Cultural History of Early South Asia

For our entire range of books please use search strings "Orient BlackSwan",
"Universities Press India" and "Permanent Black" in store.
Cultural History
Of
Early South Asia
A Reader

Edited by
Shonaleeka Kaul
Orient Blackswan Private Limited
Registered Office
3-6-752 Himayatnagar, Hyderabad 500 029 (A.P.), INDIA
e-mail: [email protected]
Other Offices
Bangalore, Bhopal, Bhubaneshwar, Chennai,
Ernakulam, Guwahati, Hyderabad, Jaipur, Kolkata,
Lucknow, Mumbai, New Delhi, Noida, Patna
© Orient Blackswan Pvt. Ltd. 2014
First published by Orient Blackswan Pvt. Ltd. 2014
eISBN 978 81 250 5601 0
e-edition:First Published 2014

ePUB Conversion: Techastra Solutions Pvt. Ltd.

All rights reserved. No part of this publication may be reproduced,
distributed, or transmitted in any form or by any means, including
photocopying, recording, or other electronic or mechanical methods,
without the prior written permission of the publisher, except in the case of
brief quotations embodied in critical reviews and certain other
noncommercial uses permitted by copyright law. For permission requests
write to the publisher.
śrīkṛṣṇārpaṇamastu

For beloved Tulip

 Contents

List of Tables and Figures

Preface
Publisher’s Acknowledgements
Introduction: Producers and Consumers of Culture
Shonaleeka Kaul

1. ‘A Figure of Speech, or a Figure of Thought?’

Ananda K. Coomaraswamy

2. Rock Paintings of the Mesolithic Period

Erwin Neumayer

3. Ornament Styles of the Indus Valley Tradition

Jonathan M. Kenoyer

4. Texts on Stone: Understanding Aśoka’s Epigraph- Monuments

Upinder Singh

5. Social Background of Ancient Indian Terracottas

Devangana Desai

6. On Modes of Visual Narration in Early Buddhist Art

Vidya Dehejia
 7. Archaeology of Early Temples in the Chalukyan Regions
Himanshu Prabha Ray

8. Ellora: Understanding the Creation of a Past

Jaya Mehta

9. Theory and Practice of Painting: Introduction to the

Viṣṇudharmottara
Stella Kramrisch

10. The Jataka as Popular Tradition

Uma Chakravarti

11. The Functions and Social Location of Kāvya

Shonaleeka Kaul

12. Bards and Bardic Traditions in Early Tamil Poetry

K. Kailasapathy

13. Who Needs Folklore? The Relevance of Oral Traditions to South

Asian Studies
A. K. Ramanujan

Glossary of Select Terms

Notes on the Contributors
Tables and Figures
TABLES

3.1 Harappa 1986–90: Bangles sorted by raw material type

3.2 Ranking of bangles by raw material and technology

3.3 Harappa 1986–90: Beads sorted by raw material type

3.4 Ranking of beads by raw material and technology

3.5 Major Harappan phase bangle types and raw materials

3.6 Selected Harappan phase bead types and raw materials

3.7 Harappa 1986–88: Primary context burials

3.8 Harappa 1986–88: Burial ornament styles

FIGURES

3.1 Ornaments of the Indus tradition

3.2 Harappan male ornament styles

3.3 Harappan female ornament styles

5.1 A goddess with partner, Ahichhatra, about second century bce

5.2 A garden party with dance and music, Mathurā, c. second century
bce
5.3 A couple from Kauśāmbī, c. second century bce

5.4 A female head from Mathurā, Kuṣāna period

5.5 A panel of the Bhitargaon temple

6.1 Monoscenic narrative, Vessantara Jataka. Bharhut coping, c. 100–

80 bce

6.2 Monoscenic narrative, Asilakhana Jataka. Bharhut coping

6.3 Monoscenic narrative, Kukkuta Jataka. Bharhut coping

6.4 Monoscenic mode, Chaddanta Jataka. Top architrave (inner face),

north gateway, Sanchi stupa, c. 50–25 bce

6.5 Static monoscenic narration, diachronic mode. Ajatashatru pillar,

Bharhut

6.6 Static monoscenic narration, synchronic mode. Prasenajit pillar,

Bharhut

6.7 Synoptic narrative, Monkey Jataka. Sanchi stupa

6.8 Synoptic narrative, Chaddanta Jataka. Amaravati, second century

ce

6.9 Conflated narrative, Dipankara Jataka. Gandhara, second/third

century ce

6.10 Conflated narrative, Deer Jataka. Bharhut

6.11 Continuous narrative, Great departure of the Buddha. Central

architrave (outer face), East gateway, Sanchi stupa

6.12 Detail of Linear narrative, Vessantara Jataka. Goli, third century ce

6.13 Detail of Linear narrative, Story of Nanda. Nagarjunakonda, third
century ce

7.1 Dolmen on Meguti hill, Aihole, in the vicinity of Jaina temple

7.2 Bhutanatha group of temples at the edge of the tank, Badami

7.3 Buddhist vihara at Aihole

7.4 Durga temple at Aihole

7.5 Lad Khan temple at Aihole

 Preface

T HE IDEA OF PUTTING TOGETHER A READER ON THE cultural

history of early South Asia—as vast a topic as there can be—arose back
in 2008, when Hemlata Shankar of Orient BlackSwan (then still Orient
BlackSwan) approached me to take up the task. This book owes itself to
Lata’s patience and faith. I was then on a visiting chair at Yale University and
Lata waited for me, not only till I returned to India, but till after I had
worked my way through critical illness and death in my immediate family,
and several months of upheaval in my circumstances thereafter. It was only
in 2010 that we got down to discussing what it was that I wanted to do with
this book.
Compiling a cultural history of early India (which I later expanded to
South Asia for reasons explained in the Introduction) was not an obvious or
easy mandate. Any number of masterly works on different aspects of India’s
vast and varied cultural past had been published over the last century. Not
least among them was a collection of essays edited, appropriately, by the
Indophile A. L. Basham, and named precisely A Cultural History of India,
which attempted a sweeping coverage of Indian religion, philosophy, art,
architecture, literature and so on, in one volume. Though obviously the
potential for fresh research was nowhere near exhausted, fresh research is
not quite what a Reader based on already published, representative works
purports to do. So what could we say that would be new and yet only
introductory? What tack or line should the Reader adopt that would
acquaint its readership with fundamental contributions to the region’s art
history, and yet do so in a way that questions and opens up received
wisdom, and initiates, hopefully, a new understanding of early cultural
processes?
 I decided that the answer may lie in somewhat democratising the study of
art and culture in early India. This would take two forms—namely,
broadening the choice of objects selected for study, and more so, the
questions asked about those objects. Hence readers will find the inclusion of
ornaments, on the one hand, and folklore, on the other, in this collection of
essays; these were and arguably continue to be widespread forms of cultural
production that emanate from people’s daily lives, but have rarely been
incorporated in standard surveys of South Asian cultural history. Then,
apart from commonplace culture, this volume also takes up more visible and
conventionally regarded cultural forms, like temples, stupas, sculptural
reliefs, or, of a different order, drama and poetry composed in ancient times.
But here too, it seeks to broaden the perspective and nudge the focus beyond
their production to the much wider phenomenon of their consumption or
reception. What did an art form mean to its audience? Indeed, who
constituted its audience, its community of response, its clientele and user
base? How did art speak to their lives—to their beliefs, practices and
identities? And how did the dialogue between culture and its communities
contribute over the centuries to the formation of South Asian traditions?
The guiding principle of the book is thus a quest for elucidating culture’s
varied contexts of consumption. Despite challenges, including the paucity of
writings that explicitly interpret culture from this vantage , this book raises
these questions about an eclectic—though, necessarily, not exhaustive—
range of cultural forms, each marking a unique innovation, from
chronologically ordered periods of South Asia’s past, beginning with
prehistory down to approximately the beginning of the early medieval.
Readers shall judge for themselves the proffered answers.
The manuscript was ready by 2011. A number of people offered
encouragement in the course of its preparation. I should thank first the
scholars whose essays are carried here, or their heirs, who were supportive
of the concept behind this Reader, and enthused at the novel line of enquiry
their work was being both deployed for and subjected to. They all gave
readily of their consent. I am thankful to my colleague Professor Kesavan
Veluthat who showed a warm interest in the project and put me through to
Professor Indrapala in Sydney who, in turn, I thank for putting me through
to the family of the late K. Kailasapathy in Colombo. My gratitude to
Professor Himanshu Prabha Ray for all the help with copyright issues and
contacting other contributors, and to Professor Uma Chakravarti for
battling commercialism in the publishing world for this Reader. Special
thanks to Dr Uma Shankar Pandey but for whose knowledge of French, I
could not have corresponded with Paléorient. Amit Chhikara thoughtfully
chipped in at a critical moment with library and photocopying work and I
thank him for that.
At Orient BlackSwan, Mimi Choudhury was receptive to all my ideas and
suggestions. Nilanjana Majumdar helped with complex negotiations
revolving around permissions that were an exercise in great patience.
Shubhro Datta copy-edited the manuscript with unflagging diligence. He
and I worked hard to bring uniformity and completion to the state of the
references and citations across the many essays, but with some of these
written a long time ago, gaps in publication details at places have been
difficult to fill. With the essays teeming with indigenous terms and
diacritical marks that accompany them, we decided not to use diacritics in
the Introduction and the Glossary, making these more accessible thereby to
the general reader. For the same reason we did not add diacritics to the odd
essay that did not originally use them, even though this then meant there
would be a slight variation in spellings in this book.
Finally, a word about my family. My father is no more but remains the
blessing in my life. My husband Nachiketa is a very loving support in all my
endeavours, and surely knows more history than I do! My sister Devika
continues to be her gentle self. My fourth standard school teacher, Mrs Usha
Chaujer, has gone from being ma’am to ma for me in the last couple of years.
In 2012, heartbreakingly, I lost little Tulip, my beloved dog of eleven years
and exemplary person; this book is dedicated to her abiding presence. I also
found littler Jim, a rescued dachshund, who has brought light and life back
into our home.
For these people and this book, I thank God.

New Delhi Shonaleeka Kaul

October 2013
 Publisher’s Acknowledgements

For permission to reproduce copyright material in this volume, the

publisher wishes to make the following acknowledgements.

Oxford University Press India for ‘Rock Paintings of the Mesolithic Period’,
in Erwin Neumayer, Prehistoric Rock Art, New Delhi, 1983, pp. 5–28. ©
Oxford University Press.

Paléorient for ‘Ornaments of the Indus Valley Tradition’. Originally

published as ‘Ornaments of the Indus Valley Tradition: Evidence of Recent
Excavations from Harappa, Pakistan’, 17(2), 1991, pp. 79–98.

Indian Historical Review for ‘Texts on Stone: Understanding Aśoka’s

Epigraph-Monuments’. Originally published as ‘Texts on Stone:
Understanding Aśoka’s Epigraph-Monuments and their Changing Contexts’,
24: 1–2, July 1997 and January 1998, pp. 1–19. © Indian Council of
Historical Research. Reproduced with permission of the copyright holder
and the publisher, Sage Publications India Pvt. Ltd.

K. P. Bagchi and Co. for ‘Social Background of Ancient Indian Terracottas’.

Originally published as ‘Social Background of Ancient Indian Terracottas
(circa. 600 bc–ad 600)’ in Debiprasad Chattopadhyay (ed.), History and
Society: Essays in Honour of Professor Niharranjan Ray, Kolkata, 1978,
pp.143–65.

Munshiram Manoharlal for ‘On Modes of Visual Narration in Early

Buddhist Art’. Originally published as ‘On Modes of Visual Narration’ in
Vidya Dehejia, Discourse in Early Buddhist Art: Visual Narratives of India,
New Delhi, 1997, pp. 3–35; 298–99 (Notes).
Aryan Books for ‘Ellora: Understanding the Creation of a Past’. Originally
published as ‘Unravelling the Picturesque: Understanding the Creation of a
Past’ in H. P. Ray and Carla Sinopoli (eds), Archaeology as History in Early
South Asia, New Delhi, 2004, pp. 98–117.

University of Calcutta for ‘Theory and Practice of Painting: Introduction to

the Viṣṇudharmottara'. Originally published in Stella Kramrisch, The
Viṣṇudharmottara Part III: A Treatise on Indian Painting and Image-Making,
Second Revised and Enlarged Edition, Kolkata, 1928, pp. 1–28.

Studies in History for ‘The Jataka as Popular Tradition’, 9(1), 1993, pp. 43–70.
© Jawaharlal Nehru University. Reproduced with permission of the
copyright holder and the publisher, Sage Publications India Pvt. Ltd.

Sarvamangalam Kailasapathy for ‘Bards and Bardic Tradition in Early Tamil

Poetry’. Originally published as ‘Bards and Bardic Traditions’ in K.
Kailasapathy, Tamil Heroic Poetry, Oxford: The Clarendon Press, 1968, pp.
94–134.

University of Hawai’i at Mānoa for ‘Who Needs Folklore? The Relevance of

Oral Traditions to South Asian Studies’, South Asia Occasional Papers Series,
1990.
 Introduction
Producers and Consumers of Culture
 SHONALEEKA KAUL

W HAT IS CULTURE? THE TWO COMMON, INTERRELATED

meanings are culture as way of life and culture as art. Broadly, the
former is an anthropological definition and the latter, aesthetic. In the
discipline of history, where the approach has tended so far to be closer to the
latter, individual detailed studies abound on early South Asian forms of art,
architecture, sculpture, music and literature. Volumes that do a sweeping
survey of all these aspects are also available, usually compiling major
theoretical principles or features of various arts and their important
specimens. This is often undertaken against the backdrop of the religious
context of these art forms—religion as a way of life that infuses and
motivates the production of these art forms—thereby making a survey of
religions, sometimes in their own right as philosophies, a regular part of
cultural histories of South Asia.
The present volume departs from these approaches; it attempts neither a
vertical nor a horizontal perspective, but one that is conceptual. It seeks to
contextualise, rather than objectify, cultural forms, looking at art less as a
static ‘heritage of the past’ with just exhibition or display value, than as
dynamic processes of meaning and communication in the past. In other
words, taking off from the contemporary discipline of cultural studies, this
book aims to connect cultural production with ordinary life, to explore
cultural form as cultural experience—the variety of roles that paintings,
sculptures, monuments and works of poetry or drama played in the lives of
their communities.
So in the several articles brought together here, art is investigated not only
as objects of aesthetic enjoyment, but also as creations of rhetorical or
philosophical moment, as well as of utilitarian value. These furnish the
variety of contexts of consumption within which this book seeks to place
early South Asian cultural forms. This is a departure from the emphasis on
patronage, or contexts of production, which has already received
considerable and able attention from scholars looking to historicise acts that
produce culture, like temple-building or composition of texts. Seeing mainly
materialist motives and processes behind the creation of culture is to adopt
the political economy approach—a cultural materialism that tends to shear
culture down to power-effects. But did all art serve little purpose other than
to perpetuate or mediate hierarchy and domination?
In departing from it, however, our emphasis actually takes the work on
patronage forward. One of the abiding contributions of the turn to analysing
the patronage of art in early South Asia has been the generation of a social
understanding of culture, an understanding of the socio-economic and
ideological underpinnings of art, thereby affording it a second dimension in
addition to that of a purely formal analysis. However, an essential but
neglected completive to any social history of art is looking at not only its
production but its reception and consumption. For art is hardly produced in
a vacuum, in oblivion of an intended or imagined audience: it is not only
about self-expression, but also, usually, about communication. It can be
regarded as an act of signification realised essentially through the practice of
viewership. And early South Asian art was largely public and performative.
This meant that its audience or clientele was an important part of
constructing its meaning(s). This is confirmed by the emphasis that Indian
treatises on art, from Bharata’s Natyashastra onwards for many centuries,
place on the rasika, who was the taster of rasa, the flavours or emotions
elicited by art forms. The very fact that appreciation of works of art evoked
enormous theorisation says a good deal about the conscious way in which
consumers of art were conceived of in South Asian traditions.
The public character of art also meant that the audience could well on
occasion have been a broad and complex social gathering, not necessarily
monochromatic in terms of class, caste, gender, ethnicity or the rural-urban
contrast. It could, therefore, have brought with it diverse interests and
expectations, which could also have changed over space and time. With this
focus on the audience or reception of cultural forms, this book examines
some entrenched assumptions such as the distinction between classical and
folk culture or high art and popular art, as also between the religious and the
secular in art. It explores the social presence of culture in early South Asia.
 This book proceeds with an understanding of culture as practice through
which a society makes sense of itself and the world it inhabits, and through
which it articulates its humanity. Art is arguably the most creative of those
articulations, usually enshrining a vision with sophistication and/or
profundity (hence ‘art’ is often used interchangeably for ‘culture’ in this
book). Accordingly, with an eye on the visions encoded in cultural forms,
some of the articles in this book look closely at representationality—not only
what is being represented but its how and why. This book thus seeks to make
a case for art as not only an aesthetic experience but also an intellectual
institution. The interaction of this institution with a heterogeneous
community as audience/clientele, over shifting spatial and chronological
contexts, can then immediately be seen as acquiring a complexity and
unpredictability that deserve to be investigated and articulated. And, given
the sometimes very widespread currency of artistic styles, icons and motifs
across early South Asia, this book also highlights the related issue of
transmission and mobility of culture.
In fact, the shared cultural forms and traditions across a landmass as vast
as the Indian subcontinent are what prompted the adoption of the
nomenclature ‘South Asia’ in preference to ‘India’ in this book. It is true that
these traditions have been, and can tenably be, described as Indic. It is also
true that distinct local and regional specificities and exceptions or
adaptations of Indic traditions obtained, often as a result of confluence with
non-Indic styles. Nonetheless, South Asia, broadly speaking, and not the
boundaries of the Indian nation alone, can be said to constitute the cultural
region.
Thus the geographical catchment area of the forms and tendencies
discussed in this book, though not always made explicit, spans Afghanistan
in the west, Bangladesh in the east, Nepal in the north, and Sri Lanka in the
south, including of course India and Pakistan. The chronological frame for
the exercise is just as sweeping, starting from the earliest known specimens
of human cultural expression, namely prehistoric rock paintings, through
protohistoric ornaments from the Harappan culture, pillar edicts of Ashoka’s
empire, post-Mauryan terracottas, the Buddhist stupas from Sanchi and
Bharhut, the ‘Brahmanical’ temples at Aihole, Badami and Pattadakal, the
caves of Ellora, the (theory of the) paintings of Ajanta, Sanskrit drama,
Tamil poetry, Pali Jatakas (birth stories of the Buddha in prose), and
variegated South Asian folklore.
The selection of the articles was not easy. I discovered through my
searches, which extended ultimately to enquiries made of experts in various
areas, that serious or sustained work on the audiences in early South Asia
‘simply has not been done’! This is only partly because the question of the
reception of art has not quite aroused the curiosity of art historians and
other scholars of early Indian culture; the other reason is the reticence of our
sources on the matter, which need to be prised open. In the event, though I
wanted to include only works which dealt with the question of the
community serviced by the art form they were discussing, the articles
selected do not always end up doing that directly or abundantly. This
introduction, therefore, tries to flesh out the issues and show the
connections. In the process it also raises questions about early cultural
production and communication that the essays here may not take up but
which belong in any history as a useful background to the cultural situation.

In the scheme of this book, the first of these questions is: Who was the artist
in early South Asia? Or, who were the producers of culture? Indeed, the
broader word ‘producer’ is more appropriate since ‘artist’ is often associated
with the performer of visual arts like painting and sculpture or dance, and
not usually understood to cover poet and playwright or folk bard, nor even
the artisan, though there is no reason why it should not include them. It is
also more appropriate because the artist does not seem to have been greatly
differentiated from the craftsperson in early South Asia, the term for both
being ‘shilpin' in all the early texts.1 Indeed, the two indigenous words for art
—shilpa and kala—connote essentially ‘skill at performance’, whether of a
practical or aesthetic or ritual kind. Thus, it appears, art was craft and craft
was art in early South Asia, the two embraced by the conventional sixty-four
kalas that are said to exist in the world. A more pervasive presence in the
day-to-day lives of people can thus be surmised for art viewed in this way.
Indeed, art was not only craft, it was also rite and ritual in early South
Asia. Its performance, and participation in it, was understood to have the
function of a yajna or sacrifice. Hence the ambivalence in the social status of
artisans: they were as an occupational group classed low as shudras,
presumably for being engaged in physical rather than intellectual labour. But
they were nonetheless regarded as having divine origins, being descendants
of cosmic architects Vishvakarma, Maya and Tvashtri. Hence also the textual
injunction that the hand of the craftsperson was ritually pure, as it was
engaged in the creation of objects of consecration.
Further, the punya or spiritual merit accruing from this ‘sacrifice’ is
believed to be the chief reason why patrons sponsored art in the first place
(this sponsorship was known as dana or donation / gift). The second reason
was likely to be public recognition and acclaim, and corresponding social
status. Royalty and others in positions of power making such donations
enjoyed the additional benefit of legitimation by association. Indeed, the
inscriptions that we find in large numbers associated with works of art are
those proclaiming the name and fame of the donor much more than that of
their producers.2 Art then certainly seems to have been an instrument for
society in more senses than one.
But was it just an instrument? Was the role of the artist merely executing
projects of power, social and political, and slipping passively back into the
wings? From what derived the capacity of art to be a vehicle of power in the
first place? First of all, cultural studies have shown that the relationship
between artist and patron—even the mightiest of patrons, the king—was
hardly a one-way affair or an unequal equation with the former being
dependent on the latter, and so his official mouthpiece. Not only was the
patron also dependent on the artist for legitimation or for propagating his
agenda, which is well known, but the artist on his own resisted the power of
the patron in a variety of ways and exercised a fair deal of critical
individuality or autonomy through his art.3 This is inevitable for a creative
medium in any case, a fact not given due consideration in modern studies
on patronage. Also worth emphasising is the fact that not all art was royal
art; most of the early South Asian art represented in this book was not. It
catered to broad sections and, evidence tells us, was also patronised by
diverse elements from among them. Indeed, it is the public nature of the arts
that made them worthy vehicles for royal messages or non-royal
advertisements of status, in the first place.
 A second clue to the power or popularity of art may lie in asking: What
did the production of art—or cultural forms—primarily involve and intend?
It is perhaps to state the obvious that a deference to aesthetics, to the
creation of beauty variously defined, lay at the heart of cultural expression.
As such, its primary function would be to produce pleasing perceptions, to
entertain. No doubt any cultural form, be it a poem or play, a sculpted relief,
a painting, or a piece of magnificent architecture, when meant for public
consumption, would invoke an element of spectacle, of entertainment. But
to conceive of so limited a purpose internal to cultural production is only to
duplicate the inadequacy of the political economy approach that emphasises
a narrow purpose external to it, like social legitimation. It may also be to
commit the fallacy of ‘making of works of art a kind of aphrodisiac’4 and of
all creators and consumers of art in early South Asia, simply aesthetes!
Certainly texts like the fourth-century Kamasutra, a treatise on pleasure and
good living for the nagaraka (the urban connoisseur), prescribe mastery
over arts and crafts as part of a cultivated, refined way of life. But witness the
earliest extant treatise on aesthetics itself, the second-century Natyashastra
that went on to influence the theory of virtually every other branch of art: it
mentions not only the production of delight (priti, vinoda) but also
anukarana or imitation/mimesis of the ways of the world, and most
significantly, upadesha or instruction, as the goals of an art form like the
drama (natya, more properly dance-drama).5 A clearly contemplative aspect
to artistic production is thus indicated. Besides, cultural forms themselves—
their themes, motifs, symbols, methods and modes of depiction—all testify
to enormous investment (of intent and attention) in representationality. And
inherent to any representation is meaning.

These are precisely the points, not always recognised or asserted in studies of
art, that the first two essays in this book make. The piece by the father of
South Asian art history, Ananda Coomaraswamy, may be six decades old
but raises a basic question of understanding instructive to this day: Is art an
aesthetic or a rhetoric, i.e., is it a feeling or a meaning? His answer is a
passionate affirmation of intellect in art, ‘a theory of art as the effective
expression of theses’, much like the Greeks understood it. Further, he
suggests regarding art as a language (rather than a spectacle) regardless of
the medium in which the artist worked. And the traditional science of that
language, alamkara shastra, he argues, corresponds not to a category of
aesthetics or empty ornamentation, so much as to the category of rhetoric.
In the latter, eloquence or beauty is not an end in itself—art for art’s sake;
nor is it just to display the artist’s skill, which is the usual
understanding/criticism brought by scholars to the study of alamkara or
ornamentation. It is essentially a means of effective communication.6
Communication of what? Coomaraswamy has been criticised for reading
too spiritual a meaning into art, as when he speaks of it in this essay as
meant to ‘assist the soul’s interior revolution, to restore it to order and
concord with itself ’, to adhere to ‘our knowledge of what is good and bad for
us’—a concept of ‘liberating art (vimuktida)’. Nonetheless, his work makes a
fundamental point by emphasising the union, not separation, of beauty,
philosophy, and utility in art. These, incidentally, constitute the very contexts
of consumption—the functions of culture—spelt out in the beginning of this
introduction.
Once we foreground meaning in our approach to cultural forms, yet
another role played by art and the artist in early South Asia emerges. It was
an overarching role, an umbrella function that subsumed all others and
located ‘meaning’ within a larger cultural process/order. This role was the
giving of form to and the transmitting of tradition. Tradition may be defined
as values and practices an individual participates in as part of a community.
What kind of values did art transmit (or what kind of meanings did art
communicate)? It would appear these were aesthetic, intellectual, ritual and
civic values, apart from the political and social overlays. Where did these
values come from? Often they came from the process of artistic
interpretation and rendition itself.
But was the making of tradition a one-way process, a handing-down of
values conceived by artistic convention or innovation? Or, more likely, did it
operate in a dialogic fashion, creating meaning only when consumed by the
community within a shared system of signification?
The very fact that, as several articles in this book underline, in early South
Asia there was not a single tradition but plural traditions in every branch of
culture, variously intersecting or diverging but always interacting—is evidence
of the central role played by variegated communities (‘consumers’) in
constructing the meanings of cultural forms. To transmit tradition was thus
not always to endorse it but to engage with it and reshape it in as many ways
as there were communities of response. In turn, traditions shaped
communities that conceived them and gave the latter their identities. In this
sense, the cultural was necessarily implicated in the social world, but
perhaps in a subtler ideational way as much as in a political/ideological way.
More on this later. For now, let us turn to the second point:
representationality and its relationship with the crucial shaping factor of the
community.

A thorough examination of Mesolithic rock paintings in the Indian

subcontinent, which were the first ‘forms of expression external to their
authors’ physical needs’, Erwin Neumayer’s essay throws up several points
important to cultural history. It concentrates on the region of the Vindhyas
which is rich in prehistoric sites like the world famous ones of Bhimbetka
and Pachmarhi. Whether it is the location of some of these paintings in
dangerous, virtually inaccessible, places in the rock shelters such as at
dizzying heights, or the use of colour made from raw materials brought
from long distances, or the abstract, symbolic representation of subjects
(humans), or the sheer detail and variety of theme and technique including
the depiction of motion and perspective—all bring home the self-conscious,
purposive nature of so-called ‘primitive’ stone-age art which is somehow
expected to be unthinking, innocent and crude. At the same time,
Neumayer tells us, rock painting does not seem to have been an activity
invested with any particular prestige or practised by an elite; it was produced
by artists who were not distinguishable from the community. Not
surprisingly then, the commonest objects of representation are mass
communal activities: dance, music and chants, with more than a hundred
participants depicted at times, are the most widely recurring cultural
symbols in Mesolithic art. Intriguingly, here is one cultural form (painting)
capturing another (group dance/ritual), which in turn displayed an inner
culture consisting of the use of masks, ornaments and body paint depicted
on the dancers and singers.7
The second major theme represented, though not in the earliest paintings,
is large hunting expeditions in carefully organised groups. It seems that both
the dances and the hunts were special, extraordinary events or gatherings
with a ritual or cult significance for the community that enacted and then,
indeed, recorded them. Neumayer believes that the paintings themselves, for
this reason, had a cultic significance, perhaps a sacredness attached to them
(though they could of course simply be for decoration). This would not be as
relevant for some of the domestic scenes of daily activity also represented
among the hundreds of cave paintings. Domesticity, however, again points
to the social character of cultural expression in this early period of creativity.
An intriguing feature is the striking uniformity of style, theme and
execution in rock paintings sometimes hundreds of kilometres apart, which
in prehistoric terms would have covered the provinces of multiple
Mesolithic groups. Thus, very early in South Asian cultural history we run
into a trend—namely, noticeable uniformity—that conspicuously marks
later periods of cultural production as well, over larger and larger areas. Is
this to be understood as an instance of cultural transmission, which in turn
would need to be explained, or as evidence of a ‘very uniform culture and
code of expression’, as Neumayer puts it, obtaining in a specific kind of
society?
It is hard to prove anything, even to argue, for a period for which we have
scarce sources; a strong element of speculation necessarily inheres in our
reading of prehistoric art, indeed prehistoric people. Often this speculation
is from the perspective of the present, or that of other, later periods of
history, with their vastly different technologies, social organisation, and
aesthetic inclinations from those of the period under study. This gives rise to
what Neumayer rightly calls ‘the totally chauvinistic division between
primitive and classical art’.
Yet, it is from a comparison of cultural practices across periods that
certain tendencies also become significantly apparent. For instance,
Mesolithic paintings invariably have overlapping layers of painting,
sometimes as many as twenty such layers, from succeeding epochs like the
Chalcolithic and the Early Historical and Medieval down to even Modern
times! This shows that local communities continued to use the same surfaces
and locations for art. clearly, prehistoric artists had inaugurated a tradition
not only of cultural form but of cultural site: places regarded as sacred or
special, often precisely because they had the sanction of antique usage.
(There could be other reasons, too, for this practice which will become clear
for other cultural forms discussed in this book that are united by this resort
to shared cultural spaces.)

As we move into the protohistoric in South Asia, namely the Harappan

culture or the Indus Valley civilisation, there is a great increase not only in
the archaeological sources available but also in the forms of cultural
production. Indeed, we get a variety of arts and crafts from the earliest
urban society of South Asia. Of these, I have selected objects usually studied
for what they reveal about the technology and material culture of the time
rather than about the beliefs of the people: ornaments. There are several
reasons for this choice. Ornaments combine the functional character of
crafts with the aesthetic appeal of the arts. What is more, as Jonathan
Kenoyer’s excellent essay tells us, they have an extremely long and
meaningful history in the whole South Asian region as social symbols—
symbols of age, status, ethnic identity, and religious affiliation. They also
served to protect the wearer in the form of amulets and serve as a means of
storing wealth. And it is somewhat easier for the historian to arrive at these
meanings because of the availability of an abundant archaeological context
and the possibility of ethno-archaeological comparison with ancient texts
and folk traditions, which teem with the use of these ornaments that are also
reflected in seals and figurines produced by the civilisation.
Utilising these methods, Kenoyer argues that the raw material used, the
technology of production, the colours and designs employed, and finally,
who wore them and how, together determined the cultural meaning of
ornaments. Accordingly, it was possible to rank ornaments made of precious
and semi-precious materials like gold, silver, copper/bronze, agate,
carnelian, jasper, turquoise, faience and steatite, down to terracotta,
probably in that order. The worth of an ornament was governed not only by
its inherent value but by the availability or rarity of the raw material in the
region and by the degree to which it had been worked on. These two factors
together defined the style of the ornament. Distinctive styles suggest a
complex ensemble of symbols and identities they represented in what was
probably a ranked, if not stratified, society in the Indus Civilisation.8 And
given the range of raw materials—from the highly precious to the humble—
in which identical shapes have been reproduced in the Indus tradition, these
symbols would have been accessible to all members of society. This would
have ensured, as it were, a joint belief system or cultural vocabulary across
the social order.
However, a study of burials and the grave goods contained in them along
with the bodies of the dead, shows that a range of roles are apparent within
this larger cultural understanding of the purpose of ornaments. Thus, while
some ornaments that have been found under the shrouds covering the
bodies must have been meant for specific ritual purposes, and those worn
under the clothes were obviously highly personalised ones, those ornaments
that were meant to display social standing and wealth were not interred in
graves at all. They have been found in hidden hoards in houses instead. They
were retained and passed on from generation to generation to be worn by
the living. Amidst the living, as explained, they would be part of what
Kenoyer believes was a ‘highly efficient form of visual communication’. They
would serve as both public identifiers of difference in an increasingly diverse
urban population as well as unifying symbols representing a shared aesthetic
prescribed by and subscribed to by the community as a whole. Here was
cultural form as lived cultural experience.

Culture as communication in early South Asia took a variety of forms.

While with the Harappan beads we see a community in conversation with
itself, or communities in conversation with themselves, as it were, the
Ashokan edicts from the early historic Mauryan period represent a new
voice on the scene: the king, addressing the multiple communities that
inhabited his empire. Among the edicts, those on pillars are unique because
they represent three different cultural media—architecture, sculpture and
text—all at once. For the first time in stone, tall, highly polished columns,
usually topped by magnificently sculpted animal capitals (standing for royal
power or the Buddhist tradition) and inscribed with orders and
proclamations of the king, loomed over the landscape, dominating it, hard
to ignore. Together they comprised an artistic and epigraphic whole that, as
upinder Singh in her essay says, conveyed its message of imperial authority
and piety to an audience as varied as officers of the state on the one hand,
and forest-dwellers on the other, with monks and laymen thrown in.
She argues that, given its elevated placement and the limited literacy of
the time, the message was not meant to be directly consumed by its intended
audience, but perhaps verbally propagated in ceremonial, congregational
readings on certain occasions. The oral and performative aspect of early
cultural communication is seen here.
The sheer form and art of the monuments then—the symbolisms and the
loftiness and power they were meant to impress upon onlookers—perhaps
conveyed more of the regal message than did the text of the message itself.
So much so that long after the Ashokan inscriptions ceased to be intelligible,
the awe-inspiring pillars bearing them continued to compel attention and
were appropriated by later kings down to the medieval period to
communicate their own power.9 In this context it is instructive that the
edict-bearing Ashokan pillars may themselves have drawn inspiration from
preceding Achaemenid royal art practices.
Interestingly, parallel to this phenomenon, the pillars seem to have
acquired an altogether different association and significance in popular
culture where they came to be memorialised as the walking sticks of
mythological characters from the Mahabharata and Ramayana! In time,
they also came to be worshipped as such, much in the fashion of the phallus
or linga.10 This would mark a radical shift in their meaning for the local
community, and yet belong broadly within the cultic role these striking
objects of ancient South Asia obviously played from nearly the beginning.
For, we know that the original Ashokan pillars were erected on major
pilgrimage routes of the time and appear to have developed as centres of
Buddhist mythology and worship. It is thus clear that the same cultural form
stood for different things for different interpretive communities11 across
space and time—a threat for tribals challenging the state; a command for
officers of the king; spiritual counsel for believers and institutions of the
Buddhist faith; moral exhortation for people at large; a tool of sovereignty
for later kings; and cult anicons and even tourist attractions for later social
groups right down to the present.
In what appears to be very different from the stately Mauryan imperial art,
our next object of study are terracottas from the same period as well as a
little before and after, i.e., from 600 bce to 600 ce. Much has been made of
the contrast between ‘art from above’ and a people’s art, the terracottas or
baked clay figurines being said to represent the latter.12 They have even been
described in the seminal article by Devangana Desai included here, as the
‘spontaneous handiwork of the simple folk’. Categorisations like these are
defied by the diversity that terracotta art evidence presents (see below).
However, the categorisations do help draw attention to the important
possibility of multiple strands of cultural production co-existing. Whether
those strands are best understood in terms of binaries like classical and folk,
elite and popular, urban and rural, or stone and clay remains to be
examined. Desai’s essay and the case study of ancient South Asian
terracottas provide an opportunity to raise these questions fundamental to
art and cultural history.
Post-Mauryan terracottas, that have been found all over, from Charsad in
Pakistan to Chandraketugarh in West Bengal and Sannathi in Tamil Nadu,
represent a ubiquitous cultural form that catered to many needs of the
communities that produced and consumed them. They served as votive
offerings, magical charms, household or roadside deities, as well as
decorative plaques for homes and toys for children. It seems that the cultic
or ritual function—epitomised by the so-called Mother Goddess figurines—
was initially the predominant one of the two kinds of roles terracotta
figurines played, historically speaking; it is said to be visible even in the
Neolithic settlements of the north-west and then in the Indus valley. Their
decorative and artistic aspect developed only as early historic urbanism grew
and developed in the Ganga valley and beyond.
Indeed, Desai emphatically characterises terracottas as ‘the art and ritual
objects of an urban culture’. They had ‘as their clientele the urban people
whose needs, interests and tastes are reflected in them’. In something of a
contradiction, however, she observes also that ‘tribal and peasant cults and
rituals are represented in terracottas, but their treatment is marked by urban
sophistication’. The reason for this tribal or peasant content presumably was
that while terracottas have been found from urban sites, they seem to have
been produced very much in the villages of potters that lay outside the
towns, the markets of which they catered to.13 Such villages are frequently
mentioned in texts of the period.
Now, either we might be completely overlooking the possibility of
terracotta production within cities. Or, if we accept the non-urban
authorship, flavour and substance of these cultic terracottas, we must ask
how far this cultural form was really urban. Does stylistic sophistication
alone constitute urbanity in art, and is ‘sophistication’ the provenance of
only the urban?14 Such a thought or expectation may come naturally to us.
But are value judgments not implicit in this association and its converse,
which would be to not attribute the coarse and the crude to the urban and
therefore be surprised at the occurrence of precisely such ‘coarse and crude’
figurines in large numbers from the highly developed urban phase of the
Kushana period? This is in fact the dilemma Desai finds herself in. Indeed,
how far are distinctions like urban and rural useful when discussing the art
of traditional communities that apparently, and not surprisingly, held ritual
beliefs, motifs and practices in common? Desai is certainly correct in
identifying ‘secular’ terracotta depictions of genteel men and women
frolicking in garden parties and cultural soirees (gosthi) and such like
convivial or erotic themes as rather urban and elite in origin. But no less
common, perhaps only more so, in the findings from urban sites, are the
figures of yakshas, yakshis, nagas, nagis, dwarves, gandharvas, and so on,
who evoke what are often called folk cults but were clearly popular in cities.
Similarly, as Desai says, the depiction of shalabhanjika-type figures
associated with vegetation may have been but a secularised and sensualised
version of agricultural fertility rites.
It is not my purpose to argue that urban and non-urban cultures,
whatever these may denote in the present state of our application to these
notions, were identical. But we should be open to the possibility that what
we are looking at may be a partially shared repertory of themes and objects
of veneration and celebration across town and country. Permeable
boundaries and integration, or at least interaction, with hinterlands and
influences near and far were realities of the early South Asian cultural
situation. The same process would explain, after all, the ‘foreign’
Achaemenian and Hellenistic elements in sculpture, including terracottas
fashioned in the subcontinent in the Mauryan and post-Mauryan periods.
 The idea that all terracotta sculpture was people’s art—while stone
sculpture was elite/royal—should perhaps also be reassessed. Desai’s
detailing of specimens shows that terracottas were produced at all levels of
technique and treatment, from large and fragile statues meant for palaces in
the Mauryan period, to dainty forms and delicate workmanship in white
clay in Satavahana realms, to run-of-the-mill, even coarse, products of
Kushana times. They were used in households, palaces, street crossings, and
temples. They could and did assume high technical perfection and finesse, as
in the Gupta period temple reliefs, where we also find reportedly folk motifs
like yakshas and nagas rendered in ‘elite’ stone. The lesson in this is obvious
but not always heeded by scholars keen to find some order in the
bewildering diversity of the early South Asian cultural scene: a cultural
medium could be versatile and used for representations in multiple registers
across time and space; it can be regarded as popular or elite in itself only at
some risk.

Coming to monuments proper, I have chosen two most representative types

of sacred South Asian architecture that arose in the last few centuries bce:
the Buddhist stupa and the ‘Hindu-Brahmanical’ temple. Needless to say,
these are not the only architectural types we have remains of; but they
dominate the archaeological landscape of the region in terms of visibility
and significance, are cultural forms invested with enormous symbolic value,
and have stood at the centre of ritual and of various complex institutions
and communal activity. Any cultural history of the subcontinent would be
incomplete without considering these major, multifaceted expressions of
faith, philosophy and iconography that sported a variety of relations with
the social world.
A large amount of interesting work has been done on the stupa.15 I have
chosen Vidya Dehejia’s essay since it is concerned centrally with the
question of viewer-response or the dialogue between the cultural form and
its audience. Studying viewer-response, in fact, as Dehejia acknowledges, is
highly appropriate in the context of the subcontinent where the rasa theory
has reigned supreme in age-old canons of culture, including those for
literature as represented in the Natyashastra onwards, and those for
sculpture and painting as represented in the Chitrasutras. As explained
earlier in this introduction, the rasa theory refers to the eliciting of
abstracted emotional responses from the reader or viewer upon his
‘consuming’ an art form. The success of the artistic endeavour is seen to
depend on the degree to which it is able to achieve this effect and evoke this
response in the audience.
Much like a literary culture, visual cultures were constituted into being by
early South Asian sculpture and painting. These had their own modes of
representation and codes of reception. Dehejia approaches some of the
earliest Buddhist stupas of South Asia in search of these modes and codes.
In her essay, she examines the narrative sculptural reliefs in stone that stupas
bear in profusion, either on the dome of the stupa itself or on other
structural parts like the railings that encircled it and the toranas or gateways
that marked off the sacred enclosure from surrounding space.
The content of these reliefs is well known—stories from the life and gospel
of the Buddha found in the later Avadana literature and from the Jatakas,
tales of the Buddha’s many prior births, each espousing a Buddhist ethic or
ideal. Incidentally, this again points to reflexivity or intertexuality, i.e., one
medium of expression (sculpture) was informed by another (text) with a
shared content. When discussing intertexuality in the South Asian context,
one must keep in mind the orality of many of these literary texts, which
would explain their wide circulation and easy accessibility to society at large
and artists in particular. Nonetheless, the phenomenon of themes and
theories common to art and literature, as also Dehejia’s larger stress on
narrative discourse in art just as in literature, highlight the close parallels
that populated apparently discrete branches of culture in early South Asia.16
What interests Dehejia is not the content of the narrative reliefs but the
ways in which it was visually represented and communicated at these stupas.
She identifies seven ‘modes of narration’ employed at different stupas, from
the very pithy and synoptic to the elaborate and sequential, and discusses
how these must have involved and drawn the viewer into the story and its
message in a number of ways. For example, the monoscenic mode, where
less of the story was shown, and more was left to the viewer to supply—to
recall, reconstruct, and imbibe till he became practically the narrator of,
even a participant in, the story!17 This would no doubt have entailed a more
intense viewing experience. In the process, the physical act of
circumambulating the stupa, lined by these reliefs, would have transformed
into a contemplative and spiritual experience for the pilgrim. For the monks,
it would have been tantamount to chanting as they surveyed and meditated
on visual reminders of the creed, like depictions of the Triratna (standing for
the Buddha, the dhamma, and the sangha).
It is further argued by Dehejia that the stories ‘narrated’ were already
known to viewers and monks alike.18 This would be, again, due largely to
oral propagation, given the presumably low literacy levels of the time as well
as the predisposition to orality as a cultural technique. It is a knowing
audience that allowed the sculptor-artist the option of a range of narrative
strategies to choose from, some of which could not have been decipherable
without this prior knowledge. The theme to be depicted was probably the
individual, personalised choice of the donors, as suggested by the repetition
of some evidently popular stories on the same stupa. The Vessantara Jataka
seems to have been a favourite; it emphasised the spiritual value of dana or
gift-giving, thereby perhaps echoing the value of the gift made by the donor
at the stupa herself/himself.
We already know from the rich inscriptional records at stupas like those
at Sanchi and Bharhut that the decoration and elaboration of stupas enjoyed
collective patronage at this time. They were sponsored by hundreds of the
faithful—men and women, lay and ecclesiast, locals and outsiders, villagers
and townspeople, craftsmen and traders, and some of the king’s men.
Reasonably assuming that patrons belonged to the ranks of visitors and
worshippers, we can infer a diverse and partially shifting composition of the
community to which the stupa catered. There would be periodic or irregular
visits by variegated lay worshippers and pilgrims from settlements,
especially urban ones, in the vicinity. The stupa’s location on major trade
routes would have been further responsible for the itinerant component of
visitors. Then, a partially fixed or bonded audience can also be assumed in
the form of monks and nuns, who resided in viharas that usually accompany
the location of stupas and chaityas. They would be required to regularly
worship at the sacred site. Dehejia tells us that there is evidence to suppose
that they may even have guided laity through the narrative friezes. Thus, a
host of sometimes-overlapping groups—artist, patron, viewer, and monk—
most from within and one from without the social world, playing various
roles, came together in the joint performance of a prominent visual culture
from early South Asia centred on the Buddhist stupa.
Though sacred shrines would have visually dominated their surrounding
physical landscape, we do not know much about that landscape itself. This is
particularly true for the Hindu-Brahmanical shrine or temple (perhaps
better described as Vaishnava or Shaiva or Shakta, etc. shrines). The
sprawling imperial temples of the eleventh to fourteenth centuries, those of
the Cholas (Tamilakam) and Gangas (Odisha), have certainly been
investigated for their links with political and socio-economic formations of
their respective regions. But the early temples of South Asia, dating as far
back as the third century bce and no later than the middle of the first
millennium ce, have not received the same attention, and almost certainly
cannot be approached with the same theoretical assumptions and
formulations as medieval temples. Indeed, when it comes to all things
Brahmanical in culture, the tendency is to see political patronage and
legitimation as the prime moving force. This is part of yet another
constructed binary, that between ‘popular Buddhism’ and ‘elite Brahmanism’,
which does not stand historical scrutiny, not least because both creeds have
had their share of royal patrons as well as popular following.
The point, therefore, is not to deny that temples had political uses for the
royalty, but also to explore the very real relationship of temples with their
communities. For there is adequate textual and epigraphic evidence of the
role played by private individuals and groups in the construction,
consecration, expansion and maintenance of Brahmanical shrines. And
there is also archaeological evidence for the role played by the temple in the
life of the people who formed its clientele.
Himanshu Prabha Ray’s essay presents some of this evidence for the
artistically rich, chronologically and stylistically intermediate temples of the
early Chalukyan regions. These include the temples at Pattadakal, Aihole
and Badami, and surrounding areas in the Deccan. She explores what she
calls the social history of these temples in the culturally unified Malaprabha
river valley, and points to a number of ‘temple writings’ in and around these
sites. Interestingly, several of these are in Kannada, the regional language,
apart from Sanskrit and the odd one in Prakrit, and are authored by a range
of social groups. Her analysis suggests no significant royal presence at these
sites for the early period. Instead, as her other analyses on the Chalukyan
temples suggest, on view are a number of individuals, sometimes in a
collective, like bhattas, settis, acharyas, sculptors, stone-cutters, architects,
etc. They inscribed not only details of gifts but also names of visitors on
monumental stones (nisidis) or temple walls.19 Ray says that the diverse
gatherings these temples would attract on festive occasions would give the
political elite a platform to address a wider public.
Ray’s archaeological analysis also points towards trading linkages and
agrarian bases in the vicinity as credible (non-royal) sources of sustenance
for the temples, much like what her earlier works have proved for Buddhist
establishments. Moreover, she alludes to the continuity of sacredness that
these sites represent by virtue of their prehistoric origins, which made the
selection of these points for the location of the early historic temples
probably a local choice—‘sanctity in a regional context’. The temples then
embodied, in a sense, the cultural identity of the locality.
Taking off from this location, the most compelling of her observations to
my mind, is that there appear to be a multiplicity of faiths sharing the same
sacred space. Of the cave temples of the early period, there was a Shaiva, a
Jaina and two Vaishnava temples at Badami; and one Shaiva and one Jaina
temple, and one Buddhist vihara at Aihole. This phenomenon gives an
altogether new dimension to the diversity of communities that could be
linked to a temple in early South Asia. Nor is it one of a kind; it holds true
also for later temple centres, like Khajuraho which saw Jaina temples co-
existing with Vaishnava and Shakta ones. In a slightly different way, we
know that Bodh Gaya was regarded as highly sacred by Buddhists and
followers of Brahmanical faiths alike from very early times. Indeed, the next
essay in this book, the one by Jaya Mehta, needs to be read together with this
argument, for it reiterates the point in the context of another culturally rich
site from the Deccan, Ellora.

Excellently put together, Mehta’s essay demonstrates the multiplicity of

associations, identities and claims to Ellora, a natural tirtha (river-ford,
hence also the word for pilgrimage spot) in the middle of the first
millennium CE. Here we see Brahmanical, Buddhist as well as Jaina
architecture and ritual (all three seen also at Mathura for an earlier period,
incidentally). Apart from showing ‘a lot of imitation, replication and sharing
of a common vocabulary of art’ and imagery, the edifices of all three faiths
display a prominent deference to the goddess tradition, carrying
representations of saptamatrikas, devis, and shaktis like Ambika, Indrani
and Saraswati. This ‘points to a strong following of the local goddess (river
goddess) tradition in the area’, a belief captured in the local texts too. It can
be reasonably inferred, then, that the local community’s cultural beliefs and
inclinations directly shaped this representational emphasis. And the
representational emphasis, willy-nilly, forged communities of response or
clientele for the faiths and their shrines. The shrines thus became a way for
communities to repeatedly reappropriate the past and renegotiate the
present of a cultural space.
What is the import of this trend for the cultural history of early South
Asia? For one, a single site could have plural affiliations. This would have
made for interaction and negotiation among diverse interest groups. At the
same time, we also get instances of seeming cross-patronage, with persons in
early Indian history, most famously kings but also commoners, sponsoring
any two or more of Vedic, Shaiva, Vaishnava, Buddhist or Jaina practices and
institutions simultaneously.20 Thus a single community could also sport
multiple allegiances.21
The sharing of cultural space would extend beyond worship and
pilgrimage to other events and activities that centred on shrines, such as
festivities and teaching/learning, which would involve the communities,
perhaps in an interactive way. These functions taken together inscribe the
temple as a quintessentially popular institution, the life of which was
dependent on the participation and practices of its clientele.
However, as scholars including Mehta above have observed, the cultural
forms of these different religious traditions, like temples and chaityas, drew
on a certain set of the same artistic signifiers and motifs. These included
yakshas, nagas, vidyadharas, apsaras, shri, swastika, kalasha, mithuna, etc.
which are found depicted as much on Brahmanical structures as on
Buddhist and Jaina ones.22 Also, the cultural forms would all have dipped
into a common pool of architects and sculptors in circulation, a professional
class of artisans believed to have shown a striking horizontal (and linear)
continuity, i.e., spread, as the characteristic of its trade in early South Asia.23
In fact, one of the reasons for the transmission of artistic styles and motifs
across this macro region is believed to be the itinerancy of artisans.
All this taken together, there is reason to speak of what scholars like Desai
have called ‘a cultural substratum’, which others like A. K. Ramanujan have
always suspected exists.24 This refers to a common reservoir of symbols and
meanings underlying differences of doctrine. In fact, the evidence reviewed
in the last several paragraphs is so leading that it begs the question: Do the
appellations Shaiva, Jaina, Vaishnava, Buddhist, and so on, represent so
many discrete communities in ancient times, after all? The boundaries among
religious/sectarian traditions that we assume today appear to have operated
rather differently then. This might help us correctly comprehend, or at least
more correctly probe, ritual-cultural praxis in early South Asia, beyond
paradigms of royal legitimation.

With Stella Kramrisch’s piece we enter the world of painting in early South
Asia as laid out in the exhaustive canon on the subject, the
Vishnudharmottara Purana. This is believed to have been compiled in the
seventh century CE. Taken together with the preceding centuries of art
practices that it would cumulatively reflect, as any technical treatise
invariably does, the Vishnudharmottara's sections on painting are thus
contemporaneous with the mature phase of painting at Ajanta and at sites
like Sigiriya and Anuradhapura, in Sri Lanka, which belong to the same
tradition as Ajanta.25 Admittedly, the text contains but the theory of
painting. However, as Kramrisch emphasises, like all theories, it is derived
from and subservient to the practice of the art. In the same context, it is
important that the theory of painting in the Vishnudharmottara itself ‘left
every freedom to the artist to work “according to his own intellect”’. The
suggested say and autonomy of the artist is instructive for all those who
believe that early South Asian culture was ruled by convention. It is much
better appreciated as the confluence of canon and practice, of tradition and
innovation.
 Kramrisch’s essay brings out very well that early South Asian painting was
also the confluence of imitation and imagination, of description and
suggestion. In its use of a variety of techniques like foreshortening,
proportion, and shading, it was science. And in its privileging of mood and
expression, it was story-telling. Indeed, the conjoint stress on observation
and imagination, and on emotion, was common to painting, sculpture,
literature, and dance in early South Asia, making these kindred arts.26 This is
proved also by the fundamental importance of the rasa theory in all of these,
and by the use of colour symbolism, i.e., the differential use of colours to
denote different characters and characteristics, detailed in the Natyashastra
and Vishnudharmottara. The latter also explicitly prescribes a knowledge of
dancing movements, body gestures, and rhythm as imperative for attaining a
knowledge of painting. Unity of early South Asian cultural forms is thus
reiterated.
Who was this exercise in realism and expressionism that painting
represented geared towards? Who were the audience or consumers of
paintings? A lot of different people were, it would seem, going by the
Vishnudharmottara and other texts like the Jatakas, kavyas and Kamasutra
which Kramrisch cites. The Vishnudharmottara speaks of the variety of
tastes, from the refined to the simple, that different aspects of a painting
appealed to and was required to satisfy (again echoing the Natyashastra).27
The range of consumers is also indicated by the range of materials or locales
that paintings appear to have been made on. They were used as decoration
or narrative, especially for festivals, on the floors, walls and ceilings of
private houses, palaces, temples and viharas alike. They were made on
boards as didactic tools used by sages, and also on canvas rolls (yamapata)
displayed by itinerant showmen in the streets and in courtyards of dwellings.
Painting thus comes across as one of the most democratic cultural forms in
early times, widely available and visible to its community, and enjoying quite
a social presence.

Let us turn now to the last four essays of this book, which cohere into a
section of sorts on literature broadly. Uma Chakravarti on Pali Jatakas; my
own essay on Sanskrit-Prakrit kavyas (ornamental poetry and drama); K.
Kailasapathy on early Tamil poetry, and A. K. Ramanujan on variegated
folklore (tales, songs, proverbs, riddles, etc.) present studies on four very
different genres of narrative and cultural expression. But they throw up
common issues, namely, orality, language, and performance; popular, folk
and elite traditions; multiplicity and questioning of tellings; cultural
appropriation, transmission and profusion, and so on. In fact, most of these
issues relate, as we have seen, to the larger cultural situation in South Asia
that we have been reviewing so far. Literature, in its different forms, helps
underline the complexity and fluidity of that situation.
So much anthropology and social history have shown the limitations of
hidebound categories like ‘popular’ and ‘high’ culture or Little and Great
Traditions, and their interpermeability. Even as she goes along with these
dichotomous categories, Chakravarti’s first-rate analysis of the multiple
layers to the structure and intentionality of the Jatakas proves their
limitations. It also instantiates the interweaving and cross-referencing of oral
and written traditions that were widespread in early South Asia. The Jatakas,
says she, are drawn from folk tales about common people that floated about
in oral form; as such, they must have been authored by ‘sections who are
normally outside the arena of intellectual production’ or ‘subordinate
classes’, which is why they constitute popular literature for her.
If this kind of statement on the authorship of a text sounds vague, it
serves to illustrate an important point for appreciating cultural dynamics in
the early period. In an ancient oral context, literary processes in the
subcontinent were marked by difficulty of pinpointing singular, well-defined
origin and character. They were inherently fluid and, be it the Vedas, the
epics, the Puranas, the Sangam anthology, or creative literature like the
Panchatantra, the Dashakumaracharita or the Kathasaritsagara, texts almost
always contained material that belonged to more than one strata of time,
place, and ideology. Thus, while Chakravarti calls the Jatakas popular
literature, she herself demonstrates that in the process of their compilation,
these tales were appropriated and critically altered by monk-editors who
were representing and advancing values of ‘high Buddhism’.
Not believing this to be a total contradiction, Chakravarti speaks of the
grey area that is the relationship between popular and high culture. She has
other reasons to maintain that the nature of the texts was popular (she
alternatively calls it ‘folk’, not distinguishing the two terms at all). Among
these, she speaks of Pali, the language in which the Jatakas are composed, as
‘closer to the language of the people’ than Sanskrit which was ‘a language of
a fairly restricted group even among the elites’. She also cites the Jatakas
being orally rendered or performed as being a sign of the popular. While we
shall return to the points about language and performance, Chakravarti’s
chief contribution in this context is emphasising the role of the audience.
Though unable to quite reconstruct the audience of the Jatakas specifically,
she argues generally, after Roger Chartier, that cultural consumption is also a
form of cultural production. Audiences, far from being passive spectators,
may recompose a text during viewership to permit a plurality of meanings.
Indeed, open-ended meaning is what lies behind another remarkable
literary phenomenon: multiplicity of tellings of a traditional text like the
Ramayana. As Paula Richman and others have shown, a rather than being a
closed text the epic represented a motif, that of the story of Rama, which was
successfully appropriated many different times, simultaneously as well as
across centuries and regions, by social groups seeking to express different
worldviews.28 This is an important new way of comprehending literary and
cultural processes. It means that one popular motif could lend itself to
multiple forms, fashioned by multiple communities of interpretation. It was
a situation of cultural profusion rather than diffusion, and may provide a
model for understanding the transmission and sprawling currency of certain
motifs, styles and icons in early South Asian sculpture, architecture and
painting too, which we have seen.

The next essay analyses a literary genre which is the seeming opposite of the
Jatakas. Kavyas were poems and plays (in the main) composed as art, i.e.,
highly literate, aestheticised compositions in Sanskrit as well as Prakrits.
Historians have typically regarded them as classical literature, born of a
courtly culture and meant for the pleasure of elite audiences. In my essay in
this volume, as well as in my book, I have urged a substantial review of this
understanding. I have specifically taken up the linguistic question, and
argued against the supposed subalternity of dialects like the Prakrits and the
‘elite-ness’ of Sanskrit. I have pointed out that the two are represented as
literary languages of comparable stature in the traditional view as laid out in
manuals like the Natyashastra and the Kavyadarsha.
Further, in my essay, the performative aspect of the plays is reconstructed
to show that they seem to have been open presentations at public venues like
temples and street crossings, and so would have enjoyed, in all likelihood,
mixed rather than small and elite audiences. A related point is the large
audio-visual component, including song, dance, acting and pantomime, in
the performance and consumption of texts. This not just demonstrates,
again, the unity of cultural enterprises in early South Asia but, significantly,
proves that the reach of literature exceeded the limits of literacy. Orality as a
cultural technique thus assumes a powerful new role and implication. It
suggests how traditions were built and fortified and knowledge propagated.
A heterogenous viewership, bringing with it varied bents and
expectations, would have made very different sense of the content of the
plays. For example, when exposed to such an audience, the extensive
critique of the political, ritual and socio-sexual institutions of authority—the
preserve of the elite—that several of these plays resort to, takes on a radical
complexion. I have argued that this must be factored into any appraisal of
the function and social location of kavyas.

We have been speaking of orality and literary performance repeatedly,

inasmuch as they appear as hallmarks of early culture. But in none of the
essays so far do we get a concrete idea of who these oral composers and
performers in fact were, and how they were related to the communities of
and for which they sang. The next essay by K. Kailasapathy fills that gap
substantially, which is why we turn back the chronological order a bit and
look now at early Tamil poetry. Popularly, though problematically, known as
Sangam literature, it was composed and sung perhaps between the first and
third centuries CE, and redacted much later. It is originally and essentially
oral literature.29

Early Tamil literature is conventionally represented as consisting of two

different parts, akam and puram, or poems of love and poems of war,
arranged into several anthologies. According to Kailasapathy, however, it
was actually a unified whole. This he describes as essentially heroic poetry.
He points out that even the akam poems which sang of love, sang of the love
of heroes, who were warriors, chiefs and princes in an age of transition to
chiefdoms in early Tamilakam.30 Replete with notions of honour, fame and
valour, these poems, for Kailasapathy, are best described as canror ceyyul,
which means ‘the poetry of the noble or excellent ones’—heroes—and is a
term used for them in later commentaries.
The society depicted in the so-called Sangam poems appears to consist of
two divisions. One was described as the canror, the other as the ilicinar,
which in Tamil means ‘base’ or ‘servile’.31 People of the latter category figure
infrequently in the poems, and never as the main subject. In an interesting
contrast, however, the bards who composed, recited and performed these
songs about ‘the noble ones’, appear to have been of a mixed, and not
exalted, social character themselves.
Kailasapathy identifies broadly two kinds of bards: the singers and
dancers, and the poets.32 Among the former were the panars who sang their
songs to the accompaniment of a lute. They appear to have been wandering
minstrels, often in chronic poverty, moving in kin groups from place to
place in search of patrons. Then there were the porunar, war-bards travelling
with the warriors, singing and beating their drums. The kuttar were dancer-
actors, male and female, also itinerant, who are often depicted as performing
in the common grounds of villages to much popular enjoyment. The
pulavar, on the other hand, were the scholar-poets as different from mere
entertainers and encomiasts. The word means ‘wise men’. They were not
associated with music and dance, but were highly respected for their
learning and insight. In the poems, they are often found in comradeship
with princes, who they were also capable of reproaching on occasion. It is
these poets, it is believed, who were later responsible for the collection and
redaction of the heroic poetry under discussion.
Correspondingly, then, the bardic performance of these poems can be
visualised as serving audiences of varied composition. This was related also
to its different functions in the early Tamil communities. Thus, Kailasapathy
speaks of a ritual aspect to the performance by these artists. A certain
association with prophecy and divination adhered to these performers—a
vestige, perhaps, of the tribal ‘prehistory of poetry’.33 Predominantly,
however, the poems served as hortation to chiefs and warriors in battle. But
they were also sung and acted out as entertainment at popular festivals. The
complex weave of an oral culture is thus well exemplified in ‘Sangam’
literature.
There are two reasons for my choice of the next, and final essay of this book.
One is the author, A. K. Ramanujan, whose body of work as a whole has
yielded many an insight into the possible structure and workings of early
South Asian culture. Concepts like context-sensitivity and reflexivity which
were emphasised by him appear veritably as organising principles that
suggest some order out of the seeming chaos of sheer numbers and variety
of cultural forms and practices in the subcontinent. The second reason is the
theme of the essay—folklore—which is understood to be ubiquitous and to
lie beneath so much of the cultural production we have reviewed, but which
is defined and appreciated only loosely by historians.
For Ramanujan, folklore includes a range of genres from riddles,
proverbs, jokes, lullabies and songs to ballads, tales, and epics—‘the
literature of the mother tongues of the village, street, kitchen, tribal hut, and
way-side teashop’. It is cultural performance that surrounds us. Moreover,
folklore studies have clarified certain fundamental but problematic cultural
categories and equations; for example, the relation between folk and classical
literature; the former being oral, colloquial and improvisable, the latter
written, and in literary languages, with greater fixity. There are also
intermediate forms that are in transition from one to the other, alternating
between properties of the two. Several early South Asian works would
perhaps fall in this category; most would have been in this category at some
point in their transmissional lives, as Ramanujan tells us.
Further, crossovers and borrowings—continuities—between the folk and
the classical are as important a fact as the distinction between them, though
the directions in which these operate remain a moot point. Sanskrit and
Prakrit texts like the Panchatantra, Hitopdesha, Jatakas, and
Kathasaritsagara are excellent illustrations. Indeed, Ramanujan suggests not
a hierarchy between the two, such as terms like ‘high culture’ and Great and
Little traditions imply, but a ‘continuum’, where the folk and the classical are
but different expressive registers occurring in suitably different contexts.34
 Folklore is inseparably connected with an immediacy of performance
(coupled, paradoxically, with anonymity of authorship).35 Elsewhere,
Ramanujan identifies two main contexts for cultural performance:
private/domestic and public.36 The audience is of the essence in the
difference between the two contexts, and is thereby of the essence in the
‘recomposing of meaning’. In the domestic context, the audience would be
the household, comprising especially children and womenfolk. (Some of the
women would also be among the performers.) The dimensions of the public
settings could vary from the neighbourhood to the whole village or locality.
Occasions and motives for folklore performances would also be duly
domestic or public. As such, narrative would often entwine with ritual and
entertainment for the community,37 and in the process, bind the
community.
Folklore has a dual character: it is traditional, yet lives through variation.
This made it an ‘effective agent of cultural innovation.’38 Ramanujan also
speaks of folklore forming a ‘countersystem’—one that was responsive to
literate traditions, sometimes subverting, often balancing, themes and
attitudes of the latter.39 By housing ideas and attitudes not in currency in the
latter, it became a ‘full fledged collaborator in the production of cultural
meaning’, as Blackburn and Dundes put it.40 Finally, given its verbal
character, folklore was also highly mobile, its types and genres being trans-
regional, if not pan-Indian. As such, it represented culture in motion—
culture composed, transmitted, performed, consumed, lived by individuals
and groups, daily. For all these reasons study of folklore both deepens and
rounds off our understanding of the cultural history of early South Asia.
1

