkṣañ śubhaikastho ’śubhojjhitaḥ || tasya kāryaṃ sadā
mantrair aśubhāṃśavināśanam |.
10 Kṣemarāja ad Svacchanda 4.85: lokamārgaḥ śrutismṛtyācāraḥ. iṣṭaṃ tīrthasnānānnadā-
nādi. pūrtaṃ kūpataḍākamaṭhādi, “The mundane path is the observance according to
śruti and smṛti. The sacred rites (iṣṭam) [consist of] such actions as bathing at a sacred
site and giving away food. The meritorious acts (pūrta) are [the donations and setting up
of] such things as wells, tanks and monasteries for ascetics.”
11 See also, Brunner 1975, 417–420.
12 See pp. 254 ff. for a discussion of this problematic passage, here italicized.
via free access
254 mirnig
other form [of bubhukṣu initiation] is the lokadharmiṇī, which destroys

both past and future demerit. That lokadharmiṇī [dīkṣā] is known to
exclude the obligation to propitiate mantras [by means of purvasevā
etc.]. However, when the current body breaks, [the candidate] expe-
riences [the series of eight supernatural natural powers] starting with
becoming very small (aṇimā). Having experienced [these] enjoyments he
moves upwards to whichever [cosmic level] the Guru has joined him [by
yojanikā]. Whether this is at the sakala or niṣkala level [of Śiva] depends
on [the preference of] the candidate and Guru.13
Thus, we learn that the lokadharmī candidate can be united with any deity
of any cosmic level at the end of the procedure, and could even choose to be
united with the highest, niṣkala form of Śiva, a form of union synonymous with
the state of ultimate liberation. In addition, while the lokadharmī initiate can-
not propitiate Śaiva mantras, unlike the Sādhaka, he is nevertheless granted
supernatural powers, which are usually said to result from mantra practice; the
difference is that he experiences these powers after death.
Regarding the śivadharmiṇī dīkṣā, we note that there is some difficulty inter-
preting how karma is held to be eliminated at this point. As evident from
the translation, the problematic phrase is 142ab, sādhakasya tu bhūtyarthaṃ
prākkarmaikaṃ tu śodhayet; this is supposed to define the difference in proce-
dure from the nirvāṇadīkṣā of the mumukṣu, for whom only prārabdha karma,
the karma that fuels the current life, is to be preserved.14 Looking at the differ-
ent treatments of this phrase, we find that commentators disagree on whether
or not any additional karmic bonds are excluded from purification—and in
effect, whether the dīkṣā of the Sādhaka is very close to that of the mumukṣu
Putraka or not. Kṣemarāja, in his commentary, wants to see a clear difference,
and interprets this passage to indicate that two kinds of karma are excluded
from purification, in case of the Sādhaka, namely (1) the karma that is nec-
essary for the attainment of supernatural powers (i.e. bhūtyarthaṃ), without
13 Svacchanda 4.141–145: prākkarmabhāvikasyātha abhāvaṃ bhāvayet tadā | mumukṣor nira-

pekṣatvāt prārabdhr ekaṃ na śodhayet || sādhakasya tu bhūtyarthaṃ prākkarmaikaṃ tu
śodhayet | prākkarmāgāmi caikasthaṃ bhāvayitvā ca dīkṣayet || śivadharmiṇy asau dīkṣā
lokadharmiṇy ato ’nyathā | prāktanāgamikasyāpi adharmakṣayakāriṇī || lokadharmiṇy
asau jñeyā mantrārādhanavarjitā | prārabdhadehabhede tu bhuṅkte sa hy aṇimādikān ||
bhogān bhuktvā vrajed ūrdhvaṃ guruṇā yatra yojitaḥ | sakale niṣkale vāpi śiṣyācāryavaśād
bhaved ||.
14 For the locus classicus regarding the mechanics of prārabdha karma, see Kiraṇa 6.20–21,
edited and translated in Goodall 1998.
via free access
making explicit what that means,15 and (2) the good portion of the prārab-
dha karma, which ensures the success of the siddhi practice by preventing
its being blocked by bad karma.16 On the other hand, Abhinavagupta, when
adopting this passage in his Tantrāloka, as discussed below, drops pādas 142cd
and rephrases 142ab as prākkarmetthaṃ tu śodhayet, “He should purify the past
karma in the same way [as in the case for the mumukṣu]”—thus indicating that
for him, the procedure for the Sādhaka is exactly the same as for the mumukṣu
Putraka, except that the prārabdha karma is directed towards supernatural
powers rather than liberation. In his commentary on this passage, Jayaratha,
in turn, criticizes Kṣemarāja’s reading of ekam instead of ittham, undermining
his understanding that the bad portion of the prārabdha karma is destroyed,
instead saying that neither is it possible to remove any part of the prārabdha
karma, since this fuels the current life force, nor does this interpretation have
scriptural support.17 At the same time, he remains silent about the fact that
the text teaches the removal of only the bad karma in case of the lokadharmiṇī
dīkṣā.18
15 Kṣemarāja ad Svacchanda 4.142a (sādhakasya tu bhūtyarthaṃ): bhūtiproyojanaṃ bhāvi-

mantrārādhanarūpaṃ yat tad api na śodhayed ity arthaḥ, “He should not purify the [por-
tion of karma] which is for the future propitiation of mantras for the purpose of super-
natural powers; this is the meaning.”
16 Kṣemarāja ad Svacchanda 4.142b (prākkarmaikaṃ tu śodhayet): bhūtyarthād yat prāg
dehārambhi śubhāśubhaṃ tata ekam aśubham evāsya śodhayet, evaṃ hi nirvighnaṃ bho-
gasiddhir bhavati || yat tu janmāntarasaṃcitaṃ śubhāśubhaṃ yac cāsmiñ janmani kari-
ṣyati, tat sarvam uktanītyā mantrārādhanavarjam asya putrakavac chodhyam evety āha,
“For the purpose of supernatural powers, from the past pure and impure [karma] that
maintains the [current] body (i.e. the prārabdha karma) he should only purify the impure
[portion] for him, for in this way (evam) the achievement of enjoyment comes about with-
out any obstacles. As for the pure and impure [karma] that is accumulated in other [past]
births and which he will do in a [future] birth, all those should be purified for him accord-
ing to the proclaimed procedure, like in the case of the Putraka, apart from [those karmas
for] the propitiation of mantras (mantrārādhanavarjam). Therefore he said, [prākkar-
māgāmi caikasthaṃ bhāvayitvā ca dīkṣayet (Svacchanda 4.142cd)].”
17 Jayaratha ad Tantrāloka 15.28: sādhakasya ca ittham eva karmaśodhanaṃ, kiṃ tu tat bhū-
tyarthaṃ, tasya hi bhogaunmukhyāt tadvāsanābhedāt phalabheda iti pratijñāto ’pi artho
nirvāhitaḥ | yad uktaṃ “sādhakasya tu bhūrtyarthaṃ prākkarmetthaṃ tu śodhayet | prākka-
rmāgāmi (corr. Hatley; prākkarmagāmi Ed.) caikasthaṃ bhāvayitvā tu dīkṣayet” iti. atra ca
udyotakṛtā yad ittham iti apāsya ekam iti paṭhitvā dehārambhiśubhāśubhakarmamadhyād
ekam aśubham api asya śodhayed iti vyākhyātaṃ, tad upekṣyam ārabdhakāryadehāramb-
hikarmocchedāśakyatvasya prāgupapāditatvāt kva cid apy evam anāmnātatvāc ca.
18 A similar problem of interpretation also occurs in the passage of the Mṛgendra, which
outlines the superior and the inferior bhautikī dīkṣā, the former being the śivadharmiṇī
and the latter the lokadharmiṇī. This passage appears to say that for the superior kind, i.e.
the śivadharmiṇī, along with the prārabdha karma some undefined additional karma is
via free access
256 mirnig
Another early source which distinguishes between the śivadharmiṇī and

lokadharmiṇī dīkṣā is the Mṛgendra, which was composed after the Svaccha-
nda. This scripture maintains the same basic set of categories and organiza-
tion of initiatory groups, even though it uses slightly different terminology.
Thus, instead of referring to initiations for the mumukṣu and the bubhukṣu,
the Mṛgendra uses the terms naiṣṭhikī and bhautikī,19 the former of which is
divided into the sāpekṣā and nirapekṣā,20 corresponding to the sabījā/nirbījā
distinction, and the latter again into śivadharmiṇī and lokadharmiṇī, for which
we find the following description:
preserved for the success of the siddhi practice (śaivasādhanasādhyena), perhaps similar
to Kṣemarāja’s bhūtyartham category. Mṛgendra 8.1146–148: evam eva kriyāyogād bhautiky
api parāparā | kiṃ tu dehāviyogārthaṃ prārabdhaṃ karma dehinaḥ || śaivasādhanasā-
dhyena saṃdhāya paripālayet | sānubandhaṃ dahed anyad dīpte dīkṣāhutāśane || kaṣṭā-
kaṣṭapadāntasthe sādhake pañcakeśvare | yogo ’syātmikadeśinyām anyasyām iṣṭavigrahe ||
(for a translation, see Brunner 1985, 258). However, Bhaṭṭa Nārāyaṇakaṇṭha, comment-
ing on this passage, is against such an interpretation, and rather awkwardly interprets
śaivasādhanasādhyena to refer to the post-initiatory discipline that the Sādhaka is to carry
out, which would seem odd given that such post-initiatory practice is also to be carried
out by the mumukṣu: … prārabdhaṃ karma prārabdhakāryaṃ dehinaḥ jantoḥ dehāviyogā-
rthaṃ śarīrapātasaṃrakṣārthaṃ śaivasādhanasādhyena samayācārapālanena saṃdhāya
saṃyojya paripālayet rakṣayet. (For a translation, see Brunner 1985, 286).
19 That the terms bhautikī and naitṣṭhikī correspond at this point to the initiations for the
bubhukṣu and mumukṣu respectively, is suggested by the commentary of Nārāyaṇakaṇ-
ṭha (ad Mṛgendra, Kriyāpāda 2: niṣṭhā kaivalyam. te bhūtiniṣṭhe prayojanaṃ yayos te tathā.
bhautikī bhūtikāmasya. niṣṭhārthino naiṣṭḥikī nairvāṇikīty arthaḥ), which also Brunner fol-
lows in her translation, translating bhautikī as “donneuse de jouissance” and naiṣṭhikī as
“donneuse de liberation” (Brunner 1985, 199). Further, from later passages it is clear that
the bhautikī is equated with the Sādhaka; e.g. in Mṛgendra, Caryāpāda 11, where the loka-
dharmiṇī is even referred to as a Sādhaka. As such, this division of the Mṛgendra would
reproduce the same structure as we found it in the Svacchanda; but internally we have
to note that the category of the bhautikī and the naiṣṭhikī, if taken in this manner, don’t
overlap with what in the Cāryapāda of the Mṛgendra is described as bhautikavratins and
naiṣṭhikavratins, as Brunner (1985, 347, note 2) already points out. Thus, here, the bhau-
tikavratin, i.e. one who has a temporary vrata, and the naiṣṭhikavratin, who has a perma-
nent vrata, are both opposed to those who have no vrata at all (avrata), a category which,
in turn, includes the lokadharmī sādhaka featuring under the bhautikī dīkṣā. However,
it may be that the Mṛgendra is simply not entirely consistent. Interpretative problems
appear also in other places regarding the definitions and procedures for the śivadharmiṇī
and lokadharmiṇī, and similarly, perhaps confusion also remains on the bhautikī and
naiṣṭhikī distinction. Goodall also draws attention to the difficulties in interpreting the
terms naitṣṭhika and bhautika in his entry on the term naiṣṭhika in Tāntrikābhidhānakośa,
vol. III.
20 However, note that in her English summary it seems that Brunner (1985, 481–482) applies
the divisions of sāpekṣā and nirapekṣā to both the naiṣṭhikī and bhautikī.
via free access
The śivadharmiṇī [initiation] is the root of success for the fruits of the
Śaiva religion for the individual soul. There is another [kind of śiva-
dharmiṇī] taught without the destruction of the body, up until the disso-
lution of the world.21 The remaining one is taught to be the lokadharmiṇī,
for the purpose of [attaining the eightfold supernatural powers] starting
with aṇimā after the current life, after all the bad portions [of karma] were
destroyed at all reality levels.22
While the description of the śivadharmīṇī initiation differs somewhat, imply-

ing that there are two levels of Śaiva Sādhaka,23 the explanation of the loka-
dharmiṇī resonates much with what we found in the Svacchanda, namely that
all bad karma is destroyed and the initiate attains supernatural powers after
death. Regarding the end point of union for the lokadharmī candidate, the
Mṛgendra offers a choice similar to that of the Svacchanda, namely that the
candidate is either lifted up to the cosmic level of his choice and absorbs the
applicable deity’s divine powers, or that he is united with the highest form of
Śiva, for liberation:
Having lifted up the lokadharmī to [the cosmic level of] the deity he
desires, he should cause [this deity’s] qualities to be present in the can-
didate, or, for those desirous of liberation, [join him] in Śiva. He should
establish the [regent] who is at the top of the [respective] path,24 together
with his powers, recite the OṂ at the end of the mantra, and then join
[him with the deity], while remaining untouched by unmeritorious
[karma].25

22 Mṛgendra, Kriyāpāda 8.6–7: śivadharmiṇy aṇor mūlaṃ śivadharmaphalaśriyaḥ | hitetarā
vinā bhaṅgaṃ tanor ā vilayād bhuvām || bhogabhūmiṣu sarvāsu duṣkṛtāṃśe hate sati |
dehottarāṇimādyarthaṃ śiṣṭeṣṭā lokadharmiṇī ||.
23 The definition of the śivadharmiṇī appears more general here, in comparison to the Svac-
chanda, since the śivadharmiṇī bhautikī initiation is simply said to confer the fruits rooted
in the Śaiva religion, presumably through the propitiation of Śaiva mantras. In addition,
the Mṛgendra also teaches a special version of the śivadharmiṇī, in which the candidate
attains the ability to keep his body until the end of dissolution, the form of the śiva-
dharmī that Bhaṭṭa Nārāyaṇakaṇṭha declares to be the superior one (see Nārāyaṇakaṇṭha
ad Mṛgendra 8.6b: itarā dvitīyā śivadharmiṇy eva tanoḥ bhaṅgaṃ vinā hitā vihitā śreyas
sādhikā.).
24 I am following Brunner’s interpretation (1985, 287) of the problematic phrase placed here
in italics.
25 Mṛgendra, Kriyāpāda 8.149–150: lokadharmiṇam āropya mate bhuvanabhartari | taddha-
rmāpādanaṃ kuryāc chive vā muktikāṅkṣiṇām || yasya yo ’dhvā tadantas tam upasthāpya
satadbalam | mantrānte tāram uccārya yojayet kaṣṭam aspṛśān ||.
via free access
258 mirnig
However, the Mṛgendra goes even further than the Svacchanda and makes
explicit that this procedure can also be applied in case a candidate wishes to
be united with a non-Śaiva brahmanical deity:
In exactly the same way (evam eva), [the Guru] may guide a devotee to
union [with a deity such as] Ambikā, Sūrya, Smara, Viṣṇu or Brahmā, after
having purified that path [up to the cosmic level of the desired deity] [of
all bad karma]. And the same procedure (evam eva hi) [is applicable] for
any other deity that is on the path [that the devotee wishes to be united
with].26
As for the mode of religious practice of the lokadharmī, we learn in the Caryā-
pāda of the Mṛgendra that the lokadharmī is avrata,27 presumably referring to
26 Mṛgendra, Kriyāpāda 8.151–152b: evam evāmbikāsūryasmaraviṣṇuprajāsṛjām | sāyujyaṃ

bhaktimān neyas tad adhvani viśodhite || mārgasthānāṃ pareṣāṃ ca devānām evam eva
hi |.
27 Mṛgendra, Caryāpāda 11: sādhako lokadharmī yaḥ putrakaḥ snātako gṛhī | samayī prāg-
gṛhasthaś ca śaivāḥ syur vratavarjitāḥ ||, “The lokadharmī Sādhaka, a Putraka who is a
married householder, a Samayin and someone who was previously a householder [and had
become a saṃnyāsin] are Śaivas without vratas.” (I follow Brunner (1985, 349) here in sep-
arating the Samayin and pṛāggṛhastha, i.e. one who was previously a householder, as two
kinds of initiates intended here; Kṣemarāja also interprets the passage thus. However,
as Brunner already points out, this category makes little sense in the present context.)
See Kṣemarāja ad loc.: sādhakasya śivadharmilaukikadharmibhedād dvaividhye sati loka-
dharmī yaḥ sādhakaḥ, putrakaḥ, tathā snātako gṛhī gṛhasthatve sati snātaḥ, samayī, prāg-
gṛhasthaḥ pūrvaṃ gṛhasthaś cety ete vratavarjitā jñeyāḥ | prāggṛhasthaś ca ity uttarakālaṃ
gārhasthyatyāgāpekṣayā | yataḥ sambhavanti kecana ājīvanam ujjhitagṛhāvasthitayaḥ |
tadvailakṣaṇyārthaṃ prāggṛhasthagrahaṇam ||. As Brunner (1985, 350, note 5) argues,
vrata appears here to refer to a Śaiva observance with mantras, as in the Svacchanda,
perhaps combined with an abstinent ascetic lifestyle, rather than to the performance of
a Śaiva post-initiatory discipline as a whole; this is indicated by the fact that the loka-
dharmī is mentioned here next to the Putraka who is a snātaka and thus married, the
Putraka who is a head of the household, the Samayin, and someone who is the head of the
household, though this last category is dubious. In any case, of those we can expect that
at least the Samayin and the married Putraka—unless he received a nirbījā initiation—
would have had to comply with the samayas. On the other hand, the other interpretation
of vrata above would fit for all those cases, since such an ascetic lifestyle would not be
appropriate for the married Putraka, and the Samayin would perhaps not be considered
able yet to do so given his neophyte status. Nevertheless, there appears also to be the
possibility of becoming a bhautikavratin with a temporary vrata, in which case the practi-
tioner is instructed to reject the paraphernalia he received for the practice and offer them
back to the deity after the successful outcome of his vrata (see Mṛgendra, Caryāpāda 9b–
10).
via free access
his not being able to perform religious observances involving propitiation of

mantras, just as we have seen in the Svacchanda.
Aside from the Svacchanda and the Mṛgendra, there is one more scrip-
ture that features the lokadharmiṇī as a type of Sādhaka initiation, namely
the Tantrasadbhāva. However, this scripture does not provide any details con-
cerning the lokadharmiṇī; it mentions this initiatory category only for the sake
of completeness while introducing the śivadharmiṇī dīkṣā, here also called
vidyādīkṣā,28 as the initiation undertaken by the Śaiva Sādhaka before receiv-
ing consecration.29
The lokadharmiṇī/śivadharmiṇī distinction between Sādhaka initiations is
carried over into Abhinavagupta’s early-11th-century Tantrāloka,30 since Abhi-
navagupta incorporates the Svacchanda passage into his work, somewhat mod-
28 Cf. the vidyādīkṣā in the Niśvāsa, Goodall et al. 2015, 280–282.

29 Tantrasadbhāva 9.516cd–522 (the variants are recorded from the etext of Dyczkowski;
emendations and conjectures are my own): sa guruḥ śivatulyas tu *śivadharmaphala-
pradaḥ (em.; śivadhāmaphalapradaḥ Ed.) ||516|| śāntyaṃte bhūtidīkṣā tu sadāśivapha-
lātmikā | *śivadharmī (kh; śivadharmo K,g) tu sā jñeyā lokadharmī matānyathā ||517||
śivadharmī tu yā dīkṣā sādhakānāṃ prakīrtitāḥ | teṣāṃ kuryābhiṣekaṃ tu sādhakatve (kh,
g; *sādhakaṃ *? K) niyojayet ||518|| sādhakasyābhiṣeko ’yaṃ vidyādīkṣād anantaram |
†vidyādīkṣā bhavet sā tu vāsanā*bhedataḥ (corr.; -bhedata mss.) sthitā ||519|| na karma-
bhede vidyeta *sarvo ’syādhvani (conj.; sarvāsyādhvani Ed.) saṃsthitaḥ | kṛtāni yāni kar-
māṇi sarvāṇy adhva*gatāni (K, kh; -sarvānya- g) tu ||520|| tāni saṃśodhya (K, g; sa -- kh)
vidhivat kalāpañcasthitāni vai† | yojanyāvasare bhedaṃ vijñātaṃ (conj. Hatley; bhedo vi-
jñānaṃ mss.) sādhakasya tu ||521|| prārabdhakarma paśvarthe tam ekasthaṃ tu bhāvayet
| śivam uccārya sakalaṃ sadāśivatanau (K, kh; sadāsiva -- g) nyaset ||522||, “That Guru is
equal to Śiva, bestowing the fruits of the Śaiva dharma. The bhūtidīkṣā contains the fruits
of Sadāśiva at the end of the Śānti [level]. This [initiation] is known as the Śivadharmī; the
other is called the Lokadharmī. The Śivadharmī is known to be the initiation for Sādhakas.
For them [the Guru] should perform consecration, and unite them with the state of being
Sādhakas (sādhakatve). The consecration for the Sādhaka [should be performed] after the
vidyādīkṣā. †The vidyādīkṣā is based upon a difference in intention (vāsanā); one should
not understand it on the basis of difference in action (karma). Everything is established on
the path: all actions which have been performed, associated with [various levels of] the
cosmic course [of tattvas] [and] established on/with the five kalās, should be duly puri-
fied.† The difference [in procedure] for the Sādhaka is known to be at the time of union
[with the deity]: he should manifest together (ekastham) the prārabdha karma for the
sake of the bound soul, pronounce the mantra of Śiva in its sakala form, and place [the
consciousness] into the body of Sadāśiva.”
30 Tantrāloka 15.23c–24b: sādhako dvividhaḥ śaivadharmā lokojjhitasthitiḥ || lokadharmī pha-
lākāṃkṣī śubhasthaś cāśubhojjhitaḥ |, “There are two types of Sādhakas. The one fol-
lowing the Śaiva dharma [i.e. Śivadharmī] is free of mundane [observances]. The loka-
dharmī Sādhaka is desirous of fruits, rooted in meritorious [actions] and free of impure
[actions].”
via free access
260 mirnig
ified.31 However, Abhinavagupta does not deal with the śivadharmī/lokadharmī

Sādhaka distinction beyond quoting this passage. On the other hand, Jayaratha
returns to this division among bubhukṣus in his commentary on the 13th chap-
ter, which is dedicated to the different degrees of śaktipāta, the descent of
Śiva’s divine power upon souls, which causes them to seek initiation. In this
chapter, Abhinavagupta ranks the intensity of śaktipāta, relating each degree—
from highest, middle, and lowest, each further divided into three levels—to
a certain kind of initiatory category. The principle is that the more intense
the descent of power, the more immediate is the candidate’s final liberation;
and by this criteria he is to receive the appropriate form of initiation.32 On
this scale, the enjoyment-seeking bubhukṣu is at the lower end, but Abhinava-
gupta himself does not make explicit how he correlates different degrees of
śaktipāta to the śivadharmī and lokadharmī initiates. Rather, he restricts him-
self to the bubhukṣu, assigning the middle (madhya-madhya śaktipāta) and
lower levels (manda-madhya śaktipāta) of the middling kind of śaktipāta to
two unspecified categories thereof. Of those, the middle level is for the bubhu-
kṣu who attains Śivahood after enjoying the siddhis of his respective tattva,33
and the lower level for the bubhukṣu who enjoys siddhis in his subsequent
31 Tantrāloka 15.27–30: abhāvaṃ bhāvayet samyak karmaṇāṃ prācyabhāvinām (< Svaccha-

nda [Sv] 4.141ab: prākkarmabhāvikasyātha abhāvaṃ bhāvayet tadā) | mumukṣor nira-
pekṣasya prārabdhr ekaṃ na śodhayet (< Sv 4.141cd: mumukṣor nirapekṣatvāt prārabdhr
ekaṃ na śodhayet) || sādhakasya tu bhūtyartham ittham eva viśodhayet (< Sv 4.142ab:
sādhakasya tu bhūtyarthaṃ prākkarmaikaṃ tu śodhayet) | śivadharmiṇy asau dīkṣā loka-
dharmāpahāriṇī (< Sv 4.143ab: śivadharmiṇy asau dīkṣā lokadharmiṇy ato ’nyathā) || adha-
rmarūpiṇām eva na śubhānāṃ tu śodhanam (< Sv 4.143cd: prāktanāgamikasyāpi adharma-
kṣayakāriṇī) | lokadharmiṇy asau dīkṣā mantrārādhanavarjitā (< Sv 4.144ab: lokadharmiṇy
asau jñeyā mantrārādhanavarjitā) || prārabdhadehabhede tu bhuṅkte ’sāv aṇimādikam
(< Sv 4.144cd: prārabdhadehabhede tu bhuṅkte sa hy aṇimādikān) | bhuktvordhvaṃ yāti
yatraiṣa yukto ’tha sakale ’kale (< Sv 4.145ab: bhogān bhuktvā vrajed ūrdhvaṃ guruṇā yatra
yojitaḥ | sakale niṣkale vāpi śiṣyācāryavaśād bhaved) ||, “He should destroy all the past
and future karmas for the liberation-seeker who is indifferent. He should only purify the
prārabdha karma. For the Sādhaka he should purify [the karmas] in the same manner for
the purpose of powers (bhūti). This is the śivadharmiṇī dīkṣā, which removes the worldly
religion (lokadharmāpahāriṇī). The purification of only the bad karma, and not the mer-
itorious, is the lokadharmiṇī dīkṣā, which is without the worship of mantras. Upon the
death of his current body he enjoys [the supernatural powers], starting with aṇimā; and
having enjoyed these he goes upwards to where he was joined [during the initiation rit-
ual], at a sakala or niṣkala level.”
32 See Takashima 1992, 72–75, who explains the different levels of śaktipāta and the corre-
sponding initiation categories.
33 Tantrāloka 13.242. See also Takashima 1992, 72–73.
via free access
rebirth and only attains union with Śiva after that.34 In his analysis of this
chapter, Takashima (1992, 73) points out that the śivadharmiṇī and the loka-
dharmiṇī division would fit these two respective levels well, yet Jayaratha’s
commentary nonetheless interprets these differently. He instead equates the
two levels of bubhukṣu with two levels of the śivadharmī Sādhaka, a distinc-
tion encountered in the Mṛgendra passage above, which spoke of an ordinary
Sādhaka and one who retains his body until the time of cosmic dissolution.35
It is of course not clear whether Jayaratha had precisely this distinction in
mind, but be that as it may, by assigning these two levels to the śivadharmiṇī
category, Jayaratha frees up the category of the lokadharmī Sādhaka to cover
the three remaining, lowest levels of śaktipāta, which are concerned appro-
priately with practices focused on worldly enjoyments.36 In order to create
three levels of lokadharmī, he uses the following distinctions, according to
which the way to liberation is increasingly remote: the highest (tīvra-manda
śaktipāta) concerns the lokadharmī who automatically attains Śivahood after
an interval of of some period, which he spends in a desired cosmic level—
a category of lokadharmī Jayaratha equates with the standard form we have
so far encountered;37 the second highest (madhya-manda śaktipāta) concerns
the lokadharmī who is initiated again in another world before reaching libera-
tion;38 and the third (manda-manda śaktipāta) concerns the lokadharmī who
first experiences enjoyments for a very long period, as well as different levels of
proximity to the deity (sālokyasāmīpyasāyujyāsādanakrameṇa)39 of the cho-
34 Tantrāloka 13.243. See also Takashima 1992, 73.

35 See p. 255. Note that this passage caused some confusion of interpretation, even for
Nārāyaṇakaṇṭha.
36 Tantrāloka 13.245c–246b. See also Takashima 1992, 73.
37 Jayaratha even quotes the Mṛgendra (cf. note 25) passage at this point, ad Tantrāloka
13.246: yaḥ kaś cit tīvramandaśaktipātavāṃl lokadharmī, sa dīkṣābalāt dehānte kva cana
yathābhimate bhuvanādau bhogān bhuktvā śivatāṃ yāyāt | yad uktaṃ: lokadharmiṇam
āropya mate bhuvanabhartari | taddharmāpādanaṃ kuryāc chive vā muktikāṃkṣiṇaṃ
||, vakṣyati ca: prārabdhadehabhede tu bhuṅkte ’sāv aṇimādikam | bhuktvordhvaṃ yāti
yatraiṣa yukto ’tha sakale ’kale ||. See also Takashima 1992, 73.
38 Jayaratha ad Tantrāloka 13.246: kaś cic ca madhyamandaśaktipātavān kva cana bhuvanā-
dau kañ cit kālaṃ bhogān bhuktvā, tadīśvaradīkṣitaḥ paryante śivatāṃ gacchet.
39 On these different grades of “theistic liberation,” see Watson et al. 2013, 249–250, note 154;
and Goodall 1998, xxxvii–xxxix, note 85. As Watson et al. (2013, 249–250 and note 167) dis-
cuss, an example of a “lower” level of liberation, associated with paurāṇic spiritual goals, is
to become a gaṇapati, or chief attendant, of the lord, thus experiencing being in the same
world (sālokya) as the Lord or in his vicinity (sāmīpya)—spiritual goals, as the authors
point out, which are also expressed in the lay literature of early medieval India, such as
the old Skandapurāṇa and Śivadharmaśāstra.
via free access
262 mirnig
sen universe, and only then receives initiation again, through which he attains
Śivahood.40 In creating these different levels, Jayaratha might have had a sys-
tem similar to the Mṛgendra in mind, where we also found various levels of
lokadharmiṇī dīkṣā, whose recipients range from those seeking final liberation
to those who want to be united with and experience enjoyments related to a
non-Śaiva deity.41
3 Tracing the Socio-historical Context of the Emergence of the

lokadharmiṇī dīkṣā
Thus, to sum up, the following five features consistently appear in descriptions
of the lokadharmī initiate:
1. His source of merit lies in lay religion, the lokadharma, namely what is
taught in śruti and smṛti.
2. During initiation all the bad karma connected with any level of reality
is to be purified, leaving good karma intact to be enjoyed by the initi-
ate.
3. He is not entitled to perform the propitiation of mantras (mantrārā-
dhana).
4. He experiences supernatural powers after death.
5. He can choose whether he eventually attains ultimate liberation, or
instead union with any deity of the cosmos he wishes; in the case of the
Mṛgendra these include even non-Śaiva, brahmanical deities.
For obvious reasons, this initiatory category is difficult to position within a Śaiva
doctrinal framework, for it offers not only enhanced merit through destruction
of all bad karma, but also attainment of any cosmic level, including libera-
tion, without the commitment to perform Śaiva ritual with mantras. In this
light, even the very label “Sādhaka” to such a candidate appears inappropriate,
given the Sādhaka’s paradigmatic association with seeking supernatural attain-
ments through the power of mantras. Perhaps with this hesitation in mind,
40 Jayaratha ad Tantrāloka 13.246: mandamandaśaktipātavān punas tatraiva sālokyasāmī-

pyasāyujyāsādanakrameṇa cirataraṃ kālaṃ bhogān bhuktvā, tata eva dīkṣām āsādya śiva-
tām iyat. See also Takashima (1992, 73), who has a slightly different interpretation of
the sālokyasāmīpyasāyujyāsādhanakrameṇa, taking it to denote “various stages of enjoy-
ment.”
41 Cf. pp. 257–256. Note that in his conclusion, Takashima (1992, 74–75) proposes that Abhi-
navagupta intended the manda level of śaktipāta for those who only received samaya-
dīkṣā, contrary to Jayaratha’s interpretation of different levels of lokadharmī initiates.
via free access
Kṣemarāja ad Svacchanda 4.85, offers a weak explanation for the lokadharmī

being called a Sādhaka, leaning on the meaning of the underlying verbal root
“sādh-,” “to be successful, achieve;” he says that the lokadharmī is referred to as
a Sādhaka “because he achieves (sādhanāt) the fruit by his pious acts alone.”42
One could argue that a stronger reason to refer to the lokadharmī as a Sādhaka
is that he will experience supernatural powers, like the Śaiva Sādhaka, except
that he does so after death. Yet this addition to the professed outcome of the
lokadharmiṇī dīkṣā seems almost to be a doctrinal elaboration added to war-
rant its place among Sādhaka initiations. Another sign that the lokadharmiṇī
dīkṣā held a doctrinally difficult position might be the meager commentary
these passages receive. Bhaṭṭa Nārāyaṇakaṇṭha, for instance, restricts himself
to rather self-explanatory comments on the entire lokadharmiṇī passage in the
Mṛgendra.
Regardless of this awkward near-silence, at least for these scriptural sources
and Kashmirian Śaiva authors, such an initiatory category appears to have been
a social reality that needed to be accommodated and acknowledged. In addi-
tion to the sources discussed, the lokadharmiṇī/śivadharmiṇī distinction also
features in a fourteenth century Kashmirian initiation manual, the Kalādīkṣā-
paddhati of Mānodatta,43 and Kṣemarāja introduces these initiatory categories
into his commentary on the Netratantra, even though the root text itself does
not contain them.44 Additionally, we have seen how Jayaratha uses the category
of the lokadharmī Sādhaka to speak of initiates who pursue mundane spiritual
goals, giving the impression that there exists a wide range of lokadharmī initi-
ates with different degrees of involvement in the Śaiva religion, similar to the
model in the Mṛgendra.
However, the question remains why such an initiatory category, which seems
so counter-intuitive and difficult to justify doctrinally, came into existence in
the first place. From a socio-historical perspective, the creation of a bubhukṣu-
orientated exoteric form of Śaiva initiation appears to be connected to the
position of tantric communities in an early phase, when the sphere of the
42 Kṣemarāja ad Svacchanda 4.85: śubhakarmaṇaiva tu phalasya sādhanāt sādhako ’yam.

43 I thank Professor Alexis Sanderson for sharing with me his handout, “Bhaṭṭa Rāmakaṇ-
ṭha’s commentary on the opening passage of the Mataṅgapārameśvara,” which quotes
the Kalādīkṣāpaddhati on this point (pp. 36–37), and through which I found out about
the usage of the śivadharmī/lokadharmī categories in this manual. For more information
on Manodatta’s work, see Sanderson 2004, 362, note 34.
44 See Kṣemarāja ad Netratantra 4.1: bhuktidīkṣā śivadharmalaukikadharmabhedena bhinnā
sādhakasya, “The bhuktidīkṣā for the Sādhaka is divided into the śivadharma and laukika-
dharma kind.”
via free access
264 mirnig
power-seeking practitioner had a more prominent role. Brunner (1975, 439)

long ago noted that the Svacchanda and the Mṛgendra—both of which contain
the lokadharmiṇī, as we have seen—appear to preserve a comparatively promi-
nent position for the power-seeking Sādhaka, which she suspects harkens back
to an early phase of the tantric traditions, when the gulf between soteriological
and power-orientated goals was not so vast. Her hypothesis finds confirmation
in findings of the past decade, particularly those resulting from study of the Niś-
vāsatattvasaṃhitā (Goodall et al. 2015). As the earliest extant Śaiva tantra, the
Niśvāsa affords important insights into the early stages of the “tantric age,”45
particularly in that—as Goodall points out—it envisages a time when tantric
communities appear to have been restricted to limited esoteric circles46 with
much focus on the propitiation of mantras for magical purposes, which were
pursued prior to liberation. He further observes that the classical fourfold hier-
archy of Samayin, Putraka, Ācārya and Sādhaka is entirely absent at this stage,
with all practitioners simply referred to as Sādhakas—a category of initiates
associated with the utilization of mantras for both supernatural powers and
liberation. The community had not yet been divided into those who sought
one to the exclusion of the other.47 That the pursuit of power and soteriological
goals belonged to the same spiritual path is reflected in the Niśvāsa’s initiation
system, which teaches rites of two types, namely vidyādīkṣā,48 through which a
candidate is admitted to the community and becomes a Sādhaka with authority
to propitiate mantras, and the muktidīkṣā, through which the confirmed Sādha-
ka attains liberation.49 Of neither initiation is there an exoteric form at this
stage.50
Theoretical repercussions caused by separating the liberation- and power-
seekers’ goals are visible, for instance, in the ambiguous treatment of karma’s
45 The Niśvāsa is the earliest extant tantric scripture, preserving a stage of the development
of tantric communities where non-tantric, Atimārgic roots are often visible, as Sanderson
(2006) observed.
46 On the absence of the domain of social ritual in the Niśvāsa, see Goodall et al. 2015, 47.
47 See Goodall et al. 2015, 47–50. As for all initiates probably being referred to as sādhakas at
this stage, see also Goodall et al. 2015, 281.
48 See Goodall et al. 2015, 50. And, in fact, the dīkṣā for the Sādhaka is sometimes referred
to as vidyādīkṣā, as in the Tantrasadbhāva. For more see Goodall’s entry nirvāṇadīkṣā in
Tāntrikābhidhānakośa, vol. III.
49 See Goodall et al. 2015, 280–282, 289. However, Goodall also notes that this interpretation
that any initiate takes first the vidyādīkṣā and then the muktidīkṣā is a working hypothesis
at the moment, not completely excluding the possibility that these may be two separate
modes of initiation, which at the same time seems less likely.
50 See also Goodall et al. 2015, 51.
via free access
elimination during the initiation ritual for the Śaiva Sādhaka.51 Nevertheless,
we find that the status and centrality of the “original” Sādhaka’s position and
practices are preserved in most early tantric scriptures, even after the commu-
nity came to be divided into different initiatory ranks, and centered around
the Ācārya and liberation-seekers. It is tempting to conclude that the creation
of the lokadharmiṇī dīkṣā is directly connected to this early sphere of tantric
religion, which put so much emphasis on the pursuit of magical powers and
benefits. While in due course Śaiva literature grew to put more emphasis on
the soteriological aspirations of the mumukṣu—reflected in the fact that the
soteriological-orientated nirbījā dīkṣā has a more prominent role—the attrac-
tion of the powers and benefits gained through mantras must nonetheless have
played an important role in the initial popularization of tantric groups. For
the lokadharmiṇī dīkṣā would hold obvious attractions for laypeople desiring
to enhance the fruits of their current practice without committing to the dis-
ciplinary regimen of tantric Śaiva ritual. Another appealing aspect may have
been that through this initiation, a lokadharmī became part of the initiatory
community and perhaps privy to services accessible only to initiates, especially
those offered by Śaiva Sādhakas. That broadening access to initiation may have
played an important role in popularizing tantric Śaivism at an early stage is
suggested by the fact that the Mṛgendra extends the scope of the lokadharmiṇī
dīkṣā to those who are not even Śiva-worshippers, people who may nonetheless
have desired specific services distinctive to the Śaiva tantric domain. However,
the doctrinally awkward position of the lokadharmī Sādhaka, combined with
the increasing emphasis on purely soteriological goals, may have led to the
gradual oblivion of this category, causing its reassignment in later sources, as
will be seen below.
51 See ibid., pp. 254 ff. I am not aware of an established, coherent theory on what kind of
karma needs to be left in place in order to pursue supernatural powers. But note that—in
quite a different context—the idea that the success of a Sādhaka’s practice is intertwined
with his karmic position is also found in the Brahmayāmala. In his analysis of the chapters
on the Sādhaka, Kiss shows how the Brahmayāmala teaches complex mechanics under-
lying a logic of karma, according to which the level of a Sādhaka’s practice is dependent
on actions in past births and on whether he had received initiation in a past life, as well
as special rituals to remove karmic bonds that would block the success of the Sādhaka’s
practice (see Kiss 2015, 35–55, in particular 52–53 concerning the rites to remove karmic
bonds).
via free access
266 mirnig
4 The lokadharmiṇī dīkṣā in the Context of the nirvāṇadīkṣā
The two other early Śaiva scriptures which make a distinction between śiva-
dharmiṇī and lokadharmiṇī initiations are the Jayadrathayāmala52 and the
rather later Bṛhatkālottara, both composed after the Svacchanda, and in the
latter case certainly after the Mṛgendra as well. However, in both of these
works the association of these initiatory categories with the Sādhaka gets lost;
they become reassigned to the context of nirvāṇadīkṣā, the initiation for those
seeking liberation rather than supernatural powers and enjoyments. However,
unlike the Bṛhatkālottara, the Jayadrathayāmala preserves in part the origi-
nal logic of karma-elimination seen in the bubhukṣu category of lokadharmiṇī
dīkṣā, despite this reassignment.
In the Jayadrathayāmala we find both of the initiatory categories in ques-
tion—here referred to as śivadharmadīkṣā and lokadharmadīkṣā—in the
description of initiation types53 outlined in the first ṣaṭka. This is the por-
tion of the text which Sanderson (2007, 235–236) identifies as its oldest layer,
deriving originally from a work called the Śiraścheda. Unlike the Svacchanda
and Mṛgendra, the Jayadrathayāmala does not here employ the broad divi-
sion between the mumukṣu and bubhukṣu in its exposition of initiatory cat-
egories, but rather makes a distinction between an initiation that bestows
adhikāra, that is to say the authority to perform certain rituals,54 and those that
bestow liberation (mokṣadā).55 In presenting the latter, the nirvāṇadīkṣā, the
52 In search for references and access to this passage in the Jayadrathayāmala I am grateful
to several people: to Dominic Goodall, who first drew my attention to the fact that the
lokadharmiṇī features in the Jayadrathayāmala; to Olga Serbaeva, for searching her etext
and locating references for me, which showed that the thirteenth chapter is relevant; to
Csaba Kiss, for sharing with me the paper manuscript of this chapter; and to Alexis Sander-
son, for sharing his preliminary edition of the passage, which is quoted below in note 57,
and for reading and discussing the passage in question with me.
53 The immediate context here is Devī’s question to Bhairava about the number of initiatory
categories as well as the issue whether the initiatory procedure is the same for all four
caste-classes.
54 The passage is problematic, and it is not specified what exactly is intended by the category
of adhikāradā initiations. We may hypothesise that this refers to a kind of initiation that
bestows to Śaiva office holders—i.e. the Ācārya and the Sādhaka—entitlement (adhikāra)
to perform rituals for others. See Goodall’s entry on niradhikāradīkṣā in Tāntrikābhi-
dhānakośa, vol. III. See also note 83 for an example of the pair adhikāradā and niradhi-
kāradā dīkṣā in a list of initiation types in the Somaśambhupaddhati.
55 The Jayadrathayāmala (1.13.3) appears to want to impose this distinction already on the
Samayadīkṣā, which would be unusual, since the Samayadīkṣā by itself typically cannot
grant adhikāra. For the Sanskrit text and translation, see note 56.
via free access
Jayadrathayāmala teaches the following options:56 (1) the śivadharmadīkṣā, (2)

the lokadharmadīkṣā, and (3) the kṣipraghnī, which appears to be a synonym
of the sadyonirvāṇadīkṣā (i.e. “the initiation that instantly grants liberation”).
56 As mentioned earlier (note 52), a preliminary edition of this passage was kindly made
available to me by Alexis Sanderson, who has also discussed the passage with me, helping
to clarify points for the translation. I quote his text below, except that I mark his conjec-
tures with “conj. S.,” and degeminate doubled consonants after “r” (any mistakes in the text
and translation are, of course, my own). Jayadrathayāmala, Ṣaṭka 1, 13.3–18: sā *samay-
ātha (conj. S.; sāmayātha Cod.) nirvāṇā dvibhinnā dvisvarūpata⟨ḥ⟩ | samayā hi punar
yāti dvibhāvaṃ phalabhedataḥ ||3|| jñānayogapravṛttisthā prāṇinām adhikāradā | tad-
vāsanavadhā *nityavyāpti (conj. S.; nityā vyāpti Cod.) pada⟨pra⟩dāyikā ||4|| *samayā (conj.
S.; sāmayā Cod.) † yac ca† vijñeyā nirvāṇāpi dvirūpiṇī | *mokṣādhikārabhedābhyāṃ (conj.
S.; mokṣārthikārabhedābhyāṃ Cod.) †sthāṇubhis tantravedibhiḥ† ||5|| bhūyo vimuktidā
dīkṣā nirvāṇā bahubhedagā | śivadharmmā *lokadharmmā (conj. S.; lokadharmmī Cod.)
kṣipraghnī piṇḍapātikā ||6|| *ekatamādhvacicchuddhyā (conj. S.; ekātamodhvacicchud-
dhyā Cod.) vyāvṛtte bandhanatraye | śivāptidarśitāśeṣā (conj. S.; śivāptidarśitāśeṣa Cod.)
samayācārapālanāt ||7|| mokṣāvāptikarī yā tu lokācāraviparyayāt | śivadharmā smṛtā dīkṣā
dvisvarūpā *maheśvari (corr. S.; maheśvarī Cod.) ||8|| nirbījākhyā *sabījā (conj. S.; savīryā
Cod.) ca dīkṣābhedaḥ śivāgame | pākamūrkhasudīrṇāṅga-*eḍaka(conj.; eḍatva Cod.)kuṇi-
yoṣitām ||9|| *sarujātyantyabhaktānāṃ (conj. S.; saruḍhātyantabhaktānāṃ Cod.) bhūbhṛt-
sa⟨ṃ⟩nyāsināṃ priye | viśuddhasamayācārā yā kṛtā daiśikena tu ||10|| nirbījā sā bhaved
dīkṣā sabījā syād vilomataḥ | *vidvad(em.; vidhi Cod.)dvandvasahānāṃ tu sundari kriyate
tadā ||11|| gurudevāgnikāryāṇi yaiḥ kāryāṇi subhaktitaḥ | akurvatāṃ hi tat teṣāṃ vāñchitaṃ
na bhavet phalam ||12|| lokadharmā tu yā dīkṣā sā syāt sthitivilomagā | śivadharmakriyā
sā tu kin tu laukikavṛttigā ||13|| śodhayed atha vā sarvaṃ dharmam ekaṃ na śodhayet |
laukikācāradharmasthaḥ tad bhuktvā nirvṛtiṃ vrajet ||14|| athāsitaṃ viśodhyaṃ tu kar-
maikaṃ daiśikena tu | malino ’dhvāthavā devi *śodhyo bhogo (conj. S; śuddho bhogyo Cod.)
na jāyate ||15|| *na bhogyaṃ tad dhi bhuktatvāt (conj. S.; tad bhogyaṃ tadvibhuktatvā Cod.)
prayāty ātmā paraṃ padam | smṛteyaṃ *lokadharmā (conj. S.; lokadharmī Cod.) tu dīkṣā
*nirvāṇagāminī (em. S.; nirvvāṇamāminī Cod.) ||16|| eṣātīte niruddhe tu prārabdhe kar-
maṇi priye | kṣīṇabhogyavaśād yā tu sā dīkṣā piṇḍapātikā ||17|| kṣurikāstrogramantraiś ca
*yogenādāya (conj. S.; yogenādhāya Cod,) prakṣipet | parayogasamāyoge sadyonirvāṇadā
hi sā ||18||, “And that [initiation] is either a Samayadīkṣā or Nirvāṇādīkṣā, divided into
two because it has two natures. Now the Samayadīkṣā is further twofold because of a
difference in the result (phala). [The first] bestows adhikāra [and] follows the prac-
tices of jñāna and yoga, [the second] destroys the latent impressions (vāsana) of that
[soul?] and bestows a state of eternal pervasion [with the deity]. † [That is known] by
the firm ones who know the tantras. † Besides (bhūyas) the nirvāṇadīkṣā bestowing liber-
ation is of many kinds: śivadharma[dīkṣā], lokadharma[dīkṣā] and the initiation which
kills quickly, causing the body to fall. [The initiation] which reveals everything through
the attainment of Śiva (śivāptidarśitāśeṣā) through the performance of post-initiatory
rites (samayācārapālanāt) once the three bonds (i.e. the three impurities (mala)) have
ceased due to the purification of the consciousness on one [of the six] paths, [that]
initiation is known to be the śivadharmadīkṣā, which bestows the attainment of liber-
ation (mokṣāvāptikarī) because it is contrary to the mundane practice. [And], O God-
dess, [the Śivadharmadīkṣā] has two forms: in Śaiva scriptures the division of initiation
via free access
268 mirnig
Of those, the śivadharmadīkṣā is defined as twofold, namely sabījā (i.e. with

the seed of post-initatory practice) and nirbījā (i.e. without the seed of post-
initatory practice), which amounts to the bifurcation we typically expect of nir-
vāṇadīkṣā. The śivadharmadīkṣā initiate therefore attains liberation at death,
either with the stipulation to perform post-initiatory rites (sabījā) or without
this stipulation (nirbījā). The lokadharmadīkṣā variations, on the other hand,
lead to liberation only after the initiate has enjoyed the merit earned through
his mundane ritual practice. In this category, two versions are taught: first is
the lokadharmiṇī kind, which we have already encountered in the Svaccha-
nda, Mṛgendra and Tantrāloka, namely an initiation involving the elimination
of all bad karma, leaving meritorious karma in place to be enjoyed before the
eventual attainment of liberation. The second appears to be a version of the
lokadharmadīkṣā in which only the impure universe is purified, the idea being
that the initiate will go to some segment of the pure universe after death.57
is called that without the seed (nirbījākhyā) and that with the seed (sabījā). The Ācārya
performs the [initiation] which contains the duty to perform post-initiatory rites purified
(viśuddhasamayācārā) for children, imbeciles, those whose limbs suffered trauma, deaf
people, women, people who are suffering from chronic illness and kings and renouncers
(saṃnyāsin) who are extremely devoted [to Śiva]; this [initiation] is the nirbījā. The sabījā
is the opposite to this and is performed, O beautiful one, for those who are learned, endure
extremes and are able bodied. By those the rituals towards the Guru, the God and the fire
have to be performed with extreme devotion, since the desired fruit will not come about
for them who don’t do [these rites]. The lokadharmadīkṣā is a Śaiva ritual [and therefore]
proceeds contrary to established [practice] (i.e. the brahmanical order), but also conforms
to worldly religion. Either [the Ācārya] should purify all [karma] or only not purify the
dharma (i.e. the auspicious karma); [then the initiate] is dedicated to [the accumula-
tion of] dharma through mundane observances (laukikācāradharmasthaḥ), and having
enjoyed this [dharma] he proceeds to liberation. Thus, the Ācārya should only purify the
bad [karma]. Alternatively, (athavā) [only] the impure path is purified, [so that] no expe-
rience (bhogaḥ) comes about [in the impure universe]. [In other words] that [experience]
does not have to be experienced [anymore in the impure universe] because it has already
been experienced [through the process of initiation]. The soul [of the initiate] (ātmā)
goes straight to the higher level (i.e. the pure universe). That is known to be the initiation
called lokadharma, which leads to liberation. Such [an initiation] (i.e. the śivadharma-
or lokadharmadīkṣā?) [is performed] when the past action has been destroyed, but the
prārabdhakarma [is present], O loved one. But that initiation which [is performed] after
[all experiences that] need to be experienced have waned, that is the piṇḍapātikā (i.e.
which causes the dropping of the body). [The Ācārya] should take hold [of the soul]
through yoga and expel it with the razor and other fierce mantras in order to bring its
union [with the deity] through the highest fusion (parayogasamāyoge). For this is the [ini-
tiation which] bestows liberation immediately (sadyonirvāṇadā).”
57 For the interpretation of the second kind of lokadharmadīkṣā I am indebted to Alexis
Sanderson.
via free access
While our sources discussed so far did not teach such a variation of the loka-
dharmiṇī, we did see that the Mṛgendra and Jayaratha conceived of different
levels of lokadharmī practice, the latter even adding a level in which the loka-
dharmī is reborn and then initiated again on another cosmic level.58 The śiva-
dharmadīkṣā and the lokadharmadīkṣā thus differ broadly from each other in
that the former bestows spiritual liberation through the power of Śaiva ritual
alone, being opposed to the worldly practice (i.e. mokṣvāptikarī yā tu lokācāra-
viparyayāt | śivadharmā smṛtā dīkṣā), while in the latter brahmanical ritual acts
as the source of the immediate spiritual benefit, even though—as in the ear-
lier cases—the power of initiation eventually brings about union with a deity
on some or another cosmic level. That in the Jayadrathayāmala too this cate-
gory holds a doctrinally awkward position can perhaps be inferred by the need
the redactors felt to stress that the lokadharmadīkṣā is indeed a Śaiva ritual
( Jayadrathayāmala, Ṣaṭka 1, 13.12).59
In the case of the Bṛhatkālottara, the śivadharmiṇī/lokadharmiṇī terminol-
ogy makes only a brief appearance. So far I have only been able to locate a
single passage featuring this distinction, at the end of the chapter on initia-
tion.60 Here too these categories are not used in the context of the Sādhaka or
bubhukṣu, but feature rather as variations of the nirvāṇadīkṣā. In contrast to the
Jayadrathayāmala, the pair is used to describe the nirbījā dīkṣā and sabījā dīkṣā,
respectively, and in the case of the lokadharmiṇī, there is no reference to the
elimination of bad karma alone. What is curious is that the text, as preserved
in the manuscript available to me, equates nirbījā dīkṣā with the śivadharmiṇī
and sabījā with the lokadharmiṇī.61 This seems counterintuitive, since in all the
classifications we have seen the sabījā is the initiation after which a candidate
is obliged to observe Śaiva post-initiatory rites (samayas) until death, whereas
in the nirbījā case, the initiate is free of this obligation and maintains his ritual
obligations in accordance with his position in society. It is tempting to conjec-
ture that the transmission of the text is simply wrong at this point—that we
should emend the order so that the nirbījā is equated with lokadharmiṇī and
the sabījā with śivadharmiṇī. However, this remains conjectural, and a more
thorough study of the initiatory categories will require critically editing at least
58 See here, pp. 255 ff. and pp. 261 ff.

59 For the Sanskrit text and translation, see above, note 56.
60 The text has not been critically edited yet and to my knowledge only some parts of it have
been transcribed electronically, kindly shared with me by Nirajan Kafle; the passage in
question appears at the end of what is termed “dīkṣāpaṭala” in the palm-leaf manuscript,
also kindly provided to me by Nirajan Kafle.
61 Bṛhatkālottara 43r, line 2–3: nirbījā śivadharmakhyā sabījā lokadharmiṇī, “The nirbījā is
called the śivadharma[dīkṣā], the sabījā is the lokadharmiṇī.”
via free access
270 mirnig
this portion of the text. Nevertheless, what we can extract for our discussion is
that here the classification is associated neither with the Sādhaka, nor with the
“karma logic” of the lokadharmiṇī dīkṣā found in the Jayadrathayāmala.
5 Pre-13th Century Saiddhāntika Manuals: Householders versus

Ascetics
Labels for the initiation categories śivadharmiṇī and lokadharmiṇī are carried
over into the Saiddhāntika manuals, though differently and inconsistently reas-
signed by various sources, and with weak links to the original Sādhaka context.
The following surveys early Saiddhāntika manuals, with particular focus on the
earlier sources, but is not exhaustive. Noticeably, the influential Somaśambhu-
paddhati, also called the Kriyākāṇḍakramāvalī, of Somaśambhu does not teach
a mundane form of the sādhakadīkṣā, nor does it include lokadharmiṇī among
the mumukṣu options; rather, the text speaks of the Śaiva Sādhaka receiving,
as in the sources discussed earlier, a modified form of the nirvāṇadīkṣā right
before his consecration to office.62 However, in the description of the sabījā
nirvāṇadīkṣā, at the point when karmas are emplaced upon the pāśasūtra—
representing all reality levels of the cosmos, which will be burnt and destroyed
in the consecrated fire—the Somaśambhupaddhati presents various options
for the karma that can be exempted from purification, namely the prārabdha-
karma, the karma for mantrasiddhi, and the karma that results from meritori-
ous acts, comparable to those of the lay devotee. While Somaśambhu himself
does not explicitly correlate these options to specific initiation types, we may
note that the latter two categories of karma, to be exempted for purification in
certain circumstances, correspond to the procedures we have seen envisaged
for the śivadharmiṇī initiation for the Śaiva Sādhaka, and the lokadharmiṇī
initiation, respectively. In fact, in his commentary on this passage, Trilocana
explicitely assigns these categories accordingly.63
62 See, Somaśambhupaddhati Sādhakābhiṣeka 1–2: sādhaskyābhiṣeke tu nivṛtyādikalātrayam

| mumukṣor iva saṃśodhya śāntyatītāṃ viśodhayet || tataḥ sadāśivaṃ dhyātvā mūlama-
ntreṇa pūrṇayā | śāntau saṃyojya kurvīta guṇāpādanam aṣṭadhā. For a tanslation and
notes, see Brunner 1977, 500–506.
63 Somaśambhupaddhati, Nirvāṇadīkṣāvidhi 31–32: sabījāyāṃ tu dīkṣāyāṃ samayācārapāśa-
taḥ | dehārambhakadharmāc ca mantrasiddhiphalād api || iṣṭāpūrtādidharmāc ca vyatiri-
ktaṃ prabandhakam | caitanyarodhakaṃ sūkṣmaṃ kalānām antare smaret ||. For a trans-
lation and annotation, see Brunner 1977, 190–196. For Trilocana’s commentary on this
passage, see Brunner 1977, 191, [32a] and notes thereon.
via free access
Another manual from this period is the Vimalāvatī of Vimalaśiva, who com-
posed his work in 1101/2 and lived in Vārāṇasī (Sanderson 2014, 22). Vimalaśiva
explicitly uses the categories of śivadharmiṇī and lokadharmiṇī; and though
he partly draws on the passage from the Mṛgendra examined earlier, he does
not assign these to different kinds of Sādhaka, but rather assigns the śiva-
dharmiṇī initiation to the Putraka and Ācārya—a feature we also see in the
Ajitāgama64—and the lokadharmiṇī simply to the Sādhaka devoted to merito-
rious action.65 However, it appears that in this passage the allocation of initi-
ation categories has gone awry, such that the Śaiva Sādhaka is missing, unless
the reason for this omission is the actual disappearance of the Sādhaka from
medieval South-India.
Jñānaśiva wrote his Jñānaratnāvalī also while residing in Vārāṇasī, probably
some decades later.66 In his application of the śivadharmiṇī/lokadharmiṇī ter-
minology he seems to be the most innovative of Saiddhāntika authors in this
period. He also quotes sources for this classification that have yet to be identi-
fied. Jñānaśiva uses this terminology to denote divisions of the nirvāṇadīkṣā—
that is to say, the full regular Śaiva initiation—in essence in order to distinguish,
broadly, between initiations performed for the ascetic practitioner, here the
śivadharmī, and those for the householder practitioner, here the lokadharmī.67
64 Ajitāgama 77.14: sabījā ca dvidhā bhinnā prathamā śivadharmiṇī || sādhakācāryayoḥ pro-

ktā deśakālādibhedanaḥ | dharmādharmātmakaṃ karma prāgāgāmi vicitrakam || sañc-
intya śodhyate yatra saivoktā śivadharmiṇī | adharmamātrasaṃśuddhau dvitīyā lokad-
harmiṇī ||.
65 Vimalāvatī f. 67r: dvividhā syāt sabījāpi tatrādyā śivadharmiṇī | *putrakācāryayor (conj.;
putrakācāryayo Ed.) mukhyā dharmādharmakṣayakarī || iṣṭāpūrtādinirate sādhake loka-
dharmiṇī || śivadharmiṇy *anor (conj.; aṇau Ed.) mūlaṃ śivadharma*phalaśriyaḥ (conj.;
*phalaṃ śriyah Ed.) | *hitetarā (conj.; hitatarā Ed.) vinā bhaṃgaṃ *tanor (conj.; tannaur
Ed.) ā vilayād *bhuvām (conj.; bhavāṃ Ed.)|| bhogabhūmiṣu sarvāsu *duṣkṛtāṃśe (conj.;
duḥskṛtāṃśa- Ed.) hate śati | dehāntarāṇimādyarthaṃ śiṣṭeṣṭā lokadharmiṇī ||.
66 See Goodall 2000, 209 for the date of Jñānaśiva. I am extremely grateful to Dominic
Goodall for sharing with me his etext of the Jñānaratnāvalī as well as his photographs
of the transcript (Re 1025/52) cited in this article.
67 Jñānaratnāvalī, p. 266: tatra tāvat *sāpekṣānapekṣā (conj.; sāpekṣāpekṣā Cod.) dvividhā
śivadharmiṇī lokadharmiṇī ceti. tatra tāvac chivadharmiṇī liṅgināṃ śikhācchedasamā-
yuktā anyā gṛhiṇāṃ tadvirahitā. uktaṃ ca: dvedhā nirvāṇadīkṣā laukikī śivadharmiṇī | gṛhi-
ṇāṃ laukikī jñeyā liṅgināṃ śivadharmiṇī || śikhācchedo na yatrāsti dīkṣā sā lokadharmiṇī
| śikhākhaṇḍena saṃyuktā dīkṣā sā śivadharmiṇī ||, “Therein, now, [the initiation types]
are twofold, [namely] dependent on [whether] there is a requirement to perfrom post-
initiatory practice or not; and [they are also twofold insofar as being] śivadharmiṇī or
lokadharmiṇī. Here [in the category of the sāpekṣā nirvāṇadīkṣā kind], the śivadharmiṇī
is for ascetics and contains the cutting off of the topknot, while the other [initiation] is
for householders and is without [cutting off the topknot]. And it is said: The nirvāṇadīkṣā
is twofold, [divided into] the mundane (laukikī) [kind] and the śivadharmiṇī. The mun-
via free access
272 mirnig
The actual procedures for the śivadharmiṇī and lokadharmiṇī initiations are
the same, being those for the nirvāṇadīkṣā, with only one difference: for the
śivadharmī, the topknot is cut off during the initiation, while for the lokadharmī
it is not.68 The topknot is believed to contain the rodhaśakti,69 that is to say the
power to delude the soul,70 which is apparently eliminated by cutting the top-
knot off in the case of the ascetic. This is not done for the lokadharmī, who, after
all, is by definition still engaged in worldly activity. In terms of post-initiatory
practice prescribed for the lokadharmī initiate, the Jñānaratnāvalī differs from
the earlier scriptural sources discussed in the first part of the paper, in which
an initiate who received a lokadharmiṇī or lokadharmadīkṣā was only required
to continue his ritual obligations of the brahmanical ritual sphere (the loka-
dharma), without Śaiva-specific additions. In the Jñānaratnāvalī, however, we
have seen that the lokadharmiṇī is clearly defined as a form of nirvāṇadīkṣā,
which—unless otherwise specified—is of the sabījā kind. Thus, the signifi-
cance here is that the initiate retains his socio-religious status in mundane
society (lokadharma) as a householder and keeps up these practices in addi-
tion to his Śaiva post-initiatory ritual obligations, a principle that is already
expressed in early Śaiva sources.71 There is no mention of the procedure to
dane is known to be for householders, the śivadharmiṇī for ascetics. In the lokadharmiṇī
initiation there is no cutting of the topknot. The initiation in which the topknot is cut off
is the śivadharmiṇī.” At the end of this section, all the sāpekṣā types are contrasted with
the nirapekṣā; see Jñānaratnāvalī, p. 269: evaṃ sāpekṣā bahudhā jñeyā nirapekṣā ca parā-
paratvena dvividhā sadyosadyonirvāṇā ceti, “In this manner the sapekṣā type is known to
be manifold, and the nirapekṣā type is twofold on account of its higher and a lower forms,
[namely] the immediate and the not-immediate types of nirvāṇadīkṣā.”
68 For the Sanskrit text and translation, see note 67. Note that this is not the case in all
manuals; for instance, the Somaśambhupaddhati does not maintain this distinction. See
Brunner 1977, 338, note 395.
69 Also referred to as tirodhānaśakti.
70 See, e.g., Somaśambhupaddhati, Nirvāṇadīkṣāvidhi, verse 219 (Brunner 1977, 338–339):
śikhāmantritakartaryā rodhaśaktisvarūpiṇīm | śikhāṃ chindyāt śivāstreṇa śiṣyasya catu-
raṅgulām ||, and also Jñānaratnāvalī, p. 344: śikhāñ chidyāt tayā śaktyā rodhaśakti-
svarūpiṇīm | kāraṇaṃ sarvapāśānāṃ sāvadhe dvādaśāṅgulāt. See also Brunner 1977, xli,
and for a discussion of the cutting of the topknot at other moments during initiation in
some sources, see Goodall et al. 2015, 282–283.
71 Sanderson (2009, 302) has demonstrated how Śaiva scriptures and commentaries early
on developed clear positions regarding the importance for Śaiva householder initiates
of maintaining their brahmanical householder duties, i.e. the lokadharma. This helped
ensure that Śaiva initiation remained attractive to the householder mainstream, who,
in this way, were not required to completely transform their public lives. However, it is
emphasised in these sources that such full Śaiva initiates were asked to maintain brah-
manical practice merely for the sake of conformity, or in order to avoid transgression, not
as a source of spiritual merit.
via free access
eliminate only the bad karma linked to all reality levels, which we have seen
associated with the lokadharmiṇī dīkṣā.
However, while this seems to be the underlying basic division of the nir-
vāṇadīkṣā in the Jñānaratnāvalī, these distinctions subsequently become
entangled with aspects of the śivadharmiṇī and lokadharmiṇī initiations stem-
ming from their original Sādhaka context. This happens in the context of cre-
ating sub-categories for both the śivadharmiṇī and lokadharmiṇī categories,
based on a division between those who desire something (sakāma), i.e. what
would otherwise be called the bubhukṣu, and those who do not desire anything
except liberation (akāma, niṣkāma), i.e. the mumukṣu. For ascetic practition-
ers, the śivadharmiṇī is divided into kinds for temporary ascetics (bhautika)
and permanent ascetics (naiṣṭḥika),72 and each of these is further divided into
the sakāma and niṣkāma type.73 In the case of the naiṣṭhikī sakāmā initia-
tion, the initiate is promised enjoyments of the pure universe, after which he
is to attain liberation.74 The naiṣṭhikī niṣkāmā initiation, on the other hand,
bestows liberation immediately. On the division of the bhautikī into sakāmā
and niṣkāmā types, Jñānaśiva quotes the first two verses of a passage from
the Mṛgendra, which mentions that there are two types of bhautikī dīkṣā, the
superior and the inferior, the first being śivadharmiṇī and the second being
the lokadharmiṇī type.75 However, since Jñānaśiva uses these as sub-categories
72 Jñānaratnāvalī, p. 266 (with a conjecture by Goodall recorded from the etext): tatra
śivadharmiṇī dvidhā naiṣṭḥikī bhautikī ceti tatra niravadhitapasvino naiṣṭhikāḥ. teṣāṃ yā
sā naiṣṭhikī niravadhīty api kathyate dvitīyā *katipayadina(conj. Goodall, gādipayadina
ms.)vratapālanād anantaraṃ gṛhapadadāyī |, “Here the śivadharmiṇī kind is twofold,
[namely] naiṣṭhikī and bhautikī. Of these (tatra) the naiṣṭhikas are permanent ascetics.
The [initiation] that is [performed] for them is the naiṣṭḥikī, that is to say the one called the
permanent. The second [kind of initiation] (i.e. the bhautikī) restores the state of being
a householder immediately after the performance of an observance for a certain number
of days.”
73 Jñānaratnāvalī, p. 266: tatra naiṣṭḥikī sakāmākāmabhedena dvidhā, “In that case, the
naiṣṭḥikī initiation is divided into one for those who desire [enjoyments and supernat-
ural powers] and one for those who don’t.”
74 Jñānaratnāvalī, pp. 266–267: tatra sakāmasya śuddhādhvabhogan datvā tadanantaraṃ
śivapadapradā sakāmā, “In that case, for the one who desires [enjoyments and supernatu-
ral powers] the sakāmā [initiation] [first] bestows the pleasures of the pure universe and
immediately after that the level of the [highest] Śiva.”
75 For the Mṛgendra passage, see pp. 256ff. Jñānaratnāvalī, p. 267: *atha (em.; athā Cod.)
bhautikī ca tathā dvividhā | uktaṃ ca: evam eva kriyāyogād *bhautiky (conj.; bhaumity
Cod.) api parāparā | kin tu dehāviyogārthaṃ prārabdhaṃ karma mohinām || śaivasā-
dhanasādhyena sandhāya paripālayet | sānubandhaṃ dahed anyad dīpte dīkṣāhutāśane
|| asyāpy ayam arthaḥ: evam eveti naiṣṭhika*sakāmākāmavad (conj.; -sakāmakāmavad-
Cod.) bhautikasakāmākāmayor api kriyāyogāt kriyā ca yogaś ceti kriyāyogam | tat kriyā
via free access
274 mirnig
of the śivadharmiṇī, this is not the way in which he wants the Mṛgendra to
be understood; rather, the prose after the quotation explains that the differ-
ence lies in the sakāmā entailing purification of the impure universe only up
to māyā, thus providing enjoyments within the sphere of the pure universe;
the niṣkāmā type, however, entails purification of all levels of the universe, and
the initiate thus attains liberation at death.76 If we are to understand these as
two sub-categories of the temporary ascetic, the second type makes little sense,
a confusion that is probably caused by the terminology bhautikī and naiṣṭḥikī
dīkṣā in the Mṛgendra, where the terms are not, in turn, synonyms of the bhau-
tikavrata and naiṣṭhikavrata,77 as they are understood in the Jñānaratnāvalī.78
The lokadharmiṇī dīkṣā, in turn, is divided into superior (para) and infe-
rior (apara) forms. Of those, the superior version is what we expect of the
nirvāṇadīkṣā, with all karma but the prārabdha destroyed, the candidate thus
being united with Śiva at the end of the initiation ritual in anticipation of his
ultimate liberation at the time of death.79 The inferior form is assigned to an
initiation in which only bad karma is destroyed, thus the version of the loka-
dharmiṇī dīkṣā seen in the Svacchanda and the Mṛgendra.80 Jñānaśiva even
sakāmasya māyāntādhvaviśodhanam | niṣkāmasya samastādhvaviśodhanañ ca yogaḥ |

punaś śuddhādhvani bhogayogas sakāmasya niṣkāmasya śivayogaś ca […], “Next, the bhau-
tikī [dīkṣā] is twofold, and it is said [in the scriptures]: ‘In the same way the bhautikī [dīkṣā
is achieved] through ritual and union [and] is also of a superior and inferior kind. Rather,
for the [still] deluded [souls] he should preserve the prārabdha karma, which has the pur-
pose of keeping [the initiate] with his [current] body, after joining it with [the karma] to
be cultivated for the practice of propitiating Śaiva mantras for supernatural powers (i.e.
śaivasādhana). The other [karmas] together with their consequences he should burn in
the blazing initiation fire.’ This is the meaning: ‘in the same way’ means through kriyāyoga,
that it to say the rituals and union [with the respective deity] for the sakāma and akāma
bhautikī are just like those for the sakāma and akāma naiṣṭhikī. Therefore the ritual for
the sakāma kind is the purification of the universe up to māyā. For the niṣkāma it is the
purification of the entire cosmic path and then union. Or rather, for the sakāma there is
union with enjoyments in the pure universe, and for the niṣkāma there is union with Śiva.
[…]”
76 Ibid.
77 See p. 254, note 19.
78 See note 72.
79 Jñānaratnāvalī, p. 268: atha lokadharmiṇī ca parāparatvena dvividhā. tatra parā śivatvadā.
uktaṃ ca: śivāśivātmakaṃ karma śodhayitvādhvamadhyagam | yā yā nirvāṇam āpnoti sā
parā lokadharmiṇī. iti, “Now the lokadharmiṇī is divided into the superior and the infe-
rior. Of these the superior bestows Śivahood. And it is said: ‘Having purified the pure and
impure karma situated on the cosmic path (adhvamadhyagam), whichever [initiation]
brings about liberation is the supreme lokadharmiṇī [initiation]’.”
80 Jñānaratnāvalī, p. 268: *śarīrapātottaram (em.; śarīrapādotaram Cod.) abhilakṣitabhu-
vane śiṣṭāparālokadharmiṇī. aparā punar abhilaṣitabhuvanaṃ yāvad adhvaśuddhau sa-
via free access
quotes a verse which is shared with or probably original to the Mṛgendra,

though embedded in a quotation from a different, unidentified text, which reit-
erates that the initiate can be united with the deity of his choice at the end of
the initiation ritual.81 However, what is not made explicit in any of these pas-
sages is what the practice after initiation consists of; since both of them are
sabījadīkṣās, we expect some form of post-initiatory practice to be applicable,
which is fitting for the superior kind, but not for the inferior kind, as we have
seen it so far. But given the context, we can probably assume that the inten-
tion is the same as it was in the original context, namely that the initiate of the
lower (apara) form simply follows his lokadharma after initiation. However,
it is difficult to know whether this reflects actual practice—that is, whether
this categorization intimates the convention of giving initiation even to lay
worshippers of non-Śaiva deities—or whether this is an artificial distinction
created in the process of accounting for the Mṛgendra’s shared terminology.
Another place where Jñānaśiva invokes the śivadharmiṇī/lokadharmiṇī dis-
tinction is regarding the post-funerary śrāddha rituals, which vary according
to initiatory types. Here, the mode of śrāddha for the śivadharmī appears to
tyām abhīṣṭabhuvane ’py *aśubhāṃśakarmanivṛttau (conj.; aśubhāṃ karmanivṛttau Cod.)

tadbhuvaneśvara*guṇāṇimādisiddhipradā (conj.; *gaṇāṇimādisiddhipradā Cod.). uktañ
ca: bhogabhūmiṣu sarvāsu duṣkṛtāṃśe hate sati | dehottarāṇimādyarthaṃ *śiṣṭeṣṭā (conj.;
śiṣṭoṣṭa- Cod.) lokadharmiṇī || iti, “The remaining one is the inferior lokadharmiṇī [initi-
ation], [which] after death (śarīrapātottaram) [leads the candidate to] the universe he
desired. But the inferior [lokadharmiṇī initiation] bestows supernatural powers, starting
with the power to become as small as one wishes, and the qualities of the deity presid-
ing over the [respective cosmic level], after the purification of the cosmic path up to
the level of the universe [the candidate] desires. And even on this desired cosmic level,
the cessation of karma is [only] of the unmeritorious part. And it is said: ‘The remain-
ing [initiation], known as the lokadharmiṇī, [is performed] for the purpose of [obtaining
supernatural powers], such as the power to become as small as one wishes, in a subse-
quent body, after the bad portion of [karma] is destroyed on all reality levels [through
initiation]’.” Note that the appropriation of the Mṛgendra verse here to denote the inferior
level of the lokadharmiṇī dīkṣā is not the original intention. Rather, the phrase śiṣṭeṣṭā
lokadharmiṇī denotes the second kind of bhautikī dīkṣā, namely the lokadharmiṇī itself,
in contrast to the śivadharmiṇī.
81 Jñānaratnāvalī, p. 268: *śivadharminam (conj.; śivadharmiṇīm Cod.) āśodhya *yojayet
(conj.; yo japet Cod.) parame pade | dehapāte ca *mokṣas (conj.; mokṣa Cod.) syāt sama-
yācārapālanāt || *lokadharmiṇam (conj.; lokadharmiṇīm Cod.) āropya mate (conj.; made
Cod.) bhuvanakartari | *taddharmāpādanaṃ (conj.; taddharmapālanaṃ Cod.) kuryāc
chive vā muktikāṃkṣiṇām ||, “Having purified the śivadharmī, he should join him with the
highest cosmic level, and after having performed his post-initiatory obligations liberation
will come about at death. Having lifted up the lokadharmī to the desired [level] of the pre-
siding deity, he should bring about the qualities of this [deity in the candidate] or [unite
him] in Śiva, for those who desire liberation.”
via free access
276 mirnig
be that which is intended for ascetic initiates, while that for the lokadharmī
is for householder initiates, thus being consistent with the basic distinction
Jñānaśiva sets up in the beginning of the initiation section (Mirnig 2013, 292–
295).
The Jñānaratnāvalī’s innovative application of the śivadharmiṇī/loka-
dharmiṇī terminology led to some interpretative confusion and inconsisten-
cies in subsequent classifications,82 perhaps especially since the influential
Somaśambhupaddhati does not mention this division of initiatory categories.83
Thus Trilocana, for instance, who is a student of Jñānaśiva and wrote a com-
mentary on the Kriyākāṇḍakramāvalī, when commenting on the initiation cat-
egories taught there, fits the śivadharmiṇī and lokadharmiṇī initiations into
Somaśambhu’s system under the category of sāpekṣā initiation, even incorpo-
rating such details as the difference in cutting or not cutting off the topknot.84
However, he runs into difficulties when mapping the initiatory categories onto
the various forms of śrāddha, which differ according to the initiate’s spiritual
status during his lifetime (Mirnig 2013, 296).
This distinctive idea that the topknot is be to cut off in the case of the śiva-
dharmiṇī initiation, but not in the lokadharmiṇī, is carried over into subse-
quent sources that include these categories in their initiatory classifications.
82 See, for instance, Brunner 1977, 194, note 71.

83 Somaśambhupaddhati, Samayadīkṣāvidhi 5–13: nirādhārā ca sādhārā dīkṣā tu dvividhā
matā | nirādhārā dvayos teṣāṃ sādhārā sakalasya tu || ācāryanirapekṣeṇa kriyate śambhu-
naiva yā | tīvraśaktinipātena nirādhāreti sā smṛtā || ācāryamūrtim āsthāya mandatīvrādi-
bhedayā | śaktyā yāṃ kurute śambhuḥ sā sādhikaraṇocyate || iyaṃ caturvidhā proktā
sabījā bījavarjitā | sādhikārā niradhikārā yathāvad abhidhīyate || samayācārasaṃyuktā
sabījā jāyate nṛṇām | nirbījā tv asamarthānāṃ samayācāravarjitā || nitye naimittike kāmye
yasya syād adhikāritā | sādhikārā bhaved dīkṣā sādhakācāryayor ataḥ || nirbījādīkṣitānāṃ
tu tathā samayiputrayoḥ | nityamātrādhikāritvād dīkṣā niradhikārikā || dvividheyaṃ dvirū-
pāpi pratyekam upajāyate | ekā kriyāvatī tatra kuṇḍamaṇḍalapūrvikā || manovyāpāramā-
treṇa yā sā jñānavatī matā | itthaṃ labdhādhikāreṇa dīkṣācāryeṇa sādhyate ||. For a trans-
lation and notes, see Brunner 1977, 4–13.
84 Thus, Trilocana’s Somaśambhupaddhativyākhyā, p. 184, is somewhat similar to Jñānaśiva’s
text (see note 67): … yā dīkṣaiva sabījā, nirbījā, sādhikārā, niradhikāra ceti caturvidhā
proktā. tatra tāvat samayācārasaṃyuktā yā sā sabījā. sā ca dvividhā śivadharmiṇī loka-
dharmiṇī ca. tatra tāvac chivadharmiṇī śikhāyāś chedasahitā. sā ca dvividhā naiṣṭḥikī bhau-
tikī ceti. tāvan naiṣṭhikī niravadhiḥ. dvitīyā katipayadinavrataparipālanānantaraṃ gṛha-
padadāyinī. And Trilocanaśiva ad Somaśambhupaddhati, Nirvāṇadīkṣāvidhi 31, as quoted
by Brunner 1977, 191, [32a]: atraivāvasare tāvan nirbījināṃ dehārambhakarmaṇaḥ | sabī-
jāyāṃ tu dīkāyāṃ śivadharmiṇyāṃ samayācāradehārambhakakarmadvayāt | atha sādha-
kadīkṣāyāṃ mantrasiddhiphalād api | lokadharmiṇyām iṣṭāpūrtādidharmāc ca vyatiri-
ktaṃ prabandhakaṃ kalānām antare smaret. For notes and a translation, see Brunner 1977,
190, note 66.
via free access
Thus, the Siddhāntaśekhara (18–22b)85 of Viśvanātha, who wrote in Vārāṇasī in

the first half of the thirteenth century (Sanderson 2014, 23), in essence teaches
a structure similar to the Jñānaratnāvalī, with the śivadharmiṇī and lokad-
harmiṇī being the broad divisions of sabījā initiation, except that the lokad-
harmiṇī is not further divided; the śivadharmiṇī, in contrast, is divided into
naiṣṭhikī and bhautikī varieties, but there is no further division into sakāma
and niṣkāma kinds.86
Even this brief survey of some early-medieval South-Indian sources, writ-
ten some centuries after the sriptural material discussed in the first part of the
paper, shows that there was no fixed consensus on what the different initia-
tory categories covered exactly; the same elements appear in various places,
assembled together differently.87 What the sources share, however, is that these
categories are no longer in any case associated with the bubhukṣu category.
6 Conclusion
The material reviewed above allows us to trace how the emergence and devel-
opment of the lokadharmiṇī initiation are linked to larger developments in
tantric Śaiva history. To sum up: the initial stage appears to reflect an early
phase of tantric Śaivism, when the power- and enjoyment-seeking aspects of
85 Siddhāntaśekhara p. 145, Naimittikakāṇḍa, chapter 2.18–22b: sabījā dvividhā proktā pra-

thamā śivadharmiṇī | dharmādharmātmakaṃ karma sañcitāgāmicitrakam ||18|| saṃ-
mṛdya sādhyate yasmāt sā dīkṣā śivadharmiṇī | adharmasañcitaṃ sarvaṃ karmāṅgam iva
śodhyate ||19|| dharmāvaśeṣito yatra sā dīkṣā lokadharmiṇī | tatrādyā saśikhācchedā na
śikhācchedanādarā ||20|| bhautikī naiṣṭikī ceti dvidhā sā śivadharmiṇī | vratamokṣasamā-
yuktā bhautikīti nigadyate ||21|| śarīrapātaparyantā yathā sā naiṣṭhikī matā |.
86 See, for instance, Nigamajñāna II’s Dīkṣādarśa, for which see Ganesh 2009, 196; see also
further mention of these categories on pp. 137, 164–165, 196, and 240.
87 For instance, in the 16th century Śaivaparibhāṣā (for the date of which see Sanderson 2014,
86) of Śivāgrayogin, the śivadharmiṇī and lokadharmiṇī are differentiated according to an
original feature: that for the former all karmas are destroyed, and for the latter only the
bad ones. The feature of cutting off the topknot, on the other hand, is connected with
the nirvāṇadīkṣā as opposed to the samayadīkṣā and viśeṣadīkṣā; however, the text does
link the cutting of the top-knot to the ascetic, saying that it should take place after his
nirvāṇadīkṣā; see Śaivaparibhāṣā, chapter 5, p. 5: iyam api śivadharmiṇī lokadharmiṇī ceti
dvividhā. dharmādharmayor ubhayor api yatra saṃśodhanaṃ sā śivadharmiṇī. adharma-
mātrasya yatra saṃśodhanaṃ sā lokadharmiṇī. kriyāvatī prakārāntareṇa trividhā. sama-
yadīkṣā viśeṣadīkṣā nirvāṇadīkṣā ceti. yatra rudrapade yojanaṃ kriyate sā samayadīkṣā.
yatreśvarapade yojanaṃ sā viśeṣadīkṣā. anayor ubhayor api nāsti śikhācchedaḥ. yatra śiva-
pade yojanaṃ sā nirvāṇadīkṣā. yatra nirvāṇadīkṣapūrvakaṃ sannyāsaḥ tatra tv asti śikhā-
cchedaḥ.
via free access
278 mirnig
tantric practice held a comparatively prominent role. It seems that the loka-
dharmiṇī dīkṣā was created in this milieu in order to extend the benefits and
services of the tantric repertoire to various groups within the brahmanical
mainstream, in this way appealing to the ordinary lay worshipper, who was
perhaps attracted by the magical and merit-enhancing aspects of the religion
but did not want to commit to the ritual life of a tantric practitioner. This ini-
tial function aligns with a gradual process of professionalization within tantric
Śaiva communities, bridging the gap between esoteric practitioners and the
laity. In this matter they followed in the footsteps of the Atimārga, for epi-
graphical evidence attests to members of these Śaiva ascetic initiatory cir-
cles serving as religious officiants for lay practitioners from as early as the
fourth century (Sanderson 2013).88 However, at the same time, the Sādhaka
form of lokadharmiṇī dīkṣā is found only in the Svacchanda, Mṛgendra, and
the Tantrasadbhāva, suggesting that in the broader Śaiva tantric world—apart
from Kashmir—its function was of marginal or no importance, a circumstance
likely contributing to its disappearance from subsequent sources.
We have seen that in other sources using the lokadharmiṇī category, this was
reassigned to the sphere of the liberation-seeker. This allocation, in turn, may
be correlated with two further developments. Firstly, increasing emphasis was
given to the liberation-seeking aspect of the religion, a development that in
the South culminates in the disappearance of the figure of the Śaiva Sādhaka,
as Brunner (1974, 440) observes. Consequently, retention of an exoteric form
of the enjoyment-seeker category may have gradually become obsolete. Sec-
ondly, we know that tantric Śaivism was successful in gradually adapting to
the ritual needs of the brahmanical mainstream, even to the point of incor-
porating its practices within the tantric ritual repertoire, so that full initiation
also became available to brahmanical householders, giving them unrestricted
access to the religion.89 This development is reflected in the use of the cate-
gory lokadharmiṇī dīkṣā in the 12th-century Jñānaratnāvalī, where it is pre-
scribed for precisely those householder initiates who had full access to the
tantric repertoire but maintained their position and brahmanical ritual obli-
gations in society (the lokadharma). Again, the need for a lokadharmiṇī dīkṣā,
which made the bubhukṣu aspect of the religion more accessible and may
originally have been a catalyst for outreach, may thereby have become less
meaningful. The increasingly large-scale inclusion of the brahmanical main-
88 See also Sanderson 2009 for the many ways in which tantric officiants created strong links
with the laity, also by acting as officiants in the domain of royal and consecration ritu-
als.
89 See Sanderson 1995 and 2009.
via free access
stream from this time onwards is also visible in other areas of ritual, especially
in South-Indian sources. Thus, for instance, we find the emergence of literature
on penance rites regulating social interactions, such as the two works known
as Prāyaścittasamuccaya by Hṛdayaśiva and Trilocana, respectively, the latter
having been edited and translated by Sathyanarayanan (2015). In his intro-
duction to this volume, Dominic Goodall draws attention to the socially all-
encompassing nature of the work, showing how its prescriptions include even
women as part of the community, and how the focus shifts to public temple rit-
ual.90 Goodall also points out another change within these tantric traditions,
which we may find reflected in the history of the lokadharmiṇī dīkṣā, namely
that along with these developments the goal of individual liberation gradu-
ally waned in importance. He observes also that the “colonization by the Śaiva
Siddhānta of many large South Indian temples, and the dominance of Vedān-
tic non-dualism appear to have diminished the significance of ritual salvific
dīkṣā.”91 As the tradition’s focus shifted from individual to public ritual life, it
may be this waning importance of dīkṣā’s salvific function that has contributed
to the confusion evident in Saiddhāntika manuals concerning how to cate-
gorize the lokadharmiṇī and śivadharmiṇī dīkṣās. The dominant concern in
allocating these initation types eventually became simply to differentiate the
householder initiate from the ascetic, rather than to distinguish the spiritual
goals attained through various initiations. This usage of the terminology is far
removed from its original purpose: designating special types of tantric intia-
tion for those seeking powers, access to higher cosmic realms, and, ultimately,
liberation.
Acknowledgements
I am very grateful to Shaman Hatley for his careful reading and extremely help-
ful comments, suggestions and edits of the text.
90 In Sathyanarayanan 2015, 15–63.

91 See Goodall’s entry on dīkṣā in Tāntrikābhidhānakośa, vol. III. For more on the influences
on the development and formation of Śaiva Siddhānta theology from the 10th century
onwards, see Goodall 2006, in particular pp. 98–103.
via free access
280 mirnig
References
Primary Sources
Jayadrathayāmala, Ṣaṭka 1. NAK 3–358; NGMPP A 995/6–996/1. Palm-leaf, Pāla script.
Jñānaratnāvalī of Jñānaśivācārya. Transcript of Re 1025/52 (57) copied by V. Rangas-
vami (GOML); photographed and transcribed by Dominic Goodall.
Tantrasadbhāva. E-text by Mark Dyczkowski, dated 1 August 2006, accessed through
the Muktabodha Indological Research Institute website, based on the following
manuscripts: NGMPP A188/22, śaivatantra 1533 (siglum k); NGMPP A44/1 (NAK 1–
363), śaivatantra (kh); NGMPP A44/2 (NAK 5–455), śaivatantra 185 (g).
Tantrāloka. Shāstrī, Mukund Rām, ed. Tantrāloka of Abhinavagupta, with the commen-
tary (-viveka) of Rājānaka Jayaratha. Kashmir Series of Texts and Studies, no. 59.
Bombay, 1938.
Netratantra. Madhusūdan Kaul Śāstrī, ed. Netratantra, with the Commentary (Netrod-
dyota) of Rājānaka Kṣemarāja. Kashmir Series of Texts and Studies, nos. 46, 59. Bom-
bay: 1926 and 1939.
Bṛhatkālottaratantra. NGMPP 4–131/VI Śaivatantra 127, Reel no. A43/1, Palm-leaf.
Mṛgendra. Bhatt, N.R., ed. Mṛgendra Kriyāpāda and Caryāpāda, with the commentary
(-vṛtti) of Bhaṭṭa Nārāyaṇakaṇṭha. Pondicherry: Institut français de Pondichéry,
1962.
Vimalāvatī of Vimalaśiva. NGMPP reel number A186/10. Eletronic text of the Mukta-
bodha Indological Research Institute typed under the supervision of Mark. S.G.
Dyczkowski.
Śaivasiddhāntaparibhāṣā of Śivāgrayogin. Oriental Research Institute Sanskrit Series,
no. 90, Madras: University of Madras, 1950. Eletronic text of the Muktabodha Indo-
logical Research Institute typed under the supervision of Mark. S.G. Dyczkow-
ski.
Siddhāntaśekhara of Viśvanātha. Sītārāma Somayājin and Śiva Śrī Talakāḍu Āvamika
Kṛṣṇadīkṣita, eds. Ubhayavedānti Viśvanāthakṛtaḥ Siddhāntaśekharaḥ. Manonma-
nīgranthamālā, no. 20. Mysore: K. Sītārāma Somayājin, 1971.
Somaśambhupaddhati. See Brunner 1977.
Somaśambhupaddhativyākhyā of Trilocana. IFP T. no. 170.
Svacchandatantra. Madhusudan Kaul Shāstrī, ed. Svacchandatantra, with the Commen-
tary (Svacchandoddyota) of Rājānaka Kṣemarāja. Kashmir Series of Texts and Stud-
ies, no. 38. Bombay: 1923.
Secondary Sources
Bhatt, N.R. 1962. Mṛgendrāgama. Kriyāpāda et Caryāpāda. Avec le commentaire de
Bhaṭṭa-Nārāyaṇakaṇṭha. Pondicherry: Institut Français de Pondichéry/ Ecole fran-
çaise d’Extrême-Orient.
via free access
Brunner, Hélène. 1975. “Le sādhaka, un personnage oublié du Sivaïsme du Sud.” Journal
Asiatique 263: 411–416.
Brunner, Hélène, ed. and trans. 1977. Somaśambhupaddhati, Troisième Partie. Rituels
occasionels dans la tradition śivaïte de l’Inde du Sud selon Somaśambhu II: dīkṣā,
abhiṣeka, vratoddhāra, antyeṣṭi, śrāddha. Pondicherry: Institut Français de Pondich-
éry/École française d’Extrême-Orient.
Brunner, Hélène. 1985. Mṛgendrāgama. Section des rites et section du comportement
avec la vṛtti de Bhaṭṭanārāyaṇakaṇṭha. Pondicherry: Institut Français de Pondich-
éry.
Goodall, Dominic. 1998. Bhaṭṭa Rāmakaṇṭha’s Commentary on the Kiraṇatantra. Vol-
ume I: chapters 1–6. Critical Edition and Annotated Translation. Pondicherry: Institut
Français de Pondichéry/École française d’Extrême-Orient.
Goodall, Dominic. 2000. “Problems of Name and Lineage: Relationships between South
Indian Authors of the Śaiva Siddhānta.” Journal of the Royal Asiatic Society 10 (2):
205–216.
Goodall, Dominic. 2004. The Parākhyatantra, a Scripture of the Śaiva Siddhānta. A Crit-
ical Edition and Annotated Translation. Pondicherry: Institut Français de Pondich-
éry/École française d’Extrême-Orient.
Goodall, Dominic, Harunaga Isaacson, and Alexis Sanderson, et al. 2015. The Niśvāsa-
tattvasaṃhitā. The Earliest Surviving Śaiva Tantra. Volume 1. A Critical Edition &
Annotated Translation of the Mūlasūtra, Uttarasūtra & Nayasūtra. Pondicherry:
Kiss, Csaba. 2015. The Brahmayāmala Tantra or Picumata, Volume Two. The Religious
Observances and Sexual Rituals of the Tantric Practitioner: chapters 3, 21 and 45.
Pondicherry: Institut Français de Pondichéry/École française d’Extrême-Orient/
Universität Hamburg.
Mirnig, Nina. 2013. “Śaiva Siddhānta Śrāddha. Towards an evaluation of socio-religious
landscape envisaged by pre-12th century sources.” In Puṣpikā. Tracing Ancient India
Through Cults and Traditions, edited by Nina Mirnig, Péter-Dániel Szántó and
Michael Williams, 283–301. Oxford: Oxbow Books.
Sanderson, Alexis. 1995. “Meaning in Tantric Ritual.” In Essais sur le Rituel III: Colloque
du Centenaire de la Section des Sciences religieuses de l’École Pratique des Hautes
Études, edited by Anne-Marie Blondeau and Kristofer Schipper, 15–95. Louvain-
Paris.
Sanderson, Alexis. 2001. “History through Textual Criticism in the Study of Śaivism, the
Pañcarātra and the Buddhist Yoginītantras.” In Les sources et le temps. Sources and
Time. A colloquium. Pondicherry 11–13 January 1997, edited by François Grimal, 1–47.
Pondicherry: Institut Français de Pondichéry/École française d’Extrême-Orient.
Sanderson, Alexis. 2004. “The Śaiva Religion Among the Khmers, Part I.” Bulletin de
l’École française d’Extrême-Orient 90: 352–464.
via free access
282 mirnig
King’s Brahmanical Chaplain.” Indo-Iranian Journal 47: 229–230.
Sanderson, Alexis. 2006. “The Lākulas: New Evidence of a System Intermediate
Between Pāñcārthika Pāśupatism and Āgamic Śaivism.”Indian Philosophical Annual
24: 143–217.
Sanderson, Alexis. 2007. “The Śaiva Exegesis of Kashmir.” In Mélanges tantriques à
la mémoire d’Hélène Brunner. Tantric Studies in Memory of Hélène Brunner, edited
by Dominic Goodall and André Padoux, 231–442. Pondichéry: Institut Français de
Pondichéry/École française d’Extrême-Orient.
Sanderson, Alexis. 2009. “The Śaiva Age—The Rise and Dominance of Śaivism in
the Early Medieval Period.” In The Genesis and Development of Tantrism, edited by
Shingo Einoo, 41–349. Tokyo: Institute of Oriental Culture, University of Tokyo.
Sanderson, Alexis. 2014. “The Śaiva Literature.” Journal of Indological Studies 24 & 25:
1–113.
Sathyanarayanan, R. 2015. With an introduction by Dominic Goodall. Śaiva Rites of
Expiation. A First Edition and Translation of Trilocanaśiva’s Twelfth-Century Prāya-
ścittasamuccaya (With a Transcription of Hṛdayaśiva’s Prāyaścittasamuccaya). Pon-
dicherry: Institut Français de Pondichéry/École française d’Extrême-Orient.
Takashima, Jun. 1992. “Dīkṣā in the Tantrāloka.” Institute for Advanced Studies on Asia
(Tohyobunkakenkyusho kiyo) 119: 45–84.
trique. Vol. III. Dominic Goodall et Marion Rastelli, eds. Beiträge zur Kultur- und
Geistesgeschichte Asiens, no. 76. Vienna: Austrian Academy of Sciences Press, 2013.
Vasudeva, Somdev. 2012. “Powers and Identities: Yoga Powers and the Tantric Śaiva Tra-
ditions.” In Yoga Powers: Extraordinary Capacities Attained Through Meditation and
Concentration, edited by Knut A. Jacobson, 265–302. Leiden: Brill.
Watson, Alex, Dominic Goodall and S.L.P. Anjaneya Sarma. 2013. An Enquiry into the
Nature of Liberation. Bhaṭṭa Rāmakaṇṭha’s Paramokṣanirāsakārikāvṛtti, a Commen-
tary on Sadyojyotiḥ’s Refutation of Twenty Conceptions of the Liberated State (mokṣa).
Pondicherry: Institut Français de Pondichéry/École française d’Extrême-Orient.
via free access
chapter 12
Innovation and Social Change in the Vale of

Kashmir, circa 900–1250C.E.
John Nemec
iti prakaṭito mayā sughaṭa eṣa mārgo navo mahāgurubhir ucyate sma śiva-
dṛṣṭiśāstre yathā |
Thus I have set forth this new (nava), easy path as it was explained by my
distinguished teacher [Somānanda] in the śāstra, [entitled] the Śivadṛṣṭi.
Īśvarapratyabhijñākārikās of Utpaladeva at 4.16ab
∵
1 Introduction
It is well known that religious agents in premodern South Asia appealed to the
purported timelessness and transcendence of their scriptural sources to secure
not only their legitimacy but also that of the ideas found in them. Equally well
known—in no small part due to the work of Alexis Sanderson on the social
history of Śaiva and other religious traditions in early-medieval Kashmir and
elsewhere—is the fact that premodern South Asian religions do not appear as
unchanging and immobile traditions in social stasis. Quite the opposite: the
various religious traditions of medieval South Asia were nothing if not inno-
vative in idea and practice, most notably in their literary productions. These
myriad religious traditions, moreover, had a measurable and not insignificant
influence on contemporaneous social life.
In beginning to address the question of religious change in premodern South
Asia I would like not merely to point out that the religious practitioners of the
day were surely able to distinguish new religious, and other, ideas and prac-
tices from received tradition—just as we are today—a fact that itself calls
into question the reification of the sort of social stasis and lack of historical
awareness posited in previous Indological scholarship.1 I also will argue that
1 Monier-Williams (1891, 38–39), for example, suggested that Brahminical theology preempts
© john nemec, 2020 | doi:10.1163/9789004432802_014
via free access
284 nemec
a self-conscious, emic theory or explanation of scriptural authority and social

change may be found in the history of religions in premodern Kashmir. I wish
to examine the significance of novelty, of innovation, as it was conceived by
the authors I propose to place under study, and to identify its role in establish-
ing, or challenging, religious authority in the period in question, all of which I
will do by exploring an exemplar that illustrates what I suggest should be taken
as a maxim in the study of South Asian religions (and religion more generally),
namely, that change is not inimical to religion, even if particular religious agents
are not infrequently inimical to change.2
In particular, I propose to examine selected writings of some among the
most prominent of the Śaiva tantric “post-scriptural” authors of the Kashmir
Valley, who thrived there in the period reserved by the title of the present
essay. They include Somānanda (circa 900–950), Utpaladeva (circa 925–975),
and Abhinavagupta (circa 975–1025), as well as Jayaratha (early 13th C.E.), the
author of the Tantrālokaviveka commentary on the Tantrāloka (hereafter TĀV
and TĀ, respectively). Their textual contributions are properly associated with
what continues, sometimes, to be labeled “Kashmiri Shaivism,”3 and they offer
an important opportunity to reevaluate the nature of various sources of reli-
any proper appreciation of history. MacDonell, in turn, both reiterates this position and adds
to it a second claim regarding what he calls the “total lack of the historical sense” in premod-
ern Indian literature: he claims that historical events in premodern South Asia were insuffi-
cient to trigger the cultivation of a properly historical consciousness. See MacDonell [1900]
1962, 8–9. This pair of views exemplifies the two types of arguments that earlier Indologists
have made regarding the supposed ahistoricality of premodern South Asian works: Schol-
ars often suggested either that the authors of premodern Sanskrit works held theoretical—
mainly religious—views that precluded them from taking interest in historical concerns; or,
they argued that historical events in premodern South Asia transpired in ways that preempted
the possibility of a properly historical response to them.
2 Scholars in the academy have given new attention to the question of the place of change in
religion in recent years. Jonathan Edelmann, formerly of Mississippi State University and now
at the University of Florida, has suggested, in an unpublished précis of concerns addressed by
four panelists of a session of the 2011 American Academy of Religion National Meeting enti-
tled “Authorizing Theologies,” that the question to hand was a matter of “newness,” of novelty,
and of “authorizing” such novelty in the language of “theology,” with the shared hypothesis
of all the panelists being that the various theological formulations they placed under exami-
nation each sought to “recast and re-contextualize influential concepts and arguments from
earlier traditions” (to quote from the panelists’ description of the session). Some of the prod-
ucts of this panel may be found in a special issue of the International Journal of Hindu Studies,
for which see Edelmann 2014. The present essay constitutes a response to and an engagement
with the broader theme first engaged by these scholars at the AAR Annual Meeting in ques-
tion.
3 It is well known by now that this term is something of a misnomer, because it is both overly
specific geographically and overly general doctrinally. See Dyczkowski 1987: 222–223.
via free access
innovation and social change in the vale of kashmir 285
gious authority in premodern South Asia, as well as the manner of establishing

and deploying such authority in practice in India’s premodernity.
This is so, firstly and primarily, because the authors in question explicitly
wedded themselves to the notion that novelty, the production of something
new, was not only acceptable, but was an asset to their religion. This they
maintained over and against the ethos of Vedic religious authority, which was
defined by its claim to offer a kind of knowledge that was beyond question,
its infallibility guaranteed by the purported timelessness and authorlessness
of the Vedic scriptures themselves. Secondly, and related to this, they lived
and wrote in the Valley at a time when Kashmir was economically dynamic
and culturally and religiously pluralistic, and was perhaps the most active cen-
ter of Sanskrit literary production in the Indian sub-continent not only in its
day, but in Indian history (after the time of the Guptas).4 Indeed, the bibli-
4 Royal patronage certainly did much to cultivate this cultural richness, as the many Kashmiri
contributions in poetry and aesthetic theory illustrate, these of course being important con-
cerns of the premodern South Asian court; but this was not the only catalyst. Five additional
elements served to make such cultural production possible. These include:
(1) The relative security of Kashmir. Closing the famed “gates” to the Valley—the moun-
tain passes or dvāras—offered reliable protection from outside military interference, though
this did not help quell internal military threats, as the Rājataraṅgiṇī makes abundantly clear.
(2) This does not mean, however, that Kashmiris were unaware of outside military and
other influences. One anecdote illustrates as much: to my knowledge the earliest use in San-
skrit of the term turuṣka appears in a Kashmiri work, the famed Kuṭṭanīmata of Dāmodara-
gupta; and the term frequently appears in later Kashmiri works such as the Rājataraṅgiṇī and
the Kathāsaritsāgara. So much signals the second catalyst: cosmopolitanism, or a cognizance
of the wider world and significant contact with it. Foreign invaders were long kept at bay, to
be sure; but, simultaneously, scholars and merchants regularly were given leave to visit the
Valley from the plains of India and from as far afield as Central Asia and China, and possi-
bly beyond. Perhaps most notably, Tibetan Buddhist pilgrims frequented Kashmir in order to
study in the Valley’s thriving monasteries, which were influential in the period in question
even if relatively few archaeological traces of them have survived to the present day.
(3) Next, the extensive and longstanding cross-pollination of Hindu and Buddhist thought
did much to enrich humanistic thinking in the Valley in the period under study; it helped that,
with relatively few exceptions, the kings of Kashmir supported a healthy religious diversity
in the Valley. (Harṣa of course is one exception to the general rule that Kashmiri kings, and
members of the court, patronized various traditions simultaneously, and certainly lived up to
the Dharmaśāstric norm of leaving religion alone short of the emergence of egregious con-
cerns therewith.) I would count this intellectual and religious pluralism the third productive
factor in Kashmiri intellectual and cultural history.
(4) A fourth influence was economic: global trade along the nearby Silk Road is likely to
have contributed to the Valley’s material prosperity, which would have facilitated the culti-
vation of Kashmir’s cultural wealth (though more work is needed better to understand this
dimension of Kashmiri social life); and the Valley itself, with its large tracts of arable land,
was agriculturally largely self-sufficient.
via free access
286 nemec
ography of Kashmiri contributions, one feels, is practically asymptotic to that

of premodern Sanskrit learning tout court.5 Quite a number of innovative lit-
erary works were produced there in this period, moreover; and some among
the Śaiva authors in question actively participated in this cultural and liter-
ary tradition by composing texts that contributed to selected non-tantric lit-
erary genres. A study of these authors thus opens a window into the ways in
which religion functioned in larger intellectual and public spheres in a period
of tremendous intellectual and cultural flourishing, even if the present essay
will examine the internal logic of the selected tradition’s philosophical and
(5) Finally, one cannot discount the significance of the location itself: Kashmir was—and
is—considered to be one of the most beautiful places of the entire sub-continent, and people
simply wanted to live there. This in part explains the long history of Brahminical migration
(at the king’s invitation) to the Valley from across the sub-continent.
5 The Kashmiri contributions are of course numerous and cover the gamut of fields of San-
skritic learning, from Dharmaśāstra to philosophical writings to vyākaraṇa and other tech-
nical literatures, to works of belles lettres in a range of genres, and, of particular note, to the
alaṃkāraśāstra. Consider the following. Two of the five works that are demonstrably based
on Guṇādhya’s Bṛhatkathā were composed in Kashmir (these are the Bṛhatkathāmañjarī of
Kṣemendra and Somadeva’s famed Kathāsaritsāgara). Virtually all the ālaṃkārikas of the
mature phase of the study of aesthetics were Kashmiri—including Vāmana, Bhaṭṭa Udbhaṭa,
Rudraṭa, Ānandavardhana, Bhaṭṭa Nāyaka, Kuntaka, Mukulabhaṭṭa, Mammaṭa, Ruyyaka, and
Pratīhārendurāja. In the various areas of Śāstric learning, the famed Naiyāyika Bhaṭṭa Jayanta
was Kashmiri, as were Varadarāja and possibly Vyomaśiva. Among the Buddhists, Arcaṭa,
Dharmottara, Śaṅkaranandana, and Jñānaśrīmitra hailed from the Valley, and Vinītadeva
(among many others) spent time there. The grammarians Jayāditya and Vāmana, authors of
the Kāśikāvṛtti, are thought by some to have been Kashmiris, and Kaiyaṭa certainly was (and
is said to have been the son of Mammaṭa). So, too, was Helārāja a Kashmiri (of course he is
the author of the Prakīrṇaprakāśa commentary on the Vākyapadīya). Medhātithi, the famed
early commentator on the Mānavadharmaśāstra, also hailed from the Valley, as, of course,
did Kalhaṇa, the author of the Rājataraṅgiṇī.
Additionally, a number of key works of unnamed authorship can be shown to have Kash-
miri origins, including the Viṣṇudharmottarapurāṇa, the Haracaritacintāmaṇi, and the Mok-
ṣopāya, which, as by now is well known, is the early recension of the Yogavāsiṣṭha. Among
Śaiva scriptural works, we can say with some confidence that the Bṛhatkālottara, the sec-
ond, third, and fourth ṣatkas of the Jayadrathayāmalatantra, the Svacchandatantra, the
Netratantra, and, in all probability, the Tridaśaḍāmaratantra (about which I wrote my M.Phil.
thesis at Oxford under the direction of Alexis Sanderson, the product of which may be found
in evidence in Nemec 2013) are Kashmiri productions. Among Pāñcarātrika scriptures (this
according to Alexis Sanderson), the Jayākhya-, Jayottara-, and Sāsvata-saṃhitās are of a
Kashmiri provenance. Finally, I note that many other works were also influential in the Valley,
and were heavily studied there, even if they cannot be proven to have a Kashmiri provenance.
For a survey of Kashmiri authors and the scholarship that treats them, see Nemec 2015. (I
thank Harunaga Isaacson for his contribution to my understanding of the provenance of
these anonymous tantric works, which came in the form of personal correspondence, via
email, in December of 2013.)
via free access
religious writings rather than explain the relationship between such writings
and those of other, more “secular” genres.
The wider cultural context of the Kashmir Valley is of note, because schol-
ars have sometimes linked the purportedly transcendent nature of religious
authority—and of textual production more generally—to a concomitant,
forced stasis in the social order. Rigid and purportedly timeless ideas about
the nature of reality and the place of various peoples situated in it, the think-
ing goes, produced and supported rigid rules concerning the innate rights and
capacities of particular individuals and groups, with such rights and capacities
inevitably being hierarchically and inequitably defined, all in a manner as resis-
tant to change as the transcendent sources of religiously sanctioned authority
were held to be. The precise social implications of a religious worldview that
welcomes innovation, in a context in which innovation in literary production
was not uncommon, is therefore of primary concern in defining the role of reli-
gion in public life in premodern South Asia.6
In what follows I will proceed in three stages. First, I will offer a précis of
the problematic associated with the issues to hand, this by examining Pollock’s
theorization of the relationship of theory (śāstra) to practice (or prayoga) in
premodern South Asia. Pollock is selected for further examination because his
writings on premodern South Asian cultural change are, in my view, the most
sophisticated and engaging of any scholarly treatment of the subject to date,
even if I will ultimately adopt his view only in part, challenging it in part as
well.
Following this, I will examine key textual passages from the writings of the
tradition here placed under study, this to illustrate the ways in which tran-
scendentally authorized religious ideas can be conceived simultaneously to
be historically situated and, indeed, new—and therefore demanding of a new
social consciousness. Here I wish to support the claim that social change in pre-
modern South Asia tends almost invariably to be incremental, not revolutionary,
in nature. It regularly—not to say exclusively—involves modifications of social
norms and strictures, not wholesale changes thereof. The relevant textual pas-
sages will be examined in two sections, the first dealing with the tradition’s
emic theory of textual authority, the second with its religious practices.
6 The idea, then, is to begin, at least, to consider the tantric Śaivas’ religion in light of the exis-
tence of their non-tantric lives (if not explicitly their non-tantric textual contributions), as
well to measure their religious writings with the wider social, cultural, and intellectual con-
texts in which they thrived in mind. I intend, anyhow, to make a start at this here by examining
the role of innovation, of change, in religion, even if any exploration of these themes more
broadly will have to wait for a subsequent study.
via free access
288 nemec
Finally, I will conclude by charting the implications of this exemplar in a

concentric series of contexts, culminating with the study of religion, broadly
conceived. If I might be allowed to argue by way of example—ab uno disce
omnes—I will suggest in conclusion that the present illustration offers insight
into the nature and functioning not only of religion in premodern India, but
also of religion tout court.
2 Theory and Practice: Sheldon Pollock on śāstra and prayoga
The issues associated with endeavors to negotiate cultural and social change
may be set in the brightest relief if we read them in light of what is in my
view the most sophisticated treatment to date of the relation of śāstra—i.e.,
(textual) theory—on the one hand, and cultural and intellectual practice or
prayoga, on the other, that of Sheldon Pollock (as articulated in a pair of arti-
cles: “Mīmāṃsā and the Problem of History in Traditional India” [1989] and,
perhaps more importantly, “The Theory of Practice and the Practice of Theory
in Indian Intellectual History” [1985]). Pollock has argued, in what is the most
significant theoretical contribution of this pair of articles, that classical San-
skrit theoretical works in premodern South Asia articulated an epistemological
formulation according to which there was no dialectical relationship between
theory (śāstra) and practice (prayoga). New practices were not seen as able to
challenge or shape theoretical models, Pollock has argued, due to the claimed a
priori status of the theoretical works themselves, which came to be understood
to represent (in, he suggests, the language of Ryle) a form of “knowing that” a
given social or cultural form was in its very nature just as it appeared in prac-
tice, as opposed merely to offering a “knowing how” the social or cultural form
in question functioned or could function (based on empirical observation).7
This is to say that the purportedly timeless and authorless works defined
the world in terms of what it is and should—and must—be, rather than artic-
ulating a contingent theory, one subject to revision on the basis of on-the-
ground practices, or in other words in a dialectical relationship with events.
“For here,” Pollock argues, “on a scale probably unparalleled in the pre-modern
world, we find a thorough transformation—adopting now Geertz’s well-known
dichotomy—of ‘models of’ human activity into ‘models for,’ whereby texts that
initially had shaped themselves to reality so as to make it ‘graspable,’ end by
7 See Pollock 1985, 504.
via free access
asserting the authority to shape reality to themselves.”8 Pollock ultimately iden-

tifies three consequences to this theory of theory and its practical application:
(1) the “‘creation’ of knowledge” was understood to be exclusively a divine activ-
ity; (2) knowledge was “by and large viewed as permanently fixed in its dimen-
sions”; and (3) third and finally, there could be “no conception of progress, of
the forward ‘movement from worse to better,’ on the basis of innovations in
practice.”9
One product of such “ahistoricality”10 was an apparent stasis in Brahminical
views of the normative social order, at least as such views were codified in the
classical Sanskrit śāstras par excellence—the Vedas, which, to reiterate, were
reified as authorless and timeless works and said therefore to convey transcen-
dent and unchanging truths.11 The śāstras were self-consciously understood to
8 ibid.
10 This is the term so often used to describe Indian attitudes toward history—and change.
Pollock is not alone in using it (though one should be careful to note that he deploys it
differently, and in a markedly more self-conscious and sensitive manner, than his prede-
cessors). So much was it perceived to be the case that India knew nothing of history, that
some Western Indologists around the turn of the last century went so far as to suggest,
per A.A. MacDonell, for example, that “early India wrote no history because it never made
any.” See MacDonell [1900] 1962, 8–9; cf. footnote 1, above; finally, see footnotes 11 and 27,
below, for Pollock’s use of the term in question.
11 As Pollock (1989, 606) understands it, the practice of producing “a content invariably
marked by ahistoricality” has its root in the early history of Vedic exegesis, which sought
to establish the authority of the Vedas by claiming them to be timeless and authorless
productions—ahistorical productions—in which any apparent references to contempo-
raneous events or persons were said to be explicable by a variety of analyses that served
to vitiate any reference to historical realities. See Pollock 1989, 608:
Mīmāṃsā holds on empirical grounds that the tradition of the recitation of the Vedas
must be beginningless (uktaṃ tu śabdapūrvatvam, PMS 1.1.29; cf. Ślokavārttika, Vākyā-
dhikaraṇa, vs. 366). But that is not sufficient to prove its transcendence and thus infal-
libility (something false can be beginningless, the jātyandhaparaṃparānyāya). It is
therefore argued that the Vedas are transcendent by reason of their anonymity. Had
they been composed by men, albeit long ago, there is no reason why the memory of
these composers should not have been preserved to us. Those men who are named
in association with particular recensions, books, hymns of the Vedas—Kaṭhaka, for
example, or Paippalādaka—are not to be regarded as the authors but simply as schol-
ars specializing in the transmission or exposition of the texts in question (ākhyā prava-
canāt, 1.1.30; pūrvapakṣa ad 1.1.27). Texts for which no authors can be identified have
no authors, and this applies to the Vedas and to the Vedas alone (which are thus pre-
sumably the only authentically anonymous texts in Indian cultural history).
See also ibid.:
The transcendent character of the Vedas, which is proved by the fact of their having
no beginning in time and no author, is confirmed by their contents: the Vedas show
via free access
290 nemec
articulate a range of social and other normative views that were consciously
held to be unassailable, because the works in which they were recorded were
said to stand entirely outside the realm of mundane and contingent existence:
History, one might thus conclude, is not simply absent from or unknown
to Sanskritic India; rather it is denied in favor of a model of ‘truth’ that
accorded history no epistemological value or social significance. The
denial of history, for its part, raises an entirely new set of questions. To
answer these we would want to explore the complex ideological forma-
tion of traditional Indian society that privileges system over process—the
structure of the social order over the creative role of man in history—and
that, by denying the historical transformations of the past, deny them for
the future and thus serve to naturalize the present and its asymmetrical
relations of power.12
no dimension whatever of historical referentiality. Allusions to historical persons or to

historical sequentiality are only apparently so. For instance, the Vedic sentence ‘Babara
Prāvāhaṇi [son of Pravāha] once desired …’ [TS 7.1.5.4]—which might establish a ter-
minus post quem for the composition of the text (i.e., after Pravāha begot Babara)—
contains merely phonemic resemblances to the names of historical persons (paraṃ tu
śrutisāmānyamātram 1.1.31; pūrvapakṣa ad 1.1.28). ‘Etymological’ analysis shows that
the references are in fact to eternally existing entities (in the case in question, to the
‘howling wind’).
Note that Pollock thus challenges the notion, maintained by earlier Indologists such as
MacDonell, that such ahistoricality reflects any lack or deficiency in Indian intellectual
historical awareness, and he clearly would disagree, as I emphatically do, with those earlier
Indologists who denigrated Indian thinkers for failing to develop a “proper” (Western-like)
historical consciousness. See, e.g., Pollock 1989, 604: “I believe the received view about
Indian historical consciousness is constructed out of a set of ideas whose truth can no
longer be taken for granted: ideas about history and narrativity as such, about ancient
historiography in general and Indian intellectual history in particular. Moreover, even if
we grant that there are idiosyncratic features about the traditional Indian response to
historical experience, the characteristics of this idiosyncrasy have never been adequately
described or convincingly explained.” (See also, ibid.: “On the face of it the reduction of
historical consciousness to a ‘zero-category’ for traditional India is improbable; from the
viewpoint of phenomenology, which offers us the most sustained analysis of such con-
sciousness, it is impossible.”) Rather, Pollock argues, any “ahistoricality” is intentionally
constructed, in his view to perpetuate the social control of those (Brahmins) who con-
structed it. See below.
via free access
3 The Kashmiri Śaiva Example in Its Theoretical Form
If I question Pollock’s model in part, it is not because it is nowhere operative in

premodern South Asia or, mutatis mutandis, in contemporary Indian religions.
For, decidedly it was, and sometimes is; and Pollock’s theory adroitly explains
a strategy for dealing with cultural and social change that in its primary fea-
tures is also operative in other religious traditions of other times and places.
Indeed, it signals what is a fundamental concern for religion and does much
to theorize and explain what is a common response to it.13 Where I differ with
Pollock is with his understanding of this model as monolithic, as functioning
equally across the gamut of premodern śāstric modes of discourse and across
textual genres and religious traditions.14 For the facts as we can discern them
“on the ground” in the period in question challenge the universal applicability
of his model, and the construction and use of religious authority in premodern
South Asia was regularly if not uniformly more complex than Pollock’s model
allows.
Two dimensions of textual production in premodern South Asia compli-
cate his view of the śāstras—the normative, technical, and for our purposes
religious writings—of classical India. These are, firstly, (1) the fact that the
canon—or, more precisely, canons—of Hindu scripture remained (and in cer-
tain instances continue to remain) open, allowing for a proliferation of scrip-
tural sources regularly to emerge “on the ground”; and, secondly, (2) the fact
that “ahistoricality”—the very basis of the authorless and timeless transcen-
dence that, Pollock has shown, furnished (religious) authority—was conceived
in multiple ways, including, in the tradition placed under consideration in
this essay, in a manner that allowed purportedly divine and timeless teachings
simultaneously to be associated with the particular biographies of historically-
located religious figures (“perfected ones” or Siddhas, in the present example).
13 The concern may be summed up in plain terms, namely, that while the world changes, the
religious ideas, ideals, practices, and principles that are meant to shape and guide human
activity are established before such changes can be known or fully anticipated. More than
this, religious authority not infrequently is derived from its very antiquity, from the fact
that it is born from what is said to be a transcendent source or sources, which can speak
“truths” neither produced by, nor subject to the revisions of, the fallible and fickle. There
is a need, then, to prove the prescience of the religiously authoritative sources, either
through exegesis or eisegesis: one may do so either by finding events as they eventually
transpired to have been anticipated in the scriptures or by reading events as they tran-
spire into the scriptures themselves.
14 On the ubiquity of this strategy, in Pollock’s view, see footnote 27, where he refers to the
use of the strategy in “whatever sort of [Sanskrit] text it might be” that one has to hand.
via free access
292 nemec
This pair of dimensions of scriptural production allowed for the recognition

of and advocacy for change in Hinduism in ways that have heretofore passed
without mention: religious authority could be conceived of as simultaneously
new and timeless; and in a context in which religions could readily proliferate
textual sources of a staggering variety, religious authority was complex, both
because it was layered and because it could and did change over time. It is
precisely for these reasons that the example offered by the Śaiva tantrics of
Kashmir both challenges and confirms elements of Pollock’s theory.
I begin the case-study with the examination of a theoretical work, a śāstra,
one inspired by Śaiva tantric scriptures and views, but one voided of explicit ref-
erences to such texts: Utpaladeva’s Īśvarapratyabhijñākārikās (hereafter ĪPK),
the preeminent expression of Pratyabhijñā philosophy, for which the author
wrote both a short and a long autocommentary, respectively the Īśvarapratya-
bhijñākārikāvṛtti (ĪPKVṛ) and the Īśvarapratyabhijñākārikā-vivṛti or -ṭīkā (the
latter of which is lost excepting that a number of fragments of the work have
been recovered in recent years; see chapter 5 of this volume). The ĪPK and ĪPKVṛ
offer an exoteric explanation for the nature of the divine and the place of the
practitioner in the world. They were written by a Brahmin who in his pub-
lic life would have adhered to caste norms and the associated rules of purity
and pollution. Yet, as is well known, Utpaladeva and the other authors of the
Pratyabhijñā simultaneously would have allowed, in theory, at least, for the
contravention of these rules in the context of the family of tantric initiates, a
complex practice summarized by the famous dictum, “internally a Kaula, exter-
nally a Śaiva, and in social practice a follower of the Veda.”15 (I will return to this
maxim in what follows in order to examine the wider textual context in which
it is recorded.)
My contention is that the ĪPK and its commentarial tradition exemplify the
possibility of self-consciously introducing religiously—indeed, what are held
to be transcendently—authorized innovations, and that they also serve to illus-
trate and explain the possibilities and limits of such innovations on contempo-
raneous social life. Consider, to begin, the by now well-known invocatory verse
(maṅgala) of Utpaladeva’s masterwork:
15 See TĀV ad TĀ 4.251ab (vol. 3, p. 278, line 5): antaḥ kaulo bahiḥ śaivo lokācāre tu vaidikaḥ.
Cf. TĀV ad TĀ 4.24cd–25ab (vol. 3, p. 27, line 11).
I note that there is a difference in the writings of Utpaladeva, Somānanda, and Abhi-
navagupta as regards the place of the norms of varṇāśramadharma; for, the latter two
authors were more explicit in their understanding of the merely conventional nature of
such social distinctions than was Utpaladeva (in the ĪPK and ĪPKVṛ, at least).
via free access
Having somehow become a servant of Maheśvara, wishing to offer assis-

tance to all people ( janasyāpi), I will facilitate the recognition of [the
Lord], which brings about the achievement of all successes.16
Utpaladeva’s stated wish to assist “all people” ( janasyāpi) requires further

examination, and I will return to this statement in what follows. More immedi-
ately, one may note two key and perhaps obvious qualities of the maṅgala as it
is offered to the reader. First, Utpaladeva claims access to the transcendent, for
he says he has “somehow become a servant of Maheśvara,” i.e., of Śiva. Second,
he promises to bring others to a similar level of accomplishment by account-
ing for his insight in the very text that is to hand, paving a path for others to the
same ennobled state.
The first quality is anticipated in the maṅgala with which Somānanda opens
his Śivadṛṣṭi (hereafter ŚD), a verse that is also by now well known:
May Śiva, who has penetrated/possessed my form by warding himself off

by means of his own self, pay homage to his (all-)extensive self by means
of his own power.17
Here, as before, the author’s association with Śiva is noted, though it is rather
more prominently asseverated than is the association with Śiva described by
Utpala in the ĪPK—for the ŚD uses the Kaula scriptural idiom of possession.
We thus clearly are not met with those who count themselves merely to be
informed intellectuals, but with those who claim to be both scholars and practi-
tioners, who understand themselves to be close to—or in fact the very embod-
iment of—God in the form of Śiva.
So much is confirmed in the history of the transmission of the teachings of
the Pratyabhijñā that is found in a textual passage appended to the Śivadṛṣṭi,
said to be an autobiographical account of Utpaladeva’s teacher, Somānanda.18
16 See ĪPK 1.1.1: kathañcid āsādya maheśvarasya dāsyaṃ janasyāpy upakāram icchan | samas-
tasaṃpatsamavāptihetuṃ tatpratyabhijñām upapādayāmi.
17 See ŚD 1.1: asmadrūpasamāviṣṭaḥ svātmanātmanivāraṇe | śivaḥ karotu nijayā namaḥ śak-
tyā tatātmane.
18 An edition and translation of this “autobiography” may be found in Nemec 2011, 22–24.
As noted there, I have my doubts about the authenticity of the passage as autobiogra-
phy, and rather think it is something probably composed after Somānanda’s day, but I
here note that it is likely to have existed by the eleventh century, as it is replicated, to
a substantial degree, in Abhinavagupta’s Īśvarapratyabhijñāvivṛtivimarśinī (ĪPVV), which
explicitly associates the narrative with the ŚD itself. (See below, especially footnote 33.)
via free access
294 nemec
While this quasi-mythological passage claims a divine origin for the Pratyabhi-
jñā, origins that were shrouded with secrecy, it emphasizes the emergent quality
of the teachings—their very entry from obscurity into circulation among men,
or, in a word, their novelty. For the teachings, we are told therein,19 were kept
by “great-souled sages” (mahātmanām ṛṣīṇām) until they retreated from the
world at the dawn of the Kali Yuga, at which point it was incumbent on God
(deva) himself to cross down to earth “in order to grace [humanity]” (devaḥ
… anugrahāyāvatīrṇaḥ … bhūtale).20 This he did by instructing the irascible
and imprecating sage Durvāsas to perpetuate the lineage of initiates. Durvāsas
is said to do so by producing a mind-born son named Tryambakāditya, who
stands at the head of a fifteen-generation lineage of such mind-born Siddhas
(all named Tryambaka).21 The last falls in love with a Brahmin woman who
bears him a son named Saṃgamāditya. He in turn is said to have moved to
Kashmir,22 and over the course of the generations the lineage and its esoteric
teachings were passed down to Somānanda. Thus, we are told, the teachings
were not lost (nocchidyate).
I would further suggest that the type of autobiographical expression offered
by Somānanda is also found in Abhinavagupta’s famed autograph verse, with
which he opens virtually all his Trika commentarial works and which Sander-
son has examined, already, in detail. Sanderson (2005, 89) has rendered this
bivalent (śleṣa) maṅgala verse as follows:
May my heart shine forth, embodying the bliss of the ultimate, [for it is]
{one with the state of absolute potential made manifest in the fusion of
these two, the ‘Mother’ grounded in pure representation, radiant in ever
new genesis, and the ‘Father’, all-enfolding [Bhairava], who maintains the
light [of consciousness] through his five faces} / { formed from the emis-
sions produced through the fusion of these two, my mother Vimalā, whose
References to verse numbers of this “autobiography” reflect the enumeration of the edi-
tion found in Nemec 2011.
19 See the “autobiography” at ŚD 7.107a–c. śaivādīni rahasyāni pūrvam āsan mahātmanām |
ṛṣīṇāṃ vaktrakuhare …
20 See the “autobiography” at ŚD 7.108–109: kalau pravṛtte yāteṣu teṣu durgamagocaram
| kalāpigrāmapramukham ucchinne śivaśāsane. kailāsādau bhraman devo mūrtyā śrīka-
ṇṭharūpayā | anugrahāyāvatīrṇaś codayām āsa bhūtale.
21 See the “autobiography” at ŚD 7.111: tataḥ sa bhagavān devād ādeśaṃ prāpya yatnataḥ
| sasarja mānasaṃ putraṃ tryambakādityanāmakam. See also the “autobiography” at
ŚD 7.114: siddhas tadvat sutotpattyā siddhā evaṃ caturdaśa | yāvat pañcadaśaḥ putraḥ sar-
vaśāstraviśāradaḥ.
22 The move to Kashmir is recounted in the narrative at ŚD 7.118cd.
via free access
greatest joy was in my birth, and my father [Nara]siṃhagupta, [when both

were] all-embracing [in their union]}.23
These expressions, in turn, reflect the tradition of the poet’s autograph verse,
which, it must be added, is something of a late phenomenon in the history of
Indian literature. I offer for comparative purposes one example thereof, that of
Ratnākara (as rendered by David Smith), which opens his famed Haravijaya:
Ratnākara, son of Amṛtabhānu who was the son of him who lived at Gaṅ-
gāhrada on Himādri’s peak and was of Śrīdurgadatta’s lineage, wrote this
poem, a lovely composition, which is beautiful because it is based on the
deeds of Śiva, whose crest is the digit of the moon.24
The similarities lie precisely in the fact that by identifying his parentage and/or
family lineage each author stakes a claim to his unique biography and thereby
his historicality, even if each does so without explicitly situating himself histori-
cally with any traceable reference to a calendrical date or reign of any particular
royal sovereign. The autograph, then, furnishes precisely what this term sug-
gests, a marker of the unique identity of the author.25
23 The verse reads as follows: vimalakalāśrayābhinavasṛṣṭimahā jananī bharitatanuś ca pa-