'A Figure of Speech, or a Figure of

Thought?'1
ANANDA K. COOMARASWAMY

W EfactAREthatPECULIAR PEOPLE. I SAY THIS WITH REFERENCE to the

whereas almost all other peoples have called their theory of
art or expression a ‘Rhetoric’ and have thought of art as a kind of knowledge,
we have invented an ‘Aesthetic’ and think of art as a kind of feeling.
The Greek original of the word ‘aesthetic’ means ‘perception by the senses,
especially by feeling’. Aesthetic experience is a faculty that we share with
animals and vegetables, and is irrational. The ‘aesthetic soul’ is that part of
our psychic make-up that ‘senses’ things and reacts to them: in other words,
the ‘sentimental’ part of us. To identify our approach to art with the pursuit
of these reactions is not to make art ‘fine’ but to apply it only to the life of
pleasure and to disconnect it from the active and contemplative lives.
Our word ‘aesthetic’, then, takes for granted what is now commonly
assumed, viz., that art is evoked by and has for its end to express and again
evoke emotions. In this connection Professor Whitehead has remarked that
‘It was a tremendous discovery, how to excite emotions for their own sake’.2
We have gone on to invent a science of our likes and dislikes, a ‘science of
the soul’ (psychology), and have substituted psychological explanations for
the traditional conception of art as an ‘intellectual virtue’ and of beauty as
‘pertaining to knowledge’.3 Our current resentment of ‘meaning’ in art is as
strong as the word ‘aesthetic’ implies; when we speak of a work of art as
‘significant’ we try to forget that this word can only be used with a following
‘of ’; that ‘expression’ can be significant only of some thesis that was to be
expressed, and overlook that whatever does not mean something is literally
in-significant. If, indeed, the whole end of art were ‘to express emotion’, then
the degree of our emotional reaction would be the measure of beauty and all
judgment would be subjective, for there can be no disputing about tastes. It
should be remembered that a reaction is an ‘affection’, and every affection a
passion, that is something passively ‘suffered’ or ‘undergone’, and not as in
operation or judgement an activity on our part. To equate the love of art
with a love of fine sensations is to make of works of art a kind of
aphrodisiac. The words ‘disinterested aesthetic contemplation’ are a
contradiction in terms and a pure non-sense.
‘Rhetoric’, of which the Greek original means ‘skill in public speaking’,
implies, on the other hand, a theory of art as the effective expression of
theses. There is a very wide difference between what is said ‘for effect’, and
what is said or made to be effective, and must work, or would not have been
worth saying or making [emphasis added]. It is true that there is a so-called
rhetoric of the production of ‘effects’, just as there is a so-called poetry that
consists only of emotive words, and a sort of painting that is merely
spectacular: but this kind of eloquence that makes use of figures for their
own sake, or merely to display the artist, or to betray the truth in courts of
law, is not properly a rhetoric, but a sophistic, or art of flattery. By ‘rhetoric’
we mean, with Plato and Aristotle, ‘the art of giving effectiveness to truth’.4
My thesis will be, then, that if we propose to use or understand any works of
art (with the possible exception of contemporary works, which may be
‘unintelligible’),5 we ought to abandon the term ‘aesthetic’ in its present
application and return to ‘rhetoric’, Quintilian’s bene dicendi scientia.
It may be objected by those for whom art is not a language but a spectacle
that rhetoric has primarily to do with verbal eloquence and not with the life
of works of art in general. I am not sure that even such objectors would
really agree to describe their own works as dumb or ineloquent. But
however this may be, we must affirm that the principles of art are not altered
by the variety of the material in which the artist works—materials such as
vibrant air in the case of music or poetry, human flesh on the stage, or stone,
metal or clay in architecture, sculpture and, pottery. Neither can one
material be called more beautiful than another; one cannot make a ‘better’
sword of gold than of steel. Indeed, the ‘material’ as such, being relatively
formless, is relatively ugly. Art implies a transformation of the material, the
impression of a new form on material that had been more or less formless;
and it is precisely in this sense that the creation of the world from a
completely formless matter is called a ‘work of adornment’.
There are good reasons for the fact that the theory of art has generally
been stated in terms of the spoken (or secondarily, written) word. It is, in the
first place, ‘by a “word” conceived in intellect’ that the artist, whether human
or divine, works.6 Again, those whose own art was, like mine, verbal,
naturally discussed the art of verbal expression, while those who worked in
other materials were not also necessarily expert in ‘logical’ formulation; and
finally, the art of speaking can be better understood by all than could the art
of, let us say, the potter, because all men make use of speech (whether
rhetorically, to communicate a meaning, or sophistically to exhibit
themselves), while relatively few are workers in clay.
All our sources are conscious of the fundamental identity of all the arts.
Plato, for example, remarks that ‘The expert, who is intent upon the best
when he speaks, will surely not speak at random, but with an end in view; he
is just like all those other artists, the painters, builders, shipwrights, etc.’;7
and again, ‘the productions of all arts are kinds of poetry, and their
craftsmen are all poets’,8 in the broad sense of the word. ‘Demiurge’ and
‘technician’ are the ordinary Greek words for ‘artist’ (artifex), and under
these headings Plato includes not only poets, painters and musicians, but
also archers, weavers, embroiderers, potters, carpenters, sculptors, farmers,
doctors, hunters and above all those whose art is government, only making a
distinction between creation, and mere labour, art and artless industry.9 All
these artists, in so far as they are really makers and not merely industrious,
in so far as they are musical and therefore wise and good, and in so far as
they are in possession of their art and governed by it, are infallible.10 The
primary meaning of the word wisdom, is that of ‘skill’, just as Sanskrit
kauśalam is ‘skill’ of any kind, whether in making, doing or knowing.
Now what are all these arts for? Always and only to supply a real or an
imagined need or deficiency on the part of the human patron, for whom as
the collective ‘consumer’ the artist works:11 when he is working for himself,
the artist as a human being is also a consumer. The necessities to be served
by art may appear to be material or spiritual, but as Plato insists, it is one and
the same art, or a combination of both arts, practical and philosophical, that
must serve both body and soul if it is to be admitted in the ideal city.12 We
shall see presently that to propose to serve the two ends separately is the
peculiar symptom of our modern ‘heartlessness’; our distinction of ‘fine’
from ‘applied’ art (ridiculous, because the fine art itself is ‘applied’ to giving
pleasure) is as though ‘not by bread alone’13 had meant ‘by cake’ for the élite
that goes to exhibitions and ‘bread alone’ for the majority and usually for all.
Plato’s music and gymnastics, which correspond to what we seem to intend
by ‘fine’ and ‘applied’ art (since one is for the soul and the other for the
body), are never divorced in his theory of education; to follow one alone
leads to effeminacy, to follow only the other, to brutality; the tender artist is
no more a man than the tough athlete; music must be realised in bodily
graces, physical power should be exercised only in measured, not in violent
motions.14
It would be superfluous to explain what are the material necessities to be
served by art: we need only remember that a censorship of what might or
ought not to be made at all should correspond to our knowledge of what is
good or bad for us. It is clear that a wise government, even a government of
the free by the free, cannot permit the manufacture and sale of products that
are necessarily injurious, however profitable such manufacture may be to
those whose interest it is to sell, but must insist upon those standards of
living to secure which was once the function of the guilds and of the
individual artist ‘inclined by justice, which rectifies the will, to do his work
faithfully’.15
As for the spiritual ends of the arts, what Plato says is that ‘we are
endowed by the Gods with vision and hearing, and harmony was given by
the Muses to him that can use them intellectually, not as an aid to irrational
pleasure, as is nowadays supposed, but to assist the soul’s interior revolution,
to restore it to order and concord with itself. And because of the want of
measure, and lack of graces in most of us, rhythm was given us by the same
Gods for the same ends’;16 and that while the passion evoked by a
composition of sounds ‘furnishes a pleasure-of-the-senses to the
unintelligent, it (the composition) bestows on the intelligent that hearts’ ease
that is induced by the imitation of the divine harmony produced in mortal
motions’.17 This last delight or gladness that is experienced when we partake
of the feast of reason, which is also a communion, is not a passion but an
ecstasy, a going out of ourselves and being in the spirit: a condition
insusceptible of analysis in terms of the pleasure or pain that can be felt by
sensitive bodies or souls.
The soulful or sentimental self enjoys itself in the aesthetic surfaces of
natural or artificial things, to which it is akin; the intellectual or spiritual self
enjoys their order and is nourished by what in them is akin to it. The spirit is
much rather a fastidious than a sensitive entity; it is not the physical
qualities of things, but what is called their scent or flavour, for example ‘the
picture not in the colours’, or ‘the unheard music’, not a sensible shape but an
intelligible form, that it tastes. Plato’s ‘heartsease’ is the same as that
‘intellectual beatitude’, which Indian rhetoric sees in the ‘tasting of the
flavour’ of a work of art, an immediate experience, and congeneric with the
tasting of God.18
This is, then, by no means an aesthetic or psychological experience, but
implies what Plato and Aristotle call a Katharsis,19 and a ‘defeat of the
sensations of pleasure (or pain)’.20 Katharsis is a sacrificial purgation and
purification, ‘consisting in a separation, as far as that is possible, of the soul
from the body’; in other words, a kind of dying, that kind of dying to which
the philosopher’s life is dedicated.21 The Platonic katharsis implies an
ecstasy, or ‘standing aside’ of the energetic, spiritual and imperturbable from
the passive, aesthetic and natural self; a ‘being out of oneself ’ that is a being
‘in one’s right mind’ and real Self, that insistence that Plato has in mind
when he ‘would be born again in beauty inwardly’, and calls this a sufficient
prayer.22
Plato rebukes his much beloved Homer for attributing to the Gods and
Heroes all too human passions, and for the skilful imitations of these
passions that are so well calculated to arouse our own sympathies.23 The
katharsis of the City is to be effected not by such exhibitions as this, but by
the banishment of artists who allow themselves to imitate all sorts of things,
however shameful. Our own novelists and biographers would have been the
first to go, while of modern poets it is not easy to think of any but William
Morris of whom Plato could have heartily approved.
The katharsis of the City parallels that of the individual; the emotions are
traditionally connected with the organs of evacuation, precisely because they
are waste products. It is difficult to be sure of the exact meaning of Aristotle’s
better known definition, in which tragedy ‘by its imitation of pity and fear
effects a katharsis from these and like passions’, though it is clear that for
him too the purification is from the passions; we must bear in mind that for
Aristotle tragedy is still essentially a representation of actions, and not of
character. It is certainly not a periodical ‘outlet’ of, that is to say indulgence
in, our ‘pent-up’ emotions that can bring about an emancipation from them;
such an outlet, like a drunkard’s bout, can be only a temporary satiation.24 In
what Plato calls with approval the ‘more austere’ kind of poetry we are
presumed to be enjoying a feast of reason rather than a breakfast of
sensations. His katharsis is an ecstasy or liberation of the ‘immortal soul’
from the affections of the ‘mortal’, a conception of emancipation that is
closely paralleled in the Indian texts in which liberation is realised by a
process of ‘shaking off one’s bodies’.25 The reader or spectator of the
imitation of a ‘myth’ is to be rapt away from his habitual and passible
personality and, just as in all other sacrificial rituals, becomes a God for the
duration of the rite, and only returns to himself when the rite is
relinquished, when the epiphany is at an end and the curtain falls. We must
remember that all artistic operations were originally rites, and that the
purpose of the rite is to sacrifice the old and to bring into being a new and
more perfect man.
We can well imagine, then, what Plato, stating a philosophy of art that is
not ‘his own’ but intrinsic to the Philosophia Perennis, would have thought
of our aesthetic interpretations and of our contention that the last end of art
is to please simply. For, as he says, ‘Ornament, painting and music made
only to give pleasure’ are just ‘toys’.26 The ‘lover of art’, in other words, is a
‘playboy’. It is admitted that a majority of men judge works of art by the
pleasure they afford; but rather than sink to such a level Socrates says, ‘No,
not even if all the oxen and horses and animals in the world, by their pursuit
of pleasure, proclaim that such is the criterion.’27 The kind of music that he
approves of is not a multifarious and changeable but a canonical music;28
not the sound of ‘poly-harmonic’ instruments, but the simple music of the
lyre accompanied by chanting ‘deliberately designed to produce in the soul
that symphony of which we have been speaking’;29 not the music of Marsyas
the Satyr, but that of Apollo.30
All the arts, without exception, are ‘imitative’: the work of art can only be
judged as such (and independently of its ‘value’) by the degree to which the
model has been correctly represented; the beauty of the work is
proportionate to its accuracy or truth. In other words, the artist’s judgment
of his own work by the criterion of art is a criticism based upon the
proportion of essential to actual form, paradigm to image. ‘Imitation’ is a
word that can be as easily misunderstood as St. Thomas Aquinas’ ‘Art is the
imitation of nature in her manner of operation’31 can be mistaken to mean
that that is the best art that is ‘truest to nature’ as we now use the word in its
most limited sense, with reference, not to ‘Mother Nature’, Natura naturans,
Creatrix, Universalis, Deus, but to whatever is ‘presented’ by our own
immediate and ‘natural’ environment, whether visually or otherwise
accessible to observation. In this connection, it is important not to overlook
that the delineation of ‘character’ in literature and painting is, just as much
as the representation of the looking-glass image of a physiognomy, an
empirical and realistic procedure, dependent on ‘observation’. St. Thomas’
‘Nature’, on the other hand, is that Nature ‘to find which’, as Meister Eckhart
says, ‘all her forms must be shattered’.
The ‘imitation’ or ‘re-presentation’ of a model (even a ‘presented’ model)
involves, indeed, a ‘likeness’ (Skr. sādṛśya), but hardly what we usually mean
by ‘verisimilitude’. What is traditionally meant by ‘likeness’ is not a copy but
an ‘image’ akin and ‘equal’ to its model, in other words a natural and
adequate symbol of its referent. The representation of a man, for example,
must really correspond to the idea of the man, but must not look so like him
as to deceive the eye; for the work of art, as regards its form, is a mind-made
thing and aims at the mind, but an illusion is no more intelligible than the
natural object it mimics. The plaster cast of a man will not be a work of art,
but the representation of a man on wheels where verisimilitude would have
required feet may be an entirely adequate ‘imitation’ well and truly made.32
 It is with perfect right that the mathematician speaks of a ‘beautiful
equation’ and feels for it what we feel about ‘art’.33 The beauty of the
admirable equation is the attractive aspect of its simplicity; it is a single form
that is the form of many different things. In the same way Beauty absolutely
is the equation that is the single form of all things; which are themselves
beautiful to the extent that they participate in the simplicity of their source.
‘The beauty of the straight line and the circle, and the plane and solid figures
formed from these . . . is not, like that of other things, relative, but always
absolutely beautiful’.34 Now we know that Plato, who says this, is always
praising what is ancient and deprecating innovations (of which the causes
are, in the strictest and worst sense of the word, aesthetic), and that he ranks
the formal and canonical arts of Egypt far above the humanistic Greek art
that he saw coming into fashion;35 the kind of art that he endorsed was,
then, precisely what we know as ‘Greek Geometric art’. We must not think
that it would have been primarily for its ‘decorative’ values that Plato must
have admired this kind of ‘primitive’ art, but for its truth or accuracy,
because of which it has the kind of beauty that is universal and invariable; its
equations being ‘akin’ to the First Principles of which the myths and
mysteries, related or enacted, are imitations in other kinds of material. The
forms of the simplest and severest kinds of art, the synoptic kind of art that
we call ‘primitive’, are the natural language of all traditional ‘philosophy’; and
it is for this very reason that Plato’s dialectic makes continual use of figures
of speech, which are really figures of thought.
Plato knew as well as the Scholastic philosophers that the artist as such
has no moral responsibilities, and can only sin as an artist if he fails to
consider the sole good of the work to be done, whatever it may be.36 But like
Cicero he also knows that ‘though he is an artist, he is nevertheless a man’,37
and if a free man, responsible as such for whatever it may be that he
undertakes to make; a man who, if he represents what ought not to be
represented and brings into being things unworthy of free men, should be
punished, or at the least restrained or exiled like any other criminal or
madman. It is precisely those poets or other artists who imitate anything
and everything, and are not ashamed to represent or even ‘idealise’ things
essentially base, that Plato, without respect for their abilities however great,
would banish from the society of rational men, ‘lest from the imitation of
shameful things men should imbibe their actuality’,38 that is to say, for the
same reasons that we, in moments of sanity see fit to condemn the
exhibition of gangster films in which the villain is made a hero, or agree to
forbid the manufacture of even the most skilfully adulterated foods.
If we dare not ask with Plato ‘Imitations of what sort of life?’, and ‘whether
of the appearance or the reality, the phantasm or the truth?’,39 it is because
we are no longer sure what kind of life it is that we ought for our own good
and happiness to imitate, and are for the most part convinced that no one
knows or can know the final truth about anything: we only know what we
‘approve’ of, i.e., what we like to do or think and desire a freedom to do and
think what we like more than a freedom from error. Our educational
systems are chaotic, because we are not agreed for what to educate, if not for
self-expression. But all tradition is agreed as to what kind of models are to
be imitated: ‘The city can never otherwise be happy unless it is designed by
those painters who follow a divine original’;40 ‘The crafts such as building
and carpentry . . . take their principles from that realm and from the
thinking there’;41 ‘Lo, make all things in accordance with the pattern that
was shown thee upon the Mount’;42 ‘It is in imitation (anukṛti) of the divine
forms that any human form (śilpa) is invented here’;43 ‘There is this divine
harp, to be sure; this human harp comes into being in its likeness’ (tad
anukṛti);44 ‘We must do what the Gods did first’.45 This is the ‘imitation of
Nature in her manner of operation’, and like the first creation the imitation
of an intelligible, not a perceptible model.
But such an imitation of the divine principles is only possible if we have
known them ‘as they are’, for if we have not ourselves seen them, our
mimetic iconography, based upon opinion, will be at fault; we cannot know
the reflection of anything unless we know itself.46 It is the basis of Plato’s
criticism of naturalistic poets and painters that they know nothing of the
reality but only the appearances of things, for which their vision is over-
keen; their imitations are not of the divine originals, but are only copies of
copies.47 And seeing that God alone is truly beautiful, and all other beauty is
by participation, it is only a work of art that has been wrought, in its kind
and its significance, after an eternal model that can be called beautiful.48
And since the eternal and intelligible models are supersensual and invisible,
it is evidently ‘not by observation’, but in contemplation that they must be
known.49 Two acts, then, one of contemplation and one of operation, are
necessary to the production of any work of art.50
And now as to the judgment of the work of art, in the first place by the
criterion of art, and in the second place with respect to its human value. As
we have already seen, it is not by our reactions, pleasurable or otherwise, but
by its perfect accuracy, beauty or perfection or truth, or in other words by
the equality or proportion of the image to its model that a work of art can be
judged as such. That is to consider only the good of the work to be done, the
business of the artist. But we have also to consider the good of the man for
whom the work is done, whether this ‘consumer’ be the artist himself or
some other patron.51 This man judges in another way, not, or not only, by
this truth or accuracy, but by the artefact’s utility or aptitude to serve the
purpose of its original intention, viz., the need that was the first and is also
the last cause of the work; accuracy and aptitude together making the
‘wholesomeness’ of the work that is its ultimate-rightness.52 The distinction
of beauty from utility is logical, not real.
So when taste has been rejected as a criterion in art, the Stranger sums up
thus, ‘The judge of anything that has been made must know its essence—
what its intention is and what the real thing of which it is an image—or else
will hardly be able to diagnose whether it hits or misses the mark of its
intention’; and again, ‘The expert critic of any image, whether in painting,
music or any other art, must know three things, what was the archetype, and
in each case whether it was correctly and whether well made . . . whether the
representation was good or not’.53 The complete judgement, made by the
whole man, is as to whether the thing under consideration has been both
truly and well made. It is only ‘by the mob that the beautiful and the just are
rent apart’;54 by the mob, shall we say, of ‘aesthetes’, the men who ‘know
what they like’?
Of the two judgements, respectively by art and by value, the first only
establishes the existence of the object as a true work of art and not a
falsification of its archetype: it is a judgement normally made by the artist
before he can allow the work to leave his shop, and so a judgment that is
really presupposed when we as patrons or consumers propose to evaluate
the work. It is only under certain conditions, and typically those of modern
manufacture and salesmanship, that it becomes necessary for the patron or
consumer to ask whether the object he has commissioned or proposes to
buy is really a true work of art; under normal conditions where making is a
vocation and the artist is disposed and free to consider nothing but the good
of the work to be done, it is superfluous to ask, ‘Is this a “true” work of art?’
When, however, the question must be asked, or if we wish to ask it in order
to understand completely the genesis of the work, then the grounds of our
judgement in this respect will be the same as for the original artist; we must
know of what the work is intended to remind us, and whether it is ‘equal’ to
(an ‘adequate symbol’ of) this ‘content’, or by want of truth betrays its
paradigm. In any case, when this judgment has been made, or is taken for
granted, we can proceed to ask whether or not the work has a value for us, to
ask whether it will serve our needs. If we are whole men, not such as live on
bread alone, the question will be asked with respect to spiritual and physical
needs to be satisfied together; we shall ask whether the model has been well
chosen, and whether it has been applied to the material in such a way as to
serve our immediate need; in other words, What does it say, and Will it
work? If we have asked for a bread that will support the whole man, and
receive however ‘fine’ a stone, we are not morally, though we may be legally,
bound to ‘pay the piper’. All our efforts to obey the Devil and ‘command this
stone that it be made bread’ are doomed to failure.
It is one of Plato’s virtues, and that of all traditional doctrine about art,
that ‘value’ is never taken to mean an exclusively spiritual or exclusively
physical value. It is neither advantageous, nor altogether possible, to separate
these values, making some things sacred and others profane: the highest
wisdom must be ‘mixed’ with practical knowledge, the contemplative
combined with the active life; the pleasures that pertain to these lives are
altogether legitimate, and it is only those pleasures that are irrational, bestial
and in the worst sense of the words seductive and distracting that are to be
excluded.55 Plato’s ‘music and gymnastics’, which correspond to our ‘culture
and physical training’, are not alternative curricula, but essential parts of one
and the same education;56 philosophy is the highest form of music (culture),
but the philosopher who has escaped from the ‘Cave’ must return to it to
participate in the every day life of the world and, quite literally, ‘play the
game’.57 Plato’s criterion of ‘wholesomeness’ implies that nothing ought to be
made, nothing can be really worth having, that is not at the same time
correct, or true, or formal or beautiful (whichever word you prefer) and
adapted to good use.
For, to state the Platonic doctrine in more familiar words, ‘It is written,
that man shall not live by bread alone, but by every word of God . . . that
bread which came down from heaven’;58 that is to say, not by mere utilities
but also by those ‘divine realities’ and ‘causal beauty’ with which the
wholesome works of art are informed, so that they also live and speak. It is
just to the extent that we try to live by bread alone and by all the other in-
significant utilities that ‘bread alone’ includes—good as utilities, but bad as
mere utilities—that our contemporary civilisation can be rightly called
inhuman and must be unfavourably compared with the ‘primitive’ cultures
in which, as the anthropologists assure us, ‘the needs of the body and soul
are satisfied together’.59 ‘Manufacture’ for the needs of the body alone is the
curse of modern civilisation.
Should we propose to raise our standard of living to the savage level, on
which there is no distinction of fine from applied or sacred from profane art,
it need not imply the sacrifice of any of the necessities or even conveniences
of life, but only of luxuries, only of such utilities as are not at the same time
useful and significant. If such a proposal to return to primitive levels of
culture should seem to be Utopian and impracticable, it is only because a
manufacture of significant utilities would have to be a manufacture for use,
the use of the whole man, and not for the salesman’s profit. The price to be
paid for putting back into the market place, where they belong, such things
as are now only to be seen in museums would be that of economic
revolution. It may be doubted whether our boasted love of art extends so far.
It has been sometimes asked whether the ‘artist’ can survive under
modern conditions. In the sense in which the word is used by those who ask
the question, one does not see how he can or why he should survive. For just
as the modern artist is neither a useful or significant but only an ornamental
member of society, so the modern workman is nothing but a useful member
and neither significant nor ornamental. It is certain that we shall have to go
on working, but not so certain that we could not live, and handsomely,
without the exhibitionists of our studios, galleries and playing fields. We
cannot do without art, because art is the knowledge of how things ought to
be made, art is the principle of manufacture, and while an artless play may
be innocent, an artless manufacture is merely brutish labour and a sin
against the wholesomeness of human nature; we can do without ‘fine’ artists,
whose art does not ‘apply’ to anything, and whose organised manufacture of
art in studios is the inverse of the labourer’s artless manufacture in factories;
and we ought to be able to do without the base mechanics ‘whose souls are
bowed and mutilated by their vulgar occupations even as their bodies are
marred by their mechanical arts’.60
Plato himself discusses, in connection with all the arts, whether of potter,
painter, poet or ‘craftsman of civic liberty’, the relation between the practice
of an art and the earning of a livelihood.61 He points out that the practice of
an art and the wage-earning capacity are two different things; that the artist
(in his sense and that of the Christian and Oriental social philosophies) does
not earn wages by his art. He works by his art, and is only accidentally a
trader if he sells what he makes. Being a vocation, his art is most intimately
his own and pertains to his own nature, and the pleasure that he takes in it
perfects the operation; there is nothing he would rather work (or ‘play’) at
than his making; to him the leisure state would be an abomination of
boredom. This situation, in which each man does what is naturally (Skr.
svabhāvatas) his to to do (Skr. svadharma, svakarma) is not only the type of
Justice,62 but furthermore, under these conditions (i.e., when the maker
loves to work) ‘more is done, and better done, and with more ease, than in
any other way’.63 Artists are not tradesmen; ‘They know how to make, but
not how to hoard’.64 Under these conditions the worker and maker is not a
hireling, but one whose salary enables him to go on doing and making; he is
just like any other member of a feudal society, in which none are ‘hired’
men, but all enfeoffed and all possest of a hereditary standing; that of a
professional whose reward is by gift or endowment and not at ‘so much an
hour’.
The separation of the creative from the profit motive not only leaves the
artist free to put the good of the work above his own good, but at the same
time abstracts from manufacture the stain of simony, or ‘traffic in things
sacred’; and this conclusion, which rings strangely in our ears, for whom
work and play are alike secular activities, is actually in complete agreement
with the traditional order in which the artist’s operation is not a meaningless
labour but quite literally a significant and sacred rite, and quite as much as
the product itself an adequate symbol of a spiritual reality; and therefore a
way, or rather the way, by which the artist, whether potter or painter, poet or
king, can best erect or edify himself at the same time that he ‘trues’ or
corrects his work.65 It is, indeed, only by the ‘true’ workman that ‘true’ work
can be done; like engenders like.
When Plato lays it down that the arts shall ‘care for the bodies and souls
of your citizens’, and that only things that are sane and free, and not any
shameful things unbecoming free men66 are to be represented, it is as much
as to say that the true artist in whatever material must be a free man; not
meaning by that an ‘emancipated artist’ in the vulgar sense of one having no
obligation or commitment of any kind, but a man emancipated from the
despotism of the salesman. Whoever is to ‘imitate the actions of Gods and
Heroes, the intellections and revolutions of the All’, the very selves and
divine paradigms or ideas of our useful inventions, must have known these
realities ‘themselves and as they really are’: for ‘what we have not and know
not we can neither give to another nor teach our neighbour’.67
In other words, an act of ‘imagination’ in which the idea to be represented
is first clothed in the imitable form or image of the thing to be made, must
precede the operation in which this form is impressed upon the actual
material. The first of these acts, in the terms of Scholastic philosophy, is ‘free’,
the second ‘servile’. It is only if the first be omitted that the word servile
acquires a dishonourable connotation; then we can only speak of labour, and
not of art. It need hardly be argued that our methods of manufacture are, in
this shameful sense, servile, or denied that the industrial system, for which
these methods are needed, is an abomination ‘unfit for free men’. A system
of manufacture governed by money values presupposes that there shall be
two different kinds of makers, privileged artists who may be ‘inspired’, and
under-privileged labourers, unimaginative by hypothesis, since they are
required only to make what other men have imagined, or more often only to
copy what other men have already made. It has often been claimed that the
production of ‘fine art’ is useless; it would seem to be a mockery to speak of
a society as ‘free’, where it is only the makers of useless things that are
supposedly free.
‘Inspiration’ is defined as ‘A supernatural influence which qualifies men to
receive and communicate divine truth’ (Webster). This is stated in the word
itself, which implies the presence of a guiding ‘spirit’ to be distinguished
from but nevertheless ‘within’ the agent who is inspired, but is certainly not
inspired if ‘expressing himself ’. Before continuing, we must clear the air by
showing, how the word ‘inspire’ has been scabrously abused by modern
authors. We have found it said that ‘A poet or other artist may let the rain
inspire him’.68 Such misuses of words debar the student from ever learning
what the ancient writers may have really meant. We say ‘misuse’ because
neither is the rain, or anything perceptible to sense, in us; nor is the rain a
kind of spirit. The rationalist has a right to disbelieve in inspiration and to
leave it out of his account, as he very easily can if he is considering art only
from the aesthetic (sensational) point of view; but he has no right to pretend
that one can be ‘inspired’ by a sense perception, by which in fact one can
only be ‘affected’ and to which one can only ‘react’. On the other hand,
Meister Eckhart’s phrase ‘inspired by his art’ is quite correct, since ‘art’ is a
kind of knowledge, not anything that can be seen, but akin to the soul and
prior to the body and the world.69 We can properly say that not only ‘Love’
but ‘Art’ and ‘Law’ are names of the Spirit.
Here we are not concerned with the rationalist’s point of view but only
with the sources from which we can learn how the artist’s operation is
explained in a tradition that we must understand if we are to understand its
products. Here it is always by the Spirit that a man is thought of as inspired.
‘The Genius breathed into my heart to weave’, Penelope says.70 Hesiod tells
us that the Muses ‘breathed into me a divine voice . . ., bade me sing the race
of the blessed Gods’.71 Christ, ‘through whom all things were made’, does not
bear witness of (express) himself, but says, ‘I do nothing of myself, but as my
Father taught me, I speak’;72 Dante is one who ‘when Love (Amor, Eros)
inspires me, attend, and go setting it forth in such wise as He dictates within
me’.73 For ‘there is no real speaking that does not lay hold upon the Truth’.74
And who is it (‘What self?’) that speaks ‘the Truth that cannot be refuted’?75
Not this man, So-and-so, Dante or Socrates or ‘I’, but the Synteresis, the
immanent Spirit, Socrates’ and Plato’s Daimon, he ‘who lives in everyone of
us’76 and ‘cares for nothing but the Truth’;77 it is ‘the God himself that
speaks’ when we are not thinking our own thoughts but are His exponents,
or priests.
And so as Plato, the father of European wisdom, asks, ‘Do we not know
that as regards the practice of the arts the man who has this God for his
teacher will be renowned and as it were a beacon light, but one whom Love
has not possest will be obscure?’78 This is with particular reference to the
divine originators of archery, medicine and oracles, music, metal-work,
weaving and piloting, each of whom was ‘Love’s disciple’. He means, of
course, the ‘cosmic Love’ that harmonises opposite forces, the Love that acts
for the sake of what it has and to beget itself, not the profane love that lacks
and desires. So the maker of anything, if he is to be called a creator, is at his
best the servant of an immanent Genius; he must not be called ‘a genius’;
but, ‘ingenious’; he is not working of or for himself, but by and for another
energy, that of the Immanent Eros, Sanctus Spiritus, the source of all ‘gifts’.
‘All that is true, by whomsoever it has been said, has its origin in the Spirit’.79
We can now, perhaps, consider with less danger of misunderstanding
Plato’s longest passage on inspiration. ‘It is a divine power that moves’80 even
the rhapsodist, or literary critic, in so far as he speaks well, though he is only
the exponent of an exponent. The original maker and exponent, if he is to be
an imitator of realities and not of mere appearances, ‘is God-indwelt and
possest . . . an airy, winged and sacred substance (Skr. brahma-); unable ever
to indite until he has been born again of God within him and is out of his
own wits and his own mind is no longer in him;81 for every man, so long as
he retains that property is powerless to make or to incant The men whom he
dements God uses as his ministers . . . It is the God himself . . . that speaks,
and through them enlightens us . . . . The makers are but His exponents
according to the way in which they are possest’.82 It is only when he returns
to himself from what is really a sacrificial operation that the maker exercises
his own powers of judgment; and then primarily to ‘try the spirits, whether
they be of God’, and secondarily to try his work, whether it agrees with the
vision or audition.
 The most immediately significant point that emerges from this profound
analysis of the nature of inspiration is that of the artist’s priestly or
ministerial function. The original intention of intelligible forms was not to
entertain us, but literally to remind us; the chant is not for the approval of the
ear,83 or the picture for that of the eye (although these senses can be taught
to approve the splendour of truth, and can be trusted when they have been
trained), but to effect such a transformation of our being as is the purpose of
all ritual acts [emphasis added]. It is, in fact, the ritual arts that are the most
‘artistic’, because the most ‘correct’, as they must be if they are to be effectual.
The heavens declare the glory of God: their interpretation in science or art
—and sine scientia ars nihil—is not in order to flatter or merely ‘interest’ us,
but ‘in order that we may follow up the intellections and revolutions of the
All, not those revolutions that are in our own heads and were distorted at
our birth, but correcting these by studying the harmonies and revolutions of
the All: so that by an assimilation of the knower to the to-be-known,84 the
archetypal Nature, and coming to be in that likeness,85 we may attain at last
to a part in that “life’s best” that has been appointed by the Gods to men for
this time being and hereafter’.86
This is what is spoken of in India as a ‘metrical self-integration’
(chandobhir ātmānaṁ saṁskaraṇa), or ‘edification of another man’ (anyam
ātmānam), to be achieved by an imitation (anukaraṇa) of the divine forms
(daivyāni śilpāni).87 The final reference to a good to be realised here and
hereafter brings us back again to the ‘wholesomeness’ of art, defined in
terms of its simultaneous application to practical necessities and spiritual
meanings; to that fulfilment of the needs of the body and soul together that
is characteristic of the arts of the uncivilised peoples and the ‘folk’, but is
foreign to our industrial life, in which the arts are either for use or for
pleasure, but never spiritually significant and very rarely intelligible.
Such an application of the arts as Plato prescribes for his City of God, arts
that as he says ‘will care for the bodies and the souls of your citizens’,88
survives for so long as forms and symbols are employed to express a
meaning, for so long as ‘ornament’ means ‘equipment’;89 and until what were
originally imitations of the reality, not of the appearance, of things become,
as they were already rapidly becoming in Plato’s time, merely ‘art forms,
more and more emptied of significance on their way down to us’,90—no
longer figures of thought, but only figures of speech.
We have so far made use of Oriental sources only incidentally, and chiefly
to remind ourselves that the true philosophy of art is always and everywhere
the same [emphasis added]. But since we are dealing with the distinction
between the arts of flattery and those of ministration we propose to refer
briefly to some of the Indian texts in which ‘the whole end or the expressive
faculty’ is discussed. This natural faculty is that of the ‘Voice’: not the audibly
spoken word, but that by which a concept is communicated. The relation of
this maternal Voice to the paternal Intellect is that of our feminine ‘nature’
to our masculine ‘essence’; their begotten child is the Logos of theology and
the spoken myth of anthropology. The work of art is expressly the artist’s
child, the child of both his natures human and divine: stillborn if he has not
at his command the art of delivery (rhetoric), a bastard if the Voice has been
seduced, but a valid concept if born in lawful marriage.
The Voice is at once the daughter, bride, messenger and instrument of the
Intellect;91 possest of him, the immanent deity, she brings forth his image
(reflection, imitation, similitude, pratirūpa, child).92 She is the power and
the glory,93 without whom the Sacrifice itself could not proceed.94 But if he,
the divine Intellect, Brahma or Prajāpati, ‘does not precede and direct her,
then it is only a gibberish in which she expresses herself ’. Translated into the
terms of the art of government, this means that if the Regnum acts on its
own initiative, unadvised by the Sacerdotium, it will not be Law, but only
regulations that it promulgates.
The conflict of Apollo with Marsyas the Satyr, to which Plato alludes,95 is
the same as that of Prajāpati (the ‘Progenitor’) with Death,96 and the same as
the contention of the Gandharvas, the Gods of Love and Science, with the
mundane deities, the sense powers, for the hand of the Voice, the Mother of
the Word, the wife of the Sacerdotium.97 This is, in fact, the debate of the
Sacerdotium and the Regnum with which we are most familiar in terms of
an opposition of ‘sacred’ and ‘profane’, ‘eternal’ and ‘secular’, an opposition
that must be present wherever the needs of the soul and the body are not
satisfied together.
 Now what was chanted and enacted by the Progenitor in his sacrificial
contest with Death was ‘calculated’ (saṁkhyānam)98 and ‘immortal’, and
what by Death ‘uncalculated’, and ‘mortal’; and that deadly music played by
Death is now our secular art of the ‘parlour’ (patnῑśālā), ‘whatever people
sing to the harp, or dance or do to please themselves’ (vṛthā), or even more
literally ‘do heretically’, for the words vṛthā and ‘heresy’ derive from a
common root that means to ‘choose for oneself ’, to ‘know what one likes and
to grasp at it’. Death’s informal and irregular music is disintegrating: on the
other hand, the Progenitor ‘puts himself together’ (composes or synthesises
himself) by means of the metres, the Sacrificer ‘perfects himself so as to be
metrically constituted’,99 and makes of the measures the wings of his
ascension.100 The distinctions made here between a quickening art and one
that adds to the sum of our mortality are those that underline Plato’s
katharsis and all true puritanism and fastidiousness. There is no
disparagement of the Voice (Sophia) herself, or of music or dancing or any
other art as such; whatever disparagement there is, is not of the instrument;
there can be no good use without art.
The contest of the Gandharvas, the high Gods of Love and Music (in
Plato’s broad sense of the word) is with the unregenerate powers of the soul
whose natural inclination is the pursuit of pleasures. What the Gandharvas
offer to the Voice is their sacred science, the thesis of their incantation; what
the mundane deities offer is ‘to please her’. The Gandharvas’ is a holy
conversation (brahmodaya), that of the mundane deities an appetising
colloquy (prakāmodaya). Only too often the Voice, the expressive power, is
seduced by the mundane deities to lend herself to the representation of
whatever may best please them and be most flattering to herself; and it is
when she thus prefers the pleasant falsehoods to the splendour of the
sometimes bitter truth that the High Gods have to fear lest she in turn
seduce their legitimate spokesman, the Sacrifice himself; to fear, that is to
say, a secularisation of the sacred symbols and the hieratic language, the
depletion of meaning that we are only too familiar with in the history of art,
as it descends from formality to figuration, just as language develops from
an original precision to what are ultimately hardly more than blurred
emotive values.
 It was not for this, as Plato said, that powers of vision and hearing are
ours. In language as nearly as may be identical with his, and in terms of the
universal philosophy wherever we find it, the Indian texts define ‘the Whole
End of the Voice’ (kṛtsnanaṁ vāgārtham). We have already called the voice
an ‘organ’, meaning to be taken in the musical as well as he organic sense. It
is very evidently not the reason of an organ to play of itself, but to be played
upon; just as it is not for the clay to determine the form of the vessel, but to
receive it.
‘Now there is this divine harp: the human harp is in its likeness . . . and
just as the harp struck by a skilled player fulfils the whole reason of the harp,
so the Voice moved by a skilled speaker fulfils its whole reason’.101 ‘Skill in
any performance is a yoking, as of steeds together’102 or in other words
implies a marriage of the master and the means. The product of the
marriage of the player, Intellect, with the instrument, the Voice, is ‘Truth’
(satyam) or ‘Science’ (vidyā);103 not that approximate, hypothetical and
statistical truth that we refer to as ‘science’, but ‘philosophy’ in Plato’s
sense,104 and that ‘meaning of the Vedas’ by which, if we understand it, ‘all
good’ (sakalam bhadram) is attainable, here and hereafter.105
The raison d’être of the Voice is to incarnate in a communicable form the
concept of Truth; the formal beauty of the precise expression is that of the
splendor veritatis. The player and the instrument are both essential here. We,
in our somatic individuality, are the instrument, of which the ‘strings’ or
‘senses’ are to be regulated, so as to be neither slack nor overstrained; we are
the organ, the inorganic God within us is the organist. We are the organism,
He its energy. It is not for us to play our own tunes, but to sing His songs,
who is both the ‘Person in the Sun’ (Apollo) and our own Person (as
distinguished from our ‘personality’). When ‘those who sing here to the
harp sing Him’,106 then all desires are attainable, here and hereafter.
There is, then, a distinction to be drawn between a significant
(padārthābhinaya) and liberating (vimuktidā) art, the art of those who in
their performances are celebrating God, the Golden Person, in both his
natures, immanent and transcendent, and the in-significant art that is
‘coloured by worldly passion’ (lokānurañjaka) and ‘dependent on the moods’
(bhāvāśraya): the former is the ‘highway’ (mārga) art that leads directly to
the end of the road, the latter a ‘pagan’ (deśῑ) and eccentric art that wanders
off in all directions, imitating anything and everything.107
If now the orthodox doctrines reported by Plato and the East are not
convincing, this is because our sentimental generation, in which the power
of the intellect has been so perverted by the power of observation that we
can no longer distinguish the reality from the phenomenon, the Person in
the Sun from his sightly body, or the uncreated from electric light, will not
be persuaded ‘though one rose from the dead’. Yet I hope to have shown, in a
way that may be ignored but cannot be refuted, that our use of the term
‘aesthetic’ forbids us also to speak of art as pertaining to ‘the higher things of
life’ or the immortal part of us; that the distinction of ‘fine’ from ‘applied’ art,
and corresponding ‘manufacture’ of art in studios and artless industry in
factories, take it for granted that neither the ‘artist’ nor the ‘artisan’ shall be
whole men; that our freedom to work or starve is not a responsible freedom
but only a legal fiction that conceals an actual servitude; that our hankering
after a leisure state, or state of pleasure, to be attained by a multiplication of
labour-saving devices, is born of the fact that most of us are doing forced
labour, working at jobs to which we could never have been ‘called’ by any
other master than the salesman; that the very few, the happy few of us whose
work is a vocation and whose status is relatively secure, like nothing better
than our work, and can hardly be dragged away from it; that our division of
labour, Plato’s ‘fractioning of human faculty’, makes the workman a part of
the machine, unable ever to make or to co-operate responsibly in the
making of any whole thing; that in the last analysis the so-called
‘emancipation of the artist’ is nothing but his final release from any
obligation whatever to the God within him, and his opportunity to imitate
himself or any other common clay at its worst; that all wilful self-expression
is auto-erotic, narcissistic and Satanic, and the more, its essentially
paranoiac quality develops, suicidal; that while our invention of
innumerable conveniences has made our unnatural manner of living in
great cities so endurable that we cannot imagine what it would be like to do
without them, yet the fact remains that not even the multi-millionaire is rich
enough to commission such works of art as are preserved in our museums
but were originally made for men of relatively moderate means or, under the
patronage of the Church for God and all men; the fact remains that the
multi-millionaire can no longer send to the ends of the earth for the
products of other courts or the humbler works of the folk, for all these
things have been destroyed and their makers reduced to be the providers of
raw materials for our factories, wherever our civilising influence has been
felt; and so, in short, that while the operation that we call a ‘progress’ has
been very successful, man the patient has succumbed.
Let us then, admit that the greater part of what is taught in the Fine Arts
Departments of our Universities, all of the psychologies of art, all the
obscurities of modern aesthetics, are only so much verbiage, only a kind of
defence that stands in the way of our understanding of the wholesome art, at
the same time iconographically true and practically useful, that was once to
be had in the market place or from any good artist; and that whereas the
rhetoric that cares for nothing but the truth is the rule and method of the
intellectual arts, our aesthetic is nothing but a false rhetoric, and a flattery of
human weakness by which we can only account for the arts that have no
other purpose than to please.
The whole intention of our own art may be aesthetic, and we may wish to
have it so. But however this may be, we also pretend to a ‘scientific’ and
‘objective’ discipline of the ‘history and appreciation of art’, in which we take
account not only of contemporary or very recent art but also of the whole of
art from the beginning until now. It is in this arena that I shall thrown down
a minimum challenge: I put it to you that it is not by our aesthetic, but only
by their rhetoric, that we can hope to understand and interpret the arts of
other peoples and other ages than our own; I put it to you that our present
university courses in this field embody a pathetic fallacy, and are anything
but scientific in any sense [emphasis added].
And now, finally, in case you should complain that I have been drawing
upon very antiquated sources—and what else could I do, seeing that we are
all ‘so young’ and ‘do not possess a single belief that is ancient and derived
from old tradition, nor yet one science that is hoary with age’108—let me
conclude with a very modern echo of this ancient wisdom, and say with
Thomas Mann that ‘I like to think—yes, I feel sure—that a future is coming
in which we shall condemn as black magic, as the brainless, irresponsible
product of instinct, all art which is not controlled by the intellect’.109
 2

Rock Paintings of the Mesolithic Period

 ERWIN NEUMAYER

I NthanTHISChalcolithic
ESSAY, ALL PAINTINGS WHICH ARE stratigraphically earlier
rock art will be treated as Mesolithic. The earliest
paintings which may have originated in a period earlier than the Mesolithic,
whose designation is still under dispute, will also be assigned to the
Mesolithic period for their cultural traits are similar to those of the
Mesolithic paintings as a whole.
The research into palaeolithic and Mesolithic cultures in India has so far
dealt only with finds of stone tools or other technological innovations found
in archaeological excavations, which provide very little clue to
understanding the lifestyles of people living in this period. The research in
stone-tool typology and the initial, but nonetheless rich, results of rock
painting research present us with two complementary sources of
information which together give us a better understanding of the people of
an age long past.
The rock paintings, engravings and bruisings were not intended by their
authors to furnish us with a complete picture of their activities. We have to
look at them as a manifestation of forms of expression, which were external
to their authors’ physical needs. The reason for this development is open to
speculation; the paintings may have been considered a means of influencing
themselves or their environment. Nor should we overlook the decorative
value of this form of expression. Many of the paintings are highly abstract, a
fact which can be regarded as a step towards symbolism and writing.
Extracting answers to our questions from these paintings is difficult, for the
modern observer of these symbols and pictures is himself largely confined to
the perceptions dictated by his own milieu. Most writers present the hunter-
gatherers of the Stone Age as being at the mercy of wild animals and all-
absorbed in the struggle for subsistence, as subject to the vagaries of nature,
and as scarcely different from wild animals themselves. This view may be a
projection of our own fears about a strange environment. The untouched
jungle is perhaps the most peaceful landscape that I know, even when it
harbours wild or dangerous animals which can be harmful to man under
certain circumstances. Although modern man has withdrawn into the cities
for his ostensible security, he is nevertheless at the mercy of dangers which,
by comparison, make the most vicious tiger or king cobra seem as harmless
as lambs!
The information that we gather from the tool typology and the rock
pictures deals with two different aspects of culture which do not overlap
very much. The rock pictures provide some detailed representations of
weapons, like spears and arrows, which give us an idea of how the microliths
we find in excavations were actually used. Querns and grinders are also
shown in the paintings. The rock pictures are rich in often intricate designs
which have yet to be found on any portable artifact from a stratified layer in
an excavation.1 Designs on bones and ostrich egg shells are confined to
cross-hatched lines and arrangements of dots. Because there is no
overlapping of design traits between excavation finds and rock paintings,
stylistic change cannot be dated on an absolute time scale. It can only be
seen in the relative context of the succession of styles.
The earliest picture complex is distinguished by the extreme naturalism of
its animal depictions. The representations of human forms, on the contrary,
are abstract and reduced to symbols. This earliest style masterfully captures
the sense of movement. The execution of the lines is extraordinarily fine
even to the smallest detail. These distinct details stand in contrast to
paintings of later styles which are often drawn in wavering lines of varying
thickness.
Not many groups of pictures from the earliest style remain. This style also
has the few polychrome pictures known to exist. In these paintings, a red
drawing usually has a green outline, sometimes joined by a yellow one. A
composition in Kharwai shows a group of dancers; some of them have
multicoloured head ornaments or multicoloured loincloths. In Jaura I
noticed the painting of a bull in which the body was shaded in with two
different pigments of red, and surrounded by a green contour line. The
painting gives a feeling of plasticity, which makes it quite unique.
The largest group of green pictures is located on the northern edge of
Lakhajoar in a shelter which contains pictures from the Mesolithic period
only. Under several layers of red paintings is a long group of green dancers,
many of whom are doing complicated steps. The ‘S’ position of the dancers’
bodies and the intricate positioning of the legs and arms immediately
recalled to me the mating dances of the cranes which I had often observed
from that shelter in the fields below.
A parallel which may approximate to the meaning of this dance depiction
is the practice, current among many existing tribal societies, of
pantomiming animal movements. Next to the isolated dancers there is a
group of seventeen dancers in a row led by one with a square head-mask.
Row dances, in which one often finds a complete village society
participating, are still practised in Indian tribal areas today. The last dancer
in this group seems to be having trouble maintaining his balance. Or what
appears to be mockery may have had a meaning which we do not
understand today. But it is a good example of the many small details which
are so typical of Indian rock art. Amazingly enough, the human figures
always wear loincloths which must have been made from a material soft
enough to swing around with the movement of the dancing. This leads me
to believe that they were actually made from some woven material. The
technique of weaving cloth was perhaps already known since it is only a
refinement of weaving.
The heads of the dancers are kidney-shaped. Often there is a small circle
in front of the mouth, which doubtless indicates that the person is singing.
The dancers are often depicted with extended headdresses. Nevertheless, it is
not evident which material was used to fashion them; they are strongly
reminiscent of the headdresses worn by some of the Naga tribes in eastern
India. In one picture at Bhimbetka, a green dancer wears a necklace with
small sticks radiating from it. In this style no sex difference is evident,
perhaps because all the humans depicted are men. These earliest paintings
give no technological information whatsoever. There is also no depiction of
food-gathering, nor are men and animals ever portrayed in a hunter-prey
relation. In one dancing scene in Lakhajoar, one of the participants holds a
musical instrument composed of vibrating prongs. In another group at
Kharwai one of the dancers is swinging a branch, but nowhere could I find a
clearly distinguishable tool. This is all in sharp contrast to later Mesolithic
paintings.
 It is important to understand the psychological motivation behind this
pictorial art. Hunting was a sacred act, hence the great number of ‘hunting
chieftain’ depictions in hunting scenes. Success at the hunt was desirable; but
for the successful hunter the emotional upheaval of inflicting wounds and
death was too strong. None wanted to be individually responsible for the act
of killing. For this reason, single hunters are only rarely depicted. Hunting
was a social act and killing was a communally shared trespass. Heroes who
encounter animal or human foes alone do not yet exist. They appear for the
first time in the Chalcolithic period. This may be the reason why dancers
and hunters are reduced to stickmen in the Mesolithic period, despite the
fact that the artists definitely had the technical capability to paint humans in
the same naturalistic style as they did the animals. Yet with the possibility of
pictorial art, the taboo against depicting scenes of a kill or of an assault on
other human beings would soon have to disappear.
Theoretically, the depiction of naturalistic animal and schematically
drawn man should have been reversed. This takes place later, to a certain
extent, when the hunters came to be depicted more bodily. When this
change occurred, the animals’ movements were shown according to the
circumstances of the hunt and the artist began treating the subject in the
context of the situation in the picture when choosing naturalistic or abstract
forms. The later Mesolithic styles reveal some of the most dramatic pictures
of animals panicking, rolling over in death, writhing in pain at an inflicted
wound and dying. In the same way as the depiction of animals was adapted
to show the hunting mêlée, the human forms had to be modified as well.
Of course, the Mesolithic artist never had any difficulty in indicating the
movement of even a stickman, but a stickman is always somehow weightless.
This does not mean that the stickman was given up totally, for artists would
occasionally still find this a convenient form. But in some of the paintings
where the hunter is running, jumping and throwing a spear, the figures are
well filled out, no longer stickmen.
Depictions of woman also become more frequent. It is worth noting that
in contrast to their spouses, women always look well-nourished and fat.
Women did not live under the threat of the killing taboo since their part in
the economic structure was to gather food, as depicted in Indian rock art;
for this reason there was nothing to prevent the artist from portraying them
in a more individualistic manner than their male counterparts. Another
difference between the earliest rock paintings and the later Mesolithic ones
is the absence of food production activities in the former. In the later
Mesolithic paintings, in contrast to the earliest ones, hunting and gathering
is the central theme. The stylistic details of Mesolithic rock paintings also
underwent some modification as time went by, as seen for example in the
modification of head forms and ornaments; but the theme remained static as
a whole.
Sometimes a rudimentary landscape will be included if necessary, as for
example in Jaora where a herd of elephants are trampling through high grass
or bamboo; a hunter in Bhimbetka ambushes some animals from behind
scrubs, and monkeys can be seen climbing trees. Water pools are not only
indicated by fish but also by an encircling shore line. Shelters and caves in
which people are engaged in some activity are shown as well, as are the caves
in which animals like rats, porcupines and scaly ant-eaters hide themselves.
An X-ray style which depicts the internal organs of animals with
anatomical accuracy is rare. More often the structure of the intestines and
the skeleton was simplified to a design-like pattern. In many instances it is
open to question whether the pattern was derived from an X-ray style.
Frequently these patterns seem to symbolise some mythological quality with
which the depicted animal is endowed. Even in the earliest styles there are
paintings replete with complicated geometric designs; these designs are
sometimes several metres square and are often the remains of oversized
animal portraits. These are mainly depictions of the horned boar, which had
a central place in the lore of Mesolithic Indian culture. The exact
stratigraphical position of these pictures is unclear; too few of them have
been found and most of those that have been found are in a very bad state of
preservation. But since many of them are polychrome, I think they probably
belong to the earliest stylistic group.
A hint as to the meaning of these designs comes from a picture in Jaora
which shows a rectangular plane divided into seven vertical design stripes;
on the upper side of the rectangle there are two wavy lines to denote water
out of which reeds are growing; fish are swimming beneath the lower
waterline. Ducks are paddling along the right and lower edge of the
rectangle. Towards the right of the design are five birds in flight. This picture
could be a depiction of the Mesolithic cosmos in which the rectangular
plane symbolises the earth, the wavy lines signify water with fish and reeds;
the air is indicated by flying birds. The seven design stripes could indicate
different features of the earth, which, when transferred to the body of an
animal or man, would show the same earth quality incarnated.
Of course, without verbal explanation from the artist, no speculation of
this kind can ever give us a definite answer; but nevertheless some
explanation should be attempted. A peculiarity of the Mesolithic paintings is
that the limbs of sitting humans or supine animals always cross each other—
which incidentally is a perspectival treatment we consider to be correct,
though it is contrary to the method employed elsewhere in Mesolithic
paintings. This perspectival treatment of crossing the limbs is used so
compulsively that one often gets the impression that the kneeling man or
resting animal have actually become entangled in their own limbs.
The concept of perspective in Mesolithic paintings as well as in paintings
of later periods followed a very complex pattern. Objects which are meant to
be furthest away from the beholder are placed towards the top of a
composition. Since this does not always solve the problem of understanding
the position of an object, a ‘mixed’ perspective sometimes had to be used, as
for example the flattening out of a group of objects which stand in some
relationship to each other. Here the main object is placed in the middle of
the composition and those related to it are arranged around it in such a way
that the relationship between them can be understood. That this was not
always easy for the artist can be seen from the many paintings where he
struggled very hard to paint things according to the then understood
concept of perspective.
Several hunting scenes employ this perspective. One group in Kathotia
shows a bovid in the centre, surrounded by hunters. In order to ensure that
the hunters behind the animal would not be concealed by its body, the artist
has drawn them above it; to show that they are dashing their spears against
it, he has drawn them with their heads towards the animal—in other words,
upside down.
 The perspective of animal horns or antlers as well as the animal’s ears is
always a twisted one. The horns must already have been imputed with a
special meaning in Mesolithic lore, for there are depictions of animals with
horns even when they actually have no horn at all, as for example the boar.
To emphasise this power attributed to the horns, they had to be shown in a
twisted perspective on the exclusively silhouetted animals. The crescent
shape of the horns gave them their full symbolic appearance.
The same treatment of perspective is given to the limbs of standing
animals, which as a rule are always shown with four legs, when actually two
should be at least partly obscured by the ones closer to the viewer. This
attribute makes the Indian rock paintings appear somehow flat when the
animal is not shown in motion. Yet when the animals are in motion, it adds
an extreme dynamism to the paintings. Exceptions to this twisted
perspective are found in the earliest paintings which, as already mentioned,
are often executed in green. Here in some cases, the horns and legs are
shown in the perspective we would consider ‘correct’. The eye, which is a
prominent organ in folk art, and incidentally in witchcraft, is totally absent
from Mesolithic and Chalcolithic art.2
Snakes are completely absent from the Mesolithic zoo. There might be
two explanations for this: either these creatures were insignificant to the
people of the time, which is not very likely, or snakes were so feared or
venerated (which amounts to the same thing) that drawing pictures of them
would have had some unpleasant consequence for the artist. However
significant a place snakes may have in Indian folklore and mythology, they
are totally absent in Mesolithic and Chalcolithic rock art. They first appear
in rock paintings contemporaneous with the Gupta period.
Domestication of animals is not what we would expect from a hunting
and gathering culture, and yet I know of three paintings which depict wild
animals which seem to be mascots or animals kept by the people for
sacrifice. Two examples come from Kathotia. One shows a pig being fed by
some people with food which seems to have been prepared by a woman
grinding some fruit on a quern. In another group of pictures a man and an
antelope appear to be playing together. In Lakhajoar there is a depiction of a
woman carrying a basket on her back and a monkey on her head (which
could also be prey, however).
The boar might have had a similar significance for the Mesolithic Indians
as the bear had among the Ainus in Siberia and Japan. The latter kept their
bear-brothers in cages until they were too big for captivity and then
‘murdered’ them for food. The depiction of the monkey and the antelope
allow us to imagine that they were mascots which were found when young
and kept as pets.
The style of Mesolithic rock paintings in the Vindhyas is remarkably uniform
[emphasis added]. Rock paintings which are located sometimes dozens or
even hundreds of kilometers apart show scarcely any difference in style,
theme or execution. This allows us to conclude that the nomadic hunters
and gatherers had a very uniform culture and code of expression.
One question which always turns up is whether women or men painted
the rock pictures. There are no really good criteria for distinguishing
women’s art from that of men. Rock art was an integral part of a culture
which had not yet consciously begun to mirror the cultural milieu. The
artists were almost certainly not an independent community as in modern
pluralistic society.
If we assume that the creators of the rock art were shamans, they could
only have communicated their mystic experiences through an established
and commonly understood code of expressions, which, it is true, was
sometimes different for women and men both in pictorial art and in
language. But this difference in expression is not conditioned so much by
sex as by the hierarchies of power current at the time, an example of which
is the range of dependency relationships within a society. (That hierarchies
are generally formed on the basis of sex is not denied; nevertheless,
hierarchical groups are not based exclusively on gender.) This would mean
that through the study of the artist’s product (which is a social product as
well, since the artist is not distingnished from the rest of the society), we
should theoretically be able to infer his position or role in society (as well as
the position of the society itself), which should answer our question as to
who did the paintings.
 The difficulty in accomplishing this is our own limitation, which stems
from the way we perceive history. Research into the history of art takes the
artistic possibilities of the present as its perspective, its yardstick by which it
analyses the art and culture of societies with other technologies, social
organisation and inclinations. From this arises the totally chauvinistic
division between primitive and classical art. Our patriarchally-oriented
social world makes it extra ordinarily difficult for us to understand the
societies that created the rock paintings.
Although most of the house-wall paintings in India are done by women,
the exceptions to this are not insignificant. Yet the theme of modern Indian
folk art is probably fundamentally different from that of the Mesolithic
period. The Mesolithic portraits of women, as well as those of men, do not
idealise (at least as our culture defines this) man or woman as a sex object. If
it were otherwise we still would not know who made the pictures, but we
would get some indications as to whom they were done for.
There is a striking segregation between hunting and gathering activities in
the Mesolithic pictures. In depictions of the hunt the man has all the skill
necessary for running, jumping, hurling weapons and retreating from a
charging animal. Even so, he is most often depicted as a stickman. On the
other side are the women gathering food, nursing babies, being pregnant,
carrying baskets and working on the quern. This observation conforms with
our concept of a sex-defined division of labour. But there is no indication of
the distribution of power in these pictures. Since the women produced the
greater part of the food requirements through their gathering activity, it
would be logical to conclude that the greater power within the group rested
with them—if our concept of power is even applicable in this situation.
Sensuality was never portrayed as something completely ecstatic in the
Mesolithic pictures. There are very few pictures of women exposing their
genitals or of ithyphallic men. Depictions of sexual activities are so natural
that they do not always fulfil our expectations of sex scenes; the possibility
often remains that we are only viewing dance scenes in which the men and
women are closer to each other than normally. Amazingly, these depictions
of sexual intercourse preserve the stylistically different ways of portraying
women and men. A stickman is playing with the breasts of a woman, whose
portrayal called for three times as much pigment as did her somehow
meagre lover.
I have to admit that all these thoughts about society and the distribution
of power did not finally bring me to a clear understanding of who did the
paintings. But assuming, on the basis of the information furnished by these
paintings, that Mesolithic society was a very egalitarian one, the question of
who made the paintings still cannot be solved because painting does not
appear to have been an activity invested with any particular prestige or
practised by an elite determined on the basis of sex [emphasis added].
The Mesolithic paintings clearly illustrate a society of hunters and
gatherers. The Indian rock paintings provide a more varied source material
on lifestyles than do the excavations carried out thus far, which generally
furnish information relating mainly to tool technology. But even with all
these exciting new discoveries, the deciphering of the data has not yet made
much progress. The pictures are certainly not illustrations of the hunter-
gatherers’ day-to-day activities. They are a part of some socio-religious
ritual; certain themes are emphasised more than others, which, though
obviously of greater economic importance, only appear as peripheral by-
products. We must try tuning our mode of perception in an attempt to filter
these facts.
These pictures may possibly be part of a ritual to magically influence
circumstances which were difficult to control rationally. Ethnological studies
of traditional societies show a direct correspondence between the
complexity of the activity performed and the magical attention given to it.
Difficult tasks require more magic, while everyday ones hardly need any.
Success on the hunt is one of the results most often invoked through magic.
Although the hunt was not the most important source of food, most of the
pictures are dedicated to this theme.
Yet we tend to overemphasise the magical content of the pictures,
implying that superstition was the dominant force in the society, and
therefore labelling the society ‘primitive’, which it definitely was not. The
pictures originated during a period of change of which even the new
technique of painting is an indication. The mental capacity to make
representations must already have been present for millennia before this art
actually appeared. The earliest specimens of rock art cannot be classed as
products of the human spirit emerging from its childhood. They are not
early steps on the way to a more highly ‘evolved’ art; rather, their symbolism
and technical proficiency indicate that these oldest examples were high points
in the field of artistic expression [emphasis added].
The rock paintings are expressions of the code which governed the life of
a microlith-using people. Magical ideas could already be expressed through
a complex symbolism which helped cope with psychological fears and
insecurities, and so helped to channelise them. This is a circumstance which
comes under the scope of the term ‘culture’. Only under these social
conditions could the development of organised food production be
accomplished. Therefore, the rock paintings are important documents of a
transitional period; they not merely furnish us with statistics about tool
types and middens, but also, and more importantly, provide us with
illustrations of the cultural circumstances of the Mesolithic people.
CULT REPRESENTATIONS AND SOCIETY

The Mesolithic rock pictures form a unity which has had to be separated
into single themes for the purpose of analysis in order to attempt an
explanation of what motivating causes lay behind their origin. The
distinction between the individual theme complexes are subject to the
intuition and experience of the modern viewers. These reflections and
speculations could either be confirmed or denied by the perceptions of
another viewer, but none can claim that his interpretations are final.
Animals pursuing man are often portrayed in the rock paintings. The
hunter is the victim of his chosen prey. Often these depictions seem to recall
a hunting accident. In Lakhajoar a buffalo is throwing a man over his back.
At Bhimbetka, a boar is chasing an unarmed man, while in another shelter
on the same hill, a rhino is hurling one of his attackers into the air. Nowhere
in the Mesolithic paintings does this turning of the tables appear to be a
desired occurrence. The men in these pictures are depicted as being inferior
to an animal, which symbolises the greater power of the latter. In the
Mesolithic paintings this fact of people being thrown around by animals
cannot be explained as the sport of ‘bull jumping’ which was practised in the
Chalcolithic period in western Asia and was, above all, a strong cultic
peculiarity in Crete. The depiction of the men being chased by animals has
to be seen in the context of the hunting complex, even if the animal is
sometimes depicted as an oversized deified being with a body composed of
many parts of different animals.
Nevertheless, pictures showing these hunting accidents were possibly
already executed as part of some cultic activity. Pictures which depict well-
organised hunting expeditions do not seem to have any mythological
content. But, as already mentioned, they could very well have had an entirely
magical purpose. It is even possible that the execution of the painting was in
itself a magical exercise, and that—not very likely, but still possible—the
finished product was of no importance after the act of execution.
But a great proportion of the rock paintings of the Mesolithic period
cannot be explained as depictions of everyday activity. The most eye-
catching scenes in this group are the occasional huge drawings of animals,
which often carry the typical attributes of some other animals as well. One
of the best preserved picture groups of this type is at Bhimbetka under a
mushroom-shaped rock some 12 m high, whose very form and setting
already convey the aura of a sacred place.
The most distinctive figure among the many paintings which this rock
shelter houses is a huge animal with the body of a boar; however, its snout is
very similar to a rhino’s—the underlip seems to derive from an elephant,
and, most surprising, it has bovine horns. This creature, for which it is
difficult even to assign a name, is chasing away a diminutive man. The
contiguous form of the boar is not placed on a horizontal plane; rather, he is
storming down at an angle from above at a man who is seeking safety in
flight. This special organisation lends immense power to the composition.
The boar is spraying his semen in long streams towards the earth, over a
number of small piglets, which makes it look as if he is generating all these
little creatures drawn beneath him. (This of course is only a personal
reflection, for it cannot be ascertained whether the small animals belong to
the same group or have just incidentally been overpainted by the big boar.)
Yet the portrait of the deified creature shows a great many cultural
accretions which were all added during the Mesolithic period.
In front of the figure of the fleeing man is the drawing of a comparatively
large crab. Between the man and the snout of the boar are radiating lines
which may indicate the snorting of the boar, or arrows which the man has
dared to shoot at his oversized foe.
There is a stylistically contemporary depiction in Jaora. In this case,
however, the animal depicted is clearly a rhino. (This is incidentally one of
the best animal portrayals in Indian Mesolithic art.) Once again we see a
man taking to his heels, closely followed by the formidable snout of the
rhino. Over his shoulder he carries a long spear. But he will not be able to
get away because a few steps ahead he will encounter a big boar, no less
formidable than the rhino. In Lakhajoar again, there is a picture of a boar
being attacked by two comparatively small hunters with microlith-tipped
spears.
 There are many more deified animals depicted at Bhimbetka but few of
them are involved in any action. Generally their bodies resemble those of
boars, but they always have horns which are sometimes even longer than
their bodies. Another group at Bhimbetka shows a boar which is placed
vertically, head down, on the shelter wall; to the left of its snout is the
drawing of a crab, and to the right a woman with a carrying net extended
from her head. This composition seems to illustrate the polarisation of ideas
in the Mesolithic mythological system. There is the huge and vengeful beast
which chases a man, ejaculating its semen on the earth; on the other hand,
the woman with the basket goes about her business unchallenged. The boar
could be the ‘Lord of the Vegetation’—to use Frazer’s term. The man
symbolises the hunter challenged in his pursuits by the deified boar, or the
equally powerful rhino, or in some cases by a buffalo bull.
The woman with the basket, which could be a symbol of plenty, is left in
peace to harvest the fruit and vegetation. The deified animals seem to have
been cult idols whose depictions were maintained and restored over a long
span of time. In Bhimbetka one of these depictions revealed successive
layers of paints and repaint due to a trickle of water having washed off the
uppermost layer. This clearly shows that the paintings were actually restored.
Of course in this process some of the features of the original painting
underwent considerable change. The deified animal or the concept it stood
for was recognised not only in the Mesolithic period; it seems that some of
the paintings were repainted well into the historic period, or overlaid by
animal paintings which were modelled on the already available outlines. Of
course we cannot expect a cult to exist unmodified for millennia, and these
paintings had to be adapted to a different symbolism.
The 4 m-long depiction of a bull in Jaora—one of the most impressive of
the Indian rock paintings—seems to be a product of continuous repainting.
The bull as we see it now is in the style of paintings which were done after
the beginning of the Christian era. The unique form and position of the
head in this painting is puzzling. The artist, or rather the restorer, seems to
have shaped the head according to outlines already available from earlier
paintings, which may have depicted a rhino rather than a bull.
 The body of this big bull is whitewashed and decorated with thousands of
red dots, which are in fact fingerprints. What is surprising is that lines of
Mesolithic animal bodies, which could easily have been whitewashed as
well, were repainted and restored. Painting compositions were adapted,
changed and enlarged even during the same style period to which the
originals belonged, in other words at a time when the symbolism of the
original paintings was still understood. This fact is clear from the group
from Bhimbetka described above. Here the lower smaller deified boar is
only an adaptation of an elephant, with whom he shares the lower lip and
one leg. It could easily be imagined that new adaptations of earlier sacred
paintings could alter the iconography of a deified animal considerably. A
freak depiction could develop into the prototype of new iconography for
sacred images.
Some of the earliest layers of paint reveal traces of larger than life-size
animals embellished with extraordinary designs which are sometimes
executed in three colours—red, green, and white. Unfortunately none of
these big paintings of the Mesolithic period are complete. And, more often
than not, the scarce remains are almost invisible. Moreover it requires
considerable knowledge of the stylistic peculiarities of the Mesolithic
paintings to recognise the almost invisible lines and reconstruct the original
shape and appearance of the animal. The execution of animal depictions of
this size, in sometimes very inconvenient places like high rock walls or
ceilings, could only have been done by using elaborate ladders or
scaffoldings. But even this does not explain the well-executed lines, seeing
that the painter had to work too close to the rock to be able to get an overall
view of the composition. The animals are well-proportioned and radiate the
same dynamic movement as do the generally much smaller works of art of
the Mesolithic period.
A modern artist would solve the problem of depicting the proportions of
the animal correctly with the help of a grid and model. It is not unlikely that
a similar method was used by the Mesolithic artists, though no trace of this
survives. It is only in the Chalcolithic period that artists constructed animal
figures from geometric forms (as unfinished animal drawings at every stage
of completion tell us) using a technique similar to that of the cubists of this
century, though for a very different reason.
 Interpreting the big boar as a deified animal—and equating him, quite
speculatively, with the Lord of the Vegetation—seems quite plausible
considering his similarities with the iconography of West Asian cultures.3
But the problem gets much more complicated with depictions of beings
which are only comparable to Rakshasas. Here we identify them with
Yakshasas only on account of the fact that these Indian demons have the
ability to change into whatever form and size they fancy.
At Jaora we find a bird-like being with human legs and a huge beak with
wicked-looking teeth. A tiny human seems to be running away from its
beak. At Kathotia we find a being composed almost entirely of jaws and
teeth. Two paintings at Kathotia show fish standing on their tailfins, which
exceed some humans standing close to them in size. In one case the fish is
surrounded by figures of the ‘hunting chieftain’ type, who are clearly women
in this case. One man is shooting a microlith-barbed arrow into the fish. In
the second painting one ‘hunting chieftain’ figure with a basket on his back
is touching the fish which is three times bigger than him.
Depictions of fish—apart from those in fishing scenes—are quite rare. In
Lakhajoar there are drawings of two snake-like fish whose forms have been
adapted to the available space, which only allowed the artist to draw the fish
snake-like. From Sauro Mauro comes the depiction of three big fish. As the
body patterns look quite Chalcolithic, it is possible that the big fish
continued to be a symbol right into the Chalcolithic period.
In Kathotia there are pictures of some cult activity which seem to be
confined to this particular site. These pictures show uniformly adorned men,
standing on a double line which moves, most often, along a zig-zag path.
Many lines radiate downwards from these double lines, which I believe are
symbolic of the flow of water or a cascade of water. The men are bent over
and knocking something against the ground with both hands. In one case it
is easy to see that they are knocking tortoises on the ground and throwing
them down in the same direction in which the water is flowing.
In one case a lizard has also been drawn between these wavy lines. In
Satkunda there is a unique group showing the same subject in a very
symbolic way. One tortoise is placed between a human figure and another
figure which looks like a human but has a long tail like a lizard. Around this
scene are dancers with interlocked arms.
In a shelter on the northern end of Lakhajoar, I noticed another unique
picture group. To investigate this group thoroughly, Dr Wakankar and I had
to construct a scaffolding, which dangled quite precariously above the
ground. But the exercise was worth the labour. While standing on this ‘air
raft’ it was not surprising to find the painting of an elaborate construction.
The picture shows a platform resting on four long poles; the outer two have
branches cut in such a way as to facilitate human climbers, like a ladder. Two
men are climbing this structure, which stands about four to five times their
own height. They carry big loads on their heads. Two men sitting on the
platform seem to be lending a hand to the overloaded climbers. To the left of
this platform is another, bigger one on which a few men are actually dancing
and beating some instrument. (This shows that the Mesolithic people had
some confidence and skill in their carpentry work.)
To the left of this Mesolithic dance platform are faded drawings of fishes
and lizards, and between this jumble of aquatic creatures is a creature, rather
human in appearance, engaged in a dance. Since the rock surface here is
crystalline and the paintings very faded, the overlapping of painting layers
cannot be discerned. Hence we cannot tell whether the left part, that is the
fishes and lizards, belongs to the composition depicting the scaffolding, even
if we assume that it belongs to the same stylistic epoch.
One surprising feature of Mesolithic art is, as already mentioned, the total
absence of any depictions of snakes. There are many writers who have tried
to explain several of the weird depictions of Mesolithic art as proto-Hindu
or even ‘Aryan’ deities. The ‘hunting chieftain’ becomes the Kurma
incarnation of Vishnu, the fish Matsya, the boar Varaha. Occasionally the
boar is said to be a bull on account of its horns and called Indra, depending
upon the fancy of the writer. The absence of the snake would be difficult to
explain if we were to assume that all these deified beings are merely early
depictions of Hindu deities. Snakes are an important iconographic
peculiarity right from the earliest examples of classical Indian art as well as
in Indian folklore.
 The reluctance to depict snakes—and scorpions as well—in the Mesolithic
period could of course be explained as being a reflection of the great
reverence in which these creatures were held by Mesolithic people. In any
case snakes and scorpions are totally absent in Mesolithic paintings. Dance,
music, and chants are the most widely recurring cultural symbols.
Mesolithic paintings show that rhythmic movements most probably
reinforced by a uniform beat, were used to forge a number of people into a
unified mass sharing a homogeneous psychological condition. This practice
is still common today. It has a simple magic which nobody can escape. I
often experienced this same spell while participating in tribal dances with a
monotonous rhythmic drum and repetitive step pattern in surroundings
which were otherwise totally strange to me (as was the deer). In the earliest
rock painting style—the one in which green colour was used—dance scenes
are almost the only theme. These pictures indicate a very complex dancing
culture with uniform step and movement patterns.
A picture done in a green pigment at Bhimbetka shows three dancing
men, two of them positioned symmetrically to each other. A depiction in
Jaora also shows two dancers in a symmetrical relationship—in this case
their posteriors form the symmetrical axis.
The Mesolithic paintings show a number of incidents in which dancers
lose their balance, which I cannot really believe is caused by the rugged
surface of the dancing ground. An example comes from one of the best
preserved green paintings found in Lakhajoar. This group shows some
people engaged in acrobatic dance steps followed by a group of seventeen
dancers who have formed a chain by placing their hands on the back of the
man in front. The last man has lost his balance. Unfortunately the reason for
this is not clear. But the jungle definitely yielded some liquids, plants and
bark during Mesolithic times, the consumption of which was intoxicating
and which would therefore have facilitated the losing of balance. Small
circles are painted in front of the mouths of many of the dancers, which I
believe to be ‘speech bubbles’ showing that the person was singing or
chanting. The long chain of dancers follows a man decorated with a square
headdress and some radiating feathers or ornaments. A small green picture
in Lakhajoar shows a group of five dancers with one holding a comb-like
rhythm instrument.
 Musical instruments are rarely depicted in any of the rock painting
periods. Besides this comb-like instrument, I only know of a harp-shaped
one depicted in a Mesolithic painting. Gongs and drums which are beaten
by sticks are the only examples of instruments I know of from the
Chalcolithic period.
In paintings of the historical period, drums are the most frequently
depicted instruments, and are found mainly in scenes of war. But a few
pictures also show the harp and wind instruments. Probably the most
beautiful picture of a musician comes from Pachmarhi. It shows a man
playing on a harp, surrounded by his family. His bow, arrows and axe lie
beside him. This painting belongs to the historical period as well, but is
executed in the graceful style characteristic of the Mahadeo hill rock
paintings.
In Kharwai we find another green painting group showing a dancing
scene; this is the most carefully executed painting of the earliest style group.
It is placed in a very low shelter and is overprinted by many later Mesolithic
compositions. The painter of this group used not only the rare green colour
but also yellow and red. Like green, yellow is a very rare pigment and is
seldom used. This yellow colour is derived from a thin patina formed on a
deposit of brown or orange chalcedon, which has been lodged for a long
time in a humid lateritic layer.
The square shape of the dancer’s head in this group differs from the usual
kidney-shaped form. This square shape reminds one of the lead dancer of
the long dancing group in Lakhajoar or other green paintings in Muni Ki
Pahar. One of the dancers in the Kharwai group has an intricate headdress
painted in three colours (green, red, and yellow).
Worthy of notice in all these early human depictions is the prominently
executed loincloth, which definitely cannot be described as practical
clothing. This view is strengthened by the figure of the green dancer in Muni
Ki Pahar who holds his loincloth between his hands while dancing; as if to
accentuate the whole dance movement by the swaying of the wavy material.
Incidentally, in the iconography of paintings of the early Buddhist period in
Afghanistan as well as in the art of Mongolia and Tibet, the swaying
garments lend a graceful touch to the otherwise static forms of the
meditating Buddha.
Many of the early dance depictions gave us only a glimpse of what may
have been the use of masks and body ornaments. Since the human body is
reduced to the simplest and barest number of lines the figures often look
over-decorated and it is not always easy to distinguish the shape of the head
from a mask. Some depictions dwell lavishly on masks, adornment, and
perhaps also body paint, neglecting the human form altogether.
A painting in Kathotia shows more than a hundred dancers in a row.
Leading this chain of people are three persons decorated with headdresses
from which long radiating red and white lines form a round ray-like halo.
The bodies of these figures are again almost hidden by lavish costumes.
These paintings remind one of some of the dance costumes of the Nagas
in eastern India with their cotton clothing dyed in red, black, and white and
their headdresses radiating bamboo-splinters. Some of the dancers in this
group in Kathotia have linked hands, while others place theirs on the back of
the man in front. It is surprising that the artist should have given some
consideration to perspective in this portrayal of dancers, for only one leg can
be seen when people are viewed from the side, and both when they face
outwards from the picture and are being seen from the front.
Few of the dance scenes give any clue as to the purpose of the dance (if it
is at all necessary to search for one). The experience of the dance itself was
probably a sufficient incentive to stage one, as it is in Indian tribal societies
today. The tribals hardly have anything to celebrate—by our standards—but
yet they dance almost every evening. One of the early painting groups from
Kathotia (which is, incidentally, one of the few groups in green not
belonging to the very first Mesolithic style) shows dancers in close proximity
to another painting of a hunting scene which belongs to the same period;
probably these two groups were actually part of the same painting, but got
separated by weathering and rock chipping. What is important in this scene
is that the dance centres around a dead boar. One of the dancers holds fork-
like instruments in his hands and another man, who is sitting, plays on a
harp. In the hunting scene nearby we see men thrusting their spears into a
bovid. Of interest in this group are two dancing men. One of them is
bending his head back below shoulder level. The other one has a ‘U’-shaped
form above the shoulders.
A similar depiction comes from Bhimbetka. I believe these figures are
arranged according to another type of perspectival treatment to show
progress of movement in much the same way as a draftsman portrays an
object from two sides in a blueprint. The artist wanted to show a man
violently throwing his head backwards and forwards. In the profile view, he
has shown the head thrown back; in the view from the front, the shoulder-
head-shoulder line becomes ‘U’-shaped. In Bhimbetka the ‘U’ shape of the
head is very clear. In Kathotia the artist put the ‘U’ shape above the shoulder,
but it still meant the same thing.
As already mentioned, row-dancers are plentiful in all the rock painting
epochs. In the Chalcolithic period, these row-dancers are often related to
cultic activities which have a strongly homoerotic character. These dance
groups are often related to chariot-processions. One can see acrobats
walking on their hands, and fist fights in progress. Many of the dancers
indulge in pederastic activities. Many men have erect phalluses of
sometimes astonishing size; this could also mean that they wore penis
sheaths.
A very well-preserved group of this kind of painting is found in Kathotia.
Here the men are walking in a row, with both hands placed on the man in
front. The leader of the group stands bent in front of a man twice his size
holding a stick in his hand. All the men have erect phalluses. The genital of
the man with the stick is very large, the testicles are shown in a twisted
perspective. To the right of this group is a row of dancers with interlocked
arms. The procession ends before a man squatting on the floor and
masturbating. Above him is the depiction of a cow with some other small
figures beneath. Some men sitting on the shoulders of other men can also be
seen in the left part of the painting.
Practices analogous to the ones shown in this picture can be found in the
ethnological literature of still existing tribal peoples. The popular form of the
spring festival Holi has in it a great number of sexual elements which today
are only manifested in verbal abuse or obscene songs. The depiction of
pederastic activities finds analogies in descriptions of the Dionysian
mysteries.
The depictions of dancing and sexuality from the Mesolithic period are
contrary to the ecstatic depictions of homosexuality in the Chalcolithic
period. The Mesolithic paintings are totally confined to heterosexuality. Sex
in Mesolithic paintings is an ‘everyday activity’ not distinct from others.
These scenes never have a rapturous air about them. Sexual intercourse is
shown in much the same way as is a dance, the only difference being that in
the former the couples get closer to each other than in a ‘normal’ dance. The
activity of the other people around seems neither to be disturbed nor
disturbing. Beside a loving couple, another woman suckles her baby, another
grinds fruit on a quern. Sexuality is not taboo, and the subject does not have
to be depicted in a voyeuristic manner. The difference between the orgiastic
homosexual activity of the Chalcolithic period and the innocent lovemaking
of the Mesolithic people illustrates a basic difference between the two
societies.
In the Mesolithic hunting and gathering culture whose subsistence came
mainly from gathering, the activity of the woman must have given women a
position equal in importance to the work they performed. (I would not
conclude from this that it was a matriarchal society, for I can imagine a
society existing without any hierarchical organisation at all.) But in order to
have a term for the society we could label it matristic; women were not
subjected to the authority of men or that of other women, nor were men
subject to anyone’s authority. This fact is strengthened by the position in
which the couples perform intercourse. I am aware that I am now moving
away from archaeological reasoning to a psychologist’s interpretation of the
paintings. The Mesolithic heterosexual couples sit while making love,
contrary to those of the Chalcolithic and historic periods where the few
heterosexual depictions show the woman subjected to the man, lying on the
ground while the man stands with an erect penis; in one case there is even a
whole row of men standing in this condition, as if waiting for their turn. All
this looks more like a rape scene than lovemaking.
In the Chalcolithic and historic periods, dance depictions are sometimes
closely related to depictions of the transportation of the soul. An interesting
Chalcolithic group of this kind is found in Lakhajoar. Here one man with a
bow in his left hand is hanging on to the neck of a bovid with six legs. Some
bovids, a pig, and some animals drawn too schematically to be recognised,
follow behind them. A long row of dancers with interlocked arms
accompany this procession. The man holding on to the neck of the bovid is
adorned with a horned mask. The dancers also wear some headgear which
has the wavy decorations so typical of designs of the Chalcolithic period.
Swastika-like symbols are drawn between some of the dancers. This group of
paintings bears so many thematic parallels to the depictions of the Cemetery
H urns from Harappa, that one can hardly help thinking that what we see
here is a ritual centred around a bull who acts as the carrier of the departing
soul.
In the historic rock paintings near Pachmarhi which were executed a few
thousand years later, we find depictions of the same type, but the soul’s
vehicle to the place of final rest has changed from the bull to the horse. Even
in the modern wooden hero boards of the Korkus, which are stored by the
hundreds beneath the big mango tree opposite the Tehsil office in
Pachmarhi, it is the horse which carries the soul to heaven.
In early historic rock paintings, we also encounter cult scenes centring
around a tree fenced by railings. In older painting groups, we see cults in the
vicinity of decorated pillars or posts. In these scenes there are people
wrestling or holding boxing competitions. These paintings remind one of
the reliefs on the toranas in Sanchi showing the bodhi tree and railings.
Dances and cult activities around a stake or a pillar could indicate a popular
axis mundi cult which later developed into the pillar-erecting activity of
which the pillars of the Maurayan emperor Ashoka are the climax.

X-RAY STYLE AND DESIGN PATTERNS IN INDIAN ROCK

ART

The depiction of the skeleton and intestines of living animals is a widely

diffused phenomenon in the art of hunting and gathering societies. Even in
the earliest paintings, depictions in X-ray style are plentiful. The style is
obviously not uniform. Uniformity is lacking even between animal
depictions within the same painting group. We find anatomically accurate
portrayals of the intestines side by side with pictures presenting a schematic
design of the animal’s insides within the contoured outline of the body. It is
possible that an anatomically accurate X-ray treatment and a schematic
design express the same idea. But the difference can be puzzling for the
modern viewer.
It is difficult to distinguish a depiction in the X-ray style from a purely
decorative design pattern. As we shall see, there are many animal depictions
in which only the windpipe in the animal’s neck has been painted, while the
pattern inside its body remained purely a design.
The assumption that schematic body patterns developed from the X-ray
style cannot be proved in the case of the Indian rock paintings, for both
ways of depicting the insides of animals were contemporary. In Chalcolithic
paintings, the body patterns are only schematic, but there are depictions of
the foetuses of pregnant animals. The X-ray style showing all the intestines
in anatomically accurate positions is rare in Indian rock art, and when found
it is confined to cervines and bovines only.
The best example of an anatomically accurate X-ray painting comes from
Kathotia. This picture is almost a textbook illustration of the digestive tract
of a deer. Even the fodder in the stomach of the animal is shown.
The figures which deal correctly with the anatomical details of the
intestines neglect the skeleton, and vice versa. Frequently the digestive path
ends in a square body-design which is formed by four or more triangles, so
that a geometric body-pattern is obtained.
X-ray depictions which show the inside of the womb appear regularly,
sometimes with a foetus inside, in the Mesolithic and Chalcolithic periods.
The foetus is generally a fully developed animal standing inside the body.
The portrayal of correctly positioned foetuses seems to be confined to the
paintings of the Kushana period.
Many animals show a continuous line from the mouth to the anus, much
like the X-ray depictions in some of the engravings in northern Europe,
where it is called the ‘lifeline’. This ‘lifeline’ in Indian X-ray depictions is
generally intersected by two lines, one deriving from the shoulder going
down to the front leg, and the other going from the back to the hind legs.
Hence again a square is obtained. Generally, the heart as well as the
windpipe and the lungs connected to it are depicted. In contrast to the heart
muscle, the lungs are not filled in with paint. Quite often, the body square is
very small and drawn in the middle of the body. In two elephant drawings
from Lakhajoar the foodpipe and a connected square in the middle of the
body are left blank, while the body is otherwise filled in with red. The small
square has a spiral drawn inside.
Most Mesolithic drawings of fish have a herring-bone pattern, which
could very well be an X-ray depiction of their skeletons. In Kathotia there
are two drawings of fish in a cultic context. Here the body patterns of the
fish are in no way reminiscent of a fish bone. The features of a fish, such as
its bones and scales, easily lend themselves to decorative treatment; yet they
would still look like X-rays. There are numerous depictions of animals with
body patterns more or less reminiscent of the X-ray style. Yet the dividing
line between the X-ray style and the purely decorative patterns can only be
determined by the viewer. Here the illustrations themselves will give a better
overall view than words. The X-ray style is only used to portray pregnant
animals in the Chalcolithic period.
The rock painting site, Ramchaja, near Raisen, has about a dozen shelters
clustered in close proximity to each other. They contain paintings of animals
which together have more than twenty different body patterns. These
schematic designs could have had symbolic value, but it would seem that the
symbolism had long since been replaced by convention. This means that the
artist could not have left the space formed by the outlines of an animal body
undecorated; it would have gone against his decorative mind. In Ramchaja,
dotted designs seem to be associated with the depiction of carnivores. But
we do not know whether this observation would hold true elsewhere, and
on the whole it would seem that there was no code as to which design went
with which animal. The decision seems to have been left to the artist’s fancy.
But these patterns are never reminiscent of actual X-ray features.
The heads of big animals like the rhinoceros or the bear are regularly
painted in X-ray style. In depictions of these big animals the cavity of the
mouth, the teeth and the tongue are always shown.
 X-ray depictions are absent in the earliest historic styles. There are, it is
true, depictions of animals with body squares and spiral body patterns, but
no anatomical details are visible. Nor do I know of any depiction of a
pregnant animal dating to this period. The surprising exceptions to this
observation are the pictures of the Kushana period.
The animal depictions of this style group are graceful and have many
anatomical details which not even the Mesolithic artist thought worth
including. Female animals are shown with their udders; the genitals of male
animals are drawn as well. From Satkunda comes the depiction of a
pregnant antelope, with the foetus in a crouched position inside. What led to
this style is unknown. But it presents us with some of the finest examples of
Indian art, which are at least on a par with the courtly art of the same period
or any subsequent one. But this style disappeared, giving way to repetitive
pictures mainly featuring war scenes with horses, riders and soldiers. There
is no design in the later historic styles which could be understood as an X-
ray depiction.
People are never depicted in X-ray style. There are depictions of humans
with body patterns which are sometimes quite similar to patterns on animal
bodies. In fact, Chalcolithic paintings of human figures have patterns
generally similar to those painted on the bodies of animals. But these
patterns are never similar to actual X-ray features.
In Mesolithic paintings almost every woman has body decorations. This is
explained by the fact that the dimensions of the female body are spacious
enough to accommodate something like a design, while most male figures
are reduced to a stick-form. Only one male figure in Nagourie has a distinct
pattern of small rhomboids as decorations. Since this figure is quite unique I
feel this body pattern is related to the minutely drawn genitals in the
depiction.4
In Lakhajoar we find the depiction of a rather fat-bellied human figure
with a body pattern somewhat resembling a baby upside down. In another
picture of a female, the inside contains something resembling a fish
(incidentally the woman is eating and there is a fish on the dining table). But
since Indian rock paintings are numerous, every detail which is to be
analysed should be observed on several depictions to avoid errors or wrong
interpretations. None of the human depictions in Indian rock art have any
body design which could be identified as an internal organ, intestines or
parts of the skeleton. Hence the X-ray style has to be understood as a
technique reflecting the relation of man towards his prey. The animal
depicted does not ultimately have to be a victim; it could also be a symbol
for food, even food of a vegetable origin.
The Mesolithic artists often used the X-ray style to depict the contents of
inanimate objects. Caves and baskets are often portrayed in a cutaway
manner to show the contents. (As, for example, rats in their hole, children or
fruit in a basket.) This technique is similar in form to the anatomical X-ray
style, but here it was definitely used because the concept of a perspective
which could deal with that problem was lacking.
HUNTING AND GATHERING

Most of the Mesolithic rock paintings deal with the relation of men to
animals. Generally this is a hunter-prey relationship. Even so, as already
mentioned, there is a whole group of pictures where the animal signifies
more than just prey.
In none of the existing hunting and gathering societies of the moderate or
tropical climatic zones is animal flesh the most important foodstuff. The
jungles of India provide a rich variety of plants, fruit and tubers which, as
the paintings show, were exploited in the Mesolithic period, just as they are
exploited today by the poor tribal population to make up for shortages in
cultivated grains and pulses or as substitutes if the crops have failed.5
It is certain that during the Mesolithic period, the greater part of the
people’s subsistence was derived from collecting plant food as well as from
different kinds of small game like rats, insects, eggs, and fish. It is true that
only a small proportion of the paintings depict this kind of activity, as
against the many which portray hunting scenes. But as already mentioned,
an activity like gathering was much less dependent on non-rational support,
and it could not have inflicted as much mental stress upon the specific group
engaged in collecting as hunting did on the hunters.
The hunt is clearly depicted as an exclusively male activity. The
organisation of the bigger hunting expeditions demanded a whole array of
special abilities—such as setting traps, preparing microliths to barb and arm
spears and arrows and getting into psychological readiness with the help of
magicians, the ‘hunting chieftains’ as I call them.
Contrary to all these preparations for the hunt, food gathering, which was
solely done by women, seems to have been a simple affair. All that was
needed was a basket and a digging stick. But it has to be kept in mind that
the women who did the gathering had to have a thorough knowledge of the
ecology of the area which they were going to exploit. And organised
exploitation of an area is not far different from organised agriculture.
Women were the backstage workers in this economical set-up without
whom the rock painting show could not have gone on. Nevertheless, they
remain more or less unnoticed by casual visitors and researchers.
The large and well-organised hunting expeditions, whose depictions form
the bulk of the Mesolithic paintings on the rock walls of central India, seem
to have been comparable in importance to Mesolithic man as the seasonal
hunts, which modern tribal societies in India still enjoy, are to the latter. In
these seasonal hunts, a number of villages join together and choose a chief
of the hunt. The success or failure of this hunt serves as an omen for the
success or failure of the next agricultural season.
Hunting scenes in Mesolithic art furnish us with the greatest amount of
technological information for this period. They also give us the better part of
the knowledge we have about body ornaments, masks and group
organisation. The overemphasis on hunting in rock paintings indicates that
the very art of drawing was closely connected with a revolution in hunting
and technology. Weapons tipped or armed with microliths became more
efficient. From now on the hunter was able to bag his prey from a much
greater distance and this made for far less emotional stress.
The ease with which man was now able to kill had to be controlled
mentally; he needed new facilities to adapt his consciousness to the newly
acquired possibilities. Man’s mental balance is protected by his cultural
background. This culture has to be adapted to new innovations. It is man’s
ultimate strength that he can accomplish this very rapidly.6
Even so, the earliest rock paintings do not depict hunting scenes. To
portray an act of killing was taboo, it would seem. Fashioning paintings was
the creative mind’s way out of this antagonistic mental crisis which the new
long-range weapons like the bow and arrow brought with them. The
Mesolithic man had to hunt, but the successful hunter became guilty. This
guilt was eased by dance and later by depictions of the hunt, showing the
hunter as a symbol not accountable for a deed accomplished—a symbol in a
mass of symbols. The hunt was the domain of the male, just as gathering was
the domain of woman. The pictures leave us in no doubt about that.
Big game hunting scenes are plentiful in Mesolithic art; unfortunately
most of them are in a bad state of preservation and only parts of
compositions which were originally many metres long remain. The most
important of these groups and also one of the earliest was found in
Lakhajoar. This group spans about 6 m of a shelter wall and is covered by a
layer of grey sinter which protected the paintings from weathering but still
left them transparent enough not to obliterate the view. The drawing was
never overpainted in a later period, but there are some green paintings
underneath it which are so faint as to be hardly noticeable.
This hunting scene gives us a clear picture of all the activities involved in a
big hunt. Most of the men are beaters standing in two rows, of whom one
group is armed with rattles; the other beaters have one hand in front of their
mouths as if to break a continuous sound by pressing the palm rhythmically
to the mouth. Both lines of beaters are parallel to each other and have
between them succeeded in trapping a number of animals. Above these two
columns of beaters, one man has broken away and caught hold of the hind
legs of a pig with his bare hands. Some other people, including one with a
basket, are engaged in bending down a tree, but it is not clear for what
reason—perhaps to make a trap. Four bovids are caught between the two
rows of beaters—two of them seem to be calves. Behind these animals is the
big figure of a hunter with a bow and arrow followed by a bull. The drawing
of this animal is interesting for its body outline is a finely scratched line
which is only visible because the sinter which has accumulated in the
indentation has a slightly brighter colour than the more recent layer of
deposit over it. Below the bull, beaters can be seen gesticulating and
running. Above them a large group of dancers is following; one of them is a
‘hunting chieftain’ figure, making ecstatic movements. Behind them, the
hunt continues. An antelope has already been hit by an arrow. A row of
beaters can be seen running. In this section the painting is so obscured and
heavily weathered that it provides no further information.
Such big hunting expeditions were definitely not part of the daily quest
for food. About seventy men are engaged in this expedition. (Originally
there must have been more, but the right side of the painting has been
damaged by water and the sinter is not oblique enough for us to see this part
of the painting.) Hence this undertaking must have been organised by a
group of at least 200 people if we take two persons for every hunter engaged
—that is, women, children and people who did not participate for whatever
reason. Surely 200 people could not have been the population of a single
nomadic hunting and gathering group. The area they would need to cover in
search of vegetable foodstuff would be too vast to constitute a single
residential unit for all of them. Storage and conservation of vegetable matter
can only have been practised on a small scale since a nomadic group cannot
transport an unlimited load. What we know about hunting and gathering
groups in the ethnographic present supports this hypothesis. The number of
members in a modern group of hunter-gatherers varies from less than thirty
to never more than seventy.
The most important technological information we get from these hunting
scenes are portrayals of various microlith-barbed weapons and implements.
In India no weapon has yet been found with its original microlith arming
still intact in its setting. It is probable that microliths were used in the same
way as those on tools found in Europe and the Middle East, as well as the
microlith-armed tools used in living memory in Australia. In India all our
information as to how the microliths were used comes from rock paintings.
Most of the arrows and spears as depicted in Mesolithic rock paintings
have harpoon-like barbs made from microliths fastened on both sides of the
shaft. Of course the depictions of these tools are never clear enough to
understand which particular microlith type was used for arrows and spears.
Triangular shaped tools would be the easiest to set in this manner.
(Incidentally triangles are the most frequently occurring tool type in the
geometric tool assemblage found at Bhimbetka.) The arrow types seen in the
paintings show a wide spectrum of variation; every hunter had more than
one type of arrow in his quiver to be used for the different prey he
encountered.
Arrows with pointed tips are the most commonly depicted. Points or
slender lunates would have been the most suitable types of microliths with
which to tip an arrow. A sharp point would easily penetrate the skin of an
animal; the barbs on the arrowshaft would cause an extensively bleeding
wound. Another arrow type has a trapeze-shaped tip. Trapezes are found in
the Mesolithic toolkit as well. Set at the end of an arrowshaft they would
have had a formidably sharp cutting edge. Arrows with blunted ends
without barbs were obviously not meant to penetrate the prey’s body, only to
hit the animal hard enough to disable it. Arrows of this type, topped by a
wooden plug, are still used for bird hunting. Many of the drawings of arrows
have fletching for proper trajectory, probably made of feathers. They also
have a notch at the shaft end to accommodate the bow’s string.
The depictions of the bow are also varied in size and form. Some of them
are bigger than the hunters who use them, while others are so small that
they look rather like toys. However, it has to be kept in mind that not every
artist attached importance to the correct portrayal of things which to him as
well as to his contemporaries would be clear enough once it could be
ascertained what the artist wanted to portray. But generally speaking, the
size of the bow as depicted in Mesolithic paintings was somewhat smaller
than the size of a man.
It is not clear from the paintings whether the bow was stressed by the
arrow or whether it had already been stressed even before the arrow was
placed on the string. Since there is a painting showing a man stringing his
bow with the help of his knee, I rather believe the latter to have been the
case. Nor do the paintings give any due as to what material the bowstring
was made from. Most of the central Indian tribal people use a bamboo-
splinter which is fastened to the bow by a fibrous cord. This detail is very
well recorded in the depiction of the harp player from Pachmarhi which, of
course, is a historic painting. But none of the Mesolithic paintings is detailed
enough on this subject.
As with the arrows, spears are also barbed with microliths. The barbs
sometimes reach right to the middle of the shaft. Spears as well as arrows are
often decorated with strings and tassels. An important preparation for the
hunt was the placing of traps on animal tracks or in the area where the
hunters wanted to chase their prey. However, depictions of traps are rare and
often schematic.
There is a picture of frame-traps at Lakhajoar. Splinters of flexible
branches or bamboo were fashioned into a frame and hung into the scrub
and bushes through which the beaters were supposed to chase the animals.
The beasts’ horns and antlers would thus get entangled in the traps, and
make it easy for the hunters to kill them. In this hunting group a deer is
actually shown with his head and antlers entangled in one of the frames and
heading straight for another.
The use of snares is clearly depicted in a painting found in Putli Karar.
This shows a row of snares arranged in a semicircle and held in place by
poles. One deer is running in the direction of these traps and a hunter has
already shot an arrow into its back. Another painting from Lakhajoar shows
a bear being driven into a circular trap fashioned from long elastic bamboo
which seems to have been placed slightly above the ground and fixed to the
undergrowth. This flexible obstacle would irritate the animal and enable the
hunters to aim at him with ease or club him down. All these methods are
still practised by tribal people today, mainly for animals smaller than the
ones in the Mesolithic paintings.
A method which must have been very successful in the area around
Bhimbetka, where the hills frequently have cliffs, was to drive a herd of
animals towards precipices where they would fall to their death. But only
one painting at Bhimbetka portrays this. A group of bovids can be seen
falling. The vertical cliff is indicated by a vertical crack in the rock.
The most important part of the preparation for an organised hunting
expedition was the conditioning and shaping of the individual hunters’
minds. This knit them into a competent and organised group which could
respond to the requirements of the hunt to come. There is no doubt that this
task was codified in customs as the pictures show. The figure which stands
out in the midst of the stick-like hunters in most hunting scenes is that of
the ‘hunting chieftain’. These figures are always covered by a full-body mask
which has the intricate designs so typical of the patterns on Mesolithic
animal depictions. Frequently the hunting chieftain carries a basket, which I
understand to be a symbol of plenty.
What their exact purpose was is not clear. It is interesting in this context
that during their seasonal hunts some tribals choose chieftains whose
function is to stand behind a cane fence which has two holes for the eyes,
and cast a spell over the animals. The depiction of these chieftain figures is
reminiscent of that of women, and in one painting in Kathotia these figures
are clearly women since the artist has taken care to show breasts. Even so,
this painting does not have a clear hunting context. This raises the question
as to whether women had a magical role to play even in the male-dominated
field of hunting. It is even possible that the artist did not generally bother to
indicate the sex of these hunting chieftains since his contemporaries knew
what he wanted to depict anyway. However there is not a single instance of a
woman carrying or using hunting weapons.
In a few hunting groups, women with baskets may be seen as if to bring
back the kill to the camping ground. In one case a woman carries a basket
with two children inside. What this might indicate in a hunting context is
not clear.
In many hunting scenes, all the hunters are heavily adorned; as already
mentioned it is not always easy to distinguish between a mask and the rather
unusual head-forms which the artist sometimes gave to his creations. Masks
and excessive garments must have hampered movement during the chase. I
am aware that fashions, clothes and jewellery have almost never developed
for practical reasons only. But the ornaments which the hunters wear in
Mesolithic rock paintings would completely hamper movement. This fact
could be another indication of the cultic significance of the hunts as well as
of the paintings.
A picture from Jaora shows hunters wearing decorated helmets with
tassels hanging down. The helmets show some of the usual intricate design
patterns. The hunters, who are armed with bows and whole bundles of
microlith-barbed arrows, are not dressed in the usual small loincloth but in
finely decorated skirt-like apparel, rather similar to that of the dancers in the
Kathotia painting described earlier.
It cannot have been easy for the hunters to follow their prey with all these
fancy decorations on. In any case, the heavily adorned hunters are never to
be seen engaged in real combat; they seem to feature in the paintings only
for show—being stiff and motionless. In a Mesolithic depiction of a hunt at
Bhimbetka, five hunters encounter a rhinoceros. One man has already been
thrown into the air by the mighty horn of the animal. The spears of the
hunters have microlithic barbs. All the hunters have mask-like head-forms
resembling the snout of a rhino. In this case, the head-form seems related to
their activity. But in Mesolithic paintings this particular head-form is not
confined to rhino hunters only.
 What is strange in this depiction is the sixth figure, whom I have
deliberately excluded from the count of the hunters. This person’s activity is
directed against the hunters. Being bigger than them he attempts to grab the
spear of one of the hunters, who tries to ward him off. So there is actually
someone to fight for the rhinos. Without verbal explanation from the
Mesolithic artist we may postulate many things, but we have to admit that
we will never know everything about these paintings.
The hypothesis that the depiction of the successful hunt in painting has a
sympathetic effect on the actual hunt to come can nowhere be proved in
Indian rock paintings. The climax of the hunting scene is not the ‘successful
hunter putting his foot on the head of the slain tiger’. Nowhere is man the
master of nature; often he is its inferior. The rhino flings him into the air,
and the deified boar or a combination of many animals chase him away.
Even so there are no pictures where we can see joy over a successful kill.
Death is seldom portrayed in Indian rock art. There are of course depictions
of animals who have been wounded during the hunt. They still have spears
or arrows implanted in their bodies, are bleeding extensively and crying in
agony. The hunter is aiming the bow at the victim. But the Mesolithic artist
never dwelt on the act of killing—there is no parallel, for example, to the
image of Durga cutting through the neck of Mahishasur. In such a
representation death becomes revenge—a sanctified cruelty. Not even in the
Chalolithic and historic rock paintings, where war-scenes are plentiful, are
warriors to be seen bringing down the sword half-way through the skull of
their foes or sticking spears into their bodies, and there are no horses
trampling decapitated and dismembered foes into the dust. Indian rock
paintings never reveal which side the painter sympathised with.
I have noticed only one Mesolithic painting which could be explained as
depicting a cult activity centring around a killed animal. This depiction
comes from the already mentioned green group at Kathotia. In one part of
the group there are three hunters engaged in spearing an animal. Behind
them the head of an animal is lying on the ground. Three persons are
moving in a file adopting strange postures. Below them is the picture of a
bovid and two men doing an ecstatic dance. Above them are a pregnant
woman and a man. The second part of this group, which is separated from
the first by an area of heavily chipped rock, shows a dead boar lying on its
back and people staging some dance in its vicinity. There is a man swinging
fork-like instruments and another man playing a harp. Elsewhere, a woman
has caught hold of a long ornament on the head of a hunter carrying bow
and arrows.
The expression of this group is definitely unique. Hunting was one aspect
of the exploitation of the environment which demanded a great deal of
magical attention. This magical attention probably accounts for there being
so many hunting depictions in Mesolithic rock paintings.
The greater part of the food requirements in the Mesolithic periods came
from the collection of vegetable foods as well as small game. Paintings of
such activity are rare.
Gathering was a simple affair. A basket and a digging stick were the
means of production. But it should not be forgotten that the women needed
a thorough and intimate knowledge of many more plants than a modern
farmer has to deal with. The dependence on a vegetable subsistence caused
the Mesolithic people to move constantly between areas where plants were
to be found. In the dry season the hunter-gatherers could live in close
proximity to rivers or perennial springs. During the rains, when every nulla
carried plenty of water, it was definitely more comfortable to live in rock
shelter areas than in constructed huts on open land. It is not impossible that
some cultic activities were related to certain seasonal periods.
Fruit-gathering depictions are related to the ripening of certain fruits.
Unfortunately the pictures are not detailed enough to tell us which fruit is
being harvested. But there are several indications that it is the mango. In
some cases these fruit-gathering scenes are drawn in close proximity to
depictions of deified animals. This closeness in space also indicates the
possibility of a spiritual closeness between them.
Under the mushroom-shaped rock at Bhimbetka near where the large
deified boar is chasing away the hunter, there is another painting in a very
inaccessible hollow. On the top it shows a boar very much like the big one,
again chasing away a man. What is of interest in this painting is that there
are some people near a fruit-laden tree.
 In the same hollow there is another painting again showing the boar, but
this time without the features of other animals. There is a fruit-laden tree
below him again, but this time with a person climbing up. Another person
holds a few fruits in his arm, while one is seen to be sitting and eating from a
fruit straightaway. A fourth person with a basket is standing nearby. This
would mean that the fruit depicted was one which could be eaten without
having to remove a hard shell.
In the Rangmahal shelter at Bhimbetka one person is climbing a tree with
a basket on his or her back. The very stylised tree has only a few fruits. But
they definitely have the shape of mangoes.
This could mean that the cult of the deified boar had some relation to the
rainy season. The presence of water animals like the crab depicted in the
paintings would then make sense. It has to be kept in mind that gathering
and hunting as shown in the paintings were a highly organised harvest of
the jungle, the rivers and the open country. The cult of the Lord of the
Vegetation and that of the great Mother could have grown from this cyclical
harvest of the jungle and the rivers, long before the people began practising
what we now call agriculture.
Collecting small game and fish has to be understood as gathering, and so
fall within the sphere of women’s activities. At Jaora, there is a picture group
showing five women unearthing a rat nest and killing its inmates. This
picture records such minute details that one could believe that all the artist
had in mind was to hand down a description of Mesolithic life to people
who came that way many thousand years later. And in this the artist has
succeeded fully. Since one part of the drama is enacted underground, the
artist had to make a cut-away drawing to show the burrow with its tube-like
entrances and the rats caught inside.
A woman is digging open one of the tunnels with the help of a digging
stick. The artist even took care to show the already excavated earth. Three
dead rats lie beside the woman, and the five rats hiding inside the burrow
will surely share the same fate. A sixth rat has tried to escape through the
other tunnel and only gets as far as a pair of fateful hands belonging to a
woman waiting above. This shows how selective details were used in
Mesolithic paintings. Generally, fingers or toes are not included in the
depiction of human figures. But details were often overemphasised if they
were considered necessary for the understanding of some specific action—as
here with the two hands of the woman lying in wait. (The figure of the
woman herself has chipped off.)
This picture group also shows the baskets then in use in great detail. In
the Mesolithic period, loads were generally carried in baskets supported by a
headband. The type of basket used in this ‘rat hunt’ is similar to the conical
baskets still used in the northeastern and eastern parts of India where people
also carry them in the same fashion as the Mesolithic people in these
paintings. carrying loads with the help of headbands is a peculiar feature of
Mesolithic life. All loads in later periods were carried with the help of
wooden poles in much the same way as porters do today.
The Mesolithic paintings show several types of baskets, carrying nets and
small bags. Baskets carried by hunting chieftains are generally flat-bottomed.
The baskets used by women on gathering trips are conical. Hunters usually
carry nets or small bags with them. A well-preserved picture group at
Lakhajoar shows people fishing. As before, all the implements and
techniques are recorded in minute detail—so much so that even the branch
where one of the fishermen has hung his bag is shown. The shoreline of a
pond or pool of stagnant water in a nulla is indicated by a circular outline.
The people fishing look bulky and are probably women. They are catching
the fish with nets extended from sticks. One person has put aside the net to
leave both hands free to catch a tortoise. One fisherwoman is holding three
fish on a rod, while she tries for more with her net. Extending from this
woman downwards is another figure which seems to depict her reflection on
the water.
In the same shelter is the depiction of a tent-like structure made of
branches. By means of a cut-away view, a man, woman and child can be seen
eating inside it. Beside their very cultured-looking dining table are two
containers probably made from dried pumpkin, just like the ones tribal
people in this area use today.
Women grinding on querns are found in several rock painting groups;
this indicates that some of the vegetable food had to be processed by
grinding. The quern and grinding stone, besides being important kitchen
utensils, must have had some further significance as well. A skeleton found
in one of the trenches at Bhimbetka was covered by a great number of
rubbing stones. One painting in Kathotia shows a woman grinding fruit
which another woman has brought in a basket. close at hand, a man is
feeding a pig. Another painting nearby shows a group of several men,
women and children engaged in all sorts of activities: dancing, suckling
babies, making love and singing. Two women are grinding down some fruit
which other women bring in baskets. Since the fruit are always shown as red
dots, it is impossible to tell what fruit is being ground. But having examined
the activities of the people in this picture, I rather feel the fruit had an
intoxicating effect on them.
In Kharwai I found a painting of a woman working inside a shelter. At the
far end of the shelter are two baskets which have something round inside
them. In the above-mentioned painting from Lakhajoar similar round-
shaped objects are placed before the two persons who are eating. It is
possible that ground tubers or grass seeds were ground to a paste-like
substance and formed into balls for later consumption.
The gathering and processing of food is well documented. In the case of
small game, the pictures tell us what sort of animals were collected. But
where fruit and plants are connected, they are not detailed enough for us to
tell what species they are. It is also not clear as to which materials were used
to produce baskets, net and garments. But these questions are really no
problem because ethnological studies of Indian tribes tell us about a large
number of crafts which could have been practised in the Mesolithic period
as well. Tribal people in eastern India fashioned almost all their
requirements from bamboo and cane. Clothes which are now produced
from cotton were made of soft fibrous bark not so long ago.
The tribal population of central India, whose small plots of land have
regularly been appropriated by moneylenders, would starve for many
months in the year if they did not know how to harvest jungle produce. The
Mesolithic jungle, which then also included the fertile alluvial lowlands
along the rivers, could feed a large population of hunters and gatherers.
In an ethnological or archaeological context, we often think of the
hunting and gathering economy as being synonymous with primitive
culture. Yet the pictures are evidence of codified socio-religious expressions
and group organisation. This cultural organisation undoubtedly reflects a
highly economical harvest of the environment in accordance with the cycle
of the seasons, a way of life we can suitably call proto-agriculture. The
question which remains is how we are to distinguish between this economy
and agriculture proper.
3

Ornament Styles of the Indus Valley

Tradition
JONATHAN M. KENOYER
INTRODUCTION

T HE ROLE OF ORNAMENTS IN PREHISTORIC SOCIETY HAS not

been given the attention it deserves and very little work has been done
on the ornament styles worn by the peoples of the Indus Valley tradition.
Although numerous ornaments have been collected from excavations at
major Indus sites (fig. 3.1), the publications do little more than list them
along with miscellaneous small finds. This situation is surprising when one
considers that ornaments are traditionally used as an outward symbol of a
person’s age, social status, ethnic identity and even religious affiliation.
While the earlier excavators were definitely interested in understanding
these aspects of ancient Indus society, their immediate goals and
methodologies overlooked the types of information that a contextual study
of ornaments could provide.
Two major exceptions to this attitude are seen in the systematic study of
beads by H. Beck1 and the careful collection and recording of ornaments
and manufacturing waste done by Ernest Mackay at Mohenjo-Daro and
Chanhu-daro.2 However, Beck’s nomenclature and terminologies were not
intended to remain unchanged and need drastic revisions, while most of
Mackay’s notes and collections are no longer accessible due to the vagaries of
time. Fortunately, new opportunities to study Indus ornaments are being
provided through recent excavations at sites in Pakistan3 and India.4
These new opportunities allow for the careful assessment of the
technology and raw materials, as well as the archaeological contexts in
which ornaments are found. Through careful recording of excavations in the
cemeteries at Harappa and Mehrgarh, it is now possible to determine
specific ways in which certain ornaments were constructed and worn. This
information can now be correlated with the ornament styles seen on
figurines and seal carvings.
Figure 3.1: Ornaments of the Indus tradition
Source: 1. Terracotta Female Figurine (S. J. Marshall, Mohenjo-daro and the
Indus Civilization, London: A. Probsthain, 1931: XCIV, 14) 2. Terracotta
Female Figurine (M. S. Vats, Excavations at Harappa, Delhi: Government of
India, 1940: HP 160.3) 3. Terracotta Female Figurine (Vats, Excavations at
Harappa: pl. LXXVII, 31) 4. Copper/Bronze Female Figurine (Marshall,
Mohenjo-daro and the Indus Civilization. XCVI, 6) 5. Terracotta Male
Figurine (Harappa, 1989) 6. White Steatite Male Figurine (Marshall,
Mohenjo-daro and the Indus Civilization: 356–57, pl. XCVIII) 7. Fired
Steatite Intaglio Seal (E. J. H. Mackay, Further Excavations ay Harappa, New
Delhi: Government of India, 1938, pl. C)

Even more important is the study of the different techniques by which

raw materials were processed to produce specific types of ornaments. The
ancient artisans used both natural and highly processed materials often
combining different materials to form elaborate ornaments.
Although there are no written texts of the Indus period to tell us the value
or meaning of specific ornaments we can make some general inferences on
the basis of problem-oriented ethnoarchaeological studies. For example,
detailed research on traditional stone bead manufacture can provide
valuable models for interpreting economic and social organisation.5
Another approach to building interpretative models regarding the role of
specific ornaments in the prehistoric period is through the critical analysis
of ancient texts and folk traditions of South Asia.
In this essay, I will first present a general methodology for the study of
Indus ornaments and then focus on specific examples of bangles and beads
to illustrate how these neglected artifacts can provide a meaningful insight
into the society of the ancient Indus Valley tradition.

CHRONOLOGICAL BACKGROUND

The term ‘Indus Valley tradition’ refers to the long series of cultural
developments taking place from around 6500 bce to approximately 1500
bce, in the areas of modern Pakistan, north-western India and parts of
Afghanistan.6 The roots of this tradition can be traced to the beginning in
the Neolithic period (6500 bce) in sites such as Mehrgarh. Specific
ornament styles recovered from the Neolithic and early Chalcolithic levels at
Mehrgarh appear to have provided the basis for the early Harappan and
Harappan ornament styles found at sites such as Nausharo, Harappa,
Mohenjo-daro, Chanhu-daro, etc.
‘Early Harappan’ is the term usually associated with the formative phase
of the Harappan or Indus civilisation which is dated from 3500–2600 bce.7
The Harappan phase is now dated from approximately 2600 to 1900 bce and
is characterised by a fully developed urban civilisation that extended over an
area that was twice the size of contemporaneous civilisations in
Mesopotamia or Egypt.8 The Indus Valley tradition continues through the
late Harappan period (as late as 1500 to 1000 bce) and overlaps with the
cultural developments that coincide with the Vedic period in the northern
subcontinent.9
Whereas in the past scholars assumed that there was a major ‘Dark Age’
between the Indus civilisation and the early Historic period in the northern
subcontinent, current research is showing that there are in fact numerous
continuities.10 When properly demonstrated, these continuities can serve to
strengthen the reliability of inferences about the Indus civilisation.11

ORNAMENTS IN MODERN AND TRADITIONAL SOCIETIES

Before attempting to understand the role of ornaments in ancient societies it

is important to establish a framework of study and identify biases that can
distort our research methods and interpretations. In modern, Euro-
American and industrialised societies, with the exception of wedding rings,
most body ornaments are regarded as non-utilitarian and essentially
optional forms of adornment. Although ornaments are worn in most
communities, heavily adorned individuals, either male or female are usually
associated with ostentatious displays of status, wealth, sexuality or fantasy.
Such displays are quite acceptable in coronation ceremonies or at festivals
such as Mardi Gras or Halloween, but are not encouraged as an everyday
phenomenon.
Furthermore, in modern industrial societies, finished ornaments usually
are purchased from a store that is far removed from the actual manufacture.
While different qualities and styles of objects may be used to represent
general social status and ethnic affiliation, most ornaments no longer have
any specific social or ritual significance. Even when ornaments do have
ritual significance, usually it is derived from the use of specific symbols and
not from the manufacture or raw material. In our modern society, most
people cannot differentiate amber, ivory or gem stones from imitations
made in glass or plastic. One reason for this is that it really does not matter,
since few people wear these materials for their magical or ritual properties.
In contrast, many of the ornaments worn in pre-industrial or traditional
communities throughout the world are worn for specific social and ritual
purposes. Traditional Native American ornaments, the ornaments of the
Pacific islanders, and in many communities of South Asia, continue to retain
specific social and ritual significance. Consequently, the manufacture, trade
and use of ornaments are explicitly or implicitly prescribed. Ornaments are
made from specific types of raw materials and they have standardised shapes
and colours. Furthermore, the technology and season of manufacture are
often regulated and in some cases, only specific individuals are allowed to
manufacture, trade and wear certain ornaments.
The recognition of these culturally defined processes by most members of
the society provides a greater sense of meaning and significance to the use of
specific raw materials, the technology of production, and eventually the use
of specific ornaments.
On the basis of observations among both settled and mobile communities
of the north-western subcontinent, it is evident that ornaments function in
several different ways. Generally speaking, valuable metals and stones are
fashioned into ornaments that depict important ritual or symbolic motifs.
These ornaments serve to protect the wearer, to identify the social and
economic status of the wearer, and as a means of storing wealth. In
displaying these ornaments the wearer is usually aesthetically pleasing to
other members of the community. Ornaments made from less valuable
materials generally function in the same ways except that they do not
necessarily represent actual wealth.
Because of the important social and ritual meanings attached to
ornaments, standardised designs and styles have extremely long life cycles.
Nevertheless, there is scope for creative variation and this is revealed in
design and technology as well as in the combinations of ornaments worn by
different individuals.
In contrast to these types of multipurpose ornaments, some ornaments
are worn primarily for their amuletic or magical properties. These may be
made from rare or common materials and often they are not displayed
openly, but are worn beneath the clothes or sewn into amulet bags. Because
of their amuletic properties these ornaments are worn for life or until they
have served their purpose. Such ornaments may be passed on to other
individuals or ritually disposed of, either by burial or by placing them on a
specific shrine.
 In such traditional communities, ornaments are not only utilitarian, but
are essential to the proper functioning of the social group [emphasis added].
They protect, identify and preserve an individual’s place in society and the
natural environment.12 similar attitudes towards ornaments are documented
from excavations and texts of the early civilisations in Mesopotamia13 and
Egypt14 and we can assume that ornaments produced by artisans of the
Indus tradition were used in much the same manner. There are, however,
important regional differences in terms of how ornaments were used and
eventually discarded, and a systematic research methodology needs to be
established to properly document and compare these regional patterns.

ORNAMENT RESEARCH METHODOLOGY AND

DEFINITIONS

When outlining specific research methods, it is always best to begin with

definitions, in order to avoid confusion and to allow other scholars to
compare data and develop more effective analytical methods. In this
discussion, I will focus only on ornaments that appear to have been worn by
humans, either as discrete objects or possibly sewn onto clothing. categories
of ornamentation that are preserved archaeologically include objects such as
beads and pendants, rings, bangles, belts, fillets, pins and various head
ornaments (fig. 3.2). These terms are used in various ways by different
cultures, but in order to avoid confusion I propose the following definitions
for the general sets of ornaments that will be discussed in this essay. The
general definition for beads and pendants are similar to those used by Beck
in 1928,15 but the specific terminology used for bead types are quite
different.
Figure 3.2: Harappan male ornament styles
Source: 1. (G. F. Dales and J. M. Kenoyer, ‘Preliminary Report on the Third
Season of Work at Harappa, Pakistan’, Pakistan Archeology, vol. 24, 1990, pp.
68–176.) 2. (Mackay, Further Excavations at Mohenjodaro: pl. LXXII, 7) 3.
(Mackay, Further Excavations at Mohenjodaro: pl. LXXIII, 1) 4. (Mackay,
Further Excavations at Mohenjodaro: pl. C) 5. (Mackay, Further Excavations
at Mohenjodaro: pl. LXXIV, 23) 6. (Vats, Excavations at Harappa: pl. LXXVI,
13) 7. (Marshall, Mohenjo-daro and the Indus Civilization: pl. CXVIII)

Bead—any object that is perforated along its major axis, generally worn on a
cord or wire, sewn onto clothing or used as an ornament.
Pendant—any object that is perforated or scored at one end and is hung or
attached to a cord or wire, sewn onto clothing or used as an ornament.
Bangle—any circlet (closed or open) made of a continuous homogeneous
material that can be worn on the arm or ankle.
Bracelet—any circlet made of components such as beads, chain or cord, etc.
that can be worn on the arm or ankle.
Due to the regional climate and various disturbance factors, only
ornaments that are made from permanent and durable materials are
available for study in the Indus sites. However, the Indus peoples may have
used other forms of ornamentation that have not been preserved, i.e., body
paint or tattoos, horn, feathers, decorative fabrics, flowers, etc. Some of these
more ephemeral forms of ornamentation, such as flowers or fabrics are in
fact depicted on figurines and seal carvings. In most cases, these less
permanent types of ornaments were used in combination with other more
durable ornaments (fig. 3.2).
Most cultures at this time in history had a similar range of ornament
categories, but each region seems to have developed a specific emphasis on
certain ornament styles. The shapes and colours of Indus ornaments, and
the way they were worn, definitely represent a regional cultural aesthetic.
These distinctive ornament styles can be defined through a careful study of
the five major variables, which include the raw materials, the technology of
manufacture, the physical shape, the style of ornaments (combinations of
colour, design and shape) and the ways in which specific ornaments were
combined and used.

RAW MATERIALS AND TECHNOLOGY

In the past, scholars have underestimated the importance of specific raw

materials and the technology used to manufacture ornaments. It is quite
clear, however, that during the Neolithic and Chalcolithic periods, certain
raw materials were selected as being culturally meaningful and were made
into specific types of ornaments. While some of these raw materials initially
may have been selected because they were easily obtained and easily
fashioned into specific types of objects, other raw materials were difficult to
obtain or were completely transformed through manufacture. By the
beginning of the urban phase of the Indus tradition, around 2600 BCE some
of the processes used to manufacture ornaments were extremely specialised.
In the context of this essay it is not possible to go into detail on all of the
different raw materials, but a few examples will serve to illustrate this point.
For example, easily collected natural sea shells or water-worn pebbles were
often perforated and worn as necklaces or bracelets without further
modification.16 Such ornaments have an exceedingly long history and
continue to be used in present day South asia.
 In contrast, during the Neolithic period at the site of Mehrgarh, Pakistan
(c. 6500–4500 BCE), soft black and grey-white steatite was used to make
small disc beads and pendants.17 During the subsequent Chalcolithic
periods, we see evidence for the firing of these same types of beads resulting
in hard white beads.18 Later on, by 2600 BCE we find the same variety of soft
steatite being shaped into beads or pendants that often were painted or
glazed. By firing these ornaments in high temperature kilns the Indus
artisans were able to produce various colours, such as red, brown, blue,
green, black or white.
Another complex technique that became common around 2600 BCE is the
manufacture of compact glazed silica or faience.19 White quartz pebbles
were ground to extremely fine powder and melted with various colourants
to make a glassy frit. This frit was then reground and glazed to produce
various colours of faience ornaments, the most common colour being azure
or blue-green.

Table 3.1: Harappa 1986–90: Bangles sorted by raw material type

Table 3.2: Ranking of bangles by raw material and technology

 The selection of steatite and faience for producing specific types of
ornaments such as steatite disc beads, square steatite seals and glazed faience
bangles serves to distinguish the Indus ornaments from those produced in
other regions at the same time. Both of these techniques became highly
specialised in the Indus cities, and although they do occur in Mesopotamia
and Egypt, the specific quality of the finished objects and the techniques of
manufacture appear to be significantly different.20

RELATIVE VALUE OR RANKING OF ORNAMENTS

Keeping these examples in mind, it is possible to assume that the value and
importance of an ornament was not simply a function of the final shape or
design, but also derived to some extent from the type of raw material,
combined with the time, effort and techniques used to produce a specific
object. Even though it is not possible to assign specific values to ornaments,
it is possible to differentiate and rank specific ornaments or ornament styles
on the basis of raw materials and technology.

Bangles
For example, at most sites of the Indus tradition, terracotta bangles in
various shapes and sizes occur in extremely large quantities. While some of
the objects classified as terracotta ‘bangles’ may actually have been used as
kiln setters21 and not actually worn, the extremely high number of terracotta
bangle-like fragments is still significant. During the course of five seasons at
Harappa, 34,127 terracotta ‘bangles’ and fragments were recovered,
constituting 97.72 per cent of the total number of prehistoric bangles
recovered. From these same excavations only 390 (1.12 per cent) faience,
340 (0.97 per cent) shell, 48 (0.14 per cent) stoneware, and 17 (0.05 per cent)
copper* bangles/fragments were recovered. Due to the fact that all of these
types of bangles, except for copper, tend to break into an equally wide range
of fragment sizes, the total numbers do in fact provide meaningful
comparisons. (*Copper appears to have been recycled and the low number
for the copper bangles is not representative).
The ranking of bangles on the basis of raw material and the level of
technology seems to be correlated with the overall abundance of bangle
fragments from the excavations. Gold, silver, stoneware, copper and faience
would be the highest ranked and terracotta the lowest, with shell falling
somewhere in between. The higher numbers of faience bangle fragments
could be a function of their tendency to break into smaller fragments than
shell bangles. While most of these bangles are relatively comparable in terms
of shape, there are specific elements of design and decoration as well as the
contexts in which they have been found, that will be further evaluated
below.

Beads
The ranking of beads is not quite as simple because of the numerous raw
materials and technologies involved (Table 3.3). On the model of bangles,
one might expect that there would be thousands of terracotta or bone beads
and relatively fewer beads made from materials that were more difficult to
acquire, and manufacture. However, this pattern is not evident, and steatite
beads appear to be the most common, comprising 77.30 per cent of all
recovered beads and pendants. Most of these beads are made from white
fired steatite and have a very short cylinder or disc shape. This shape and
colour of bead was first made from shell during the Neolithic and then made
from steatite and fired steatite in the Chalcolithic and early Harappan
periods. Except for a few examples in faience or paste, disc beads came to be
made exclusively from fired steatite during the Harappan phase. It is quite
clear that these beads were produced in extremely large quantities, and worn
in long necklaces or anklets. Two such ornaments recovered in the cemetery
at Harappa contained 340 and 297 beads respectively.22

Table 3.3: Harappa 1986–90: Beads sorted by raw material type

By ranking the beads on the basis of raw material and technology (Table
3.4) terracotta and bone fall below the other raw materials though they were
not the most abundant. shell and some varieties of stone that can be worked
with relatively simple technologies fall in the middle, but most of the beads
are made from raw materials that require relatively complex technologies
and high temperature kilns.
The objective of these examples is to demonstrate the relationships that
can be found between general categories of ornaments, raw materials and
technology. the sample of material collected from recent excavations at
Harappa is unique in that it represents the total number of artifacts
recovered by a uniform recovery and recording method. Future research at
sites in South Asia, Mesopotamia and Egypt may some day provide
comparable samples, but at this time there is no such published data. In the
absence of such comparative samples, the meaning of these patterns can be
further understood by examining the ornaments on the basis of shape,
colour and the ways in which ornaments were combined and worn.

Table 3.4: Ranking of beads by raw material and technology

ORNAMENT SHAPES AND STYLES

There are several levels at which ornament shapes and styles can be studied.
In most cases, archaeologists discover single ornaments, such as beads or
broken bangles that were discarded or lost and became buried along with
domestic garbage. Each of these individual ornaments preserves a total
shape and colour that can be analysed and classified according to specific
ornament types. Such studies are currently underway for the site of Harappa
and the Indus tradition in general.
At a different level we can study the ways in which these separate
components were combined by different individuals to form complete
necklaces or body ornaments. Ornaments found in burials and depicted on
figurines and carvings demonstrate that the Indus people had developed
distinctive ornament styles. Ornament style refers to the combination of
different components to produce specific patterns derived from shapes and
colours. Preliminary results from the study of Indus bangles and beads
indicate that both the raw materials and the varieties of ornament types
changed significantly over time. In this essay, most of the discussion will
focus on specific examples from recent excavations at Harappa, with some
correlations made to other sites.

Bangle and Bracelet Types

The earliest bangles from the aceramic Neolithic period at Mehrgarh (Period
IA) appear to have been wide shell bangles and bracelets from shell and
stone beads.23 After ceramic production became well established in the later
Neolithic and Chalcolithic periods, terracotta bangles were introduced, but
shell bangles and bead bracelets continued to be used. During the early
Harappan phase there is a gradual diversification in raw materials and
bangle styles, but the real burst of design and variation is seen in the urban
Harappan phase, circa 2600 BCE (Table 3.5). Harappan bangles were made
from a wide range of raw materials; terracotta, fine terracotta, faience,
stoneware, shell, copper/bronze, gold and silver. The designs for the different
bangles are quite varied and range from simple circlets with round section,
to decorated bangles with incised, glazed, pinched and moulded motifs.

Bead Types
During the Neolithic and early Chalcolithic at Mehrgarh beads were made
from both locally available and exotic raw materials and in a relatively
limited range of shapes and sizes.24 The vast majority of the beads were
made in short or long cylindrical shapes, though there are some other
varieties. Most of the beads were made from relatively soft raw materials;
shell, limestone, steatite, serpentine, lapis lazuli and even turquoise. There
are some examples of hard carnelian beads in the Neolithic, but they all
appear to have been short biconical shapes that can be perforated by
chipping rather than drilling. Later, during the Chalcolithic period (4200
BCE Period III) there is evidence for the use of hard stone drills and the
production of longer bead shapes in agate and carnelian. A very important
development during this same period is the firing of steatite to produce
white steatite beads25 and also probably the intentional heating of agate to
produce a deeper red-orange carnelian. The raw materials, techniques of
manufacture and even the bead styles of these earlier periods, undoubtedly
formed the basis for later developments in the early Harappan and
Harappan phases.26
Current research indicates that the production of beads in the early
Harappan and particularly the Harappan phase sees a dramatic increase in
the variety of raw materials and also an increase in the shapes and varieties
of beads being produced. One of the important features of the beads made
during the Harappan phase is that many bead types, defined by shape, size
and decoration, were made in different qualities of raw material (Table 6).
Continuing in the trends established during the Neolithic and chalcolithic
periods, the Harappan bead makers chose to use steatite, fired steatite,
banded agates, carnelian and other multi-coloured rocks. By careful
chipping and grinding they were able to accentuate certain patterns of
banding, dots, circles or mottling that were present in the natural stones.
However, the Harappan artisans did not limit themselves to the natural
rocks alone, and by using other raw materials and new techniques they
made copies or imitations of the natural stone beads in terracotta, fired
steatite, or faience (Table 3.6).

Table 3.5: Major Harappan phase bangle types and raw materials
Table 3.6: Selected Harappan phase bead types and raw materials
Styles of Adornment: Bangles
In order to better understand the role of these ornaments in the Indus
tradition and especially the urban Harappan society, we need to determine
how these ornaments were worn and who wore them. Bangles, by definition,
refer to circlets worn on the arms, but there are ethnographic examples of
circlets worn in the hair, on belts, on the ankles, or sewn onto clothing. The
evidence from burials and figurines from the Neolithic through the urban
Harappan phase suggest that circlets were worn primarily as bangles on the
lower and upper arms. In Neolithic Mehrgarh, a single wide shell bangle was
worn on the wrist and bead bracelets often were worn on both wrists. Later,
with the introduction of narrow bangles of shell or terracotta, several
bangles appear to have been worn at the same time, often on both wrists.27
This pattern is well documented from figurines and seal carvings of the
Harappan phase (figs. 3.1, 3.2 and 3.3).
During the Harappan phase, some female figurines depict three of four
bangles at the wrist and two or more bangles above the elbow, often with
equal numbers of bangles worn on both arms.28 A similar style is seen on a
male figurine from the early Harappan phase at Nausharo (Period ID).29 A
slightly modified pattern is depicted on the famous copper figurines from
Mohenjo-Daro, where several bangles were worn on the right arm at the
wrist and elbow, but the left arm was filled with bangles from wrist to
shoulder.30
The famous seals depicting a seated male with horned headdress show
both arms filled with what have been interpreted as bangles. In one seal, the
bangles appear to be of different types. A bangle with projecting section was
worn at the wrist, followed by four flat bangles on the forearm. Another
projecting bangle is seen above the elbow followed again by four flat bangles.
A final projecting bangle is worn on the upper arm.31 A second seal shows a
similar individual with bangles from wrist to shoulder, seven on the right
arm and eight on the left arm.32

Table 3.7: Harappa 1986–88: Primary context burials

 Figurines and seals show that bangles were worn on both arms from wrist
to shoulder, or as one or two bangles at the wrist of one or both arms.
However, they do not allow the identification of what types of bangles were
being worn or the combinations of design and colour being used. The only
concrete evidence for the types of bangles being worn comes from burial
ornaments (Table 3.8). In reviewing the reported grave goods from all of the
cemeteries of the Harappan phase, only white shell bangles appear to have
been buried with the dead. In recent excavations at the cemetery of Harappa
itself, white shell bangles were found on the left arm of middle-aged adult
women (age 35–55). Sometimes they were worn on the lower arm or wrist,
but in two cases bangles were worn both above and below the elbow. The
bangles in the earliest burials are slightly wider than those found in later
burials, and the thinnest bangles are found in the latest burial. However,
even though the width of the bangles changed, all of them were incised with
the same style of chevron motif and all of the bangles were worn with the
chevron motif oriented in the same direction.
There is only one example of a middle-aged adult male (Burial 147A) with
a broken shell bangle that appears to have been worn on the left wrist. This
individual was buried in a wooden coffin and covered with some form of
shroud. Another elaborate ornament was found at the right side of the skull
towards the back of the head. This apparent head ornament was made from
thousands of steatite microbeads and a jasper bead combined with two or
three shell circlets.33

Table 3.8: Harappa 1986–88: Burial ornament styles

 Another instance of shell circlets is reported from Wheeler’s earlier
excavations in this same cemetery. The burial, ‘probably that of a female’ was
interred in a coffin with what appears to have been a reed shroud. ‘On the
right middle finger was a plain copper ring; one shell ring (probably earring)
lay to the left of the skull and two to the left of the shoulder’.34 The
similarities between these two burials are striking and since the sex
identification proposed by Wheeler is somewhat uncertain, it is difficult to
determine if the shell circlets worn in a hair ornament were used by both
men and women or not.
On the basis of burials from the Neolithic period at Mehrgarh and later
figurines from the Chalcolithic period, we can determine that bangles or
bracelets were worn by both men and women, adults and children. During
the Harappan period, bangles are seen primarily on female figurines and in
female burials, although some males do appear to have worn bangles.
occasionally, shell circlets were worn as hair ornaments on men. Disc bead
ankle bracelets were found on a female burial at Harappa in 196635 and one
possible example of a similar ornament was found in the recent excavations
(Table 3.8).36 It is important to note that none of the individuals buried at
Harappa or any other reported site have been found wearing terracotta,
faience, copper or stoneware bangles.

Styles of Adornment: Beads

While bangles are generally limited to the arms, beads or pendants can be
worn in numerous ways and allow for considerable individual variation. In
assessing the ways in which beads were commonly worn in the Indus
tradition, we can look at the general styles reflected by figurines (figs. 3.1
and 3.2) and compare these with what is seen in ornaments found in burials.
During the Neolithic and Chalcolithic period at Mehrgarh, beads were
used in a wide variety of ornaments. They were worn by infants, children
and adults, both male and female. The earliest use of ornaments is in fact
recorded only in the Neolithic burials, where we see the use of shell and
limestone beads in head bands, necklaces, belts, bracelets and anklets.37 The
wide range of styles represented in the burials indicate that there was
considerable individual choice as to how beads were strung and worn. A full
analysis of the burial ornaments will hopefully provide a better
understanding of these early ornament styles.
 On the basis of terracotta figurines of the later Chalcolithic and early
Harappan periods,38 it is clear that individuals often wore numerous
necklaces and pendants. However, large quantities of ornaments as depicted
on the figurines have not been found in any burials. This suggests that
certain ornaments, presumably the ones which represented valuable wealth
or socio-ritual status, may have been passed on to living relatives rather than
being buried.
Most of the females from Period V and VI are depicted with wide torques
of five to seven massive strands of matched or graduated beads. In period
VII these same torques are depicted with additional pendants or discs.39
Some figurines have, longer necklaces made with double or triple strands of
beads that support a pendant. These longer necklaces reach to the middle of
the breasts or even to the stomach. During Period VII, male figurines are
depicted with three to five discs and a single pendant bead reaching to the
middle of the chest.
This style of wearing several necklaces with discs or long beads as
pendant is continued in the Harappan phase, where both male and female
figurines are depicted with chokers or short necklaces bearing three or more
long pendant beads. Some of the more elaborately adorned female figurines
have layers of necklaces extending to the waist, each with several long
pendant beads.
In addition to being worn around the neck, beads may have been used to
make girdles or belts that are often depicted on female figurines. These belts
consist of three or more strands with a central disc or buckle. Beads also
appear to have been worn as hair ornaments, either as single beads or
pendants, or as long strands of beads that encircle the elaborate headdress.
The elaborate ornaments depicted on figurines were not simply the
fantasy of some sculptor, but are substantiated by the occasional hoards of
jewellery found hidden in jars or buried under the floors of ancient
Harappan houses.40 These hoards usually comprise gold, silver and
copper/bronze ornaments as well as exquisite stone beads made from agate,
carnelian, jasper and turquoise. Although there are some examples of
necklaces or bracelets made exclusively of gold components, most of the
ornaments are made from several varieties of raw material.
 Some of these ornaments have been carefully reconstructed and
demonstrate that the necklaces depicted on the figurines probably were
constructed of many different varieties of high quality raw materials,
including copper, gold, silver, faience, fired steatite, agate and other
colourful rocks. The manner in which these various raw materials were
combined is distinctive. Single strand necklaces were usually composed of
gold beads alternating with natural stone beads, faience or fired steatite
beads. Many of these stone or faience beads were capped at both ends with
gold. The gold capped beads, particularly the cylindrical blue-green faience
beads are relatively common and are found in all of the different hoards. In
one hoard, they are even used as pendants,41 while in another example from
Harappa they were added to the end of pendant agate beads.42 Pendant
beads are easily identified because they were usually attached to the necklace
by a gold or copper wire that was run through the bead and coiled at one
end to form an eye.
Multiple strand necklaces were also quite common and were generally
made up of matched or graduated beads, usually of the same shape and raw
material. They were attached to terminal beads that had multiple holes and
the different strands were kept separate by spacer beads with multiple holes.
This pattern was used for making long necklaces as well as bracelets or
armlets.
One of the most exquisite ornaments is a massive necklace or belt made
from forty-two very long bicones of carnelian arranged in six strands.43
These relatively heavy carnelian beads were combined with small spherical
beads, multiple hole spacers and terminals, all made from copper/bronze.
Traces of gold leaf indicate that these copper/bronze components were
originally covered with gold.44 A similar necklace or girdle, also with
copper/bronze components, was found in excavations at the small site of
Allahdino, Pakistan.45
One important feature of these types of long carnelian beads is that
generally they show considerable wear, and broken beads were often
reground to make into shorter beads. The original shape of the beads was a
long truncated bicone, but over time the central ridge was worn down so
that they were often referred to as long barrel bicones.46 Almost identical
carnelian beads have been found in Mesopotamia, but they are relatively
unworn and the central ridge is well defined, like the unfinished and broken
beads from the ancient workshop in Chanhu-daro (personal observations).
This difference can be explained by the fact that in Mesopotamia, the beads
were taken out of circulation and buried with elites, while in the Harappan
context, they were worn for many generations and only went out of circulation
when they were lost or hidden in hoards [emphasis added].

Beads and Pendants Found in Burials

The important contrast between the types of ornaments depicted on
figurines and those found in the burials reveals an important facet of
Harappan belief and value systems. Excavations at Harappan cemeteries
(Harappa, Lothal, Kalibangan and Rupar) demonstrate that large quantites
of status goods and elaborate ornaments were not included with the dead.
However there are occasional beads and necklaces that were made from
relatively valuable materials, such as fired steatite, natural stones and even
gold (Table 3.8).
Steatite disc beads were included with many of the male and female
burials. In most cases the steatite beads were worn as necklaces or ankle
bracelets, and these ornaments invariably comprised several hundreds of
beads. Although these disc bead ornaments were probably less valuable than
long carnelian beads or gold ornaments, they do represent considerable
technological effort and their inclusion in the burials is quite intriguing.
There are relatively few other types of beads with the burials, but they may
provide a clue to the meaning of burial ornaments. Some individuals, both
male and female were buried with one to five small beads tied at the waist
(see Table 3.8). These beads were made from copper or natural stones such
as carnelian, agate and lapis lazuli. Since these beads were generally found
on the pelvis or at the lower back, it is possible that they were worn beneath
the clothing and next to the skin. As such, they would represent amulets that
were worn, not for display or communication, but rather for protection or
good luck.
In some male burials, two or three beads made from natural stones were
worn around the neck or on the chest. One adult male (196A) was buried
with three natural stone beads and three tiny gold beads. This same
individual was wearing a single copper bead at the waist. None of these
beads appear to be part of a complex ornament as is depicted on the
figurines and may have been worn as amulets rather than as displays of
wealth and status. The fact that the stone beads were well rounded and the
perforations were worn and polished by the cord suggests that they may
have been passed on for many generations before being buried with this
individual.
An important type of pendant is the small truncated cone, with a single
line incised around the top. This pendant is usually made of black basalt,
black steatite or dark green serpentine, and five out of six examples from the
cemetery have been found in association with female burials. one of these
pendants was found under the chin of burial 134A and suggests that they
were worn at the throat. Since only one example has been found in burial fill
associated with a possible male burial, it is probably safe to assume that
these pendants represent a form of ornament or amulet associated with
women.
Although there are similarities between the ornament styles found in the
burials and those represented on figurines and seals, it is evident that the
general function of burial ornaments was quite different. Some of these
ornaments may have been worn beneath clothing or a shroud and most of
the corpses with ornaments were enclosed in coffins as well. This pattern
suggests that instead of functioning as displays of social status and wealth,
the ornaments were included in the burials for specific ritual purposes, or
were highly personalised and could not be transferred to living relatives.

THE ROLE OF HARAPPAN ORNAMENTS

The systematic analysis of raw materials, technology, ornament shape and

style of use, provide the necessary background for discussing the role of
ornaments in the newly emerging Harappan urban context. During the early
Harappan and Harappan phases we see for the first time in South Asian
history, large populations aggregated in regional urban settlements. It was
during this time that the rules defining civic order and interaction were in
the process of being refined and codified. In this urban context we see the
development of relatively standardised shapes and styles of ornaments that
continued to be worn for hundreds of years. As decorative objects that
would have been openly displayed, these ornament styles represent a highly
efficient form of visual communication and public identification [emphasis
added] (figs. 3.2 and 3.3).

Figure 3.3: Harappan female ornament styles

Source: 1. (Mackay, Further Excavations at Mohenjodaro: pl. LXXV, 3) 2.
(Mackay, Further Excavations at Mohenjodaro: pl. LXXXVII, 41) 3. (Mackay,
Further Excavations at Mohenjodaro: pl. LXXII, 4) 4. (Mackay, Further
Excavations at Mohenjodaro: pl. LXXV, 1) 5. (Mackay, Further Excavations at
Mohenjodaro: pl. LXXVII, 58) 6. (Vats, Excavations at Harappa: LXVII, 35)
7. (M. S. Vats, 1940: unpublished) 8. Necklace and bracelets (Marshall,
Mohenjo-daro and the Indus Civilization, pl. CXIV, 6), anklets (Mackay,
Further Excavations at Mohenjodaro: pl. LXIII, 5) 9. (Mackay, Further
Excavations at Mohenjodaro: pl. LXXVII, 51, 53) 10. (Vats, Excavations at
Harappa: HP 160.3) 11. (Mackay, Further Excavations at Mohenjodaro: pl.
LXXV, 5) 12. (Marshall, Mohenjo-daro and the Indus Civilization: pl. XCIV,
14) 13. Headdress and necklaces (Marshall, Mohenjo-daro and the Indus
Civilization, pl. XCIV, 1), necklaces and belt (Mackay, Further Excavations at
Mohenjodaro: pl. LXXV, 17) 14. (Mackay, Further Excavations at
Mohenjodaro: pl. LXXV, 5) 15. (Mackay, Further Excavations at
Mohenjodaro: pl. LXXV, 19) 16. (Mackay, Further Excavations at
Mohenjodaro: pl. LXII, 5) 17. (Mackay, Further Excavations at Mohenjodaro:
pl. LXXV, 21) 18. (Mackay, Further Excavations at Mohenjodaro: pl. LXXV,
1)

The increased diversity of bangle and bead styles and raw materials
during the Harappan phase can be attributed to a need for more outward
symbols of identity to differentiate an increasingly diverse urban population.
This sort of identification would have been extremely important in order to
avoid conflict and confrontation between socially stratified and ethnically
diverse populations.
On the other hand, the reproduction of identical shapes and styles using
different raw materials can be interpreted as the need to reinforce general
cultural beliefs or aesthetics. The relative value of the raw materials
themselves could serve to reinforce ranking or stratification within the society
as a whole [emphasis added].
The examples of bangles and beads presented above show how several
varieties of raw materials were used to produce ornaments that had the same
basic shape and often the same colour. From a distance these ornaments
would have been indistinguishable and would communicate similar
messages about the wearer. Such ornaments may have served as symbols to
express overarching cultural beliefs or aesthetics. Because of the fact that
they were made in many different raw materials, these symbols would have
been accessible to all members of the society, thereby reinforcing important
belief systems and the social order.
 Upon close examination, however, the discerning eye would be able to
distinguish the precise nature of the ornament, its relative value and
presumably the economic and socio-ritual of the wearer. The manufacture of
similar beads or bangles from different raw materials is not unique to the
Indus civilisation, but is seen in all societies where valuable raw materials
are not equally accessible to all members of a community. In such contexts,
different raw materials can be used to differentiate individuals on the basis
of economic, social and ritual affiliations.
Based on the discovery of hoards containing exquisite ornaments of
carnelian, turquoise, copper, silver and gold, we can assume that these raw
materials were relatively more valuable than the common forms made from
steatite and terracotta. Furthermore, we can assume that individuals who
owned and/or wore these exquisite ornaments would have been ranked
relatively high in terms of economic or socio-ritual power. The distinct
ranking in ornament styles on the basis of raw material and technology
most likely reflects social ranking or even stratification within Harappan
society. This ranking or stratification may have applied both to individuals
who wore the ornaments, and in a different context, to the artisans who
manufactured them.
In other words, an individual wearing numerous carnelian and gold
ornaments was undoubtedly of a higher status than an individual wearing
similar ornaments made from steatite or terracotta. Similarly, the artisans
who worked on carnelian and gold may have been ranked higher than the
artisans who made red painted terracotta imitations. These examples may
seem a trifle obvious given the ranking of gold in modern societies, but it is
important to point out that the interpretations presented above rely
primarily on archaeological patterning and not simple analogies to modern
value systems.
There is little doubt that ornaments played a significant role in the
reinforcement of the Harappan socio-ritual order, but there must have been
some fluctuation over the course of 700 years. Detailed chronological studies
of ornament styles are under way and will be extremely important for
identifying the role of specific ornaments, both as unifying symbols and
ethnic identifiers. A brief discussion of the role of shell bangles and bangles
in general will illustrate this point.
The shell bangle appears to have been used as a symbol that expressed an
overarching unity as well as sexual and possibly ethnic distinctiveness. Shell
bangles with the incised chevron design are found at most settlements of the
Harappan phase, and although the symbolic meaning of shell bangles could
have differed from region to region or over time, they do represent a shared
aesthetic.
On the other hand, the limited quantity of such bangles indicates that
they were not available to all members of the community. The recent
cemetery excavations at Harappa indicate that this shell bangle style was
worn primarily by middle-aged adult women and that this use remained
quite uniform for several generations, if not longer. Although the width of
the shell bangles decreased over time, the continued use of the chevron
motif and the uniform manner of wearing them on the left arm, suggests
that the women buried with these ornaments came from a stable and well
established community. It is not clear when this design of shell bangle
became common, but it is reported from early Harappan sites and continues
throughout the Harappan phase.47
The extensive manufacture and use of other types of bangles in contexts
outside of burials is also quite significant and can be interpreted in several
ways. It is possible that women who wore shell bangles did not wear other
types of bangles, and that the cemeteries represent only a limited segment of
the population. Alternatively, if they did wear other types of bangles, these
were systematically removed or broken before the women were buried.
The distinctive patterns of shell bangle use and discard suggest that they
were more than simply decorative ornaments. In view of the extensive use of
other types of bangles it is possible to propose that bangles, as an ornament
style, came to be used as important symbols for defining and reinforcing the
socio-ritual order during the Harappan phase.
When compared with contemporaneous sites in Mesopotamia and the
Persian Gulf, the use of bangle in the Indus region appears to be quite
distinctive. It has not been possible to make a detailed comparative study of
archaeological materials from Mesopotamia, but the available published
materials and personal communications with archaeologists working in
Mesopotamia indicate that while bangles were made in a range of materials,
the variety and quantity of these ornaments were significantly less than what
is seen in the Indus sites.
Furthermore, in direct contrast to the Indus, valuable ornaments,
including carnelian, lapis lazuli, copper and gold were often buried with the
dead.
The distinct patterns of ornament use established during the first urban
development in the Harappan phase undoubtedly had an important effect
on later developments in South Asia. Excavations at early historic sites
throughout South Asia combined with the evidence from early Vedic and
epic literature indicate that bangles continued to play an important role in
rituals, sensuality and general social and economic identification.
From infancy, children wear specific types of bangles as amulets for health
and to enhance beauty. A young woman wears bangles during courtship,
and at marriage these bangles are replaced by different types of bangles to
symbolise her changed status. Throughout a woman’s life, bangles are worn
as ornaments and also to protect and preserve her family’s well-being. These
bangles are removed or broken at the death of her husband, and all valuable
ornaments are passed on to subsequent generations. Men often wear bangles
for physical protection in battle, for amuletic purposes, for defining status
and ethnic affiliation, and simply as ornaments.
While the use of bangles can differ from region to region, and definitely
fluctuated over time, it is apparent that bangles as an ornament style have an
extremely long and meaningful history in South Asia. Furthermore,
throughout history, the degree to which bangles have been used as socio-
ritual and economic symbols in South Asia appears to be significantly higher
than in other adjacent regions.

CONCLUSION
Through these various examples, I have tried to demonstrate how a
systematic analysis of ornaments can provide important new insights into
the character of the earliest urban society in South Asia. The methods used
to analyse these ornaments need to be refined and adapted to other objects,
and the initial interpretations need to be tested through comparative studies
at other Indus sites. Through such studies of the different components of the
ancient Indus tradition it is possible to break through the barriers of
understanding that have resulted from the absence of literary documents. As
we gradually understand more about the nature of this early urban
civilisation we can begin to define its relationships to contemporaneous
civilisations in West Asia and its contributions to later cultural
developments in this region of the world.
4