ñcamukhaguptarucir janakaḥ | tadubhayayāmalasphuritabhāvavisargamayaṃ hṛdayam
anuttarāmṛtakulaṃ mama saṃsphuratāt. See Sanderson 2005, 89.
24 The verse reads as follows: śrīdurgadattanijavaṃśahimādrisānugaṅgāhradāśrayasuta-
̄mṛtabhānusūnuḥ | ratnākaro lalitabandham idaṃ vyadhatta candrārdhacūlacaritāśraya-
cāru kāvyam. See Smith 1985, 104.
25 One may observe that only authors of particular genres reveal much of anything of them-
selves: some authors never compose autograph verses, or anything similar to them. Thus,
while the poets who clearly worked in the royal courts of Kashmir reveal parts of their
own biographies to the reader, or at the least are willing to acknowledge themselves as
biographical figures, we conversely know little from the playwrights of their parentage
or their place in the world, nor do the philosophers reveal much of themselves (though
Kalhaṇa places many such figures in history in his Rājataraṅgiṇī).
It is clear that the Śaivas of concern in this essay were willing to think of and talk
about themselves in a language that best pairs—at least up until the time when they
lived—with the literary agents who operated in the “secular” arena of the royal court. In
Somānanda’s case, his personal story is, as we have seen, offered in semi-mythological
terms; Utpaladeva’s is presented in the maṅgala of a text that he surely would have had
circulated in intellectual circles, probably those connected, in some significant ways, to
the royal court. All of this clearly signals some sense of—some self-conscious awareness
and articulation of—historical location, which is proximate to a self-conscious awareness
and articulation of novelty. Abhinavagupta’s autograph verse in particular is, in my view,
most like those of the poets, and it certainly indicates his very entrance into historicality,
however extraordinary and exulted he wished to suggest his entrance may have been.
via free access
296 nemec
It bears reiterating that the historical accuracy of such accounts is not here
pertinent; what is of concern in these and other similar expressions is the fact
that their authors conceive of themselves not only in terms of their historicality,
the mere fact that they appear in time, but also in terms of their historicity:26 the
authors in question claim for themselves—and whatever (religious) authority
they claim—the possibility of being mapped historically, located in particular
social, political, religious, and other contexts due to their own distinguishable
personages and individual biographies.
Most significant is the fact that Utpaladeva is explicit in the ĪPK about the
nature and significance of the spiritual path he describes: it is not only efficient
and effective, but it is explicitly new.27 This Utpaladeva tells us in the concluding
passage of the fourth and final section (adhikāra) of the work, where, reflecting
on his completed text, he says:
26 I here adopt the distinction between historicality and historicity as defined by J.N.
Mohanty (2010, 334): “Historicality is the property of being in history. Historicity is the
condition of the possibility of history. One can distinguish between three levels of his-
toricity: historicity of human existence; historicity of a culture; and historicity of scientific
thinking and, in the long run, philosophy. A mere fact or an event does not have historicity.
Only an ideal meaning, as Derrida insists in his work on Husserl, can have it.”
27 That this is so is significant, for it is indeed the case that such expressions of novelty are
relatively scarce in premodern Sanskrit sources, as Pollock has noted. But they are not
entirely absent, and one suspects that scholars will find a greater number of such pas-
sages as they continue to look for them. See Pollock 1989, 606: “To an astonishing degree
Sanskrit texts are anonymous or pseudonymous, or might just as well be. The strategy
of eliminating from the text—whatever sort of text it might be—the personality of the
author and anything else that could help us situate the text in time is a formal correlate of
a content invariably marked by ahistoricality. Works on statecraft, for example, describe
their subject without specific reference to a single historically existing state. Books on law
expatiate on such crucial questions as the relationship between local practices and gen-
eral codes of conduct without adducing any particularized events or cases. Belles-lettres
seem virtually without date or place, or indeed, author. Literary criticism prior to the
tenth century (Ānandavardhana) neither mentions the name nor cites the work of any
poet, the ālaṃkārikas themselves supplying all examples. Philosophical disputation takes
place without the oppositional interlocutor ever being named and doxographies make
no attempt to ascribe the religious-philosophical doctrines they review to anyone, unless
a mythological personage. Even in those texts whose historical vision I suggest merits
particular reconsideration—Raghuvaṃśa, for instance—referentiality remains somehow
unanchored: We are indeed told that it is the Bengalis that Raghu uproots (4.36), the Kalin-
gas he attacks (38), the Pandyas he scorches (49), the hair of the Kerala women upon which
he sprinkles the dust of his army (54), and so on, but if the digvijaya has local contours, it
remains essentially timeless. In short, we can read thousands of pages of Sanskrit on any
imaginable subject and not encounter a single passing reference to a historical person,
place, or event—or at least to any that, historically speaking, matters.”
via free access
Thus I have set forth this new (nava), easy path as it was explained
by my distinguished teacher [Somānanda] in the śāstra, [entitled] the
Śivadṛṣṭi.28
We see this sentiment reiterated, moreover, in the ĪPKVṛ, where Utpaladeva

glosses the term “new” (nava) with the synonym abhinava;29 and Abhinav-
agupta reproduces this latter term in the shorter of his two sub-commentaries,
the Īśvarapratyabhijñāvimarśinī (ĪPV), where he adds that the path in question
is both “included in” all the various secret śāstras (sarvarahasyaśāstrāntargata)
and, because it has been concealed from public view, is not well known (san-
nigūḍhatvād aprasiddha).30
Isabelle Ratié, the first scholar explicitly to note the significance of this
passage, has suggested that Utpaladeva refers to the novelty of the path he
described in the ĪPK because he wishes to indicate that it does not authenti-
cate itself by any overt appeal to scripture.31 Now, it is undoubtedly the case
that the Kārikās are unmistakably philosophical in tone and content, and that
they require no special or extensive knowledge of esoteric scriptural sources for
their comprehension; and one may note that Utpaladeva’s teacher, Somānanda,
similarly obscured the scriptural influences on his articulation of the Pratya-
bhijñā.32
28 See ĪPK 4.16ab: iti prakaṭito mayā sughaṭa eṣa mārgo navo mahāgurubhir ucyate sma
śivadṛṣṭiśāstre yathā |.
29 See ĪPKVṛ ad ĪPK 4.16: so ’yam avakra evābhinavo mārgaḥ sākṣātkṛtaparameśvarabhaṭ-
ṭārakākārair bhaṭṭaśrīsomānandapādaiḥ śivadṛṣṭināmni prakaraṇe nirdiṣṭo mayā yuktini-
bandhanena hṛdayaṃgamīkṛtaḥ. etatpariśīlanena śivatāveśāt jīvann eva mukto bhavati.
“This new, direct path was foretold in the treatise entitled the Śivadṛṣṭi by the venerable
Somānanda, whose very appearance is that of the great lord Parameśvara in front of one’s
eyes; I have made it [i.e., this path] enter the heart(s) (of men) by furnishing a logical justi-
fication for it. By pursuing this [path] one becomes liberated in this very life [ jīvann eva],
this as a result of being (fully) penetrated by Śiva-nature.”
30 See ĪPV ad ĪPK 4.16: asmatparasmeṣṭhibhaṭṭārakaśrīsomānandapādaiḥ śivadṛṣṭiśāstre
’yam abhinavaḥ sarvarahasyaśāstrāntargataḥ sannigūḍhatvād aprasiddho bāhyāntara-
caryāprāṇāyāmādikleśaprayāsakalāvirahāt sughaṭas tāvad uktaḥ … “This new, easy
[path]—(easy) because it lacks in the (need for) skill in the external and internal exertions
(usually required) for the (removal of one’s) afflictions, [practices] such as appropriate
conduct [caryā] and breath exercises [prāṇāyāma]—which is included in all the secret
śāstras, (and) is not well known since it has been concealed from public view, was first
explained in the śāstra (entitled) the Śivadṛṣṭi by the venerable Somānanda, our great-
grand guru.”
31 See Ratié 2011, 6 (and especially note 8): “Pourtant, Utpaladeva lui-même présente la
Pratyabhijñā comme une ‘voie nouvelle’ (mārgo navaḥ). Nouvelle, elle l’est avant tout au
sens où elle ne fait pas appel à l’ autorité scripturaire.”
32 About this phenomenon, see Nemec 2011, 12–19.
via free access
298 nemec
I would propose, however, that one must understand a rather different jus-
tification for Utpaladeva’s description of the path he describes as “new.” It is
not—or, rather, it is not merely—the fact that one need not appeal to Śaiva
scriptures to follow it; rather, I suggest that the primary sense of Utpaladeva’s
description of the path as “new” is that it is new to humanity. As is evinced
in Somānanda’s “autobiography,” the Pratyabhijñā teachings were said to have
been concealed by great sages at the beginning of the Kali Age, preserved as
they were by means of the secret lineage described therein, and were only
revealed subsequently for the benefit of those who would come to be initiated
into its ways. Some justification for linking these mythological origins to the
attested novelty of the path may be found, moreover, in the fact that Abhinava-
gupta explicitly invokes the narrative found in Somānanda’s “autobiography” in
a passage of the ĪPVV that comments on ĪPK 4.16, precisely the verse in which
the novelty of the Pratyabhijñā’s path is proclaimed. And in doing so Abhinav-
agupta may be understood himself to associate the lineage with Utpaladeva’s
claim to novelty.33
The Pratyabhijñā thus presents the reader with a self-conscious (theoretical)
understanding of their śāstra as divinely sanctioned, guaranteed by the iden-
tity of the authors in the lineage with Śiva himself; and, while thus divinely
sanctioned, the śāstras are intimately, inextricably, tied to the biographies of
individual, monadic agents—Utpaladeva, Somānanda, and those preceding
and following them in their lineage. Thus, and perhaps not unlike the theology
33 ĪPVV ad ĪPK 4.16 (vol. 3, p. 402, lines 14–20): śivadṛṣṭīti tadgatam ā śrīkaṇṭhanāthāt guru-
parvakramaṃ sūcayati. tatra hi śrīśrīkaṇṭhanāthaḥ śāsane samutsanne śrīdurvāsomuniṃ
tadavatāraṇāya dideśa, so ’pi śrītryambakādityaṃ traiyambakākhye lokaprasiddhyā tairim-
bābhidhāne gurusantāne pravartayitāraṃ mānasaṃ sasarja—ityādi vitatyoktam. śrīkaṇ-
ṭhanāthaś cādhigatatattvaḥ śrīmadanantanāthāt, so ’pi śrībhagavacchaktita ityādy āga-
meṣu nirūpitam iti saṃpūrṇo guruparvakrama uktaḥ.
Compare, in the following graph, the lineages as articulated in the ĪPVV and the ŚD
“Autobiography,” with coincident names in the respective lineages underlined:
ĪPVV: Bhagavacchakti → Anantanātha → Śrīkaṇṭhanātha → Durvāsas →
Tryambakāditya.
ŚD “Autobiography”: Śrīkaṇṭha → Durvāsas → Tryambakāditya → 14 Tryambaka-s →
Saṃgamāditya → Varṣāditya → Aruṇāditya → Ānanda → Somānan-
da.
Clearly, Abhinavagupta wishes both partially to confirm the lineage offered in the “autobi-
ography” and also to add to it at the top the persons Bhagavacchakti and Anantanātha. (Or,
if the composition of the ŚD “autobiography” postdates that of the ĪPVV, one may instead
conjecture that the author of the former wished to add to the lineage of the latter at the
tail end of the same.)
via free access
associated with, e.g., the avatāra theory, we are here offered a view of Siddhas or
“Perfected Ones” who have emerged in human form in the world in their partic-
ular historical contexts, contexts to which they could—and did—actively and
explicitly respond. Finally, this concurrence of historicity and divine author-
ity allows the ĪPK in particular to claim to offer something that is plainly and
explicitly “new” (nava), yet nevertheless “transcendent,” even “timeless,” as it
were, insofar as authorized by Śiva himself. The Pratyabhijñā thus achieves in
its explicit and implicit theoretical formulations the apparently paradoxical
aims of offering something both historically located and new, on the one hand,
and transcendent, on the other, the teachings being ultimately authored and
authorized by Śiva himself and thus held to be uncontained by any historical
bounds.
4 The Kashmiri, Śaiva Example in Practice
This much, then, represents the Kashmiri non-dual Śaivas’ theory, their self-
conscious view of the place of their śāstra and those who carry it in history.
But, what of their practices? What did their theory suggest they should do, and
what, so far as can be known, did they do?
As is well known, the Śaivas of the Pratyabhijñā understood the various prac-
tices related to varṇa or birth to involve mere social custom, their view being
that caste identity was not based in any real difference in the natures of the
individuals so marked by it. We see this view exemplified perhaps nowhere
more clearly than in a passage of Somānanda’s ŚD (at 3.42cd–47), where the
question of the need to purify a fire installed in an outcaste’s, a caṇḍāla’s,34
house is addressed:
svecchāto bhāvarūpatve parādhīnā kutaḥ sthitiḥ 3.42

kṣīravad yadi vocyeta parādhīnaṃ jaḍaṃ bhavet
etayaiva diśā śodhyaṃ śuddhanyūnādidūṣaṇam 3.43
abhagne ’sya svarūpatve śuddhanyūnādikaṃ kutaḥ
patadgrahādike hemni hematvaṃ mukuṭādike 3.44
sthitam eva na hemno ’sya kācid asti vibheditā
caṇḍālasadmago vahnir na vahnir yadi kathyate 3.45
tad evaṃ syād athocyeta vahneḥ saṃskāracodanā
śāstreṣu varṇitā kasmāt kāryārthaṃ kāryam eva tat 3.46
34 A caṇḍāla in this context is the offspring of a Brahmin woman and an outcaste male.
via free access
300 nemec
na svarūpavibhāgo ’tra tathā tatra vyavasthiteḥ

saṃjñākaraṇamātraṃ tad vyavahārāya kalpitam 3.47
Given that he [i.e., Śiva] exists of his own volition in the form of (all)
the entities (that make up the universe), how is existence dependent
on another than himself? If, for example, you say it [i.e., the purported
dependence] is one similar to (the example of curds, whose genesis
depends on the) milk (of which they are comprised), it [i.e., the universe]
would be insentient, dependent on another. The fault (attributed to our
system) that must be corrected—being pure, being diminished, or the
like—is precisely the result of this (wrong) point of view. How can there
be something pure, something diminished, etc., when his nature is undi-
vided? The fact of being gold simply exists in gold, (be it) in (the form of)
a golden spittoon, etc., or in (the form of) a tiara, etc. The (fact of being)
gold is in no way divided [i.e., differentiated] whatsoever. If you argue that
a fire installed in an outcaste’s house is not (properly) called a fire [i.e., it
is not a proper, ritually-purified fire], we reply: that may be so [i.e., this
does not contradict our notion of the uniformity of the nature of fire as
such]. But, if you say (in reply) that injunctions involving the rites of pas-
sage for fire are explained in the (Śaiva) teachings, (we reply:) what is the
goal of the (ritual) action (in question)? It is the action itself.35 There is
no division of its [i.e., the fire’s] nature, here. It is the same for his [i.e.,
Śiva’s] abiding there [in the world]: that [i.e., the distinguishing of “pure”
from “impure” elements in the world, or the distinction of that which is
said to be Śiva and that which is said not to be] is conceived of merely
as the assignation of names for the purpose of everyday speech/everyday
activity [vyavahāra].36
35 In other words, there is no substantial change in the fire qua being fire that results from the
performance of the ritual. It is a social custom, not something that materially transforms
the fire so treated.
36 See also the ŚDVṛ on the same passage: atha vahner yady aśuddhatā na syāt tat tasya
kuto mantraiḥ saṃskārayogaḥ kāryasaṃpādanārthaṃ śāstre codyeta tasmāt tasyāśud-
dhatā saṃbhavet. naivam kāryam evānuṣṭheyam eva tad vyavahārāya na tu svarūpe vah-
ner vibhāgaḥ, svarūpe nijarūpe sthite tu vahnāv upagamyamāne tatsvarūpatā vahnirū-
pataiva sarvatra. tadvac chivarūpatā jagatas tasya, tatheśvaratanmātrapṛthivyādirūpatayā
vyavasthānāt parasthūlasūkṣmatādibhedaḥ. tathā ca sarvatraikye saṃsāravyavahārāya
saṃjñākaraṇamātraṃ tat kalpitaṃ bhavati. “Now, if you argue that, if fire cannot be
impure, then it is inexplicable why the rites of passage, replete with the mantras, are
enjoined to it in the (Śaiva) teaching(s), this for the purpose of accomplishing their (puri-
fying) effect; thus, it must be possible that [fire] can be impure, (we reply:) this is not
so. The rite itself is simply established for the purpose of everyday speech/activity, while
via free access
That caste distinctions should not disqualify anyone from hearing the teach-
ings of the ĪPK has also been stated plainly enough, if somewhat obliquely, in
the maṅgala verse of the ĪPK, where, as noted above, the “new path” (nava
mārga) is explicitly said to be offered to “all people” ( janasyāpi). While the
maṅgala is merely suggestive of any implications regarding caste, however, the
ĪPKVṛ and Abhinavagupta’s ĪPV and ĪPVV further clarify what is intended by
the expression; and what Utpaladeva means by “all people” is explained with
increasing clarity as one delves into the succeeding levels of commentary.
The ĪPKVṛ indicates that one should understand the term to refer to all
people without exception: Utpaladeva glosses janasyāpi with imam akhilam.37
Abhinavagupta, in turn, addresses the caste implications of this pledge—to
help all people—in a more explicit fashion. In the ĪPV subcommentary, he indi-
cates that the term jana refers to any person regardless of the circumstances of
his birth ( janasyeti yaḥ kaścij jāyamānaḥ tasya); and he further specifies that
no distinction by the “office” held by a candidate for initiation (adhikāriviṣaya)
may serve to restrict access to the teachings of the Pratyabhijñā.38 He later adds
that the term in question should be understood to refer to anyone afflicted by
the suffering of incessant births and deaths.39
Finally, Abhinava also is explicit, in both the ĪPV and ĪPVV, in stating that
jāti should not be considered an appropriate measure of a candidate’s eligibil-
ity for entry into the soteriological path of the Pratyabhijñā. This he does in the
passages of his two sub-commentaries that address the concluding verse of the
Kārikās, where Utpaladeva has again indicated that the text is meant for “[all]
people.” Referring to himself in the third person, he says (at ĪPK 4.18):40
there is no change in the nature of the fire. Rather, it being established that fire is fixed in
its nature, i.e., in its own form, the fact of having that nature, i.e., the very fact of being fire,
exists in every instance (of fire). Similarly, the universe has Śiva as its nature. Thus, divi-
sion into the fact of being supreme, coarse, and subtle, etc., exists as a result of (the same
Śiva) being established in the form of the Lord, the subtle elements, the earth, etc. And
thus, there being oneness everywhere, that [i.e., such categorization or division of Śiva-
nature] is conceived of as merely the assignation of names for the purpose of everyday
speech/everyday activity in saṃsāra.”
37 See ĪPKVṛ ad ĪPK 1.1.1. parameśvaraprasādād eva labdhātyantadurlabhataddāsyalakṣmīr
aham ekākisaṃpadā lajjamāno janam apīmam akhilaṃ svasvāminaṃ vakṣyamāṇopāyena
pratyabhijñāpayāmi yena tasyāpi paramārthalābhena parituṣyeyam.
38 See ĪPV ad ĪPK 1.1.1. (vol. 1, p. 14, lines 10–12): janasyeti yaḥ kaścij jāyamānaḥ tasya, ity ane-
nādhikāriviṣayo nātra kaścin niyama iti darśayati …
39 See ĪPV ad ĪPK 1.1.1. (vol. 1, p. 15, line 9) where Abhinava glosses jana with anavaratajanana-
maraṇapīḍita.
40 See ĪPK 4.18: janasyāyatnasiddhyartham udayākarasūnunā | īśvarapratyabhijñeyam utpa-
lenopapāditā.
via free access
302 nemec
In order that [all] people ( jana) can attain perfection effortlessly, Utpala,
the son of Udayākara, has presented this Īśvarapratyabhijñā [treatise].
Abhinava’s ĪPV suggests that with “people” ( jana) Utpaladeva refers to any
person, for neither jāti nor any other criterion is innately disqualifying.41 He
reiterates the same in the ĪPVV as well, where, additionally, he excludes the
individual’s “conduct” (ācāra) from consideration in determining whom the
teachings may help.42 Note also that the ĪPVV explicitly links the present pas-
sage with the one found at the beginning of the ĪPK, saying: etac ca prārambha
evāsmābhir vitatam.43 Simply, distinctions marked by birth—including caste
distinctions—are to be ignored.
The professed catholic reach of the work is significant, not only due to its
inclusivism, a quality well attested in the context of tantric rules of initiation,
even if this cannot be taken to imply the existence in practice “on the ground”
of a concomitant equality of participation, or status, within the context of the
initiatory “family” (kula); but it is further significant that it is declared, albeit
obliquely, in the ĪPK (and ĪPKVṛ), a text (and commentary) that has all the
hallmarks of works intended for audiences of non-initiates.44 I submit that
there must have been some appreciation for a certain novelty in the appar-
ent inclusivism professed in the work. This, then, may constitute another rea-
son for Utpaladeva’s description of the Pratyabhijñā path as “new”: it offers a
certain novelty in being (more or less explicitly) indifferent to caste distinc-
tions.
While the use of the term jana in the ĪPK offers a certain ambiguity of
expression, Abhinava’s subcommentaries are explicit in stating the inclusivism
of the śāstra. One can thus see operative here a certain stratification of expres-
sion, a certain obliqueness with regard to any reference to caste in the ĪPK and
ĪPKVṛ, but a more explicit asseveration of the sociological parameters of the
tradition in Abhinavagupta’s sub-commentaries. So much of course reflects the
41 See ĪPV ad ĪPK 4.18 (vol. 2, p. 276, lines 5–6): yasya kasyacij janor iti nātra jātyādyapekṣā
kācid iti sarvopakāritvam uktam. See also the avataraṇikā to ĪPK 4.18 (ĪPV, vol. 2., p. 275,
line 14), where Abhinava describes the śāstra in question as one that can help “everyone”:
sarvopakārakaṃ mahāphalam idaṃ śāstraṃ …
42 See ĪPVV ad ĪPK 4.18 (Vol. 3., p. 404, lines 6–8): yaḥ kaścij jananadharmā, tasyātaḥ siddhiḥ,
na tv atra jātyācārādau bhara iti sarvānugrāhakatvam uktam. etac ca prārambha evāsmā-
bhir vitatam.
43 See footnote 42.
44 On the intended audiences of the ĪPK and ĪPKVṛ, on the one hand, and the ŚD (and ŚDVṛ),
on the other, see Nemec 2011, 12–19.
via free access
complex of practice of the tradition, exemplified as it is by the dictum quoted

above, “internally a Kaula, externally a Śaiva, and in social practice a follower
of the Veda.”
This dictum and the complex practice it invokes—one of engaging differ-
ent rules and norms in differing contexts—is explicitly tied to the question of
social practices, those associated with caste and life-stages, as the context pro-
vided by Jayaratha for this quotation clearly indicates. For he considers the sig-
nificance of preserving the social order—varṇāśramācāra—precisely where
he quotes the complex form of practice that is here of concern.45 Moreover,
the entire matter is raised in the context of coming to understand precisely
whether the Śaiva practitioner should respect the Veda-based rules of purity
and impurity. For the relevant section of the fourth āhnika of the TĀ is governed
45 See TĀ (and TĀV ad) 4.251ab (vol. 3, p. 277, line 9 to p. 278, line 6):
evaṃ sarvotkṛṣṭatvāc chaiva eva śāstre mukhyayā vṛttyā niṣṭhā kāryā, nānyatrety āha
tasmān mukhyatayā skanda lokadharmān na cācaret | 4.251ab
niṣṭhāśūnyatayā tu gauṇyā vṛttyā lokasaṃrakṣaṇārthaṃ lokadharmān ācarato na kaś-
cid doṣa iti bhāvaḥ. tad uktaṃ tatra
ye tu varṇāśramācārāḥ prāyaścittāś ca laukikāḥ |
saṃbandhān deśadharmāṃś ca prasiddhān na vicārayet.
garbhādhānāditaḥ kṛtvā yāvad udvāham eva ca |
tāvat tu vaidikaṃ karma paścāc chaive hy ananyabhāk.
na mukhyavṛttyā vai skanda lokadharmān samācaret.
iti. ata eva
antaḥ kaulo bahiḥ śaivo lokācāre tu vaidikaḥ |
sāram ādāya tiṣṭheta nārikelaphalaṃ yathā.
“Thus, [Abhinavagupta] says that, since it is superior to all (other śāstras), what is to
be done is fixed in its primary sense in the Śaiva śāstra alone and not elsewhere:
‘Therefore, O Skanda, one need not observe the rules of the (everyday) world as
though they were of primary importance.’
However, the sense (of this passage) in its secondary meaning, absent of what is
fixed (to be done), is that no fault accrues to the one who performs lokadharma for
the purpose of protecting the people/world [loka]. The following is said on the mat-
ter:
‘Moreover, one should not cause the well-known, proper customs of the country
(in which one lives) to go awry, these being ( ye) the societal [laukikāḥ] acts of conduct
associated with varṇa and āśrama and the expiatory acts. From the time beginning
with the ceremony before conception and until one has performed the wedding cere-
mony, Vedic acts (are to be enjoined). Following this, one is to be devoted to the Śaiva
[path] and no other. Indeed, O Skanda, one need not perform the rules of the (every-
day) world as though they were of primary importance.’
For this very reason:
‘Internally a Kaula, externally a Śaiva, and in social practice a follower of the Veda,
having extracted the essence, one should remain, as does the fruit of the coconut.’”
via free access
304 nemec
precisely by the question of how the jīvanmukta should behave in the world,
and Abhinavagupta explicitly cites the eighteenth chapter of the Mālinīvijay-
ottaratantra there as an authoritative voice in the matter.46 Note, in addition,
46 See TĀ 4.213–221ab (vol. 3, pp. 241 ff., which quote the Mālinīvijayottaratantra, verses 18.74–
81):
atra nāthaḥ samācāraṃ paṭale ’ṣṭādaśe ’bhyadhāt |
nātra śuddhir na cāśuddhir na bhakṣyādivicāraṇam 4.213
na dvaitaṃ nāpi cādvaitaṃ liṅgapūjādikaṃ na ca |
na cāpi tatparityāgo niṣparigrahatāpi vā 4.214
saparigrahatā vāpi jaṭābhasmādisaṃgrahaḥ |
tattyāgo na vratādīnāṃ caraṇācaraṇaṃ ca yat 4.215
kṣetrādisaṃpraveśaś ca samayādiprapālanam |
parasvarūpaliṅgādi nāmagotrādikaṃ ca yat 4.216
nāsmin vidhīyate kiṃcin na cāpi pratiṣidhyate |
vihitaṃ sarvam evātra pratiṣiddham athāpi ca 4.217
kiṃ tv etad atra deveśi niyamena vidhīyate |
tattve cetaḥ sthirīkāryaṃ suprasannena yoginā 4.218
tac ca yasya yathaiva syāt sa tathaiva samācaret |
tattve niścalacittas tu bhuñjāno viṣayān api 4.219
na saṃspṛśyeta doṣaiḥ sa padmapatram ivāmbhasā |
viṣāpahārimantrādisaṃnaddho bhakṣayann api 4.220
viṣaṃ na muhyate tena tadvad yogī mahāmatiḥ. 4.221
“Here, the Lord has explained the appropriate conduct (of the accomplished yogin) in
the 18th chapter (of the Mālinīvijayottaratantra): ‘There is no purity here, nor impu-
rity, no consideration of what is to be eaten, etc. There is no duality, nor non-duality,
and no (requirement to perform) acts of devotion to the liṅga, etc. There is similarly
no (requisite) abandoning of those [acts], nor the (required) renunciation of mate-
rial possessions, nor again any (requirement regarding the) accumulation of material
possessions. There is no (requisite) maintenance of twisted locks of hair [ jaṭā], of
(smearing oneself with) sacred ashes, or the like, nor any (requisite) abandoning of the
same. And as regards the performance or non-performance of vows, etc., and entrance
into sacred places, etc. [i.e., kṣetras, pīṭhas, and upapīṭhas; see TĀ 4.259cd and the TĀV
thereupon], the observance of rules of action, and (those rules associated with) initia-
tory name, initiatory lineage [gotra], or the like [i.e., according to the lodge (ghara)
and the like of the initiate; see TĀ 4.267cd], whether the form, sectarian marks, and so
on be one’s own or another’s—nothing is prescribed here regarding these, nor, con-
trariwise, prohibited. Absolutely everything is performed here [according to the rules
of the Mālinīvijayottara], and, contrariwise, omitted. Yet, this (alone) is necessarily
enjoined here [in the Mālinīvijayottara], O Goddess, that the wholly pleased yogin
must fix his consciousness [cetas] on reality; and he should therefore act only in accor-
dance with that [reality (tattva)], whatever that may be for him. Moreover, the one
whose consciousness [citta] is fixed on reality, partaking even in the pleasures of the
senses [viṣaya], is not touched by bad consequences, just as the petal of a lotus (is not
affected) by water. The yogin who has great understanding [mahāmati] is the one who
is similar to the person who, armed with mantras that counteract poison and the like,
is not deluded by the poison even while devouring it.’”
I thank Shaman Hatley for answering several queries I had about this passage.
via free access
that it is the matter of śuddhi and aśuddhi, as in was in ŚD 3.42cd–47, which is

here (at TĀ 4.213cd) explicitly placed in question. The TĀ and TĀV go on to sug-
gest that yogis know for themselves what is pure, what impure; and they further
state that it is their very awareness that should determine how they act, for the
normative rules of purity do not apply to them.47
Following this, the key objection is raised: are there not (Vedic) injunctions
that require one to follow the rules of purity?48 The opponent asks, that is,
whether tantric Brahmins must respect Vedic authority as well as Śaiva truths.
How is it that, given the apparent validity or “truth” of both sets of śāstric
injunction, the Vedic and the Śaiva, one can be invalidated, the other affirmed?
How can one be internally a Kaula, externally a Śaiva, and a Vaidika in social
practices when to be one sometimes requires one to contravene the strictures
of the other?
The answer given not only suggests that it is not always inappropriate in
practice to contravene the rules of purity and impurity, of śuddhi and aśud-
dhi; it further offers a rationale—a theory—for this complex of practice. This
it does by invoking a general, and common, śāstric sensibility, namely, that the
śāstras operate by first issuing blanket rules, which may later be superseded
in certain specified contexts. This is a fundamental organizational structure of,
among other śāstric works,49 the system of Pāṇini’s Aṣṭādhyāyī—which (as is
well known) sets out exhaustively to account for correct speech not by listing
every possible form of speech, but by structuring classes of rules that first define
the utsarga or generally applicable rule, only then to itemize the apavādas or
47 See TĀ (and TĀV ad) 4.228ab (vol. 3, p. 249, lines 4–12):

yoginaṃ prati sā cāsti bhāveṣv iti viśuddhatā | 4.288ab
paśuprāyāṇāṃ hi mantreṣv api śivātmatvena parijñānaṃ nāsti iti tān prati teṣāṃ svakā-
ryakāritvābhāvāt saṃbhāvanīyam api aśuddhatvam. dharādīnāṃ ca yoginaṃ prati
tatparijñānam asti iti teṣām api viśuddhatvam. etad eva hi nāma yogino yogitvam,
yat nikhilam idaṃ viśvaṃ śivātmatayā parijānāti iti.
“ ‘As regards the yogin, moreover, that purity exists with respect to (all) entities.’
Indeed, the majority of bound souls do not perceive even (Śaiva) mantras as hav-
ing Śiva-nature, and therefore they suppose them to be impure, since they fail (even)
to perform their own duties (in employing these mantras). But, as for the yogin, he
perceives (everything,) beginning with the earth(-tattva), as having that [Śiva-nature].
Therefore, (he perceives that) all of them without exception are completely pure.
Indeed, this alone is the very nature of the yogin as a yogin, that he perceives this entire
universe as possessed of Śiva-nature. This is definitive.”
48 See TĀ 4.228cd (vol. 3, p. 250, lines 4–5): nanu codanayā śuddhyaśuddhyādikaviniścayaḥ.
“Objection: a fixed knowledge of purity, impurity, and so on exists by way of (Vedic) injunc-
tion(s) [codanayā].”
49 It is of course a model found in the Mīmāṃsā, as well, for example.
via free access
306 nemec
exceptional rules that can override the former in the appropriate contexts. So
much is explained in Patañjali’s Mahābhāṣya as follows:50
Now, one may object by asking whether it isn’t the case that, when this
(system of) instruction of (correct) speech is operative, one must recite
each word in the course of acquiring the (correct forms of) words. (That
is:) Every word must be recited, as in ‘cow, horse, man, elephant, bird, deer,
Brahmin.’ Reply: Certainly not! It is not expedient to recite each word in
the course of acquiring the (correct forms of) words. For, so much has
been communicated as follows: ‘Bṛhaspati imparted to Indra the study
of words by uttering each one individually over the course of a thou-
sand divine years, and he did not reach the end of it.’ Bṛhaspati was the
teacher [pravaktṛ], and Indra was the student [adhyetṛ]; and the time
of instruction did not reach its terminus over the course of a thousand
divine years. What of (the use of this method) today? One who plainly
lives a long time (today) lives for (only) a hundred years … How, then,
should these (correct) words be acquired? A certain characteristic asso-
ciated with the generic [sāmānya] and the particular [viśeṣa] should be
activated, by means of which [people] can—with little effort—become
acquainted with the extremely massive flood of words. What, pray tell,
is that? Utsarga and apavāda. A general rule [utsarga] is (first) to be
employed; (then,) an exception [apavāda](, which overturns the general
rule). But under what circumstances is the general rule to be used, under
what circumstances the exception? The general rule is to be used generi-
cally. E.g.: karmaṇy aṇ (A 3.2.1). The exception to it (should be used) in a
particular circumstance. E.g.: āto ’nupasarge kaḥ (A 3.2.3).
50 See Paspaśāhnika (7) of Patañjali’s Mahābhāṣya: utsarga and apavāda, vol. 1, p. 5, line 24
and following of the Kielhorn edition: athaitasmiñ śabdopadeśe sati kiṃ śabdānām prati-
pattau pratipadapāṭhaḥ kartavyaḥ. gaur aśvaḥ puruṣo hastī śakunir mṛgo brāhmaṇa ity
evamādayaḥ śabdāḥ paṭhitavyāḥ. na iti āha. anabhyupāyaḥ eṣaḥ śabdānām pratipattau
pratipadapāṭhaḥ. evam hi śrūyate. bṛhaspatir indrāya divyaṃ varṣasahasraṃ pratipado-
ktānāṃ śabdānāṃ śabdapārāyaṇaṃ provāca na antaṃ jagāma. bṛhaspatiś ca pravakten-
draś ca adhyetā. divyam varṣasahasram adhyayanakālaḥ na ca antaṃ jagāma. kim punar
adyatve. yaḥ sarvathā ciraṃ jīvati sa varṣaśataṃ jīvati … kathaṃ tarhīme śabdāḥ pratipa-
ttavyāḥ. kiñcit sāmānyaviśeṣaval lakṣaṇaṃ pravartyaṃ yenālpena yatnena mahato maha-
taḥ śabdaughān pratipadyeran. kiṃ punas tat. utsargāpavādau. kaścid utsargaḥ kartavyaḥ
kaścid apavādaḥ. kathañjātīyakaḥ punar utsargaḥ kartavyaḥ kathañjātīyako ’pavādaḥ.
sāmānyenotsargaḥ kartavyaḥ. tad yathā. karmaṇy aṇ. tasya viśeṣeṇāpavādaḥ. tad yathā.
āto ’nupasarge kaḥ.
via free access
Here, we see that it is for efficiency’s sake that this model is adopted. And
we are given a sample application in Pāṇini’s grammar, the details of which
are explained in Appendix A. Essential to recognize is the fact that the pre-
scription delineated by the general rule, the utsarga, always stands; yet, on
some occasions its application is suspended by a special prescription, in the
defined context of the apavāda or exceptional rule.51 Whenever no apavāda
applies, however, the utsarga automatically prevails, for whenever or wherever
the special rule does not apply (or must itself be suspended), one reverts by
default—invariably and automatically—to the general, standing rule.
It is this very śāstric mode that is explicitly invoked by Jayaratha in reference
to the very phenomenon we have here, that of a complex of religious practice,
Vedic and Śaiva.52
TĀ 4.230ab: ‘If you argue that this [Śaiva rule] is invalidated by the Vedic
one, (we reply:) why shouldn’t it be the other way around?’53
…
Considering that one may object by asking how it is that both [śāstras] are
equally real, given that in certain contexts the injunction(s) [vidhi] asso-
ciated with purity and the like are invalidated, even though they apply
universally, for all people, [Abhinavagupta] says:
51 Note, therefore, that the domain of application of the general rule is said to be ubiquitous
(sarvatra) at both Kāśikāvṛtti ad A 3.2.1 (which reads, in part: sarvatra karmaṇy upapade
dhātor aṇpratyayo bhavati) and in Jayaratha’s explanation of the matter, at TĀV ad TĀ
4.230cd–231ab, for which see below.
52 See TĀ 4.230–232ab (vol. 3, p. 253, line 1 to p. 255, line 16), along with the corresponding pas-
sages of the TĀV (though one should note that, for the sake of brevity, none of the commen-
tary is here cited of what follows TĀ 4.230ab, excepting the avataraṇikā that introduces
4.230cd–231ab; and only part of the commentary that appears prior to TĀ 4.231cd–232ab
is quoted here below):
vaidikyā bādhiteyaṃ ced viparītaṃ na kiṃ bhavet |
… nanu katham anayor aviśiṣṭaṃ sattvaṃ śuddhyādividheḥ sarvapuruṣaviṣayatayā pra-
vṛttāv api kvacid viṣaye bādhāt ity āśaṅkyāha
samyak cen manyase bādho viśiṣṭaviṣayatvataḥ. 4.230
apavādena kartavyaḥ sāmānyavihite vidhau |
yadi nāma bādhāvṛttaṃ samyag avabuddhyase, tan na kasyā api codanāyāḥ sattva-
hāniḥ. tathā hi—niravakāśatvād viśeṣātmā apavādavidhiḥ sarvatra labdhāvakāśaṃ
sāmānyātmakam utsargavidhiṃ bādhate, iti vākyavidaḥ …
śuddhyaśuddhī ca sāmānyavihite tattvabodhini. 4.231
puṃsi te bādhita eva tathā cātreti varṇitam |
53 This is of course to say, in a rhetorical manner, that it is rather that precisely the opposite
must be the case.
via free access
308 nemec
TĀ 4.230cd–231ab: ‘If you think about it clearly, a rule that is an excep-

tion [apavāda] invalidates a rule generally applied [sāmānyavihita
vidhi], because it applies in a particular domain.’
So, if you properly consider the procedure of invalidation, then (you will
realize that) no injunction [codanā] whatever loses reality [sattvahāni].
To explain: the rule that is the exception [apavādavidhi]—by nature spe-
cific because it is (generally) void of any occasion (for application)—
supersedes the general rule [utsargavidhi], which, being one that always
has met with its occasion (for application), is by nature generally appli-
cable [sāmānyātmaka]. This is what those who know language say …
TĀ 4.231cd–232ab: ‘Moreover, purity and impurity, which are generally
enjoined, are simply superseded when a man knows reality. This is how
it has been explained here [in the Mālinīvijayottara].’
In the face of the objections of a more orthodox opponent who demands to

know how the less permissive rules of the Veda cannot prevail over Śaiva scrip-
tures, Abhinavagupta brilliantly inverts the hierarchy by appealing to a main-
stream sāstric interpretive rule: it is the exception that preempts the normative,
overarching rule and not the other way around. The rule that applies most gen-
erally, the Vedic one, is the utsarga. It is invariably prescribed—Jayaratha here
described the utsarga as sarvatra labdhāvakāśa and sāmānyātmaka—but it
is occasionally superseded by the exceptional rule or apavāda (described by
Jayaratha as niravakāśatvād viśeṣātmā), where in this instance the apavāda is
the Śaiva rule, occasioned as it is by the special knowledge of the accomplished
yogin who sees reality for what it is, as transcending the conventions of social
position as determined by birth.
Nevertheless, whenever this special rule does not prevail—viz., outside the
confines of tantric knowledge and practice—one must revert by default to con-
forming to the general rule, the utsarga (i.e., the Veda-based strictures), which
is never nullified but only occasionally and temporarily superseded. So much is
the position Abhinavagupta articulates in the TĀ, and Jayaratha indicates that
in doing so he responds to the potential criticism that tantric rules authorize
practices that are proscribed by Vedic norms, implying thereby that the tantric
practices should be abandoned. This complex of practice, offered by Śaivas who
engaged in rites that transgressed social and other contemporaneous cultural
norms, was thus said to be the necessary product of the mainstream theoretical
model that they understood to prescribe it.
The implications of this exemplar, then, are evident. Divinely sanctioned
(in no small part by the authors’ identification with Śiva), the Pratyabhijñā
offers a tradition that prescribes exceptional practices that were introduced
via free access
in historical time but are nevertheless rooted in an “ahistorical” authority. The

knowledge it offers is fixed in its dimensions: there is no theology of “progress”
in this śāstra. Each śāstra, moreover, Vedic and Tantric, speaks in a voice that in
some important sense offers not merely a “knowing how,” but rather a “knowing
that”; for each individually denies the possibility of a dialectical relationship
between theory (śāstra) and practice (prayoga) in Pollock’s sense of the prob-
lematic. (New political developments could not alter its vision of the nature of
Śiva as a ubiquitous consciousness, for example.) To this extent, then, Pollock’s
model is applicable to these śāstras.
Yet, the tantric śāstra here placed under study nevertheless allows for a the-
ology of novelty, because its “ahistorical” authority was said to have been intro-
duced by particular agents in particular historical moments. The system thus
allowed for, indeed anticipated, the emergence of new exceptional rules, and
because the canon remained open, it was ever possible that another text, a new
apavāda as it were, could emerge in a context in which the general parameters
of social action—of religious practice or prayoga—otherwise and inevitably
would remain operative.
A dialectic, then, between theory and practice in premodern South Asia,
while it might not be found in the self-representation of any single given work,
can be found in the proliferation of multiple śāstras, which in supplementing as
well as competing with one another could accommodate innovation, explicitly
or implicitly, by introducing novel theoretical models as circumstances war-
ranted. The dialectic may only be found in the scriptural gestalt, however; for,
it is only by reading multiple scriptural sources with an awareness of their
mutually shared contexts, pedigrees, and, most importantly, structured mutual
relations that one might begin to identify patterns of religious (and other forms
of) change in Indian premodernity.54
54 In response to two important observations of Shaman Hatley, who read an earlier draft
of this essay, two caveats are here required. First, I wish herein to indicate not that the
Pratyabhijñā authors innovated the practice of deriving authority from their siddha-
hood. This is a practice that indeed predates these authors. Rather, I wish to suggest that
such a practice, and others like it such as the avatāra theory, allowed for innovations
in tradition—they allowed authors self-consciously and explicitly to introduce unprece-
dented practices and ideas, what would be new in human history, as those that were
religiously—transcendently—sanctioned. Thus, in an example offered from the avatāra
theory, Kṛṣṇa deploys novel—and context-appropriate—ways of upholding dharma, even
if he is himself ultimately considered to be an emanation of the timeless and transcendent
Viṣṇu himself.
Second, it can be observed (as Hatley did in a personal communication, via email, of
November 11, 2015) that the social model that Abhinavagupta has developed is in fact also
quite conservative, for while allowing for an innovative social institution that ignores caste
via free access
310 nemec
5 Conclusion
I note, in closing, a concentric series of five implications of this phenomenon,

each of an increasing degree of generalization, and in doing so I intend to
point to possible avenues of further inquiry that lie beyond the scope of the
present (and admittedly particular) case-study. The first involves a relatively
narrow methodological concern in the study of Śaiva post-scriptural sources:
the present exemplar stands as a reminder that knowledge of the wider con-
text of “orthodox” śāstric texts and concerns should inform scholarly inter-
pretations of the post-scriptural literature; for even works that are primarily
rooted in and explicitly concerned with esoteric scriptures, as is the TĀV, fre-
distinctions it also synchronizes such practices with the caste-based Brahminical stric-
tures that were not, historically speaking, espoused in the Kula-Kaula branch of Śaivism;
and it limits the transgressive practices by defining them as exceptional, as the (rare)
contravention of the overarching, normative structures of Brahminism. (Similarly, Alexis
Sanderson, in response to the shortened version of this essay that I presented at the confer-
ence held at the University of Toronto in his honor, suggested that what is offered here is a
“vertical” model of authority and innovation, as opposed to the more radical, “horizontal”
models found elsewhere in the Śaiva traditions, which more fully unshackled themselves
from Brahminical norms.) This cannot be denied. There is, in the complex of theory and
practice offered by the Pratyabhijñā, a certain preservation of Veda-based tradition. And it
is true that there were in premodern South Asia models of change that sought not to mod-
ify (as Abhinavagupta does) but to upend Brahminical norms. Yet, a distinction here must
be made between the degree to which these traditions in fact changed social norms and
the fact that they explicitly and self-consciously wedded themselves to the idea that reli-
gion could offer ideas and practices that were explicitly said to be new. To put the matter
differently, one might reasonably argue that Abhinava was somewhat inimical to (social)
change—not entirely so, as what he endorses clearly also allows for the contravention of
Brahminical norms, in part at least; but he also explicitly does not espouse a view that
change is inimical to religious ideas and practices. If it is indeed the case that he—and
Utpaladeva—sought to preserve a certain Brahminical order (in part at least), reigning in
or domesticating, as it were, the more radical practices of Śaiva and Śākta traditions, it is
all the more noteworthy that they did so by explicitly endorsing the possibility of religious
change—of changes in ideas and social and other practices—in doing so; for this strategy
speaks to the very fact that change, in premodern South Asia, was not inimical to religion,
even if, as one might argue, the fact that they said as much was done in service of limiting
social change to modifications of Brahminical norms, rather than allowing for the whole-
sale rejection of them. My own view, as is clarified in the conclusion, is that the model here
presented—of change through modification of the normative order—both embraced the
idea of change and ushered in changes “on the ground” in religious and social practice;
moreover, I maintain that it offers what was the prevalent model of change in premod-
ern South Asia—that of change through modification as opposed to wholesale change or
revolution—and scholars should look for such types of change in places where previous
scholarship has only seen a static social and intellectual order.
via free access
quently record a substantial interaction with the mainstream systems of learn-

ing (including both various śāstric systems such as Pāṇinian grammar and the
classical philosophical darśanas).
Second, there is the matter of the social context, the Kashmir Valley of the
period in question, in which the authors of the Pratyabhijñā lived and wrote.
While it is worth noting that the tantric initiatory “family” (kula) would have
frequently offered an uncomfortable home for low-caste and outcaste initiates
(as many or few as there might have been in practice), entry into the tantric
kula could also have afforded the initiate a certain place of privilege. Initiation
was, from the perspective of the tradition, a significant institution, one that
the tantric practitioner would have coveted and shared sparingly; and member-
ship would have conferred not only a certain prestige but also a special form of
knowledge that those within the tradition would have held to be paramount: a
certain access to the divine.
Thus, while this form of privilege does relatively little globally to address the
social inequities enshrined in normative, caste-based strictures, the initiation
of low-caste or outcaste individuals into the secret śāstras nevertheless consti-
tutes a social innovation over and against preexisting Vedic (or, perhaps more
accurately, dharmaśāstric) norms. The change it offers, however, comes not in
the form of social revolution but in the form of a particular and circumscribed
social inclusion. Generalizing from the present exemplar, one suspects that
social change in premodern South Asia will most often be found in its incre-
mentalism, as opposed to efforts in total revolution such as the toppling of the
institution of royal sovereignty or the wholesale displacement of Brahminical
authority. Social change, simply, will likely be found, where it is to be found, in
the often-subtle exceptions that contradict overarching and often oppressive
rules that were more globally applied, in Kashmir or elsewhere, in premodern
South Asia.
Third, the model of social conduct witnessed in the Pratyabhijñā exemplar
finds correspondences in the narrative tropes of (non-tantric) Śaiva mythology.
Śiva is not only paradoxically erotic and ascetical in the mythological literature,
as is well known, but he also appears there as both asocial or anti-social and
integrated into society (and often simultaneously so), as is also well known.
Take, for example, the insult of Dakṣa’s sacrifice, where Śiva is depicted as
a pariah while simultaneously standing as the ideal husband in the eyes of
the eminently acceptable Satī. One may therefore query whether the sort of
complex of social practice examined in the present essay constitutes a phe-
nomenon more widely attested across various Śaiva traditions.55
55 I thank Shaman Hatley for suggesting that, among the various candidates that present
via free access
312 nemec
Fourth, the exceptionalism offered by the utsarga-apavāda model is com-

monly attested across a range of premodern Indian cultural systems, most
notably, for the purposes of the present essay, those that address social rules
and life. The Kṛṣṇa of the Mahābhārata plainly contravenes a series of gener-
ally applicable social strictures, for example, rules that he otherwise would wish
vigorously to maintain; and the rules of conduct for crisis-situations (āpad-
dharma) more generally are premised on the idea that the prescribed actions
are exceptional both because they contravene otherwise normative behaviors
and because they are deemed inactionable outside the only context in which
they may apply: that of a social emergency. And one cannot overlook, in this
context, the frequency with which the Veda is acknowledged as authoritative in
Hindu traditions while the contents of Vedic works are nearly entirely ignored.
It is common, in sum, to find a layering of practice in South Asian religion
and social life; and such complexes of practice, allowing as they do for the
exception, in its place, to supersede normative strictures, should probably be
counted as the normative model of social organization in premodern South
Asia.56
Fifth and finally, I address an issue relevant to the study of religion, broadly
conceived. It is instructive that Pollock appealed to Geertz in formulating
the theoretic that defines his understanding of the relationship of śāstra to
prayoga. Indeed, he finds Geertz’s famed distinction of “models of” from “mod-
els for” precisely in the article in which he famously defines religion as “(1) a
system of symbols which acts to (2) establish powerful, pervasive, and long-
lasting moods and motivations in men by (3) formulating conceptions of a
general order of existence and (4) clothing these conceptions with such an
aura of factuality that (5) the moods and motivations seem uniquely realis-
tic.”57
Geertz apparently did not anticipate the possibility of a complex of (some-
times mutually-contradictory) beliefs or a complex of (sometimes mutually-
contradictory) practices. His model instead elaborates a more singular structure
or phenomenon, referring as it does to “a system of symbols” that establishes
powerful moods and motivations in men by formulating “a general order of
themselves for examination along these lines, the anti-caste Vīraśaivas immediately come
to mind.
56 Indeed, the sort of “And” approach that Pollock identifies and endorses (over and against
an “Either/Or” approach) elsewhere in his writings owes something, I think, to this intel-
lectual practice of allowing for exceptions in the context of an overarching normative
practice that otherwise prevails. See Pollock 2006, 574–580. Cf. footnote 59.
57 See Geertz 1973, 89.
via free access
existence,” which is clothed in “an aura of factuality,” all of which implies a

model of a singular source of religious authority. There is in this, as in other sim-
ilarly situated definitions,58 an understanding of religion as involving an appeal
to an ultimate that does not seem naturally to allow for variegated and equally
legitimate sources of authority, that fails to account for the type of “And” model
found in South Asia and, one imagines, elsewhere, as well;59 and in defining
religion in this manner, we see foreclosed any possibility that a religious tradi-
tion might innovate or simultaneously access disparate sources of authority in
order to recommend diverse—or even new—doctrines and/or practices, as we
see in our premodern Indian exemplar.
The example here studied, then, begs for novel theoretical models of reli-
gion, ones that could attend more fundamentally or at least more explicitly
to the capacity of religion to accommodate, indeed self-consciously to cul-
tivate, change. To proffer as much would refocus our possible ways of read-
ing premodern religious traditions, allowing us to identify patterns of social
and religious change that are not defined by modern understandings of the
same (as exemplified in Pollock’s dependence on the concept of “progress” to
measure any conceptual openness to social change in India’s premodernity).
It would also require one to cultivate a view of religion that emphasizes not
its interest in certainty, in its capacity to memorialize particular ways of act-
ing or models of knowing—be they models of “knowing that” or “knowing
how”—but its interest in possibility. Cultivating such a new model (and thereby
a new interpretation, even a new scholarly definition) of religion could, that
is, nurture a view of religion that emphasizes its more active and dynamic,
and malleable and practical, dimensions, the capacity of religious practitioners
and authors themselves self-consciously to formulate systems of understand-
ing and to stake claims where any number of claims could have possibly been
made.
58 For example, Durkheim’s definition of religion reads as follows: “A religion is a unified sys-
tem of beliefs and practices relative to sacred things, i.e., things set apart and forbidden—
beliefs and practices which unite in one single moral community called a Church, all those
who adhere to them.” So too could it be said of a number of others, such as Melford Spiro’s
famed and widely accepted definition of religion as “an institution consisting of culturally
patterned interaction with culturally postulated superhuman beings.” These and other,
similarly-oriented definitions all, if they do not demand it, implicitly lean on the notion
that what can be true or significant is something similar to a singular ideal, rather than
allowing for the sorts of nuanced and contextually situated notions of what counts that
we can find in our Indian exemplar.
59 I here follow Pollock in distinguishing between the “And” model offered in South Asia and
the “Either/Or” model articulated elsewhere. See Pollock 2006, 574–580; cf. footnote 56.
via free access
314 nemec
This would, in short, require one to understand religion not primarily as the
maps of systems and symbols and practices that are charted by religious agents,
but instead in terms of the strategies they use to sketch their maps. Those who
have crafted religious ideas and ideals, I think, were—and are—well aware of
the fact that religions offer changing models of action and modes of reflection
in the context of a changing world that is perennially met with novel problems,
begging resolution. Those of us who read and theorize their practices should, in
turn, seek to shape our reflections on religion in a manner that allows more nat-
urally for the innate flexibility, the freedom of possibility—of change—in idea
and practice, with which religious agents developed and develop their models
for action.
Abbreviations
A Aṣṭādhyāyī of Pāṇini
ĪPK Īśvarapratyabhijñākārikā of Utpaladeva
ĪPKVṛ Īśvarapratyabhijñākārikāvṛtti of Utpaladeva
ĪPV Īśvarapratyabhijñāvimarśinī of Abhinavagupta
ĪPVV Īśvarapratyabhijñāvivṛtivimarśinī of Abhinavagupta
KSTS Kashmir Series of Texts and Studies
ŚD Śivadṛṣṭi of Somānanda
ŚDVṛ Śivadṛṣṭivṛtti of Utpaladeva
TĀ Tantrāloka of Abhinavagupta
TĀV Tantrālokaviveka of Jayaratha
Acknowledgements
I would like to thank Srilata Raman and Shaman Hatley for inviting me to the
conference in honor of Alexis Sanderson, titled “Śaivism and the Tantric Tra-
ditions,” at which I presented a shortened rendering of this essay. I also thank
Eric Steinschneider for all he did to facilitate the running of this well-organized
event. Prior to this, I delivered an earlier version of this essay in the Hindu Stud-
ies Colloquium at Harvard Divinity School on February 20, 2014. I thank Anne
Monius for the invitation to speak there and for her thoughtful and instruc-
tive comments on my lecture. I also thank Ben Williams and James Reich for
their comments on my lecture and for their hospitality during my time in Cam-
bridge, MA. Finally, I thank Shaman Hatley for having read a draft copy of this
essay and for the insightful comments and suggestions he offered in doing so.
via free access
Appendix A: utsarga and apavāda as Exemplified in Aṣṭādhyāyī

(A) 3.2.1 and 3.2.3
3.2.1. karmaṇy aṇ.

3.2.3. āto ’nupasarge kaḥ.
1. The Pāṇinian example here offered is of a general rule, or utsarga, that

is preempted by a special rule, or apavāda, in a particular circumstance.
The general rule is offered at Aṣṭādhyāyī (A) 3.2.1; the special rule is sub-
sequently given at A 3.2.3. The rules here are used to generate what are
referred to as upapadasamāsas.
2. A principal purpose of the grammar, of course, is to allow one to gen-
erate the correct form of speech by the application of a series of rules
that build, as it were, the correct form of speech in a series of operational
steps. The point of the present example is that the general rule, A 3.2.1,
regularly—indeed, always—applies,60 unless the parameters of the spe-
cial rule, A 3.2.3, apply, in which case the general rule is suspended. If the
special conditions that trigger the special rule are not met, the general
rule is reverted to as authoritative and is applied.
3. A 3.2.1. states that the affix aṇ is added after a verbal root (dhātoḥ is to
be read into A 3.2.1 by anuvṛtti, or ellipsis, from a preceding rule) when
the object of the verb is in composition with it as an upapada. This
yields, e.g., kumbhakāra (“potter”) and nagarakāra (“city-builder”); kāṇ-
ḍalāva (“branch-cutter”) and śaralāva (“reed-cutter”); and vedādhyāya
(“the learning of the Veda”) and carcāpāṭha (“the reading of Carcā”). In
each of these cases, the final “a” of each word is present because of the
affixation of aṇ. (The a of aṇ is precisely this letter or akṣara, while the
ṇ of aṇ is a “marker,” an anubandha or it, which is dropped out in the
derivational process and does not appear in the final form of the given
word or words being derived, though it does serve to trigger certain rules
and operations in the grammar, and so has a meta-function in the deriva-
tional process.)
4. On the other hand, the affix aṇ is not allowed to be added in cases where
the special rule found in A 3.2.3 does apply. This rule states that the affix
ka will come after a verbal root that ends in “ā,” if and only if the verbal
root in question is not preceded by a preverb (upasarga), and only when
60 Note that so much is made clear by the Kāśikāvṛtti ad A 3.2.1, which reads in part as follows:
sarvatra karmaṇy upapade dhātor aṇpratyayo bhavati.
via free access
316 nemec
the object of the verb is in composition with it (karmaṇi here is extended

from A 3.2.1 to the present rule by way of anuvṛtti or ellipsis). Examples of
the application of this special rule are: goda (“giver of cows”) and kambal-
ada (“giver of a blanket”). In both of these examples, the final “a” of each
word is present because of the affixation of ka. (The a of ka is precisely
this akṣara, while the k of ka is again a “marker” or anubandha/it that is
dropped out in the derivational process and does not appear in the final
form of the word being derived.)
5. If we were to have been able to use the affix aṇ instead of the affix ka in
the present pair of examples, we would have been required to preserve the
long ā in the verbal root dā, which is present in both examples (goda and
kambalada), generating thereby the wrong forms. Using rule 3.2.1 instead
of 3.2.3 (and thus using aṇ instead of ka) would give us the wrong forms
*godā and *kambaladā in place of goda and kambalada. This would be
bad!
6. Similarly, in an instance where the verb employed does take an upasarga
or preverb, we have an exception to the special rule, which prohibits the
use of ka in place of aṇ precisely when there is a preverb affixed to the
verb. Thus, even when the verbal root in question ends in long ā and is
in composition with its object as an upapada, rule 3.2.3 cannot apply. A
pair of examples is given with the verbal root dā + sam. The apavāda rule
(A 3.2.3) is not allowed to take effect, and therefore ka is not appended
to the verb. Thus, we derive gosaṃdāya (“who ceremoniously gives a
cow”) and vaḍabasaṃdāya (“who ceremoniously gives a vaḍaba horse”),
and here the final vowel (a) of each term is the product of aṇ and not
ka.
7. What is at stake in this last pair of examples (viz., those of gosaṃdāya
and vaḍabasaṃdāya) is the status of the long vowel (ā) in the verbal root
dā + sam. If the affix ka were used in the present examples, then the
“marker” (anubandha, it) k would have triggered another rule of the gram-
mar (A 6.4.64, see paragraph 10, below), one that deletes this long vowel.
If this marker (k) is not in play, on the other hand, as it is not, for example,
when the affix marked with the anubandha ṇ is used (as in aṇ of A 3.2.1.),
then the rule in question that deletes the long ā does not apply.
8. So, to review before concluding: we here have a general rule, A 3.2.1, which
applies generally. It is the utsarga. We have a special rule or apavāda
(A 3.2.3) that applies within what otherwise would have been the domain
of this general rule, but which is commanded by the special rule under
special circumstances, with the application of the special rule blocking
the application of the general rule or utsarga.
via free access
9. How does this figure in the present example? Well, A 3.2.3 should apply
wherever we have the requisite special circumstances, namely whenever
a verbal root ending in “ā” is in composition with its object as an upapada,
as it did in the derivation of goda and kambalada. But the apavāda rule
(A 3.2.3) does not apply in the derivations of gosaṃdāya and vaḍabasaṃ-
dāya, even though in both cases we have a verbal root that ends in ā and
that is in composition with its object as an upapada. This is so, because in
both cases the verbal root appears with a preverb (upasarga), and A 3.2.3
specifies that it can apply only in instances when there is no preverb (it
reads: anupasarge).
10. Because the examples gosaṃdāya and vaḍabasaṃdāya include a verbal
root that appears with an upasarga (the verb is dā, the upasarga or pre-
verb is sam, of course), they are no longer to be marked by ka. This is
significant because of what k does: according to A 6.4.64 (āto lopa iṭi
ca), the final ā would have been elided, the rule effecting as much either
because it appears before an ārdhadhātuka suffix (defined at A 3.4.114)
augmented with iṭ, or when the affix begins (as would have been so in
the present instance) with a vowel and has an indicatory marker (aun-
bandha or it) of k or ṅ (the affix a in ka begins with a vowel—a of course
being a vowel and the first member of a single-lettered affix—this after
taking into account that the k is a marker and not a part of the affix
itself).61
11. Thus, in these instances A 3.2.3 does not apply. And here is the key to
understanding the present example of utsarga and apavāda: the utsarga
or general rule (i.e., A 3.2.1) is thus again deemed operative and auto-
matically so, by default. This is so because it is understood always to
apply, only excepting when an apavāda overrules or suspends it. What
this means is that there is no need to write yet another rule to account for
the derivation in instances where the rule of exception or apavāda is itself
rendered inoperative by an exception; for upon the suspension of the
apavāda rule, the utsarga is immediately and automatically understood
once again to apply. As such—returning now to the pair of examples here
reviewed—the final a of gosaṃdāya and vaḍabasaṃdāya results from aṇ
being applied, not ka. One returns to the original rule by default, thus pro-
viding for the final letter a of both gosaṃdāya and vaḍabasaṃdāya. And
the anubandha ṇ, unlike k, does not trigger A 6.4.64 and thus does not lead
61 Note that the Kāśikāvṛtti ad A 6.4.64 reads in part as follows: iḍādāv ārdhadhātuke kṅiti
cākārāntasyāṅgasya lopo bhavati.
via free access
318 nemec
to the elision of ā in the verbal root dā appearing in both compounded

words. In a word, apavādas can only be written in contexts where an
utsarga is always otherwise operative.
References
Primary Sources
Aṣṭādhyāyī. See Kāśikāvṛtti.
Īśvarapratyabhijñākārikā. Raffaele Torella, ed. The Īśvarapratyabhijñākārikā of Utpala-
deva with the Author’s Vṛtti: Critical Edition and Annotated Translation. Rome: IsMEO,
1994.
Īśvarapratyabhijñākārikāvṛtti. See Īśvarapratyabhijñākārikā.
Īśvarapratyabhijñāvimarśinī. Madhusudan Kaul Shāstrī, ed. The Īśvarapratyabhijñā of
Utpaladeva with the Vimarśinī by Abhinavagupta. KSTS 22 and 23. Pune: Arya-
bhushan Press, 1918–1921.
Īśvarapratyabhijñāvivṛtivimarśinī. Madhusudan Kaul Shāstrī, ed. The Īśvarapratyabhi-
jñā Vivṛtivimarśinī by Abhinavagupta. 3 volumes. KSTS 60, 62 and 65. Pune: Arya-
bhushan Press, 1938, 1941, and 1943.
Kāśikāvṛtti. Swami Dwarikadas Shastri and Pt. Kalikaprasad Shukla, eds. The Kāśikāvṛtti
of Vāmana-Jayāditya, with the Nyāsa or Pañcikā of Jinendrabuddhipāda and the Pada-
mañjarī of Haradatta Miśra. 6 vols. Prachya Bharati Series 2–7. Varanasi: Pracya
Bharati Prakashan, 1965.
Tantrāloka. Paṇḍit Mukund Rām Shaāstrī (vol. 1) and Paṇḍit Madhusudan Kaul Shāstrī
(vols. 2–12), eds. The Tantrāloka of Abhinava Gupta, With Commentary by Rājānaka
Jayaratha. 12 vols. KSTS 23, 38, 30, 36, 35, 29, 41, 47, 59, 52, 57 and 58. Allahabad:
The Indian Press Ltd. (vols. 1, 5–6); Bombay: Nirnaya Sagar Press (vols. 7, 10–12),
Tatva-Vivechaka Press (vols. 3–4, 8–9), and Shri Venkateshvar Steam Press (vol. 2),
1921–1938.
Mahābhāṣya. F. Kielhorn, ed. [1880–1885], revised by K.V. Abhyankar (1972–1996). Vyā-
karaṇa-Mahābhāṣya of Patañjali. 3 vols. Bhandarkar Oriental Research Institute,
Poona, 1996–2005.
Śivadṛṣṭi. John Nemec, ed. The Ubiquitous Śiva: Somānanda’s Śivadṛṣṭi and His Tantric
Interlocutors. New York: Oxford University Press, 2011.
Śivadṛṣṭi. Madhusudan Kaul Shastri, ed. The Śivadṛṣṭi of Somānandanātha with the Vṛtti
by Utpaladeva. KSTS 54. Pune: Aryabhushan Press, 1934.
Śivadṛṣṭivṛtti. See Śivadṛṣṭi.
via free access
Secondary Sources
Dyczkowski, Mark. 1987. The Doctrine of Vibration: An Analysis of the Doctrines and
Practices of Kashmir Shaivism. State University of New York Series in the Shaiva Tra-
ditions of Kashmir. Albany: State University of New York Press.
Edelmann, Jonathan. 2014. “Introduction: Innovation in Hindu Traditions.” Special
issue: Innovation in Hindu Traditions. International Journal of Hindu Studies 18 (2):
113–118.
Geertz, Clifford. 1973. “Religion as a Cultural System.” In The Interpretation of Cultures:
Selected Essays. New York: Basic Book.
Macdonell, A.A. [1900] 1962. A History of Sanskrit Literature. Reprint, Delhi: Motilal
Banarsidass.
Mohanty, J.N. 2010. A Dictionary of Cultural and Critical Theory, edited by Michall Payne
and Jessical Rae Barbera. Second Edition. Oxford: Wiley-Blackwell.
Monier-Williams, M. 1891. Brāhmanism and Hindūism; Or, Religious Thought and Life in
India, As Based on the Veda and Other Sacred Books of the Hindūs. 4th edition. New
York: Macmillan and Co.
Nemec, John. 2011. The Ubiquitous Śiva: Somānanda’s Śivadṛṣṭi and his Tantric Interlocu-
tors. New York: Oxford University Press.
Nemec, John. 2013. “On the Structure and Contents of the Tridaśaḍāmaratantra, A
Kaula Scriptural Source of the Northern Transmission.” Journal of Hindu Studies 6
(3): 297–316.
Nemec, John. “Kashmir.” 2015. Oxford Bibliographies: Hinduism, edited by Alf Hiltebei-
tel. New York: Oxford University Press. October 26, 2015. doi: 10.1093/OBO/97801953
99318-0156
Pollock, Sheldon. 1985. “The Theory of Practice and the Practice of Theory in Indian
Intellectual History.” Journal of the American Oriental Society 105 (3): 499–519.
Pollock, Sheldon. 1989. “Mīmāṃsā and the Problem of History in Traditional India.”
Journal of the American Oriental Society 109 (4): 603–610.
Pollock, Sheldon. 1993. “Deep Orientalism? Notes on Sanskrit and Power Beyond the
Raj.” In Orientalism and the Postcolonial Predicament, edited by Carol Brecken-
ridge and Peter van der Veer, 76–133. Philadelphia: University of Pennsylvania
Press.
Pollock, Sheldon. 2006. The Language of the Gods in the World of Men: Sanskrit, Culture
and Power in Premodern India. Berkeley and Los Angeles: University of California
Press.
Ratié, Isabelle. 2011. Le Soi et l’Autre: Identité, différence et altérité dans la philoso-
phie de la Pratyabhijñā. Jerusalem Studies in Religion and Culture, vol. 13. Leiden:
Brill.
Sanderson, Alexis. 2005. “A Commentary on the Opening Verses of the Tantrasāra
of Abhinavagupta.” In Sāmarasya: Studies in Indian Arts, Philosophy and Interreli-
via free access
320 nemec
gious Dialogue, edited by Sadananda Das and Ernst Fürlinger, 89–148. New Delhi:
D.K. Printworld.
Smith, David. 1985. Ratnākara’s Haravijaya: An Introduction to the Sanskrit Court Epic.
Delhi: Oxford University Press.
via free access
chapter 13
Toward a History of the Navarātra, the Autumnal

Festival of the Goddess
Bihani Sarkar
In this essay, I wish to address the problems a historian encounters while

explaining the function and origin of ancient rituals. One is particularly con-
fronted by these problems when dealing with a ritual such as the Navarātra.
With regard to its function, the festival resists sharp distinctions between the
sacred and the temporal because it simultaneously propitiates a deity and sol-
emnizes the authority of a ruler. It seems to be two things at the same time:
a rite of religious power and a rite of political power. In fact, in the South-
ern Navarātra, for instance as celebrated in Vijayanagara, the worship of the
Goddess would take place largely out of view in a private shrine, while all the
individual rites of the festival appeared to publicly celebrate a cult of the king in
the larger communal area: the Navarātra thus appeared, as it did to Portuguese
and Persian visitors to the Vijayanagara court, to be a political festival with
a minor religious dimension.1 While explaining its origins, we run into even
greater difficulty as the earliest traces of the Navarātra are found in more than
one distinctive religious tradition: the Vaiṣṇava, the Śaiva, the Purāṇic and even
possibly in regional traditions of communities outside mainstream ‘Hindu’ tra-
ditions. Where it truly “originates” is therefore difficult to see, though for clar-
ity’s sake I have proposed here that the Vaiṣṇava domain was where a mature,
theologically coherent2 conception of the rite evolved. On the whole the ritual
appears to have been of a composite character at each stage of its manifesta-
tion. The overall impression is that we are looking at many permutations of dif-
ferent rites with different origins that attached themselves around the central
figure of the Goddess, and through her, and the demon-slaying mythologies sur-
rounding her, acquired a structural and thematic unity. In the following, I shall
1 Stein 1983, 78, 80, Sarkar 2017, 211–212, 261. Similarly, Abbe Dubois, a visitor to the Mysore
Navarātra in the early nineteenth century, described it as a “soldier’s feast,” and as “entirely
military” (Kinsley 1988, 106).
2 For instance the conception of Māyā that we find in earlier speculative traditions on cos-
mogony was added to the overall presentation of the goddess.
© bihani sarkar, 2020 | doi:10.1163/9789004432802_015

via free access
322 sarkar
present and probe these ambiguities of function and origin through a scheme
of the richly varied regional traditions of the Navarātra that emerged in the
course of its history.
Navarātra, the autumnal festival of the warrior goddess Caṇḍikā, is today
one of the most eagerly anticipated events of the Hindu calendar. Built up over
nine lunar days and culminating on a tenth, the festival fulfills several, appar-
ently disparate, purposes: it offers obligatory worship to the Goddess, without
which her wrath could become implacable (so legends warn in dire tones);
wards away omens from—and thereby symbolically cleanses and renews—a
community of people; and bestows the ritual stamp of victory on the mili-
tary forces of a kingdom. What is particularly noteworthy even in the mod-
ern ceremony is the symbolic connection between the political, the martial,
and the religious, manifested by a priest through a sequence of meaningfully
choreographed rites and staged within a lavishly ornamented arena of worship
publicly open to all. Facilitating this connection is the Goddess herself, who
intertwines in her being an image of secular rulership and transcendent, or
spiritual, sovereignty. Little, though, is understood about the historical reasons
for the culmination of this overlap in the figure of Durgā and its ritualized real-
ization in the ceremony of the Navarātra. Doubtless, the Navarātra, given that it
was dedicated to a war-goddess, played a significant role in preparations under-
taken by a medieval kingdom to wage war, and furthermore, to affirm social
structure, as notable studies in the past by Alexis Sanderson (2007, 195–311),
Shingo Einoo (1999, 33–70) and Ralph Nicholas (2013) have shown. The sea-
son of autumn, which in many cultural traditions, and also in classical India,
was when armed campaigns would take place, must also have formed a rea-
son for the presence of military rituals such as the lustration of weapons and
war-animals during Durgā’s Navarātra: following the monsoon, during which
it is notoriously difficult to make journeys, the autumn, when the skies are
clear and the weather cooler, formed the perfect season to venture forth on
campaign.3 Doubtless there must also have been an association between social
governance, its urban political locus, and the Goddess, as there is in the modern
ceremony. But when did the autumnal festival acquire such a role in sanctify-
ing heroic endeavor, and in affirming roles and functions dispersed within the
social organization? For in many Navarātras certain groups and lineages were
traditionally associated with particular rituals: for example the priest’s duty, the
3 Dominic Goodall drew my attention to this point, while noting examples from the Raghu-
vaṃśa, chapter 4, and Cambodian inscriptions; see Goodall 2014, 187–188.
via free access
toward a history of the navarātra 323
cutting of the head of the animal, the making of the effigy, or the provision
of virgin girls for worship, etc. This still remains at the forefront of any basic
inquiry into the nature of this ritual.
However, answering this question presents certain methodological compli-
cations. Literature (primarily in Sanskrit) indicates that during the period the
festival developed and was popularized throughout South Asia, viz. the 5th
to 12th centuries CE, it grew into a locally diverse tradition. At present these
regional traditions seem on the surface to be but tenuously interrelated, and
in their diversity forestall our entertaining the possibility of there having been
common templates of origination.
Apart from the Navarātra’s multiplicity of form, other factors have pre-
vented, it would seem, a full history of the festival from being undertaken—
apart from, it is important to note, Einoo’s (1999) pioneering study, “The
Autumn Goddess Festival Described in the Purāṇas.” These factors are as fol-
lows: difficulty in interpreting and evaluating sources; confusion prompted by
the presence of non-Brahmanical rituals within an outwardly Purāṇic-Brah-
manic ritual framework; and ambivalence in status because of the important
roles played by people outside the caste-system in the ritual sequence. How-
ever, these difficulties, confusions and ambivalences are not insuperable, and
do in fact point to an important characteristic of the ritual: that its position
within either the Brahmanical or the non-Brahmanical realms was never very
clear. Both sides claimed certain aspects of the ritual as theirs, and in fact
operated in tandem within its domain. A political synergy between different
power-groups was effected through the course of the festival, as indeed has
been recently shown by Nicholas (2013) and which will grow even more evi-
dent through surveying the different traditions.
However, if we look at a wide range of ritual descriptions in Sanskrit con-
tained in Purāṇas for which the conjectured dates seem reliable, in their reused
forms in Dharmaśāstric compendia (nibandhas) and in ritual manuals (pad-
dhatis), together with ethnographic accounts of ceremonies, where available, a
historical pattern begins to emerge. The full analysis of that pattern is treated in
my recent book, Heroic Shāktism: the Cult of Durgā in Ancient Indian Kingship,
including texts and translations from the relevant ritual descriptions.4 Since it
may also be useful to offer a succinct overview of the historical pattern, here
I offer a condensed summary of that larger narrative, presented in schematic
form below. I also take the opportunity here to include three very early sources
that I did not have the opportunity to consider for the arguments made in
4 Sarkar 2017, 210–274.
via free access
324 sarkar
my book: the Southern Cilapattikāram,5 and two passages from the Mahāb-
hārata and the Kādambarī. These offer fresh insights int0—if not the Navarātra
proper—the ritual background of the Goddess: her worshippers, their prove-
nance and the purpose for which she may have been worshipped prior to the
development of the liturgical materials on the Navarātra. The consideration of
the Southern material in particular leads to re-conceiving the historical devel-
opment of the Goddess’s worship south of the Vindhya mountains. In the book,
I had suggested that
The Deccan seems to have followed in the wake of the eastern form of the
Navarātra outlined in the Devī[purāṇa] and the Kālikā[purāṇa] until at
least the early half of the 14th century [on the basis of passages from those
works appearing in Deccan Dharmaśāstric nibandha-literature] … The
gradual independence of the southern tradition and its advocacy by the
15th century of a Navarātra that was qualitatively different from the east-
ern tradition in that it celebrated Daśamī differently and eschewed rites
that were Tantric in their tone are attested by the eyewitness accounts of
the Navarātras of the Vijayanagara kingdom, of Mysore under the Wode-
yars, of Ramnad and Śivagaṅgai in Tamil Nadu.
Sarkar 2017, 258–259
On the other hand, the Cīlapattikāram suggests that already in the early cen-
turies of the common era, a local form of worshipping the Goddess for power
in battle, like the Navarātra, was celebrated in the Tamil country, and that this
included possession, trance and bacchanalia. On this basis, it is possible to sug-
gest, first, that long before descriptions of Durgā’s worship appeared in mate-
rials of an Eastern provenance, she was popular in the south and, second, that
the direction of liturgical influence could have been the other way: the ritual
of bacchanalian enjoyment offered to a goddess of battle could have entered
into Eastern liturgies, in which they occupy a prominent place, from Southern
prototypes. Moreover, all three textual examples are magnificently composed,
and exemplify all that is most vivid and energetic in poetry about the Goddess.
In some of our earliest sources, the rite did not begin in autumnal Āśvina, the
month usually associated with the Navarātra. Rather it began in the monsoon
month of Śrāvaṇa, and prior to that seems to have been a popular festival cel-
ebrated by everyone regardless of sectarian affiliation. It then came to replace
a more established set of Brahmanical military traditions (such as the worship
5 The arguments made about this Southern work are hindered by my lack of knowing Classical
Tamil. The reader is asked to treat them as preliminary and to refer to the original source.
via free access
of weapons and the lustration of the army) practised in Āśvina. Once these
Vedic royal traditions were harmonized with the worship of the Goddess, they
altered in character to become a goddess-centred heroic tradition, in which
sanguinary rites to calm down Durgā’s fiery nature began to dominate. But fur-
ther alterations in the character of the ritual followed, as it was incorporated by
other religious “specialisms.” One of the most critical transformations to have
occurred in the structure of the Navarātra is the appearance of Tantric rituals in
descriptions of the rite emerging from East India, notably from Orissa and the
kingdom of Mithilā. Compared to the military festival of the earlier Vaiṣṇava
and then the Vedicized formats, these rituals amplify the power-bestowing effi-
cacity of the ritual by including rituals that grant siddhis (powers). Moreover,
there is no single goddess, but many, and of many forms, names and natures.
In literature from Mithilā, the rite expands to nine days to include, apart from
the worship of nine forms of Durgā of different colours, an array of Tantri-
cized rituals such as the purification of elements (bhūtaśuddhi), the worship
of the sixty-four yoginīs, the installation of mantras in the body (nyāsa), self-
identification with the deity (a ritual that, although also found in pre-tantric
materials, came to be associated with tantric practice6), rites bestowing pow-
ers (siddhi) held at midnight, and the heightening of the Goddess’s personality
so that her ferocious properties are thought to take over. She is invoked as Kālī-
Lauhadaṇḍā (Kālī, goddess of the iron rod) in mantras in the mediaeval Bengali
rite. In the Maithila rite she is even summoned as Cāmuṇḍā into a bel branch,
which is then worshipped as the vehicle of her essence throughout the dura-
tion of the worship. In Orissa, as shown by Sanderson (2007), mantra elements
from the Kashmirian Kālīkula were incorporated into the Mahānavamī tradi-
tions of Bhadrakālī. The effect of this Tantricization was the enhancement of
the power-bestowing agency of the ritual, desirable no doubt for rulers eager
to achieve victory in the battles they were about to undertake.
The sources in which the above ritual patterns are described are as follows
(the specific references with emended Sanskrit texts and analyses are to be
found in the locus indicated in the accompanying footnotes):
I. Early Vaiṣṇava phase in the monsoon7

Harivaṃśa 57.35–36; Mahābhārata 4.5.29ff. and 6.22.6ff., old Skandapu-
rāṇa 60.46; Kādambarī pp. 30–31; Harṣacarita p. 126; Caṇḍīśataka 16;
Gaüḍavaho 318, 319, Purāṇic citations in Dharmaśāstric compendia from
Mithilā and Bengal.
6 Goodall et al. 2005, 13, note 5.

7 Sarkar 2017, 214–221.
via free access
326 sarkar
table 13.1 Developmental phases of the Navarātra
I. circa 4th century II. circa 5th century III. circa 8th century CE: IV. circa 14th century CE: The
CE: Early Vaiṣṇava CE: Incorporation Expansion and inclusion of Southern and Western Court
Phase in the mon- with a pre-existing Tantric power-rituals in East- Traditions of the Kingdoms
soon Brahmanical military ern Court Traditions, notably of Devagiri and Vijayanagara
festival in Āśvina in the kingdom of Mithilā8
Worship of Kṛṣṇa’s Worship of Bhadra- Appearance of a ten day On Pratipat: King enthroned
sister Kālarā- kālī is attached to a structure spread out over and given an amulet empow-
tri/Nidrā, a dark, pre-existent strata of the First lunar day (Prati- ered by the goddess’s mantra;
blood-thirsty, brahmanical rituals pat) to the Tenth lunar day vow of fasting and abstinence
alcohol-loving of kingship advocated (Vijayadaśamī) in the bright to be undertaken by him;
goddess associ- in Vedic literature. phase of Āśvina king’s sword and sceptre cer-
ated with night, (Bṛhatsaṃhitā 43; emonially presented to him
sleep, hallucination Kāṭhakagṛhyasūtra recitation of the Devīmāhāt- and placed at the base of the
and enchantment 57.1 and Arthaśāstra mya (caṇḍīpāṭha) lineage goddess’s image; sum-
(māyā) 2.30.519) moning of the goddess in the
The first seven days involve: person of the king
Buffalo-sacrifice by Celebrated on kalaśapūjā (worship of deities
a ruler to propitiate the Eighth and including the goddess, the
King worships Durgā, Lakṣmī
the Goddess before Ninth lunar days Mothers and waters from and Vāgdevī in lidded pots
the onset of battle (Mahāṣṭamī and the sacred fords in a vase); a
in a flower pavilion (puṣpa-
Mahānavamī) of the king bathing in the sanctified
maṇḍapam) specially built
Incorporation of bright half of Āśvina waters from the kalaśapūjā;
to worship the goddess and
outcaste groups fasting, worshiping Śiva thrice
the king; worship of the king’s
referred to in San- Worship of the God- daily, animal sacrifice (paśub-
thirty two weapons (lohāb-
skrit literature as dess in a cloth in a ali); daily worship of the royal
hisārikapūjā), worship of the
“Śabaras” shrine built in the horses; fire oblations and feed-
royal insignia, worship of the
north-eastern part of ing a maiden royal horses and elephants;
Association with a military encamp- court assembly at the puṣ-
the Vindhya moun- ment The sixth (Ṣaṣṭhī) and the sev- pamaṇḍapa; king travels in
tain enth (Saptamī) lunar days pomp to an assembly hall
Worship of weapons involve awakening the goddess (āsthānamaṇḍapa) built for
Rites of self- with flowers, perfume in a bilva tree (bodhana), wor- the festival at which a dur-
mortification to and food, adapted ship of goddess as Cāmuṇḍā bar is held; public display of
be performed by from the pre-existing and Kālī in the branch, sum- goddess’s image next to the
heroes to demon- Vedic model of mili- moning her nine radiations enthroned king; spectacles in
strate their valour tary festivities. in nine leaves (navapatra- front of the āsthānamaṇḍapa
pūjā/patrikāpūjā),
8 The Nepalese Tradition, though deriving in the main from the Maithila tradition as embod-
ied in the Kārṇāṭa royal ceremony of the Kṛtyaratnākara and Durgābhaktitaraṅgiṇī, is much
more Tantric in character, involving mantra elements from the Kubjikā cult.
9 All three are cited in Sanderson 2005, 229–300. It was Professor Sanderson who pointed out
the existence and relevance of this archaic military stratum of rituals to me (personal com-
munication).
via free access
Table 13.1 Developmental phases of the Navarātra (cont.)
festival in Āśvina in the kingdom of Mithilā
Celebrated on the Staged entertainment enlivening an unfired From Pratipat to Navamī:

Ninth lunar day of of the Goddess clay image of the goddess daily worship by the king of
the dark half of Śrā- (prāṇapratiṣṭhā) nine maidens (kumārīpūjā)
vaṇa, the day after Kings to keep a night- as nine goddesses, Kumārī,
Janmāṣṭamī vigil on Mahāṣṭamī Nine wooden shrines to be Trimūrti/Trimurtinī, Kalyāṇī,
night and main- built on the Eighth lunar day Rohiṇī, Kālikā, Caṇḍikā,
Possession and bac- tain a fast for victory (Mahāṣṭamī), and the goddess Śāmbhavī, Durgā, Bhadrā
chanalia involving (śauryavrata) is to be installed in a gold or
dancing and the silver image, in a sword or in a Worship repeated till Navamī
singing of hymns Worship repeated on trident; worship involves char-
(suggested by the Mahānavamī followed iot and palanquin processions On Navamī: fire oblation to
Cīlapattikāram) by a parade the goddess (caṇḍīhoma);
On Mahāṣṭamī: worship of paśubali,
Buffalo-sacrifice the Nine Durgās (Rudracaṇḍā, king removes the amulet
Pracaṇḍā, Caṇḍogrā, Caṇ-
Festival performed ḍanāyikā, Caṇḍā, Caṇḍavatī, On Vijayadaśamī: worship
for the sake of victory Caṇḍarūpā, Aticaṇḍikā, Ugra- of a śamī tree according to a
and pacification caṇḍā), the eight mothers, tradition attributed to the
the sixty-four yoginīs, purifi- Gopathabrāhmaṇa; king
cation of the gross elements given weapons including
(bhūtaśuddhi), installation of five arrows by the priest; king
mantras on the body (nyāsa); goes to the śamī in pomp
restraining the breaths with his army; shooting
(prāṇāyāma); visualization of arrows in every direc-
and self-identification with tion to destroy enemies;
the deity, rite of the sword evening court assembly at
(khaḍgapūjā) in Nepal for the āsthānamaṇḍapa
powers (siddhis); paśubali
(animal sacrifice) and offer-
ing blood from a king’s arms
and naraśiraḥpradāna (offer-
ing a human head); worship
of weapons (astrapūjā/śas-
trapūjā); goddess is believed
to morph into a more uncon-
trollable presence requiring
constant placation
via free access
328 sarkar
Table 13.1 Developmental phases of the Navarātra (cont.)
festival in Āśvina in the kingdom of Mithilā
Blood sacrifice to pacify

demons in various direc-
tions and the sacrifice of a
dough image of the king’s
enemy (śatrubali) for “uni-
versal power” (sarvavaśyatā)
to take place at midnight
(ardharātrapūjā), when
the asterism Kanyā (Virgo)
joins aṣṭamī; navadurgāpūjā
again
On Mahānavamī: worship
of Bhadrakālī with mantras
from the Kālīkula in Orissa
(Sanderson 2007, 255–295);
worship of the Goddess in a
trident; repetition of rites on
Mahāṣṭamī; kumārīpūjā (wor-
ship of a maiden); rathayātrā
(chariot procession) of the
Goddess
On Daśamī: worship of god-

dess Aparājitā; śābarotsava;
royal consecration (abhiṣeka)
of king with empowered water
from the opening kalaśa-
pūjā
II. Incorporation with a Brahmanical military festival in Āśvina10

Viṣṇudharmottarapurāṇa 2.158.6cd–7, Agnipurāṇa 267.13cd–16ab (re-
peating Viṣṇudharmottara); Varāhapurāṇa cited in the Kṛtyaratnākara,
pp. 364–365.
10 Ibid., 221–226.
via free access
III. Expansion and inclusion of Tantric power-rituals in Eastern Court

Traditions11
Devīpurāṇa, Kālikāpurāṇa, Kṛtyakalpataru, Durgābhaktitaraṅgiṇī, Dur-
gāpūjātattva, Durgāpūjāviveka, Bhadrakālīmantravidhiprakaraṇa in San-
derson (2007); account of the Durgā Pūjā in Kelomal, West Bengal (Nich-
olas 2013).
IV. The Southern and Western Court Traditions12

Caturvargacintāmaṇi, Sāmrājyalakṣmīpīṭhikā, Puruṣārthacintāmaṇi, ac-
counts of ceremonies in Śivagaṅgai and Ramnad, Tamil Nadu (Price
1996), Portuguese traveller accounts from the Vijayanagara Empire (Stein
1983).
To the above sources, I would like to add a passage from the Mahābhārata
(first noticed and pointed out to me by Sahiṣṇu Bhaṭṭācārya, Bardhaman, West
Bengal in a personal communication), whose importance in regard to the wor-
ship of Nidrā, Durgā’s early form, requires emphasis. This passage appears in
the Sauptikaparvan (Mahābhārata X, 8.64–68) and suggests that the Vaiṣṇava
Nidrā, goddess of Sleep and Death, presided over and blessed battle as a dan-
gerous spirit (kṛtyā). Called Kālarātri, identified with apocalyptic destruction,
adorned with a peacock feather (śikhaṇḍinīm) that evokes her alliance with
her brother Kṛṣṇa, as prevalent in this period,13 she manifests herself when
Aśvatthāman, the son of Droṇa, secretly enters the Pāṇdava camp and goes on
a murderous rampage. Hers is a strange, menacing apparition:
kālīṃ raktāsyanayanāṃ raktamālyānulepanām |

raktāmbaradharām ekāṃ pāśahastāṃ śikhaṇḍinīm ||
dadṛśuḥ kālarātriṃ te smayamānām avasthitām |
narāśvakuñjarān pāśair baddhvā ghoraiḥ pratasthuṣīm ||
harantīṃ vividhān pretān pāśabaddhān vimūrdhajān ||
svapne suptān nayantīṃ tāṃ rātriṣv anyāsu māriṣa |
dadṛśur yodhamukhyās te ghnantaṃ drauṇiṃ ca nityadā ||
yataḥ pravṛttaḥ saṃgrāmaḥ kurupāṇḍavasenayoḥ |
tataḥ prabhṛti tāṃ kṛtyām apaśyan drauṇim eva ca ||
11 Ibid., 226–258.
12 Ibid., 258–270.
13 See Couture and Schmid 2001, Schmid 2002, Yokochi 2004 and Sarkar 2017, 41–69, for the
incorporation of Durgā in Vaiṣṇava traditions.
via free access
330 sarkar
tāṃs tu daivahatān pūrvaṃ paścād drauṇir nyapātayat |

trāsayan sarvabhūtāni vinadan bhairavān ravān ||
Mahābhārata X, 8.64–68
Good sir, they saw her, Kālarātri, standing, smiling, alone, blue-black in
hue, with red mouth and eyes, garlands and unguents of crimson, red
robes, a noose in one hand, a peacock feather [in her hair], binding men,
horses and elephants with her horrifying fetters while she stood, captur-
ing many headless ghosts trapped in her noose, leading those asleep in
their dreams to other Nights (rātriṣv14 anyāsu). And at all times the best
soldiers saw the son of Droṇa slaughtering. From the time when the battle
between the Kuru and Pāṇḍava armies began, they saw [both] that evil
spirit (tām kṛtyām) and the son of Droṇa. The son of Droṇa later felled
those who had first been struck by this divinity [Kālarātri], terrorizing all
creatures while shouting out ferocious bellows.
Towering over the nightmarish battlefield, a grinning image of Death, Kālarātri

governs both sleep and death, ensuring the interceptor certain triumph during
his secret raid.
The Tantricization of Durgā’s worship must have been well established by
700CE, by which time it must have already been correlated with worship of
Śiva, rather than Viṣṇu. Such is the impression created by a minutely detailed
description of Durgā’s shrine and her worshipper in Bāṇa’s Kādambarī
(pp. 224–22815). On his way from Hemakūṭa to Ujjayinī, the hero of the work,
Candrāpīḍa, stops for shelter at a shrine of the Goddess that comes midway.
The shrine is nestled in the midst of a densely wooded forest, and the narrative,
through a telescoping of perspective from outside to inside the shrine, provides
a leisurely description of its design, appearance and atmosphere. From afar
Candrāpīḍa first sees a “crimson ensign,” “inscribing the sky with a gold trident,
from which swung a terrifying bell making a raucous clanging (ghargharar-
ava) that dangled down from an iron chain attached to the tip, arranged with a
yak-tail whisk as splendid as a lion’s mane” (dolāyitaśṛṅgasaṅgilohaśṛṅkhalā-
valambamānagharghararavaghoraghaṇṭayā ca ghaṭitakesarisaṭāruciracāma-
rayā kāñcanatriśūlikayā likhitanabhaḥsthalam … raktadhvajam; p. 224). Going
ahead a little, he then sees that the Goddess Caṇḍikā “was enclosed by a
door made from the ivory of wild elephants, as yellowish-white as fragments
14 This appears to be a play with Rātri, another name for this goddess.
15 All references to the Kādambarī in this paragraph are from Peterson’s edition of 1889.
via free access
of ketakī filaments, and an iron architrave (toraṇa) bearing an ornamental

garland of black iron mirrors surrounded by a row of red yak tail whisks
resembling a garland of Śabara heads horrific with tawny hair” (ketakīsūcikha-
ṇḍapāṇḍureṇa vanadviradadantakapāṭena parivṛtāṃ lohatoraṇena ca raktacā-
maraparikarāṃ kālāyasadarpaṇamaṇḍamālāṃ śabaramukhamālām iva kapi-
lakeśabhīṣaṇāṃ bibhrāṇena … caṇḍikām; pp. 224–228). Then he notices the
dvārapāla (guardian of the gate), about which it is said that “[Caṇḍikā] had
protected her entrance with an iron buffalo installed in front, which, because of
the fact that it had been marked by palms [dyed with] red-sandalwood, seemed
to have been stamped by Yama’s hand-prints red with blood, the red eyes of
which were being licked by jackals greedy for drops of blood” (sanāthīkṛta-
dvāradeśām abhimukhapratiṣṭhena ca vinihitaraktacandanahastakatayā rud-
hirāruṇayamakaratalāsphāliteneva śoṇitalavalobhalolaśivālihyamānalohitalo-
canena lohamahiṣeṇa; Kādambarī, p. 224). Then through the main entrance,
the temple yard: “Her courtyard was adorned with thickets of red aśoka trees,
the spaces between the branches of which were made gapless by flocks of
perching red cockerels, [trees] which appeared to reveal unseasonal clusters
of blooms in their fear” (śākhāntarālanirantaranilīnaraktakukkuṭakūlaiś ca
bhayād akāladarśitakusumastabakair iva raktāśokaviṭapair vibhūṣitāṅgaṇāṃ;
p. 225). (More in fact is said about the overflowing mass of flowers, trees and
even lion cubs that populate her front courtyard, slippery with blood.) Then
the portal to the sanctum sanctorum, a riot of colour and form: “She was being
illuminated by the entrance, on which there were hanging cloths reddened by
lamp-smoke, a row of bracelets made of peacock-throats festooned [over it],
a garland of bells closely-set and pale with powdered flour-cakes, which sup-
ported two door-panels, [studded] with tin lion heads with thick, iron pins in
their centres, barricaded with an ivory-rod bolt, carrying [what seemed to be]
a necklace of sparkling bubbles that were mirrors oozing yellow, blue and red
[light]” (avalambamānadīpadhūmaraktāṃśukena grathitaśikhigalavalayāva-
linā piṣṭapiṇḍapāṇḍuritaghanaghaṇṭāmālabhāriṇā trāpuṣasiṃhamukhama-
dhyasthitathūlalohakaṇṭakaṃ dattadantadaṇḍārgalaṃ galatpītanīlalohitadar-
paṇasphuritabudbudamālaṃ kapāṭapaṭṭadvayaṃ dadhānena garbhagṛhadvā-
radeśena dīpyamānāṃ; ibid.).
Then follows the image of the Goddess, which in its association with the
terrible, and in the predominance of the colour red, matches the concep-
tion of Kālarātri in the passage from the Mahābhārata: “She was installed on
an altar of black stone” (adhyāsitāñjanaśilāvedikām, Kādambarī, p. 224). “Her
feet were never bereft of cloths [dyed with] red lac thrown upon the mound
of her seat [on the altar] as if they were the lives of all creatures arrived
there for shelter; she resembled an inhabitant of the Underworld because of
via free access
332 sarkar
the intense darkness obstructed [only] by the flashes from axes, spears, etc.,
weapons deadly for beings, that seemed to hold nets of hair stuck from decap-
itations because of the reflections of black yak-tail whisks cast [upon their
surfaces]; she was adorned in garlands of bilva leaves furnished with gleam-
ing fruits and buds anointed with red sandalwood, that were like hanging gar-
lands of infant-heads; she expressed cruelty with limbs worshipped with clus-
ters of kadamba flowers ruddy with blood, which horripilated, it seemed, at
the thrill of the flavour of the keen roar of drums during the animal-offering;
she bore the coquettish apparel of a woman going out to meet Mahākāla at
night, with a vine-like body furnished with a raiment reddened with saffron-
dye, with a face with red eyes, whose brows were furrowed into a frown,
whose lip was crimsoned with betel that was blood, whose cheeks were red-
dened by the light shed from ear-ornaments of pomegranate flowers, with
a forehead on which there was a tilaka dot of vermillion made by a Śabara
beauty, covered by a magnificent gold turban (cāmīkarapaṭṭa). She was wor-
shipped by goats … mice … antelope and black serpents … She was praised on
all sides by flocks of old crows.” (piṇḍikāpīṭhapātibhiś ca sarvapaśujīvitair iva
śaraṇam upāgatair alaktakapaṭair avirahitacaraṇamūlāṃ patitakṛṣṇacāmara-
pratibimbānāṃ ca śiraśchedalagnakeśajālakānām iva paraśupaṭṭiśaprabhṛtī-
nāṃ jīvaviśasanaśastrāṇāṃ prabhābhir baddhabahalāndhakāratayā pātālani-
vāsinī, ivopalakṣyamāṇāṃ raktacandanakhacitasphuratphalapallavakalitaiś ca
bilvapattradāmabhir bālakamuṇḍaprālambair iva kṛtamaṇḍanāṃ śoṇitatā-
mrakadambastabakakṛtārcanaiś ca paśūpahārapaṭahapaṭuraṭitarasollasitaro-
māñcair ivāṅgaiḥ krūratām udvahantīṃ cārucāmīkarapaṭṭaprāvṛtena ca lalā-
ṭena śabarasundarīracitasindūratilakabindunā dāḍimakusumakarṇapūrapra-
bhāsekalohitāyamānakapolabhittinā rudhiratāmbūlāruṇitādharapuṭena bhṛ-
kuṭikuṭilabhruṇā raktanayanena mukhena kusumbhapāṭalitadukūlakalitayā ca
dehalatayā mahākālābhisārikāveṣavibhramaṃ bibhratīṃ … chāgair … ākhubhir
… kuraṅgair … kṛṣṇasarpair … ārādhyamānāṃ sarvataḥ kaṭhoravāyasagaṇena
… stūyamānām; pp. 225–226).
A Draviḍa ascetic, portrayed as a comical figure, is said to be her priest; per-
haps Bāṇa was conscious in making the priest a Draviḍa, on account of the
widespread worship of the goddess Koṛṛavai, later correlated with Durgā, in
the South? There are apparently several Tantric rites that Bāṇa pejoratively as-
sociates with the priest: he, “the ageing Draviḍa religious man” “demeans Durgā
with his prayers for the boon of sovereignty over the Southern lands” (dak-
ṣiṇāpatharājyavaraprārthanākadarthitadurgeṇa … jaraddraviḍadhārmikeṇa;
p. 226); “he had copied a hymn to Durgā on a strip of cloth” (paṭṭikālikhitadur-
gāstotreṇa; ibid), “he had collected palm-leaf manuscripts of spells, Tantras
and jugglery the letters of which were written in red lac and fumigated with
via free access
smoke” (dhūmaraktālakākṣaratālapattrakuhakatantramantrapustikāsaṃgra-
hiṇā; ibid); “he had written down the [work known as ] the ‘Doctrine of Mahā-
kāla’ instructed to him by a withered Mahāpāśupata mendicant” ( jīrṇamahā-
pāśupatopadeśalikhitamahākālamatena, pp. 226–227); “he was one in whom
the disease of talking about [finding] treasure had arisen” (āvirbhūtanidhivā-
davyādhinā, p. 227); “in him the wind [disease] of alchemy had grown” (saṃ-
jātadhātuvādavāyunā, ibid.); “he entertained the deluded desire of becoming
the lover of a Yakṣa maiden” (pravṛttayakṣakanyakākāmitvamanorathavyāmo-
hena, ibid.); “his collection of practices for mastering mantras for invisibil-
ity had grown” (vardhitāntardhānamantrasādhanasaṃgraheṇa, ibid.); “he was
acquainted with a hundred tales about the marvels of the Śrīparvata mountain”
(śrīparvatāścaryavārttāsahasrābhijñena, ibid.); “his ear-cavities were punched
by those possessed by piśāca-demons, who had run to him when struck by
white mustard seed he had empowered with mantras more than once” (asakṛd-
abhimantritasiddhārthakaprahitapradhāvitaiḥ piśācagṛhītakaiḥ karatalatāḍa-
nacipiṭīkṛtaśravaṇapuṭena ibid.); and “he had used magic powders for snaring
women many times on aging mendicant ladies, who having arrived from other
lands retired [there to rest]” (anyadeśāgatoṣitāsu jaratpravrajitāsu bahukṛtvaḥ
saṃprayuktastrīvaśīkaraṇacūrṇena).
While it would be imprudent to treat this example of poetic literature as
a bald record of fact, it is possible to see Bāṇa’s extensive and richly crafted
episode of this horrific, yet magnificent, temple as a reflection of social atti-
tudes to the Goddess and her worship. There is a mix of suspicion, fear and
reverential awe underlying the image of the forbidding shrine tucked away in
the wilds, with its Tāntrika priest who knows not how ‘appropriate’ worship
should be conducted, and its blood-spattered, grisly interiors. The very opposite
of this ambivalent attitude surfaces in Bāṇa’s unequivocally laudatory poem
to Durgā, the Caṇḍīśataka—verse 8 of which is consciously alluded here in
“she seemed to be scolding the wild buffalo who had offended by moving the
trident-shaft by scratching his shoulders [on it]” (skandapīṭhakaṇḍūyanacalita-
triśūladaṇḍakṛtāparādhaṃ vanamahiṣam iva tarjayantīm; Kādambarī p. 226).
The topos of Mahiṣa scratching his back on the post appears in Caṇḍīśataka 8
too,16 in which there is a mischievous pun with the word sthāṇu that means
both “post” and “Śiva.” The saviour of Dharma in the Caṇḍīśataka contrasts
with the menacing though beauteous figure here. One may suggest that the
16 grastāśvaḥ śaṣpalobhād iva haritaharer prasoḍhānaloṣmā

sthāṇau kaṇḍuṃ vinīya pratimahiṣaruṣevāntakopāntavartī |
kṛṣṇaṃ paṅkaṃ yathecchan varuṇam upagato majjanāyeva yasyāḥ
svastho’ bhūt pādam āptvā hradam iva mahiṣa sāstu durgā śriye vaḥ ||
via free access
334 sarkar
wider context for this attitude of ambivalence is a historical transition: the God-
dess first imagined as we have seen as a kṛtyā in the Mahābhārata was being
absorbed within mainstream devotional practices, through which her demonic
attributes became ‘toned down,’ balanced by the benevolent and the charming,
but nevertheless remained, at the stage of Bāṇa’s compositions, tinged with a
degree of the terrible. The comic portrayal of the priest registers the fact that by
Bāṇa’s time Durgā’s worship had acquired firm cultural associations with Śaiva
tantric rites. There was also an association of her site of worship with wild envi-
ronments inhospitable to people, to flora and fauna in general (note that most
of the ornaments in her shrine are of plants and flowers) and even, it seems,
with a peculiarly Southern religious attitude.
Regarding the Southern Navarātra, it is tempting to conjecture that the roots
of worshiping the Goddess in Devagiri and Vijayanagara drew also upon the
older cult of Koṛṛavai, the stag-borne goddess described in the old Tamil poem,
the Cīlapattikāram.17 In fact many salient elements of Durgā’s rituals in gen-
eral (especially, though, in the East), such as the transactional nature of wor-
ship, trance, possession, ecstatic dancing, singing hymns, the important role
of women, virgin-worship and heroic self-sacrifice involving blood are to be
found even here, which suggest that among all elements of the Navarātra, these
appear to be the earliest. In Canto XII Koṛṛavai is said to be worshipped by
cattle-raiders for victory in their missions.18 The canto, called “Vēṭṭuvavari”
(The Hunter’s Song), portrays the Eyinar community worshipping their pro-
tective goddess for victory before setting off on a raid. The chapter describes
vividly the stages of pūjā at the shrine of Aiyai (Koṛṛavai), eulogized throughout
the canto as Durgā, the slayer of Mahiṣa, the sister of Viṣṇu and the consort of
Śiva. First, a respected Maravar lady Śālinī, an oracle, becomes possessed and
dances, singing a hymn urging the hunters to offer tribute to the Goddess.19
In the hymn Śālinī rebukes the men for growing weak and no longer robbing
passers-by,20 the implication being that the Goddess will re-invigorate them
with heroic zeal. After the oracle performs, a virgin is selected from the Eyi-
nars, in what appears to be an early form of the kumārīpūjā, and treated with
especial care as the Goddess. Dressed in tiger skin, with a snake of silver and a
wild hog’s tooth in her matted hair, a necklace of tiger-tooth, a bow of wood and
17 See Dikshitar 1939 and Danielou 2009. All references to the Cīlapattikāram are from Dik-
shitar 1939. I am grateful to Professor Goodall for kindly indicating the need to include
mention of this work within this account.
18 Cīlapattikāram, p. 180. See also Danielou 2009, 76–85, and Mahalakshmi 2011, 68–71.
19 Cīlapattikāram, XII.6–11; Danielou 2009, 77; Mahalakshmi 2011, 69.
20 Cīlapattikāram, XII.12–19; Danielou 2009, 77.
via free access
seated on a stag, the kumārī is brought before the shrine of the Goddess, set in
lush and verdant groves of fragrant and flowering trees.21 The women offer her
various gifts of dolls, beautiful birds, paints, scents, food and flowers with much
fanfare and the beating of drums.22 After worshipping the Goddess, the virgin
goes into a trance, and speaks to the heroine of the poem Kaṇṇakī, introducing
her to Koṛṛavai, who from this moment in the ritual, it is suggested, becomes a
living presence.23
Koṛṛavai appears. She bore a moon on her hair, a third eye on her forehead;
her lips were red, her throat blue with poison like that of her consort Śiva. The
snake Vāsukī was her girdle and she wore a bodice resembling snake-teeth, an
elephant’s hide over her upper body and a tiger skin over her hips; she carried
a trident. There are rich ornaments on her feet. Dark in hue as a sapphire,24
bejewelled, youthful, beautiful, ascendant on the head of the buffalo demon,
she is called, among many names, the sister of Kṛṣṇa, Durgā, Gaurī, the giver of
victory, worshipped by Viṣṇu and Brahmā,25 and also the defeater of Kaṃsa.26
The names and descriptions indicate that even at this early period, circa 450 CE,
when the poem is thought to have been composed,27 the Goddess, whose ini-
tial sectarian affiliation was with Kṛṣṇa, had already become associated with
Śiva, and moreover had acquired an independent identity as a supreme divin-
ity, worshipped by all the gods. Another hymn is sung by a girl to the virgin
dressed as the Goddess, in which the duality of Koṛṛavai-Durgā is emphasised
in a series of rhetorical questions or paradoxical contrasts: she is worshipped
by gods and is an exalted repository of Vedic knowledge, yet also stands on a
buffalo head adorned with wild animal hides; standing as light above the trinity
of Viṣṇu, Śiva and Brahmā, she stands also on a humble stag with twisted black
horns, holding aloft, with hands adorned with delicate bangles, a cruel sword;
she who is consort of Śiva with three eyes also has a fierce red-eyed lion and the
Vaiṣṇava conch and the discus.28 She is also said to have danced the marakkāl,
21 For the trees see Cīlapattikāram XII verse not indicated, p. 184; Danielou 2009, p. 79.
22 Cilapattikāram XII.20–53; Danielou 2009, 78; Mahalakshmi 2011, 71.
23 Cilapattikāram XII.51–53; Danielou 2009, 78.
24 Cīlapattikāram XII, verse not indicated, p. 188; Danielou 2009, 83–84.
26 Cīlapattikāram XII. verse not indicated, p. 188; Danielou 2009, 84.
27 Regrettably the dating of this fine work has not yet been settled. Dikshitar (1939, 8–10)
suggests sometime in the second century CE. Here I have cited the date proposed by
Zvelebil (1977, 132), which nevertheless is inconclusive. I am grateful to Dominic Goodall
for explaining the issues concerning the problem of dating this text and providing me with
Zvelebil’s study.
via free access
336 sarkar
a dance on wooden legs, to defeat demons. If anyone invokes her wearing a

victory garland (veṭci) before setting forth to seize cattle, omens of defeat will
appear in the enemy’s village, and the Goddess will accompany the hero on
his quest before his bow.29 In the song, the plenitude of captured cattle is then
praised, and the Goddess is asked to accept the raiders’ blood offered by cut-
ting their necks to her in thanks.30 This offering of flesh and blood is described
as the Goddess’s price for the victory conferred on the warrior and outlines the
transactional nature of the worship.31 The hymn becomes hypnotic at this cli-
mactic moment of blood-offering as in verse after verse the Goddess is asked
to accept the blood.
Four things illuminated by the description are worth pointing out. Koṛṛavai
is already treated as an eclectic deity merged with Durgā, herself a cluster of
Vaiṣṇava and Śaiva elements. From the context here it is tempting to specu-
late that Durgā as a deity of royal power and kingship may trace her roots from
local warrior communities such as the Eyinār, who sustained themselves by
periodic looting necessitating armed confrontation. In fact, the evidence of the
old Tamil text prompts us to refine the schematic diagram, and suggest that the
first phase of the ritual going back to the early centuries of the common era was
widely practised in the South and, along with blood sacrifice, included ecstatic
singing, dancing and deity-possession. When this archaic phase of worship is
described in literature, Durgā has already ceased to be a peripheral deity, as I
had assumed she still was at this time when I wrote my book, and is considered
both the consort of Śiva and Kṛṣṇa’s sister, and empress of the gods.
A historical survey of the Navarātra’s developmental pattern reveals that
it acquired its characteristic shape from a confluence of two different tradi-
tions: first, a festival of the goddess Nidrā-Kālarātri on the ninth tithi of the
dark half of a monsoon month, centred on sanguinary rites exhibiting heroism,
and second, a military tradition celebrated in autumn centred on lustrating
weapons and armies with fire to ward away ill omens. In this way, through a
gradual process of coalescence and subsequent transformation, the Navarā-
tra acquired two of its hallmarks: the central place of the Goddess as the
deity to whom all the rituals were dedicated, and the autumnal season as its
most favourable and appropriate time. The phases by which this confluence
occurred and then further developed show us how a relatively small single-
day civic festival, performed by Vaiṣṇavas as part of the celebrations of Janmā-
29 Cīlapattikāram XII, verse not indicated, p. 186; Danielou 2009, 82.