Texts on Stone
Understanding Aśoka’s Epigraph-Monuments
 UPINDER SINGH

T HIS ARTICLE MAKES FOUR INTERRELATED CONTENTIONS. The

first is that any interpretation of the dhamma of the Mauryan emperor
Aśoka (269–32 BCE) must accord importance to the manner in which the
dhamma message was transmitted to and received by his audience. The
second contention is that Aśoka’s edict-bearing rocks and pillars represent a
unique combination of inscriptional and monumental aspects and should
therefore be regarded as (for want of a more elegant epithet) ‘epigraph-
monuments’. Thirdly, it is argued that the sculptural and inscriptional
elements of the Aśokan edict-bearing rocks and pillars must be considered
as part of an integrated whole and that this would help in understanding the
nature of Aśoka’s dhamma. And the fourth point is that Aśoka’s epigraph-
monuments have a post-Mauryan history which illustrates how changes in
historical context may be accompanied by very dramatic transformations in
the significance and meaning of monuments.
 I
One of the consistent features of Aśokan studies has been the meticulous
analysis of the text of the inscriptions.1 Historians and epigraphists have
made detailed lexical and grammatical analyses, and have carefully
examined the meaning of specific words, phrases, and sentences of the
epigraphs. No one would maintain that such an exercise is inessential.
However, if we recognise the fact that the Aśokan inscriptions seek to make
a statement or a series of statements, we should also consider the possibility
that the manner in which that statement/those statements was/were read or
understood perhaps had less to do with the precise phraseology of the text
of the inscriptions than with certain other things.
In the third century BCE context of limited literacy, a very small section of
the audience that Aśoka was addressing through his inscriptions would have
been in a position to actually read his messages. In fact, the text of some of
the edicts could not have been read easily in entirety by even a literate
person. In the case of the pillars, the arrangement of the edicts in four
narrow columns (except for the Delhi-Topra pillar where the seventh edict
runs right across the circumference of the pillar) and deep, clear incising (as
is the case with the Delhi-Topra pillar) would have favoured the reader. On
the other hand, the six or seven (in the case of the Delhi-Topra pillar) edicts
of varying length were inscribed at an elevated height on the pillar face,
which meant that the writing of the edicts was not at immediate eye level,
and could only be read from a fair distance. In the case of the rock edicts as
well, the uppermost sections of the inscriptions may not have been easy to
read.2
It is against this background that we can take note of the various
references in Aśokan edicts directing that the edicts be orally propagated.3
Separate Rock Edict I (addressed to the mahāmattas of Tosali) and Jaugada
(addressed to the mahamattas of Samāpā) states that this edict (līpī) was to
be listened to on the day of the Tiṣya constellation and also on any other
suitable occasion).4 Separate Rock Edict II at the same places (addressed to
the kumāra and mahāmattas of Samāpā in the first case and the mahāmattas
of Samāpā in the second) advocates that this edict should be listened to on
every Catummāsī (Sanskrit, Caturmāsī) day, on the day of the Tiṣya
constellation, and in between.5 These references seem to suggest some sort
of ceremonial, congregational reading of the edicts on certain occasions,
either in select gatherings consisting of high-level Mauryan officials or in
larger gatherings in which such officials played a leading role. What is
significant is that the references in the Separate Rock Edicts are to groups of
people or individuals being exhorted to listen to the edicts.6
The message of the Aśokan edicts was also orally propagated by various
Mauryan officials—kumāras, yutas, lajūkas, padesikas, mahāmattas,
dhamma-mahāmattas, anta-mahāmattas, the pulisāni, and members of the
parisā. In most cases, these officials were entrusted with this task in addition
to their regular duties, while the dhamma-mahāmattas were supposed to
exclusively devote themselves to spreading dhamma. Of course, the arch-
propagator of Aśoka’s dhamma was Aśoka himself. And if we connect the
dhamma-yātās of Major Rock Edict VIII with the vivāsā of Minor Rock
Edict I (which refers to 256 nights having been spent on tour), we get an
image of a king who was propagating dhamma with an ardour that could
only have been to the detriment of his routine administrative duties. In all
these cases, whether transmitted by the king himself or by various Mauryan
officials, such renderings would no doubt have conveyed an amplified and
idiosyncratic version of the dhamma message. Aśoka’s inscriptional message
would have been accompanied by an oral commentary that would have
varied considerably in length, interpretation, and emphasis. It is more than
likely that the geographical range and effectiveness of such an oral
propagation would have surpassed that of Aśoka’s inscriptions.
In view of the factors of limited literacy, the practical problems involved
in reading some of the edicts, the evidence of an extensive network for the
oral transmission of Aśoka’s message, and also certain other factors
discussed in the succeeding paragraphs, there are good reasons to consider
the text of Aśokan inscriptions in conjunction with the surface on which
they were engraved and to view them together as ‘epigraph-monuments’.
How these epigraph-monuments were perceived and the impact that they
had depended not a little on things other than the words they bore. What
would have been of paramount significance for the person standing in front
of the epigraph-monument would have been a variety of other factors: the
exposure to the oral version of the message, and the symbolisms and
meanings conveyed by the sculptural embellishments, wherever these were
present.
Not all ancient Indian inscriptions qualify as epigraph-monuments. The
copper plate land grants, for instance, do not qualify as monuments of any
kind. Inscriptions on temple walls would also not count as epigraph-
monuments because the monument in itself (that is, the temple) has a more
or less established identity, meaning, and function of its own through its
architecture and the iconography of its sculpture. The inscriptions here are
information-conveying documentary adjuncts to the monument in
question. The Aśokan inscriptions, on the other hand, are different. The text
of the inscription is an intrinsic part of the monument, defining its
personality in a major way. Text and stone together go to form an epigraph-
monument.
 II
The most basic element of the Aśokan epigraph-monument was the surface
on which the inscriptions were engraved. It is recognised that the Rock
Edicts, Major and Minor, preceded the Pillar Edicts.7 The pillar had certain
obvious advantages over natural rock as the bearer of the royal edicts. It
allowed for a flexibility in the choice of the site of the inscriptions, and in
terms of architectural monumentality—the tall, slightly tapering, polished
pillar, especially the pillar with capital, its height compelling attention from a
distance—definitely scored over stark, rugged rock.
Much has been written about the ancestry of the Aśokan pillar. It has been
connected variously to the Vedic yūpa, the Indra dhvaja, an ancient pillar
cult; its symbolism linked to various cosmogonic myths.8 It is also
recognised that in certain cases, as is suggested by Pillar Edict VII and
Minor Rock Edict I, the Aśokan edicts may have been incised on pre-
existing pillars.9 However, delving into the ancestry of the Aśokan pillar,
although an important exercise, is useful only up to a certain point. A
symbol may be a part of a long cultural tradition and at the same time the
context and manner of its deployment may mark a major break with that
tradition. Even if we accept the fact that the pillar (of wood or stone)
antedates Aśoka, it must be emphasised that by having the pillar inscribed
with his edicts, Aśoka transformed the pillar into something new—a
didactic monument and an imperial symbol. This point is relevant to the
debate regarding the influence of Achaemenid art on Aśokan court art in
general and the Mauryan pillars in particular. There is plenty of evidence
from fairly early times of interaction of different kinds between West Asia
and the Indian subcontinent. There is also the evidence from various parts
of the ancient world of the employment of foreign artisans in the building of
monuments. Aśoka’s edicts reveal him to be a king who was well aware of
and in interaction with many of his contemporaries whose domains lay to
the west of his empire. In Rock Edict II, he speaks of providing medical
services for men and animals in the domain of Antiochus (II) and his
neighbours. Rock Edict V speaks of the activities of the dhamma-
mahāmattas among the dominions of the Yoṇas, Kambojas, Gandhāras,
Risṭikas/Raṭhikas, Pitiṇikas/Peteṇikas, and other peoples.10
In view of all this, certain similarities between the Mauryan and
Achaemenid pillars (for instance their polished nature and in their use of
certain common sculptural motifs such as the lotus) can, to a certain extent,
be seen either as a product of cultural interaction and influence or as a
reflection of the existence of a large cultural interaction sphere that
encompassed Iran and India. However if, as has been argued here, we look
at the Aśokan pillars not in purely sculptural or architectural but in broader
cultural terms, we have to concede that notwithstanding certain elements of
architectural similarity—in spite of the possibility that the very idea of
issuing the edicts as also certain aspects of their style may have been
inspired by Achaemenid example, in view of the architectural differences that
also exist11 and more importantly the nature of the message that they convey
through the totality of shaft, abacus, capital, inscriptions, and context—the
Aśokan pillars have a distinct identity of their own. This difference in
identity corresponds to two very different sorts of messages that they sought
to convey, and to two very different conceptions of kingship, the
Achaemenid and the Aśokan.12 One was martial and dynastic in nature; the
other was moral and didactic, and while not devoid of allusions of an
abstract kind to successors, is indifferent to matters of lineage and
predecessors.
Thus, although it has been argued in earlier paragraphs that there is more
to the Aśokan epigraph-monuments than just the text of the Aśokan
inscriptions, it is simultaneously being suggested that the text of the
inscription transformed the pillar, regardless of the indigenous or foreign
source of inspiration of its various parts, into something new. Or rather,
pillar and inscription together went to create a monument of unique cultural
meaning.13
 III
The choice of the decorative elements of the Aśokan pillars has to be seen as
an important part of his epigraph-monuments. These elements must have
been conceived of as being connected to and in harmony with the message
of the inscriptions. The animals crowning the pillars thus assume great
importance. The lion appears on the capitals of the Basarh, Lauriya-
Nandangarh and one of the Rampurva pillars, quadruple lions on the Sanchi
and Sarnath capitals, and the bull on the second pillar found at Rampurva.
Mention may also be made of the elephant capital (sans pillar) found at
Sankisa. That the symbolism of the animal figures was a very important part
of the imperial message is suggested by the uninscribed pillars at Basarh and
Rampurva. This indicates that the pillar and the surmounting animal, even
without an inscription, conveyed a certain message.
The cakra (wheel) of the Sarnath capital has generally been identified as a
Buddhist symbol, representing either the Buddha or the first sermon.
However, V. S. Agrawala has discussed in some detail the cakra as a symbol
of creation and time in Vedic literature.14 The floral and faunal motifs that
appear on the Aśokan abacii and capitals are the lion (single or quadruple),
bull, deer, elephant, horse, geese (haṁsas), the honeysuckle/ palmette; to
these we may add the inverted lotus that supports the abacus. On the
Rampurva lion capital, there are engravings (poor in quality) of peacocks
and fish between the belly of the crowning lion and the top of the abacus.15
The decorative elements noted by the Chinese pilgrims Fa-hsien and Hsüan-
tsang on Aśokan pillars were the lion, wheel, ox, horse, and elephant.
The sculptural elements associated with the Aśokan pillars have a long
and varied ancestry in ancient Indian religious traditions as also in the
archaeological record. The lotus, honeysuckle/palmette, and geese are
associated in various Indian traditions with water cosmology and fecundity.
The lion, horse, elephant, and bull are associated with the guardians of the
four quarters. S. P. Gupta suggests that the lion may have represented the
sun and royalty, while the animals on the abacus may have been associated
with the cardinal directions.16 Agrawala’s discussion of the rich symbolism
of the sculptural elements of the Aśokan pillars includes detailed references
to the frequent appearance of these animals, individually and in groups in
various contexts in Vedic, Paurāṇic, and Buddhist literature.17
Looking for evidence of the depiction of these animals in the early
archaeological record of the Indian subcontinent, we find that the bull,
elephant, horse, tiger/lion, and peacock are represented (along with a great
variety of other animals and birds) on seals and copper tablets, and among
figurines of terracotta, faience, and copper/bronze found in Harappan
contexts.18 In some cases, the way these animals are depicted on seals (for
instance, where they are associated with a ‘standard’ or ‘manger’) suggests
the likelihood of some sort of religious significance. Particular reference may
be made to the terracotta bust of a double-headed lion (or tiger?), tongue
protruding out of mouth, found at Harappa. Its kinship with the addorsed
Aśokan lions has been commented on by many scholars. Post-Harappan
chalcolithic and early iron-using cultures have yielded several terracottas of
bulls and elephants, and bone and ivory figurines of elephants, some of
which may belong to the Mauryan period.19 Thus, representations of the
bull (which appears most frequently), elephant, horse, and lion are attested
to in Harappan and post-Harappan (pre-Mauryan) cultural contexts.
Furthermore, all the floral and faunal motifs of the Aśokan abacii and
capitals—the lion, elephant, geese, horse, deer, bull, honeysuckle/palmette,
and more—appear on Mauryan, late-Mauryan, and post-Mauryan ring-
stones and disc-stones, very often in association with a goddess. The bull
and geese also appear in the stone sculpture of these periods. Terracotta
figurines of the time include elephants, horses, and bulls. The honeysuckle/
palmette and geese also appear as part of the sculptural embellishment on
the Vajrāsana at Bodh Gaya.20
It may be added that if we examine the motifs on early Indian punch-
marked coins, we find that the wheel, bull, and elephant appear frequently in
the Kosalan and Magadhan series (pre-Mauryan and Mauryan), the
elephant and wheel in the series of Aṅga, the lion and bull in the Avanti
series, the bull and elephant in the Śūrasena series, the bull in the Koṇkan
series, the deer, elephant, and bull in the Kaliṇga series, the bull and
elephant in the Pāṇḍyan series, and the wheel, lion, and elephant in early
(pre-Christian era) non-Pāṇḍyan south Indian coins. The horse makes its
appearance only on coins found in the south-east Afghan plateau.21 It may
be noted that while the motifs on the middle and late Magadhan coinage
(460–321 BCE) include the elephant, bull, bull-on-hill, fish, peacock-on-hill,
three-arched stūpa, hare-on-hill, frog, wheel, serpent, and tree-on-hill, out
of which Mitchiner identifies the peacock-on-hill and three-arched stūpa as
Mauryan emblems, the lion does not feature at all. This would lead to the
questioning of Irwin’s hypothesis that the Mauryas adopted the lion as the
new symbol of their sovereignty.22
Many scholars have suggested that the animals of the Aśokan pillars have
a specifically Buddhist symbolism. Radhakumud Mookerji suggests that the
elephant typifies the Conception, the bull as presiding over the Nativity, the
horse the Great Departure, and the lion the Buddha (the lion among the
Śākyas). A little more far-fetched is his suggestion that haṁsas pecking seed
on the abacus of the Rampurva lion capital represent the flock of which the
Buddha is the shepherd.23 The contradiction that seems to have been
overlooked is that while Mookerji argues in favour of a very definite
Buddhist symbolism for the sculptural elements of the Aśokan pillars, he
denies that the dhamma of the edicts has any specifically Buddhist content.
According to him, this dhamma was ‘completely cosmopolitan, capable of
universal application and acceptance as the sāra or essence of all religions’.24
There are other interpretations. According to Foucher, the bull represents
the traditional date of the Buddha’s birth, the full moon day of Vaiśākha.25 S.
P. Gupta suggests that apart from being the lion of the Śākyas, the Buddha
was also the bull and the musk elephant among great leaders, and a steed of
men.26 Gupta relates the Sarnath sculptural motifs to the inscription incised
on the pillar, connecting the symbolism of the capital with the chariot of
time, the eternal movement of the wheel of time. This symbolism, according
to him, belongs to the realm of Buddhist cosmology which drew on, but was
not identical with, Vedic cosmology. The lion, horse, and elephant have also
been connected with the Anotatta lake of Buddhist tradition.27
What do we make of this multiplicity of possibilities? A solution to the
problem may be provided by the animals on the capitals of pillars that bear
specifically Buddhist edicts. At Sarnath (the Sarnath pillar bears the Schism
Edict addressed to the Buddhist saṅgha) appear the lotus, bull, deer, horse,
quadruple addorsed lions, and the dharmacakra. The Sanchi pillar (which
also bears a version of the Schism Edict) shows the lotus,
honeysuckle/palmette, pecking geese, and the quadruple addorsed lions. A
broken lotiform capital was found in the vicinity of the Nigali-Sagar pillar
(which records Aśoka’s enlargement of a stūpa).28 The Rummindei pillar
(which records Aśoka’s visit to Lumbini, the birth place of the Buddha) is
fragmentary. If we argue for a unity in the inscriptional and sculptural
programme of the Sarnath and Sanchi pillars, we would have to see the
sculptural motifs, especially the crowning lions, as having a symbolism
inspired by Buddhism. We would also, then, note that the lion appears (this
time singly) on the capitals of the Lauriya-Nandangarh and Rampurva
pillars which bear the six pillar edicts. Extending the argument, one could
then see the appearance of the lion on these pillars as evidence of the
Buddhist nature of the dhamma that the edicts on these pillars speak of.29
This would tie in with the post-Mauryan evidence. All the sculptural
motifs associated with the Aśokan pillars (especially the elephant, lion,
dharmacakra, and lotus) are a conspicuous part of the sculptural
programme of the early Buddhist sites of Sanchi and Bharhut. Apart from
the appearance of these motifs singly in various places at these sites, we may
note that monolithic pillars surmounted by an inverted lotus and elephant
capital appear at two places among the Bharhut reliefs.30 More numerous
and striking instances can be cited from the Sanchi reliefs. Pillars (the shaft
divided into three parts by flutes) surmounted by an inverted lotus and
crowned by quadruple addorsed lions supporting a dharmacakra appear at
prominent places as on the pillars of the northern and southern gateways of
Stūpa nos. 1, 2, and 3 at Sanchi, and are strongly reminiscent of the Sanchi
and Sarnath pillars of Aśoka.31 The affinities between these and the Aśokan
pillars were noted by Sir John Marshall.32 If does not in fact seem too far-
fetched an idea to see in the depiction of such pillars at early Buddhist sites
the memory of Aśokan pillars. And these pillars could have only made their
way to occupy such an important place in the sculptural programme of these
sites if at the time they were executed, they were seen as part of a Buddhist
inheritance. Whether this had to do with the original purport of the Aśokan
pillars, or whether this was a post-facto appropriation can be debated, but
the former possibility seems to be a strong one.
 IV
Having looked at various aspects of the sculptural elements of the Aśokan
pillars, and identified some evidence of their Buddhist affiliations, let us turn
to the text of the Aśokan inscriptions. Certain references in these
inscriptions seem to suggest that the dhamma that they speak about should
not to be identified in a restricted sense with the Buddhist dhamma, and
that it was something broader in scope. This is the impression we get from
the account of the duties of the dhamma-mahāmattas specifying that they
are to occupy themselves with all sects (the term used is pāsaṇḍa) (Rock
Edict V). Rock Edict XII which begins with Aśoka’s statement that he
honours all sects, both ascetics and householders, with gifts and honours of
all kinds, goes on to express his interest in advancing the essentials of all
sects (it is something much more substantial and positive than mere
‘toleration’ that he is expressing). It exhorts men to respect one another’s
dhamma, to maintain reciprocal concord, and mentions various officers
concerned with all this. The fruit of this, the inscription goes on to say, is the
promotion of one’s own sect and glory to dhamma. Pillar Edict VII describes
the dhamma-mahāmattas as occupied with all sects; it states that some are
to busy themselves with affairs of the saṅgha, others with the Brāhmanas,
Ājīvikas, Nirgranthas, as well as with other sects.
On the other hand, apart from the well-established evidence of Aśoka’s
personal faith in Buddhism, there is the very evident Buddhist element in
the dhamma he sought to propagate through his edicts. (We are not talking
here about Aśoka’s motives or strategies, conscious or unconscious, in
issuing his edicts.33) The fact that the first edict in the Major Rock Edict
series has ahiṁsā as its theme is surely not insignificant, as is the recurrence
of this theme in numerous other edicts (such as Rock Edicts I, III, IV, Pillar
Edict V, and the Minor Edict found at Kandahar34). The close parallels
between some of the elements of the dhamma of the edicts (as enunciated,
for instance, in Rock Edicts III, IX, XI, Pillar Edict VII, and the Minor Rock
Edict found at Brahmagiri and other places35) with the upāsaka dhamma as
enunciated in the Sigālavāda sūtta has been noted by many scholars. So has
the fact that the Minor Rock Edict at Bhabru, in which Aśoka addresses the
sangha, lists six Buddhist texts as texts of dhamma. Bühler traces the terms
dhamma-dāna and dhammanuggaha in line 7 of Rock Edict IX to the
Itivuttaka, a Buddhist text.36 Then again, Rock Edict VIII connects Aśoka’s
visit to Bodh Gaya with the beginning of his series of dhamma-yātās. The
Ahraura version of Minor Rock Edict I contains in its last line an
amplification that does not occur elsewhere: it refers to this declaration
having been made while Aśoka was on a tour of pilgrimage for 256 nights
‘since the relics of the Buddha ascended the platform’,37 possibly referring to
Aśoka’s installation of certain relics of the Buddha (in a stūpa?). If these
tours can be connected with the dhamma-yātās referred to in Rock Edict
VIII, we get further evidence for the Buddhist aspect of Aśoka’s dhamma.
Minor Rock Edict I in all its versions also connects Aśoka’s increasing
ardour for Buddhism with the dhamma proclamations. Buddhist resonances
can be detected in Rock Edict VI where the king speaks of the debt he owes
to all living beings and his concern for the welfare of the whole world, in the
virtues (self-control and purity of mind) he prescribes in Rock Edict VII.
There is another piece of evidence which suggests that the dichotomy
some historians have suggested between Aśoka’s personal faith in Buddhism
and the dhamma he propagated through his edicts is somewhat artificial.
This evidence comes from the Allahabad-Kosam pillar. Here, we find a set of
the six pillar edicts as well as two other inscriptions. One is known as the
Queen’s Edict and talks about certain gifts made by Queen Kāruvākī. The
other is the Schism Edict in which Aśoka addresses the Buddhist saṅgha and
warns monks and nuns of the consequences of creating dissension within
the Order. This edict is also found at Sanchi and Sarnath, where it occurs on
its own. But on the Allahabad pillar, it occurs along with the pillar edicts.
This indicates that we are not looking here at two separately placed and
unconnected sorts of messages addressed to two different audiences, one
consisting of the general populace and the other of the Buddhist
community.
The hypothesis that the dhamma of Aśoka’s edicts is essentially a form of
the Buddhist upāsaka dhamma is strengthened when we take note of the
following examples where a close connection can be established between
animal motif and text of inscription. At Girnar, beneath Rock Edict VIII, to
its right, is a fragmentary line referring to the white elephant bringing
happiness to the whole world. It is well known that the series of rock edicts
at Kalsi are accompanied by a carving on the north face of the figure of an
elephant and the word gajatame (the best elephant).38 At Dhauli, the
elephant appears again, at the top of the rock, and the word seto (the white
one) which appears after the sixth edict obviously refers to this elephant.39
Hultzsch points out that like the Kalsi and Dhauli rocks, the one at Girnar
must also have borne the figure of an elephant. He suggests that this figure
(along with certain parts of Rock Edicts V and XIII that lay adjacent to it)
was probably destroyed during the construction of a causeway for pilgrims
from Junagadh to Girnar.40
The elephant appears in ancient Indian art, literature, and mythology in
various contexts, including as the god Indra’s vāhana, Airavata, as also as
part of the Gaja-Lakṣmī motif. Yet, while we may be able to read manifold
symbolic meanings into a certain motif, in a particular context, one of these
many possible meanings may appear to be more relevant than the others.
The testimony of the Girnar, Kalsi, and Dhauli rocks clearly connects the
dhamma of the inscriptions with the elephant, specifically described on the
Dhauli rock as white in colour. It appears quite reasonable to conclude that
there is a relationship between the elephant and the text of the inscription,
and that relationship and connection involves an understanding of the
elephant here as a Buddhist symbol, the symbol of the bodhisattva who
entered the womb of Māyā in the form of a white elephant. Surely the
appearance of the elephant as the prologue/ epilogue to the set of Major
Rock Edicts at these sites is a point of great significance and further
substantiates the pronounced Buddhist aspect of Aśoka’s dhamma.41
The evidence of the sculptural and inscriptional aspects of the Aśokan
epigraph-monuments is in fact quite consistent. In both cases, there is a
definite Buddhist core, but this subtly blends into a broader, non-sectarian
fabric of cultural expression. It may be pointed out that the difficulty in
categorising and classifying the sculptural elements is to a large extent
because of the fact that the Mauryan period marks the first stage in the
evolution of a specifically Buddhist sculptural vocabulary in stone. As for
the dhamma teaching of the Aśokan inscriptions, it is clearly a version of the
Buddhist upāsaka dhamma with its emphasis on a set of societal ethics,
modified marginally by the emperor to elicit a broader social appeal.
The Buddhist connections of the Aśokan pillars and their message also
come out strongly on a careful reading of the information documented by
Alexander Cunningham and others in the Archaeological Survey reports
that describe the details of the circumstances and surroundings of the finds
of many of the Aśokan pillars. Even in cases where the site in question seems
to have had an important position on the trade routes of the time, we
encounter again and again the additional evidence of ancient Buddhist
remains. Again, there are two ways of understanding this congruity. Did the
pillars mark already existing Buddhist establishments or did such
establishments emerge there subsequently? Either way, the Buddhist
connection of sites where the Aśokan pillars have been found is a point that
seems to have been quietly buried in most writings on the subject.42 In a
highly influential work on Aśoka, Romila Thapar has argued that there need
be no direct connection between the personal religious beliefs and public
policies of a statesman.43 In Aśoka’s case, there is just too much evidence to
deny that such a connection did in fact exist.
And yet, there seems to have been more to the Aśokan inscriptions than
their dhamma message with its pronounced Buddhist tone. When
considered as epigraph-monuments, the inscriptions and the surface on
which they were engraved can be interpreted as having an additional
political dimension. They can be seen as politico-religious, imperial
monuments, similar in this sense to the much later imperial temples of the
early medieval period, conveying, apart from their dhamma-teachings of
Buddhist inspiration, a certain image of kingship. If considered as such, the
shift from rock-face to pillar can be seen as a new, more refined, step in the
articulation and projection of a more grandiloquent articulation of royal
power.44 We can also correlate with this the comparatively static tone of the
pillar edicts as compared with the major rock edicts, the absence in the
former of the sensitivity, the disarming frankness, the eagerness to examine
all possibly related aspects of dhamma, the spirit of exploration and the
patient attempt to explain and convince—which are all aspects of the Major
Rock Edicts. The pillar (inscribed or otherwise) was chosen as the symbol of
Aśoka’s imperial power and dominion; and the fact that it was not the figure
but the voice of the emperor that was rendered in stone was perhaps an even
more eloquent testimony to his authority over his dominion, his people, and
the higher sphere of dhamma. While the king as custodian of dharma is a
very familiar figure in ancient Indian literature and inscriptions, the manner
in which this relationship was expressed by Aśoka remains unique.
Aśoka himself must have been the architect of this articulation and
projection of his image as king, as his inscriptions resound very clearly with
his personal voice in contrast to the impersonal, often tediously repetitive
and formulaic panegyric penned by later court poets. What is important is
that as far as articulations of royal power go, this is the only one of its kind.
It has been suggested above that the message conveyed by Aśoka’s
inscriptions was neither entirely encompassed nor circumscribed by the text
of his edicts. It may be added that although the edicts are not devoid of
chronological references, Aśoka’s epigraph-monuments, their sculptural
symbolism, the content and phraseology of their text and message all strive
powerfully to rise above the mundane and particular, succeeding in
combining simplicity and directness with profundity and grandeur. Their
success in achieving this delicate balance is what sets them and their author
apart from others.
 V
Several centuries later, the accounts left by the Chinese pilgrims Fa-hsien
and Hsüan-tsang indicate that stūpas and pillars built by Wu-yau (Chinese
for ‘sorrowful’, that is, Aśoka) had become places of pilgrimage associated
with miraculous happenings, past and contemporary. It may be noted that
the Chinese pilgrims invariably describe the Aśokan pillars as surmounted
by animal capitals (most often a lion). They also invariably refer to the fact
that these pillars bore inscriptions. Of course, they do not necessarily give an
accurate rendering of the text of these inscriptions, which they would not
have been able to decipher themselves. Further, although the description of
the pillars does not in certain respects match what we know about the
Aśokan pillars—as in the occasional references to carvings of the Buddha
executed in niches around their contours—there are rather specific
references to the unusually smooth and lustrous surface of the pillars. The
Chinese travellers connect the sites of Aśokan pillars with sites of stūpas and
vihāras built by Aśoka, and go on to connect these places with events in the
life of the Buddha and certain Jātaka stories.
In Chapter XVII of Fa-hsien’s early-fifth-century account, there is a
description of Sankasya (which can be identified with Sankisa, where an
elephant capital has been found).45 Fa-hsien describes this as the place
where the Buddha came down after ascending to the Trayastriṁśa heaven
and preaching dhamma to his mother for three months. It is here, he tells us,
that the Buddha, followed by a multitude of devas, descended three flights of
steps made of precious substances. After he had descended, all the steps but
seven disappeared into the ground. Many years later, Fa-hsien reports,
Aśoka had the place dug in order to find the steps. While the steps were
found, their limits were not, and the king’s reverence for the Buddha was
increased by this miracle. He built a vihāra at the spot and behind it erected
a stone pillar, about 50 cubits high, with a lion on top. Fa-hsien tells us that
some teachers of another doctrine once had a dispute over this place with
the Buddhist śramaṇas. Both sides agreed that if the spot did indeed
rightfully belong to the śramaṇas, this should be proved by some miracle.
Scarcely had these words been spoken, we are told, when the lion
surmounting the pillar gave a great roar, settling the matter quite firmly.
The seventh century account of Hsüan-tsang also refers to several Aśokan
pilIars. Among these was a stone pillar at Kapitha, about 70 ft high, shining
as if with moisture, surmounted by a lion. The pillar had certain figures
carved on it, which manifested themselves to different people in different
ways depending on their moral character.46 Hsüan-tsang speaks of a pillar at
Lumbini, close to a stūpa built here. The pillar was surmounted by a horse,
which due to the contrivance of a wicked nāga, was broken off in the middle
and fell to the ground.47 Hsüan-tsang tells us of a pillar near a vihāra and
stūpa at Kusinagara, which bore an inscription recording the Buddha’s
parinibbāna.48 To the north-west of Varanasi, he says he saw a stone pillar in
front of a vihāra. He describes the pillar as 100 ft high, bright and shining as
a mirror, its surface as glistening and smooth as ice.49 In the deer park at
Varanasi, where the Buddha set in motion the Wheel of the Law after his
Enlightenment, Hsüan-tsang tells us that he saw an Aśokan stūpa and pillar.
The pillar was some 70 ft high. Its stone glistened and sparkled like light. All
those who prayed fervently before it saw, according to their petitions, good
or bad signs.50
While Aśoka may have been quite gratified by the reverence inspired by
his pillars, at least in the hearts of the Chinese pilgrims, he would probably
have been amazed (or more likely, extremely annoyed) if he knew what
would happen to some of his edict-bearing pillars in subsequent centuries.
Hsüan-tsang refers to a fallen pillar at Lumbini, hinting at what was to be
eventually the fate of many an Aśokan pillar, several of which were uprooted
or damaged. Three of the pillars (the Allahabad-Kosam, Delhi-Topra, and
Delhi-Meerut) were removed from their original place and installed
elsewhere, and most of them saw the addition of new inscriptions. The later
inscriptions can be divided into three categories— royal inscriptions,
inscriptions of ordinary people, and the graffiti left variously by ancient,
medieval, and modern pilgrims and/or tourists. The shorter inscriptions
were held by Cunningham to be unimportant, and are likewise ignored by
historians. But they do form a part of the history of the Aśokan pillars.
 The Delhi-Topra pillar bears three twelfth-century inscriptions of the
Chauhān Rajput king Vīsaladeva alias Vigraharāja (IV).51 In these the King
boasts of his conquests from the Himalayas to the Vindhyas and of his
exterminating the mlecchas and restoring Āryavarta to what its name
signified. The pillar that had, almost a thousand years earlier, proclaimed the
dhamma of Aśoka, and symbolised the majesty of the Mauryan empire, now
proclaimed the conquests and dominion of an early medieval Rajput king.
But the story does not end here.
We are extremely fortunate in having an account of how in the fourteenth
century, the Delhi-Topra and Delhi-Meerut pillars were transported from
their original sites to Delhi with great care and effort and installed there
during the time of Sul ān Fīrūz Shāh Tughluq. Shams Sirāj ‘Afīf tells in his
chronicle Tārikh-i-Fīrūz Shāhi of how the Sul ān noticed the two monolithic
columns in the course of his military campaigns and how he had one (which
came to be called the Mīnār-i-Zarīn or Golden Column) transported to and
erected in his palace at Fīrūzabad near the Jam‘a Masjid, and the other
placed in his Kūshk-i-shikār or Hunting Palace.
‘Afīf writes that these two pillars dated from the time of the Pāṇḍavas, and
recounts a tradition that they had been the walking sticks of the Pāṇḍava
hero Bhīma. We are told that the Sul ān was filled with admiration when he
saw the monolithic pillars and decided to move them to Delhi as trophies.
The description of the moving of the Delhi-Topra (Mīnār-i-Zarīn) pillar tells
of orders issued to people living in and around Topra village, and also to
soldiers, directing them to assemble near the column, bringing with them
various implements and materials, including large quantities of silk cotton
from the silk cotton tree. When the earth around the column was carefully
removed, the column fell on the bed of silk cotton that had been prepared
for it. It was then encased in reeds and raw hides and carefully moved onto a
specially constructed carriage with forty-two wheels. Men pulled at the
ropes attached to the wheels, and the pillar slowly made its way in this
manner to the banks of the Yamuna. Here the Sul ān came in person to meet
it. The pillar was then heaved onto several boats tied together and taken by
river to its new home in Delhi. At Fīrūzabad, it was raised to its present
position in the palace complex with great ingenuity, skill, and labour. The
writing on the Aśokan pillars could no longer be read by Fīrūz Shāh’s time.
‘Afīf tells us that some Brahmins gave the interpretation that it contained a
prophecy that no one would be able to remove this pillar from its place till
the time of a great king named Sul ān Fīrūz. The reading was too
unimaginative and contrived to convince anyone, even a king who might
have found it flattering. Some years later, when the Mongol Timur invaded
India, ‘Afīf tells us that he visited the monuments of former kings, and was
very impressed by these remarkable pillars.52
In addition to the seven Aśokan edicts and the three inscriptions of
Vīsaladeva, the Delhi-Topra pillar also bears certain other inscriptions. The
first is a short sixteenth-century Sanskrit inscription containing the date and
the name of the writer, a person named Ama. The longer inscription (also of
the sixteenth century) in mixed Sanskrit and Persian refers to Sultān
Ibrāhim (who may be identified with Ibrahim Lodi) and states that the
inscription was written by Viyadā, son of Sāyana. There is also a reference to
a person named Bahādur Khān, who may have been a nobleman during the
reign of the Lodi king.53 There are also several short inscriptions on the
pillar, three of which refer to persons who were goldsmiths, and one which
consists of the name of one Bhayankar Nāth Jogi, who also left his name in
an inscription in the Barabar caves.54
As for the post-Aśokan inscriptions on the Delhi-Meerut pillar, there are
three short early medieval Sanskrit epigraphs.55 The first refers to a person
named Vīrapāla, son of a ruler of Sind. The second inscription, also of the
fourteenth century, refers to the writer as a goldsmith named Mala Sāhā. The
third inscription belongs to the sixteenth century. It says that it was written
by a person named Amara and contains certain words the meaning of which
is unclear.
The medieval accounts of the moving of the Aśokan pillars to Delhi and
the medieval inscriptions they bear reflect an important aspect of certain
historical monuments—the redefinition of what they mean and what they
signify, over time. In this particular case, Fīrūz Shāh Tughluq was impressed
by two monolithic pillars made and erected over a thousand years before his
time by a Mauryan emperor. He had no idea what the original purpose of
the pillars may have been; neither did he have any idea of what the writing
on the pillars said. He brought the pillars to Delhi and had them set up in
two prominent places. When the pillars were fixed in their new locales, ‘Afīf
tells us that certain additions were made—ornamental friezes of black and
white stone were added to the top, and a guilded cupola was placed on the
pinnacle. Changes in the physical appearance of the Aśokan pillars were
accompanied by continuity and change in the meaning of the monuments
[emphasis added]. The pillars had been politico-religious imperial
monuments, symbols of a king’s power and dominion in Aśoka’s time. Fīrūz
Shāh Tughluq took them over as monuments signifying his own power and
dominion. But, in their newer medieval setting, the edict-bearing pillars
were no longer conveyors of the dhamma message of the earlier king.
The Kosam pillar bears no Aśokan inscription, but does have a series of
small inscriptions, some damaged and illegible, which range (judging from
their script) from the Gupta period to modern times.56 There is also a
medieval inscription dated in the reign of the Mughal emperor Akbar. This
was read by Cunningham as: ‘Mogul Pādshāh Akbar Pādshāh Ghazi’. This is
followed by a three-line inscription of a soni or goldsmith. Below this is a
longer inscription dated in Samvat 1621 (that is, 1564 CE) giving the
genealogy of a family of goldsmiths, beginning with a person named Anand
Ram Das, who, we are told, died at Kosambipura. Cunningham records that
the Kosam pillar was locally know by the names of ‘Rām- kī-Charrī’ or the
alternative ‘Bhīm-kā-gadā’.
The Allahabad-Kosam pillar is located inside Akbar’s fort.57 The fact that
it originally stood at Kauśāmbī is suggested by the Schism Edict inscribed
on its surface which refers to the mahamattas of Kauśāmbī. The pillar carries
the inscriptions of three kings, separated from each other by vast stretches of
time. In addition to the Aśokan inscriptions, it bears the famous Allahabad
praśasti of Samudragupta, which details the military successes and
personality of the Gupta king Samudragupta. There is also on this pillar an
inscription of the time of Jahāngīr, giving the genealogy of this Mughal
emperor. The fact that the pillar carries the inscriptions of three emperors of
different ages makes it a remarkable and unique monument. In spite of the
fact that the meaning of the original Aśokan inscription was probably
unreadable by the time of Samudragupta and the text of both Aśoka’s and
Samudragupta’s inscriptions was certainly unreadable by the time of
Jahāngīr, the pillars—because of their impressive architectural attributes, or
location, or importance ascribed to them by local tradition—were seen as
worthy bearers of an imperial statement.
The Allahabad-Kosam pillar also has a number of short inscriptions of
names of people, ranging from the beginning of the Christian era to modern
times. Cunningham very ingeniously reconstructed the various stages of the
writing on the pillar, which part of it was executed while the pillar was
standing, and which part inscribed when it lay flat on the ground. He
suggested at least three phases of the re-erection of the pillar, one during the
time of Samudragupta, and the second during the time of Jahāngīr. The
pillar was again thrown down in the early nineteenth century by one
General Garstin, because it got in the way of the new line of rampart he was
having built to strengthen the Allahabad fort. Then, in 1838, it was re-
installed by one Captain Edward Smith. There was a plan to crown the pillar
with a newly-constructed capital shaped in the form of a flower. But,
Cunningham tells us, since local memory insisted that it was a lion that had
originally been there, a committee of the Asiatic Society suggested that the
new capital should be on the lines of the Bakhira and Lauriya pillars.
Cunningham observes that the new capital designed by Captain Smith was
‘a signal failure’, and that the lion looked like ‘a stuffed poodle stuck on the
top of an inverted flower pot’.58
The Lauriya-Nandangarh pillar bears, in addition to Aśoka’s six pillar
edicts, a Persian inscription of the time of Aurangzēb. Cunningham read its
text as follows: ‘Mahi-ud-din Muhammad Aurangzib Pādshāh Alamgir
Ghāzi Sanh 1071’.59 Another inscription (this one in the Nagari script) refers
cryptically to a king (nṛpa) Amara Singha, son of King Nārayāṇa. There is
also on this pillar an inscription in English which simply reads: ‘Reuben
Burrow, 1792’. Reuben Burrow evidently also paid a visit in the same year to
Basarh, where he left his name neatly carved on the Aśokan pillar there.
To the inscriptions which definitely may be ascribed to Aśoka Maurya
may be added the fragments of what may be Aśokan pillars found at Hissar
(Haryana) and Fatehabad (45 kilometres north-west of Hissar).60 The
former is located in front of a mosque (Lāṭ-kī-masjid) built by Fīrūz Shāh
Tughluq. The ‘Aśokan’ fragment of yellow sandstone forms the lowest part of
a composite pillar, the rest of which consists of three registers of red
sandstone separated by white stone discs. The Fatehabad fragment is also
part of a composite pillar. It is located in the middle of a praying ground
with a late Mughal wall. The pillar bears on its surface a Persian inscription
which gives genealogical information about Fīrūz Shāh Tughluq. Falk has
pointed out that this is the longest historiographical inscription of the Delhi
Sul ānate. He also gives evidence that suggests that the Hissar and Fatehabad
fragments were two one-third parts of one and the same Aśokan pillar. Once
again we witness Fīrūz Shāh Tughluq’s penchant for Aśokan pillars resulting
in a striking change in their context and meaning.
One of the recurrent themes that emerges in modern perceptions of the
Aśokan pillars is their being described by local inhabitants as walking sticks
of Bhīma (or some similar thing). This interpretation goes back at least to
the fourteenth century as it is referred to in the account of Shams Siraj ‘Afīf
cited above. It was still going strong in the late nineteenth century, because
Cunningham and others had also encountered it repeatedly. The Rampurva
pillar was discovered by Carlleyle when local informants told him about a
stone called ‘Bhīm’s Lāṭ’ at the place.61 The Lauriya-Nandangarh pillar was
locally known as the ‘Bhīm-Mari-ka-Lāṭ’, the Basarh pillar as ‘Bhīm-Sen-kā-
lāṭ’ or ‘Bhīm-sen-kā-ḍaṇḍā’.62
Another interesting aspect of the redefinition of the meaning of the
Aśokan pillars is the fact that late-nineteenth-century and twentieth-century
accounts frequently refer to entire Aśokan pillars or their fragments being
actively worshipped in various parts of India as Śiva liṅgas. Cunningham
tells us that the Lauriya-Araraj pillar was locally known as laur, that is,
phallus.63 He also narrates how, while he was engrossed in copying the
inscriptions on the Lauriya-Nandangarh pillar, a man, two women, and a
child set up a small flag before the pillar, placed offerings before it, and
offered worship. The pillar was being worshipped as a liṅgam.64
Thus, the Aśokan pillars are not only important sources of information
for various aspects of the Mauryan period. They have a post-Mauryan
history that is no less interesting or important. As they came to be
appropriated by different ages, these pillars were redefined and reinterpreted
in a dramatically varied manner. This redefinition and reinterpretation
involved an interface between the ancient and medieval past and between
history and local traditions.
5

Social Background of Ancient Indian

Terracottas
 DEVANGANA DESAI

T ERRACOTTAS OR BAKED CLAY FIGURINES, WHICH APPEAR to

be spontaneous handiwork of the simple folk, can be produced on a large
scale only when certain social conditions are fulfilled. Their making involves
‘technical equipment at an artisan level which goes beyond the independent
domestic pattern of production’ and is possible in a society which is well-
settled and differentiated into specialised professions including that of
potters. It is undertaken when there is a demand arising from
institutionalised religious cults which require the use of clay figurines as
votive offerings, magical charms or household deities, and from a public
who would buy secular figurines and plaques for decoration of homes, toys
for children and varied other purposes. Such religious and secular demands
for terracottas can be efficiently fulfilled on a large scale in urban societies.
The large-scale terracotta production is closely related to the process of
urbanisation and the development of markets. The establishment and
growth of well-organised markets, the most essential feature of urban
economy, is a stimulus to the expansion of industries in general, including
that of terracottas.1
The history of terracottas in India shows that large-scale production is
associated with the rise and flourishing of urban cultures. The early settled
cultures of Baluchistan and Afghanistan began to produce terracottas in
their later stages when they were evolving from village to town level and
when religion was institutionalised. Large-scale production of terracottas is
seen in the urban culture of the Indus valley. It is significant to note that
comparatively very few terracottas are found from the village settlements of
the post-Harappan Chalcolithic cultures, although these were river valley
sites with easy availability of clay. Similarly, there is a conspicuously scanty
production of terracottas among communities of the Ochre Coloured
Pottery (OCP) and the Painted Grey Ware (PGW).2 Terracottas appear again
with the second urbanisation in northern India from circa 600 BCE and are
seen on a mass scale from 200 BCE to 300 CE when the urbanisation was at its
peak. But they became less popular in the post-Gupta period with the decay
of towns and decline in urban culture. Terracottas of ancient India are the
art and ritual objects of an urban culture that flourished from about 600 BCE
to CE 600.
From circa 600 BCE there was a ferment in the social climate of northern
India which was a part of the ferment of the Iron Age in the civilised world
from the Mediterranean to the Jaxartes in Central Asia. The regular use of
iron technology ushered in a process of gradual urbanisation in the Gangā-
Yamunā valley which reached its peak in the period after 200 BCE. The use of
iron implements in agriculture and in the clearance of dense forests made
agricultural production on a large scale possible which was hitherto
unknown in earlier periods. With extensive cultivation a surplus was created
which was large enough to sustain non-agricultural groups. Specialisation in
industries and crafts became possible, which gave an impetus to trade.
Transport was made more efficient because of better equipment. Coins
began to appear for the first time in the levels associated with the Northern
Black Polished Ware (NBPW) (550–200 BCE) in many sites of Bihar and
Uttar Pradesh.3 The use of coins, in its turn, brought revolutionary changes
in social and economic life. Even small artisans and workers could save for
future use and amusements. Manufacture of cheap goods for popular
consumption became profitable with money economy. Commodity
production developed considerably with coinage.
Specialisation and concentration of industry and crafts along with the
growing volume of trade made the centres of production grow into busy
towns. Vārāṇasi is such an example which became one of the six cities in
Buddha’s time by its specialisation in textile industry. We find that gradually
there were jalapaṭṭanas or towns where goods were brought by boats,
sthalapaṭṭanas where goods were brought by roads, droṇamukhas where
goods were brought both by sea and land, putṭabhedanas or mercantile
centres where seals of packed goods were broken,paṇyapaṭanas or market
towns and rājadhaānīs or capital towns.4 There were two long-distance trade
routes connecting a number of towns: Uttarāpatha or northern route from
Rajgir to Pushkalāvatī (Charsada), and a southern route from Rajgir and
Śrāvastī to Pratiṣṭhāna. There were also four minor routes from Rajgir to
Kapilavāstu, Śrāvastī, Campā, and to Kaliṅga.5 Thus, there was a network of
towns and markets in the early historic period.
The urban life created multifarious demands in response to which various
industries and crafts sprang up. The Jātakas mention eighteen crafts which
kept on increasing in number. The Dīgha Nikāya speaks of twenty-eight
crafts and the Mahāvastu speaks of thirty-six kinds of workers. The
Milindapañha, towards the beginning of the Christian era, enumerates
seventy-five occupations and describes a busy town market.6 Luxury
industries such as perfumery, ivory-carving, fine textile industry, jewellery-
making, etc. considerably developed in the post-Maurya period.
Pottery was also recognised in ancient India as one of the eighteen
industries and was undertaken on a large scale. The deluxe pottery generally
known as the NBPW, produced by refined iron technology, was traded from
Magadha (south Bihar) where major centres of its production were located,
to faraway places like Taxila. That it had reached several central Indian sites
such as Kayatha, Ujjain and Besnagar by circa 500 BCE can be evidenced by
carbon-14 dates.7 Trade in this fine pottery must have made potters rich.
Besides this luxury ware, potters made red, black and grey wares as seen
from excavations in Bihar sites. They also earned from the making of ring-
wells which were used for sanitary arrangement in towns. An early Jain text,
Uvāsagadasāo, speaks of one potter Saddalaputta who owned 500 workshops
and had a large number of potters working under him. The Mahāummagga
Jātaka refers to a potter’s hireling, which suggests that there was a thriving
business of potters. From the Buddhist literature we know that potters were
well organised into guilds, and that there were villages (suburbs) of potters
outside the gates of cities. Thus, there was a demand for the potter’s craft in
the urban culture of the period.
In understanding early Indian terracottas, three points deserve special
consideration. The first is that the terracottas that we have come across so far
have been found from urban sites and have as their clientele the urban
people whose needs, interests and tastes are reflected in them. Secondly, they
were produced on a large scale because there was a market for them. There
was a demand for terracotta figurines both at religious and secular levels,
which could be met because of the existence of a well-organised class of
potters. Thirdly, tribal and peasant cults and rituals are represented in
terracottas, but their treatment is marked by urban sophistication.
We shall study here in a broad outline the social history of terracottas.
Terracotta-making is viewed as a social process in the context of the art-
loving public and the potter-artist of the period. The period circa 600 BCE-
600 CE has been divided into five sub-periods on the basis of socio-
economic and political changes. These are (i) circa 600–320 BCE; (ii) circa
320–200 BCE; (iii) circa 200 BCE 50 CE; (iv) circa CE 50–300, and (v) circa CE
300–600.

CIRCA 600–320 BCE

We find a few specimens of terracottas of this period before the consciously

organised movement in art begins under the Mauryas. As there is a
controversy among art historians about the existence of terracottas in the
pre-Maurya period, it is best to give data from archaeological excavations of
sites. The pre-Maurya terracottas are found mainly from sites in Bihar.
Among the earliest are terracottas from Buxar, ancient Charitravana on the
Gaṅgā, in Shahabad district. Here four animal figurines and two human
figurines have been found from pre-NBPW and pre-coin levels of period I.8
From Pāṭaliputra (Patna), which was Magadhan capital from the fifth
century BCE, we have human figurines and nāgas (snakes) from its pre-coin,
earliest level of period I.9 From Vaiśāli (Basarh in Muzaffarpur district), the
Licchavi capital, we get terracotta nāgas in its pre-coin and pre-NBPW level
of period I.10 From Champā, 5 km west of Bhagalpur, we get a terracotta
nāga (female) from its early NBPW level, dated by the excavator to sixth-
fifth century BCE. 11 From Chirand in Saran district and Sonpur in Gaya
district nāga figurines have been found along with NBPW ware.12 At
Chirand, this phase is pre-structural.
The nāga figurines found at Vaiśālī, Pāṭaliputra, Champā, Sonpur and
Chirand are stylised and show an iconographic form consisting of hood of a
snake and human body. There are punched circlets for breasts and navel and
incised lines to mark parts of the body. The figurines were possibly used as
votive offerings or as images in the nāga cult. So powerful must have been
this cult in this region and other parts of the Middle Gaṅgā Valley, that it
survived for more than 700 years, as we can see terracottas of the same form
and style up to the Kuṣāṇa period.
The Mother Goddess cult was yet to develop in its full-fledged form. The
female figurines are not widely prevalent in this period. The Buxar female
figurine has a round face, wide eyes which are incised, punched earlobes,
prominent breasts and perforations over the head for decoration.13 This type
is also seen from Pāṭaliputra (Patna Museum, Arch. No. 4330 from
Kumrahar) and Kauśāmbī.14 That this type of female was associated with a
Mother Goddess cult is certain, as we have another example of such a hand-
made female from Buxar who is sitting on a stool and holding a child in her
left hand (Patna Museum, Arch. No. 6303). There is another type of animal-
faced hand-made figurine from Pāṭaliputra.15 The excavators identify the
figurines as males, but the punch-marks for breasts and navel indicate that
they are females.
All the terracotta figurines of this period are hand-made. They are not
decorative or artistic in intent and are not produced on a large scale. The
society had not yet reached the stage of commodity production. The process
of urbanisation had just begun, but it had not made much of an impact on
the cultural life of the people. It will be noticed that figurines from Buxar,
Pāṭaliputra and Vaiśālī have been obtained from pre-coin levels. Earlier
traditions and Chalcolithic technology continued along with iron tools and
NBPW in several sites of Bihar.16 This probably explains the archaic nature
of terracottas in this period.

CIRCA 320–200 BCE

The tribal and peasant cults continue in the Maurya period, but for the first
time we see terracottas as objects of art. Secular and artistic terracottas are
mainly confined to Magadha (south Bihar) which had become the most
important economic and political centre of India under the Maurya empire.
As Professor Niharranjan Ray has shown, court influence was the
predominating feature of the Maurya art in stone, commissioned by King
Asoka for the propagation of Buddhism. The tall and elegant columns with
their naturalistic perfection in animal sculpture reflect austerity and dignity
of the royal court, and also the love of foreign art, Achemenian and
Hellenistic.17 The Mauryas had established diplomatic and economic ties
with West Asiatic and Hellenistic countries. So great was the influx of
foreigners in the Maurya capital, Pāṭaliputra, that, as Megasthenes notes, the
municipal board had to set up a special committee to look after them.
Foreign influence is reflected in terracottas of Pāṭaliputra, (Bulandibagh,
Patna) ‘with definitely Hellenistic heads and faces, Hellenistic modelling
and, in a few cases, also Hellenistic drapery’.18 They remind us of the
Tanagra terracottas of the fourth-third century BCE. 19 The emphasis on
drapery and elongated physique is a common feature of both Pāṭaliputra and
Tanagra figurines.
These dancing girls and graceful ladies of the Maurya capital breathe the
air of urban sophistication with their smooth and sensitive modelling,
dynamic dancing movements, delicate neck, and rich, probably starched,
drapery. They reflect the tastes and interests of the royal family and the
śreṣṭhin class who had acquired a dominant position in the social life of the
period. As these terracottas are large and fragile, they could not have been
preserved in small houses or huts of the common folk. They were meant for
decorating large palaces of royal and aristocratic families, and were most
probably produced by rājakumbhakāras or royal potters of whom Pāṇini
speaks about.20 These potters did not produce for the market but produced
for the individual needs of the royal family. The Cullakaseṭṭhi Jātaka also
refers to a king’s potter. There were artisans attached to wealthy sreṣṭhins
also.21
We may mention here that the Maurya date of the Patna dancing figures
is sometimes questioned by art historians. But the find of a similar figure
from Sonpur (Gaya district) in its period II along with NBPW shows its
Maurya date, while figurines of period III at Sonpur show Śuṅga
characteristics.22
Sophistication in its highest form is seen in terracottas of Buxar mainly
representing fashionable ladies with elongated foreign facial features and
elaborate hairstyles in numerous varieties.23 The foliage and floral designs in
their ornaments and headdresses suggest that they could be females
associated with vegetation and agricultural rites. Pāṇini informs us that the
Śālabhaṅjikā festival was peculiar to the eastern people. From the Buddhist
literature it is known that the people plucked śāla flowers and celebrated the
festival amidst merrymaking. We learn from the literature (Meghadūta) of a
later period that in a similar vegetation festival of Aśokabhaṅjikā, aśoka
leaves were worn on the ears by the girl, who performed this rite of
rejuvenating the tree. The Buxar figurines with floral and foliage motifs seem
to be a sophisticated version of a fertility maiden.
Along with these urbanised figurines, we also see a continuation of
archaic forms in terracottas. The nāga figurines seen in the earlier period
continue to appear in several sites of the Middle Gaṅgā Valley. They retain
the linear abstraction and stylised form of the earlier period. The nāga
worship on the new and full moon days was known to Kauṭilya. He says that
persons knowing Atharvavedic magic should perform auspicious rites in
honour of snakes to ward off their danger (Arthaśāstra, IV–3).
The cult of the Mother Goddess was widely prevalent in the Maurya
period the evidence of which is found not only from terracotta figurines but
also from numerous stone discs representing the goddess, sometimes with
her partner.24 Such discs are found in a number of sites from Bihar to
Punjab and Taxila. The largest number of Mother Goddess terracotta
figurines are found at Mathurā, which, in the Maurya period, became an
important terracotta-making centre outside Magadha. Here, Mother
Goddess figurines have been found along with NBPW and coins in the early
and middle phases of period II.25 Their archaic and vital form is altogether
different in idiom from Magadhan figurines. The inspiration is clearly
religious. There is no emphasis on drapery as in Patna figurines. The
Mathurā figurines represent a fat steatopygous female, with prominent
breasts, an applique collar or necklace, an applique girdle often punch-
marked or incised, and a well-marked navel. The earlier examples have
animal or bird-like face which is replaced by moulded oval face in later
examples. Rosettes and large round earrings are peculiar decorations of this
type of goddesses.
 The Mathurā types of the goddess are seen in several sites of Uttar
Pradesh. V. S. Agarwala has pointed out the similarity of Ahicchatrā
figurines (stratum VIII) with those of Mathurā.26 A figurine showing a
broad-jewelled girdle from Hastinapur (mid-level of period III) is similar to
one of the types of the goddess of Mathurā.27 Sarairnohana near Vārāṇasī
(period I-B with NBPW)28 and Vaiśālī (early level of period II)29 also show a
Mathurā type of female figure with applique necklace and girdle and incised
eyes. Similarly, Masaon30 near Vārānasī has, along with NBPW and punch-
marked coins, female figures with pinched up nose, big neck and prominent
breasts, which have resemblance to Mathurā figurines. Similar figurines have
been found at Kauśāmbī.31
We may mention here that Charsada (Puṣkalāvatī) in Gandhāra on the
junction of several trade routes, has typical Mother Goddess figurines of
smooth and levigated clay excavated from the levels dated between circa
250–100 BCE.32 But these figurines with small breasts, applique eyes
resembling inverted shells, and without an indication of navel are different
from female figurine of Mathurā and Pāṭaliputra.
A popular theme in the Maurya period is animals: elephants, horses,
rams, bulls, dogs, etc. Animal figurines of Pāṭaliputra, Buxar, Vaiśālī,
Mathurā, Hastināpur, and Atranjikhera are meticulously decorated as in
cases of ceremonial animals. The decoration consists of the prevalent
applique and technique and punch-marked and incised circlets. It is
interesting to see that the animal figurines of Mathurā have rosette motifs
similar to those seen on the Mother Goddess figurines. Elephants of
Hastināpur and rams of Vaiśālī are decorated with vegetation motifs
reminding us of the same motifs on the Buxar female figurines. It is possible
that animal terracotta figurines were connected with Mother Goddess
worship. The contemporary stone-discs show animals such as lion, elephant,
horse, ram, and deer surrounding the goddess. Animal figurines may have
also been used as toys. The Vessantara Jātaka refers to toys consisting of
horses, bulls and elephants.
Thus we see that in the Maurya period, secular and artistic figurines have
been mainly confined to Magadha whereas other sites of the period have
generally ritual figurines. The monopoly of the Magadhan state in economic
and political fields and the resultant social inequality are reflected in
terracottas of the period. The royal society and the upper class of Magadha
commissioned special potters to produce artistic terracottas resembling
Hellenistic figurines, whereas the country outside Magadha was producing
crude terracottas for cults and rituals. Magadhan (Patna, Buxar) terracottas
have been found from Kauśāmbī, Bhita and Rajghat, but the reverse is not
the case.33 This indicates that the Magadhan terracottas were popular in
other regions and possibly also that they were traded. Candraketugarh and
Tamluk in Bengal have some specimens of this period, but these also show
influence of the Magadhan style. The economic and political centre of
gravity was located in Magadha which dominated the cultural scene in this
period.

CIRCA 200 BCE–50 CE

There is a spectacular and unprecedented growth of terracotta industry in

this period as a consequence of the rapid progress of urbanisation.
Numerous centres between Bengal and the Punjab came into prominence.
The rule of the Śuṅgas which followed the imperial Mauryas lasted for a
short time. After Puṣyamitra Śuṅga’s death in the middle of the second
century BCE, we see the rise of numerous small kingdoms of Kauśāmbī,
Kośala, Mathurā, Pañcāla, etc. having independent coinage systems. The
political and economic dominance of the Magadhan state broke down and
the surplus was widely distributed among other regions. Even private
individuals amassed considerable wealth.
The vigorous trade and the growth of numerous industries and crafts
accelerated the pace of urbanisation. The literature of the period records a
larger number of industries and crafts than those noted in the Arthaśāstra.34
The growth of trade and industries considerably improved the position of
the mercantile and artisan groups. The ruler’s control over artisans was
weakened by the increase in the number of guilds. The heads of artisan
guilds, the jeṭhakas were accorded high social status by the king. Artisans
worked as independent persons and earned cash income in contrast to dāsas
and karmakāras who worked for getting food and clothes.35 The payment in
cash naturally served as an incentive to the artisan to work hard and
produce enough to cater to the demands of the market.
The demand also increased as the purchasing power of the people
increased. In the Maurya period it was mainly the royal house and wealthy
śreṣṭhins that indulged in art and luxuries. In this period there were many
other sections of people who were in a position to afford luxuries. There was
now a class of nāgarakas (citizens) who had leisure and money to afford art.
Even common people including potters, ivory-carvers, masons and weavers
had money to spend on decoration of religious monuments, as evidenced
from the inscriptions of Bharhut and Sā chī. In respect of terracottas, there
were more people to buy them, not only for religious but also for decorative
purposes.
The potter in this period had to meet the mass demand for terracottas.
This he was able to do by adopting the mould technique in the manufacture
of terracottas. The adoption of this technique gave such a great boost to the
production that terracotta-making rose to the level of an industry.
Terracottas became commodities for the market. The direct control of the
royal patron over the potter was weakened in this period. An impersonal
relationship of the potter and his clientele developed through the agency of
the market.
The use of the mould was also made possible because the public was not
particularly interested in the individual creation of an artist. The
reproduction of the same subject and composition did not affect the market
value of terracottas.
The adoption of moulds in terracotta production facilitated the depiction
of narrative and illustrative themes in which the public were interested. The
round figures of the earlier period gave place to compositions in flat reliefs,
consisting of several figures. Exhaustive details of dress, ornaments, coiffure,
etc. were minutely worked out in terracotta plaques as in ivory-carving.
Even the stone art reflects the same mental attitude and the same formal
principles of style. The gap between the stone and terracotta art almost
disappears in this period. One can see much similarity, for instance, between
the stone reliefs of Bharhut and the clay art of Balirajgarh in Bihar.36 There
is, however, a difference in the thematic content of these arts, as their
functions are different.
Religious themes are represented in a large number of terracotta plaques,
but their treatment is influenced by urban sophistication. Almost all the
excavated sites have yielded female figurines associated with the Mother
Goddess and vegetation cults. But the goddesses are decorative female
figurines in urbanised garbs. Instead of the primitive archaic goddesses of
the Maurya period, we have beautifully decked and richly bejewelled female
figures with elaborate hairstyles. Nudity is often shown through their
transparent fine apparel. Fine textiles and a large variety of rich ornaments
which we have seen in this period were possible because of the growth of
industries of luxury. So many ornaments are rarely represented in art of
other periods.
Śrī-Lakṣmī, the goddess of prosperity, wealth and beauty became a
popular goddess with the rise of commercial and mercantile communities.
Her representations are seen in numerous terracottas of Tamluk,
Candraketugarh. Bangarh, Harinarayanpur, Lauriya-Nandangarh,
Kauśāmbī, Mathurā, Awra, etc. Another auspicious goddess Vasudhārā,
probably related to Anāhitā, who holds two fishes in her hand, became
popular in Rajasthan, Uttar Pradesh and Bengal. The cult of the goddess
with five āyudhas or weapons in heads who may be called Pa cacūdā for
convenience, was widely prevalent from Bengal to Punjab (Rupar). The
representation of her partner in plaques of Ahicchatrā and Candraketugarh
suggests that she was a fertility goddess whose ceremonial or symbolic
marriage was celebrated for the general welfare of the community and for
agricultural and vegetation fertility,37 as in the West Asiatic cultures. The
male partner of the goddess holds a musical instrument reminding us of the
role of music in fertility rites. There is another vegetation goddess at
Mathurā, Kauśāmbī, Ahicchatrā, Rajghat, Rupar, etc. who has elaborate
headdress with palm fronds or corn sheaves. Auspicious and fertility
symbolisms are shown in association with goddesses of this period. They
touch their own mekhalā (girdle) or earring, hold flower in one hand, hold
two fishes, wear auspicious symbols in headdress and have symbolic amulets
and ornaments.
 Several male deities also make their appearance, not only as the consorts
of the goddess but also as independent figures. We see plaques representing
Sūrya (Chirand, Candraketugarh), winged deity (Vaiśālī, Kauśāmbī,
Balirajgarh, Candraketugarh, Tamluk) and squatting yakṣa figures probably
representing Kubera (Candraketugarh, Kauśāmbī, etc.). There are also male
figures standing on wheeled ram, elephant and horse.

Figure 5.1: A goddess with partner, Ahichhatra, about second century bce
Source: Archeological Survey of India (ASI).
The cults of these gods and goddesses were associated with festive
gatherings (samājas) in city gardens (nagaropavana), which become a
favourite subject with the terracotta public. In urban atmosphere, remote
from the actual relation with nature and growth of plants and corn, the
celebration of these fertility festivals tended to get secularised and
sensualised. They were treated as krīḍas (sports), to use the word from the
Kāmasūtra, for such fertility festivities. The garden party scenes of Mathurā
breathe secular air. Similarly, a plaque from Kauśāmbī showing a richly
dressed couple sitting on a chair manifests urban tastes and vision in its
depiction and seems to be associated with the sophisticated nāgaraka class.38
The ritual significance is retained by showing the woman touching her own
earring which is a symbolic gesture associated with fertility goddesses such
as Śrī.
The taste of the nāgaraka class is reflected in some of the plaques showing
goṣṭhi or cultural parties (Kauśāmbī), wrestling, animal fights (Kauśāmbī ),
palace scenes, ladies decorating themselves (Kauśāmbī, Mathurā, Rajghat), a
well-dressed nāgaraka with a parrot in hand, a nāgaraka with elephant
(Candraketugarh), a child writing an alphabet (Haryana), Kāmaśāstrīya
scenes (Chandraketugarh, Tamluk, Bhita, Kauśāmbī), love-making couples
(Kauśāmbī, Sankisa, Mathurā, Ahicchatrā), udayana- Vāsavadattā theme
(Kauśāmbī), etc. The vast array of themes is baffling. The potter-artists of the
period have succeeded in presenting the dynamic quality of popular culture
in terracottas.
Figure 5.2: A garden party with dance and music, Mathurā, c. second
century bce
Source: National Museum, New Delhi.

CIRCA 50–300 CE

Sātavāhana terracottas
Active trade with Rome and the consequent commercial prosperity in
southern regions created a favourable situation for the manufacture of
terracottas during the early centuries of the Christian era. We see terracottas
(kaolin or rich white clay figures) in a number of towns such as Kondapur,
Yellesvaram, Nagarjunakonda and Chebrolu in Andhra Pradesh, Sannathi in
Mysore, and Ter, Paithan, Kolhapur and Nevasa in Maharashtra.
From the Periplus of the Erythraean Sea, we get detailed information
about the maritime trade between India and the Roman empire and the
number of ports and inland market towns which had come up as a result of
this trade. The discovery of the Roman coins of the first century ce in south
India and the well-known reference of Pliny of the drainage of gold from
Rome to India indicate the favourable balance of trade and the resultant
wealth and prosperity.
It is in this period of urbanisation in the south that we get one of the
earliest inscriptions referring to a guild of potters. This inscription from a
cave at Nasik records that money was deposited with a guild of potters for
the benefit of the Buddhist sect residing in the monastery.39 This means that
the potters who flourished in this period, were well-organised in a guild and
that money was invested with the guild as in other commercial activities.
Besides the fine pottery with red lustrous polish, potters also made beautiful
bangles, amulets and beads, as a large number of these have been found
from Sātavāhana sites. Kondapur and Maski were known for terracotta bead
industry.40
Figure 5.3: A couple from Kauśāmbī, c. second century bce
Source: ASI.
Making of terracotta figurines with kaolin was also an industry which
catered to the tastes and interests of the wealthy mercantile community. This
is evident in the thematic content of terracottas, most of which is secular.
Among the themes most popular, are joyous aristocratic men and women
(whose heads are mostly found), wearing rich ornaments with neatly
arranged coiffure. Another theme representing an aristocratic couple riding
a horse reminds us of a similar motif on the capitals of the contemporary
caves of Karle and Kanheri, though however there are stylistic differences.
There are figurines representing a male riding a horse, children, gaṇas, and
animals such as horses, rams, lions and buffaloes. These dainty figurines
could not have been used as toys as they are too delicate for children’s play.
They seem to be decorative pieces.
The Sātavāhana terracottas have a totally different idiom from that of the
north Indian terracottas. The difference is not so much due to the adoption
of the double mould technique in the south as the practice of double mould
was also prevalent at many places in the north such as Mathurā, Ahicchatrā,
Purana Qila, etc. It is rather the dainty forms and delicate workmanship of
Satavahana figurines and the representation of a distinct ethnic type of
features and physiognomy which distinguishes them from those of northern
India.
The cultic figurines showing a nude goddess sitting with legs apart are
Roman-Egyptian in inspiration and are found at Ter, Nevasa, Yellesvaram,
Nagarjunakonda, etc.41 That this cult was spread among the aristocracy is
evident from an inscription at Nagarjunakonda which mentions a queen
who was avidhavā and jīvaputā (with her husband and sons alive) in
connection with this type of figurine. The image seems to have been
dedicated by the queen as an offering for the fulfilment of certain desires.42

Kuṣāṇa terracottas
Under the Kuṣāṇa, India was materially a prosperous country having trade
links with Rome and the Hellenistic west. A century before and after the
Christian era, there were several invasions from the nomadic tribes—the
Śakas, Parthians and Kuṣāṇa as a result of events in Central Asia. The period
witnessed the acculturation process when foreign religious cults, motifs and
forms were gradually being assimilated into the cultural pattern of India.
Terracottas vividly document the interesting results of the mixing of
cultures.
Two distinct trends are noticeable in the Kuṣāṇa terracottas. One is the
continuation of the old tradition of moulded plaques as we saw in the Śuṇga
or post-Maurya period. The other is the emergence of the new trend due to
the cultural changes brought about by the influx of new races.
 The older tradition of terracotta reliefs is widely seen at Candraketugarh,
Tamluk, Rajghat, Kauśāmbī, Bhita and Mathurā. There is a further
improvement in the formal quality. The flattened reliefs of the earlier period
give place to reliefs in depth. The themes reflect the tastes of the affluent
nāgaraka class. We see a well-dressed gay nāgaraka in a dancing pose
(Candraketugarh), erotic plaques (Chandraketugarh, Bhita), a woman with
parrot on the right hand (Kauśāmbī), a woman holding flowers (Mathurā), a
salabhanjika under a tree (Kauśāmbī, Rajghat), a drunken woman helped by
her paramour or husband (Ahicchatrā), etc. Some of the subjects are seen in
Mathurā stone art.
The other trend in terracottas represents influence of the new races which
had infiltrated into India. This second group of terracottas is mainly of two
types, one produced by double moulds and the other modelled completely
by hand. Seated gaṇas and yakṣas, grotesque figures, musicians and riders on
horseback were produced by the double mould technique which gives depth
and roundness to figures. However, the figures are often crudely produced,
unlike the Sātavāhana terracottas of about the same period. They remind us
of the Seleucian terracottas of the Parthian period.43
The completely hand-modelled figures were more widely prevalent and
represented both religious and secular demands of the new public. The
newly rising Brahmanical cults of Śāivism, Vaiṣṇavism and Śāktism as well
as Buddhism were supported by the foreign rulers. Religion was ‘humanised
and emotionalised’ as a result of the development of bhakti (devotion).
There was a demand for images of gods and goddesses for household
worship. Several Hindu gods and goddesses received their shape and form in
this period. Ekamukhī Śiva liṅgas are seen in large numbers at various sites
such as Mathurā, Rajghat, Kauśāmbī, Bhita, etc. Balarāma, Viṣṇu,
Kārttikeya, Gaja-Lakṣmī, Sapta Mātṛkās and Devī Mahiṣāsuramardini were
seen both in terracotta and stone. Some of the sites of Bengal such as
Bangarh produced powerful Mother Goddesses with archaic features.
Buddha and Bodhisattva images appeared for the first time. Mathurā and
Hastinapura have good terracotta Bodhisattvas.
Some of the foreign cults were assimilated into Indian religious practices.
Parthian influence is seen in terracotta votive tanks enshrining a Mother
Goddess surrounded by birds and musicians.44 These votive tanks have been
found from various sites such as Ahicchatrā, Hastinapura and Taxila
(Sirkap). The cult of the nude goddess having Roman-Egyptian influence is
represented in clay plaques of Bhita, Kauśāmbī and Jhusi.
The public of the period were interested in portraiture. From the
dramatist Bhāsa who flourished in this period we learn about the rājakula
housing statues of ancestors. Kuṣāna royal portraits in stone are well-
known. In terracottas, we see a number of heads (with tenons) of both men
and women with expressive facial features. Sites of Uttar Pradesh have
yielded better figurines compared to those of Bihar, as the centre of
economic and political activity now shifted to the north. Terracotta heads
from Mathurā, Kauśāmbī, Bhita, Masaon, Atranjikhera, Ahicchatrā, etc.
reveal the artistic talent of the potter. ‘Faces of female figures are
characterized either by a smile or at least an animation noticeable on the
cheek-bones set below a pair of flat wide-open eyes.’45 Some of the heads
represent distinctly foreign ethnic features. But there are a number of
crudely made heads having heavy and harsh facial features. One wonders
whether they were meant to be portraits.
The appearance of a large number of crude and coarse terracottas is
somewhat surprising, as this was a period of great prosperity. Numerous
gold and copper coins were issued by the Kuṣāṇa rulers. Even the
indigenous dynasties issued copper coins. As Prof. R. S. Sharma says, ‘This
would suggest that perhaps in no other period had money economy
penetrated so deeply into the life of the common people of the towns and
suburbs as during this period, a development which fits well with the growth
of arts and crafts and the country’s flourishing trade with the Roman
empire.’46 Urbanisation was at its peak in this period. Most of the excavated
sites show large brick structures and baked tiles for roofing and flooring.47
Probably potters were preoccupied with the making of bricks, tiles and pots
(red ware) at the cost of the artistic quality of terracottas. It seems that new
forms of images which were emerging in this period required a certain
degree of experimentation, and hence hand-modelling was resorted to for
the purpose. When the potter had to meet the large demand for terracottas
raised by the increase in the purchasing power of the people as well as the
increase in population, he was forced to pay attention to the production
rather than to the quality of terracottas. However, he could afford to pay
special attention to images and secular figures had he been paid more for
these. Some of the portrait figures or heads representing noblemen and
aristocrats are well-executed, especially at Kauśāmbī, Bhita, Ahicchatrā and
Mathurā. It is for the first time that one comes across the powerful intensity
in the facial expression of Kuṣāṇa terracottas. It is not the ornamental
quality but the massive and sturdy forms and expressive facial features that
distinguish Kuṣāṇa terracottas from those of other periods.

Figure 5.4: A female head from Mathurā, Kuṣāna period

Source: ASI.

CIRCA 300–600 CE

This is a period of transition when on the one hand there was a tremendous
prosperity on account of industries and trade, and on the other hand were
appearing signs of decline leading towards feudal conditions. Along with the
urban mercantile class which was prosperous and ranked high in social
circles, a new class of landed aristocrats was coming up towards the end of
the period on account of numerous land grants of the Gupta rulers and their
feudatories. Land, rather than cash was becoming the medium of grants and
rewards both in religious and secular spheres. Land grants to Brāhmaṇas
transferring revenue and surrendering police and administrative functions
to the donee weakened the position of the state and led towards partial
feudalisation.48
In this period of transition, when the upper classes—both the old
commercial class and the newly emerging landlords—had amassed wealth,
there was an overall development in the arts of the period—both visual and
literary. Terracotta art reached the highest technical perfection and
refinement in the Gupta period. A study of the Gupta terracottas (preserved
in the Bharat Kala Bhavan, Vārānasī) has shown that they have superfine
texture and do not have admixture of rice-husk in their clay. There is also an
absence of air-holes and grit in the clay.49 Again, for baking large-sized
terracottas such as the Gaṅgā and Yamunā of Ahicchatrā, there were large
cylindrical pits for special kilns ten to twelve feet in depth.50 Lepya- karma,
terracotta manufacture, was an extant technical term in the artistic
vocabulary of the period.51 Perfection in terracotta and other arts was
accompanied by the general improvement in techniques of metal working.
Iron tools of this period show considerable improvement. Knowledge in
iron technology had attained a high standard as seen from the Mehrauli iron
pillar of Candragupta. That the position of the artisan class had improved
can be inferred from the fact that they had to pay taxes now.52
Two distinct types of terracottas are seen in the Gupta period. The first
consists of the moulded and modelled (heads of) figurines and small
plaques used by individuals for household decoration or religious purpose.
The second consists of the large-sized figures and plaques used for
decorating temples and monasteries.
The first group seems to have catered to the tastes of the urban class, the
nāgarakas. In this group, most of the themes are secular, barring a few which
are religious, such as Naigameśa and Hara-Gaurī. We come across beautiful
and neatly produced heads of aristocratic men and women with varieties of
fashions in the coiffure which one can associate with the nāgarakas of
Vātsyayāna who flourished in this period. They are found from various sites
such as Vaiśālī, Rajghat, Bhitari, Bhita, Kauśāmbī, Jhusi, Śrāvastī,
Ahicchatrā, Mathurā, etc. At Bhita, marks of paint are still visible on
terracotta heads which indicate that they were often painted. The small
reliefs found from various sites consist of mithunas (Śrāvastī, Bhita, Rajghat,
Mathurā), kinnara-mithuna (Ahicchatrā), erotic scenes (Candraketugarh),
lady with pitcher (Pāṭaliputra-Kumrahar, Masaon), mother and child
(Vaiśālī, Bhita, Ahicchatrā), śālabha jikā (Pāṭaliputra- Kumrahar), etc.
Gaṇas and horse-riders are also found from numerous sites.
On the other hand, we have large-sized figures and plaques associated
with Brahmanical temples which were increasingly patronised by the newly
emerging feudal class and which reflect the Smārta-Paurāṇic ideology and
outlook. It is significant that the themes of epics which were known since at
least a thousand years, now received sudden prominence in visual arts.
Kṛṣna themes are seen in stone art from the Kuṣāṇa period and more widely
from the Gupta period in terracottas of Rajasthan (Rangamahal). The
Rāmāyaṇa scenes have been mainly found in terracottas from Śrāvastī in
Uttar Pradesh, Chausa and Ashad in Bihar and Barehat in Madhya Pradesh.
The Mahābhārata themes and Śaiva mythology are illustrated in terracottas
of Ahicchatrā. Along with religious subjects, there are also secular scenes on
plaques. For instance, the palace scene of Mathurā in which a woman is seen
in company probably of the Vidusaka. Most of the temples with terracotta
decoration are in ruins. We get only loose plaques which must have once
formed part of temple decoration. The extant temple of this period at
Bhitargaon near Kanpur gives us an idea of terracotta ornamentation of
temples.
Figure 5.5: A panel of the Bhitargaon temple
Source: ASI.
In comparison with the terracottas of the first group which are
sophisticated in outlook, the temple terracottas are more homely in
character, more intimately connected with the daily life of the people and
more playful in attitude. This is evident from scenes of the Bhitargaon
temple depicting; Viṣṇu Śeṣaśāyī or Gaṇeśa running away with the laḍḍus
while his brother is chasing him, or Śiva and Pārvatī in a family scene. The
same attitude pervades Rajasthan terracottas. On the other hand, spiritual
quality is captured in the Buddhas of Mirpur Khas in Sind and to some
extent of Devni Mori in Gujarat. Some of the large heads of Panna (Bengal)
and Ahicchatrā also reflect equipoise and self-composed disposition.
Decorating a temple with large figures and plaques must have involved
joint effort of several potters. A guild of potters rather than individual ones
must have been entrusted with the work of decorating temples and
monasteries. Again, the potters must have been in direct contact with their
patrons unlike the case of small, moulded terracottas which were produced
for the market.
 The output of the small mould-made terracottas dwindled in the post-
Gupta period with the decline of urban conditions. The rise of local units of
production led towards self-sufficiency and minimised the exchange of
goods and trade. Trade was also affected by the Hūna invasions. Decline of
trade led to decay of towns which were the centres of crafts and commerce.
The Mṛcchakaṭika, about the fifth century CE, portrays the decline in
prosperity of the merchant Carudatta. In this play there is a trader of
Pāṭaliputra who, because of the reverse of fortune had to migrate and to
work as a shampooer in Ujjain. From the study of the archaeological
material it has been shown that Pāṭaliputra, Vaiśālī, Chirand, Rajghat,
Kauśāmbī, Śrāvastī, Hastinapura, Mathurā, Purana Qila and several towns in
Haryana and the Punjab which had thrived in the Kuṣāṇa age began to
decline in the Gupta period and almost disappeared in post-Gupta times.53
With the decline of towns, terracotta art lost its main clientele, the
nāgarakas. Money economy was already shrinking in the Gupta period.
There were numerous gold coins but copper coins used in daily transactions
were less in number. Coins were becoming scarce from the times of
Harsavardhana. The purchasing power of the people declined. The class
which could purchase terracottas in the market had almost disappeared.
With the decline of towns, artisans dispersed to the countryside. They
became village servants and received their share in kind and not in cash.
Instead of producing for the market, they now produced for the village, the
landlord, the Brahmanical temple or the Buddhist monastery which also
possessed villages in large numbers and were feudal in character.54 With the
donation of villages, artisans came more and more under the domination of
donors and had to perform forced labour. Their mobility was restricted to
maintain the self-sufficiency of local units.
There was no scope tor specialised production for the market. Terracotta
was no longer the art of the people but was used by landed aristocrats and
kings to decorate religious buildings and their own palaces on auspicious
occasions like marriage, as recorded by Bāṇa in the Harṣacarita. Terracotta
acquired the character of elite art and was preserved in feudal headquarters
and religious centres such as Paharpur (Bengal), Antichak (Bihar), Akhnur
and Uskar (Kashmir). The main period of terracotta production was over
with the decline of urban culture.
 6

On Modes of Visual Narration in Early

Buddhist Art
 VIDYA DEHEJIA

N ARRATIVE IS A UNIVERSAL PHENOMENON WHICH, AS Roland

Barthes remarked, ‘is international, transhistorical, transcultural: it is
simply there, like life itself ’.1 Every nation has its corpus of narratives that
includes popular tales and fables, epics and romances, sacred myths and
legends. While the appreciation of written narrative is necessarily restricted
to the literate among the population, down the ages narrative has found two
other forms of expression that are accessible to all. The first is the tradition
of oral narration which may either take on a formal character as with the
heroic narration of epics, or assume the informal mode of story-telling. The
other readily accessible narrative tradition, prevalent in most cultures, is
visual presentation, whereby stories are communicated to an audience in
direct terms through two and three dimensional paintings and sculptures.
Stories revolve around actions, human or otherwise, that occur in space
and unfold in time. For the artists, the three major components of narrative
are the protagonists of a story, together with the elements of space and time.
Artists have to decide how to portray their actors, how to represent the
space or spaces in which the story occurs, and how to shape the time during
which the story unfolds. Artists may also arrange their story in a series of
more or less discrete episodes; if so, they must decide on the manner in
which they wish to compose these episodes within the visual field. The
sculptors or painters can also adopt a variety of modes to present the same
or similar narratives to their viewers.
Sometime after the death of the Buddha, which current research suggests
may have occurred around the year 400 BCE,2 a canonical body of literature
known as the Tipitaka or ‘Three Baskets’ began to be compiled, probably in
the language of the people, some form of Prakrit, and transmitted orally. At
a still later date, sacred texts also began to be composed in, or ‘translated’
into, the courtly refined language, Sanskrit. From this vast collection of
sacred literature, dozens of times the size of the Old and New Testaments
put together, artists concentrated on presenting two sets of legends. One was
the life of Prince Siddhartha who renounced the world and achieved
supreme wisdom, being known thenceforth as Buddha or ‘Enlightened One’.
The second was the long series3 of prior lives (jatakas) of the Buddha, in
which he took birth in a variety of forms, animal and human, until he was
ultimately born as Prince Siddhartha, achieved enlightenment, and put an
end to his cycle of rebirth. Between the first century bce and the fifth
century CE, the numerous Buddhist monastic establishments constructed
across India were extensively decorated with visual narratives. Most early
sites also portray a range of non-narrative themes, particularly semi-divine
figures of pan-Indian significance such as guardian deities, yakshas, yakshis,
and serpent beings, who were partially incorporated into Buddhist lore.
With the passage of time, the person of the Buddha began to be portrayed in
iconic and increasingly hierarchic form that emphasised a unique being, and
this alternative to narrative became the dominant mode in later Buddhist
art. My emphasis, however, is on narrative material, and this essay is devoted
to an analysis of the distinctive ways in which ancient Indian artists
presented Buddhist legends to their audience of monks, pilgrims, and
patrons.
There is universal awareness of the wide range of modes within the field
of literary narrative, and the distinctions that exist between a short story and
a saga, an epic and novel, a poem and drama, all of which may be used with
equal effect as alternative modes to narrate a story. A comparable range of
variation exists in visual narration, and an idea of the narrative modes
available to the artist may be gained from a study of two depictions of the
Buddha’s prior life as Prince Vessantara whose extraordinary acts of
generosity made his name a legend in Buddhist circles. Banished by his
father, the king, for giving away the kingdom’s prized rain-bestowing white
elephant, Vessantara donates all his wealth as he leaves for exile with his wife
Madri and their two children. On the way, he parts with the horses drawing
their chariot, and then the chariot itself, so that the family must proceed on
foot. In their peaceful forest abode, Vessantara gives away the two children,
and then his wife Madri. The final intervention of god Indra and the king
results in the happy reunion of the entire family. The story of Vessantara is
generally broken down by the artist into a series of episodes, prime among
them being the donation of the auspicious elephant, the banishment of
Vessantara and his family, his donation of horses, chariot, children, and wife,
and the final reunion in the palace. Fundamental to the meaning of the term
‘episode’ is the fact that it comprises several parts or events which may be
described in artistic terms as ‘scenes’. Thus the episode of the donation of the
children may be broken down by the artist into three (or more) scenes: the
evil brahmana making his demand, Vessantara granting his wish, and the
cane-wielding brahmana departing with the children.
Artists desirous of portraying this legend had a number of options
available. They could, for instance, decide in favour of brevity and use the
monoscenic mode of narration to tell the tale, utilising a space just 12 inches
square. In this mode, a single, easily identifiable scene, excerpted from one
of the episodes of the narrative, is presented to stimulate recognition of the
story in the viewer. A Bharhut artist portrayed the single scene of
Vessantara’s gift of the white state elephant. The viewer is presented with just
three figures, albeit unmistakable ones—the elephant, the brahmana who
receives the gift, and Prince Vessantara pouring water to ratify the gift.
Having stimulated viewers into identifying the tale, the artist leaves them to
narrate the story to themselves, and to remind themselves of the importance of
charity, the most valued of the Buddhist virtues (paramitas) [emphasis
added].4
on the other hand, artists could choose the expanded mode of continuous
narration, presenting their viewers with the entire series of episodes listed
earlier, leading up to the climactic conclusion of the tale. To do this, a Sanchi
artist spread out the narrative across two faces of a gateway architrave so
that the Vessantara story occupies a space some 22 feet in length and 2 feet
wide. Each episode consists of more than one scene, and in each scene the
figure of the protagonist is repeated. There are, however, no framing devices
to demarcate one scene from another, or one episode from the next, and the
story flows ‘continuously’ across the available space. To decipher the
presentation, to which we shall return later, the viewer must be aware that
the repetition of the figure of Vessantara indicates that we are seeing him in
different spaces at successive moments of time.
Figure 6.1: Monoscenic narrative, Vessantara Jataka. Bharhut coping, c. 100–
80 bce
Source: author.
The gulf that exists between these two modes of narration in the field of
Western art has been commented upon by scholars who see a major
distinction between the isolating, monoscenic mode of narration and the
expanded method of continuous narration.5 Others see the excerpted
method of monoscenic narrative as a degeneration of prior narrative cycles.6
In the Indian context, however, these two modes of narration exist side by
side, frequently on the same monument. Considerations of space and the
restrictions imposed thereby may have been partly responsible for the choice
of narrative mode, but it does not seem to have been the deciding factor. At
Sanchi, for instance, the architraves of the gateways offer a span 8 feet across
that would seem ideal for the method of continuous narration. Some Sanchi
artists chose to use this method, capitalising on the available space, but
others opted for the monoscenic mode, representing just a single scene from
a story to suggest the entire narrative. One such instance is the story of the
Buddha’s prior birth as six-tusked elephant chaddanta who lived in a lotus-
filled lake beside a giant banyan tree together with his two wives and his
devoted herd of elephants. When Chaddanta gave a lotus to his senior
queen, Maha-subhadda, he aroused the intense jealousy of the junior queen,
Chulla-subhadda who lay down to die determined to seek revenge in her
next birth. Reborn as the queen of Benares, she sends hunter Sonuttara to
bring her Chaddanta’s tusks and in his magnanimity, chaddanta aids the
hunter in sawing them off. ultimately, the queen dies of remorse upon
viewing Chaddanta’s tusks. In two of the three Sanchi architraves that depict
the story of Chaddanta, the artist chose to use the monoscenic mode of
narration, extending the single introductory scene of Chaddanta and his
horde sporting in the lotus pond to occupy the entire space available. The
third architrave, however, presents the viewer with four scenes strung
together as a continuous narrative. When one takes a popular tale such as
that of Prince Vessantara or elephant Chaddanta, and surveys the range of
early Buddhist art, we find it presented to the viewer in all of the modes of
visual narration discussed below. In addition, stories are often repeated at a
site using differing narrative modes. Did the artist as well as the viewer
regard these modes of narration as equally viable alternatives? Certainly,
choices were being made, and we need to probe further into the function
and position of the reliefs, the role of the patron, and other related
questions.
It is traditionally accepted that a narrative has two aspects—a story or
content that generally consists of a sequence of events, and the form or
expression which is the means by which the story is communicated and its
actions presented. The dichotomy between content and form has been
expressed by a range of scholars: there is the fabula (raw materials of a story)
and syuzhet (procedures used to convey them) of the Russian formalists; the
histoire and discours of the French structuralists; or the story and discourse
proposed by Seymour Chatman.7 While it might seem simplistic to stress
this dualism, it seems necessary to do so in the complete absence of the
study of form, or procedure or discourse in Indian visual narrative. Studies
of India’s painted or sculpted narratives are studies of stories, identifications
of legends, quests for textual versions that might have been followed,
analyses of the sequence of Buddhological cycles at various sites, or
refinements of typologies of Buddhist imagery. While such studies are
undoubtedly valuable, they do not aid an appreciation of the structure of
narrative by addressing the mode of presentation of a story, and it is this
aspect of discourse that is the focus of this essay. Discourse itself has two
important aspects: first is the selection of events from among those available
in a story, and secondly, the ordering of the chosen events.
I shall demonstrate in this essay that seven modes of visual narration were
used by the ancient artist to tell tales from Buddhist legend; some of these
modes are closely allied, while others are vastly different. Thus, the story of
Prince Vessantara is presented using the entire range of modes of narration;
apart from monoscenic and continuous narrative referred to above, the
artist made use of sequential, synoptic, and conflated modes, as also
‘narrative networks’, each of which we shall explore in detail. It is not my
intention to insist that the narrative mode utilised in the presentation of
every Buddhist tale may simply and easily be identified. While such is
indeed the case with a large number of narratives, a considerable corpus
remains that defies neat categorisation. Such narratives tend to combine
elements from varying modes which occasionally overlap and are not as
sharply differentiated as might appear from my analysis. Yet, the modes
possess a distinct validity. The emotional response triggered in the viewer-
cum-worshipper by the use of one mode as against another, as also the
stylistic judgments evoked in the scholar, are significant. In concluding this
essay, I shall turn to the relation between content and form which is, in sum,
the affect, leading to a consideration of the viewer’s reception and response.
Does the identification of seven modes of visual narration merely provide
modern scholars with a useful tool or were such modes ancient indigenous
distinctions? No texts survive that advise artists on modes of presenting
narratives; one might remark that such evidence is equally lacking in the
case of Egyptian, Roman, Etruscan or Christian narratives. On the other
hand, it is clear that artistic choices were being made. One Bharhut artist
adopted the monoscenic mode to present the Vessantara Jataka, while
another chose sequential narrative. Two Sanchi sculptors opted for the
monoscenic mode to portray the chaddanta story while a third decided on
continuous narrative. Once we acknowledge that ancient artists were
making deliberate decisions on the manner in which they wished to
communicate their message, it seems necessary to assume the existence of a
theoretical framework. The visual evidence provided by artistic material,
which is surely as revelatory as textual sources,8 speaks for itself; it seems
that we must admit the existence, in ancient India, of a sophisticated concept
of visual narration.