30 Cīlapattikāram, XII, verse not indicated, p. 186; Danielou 2009, 83; Mahalakshmi 2011, 68.
31 Cīlapattikāram XII, verse not indicated, pp. 187–188; Danielou 2009, 84–85.
via free access
ṣṭamī, gradually expanded into a much longer rite thought to safeguard soci-
ety and the political class, which then came to be performed in the month of
Āśvina. This development paralleled the attenuation of the rite’s importance
for the Vaiṣṇavas and its absorption, first by Śaivas, who would promote the
worship of the Goddess on Navamī by fasting, in literature about rites for lay
devotees,32 and then by the more widespread Purāṇic-Brahmanical tradition.
These sectarian absorptions provided impetus to the popularity of the rite
among rulers of upcoming kingdoms eager to cultivate the ritual apparatus of
goddess-sanctified kingship.
We can map out two broad regional traditions to have matured later in
courts: an Eastern and a Deccan one. By and large these formed the basic
blueprint for localized variations. As descriptions of Rajput rites of the śamī
tree and the weapon-shooting on Daśamī from colonial ethnographic reports
from the nineteenth century show, many of the Rajasthani royal rituals were
extremely similar to the template of the Southern Navarātra.33 The most splen-
did Navarātras seem to have flowered in the ornately ritualized Tantricized
environments of Eastern India, among which the kingdom of Mithilā provides
us with the most detailed testimonia. These appear to have percolated (in
as much as this is reflected through citations) into traditions as far afield as
the Deccan. In Mithilā the ten-day structure seems to have matured, and the
ferocious identity of the Goddess took a central place in the Tantric rites of
Mahāṣṭamī and Mahānavamī. This North-Eastern tradition developed into the
Eastern or Gauḍīya tradition—a trend evidenced by a fifteenth-century Ben-
gali work, the Durgāpūjātattva by Raghunandana, which incorporates rituals
of a Bengali character, such as summoning goddesses and worshipping them
in nine leaves from crops. A separate, even more markedly Tantric tradition,
with elements borrowed from the Kubjikā cult, developed later in the Navarā-
tra or Dasain of Nepal, but for the time being this will remain excluded from the
discussion. In the South, rituals of a Tantric character were largely eschewed.
The Navarātra was choreographed around the public display of the king, his
court, his weapons and war-animals and magnificent parades, while the wor-
ship of the Goddess, and indeed the summoning of the Goddess into the king,
occurred privately. During these days the king would worship nine virgins con-
sidered vessels of nine forms of the Goddess (different from the Navadurgās of
Mithilā and Nepal) for powers such as mastery over enemies, knowledge, riches
32 The rituals of Navamī appear in the Śivadharma, and in a parallel in an early Śaiva scrip-
ture, the Niḥśvāsamukhatattva; these are treated in greater detail in Sarkar 2017, 72–76.
33 Report of Alexander Forbes, in Kinsley 1988, 106–107.
via free access
338 sarkar
and an abundance of slaves and slave-girls, as described in ritual instructions in

Sanskrit. The most resplendent examples of this version of the rite took place,
it seems, in the Vijayanagara empire.
In this way what we find are many rituals of a large-scale, communal, pub-
lic character clustering and growing according to differing political environ-
ments (as the Navarātra was chiefly promoted by the court) around the figure of
Durgā. This process allowed a more sophisticated ritual interaction to develop
between political agents (chiefly the ruler, then the army and the polis) and the
Goddess (and her forms), who in the course of the Navarātra’s transformations
cements her role as the deity who grants the goals of kingship (military victory;
territorial protection) and protects communal areas such as fortresses, citadels
and palaces. However, in spite of its proliferation one element remained the
cornerstone of the rite: blood sacrifice. A key aspect of the nature of the ritual
as a pact between the Goddess and solicitor of rewards, this remained constant
throughout the development of the Navarātra and indeed even today is seen to
be critical to its success. It is possible that before its appearance in Vaiṣṇava
sources, the worship of Nidrā-Kālarātri in the monsoon was a widespread pop-
ular festival of heroism based on blood-sacrifice, including ecstatic communal
bacchanalia, that could have formed part of a marauder’s cult, as in the worship
of the stag-riding goddess Koṛṛavai. It was gradually absorbed into the influ-
ential sectarian traditions when goddess-cults came to be elevated during the
Gupta period, as inscriptions from Valkhā, Madhya Pradesh, in the late Gupta
period attest.34
Recently, the historian Kunal Chakrabarti (2001) has suggested that the ori-
gins of the Navarātra lie in indigenous practice, and that its late emergence in
Sanskritic literature is the culmination of a long process whereby the Goddess
was gradually brought into the Sanskritic sphere. The Goddess, Chakrabarti
argues, is a strategic means whereby peripheral and popular deities and tra-
ditions can be absorbed into the mainstream. Her festival was the time of the
year when these popular traditions could be made public and shown to cohere
around her.
Indeed this was the case, and such is also made evident in the description
in the Cīlapattikāram. The festival of the Goddess, unlike its Vedic autumnal
ancestor, integrated rites of different affiliations. Apart from the Tantric, other
rites, performed by indigenous groups, would regularly be incorporated into
the ritual sequence. In the Cīlapattikāram we are shown that, while the main
34 For a further discussion of this, see Chapter 1 of Heroic Śāktism (Sarkar 2017).
via free access
community profiting from the worship are the Eyinār hunters, a representative
of another group, the Maravars, plays a critical ritual role as the oracle. Cer-
tain social groups, for example, were authorized to carry through the animal
sacrifice. In this sense the Navarātra united disparities within the social can-
vas in which it was embedded. The obligatory performance of rites that would
involve everyone regardless of their caste represents a social inversion that the
Navarātra set into motion during the classical period. It was at this time that
the strict hierarchies enforced by the orthodox social order were overturned,
albeit for a limited period, as the single day Śābarotsava attests. This leads us
to question the long-held assumption that the Navarātra is a Purāṇic festival.
Indeed, although in outward character it was, since it was taught in texts that
were Purāṇic, and since it was further elaborated on by Sanskrit writers work-
ing to strengthen the Brahmanic order, it nevertheless was elusive in essence. I
would argue that this lack of affiliation was one of the chief reasons why it grew
into the most important ritual of political and communal affirmation. That it
was one of the few rituals of elevated, that is to say Sanskritic status, that solid-
ified the status and place of outcaste groups, and publicly displayed subversive
rites that would otherwise have been deemed suspicious by brāhmaṇas, such
as the caste-dissolving, orgiastic śābarotsava (The Festival of Śabara-tribes) on
Daśamī taught in the Kālikāpurāṇa and Dharmaśāstric literature, served to
identify it as a ritualized act of cohesion.35 The Goddess herself was a metaphor
of this cohesion, worshipped by both outcastes and people within the caste
hierarchy. Indeed, literature, particularly classical kāvya, shows that her role
as an outcaste deity preceded that of the Goddess of special importance to
a kṣatriya. The importance for kṣatriyas is emphasized in the Devīmāhātmya,
from perhaps the late-eighth century CE. Though images of the Goddess in
the presence of warriors offering their blood to her appear from as early as
the seventh century, it is in this work that we are first presented with what
became a canonized narrative of the Goddess blessing a kṣatriya king and a
vaiśya merchant, thereby being firmly associated with the power-model of the
caste system. On the other hand, that the Goddess’s worship was meant for
all varṇas and also heretics (pāṣaṇḍas), Tantric physicians (gāruḍikas) and
Buddhists, is still registered by the slightly later Devīpurāṇa, in which an ecu-
menical devotee-base, including even women, is envisaged in such verses as
35 Spring rites to Kāma were other orgiastic public celebrations of this kind discussed in
brahminical prescriptive literature, but unlike the Navarātra, which survived, they con-
tinued in adapted form within Śaiva rituals as the festival of the damana plant (damanot-
sava): see Goodall forthcoming for this argument.
via free access
340 sarkar
Devīpurāṇa 91.136 and 35.17 cd37 (on the right of women to worship and the
inherence of the Goddess in girls), 22.24 ab38 (on the worship of the Goddess by
all varṇas including śūdras) and 88.1–339 (on the Goddess’s worship by heretics,
Tantric physicians, Buddhists and those engaged in other faiths). After the four-
teenth century, and the rise of the Rajput lineages in Rajasthan, she became,
nonetheless, even more strictly connected with a specifically kṣatriya-ethos
despite being, in practice, a non-sectarian deity in the earlier classical period.40
One manner in which the Navarātra negotiated the boundary between dif-
ferent religious affiliations is through allowing optionality: nearly all manuals
36 Devīpurāṇa 91.1: brāhmaṇaḥ kṣatriyo vaiśyaḥ śūdro vā yadi vā striyaḥ | pūjayen mātaro
bhaktyā sa sarvāṃl labhatepsitān || (labhatepsitān should be understood as labhate+īpsi-
tān; the Sanskrit of this work is idiosyncratic).
37 Devīpurāṇa 35.17cd: kanyā devyā svayaṃ proktā kanyārūpā tu śūlinī |.
38 Devīpurāṇa 22.24: sarveṣu sarvavarṇeṣu tava bhaktyā prakīrtitā | kṛtvāpnoti yaśo rājyaṃ
putrāyurdhanasampadaḥ || (-sampadaḥ is the reading of Sharma’s edition, adopted by the
suggestion of S. Hatley; the Bengali edition reads -sampannaḥ. Bhaktyā is used as nomi-
native singular for bhakti).
39 Devīpurāṇa 88.1–3:
vedaiś śivāgamais tv etāḥ pūjitāś ca mumukṣubhiḥ |
gāruḍe bhūtatantre ca bālatantre ca pūjitāḥ |
sādhyante sarvakāryāṇi cintāmaṇisamāḥ śivāḥ ||
pāṣaṇḍibhir bhaviṣyais tu bauddhagāruḍavādibhiḥ |
svadharmaniratair vatsa svena nyāyena pūjitāḥ ||
yena yena hi bhāvena pūjayanti manīṣiṇaḥ |
tena tena phalaṃ dadyuḥ dvijānām antyajām api ||
Quoted from unpublished draft critical edition prepared by S. Hatley. This passage con-
cerns worship of the Seven Mothers, who are included in the worship of Durgā, even
during the Navarātra, as her attendants. A translation, citing the working draft of Hat-
ley ( forthcoming), is as follows: “People desiring liberation worship the Mothers by way of
the Vedas and the Śaiva Tantric revelation. They are also worshipped in accordance with
the Gāruḍatantras, Bhūtatantras, and Bālatantras. Beneficent, they bring all endeavors to
fruition, and are like wish-fulfilling jewels. Heretics of the future—[viz.] the Buddhist pro-
ponents of Gāruḍa Tantra—will worship them according to their own methods, devoted
to their own ways, dear child. They give rewards that accord with any disposition wise
people worship them with, whether they be Brahmins or even lowborn outcastes.” I am
grateful to Dr. Hatley for sharing his draft translation and edition with me and indicating
the need to include mention of the Devīpurāṇa.
40 For a discussion of the iconography, kāvya and narratives in Cālukya-era inscriptions por-
traying the goddess favouring a ruler, see the Introduction, Chapter 1, Chapter 3 and Chap-
ter 6 of my book Heroic Shāktism (Sarkar 2017). One of the arguments therein is that prior
to the Devīmāhātmya, the worship of the goddess seems to have been non-sectarian and
open to all rulers regardless of their caste. It is from the fourteenth century that we find
restrictions concerning who could worship the goddess and in what way in Dharmaśāstric
literature such as the Puruṣārthacintāmaṇi, also treated in my book.
via free access
include options allowing the substitution of animals with vegetables, and if one
was averse to cutting a human head, about which the Eastern manuals are quite
direct, one could just as well manage with a pumpkin. The template was fluid
and could be adapted to differing tastes and needs.
Conclusion
To sum up, I wish to take a step back and reflect on the ritual as a polit-
ical moment—which is what the Navarātra encapsulates in all its regional
forms. The political ritual concentrates divine power in the king and simulta-
neously disperses it within the body politic, thereby integrating all its aspects
within one divine body. Cycles of nature were renewed thereby, but so also
were political cycles, such as the military year. Forces of nature and the divine
that were held to be whimsical were placated, and crises—ill omens, disasters,
and calamities—that could potentially damage entire kingdoms were averted.
All this was effected within the controlled environment of the ceremony. The
charismatic heart of this ceremony was the Goddess herself: elusive because
she integrated the essences of other goddesses, and yet powerfully coherent.
Her coherence came from a representation of death, and the ceremony became
an enactment of her triumph over death. The buffalo, a vāhana of Yama, was
a symbol of death. Durgā’s slaying the buffalo symbolized both her mastery
over and her association with death and danger. In this respect, the character
of the Goddess that came alive during the Navarātra was that of a capricious
and fierce deity. If we study the most archaic layer of the buffalo-sacrifice, and
the words used in the hymns accompanying the offering of the animal’s blood,
we find an old conception of the Goddess to emerge. She is thought to stand
at the centre of an essentially cruel natural universe that could only be coaxed
into a truce through placatory worship, and through the establishment of a pact
between man and deity. That pact, if regularly and respectfully maintained dur-
ing the Navarātra, generated the goodwill of divine power.
via free access
342 sarkar
Acknowledgements
My deep gratitude to Professor Dominic Goodall who kindly went through

many versions of this article in great depth and made detailed comments. The
paper evolved as a result of his engagement with it.
References
Primary Sources
Kathāsaritsāgara of Somadeva Bhaṭṭa. Pandit Durgāprāsāda and Kāśīnātha Pāṇḍuraṅg
Parab, eds. 4th ed. revised by Vasudeva Lakṣmaṇa Śāstrī Paṇśīkar. Bombay: Nirṇaya
Sāgara Press, 1930.
Kādambarī of Bāṇa, P. Peterson, ed. Bombay: Government Central Book Depot, 1889.
Kālaviveka of Jīmūtavāhana. See Durgāpūjāviveka, pp. 30–41.
Kālikāpurāṇa. B. Shastri, ed. Varanasi: Chowkhamba Sanskrit Series. 1972.
Kṛtyakalpataru of Lakṣmīdhara. R. Aiyangar, ed. Vol. 11, Rājadharmakāṇḍa. Baroda: Bar-
oda Oriental Institute, 1943.
Kṛtyaratnākara of Caṇḍeśvara. Ṭhakkura. K.K. Smṛtitīrtha, ed. Calcutta: Asiatic Society
of Bengal, 1925.
Gaüḍavaho of Bappaī/Vākpati. S. Pandurang, ed. The “Gaüdavaho,” a Historical Poem
in Prâkṛit by Vâpati, with a Commentary of Haripâla. Bombay Sanskrit Series, no. 34,
Bombay: Government central Book depot, 1887.
Caṇḍīśataka of Bāṇa. Durgāprasāda and Kāśīnātha Pāṇḍurang Parab, eds. Kāvyamālā,
no. IV. Varanasi: Chaukhamba Bharati Academy, 1988.
Caturvargacintāmaṇi of Hemādri. Yajñeśvara Smṛtiratna and Kāmākṣyānātha Tarkav-
āgīśa, eds. Bibliotheca Indica, no. 72. Part II, vol. II. Calcutta: K.N. Bhattacharya at
the Gaṇeśa Press, 1879.
Cīlapattikāram. See Dikshitar 1939.
Daśakumāracarita of Daṇḍin. M.R. Kale, ed. Reprint, Delhi: Motilal Banarsidass, 1986.
Durgābhaktitaraṅgiṇī of Vidyāpati. Īśāna Candra Śarman, ed. Calcutta: Saṃskṛta
Sāhitya Pariṣad, 1856 Śaka Era [1932CE].
Durgāpūjātattva of Raghunandana Bhaṭṭācārya. Satīśa Candra Siddhāntabhūṣaṇa, ed.
Calcutta: Calcutta Saṃskṛta Sāhitya Pariṣad: Bengali Saṃvat 1331 [1922CE].
Durgāpūjāviveka. Satīśa Candra Siddhāntabhūṣaṇa, ed. Saṃskṛta Sāhitya Pariṣad
Series, no. 7. CalcuttaSaṃskṛta Sāhitya Pariṣad, 1331 Bengali Saṃvat [1922CE].
Durgotsavaviveka of Śūlapāṇi. See Durgāpūjāviveka, pp. 1–27.
Devīpurāṇa. P. Tarkaratna and S. Nyāyatīrtha, eds. Reprint, Navabhārata Publishers: Cal-
cutta, Bengali Saṃvat 1400 [1991CE].
Devīmāhātmya. Draft electronic edition of Yuko Yukochi based on the following three
via free access
manuscripts: NAK Nr. 1–1077= NGMPP Nr. A 1157/; NAK Nr. 1–1534 = NGMPP Nr.
A 1157/12; palm-leaf manuscript in the possession of Sam Fogg London.
Puruṣārthacintāmaṇi of Viṣṇubhaṭṭa Āṭhavaḷe. Vyāsa Miśra, ed. Varanasi: Sampūrṇā-
nanda Saṃskṛta Viśvavidyālaya, 2006.
Mahābhārata. Vishnu S. Sukthankar et al., eds. Bhandarkar Oriental Research Institute:
Poona, 1933–1959.
Vāmanapurāṇa. A.S. Gupta, ed. Varanasi: All India Kashiraj Trust, 1968.
Viṣṇudharmottaramahāpurāṇa. Kṣemarāja Śrīkṛṣṇadāsa, ed. Venkateshvara Steam
Press. Reprint, Bombay: Nag Publishers, 1985.
Skandapurāṇa. Yuko Yokochi, ed. Skandapurāṇa, Vol. III: Adhyāyas 34.1–61, 53–69. The
Vindhyavāsinī Cycle. Leiden: Brill, 2013.
Skandapurāṇa. Kṛṣṇaprasāda Bhaṭṭarāī, ed. Skandapurāṇasya Ambikākhaṇḍa. Kath-
mandu, 1988.
Haravijaya of Ratnākara. Pandit Durgaprasad and Kashinath Pandurang Parab, eds. The
Haravijaya of Râjânaka Ratnâkara with the Commentary of Râjânaka Alaka. Kāvya-
mālā, no. 22. Bombay: Nirṇayasāgara Press, 1890, Reprint 1982.
Harivaṃśa. Parashuram Lakshman Vaidya et al., eds. 2 vols. Poona: Bhandarkar Orien-
tal Institute, 1969–1971.
Harṣacarita of Bāṇa. P.V. Kane, ed. The Harshacarita of Bânabhatta: Text of Ucchvāsas
I–VIII. Reprint, Delhi, 1986 [1918].
Secondary Sources
Baldissera, Fabrizzia. 1996. “Caṇḍikā/Caṇḍī Vindhyavāsinī and Other Terrific God-
desses in the Kathāsaritsāgara.” In Wild Goddesses in India and Nepal, Proceedings of
an International Symposium in Berne and Zürich 1994, edited by A. Michaels, C. Vogel-
sanger, and A. Wilke, 73–103. Studia Religiosa Helvetica. Berne and Zürich: Peter
Lang.
Chakrabarti, Kunal. 2001. Religious Process: The Purāṇas and the Making of a Regional
Tradition. New Delhi: Oxford University Press.
Couture, A. and Schmid, C. 2001. “The Harivaṃśa, the Goddess Ekānaṃśā, and the
Iconography of the Vṛṣṇi Triads.” Journal of the American Oriental Society 121 (2): 173–
192.
Danielou, A. (trans.). 2009 Shilapaddikaram (The Ankle Bracelet) by Prince Ilango Adi-
gal. New Delhi: Penguin Books.
Dikshitar, V.R.R. (trans.). 1939 The Śilapaddikāram. Madras: Humphrey Milford Oxford
University Press.
Einoo, Shingo. 1999. “The Autumn Goddess Festival described in the Purāṇas.” In Liv-
ing with Śakti: Gender, Sexuality and Religion in South Asia, edited by M. Tanaka and
M. Tachikawa, 33–70. Tokyo: National Museum of Ethnology, Tokyo.
Goodall, Dominic, N. Rout, et al. 2005. The Pañcāvaraṇastava of Aghoraśiva: A Twelfth
via free access
344 sarkar
Century South Indian Prescription for the Visualisation of Sadāśiva and his Retinue.
Pondicherry: Institut Français de Pondichéry.
Goodall, Dominic. 2014. “Communication: Des saisons dans la poésie Sanskrite du
Cambodge.” Comptes Rendus de l’Académie des Inscriptions I (January–March): 175–
188.
Goodall, Dominic. Forthcoming. “Damanotsava: on Love in Spring, on What Jñāna-
śambhu Wrote, and on the Spread of Public Festivals in the Mantramārga.” In Tantric
Communities: Sacred Secrets and Public Rituals, edited by Vincent Eltschinger, Nina
Mirnig, and Marion Rastelli, 385–424. Vienna: Austrian Academy of Sciences Press.
Hatley, Shaman. Forthcoming. “The Seven Mothers (Sapta Mātaraḥ): Origin Tales from
the Old Skandapurāṇa and Devīpurāṇa.” In A Garland of Goddesses: Hindu Tales of
the Divine Feminine from India and Beyond, edited by Michael Slouber. Berkeley: Uni-
versity of California Press.
Kinsley, David. 1988. Hindu Goddesses: Visions of the Divine Feminine in the Hindu Reli-
gious Tradition. Berkeley: University of California Press.
Mahalakshmi R. 2011. The Making of the Goddess: Koṛṛavai Durgā in Tamil Traditions,
New Delhi: Penguin Books.
Nicholas, R. 2013. Night of the Gods: Durga Puja and the Legitimation of Power in Rural
Bengal. Delhi: Orient Blackswan.
Price, Pamela. 1996. Kingship and Political Practice in Colonial India. Cambridge: Cam-
bridge University Press.
Sanderson, Alexis. 2007. “Atharvavedins in Tantric Territory: The Āngirasakalpa Texts of
the Oriya Paippalādins and their Connection with the Trika and the Kālīkula, with
critical editions of the Parājapavidhi, the Parāmantravidhi, and the *Bhadrakālī-
mantravidhiprakaraṇa.” In The Atharvaveda and its Paippalāda Śākhā: Historical
and Philological Papers on a Vedic Tradition, edited by Arlo Griffiths and Annette
Schmiedchen, 195–311. Geisteskultur Indiens: Texte und Studien 11, Indologica
Halensis. Aachen: Shaker Verlag.
Sanderson, Alexis. 2009. “The Śaiva Age—The Rise and Dominance of Śaivism Dur-
ing the Early Mediaeval Period.” In Genesis and Development of Tantricism, edited
by Shingo Einoo, 41–348. Tokyo: Institute of Oriental Culture, University of Tokyo.
Sarkar, Bihani. 2012. “The Rite of Durgā in Medieval Bengal: An Introductory Study of
Raghunandana’s Durgāpūjātattva with Text and Translation of the Principal Rites.”
Journal of the Royal Asiatic Society 22 (02): 325–390.
Sarkar, Bihani. 2017. Heroic Shāktism: the Cult of Durgā in Ancient Indian Kingship.
Oxford: Oxford University Press.
Schmid, C. 2002. “Mahiṣāsuramardinī: A Vaiṣṇava Goddess?” In Foundations of Indian
Art: Proceedings of the Chidambaram Seminar on Art and Religion, Feb. 2001, edited
by R. Nagaswamy, 143–168. Chennai: Tamil Arts Academy.
Stein, Burton. 1983. “Mahānavamī: Medieval and Modern Kingly Ritual in South India.”
via free access
In Essays on Gupta Culture, edited by Bardwell L. Smith, 67–90. Delhi: Motilal Banar-
sidass.
Yokochi, Yuko. 2004. The Rise of the Warrior-Goddess in Ancient India: A Study of the
myth of Kauśikī-Vindhyavāsinī in the Skanda-Purāṇa. Ph.D. Thesis, University of
Groningen.
via free access
via free access
part 4
Mantra, Ritual, and Yoga
via free access
via free access
chapter 14
Śārikā’s Mantra
Jürgen Hanneder
The goddess Śārikā belongs to a group of Kashmirian lineage deities (kuladevī

or vaṃśadevī) who, like Bālā, Rājñī, and Jvālā, are identified with particular
locations.1 Śārikā resides on the Pradyumna peak in Śrīnagar, also known as the
“Śārikā peak,” and is worshipped there in the form of a large stone around which
a temple has been constructed. If one wishes to identify the cult practiced there
and locate it within the religious landscape of the valley, a visitor might start
with the modern inscription shown in figure 14.1, which reads as follows:
bindutrikoṇavasukoṇadaśārayugma-
manvaśranāgadalasaṃyutaṣoḍaśāram |
vṛttatrayaṃ ca dharaṇīsadanatrayaṃ ca
śrīcakrarājam uditaṃ paradevatāyāḥ ||
This verse describes the śrīcakra, the yantra of the “supreme deity” (parade-
vatā) commonly known as Tripurā, by merely listing in the first three pādas the
geometrical elements of the yantra, for instance bindu (“dot”) and trikoṇa (“tri-
angle”). It is quoted often, with variations,2 and attributed to several scriptural
sources in Tantric literature.
What we are to understand here, one must suppose, is that in modern times
Śārikā was understood or presented as a form of Tripurā, or as belonging to her
cultic context.
Unfortunately fieldwork of this kind does not disclose the history of the cult.
We know that the first scriptures of the cult of Tripurā were written not before
the eleventh century;3 needless to say, they make no mention of Śārikā. On the
other hand, the local Kashmirian cult of Śārikā is at least as old as the Kathā-
saritsāgara (second half of the twelfth century),4 and it is rather doubtful that
the two goddesses’ association is this old.
1 See Sanderson 2009, 111.

2 For instance, Lakṣmīdhara in his commentary on the Saundaryalaharī reads the third pāda
as vṛttatribhūpurayutaṃ paritaś caturdvāḥ and in the fourth śrīcakram etad […].
3 Sanderson 2015, 32.
4 See 12.6.111: yena pradyumnaśikharaṃ śārikākūṭam ity api […], and 12.6.116: tatra snātvā vita-
stāyām arcayitvā vināyakam saṃpūjya śārikāṃ devīṃ digbandhādipuraḥsaram.
© jürgen hanneder, 2020 | doi:10.1163/9789004432802_016

via free access
350 hanneder
figure 14.1 Śārikā stone

Photograph by Walter Slaje
For finding out more about the cult of Śārikā, including her ritual and pos-
sibly doctrinal details, we have a single published source, the Devīrahasya,5
which contains passages on the cult of the Kashmirian lineage deities. In par-
ticular, the ritual texts appended to the edition of the Devīrahasya give the
mantroddhāra, sahasranāma and other typical elements of the worship of
these deities. In this text Śārikā is clearly identified as having the form of a stone
(śilārūpā)6 on the Pradyumna hill. Her worship is said to remove the impurity
(mala) stemming from such capital offences as the murder of a Brahmin or
drinking alcohol, or eating what is forbidden.7 The details of the mantra are
given, as are the Ṛṣi, etc., a dhyānaśloka, her yantra, and how to employ the
mantra for the magic acts of immobilizing (stambhana) and so forth.8 This is
followed by a Śārikāpūjāpaddhati (pp. 412–419).
5 Devīrahasya, pp. 407 ff.

6 Op. cit., p. 407.
7 Op. cit., p. 408.
8 Op. cit., pp. 410 ff.
via free access
śārikā’s mantra 351
Further sources are difficult to trace or remain unpublished. First is a so-

called Śārikāstrotra attributed to the Pradyumnāvatāra. This text is “a mālā-
mantra in prose with a dhyānaśloka in the beginning and two verses at the
end.”9 Apart from its manuscript in the Bodleian, there are two more in the
Lindenmuseum (Stuttgart),10 and one in Śrīnagar.11 Judging from the excerpts
given in the Verzeichnis der orientalischen Handschriften in Deutschland
(VOHD) catalogue, there are variations such as insertions of new names and
additional phrases that cannot be reduced easily to a single critical text.12 In
any case, for our purposes a transcript of the Śrīnagar manuscript (S) is suffi-
cient:
bījaiḥ saptabhir ujjvalākṛtir asau yā saptasaptidyutiḥ13

saptarṣipraṇatāṅghripaṅkaja14yugā yā saptalokārtihā15 |
kāśmīrapravareśamadhyanagare pradyumnapīṭhe sthitā
devīsaptakasaṃyutā bhagavatī śrīśārikā pātu naḥ ||
oṃ jaya bhagavatyai vindhyavāsini kailāsavāsini śmaśānavāsini huṅkāriṇi kā-

lāyini kātyāyani himagiritanaye kumāramātaḥ govindabhagini śitikaṇṭha16
bharaṇe aṣṭādaśabhuje bhujagavalayamaṇḍite keyūrahārābharaṇe jaya-kha-
ḍgā-triśūla-ḍamaru17-mudgara-paraśu-caṣaka-kalaśa-śaracāpa-varadābhaya-
pāśa-pustaka-kapāla-khaṭvāṅga-gadā-musula-tomaravarahaste kṛpāpare pra-
bhūtavividhāyuddhe kṛpāprāgbhūtavigrahe caṇḍike caṇḍaghaṇṭe kirātaveśe
brahmāṇi rudrāṇi nārāyaṇi brahmacāriṇi divyatapovidhāyini vedamātaḥ gāya-
tri bhāvitri sarasvati sarvādhāre sarveśvari viśvakartā (sic) samādhiviśrāntimaye
cinmaye cintāmaṇisvarūpe kaivalye kaivalyasvarūpe śivasvarūpe śive nirāśraye
nirupādhimaye nirāmayapade brahmaviṣṇumaheśvaranimitte18 mohani tosaṇi
9 Aithal 1999, 35. Aithal gives a Kashmirian edition (Bhaktivivekasāra, Śrīnagar/Bombay:

1927), but this work has not made it into a catalogued and accessible library.
10 Verzeichnis der orientalischen Handschriften in Deutschland 2.2.711 and 712.
11 Oriental Research Library, University Campus Hazaratbal, Srinagar, National Mission for
Mss. DSO 0000/5517.
12 There are other sources that utilize the same material, but with variations: the Siddha-
lakṣmīpūjāpaddhati—I have merely access to the transcription of ms. “Kashmir Research
Center accession no: 2376” by the Muktabodha Institute—has for instance a completely
different third pāda of the starting verse. The Agnikāryapaddhati also transmits this verse.
13 saptasapti] em. The ms. reads saptasapta as does the one in the Lindenmuseum, Stuttgart.
14 paṅkaja] em.; paṅka ms.
15 ārtihā] S; āśrayā VOHD.
16 śiti] em.; śati ms.
17 ḍamaru] em.; ḍumara ms.
18 nimitte] em.; nimite ms.
via free access
352 hanneder
bhayaṅkaranāśini ditisutapramathani kāle kālakiṅkarabhaṣini kālāgniśikhe kā-

larātri aje nitye siṃharathe yogarate yogeśvari19nimite (?) bhaktajanavatsale
surapriyakāriṇi durge durjaye hiraṇye śaraṇye kuru me dayāṃ kuru me jayam |
oṃ pradyumnaśikharāsīnāṃ mātṛcakropaśobhitām |
pīṭheśvarīṃ śilārūpāṃ śārikāṃ praṇamāmy aham ||
amā caiva u kāmā ca cārvāṅgī ṭaṅkadhāriṇī |
tārā ca pārvatī caiva yakiṇī śārikāṣṭamī ||
iti śrīśārikāstotram
From this stotra we can gather something about the iconography of the deity,
for instance the attributes held in her eighteen arms,20 her names, the seven
goddesses (amā etc.) that form her retinue and their symbolism.
The main information on a deity for ritual purposes is of course her mantra.
Here one unpublished source21 gives the mantra in a Vedic style by stating its
Ṛṣi, metre, and deity, but augmented with the tantric elements bīja, śakti and
kīlaka,22 and there is of course also a tantric Gāyatrī devoted to Śārikā.23
So it seems the cult of Śārikā is fairly old, but it has been influenced by the
cult of Tripurā or Śrīvidyā. One such influence must have been the Kashmirian
Kauls, a clan that migrated to Kashmir not before the fifteenth century24 and
brought their own cults with them, which were then fused with the local Kash-
mirian cults.25 The most important figure in this group, Sāhib Kaul, is credited
with three ritual handbooks and furthermore fused Advaita Vedāntic ideas with
the Kashmirian Pratyabhijñā.26 Sāhib Kaul has also composed a further text on
the deity, a Śārikāstotra27 in eighteen verses which, as he says, gives the deriva-
tion of her Mantra.28
19 yogeśvari] em.; yogaśvari- ms.

20 We have to count the bow and arrow as one item held in the same hand.
21 Agnikāryapaddhati. Manuscript no. 781, Research Library of the Jammu and Kashmir Gov-
ernment, Shrinagar. Transcription of the Muktabodha Indological Research Institute.
22 asya śrīśārikāmantrasya | śrīmahādeva ṛṣiḥ | triṣṭupchandaḥ || śrīśārikābhagavatī devatā ||
śāṃ bījaṃ || āṃ śaktiḥ hrāṃ kīlakam || home viniyogaḥ ||
23 oṃ hrīṃ śārikāyai vidmahe | saptākṣaryai dhīmahi || tan naḥ śilā pracodayāt ||3||
24 Sanderson 2003–2004, 362–366.
25 Sanderson 2009, 124–125.
26 See Hanneder 2001.
27 Not to be confused with the Śārikāstotra given above.
28 There are further stotras and other minor works of Sāhib Kaul, an edition of which is under
preparation by the present author.
via free access
1 Sāhib Kaul’s Śārikāstotra
vande devīṃ śārikāṃ mokṣadātrīṃ

sarvasthāṃ tāṃ sarvato mohadātrīṃ |
mithyāmṛtyuprāptidurbhītidātrīṃ
sadbhaktyāhaṃ mūrdhni candrāṃśadhātrīm ||1||
Mss.: O3 (CSS e.264, ff. 529r–531v) L1 (SOAS 44389, ff.1–5) B1 (Berlin Hs. or. 12509). 1b moha]
O3B1L1pc; mokṣa L1ac 1d mūrdhni] O3 B1; mūrti L1 1d dhātrīṃ] L1B1; dātrīṃ O3
With true devotion I worship that divine and omnipresent Śārikā, who
bears the crescent moon on her head, who grants liberation, destroys
delusion everywhere, destroys the bad fear of meeting a wrong death.
Since we have to suppose a Kashmirian pronounciation, which does not prop-

erly distinguish aspirated from non-aspirated stops, all four lines are real-
ized as an end rhyme. In this verse the goddess is adored as granting liber-
ation, removing29 confusion etc. One iconographical detail known from the
other sources, that she bears the crescent moon on her head, is alluded to as
well.
After the introductory stanza the main topic of the stotra, the mantroddhāra
explicitly mentioned in v. 17, commences:
tāraṃ bījaṃ yo japed amba bhaktyā

saṃsārābdhes tārakaṃ śārike te |
brahmajñānaprauḍhayā prajñayāyaṃ
vācāṃ nāthasyāpi kuryād vihāsam ||2||
2b ābdhes] L1; ābdhaṃ O3
O mother Śārikā, whoever devotedly recites your tāra-syllable, which car-

ries one across (tāraka) the ocean of transmigration, may, when his wis-
dom is ripened through the knowledge of the absolute, even put to shame
the Lord of the Word (bṛhaspati).
“Your bīja” means the (first) syllable of the mantra of Śārikā. Sāhib Kaul uses
code words for the syllables that make up the mantra, a common practice
29 In verse 1a the root dāṇ dāne (Dhātupāṭha 1.977) is used, while in 1b and 1c the root is dāp
lavane (Dhātupāṭha 2.50).
via free access
354 hanneder
which enables a writer to—as it were—speak of a mantra without actually pro-

nouncing it. Thus, to those uninitiated and ignorant of the codes, the true form
of the mantra must have, at least in theory, remained unknown.
tāra is a common code word for the syllable oṃ and poses no problems. The
codes that follow are partly difficult to interpret and will be discussed below.
īśaṃ sābjaṃ vahnisaṃsthaṃ sapadmaṃ

bījaṃ te ’nyad yaḥ smaret taṃ smaranti |
nāke devyo bhūtale nāgakanyā
bhūmau nāryo vihvalā mārabāṇaiḥ ||3||
3c kanyā] O3; patnyo L1B1
He who remembers your next syllable, which is īśa with abja, vahni, and
padma, is remembered by goddesses in heaven, Nāga maidens in the
netherworld, and women on earth confused by the arrows of Kāma.
lakṣmībījaṃ durlabhaṃ durjanānāṃ

samyagbhaktyā yo japec chuddhabuddhiḥ |
padmā nityaṃ darśanaṃ vīpsur asya
dvāre tiṣṭhaty ādarāc cañcalāpi ||4||
4d ādarāc] L1 B1; ādarā O3
One of pure mind who recites with complete devotion the lakṣmī-syllable,
which is difficult for bad people to obtain, him the goddess of good for-
tune will always be eager to see, and although unsteady (by nature) she
will remain at his doorstep out of devotion.
īśaṃ sābjaṃ vāmakarṇordhvasaṃsthaṃ

bījaṃ te ’nyad yo japet tasya śatruḥ |
sarvair devair apy ajayyaḥ kṣaṇena
dṛṣṭeḥ pātād yāmagehātithiḥ syāt ||5||
5a karṇordhva] L1 B1; karṇardhva O3 5d dṛṣṭeḥ] B1; dṛṣṭoḥ O3L1 5d ātithiḥ] L1B1; ātithi O3
He who recites your next syllable, which is īśa with abja and the one
above the left ear, his enemy, although invincible even for all the gods, will
instantly, in the wink of an eye, become a guest in the house of Death.
via free access
īśaṃ bījaṃ vaktravṛttena yuktaṃ

sābjaṃ yo ’nyat saṃsmared vahnisaṃstham |
bhuktir muktiḥ sadvicārasya yuktir
bhaktiś caitaddhastayātā bhavanti ||6||
6b saṃsmared] O3; saṃjaped L1B1 6d bhaktiścai] L1; bhakticai O3
He who remembers your next syllable, which is īśa together with vak-
travṛtta and vahni, will have at his disposal “enjoyment”30 (bhukti), lib-
eration, the method of real vicāra,31 and devotion.
sābjaṃ bījaṃ vaktravṛttaṃ japed yaḥ

samyagbhaktyā śuddhahṛc chārike te |
vāṇī nānāsadrasair jṛṃbhitaśrīr
nityaṃ vaktre tiṣṭhati prauḍham asya ||7||
7a sābjaṃ] O3; īśaṃ L1 B1
He who recites your syllable with pure heart and proper devotion, O
Śārikā, which consists of abja and vaktravṛtta, in his mouth a fully devel-
oped32 voice stays, which has the beauty of unfolding through various
good emotions.
sābjaṃ bījaṃ vaktravṛttena yuktaṃ

asthyātmākhyaṃ yo japec chārike te |
jīvanmuktaś ceha bhuktvātibhogāl
līnaḥ paścāt tvatpade syād bhavāni ||8||
8c bhuktvā] L1 O3; bhuktā B1
He who recites your syllable, consisting of abja and vaktravṛtta, and called
asthyātmā, O Śārikā, is liberated in life and, enjoying supreme bhogas,33
will later dissolve in your state, O Bhavānī.
30 In context, this may mean the enjoyment of powers (siddhi).

31 I understand the term in the sense of ātmavicāra as used in the Mokṣopāya (5.5.27), a text
which Sāhib Kaul occasionally refers to.
32 prauḍham is of course an adverb, but to translate “stays ripely” would not really capture
the intended sense.
33 This could again denote the siddhis, but in a less technical register the enjoyment of the
pleasures of liberation.
via free access
356 hanneder
etadbījaprāntagāṃ śārikāyā
ityākhyāṃ te saṃjaped yo namontāṃ |
tat te dhāma prāpyate tena śaśvad
gatvā bhūyaḥ śocyate naiva yatra ||9||
9b ityā] L1 O3 B1; yityā O3
He who recites after that syllable your name, Śārikā, followed by namaḥ,
attains forever to that abode where, when reached, one never suffers
again.
saṃstaumi tvāṃ tvām imām āśraye ’haṃ

seve devīm eva sarvaikaśaktim |
tvām atyuccair gadgadaṃ saṃbravīmi
sarvāṃ sārvāṃ sarvato bhāvayāmi ||10||
10a tvāṃ tvām imām ] L1 B1; tvā tvāmim O3 10b seve] L1 B1; seved O3 10c tvām atyuccair ]
L1 B1; tvāṃ mṛtyuccair O3 10d sarvāṃ sārvāṃ] L1; sarvāṃ sarvāṃ O3 B1
I praise you; it is you in whom I take refuge. I serve the Goddess alone, the
one power of all (powers). I utter my noisy stammering to you; I contem-
plate (you) who are everything, suitable for all (sārva), and everywhere.
jātaḥ so’ sau satkule tasya dhanyā

mātā tasmin saṅgatā āśiṣaś ca |
jñātaṃ sarvaṃ tena śambhupriyāyāḥ
samyakprajñā nirmitā yena bhaktyā ||11||
11a satkule] L1 O3; satkulye B1 11d prajñā] O3; prajā L1 B1
He is born in a good family, his mother is blessed, and he receives good

wishes. He knows everything about [Śārikā,] the beloved of Śiva, who has
fathomed true knowledge through devotion.
bhaktiḥ puṣṇāty anvahaṃ māṃ tvadīyā

nityaṃ yadvat pūrṇacandrodayo ’bdhim |
tvadbhakteḥ satsaṃpadā prāptayāhaṃ
jiṣṇor lakṣmīṃ cāpy upekṣe purogām ||12||
12a puṣṇātya] L1 B1; puṣṇātyu O3 12b candrodayo’bdhiṃ] L1 B1; candrodayābdhiṃ O3
via free access
My devotion to you nourishes me every day, as the rise of the full moon
always nourishes the ocean. On account of the true affluence of victorious
devotion to you I even ignore the excellent Lakṣmī.
etat sarvaṃ tvanmayaṃ devadevi

tvaṃ ciddehā kevalā suprasiddhā |
nāsty ajñānaṃ kvāpi tasmāt kva dṛṣṭo
vandhyāputraś cāpam āropya dhāvan ||13||
13a etat L1 B1; et O3 13c nāstya L1 B1; nāsti O3 13d cāpam āro L1; apasāro B1; cāpimāro O3
The whole world (etat sarvaṃ) consists of you, Goddess of Gods! Your
body is consciousness, you are alone and perfectly established. Nowhere
is there ignorance. Thus, where do we see the son of a barren woman run
and raise his bow?
āśāste me devabhāvaṃ na cittaṃ

sūteḥ kāle strīva bhogaṃ kadācit |
tvatsadbhaktyā nṛtyati prāptayālam
gāyañ chrutvā meghanādaṃ śikhīva ||14||
14a cittaṃ L1 B1; citta O3
My mind does not strive after the divine state, just as a woman giving birth
never craves enjoyment. Having gained perfect (alam) devotion to you it
sings like a peacock who has heard the sound of the rain clouds.
sā kā bhūmir yatra nāsti sthitis te

sā kā vāṇī nocyase vā yayā tvaṃ |
ko ’sau śabdaḥ śrūyase yatra na tvaṃ
ko ’sau bhāvo yatra te bhāsanaṃ no ||15||
15a bhūmiḥ] L1; bhūmi O3 B1
There is no place where you do not reside; there is no voice in which you
are not expressed. There is no word in which you are not heard; there is
no thing in which you do not shine.
tvatsadbhaktyarkodayāt saṃpraphullaṃ
hṛtpadmaṃ me ’tyadbhutāt sadraseddham |
via free access
358 hanneder
nityaṃ yasmin sarvadā kāṅkṣaṇīyā

sarvais tiṣṭhaty ādarān mokṣalakṣmīḥ ||16||
16a tvatsat] conj.; tvatpad L1 B1; unclear in O3 due to damaged folio
When the marvelous sun of true devotion to you rises, the lotus of my
heart is inflamed through true emotion (rasa). In it always resides, out of
respect, the good fortune of liberation that is coveted by all.
jñānasvāmiprāptasadbuddhisāro
jñātajñeyaḥ sarvataḥ svātmabhāvī |
stotraṃ mantroddhāry adaḥ śārikāyāḥ
sāhibkaulo vaṃśadevyāś cakāra ||17||
17a sāraḥ] conj.; sāra L1 O3 B1 17c mantroddhāry adaḥ] L1 B1; mantrodāry adaḥ O3 17d
devyāś cakāra O3B1] devāś cakāra L1
Having attained the strength of true intelligence through Jñānasvāmin, I

know what there is to know and everywhere contemplate my own self. I,
Sāhib Kaula, have composed this hymn to the lineage deity Śārikā, which
contains the construction of her Mantra.
Sāhib Kaul’s Śārikāstava deals with some aspects of the worship of this deity;
most importantly, as the author states in verse 17, it gives the mantroddhāra of
the Śārikāmantra. The author’s teacher Jñānasvāmin, according to Madhusu-
dan Kaul,34 was his maternal uncle.
yo vāpy etaṃ kīrtayet stotram āḍhyaṃ

samyagbhaktyā śroṣyati śrāvayed vā |
nirmantro ’pi prāpnuyād devadevi
niḥsandehaṃ mantrajaṃ satphalaṃ saḥ ||18||
18a vāpyetaṃ] L1; vyāpyetat O3B1 18a kīrtayet] L1; kīrtaye O3 18a āḍhyaṃ] L1; mādyaṃ
O3 18b śroṣyati] conj.; śrośyati L1; śriṣyati O3 18b vā] L1; vāvā O3
Whoever chants this rich hymn of praise with perfect devotion, hears it or
has it recited, even if he be without mantra, he will, O supreme Goddess,
without doubt reap the great fruit of this mantra.
34 See above.
via free access
This final stanza explains the idea behind this work. A person who is not ini-
tiated into the recitation of the mantra of Śārikā and may not even know how
to decode the mantroddhāra can still benefit from this type of substitute recita-
tion. In this the work is similar in approach to the Sūryastutirahasya of Sāhib
Kaul’s contemporary Ratnakaṇṭha, where the Vedic Gāyatrī-mantra is hidden
within a hymn addressed to the sun as an acrostichon.35
The aim in both cases is apparently to enable persons who lack proper
adhikāra—for Vedic mantras, in the case of Ratnakaṇṭha, or Tantric mantras, in
the case of Sāhib Kaul—to gain at least some kind of access to these restricted
parts of the religion. This technique of “hiding” the actual form of the mantras
in a stotra meant for religious recitation is not so much a way to conceal it from
the outsider, but a method to enable him or her to use it without breaking reli-
gious rules, in other words a method to bypass religious and social restrictions.
Theologically the matter is of course complicated, because Ratnakaṇṭha’s sto-
tra actually contains the sounds that make up the Gāyatrī, so in a sense by
reciting the stotra one does recite the Gāyatrī. In the case of the Śārikāstava,
since only code names are given, one does not utter the sounds that make up
the mantra of Śārikā.
Despite the fact that the stotra contains the mantroddhāra, it is quite diffi-
cult to decipher the mantra from Sāhib Kaul’s stotra alone, for the system of
codes is not otherwise known. The most obvious place to search for a solution
would be the mantroddhāra in the Devīrahasya, for this text deals in detail with
the Kashmirian lineage goddesses, and according to Aithal,36 the Śrīvidyānitya-
pūjāpaddhati of Sāhib Kaul as available in the ms. Chandra Shum Shere c. 264
is roughly identical with the ritual manuals printed in the appendix to the Devī-
rahasya.37 This is the relevant verse:38
tāraṃ parā-mā-taṭa-sindhurārṇāḥ
khaṃ śarma tanmadhyagataṃ ca nāma |
ante ’śmarī pārvati śārikāyās
trayodaśārṇo manur asti gopyaḥ ||
35 See Stanislav Jager in Hanneder, Jager and Sanderson, 2012, 23.

36 Aithal, 1999, 35.
37 Judging from the excerpts a similar ms. is described in Catalogue of the Sanskrit Manu-
scripts in the Library of the India Office. Part IV. A.VIII Tantra. See Windisch, 1894, 861–
862.
38 Devīrahasya, p. 407.
via free access
360 hanneder
table 14.1 The bījas as defined in the Śārikāstotra
Verse bīja Explanation
2 oṃ tāraṃ
3 hrīṃ īśaṃ sābjaṃ vahnisaṃsthaṃ sapadmaṃ
4 śrīṃ lakṣmībijaṃ
5 hūṃ īśaṃ sābjaṃ vāmakarṇordhvasaṃsthaṃ
6 phrāṃ īśaṃ vaktravṛttena yuktaṃ sābjaṃ
7 āṃ sābjaṃ vaktravṛttaṃ
8 śāṃ sābjaṃ vaktravṛttena yuktaṃ asthyātmākhyaṃ
9 etadbījaprāntagāṃ śārikāyā ityākhyānte saṃjaped yo
namo’ntāṃ
The code words—parā for hrīṃ, aśmarī for namaḥ etc.—are explained
in the edition of the Devīrahasya.39 Furthermore, in its second chapter the
mantras are given also in plain language (spaṣṭam), as is the mantra of Śārikā:
oṃ hrīṃ śrīṃ hūṃ phrāṃ āṃ śāṃ śārikāyai namaḥ.40 We might thus conclude
that all is well, and that since the wording of the mantra itself is not (and, the-
ologically speaking, should not) be in doubt, we have a good chance to under-
stand Sāhib Kaul’s mantroddhāra. Since these sets of codes cannot easily be
corrupted in the course of transmission, no banal error should have crept in.41
However, if we look at the definitions in Sāhib Kaul’s Śārikāstava, we find
that not all can be brought into accord with this form of the mantra. Table 14.1
shows the expected bījas and their definitions in the verses. We may leave out
verses 2, 4 and 9, because they do not use complicated codes: tāra usually
means “oṃ,” lakṣmī stands for “śrī,” and the conclusion of the mantra (dative
of the deity and namaḥ) is as expected.
The problem arises when we try to decode the remainder as follows: (1) the
element present in all definitions is abja and there is only one sound common
to all five bījas, that is, ṃ. Apart from that nos. 6–8 have only one element in
common, so vaktravṛtta stands for ā. Taking now elements that occur only once,
it follows that in 8 asthy- must signify ś, vāmakarṇa- in 5 means ū, and vahni
39 Op. cit., pp. 19–21.

40 Op. cit., p. 13.
41 There is, of course, the counter-evidence of one manuscript of a Śārikāstavarāja (VOHD
2.2.116), which gives the mantra with the misspelling or wrong transcription of hrūṃ for
hūṃ.
via free access
as expected stands for r. Now only īśa remains, which stands for h in 3 and 5,
but—in view of our mantra—should stand for ph in 6.
Let us now look at another source. The mantra is also given in the Dakṣiṇa-
mūrtyuddhārakośa, where we find the following definition:42
tāraṃ māyāṃ śriyaṃ kūrcaṃ sindhuraṃ śūnyam eva ca |

kalyāṇaṃ śārikādevyā bījaṁ saptākṣaraṃ smṛtam ||
According to the index of code words in the appendix and the “prakāśam” ver-
sion (ibid., 16) this translates into the seven bījas “oṃ hrīṃ śrīṃ hūṃ hrāṃ āṃ
śāṃ.” The manuscripts reported have three other options for hrāṃ, one being
phrāṃ, our reading from the Devīrahasya.
Now, in the index of the edition of the Dakṣiṇamūrtyuddhārakośa the code
word sindhura is given as hrāṃ, while according to the bījākṣarapāribhāṣikasūcī
contained in the Devīrahasya (p. 21) it means phrāṃ. In other words we find two
versions of the mantra in the available sources, unless the editors have misread
their manuscripts. What we can say is that the evidence from Sāhib Kaul con-
firms the version of the Dakṣiṇamūrtyuddhārakośa.
Abbreviations
conj. conjecture
em. emendation
ms(s.) manuscript(s)
VOHD Verzeichnis der orientalischen Handschriften in Deutschland
References
Primary Sources
Agnikāryapaddhati
Manuscript no. 781, Research Library of the Jammu and Kashmir Government, Shrina-
gar. Transcription of the Muktabodha Indological Research Institute.
Uddhārakośa of Dakṣiṇāmūrti
Raghu Vīra and Shodo Taki, eds. Śrī-Dakṣiṇāmūrti-viracita Uddhārakośaḥ. Reprint, New
Delhi: Mushiram Manoharlal, 1978 [1938].
42 Dakṣiṇāmūrti’s Uddhārakośa, p. 15. The first line up to sindhura is identical with another
mantroddhāra in Devīrahasya, p. 13.
via free access
362 hanneder
Kathāsaritsāgara
Pandit Durgāprasād, ed. The Kathāsaritsāgara of Somadevabhaṭṭa. Bombay: Tukārām
Jāvajī, 1903.
Devīrahasya
Ramchandra Kak and Harabhatta Shastri, eds. Śrīdevīrahasyam. Vrajajivan Prachyab-
harati Granthamala, no. 33. Reprint, New Delhi: Chaukhamba Sanskrit Pratishthan
1993 [Srinagar: Vishinath & Sons, 1941].
Dhātupāṭha
In Otto Böhtlingk, ed. Pâṇini’s Grammatik, pp. 61–145 (appendix). Leipzig: H. Haessel
1887.
Mokṣopāya
Krause-Stinner und Peter Stephan, eds. Mokṣopāya. Das Fünfte Buch. Upaśāntipraka-
raṇa. Kritische Edition. Anonymus Casmiriensis: Mokṣopāya. Historisch-kritische
Gesamtausgabe. Herausgegeben unter der Leitung von Walter Slaje. Textedition. Teil
4. Akademie der Wissenschaften und der Literatur, Mainz. Veröffentlichungen der
Indologischen Kommission. Wiesbaden: Harrassowitz, 2013.
Śārikāstotra of Sāhib Kaul

(O3) Bodleian Library, Oxford, Chandra Shum Shere e.264, ff. 529r–531v.
(L1) School of Oriental and African Studies, London, No. 44389, ff. 1–5.
(B1) Staatsbibliothek zu Berlin, Preussischer Kulturbesitz, Sammlung Janert, Hs. or.
12509.
Siddhalakṣmīpūjāpaddhati
Manuscript, Kashmir Research Center accession no. 2376. Transcription of the Mukta-
bodha Indological Research Institute.
Saundaryalaharī
Anantakṛṣṇa Śāstri, ed. Saundarya-laharī of Śrī Śaṃkarācārya with Commentaries,
Saubhāgyavardhanī of Kaivalyāśrama, Lakṣmīdharā of Lakṣmīdharācārya, Aruṇā-
modinī of Kāmeśvarasūrin. Madras: Ganesh, 1957.
Secondary Sources
Aithal, K. Parameswara. 1999. A Descriptive Catalogue of the Sanskrit and Other Indian
Manuscripts of the Chandra Shum Shere Collection in the Bodleian Library. Part 3: Sto-
tras. Oxford: Clarendon Press.
Hanneder, Jürgen. 2001. “Sāhib Kaul’s Presentation of Pratyabhijñā Philosophy in his
via free access
Devīnāmavilāsa,” pp. 414–415. In Le Parole e i Marmi. Studi in onore di Raniero Gnoli

nel suo 70 compleanno, edited by R. Torella, 399–418. Serie Orientale Roma, XCII, 1/2.
Roma: Istituto italiano per l’Africa e l’Oriente.
Hanneder, Jürgen, Stanislav Jager and Alexis Sanderson. 2012. Ratnakaṇṭhas Stotras
Sūryastutirahasya, Sūryaśataka und Śambhukṛpāmanoharastava. Indologica Mar-
purgensia, no. 5. München: Kirchheim Verlag.
Sanderson, Alexis. 2003–2004. “The Śaiva Religion Among the Khmers, Part I.” Bulletin
de l’Ecole française d’Extrême-Orient 90–91: 362–366.
Sanderson, Alexis. 2009. “Kashmir.” In Brill’s Encyclopedia of Hinduism, vol. I, edited by
Knut A. Jacobsen, 99–126. Leiden; Boston: Brill.
Sanderson, Alexis. 2015. “Śaiva Texts.” In: Brill’s Encyclopedia of Hinduism, vol. 6, edited
by Knut A. Jacobsen, 10–42. Leiden; Boston: Brill.
Verzeichnis der orientalischen Handschriften in Deutschland.
Windisch, Ernst and Julius Eggeling. 1894. Catalogue of the Sanskrit Manuscripts in the
Library of the India Office. Part IV. A.VIII Tantra. London: Printed by Gilbert and Riv-
ington by order of the Secretary of State for India in Council.
via free access
chapter 15
The Kāmasiddhistuti of King Vatsarāja

Diwakar Acharya
This essay concerns a pūjāstuti1 that guides its reciter through the mental or
actual worship of the goddess Nityā. The text is composed in the first person but
the author does not name himself in the text. The text is named Vāmakeśvarī-
stuti and attributed to Mahārājādhirāja Vidyādharacakravartin Vatsarāja in the
colophon of the sole palm-leaf manuscript of the text available to me. How-
ever, the last verse of the text calls it Kāmeśvarīstuti and describes it using two
adjectives, kāmasiddhi and atimaṅgalakāmadhenu. It is not unnatural, I think,
to name this stuti using its first adjective.2
The manuscript containing this stuti text is preserved in the National Ar-
chives, Kathmandu. It bears accession number 1–1077 and can be found micro-
filmed under NGMPP reel number A 39/15. The same manuscript also contains
a paddhati text called Aśeṣakulavallarī that dwells on the worship of the god-
dess Tripurā, but this text remains incomplete as the folios following the six-
teenth are absent. Our text begins on the verso of the first folio and ends in
the third line of the recto of the fourth, with a colophon and a decorative sym-
bol. The other text immediately follows in the same hand with a salutation to
the goddess Tripurā. The manuscript is written in a variety of North Indian
script close to Newari with frequent use of pṛṣṭhamātrās. It is possible that
this manuscript was copied by an immigrant or pilgrim in Kathmandu valley.
It measures 33×4.5cm and has a binding hole to the left of the centre. It bears
foliation in numerals in the left margin and in numbers in the right margin
of verso folios. The text in the manuscript is dotted with scribal errors, but no
secunda manus corrections are seen. On palaeographical grounds I place the
manuscript in the late fourteenth century.
This manuscript contains 46 verses of the stuti and one more verse (num-
bered here as 38a) can be retrieved from a citation.3 A little less than the half
of the stuti covering the first 21 verses is in Anuṣṭubh metre and the rest in
1 A number of pūjāstutis of the Tripurā traditon can be found in the appendix section of
Dwivedi 1985. Aghoraśiva’s Pañcāvaraṇastava, published from Pondicherry (see Goodall et
al. 2005), is a good example of a Siddhānta Śaiva pūjāstuti.
2 See footnote 32 for further discussion on the name, extent and circulation of the text.
3 See foonote 32 for details.
© diwakar acharya, 2020 | doi:10.1163/9789004432802_017

via free access
the kāmasiddhistuti of king vatsarāja 365
Vasantatilakā. Verses 31 and 32 form a yugalaka as the finite verb comes only
in the second verse. The author plays now and again with syllabic rhyming
(anuprāsa), and his language is beautiful, though sometimes elliptical.
The stuti opens with a pair of verses invoking Paramaśiva and Nityā Śakti.
These verses already tell us of the poet’s understanding of the nature of Nityā
and inseparability of Paramaśiva and Śakti, a point highlighted in the second
half of the text, particularly verses 31–32 and 42. In verse 3 the poet states that
he approaches the temple of Mṛḍānī from the west gate (paścimadvāra).4 The
next two verses invoke Gaṇeśa and Kṣetreśa. The latter, who has the form of
Bhairava, can be identified as Baṭuka. Gaṇeśa and Baṭuka together are iden-
tified as the goddess’s sons in Śākta systems and serve as her doorkeepers.5
To our surprise, verse 6 invokes the Vaiṣṇava doorkeepers Śaṅkhanidhi and
Padmanidhi, who bear the Vaiṣṇava emblems of the conch and lotus on their
heads.6 Verses 7–9 invoke respectively three goddesses: Padmā, a Vaiṣṇava ver-
sion of Durgā carrying a conch and discus, and Bhāratī. Verses 10 and 11 invoke
Manobhava, namely, the Indian love-god Kāmadeva, and describe him as the
4 This should be the intended meaning, because one is supposed to enter a temple from the
western or southern gate facing east or north. Therefore, many of the early Śaiva-Śākta tem-
ples, even though they face east, have an older western or southern entry. For more discus-
sions, see Goodall et al. 2005, 103–107 and Goodall et al. 2015, 366 (Niśvāsa, Uttarasūtra 3:8
and annotation thereon). Another possible interpretation of paścimadvāra is “the last door
to resort to.” Perhaps, the poet is punning.
5 For Gaṇeśa and Baṭuka as the Goddess’ sons, See, e.g., Jayaratha on Tantrāloka 1.6b.
6 Śaṃkhanidhi and Padmanidhi have strong associations with the cult of Yakṣas. In the Megha-
dūta, Kālidāsa’s Yakṣa tells the cloud-messenger that the marks of conch (śaṃkha) and lotus
(padma) are painted on the sides of the gate of his house in the city of Alakā, as he pro-
vides a number of clues for the identification of his house. In the form of emblems as well as
human forms, Śaṃkhanidhi and Padmanidhi are depicted in the Ajaṇṭā caves and are asso-
ciated with Yakṣa deities (cf. Bautze-Picron 2002, 225–231). Besides, the Buddhist Vasudhārā
Dhāraṇī enjoins worship Śaṅkhanidhāna and Padmanidhāna with the goddess Vasudhārā
encircled by a group of eight unspecified Yakṣiṇīs. Some other texts name Śaṃkhanidhi and
Padmanidhi as male consorts of Vasudhārā and Vasumatī, respectively. Anyway, these two are
adopted by the Vaiṣṇavas as doorkeepers or attendants of Viṣṇu along with the other pairs of
Jaya and Vijaya, Caṇḍa and Pracaṇḍa, Nanda and Sunanda. They also feature in some compar-
atively late Tantric texts of other traditions, particularly those from the south. They are listed
also among the twelve Vaiṣṇava nidhis found in some Puranic and Vaiṣṇava texts. Professor
Dominic Goodall kindly informs me (personal communication of November 20, 2019) that
what is now called the Kailāsanātha temple in Kancheepuram seems to have Śaṅkhanidhi
and Padmanidhi framing the doorway. According to him, that temple now has an eastern
entrance to the enclosure, but there is an older western entry, now blocked up.
For an example of images of Śaṃkhanidhi and Padmanidhi from Anurādhapur, Sri Lanka,
See Paranavitana 1955.
via free access
366 acharya
beloved husband of Rati and Prīti.7 Here we are told that the love-god forms
the circular base of the Śrīcakra, the maṇḍala of the goddess Nityā Sundarī.
With these verses the text enters the process of installation of various deities
in the Śrīcakra. It does not specify where these deities are installed, but from
the order of verses we know that we are starting from the periphery and mov-
ing towards the centre. Verses 12–14 respectively praise eight siddhis, beginning
with Aṇimā (in personified forms), eight mother-goddesses, and the deities of
ten gestures of the goddess.8 Verses 15 and 16 venerate sixteen goddesses of
attraction (ākarṣaṇa) and eight powers of the bodiless love-god (anaṅgaśakti),
respectively, all in personified forms.9 We know from the Vāmakeśvaratantra
and other Tripurā texts that these are installed on the petals of the sixteen- and
eight-petalled lotuses. The next four verses, 17–20, respectively praise the set of
fourteen goddesses/powers (śaktis) headed by Sarvasaṃkṣobhaṇī,10 ten Kula
7 For Rati and Prīti as Kāmadeva’s wives, see, e.g. Brahmāṇḍa Purāṇa III.44.33. A Śuṅga
period teracotta plaque of Kāmadeva with Rati and Prīti is also preserved in the Mathura
Musuem (accession no. 34–2552).
8 It is possible that these three sets of deities are installed on the three lines forming the out-
ermost retinue of the rectangular boundary. The Vāmakeśvaratantra, also known as Nityā-
ṣoḍaśikārṇava, enjoins installing the eight mother-goddesses as well as the eight siddhis
in the four directions and four sub-directions, and does not instruct one to worship the
goddesses of the gestures. Bhāskararāya (p. 99), however, mentions that according to some
other system the outermost boundary is made of three lines and these three sets of god-
desses are installed there. According to its commentators, the Vāmakeśvaratantra teaches
that one should build the boundary with only two lines. Although the Vāmakeśvaratantra
does not assign a place for the gestures (mudrā) in the maṇḍala, it does describe them
and asks the worshipper to use them during the worship. As found in the third chapter
of the Vāmakeśvaratantra, these ten gestures are trikhaṇḍā, kṣobhiṇī, vidrāviṇī, ākarṣiṇī,
āveśakarī, unmādinī, mahāṅkuśā, khecarī, bīja, and yoni.
As listed in many texts, including the Niśvāsaguhya (7.204–205), the eight siddhis are
aṇimā, laghimā, mahimā, īśitva, vaśitva, prāpti, prākāmya, and yatrakāmāvasāyitā. The
Vāmakeśvaratantra (1.153–155) makes them ten by adding two more, bhukti and icchā,
and prescribes worshipping them in ten directions. According to the latter (1.156–157),
the eight mother-goddesses are Brahmāṇī, Māheśī, Kaumārī, Vaiṣṇavī, Vārāhī, Indrāṇī,
Cāmuṇḍā, and Mahālakṣmī.
9 These are not individually named in this text, but, as listed in the Vāmakeśvaratantra, the
first set is made of Kāmākarṣiṇī, Budhyākarṣiṇī, Ahaṃkārākarṣiṇī, Śabdākarṣiṇī, Sparśā-
karṣiṇī, Rūpākarṣiṇī, Rasākarṣiṇī, Gandhākarṣiṇī, Cittākarṣiṇī, Dhairyākarṣiṇī, Smṛtyā-
karṣiṇī, Nāmākarṣiṇī, Bījākarṣiṇī, Ātmākarṣiṇī, Amṛtākarṣiṇī, and Śarīrākarṣiṇī (cf. 1.158–
161), and the second set is made of Anaṅgakusumā, Anaṅgamekhalā, Anaṅgamadanā,
Madanāturā, Anaṅgarekhā, Anaṅgaveginī, Anaṅgāṅkuśā, and Anaṅgamālinī (cf. 1.163–
164).
10 We know only the name of the first from this text but the rest can be known from
the Vāmakeśvaratantra (1.165–168). They are: Sarvavidrāviṇī, Sarvākarṣiṇī, Sarvāhlādinī,
Sarvasaṃmohinī, Sarvastambhanī, Sarvajambhanī, Sarvatovaśinī, Sarvarañjanī, Sarvon-
via free access
goddesses (kuleśvarī) headed by Sarvasiddhipradā,11 ten goddesses headed by

Sarvajñā,12 and eight goddesses of speech, headed by Vaśinī.13 They are sta-
tioned in the four consecutive retinues of fourteen, ten, another ten, and eight
triangles. All deities in a group (see verses 12–20) are visualised in the same
way; for example, all mother-goddesses (mātṛ) have the same appearance.14
Verse 21 invokes and asks the deities of four weapons of the goddess for their
permission. It is known from other sources that they are placed around the
central triangle (cf., e.g., Vāmakeśvaratantra 1.179–180). The next three verses,
22–24, praise Kāmeśvarī, Vajreśvarī, and Bhagamālinī, and urge them to fulfill
the reciter’s desires. Unlike previous ones, these verses also name the three cor-
ners of the central triangle as the homes of these goddesses. Verse 25 is in praise
of Nityā Sundarī, the goddess in the centre. From here onward, until the second
to last verse (45), the poet praises Nityā in various ways. He first invokes the god-
dess as Nityā (verse 25) and later as Śrīsundarī (verse 30), and describes her as
“the felicitous banner of the Love-god.” Verses 25–28 describe the beauty of the
goddess, and verses 29–45, with the exception of verse 33 (which describes the
Śrīcakra made of 43 triangles as her abode), exalt her in various ways, identi-
fying her as the ultimate reality of the external as well as internal worlds. She
is described as the primordial light (ādyamahas) and paramārthavidyā, which
can be interpreted as the highest mantra, the mantra leading to the highest, or
the ultimate gnosis. The last verse is a fine eulogy of the stuti itself, describing
its reward and thus encouraging people to recite it.
It has been already pointed out by Sanderson and also Golovkova that the
mature cult of Tripurasundarī developed against the backdrop of the nityā cult,
evidence for which is available in the Nityākaulatantra and the Siddhakhaṇḍa
of the Manthānabhairavatantra. In those texts Tripurasundarī is accompa-
mādinī, Sarvārthasādhanī, Sarvasampattipūraṇī, Sarvamantramayī, and Sarvadvandva-

kṣayaṃkarī.
11 Again, the list can be completed with the help of the Vāmakeśvaratantra, but these god-
desses are here simply called śaktis. The other nine following Sarvasiddhipradā are: Sar-
vasampatpradā, Sarvapriyaṃkarī, Sarvamaṅgalakāriṇī, Sarvakāmapradā, Sarvaduḥkha-
vimocinī, Sarvamṛtyupraśamanī, Sarvavighnanivāriṇī, Sarvāṅgasundarī, and Sarvasau-
bhāgyadāyinī (cf. 1.169–171).
12 Sarvajñā is followed by Sarvaśakti, Sarvaiśvaryapradāyinī, Sarvajñānamayī, Sarvavyādhi-
vināśinī, Sarvādhārasvarūpā, Sarvapāpaharā, Sarvānandamayī, Sarvarakṣāsvarūpiṇī, and
Sarvepsitaphalapradā (cf. Vāmakeśvaratantra 1.173–175).
13 The names of these eight can be retrieved from the mantroddhāra section of the Vāma-
keśvaratantra (cf. 1.77–80). They are Vaśinī, Kāmeśvarī, Modinī, Vimalā, Aruṇā, Jayinī,
Sarveśvarī, and Kaulinī.
14 Neither the Vāmakeśvaratantra nor any of the paddhatis of that tradition give visualisa-
tions of these deities.
via free access
368 acharya
nied by a retinue of eleven and nine nityās, respectively, and worshipped with
Kāmadeva.15 Our text identifies Kāmadeva as the husband of Rati and Prīti,
places him on the base of the Śrīcakra (cf. verses 10–11), and installs Nityā Sun-
darī at the altar of worship in the centre of the maṇḍala without a consort, inde-
pendent and supreme. However, in verses 31–32 she is described as devamahiṣī,
although it is said that their body is one and undifferentiated. In verse 2 the poet
names the goddess Nityā and invokes her as the Śakti of Paramaśiva possessing
all powers and carrying out the five tasks (pañcakṛtya) for him. In verse 34 the
poet invokes her as Maheśvarī but states that some royal people in this world
call her Lakṣmī and Parā Prakṛti. In verse 40 she is described as Atibhavā, high-
lighting her transcendent nature, and in verse 42 she is invoked again as Gaurī.
It is thus clear that the poet of our text is a Śaiva devotee of goddess Nityā. It
is important to note that in the system known to our poet there is only one
Nityā, simply called Sundarī, and that the Śrīcakra is also already known. Our
poet appears unaware of the sixteen nityās, who are worshipped in the tradi-
tion of the Vāmakeśvaratantra. It thus appears that the tradition this stuti text
represents is different from both the cult of nityās and that of Tripurā.
The inclusion of Śaṅkhanidhi and Padmanidhi (verse 6), Padmā (verse 7),
and the Vaiṣṇava Durgā (verse 8) suggests that the goddess Nityā is somehow
linked to the Vaiṣṇava tradition as well. In fact, in verse 34 the poet mentions
that some people call her Lakṣmī and Parā Prakṛti, but we are not aware of sur-
vival of any Vaiṣṇava paddhati of Nityā.
Now I come to the issue of the poet’s identity. The fact that he is a king and
was perhaps somewhat distressed at the time of composition of the stuti can
be known from the text itself (cf. verse 40). Furthermore, in the colophon the
text is attributed to Mahārājādhirāja Vidyādharacakravarti Vatsarāja.16 Appar-
ently, the first epithet is royal—he is the king of great kings—while the second
is mantric: he is sovereign among the vidyādharas, who are supposed to pos-
sess esoteric mantric knowledge and due to this have supernatural powers.
Vatsarāja is his personal name. The most famous Vatsarāja, the mythical king
of Ujjayinī, does not fit the context. Another is King Vatsarāja of the Gurjara-
Pratihāra dynasty (c. 775–805ce), the father of Mahārājādhirāja Nāgabhaṭa
II (805–833ce). Vatsarāja is always called paramamāheśvara, but in the Pra-
tāpagaḍh Stone Inscription of Mahendrapāla II (dated Year 1003 = 946 ce),
15 cf. Sanderson 2009, 47–49; Golovkova 2012, 816–817.

16 It is interesting to note that a fifteenth-century inscription from Vijayanagara remembers a
king called Vatsarāja blessed by Tripurāmbā. As Sinopoli (2010, 22) cites, “‘As Vani blesses
king Bhoka, Tripuramba king Vatsaraja, and Kali king Vikramarka, so does Pampa now
bless Devaraya’ (trans. Rajasekhara 1992: 27).”
via free access
Nāgabhaṭa II is called paramabhagavatībhakta.17 It may be a coincidence, but

the latter’s mother is named Sundarī. In any case, this Vatsarāja could be our
poet.18 Our text represents an archaic tradition that does not even know the
name Tripurasundarī, and thus this date in the early-ninth century ce fits it
well.
1 Text and Translation
19niḥsīmānandayā devyā nityaṃ samarasātmanaḥ |

paramasya śivasyāhaṃ śraye śrīpādukādvayam ||1||
a. niḥsīmā°] em.; niḥśāmā° Ms. b. °rasātmanaḥ] em.; °rasānmanaḥ Ms.
I resort to the glorious sandals of Paramaśiva (Śiva in the Ultimate state),

who is eternally in equilibrial union with the Goddess characterised by
boundless bliss (niḥsīmānanda).20
sarvānugrāhiṇīṃ nityāṃ sarvamaṅgalamātaram |

sarvaśaktiṃ bhaje śaktiṃ pañcakṛtyakarīṃ prabhoḥ ||2||
a. sarvānugrāhiṇīṃ] em.; sarvān grāhiṇī Ms. d. °karīṃ] em.; °karī Ms.
I revere the Nityā Śakti of the Lord, i.e. Paramaśiva. She possesses all pow-
ers and carries out the five tasks [for him].21 She bestows grace upon all,
is eternal, and is the motherly origin of all good.
17 cf. Sircar 1983, 251.

18 There is another poet of the same name who flourished in the second half of the twelfth
and the first quarter of the thirteenth century ce (cf. Dalal 1918, vi–vii), but he is a minister,
not a king. He served the Kālañjara King Paramardideva and wrote some dramatic pieces.
Six of such pieces have been published in one volume under the title Rūpakaṣaṭkam (see
Dalal 1918). He does not mention Nityā, Sundarī, or Tripurasundarī in his dramas.
19 The manuscript begins with an invocation, ||oṃ namo gaṇapataye||, preceded by a siddhi
sign. I do not think that this invocation is part of the text.
20 The manuscript reads niḥśāmānanda- and I have emended it to niḥsīmānanda. I have
found this compound used at least in one more text, the Adhikaraṇasārāvalī of Vedān-
tadeśika.
21 The five tasks of Śiva include punishment (nigraha) and grace (anugraha), besides cre-
ation, maintenance, and destruction.
via free access
370 acharya
pālitaṃ bahir indrādyaiḥ paramaiśvaryaśobhitam |

prapadye paścimadvāraṃ mṛḍānyā mandiraṃ mahat ||3||
b. paramaiśvarya°22] em.; parameśvarya° Ms.
I approach the great temple of goddess Mṛḍānī that opens to the west. It is
guarded outside by Indra and the other [gods who guard the directions],
and shines beautifully with utmost richness.
pāśāṅkuśaphalāmbhojaiḥ pāṇipadmaṃ tu maṇḍitam |

bālaṃ vighnacchidaṃ vande gajavaktraṃ gaṇeśvaram ||4||
a. °phalāṃbhojaiḥ] conj.; °phalāṃbhoja Ms. b. pāṇipadmaṃ tu maṇḍitam] conj.; pāṇi-

pādan tu nanditam Ms. c. bālaṃ] conj.; bāraṃ Ms.
I venerate the young elephant-faced master of Śiva’s gaṇas, the destroyer

of obstacles. His lotus-hands23 are decorated with a noose, goad, fruit, and
lotus.
kapālaśūlau bibhrāṇaṃ kṛpāluṃ kṛṣṇavigraham |

trīkṣaṇaṃ tīkṣṇam arcāmi kṣetreśaṃ kṣatavidviṣam ||5||
a. °śūlau bibhrāṇaṃ] em.; °śūlo bibhrāṇa Ms. b. kṛpāluṃ] em.; kṛpālu Ms. b. kṣetreśaṃ
kṣata°] em.; hyetreśaṃ hyata° Ms.
I worship the three-eyed sharp-natured Kṣetreśa. His body is black, he has

destroyed his adversaries, he carries a skull-bowl and a spear, [but] he is
compassionate.24
varābhayadharau dhīrau viśvasaṃkalpakalpakau |

śaṅkhapadmāsanagatau śaṃkhapadmanidhī śraye ||6||
I resort to Śaṅkhanidhi and Padmanidhi, who who sit upon a conch and
lotus [respectively] as their seats. They are patient, bear the gestures
of generosity and protection in their hands, and bring about everyone’s
dreams.
22 pāramaiśvarya° is equally possible.

23 Finding the manuscript reading pāṇipādan tu nanditam problematic, I have conjectured
pāṇipadmaṃ tu maṇḍitam.
24 This description reveals that Kṣetreśa has the form of Bhairava. I assume that he is Baṭuka-
bhairava on the basis of his Bhairava-like appearance and companionship with Gaṇeśa.
via free access
padmadvayavarābhītibhāsvatpāṇicatuṣṭayam |
padmavallīṃ bhaje padmāṃ padmākṣīṃ padmavāsinīm ||7||
c. °vallīṃ bhaje padmāṃ] em.; °vallī bhaje padmā Ms. d. padmākṣīṃ25] em.; padmāhyā
Ms.
I honour Padmā, [beautiful and tender like] a lotus plant. Her eyes are
lotus-like and she dwells in a bed of lotuses. Her four arms look splendid
with two lotuses [in two hands] and the gestures of grace and safety [in
two others].
śaṅkhacakrāṅkitakarā kumārī kuṭilālakā |

mṛgendravāhanā devī durgā durgāṇi hantu me ||8||
b. kuṭilā°] em.; kuḍhilā° Ms.
May the virgin goddess Durgā annihilate my hardships (durgāṇi), I pray.

Her hands are marked by the conch and discus. She has curly locks and
rides [a lion,] the king of wild animals.
akṣasrakpustakadharā pūrṇacandrāmaladyutiḥ |
viśvavidyāmayī devī bhāratī bhāsatāṃ mayi ||9||
a. akṣa°] em.; ahya° Ms. d. bhāratī] em.; bhāratā Ms.
May goddess Bhāratī shine upon me, I pray. She carries a rosary and a book
in her hands, she has the stainless complexion of the full moon, and she
embodies the entirety of knowledge.
palāśapaṭalacchāyaṃ ramaṇīyaṃ ratipriyam |

puṣpeṣucāpaṃ puṣpeṣumantaṃ vande manobhavam ||10||
prapadye prītidayitaṃ pūrṇendum iva veṣṭitam |
āśrayaṃ śakticakrasya śrīcakrāñcitaye śriye ||11||
10c. puṣpeṣu°] conj.; puṇḍeṣu° Ms. ● °mantaṃ] em.; °mattaṃ Ms. 11b. pūrṇendum iva
veṣṭitam] em.; pūṇṇendur i veṣṭitam Ms.
25 padmākṣāṃ is possible, but as I have observed confusion between the mātrās of ā and ī
in this manuscript, I opt for padmākṣīṃ.
via free access
372 acharya
I venerate the beloved husband of Rati, the beautiful Mind-born [God

Kāmadeva]. He carries a bow and arrows of flowers and his complexion
resembles the petals of Dhak. [Again,] I approach the beloved husband of
Prīti, bent round (veṣṭitam) like the full moon,26 [serving as] the base for
the ring of goddesses, in order to draw the Śrīcakra for the sake of pros-
perity.
cintāratnobhayakarāś candrottaṃsās trilocanāḥ |

aṇimādimahāsiddhīr aruṇāḥ siddhaye śraye ||12||
a. °karāś] em.; °karā Ms. c. aṇimā°] em.; animā° Ms.
I take refuge with the goddesses of becoming minute (aṇimā) and other
great accomplishments (mahāsiddhis) for the sake of success. They hold
wish-fulfilling jewels in both hands. They are moon-crested, three-eyed,
and red in complexion.
vaṃśīdalaśyāmalāṅgīḥ kapālotpaladhāriṇīḥ |
brahmāṇyādīr bhaje mātṝr bandhūkarucirāmbarāḥ ||13||
c. °ṇyādīr bhaje mātṝr] em.; °ṇyādir bhaje māntar Ms.
I revere Brahmāṇī and the other mother-goddesses. They carry a skull-

bowl and red lily in their hands, their bodies are dark-colored like the
leaves of bamboo, and they are clad in lovely [red] clothes resembling
bandhūka flowers.
dāntāḥ pāśāṅkuśadharāḥ svasvamudrāvaśoditāḥ |

anugṛhṇantu me ’bhīṣṭaṃ mudrādaśakadevatāḥ ||14||
a. dāntāḥ] em.; dantāḥ Ms. b. °vaśoditāḥ] em.; °vadhoditaḥ Ms. c. °gṛhṇantu] em.;
°gṛhṇaṃnta Ms.
May the deities of the ten gestures (mudrās), who [have forms that] are
in accordance with the powers of their respective gestures, are mild, and
carry a snare and goad, endow me with the object of my desire.
26 It appears that Kāmadeva is bent round in the shape of the moon inside the sixteen-
petalled and eight-petalled lotuses, and forms the base for the Śrīcakra in the form of
nested triangles.
via free access
raktāḥ pāśāṅkuśadharāḥ kalā nityāḥ kalānidheḥ |

ākarṣantu mamābhīṣṭaṃ ṣoḍaśākṛṣṭiśaktayaḥ ||15||
b. °nidheḥ] em.; °nidhiḥ Ms. c. ākarṣantu] em.; ākarṣanta Ms.
May the sixteen goddesses of attraction, [representing] the perennial

constitutive digits (kalā nityāḥ) of Kalānidhi [i.e. the Moon], draw towards
me the object of my desire.
aruṇāḥ karuṇāvṛttīr devyāś chāyā ivoditāḥ |

anaṅgaśaktīr aṣṭau tāḥ pūjayāmi sudurjayāḥ ||16||
a. aruṇāḥ] em.; aruṇā Ms. ab. °vṛttīr devyāś chāyā] em.; °vṛtti ddaivyā cchāyā Ms. c.
aṣṭau tāḥ] em.; aṣṭaustā Ms.
I worship those compassionately-disposed goddesses of red-complexion,

the eight powers of the bodiless [love-god Kāmadeva], who have arisen
like shadows of the goddess [Nityā Sundarī] and are very difficult to con-
quer.
sarvasaṃkṣobhaṇīpūrvāḥ śoṇabāṇadhanurdharāḥ |
caturdaśa bhaje śaktīś caturdaśajagannutāḥ ||17||
a. sarvasaṃ°] em.; sarvasa Ms. b. śoṇa°] em.; śoṇo° Ms. ● °rdharāḥ] em.; °dharāḥ Ms. c.
caturdaśa] em.; vantadaśa Ms. ● śaktīś°] em.; śakti Ms. d. jagannutāḥ] em.; jagantutāḥ
Ms.
I venerate those fourteen goddesses, with Sarvasaṃkṣobhaṇī at the fore,

to whom [all] fourteen worlds bow. They carry a bow and arrows made of
sugarcane.
śubhā varābhayabhṛto vande viśvakuleśvarīḥ |

sarvasiddhipradādyās tā bahirdaśakadevatāḥ ||18 ||
a. śubhā] em.; śucā Ms. ● °bhṛto] em.; °bhūto Ms. (unmetrical) c. °pradādyās] em.;
°pradadyās Ms.
I venerate all [ten] Kuleśvarīs, starting with Sarvasampatpradā, the god-

desses of the external ring of ten. They are auspicious and display the
gestures of boon-giving and safety.
via free access
374 acharya
akṣasrakpustakadharāḥ karpūrarucirākṛtīḥ |
antardaśakadevīs tāḥ sarvajñādyāḥ samāśraye ||19||
b. karpūra°] em.; kappūra° Ms. c. antardaśaka°] em.; antarddhaśaka° Ms.
I resort to Sarvajñā and other goddesses situated in the internal ring of ten.
They carry a rosary and a book [in their hands], and their appearance is
charming like camphor.
cāpeṣupustakākṣasrakcārupāṇicatuṣṭayāḥ |
raktā vāgīśvarīr vande vaśinyādyaṣṭadevatāḥ ||20||
a. °srak°] em.; °srakta° Ms. (unmetrical) b. °catuṣṭayāḥ] em.; °cantaṣṭayāḥ Ms. c.

°vāgīśva°] em.; °vāgeśva° Ms.
I worship those eight goddesses of speech, Vaśinī and others, whose com-
plexion is red. They carry in their four lovely hands a bow, arrows, a book,
and a rosary.
raktā varābhayabhṛtaḥ svarūpāṅkitamastakāḥ |

catasro me diśantv ājñām iṣvādyāyudhadevatāḥ ||21||
a. °bhṛtaḥ] em.; °bhūtaḥ Ms. (unmetrical) b. svarūpā°] em.; surūpā° Ms. c. catasro] em.;
cartasro Ms. d. iṣvādyā°] em.; iṣṭādyā° Ms.
May the four deities of the weapons [of the Goddess]—the arrows and
others—red in complexion, displaying the gestures of generosity and pro-
tection, marked on the head by their own respective weapon-forms, grant
me permission [for worship of the Goddess].
pāśāṅkuśāmṛtakapālavarābhayāṅkair
hastaiś caturbhir abhirāmadṛśāruṇāṅgī |
koṇāgragā trinayanā taruṇenducūḍā
kāmeśvarī mama dadātu samastakāmān ||22||
ab. °yāṅkair hastaiś] em.; °yāṅke haste Ms. b. °dṛśā°] conj.; °bhṛśī° Ms. d. kāmeśvarī
mama dadātu] em.; kāmaśvarī mama dadānta Ms.
May the goddess Kāmeśvarī, who dwells at the front angle [of the central
triangle], give me all objects of my desire. She is three-eyed, her eyes are
beautiful and her limbs are ruddy. She has the crescent moon on her crest.
via free access
She looks beautiful with her four hands marked with a snare together with
a goad, a plate with the nectar of immortality, the gesture of boon-giving,
and the gesture of safety.
bālaprabālarucirā karaṣaṭkasakta-
cāpeṣupāśaśṛṇipālakamātuluṅgā |
vajreśvarī prathitadakṣiṇakoṇavāsā
vajrojjvalā vidiśatāṃ mama vāñchitāni ||23||
a. °sakta°] conj.; °saddhi° Ms. b. °mātuluṅgā] em.; °māntaluṅgā Ms. d. dadātu] em.;
dadānta Ms.
May the goddess Vajreśvarī give me all objects of my desire. She is known
to have her abode at the right corner [of the central triangle]. She is
resplendent like a thunderbolt, beautiful like fresh coral, and has a bow,
arrows, a snare, a hook, a shield, and a mātuluṅga fruit attached to her six
arms.
tryakṣā śaśāṅkarucirā śritavāmakoṇā

pāśāṅkuśekṣuguṇapustakaśastrahastā |
udbhāvayatv aniśam adbhutabhūriśaktir
bhāgyaśriyaṃ bhagavatī bhagamālinī me ||24||
a. °vāmakoṇā] em.; °vāmako Ms. (unmetrical) c. °adbhutabhūriśaktir] em.; °adbhṛta-

bhūriśakti Ms. c. bhagavatī] em.; bhagavatā Ms.
May the three-eyed goddess Bhagamālinī give the glory of good fortune.
She possesses abundant miraculous power and is as lovely as the moon.
She is stationed in the left corner [of the central triangle] and holds in the
row of her arms a snare, a goad, a sugarcane, ropes, a book, and a sword.
sūryenduvahnimayabhāsurapīṭharohāṃ
svacchāṃ gṛhītaśṛṇipāśaśareṣucāpām |
bālendumaulim alakāgralalāmanetrāṃ
nityāṃ namāmi satataṃ mahanīyamūrtim ||25||
a. °rohāṃ] em.; °rohaṃ Ms. b. svacchāṃ] em.; svacchā Ms. °cāpām] em.; °cāpaṃ Ms.
cd. satataṃ mahanīya°] conj.; mahanīya° Ms. (unmetrical)
I uninterruptedly bow to Nityā who has a form worthy of worship. She has
ascended the shining throne made of the sun, moon, and fire. She holds
via free access
376 acharya
in her hands a hook, a snare, arrows, and a bow, and carries the crescent
moon on her crest. She is pure and clean, and her eyes, adorned with the
tips of the locks of hair, are very beautiful.
sindūrasundaratanuṃ tanumadhyabhāgāṃ
kāntyāśrayāṃ kalabhavatkucakumbhanamrām |
candrānanāṃ calakuraṅganibhāyatākṣāṃ
mandasmitāṃ madanamaṅgalavaijayantīm ||26||
a. °tanuṃ] em.; °tanu Ms. b. kāntyāśrayāṃ kalabhavatkuca°] conj.; kāntiśrayaḥ kulabha-

vatkaca° Ms. d. °vaijayantīm] em.; °vejayantīṃ Ms.
Her body is beautiful and bears the hue of vermillion. Its middle part is
slim, [and] she is the repository of beauty. She is slightly bent like a young
elephant because of her pitcher-like breasts, resembling the temples of a
young elephant. Her eyes are moving and wide like those of a deer. She is
moon-faced, her smiles are gentle, and she serves as the felicitous banner
of the Love-god.
koṭīriṇīṃ kaṭakakuṇḍalahāravallī-
kāñcīkalāpamaṇinūpuramaṇḍitāṅgīm |
bandhūkabandhuvasanāṃ bahalānurāgāṃ
kāśmīracandanasamullasitāṅgarāgām ||27||
a. °riṇīṃ kaṭaka°] conj.; °riṇī kaṭa° Ms. (unmetrical) ● °vallī°] conj.; °vallīṃ Ms. b.
°kalāpa°] em.; °kālāpa° Ms. c. °nurāgāṃ] em.; °nurāgā Ms.
She has braided hair. Her limbs are adorned with bracelets, earrings, neck-
laces, twining laces, girdles, jewels, and anklets. Her clothes resemble
Bandhūka flowers. She is full of affection, and the hue of her body is
brightened up with saffron and sandal paste.
muktāvitānamahite maṇiviṣṭarāgre
paryaṅkaparṣarucire surasopaviṣṭām |
paryaṅkapaṅkajamukhīṃ dhutacāmarālāṃ
hāse vilaṅghitalasadvadanāravindīm ||28||
b. paryaṅkaparṣa°] em.; paryaṅkaparya° Ms. ● suraso] em.; suvaso ° Ms. c. °mukhīṃ

dhutacāmarālīṃ] em.; °mukhī dhutacāmarālā Ms. d. hāse] em.; hasi Ms. (unmetrical)
via free access
She is elegantly seated on a lofty couch studded with jewels, furnished

with seats (paryaṅka) and pillows (parṣa), and decorated with a canopy
of pearls. Her face is a fully developed lotus. She has a row of chowries
being shaken around her, and her beaming lotus-face surpasses beautiful
lotuses.
ārādhanamravibudhendrabudheśavṛnda-
sīmantaratnarucirañjitapādapīṭhām |
loladviśālanayanāṃ calakelikḷpta-
svārājyavaiśravaṇatādivarapradānām ||29||
a. °budheśavṛnda°] em.; °buddhasamuda° Ms. (unmetrical) d. °kḷpta°] em.; °klupta° Ms.

● °vaiśravaṇa°] em.; °vaiśramaṇa° Ms. ● °pradānām] MSpc, °pradānaṃ MSac
Her foot-rest is illuminated by the rays of jewels on the forehead of the

king of gods and other gods as they bow in devotion [to her feet]. She has
roving, wide eyes, and she bestows as boons the sovereignty [of Indra]
and status [equal to] Vaiśravaṇa, lord of riches, and still more, which she
creates in fleeting acts of amusement.
ānandasāndraparamodyamadīpyamāna-
svacchandasaṃsphuradamandataraprakāśām |
devīṃ dayārdrahṛdayāṃ hṛdayaṃ rahasyaṃ
śrīsundarīṃ śivakarīṃ śaraṇaṃ śrayāmi ||30||
a. °paramodyama°] em.; °paramedyama° Ms. b. °saṃsphura°] em.; °sasphura° Ms. c.

devīṃ] em.; devī Ms. ● hṛdayaṃ rahasyaṃ] em.; hṛdaya rahasya Ms.
I seek refuge with the glorious goddess Sundarī, the benefactress of pros-
perity, the secret heart, whose heart is soaked with compassion. She is
blazing with an utmost tenacity steeped in joy, and consequently beam-
ing with plenteous light that shimmers spontaneously.
tvāṃ devi devamahiṣīm avibhāgabhogāṃ

bhogāpavargaphaladāṃ bhuvaneśi dhārām |
śaivādhibhūmyavadhiṣaḍguṇitātmaṣaṭka-
vaicitryacitraracanodbhavabhāvamārgām ||31||
ādhāravāriruhaṣaṭkavilāsasaumye
sauṣumṇavartmani sudhāṃśurasān śravantīm |
via free access
378 acharya
ānanditatribhuvanām aruṇāruṇāṅgīṃ
vande ’ham ādyamahasaṃ manasāpi vācā ||32||
31b. dhārām] conj.; dhārāḥ Ms. 31c. śaivādhibhū°] conj.; śivādhibhū° Ms. 31d. vaicit-
rya°] em.; vecitra° Ms. ● °bhāvamārgām] conj.; °bhivamārggaḥ Ms. 32b. sauṣumṇa°] em.;
sauṣyamna° Ms. 32c. ānanditatri°] em.; ānanditastri° Ms. ● °ruṇāṅgīṃ] em.; °ruṇāṅgī
Ms. 32d. vande ’ham ādya°] conj.; m ādya° Ms. (unmetrical) ● vācā] em.; vāpi Ms.
O goddess, I praise you with mind and speech. Your greatness is primor-
dial. Your limbs are slightly ruddy like the morning sun, and you have
made the triple world happy. You are the bride of the god [i.e., Śiva], and
possess a body inseparable [from his]. You bestow worldly enjoyment and
also liberation from [the world]. You are the stream [of consciousness or
immortality] (dhārā),27 O ruler of worlds. Dwelling originally in the abode
of Śiva, you multiply yourself sixfold and prepare the path of existence
where you nurture wonderful and manifold creation with your own six
forms. You shed moonlight on the path of Suṣumṇā that is charming due
to the beautiful appearance of the six lotuses serving as [your] bases.
ekaikavaty api navāsi daśāsi devi

bhūyo daśāsi punar eva caturdaśāsi |
itthaṃ trikādhikadaśadvitayadvayāṅke
śaktyarṇave vasasi śarmakarī kavīnām ||33||
a. ekaikavaty api] conj.; ekekavaty asi Ms. c. itthaṃ trikā°] conj.; itthan nrikā° Ms. ●
°dvayāṅke] em.; °dvayāke Ms. d. śaktya°] conj.; śaktyā° Ms. ● kavīnāṃ] em.; kavānāṃś
Ms.
O goddess, though you are one and simple,28 you are [also] nine,29 you are
ten, you are again ten, and again you are fourteen. Thus you, the benefac-
tor of poets, dwell in the sea of Śaktis marked with forty-three triangles.
27 These two verses depict the goddess as the stream of consciousness or immortality in the
human body, known widely as Kuṇḍalinī, originating from the brahmarandhra, the abode
of Śiva, flowing through various channels and reaching to the six bases. It is in this light
that these verses should be read.
28 I have conjectured api in place of asi to provide a concessive tone. Perhaps this is not even
necessary. In any case, on her own the goddess is singular and unembellished, but the poet
appears to imply that all goddesses in different retinues of the Śrīcakra are her projections.
29 The central triangle and the immediately following retinue of eight triangles are obviously
counted together as nine.
via free access
lakṣmīṃ parāṃ prakṛtim atra jagaty aśeṣam

ekātapatradharacāmaracārucihnām |
mātar maheśvari yaśaskṛtam ādyaśaktiṃ
prāhuḥ parāparadṛśaṃ parameśvari tvām ||34||
a. lakṣmīṃ] em.; lakṣmīḥ Ms. ● °aśeṣam] conj.; aśiṣam Ms. a. °cihnām] em.; °cihnāḥ Ms.
c. mātar mahe°] em.; māta’rmmahe° Ms.
O Mother! Great Goddess! Supreme Goddess! People proclaim you

Lakṣmī, Parā Prakṛti,30 who has chowries as lovely distinctive marks and
who bears a sole [royal] parasol covering the entire world. They proclaim
you as the conferer of fame, the primordial power, and the supervisor of
both higher and lower realms.
śrīs tvaṃ śriyas tvayi giras tvayi gīs tvam ājñā

dhīs tvaṃ dhiyas tvayi puras tvayi pūs tvam ādyā |
śaktiḥ parā tvam asi śaktiguṇās tavaite
kiṃ vistareṇa nanu sarvam idaṃ tvam eva ||35||
a. ājñā] em.; ājnā Ms. b. pūs tvam] conj.; pus tam Ms. c. tavaite] em.; tavete Ms.
You are the goddess of prosperity, and prosperities depend on you. You are
the goddess of speech, and authority and words depend on you. You are
the goddess of wisdom, and wise ideas depend on you. You are the fore-
most fortress, and towns depend on you. You are the primordial power,
and yours are all the properties of power. What is the use of any further
explanation: this entire world is nothing but you.
tvām āvadanti munayas tamaso nihantrīm

āhlādinīṃ smṛtimatām amṛtaṃ duhānāṃ |
nityoditām anuditāstamayaprasaṅgām
antaścarīṃ śaśikalām akalaṅkaśaṅkām ||36||
a. tamaso nihantrīm] em.; tamasā nihandrīm Ms. d. antaścarīṃ] conj.; mantaścarī Ms.
30 A number of older texts, including the Śāṃkarabhāṣya (on Brahmasūtra 2.2.42), state that
the Pāñcarātrikas identify Vāsudeva as Parā Prakṛti, the supreme cause (for a discussion on
Vāsudeva as Parā Prakṛti, see Watson, Goodall and Sharma 2013, 30–31, 241–246). However,
according to a verse attributed to the Skandapurāṇa and cited in the Bhagavatsandarbha
of Jīva Gosvāmin, Śrī is Parā Prakṛti, the consciousness associated with Viṣṇu (śrīḥ parā
prakṛtiḥ proktā cetanā viṣṇusaṃśrayā |; Bhagavatsandarbha, p. 278). It is possible that our
via free access
380 acharya
Sages address you as the destroyer of darkness, the bestower of delight,

yielding the immortal nectar to all those who remember you. They ad-
dress you as the ever-risen one with no possibility of rising and setting, as
the underlying digit of the moon31 never suspected to have a stain.
tvattejasā tapati devi patir dinānām

āpyāyayaty api karair amṛtaṃ himāṃśuḥ |
prāṇās tapanta iha vāyuvaśāc charīre
tvām antareṇa na hi kasya cana pravṛttiḥ ||37||
a. tvattejasā] em.; tvattojasā Ms. ● patir di°] em.; pati ddi° Ms. b. āpyāyayaty] em.; āpyā-
payaty MSpc, āpyādapayaty MSac (unmetrical) c. prāṇās tapanta iha] conj.; prāṇā tapan nta
iti Ms. ● charīre] em.; charāre Ms. d. cana pravṛttiḥ] em.; jana pravṛrttiḥ Ms.
O Goddess! With your energy the sun burns, the moon expands the
immortal essence with his beams, and here in our body the vital func-
tions glimmer under the control of the vital air. For, without you none
can function at all.
[lokāś caturdaśa mahendramukhāś ca devāḥ

mūrtitrayaṃ munigaṇāś ca vasiṣṭhamukhyāḥ |
sadyo bhavanti na bhavanti samastamūrter
unmīlanena tava devi nimīlanena|| 37a ||]32
poet is telling us about two different identifications: Nityā as Lakṣmī, the consort of Viṣṇu,
and Nityā as Parā Prakṛti Vāsudeva, the supreme cause of both sentient and insentient
beings.
31 The underlying digit of the moon (antaścarī śaśikalā) in all likelihood is the sixteen inner-
most digit beyond the waning and waxing process.
32 The late Pundit Vraja Vallabha Dwivedi (1985, 45) presents this verse in his preface (origi-
nally written in 1968) to the Nityāṣoḍaśikārṇava as cited in the Aruṇāmodinī commentary
of the Saundaryalaharī and attributed to the Kāmasiddhistotra of Vatsarāja (cf. Śāstrī
1957, 221), and suggests that it should be located in the Nepalese palm-leaf manuscript
of the text (the same manuscript I am editing now). However, in 1983 in the Luptāga-
masaṃgraha, a collection of citations from lost Āgamic texts he prepared, he writes that
the verse is not found in the palm-leaf manuscript and so must come from a different
text (cf. Dwivedi 1983, 25). I think Dwivedi arrived at this conclusion without reading the
implied name of the stuti. The author of the Aruṇāmodinī writes that it is a verse from
the Kāmasiddhistotra of Vatsarāja, and the same name is alluded to in the last verse of our
text. I conclude that the verse therefore belongs to this text even though it is not found in
the palm-leaf manuscript. I assume that it was dropped in the process of transmission. It
is thus just possible that there are still a few more verses missing from the latter part of
the stuti.
via free access
The fourteen worlds, all gods headed by Mahendra, the three embodi-
ments [of the ultimate reality], and also the groups of sages headed by
Vasiṣṭha, come into existence or cease to exist, O goddess, by the opening
and closing of your eyes, because you embody all.
vedā vibhinnagatayo viduṣaḥ svatantrās

tantrāṇi mantranivahā mahitaprabhāvāḥ |
bhāvā vibhaktiviṣayāḥ kavigumphanāś ca
mātaḥ paraṃ pariṇamanti tavāyutāṃśāt ||38||
b. prabhāvāḥ] em.; prabhāvoḥ Ms. c. ca] em.; va Ms. d. °ṇamanti] em.; °namanti Ms.
The Vedas, independent scholars of different capabilities, the Tantras, the

collection of mantras with celebrated powers, and thoughts and feelings
concerning syntax and grammar (vibhaktiviṣaya) and poetic composi-
tions, all these, O mother, evolve to excellence from a millionth part of
you.
yas te vibhūni paramapravilambitātmā

buddhyā pramitsati pumān puruṣaḥ pramāyāḥ |
saṃspṛśyati33 sphuṭam asau bhuvane paṭīyān
cchāyāṃ svakīyaśirasaḥ svapadakrameṇa ||39||
b. pramitsati pumān] em.; pramicchati pumāna Ms. (unmetrical) b. svapadakrameṇa] em.;

svayadakramena Ms.
Any person who is hanging on to the ultimate [reality of yours] (parama-

pravilambitātmā) and wishes to perceive and measure your proportions
(vibhūni)34 with his [limited] mind, that man of perception (pramāyāḥ
puruṣaḥ) smarter [than everyone] in this world evidently touches the
shadow of his own head in a series of his own footsteps.35
abhyarcya devi bhavatīṃ vibhavāmi bhūmim

āmnāyadarśitapathena yathā-kathañcit |
33 Our poet is using the root spṛś as if it belongs to the fourth class. We cannot emend it to
saṃspṛśati, because that would be unmetrical.
34 The word vibhu is treated here clearly as a noun.
35 This reminds me of a verse attributed to the now lost Trikahṛdaya and cited by Kṣemarāja
in his commentaries (e.g., Śivasūtravimarśinī, p. 9): svapadā svaśiraśchāyāṃ yadval laṅghi-
tum īhate | pādoddeśe śiro na syāt tatheyam baindavī kalā ||.
via free access
382 acharya
vācoratair atibhavābhidhakanyakāyā
vāllabhyavartmani janair na vigaṇyate kaḥ ||40||
c. atibhavābhidha°] conj.; adhibhavāvidha° Ms. d. °vartmani janair na] em.; °vatmani

janai nna Ms.
O goddess! Having worshipped you I somehow manage to govern [my]

land following the path dictated by the scriptures. [But] who among
the people on the path of winning the love of the young lady called
Atibhavā—the one who has transcended the world—is not disrepected
by those people who are engaged in gossip?
sā tvaṃ samāhitadhiyo hṛdayaṅgamāsi

gāyanti gauri madhurās tava kīrtigāthāḥ |
hālāsugandhiharicandanavāṭikāsu
vidyādharā vibudhasindhutaṭasthalīṣu ||41||
a. sā tvaṃ samāhitadhiyā] conj.; sā ddhaṃ samāhitadhiyo Ms.
O goddess! You enter the heart of a man whose mind is composed. Sweet
ballads of your renown, O Gaurī, the vidyādharas sing in the groves of
Haricandana trees that emit the sweet fragrance of liquor on the banks of
the heavenly river.
devi tvadīyamahasā mahitā mahānto

bhālekṣaṇāḥ śaśibhṛto bhujagendrabhūṣāḥ |
siddhāntasiddhiparamārthavidhiprasiddhāḥ
siddhāḥ sudhāṃśuvadane śivatāṃ vrajanti ||42||
a. tvadīya°] em.; tadīya° Ms. ● mahānto] em.; mahāmbho Ms. b. bhāle°] em.; bhālo° Ms.
● °bhūṣāḥ] em.; °bhūṣaḥ Ms. d. sudhāṃśuvadane śivatāṃ] em.; sudhāśuvadane śivantā
Ms. (unmetrical)
O goddess, those great people who are honoured with your greatness
achieve Śivahood as they attain perfection. They are the people who have
attained perfection following the regimen prescribed in the [system of]
Siddhānta for the purpose of supernatural powers as well the ultimate
goal [of liberation]. O moon-faced [goddess, they] bear [the characteris-
tics of Śiva]: the third-eye on the forehead, the moon on the head, and the
ornaments of serpents.
via free access
dṛṣṭiprasādalavalābhavatāṃ tavāmba
pādau namaty amararājagaṇo ’pi puṃsām |
prottālamaulimaṇimaṇḍitapādapīṭhāḥ
sarve kim aṅga caturabdhivasundharendrāḥ ||43||
a. °lavalābhavatāṃ] em.; °lavālābhavatāṃ Ms. (unmetrical) c. prottālamauli°] em.;

prottvālamoli° Ms.
O mother! Even the kings of gods bow to the feet of those men who have
acquired a drop of the grace of seeing you. Kings of all the rich lands
extending to the four oceans [bow to them] all the more, illuminating
their footrests with the studded jewels of their elevated crowns.
tvāṃ kledanīti kulakuṇḍalinīti keti

nityeti nītir iti naur iti nāviketi |
vidyeti saṃvid iti viśvamayīty umeti
kāmeśvarīti kamaleti vadanti santaḥ ||44||
a. tvāṃ] em.; tvā Ms. b. nītir iti] conj.; nityar iti Ms. ● naur iti] conj.; gaurīti MSpc
(unmetrical), gaur iti MSac d. vadanti] em.; vadānta Ms. ● santaḥ] bottoms damaged
Mindful men call you Kledanī, Kulakuṇḍalinī, Kā, Nityā, Nīti, Nau,36 Nā-
vikā, Vidyā, Saṃvid, Vīśvamayī, Umā, Kāmeśvarī,37 and Kamalā.
ekām anuttarakalāṃ kulanāthakāntāṃ

bālām atulyamahasaṃ bahumaṅgalāḍhyām |
ānandamūrtim akhilārthavilāsahetum
ādyāṃ smarāmy avirataṃ paramārthavidyām ||45||
a. °kalāṃ] em.; °kalā Ms. b. bālām atulya°] em.; bālātulya° Ms. (unmetrical) ● °bahumaṅ-
galāḍhyām] conj.; °bahu[pa]lāḍhyā MSpc, °bahulāḍhyā MSac (both unmetrical) cd. °hetum
ādyāṃ°] em.; °hetuṃm ādyāṃ Ms.
36 The original reading of the manuscript gaurīti is unmetrical. The scribe has corrected it
to gaur iti which is just possible, but I conjecture naur iti because of the following word
nāviketi. Thus, also, the syllabic rhyme of the line is restored.
37 Thus, there are three deities in this tradition who can be called by this name: the chief
goddess Nityā, one of the goddesses in the central triangle, and one of the goddesses of
speech in the retinue of eight triangles.
via free access
384 acharya
I uninterruptedly remember [you], the Vidyā leading to the ultimate well-

being, embodiment of bliss, the cause of the extension of all prosperities.
[You are] the primordial one, the insurpassable Kalā. You are Bālā, the
beloved of Kulanātha (namely, Śiva). [Your] glory is incomparable, and
you are filled with many felicities.
yaḥ kāmasiddhim atimaṅgalakāmadhenuṃ

kāmeśvarīstutim imāṃ paṭhati pratītaḥ |
kāntyā śriyā kavitayā guṇasampadā ca
so ’yaṃ svayaṃ vivṛta eva kimu priyābhiḥ ||46||
a. kāmasiddhim ati°] em.; kāmasiddhir iti MSpc, kāmaḥsiddhir iti MSac (unmetrical)
One who recites this eulogy of Kāmeśvarī called Kāmasiddhi, which

serves as a very auspicious wish-fullfilling cow, placing trust [in her], is
specially chosen (vivṛta) by [the goddesses of] Beauty, Prosperity, Elo-
quence, and Treasury of Qualities. So, what would he do with any [other]
lovers?
iti śrīmahārājādhirājavidyādharacakravartivatsarājaviracitā śrīvāmakeśvarī-

stutiḥ samāptā ||⊗||
Here ends the Vāmakeśvarīstuti composed by Vatsarāja, the king of great kings,
the sovereign among the vidyādharas.
References
Primary Sources
Adhikaraṇasārāvalī of Vedāntadeśika. Kanchi P.B. Annangaradharyar, ed. Vedāntade-
śikagranthamālā. Madras: Sannidhi Press, 1940.
Aruṇāmodinī. See Śāstrī 1957.
Niśvāsaguhya. The Guhyasūtra of the Niśvāsatattvasaṃhitā. NAK MS 1–227. (The Guhya-
sūtra starts on folio 41). NGMPP Reel No.A 41/14. Palm-leaf, Folios 117, early Nepalese
(Kuṭilā) script. There are two apographs available, both in Devanagari and on paper:
NAK MS 5–2401, NGMPP Reel No. A 159/18, and Welcome Institute for the History of
medicine, London, Sanskrit MS I.33. The verse and chapter numeration used in this
paper is that of Dominic Goodall’s edition in progress.
Tantrāloka. Mukund Ram Shastri, ed. The Tantrāloka of Abhinavagupta With commen-
tary of Rājānaka Jayaratha, vol. 1. Kashmir Series of Texts and Studies, no. 23. Śrīna-
gara: 1918. Reprint, Delhi: Motilal Banarsidass, 1987.
via free access
Brahmāṇḍapurāṇa. J.L. Shastri, ed. Delhi: Motilal Banarsidass, 1973.