Precursors of Narrative in Stone: Painted Scrolls and Picture-

Showmen
The painted narrative scroll, known as the pata, took a vertical format in
India, and was generally of a size akin to the standard modern poster. To
judge from Buddhist legend, the pata appears to have been in use during the
time of the Buddha. Faced with the task of announcing to king Ajatashatru,
a devout follower and friend of the Buddha, that the Master had passed
away, and fearing a violent reaction, the venerable monk Mahakasyapa came
up with a novel way of gently breaking the news. He instructed minister
Varshaka to have a painted scroll prepared with depictions of the four great
miracles of the Buddha’s life—the birth, enlightenment, first sermon, and
the great decease. When it was held up before the monarch, Ajatashatru
would realise that the Buddha was no more. This legend of uncertain date,
but narrated in the Mulasarvastivadin Vinaya,9 is dramatically portrayed in
the murals at the Central Asian cave site of Kizil that, according to recent
evidence, belongs to the fourth and fifth centuries CE. 10
The earliest surviving visual narratives in the medium of stone were
produced at the commencement of the first century bce, but Buddhist
legends had been orally narrated for the prior four hundred years. In this
context, the tradition of patas and picture-showmen is of special relevance.
In a 1929 article, Coomaraswamy pointed out that the Sanskrit commentary,
Mahabhashya, written by the grammarian Patanjali around 140 bce,
contained evidence for the existence of picture-showmen. In elucidating the
use of the ‘historical present’, Patanjali gave as one example the display of
paintings representing the slaying of the evil king Kamsa, a climactic event
in the story of the youthful life of Hindu god Krishna. The relevant portion
of the text reads thus in literal translation: ‘How in the respect of the
paintings? [Here too the historical present is used, for] in the pictures
themselves men see the blows rained down on Kamsa, and how he is
dragged about.’11 Patanjali’s text refers, in fact, to a practice that remained
widely prevalent in India until the recent introduction of television, a
practice by which itinerant picture-showmen toured the countryside, telling
tales, and using painted scrolls to illustrate their telling. Jaina texts of the
early centuries of the Common Era refer to such picture-showmen as a
distinct category, setting them apart from actors, dancers, and ordinary
storytellers.
The travelling storyteller continued to be popular down the ages. The
dramatist Bana (seventh century CE) speaks of a category of picture-
showmen known as Yama pattakas, who held a painted cloth stretched out
on a support of reeds while, with a reed wand, they expounded the features
of the next world which formed the subject of their scrolls.12 One may also
point to Paithan paintings, roughly 12 inches by 15 inches, with imagery on
both sides, produced into the late nineteenth century; these were held up by
picture-showmen while they narrated the tales depicted thereon, which were
from the epics or from a variety of local legends.13 And indeed, there is the
famous Rajasthani Pabuji-ka-par, in which a quasi-historical romance is
similarly narrated, except that the painting, made on cloth, extends across a
span some 15 feet in length and 5 feet high.14 Two persons are involved in
this narration which continues to be a living tradition, a male bhopa who
tells the tale and a female bhopi who both sings and holds up an oil lamp to
illuminate different sections of the painting.
Did picture-showmen exist in an early Buddhist context? An isolated
passage in the Samyutta Nikaya section of the Pali canon contains a
reference to picture-showmen.15 When his monks reply in the affirmative to
the question as to whether they have seen carana-citras or ‘wandering
pictures’, the Buddha compares the work of the painter to the ability of the
mind to conjure up a world of illusion. Fifth century commentator
Buddhaghosha clarifies that the Buddha’s reference was to the pictures
(citras) of an itinerant or wandering (carana) artist. In a society with a
strong oral tradition, in which picture-showmen were common in a Hindu
context, it seems reasonable to assume that Buddhist legends were
disseminated in a similar manner. Such painted scrolls would then have
been the precursors of our earliest stone narratives. We have no indication
as to whether written captions accompanied such paintings; however they
do not exist today on either Paithan paintings or on the Rajasthani pars.
 Of interest in this context are the inscribed labels that appear on the early
narrative reliefs at Bharhut and a few other sites including Pauni and
Amaravati’s earliest monument (c. 100–50 bce), but disappear completely by
the time of the Sanchi stupa whose carvings date some fifty years later. Some
of the Bharhut inscriptions are neatly placed above or below a panel or
medallion, and are clearly in the nature of captions, stating, for instance,
‘Deer Jataka’ or ‘Descent of the Buddha’ above medallions depicting those
scenes. A crowded scene of worship has a label that informs the viewer that
the divine being Arahadguta has just announced, to the assembly of the
gods, the Buddha’s decision to be born again on earth. Other inscriptions,
while identifying elements of a narrative, are intrusive into the visual field. A
panel which depicts the heavenly nymphs celebrating the enlightenment of
the Buddha, has the names of the four dancers inserted beside each nymph;
in addition, a label stating that this is a divine pantomime is inserted
sideways into a vertical band of railing at the lower edge of the panel. In
other reliefs, palaces and mansions are identified by inscriptions placed on
their roofs, while serpent-king and earthly monarchs have their names
inserted beside them.16
The Bharhut carvings were executed at a time when the Pali canon was
transmitted solely as an oral tradition. Literacy, as such, could not have been
sufficiently high to warrant the suggestion that every devotee visiting the
shrine would have been able to read these labels. Yet, there seems little doubt
that the inscriptions are intended as aids to identification. They are not of
the category that served as instructions for the sculptors, as seems to have
been the case with the basement inscriptions on the eighth century stupa of
Borobudur in Java. The manner in which the Bharhut inscriptions are
inserted into the visual field, placed behind a kneeling figure, or running
sideways along the length of a relief pillar, indicates that the epigraphs were
added after the carving was done. The most plausible explanation is that the
captions were intended for the literate monks who used them as prompts
when they acted as spiritual mentors and guided worshippers through a
selection of the two hundred or so narratives presented on this stupa.
It is, however, perplexing to note that half-a-century later, on the Sanchi
stupa, not a single identifying label was added, and that such captions were
not used again in the context of Buddhist art until the days of the Ajanta
cave monasteries of the fifth century ce What happened in the period of
two, perhaps three generations, that intervened between Bharhut and Sanchi
to justify the total abandoning of a system of captions? Does the answer lie
merely in the increasing familiarity of the Buddhist legends? Or is it possible
that the Pali canon was committed to writing in India during this
intervening period, and that with the existence of a written text, inscribed
captions were considered redundant for the monks who had used them as
aids?
Five hundred years later, artists of the murals at Ajanta added painted
inscriptions of two types. One consists of verses in courtly Sanskrit painted
beside certain narrative murals. These would have been read and
appreciated only by the learned ecclesiastic or patron; it is unlikely that they
would have been translated for the benefit of the worshiper for whom the
monk may have served as guide. The second type, intrusive into the visual
field, identifies individual figures of a story. For instance, the words ‘King
Sibi’ are painted four times beside the figure of that monarch in his story in
Cave 17. It is intriguing that the label should have been thus repeated; once
would seem sufficient for the monk-cum-guide.
After fifth-century Ajanta, labels once again disappear in the Buddhist
context to recur only in the seventeenth century in Nepalese and Tibetan
scroll paintings, and on an occasional book cover. Tibetan tankas that depict
jatakas and avadanas with anywhere from five to ten stories on each scroll,
frequently have Sanskrit or more commonly Tibetan inscriptions painted in
gold letters beside the story. Belonging to a time when Buddhism had
developed a vastly expanded pantheon of deities, one reason for the labels
may be unfamiliarity with the stories. Ironically, the reason for adding labels
was probably the same in first century bce Bharhut as in seventeenth
century Nepal and Tibet; in the one case, the unfamiliarity was due to the
relatively recent formulation and dissemination of Buddhism and its
legends, while in the other, it was due to its remoteness from the original
conception.
MODES OF NARRATION

Monoscenic Narratives: Theme of Action

The monoscenic mode centres around a single event in a story, one that is
generally neither the first nor the last, and which introduces us to a theme of
action. Such a scene is usually an easily identifiable event from a story and it
serves as a reference to the narrative. This system of representation
functioned well in India, as it did in the classical world, where tales were
generally familiar to the viewer. The final outcome of a story was known
prior to its oral telling and prior to viewing its depiction in sculpture or
painting, or indeed as a theatrical performance. Monoscenic narratives
must, of course, contain sufficient narrative content to stimulate the story-
telling process in the mind of the observer. Viewers are introduced to the
story in the middle of the action, and the relation of key figures to scenic
details must be unmistakable and stimulate viewers into telling themselves
the story.
Several instances of monoscenic narrative involving a theme of action
exist along the lengths of coping at Bharhut, some 3 feet above eye-level.
One such that we have already seen is the Vessantara Jataka where the artist
presents the single scene of the gift of the state elephant, which is in the
nature of a prelude to the story. Another instance of monoscenic narrative
along the Bharhut coping is the Mahabodhi Jataka where the artist portrays
a young brahmana holding parasol, stick, bowl, sandals, and antelope skin,
while a barking dog confronts him; to the left stand a richly attired couple.
The knowing viewer will immediately recall the story of Mahabodhi,17 who
prepared to leave the royal kingdom after the monarch’s mind had been
poisoned against him by his counsellors. The king’s plan to have the young
recluse killed is heard by the dog whom the ascetic had befriended, and the
dog barks to warn him of their dastardly plans. The royal couple go to the
park where Mahabodhi resided, only to see him on the verge of departing,
and ask:
 What mean these things, umbrella, shoes, skin-robe and staff in hand?
What of this cloak and bowl and hook? I fain would understand.18
It is this single verse that the artist has chosen to present visually in this
instance of monoscenic narrative. The combination of key figures indicating
narrative elements is so specific that it must have aroused immediate
recognition of the story in the viewer. The process is somewhat different
from that of the Vessantara tale in which the artist presents viewers with a
scene that occurs at the very start of the action, leaving them to narrate the
entire story to themselves. Here, the scene presented occurs towards the
conclusion of the story and the viewer must recall the events that preceded
it.
A third instance will demonstrate the varied character of monoscenic
narrative. The Kukkuta Jataka tells the tale of a she-cat who tried
unsuccessfully to induce a cock to become her mate with the intention of
devouring him. The Bharhut coping panel merely depicts the two
characteristic animals, the one at the foot of a tree and the other perched
upon it, as sufficiently distinctive to stimulate recognition of the story.
Viewers are left to narrate the entire tale to themselves and to remind
themselves of the moral—the danger of succumbing to sensual desires. I
have suggested that the captions attached to the Bharhut scenes might have
been intended for the literate monk who took the worshipper around on the
first few visits, aiding in the identification of the stories. It is curious that the
easily ‘readable’ and instantly recognisable cock and cat story has an
inscribed label which reads ‘Cat Jataka, Cock Jataka’ (Bidala Jataka, Kukkuta
Jataka).
Monoscenic narratives contain fragments excerpted from a narrative that
function as signal references to viewers and rely on their prior knowledge of
the tales both for their narrative completion and for reminding themselves
of their moral significance. The extraordinary number of such monoscenic
narratives along the coping of the railing at Bharhut (some 45 such scenes in
the surviving two-fifths of the original 330 feet of coping), should be
considered in conjunction with the character of the rite of
circumambulation. As worshippers/viewer’s walked slowly around the stupa,
repeating the ritual circling three or more times, they could stop and
selectively absorb the stories. The process of telling themselves all the stories
represented on the railing would undoubtedly involve several visits to the
stupa, but then this was indeed expected of the devotee. As viewers
circumambulated the stupa, they were induced into extended contemplation
of Buddhist legends and Buddhist virtues by the presence of the many
narratives.
No perceptible hierarchy is evident at Bharhut in the presentation of these
Buddhist legends. There is no indication of a planned continuity that might
have commenced, for instance, with the earlier of the 550 Jatakas and
concluded with the final ten which tell of the human lives of the Buddha.
Nor are animal tales grouped in specific segments of the railing, as against
those featuring humans. The apparent absence of planning may partly be
explained by the collective and popular patronage of early Buddhist
monuments. The Sanchi stupa, for instance, contains no less than 631
individual donative inscriptions,19 each recording the gift of the pillar,
crossbar, or coping stone upon which it is engraved. While at Bharhut a
lesser number of such records has survived, there is a comparable variety
and range in the donations.20 From these inscriptions one may suggest the
likelihood that themes were the choice of individual donors.21 Specific
Jatakas were probably carved along a length of coping as donations came in
with requests for a particular legend. The fact that themes are repeated—
three Chaddanta Jatakas at Sanchi, two Vessantara legends at Bharhut, two
Mandhata Jatakas at Amaravati, three Maitribala stories at Ajanta—lends
support to such a practice of individual and personalised gifting.
Figure 6.2: Monoscenic narrative, Asilakhana Jataka. Bharhut coping
Source: American Institute of Indian Studies (AIIS).
Figure 6.3: Monoscenic narrative, Kukkuta Jataka. Bharhut coping
Source: AIIS.
The instances of monoscenic narrative thus far considered were all
inserted into restricted spaces, at most 15 inches across, provided by the
undulating stem of the lotus that meanders its way along the coping at
Bharhut. Since such examples seem to suggest that choice of the monoscenic
mode was dictated by restrictions of space, it is necessary to consider at least
one example of the use of the mode where ample space is available. Artists
carving the Chaddanta story along the inner face of the top architrave of the
north gate at Sanchi, some 30 feet above the ground, had at their disposal
the 8 feet central span of the architrave as well as its two extensions, which
provided them a total space of around 12 feet. I am assuming that the story
was chosen by the donor, but that the mode of narration was left to the
artists who decided on the monoscenic mode and portrayed the
introductory scene of elephants in a lotus-filled lake beside a banyan tree to
serve as a signal reference to this Jataka tale. The artist carved no less than
twenty elephants to represent Chaddanta and his followers, while the lotus-
filled lake is extended to cover the entire available space.22 The extensive
space available does indeed lead to a difference in the treatment of the
monoscenic mode. There is a strong centralised focus around the banyan
tree at the middle of the architrave, while the two extensions of the
architrave display a lateral discursive movement.

Figure 6.4: Monoscenic mode, Chaddanta Jataka. Top architrave (inner

face), north gateway, Sanchi stupa, c. 50–25 bce
Source: Archaeological Survey of India (ASI).
By its choice of a single scene to recall the entire story, the monoscenic
mode effectively eliminates the need to deal with the element of time that is
the most challenging problem for the artist confronted with a two-
dimensional surface, be it stone, cloth, or paper. What is the affect of the use
of the monoscenic mode? A story is presented in so succinct a manner that
it requires the viewer to act also as the narrator in order to experience it in
full.
Monoscenic Narratives: Being in State versus Being in Action
A static mode of monoscenic narration is frequently used by artists to
present the viewer with scenes from the Buddha’s life when the supremacy of
the Buddha is the prime concern. In this mode, artists generally present the
single, culminating episode of a story and focus thematically on the wisdom
and presence of the Buddha. In such depictions, the narrative content is
sharply reduced, and the reliefs represent scenes in which the action has
already taken place. The artist presents us with the result of a narrative
episode, or with the situation that immediately follows that narrative
episode. In this context, the term ‘being in state’, as contrasted with ‘being in
action’, seems suitable. An instance of the use of the static monoscenic mode
from Bharhut is the portrayal of the miracle at Sravasti, on the lowest panel
of the Ajatashatru pillar, in which the Buddha caused a full-grown mango
tree to emerge instantly from a newly-planted mango seed. The presence of
the Buddha is indicated in this panel by the combined indexical signs of the
footprints, the seat beneath the tree, and a regal parasol. The artist of this
Sravasti miracle was not interested in presenting the viewer with the
sequence of events (which constitute the episode) that led up to the
performance of the miracle. The brahmana’s act of cutting down every
mango tree in Sravasti in an attempt to foil the Buddha’s promise of
performing a miracle beneath such a tree, and gardener Ganda’s production
of a mango seed, were of little interest. Instead, the state after the miracle,
when the mango tree had already sprung up, is presented; the panel focuses
on the presence of the Buddha surrounded by worshippers offering him
homage. It is the divine power of the Buddha that is highlighted in this
narrative in which action is totally absent. Meyer Schapiro, who discusses
differences between representations revolving around ‘being in state’ and
‘being in action’, establishes that static depictions were regularly used in
situations where theological concerns were predominant.23 It appears that
the artist of the Bharhut Sravasti panel was interested in emphasising the
supremacy and power of the Buddha at the expense of the fascinating series
of events leading up to the Sravasti miracle, events in the life of a human
Buddha.
 The Sravasti miracle is one of three scenes, all emphasising ‘being in state’,
that stand in a diachronic relationship on the Ajatashatru pillar.
Commencing with the miracle of the mango tree at the bottom of the pillar,
the viewer must move to the topmost panel where the Buddha preaches to
the gods in the Trayastrimsa heavens, and end with the central panel that
depicts the descent from the heavens at Sankissa. The preaching in the
heavens, that followed the Sravasti miracle, likewise emphasises the great
wisdom of the Buddha and belongs to the static category of narration. Its
compositional similarity to the Sravasti miracle is striking and, in isolation,
would present problems of identification. Its position in conjunction with
the Sravasti and Sankissa miracles enables those familiar with the
Buddhological cycle to place it as the intervening episode when the Buddha
went up to the heavens to preach to the gods and to his mother who had
died soon after he was born. At the end of his stay in the heavens,
accompanied by Indra and Brahma, the Buddha descended to earth at
Sankissa, upon ladders created by the gods. Crowds of worshippers
thronged to greet him, and the Buddha preached a sermon at the site. The
panel of the descent at Sankissa includes elements of action in presenting the
footprint of the Buddha on the topmost and lowest rungs of the central
ladder as references to his act of descent. Yet, the action is complete, as
indicated by the lower footprint, while the ensuing worship of the Buddha,
portrayed to the left of the panel by the indexical signs of the throne,
parasol, and tree, reinforces the fact that the descent is complete. Thus, even
in a panel that narrates an event of action, the static mode is predominant.
Figure 6.5: Static monoscenic narration, diachronic mode. Ajatashatru pillar,
Bharhut
Source: ASI.
 This 6 feet 8 inch pillar presents the viewer with a narrative sequence that
involves a vertical movement; it commences its diachronous narrative at the
bottom, and requires an ascent to the topmost panel before culminating at
the centre. The choice of vertical movement as opposed to the lateral
movement apparent on the Sanchi architraves is explained by the particular
architectural unit that was being covered with narrative reliefs, and on the
obvious compositional difference imposed by the vertical format of a pillar
as compared to the horizontality of a gateway architrave. It must be
mentioned that the panels on a pillar are not invariably related; for instance,
the adjoining face of the Ajatashatru pillar portrays, from top to bottom, a
scene of heavenly worship of the Buddha’s hair-lock (cut off when he left the
palace in search of enlightenment), the announcement in the heavens that it
was time for the Buddha to take birth on earth, and king Ajatashatru’s visit
to the Buddha. While pertaining to events in the life of the Buddha, the
three have no integral relationship one to the other.
Figure 6.6: Static monoscenic narration, synchronic mode. Prasenajit pillar,
Bharhut
Source: ASI.
 Relief panels from the Sanchi stupa and the Gandharan region further
demonstrate the nature of monoscenic narratives that focus on being in
state. The topmost panel of the eastern pillar of Sanchi’s north gateway
portrays the Sravasti miracle and centres upon a seat beneath a mango tree
that bears clusters of fruits and is surmounted by a parasol flanked by
chowris. Rows of seated figures join their palms in respectful adoration
while two divine flying beings beat upon large drums (dundubhi) to
proclaim a miraculous happening. As with the Bharhut example, the Sanchi
panel dispenses with the events that led to the performance of the miracle;
instead it presents the moment of the miracle with the emergence of the full-
grown mango tree.
Figure 6.7: Synoptic narrative, Monkey Jataka. Sanchi stupa
Source: ASI.
Monoscenic narration in the theme of state is the predominant mode
used to decorate a small stone stupa from the Gandharan site of Sikri, that
now stands reconstructed within the Lahore Museum. A panel that is typical
of this group depicts the Buddha, carved somewhat larger than the figures
around him, and enthroned beneath a centrally placed mango tree. The
presence of a female figure in the left foreground, pouring water which
signifies the ratification of a gift, indicates that this is not the Sravasti
miracle; it depicts Amrapali’s gift of her mango grove. The artist was not
interested in the fascinating story of Vaisali’s most renowned courtesan or of
the Buddha’s acceptance of her friendship and hospitality. The sole concern
was an emphasis on Amrapali’s status as a Buddhist convert, and a
demonstration thereby of the supremacy of the Buddha and his doctrine.
The static monoscenic mode effectively and simply emphasises
theological concerns. The centralised composition, and the repetitive
character of the motifs from one panel to the next, frequently tend to convey
an impression of a dearth of artistic inspiration. When viewed, however, in
the context of an intended emphasis on the supremacy of the Buddha, the
evaluation of artistic success and failure may need to be amended.

Continuous Narrative Continuous narratives depict successive events of

an episode or successive episodes of a story within a single enframed unit,
repeating the figure of the protagonist in the course of the narrative.
Consecutive time frames are presented within a single visual field, without
any dividers to distinguish one time frame from the next; however, temporal
succession and spatial movement are generally clearly indicated. The
comprehension of continuous narrative requires awareness that more than
one moment of time is presented within a single visual frame, and that
multiple appearances of the protagonist indicate successive moments of time
as well as successive spaces in which action occurs. In Otto Pacht’s words, ‘it
is the very essence of continuous narrative to render changes visible by
comparing the same person in different movements or states’.24 Yet, the
viewer unfamiliar with such presentation may indeed find these repeated
appearances of the protagonist within a single enframed setting both
illogical and incoherent.
A classic instance of continuous narrative which depicts successive phases
of a single episode, and in which spatial movement and temporal
development are clearly presented, is the Great Departure of the Buddha,25
portrayed on the outer face of the central architrave of the east gateway at
Sanchi. The narrative moves from left to right along the 8 feet span of the
architrave; and commences with the Buddha-to-be emerging on horseback
from the gates of his palace with the groom leading the horse whose hooves
are supported by the gods to muffle their sound and allow the prince an
undetected departure. The indexical sign of the Buddha’s presence on
horseback is an empty space with a regal parasol held above at an
appropriate height. To depict the progressive ride away from the palace, the
artist has repeated the figure of the progonist horse-and-rider another three
times across the span of the architrave. The completion of this episode at the
far end of the architrave, when the Buddha dismounts from the horse, is
represented by a parasol poised above a pair of footprints. To indicate that
the riderless horse (without the parasol) and groom then return to the
palace, the artist has placed them facing left, counter to the movement of the
ride away from the palace. Temporal succession is clear and spatial
progression is equally straightforward in this depiction in which horse and
rider are depicted a total of five times without any internal framing devices
to distinguish their various appearances.
Figure 6.8: Synoptic narrative, Chaddanta Jataka. Amaravati, second century
CE
Source: ASI.
I stated earlier that on occasion narrative modes overlap and may not
always be sharply differentiated. The Great Departure architrave, an
apparently straightforward instance of continuous narrative, contains at its
very centre a reference to an earlier incident. The rose-apple tree surrounded
by a railing and topped with a parasol makes reference to the first
meditation of Prince Siddhartha, and comprises an instance of what Genette
terms analepsis.26 It was that first meditation which led to his determination
to abandon his princely existence and set forth in search of ultimate truth—
the subject of our continuous narrative. While using continuous narration to
present us with a clear statement of a temporal sequence, the artist then
inserts a ‘flash-back’ to an earlier time, an analeptic reminder of an incident
with an integral connection to the main narrative.
The dominant or prevailing movement in Buddhist visual narratives is of
obvious relevance, but the nature of the rite of circumambulation rules out
any categorical statement on the matter. A stupa is circumambulated in
clockwise manner, both from the exterior of the sacred area, and from
within its railed enclosure. When outside the sacred area, pilgrims will view
the reliefs on the railing (its outer face) from right to left, and its narratives
generally follow this direction of movement. Once pilgrims enter the sacred
enclosure, they will view the narratives on the same railing (its inner face)
from left to right, while experiencing reliefs placed against the stupa itself
from right to left. Thus the direction of movement in a narrative frieze or
painting may commence from either left or right, depending on its position
relative to the stupa or other sacred object.27 However, in several small
panels that measure only a foot or so in length and present three or four
scenes from a story, we find that the movement is from left to right
regardless of the position these occupy upon the monument. One may note
in this context that the Indian system of writing moves from left to right
across a page or a stone tablet; the influence of this factor may not be
entirely discounted.

Figure 6.9: Conflated narrative, Dipankara Jataka. Gandhara, second/third

century CE
Source: author.
A second and more extensive example of continuous narrative is the
portrayal of the Vessantara Jataka, in a series of twenty-one scenes, on the
north gateway at Sanchi. Both sides of the lowest architrave and its
extensions (with the exception of the eastern extension of the outer face) are
utilised to present the gradual unfolding of the story. The figures of the
protagonists, Prince Vessantara, his wife Madri and their two children, are
repeated no less than fifteen times. On both faces of the architrave, located
some 20 feet from ground level, the movement is from right to left. While
such movement would be appropriate for the outer face, in which the viewer
would be circumambulating the stupa on the exterior of the enclosure, it is
no longer suitable for the inner face. Here, the circumambulating viewer,
keeping the stupa to his right, would view the 12 feet of narrative (the
architrave and its extensions) from left to right. The sculptor appears to have
carved one face of the architrave, and then commenced the inner face at the
end just completed. Viewers must perforce abandon the general movement
of circumambulation if they wish to experience this story in its entirety.
Once they have followed Prince Vessantara and his family into banishment,
from outside the sacred enclosure, they must stop and enter through the
gateway in order to follow the denouement of the Jataka tale along the inner
face of the same architrave. The stop, an important element in the
construction of literary narratives, comes here at an appropriate juncture.
Figure 6.10: Conflated narrative, Deer Jataka. Bharhut
Source: ASI.
Scenes 1 to 8 on the outer face of the gateway are relatively easy to read
and move from right to left; in the background, however, movement from
left to right is used to depict the completion of episodes that commenced in
the foreground. Prince Vessantara, riding the auspicious state elephant, is
beseeched by brahmanas to donate it (1), and pours water to ratify the gift
(2). The banished Vessantara family takes leave of the king (3), and departs
in a horse-drawn chariot (4). Vessantara gives away the horses (5) and the
chariot (6, 7), and the family proceeds on foot towards the forest (8). Scenes
5 and 7, depicting brahmanas with the gifted horses and chariot, are placed
counter to the general right to left movement; since they portray the
recipients of the gifts returning to the town which the Vessantara family has
just left, such a directional placement is appropriate. The inner face of the
architrave, which contains scenes 9 to 21, is slightly more complex in its
reading of temporal development. The narrative commences with a clear
right to left movement, and uses the back ground extensively for the
completion of the various episodes. Vessantara and his family reach the
forest (9, 10) and are seen in their hermitage (11). The evil brahmana
outwits a guard (12), and lions prevent Madri’s timely return (13), so that
Vessantara gifts the children (14), who are led away by the cane-wielding
brahmana (15). Madri returns to the hermitage (16), to learn the news from
Vessantara; the prince finally gifts his wife (17) only to receive her back from
Indra (18). The artist now requires us to move to the left extension of the
architrave which contains the penultimate episode of the legend in which
the evil brahmana sells the children to their grandfather, the king (19).
Returning to the central span of the architrave, the artist introduces a
movement from left to right, with the monarch on horseback, together with
his grandchildren on elephants, riding into the forest (20). Even viewers
quite familiar with the story of Prince Vessantara find themselves
momentarily challenged with the decipherment of this visual narrative
towards its completion; the reason for the left to right movement appears to
lie in the geographical placement of the palace at the far left. The final scene,
in the rear ground, returns to a right to left movement and depicts the
monarch, the children, Vessantara, and Madri, riding towards the palace
(21). There are no internal scene dividers, and the viewer unravels the story
merely by moving from one depiction of the Vessantara family to the next.
Yet the spatial treatment of the story is quite logical. The narrative
commences with the palace from which Vessantara is banished (outer face
of architrave) and culminates with the palace into which Vessantara is
welcomed back (inner face, left extension). The appearance of a palace thus
signals to the viewer either the commencement of the narrative or its
conclusion.
Figure 6.11: Continuous narrative, Great departure of the Buddha. Central
architrave (outer face), East gateway, Sanchi stupa
Source: author.
A variation on the idea of continuous narration, which may indeed be
termed continuous action, is seen in the portrayal of two separate figures
that are actually to be read as a single figure completing a set of movements.
One such example is provided by a medallion from Amaravati which
portrays at its lower edge two female forms, one kneeling and the other bent
backwards. Confirmation of such an interpretation comes from an Indian
dancer who, having viewed the medallion, spontaneously and immediately
read the image as that of a dancer completing a dance sequence.28 Joanna
Williams, in her studies of Ramayana manuscripts, points out that images of
two-headed deer may be read the same way; the second head, turned
backwards, indicates that the fleeing deer is looking back at its pursuers. She
remarks that the appearance of multiple body parts, throughout the history
of Indian art, may often be read in this manner.29 The manner of portrayal
of both the Amaravati dancing image and the two-headed Ramayana deer
may be viewed as unique attempts to portray the passage of time; both may
be considered examples of continuous action.

A Beginning in Medias Res?

A narrative that commences in medias res starts with an important event of
a story rather than with the first event in time; such an event is generally
followed by a return to an earlier period of time. The question posed here is
whether such a narrative construction was utilised by the artists working at
the ancient Buddhist sites.
Along the right wall of Cave 10 at Ajanta, in a band 4 feet wide and 60 feet
in length, is a painted mural that tells the Jataka tale of elephant Chaddanta
in a set of twelve scenes; While consideration of the mural at this particular
stage of our discussion suggests the use of continuous narration, the story
does not commence at one end and move logically and sequentially to the
farther end. Instead, circumambulating viewers encounter the story towards
its climactic conclusion. They find themselves faced with a dense fearsome
forest in which an elephant and a crocodile fight furiously, while a vicious
serpent grabs another elephant’s hind foot. In this ominous setting viewers
first encounter the six-tusked elephant king Chaddanta surrounded by his
herd (8). In quick succession, the artist presents the tragedy of the tale. We
are first shown Chaddanta prostrate on the ground, submitting meekly while
hunter Sonuttora saws off his tusks (9); then we see the hunter tying the
tusks together in bundles (10); and finally departing with the tusks slung
from a pole balanced on his shoulders (11). How and why did all this
happen? The original cause of these events now follows, commencing with
Chaddanta in happier days presenting the lotus that caused all the trouble to
his senior queen Maha-subhadda (3) and thus arousing the consuming
jealousy of junior queen Chulla-subhadda. Moving right, we see Chaddanta
sporting with his horde in the lotus pond (2), and next beneath the banyan
tree (1). If we move right again, we find the conclusion of the tale when the
queen of Benares (Chulla-subhadda reborn to seek revenge) swoons at the
sight of the tusks presented by the hunter (12). To trace the series of events
leading up to this event, we must move further right and view a set of scenes
that occurred at the court of Benares (4–7), culminating in the queen’s own
request for the tusks.
Viewers wishing to follow the story in the sequence of its occurrence face
a most convoluted narrative; they would have to walk backwards and
forwards more than once along the 60 feet length of wall. If however, the
artist intended to present the narrative using the construction of medias res,
the entire experience is different. Viewers merely allow themselves to be
plunged into the story towards its climax, and see it slowly unravel before
their eyes. An issue worth consideration is whether alternative recensions,
particularly continuing oral ones, might offer scenes in a variant order from
the Pali canonical or post-canonical written versions. While the Chaddanta
tale in the Jataka collection follows the temporal sequence of events, we may
recall that the Buddha’s narration of the events was occasioned by the loud
weeping of a female novice who remembered how, in her previous life, she
had caused the death of elephant Chaddanta by sending a hunter after him.
This preliminary reference to the climax of the story, found in other Jatakas
as well, may perhaps be viewed as a variation on the technique of
commencing a story in medias res.
Figure 6.12: Detail of Linear narrative, Vessantara Jataka. Goli, third century
CE
Source: ASI.
Sequential Narrative
Sequential narrative,30 like continuous narrative, contains the repeated
appearance of the protagonist at different times and places; the distinction
between the modes is compositional in nature, and revolves around the
principle of enframement. In continuous narrative, temporal development is
to be understood by means of intrinsic criteria and requires an integrating
effort of mind and eye on the part of the viewer. In sequential narrative, on
the other hand, extrinsic criteria are used to demarcate temporal divisions.
Scenes are separated from one another by a variety of compositional means,
and generally each episode is contained within a separate frame. In
sequential narrative, each scene is a unit in itself; each event occurs at one
specific moment, in one particular space. Frequently, sequential narrative is
found in the format of a horizontal frieze, although on occasion it may
assume a vertical layout.
The Vessantara Jataka from Goli in south India makes effective use of this
mode, and in a panel over 10 feet long the artist depicts the story of the
prince whose charity is a byword in Buddhist legend. Different episodes of
the story, to be read from left to right, are neatly demarcated by columns,
trees and other vertical devices, so that the viewer is led from one scene to
the next in clear succession. Commencing with column dividers, we see
Prince Vessantara riding his elephant, and then giving it away. We are next
shown the departure of the banished prince with his wife and two children
in an ox cart, followed by Vessantara giving away first the oxen, then the
cart, and finally the two children. This central section of narrative uses a
series of subtle markers—rocks in a pyramidal formation surmounted by a
tree—to divide the various episodes. It could indeed be seen as the use of the
continuous mode within a larger scheme of sequential narrative. The final
episodes, culminating in the reunion of the Vessantara family with the king,
are once again demarcated by columns. These divisions introduced by the
artist are characteristic of the mode of sequential narrative; they could be
said to resemble the subheadings of an article introduced by an author to
facilitate the readers’ comprehension of the whole.
 Instances of sequential narrative from the site of Nagarjunakonda
demonstrate even more pointedly the extrinsic nature of the temporal
divisions introduced. A horizontal sequence 10 feet long that presents four
episodes to tell the story of the Buddha’s half-brother Nanda, is punctuated
by pilasters flanking an amorous couple. Couples, representing fertility and
hence signifying growth, abundance, and prosperity, are symbols of the
auspicious in the art of India; they had clearly become conventional scene
dividers in the art of the southern region of Andhra. Each enframed
narrative unit of the Nanda story is complete in itself, with the repetition of
the figures of Nanda and the Buddha from one unit to the next. The first
scene, somewhat damaged, portrays the Buddha carrying an alms bowl,
with Nanda beside him. The next episode tells of Nanda’s forced induction
into the Buddhist Order with his head shorn in the Buddha’s presence. The
artist then moves on to the flight of the Buddha and Nanda to Indra’s
heaven, leaving the viewer to fill in that part of the story which tells of how
Nanda was unable to forget his beautiful wife and wished to leave the Order.
It was to cure him of his love-sickness that the Buddha decided to confront
him with the beauty of the nymphs of heaven. The artist portrays the flying
figures of Buddha and Nanda, and a group of heavenly nymphs. The
prominent figure of a monkey at the feet of the flying figures reminds us that
compared to the nymphs, Nanda now found his wife more akin to the
monkey. The final segment of the sequence portrays a ‘cured’ Nanda being
welcomed by the residents of his town.
The most extensive instance of sequential narrative may be seen in the
panels that cover the many levels of terraces of the great stupa at Borobudur
in central Java built around 800 CE. Running around the walls of the first
terrace, along the upper section of the wall (the lower depicts the Jatakas),
are a complete series of scenes from the life of the Buddha, commencing
with the scene in the Tushita heavens when the Buddha announces that it is
time for him to be born on earth as Gautama, and concluding with the first
sermon.31 The artists presented their viewers with a sequence of one-
hundred-and-twenty coherent independent episodes, each contained within
the boundaries of a panel, 10½ feet long and 3 feet high, and extending
1,260 feet in their totality. The circumambulating pilgrim was provided with
a unique opportunity to contemplate the life-cycle of the Buddha.
Figure 6.13: Detail of Linear narrative, Story of Nanda. Nagarjunakonda,
third century CE
Source: ASI.
Synoptic Narratives
In the synoptic mode of narration, multiple episodes from a story are
depicted within a single frame, but their temporal sequence is not
communicated and there is no consistent or formal order of presentation
with regard to either causality or temporality. The scant attention paid to the
element of time is characteristic of this mode. The multiple episodes of a
story generally contain the repeated figure of the protagonist.32
Five episodes from the Monkey Jataka are depicted within a rectangular
panel from the Sanchi stupa, with no indication whatsoever as to temporal
succession or causality. The viewers’ attention is likely to be caught by the
river that curves across the panel, the six prominent foreground figures, and
perhaps by the monkey straddled across the panel at the very top. However,
even the knowing viewer must closely scrutinise the panel to ‘read’ the story
accurately, and to realise that the foreground figures are of little relevance to
the action, or that the entire section to the far side of the river is marginal to
the tale. To the lower left, behind a group of soldiers and musicians, a
monarch arrives on horseback (1). Roughly at the center of the panel is the
half-hidden figure of an archer, bending backwards as he aims his arrow
directly upwards (2). If we realise the importance of the archer and follow
the upward movement of his bow, we see a monkey who has stretched
himself out to form a bridge across the river below, and we surmise that his
monkey friends are escaping from the archer to the safety of a tree on the
opposite bank (3) where deer slumber in a quiet forest environment. Below
the monkey, two men hold a stretcher (4), and to the upper left of the panel
we see the seated figures of the monkey and the monarch (5). Further
analysis will make it evident that the river meandering its way across the
panel creates a distinction between the area where all the activity takes place
and a distant safe environment. The artist has left much unsaid and given
prominence to subsidiary figures so that even those familiar with the
Buddha’s previous life as the monkey king will find it difficult to read this
narrative correctly. They will need to supply more than one key element of
the story, including the fact that the arch enemy monkey jumped so heavily
upon the monkey-king’s back while escaping across the river that the
monkey king was mortally wounded. They will need to supply the human
monarch’s admiration of the scene he has witnessed, understand that the
stretcher was to rescue the injured monkey, and view the last episode as the
monkey-king preaching to the monarch on the vital importance of
attending, at any cost, to the welfare of his people. It is this virtue that the
story exemplifies, and of which the viewer has to remind himself while
stringing the episodes together. While the artist has undoubtedly supplied
clues to reading the elements in some order, and while a certain structure
appears to underlie that order, the viewer is able to ‘read’ the tale in a
coherent manner only after a recapitulation such as that presented above.
One may remark that the irregular pattern of staging that would confuse the
uninitiated, creates interest for the knowledgeable viewer, offering spice and
enjoyment to the viewing!
One of the complex issues that confronts narrative artists working in a
spatial medium is the manner in which to convey the essential added
dimension of time. We have seen that artists using the monoscenic mode
bypassed the entire problem. Those using the continuous and sequential
modes conveyed the passage of time in a clear manner by repeating a set of
successive occurrences, creating an interesting parallel to the artist working
on a cartoon film who produces a series of successive drawings, each a
moment in time ahead of the previous. Artists using the synoptic mode
ignore the viewers’ problems in deciphering a narrative and seem to display
a total indifference to the presentation of the element of time. While the
repetition of the protagonist is indeed an indication of a different moment in
time, the artists gave the viewer few clues as to the sequence in which to
view the various repetitions of the protagonist, or where exactly to
commence and conclude their viewing.
A second instance of synoptic narrative that once again presents a
challenge in decipherment, even to those familiar with the story, is a
medallion from the stupa at Amaravati, belonging to the second century CE.
In a total of seven scenes that may be grouped into three episodes, the artist
presents the story of the Buddha’s previous birth as the elephant Chaddanta.
In view of the abraded condition of the original medallion and the complex
nature of the visual treatment, I have superimposed my numbering scheme
upon a drawing of the image in order to facilitate reading of the narrative.
The centre of the medallion is ‘dead’ space, and the artist commences to the
right of the central zone depicting Chaddanta in the forest, with the
identifying regal parasol above him (1). The lower zone is treated as a locus-
filled pond, and here we see the occurrence of an important episode of this
story, treated in three scenes. To the right, Chaddanta presents his chief
queen, Maha-subhadda, with the trouble- causing lotus (2), while to the left,
his offended and jealous junior queen Chulla-subhadda leaves the pond (3).
Her figure is repeated to the extreme left (4), just beyond the pond, where
we see her lying down to die (praying for revenge in a future birth). The
story now moves to the upper zone where the next episode is depicted to the
far right. We see the unsuspecting Chaddanta, at whom hunter Sonuttara
(sent by Chulla-subhadda reborn as the queen of Benares) aims an arrow
from his hideout in a concealed pit (5). To the left of the upper zone, the
hunter saws off the tusks requested by the queen, while Chaddanta quietly
acquiesces (6). The final scene chosen by the artist, located at the very top of
the medallion, portrays the hunter departing with the tusks (7). This
example of synoptic narrative, presented within the circular space of a
medallion, provides little indication to the viewer of either temporal
sequence or causality.
The two instances of synoptic narrative that we considered present us
with multiple episodes placed in varied patterns within a single space. One
of the implications of this method, and indeed of several other narrative
modes, is that the artist relied on a knowing viewer. These ‘illegible’
narratives could be read only by those who previously knew the story and
could hence read the episodes in their correct sequence, supplying the
missing narrative elements from their memory. The viewer was given aids to
stimulation of narrativisation by being presented with more than one image
of the same characters in diverse spaces and actions. Yet the depictions
depended for their decipherment on the prior knowledge of the viewer
[emphasis added].
The second implication of the synoptic method concerns the way it
undermines chronology and influences the manifestation of a story since its
outcome is visibly seen at the outset. The narrative layout of these
medallions and panels, and their relatively restricted size, is such that
different episodes are perceived simultaneously by the viewer. The manner
in which one approaches synoptic visual narrative is, in significant ways,
almost the reverse of the way one hears or reads a story. Rather than putting
elements together to make a whole, we are presented with a whole which has
to be taken apart in order to render it intelligible. As far as time sequence is
concerned, one may first view an episode from the middle of a story, next
one from the end, and only then one from the beginning. As Richard
Brilliant observed in a slightly different context, ‘temporal succession can be
conceived as a road that neither comes from somewhere nor goes anywhere
but is comprehended, as if from far away, as a whole’.33 The various events of
these Jatakas, temporally overlapped, are meant to communicate a religious
and moral message—the victory of good over evil. It is not the revealing of
the facts that is important since the facts are known to the viewer. It is the
telling of the story, or stimulating the viewer into telling himself the story,
that is the significant experience. It is an end game, a teleological scheme.
Synoptic visual narrative, by its choice of episodes depicted, often gives
priority to events that a literary narrative, whether verbal or written, does
not, and it may, by its presentation, persuade the viewer to consider the
‘story’ differently. A comparison of the Sanchi Monkey Jataka just
considered with a Bharhut medallion presentation of the same story
highlights the manner in which the choice of episodes, as well as their
placement within the space of a panel or medallion, influences the visual
reception of a story. The Bharhut artist omits the first two episodes
presented at Sanchi—the arrival of the monarch, and the archer shooting up
at the monkeys. The major portion of the medallion is devoted to the escape
of the monkeys who use their monkey-king as a bridge to safety. The centre
foreground depicts the Benares king’s followers holding a stretcher in which
they will rescue the injured monkey-king, while the very front portrays the
monkey-king preaching to the human monarch. The Bharhut artist’s
emphasis is on the predicament of the monkeys, while the Sanchi artist
relegated the escaping monkeys to the top margin of his panel. The Sanchi
artist gave pride of place to the monarch and his retinue whose arrival
occupies the left foreground of the panel. (The Bharhut medallion may, in
fact, be read from top to bottom; the Sanchi panel presents a complex
movement in time and space.) The Bharhut portrayal conveys an impression
of the heroic deed of the monkey-king; the Sanchi panel emphasises, rather,
the aspect of kingship. What catches the eye first is undoubtedly significant;
it does so precisely because the artist intended it that way. A figure or an
event is placed in a position where it could not be ignored, precisely because
the artist wished to emphasise its relevance. Equally, what is left out is also of
consequence. As Sherlock Holmes remarked cryptically in solving the case
of The Silver Blaze, ‘The dog did nothing in the night-time!’
Synoptic narrative also obliges viewers to enter into an active relationship
with the content of a story [emphasis added]. In monoscenic narrative,
viewers have to remind themselves of the story; in synoptic narrative,
viewers must engage in the game of unraveling the story. Viewers are
required to become more or less active participants; as visual readers, they
are compelled to be also storytellers.
It is interesting to ponder over the reasons for the use of the synoptic
mode that subverts the element of time. One cannot but wonder if there is
any underlying cultural explanation for its popularity in Buddhist art. The
fact that time in India is considered to be cyclical, rather than linear as it is
in several major civilisations, may possibly have something to do with its
apparent disregard in several modes of visual narrative. At the same time,
one wonders if the ancient artists would have been amused, or peeved, if
asked why they had not presented temporal progression more explicitly! We
shall return to this issue later in this essay.
Conflated Narrative
Conflated narrative is complementary to the synoptic mode with which it
shares many features. However, while multiple episodes of a story or
multiple scenes of an episode are presented, the figure of the protagonist is
conflated instead of being repeated from one scene to the next; this
characteristic overlapping manner of presentation subverts temporal
succession even further.
Typical of conflated narrative is the portrayal of the Dipankara Jataka by
the Indo-Hellenistic school of Gandhara during the second and third
centuries CE. The tale relates to the young brahmana Sumedha’s worship of
Dipankara Buddha who declares that Sumedha will be reborn as the
historical Gautama Buddha. The right half of the panel is occupied by an
enlarged figure of Dipankara Buddha followed by a monk. At the extreme
left, Sumedha buys lotuses from a young woman. He then throws the lotuses
at Dipankara (they remain suspended around the Buddha’s head), spreads
out his long hair upon the slushy ground for the Buddha to step upon, and
rises up into the air upon hearing Dipankara’s pronouncement. The single
large image of Dipankara is to be read as receiving the lotuses, perceiving
Sumedha with his hair spread out upon the slushy ground, as well as making
the prediction of the future birth, and thus, as participating in three scenes
in which Sumedha is portrayed three times.
In this and other parallel instances of conflation revolving around the
person of the Buddha, one may view the choice of the mode as intended to
convey the Buddha’s transcendence over both space and time. The single
image of the Buddha is the visual equivalent, spatially, of three Sumedhas;
and the single image possesses the ability to participate in more than one
temporal moment of time. It may be possible to view such a use of
conflation as reflecting the values, vis-à-vis its founder, that are central to
any system of religious beliefs.
It seems, however, that an explanation of a different nature must account
for the use of conflation in narratives where a character other than the
Buddha as protagonist is subject to conflation. One such instance is a
Bharhut medallion presentation of the Quail Jataka, where the quail (who is
not the Buddha in a previous life) is placed on a tree to the right of centre; a
set of six scenes that unfold around her all refer to her single presence. In the
lower half of the medallion are two elephants, representing the herd of the
bodhisattva elephant to whom the quail appealed and who peaceably walked
past her nest on the ground in which were baby quails too young to fly.
Above is the rogue elephant who followed shortly thereafter; ignoring the
quail’s appeal he deliberately trampled the nest seen to the left at the point
where the medallion is damaged. The quail swore revenge and sought the
help of three friends, a crow, a fruit fly, and a frog, to destroy the elephant.
The medallion contains a second instance of conflation in the figure of the
rogue elephant, who is part of three different episodes, all of which refer to a
single image. The same elephant crushes the nest of baby quails, has his eyes
pecked out by a crow, and eggs laid in the sightless eye by a fruit fly. The
croaking frog at the top of the medallion, seated on the edge of a precipice,
brings about the elephant’s death; the half-blinded elephant follows the
sound, expecting to find water, but plunges instead to his death. The artist
provides us with a second image of the evil elephant whose rear end is seen
as he falls over the cliff. The explanation for the use of conflation in this
medallion, and the consequent compression and undermining of time, is
clearly different from that proposed when conflation occurs with the image
of the Buddha. The centralised positioning of the single image of the quail,
referring to six different episodes that unfold around her, may have been
dictated by artistic considerations and the fact that she provided a logical
point of reference. When a single character takes part in a number of
episodes, the artist could avoid needles repetition by making that character
the central or most prominent motif. The conflation of the figure of the
rogue elephant is mystifying, unless the device was adopted merely due to
shortage of space within the medallion.
A variation on conflated narrative that shares elements from the two
instances considered may be seen in a medallion from Bharhut which
presents four episodes from the Deer Jataka. The artist sets the scene by
portraying a forest in which a large deer (the bodhisattva) occupies the
centre of the space while others of the herd, all of a smaller size, cluster
beside him. Having communicated the setting, the artist moves to the lower
strip of the medallion schematically treated to depict a river, where a deer
rescues a hunter from drowning. Moving to the far right of the medallion,
the artist positions the ungrateful hunter pointing out the deer to the king’s
archers who stand with strung bows, ready to shoot. It is the same deer who
set the scene for the story at whom the hunter is pointing with raised index
finger. For the final episode of the Jataka in which the deer preaches to the
king and his followers, the artist uses this same central deer; the king and his
followers, with hands joined in worship, are placed towards the centre of the
medallion, beside the deer. We see here the conflation of more than one
episode of the story, with the central enlarged bodhisattva deer intended to
be read as participating in three episodes. The centralised position of the
deer reminds us of the positioning of the quail in the previous example,
while the fact that the conflated figure is the Buddha in a previous life may
be read as all indication of his ability to participate in more than one
moment of time. While the various appearances of the deer have been
conflated into a single figure, the narrative events are quite distinct as
compared to the Dipankara Jataka panel. The composition of the Deer
Jataka, revolving around the central conflated deer, appears eminently
suitable for the format of a circular medallion; however, the conflation of the
figure of the protagonist certainly confuses a reading of temporal succession.
Narrative Networks
A complex variety of story-telling, that may be described as a system of
networks, is seen at its most expressive in the murals at the monastic site of
Ajanta, executed towards the end of the fifth century CE. While all the caves
at Ajanta, including the chaitya halls, are covered with painted scenes, the
most ambitious narrative sequences are found in the residential viharas at
the site. On occasion, the entire side walls of such caves, interrupted by the
doorways that lead into cells for individual monks, are given over to the
depiction of a single narrative, and the result is frequently one of
labyrinthine complexity. These interlaced sequences meander across the
walls of Ajanta in such a manner that several murals remain unidentified.
With few clues as to where a story is likely to commence or where one might
expect it to conclude, it appears that only an intimate knowledge of the
Buddhist texts popular in the fifth century CE. will enable one to unravel
these tortuous sequences.
An idea of the organisation of narrative networks may be derived from
two examples painted upon the two side walls of Cave 17 at Ajanta, a
spacious vihara about 70 feet square produced under the patronage of
Prince Upendragupta, a feudatory of Vakataka king Harisena who
dominated the region. The story of Simhala, told in twenty-nine scenes,
extends 45 feet along the right wall of the cave, covering the entire 13 feet
span from floor to ceiling. While the protagonist, the merchant Simhala, is
repeated in several scenes that merge and flow into one another, there is
none of the coherence that accompanies continuous narrative with its clear
depiction of temporal succession and spatial movement. Instead, the action
moves across the 45 feet of wall in an unpredictable manner, commencing at
the lower level of the right end and moving upwards, then working its way
across the upper segment of wall to the left where it meanders downwards,
finally culminating in the central section of the available space. Within each
of these three segments—right, left and center—the action moves in criss-
cross fashion, and no specific pattern emerges from a close study of the
painted wall. In fact, one is confronted with a complete network of
movement in space and time.
 A line drawing of the Simhala story as analysed by Schlingloff,34
accompanied by an abbreviated summary of the story, will enable an
appreciation of the complexity of the action. Schlingloff ’s drawing of the
architectural motifs (minus the human figures) that forms the background
for this complex narration, demarcating and setting off certain segments and
scenes, is helpful. Yet, these architectural frames do not stand out sufficiently
on the painted wall, and the viewer is left with doubts as to where one scene
ends and another starts.

1. Simhala and fellow traders shipwrecked on island.

2–6. Witches transform themselves into beautiful women and entice traders
to live with them.
7. [destroyed] Simhala discovers imprisoned merchants in iron chamber and
is told the truth about the witches. Only the magic horse Balhala can save
them.
8. Simhala takes news to traders.
9–13. Balhala agrees to ‘fly’ traders to India, but if they look down with
longing at the witches, they will fall and be devoured by them. Simhala alone
reaches home.
14. Simhala’s witch-wife arrives in India with her son, and he rejects her as a
witch.
15–18. King of the land takes her into his harem.
19–23. She admits fellow-witches who devour king and nobles.
24–26. People decide to crown Simhala.
27–29. Simhala takes his army to the witches’ island and defeats them.

One organising principle that emerges from the study of the Simhala wall
is that of spatial geography; the artist appears to have arranged the episodes
of his story along one or other portion of the wall in terms of their
topographical relationships. The right third of the wall represents the
witches’ island and contains all events that occurred there, while the left
third is to be viewed as the palace of the king. The central area, containing
the conclusion of the story, as also earlier episodes relating to Simhala, is
space that belongs to Simhala. The upper zone contains Simhala’s home, the
central zone his coronation, while the lower area depicts the climactic
confrontation between Simhala’s army and the hordes of witches.
The second major narrative in Cave 17 that partakes of the nature of the
network is the well-known Vessantara Jataka, the various episodes of which
it is well-nigh impossible to mistake. The story unravels in a set of twenty-
five scenes along 45 feet of the left wall of the cave, directly opposite the
Simhala legend. Unlike the centralised placing of the climax seen in the
Simhala network, the vessantara story basically commences at the left and
moves gradually across the wall to the far right.35 Yet, its frequent
achronological and apparently random arrangement of episodes is curious.
For instance, the scene depicting the evil brahmana Jujuka asking Vessantara
for his children is placed high up on the wall above the doorway of the third
cell, while the actual gift of the children, with the pouring of water to
legitimise it, is placed at ground level beside that same doorway. The events
of the Vessantara story are, broadly speaking, arranged geographically, with
the left and right segments of the wall given over to palace scenes, and the
space between reserved for the events in banishment. The artist appears to
have conceived of the story in terms of thematic clusters which he placed at
an easily readable eye-level. To the right of doorway 1 is the gift of the ‘700s’
in which Vessantara gives away all his property; to the right of Doorway 2 is
the gift of the horses and the gift of the chariot, and higher up on the wall is
the gift of Madri; while to the right of doorway 3 are scenes relating to the
gift of the children. Most acts of giving (charity is the moral exemplified by
this tale) are in readily viewable locations; however, the gift of Madri and the
initial gift of the state elephant are placed above the level of the cell
doorways. The temporal displacement of the gift of Madri, to precede the
gift of the children, seems to be the result of the artist’s desire to place all
scenes connected with Madri in one place. The space to the left of doorway 3
contains various forest scenes revolving around Madri; the artist apparently
thought this to be the logical place for scenes of her donation.36
It would appear that in certain ways the artists painting Ajanta’s extensive
narrative networks failed to achieve narrative coherency. There are, for
instance, multiple points of entry into the story of Simhala, and thus too
many immediate options available to the viewer. If one walks around the
cave in the circumambulatory mode, such movement will bring us via the
conclusion of the story to its commencement. Assuming that
circumambulation had no relevance within a vihara, and that viewers
walked to the right Simhala wall immediately upon entering, they will first
view scenes on the witches island, then the conclusion of the tale, and finally
move to events in the king’s palace. Perhaps the artist’s intention was merely
to have viewers recognise the story and view its major episodes, rather than
follow its detailed denouement.
The principle of topographical location evident in these narrative
networks, or the ‘primacy of place over time’,37 is seen to be the guiding
principle in the large painted scrolls of the contemporary Rajasthani Pabuji-
ka-par tradition. John D. Smith’s study indicates that the central space of a
15 feet long par is occupied by the court of the protagonist, Pabuji; the left
and right extremes represent the territories of his chief enemies; while the
space between is occupied by the courts of various neighbours and allies.
Episodes 10 and 11 that follow each other in the temporal unravelling of this
extensive narrative are not located next to each other as they would be if the
artist had used the system of continuous narration. Instead, number 10 is to
the left of Pabuji’s court while number 11 is to the far right. Such a
placement is dictated by the geographical location of scene 10 in Patan, to
the west, and the succeeding scene 11 in Pushkar, to the east. Equally,
episodes placed next to each other may have no temporal connection, their
placement merely reflecting their occurrence in the same geographical
space. Scene 46 (news from Lanka presented to Pabuji), scene 59 (wedding
proposal presented to Pabuji, and scene 100 (enemy’s severed head
presented at Pabuji’s court) are temporally separated by considerable spans
of time; their contiguous placement reflects the fact that each occurred at
the identical location of Pabuji’s court. The topmost border of the par is
celestial space occupied by the various deities invoked during the course of
the story; their numbering reflects their temporal sequence in the legend of
Pabuji. While geographical space is tightly organised in the par tradition,
Ajanta’s Simhala legend, while making use of this principle, is less strictly
regulated by it. Within each of its three geographical areas, there exists a
network of movement in space and time.
 Narrative networks are in many ways similar to synoptic narratives,
except for the major distinction that they are substantially larger and
considerably more complex. They possess a distinct validity as an
independent category. Synoptic narratives, as the very term used to describe
them suggests, may be viewed in their entirety and taken in (if not
comprehended) from a single viewpoint. Narrative networks extend across
an expanded span that does not allow the viewer to see the conclusion of the
tale at the same time as he views the beginning. The viewers’ experience of
the two modes is quite dissimilar.
Is it possible to see anything in common between visual narrative
networks and the stream-of-consciousness apparent in novels like James
Joyce’s Ulysses where a reader can go fifty pages without finding
punctuation? What was Joyce’s, or for that Gabriel Marquez’s intention in
the use of such a method? And more importantly, what is the effect of such a
narrative upon its audience? Verbally, one feels submerged in the events,
surrounded by the characters, and practically one of performers. Visually,
the effect is strikingly similar. Viewers are not given the choice of stepping
away at the end of one episode; before they realise it, they have already
entered into the following one. The viewer becomes a participant in the
narrative.
It is interesting to speculate on whether the lesser importance assigned to
the element of time in India’s visual narratives is partially due to the system
followed by much of its literary narrative in which plots contain sub-plots,
and stories are contained within stories. In the compendium of animal tales
and fables known as Hitopadesa, for instance, the story-line of each of its
four books is interrupted so often by the narration of other tales that
listeners and readers have often forgotten the original story and have
certainly lost track of its time sequence. Its first section titled ‘Acquisition of
Friends’ is basically the story of the friendship that developed between a
crow, a mouse, a tortoise, and a deer, but its narration is interrupted by the
presentation of six other stories,38 that make it difficult to keep track of the
temporal and causal sequence of the main legend! This system of literary
narrative, oral and written, was a standard pattern in India, and the great
Mahabharata epic, for instance, proceeds along similar principles. Popular
oral tellings and dramatised versions of various works take great satisfaction
in presenting their numerous sub-plots. Versions staged in villages and small
towns often introduce a character whose sole purpose is to ensure that the
action is frozen in order to introduce a sub-plot; frequently such characters
enlist the support of the audience in urging the narration of such sub-
stories.
This literary model of pan-Indian occurrence is also found in a
specifically Buddhist context. The very life story of the Buddha, as presented
in the Mahavastu text, is interrupted at regular intervals by none other than
the Buddha himself who narrates legends of his previous lives.39 In the later
Pali Jataka collection too, on occasion stories are stalled in order to narrate
another tale.40 Being part of a milieu in which stories, written, oral, or
dramatic, rarely proceeded in a direct and uninterrupted fashion, ancient
sculptors and painters may have attached little consequence to the
presentation of temporality in their visual narratives.

VIEWER-RESPONSE

A work of art may be viewed from two angles—artistic and aesthetic; the
artistic aspect may be said to refer to the product of the artist, while the
aesthetic aspect refers to the effect it produces upon the viewer.41 The aspect
of viewer-response would find easy and obvious acceptance in the Indian
context. For more than two thousand years, Indian aesthetics has
propounded the theory of rasa that revolves around audience participation.
Indian theorists spoke of nine bhavas or sentiments and of the
corresponding nine rasas or emotions that these should arouse in its
audience. The nine rasas are Shringara (erotic), Hasya (amusing), Karuna
(pathetic), Raudra (furious), Vira (heroic), Bhayanaka (fearsome), Bibhatsa
(odious), Adbhuta (wondrous), and Shanta (quiescent). The theory of rasa
was originally propounded for the performing arts of music, dance and
drama, but extended to painting and sculpture by the sixth/seventh century.
In the Chitrasutra, the painting section of the Vishnudharmottara Purana,
the rasas are discussed in relation to painting, with a postscript to the effect
that its analysis applies also to sculpture. While ancient authors, adhering to
aesthetic principles, narrated stories with a view to producing certain rasas
in their viewers, the rules governing aesthetic response cannot be fully
applied to ancient sacred art. Whether it be the believing Christian viewing
the miracle of the fishes and loaves, or the Buddhist devotee viewing the
multiplication miracle at Sravasti, the immediate response was not an
appreciation of aesthetic quality. Rather, the ancient viewer looked upon
sacred art primarily as a demonstration of religious legitimacy.
It is of interest to discuss the reception of religious visual narratives and
the viewer’s response to the manner in which tales are presented. The
situation is somewhat complicated by our uncertainty regarding the exact
audience for whom the narratives were intended. Equally complex is the
problem of the manner in which that audience was exposed to viewing the
narratives. Probing into both these issues is necessary if we wish to address
the motivation that lay behind the lavish portrayal of visual narratives at
India’s Buddhist monastic sites.
The many Buddhist pilgrims and a few non-Buddhist visitors attracted to
sites of religious and artistic importance constitute a significant section of
the audience towards whom the narratives were directed. In such a context,
it is obvious that visual narratives could serve as a powerful proselytising
tool. For instance, the numerous animal fables in their Jataka guise may be
seen as part of a desire to present popular material for easy reading by
viewers. Assuming an audience composed largely of laity who came to pay
homage to the relics of the Buddha, we may theorise upon the manner in
which viewers would be exposed to the narratives portrayed at a stupa.
Devotees would have followed the prescribed rite of circumambulation,
viewing the narrative reliefs slowly and at leisure as they made their repeated
rounds of a monument that was located in the open air. Certainly, reading
the imagery of a stupa, unlike reading a novel, should not be assumed to be
a single continuous activity. Having completed the ritual devotional act of
circumambulation, we may assume that pilgrims, while following the basic
clockwise circling of the stupa, would often read narratives at random,
concentrating on obvious locations like entranceways, and places pointed
out to them by a spiritual guide or by fellow pilgrims. The existence of such
monk-guides is corroborated by the Mulasarvastivadin Vinaya’s account of
the wheel of life painted on the porch of a monastery, where a monk is
assigned the task of explaining its significance to visitors.42
 At the monastic caves at Ajanta, the problem of viewer-response is more
intricate. The narrative murals are located within viharas which are
primarily residential caves, even though each contains, in its back wall, a
shrine with a Buddha image. The rite of circumambulation may not have
held much meaning in such a cave. While it would be appropriate to pass
the monks, keeping them on your right, it would presumably have been
acceptable to walk directly to the shrine and then visit a monk, or an image,
at random. One may then set aside the mode of circumambulation when
considering the manner in which narratives unfold within Ajanta’s viharas.
Indeed, we have seen when considering the Vessantara Jataka on both sides
of a Sanchi architrave, that the circumambulatory mode has occasionally to
be abandoned even when viewing the reliefs that adorn a stupa.
A second problem at Ajanta pertains to the darkness that envelops the
interior of the viharas. Even the electric flashlights provided today in certain
caves by the Archaeological Survey of India do little more than light up a
limited section of a mural. When lit only by flickering oil lamps, it would
have required ‘strong eyes, great persistence and an excellent retentive
memory’43 to follow the narrative course. Like Trajan’s column, it is all there
though it is very hard to ‘see’ in its entirety. As a purely practical matter, one
wonders if any viewer was inclined to put in so much effort. Yet, the sense of
the narrative may have been clear to him through familiarity with the
subject. Perhaps identifying the story and gathering its general sense was all
that interested the viewer, or indeed was expected from him. Perhaps too, it
is in this context that one may best understand the thematic clusters in Cave
17’s treatment of the Vessantara Jataka. In a dimly lit cave where it is
impossible to view the entire narrative, clusters of episodes dealing with
various acts of giving becomes a logical way of presenting an extensive
legend. It is possible that the viewer was inducted into the experience of
understanding these ext ended narratives by a monk who held aloft oil
lamps and guided the viewer through the course of the narratives. In fact, it
seems near impossible for a viewer to manage without such direction. One
many perhaps surmise that the painted labels like ‘Sibiraja’ placed along a
king’s seat, or ‘Indrah’ beside a figure of that god, were added to serve as
prompts for the literate monk.
 It is, however, a distinct probability that Ajanta’s painted narratives were
not intended exclusively or even primarily for public viewing. In view of
their location within residential viharas, it is plausible that the murals were
primarily to be experienced by the monastic community of Ajanta.
Certainly, the Sanskrit verses from Aryasura’s Jatakamala, painted alongside
two narrative murals in Cave 2, one telling the tale of Ksantivadin and the
other of Maitribala,44 would have been appreciated only by the highly
educated ecclesiastic. It seems unlikely that any monk would have translated
the verses for worshippers while guiding them through the accompanying
visual narrative. The narratives may have been addressed to the resident
monastics with the intention that they benefit from the Buddhist message of
these tales. Contemplating the fortitude displayed by ascetic Ksantivadin
who, when his limbs were torn apart, only blessed the king who passed the
order, may have served as an inspiration to the Ajanta monks and facilitated
their spiritual advancement. In this context, one may recall a passage in the
Mulasarvastivadin Vinaya where the Buddha instructs Anathapindaka on
the themes to be used to decorate monasteries. Specially noteworthy is his
admonition to the monks to beware of damaging the murals while washing
or lighting fires within the premises since the paintings would prove
effective aids to meditation.45 Here is a clear indication that visual narratives
were addressed to an ecclesiastic audience. In other words, Ajanta’s narrative
networks may indeed have been experienced by competent ‘readers’ and
hence functioned in the manner in which they were intended.
Apart from their significance for lay worshippers and monks, the visual
narratives are also closely linked with the patrons and donors active at a site.
Gifts to a religious establishment were always made for the accumulation of
merit that would serve the donor and his family, not just in this life, but
more importantly in the next birth. A gift to a Buddhist monastic
establishment would bring merit whether or not donors were themselves
Buddhist.46 While plans for the construction of a monastery were often in
the hands of an ecclesiastic, it would appear that its narrative decoration
often did not proceed according to a pre-planned programme; rather, stories
that were particular favorites of individual donors found a place upon
monuments to which they contributed. A section of the monument may
have been set aside for Jataka tales; yet if three individual donors requested
the same Jataka, the evidence of the three architraves at Sanchi devoted to
the Chaddanta story suggests that donors were indeed permitted to have
their way. Narratives may, in fact, be connected in multiple ways with
patrons. The frequent portrayal of the Vessantara story, with its emphasis on
generous gifting (and the negative portrayal of brahmanas), may be
considered in this context. If carved by order of the ecclesiastic community,
as may have been the case at the Sanchi stupa, the Vessantara narrative was
possibly intended to present an exemplary example to wealthy pilgrims,
hoping thereby to induce them to make donations to the monastery! On the
other hand, if ordered by a wealthy donor, as in Ajanta Cave 17, it may have
been the donor’s intention to set up a parallel between his own gifting and
that of the great Prince Vessantara. Certainly the patron’s desire for glory
was a significant motivation for Ajanta’s magnificent picture galleries that
were mostly dedications from the aristocracy. The decision to decorate a
cave with stories focussing on royalty, as evident in Ajanta Cave 1 for
instance, may have had more to do with the patrons’ prestige and their
desire for reflected glory than with consideration for either visiting pilgrims,
or the monks’ use and the monastic community’s dormitory needs.
A final reflection is that certain narratives were probably placed upon
monuments because they needed to be there for the value of the monument
itself, rather than for reading by monks, pilgrims, or patrons. It is likely that
carved or painted guardians were placed at entrances to stupas and
monasteries because such figures were considered propitious. Probably, it
was deemed equally beneficial for these sacred monuments to be decorated
with Jataka tales and life scenes of the Buddha. When one contemplates the
many narratives that are placed beyond the reach of easy observation by
pilgrim, monk or patron, it seems that this factor may not be discounted. A
case in point is provided by the plaques decorating the six inaccessible roof
terraces of the Ananda temple at Pagan in Burma. In the absence of
stairwells, neither pilgrims nor monks could view the complete set of 537
Jataka plaques adorning the three lower terraces, nor the 375 plaques
portraying the last ten Jatakas placed against the three upper terraces.
Clearly, the placement of these narratives was not motivated by the aspect of
viewing; rather, in the Burmese Buddhist scenario, Jatakas seem to have
been a necessary and efficacious part of decoration. One may cite also a
practice of comparable significance in the oral narration of the Hindu epic
Ramayana at certain ritual occasions. Instead of serving as a focus in itself,
the recitation serves as a background chant that is scarcely attended to in
any active fashion. Yet, the narration is needed to add to the validity and
significance of the occasion.47 It is likely that the visual narratives that form
the subject of this study were intended for a diverse audience, and that
several of the factors discussed above functioned in combination, resulting
in the rich decoration of the early Buddhist monuments of India.
7

Archaeology of Early Temples in the

Chalukyan Regions
 HIMANSHU PRABHA RAY

This is a revised version of a Special Lecture delivered at the Indian

Council for Historical Research Southern Regional Centre, Bangalore
on 8 December 2006. The author gratefully acknowledges a grant from
the Jawaharlal Nehru University under the UPOE (University with
Potential of Excellence) Scheme that enabled fieldwork and purchase of
satellite images and thanks Dr Diwakar Parsi and Dr Sudha
Ravindranath RRSSC-B for their support and help.
In the elder days of Art
Builders wrought with greatest care
Each minute and unseen part
For the gods see everywhere.1

T HE SPECTACULAR TEMPLE COMPLEXES OF AIHOLE, Badami and

Pattadakal located along the 25 km long fertile valley of the Malaprabha
river (a tributary of the Krishna) in Bagalkot district have long been
admired for their distinctive architecture and sculptural exuberance. The
hills and rocky outcrops of reddish-golden sandstone provided the stone
used for the temples, and the large number of inscriptions has linked several
of these to the early Chalukya rulers.
It was along the Malaprabha river in north Karnataka that Pulakesi I
established his capital at Vatapi or present Badami and built a fort on top of
the sandstone cliff in circa 543 CE. The name Aryapura (Aihole) occurs in an
undated inscription of seventh–eigth century CE on one of the entrance
lintels of the east-facing temple of Gaudargudi, Aihole situated at a little
distance to the south-east of the Lad Khan temple. The inscription refers to a
donation by the citizens of Aryapura to Durga Bhagavati.2 Thus, the attempt
by the early Chalukya rulers to establish a base in the fertile Malaprabha
valley is undeniable. It is also evident that in the seventh-eighth centuries CE,
the Malaprabha valley acquired a cultural identity that has continued to
mould the lives of the communities both within and outside it. This cultural
identity included the demarcation of a territorial boundary within the
enclosed terrain of the valley, adoption of a new mode of worship in the
temple, the use of the Kannada language on coins and inscriptions, and
assumption of a sculptural programme based on the epics and the Puranas.
The objectives of this essay are twofold: one, to use the temple as an entry
point to underscore the cultural unity of the Malaprabha valley, which
emerged as the core area under the Chalukyas; and second, to highlight the
diverse social base of the religious shrines that dotted the landscape, thus
shifting the emphasis from viewing temples as instruments of political
legitimisation to understanding them as ritual instruments that provided
anchorage to the social fabric. In this endeavour, high-resolution satellite
images have proved to be an invaluable source in situating the temples
within their geo-physical habitat and it is the results of this collaborative
venture that are presented here.
Available studies on the temple complexes in the Malaprabha valley have
focused largely on chronology and style, but it needs to be stressed that
archaeological data encompasses not just standing monuments, but more
importantly it also includes an analysis of the location of religious
architecture within the social domain. Religious architecture is thus an
important indicator of interaction with diverse interest groups, such as
worshippers, ritual specialists, patrons, artisans, etc.3 No religious
architecture can survive without adequate maintenance and we do know
that shrines and other sacred architecture far outlived their patrons.
It is this social base of religious shrines that anchors them as crucial
indicators of cultural unity, both temporally and spatially. It is important to
underscore the role of religious architecture in demarcating cultural unity.
The classification of architecture within a racial-religious framework is an
enduring legacy of the colonial period and it is intriguing that this
connection has survived into the present. James Fergusson was a prominent
votary of the racial-religious framework and defined the limits of Dravidian
architecture by the spread of people speaking Tamil, Telugu, Malayalam and
Kannada.4 A characteristic feature of Dravidian temples, as he described it,
was the emphasis on ornamentation, and almost every nook and corner was
elaborately sculpted to the detriment of architectural planning. Another
‘failing’ of the region was that the Dravidian temple was a fortuitous
aggregation of parts, arranged without plan unlike European architecture,
which adopted uniform plan, changing only with the progress of time.
In this essay, we interrogate this framework by focussing on the social
history of the monument, which, however, cannot be understood solely in
terms of linear time and needs to be situated in both context and space. One
of the primary issues in a study of the past is not merely to locate sites
accurately but more importantly to relate these to other sites in the region.
This is especially important in the context of Aihole where temples are
located across the entire village and sit cheek by jowl along with houses and
other structures. It is also significant that the setting up of religious sites not
only altered the physical and cultural landscape of the Malaprabha valley in
the seventh–eighth centuries CE, but that these monuments continued to be
modified and reinvented. This essay locates intersections of these different
time-scales horizontally across the valley for an appreciation of the extent to
which the early religious shrines shaped the landscape.
Perhaps the first issue is to define the social history of the temple and to
delineate its various constituent factors. I would suggest that it is not
adequate to focus on origins, patronage, styles and iconographic
programmes of shrines, but what is essential is to trace their biography and
linkages established over time. The temple was not merely a place of
worship; it played a much larger role in society. For this essay, its cultural
network and communication with a host of audiences is relevant, such as the
lay worshipper, the ritual specialist, the scribe, the royalty and the elite, as well
as the artisan and the sculptor [emphasis added]. How did these diverse
communities relate to these religious structures over time? What
information is available regarding settlements in the region and the routes of
travel and communication? How did these settlements sustain themselves
and what was the nature of the resource base of these temple complexes? We
have addressed many of these questions elsewhere;5 here the emphasis will
be on the manifold audiences of the temple.
Worship in the temple from around the fifth-sixth centuries CE involved
bathing of the image, application of sandal paste, offering of flowers, lighting
of oil lamps, presentation of vocal and instrumental music and dances,
performance of sacrifices and offerings (bali-caru-satra), and maintenance
of servants of the gods. Recitation from sacred texts including the epics, the
Ramayana and the Mahabharata was a crucial part of ritual activities further
reinforced by representations of themes from literature in narrative panels
on temple walls. In several cases mahatmyas of major shrines appeared,
placing the temple within the larger domain of Puranic bhakti. In other
cases, pilgrimage became a significant link between the lay devotee and
sacred space. Another facet of this discussion relates to the presence of
donatory inscriptions at several religious sites, which have often been used
for classifying the patrons of shrines or as identification labels. But perhaps
the critical issue is of the interface between the community and the religious
shrine located within the macro context of the Malaprabha valley, and it is
this interface that we attempt to comprehend in this essay [emphasis added].
The essay is divided into three sections: in the first part, we discuss the
modern history of the temples from their discovery to their identification as
heritage sites. In the second part, we highlight details regarding the temples
themselves and their dynamic social history, while in the final section, the
focus is on the distinctive nature of the three complexes of Aihole, Badami
and Pattadakal. It is hoped that this essay will generate interest in the
utilisation of scientific tools for both understanding and protecting our
cultural heritage. We should start with a brief introduction to the discovery
of the temples in the Malaprabha valley.
THE DISCOVERY

The architectural ruins of Aihole and Pattadakal were first noticed and
photographed in the middle of the nineteenth century by Colonel Briggs.6
Meadows Taylor and James Fergusson published these photographs, but at
this time all that could be said about the temples was that these were pre-
tenth century CE structures.7 The connection with the dynasty of the
Chalukyas had yet to be established. In 1874, James Burgess visited Badami
and reported the story narrated to him by the villagers that the Jina image in
one of the caves was that of the raja, who had built the tank at the site.8 He
made an estampage of the inscription in Cave III and dated the record to
578 CE. Even though the caves and the temples at Badami were no longer
places of worship, Burgess records the annual pilgrimage at Banashankari, a
short distance to the south of Badami. A prominent structure at the spot was
a tank about 400 feet square, surrounded on all sides except the west by a
colonnade or covered walk.9
At Pattadakal, opposite the village of Katapur, Burgess mentions brisk
iron-smelting activity carried out by villagers as well as the continuation of
worship in the largest temple at the site. He also describes several megalithic
monuments termed ‘dolmens’ in the region, especially south-east of the
village of Aihole on a rocky hill, which we now know to be Meguti hill
(Fig.7.1). To the west of Aihole on rocky rising ground on the banks of the
Malaprabha, he refers to more such structures: ‘among a group of very old
temples in the same place is one raised on four unhewn pillars—but this
may have been the shelter of some bygone devotee’.10 This structure is in the
Galaganatha group of temples, which date stylistically to the second quarter
of tenth century CE.11 At the time of their discovery, however, the temples
had long been abandoned.
Henry Cousens visited Aihole and several other Chalukyan temples in
Dharwar district and focused his discussion largely on the Dravidian style of
architecture of these temples following Fergusson’s racial-religious
categorisation of architecture.12 Nevertheless, Cousens identified numerous
features that provided a unitary style to the temples, such as the use of
massive blocks of locally available sandstone, ‘an exuberance of lace-like
carving’, profusely decorated doorways, and the adoption of domical
ceilings. He suggested that the size of the temples was governed by the size
of the pillar, which was made of the most convenient length of block
available in the quarries.13 The temples were erected with no cementing
material and no clamps. Another significant observation related to the
sculpted panels was the fact that while the earlier temples carried narratives
from the epics, this practice was given up in the later tenth-eleventh century
structures, where the walls were decorated entirely with images of deities.14

Figure 7.1: Dolmen on Meguti hill, Aihole, in the vicinity of Jaina temple
Source: Author.
Within its (the old village of Aihole’s) crumbling walls, which enclose a
space little more than five hundred yards across, it has more than 30
ancient stone temples, all jostling one another for room. They are
embedded amongst closely-packed dwellings, and are as a rule,
desecrated and ruined, many having been converted into dwellings and
cowsheds . . . . Around the village, without the walls, are many more
dilapidated shrines, embosomed in prickly-jungle or standing uncared
for in outlying fields.15
This continued to be the state of affairs until the Ancient Monuments
Preservation Act was passed in 1904 under Lord Curzon, ‘to provide for the
preservation of ancient monuments, for the exercise of control over traffic in
antiquities and over excavation in certain places and for the protection and
acquisition in certain cases of ancient monuments and of objects of
archaeological, historical or artistic interest’; and this formed the basis of
interest in the monuments by the Archaeological Survey of India.
In the post-Independence period, research on the Aihole temples
continued, and scholars such as K. V. Soundara Rajan16 and M. Rama Rao
added several new dimensions to the study of the region.17 They questioned
earlier interpretations on the basis of fieldwork and adopted Sanskrit
terminology as cited in the Agamas and the Vastu Sastras. K. R. Srinivasan
further refined the analysis by Rama Rao, while continuing his methodology
of first-hand observation and reliance on Sanskrit treatises.18 Tartakov’s
research has not only defined the chronology of the temples, but has also
effectively shown the extent to which early beginnings impinged on
subsequent reflections in the context of the eighth century Durga temple at
Aihole. The temple’s apsidal form led James Fergusson to suggest that it was
a Buddhist structure subsequently appropriated for the worship of Siva and
by the 1860s the temple featured ‘as an inglorious, structural version of a
Buddhist caitya hall, appropriated by Brahmanical Hindus and buried under
rubble at a site of the ancient Chalukya dynasty’.19 As a result of subsequent
investigation and research not only on the plan of the temple, but also its
rich imagery, it is now evident that the Durga temple is one of nearly 150
temples built across 450 square km of the Deccan in the seventh-eighth
centuries CE, albeit the Durga temple is the largest and most lavishly
constructed monument dating to around 725–30 CE.
In the second half of the twentieth century, the focus was on a detailed
analysis of the sculptures at Aihole,20 a comparative study of the figurative
imagery of the temples,21 and a thorough analysis of the architecture of the
temples,22 as well as the scribes who have left inscriptions on the temples
walls and pillars.23 Adam Hardy suggested a different configuration for the
temples of north Karnataka by terming them as belonging to the Karnāta-
Drāvida tradition that flourished between the seventh and thirteenth
centuries, a period for which no Vastu Sastra texts survive from Karnataka.
In all about 200 temples are known: seventeen Early Chalukyan; thirty-one
Rashtrakuta, including those at Ellora and Sirval; 104 Later Chalukyan;
nineteen early phase, thirty-eight mature mainstream, forty-seven non-
mainstream, and seventy-two Hoysala.24 Following Stella Kramrisch, he
reiterated the adherence of religious architecture to underlying philosophic
principles as enshrined in the Upanishads, the ritual methodology of the
Brahmanas and the mythology of the epics.25
The architectural forms embody an Indian concept of manifestation of
the coming into concrete form of the divinity. But this symbolism of
formal structure is not entirely abstract, divorced from imagery.
Perhaps more than any other architecture, Indian temple architecture
notwithstanding a process of abstraction, which takes place, is
representational, made up of images. Temples make tangible the same
ungraspable truth that mythological and philosophical formulations, in
their own different ways, attempt to bring to human understanding.26
It would be worth analysing this principle through the temples at Aihole
characterised by diversity in architectural form. But to put this diversity in
perspective, we need to trace the location of religious architecture in
peninsular India and to especially analyse its oft-quoted link with the
political elite.
SOCIAL HISTORY

The period between the second–first centuries BCE and fourth century CE
was a period of creativity and experimentation as far as religious
architecture in the subcontinent was concerned. It was at this time that the
concept of the superstructure evolved and can be traced through several
formative stages as represented in reliefs from Bharhut, Sanchi, Mathura,
Ghantasala and Jaggayyapeta. A survey of the archaeological record
establishes the diverse nature of the sacred landscape in the early centuries
of the Christian era, with the Hindu temple coexisting with Buddhist and
Jaina shrines, memorial pillars and in many cases Iron Age burial structures.
Clearly these structures were sites of rituals and the performance of Vedic
sacrifices, but there is little evidence to support intervention by the royalty
or the establishment of temples as a result of royal generosity in this early
period.
It is also evident that the religious shrine did not emerge out of a void in
the sixth–seventh centuries CE, instead its beginnings may be traced to an
earlier period. Archaeological clearance work at Ambiger-gudi, Aihole
exposed a brick wall underlying the rear side of the sanctum of the stone
temple. The layer between the brick wall and the stone temple yielded
typical red ware bowls of the first–third century CE. Similar evidence was
also obtained in archaeological excavations at Pattadakal, where Rao
identified an underlying brick temple plan dated to third century CE based
on ceramic sequence, and suggested that it was a precursor to the earliest
stone temples of the Chalukyan period at the site.27 He further indicated
that a mound across the temple complex provided evidence of pre-chalukya
occupation in the area based on second-third century ceramics revealed in
the cutting.
Recent archaeological research has shown that the Malaprabha valley
formed a part of the larger settlement network in the Deccan in the early
centuries CE. The site of Cotta Chandor in Chandor district about 300 km
from Aihole towards the west coast was subjected to excavation for two field
seasons in 2002–04. The complete plan of a brick temple complex datable
from third to eleventh century CE was unearthed and five phases of
structural activity were identified. In the earliest Phase I, the temple was
cruciform in plan and this marks it out as perhaps one of the earliest
structures to adopt this plan of construction. Though three phases of
construction were identified, these were marked by continuity of religious
beliefs and in the last phase the sculpture of Nandi was added to the temple
complex.28
Vadgaon-Madhavpur is a forty-hectare site in the suburb of Belgaum,
which was discovered and excavated in 1945 by the Kannada Research
Institute, Dharwad, though the report has yet to be published, and again
from 1972 to 1977. A hexagonal stone pillar from the site carries an
inscription dated to first century BCE and records the performance of
Vajapeya and other sacrifices by a brahmana of the Kasyapa gotra. Based on
an analysis of ceramics from the site, three phases of occupation have been
identified dating from third century BCE to first century CE. The site has
yielded a rich hoard of coins and pottery, but relevant for this essay are the
find of two brick structures, one apsidal and the other consisting of a
spacious hall, which have been identified as religious in nature.29
The University of Mysore in collaboration with the State Department of
Archaeology excavated Banavasi near Sirsi in North Kanara district in the
1970s and unearthed an apsidal brick temple of the Satavahana period.30 A
number of mounds spread across one square km area along the river Varada
were explored and these yielded ceramics, such as Rouletted Ware and
Russet-coated Painted Ware, in addition to providing information on brick
structures. A square inscribed pillar bore the names of Kakusthavarman,
Santivarman and Mrgesavarman of the Kadamba dynasty and their victory
over the Pallavas.
Two brick temples were also excavated at Paithan in the Aurangabad
district of Maharashtra as early as 1937. The temples stand next to each
other, but their orientation is slightly different and there are also differences
in their plans and five separate phases of development that have been
identified. Subsequent excavations at the site have added to the information
available, but dating continues to be somewhat problematic, as much of the
information from the earlier excavations was not recorded. A broad time-
span of fourth to eighth centuries CE has been suggested for the two
temples.31
Another early temple site is that of Siddhapur, located on the right bank
of the Bhima, 56 km southeast of Solapur. Archaeological excavations
revealed two cultural phases, the first dated to first–third centuries CE and
the second to eleventh–twelfth centuries CE. A burnt-brick structure was
uncovered in the centre of the settlement, roughly rectangular on plan and
inside it a terracotta plaque of what has been identified as a syncretic image
of Lakshmi-Parvati was found along with a terracotta lamp and an incense
burner.32
Intense archaeological exploration at the confluence of the Krishna and
Tungabhadra rivers has brought to light a remarkable series of brick temples.
Two phases of structural activity marked the construction of temples at
Veerapuram in the Krishna valley. Mostly square on plan, these contained
pebble lingas and have been dated between the second to sixth centuries
CE.33 Similarly in Andhra, the earlier tradition of erecting memorials for the
dead re-emerged in the form of chhaya-stambhas or memorial pillars
coinciding with the advent of the temple.
Thus, by the fourth–fifth century CE, the temple had already emerged as a
well-defined unit positioned in proximity to water bodies, as also at
Nagarjunakonda, and it is within this changing sacred landscape in the
Deccan that the temples at Aihole need to be located. These temples are
situated along a north-south axis on the banks of the Malaprabha with
Meguti hill forming one end of the axis, past the caves and temples at
Ravalaphadi to Tarabasappa at the northern end. Among the three or four
small shrines in front of Ravalaphadi cave, the oldest is the west-facing
structure to the south.34 In the centre of the ancient town of Aihole stand
four structures: Gaudargudi temple or Lad Khan to the north-west;
Suryanarayana to the north; an ancient tank to the south, and Cakragudi
close to the southern edge of the tank.35 The presence of water bodies close
to Mallikarjuna temple at the foot of Meguti hill, Tarabassappa and
Huccimalliguddi is significant and is in keeping with the close association
between religious architecture, both Buddhist and Hindu, and the
community as evident from the archaeological record of shared agrarian
spaces at a range of other sites (Fig. 7.2).36

Figure 7.2: Bhutanatha group of temples at the edge of the tank, Badami
Source: Author.
The location of water bodies and temples indicates the beginnings of
temple construction on Meguti hill and its extension, the Ravalaphadi caves,
as well as in the centre of the village at Lad Khan where wells were dug for
the purpose of providing a source of water. The natural source, viz. the
Malaprabha river does not appear to have been a favoured early core area as
only one small temple in the Ramalinga group near the Malaprabha river,
viz., the Puspabhadra temple to the south of the village of Aihole dedicated
to the goddess Sri in the form of a lotus and oriented to the west to face the
river is dated to the first phase of construction.37
A characteristic feature of early religious architecture at Aihole is its
diversity in terms of religious affiliation and this is evident both from the
shrines at Siddhanakolla and the earliest constructions on Meguti hill. The
site of Siddhanakolla is located 2 km from the village on the Pattadakal–
Aihole road and includes a rock shelter at the site with a natural trough in
front, which collects and stores rain water. A Lajja-Gauri image is carved
inside the rock shelter, while two temples, i.e., Sangameshwar and
Siddeshwar of the Chalukya-Rashtrakuta period are located nearby. An
annual mela is held at the time of Sankranti when devotees come to the site
in large numbers and a modern temple has been built at the site.
The Jaina cave at Meguti hill at Aihole faces south-west and comprises a
garbhagrha, two side chambers, a common hall and a rectangular verandah.
Another two-storey Jaina shrine is located on top of the hill and bears an
inscription in Sanskrit on stone on the outside east wall. It mentions the
kings Jayasimha I, Ranaraga, Pulakesi I, Kirttivarma and Mangalisa (whose
horses were picketed on the shores of the oceans of the east and the west—
pūrva-paścima-samudratatositāśvah)—and Pulakesi II. The objective is to
record the erection of a stone temple of Jinendra by a certain Ravikirtti, the
composer of the eulogy who has attained the fame of Kalidasa and Bhairavi
by his poetry, during the reign of Pulakesi II in Saka year 507 (585–86 CE).38
The upper structure of the temple bears two inscriptions in Kannada, which
read Sri Rajakomara and Sri Bai.39 The temple walls are plain and there is an
absence of decoration on the pillars and door-frames as well.
On the northern slope of Meguti hill is a Buddhist vihara, which is part
rock-cut and part structural (Fig.7.3). It has an upper storey, and depicted
on the ceiling is a seated figure of Buddha, while scenes from the life of the
Buddha are sculpted on the door-jambs of the cave. A short seventh-century
inscription on one of the pillars of the ground floor mentions Mahendra of
the Pinthavadi (Pindapati) school and a disciple of Sthavira Ananda.40
Another rock shelter site is that of Bhadranayakana Jalihal, 3 km from
Pattadakal, which also has a spring and a water body in front. Inside the
natural cavern are several early Chalukya sculptures, such as a Saptamatrka
set carved in the wall, Ganesa, Siva, Parvati, Brahma, Visnu, Nandi, Vasuki,
Nagaraja, Lajja-Gauri, etc., while in front of the cave are eleven small
Chalukyan temples.
Figure 7.3: Buddhist vihara at Aihole
Source: Author.
Nine have an open porch and a vimana with tiered pyramidal
sculptures. One enshrines a standing Siva and Parvati with Nandi in a
style datable to 720 CE. The temples have no exterior niches. One of the
temples is interesting for its inscription by either its donor or builder. It
is referred to as a funerary shrine or karandaka (casket) bearing shrine
of Vikramaditya II who ruled from 733–45 CE. Nearby carved on a
boulder in a vertical rectangle that looks like a hero-stone are figures of
a seated king flanked by female cauri-bearers.41
The inscription from Gaudargudi temple dedicated to Durga Bhagavati
has been referred to earlier, thus making it one of the temples affiliated to
the goddess. The affiliation of other temples at Aihole is varied. For example,
in the Mallikarjuna temple, the floor of the garbhagrha, the pitha and the
linga are all carved from a single block of stone.42 The Durga temple is
among the largest Chalukya buildings and the most finished of all the
Chalukya temples (Fig. 7.4). The temple may have had a prakara, the pratoli
of which stands on the south. The north wall of the pratoli has a donative
inscription of the time of Vikramaditya II (733–45 CE), which records a gift
to a sun temple founded by a kumara, possibly a revenue officer. An undated
foundation inscription (recently discovered by Ramesh) states that the
kumara built this ālaya for Aditya. The characters of the inscription are no
later than 700 CE. On the north kapota face an engraved word ‘Jinālayan’ in
seventh–eighth century characters denotes a personal name, perhaps of an
architect, and does not denote a Jaina sanctuary as once supposed.43
Several phases of temple construction have been identified at Aihole
starting from the late sixth century, which continued until the twelfth
century with several renovations and additions. In the Durga temple, for
example, a step-well and a shrine were added in the later Kalyani-Chalukya
period (973–1198 CE), while Konti-guddi group received additional
structures in the Rashtrakuta period (757–973 CE). No stone temples were,
however, created at Aihole after the fall of the Kalyani Chalukyas at the end
of the twelfth century. Thus the early temples are distributed over an
extensive area and cover large parts of the Deccan uplands, while in the
middle of the first millennium CE there seems to be a concentration in the
Malaprabha valley. How is this concentration to be understood?

Figure 7.4: Durga temple at Aihole

Source: Author.
Historians associate Buddhism and Jainism with the growth of trade
resulting from the development of an agrarian surplus around the middle of
the first millennium BCE. This was followed by a decline of trade, rise of
feudal tendencies in Indian society, and the decay of towns from 300 CE
onwards.44 Buddhism is said to have declined and the post-fourth century
period is seen as marking a transitional phase between the sacrificial Vedic
religion and the emergence of Puranic worship characterised by the
migration of brahmanas from the towns and the development of tirthas or
sacred spots.45 It is suggested that at numerous feudal centres temples were
constructed in permanent material such as stone for the first time in the fifth
century CE.46 The period from sixth-seventh to twelfth- thirteenth centuries
CE in Indian history has often been termed as feudal and is said to have been
based on agrarian expansion and self-sufficient village economy, which
lasted until 1000 CE when foreign trade revived under the Arabs.47
Though not all historians accept the feudalism model and the
fragmentation of society that it suggests, yet the migration of brahmanas
and acculturation of tribal communities has generally been supported.
Chattopadhyaya, for example, explains this movement with the formation of
regional and sub-regional kingdoms and their requirement for
legitimisation.48 In effect then historians have explained the origins of the
temple in terms of the requirements of the local political elite or landed
intermediaries for legitimisation of their newly emerging status in a period
of urban decay, decline of trade, and agrarian expansion. There is unanimity
also on the association with migrations of brahmanas and their role as
priests in consolidating the new cults, frequently described as of Puranic
affiliation.49
In this essay, we question this association of temple with the ruler, as only
seven out of 140 temples bear an inscription recording a dedication of the
royal family of the Chalukyas and three of these are at Pattadakal, two at
Badami, one at Mahakuta and one at Alampur. What then was the role of the
kings, considering that very few temples provide evidence of being patronised
by the ruling elite [emphasis added]? A response to this question necessarily
involves an examination of the spatial context of these sites and the diverse
nature of these shrines, which were not only dedicated to a host of Hindu,
Buddhist and Jaina deities, but also owed affiliation to fertility cults such as
that of Lajja-Gauri. The beginnings of many of the sites date to a prehistoric
past and indicate continuity in terms of sanctity in a regional context. The
water bodies, such as tanks fed by underground springs continued to be
used over time and provided moorings.
It is evident then that rather than being at the mercy of the political elite,
the temple occupied a central place in the life of the community. The shrine
continued to retain this eminent position even after the crystallisation and
proliferation of a common political culture throughout all major regions of
the subcontinent from 350 to 750 CE.
A major source to investigate the history of the Chalukyas of Vatapi50
comes from their approximately 128 inscriptions and inscriptions of their
rival polity, the Pallavas, with whom they were perennially at war in south
India. It was along the Malaprabha river in north Karnataka that Pulakesi I,
performer of the Asvamedha and other sacrifices according to srauta rites
and born of Hiranyagarbha, established the capital at Vatapi and built a fort
on top of the sandstone cliff called the north fort in the sixth century CE.51
Dharādharendram Vātāpim-ajeyam-bhūtaye bhuvah
Adhastād-uparistāc-ca durggam-etad-acīkarat
A second inscription was subsequently discovered on the rockface at
Badami and deciphered as follows:
Svasti-Śaka-varsesu catuś-śatesu pancasasti-yutesu
Aśvamedha-ādi yajnānām yajvā śrauta-vidhānatah
Hiranya-garbha-sambhūtaś-ChalikyoVallabheśvarah52
There is no inscriptional evidence to associate Pulakesi I with any of the
religious shrines at Badami; perhaps the earliest evidence of a link is
provided by the Mahakuta pillar inscription, which refers to the
construction of a temple as well as a tank and the small mandapa
(devadroni) in the midst of the tank by Mangalesa, the son of Pulakesi.53
The religious development of the early Chalukya period was diverse and
included Jaina, Buddhist, Saiva and Vaisnava temples at a time when the
early rulers proclaimed themselves as paramabhāgavata. The 140 early
Chalukya temples are spread unevenly across 450 km from Banavasi in the
west to Satyavolu in the east, covering two major linguistic cultures.54 Based
on the presence of inscriptions in seven cases out of the known three dozen
examples, Burgess identified early Chalukya temples as a set and dated them
from late sixth to mid-eighth century CE. Cousens added to their number on
the basis of style, the style they shared and the styles of the neighbouring
temples with which they contrasted. These common visual characteristics
included locally quarried buff and gold or plum-coloured sandstone used in
their construction, but those identified as early Chalukyan were constructed
of significantly larger blocks than the other apparently later temples. Their
decoration was consistently more massive and less linear and they were less
ornate, with simpler plans and profiles. They carried more figure sculpture
and their figures were larger in size. Cousens enumerated the domical
ceiling and the profusely decorated doorways as the prettiest parts of the
early temples in Karnataka.55
The resources for the temple came from both agriculture as well as
trading activity, though detailed information on the nature of crops and the
yield is absent. There are nevertheless references to donation of land and
villages, for example, 25 nivartanas of land in the vicinity of Badami were
granted to a Vedic brahmana.56 The pillar inscription from Mahakuta refers
to the grant of ten villages to supplement the property of the temple. These
included among others Sriyambataka, Vrihimukhagrama, Kesuvolala
(Pattadakal), Kendoramanya and Nandigrama, possibly present
Nandikeshwar.57 Similarly 2 kulas of millet for every mattar of land owned
in the village of Nareyangal were granted to Virupaksa temple at
Pattadakal.58 The record of Pulakesi II from the Jaina temple at Meguti hill
mentions several villages that were gifted to the temple for its upkeep.
Interestingly it mentions several types of settlements, such as village, hamlet,
town, etc. thereby indicating the complex settlement pattern that had
emerged around the seventh century CE.59 At other times, donations took
the form of commodities, such as oil. A Kannada inscription from
Huchhcimalligudi of the period of King Vijayaditya records the grant of oil
to one who was evidently a priest of his temple and is dated to Saka era
630.60
Patrons and donors belonged to a wide social base. The largest temple in
the complex is that of Virupaksa and it is on its walls and pillars that many
of the inscriptions are engraved, especially in the east porch. The shrine is
asāndhāṛāvīṃāna with a large gūdhamandapa or closed hall with three
porches and a nandimandapa.61 Two inscriptions in old Kannada on
pilasters in the eastern gateway to the Virupaksa record that the Sutradhari
(assistant of the sthapati) Sri Gunda built the temple for Lokamahadevi, the
queen consort of the western Chalukya king Vikramaditya II in
commemoration of her husband having three times conquered Kanchi. The
inscription seems to be intended to record the re-admission into caste of the
artisans of the locality who had been proclaimed outcastes for some act,
which is not stated. The purport of this portion hinges entirely upon the
meaning to be given to the word ‘balligavarte', which is not a dictionary
word.62
An inscription on the north side of the porch of the Virupaksa temple
states that the queen-consort, Lokamahadevi, of Vikramaditya II confirmed
to the singers (gandharvavvarge) of the locality the enjoyment of the grants
and privileges that had been conferred on them by Vijayaditya. On the same
pillar there is another short ninth–tenth century record, which refers to a
certain Dhuliprabhu, who may have been a visitor to the temple.63 On the
corresponding back face of the pillar, there are two eighth–ninth century
Sanskrit verses in praise of Achalada Bharata, the author of a work on
dramatic composition.64
In the courtyard of the Meguti temple is a different kind of short Kannada
dedication of twelfth–thirteenth centuries on a monumental stone, which
shows a shift not only in the language used, i.e., from Sanskrit to Kannada,
but more significantly in the nature of the record itself. The stone is in three
compartments. The upper compartment contains a figure of Jinendra with
two yaksas above it; and on the right and left a kneeling figure, apparently of
a woman facing towards the image. The central compartment contains the
writing. In the lower compartment there is a Siva linga in the centre and
kneeling figures on both sides. The inscription refers to the stone as the
nisidi of Ramisetti son of Aibhasetti.65
The early trading links continued in the later centuries and were perhaps
consolidated in guilds, such as that of the Ayyavole as reflected in the eighth
century record from the Lad Khan temple at Aihole (Fig. 7.5), which refers
to:
 The grant given to the Five Hundred who constituted the great body of
Caturvedins (mahācāturvidyasamudāyamai-nuruvarkam): a dharana at
the ceremony of feeding a child with boiled rice, and at the festival held
when the first signs of life are perceived in the foetus; and at the
ceremony of tonsure. A gadyana at the ceremony of investiture with the
sacred thread and at the rites performed when the religious student
returns home after completing his studies. Two gadyanas at marriage
and at the ceremony performed on the first sign of conception and at
the celebration of an animal sacrifice. Three gadyanas at the celebration
of the caturmasya sacrifices and five gadyanas at the celebration of the
agnistoma sacrifice.66

Figure 7.5: Lad Khan temple at Aihole

Source: Author.
While this particular Kannada inscription refers merely to the mahajanas
as administrators of the agrahara, another epigraph on a beam of the
Gaudargudi temple at Aihole provides evidence of the mahajanas
collaborating with other local groups, such as the eight nagaras and the 120
uralis to make a grant to the goddess Durga-Bhagavati who later became the
patron deity of the Ayyavole guild.67 It is a twelfth century inscription
engraved on the walls of Carantimatha, Aihole that finally confirms the
common identity of the two groups when it refers to a Jaina merchant as the
son of the five hundred svamis of Ayyavole,68 thus emphasising the
commercial nature of the settlement at Aihole.
Gadyana and dharana referred to in the Lad Khan inscription were gold
coins that find mention in several other inscriptions from the region as well.
The Kharepatan plates of Rattaraja from Ratnagiri district of Maharashtra
dated to 1008 CE list gifts to the temple of Avvesvara built by Rattaraja’s
father and situated inside the fortifications. These included a measure
(gadyana) of gold from every vessel coming from foreign lands and a coin
(dharana) of gold from those coming from the coast, except those from
Chaul and Candrapura. Also located within the fortifications were
settlements of female attendants, oilmen, gardeners, potters and
washermen.69 Interestingly, the gadyana was a punch-marked gold coin
unlike the die struck coins, which were struck by their contemporary rulers
of north India. These are rare gold coins of ~3.5 to 4 g, which were later
adopted by various kingdoms of south India, including those who
acknowledged suzerainty of Chalukyas. These gold coins are broad and
circular in shape with various punches at the edge and a central punch
depicting a varaha or boar, the royal emblem of the Chalukyas.70 Thus by
the seventh–eighth centuries CE the early trading contacts had been
consolidated and established on a firm footing, while at the same time the
agrarian resource base had been expanded.

THE DEMARCATION OF A TERRITORIAL BASE

It is then evident that in the seventh–eighth centuries, the Malaprabha valley

acquired a cultural identity that it retained for at least the next four to five
hundred years. This demarcation of space included not only political centres
such as Badami, but also centres of pilgrimage like Mahakuta and places like
Pattadakal where royalty established temples. In addition there was a diverse
range of sites of worship and settlement that came to be linked through the
temples. This section attempts to define this newly emerging nuclear area.
At both ends of the valley, the hills close in and Badami lies some distance
outside the southern entrance to the enclosed terrain. Here, between the
cliffs the Chalukyan rulers made a huge tank and temples are scattered
around the tank. Aihole (Aryapura/Ahivolal) is located at the opposite
entrance to the valley and in the middle of the valley lies Pattadakal (ancient
Kisuvolal), where a complex of temples was built on the banks of the
Malaprabha, which at this point runs auspiciously from south to north.
Mahakuta lies 5 km east of Badami as the crow flies in a cleft in the hills, and
continues to be a place of pilgrimage, with most of the shrines built around a
sacred tank within a walled enclosure. Other significant places include the
site of Naganatha temple in the forested area; Nagral and Siddhanakolla on
the hills near Aihole; Sandur in Bellary district; Banavasi, the ancient
Kadamba capital; Kudaveli Sangam; Satyavolu, and Alampur, 150 km or so
to the east.
Badami is nearly 4.5 km from the Malaprabha river at the outlet between
two rocky hills on its north and south sides. The inscription of Pulakesi, the
founder of the Chalukyan dynasty occurs on the fortification and is dated
543/4 CE. A century later around 642 CE, the Pallavas conquered Badami and
it was only later under Vikramaditya I that Chalukya rule was re-established
in 654/5 CE in the Deccan-Andhra-Telengana regions. The nine rock-cut
temples date from the third quarter of the sixth century and the cutting of
cave III at Badami marks the culmination of this phase in 578 CE.71
A dam to the east of the present town between the bases of the hills forms
a large tank, which at present provides the supply of water to the town. All
along the north side of the Bhutanatha tank are two groups of temples and
two rock-cut bas-reliefs, while the hill behind them is a ruined fort, taken by
a British detachment under Sir Thomas Munro in 1818. At the east end of
the tank, there is the temple of the god Bhutanatha and there are some short
damaged inscriptions on the columns of the mandapa. The only one of
importance is a ninth–tenth century inscription on the outside of one of the
stones in the north wall of the temple, which records the gift of eight mattars
of cultivable land to the venerable Sridharabhutesvara.72
On the way to the northern fort, near the figure of Hanuman cut on the
rock an inscription reads: ‘. . . (of the village of) Mudugal, who was the
worshipper of the holy feet which are like lotuses of the goddess Kālikā and
of the god Sri Kamathesvara’.73
 In and about this rock fort are some temples also, but it is in the scarp of
the hill to the south-east that the cave temples are excavated. They are four
in number: the lowest on the west end of the hill is a Sivalaya or Saiva cave;
the next is a Vaisnava temple considerably higher up in the rock and to the
north-east of the Sivalaya; the largest, also Vaisnava, is still further to the
east on the north face of the hill, and the last is a little beyond it, but it is a
Jaina cave and of much smaller dimensions than the preceding Brahmanical
ones. It is on a pilaster in cave III that the inscription of Mangalisa, the son
of Pulakesi I is found.
Located midway in the enclosed terrain of the Malaprabha valley in the
region is Pattadakal, and one of the first shrines to be built at Pattadakal was
that of Papanatha. It seems that the Papanatha temple grew from a small
two-unit temple into the grand size it now has through expansion, re-
modelling and adaptation.74 An outer mandapa was added to the west of the
conventional one and together with the inner one added to its present
dimensions.
The Virupaksa, on the other hand, is the largest of all Chalukya temples
and consists of a sandharavimana with a large gudhamandapa or closed hall
with three porches and a nandimandapa.75 The Lakshmi pillar inscription at
Pattadakal dated 745 CE refers to three stone temples sponsored by royalty.
Vijayaditya erected the stone temple of the god Vijayesvarabhattaraka and to
the south of this, the consort of Vikramaditya II erected a stone temple to
Lokesvarabhattaraka to commemorate the ruler’s victory over the Pallavas.
To the north, the younger consort, Lokamahadevi erected the temple of
Trailokesvarabhattaraka. The Lokesvara and Trailokesvara temples are now
known as Virupaksa and Mallikarjuna respectively, while the Sangamesvara
temple bears on some of its interior pillars references to the temple as being
that of god Vijayesvara, Siva, lord of king Vijayaditya. The last named
temple, however, does not contain any inscription specifically associating it
with the king.76
In contrast to Pattadakal where the cluster of temples is located along the
river, most of the shrines at Mahakuta are within the confines of the walled
compound around the VisnuPuskarini (pond). Mahakuta contains remains
of nearly thirty–temples, twenty-two of these being attributed to the early
Chalukya period. The dharma-jayastambha, which Mangalesa claims to
have erected, is a free-standing pillar similar to the one in the Ravalaphadi
complex at Aihole. The message on the pillar records Mangalesa’s victory
over Buddharaja, a Kalacuri ruler and his subsequent gift of ten villages to
the god Mahakutesvaranatha made in association with and in the presence
of Durlabhadevi, the wife of Pulakesi I, and others. It refers to a god already
established at the site and previously patronised by Kirtivarma and includes
images substantial enough to be taken out in procession.77
Local tradition as recorded in the Mahakuta Mahatmya narrates that this
was the scene of destruction of the demon brothers Vatapi and Ilvala by the
saint Agastya. Near the steps on the west side of the tank there stands a
linga-shrine in the water with a pancamukhalinga, while in the courtyard are
a number of small lingatemples and a cell with an image of Lajja-Gauri. The
only two temples of any size in the courtyard are those of Mallikarjuna and
Mahakutesvara and it is the latter that has three inscriptions. The first is in
the porch of the temple and belongs to the time of the Chalukya king
Vijayaditya. It records that one of his concubines, Prāna-vallabhā Vināpoti,
bestowed the entire gift of a hiranyagarbha and having made a pedestal for
the god with rubies and having set up its silver umbrella, gave a field called
Mangalulle (of the measure of) eight hundred.
The second inscription is on the front or north face of another pillar in
the porch. It appears that the pillar was a votive offering of a certain
Puttimaninage and the inscription mentions a certain Mahasamanta named
Ereve, who seems to have issued some edict regarding the pillar or the
temple to the guild of the 1,966. The third inscription is on the east face of
the pillar inside the same temple and records a grant of three rice fields at
the village of Nandikesvara made to the temple by the Mahasamanta
Bappuvarasa in Saka era 856 (934–35 CE) when he visited it.78 These
inscriptions not only highlight the role of the royalty in patronising the
religious structures, but also more significantly link the shrines to the larger
social world of agrarian villages and trading networks and guilds. While the
tank at Mahakuta is praised in the Puranas and the Mahatmyas, water
bodies at other sites, especially at Badami become crucial sources of water
for the sustenance of the villages and communities located in the area. It is
these inter-linkages between varied religious, social and political spaces that
are highlighted when analysed within the larger context of the Malaprabha
valley and satellite images provide significant tools and techniques for an
appreciation of the spatial intersections in history and archaeology.
As discussed above, the temple complexes went through varied life
histories and at almost all the sites there is evidence for additions and
reconstruction. In the Durga temple at Aihole, for example, a step-well and a
shrine were added in the tenth–eleventh century CE, while the Kont-guddi
group received additional structures in the eighth–tenth centuries CE.
Within the temple an inscribed slab placed against a pillar contains a record
of Chamunda II (1169 CE), one of the Sinda chiefs. Though much damaged,
it begins with an invocation to Siva, indicating perhaps the twelfth century
change in the nature of the shrine to Saivism. At other shrines, such as the
Lad Khan in Aihole, a linga was placed in the temple in later times and a
large Nandi enshrined.79
This change in the nature of affiliation of shrines becomes more
noticeable over time. The early examples, for instance, were diverse and even
though no cult images survive in the shrines, their affiliation becomes
evident from the imagery of the temple. Of the important cave temples of
the early period, two were Saiva (cave I at Badami and the Ravalaphadi,
Aihole), two Vaisnava (caves II and III at Badami) and two Jaina (cave IV,
Badami and the Jaina cave temple at Aihole). The Meguti in Aihole was also
a Jaina temple, while the Upper Sivalaya at Badami was originally Vaisnava
and there was a two-storeyed Buddhist vihara at Aihole, built against the
hillside beneath the Meguti. Most temples of the later phase, however, were
dedicated to Siva and the three garbhagrhas of Jambulingesvara in Badami
housed the trinity of Brahma, Visnu and Siva. In addition there were
temples dedicated to Surya, the most prominent of these being the Durga
temple at Aihole, another temple in the same complex located next to Lad
Khan, as also the Malegitti Sivalaya at Badami.80 In addition to differences
over time, there is evidence of spatial variation. Ganesa images, for example,
are well-represented at Badami (cave I), Mahakuta and Aihole, but the deity
is absent in early Pallava rock-cut temples and first occurs at Kanci in the
eighth century.81 From the eighth–tenth centuries, the temples were
distributed widely across the Deccan. In addition to the continuation of
earlier shrines, new constructions also came up and especially significant are
the two relatively large Jaina shrines at Melagudi at Hallur and the Jaina
temple at Pattadakal, a few hundred metres from the complex dated to the
early Chalukya period. While settlements and temples at Badami, Alampur
and Mahakuta continued down to the Vijayanagara period without
disjunction, the same was not valid for other sites, such as the shrines at
Aihole. No new construction took place at Aihole after twelfth century CE
and developments in the subsequent period are not known. The final point
is evident—without the support of the community the temples at Aihole and
Pattadakal lost their relevance and survived only as relics of the past. In
contrast, those at Badami and Mahakuta underwent several changes,
reinvented themselves and continued almost uninterrupted. The strategic
location of Badami perhaps made it invaluable to succeeding generations,
but both Alampur and Mahakuta owed their survival in no small measure to
the Puranas and the Mahatmyas that extolled their virtues.
8

Ellora
Understanding the Creation of a Past

JAYA MEHTA

V ARIOUS SCHOLARS HAVE DOCUMENTED THE EARLY

explorations of cave sites like Ellora, Elephanta and Kanheri by
European travellers, painters and scholars. Partha Mitter talks of these
explorations in the context of colonialism and its impact on the study of
Indian art.1 Others like Vidya Dehejia look at the nature of representations
of the Indian views by British artists of the eighteenth century.2 This essay
goes beyond these studies in its twofold aim. The first is to view the
interconnectedness of this history of past approaches in order to understand
its sedimentation into present day studies. The various explorations by
European travellers, artists and scholars went on to influence the works of
British archaeologists. Much of these archaeological reports are accepted as
major archaeological documents to date. The ‘interconnectedness’ thus
highlights how these perceptions have shaped these archaeological reports
and further influenced major criteria for understanding Indian art and
architecture.
In the post-Independence period, various perspectives have emerged
concerning the criteria of classification of Indian art. These delve into
various issues like religious styles, dynastic classification, patronage, etc.
While some scholars like K. V. Soundara Rajan perceive Buddhist,
Brahmanical and Jaina art as sharply bracketed in terms of style and
chronology,3 others like Pramod Chandra point out that it would be
incorrect in India to say that essential differences among various styles
existed on account of religion.4 In her studies on Khajuraho, Devanagana
Desai points out many commonalities between various religious styles, both
in terms of architectural plans as well as sculpture.5 Regarding patronage,
scholars like Krishna Deva6 and Soundara Rajan7 continue to impose a rigid
dynastic system of classification, implying an immediate connection
between the dynasty and the art produced in the period. Yet, others like
Joanna G. Williams8 accept that the political connection is a secondary
aspect of the images and temples considered. Her work on Gupta art is not
limited to the territory of the Gupta empire and she does not expect Gupta
patronage for principal monuments. Pramod Chandra also points to the vast
number of monuments from ancient India built for those other than the
royalty.9 Thus, vastly different views on Indian art and architecture have
emerged revealing the inadequacies of our perceptions. We need to examine
which of these perceptions are legacies of the past and how they came to be
constructed in the first place.
Connected to this is the second aim of this essay: to move away from
established criteria and highlight alternative ways of looking at a site, with
sensitivity to the nature of varied contexts. When looking at a temple site, art
historians like Michael Meister have raised the point that they need to be
seen as fluid entities in time.10 A temple is often perceived as linked to a
king or community of one period. Yet, issues of worship, patronage and
control cannot be limited to one king or community. There often are multiple
claims involving the identities of many communities over time [emphasis
added]. Meister also states that a temple serves as a means through which
every king or community seeks to redefine the past and renegotiate the
present. The history of a temple site thus involves many layers of identities
and associations as well as conflicting claims to patronage, worship and
control. While historians look at issues regarding patronage in terms of royal
patronage and brahmanisation, art historians have discussed the notion of
patronage and its diversity. Hence, it is only through a multi-disciplinary
approach that we can get a more holistic picture of the past. This essay thus
has a bearing on several contexts, but the issues to be taken up here are only
the works of the British archaeologist James Burgess and the study of the site
of Ellora. In the first section of the essay, we look at the explorations by
European travellers and painters of various cave sites in the Deccan. Their
perceptions of Indian art and architecture are vital precursors of
archaeological studies of the region by Burgess. The second section
illustrates this connection with Burgess’s work and his criteria for assessing
Indian art and architecture, including whether it signified a colonial
enterprise. In the final section, the aim is to see how far the positive as well
as negative aspects of Burgess’s work are influential in the present-day
studies of Ellora. It highlights how various scholars have looked at Ellora
and presents alternative ways of studying the site.

RELATING TO MUCH MALIGNED MONSTERS

In his book Much Maligned Monsters, Partha Mitter highlights the Western
perceptions of Indian art, right from the end of the Middle Ages, to the
changed perceptions of the eighteenth century. Medieval texts like Marvels
of the East talked of Indian monsters, a notion inherited from diverse
classical sources like Pliny and Solinus. In these works, the Greeks
rationalised their instinctive fears in the invention of the monstrous races
that they imagined lived at a distance in the East, above all in India. The
Middle Ages thus received a large number of monsters from this source. The
concept of a monster was a means to describe what was not considered
natural or rational. Indian gods with their many arms were regarded as
monstrous because they looked unnatural and could not be rationalised.
With these stereotypes in mind, later European travellers took Indian gods
to be an example of multi-limbed pagan monsters known to exist in the
East. Indian gods thus acquired malevolent attributes in travel literature
after the Middle Ages.11
How would one then explain the totally contrasting appreciation of
Indian art and architecture from the eighteenth century? Mitter illustrates
the key factors behind this changed approach. The major backdrop to the
period was the development in Western aesthetics. The aesthetic movements
of the eighteenth century heralded a new awareness of non-classical
traditions, and the aesthetic notion of ‘sublime’ was placed on the same level
as beauty. The main features of sublime architecture like enormous size,
surrounding darkness, silence and solitude were all heightened in the Indian
cave architecture. This sense of awestruck wonder at the scale of conception
is evident in the fervent desire in travellers to take measurements, right from
the earliest times.12 The Western Indian caves thus came to be viewed as a
fine example of the ‘sublime’ and ‘picturesque’ nature of Indian architecture.
In my view, there would be yet another reason for the cave sites in Western
India to become a popular point of interest and study for the Europeans.
These sites also had a long history of exploration from the Portuguese times
onwards, as they were located around the first port of entry, that is, Bombay.
Thus, accessibility would be a major factor in making these sites get noticed
—more importantly—vis-à-vis other sites in Western India.
Mitter further illustrates the significant point of the nature of perceptions
towards Indian art and architecture at this point in time. Western analogies
and religious studies13 were utilised to classify Indian art and fit it into
Western art history. Elephanta was even identified as an antique Roman
temple14 and its Shiva Ardhanarisvara sculpture as the Amazon of classical
mythology.15 Mitter’s concern was to examine the issues linked to
colonialism and how these influenced the study of Indian art. At this point, I
would risk a digression by drawing attention to a very different later
nationalist vision of looking at Indian art. Tapati Guha-Thakurta studies the
way in which Orientalist and nationalist discourse in Calcutta at the turn of
the twentieth century tried to reconstitute a ‘great’ tradition for Indian art in
the past. Powerful equations emerged between ideas of ‘art’, ‘tradition’ and
‘Indian-ness’, as the image created of Indian art was integrally linked with
antiquity, religion and transcendental philosophy. Instead of the
longstanding image of the monstrous and barbaric, Indian art now receded
into the opposite pole of the abstruse and metaphysical.
The answer why the two issues raised both by Mitter and Guha-Thakurta
gain importance lies in the context of Burgess’s works. In the next section,
we look at how they reflected these issues. Do we look at his work as a
colonial enterprise or as reflecting sensitivity to its Indian context? How
much do the dynamics of religion and local identities enter his work? A
point very vital to notice regarding Burgess is that he was fully aware of the
travel accounts as well as the paintings of British artists describing Western
India. Of Elephanta, Burgess said in his Cave Temples of India: ‘The great
cave at Elephanta . . . has been so long known to Europeans, and has been
consequently so often described and so fully illustrated by Daniells and
others, that it is hardly necessary to say anything about it in the present
work.’16 As regards Ellora, Burgess is keenly aware of accounts of travellers
like Thevenot, Anquetil du Perron (1758), Sir Charles Malet (1794) and
Colonel Sykes (1820). He describes the twenty-four views published by the
artist Daniells, accompanied by plans by James Wales as ‘the most splendid
and accurate account of these caves, as a whole, which has yet been
published’.17 Burgess himself was an architect by training. Like the travellers
and painters, he was concerned with understanding architectural details.
Like the traveller, his focus was on the tremendous grandeur of the total
scale of conception and the wealth of exquisite details. Things which did not
fall into these categories were of minor importance to him. Of Kanheri,
Burgess remarks that the 109 separate caves were small and architecturally
unimportant though easy access from Bombay attracted Europeans visitors
to it for a long time.18 A similar approach is betrayed in his assessment of
the Jaina caves of Ellora.19 Archaeology, in fact, was in his view the history
of art, and he considered it his aim to provide a full illustration and history
of ancient and medieval architecture down to the decline of Muhammedan
styles.20
Equally significant was the impact of religious and epigraphic studies on
the work of Burgess. His focus on Buddhism was vitally linked to James
Prinsep’s discovery of the Brahmi script. In the joint volume Cave Temples of
India by Burgess and Fergusson, the latter wrote that when James Prinsep
deciphered the Asokan inscriptions:
The history of Buddha and of early Buddhism, which before had been
mythical and hazy in the extreme, now became clear and intelligible
and based on recognisable facts. The relation of Brahmanism and other
Hindu religions to Buddhism and to each other were now for the first
time settled, on a basis that was easily understood and admitted of a
logical superstructure raised upon it. When all this was done the
remaining task was easy. It only required that someone should visit the
various localities where the caves where situated, and apply, the
knowledge so amassed, to their classification (sic).21
 It is these influences and ideas that need to be kept in mind while using
Burgess ’ archaeological reports. It was an influence of a variety of sources
and factors that conditioned his perceptions of Indian art and architecture.
The works of earlier European travellers and artists were a predecessor that
shaped his own understanding of the cave sites of western India. The
religious and epigraphic studies of his time were extremely influential in
shaping his ideas on the relation of different religions in India. The next
section discusses the positive as well as the negative aspects of his work and
his criteria of understanding Indian art.

COLONISING THE PICTURESQUE

Much before the Archaeological Survey of Western India was set up in 1873
and handed over to him, Burgess had already done extensive work on
western India.22 From 1886, he assumed the post of Director-General of the
Archaeological Survey of India, succeeding Alexander Cunningham. Yet,
unlike Cunningham, he placed added emphasis on architectural survey
probably due to his early training as an architect. Unlike Cunningham,
Burgess found hardly any time to take much active interest in excavation
and his method of bringing out results was also different. Instead of
publishing periodical reports of his discoveries at the time they were made,
he kept his materials with him till he could present a complete monograph.
Within fifteen years, starting from his Western Indian post, he produced
twenty volumes of which seven formed part of the Archaeological Survey of
India, New Imperial Series.23
The influence of the parameters of Western art history is evident in
Burgess’s belief in architectural style as a means to chronology. Such was his
focus on style that he considered it to be in both Europe24 and Asia as ‘. . .
infallible, yielding results that admit no dispute, and which are more
generally relied upon by antiquaries than those derived from any other
sources’.25 Rather than trusting inscriptional data, Burgess had faith in
tracing the development of styles as the true guide to chronology. Burgess
continued with the trend of an evolutionary framework in his
understanding of style: ‘Cave architectural features, as in all true styles, all
over the world change according to certain laws of progression’.26 To him,
architectural style was not only the reliable means of establishing
chronology in Europe but the standard criteria for Indian architecture as
well.
Along with this stylistic focus came his imposition of religious and
epigraphic studies onto the study of Indian art and architecture. Burgess
claimed that the peculiarity of the western Indian caves was in affording a
vivid illustration of the rise and progress of three great religions of India. He
said: ‘The caves show how Buddhist religion rose and spread and its later
form became corrupt. Consequently it was superseded by the nearly cognate
form of Jainism and antagonistic development of the revived religion of the
Brahmins.’27 As mentioned earlier, this focus on Buddhism had a lot to do
with Prinsep’s discovery of the Brahmi script. Architectural space was thus
understood in the light of a predetermined and fixed concept of Indian religion
[emphasis added]. Buddhist, Brahmanical and Jaina religions were posited
as distinct periods inimical to each other; their representation in caves was
sharply bracketed in terms of style and chronology.
Burgess’s treatment of religious art reveals a marked bias towards
Buddhist art as being the true representative of western Indian cave
architecture. In terms of both architecture and comparisons of Buddhist
paintings and Brahmanical sculpture, the accent is on the superiority of the
former. In stylistic terms, Buddhism was correlated to simplicity of sculpture
and paintings whereas Brahmanism was linked to ornate sculpture. Burgess
found Buddhist stories to be manifested through paintings, while
Brahmanism, which had no stories but single images, manifested itself in
sculpture. He remarked, ‘If paintings of Buddhist caves had existed fully,
they would afford illustration of history and mythology far more complete
than the limited scope of Brahmanical sculpture.’28 His focus on Buddhism
is also evident in his intensive study of the stylistic variations in Buddhist
sculpture and architecture at different sites.29
Burgess also pointed to the notion of diffusion of styles through conquest
by rulers. He attributed the change of style at Ellora’s Cave 10, Kailasa, to the
change of dynasty, that is, with the rise of seventh- century Rashtrakutas and
their king, Dantidurga. According to him, the Pattadakal temple at Badami
made during Dantidurga’s reign was identical in style, plan and dimensions
to the Kailasa temple at Ellora. This he took to be a confirmation of the fact
that Kailasa was made in his reign; maybe by the same architect.30 He thus
linked the notion of style to dynasty, the eclectic nature of inputs evident at
Kailasa notwithstanding.
Though Burgess essentially linked Indian art to notions of antiquity and
religion, he probed little to explore the aesthetics of Indian art, its unique
style and sensibility. He did not intend to go into the regional context and
understand how these examples of architecture met the needs of the
community. Instead, in 1905, Burgess was already complaining of
antiquarians like Coomaraswamy who, ‘first formed theories more
wonderful than natural and then tried to make facts and inscriptions
support them’.31 He continued to ignore Indian sources like texts and
traditional architects. Burgess’s assistant Henry Cousens attempted to use
Indian architectural terms to focus more precisely on the detailed structure
of a temple, something that was difficult with inept European terminology.
Burgess dismissed the effort on the grounds that ‘few of these terms are to be
found in our lexicons, and their precise forms can hardly be controlled out
of India’.32 The study of Indian architecture for Burgess was really for non-
Indians by non-Indians. E. B. Havell attacked Burgess and Fergusson in this
failure to approach Indian architecture from an Indian point of view. Havell
also rejected the sectarian division of styles and the persistent habit of
looking for foreign influences. While Burgess represented the dry
archaeological approach that focussed on documentation, Havell
exemplified the nationalist vision, which focused on an aesthetic
appreciation of Indian art that highlighted its unique and exclusive aesthetic
philosophy.33
Burgess’s contributions thus lay in the vast amount of documentation of
sites in Western India and his analysis of the development of Buddhist
iconography in different contexts. Yet, his imposition of religious studies
onto architectural space created notions of essentialism tied to community
and creed. Instead of viewing Indian monuments in their larger context, he
tied them down with unchanging conceptions of what was properly
‘Buddhist’, ‘Hindu’ and, ‘Jaina’. He did not take notice of the alternative that
styles spread as craftsmen moved from one regional workshop to another,
and instead stressed the simplistic notion that styles were diffused through
the interaction and conquests of rulers. His imposition of Western art
history parameters onto Indian art and architecture was evident in his
exclusive focus on comparison of styles. The colonial nature of his
undertaking was reflected in his lack of sensitivity to local texts and Indian
architectural terms. He thus reflects a very different approach from the
nationalist vision of art historians like A. K. Coomaraswamy and E. B.
Havell, whose concern was the community sponsoring the monument. Both
the positive and negative aspects of Burgess’s work are significant in their
impact on future studies of Indian cave sites. The next section will deal with
these issues in the context of one site, namely, Ellora. The aim shall be to see
how such issues that were a part of Burgess’s approach continue to influence
present studies on Ellora.

SITUATING ELLORA
Both the positive and the negative aspects of Burgess’s work have influenced
major approaches towards the study of Ellora. The primary archaeological
report of the Archaeological Survey of India on Ellora is by James Burgess,
and various scholars vouch for the accuracy of its factual data34 and its
comprehensive nature, despite the fact that Burgess’s proposed hypothesis
was later discarded. Burgess posited that the caves were excavated in a
sequence from south to north and that determined their chronology. This
represented a neat sequence where all the Buddhist, Brahmanical and Jaina
caves followed each other in excavation. Later scholars have corrected this
with a ‘fan theory’ in which early caves have been posited near the centre,
from where later excavations spread out.35 It is also vital to keep Burgess’s
aims in mind when he was writing the reports. As mentioned earlier, his
studies sought to chart the rise and fall of India’s three religions. This
application of religious studies onto architecture takes as its basis the
presumption of decline and notions of ‘essentialism’ tied to creed. This
limited context of looking at a monument resulted in a sectarian
understanding of Ellora.
 These restrictive categories need to be abandoned in favour of notions of
regional styles cutting across communal lines [emphasis added]. Yet, scholars
like Soundara Rajan have limited themselves to a narrow, sectarian
understanding of Ellora. According to him, ‘The fact that even the Buddhist
caves at Ellora have a certain richness of detail and figure carving, would go
to show that their designers and architects were having their primary
tutelage under Brahmanical art undertakings, which is so different from the
sobriety and fluidity of architectural contours of Buddhist art.’36 That these
undertakings were a part of a larger regional art activity is not considered
here. More importantly, Soundara Rajan continues Burgess’s point of linking
the diffusion of styles to political dynasties. For him, the development at
Ellora after the Brahmanical phase is explained by ‘rulership (sic) of the area
which had the chance and circumstance of displaying its proven expertise in
rock art in the region’.37 In his book Cave Temples of the Deccan, he
illustrates the entire art activity in the region, and at Ellora, only in terms of
dynastic patronage of the Calukyas of Badami and the Rashtrakutas. The
basic approach behind these criteria of classification also shows Burgess’s
tremendous influence. Dynastic patronage at Ellora is proven through an
exclusive focus on the comparison of styles, notwithstanding the fact that no
epigraphic or inscriptional data links any other dynasty except the
Rashtrakutas to the site.
Most studies on Ellora are characterised by Burgess’s obsessive focus on
formal style, and scholars are continuously seeking to establish chronology
through it. Very few studies have tried to look at alternative approaches such
as criteria other than style and an understanding of the local context and
associations. Instead of Burgess’s dry archaeological approach of
documentation, the focus needs to be on the connection between the
monument and community for whom it was created. Scholars like G. H.
Malandra and Michaela Soar represent a shift in this direction. Michaela
Soar investigates the local myths and associations at Ellora by studying the
local mahatmya. This results in the very significant understanding of the
age-old goddess tradition of the region that became an important part of all
the religious sects of the region. G. H. Malandra illustrates a shift in
understanding through the use of approaches other than style to examine
Ellora’s larger network.38 These findings represent a qualitative shift away
from the assumptions of British archaeologists like Burgess and studies after
him that followed his approach.
In this attempt to move away from previous standard conceptions of
Ellora, we require a fresh outlook towards the site. It is vital to keep in mind
that apart from a few sites like Mathura, we do not have sites where three
different religions are represented. Instead of narrow and insular approaches
of looking at only the Buddhist, Brahmanical or Jaina caves at Ellora, it
would be better to examine the shared culture at Ellora. In what ways could
this shared culture of Ellora be defined? Firstly, this would be in terms of the
location shared by the three religious sects. The natural landscape of Ellora
is marked by the vital characteristics of a tirtha, with its mountains, forests
and pools. The site also had tremendous sanctity due to its association with
the river goddess as pointed out by the mahatmya. The local mahatmya
points to the sacred associations with the site for the Brahmanical sect.
Though we do not have similar texts for the other two religious sects,
excavation of the caves indicates that it was an important spot for the
Buddhists and Jains as well. This sacredness of association with Ellora is very
evident in the fact that three such different religious sects desired to fit into
one landscape. The second interesting aspect of this shared culture lies in the
goddess tradition, which cuts across all religious styles. The strength of the
local goddess tradition is visible in the concerted effort of all religions to
include the representations of the goddesses in their imagery. Even Burgess
remarked that the singularly striking feature of the Buddhist sculpture at
Ellora vis-à-vis other Buddhist sites is the presence of numerous female
deities.39 Similarly, five sets of Saptamatrikas are found in Brahmanical caves
14, 21, 22, 16-B and 16 and indicate their concentrated development as a
theme. The Jaina caves also depict many Jaina goddesses like Ambika,
Indrani and Saraswati. The preponderance of female deities is striking as a
common point among all religious styles. This fact points to a strong
following of the local goddess tradition in the area. In order to involve these
communities and their local associations, the goddess tradition was
incorporated into all the religions represented at Ellora [emphasis added].40
Also important to notice is that local text itself points to the multiple
levels of associations of the site. The local mahatmya of the Skanda Purana
seeks to extol a Sivalaya Tirtha, yet alludes continously to the sanctity of the
local river goddess Ila. Michaela Soar indicates the nuances of these
associations. She points out that the origins of the river Ila is attributed to
two reasons—one as being born from the goddess and the other from the
local king. The story linking the origin from the goddess points to the local
goddess tradition being appropriated into the mythology of Saivism. The
second reason for the origin of the river Ila was that it was brought forth by
the king, Ila, just as Bhagirath brought Ganga. Thus, the goddess was
brought into the mythology and tradition of kingship. She also highlights
the second point explained by the text, that is, the establishment of the city
of Elapura by a king of the same name. According to this a mythical king Ila
enters the Sivalaya Tirtha and renames it after his own name, after adopting
the local worship of the linga and gifting land to brahmanas and holy men.
The important point here is that he is keen to take over the earlier allegiance
of the site to the goddess, and he adopts the local name of Ila, the local
goddess and the river.
Thus, apart from the religious affiliations as indicated by Ellora’s
architecture, I would like to draw attention to how the text points to several
layers of narratives relating the identities and associations of various interest
groups to Ellora. The important point raised by J. E. Cort that communities
express different sense of identities by allegiance to different deities is very
significant in this context.41 The multiplicity of associations to the site of
Ellora point to the probability of communities expressing multiple forms of
allegiance to the deities represented. The goddess tradition and the
subsequent religions would thus be a part of a shared culture space at Ellora
even for the interest groups coming to the site [emphasis added].
We now turn from the multiplicity of associations to probe the shared
culture in the architectural space of the site. T. V. Pathy has demonstrated
the need to see the architectural interactions among the Buddhist, Hindu
and Jaina cave temples at Ellora. In terms of architecture, he underlines the
development of the cruciform plan in both Buddhist and Brahmanical caves.
The Buddhist cave Tin Taal also demonstrates the transition where cells are
substituted for sculpture, like in the nearby Brahmanical Dasavatara cave.
Interestingly, the facade of the Dasavatara cave is identical to that of the
Buddhist Don Taal and Teen Taal caves. Even Burgess noted that
Brahmanical caves like Cave 27 show strongly Buddhist cave features.42 The
Jaina cave Indra Sabha presents an interesting phenomenon. Its ground plan
resembles that of Kailash while its upper floor is similar to Buddhist plans
like that of Caves 2 and 3.43 Similarly, features like cushion capitals, the
ghatopallava motif on pillars and the vidyadharas are motifs in all religious
cave styles. In terms of sculpture, apart from the common representation of
female goddesses, scholars have also pointed to borrowings in imagery.44
Thus what we see at Ellora is a lot of imitation, replication and sharing of
a common vocabulary of art. These are not three distinct cave styles
developing in isolation from each other or representing a contestation.
Scholars like Soundara Rajan seek to amplify the contestation aspect by
pointing to the way the Brahmanical authorities appropriated Dasavatara,
which had been a Buddhist cave. Yet, one needs to bear in mind that the
Buddhist structures of this cave remain unaltered. These include the vihara
in the open court, the entire ground floor and even various Buddhist pillars
and pilasters on the upper Brahmanical floor. In case of contestation and
appropriation, the Buddhist features of the cave could have been easily
demolished. Instead, all have been allowed to coexist—it is possible that the
Buddhist cave was left incomplete.
All of the above examples strongly suggest that different religious sects
shared a space and had local associations with it. It is very probable that they
shared between them the allegiance and patronage of various communities.
Instead of routinely attributing patronage of the site to ruling dynasties, we
need to recognise the role of wider interest groups. There is a clear lack of
inscriptions to determine Ellora’s patrons. It is only the Brahmanical cave
Kailasa whose patronage is directly linked to the Rashtrakuta king,
Krishnaraja I, through a Baroda copperplate. Yet, Kailasa as a structure
displays an extremely eclectic nature of styles. It does not cater to any one
taste and reflects various idioms.45 Even on Kailasa’s patronage by
Krishnaraja I, Gary Tartakov suggests he was probably responsible only for
the cave’s inauguration.46 With the help of examples from various texts,
Tartakov proposes that the patron who conceives a project and lays the first
brick gets the credit for it, whether he completes it or not. As regards other
caves, Rashtrakuta patronage has often been further extended to the Jaina
caves. Yet the Jaina caves contain no references to royalty but to individual
donors.47 More importantly, the imagery of the seventeen Bahubali figures
in the Jaina caves of Ellora suggests a north-south network. M. N. Tiwari
and K. Giri argue that though the Bahubali cult was popular in the south, it
is only later in the ninth century that his images acquired the status of the
Jina in the north.48 This process of Bahubali images of the north acquiring
attributes of the Jina is first noted in Ellora. It is thus from here that the cult
and imagery travelled up north.
The important point here is Ellora’s proximity to the trade route on which
Ajanta lay, linking the cis- and trans-Narmada zones. This ancient trade
route passed via the ranges of Ajanta and Balaghat, the centres of
Daulatabad, Aurangabad and Jalna to touch the Godavari at Nanded. This
was an important trade route since very ancient times,49 linking the
Godavari valley with political centres like Ujjayini and Vidisha. That Ellora’s
network also included the east is demonstrated by G. H. Malandra through a
comparison of Buddhist iconography at Ajanta, Ellora, Sirpur and
Ratnagiri.50 Ellora must thus not be understood in terms of locally limited
sects or dynastic equations. Instead, it is clearly important to dwell on its
wider network as a religious centre, sacred site, and centre of art activity. It
demonstrates linkages with the north, south, and east through iconography,
inscriptions and architectural data. As a thriving centre surviving on not just
royal patronage, the role of the pilgrims and a wider network became
immensely important.
Certain other issues regarding Ellora not addressed here but making
important points of enquiry are about its architectural space. The various
overlaps among religious styles must be inspected more carefully in terms of
their relation. In this regard, the paper ‘Sharing, Intrusion and Influence’51
by Gary Tartakov and Vidya Dehejia seeks to refine the uncritical
vocabulary often used to discuss stylistic continuities. The point is that
instead of using vague terms like ‘influence’, more specific terms need to be
utilised in different contexts.52 It would help immensely to analyse Ellora’s
space with such a critical vocabulary to gauge the relations of the
communities involved. Another approach, which is required, is to analyse
the placement of images. Devangana Desai has done this very effectively in
the case of Khajuraho.53 An attempt has been made to grasp the meaning
and context of the images through the study of the placement of divinities
and their configurations in the architectural scheme of the temples. Perhaps
such an exercise in the case of Ellora could provide more fruitful results
than the repeated focus on the sectarian affiliations of the cave sites.
The important point is to grasp what the site meant to the interest groups
involved, which influenced the course of its creation, preservation and
renovation. Not only did building activity continue at Ellora from the fourth
to the tenth century CE, later texts continued to associate it with different
groups. The local mahatmya of the Skanda Purana from the area also points
to the multiple claims to the site. Not only is there a strong tradition of
worship of the river goddess, it is reiterated as a Shiva shrine and the local
king seeks power by association with both. P. V. Ranade54 points to texts
from the thirteenth century CE as well as local Marathi folk tales55 that
further illustrate interesting associations with the site. The thirteenth-
century text Lilacharita mentions the visit of a saint, Sri Chakradharswami,
to Ellora in 1268 CE. It gives a fanciful account of the strange deities of the
place that take form of spirits and creatures. Interestingly, the authorship of
the caves is ascribed to a carpenter, Kokas.56 Kokas is also found in the text
Kathakalpataru of Krishna Yajnavalki (c. 1470–1535 CE), which is the
earliest Marathi legend on the excavation of the Kailasa cave. It narrates the
story of a king, Yelurai, who is struck by a disease due to the sins of his past
birth. His queen vowed that if her husband was cured she would not eat till
she saw the pinnacle of a Shiva shrine made. When the king got
miraculously cured he got a carver called Kokas Vadhai from Paithan, who
with his 7,000 artisans erected the tower of Kailasha within a week. The
carver is then said to have created other caves; and the settlement of Yelura
thus came about due to the efforts of King Yelurai. It is interesting to note
the persistent reference in both texts to the carver Kokas Vadhai who, as
Ranade points out, is a legendary figure of Indian literature found in many
Jaina texts and is epigraphically recorded in the central Indian inscription of
1155 CE.57 The important point is that till as late as fifteenth and sixteenth
centuries, there were claims and counter claims to the creation of the site.
These texts thus point to continued associations with Ellora and its
creation. That it continued to figure in the popular imagination of regional
communities illustrates the multiplicity of connections with it. An
important digression needed at this point is the study of four Western
Indian temples done between 1996 and 1998 by John E. Cort, L. A. Babb and
Michael W. Meister.58 Instead of clear-cut continuities of caste and
community patronage, the temples present a different picture. The Sacciya
Mata temple in Osian, for example, had for long been tied to the origin
narrative of the Osval community, since Sacciya goddess was described as
their kuladevi (associated with their ‘home village’). But the author’s
research indicated that the Osian temple was largely in the hands of Bhojak
priests, and patronage had been more from local Bhatti Rajputs, Mahesvaris
and other castes than from non-local Osvals. The Dadhimati Mata temple in
Goth-Manglod presented a similar case. It had presumably always been
worshipped by the Dahima brahmana who lived in the regions around the
villages Goth and Manglod, between which the temple is located. But
research indicated it was ritually important to Bidyasar Jats, and its control
may have been in the hands of non-Dahima brahmana priests until early
twentieth century. Also, the construction of pilgrim rest houses and various
other facilities by Mahesvaris, Khattis, Svarnkars and other castes at the site
indicates that members of these castes attached importance to the
Dadhimati temple.
These temple studies present cases of continuous contestations of
patronage, worship and control. These multiple claims involve differing
understandings of the relationships between the deity and castes, between
worshippers and worshipped, between the donor and the donated. The
temple site thus means very different things to different communities and
serves to redefine their past and renegotiate their present. Michael Meister
points to the life a monument has after its making, and argues that the
monument forms a ‘series’ as it interacts with different contexts through
time. It is imperative to look at Ellora as such a series that interacts with
different communities in time. In terms of the shared culture of architectural
space at Ellora and the text and tales associated with it till a later period, we
see several layers of narratives relating the identities and associations of
various interest groups to Ellora. The monument would thus be a part of
multiple identities of communities involved, and the community concerned
would re-invent each association. It is vital to weave the monument into the
lives of the community without which it continues to have no meaning. This
is only possible by connecting Ellora to a larger network with the interest
groups coming to it, and examining their multiplicity of associations with
the site. Through this, we can examine Ellora not as a static monument
linked to one period or a singular dynasty but a fluid entity that constantly
grew as a ‘series’ as it interacted with many communities over time. In this
understanding of the site we need to re-look at a whole series of associations
with Ellora prior to Burgess’s reports. The local texts and Marathi literature
present an Ellora that is vividly alive in the minds of the local community.
This relation and continuity is lost with Burgess and the subsequent trend of
documentation.
The important point is that we are not just looking at the art historical
remains at a site. By divorcing the gamut of associations made with the site
over time, we lose much of the dynamism of these architectural remains.
The essay has thus sought to move beyond strategies of research and
documentation established by Burgess and to enlarge the scope by including
a study of local myths and legends as well as the reinvention of sites over
time. It is only through this different dimension to the study of architecture
that we can answer vital historical questions regarding the interaction of the
monument and the community.
 9

Theory and Practice of Painting

Introduction to the Viṣṇudharmottara
stella kramrisch

He who paints waves, flames, smoke, and streamers fluttering in the air,
according to the movement of the wind, should be considered a great
painter. VDhP 43.5.28
Painting is the best of all arts, conducive to dharma, pleasure, wealth,
and emancipation. It gives the greatest pleasure, when placed in a
house.VDhP 43.5.38

T HE VIṢṆUDHARMOTTARA IS A SUPPLEMENT OR APPENDIX to

the Viṣṇupurāṇa. Part 3 of the Viṣṇudharmottara gives the fullest
account known of the various branches, methods, and ideals of Indian
painting. It deals not only with its religious aspect but also, and to a far
greater extent, with its secular employment. It proclaims the joy that colours
and forms and the representation of things seen and imagined, produce.
Speaking of artistic representation in relation to religion, it points out their
mutual limitations:
Vajra said—The Supreme Deity has been described as devoid of form,
smell and emotion and destitute of sound and touch—so how this form
can be (made) of Him?
Mārkaṇḍeya replied—Prakṛti and Vikṛti (come into existence)
through the (variation in) the form of the Supreme Soul. That form of
Him (which is) scarcely to be perceived is called Prakṛti. The whole
universe should be known as the Vikṛti (i.e., modification) of Him,
when endowed with form. Worship and meditation (of the Supreme
Being) are possible (only when He is) endowed with form . . . . The best
position of the (Supreme) Soul (however) is to be imagined without
form. For seeing the worlds (He) possesses eyes closed in meditation . .
.1
 This concession made, life in its entirety becomes fit for artistic
representation, and the realm of imagination is as close within the reach of
the artist as nature that surrounds him, for tradition guides him in the one
case and observation checks and inspires him in the other.
The Viṣṇudharmottara admits in several places that it is but repeating and
compiling from older sources. These being lost to us, our text represents the
earliest exhaustive account of the theory of painting. Its date can be
ascertained partly from chapters copied from earlier sources, and partly
from a custom of setting up statues to renowed personages with which the
text deals.
Vyāsa, the reputed author of the Mahābhārata, was worshiped as a deity.
Directions for making the image of Vyāsa are given in the chapters dealing
with image-making, following the chapters on painting.2 Directions are also
given for making images of Yudhiṣṭhira, Bhῑma, Arjuna, Nakula, Sahadeva,
Krishna; of Sumanta, Jaimini, Paila, Vaiśampāyana (the four disciples of
Vyāsa); and of Devala, Balarāma, Rukmiṇῑ, Satyabhāmā, Śāmba, and
Aniruddha.
At the time of the compilation of the Viṣṇudharmottara, the Mahābhārata
must have been held sacred to such an extent that not only its heroes, but
also its so-called author and his disciples were worshipped as deities.
Vālmῑki, too, was worshiped as a deity and the Viṣṇudharmottara also tells
how to make his image.3
The complete book of the Rāmāyaṇa is not anterior to the Mahābhārata,4
and since the Mahābhārata in its present shape is assigned to a period
between 200 and 400 CE, the Viṣṇudharmottara cannot date earlier than the
fifth century CE. This date is also evident from that of the Viṣṇupurāṇa, of
which the Viṣṇudharmottara is an appendix. The Viṣṇupurāṇa bases its
genealogical accounts on those of the Bhaviṣya, Matsya, and Vāyu. ‘The
Vāyu borrowed the Bhaviṣya's augmented account about or soon after the
year 330 or 335 CE.’5 The Viṣṇupurāṇa cannot, therefore, be earlier than the
second half of the fourth century CE.
The lower age limit of the Viṣṇudharmottara must, however, be pushed
still further. For Part 3, chapter 27, dealing with colours, is borrowed
verbatim from Bharata’s Nāṭyaśāstra, and the number of rasas in the
Viṣṇudharmottara is nine, while only eight are dealt with by Bharata.
The upper age limit of the text may be conceived from an iconographic
peculiarity of its pantheon. In it every god, hero, philosopher, or sage finds
his place somewhere round the central figure of Vishnu, the God Supreme.
Yet, no mention is made of Śaṅkara, who up to the present day is
worshipped in the form of an image. Had Śaṅkara existed at the time of the
compilation of the chapters on painting, the author would have incorporated
him into the Vaiṣṇava pantheon in spite of his being an incarnation of Śiva.
The chapters of the Viṣṇudharmottara dealing with painting must have
been compiled in the seventh century, contemporary with the latest
paintings of Ajanta; and so we get acquainted with the theories prevalent at
the time of the full maturity of their practice. But it must not be forgotten
that our text is but a compilation and that its recipes and prescriptions go
back to a remoter past. Valuable as these various kinds of information are,
we have to be aware that, like all theories, they are derived from, and
subservient to, the practice. They left every freedom to the artist, to work, as
the text says, ‘according to his own intellect’. What Śrῑ Kumāra modestly
states in his Śilparatna (verse 14) also holds good for our text: ‘I describe the
methods of Chitra for the benefit of the ignorant.’
Painting in ancient India, especially in the Gupta age, was of great
importance in the life of the citizen. The interest taken in pictures varied
with the education of the spectator. ‘The masters praise the rekhās . . .
(delineation and articulation of form), the connoisseurs praise the display of
light and shade . . ., women like the display of ornaments, to the rest of the
public richness of colours appeals.’ The artists, therefore, should take great
care that the painting may be appreciated by everyone. There was ample
opportunity for contemplating and appreciating paintings.
From the great hall built by the Bodhisat—according to the Mahā-
Ummagga-Jātaka6—painted with beautiful pictures, and the subterranean
palace of the same Jātaka, with its stucco-coated walls, bearing paintings of
the splendour of Sakka, the zones of Mount Sumeru, the sea and the ocean,
the four continents, the Himavat, the lake Anotatta, the vermillion
mountain, the sun and the moon, the heaven of the great kings with the six
heavens of sense and their divisions—to the picture gallery (cittāgāra) in the
royal pleasure grove of Prasenajit, king of Kosala, where many people used
to go, among them the bhikkhuṇis, who were forbidden to do so7—to those
many portraits and landscapes painted by artists of the royal and the servant
class alike, as mentioned in the Ratnāvalῑ, Raghuvaṁśa, Śākuntala, and
Uttararāmacarita—we see an unfading delight taken in the magic and the
sensuousness of painting. Wherever there was a festival painting enhanced
its mood—‘from the city-gate to the palace, and from the palace to his own
house, on both sides of the road, he erected lattice-work, and covered all
over with mats, covered all with pictures, scattered flowers upon the ground,
hung flags and banners’.8
As permanent or temporary decoration, on the floors, on the walls and
ceilings of private houses, palaces, and temples, and in the streets, paintings
instructed and enlivened the mind of the public. Even religious teachers
used painting as the most popular means of communication that could be
understood by the illiterate and the child. ‘There is a class of Brahmanical
teachers, known by the name of Nakha. They make a (portable) framework
upon which they cause to be drawn a variety of pictures, depicting scenes of
good and evil destinies, of fortunes and misfortunes, and causing the labels
to be inscribed: “By doing this deed one attains this”, “By doing that, one
attains that”—thus showing different destinies, they wander about with
these pictures.’9
That every cultured man had in his house drawing board and a vessel for
holding brushes and other requisites of painting is evident from Vātsyāyana’s
Kāmasūtra.10 But one should not have a painting by one’s own hand in one’s
house, says the Viṣṇudharmottara. Certain objects are only to be painted in
private residences, suggesting love, gaiety, and peace, while the supernatural
and the terrible aspects of life were reserved for the walls of temples and
royal audience halls.
The paintings were executed in various types; wall paintings, pictures on
board, and on canvas were equally frequent (cf. Kāmasūtra and
Viṣṇudharmottara). The latter were sometimes in the shape of rolls,
exhibiting continuous representation. Such a roll was spread out by a spy of
Cāṇakya before the people in Candanadāsa’s house and was exhibited by
him with songs.11 The figures were of oblong, squat, and rounded shape and
the Viṣṇudharmottara accordingly enumerates four types of depictions: (1)
satya—true, realistic, as we may say, of an oblong shape; (2) vaiṇika, which
may mean lyrical, of squat shape, and (3) nāgara, of the citizen, gentry
pictures of rounded limbs; while the fourth type simply is (4) miśra, mixed.
In connection with wall paintings the Viṣṇudharmottara also alludes to
floors inlaid with precious jewels. From the Śilparatna, on the other hand,
we know that dhūlicitra, powder painting, familiar to Bengali ladies as
alpona, was applied as temporary coating of powdered colours on a beautiful
piece of ground.12
Since painting took such a wide part in secular and religious life, it was
only natural that legends were invented to explain the origin of the art. The
Viṣṇudharmottara13 gives a long account of how the sage Nārāyana, in order
to put the apsarases to shame, created the most beautiful nymph, Urvasῑ, by
drawing her outline with mango juice. The Citralakṣaṇa14 again tells us how
King Nam-grags a Jigs t’ ul, ordered by Brahma, painted the likeness of a
deceased son of a brahmana; whereupon Brahma made it come to life, and
having thus defeated Yama, restored the son to his father. In either legend
the origin of the art of painting is seen in the outlining of a human figure15
for the purpose of creating a living human form. This reconstructed origin is
magic and non-aesthetic. A similar notion is to be found in Bhāsa’s
Svapnavāsavadattā,16 where King Udayana and Princess Vāsavadattā, with
whom he had eloped, are, though absent, married by their parents in
effigies, by drawing the portraits of the two on a board. These instances
prove that the artist draws from his memory when visualising a portrait.
With this tradition another has to be compared. A Tibetan text of the
eighteenth century, the D’pag bsam ljon bzair17 tells us how King Udrayasa
of Raura caused a picture of the Buddha to be made by taking a reflex of the
figure of the Daśabala as his model. This picture has become known under
the name of l’u lon-ma (derived from the water). The Śilparatna accordingly
defines painting as what bears a resemblance and looks like a reflex in a
mirror.18 The imitative and the imaginative origins of pictorial
representation, therefore, were felt as equally true. The Viṣṇudharmottara
quotes the Urvasῑ legend, yet the text never grows tired of pointing to things
seen as ever-fresh sources of artistic inspiration, when dealing with the dṛṣṭa
(i.e., with things seen). But not only the twofold origin of painting in
observation and imagination was theoretically known to the authors of the
various treatises; the Viṣṇudharmottara, moreover, introduces its chapters
on painting with a discourse, where Mārkaṇḍeya instructs King Vajra that
without a knowledge of the science of dancing, the rules of painting can
scarcely be understood [emphasis added]. In another passage, the
observation of nature and of the rules of dancing are indicated as the
ultimate resources of the painter. This does not mean that the positions of
dancers have to be painted. None of the nine positions of the treatise on
painting in the Viṣṇudharmottara coincides with any of the 101 positions
explicitly described in Bharata’s Nāṭyaśāstra. What is meant by the
derivation of painting from dancing is the movement in common to both
these expressive forms; it asserts itself in purity through dancing; it guides
the hand of the artist, who knows how to paint figures, as if breathing, the
wind as blowing, the fire as blazing, and the streamers as fluttering. The
moving force, the vital breath, the life movement (cetanā), that is what is
expected to be seen in the work of a painter, to make it alive with rhythm
and expression. Imagination, observation, and the expressive force of
rhythm are meant by the legends of the origin of painting to be its essential
features.
The Viṣṇudharmottara clearly distinguishes between dṛṣṭa and adṛṣṭa, the
latter comprising things invisible or rarely to be seen. The dṛṣṭa, things that
are seen easily by ordinary mortals, excel in what we call landscape painting.
The hours of day and night and the seasons are described (chapter 42).
There we find a close connection of mood and time, which reached its
height in the Rāgamālā pictures, where season, hour, emotion, and music
became fused as painting. At the same time details are observed with the
same sincerity as we find in the pictures of Dutch masters, for instance, in
the description of the drinking place. The light effects sought to be produced
show a very sensitive reaction to optic effects, as the faded light of the candle
in the morning dawn. Yet we are told that moonshine should be shown by a
Kumuda flower in full bloom, and sunshine by drawing creatures suffering
from heat. In one instance, atmospheric effects are observed, while in the
other, the behaviour of one object or the other, reacting to the atmospheric
change, is represented suggestively. This interest in the living individuality of
the single forms of nature gives to Indian landscape the charm of
storytelling.
Side by side with the naïve joy in the variegated forms of nature, we learn
that rivers are to be represented in human shape, as was the case with them
in Greece, but they should stand on their vāhanas, their knees should be
bent, and their hands should hold full pitchers. What an amazing
association of ideas! The personification of the river put again into action as
an ordinary human being, bending down under the load of the full pitcher
of water drawn from the river. This versatility in visualising abstraction and
actual action replaces the mere observation of nature. That seas should have
water depicted instead of a halo, or that an artist should show a pitcher to
suggest a tank, a conch shell in representing a conch shell, and a lotus flower
in representing a lotus flower, once more points to a matter of absorbing
interest, namely, the single form of nature exercised on the mind of the
artist. He rendered it faithfully. Yet, where large appearances like whole
rivers and seas, or landscapes with rising suns had to be painted, he took his
refuge and delight in introducing personifications or such actions of some
members of the scene appropriate to, and indicative of, their surroundings.
The Indian artist never took the world at a sweeping glance.
Observing the details of appearance, the author of the Viṣṇudharmottara
describes the different types of men. Country people, the nobility, widows,
courtesans, artisans, wrestlers, soldiers, and so forth, are vividly described in
movements, habits, and features peculiar to their class, while, at the same
time, most of them belong to one of the five standardised types of men,
called (1) Haṁsa, (2) Bhadra, (3) Mālavya, (4) Ruchaka, and (5) Śaśaka.
Their respective measures should be 108, 106, 104, 100, and 90 aṅgulas in
contradistinction to the measurement given in the Bṛhatsaṁhitā, where the
relation of sizes is inverse, 96, 99, 102, 105, and 108 aṅgulas respectively. A
detailed description of the five types is given there.19
The figures may appear in various positions, of which nine are the leading
attitudes:
(1) Ṛjvāgata—the front view;
 (2) Anṛju—the back view;
(3) Sācῑkṛtaśarῑra—a bent position in profile view;
(4) Ardhavilocana—the face in profile, the body in three-quarter profile
view;
(5) Pārśvāgata—the side view proper;
(6) Parāvṛtta—with head and shoulder belt turned backwards;
(7) Pṛṣṭhāgata—back view with upper part of the body partly visible in
forlorn profile;
(8) Parivṛtta—with the body sharply turned back from the waist upwards,
and lastly
(9) Samānata20—the back view, in squatting position with body bent.
In a further passage thirteen positions are enumerated—which, however,
is clearly an interpolation.
These positions are obtained in painting with the help of kṣaya and
vṛddhi, decrease and increase, that is to say, the science of foreshortening.
Foreshortening (kṣaya and vṛiddhi) and proportion (pramāṇa) constitute
with regard to single figures the working of observation and tradition. The
law of kṣaya and vṛddhi was as intensely studied by the ancient Indian
painter as was perspective by the early Italian masters. Pramāṇa, on the
other hand, was the standardised canon, valid for the upright standing figure
and to be modified by every bent and turn. The Haṁsa measure is detailed
in full; it is of the uttama navatāla type.21
Though not distinctly mentioned, it is evident that the body of a deity
should be of the size of a Haṁsa. Ṛṣis, gandharvas, vidyādharas, daityas,
dānavas, ministers, brahmanas, Saṁvatsara (i.e., Śiva), and the family priest
(purohita) should be of the size of a Bhadra. Yakṣas, prostitutes, and Vaisyas
should have the size of a Ruchaka. Kinnaras, rākṣasas, nāgas, and domestic
women should have the size of a Mālavya, while the chief among men and
the shudras should be of the size of a Śaśaka (chapter 42).
 Together with pramāṇa and kṣaya and vṛddhi, the knowledge of modeling
or shading (varttanā) was fully known to the Indian masters of old. It is
stated to be threefold:22 (1) patraja—cross lines; (2) airika—stumping; and
(3) vinduja—dots (chapter 41). How much observation and technical
experience is needed to state their results in such clear terms will be
understood. Yet, ‘whatever the artist represents he should avoid placing one
figure in front of another’ (chapter 43). Once more, one notices the same
counteraction of abstraction and observation as in the case of landscape
painting. A logical employment of kṣaya and vṛddhi would have implied
oversecting. But the Indian artist cherishes every single form as a whole, as
containing all he has to express and as containing the whole of nature. So he
cannot bring himself to cover and hide one of its parts.
As for the employment of colour, we have the following information from
the Viṣṇudharmottara and other sources. Five primary colours consist of
white, yellow, black, blue, and the colour of the myrobalan, according to our
text (chapter 40), or white, red, yellow, black, and green, according to our
text (chapter 27) and the Nāṭyaśāstra; white, red, yellow, soot, and dark blue
or black (śyāma) according to the—Śilparatna. The Abhilaśitārtha Cintāmaṇi
knows as pure colours white made of conch shell, red (prepared from red
lead or from alaktaka juice, i.e., lac, or from red chalk—gairika), green-
brown (haritāla), i.e. sulphureted arsenic, and black from kajjvala (soot,
used as eyeblack).23
The mixing of these colours is left to the ingenuity of the artist. He may
coat them with lac and resin. Colouring substance are gold, silver, copper,
mica, deep coloured brass, red lead, vermillion, tin, yellow orpiment, yellow
myrobalan, lac, indigo, and some kind of iron oxide. Prescriptions for the
preparation of these vegetable and mineral colours are given in the
Śilparatna.24 Red chalk, for instance, ought to be ground on stone and left
covered with water for a day; red lead also ought to be ground and covered
with water for half a day; red arsenic, however, should be made into powder,
but it should not be covered with water. Then the colours should be ground
once more and again put into water for five days. Afterwards they should be
mixed with the exudation of the nimba tree, then at last they are ready to be
applied on walls and in pictures.
 Gold leaf should be divided into finest leaflets, and then ground with a
well-smoothed stone; afterwards it has to be mixed with water and with a
small amount of sand. When this paste is well ground, it should again be put
into water so that all dirt may rise to the surface. The gold paste should then
be pressed, mixed with vajralepa and applied with suitable brushes. There
should be nine brushes for every colour.25 When dry, the artist should rub it
slowly with a boar tusk until it becomes bright. Then again he may place
over this very fine gold leaves and rub them with hard cotton. The same
prescription is to be found in the Abhilaśitārtha Cintāmaṇῑ.
As to the preparation of the ground, buffalo skin has to be boiled in water
until it becomes soft as butter. The water then has to evaporate and sticks
have to be made of the paste and dried in the sunshine. This hard plaster is
called vajralepa.26 If then boiled in a mud vessel with water, it will make fast
any colour with which it is mixed. If mixed with white mud, it has to be used
as coating for the wall, in three layers, each layer being allowed to dry before
the application of the next. On this the painting may be applied.27 Previous
to the process of vajralepa coatings, the wall has to receive a thick coating,
consisting of bricks, burnt conches, and the like, powdered according to the
Śilparatna; and mixed with sand, the watery, preparation of molasses, and
drops of the decoction of mudga (phaseolus munga, mung pulse) amounting
to a fourth part of the mortar powder. Into this, smashed ripe banana fruits
have to be put, also a fourth part of the amount of the mortar. After three
months, when this mixture has dried up, it shall be ground again. Then it
must be mixed once more with molasses water until it gets the touch of fresh
butter. In this stage it should be applied to the wall with a spoon, the wall
having been cleansed with coconut fibres and having been sprinkled for
some time with molasses water. This is the twofold process by which the wall
is made ready for the drawing and the application of colours.28
The outlines ought to be drawn in yellow and red colour, as a rule
The painter should think of the proportionate size of the thing to be
painted, and think of it as having been put on the wall. Then calculating
its size in his mind, he should draw the outline marking all the limbs. It
should be bright in prominent places and dark in depressed places. It
may be drawn in a single colour where comparative distinction is
 required. If depressed places are required to be bright; jet black should
be used.29
The modeling capacity of the outline is also described in the
Vṛddhaśālabha jikā. This outline has to be filled with the first colour wash
which, as a rule, is white, but according to the Viṣṇudharmottara also may
be green.
The colouring of things seen, says the Viṣṇudharmottara, is true to nature.
Great emphasis is laid on the thousandfold mixtures of colours left to the
imagination of the artist, and on the light and dark shade of every tone. The
range of colours must have been wide enough to render with subtlety the
local colour of objects. The different tribes and castes of India are thus
distinguished as dark, when belonging to the Pulindas and people of the
South, to Pa chālas, Śūrasenas, and Mag dhas, to Aṅgas, Vaṅgas, and
Kaliṅgas, to shudras—to sick men and to family men engaged in toilsome
work. Śakas, Yavanas, Pallavas, and the Vālhikas should be predominantly
white, and so are the twice-born and the kshatriyas, kings, and prosperous
people. Those who are oppressed by evil stars are of dark colour. And it is
also clear that evil-doers ought to be of a dark complexion. The colour thus
has partly descriptive and partly suggestive significance. The dṛṣṭa and adṛṣṭa
hold their sway; symbol and illustration are amalgamated into an expressive
language, keenly alive to all those visual impressions that are on a small
scale, obtrusively finite, and seem to carry their meaning expanded within
their outlines, as local colour. But this ambiguity of the colour in its
suggestive and descriptive faculty was clearly kept apart. Taken in a
naturalistic and descriptive sense, the sky or the atmosphere has to be
painted as almost without any special colour.30 The sky, on the other hand, is
of the colour of the blue lotus and wears a garment of that colour, if
represented as a statue, when it should carry the sun and the moon in its
hands.31 Colour symbolism underlies not only the painting of statutes
which, according to their sāttvika, rājasika, and tāmasika aspects, had to be
painted white, red, or dark, but was respectively selected for rasa-citras, the
pictures of emotions, which, according to the Śilparatna formed a group by
themselves, distinct from the realistic paintings that were resembling what is
actually seen in nature and looked like a reflex in a mirror.32 Each rasa
(emotion) had to be painted in its expressive colour, the śṛṅgāra (erotic) was
of śyāma hue, the laugh-exciting (hāsa) of white colour, the pathetic
(karuṇa) of grey colour, the furious (rudra) of red colour, the heroic (vῑra) of
yellowish-white colour, the fearful (bhayānaka) of black colour, the
supernatural and amazing of yellow colour, and the repulsive (loathsome,
bῑbhatsa) of blue colour.33
The expressionism of colours visualises a temperamental attitude and is
concerned with the wide range of emotions. Yet, side by side with it, colour
in its descriptive quality was made use of to a large extent. It was not only
known as local colour, distinctive of, and unchanging with, the various
objects, but also its modifications due to light and surroundings were
considered: ‘Vajra said: my curiosity (runs) high, and I wish to hear (more)
about the true and untrue colours of water, mentioned by you. Mārkaṇḍeya
replied: The untrue colour of water resembles that of lapis lazuli. It is the
effect of the reflection of the sky in water. But the natural colour of water is
seen in the falling down of water-falls; it resembles moonlight.’34
The abstract and the realistic vision, which as a rule, we hold apart as
poles in the evolution of art, isolated from one another by gradual steps of
development or by the sudden gap of reaction, are but the two sides of
Indian art, contemporary and organic, for the obverse is turned toward that
which lies outside, changeable, alluring in its variety and provoking
observation, while the reverse faces the within, essentially unchangeable,
because continually stirred up by emotions, of which cetana, the life
movement, is the common source. To do justice to them, a language of
symbols comprises colours and measurements in solemn hierarchy.35
10

The Jataka as Popular Tradition

 UMA CHAKRAVARTI

This essay was first published in Studies in History, vol. 9, no.1, n.s.,
1993, pp. 43–70.
I am grateful to Kumkum Sangari, Neeladri Bhattacharya and the
anonymous referee of this paper whose comments on an earlier draft
were extremely useful in revising it.