Bhagavatsandarbha of Jīva Gosvāmin. Haridāsa Śāstrī, ed. Bhāgavatasandarbhe dvi-
tīyaḥ śrībhagavatsandarbhaḥ. Vṛndāvana: Gadādhara Gaurahari Press, 1984.
Vāmakeśvaratantra. Vraja Vallabha Dwivedi, ed. Nityāṣoḍaśikārṇava with Two Com-
mentaries Ṛjuvimalā by Śivānanda & Artharatnāvalī by Vidyānanda. Second Edition.
Benares: Sampūrṇānanda Sanskrit University, 1985.
Śivasūtravimarśinī. J.C. Chatterji, ed. The Śiva Sūtra Vimarshinī, being the Sūtras of Vasu
Gupta with the Commentary called Vimarshinī by Kshemarāja. Kashmir Series of
Texts and Studies, no. 1. Bombay: Nirnaya Sagar Press, 1911.
Secondary Sources
Bautze-Picron, Claudine 2002. “‘Nidhis’ and Other Images of Richness and Fertility in
Ajaṇṭā.” East and West 52 (1): 225–284.
Dalal, Chimanlal. 1918. Kālañjarādhipatiparamarddidevāmātyakavivatsarājapraṇītarū-
pakaṣaṭ[k]am. A Collection of Six Dramas of Vatsarāja. Gaekwad Oriental Series,
no. 8. Baroda: Central Library.
Dwivedi, Vraja Vallabha, ed. 1983. Luptāgamasaṃgraha, part II. Yogatantra Grantha-
mālā, no. 10. Benares: Sampūrṇānanda Sanskrit University.
Dwivedi, Vraja Vallabha. 1985. See Vāmakeśvaratantra.
Golovkova, Anna. 2012. “Śrīvidyā.” In Brill’s Encyclopedia of Hinduism, vol. IV, edited by
K.A. Jacobson, 815–822. Leiden: Brill.
Goodall, Dominic, Nibedita Rout, R. Sathyanarayanan, S.A.S. Sarma, T. Ganesan, and
S. Sambandhaśivācārya. 2005. The Pañcāvaraṇastava of Aghoraśiva: A twelfth-cen-
tury South Indian Prescription for the Visualisation of Sadāśiva and his Retinue. Col-
lection Indologie, no. 102. Pondicherry: Institut Français d’Indologie/École française
Goodall, Dominic, in collaboration with Alexis Sanderson and Harunaga Isaacson
with contributions of Nirajan Kafle, Diwakar Acharya and others. 2015. The Niś-
vāsatattvasaṃhitā. The Earliest Surviving Śaiva Tantra. Volume 1. A Critical Edition
and Annotated Translation of the Mūlasūtra, Uttarasūtra and Nayasūtra. Collec-
tion Indologie, no. 128 (Early Tantra Series, no. 1). Pondicherry: Institut Français
d’Indologie/École française d’Extrême-Orient/Universität Hamburg.
Paranavitana, S. 1955. “Śaṃkha and Padma.” Artibus Asiae 18 (2): 121–127.
Sanderson, Alexis. 2009. “The Śaiva Age: An Explanation of the Rise and Dominance of
Śaivism during the Early Medieval Period.” In Genesis and Development of Tantrism,
Tokyo.
Śāstrī, Anantakṛṣṇa. 1957. Saundaryalaharī of Śrī Śaṃkarācārya with Commentaries:
Saubhāgyavardhanī of Kaivalyāśrama, Lakṣmīdharā of Lakṣmīdharācārya, Aruṇā-
modinī of Kāmeśvarasūrin. Edition, English Translation and Notes. Madras: Ganesh
& Co.
via free access
386 acharya
Sinopoli, Carla M. 2010. “Echoes of Empire: Vijayanagara and Historical Memory, Vija-
yanagara as Historical Memory.” In Contemporary Archaeology in Theory: The New
Pragmatism, edited by Robert W. Preucel and Stephen A. Mrozowski, 17–33. New
York: John Wiley & Sons.
Sircar, Dines Chandra. 1983. Select Inscriptions bearing on Indian History and Civilization
From the Sixth to the Eighteenth Century A.D. Volume II. Delhi: Motilal Banarsidass.
Watson, Alex, Dominic Goodall, and S.L.P. Anjaneya Sarma. 2013. An Inquiry Into the
Nature of Liberation. Bhaṭṭā Rāmakaṇṭha’s Paramokṣakārikāvṛtti, a Commentary on
Sadyojyotiḥ’s Refutation of Twenty Conceptions of the Liberated State (mokṣa). Col-
lection Indologie, no. 122. Pondicherry: Institut Français d’Indologie/École française
via free access
chapter 16
The Lotus Garland (padmamālā) and Cord of

Power (śaktitantu): The Brahmayāmala’s
Integration of Inner and Outer Ritual
Shaman Hatley
The significance of the Brahmayāmala to the history of Śaivism was first iden-
tified by Alexis Sanderson (1988), who was gracious enough to read a section of
this voluminous text with me at Oxford, in 2004, at an early stage of my doctoral
research. This was a formative professional experience, and I remain touched
by his generous hospitality towards me as an unknown visiting student. In the
present essay, I revisit a particular thread which emerged from these reading
sessions: the śaktitantu or śaktisūtra, the “thread” (tantu) or “cord” (sūtra) of
divine power (śakti). This is a technical term of ritual distinctive to the Bra-
hmayāmala to which Professor Sanderson first drew my attention. Inquiry into
the Cord of Power leads me to examine the ways in which the Brahmayāmala
(hereafter BraYā) integrates meditational discipline with the somatic perfor-
mance of ritual, and to query its understanding of the category yoga.
In chapter 15 of the Tantrāloka, Abhinavagupta invokes the authority of
the BraYā concerning the inseparability of “external” (bāhya) and “inner” or
“internal” (adhyātma) worship. The passage (15.43cd–44) reads as follows in the
printed (Kashmir Series of Texts and Studies) edition:
nādhyātmena vinā bāhyaṃ nādhyātmaṃ bāhyavarjitam ||43||

siddhyej jñānakriyābhyāṃ tad dvitīyaṃ saṃprakāśate |
śrībrahmayāmale deva iti tena nyarūpayat ||44||
The fourth verse-quarter (44b) is problematic; following the commentator

Jayaratha,1 the passage might be interpreted as meaning,
“Not without the inner (adhyātma) would the outer succeed, nor the
inner devoid of the outer. The second [i.e. the inner] manifests through
1 Tantrālokaviveka ad 15.43c–44: tad dvitīyam iti adhyātmalakṣaṇam, “‘the second one’ means
‘characterized by being inner (adhyātma)’.”
© shaman hatley, 2020 | doi:10.1163/9789004432802_018

via free access
388 hatley
both gnosis and ritual action”—the Lord has explained accordingly in the
revered Brahmayāmala.
Both the sense and syntax appear doubtful, however. On the grounds of coher-
ence, it seems significantly better in 44b to read tad dvitayaṃ (“that pair [of
inner and outer]”) for tad dvitīyaṃ (“the second one”), an emendation sug-
gested by Harunaga Isaacson.2 In this case the passage might be interpreted
as follows:
“Not without the inner (adhyātma) would the outer succeed, nor the inner
devoid of the outer. That pair [of inner and outer] manifests through gno-
sis and ritual action, [respectively]”—the Lord has explained accordingly
in the revered Brahmayāmala.
Thus it seems that Abhinavagupta places the dichotomy of outer (bāhya) and
inner (adhyātma) worship in correlation with that of kriyā and jñāna: ritual
action and gnosis, respectively. This accords with his remarks prefacing cita-
tion of the BraYā, which clarify that the performative acts of ritual are valuable
only as ancillaries to śivābhimānatā, conviction of one’s identity with Śiva. This
inner conviction alone, a form of knowledge, is the real means of liberation.3
One need not follow Abhinavagupta entirely in reading the BraYā’s dichot-
omy between ‘inner’ and ‘outer’ processes in terms of gnosis and ritual: his
source scripture draws no such distinction overtly. Nonetheless, Abhinava-
gupta’s selection of the Brahmāyamala is by no means contrived, for in invok-
ing the text thus he highlights a premise central to its systems of practice: the
integration of internal and external ritual processes, which mirror each other
closely.
Although many of Abhinavagupta’s citations of the BraYā may be identified
precisely,4 in this case the passage he had in mind remains uncertain. BraYā
87.140 is perhaps the strongest candidate:
2 Personal communication, 2003.

3 Tantrāloka 15.42–43b: śivābhimānitopāyo bāhyo hetur na mokṣadaḥ | śivo ’yaṃ śiva evāsmīty
evam ācāryaśiṣyayoḥ || hetutadvattayā dārḍhyābhimāno mocako hy aṇoḥ | (“Conviction of
one’s identity with Śiva is the means; no external cause bestows liberation. ‘He is Śiva, and
I am nothing but Śiva’: [in the rite of initiation,] this firm conviction of the guru and disciple
is indeed what liberates a soul, by way of [the guru’s conviction] being the cause and [the
disciple’s] the effect (? hetutadvattayā)”).
4 See Hatley 2018, Appendix B.
via free access
the lotus garland and cord of power 389
ādhyātmañ cintayed bāhyaṃ bāhyam adhyātmikāṃ tathā |

cakre samānabhāvena tato vinyāsam ārabhet ||140||
One should meditate upon the internal [wheel/cakra] as external, and

the external likewise as internal. Considering [these] to be identical, one
should then commence installation [of the mantra-deities] on the cakra.
The degree of correspondence is merely suggestive, and Abhinavagupta might

have had in mind other BraYā passages. His language is actually somewhat
closer to a verse from the Tantrasadbhāva, however, another early scripture
of the Vidyāpīṭha division of Bhairavatantras. This is highly likely to postdate
the BraYā and to reflect a degree of historical development vis-à-vis the latter.5
Tantrasadbhāva 6.218 may read as follows:
adhyātmaṃ bāhyato dṛṣṭvā bāhyam adhyātmasaṃsthitam |

yo jānāti sa siddhyeta tadbhāvabhāvabhāvitaḥ6 ||218||
After viewing the inner (adhyātma) externally, one who [also] knows the
outer to be located internally will attain success, purified by contempla-
tion of their identity (?).7
It is conceivable that Abhinavagupta has conflated these two sources, though

he may instead have had in mind a verse of the Brahmayāmala not preserved
in its oldest manuscript. More probably, he simply paraphrases his source with
unusual freedom.
Regardless of which passage Abhinavagupta had in mind, a point of signifi-
cant interest emerges from considering this question, namely that the Tantrā-
loka, Tantrasadbhāva, and Brahmayāmala all appear to mean something differ-
ent by the distinction between adhyātma, “inner/directed towards the self,” and
bāhya, “external.” For the Tantrasadbhāva, the performative acts and parapher-
nalia of ritual have superior, inner (adhyātma) equivalents: the outer (bāhya)
finds its true meaning in the inner processes of yoga. Thus, for instance, in the
section of Tantrasadbhāva 6 quoted above in excerpt, the external rosary made
5 See the discussion in Hatley 2018, 100–104.

6 For the full passage with critical apparatus, see note 8 below.
7 Tantrasadbhāva 6.218d—tadbhāvabhāvabhāvitaḥ (or tadbhava°)—is unmetrical and prob-
lematic; I have understood tad as tayoḥ, referring to the pair of adhyātma and bāhya (inner
and outer).
via free access
390 hatley
of conch (śaṃkha) has as its inner form the subtle śaṃkhā or śaṃkhiṇī channel
(nāḍī) of the yogic anatomy, which extends into the crown of the head.8 The
‘outer’ thus finds true purpose in the ‘inner’ realities of yoga, which supersede
the outer entirely. This principle applies to sacred geography as well, for the
Tantrasadbhāva transposes into the yogic body a series of cremation grounds
closely related to those of the BraYā.9 Access to these inner levels of meaning
is a form of knowledge: the inner sacred geography is understood ( jñātavyaṃ)
by those who know the self (viditātmanāṃ), while only “one who knows [the
identity of inner and outer] attains success” ( yo jānāti sa siddhyeta). Knowl-
edge ( jñāna) thus transforms ritual into an inner process through which it
8 Tantrasadbhāva 6.217–220b (based on the draft edition of Marc Dyczkowski, which reports
the readings of three manuscripts, k, kh, and g):
ete cāṣṭādaśa proktā akṣasūtrā bahisthitāḥ |
adhyātme ca gatās tv ete sāṃprataṃ tān nibodha me ||217||
adhyātmaṃ bāhyato dṛṣṭvā bāhyam adhyātmasaṃsthitam |
yo jānāti sa siddhyeta tadbhāvabhāvabhāvitaḥ ||218||
ekaivādhyātmagaṃ sūtraṃ prāg eva kathitaṃ mayā |
śaṃkhāvartā tu yā nāḍī śikhānte tu vyavasthitā ||219||
tena śaṃkhamayī proktā akṣasūtraṃ varānane |
217b bahisthitāḥ] em.; vahisthitā k, kh, g 217c tv ete] k, g; tvate kh 218a adhyātmaṃ] corr.;
adhyātma mss. 218b °saṃsthitam] em. (Dcyzkowski); saṃsthitā mss. 218d tadbhāva°]
k; tadbhava° kh, g 219b prāg eva] kh; prog eva k, g 219c śaṃkhāvartā] kh; saṃkhyāvartā
k, g 219d śikhānte tu] k, g; śikhāntera kh 220a śaṃkhamayī] k, g; śaṃkhamayaṃ kh
220b akṣasūtraṃ] em. (Dyczkowski); makṣasūtraṃ mss.
On śaṃkhāvartā as a nāḍī, see Amaraughaśāsana 60, which describes it as “having the mea-
sure of a lotus fibre” (mṛṇālasūtraparimāṇā). The more commonly attested name is śaṃkhiṇī.
9 See especially Tantrasadbhāva 15.62–67:
prayāgā nābhisaṃsthā tu varuṇā hṛtpradeśata[ḥ] |
kolagiryāṃ tu kaṇṭhasthaṃ bhīmanādaṃ tu tāluke ||65||
bindusthāne jayantyāṃ tu nādākhye tu caritrakam |
ekāmrakaṃ śaktimadhye jñātavyaṃ viditātmanām ||66||
guruvaktragataṃ proktaṃ koṭivarṣaṃ tathāṣṭamam |
ete sthānā mayā proktā adhyātme pudgalāśrayāḥ ||67||
For those who know the Self, Prayāga should be understood as located in the [cakra of the]
navel, Varuṇā [i.e. Vārāṇasī] in the heart region, Kolagiri in the throat, Bhīmanāda in the
palate, Jayantī in the place of Bindu, Caritra in [the plexus] called Nāda, and Ekāmraka in
[the plexus of] Śakti. The eighth, Koṭivarṣa, is likewise said to be in the Mouth of the Guru.
These are the places I have declared to be present in the person internally (adhyātme).
This list of eight pīṭhas overlaps with the nine śmaśānas or pīṭhas of the BraYā’s principal
maṇḍala (as outlined in chapter 3; see Kiss 2015, 24); however, it corresponds more precisely
to the eight delineated in BraYā 87 (see Hatley 2018, 134, table 1.16). Cf., also, Tantrasadbhāva
15.70: kulūttamoḍḍiyānaṃ ca eruḍī pulivallabham | tāny eva tu samastāni [em.; saman tāni
mss.] svadehe saṃsthitāni tu ||.
via free access
ultimately achieves fruition—an orientation consonant with the Kaula turn

towards interiority, which may presage aspects of Abhinavagupta’s gnostic
nondualism. Yet in contrast to the latter, for the Tantrasadbhāva, the inner
is nonetheless conceived of in terms of ‘doing’ (kriyā) rather than ‘knowing’
( jñāna). The inner and outer, respectively, represent higher and lower arenas of
activity, for the mental processes of meditation are also forms of action (kriyā).
Chapter 26 of the Tantrasadbhāva colourfully proclaims this principle:
kriyā tu phaladā10 puṃsāṃ na vijñānaṃ phalapradam11 ||19||

yathā strībhakṣyabhogajño na sukhī kriyayā vinā |
kriyā tu dvividhā jñeyā bāhyā adhyātmikā smṛtā12 ||20||
adhyātmā dhyānayogena bāhyā13 pūjāvratādibhiḥ14 |
Doing (kriyā) is what gives people results; knowledge does not produce
results, just as a man knowledgable in the sexual enjoyment of women
is not happy without doing it (kriyā). But doing should be understood as
twofold: it is held to be outer and inner. Inner action (kriyā) is through
yogic meditation, while outer action is through worship, ascetic obser-
vances, etc. …
Knowledge is here conceptualized neither in processual nor gnostic terms:

jñāna and kriyā represent a dichotomy along the lines of “theory” and “prac-
tice.”
The BraYā likewise embraces the twofold division of kriyā into inner and
outer—adhyātma and bāhya—which refer in its usage to subjective processes
performed mentally and to the somatic performance of ritual, respectively. In
contrast to the Tantrasadbhāva, the BraYā draws no overt hierarchy between
the two. The inner does not represent a higher meaning or reality; the inner
and outer rather represent complemetary arenas for the performance of ritual
whose integration is essential. The inner and outer find their fullest integration
in the performance of yāga or yajana, “pantheon worship,” the paradigmatic
mode of ritual whose limitless possibilities for inflection account for much of
the BraYā’s bulk. Yet the principle of their unity applies widely, extending to all
10 phaladā] em.; phaladāḥ mss.

11 Cf. Śrīkaṇṭha’s Ratnatrayaparīkṣā, 129ab: kriyā hi phaladā puṃsāṃ na jñānaṃ syāt pha-
lapradam.
12 smṛtā] em.; smṛtāḥ mss.
13 bāhyā] em.; bāhya mss.
14 °vratādibhiḥ] k, kh; vratedibhiḥ g.
via free access
392 hatley
of the various practices encompassed by the terms kriyā and karman, includ-
ing fire ritual (homa), rites of installation (pratiṣṭhā), initiation (dīkṣā), dura-
tional observances (vrata), rites using deity images (pratimākarman) or two-
dimensional diagrams ( yantrakarman), and so forth.15 This integration of inner
and outer practice makes a rigid distinction between “ritual” and “meditation”
artificial: in the context of the BraYā, ritual—kriyā or karman—encompasses
both outer and inner forms of action. These are remarkably parallel in structure
and process, and aim at the unification of subjectivity, body, ritual space, para-
phernalia, and the hierarchy of ontic principles (tattva) which comprise the
cosmos (adhvan). In this respect, despite its cultic affinity to later Kaula sys-
tems, the BraYā’s harmonious integration of the inner and outer seems largely
congruent with classical Śaivasiddhānta ritualism. Arguably, such integration
of inner and outer processes is characteristic of tantric ritual, broadly con-
ceived, as reflected in the ubiquitous dichotomy of antaryāga (“inner worship”)
and bahiryāga (“external worship”). What is most distinctive to the BraYā is its
unique manner of their integration: the “method of the lotus garlands” (padma-
mālāvidhi).
1 Method of the Lotus Garlands (padmamālāvidhi)
Patterning the processes of inner and outer ritual is the BraYā’s pantheon of
mantra-deities, whose core comprises the Four Goddesses (devī) or Guhyakās,
Four Consorts (dūtīs) or Handmaidens (kiṅkarī), and their lord, Kapālīśa-
bhairava. Secondary members of the pantheon are a sextet of Yoginīs and an
octad of Mother-goddesses (mātṛ). These (see table 16.1) are the mantra-ele-
ments manipulated in all practice, both inner and outer, and their permuta-
tions are the principal ritual variables. Collectively, the deities comprise the
Nine-Syllable Vidyā (navākṣarā vidyā), mantra of the supreme goddess, Caṇḍā
Kāpālinī, as first identified by Sanderson (1988, 672) in his pioneering remarks
on the BraYā. In syllabic terms, the vidyā is [oṃ] hūṃ caṇḍe kāpālini
svāhā.
15 Cf. BraYā 90.101: anena vidhinā devi japahomādikarmasu [corr.; karmaṣu ms.] | bāhyā-
dhyātmeva mantrajñaḥ pūjāṃ [em.; pūjyāṃ ms.] kurvan prasidhyati ||101|| (“Through this
procedure, O goddess, in mantra incantation, fire sacrifice, and other rites, the knower of
mantras achieves siddhi, practicing both external and internal worship”). In 101c, bāhyā-
dhyātmeva should be understood in the sense of bāhyādhyātmām eva—nominative for
accusative, metri causa.
via free access
table 16.1 Deities of the Nine-Syllable Vidyā
Kapālīśabhairava (hūṃ)
four devīs four dūtīs
Raktā (caṃ) Karālā (li)

Karālī (ḍe) Danturā (ni)
Caṇḍākṣī (kā) Bhīmavaktrā (svā)
Mahocchuṣmā (pā) Mahābalā (hā)
six yoginīs eight mātṛs
Kroṣṭukī (oṃ hūṃ caṇḍe Maheśvarī (aṃ)

kāpālini namaḥ) Brāhmī (e)
Vijayā (oṃ … svāhā) Vaiṣṇavī (ā)
Gajakarṇā (oṃ … hūṃ) Kaumārī (a)
Mahāmukhī (oṃ … vauṣaṭ) Vaivasvatī (ī)
Cakravegā (oṃ … vaṣaṭ) Indrāṇī (i)
Mahānāsā (oṃ … phaṭ) Caṇḍikā (svā)
Paramā/Aghorī (hā)
Virtually all ritual begins with the instruction to enter into a state of med-
itative concentration, called nirācāra, and to take on a body of śakti, called
the avadhūtatanu.16 Śiva is the nirācārapada, “the state beyond regulated con-
duct,” while the Goddess is avadhūtā, “the stainless/unblemished one.”17 The
avadhūta-body, the body of śakti, is a body of mantra (mantravigraha) engen-
dered by the placement of mantra-syllables in a series of bodily lotuses. The
process of taking on the avadhūta-body culminates in assumption of one’s
inner identity as Kapālīśabhairava at the heart of a maṇḍala of goddesses, all
of whom collectively comprise the Nine-Syllable Vidyā. Although framed as a
16 Note, for example, BraYā 3.187c–189b: ekākī vijane tasmiṃ dakṣiṇābhimukhasthitaḥ ||187||
muktakeśaś ca digvāsaḥ kṛtanyāso vidhānavit | avadhūtanur bhūtvā nirācāras tu sādhakaḥ
||188|| prathamaṃ pūjayed devaṃ karṇṇikāyāṃ paraṃ śivaṃ |. Cf., e.g., 8.3–4b: tato hy
ekamanā bhūtvā avadhūtatanuḥ sthitaḥ | nirācāreṇa bhāvena yadā paśyati sarvvataḥ ||3||
tadā karoti karmāṇi vicitrāṇi mahītale |; cf. also 47.17c–18b: tataḥ ekāgracittas tu avadhūta-
tanusthitaḥ ||17|| nirācāreṇa bhāvena smared vidyāṃ suyantritaḥ |.
17 BraYā 2.2cd: avadhūtā tu sā śakti nirācārapadaḥ śivaḥ (= 62.98ab).
via free access
394 hatley
preliminary to ritual action, this process of mantra-installation (nyāsa) is in

fact the very template of ritual, both inner and outer; it has both elaborate and
abbreviated forms, with numerous inflections. I here summarize one of several
long descriptions of this process, the padmamālāvidhi (“method/procedure of
the lotus garlands”) of BraYā, chapter 4 (vv. 497ff.):
First one visualizes the avadhūtaṃ (i.e. the avadhūtā śakti), whose func-
tion in this application is analagous to the ādhāraśakti (“basal power”) of
most Śaiva systems.18 This begins preparation of the practitioner’s body
as locus for installation of the deities, from the crest (śikhā) of the head to
the feet. A series of nine lotuses is visualized situated at points in the body
called granthis (knots or joints). These are located at the crown of the
head (śikhā), the forehead (lalāṭa), throat (kaṇṭha), navel (nābhi), knees
( jānu), mouth (vaktra), heart (hṛd), genitals (guhya), and feet (pāda), fol-
lowing the order of their sequence in nyāsa. The eight-petalled lotuses
situated therein are loci for installation of the principal nine deities:
Kapālīśabhairava, who is installed in the crown lotus, and two sets of four
goddesses, the Devīs and the Dūtīs. In each lotus one prepares a mantric
seat for the deity, whose complete mantra-forms are then installed, inclu-
sive of their ancillary mantras (aṅga).19 Each of the eight goddesses is
installed in a lotus along with the mantra-body of Kapālīśabhairava, who
is thus coupled with each goddess.20 This series of nine forms the Brahma-
yāmala’s principal padmamālā, the garland of Devīs and Dūtīs, which is
illustrated in figure 16.1. While one might associate bodily lotuses with
the body seated in yogic meditation, here a standing position with the
legs together is implied, for a single lotus presides over the two knees and
likewise feet.
Next is installed a second series of seven lotuses, the garland of Yoginīs
(figure 16.2b).21 In contrast to the first lotus garland, these do not lie in a
vertical axis. Three form a kind of girdle: one lotus is placed in the center
18 Concerning the ādhāraśakti, see the article on this term in Tāntrikābhidhānakośa, vol. I.
19 BraYā 4.497c–501b: adhunā [em.; adhuna ms.] sampravakṣyāmi padmamālāvidhikramaṃ
[em.; °vidhiḥ kramaṃ ms.] ||497|| śikhādipādayor antaṃ [em.; antā ms.] avadhūtaṃ vi-
cintayet | navagranthivibhāgena tayā [?] proktā punaḥ kramāt ||498|| navapadmāni saṃ-
cintya aṣṭapatrāṇi sādhakaḥ | yuktāni keśaraiś caiva caturvvinsatibhiḥ kramāt ||499|| ka-
rṇṇikāyāṃ yutānīha cintanīyāni mantriṇā | śikhāpadmaṃ [em.; °padma ms.] samārabhya
āsanāni prakalpayet ||500|| vakṣyamānena nyāyena padme padme na saṃśayaḥ |.
20 BraYā 4.516–517b: vāmapārśve tatas tasyā kapālīśasya vinyaset | bījamātraṃ mahāprājña
vaktranetrāṅgavarjitam ||516|| tasyopari nyased devi bhairavaṃ mantravigraham |.
21 BraYā 4.523c–529.
via free access
of the waist, on the back (kaṭipṛṣṭhe), while the other two lie on either side
of the waist. The remaining four lotuses are situated on the sides of the
knees and feet. Installed upon these lotuses are goddesses known as the
Six Yoginīs, led by a male deity, Ādivīra (“Primordial Hero”), positioned in
the lotus on the back of the waist (see figure 16.2b).
A third series of lotuses serves as the locus for installation of the
Eight Mother-goddesses (aṣṭa mātaraḥ), accompanied by the male deity
Mātṛvīra (figure 16.2a).22 In this case as well the lotuses do not form a
vertical axis, lying at various positions on the head and torso: the points
between the eyebrows (bhrūmadhya) and between the eyes, the tip of the
nose, and each ear. Next are a point between the throat and heart, another
between the heart and navel, and an unspecified place on the back or
spine,23 locus of the goddess Carcikā or Cāmuṇḍā. The eighth mātṛ, the
supreme śakti, Paramā, also called Aghorī or Yogeśī, pervades the entire
body, lacking a lotus base and being devoid of ancillary mantras.24
In figure 16.2, the second and third lotus garlands are shown together, thus illus-
trating the empowerment of the upper and lower bodies by mantra.
2 The Cord of Power (śaktitantu, śaktisūtra)
What renders the lotuses into garlands (padmamālā) is the thread which binds
them. The BraYā first introduces this idea in presenting the second garland
(that of the Yoginīs), describing the lotuses as “bound together by the cord
of śakti, like gems [strung] by a cord” (śaktitantunibaddhāni sūtreṇa maṇayo
yathā, 4.526cd). The terms utilized are śaktitantu and śaktisūtra, meaning,
respectively, a thread or a cord of śakti. This divine power binding the lotuses
together is consubstantial with the supreme Goddess herself, the Nine-Syllable
Vidyā whose being encompasses the deities of the maṇḍala.
A concept seemingly unique to the BraYā, the śaktitantu or śaktisūtra re-
ceives minimal explication. References to it occur almost entirely in the con-
text of the “method of the lotus garlands” (padmamālāvidhi). In what manner
the cord connects the lotuses of the three garlands is somewhat ambiguous.
For the first garland, whose lotuses form a vertical series, the śaktitantu must
22 BraYā 4.530–538.
23 BraYā 4.532cd: aṣṭaman tu tathā pṛṣṭhe kalpayen mantravit kramāt.
24 BraYā 4.538: sarvvāṅge paramā śakti vaktranetrāṅgavarjitā | padmāsananvihīnā tu vinya-
sen mantravit kramāt ||.
via free access
396 hatley
figure 16.1 Garland of the Devīs and Dūtīs (padmamālā I)
via free access
figure 16.2 Garlands of the Mātṛs and Yoginīs (padmamālās II–III)
via free access
398 hatley
figure 16.3 The pure body of power (avad-

hūtatanu)
via free access
string them together vertically from crest to feet (figure 16.1). In the case of the
third padmamālā, the śaktisūtra likely begins from the lotus between the eye-
brows (bhrūmadhya), the uppermost of this series and locus of the male deity
Mātṛvīra, lord of the Mother-goddesses (mātṛ). The cord pierces (bhid-) and
thus strings together eight lotuses (figure 16.2a).25 Unlike those of the first gar-
land, these do not form a vertical axis in the body, and it is unclear precisely how
and in what sequence the śaktitantu links them together: does the cord form
a garland-like closed circuit, or connect the lotuses like a strand? This śakti-
tantu extends through the body in three dimensions, for the seventh lotus is
located on the back, forming the locus of Carcikā, i.e. Cāmuṇḍā.26 In the case
of the second padmamālā, that of the yoginīs, the manner in which the śakti-
tantu connects the lotuses seems less ambiguous (figure 16.2b). The garland is
threaded from a lotus on the back of the waist,27 extending outwards to lotuses
on the sides of the waist, then downwards to lotuses on the sides of the knees
and feet. Although this is not explicitly stated, the cord might connect the foot-
lotuses together in a garland-like manner, so forming a closed circuit.
The principal series of nine lotuses spans the body’s axis from crest (śikhā)
to feet, linked by the śaktitantu and thus forming a vertical strand (figure 16.1).
This vertical sequence of bodily lotuses connected by a cord of śakti has obvi-
ous similarities with models of the yogic body in which suṣumnā nāḍī, the
central channel, links together an ascending series of cakras or lotuses. This
ubiquitous paradigm is exemplified by the system of seven cakras common to
Śrīvidyā and Haṭhayoga, identified by Sanderson (1988, 687–688) as being first
attested in the Kubjikāmata. Like the śaktitantu, the suṣumnā is closely iden-
tified with śakti, especially in the form of the bodily kuṇḍalinī. The suṣumnā
of yoga and the BraYā’s śaktitantu both unite a vertical series of lotuses situ-
ated at particular “knots” (granthis) in the body, including such standard loca-
tions as the crest, forehead, throat, heart, navel, and genital region. While the
suṣumnā came to be envisioned as a vertical channel extending upwards from
the heart, navel region, or base of the torso,28 there is an old precedent for the
idea that it extends, like the śaktitantu, to the feet: the Mataṅgapārameśvara,
a comparatively early Siddhāntatantra, envisions the suṣumnā running from
25 BraYā 4.530: bhruvo madhye tataḥ padmaṃ śaktiśūtreṇa pūrvvavat | bheditaḥ -m- āditaḥ
(?) kṛtvā anyāny api tathaiva hi ||530||.
26 BraYā 4.536cd: pṛṣṭipadme tu devesi carccikāṃ vinyased budhaḥ.
27 BraYā 4.527c–529: vīran tu trikapadme tu kaṭipṛṣṭhe tato nyaset ||527|| vaktranetrāṅgasa-
ṃyuktaṃ vīraśaktyā samanvitaṃ | bījamātraniviṣṭāyāṃ vāmapārśve tathaiva ca ||528||
vāmakaṭyādim ārabhya yoginyo vinyaset tathā | vaktranetrāṅgasaṃyuktāḥ padme padme
na saṃśayaḥ ||529||.
28 See Tāntrikābhidhānakośa, vol. III, entries for piṅgalā, nāḍī, nābhi, and nābhikanda.
via free access
400 hatley
the tips of the big toes to the crown of the head via the navel and heart.29 This
archaic model of a central channel extending to the feet, linking together vari-
ous granthis, may underlie the BraYā’s conception of the śaktitantu.
It should perhaps be emphasized that the BraYā’s series of lotuses do not
in any simple sense represent “structures” of a subtle or “yogic” anatomy. The
lotus-seats (āsana) of the mantra-deities are created through meditative visu-
alization: one actively engenders a divine body of mantra rather than reifying
a subtle reality already latent in the body. More precisely, through the padma-
mālāvidhi, one imaginatively superimposes the mantric body (mantravigraha)
of Bhairava upon one’s own corpus and psyche. That the deity-lotuses do not
represent fixed structures of a subtle body is illustrated by variations in their
sequence: chapter 21 provides an alternative order of the principal nine deities,
placing Bhairava in the heart,30 while an inflection of the padmamālāvidhi in
chapter 45 inverts their typical sequence, placing Bhairava in the lotus of the
feet.31 An alternative version of the second garland has the lotuses of the Six
Yoginīs encircle the waist like a girdle.32
Nonetheless, while the BraYā does not treat the lotus garlands as fixtures of
a subtle body, it does posit the existence of such structures: the body’s channels
(nāḍī) and vital airs (vāyu, prāṇa), for instance, and more pertinently, points
known as granthis (“knots” or “joints”). A particular series of nine granthis
forms the locus for installing the nine lotuses of the primary padmamālā. That
granthis were considered to be anatomical realities is suggested by their treat-
ment as points in the body rather than as objects to be placed/installed (nyas-)
in the body or engendered through meditation (kalp-, cint-, etc.).33 Integral to
the idea of the padmamālā is this correlation between a vertical sequence of
29 Mataṅgapārameśvara, vidyāpāda 20.21c–22b: pādāṅguṣṭhāgrato vyaktā nābhito hṛdayaṃ

gatā || 21|| suṣumnā nāma sā jñeyā brahmarandhrābjanirgatā |.
30 BraYā 21.125ab: hṛdaye bhairavo nyasya vaktranetrāṅgasaṃyutaṃ (understanding bhai-
ravo as accusative in sense).
31 In BraYā 45.390–397, the sequence of installation (nyāsa) for inner worship is from feet
(Bhairava) to head (Paramā Śakti). See the notes of Kiss (2015, 281–282).
32 BraYā 21.128–129: nābhyām vai vīra vinyasya tato vāmādi-m-āśṛtāḥ | mekhalākṛtisaṃsthānā
kaṭyāṃ vai yoginī nyaset || 128|| dakṣiṇāpārśvakaṃ yāvad vāran tu varavarṇini | vaktrane-
trāṅgasaṃyuktā kartavyā nātra saṃśayaḥ ||129||.
33 The granthis are referred to solely by their locations in the body. Take for example BraYā
4.564c–565, which specifies the “knot of the throat” (kaṇṭhagranthi) as locus for the lotus
of goddess Karālī: anenaiva vidhānena karālīṃ [em.; karālī ms.] vinyaset punaḥ || kaṇṭha-
granthigate [em.; °sate ms.] padme nātra kārya vicāraṇāt | nābhipadme [em.; °padmes
ms.] tathaiveha caṇḍākṣīṃ [em.; °ākṣī ms.] vinyased budhaḥ ||. (Emendation of sate to
gate is trivial, given the frequent confusion of śa and ga in the old manuscript, and the
interchangability of śa and sa.)
via free access
lotuses and the body’s nine knots, which are repeatedly described as dividing
the garland.34 In the case of the two secondary padmamālās, the lotuses’ posi-
tions in the body do not on the whole correlate with granthis. This supports
the idea that the primary padmamālā is rooted in a model of the yogic body
in which nine granthis are united by a central channel (nāḍī), which the BraYā
reconceives of as the śaktitantu. This cord of śakti, unique to the BraYā, appears
to combine what in later traditions would be differentiated as the suṣumnā nāḍī
and the bodily (as opposed to cosmogonic) kuṇḍalinī.35
Although archaic tantric conceptions of the yogic body may have inspired
the BraYā’s lotus garlands and cord of power, the concept primarily describes
the practitioner’s assumption of a divine body of mantra. This avadhūtatanu,
“body of pure śakti,” is formed by lotuses of the three garlands and united by the
śaktitantu, shown as whole in figure 16.3. This embodies the entire pantheon of
deities as well as the hierarchy of ontic levels (tattvakrama), from the earth ele-
ment to paramaśiva.36 The avadhūtatanu taken on by the sādhaka mirrors, in
part, the mantra-body of the deity as Sadāśiva; his divine form is composed
of a garland of nine lotuses pierced by the cord of power, further augmented
by a triad of cosmological powers—vāmā, raudrī, and jeṣṭhā.37 To assume the
34 Note for instance BraYā 21.123c–124b: śaktitantunibaddhān tu padmamālāṃ [em.; mālā

ms.] vicintayet ||123|| navagranthivibhāgena tato nyāsaṃ prakalpayet | (“One should visu-
alize the garland of lotuses as bound by the cord of śakti, then one should perform
mantra-installation, according to the division of the nine knots [where are positioned the
lotuses]”).
35 Key to this issue is the question of whether and by what other names the BraYā conceives
of a central channel (nāḍī). While the BraYā does attest the common nāḍī triad of iḍā,
piṅgalā, and suṣumnā, the expression suṣumnā (which in most sources refers to the central
channel) occurs only once (18.80) in the earlier stratum of the BraYā—its first fifty chap-
ters. Its meaning in later chapters, where it does occur several times, is ambiguous. It may
not (always) be the middle channel: it is never specifically designated as such, and BraYā
85.138cd seemingly refers to an unmamed middle channel between iḍā and suṣumnā (iḍā-
suṣumnayor madhye muktimārggānusāriṇī). Concerning the early history of suṣumnā, see
the preliminary remarks of Goodall et al. (2015, 33–34).
36 Note e.g. BraYā 45.104ab: śaktitantu tato dhyātvā śivādyavanigocare. Similar expressions
abound in the BraYā.
37 BraYā 32.42–46:
evaṃsvarūpasaṃpannaṃ cintayīta sadāśivam |
evaṃsthito mahādevi ādidevaḥ sadāśivaḥ ||42||
svayaṃsthito sthitiṃ kuryā trailokye tu na saṃśayaḥ |
saṃhāraṃ tu sadā kuryād icchayā caiva saṃharet ||43||
dahyātmanaḥ punas sarvaṃ trailokyaṃ sacarācaram |
sṛjate tu yadātmānaṃ trailokyaṃ sṛjate sadā ||44||
vāmayā sṛjate sarvaṃ raudryā caiva tu saṃharet |
jeṣṭhayā ca sthitiṃ kuryā tritattvatanusaṃsthitaḥ ||45||
via free access
402 hatley
avadhūtatantu is, more specifically, to take on the mantra-body (mantravi-

graha) of Kapālīśabhairava, the deities of whose largely female maṇḍala per-
vade the lotus garlands. Strung together by the cord of śakti, the lotuses of
all the goddesses are also simultaneously loci of Bhairava, whose seed-syllable
hūṃ appears in each, paired with the presiding goddess.38 Thus it is that the
BraYā’s opening benediction invokes the bhairavatattva as sporting in the form
of the liṅgam in the lotuses of his consorts.39
While the sādhaka seeks apotheosis as Bhairava, the divine body he takes on,
the avadhūtatanu, in fact consists of a feminine power which ultimately tran-
scends the male deity. The “unblemished” or “pure” (avadhūtā) śakti is present
not only as the cord which unites the lotus garlands: in mantric terms, she
also pervades the garlands as hāṃ, the final syllable of the Nine-Syllable Vidyā
(with the addition of anusvāra). In this form she is known as Paramā or Yogeśī,
the eighth and highest Mother-goddess from whom the others emerge,40 and
as Aghorī or Bhairavī, consort of Bhairava. Her simultaneous pervasion of the
body and syllabic presence in each lotus are intimated in BraYā 4.595–596b:
sarvvāṅge paramā śaktir41 vaktranetrāṅgavarjitā42 |

padmāsanavihīnā tu vinyaset sādhakottamaḥ ||595||
padmamālaiḥ samāyuktā padme padme vikalpayet |
The best of sādhakas should install the supreme sakti [hāṃ] on the entire
body, without face, eye, or limb mantras, devoid of a lotus seat; one should
[also] envision her in each lotus, conjoining/possessing the lotus gar-
lands.
navāṃbhojakṛtāṃ mālāṃ śaktisūtreṇa bheditām |

navagranthivibhāgena bibhrāṇo parameṣvaraḥ ||46||
43d icchayā] em.; icchayo ms. 44a sarvaṃ] em.; savaṃ ms. 44c sṛjate] corr.; śṛjate
ms. 45a sṛjate] corr.; śṛjate ms. 45c jeṣṭhayā ca sthitiṃ] em.; jeṣṭayāvasthitaṃ ms.
46d bibhrāṇo] em.; bibhraṇau ms.
38 E.g. BraYā 4.515, concerning the lotus in the forehead of goddess Raktā: vāmapārśve tatas
tasyāḥ kapālīsasya vinyaset | bījamātraṃ mahāprājña vaktranetrāṅgavarjjitaṃ ||.
39 BraYā 1.1b: dūtīnāṃ padmaṣaṇḍe ’samasukhavilasal liṅgarūpaṃ bibharti; see the discus-
sion in Hatley 2018, 383–385.
40 BraYā 2.17–18: hākāreṇa parā śaktir etā yasyā vinirgatāḥ | mātaras te mayā proktā yāge
ucchuṣmapūjite ||17|| yā sā eva mayā proktā mātṝṇāñ caiva pūraṇī | tasyedaṃ kathitaṃ sar-
vaṃ yaṃ jñātvā nāvasīdati ||18||.
41 śaktir] em.; śakti ms.
42 varjitā] corr.; varjitāḥ ms.
via free access
As holds true for the deity maṇḍala in all its forms, the supreme goddess has
a transcendent, subtle presence, lacking a singular locus.43 Thus while Kapāl-
īśabhairava reigns from the uppermost lotus, the formless, all-encompassing
reality to which the sādhaka aspires is “a feminine power which transcends the
male-female dichotomy which patterns the lower revelations” (Sanderson 1988,
672).
3 Yoga and the Integration of Inner and Outer Ritual
The “method of the lotus garlands” (padmamālāvidhi) has more and less elab-
orate forms and numerous inflections. At its simplest, one installs the seed-
syllables of the principal deities in nine lotuses.44 The procedure of the lotus
garlands applies not only to the body, but to ritual involving external supports
as well. Thus for the worship system ( yāga) taught in chapter 12, one begins
by laying out nine lotuses in a square maṇḍala on a substrate, then visualizing
the śaktitantu and installing the mantra-deities.45 After envisioning an elabo-
rate mantric throne for installation of Bhairava in the central lotus, one engages
in the somatic performance of worship. Subsequently, the procedure is repli-
cated in full detail inwardly “by the path of yoga” ( yogamārgeṇa), seated in the
lotus posture.46 ‘Outer’ ritual thus begins with meditative visualization and is
followed by the rite’s recapitulation internally.
While the padmamālāvidhi represents a basic template for ritual and the
empowerment of the body, ritual may be patterned by divergent pantheons,
especially the practitioner’s personal pantheon (svayāga), a configuration of
the mantra-deities established through initiation. These personalized inflec-
tions of the root pantheon (the mūlayāga) are nine in number, based on the
predominance of each of the nine major deities in turn.47 In the obligatory
daily worship (nityakarman) and much other ritual, inner worship (hṛdyāga) of
this specific pantheon forms the standard preliminary to bahirnyāsa, the act of
43 See the discussion of Kiss (2015, 20–22), a section fittingly titled “Where is Caṇḍā Kāpā-
linī?”
44 This minimal (svalpa) form of nyāsa is described in BraYā 18.75–79.
45 BraYā 12.1–3: athātaḥ saṃpravakṣyāmi āsanaṃ pūrvvam eva hi | yāgaṃ caiva mahādevi
sādhakānāṃ hitāya vai ||1|| pūrvvokte maṇḍale caiva gandhamaṇḍalake pi vā | puṣpa-
maṇḍalake vāpi śaktitantu vicintayet ||2|| tasyādho praṇavaṃ dadyā kālāgnin tatra vinyaset
| kṣīrodaṃ tan tu vinyasya avadhūtaṃ tato nyaset ||3||.
46 BraYā 12.29 ff.
47 Concerning the nine pantheons (navayāgāḥ), see the brief remarks in Kiss 2015, 16–18, and
the article navayāga in Tāntrikābhidhānakośa, vol. III.
via free access
404 hatley
installing the mantra-syllables upon the body or another substratum, whether

the fire pit, ritual vessel, or icon.48 Applications of mantra-deity pantheons not
based upon the Nine-Syllable Vidyā also follow this alternating pattern of inner
and outer ritual. While worship ( yāga, yajana) provides the essential paradigm
for the BraYā’s integration of inner and outer ritual, the principle and prac-
tice of their integration applies more widely. This is abundantly evident in the
domain of “magic,” including all manner of rites involving alphabetical wheels
(cakra), images of the deities (pratimā), and two-dimensional figures ( yantra).
These processes may entail, for example, the fusion of the channels of one’s
own body with the spokes drawn in the lotus of a cakra—the term nāḍī refers
to both—which may then be fused with the channels of the targeted victim
(sādhya).49
Given the integral unity of outer and inner ritual, of somatic acts and sub-
jective processes, the question arises as to whether and in what manner ‘yoga’
represents, for the BraYā, a domain of practice meaningfully distinct from ‘rit-
ual’ (kriyā or karman). The evidence is ambiguous, at least for what seems
likely to be the core, early stratum of the text (chapters 1–50, more or less).50
Occurrences of the term yoga in the sense of meditational practice—often in
the phrase yogamārgeṇa, “via the path of/by way of meditation”—generally
seem synonymous with inner ritual (adhyātmakriyā). Chapter 12, for instance,
describes the procedure of the inner padmamālā as “installation of the sakti”
(śaktinyāsa) through “visualizing by way of yoga” ( yogamārgeṇa saṃciṃtya,
12.35cd).51 On the whole, meditational disciplines are so well integrated into
48 E.g. BraYā 4.366c–367b: yāgasthānaṃ tato gatya hṛdyāgan tu prakalpayet ||366|| svayāgo-
tthena mārgeṇa bahi nyāsaṃ tathaiva ca |. A detailed description of inner worship (hṛd-
yāga) of one’s own pantheon (svayāga) appears in BraYā 38.24c–31, in this case as a pre-
liminary to fire ritual (agnikārya).
49 To give an example, I quote from my remarks on parakāyapraveśa (“entry into the body of
another”) in Tāntrikābhidhānakośa, vol. III: “PBY(H) [Picumata-Brahmayāmala] 14.254–
263ab describes a practice in which one should fuse ( yojayet) the nāḍīs of the yogic body
with those of a cakra of the kulavidyā-mantra of this system, inscribed on cloth, metal
or wood (PBY(H) 14.240). In this case the purpose is the extraction of ‘nectars’: ‘The sād-
haka, thus exiting the body through the tip of the nose, should enter the body of another
and perform the extraction of nectars, after fusing [his with the victim’s] nāḍīs; about
this, there is no need for deliberation’ (evaṃ dehā[d] viniṣkramya nāsikāgreṇa sādhakaḥ
|| paradeham vi[ś]et mantrī amṛtākṛṣṭiñ ca kārayet | nāḍīsaṃdhānakaṃ kṛtvā nātra kārya
vicāraṇāt, PBY(H) 14.259cd–260).”
50 See Hatley 2018, 64–71, concerning the BraYā’s likely stratification.
51 BraYā 12.35c–40b: yogamārggeṇa saṃciṃtya śaktinyāsan tu kārayet || 35|| padmāsano-
paviṣṭas tu cintayitvā tu sādhakaḥ | avadhūtaṃ nyasen mantrī pādayor ubhayor api ||36||
pañcātmikā mahādevi guhye caturgguṇan tathā | hṛdayaṃ tṛguṇaṃ nyasya lalāṭe dvigu-
via free access
the BraYā’s ritual systems that to treat these as a distinct domain of practice—
whether or not designated yoga—seems artificial. However, three brief chap-
ters of the core text do focus on meditational practices which, exceptionally,
seem virtually independent from somatic acts and external supports.52 More-
over, a passage in chapter 25 specifies that a sādhaka who aspires for libera-
tion (the mumukṣu), rather than supernatural experiences and powers (bhoga),
should devote himself to a simplified pantheon worship ( yāga), giving up all
other rituals (karman) and “abiding by the path of yoga.” While not abandoned,
outer ritual here assumes a simplified form, the emphasis shifting decidedly
toward inner practice.53 Thus the degree to which meditational disciplines
feature primarily as integral elements of ritual may simply reflect the BraYā’s
emphasis on supernatural attainment (siddhi): the path of yoga, ultimately, is
for seekers of liberation.
The Brahmayāmala likely has two or more distinct textual strata,54 and
yoga proves to be an area in which stark contrasts emerge between these:
later chapters of the text introduce yogic practices and emphases seemingly
unanticipated in the core fifty-odd chapters. These include systems of med-
itation focused on a series of inner voids and resonances,55 a yoga system
based on internalization of visionary encounters with the goddesses ( yogin-
īmelaka),56 yogic practices for cheating death (a system which, in contrast
ṇaṃ nyaset|| 37|| ekātmikā tathā caiva śikhāyāṃ sādhakottamaḥ | sakṛt sakṛt tathā caiva
vaktrasthāneṣu vinyaset ||38|| netrasthāneṣu vai dadyād aṅgasthāneṣu caiva hi | śakti-
tantu tato dhyātvā nmastakāt pādayo ntikā ||39|| bahirbhūtaṃ śarīrasthaṃ padmamālān
tu cintayet |.
52 These are chapters 36 (nāḍīsañcārapaṭala), 41 (bindupaṭala), and 44 (krīḍākarmapaṭala).
The latter, for instance, describes meditation upon the self in the lotus of the heart; merg-
ing with bindu, one gains the ability to traverse the universe in various forms. BraYā 44.2–
3, 8: hṛddeśe kamalaṃ dhyātvā vyomapaṅkajasaṃyutaṃ | bindumadhye nyase ’tmānaṃ
[em.; nyasen mānaṃ ms.] viśvadeham ayaṃ śubhaṃ ||2|| śaktibhiḥ kiraṇopetaṃ tārā-
ṣṭakavibhūṣitaṃ | taṃ dhyāyet paramaṃ rūpaṃ bindulīnaṃ śivātmakaṃ ||3|| … antarīkṣe
tathā bhūmau pātāleṣu ca dehiṣu | anya-m-anyeṣu rūpeṣu vicaren nātra saṃśayaḥ ||8||.
For another example of the separation of meditation and ritual, BraYā 24.16c–17b speaks
of meditative absorption (samādhi) as an activity one may engage in during interludes
between the daily rites: japayukto kṣapen mantrī prāta[r]madhyāhṇikottaraṃ ||16|| śāstra-
saṅgena vā saṃstho atha vā samādhiṣu sthitaḥ [em.; sthitau ms.] |.
53 BraYā 25.341c–342: kevalasya mahādevi śivasya paramātmanaḥ || 341|| bhairavākhyasya
saṃprokto yāgo ’yaṃ muktilakṣaṇaṃ | etad yāgatrayaṃ proktaṃ mumukṣo sādhakasya tu
| nānyaḥ karmanivṛttasya yogamārgasthitasya tu ||342||. (Perhaps read anyakarmanivṛtta-
sya, “desisting from other rituals”?)
54 As argued in Hatley 2018, 64–71.
55 Chapters 92 and 99, respectively.
56 Chapter 100.
via free access
406 hatley
to the core chapters, attests the bodily kuṇḍalinī),57 and a kind of “yoga of
absorption” (layayoga) based on the granthis of the principal padmamālā.58
While not embeded within critiques of outer ritual, these diverse yogas of the
text’s later stratum tend to eschew external supports and somatic, performa-
tive acts. Far from being disciplines limited to the liberation seeker, these yogas
offer the possibility of accomplishing the sādhaka’s aims through inner acts
alone, potentially superceding outer ritual. This is particularly evident in chap-
ter 100, which teaches a “rite for mastery of the clans of goddesses through
yoga” ( yogena kulasādhanam). Attainment of direct, power-bestowing encoun-
ters (melaka) with the goddesses is one of the BraYā’s dominant ritual aims, as
illustrated by the “rite for the mastery of vetālas” (vetālasādhana, ch. 15), “rite
of the great churning” (mahāmanthāna, ch. 46), “pavilion of power” (siddhi-
maṇḍapikā, ch. 47), and “worship in the pit [of power]” ([siddhi]garttāyāga,
ch. 48). These virtuoso and macabre performances may culminate with the
goddesses manifesting bodily before the sādhaka and granting boons. Belying
this pattern, the yoga of chapter 100 offers the possibility of accomplishing mas-
tery over the clans of goddesses through a process of inner realization alone.59
Similarly, the yoga of BraYā 99 (called vijñānapañcaka, “the five knowledges”)
promises the ability to enter the bodies of others without recourse to external
supports (cakra or yantra), through manipulation of a series of inner reso-
nances (rāva).60 Incorporation of such meditational disciplines into the text’s
latter strata marks a trend towards increasing differentiation between medita-
tion and ‘ritual’—between yoga and kriyā—and the subversion of outer forms
of ritual. The possibility of attaining all ritual aims through yoga alone brings
the vulgate BraYā closer to the Tantrasadbhāva’s hierarchical dichotomy of the
inner and outer, whereby the externalities of ritual are subsumed by inner real-
ities. It is thus possible in this literature to trace early steps in the direction of
Abhinavagupta’s gnostic nondualism, which further overlays the dichotomy of
jñāna and kriyā upon that of inner and outer ritual.
57 Chapter 104.
58 Chapter 53 (cf. 99.19–35).
59 BraYā100.2–3b, 25–26: kulānāṃ sādhanaṃ nātha kathitan tu purā yathā | tat tathā vidi-
taṃ sarvaṃ kulasiddhipradāyakam ||2|| sāmprataṃ sarahasyan tu yogena kulasādhanam
| … mātṛyoginikāyāni śākinīnāṅ kulāni tu | sidhyanti sādhakendrasya yogenānena suvrate
||25|| yena [conj.; – na ms.] sarvagato bhūtvā yoginīsiddhim āpnuyāt | kathayanti ca sad-
bhāvaṃ kulajaṃ jñānam uttamam ||26||.
60 BraYā 99.30cd: kṛtābhyāsas tu vai mantrī dehād deham vrajet kṣaṇāt ||.
via free access
Acknowledgements
I would like to thank Alberta Ferrario, Dominic Goodall, Csaba Kiss, James
Mallinson, and Alexis Sanderson for their stimulating comments on this essay.
I would also like to thank Tadeusz Majewski for producing the line drawings
which I have used for figures 16.1–3.
Abbreviations
BraYā Brahmayāmala
conj. conjecture
corr. correction
em. emendation
ms(s). manuscript(s)
References
Primary Sources
Amaraughaśāsana. Mukunda Ram Shastri, ed. The Amaraugha Shāsan of Gōraksha-
Nātha. Kashmir Series of Texts and Studies, no. 20. Bombay: the Research Depart-
ment of Jammu and Kashmir State/Nirnaya-Sagara Press, 1918.
Tantrasadbhāva. Mark Dyczkowski, ed. Draft edition; e-text available from the Digital
Library of the Muktabodha Indological Research Institute. https://etexts.muktabod
ha.org/digital_library.htm.
Tantrāloka of Abhinavagupta. Mukund Rām Shāstrī (v. 1), Madhusūdan Kaul Shās-
trī (v. 2–12), eds. The Tantrāloka of Abhinava-Gupta, with Commentary by Rājānaka
Jayaratha. ksts, nos. 23, 28, 30, 36, 35, 29, 41, 47, 59, 52, 57, 58. Allahabad; Bombay:
the Research Department of Jammu and Kashmir State, 1918–1938.
Niśvāsatattvasaṃhitā. Dominic Goodall, et al., eds. The Niśvāsatattvasaṃhitā. The Earli-
est Surviving Śaiva Tantra, volume 1. A critical edition and annotated translation of the
Mūlasūtra, Uttarasūtra, and Nayasūtra. Collection Indologie, no. 133 (Early Tantra
Series, no. 5). Pondicherry: Pondicherry: Institut Français d’Indologie/École fran-
çaise d’Extrême-Orient/Universität Hamburg, 2015.
Brahmayāmala. National Archives of Kathmandu manuscript no. 3–370 (Nepal-Ger-
man Manuscript Preservation Project reel no. A42/6).
Brahmayāmala. Shaman Hatley, ed. The Brahmayāmala Tantra or Picumata, volume I:
Chapters 1–2, 39–40, & 83. Revelation, Ritual, and Material Culture in an Early Śaiva
Tantra. Collection Indologie, no. 128 (Early Tantra Series, no. 1). Pondicherry: Insti-
via free access
408 hatley
tut Français d’Indologie/École française d’Extrême-Orient/Universität Hamburg,

2018.
Brahmayāmala. Csaba Kiss, ed. The Brahmayāmala Tantra or Picumata, volume II. The
Religious Observances and Sexual Rituals of the Tantric Practitioner: Chapters 3, 21,
and 45. Collection Indologie, no. 130 (Early Tantra Series, no. 3). Pondicherry: Institut
Français d’Indologie/École française d’Extrême-Orient/Universität Hamburg, 2015.
Mataṅgapārameśvara. N.R. Bhatt, ed. Mataṅgapārameśvarāgama (Vidyāpāda), avec le
commentaire de Bhaṭṭa Rāmakaṇṭha. Publications de l’Instutut français d’Indologie,
no. 56. Pondicherry: Institut français d’Indologie, 1977.
Ratnatrayaparīkṣā of Śrīkaṇṭha. Vajravallabha Dvivedi, ed. Aṣṭaprakaraṇam. Yogatan-
tragranthamālā, no. 12. Varanasi: Sampurnanand Sanskrit University, 1988.
Sārdhatriśatikālottara. N.R. Bhatt, ed. Sārdhatriśatikālottarāgama, avec le commen-
taire de Bhaṭṭa Rāmakaṇṭha. Publications de l’Institut français d’Indologie, no. 61.
Pondicherry: Institut français d’Indologie, 1979.
Secondary Sources
Hatley, Shaman. 2007. “The Brahmayāmalatantra and Early Śaiva Cult of Yoginīs.” PhD
dissertation, University of Pennsylvania.
Hatley, Shaman. 2018. See Brahmayāmala.
Kiss, Csaba. 2015. See Brahmayāmala.
Sanderson, Alexis. 1986. “Maṇḍala and Āgamic Identity in the Trika of Kashmir.” In
Mantras et Diagrammes rituels dans l’Hinduisme, edited by Andre Padoux, 169–214.
Paris: Editions du cnrs, 1986.
Sanderson, Alexis. 1988. “Saivism and the Tantric Traditions.” In The World’s Religions,
edited by S. Sutherland et al., 660–704. London: Routledge and Kegan Paul.
trique. Vols. I–II. H. Brunner, G. Oberhammer, and A. Padoux, eds. Beiträge zur
Kultur- und Geistesgeschichte Asiens, nos. 35 and 44. Vienna: Austrian Academy of
Sciences Press, 2000 and 2004.
trique. Vol. III. Dominic Goodall and Marion Rastelli, eds. Beiträge zur Kultur- und
Geistesgeschichte Asiens, no. 76. Vienna: Austrian Academy of Sciences Press, 2013.
via free access
chapter 17
The Amṛtasiddhi: Haṭhayoga’s Tantric Buddhist

Source Text
James Mallinson
Like many of the contributors to this volume, I had the great fortune to have
Professor Sanderson as the supervisor of my doctoral thesis, which was a critical
edition of an early text on haṭhayoga called the Khecarīvidyā. At the outset of
my work on the text, and for several subsequent years, I expected that Sander-
son’s encyclopedic knowledge of the Śaiva corpus would enable us to find
within it forerunners of khecarīmudrā, the haṭhayogic practice central to the
Khecarīvidyā. However, notwithstanding a handful of instances of teachings
on similar techniques, the fully-fledged practice does not appear to be taught
in earlier Śaiva works. In subsequent years, as I read more broadly in the corpus
of early texts on haṭhayoga (which, in comparison to the vast Śaiva corpus, is
relatively small and thus may easily be read by one individual), I came to the
realisation that almost all of the practices which distinguish haṭhayoga from
other methods of yoga were unique to it at the time of their codification and
are not to be found in the corpus of earlier Śaiva texts, despite repeated asser-
tions in secondary literature that haṭhayoga was a development from Śaivism
(or “tantra” more broadly conceived).
The texts of the haṭhayoga corpus do, however, couch their teachings in
tantric language. The name of the haṭhayogic khecarīmudrā, for example, is
also that of an earlier but different Śaiva practice. When I was invited to speak
at the symposium in Professor Sanderson’s honour held in Toronto in 2015,
I decided to try to articulate my rather inchoate thoughts on this subject by
presenting a paper entitled “Haṭhayoga’s Śaiva Idiom.” The inadequacy of my
theories was brought home to me some months after the symposium when I
started to read, together with two other former students of Sanderson, Péter-
Dániel Szántó and Jason Birch,1 a twelfth-century manuscript of the Amṛtasid-
dhi (AS), the earliest text to teach many of the key principles and practices of
1 We were joined at our reading sessions by Sam Grimes, Diwakar Acharya, Camillo Formi-
gatti, Anand Venkatkrishnan and Paul Gerstmayr, whom I thank for their valuable com-
ments.
© james mallinson, 2020 | doi:10.1163/9789004432802_019

via free access
410 mallinson
haṭhayoga.2 I had already read much of the text with Sanderson and others, but
only from later manuscript sources. As we read the older manuscript it gradu-
ally became clear that the Amṛtasiddhi was composed in a Vajrayāna (tantric
Buddhist) milieu.
Thus my notion of haṭhayoga having a Śaiva idiom needed readdressing.
One might perhaps talk instead of its “tantric idiom.” But I shall leave reflections
on that topic for a later date and in this short paper focus on the Amṛtasiddhi
and, in particular, the features of it which make it clear that it was composed
in a Vajrayāna milieu. I am currently preparing a critical edition and anno-
tated translation of the text with Dr Szántó; what follows here results from
our work in progress. Despite our edition being incomplete, I am confident
that the conclusion drawn here about the origins of the text is sound (and
that further work on the text will provide additional and complementary evi-
dence) and I think it important enough to warrant preliminary publication.
Subsequent publications will address this unique text’s many other remarkable
features.
1 The Amṛtasiddhi
The importance of the Amṛtasiddhi was first brought to scholarly attention by

Kurtis Schaeffer in an article published in 2002.3 Here I shall reprise as little
of his rich and dense article as is necessary to provide the background to what
follows. Schaeffer focuses on the twelfth-century4 manuscript of the text, pho-
2 I thank Professors Kurtis Schaeffer and Leonard van der Kuijp for sharing with me pho-
tographs of printouts from a microfilm copy of this manuscript. Schaeffer also kindly shared
his draft edition of the Tibetan translation of the Amṛtasiddhi given in this witness. We read
the manuscript together with a collation of other witnesses, including a transcription of the
Grantha manuscript M2 prepared by Viswanath Gupta, whom I thank for his assistance. Con-
cerning manuscript sigla for the Amṛtasiddhi, please refer below to the section “Witnesses of
the Amṛtasiddhi.”
3 Prior to Schaeffer’s article, the only mention of the text of which I am aware (other than in
manuscript catalogues) is Gode 1954, 22, in which its citations in the Yogacintāmaṇi are noted.
4 Schaeffer (2002, 517) says that the manuscript’s colophon gives a date which “may read
1159 c.e.” The reading is clear: ekāśītijute [°jute is Newar scribal dialect for Sanskrit °yute] śāke
sahāsraike tu phālgune | kṛṣṇāṣṭamyāṃ samāpto ’yaṃ kṛtvāmṛtasiddhir mayā || (f.37v). The
eighth day of the dark fortnight of the lunar month of Phālguna in Śāka 1081 corresponds
to March 2nd 1160 ce (according to the calculator at http://www.cc.kyoto‑su.ac.jp/yanom/
pancanga/). It is possible that the colophon has been copied from an examplar and that
the manuscript itself does not date to 1160. The mansucript’s Tibetan colophon says that the
Tibetan translation is that of the “monk of the Bya [clan]” (Bya ban de) Pad ma ’od zer, who
via free access
the amṛtasiddhi: haṭhayoga’s tantric buddhist source text 411
tographs of printouts from a microfilm of which he and Leonard van der Kuijp
have kindly shared with me. At the time that the microfilm was made, the
manuscript was in Beijing, although Schaeffer believes that it has since been
returned to Tibet. The manuscript is unique in that it is bilingual, with three
registers: the Sanskrit text in a Nepali or east Indian script, a transliteration of
the Sanskrit in Tibetan hand-printing script and a translation5 into Tibetan in
the Tibetan cursive script.
This manuscript is referred to in what follows by the siglum C. The other
witnesses of the text which have been collated are considerably later than C
(the oldest is perhaps the c. 17th-century K1). They present versions of the text
in which redaction has removed or obscured some of the Buddhist features
evident in C. These witnesses may be divided into two groups. The first is a sin-
gle Grantha manuscript from the Mysore Government Oriental Library (M2),
the second seven north Indian and Nepali manuscripts, two from Jodhpur’s
Maharaja Man Singh Pustak Prakash (J1 and J2 = J) and four from the Nepal-
German Manuscript Preservation Project (K1–K4 = K).6
The text of the Amṛtasiddhi consists of 303 verses divided into 35 short
vivekas.7 The first ten vivekas teach the constituents of the yogic body. Vivekas
11–13 teach three methods of manipulating those constituents (mahāmudrā,
mahābandha and mahāvedha) and viveka 14 teaches the practice (abhyāsa),
i.e. how the three methods are to be used together. Vivekas 15–18 teach the four
grades of aspirant, 19–33 the four states (avasthās) of yoga, and 34–35 the final
transformation of the body leading up to nirvāṇa.8
worked towards the end of the eleventh century, which provides us with an earlier terminus
ante quem for the text than the date of the manuscript itself.
5 As noted in the manuscript’s Tibetan colophon, the translation is of a different recension of
the Sanskrit text from that given in the manuscript. At some places, e.g. 7.10 and 7.26, the
translation corresponds to the text as found in the other witnesses, but not that in C.
6 Full details of these witnesses are given at the end of this article.
7 There are 35 vivekas in the Beijing ms and 38 in the others. All verse numbering given here
corresponds to the order of verses in C (which does not itself give verse numbers).
8 Vivekas 19–35 are interspersed with very short chapters on a variety of topics. In the first viveka
(vv. 10–13) there is a list of the topics to be taught in the text. The list corresponds exactly to
the vivekas up to viveka 19, but then goes awry. More analysis is needed to be sure, but it seems
likely that at least some of the viveka divisions after 19 are later additions to the text.
via free access
412 mallinson
2 The Amṛtasiddhi in the Haṭhayoga Tradition
2.1 Citations and Borrowings

The Amṛtasiddhi is a seminal work in the haṭhayoga textual tradition.9 Scha-
effer (2002, 518–519) mentions its citations in the Yogacintāmaṇi (c. 1600 ce)10
and Haṭhapradīpikājyotsnā (1837ce).11 In addition, several haṭhayoga texts bor-
row directly from the Amṛtasiddhi without attribution. The c. 13th-century
Gorakṣaśataka shares three half-verses with it.12 The Vivekamārtaṇḍa, which
is also likely to date to the 13th century redacts four of the Amṛtasiddhi’s verses
into three.13 The c. 14th-century Amaraughaprabodha shares six verses with the
Amṛtasiddhi and paraphrases it extensively elsewhere.14 The Gorakṣayogaśās-
tra (15th century or earlier) borrows two and a half verses15 and extensively
paraphrases other parts of the text. The c. 15th-century Śivasaṃhitā is much
the biggest borrower from the Amṛtasiddhi, sharing 34 verses with it.16 The
Haṭhapradīpikā shares five half-verses with the Amṛtasiddhi, but these may be
borrowed from the Amaraughaprabodha since all the shared passages are also
in that text.17
2.2 Doctrinal Innovations

Several of the Amṛtasiddhi’s teachings have no prior attestation and are central
to teachings on haṭhayoga in later texts, where they are either reproduced ver-
9 Despite the compound haṭhayoga being found in earlier Vajrayāna works (Birch 2011, 535–
536) and its teachings being central to later haṭhayogic texts, the Amṛtasiddhi does not call
its yoga method haṭha. This paradox will be addressed in subsequent publications.
10 Yogacintāmaṇi p. 13 [AS 6.11, 6.13], p. 26 [14.4, 14.8–12, 14.16c–19d], p. 34 [7.25, 7.15, 7.18,
7.8–9, 7.12, 7.5, 7.2cd, 7.3cd, 7.6–7, 7.16c–17d, 7.19ab, 7.17cd, 7.20–24], p. 39 [32.3–4], p. 40
[33.1, 34.1] p. 101 [6.10], 107 [3.1–4, 4.1a–4.2b, 4.4c–4.12d], p. 112 [5.1, 5.3–4], p. 213 [25.3c–
4d, 26.1–2, 27.1, 28.1, 29.1, 31.3c–4b, 31.5ab, 31.5c–7b (with significant differences), 31.8c–
9b, 3.10, 3.12], p. 218 [15.3a–4b, 16.1ab, 16.2a–3b, 17.1, 17.2ac (with differences), 17.3, 18.1–5,
19.1ab].
11 Haṭhapradīpikājyotsnā ad 3.100 [AS 7.8c–9d, 7.12, 7.5a–, 7.2cd] and 4.1 [AS 33.1, 32.3–4, 7.23,
34.1, 7.17, 7.20, 7.6c–7d, 7.16cd].
12 AS 6.7a–6.8b = Gorakṣaśataka 34a–35b. This verse is also found at Gheraṇḍasaṃhitā 5.62.
13 AS 7.16–20 ≈ Vivekamārtaṇḍa 90–92.
14 AS 16.1c–2b, 11.3, 11.9cd, 11.3cd, 14.6, 13.5cd, 13.7cd, 19.2 = Amaraughaprabodha 20, 29, 32cd,
37ab, 38, 39cd, 40ab, 45.
15 AS 3.1, 6.11ab, 11.4 = Gorakṣayogaśāstra 5, 13ab, 17.
16 AS 1.15b–1.16d, 1.17c–1.18b, 1.19ab, 3.1–4, 4.3–4, 11.1ab, 11.3cd, 11.4bc, 11.5ab, 11.6, 11.7cd, 12.6,
15.1, 16.1–3, 19.2 = ß 2.1b–2.2d, 2.3, 2.4ab, 2.6c–9, 2.11–12, 4.28cd, 4.27ab, 4.27dc, 4.28ab, 4.31,
4.34cb, 4.38, 5.13, 5.17c–5.20b, 3.31.
17 AS 11.3, 11.9cd, 19.2 = Haṭhapradīpikā 3.9, 3.13cd, 4.69.
via free access
batim, as noted above, or incorporated into new compositions. These may be

summarised as follows.
1. The Yogic Body
a. The Amṛtasiddhi is the first text to relocate to the body the old
tantric triad of sun, moon and fire.18 The idea of a moon in the skull
dripping amṛta is found in many earlier tantric works, but that of
the sun in the stomach consuming it is new, as is the conflation of
the sun and fire.
i. The Moon
meruśṛṅge sthitaś candro dviraṣṭakalayā yutaḥ |
aharniśaṃ tuṣārābhāṃ sudhāṃ varṣaty adhomukhaḥ ||3.1||
The moon is on the peak of Meru and has sixteen digits.

Facing downwards, it rains dewy nectar day and night.
ii. The Sun
madhyamāmūlasaṃsthāne tiṣṭhati sūryamaṇḍalaḥ |
kalādvādaśasaṃpūrṇo dīpyamānaḥ svaraśmibhiḥ ||4.1||
ūrdhvaṃ vahati dakṣeṇa tīkṣṇamūrtiḥ prajāpatiḥ |
vyāpnoti sakalaṃ dehaṃ nāḍyākāśapathāśritaḥ ||4.2||
grasati candraniryāsaṃ bhramati vāyumaṇḍale |
dahati sarvadhātūṃś ca sūryaḥ sarvaśarīrake ||4.3||
2d °pathāśritaḥ ] CK4; yathāśritaṃ K3, yathāśritaḥ cett.

3b °maṇḍale ] M2; °maṇḍalaiḥ C, °maṇḍalaṃ cett.
(1) The sphere of the sun is at the base of the Central Chan-
nel, complete with twelve digits, shining with its rays. (2)
The lord of creatures (Prajāpati), of intense appearance,
travels upwards on the right. Staying in the pathways in the
spaces (ākāśapatha)19 in the channels it pervades the entire
body. (3) The sun consumes the lunar secretion, wanders
in the sphere of the wind and burns up all the bodily con-
stituents in all bodies.
iii. Fire
kalābhir daśabhir yuktaḥ sūryamaṇḍalamadhyataḥ |
vasati vastideśe ca vahnir annavipācakaḥ ||5.1||
18 This triad is mentioned at Niśvāsatattvasaṃhitā Nayasūtra 4.147 and in many subsequent

tantric works.
19 This is a śleṣa: ākāśapatha can also mean the sun’s orbit in the sky.
via free access
414 mallinson
yo vai vahniḥ sa vai sūryo yaḥ sūryaḥ sa hutāśanaḥ |

etāv ekatarau dṛṣṭau sūkṣmabhedena bheditau ||5.2||
1b °madhyataḥ ] CJK; °madhyagaḥ M2Y

1c vasati vastideśe ] conj.; vasati vatideśe C, vasate vastideśe M2Y, vasatir
asthideśe cett.
2d sūkṣmabhedena bheditau ] C; sūkṣmāt sūkṣmatarau nṛbhiḥ M2, sūkṣ-
mabhedena bhedinau cett.
(1) Endowed with ten digits, in the middle of the sphere

of the sun in the region of the stomach dwells fire, which
digests food. (2) Fire is the sun; the sun is fire. The two look
almost the same [but] differ subtly.
b. The use of the word bindu for semen, bindu’s identification with the
amṛta dripping from the moon, its preservation being essential for
life and its division into male and female are all innovations of the
Amṛtasiddhi which are widely adopted in later haṭhayoga texts.
i. adhaś candrāmṛtaṃ yāti tadā mṛtyur nṛṇāṃ bhavet ||4.11||
11a yāti ] M2K; yati C, °mṛtaṃ yasya J
The nectar of immortality in the moon goes downwards; as a

result men die.
ii. bindupātena vṛddhatvaṃ mṛtyur bhavati dehinām ||21.3||
The fall of bindu makes men grow old [and] die.
iii. sa bindur dvividho jñeyaḥ pauruṣo vanitābhavaḥ |
bījaṃ ca pauruṣaṃ proktaṃ rajaś ca strīsamudbhavam ||7.8||
anayor bāhyayogena sṛṣṭiḥ saṃjāyate nṛṇām |
yadābhyantarato yogas tadā yogīti gīyate ||7.9||
kāmarūpe vased binduḥ kūṭāgārasya koṭare |
pūrṇagirimudāsparśād vrajati madhyamāpathe ||7.10||
yonimadhye mahākṣetre javāsindūrasannibham |
rajo vasati jantūnāṃ devītattvasamādhṛtam ||7.11||
binduś candramayo jñeyo rajaḥ sūryamayas tathā |
anayoḥ saṃgamaḥ sādhyaḥ kūṭāgāre ’tidurghaṭe ||7.12||
9cd yadābhyantarato yogas tadā yogīti gīyate ] CHJ; yadābhyantarato yogas

tadā yogī sa gīyate M2, yadā tv abhyantare yogas tadā yogo hi bhaṇyate cett.
10a kāmarūpe ] CM2; kāmarūpo cett.
10b kūṭāgārasya° ] CM2; kūṭādhāraṇya J, kūṭādhārasya K
10d °mudā° ] C; °sadā° J, °guhā° cett.
10d vrajati ] C; vrajate M2, rājanti cett.
11d °samādhṛtam ] C; °samāvṛtaṃ M2, °samāvṛta K1, samāvṛtaḥ cett.
via free access
(8) Know bindu to be of two kinds, male and female. Semen

(bīja) is said to be the male [bindu] and rajas (female gener-
ative fluid) is female. (9) As a result of their external union
people are created. When they are united internally, then one
is declared a yogi. (10) Bindu resides in Kāmarūpa in the hollow
of the multi-storied palace (kūṭāgārasya).20 Through pleasur-
able contact at Pūrṇagiri it travels along the Central Channel.
(11) Rajas resides in the great sacred field in the perineal region
( yonimadhye). It is as red as a javā flower21 and is supported
by the Goddess element (devītattvasamādhṛtam). (12) Know
bindu to be made of the moon and rajas to be made of the
sun. Their union is to be brought about in the very inaccessible
multi-storeyed palace.
c. A connection between the mind and breath is taught as early as
the Chāndogya Upaniṣad (6.8.2). The Amṛtasiddhi is the first text to
teach that mind, breath and bindu are connected, a notion found in
many subsequent haṭhayoga texts.
calaty ayaṃ yadā vāyus tadā binduś calaḥ smṛtaḥ |
binduś calati yasyāyaṃ cittaṃ tasyaiva cañcalam ||7.17||
17a calaty ayaṃ yadā ] C; yadāyaṃ calate M2, yadā caṃcalate JK, calaty eṣa yadā
YHJ
17b calaḥ smṛtaḥ ] JKYHJ; calaḥ smṛ˹ta˺ḥ C, ca cañcalaḥ M2
17c binduś calati yasyāyaṃ ] C; yasyāyaṃ calate binduḥ M2, yasyāyaṃ calate
binduś JK, binduś calati yasyāṅge YHJ
17d tasyaiva ] CKYHJ; tasthyai∗ M2, tathaiva J
It is taught that when the breath moves bindu moves; the mind of
he whose bindu is moving is restless.
d. The three granthis.
The Amṛtasiddhi’s system of three granthis, brahma°, viṣṇu° and
rudra°, which are situated along the central channel of the body and
are to be pierced by the mahāvedha (13.10–11), is very common in
subsequent haṭhayoga texts.22
20 On the kūṭāgāra, see below, p. 418.

21 The bright red javā flower (Hibiscus rosa-sinensis L.), popularly known as the China Rose,
is common throughout south, southeast and east Asia.
22 Granthis are mentioned in many earlier Śaiva texts, some of whose lists include brahma,
viṣṇu and rudra granthis but not in the Amṛtasiddhi’s configuration. See e.g. Kubjikā-
via free access
416 mallinson
2. The three practices, mahāmudrā, mahābandha, mahāvedha (vivekas 11–

13).
These practices, which involve bodily postures and breath control, are
used to make the breath enter the central channel and rise upwards. They
are an innovation of the Amṛtasiddhi and are taught in all subsequent
haṭhayoga texts, albeit sometimes with different names.
3. The four avasthās
The four avasthās, “states” or “stages” of yoga practice (ārambha, ghaṭa,
paricaya, niṣpanna/niṣpatti) introduced in the Amṛtasiddhi (vivekas 19–
33), are taught in many Sanskrit haṭhayoga texts; they are also mentioned
in the old Hindi Gorakhbāṇī (śabds 136–139).
In addition to these innovations, in viveka 14 (abhyāsa, “practice”) the Amṛ-
tasiddhi describes, at a level of detail unparalleled in other texts, the internal
processes brought about by its methods, in particular the movement of the
breaths.
3 Buddhist Features of the Amṛtasiddhi
In Schaeffer’s analysis of the Amṛtasiddhi (2002, 521–524), he notes how it

is unique amongst Tibetan Buddhist works because its teachings are said to
bestow jīvanmukti, “liberation while living,” and make the yogi identical with
Śiva. Despite these Śaiva features, however, close reading of manuscript C, the
twelfth-century bilingual witness of the text, shows that the text was composed
within a Vajrayāna milieu. Furthermore, it pits its teachings against those of
other Vajrayāna schools, not Śaiva ones.
As can be seen in the examples given below, manuscript C generally has the
best readings of the text and presents its Buddhist teachings intact. In the other
manuscripts the specifically Buddhist doctrines found in C are either unwit-
tingly included, misunderstood (and sometimes presented in corrupt forms as
a result) or deliberately changed or omitted.
Some of the text’s Buddhist features are ambiguous or obscure enough for
them to have been preserved by the redactors of the text as presented in the
later witnesses. Thus we find multiple examples of Vajrayāna (or more broadly
Buddhist) terminology such as mahāmudrā (viveka 11 and 31), vajrapañjara
(7.26d), jñānasaṃbhāra (6.9c, 20.2bc), śūnya (8.2a, 8.8d, 8.10d, 19.15a, 20.7b,
matatantra 17.61–84, in which there are sixteen granthis and Netratantra 7.22–25, in which
there are twelve.
via free access
25.1c), niṣpanna (19.2c, 31.1c) and abhiṣeka (13.15a). Similarly, Amṛtasiddhi 7.4
mentions the very specifically Vajrayāna notion of the four blisses:23
ānandā ye prakathyante viramāntāḥ śarīrataḥ |

te ’pi bindūdbhavāḥ sarve jyotsnā candrabhavā yathā ||7.4||
4c °viramāntāḥ ] C; ciram antaś M2, viramāṃtā JK
The [four] bodily blisses whose last is [the bliss of] cessation all arise from
bindu, just as moonlight arises from the moon.
Other Buddhist features of the text as found in C are deliberately omitted or

altered in the later witnesses. Examples of these are listed below. This list is not
exhaustive; further close reading of the text is likely to reveal more examples.
1. Chinnamastā
Manuscript C opens with a sragdharā maṅgala verse in praise of the god-
dess Chinnamastā:
nābhau śubhrāravindaṃ tadupari vimalaṃ maṇḍalaṃ caṇḍaraśmeḥ
saṃsārasyaikasārā tribhuvanajananī dharmavartmodayā yā |
tasmin madhye trimārge tritayatanudharā chinnamastā praśastā
tāṃ vande jñānarūpāṃ maraṇabhayaharāṃ yoginīṃ yogamudrām ||
a śubhrā° ] C; candrā° M2 ● vimalaṃ ] C; vivaraṃ M2

c tasmin ] C; tasyāṃ M2 ● tri° ] M2; tre° C ● chinnamastā praśastā ] C; cittahasthāṃ
praśastāṃ M2
d tāṃ vande jñānarūpāṃ ] C; vande jñānasvarūpāṃ M2
At the navel is a white lotus. On top of that is the spotless orb of the
sun. In the middle of that, at the triple pathway, is she who is the sole
essence of samsara [and] the creator of the three worlds, who arises
on the path of dharma, who has three bodies [and] who is lauded as
Chinnamastā, “she whose head is cut.” I worship her, she who has the
form of knowledge, who removes the danger of death, the yoginī, the
seal of yoga.
Until the 16th century, Chinnamastā is not mentioned in non-Buddhist
texts (Bühnemann 2000, 37). Her Vajrayāna origins have been demon-
strated by Sanderson (2009, 240–241), who notes how the epithet dhar-
modayā, found in the Amṛtasiddhi as dharmavartmodayā, is “strictly Bud-
23 On the four blisses see Isaacson and Sferra 2014, passim.
via free access
418 mallinson
dhist.” One might argue that this maṅgala verse could be an addition
to the text when it was redacted by a Vajrayāna tradition, but the verse
is also found in the Grantha manuscript M2 in a corrupt form. Chin-
namastā’s name is given therein as Cittahasthā, but the epithets dhar-
mavartmodayā and tritayatanudharā are preserved. The Rajasthani and
Nepali manuscripts omit the verse.
2. chandoha
At Amṛtasiddhi 1.16, manuscript C uses the specifically Buddhist term
chandoha:24
sāgarāḥ saritas tatra kṣetrāṇi kṣetrapālakāḥ |
chandohāḥ puṇyatīrthāni pīṭhāni pīṭhadevatāḥ ||1.16||
16c chandohāḥ ] em.; chandohā C, saṃbhedāḥ M2JK
There are oceans, rivers, regions [and] guardians of the regions; gath-
ering places (chandohāḥ), sacred sites, seats [of deities and] the deities
of the seats
In Śaiva texts chandoha is found as saṃdoha.25 That the manuscripts
other than C read saṃbhedāḥ, which makes no sense, suggests that they
may derive from an archetype that had saṃdohāḥ, which subsequent
copyists did not understand.
3. The four elements
Amṛtasiddhi 6.2 refers to four physical elements:
pṛthivyādīni catvāri vidhṛtāni pṛthak pṛthak ||6.2||
2a catvāri ] C; tattvāni cett.
The four [elements] earth etc. are kept separate [by the breath].
In Śaiva and other Hindu traditions there are five primary physical ele-
ments. The later manuscripts therefore change catvāri, “four,” to tattvāni,
“elements.”
4. kūṭāgāra
This is a common term in the Pali Canon, meaning “a building with
a peaked roof or pinnacles, possibly gabled; or with an upper storey”
(Rhys Davis and Stede 1921–1925, s.v. kūṭāgāra). It is also found in sev-
eral Vajrayāna texts, where it refers to a “multi-storeyed palace” in the
middle of a maṇḍala (Reigle 2012, 442). It is not found in Śaiva texts
24 Sanderson 2009, 180 note 436.

25 Sanderson loc. cit.: “This substitution of initial ch- for s-/ś- is probably an east-Indianism.”
via free access
and is not recognised by the later north Indian and Nepali witnesses of
the Amṛtasiddhi.
7.10ab kāmarūpe vased binduḥ kūṭāgārasya koṭare |
10a °rūpe ] CM2; °rūpo JK

10b kūṭāgārasya ] C, ∗ūṭāgārasya M2, kūṭādhāraṇya° J, kūṭādhārasya K
Bindu resides at Kāmarūpa,26 in the hollow of the multi-storeyed pal-

ace.
5. trivajra
8.21 in C mentions the three vajras, i.e. the common Vajrayāna triad of
kāya, vāk and citta. In the other witnesses trivajrāṇāṃ is found as trivar-
gāṇāṃ.
trivajrāṇāṃ samāveśas tadā vai jāyate dhruvam ||8.21||
21c trivajrāṇāṃ ] C; trivargāṇāṃ M2JK
Then absorption into the three vajras is sure to arise.

6. trikāya
A reference to the Buddhist notion of the triple body is expunged in the
later witnesses:
sarvajñatvaṃ trikāyasya sarvajñānāvabodhakam |
lakṣaṇaṃ siddhacittasya jñātavyaṃ jñānaśālibhiḥ ||29.2||
2a °kāyasya ] C; °kālasya M2, °kāryasya JK

2b °bodhakam ] CM2; °bodhanam JK 2c siddhacittasya ] C; siddhivit tasya JK
Omniscience, which brings about complete understanding of the

triple body, should be known by the knowledgable to be the mark of
he whose mind has been mastered.
7. buddha
Verses in which C has (or its archetype is likely to have had) buddha are
reworked in the later witnesses.
bindur buddhaḥ śivo bindur bindur viṣṇuḥ prajāpatiḥ |
binduḥ sarvagato devo bindus trailokyadarpaṇaḥ ||7.15||
15a buddhaḥ ] em.; vṛddhaḥ C, ūrdhvaḥ cett.
26 The Mahāmudrātilaka (draft edition of Staatsbibliothek zu Berlin Preuss. Kulturbesitz

Orientabteilung Hs. or. 8711, folio 17 verso) locates the bodily Kāmarūpa between the eye-
brows.
via free access
420 mallinson
Bindu is Buddha, bindu is Śiva, bindu is Viṣṇu, the lord of creatures,

bindu is the omnipresent god, bindu is the mirror of the three worlds.
tāvad buddho ’py asiddho ’sau naraḥ sāṃsāriko mataḥ | 32.3ab
3a buddho ] C; ∗ddho M2, ⊔dvo J1, siddho J2, vaddho K
Even a Buddha, as long as [he remains] unperfected [by means of the

practice taught in the Amṛtasiddhi ], is considered a worldly man.
8. svādhiṣṭhāna yoga
In two places the Amṛtasiddhi mentions svādhiṣṭhāna yoga. This is a
method of visualising oneself as a deity which is central to the teach-
ings of a wide variety of Vajrayāna texts (e.g. Guhyasamāja 7.2, where it
is called svādhidaivatayoga, and the Pañcakrama, whose third krama is
called the svādhiṣṭhānakrama). In the two verses from the Amṛtasiddhi
given below, the methods of svādhiṣṭhāna yoga are said to be ineffective;
to achieve the goals of yoga one must use the practice taught in the Amṛ-
tasiddhi. The later witnesses of the text do not understand the phrase
svādhiṣṭhānena yogena and, presumably surmising svādhiṣṭhāna to refer
to the second of the six cakras in a system taught in many haṭhayoga texts
(but not in the Amṛtasiddhi, which makes no mention of cakras), they
change yogena to mārgeṇa in an attempt to make the phrase refer to a
pathway in the yogic body.
svādhiṣṭhānena yogena yasya cittaṃ prasādhyate |
śilāṃ carvati mohena tṛṣitaḥ khaṃ pibaty api ||8.9||
9a yogena ] C; mārgeṇa M2JK 9b yasya ] JK; yastuś C, yatna M2 ● prasādhyate ] M2JK;

prasādhyati C
He who tries to master his mind by means of self-established yoga

deludedly chews a rock and, thirsty, drinks the sky.
svādhiṣṭhānena yogena na kṣīyete guṇau nṛṇām |

asti mudrā viśeṣeṇa gurumukhābjasaṃbhavā ||10.11||
11a yogena ] C; mārgeṇa M2JK 11b na kṣīyete ] em.; na kṣīyate C, prakṣīyante M2,
nākṣipeti JK ● guṇau ] C; guṇā M2, guṇo JK 11c viśeṣeṇa ] CJK; viśeṣād vā M2 11d
guru° ] CJK; guror M2 ● °mukhābja° ] C; °vaktrābja° M2, °mukhāt tu JK ● °saṃbhavā ]
JK; °saṃbhavāṃ C, °saṃbhavāt M2
via free access
The two [unwanted] guṇas [rajas and tamas] in men are not destroyed
by self-established yoga. There is a mudrā especially [for that], born
from the lotus-mouth of the guru.
4 Conclusion
The Amṛtasiddhi was composed in a Vajrayāna Buddhist milieu and its in-
tended audience was other Vajrayāna Buddhists. Its teachings are subsequently
found in haṭhayoga texts from a wide range of non-Buddhist traditions. This
does not mean, however, that haṭhayoga itself was a product of Vajrayāna
Buddhists. I have argued elsewhere (e.g. Mallinson 2015) that some haṭhayoga
techniques were current among ascetics long before their codification. The
Amṛtasiddhi was the first text to codify many of haṭhayoga’s distinctive prin-
ciples and practices and was thus the first to assign names to them. As a result
the Amaraughaprabodha, the first text to teach physical yoga methods under
the name haṭha, includes among its techniques the Amṛtasiddhi’s mahāmudrā,
mahābandha and mahāvedha (with slight variations in their methods). In addi-
tion to these physical techniques, the Amaraughaprabodha also adopts from
the Amṛtasiddhi the more theoretical doctrine of the four avasthās or stages of
yoga, showing that the Amṛtasiddhi’s influence was more than simply termino-
logical.
Because they share traditions of 84 siddhas, several scholars have posited
connections between Vajrayāna Buddhists and Nāth yogis,27 with whom the
practice of haṭhayoga has long been associated. The Amṛtasiddhi’s Vajrayāna
origins and its borrowings in subsequent haṭhayoga texts, some of which are
products of Nāth traditions, provide the first known doctrinal basis for this con-
nection and a stimulus for its further investigation.28
27 Although such usage is not found in pre-modern texts, to avoid confusion I use the word
“Nāth” to refer to ascetics usually called yogīs or jogīs in texts and travellers’ reports and
whose traditions, with some exceptions such as those which trace their lineages to Kā-
nhapa or Kṛṣṇācārya, came, by the sixteenth century at the latest, to be grouped together
in twelve panths or lineages. On the Nāth Saṃpradāya, see Mallinson 2011.
28 The historical context of this connection is explored in Mallinson 2019, in which the
Konkan site of Kadri (in present-day Mangalore) is proposed as the location of the tran-
sition from Vajrayāna Buddhism to Nāth Śaivism evinced by the Amaraughaprabodha’s
reworking of the teachings of the Amṛtasiddhi.
via free access
422 mallinson
5 Witnesses of the Amṛtasiddhi
5.1 Manuscripts Collated

– (C) China Nationalities Library of the Cultural Palace of Nationalities
Ms. No. 005125 (21). Paper. Sanskrit text in both Nepali (or perhaps East
Indian) and Tibetan hand-print scripts, Tibetan translation in Tibetan cur-
sive script.
– Maharaja Man Singh Pustak Prakash, Jodhpur
1. (J1) 1242. Paper. Devanagari.
2. (J2) 1243. Paper. Devanagari.
– Nepal-German Manuscript Preservation Project. All entitled Amṛtasiddhi.
1. (K1) E655/39. Paper. Devanagari.
3. (K3) H232/37. Paper. Newari.
– (M2) Mysore Government Oriental Manuscripts Library D-4342 (ff. 21v–
40v). Palm leaf. Grantha.
5.2 Other Collated Witnesses

These two texts are mentioned in the apparatus only in the small number of
instances that they provide readings.
– (Y) Yogacintāmaṇi ed. Haridās Śarmā, Calcutta Oriental Press, n.d.
– (HJ) Haṭhapradīpikājyotsnā of Brahmānanda. Svāmī Maheśānand, Dr. Bābu-
rām Śarmā, Jñānaśaṃkar Sahāy, and Ravindranāth Bodhe, eds. Brahmānan-
dakr̥ tā Haṭhapradīpikā Jyotsnā: Locanātmaka Saṃskaraṇa (Hindī). Lonavla:
Kaivalyadhām S.M.Y.M. Samiti. 2002.
5.3 Manuscripts Not Yet Collated

1. Mysore Government Oriental Library D-4341. Paper. Grantha.
2. Mysore Government Oriental Library R-2881(n). Palm leaf. Grantha. In-
complete.
3. Adyar Library 75278. Palm leaf. Grantha.
4. Baroda Oriental Institute 7970(b). Palm leaf. Grantha.
Acknowledgements
I thank Dominik Wujastyk for his comments on a draft of this article, the
research for which was carried out as part of the Hatha Yoga Project (hyp.soas
.ac.uk). This project has received funding from the European Research Coun-
via free access
cil (ERC) under the European Union’s Horizon 2020 research and innovation
programme (grant agreement no. 647963).
References
Primary Sources
Amaraughaprabodha of Gorakṣanātha. K. Mallik, ed. The Siddha Siddhānta Paddhati
and Other Works of Nath Yogis. Poona: Poona Oriental Book House, 1954.
Kadalīmañjunāthamāhātmyam. Śambhu Śarmā Kaḍava, ed. Kāśī: Gorakṣa Ṭilla Yoga
Pracāriṇī, 1957.
Kubjikāmatatantra. T. Goudriaan and J.A. Schoterman, eds. The Kubjikāmatatantra:
Kulālikāmnāya Version. Leiden: E.J. Brill, 1988.
Guhyasamājatantra. Yukei Matsunaga, ed. Osaka: Toho Shuppan, 1978.
Gorakṣayogaśāstra. National Archives of Kathmandu 5–332; Nepal-German Manu-
script Preservation Project 24/44.
Gorakṣaśataka. Government Oriental Manuscripts Library, Madras, Ms. No. R 7874.
Gorakhbānī. P.D. Baḍathvāl, ed. Prayāg: Hindī Sāhity Sammelan, 1960.
Gheraṇḍasaṃhitā. J. Mallinson, ed., trans. New York: YogaVidya.com, 2004.
Chāndogya Upaniṣad. Patrick Olivelle, ed., trans. The Early Upaniṣads: Annotated Text
and Translation. New York; Oxford: Oxford University Press, 1998.
Dattātreyayogaśāstra. Unpublished edition by James Mallinson.29
Navanāthacaritramu of Gauraṇa. K. Ramakrishnaiya, ed. Madras University Telegu
Series, no. 7. Madras, 1937.
Niśvāsatattvasaṃhitā. Dominic Goodall, Harunaga Isaacson and Alexis Sanderson, eds.
The Niśvāsatattvasaṃhitā: The Earliest Surviving Śaiva Tantra, volume 1. A critical
edition and annotated translation of the Mūlasūtra, Uttarasūtra, and Nayasūtra. Col-
lection Indologie, no. 128 (Early Tantra Series, no. 1.) Pondicherry: Institut Français
d’Indologie/École française d’Extrême-Orient, 2015.
Netratantra. Madhusūdan Kaul Śāstrī, ed. The Netra Tantram, with Commentary by
Kshemarāja. 2 vols. Kashmir Series of Texts and Studies, nos. 46, 61. Bombay: the
Research Department of Jammu and Kashmir State, 1926, 1939.
Pañcakrama. Katsumi Mimaki and Toru Tomabechi, eds. Pañcakrama: Sanskrit and
Tibetan Texts Critically Edited with Verse Index and Facsimile Edition of the San-
29 This edition was read with Professor Alexis Sanderson, Jason Birch, Péter-Dániel Szántó
and Andrea Acri in Oxford in early 2012, all of whom I thank for their valuable emenda-
tions and suggestions.
via free access
424 mallinson
skrit Manuscripts. Tokyo: The Centre for East Asian Cultural Studies for Unesco,
1994.
Yogacintāmaṇi of Śivānandasarasvatī. Haridās Śarmā, ed. Calcutta: Calcutta Oriental
Press, n.d.
Vivekamārtaṇḍa of Gorakṣadeva. Oriental Institute of Baroda Library, Acc. No. 4110.
Haṭhapradīpikā of Svātmārāma. Svāmī Digambarjī and Dr Pītambar Jhā, eds. Lonavla:
Kaivalyadhām S.M.Y.M. Samiti, 1970.
Haṭhapradīpikājyotsnā of Brahmānanda. Svāmī Maheśānand, Dr. Bāburām Śarmā,
Jñānaśaṃkar Sahāy, and Ravindranāth Bodhe, eds. Brahmānandakr̥ tā Haṭhapradī-
pikā Jyotsnā: Ālocanātmaka Saṃskaraṇa (Hindī). Lonavla: Kaivalyadhām S.M.Y.M.
Samiti, 2002.
Secondary Literature
Birch, Jason. 2011. “The Meaning of haṭha in Early Haṭhayoga.” Journal of the American
Oriental Society 131 (4): 527–554.
Bühnemann, Gudrun. 2000. The Iconography of Hindu Tantric Deities, volume I. The Pan-
theon of the Mantramahodadhi. Groningen: Egbert Forsten.
Gode, P.K. 1954. Studies in Indian Literary History, Vol. II. Bombay: Bharatiya Vidya Bha-
van.
Isaacson, Harunaga and Francesco Sferra. 2014. The Sekanirdeśa of Maitreyanātha
(Advayavajra) with the Sekanirdeśapañjikā of Rāmapāla. Critical Edition of the San-
skrit and Tibetan Texts with English Translation and Reproductions of the MSS. (With
contributions by Klaus-Dieter Mathes and Marco Passavanti). Serie Orientale Roma
fondata da Giuseppe Tucci Vol. CVII. Napoli: Universita degli Studi di Napoli
“L’Orientale.”
Mallinson, James. 2011. “Nāth Saṃpradāya.” In Brill’s Encyclopedia of Hinduism, Vol. 3,
edited by Knut A. Jacobsen, 407–428. Leiden: Brill.
Mallinson, James. 2015. “Śāktism and Haṭhayoga.” In Goddess Traditions in Tantric Hin-
duism, edited by Bjarne Wernicke Olesen, 109–140. London: Routledge.
Mallinson, James. 2019. “Kālavañcana in the Konkan: How a Vajrayāna Haṭhayoga Tra-
dition Cheated Buddhism’s Death in India.” Religions 10 (4), 273. https://doi.org/10
.3390/rel10040273
Reigle, David. 2012. “The Kālacakra Tantra on the Sādhana and Maṇḍala.” Journal of
the Royal Asiatic Society Series 3, 22 (2): 439–463.
Rhys Davis, T.W. and William Stede. 1921–1925. Pali-English Dictionary. London: Pali Text
Society.
Sanderson, Alexis. 2009. “The Śaiva Age.” In Genesis and Development of Tantrism,
Tokyo.
Sastri, V.V. Ramana. 1956. “The Doctrinal Culture and Tradition of the Siddhas.” In The
via free access
Cultural Heritage of India. Vol. 4: The Religions, edited by Haridas Bhattacharyya,

300–308. Calcutta: The Ramakrishna Mission Institute of Culture.
Schaeffer, Kurtis R. 2002. “The Attainment of Immortality: From Nāthas in India to Bud-
dhists in Tibet.” Journal of Indian Philosophy 30 (6): 515–533.
via free access
chapter 18
A Sexual Ritual with Māyā in

Matsyendrasaṃhitā 40
Csaba Kiss
1 Introduction
In this short article I revisit the Matsyendrasaṃhitā (hereafter MaSaṃ), a

Kubjikā-Tripurā oriented tantric yoga text of the Ṣaḍanvayaśāmbhava tradition,
probably from thirteenth-century South India, core chapters of which I edited
and translated for my PhD studies under the supervision of Professor Alexis
Sanderson. My purpose there was to demonstrate that this text provides evi-
dence for a transitional phase in the history of Śaiva Tantra, revealing aspects
of a transition from Kaula practices to early Haṭhayoga. In the present essay I
analyse, and partly edit, MaSaṃ chapter 40, in which a unique and somewhat
ambiguous variant of a Śrīvidyā-type sexual ritual is described, and which Pro-
fessor Sanderson was kind enough to read with me in Oxford in 2005. I would
like to dedicate this paper to him, offering new interpretations of some key ele-
ments, and thus updating my previous analysis (Kiss 2009, 66).1
My approach is based on textual criticism: by restoring the text using four
available manuscripts, I aim at giving an accurate translation and interpreta-
tion, which then enables me to draw some modest conclusions regarding the
history of Śaivism around the thirteenth century.2 I would like to contribute
to the contemporary research on Śaiva sexual rituals which focuses on their
religio-historical importance. For although it is now generally acknowledged
that sexual rituals are a distinctive feature of some of the Śaiva tantric tradi-
1 On the Ṣaḍanvayaśāmbhava tradition, see Sanderson 1988, 687; 2002, 2–3; 2014, 72–73, 76–77,
80. On details concerning the MaSaṃ, see Sanderson 2014, 80, Kiss 2007, 2009, 2011, and 2020
(forthcoming). All quotations from the MaSaṃ are either from Kiss 2020 (forthcoming) or
from my draft edition of the text.
2 I would definitely like to steer clear of some of the nineteenth- and twentieth-century, and
contemporary approaches to the topic of tantric sexual rituals, which include, as White
observes (2000, 4–5): (1) denial, (2) emphasis on the philosophical reinterpretation of these
rites, “while generally denying the foundational importance of transgressivity or sexuality to
the traditions themselves,” and (3) the commodification of New Age “Tantric Sex” as a com-
mercial product.
© csaba kiss, 2020 | doi:10.1163/9789004432802_020

via free access
a sexual ritual with māyā in matsyendrasaṃhitā 40 427
tions, much work remains to be done on this topic.3 Alongside Sanderson’s find-
ings in many of his publications,4 as well as publications by Dupuche (2003),
White (2003) and Biernacki (2007), Hatley’s work, especially on the asidhā-
rā ̆vrata (Hatley 2018, 195–215), a sexual ritual attested in the Picumata/Brah-
mayāmala (BraYā), chapter 40, as well as in the Niśvāsatattvasaṃhitā and the
Mataṅgapārameśvara, is fundamental and is a source of inspiration and in
many respects a model for this essay. Mallinson’s exploration of the haṭhayogic
techniques of khecarīmudrā, vajroli and amarolī is closely related to this field of
research (Mallinson 2007, 221–223, notes 333–334 and 336–337). My own contri-
bution so far (Kiss 2015) comprises an analysis of the BraYā’s relevant teachings
in BraYā 45 on sexual encounters that involve the gathering and magical use of
sexual fluids.
A detailed overview of the types of sexual ritual found in tantric and haṭha-
yogic texts is beyond the scope of this short essay,5 but a number of their dis-
tinctive features can be listed here. This list mainly concerns the variable nature
of the female and male partners, and the nature of the sexual act. Note that cat-
egorisation of a phenomenon like this is greatly complicated by that fact that
many features overlap:
(1) sexual rituals involving restraint or celibacy;6
(2) a sexual act at the end of which the practitioner applies the vajrolimu-
drā, “the practice of urethral suction […] to draw up the combined sexual
fluids”;7
(3) sexual rituals producing male and female sexual fluids, which are then
used and consumed for magical purposes;8
(4) sexual rituals with one’s own wife/partner vs. other partners;9
3 As Shaman Hatley also remarks (2018, 195–196).