INTRODUCTION: TEXT AND CONTEXT

T HE JATAKAS, A COLLECTION OF 550 STORIES, CONSTITUTE a

significant part of the Buddhist tradition. It is probably one of the most
widely known compendiums of the Buddhist world of which many of the
stories form part of the oral tradition.1 Despite their popularity historians
have been wary of using them in recounting the past. The scepticism in
using these texts is not without reason; as narratives, they present their own
problems as sources. Further, they contain many layers and are therefore
very difficult to stratify or to arrange chronologically. These are major
problems, especially for historians of early Indian history struggling against
great odds in their attempts to tell us something coherent about social
processes. Thus, evidence from the Jatakas, to date, has been used only in
conjunction with evidence from other literary texts; the usual practice is to
pick up a statement from the Jatakas to bolster a point that the historian is
trying to make. As narratives, however, the Jatakas open up other
possibilities. What I intend to do here is to use the Jatakas more fully, using
each segment of a narrative in its entirety rather than its isolated statements,
to tentatively explore the everyday lives and experiences of people in early
India. I will focus on the labouring poor, who hardly figure in the texts of
high culture except as a collective category upon whom certain rules were
sought to be imposed. The Jatakas represent a genre of literature, possibly
the only one of its kind for early India, that enables us to get a glimmer of
the total experience of men and women who inhabited the world for which
the texts of high tradition made rules.
The difficulty in dating the Jatakas is mainly attributable to its many
layers.2 A collection of stories is intrinsically open-ended, and can grow (as
the Jatakas clearly did) over many hundreds of years till it is finally compiled
in the form in which it comes down to us. According to scholars, there are
different strands within the Jataka texts that are earlier and later; the core of
the narratives was widely known by the third century BCE, by which time
they were being sculpturally represented on Buddhist monuments and are
identifiable by their titles which are inscribed at the top end of the carvings.3
The portrayals clearly utilise a grammar understood alike by the sculptor,
the bhikkhus (Buddhist monks) and the general public who congregated at
these spots. The above evidence indicates that there is a specific association
between these narratives and Buddhism. By about the middle of the third
century CE, some of the Jataka stories were available in Chinese versions,4
and by roughly the middle of the sixth century CE, an elaborate commentary
linking up various parts of the narratives, which was itself inherited as part
of the Jataka legends from earlier times, had been finalised.5
Although the various layers of the Jatakas cannot be chronologically
unravelled very effectively, the text as we have it provides evidence of the
broad characteristics of the society that was the locus of the narratives.
Sharp differences based on class and caste stratification were deeply
entrenched. Hierarchy and pollution taboos are themes that recur in the
narratives of the Jatakas. Further, the brahmana’s claim to pre-eminence,
which is a running theme in the high Buddhist textual tradition,6 frequently
appears in the Jatakas too, only to be contested and exposed within the
narrative rather than through rational arguments of the Buddha as in texts
of the high tradition. Thus, while the major themes of the Jataka stories bear
some relationship to the texts of the high Buddhist tradition, the treatment
of the themes is different in each case.
Another way of locating the Jatakas is in relation to the body of texts
produced by the brahmanas. What is significant about the brahmanical
texts, especially the narrative texts composed roughly during the period that
the Jatakas were assuming some shape, is that they represent a fairly unified
ideology and bear the definite stamp of the traditional literati, men schooled
in the prevalent system of learning available to the brahmanas. The Jatakas,
on the other hand, are marked by a comparative looseness of structure and a
somewhat dispersed ideological content, even though the various strands
are gathered up and held together within an overall Buddhist worldview. The
Buddhist worldview may be summed up as an emphasis on the rejection of
the claim of the brahmanas to social and intellectual pre-eminence based on
birth, and, in its stead, the advocacy of acquired spiritual merit, potentially
available to all who are willing to strive for it. This emphasis on equality in
the path to salvation, however, went along with an acceptance of social
inequalities (between rich and poor, high-born and low-born) and a notion
of karma, or the fruit of action, which determined the status of an individual
in society, but also made it possible for the same individual to shape his/her
ultimate destiny. Generosity and moderation were deeply cherished virtues,
and it was the selfishness and lack of moderation in the brahmanas’ conduct,
along with their unashamed pursuit of material goals and their claim to
inherent social pre-eminence, that marked them off as a group with whom
the Buddhists were in opposition.7
While there is a contrast between the Jatakas and the Brahmanical texts,
there is also a contrast between the former and Buddhist canonical texts
such as the Sutta and Vinaya Pitakas. These deal explicitly with points of
doctrine, essentials of Buddhist philosophy and rules of governance. Like
the Brahmanical texts, they are characterised by a fairly tight structure and
formal presentation even while using, unlike the Brahmanical normative
texts, a narrative mode. The Jatakas, on the other hand, hardly ever make
direct references to the central themes of Buddhist philosophy, but they do
take up Buddhist ethics and deal with them using a more popular idiom, as I
shall show. The minimal presence of philosophical content and the adoption
of a narrative strategy makes it possible for more of everyday life to be
represented in the Jatakas.

AUDIENCE, LANGUAGE AND AUTHORSHIP: THE SOCIAL

CREATIVITY OF THE INARTICULATE
One reason for this heightened representation of everyday life might be that
while it is clearly not possible to identify folk narratives like the Jatakas as
the production of a particular set of people (they are the common property
of the people with common authorship), it is possible to see in this common
authorship marked participation by sections who are normally outside the
arena of intellectual production. It appears to me that at least some of the
narratives in the Jatakas, where the subordinate classes are sometimes
represented with a sensibility that is suggestive of emanating from
experience, point to the ‘social creativity of the so-called inarticulate’.8
However, one can never tell with any certainty as far as the Jatakas are
concerned, so I shall return to this grey area later.
Keeping in mind the general constraints of text formation within a
religious tradition in early India, the Jatakas come as close to popular
literature as possible. Containing a strong component of folk tales, the
Jatakas are probably one of the oldest of such collections in the world. Many
of the stories are pieces of folklore which had floated about for ages as waifs
and strays of literature, to be appropriated by any casual claimant.9 In the
case of the Jatakas, the appropriation took place at the hands of the Buddhist
bhikkhu. The bhikkhu mediated between the folk tale and the canonical
Buddhist text to construct a unique set of narratives; he mediated also
between the great and little traditions within Buddhism which, as a
peripatetic, he was ideally suited to do. The bhikkhu was on the move
constantly and through the daily alms-round, was placed in close
relationship with common folk, whatever his own origins may have been.
Like the medieval friars whom Burke describes as amphibians, men of the
marketplace as well as the university,10 the bhikkhu straddled the divide
between the common folk and the learned; functioning as an active bearer
of the little tradition upwards. Ultimately, the Jatakas were officially accepted
and classified among the miscellanous texts of Buddhism. Despite this
official acceptance of the stories, the relationship between the narratives and
high Buddhism, which the mediators strove hard to achieve, remains
somewhat stretched when compared to the authoritative Buddhist texts. It
appears that the folk narratives were not entirely successfully homogenised
by the master narrative and retained a certain measure of autonomy from
the latter.
 There are other aspects that lend a popular face to the Jatakas. One is the
use of Pali, the language of all Buddhist texts. Though Buddhism cannot be
divested from the elites, pali, the language of the texts, has been described as
a compromise of various spoken dialects; it was seen as an alternative to
Sanskrit, the language of a fairly restricted group of people even among the
elites.11 Sanskrit was consciously rejected by the Buddha for the language of
the bhikkhu and the lay folk. Closer to the language of the people, Pali
would have enabled more easy incorporation of the folk tale in its original
form, with more people participating in the process of incorporation.
Another characteristic that associates the Jatakas with the popular is its
oral narration, or its link with performance. In the seventh century CE, Bana,
author of the Harshacharita, referred to owls repeating the Jataka stories,
having heard them continuously narrated.12 There is also a phrase, iti
vitharstabbham, which suggests that the narrator was to amplify the details
at his discretion.13 This suggests that there was no fixed text but, rather, one
that provided a broad structure, leaving considerable room for improvisations
at the time of each performance [emphasis added].
The association of the Jataka stories with narration to an audience, set in
the open, is fairly strong, but what is not very certain is who constituted the
audience. The location of the opening narrative is almost invariably the
sangha (the Buddhist monastic organisation), and the context suggests that
the second narrative is being told to the bhikkhus whom the Buddha is
guiding in their everyday problems. This, however, may equally be a literary
device to link the appropriated narratives together into a text that served a
special purpose in Buddhism. Thus, it is fairly plausible that a Jataka story
was told to larger lay audiences in its entirety. This appears to be the case
with some of the Jataka stories, like that of the famous Vessantara. The ploy
of setting a narrative within a narrative appears as a device to keep audience
interest alive: so, whether the audience consisted of bhikkhus or of common
folk whom the bhikkhus addressed, it was a ‘popular’ audience, drawn from
a heterogeneous social background.
The presence of an audience would have had a crucial bearing on shaping
the text of the Jatakas, making it somewhat fluid and contextual, wherein
differences in narration and style would be improvised to make for
particular emphasis according to the nature of the audience. Thus the
audience response would have been recycled as an input into the original story,
leading to considerable interpenetrations in the agendas of the narrator and of
the audience [emphasis added]. The performance aspect of a narrative
necessitates a two-way interaction: between the text and the audience, and
between the narrator and the ‘consumer’ of the text. Central to this two-way
interaction is the problem of resolving the gap between the message of the
narrative and the meaning actually derived by the listener/reader of the text.
This was a continuing problem that the Jatakas attempted to resolve, as we
shall see.

TEMPORALITY, STRUCTURE AND NARRATIVE

There are certain special characteristics in the Jatakas, the most striking
being the unique use of time and the handling of many levels of narration,
with the establishment of links between the various levels. Each Jataka story
is divided into four parts: the main story from which the title of the Jataka is
usually derived is preceded by an introductory story that is set in the
Buddhist present, that is, the age of the Buddha. It is called the
paccuppannavatthu, or the story of the present, and recounts where and in
what circumstances the main story was told by the Buddha. Then follows the
main story, which is called the atitavatthu, or the story of the past. It always
begins by mentioning the place and the period when the events took place,
and is often set in a stock mythical past, located in the reign of King
Brahmadatta of Banaras. It often begins with the phrase, ‘once upon a time
in the reign of King Brahmadatta’ the Bodhisattva (the Buddha in his
previous incarnation) was born as such and such. It then proceeds to
introduce other characters in the narrative. The Buddha himself is thus more
or less always involved, either as an actor or as a witness, in the events of the
atitavatthu, and it is he who recounts the tale of the past. At the end of this
narrative there is a verse, which is the essence and sometimes the moral of
the story; finally, there is a linking of the stories of the past and the present
through an identification of who is who among the major characters in the
two narratives, called the samodhanam, and is attributed to the Buddha.14
Multiple levels of the narrative are made possible within the larger
framework of karma, the cycle of birth, death and rebirth, plus the Buddhist
notions of the cyclical and therefore incalculable nature of time. These also
enable each Jataka story to move with ease from the world of animals to the
world of gods and demons, respectively, and return once more to the world
of men.
The linking together of different time-frames, different regions, and
different animal and human characters within a single narrative is a unique
achievement of the Jatakas, making it possible for them to contain a
simultaneity of forms. There is a mythical quality to the narrative as it
evokes a timelessness or immeasurable units of time, and the long durations
of past existences in heaven and hell; these go alongside a more realistic
delineation of human existences on earth. As Feer has pointed out in his
study of the Jatakas, the chronology within the Jatakas is made possible
through a succession of existences extending over a great number of years
and kalpas.15 Each kalpa, which is only one day in the life of Brahma,
comprises a colossal number of years which the Buddhist imagination has
some difficulty in filling with events.16 The narrative structure of the Jatakas
is a modest way of doing so. For example, in one Jataka, the Bodhisattva is
described as spending seven full kalpas in the heaven of Brahma before
fulfilling all his existences in the world of men. In the interim, a long period
is spent in the Himalayas. In another Jataka, the Buddha is born as a yaksha
500 times, then as a dog 500 times, with an interval between the two series
of existences of 100,000 years in hell. The totality of the past adding up to an
incalculable number of existences, fulfilling immense periods of kalpas, is
preserved in memory and not in historical time, and is punctuated with
formulaic descriptions of ‘when I resided in the forest of Brahma’, for a
mythical existence, and ‘once again when I am’, for an earthly existence.17
This unique handling of time makes the narrative structure more
interesting, but also more complex, and added to the problem of ensuring
firm control over the ‘correctness’ of meaning to be derived from the
narrative by the ‘receiver’.

CULTURAL PRODUCTION AND CONSUMPTION:

ANCHORAGE AND CLOSURE IN THE JATAKAS
I have already argued that the Jatakas appropriated a considerable amount of
folklore and that this appropriation took place through the mediation of the
bhikkhu. The Buddhists used the dual characteristics of folklore, having
both fixity of form and capacity for variations in content, to insert their own
messages. The traditional narratives thus became vehicles for new ‘cultural
meanings’18 that were intended to reach the largest audience possible, in
contrast to the texts of the high tradition in Buddhism, which would have
had a much more restricted clientele. But there still remained the problem of
ensuring that these Buddhist messages and new cultural meanings were
actually received as they were intended.
In this context, the autonomy of the listener/reader of a narrative needs to
be considered, and it might be useful to note the distinction between
‘creation’ and ‘consumption’ made by Chartier. The people, according to
Chartier, are not passive consumers; they are not like wax to be inscribed
over in whichever way the author likes. The text (in our case, the Jataka
stories) may be ‘appropriated’ differently, just as, within the narratives of the
Jatakas, the creations of the people may have been appropriated by the
bhikkhus to serve the goals of high tradition.
Chartier further argues that intellectual consumption should be taken as a
form of production. He suggests that although the consumers may
manufacture no object, they constitute representations of the object that are
never identical to those that the producer had implanted in the work.19
Similarly, Michel de Certeau argues: ‘To a rationalised production
corresponds another production, qualified as consumption. The latter is sly,
it is dispersed, but it insinuates itself everywhere; it is quasi-invisible since it
is not recognisable by its products per se, but by ways of using the products
imposed by a dominant order.’20
That readers (or audiences) may recompose the text to permit a plurality of
meaning, is now widely accepted [emphasis added]. Ginzburg has shown how
the sparse intellectual elements of a literate culture, when they reach a man
of the people, can be used by him to read a text differently. Menochio the
Miller had a different reading code from that of the learned man; by
breaking up the text into individual sections that acquired their own
meaning, he changed the hierarchy of themes and radically altered their
meaning. His reading code flowed out of his experiences rather than being
one that reflected the external system of knowledge.21 Thus, despite the fact
that the relationship of the popular to high culture remains a grey area
within the Jatakas, there is a strong possibility that the narratives contained
in them could be decoded and used differently from the way the bhikkhus
had intended. So it had to be ensured that the intended message (which took
recourse to an appropriation of folklore) was not left too open-ended and
thereby capable of a diametrically opposite meaning. They did this through
a creative narrative strategy set within a tightly structured form in which the
narrative unfolded.
A number of strategies were adopted by which the narratives in the
Jatakas were sought to be provided with fixed meanings, rather than leaving
the message dispersed. First, the ‘true’ subject matter of the narratives is
Buddhist sila or ethics, although it is not often specifically mentioned, and
the structure of the Jatakas is designed to give prominence to it. The
Buddha’s ‘resplendence contrasts with the shadow which the opponents of
the Buddha cast upon the narrative’, as Feer points out.22 The circumstances
that provoke the Buddha’s narration of a story often are the evil intentions of
the villainous Devadatta, or the vices of lazy bhikkhus, or the virtuous
actions of certain bhikkhus, and these are woven into the narrative to
highlight the Buddhist sila. Second, there is consistency in the behaviour of
the characters who are linked in the stories of the past and the stories of the
present, and this too helps to provide fixity of meaning. Third, since the
present in the Jatakas is always contemporaneous with the Buddha and this
is the starting point of the narration, and also because the entire set of
narratives concludes with a return to the Buddha, the beginning and end of
a Jataka story is always in the hands of Buddhists.
The strategy works to provide anchorage and succeeds in imposing
Buddhist ethics upon different sub-narratives which might be read in a
variety of ways. There is thus a master narrative that seeks to provide both
anchorage and closure to units which may otherwise be open-ended, and
capable of being ‘consumed’ differently.
What is most significant about the way each Jataka story is structured is
that it lends itself to a sedimenting of meaning. The Buddhists attempted to
retain control over meaning not only by providing a closure for the entire
narrative, but by providing that each segment and each individual story in
the larger account had its own closure. Thus the closure comes at different
points: it is provided to the original narrative or the paccuppannavatthu, to
the second narrative or the atitavatthu, and, finally, in the linking of the two.
These closures anchor the narrative, as they come in the form of evaluative
statements. The introduction of Buddhist characters linking the major
figures, the insertion of ethics, with the good being marked off from the bad
as they stand above ordinary mortals who cannot rise above the selfishness
and narrowness of men and women, further try to ensure that the ‘correct’
meaning reaches the audiences of the Jatakas. In these ways, the Buddhists
consciously tried to maintain a firm grip on the way the narratives were
understood. However, only an analysis will indicate whether they fully
succeeded in achieving controlled ‘consumption’ of the texts.

RECURRENT MOTIFS: MEN, WOMEN AND BEASTS IN THE

JATAKAS

Certain themes and patterns have been identified as recurrent while

indexing Indian folk tales. These have been analysed by Beck to provide a
new coding approach where the basic unit is the ‘social dyad’. A social dyad
involves any two characters in the story, taken as a pair, and it makes that
relationship a primary unit of study. The dyads of Beck isolate the
protagonist as human or animal, or one that involves an interaction between
the two.23 Beck uses the moral relationship governing a specific dyad in folk
tales as her point of departure to explore the plots along a number of points,
and comes to certain conclusions. For the purpose of exploring the Jatakas,
the following arguments of Beck have a special interest.
According to Beck, while the actual events and personages described in
folk tales sometimes seem extraordinary, the social concerns they represent
are commonplace. Animals and ogres, for example, are usually guided by
human emotions and culturally familiar anxieties. Stories about troubled
relationships are more frequent than those depicting contentment. The body
size of the animals involved is often a key issue, and there are many stories
where a jackal is placed in a dyadic relationship with a large violent animal
such as a lion. These dyads relate to a greater/lesser relationship.24
An interesting aspect of the stories is that they run counter to social
norms. As may be expected, a number of themes deal with unequal
characters in a hierarchical situation, and many stories project negative
feelings about exploitation of a ‘weak’ (or inferior) partner and display
support for the underdog, quite unlike the social norm. Folk tales of
inferiors against superiors are often depicted as avoiding open
confrontation, and taking recourse to deception and cheating, especially in
the case of animals. Further, even though the weaker may not succeed
against the stronger or larger animals, the latter are forced to defend
themselves.
Although such themes occur across cultures, the emphasis they receive in
South Asia suggests that the pattern holds a special importance for this
region. Beck argues that folk tales often project attitudes not allowed full
expression in everyday life and hence it is important to note that folk
interest in the success of an inferior finds its maximum expression here.
Most importantly, when a tale unfolds from the point of view of a lower-
ranking figure, the challenge and confrontation often originate from below.
But, at the same time, alliances between masters and servants also figure
commonly in the Indic story corpus, suggesting a certain ambiguity in
handling the master-servant relationship as merely hierarchical.
Dyads involving spouses are most significant. The husband-wife bond is
characterised most often by negative attitudes rather than neutral or positive
ones, involving adultery, suspicion or outright physical aggression; nearly all
the adultery stories highlight a woman’s and not a man’s transgression. In the
brother-sister dyad, a frequent concern seems to be the protection of a
younger sister’s sexuality from outside aggression or violation.
Beck concludes that Indic tales, both about humans and animals, often
stress relationships of inequality. The great majority of these bonds depict
distrust: fear either of abandonment by a superior or of rebellion on the part
of an aggressive and recalcitrant inferior. Rather than expressing non-strife
situations, they reflect widespread suspicion about the structures and
motivations that underpin daily life.25
 The most direct appropriation of narratives from the folk tradition in the
Jatakas are the animal stories; these are some of the older stories in the
collection. Nowhere else in Buddhist texts do we get animals so dramatically
depicted; even more significant is the fact that the Buddha himself is the
main character in the animal stories. This is also one of the elements that
helps define the popular character of the Jatakas. Recourse to animal stories
was recommended to preachers to use as moral stories to enliven sleepy
audiences in Europe, and may have served the same purpose here;26 there is
no doubt that pigs who preach, parrots that watch over women’s conduct
while their husbands are away, oxen and elephants who labour, and jackals
that attempt to subvert through cunning, when used as literary motifs or
narrative devices, would have a lively impact on listeners. But what is most
significant is that these animal characters are free from the pressures of the
human world to make both social and moral comments on the state of
human society. The narrative can therefore push forward its theme more
fully and more forcefully than otherwise.
As a device with popular appeal, the use of the animal story can be
explained by the real and collective memory of the close relationship
between the animal and the human world at some time in the past. Even at
the time that the Jatakas were finalised, men, women and beasts continued
to live in close proximity to each other, given the settlement pattern reflected
in Buddhist texts. Between each settlement and the cultivated area
surrounding it was what the bhikkhus called aranna or forest land, teeming
with animal and vegetable life.27 The bhikkhus, in particular, lived on its
edges and traversed it often in the course of their peripatetic existence. The
animal world was thus integral to the experience of the early Buddhists and
this may partially explain its prominence in the narratives. However, animal
stories are also an important part of the Indic folklore tradition (as stated
earlier), so their incorporation into the Jatakas is not surprising.
Stories about animals exclusively reflect norms and values that prevail in
the world of men. They express the same anxieties and modes of conduct:
like women who lure unsuspecting men, the female deer lures the stag away
and leads him to his doom;28 the male fish when caught in a net worries that
his wife-fish will think that he has abandoned her for another female fish.29
Just as in the human world, in the animal world too, friendship can be
established only with one’s own kind, certainly not with ‘low’ creatures.30
Thus there can be no friendship between an iguana and the low and base
chameleon.31 No mixing of classes is permissible: the golden goose who
cohabits with a crow when he goes ‘upcountry’ tells his offspring that he
must know his ‘station’, which is not the same as that of his legal and pure
offspring.32
Thus the animal kingdom has thoroughly internalised the social value
system of the humans. But while the animals occupying lower positions
cannot escape the hierarchy that governs them, they too, like men, attempt
to break loose from their imposed identity. The offspring of the golden
goose and the crow, who is neither like a crow or a goose (and is therefore
named ‘Dingy’) aspires for a different place for himself, and for a brief
moment, likens himself to the king of Videha as he is being flown over it by
his stepbrothers on a visit to his father. Enraged by the verbalisation of his
aspirations, the stepbrothers report the matter to their father. Rebuked and
rejected by his father, Dingy realises that he will be acceptable only in the
world of his mother.33
The danger in transgressing social barriers is the theme of another animal
story. A jackal who occupies the place of the untouchable in the animal
world falls in love with a lioness (the lion represented royalty, the aristocracy
among beasts). The jackal proposes to the lioness that they should mate.
Says the jackal, ‘Oh lioness, I am a four-footed creature and so are you;
therefore be my mate.’ Unfortunately for the jackal, the lioness does not see
the essential similarity between them. Horrified by the transgression of the
social code by the low-down jackal and dishonoured by his proposition, the
lioness contemplates suicide. Fortunately for her, however, her eldest brother
vindicates his sister’s honour and the jackal receives a fitting end for his
transgression.34
Only in one narrative does a low-down animal, this time a boar, succeed
in staving off his ignominious end for attempting to assert the essential
similarity of all four-footed beings and thus claim equality with the lion, the
noblest of all creatures in the animal world. The lion challenges the boar to
fight a duel; the terrified boar is convinced that his end is near. But all the
other boars devise a ruse that will surely save him from the lion. They advise
the boar to roll about in dung for seven consecutive days, expecting that the
lion, who is a clean creature, will refuse to fight the stinking boar. The ruse
works and the lion proclaims, ‘Oh dirty boar, your hide is foul, the stench is
horrible to me; if you would fight I yield me quite and own you have the
victory.’35 When faced with sure death, the only weapon the weak can
successfully take recourse to in counteracting the strong is strategy and low
cunning.
The animal motif serves a different function in narratives where animals
and humans interact. These are occasions when animals can express what
the lower classes must be silent about. One of the Jataka stories suggests that
the desire for a better life weighs over emotional considerations in parting
from loved ones. It describes how a woman who had fallen on bad times
rents out her lodgings to make ends meet. She receives from a horse-dealer,
in exchange for the hire of lodgings, a young foal, whom she loves like a son
and feeds with whatever is available to her. A travelling merchant notices the
value of the horse and asks to buy it for a good price from the old woman.
‘What are you saying’, says the old woman, horrified by the idea, ‘one mustn’t
sell one’s own foster child.’ But the merchant makes her conscious of the
riches and honour that the foal could have if he lives with the merchant. The
old woman then agrees to give up the foal, saying, ‘then take this child of
mine and may he be happy.’36
The story of this mother-son dyad which bridges the animal and human
world points to the nature of transactions between social classes. To the old
woman, affection cannot have a monetary value, but the merchant is able to
persuade the old woman to part with her beloved foster son because her
present loss can be offset by the reassurance she derives of the pleasure and
happiness her foster son will gain in the future, when his value is recognised.
Implicit in the story are certain notions of ‘fair’ and ‘just’ transactions, and
the differences between the poor and the rich in their ways of transactions.
The transaction itself is both unjust and just, equitous and non-equitous at
the same time. What appears to be just to the merchant is clearly not so to
the old woman initially, but she comes to accept the merchant’s notion of
value later, for it will ensure the happiness of her foster child.
 The bonds of solidarity between ‘good’ labouring men and an elephant is
the theme of another story. The elephant is saved from a painful wound
suffered by him from a shrapnel of wood being worked by some carpenters.
In great agony, he brings the wound to the notice of the carpenters, who
doctor it carefully. Grateful for the help, the elephant labours for the
carpenters and, in return for the labour, the carpenters feed him, each with a
portion of their own meal. When the elephant grows old he hands over his
young son, a marvellous white specimen, who in turn works and lives with
the carpenters, playing with their children when his work is over. The king
of the land comes to hear of the magnificent elephant. Honoured by the
king’s request, made in person, the carpenters decide to gift him to the king.
But the elephant refuses to go with the king until the king pays to the
carpenters what they had spent on him over the years. Even after the money
is paid he refuses to budge till the king accedes to his second demand, which
is clothes for the carpenters, their wives and their children. Then, with many
a loving look backwards, he bids a fond farewell to the carpenters and goes
off with the king.37
The narrative in this story brings together animals, humans and the king
in a three-cornered relationship. As in the earlier story of the old woman of
humble means and the young foal of noble lineage, where a relationship
comes into being between the high and the low, here too, the elephant, a
great and noble animal, receives affection from poor carpenters, establishing
an enduring reciprocal bond between the high in the animal world and the
humble in the human world. In both stories, the affectionate world of love
and happiness is disrupted by powerful and rich figures from the human
world who use their power and authority to break these bonds. However,
while in the story of the poor old woman the young foal accepts the decision
of his foster mother, in the story of the elephant the disruption caused by the
figure of power is apparently accepted by the humans but is contested by the
elephant. The elephant’s contestation draws attention to the injustice of the
king’s action in arrogating to himself everything that is magnificent in the
land, and critiques the notion that the poor have no right to enjoy such
magnificence. However, neither the humans (whose capitulation to the king
is depicted by the narrator as their being honoured by the king’s request) nor
the elephant can resist the king’s power. Further, the humble carpenters
cannot even make a statement about their condition to the king. It is the
elephant who manages to convey to the king what his obligations to the
carpenters should be. The elephant’s contestation has its effect.
There is also a plea for justice, and for reciprocity between the king and
his people, even as the loving world of reciprocity between the carpenters
and the elephant is being destroyed by the king. The critique of kingly power
in this narrative echoes a similar critique of kingly power and the setting of
its limits, the laying down of norms for the legitimate exercise of power and
the creation of a just social order, in the Digha Nikaya,38 a crucial text of the
high Buddhist tradition. Characteristically, the story tells us that the
elephant is actually the Buddha in a previous incarnation. This is a
marvellous example of the coalescing of the Buddhist high tradition with the
popular tradition prevalent in the region.
Protesting with the feet to seek one’s due is the theme of another story
where animals and men constitute a close community. A poor old woman
receives as payment for letting out her lodgings, a young bull-calf. The old
woman rears him like her own child, feeding him with rice gruel. He is
called ‘Grannie’s Blackie’ as he is black as jet, and thus he grows up with the
village urchins playing with him as if he is one of them. One day he thinks to
himself that he should ease the hard lot of the old woman who has so
painfully reared him as if he were her own son. The bull finds a suitable
opportunity when a young merchant with 500 wagons gets stuck in a ford as
his oxen cannot pull the weight of the wagons. The merchant notices
Grannie’s Blackie who has strategically placed himself in sight of the
merchant; the merchant asks the cowherds who he belongs to, so as to settle
the wages for his services. On finding that there is no one there to claim
him, the young merchant attempts to yoke the bull. The bull, however, will
not budge till his wages are fixed. The merchant, understanding the bull’s
gesture, fixes the wages at two coins per cart. Thereupon Grannie’s Blackie
goes into action and pulls all the wagons across with one heave each. Having
got his work done so efficiently, the merchant tries to backtrack on the
contract by paying only half the amount he had earlier settled, since he was
dealing only with a ‘dumb’ animal. Grannie’s Blackie, however, resists the
infringement of the contract: he stands across the path of the first wagon,
blocks the way and refuses to budge from there. Unable to proceed, the
merchant gives in to the resistance put up by the bull and pays up the due
amount. Away then goes the bull with the 1,000 pieces of money tied to his
horns to hand over to his ‘mother’, who is much moved by his gesture. She
rubs him with warm water and oil, and feeds him with dainties.39
The key point in the narrative is where the merchant who needs the
services of the bull uses his labour and then does not pay for it adequately. It
serves as an allegory for the rights of the voiceless, for all those who work
and are not given their due. The action of the bull is symbolic of the
potential of resistance; even the seemingly dumb can protest against an
infringement of a collective understanding of what is right.
There is a play in this narrative between apparent weakness and actual
strength, which the merchant cannot distinguish. As in the earlier two
narratives, here too, it is a powerful outsider who disrupts the
understanding, generosity and sharing that have been established earlier.
The outsider follows a different code, either subscribing to a different system
of values, or displaying the arrogance of royal power, or as in the story of the
bull, the fraud, manipulation and greed of the merchant’s world.
Significantly, all human figures of power treat animals as dumb and
voiceless, as those whose value/labour can be appropriated. But the
appropriation is contested by the elephant and the bull.
Among these three narratives, the most effective resistance to disruption
comes from the bull, and it speaks most eloquently about labour, wages and
social norms. Most importantly, when the bull silently resists and, in the
end, triumphs over the merchant, he reveals the potential of resistance. He
also shows the power of love and reciprocity against the forces of greed and
injustice. What is interrogated in the narrative is the deceit and exploitation
in the relations between the users of labour and those who must sell their
labour in order to earn. What is celebrated is the more humane world of the
poor.
The story also elaborates another set of themes: there is the familiar
relationship between the old and the young, which is a cyclical relationship
of reciprocity. The old woman, as female, may be regarded as ‘weak’ but she
begins the relationship of reciprocity by nurturing the young bull-calf who,
as male, may be regarded as strong but initially requires nurture. As the
narrative proceeds, the roles are reversed. The bull-calf now supports the
nurturing mother and alleviates her poverty. He labours for her sake; she in
turn rubs him with oil and feeds him with dainties. The mother-son dyad
retains the roles but changes the elements so that the relationship goes full
circle. The son is now the protector, the actual provider, which originally the
mother was.

REPRESENTATIONS OF WOMEN: THE INTERFACE BETWEEN

HIGH CULTURE AND POPULAR CULTURE, AND THE
COALESCING OF VALUES

Narratives relating to women in the Jatakas provide a valuable insight into

the points of intersection and divergence between popular culture and the
culture of the higher classes, as well as between the Buddhist and
Brahmanical traditions. A valid point of departure in reading the Jataka
stories about women could be the relationship between the normative and
the narrative literature: the one details the code of conduct for women, while
the other draws our attention to the extent of subversion of such rules by
everyday acts of resistance. These subvert the rules for the control of women
—the severe sexual control imposed upon them, especially if they belong to
the upper classes, and also on their speech and everyday actions. The
narratives represent women as being prone to ‘lazing’ or ‘gadding’ about,40
slovenly in their householding, dissipating the store won by the hard work of
the husband in the field,41 or drinking when they are not being downright
adulterous.42 More often than not, they combine their slovenliness with
displaying an inordinate and ghoulish sexual appetite.43 On the whole, the
representation of women in the Buddhist Jatakas reiterates the apparent
obsession in the normative literature with the need to control women by
constantly guarding them and restricting their freedom.44 The only
distinction between the texts of Brahmanic culture and the more popular
Jataka stories is that between prohibition and the representation of
transgression: in the former, transgression is only implied or viewed as a
possibility; in the Jatakas, it is actually enacted by women. Apart from this,
there is a coalescing of the values of high culture and the more popular
beliefs about the ‘essential’ nature of women.
The Jataka narratives about women articulate the monastic prejudice
against women:45 by their very presence, women are perceived as an
extremely potent threat to the bhikkhu seeking salvation. Almost invariably,
the many stories dealing with the insatiability of women’s sexual appetite
and their habitual proneness to adulterous liaisons are an account of one
bhikkhu or another pining for the wife of his lay life. These introductory
stories indicate the rigours of monastic existence, and the difficulties of
convincing men of the need for chastity and asceticism as practices that
must be followed. The mediation of the Buddhist monk in the production of
the Jataka literature is thus most noticeable in the narratives on women; the
stringing together of the various layers of the narratives bears the stamp of
Buddhism and its insistence on the need to reject the householder’s life
when pursuing the goal of salvation. Within these narratives, a story of the
past is effectively utilised to warn the bhikkhus of the doom women signify.
The incorporation of sub-narratives into the master narrative is thus much
more coherent and effective in these tales.
It is in stories of the past that representations of women abound. Women
of the upper classes are almost always associated with adulterous behaviour,
seeking opportunities for such ‘misbehaviour’. No matter how much men
may do for them, women are never grateful, never satisfied.46 They manage
to circumvent the most elaborate ways of guarding them, showing no
discretion in their choice of partners: just about anyone will do. Men, in
turn, are advised to be constantly on the alert, and to devise more and more
stringent ways to guard or discipline and punish women. One husband who
has problems with his wife is advised by the Bodhisattva to cure his wife,
who ‘gads’ about at night and lazes about feigning illness during the day, by
threatening to feed her with pickled dung.47
In the Cula Paduma Jataka, we are provided with a long and complicated
account of the despicability of women. A group of princes are banished by
their father. They leave the kingdom with their wives and wander about. As
their hardships mount and they are left without any food, they resort to
killing their wives and eating them one by one, beginning with the youngest.
One of the brothers saves his wife by saving a portion of the food each day
and compensating for her life when her turn comes to be killed. He then
leaves his brothers and travels further with his wife, whose life he saves again
by giving her blood from his knee to drink when she is dying of thirst. Yet,
the wife resorts to adultery at the very first opportunity when her husband
shelters a criminal whose life too he has saved. The wife then attempts to kill
her husband by pushing him down a precipice. The prince escapes and is
installed as king later on. He then punishes the wife by ordering her killed.48
In the Bandana Mokkha Jataka, a queen makes the king promise exclusive
fidelity to her but herself commits adultery with each one of the royal
messengers sent by the absent king to enquire about her welfare. The verse at
the end of the narrative proclaims:

The passions of woman are insatiate

and she does but act according
to her inborn nature.49

Similarly, the wife of a brahmana who has to go out on work misbehaves

every day. Unfortunately for her, a pair of parrots are left to watch over her
and they duly report her activities to the brahmana when he returns. The
parrots comment that it is impossible to guard a woman: ‘You might carry a
woman about in your arms and yet she would not be safe.’50
In contrast to the wives of kings, brahmanas and gahapatis, who are most
often associated with the adultery stories, women of the unpropertied
classes are depicted as engaged in labouring51 or in eking out an existence,52
and also in searching for occasions to snatch a few small pleasures together
with their menfolk. Their sexuality represents pleasure;53 this takes us to the
stories of men and women of the lower classes.

THE LOWER ORDERS: LABOUR, DRUDGERY AND SURVIVAL

STRATEGIES
The lower classes are represented in the Jataka stories as engaged in
agriculture, tending livestock, craft manufacture and performing drudge
labour. ‘Working for wages’ to support themselves and their families is a
very common description of the poor.54 Some of the stories impose a certain
identity upon them in representing social relations: loyalty and meekness are
their stock characteristics.55 ‘Meek as a 100-piece dasi is a common
expression in the Jatakas.56
Representations of loyalty include an occasion when the ex-farm servants
of a gahapati (who has fallen on bad times and become a victim of injustice)
take the initiative in enforcing moral order. They march in a body to seek
the king’s intervention to ensure that justice is done to their ex-master.57
This is in conformity with the moral order as prescribed by Buddhists: kind
masters and loyal, hardworking dasas, not envious of their masters’ position
and wealth, make for social harmony.58
However, even in the canonical texts, which represent the official voice of
the Buddhist social philosophy, there is a suggestion that the portrait of
social harmony is only an idealisation emanating from men of property who
are the major employers of servile labour. It is this disjunction between
imposed identity and the experiences, fears and desires of the lower classes
that the Jatakas manage to represent by the simultaneous incorporation of
many different levels of popular consciousness into the text.
The quality of life of the poor is one feature that the Jatakas capture.
Abject poverty defines the lives of the lower classes. By their food, clothing
and dwelling, they are marked off from those who have money. ‘In my whole
life’, says one of them, ‘I have possessed nothing of value.’59 Often beaten,
reviled and abused by their masters, drudgery distinguishes their labour.
Their experiences are so limited that a poor rustic who became a bhikkhu
(in order to escape his abject poverty) is represented as being able to see
only the shaft of a plough in every object that he is shown.60
With no real choices available to the poor, one such man, wanting to be
eased of his poverty, tells the Buddha that he worships a tree sprite to deliver
him from his misery.61 On the other hand, a young dasi, probably the most
powerless in the social hierarchy, who has known of no other existence, can
conceive of no escape. Her story makes it clear that the realm of the possible
for her employers is vastly different from her own.62
How did the lower classes cope with the ‘daily anguish of the morrow’63 in
a life that denied them any self-esteem?
In the Gangamala Jataka, a poor water-carrier is so ecstatic at the prospect
of enjoying a few hours at a festival that the hot sun beating down in the
heat of the noon-day, the burning sand beneath his feet as he runs across the
city to retrieve his carefully hoarded saving of half a penny, and the king’s
order to stop in his tracks, are all obscured by the pleasure and fulfilment in
the here and now. When the king asks him why he is so ecstatic, what
delights him so much that he sings joyfully in total disregard of the sun and
wind, the water-carrier says:
Oh king, I live by the South gate with a poor woman. She proposed that
she and I should amuse ourselves at the festival and asked if I had
anything [saved]. I told her I had a treasure stored inside a wall by the
North gate. She sent me for it to help us amuse ourselves: those words
of hers never leave my heart and as I think of them hot desire burns
me. Oh king, I sing to think that when I fetch my treasure I shall amuse
myself along with her.64
Earlier, the narrative informs us that the water-carrier and his companion
planned to spend a third of their savings on a garland, a third on perfume
and the rest on strong drink. He runs across the city, dressed in yellow
clothes and with a palm-leaf fastened in his ears, in pursuit of his carefully
hidden treasure.
But even such brief moments of joy are denied to the poor. A poor
woman longs to go to a festival in garments dyed yellow (the colour of
festivity) with her husband, but all she has are worn-out, colourless rags. Her
husband tries to persuade her to manage with what she has but the wife is
adamant: ‘I want a safflower-coloured cloth to wear . . . as I go about the
festival hanging round your neck.’ Finally the husband tries to get the
safflowers for the dye from the king’s garden, but is caught and taken out of
the city to be impaled alive. The narrative ends with a description of his
agonies. His dying regret expressed to his wife is: ‘Alas, I shall miss going to
the festival with you arrayed in safflower-coloured clothes, with your arms
twined round my neck.’65
There is, in this narrative, a celebration of the physical, a desire to
experience a few snatched moments of ecstasy. Clearly a means of escaping
the drudgery of existence, the festival acts as an ‘investment’ in survival.66
But what may have been an investment in survival for the poor is treated by
the high tradition of Buddhism as a low form of existence and, in reality, a
trap. By linking pleasure with death, the master narrative of Buddhist
monastic values imposes its message: pleasure can only be momentary; it
will be cut off by dukkha, or sorrow, unless the individual seeks the only
enduring end to dukkha, that is, cessation of all desires.

AUTHORITY AND DOMINANCE: TRANSGRESSION AND ITS

LIMITS

The possibility of staving off the authority of the powerful with deviousness
rather than defiance is also played out in some of the narratives. The lower
orders use Brechtian cunning to try and get rid of their oppressors. One
account tells how the harassed servants of a cruel mistress, who beat them
for the slightest reason, use the first opportunity to get rid of her. Once,
while the mistress is bathing in the Ganga, a storm breaks out and everyone
scampers away. Finding themselves alone, the attendants say to themselves,
‘Now is the time to see the last of this creature’; they throw her into the river
and hurry off.67 Similarly, the cruel son of a king orders his servants to bathe
him in the middle of the river and bring him back. The servants take him in,
but midstream they talk amongst themselves and say, ‘What will the king do
to us? Let us kill this wicked wretch here and now. So in you go, you pest!’, as
they fling him into the water.68 It is significant, however, that these accounts
do not refer to defiance of all forms of authority and oppression, but only of
a ‘cruel’ mistress or a ‘wicked’ prince. I have already referred to stories of
loyalty; so, taken together, these stories seem to define the limits of
authority, rather than interrogate authority itself. They depict forms of
authority that are acceptable and others that are not. At the same time, the
narratives also set the limits and legitimate bases of defiance and subversion.
 In the Jatakas, the poor desire entry into the world of the well-to-do.
Almost invariably, entry into this ‘other’ world is gained through a beautiful
nymph-like woman of the upper class.
A barber lived at Vesali shaving and hairdressing for the royal household,
for the king and queen, princes and princesses. A true believer, he listens to
the Buddha’s discourses from time to time. One day, he takes his son with
him to the palace. The young fellow, seeing a Lichchhavi girl, all dressed up
fine and grand like an apsara (nymph), falls madly in love with her. He says
to his father: ‘There is a girl—if I get her I shall live, but if I don’t there’s
nothing but death for me.’ He does not touch a morsel of food and lies down
hugging the bedstead. His father says to him: ‘Why son, don’t set your mind
on forbidden fruit. You are a nobody, a barber’s son. This Lichchhavi girl is a
highborn lady. You’re no match for her. I’ll find you someone else, a girl of
your own place and station.’ But the lad does not listen to him; others try to
pacify him but cannot. So he pines and pines away, and lies there till he dies.
Then the father performs his obsequies and after the first edge of grief has
worn off, goes to see the Buddha. The Buddha tells him, ‘This is not the first
time he has perished by setting his heart on what he must not have; he has
done this before.’ He then relates to the barber the jackal and lioness story
(which I have cited before), and, at its conclusion, the layman is suitably
elevated: he moves to the first step in the four stages to salvation.69
The narrative in this story ends by reiterating that social space cannot be
transgressed: true wisdom lies in understanding and accepting this
inevitability. For those who do transgress (like the young son), desire will
lead to death.
Another story dealing with desire leading to the transgression of social
space introduces a third element in the development of the plot and its
ultimate resolution.
The son of a dasi is brought up along with the son of her master, so he
picks up writing and other skills, grows up to be ‘fairspoken’ and ‘handsome’,
and is employed as a private secretary in the master’s household. The young
man, however, fears that his comfortable situation can only be temporary
and that, at any time, he could revert to the status of the other servants, be
beaten, imprisoned, branded and fed on slaves’ fare. He then thinks that the
way to secure his future is to pass himself off as the son of his master with a
merchant friend of the latter who lives far away, marry his daughter and live
happily ever after. The young man carries off the deceit and settles down
with the merchant’s daughter in great style. He is soon found out by his ex-
master but, by falling at his feet, he manages to beg the master to keep his
secret. He also manages to keep the secret because the master forgives him
after making sure that the young man ‘behaves’ himself with his wife, who
was finding it difficult to satisfy his finicky food requirements.70
A key problem for the young man in the narration is the need to conduct
himself publicly in a slave-like manner with his ex-master; this is required of
him in the presence of someone who knows his ‘true’ station. Others who
are not in the know of his secret, including his father-in-law, have thus to be
given some kind of explanation for his humble behaviour in the presence of
his ex-master, who is supposedly his father.
Another version of the same story ends with the dream crashing; the
truth is revealed by a parrot who is sent out in search of the impostor. The
parrot succeeds in locating him because the runaway slave’s lack of breeding
becomes apparent; unlike the others, who are born into wealthy and genteel
households, with whom he is sporting in a boat party, the slave is not used
to drinking milk containing a strong drug (customary among the gentlefolk
to avoid the cold). He spits out the drink on the merchant’s daughter’s head.
This ends his runaway days; he is hauled back to his old master where,
according to the narrative, ‘once more he has to put up with slaves’ food’.71
The three major elements in this second version thus are deceit, discovery
and a return to the original status. Also, it is significant that the two versions
deal with relations between the classes differently. While in the first the
problem of social distance is circumvented by an elaborate ruse, in the
second servility is so deeply internalised that it is impossible to carry off the
deceit. Ultimately breeding shows: no amount of education or training can
effectively erase it.
Desire may only be latent, as in the story of Nanda, the household servant
of a gahapati. Nanda holds the key to the knowledge of the whereabouts of
his young master’s assets left for him by the dead father. When the young
master tries to get Nanda to reveal the treasure, and the master and servant
reach the spot where the treasure is buried, there is a dramatic development.
As Nanda stands on the treasure there is a reversal of the master-servant
roles and Nanda addresses the master with authority: ‘You servant of a dasi’s
son, how should you have any money here?’ This happens repeatedly each
time the dasa stands on the spot where the treasure is hidden, and each time
he reverts to his normal servile mien when they return home. Finally the
young master is advised to assert his power. There is a recognition that the
spot on which the reversal takes place is the spot where the treasure is
buried, so the young master must pull down the dasa from his perch, dig
down, remove the family treasure and then make the dasa carry it back
home for him.72 There is no actual transgression of the social space in this
narrative, only a temporary reversal; at best a latent desire, a psychological
transformation and, perhaps, a recognition on the part of the dasa, as well as
the propertied, that the key to power and control over others lies in control
over wealth. This is evident from the servant’s appropriation of the mode of
address of the high when he has symbolic physical control over the treasure.
What is significant in this narrative is that when the story begins the
ageing gahapati considers the dasa a more reliable trustee of his property
than his wife. The subversive potential of the wife is recognised, but not that
of the dasa. It appears that the lower classes are not perceived as a threat; the
threat comes from one who is right inside the home. The dasa’s gesture of
holding on to the treasure is unexpected and is described as akin to one who
is possessed. It comes as a total surprise to all, even, maybe, to the dasa
himself.

BRAHMANIC HEGEMONY AND BUDDHIST NARRATIVE

STRATEGIES: SUBVERSION AND REAFFIRMATION

A question may be asked at this point. Was the subversive potential of the
lower classes not recognised in the Jatakas because this might reflect
disharmony? Or was it that the bhikkhus shifted the possible confrontation
of social groups, even at the level of representation, on to another plane in
the Jataka narratives? On the basis of other narratives, the latter appears to
be a possibility. I shall argue that direct confrontation between the high and
the low is represented in the Jatakas only on the intellectual plane, and it is
only at this level that the confrontation is permitted a resolution in favour of
those who occupy the lower rungs of the social hierarchy. The early
Buddhist texts often dwelt on the theme of the brahmana’s claim to high
status based on inherent criteria, which of course was contested by the
Buddha, as I have stated earlier.73 Nevertheless, social hierarchy based on
different criteria was recognised by the Buddha in the social world.74 It was
only in the world of the sangha that the principle of equality was recognised,
and upheld.75 Confrontation between the high and the low, or the brahmana
and the dasa, was not only to be represented but also had to be resolved in
favour of the lowly within the space created by the Buddhists through their
constant interrogation of the brahmana’s claim to social and intellectual
supremacy.
In the Setaketu Jataka, a famous teacher has a brahmana pupil who is very
proud of his high-caste status. One day he sees a chandala (untouchable)
and, fearing that the wind after striking the chandala’s body might strike his
own and pollute him, he cries out, ‘Curse you, you ill-omened Chandala, get
to leeward’, and goes quickly windward himself. The chandala, however, is
too quick for the brahmana and stands to windward of him; clearly defying
the brahmana in doing so. The chandala then offers to do as the brahmana
says if he can answer his question. In making this offer, he is clearly carrying
his contest with the brahmana into the intellectual arena which is
supposedly the basis for the brahmana’s social supremacy. The brahmana is
so confident of his intellectual powers that he agrees to answer the question
that the chandala puts to him in the presence of an audience. However, the
brahmana fails to answer the question and the chandala then forces the
proud brahmana to get down on his knees and humiliates him in full view of
the gathering of pupils.76
A more tempered version of an intellectual confrontation between a high-
caste brahmana and a poor dasi girl is found in a different genre of Buddhist
texts, the Therigatha. This text contains a collection of verses composed by
women who had joined the Buddhist monastic organisation. Verses
attributed to Punna, who had been a dasi in her lay existence and had
hauled water for the family of her master, query a brahmana bathing in icy
cold water in the river about why he subjects himself to such misery when
he has the option not to do so (this is an option she does not have). Says
Punna:
Drawer of water, I (go) down to the stream
Even in winter in fear of blows,
Harassed by fear of blame from mistresses,
What, brahmin, fearest thou that ever-thus
Thou goest down into the river?
Why with shivering limbs dost (thou) suffer bitter cold?
The brahmana explains that he bathes in the icy cold water in order to wash
off his sins, to seek release from his ill deeds. Punna’s response then is:
Nay now, who, ignorant to the ignorant,
Hath told thee this: that water baptism
From evil karma can avail to free:
Why then the fishes and the tortoises
To frogs, the watersnakes, the crocodiles
And all that haunt the water
Straight to heaven will go.
Yea, all who evil karma (deeds) work . . .
Thieves, murderers—so they but splash themselves
With water, are from evil karma free;
Evil thou hast wrought, they’d bear away
Thy merit too, leaving thee stripped and bare
To bathe and shiver here
That even that, leave thou undone,
And save thy skin from frost.77
Punna, by thus arguing, establishes her intellectual superiority and
succeeds in making the brahmana understand that there is no easy
absolution from evil actions. She also establishes her moral superiority by
declining to accept a robe that the brahmana offers her as a token of his
esteem. In this too she scores a moral victory over the brahmana whose
caste is associated in the Buddhist texts with the unashamed pursuit of
material gains from others in the form of dakshina (sacrificial fee) or dana
(gift). Even in her poverty and misery Punna, in a quiet and dignified
manner, resolves the intellectual confrontation between the low and the
high. But while she establishes her intellectual and moral superiority over
the high-caste brahmana who stands at the apex of society, Punna manages
to simultaneously sidestep the social confrontation inherent in the
interrogation of the brahmana’s system of beliefs.
Punna’s confrontation with the brahmana is thus a contrast to the
chandala story; it remains confined to the intellectual plane. In the chandala
story there is a very interesting shift. It begins with a social confrontation:
the chandala uses his polluting power as an effective weapon of harassment.
This is followed by an intellectual confrontation in which the chandala
outwits the brahmana and then humiliates him—thus carrying the
confrontation back to the social plane. The chandala, humiliated in everyday
life, now has the humiliator in his power. In Punna there is a mature
intellectual interrogation of the beliefs and rituals of the brahmana. With
simplicity and dignity, using reason rather than abstruse philosophy, Punna
shows up the shallowness and ridiculousness of the brahmana’s beliefs, who
ends up looking comic rather than wicked or villainous; he is certainly no
figure of authority having monolithic control upon people’s minds. Punna
may be a lowly dasi having no control over her time and labour, but she has
retained through the sheer drudgery of her existence an inner space, a
common sense, which no amount of hegemonic control can reach or
destroy.78 Seen through Punna’s eyes, Brahmanic ideology itself hardly
appears hegemonic but, rather, weak and tenuous in terms of effective
control.
The major problem in the narrative of Punna is to locate how much of her
statement is the voice of the dasi without the mediation of the bhikkhu/
bhikkhuni; or does the text represent the voice of the bhikkhuni put into the
mouth of Punna, the dasi? The fact that the verses are contained in the
Therigatha, and not the more popular Jatakas, might account for the
generally restrained speech and more subdued handling of the intellectual
confrontation between the high-status brahmana male and one who, as a
dasi, represents the lowest rung in the social ladder; Punna’s low status is
compounded by her gender subordination. Or is Punna’s voice a bit of both,
that is, the voice of a dignified bhikkhuni verbalising the ‘commonsense’
perception of a dasi? We are back to the grey area of the relationship of
popular culture to the culture of the learned.
In this context, it is significant that the general Buddhist critique of
Brahmanical hierarchy and intellectual arrogance coalesces with the popular
criticism of the Brahmanical elite tradition in the Jatakas, of which there is
much evidence in the high Buddhist literature. However, the popular
criticism of Buddhist monks for not upholding the renouncer tradition with
sufficient rigour is represented in the Vinaya Pitaka, the text for the monks,
as an occasion for the Buddha ruling for more rigorous behaviour. Such
concerns hardly make an appearance in the Jatakas. Here, popular criticism
is judiciously and selectively woven into the Buddhist critique of
Brahmanical misconduct. The attempt at maintaining control over the
various levels of the narratives is at all times apparent in the way the themes
are handled.
To sum up, we may draw attention to certain issues that arise in this
reading of the Jatakas. First, in these narratives the low-status groups are
released from their roles as objects of the rules, attention and focus of the
elites, and acquire a subjectivity. They think, act, laugh, weep, subvert and
resist as much as they fall in line, but even as they do so, they retain a degree
of autonomy from the consciousness and perceptions of those who represent
high culture. In fragments one can see elements of a popular vision of the
world from below, but since the narratives are as much the creation of men
who share, in part, the values of the upper classes, the representations of the
subjectivity of those at the bottom end stop short of complete coherence.
The texts ultimately succeed in imprisoning them once more in the
anonymity and obscurity from which the Jatakas had momentarily rescued
them. Once the purpose of the appropriation is achieved, the narrative itself
is often cut short without telling us what happened to the key characters
with whom we have begun to identify. We are left with a faint glimpse of
what it was like to be at the bottom end of society.
To return to the theme of the relationship between the Jatakas and other
Buddhist texts, we can attempt to highlight the nature of the appropriation
or utilisation of different elements in the interaction between the culture of
the people and the culture of the bhikkhu. Albeit loose in structure and a
dilution of the specific philosophical content of Buddhism, the stories are
cast within its overarching world view. They present Buddhist ethics in a
popular idiom and style. In the narratives about women, the overlaps
between the two visions make possible a fairly well-synchronised
representation. The fear of women’s sexuality, which was probably widely
shared by all categories of men, is put to excellent use in bringing home to
the bhikkhus, and to the pious, the inherent danger that women represent.
Through the use of popular idiom, the rigours of enforcing chastity upon
bhikkhus are mediated successfully. But there are areas of tension too
between the values of the bhikkhus and the values of the common folk, as in
the narrative of the poor man who is executed for stealing flowers and his
dying regret about not being able to attend the festival with his wife, which
leads him, according to the bhikkhus, to hell. Thus, while the people might
see pleasure, even when momentary, as a strategy for survival, the bhikkhus
reinscribe it as the path to moral degeneration. Similarly, in the story of the
water-carrier, the festival episode is pushed aside, the woman disappears
from the story and the spotlight shifts to the narrative dealing with the
other-worldly concerns of the bhikkhus.
The Jatakas also retain the notion of social hierarchy prevalent among the
Buddhists (where there were rich and poor according to differing access to
the means of production, and high and low status according to the nature of
work pursued) and, at the same time, provide for the playing out of the
dreams and fears of the lower classes (without, however, allowing
transgressions to go beyond permissible limits). The moral world of the
Buddhists is reaffirmed, especially since the narratives simultaneously limit
the playing out of social transgressions and yet permit, or even encourage,
the countering of the brahmana’s hegemonic control over ideology through
transgressions originating in the intellectual arena. All this is achieved while
retaining a popular genre of storytelling, and letting the text give out a
multiplicity of meanings at the same time: ambiguous, subversive and
reaffirmative.
11

The Functions and Social Location of

Kāvya
SHONALEEKA KAUL

This paper was presented at an international seminar on ‘Revisiting

Kālidāsa’s Abhijñāna-śākuntalaṃ'. Land, Love, Languages: Forms of
Exchange in Ancient India’, organised by the Department of English,
Miranda House, University of Delhi, on 20–21 January 2010.

K ĀLIDĀSA’S WORLD FAMOUS ABHIJÑĀNAŚĀKUNTALAṂ is a

kāvya—a dramatic kāvya or nāṭaka. The influential tendency in
1

scholarly circles has been to interpret its contents in the light of historical
contexts external to it, such as monarchy, courtly culture and patriarchy,
which are seen to be directly expressed in the theme, treatment and
characterisation of the play. The primary constitutive factor in the formation
of the play has been understood to be royal patronage, and for this reason it
is believed to be elite literature, addressed to the court and mirroring its
ideology.
This essay suggests that the primary perspective in which the play needs
to be located is internal to the Sanskrit textual tradition. In other words,
before interpreting the play it is necessary to defer to its genre, kāvya, and to
the rules and functions of the latter that may have informed its composition.
It is also necessary to investigate assumptions about the social location of
kāvya—not only questions about who wrote them and for whom, but also
who got to watch their performance, where they were performed, and
indeed, where the text locates its plot, or the interior locale. Taking cues
from the indigenous Sanskrit rhetorical literature like the Nāṭyaśāstra
(henceforth NS), and examining aspects of language use in kāvya plays, this
essay raises questions about how we have been looking at classical Sanskrit
literature. It also urges a greater appreciation for the critical apparatus of
kāvya vis-à-vis the court, as also for its not-inconsiderable social interface.
This essay is thus not about a text called the Śākuntalaṃ, but about a
textual genre, a sub-genre of which the Śākuntalaṃ is one example. This
distancing from a play in order to study it is deliberate—and purposive. For,
locating a work of literature in its generic context makes sense not only from
a literary-critical perspective but from a historical perspective as well. Social
historians have often brought assumptions to the interpretation of the
Śākuntalaṃ that could benefit from reappraisal in the light of other
contemporary works, or aspects of the literary movement that the play was a
part of. Given that it is difficult in any case to ascertain the specific
background of early Indian texts, the less we confine our inferences to a
single text and the more we broaden our view to encompass, say, the whole
genre, the more sound the basis on which we can attempt conclusions.
The genre in question is the Sanskrit kāvya. In keeping with the usage in
Bharata’s Nāṭyaśāstra and Bhāmaha’s Kāvyālaṁkara, the earliest extant
works on rhetoric and poetics,1 kāvya includes not only poetry
(sargabandha or mahākāvya) but drama (nāṭya), tale (kathā), and biography
(ākhyāyikā). The Śākuntalaṃ, as observed above, is a nāṭaka, one of the ten
types of plays the Nāṭyaśāstra enumerates. A nāṭaka is defined as the five-or
ten-act play par excellence with a well-known (prakhyāta) story taken from
tradition or history (itihāsa).2 It is no surprise then, and well-known, that
the Śākuntalaṃ bases itself on a plot found in the Mahādbhārata. But it is a
retelling said to have marked and historically significant differences in
thematic detail and characterisation from the older version. The
conventional view is to attribute the differences to a changed historical and
ideological context from Vyāsa to Kālidāsa, so to speak. The change is
supposed to be the full development of courtly culture under the Guptas,
representing the high point of monarchy, Brahmanism, and patriarchy in
concert—ideologies which are believed to have worked their way into
literary compositions through royal/elite patronage.
 It is tempting but not foolproof, however, to read so direct a translation of
socio-political context into literary content. Based on my own study of more
than two dozen kāvyas, this essay goes into some of the problems this
approach presents. Further, drawing closely on the defining indigenous text
the NS, it queries some set notions about kāvya, which have had
implications for understanding the social location of the genre, and willy-
nilly shaped our interpretation of texts like the Śākuntalaṃ.
The influential approach to interpreting kāvya literature for historical
purposes regards it as ‘courtly literature’ ‘intended for the royal court’, or
‘class literature’ composed by poets who are no more than ‘housebirds’ of
patricians.3 Focused on the question of patronage, this view emphasises
‘who creates the form and who orders it’ as the key to understanding
literature.4
There is no doubt that artistic forms like kāvyas, that evidently require a
great degree of learning and leisure to compose, would have been patronised
or cultivated by affluent individuals and groups. This would mean primarily
the king or ministers of his court on the one hand, and wealthy magnates
(the śreṣṭhῑs and sārthavāhas) on the other. some historical examples are
available of these categories of patrons and authors. The ‘evidence’, though
not substantial by way of information, certainly serves to confirm the
picture of upper class patronage. Elite patronage of kāvyas, however, can be
interpreted in different ways. I believe the significance of elite or ‘court’
patronage, as it is generically called, lies in both its contribution to and
partaking of the prestige that Sanskrit kāvya acquired as classical literature.
True to classicism, kāvya inaugurated enduring literary standards by taking
the cultivation of language and theme to unprecedented heights and
exuding an appeal that was rooted in scholarship, yet chiefly aesthetic. For
the cultural elite to back this powerful and promising creative literary trend
would have been natural and fashionable. Their association with it, in turn,
would have fortified its status as belles-lettres.
The other interpretation of ‘court literature’ is the one I earlier cited,
which suggests that kāvyas are the expression of dominant ideologies asserted
through patronage. While this cannot be ruled out, there are difficulties in
accepting this as forming exclusively the constitutive knowledge of kāvya
literature. In its focus on material creation, this perspective has the effect of
restricting our sense of the meaningful possibilities of a text. It tends to
reduce kāvyas to pre-textual imperatives (like the patron’s needs or the
author’s class affiliation), which is to ride roughshod over intricate and
intuitive literary processes of signification. It runs against the idea of
interpreting literature on its own terms, and tends to deny autonomy and
critical agency to the author.
There is also a latent collapsing of ‘patron’ and ‘audience’ in this
understanding. This is not justified, at least and especially for the plays
among the kāvya repertory, for they were performed at festivals, among
other occasions, as the NS tells us, plausibly in public places like temples and
city squares.5 Indeed, Bhavabhūti’s Mālatῑmādhava spells out its audience
and the occasion thus: ‘A large conclave of men residing in different quarters
(nānādigantavastavyo mahājanasamājaḥ) has gathered here in connection
with the festival of Lord Kālapriyanātha and I (the presenter) am ordered by
the assembly of the learned (vidvajjanapariṣada) to entertain them by the
performance of some new play.’6 It may be observed that the audience and
the sponsors are clearly distinguished. Similarly, Vijayā’s Kaumudῑmahotsava
alludes to the autumn ‘public festival’ (sarvajanasāmānyamahotsava) as the
occasion on which it is performed, while Kālidāsa’s Mālavikāgnimitra refers
to the ‘spring festival’ in the same context.7 Interestingly, Bāṇa’s Harṣacarita
has vividly described festivals as occasions when the high and low mingle.8
On the theoretical side, types of plays like the vῑthi (‘street play’), bhāṇa
(erotico-comic monologue) and prahasana (farce), with their lively,
irreverent, often bawdy content, are suggestive of general lay audiences.
Moreover, the NS explicitly describes drama as ‘the fifth Veda for all classes
of society’ (vedaṃ pañcamaṃ sārvavarṇikaṃ).9 Far from envisaging, as it is
often claimed it does, a purely learned and cultured, connoisseurial
audience,10 this text speaks of the presence of various categories of
spectators in the assembly: the superior, the inferior, and the middling
(uttamadhamamadhya); in fact, there is an explicit reference to ‘fools’
(mūrkha) and children in the audience!11 It is true that the NS first
prescribes a long list of qualities like high birth, good character, proficiency
in music, prosody, and all the śāstras, a sensitive nature, etc. for a spectator.
But this is only an ideal, as the text goes on to realistically observe: ‘All these
various qualities are not known to exist in any single spectator’. It also
distinguishes spectators (prekṣaka) from ‘assessors’ (prāśnika), who are the
experts, like the actor, prosodist, grammarian, musician, courtesan, etc. who
are called in to judge controversial performances during dramatic
competitions.12 Indeed, the NS's final take on the matter of audience is that
‘different are the dispositions of women and men, young and old, who may
be of the superior, middling and inferior type, and on such dispositions (the
success of) drama rests’.13 The Mālavikāgnimitra echoes this when it says:
‘Drama is the one entertainment for a variety of people with very different
tastes’ (nāṭyaṃ bhinnarucerjanasya bahudhāpyekaṃ samārādhanaṃ).14
Thus, the audience (and, therefore, the materialising context) for the plays
was most likely to be mixed—and not necessarily literary or even literate. By
this last observation I draw attention to the highly visual component of
theatre in early India. Language was considered but one means among
several others equally symbolic. Thus, enactment through dialogue or
speech (vācika) was only one out of four modes of histrionic representation
(abhinaya) mentioned by the NS. The other three were gestural (āngika),
including movements of the face, limbs and body; sartorial and make-up
(āhārya); and significatory (sāttvika), including tears, horripilation, etc.15
Even as ‘words were the body of the dramatic art’, the latter three ‘clarified
the meaning of words’.16
Indeed, the words were to be sung or recited to a variety of metres, so that
together with the importance of body movements, the central role of music
and dance in any play is indicated. Altogether then, dramatic performances
had a very large element of pantomime and spectacle: not for nothing was
drama known as dṛśyakāvya (visual kāvya). This should qualify the
assumption that literacy and learning were prerequisites for the
comprehension and consumption of this kind of kāvya, even if they were for
its creation. Note also that those who performed the plays—the actors and
actresses (naṭa, naṭῑ)—were traditionally regarded as belonging to low class
and caste. That they still did justice to these plays and presumably to the
laborious instructions of the NS, underlines the largely oral character of the
theatrical enterprise to which formal literacy may or may not have been
relevant.
The point then is that aspects like venue, audience and pantomimic
nature of the performance of texts like the Śākuntalaṃ suggest that the
context of consumption of these works—as I like to call it—unlike the
context of production or patronage, cannot be assumed to be confined to
aristocratic circles. It could have been distinctly non-elite or mixed. And this
must have shaped the reception of ideas and representations in plays, which
reception can, in turn, be expected to have influenced/informed the intent
and content of the compositions, especially parts that were critical of
authority, bringing out its often illegitimate doings, as well as revealing the
contradictions and contraventions in relations like gender. In this
connection, note that the functions or goals of kāvya as set out in the
rhetorical literature include entertainment (creating delight—prῑti, harṣa,
vinoda), which underlines the centrality of the audience to drama, and
instruction (upadeśa), which points to both the communicative and critical
goals of the genre.
However, I do not argue for seeing kāvyas as a direct part of anything
approximating mass culture, if we can even speak of such a category for
early India. One reason for this is that kāvya includes not only drama but
also poetry and prose, some of which, notoriously like the kādambarῑ,
undoubtedly gets a great deal more complex and dense, both structurally
and linguistically, and would have been proportionately less accessible, than
the plays. The other, more important reason for withholding any sweeping
suggestion as to the social location of this literature is that intriguingly little
is definitely known about literary transmission and linguistic practice in
early India.
This may not be immediately obvious from the reams of serious
scholarship on the history and sociology of languages in India. Or from
well-established views about the ideology of languages like Sanskrit, which is
believed to have been not a spoken language but an esoteric language of
learning and ‘in particular the language of the brahmin caste and of its
religion’, while Prakrit is deemed a popular tongue, ‘the language of the
masses’.17 It can be seen how this view both feeds into the courtly assessment
of kāvya and proceeds from it.
Specifically, one may note the related, popular theory that in the plays,
Sanskrit is spoken only by men of high status while lowly characters and
women speak Prakrits.18 This, it is maintained, reflected the reality of the
elite language/popular tongue divide in early India. In other words, a
hierarchy is assumed to exist in the way Sanskrit and Prakrit are mutually
deployed, and this is believed to correspond to socio-economic hierarchy.
This is such a persuasive interpretation that the fact that there are telling
exceptions drops quite from view. For example, the vidūṣaka or the jester
and hero’s companion is a brahmana, but always speaks in Prakrit.
Merchants and courtiers are also required to do so.19 On the other hand,
parivrājikās or female ascetics use Sanskrit, even though queens usually do
not and courtesans may, while children—even when male and high-caste—
speak only Prakrit. Indeed, the section on use of languages in the NS attests
to the large number of characters and situations that can be seen as
exceptions. The grounds for allocating languages range from
characterisation, ethnicity and vocation of protagonists to regional phonetic
peculiarities of audiences, as well as blanket considerations like ‘special
times and situations’ or ‘reasons and needs’.20 Clearly, kāvya’s understanding
of the use of Sanskrit and Prakrit was a self-conscious one, and so too its
choice to be bilingual (or rather, multilingual) in the first place; but the logic
was apparently not as simple or stock as reflecting social status.
This is reinforced by the fact that, as several Sanskritists have pointed out,
the Prakrits in kāvyas are very much literary, not ‘popular’, dialects—as
cultivated, standardised and universalised as Sanskrit.21 At the same time,
the term that Patañjalῑ down to Bhartṛihari and other grammarians use for
Sanskrit—‘bhāṣā’ or spoken language—shows that Sanskrit, in its turn, may
well have been generally known and actually used, though its correct usage
may have been within a restricted circle. Scholars have cited extensive usage
to this effect, even though a majority of others may not agree.22 The point is
simply that when this is seen together with the confusion over Prakrit, we
cannot say with certainty what was indeed the spoken tongue(s) in early
India. The Kāvyādarśa, one of the most influential theoretical works on
kāvya composition, is understood to refer to both Sanskrit and Prakrit as
‘śṣṭta’ or (grammatically) disciplined/systematised.23 They were both (as well
as Apabhraṁsha) regarded as languages of kāvya composition, and were
used as such in actual practice, for example, compositions like the
Setubandha, Haravijaya and Gāthāsaptaśatῑ in Prakrit, so that no literary
privilege is inherent in the choice of the languages.24 The kāvyādarśa does
describe Sanskrit as ‘daivῑ vāk' or divine speech but this seems to have been
on account of it being perfected speech ‘worked out by the sages’
(anvākhyātā maharśibhiḥ).25 The choice of language seems to have been
more a matter of genre and metre used.26
Moreover, the fact that they were employed in the same play shows that
Sanskrit and Prakrit must have belonged to a common speech community.
A mutual intelligibility for the two on the part of the audience must be
assumed.27 One may also note the NS’s instruction that a play should
‘contain no obscure and difficult word and be intelligible to country people’
(gūḍhaśabdārthahῑnaḥ janapadasukhabhogyaṃ).28 Together, these
observations make a case for not only reviewing the supposed subalternity
of Prakrits but rethinking Sanskrit as a rarefied language and literary culture
of little wider reach or relevance,
Certainly, most of the main characters and plots of kāvyas comprise what
would be called an elite line up. But in that case, labelling the texts as elite
literature does not tell us anything more than what the texts already tell us.
Instead, it may mislead one into the expectation that received codes of high
culture or behaviour will be confirmed. As I have tried to show in my book,
this is a gross assumption for a highly sophisticated literature such as kāvya:
the texts do not always appear committed to reproducing power structures
and hierarchies determined by caste, wealth, gender, ritual authority, and
political institutions like monarchy, but instead critique them at every turn.
Limitations of space do not allow examples here, especially of the stinging
critique of the acts of the monarch in the Śākuntalaṃ and several other
kāvyas like the Mālavikāgnimitraṃ, Mṛcchakaṭikaṃ, Daśakumāracaritaṃ,
Mālatῑmādhava, Vāsavadattā, and so on. Suffice it to say that there is a need
to separate the literate and ornamented form of kāvya from the concerns and
vantages of at least some of the texts, which could extend to the underdog,
the menial, the criminal, the rebel, even the feminist, and so on.29 Hence the
‘courtly’ label should not obscure the polyphonic and critical apparatus of
the texts: the social limits of kāvya do not deny its elite origin but were not
identical with it.
At the least, then, the label ‘elite’ or ‘courtly’ for kāvya could do with some
qualifying. As high class or caste, linked with an emphasis on patronage, I
believe it is helpful, but not as much as it is leading;30 as meant for the
intelligentsia, it can have potential. While it can be argued that the two
almost always overlapped in early India, the latter allows for the attribution
of a discerning voice to the kāvyakāras. And it is this voice more than
anything else which gives to this body of literature its rightful place as a rich
historical document.
12

Bards and Bardic Traditions in Early

Tamil Poetry
K. KAILASAPATHY

O Fmost
ALL EPOCHS IN THE HISTORY OF A NATION, THE ONE that is
conducive to the cultivation of the art of minstrelsy is perhaps the
Heroic Age. The age itself seems to have been a transitory period. But within
the short space of time conditions were usually favourable to the minstrels.1
The kings and chieftains vied with one another in their generosity to the
minstrels and bards: partly as a means to get their own fame sung and
perpetuated. In the event, the royal support proved beneficial for the art.
Under these ideal conditions ‘it attained a perfection which has not since
been equalled’.2 However, this high development was not a sudden
achievement. It was the result of generations of artistic effort, itself sustained
by a tradition, the origin of which can be seen only dimly. As might be
expected, minstrelsy was a specialised discipline at this stage, instructing
and exercising its practitioners over a period with a view to perfecting their
art. One is afforded a glimpse of these things in the poems. In one sense this
art was the metrical expression of the imagination of a group through the
medium of individuals.
Early Tamil poetry mentions many types of bards or minstrels. The most
important ones amongst them were pāṇar, kūttar, porunar, viṟaliyar, pulavar,
and akavunar. There existed differences among these types of bards, both in
their way of life and in the nature of their craft. But all evidences point to the
fact that they were oral bards. In fact, nowhere in the texts of the extant
poems is writing mentioned in the context of the making of a poem or a
song. Besides this negative evidence of writing not being mentioned in the
poems, the various adjectives qualifying the poems and bards themselves
conclusively denote oral composition. The bards always mention the singing
of songs. It is either the tongue or the mouth that is spoken of as the
medium of the poems. We may cite two examples. Lamenting the death of
an unknown hero, a bard says ‘having established his fame far and wide, he
now lives in words that come from the mouths of the bards whose tongues
are well trained’.3 In describing the court of the learned men that helped a
king a bard says ‘the assembly of the nobles with well-trained tongues’.4
These descriptions show us that the predominant mode of learning and
expression was oral in character. Viewed in the context of the vestigial
remnants of mantic wisdom which the heroic bards had inherited, this
repeated statement about oral composition and oral transmission of
knowledge becomes very clear. With these preliminary comments, we may
now examine the different kinds of minstrels mentioned in the poems.