4 E.g. Sanderson 2005a, 110–114, note 63; 2007, 284–287; 2009, 132ff., 294, note 699; and 2014, 57.
5 See a similarly non-comprehensive but more detailed list of types of tantric sexual rituals in
Hatley 2018, 196–199.
6 E.g. the asidhāravrata, “Razor’s Edge Observance,” as taught in BraYā 40, which involves sex-
ual contact with a beautiful woman without the sādhaka’s engaging in orgasm. (See Hatley
2018, 195–215.) See also BraYā 45.270–296ab, in which a sexual ritual involving “restraint” (ava-
graha) is taught (Kiss 2015, 49).
7 See Mallinson (2007, 189 note 149). Vajroli is described e.g. in the Vaiṣṇava Dattātreyayogaśā-
stra, vv. 299–314.
8 E.g. BraYā 45, where several variants of a basic type of sexual ritual are described (see Kiss
2015).
9 See for example BraYā 24.32cd, where one’s mother, sister, daughter and wife are listed as pos-
sible partners (figuratively or otherwise): mātā ca bhaginī putrī bhāryā vai dūtayaḥ smṛtā[ḥ].
See also Jayaratha ad Tantrāloka 29.102 addressing the question after citing the line svapatnī
bhaginī mātā duhitā vā śubhā sakhī: ityādyuktyā svapatny api atra kasmāt na parigaṇitā … (“In
via free access
428 kiss
(5) sexual rituals with one partner vs. several partners;10

(6) sexual encounters with human or visionary Yoginīs, or other divine
beings;11
(7) highly æstheticized, philosophic descriptions of Śaiva sexual rituals;12
(8) sexual rituals involving the visualisation of a goddess or the mental pro-
jection of the image of a goddess onto the female partner,13 and so forth.14
The passages from MaSaṃ 40 analyzed and presented below seem to represent
a variety of the last type listed above, but they are somewhat ambivalent: after
mentioning sexual encounters with (human) Yoginīs (verses 38 and 40–41), the
text focuses on a somewhat ambiguous sexual ritual which could be read either
as involving an exclusively visualised goddess as partner or as prescribing the
projection of an image of the goddess onto a human sexual partner. It also
teaches the magical use of male sexual fluid (real or imaginary). MaSaṃ 40 is
somewhat ambiguous in other aspects as well, and it seems that it is through
the analysis of this ambiguity that we may gather clues concerning the his-
tory of late Śaiva traditions and the transition between Śaiva Tantra and early
Haṭhayoga.
2 Details and Ambiguities
The beginning of the chapter in question, MaSaṃ 40.1–28, stands somewhat

apart from the rest of the chapter and does not discuss sexual rituals. Never-
theless, a short analysis of it may be useful here. The text starts (1–6ab) with
the proclamation of its topic: the rituals of Kula/Kaula conduct (kulācāravidhi)
characterised by “the great non-dualist practice” (mahādvaita) and “freedom
these kind of statements, why is one’s own wife not enumerated?”) See also Dupuche 2003,
249 ff.
10 E.g. BraYā 45.574cd–557ab, where sexual rituals with four to eight women are taught. See
also Jayadrathayāmala, Ṣaṭka 4, National Archives, Kathmandu, MS 1–1468, ff. 206v3–
207v5 and Kṣemarāja’s Daśāvatāracarita 10.26. For these references, see Sanderson 2007,
284–287; and 2009, 294 note 699.
11 E.g. the sādhaka engages in “great amusement” with Nāga girls and Āsurīs (demonesses)
in BraYā 59.107cd (f. 254r): nāgakanyais mahākrīḍā āsurībhiś ca jāyate.
12 E.g. Tantrāloka 29 (see Dupuche 2003).
13 E.g. Nityotsava, p. 60: atha tāṃ devarūpāṃ vibhāvya … (“And visualising her [the sexual
partner] in a godly form …”).
14 This list could easily be expanded by using more variables, as Shaman Hatley has sug-
gested (personal communication): partners; place/space/circumstances; roles of mantra,
visualization, accoutrements; kinds of meaning given to the practice; aims/goals; fluids;
etc.
via free access
from conventional practices” (nirācāra), both familiar terms from earlier Śaiva
tantras such as the BraYā.15 Verses 6cd–16 praise and recommend ritual bathing
at a special śivatīrtha, in “Śiva’s water” or “water embodying Śiva” (śivamaye
jale). In the context of the MaSaṃ, it is very probable that what is meant here
is bathing in or with, and consuming, urine or other bodily fluids. Mallinson
(2007, 221–223, notes 333–334, 336–337) has shown convincingly that the prac-
tice of bathing with urine was not unknown in Kāpālika and haṭhayogic tradi-
tions.16 On the other hand, in light of the second half of the chapter, it is not
inconceivable that semen is what is hinted at here. In either case, the appli-
cation of this magical fluid involves transgression and thus should be carried
out in a secret place (16cd). Verse 18 names the miraculous fluid as amarī, a
term echoed in MaSaṃ 27.5 as amarīrasa. That chapter, MaSaṃ 27, teaches
concoctions of herbs and physical secrations such as fæces, urine, menstrual
blood, phlegm (?) and semen (?) (viṅ-mūtra-rajo-recaka-sārakāḥ) associated
with Lokeśa, Keśava, Rudra, Īśa and Sadeśvara, respectively (27.2, see Mallinson
[2007, 220 note 328]). In MaSaṃ 27.5a Sadāśiva (i.e. probably Sadeśvara, or
rather the substance associated with him, probably semen) is said to be the
best among them (sadāśivo varo jñeyas). This may indicate that the meaning
of amarī (and sudhā, amṛta etc.) is flexible; it may refer not only to urine, but
to other bodily fluids as well. Amarī should be drunk after reciting the appro-
priate mantra and should be massaged on one’s body (27.21–26ab), similarly to
what is taught in MaSaṃ 40.64–65, where it is clearly semen.
The second part of our chapter, MaSaṃ 40.29 ff., commences to further
describe kulācāra, and claims that sexual rituals should be performed either
with Yoginīs or with Māyā-type women (verse 38). Here follows an edition and
translation of verses 29–38:
[kulācāranirṇayam]17
ataḥ paraṃ śṛṇu śive kulācārasya nirṇayam |

samprāpya siddhasaṃtānaṃ guruṃ daivaṃ sadā yajet ||40.29||
15 E.g. BraYā 3.227, 45.159 and Kubjikāmatatantra 2.107 etc.

16 In a similar fashion, the BraYā (e.g. in 45.427 and 45.456) prescribes the praise of urine to
be performed by the sādhaka.
17 This edition of MaSaṃ 40.1–69 is based on four manuscripts: Ja, Jb, Jc, Well (see the bibli-
ography; Jd is not available for chapter 40). The transcription of the text of the MaSaṃ
both in the textus acceptus and in reporting variants involves some inevitable falsifi-
cation, of which the most important are the following: I have not always attempted to
report differences in readings between akṣaras that are usually interchangable in MSS (b-v,
via free access
430 kiss
gurupūjārato yogī parāṃ siddhim avāpnuyāt |

guroḥ khaṭvāṃ tathā śayyāṃ vastram ābharaṇāni ca ||40.30||
pādukāṃ chattram ajinaṃ pātram anyad athāpi vā |
pādena spṛśate yas tu śire dhṛtvāṣṭakaṃ japet ||40.31||
gurunindāparaṃ dṛṣṭvā ghātayed athavā śapet |
sthānaṃ vā tat parityajya gacched yady akṣamo bhavet ||40.32||
kulapūjāṃ na nindeta yoginaṃ yoginīm api |
unmattāṃ puṣṭitāṃ nārīṃ surākumbhaṃ śivaṃ gurum ||40.33||
nindanād bhraśyate sadyaḥ paratreha ca pārvati |
paśumārgaṃ na seveta paśūcchiṣṭaṃ na kāmayet ||40.34||
yoginīmelanaṃ kāryaṃ na seveta paśustriyam |
nocchiṣṭaṃ paśave dadyāt na nārīṃ nindayet kvacit ||40.35||
ekībhāvaṃ prakurvīta pṛthagbhāvaṃ na kārayet |
vṛthā pānaṃ na kurvīta na vṛthā māṃsabhakṣaṇam ||40.36||
asaṃskṛtaṃ na piben madyaṃ tatpūtaṃ māṃsam āgraset |
na kuryān mantragoṣṭhīṣu śaucam ācamanādikam |
yadi kuryāt pramādena yoginīśāpa āpatet ||40.37||
bhuñjīyād yat striyaṃ devi śivaśaktyātmabhāvayā |
yoginīṃ sevayen nityaṃ māyāṃ vā pāśavīṃ na ca ||40.38||
Witnesses for MaSaṃ chapter 40: MS Ja: ff. 62r–64r, MS Jb: ff. 126v–130v, MS Jc: ff. 132r–136r,
MS Well: ff. 89v–91v.
29a śive ] J; śide Well (pāda a is a na-vipulā) 29b nirṇayam ] JaJcWell; nirṇayaḥ Jb 29c
siddha° ] J; siddhi° Well; cf. Kubjikāmata 3.98ab: atraiva siddhasantāne pratyakṣo ’haṃ
vyavasthitaḥ 29d guruṃ ] JaJcWell; guru Jb ● yajet ] J; yatet Well 30c khaṭvāṃ ]
J; khadvāṃ Well 31b pātram anyad ] J; pātramadhyam Well 31a na-vipulā 31d tu
śire dhṛtvāṣtakaṃ ] conj.; tu śirasi ghṛtvāṣtakaṃ J, u śira ghṛtvāṣtaka Well; cf. Kubjikāmata
3.133cd–134ab [Manthānabhedapracāraratisaṅgama chapter]: pādukopānahau chattraṃ
śayyāpaṭṭo ’tha bhājanam | pādena saṃspṛśed yas tu śire dhṛtvāṣṭakaṃ japet || 32a
°nindā° ] J; °nidā° Well (Well’s reading would result in two laghu syllables) ● dṛṣṭvā ] J;
v-c, t-n, y-p), but I always report them when both readings are theoretically possible (e.g.
candana-vandana, jaya-japa); I have ignored most instances of gemination of consonants
in ligature with semivowels (e.g. dharmma); I have treated anusvāras and homorganic
nasals as interchangeable; I have altered them, as well as word-final anusvāras and m-s,
silently as required by standard orthography; instances of confusion between ś and s are
reported only in the non-accepted variants; avagrahas are mostly missing in the MSS and I
have always silently supplied them in the textus criticus and in the lemmata, but not when
reporting variants; in the apparatus: em. = emendation; conj. = conjecture; corr. = correc-
tion; ac = before correction (ante correctionem); pc = after correction (post correctionem);
†…† enclose corrupted text which I have not been able to improve upon; ° indicates that
the word is part of a compound or phrase; × stands for an illegible akṣara; Jab = MSS Ja
and Jb; Jall = all available Jodhpur MSS (= Ja, Jb, Jc).
via free access
dṛṣṭā Well 32d akṣamo ] JaJcWell; amo Jb 33ab nindeta yo° ] JWellpc; nindeta yogin-
īmelanaṃ kāryam na seveta paśus trayaṃ nocchiṣṭaṃ pasavo dadyāte Wellac (eyeskip to
verse 35) 33b yoginīm ] Jab; yogibhīm JcWell 33c unmattāṃ ] J; unmatāṃ Well ●
puṣṭitāṃ ] JaJcWell; puṣṭito Jb ● nārīṃ ] Jab; nārī JcWell 33d surākumbhaṃ ] Jab;
sur×kumbhaṃ Jc, surīkumbhaṃ Well 34a nindanād ] JaJcWell; nindanāt Jb ● bhraśy-
ate ] JaJcWell; praśyate[?] Jb 34d paratreha ] J; paratredva Well 34d paśūcchiṣṭaṃ ]
J; paścacchiṣṭaṃ Well 35b °striyam ] corr.; °striyām J, °strayam Well 35c paśave ] J;
pasavo Well 35cd note the absence of sandhi between the two pādas. 36a ekī° ] J; ekvī°
Well 37a asaṃskṛtaṃ ] J; aṃskṛtam Well ● pāda a is hypermetrical. Cf. Kulapradīpa
3.48: asaṃskṛtaṃ piven madyaṃ balātkāreṇa maithunam | svapriyeṇa hataṃ māṃsaṃ
rauravaṃ narakaṃ vrajet 37c kuryān ] Jab; kuryon JcWell ● °goṣṭhīṣu ] Jab; goṣṭhīvu
Well 37 f. yoginīśāpa āpatet ] Jab; yoginī śāyatet JcpcWell, yogānī śāyatet Jcac 38a bhuñ-
jīyād ] J; bhujīyād Well ● yat ] Jab; ya JcWell 38b °bhāvayā ] conj.; °bhāvaya JaJc, °bhāvaye
Jb, °bhāvayaṃ Well 38c yoginīṃ ] Jab; yoginī Well ● sevayen ] Jab; sevayon Well 38d
māyāṃ vā ] Jab; māyāṃ cā Jc, māyāṃ ca Well ● pāśavīṃ ] JapcJbJcWell; pośavīṃ Jaac ●
na ca ] J; na ce Well
[The exposition of Kula Conduct]
After this, O Śivā, hear the exposition of the Kula Conduct (kulācāra).
After he has joined the tradition of the Siddhas, he should worship
his guru as divine. (40.29)
The yogin who is engaged in the worship of his guru can obtain the
highest Power (siddhi). The guru’s bedstead, his bedding, clothes,
ornaments, sandals, parasol, antilope-skin, bowl or anything else: if
he touches any of these with his feet, he should place them on his
head and recite [mantras] eight times. (40.30–31)
If he sees anybody who is abusing the guru, he should beat him or [at
least] curse him. Or, if he is unable [to do so], he should leave the
place. (40.32)
He should not ridicule the worship of the [Yoginī] clans (kulapūjā), or
despise yogins or yoginīs, women when they are intoxicated, or nour-
ished,18 or the wine-pot, or Śiva, or the guru. (40.33)
Contempt [for these] will make him fall immediately here in this world
and in the other world, O Pārvatī. He should not follow the path of
the paśus [i.e. that of the uninitiated] and he should not long for the
leftover of paśus. (40.34)
He should strive for an encounter (melana) with the Yoginīs. He should
not have sex (na seveta) with uninitiated women (paśustrī).19 He
18 Perhaps: “pregnant.” Alternatively, some kind of sexual interpretation is needed here.

19 See 40.38.
via free access
432 kiss
should not give leftovers to the uninitiated (paśu). He should never

abuse women. (40.35)
He should treat [all phenomena] as one, not as separate.20 He should
not drink [alcohol] or eat meat idly [with no ritual purpose].
(40.36)
He should not drink wine without first purifying it [with mantras], and
he should consume meat after he has purified it with that [wine]. He
should not answer the call of nature, should not sip water, etc., while
reciting mantras or in an assembly.21 If he does so out of folly, the
curse of the Yoginīs will fall on him. (40.37)
When ( yat) [the yogin] wants to enjoy (bhuñjīyāt) a woman, O Goddess,
visualising himself as Śiva [and her as] Śakti,22 he should always have
sex with a Yoginī or with a Māyā [type of woman], and never with a
Pāśavī [i.e. a paśu-natured woman, or more precisely someone who
has not been initiated]. (40.38)
Noteworthy among the above are verse 31, which is more or less parallel with
Kubjikāmatatantra 3.133cd–134ab, reconfirming that the cult of the MaSaṃ
draws heavily on the Paścimāmnāya;23 verse 33, which warns against the abuse
of women; and verse 35, which seems to mention only Yoginīs (recommended
for sexual encounters) and uninitiated women (not recommended), and seems
to be silent on any other category, in contrast to verse 38, in which the third type,
Māyā, first appears.
The subsequent verses, 40 and 41, define a Yoginī: she follows the Kula path,
is initiated, uses alcohol in her rituals, and generally gives the impression of
being a human female practitioner. Rather awkwardly, the text does not give
any other detail, but goes on defining her antithesis, the Pāśavī: an uninitiated
woman, who is hostile to Śaivism and who should be avoided (41cd–42).
A sexual ritual with the third type, Māyā, is what the rest of the text focuses
on. It is here that major ambiguities come into the picture. The key aspects
mentioned in the description of the Māyā-type woman are: she has all the
20 I suspect that here the principles of non-hesitation (nirvikalpa) and freedom from conven-
tional practices (nirācāra) are being reaffirmed: the practitioner should not distinguish
between right and wrong, pure and impure etc. See 40.1–28.
21 The interpretation of mantragoṣṭhīṣu is highly tentative.
22 The compound śivaśaktyātmabhāvayā (conj.; °bhāvaya JaJc, °bhāvaye Jb, °bhāvayaṃ Well)
is rather unusual, and my conjecture is insecure. Perhaps °bhāvayā stands for bhāvanayā
metri causa. The intended meaning seems to be clear, though.
23 See Kiss 2009, 25 ff.
via free access
auspicious characteristics required (sarvalakṣaṇasampannā), has neither rūpa

nor kula (kularūpavivarjitā),24 and is to be approached by bhāva (bhāvagamyā,
40.43).25 All of these terms remain rather ambiguous without the help of clar-
ification based upon other passages. Here follows the text and translation of
this section:
[yoginī māyā pāśavī ca]
devy uvāca |
kā nārī yoginī deva kā māyā kā ca pāśavī |
etāsāṃ saṃgame doṣaṃ guṇaṃ ca vada bhairava ||40.39||
[yoginī]
bhairava uvāca |
kulamārgagatā nārī paśumārgavivarjitā |
mādhvīmadasamādhmātā paśupāśavivarjitā ||40.40||
madirānandacetaskā yoginī śivaśāsane |
[pāśavī]
vikalpakuṭilā pāpā kulācāraparāṅmukhī ||40.41||

śivanindāparā devi tatsaparyāvirodhinī |
pāśavī sā mayākhyātā māyākhyāṃ śṛṇu suvrate ||40.42||
24 That kula and rūpa are to be understood as a dvandva compound is confirmed in 40.44ab:
yā māyā rūparahitā kulahīnā maheśvara.
25 Elsewhere I have argued (Kiss 2009, 57–60) that while bhāva in the MaSaṃ is mostly to
be interpreted as “visualisation,” which is a rather common meaning for this word, an
additional specification may be required: bhāva refers to a particularly emotive process
of creating mental images. I suspect that bhāva is preferred in the MaSaṃ to dhyāna
(although dhyāna, dhyāyet etc. also abound) for its extremely wide range of meanings.
One of the basic meanings of bhāva is “being, becoming.” Another is “emotion, sentiment”
or even “passion, love.” I think that all these rather different senses of the word are con-
densed in the yogic key term of bhāva in the MaSaṃ. It is a creation of something in the
mind by the yogin, towards which he should also create a feeling, an empathic attitude,
perhaps passionate devotion, which will result in the ultimate condition, the desired state
of mind: the union with the object, with the deity, with Śiva, or in our case, with a visu-
alised goddess.
via free access
434 kiss
[māyā]
sarvalakṣaṇasampannā kularūpavivarjitā |
bhāvagamyā maheśāni yā sā māyā nigadyate ||40.43||
39 devy ] JaJcWell; śrīdevy Jb 39a yoginī ] JapcJbJcWell; yogi Jaac 39b kā ca pāśavī ] JaJc;
kāśavī Jb, kā ca pāśavīḥ Well 39d bhairava ] J; bhairavaḥ Well 40a nārī ] JbJcWell; ttārī
Ja 40b °vivarjitā ] J; °vivarjitāḥ Well 40cd mādhvīmadasamādhmātā paśupāśavivarjitā
] em.; mādhvīmadasamādhyātā paśupāśavivarjitā Jab, JcWell omit this line 41b °śāsane
] JaJcWell; °śāsanāt Jb 41c °kuṭilā ] J; °kuṭikā Well 41d °parāṅmukhī ] JapcJbJcWell;
°pa×ṅmukhī Jaac 42a °parā ] J; °paro Well ● devi ] JaJcWell; devī Jb 42b °virodhinī ]
corr.; °virodhitī J, °virodhinīm Well 42c pāśavī ] J; pāśavīḥ Well ● sā ] J; sa Well 42d
°ākhyāḥ ] corr.; °ākhyaḥ JWell 43b °vivarjitā ] J; °vivarjitāḥ Well
[Yoginī, Māyā and Pāśavī]
Devī spoke:
O God, what kind of a woman is a Yoginī? Who is Māyā and who is
Pāśavī? Tell me, O Bhairava, the pros and cons of having sex with
them. (40.39)
Bhairava spoke:
A woman who is on the Kula Path [of the Yoginī clans] (kulamārga),
who avoids the path of bound souls (paśumārga) [i.e. the path of the
uninitiated], who is elevated by intoxication induced by liquor, and
is free of the bonds that fetter the soul (paśupāśa), and whose mind
is filled with the bliss of wine (madirā), is [called] a Yoginī in Śiva’s
teaching. (40.40–41ab)
[Pāśavī:] her mental attitude is dishonest, she is wicked, hostile to
Kaula Practice (kulācāra). She tends to abuse Śiva, O Goddess, and
to obstruct his worship. This [type], the Pāśavī, has been [now]
taught by me. O Suvratā, hear the one that is called Māyā. (40.41cd–
42)
A woman who possesses all favourable characteristics (lakṣaṇa) [but]
has neither a [Yoginī] Clan/noble family (kula) nor a [human/mate-
rial] form/beauty (rūpa), and who is to be approached by empathic
imagination (bhāva), O Maheśānī, is called Māyā. (40.43)
The term rūpa in 43b may refer to “material form” and a lack of rūpa would then
indicate that Māyā, the preferred sexual partner, is solely imaginary. In this case
sarvalakṣaṇasampannā indicates that she is to be visualised as a divine being
with great beauty. But how to interpret “devoid of kula” then? The term kula is
via free access
often used in the sense of “a clan of Yoginīs.”26 Is she then not a member of any
Yoginī clan? Does she transcend the clans of Yoginīs, being the supreme God-
dess?27 This interpretation is supported and at the same time refuted by MaSaṃ
22.24 (which in fact is a citation of Nityāṣoḍaśikārṇava, alias Vāmakeśvarīmata,
4.14):
kulayoṣit kulaṃ tyaktvā paraṃ puruṣam eti sā |

nirlakṣyaṃ28 nirguṇaṃ caiva kularūpavivarjitam ||22.24||29
That noble lady abandons her family (kula) and goes to the highest man,
who is invisible, who lacks qualities, and is devoid of kula and form.
Here, in the MaSaṃ as well as the Nityāṣoḍaśikārṇava, the supreme soul

(paraṃ puruṣam) is described as having neither kula nor rūpa: as being imma-
terial and formless.30 But the word kula is used in a double sense in the case
of the female subject of the sentence (kuṇḍalinī in fact): she leaves her/the
body to unite with the Supreme Soul as a noble lady (kulayoṣit) leaves her
family (kulaṃ tyaktvā) to unite with her husband. This does not make it eas-
ier to interpret MaSaṃ 40.43b (kularūpavivarjitā): the sexual partner might be
a woman who was not born in an eminent family, i.e. is of low birth (kula-
vivarjitā), and who lacks beauty (rūpavivarjitā), “beauty” being a natural way
26 See e.g. Sanderson (1988, 671–672): “All Yoginīs belong to the family (kula) or lineage
(gotra) of one or other of a number of ‘higher’ maternal powers, and in any instance this
parentage is ascribed on the evidence of certain physical and behavioural characteris-
tics. An adept in the cult of Yoginīs can identify members of as many as sixty-three of
these occult sisterhoods, but is most vitally concerned with the eight major families of
the Mothers (mātṛ) Brāhmī, Māheśvarī, Kaumārī, Vaiṣṇavī, Indrāṇī, Vārāhī, Cāmuṇḍā and
Mahālakṣmī.” See also Hatley (2007, 13–23, 33, 392, etc.).
27 Cf. Tāntrikābhidhānakośa, vol. I, sv. akula: “Dans les traditions (āmnāya*) du Kula, c’est la
Réalité lumineuse suprême, l’absolu inconditionné: anuttaraṃ paraṃ dhāma tad evāku-
lam ucyate (Tantrāloka 3.143ab),” etc.
28 em.; nirlajjaṃ Cod.
29 Cf. Nityāṣoḍaśikārṇava 4.14: kulayoṣit kulaṃ tyaktvā paraṃ puruṣam eti sā | nirlakṣaṇaṃ
nirguṇaṃ ca kularūpavivarjitam ||.
30 Cf. Jayaratha ad Nityāṣoḍaśikārṇava 4.14: tadā kulaṃ śarīram apahāya akulapadāvasthi-
taṃ paraṃ pūrṇam […] kulena śarīreṇa taddharmeṇa rūpeṇa ca vivarjitaṃ nirāvaraṇasva-
bhāvam, ata eva nirguṇam puruṣaṃ, paraṃ pramātāram, eti tadaikātmyam āsādayatīty
arthaḥ (“Then leaving behind the kula, i.e. the body, she goes to the one who is in the
realm of akula, the supreme, i.e. full […] Person, the highest authority, who is without
a body and without bodily form, with his innate nature manifest and therefore lacking
qualities, i.e. she reaches oneness with Him. This is the meaning [of this verse].”)
via free access
436 kiss
of translating rūpa. In this case, sarvalakṣaṇasampannā (40.43a) refers to the

form she assumes during the yogin’s visualisation, and bhāvagamyā (40.43)
may suggest that a sexual ritual with her requires this visualisation. Alterna-
tively, the ideal sexual partner might be one who is immaterial and formless
(kularūpavivarjitā): a visualised goddess, or, as a matter of fact, kuṇḍalinī. As we
have seen, in the context of the yogin’s sexual rituals, first two, then three types
of partners are enumerated: one, who is hostile to Śaivism, is to be avoided;
initiated human Yoginīs are ideal, but are not dwelt upon in the text, perhaps
because they were less and less available at the time of the composition of the
MaSaṃ; and as a third alternative, the text either suggests pure visualisation or
an uninitiated woman of low birth, without any particular charm, as the locus
of visualisation of the Goddess.
The text goes on to give instructions on visualisation needed for the sexual
ritual. The yogin should visualise himself as Śiva in the form of Kāmeśvara, and
his partner as a goddess:
[māyayā saṃgaḥ]
devy uvāca |
yā māyā rūparahitā kulahīnā maheśvara |
yoginaḥ saṃgamas tasyāḥ kathaṃ bhavati tad vada ||40.44||
bhairava uvāca |
śṛṇu devi pravakṣyāmi māyayā saṃgam adbhutam |
yad amoghaṃ maheśāni durvijñeyam utāparaiḥ ||40.45||
yogasiddhivihīnaiś ca yogibhiḥ suranāyaki |
44 uvāca ] JaJc; u- Jb, uvā- Well 44a yā ] J; omitted in Well (pāda a is a na-vipulā) 44b
maheśvara ] Well; maheśvaraḥ J 44c yoginaḥ ] JaJc; yogina Jb, omitted in Well 44d kathaḥ
] conj.; vāthaḥ J, cāthaḥ Well ● vada ] Jab; vadaḥ JcWell 45 uvāca ] JaJcWell; u- Jb 46a
°vihīnaiś ] JaJc; vihītaiś Jb, vihīnaiḥś Well 46b yogibhiḥ ] J; yogibhīḥ Well ● suranāyaki ]
JaJcWell; suranāyakiḥ Jb
[Sexual ritual with Māyā]
The Goddess spoke:

Tell me, O Maheśvara, how should the yogin sexually approach the
one who is called Māyā, who has neither form/beauty (rūpa) nor a
clan/noble family/body (kula)? (40.44)
Bhairava spoke:
Listen to me, O Goddess, I shall teach you the extraordinary intercourse
(saṃga) with Māyā. It is fruitful, O Maheśānī, and difficult to learn
via free access
by others and yogins without yogic Powers (siddhi), O Suranāyakī.

(40.45–46ab)
[parameśvaradhyānam]
sugupte mandire mantrī mṛdvāsanaparigrahaḥ ||40.46||

bhāvayec ca svam ātmānaṃ parameśvaravigraham |
kāmeśvaram ivādyantaṃ sūryāyutasamaprabham ||40.47||
cārumañjīrakeyūrakuṇḍalāṅgadabhūṣitam |
mudrikāchannahīrādikirīṭamukuṭojjvalam ||40.48||
prasannavadanaṃ kāntaṃ tāmbūlāpūritādharam |
madirānandacetaskaṃ paramānandavigraham ||40.49||
navayauvanasampannaṃ sarvalakṣaṇabhūṣitam |
svavāmabhāgavinyastamahājagavakārmukam ||40.50||
dakṣabhāgojjvalatpañcaśaram indīvaradyutim |
nīlotpalalasanmālābhūṣitoraskam īśvaram ||40.51||
madena kṣubdhahṛdayam īṣāsmitamukhāmbujam |
evaṃ dhyāyec ciraṃ yogī svaśarīraṃ śivātmakam ||40.52||
candanāgarukarpūrakuraṅgajayakuṅkumaiḥ |
adhivāsitasarvāṅgaṃ cāruvaktravirājitam ||40.53||
ratnadvīpāyutayutaṃ gehe sattalpamadhyagam |
46c sugupte mandire ] J; sugupto māḥdere Well 46d mṛdvāsana° ] J; mṛddhāsana° Well
47b °eśvara° ] JaJbpcJcWell; °eśva° Jbac 47d °sama° ] J; omitted in Well 48b °kuṇḍalāṅ-
gadabhūṣaṇam ] Jab; °kuṇḍalāṅgadabhūṣitam Jc, °kumbhalāṅgadabhūṣitam Well 48d
°ojjvalam ] Jc; °ojvalam JabWell 49a prasannavadanaḥ ] JaJc; prasaḥnaḥ vadanaḥ Jb,
prasannavadanāḥ Well 49b tāmbūlāpūritādharam ] JabWell; tā×būlāpūridharam Jcac,
tā×būlāpūritādharam Jcpc 50d °mahājagava° ] conj.; °mahadaikṣava J, °mahavaivakṣa°
Well 51a dakṣa° ] J; rakṣa° Well ● °ojjvalat° ] corr.; °ojvalat° JWell 51c nīlotpala° ]
JbJcWell; nīlo×la° Ja ● °lasan° ] J; °lasam° Well 52a kṣubdha° ] JaJc; kṣucca° Jb, kuṣu[?]°
Well 52b īṣāsmita° ] J; īsmita° Well ● mukhāmbujam ] JaJcWell; mukhāḥ Jb 52a na-
vipulā 52c ciraḥ ] J; cīraḥ Well 53a °āgaru° ] J; °āgara° Well ● °karpūra° ] JaJcWell;
°karpūraḥ Jb 53d cāru° ] J; cāruḥ Well ● °virājitam ] J; °virājitaḥḥ Well 54a na-vipulā
54b gehe sattalpa° ] Ja; gehe satralpa° JbJcWell
[Visualisation of Parameśvara]
In a hidden sanctuary, the mantra master should sit on a soft cushion31

and should visualise himself as having the body of Parameśvara, as if
31 The element °parigrahaḥ in the compund mṛdvāsanaparigrahaḥ is suspicious. A word

meaning “seated” would fit the context better.
via free access
438 kiss
[he were transformed into] Kāmeśvara,32 having no beginning and no

end, shining like millions of suns. (40.46cd–47)
He is adorned with nice anklets, armlets, rings and bracelets, and he
shines with small toe rings (mudrikā),33 channahīras,34 etc., and
diadems and a crown. (40.48)
His face is gracious, beautiful, his lips are smeared with betel leaves. His
mind is filled with the joy of wine,35 and his body is supreme bliss
[itself]. (40.49)
He is in the prime of his youth and has all the auspicious characteristics.
He has the great Ajagava36 bow placed on his left side. (40.50)
On his right, he has five glowing arrows. He is shining like a blue lotus.
On his chest there is a glittering garland of blue lotuses. He is the
Lord. (40.51)
His heart is agitated with sexual desire. His lotus face displays a faint
smile.37 This is how the yogin should visualise his body for a long
time, as transformed into Śiva. (40.52)
All his limbs are perfumed with sandal, aloe, camphor, musk38 and saf-
fron. He has a beautiful face. (40.53)
He is surrounded by millions of gem islands, in a chamber on a fine
bed.39 (40.54ab)
32 Note that Kāmeśvara features as Śiva or the central deity in the pre-Śrīvidyā Dakṣiṇām-
nāya tradition, with which the MaSaṃ is clearly affiliated. See Sanderson (1988, 688), and
Kiss (2009, 18 and 42–43).
33 This is somewhat tentative. See mudrikā as an ornament in Brahmayāmala 21.63cd:
mudrikām aṅguliś caiva pādau laktakarañjitau (Kiss 2015, 219).
34 A channahīra, or rather a channavīra, is made up of two sacred threads ( yajñopavīta)
worn over the two shoulders and across the chest. Bunce (1997) provides two definitions:
“Channavira–(Ind.: channa-vīra) A Hindu iconographic object for bodily adornment. The
term channavira refers to a chain worn by both male and female deities. It is made up of
two chains crossed over the chest, a disc covers the front crossing” (Bunce 1997, 58). “The
term chhannavira refers to two sacred cords similar to yajñopavita. One is placed over each
shoulder, crossing on the chest and back and looping as low as the hips” (Bunce 1997, 63).
See also Rao 1914, vol. 1.2, xxxi (Addenda), where it is defined as a double yajñopavīta. See
channavīra mentioned in e.g. Rauravāgama, Kriyāpāda 10.52d.
35 Note that here intoxication is something only to be visualised.
36 Note that mahājagava is a rather insecure conjecture for mahadaikṣava and mahavai-
vakṣa. Other variants of the name of Śiva’s bow are ajakava and ajīkava.
37 Note the slightly odd form īṣāsmita° metri causa for the standard īṣadasmita°.
38 I take kuraṅgajaya in the sense of “musk,” although I have not found any evidence for this
compound being used instead of the well-known kuraṅganābhi.
39 I am grateful to Harunaga Isaacson for his assistance with this passage.
via free access
[śaktidhyānam]
tatra svavāmabhāgasthāṃ śaktiṃ bhuvanamohinīm ||40.54||

sarvalakṣaṇasampannāṃ navayauvanagocarām |
nīlālakasamābaddhamālālolupaṣaṭpadām ||40.55||
kastūrīsāndrakarpūratilakāṃ kamalekṣaṇām |
kuṇḍalāṅgadakeyūragraiveyādivibhūṣitām ||40.56||
akalaṅkaniśānāthasadṛśaśrīmukhāmbujām |
tāmbūlāpūrṇavadanāṃ svarṇakumbhopamastanīm ||40.57||
divyānulepavastrāḍhyāṃ vistīrṇajaghanāntarām |
cārūrujaṅghāṃ saundaryasārasarvasvavigrahām ||40.58||
sarvalāvaṇyasaundaryasārasarvasvavigrahām |
mañjīrāñcitapādāḍhyāṃ divyamālyānulepanām |
madirāsvādamuditāṃ madanāviṣṭavigrahām ||40.59||
vilāsavibhramāṃ kāntāṃ dhyāyet śaktiṃ maheśvari |
54c sva° ] JaJcWell; omitted in Jb 55b °gocarām ] J; °gaucarām Well 55c nīlā° ] conj.; līlā°
JWell ● °baddha° ] J; °baddhaḥ Well 55d °mālā° ] J; °māla° Well ● °lolupa° ] conj.;
°lotuya° JWell 55d°padām ] em.; °padam JWell; 56a kastūrī° ] JaJc; kastū° Jb, kastūra°
Well 56b °tilakāḥ ] J; °tilaḥkāḥ Well ● °ekṣaṇām ] Jab; °ekṣaṇam JcWell 56c °keyūra° ]
J; °keyūraḥ Well 57a °nātha° ] J; °nāthā° Well 57b °mukhāmbujām ] JcWell; °yukhām-
bujāḥ Ja, °yukhāmbujaḥ Jb 57d °opamastanīm ] Ja; °opastanīm Jb, °opamastakīm JcWell
58a °āḍhyāḥ ] conj.; °ādyāḥ JabcWell 58b °jaghanā° ] em.; °jayanā° JWell; see Kauṇḍinya’s
commentary on Pāśupatasūtra 1.9: adhomukhenādaṃṣṭreṇa jaghanāntaracāriṇā 58b
°ntarām ] JaJcWell; °ntaram Jb 58c saundarya° ] J; soṃdaryaṃ° Well 58d °vigra-
hām ] Jab; °vigraham JcWell 58c ma-vipulā 59a °pādāḍhyāṃ ] Ja; °pādādyāṃ JbJcpc,
°pādyādāṃ Jcac, °pādāyāṃ Well 59b °mālyānulepanām ] JapcJb; °mālyānulepanam Jaac,
°mālānulepanām JcWell 59d madanāviṣṭa° ] Jab; madanā° JcWell 59a na-vipulā 60b
dhyāyet ] JapcJc; dhyā Jaac, dhyāye Jb, madhyāt Well ● śaktiṃ ] J; sakti Well
[Visualisation of Śakti]
On his left side, [he should visualise] Śakti, who infatuates the world.
(40.54cd)
She has all the auspicious characteristics. She is in the prime of her
youth. She has bees longing for the garland tied in her black locks.
(40.55)
The tilaka-mark on her forehead is made with musk thickened with
camphor. She has lotus-eyes. She is adorned with rings, armlets,
anklets, necklaces etc. (40.56)
Her beautiful lotus face resembles the spotless moon. Her mouth is
filled with betel. Her breasts are like golden jars. (40.57)
via free access
440 kiss
She is anointed with divine ointments and she is dressed in divine

clothes, with her loins exposed.40 Her thighs and shanks are beau-
tiful. Her body is the ultimate essence of gracefulness. (40.58)
Her feet are embellished41 with anklets. She wears divine garlands and
[has been anointed] with divine ointments. She is delighted by the
wine she is enjoying. Her body is filled with passion. She is restless
with wantonness. [This is how the yogin] should visualise his lover
(kāntā) as Śakti, O Maheśvarī. (40.59–60ab)
The appearance of Kāmeśvara and the mention of a gem-island (ratnadvīpa)

in the above verses suggest an affiliation with the Śrīvidyā tradition and with
love magic, perhaps with that of the pre-Śrīvidyā Dakṣiṇāmnāya tradition.42
Verse 46cd (“in a hidden sanctuary”) may also be revealing: if the whole ritual
were purely imaginary, it would be less important to perform it in a secret place.
Going further in our text, MaSaṃ 40.60cd–68 may also suggest that a real
sexual encounter is being described by using words such as āśliṣya (“embrac-
ing”), samācaret (“he should perform”), bahiḥ (“outwardly”), kṣipet (“place [his
hand]”), vimuñcati (“ejaculates”), by giving instructions in 40.61 to stimulate
the partner, and also by avoiding words that would refer to visualisation, except
dhyāne, probably hinting at the fact that the whole process is accompanied by
a projection of the image of the Goddess onto the female partner. On the other
hand, the formulation of verse 40.65 may cast some doubt upon the real-life
presence of the female partner: the yogin should rub his semen on his body,
and there is no mention of the female partner’s sexual fluids or her receiving
or consuming any of the magical mixture, a common practice in Śaiva sexual
rituals.43 Here follows the end of the chapter, describing the sexual ritual:
40 Ex em. Compare Kauṇḍinya’s commentary ad Pāśupatasūtra 1.9 (p. 14): adhomukhenā-

daṃṣṭreṇa jaghanāntaracāriṇā | sarvaśāstrācikitsyena jagad daṣṭaṃ bhagāhinā ||, which
is translated by Hara (1966, 196) as “The world is bitten by a snake whose mouth is below,
toothless, who has crawled between the loins and whose poison can be cured by no sci-
ence. Its name is the vulva.” The passage can also interpreted as “The world is bitten by
the snake of the vagina whose toothless mouth points downward, and who moves in the
genital area, and for which there is no antidote.”
41 I have translated añcita as “embellished,” suspecting that the word may need some emen-
dation (aṅkita/rañjita?).
42 See Sanderson 1988, 688–689 and 2009, 47 ff.
43 See e.g. BraYā 45.201–202ab (Kiss 2015, 142, 258): upaviśyāpayet tatra cumbanādyāvagūha-
nam | kṛtvā kṣobhaṃ samārabhya pavitraṃ gṛhya sādhakaḥ || prāśayitvā tu tau hṛṣṭau
yāgadravyāṇi prokṣayet |. “The Sādhaka should make [her] sit down there. He should start
kissing and embracing her and stimulating her. He should collect the purifying [substance,
i.e. the sexual fluids]. Overjoyed, they should consume [the fluids] …”
via free access
[saṃgaḥ]
āśliṣya cumbanādīni yogī samyak samācaret ||40.60||

dhyāne bahiḥ susaṃsnigdhaṃ tayā saṃgaṃ samācaret |
gajahastaṃ kṣipet tasyāḥ śrīmanmadanamandire ||40.61||
mantram enaṃ smared yogī madaṃ yāvad vimuñcati |
bhautikavyoma †lā aiṃḍi† kevalārṇas tu bhautike ||40.62||
aparo vahni vāmākṣi bindu yukto maheśvari |
calepadaṃ calapadam citrepadam anantaram ||40.63||
retopadaṃ muñcapadayugaṃ pūrvabījā vilomagāḥ |
evaṃ krameṇa yo yogī māyāsaṃgaṃ samācaret ||40.64||
tadvīryaṃ svarṇakarpūrakuṅkumādiviloḍitam |
svadehaṃ mardayet kāntiś candravat samprajāyate ||40.65||
samūlāṃ brahmamaṇḍūkīṃ chāyāśuṣkāṃ prasādhayet |
mṛdvīkārasasaṃmiśrāṃ śarkarāghṛtamelitām ||40.66||
trimāsaṃ bhakṣayet kālatrayam akṣapramāṇakam |
annapānaṃ payaḥ pītvā nāsya śukraḥ kṣayaṃ vrajet ||40.67||
śatavarṣasahasrāntaṃ yadi saṃgaṃ samācaret |
nāsya prakṣīyate satyaṃ navatvāptyai tanoḥ priye ||40.68||
rasāyanam idaṃ guhyam anyeṣāṃ na prakāśayet |
etad rahasyaṃ vyākhyātaṃ durlabhaṃ siddhasaṃtatau |
gurukāruṇyasaṃlakṣyaṃ kiṃ bhūyaḥ śrotum icchasi ||40.69||
|| iti śrīmatsyendrasaṃhitāyāṃ catvāriṃśaḥ paṭalaḥ ||
60c cumbanā° ] J; cumbunā° Well 60d samyak ] JaJc; samyā[?]k Jb, samyas Well ●
samācaret ] JapcJbWell; samācare Jaac 61a dhyāne ] Jab; dhyāye JcWell ● bahiḥ ] corr.;
bahi J, vadi Well ● °snigdhaḥ ] Jab; °śligdhaḥ Well 61d śrī° ] J; ślī° Well 62b vimuñcati ]
JaJcWell; vimuḥtica Jb (metathesis) 62d kevalārṇas ] J; kaivalārṇas Well ● tu bhautike ] J;
tv abhautike Well 62 marginalia in f. 64r of Ja (top): hraiṃ tvaiṃ rīṃ cate vata citte
reto muñca 2 rīṃ klaiṃ hraiṃ. 63a vahni° ] Well; vahvi° J 63b bindu° ] J; bindud°
Well ● maheśvari ] J; maheśvarī Well 63c cale° ] J; cile° Well ● cala° ] J; calā° Well 64a
muñcapada° ] corr.; muñcapadaṃ J, mucapadaṃ Well 64a hypermetrical 64b °yugaṃ
] J; omitted in Well 65a °vīryaṃ ] JaJcWell; °bījaṃ Jb ● °karpūra° ] Well; °karpūraṃ
J 66b chāyā° ] JaJcWell; chāyāṃ° Jb ● °śuṣkāṃ ] J; °śu×āṃ Well 65c mardayet ] J;
marddhayet Well 66c °miśrāṃ ] J; °miśroṃ Well 66d °melitām ] JaJcWell; °melikām
Jb 67a trimāsaṃ ] J; trimāse Well ● kāla° ] JaJcWell; kā° Jb 67d °pramāṇakam ] Jab;
°pramāṇakraṃḥ Well 67c °pānaṃ ] J; °pātaṃ Well 68b saṃgaṃ ] corr.; saṃga JWell
68c nāsya ] Ja; nākṣasya Well ● satyaṃ ] JbWell; ×yaṃ Ja 68d navatvāptyai ] JaJcWell;
navātvāstha Jb 69a guhyam ] J; guhyaṃm Well 69d °labhaṃ ] J; °lebhaṃ Well 69e
°kāruṇya° ] JabWell; °kāruṇyaṃ Jc 69 f. bhūyaḥ ] J; bhūya Well ● śrotum ] JaJcWell;
śrītum Jb Colophon: catvāriṃśaḥ ] J; catvāriṃśat Well
via free access
442 kiss
[The intercourse]
The yogin should embrace and kiss her, etc., properly. (40.60cd)
[Then] he should have sex with her outwardly, very gently, while
[performing] visualisation. He should apply the “elephant trunk”
[method]44 on her divine love temple [i.e. her genitalia]. (40.61)
The yogin should recall this mantra when he ejaculates: Bhautika
[ai], Vyoman [h], […], tu verbatim and Bhautika [ai; i.e. hraiṃ
tvaiṃ].45 (40.62)
Moreover: Vahni [r] and Vāmākṣi [ī] with a Bindu [ṃ; i.e. rīṃ], O Mah-
eśvarī, the words cale and cala and immediately citre [citte?],
(40.63)
the word reto, muñca twice,46 and the previous seed-mantras back-
wards [rīṃ tvaiṃ hraiṃ]. The yogin who has had sex with Māyā
should rub his semen mixed with gold, camphor and saffron on his
own body: [his] beauty will become moon-like. (40.64–65)
He should dry brahmamaṇḍūkī47 together with its roots in the shade.
He should mix it with grape-juice, candied sugar and ghee. (40.66)
He should have it three times [a day] for three months in portions mea-
suring a dice as food and drink and he should drink milk. His semen
will not deteriorate in millions of years if he practises sex [with
Māyā]. His [semen] will never ever wane. It is for the rejuvenation
of the body, O Priyā. (40.67–68)
This is the secret of alchemy. He should not reveal it to others. This
secret of the Siddha tradition, which is difficult to obtain, has now
been taught. It is to be revealed through the compassion of the guru.
What else do you wish to hear? (40.69)
Here ends the fortieth chapter of the Matsyendrasaṃhitā.
44 See p. 443 below.

45 If the decoding of this mantra in the marginalia in f. 64r of MS Ja (top, hraiṃ tvaiṃ
rīṃ cate vata citte reto muñca 2 rīṃ klaiṃ hraiṃ) is more or less correct, the
puzzling syllables lā aiṃḍi must stand for r. Note the slight differences between the code
in the text and the mantra given in the marginalia. The reconstruction of the mantra is
somewhat tentative.
46 See Kaulāvalīnirṇaya 5.51: amukīṃ drāvaya svāhā vinyaset sādhakottamaḥ | vāmāyā[?]
capalacitte reto muñcadvayaṃ paṭhet.
47 Clerodendrum Siphonantus?
via free access
Rubbing one’s own semen on one’s body (verse 65) is an old custom, and
was probably not considered a transgressive practice at all. Bṛhadāraṇyaka-
upaniṣad 6.4.4–5 describe a practice to be followed in case one discharges
semen:
bahu vā idaṃ suptasya vā jāgrato vā retaḥ skandati || 6.4.4 ||

tad abhimṛśed anu vā mantrayeta—yan me ’dya retaḥ pṛthivīm
askāntsīd yad oṣadhīr apy asarad yad apaḥ | idam ahaṃ tad reta
ādade | punar mām aitu indriyaṃ punas tejaḥ punar bhagaḥ | punar
agnir dhiṣṇyā yathāsthānaṃ kalpantām | ity anāmikāṅguṣṭhābhyām
ādāyāntareṇa stanau vā bhruvau vā nimṛjyāt ||6.4.5||
In Hume’s translation (1921, 168–169):
“[If] even this much semen is spilled, whether of one asleep or of one
awake, [5] then he should touch it, or [without touching] repeat:—
‘What semen has of mine to earth been spilt now,
Whate’er to herb has flowed, whate’er to water—
Again to me let vigour come!
Again, my strength; again, my glow!
Again the altars and the fire
Be found in their accustomed place!’
Having spoken thus, he should take it with ring-finger and thumb and
rub it on between his breasts or his eye-brows.”48
As can be seen from this passage, semen was probably never considered so
impure as to forbid its magical application on the body, and this practice was
recommended to regain strength—exactly as in MaSaṃ 40.68 (navatvāptyai
tanoḥ). The question is inevitable: is the somewhat similar practice in MaSaṃ
40.65 recommended for a similar situation, namely for ejaculation in a state of
mind comparable to sleep, i.e. visualisation? Does this similarity between the
two instructions suggest the absence of the female partner in the MaSaṃ?
To return to 40.61cd, where stimulation of the partner is probably hinted at:
this gives the instruction to “cast an ‘elephant trunk’ on the partner’s divine
love temple” (gajahastaṃ kṣipet tasyāḥ śrīmanmadanamandire). This again
is ambiguous. Gajahasta could be a mudrā to be shown during the ritual,49
48 I find Olivelle’s translation (1996, 88) slightly less accurate at this point, although the dif-
ferences are minor.
49 See e.g. Kaulāvalīnirṇaya 17.133–135ab: eṣā tu paramā mudrā sarvasaṅkṣobhaṇī matā |
via free access
444 kiss
but could as well be the technique mentioned in Jaśodhara’s commentary ad

Kāmasūtra 7.2.2:
ratasyopakrame sambādhasya kareṇopamardanaṃ tasyā rasaprāptikāle

ca ratayojanam iti rāgapratyānayanam |
Commentary: ratasyeti | samprayogasya upakrama iti | ayam ārambhe,

yady api mando rāgo rate pravartayati stabdhaliṅgatvāt, tathāpi pratha-
mataḥ sambādhasya bhagasya kareṇopamardanaṃ gajahastena kṣobha-
ṇaṃ kāryaṃ …
[When you are] about to practise sex, [first you should] rub her genitalia
with your hand, and when there is dampness, the sexual act can be com-
menced. This is the restoration of passion.
Commentary: “about to practice sex”: at the beginning of the sexual act.

This is at the start [of the sexual act]. Even if the passion is weak with
regards to sex because the penis is inert, first “her genitalia,” i.e. her vulva,
should be rubbed with his hand, should be stimulated with the “elephant
trunk” [method] …
The possible hint in MaSaṃ 40.61cd at a Kāmaśāstric technique again suggests

that we are dealing with the description of a sexual ritual involving a human
female partner. Or should this also be only visualised?
3 Why All These Ambiguities?
I think it is safe to say that the teaching of MaSaṃ 40 is ambiguous to an

extent that makes it rather difficult to decide on one or another exclusive
interpretation. The question to answer is rather: why is it so ambiguous? Is it
deliberately so? Is it so due to bad writing, to sloppy composition? Are essen-
tial details left out because they were well-known to gurus and pupils at the
time of composition of the text? Did revisions/insertions during the course
of transmission cause the ambiguity (deliberately or accidentally)? Is the text
kṣobhayed athavā mantrī gajahastākhyamudrayā || adhomukhaṃ dakṣapāṇiṃ nidhāyā-

ṅguṣṭhake same | niḥkṣiped aṅgulīḥ sarvā gajatuṇḍākṛtir yathā || gajahastā mahāmudrā
kathitā siddhidāyikā |.
via free access
ambiguous because of the uncertainty of its redactors, i.e. they were rephras-
ing old teachings but were not sure of the details? Is this ambiguity the result
of the redactors’ diffidence in expressing secret teachings on sexuality? Could
this ambiguity be seen as indicative of some major change in the tradition?
The first possibility, namely that the text is ambiguous deliberately, would
imply that the author(s) or redactor(s) wanted to hide their secret teaching
from unauthorised eyes. This is possible but not as typical as one would think.
For instance, while they are difficult for the modern reader to decipher, the
BraYā’s radical teachings on sexual ritual are relatively straightforward con-
cerning what is real and what is imaginary, and while there are technical terms
which are not openly discussed in the text,50 and its language is far from being
standard Pāṇinian Sanskrit, it is generally possible to understand how sexual
rituals were supposed to be performed.
The next possibility hinted at above, i.e. the effect of bad writing, is possible,
but the MaSaṃ is far from being a very cryptic or confused text. It would have
required minimal effort and ability on behalf of the redactors to clarify details
that we miss: a few lines on how to acquire or invite a Māyā woman, similar to
the BraYā’s instructions on finding a partner,51 a verse on her role and position
during the ritual, or a clear remark stipulating the yogin’s solitude would have
been enough.
On the other hand, one should not forget that texts like the MaSaṃ were
definitely not written with an outsider reader in mind who would try to under-
stand them several hundred years later. Essential details could have been left
out because they were obvious to the redactors.
As regards possible revisions and insertions, there are signs that MaSaṃ 40 is
made up of at least three distinct parts. Verses 40.1–28 constitute a small chap-
ter in themselves with weak links to the rest of the chapter. What they have
in common is the mention of “Kaula conduct” and bodily fluids. 40.29–37 and
the rest of the chapter are more closely related. Both mention and discuss sex-
ual rituals, but the first section, while mentioning sex with Yoginīs, is silent on
Māyā, the focus of the second part, i.e. verses 40.38–69, which seem as if they
were an alternative and additional teaching. But this additional section may
again be made up of passages drawn from various sources. The visualisation
of Kāmeśvara and his partner (40.46cd–60ab) may come from a source dif-
ferent from that of the instructions on the sexual act itself, and this in itself
would provide some explanation for discrepancies in the text. The text may
50 E.g. the term avagraha (“restraint”) or pīṭha (“external genital organ of the female part-
ner”). See Kiss 2015, 49, 47–48.
51 See BraYā 45.185cd–189ab.
via free access
446 kiss
have originally described a sexual ritual with a human partner, but during the
transmission of the text some passages teaching new ideas (such as a visualised
partner) were inserted, thus making the text ambiguous.
This may lead us to another possibility, namely that at some point the redac-
tors of the text became uncertain of the exact details of the ritual and when
they tried to solve the problem they may have ended up obscuring it even
further. Finally, the possibility of diffidence may also have played some role,
especially if the female partner was meant to be purely imaginary, but other
details of the ritual were not.
It seems that a wider perspective may be required to see what this chapter
of the MaSaṃ signifies. The ambiguity between actual sex and visualisation
in chapter 40 may have its roots in the tension between sexuality and asceti-
cism which is clearly manifest in the frame story of the text (chapters 1 and 55).
The frame story contains a unique version of the legend of Matsyendra and
Gorakṣa: Matsyendra occupies the body of a dead king and indulges in sensual
pleasures. It is Gorakṣa, his disciple, who “rescues” Matsyendra from the trap of
sexuality and power, and leads him back to an ascetic life.52 Taking into account
this wider context, it is possible to discover the same tension in the teaching of
MaSaṃ 40 between a sexual ritual that may have originated in an earlier tantric
strata of the cult, and its probably later haṭhayogic layers.
As I suggest elsewhere,53 the MaSaṃ could provide clues about the transi-
tion of a tantric cult from Kaula practices, often involving transgressive ele-
ments, to early Haṭhayoga, often associated with brahmacārin practitioners.
The text may be echoing or quoting old tantric texts with such descriptions
of sexual rituals that aim at obtaining sexual fluids, but the redactors of the
MaSaṃ were perhaps in a transition towards more ascetic or brahmacaryā-
oriented teachings, and as a result, they come up with a fairly obscure variant of
the figure of the tantric Yoginī: Māyā, first described as only a phantom, resem-
bles the wholly mental visualisation of goddesses, but at the same time takes
part in a human sexual ritual. The redactors may have had reservations about
a sexual ritual with a low-caste woman, and tried to conceal this with instruc-
tions on visualisation to such an extent that even the presence of a human part-
ner is now doubtful. They may also have had kuṇḍalinī in their thoughts: the
52 See Kiss 2009, 222–233 and 317–321.

53 Kiss 2009, 9: “[…] the MaSaṃ provides some clues for, among other things, the under-
standing of the transition from the early Indian yoga traditions (Pātañjala and Śaiva) to
the late and fully developed haṭha-yogic teachings as well as of the transition from the
early Kula traditions to the later Kaula teachings associated with the figure of Matsyen-
dra.”
via free access
text as it stands now could be a metaphor for meditation on her. The manner in
which they reconcile two (or three) attitudes, in this case those of explicit sexu-
ality and of brahmacārin yogins’ mental worship of a goddess (or of kuṇḍalinī),
is, as so often in tantric texts, less than convincing. But this imperfection, this
ambiguity, is exactly the feature which seems to tell us something about the
history of the cult, its transition from one phase to another.
Acknowledgements
I would like to thank Judit Törzsök, Dominic Goodall, Harunaga Isaacson and
Gergely Hidas for their constant help and support. I am especially grateful to
Shaman Hatley and Gergely Hidas for reading and commenting on a draft of
this article.
References
Primary Sources
Upaniṣads
Patrick Olivelle, trans. Oxford World’s Classics. Oxford: Oxford University Press, 1996.
Kāmasūtra
Srī Gosvamī Dāmodar Shastri, ed. The Kāmasūtra by Srī Vātsyāyana Muni. With the
Commentary Jayamangala Of Yashodhar. Kāshi Sanskrit Series, no. 29. Benares:
Chowkhamba, 1929.
Kubjikāmata
T. Goudriaan and J. Schoterman, eds. Kubjikāmatatantra. The Kulālikāmnāya version.
Leiden: E.J. Brill, 1988.
Kaulāvalīnirṇaya
Arthur Avalon, ed. Calcutta: Sanskrit Press Depository, 1928.
Khecarīvidyā
See Mallinson 2007.
Tantrāloka
Madhusūdan Kaul Śāstrī, ed. Tantrāloka of Abhinavagupta with commentary (-viveka)
of Rājānaka Jayaratha. KSTS nos. 23, 28, 30, 35, 29, 41, 47, 59, 52, 57 and 58. Bombay
and Srinagar, 1918–1938.
via free access
448 kiss
Nityāṣoḍaśikārṇava/Vāmakeśvarīmata
Madhusūdan Kaul Śāstrī, ed. Vāmakeśvarīmatam, with the commentary Rājanaka Jaya-
ratha. Kashmir Series of Texts and Studies, no. 66. 1945.
Nityotsava of Umānandanātha
A. Mahadeva Sastri, ed. Baroda: Oriental Institute, 1948.
Pāśupatasūtra
Pāśupatasūtram. Śrīmaharṣikauṇḍinyakṛtabhāṣyopetam. Śrīmādhavānandagrantha-
mālā, no. 10. Varanasi: Shri Krishnananda Sagar, 1987.
Bṛhadāraṇyaka-upaniṣad
Nārāyaṇ Rām Ācārya, ed. Īśādiviṃśottaraśatopaniṣadaḥ. A Compilation of Well-Known
120 Upanisads. Bombay: Nirṇay-sāgar, 1948.
Brahmayāmala
National Archives Kathmandu, ms. no. 3–370.
See Hatley 2007; 2018.
See Kiss 2015.
Matsyendrasaṃhitā
See Kiss 2009; 2020 (forthcoming).
(Ja) Maharaja Man Singh Library, Jodhpur (MMSL), No. 1784.
(Jb) No. 1783, MMSL.
(Jc) No. 1782, MMSL.
(Jd) No. 1785, MMSL (not available for chapter 40).
(Well) β 1115, Wellcome Library for the History and Understanding of Medicine, Lon-
don.
Rauravāgama
3 vols. N.R. Bhatt, ed. Pondichéry: Institute Français d’Indologie, 1985–1988.
Secondary Sources
Biernacki, Loriliai. 2007. Renowned Goddess of Desire: Women, Sex, and Speech in Tantra.
New York and Oxford: Oxford University Press.
Bunce, Frederick W. 1997. A Dictionary of Buddhist and Hindu Iconography. New Delhi:
D.K. Printworld Ltd.
Dupuche, John R. 2003. Abhinavagupta: The Kula Ritual, as Elaborated in Chapter 29 of
the Tantrāloka. Delhi: Motilal Banarsidass Publishers.
Hara, Minoru. 1966. “Materials for the study of Pāśupata Śaivism.” Ph.D. thesis, Harvard
University.
via free access
Hatley, Shaman. 2007. “The Brahmayāmalatantra and Early Śaiva Cult of Yoginīs.” Ph.D.
thesis, the University of Pennsylvania.
Hatley, Shaman. 2018. The Brahmayāmalatantra or Picumata. Volume I. Chapters 1–2,
39–40 & 83. Revelation, Ritual and Material Culture in an Early Saiva Tantra. Collec-
tion Indologie, no. 133 (Early Tantra Series, no. 5). Institut Français d’Indologie/École
française d’Extrême-Orient/Universität Hamburg.
Hume, Robert Ernest. 1921. The Thirteen Principal Upanishads. Oxford: Oxford Univer-
sity Press.
Kiss, Csaba. 2007. “Notes on the Matsyendrasaṃhitā.” In Indian Languages and Texts
through the Ages. Essays of Hungarian Indologists in Honour of Prof. Csaba Töttössy,
edited by Csaba Dezső, 147–184. Delhi: Manoharlal.
Kiss, Csaba. 2009. “Matsyendranātha’s Compendium (Matsyendrasaṃhitā): A Critical
Edition and Annotated Translation of Matsyendrasaṃhitā 1–13 and 55 with Analy-
sis.” Ph.D. thesis, University of Oxford.
Kiss, Csaba. 2011. “The Matsyendrasaṃhitā: a Yoginī-centered 13th-century Yoga text of
the South Indian Śāmbhava Cult.” In Yogi Heroes and Poets: Histories and Legends
of the Nāths, edited by David Lorenzen and Adrián Muñoz, 143–162. Albany: State
University of New York Press.
Kiss, Csaba. 2015. The Brahmayāmala Tantra or Picumata, Volume II. The Religious
Observances and Sexual Rituals of the Tantric Practitioner: Chapters 3, 21, and 45. Col-
lection Indologie, no. 130 (Early Tantra Series, no. 3). Pondichéry: Institut Français
d’Indologie/École française d’Extrême-Orient/Universität Hamburg.
Kiss, Csaba. 2020 (forthcoming). The Yoga of the Matsyendrasaṃhitā. A Critical Edition
and Annotated Translation of Matsyendrasaṃhitā 1–13 and 55. Pondichéry: Institut
Français d’Indologie/École française d’Extrême-Orient.
Mallinson, James. 2007. The Khecarīvidyā of Ādinātha. A Critical Edition and Annotated
Translation of an Early Text of Haṭhayoga. London; New York: Routledge.
Rao, T. Gopinath. 1914. Elements of Hindu Iconography. 2 vols. (4 parts). Madras: Law
Printing House, 1914–1916.
edited by S. Sutherland, L. Houlden, P. Clarke and F. Hardy, 660–704. London: Rout-
ledge and Kegan Paul.
Sanderson, Alexis. 2002. “Remarks on the Text of the Kubjikāmatatantra.” In Indo-
Iranian Journal 45: 1–24.
Sanderson, Alexis. 2007. “Atharvavedins in Tantric Territory: The Āṅgirasakalpa Texts of
the Oriya Paippalaadins and their Connection with the Trika and the Kālīkula, with
Critical Editions of the Parājapavidhi, the Parāmantravidhi, and the *Bhadrakālī-
mantravidhiprakaraṇa.” In The Atharvaveda and its Paippalāda Śākhā: Historical
and Philological Papers on a Vedic Tradition, edited by Arlo Griffiths and Annette
Schmiedchen, 195–311. Geisteskultur Indiens: Texte und Studien, no. 11; Indologica
Halensis. Aachen: Shaker Verlag.
via free access
450 kiss
Sanderson, Alexis. 2009. “The Śaiva Age: The Rise and Dominance of Śaivism during the
Early Medieval Period.” In Genesis and Development of Tantrism, edited by Shingo
Einoo, 41–350. Institute of Oriental Culture Special Series, no. 23. Tokyo: Institute of
Oriental Culture, University of Tokyo, 2009.
Sanderson, Alexis. 2014. “The Śaiva Literature.” Journal of Indological Studies (Kyoto)
24–25 (2012–2013): 1–113.
Tāntrikābhidhānakośa. Dictionnaire des termes techniques de la littérature hindoue tan-
trique. Vol. 1. H. Brunner, G. Oberhammer et A. Padoux, eds. Österreichische Aka-
demie der Wissenschaften, Philosophisch-historische Klasse, Sitzungsberichte, Bd.
681. Beiträge zur Kultur- und Geistesgeschichte Asiens Nr. 35. Vienna: Austrian
Academy of Sciences Press, 2000.
White, David Gordon, ed. 2000. Tantra in Practice. Princeton; Oxford: Princeton Uni-
versity Press.
White, David Gordon. 2003. Kiss of The Yoginī: “Tantric Sex” in its South Asian contexts.
Chicago; London: The University of Chicago Press.
via free access
chapter 19
Haṭhayoga’s Floruit on the Eve of Colonialism

Jason Birch
My doctoral thesis (Birch 2013), which was supervised by Alexis Sanderson at

the University of Oxford, contained a survey of texts on Haṭha- and Rājayoga.
One of the challenges of completing such a survey was that very few of the texts
composed from the sixteenth to eighteenth century had been critically edited
or studied academically. Inspired by several exemplary surveys of Śaiva litera-
ture in Sanderson’s articles (e.g. 2001, 2007 and 2014), I visited a large number of
libraries in India in an effort to consult manuscripts of unpublished yoga texts.
By the end of my doctorate, it was apparent to me that yoga texts composed on
the eve of colonialism provided new insights into the history of yoga and, more
specifically, are crucial for understanding how Haṭhayoga changed after it had
been codified by Svātmārāma in the Haṭhapradīpikā (circa mid-fifteenth cen-
tury). In fact, after Svātmārāma had successfully transformed Haṭhayoga from
an auxiliary practice into a complete soteriological system, there began what
might be considered the floruit of Haṭhayoga, insofar as its literature flour-
ished, its systems of practice accumulated more techniques and it became,
particularly in scholarly compendiums on yoga, almost synonymous with the
auxiliaries of āsana and prāṇāyāma.
Building on my doctoral research, this article aims to provide a framework
for examining the textual sources of Haṭhayoga that were composed from the
sixteenth to eighteenth century. After a brief summary of the early literature
of Haṭhayoga, I shall discuss some of the salient features of the late literature
by dividing the texts into two etic categories; ‘extended works’ and ‘compendi-
ums.’ The extended works expatiate on Haṭhayoga as it was formulated in the
Haṭhapradīpikā, whereas the compendiums integrate teachings of Haṭhayoga
within a discourse on yoga more broadly conceived. Both categories include
scholarly and practical works which, when read together in this way, reveal sig-
nificant changes to both practical and theoretical conceptions of Haṭhayoga.
Such a reading also illuminates several developments of this time that fore-
shadowed, and in some cases inspired, the transnational yogas of the twentieth
century. The article concludes with a brief discussion on the regional extent of
the literature on Haṭhayoga during this period and how the codification of its
praxis and theory appears to have diverged in different regions.
© jason birch, 2020 | doi:10.1163/9789004432802_021

via free access
452 birch
1 Conceptions of Haṭhayoga before the Haṭhapradīpikā
The earliest references to the term haṭhayoga are found in some Buddhist
Tantras, most notably the Guhyasamājatantra and the Kālacakratantra, which
date to the eighth and eleventh centuries, respectively, and mention it as a
method of last resort when the primary techniques of these traditions had
failed (Birch 2011, 541–542). An eleventh-century commentary on the Kāla-
cakratantra, called the Vimalaprabhā, explains the term haṭhayoga as the name
of a type of yoga that forces prāṇa (‘vitality’) into the central channel through
a practice involving nāda (‘internal resonance’) and retention of bindu (‘gen-
erative fluids’). The earliest known Śaiva work to teach Haṭhayoga is the circa
twelfth-century Amaraughaprabodha (Birch 2019). Its Haṭhayoga is somewhat
consistent with that of the Kālacakra tradition insofar as both are auxiliary
practices that induce nāda. Nonetheless, a much closer counterpart to the
Amaraughaprabodha’s Haṭhayoga exists in an eleventh-century Vajrayāna
work called the Amṛtasiddhi (Mallinson 2020). Both have similar accounts of
three complex mudrās and a system of sounds (nāda), blisses (ānanda) and
voids (śūnya). It is important to note that the author of the Amṛtasiddhi does
not identify its yoga as Haṭhayoga. The reason for this is not stated in the text,
but Haṭhayoga appears to have been a controversial practice among some Bud-
dhist exegetes,1 and it is also possible that Haṭhayoga, or at the very least some
of its techniques of that time, had older associations with other religious tradi-
tions.2
Unlike earlier Buddhist works, Haṭhayoga in the Amaraughaprabodha is
embedded in a fourfold hierarchy in which Rājayoga is the principal yoga.
Rājayoga is defined as the absence of mental activity,3 a meditative state that
was known by this name in other Śaiva yoga texts of the same era.4 In this
hierarchy, Haṭhayoga was not the sole means to Rājayoga, because the latter
could also be achieved by Mantra- and Layayoga. Judging by later works, such
1 On Maitreyanātha’s and Rāmapāla’s rejection of Haṭhayoga, see Isaacson and Sferra 2014, and
Mallinson, forthcoming 2020.
2 On the prehistory of certain techniques which were integrated into Haṭhayoga, see Mallinson
2016, 120–122. In the case of the three physical techniques taught in the Amṛtasiddhi and the
Amaraughaprabodha, it seems possible to me that the Amaraughaprabodha may have bor-
rowed from a source that was older than the Amṛtasiddhi (Birch 2019, 964–966) and that the
physical practices themselves were not the preserve of esoteric Buddhists.
3 Amaraughaprabodha 3d ( yaś cittavṛttirahitaḥ sa tu rājayogaḥ).
4 The earliest work to teach Rājayoga by name is the Amanaska, which can be dated to
the eleventh or early twelfth century (Birch 2014, 406–409). In nearly all texts that teach
Haṭhayoga, Rājayoga is mentioned as the goal of Haṭhayoga.
via free access
haṭhayoga’s floruit 453
as the Śivasaṃhitā (5.13–28), which explain the rationale behind this hierarchy,
it appears that the characteristics of the student were the basis for determining
which yoga was taught to an individual, and it seems likely that Mantra-, Laya-
and Haṭhayoga were superfluous to students of extraordinary capability who
could achieve Rājayoga without an auxiliary practice.5
The praxis common to both the yoga of the Amṛtasiddhi and the Haṭhayoga
of the Amaraughaprabodha is three techniques called mahāmudrā, mahā-
bandha and mahāvedha. A repertoire larger than this rudimentary one appears
in all systems of Haṭhayoga that followed, such as that of the Dattātreyayo-
gaśāstra (circa 13th-century), a Vaiṣṇava work in which a collection of ten
mudrās, referred to as the Haṭhayoga of Kapila, was integrated with a Vai-
ṣṇava form of aṣṭāṅgayoga attributed to Yājñavalkya.6 Kapila’s collection of
mudrās consists of khecarī, viparītakaraṇī, the three bandha (“locks”), and
three variations of vajrolimudrā, in addition to mahāmudrā, mahābandha and
mahāvedha. Combinations of some of these mudrās appear in contemporary
Śaiva works, such as the Yogatārāvalī and the Yogabīja,7 which teach basic sys-
tems of Haṭhayoga. The latter text is known for its definition of Haṭhayoga as
the union of the sun and moon, which are represented by the syllables ha and
ṭha, respectively. This definition is absent from the earliest recension of the
Yogabīja, which simply defined Haṭhayoga as forcefully consuming the gross
elements of the body.8
Some of the mudrās in the Dattātreyayogaśāstra also appear in systems of
yoga of the same era that were not called Haṭhayoga as evinced, for example,
5 The Dattātreyayogaśāstra (14) states that Mantrayoga is for the lowest type of practitioner,
who has a weak intellect. This is why it is the lowest yoga of the hierarchy (alpabuddhir imaṃ
yogaṃ sevate sādhakādhamaḥ | mantrayogo hy ayaṃ prokto yogānām adhamas smṛtaḥ ||). A
passage in the long recension of the Amaraughaprabodha (17cd–24), which might postdate
the Haṭhapradīpikā, explicitly connects each of the four types of student to one of these four
yogas. Both this passage and the similar one in the Śivasaṃhitā, mentioned above, appear
to have been inspired by the Amṛtasiddhi’s discourse (chapters 15–18) on the four types of
student in relation to the four stages of yoga.
6 This Vaiṣṇava form of aṣṭāṅgayoga is taught (without the Haṭhayogic mudrās) in the Vasiṣṭha-
saṃhitā and Yogayājñavalkya, which were probably composed in the twelfth and fourteenth
centuries, respectively.
7 Recent work on nineteen manuscripts of the Yogabīja by the Haṭha Yoga Project has revealed
an early recension that does not teach these bandhas and mudrās. Nonetheless, it is likely that
a section on four kumbhakas and the three bandhas was added to the text before the time of
the Haṭhapradīpikā, in an attempt to explain the practice of śakticālana and Haṭhayoga.
8 Yogabīja, ms. no. 29917, f. 11v, line 5 (haṭhena grasyate jāḍyaṃ haṭhayogaḥ sa ucyate). The
“union of the sun and moon” definition was added to later recensions of the Yogabīja, one
of which may still predate the Haṭhapradīpikā.
via free access
454 birch
by two Śaiva works, the Vivekamārtaṇḍa and the Gorakṣaśataka.9 This suggests
that from the twelfth to the fifteenth century the practice of these particular
mudrās was more widespread than the use of the term haṭhayoga for designat-
ing a system of praxis. In fact, according to the available evidence, haṭhayoga is
used in this sense in only four non-Buddhist Sanskrit yoga texts that are likely
to predate the Haṭhapradīpikā, and two others whose dating is less certain.10
Unlike the asceticism and yoga of esoteric traditions, the texts of Haṭhayoga
do not mention the need for initiation (dīkṣā) for its practice, a characteristic
that appears to reflect haṭhayoga’s role as an auxiliary practice for people of
various religions and social status, including householders (Birch 2015, 8–10).
Although some of the distinguishing mudrās of Haṭhayoga, such as inverting
the body, may be similar to techniques of older traditions of asceticism (tapas),
the mudrās had been adapted and repurposed by tantric Buddhist and Śaiva
sects by the time texts such as the Amṛtasiddhi and the Amaraughaprabodha
were composed. None of the early teachings on Haṭhayoga refer to tapas and,
in contrast to the mortifying effects of extreme methods of tapas, the propo-
nents of Haṭhayoga claimed that this type of yoga would not afflict the body
and would, in fact, bring about health and jīvanmukti (“liberation-in-life”) rel-
atively quickly. Nonetheless, in this period there were opponents to Haṭhayoga
9 One might also include the Śivasaṃhitā as an example here. The fourth chapter of this
work teaches āsanas, prāṇāyāma and ten mudrās, as well as yonimudrā, which became
an integral part of the typology of the Haṭhapradīpikā. Chapters 1–4 of the Śivasaṃhitā,
in which these techniques are taught, do not refer to Haṭhayoga, which is mentioned only
briefly in the fifth chapter. The first four chapters may have been an original text that
was combined with the fifth sometime before the seventeenth century (Birch 2018, 107
note 13).
10 These are the Amaraughaprabodha, the Dattātreyayogaśāstra, the Yogabīja and the Yoga-
tārāvalī. The other two are the Śivasaṃhitā and the Aparokṣānubhūti, whose verses on
Haṭhayoga may not predate the Haṭhapradīpikā. On the dating of the Śivasaṃhitā, see
Birch 2018, 107, note 13. As far as I know, the date of the Aparokṣānubhūti is uncertain
and it is possible that its verses on Haṭhayoga were added more recently (Birch 2011, 540,
notes 98–100). A Sanskrit Vīraśaiva work called the Śaivaratnākara by Jyotirnātha men-
tions in passing the four yogas in the same order as the Yogabīja (i.e., Mantra, Haṭha-,
Laya and Rājayoga). According to Elaine Fisher (personal communication, 10 March 2019),
the Śaivaratnākara may have been composed in the late-thirteenth or early-fourteenth
century, which so happens to be the likely date of the early recension of the Yogabīja. A
large Sanskrit compendium called the Śārṅgadharapaddhati, probably dated to 1363CE,
mentions Haṭhayoga in a syncretic section on yoga that borrows from earlier yoga texts.
I am also aware of a Marathi text on yoga that may predate the fifteenth century, namely,
the Vivekadarpaṇa. This work defines but rejects Haṭhayoga in favour of a gnostic type
of Rājayoga. I would like to thank James Mallinson for drawing my attention to the
Vivekadarpaṇa and Elaine Fisher for the Śaivaratnākara.
via free access
who considered it to be like hard asceticism insofar as it was a cause of suffering

and unnecessary exertion. These opponents usually favoured effortless gnostic
methods for attaining liberation.11
The conception of Haṭhayoga in the Haṭhapradīpikā represents a turning
point in its history. The author of this work Svātmārāma incorporated a larger
repertoire of techniques than earlier works and synthesized diverse teachings
of various yoga traditions into a cohesive system, which he called Haṭhayoga.
Before Svātmārāma’s efforts, Haṭhayoga had been conceived as one of several
auxiliaries in hierarchical models of yoga. In the Haṭhapradīpikā, Haṭhayoga
is the sole means to Rājayoga and a complete soteriological system. The suc-
cess of Svātmārāma’s interpretation is attested by the fact that subsequent
authors borrowed much of the structure and content of the Haṭhapradīpikā
in creating more extensive works on Haṭhayoga. Also, the Haṭhapradīpikā was
widely quoted as an authority on its subject, in particular by erudite authors
of compendiums on yoga that were composed after the fifteenth century, and
it appears to have spread throughout most of India. Recently published cat-
alogues indicate that nearly two hundred manuscripts of the Haṭhapradīpikā
are held in libraries throughout India, from Kashmir to Tamil Nadu and Gujarat
to West Bengal in various scripts, as well as a few vernacular commentaries.
The Haṭhapradīpikā might be considered the culmination of a formative
period in the development of Haṭhayoga as a system of praxis. In this sense,
it marks the beginning of Haṭhayoga as a distinct method that combined both
seated and non-seated āsanas, the eight kumbhakas and the ten mudrās. In
keeping with the earlier literature, the main goal of these techniques was
Rājayoga. However, in addition to the mudrās of Haṭhayoga, Svātmārāma also
stipulated the practice of three other mudrās for Rājayoga, namely, śāmbhavī
mudrā, khecarīmudrā12 and a simplified form of ṣaṇmukhīmudrā, which is not
named as such, but is prescribed for “fusing the mind with the internal res-
onance” (nādānusandhāna).13 These “meditational” mudrās probably derive
11 Early examples of texts that critique and reject Haṭhayoga and its methods include the
Mokṣopāya/Laghuyogavāsiṣṭha and the Amanaska (Birch 2011, 531, 544–545).
12 Khecarīmudrā is taught in chapters three and four of the Haṭhapradīpikā. In the fourth
chapter, the practice of this mudrā consists of focusing the mind between the eyebrows,
rather than the Haṭhayogic khecarīmudrā of inserting the tongue into the nasopharyngeal
cavity, which is taught in the third chapter.
13 The Haṭhapradīpikā (4.68) states that the ears, eyes, nostrils and mouth should be blocked
in order to hear the internal resonance (nāda) in the suṣumnā channel. Brahmānanda’s
Jyotsnā commentary on this verse adds that the senses are blocked by the fingers, and
he calls this practice ṣaṇmukhīmudrā. A simpler technique is also mentioned at Haṭha-
pradīpikā 4.82ab, in which the yogin presses the ears with the hands to initiate nādānu-
via free access
456 birch
from older traditions of Rājayoga, one of which explicitly rejects the mudrās
of Haṭhayoga.14 Their inclusion in the Haṭhapradīpikā reflects Svātmārāma’s
efforts to bring together techniques of diverse traditions under the umbrella
term haṭhayoga.
After the fifteenth century, the composition of yoga texts which teach or
integrate Haṭhayoga flourished. Although these texts reveal significant efforts
at augmenting Haṭhayoga’s repertoire of techniques and synthesising it with
other yogas, religions and philosophies, the schema of the Haṭhapradīpikā was
most often the starting point and, in many cases, the prevailing paradigm for
Haṭhayoga in both the scholarly and more praxis-orientated works that will be
discussed below.
2 Post Fifteenth-Century Literature of Haṭhayoga
A notable change in the literature that followed the Haṭhapradīpikā was the
composition of more comprehensive works on Haṭhayoga and large compendi-
ums on yoga that integrated Haṭhayoga. The early literature on Haṭhayoga con-
sists mainly of short pithy texts that provide skeletal systems of practice and
rudimentary theoretical details. The later literature incorporates more tech-
niques and theory, as well as more elaborate systems of practice. The consider-
able growth in the length of the later works can be seen in Table 19.1.
It should be noted that, on the whole, the early works were composed in a
low register of Sanskrit and anuṣṭubh metre.15 They are prescriptive and ele-
mentary, which suggests that they were probably written for practitioners. In
contrast to this, many of the later works, such as the Haṭhatattvakaumudī and
the Yogacintāmaṇi, are more scholarly and tend to be written in higher registers
of Sanskrit. Their authors utilize more complex metres, compile their mate-
rial from a wider range of sources and often include commentary on the older
sources, which are frequently cited with attribution.
sandhāna (karṇau pidhāya hastābhyāṃ yaḥ śṛṇoti dhvaniṃ muniḥ). A similar technique is
described in the Śivasvarodaya (383), the Rājayogāmṛtasāra of Ānandānandanātha (1.17–
18, f. 9r–9v), the Maṇḍalabrāhmaṇopaniṣat (2.2.2) and the Jogapradīpyakā (685–688), and
it is depicted in an illustrated manuscript called “seven āsanas of yoga” at the National
Museum of New Delhi, PCWA gallery.
14 For example, one of the earliest yoga texts to teach a mudrā called śāmbhavī is the
Amanaska, a text on Rājayoga that explicitly rejects prāṇāyāma and the mudrās and
karaṇas associated with it (Birch 2014, 406–408).
15 The exception is the Yogatārāvalī, which is composed in triṣṭubh and incorporates poetic
images, etc.
via free access
table 19.1 A comparison of the number of verses in early and late texts on Haṭhayoga
Texts on Haṭhayoga No. of verses
Pre-16th c. texts Yogatārāvalī (14th c.) 29

Amaraughaprabodha (12th c.) 46a
Dattātreyayogaśāstra (13th c.) 169
Yogabīja (14th c.) 170
Haṭhapradīpikā (15th c.) 392
Post-16th c. texts Haṭharatnāvalī (17th c.) 404

Haṭhapradīpikā, 10 chs (18th c.) 595
Siddhāntamuktāvalī (18th c.) 1553b
Haṭhatattvakaumudī (18th c.) 1680c
Yogacintāmaṇi (17th c.) 3423d
a This number of verses is based on a short recension of the Amaraughaprabodha, which is

older than the recension published by Mallik (1954). The short recension is preserved by
two manuscripts (Ms. No. 1448 at the Government Oriental Manuscript Library, University
of Madras, Chennai, and Ms. No. 70528 at the Adyar Library and Research Centre, Chennai).
For more information on this short recension, see Birch 2019.
b This estimate is given by Gharote, et al. 2006, xvi.
c This is an approximate number of ślokas for the Haṭhatattvakaumudī, which has fifty-six
chapters, ranging from 5 to 150 ślokas each. Most are around 20. So, I have taken an aver-
age of 30. The Haṭhatattvakaumudī is undoubtedly a large yoga text, and I have felt no the
need to add up all of its ślokas to prove this point.
d This estimate is given in a scribal comment at the end of a manuscript of the Yogacintāmaṇi
held at the Kaivalyadhama Yoga Institute (ms. No. 9785 p. 257, line 14).
In order to discuss the salient characteristics of the late literature on Haṭha-

yoga, I shall divide the texts into two categories. The first consists of “extended
works” on the topic of Haṭhayoga and the second, “compendiums” that borrow
from Haṭha- and Rājayoga texts. There are some texts, such as the Haṭhasaṅke-
tacandrikā, which could be placed in either category. Nonetheless, the purpose
of introducing these etic categories is to reveal particular stylistic features and
content that are characteristic of the works in each group.
2.1 Extended Works

The extended works expatiate on the type of Haṭhayoga that was outlined in
the Haṭhapradīpikā. Many of these works borrow verses from the Haṭhapradī-
pikā and their discourse centres on the praxis and theory of Haṭhayoga. Exam-
ples include:
via free access
458 birch
Haṭharatnāvalī (17th c.)16

Haṭhayogasaṃhitā (17th c.)17
Haṭhābhyāsapaddhati (18th c.)18
Haṭhayogamañjarī (18th c.?)19
16 On the date of the Haṭharatnāvalī, see Birch 2018, 109 note 24.
17 The Haṭhayogasaṃhitā is a compilation that borrows extensively from the Haṭhapradī-
pikā. The opening verses (1.2–3) acknowledge the seven sages, namely Mārkaṇḍeya,
Bharadvāja, Marīci, Jaimini, Parāśara, Bhṛgu and Viśvāmitra, for spreading Haṭhayoga in
the world. The stated aim of Haṭhayoga is to achieve purification (śodhanaṃ), firmness
(dṛḍhatā), steadiness (sthairya), constancy (dhairya), lightness (lāghava), direct percep-
tion (pratyakṣa) and liberation (nirlipta) of the body (ghaṭa). Its Haṭhayoga has seven aux-
iliaries: the ṣaṭkarma, āsana, mudrā, pratyāhāra, prāṇasaṃyāma, dhyāna and samādhi.
The Haṭhayogasaṃhitā appears to have been the basis of the Gheraṇḍasaṃhitā (eigh-
teenth century), which adds a new frame story (viz., a dialogue between the teacher
Gheraṇḍa and a student Caṇḍakāpāli), several elaborate visualization practices and a
six-fold Rājayoga. The Gheraṇḍasaṃhitā calls its yoga ghaṭasthayoga, omits the Haṭhayo-
gasaṃhitā’s teachings on vajrolī and redefines this mudrā as a handstand, thus revealing
a reluctance to adopt the transgressive practices of Haṭhayoga.
18 The terminus ad quem of the Haṭhābhyāsapaddhati is the Śrītattvanidhi, which was a com-
pendium composed by Kṛṣṇarāja Waḍiyar III, the Mahārāja of Mysore, who was active in
the mid-nineteenth century (Sjoman 1996, 40). The Haṭhābhyāsapaddhati was a source
text of the Śrītattvanidhi (see Birch 2018, 131–134), and probably predates it by a hundred
years or so (Birch and Singleton 2019, 14–16).
19 Not much scholarly attention has been given to this work. Its name in the published edi-
tion is not entirely certain according to the colophons. The text begins with the heading
Haṭhayogamañjarī. However, the second chapter’s colophon refers to the Jogī-cintāmaṇi
of Śrīsahajānandanātha, the third chapter’s colophon to the Śrīsarvopaniṣat, and the final
colophon to the Gorakhajogamañjarī. The text could be a composite work consisting of
summaries or extracts of different texts. Nonetheless, the Haṭhayogamañjarī styles itself
as a work on Haṭhayoga. The terms haṭhayoga and haṭhavidyā are used in each chap-
ter and it contains a description of the haṭhayogī (p. 32). Its opening verses (2–5) state
that it is an explanation of the Haṭhapradīpikā in a vernacular language (bhāṣā). It cer-
tainly covers most of the content of the Haṭhapradīpikā, but also includes additional
material on yama, niyama, pratyāhāra, dhyāna, etc. A significant difference is that much
of the Haṭhapradīpikā’s discourse on Rājayoga has been omitted. Like other Brajbhāṣā
texts, such as the Jogapradīpyakā (1737 CE), the author of the Haṭhayogamañjarī equates
vajrolīmudrā with Rājayoga. I am yet to consult a manuscript of this work or even find
a reference to it in a manuscript catalogue or another yoga text. It may be the work
referred to as the Jogamañjarī (acc. no. 6543, Rajasthan Pracya Vidya Pratisthan, Bikaner,
Rajasthan) by Gharote et al. (2006a, lxvii). However, the librarian at this library in Bikaner
was unable to locate this manuscript when James Mallinson and I visited on separate occa-
sions in 2018. The Haṭhayogamañjarī mentions devotion to Rāma (e.g., p. 10, v. 33 and p. 18,
v. 25 and the last line) and was probably compiled in the same period and milieu as the
Jogapradīpyakā.
via free access
Haṭhapradīpikā with ten chapters (18th c.)20

Siddhāntamuktāvalī (18th c.),21 etc.
In particular, the last two texts are literally extended versions of the Haṭha-
pradīpikā because their authors simply added more verses to the original work
and created additional chapters on related topics. In fact, these “extended texts”
enlarge on the Haṭhapradīpikā in two ways. Firstly, they integrate other types
of yoga and various related topics. For example, the Haṭharatnāvalī combines
the fourfold system of yoga (i.e., Mantra-, Laya-, Haṭha- and Rājayoga) of earlier
works, such as the Amaraughaprabodha, with the aṣṭāṅga format. The author
of the Haṭharatnāvalī borrowed over one hundred and thirty verses from the
Haṭhapradīpikā and mentions Svātmārāma’s views on several specific matters
(Gharote et al. 2002, xx). The Siddhāntamuktāvalī significantly extends the
original Haṭhapradīpikā by adding sections on the purification of the channels
(nāḍīśuddhi), meditation (dhyāna), cheating death (kālavañcana) and indiffer-
ence (audāsīnya). A similar array of topics is seen in the Yogamārgaprakāśikā
(16–18th c.22), which adopts the fourfold system of yoga noted above. Its teach-
ings on Haṭhayoga follow for the most part the Haṭhapradīpikā.
Secondly, the repertoire of techniques in most of the texts which follow the
Haṭhapradīpikā became larger. As shown in Table 19.2, the number of āsanas
increases most significantly.23 However, it is also the case that techniques were
added to the standard collections of the ṣaṭkarmas and mudrās. The original six
therapeutic interventions known as the ṣaṭkarma form the basis of a repertoire
of twenty-one techniques in the Haṭhayogasaṃhitā. This work also adds fifteen
mudrās to the usual ten that are taught in Haṭhayoga. Furthermore, these texts
provide greater detail on many of the mudrās. For example, the Haṭhābhyās-
apaddhati contains the most elaborate teachings on vajroli, which is taught in
great detail along with its preliminary practices and medical applications. In
table 19.2 the number of kumbhakas remains almost the same but, generally
speaking, these texts contain many more verses on prāṇāyāma.24
20 On the date of the Haṭhapradīpikā with ten chapters, see Birch 2018a, 8 note 32.
21 On the date of the Siddhāntamuktāvalī, see Birch 2018, 127.
22 On the date of the Yogamārgaprakāśikā, see Birch 2018a, 8 note 29.
23 The Haṭharatnāvalī lists eighty-four āsanas but describes only thirty-six of them. Other
examples from this period of yoga texts with the names and descriptions of large num-
bers of āsanas include the Siddhāntamuktāvalī (96 āsanas), the Haṭhābhyāsapaddhati
(112 āsanas), the Yogāsana (108 āsanas), the Yogāsanamālā (110 āsanas) and the Ujjain
manuscript (No. 3537) of the Yogacintāmaṇi (54 āsanas described and two lists of over
eighty names of āsanas). For further information on this, see Birch 2018.
24 For example, the Haṭharatnāvalī has 97 verses in its chapter on prāṇāyāma whereas
via free access
460 birch
table 19.2 The proliferation of post-fifteenth Haṭhayogic techniques
Texts Number of techniques
āsana ṣaṭkarma kumbhaka mudrā
Haṭhapradīpikā (15th c.) 15 6 8 10

Haṭharatnāvālī (17th c.) 84 8 9 10
Haṭhayogasaṃhitā (17th c.) 32 21 8 25
Haṭhābhyāsapaddhati (18th c.) 112 9 8 10
The period in which these extended works arose was one in which physical
practices were documented on an unprecedented scale. Monographic works
were composed on particular techniques that had become, by this time, closely
associated with Haṭhayoga. Examples include the following:
Kumbhakapaddhati (17th c.)25

Satkarmasaṅgraha (18th c.)26
Yogāsanamālā (18th century)27
Yogāsana (19th century),28 etc.
The composition of such works indicates ongoing innovation and syncretisa-

tion in the practice of āsana, prāṇāyāma and the ṣaṭkarma that is also reflected
in the extended Haṭhayoga texts mentioned above. The Kumbhakapaddhati
describes over seventy varieties of breath retention (kumbhaka) and the Satkar-
masaṅgraha borrowed many of its additional therapeutic interventions from
the Haṭhapradīpikā has 78. The Haṭhatattvakaumudī has five chapters on prāṇāyāma (9,
10, 12, 37–38), namely, the preliminary auxiliaries and rules of practice for prāṇāyāma
(prāṇāyāmapūrvāṅgasādhanavidhi), an explanation of the names, nature and charac-
teristics of kumbhakas (kumbhakanāmasvarūpaguṇa), breathing methods for quelling
suffering (kleśaghnavāyusādhana), necessary rules for prāṇāyāma (prāṇāyāmakartavya-
vidhi) and an explanation of prāṇāyāma (prāṇāyāmavivecana), which total more than
240 verses. The Siddhāntamuktāvalī (ff. 53r–86v), the Yogacintāmaṇi (pp. 161–220) and
the Yuktabhavadeva (pp. 107–143) have large sections on prāṇāyāma as well.
25 On the date of the Kumbhakapaddhati, see Birch 2018a, 9 note 41.
26 On the date of the Satkarmasaṅgraha, see Birch 2018a, 50.
27 The one available manuscript of the Yogāsanamālā was completed on Wednesday, 20 Jan-
uary 1790 CE (miti mahīṣa sudī 5 budhavasare saṃvat 1846).
28 See below for information on the date of this text.
via free access
Āyurveda.29 All of these works build on the content of the Haṭhapradīpikā that
was relevant to their respective topics.
2.2 Extended Works: Scholarly vs Praxis-Orientated

Some of the extended works are the result of scholarly efforts to synthesise
and elaborate on material from earlier works, whereas others appear to doc-
ument, perhaps for the first time, a practice that was in use at the time of
writing. A good example of a more scholarly extended work is the Haṭharat-
nāvalī, which was composed by Śrīnivāsa. At the beginning of the first chap-
ter, he presents himself as a learned writer by informing the reader that he is
an eminent astrologer who excels in the Vedas, Vedānta, the works of Patañ-
jali, Vyākaraṇa, Tantra, Sāṅkhya, Nyāya, Vaiśeṣika, as well as various other
texts and philosophies.30 His remarks on the Haṭhapradīpikā indicate that
he was responding to deficiencies which he perceived in that work.31 In par-
ticular, his list of eighty-four āsanas appears to have been the creation of a
scholar who felt compelled to elaborate on references to this canonical number
of postures in early works, such as the Dattātreyayogaśāstra and the Viveka-
mārtaṇḍa.32 In fact, Śrīnivāsa’s list of eighty-four names of postures is pre-
ceded and followed by verses of the Vivekamārtaṇḍa and the Haṭhapradīpikā,
which state that Śiva taught eighty-four āsanas.33 The compilatory nature of
his collection is revealed by explicit references to and tacit borrowings from
the Haṭhapradīpikā, the Yogayājñavalkya, and the Dattātreyayogaśāstra in his
descriptions of thirty-six āsanas.34 Also, Śrīnivāsa incorporated into the list at
least twelve variations of certain basic postures, which further suggests that
this list was a scholarly contrivance arising from his intention to expatiate on
earlier textual references to eighty-four āsanas.35
29 For further details on this, see Birch 2018a, 49–56.

30 Haṭharatnāvalī 1.2 (vede vedāntaśāstre phaṇipatiracite śabdaśāstre svaśāstre tantre prā-
bhākarīye kaṇabhugabhihite nyāyaratnārṇavenduḥ | sāṅkhye sārasvatīye vividhamatimate
tattvacintāmaṇijñaḥ śrīmajjyotirvid agre saravaratanujo rājate śrīnivāsaḥ ||).
31 For example, see Śrīnivāsa’s commentary to Haṭharatnāvalī 1.27 and 2.86.
32 For references and translations of the relevant verses, see Birch 2018, 107–108.
33 Haṭharatnāvalī 3.7 (~ Vivekamārtaṇḍa 8–10) and 3.23 (= Haṭhapradīpikā 1.35).
34 In the Haṭharatnāvalī’s section on āsana, the Haṭhapradīpikā, the Yogayājñavalkya, and
the Dattātreyayogaśāstra are cited by name at 3.23, 3.35 and 3.36, respectively. Haṭharat-
nāvalī 3.42–43, 51–54, 57–58, etc., are tacitly borrowed from the Haṭhapradīpikā. Also,
Haṭharatñāvalī 3.77 is a rewriting of Vivekamārtaṇḍa 92.
35 Six types of mayūrāsana are mentioned in Haṭharatnāvalī 3.10 and five are described at
3.42–47; three of matsyendrāsana are mentioned in 3.12 and described at 3.58–60; two of
paścimatānāsana are mentioned in 3.13, and both are described at 3.66–68; and possibly
five of kukkuṭāsana are mentioned in 3.17 (if one assumes that ākārita and bandhacūlī,
via free access
462 birch
The scholarly efforts of Śrīnivāsa in compiling a list of āsanas can be con-

trasted with what one might call “praxis-orientated works” in this category,
such as the Haṭhābhyāsapaddhati and the chapter on āsana in the Siddhānta-
muktāvalī. These works are composed in the same style as the early Haṭhayoga
texts; they are prescriptive and focus on praxis rather than theory. Their col-
lections of āsanas do not appear to be a synthesis of earlier textual sources
because their authors do not cite or allude to any sources, and a large propor-
tion of their āsanas is not attested in earlier texts. In the case of the Haṭhā-
bhyāsapaddhati, innovations seem apparent in the use of moving āsanas,
sequencing, linking postures, counterposing and the use of props, such as
ropes and walls (Birch 2018, 134–135). Therefore, it is likely that innovation
played a significant role in the proliferation of techniques seen in these praxis-
orientated extended works which postdate the Haṭhapradīpikā. Although
some of these complex postures may predate the fifteenth century, as revealed
by iconography, the codification of large numbers of complex postures in texts
specifically on Haṭhayoga, as well as works on yoga broadly conceived, such
as the Jogapradīpyakā, appears to have emerged from the seventeenth century
onwards. In fact, only a small portion of the aggregate number of these āsanas
are anticipated by the earlier sculptural collections of the Mehudi gate, Brah-
manath temple and Hampi,36 which further suggests that a certain degree of
innovation was probably at play during Haṭhayoga’s floruit.
There is little evidence to indicate that scholarly extended works, such as
the Haṭharatnāvalī, and the compendiums mentioned below were sources of
information for Kṛṣṇamācārya, Swāmī Śivānanda and other gurus who popu-
larized physical yoga in the early twentieth century. These gurus deferred to the
Pātañjalayogaśāstra and the Haṭhapradīpikā in their publications. Nonethe-
less, there is significant evidence that they surreptitiously adopted postures
from traditions of āsana practice that were prevalent in certain regions of India
in the eighteenth and nineteenth century. The traces of these traditions are dis-
cernible through texts such as the Haṭhābhyāsapaddhati, the Jogapradīpyakā
and the Yogāsana,37 which preserve teachings on large distinct collections
which are included among kukkuṭa, ekapādakakukkuṭa and pārśvakukkuṭa, are variations
of the same posture).
36 On large collections of āsanas at these locations, see Vijaya Sarde 2015 and 2017, and Seth
Powell 2018.
37 The āsanas of the Yogāsana, most of which correspond to untraced Sanskrit descrip-
tions quoted in a commentary on the Yogasūtra by Śrīkṛṣṇavallabhācārya (1939), were
reproduced in several early twentieth-century publications, one of which was Swāmī
Śivānanda’s book Yoga Asanas. This will be discussed at length in a forthcoming publi-
cation.
via free access
of complex āsanas (Birch, forthcoming). The growing emphasis on complex

āsanas and mudrās in practical works such as the Siddhāntamuktāvalī also
occurs for the first time in the literature of erudite Brahmins of the same era.
I shall now turn my attention to these more scholarly works which integrated
Haṭha- and Rājayoga into more orthodox conceptions of yoga.
2.3 Yoga Compendiums

A century or so after the Haṭhapradīpikā, a number of lengthy compendi-
ums on yoga were composed. Most of these compendiums do not focus on
any particular type of yoga, but treat the subject more generally by combin-
ing many sources, including texts on Haṭhayoga. Generally speaking, these
compendiums are scholarly works of literature, which incorporate philoso-
phy and metaphysics on a more ambitious scale than earlier texts, such as the
Haṭhapradīpikā. In a sense, their authors produced a new discourse on yoga
by combining the teachings of Haṭha- and Rājayoga with those of the Pātañ-
jalayogaśāstra and various Brahmanical texts, including the Bhagavadgītā,
the Mahābhārata, early Upaniṣads, Purāṇas and Dharmaśāstras. In the early-
twentieth century, a synthesis of the same genres was favoured by Brahmin
gurus such as Kṛṣṇamācārya, Swāmī Śivānanda and Swāmī Kuvalayānanda,
who combined physical yoga techniques with concepts from Pātañjalayoga and
Advaitavedānta, as well as metaphysics from tantric traditions, to teach yoga to
an international audience.38
The compendiums to which I am primarily referring are:39
Yogacintāmaṇi of Godāvaramiśra (16th c.)

Yogapañcāśikā (early 16th c.)40
38 One should consult the work of Elizabeth de Michelis (2004) and Mark Singleton (2010)
for the modern elements, such as neo-Vedānta and physical culture, that also shaped the
teachings of these gurus. Some significant similarities in their teachings with the sources
I am discussing include the use of the aṣṭāṅga format, the integration of bandhas and
mudrās with āsana and prāṇāyāma, the importance of inverted āsanas, the identification
of samādhi with Advaitavedānta concepts and, above all, the trans-sectarian approach to
compiling their teachings.
39 For dating, please refer to Birch 2018a, unless otherwise indicated.
40 The Yogapañcāśikā might be the earliest attempt to integrate Haṭha- and Rājayoga with
Pātañjalayoga. The text is cited by name in a Sanskrit work called the Vivekamukura, which
was composed by Nṛsimha Bhāratīya, according to its last verse (97). If this is the same
author who wrote the commentary called the Subodhinī on the Vedāntasāra, as stated by
Thangaswami (1980, 360–361), then Nṛsimha Bhāratī of Varanasi was active in the late
sixteenth century. Unlike other compilations in this list, the Yogapañcāśikā is a short
via free access
464 birch
Upāsanāsārasaṅgraha (16th c.)41

Yogacintāmaṇi of Śivānandasarasvatī (17th c.)
Yuktabhavadeva (17th c.)
Tattvabinduyoga of Rāmacandra (17–18th c.)42
Haṭhasaṅketacandrikā (18th c.)
Haṭhatattvakaumudī (18th c.)
Yogasārasaṅgraha (18th c.)43
Yoga Upaniṣads (18th c.)44
Rājatarala (late 18th–19th c.)45
work of merely fifty verses that cites only the Pātañjalayogaśāstra. It teaches an aṣṭāṅ-
gayoga. The first four auxiliaries are Haṭhayoga, and the second four, Rājayoga. It is a Śaiva
work that aims at raising kuṇḍalinī, uniting Śakti with Śiva and attaining jīvanmukti, fol-
lowed by videhamukti when all prārabdakarma is extinguished. Its yoga is not intended
for those who deny the validity of scripture (nāstika), but for male life-long brahmacārins
(nṛnaiṣṭhika). This text is unpublished and, as far as I am aware, preserved by only one
manuscript, which has several small lacunae.
41 Bouy 1994, 89–92.
42 Birch 2014, 415, 434 note 71.
43 The Yogasārasaṅgraha undoubtedly postdates the Haṭhapradīpikā and the Śivayogapra-
dīpikā. The latter was probably composed in the late fifteenth century. However, the
Yogasārasaṅgraha may post-date the Haṭharatnāvalī (seventeenth century), as it shares
a verse on bhujaṅgīkaraṇa, a technique that is only taught in the Haṭharatnāvalī (2.31), as
far as I am aware (cf. Yogasārasaṅgraha p. 28, lines 4–5). Also, there are other verses on
Haṭhayoga that seem to follow the Haṭharatnāvalī rather than the Haṭhapradīpikā (e.g.,
Yogasārasaṅgraha p. 55, lines 8–14 ≈ Haṭharatnāvalī 2.32–35). The Yogasārasaṅgraha’s ter-
minus ad quem would probably be one of its manuscripts. Several appear to be reported
in the New Catalogus Catalogorum (hereafter NCC), but I have not had the opportunity to
consult any of them.
44 These so-called Yoga Upaniṣads are part of a recent recension compiled in South India
in the first half of the eighteenth century and commented on by Upaniṣadbrahmayo-
gin (See Bouy 1994). They include the Yogatattvopaniṣat, the Dhyānabindūpaniṣat, the
Nādabindūpaniṣat, the Śāṇḍilyopaniṣat, the Yogacūḍāmaṇyupaniṣat, the Yogakuṇḍalinyu-
paniṣat, the Yogaśikhopaniṣat, the Darśanopaniṣat, the Maṇḍalabrāhmanopaniṣat, the
Saubhāgyalakṣmyupaniṣat and the Varāhopaniṣat.
45 The Rājatarala is a lengthy commentary on the Yogatārāvalī (circa 14th c.) that was com-
posed by Rāmasvāmipaṇḍita, who is described as a worshipper of Śaṅkarācārya’s feet
(śrīśaṃkarācāryapādakiṃkara). He cites the Maṇḍalabrāhmaṇopaniṣat (ms. no. 72330,
f. 29v), which means that the Rājatarala was composed after the corpus of one hundred
and eight Upaniṣads, that is, the mid-eighteenth century (Bouy 1994, 6, 34, etc.). Also, a
verse pays homage to a Dakṣiṇāmūrti in the city of Śrīśaila, near Kadalī, which appears to
locate the work in Andhra Pradesh (Mahadevan 2018, 68). It is preserved by one undated
palm-leaf manuscript (No. 72330) in Telugu script at the Adyar Library and Research Cen-
tre, Chennai, and an undated transcript (No. B378) in Devanagari (circa 20th century) at
the Oriental Research Institute, Mysore.
via free access
Yogasandhyā (early 19th c.)46

Gorakṣasiddhāntasaṅgraha (19th c.)47
The authors of these compendiums often combined yoga teachings from differ-
ent traditions seamlessly. For example, in the Yogacintāmaṇi, Godāvaramiśra
integrated the physical methods of Haṭhayoga with the auxiliaries of āsana and
prāṇāyāma in Patañjali’s aṣṭāṅga system (see below). Similarly, in Śivānanda’s
Yogacintāmaṇi and Bhavadeva’s Yuktabhavadeva, the meditative state of
Rājayoga became the equivalent of Patañjali’s highest stage of samādhi, called
asaṃprajñātasamādhi.48 Most authors of these works were inclined towards
Vedānta. They cited the teachings of the Upaniṣads to express the gnostic
insights that arise from samādhi. Also, they incorporated theistic teachings on
yoga from the Purāṇas and Tantras, and were comfortable with defining yoga
as meditation (i.e., cittavṛttinirodha), on the one hand, and then as the union of
the self with a deity, on the other hand. Likewise, their descriptions of dhāraṇā
and dhyāna juxtapose Patañjali’s definition of “binding the mind to one place,
etc.,” with tantric visualizations of the five elements and deities.
2.4 Godāvaramiśra’s and Śivānanda’s Thought Gems on Yoga

In the Yogacintāmaṇi, Godāvaramiśra’s method of synthesising Haṭhayoga with
other yogas is typical of this genre.49 Godāvaramiśra was a chief minister
(mantrivara) and preceptor (rājaguru) to the Orissan king Pratāparudradeva,
46 According to the final colophon, the Yogasandhyā was composed by Śrīsadāśivanārāyaṇa-

brahmacāri, whose guru was Śrīmajjagannāthacaitanyabrahmacāri. They belonged to the
tradition of the Śṛṅgerīmaṭha. The final verse mentions the date of composition in the
bhūtasaṅkhyā system. If the term dharma represents the number eight, it would have been
completed in VS 1861 (= 1804CE), when the sun was in the ninth nakṣatra (tapas) and
the moon full. See Yogasandhyā, p. 203 (rākeśarasadharmorvvīsammite vaikrame ’bdake
| tapasīne ca rākāyāṃ satkṛtiḥ pūrṇatāmitā). I would like to thank for their comments
on this verse Somadeva Vasudeva, Péter Szántó, James Mallinson and, in particular, Chris
Minkowski, who suggested that dharma could mean 4, 6 or 8.
47 The date of the Gorakṣasiddhāntasaṅgraha is not certain, although it post-dates the Sid-
dhasiddhāntapaddhati, which might be as late as the eighteenth century (Mallinson 2014,
170–171).
48 Śivānanda’s Yogacintāmaṇi, p. 9 (na tatra kiñ cid vedyaṃ saṃprajñāyata ity asaṃprajñātaḥ
samādhiḥ | ayaṃ nirbīja iti nirvikalpa iti nirālamba iti rājayoga—iti cocyate) and Yuktab-
havadeva 1.33 ([…] sa eva nirvikalpaḥ samādhiḥ sadā savikalpakajñānābhāvāt || sa eva ca
rājayogaḥ).
49 This work survives in a single, incomplete manuscript, which is dated Wednesday, 16th
November 1715CE. Yogacintāmaṇi (f. 132v, lines 7–8): saṃvat 1772 varṣe kārttika vadi mā-
vāsyā budhavāsare likhitaṃ.
via free access
466 birch
who ruled in the early sixteenth century. In addition to yoga, Godāvaramiśra

wrote works on various topics, including Advaitavedānta (the Advaitadar-
paṇa50), Tantra (the Tantracintāmaṇi) and an extensive treatise on politics and
warfare (the Hariharacaturaṅga).51 His view of yoga was mainly shaped by the
Pātañjalayogaśāstra. He acknowledges this at the beginning of his work:
In this text, I summarise and examine Patañjali’s doctrine, which was

explained by Vyāsa, Vācaspati and Bhojadeva and which is validated and
[yet] overlooked elsewhere.52
Although Godāvaramiśra discusses only some of the sūtras of the Pātañjalayo-

gaśāstra in an order that fits the design of his work, he begins with the first one
(atha yogānuśāsanam) and, in his discussion of it, cites with attribution pas-
sages from the Gāruḍapurāṇa (sic), the Bhagavadgītā, the Brahmāṇḍapurāṇa
and the Kūrmapurāṇa, as well as the sages Yājñavalkya (i.e., the Yājñavalkyasm-
ṛti) and Maharṣimataṅga.53 The first half of the Yogacintāmaṇi is concerned
with general topics, such as definitions of yoga, the types of samādhi and so
on. The latter half is structured on the eight auxiliaries (aṣṭāṅga) of Pātañ-
jalayoga, and it is in his discussions of āsana, diet and prāṇāyāma that he cites
50 The Yogacintāmaṇi (f. 131v lines 7–8) of Godāvaramiśra states that he wrote the dvaitadar-
paṇa: “Now, they have been explained together by me in the Dvaitadarpaṇa” ([…] asmāb-
hir atha dvaitadarpaṇe yugapadanuvarṇitāḥ […]). However, it seems that atha dvaitadar-
paṇe is a scribal error for athādvaitadarpaṇe, because yugapadanuvarṇitāḥ refers back
to two works: the Saṅkṣepaśārīrakavārttika, which was composed by his paternal grand-
father (pitāmahacaraṇa) and the Advaitacintāmaṇi by his father (pitṛcaraṇa). The NCC
(vol. 6, 126) reports that Godāvaramiśra wrote the Advaitadarpaṇa because it is quoted in
his Hariharacaturaṅga (p. 178, v. 22; p. 196, v. 502). This is affirmed in a summary of the
Hariharacaturaṅga (Meulenbeld 2000, 562–563).
51 For more information on Godāvaramiśra’s family and works, see the NCC vol. 6, 1971, 126
and Meulenbeld 2000, 562–563.
52 Yogacintāmaṇi, f. 1v, lines 4–5 ( yad vyāsavācaspatibhojadevaiḥ pātañjalīyaṃ niraṇāyi
tattvaṃ | anyatra siddhaṃ yad upekṣitaṃ ca tad atra saṅkṣipya nirūpayāmi ||3||) The codex
reads upekṣitaṃ, but Gode (1953, 474) transcribes it as apekṣitaṃ. Whether he was tacitly
emending upekṣitaṃ to apekṣitaṃ is not clear. He may have emended because the mean-
ing of apekṣita is more consistent with siddha, but it seems possible that upekṣita was
intended to contrast with siddha, as I have translated.
53 One would expect that the Mataṅgapārameśvaratantra is meant by this attribution. How-
ever, I have not found the cited verse in the published edition of this tantra. The verse in
the Yogacintāmaṇi (f. 3v, lines 5–7) is agniṣṭomādikān yajñān vihāya dvijasattamaḥ yogāb-
hyāsarataḥ śāntaḥ paraṃ brahmādhigacchati || brāhmaṇakṣatriyaviśāṃ strīśūdrāṇāṃ ca
pāvanaṃ śāntaye karmaṇām anyad yogān nāsti vimuktaye ||. This verse is found in the
Viṣṇudharma (98.016).
via free access
two works on Haṭhayoga: the Haṭhapradīpikā (haṭhayoge—f. 39v, l. 8) and the

Dattātreyayogaśāstra (dattātreyaḥ—f. 36r, line 1; f. 40v, line 3). He also cites
the Yogayājñavalkya ( yājñavalkyaḥ—f. 36r, line 6; f. 36v, line 6; f. 37v, line 7;
f. 38v, line 7), which was a source text for the Haṭhapradīpikā. The other sources
on these topics are the Dharmaputrikā, the Pavanayogasaṅgraha, the Āgneya-
purāṇa and the Mataṅgapārameśvaratantra. Therefore, on the topic of āsana,
Godāvaramiśra created a seamless synthesis of haṭhayogic teachings with those
of tantric and brahmanical sources.
Godāvaramiśra’s work became the basis of an ambitious attempt by Śivānan-
dasarasvatī to integrate a more extensive and diverse array of Haṭhayoga texts
with Pātañjalayoga and Brahmanical works. Śivānanda also named his compi-
lation the Yogacintāmaṇi. He probably lived in the early-seventeenth century in
Varanasi.54 At the end of this text, Śivānanda informs the reader of the material
he has included and excluded. He says:
Meditation along with the practices [ancillary to it] have been explained
briefly by me according to scripture and my understanding. Listening to
and contemplating [the teachings] which are seen in detail and at length
only in the Upaniṣads, have not been discussed for fear of prolixity. I
have revealed here all that which is secret in Haṭha- and Rājayoga for
the delight of yogins. However, that Haṭhayoga which was practised by
Uddālaka, Bhuśuṇḍa and others has not been mentioned by me, because
it cannot be accomplished by contemporary [practitioners. Also], the pro-
cedures and so forth promoted by the kāpālikas have not been mentioned
[because] they contravene the Vedas, Dharmaśāstras and Purāṇas.55
54 The terminus a quo of Śivānanda’s Yogacintāmaṇi is Godāvaramiśra’s Yogacintāmaṇi (see

above). The terminus ad quem is 1630 CE, based on two dated manuscripts of the Yogacintā-
maṇi. The first is reported in the catalogue of yoga manuscripts by Kaivalyadhama (2005,
226–227), which gives the author (Śivānandasarasvatī), the library (the Asiatic Society
in Mumbai), the manuscript number (1083) and the date (VS 1687 = 1630CE). The sec-
ond manuscript is held at the Panjab University Library and has been catalogued by the
Woolner Project (Ms. no. 6922). After this manuscript’s final colophon, a scribal comment
indicates that it was copied by Rāma on the 13th of the month, Kārttika, in Śāka 1552 (i.e.,
Sunday, 17 November 1630CE). The possibility that Śivānanda was a resident of Varanasi
is supported by a reference to his devotion to Viśveśvara, a standard claim of Śaivas who
resided there. I would like to thank Alexis Sanderson for pointing this out to me, and he
also noted that similar references to Viśveśvara in works of Śaivas who resided in Varanasi
can be found in Jñānaśiva’s Jñānaratnāvalī and Viśvanātha’s Siddhāntaśekhara, which are
both Saiddhāntika Paddhatis (personal communication, 24 April 2013).
55 The Yogacintāmaṇi pp. 281–282 (with verse numbers added for the clarity of the critical
apparatus): nididhyāsanam etat tu mayā sādhanasaṃyutam | yathāśāstraṃ yathābodhaṃ
via free access
468 birch
Both Godāvaramiśra and Śivānanda excluded the Haṭha techniques of vaj-

rolī, amarolī and sahajolī, presumably because these are the kāpālika practices
which would be unacceptable to his brahmanical audience.56 The main differ-
ence between Godāvaramiśra’s and Śivananda’s compilations is that the former
focused on Pātañjalayoga and cited Haṭhayoga texts sparingly, whereas the lat-
ter discussed Haṭhayoga as forthrightly and comprehensively as the yogas of
other traditions, and cited its texts profusely. This is also the case for similar
compilations of the time, such as Bhavadevamiśra’s Yuktabhavadeva and Sun-
daradeva’s Haṭhasaṅketacandrikā.
2.5 The Polymathy of Bhavadevamiśra

The synthesis of various philosophies and genres of literature in the yoga com-
pendiums under consideration was created by well-educated Brahmins who
were knowledgeable in a wide range of scholarly subjects. A good example of
this is the seventeenth-century Bhavadevamiśra, whose Yuktabhavadeva is a
digest (nibandha) that integrated teachings of Haṭha- and Rājayoga with those
of the Pātañjalayogaśāstra and various Upaniṣads, Purāṇas, Tantras, Dhar-
maśāstras and the Epics. Apart from the fact that Bhavadeva cited a wide
range of Sanskrit works, the breadth of his learning is attested by the com-
mentaries attributed to him on various śāstras. Manuscript colophons state
that he was a Brahmin from Mithila and that his father was Kṛṣṇadevamiśra,
his elder brother Baladevamiśra and his teacher Bhavadeva Ṭhakkura. Based
on this information, the Bhavadeva who wrote the Yuktabhavadeva also wrote
commentaries on the Pātañjalayogaśāstra,57 the Brahmasūtra,58 the Kāvyapra-
saṅkṣepeṇa nirūpitam ||1|| śravaṇaṃ mananaṃ caiva saviśeṣaṃ savistaram | vedānteṣv eva
draṣṭavyaṃ noktaṃ vistarabhītitaḥ ||2||| rahasyaṃ rājayogasya haṭhayogasya yat sthitam |
prakāśitaṃ mayā sarvaṃ prītaye yoginām iha ||3|| arvācīnair asādhyatvāl likhito na mayā
hi saḥ | uddālakabhuśuṇḍādyair haṭhayogas tu yaḥ kṛtaḥ ||4|| kāpālikapraṇītās tu itikar-
tavyatādayaḥ | likhitā na mayā te tu śrutismṛtivirodhinaḥ ||5||
2b saviśeṣaṃ savistaram ] 6922, Ed.; savistaraviśeṣakam 3537. 4a arvācīnair ] 3537, 6922;
sadhrīcīnair Ed. 4b asādhyatvāl ] 3537; asādhyatvāt Ed.; asādhyatvān 6922. 4b likhito na
mayā hi saḥ ] Ed.; likhitvo na mayā hi saḥ 3537; na likhito hi mayā tu saḥ 6922 (hypermetri-
cal). 5d virodhinaḥ ] 3537, 6922; virodhanāḥ Ed.
56 The Haṭhapradīpikā (3.96d) contains a verse that points to the Kāpālika origins of amarolī
(… kāpālike khaṇḍamate ’marolī). Amarolī and sahajolī are generally considered to be vari-
ations of vajrolī (e.g., Dattātreyogayogaśāstra 31 and 158, Haṭhapradīpikā 3.92, etc.).
57 This commentary is called the Pātañjalīyābhinavabhāṣya and several manuscripts of it are
held in the Kathmandu National Archives. The chapter colophons of E 1819-9 and A 554-
522 affirm that Bhavadeva was the son of Kṛṣṇadeva and the favourite student (priyaśiṣya)
of Ṭhakkuraśrībhavadeva.
58 This commentary is called the Candrikā, and the NCC (vol. 15, 12) reports that it is by
via free access
kāśa59 and the Vājasaneyīsaṃhitā60 as well as a work on Dharmaśāstra called

the Dānadharmaprakriyā61 and another on what appears to be Vaiśeṣika phi-
losophy, the Vaiśeṣikaratnamālā.62 Some manuscript catalogues also attribute
to a “Bhavadeva” a commentary called the Abhinavabhāṣya on the Śāṇḍilya-
sūtra, also known as the Bhaktimīmāṃsāsūtra,63 and some other works,64 but
the catalogues I have consulted do not provide enough biographic informa-
tion to prove that this was the same Bhavadeva who composed the Yukta-
bhavadeva. Nonetheless, it is clear that Bhavadeva was a scholar whose knowl-
edge extended far beyond the theory and practice of yoga traditions.
Bhavadeva, who was the son of Sanmiśrasrīkṛṣṇadeva and the disciple of Ṭhakkura-
bhavadeva.
59 This commentary is called the Līlā. The NCC (vol. 4, 98) reports that it is by Bhavadeva,
son of Kṛṣṇadeva of Mithila and pupil of Bhavadeva Ṭhakkura.
60 This commentary is called the Vyākhyānaratnamālā. The NCC (vol. 28, 60) reports that it
is by Bhavadeva of Mithila, son of Kṛṣṇadeva and disciple of Bhavadeva Ṭhakkura.
61 Dānadharmaprakriyā was composed by Bhavadevabhaṭṭa, son of Kṛṣṇadevamiśra (NCC
vol. 9, 6) at the request of Rudradāsaśreṣṭhin in 1636–1637CE (NCC, vol. 16, 172). Kane (1930
vol. 1, 560) points out that this work was by Bhavadeva, son of Kṛṣṇadeva of Mithila. In a
latter volume (1962, vol. 5, part ii, 28), he attributes the Prāyaścittaprakaraṇa to Bhavadev-
abhaṭṭa.
62 The NCC (vol. 32, 64) reports that the Vaiśeṣikaratnamālā was written by Bhavadeva Paṇ-
ḍita, son of Kṛṣṇadevamiśra and disciple of Bhavadeva Ṭhakkura. This appears to be based
on a sole manuscript at the Bhandarkar Oriental Research Institute (123 of 1881–1882).
63 In its entry on the commentary called the Abhinavabhāṣya on the Śāṇḍilyasūtra or Bhak-
timīmāṃsāsūtra (also called the Śatasūtī or Bhaktisūtra), the NCC (vol. 15, 152) reports that
this work is found in the manuscript libraries of the Oriental Institute in Baroda, the Pra-
jñā Pāṭhaśālā Maṇḍala at Wai and the Sampūrṇānanda library in Varanasi. The catalogues
of the first two do not give any biographic information for Bhavadeva and I have not been
able to consult catalogues of the Sampūrṇānanda library (SB New DC XII 44408, 44416. ii.
107900. 107911).
64 Other works attributed to a Bhavadeva in the NCC (vol. 16, 172) include the Yogasaṅ-
graha, the Vyāptivāda and commentaries on the Yogadarpaṇa, the Yogabindu, the Raghu-
vaṃśa (called the Subodhinī) and the Ṣaḍaṅgarudra. Karl Potter’s Encyclopedia of Indian
Philosophies Bibliography (1983, vol. 1, 475) adds commentaries by a Bhavadeva (dated
to 1650) on Bhavānanda’s Kārakacakra, Īśvarakṛṣṇa’s Sāṅkhyakārikā and the Pañcalak-
ṣaṇa section of Gaṅgeśa’s Tattvacintāmaṇi, as well as a work on Nyāya called the Anu-
mānaprakaraṇavyākhyā.
via free access
470 birch
3 The Regional Extent of Haṭhayoga’s Literature on the Eve of

Colonialism
The compendiums of Godāvaramiśra, Śivānanda, Bhavadevamiśra and Sunda-

radeva endorsed the teachings of Haṭhayoga.65 These works were part of a con-
certed effort among some erudite Brahmins to make Haṭhayoga’s physical prac-
tices an integral part of the Brahmanical view of yoga, much like the so-called
“Yoga Upaniṣads” that were compiled in South India in the first half of the eigh-
teenth century. Christian Bouy’s work (1994) on these Upaniṣads informs us
that the compiler liberally borrowed material from earlier Haṭha- and Rājayoga
texts, often presenting the physical practice as a way of purifying the mind that
would then lead to the realization of vedantic truths. As Bouy (1994, 72) notes,
the prominence of Haṭhayoga teachings in these Upaniṣads indicates that this
type of yoga had come into vogue in vedāntic milieus. He says that the inter-
est of vedāntists in Haṭhayoga may have started in the fifteenth and sixteenth
centuries, and then flourished in the seventeenth and eighteenth centuries.
In addition to the Yoga Upaniṣads and the Upāsanāsārasaṅgraha, which
were examined by Bouy, the Yogasandhyā and the Rājatarala are large exeget-
ical works composed in South India that weave together Pātañjalayoga and
Haṭhayoga within a vedāntic framework. The other compendiums discussed in
this article support Bouy’s observations and extend his hypothesis to the region
of Northeast India. Godāvaramiśra lived in what is now Orissa, Bhavadeva
in Mithila and both Śivānanda and Sundaradeva in Varanasi. The vedāntic
overtones in the works of these authors include references to the Upaniṣads
and an emphasis on achieving the liberating gnosis of Brahman. Also, various
prosopographic details associate them with Advaitavedānta. Godāvaramiśra’s
father Balabhadra wrote a text called the Advaitacintāmaṇi and he himself
the Advaitadarpaṇa,66 Śivānanda mentions Ādiśaṅkara in his lineage67 and,
according to manuscript catalogues, Bhavadeva wrote a commentary on the
Brahmasūtra. Unlike the Yoga Upaniṣads, their works explicitly cite Haṭhayoga
texts.
65 For information on Sundaradeva and his works, see Birch 2018a, 58–61.
66 Yogacintāmaṇi (ms. no. 220 of 1882–1883) f. 131v.
67 Yogacintāmaṇi p. 2: “Having bowed to Śrīvyāsa, the ascetic Śaṅkara, the teacher of the
world, [my] teacher Śrīrāmacandra, whose lotus feet are intense bliss, and all of the gods
of yogins, the ascetic Śivānanda has written clearly the great Yogacintāmaṇi, which had
fallen into an ocean of various texts and has the power to explain everything” (śrīvyāsaṃ
yatiśaṅkaraṃ bhavaguruṃ śrīrāmacandraṃ guruṃ sāndrānandapadāmbujañ ca nikhi-
lān natvā hi yogīśvarān | nānāgranthapayodhimadhyapatitaṃ śrīyogacintāmaṇiṃ niḥ-
śeṣārthasamarthakaṃ yatiśivānandaḥ karoti sphuṭam ||).
via free access
In contrast to this, I am yet to locate any such compendium in Northwest

India. However, some of the extended works on Haṭhayoga can be traced to this
region, such as the Siddhāntamuktāvalī, the Haṭhapradīpikā with ten chapters
and the Haṭhayogamañjarī. These works, along with the Haṭhābhyāsapaddhati
in Maharashtra and Karnataka, are less scholarly and more likely influenced by
practitioners of the time in which they were composed.
4 Conclusion
The flourishing of literature on Haṭhayoga in both North and South India from
the sixteenth to eighteenth centuries was concomitant with a growing perva-
siveness of references, over the same period, to Haṭhayoga in Sanskrit literature
of various religions and philosophies.68 The findings of this paper suggest that
Haṭhayoga became more prevalent in literature composed during this period
and that the Haṭhapradīpikā was instrumental in defining the techniques and
structure of practice for this type of yoga. Moreover, distinct physical tech-
niques that became closely associated with Haṭhayoga, such as non-seated
āsanas and mudrās, had become integral to broader conceptions of yoga on
the eve of colonialism.
In the period following the Haṭhapradīpikā, it is possible to discern that
the praxis and theory of Haṭhayoga developed in different ways as it became
more widely disseminated. In scholarly circles of Northeast India, its codifi-
cation took on some of the characteristics of the philosophical yogas, as it
68 Beyond the texts I have mentioned above, the following literature of the sixteenth to eigh-
teenth centuries also integrated haṭhayogic teachings. I mention here only a few exam-
ples without detailed references, which will appear in a forthcoming publication. Exam-
ples include the Puraścaraṇacandrikā (late-fifteenth century) and the Puraścaraṇārṇava
(eighteenth century), two Śaiva ritual compilations that incorporated verses on āsanas,
some of which occur only in the Haṭhapradīpikā; the Merutantra, a relatively recent Śaiva
work, which mentions Haṭhayoga in relation to prāṇāyāma; Rāmatoṣaṇa Bhaṭṭācārya’s
Prāṇatoṣiṇī (1820 CE), which has numerous references to Haṭhayoga; Narāyānatīrtha’s
Yogasiddhāntacandrikā, a commentary on the Pātañjala Yogasūtra, which integrates fif-
teen yogas with Patañjali’s aṣṭāṅga format and defines Haṭhayoga as the auxiliaries of
āsana and prāṇāyama; Vijñānabhikṣu’s Sāṅkhyasāra, a philosophical treatise that men-
tions both Haṭha- and Rājayoga; Narahari’s Bodhasāra, a philosophical compendium that
has sections on Mantra-, Laya-, Haṭha- and Rājayoga; the Bhāvanāpuruṣottama, a San-
skrit drama in which a Kāpālika mentions haṭhavidyā as a ladder ascending to Rājayoga
(in terms similar to those of the Haṭhapradīpikā); and the Vāsiṣṭhamahārāmāyaṇatāt-
paryaprakāśa, a commentary on the Yogavāsiṣṭha, which mentions both Haṭha- and
Rājayoga, etc.
via free access
472 birch
was integrated with the Pātañjalayogaśāstra and Brahmanical sources. This

development resulted in a more syncretic and sophisticated discourse around
the physical techniques. The social background of the authors of these com-
pendiums was diverse. Some were sannyāsins, such as Śivānandasarasvatī and
Śrīsadāśivanārāyaṇabrahmacāri, whereas others identified themselves accord-
ing to their professions, which included medicine, politics and astronomy (i.e.,
Sundaradeva, Godāvaramiśra and Śrīnivāsa, respectively).
In contrast to the scholarly compendiums of the northeast, the extended
works on Haṭhayoga that arose in the northwest of India retained the more
praxis-orientated focus of the Haṭhapradīpikā, which they enlarged upon by
adding more techniques and other auxiliaries, such as yama and niyama.
Although not much is known of the authors of these works, renunciant tradi-
tions in this region seemed reasonably active in producing manuals on yoga,
such as the Haṭhayogamañjarī and the Jogapradīpyakā, and monographic
works, such as the Yogāsanamālā and the Yogāsana, whose content is related
to the Haṭhapradīpikā and its extended texts.
From the Amaraughaprabodha to the current day, Haṭhayoga has been dis-
tinguished by physical methods of practice. In the early period the practice cen-
tred on breath retentions with physical locks (mudrā), and it burgeoned over
the centuries to include an array of complex postures, ṣaṭkarma and mudrās.
Although these physical techniques were never particular to Haṭhayoga, the
textual evidence suggests that after the sixteenth century Haṭhayoga became a
dominant paradigm for the practice of physical yoga across most of the Indian
subcontinent, and this paradigm was significantly shaped by the content of
the Haṭhapradīpikā. As outlined in this article, the floruit of Haṭhayoga was
a period in which its techniques proliferated, particularly in praxis-orientated
manuals, and its literature diversified as authors of various backgrounds, most
notably erudite Brahmins, attempted to expand and integrate it with other
yogas and different religions. By the eighteenth century, this extensive liter-
ary activity appears to have peaked, but the momentum behind it carried the
notion of Haṭhayoga into the royal courts of Mysore and Jodhpur in the nine-
teenth century,69 and placed it firmly at the centre of the revival of postural
practice in the twentieth century.70
69 In the mid-nineteenth century, the Mahārāja of Mysore, Mummaḍi Kṛṣṇarāja Woḍeyar

III, commissioned a royal compendium called the Śrītattvanidhi, which had a chap-
ter on āsanas that was based on the Haṭhābhyāsapaddhati (Birch 2018, 131–132). In the
early nineteenth century, the Mahārāja of Marwar had built two temples in Jodhpur (the
Mahāmandir and the Udai Mandir) with murals of eighty-four Siddhas in complex āsanas,
many of which correspond to those in the Jogapradīpyakā (Bühnemann 2007, 102).
70 Recent research is starting to reveal more premodern sources of the āsanas adopted by
via free access
Acknowledgements
I would like to thank Jacqueline Hargreaves for her comments on numerous

drafts, and also Gudrun Bühnemann, Elizabeth de Michelis, Dominic Goodall,
Shaman Hatley and Mark Singleton, who commented on earlier drafts. I’m par-
ticularly grateful to James Mallinson for his detailed comments and for sharing
unpublished research. My research for this paper was funded by the European
Research Council (ERC) under the European Union’s Horizon 2020 research
and innovation programme (grant agreement No 647963).
Abbreviations
NCC New Catalogus Catalogorum: An Alphabetical Register of Sanskrit and

Allied Works and Authors. All Volumes. Madras: University of Madras,
1968 onwards.
References
Primary Sources
Aparokṣānubhūti. Kamla Devi, ed. Aparokṣānubhūtiḥ—Vidyāraṇyakṛtayā Aparokṣadī-
pikākhyaṭīkayā Saṃvalitā. Allahabad: Akshayavaṭa Prakāśana, 1988.
Amanaska. See Birch 2013.
Amaraughaprabodha. See Mallik 1954.
Amṛtasiddhi. Unpublished working edition by James Mallinson and Péter-Dániel
Szántó for the Haṭha Yoga Project, 2019.
Ānandakanda. S.V. Radhakrsnasastri, ed. Ānandakanda. Tañjapura: Tañjapura Sarasva-
timahalgranthalayanirvahaka-samiti, 1952.
Kālacakratantra. Viśvanātha Devaśarma, ed. Kālacakratantrarāja. Calcutta: Asiatic
Society, 1985.
Kumbhakapaddhati. M.L. Gharote and P. Devnath, eds. Kumbhaka Paddhati of Raghu-
vira: Science of Prānāyāma. Lonavla: Lonavla Yoga Institute, 2000.
Garuḍapurāṇa. Part 1. Bombay: Venkatesvara Steam Press (or reprint thereof), 1907. E-
gurus who led the revival of physical yoga in the early twentieth century. For example,
Kṛṣṇamācārya was inspired by the Haṭhābhyāsapaddhati (Birch and Singleton 2019, 49–
64) and Swāmī Śivānanda of Rishikesh borrowed the majority of the āsanas in his book
Yoga Asanas (1934) from earlier sources, one of which dates to the mid-nineteenth century
(Birch forthcoming).
via free access
474 birch
text, Göttingen Register of Electronic Texts in Indian Languages. http://gretil.sub.uni

‑goettingen.de/gretil/1_sanskr/3_purana/garup1_u.htm. Accessed 1 January 2020.
Guhyasamājatantra. Yukei Matsunaga, ed. Guhyasamājatantra. Osaka: Toho Shuppan
Inc., 1978.
Gorakṣaśataka. Ms. No. R 7874 (IFP transcript T1095). A transcript of Ms. No. 2831(p),
Government Oriental Manuscript Library, Madras (Chennai).
Gheraṇḍasaṃhitā. See Mallinson 2004.
Jogapradīpikā. M.L. Gharote, ed. Jogapradīpikā. Jodhpur: Rajasthan Oriental Research
Institute, 1999.
Jogapradīpyakā. S. Maheśānanda, B.R. Śarmā, G.S. Sahāya and R.K. Bodhe, eds. Jogapra-
dīpyakā of Jayatarāma. Lonavla: Kaivalyadhama Śrīmanmādhava Yogamandira
Samiti, 2006.
Jyotsnā. S. Maheśānanda, B.R. Śarmā, G.S. Sahāya and R.K. Bodhe, eds. Brahmānan-
dakṛtā Haṭhapradīpikā Jyotsnā. Lonavla: Kaivalyadham Śrīmanmādhav Yogamandir
Samiti, 2002.
Tattvabinduyoga. Ms. No. 664 (1883–1884), Bhandarkar Oriental Research Institute,
Pune.
Dattātrayayogaśāstra. Unpublished edition by James Mallinson (with the help of Alexis
Sanderson, Jason Birch and Péter Szántó) for the Haṭha Yoga Project, 2019.
Dharmaputrikā. Yogī Naraharinātha, ed. Śivadharma Paśupatimatam Śivadharmama-
hāśāstram Paśupatināthadarśanam. Kathmandu, 1998.
Pātañjalayogaśāstra. Kāśinātha Śāstrī Āgāśe, ed. Vācaspatimiśraviracitaṭīkā-saṃvali-
tavyāsabhāṣyasametāni pātañjalayogasūtrāṇi, tathā bhojadevaviracitarājamārtaṇḍ-
ābhidhavṛttisametāni pātañjalayogasūtrāṇi. sūtrapāṭhasūtravarṇānukramasūcībh-
yāṃ ca sanāthīkṛtāni. Ānandāśrama Sanskrit Series, no. 47. Pune: Ānandāśramamu-
draṇālaye, 1904.
Puraścaraṇacandrikā. NGMPP A42/5, National Archives of Kathmandu, Nepal.
Puraścaraṇārṇava. Pratāpasiṃha Sāhadeva and Muralīdhara Jhā, eds. Puraścaryār-
ṇava: a Treatise Dealing with Theory and Practice of Tantric Worship. Delhi: Chauk-
hamba Sanskrit Pratishthan, 1985.
Prāṇatoṣiṇī. Rāmatoṣaṇa Bhaṭṭa and Saumyānanda Nātha, eds. Prāṇatoṣiṇītantra. Cal-
cutta: Nababhārata Pābaliśārsa, 1991.
Bodhasāra. J. Cover, G. Cover, and K.V. Mande, eds. Bodhasāra: an Eighteenth Century
Sanskrit Treasure by Narahari. Charleston, S.C.: CreateSpace, 2010.
Bhāvanāpuruṣottama of Ratnakheṭaśrīnivāsadīkṣita. Siromani S. Swaminatha Sastri,
ed. Bhāvanāpuruṣottamam. Tañjāvūr: Sarasvatīmahālaya Nirvāhakasamiti, 1979.
Maṇḍalabrāhmaṇopaniṣat. Pandit A. Mahadeva Sastri, ed. Yoga Upaniṣads with the
Commentary of Śrī Upaniṣadbrahmayogin. Madras: The Adyar Library and Research
Centre, 1968.
Mataṅgapārameśvaratantra. N.R. Bhatt, ed. Mataṅgapāramśvarāgama, avec le com-
via free access
mentaire de Bhaṭṭa Rāmakaṇṭha. 2 volumes. Pondicherry: Institut Français d’Indolo-

gie, 1977–1982.
Merutantra. Paṇḍita Raghunāthaśāstri, ed. Merutantra. Bombay: Srivenkatesvara
Steam Press, 1908. E-text of the Muktabodha Digital Library. http://muktabodha.org.
Yājñavalkyasmṛti. The Śāstris at the Santurāmātmajasundaramalakheḍa, eds. Maharṣi-
varyaśrīyogiyājñavalkyaśiṣyaviracitā Yājñavalkyasmṛtiḥ Vijñāneśvara viracitamitāk-
ṣarāvyākhyāsamalaṅkṛtā. Mumbai: Śrīveṅkaṭeśvara Mudraṇayantrālaya, 1919.
Yuktabhavadeva. See Gharote et al. 2002.
Yoga Upaniṣads. Pandit A. Mahadeva Sastri, ed. Yoga Upaniṣads with the Commentary
of Śrī Upaniṣadbrahmayogin. Madras: The Adyar Library and Research Centre, 1968.
Yogacintāmaṇi of Godāvaramiśra. Ms. No. 220 (1882–1883), Bhandarkar Oriental
Research Institute, Pune.
Yogacintāmaṇi of Śivānandasarasvatī. Haridās Śarma, ed. Calcutta: Calcutta Oriental
Press, n.d.
Yogacintāmaṇi of Śivānandasarasvatī. Ms. No. 3537, Scindia Oriental Research Institute,
Ujjain, India.
Yogacintāmaṇi of Śivānandasarasvatī. Ms. No. 6922, Panjab University Library, Lahore,
Pakistan.
Yogatārāvalī. Swāmī Śrīdayānanda Śāstrī, ed. Śrīmacchaṅkarābhagavatpādaviracitā
Yogatārāvalī. Varanasi: Vārāṇaseya Saṃskṛta Saṃsthāna, 1982.
Yogapañcāśikā. Ms. No. 29917 (begins at f. 7v), Sarasvati Bhavan Library, Sampurnanand
Sanskrit University, Varanasi.
Yogabīja. Rāmalāla Śrīvāstava, ed. Yogabīja of Gorakhanātha. Gorakhapur: Gorakha-
nāth Mandir, 1982.
Yogamārgaprakāśikā. Giradhara Śāstri, ed. Śrīmahāntayugaladāsanirmita-yogamārga-
prakāśikā. Bombay: Śrīvenkaṭeśvara Steam Press, 1904.
Yogayājñavalkya. Sri Prahlad Divanji, ed. Yogayājñavalkya: A Treatise on Yoga as Taught
by Yogī Yājñavalkya. B.B.R.A. Society’s Monograph, no. 3. Bombay: 1954.
Yogasandhyā. Gaṅgāviṣṇu Śrīkṛṣṇadāsa, ed. Yogasandhyā bhāṣāṭīkāsahitā |(sādhana
karanevāloṃko amṛtakī latā) vedaśāstrasaṃpanna dharmamūrti śrīyuta jagannātha
caitanya brahmacārījīke caraṇāravindānurāgī aṣṭāṅga yogameṃ kuśala śrīsadāśiva
nārāyaṇa brahmacārī nirmita. Mumbai: Lakṣmīvenkaṭeśvara Steam Press, 1954.
Yogasārasaṅgraha. IFP transcript T859 (based on D4373). E-text of the Muktabodha
Digital Library. http://muktabodha.org.
Yogasiddhāntacandrikā. Paṇḍita Ratna Gopāla Bhatta, ed. Yogadarśanam with a Com-
mentary Called the Yogasiddhāntacandrikā of Nārāyaṇatīrtha. Benares: Vidyā Vilāsa
Press, 1910.
Yogāsana ( jaina). A photocopy (R635Y8, 15294) held at the library of the Kaivalyad-
hama Yoga Institute. A stamp on a folio indicates the Rajasthān Prācya Vidyā Prati-
ṣṭhān, Bīkāner.
via free access
476 birch
Yogāsanamālā. A photocopy (R635Y8) at the library of the Kaivalyadhama Yoga Insti-

tute, Lonavla.
Rājatarala. Ms. No. 72330, Adyar Library and Research Centre, Chennai.
Rājayogāmṛtasāra. Ms. No. 22/211, Bhārat Itihās Samśodak Maṇḍal, Pune.
Vasiṣṭhasaṃhitā. Philosophico-Literary Research Department, ed. Vasiṣṭha Saṃhitā
(Yogakāṇḍa). Revised edition. Lonavla: The Kaivalyadhama S.M.Y.M. Samiti, 2005.
Vivekamārtaṇḍa. Ms. No.4110 at the Central Library, Baroda.
Vivekamukura. Padma Padmanabhan, ed. “Vivekamukura of Nṛsiṃha Bhāratī (Edited
with Introduction, Notes and Index),” The Adyar Library Bulletin 78–79 (2014–2015):
617–665.
Śāradātilakatantra. Arthur Avalon, ed. Śārada-Tilaka-Tantra. Delhi: Motilal Banarsi-
dass, 1996.
Śivasaṃhitā. James Mallinson, ed. The Śiva Saṃhitā: A Critical Edition and an English
Translation. Woodstock: YogaVidya.com, 2007
Śivasvarodaya. Swami Maheshananda, et al., eds. Śivasvarodayaḥ: A Critical Edition—
English Version. Lonavla: Kaivalyadhama S.M.Y.M. Samiti, 2015.
Śrītattvanidhi. The folios on āsana of a manuscript of this work at the Mysore Palace
were photographed and published in Sjoman 1996.
Sacitra Vyāvahārika Yoga athavā Samādhi-Mārga. Pāṇḍuraṅga Gopālabāla Mahājana,
ed. Sacitra Vyāvahārika Yoga athavā Samādhi-Mārga. Pune, Mumbai: Pustake
Milaṇyāce Ṭhikāṇa (undated).
Satkarmasaṅgraha. See Harshe 1970.
Sāṅkhyasāra. Badarī Nārāyaṇa Pañcolī, ed. Sāṅkhyasāra: Hindi anuvāda sahita. Delhi:
Pañcolī Pustakamālā Prakāśana Samiti, 1986.
Siddhāntamuktāvalī. Catalogued under the title Haṭhayogapradīpikā. Ms. no. 6756,
Rajasthan Oriental Research Institute, Jodhpur.
Haṭhatattvakaumudī. See Gharote et al. 2007.
Haṭhapradīpikā. Swami Digambaraji and Pt. Raghunatha Shastri Kokaje, eds. Haṭha-
pradīpikā of Svātmārāma. Lonavla: Kaivalyadhama S.M.Y.M. Samiti, 1998.
Haṭhapradīpikā. K. Kunjunni Raja, ed. Haṭhapradīpikā of Svātmārāma with the Com-
mentary Jyotsnā Brahmānanda. Madras: Adyar Library and Research Centre, 1972.71
Haṭhapradīpikā (with 10 chapters). See Gharote et al. 2006.
Haṭhayogamañjarī. Hatha Yoga Manjari. Lonavla: Kaivalyadhama S.M.Y.M. Samiti,
2006.
Haṭharatnāvalī. See Gharote et al. 2002.
Haṭhasaṅketacandrikā. Ms. No. B220, Oriental Research Institute, University of Mysore.
Haṭhasaṅketacandrikā. Ms. No. 2244, Man Singh Pustak Prakash Library, Jodhpur.
71 Verse numbers are cited from this edition.
via free access
Haṭhasaṃhitā. Paṃ. Nārāyaṇarādha Agnihotrī, ed. Haṭhayoga Saṃhitā, Bhāṣānuvāda

sahita. Varanasi: Śrībhāratadharma Press, 1921.
Haṭhābhyāsapaddhati. Catalogued under the title Āsanabandha. Ms. No. 46/440,
Bhārat Itihās Samśodhak Maṇḍal, Pune.
Secondary Sources
Birch, Jason. 2013. The Amanaska: King of All Yogas. A Critical Edition and Annotated
Translation with a Monographic Introduction. Ph.D. thesis, University of Oxford.
Birch, Jason. 2014. “Rājayoga: The Reincarnations of the King of all Yogas.”International
Journal of Hindu Studies 17, 3 (2013): 401–444.
Birch, Jason. 2015. “The Yogatārāvalī and the Hidden History of Yoga.” Nāmarūpa 20:
4–13. https://www.academia.edu/12099338/ (accessed 23 December 2017).
Birch, Jason. 2018. “The Proliferation of Āsana in Late Mediaeval India.” In Yoga in Trans-
formation: Historical and Contemporary Perspectives on a Global Phenomenon, edited
by Karl Baier, Philipp A. Maas and Karin Preisendanz, 101–180. Göttingen: V&R Uni-
press (Vienna University Press).
Birch, Jason. 2018a. “Premodern Yoga and Āyurveda: Preliminary Remarks on their
Shared Terminology, Theory and Praxis.” History of Science in South Asia 6: 1–83. doi:
10.18732/hssa.v6i0.25.
Birch, Jason. 2019. “The Amaraughaprabodha: New Evidence on the Manuscript Trans-
mission of an Early Work on Haṭha—and Rājayoga.” Journal of Indian Philos 47:
947–977. doi: 10.1007/s10781-019-09401-5.
Birch, Jason. Forthcoming. “The Confluence of Haṭhayoga, Tapas and Modern Postural
Practice: Distinct Regional Collections of Āsanas on the Eve of Colonialism.” Journal
of Yoga Studies.
Birch, Jason and Singleton, Mark. 2019. “The Yoga of the Haṭhābhyāsapaddhati: Haṭha-
yoga on the Cusp of Modernity.” Journal of Yoga Studies 2: 1–69. doi: https://doi.org/
10.34000/JoYS.2019.V2
Bouy, Christian. 1994. Les Nātha-Yogin et Les Upaniṣads. Paris: Diffusion De Boccard.
Bühnemann, Gudrun. 2007. Eighty-four Asanas in Yoga: A Survey of Traditions (with
Illustrations). New Delhi: D.K. Printworld.
De Michelis, Elizabeth. 2004. A History of Modern Yoga: Patañjali and Western Esoteri-
cism. London: Continuum.
Gharote, M.L. 1999. See Jogapradīpyakā.
Gharote, M.L. and Jha, V.K. 2002. Yuktabhavadeva of Bhavadeva Miśra. Lonavla: Lonavla
Yoga Institute.
Gharote, M.L. and P. Devnath. 2006. Haṭhapradīpikā (with 10 Chapters) of Svātmārāma:
with Yogaprakāśikā, a Commentary by Bālakṛṣṇa. Lonavla: Lonavla Yoga Institute.
Gharote, M.L., P. Devnath, and V.K. Jha. 2002a. Haṭharatnāvalī of Śrīnivāsayogī. Lonavla:
Lonavla Yoga Institute.
via free access
478 birch
Gharote, M.L. 2007. Haṭhatattvakaumudī: A Treatise on Haṭhayoga by Sundaradeva.

Lonavla: Lonavla Yoga Institute.
Gharote, M.L., Jha V.K., Devnath, P., and Sakhalkar, S.B. 2006a. Encyclopaedia of Tradi-
tional Asanas. Lonavla: Lonavla Yoga Institute.
Harshe, R.G. 1970. Satkarmasaṅgrahaḥ. R.G. Lonavla: Yoga-Mīmāmsā Prakāśana.
Isaacson, Harunaga and Sferra, Francesco. 2014. The Sekanirdeśa of Maitreyanātha
(Advayavajra) with the Sekanirdeśapañjikā of Rāmapāla: Critical Edition of the San-
skrit and Tibetan Texts with English Translation and Reproduction of the MSS. Napoli:
Università degli studi di Napoli “L’Orientale.”
Kuvalayananda, Swami. 1933 [1931]. Asanas. Revised edition. Bombay: Popular Praka-
shan.
Mallik, Kalyani. 1954. Siddha-Siddhānta-Paddhati and Other Works of the Nātha Yogīs.
Pune: Poona Oriental Book House.
Mallinson, James. 2004. The Gheraṇḍa Saṃhitā: The Original Sanskrit and an English
Translation. Woodstock: YogaVidya.com.
Mallinson, James. 2014. “The Yogīs’ Latest Trick.” Journal of the Royal Asiatic Society of
Great Britain & Ireland 24 (1): 165–180.
Mallinson, James. 2016. “Śāktism and Haṭhayoga.” In Goddess Traditions in Tantric Hin-
duism: History, Practice and Doctrine, edited by Bjarne Wernicke Olesen, 109–140.
Oxford: Routledge.
Mallinson, James. 2020. “The Amṛtasiddhi: Haṭhayoga’s Tantric Buddhist Source Text.”
In Śaivism and the Tantric Traditions: Essays in Honour of Alexis G.J.S. Sanderson,
edited by Dominic Goodall, Shaman Hatley, Harunaga Isaacson, and Srilata Raman.
Leiden: Brill.
Mallinson, James. Forthcoming, 2020. “Early Haṭhayoga.” In Religious Practice, The
[Oxford] History to Hinduism, edited by Gavin Flood. Oxford: Oxford University
Press.
Powell, Seth. 2018. “Etched in Stone: Sixteenth-century Visual and Material Evidence
of Śaiva Ascetics and Yogis in Complex Non-seated Āsanas at Vijayanagara.” Journal
of Yoga Studies 1: 45–106. https://journalofyogastudies.org/index.php/JoYS/article/
view/2018.v1.Powell.EtchedinStone. Accessed 14 November 2019.
Sanderson, Alexis. 2001. “History through Textual Criticism in the study of Śai-
vism, the Pañcarātra and the Buddhist Yoginītantras.” In Les Sources et le temps.
Sources and Time, edited by F. Grimal, 1–47. Publications du département d’Indo-
logie 91. Pondicherry: Institut Français de Pondichéry/ École française d’Extrême-
Orient.
Sanderson, Alexis. 2007. “The Śaiva Exegesis of Kashmir.” In Mélanges tantriques à la
mémoire d’Hélène Brunner/Tantric Studies in Memory of Hélène Brunner, edited by
Dominic Goodall and André Padoux, 231–442, 551–582 (bibliography). Pondicherry:
Institut Français de Pondichéry/École française d’Extrême-Orient
via free access
Sanderson, Alexis. 2014. “The Śaiva Literature.” Journal of Indological Studies (Kyoto) 24
& 25 (2012–2013): 1–113.
Sarde, Vijaya. 2015. “Archaeological Signatures of the Nath Cult: A Study of the Yogic
Postures and Rituals Depicted on the Brahmanath Temple at Parunde, District
Pune.” Journal of Multidisciplinary Studies in Archaeology, 3: 232–254.
Sarde, Vijaya. 2017. “‘Yoga on Stone’: Sculptural Representation of Yoga on Mahuḍī Gate
at Dabhoī in Gujarāt.” Journal of Multidisciplinary Studies in Archaeology 5: 656–675.
Singleton, Mark. 2010. Yoga Body: The Origins of Modern Posture Practice. New York:
Sjoman, Norman. 1996. The Yoga Tradition of the Mysore Palace. New Delhi: Abhinav
Publications.
Śrīkṛṣṇavallabhācārya. 2001. śrīmanmaharṣipravarapatan͂ jalipraṇi̅tam yogadarśanam
śrīmaddārśanikapañcānanaṣaḍdarśanācāryanavyanyāyācārya-sāṅkhyayogavedā-
ntamīmāṃsātīrthapaṇḍitaśrīkṛṣṇavallabhācāryaviracitena vedāntatīrthasāṅkhya-
yogaratnapaṇḍitaśrīśvetavaikuṇthaśāstrisvāminārāyaṇa-samśodhitena bhāṣyena
sahitam tathā tarkavedāntamīmāṃsāsāṅkhyayoga-tīrthapaṇḍitaśrīnārāyaṇacara-
ṇaśāstrīsvāminārāyaṇasamśodhitena—paṇḍitaśrīkṛṣṇavallabhācāryasvāmi-nārā-
yanạviracitena paṇḍitapravarabhojarājakṛtarājamārtāṇḍākhyavṛtteḥ kiraṇena sa-
manvitam. Varanasi: Caukhambā Vidyābhavana, 2001.
Thangaswami, R. 1980. Advaita-Vedanta Literature: A Bibliographical Survey. Madras:
University of Madras.
via free access
via free access
part 5
Art and Architecture
via free access
via free access
chapter 20
The Early Śaiva Maṭha: Form and Function

Libbie Mills
We should begin by determining what we think a maṭha to be in the early Śaiva

context. In the seventh and eighth centuries, maṭhas began to receive royal
patronage. By the ninth and tenth centuries maṭhas collected taxes and agri-
cultural profits (Sears 2014, 6). In the later period and in the south, maṭhas
come to be a place for pilgrims passing through, or an institution for profes-
sional adepts, a place one might abide in on a hereditary basis. There is a rise in
endowments for maṭhas in the eleventh and twelfth centuries, with the head of
the maṭha perhaps being the rājaguru, the royal guru, of the dominion, hence
a figure of political importance (Nandi 1987, 194–195).
But the maṭha of the earlier Śaiva world is at base something simpler. It is
a place for an initiate to stay in for an extended period for the purposes of
study, as an āśramin. It is housing for initiated students, with a guru offering
teaching. Brunner-Lachaux (1998, 380), describing the situation as given in the
Somaśambhupaddhati, tells us: “Celui dont Somaśambhu parle (et qu’ il nomme
āśrama en SP1, p. 316) est d’abord un lieu où vivent des initiés de différents
grades, sous la conduite matérielle et spirituelle d’ un guru.” This was not a shel-
ter for itinerant ascetics, not a locus of political power, but both a residence and
a school for initiates or, in Sanderson’s (1988, 681) terms, a lodge for cult lineage
members.
Next, let us look more closely at these maṭha residents. The residents are
initiated āśramins assigned, as Brunner-Lachaux described, into four levels
according to their type of initiation. The samayin has received the samaya
dīkṣā, the initiation for the pledge-holder or neophyte, and is qualified to study
the teachings. The putraka has received the nirvāṇa dīkṣā and will thus be lib-
erated at the moment of death. The ācārya has received the consecration for
officiants (ācārya abhiṣeka) and is qualified to teach and give dīkṣā. And the
sādhaka has received the sādhaka abhiṣeka and is qualified to practice rituals
in order to obtain supernatural powers.
As we see stated at Mohacūrottara 4.243, all these initiates, at whatever their
level, are further regarded as being veritable liṅgas, “markers” or sacred images,
of Śiva. They are mobile ( jaṅgama) liṅgas, as opposed to the fixed (ajaṅgama or
sthāvara) liṅgas which are images installed permanently in a temple. To estab-
lish either is an act of great piety:
© libbie mills, 2020 | doi:10.1163/9789004432802_022

via free access
484 mills
saṃsthāpya sthāvaraṃ liṅgaṃ prāsāde yad bhavet phalam

tat phalaṃ labhate vidvān maṭhe saṃsthāpya jaṅgamam 243
The reward that a wise man gains from establishing a mobile image
( jaṅgamam liṅgam) [i.e. an ascetic] in a maṭha is the same as the reward
that he gains from establishing a fixed image (sthāvaraṃ liṅgam) in a tem-
ple. (243)
Having found the residents to be initiates of different types, all considered to

be mobile liṅgas, let us think about their initiation into that role, and why one
might build a maṭha for them.
If the adept is a mobile liṅga, is, then, his dīkṣā (initiation) equivalent to
the pratiṣṭhā (installation) of an immobile liṅga, an image in a temple? The
matter has been considered by, among others, Hikita (2005, 193), and Mori
(2005, 232); the latter observes that pratiṣṭhā brings the deity into a fixed
liṅga, while dīkṣā does the same thing into a mobile one. As Mori (2005, 202–
203) notes from the twelfth-century Vajrāvalī of Abhayākaragupta: “and [the
ācārya] carries out also the installation of an image, etc., like the installation
of a disciple (śiṣyapratiṣṭhām iva pratimādipratiṣṭhāṃ kuryāt).” Mori goes on
to demonstrate that “in actuality there are many correspondences between the
installation and consecration ceremonies.” Given that dīkṣā (initiation of an
adept) and pratiṣṭhā (installation of an image) are parallel processes, the place
of installation, the home for the initiate, is important, just as the temple, the
shelter for a fixed image, is a place of consequence.
If the place has consequence, then there should be a reward from its estab-
lishment. An immobile liṅga is installed in a private shrine for the benefit of the
commissioning sādhaka. What of the installation of a mobile liṅga, an adept, in
a maṭha? Who benefits from that? What is the incentive? As we saw above, the
Mohacūrottara states at 4.243 that the installation of a mobile liṅga in a maṭha
brings merit to the patron who commissions and funds it. So we here move
on from the idea of installation of an immobile liṅga in the private shrine, for
the benefit of the sādhaka patron, to something else, to a liṅga that can move
about, but must still be installed and housed, bringing reward to the patron
who houses it.
In looking for accounts of this housing for adepts, I am drawing upon six
early Śaiva records, the same six on which Professor Sanderson worked with me
some two decades ago (for a study published recently as Mills 2019): the Kiraṇa,
Devyāmata, Piṅgalāmata, Bṛhatkālottara, Mayasaṃgraha, and Mohacūrottara.
I will go through what we can learn about the maṭha from these texts. We will
find the most useful material in the Mohacūrottara (10th- or 11th-century) and
via free access
the early śaiva maṭha 485
Devyāmata (in which the prāsādalakṣaṇa material shows signs of being sub-
stantially earlier; see Mills 2019).
1 Housing in General
To begin, I should give a very brief outline of how these texts describe building
plans for houses in general: normal housing, not maṭhas in particular. The texts
give accounts of the elevation, the vertical design, that are easily followed. But
descriptions of the plan, the horizontal design, rely on some basic background
knowledge, which I now supply.
Measurements are made in hastas, hands, a measure from the elbow to the
tip of the middle finger of the patron. Those measurements are checked for
their āya. The āyas are formulae used to test measurements, to ensure that they
are suitable for use. A common presentation is of six sets of āyas: āya, vyaya,
ṛkṣa, yoni, vāra, and tithi or aṃśa. There are 12 āyas in a list of items beginning
with siddhi; 10 vyayas in a list of items beginning with śikhara; 27 ṛkṣas in a list
of the nakṣatras; 8 yonis in the list of 8 that is dhvaja, etc.; 7 vāras in a list of the
days of the week; 30 tithis in a list of the lunar days in a month; and 9 aṃśas in
a list of items beginning with taskara. In each list, some members are regarded
as auspicious, some as inauspicious.
The measurement to be tested is multiplied by a set number. The product of
that multiplication is then divided by the number of items in the āya set. The
remainder is checked against the corresponding āya in the set to determine
whether the measurement is suitable or not. Let me give an example: in the
yoni āya list, the listed yonis are numbered from 1 to 8. Dhvaja (flag) is 1, dhūma
(smoke) is 2, siṃha (lion) is 3, śvan (dog) is 4, vṛṣabha (bull) is 5, khara (donkey)
is 6, gaja (elephant) is 7 and khaga (bird) is 8. The yonis with an odd number
are regarded as auspicious. Those with an even number are considered inauspi-
cious. A measurement to be checked against the yoni āyas is multiplied by the
number 3. The product is then divided by the number of yonis, 8. If the remain-
der is 1, the yoni āya for that measurement is dhvaja, which is auspicious; if it
is 2, the yoni āya for that measurement is dhūma, which is inauspicious, and
so on. Each yoni āya, from dhvaja onward, is associated with a planet, and also
with the cardinal and intermediate directions from the east onwards in a clock-
wise direction, according to the positions assigned to those planets. Below, we
will see these associations used to indicate directions in house construction.
Moving now from measurements to design, the plan for the construction
is always in the form of a square. That square is divided into a grid of cells
(padas), with the same number of cells along each side. In the pattern that we
via free access
486 mills
1 Brahmā 16 Jaya 31 Dauvārika

2 Marīcaka 17 Mahendra 32 Sugrīva
3 Vivasvant 18 Sūrya 33 Puṣpadanta
4 Mitra 19 Satya 34 Pracetas
5 Pṛthivīdhara 20 Bhṛśa 35 Asura
6 Āpa 21 Antarikṣa 36 Śoṣa
7 Āpavatsa 22 Agni 37 Roga
8 Savitṛ 23 Pūṣan 38 Vāyu
9 Sāvitri 24 Vitatha 39 Nāga
10 Indra 25 Gṛhakṣata 40 Mukhya
11 Indrajit 26 Yama 41 Bhallāṭa
12 Rudra 27 Gandharva 42 Soma
13 Rudradāsa 28 Bhṛṅga 43 Ṛgi
14 Īśa 29 Mṛga 44 Aditi
15 Parjanya 30 Pitṛ 45 Diti
38 39 40 41 42 43 44 45 14
37 13 7 15
5
36 12 6 16
35 17
34 4 1 2 18
33 19
32 10 8 20
3
31 11 9 21
30 29 28 27 26 25 24 23 22
figure 20.1 The 9 × 9 plan, Bṛhatkālottara, chapter 112 (the vāstuyā-

gapaṭala)
via free access
figure 20.2 Deity, nakṣatra, and consequence of door position. Bṛhatkālottara, prāsāda-
lakṣaṇapaṭala 238c–243b. Positions with an asterisk are those recommended
in 243c–245b
will see here, the pattern for the construction of housing, as opposed to tem-
ples or funerary grounds or other things, there are 9 cells along each side of the
square, producing a grid of 81 cells in total. See figure 20.1 as an example from
the Bṛhatkālottara of such a configuration. Once these padas have been laid
out, deities are installed in them. In the most common pattern 45 deities are
installed. 32 of those deities are placed in the 32 padas around the outer edges
of the plan, and 13 deities are placed inside that framework, with Brahmā at
via free access
488 mills
the very centre. When building houses, particular attention is paid to the con-
sequences of a doorway placed at any of the 32 padas around the periphery.
See figure 20.2 for an example of an account of doorway consequences, again
from the Bṛhatkālottara.
This fundamental 9-by-9-part deity map is used as the basis for plans of
greater or lesser complexity. For housing, we will see descriptions of three types
of design.
2 Type 1: The 9-by-9 pura
The most involved, which I will call type 1, is the design for a 9-by-9-part pura, a
residential complex for a community and its lead figure. That lead figure might
be a king or a lesser noble, an important functionary such as a general, or a
guru. Other members of the community and all the functions of community
life are arranged on the 81 cells of the 9-by-9 plan just described, with careful
specifications as to what should be placed in each cell of the periphery in par-
ticular.
Here is an account of a type 1 complex from the Mayasaṃgraha, 5.156–159
and 181–187 (between verses 159 and 181, the text digresses to describe maṇḍa-
pas and column types):
puri vā devagarbhāyāṃ niveśo vo nigadyate

mata1viṣkambhamānena kṛtasīmni mahītale 156
śaṅkvādinā gṛhāṇīśapadādīni prakalpayet
tatraiśe japahomārcādhāmopaskaraṇānvitam 157
parjanye sarvavādyāni2 vidheyāni vipaścitā
snānārghasādhanaṃ vastu jayante marubhṛnmukhe3 158
traye pratolī saddvārakapāṭārgalaśobhitā
maṇḍapaś ca vibhūtyarthaṃ geyanāṭyādisiddhaye 159
…
bhṛśāditritaye kuryāt pākasthānaṃ yathepsitam
pūṣṇi bhojanabhāṇḍāni vitathe salilāśrayaḥ 181
dhanuḥkhaḍgaśarādīni vidadhyāt tu gṛhakṣate
yame saṃyamināṃ sthānam ātmālokanasiddhaye 182
1 156c mata ] em.; matetyādi C; mano A

2 158a ºvādyāni ] C; ºpākāni A
3 158d ºbhṛnº ] C; ºtvan A
via free access
gandharve gāyakasthānaṃ bhṛṅge vyākhyānasaṃśrayaḥ4

snānadhāma mṛgasthāne koṇe śaucagṛhaṃ tataḥ 183
maṭhaṃ vā vipulaṃ kuryād gṛhakṣatacatuṣṭaye
tāmbūladantakāṣṭhādidhāma dauvārike5 hitam 184
sugrīve puṣpadante ca pracetasi ca bhojanam
abhyāgatāśrayaṃ kuryād vidvān asuraśoṣayoḥ 185
muṣalolūkhalakumbhaśilāyantrādikaṃ ruji
gandhasthānaṃ gandhavahe puṣpāyanam ahitraye 186
somadvaye kośagṛhaṃ vidyādhāmāditidvaye
brahmādiṣu padeṣv antar amarālayamaṇḍapaḥ6 187
This record lists a place for worship at Īśa; music at Parjanya; ritual bathing
and offering at Jayanta; a gateway at Indra, Sūrya and Satya (marubhṛnmukhe
traye); the kitchen at Bhṛśa, Antarikṣa and Agni (bhṛśāditritaye); eating vessels
at Pūṣan; a water tank at Vitatha; a store room at Gṛhakṣata; a place for ascetics
to achieve contemplation of the self (saṃyāmināṃ sthānam ātmālokanasid-
dhaye) at Yama; singers at Gandharva; a place for teaching the śāstras (vyākhyā-
nasaṃśrayaḥ) at Bhṛṅga; bathing at Mṛga; toilets at Pitṛ (koṇe); betel, etc., at
Dauvārika; food storage at Sugrīva, Puṣpadanta and Pracetas; a room for visitors
at Asura and Śoṣa; tools at Roga (ruji); perfumes at Vāyu (gandhavahe); flowers
at Nāga, Mukhya and Bhalvāṭa (ahitraye); the treasury at Soma and Ṛgi (soma-
dvaye); and a school (vidyādhāma) at Aditi and Diti (aditidvaye). Within this
framework, at Brahmā, etc., are a temple and maṇḍapa. At Gṛhakṣata, Yama,
Gandharva and Bhṛṅga, a maṭha is introduced as an alternative at 184ab.
3 Type 2: The 5-by-5 nandyāvarta / nandikāvarta
A second design, somewhat less elaborate, is termed the nandyāvarta or nandi-

kāvarta. This serves as a simpler residence to house higher-ups—nobility, army
generals, and gurus—and their households. In this case, laid over the funda-
mental 9-by-9-part deity map is a building design of 5 parts by 5 parts, with
spacing between chambers that produces an array of nine chambers in total,
or eight if the central position is left undesignated. Each of the eight or nine
chambers is assigned a function. See figure 20.3 for some examples of these 5-
by-5 designs.
4 183b ºsaṃśrayaḥ ] em.; ºsaṃśraye A

5 184d dauvārike ] em.; daurike A
6 187d maṇḍapaḥ ] C; maṇḍalam A
via free access
490 mills
nandyāvarta nandyāvarta nandyāvarta

Mayasaṃgraha 5.202– Bṛhatkālottara prāsādalak- Kiraṇa 19.15–17b
206b ṣaṇapaṭala 218–221
1 shrine shrine shrine

2 assembly assembly assembly
3 kitchen kitchen kitchen
4 sleeping quarters sleeping quarters sleeping quarters
5 lavatory lavatory weaponry
6 dining hall cow shed dining hall
7 treasury treasury treasury
8 tool storage granary / tool storage granary
9 king’s quarters
figure 20.3 The nandyāvarta/nandikāvarta set of nine chambers
Take for example the description of a type 2 design from Piṅgalāmata

10.96c–114b:
aṣṭaśālayutaṃ dhanyaṃ nandikāvartam7 ucyate 96cd

āyacakraṃ vibhajyādau śālāsādhanahetukam
pūrve tu śrīgṛhaṃ kuryād dhvajāyena vipaścite 97
yāmye śayyāgṛhaṃ proktaṃ siṃhāyena vijānataḥ
paścime bhojanaṃ gehaṃ vṛṣāyena8 vidur budhāḥ 98
nyāsaṃ gṛhottare kuryād gajāyena na saṃśayaḥ
dhūmāyena prakartavyaṃ pākaṃ dahanagocare9 99
pāyudaṃ pitarasthe tu śvanāyena na saṃśayaḥ
7 96d nandikāvartam ] AC; vandikāvarttam B

8 98d vṛṣāyena ] BC; vṛṣayena A
9 99d dahanaº ] AC; hadahana B • ºgocare ] BC; ºgocaraṃ A
via free access
dhānyādikṣodanaṃ gehaṃ vāyavyāṃ rāsabhena tu 100

yāgamaṇḍapa aiśānyāṃ dhvāṅkṣāyena tu sundari
munitārākareṇaiva rudras tridaśa eva ca 101
daśapañcakareṇaiva daśasaptadikchaktikam
ekaviṃśatkareṇaiva dhvajāyaṃ pūrvato bhavet 102
caturaśraṃ samantāt tu kartavyaṃ śrīgṛhottamam10
tatra sthāne sadā bhadre11 śrīmantraiḥ saha saṃyutaḥ12 103
śriyaṃ tatra likhet13 sākṣāc chrīgṛhaṃ tena cocyate
ekona14viṃśahastena pūrvapaścāyataṃ15 bhavet 104
daśasaptakareṇaiva vistaraṃ dakṣiṇottaram
siṃhāyaṃ ca bhavet tena dakṣiṇe śayanāśrayaḥ16 105
daśapañcakareṇaiva sadīrghaṃ dakṣiṇottaram
rudrasaṃkhyākareṇaiva vistaraṃ pūrvapaścimam 106
vṛṣāyaṃ ca bhavet tena paścime bhojanāśrayaḥ
trayodaśakareṇaiva pūrvapaścāyataṃ kuru 107
rudrasaṃkhyākareṇaiva dakṣiṇottaravistaram
gajāyaṃ jāyate tena bhāṇḍāgārottareṇa tu 108
daśāṣṭakaradīrghaṃ tu vistareṇa trayodaśaḥ
dhūmāyaṃ jāyate tena vahnau pākaṃ vidur budhāḥ 109
pradīrghaṃ daśahastaṃ tu ṣaṭkaraṃ vistareṇa tu
śvānāyaṃ jāyate tena pitṛsthāne tv avaśyakam 110
daśahastaṃ bhaved17 dairghyaṃ18 svaravad vistaraṃ priye
kharāyaṃ jāyate19 vasyaṃ vāyavyāṃ kaṇḍanīgṛham20 111
dviraṣṭakaradairghyaṃ tu dviṣaṭkaṃ vistaraṃ priye
dhvāṅkṣāyaṃ jāyate ’traiva aiśānyāṃ yāgamaṇḍapaḥ 112
śrīgṛhe21 vāthavā taṃ tu viṣamaṃ caturaśrakam
vittānusārato budhvā coktā nyūnādhikaṃ bhavet 113
tṛtīyaṃ nandikāvartaṃ cumbakasya prakīrtitam 114ab
10 103b ºgṛhottamam ] BC; ºgṛhottamaḥ A

11 103c bhadre ] BC; bhadraiḥ A
12 103d saṃyutaḥ ] BC; saṃyutaiḥ A
13 104a likhet ] C; likṣet AB
14 104c ekonaº ] A; ekonaṃ C; ekūnaṃ B
15 104d ºpaścāyataṃ ] A; ºpaścāyaṃ BC
16 105d śayanāśrayaḥ ] A; sayanāyayaṃ BC
17 111a bhaved ] AC; bhave B
18 111a dairghyaṃ ] BC; dīrghyaṃ A
19 111c jayate ] AC; jayante B
20 111d kaṇḍanīgṛham ] BC; kaṇḍaṇīgṛhī A
21 113a śrīgṛhe ] A; gṛhe BC
via free access
492 mills
This account lists the eight chambers of the nandikāvarta for a cumbaka as
follows: the assembly chamber is in the east, the sleeping quarters are in the
south, the dining hall in the west, the treasury in the north, the kitchen is in
the southeast, the lavatory in the southwest, the granary in the northwest, and
the shrine in the northeast. For each chamber, the correct āya proportion is
assigned.
4 Type 3: The house with four, three, two, or one rooms
The third, lowest, level of complexity produces housing for normal citizens,
graded according to either caste or initiation class. Here we see a design, again
laid over the basic 9-by-9 deity map, for a residence with only four rooms, or
three, two, or one, in descending order according to levels of caste or initia-
tion.
An example of a type 3 presentation for different initiation levels is given at
Piṅgalāmata 10.93–95 and 114c–128 (verses 96–114b cover the Nandikāvarta, as
seen above):
sāmānyaṃ saṃpravakṣyāmi cumbakādyāśrayaṃ priye

svakṛtānyakṛtāṃ22 vāpi śālāṃ caivādhunā śṛṇu 93
catuḥśālaṃ triśālaṃ ca dviśālaṃ caikaśālakam
śālāsaṃkhyā bhavanty23 etā vibhāgas tv adhunocyate 94
ācāryasya catuḥśālaṃ triśālaṃ24 sādhakasya tu
putrakasya dviśālaṃ ca samayī hy ekaśālakaḥ25 95
…
svastikāvartam anyac ca tasyaiva catuḥśālakam 114cd
vittahīno yadā bhadre tadā tat kathayāmi te
sayāgāsthānapūrve26 tu sapākaśayanāntake27 115
sapātribhojanaṃ cāpye sayantranyāsam uttare28
yady asya saṃnikṛṣṭaṃ tu taddiśāyāṃ prakalpayet 116
taddiśāya yutaṃ caiva kalpanaṃ gṛhavarjitam
22 93c svakṛtānyakṛtāṃ ] A; svakṛtānyakṛtam BC

23 94c bhavanty ] AB; bhavaty C
24 95b triśālaṃ ] AB; triśāla C
25 95d ekaśālakaḥ ] em.; ekeśālakam ABC
26 115c sayāgāsthānaº ] A; sayāgasthānaº BC • ºpūrve ] A; ºpūrvaṃ BC
27 115d ºśayanāntake ] AB; ºsamayāntake C
28 116b uttare ] AB; uttaram C
via free access
triśālaṃ sādhakasyaiva29 kathayāmi sadādhunā 117

hiraṇyāvartakaṃ caiva hīnaṃ cottaramandiram
kartavyaṃ sādhakendreṇa bhogamokṣaphalārthinā 118
suprabhāvartakaṃ vāpi prāgghīnaṃ sukhadaṃ bhavet
cullikāvartakaṃ caiva yāmyahīnaṃ na śobhanam 119
pakṣaghnāvartakaṃ devi na śastaṃ30 cāpyahīnakam
dvayaṃ grāhyaṃ dvayaṃ varjyaṃ yato vai sukhaduḥkhadam 120
dviśālaṃ putrakasyaiva śṛṇuṣva varavarṇini
vṛṣasiṃhayutaṃ dhāmasiddhārthaṃ tat prakīrtitam 121
sukhamokṣakaraṃ nityaṃ putrakasya na saṃśayaḥ
gajānaḍvānsamāyuktaṃ yamasūryaṃ vidur budhāḥ 122
mṛtyudaṃ ca yato devi varjitavyaṃ prayatnataḥ
gajadhvajasamāyuktaṃ daṇḍākhyaṃ tad vijānataḥ 123
rājadaṇḍakaraṃ nityaṃ na prāptaṃ viśeṣataḥ
dhvajasiṃhasamāyuktaṃ vātākhyaṃ tad gṛhaṃ bhavet 124
kalahaṃ ca bhaven nityaṃ na śastaṃ varjayet sadā
vṛṣadhvajasamāyuktaṃ pakṣi31nāmnā ca viśrutam 125
vittanāśakaraṃ nityaṃ varjitavyaṃ prayatnataḥ
gajasiṃhasamāyuktaṃ kākīnāmnā ca tad gṛham 126
janaiḥ saha virodhaṃ32 tu tyaktavyaṃ taṃ na saṃśayaḥ
samayinaikaśālaṃ tu tac chṛṇuṣva varānane 127
dhvajaṃ vā paścimāsyaṃ tu siṃhaṃ vā cottarānanam
vṛṣaṃ vā prāṅmukhaṃ bhadre dakṣiṇāsyaṃ gajaṃ na hi 128
In verse 10.95 we are told that the house for an ācārya has four rooms, that for
a sādhaka has three, and that for a putraka has two, while the samayin has one
room. The verses from 114cd onward describe the four-roomed, three-roomed,
two-roomed and one-roomed house. In the case of the four-roomed one, rec-
ommended for the cumbaka when money is wanting, we are given an account
of the rooms at 115cd–116ab. The shrine is in the east, the kitchen and bedroom
are in the south, the vessel store and dining room are in the west, and the uten-
sils and treasury are in the north. In the case of the three-roomed house for the
sādhaka, we are told that there may be a room lacking in the north or in the
east, but not in the south or in the west. The house with no room in the north
29 117c sādhakasyaiva ] em.; sādhakaścaiva ABC

30 120b śastaṃ ] A; saptaṃ BC
31 125d pakṣiº ] BC; pakṣī A
32 127a virodhaṃ ] A; virodhe BC
via free access
494 mills
is termed the hiraṇyanābha, while that without a room to the east is termed a
suprabhāvartaka. In the case of the 2-roomed and 1-roomed house, the direc-
tion of the rooms is described in terms of the 8 āya direction associations, from
dhvaja in the east onwards in a clockwise rotation.
5 Maṭhas
Having looked at accounts of construction of housing in general, in types 1, 2

and 3, we now are ready to look at what the texts have to say about maṭhas
in particular. The Bṛhatkālottara and Kiraṇa give details on the design of tem-
ples and also on the construction of domestic buildings for different members
of society, from kings down to ordinary caste members. But these texts do not
refer to a maṭha or anything that could be understood as a maṭha.
The Mayasaṃgraha mentions the maṭha, but quite briefly. As we saw above,
Mayasaṃgraha 5.156–159 and 181–187 describes the pura on a 9-by-9-plan, list-
ing the uses to which each of the 32 padas around the outskirts of the plan is
put: kitchen, storage areas, armories, meeting rooms, etc., in a type 1 design. On
reaching those padas at the centre of the south side, the text tells us that:
dhanuḥkhaḍgaśarādīni vidadhyāt tu gṛhakṣate

yame saṃyamināṃ sthānam ātmālokanasiddhaye 182
gandharve gāyakasthānaṃ bhṛṅge vyākhyānasaṃśrayaḥ 183ab
…
maṭhaṃ vā vipulaṃ kuryād gṛhakṣatacatuṣṭaye 184cd
At Gṛhakṣata one should set up [a storeroom for] bows, arrows, swords,

and other weapons. At Yama there should be a place for ascetics to
achieve contemplation of the self (saṃyāmināṃ sthānam ātmālokanasid-
dhaye). Singers are stationed at Gandharva. At Bhṛṅga is a hall for the
exposition [of the śāstras]. Or one may construct a large maṭha on the
four [positions] which are Gṛhakṣata and [Yama, Gandharva and Bhṛṅga].
Here, the Mayasaṃgraha is giving the maṭha as an option, to be placed on the

south edge of a pura, as an alternative to a combination of items: weapons,
ascetics, singers, and a space for teaching. Why here, on the south side? Bakker
(2004, 118) has pointed out that the south, a direction associated with the inaus-
picious, with death, is also viewed as being the benign right side of the deity or
guru as he faces east. Thus it is a suitable position for the initiate, at the right
side of the deity or guru.
via free access
The Piṅgalāmata, in chapter 10, verses 3–77, in describing a 9-by-9 type pura,
lists the uses to which each of the 32 padas all around the outskirts of the plan
are used. In verses 33–36 the Piṅgalāmata, like the Mayasaṃgraha, places the
maṭha on the south side, at the padas of Yama and Gandharva.
yame gandharvake kuryān maṭhaṃ bhūmitrayānvitam 33cd

athavā dvayasārdhaṃ tu bhūmikaikāthavā priye
uttamaṃ madhyamaṃ caiva kanyasaṃ ca yathākramam 34
ācāryasya tad evoktaṃ nijaṃ śayanahetukam
iṣṭāhnikaṃ jayaṃ dhyānaṃ yogābhyāsaṃ tu tatra vai 35
vīrabhojyānnapānādyair vīraiḥ saha samācaret 36ab
My dear, at Yama and Gandharva one should make a maṭha with three
storeys, two [storeys] or one storey. [These are] the best, middling and
least [maṭhas] in turn. That is the place for the ācārya to sleep, for [prog-
nostication of] auspicious days, triumph, meditation, and the practice of
yoga. [There the teacher] may associate with vīras (vīraiḥ saha), sharing
vīra food and drink, etc.33
Altogether, so far, from the Piṅgalāmata and Mayasaṃgraha, we have learned

that the maṭha is generally placed in the south. The Mohacūrottara and Devyā-
mata give more information on the nature of the construction itself. The
Mohacūrottara uses the term maṭha. The Devyāmata does not.
First let us look at what the Mohacūrottara has to say. At Mohacūrottara
4.234–243 we get a more detailed depiction of the maṭha:
maṭhaś cāntakadigbhāge liṅgināṃ sthitaye hitaḥ

yatas te dakṣiṇāśāyāṃ vaseyuḥ śivabhāvitāḥ 234
prāsādavistaraṃ sūtraṃ tanmānaṃ jagatībahiḥ
prākāraṃ kārayet tyaktvā tataś cāśramiṇāṃ gṛham 235
maṭhāgre tatsamaṃ tyaktvā siṃhāyaṃ dakṣiṇe sthitam
vṛṣāyaṃ34 paścime jñeyaṃ dhvajāyaṃ35 pūrvataḥ sthitam 236
vipulaṃ vā prakartavyaṃ kartur icchāvaśena tu
33 Shaman Hatley notes that “Vīra likely refers to sādhakas: ‘heroic sādhakas’ might be a
good rendering. Vīrapāna refers to impure liquids such as alcohol, used in rituals of the
Bhairava- and Kulatantras.”
34 236c vṛṣāyaṃ ] F; vṛṣoyaṃ H
35 236d dhvajāyaṃ ] F; dhvajeyaṃ H
via free access
496 mills
caturaśre śarair bhakte madhyaṃ tyaktvā vilopayet 237

gṛhāṇāṃ svecchayā nyāsaḥ saumye syād veśanirgamaḥ
ekabhaumaṃ dvibhaumaṃ vā tribhaumaṃ vā yathāsukham 238
dīrghaśālāvṛtaṃ bāhye prāci yāgālayānvitam
pākādigṛhavinyāsaṃ yathāyogaṃ niveśayet 239
pūrvoktam antaraṃ sārdhaṃ36 tasyāpy ardhaṃ yathāyatham
maṭhikaikātra†yaṃ kāryaṃ† paṭṭaśālā catuṣkikā 240
bhūmayaḥ pūrvam uddiṣṭā vittābhāve kuṭī matā
samasūtraṃ susaṃsthānaṃ vāstupūjāpuraḥsaram 241
liṅgināṃ ca gṛhaṃ kāryaṃ mahāpuṇyajigīṣayā
etad37 eva mahāpuṇyaṃ kathayāmi tavākhilam 242
saṃsthāpya sthāvaraṃ liṅgaṃ prāsāde yad bhavet phalam
tat phalaṃ labhate vidvān maṭhe saṃsthāpya jaṅgamam 243
And a maṭha for ascetics to stay in (liṅgināṃ sthitaye) should be in the

south. For they, as devotees of Śiva, should reside to the right [of
Śiva]. (4.234)
One should build a wall (prākāram) at a distance 1 temple-width
beyond the temple base ( jagatī). At a distance from there (tyaktvā
tataḥ) is the housing for ascetics (āśramiṇāṃ gṛhaṃ). (4.235)
In front of the maṭha (maṭhāgre), leaving a distance of the same [size]
(tatsamaṃ tyaktvā), [houses should be built that are sized according
to] the siṃhāya in the south, the vṛṣāya in the west, and the dhvajāya
in the east. (4.236)
Or they may be made as large as the patron wishes. They are on a square
site divided into five [parts along each side]. One should leave the
intermediate spaces empty (madhyaṃ tyaktvā vilopayet). (4.237)
The installation of the houses is according to the wishes [of the patron].
There should be a [door for] entry and exit to the north. [The houses]
may have one, two, or three floors, or as is pleasing. (4.238)
Externally, [the building] is surrounded by a long hall (śālā). In the
eastern side of the building is the place for worship ( yāgālaya). One
should install the kitchen and so forth as appropriate. (4.239)
At a distance of 1½-times the previously given distance, and half that, as
is suitable, is a single maṭhikā, in the form of a set of four (catuṣkikā)
awnings (paṭṭaśālā). (4.240)
36 240a sārdhaṃ ] em. Sanderson; sārdhaṃ Ātmārthapūjāpaddhati; cārddhaṃ FH

37 242c etad ] H; tad F
via free access
The storeys (bhūmi) are as have already been taught. If money is lacking,
a hut (kuṭī) is approved. (4.241ab)
One should build the dwelling for ascetics with the same measurements
and a good design, [and] performing the veneration of the site (vās-
tupūjā), out of a desire to attain great merit. I will now describe to
you this great merit in full. (4.241cd–242)
The reward gained from establishing a mobile image ( jaṅgamaṃ
liṅgam) [i.e., an ascetic] in a maṭha is the same as the reward gained
from establishing a fixed image (sthāvaraṃ liṅgam) in a temple.
(4.243)
So, here, we seem to have a maṭha housing complex to the south of the temple,
bracketed to the east, south and west by actual residences. Verse 237 indicates
that each individual residence is of the 5-by-5 type 2 variety. The term maṭhikā
is introduced, referring to a set of 4 awnings, and a simple hut, kuṭī is given as
a cheaper alternative.
And, finally, in Devyāmata, chapter 105 we are given another account of the
residence for initiates. The terms used for the residence are gṛha (verses 1, 15, 17),
āśrama (verses 15, 17), and veśman (verse 41). The term maṭha is not used. The
terms used for the residents of these domiciles are āśramin (verse 1), prāsādāś-
ramin (verse 82), dīkṣita (verses 16, 17), gṛhin, and gṛhamedhin (verse 72). In
verses 15–19b is given an account of the residence (gṛha) for the initiate (dīk-
ṣita), outside the temple exterior wall and to the south of the temple:
dakṣiṇāyāṃ diśi śastaṃ gṛham āyatanasya tu

prākārasya bahiḥ kāryam āśramāyatanasya tu38 15
vastavyaṃ liṅgam āśṛtya dīkṣitaiḥ sujitendriyaiḥ
tadabhāve ’thavānyasmiṃ pradeśe sumanorame 16
nātidūrāmbhasaḥ kāryam āśramaṃ dīkṣitātmanāṃ
asaṃkīrṇẹ śubhe deśe vastavyaṃ dīkṣitātmabhiḥ 17
ekaśālaṃ dviśālaṃ vā triśālam athavā gṛham
catuḥśālagṛhaṃ vāpi kāryaṃ vittānusārataḥ 18
kāryaṃ hiraṇyanābham vā sukṣetram vā manoramam 19ab
A residence (gṛham) is recommended to the south of the temple. The

residence (āśramam) should be built beyond the outer wall of the
temple. (15)
38 Sanderson, by email communication, points out that the sense intended is certainly āśra-
mam āyatanasya tu.
via free access
498 mills
It is to be dwelt in by initiates (dīkṣitaiḥ), their senses well-subordinated,

who have come to (āśṛtya) the image (liṅgam). Or, in its absence
(tadabhāve), [they should dwell in] another pleasant place (pradeśe
sumanorame). (16)
The residence (āśramam) for the initiates (dīkṣitātmanām) should be
built not too far from water. Initiates (dīkṣitātmabhiḥ) should live in a
fine, unpolluted place. (17)
The residence (gṛham) should have one, two, or three rooms. Or a four-
roomed residence should be built, according to funding. (18)
A pleasing hiraṇyanābha or sukṣetra39 may be built. (19ab)
In verse 18, the housing described is of type 3. In verses 54–74b the conse-
quences of a doorway at each of the peripheral padas of the 9 by 9 deity map
are given in some detail:
evaṃ yathākrameṇaiva dvārāṇāṃ phalam ucyate

īśe hy agnibhayaṃ vidyāt parjanye strīviṣo gṛhī 54
jaye ca dhanasampanno māhendre nṛpavallabhaḥ
krodhaparas tathāditye satya ṛta40paro bhavet 55
bhṛśe tasya bhavet krauryaṃ cauryaṃ caiva tathāmbare
alpasutas tathā cāgnau pūṣākhye preṣyatāṃ vrajet 56
vitathe ’vinītatāṃ yāti gṛhī gṛhakṣate sudhī
yame ca raudratāṃ yāti gāndharve śrīm avāpnuyāt 57
bhṛṅgarāje bhaven nisvo mṛgākhye41 nṛpapīḍitaḥ
uktaṃ dvārāṣṭakaṃ tubhyaṃ gṛhe ’smin dakṣiṇāmukhe 58
procyate saviśeṣeṇa gṛhe ’smim paścimāmukhe
paścime pitṛdevatye gṛhī syāt sutapīḍitaḥ42 59
ripuvṛddhis tathā tasya vidyā dauvārike pade
sugrīve dhanasampat syād gṛhiṇaḥ sarvadā bhavet 60
sutārthabalasampat syāt pade ’smim puṣpadantake
vāruṇe dhanasampattiṃ nṛpabhayaṃ tathāsure 61
39 As recorded, for example, in Bṛhatkālottara, prāsādalakṣaṇapaṭala 223–224 and Kiraṇa

19.18–19, the hiraṇyanābha is a building with three rooms, in the east, west and south, and
the sukṣetra is a building with three rooms, in the west, south and north. At Piṅgalāmata
10.118–119 (above), the building without a room in the north is termed a hiraṇyanābha,
while that without a room to the east is termed a suprabhāvartaka.
40 ṛtaº ] em.; nṛtaº MW; nanṛṃta N. The emendation to ṛta is supported by the fact that at
Bṛhatkālottara, prāsādalakṣaṇapaṭala 239 we see dharma at Satya.
41 58b mṛgākhye ] em.; mṛṣākhye NM; mṛṣākhya W
42 59d sutapīḍitaḥ ] N; sutapātitaḥ MW
via free access
dhanakṣayaṃ tathā śoṣe rogaḥ syāt pāpayakṣmaṇi

aṣṭau devāḥ samākhyātā gṛhe ’smin paścimāmukhe 62
vāyavyādikrameṇaiva procyate hy uttarāmukhe
baddhabandhas tathā roge ripuḥ syān nāgasaṃjñike 63
mukhye sutārthalābhaṃ syāt sampad bhalvāṭake tathā
dhanasampat tathā some putravairam anantake 64
strīdoṣaś cāditau jñeyo daridrā gṛhiṇo ditau
kathitāni viśeṣeṇa svadevatānvitāni tu 65
hitāvahitāni yāni syūr dvārāṇi śṛṇu yatnataḥ
jayākhyaṃ yat tṛtīyaṃ tu suprabhūtadhanapradam 66
māhendrākhyaṃś caturthaṃ tu gṛhiṇāṃ sarvakāmikam43
gṛhakṣataṃ caturtham tu gṛhe ’smin dakṣiṇāmukhe 67
bhakṣyapānasutavṛddhiṃ karoti gṛhamedhinām
gandharvākhyaṃ tathā ṣaṣṭaṃ śrīsaukhyaś ca sukhapradam 68
dvāradvitayaṃ śaṣṭaṃ hi gṛhe ’smin dakṣiṇāmukhe
dhanasampatkaraṃ dvāraṃ tṛtīyam paścimāmukhe 69
caturthaṃ puṣpadantākhyaṃ sutārtha44balavardhanam
pañcamaṃ vāruṇaṃ dvāraṃ dhanasampatkaraṃ nṛṇām 70
dvāratṛtayam ākhyātaṃ gṛhe ’smin paścimāmukhe
mukhyādhidevataṃ dvāraṃ tṛtīyaṃ cottarāmukhe 71
dhanasutārthasampattiṃ karoti gṛhamedhinām
bhalvāṭākhyaṃ caturthaṃ tu gṛhiṇāṃ sarvakāmadam 72
dhanasampatkaraṃ proktaṃ pañcamaṃ somadevatam
evaṃ jñātvā45 viśeṣeṇa yathoktaṃ46 dvāralakṣaṇam 73
guṇādhikaṃ47 tato vidvān sthāpayed dvāram ādarāt 74ab
Thus, in due sequence, the consequences (phalam) of doorways are

given. [With a doorway] at Īśa, the householder will have the risk of
fire; at Parjanya, harm from women. (54)
At Jaya [the householder] is endowed with wealth. At Māhendra he is
dear to the king. At Āditya there is anger. At Satya there is lawful con-
duct. (55)
At Bhṛśa is awfulness. And at Ambara there is theft. At Agni there is a
lack of sons. At Pūṣan is servitude. (56)
43 67b sarvakāmikam ] em.; sarvakarmikam N; sakāmikam MW

44 70b sutārthaº ] MW; sutākhyaṃ N
45 73c jñātvā ] N; jñāna MW
46 73d yathoktaṃ] N; yathoktā MW
47 74a guṇādhikaṃ ] N; gaṇādhikan M; gaṇādhika W
via free access
500 mills
At Vitatha the householder comes to a lack of decorum, at Gṛhakṣata he

gains wisdom. At Yama he attains savagery. At Gāndharva he acquires
glory. (57)
At Bhṛṅgarāja there is malady. At Mṛga one is oppressed by the king.
The set of 8 doorways have been described to you, in the house facing
south. (58)
Next it will be specifically described for the house facing west. In the
west, at the Pitṛdeva position, the householder will be oppressed by
his sons. (59)
There is an increase in the enemy and his knowledge at Dauvārika. At
Sugrīva is always an increase of wealth for the householder. (60)
At Puṣpadantaka is a gain in sons, wealth and power. At Vāruṇa is an
increase in wealth. At Asura is danger from the king. (61)
There is loss of wealth at Śoṣa and disease at Pāpayakṣman. Eight deities
have been listed, in the house facing west. (62)
Those facing north are listed next, in sequence, from the northwest
(vāyavya) on. At Roga is bondage. At Nāga (Vāsuki) is an enemy. (63)
At Mukhya is an increase in sons and wealth. At Bhalvāṭa is gain. At
Soma is a gain in wealth. At Anantaka is heroism in sons. (64)
At Aditi is trouble from women. At Diti is poverty. Specifically listed
with their own deities are those doorways which are especially good.
Listen with care. The third one, named Jaya, brings great power and
wealth. (65–66)
The fourth one, named Māhendra, fulfills every desire for the house-
holder. The fourth one in the house facing south, Gṛhakṣata,
increases food, drink and sons for householders. The sixth one, called
Gandhārva, brings glory, pleasures and contentment. (67–68)
The second set of doorways has been declared, on the south side. On
the west side, the third doorway (i.e., at Sugrīva) brings an increase in
wealth. (69)
The fourth, called Puṣpadanta, increases sons, power and strength. The
fifth doorway, Vāruṇa, brings increased wealth for men. (70)
The third set of doorways has been declared, on the west side. And on
the north side, the doorway governed by Mukhya brings an increase
in wealth, sons and property to householders. The fourth one, named
Bhalvāṭa, gives men every desire. (71–72)
The fifth one, whose deity is Soma, brings an increase in wealth. After
learning the features of doorways as described, specifically, the wise
man should carefully establish a doorway with ample good qualities.
(73–74ab)
via free access
One should note that, while these are the consequences for doorway posi-
tions in a residence for ascetics, the consequences do not fall on the ascetic
residents themselves, but on the patron who commissions and funds the build-
ing of the residence, and who gains the benefit from it.48 So, the consequences
are not in any way to be connected to the lives of the residents. None-the-less,
it is worth noting that these are the same consequences that we see repeated
over and again for domestic buildings of all sorts. The model that is being used
is that for normal housing.
The portion from 81 to 86 describes the layout of the residence (gṛha) for the
prāsādāśramin:
gṛhaṃ niṣpādayed yatnād yathokta49lakṣaṇānvitam

antaradiśvibhāgaṃ ca gṛhaprāsādayos tataḥ 81
uktaṃ yathākrameṇaiva prāsādāśramiṇāṃ gṛham
āgneyyāṃ mahānasaṃ śastam īśānyāṃ yāgamaṇḍapam 82
ratnahiraṇyavastrāṇām aindrādiśi praśasyate
yāmyāśre ’py uttare50 kāryaṃ sthāpanaṃ salilasya tu 83
dhānyānāṃ sthāpanaṃ śastaṃ vāruṇyāṃ sarvadāhitam
udūkhalasya vāyavyāṃ sthāpanaṃ samudāhṛtam 84
sarvavastuṣpadānāṃ tu kauberyāṃ nilayaṃ smṛtam
nātidūraṃ na cāsannaṃ pracchannaṃ parivāritam 85
gṛhasyāvāhṛtaṃ kāryam avaśyakāraṇaṃ gṛham
pūrveṇa vanaṣaṇḍaṃ tu tathā puṣpaphaladrumāḥ 86
One should carefully arrange the residence (gṛham) in such a way that
is has the characteristics that have been taught. Then [one should
arrange] the area in between (antaradiśvibhāgam) the residence and
temple. (81)
The residence for those who come to the temple (prāsādāśrami-
ṇām) is described in due sequence. In the southeast is the kitchen
(mahānasam). In the northeast is the space for worship ( yāgamaṇḍa-
pam). (82)
Storage for gems, gold and cloths (ratnahiraṇyavastrāṇām) is recom-
mended in the east, and for water in the south and centre.51 (83)
48 I thank Shaman Hatley for this observation.

49 81b yathoktaº ] MW; antaraṃ N
50 83c yāmyāśre ’py uttare ] MW; yāmyāśreruttare N
51 I take uttare to refer to the position to the north of the southern cell. That is, the centre. I
via free access
502 mills
Grain storage (dhānyānām) is recommended in the west. In the north-

west is storage for the mortar (udūkhalasya). (84)
To the north is general storage (sarvavastuṣpadānām). Not too far away
(nātidūram), nor adjoining (na cāsannam), is a secluded (pracchan-
nam), sheltered (parivāritam) (85)
lavatory building (avaśyakāraṇaṃ gṛham), aside from the residence
(gṛhasyāvāhṛtam). To the east (pūrveṇa) should be made a copse
(vanaṣaṇḍam), and trees with flowers and fruit (puṣpaphaladrumāḥ).
(86)
This is an account of a type 2 nandikāvarta type of construction, with 8 rooms

on a 5-by-5 plan.
From verse 87 on we get a detailed description of the trees for the surround-
ing gardens. Then the chapter ends:
kathitaṃ saviśeṣeṇa gṛham āśramiṇāṃ tava52 94cd

antaraṃ diśvibhāgaṃ ca prāsādagṛhayor drumān
prākāraḥ svapramāṇena prāsādānāṃ prakīrtitaḥ 95
prākārasya bahiḥ proktam udyānaṃ sumanoramam
kartavyaṃ ca tathodyānaṃ prākārapariveṣṭitam 96
pūrvavac ca ṛjuḥ kāryaḥ prākāraḥ sucayas samaḥ
prākārasya bahiḥ sthāpya parivārālayāḥ priye 97
evaṃ yathākrameṇaiva kāryaṃ sarvam aśeṣataḥ
gṛhād udyānaparyastaṃ prākāreṇa samanvitam 98
tat sarvaṃ samudāyena kartavyaṃ samudāhṛtam 99ab
iti gṛhavāstupaṭalaḥ
The residence (gṛham) for the āśramins has been described to you in
particular; (94cd)
and [also] the intervening (antaram) area (diśvibhāgam) between the
temple and residence (prāsādagṛhayoḥ), and trees. The surround-
ing wall (prākāraḥ), with its measurements, has been described for
temples. (95)
The pleasing garden (udyānam) beyond the surrounding wall has been
described. The garden too is to be surrounded by a wall (prākāraveṣṭi-
tam). (96)
find support in the fact that, in verse 22 of this chapter, we were informed that the water
supply should be in the centre of the house (gṛhamadhye).
52 94d tava] em.; bhavaḥ MW
via free access
As before, the surrounding wall should be made straight, well erected

and level. My dear, having established the subsidiary shrines (parivār-
ālayāḥ) beyond the wall, (97)
everything should be made thus, complete, and in due sequence.
Beyond the residence, it is surrounded by a garden and has a sur-
rounding wall. Everything to be done has been altogether declared.
(98–99ab)
Thus ends the chapter on the residence.
Looking back over what we have seen, in the Bṛhatkālottara and Kiraṇa we saw
no mention of a maṭha, but from the Mayasaṃgraha and Piṅgalāmata came
the information that the maṭha should be on the south side. When it came
to details of the maṭha design, we saw in the Mohacūrotttara and Devyāmata
descriptions that looked very much like those for houses for any other person,
in types 2 and 3, to the south of a type 1 complex.
We have been looking at the building designs for clues as to what went on
inside them, following the sensible line of thinking of Sears (2014, 76), who
writes, “the architecture of the monastery indexes the concerns of its residen-
tial community.” But perhaps all we have learned from the building designs
for the maṭha is that dorms are dorms, in the end. While the installation of a
jaṅgama liṅga, an initiate, is equal in merit to the installation of an ajaṅgama
liṅga, an image, there is by no means the same glamour in its housing. This
proves to be a practical domestic establishment entirely like that of an alto-
gether ordinary person who is not initiated—not a jaṅgama liṅga.
Acknowledgment
My gratitude to Professor Sanderson is enormous. His guidance is invaluable

in many ways, but I will point out just two. On the one hand there is his stern
search for error in the material, on the other is his patient tolerance of error
in the student—a tolerance that I put to some pretty severe testing, of course.
None of this work would have been possible without it.
References
Primary Sources
Kiraṇatantra
Dominic Goodall, ed. Bhaṭṭarāmakaṇṭhaviracitā Kiraṇavṛttiḥ. Bhaṭṭa Rāmakaṇṭha’s
Commentary on the Kiraṇatantra. Volume I: Chapters 1–6. Critical Edition and Anno-
via free access
504 mills
tated Translation. Publications du département d’indologie, no. 86.1. Pondichéry:

Institut français de Pondichéry/École française d’Extrême-Orient, 1998.
Śrīmat-Kiraṇāgama-mahātantram. Pañcāpageśa śivācārya and K.M. Subrahmaṇyaśā-
strī, eds. Devakoṭṭai: Śivāgamasiddhāntaparipālanasaṅgha, 1932.
NAK 5–893; NGMPP reel no. A40/3.
NAK 4–13; NGMPP reel no. B117/11.
IFP T. no. 47637.
IFP T. no. 39802.
IFP T. no. 998.
IFP T. no. 0746.
IFP T. no. 0401.
Devyāmata
(N) NAK 1–279; NGMPP reel no. A 41/15.
(M) NAK 1–1003; NGMPP reel no. B 27/6.
(W) Wellcome Institute, gamma 607.
Piṅgalāmata
(A) British Library OR 2279
(B) NAK 3–376; NGMPP reel no. A42/2
(C) NAK 5–1929; NGMPP reel no. A166/5
Bṛhatkālottara
NAK 1–273; NGMPP reel no. B24/57
NAK 5–778vi; NGMPP reel no. A42/8
NAK 4–131; NGMPP reel no. A43/1
Bṛhatsaṃhitā of Varāhamihira
Avadha Vihārī Tripāṭhī, ed. Bṛhatsaṃhitā of Varāhamihira with the Commentary of Bhaṭ-
ṭotpala. Varanasi: Varanaseya Sanskrit Vishvavidyalaya, 1968.
M. Ramakrishna Bhat, ed., trans. Varāhamihira’s Bṛhatsaṁhitā, 2 vols. Delhi: Motilal
Banarsidass, 1981, 1982.
Kṛṣṇacandra Dvivedī, ed. Bṛhatsaṃhitā by Śrī Varāhamihira with Commentary of Bhaṭ-
ṭotpala. Varanasi: Sampurnanand Sanskrit University, 1996.
Mayasaṃgraha
(A) NAK 1–1537; NGMPP reel no. A31/18
(C) A copy of a commentary to the Mayasaṃgraha: the Bhāvacūḍāmaṇi of Bhaṭṭa
via free access
Vidyākaṇṭha. Shri Raghunath Temple Manuscript Library no. 5291, now in the col-
lection of the Ranbir Research Institute, Jammu.
Mohacūrottara
(F) NAK 1–1633; NGMPP reel no. B26/29
(G) NAK 4–1622; NGMPP reel no. B27/18 (missing the section of chapter 4 discussed
here).
(H) NAK 5–1977; NGMPP reel no. A182/2.
Somaśambhupaddhati
Hélene Brunner-Lachaux, ed. 4 volumes. Collection Indologie, nos. 25.1–4. Pondichéry:
Institut Français de Pondichéry, 1963, 1968, 1977, 1998.
Secondary Sources
Bakker, Hans. 2004. “At the Right Side of the Teacher: Imagination, Imagery and Image
in Vedic and Śaiva Initiation.” In Images in Asian Religions: Texts and Contexts, edited
by Koichi Shinohara and Phyllis E. Granoff, 117–149. Vancouver, B.C.: University of
British Columbia Press.
Granoff, P.E. 2004. “Images and their Ritual Use in Medieval India: Hesitations and Con-
tradictions.” In Images in Asian Religions: Texts and Contexts, edited by Koichi Shi-
nohara and Phyllis E. Granoff, 19–55. Vancouver, BC: University of British Columbia
Press.
Hikita, Hiromichi. 2005. “Consecration of Divine Images in a Temple.” In From Material
to Deity, Indian Rituals of Consecration, edited by Shingo Einoo and Jun Takashima,
143–198. Manohar: New Delhi.
Mills, Libbie. 2019. Temple Design in Six Early Śaiva Scriptures. Critical Edition and Trans-
lation of the Prāsādalakṣaṇa-portions of the Bṛhatkālottara, Devyāmata, Kiraṇa, Mo-
hacūrottara, Mayasaṃgraha & Piṅgalāmata. Collection Indologie, no. 138. Pondich-
éry: Institut Français de Pondichéry/École française d’Extrême-Orient.
Mishra, R.N. 1993. “The Saivite Monasteries, Pontiffs and Patronage in Central India.”
Journal of the Asiatic Society of Bombay 64–66: 108–124.
Mori, Masahide. 2005. “The Installation Ceremony in Tantric Buddhism.” In From
Material to Deity, Indian Rituals of Consecration, edited by Shingo Einoo and Jun
Takashima, 199–240. Manohar: New Delhi.
Nandi, R.N. 1987. “Origin and Nature of Śaivite Monasticism: The Case of the Kālā-
mukhas.” In Indian Society: Historical Probings, In Memory of D.D. Kosambi, edited
by R.S. Sharma, 190–201. New Delhi: People’s Publishing House.
edited by S. Sutherland, et. al., 660–704. London: Routledge and Kegan Paul.
Sanderson, Alexis. 1995. “Meaning in Tantric Ritual.” In Essais sur le Rituel III, edited by
via free access
506 mills
A.-M. Blondeau and K. Schipper, 15–95. Bibliothèque de l’École des hautes études,
Sciences religieuses, vol. CII. Louvain-Paris: Peeters.
Sanderson, Alexis. 2001. “History Through Textual Criticism in the Study of Śaivism,
the Pañcarātra and the Buddhist Yoginītantras.” In Les Sources et le temps, edited by
François Grimal, 1–47. Publications du département d’Indologie, no. 91. Pondichéry:
Brahmanical Royal Chaplain (with an appendix on the provenance and date of the
Netratantra).” Indo-Iranian Journal 47: 229–300.
Sanderson, Alexis. 2007. “The Śaiva Exegesis of Kashmir.” In Melanges tantriques a
la memoire d’Helene Brunner/Tantric Studies in Memory of Helene Brunner, edited
by Dominic Goodall and Andre Padoux, 231–442, 551–582 (bibliography). Collec-
tion Indologie, no. 106. Pondichéry: Institut Français de Pondichéry/École française
Sanderson, Alexis. 2009. “The Śaiva Age: The Rise and Dominance of Śaivism during the
Oriental Culture, University of Tokyo.
Sears, Tamara I. 2007. “Śaiva Monastic Complexes in Twelfth-Century Rajasthan: The
Pāśupatas and Cāhamānas at Menāl.” South Asian Studies 23: 107–126.
Sears, Tamara I. 2014. Worldly Gurus and Spiritual Kings, Architecture and Asceticism in
Medieval India. New Haven; London: Yale University Press.
via free access
chapter 21
The Kriyāsaṃgrahapañjikā of Kuladatta and its

Parallels in the Śaiva Pratiṣṭhātantras
Ryugen Tanemura
1 Introduction
Most people think that the prominent features of Tantric religions are their
esoteric teachings. This might mislead us into supposing that these religions
were limited to restricted communities. But in fact these religions extended
over a much wider domain. We see this if we consider the fact that both Śaivism
and Tantric Buddhism offered a wide range of public social rituals.1 Following
1 Sanderson proposes that what kept Śaivism alive, and enabled it to exert its influence, was
ritual for others, as the professional activity of officiants who operated outside the narrow
confines of self-cultivation (Sanderson 2010, 12).
Generally speaking, rituals for others, i.e. rituals performed for the benefit of donors,
were formed through modification of rituals for personal salvation. In the case of Tantric
Buddhism, the pratiṣṭhā ritual is a modification of the utpattikrama practice. See the fol-
lowing three quotations: Ratnākaraśānti’s Bhramahara: tasyānandina āsyena dvihoḥkāravi-
darbhitam | jvalad bījadvayaṃ rāgāt padmāntaḥ praviśad dravet || tato vajrī mahārāgād
vilīya saha vidyayā | śaraccandradravanibhāṃ tiṣṭhen maṇḍalatāṃ gataḥ || athotthānāya taṃ
devyaḥ sthitvā koṇāsanenduṣu | codayeyuś catasṛbhiś catasro vajragītibhiḥ || (Isaacson 2002,
162, lines 9–15); Vajrāvalī (abhiṣeka section): tais tathāgataiḥ prajñāsamāpannair mahārā-
geṇa dravībhūya vairocanadvāreṇāntarniviśya vajramārgeṇa nirgatya taddravair devīpadme
mukhena praveśitaṃ śiṣyaṃ jhaṭiti śūnyatānantaraṃ hūṃvajrajātasaprajñākṣobhyarūpiṇaṃ
jñānasattvābhinnam abhiṣicya punar bhujamukhādimūrtibhiḥ *padmān (corr.; padmāt EM)
niḥsṛtya gaganam āpūrya sthitair locanādividyāsahitaiś chatradhvajapatākāvastravāditragī-
tanṛtyapuṣpakuṅkumādivṛṣṭibhiḥ karakiśalayāvarjitabodhicittāmṛtapūrṇasitakalaśais taṃ śi-
ṣyaṃ *padmān (corr.; padmāt EM) niḥsṛtam abhiṣicyamānaṃ … (EM §24.2, vol. 2, p. 341,
lines 6–11); Vajrāvalī (pratiṣṭhā section): taiś ca tathāgataiḥ prajñāsamāpannair mahārāgeṇa
dravībhūya svasya vairocanadvāreṇa praviśya vajramārgeṇa nirgatya taddravair devīpadme
mukhena praveśitaṃ pratimādikam abhiṣicya punar bhujamukhādimūrtibhiḥ *padmān (corr.;
padmāt EM) niḥsṛtya bahir ambaram āpūrya sthitair locanādividyāsahitaiś cchatrapatākānṛ-
tyagītavāditrakusumakuṅkumādivṛṣṭiparikaritakarakiśalayāvarjitabodhicittāmṛtapūrṇasita-
kalaśais tat pratimādikaṃ padmād bahir niḥsṛtam abhiṣicyamānaṃ … (EM §17.3, vol. 2, p. 416,
line 17–p. 417, line 2). The first passage quoted from the Bhramahara teaches how the practi-
tioner should generate himself as Hevajra in the First Union (ādiyoga). The practitioner, who
has the form of the seed syllables, should enter the womb of Nairātmyā, Hevajra’s consort,
through Hevajra’s mouth, become liquid (i.e. the state of śūnyatā), be emitted outside the
© ryugen tanemura, 2020 | doi:10.1163/9789004432802_023

via free access
508 tanemura
Śaiva models, Tantric Buddhism offered various kinds of consecration cere-

monies (pratiṣṭhā).2 Śaivism produced Paddhatis and Pratiṣṭhātantras3 which
teach the details of these public social rituals. Tantric Buddhism also produced
a number of manuals which are closely comparable to Śaiva Pratiṣṭhātantras
and Paddhatis. Of these, three manuals are particularly rich in information: the
Kriyāsaṃgrahapañjikā of Kuladatta, the Vajrāvalī of Abhayākaragupta, and the
Ācāryakriyāsamuccaya of Jagaddarpaṇa or Darpaṇācārya, which incorporates
much of the Vajrāvalī but adds some new materials (Sanderson 2009, 126–127,
note 293).
The purpose of this paper is to present a variety of Śaiva parallels in the
Kriyāsaṃgrahapañjikā, especially textual parallels between the Kriyāsaṃgra-
womb, and have the form of Hevajra. The second passage quoted from Vajrāvalī teaches how
the master should visualise the initiand in the udakābhiṣeka. The master should visualise that
the initiand is made to enter the womb of the goddess, becomes liquid (śūnyatā), and is emit-
ted outside the womb. This is a modification of the meditation in the utpattikrama practice.
The third passage quoted from Vajrāvalī teaches how the officiant should visualise the deity
of the image in the udakābhiṣeka of the pratiṣṭhā. The same method as the second passage
is applied here. For the relationship between the utpattikrama practice and the pratiṣṭhā in
Tantric Buddhism, see Tanemura 2004, 85–90.
2 For the parallel repertoire of rituals between Śaivism and Tantric Buddhism, see Sanderson
2009, 124–127. To add some more information about the manuals of the Tantric Buddhist
funeral rite (given on p. 126, note 294), after the publication of Sanderson 2009, I published a
critical edition of and notes on Śūnyasamādhivajra’s Mṛtasugatiniyojana (Tanemura 2013a),
an annotated Japanese translation of the same text (Tanemura 2013b), and a preliminary
edition and annotated Japanese translation of the Antasthitikarmoddeśa of Padmaśrīmitra’s
Maṇḍalopāyikā (Tanemura 2012). The visualisation taught in verses 12–13 of the Mṛtasuga-
tiniyojana is a modification of the mṛtasaṃjīvana practice elaborated in commentaries on
the Guhyasamājatantra, which are classified as works of Jñānapāda school in the Tibetan
canon. The utkrānti (intentional death of a yogin) is applied to the visualisations taught in
verses 14–16.
3 Sanderson proposes that the fundamental reason for Śaivism’s success was “that it greatly
increased its appeal to royal patrons by extending and adapting its repertoire to contain a
body of rituals and theory that legitimated, empowered, or promoted key elements of the
social, political and economic process that characterizes the early medieval period (Sander-
son 2009, 253).” With regard to the second element, the proliferation of land-owing tem-
ples, “[t]he Śaivas of the Mantramārga provided specialized officiants and rituals to estab-
lish these Śivas [= Liṅgas], developing in the course of time a secondary body of scriptural
authorities, the Pratiṣṭhātantras, devoted exclusively to this domain, setting out the rituals
of installation (pratiṣṭhā) and defining the norms for the form of the Liṅga, the iconogra-
phy of ancillary images, and the architectural design of the various temple types” (Sanderson
2009, 274; the word in square brackets is supplied by the present author). The characteris-
tics of the Kriyāsaṃgrahapañjikā are very close to those of the Pratiṣṭhātantras mentioned
above. With regard to the contents of the whole Kriyāsaṃgrahapañjikā, see Tanemura 2004,
12–42.
via free access
the kriyāsaṃgrahapañjikā of kuladatta 509
hapañjikā and the Devyāmata, a Śaiva Pratiṣṭhātantra, as materials to consider

concerning the relations between Śaivism and Tantric Buddhism.
2 Rituals in the Public Domain
First, I would like to present parallels at the scriptural level between Śaivism
and Tantric Buddhism. Just as in Śaivism, where the Saiddhāntika religion,
which non-Saiddhāntikas considered to be a fundamental but exoteric and
lower Śaiva teaching, is involved in the rituals in the public domain,4 so too the
ritual system of the consecration ceremonies prescribed in the Kriyāsaṃgra-
hapañjikā is based on the Yogatantras (more precisely, the Vajradhātumaṇḍala
system prescribed in the Sarvatathāgatatattvasaṃgraha, the principal scrip-
ture of the Yogatantra class), which was considered to be the Vajrayāna’s fun-
damental authority by the “higher,” more esoteric tantras, i.e. Yogottara- and
Yoginītantras.5
4 For this non-Saiddhāntika view on the Saiddhāntika religion, see, e.g., Sanderson 2007, 231.
On the public character of the Saiddhāntika religion, see, e.g., Sanderson 2007, 238–239 and
Sanderson 2014, 13.
5 For instance, the Sūtaka(-melāpaka) (commonly known as “Caryāmelāpakapradīpa,” which
is not attested in Sanskrit primary sources) calls the Sarvatathāgatatattvasaṃgraha the “root
tantra.” See Sūtakamelāpaka (chapter 3 Vāgviveka): vajragurur āha—sādhu sādhu mahāsatt-
va śrītattvasaṃgrahādau mūlatantre cottaratantre ca vāyutattvaṃ na vispaṣṭenoktam, saṃ-
dhyāyabhāṣitatvāt (EW p. 375, lines 13–15); vajragurur āha—sādhu sādhu mahāsattva mantra-
tattvaṃ nāma tattvasaṃgrahādau mūlatantre cottaratantre ca kevalaṃ mantramātram udī-
ritaṃ mantroddhāro na pradarśitaḥ (EW p. 378, lines 13–15). That the system of the Yogatantra
is the Vajrayāna’s fundamental authority might also be implied by the following verse of
the Saṃvarodayatantra (21.2): sāmānyayogatantrānāṃ rahasyaṃ na vipañcitam | siddhīnāṃ
paramā siddhir vratnānāṃ paramaṃ vratam || (ET p. 134).
That the system of the Vajradhātumaṇḍala taught in the Sarvatathāgatatattvasaṃgraha
is employed for rituals in the public domain is inferred from the fact that the fundamental
system of the Kriyāsaṃgrahapañjikā is the Vajradhātumaṇḍala (See the citation from the
Kalaśādhivāsanā section of the Kriyāsaṃgrahapañjikā below).
The verses quoted below from Padmaśrīmitra’s Maṇḍalopāyikā suggest that the rituals
prescribed in his manual, which are performed in public, are based on the Sarvatathā-
gatatattvasaṃgraha. See Padmaśrīmitra’s Maṇḍalopāyikā 2.39–41: ādāv arghavidhiḥ proktaḥ
dvitīye bhūparigrahaḥ | tṛtīyaṃ *ṭippa(em.; tippa- Ms.)sūtraṃ tu jñānasūtraṃ caturthakam
||2.39|| pañcamo rajasāṃ pātaḥ ṣaṣṭhaṃ kalaśādhivāsanam | saptamaḥ kalaśanyāso maṇ-
ḍalasādhanam aṣṭamam ||2.40|| pratiṣṭhā navamī caiva daśamī homakriyā matā | ekādaśī
visṛṣṭiḥ syād ity uktaṃ tattvasaṃgrahe ||2.41|| (2r7–8). The eleven rituals from the arghavidhi
to the visṛṣṭi mentioned in the above-quoted verses are rituals related to the pratiṣṭhā (cf.
the structure of the rituals prescribed in the Kriyāsaṃgrahapañjikā, for which see Tanemura
2004, Introduction). The phrase “taught in the (Sarvatathāgata-)Tattvasaṃgraha” does not
via free access
510 tanemura
The Kriyāsaṃgrahapañjikā as a whole is a kind of construction manual

for monasteries, and the author Kuladatta teaches details of various kinds of
rituals within this framework.6 In chapter 2, having examined the site for a
monastery and removed from the site extraneous substances that cause vari-
ous calamities, the officiant (ācārya) should visualise the site, which has been
divided into eighty-one compartments, as the Vajradhātumaṇḍala. He should
visualise Vairocana in the centre and the other deities in the rest of the com-
partments. In chapter 3, the officiant should prepare water jugs which are to be
used in the ritual. These water jugs represent the deities of the Vajradhātumaṇ-
ḍala, and he should draw the symbols of the deities on them. In the pratiṣṭhā of
images, although images (or rather the deities of images) go through the brah-
manical life cycle rites, they are sprinkled with water from the water jugs which
represent the deities of the Vajradhātumaṇḍala.
The fact that the system of the Vajradhātumaṇḍala is fundamental is implied
by the following remarks of Kuladatta.7
(1) Kriyāsaṃgrahapañjikā, Kalaśādhivāsanā in chapter 3:

yasyācāryasya vajradhātau nādhimokṣas tasya sveṣṭadevatādhimokṣeṇa
bhūśodhanapratiṣṭhāparyanteṣu sarvakriyākaraṇam aviruddham.8
Tanemura 2004, 135
necessarily mean that the scripture prescribes the eleven rituals listed in the verses, probably
only that the mantra-visualisation system employed in those rituals is that of the Vajradhātu-
maṇḍala taught in the Sarvatathāgatatattvasaṃgraha. (Cf. Kimiaki Tanaka’s understanding
of the words from ādau to visṛṣṭiḥ syāt as being directly quoted by Padmaśrīmitra from the
scripture; see Tanaka 2010, 562.) For instance, the samaya which the officiant causes the deity
of the images to listen to is “oṃ hūṃ trāḥ hrīḥ aḥ,” the syllables of the Five Buddhas of the
Vajradhātumaṇḍala (9r9).
6 See Tanemura 2004, Introduction § 1.5.
7 After the publication of Tanemura 2004, I was able to access the following two manuscripts
of the Kriyāsaṃgrahapañjikā. N2: Kaiser Library Access No. 109; N3: NGMPP E365/12 (private
collection). I report on the readings of these manuscripts when I quote the texts from Tane-
mura 2004. With regard to N2, the first forty-five leaves, of which the material is paper, are
a later addition to the original palm-leaf manuscript. The leaves before folio 35 of the origi-
nal palm-leaf manuscript have been lost. The original manuscript was copied in samvat 336.
According to Petech, the date of copying is verified for Thursday, February 11th, 1216 (Petech
1984, 81). With regard to N2, I report only on the readings of the original palm-leaf manuscript.
8 Mss. N2 and N3 read as follows. N2: yasyācāryyasya vajradhātau nāmādhimokṣaḥ tasya sveṣ-
ṭadevatādhi*mokṣeṇa(N2pc; mokṣauṇa N2ac) bhūśodhanapratiṣṭhāparyyanteṣu sarvve kriyā-
karaṇam aviruddhaḥ || (30v1–2); N3: yasyācāryasya vajradhātau nādhimokṣaḥ tasya sveṣṭade-
vatādhimokṣeṇa bhūśodhanapratiṣṭhāparyanteṣu sarvvakriyākaraṇam aviruddhaṃ || (24v1–
2).
via free access
If an ācārya does not have a strong conviction in the Vajradhātu, there is

no obstacle to his doing all the rites from purification of the site to conse-
cration [of images etc.] with a strong conviction in his own chosen deity.
The fact that the pratiṣṭhā prescribed in the Kriyāsaṃgrahapañjikā is a public

social ritual is implied in some parts in the text.
(2) Kriyāsaṃgrahapañjikā, chapter 4:

tato bhūpatiḥ sthapatikarmakaraprekṣakalokān yathārhaṃ kaṅkaṇāṅgu-
līyakavastrahiraṇyasrakpūgatāmbulādibhiḥ samyak paritoṣayet.9
● bhūpatiḥ ] em.; bhūpati° N N2 O C1 C2 C3 T1 T2 T3; bhūpatiṃ N3 ] ● °karmakara° ] N

N2 O C1 C2 C3; °karmakarau N3 °karmakaraḥ T1 T2; karmakare T3
Then the king should satisfy the architects, the assistants, and the specta-
tors with a bracelet, a finger-ring, a garment, gold, heap of chaplet, tām-
būla, or other [articles] according to [the donor’s] wealth.
(3) Kriyāsaṃgrahapañjikā, Pratiṣṭhā in chapter 6:

prekṣakajanāṃś ca tāmbūlādibhiḥ saṃtoṣya śreyase bhojanaṃ baliś ca
deyaḥ.10
Tanemura 2004, 162
[The ācārya should] also entertain spectators with tāmbūla etc. [In addi-
tion,] food and a bali should be offered for [their] good fortune.
(4) Cf. Vajrāvalī (Vihāragandhakūṭīcaityāvasathāśramārghavidhi)

pūgasrakcandanaiḥ prekṣakalokān sampūjya
EM § 1.1.6, vol. 1, p. 58; A 4r1; B 3v4
In the above-quoted passages from the Kriyāsaṃgrahapañjikā, it is envisaged

that the ritual is performed in the presence of spectators. The same character-
istic of the ritual is also found in the above-quoted passage of the Vajrāvalī. It is
also envisaged from the first quotation that the sponsor of the ritual is a king. I
shall come back to this point later.
9 The edition is based on the following witnesses (for sigla, see the bibliography): N 37v6,
N2 39v5, N3 44v1–2, C1 33r5, C2 51v5, C3 43r5, T1 43v4, T2 40v2–3, T3 37v5.
10 Mss. N2 and N3 read as follows. N2: prekṣakajanāṃś ca tāṃvūlādibhiḥ saṃtoṣya śreyase
bhojanam valiś ca deyaḥ (150v3); N3: prekṣakajanāś ca tāmvūlādibhiḥ santoṣya śreyase bho-
janaṃ valiś ca deyaḥ (174r3–4).
via free access
512 tanemura
From the passage quoted below, we see that people are not only passive spec-
tators but also active performers of the ritual.
(5) Kriyāsaṃgrahapañjikā, Vanayātrā in chapter 5:

tato nagarapraveśasamaye vṛkṣāṇāṃ śilānāṃ vā rājani paurajaneṣu vā
vārttāṃ vidhāya madanasphūrtimūrtibhiḥ puṣṭacittair janair vāhayet.11
● °samaye ] N N3 K O C1 C2 T1 T2; °samaya° N2; °sa++ C3; °samayai T3 ● śilānāṃ ] N N2
N3 K O C1 T1 T2 T3; śilānā C2; [śilānā] C3 ● paurajaneṣu vā ] N N3 C1 T2; paurujaneṣu vā

N2; porajaneṣu vā K; porajaneduvā O; poruṣajaneṣu vā C2; paurajaneṣu C3; purajaneyuvā
T1; porajaneṣu cā T3 ● vārttāṃ ] K C2 C3 T2 T3; vārttā N N3 C1; vattāṃ O; vartta T1 ●
vidhāya ] N N3 K O C1 C2 C3 T1 T2 T3; vidhātāṃ N2 ● madana° ] em.; madanamada° N
N2 N3 K O C1 C2 C3 T1 T2 T3 ● °mūrtibhiḥ ] N N3 K O C1 C2 C3 T1 T3; mūrttibhi N2 ●
puṣṭacittair ] N2 K O C2 C3 T2 T3; puṣṭacittai N; puṣṭicittair N3; puṣṭacirtte C1; puṣṭicitter
T1 ● janair ] N N3 K O C1 C3 T1 T2 T3; janai N2 C2; janer
When the wood [to be used for the construction of a monastery] or the
stones [to be used for the construction of a caitya] are brought into the
city, [the ācārya] should send a message [that these materials are being
brought into the city] to the king or the citizens. He should make peo-
ple with joyful minds whose bodies quiver with excitement carry [these
materials].
3 Royal Patronage
The above-quoted passage (5) too implies not only that the pratiṣṭhā prescribed
is a public social ritual, but also that the king might be envisaged as a donor.
Royal patronage of the ritual is also implied by the following passage in the
nimittokti section of chapter 3:
śiraḥkaṇḍūyamānaṃ yady ācāryaśilpiyajamānatanniyogijanānāṃ ma-

dhye kaścit karoti tadaikapauruṣād adhaḥ śalyam asti.12
Tanemura 2004, 150
If someone, either the tantric officiant, a craftsman [involved in the rite],

the donor or his officials scratches his head [in the site for a monastery
11 N 41v1, N2 43v4–5, N3 49r1–2, K 45r1–2, O 38v5–6, C1 36v2–3, C2 56r3–4, C3 47v2–3, T1

47v4–5, T2 44v1–6, T3 41v4–5.
12 For the reading of Ms. N2 see footnote 28 in this paper.
via free access
etc.], then there is an extraneous thing [that causes a calamity at a depth

of] the full height of a man underground.
While an ordinary donor might be present on his own, the king would never be
seen without his retinue of officials.
As already mentioned in footnote 3 in this paper, Sanderson proposes that
the fundamental reason for Śaivism’s success was “that it greatly increased its
appeal to royal patrons by extending and adapting its repertoire to contain a
body of rituals and theory that legitimated, empowered, or promoted key ele-
ments of the social, political and economic process that characterizes the early
medieval period (Sanderson 2009, 253).” With regard to the second element, he
states as follows:
The second element of the early medieval process to which I have drawn
attention is the proliferation of land-owning temples. All but the most
ephemeral sovereigns during this period, both in the subcontinent and
in Southeast Asia, gave material form to the legitimacy and solidity of
their power by building grand temples in which images of their chosen
God were installed, animated, named after themselves (svanāmnā), and
endowed with land and officiants to support their cult. As we have seen,
the great majority of these temples enshrined Śiva [in the form of a Liṅga].
Sanderson 2009, 274
The first line after the opening verses of the Kriyāsaṃgrahapañjikā might
reflect similar activities done by royal patrons for tantric Buddhism.
Kriyāsaṃgrahapañjikā, Ācāryaparīkṣā in chapter 1:

tatra vihārādikam abhidhātukāmo yajamāna ādāv ācāryaṃ parīkṣayet.13
●tatra ] C1 C2 T2; tatrā K T1; tato C3 T3 ● vihārādikam ] K C1 C2 C3 T1 T3; vihārādikrama

T2 ● abhidhātukāmo ] C1 C2 T3; vidhātukāmo K T2; katukāmo C3; vidhātukāme T1 ●
yajamāna ] K C1 C2 T1 T2 T3; yajamāne C3 ● ādāv ] K C1 C2 T1 T3; n.e. C3 T2 ● ācāryaṃ ]
K C1 C2 C3 T2 T3; ācārya T1 ● parīkṣayet ] K C1 C2 C3 T1 T2; parīkṣa[ye]+ T3
In this case (tatra), a donor who wants to name a monastery and other
[thing for religious purpose after himself] should, first of all, choose an
[appropriate] officiant.
13 N folio missing, N3 folio missing, O folio missing, K 1v2–3, C1 1v2–3, C2 1v3, C3 1v3–4, T1 1v3,
T2 1v3, T3 1v2–3.
via free access
514 tanemura
The manuscripts are divided into two groups with regard to the reading of
the third word. Mss. C1 C2 T3 read abhidhātukāmo, which is employed in the
above quotation, and Mss. K T2 vidhātukāmo. (The reading of T1 is a corrup-
tion of vidhātukāmo, and that of C3 is a corruption of kartukāmo, which is a
synonym of vidhātukāmo.14) I suspect that the author, Kuladatta, envisages a
king as a donor of a monastery, which should be named after the king. The
original reading of the third word is therefore abhidhātukāmo. The prescrip-
tions in the Kriyāsaṃgrahapañjikā are applied to other rituals related to the
construction of a temple. In some cases, objects of pratiṣṭhā are not named
after the donor. This might have changed the reading abhidhātukāmo to vi-
dhātukāmo. Alternatively, it might be the case that the custom that monas-
teries, caityas, and other religious objects are named after the donor had died
out or was dying out in Kathmandu in Kuladatta’s time. Sanderson points
out that the Licchavis of Nepal supported Buddhism (Sanderson 2009, 74–77).
According to the Gopālarājavaṃśāvalī, the earliest local chronicle, the follow-
ing monasteries and caitya were named after the donor: the Mānavihāra by
Mānadeva, the Dharmadevacaitya by Dharmadeva, and the Devalavihāra by
Devaladeva (Sanderson 2009, 74). The first one is confirmed by its mention in
an undated inscription assigned to his reign (Sanderson 2009, 75). Several of
the monasteries of the Kathmadu valley are attributed to kings of the period
of the Ṭhākurī kings—most probably Kuladatta flourished in this period—in
inscriptions, palm-leaf deeds, manuscript colophons, or their own tradition.
But no monastery or caitya named after a king is reported (Sanderson 2009, 77–
80). I am not able to draw a firm conclusion, but there might be factors which
changed the reading abhidhātukāmo to vidhātukāmo.15
4 Textual Parallels
Next, I would like to present textual parallels between the Kriyāsaṃgraha-

pañjikā and Śaiva Pratiṣṭhātantras. The section which contains these Śaiva
parallels is called the nimittokti. In chapter 3, the ācārya should divide the
14 This kind of “unfaithful copy” is found in various places of Ms. C. See Tanemura 2004, 102.
15 I do not conclude that the prescriptions of the Kriyāsaṃgrahapañjikā are merely ideal.
Rather, as I have already pointed out, the Kriyāsaṃgrahapañjikā is practical in its char-
acter (Tanemura 2004, 104–111). For instance, the measurements of a monastery by calcu-
lation based on the prescriptions of the vāstunāga are very close to those of the plans of
Cha Bahi and I Baha Bahi in the Kathmandu valley. See Tanemura 2002, 572, note 29. For
the plans of Cha Bahi and I Baha Bahi, see the plates attached to Watanabe et al. 2009.
via free access
site into thirty-six compartments. He should drive ritual spikes (kīla) sym-
bolising the thirty-two wrathful deities into the compartments, excluding the
four central ones, and worship the spikes. Then he should visualise himself
as Vajrahūṃkāra in order to remove obstacles from the site. Then the ācārya
should re-arrange the placement of the spikes in a proper way. After that, the
ācārya should connect the pañcasūtras—the cords of Brahman, the root cords,
the direction cords, and the side cords—to the spikes driven to the ground
(sūtrapātana). The nimittokti explains various kinds of good and bad omens
during the sūtrapātana. I have already shown parallels in the Piṅgalāmata, a
Śaiva Pratiṣṭhātantra, the Īśānaśivagurudevapaddhati, and some other sources
(Tanemura 2004, 148–155), and I have found yet other parallels in the Devyā-
mata, another Śaiva Pratiṣṭhātantra.
4.1 The Nimittokti of the Kriyāsaṃgrahapañjikā and the

Śalyoddhārapaṭala of the Devyāmata
First I shall quote the relevant part of the Kriyāsaṃgrahapañjikā (abbreviated
as KSP) from Tanemura 2004, 148–155. The corresponding verse numbers of the
Devyāmata (DM) are indicated at the end of each section. For the convenience
of readers, the corresponding section number of the Kriyāsaṃgrahapañjikā are
also indicated at the end of each section of the Devyāmata, both in the edition
and in the translation.
(1) The Nimittokti of the Kriyāsaṃgrahapañjikā16 (Tanemura 2004, 148–155)
[0] animittair asiddhiḥ syāt sūtracchede guroḥ kṣayaḥ17

ity vacanān nimittāny upalakṣayet. liṅgāni sūtracchedanarodanasūtrasa-
mullaṅghanagātrasparśananāmasaṃkīrtanādīni.18 (= DM vv. 9–10)
[1] tatra sūtracchedanenācāryasya maraṇam.19 (= DM v. 7ab)
[2] śvaśṛgālagṛdhrakaṅkarutair yajamānasya maraṇam āhuḥ.20
[3] (1) mārjāreṇa sūtralaṅghane tadasthi rāsabhasya vā. (= DM v. 18) (2) gar-
dabhena laṅghane tadasthi. (= DM v. 23b) (3) kukkureṇa laṅghane tadasthi. (=
16 The readings of N3 are presented in the footnotes at the end of each section.
17 Guhyasamājamaṇḍalavidhi v. 229ab (Ms. 12r2, ES 134). The numeration follows that of the
Sarnath edition (ES).
18 For §[0] N3 reads as follows: animittair asiddhiḥ syāt sūtraccheda guroḥ kṣaya iti vacanāt
nimittāny upalakṣayet || liṅgāni sūtracchedanarodanasūtrasamullaṅghanagātrasparśana-
nāmasaṅkīrttanādīni || (33v1–2).
19 For §[1] Ms. N3 reads as follows: tatra sūtracchedanenācāryasya maraṇaṃ | (33v2).
20 For §[2] Ms. N3 reads as follows: śvaśṛgālagṛdhrakaṅkarunair yajamānasya maraṇam
āhuḥ | (33v2).
via free access
516 tanemura
DM v. 19ab) (4) ajāvibhyāṃ laṅghane tayor asthi gor asthi vā. (= DM v. 20a) (5)
aśvenāśvāsthi. (= DM v. 20cd) (6) hastinā hastyasthy uṣṭrāsthi vā. (= DM v. 22c)
(7) aśvatareṇa tadasthi. (8) mahiṣeṇa śṛgālāsthi. (= DM v. 21ab) (9) mṛgena
mṛgāsthi. (10) ṛkṣena ṛkṣāsthi. (11) varāheṇa vyāghrāsthi. (= DM v. 22a) (12)
vyāghreṇa gajāsthi. (= DM v. 22b) (13) mūṣakeṇa mūṣakāsthi. (= DM v. 19cd) (14)
sarpeṇa sarpāsthi. (15) kacchapena kacchapāsthi.21
[4] (1) śiraḥkaṇḍūyamānaṃ yady ācāryaśilpiyajamānatanniyogijanānāṃ
madhye kaścit karoti tadaikapauruṣād adhaḥ śalyam asti. (= DM vv. 63c–64b)
(2) bhrūsparśe suvarṇaṃ hastatrayāt kācaṃ vā.22 (3) netrasparśe netraparyan-
tādhastān muktā.23 (4) mukhasparśe keśaṃ kāṣṭhaṃ vā trikarādhare. (= DM
vv. 64c–65b) (5) dantasparśāt tribhir hastair dantam anumīyate. (= DM v. 65cd)
(6) karṇakaṇḍūyane karṇāntādhastād rūpyaṃ suvarṇaṃ vidrumaṃ vā bha-
vet.24 (7) galasparśena tatpramāṇādhaḥ kaṇṭhikā lohaśṛṅkhalā vā, mārjārakaṅ-
kālaṃ vā trikarādhare. (= DM vv. 66c–67b) (8) aṃsasparśe tatpramāṇādhare
tadābharaṇam. (9) kakṣasparśe kakṣāntādho loham.25 (10) bāhupīḍane kaṇ-
ṭhapramāṇādhare tadābharaṇam. (= DM vv. 69c–70b) (11) dakṣiṇakarasparśe
kaṭimātrādhaḥ †pṛṣṭha†kapālaṃ mṛnmayakapālaṃ vā. (= DM vv. 71c–72b) (12)
vāmahastasparśe jānumātrādhaḥ kaṭvāpādaḥ. (= DM v. 70cd) (13) pārśvakaṇ-
ḍūyane narārdhamātrādho dhūlī. (= DM vv. 74c–75b) (14) uraḥsparśe kaṭimā-
trādhaḥ paśukīkasam. (= DM vv. 72c–73b) (15) pṛṣṭhasparśe pṛṣṭhāsthi tatpra-
māṇādhaḥ. (= DM vv. 74ab) (16) kaṭisparśe dvikarādhaḥ pradeśe lohakaṇṭakam.
(= DM vv. 75c–76b) (17) liṅgasparśena hastapramāṇādhare trilohaśalyam.26
(18) jaṃghāsparśe tadasthy ekādaśāṅgulādhare. (= DM vv. 78c–79b) (19) gul-
21 For §[3] Ms. N3 reads as follows: mārjjāreṇa sūtralaṃghane tadasthi rāsasya vā || gard-
dareṇa laṃghane tadasthi || kujjureṇa laṃghane tadasthi || ajāvibhyāṃ laṃghane tayor
asthi vā || aśvenāśvāsthi || hastinā hastyasthi || uṣṭrāsthi vā | aśvatareṇa tadasthi mahiṣeṇa
śṛgālāsthi || mṛgeṇa mṛgāsthi || rekṣeṇa ṛkṣāsthi | varāheṇa vyāghrāsthi || vyāghreṇa gajā-
sthi || mūṣakeṇa mūṣakāsthi | sarppeṇa sarppāsthi || kacchapena kacchapāsthi || (33v2–5).
22 Though the Devyāmata does not have a parallel to this teaching, Piṅgalāmata has a close
parallel: bhruvoḥ saṃsparśanād bhadre kācaśalyaṃ trihastakam (68r3). See Tanemura
2004, 150.
23 Though the Devyāmata does not have a parallel to this teaching, the Piṅgalāmata has a
close parallel: īśa*sthe (conj. Sanderson; sthaś Ms.) cakṣuḥsaṃsparśāt tanmānān mauk-
tikaṃ bhavet (68r1). See Tanemura 2004, 150.
close parallel: śruti*sthe (conj. Sanderson; sthaḥ Ms.) śrutisaṃsparśāt pravālaṃ vātha kāñ-
canam | rajataṃ ca śubhā hy ete karṇamātrāt samuddharet | (68r1). See Tanemura 2004,
150.
25 Though the Devyāmata does not have a parallel to this teaching, tthe Piṅgalāmata has
a close parallel: kakṣau kakṣākṛtiṃ vindyāt kṛṣṇalohaṃ karatrayāt (68r3). See Tanemura
2004, 151.
via free access
phasparśe ’ṣṭādaśāṅgulādhare ’śvakhuraḥ.27 (20) pādasparśād dvādaśāṅgulā-

dhare śālmalī kaṇṭako vā. (= DM vv. 79c–80b) (21) pādakaniṣṭhāṅgulīsparśe
’ṣṭāṅgulādhare kāṃsyam. (= DM vv. 82c–83b) (22) pārṣṇisparśe dvādaśāṅgulād-
hare ’bhrakam.28
[5] sūtrapātanasamaye yajamānasya pārśve sthitvā kenacid anyena puru-
ṣeṇa yasya prāṇino nāma saṃkīrtyate tadasthi tatrāstīti niścayaḥ.29 (= DM
v. 12cd)
[6] (1) akasmād gaur āgatya yadi viṣṭhām utsṛjati tadadho ’vaśyaṃ tatpra-
māṇaṃ kanakam astīti niścīyate. (= DM v. 32ab) (2) yady akasmād āgatya
purīṣam utsṛjati bālakumārikā tadā tadadho ’vaśyaṃ tatpramāṇaṃ rūpyaṃ
bhavet. (3) bhekarutena jalabhayam. (4) śukaśarikāhaṃsakokilamayūrajīva-
ṃjīvakacakravākavṛṣabhāṇāṃ hṛdayopakūjanaṃ kalyāṇāya bhavati. (= DM
v. 34abc) (5) siṃhagajameghamanojñasvano dhanadhānyārthalābhodayāya
bhavati. (6) śaṅkhamaṅgalagītikābālakrīḍanair arthāptiḥ. (7) dhūmadarśane
cittapīḍā. (8) hīnadīnavyādhiparipīḍitajanadarśane rogaḥ. (9) dhvajacchatra-
patākāmadyamāṃsaghaṇṭālaṅkārāmbhojadadhīndravahnijvālāphalamīnayu-
garājāṅganadīnāṃ saṃdarśane śubhaṃ bhavati. (10) vidvadbrāhmaṇabhikṣu-
sādhujanānāṃ saṃdarśane dharmaḥ syāt.30
iti nimittoktiḥ.
close parallel: śiśne tu vikṛtiṃ yāte trilohaṃ tatra jāyate | trikarādhaḥ samuddhāryam iti
śāstrasya niścayaḥ | (68r3–4). See Tanemura 2004, 152.
close parallel: gulphasthe gulphasaṃsparśād dhayapādaṃ vinirdiśet | daśāṣṭāṅgulamā-
nena tatra *khātvā (em. Sanderson; khaṭvā Ms.) samuddharet | (67v4) See Tanemura 2004,
153.
28 For §[4] N3 reads as follows: śiraḥkaṇḍūyanaṃ yady ācāryaśilpiyajamānatanniyogijanā-
nām madhye kaścit karoti || tadā ekapauruṣād adhaḥ śalyam asti || bhrūsparśe suvarṇ-
ṇaṃ hastatrayāt kācam vā || netrasparśe netraparyāntādhastān muktā || mukhasparśe
keśaṃ kāṣṭham vā *trikādhare (N3ac; trikarādhare N3pc) || dantasparśāt tribhir hastair dan-
tam anumīyet || karṇṇakaṇḍūyane karṇṇāntādhastād rūpyaṃ || suvarṇṇaṃ vidrumam vā
bhavet || galasparśena tataḥ pramāṇādhaḥ kaṇṭhikā || lohaśṛṅkhalā vā aṅśākaṃkālam
vā trikarādhare || aśasparśe tatpramāṇādhare tadābharaṇaṃ || kakṣasparśe kakṣāntādho
lohaṃ || dakṣiṇakarasparśe kaṭimātrādhaḥ pṛṣṭhakapālaṃ mṛnmayakapālam vā | vāma-
hastasparśe jānumātrādhaḥ khaṭvāpādaḥ || pārśvakaṇḍūyane navārddhamātrādho dhūlī
|| uraḥsparśe kaṭimātrādhaḥ paṇḍuḥ kīkaśaṃ || pṛXṣṭhasparśe pṛṣṭhāsti tatpramāṇādhaḥ
|| kaṭisparśe dvikarādhaḥ pradeśe lohakaṇṭhakaṃ || liṅgasparśena hastapramāṇādhare
trilohaśalyaṃ || jaṃghāsparśe tada*X(N3ac; tyo N3pc)kādaśāṅgulādhare gulphasparśe
’ṣṭādaśāṅgulādhare ’śvakuraḥ || pādasparśāXt dvādaśāṅgulre ’dhare śālmalī kaṇṭhako vā
|| pādakaniṣṭhāṅgulīsparśe ’ṣṭāṅgulādhare kāṃśyaṃ | pārṣṇisparśe dvādaśāṅgulādhare
’bhrakaṃ || (33v5–34v1).
29 For §[5] N3 reads as follows: sūtrapātanasamaye yajamānasya pārśve sthitvā kenacid
anye[na pu]ruṣeṇa⊔ yasya prāṇino nāma sa[kīttyata tada]sthi tatrā[stī]ti niścayaḥ || (34v1).
30 For §[6] Ms. N3 reads as follows: akasmād gaur āgatya yadi viṣṭhām utsṛjati tadadho
via free access
518 tanemura
(2) The Śalyoddhārapaṭala of the Devyāmata (excerpted)

(Ms. A 91r5–93v5, Ms. B 50r2–54v5)31
Preliminary Edition
ataḥ paraṃ pravakṣyāmi yathoktaṃ śalyalakṣaṇam |

sattvānām apakārāya vāstumadhye vyavasthitam ||1||
1b yathoktaṃ ] A; yathoktaX B 1c sattvānām ] B; satvānāṃ m A 1d vāstumadhye ] A;

vāstumadhya B
caturasrīkṛte kṣetre sūtrite śakunādibhiḥ |

vāstudehavibhāgajño vāstuśalyaṃ nirūpayet ||2||
gṛhaprāsādayor vidvān ārambhe sūtrakarmaṇi |
lakṣayec chakunaṃ samyag nimittaṃ copalakṣayet ||3||
darśanaṃ kīrtanaṃ śabdaṃ yajamānasya ceṣṭitam |
vastudehe yathāvasthaṃ lakṣaye śalyam ādarāt ||4||
4a darśanaṃ ] A; daśanaṃ B 4c lakṣaye śalyam ] A; lakṣayec chalyam B
pāṣaṇḍidarśane neṣṭaṃ gṛhiṇām asukhāvaham |

hataṃ kṣataṃ mṛtaṃ bhagnaṃ śrutvā na sūtrayed gṛham ||5||
5a pāṣaṇḍidarśane ] em.; pāṣaṇḍidarśanaṃ A; pāṣaṇḍidaśane B 5a neṣṭaṃ ] A; neṣṭhaṃ

B
†aśastān† ye ’pi ye ’śastā neṣṭā sattvāś ca garhitāḥ |

darśanaṃ kīrtanaṃ śabdaṃ †sattvās† teṣāṃ vivarjayet ||6||
6a aśastān ye ’pi ] B; aśastāmy api A 6d vivarjayet ] B; vivajayet A
vaśyaṃ tatpramāṇaṅ kanakam astīti niścīyate | yady akasmād āgatya [pu?]rīṣam utsṛ-
jati vālakumārikā tadā tadadho ’vaśyaṃ tatpramāṇa rūpyam bhavet || bhekarutena jala-
bhayaṃ || sukasīrikāhaṃsakokilamayūjīvaṃjīvakacakravākavṛṣabhāṇāṃ hṛdyopakūja-
naṃ kalyāṇāya bhavati || siṃhagajameghamanojñasvano dhanadhānyārthalābhodayāya
bhavati || śaṅkhamaṅgalagītikāvālakrīḍaner arthāptiḥ || dhūmadarśane cittapīḍā || hīnadī-
navyādhiparipīḍitajanadarśane rogaḥ || dhvajacchatrapatākāsadyamāṃsaghaṇṭālaṅkā-
rāmbhojadadhīndravahnijvālāphalamīnayugarājāṅganādīnāṃ sandarśane śubham bha-
vati || vidvadbrāhmaṇabhikṣusādhujanānāṃ sandarśane dharmmaḥ syāt || (34v1–5).
31 The preliminary edition of the Devyāmata is based upon the two manuscripts listed below
under References. There is another incomplete palm-leaf manuscript of the same scrip-
ture (NAK 5–446/vi. śaivatantra 105, catalogued under the title Niśvāsākhyamahātantra =
NGMPP A41/13), which does not, unfortunately, contain the text of the relevant chapter.
via free access
sūtracchedena maraṇaṃ duḥkhaṃ vā maraṇāntikam | (v. 7ab = KSP [1])

evaṃ jñātvā vidhānajñaḥ śāntihomaṃ tu kārayet ||7||
sarvasukhāvahaṃ yasmāt samaṃ śriyānvitaṃ gṛham |
tasmāt susūtritaṃ kṛtvā śalyaṃ veśmani lakṣayet ||8||
8c susūtritaṃ ] A; svasūtritaṃ B 8d śalyaṃ ] B; śalya A
sūtrasya laṅghanād vāpi darśanān nāmakīrtanāt |

śabdasaṃśrāvaṇād vāpi lakṣaye śalyam ādarāt ||9||
9b nāmakīrtanāt ] em.; nāmakīrtanā A; nāmakīrttināt B 9d lakṣaye śalyam ] A; lakṣayec

chalyam B
laṅghanaṃ darśanaṃ yasya ruditaṃ nāmakīrtitam |

tasya sattvasya tac chalyam ādiśel laṅghanādibhiḥ ||10|| (vv. 9–10 = KSP
[0])
10a darśanaṃ ] A; darśanaṃ [ve] B
anyasya prāṇino ’py aṅgaṃ vidyād anyasya laṅghanāt |

gṛhiṇo ’ṅgavikāreṇa vāstunaḥ śalyam ādiśet ||11||
11c gṛhiṇo ’ṅgavikāreṇa ] B; gṛhiṇodbhavikāreṇa A
dṛśyate śakuno vāpi sa yasya śrūyate svanaḥ |

nāmasaṃkīrtanaṃ yasya tasya śalyaṃ vinirdiśet ||12|| (v. 12cd = KSP [5])
12c nāmasaṃkīrtanaṃ ] em.; nāmasaṃkīrtta[na] A; nāmasaṃkīrttate B
…
sūtre prasāryamāṇe tu mārjāro yadi laṅghanam |
rāsabhāsthi vijānīyā tadaṅge vāstuno hy adhaḥ ||18|| (v. 18 = KSP [3](1))
18b laṅghanam ] A; laṃghayete B (hypermetrical) 18c rāsabhāsthi ] A; rāsabhāsthim B

18d tadaṅge ] A; tadaṅgo B
yadi śvā laṅghate sūtraṃ tasmiṃ śvānāsthim ādiśet | (v. 19ab = KSP
[3](3))
mūṣikālaṅghanenaiva ajāvikāsthim ādiśet ||19|| (v. 19cd = KSP [3](13))
19a laṅghate ] A Bpc; laṃghayete Bac 19b śvānāsthim ] em.; śvāno sthim A; śvāno ’sthim B
19c mūṣikālaṅghanenaiva ] A; mūsakālaṃghanenaiva B
via free access
520 tanemura
ajāvikāś ca gosthi syād gāva-m-aśvāsthim ādiśet | (v. 20a = KSP [3](4))

aśvasya laṅghanenaiva māhiṣaṃ śalyam ādiśet ||20|| (v. 20cd = KSP
[3](5))
20b aśvāsthi ] A; Bpc; āśvāsthi Bac B 20d māhiṣaṃ śalyam ādiśet ] A; māhiśet B (haplo-
graphical error)
mahiṣalaṅghanenaiva asthi syāj jambukasya tu | (v. 21ab = KSP [3](8))

jambukalaṅghanenaiva śūkarāsthi samādiśet ||21||
śūkarākramaṇe vyāghraṃ vyāghreṇaiva tu kuñjaram | (v. 22a = KSP
[3](11), v. 22b = KSP [3](12))
kuñjarākramaṇe hy uṣṭram †aṅgāraso†ṣṭralaṅghanāt ||22|| (v. 22c = KSP
[3](6))
22a vyāghraṃ ] A; n.e. B (caused by eye-skip) 22c uṣṭram ] A; aṣṭam B
nṛlaṅghanān narāsthi syāt kharāsthi kharalaṅghaṇāt | (v. 23b = KSP

[3](2))
evaṃ saṃlakṣayec chalyaṃ laṅghanādarśanādibhiḥ ||23||
23a nṛlaṅghanān ] A; nṛlaṃghanā B 23a narāsthi ] A Bpc; nṛrāsthi Bac 23c saṃlakṣayec ]
A; śalakṣayec B
…
gavāṃ mūtreṇa rūpyaṃ syāt purīṣeṇaiva kāñcanam | (v. 32ab = KSP
[6](1))
lohaṃ mārjāramūtreṇa purīṣenāśam ādiśet ||32||
32a gavāṃ ] A; gavā B 32a rūpyaṃ ] conj.; ruśmaṃ A; rugmaṃ B 32c lohaṃ ] em.; loha
A B 32c mārjāramūtreṇa ] B.; mārjāmūtreṇa A 32d purīṣenāśam ] A; purīṣenāgam B
…
śāntāyāṃ diśi śakuno madhuraṃ ravate yadi |
arthaṃ tatra vijānīyād … ||34|| (v. 34abc = KSP [5](4))
34a śāntāyāṃ ] A; śāntāyā B
…
śiraḥkaṇḍūyamāne tu śirasi śalyam uddharet ||63||
63c śiraḥkaṇḍūyamāne ] A; śirakaṇḍuyamā*ṇde(?) B 63d śirasi ] B; ++[sa] A
via free access
asthiśalyaṃ tu taṃ jñeyaṃ puruṣārdhena tiṣṭhati | (vv. 63c–64b = KSP

[4](1))
mukhe kaṇḍūyamānena kāṣṭhajaṃ vā śirobhavaḥ ||64||
64b tiṣṭhati ] A; tiṣṭati B 64c kaṇḍūyamānena ] A; kaṇḍuyamānena B 64d kāṣṭhajaṃ ]

B; kāṣṭajaṃ A
adhasthād dhastadvayenaiva tiṣṭhate nātra saṃśayaḥ | (vv. 64c–65b =

KSP [4](4))
hanujaṃ dantasaṃsparśād uddhare tatpramāṇataḥ ||65|| (v. 65cd = KSP
[4](5))
65a adhasthād ] A; adha B 65b tiṣṭhate ] em.; tiṣṭate A B 65b saṃśayaḥ ] A; saśayaḥ B
65c dantasaṃsparśād ] B; dattasaṃspārśād Aac; dattasaṃsparśād Apc 65d uddhare ] em.;
urddhare A; dhaddhare B
yadaṅgaṃ spṛśate hy arthaṃ tadaṅge śalyam ādiśet |

yadi kaṇḍūyate grīvā śṛṅkhalālohajaṃ viduḥ ||66||
66a spṛśate hy ] A; spṛtete ty B
hastatrayeṇa sā jñeyā śṛṅkhalā nātra saṃśayaḥ | (vv. 66c–67b = KSP

[4](7))
aṅge kaṇdūyāmāne tu aṅgajaṃ śalyam ādiśet ||67||
67b nātra ] A; śātra B
adhastāt tatpramāṇe tu uddharec chalyam ādarāt |

sphaṇake spṛśyamāne tu sphaṇajaṃ śalyam ādiśet ||68||
68b uddharec chalyam ādarāt ] A; śiraśi śalya sudarāt B 68cd Sphaṇa(ka) is a corruption
of skandha?
tiṣṭhate tatpramāṇe tu samyag jñātvā samuddharet |

bāhukaṇḍūyamāne tu bāhuke kaṭakaḥ sthitaḥ ||69||
69b jñātvā ] A; kṛtvā B 69c bāhu° ] B; vāhū° A 69d bāhuke kaṭakaḥ ] conj.; vāhuje
nakalaḥ A; vāhyaje nalakaḥ B
hastatrayeṇa sārdhena tiṣṭhate nātra saṃśayaḥ | (vv. 69c–70b = KSP

[4](10))
via free access
522 tanemura
haste kaṇḍūyamāne tu khaṭvāpādaṃ vinirdiśet ||70|| (v. 70cd = KSP

[4](12))
70a sārdhena ] A; sāddhena B 70b tiṣṭhate ] A; tiṣṭate B
jānumātre sthitaṃ vidyāt karajāṅgulisparśanāt |

haste kaṇḍūyamāne vā kaṭimātre sthitaṃ viduḥ ||71||
71a vidyāt ] B; vindyān A 71c haste kaṇḍūyamāne vā ] A; hastena kaṇḍūyamāne *vā(?) B

(hypermetrical)
śalyaṃ śamuddhared vidvān kapālaṃ vātha mṛnmayam | (vv. 71c–72b =

KSP [4](11))
uraḥkaṇḍūyamāne tu paśuśalyam athāṅgajam ||72||
72a śalyaṃ ] A; śalya B 72c uraḥkaṇḍūyamāne ] em.; uraḥkaṇḍūyamānena A; urakaṇ-

ḍūyamāne B
sārdhahastadvaye ’dhastāc chalyaṃ yatnāt samuddharet | (vv. 72c–73b =

KSP [4](14))
hṛdaye hṛdayasparśāt tatpramāṇena tanmayam ||73||
73a sārdhahastadvaye ] A; sārddhahastādvayo B 73b yatnāt ] em.; yannān A; yannā B 73c

hṛdayasparśāt ] em.; hṛdayaṃ sparśāt A B
pṛṣṭhajaṃ pṛṣṭhasaṃsparśād udare tatpramāṇataḥ | (v. 74ab = KSP

[4](15))
pārśve saṃsparśanād vidyāc chalyaṃ pāṃsulikodbhavam ||74||
74a pṛṣṭhajaṃ ] B; spṛṣṭhajaṃ A 74a pṛṣṭhasaṃsparśād ] em.; spṛṣṭhasaṃsparśād A B

74d chalyaṃ ] A; chalyāṃ B
tatpramāṇe sthitaṃ śalyam uddharec chalyavittamaḥ | (vv. 74c–75b =

KSP [4](13))
kaṭijaṃ kaṭisaṃsparśād athavā lohakaṇṭakam ||75||
hastadvayapramāṇe tu śalyaṃ tatra samuddharet | (vv. 75c–76b = KSP
[4](16))
ūrukaṇḍūyamāne tu ūrujaṃ vātha dārujam ||76||
76b śalyaṃ ] A; śalya B 76c ūrukaṇḍūyamāne ] A; urukaṇḍūyamāṇe B 76d ūrujaṃ ] A;

urujaṃ B
via free access
sārdhahastapramāṇena śalyaṃ yatnāt samuddharet |

jānukaṇḍūyato dṛṣṭvā sthāṇujaṃ vātha jānujam ||77||
77a sārdhahastapramāṇena ] A; sāddhastapramāṇena B 77b yatnāt ] A; yannāt B 77c

jānukaṇḍūyato ] A; jānukaṇḍūto B 77d sthāṇujaṃ ] corr.; sthānujaṃ A B
†nādhitopaskarasvāpi† hastamātre samuddharet |

yadā saṃspṛśyate jaṅghāṃ jaṅghāsthiṃ tatra nirdiśet ||78||
78a nādhitopaskarasvāpi ] A; nāditopaskaramvā B (The readings of both A and B are

suspected but undiagnosed corruption.)
ekādaśāṅgule ’dhastāt tiṣṭhate nātra saṃśayaḥ | (vv. 78c–79b = KSP

[4](18))
pāde kāṇḍūyamāṇe tu kauñjaraṃ śalyam ādiśet ||79||
79a ekādaśāṅgule ’dhastāt ] A; ekādaśāṅgulodhastā B 79b tiṣṭhate ] A; tiṣṭate B 79d

kauñjaraṃ ] B; kaujeraṃ A
dvādaśāṅgulamāne tu śalyaṃ kaṇṭakam uddharet | (vv. 79c–80b = KSP

[4](20))
aṅguṣṭake yadā kaṇḍū khaṭikāśalyam ādiśet ||80||
80b śalyaṃ ] A; śalya B
rītikācitrasaṃmiśraṃ lohaṃ vā tatra nirdiśet |

aṅgulyāṃ yadi kaṇḍūya aśvapādaṃ vinirdiśet ||81||
81a rītikācitrasaṃmiśraṃ ] B; rītikācitrasanmriśraṃ A 81b lohaṃ ] A; loha B 81d

aśvapādaṃ ] B; aṣṭapādaṃ A 81d vinirdiśet ] A; vinidiśet B
sārdhavitastimātreṇa tiṣṭhate nātra saṃśayaḥ |

kaniṣṭhikāyāṃ kaṇḍūya kāṃsyaṃ tatra vinirdiśet ||82||
82a sārdha° ] em.; sārddhaṃ A; sāddha° B 82b tiṣṭhate ] A; tiṣṭate B 82c kaniṣṭhikāyāṃ
] A; kaniṣṭhikāyā B 82d kāṃsyaṃ ] B; kāṃsaṃsyavit A
tac cāṣṭāṅgulamāne tu tiṣṭhate nātra saṃśayaḥ | (vv. 82c–83b = KSP

[4](21))
adhopādasya kaṇḍūya carma śalyaṃ samādiśet ||83||
83a tac ] A; ta[ṃ] B 83d carma ] A; cama B
via free access
524 tanemura
aṣṭāṅgulapramāṇe tu tiṣṭhate śalyakaṇṭakam |

…
iti śalyoddhārapaṭalaḥ.
Provisional translation
Next, I shall, as told before, teach the characteristics of extraneous substances,
which exist beneath the site (vāstumadhye) and cause calamities to people.
(1)
When the site, which has been made square, is being divided with cords, [the
officiant] who has knowledge of divisions of the site (vāstudehavibhāga-
jñaḥ) should investigate extraneous substances by omens, etc. (2)
When the division of [the sites for] a house and a shrine with cords has been
commenced, the wise man [i.e. the officiant] should notice an omen and
observe it correctly. (3)
[The omens are] seeing [someone or something], announcing [a creature’s
name], cries [of animals], and the actions of a donor. [The officiant] should
carefully notice an extraneous substance as situated beneath the site. (4)
If a heretic is seen, that brings an undesirable outcome to householders. If one
hears someone hurt, wounded, or killed, or something broken, then [the offi-
ciant] should not divide the site with cords. (5)
If there are persons who are not praised, undesirable, or blameworthy, then
one should avoid seeing such persons, hearing [the names of] such persons
announced, and hearing the voices of such persons. (6)
If a cord is cut, there is death or deadly pain.32 (= KSP [1]) (7ab)
[The officiant] who has knowledge of the ritual should perform the fire rite for
quelling of calamities, if he becomes aware of such [omens]. (7cd)
Since a levelled house brings every comfort and prosperity [to the residents],
one should divide the site properly with cords and examine extraneous sub-
stances beneath the site (veśmani). (8)
[The officiant] should carefully prognosticate the extraneous thing [under-
ground] by observing [a creature] step over a cord, seeing [an auspicious
or inauspicious thing], announcing a [creature’s] name, or hearing [an aus-
picious or inauspicious sound]. If [a creature] steps over [a cord] or is seen,
or if one [hears] a cry of [a creature] or announce a [creature’s] name, then
[the officiant] should prognosticate the extraneous thing [related to] that
creature according to the stepping over and other [omens]. (= KSP [0]) (9–
10)
32 The second outcome, deadly pain, is not mentioned in the Kriyāsaṃgrahapañjikā.
via free access
If a creature [intrudes into the site] stepping over [a cord], then [the officiant]
should know that there is the body [of that creature, i.e. bones of that crea-
ture beneath the site]. He should prognosticate an extraneous substance
beneath the site (vāstunaḥ) by the bad condition of the householder’s body.
(11)
If an omen is seen, or if [a creature] cries out, or if [someone] announces a
[creature’s] name, then [the officiant] should prognosticate an extraneous
thing [related to] that [creature]. (= KSP [5]) (12)
…
If a cat [intrudes into the site] stepping over [a cord] while a cord is being cast,
it should be understood that there is the bone of an ass beneath that spot of
the site.33 (= KSP [3](1)) (18)
If a dog steps over a cord, [the officiant] should prognosticate the bone of a dog
[beneath] the [spot of the site]. (= KSP [3](3)) (19ab)
If a mouse passes [over a cord], [the officiant] should prognosticate bones of
goats and sheep [beneath the site].34 (= KSP [3](13)) (19cd)
If rams or sheep [step over a cord], there is the bone of a cow [beneath the
site].35 (= KSP [3](4)) (20a)
If cows [step over a cord], [the officiant] should prognosticate bones of a horse
[beneath the site]. (20b)
If a horse steps over [a cord], [the officiant] should prognosticate an extraneous
thing related to a buffalo[, i.e. the bone of a buffalo beneath the ground].36
(= KSP [3](5)) (20cd)
If a buffalo steps over [a cord], there is the bone of a jackal [beneath the site].
(= KSP [3](8)) (21ab)
If a jackal steps over [a cord], [the officiant] should prognosticate the bone of
a boar [beneath the site]. (21cd)
If a hog steps over [a cord], there is [the bone of] a tiger [beneath the site]. (=
KSP [3](11)) (22a)
33 Although the Devyāmata does not mention the bone of a cat as an extraneous thing, it
should also be prognosticated if we consider v. 10 of the Devyāmata above.
34 The Kriyāsaṃgrahapañjikā mentions the bone of a mouse, which is not mentioned in the
Devyāmata, as the extraneous thing in the case that a mouse passes over a cord. If we con-
sider v. 10 of the Devyāmata above, the bone of mouse should also be prognosticated in
this case.
35 The Kriyāsaṃgrahapañjikā mentions bones of rams and sheep, which are not mentioned
in the Devyāmata as extraneous things. Probably, in this case too, the rule of v. 10 above
should be applied.
36 If the rule of v. 10 is applied, the bone of a horse should also be prognosticated in this case.
The bone of a buffalo is not mentioned in the corresponding part of the Kriyāsaṃgraha-
pañjikā.
via free access
526 tanemura
If a tiger [steps over a cord], there is [the bone of] an elephant [beneath the
site]. (= KSP [3](12)) (22b)
If an elephant steps over [a cord], [there is the bone of] a camel [beneath the
site].37 (= KSP [3](6)) (22c)
If an camel steps over [a cord], there is †aṅgārasa† [beneath the site]. (22d)
If a man steps over [a cord], there is a human bone [beneath the site]. (23a)
If an ass [intrudes into the site] stepping over [a cord], there is the bone of an
ass [beneath the site]. (= KSP [3](2)) (23b)
In this way, [the officiant] should examine extraneous substances by [the
omens] such as stepping over and seeing. (23cd)
…
If a cow [which has entered the site] urinates or drops dung, there are pieces
of silver or gold [beneath the site, respectively].38 (= KSP [6](1)) (32ab)
If a cat urinates or drops dung, [the officiant] should prognosticate a piece
of iron or an inauspicious thing (? aśam)39 [beneath the site,] respectively.
(32cd)
…
If a bird sings sweetly in an auspicious direction, then [the officiant] should
prognosticate a treasure there. (v. 34abc = KSP [5](4)) (34a–c)
…
If [someone] scratches his head, [the officiant] should remove an extraneous
thing at a depth of the full height of a man (śirasi). On the other hand, it
should be understood that the extraneous thing which is a bone exists [at
a depth of] a half [of the height] of a man [underground].40 (= KSP [4](1))
(63c–64b)
If [someone] touches his mouth (or face), there must be [an extraneous thing]
which is a piece of wood or hair (? śirobhavaḥ) [at a depth of] two cubits
underground.41 (= KSP [4](4)) (64c–65b)
If [someone] touches his teeth, there is [an extraneous thing] which is a tooth
(hanujam). [The officiant] should remove [it from a depth of] that measure-
ment [= up to the teeth].42 (= KSP [4](5)) (65cd)
37 The Devyāmata does not mention the bone of an elephant, which is mentioned in the
Kriyāsaṃgrahapañjikā. Probably, in this case too, the rule of v. 10 should be applied.
38 The Kriyāsaṃgrahapañjikā does not mention the former omen, i.e. the urination of a cow.
39 There might be a corruption here. It is expected that the extraneous thing is a certain kind
of metal in this case.
40 The Kriyāsaṃgrahapañjikā teaches only that there is an extraneous thing at a depth of
the full height of a man in this case.
41 The Kriyāsaṃgrahapañjikā teaches that the depth is three cubits in this case.
42 The Kriyāsaṃgrahapañjikā teaches that the depth is three cubits in this case.
via free access
If [someone] touches a part of his body, [the officiant] should prognosticate an

extraneous thing (arthaṃ … śalyam) [at a depth] up to the part. (66ab)
If [someone] touches his neck, they know that there is [an extraneous thing]
which is an iron chain. It should be understood that that iron chain [exists
at a depth of] three cubits [underground]. There is no doubt about it.43 (=
KSP [4](7)) (66c–67b)
If [someone] scratches a part of his body, [the officiant] should prognosticate
an extraneous thing related to the part (aṅgajam) at a depth up to the part
(adhastāt tatpramāṇe). [The officiant] should remove the extraneous thing
carefully. (67c–68b)
If [someone] scratches his shoulder (?), [the officiant] should prognosticate an
extraneous thing related to the shoulder (?), which is at a depth up to the
[shoulder (?)]. If he knows it correctly, he should remove it.44 (68c–69b)
If someone scratches his arm, there is armlet [at the depth] up to the arm.45
[That extraneous thing] exists [at a depth of] three and a half cubits [under-
ground].46 There is no doubt about it. (= KSP [4](10)) (69c–70b)
If [someone] touches his [left?] hand, [the officiant] should prognosticate the
leg of a couch [beneath the site]. If [someone] touches his finger (karajāṅ-
guli°), [the officiant] should know [that the extraneous thing] is situated at
a depth up to the knee.47 (= KSP [4](12)) (70c–71b)
Alternatively, if [someone] scratches his [right?] hand, it is understood that
there is an extraneous thing, i.e. a skull or [a bowl] made of clay [at a depth]
just up to the buttocks [underground]. The wise man [i.e. officiant] should
remove it.48 (= KSP [4](11)) (71c–72b)
If [someone] scratches his breast, there is the bone of an animal (paśuśalyam)
or hair (aṅgajam)49 [at a depth of] one and a half cubits underground.50
43 The Kriyāsaṃgrahapañjikā additionally mentions a necklace and skeleton of a cat as

extraneous things in the case that someone touches his neck.
44 I am not sure what sphaṇa or sphaṇaka means. The preceding verse refers to the neck and
the following the arm, so this word probably means shoulder.
45 If we refer to the parallel in the Kriyāsaṃgrahapañjikā, the extraneous thing to be prog-
nosticated is an ornament related to the arm.
46 The Kriyāsaṃgrahapañjikā teaches that the depth is the measurement up to the neck.
47 The Kriyāsaṃgrahapañjikā limits the first condition to the left hand, but does not men-
tion the second condition, i.e. touching a finger.
48 The Kriyāsaṃgrahapañjikā limits the condition to the right hand. The relevant part of the
Devyāmata does not have a word corresponding to †pṛṣṭḥakapālam† in the Kriyāsaṃ-
grahapañjikā, i.e. the Devyāmata does not give a clue to solve the textual problem in the
Kriyāsaṃgrahapañjikā.
49 The Kriyāsaṃgrahapañjikā does not mention this second extraneous thing.
50 The Kriyāsaṃgrahapañjikā teaches that the depth is the measurement up to the buttocks
in this case.
via free access
528 tanemura
[The officiant] should remove that extraneous thing carefully. (= KSP [4](14))
(72c–73b)
If [someone] touches his heart, there is [an extraneous thing] related to it (tat-
mayam) at a depth up to the heart (hṛdaye … pramāṇena). (73cd)
If [someone] touches his back, there is [an extraneous thing] arising from the
back[, i.e. a back-bone at the depth up to the back]. If [someone touches]
his belly, [there is an extraneous thing related to the belly] at the depth up
to the [belly]. (= KSP [4](15)) (74ab)
If [someone] touches his side, one should prognosticate that there is an extra-
neous thing arising from dust. The best knower of extraneous things [= the
officiant] should remove that extraneous thing which exists [at a depth of]
that measurement [= up to the side] [underground]. (= KSP [4](13)) (74c–
75b)
If [someone] touches his buttocks, there is [an extraneous thing] arising from
the buttocks[, i.e. coccyx?] or an iron nail at a depth of two cubits [under-
ground]. [The officiant] should remove that extraneous thing from there. (=
KSP [4](16)) (75c–76b)
If [someone] scratches his thigh, there is an extraneous thing related to the
thigh or piece of wood at a depth of one and a half cubits. [The officiant]
should remove it carefully. (76c–77b)
If [someone] is seen to scratch his knee, there is an extraneous thing, i.e. a
stump (sthāṇujam) or a knee bone ( jānujam) at a depth of one cubit. [The
officiant] should remove it.51 (77c–78b)
If [someone] touches his shank, [the officiant] should prognosticate a bone of
the shank [at a depth of] eleven digits underground in that place. There is
no doubt about this. (= KSP [4](18)) (78c–79b)
If [someone] scratches his foot, [the officiant] should prognosticate an extrane-
ous thing related to an elephant[, i.e. a born of an elephant (kauñjaraṃ)].52
He should remove the extraneous thing, i.e. a thorn [at a depth of] twelve
digits [underground]. (= KSP [4](20)) (79c–80b)
If [someone] scratches his big toe, [the officiant] should prognosticate an extra-
neous thing, i.e. a piece of chalk. Alternatively, he should prognosticate a
piece of iron mixed with various calxes of brass there. (80c–81b)
51 I have not translated the corruption, nādhitopaskarasvāpi.

52 An extraneous thing prescribed in the corresponding part of the Kriyāsaṃgrahapañjikā
is a piece of silk-cotton wood (śālmalī).
via free access
If [someone] scratches his toe, [the officiant] should prognosticate a foot of a

horse [beneath the site].53 It exists at a depth of one and a half vitastis. There
is no doubt regarding this. (81c–82b)
If [someone] scratches his little toe, [the officiant] should prognosticate a piece
of bell-metal [beneath] the spot. That [extraneous thing] exists [at a depth
of] eight digits [underground]. There is no doubt about it. (= KSP [4](21))
(82c–83b)
If [someone] scratches his sole, [the officiant] should prognosticate an animal’s
hide as the extraneous thing. There is the extraneous thing at a depth of eight
digits. (83c–84b)
…
These are the rules for the removal of extraneous substances.
5 Concluding Remarks
Whereas Śaivism produced Pratiṣṭhātantras, scriptures which specialise in

temple construction and installation, Buddhism did not produce a scripture
in this domain.54 The author of the Kriyāsaṃgrahapañjikā was aware that
there is no scriptural authority at least with regard to the nimittokti section.
Thus he states that one should consider various omens based on the half
stanza from Dīpaṅkarabhadra’s Guhyasamājamaṇḍalavidhi. He also employs
the word liṅga as a synonym of nimitta and uses this word not only for an aus-
picious or inauspicious sign or omen, but also as a term from logic (inferential
sign). His intention is probably to state that the word nimitta in the half stanza
quoted from the Guhyasamājamaṇḍalavidhi should be understood as a liṅga
and, therefore, things caused by those liṅgas are correctly inferred (anumīyate)
based on the liṅgaliṅgisaṃbandha. However, this might not necessarily mean
that Kuladatta thinks his teachings are non-Buddhist. For in ritual manuals
53 A foot of a horse (aśvapādaṃ) is supported by Ms. B and Ms. A reads aṣṭapādaṃ (a spi-
der). Since the omen is scratching the toe, an extraneous thing related to the foot might
be better.
54 The Śaivas of Mantramārga produced a secondary body of scriptural authorities, the
Pratiṣṭhātantras, devoted exclusively to the domain of construction of royal temples. They
also asserted the principle that the Śaiva sthāpaka, the specialist who performs the rit-
uals related to temple construction and installation, is competent not only for the Śaiva
domain but also for all the levels that Śaivas ranked below this (Sanderson 2009, 274–275).
Cf. Devyāmata: pāṣaṇḍidarśanaṃ neṣṭaṅ gṛhiṇām asukhāvahaṃ (Ms. A 91v1); Kriyāsaṃ-
grahapañjikā, nimittokti: vidvadbrāhmaṇabhikṣusādhujanānāṃ saṃdarśane dharmaḥ
syāt (Tanemura 2004, 155).
via free access
530 tanemura
it is important that Buddhist mantra-visualisation systems are employed.55 In

the case of the Kriyāsaṃgrahapañjikā, the mantra-visualisation systems of the
Yogatantra and the “higher” tantras are employed, and in this sense the rituals
prescribed function as Buddhist. This syncretism of different classes of scrip-
tures is common to both Śaivism and Tantric Buddhism.
The nimittokti and other sections (e.g. bhūmiparīkṣā) in which close tex-
tual parallels to Śaiva Pratiṣṭhātantras and Paddhatis are found have little to
do with Śaiva and Buddhist doctrines and those sections have close parallels
to jyotiḥśāstras and śilpaśāstras.56 One might suppose that this is evidence
of the fact that both Śaivism and Buddhism used this kind of literature as
common sources and established their own respective ritual systems. This is
probably not the case. Although the non-sectarian parallels are common to
the Śaiva Pratiṣṭhātantras or Paddhatis and the Buddhist ritual manuals, and
these two religions employ different mantra-visualisation systems, the struc-
tures or styles of the rituals prescribed are similar to each other. For example,
while the Kriyāsaṃgrahapañjikā teaches that in the preparation of water jugs
the officiant should make the assistants recite Mahāyāna sūtras, the Somaśam-
bhupaddhati prescribes a preparation ritual for consecration of the Śivaliṅga in
which mantras of the four Vedas are recited in the four directions (Tanemura
2004, 235, note 50). Considering the parallels on the scriptural level mentioned
above, the parallel repertoire of rituals prescribed in Śaiva ritual texts and the
Kriyāsaṃgrahapañjikā, and the similarities of structures or styles of the rituals,
it is not implausible that one religion, probably Buddhism, followed examples
of the other. It is, of course, important to consider various parallels of the kinds
presented in this paper in greater detail in order to understand the relationship
between Śaivism and Tantric Buddhism.
55 The sūtrapātanavidhi, in which the nimittokti section is included, is based on the mantra-
visualisation system of the Vajradhātumaṇḍala taught in the Sarvatathāgatatattvasaṃ-
graha. See Tanemura 2004, 139–155, 237–250.
56 For example, descriptions similar to those of the nimittokti of the Kriyāsaṃgrahapañjikā
are found in the Bṛhatsaṃihtā 53.105–110 (EB vol. 1, pp. 489–491). See also Tanemura 2004,
245–250.
via free access
Appendix: A Provisional Edition and Translation of the

Bhūśalyasūtrapātananimittavidhi of the Ācāryakriyāsamuccaya
After I had finished writing a draft of this article, I found a small section of
Jagaddarpaṇa’s Ācāryakriyāsamuccaya which teaches about extraneous things
beneath the site for a maṇḍala or monastery, and the omens for those extrane-
ous things. Here I present a preliminary edition and provisional translation of
the relevant section. This is written in verse, and the metre is śārdūlavikrīḍitā.
Most probably the material is silently quoted from the work of a predeces-
sor; the section colophon of Ms. K states that the Ācāryakriyāsamuccaya is a
compilation of various teachings (for this colophon, see the apparatus of this
preliminary edition).
via free access
532 tanemura
sūtraṃ tatra nipātayet suvidhinā tullīkṛte bhūtale

pañcajñānamayaṃ viśuddhavalitaṃ nirgranthi nābhyūrdhvagam |
sthairyadhyānasamanvito varagurur nāsāgramadhye ’kṣiṇī
ghaṇṭāmaṅgalagītaśaṅkhapaṭhanaiḥ saṃstūya vaṃśasvanaiḥ ||1||
5 airandhrīkarapallavoddhṛtapayomiśrapraphullojjvalair
arghaṃ ratnasuvarṇagandhakusumaiḥ saṃdāpayed dāyakaḥ |
naimittaṃ ca nirūpayed gurumṛtiḥ sūtrasya saṃcchedane
nūnaṃ jambukagṛdhrakaṅkaruditair mṛtyur bhavet svāminaḥ ||2||
2c gurumṛtiḥ … saṃcchedane] = KSP [1] 2d = KSP [2]
10 yajjātīyaviśeṣasattvajanitaiḥ sūtraṃ samullaṅghyate

tajjātīyakam asthi tatra niyataṃ sūtrādhare vāstuni |
dātuḥ pārśvagato hi sūtravitate yannāma saṃkīrtayet
tannāmānugasattvakīkasamalaṃ dātṛsthabhūkhaṇḍale ||3||
3ab = KSP [3] 3cd = KSP [5]
15 svāṅgaṃ vā spṛśati drutaṃ vidhivaśāt tanmānam ākhanya tac

chalyānāṃ bahudhā nimittam uditaṃ samkṣepamātraṃ tv iha |
gaur āgatya tadā purīṣam asṛjat tanmānahema sthitaṃ
yadvā bālakumārikā ca visṛjet tanmānarūpyaṃ bhavet ||4||
4a = KSP [4] 4c = KSP [6](1) 4d = KSP [6](2)
2 viśuddhavalitaṃ] Kpc N1 N2 (The relevant word in Ms. K is corrected at least three times: viX →
vidha → visuvalitaṃ → visuddhavalitaṃ) ‖ nirgranthi] N1; nigranthi K; nitranthi N2 3 nāsāgra-
madhye] K N1; nāśāgramadhya N2 ‖ ’kṣiṇī] conj.; ’kṣaṇo K; kṣaṇo N1 N2 4 ° paṭhanaiḥ]
N1 N2; °pradhvanaiḥ K ‖ saṃstūya] em.(←Tib. bstod pas); santūryya K; santūrya N1 N2 5
airandhrīkara°] N1 N2; airandhrikara° K ‖ ° praphullojjvalair] N1 N2; °prasphullojjvalair K 6
dāyakaḥ] em.; dāyakaṃ K N1 N2 7 gurumṛtiḥ] N2; gurumṛti K 8 jambuka°] N2; jam-
būka° K ‖ ° ruditair] K; °rudite N2 ‖ mṛtyur] K; mṛtyu N2 10 ° sattva°] N2; °satya° 12
dātuḥ] Kpc N2; pādātuḥ Kac ‖ pārśvagato] K N2pc; śvagato N2ac 15 ākhanya] N2; ākhana K
16 chalyānāṃ] K; chalyānāṃ m N2 ‖ bahudhā] Kpc N2; bahuvidhā Kac ‖ nimittam uditaṃ] K;
nimittavyditaṃ or nimittamaditaṃ N2 ‖ tv] N2; n.e. K 17 asṛjat] N2; asṛjan K 18 visṛjet] K;
visṛjat N2 ‖ tanmānarūpyaṃ] N2; tanmānarūpa K ‖ bhavet] N2; bhave K
7 naimittaṃ] The folios which contain the text from nimittam to the end of this chapter are missing
from Ms. N1.
via free access
The excellent master [= officiant] in steady meditation, gazing upon the centre
of the tip of his nose, should cast the cord on the surface of the site which
has been levelled following the rules exactly. [The cord,] into which [the five
threads of the five colours] are twined, has as its nature the five wisdoms and
is purified. [It] does not have a knot, and is placed in the centre [of the site
before casting]. (1a–c)
Having praised [the cord] with the sounds of a bell, auspicious song, conch
shell, and bamboo flute, the donor should offer guest water [to the cord]
together with jewels, gold, and fragrant flowers, which are blooming and
beautiful, and mixed with the juice extracted from the sprouts of the aira-
ndhrīkara. (1d–2b)
[The officiant] should examine omens. If a cord is cut, the death of a master
[will take place]. (= KSP [1]) If the cries of a jackal, a vulture and a heron
[are heard], then the death of a lord [will] definitely [take place]. (= KSP [2])
(2cd)
If a cord is stepped over by a specific kind of creature, then there must be a
bone of that creature beneath the site (vāstuni) on which the cord is being
cast (sūtrādhāre). (= KSP [3]) (3ab)
If [some other man] who stands beside the donor announces a [creature’s]
name while a cord is being cast, then there is an impure substance, i.e. a
bone of the creature of the name beneath the site on which the donor is
standing. (= KSP [5]) (3cd)
If [someone] touches [a particular part of] his body and [the site] is quickly
dug to a depth up to that [particular part of the body] according to the rules,
then there is the [extraneous thing corresponding to the omen]. (= KSP [4])
(4a)
[With regard to bodily sensations,] various omens of extraneous things [be-
neath the site] are taught. In this [short section], however, [the explanation
is] just abridged. (4b)
If a cow comes and drops dung, then there is the same amount of gold as the
[dung beneath the site]. (= KSP [6](1)) Alternatively, if a young girl [comes
and] urinates, then there must be the same amount of silver as [the urine
beneath the site]. (= KSP [6](2)) (4cd)
via free access
534 tanemura
maṇḍūkadhvaninā gṛhe jalabhayaṃ dhūmena cittākulaṃ

rogārtāṅgavihīnakuṣṭhavivaśastrīdarśaṇe rogabhāk |
jīvaṃjīvamayūrakokilaśukāś cakrāṅkahaṃsarṣabhās
teṣāṃ kaṇṭharutaṃ śubhodayakaraṃ saṃpatkaraṃ darśaṇam ||5||
5 5a maṇḍūkadhvaninā … jalabhayaṃ ] = KSP [6](3) 5a dhūmena cittākulaṃ ] = KSP
[6](7) 5b = KSP [6](8) 5cd = KSP [6](4)
siṃhāmbhodhararājakuñjararavair dhānyārthalābhodayaṃ
bālakrīḍanaśaṅkhamaṅgaladhvanau dravyāgamas tadgṛhe |
chatrāmbhojapatākāmurajadhvajālaṅkārarājāṅgaṇā-
10 matsyakṣīradadhīndramadyadahanajvālāphalānāṃ jayam ||6||
6a = KSP [6](5) 6b = KSP [6](6)
dhānyadravyasutādivṛddhir atulā niṣpannakāryaṃ tadā

bhikṣubrāhmaṇadhīdhanottamajanaiḥ saṃdarśaṇe dharmabhāk |
prārambhe bhuvi sūtrapātanavidhau devādisaṃsthāpane
15 śreyoliṅgam idaṃ hitodayakaraṃ saṃvīkṣya kuryāt tadā ||7||
6c–7a = KSP [6](9) 7b = KSP [6](10)
śilpācāryavicāracārucaturais tyaktvāśubhaṃ sarvathā

yena sthānanivāsisarvajanatārājñāṃ ca dātur yathā |
kalyāṇāya śubhodayāya nitarām ādau vicāryaiva tat
20 sattārāgrahayogavāsaraśubhe kāryāṃ samārabhyatām ||8||
iti bhūśalyasūtrapātananimittavidhiḥ.
1 maṇḍūkadhvaninā] N2; maṇḍakadhvaninā K ‖ cittākulaṃ] em.; cintākulaṃ K; citākulaṃ N2

2 rogārtāṅga°] K; rogārttaṅga° N2 ‖ ° vihīna°] N2; °vihina° ‖ ° vivaśa°] em.; °vivaśaḥ K; °viva-
gaḥ N2 ‖ ° strīdarśaṇe] N2; °strīdarśaṇa K 3 cakrāṅka°] K; cakrāṃṅka° N2 ‖ ° rṣabhās] N2;
°gaṇāḥ K 7 ° ravair] em.; °ravai K N2 8 ° dhvanau] N2; °dhvanair K 9 chatrāmbhoja°] Kpc;
cchatrāmbhojakha° Kac; chatrāmbhojakhama N2 ‖ ° muraja°] K; °jera° N2 ‖ ° dhvajālaṅkāra°]
K; °dhvajaghaṭālaṅkāra° N2 10 ° dadhīndra°] N2; °dadhisu° K 12 dhānya°] K; dhānyārtha°
N2 ‖ niṣpanna°] K; niṣpan° N2 14 bhuvi] N2; guru K 15 śreyoliṅgam] K; śriyoliṅgam N2 ‖
saṃvīkṣya] N2; saṃvidya K 17 śilpācārya°] em.; śiṣyācāryya° K; śilppācārya° N2 ‖ ° caturais]
K; catures N2 18 yena] N2; ye K ‖ sthānanivāsi°] N2; sthānavāsināṃ K ‖ ° sarvajanatā°]
N2; sarvvajanānāṃ Kac; sarvvajanatānāṃ Kpc ‖ ca dātur] N2; pradābhūr K 19 tat] K; tata
N2 21 iti bhūśalyasūtrapātananimittavidhiḥ] em. based upon N2; nānāmatāt samākṛṣya maṇ-
ḍalācāryadarppaṇaviracite kriyāsamuccaye bhūśalyasūtrapātananimittavidhiḥ ṣaṣṭamapaṭalaḥ
K; iti bhūśalyasūtrapātananimitavidhi N2
8 ° maṅgala°] la of maṅgala must be short (dhv does not make the vowel heavy). 9 muraja] la
of muraja must be short (dhv does not make the vowel heavy).
via free access
If a frog croaks, there is danger of water in the [donor’s?] house. (= KSP [6](3))
If smoke [is seen], there is distraction of mind. (= KSP [6](7)) If a person
suffering from a disease (rogārtāṅga°), a person of a lower [class], a person
suffering from leprosy, a deranged person (°vivaśa°), and a woman are seen,
then it causes disease (rogabhāk). (= KSP [6](8)) (5ab)
Songs (kaṇṭharutam) of a jīvaṃjīvaka bird, peacock, kokila bird, parrot, ca-
krāṅka, haṃsa, and a bull bring auspiciousness. If [these creatures] are seen,
it brings prosperity. (= KSP [6](4)) (5cd)
The roar of a lion, the sound of thunder, and the roar of a royal elephant bring
the gain of grain and property. (= KSP [6](5)) (6a)
If the voices of children playing, the sound of a conch-shell, or an auspicious
[song are heard], it brings wealth to the [donor’s] house. (= KSP [6](6)) (6b)
If a parasol, lotus, banner, muraja drum, flagpole, ornament, a woman of the
court, fish, milk, the best curd, wine, blazing fire, and fruits [are seen], then
there are victory, extraordinary increase of grain, property, [the number of]
sons, and other [merits], and the completion of duties. (= KSP [6](9)) (6c–
7a)
If a bhikṣu, brāhmaṇa, wise man (°dhī°), or a wealthy man (°dhanottamajana-)
is seen, it brings virtue. (= KSP [6](10)) (7b)
In the consecration of [images of] deities and other [sacred objects], [the offi-
ciant] should examine the [above-mentioned] omens which bring merits
[to the donor] in the commencement of the rite of the casting of cords, and
then perform [the casting of cords]. (7cd)
The officiant with special knowledge of architecture57 who is skilled in the
examination [of omens] should abandon inauspicious[, extraneous] things
by all means. By doing this ( yena), fortune and auspiciousness will certainly
be brought to the donor, the king, and all people who live in the region.
[Therefore, the officiant] should first examine the [omens], and then under-
take the rite [to follow] when the combination of fixed stars and planets, and
the day are auspicious. (8)
These are the rules for extraneous things beneath the site and the omens
[observed] in the rite of casting of cords.
57 I have not seen the word śilpācārya elsewhere. If this is not a corruption, it probably refers
to a particular class of officiant which is equivalent to the sthāpaka, the Śaiva officiant who
specialises in the installation of images and consecration of temples.
via free access
536 tanemura
Abbreviations
ac before correction
conj. a diagnostic conjecture
corr. a correction
DM Devyāmata
em. an emendation
KSP Kriyāsaṃgrahapañjikā
NAK National Archives, Kathmandu
n.e. not existent
NGMPP Nepal-German Manuscript Preservation Project
pc after correction
X an illegible akṣara
+ an akṣara lost due to damage to the manuscript.
†…† suspected but undiagnosed corruption
References
Primary Sources and Sigla

Ācāryakriyāsamuccaya by Jagaddarpaṇa or Darpaṇācārya
(K) Manuscript preserved in the Kyoto University Library, No. 7.
(N1) Manuscript preserved in Kaiser Library, Kathmandu, Access No. 110.
(N2) Manuscript preserved in NAK, No. 4–123/vi. bauddhatantra 16; NGMPP B31/5.
Kriyāsaṃgrahapañjikā by Kuladatta
See Tanemura 2004.
For sigla N K O C1 C2 C3 T1 T2 T3, see Tanemura 2004, 103–103.
(N2) Manuscript preserved in Kaiser Library, Kathmandu, Access No. 109.
(N3) NGMPP E365/12. According to the card catalogue, N3 is a private manuscript of
B.H. Bajracharya. The short title given in the card catalogue is Pañjikā.
Guhyasamājamaṇḍalavidhi by Dīpaṅkarabhadra
(Ms.) Niedersächsishe Staats- und Universitätsbibliothek, Göttingen, Cod. Ms. Sanscr.
257, folios 6v4–16v.
(ES) “Śrīguhyasamājamaṇḍalavidhiḥ.”Dhīḥ: Journal of Rare Buddhist Texts Research Unit
42 (2006): 109–154.
via free access
Devyāmata
(A) NAK 1–279 vi. śaivatantra 104; NGMPP A41/15–42/1 (catalogue title: Niśvāsākhyama-
hātantra).
(B) NAK 1–1003; NGMPP B27/6 (catalogue title: Pratiṣṭhātantra).
Piṅgalāmata
NAK 3–376; NGMPP A42/2.
Bṛhatsaṃhitā by Varāhamihira
(EB) M. Ramakrishna Bhat, ed. Varāhamihira’s Bṛhatsaṃhitā. 2 vols. Reprint, Delhi:
Motilal Banarsidass, 1997 [1981].
Bhramaharanāma Hevajrasādhana (Bhramahara) by Ratnākaraśānti

See Isaacson 2002.
Maṇḍalopāyikā by Padmaśrīmitra
Ms. preserved in the Tokyo University Library, No. 280 (New Number).
Vajrāvalī by Abhayākaragupta
(EM) Masahide Mori, ed. Vajrāvalī of Abhayākaragupta: Edition of Sanskrit and Tibetan
Versions. 2 vols. Buddhica Britannica Series Continua XI. Tring: Institute of Buddhist
Studies, 2009.
(A) NAK 3–402 vi. bauddhatantra 76 = NGMPP A48/3.
(B) NAK 5–84 vi. bauddhatantra 78 = NGMPP B31/14.
Somaśambhupaddhati
For the pratiṣṭhā section see Brunner-Lachaux 1998.
Secondary Sources
Brunner-Lachaux, Hélène. 1998. Somaśaṃbhupaddhati: Rituels dans la tradition sivaïte
selon Somaśambhu, quatrième partie, rituels optionnels: pratiṣṭhā. Publications du
département d’indologie 25.4. Pondichéry: Institut Français de Pondichéry/École
Isaacson, Harunaga. 2002. “Ratnākaraśānti’s Bhramaharanāma Hevajrasādhana: Crit-
ical Edition (Studies in Ratnākaraśānti’s tantric works III).” Kokusai Bukkyōgaku
Daigakuin Daigaku Kiyō [Journal of the International College for Advanced Bud-
dhist Studies] 5: 80–55.
Petech, Luciano. 1984. Medieval History of Nepal (c. 750–1480). 2nd edition. Serie Ori-
etale Roma, no. 54. Roma: Istitute per il Medio e Estremo Oriente.
Sanderson, Alexis. 2007. “The Śaiva Exegesis of Kashmir” In Mélanges tantriques à
via free access
538 tanemura
la mémoire d’Hélène Brunner/Tantric Studies in Memory of Hélène Brunner, edited

by Dominic Goodall and André Padoux, 231–442, 551–582 (bibliography). Collec-
tion Indologie, no. 106. Pondicherry: Institut Français d’Indologie/École française
Sanderson, Alexis. 2009. “The Śaiva Age: The Rise and Dominace of Śaivism During the
Oriental Culture, University of Tokyo.57
Sanderson, Alexis. 2010. “Ritual for Oneself and Ritual for Others.” In Ritual Dynam-
ics and the Science of Ritual Volume II. Body, Performance, Agency, and Experience,
edited by Axel Michaels, et al., 9–20. Wiesbaden: Harrassowitz Verlag.
Sanderson, Alexis. 2014. “The Śaiva Literature.” Journal of Indological Studies (New title
for Studies in the History of Indian Thought) 24 and 25 (2012–2013): 1–113.
Tanaka, Kimiaki. 2010. Indo ni okeru Mandara no Seiritsu to Hatten [Genesis and Devel-
opment of the Maṇḍala in India]. Tokyo: Shunjūsha.
Tanemura, Ryugen. 2004. Kuladatta’s Kriyāsaṃgrahapañjikā: A Critical Edition and
Annotated Translation of Selected Sections. Groningen Oriental Studies, no. 19. Gro-
ningen: Egbert Forsten.
Tanemura, Ryugen. 2012. “Padmaśrīmitra Saku Maṇḍalopāyikā no Antasthitikarmod-
deśa: Preliminary Edition oyobi Shiyaku” [*Antasthitikarmoddeśa of Padmaśrīmi-
tra’s Maṇḍalopāyikā: A Preliminary Edition and Annotated Japanese Translation].
Gendai Mikkyō 23: 103–121.
Tanemura, Ryugen. 2013a. “Śūnyasamādhivajra Chosaku no Sōgi Manyuaru Mṛtasug-
atiniyojana: Sansukurittogo Kōtei Tekisuto oyobi Chū” [Śūnyasamādhivajra’s Mṛta-
sugatiniyojana: A Critical Edition and Notes]. Tōyō Bunka Kenkyūjo Kiyō [The Mem-
oirs of Institute for Advanced Studies on Asia] 163: 136–110.
Tanemura, Ryugen. 2013b. “Śūnyasamādhivajra Chosaku no Sōgi Manyuaru Mṛtasuga-
tiniyojana: Shiyaku oyobi Chū” [Śūnyasamādhivajra’s Mṛtasugatiniyojana: An Anno-
tated Japanese Translation]. Acta Tibetica et Buddhica 6: 21–60.
Watanabe, Katsuhiko et al. 1998. Nepāru no Bukkyō Sōin: I Baha Bahi Shūfuku Hōkokusho
[The Buddhist Monasteries of Nepal: a Report on the I Baha Bahi Restoration
Project]. Tokyo: Chūō Kōron Bijutsu Shuppan.
57 Genesis and Development of Tantrism is also published by Sankibo Press, Tokyo, under the
Japanese Title Tantora no Keisei to Tenkai.
via free access
chapter 22
Mañjuśrī as Ādibuddha: The Identity of an Eight-

armed Form of Mañjuśrī Found in Early Western
Himalayan Buddhist Art in the Light of Three
Nāmasaṃgīti-Related Texts
Anthony Tribe
In this article I suggest that an eight-armed Mañjuśrī found in early Western

Himalayan Art be identified not as the bodhisattva Mañjuśrī but as the Ādibud-
dha.1 This figure is distinctive in that it holds four swords, one in each of its four
right hands, and four book volumes, one in each of its four left hands. Previously
identified as a form of Dharmadhātuvāgīśvara Mañjuśrī,2 this image is found
in three locations in the Western Himalayas. Two of these are in Ladakh: one
at Alchi in the Sumtsek (Gsum-brtsegs, “Three-Storeyed”) Temple; the other
at Mangyu, where there are two images, the first in the Two-armed Maitreya
Chapel, the second in the Village Stūpa.3 These three images are murals. The
third location is in Spiti, where there is a clay sculpture in the Golden Temple
or Serkhang (Gser-khang) at Lalung.4
1 The expression “Ādibuddha,” which may be rendered in English by “Original Buddha,” de-
notes, in the present context, a figure seen as the embodiment of the gnosis ( jñāna) under-
lying the state of Buddhahood. It is not surprising, perhaps, that Mañjuśrī, as the bodhisattva
of wisdom par excellence, would be reconfigured to function additionally as the Ādibuddha.
This article, which, as will become evident, very much represents work in progress, has its
origins in research on Vilāsavajra’s Nāmasaṃgīti commentary, research that was supervised
some twenty-five years ago by Professor Sanderson. I undertake this foray into the field of
early Western Himalayan art with some trepidation: it is not an area in which I possess exper-
tise. I have tried not to go beyond the limits of what I feel confident in claiming; nonetheless,
there are bound to be errors, both of omission and commission. I offer advance apologies!
2 The Dharmadhātuvāgīśvara-maṇḍala as described by Abhayākaragupta has an eight-armed
form of Mañjuśrī, named Mañjughoṣa, as its central deity (Niṣpannayogāvalī—hereafter
NYĀ—54).
3 “Two-armed Maitreya Chapel” is the nomenclature of Luczanits 2004; van Ham 2011 uses
“Maitreya Tower I.” With “Village Stūpa,” however, I follow van Ham’s (ibid.) terminology.
Luczanits (ibid.) has “Four-image Chörten,” which is somewhat misleading, as while the build-
ing contains four clay images there are also mural images.
4 Regarding the dates of these temples, that of the Alchi Sumtsek is still a matter of contro-
© anthony tribe, 2020 | doi:10.1163/9789004432802_024

via free access
540 tribe
figure 22.1 Eight-armed Mañjuśrī, uppermost level, Sumtsek, Alchi

Photograph by Jaroslav Poncar ( JP93 26.4.05 WHAV)
The Sumtsek figure, perhaps the best known of the three, is the central deity
of a fifty-three deity maṇḍala on the top (i.e., the third) level of the temple
(Figs. 22.1–2).
Being on the highest level and also on the wall opposite the temple’s en-
trance, it occupies the place of greatest symbolic importance in the building.
versy. While the eleventh and thirteenth centuries both have their advocates (see Levy and
Fidler 2014), I am inclined to agree with Linrothe’s (2011) assessment of a mid-twelfth to early-
thirteenth century time frame. For the Two-armed Maitreya Chapel and the Village Stūpa at
Mangyu, Luczanits (2004, 170, 173) has suggested circa 1225 as a date, and the second half of
the twelfth century for the Lalung Serkhang (ibid., 106). The eight-armed Mañjuśrī under dis-
cussion is also present in the murals of the mid-fifteenth century “Great Stūpa,” or Kumbum
(sKu ’bum), at Gyantse, Tibet (see Ricca and Lo Bue 1993, plates 5–9; and also Tribe 2016, 89–
90, for a brief discussion).
via free access
mañjuśrī as ādibuddha 541
figure 22.2 Maṇḍala of eight-armed Mañjuśrī, uppermost level, Sumtsek, Alchi

Photograph by Jaroslav Poncar ( JP83 26.4.01 WHAV)
This position has been puzzling since one might expect to find a Buddha in
such a location, not a bodhisattva. Emphasizing this apparent oddity is the plac-
ing of a maṇḍala of Vairocana on the side wall to Mañjuśrī’s (proper) right.
As the cosmic Buddha, Vairocana has seemingly been demoted from a more
appropriate position on the back wall.5
5 For further photographs of the Sumtsek figure and the associated maṇḍala, see Goepper and
Poncar 1996, 222–223, and van Ham 2019, 344–345, 354–355. Another mural of a multi-armed
Mañjuśrī holding swords in the Sumtsek should be mentioned. To the proper right of the
monumental clay figure of Maitreya on the ground floor, there is, close to the floor, a seated
ten-armed, five-headed, white Mañjuśrī, with each hand holding a sword. Goepper and Pon-
car have just two views of this figure (1996, 127, 135), one very partial, the other small and
indistinct: the location makes photography especially challenging. Van Ham (2019, 237) has
a better image, which shows most of the figure clearly. Four of the left hands are not visi-
ble, occluded by part of the Maitreya statue. Both van Ham (ibid.) and Goepper and Poncar
(1996, 132) make slips in their descriptions of the figure, however, the former describing it as
eight-armed, the latter as six-headed. It is clearly five-headed and the five right hands, as well
as single visible left hand of the photograph, each bear a sword (van Ham states that all the
hands have swords). While neither Goepper and Poncar nor van Ham attempt to identify this
unusual image, van Ham (ibid.) nicely demonstrates the stylistic affinity between it and the
seated figure at Mangyu by placing photographs of the two figures on the same page.
via free access
542 tribe
figure 22.3 Eight-armed Mañjuśrī, Two-armed Maitreya Chapel, Mangyu

Photograph by Christiane Papa-Kalantari (CP02 44,39 WHAV)
At nearby Mangyu, the figure (Fig. 22.3) in the Two-armed Maitreya Chapel
is found on the (proper) left panel of two narrow panels framing the halo of
the large standing clay figure of Maitreya. It is one of a number of single figures
stacked vertically, and is not surrounded by a maṇḍala.
The second Mangyu figure (Fig. 22.4), in the Village Stūpa, is standing rather
than seated, and is stylistically related to the seated figure. This form, which I
have not seen elsewhere, appears to be a variant of the seated figures that were
the initial and primary focus of this article.6
6 This figure is, perhaps, particular to the context of the three other murals of standing fig-
via free access
figure 22.4 Standing eight-armed Mañjuśrī, Village Stūpa,

Mangyu
Photograph by Christiane Papa-Kalan-
tari (CP02 41,14 WHAV)
The figure in the Lalung Serkhang (Fig. 22.5) is a clay sculpture. It no longer
holds any attributes. However, like those at Mangyu and Alchi, it does not dis-
play the dharmacakra-mudrā, the standard mudrā in early Western Himalayan
art for the principal hands of figures of Dharmadhātuvāgīśvara Mañjuśrī.7 Each
ures—of Prajñāpāramitā, Avalokiteśvara and Tārā—in the Village Stūpa. Van Ham (2011, 45,
144) identifies both Mangyu images as Arapacana Mañjuśrī variations. For photographs of the
standing Mañjuśrī see van Ham ibid., 148–149.
7 For example, in the assembly halls (’Du-khang) in Alchi and Sumda, the Two-armed Maitreya
via free access
544 tribe
figure 22.5 Maṇḍala of eight-armed Mañjuśrī, Serkhang, Lalung

Photograph by Jaroslav Poncar ( JP92 500,15
WHAV)
of the eight hands of the Serkhang figure displays the same mudrā—the mid-
dle and ring fingers curled down, the middle finger touching the thumb, and
the index and little fingers extended. Christian Luczanits (2004, 98–101) has
convincingly argued that the Serkhang and Sumtsek figures are the same.8
Chapel in Mangyu, the Translator’s Temple at Nako, and the main temple’s assembly hall at
Tabo. This iconography reflects the Dharmadhātuvāgīśvara-maṇḍala as described by Abhayā-
karagupta (NYĀ 54).
8 Luczanits (2004, 99) observes that the position of the principal right arm does not conform
to his interpretation, but that there is evidence of repair work at the elbow that likely altered
the arm’s position. Assuming Luczanits is referring here to the figure’s proper right arm, I take
via free access
Giving further support to the identification he notes that the Serkhang Mañ-
juśrī is flanked on each side by two columns of four further clay sculptures
that form a surrounding maṇḍala of sixteen figures. The inner two columns
are comprised of four goddesses9 and the four directional tathāgatas.10 These
eight deities also form the first circle surrounding Mañjuśrī in the Sumtsek
maṇḍala.
In what follows I will first examine the identity of the central eight-armed
figure of Mañjuśrī, after which I will turn to the question of the nature of the
maṇḍalas surrounding the figures at Alchi and Lalung, as well as their lack
around the Mangyu figures. The eight-armed figure, as has been noted, has
previously been identified as Dharmadhātuvāgīśvara Mañjuśrī.11 This is pri-
marily in connection with the Sumtsek mural, and seems to be largely on the
basis of the two figures having the same number of arms, and, in the case of
the Sumtsek Mañjuśrī, also having the same number of heads and each being
white in colour. Thus Snellgrove and Skorupski (1977, 64) suggested the Sum-
tsek Mañjuśrī corresponds to the central figure of the Dharmadhātuvāgīśvara
Mañjuśrī of maṇḍala no. 2 in the Alchi Assembly Hall (’Du-khang).12 The two
him to be suggesting that the arm/hand is in an inappropriate position to hold a sword.

While he identifies the Serkhang and Sumtsek figures as the same, Luczanits identifies
them as an alternative form of Dharmadhātuvāgīśvara Mañjuśrī.
9 Luczanits (2004, 100) suggests that these are Locanā, Māmakī, Pāṇḍaravāsinī, and Tārā.
However, if it is right, as I will argue, that these eight-armed Mañjuśrī figures are pri-
marily associated with variants of the Vajradhātu-maṇḍala of the main yogatantra, the
Tattvasaṃgraha, I think the four female figures are more likely to be the Vajradhātu-
maṇḍala’s four Perfections (pāramitā), known also as seals (mudrā) and Family-mothers
(kulamātṛ). Locanā et al. are not witnessed as a group of four until later, in texts of the
yogottara/mahāyoga Guhyasamāja cycle, and likely first appearing in chapter one of the
Guhyasamājatantra. Tanaka Kimiaki (2018, 134–135) suggests the two groups were origi-
nally from different traditions, with the Guhyasamājatantra group evolving from a triad
of kulamātṛs—Locanā, Māmakī and Pāṇḍarā—in the Susiddhikaratantra, and, as a result,
bypassing the Tattvasaṃgraha.
10 I.e., Akṣobhya (E), Ratnasaṃbhava (S), Amitābha (W), and Amoghasiddhi (N).
11 Snellgrove and Skorupski (1977, 64), Goepper and Poncar (1996, 223), Luczanits (2004, 98–
99). Snellgrove and Skorupski’s identification has, in lieu of alternatives, generally been
accepted by subsequent writers. More recently however, given the obvious iconographical
discrepancies between the Sumtsek Mañjuśrī (plus maṇḍala) and the Dharmadhātuvāgīś-
vara-maṇḍala, the identifications have been made with increasing degrees of caution (see,
for example, Goepper and Poncar ibid., notes 151 and 153). Linrothe (1996, 272) does not
follow the Dharmadhātuvāgīśvara identification, and although he does not offer a con-
crete alternative, he usefully explores reasons why Mañjuśrī was of significance not only
in the Sumtsek and but in the Alchi complex as a whole.
12 For a photograph of this figure and first circle of the surrounding maṇḍala see Pal and
Fournier 1988, plate D 14; also van Ham 2019, 114–121, for the whole maṇḍala and details.
via free access
546 tribe
principal hands of the latter figure are in dharmacakra-mudrā. Of the remain-

ing six, the three on the (proper) right hold in turn a sword, an arrow, and a
vajra, while those on the left grasp a book, a bow, and a bell. This is standard
iconography for Dharmadhātuvāgīśvara Mañjuśrī.13 Yet, as far as I am aware,
a description of him holding four swords and four book volumes is textually
unattested.
However, an eight-armed Mañjuśrī holding four swords and four books is
described by Mañjuśrīmitra, Agrabodhi, and Vilāsavajra, all in works associ-
ated with the Nāmasaṃgīti (hereafter NS), “The Chanting of the Names.”14
The three works are Mañjuśrīmitra’s Ākāśavimala (D 2543, a NS maṇḍalavidhi),
Agrabodhi’s *Sādhanaupayika15 (Sgrub pa’i thabs, D 2579, a NS sādhana), and
Vilāsavajra’s Nāmamantrārthāvalokinī (NMAA) (D 2533, a NS commentary).
They were most likely composed between the mid-eighth and early-ninth cen-
turies, and were each translated into Tibetan in the early part of the eleventh
century.16 Only Vilāsavajra’s text is known to survive in Sanskrit. All three
authors agree in identifying this eight-armed Mañjuśrī as the Ādibuddha. Thus
understood, Mañjuśrī’s location in the position of highest status in the Sumtsek
makes sense. As the Ādibuddha, Mañjuśrī becomes the source of all Buddhas,
including Vairocana to his right, as well as Prajñāpāramitā, who is at the centre
The maṇḍala numbering is that of Snellgrove and Skorupski (1977, 38). It should be noted
that it is mistakenly numbered as “maṇḍala 1” (ibid., 64), which in their own number-
ing is a maṇḍala with Vairocana as its central figure. Snellgrove and Skorupski state that
the Sumtsek Mañjuśrī is “not [in the location he is] simply in his own right however, but
in terms of the Sarvavid (Vairocana) tradition, which controls all the Alchi iconography”
(ibid.). They may here be tacitly acknowledging that the maṇḍala surrounding Mañjuśrī
bears little resemblence to the Dharmadhātuvāgīśvara-maṇḍala in the ’Du-khang, and
presumably suggesting that Mañjuśrī as Dharmadhātuvāgīśvara has been appropriated
into a Sarvavid context.
13 This matches the description given by Abhayākaragupta (NYĀ 54, l. 6–7).
14 Also known as the Mañjuśrīnāmasaṃgīti, the NS was translated into Tibetan by Rinchen
Zangpo (Rin chen bzangpo, 958–1055 ce), although it is likely that his was not the first
translation. Rinchen Zangpo also translated the NS commentaries of Mañjuśrīmitra and
Mañjuśrīkīrti (respectively D 2532 and D 2534), the latter being the source of the Dharma-
dhātuvāgīśvara-maṇḍala. For an English translation and edition of the Sanskrit text of the
NS see Davidson 1981. For a recent overview of the text and its contents see Tribe 2015. Very
much on the right track, Linrothe (1996, 272) identified the NS as a crucial text in trying to
understand the reason for the location of the eight-armed Mañjuśrī at the top of the Alchi
Sumtsek.
15 For comment on *Sādhanaupayika as a Sanskrit reconstruction of Sgrub pa’i thabs, see
“Note for readers of Tibetan: What is a no pyi ka?” in van Schaik 2009.
16 On Mañjuśrīmitra and Vilāsavajra see Davidson 1981, 5–8. For a more extended discussion
of Vilāsavajra’s dates and life, see Tribe 2016, 21–33.
via free access
of a maṇḍala to his (proper) left. Mañjuśrī should no longer, at least in this form,
be seen as a bodhisattva.17
The depictions of Mañjuśrī as Ādibuddha by Vilāsavajra, Mañjuśrīmitra, and
Agrabodhi will be examined next. I will also comment on the relationship
between the textual descriptions and the artistic depictions at Alchi, Mangyu
and Lalung.
1 Vilāsavajra’s Nāmamantrārthāvalokinī
Of the three authors, Vilāsavajra’s description of the Ādibuddha is the most

extensive. The NMAA interprets the Nāmasaṃgīti within the context of a tantric
sādhana,18 one based on an expanded version of the Tattvasaṃgraha’s yoga-
tantra Vajradhātu-maṇḍala.19 The maṇḍala’s principal deity is a four-faced
Mahāvairocana, and it is in his heart that Vilāsavajra locates the Ādibuddha,
depicted with eight arms holding four swords and four book volumes. The
Ādibuddha is not the terminus of the interiorization process in Vilāsavajra’s
sādhana, however. At the Ādibuddha’s heart a prajñācakra (“wisdom wheel”)
is visualised, on which mantras assosiated with the Nāmasaṃgīti are placed.
At the centre of the prajñācakra a final form is generated, that of Mañjuśrī-
jñānasattva (“the gnosis-being Mañjuśrī”), who is seen as the embodiment of
non-dual gnosis (advayajñāna). Mañjuśrī-jñānasattva is visualised as six-faced,
and holding a blue lotus in each of his two hands. Each lotus is crowned with
a volume of the Prajñāpāramitā. As a result, although the Ādibuddha is, in a
sense, the cakreśa (“lord of the maṇḍala”) of Vilāsavajra’s Vajradhātu-maṇḍala,
he is the intermediate figure of three forms, all of whom have, or share, that
role.
In the fourth chapter of the NMAA, Vilāsavajra’s description of the Ādibud-
dha directly follows that of Mahāvairocana:
17 Mañjuśrī thus transitions from being the bodhisattva of wisdom (prajñā) to being the wis-
dom (now jñāna) that underlies, and is therefore conceptually prior to, Buddhahood.
18 For a critical edition and translation of the Sanskrit text of the first five chapters of the
Nāmamantrārthāvalokinī that contains a more detailed analysis of Vilāsavajra’s sādhana,
see Tribe 2016. A summary of it is also available in Tribe 1997.
19 The Vajradhātu-maṇḍala is more accurately called the Vajradhātu-mahāmaṇḍala. For con-
venience I use the shorter and more familiar form. The Tattvasaṃgraha is known also by
the more extended title, Sarvatathāgatatattvasaṃgraha. The shorter form is commonly
found in Sanskrit sources, including the NMAA.
via free access
548 tribe
tam evaṃbhūtaṃ mahāvairocanam ātmānam adhimucya caturbuddhā-

sanayogena taddhṛdaye candramaṇḍalaṃ tadupari dhīḥkāreṇa pariniṣ-
pannam ādibuddhaṃ bhagavantam | pañcānanam iti pañcamukham |
pañcaśikham iti pañcacīrakam | tasyaiva bandhanāt pañcacīrakaśekha-
ram | pañcavarṇopetam | pūrveṇa nīlaṃ dakṣiṇena pītaṃ paścimena rak-
tam uttareṇa haritam | mūrdhni paramāśvamukhavad avasthitaṃ mu-
khaṃ śvetaṃ ⟨yasya taṃ⟩ | śāntaṃ kumārābharaṇopetaṃ saśṛṅgāraṃ vic-
itravastraparidhānam aṣṭabhujaṃ śatasāhasrikāṃ prajñāpāramitāṃ ca-
turdhā vibhajya caturbhiḥ karair hṛdi saṃdhār⟨a⟩yamānam | aparaiś ca-
turbhiḥ karaiḥ prajñākhaḍgaṃ praharaṇābhinayena dhārayantaṃ catur-
buddhāsanayogena vyavasthitaṃ bhāvayet ||
Tribe 2016, 255–256, ll. 105–116
[The sādhaka,] on generating the conviction that he himself is Mahāvai-

rocana as [previously] described, via the yoga of the four Buddha-thrones,
should visualise a moon-disc in his heart. Above that, transformed out
of the syllable dhīḥ, [he should visualise] the lord, the Ādibuddha. [The
Ādibuddha] has five faces (pañcānana > pañcamukha). [He also] has
five crests (pañcaśikha)—in other words, five hair-braids. It is through
tying up those [hair-braids that he] has a crown of five hair-braids (pañ-
cacīrakaśekhara). [His five faces] have five [different] colours: dark blue
for the east [and forward-facing face], yellow for the south, red for the
west, [and] green for the north. On the top, he has a white face, the face
of [the deity] Paramāśva. He is tranquil, with the ornaments of a youth,
in fine clothing, wearing about himself a many coloured garment. He
has eight arms, holding at his heart with four hands the Śatasāhasrikā-
prajñāpāramitā (“The Perfection of Wisdom in One Hundred Thousand
Lines”) divided into four parts, [and] carrying, in each of the other four
hands, a sword of wisdom in the gesture of striking. [All this is to be] put
in place [i.e. visualised] via the yoga of the four Buddha-thrones.20
This passage is not free from textual or interpretive problems but none of them
materially affects the present discussion. Although Vilāsavajra’s figure corre-
sponds to the Alchi and Mangyu Mañjuśrī figures in holding the four swords
and four text volumes, other aspects of the description do not completely
match. Vilāsavajra’s Ādibuddha has five faces, differently coloured. The Sum-
20 The translation is a slightly adapted version of the translation that accompanies the San-
skrit edition in Tribe 2016.
via free access
tsek figure has four faces, all white. The Lalung Serkhang Mañjuśrī has three
faces visible, each painted golden.21 At Mangyu, both the seated and standing
Mañjuśrī figures do have five faces, albeit arranged differently from the NMAA
description. They likely also have the same colours,22 those of the four direc-
tions plus white for the main forward-looking face, matching their body colour.
In addition, it is worth noting that in the NMMA the Ādibuddha is said to hold
all four volumes of the Śatasāhasrikā-prajñāpāramitā to his heart. While the
seated Mangyu figure holds one book to his heart, the standing Mangyu and
the Sumtsek figures hold a sword to their hearts (and no book volumes), and
the Serkhang figure has one (empty) left hand raised to heart level but just to
the side of his torso.23
A note of caution should perhaps be added here. It has been observed that
the murals and clay sculptures of early Western Himalayan art reflect a process
of Tibetan integration and adaptation of their Indian sources.24 This means
that it may not always be feasible to pinpoint precise textual sources for iconog-
raphy. Elements of a maṇḍala’s or image’s iconography—number of heads,
or colour of faces, for example—may be local adaptations, or inflections, of
descriptions transmitted originally via Indian teachers and their Sanskrit texts.
Bearing this proviso in mind, I hope to show that it does seem possible to rule
some (broad) iconographic identifications in, and others out.
21 Luczanits (2004, 99) suggests a fourth head may be hidden by the (cloth-draped) umbrella
above the eight-armed figure’s three heads. It should be noted that the clay sculptures of
early Western Himalayan art have generally been repainted, and not necessarily in their
original colours. The Lalung Serkhang was completely repainted in the early 20th century
Luczanits (ibid., 93–94).
22 For discussion of the facial colours of the Mangyu figures see the section on below on
Mañjuśrīmitra’s Ākāśavimala. While this description of the Ādibuddha does not specify
whether the figure is seated or standing, I think it fairly certain that the former is pre-
supposed. Prior to the visualisations of Mahāvairocana and the Ādibuddha, the NMAA
describes how the maṇḍala, as a residence, should be visualised. The description con-
cludes with an account of the thrones (āsana) of the deities, with a lion throne in the
centre.
23 The positions of the principal hands of the seated Mangyu and Sumtsek and standing
Mangyu figures are exactly reversed: left hand to heart with book, right hand to hip with
sword (seated Mangyu), and left hand to hip with book, right hand to heart with sword
(Sumtsek and standing Mangyu). The seated figure at Mangyu with a book held to his
heart is thus closer to Vilāsavajra’s description, as well as in alignment with Mañjuśrī’s
primary association with wisdom. In this respect the Lalung figure appears (assuming
an unchanged principal left arm position) to be more affiliated with the seated Mangyu
iconography.
24 See Luczanits (2004, 72) for some comments on this in relation to Tabo.
via free access
550 tribe
In the above description of the Ādibuddha by Vilāsavajra, the words in bold

typeface reference NS 93 (pañcānanaḥ pañcaśikhaḥ pañcacīrakaśekharaḥ).
When Vilāsavajra specifically comments on NS 93 later in the NMAA, unfor-
tunately he does not add anything, but simply refers the reader back to the
present description:
pañcānanaḥ pañcaśikhaḥ pañcacīrakaśekhara iti || ayam aparārdhaś

cādibuddhasya gurūpadeśenādiyoge draṣṭavyaḥ | tatraiva vyākhyātatvāc
ca punar iha nocyate ||
iti] AD; ti B • ayam] AD; aryam B • aparārdhaś] corr.; aparārddhaś B; ana(ṣṭhu?)varddhaś A;

anastavarddhaś D • ihaiva māyājālābhisambodhikramoktasya mahāvairocanahṛdgatasya
A (as a marginal addition with an insertion mark after cādibuddha) • yoge] AD; ge B
NMAA chapter 8, on NS 93cd; mss: A.69v4, B.48v3, D.110r2
With five faces. With five crests [of hair]. With a crown of five hair-
braids. And this second half [of the verse] should be understood with
reference to the Ādibuddha, via the teacher’s instruction on the begin-
ning yoga (ādiyoga) [phase of the sādhana]. And as it is explained there,
it is not restated here.
Ms. A’s marginal addition (whether scribal or a correction) somewhat super-

flously adds that the Ādibuddha is “in the heart of Mahāvairocana [as] stated
in the “Method of Awakening according to the Māyājāla” [i.e., in NMAA ch. 4,
where the Ādibuddha is described].” Although the presence of swords and
book volumes makes it hard not to see Vilāsavajra’s Ādibuddha as a form of
Mañjuśrī, he does not here explicitly say that he is. This is remedied later in his
comment on the word ādibuddha (NS 100ab: anādinidhano buddha ādibuddho
niranvayaḥ):
ādibuddha iti || ādāv eva buddha ādibuddhaḥ | sa ca pañcajñānātmakaḥ

…25 evaṃ pañcajñānātmakaḥ pañcavarṇātmakaś cāsau bhagavān | sar-
25 This ellipsis contains a summary of the five jñānāni (gnoses) in terms of various master-
ies. Thus the ādarśajñānam (mirror-like gnosis) is associated with the five balāni (powers);
samatājñānam (gnosis of equality), with the four vaiśāradyāni (confidences); pratyavekṣ-
aṇājñānam (discriminating gnosis), with the four pratisaṃvidaḥ (special knowledges);
kṛtyānuṣṭhānajñānam (praxis gnosis), with friendliness (maitrī) and compassion (ka-
ruṇā); and suviśuddhadharmadhātujñānam (gnosis of the completely pure dharma-
sphere), with overlordship of everything (sarvādhipatya). These associations are refer-
enced to the Māyājālatantra in ms. A (alone).
via free access
vadharmasamatāmukhenaikālambanatvena draṣṭavyaḥ sa ca mañjuśrīr

ata evāsau niranvaya iti ||
varṇātmakaś] A; varṇārtmakaś B; varṇāś D • ālambanatvena] A; ālambanatve D; ārthatve-

na B • niranvaya] BD; niratvaya A.
NMAA chapter 8, on NS 100ab; mss: A.75r1, B.53r1, D.120v1)
Ādi-buddha: [the word] ādibuddha means [he who is] “awakened from
the very beginning,” and that one has the five gnoses ( jñānāni) as his
nature … So that one, who has the five gnoses as his nature and [also]
the five colours as his nature, is the lord (bhagavān). And he should be
understood to be Mañjuśrī, since as the equality of all dharmas he is the
unique ground [of all phenonena]. For that very reason he is [described
as] free from [causal] connection (niranvaya).
To summarize, Vilāsavajra has clearly identified an eight-armed figure holding

four swords and four books as the Ādibuddha, and has stated that the Ādibud-
dha is also Mañjuśrī. At the same time, there are some discrepences between
Vilāsavajra’s description and the images at Alchi, Mangyu and Lalung. There is
also some variance of appearance between each of the four (three mural and
one clay) temple figures.
The likelihood of Vilāsavajra’s text—or Agrabodhi’s, as we shall see—being
of influence in the iconography of these figures is given support by the exis-
tence of a small clay figure directly above the eight-armed Mañjuśrī in the
Lalung Serkhang (Fig. 22.6). The figure is two-armed, each hand holding a lotus
stem. The two stems each rise to a flower base, one on each side of the figure a
little above its shoulders.26
If it is correct to identify the central eight-armed figure as the Ādibud-
dha, then I think the two-armed figure may be Mañjuśrī-jñānasattva, who in
26 This distinctive figure was noted by Tucci (1988, 118), who, having identified the Ādibud-
dha figure as “Vairocana or one of his emanations” (ibid., 117), on the basis that the central
figure was surrounded by the four directional tathāgatas, briefly observed that the fig-
ure above was Sarvavid. More recently this image has been discussed at greater length by
Luczanits (2004, 99–100, 208–209), who notes the figure’s unusual iconography and sug-
gests a possible link with the sun-god Sūrya, who holds two lotuses similarly. This in turn,
he suggests, supports the hypothesis that the figure is a form of Vairocana given the word
vairocana means “resplendent, exceeedingly bright.” In fact, the link between Vairocana
and the sun is stronger than Luczanits perhaps realised: the literal Sanskrit meaning of
vairocana is “coming from or belonging to the sun (virocana).” Also, Edgerton (1953 II, 512)
suggests that in Buddhist contexts virocana, “the sun,” and its derivative vairocana, can be
synonymous.
via free access
552 tribe
figure 22.6 Two-armed Mañjuśrī (as Jñānasattva), Serkhang, Lalung

Photograph by Christian Luczanits (CL91 52,6 WHAV)
the NMAA is visualised in the heart of the Ādibuddha. Vilāsavajra describes

Mañjuśrī-jñānasattva as follows:
tasya cakrasya madhyavartinam akārajaṃ bhagavantaṃ ṣaṇmukhaṃ

jñānasattvaṃ śaracchaśāṅkaprabham indranīlāgrasaccīraṃ bālārkama-
ṇḍalacchāyaprabhāmaṇḍalaṃ sarvatathāgatamayābharaṇaṃ samādhi-
samāpannaṃ vicitrapadmāsanopaviṣṭam ubhayakarāsaktanīlotpalor-
dhvasthitaprajñāpāramitāpustakadvayaṃ śāntarasopetam ātmānaṃ vi-
cintya |
Tribe 2016, 260, ll. 172–176
via free access
[Next] he should visualise himself as the fortunate one, the gnosis-being

[Mañjuśrī], born from the syllable a situated in the middle of that [wis-
dom-]wheel [situated in the heart of the Ādibuddha]. He has six faces, is
radiant like the autumn moon, with the best of sapphires in his beauti-
ful hair, with a halo that has the brilliance of the orb of the newly risen
sun, with all the tathāgatas as [head-]ornaments, immersed in medita-
tive concentration, seated on a variagated lotus throne, in tranquil mood,
with a pair of books of the Prajñāpāramitā above blue lotuses held in his
two hands.
The figure in the Serkhang is missing the volumes of the Prajñāpāramitā, as

is the eight-armed Mañjuśrī below him.27 There is also a discrepancy in the
count of heads. Vilāsavajra has six for Mañjuśrī-jñānasattva (one more than
the Ādibuddha, who in turn has one more than Mahāvairocana, the outermost
figure of the central nested triad). Although the Serkhang figure has just four
heads, the fourth is about a head’s distance above the tops of the first three
heads. The result is that the figure looks distinctly odd, with one head float-
ing above the other three. It does not seem impossible (and it would certainly
improve the aesthetics) that there were originally two further heads in between
the bottom three and single top head, giving a pyramidal structure of three +
two + one heads, i.e. six altogether, matching Vilāsavajra’s description.
There is also a small figure in direct vertical alignment with the two-armed
and eight-armed sculptures, immediately beneath the (let us assume) Ādibud-
dha’s lion throne. It is a mural of a white four-faced Mahāvairocana, two-armed
and simply attired, with his hands in dharmacakra-mudrā (see Fig. 22.5).28
His location may be coincidental, but if not, he completes Vilāsavajra’s triad
of Mahāvairocana, the Ādibuddha, and Mañjuśrī-jñānasattva. The location of
Mahāvairocana beneath the other two figures is hard to explain unless he
is somehow understood to be subordinate to them, which is precisely what
Vilāsavajra is saying.29
Vilāsavajra describes Mahāvairocana as follows:
27 That the Ādibuddha lost the four volumes of the Prajñāpāramitā apparantly without dam-
aging the hand mudrās suggests that they were easily removable. The same detachability
may have applied to the two-armed figure.
28 Luczanits (2004, 93–94) notes that while the Serkhang murals were repainted crudely dur-
ing the early twentieth century, it appears that the repainting followed what was present
before.
29 Namely that Mahāvairocana (= all samādhis) has the Ādibuddha (= the pañcajñāna) as
his nature, who in turn has Mañjuśrī-jñānasattva (= advayajñāna) as his nature. It is pos-
sible that the Mahāvairocana mural is not part of the clay sculpture maṇḍala at all, and
via free access
554 tribe
tataś ca pradhānāsana āḥkāreṇa pariniṣpannaṃ mahāvairocanam | yato

yad eva cittaṃ prakṛtyā grāhakādyākāravirahād dharmadhātusvabhāvaṃ
tad eva caturmukhaṃ śūnyatādicaturvimokṣamukhānāṃ dharmadhātor
ālambanatvena sarvasamādhiprasūtihetutvāt | śuklavarṇaṃ dharmadhā-
tusvabhāvatvāt | jaṭāmakuṭopetaṃ nirābharaṇaṃ ca śāntacittatvāt | bodh-
yagrīmudrāyuktaṃ prajñopāyātmakatvāt
Tribe 2016, 255, ll. 100–105
And then [the sādhaka should visualise] Mahāvairocana on the princi-

pal seat, generated by means of the syllable āḥ. [Why has he four faces?]
Since consciousness—which is of the nature of the Dharma-Sphere since,
by its nature, it lacks such forms as the grasped [i.e., the subject-object
duality]—is four-faced. [This is] because the four liberation faces
[/doors] (vimokṣamukha)—emptiness and the rest—are the cause of the
origination of all meditative concentrations (samādhi), [and this in turn
is] because their ground is the Dharma-Sphere. He is white in colour
because he has the Dharma-Sphere as his nature. He has braids of hair
[stacked up on his head] as a crown and is unadorned because he is
one whose mind is tranquil. Since he has both wisdom and means as his
nature he makes the bodhyagrī (“highest awakening”) hand gesture.30
In contrast with Vilāsavajra’s description, the Serkhang Mahāvairocana is not

“unadorned”. He has armlets, bracelets and two necklaces. Yet he is is still
fairly simply adorned, with no clothes on his upper body and with his hair
stacked behind a minimal crown.31 He is two-armed but with his hands in
that the significance of his location should be understood just in relation to the two rows
of deities that flank him to each side. However, this same vertical alignment of the three
figures of Vilāsavajra’s NMAA is also found in a Tibetan thangka of unknown provenance,
a photograph of which was kindly sent to me by Tanaka Kimiaki. In the thangka the cen-
tral figure is Mañjuśrī-jñānasattva, however, with the eight-armed Ādibuddha above him,
and a small four-headed Mahāvairocana beneath him; and as in the Serkhang, the three
central figures are surrounded by the four directional tathāgatas and four kulamātṛs.
30 bodhyagrī-mudrā. In this mudrā the extended and raised left forefinger is grasped and
encircled by the fingers of the right hand, the hands being held at the heart. The Tattvasaṃ-
graha (89, 4–5) describes it as follows: “The raised left vajra-finger should be grasped with
the right [hand]. This mudrā, which bestows the awakening of the Buddhas, is called bodh-
yagrī” (vāmavajrāṅgulir grāhyā [em.; grāhya Ed.] dakṣineṇa samutthitā | bodhyagrī [em.;
bodhāgrī Ed.] nāma mudreyaṃ buddhabodhipradāyikā). For discussion of variants of this
mudrā and confusion surrounding its name, see Tribe 2016, 90–92.
31 A good example of a mural of an unadorned (i.e., with no jewelry and simple clothing)
Mahāvairocana with a crown of stacked hair ( jaṭāmakuṭa) is present in the Gyantse Kum-
via free access
dharmacakra-mudrā rather than bodhyagrī-mudrā.32 Nevertheless, although

Vilāsavajra’s description does not exactly match the Serkhang figure, its broad
correspondence, and especially its location, is striking.
Before discussing Mañjuśrīmitra’s and Agrabodhi’s texts, it is worth noting
that Vilāsavarja makes a clear distinction between Mañjuśrī as jñānasattva and
as bodhisattva:
jñānasattva iti sarvatathāgatahṛdayavihāritvāt | mañjuśrīś cāsau jñāna-

sattvaś ceti mañjuśrījñānasattvaḥ | nāyaṃ daśabhūmīśvaro bodhisattvaḥ
kiṃ tarhy advayajñānaṃ prajñāpāramitā saiva mañjuśrījñānasattvaḥ
Tribe 2016, 236, ll. 215–218
He is [described in NS 10 as] the jñānasattva since he dwells in the heart

of all the tathāgatas. The jñānasattva Mañjuśrī is not the bodhisattva that
is the master of the ten [bodhisattva] stages. Rather, he is non-dual gnosis
(advayajñāna), the perfection of wisdom (prajñāpāramitā) itself.33
If Mañjuśrī-jñānasattva should not be identified with the bodhisattva Mañjuśrī,

then by extension, this should equally apply to Mañjuśrī as the Ādibuddha.
Understanding this helps makes sense of the iconographic programme of the
Sumtsek (and by extension of the three figures in vertical alignment in the
Lalung Serkhang). Tucci (1988, 117–118) proposed that the two clay figures in the
Lalung Serkhang were manifestations or forms of Vairocana or Mahāvairocana.
In the present reading the inverse is true: Mañjuśrī the Ādibuddha manifests
Mahāvairocana. And by implication, he also manifests Mañjuśrī the bodhi-
sattva along with all the tathāgatas and other bodhisattvas. Thus at Lalung it
is the jñānasattva Mañjuśrī, as the topmost figure, that can be read as mani-
festing, in turn, the Ādibuddha and Mahāvairocana.
bum (see Ricca and Lo Bue 1993, plate 5). If jaṭāmakuṭa is analysed as a dvandva rather
than a karmadhāraya compound it can be translated as “having a crown and stacked hair.”
For a Mahāvairocana depicted in this way, also in the Kumbum, with a large crown in front
of stacked hair see Ricca and Lo Bue ibid., plate 4.
32 The Alchi ’Du-khang has a mural of four-faced, two-armed Mahāvairocana with his hands
in bodhyagrī-mudrā (see Pal and Fournier 1988, plate D 5; van Ham 2019, 94–95). Like the
Sumtsek eight-armed Ādibuddha, this figure has its fourth head stacked centrally above
the others. In contrast, the Serkhang Mahāvairocana has all four heads on the same level,
with two to the proper right of the front face, and one to the left.
33 The karmadhāraya analysis of mañjuśrījñānasattva is omitted from the translation.
via free access
556 tribe
2 Mañjuśrīmitra’s Ākāśavimala
Mañjuśrīmitra, who may have been an older contemporary of Vilāsavajra,

wrote extensively on the Nāmasaṃgīti.34 In the Ākāśavimala, he describes both
the Ādibuddha and Mañjuśrī-jñānasattva, albeit more briefly than Vilāsavjara.
First, the Ādibuddha:
dang po’i sangs rgyas zhal lnga pa | phyag brgyad pa | g.yas bzhin shes rab
kyi ral gri | g.yon bzhin shes rab kyi pha rol tu phyin pa’i po ti yod pa | zhal
bzhi phyogs dang mthun pa | dbus dmar ser ro |
D 2543, 4r3–4
The Ādibuddha has five faces and eight arms. In the right [arms] are
sword[s] of wisdom; in the left volume[s] of the Prajñāpāramitā. The four
faces correspond [in colour] to the directions. The central face is orange.
Mañjuśrīmitra and Vilāsavajra agree on the key iconographical elements. There

is one significant difference in that Vilāsavajra’s fifth face is white, Mañjuśrī-
mitra’s is orange. As noted earlier, the seated figure at Mangyu has five faces
(unlike the four-faced Sumtsek figure), and the top and fifth face appears to
be orange-brown. However, it is likely to have originally been green. If it were
orange it would point to Mañjuśrīmitra as a possible textual source, although
he does not describe a white face, which is the colour of the principal faces for
both Mangyu figures. The likely explanation of the orange-brown colour of the
top face of the seated Mangyu figure is that the original colour has faded over
time. It is clear from the Sumtsek maṇḍala that while reds and blues preserve
their colour fairly well over time, greens and yellows can fade; in the Sumtsek
instance, they both faded to a pale reddish brown.35 The arrangement of heads
in both the Mangyu figures (seated and standing) is the same: three heads at
the same level (front, left and right facing), surmounted by two further heads,
one above the other. The topmost face of the standing Mangyu figure is clearly
green, in contrast to the orange-brown of the seated figure. The intermediate
face colouring for both is similar: very pale brown for the seated figure and off-
white for the figure standing. The colours of the three horizontal lower faces
are identical: white (to the front), flanked by blue (to the proper right) and red
(to the left). If green is correct for the topmost face, and if it is assumed that an
34 Davidson (1981, 5–6) dates Mañjuśrīmitra to the mid-eighth century and notes that Buston
records a tradition that he may have been Vilāsavajra’s teacher.
35 See below, note 40. In the Mangyu figures yellow appears to fade almost to white.
via free access
original yellow of the intermediate faces has faded, then both figures have the
same head arrangement and colouring. These colours are those described by
Vilāsavajra. The head arrangement differs, however: Vilāsavajra has four heads
(representing the four directional tathāgatas) at the same level, each facing
one of the four directions, and the fifth central (principal) white head above
the four. The Mangyu arrangement could be understood as an adaptation to
the constraints of two dimensional mural painting.36
Turning to Mañjuśrīmitra’s description of Mañjuśrī-jñānasattva, it is also
structurally parallel to that of Vilāsavajra and immediately follows his descrip-
tion of the Ādibuddha:
de’i thugs kar shes rab kyi ’khor lo rtsibs drug pa la | gsang sngags rgyal po
drug gis mtshan pa | de’i kyil du ’jam dpal ye shes sems dpa’ zhal drug phag
gnyis pa | g.yas ral gris g.yon po ti ’dzin pa
D 2543, 4r4
In his (i.e. the Ādibuddha’s) heart is a wisdom wheel (prajñā-cakra), six-

spoked, ornamented with the six secret mantra kings. In its centre is
Mañjuśrī-jñānasattva, with six faces [and] two hands. The right [hand]
carries a sword, the left a book.
Again there is a significant difference between Mañjuśrīmitra and Vilāsavajra.

Vilāsavajra’s Mañjuśrī-jñānasattva does not hold a sword and a book, but has a
lotus stem in each hand, each topped with a book. If it is correct to identify the
two-armed figure in the Lalung Serkhang above the Ādibuddha as Mañjuśrī-
jñānasattva, Vilāsavajra seems a more likely iconographical source than Mañ-
juśrīmitra.
3 Agrabodhi’s *Sādhanaupayika
Agrabodhi (Tib. Byang chub mchog) is described as Vilāsavajra’s maternal uncle

in the NMAA’s colophon.37 This makes him a contemporary of both Mañjuśrī-
36 In the murals of these eight-armed Mañjuśrī figures at Mangyu and in the Sumtsek the
principal face and body is white in each case, indicating their alignment with Mahāvairo-
cana, whose distinctive colour is also white, and who is so described by Vilāsavajra
above: śuklavarṇaṃ dharmadhātusvabhāvatvāt, “He is white in colour because he has the
Dharma-Sphere as his nature.”
37 Byang chub mchog has been standardly reconstructed as *Varabodhi. For discussion of
the NMAA’s colophon see Tribe 2016, 25–28 and Appendix 3.
via free access
558 tribe
mitra and Vilāsavajra, and probably the latter’s elder. In his Nāmasaṃgīti *Sā-
dhanaupayika Agrabodhi also describes the Ādibuddha and Mañjuśrī-jñāna-
sattva. Again, first the Ādibuddha:
dang po’i sangs rgyas zhal lnga pa | | byis pa’i rgyan gyis rnam brgyan cing
| | rin chen sna tshogs na bza’ can | | phyag brgyad mnga’ ba’i phyag mtshan
ni | | g.yas pa bzhi ni go rims bzhin | | shes rab kyi ni ral gri bsnams | | de
bzhin g.yon pa’i phyag bzhi na | | shes rab pha rol phyin pa yi | | po ti re re
bsnams pa’o | |
D 2579, 62v7–63r1
The Ādibuddha has five faces. He is adorned with the ornaments of youth,
and [wears] garments [decorated] with various gems. He has eight hands.
As for the attributes held by the hands: the four right hands in turn carry
a sword of wisdom; similarly, the four left hands each carry a volume of
the Prajñāpāramitā.
The basic iconography of five faces and eight arms (with swords, and volumes
of the Prajñāpāramitā) follows Vilāsavajra and Mañjuśrīmitra. Although the
description is a little more elaborate than Mañjuśrīmitra’s, Agrabodhi omits
colours for the Ādibuddha’s faces. Agrabodhi’s description of Mañjuśrī-jñāna-
sattva follows a half folio later:
ye shes sems dpa’ zhal drug pa | | ston ka’i zla ba ltar gsal zhing | | indra nīla’i
gtsug phud can | | nyi ma ’char ka’i ’od ’dras bskor | | zhi ba’i ngang tshul dang
ldan zhing | | g.yas dang g.yon gyi phag gnyis kyis | | utpala dmar po g.yas
pa ni | | g.yon pa na ni sngon po nyid | | utpala de gnyis steng nyid na | | shes
rab po tis mdzes pa’o | |
D 2579, 63v.2–3
Mañjuśrī-jñānasattva has six faces. He is luminous like the autumn moon,

with the choicest sapphire as a [head] crest. He is encircled as if by the
light of the rising sun, and he has a peaceful disposition. With [his] two
hands—right and left—[he holds] a red lotus in the right, a blue one in
the left. Adorning the top of [each of] those two lotuses is a volume of
Prajñā[paramitā].
All three authors agree, then, that Mañjuśrī-jñānasattva has six heads and
two arms; and Agrabodhi and Vilāsavajra concur (against Mañjuśrīmitra) that
he holds two lotuses topped with books. Agrabodhi and Vilāsavajra differ,
via free access
however, in the colour of the lotuses. Agrabodhi has one red and one blue;
Vilāsavajra has them both be blue. As noted, the Serkhang scultpures have been
repainted in the twentieth century38 and at present the Mañjuśrī-jñānasattva
figure has two green lotuses. In conclusion, either Agrabodhi or Vilāsavajra or
both could be a source for the Serkhang figures. All three authors also share
the same core structure for their sādhanas, with the Ādibuddha in Mahāvairo-
cana’s heart, a wisdom-wheel within the Ādibuddha’s heart, and Mañjuśrī-
jñānasattva in the centre of the wisdom-wheel. And like Vilāsasvajra, both
Mañjuśrīmitra and Agrabodhi describe Mahāvairocana with four heads, and
with hands in bodhyagrī-mudrā.39
4 Broader Iconographic Contexts
Some remarks about the maṇḍalas surrounding the Mañjuśrī Ādibuddha fig-
ures at Alchi and Lalung, and the lack of any at Mangyu, follow. First, the Alchi
Sumtsek. While I am not able to clearly identify either Mañjuśrīmitra, Vilāsa-
vajra or Agrabodhi as providing the textual source for the maṇḍala, one thing
is clear: it is not a maṇḍala of Dharmadhātuvāgīśvara Mañjuśrī. This is the case
irrespective of the identity of the central deity.
The Sumtsek maṇḍala (Fig. 22.2) contains fifty-three figures in total, all con-
tained within a single four-gated square courtyard. Within the courtyard is
another (ungated) square that contains two circles surrounding the central
deity. The first of these circles contains eight figures, the second sixteen. In
the first circle in the intermediate directions are the four family mothers (kula-
mātṛ), identifiable by the family symbols they hold (vajra, gem, lotus, viśva-
vajra). In the four cardinal directions are the four directional tathāgatas, indi-
cated by their mudrā, animal/mount (vāhana) and colour.40 The second circle
may well contain the sixteen samādhi deities of the Vajradhātu-maṇḍala. Each

39 See Ākāśavimala D 2543, 4r2–3; *Sādhanaupayika D 2579, 62v6–7
40 While Akṣobhya (blue) and Amitābha (red) have their colours, Ratnasambhava and
Amoghasiddhi are a similar pale reddish brown. Their original colours have likely faded,
both here and throughout the maṇḍala. The colour of the garment on the Ādibuddha’s
legs is a similar, presumably faded, colour; it is possible to discern the remains of a floral
pattern on the leg coverings. The high status of this figure is perhaps further reflected in
the use of gold paint, not only for the slightly embossed circle that surrounds him and for
the crowns on his heads, but also for the swords and book volumes. Close inspection also
reveals that both ends of each sword and book volume are adorned with a terminus of
vajra prongs.
via free access
560 tribe
figure is iconographically distinct, although with the exception of Vajrarāga (E),

Vajraratna? (SE), Vajratejas/Vajrasūrya (SSE), Vajratīkṣṇa (S), and Vajrayakṣa
(N) I have not been able to make clear identifications.41 In the corners between
the second circle and the ungated square are four offering goddesses.
Beyond the square, in the space between it and the main gated courtyard
walls are the remaining figures, twenty-four altogether. The corners are occu-
pied by four further offering goddesses. This leaves five figures on each side.
The members of each group of five have the same colour (that of their presid-
ing tathāgata) and hold the attribute definitive of their family (i.e. vajra, gem,
lotus, or viśvavajra).42 It is tempting to identify sixteen of these figures as (non
iconographically individuated) bodhisattvas. This would leave the remaining
four as door guardians.43
The Dharmadhātuvāgīśvara-maṇḍala in the Alchi ’Du-khang that, accord-
ing to Snellgrove and Skorupski (1977, 64), corresponds to the Sumtsek maṇ-
ḍala also has eight figures in a circle surrounding the central deity. In this the
two maṇḍalas are structurally alike. The figures in the ’Du-khang circle, how-
ever, are the eight uṣṇīṣa deities, iconographyically identical. Beyond this, and
within the first of three gated courtyards are four eight-armed tathāgatas and
four eight-armed goddesses.44 As the existence of three courtyards suggests,
the Dharmadhātuvāgīśvara-maṇḍala is large, with over two hundred deities.45
The Sumtsek maṇḍala should, I believe, be identified instead as a Vajradhātu-
maṇḍala variant. If the sixteen putative bodhisattva figures are excluded, the
maṇḍala is identical to the thirty-seven deity Vajradhātu-maṇḍala of the
Tattvasaṃgraha. A set of sixteen bodhisattvas often supplements the deities of
the Tattvasaṃgraha in later Vajradhātu-maṇḍala descriptions. These are gen-
erally the sixteen bodhisattvas of the present aeon (bhadrakalpa), as described
for example in Abhayākaragupta’s Niṣpannayogāvalī Vajradhātu-maṇḍala.46
41 Also, some of the directional locations of the identifiable figures are unusual: for example
Vajratīkṣṇa, standardly associated with Amitābha and the west, is in the south.
42 As with their tathāgatas, the colours of the southern and northern deities appear to have
faded.
43 However, those that occupy the positions of door guardians are not iconographically dis-
tinct from their neighbours.
44 For photographs of Dharmadhātuvāgīśvara Mañjuśrī and the uṣṇīṣa deities, see van Ham
2019, 115, and Pal and Fournier 1988, plate D 14. For examples of the tathāgatas and god-
desses see van Ham ibid. 115–116, and Pal and Fournier ibid., plates D 15–18.
45 For Abhayākaragupta’s description of the Dharmadhātuvāgīśvara-maṇḍala, which con-
tains two hundred and sixteen figures, see NYĀ 54 ff.
46 See NYĀ 44 ff. The same list is also present in Abhayākaragupta’s Durgatipariśodhana-
maṇḍala (NYĀ 66 ff.).
via free access
In assessing possible textual sources for the Sumtsek and Lalung maṇḍalas,
at present I am only able to comment on the maṇḍala elaborated by Vilāsa-
vajra around the triple central figure(s) of Mahāvairocana, the Ādibuddha and
Mañjuśrī-jñānasattva.47 It too is a fifty-three deity (if the central triad is counted
as one) Vajradhātu-maṇḍala variant. The deities are enumerated/emanated
in the following order: four family mothers (kulamātṛ),48 four tathāgatas, six-
teen samādhi deities, eight offering goddesses (pūjādevī), four gate guardians
(dvārapāla), and sixteen bodhisattvas. While Vilāsavajra elaborately connects
these deities with doctrinal categories and Nāmasaṃgīti “names,” he gives no
iconographical descriptions. Also, while the sixteen putative bodhisattva fig-
ures of the Sumtsek maṇḍala are not distinguished beyond their family affilia-
tion, the NMAA set is not the same as the bhadrakalpa group.49 Thus while the
Sumtsek maṇḍala could be derived from Vilāsavajra’s NMAA, further investiga-
tion may reveal a more immediate source.
Turning very briefly to the deities surrounding the Ādibuddha figure in the
Lalung Serkhang, it was seen earlier that (again counting the central triad as
one figure) they comprise a maṇḍala of seventeen figures, two columns of four
figures positioned to each side of the Ādibuddha. It was also noted that Luczan-
its (2004, 98–101) identified the eight figures of the two inner columns as identi-
cal with the figures in the first circle surrounding the Ādibuddha in the Sumtsek
47 Vilāsavajra enumerates the NMAA maṇḍala deities in chapter five (associating each with
a “name” from the NS).
48 Namely Sattvavajrī, Ratnavajrī, Dharmavajrī and Karmavajrī. Although the core thirty-
seven deities of Vilāsavajra’s maṇḍala (i.e., discounting the sixteen bodhisattvas) are iden-
tical in name and number with those of the Tattvasaṃgraha, the order of emanation of
the four family mothers—also described in both texts as Perfections (pāramitā)—differs.
In the Tattvasaṃgraha they appear after the sixteen samādhi deities (and before the eight
offering goddesses and four door guardians). In the NMAA they comprise the initial mani-
festation of non-dual gnosis. Their more central position can perhaps be seen as reflecting
the changing status—the increasing centrality—of the feminine within tantric Buddhism
during this period.
49 Maitreya and Amoghadarśin are the first two members of the sixteen bhadrakalpa bo-
dhisattvas (followed by Apāyañjaha, Sarvaśokatamonirghātamati, Gandhahastin, Suraṅ-
gama, Gaganagañja, Jñānaketu, Amṛtaprabha, Candraprabha, Bhadrapāla, Jālinīprabha,
Vajragarbha, Akṣayamati, Pratibhānakūṭa, and Samantabhadra). The NMAA list is headed
by Maitreya and Mañjuśrī (followed by Gandhahastin, Jñānaketu, Bhadrapāla, Sāgaramati,
Akṣayamati, Pratibhānakūṭa, Mahāstāmaprāpta, Sarvāpāyañjaha, Sarvaśokatamonirghā-
tamati, Jālinīprabha, Candraprabha, Amitaprabha, Gaganagañja, and Sarvanivaraṇaviṣ-
kambhin). That Mañjuśrī is enumerated here underlines the notion that Vilāsavajra sees
Mañjuśrī-as-jñānasattva emanating Mañjuśrī-as-bodhisattva. The NMAA set is also found
in Abhayākaragupta’s forty-three deity Mañjuvajra-maṇḍala (Tricatvāriṃśadātmakamañ-
juvajra-maṇḍala: see NYĀ 50).
via free access
562 tribe
at Alchi. He also identified the outer eight figures as offering goddesses. This
seventeen-figure maṇḍala could be derived from Vilāsavajra’s NMAA maṇḍala,
as an abbreviated version. Alternatively, the two rows of mural figures painted
below the Ādibuddha, which flank Mahāvairocana may be part of a larger maṇ-
ḍala.50 Whether the maṇḍala, larger or smaller, is described more precisely by
Agrabodhi or Vilāsavajra requires further investigation.
While the Alchi and Lalung figures each have a maṇḍala as their immedi-
ate iconographic context—a context that makes sense of, and supports, their
identification as Mañjuśrī Ādibuddha—this is not the case for the two figures
at Mangyu, neither of which is at the centre of a maṇḍala. In the Two-armed
Maitreya Chapel the seated Ādibuddha figure has no especial prominence—
being one among ten deities that flank the large two-armed clay statue of
Maitreya, five on each side. The iconographic programme of these figures is
unclear, as they do not appear to constitute a maṇḍala. Among them are also
two additional Mañjuśrī figures, one of which is a six-headed, six-armed, white
Mañjuśrī at the same level as the Ādibuddha figure, and on the other side of the
Maitreya statue.51 The iconographic situation of the standing Ādibuddha figure
in the Village Stūpa is not dissimilar. The stūpa contains three other standing
mural figures, those of Prajñāpāramitā, Avalokiteśvara and Tārā. These four fig-
ures, each of the same size, flank, in two pairs, two clay statues placed on the
main axis of the stūpa, one at each end. There are eight further murals on the
side walls (that also flank two clay statues), each of which is half the size of the
standing figure murals. The Mañjuśrī Ādibuddha figure is prominent, therefore,
but not especially so. It appears to be of equivalent status to the other three
standing figures depicted as murals. It is not clear how to read the apparent
lack of high status accorded to these two Mañjuśrī figures at Mangyu. It seems
possible that they were not understood as depictions of the Ādibuddha—and
this despite their textual context, and in spite of their iconographic context in
50 Study of these two rows of repainted figures might clarify whether they are part of the
maṇḍala or not. A combination of sculptures and murals is found elsewhere: for exam-
ple, in the Translator’s Temple at Nako subsidiary deities of the Vajradhātu-maṇḍala are
murals while the five tathāgatas are sculptures (Luczanits 2004, 79–80).
51 This intriguing figure, which as far as I am aware has not been identified, has six arms: the
principal pair in his lap in dhyāna-mudrā; an upper pair with an arrow (proper right) and
bow (left), and a lower pair holding a lily (right) and a lotus flower (left). The six faces, in
two rows of three, have two central white faces. The remaining four have the colours of the
directions. The correspondence of the number of heads with the Mañjuśrī-jñānasattva as
described by our commentators is suggestive. Could this be a Mañjuśrī-jñānasattva vari-
ant, or even an Ādibuddha variant, perhaps from another Nāmasaṃgīti commentarial
tradition?
via free access
the Alchi Sumtsek and the Lalung Serkhang. Further work on the iconographic
programmes at Mangyu might clarify their roles.52
5 Addendum: The Relationship between Agrabodhi’s

*Sādhanaupayika and Vilāsavajra’s Nāmamantrārthāvalokinī
Agrabodhi’s description of Mañjuśrī-jñānasattva, using images of the moon, a

sapphire and the sun (“He is luminous like the autumn moon, with the choicest
sapphire as a [head] crest. He is encircled as if by the light of the rising sun.”), is
very close to that of Vilāsavajra, close enough for one to be borrowed from the
other. What makes the comparison more striking, and the likelihood of bor-
rowing more certain, is that the three descriptions are also allusions—more or
less quotations—to three Nāmasaṃgīti verse quarters, which are taken in the
same order by both authors, but which are not in the order of the Nāmasaṃgīti.
The three pādas are śaraccandrāṃśusuprabhaḥ (NS 125b), indranīlāgrasac-
cīraḥ (NS 126a), and bālārkamaṇḍalacchāyaḥ (NS 125c). Vilāsavajra’s NMAA
(see above) has śaracchaśāṅkaprabham indranīlāgrasaccīraṃ bālārkamaṇḍa-
lacchāyaprabhāmaṇḍalaṃ. The second description, except for the change in
case ending, quotes NS 126a. Both authors’ descriptions also immediately follow
a statement that Mañjuśrī-jñānasattva is six-faced. If we compare Agrabodhi’s
descriptions with the Tibetan translation of the NMAA their closeness becomes
particularly clear:
| | ston ka’i zla ba ltar gsal zhing | | indra nīla’i gtsug phud can | | nyi ma ’char
ka’i ’od ’dras bskor | |
Agrabodhi
| ston ka’i zla ba’i mdog can | indra nīla mtshog gi gtsug phud [|] nyi ma ’char
ka’i ltar ’od kyi dkyil ’khor gyis bskor ba |
Vilāsavajra. D 2533, 39r1
52 Rather than exploring in any depth the iconography surrounding the two Mangyu figures,
my goal in this paragraph has been to focus on the apparent oddness of their location, on
the assumption, that is, that they should be identified as representations of Mañjuśrī as
the Ādibuddha. Rather more has been published on the Village Stūpa than the Two-armed
Maitreya Chapel. On the latter, see Luczanits 2004, 167–170; van Ham 2011, 42–55. On the
former, see Linrothe 1994 and 1999, 173–174; Luczanits ibid., 170–174; van Ham ibid., 138–
158. Luczanits usefully comments that the square Village Stūpa has a main axis. Linrothe’s
1994 article unfortunately came to my attention too late for me to consult.
via free access
564 tribe
Alone, this passage does not provide sufficient evidence to establish a direc-
tion of borrowing. Either could be an expansion or contraction of the other.
Vilāsavajra is the more descriptively elaborate of the two. Agrabodhi’s text is
in a seven-syllable metrical form (which is the case for his descriptions of
the Ādibuddha as well as Mañjuśrī-jñānasattva), suggesting that the original
Sanskrit passage was likely in verse. Further comparison of Agrabodhi’s and
Vilāsavajra’s texts is required.
6 Conclusions
In summary, the Nāmasaṃgīti-related texts of Vilāsavajra, Mañjuśrīmitra, and

Agrabodhi indicate that the eight-armed Mañjuśrī[-like] figures at Alchi,
Mangyu and Lalung be identified as the Ādibuddha. Such an identification
explains why the Alchi figure occupies the most prominent location in the
Sumtsek. It makes sense of the organization of the murals of the topmost
storey, with the Ādibuddha representing the deeper nature of Vairocana to
his (proper) right and Prajñāpāramitā to his left. It also makes sense of the
vertical organization of the storeys of the back (and most important wall),
with the Ādibuddha vertically above, and again representing the underlying
nature of, both Śākyamuni (on the second storey) and Maitreya (the future
Buddha, whose large clay figure faces the entrance to the Sumtsek on the first
storey). However, only the five-headed Mangyu figures have the number of
heads described by our authors. The Alchi mural figure has four, and the clay
figure at Lalung just three (or possibly four). These variations, which may or
may not reflect local adaptation, remain to be accounted for.
Differences between the three textual descriptions examined both allow
and disallow the possibility of differing influences. Thus, for the two figures
at Mangyu, the colouring of their five faces points to a tradition associated
with Vilāsavajra as a source, and not Mañjuśrīmitra: Agrabodhi, who does not
describe the colours, is neither ruled in nor out. Also, the small clay figure above
the Ādibuddha in the Lalung Serkhang is likely to be Mañjuśrī-jñānasattva
as depicted by Agrabodhi or Vilāsavajra rather than Mañjuśrīmitra.53 Further,
53 Both Vilāsavajra’s NMAA and Agrabodhi’s *Sādhanaupayika were translated into Tibetan
in the early eleventh century by Smṛtijñānakīrti, who was working in eastern Tibet. It may
seem unlikely that their tradition of NS interpretation could be the source for the Lalung
Serkhang images. However, Luczanits (2004, 122) has argued that the sculptures in Cave
2 at nearby Dunkar reflect influence from north-east India (via Nepal and central Tibet),
via free access
if the Mahāvairocana mural beneath the clay Ādibuddha in the Serkhang is

part of the main maṇḍala, then together the three figures represent the tri-
adic and complex deity described in each of the three texts discussed, but
here as inflected by those of Agrabodhi and Vilāsavajra. Additionally, the evi-
dence suggests that the maṇḍalas surrounding the Alchi and Lalung figures
are related to the Vajradhātu-maṇḍala of the Tattvasaṃgraha rather than the
Dharmadhātuvāgīśvara-maṇḍala. A remaining puzzle concerns the position-
ing of the two Mangyu figures: their respective iconographic contexts raise
questions about whether they were intended as, or perceived to be, depictions
of Mañjuśrī as the Ādibuddha.
In wider perspective, the mid-eighth to early-ninth century Indian textual
portrayal of Mañjuśrī as the Ādibuddha, and as the Jñānasattva, by Mañjuśrī-
mitra, Agrabodhi, and Vilāsavajra (to place them in their likely chronological
order) reveals what is hard not to read as a concerted effort during this period
to promote Mañjuśrī as the key tantric deity. While the effort ultimately failed,
as a result of competing claims for the role of Ādibuddha (for example by sup-
porters of Vajrasattva, Vajradhara, and Samantabhadra) and by the increasing
dominance of the newer yoginītantra traditions, it is nonetheless significant
that this yogatantra-based vision of Mañjuśrī as the Ādibuddha still had cur-
rency some two to four hundred years later in the artistic portrayals found in
early Western Himalayan Buddhist art.
The findings of the present investigation also give clear support to the sug-
gestion, made by both Linrothe and Luczanits,54 of the importance of the
Nāmasaṃgīti and its associated literature in understanding the role of Mañju-
śrī in Western Himalayan Buddhist art.
and although he sees Lalung’s artistic origins as being in north-west India, perhaps some
textual/iconographic influence may have come from the east. If the Lalung Serkhang dates
from the second half of the twelfth century (see above, note 4) there would be ample time
for transmission of a Vilāsavajra/Agrabodhi based iconographical tradition. In the case of
Mañjuśrīmitra’s text there is no equivalent issue of geographical transmission. Rinchen
Zangpo played a major role in the development of Buddhism in the Western Himalayas,
becoming associated with many of its temples and monasteries, and although he did not
translate the Ākāśavimala, he did translate both the NS and Mañjuśrīmitra’s NS commen-
tary (see note 14 above).
54 See Linrothe 1996, 272 and Luczanits 2004, 212–214.
via free access
566 tribe
Abbreviations
D Catalogue of the Derge edition of the Kanjur and Tenjur published

by Tōhoku Imperial University. See Ui Hakuju et al. 1934.
NMAA Nāmamantrārthāvalokinī of Vilāsavajra
NGMPP Nepal-German Manuscript Preservation Project
NS Nāmasaṃgīti
WHAV Western Himalaya Archive Vienna
Acknowledgements
As my doctoral supervisor, Professor Sanderson gave of his time, enthusiasm,

deep knowledge and friendship with extraordinary generosity. I owe him a very
large debt of gratitude. For help on the present article thanks are due to Ver-
ena Widorn, director of the Western Himalaya Archive Vienna (WHAV) at the
University of Vienna, for providing invaluable photographs; to Cécile Ducher
for her overview of the Tibetan translations; and to Jaroslav Poncar, Christian
Luczanits and Christiane Papa-Kalantari for photograph permissions.
References
Primary Sources
Ākāśavimala of Mañjuśrīmitra
Nāmasaṃgītimaṇḍalavidhyākāśavimalanāma; Mtshan yang dag par brjod pa’i dkyil
’khor cho ga nam mkha’ dri med ces bya ba, trans. Suvajra pa and Chos kyi shes ra.
Sde-ge Bstan-gyur, Rgyud-’grel, vol. ngu, ff. 1v1–13v6 (D 2543).
Tattvasaṃgraha
Yamada Isshi, ed. Sarvatathāgata-tattvasaṃgrahanāma-mahāyānasūtra: a Critical Edi-
tion Based on a Sanskrit manuscript and Chinese and Tibetan Translations. Śatapiṭaka
Series, no. 262. New Delhi: Sharada Rani, 1981.
Nāmamantrārthāvalokinī (NMAA) of Vilāsavajra

See Tribe 2016.
Manuscript belonging to the University of Cambridge (Bendall, 1883: Add. 1708). Palm-
leaf, Nevārī script, 115 folios (c. 1450CE). (siglum A)
Manuscript belonging to P.R. Vajrācārya, Kathmandu (NGMPP reel no. E 360/16). Paper,
Nevārī script (first 13 folios lost), undated. (siglum B)
via free access
Manuscript belonging to Dharmavajrācārya, Kathmandu (NGMPP reel no. E 1369/3).

Paper, Devanāgarī script, 194 folios, undated. (siglum D)
Āryanāmasaṃgītiṭīkā Nāmamantrārthāvalokini; ’Phags pa mtshan yang dag par brjod
pa’i rgya cher ’grel pa mtshan gsang sngags kyi don du rnam par lta ba, trans. Smṛtij-
ñānakīrti. Sde-ge Bstan-gyur, Rgyud-’grel, vol. khu, ff. 27v1–115v3 (D 2533).
Nāmasaṃgīti (NS)
See Davidson 1981.
Niṣpannayogāvalī (NYĀ) of Abhayākaragupta

Benoytosh Bhattacharyya, ed. Niṣpannayogāvalī of Mahāpaṇḍita Abhayākaragupta.
Gaekwad’s Oriental Series, no. 109. Baroda: Oriental Institute, 1949.
*Sādhanaupayika of Agrabodhi
Āryamañ juśrīnāmasaṃgītinopika; ’Phags pa ’jam dpal gyi mtshan yang dag par brjod
pa’i sgrub pa’i thabs, trans. Smṛtijñânakîrti. Sde-ge Bstan-gyur, Rgyud-’grel, vol. ngu,
ff. 59r4–70v2 (D 2579).
Secondary Sources
Bendall, Cecil. 1883. Catalogue of the Buddhist Sanskrit Manuscripts in the University
Library, Cambridge. Cambridge: Cambridge University Press.
Davidson, Ronald M. 1981. “The Litany of Names of Mañjuśrī.” In Tantric and Taoist
Studies in Honour of Professor R.A. Stein, vol. 1, edited by Michel Strickmann, 1–69.
Mélanges chinois et bouddhiques 20. Brussels: Institut belge des hautes études chi-
noises.
Edgerton, Franklin. 1953. Buddhist Hybrid Sanskrit Grammar and Dictionary. 2 vols. New
Haven: Yale University Press.
Goepper, Roger, and Jaroslav Poncar. 1996. Alchi: Ladakh’s Hidden Buddhist Sanctuary.
The Sumtsek. Chicago: Serindia Publications.
Levy, Rachel Q., and Luke A. Fidler. 2014. “The Date of the Alchi Sumtsek Murals: 11th
or 13th Century?” International Association for Ladakh Studies 31: 37–38.
Linrothe, Robert. 1994. “The Murals of Mangyu. A Distillation of Mature Esoteric Bud-
dhist Iconography.” Orientations 25 (11): 92–102.
Linrothe, Robert. 1996. “Mapping the Iconographic Programme of the Sumstek.” In
Alchi: Ladakh’s Hidden Buddhist Sanctuary. The Sumtsek, by Roger Goepper and
Jaroslav Poncar, 269–272. Chicago: Serindia Publications.
Linrothe, Robert. 1999. Ruthless Compassion. Wrathful Deities in Early Indo-Tibetan Eso-
teric Buddhist Art. London: Serindia.
Linrothe, Robert. 2011. “Dating the Mangyu Style.” In Heavenly Himalayas: the Murals
of Mangyu and Other Discoveries in Ladakh, by Peter van Ham, 158–161. Munich and
New York: Prestel.
via free access
568 tribe
Luczanits, Christian. 2004. Buddhist Sculpture in Clay: Early Western Himalayan Art,
Late 10th to Early 13th Centuries. Chicago: Serindia.
Pal, Pratapaditya, and Lionel Fournier. 1988. Marvels of Buddhist Art: Alchi-Ladakh. New
York and Paris: Ravi Kumar.
Ricca, Franco, and Erberto Lo Bue. 1993. The Great Stupa of Gyantse: A Complete Tibetan
Pantheon of the Fifteenth Century. London: Serindia.
Snellgrove, David L., and Tadeusz Skorupski. 1977, 1980. The Cultural Heritage of Ladakh.
2 vols. Warminster: Aris and Phillips.
Tanaka Kimiaki. 2018. An Illustrated History of the Maṇḍala: From its Genesis to the Kāla-
cakratantra. Somerville, MA: Wisdom Publications.
Tribe, Anthony. 1997. “Mañjuśrī and ‘The Chanting of Names’ (Nāmasaṃgīti): Wis-
dom and its Embodiment in an Indian Mahāyāna Buddhist Text.” In Indian Insights:
Buddhism, Brahmanism and Bhakti. Papers from the Annual Spalding Symposium on
Indian Religions, edited by Peter Connolly and Susan Hamilton, 109–136. London:
Luzac Oriental.
Tribe, Anthony. 2015. “Mañjuśrīnāmasaṃgīti.” In Brill’s Encylopedia of Buddhism, vol. 1:
Literature and Languages, edited by Jonathan Silk, Oskar von Hinüber, and Vincent
Eltschinger, 353–359. Leiden: Brill.
Tribe, Anthony. 2016. Tantric Buddhist Practice in India: Vilāsavajra’s Commentary on
the Mañjuśrī-nāmasaṃgīti. A Critical Edition and Annotated Translation of Chapters
1–5 with Introductions. London and New York: Routledge.
Tucci, Giuseppe. 1988. The Temples of Western Tibet and their Artistic Symbolism. Indo-
Tibetica III.1: The Monasteries of Spiti and Kunavar. English translation, edited by
Lokesh Chandra. Śatapiṭaka Series, no. 349, New Delhi: Aditya Prakashan. Originally
published 1935. I templi del Tibet occidentale e il loro simbolismo artistico. Parte 1. Spiti
e Kunavar. Roma: Reale Accademia d’Italia.
Ui Hakuju, Suzuki Munetada, Kanakura Yensho, and Tada Tokan, eds. 1934. A Complete
Catalogue of the Tibetan Buddhist Canons (Bkaḥ-ḥgyur and Bstan-ḥgyur). Sendai:
Tōhoku Imperial University.
van Ham, Peter. 2011. Heavenly Himalayas: the Murals of Mangyu and Other Discoveries
in Ladakh. Munich and New York: Prestel.
van Ham, Peter. 2019. Alchi: Treasure of the Himalayas. Munich: Hirmer.
van Schaik, Sam. 2009. “A Tibetan Book of Spells.” Blog entry, https://earlytibet.com/
2009/02/19/a‑tibetan‑book‑of‑spells/ (accessed May 15th 2020).
via free access
chapter 23
Life and Afterlife of Sādṛśya: Revisiting the

Citrasūtra through the Nationalism-Naturalism
Debate in Indian Art History
Parul Dave-Mukherji
This paper sets out to revisit the Citrasūtra, a seminal section on painting from
the Viṣṇudharmottarapurāṇa, in the light of key concerns around the cultural
politics of art historiography, the śāstra-prayoga debate (Maxwell 1989, 5–15),
and the related question of interpretative frameworks for studying early Indian
art. The latter concern has lately come to the forefront in the context of post-
colonial studies and global art history. It is critical of intellectual parasitism
(Dhareshwar 2015, 57–77) and pushes postcolonial thought to explore ‘native’
interpretative frames to study Indian art (Asher 2007, 12).
This paper attempts to complicate the search for alternative frameworks by
underlining gaps and slippages that surround the meaning of terms in a given
text and their modern appropriations. To this end, it traces the genealogy of
the term sādṛśya, from the śilpaśāstric lexicon through its twentieth-century
reception in art-historical discourse. How does a term acquire an afterlife when
it enters into the force field of reinterpretation steeped in cultural nationalism?
How could a newly “discovered” Sanskrit text function in such a space?1
In this paper, I also intend to address the larger question: what is the geneal-
ogy of the view of India’s cultural past, and specifically its “art,” as transcen-
dental/ idealistic/spiritual, which has translated itself into a belief? And why
does this belief persist, although in different configurations? In more recent
times, an ethnographic approach to the study of texts has emerged as a cor-
rective, which I will critically examine for its relevance for alternative inter-
pretative frames for the study of Indian art. In the end, I will conclude by
relating Coomaraswamy’s transcendentalism to David Shulman’s recent dis-
1 The first printed text of the Viṣṇudharmottara Purāṇa, edited by Pandit Madhusudhan and
Madhavaprasada Sarma in 1912 (Venkateshwar Press, Bombay), is the one that caught the
attention of a pioneering art historian, Stella Kramrisch, who had arrived in India from
Vienna. It was her English translation which brought the text into the discourse of art his-
tory and Indology.
© parul dave-mukherji, 2020 | doi:10.1163/9789004432802_025

via free access
570 life and afterlife of sādṛśya
course around the ‘more than real’ (Shulman 2012), and the latter’s implication
for interpretative frameworks for Indian art.
1 The Discovery of the Citrasūtra
It is around the first quarter of the twentieth century that some major textual
sources, either complete or as fragments, were “discovered,” edited and trans-
lated. They began to acquire tremendous cultural significance as carriers of
authentic meaning. One such text was the Citrasūtra of the Viṣṇudharmottara,
which emerged on the stage of art history in India when it was first translated
into English in 1924 by Stella Kramrisch, the pioneering historian of Indian art.
This art historian from Vienna chanced upon this text soon after her arrival
in India in 1919. The first printed text of the Viṣṇudharmottara Purāṇa, edited
by Pandit Madhusudhan and Madhavaprasada Sarma and published by the
Venkateshwar Press, Bombay in 1912, is the one that caught attention of Kram-
risch. It was her English translation (Kramrisch 1928 [1924]) which brought the
text into the discourse of art history and Indology.
It was almost a decade later that another pioneering art historian, A.K. Coo-
maraswamy, turned his attention to this text, singling out one of its chapters,
adhyāya 41, for translation and commentary (Coomaraswamy 1933, 13–21). This
chapter of the text deals with the classification of painting into four types—
Satya, Vaiṇika, Nāgara and Miśra. According to Coomaraswamy, the first two
types corresponded to the pictorial tradition of the Ajanta caves.
By now, the two most eminent art historians of Indian art who played a semi-
nal role in establishing the discipline of art history in India were involved in the
interpretation of the Citrasūtra. This, in turn, exalted the status of the text and
transformed it into an Ur-text for a wide-ranging extrapolation about Indian art
and aesthetics, which continued into the first decade of the 21st century (Nardi
2006).
Almost twenty-five years separate Coomaraswamy’s commentary on the text
and its first critical edition, brought out by a Sanskritist, Priyabala Shah in
1958 and 1961. Shah’s edition broke fresh grounds in textual criticism when she
incorporated the readings from six new manuscripts. However, her attempt
at theorization was restricted, interestingly, to the same chapter selected by
Coomaraswamy, which dealt with the classification of paintings. She was far
too involved with connecting the types of paintings with types of architectural
styles to pay attention to whether Coomaraswamy’s metaphysical readings of
the terminology were borne out by the edited text. Exactly two decades later,
interest in this text was renewed when another art historian, C. Sivaramamurti
via free access
dave-mukherji 571
published a translation and commentary of all nine chapters of the Citrasūtra

(1978). A Sanskritist and an art historian of post-independent India, Sivarama-
murti assigned a diametrically opposite significance to the text, one validating
“Indian naturalism” as opposed to the manner in which Coomaraswamy used
this text as a support for his claims of transcendentalism during the colonial
period.
2 The Citrasūtra Turn in Early Indian Art History
The interpretation of the Citrasūtra is further complicated by the fact that it

was “discovered” in colonial India; this implied that its interpretation would be
caught in a comparative framework involving the art of the colonizer and that
of the colonized. The central question around which the art of India and that
of the West were compared and judged was that of ‘naturalism,’ a term that was
seen as intrinsic to the identity of western art.
When the Citrasūtra came to light in the early decades of the twentieth
century, cultural nationalism was at its height and art history as a discipline
was being introduced into Indian universities. Naturalism supplied a key term,
often as an antithesis to ornamentalism, around which debates on the worth of
Indian art and craft were being carried out amidst the ascendency of cultural
nationalism. Transcendentalism and naturalism are dialectically connected as
part of a discourse largely constructed by A.K. Coomaraswamy to place Indian
art on a morally higher plane. The former stands for Indian art and its spiritual
interiority, defined in contrast to the naturalism ascribed to Western art. There
prevailed a climate of contestation rife with orientalist binaries that pitted the
‘rational,’ ‘scientific’ West with the ‘mystical,’ ‘irrational’ India (Masson-Oursel
1925).2 If Indian art is believed to be the result of the artists’ power of medita-
tion, Western art is assumed to capture only the surface of nature or the visible
world but never its inner essence. Hence the imbrication of the naturalism
debate within nationalism. In the battle between the superiority of naturalism
in art as a hallmark of Western control over representation and its rejection by
2 Paul Masson-Oursel, a French Orientalist, exerted a strong influence on Coomaraswamy. It

is to be noted that Coomaraswamy found strong support for his views in Masson-Oursel’s
writings and in fact translated Masson’s 1925 article “Une connexion entre l’esthétique et la
philosophie de l’ Inde: la notion de pramâna” (Review des Arts Asiatique 2, 1: 6–9) into English
as “A Connection between Indian Aesthetics and Philosophy in India” (Rupam 27 & 28 [1926]:
91–94).
via free access
Indian nationalists for the higher civilizational ideal of transcendentalism, the

latter won the day. Its victory placed Coomaraswamy at an advantageous posi-
tion, and today this discourse has come to assume truth value, constituting a
common sensibility concerning Indian art (Dehejia 1996).
For transcendentalism to emerge as an effective discourse to counter colo-
nial presuppositions about Indian art, such as the denial of the existence of fine
arts in India, an alliance of Art History with Indology was the need of the day.
Already by the middle of eighteenth century, the discipline of Indology had
been founded, having as its main objective the study of Indian culture through
texts. Indology, right from its inaugural moment, operated with the assump-
tion that it was ancient and classical Indian texts that were the most authentic
means of recovering India’s past (Halbfass 1981).
Around the early 1920s, art history came to be introduced in an Indian
university when the Vienna-trained art historian, Stella Kramrisch, began to
teach this discipline in Kalabhavana, Visvabharati University, Santiniketan, at
Rabindranath Tagore’s behest. In other universities where it was taught, such
as Calcutta University and Banaras Hindu University, it was as an adjunct to
ancient Indian history. It was in 1950s that an independent department of Art
History and Aesthetics came to be established in the Faculty of Fine Arts, MSU,
Baroda. This discipline was mobilized for the reconstruction of India’s past,
part of the nationalist project to regain the lost essence of Indian-ness as a
means of reclaiming subjectivity (Guha-Thakurta 1991).
3 Coomaraswamy’s Reading of the Citrasūtra: A Founding Moment

in Indian Art History
So strong was Coomaraswamy’s commitment to transcendentalism that it im-

pelled him to interpret the traditional śilpaśāstras through the same lens. I have
discussed his negotiation with the anomaly posed by the śilpaśāstras elsewhere
(Dave-Mukherji 2008, 132–134). Let us see how the term sādṛśya gets inflected
by his translation. As a case in point, let us turn to Coomaraswamy’s transla-
tion of adhyāya 41 of the Citrasūtra, which primarily defines four categories of
painting: Satya, Deśika, Nāgara and Miśra. He particularly focused on the first
category, as it seemed to pose a challenge to his assumption that there is no
place for naturalism in Indian art. In fact, the very term Satya may be translated
as “real” or “naturalistic” in light of its definition—that any painting bearing
resemblance to the world belongs to this category:
yat kiñcil lokasādṛśyaṃ citraṃ tat satyam ucyate
via free access
dave-mukherji 573
I have translated this as follows (Dave-Mukherji 2001, 159):
Whichever painting that bears a similarity with the world [that painting]
is called Satya (“Naturalistic”).
This is in fact in wide variance with Coomaraswamy’s translation (1933, 13):
Painting that represents any of the worlds (kiñcilloka-sādṛśya) … is called

Pure or Sacred (satya).
How satya comes to translated as “Pure or Sacred” is not a matter of willful

mistranslation. Rather, we need to historicize this mode of interpretation and
ask, under what conditions may satya be taken as “Pure or Sacred”? Satya could
no longer be accepted as “True to life”, as interpreted by Kramrisch (1928, 51).
Another term, which intensified this problem, was sādṛśya, which could not be
allowed to retain its usual sense (Coomaraswamy 1933, 21):
If we understood sādṛśya then to mean “illusion” or “realism,” verisimil-

itude of any crude or naive sort, we should be contradicting all that we
know of the oriental conception of art.
It was the co-existence of satya with sādṛśya (resemblance/likeness) which

compounded Coomarsawamy’s problem of avoiding literal translation. Hence
Coomaraswamy was compelled to connect yatkiñcilloka- with sādṛśyam,
against the rules of Sanskrit syntax, in order to yield the meaning “any of the
worlds” in place of “any painting.” His translation (Coomaraswamy 1933, 13)
continues as follows:
Painting that represents any of the worlds (yatkiñcillokasādṛśya), that is

elongated and has ideal proportions … [emphasis added]
In fact, the description of the painting as “elongated” (dīrghāṅgam) and hav-

ing “ideal proportion” (supramāṇam) belongs to the next category of painting,
termed Deśika. By misreading in this manner, he could explain away the natu-
ralistic potential in the term sādṛśya and impose his view of idealism being the
essential feature of Indian painting (Coomaraswamy 1933, 13):
So also in verse 2 above, where kiñcilloka must include devaloka [“abode

of the gods”], sādṛśya cannot be interpreted as “naturalistic.”
via free access
Coomaraswamy’s departure from normal Sanskrit syntax and semantics did

not go unnoticed, but invited ridicule from his contemporary, a well-respected
Sanskrit scholar (Raghavan 1933, 905):3
Satya cannot be interpreted as a Sattvika picture. One can as well derive

it from Sat and say it is the picture of the Upanisadic Brahman.
In response to Raghvan’s criticism, Coomaraswamy revised his translation of

satya from “Pure or Sacred” to “intellectual and ideal” (Coomaraswamy 1933, 26),
and rejected its literal meaning, which was however subsequently reinstated by
Priyabala Shah (1961, 120) in the first critical edition of this text:
Whatever ( yatkiñcid) painting depicts semblance of the world (Loka-

sādṛśyam citram) is called Satya.
Coomaraswamy evidently felt in need of more support for his elimination

of the concept of resemblance than the Citrasūtra could provide. He began
to look to the idealist schools of Upanisadic philosophy and epistemology
(Coomaraswamy 1933, 26) to arrive at a metaphysical reworking of the term
satya:
Satya is “real”, “actual”, “intelligible”, “ideal”; nāmarūpa, “form” as distinct

from the natural disorder (anṛta) of the sensible world (loka) (Bṛhadā-
raṇyaka Upaniṣad, 1, 6, 3: Taittiriya Upaniṣad, II, 6).
In his recasting of sādṛśya, Coomaraswamy turned to a rather unexpected

source for an alternative definition of imitation in art that could preserve the
term, but be free of the unwanted association of naturalism. He found what he
was looking for in western medieval sources. In my view, this detour to the west-
ern medieval sources was not fortuitous but compelled by Coomaraswamy’s
confrontation with the Citrasūtra. This is confirmed by evidence of the period
for his new interest, as given by his biographer, Roger Lipsey. According to
Lipsey (1977, 73), the beginning of the 1930’s marks the period of transformation
for Coomaraswamy: “… it is just at this point that the relation to medieval Chris-
tian thought changed from an enthusiastic but transient acknowledgement of
3 Raghavan further comments, “My impression on reading the Viṣṇudharmottara is that even to
its author the exact import of these names was not clear. The text seems to have been written
after a cut in the flow of tradition of the artists who were using these words as paribhāṣās.”
via free access
dave-mukherji 575
the value of Meister Eckhart in particular to a permanent and scholarly inter-

est in the entire scholastic tradition from its origins in the St Augustine to its
flowering in the thirteenth century.”
For Coomaraswamy, medieval Christian thought best illuminated how to
interpret sādṛśya without any ‘naturalistic’ connotations (1933, 26):
Now as to Visnudharmottara, III, 41, 1 and 9, kimcit is not “any” but “some-
what”; loka not here “worlds” but the sensible (not alaukika) aspect of the
world; sādṛśya is not “resemblance” but consonantia, adaequatio rei et
intellectus, …
Combining references to western medieval sources with the terms from Indian
epistemology, Coomaraswamy (1933, 26) arrives at a new definition of the satya
type of painting as
… sāhitya, sādhārāṇya, all of which have reference to unity self-contained

in art, and not to likeness (sādṛśatā) to a model. Kimcilloka-sādṛśya is then
“the unity of which is only somewhat as to the world, …”.
For Coomaraswamy, sādṛśya with its naturalistic implication and its coexis-
tence with satya, translated as “Pure,” was contradictory. Here, purity is taken
as a “unity self-contained in art,” and for that reason the potential of referen-
tiality to the world outside implied by the term sādṛśya was viewed as posing a
threat to his construct of art as the realm of the mind. In another instance, he
qualified this term as “a consent (sādṛśya) of pictorial and formal elements in
the substance (śarīra) and essence (ātman) of the work” (Coomaraswamy 1933,
27). While he conceded to some presence of pictorial realism in the other three
types of painting, it is the first type or Satya through which the ethical notion
of art as “Truth” could be admitted as that which had to be elevated beyond any
reference to this realism (ibid.):
… and this contrasts with vaiṇika and nāgara painting in which it is to be

understood that the realistic, pictorial (sādṛśa) element is much greater,
where accordingly an adaequatio (tadākāratā) rei et intellectus, sādṛśya,
is only partially attainable.
Notice a distinction created between sadṛśa (which allows an element of pic-

torial realism) and sādṛśya (a consent of pictorial and formal elements) was
crucial for his interpretation to avoid contradicting “all that we know of the
oriental conceptions of art.” Rather than unpacking “the oriental conceptions
via free access
of art,” Coomaraswamy ended up congealing the old binaries between a ratio-

nal occident and a spiritual orient. His ‘cultural unconscious’ is betrayed in a
footnote in which he claims to have captured the voice of tradition through his
alleged fidelity to the textual sources (Coomaraswamy 1933, 27).
What follows is derived from all these sources without the addition of any
thought or phrase of my own. Verbal authority could be cited for any state-
ment.
What relevance this kind of interrogation has for us today, this reaching back to
the cultural politics of the early-twentieth century, also needs to be addressed.
The transcendentalist claim made by Coomaraswamy on the basis of a “found”
textual tradition can be explained as a political exigency for combating the
colonial representation of Indian art. However, when such reliance on ‘scrip-
tural evidence’ is perpetuated in postcolonial times, its inbuilt essentialism
can have serious implications today in an India that is witnessing an unprece-
dented obsession around nationalism.
Foregrounding the claim that rejection of naturalism by Indian artists was
intentional, Coomaraswamy effectively undercut the criticism that Indian art
was deficient in naturalism. He suggested, instead, that it was never the inten-
tion of the Indian artists to imitate the visible world but to create a symbolic
image based on a supramundane ideal which transcended the world of appear-
ance. The charge leveled against Indian art of crude execution and a lack of
anatomical accuracy (Ruskin 1905, 347) in the rendering of human and animal
form was answered not by questioning the criteria of execution and cultural
knowledge of bodies, but by claiming that these so-called “deficiencies” were
deliberate, a result of a “specialized technique of vision” (Coomaraswamy 1934,
166):
Technical production is thus bound up with the psychological method

known as yoga. In other words, the artist does not resort to models but
uses a mental construction and this condition sufficiently explains the
cerebral character of the art, …
The vision of the classical Indian artist that emerges from Coomaraswamy’s
early writings appears to be modelled upon the idea of sculpture of dhyāni
Buddhas, with artistic activity relegated to the domain of pure mental contem-
plation (Coomaraswamy 1908). The eyes of the artist appear to be visualized
as half-closed, focused more on the inward, contemplative space of the mind
and less on the world “out there,” the domain of naturalism. With an ingenious
via free access
dave-mukherji 577
reversal of binaries set up by the colonial critics, Coomaraswamy constructed a

compelling counter-narrative, but in the process, he reproduced the very colo-
nialist logic he had set out to resist.
Certainly, Coomaraswamy’s transcendentalist claims for Indian art and aes-
thetics did not remain uncontested, as demonstrated by V. Raghavan’s (1933)
criticism of Coomaraswamy’s interpretation and by the translators of the Citra-
sūtra. However, such voices of dissent were relegated to the margins and rarely
made an impact on the disciplinary framework of Indian art history, where
Coomaraswamy’s theoretical framework has retained a canonical status. Most
curiously, Kapila Vatsyayan, in her very Foreword to my edition of the Citra-
sūtra has cited the same verse that I have used to critique Coomarawamy’s
transcendentalist standpoint in order to justify it. The tenacity of this view can
be gauged by the manner in which Coomaraswamy’s misreading of pratyakṣa,
direct perception, as the least important pramāṇa or source of knowledge in
Indian epistemology, is ratified by Vatsyayan (Dave-Mukherji 2001, xiii–xiv). In
fact, all the six classical darśanas of Indian philosophy consider pratyakṣa the
most salient source of knowledge (Matilal 1986).
4 The Citrasūtra in Postcolonial Art Historiography:

C. Sivaramamurti
The return of interest in the Citrasūtra in the work of C. Sivaramamurti in

the late 1970s signals a re-appropriation of this text within his agenda of cul-
tural nationalism. Ostensibly, his turn to this text was to eradicate false read-
ings and restore it closer to its original state (Sivaramamurti 1978, xv). Under-
scoring his interpretation of the Citrasūtra was the claim that naturalism was
accorded a positive role both in Indian art and in the text, which resonated
with the stand taken by him more than four decades earlier (Sivaramamurti
1934, 189):
Questions of perspective and foreshortening do not appear to have baf-

fled the old artists of our land as in evident from their talks on such tech-
nical details as kṣayavṛddhi and the actual conformance of the pictures to
those rules so elaborately discussed in books on theory.
It is largely the presence of terms such as sādṛśya and kṣayavṛddhi (literally

“decrease and increase,” i.e. “foreshortening,” a technique used to show appar-
ent depth in naturalistic art) that indicated to Sivaramamurti that the tradi-
tional Indian artist was no less proficient in representing the world “naturalis-
via free access
tically” than his western counterpart. In a similar move, he looks beyond the
context of the śilpaśāstras, examining classical Sanskrit literature, and finds
further evidence of imitation in art, as for example, in the Mṛcchakaṭikā (Sivara-
mamurti 1934, 189):
The tendency of imitation in an artist given in the Mṛchhakațikā is healthy

sign of progress at a particular stage provided it is directed in the right
direction.
Thus by placing the śilpaśāstras in this wider context, Sivaramamurti retrieved

key terms such as sādṛśya, anukṛti and kṣayavṛddhi from the framework of tran-
scendentalism. Sivaramamurti reinstated naturalism in post-colonial India as
an index of ‘progress,’ as proof that India could measure up, and more, to the
categories and standards of art set up in and by the West. Indeed, he writes
naturalism back into the art-historiographical demands of post-colonial India,
mapping the term powerfully onto nation building.
For me, the rejection of naturalism in Indian art on grounds of its western
identity is as problematic as its nativist retrieval in Sivaramamurti’s interpre-
tation of the Citrasūtra. Interestingly, Sivaramamurti’s espousal of an Indian
naturalism met with deep skepticism from Stella Kramrisch, who was invited
to write the foreword by the author (Sivaramamurti 1934, 10):
If as the text shows and Dr. Sivaramamurti stresses, realism was a main
consideration with the painters, their criteria of verisimilitude were, no
doubt, met in practice, although no object painted in the murals of Ajan-
ta, which are roughly contemporary with the Chitrasūtra would strike a
spectator today as being realistically painted. The realism is in the eye of
the beholder and pious stories told, though not in the Chitrasūtra …
Kramrisch almost poses as a paradox the gap between the ancient painters’
intentions to accomplish realism and their actual practice, and argues that
the premodern criteria of representation do not match up with our modern
expectations of verisimilitude. Today, it is possible to see an overlap between
Kramrisch’s skepticism about Sivaramamurti’s literal reading of the mimetic
terms in the Citrasūtra and David Shulman’s (2012) recent problematization of
‘the real’ in the context of his study of the South Indian models of mind.
In his reading of a seventh-century playwright, Harṣadeva’s classical Sanskrit
drama, Ratnāvalī, Shulman foregrounds the role of painting in the way the plot
unfolds. About the question of verisimilitude, which emerges when the char-
acters recognize each other in their respective portraits, Shulman arrives at a
via free access
dave-mukherji 579
paradox similar to Kramrisch’s, but resolves it by stressing the realm of imagi-

nation (Shulman 2012, 28):
Apparently, the painted image is recognized not so much by virtue of how

it is painted but mostly as a prop for projection, …
At one stroke, if Kramrisch evacuates realism/naturalism (taken interchange-

ably in this context) from both the walls of Ajanta and the text of the Citrasūtra,
Shulman edges it from the painted surface to the space of literary imagination.
Naturalism or ‘the real’ is discursively edged off to the realm of the mind via
the “eyes of the beholder” and the physical entity of a painting reduced to a
“prop for projection.” Despite her reservations about claims of verisimilitude
in the Citrasūtra, Kramrisch had earlier accepted cultural specificity in visual
representation. This view implicitly allowed for “Indian naturalism,” even if her
example draws from Chinese and European naturalism (Kramrisch 1922, 25):
The Chinese naturalism to the European eye appears as an idealistic

abstraction, while the European naturalism strikes the Chinese as ugli-
ness.
However, it is instructive to recognize the difference between Kramrisch and

Shulman’s skepticism about the claims of the real. Kramrisch, to some extent,
was able to accommodate naturalism as a culturally specific phenomenon that
manifested in art theory, if not in practice, while Shulman relegates the ques-
tion of painterly resemblance solely to the domain of imagination. However,
such a categorical dismissal of verisimilitude assumes a wide gap not only
between discourse and practice but also between how the Indian mind thinks,
how the Indian eyes see and how Indian bodies act; thinking, vision and agency
are assumed to fall outside of any rational intentionality.
What needs examination is the persistence of a belief that ancient Indian
art willfully rejected any empirical interest in the world that surrounded the
artists. In my view, the theoretical framework of what this belief (i.e. in tran-
scendentalism) was once a part of has faded into the background, turning the
belief and the problematic produced by that theory into a widely-shared yet
questionable sensibility concerning India’s past and art traditions. Rather than
a simplistic retrieval of an “Indian naturalism,” as attempted by Sivaramamurti,
the very semantic framework of the real operating in the śilpaśāstras and liter-
ary texts needs to be unpacked and thought through.
via free access
5 The Śāstra-Prayoga Debate
Since the inception of art history in India in the late eighteenth century, the
core of the discipline came to be defined by a “detailed study of archeological
evidence using a stylistic approach” (Maxwell 1989, 5). Despite the momen-
tous “discovery” of the śilpaśāstras inaugurated by the coming to light of the
Mānasāra in 1834, and by subsequent work on the Citralakṣaṇa, Citrasūtra,
Mānasollāsa and other sources around the early-twentieth century, these have
still not been adequately explored for developing interpretative frameworks
of art history in India. As pointed out by Tapati Guha Thakurta (1991, 170) in
the context of Bengal, a sharp divide arose between history and aesthetics, as
between archeology and the study of śilpa texts, when modern commentaries
on ancient texts came to be written by the late-nineteenth and early-twentieth
centuries:4
Art history and aesthetics, even as they grew out of the same institu-
tionalized scholarly sphere, would henceforth always be marked out as
a discretely differentiated field.
The separation of aesthetics and archeology got mapped onto the śāstra (the-
ory) and prayoga (practice) divide, and led to a division of labour in which the
question of meaning and interpretation was seen as the realm of specialized
Sankrit-based scholarship dissociated from the empirical study of artifacts,
with the latter thought to lie in the domain of archeology.
Reflecting on Indian intellectual history, Sheldon Pollock touches upon the
imbrication of power in the Indian intellectual tradition around śāstra-prayoga
(Pollock 1984, 499–519). Seldom has modern art history in India responded to
the issue of the caste hierarchy that has existed between the Brahmin authors
of the śilpaśāstras and the low caste śilpins or artisans who practiced art (Misra
1975). Kramrisch (1985, 61), in fact, was one of the first art historians to exam-
ine caste in the context of art practice and point out contradictory views about
labour and caste often held by the Dharmaśāstras:
4 Abanindranath Tagore, who spearheaded the Bengal School, wrote an article, “Shadangas or
The Six Limbs of Painting” in 1915. It claimed to base its interpretation of Indian aesthetics
on a verse culled from the Kāmasūtra of circa 500 CE. This elicited a vehement critique from
an archeologist, Akshay Kumar Maitreya, and illustrates the conflict between aesthetics and
archeology, as pointed out by Guha-Thakurta (1991).
via free access
dave-mukherji 581
Manu says that the hand of a craftsman engaged in his work is always ritu-
ally pure. The Gautama Dharmaśāstra postulates that a Brahman may not
accept food from an artisan. The law books thus distinguish the craftsman
in his social position on one hand, and in his state of grace on the other—
when he is engaged in his work, when he creates, and, thereby, gives effect
to his being an embodiment of Viśvakarma.
Despite Kramrisch’s attention to caste and its implication for the system of
patronage and art practice, the model created by Coomaraswamy of the śilpī
or an artist/artisan aligned with transcendentalism remained dominant. Even
Kramrisch later began to follow this line of thought (1985, 57):
As artists and magicians, the Indian craftsmen transmuted matter into

form, vision into concrete shape, and man beyond his earthy competence
… Charged with this power their work became concretely real, subtly
effective, and transcendentally existent ….
6 Reading Sādṛśya through Ethnography
Since the time of the modern discovery of the śilpaśāstras and particularly with
reference to texts on architecture, ethnography has emerged as a valid method
of interpreting difficult terminology. As early as Ram Raz’s attempt to make
sense of obsolete terminology around 1830s, recourse to extant practice was
sought as a way of checking the extent to which śāstric terms had purchase
with craftsmen. The question that I would like to raise here is the usefulness
of ethnography in the interpretation of terms that deal with the ‘resemblance’
or ‘truthfulness’ of visual representation. Where are we placed today in terms
of our understanding of this terminology, and to what extent is combining
ethnography with the study of śilpa texts useful for shedding light on its slip-
pery semantics?
Samuel Parker (2003, 5–34) has been at the forefront of embracing the
ethnographic approach in the study of texts mainly on architecture. Can this
method be equally illuminating in the study of the śilpaśāstras? Do critical
terms that relate to visual representation—anukṛti (imitation or mimesis),
sādṛśya (resemblance or verisimilitude), viddha (literally “pierced,” or captur-
ing resemblance), aviddha (that which lacks resemblance), etc.—hold reso-
nance for traditional idol makers? Here perhaps we need to distinguish be-
tween technical terms for parts of a building and terms that deal with aesthetics
(beauty) and visual representation. Terms that deal with complex semiotics
via free access
such as sādṛśya, beauty, etc., are too culturally loaded for them to travel well
across time. Perhaps in this regard it is instructive to turn to British art historian,
Michael Baxandall (1972, 26), who has brought to our notice how some art ter-
minologies common in fifteenth-century Italy, which had made perfect sense
to their contemporary public, resist easy comprehension for today’s audience.
Ethnography, as employed by some Indologists, assumes continuity of
meaning across time such that by recourse to extant practice by modern stha-
pati (traditional architects) the meanings of traditional treatises can be un-
locked. Terms such as sādṛśya are ensconced in a whole web of meanings and
practices of viewing that make them intelligible, and in the absence of the lat-
ter, they often give rise to skepticism around śilpaśāstras—as if what they say
fails to coincide with what they really mean.
Besides, do terms like sādṛśya or satya, no doubt used in classification of
paintings in terms of degrees of correspondence to the visible world as far as the
Citrasūtra is concerned, have the same valence as technical terms that classify
a building? Or, as demonstrated by the survey of the historiography of textual
interpretations, terms dealing with visual representation are seen by cultural
nationalists as too invested in civilizational identity and the marking of cul-
tural difference to yield an ‘objective’ meaning. More recently, Isabella Nardi
has ventured into this fraught zone and attempted to employ the ethnographic
approach to analyze how a present-day traditional idol maker engages with
visual representation and beauty. Nardi has interacted with local sculptors in
Orissa and Rajasthan to explore the relevance of the śilpaśāstras to their cur-
rent practice (Nardi 2006, 58–63).
In her conversation with a local sculptor, Ram Prasad Sharma, Nardi was
struck by Sharma’s comparison of the sculpture of Lakshmi with a European
Renaissance master, Giovanni Bellini’s Young Woman with a Mirror (painted in
1515), only to demonstrate the greater beauty of the former (2006, 60):
Applying his rules and ideals, the sculptor explains that the major defects
of this Renaissance figure are that the woman has a huge belly and very
big arms.
Likewise, Nardi (ibid.) attributes this sculptor’s failed attempt at copying

Michelangelo’s sculptures to “different vocabulary and aesthetic ideals of the
Indian sculptor and [the] Renaissance [sculptor].” Such an ethnographic exer-
cise, while shedding light on the gap between text and practice, serves to under-
line cultural difference and tends more towards truisms about how each cul-
tural practitioner responds to aesthetics through what is locally familiar than
to clarify meanings embedded in texts.
via free access
dave-mukherji 583
7 “Mistranslation” of Sanskrit Terms: A Paradigm for Rethinking

Visual Representation in Indian Art
There is no simple, pure unproblematic naturalism in the Indian tradition wait-

ing to be recuperated. The very fact that there is no one-to-one translation
possible in Sanskrit, for example, opens up an area of interrogation and also
a paradigm for interpretation. Perhaps we have neither the words nor the con-
cepts to capture the domain of “naturalism.”
Kapila Vatsyayan insightfully problematizes the question of naturalism in
the context of premodern India thus (Dave-Mukherji 2001, xii):
For a fuller examination, it would be necessary to reopen the question

of what constitutes as Nature in the Indian mind-scape, more precisely,
the mahābhūtas and prakṛti. This comprehension has to be clearly distin-
guished from the emergence of ‘naturalism’ as an art movement in Europe
in the 18th and 19th centuries.
Of course, the absence of a single term does not signify that the concept does
not exist. Rather, the absence itself can be seen as an important clue. Perhaps
what we have is a much more complex experience which can be described
by using a combination of sādṛśya, satya and anukṛti. In the same way, San-
skrit terms such as kṣayavṛddhi (literally, diminution and expansion) or vartanā
need not be directly taken as equivalents of “foreshortening” or “shading.” The
lack of a close fit between them and the gaps opening up as we attempt to map
one set of Sanskrit terms on another set of English terms may itself signify a
rich terrain to explore.
Once it is acknowledged that that no mode of visual representation in any
given culture in the West or elsewhere has a direct, unmediated and privileged
access to the visible, it makes futile any defensive and anxious search for an
“Indian naturalism.” In Art History, Norman Bryson (1983) has engaged with
a post-structuralist critique of representation and argued against any unmedi-
ated access to the real, a view that helps us to complicate the issue of naturalism
even in the context of premodern Indian art.
In more recent times and a context closer to South Asia, the question of ‘the
real’ has assumed new significance in the writings of David Shulman, who has
added a more specific dimension to this inquiry through his focus on South
Indian models of mind. I would conclude by juxtaposing Coomaraswamy’s
anti-naturalism with Shulman’s evocation of the ‘more than real.’ At one level,
one may discern a certain correspondence between Coomaraswamy’s rejection
of naturalism as alien to the Indian sensibility, a construct created under the
via free access
sway of cultural nationalism, and Shulman’s recognition of a different model

of mind for South Asia that allows for an interplay between the real and the
imagined. At the same time, it is possible to recognize a distinction between
their models. While Coomaraswamy’s view of naturalism arose out of a desire
to defend Indian art from colonial denigration, Shulman—despite his focus
on South India—complicates any simple binaries of the real and the imagi-
nary; this he does by refusing a separation between an observing subject and
observed real, instead posing their mutual permeability. Notwithstanding the
fact that Shulman arrives at this model through a close reading of literary texts,
it has important implications for notions of visual representation.
Once we overcome the defensive rejection of ‘the real’ and resituate the real
at the cusp of the existing and the imaginary, the sāstric terminology can be
accessed in new ways. The term vāyugati is a case in point. I would like to
express my gratitude to Alexis Sanderson for pointing out the ‘literal’ mean-
ing of vāyugati when I was editing the Citrasūtra as his D.Phil. student in
Oxford in the late 1980s. This term appears in the description of hallmarks of
a skilled artist, the citravid, in the last adhyāya of the Citrasūtra, and sets up
a high standard of verisimilitude for artistic practice: an expert must have an
ability to depict parts of the body (or difficult transitional points) normally
covered by ornaments like the neck, wrists, feet and ears; a body wounded
by an arrow; an old person; he must create differentiation between a sleep-
ing and a dead body, and an illusion of an apparent undulation of a flat sur-
face. It is in this context of technical mastery over representation that vāyugati
appears as a further marker of a skilled painter (Dave-Mukherji 2001, 250–
251):
taraṅgāgniśikhādhūmaṃ vaijayantyāmbarādikam
vāyugatyā likhed yas tu vijñeyaḥ sa tu citravit
He who is able to paint waves, flames, smoke, flags and garments etc. with
the speed of the wind (vāyugatyā) is considered to be an expert.
Previous editors of the Citrasūtra, who include Stella Kramrisch, A.K. Cooma-
raswamy, C. Sivaramamurti and Priyabala Shah, have overlooked the performa-
tive dimension of vāyugati. They have considered its representational mean-
ing and connected it with the objects to be depicted, such as flames, banners
and clothes that are windswept (vāyugati). Vāyugati can also connect with the
agent of representation, i.e. the citravid or the expert in painting, who must
paint flames, banners and clothes with the speed of the wind. If the first sense
locates the objects to be drawn, such as flames, banners and clothes, in the lived
via free access
dave-mukherji 585
and observable world, the second meaning implies artistic labour and skill, and
above all points to the temporality of execution as an index of mastery over rep-
resentation. Either way, it helps to recast the image of a traditional painter, not
as creating images out of a meditative trance through half open eyes and dupli-
cating formula from the past, but as one who encounters the world with keen
eyes, trained hands and a vibrant imagination.
References
Primary Sources
Viṣṇudharmottarapurāṇa. Pandit Madhusudhan and Madhavaprasada Sarma, eds. Vis-
nudharmottara Purana. Khanda III. Chapters 1–118. Bombay: Venkateshwar Press,
1912.
Viṣṇudharmottarapurāṇa. Parul Dave-Mukherji, ed. Citrasūtra of the Viṣṇudharmottara
Purāṇa. Khaṇḍa III. Chapters 35–43. New Delhi: Motilal Banarsidass, 2001.
Viṣṇudharmottarapurāṇa. Priyabala Shah, ed. Viṣṇudharmottara-Purāṇa. Third Khaṇ-
ḍa. Chapters 1–118. 2 vols. Gaekwad Oriental Series, nos. 130 and 137. Baroda: 1958 and
1961.
Secondary Sources
Asher, Frederick, 2007. “The Shape of Indian Art History.” In Asian Art History in the
Twenty-First Century, edited by Vishakha Desai, 3–14. New Haven: Yale University
Press.
Baxandall, Michael. 1972. Painting and Experience in Fifteenth Century Italy. Oxford:
Clarendon Press.
Bryson, Norman. 1983. Vision and Painting: The Logic of the Gaze. New Haven: Yale Uni-
versity Press.
Coomaraswamy, A.K. 1908. The Aims of Indian Art. London: Essex House.
Coomaraswamy, A.K.. 1933. “Vishnudharmottara Chapter 41.” Journal of American Ori-
ental Society 52: 13–21.
Coomaraswamy, A.K.. 1933. “The Painter’s Art in Ancient India, Ajanta.” Journal of the
Indian Society of Oriental Art I: 26–29.
Coomaraswamy, A.K.. 1934. The Transformation of Nature in Art. Cambridge, MA: Har-
vard University Press.
Dave-Mukherji, Parul. 2008. “The Citrasūtra and the Politics of Authenticity.” In Tattv-
abodha, edited by Kalyan Kumar Chakravarty, 125–140. New Delhi: Munshiram
Manoharlal Publishers.
Dehejia, Harsh. 1996. The Advaita of Art. New Delhi: Motilal Banarsidass.
Dhareshwar, Vivek. 2015. “Sites of Learning and Intellectual Parasitism: The Case for
via free access
New Humanities.” Journal for Contemporary Thought (A Special Number on Critical

Humanities), 41 (Summer): 57–77.
Guha-Thakurta, Tapati. 1991. The Making of a ‘New’ Indian Art: Artists, Aesthetics, and
Nationalism in Bengal, C. 1850–1920. Cambridge: Cambridge University Press.
Halbfass, Wilhelm. 1981. India and Europe: An Essay in Understanding. Albany: State
University of New York.
Kramrisch, Stella. 1922. “The Expressiveness of Indian Art.” Calcutta Review 3 (October):
1–46.
Kramrisch, Stella. 1928 [1924]. The Visnudharmottaram: A Treatise on Indian Painting
and Image-Making. Calcutta University Press. Originally published as “A Treatise on
Indian Painting.” The Calcutta Review 2 (February 1924): 331–386.
Kramrisch, Stella. 1983. “Traditions of the Indian Craftsmen.” In Exploring India’s Sacred
Art, edited by Barbara Stoler Miller, 59–68. Philadelphia: University of Pennsylva-
nia.
Lipsey, Roger. 1977. Coomaraswamy. Volume 3: His Life and Work. Princeton, NJ: Prince-
ton University Press.
Masson-Oursel, Paul. 1925. “Une connexion entre l’esthétique et la philosophie de
l’Inde: la notion de pramâna.” Review des Arts Asiatique 2 (1): 6–9. Translated into
English by Coomaraswamy as “A Connection between Indian Aesthetics and Phi-
losophy in India.” Rupam 27 & 28 (1926): 91–94.
Matilal, B.K. 1986. Perception: an Essay on Classical Indian Theories of Knowledge.
Oxford: Clarendon Press.
Maxwell, T. S, 1989, “Śilpa versus Śāstra.” In Shastric Traditions in Indian Arts, vol. 1,
edited by Anna Libera Dallapiccola, 5–15. Stuttgart: Steiner Verlag Wiesbaden
GMBH.
Misra, R.N. 1975. Ancient Artists and Art Activity. Shimla: Indian Institute of Advanced
Studies.
Nardi, Isabella, 2006. The Theory of Citrasutras in Indian Painting: A Critical Re-evalua-
tion of their Uses and Interpretations. London, New York: Routledge.
Parker, Samuel K. 2003. “Text and Practice in South Asian Art: An Ethnographic Per-
spective”: Artibus Asiae 63 (1): 5–34. www.jstor.org/stable/3249692.
Pollock, Sheldon. 1984. “The Theory of Practice and the Practice of Theory in Indian
Intellectual History.” Journal of the American Oriental Society 105: 499–519.
Raghavan, V. 1933. “Some Sanskrit Texts on Painting.” Journal of Indian Historical Quar-
terly 9: 904–905.
Ruskin, John. 1905. “Aratra Pentilici.” Chapter 6 in The Collected Works of John Ruskin,
vol. 20. London: George Allen; New York: Longmans, Green, and Co.
Shah, Priyabala. 1958/1961. See Viṣṇudharmottara Purāṇa.
Shulman, David, 2012. More than Real: A History of the Imagination in South India. Cam-
bridge, MA: Harvard University Press.
via free access
dave-mukherji 587
Sivaramamurti, C. 1978. The Chitrasutra of the Visnudharmottara. New Delhi: Kanak

Publications.
Sivaramamurti, C. 1934. “The Artist of Ancient India.” Journal of Oriental Research 8:
168–199.
via free access
via free access
Index
Abhidharmakośabhāṣya 87, 88n, 90n, asidhārāvrata 64, 66, 427

124n89 āśrama 17, 483, 497–498
Abhinavagupta xxviii, 3–6, 106–135, 144– āśramin 483, 496–497, 501–502
167, 186, 255, 259–260, 262n41, 284, Aṣṭādhyāyī (A) 89n5, 305, 315
292–310, 387–389, 391, 406 aṣṭāṅga 453, 459, 463n38, 465–466, 471n8
abhiṣeka, see Caturthābhiṣeka; consecration; Āṭānāṭiya–sutta 235n4
kings: royal consecration Atharvaveda 5, 218–220, 224–225
ācārya, see guru Atimārga 1–2, 15–29, 33–34, 38, 39n19, 278
ācāryābhimāna 5, 208–209 ātman, see self
Ācāryahṛdayam 200, 211 ātmavāda 3, 87, 98n29
Ācāryakriyāsamuccaya 10, 508, 531
Ādhyātmika 17 Bakker, Hans xi, xv, xx, xxiii, 35–38, 44, 494
Ādibuddha 10–11, 539, 546–565 bandha 8, 453, 463n38
Advayavajra 172, 180 Banaras, see Vārāṇasī
agency; agent 6, 89–90, 95–101, 122, 146, 165, banner 36, 220, 237–238, 240, 367, 376, 485,
209, 283–284, 295n25, 298, 309, 314, 535, 584
325, 338, 579, 584 bārhaspatya 110, 125n
Aghorī 40–42, 393, 395, 402 bathing, ritual 26, 37, 69, 73–74, 236–237,
Agrabodhi 11, 546–547, 551, 555, 557–559, 253, 326, 429, 489
562–565 Beyond the Fourth, see turyātīta
Ākāśavimala 11, 546, 549n22, 556, 559n39, Bhadrakālī 7, 326, 328
565n53 Bhagavadgītā 22n21, 26, 119n51, 202, 205,
Aḻakiya Maṇavālapperumāḷ Nāyaṉār 200, 211 226, 463, 466
Āḷavantār 199, 204 Bhairava 2, 20–21, 28, 34, 35n9, 40–43,
Alchi 539–548, 551, 555n32, 559–560, 562– 69n66, 71, 266n53, 294, 365, 370n24,
565 393–394, 400, 402–403, 433–434,
alcohol 2, 40–43, 326, 350, 432, 495n33 436
Amaraughaprabodha 412, 421, 452, 454, Bhairavaprādurbhāva 19–20
459, 472 bhairavas 2, 20–21, 28, 34n9, 40–43, 68–71,
Amareśvara 28, 30, 38–43 75, 266n53, 294, 365, 370n24, 393–394,
amarī 429 400–403, 433–436, 495n
Amoghapāśakalparāja 240, 241n32 Bhaiṣajyagurusūtra 240
amṛta 2, 8, 38–43, 52, 197, 206, 375, 379–380, Bhāratī 365, 371
404n49, 413–414 Bhartṛhari 110, 124n89
Amṛtakaṇikā (Aka) 172–173, 176–177 bhāva 433–434, 436
Amṛtakaṇikoddyota 173, 176 Bhavadevamiśra 468, 470
anumāna, see inference bhoga 16, 163, 211n, 255n17, 257, 261–262,
apoha 127n, 135n117 268n56, 271n65, 273–275, 355, 357, 378,
Arthaśāstra 5, 219–221, 224, 326 391, 405, 493
āsana 451, 454n9, 455, 456n13, 458–463, see also siddhi
465–467, 471, 473n, 549n22 bindu 8, 390, 405n52, 414–442, 452
ascetic(s) 1, 5, 9, 15–17, 23–27, 29n36, 33–34, Brahmayāmala (BraYā) 8, 39n19, 40, 74–
39, 54–55, 58, 68, 75, 221, 223, 229, 231, 75, 265n, 387–406, 427, 428n10, 429,
250–251, 253n10, 258n27, 270–279, 311, 438n33, 440n43, 445
332, 391, 421, 446, 454–455, 483–484, Bṛhatkālottara 19n12, 266, 269, 286n5, 484–
494–501 488, 490, 494, 498nn39–40, 503
via free access
590 index
Brunner, Hélène; Brunner-Lachaux, Hélène consecration (abhiṣeka) 66, 75, 172, 182,
52, 71n, 252n6, 253n11, 256nn18–20, 224–225, 243, 259, 270, 278n89, 328,
257n24, 258n27, 264, 270nn, 272nn68– 417, 483–484, 507–509, 530, 535
70, 276nn82–83, 85, 278, 483 see also Caturthābhiṣeka; royal consecra-
bubhukṣu 250–256, 260–263, 266, 269, 273, tion
277–278, 405 Coomaraswamy, A.K. 569–577, 581, 583–584
see also sādhaka Cord of Power, see śaktitantu
Buddhacarita 5, 229–230
Buddhism xxvii–xxviii, 1, 3, 6, 8, 10, 48, Dakṣiṇāmnāya 438n32, 440
53–56, 78–79, 87–94, 97–100, 107n9, Dakṣiṇamūrtyuddhārakośa 361
108–132, 147, 148n, 170–172, 182–188, Dattātreyayogaśāstra3 8, 453, 454n10, 457,
234–244, 252n8, 285n, 286n5, 339–340, 461, 467
365, 409–421, 452, 454, 507–514, 529– deśika, see guru
530, 539–565 Devīmāhātmya 326, 339, 340n40
Buddhist–Brāhmaṇical Ātman debate 87– Devīpurāṇa 324, 329, 339–340
88, 93 Devīrahasya 350, 359–361
Devyāmata (DM) 10, 484–485, 495, 497–
caitya 240, 512, 514 503, 509, 515–529
cakra 389, 390n9, 393, 399, 404, 406, 420, dharmabhāṇaka 236
547, 557 Dharmakīrti 88n, 107n9, 115–116, 127n,
see also Śrīcakra 135n117
Cāmuṇḍā 38, 42, 75, 325–326, 366n8, 395, Dharmottara 115, 286n5
399, 435n26 Dhvajāgrakeyūradhāraṇī 237–239, 241n5
Caṇḍikā 40–42, 322, 327, 330–331 dīkṣā 2, 4, 6, 33–38, 42–43, 48–52, 54, 75,
Caṇḍīśataka 325, 333 149n11, 154n20, 158n29, 171–188, 243,
caryā 3, 47, 50–59, 74, 77–79, 297n30 249–279, 301–302, 311, 388n, 392, 403,
see also puraścaryā, vratacaryā 454, 483–484, 492
caryāpāda 52–56, 60n16, 256n19, 258 adhikāradā 266
Caturthābhiṣeka 172–178, 182–183, 188 niradhikāradā 266n54
see also consecration (abhiṣeka) nirbījā 250–252, 256, 258n27, 265, 268–
Caturthāloka, see Tattvaratnāvaloka 269
Caturthasadbhāvopadeśa 175, 178–179, 181– nirvāṇadīkṣā 254, 264n48, 266–274,
183 277n88
causality 89–90, 97, 108, 112–116, 130, 134– naiṣṭhikī 256, 273–274, 277
136, 551 bhautikī 255n18, 256, 257n23, 258n27,
chandoha 418 259n29, 273–274, 275n80, 277
Chinnamastā 417–418 lokadharmadīkṣā 252, 266–269, 272
Christian thought, medieval 574–575 lokadharmiṇī 6, 249–279
Cīlapattikāram 324, 327, 334–336, 338 sadyonirvāṇadīkṣā 267
Citrasūtra 11, 569–585 sabījā 252, 256, 268–272, 275, 277
cognition(s) 88–97, 99–101, 118n50, śivadharmadīkṣā 266–269
124n89, 127, 128n94, 132, 134–135, 136n, śivadharmiṇī 6, 251–279
147n5 dīkṣita 497–498
colonialism 9, 11, 470–471 Durgā 322–341, 365, 368, 371
see also postcolonialism
consciousness 3–4, 62n, 89–101, 108–137, ethnography 11, 323, 337, 569, 581–582
146–162, 165–166, 174, 177, 180, 188,
259n29, 267n, 284n1, 287, 294, 309, female 2, 9, 34, 39, 71, 221, 238, 402–403, 414
357, 378, 379n, 554 see also woman
via free access
index 591
five nectars 39–40 Hatley, Shaman xxv, 8, 39n19, 40, 41n22,

see also amṛta 255n17, 259n29, 304n, 309n, 311n,
Five States; Five Mystic States 160–161 340nn38–39, 388n4, 389n5, 390n9,
flag, see banner 404n50, 405n54, 427, 428n14, 435n27,
Four Great Kings 235–236, 240, 243 495n33, 501n48
Fourth Initiation, see Caturthābhiṣeka houses; housing 237, 240, 242, 483–503
Fourth State, see turya see also gṛha, kuṭī, maṭha, maṭhikā, pura,
funeral 10, 222, 225, 508n2 veśman
householder 16, 69, 250–251, 258n27,
Gāhaḍavāla 20 270–273, 276, 278–279, 454, 499–500,
gajahasta 441, 443–444 524–525
gnosis 54, 367, 388, 391, 406, 470, 539n1, 547,
550n, 553, 555, 561n48 idealism 129n96, 573
see also jñāna imitatio dei 62–63, 71, 75–76, 78
Godāvaramiśra 463–472 indriya 109–110, 115, 121, 124n89, 125n,
Goodall, Dominic xxv, xxix, 3, 33n2, 48, 126n90, 134, 443
49n, 52n5, 54, 59n15, 60n16, 65n60, inference 4, 110–115, 121–122, 124n89, 125n,
71n, 72, 78, 93n17, 97n, 158n29, 218n3, 126n90, 127, 130n89, 132–136
221n14, 222n, 223nn17–18, 226nn23, 25, initiation, see dīkṣā
227nn26, 29, 228n30, 249n, 250, 252n8, Irāmāṉuja Nūṟṟantāti (IN) 5, 195–199, 201–
254n14, 256n19, 261n39, 264, 266n52, 209
271n66, 272n70, 273n72, 279, 322n, Isaacson, Harunaga xxv, xxix, 19n11, 22n21,
325n6, 334n17, 335n27, 339n, 364n1, 35, 48, 49n, 54, 59n15, 65n60, 72,
365nn4, 6, 379, 401n35 78–79, 95n20, 158n29, 170–72, 212,
Gorakhbāṇī 416 219n3, 221n14, 222n, 223n18, 226–
Gorakṣa 9, 446 228, 245, 286n5, 388, 417n, 447, 452n1,
Gorakṣaśataka 412, 454 507n
granthi 8, 394, 399–401, 406, 415, 416n Īśvarapratyabhijñākārikā (ĪPK) 4, 144–150,
gṛha 488–502 155, 157, 163–166, 283, 292–302
gṛhastha, see householder Īśvarapratyabhijñākārikāvṛtti (ĪPKVṛ) 292,
Gūḍhapadā 172–173, 178, 183 297, 301–302
Guhyasamāja 53n8, 171–172, 184–187, 420, Īśvarapratyabhijñāvimarśinī (ĪPV) 4, 106,
452, 508n2 108, 144, 147, 297–298, 301–302
Guhyasamājamaṇḍalavidhi 243, 515n17, 529 Īśvarapratyabhijñāvivṛtivimarśinī (ĪPVV)
Gurjara-Pratihāra 7, 368 167, 293n18, 297–298, 300n36, 301–302
guru xxv, 5, 10, 52, 57, 61–62, 66–67, 145, 148,
176, 180–182, 184–187, 195–212, 217–221, jaṅgama liṅga 483–484, 497, 503
224–225, 229, 250, 253–254, 258, 264– see also ascetic(s)
271, 297n28, 388n3, 390n9, 430–421, Jayadrathayāmala 19n12, 33n3, 251, 266–
431, 442, 444, 462–463, 473n, 483–484, 267, 269–270, 286n5, 428n10
488–489, 493–495, 510–515, 572–573 Jayaratha 6, 166n43, 255, 260–263, 269,
guruparaṃparā 203, 207, 212 284, 303, 307–308, 365n5, 387, 427n9,
Gyantse 540n, 554n31 435n30
jñāna 3, 8, 17, 23–29, 47, 50–55, 68, 69n67,
Halāyudhastotra 28–30 75, 78, 90–94, 101, 114–115, 130–133, 144,
Harṣacarita 68, 242n40, 325 146, 152, 180, 202–204, 205n18, 285, 289,
haṭha, haṭhayoga, see under yoga 297, 335, 337, 353, 356, 368, 370, 387–
Haṭhapradīpikā 8–9, 412, 422, 451–464, 467, 403, 406, 417, 539, 547n17, 596
468n56, 471–472 see also gnosis
via free access
592 index
jñāna and kriyā, dichotomy of 8, 388–392, Kiraṇa 53, 56, 59, 62, 67, 75, 79, 158n29,
406 255n14, 484, 490, 494, 498n39, 503
jñānapāda 47, 182, 187, 508n2 Kiraṇavṛtti 92nn10,12, 93n16, 94n19, 95n22,
Jñānaratnāvalī 251, 271–278 97n
jñānasattva 547, 551–559, 561–565 knower 91, 99, 146, 161, 166, 180
Jñānasvāmin 358 knowledge, see jñāna
Kramrisch, Stella 569n, 570, 582–573, 578–
Kādambarī 324–325, 330–333 581, 584
Kadri 421n28 kriyā 3, 8, 47, 54, 78, 387–388, 391–406
Kālabhairava 21 see also ritual
Kālacakratantra 172n, 173, 187, 244, 452 kriyāpāda 47, 52, 66
Kālarātri 326, 329–331, 334, 336, 338 Kriyāsaṃgrahapañjikā (KSP) 10, 508–530
Kalhaṇa 286n5, 295n25 Kṣemarāja 130n89, 145, 149n12, 159, 162–165,
Kālidāsa 217–231, 365n6 242n5, 253–255, 256n18, 258n27, 263,
Kālī 166n43, 325 381n35, 428n10
see also Bhadrakālī, Cāmuṇḍā, Kālarātri, Kubjikā 8–9, 326n8, 337, 426
Kālīkula Kubjikāmata 68–69nn, 75, 399, 429n15, 430,
Kālīkula 150, 325, 328 432
Kāmadeva 365–368, 372–373 kula 162n, 302, 310n, 311, 366, 428–436, 445,
Kāmeśvara 436, 438, 440, 445 446n53
Kāmeśvarī 364, 367, 374, 384 see also Kaula
Kaṇṇinuṉciṟuttāmpu 197–199, 201 Kulanātha 383–384
kapāla, see skull(s) Kumārila 87, 98
Kāpālika 2, 24, 28, 33–44, 75, 429, 467–468, kuṇḍalinī 9, 155n23, 378n27, 383, 399, 401,
471n 406, 435–436, 446–447, 464n40
karma 22, 47, 49, 198, 201–202, 227, 250– kuṭī 497
270, 274, 275n80, 392, 404–405
see also prārabdha Laghusaṃvaratantra 244
Kashmir xxv–xxvi, xxviii, 4, 6–7, 79n, 101, Lakulīśa 35–38
126n90, 144, 146, 170–176, 180–184, 263, Lalung 539, 540n, 543–552, 555–565
278, 283–299, 311, 325, 349–353, 369, lament 225–226
455 Laukika 17, 23–27
Kāśikāvṛtti 286n5, 307n51, 315n, 317n laya, see under yoga
Kaul, Sāhib 352–353, 355n31, 358–361 liṅga 26, 30, 48–49, 61–66, 229, 402, 483–
Kaula 33n3, 150, 156, 161–162, 292–293, 305, 484, 497–498, 503, 508n3, 513n, 529,
310n, 391–392, 426, 428, 434, 445–446 530
see also kula see also jaṅgama liṅga, sthāvara liṅga
Kevalajñāna; Kevalijñāna 25, 29–30 liṅgin 271n67, 496, 498
khaṭvāṅga 36n11, 58, 65 Linrothe, Robert 540n, 545n11, 546n14,
Khecarīvidyā 409 563n, 565
Khmer Empire xxviii, 4, 170, 184, 186–187 lokadharma 6, 249, 251, 262, 275, 278,
kings; kingship 5–6, 20, 181, 217, 226, 229– 303n45
231, 234–243, 268n56, 285n, 321–327, lokadharmī 6, 251–254, 257–265, 269–272,
336–341, 368, 383–384, 494, 514, 557 275n81, 276
royal consecration 328 lokamārga 22–26, 27n32, 28–29, 251n
royal palace 5, 236, 242n40, 244 Lokamārga 22–26, 27n32, 28–29
royal patronage 242, 285n4, 483, 508n3, Lokātīta 22, 24, 25n25, 26–27
512–514, 529n54 Lokottaramārga 23, 27–28
see also rājaguru, purohita Lorenzen, David 33n5, 34, 34n6
via free access
index 593
Luczanits, Christian 539n3, 540n, 544, Matsyendrasaṃhitā (MaSaṃ) 8–9, 426–447

545nn8–9, 11, 549nn21, 24, 549n24, Maturakavi āḻvār 197, 201
551n, 552, 552n28, 561, 562n50, 563n, Māyā 50–51, 93, 98, 146, 274, 326, 429–437,
564n, 565 442, 445–446
Mayasaṃgraha 484, 488–490, 494–495,
Mahābhārata 22n21, 26n28, 54, 205n18, 218, 503
312, 324–325, 329–331, 334, 463 Māyōṉ 197, 200–203
Mahābhāṣya 306 Mohacūrottara 483–497
Mahākāla 21, 30, 332, 333 moon 2, 8, 35–38, 43, 204–205, 295, 335,
Mahāmāyāvijayavāhinīdhāraṇī, see Nārā- 353, 357, 371–382, 413–417, 439,
yaṇaparipṛcchā 442, 453, 465n46, 548, 553, 558,
Mahāmāyūrīvidyārājñī 239n23, 242n40 563
Mahānavamī; Navamī 7, 325–328, 337 Mṛgendra 52, 55, 93nn16–17, 255n18, 256–
Mahāpratisarāmahāvidyārājñī 241 278
Mahāsāhasrapramardanasūtra 237, 239, Mṛtyuvañcanopadeśa 171, 187
242n41 mudrā(s) 8, 366, 372, 409, 411, 416–417,
Māhātmya 2, 16–25, 28 421, 427, 443, 444n, 452–472, 543–559,
Mahāvairocana 11, 510, 541, 546–551, 562n51
553–555, 557n36, 559–562, 564– mumukṣu 250–256, 265–266, 270, 273, 405
565 see also putraka
mahāvrata 35–36, 37n14, 66
Mālinīvijayottara 54, 159, 164, 166, 304, 308 nāḍī 8, 378n27, 390, 399–401, 404, 413, 415–
Mangyu 10, 539, 540n, 541n, 542–551, 556– 416, 452, 456n13, 459
559, 562–565 Nāgabhaṭa II 7, 368–369
Mañjuśrī 10–11, 539–565 Nāmamantrārthāvalokinī (nmaa) 11, 174,
Mañjuśrīmitra 11, 546–547, 549n22, 555– 546–549, 552, 554n29, 557, 561–563,
559, 564–565 564n
Mañjuśriyamūlakalpa 48, 78, 243 Nāmasaṃgīti (ns) 11, 539n1, 546–547, 556,
mantra 1, 5–8, 15–16, 48–50, 57–58, 61, 63, 558, 561, 562n51, 563–565
66–79, 181, 198, 242, 244, 254, 257, Nammāḻvār 197, 199, 201
259n29, 325–326, 349–354, 358–361, nandyāvarta/nandikāvarta 489–492, 502
367, 389, 392–404, 428n14, 429, 437, Nārāyaṇakaṇṭha 256n18, 257n23, 261n35,
442, 452–453, 454n10, 459, 471n, 459n5, 263
530, 557 Nārāyaṇaparipṛcchā 237, 238, 239n26,
Mantramārga 1–3, 15–18, 24, 27–28, 47, 242n38
54–56, 59, 68–70, 75, 78–79, 508n3, Nardi, Isabella 570, 582
529n54 Nareśvaraparīkṣāprakāśa 87, 92n9, 93n18,
mantroddhāra 7, 350, 353, 358–361, 367n13 94n19, 95n22, 96n25
Manusmṛti 17, 18n7, 19n12, 69, 75, 224 nationalism 11, 569, 571–572, 576–577, 582,
marriage 220, 222, 225, 303n 584
Mataṅga 52n4, 54, 66, 72, 93n17, 263n43, naturalism 11, 571–579, 583–584
399, 400n29, 427, 466n53, 467 Navadharma 244
Mataṅgavṛtti 92nn9, 11, 13–14, 95n22, Navarātra 7, 321–341
96nn23, 25 nectar, see amṛta; five nectars
maṭha 10, 483–485, 489, 494–497, 503 Nidrā 326, 329, 336, 338
maṭhikā 496–497 Niśvāsatattvasaṃhitā (Niśvāsa) 1, 3, 8, 17,
mātṛkā 77–78 47–48, 50n, 52–53, 59, 62, 70, 72, 75,
Matsyapurāṇa 19 78–79, 252n8, 258n27, 264, 365, 413n18,
Matsyendra 9, 446 427
via free access
594 index
Guhyasūtra 2–3, 48–52, 62–67, 68n65, prāṇāyāma 327, 451, 454n9, 456n14,
70, 72, 75, 78–79 459–460, 461n38, 463n38, 465–466,
Niśvāsamukha 17–18 471n68
Nityā Sundarī 7, 364–369, 373, 375, 377, prapatti 5, 194, 195n, 198–199, 208–
380n30 212
nityānusaṃdhānam 198, 212 prārabdha 253–255, 259n29, 260n31,
Nyāya; Naiyāyika 3, 87–101, 125n–126n, 268n56, 270, 274
130n98, 210, 286n5, 461, 469n64 see also karma
Nyāyabhāṣya 98n32 pratiṣṭhā 10, 66, 392, 484, 507–514, 530
Nyāyamañjarī 89n3, 95n20, 98nn30– pratiṣṭhātantras 508–509, 514–515, 529–
32, 107n9, 125n, 126n90, 129n98, 530
130n98 Pratyabhijñā 3, 106, 113n31, 118n47, 124n89,
145, 146n, 159, 292–294, 297–302, 308,
omen 219–221, 524–525, 526n38, 529, 533 309n, 310n, 311, 352
pratyakṣa 458n17, 577
paddhatis 10, 364, 367n14, 368, 467n54, 508, Prīti 366, 368, 372
530 pura 488, 494–495
see also Somaśambhupaddhati, Śrīvidyān- purāṇas 19–20, 26, 323–324, 329, 339–340,
ityapūjāpaddhati 463–468
Padmā 365, 368, 371 puraścaraṇa 68, 70
padmamālā 8, 392–406 puraścaryā 3, 68, 79
Padmanidhi 365, 368, 370 purohita 5, 217–225, 229–231
Pañcarakṣā 239n27, 244–245 pūrvasevā 3, 68, 70, 74, 79
Pāṇini xxvin, 305, 307, 311, 315–318, 445 putraka 57, 59, 254–255, 258n27, 264, 271,
Parā Prakṛti 7, 368, 379, 380n30 483, 493
Paramaśiva, see under Śiva see also mumukṣu
pāramitā 240, 545n9, 561n48
Paramokṣanirāsakārikāvṛtti 92n9, 97n Raghavan, V. 194n2, 574, 577
Parker, Samuel 581 Raghuvaṃśa 5, 217–221, 224–225, 229–230,
Paścimāmnāya 432 296n27
Pāśupata 2, 15–18, 21–25, 28–30, 33–35, rājayoga, see under yoga
37n14, 38, 73, 333 rājaguru 465, 483
Pāśupatasūtra 17, 24–25, 29, 439, 440n40 see also purohita
Pātañjalayogaśāstra 462–463, 464n40, 466, Rājataraṅgiṇī 285n4, 286n5, 295n25
468, 471, 472 Rāmakaṇṭha 3, 67, 87–101
Patañjali 306, 465–466, 471n Rāmānuja 5, 194–212
Piṅgalāmata 484, 490–495, 498n39, 503, Ratanasutta 242
515, 516nn21–26, 517n27 Rati 366, 368, 372
Pollock, Sheldon 6, 287–313, 580 Ratnakaṇṭha 359
possession 40, 226, 293, 304n, 324, 327, 334, Ratnaketuparivarta 240
336 Ratnavajra 4, 175–183
see also samāveśa Raviśrījñāna 172–173, 176–178, 183
postcolonialism 11, 569, 576–579 ritual xxviii, 1, 5–11, 27n32, 28, 33–43, 47,
Power of Awareness 145–146, 154 52, 54, 57, 61, 63, 73, 75, 78, 187, 198,
Pradyumna peak 349–352 218–225, 234–244, 249–251, 262–275,
Prajñāpāramitā 240, 244n, 543n6, 546–549, 278–279, 303n, 300, 321–341, 350, 352,
553, 555–558, 562, 564 359, 387–394, 403–406, 426–447, 471n,
prakāśa xxx, 108, 112, 118, 119n51, 126, 483, 489, 495n33, 507–515, 524, 529–
128n94, 133, 155, 160, 163 530
via free access
index 595
see also asidhārāvrata; bathing ritual; 150, 217, 220n9, 241n37, 243, 249, 250n3,
caryā; Caturthābhiṣeka; conse- 263n43, 264n45, 266, 267n, 268n57,
cration (abhiṣeka); dīkṣā; funeral; 271, 272n71, 277–279, 283, 286n5, 294,
jñāna and kriyā, dichotomy of; kriyā; 295n23, 310n, 314n60, 322, 325, 326n9,
Mahānavamī/Navamī; mahāvrata; 328, 349nn1, 3, 352nn24–25, 359, 367,
mantra; puraścaraṇa; puraścaryā; 368n15, 387, 392, 399, 403, 409–410,
pūrvasevā; pratiṣṭhā; sādhana; sex- 417, 418nn24–25, 426–427, 428n10,
ual ritual; śrāddha; vidyāvrata; vrata; 435n26, 438n32, 440n42, 451, 467n54,
vratacaryā; worship, outer 483–484, 496n36, 497n, 507n, 508,
royalty, see kings; kingship 509n4, 513–514, 516nn23–24, 517n27,
Rudra 16, 58, 242n40, 277n88, 415, 429, 486 529n54, 539n1, 584
Saptāṅga (SaA) 171, 177, 179, 183–187
Sab Bāk inscription 170, 184 Sarasvatī 235–236, 240n28
Ṣaḍanvayaśāmbhava 9, 426 Śārikā 7, 350–361
Ṣaḍdhātusamīkṣā 110, 123n88, 124nn88–89 Śārikāstava, Śārikāstrotra 351–358, 360
sādhaka 40–41, 48–49, 57–63, 66, 70–71, Sarvadurgatipariśodhanatantra 243,
77–78, 250–273, 278, 401–406, 427n6, 560n46
428n11, 429n16, 440n43, 483–484, 492– Sarvajñānottara 33n2, 53, 57n, 59, 62, 66–
493, 495n33, 548, 554 67, 72, 75, 155n22, 158n29
see also bubhukṣu Sarvatathāgatādhiṣṭhānavyūha 240
sādhana 48–49, 63, 547, 550 Sarvavajrodaya 243
*Sādhanaupayika 11, 546, 557, 559n39, 563, śāstra-prayoga 6, 11, 287–290, 309, 312, 569,
564n 580–581
sādṛśya 11, 569, 572–583 satya 486, 489, 499, 570–575, 582–583
Saiddhāntika 3, 51, 59, 72, 75, 92–93, 96, Sautrāntika 3, 88n, 108–116, 135, 188
100–101, 126n90, 161–162, 270–271, 279, Sears, Tamara 10, 483, 503
467n54, 509 self 4, 24–25, 87–101, 119n51, 125n, 126n90,
Śaiva; Śaivism xxv–xxviii, 1–3, 6–10, 15–30, 145–166, 293, 389–390, 465, 489, 494
33–44, 47–79, 87n2, 92–95, 99–101, Self-that-is-awareness 146, 165
107n9, 108n, 112–117, 127n, 128n94, sexual ritual 9, 426–447
130n98, 131n100, 134n113, 144–147, 159– Shah, Priyabala 570, 574, 584
167, 186, 202, 217, 242n40, 249–279, Shulman, David 569–570, 578–579, 583–584
283–311, 321, 334–337, 339n, 340n39, Siddhaikavīratantra 244
364–384, 387–406, 409–421, 426–447, siddhi 7, 16, 48–49, 52, 58, 63–66, 70, 74, 77–
451–454, 464n40, 467n54, 471n68, 483– 78, 149, 154n21, 163, 252n8, 255, 256n18,
503, 508–530 260, 325, 327, 335, 355n30, 366, 369n19,
Śaivasiddhānta 47, 100–101, 392 392n, 405–406, 431, 437, 485
śaktipāta 158, 260–261 see also bhoga
śaktitantu; śaktisūtra 8, 387, 395–403 śilpācārya 534, 535n
samāveśa 4, 144, 146, 150–152, 157, 160, 163, Śiva
165–166, 419 -tattva 159, 163–164
see also possession as sthāṇu 333
Śaṃkhanidhi 365n6, 368, 370 consort of 34, 334–336, 356, 378, 384
Sāṃkhya 3, 25, 93, 94n18, 96–101, 124, 147, 461 devotion to 268, 293, 496
Sanderson, Alexis xxv–xxxiv, 1, 7–10, 15– mythology of 20, 25–27, 29, 35–36, 294–
16, 17n6, 18n9, 25n26, 28, 33n1, 35, 37, 295, 311, 370
40n21, 44, 47–48, 49n, 54–55, 59n15, nature of 17, 24–25, 27, 30, 49–51, 119n51,
63nn46–47, 64n58, 65n60, 68, 70–72, 146, 159, 162, 202, 300–301, 309, 369, 421
75, 78nn78–79, 79n, 87, 95n21, 144–145, Niṣkala- 51, 161, 254, 260n31
via free access
596 index
Parama- 365, 368–369, 401 Śrīvidyānityapūjāpaddhati 359

Sadā- 50–51, 157, 159, 164, 259n29, stotra; stuti 5, 7, 28–30, 165, 195–202, 207,
270n62, 401, 429 212, 240, 353, 358–360, 364–368,
Sakala- 58, 254, 259n29, 260n31 380n32, 384
teachings of 16, 33n5, 43, 50, 294, 299, see also Halāyudhastotra; Kāmasiddhis-
434, 461 tuti; Śārikāstava; Sūryastutirahasya
temple of 61, 67, 242n40, 513 sthāpaka 529n54, 535n
union or identity with 42–43, 52, 62, sthāvara liṅga 483–484, 497, 503
93, 149, 157, 158, 166, 250, 257, 259n29, stuti, see stotra
260–262, 267n, 273–275, 293, 298, Sūkṣmāvarttabhaṭṭa, see Ratnavajra 174–
305n47, 308, 382, 388, 393, 405nn52, 176, 183
53, 416, 432–433, 436, 438, 464n40 Sundaradeva 470, 472
worship of 2, 28, 61, 73, 265, 326, 330, 431 Sūryastutirahasya 359
see also Bhairava; Kāmeśvara; Kulanātha; Svacchanda 2, 18, 20n14, 24–25, 28, 250n4,
imitatio dei, liṅga; Mahākāla; Rudra; 251n, 252–259, 263–266, 268, 274, 278,
samāveśa; śivatva; Śiva-water; Soma 286n5
śivatva 149, 155n22, 158n29, 260–262, svādhiṣṭhāna yoga 420
274n79, 297, 301n36, 305n47, 382 Svāyambhuvasūtrasaṅgraha 69–72, 75,
see also Śiva, union or identity with 93n17
Śiva-water 73–74, 429
śivadharmī 252, 257–261, 263n43, 271–272, Tantrāloka (TĀ) xxviii, 118n47, 130, 154n20,
275 158n29, 165–167, 255, 259–262, 268,
see also sādhaka 284, 365n5, 387, 389, 427n9, 428n12,
Śivadṛṣṭi (ŚD) 107n9, 119n51, 124n89, 293, 435n27
294n19, 297, 298n, 299, 302n44, Tantrālokaviveka (TĀV) 130n98, 165, 284
305 Tantrasadbhāva 68n63, 69n66, 75, 259,
Śivadṛṣṭivṛtti (ŚDVṛ) 124, 300n36, 302n44 264n48, 278, 389–391, 406
Śivānandasarasvatī 464, 467n, 472 Tattvaratnāvaloka (TaRaA) 171, 173, 176–180,
Sivaramamurti, C. 570–571, 577–579, 584 182–183
Śivasaṃhitā 412, 453, 454n9 Tattvaratnāvalokavivaraṇa (TaRaVi) 171,
Śivasūtra 150 176, 183, 185
skull(s) 2, 18, 21–26, 33, 36–39, 58, 65, 370, Tattvasaṃgraha (Sarvatathāgata-) 509,
372, 413, 527 545nn9–10, 547, 560, 561n48, 565
see also khaṭvāṅga Tattvasaṃgraha (by Sadyojyotis) 93n16
Soma 2, 33–39, 42–44, 486, 489, 500 textual parallels 10, 24, 78, 92, 106n2, 151–
Somānanda 6, 119n51, 124n89, 283– 152, 158n29, 432, 508–509, 514–517, 530
284, 292n, 293–294, 295n25, 297– Tibetan 8, 148n10, 170–172, 175–181, 188, 186,
299 234, 235n2, 238–239, 285n, 410nn2,
Somaśambhupaddhati 266n54, 270, 4, 411, 416, 508n2, 546, 549, 554, 563,
272n68, 276, 483 564n
Somaśarman 2, 35–38, 43 Tiruvaraṅkattamutaṉār 5, 195, 196, 197n5,
Somasiddhānta 2, 33, 43 198nn10–11
śrāddha 275–276 Tiviyappirapaṉtam 197–201, 205, 207, 211
Śrī 237, 240 Torella, Raffaele 3, 106, 109n12, 115n35,
Śrīcakra 349, 366–368, 372, 378n28 117n41, 124n89, 129n97, 134n, 144n111,
Śrīmahādevīvyākaraṇa 240 113–115, 146, 147n7, 148n9, 149, 150,
Śrīnagar 7, 349, 351 151n15, 152–159, 162–164
Śrīvidyā 352, 359, 399, 426, 438n32, 440 transcendentalism 11, 569–572, 576–577, 581
see also Dakṣiṇāmnāya Trika-Krama synthesis 162, 165
via free access
index 597
Trilocana 270, 276, 279, 372 Vidyāpīṭha 39n19, 389

Tripurā; Tripurasundarī 8, 349, 352, 364, Vijñānavāda; Vijńānavādin 4, 88n, 112,
366–369, 426 113n31, 147, 188
turya 4, 144–167, 177 vikalpa 121, 127n94, 432n20
turyātīta 4, 144–167, 180 Vilāsavajra 11, 174, 539n1, 546–548, 549n23,
550–565
Udbhaṭa 126n90, 286n5 Vimalāvatī 271
Utpaladeva 3–6, 101, 106–135, 136n, 144–145, vimarśa; vimarśana 95, 110, 112, 121, 132–133
155n23, 157, 158n30, 161, 165, 283–284, Viṣṇudharmottara 11, 286n5, 328, 569–585
292–302, 310n see also Citrasütra
Vivekamārtaṇḍa 412, 454, 461
Vāgīśvarakīrti 4, 170–187 void 4, 145–148, 152–153, 160–161, 405, 452
Vaidika; vedic 2, 6, 16–17, 20n15, 25, 34, 42– Vraḥ Dhanus 184, 187
43, 54n, 69, 205–206, 211, 218, 242n40, vrata 3, 35, 48, 53n8, 54, 56–68, 74–78,
285, 289, 290n11, 303n, 305, 307–312, 256n19, 258n27, 392
325–326, 335, 338, 352, 359 see also asidhārāvrata, mahāvrata,
see also Atharvaveda; Vedas vidyāvrata, vratacaryā
Vaiṣṇava 1, 5–8, 20, 194–199, 204n15, 207– vratacaryā 54–59, 67–68, 71
212, 321, 325–329, 335–338, 365, 368, see also caryā, puraścaryā, vrata
453
Vajradhātumaṇḍala 509–510, 530n55 war; warfare 240, 244, 245n48, 322, 337, 466
Vajrāvalī 10, 484, 507n, 508, 511 Watson, Alex xxix, 3, 87, 89n3, 91n, 92n9,
Vajrayāna xxviii, 410, 412n9, 416–421, 452, 93nn15–16,18, 94n19, 95n22, 97n, 99,
509 101n, 147, 261n39, 379n30
Vāmakeśvaratantra 364, 366–384, 435 Wedemeyer, Christian 53, 55, 59n14, 62, 68,
Vāmakeśvarīmata, see Vāmakeśvaratantra 72
Vāmanapurāṇa 19–20 woman 9, 65–66, 77, 221, 225, 250, 268n,
Vārāṇasī 2, 16–25, 28, 271, 277, 390n9 279, 294, 296n27, 299n, 332–335, 357,
Vārāṇasīmāhātmya (VM) 19, 24–27 339, 340, 354, 391, 427nn, 428n10, 429,
Vasubandhu 87, 88n, 110, 112, 130n100, 431–432, 434–436, 445–446, 499–500,
131n100 535, 582
Vaṭakalai 194n2, 195–198, 209–212 see also female
Vatsarāja 7, 364, 368–369, 380n32, 384 worship, inner 387–392, 403–406
vāyugati 584 worship, outer 387–392, 403–406
Vedas 69, 181, 198, 200, 205–207, 211, 230,
242n40, 285, 289, 340, 381, 461, 467, Yatirājasaptati (YS) 195, 203–207
530 Yatirājabhairavam 204
see also Atharvaveda; Vaidika; Vedānta yoga xxix, 1, 3, 7–9, 25, 38, 47, 50–52, 78, 144,
Vedānta 147, 205–206, 461–466, 470 153, 163, 220, 267n, 268n56, 387–390,
Vedānta Deśika 5, 194–197, 203, 206–212 394, 399, 403–406, 409, 411, 412n9, 416–
veśman 223, 497, 519, 524 421, 426, 446n54, 451–472, 495, 548,
Vibhūticandra 173, 176 550, 576
vidyā (mantra) 47–51, 66–68, 69n66, 72–77, haṭha- 8–9, 39n18, 399, 409–416, 420–
384, 392–393, 395, 402, 404 421, 426–429, 446, 451–472
vidyāṅgamantra 68, 71–75 laya- 406, 452–453, 454n10, 459, 471n
vidyāvrata 3, 54–56, 62, 66–79 rāja- 9, 451–459, 463, 464n40, 465, 467–
vidyādhara 7, 239n24, 241n33, 364, 368, 382, 470, 471n
384 see also āsana; aṣṭāṅga; bandha; bindu;
see also Vatsarāja granthi; kuṇḍalinī; mudrā; naḍī;
via free access
598 index
prāṇāyāma; śaktitantu; yogic body; yogatantra 11, 509, 530, 545n9, 547, 565
worship, inner yogic body 8, 390, 393–401, 404n49, 411, 413,
Yogācāra, see Vijñānavāda 420
Yogacintāmaṇi 410n3, 412, 456–457, 459n23, yoginī 9, 71, 325, 327, 393, 399, 417, 428–436,
461n24, 463–467, 470n66–67 445–446
via free access

Shaivism and The Tantric Traditions PDF

Uploaded by

Copyright:

Available Formats

Shaivism and The Tantric Traditions PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Shaivism and The Tantric Traditions PDF

Uploaded by

Copyright:

Available Formats

Śaivism and the Tantric Traditions

Peter C. Bisschop (Leiden)

Hans T. Bakker (Groningen)

The titles published in this series are listed at brill.com/gis

Library of Congress Cataloging-in-Publication Data

Names: Sanderson, Alexis, honouree. | Goodall, Dominic, editor. | Hatley, Shaman,

This book is printed on acid-free paper and produced in a sustainable manner.

1 From Mantramārga Back to Atimārga: Atimārga as a Self-referential

2 Why Are the Skull-Bearers (Kāpālikas) Called Soma? 33

3 Dressing for Power: On vrata, caryā, and vidyāvrata in the Early

4 Further Thoughts on Rāmakaṇṭha’s Relationship to Earlier Positions in

5 Some Hitherto Unknown Fragments of Utpaladeva’s Vivṛti (II): Against

6 Alchemical Metaphors for Spiritual Transformation in Abhinavagupta’s

7 On Vāgīśvarakīrti’s Influence in Kashmir and among the Khmer 170

8 Reflections on the King of Ascetics (Yatirāja): Rāmānuja in the

10 Buddhism, Kingship and the Protection of the State: The

11 Adapting Śaiva Tantric Initiation for Exoteric Circles: Lokadharmiṇī

12 Innovation and Social Change in the Vale of Kashmir, circa

13 Toward a History of the Navarātra, the Autumnal Festival of the

14 Śārikā’s Mantra 349

15 The Kāmasiddhistuti of King Vatsarāja 364

16 The Lotus Garland (padmamālā) and Cord of Power (śaktitantu): The

17 The Amṛtasiddhi: Haṭhayoga’s Tantric Buddhist Source Text 409

18 A Sexual Ritual with Māyā in Matsyendrasaṃhitā 40 426

19 Haṭhayoga’s Floruit on the Eve of Colonialism 451

20 The Early Śaiva Maṭha: Form and Function 483

21 The Kriyāsaṃgrahapañjikā of Kuladatta and Its Parallels in the Śaiva

22 Mañjuśrī as Ādibuddha: The Identity of an Eight-Armed Form of

23 Life and Afterlife of Sādṛśya: Revisiting the Citrasūtra through the

This volume results from a symposium held at the University of Toronto in

4.1 Self as essence 88

6.1 Two types of Turyātīta in the ĪPVV 164

“Who is Afraid of Mimesis? Contesting the Common Sense of Indian Aesthetics

projects: writing entries for the Tāntrikābhidhānakośa, contributing to the edi-

India: Vilāsavajra’s commentary on the Mańjuśrī-nāmasaṃgīti. A critical edition

On the occasion of the Symposium organised by Srilata Raman and Shaman

I had the great good fortune to begin my studies of Sanskrit at Oxford

We do not always find excellence in research combined with excellence

Sanskrit texts marshalled to demonstrate ideas, doubts and conclusions, will

mantravidhiprakarana.” In The Atharvaveda and its Paippalāda Śākhā: Histor-

2002 “Remarks on the Text of the Kubjikāmatatantra.”Indo-Iranian Journal 45 (2002):

Academic study of Asia’s tantric traditions has blossomed in recent decades.

© dominic goodall et al., 2020 | doi:10.1163/9789004432802_002

Chapter 3, by Dominic Goodall, is entitled “Dressing for Power: on vrata,

order to account for phenomenal variety. In these fragments Utpaladeva shows

Chapter 8, by Srilata Raman, is entitled “Reflections on the King of Ascetics

rulers and their realms is a long-established practice in South Asian Buddhism,

Chapter 13, by Bihani Sarkar, is entitled “Toward a history of the Navarā-

Chapter 16, by Shaman Hatley, is entitled “The Lotus Garland (padmamālā)

kā-Tripurā-oriented tantric yoga text of the Ṣaḍanvayaśāmbhava tradition, de-

ure in three Nāmasaṃgīti-related texts—the Ākāśavimala of Mañjuśrīmitra,

From Mantramārga Back to Atimārga:

The impact of Alexis Sanderson’s scholarship can be easily measured by the

© peter c. bisschop, 2020 | doi:10.1163/9789004432802_003

a development from within the Pāśupata tradition; both are concerned

2 Atimārga and Mantramārga

The Teaching of Śiva (śivaśāsana) which defines the Śaivas is divided

3 Quoted from the electronic version: http://www.oxfordreference.com/view/10.1093/acref/978

atidattam atīṣṭam atitaptaṃ tapas tathā atyāgatiṃ gamayate