THE PĀṆAR

The pāṇar were minstrels who sang their songs to the accompaniment of the
yāḻ, ‘lute’. In medieval times the word came to denote a lower caste. But in
the early poems not only do we note the absence of the caste system, but also
find that the pāṇar were held in high esteem as a vocational group. In a
poem enumerating the names of excellent flowers, food, gods, and other
things, a bard speaks of the pāṇar as one of the four noble clans. In passing,
it is interesting to note that another ancient clan—the paṛaiyar,
‘drummers’—enumerated under the category in the same poem was also
degraded in the caste hierarchy of latter times.5 Out of it arose the modern
usage of the word ‘pariah’. In the heroic poems the pāṇar are seen to be
singers par excellence though often in chronic poverty. Derived from the
word pāṇ is the word pāṇi, meaning ‘beat, time, melody, music, song’, etc.6
Later, traditions classify these minstrels into three groups: ‘vocalists’, ‘singers
with the lute’, and ‘singers with the begging-bowl’. In the heroic poems we
see the minstrels who sing to the accompaniment of the lute. The instrument
was of two kinds: the ‘little lute’ and ‘large lute’. The minstrels were called
‘singers with the little lute’ or ‘singers with the large lute’, according to the
instrument they adopted for their recitals. This is evidenced by such
expressions as ‘your left hand supports the sweet-tuned little lute, well set
with twisted strings that look like golden twines’7 and ‘you play the large lute
constructed in harmony with the prescribed rules’.8
That the lute seems to have been a favourite instrument in that age is also
proved by Tolkāppiyar’s reference to it under the category of Karupporuḷ,
which literally means ‘seed topics’, but has the wider meaning of the various
objects—gods, flora, fauna, occupation, music, and the like—associated with
the five physiographical regions. Another fact mentioned by the
Tolkāppiyam and also documented by several poems gives us a glimpse of
the social position of these minstrels: they were, as it were, intermediaries
between the heroes and heroines in their love affairs and domestic quarrels.
To that extent they must have had easy access to the inner houses of rich
families. It has been seen that a bard like Kapilar was respected sufficiently
by Pāri to have been entrusted at his death with the care of his two
daughters. But Kapilar was a court bard who lived in the company of his
patron for several years. The pāṇar were mostly wandering minstrels,
travelling in groups from one place and patron to another in search of
largess. These wandering minstrels are often compared to ‘bats which seek
the trees with ripe fruits’.9 In fact, the poems clearly suggest that hunger and
starvation cohabited with them. One poem says that King Kōpperuñcōḻaṉ
was an enemy to the hunger of minstrel families.10 Likewise, Nalaṅkiḷḷ i is
praised as the destroyer not only of his foes, but also of the hunger of his
bards.11 A description in theMaturaikkānci indicates that these minstrels
lived on the outskirts of the cities in small settlements. The suburbs of
Maturai and Pukār had minstrel ‘colonies’. Being professionals, they do not
seem to have been engaged in any other activity. However, a few references
in the poems show us some of them trying to fish in rivers with rods.12 A
young girl of the minstrel family is said to barter fish for jewellery.13
They travelled about all over the lands in groups. The repeated reference
to the ‘very large body of kinsfolk’ travelling with the minstrel calls for some
comment. The wife of a pāṇaṉ was called pāṭiṉi or pāṇicci, ‘woman minstrel’.
It is probable that she sang with her husband or mimed and danced to
songs.14 It is not clear what the very large group of kinsfolk did as far as the
performances were concerned. But the youths in the troupe appear to have
been travelling with a purpose: they were apprentices, doing odd jobs for
their elders as well as learning the art of minstrelsy. One poem says that
while a minstrel and his female companion were presented with gifts by a
king, youthful boys sang the glories of the battlefields with tender words and
their hearts full of joy.15 The context clearly suggests the inexperienced but
pleasant singing by young minstrels who were probably at the stage of
apprenticeship. It is also plausible that these minstrels belonged to
vocational groups organised on the basis of clans.
THE PORUNAR

Porunar were also minstrels whose exact identity is not easy to ascertain.
The word itself means several things: ‘a warrior, a person who plays on the
small drum, a person who is compared to another, an incomparable one, an
enemy’.16 It is also the proper name of a minor chieftain.17 The word root
poru means among other things, ‘to fight’ and ‘to resemble’.18 From this root
is derived the noun porunaṉ, ‘a warrior’ and ‘a person who is compared to
another’. These two meanings seem to provide the clue to its signification of
a minstrel or bard. The usage of the word meaning bards, in
contradistinction to its connotation of warriors, is clearly brought out in a
poem.19 The word porunaṉ is qualified adjectivally by a negative relative
participle porā, literally ‘fighting-not’ or ‘non-combating’. Hence the
expression means ‘non-combating warrior’, i.e., ‘war-bard’ who travelled
with warriors. We are told by Naccinarkkiniyar that the bards were of
different kinds: those singing of ‘ploughed fields’, battlefields, etc. In the early
poems now available to us there is no example of bards singing in praise of
ploughed fields. But there are faint signs of it. One of the expressions used
for praising persons or things is the optative polika, ‘may [it] prosper’. The
verb poli means ‘to flourish, prosper, abound, increase, live long and
prosperously’.20 The meanings of its cognates in the other Dravidian
languages strongly suggest that its primary meaning has agricultural
implications: e.g., Malayalam polikka, ‘to measure corn heaps, paying the
reapers in kind’, etc. Furthermore, in the bardic poetry itself we have a few
specimens of expressions which indicate that there was a practice of hailing
paddy fields, rain, and fertility.21 We may cite an instance from the Paripātal,
which incidentally contains apostrophes to the river Vaiyai. People dropped
in the river conches, crabs, prawns, and fishes made of gold, with chants of
‘viḷaikal! polika!’. It is clear that these objects were intended to promote
fertility.22 What is relevant to us here is the fact that the same form of
expression is used in praising the bounty of kings in battlefields.23 It is likely
that these expressions were but parts of a larger group that was commonly
used for the purpose of hailing, praising, and invoking natural
phenomena.24 Another indirect evidence that points to the prevalence of
such bards, and thereby ‘agricultural’ poems, is the presence in the heroic
poems of a few detailed and elaborate similes drawn from the fields—
particularly the scene of the harvest festival. The Tolkāppiyam and the
Purapporul Veṇpāmālai have noted them.25 The scholiasts, too, have pointed
out that these similes are analogous to harvest rituals. In fact, the analogy is
so complete that one could without difficulty reconstruct the harvest ritual
from the descriptions. From the foregoing evidences, we may not be wrong
in inferring that the bards impersonated the people they were singing of. It
is also probable that when they sang of the ploughed fields, they dressed up
and appeared like farmers; similarly, their role on the battlefield made them
appear like soldiers. In other words, they impersonated soldiers. Only such
an explanation—tentative though it may be—would clarify the meaning of
the word porunaṉ as a person comparable to another. This aspect of the life
of the war-bards is made still clearer by the fact we know that they
accompanied the warriors to the battle fronts, stirring them up with their
martial songs. Here again it may be said that they were singing, like
warriors, of the exploits and glories of their forefathers. This would have
made the identification very close.
Perhaps the principal factor that characterised these bards—
distinguishing them from the other types of minstrels—was their closeness
to the princes and chiefs. As has already been noted, they formed part of the
retinue of kings both at home and in war. In this sense they may properly be
called war-bards. Compared to the minstrels their life was relatively secure
and stable. Probably they also wandered less than the others. A poem
referring to a group of war-bards makes this point abundantly clear:
I am the leader of the war-bards living in the cool and fertile Cōḻa
country, mighty in power and flooded with goods brought back from
enemies’ ships defeated on the seas. We are the privileged war-bards of
Nalaṅkiḷḷi, chieftain of steeds bedecked with head-jewels that swing to
and fro. We do not sing about others; nor do we wish to receive gifts
from anyone else. We always sing the glories and praise the victories of
him our lord . . . .26
Similar sentiments are expressed by other war-bards claiming special
patronage from one king or another.27
The considerable security of livelihood at the courts of kings was perhaps
not given to every war-bard. Consequently, we also see groups of itinerant
bards. The festivals held in various parts of the country attracted them. A
war-bard is thus addressed in a poem:
O bard! you are wise enough to quit rich, great, wide-placed hamlets at
the end of the festivals, not caring for the post-festival food. You have
the knowledge to seek new places holding festivals.28
If the lute was the primary instrument of the minstrels the taṭāri or kiṇai,
‘small drum’, was the basic accompanying instrument of the bards. This
difference in the type of instrument may be said to indicate some difference
in their respective arts too. The lute is a stringed instrument producing a soft
and mellifluous music; the drum produces a rough and ‘martial’ effect. This
being so, it is not surprising that the bards have contributed a good deal to
heroic literature. Like the minstrels, these bards, too, operated in groups. A
group consisted of whole families, including the kinsfolk. The
Poruṇarārruppatai presents a picture of a large bardic ‘family’ on the
move.29 Among them, too, the young ones cultivated the art of bardism by
listening to the elders and by practice. We have a reference to young
apprentices singing in the court of the great Karikālaṉ. ‘Hearing my pupils
who stand beside me sing his praise, he sent for us and we greeted him with
due respect.’30 It is not possible to determine the exact number of pupils or
apprentices accompanying a leading bard. All that one can say is that they
are invariably mentioned in the plural.

THE KŪTTAR

Kūttar were a type of minstrel who danced as well as sang. Kūttu means
‘dance, dramatic performance’, etc.31 In earlier times it covered ritual dances
as well. But different rituals had different names for their dances. In the
heroic poems kūttar seem to be dancers-cum-actors, who performed some
sort of choral dramas. The themes were probably drawn from heroic and
love stories. Kūttar and āṭunar are synonyms. There were also other words
with minor differences in meaning, of which we shall speak shortly. In
general, there does not seem to be much difference between the pattern of
life of these ‘dancers’ and that of the minstrels and the bards. Whatever
differences existed concerned the nature of their art. Naccinarkkiniyar has
something interesting to say on their art and life: ‘The kūttar have no caste
distinctions; their dance [acting] expresses the eight types of [basic]
sentiments and all the implications of the mind.’ The subject of sentiment is
treated by Tolkāppiyar in ‘the chapter on the physical manifestation of
sentiments’. Perhaps, the most laconic definition of the subject is found in
Ceyiṛṛiyam—the lost book on ancient dance and drama—quoted in
fragments by medieval glossators and scholiasts: ‘The experience of the
subject, made apparent to the viewers, is what the wise-ones called the
physical manifestation.’ From what Naccinarkkiniyar has pointed out, the
dancers must have been professionals, acting and singing, or rather
performing, choral dramas. As for Naccinarkkiniyar’s comment on their
social position, it is likely that in his time they did not form a caste like, for
instance, the pāṇar. But we do know that like the minstrels and the
‘drummers’ they, too, had by this time become degraded in the social
hierarchy. The degradation was so far complete as to associate them with
drunkenness and immorality. As for the word kūtti, the feminine of kūttan,
it acquired the meaning of a prostitute.32 It is probable that the kūttar were
not organised on any social basis, although it has been suggested that they
might have had corporate organisations like guilds.33 Furthermore, it is
likely that, while the cultivation and continuation of poetry had become a
highly specialised art of certain clans and families, dancing continued to be
more free and popular with greater emphasis on communal aspects. The
frequent references in the poems to large groups of men and women
dancing in public places on festive occasions would also tend to suggest that
dancing was popular from early times. In any case, in Tamil literary
tradition, these dancers were considered a type of minstrels, the reason
perhaps being that their performance included a good deal of singing. In the
same way as most of the singing of all types of bards was accompanied by
one instrument or another, dancing too was linked with singing. Moreover,
it is difficult to separate song, dance, and music in the early phases of their
development; earliest drama combines song, dance, and impersonation
[emphasis added].
It is not possible to know, in the present state of our knowledge, details
about the choral dramas they performed or had in their repertoire. None of
them have survived.34 The extant heroic poems are in the Akaval metre, best
suited for narration and declamation rather than for intricate singing or
dancing. The kali and pari metres certainly throw some light on the
development of choral lyrics, but their relation to choral drama is yet to be
determined. Hence, we should be correct in assuming that the heroic poems
had a separate development, at least as far as the metre was concerned. The
lyrical and musical aspects of the songs of the dancers are indicated by
recurrent references to various types of paṇ, ‘melody type’, and palliyam,
literally ‘many instruments’, meaning an ‘orchestra’.35 This composite nature
of the ‘dramatic’ art of the dancers necessitates the co-ordination of different
talents: consequently a troupe of dancers is considerably large. As Dr M.
Varadarajan remarks, the variety and complexity of themes, too, might have
required a large number of actors.36 Here again we may not know the
precise number of artists that made up an average troupe. From various
descriptions of them we get some idea of the instruments they used. It was
suggested that the lute and the small drum were the primary
accompaniments to the pāṇar and the porunar respectively. It is here
suggested that the muḻavu, ‘drum’ of a medium size, was probably the basic
instrument used by the dancers to keep the time measure or beat. In several
poems this drum is given the first place when the various instruments of
dancers are mentioned or enumerated.37 A similar procedure may be
observed in the case of the lute, whenever the minstrels are described. We
may cite two examples of the use of muḻavu: ‘The melodious well-strapped
drum is used by the actresses on the dancing floor.’38 ‘The dancing girls with
bright foreheads danced keeping time to the tuneful lute and paste-smeared
drum.’39
Various instruments were used by all kinds of minstrels, but the dancers
seemed to have used the most. A poem says,
the lute, medium-sized drum, one-sided drum, hautboy, flute, and
many other instruments, were carried in bags by them.40
Another poem says that the wife of a dancer, herself a danseuse, had a one-
sided drum, medium-sized drum, and a flute.41 The bard Vañparaṇar
describes some instruments of these dancers. Impersonating a dancer and
addressing the accompanists, he says, ‘0 danseuse! I shall sing a song; you
apply the paste to the muḻavu; you set the tune in the lute; you play the
elephant-trunk-shaped hautboy; you play the ellari;42 you play the ākuḷi;43
you gently tap thepatalai on one side.’44 Referring to a group of dancers
carrying many well-tied bags full of musical instruments, a bard compares
them to jack trees studded with fruits.45 It is from such references that we
gather that they had an orchestra playing for their performances. The
accompaniment of an orchestra of different instruments is itself an evidence
of the elaborateness of the dance form. There appear to have been two kinds
of orchestra: the big and the small. The orchestra continued to become more
complex, and by the time of the Cilappatikāram there were as many as
thirty-one types of drum, besides other instruments.46 Likewise dancing,
too, had developed to a fine art.47 The music had already developed many
melodies. One poem refers to twenty-one melody types and the scholiast
gives details about them.48 Elsewhere, the dancers are said to sing the
melodies of kuṟiñci and marutam, two of the four major melodies.49
Like most of the other minstrels the dancers, too, were itinerant
entertainers, constantly seeking new patrons and audiences.50 It is probable
that they enjoyed a great deal of popular support, since they performed in the
common-grounds of villages [emphasis added]. These grounds were called
maṉṟu or maṉṟam.51 It is said that their presence in the village common-
grounds made the people mirthful, and their departure was most regretted.
It made the place empty and lifeless.52 The frequent reference to public
dancing clearly proves the fact that it was a general means of enjoyment and
had social significance too. The regular place of their performance was
called āṭukaḷam,53 literally ‘dance-field’, like ērkhaḷam, ‘ploughed field’,
pōrkkaḷam, ‘battlefield’, etc.; kaḷam also means floor.54 It has been seen that
the ritual dancing-floor was called ‘the floor of the frenzied dance’. It is likely
that a non-religious or secular dance floor was termed just ‘dance floor’. The
dancers also frequented festivals, where they would find a ready audience.
The borderline between dance and acrobatics is admittedly difficult to draw,
and, not surprisingly, some of these dancers appear to have performed rope-
dancing as well. A simile refers to ‘a dancing girl who feels much exhausted
dancing on a tightrope to the music played by high-pitched, tuneful
orchestra’.55 Elsewhere, reference is made to a dancing girl performing on
the tightrope to the accompaniment of many instruments.56 On these
occasions they travelled in groups. Expressions like ‘very large body of
kinsfolk’, ‘the relations’, ‘the large family of dancers’, etc., illustrate this fact.
Festivals, games, and encampments of soldiers had a natural attraction for
them.57 A bard praises the Cēral country as an extensive fertile land, holding
festivals where the kūttar danced with their drums.58 Several poems record
the fact that, besides kings and nobles, even ordinary people patronised these
dancers. There are suggestions of their travelling some distances to watch
dance performances. Groups of people in the agricultural tracts, who had
worked hard dumping hay and sand to prevent the river banks from
overflowing, returned to their places (homes) after attending the festival
where pleasant dance-drums resounded in the old settlements.59 At the end
of the working season, when the ploughs are idle, when there is no rain and
the sky is clear; when the moon is bright and full, the streets are lit in rows
and garlands are hung, the people join in merrily to celebrate the festivals.60
All this makes it quite clear that dancing was extremely popular. As just
observed, the common people too amused themselves by dancing. Four
types of dancing have been distinguished: folk dances, ritual dances, demon
dances, and war dances. As has been aptly remarked, dancing was closely
connected with every incident in the lives of the people.61 It may be noted in
passing that there is reference to performance of puppet-plays dealing with
religious themes.62
Among the dancers, too, the young ones travelled along with their elders,
learning the art by imitation and practice. Like young apprentices among the
other types of minstrel, they did odd jobs like carrying the musical
instruments, etc. They also sang on some occasions.
The dexterous youths carrying tight bags holding all the instruments
needed for the performances—melodious muḻavu, patalai, and others,
along with the long flute made by piercing the joints in the bamboo—
sing the glory of god.63
Another poem states that the shoulders of these youths had scars as a result
of carrying many bags or poles.64 According to the descriptions, they carried
their things in bags hanging on staffs over their shoulders.

THE VIṞALIYAR

The viṟaliyar were female dancers and singers. The word is always explained
as the feminine of kūttar and porunar. No doubt they were the wives of these
dancers and bards. But they seem to have lived and functioned as artistes in
their own right, and not merely as the talented wives of dancers and bards.
As has been pointed out earlier, the Tolkāppiyam recognises them as an
important class of minstrels. So does the Purapporul Veṇpāmālai, which says
that a viṟali sings the praise of mighty kings.65 The word is probably derived
from viṟal meaning ‘victory, strength, excellence’, etc. We may suspect here
an analogy with the word porunar, which denotes bards impersonating
victorious warriors, that is ‘war-bards’. It is suggested that these women
dancers or singers dealt with heroic and martial themes. What has been said
while arguing the case for the probable development of the meaning of the
word porunar may mutatis mutandis be applied here. Medieval
commentators have attempted to explain the word by an etymology which
seems to be doubtful. They equate the word viṟal with the Sanskrit satva,
meaning physical expression of sentiments. Thus they would explain
viṟaliyar as ‘women capable of physical expression of sentiments’. While the
description may be correct, the derivation is open to question. The word
satvam is an intruder in Tamil dramaturgy and cannot be postulated for this
period. One has to explain the word in the light of its ancient meaning. It
may be argued that, while dancing and singing was undoubtedly their
profession, they were called viṟaliyar because of the heroic themes they
performed—the victories and glories of princes and chieftains. And if it may
be argued that preceding the heroic songs there existed magical and ritual
literature, then they may be said to have sung and danced the glory of gods
too. There is some evidence pointing to primitive ritual dances in which
female dancers (demonesses) celebrated the victory of the youthful
Murukaṉ over evil Avuṇar, ‘Titans’. They are said to sing and dance on the
‘victorious battlefield’. The Tirumurukārruppatai has described the scene in
some detail.66 Thus, we may be correct in concluding that these artistes were
female singers and dancers who largely, and at any rate originally, dealt with
the theme of ‘victory’, either of gods or of men.
The sweet voice of these dancers and their knowledge of music is
especially mentioned in the poems.67
The sweet voiced danseuse sings and dances in harmony with the small
lute with a black handle, on which is played the marutam melody. The
songstresses who know how best to do their duty, praise first the
almighty god in accordance with age-old tradition; then they sing new
songs and praise the king.68
Here we clearly see that though they were basically classified as dancers their
virtuosity in music is especially mentioned. It is not surprising since they
were artistes performing choral dramas.
Like the other types of minstrel of that age they were also great wanderers.
The ‘guidance’ poems purporting to guide minstrels to patrons of art and
literature deal with female dancers too. Many poems describe their travels
and experiences. They carried their instruments.69 They travelled across
forests and arid lands.70 There is some suggestion that they also visited
battlefronts and military camps. In a poem addressed to a danseuse, a
minstrel suggests that they visit the Pāṇṭiyaṉ king, Peruvaḻuti, at the
battlefield, where he has had a thundering victory.71
As far as the musical instruments were concerned, there does not seem to
have been any difference between the male and the female dancers. Like
others, they carried their yaḻ, ākuḷi, patalai,72 flute, and muḻavu.73 Along
with their unquestionable talent for dancing, their excellent singing is often
singled out and mentioned. Their voice at times resembled the sound of the
lute.74 Their physical beauty and cultivated charm appear to have enhanced
their art. In most references to them, their long, twisted, dark plaits, pleasing
smile, bright forehead, painted eyes, slender waist, white teeth,75 and other
features and significant details are vividly described.76 These qualities and
features were complementary to their singing and dancing. In later times
they became courtesans and concubines, and by medieval times had been
reduced to harlots or prositutes, with all the attending associations of
immorality and depravity. Among the factors that created an unfavourable
climate for the bards, singers, and other artistes, must be included the strong
puritanical attitude of the Jainas and Buddhists, who treated them as
emblems of immorality.77
The female minstrels were highly respected in early Tamil society, and
their contribution to poetry and music is considerable. In the extant works
we have more than twenty women minstrels and about 140 poems attributed
to them. It has long been recognised that women minstrels have played an
important role in early Tamil poetry. And it was from their ranks that many
rose to eminence. Perhaps no better example exists than the case of Auvaiyār
—undoubtedly a great woman minstrel whose range included the whole of
Tamil literature.78 In a poem singing the praise of Atiyamāṉ Añci she
identifies herself as a danseuse.

KŌṪIYAR, VAYIRIYAR, KAṆṆUḶAR

Along with the dancers are mentioned these three types of artistes. It has
been observed that they are synonymous with kūttar. No doubt in the
poems they are treated like them. But an examination of the words would
show that the first two, at least, connote instrumentalists. The word kōṭiyar is
derived from kōṭu, ‘blowing horn’.79 The horn of animals is perhaps its
primary meaning. Thus, strictly the word would mean ‘those blowing the
horn’. Dancers blew the horn during rituals. This is clearly seen in the
Tirumurukārruppatai: during the worship when Murukaṉ is invoked, many
tuneful instruments are played. While the dancing-arena resounds with
sweet music and frenzied dances, numerous horns blow and harsh bells
ring.80 Elsewhere in the same song, the priest-dancer plays the flute and
blows the horn and sounds other minor instruments. Thus, we see that the
horn was used during ritual dancing. It is not possible to trace back how the
people blowing the horn came to be identified with regular dancers. But the
meaning is quite clear. It might, of course, go back to a stage when music
and dance were undifferentiated.
The word vayiriyar is derived from vayir, ‘bamboo, bugle’.81 Hence
vayiriyar are buglers or trumpeters. In the Tirumurukārruppatai a vayir is
mentioned as one of the instruments played to greet Murukaṉ. ‘In Alaivāy,
towards which Murukaṉ travels fast in accordance with his usual practice,
celestial drums sound; bugles blare loud; white conches call.’82 In most
instances vayir is mentioned together with vaḷai, ‘conch’. They are also said
to sound loud in battlefields.83 Here again, as in the poems, we do not know
how exactly the ‘buglers’ came to be identified with the dancers. The sweet
sound of a bugle on the dance floor is referred to in a simile in a love song.84
The buglers seem to have played the drum too. The thunder of rain clouds is
compared to the sound of their drums.85 In another song their singing is
praised as faultless.86 This shows that besides playing the instrument and
dancing they also used to sing.
The derivation of kaṇṇuḷar is not clear. The word is not found in A
Dravidian Etymological Dictionary. The Tamil Lexicon gives the meaning as
actors, dancers, masqueraders, and filigree workers.87 But its treatment of
this word would seem inadequate, in that it falls back on the Cilappatikāram
for explaining the word’s meaning. Visvanatha Pillai gives the meaning of
the word as ‘artificers, persons in masquerade’.88 Likewise he gives the
meaning of kaṇṇuḷ (kūttu) as ‘a masquerade ball’. Unfortunately he does not
cite any authority for such a derivation. Apparently he translates it by
splitting it, kaṇ, ‘eye’ + uḷ, ‘inside’ = ‘eyes inside’, thus concluding that
kaṇṇuḷar were dancers covered by masks. Ingenious as if may seem, the
interpretation cannot be accepted. Naccinarkkiniyar suggests yet another
etymology: ‘Since they “capture” the eyes of those who look at their art, they
are called kaṇṇuḷar.'89 In deference to the celebrated commentator, the least
that can be said of his explanation is that it is a classic instance of folk-
etymology. It appears that the word had undergone a radical change of
meaning by the time of the post-Heroic epic Cilappatikāram, in which it is
used in the sense of talented painters or filigree workers.90 Naccinarkkiniyar
himself in his commentary takes the word to mean dancers.91 In view of the
unsatisfactory nature of the etymologies, it is perhaps better to reserve
judgement on the meaning of this word.
It is reasonable to assume that there were at a very early date subtle
differences between these apparent synonyms. They are synonymous to the
extent that they all denote dancers. But from stray comments here and there
in the works of medieval commentators we are made to understand that
they meant different types of dancers. For example, Aṭiyārkku Nallār in a
gloss in his great commentary on Cilappatikāram says that the kaṇṇuḷar
performed the Cāntikkūttu—a type of primitive ritual dance.
Naccinarkkiniyar says that there were many types of dance, like animal
dance, forest dance, pole dance, and others. But what is of interest to us is
the fact that they were not merely dancers but also singers. On reaching the
centre of a village or settlement, the ‘buglers’ sang along the streets flanked
on both sides by houses. Later they also danced.92 Commenting on the word
‘buglers’ the annotator of Paripāṭal says that it connotes singers and dancers
endowed with a knowledge of faultless music.93 From about the eighth
century CE, when lexicons like Tivākaram came to be compiled, these words
along with their Sanskrit equivalents in literary usage were grouped together
as synonyms. Under the heading ‘dancer’ Piṅkalantai has the following:
‘Nāṭakar, Nirutar, Naṭar, Kaṇṇuḷar, Kōṭiyar, Vayiriyar, and kūttar'.
THE AKAVUNAR

This class of bards is not mentioned by Tolkāppiyar in the list of wandering

minstrels who are directed by their fellow professionals to potential patrons.
Nevertheless, in the poems they are an important class of bards. The word
akavunar, akavalar, or akavar is derived from the verb akavu, ‘to utter a
sound as a peacock, sing, call, summon’, etc.94 Hence, it literally means
‘callers or summoners’. While discussing the mantic nature of the earliest
Tamil poetry it has been pointed out that the phrase akavaṉ makaḷ meant
‘woman diviner-cum-poetess’. While it throws light on the prehistory of
akavalar, it must be pointed out that in the heroic poems and choral lyrics
themselves the word clearly means a type of bard. Compared to the pāṇar,
porunar, and others, akavunar are mentioned less frequently. This and a few
other factors suggest some difference in the nature of their respective arts. In
a gloss on the word ahavar, Naccinarkkiniyar says that they are so called
because they recall the memory of the forefathers of the family of the person
they are praising.95 In the characteristic manner of his comments, he equates
and identifies akavar with sūtas and māgadhas, professional court minstrels
mentioned in the Sanskrit epics, and Purāṇas. While there inevitably seem
to be a few common features between the akavar and the other two, their
complete identification is open to question. In Sanskrit texts, there arises the
problem of the caste position of these minstrels.96 It does not apply to the
early Tamil minstrels. Being only too familiar with the Sanskrit scholarship
of medieval times, Naccinarkkiniyar brings Sūtas, māgadhas, vandins, and
vaitāligas into his discussion—names and concepts alien to early Tamil
poetry. Notwithstanding this confusion, his comment—in relating the word
to the verb akavu—is on the whole sound. One important, and apparently a
basic, difference between akavunar and other classes of minstrels in the
Tamil country is the fact that the singing of the former seems to have been
less wedded to music and dance. Not that it was entirely divorced from
music, but its rhythm was perhaps more simple. The words assumed greater
importance. Another salient factor is the great importance given in later
times to the ōcai, ‘rhythm’, peculiar to the akaval metre. So much so that the
name of the ‘rhythm’ itself—akaval—came to stand for the metre. Such
being the case, we are not surprised to find these bards not being associated
with any particular musical instrument or form of dance as are the others.
They were, we may suspect, fast becoming declaimers of poetry. They
‘doubtless recited “histrionically” with rhythmic accompaniment’. Dr Marr is
correct when he points out that ‘such a recitation is suggested by the ōcai-
term akaval that is applied to it’.97
In the light of the foregoing observation it is interesting to note that some
poems describe these bards carrying a wand or a staff instead of musical
instruments. The Tamil word is kōl, ‘rod, stick, branch, staff ’, etc.98 Holding a
well-cut slender rod, the bards praised the victors on the battlefields.99 In
another poem the rod is said to be ‘a slender bamboo with joints, selected
and well cut, from the thicket where tall bamboos grow’.100 The epithet
nuṇkōl, ‘slender rod’, is prefixed to the noun akavunar, meaning ‘bards
possessing slender rod’.101 Akavaṉ makaḷir, ‘women bards’, also carried the
rods. It is variously described as ‘slender wand with tender shoot’ and
‘slender wand with white shoot’. The reference is to bamboo shoots.
That the slender rod carried by these bards had definite significance is
pointed out by the later commentators. One says, ‘the rod makes known the
circumstance of birth’.102 Practically the same thing is said by another.103 In
the medieval context of these comments what is meant by ‘circumstance of
birth’ is the caste rank or position of the person concerned. But such an idea
is not applicable to the early period. What is more plausible and perhaps
probable is that the staff indicated the bardic profession. It has been
suggested that in the Greek tradition the wand or staff was later than the
harp or lyre and was actually a substitute.104 Whatever may have been the
precise significance of the stick, one thing seems certain. It was a special
symbol and property of this category of bards. But such a distinction does
not appear in the works either of medieval commentators or modern
writers. No doubt the akavalar had certain characteristics in common with
the minstrels, war-bards, and dancers. But the fact that the latter groups are
associated with one musical instrument or another, while the former are
identified by their staff, is a crucial distinction to be recognised. It is
proposed to argue shortly that, freed to a greater extent from intricate music
and dance, the wand-carrying bards developed the art of minstrelsy on a
greater scale and to a greater depth, where words and techniques of
narration became more important.
THE PULAVAR

Like the word akavunar, the word pulavar is used less frequently in the
poems. But of all the synonyms that were current in the early period, this
alone survived through the centuries right into modern times. The standard
word for a poet in modern Tamil is pulavaṉ—just as it was in the early
poems. The word is derived from pulam, ‘sense, sensation, knowledge,
wisdom’, etc.105 Pulavar means ‘wise men, the learned’. Pulavar is another
word that indicates the fact that the ideas of wisdom, knowledge, and
learning were considered inherent in the person of the poet. Both early
poetry and medieval commentaries treat the word as signifying wise men,
implying poets [emphasis added]. This dual meaning of the word has
survived to this day.
That the ancients used the word with absolute reverence is proved by its
application to God Murukaṉ, who is described as ‘the wise one knowing the
lore’, ‘lion among the wise ones’, and ‘the incomparable wise one’.106 Such
descriptions are not fortuitous. It has already been stated that the wise ones
and diviners were believed to receive their powers by propitiating Murukaṉ.
Hence it is but logical that he should be described as the fountain of all
learning and mantic wisdom. It is also of some significance that through all
the ages, Tamil literary and religious traditions have enshrined the belief
that he was the ‘prince of poets’. The commentary on Iraiyanār Akapporul
places him among the presiding bards in the First Academy (Caṅkam or
Sangam). The late medieval poet Kumara Kuruparar (c. 17 CE) echoes the
same belief when he sings, ‘O Prince Bard of Caṅkam Literature’. It has
already been observed that in earlier times the Tirumurukārruppatai was
also called Pulavarāṛṛuppatai, which means that a devotee of Murukaṉ
advises another on the ways and means of obtaining the god’s blessings.
Owing to this the poem has become the liturgy of the devotees of Murukaṉ.
All these evidences unavoidably point to one fact; poets have always been
associated with profound learning and wisdom, which was originally largely
religious in character, and connected with the cult of Murukaṉ.
 A perusal of the poems and the contexts where reference is made to the
poets shows us the high esteem in which they were held in society [emphasis
added]. This is in contrast to the treatment of other types of minstrels who
give the impression of being patronised as mere entertainers and
encomiasts. The respective words speak for themselves. Pāṇar, ‘singers’,
porunar, ‘war-bards or impersonators’, kūttar, ‘dancers’, viṟaliyar, ‘female
dancers’. It need not be emphasised any more that by being termed ‘wise
men’ pulavar automatically became a distinct category of poets. These poets
did indulge in the art of eulogy which was characteristic of their age. Yet
their position and respectability must have spared them the professional
need to live by flattering patrons. Unlike the professional minstrels, pulavar
are not associated with music or dance. They also seem to be distinct from
the akavunar, by the fact that they do not carry a staff or wand. In other
words they had no specific and distinguishing symbol or property. Their
hallmark was wisdom. It is to be presumed that this gave them greater
freedom over their vocation.
In a poem of ‘character’ referred to earlier, in another context, it was
pointed out that King Neṭuñceḷiyaṉ cherished the praises of poets. In the
course of his oath the king says,
If I break this oath, let the poets honoured by the whole world, chief
among whom is the well-informed and wise Marutaṉ of Māṅkuṭi, quit
me and my kingdom as unworthy of their songs.107
It is evident that they were highly respected for their wisdom and
judgement of men and matters, which was also of considerable use and
value to the monarchs. The activities of Kapilar substantiate this. The poem
quoted above has often been cited as evidence for the existence of schools of
poets who were attached to one master poet.108 This implies a form of
discipleship on the part of junior poets, which is in accordance with the
nature of bardic training. It has also been argued that such schools of poets
eventually developed and merged into the Literary Academy modelled
largely on the Buddhist and Jaina Saṅghas.109 Such developments are
probable. But what may be pointed out here is that even such relatively
‘sophisticated’ schools of poets are analogous to the other types of minstrels
and bards who learnt their art from older singers and often functioned as
groups.
These and other evidence related to them throw some light on the
organisation and function of poets. For whatever characteristics they had in
common with the other types of minstrels their preoccupation with learning
and perhaps the exclusive knowledge of certain treatises and lores set them
apart as sophisticated men. This, it may be said, gradually enabled them to
become the acknowledged masters of all studies. For example, in the
Tolkāppiyam, the stock phrase invoking the authority for the rules described
and prescribed in it is ‘thus say the wise ones’. Other noun-epithets of similar
meaning in the same work clinch this point: ‘treatise-learned poets’,
‘eminently-learned poets’, ‘poets knowing the ancient sayings’. Though they
are stock phrases, the meaning is quite clear: the poets had become eminent
teachers. Perhaps because of the aura attached to them, their words drew
greater attention than those of other types of minstrels. While the encomia
of the others were considered pleasantly laudatory, those of poets were
considered learned; and, if we may use the word, more ‘spiritual’. A poem in
praise of the Cōḻa prince, Nalaṅkiḷḷi, illustrates this.
I have heard it being told that, those whose fame is praised by the wise
ones, shall at the end of their lives, have the privilege of travelling in the
self-driven vehicle in the sky.110
CLASSIFICATION OF MINSTRELS

The minstrels’ and bards’ description of the varieties enumerated in the

preceding section gives us a fairly adequate picture of their life and work. On
the basis of this data it is now proposed to classify them functionally. It was
mentioned earlier that the medieval commentators never attempted to draw
any distinction between these various minstrels except for the broad
division of singers and dancers. At all events this division became a
convention faithfully followed by all later writers. Analysing the data
presented earlier, it is possible to divide the minstrels into three main
categories.

1. singing minstrels: non-choral and choral

2. bards who declaimed poetry
3. wise men and scholars who were poets

Such a classification is also in keeping with the general development of

poetry in other ancient societies such as the Greek, Irish, etc. A four-stage
evolution—represented respectively by (1) magico-ritualistic singers, (2)
bards, (3) poets, and (4) philosophers and religious teachers—has been
postulated.111 The bulk of the extant poems, as elsewhere in heroic poetry, is
anthropocentric—in praise of the deeds of famous men. Nevertheless, as has
been pointed out, one can glimpse the vestiges of the earlier stage. The last
stage, of course, coincides with the age of the Tolkāppiyam and leads up to
the twin epics, a period dominated by religious teachers, Brahmanic,
Buddhist, and Jaina alike.

Minstrels—Non-choral: pāṇar, Porunar, Pāṭini, etc.

These singers either wandered over the country in search of patrons and
audience or were lucky and received the patronage of some benevolent
chieftain. They are identifiable by the primary musical instrument they used
in their performances. Among them the war-bards may be termed ‘mimetic
artists’, specialising in themes of war and heroism. A good portion of the
heroic songs extant may have originated with them. It is probable that they
composed and recited songs and had a rich and varied repertoire. But they
were largely dependent on patronage for their livelihood. Auvaiyār’s
somewhat bitter words portray the sad dependent life of such singers: ‘This
minstrel’s life, that pines for honour.’112 The very same words are echoed in
another poem by a different bard.113 The scenes of chronic poverty so well
detailed in the ‘guidance’ poems leave one in no doubt as to the hunger and
misery which were the lot of many of these singers. Though kings and
princes entertained them to lavish feasts and showered immense largess on
them, it was in return for the delightful songs which the minstrels provided.
Their talent was apparently limited to their art, and that was all they had.
They were excellent entertainers. Their clan organisation gave them some
advantage in this respect, since their art was hereditary and cultivated
almost exclusively. In tribal society the chieftains had joined in the dancing
and singing of minstrels. Tolkāppiyar speaks of two types of tērk kuravai,
‘dance of the chariot’, in which a victorious chief joins the warriors and
minstrels in a frenzied dance either in front of the chariot or behind it. But
as the aristocracy became more advanced and stabilised, the minstrels
entertained while the kings and princes patronisingly watched them. A
poem in praise of a Cēral king says that although he is not capable of
dancing to the beat of the drum of the dancers, he would dance in joy over
his victory on the battlefield.114 These singers travelled about in groups and
their art was largely impersonal. We have already referred to Tamil minstrels
beginning the praise of kings with due regard to the praise of gods.115 The
Tolkāppiyam bears further evidence for this practice. In the section ‘on
poetry of praise’ is included the praise of both men and gods. Kaṭavuḷ vāḻttu
and tēvarp parāval are two of the phrases meaning ‘the praise of god’. There
is also reference to competition among poets: ‘as the king defeats his rivals,
so does a bard encounter and defeat his rival singer.’116 The winner was
awarded prizes.

Minstrels—Choral Singers: kuttar, Viṛliyar, etc.

It has been noted above in connection with the description of the pattern of
life of choral singers and dancers that it did not vary very much from that of
non-choral singers. Dancing was their speciality and apart from heroic
themes they also dealt with love themes.117 This class of minstrels perhaps
contributed much to the development of early Tamil drama, though our
knowledge of this is very limited. These choral dancers performed both
indoors and in public places in the open.

Declaimers and Narrators: Akavunar

A detailed description of the akavunar has already been given. It was
pointed out that in contradistinction to the choral and non-choral singers,
these bards carried a wand or a staff instead of musical instruments. They
appear to have been the specialists in the recitation of songs in the Akaval
metre.

Poets and Wise Men: Pulavar

The pulavar, it was pointed out, were a class of poets who actually became
more important in the post-Heroic era, though their origins hark back to
earlier times. Some of these learned poets were of a definite mantic
character. With progressive intermixture of Sanskrit or Vedic and Dravidian
cultures, brahmanas came into this category. Stories and legends about a
number of Heroic and post-Heroic poets illustrate this trend. The legend
concerning Akattiyar, who is supposed to have been a brahmana sage and
the father of Tamil, is a case in point.118 In the post-Heroic age, the learned
poets with their mantic background and associations had a new role to play.
The Chadwicks have juxtaposed the Heroic and post-Heroic poetic milieu
very succinctly. Heroic poetry ‘represents the intellectual life—primarily the
entertainment—of court circles in the days of military kingship. It is the
product of a society whose governing principles are prowess in arms, loyalty,
generosity—in short, action, not thought.’ This spirit and life was conducive
to the professional minstrels, who thrived in it. But the ascendancy of
learned poets, not necessarily professionals, ‘represents the activities of the
learned in barbaric times and is intended for instruction. It arises in a milieu
where the governing principle is thought, accompanied by inspiration or
second sight, or at least by the cultivation of mantic lore.’119
In early Tamil the poets represent this second stage. It was not their
capacity to play on the lute or the tambour or beat the drum or dance that
found them patronage and respect. They were not professionals in any art or
craft. This presumably gave them a certain amount of independence, which
in the case of a few manifested itself in the form of arrogance. As an instance
of this we may cite the example of a poet who dared to snub princes.120 He is
said to have reproached a well-known chieftain: ‘I am not a poet singing for
money’ or ‘I am not a poet who barters his art.’121 These words, while on the
one hand portraying the poet’s confidence, imply on the other the existence
of a number of ‘mercenary’ singers.
This is perhaps the place to deal with the probable connections that
existed between the bards and the poets. From what has been discussed so
far it may be held that the power and position of the poets were posterior to
that of the bards. But once the position of the ‘wise ones’ was well
established, they seem to have taken over the functions and encroached on
the domain of professional minstrels and bards. The clue to this
development comes from the Tolkāppiyam. In the chapter on prosody or
composition, the akaval metre is singled out for special elucidation. ‘The
name Akaval is but another name for Āciriyam.122 This is very revealing. No
one doubts the fact that akaval is one of the older metres known in Tamil, if
not the oldest, and certainly the metre in which the entire bardic poetry of
the Heroic Age is sung.123 It has also been pointed out that the bards were
specialists in the handling of this metre. Yet Tolkāppiyar treats it as an
obsolescent word and gives an explanation of its meaning almost in the form
of a gloss. This clearly shows that at his time the term akaval had already
been supplanted by āciriyam, meaning ‘that of the teachers’. Āciriyar, the
Tamilicised form of the Sanskrit ācārya, means ‘teacher, preceptor’, etc. In
fact, from then on the metre came to be known as ‘the master’s metre’,
because of its use in grammars and didactic works. In that sense it may be
called an ‘aphoristic’ metre.124 Thus we see the original meaning of ‘call,
summon, and sing’ changed into ‘explain, teach, and sermonise’. In other
words, the milieu is different, showing many characteristics of changed
conditions and conventions. This shift of and change in the meaning of the
word itself exemplified the relative vicissitudes of the fortunes of these two
classes of bards. It must of course be pointed out unambiguously that this
change must have been gradual, and must be seen against the background of
a long and complex literary tradition. The ascendancy of the poets coincided
and overlapped with the influx and spread of Jaina and Buddhist religious
philosophies in the Tamil country. This again was contemporaneous with
the rise of a powerful merchant class.125 And it is not surprising that during
this period of the predominance of the poets we get the gnomic and didactic
works in Tamil.
In fact, no better testimony to the great role played by these poets and
teachers exists than the work of collecting, codifying, redacting, and
harmonising the poetry of the Heroic Age which they accomplished. In
doing so they have no doubt left the imprint of their own attitudes and age
on the work: the fundamental division of all early poetry into the two
categories of ‘love’ (akaṃ) and ‘war’ (puraṃ), which were again minutely
subdivided into numerous situations of human conduct or behaviour. The
guiding principle behind such neat—though at times artificial and idealistic
—schematisation was, of course, oḻukkam, ‘acting according to established
rules, good conduct, decorum propriety’,126 with which they were
completely concerned. It is epitomised in an aphorism of Tiru Vaḷḷuvar.
Correct conduct ennobles one; so it must be held more dear than life
itself.127
The Tolkāppiyam, which successfully effected the synthesis of Sanskrit and
Tamil literary traditions, is perhaps the finest example of harmonising the
old and the new. It has been recognised that a moral and philosophic
attitude runs through the whole of the Tolkāppiyam.128 For the first time in
Tamil it sets forth an organic view of literature.129 This concept of the
organic unity of a poem or a literary composition would itself appear to be
the result of the application to literature of the philosophical speculations of
the Jaina teachers, whose influence is unmistakably evident in the
Tolkāppiyam, For example, the Jaina classification of lives, jῑva, and non-
lives, ajῑva, is treated in the chapter on tradition and literary usage.130
From what has been discussed so far it may be held that, though medieval
scholiasts and modern writers failed to classify the minstrels and bards seen
in the earliest Tamil literature, they could be conveniently divided into three
major categories or types. It was mentioned at the outset of this chapter that
heroic poetry was traditional and sustained by tradition. We may now
discuss some aspects of the process of the transmission of this material. The
literary climate of the age and the general condition of the minstrels and
bards may be delineated here.
It has been sufficiently demonstrated that the learning of minstrels was
oral—by imitation and practice. The basis of this knowledge was inherently
auditory. As much as the mouth is the medium of expression, the ear is the
medium of comprehension. One of the words denoting learning is kēḷvi,
from kēḷ, ‘to hear, listen, ask, learn’, etc.131 This clearly indicates the oral
nature of learning. The significance of this becomes more obvious when we
see it often used for the learning of poets. As has already been noted, the
poet Māṅkuti Marutaṉ was held in high esteem by Neṭu ceḻiyaṉ. The king
refers to him as ‘the highly eminent and well-informed Māṅkuṭi Marutaṉ’. In
this instance the poet’s learning is denoted by the word kēḷvi. Elsewhere,
Kapilar is described thus: ‘Kapilan of immense learning and renowned
fame’.132
In course of time the word’ came to be identified with the Sanskrit word
śruti, ‘anything heard, oral account; the Veda’.133 Accelerated by the spread of
Brahmanic ideas in the Tamil country, it soon acquired the specific meaning
of ‘the Veda’. It is found used in this sense in some of the early poems. For
example, the noun-epithet, kēḷvi antaṇar, ‘brahmins versed in the Vedas’ is
found in more than one place.134 The same idea is also expressed in the
Purapporul Veṇpāmālai, which says that the brahmana is full of learning.135
From these evidences one fact emerges clearly: the brahmins popularised
the Vedic sacrifice among the kings. And it is in the context of the Vedas and
Vedic rituals that their scholarship is mentioned.136
Besides poetry, canonical and exegetical material was also studied and
cultivated orally. The commentary on Iraiyanār Akapporul records that it
was handed down orally from Nakkirar to Nilakantar, a period spanning
several generations. In connection with this work and some others, the oral
nature of the early literature has already been noticed previously by some
modern writers.137
GENEALOGIES

An important aspect of traditional knowledge deemed necessary for

ministrels and bards dealing especially with the heroic lore was royal
genealogies. While the panegyric verse composed by the court poets gave a
picture of the contemporary world, the names and exploits of the ancestors
of the persons concerned recalled the past. Thus both the panegyric and
genealogy were interrelated and together formed composite ‘traditional’
material. In due course these genealogists naturally developed into
‘historiographers’. In several poems, ancestors of kings not known to us
through any other poems are mentioned. Doubtless some names are vague
and not easy to determine. Such, for example, is the reference to some
bygone ancestors of the family of Atikamāṉ of Takaṭūr: Auvaiyār sings of her
patron’s glorious ancestors who first planted sugarcane in the Koṅku region;
perhaps an oblique allusion to the introduction of agriculture.138 This story,
sustained by a strong tradition that persisted all through the ages—
inevitably with embellishment—has been recorded by Logan in his Malabar
Manual.139 Nevertheless, many of the genealogical references and pieces of
information seem to be consistent and meaningful. This would imply the
existence of oral genealogical material used by the bards. Since most of the
allusions were expected to be familiar to the audience, explanatory details
were kept to the minimum. For example, the Pāṇṭiyaṉ king, Neṭuñceḻiyaṉ, is
praised as the worthy descendant of one Neṭiyōn, of whom nothing special
is mentioned.140 Yet, by a reference to the same person in a poem on
another well-known Pāṇṭiyaṉ king we understand that he had initiated a
festival to the god of the sea.141 That some form of genealogical lore was
‘accurately’ maintained is suggested in a poem by Kapilar. He speaks of
Pulikaṭimāl, a Vēḷir prince, as the scion of a family that has ruled for forty-
nine generations.142 The Patiṟṟuppattu affords a good illustration of the way
in which bards used the genealogical material. The patikams, strictly
exegetical material, which gives the parentage of the hero celebrated in that
decade, some aspects of his life, the number of years he ruled, and the name
of the poet, shows signs of having drawn on traditional information.
Scholars have long argued about the authenticity of these ‘epilogues’
pointing out discrepancies between them and the decade poems. Some tend
to reject the ‘epilogues’ as later fabrications. But the matter does not end
there. For if one admits that the epilogues are spurious, one is obliged to
reject the whole chronology of the early Cērals: without the facts mentioned
in them, the decade poems themselves would become devoid of any
historical sense and sequence. The dilemma is unavoidable and is
encountered by those who reject the poems solely because of their
lateness.143 The whole problem seems to arise because of the fallacy that late
material is necessarily unauthentic, in other words, because of the
contention that the content of a work cannot be older than the form.
Avoiding these two rigid, if not arid, alternatives, it may be argued that
such material as is dealt with in the epilogues was oral and traditional: in
this case they would assume greater significance. Marr is right when he says
‘the writers of the patikams may have drawn on material not available to the
poets who composed the poems or not used by them. This would account
for the discrepancies between the patikams and the decade poems.’144 In
fact, besides the writer(s) of the ‘epilogues’, a later poet like the author of the
literary epic Cilappatikāram must have made use of similar material in
constructing the royal genealogies.
Analogous to the epilogues added to the Patirruppattu are the colophons
attached to the songs in the Puranānuru. These colophons mention the
personnel and occasion of each poem. Though divided on the probable date
of these colophons, scholars are generally agreed that they are later than the
poems themselves. These colophons, which alone furnish the vital and
invaluable information for any ‘historical’ reconstruction, were probably
based on traditional material. Modern writers have demonstrated them to
be exceptionally accurate and trustworthy. ‘Instances fully borne out by the
texts of poems in different Anthologies are’, says Professor Nilakanta Sastri,
‘sufficient proof that the colophons embody genuine history.’145
The singular felicity and confidence with which bards sang of their
patrons and their ancestors remind us once again of the observation made
by Naccinarkkiniyar: the akavar were bards who recalled the memory of the
ancestors of kings. In passing, mention may also be made of the account of
the Sangams given in the commentary on Iraiyanār Akapporul. The
impressive array of sages and chief poets who are alleged to have
participated in the proceedings of those Sangams might have owed its
origins to some form of genealogical lore dealing with the achievements of
Pāṇṭiyaṉ kings. The fact that gods and sages are included in the list of poets
need not deter us from evaluating it seriously, as it has some modern
writers.146 Even genuine traditions have to have a beginning, and, as a rule,
beyond a number of generations the past becomes mythical.147 As Finley
has stated with reference to the early Greek historiographers, ‘the
atmosphere in which the Fathers of History set to work was saturated with
myth. Without myth, indeed, they could never have begun their work.’148 In
a few instances, fact and fiction are certainly intertwined. We may cite an
example. The Perumpānārrupapaçai mentions the hero’s ancestors. He is the
descendant of the Toṇṭaiyōr.149 Toṇṭai, a common hedge-creeper,150 was the
emblem of the clan.151 It probably had totemistic origins, as had many of the
emblems of tribes and clans of the early Tamil country. Now, the prince is
said to be the scion of the Toṇṭaiyōr, who conquered the land by their sword
and lived on plunder.152 Going still further back, their ancestors are said to
descend from the ‘god who spanned the earth, and whose breast bears the
sacred mark and who has a sea-dark skin’.153 The reference is unmistakable.
The origin is traced to God Vṣṇnu or Māyōṉ, ‘the dark one’ as he is referred
to in early Tamil literature. The Toṇṭaiyōr have had a distinguished career in
the history of the Tamil country. Toṇṭai Nāṭu or Toṇṭai Maṉtalam has
remained an important province in the Tamil land.154 But we notice their
beginnings merging into myths.
Catalogues
Connected with genealogical lore are catalogues. In one sense genealogies
are also catalogues. But other things remembered in the form of catalogues
form an important and integral part of antiquarian learning. It is held that
catalogues are among the earliest informative elements in heroic stories.155
In post-bardic literature catalogues are more prominent and frequent, and
seem to be ideally suited, to gnomic literature, where human conduct, belief,
and needs are neatly listed and are either prescribed or proscribed. The bulk
of the ‘Eighteen Minor Books’ falls into this category.156 But bardic poetry,
too, has elements of catalogue material in it, which the minstrels and poets
had to know as part of their sources. These catalogues are generally short
and comprise lists of things necessary for the smooth running of the poems
and lays.
From the earliest times Tamil literary tradition has enshrined the memory
of seven chieftains renowned for their liberality. The expression ‘the Seven
Munificent Ones’, or ‘the Seven Great Donors’, has remained the proverbial
standard of generosity. In spite of the overwhelming power and fame
attained by the three Crowned Kings—Cērals, Cōḻas, and Pāṇṭiyas—these
seven chieftains, some of whom were probably no more than petty
chieftains, have held their own in popular imagination. These seven
chieftains became legendary and seem to have been remembered
collectively. Their memory appears to have been transmitted in the form of a
catalogue. Peruñcittiraṉār lists them in a poem. Pāri, Ōri, Kāri, Eḷiṉi, Pēkaṉ,
Āy, and Naḻḻi are the seven names enumerated with befitting epithets.157 The
same list is repeated, with slight changes in the attributes; in one of the lays
in the Ten Songs.158 It is also found in a catalogue form in an anonymous
poem quoted by Per.159
Perhaps the longest catalogue in our sources is to be found in the
Kurincippāttu. In the midst of a description of a girl and her foster-sister
enjoying themselves on the banks of a stream, the song bursts into a
methodical catalogue of the better-known flowers of the hilly region.
Ninety-nine flowers are enumerated—some of them with descriptive
epithets.160 The presence of a catalogue has discomforted many modern
critics. The English translator of the Ten Songs says ‘this list seems an
intrusion, and somewhat detracts from the high poetic level of the poem’.161
Bardic training included information pertaining to flora, fauna, etc.
Accordingly, the presence of the catalogue need cause no surprise. There are
reasons to think that, whoever composed the poem, it was meant to be an
exercise in singing the kuṟiñci theme. Another name of this poem is
Perunkuṟiñci, ‘the large kuṟiñci, which, in contrast to the short poems on the
same theme, suggests a fuller treatment. It gives an impression of a model
poem, illustrative of a type. It is an informative poem on the kuṟiñci
region.162 It was intended, to illustrate most of the rules regarding that class
of poems.163 A later work—Maṇimēkalai—sung in the same metre furnishes
an example of a catalogue of flowers,164 echoing the one in Kurincippāttu.
The shorter catalogues bearing on heroic themes are of kings and
chieftains in battle. For example, Naṉṉaṉ, Eṟṟai, Atti, Kaṅkaṉ Kaṭṭi, and
Puṉṟurai were six tribal chiefs mentioned in connection with a famous battle
in which Paḻaiyan—the renowned commander of the Cōḻa armies—was
killed.165 Another catalogue which appears to have been popular describes
the alliance of kings and chieftains who were defeated by Neṭuñceḻiyaṉ at the
famous battle of Ālaṅkāṉam. The seven leaders, Cēral, Cempiyaṉ, Titiyaṉ,
Eḻiṉi, Erumaiyūraṉ, Iruṅkōveḷ, and Porunaṉ, are listed with appropriate
epithets.166 Elsewhere, other statements like, ‘the fourteen united Vēḷir,167
‘the shameless nine’,168 ‘the two kings and the Velir chieftains’169 suggest that
catalogues are implied.170
A slightly different sort of catalogue is found in the Tirumurukārruppagai.
In this list is given a few of the god’s names—attributive and descriptive. The
fact that many of the twenty-three names mentioned in the list are to be
found in other poems of the period indicates the prevalence of catalogues,
perhaps even pre-Heroic. These names, running in steady flow, suggest that
they were meant for chanting.171 In passing, it may be pointed out that the
Tolkāppiyam employs the cataloguing method freely. Being a book of
instruction for bards, it lists various types of information necessary for their
compositions.
All the factors enumerated and discussed above have conclusively shown
the traditional nature and almost unconscious manner of the learning of the
bards, of the Tamil Heroic Age. The art of minstrelsy—singing, miming, or
declamation—was learnt orally from the elders, and all the ingredients in
the learning—formulae, themes, genealogies, catalogues, etc.—were
mnemonic compendia of information, equally shared by the audience and
the singers. Under these circumstances it is not surprising to find ‘education’
absent in that age. A modern writer makes the following comment. ‘There is
hardly any information in the Pattuppāṭṭu regarding the exact nature of the
system of education, its principles and methods, etc.’172 Education, as we
understand it in a literate society, was, perhaps, not found in the Tamil
Heroic Age. And yet, according to the same writer, the extant poems ‘give
ample evidence of a very well-developed literary tradition’.
This literary tradition was first treated organically and rationally at the
end of the Heroic Age. Under the influence of Buddhist and Jaina
missionary activities,173 schools174 of learning arose in many parts of the
country. These might have incorporated whatever systems there were for
literary and religious instructions. They attracted students even from distant
places. This is borne out by a reference in the Tolkāppiyam. Illustrating the
circumstances under which a man might be away from his spouse, it names
education, war, diplomacy, and search of wealth. These took a man away
from his home to distant places. Philosophy, logic, and literature were the
main subjects taught in these schools. Metrical studies formed an important
and integral part of literary instruction. Such a world is vividly shown in the
Tolkāppiyam, which sums up the experience of the previous age and its
wisdom. In so doing, it became the most traditional of books—at times even
overshadowing the very works it intended to elucidate—in which were
resolved all the varied strains of the past, and without the aid of which the
past itself became less meaningful. This all-pervading influence of the
Tolkāppiyam175 and the redaction of the heroic and bardic love poems also
determined and secured subsequent countinuity of these works, or at least
that of a fair proporation of them: the poems became not merely the
remnants of a bygone age, but very much part of the new, albeit different,
type of education. Like Homer, who formed the staple of Greek education in
classical Greece,176 the redacted Tamil bardic corpus became Mēl Kaṇakku,
‘The Major Books’, according to the pattern of which the later Kῑḻk Kaṇakku,
‘The Minor Books’, seem to have been arranged.
13

Who Needs Folklore?

The Relevance of Oral Traditions to South Asian
Studies

A. K. RAMANUJAN

I NIndian
THIS ESSAY, MY THEME IS NOT FOLKLORE IN GENERAL, but
folklore within the context of Indian studies, using Kannada and
Tamil examples from my field notes. I wish to do several things: (1) give a
state-of-the-art report on the field of Indian folklore; (2) clarify some
notions and add some, and (3) generally ask and answer questions about
what the study of folklore, as a subject matter and as a discipline, would do
to some of the notions of humanists and social scientists about Indian
civilisation.
When some years ago I first approached this subject—the place of folklore
in the study of Indian civilisation—I heard a little sceptical voice from my
past say ‘Folklore? Who needs folklore? Old wives’ tales and peasant
superstitions, who needs them?’ As you know, the past never quite passes.
We may hear that voice again. Here, I am going to take that question literally
and answer it.

WHY FOLKLORE?

For starters, I for one need folklore as an Indian studying India. It pervades
my childhood, my family, my community. It is the symbolic language of the
non-literate parts of me and my culture. Even in a large modern city like
Bombay or Madras, even in Western-style nuclear families with their 2.2
children, folklore is only a suburb away, a cousin or a grandmother away.
One of the best folk plays I have seen was performed in the back streets of
Madras city by terukkuttu troupes. When a friend of mine in Bangalore, the
capital city of Karnataka state, said to me, ‘How can you collect folklore in a
big city?’, I asked him to try an experiment. He was a professor of Kannada,
and he had a composition class that afternoon at his college. I asked him to
set a composition exercise to his class of urban students. Each of them
should write down a folktale they had heard and never read. That evening,
my friend sought me out excitedly to show me a sheaf of forty tales his
students had written down for him in class from memory.
I shall not speak here of Indian urban folklore, for wherever people live
folklore grows—new jokes, proverbs (like the new campus proverb, ‘To xerox
is to know’), tales and songs circulate in the oral tradition [emphasis added].
Similar to chain-letters, Murphy’s Law, and graffiti, folklore may also
circulate on paper or on latrine walls.1 You don’t have to go to Pompeii to
see graffiti. Verbal folklore, in the sense of a largely oral tradition with
specific genres (such as proverb, riddle, lullaby, tale, ballad, prose narrative,
verse, or a mixture of both, and so on), non-verbal materials (such as
dances, games, floor or wall designs, objects of all sorts from toys to outdoor
giant clay-horses), and composite performing arts (which may include
several of the former as in street magic and theater)—all weave in and out of
every aspect of living city, village and small town. What we separate as art,
economics and religion is moulded and expressed here. Aesthetics, ethos
and worldview are shaped in childhood and throughout one’s early life by
these verbal and non-verbal environments. In a largely non-literate culture,
everyone—poor, rich, high-caste and low-caste, professor, pundit or
ignoramus—has inside him or her a large non-literate subcontinent.
In a south Indian folktale, also told elsewhere, one dark night an old
woman was searching intently for something in the street. A passerby asked
her,
‘Have you lost something?’
She said, ‘Yes, I’ve lost some keys. I’ve been looking for them all
evening.’
‘Where did you lose them?’
 ‘I don’t know. Maybe inside the house.’
‘Then, why are you looking for them here?’
‘Because it’s dark in there. I don’t have oil in my lamps. I can see much
better here under the street lights,’ she said.
Until recently many studies of Indian civilisation have been done on that
principle: look for it under the light, in Sanskrit, in literary texts, in what we
think are the well-lit public spaces of the culture, in things we already know.
There we have, of course, found precious things. Without carrying the
parable too far one may say we are now moving inward, trying to bring
lamps into the dark rooms of the house to look for our keys. As often
happens, we may not find the keys and may have to make new ones, but we
will find all sorts of things we never knew we had lost, or ever even had.
INDIAN REGIONAL LANGUAGES

Four centuries ago, just a century after Vasco da Gama landed on the west
coast of India, just decades after Gutenberg had printed his first Bible in
Europe, Christian evangelists had begun to study our mother tongues,
compile dictionaries, make grammars and even print them in India. Yet,
until recently, Sanskrit almost exclusively represented India to most people
in the West.
In America, it was only about twenty-five years ago* that universities
began to study Indian regional languages. At least three or four major
languages, such as Tamil, Hindi and Bengali, began to appear in course
listings. Both linguists and anthropologists went to these language- regions,
studied the languages in the field and wrote about the texts and the cultures.
These languages are only a minute fraction of those spoken in the
subcontinent. In the 1971 census more than 3,000 mother tongues were
recorded with the names of the speech varieties that the speakers said they
spoke. Linguists have classified and subsumed these speech varieties, or
dialects; under 105 or so languages which belong to four language families.
Of these 105 languages, ninety are spoken by less than 5 per cent of the
entire population; sixty-five belong to small tribes. Including Sanskrit,
fifteen of the languages are written, read and spoken by about 95 per cent of
the people. We, in universities outside India, have just begun to study a few
of these fifteen languages.
The literatures of these fifteen, some of which have long histories, are just
beginning to be taught and translated. Literature in a language like Tamil
goes back 2,000 years, and in several others, like Bengali and Gujarati, at
least 800 years. In addition to these literatures there are oral traditions,
riddles, proverbs, songs, ballads, tales, epics and so on, in each of the 3,000-
odd mother tongues that we have classified under the 105 languages. It is
true, as they say, a language is a dialect that has acquired an army, but all
these myriad dialects carry oral literature, which is what I call folklore. One
way of defining verbal folklore for India is to say it is the literature of the
dialects, those mother tongues of the village, street, kitchen, tribal hut, and
wayside tea shop. This is the wide base of the Indian pyramid on which all
other Indian literatures rest [emphasis added].
We have valued and attended only to the top of the pyramid. Robert
Redfield, the Chicago anthropologist who influenced Indian anthropology
in the 1950s and 1960s, said, ‘In a civilization, there is a great tradition of the
reflective few and there is a little tradition of the largely unreflective many’.
That is a famous formulation that deserves to be infamous. Traditionally
Indians also make a distinction between mārga, ‘the high road’, and desi, ‘the
byway, the country road’, in their discussion of the arts. The ‘Great Tradition’,
with capitals and in the singular, said to be carried by Sanskrit, is pan-
Indian, prestigious, ancient, authorised by the cultivated and carried by what
Redfield calls ‘the reflective few’. The ‘Little Tradition’, or traditions in the
plural, are local, mostly oral, and carried by the illiterate (the liberal would
call them non-literate) and the anonymous ‘unreflective many’. Redfield
himself and Milton Singer later modified these notions and others have been
critical of them. They were seminal at one time, especially because they
urged anthropologists not to ignore the ‘texts’ of a culture in favour of
‘fieldwork’.
CULTURAL PERFORMANCES AS TEXTS

Now we need a new emphasis, a larger view regarding texts themselves, as

text-theory in literary criticism and philosophical analysis urge us to do.
Written and hallowed texts are not the only kinds of texts in a culture like
the Indian. Oral traditions of every kind produce texts. ‘Cultural
performances’2 of every sort, whether they are written or oral acts of
composition, whether they are plays, weddings, rituals or games, contain
texts. Every cultural performance not only creates and carries texts, it is a
text.
When we look at texts this way we can modify terms such as great and
little traditions and see all these performances as a transitive series, a ‘scale
of forms’ (a phrase in a different context, from Collingwood)3 responding to
one another, engaged in continuous and dynamic dialogic relations. Past
and present, what is ‘pan-Indian’ and what is local, what is shared and what
is unique in regions, communities, and individuals, the written and the oral
—all are engaged in a dialogic reworking and redefining of relevant others.
Texts then are also contexts and pretexts for other texts.4 In our studies now
we are beginning to recognise and place folk texts in this ever-present
network of intertextuality. For folk texts are pervasive, behind, under,
around all the texts of our society, and in all its strata, not merely among the
rural and the illiterate, the ‘unreflective many’. City and village, factory and
kitchen, Hindu, Buddhist and Jaina, Christian and Muslim, king, priest and
clown, the crumbling almanac and the runaway computer—all are
permeated by oral traditions, tales, jokes, beliefs and rules of thumb not yet
found in books. I shall say more later about the dialogic relation between
folklore and other parts of this Indian cultural continuum.

INTERACTIVE PAN-INDIAN SYSTEMS

In the view being developed here, even what’s called the Great Tradition is
not singular but plural—it is a set of interactive pan-Indian systems,
Brahmānism, Buddhism, Jainism, with tantra and bhakti interacting
variously with these. To be comprehensive we should add Islam,
Christianity, et cetera, and modernity itself as the other active systems that
participate in this give-and-take.5
Let’s examine briefly the idea that some traditions are pan-Indian and
some are not. Sanskrit and Prakrit, though they have a pan-Indian
distribution, still originate in particular regions; Sanskrit itself, though
trans-local and apparently a-geographic, has varieties of prounciation that
can be identified as Bengali, Malayali or Banarasi.6 Nor are the so-called
‘Little Traditions’, especially folk traditions, necessarily or usually confined
to small localities or dialectal communities. Proverbs, riddles, and stories,
and tunes, motifs, and genres of songs and dances are not confined to a
region, even though they may be embodied in the non-literate dialects and
may seem to be enclosed in those mythic entities called self-sufficient village
communities. It is well known that folklore items, like many other sorts of
items in cultural exchange, are autotelic, that is, they travel by themselves,
without any actual movement of populations [emphasis added]. A proverb, a
riddle, a joke, a story, a remedy, or a recipe travels every time it is told. It
crosses linguistic boundaries any time it is told. It crosses linguistic
boundaries any time a bilingual tells it or hears it.
Neighbouring languages and regions have, therefore, a large stock of
shared folk materials. Collections, for instance, have been made of the
proverbs shared by the four Dravidian languages. Similar ones can be made
for other genres and for other neighbouring language areas, and indeed for
the whole subcontinent. A proverb such as ‘It’s dark under the lamp’ (dipada
kelage kattale, in Kannada) has been collected in Kannada and in Kashmir,
at two ends of the Indian subcontinent. The sentence is the same in each
place, but it means different things. The reference is the same, but the sense
is different. In Kannada it means that a virtuous man, like a lighted lamp,
may have dark hidden vices. In Kashmiri, I am told, ‘It’s dark under the
lamp’ has a political sense—that a good-natured king may have evil
counsellors. This is, of course, characteristic of cultural forms. The signifiers,
of which even the so-called structures and archetypes are instances, may be
the same in different periods and regions, but the signification may go on
changing. You cannot predict the one from the other. For the meaning of a
sign is culturally and contextually assigned. A sign requires an assignment.
 Not only do folklore items—arising and current in apparently narrow
incommunicable corners and very localised dialects—travel within the
country or culture area, they are also part of an international network.
Archer Taylor’s English Riddles7 gives us current English riddles and their
centuries-old written variants, as well as variants from Africa, India, and the
New World. One can collect today, as I know from experience, oral tales
from illiterate women in Kannada villages that are similar, motif for motif,
to the tales of the Greek Oedipus or to Shakespeare’s King Lear or All’s Well
That Ends Well.
Here we begin to glimpse a paradox: where the so-called pan-Indian
Hindu mythologies of Viṣṇu or Śiva, or the great classics like the
Mahābhārata and the Rāmāyana are unique to India, folklore items such as
proverbs and tales participate in an international network of motifs, genres,
types, and structures—using them all, of course, to say something particular,
local, and unique. One arrives at the paradox that the classics of a culture,
like the well-wrought epics or plays and poetry, are culture- bound forms,
but large portions of the so-called ‘Little Traditions’ are not. The latter
mould and express the values and concerns of the culture nonetheless. Their
forms, their signifiers, however, are not ethno-centred.
One has to resort to subterfuge and theoretical acrobatics to compare the
Sanskrit Mahābhārata and the Greek Iliad or invoke ancient Indo-European
structures (such as the tripartite division of priest, warrior and service
classes) as Dumezil8 does. But the comparison of Cinderella tales from
China to Peru begins with transparent structural resemblances and may end
with significant contrast between one culture’s assignment of meanings and
another’s.9 Unfortunately, comparatists have not paid attention to Indian
folklore and folklorists have usually stopped at identifying types, rarely
going further to ask questions of cultural significance. Detailed comparative
studies of particular proverbs, tales and so on, for which there are well-
attested comparative materials, are called for and would greatly enhance our
understanding of what is specifically Indian or Tamil or Bengali. Because
some of these tales, for example, can be identified in European languages,
classical Sanskrit and in our mother tongues, we can arrive at a most useful
three-way comparison between what is Indian and what is Western; and
within India, between what is Sanskritic and what is characteristic of a
regional culture and a mother tongue—and of course the dialogues and
exchanges among these. Such triangulations, if replicated for several tales,
would give us a body of unique comparative data and analyses.

WRITTEN AND ORAL MEDIA: DIFFERENCES AND

RELATIONS

Folklore also raises and makes us face other central questions; for instance,
questions regarding the differences and relations between written and
spoken media in Indian oral culture.
The relations between oral and written traditions in any culture are not
simple oppositions. They interpenetrate each other and combine in various
ways. Each of us produces more oral materials in our lives than written. We
begin our lives in an oral universe, learn our mother tongues orally first and
imbibe our culture through it. As adults, on any day or occasion, we say
much more than we write. Talk surrounds us and we talk to ourselves, not
only to others, not always even silently, and often we do not even stop when
we fall asleep. Our dreams are filled with speech. Yet, writing is more
permanent; it takes us out of a face-to-face communication and can reach
people far away and centuries later, in ages unborn and accents yet
unknown, as Shakespeare would say. In Sanskrit, a written letter is called
akṣara, ‘imperishable’. Written traditions live surrounded by oral ones and,
are even carried by oral means. As in many other languages, in Kannada the
word for writing (bare) is the same as that for drawing; and until recently to
read meant to read aloud. I have heard of a grand-uncle who would say he
could not read a novel because he had a sore throat. So, too, to write meant
to write down. Writing was an aide memoir, a mnemonic device, for
materials to be rendered oral again. Speech lies dormant in writing until it is
awakened again by one’s own or another’s voice, like these words on this
page as you or I read them.
Sometimes it is thought that the so-called classical texts are fixed and the
so-called folk-texts are constantly changing. Similarly, writing is thought to
be fixed and speech constantly changing. One often identifies the ‘classical’
with the written and the ‘folk’ with the oral. But, for India, we should
distinguish between three sets of independent oppositions. We may then
proceed to examine, complicate and dissolve them. The three are classical vs.
folk, written vs. spoken, fixed vs. free or fluid. The classical, the written and
the fixed do not necessarily belong together [emphasis added]. A text like a
Veda is fixed but was not written down until two thousand years after its
composition. The Vedas were esoteric and credited with magical properties
that would devastate anyone who mispronounced them. They were
transmitted orally but rigorously in elaborate teaching systems from guru to
disciple. Pundits and Vedic experts had what Narayana Rao calls ‘oral
literacy’; they used an almost entirely oral medium, but were learned in
grammar, syntax, logic, and poetics. Their literacy was, as it were, imbued in
their bodies. We speak of a learned man having all his text in his throat,
kanṭhastha; when one is ignorant, one is called ‘a fellow who has no letters in
his belly’ or a nirakṣarakukṣῑ.
Although such oral literacy produced texts that were carefully preserved
verbatim, allowing little change, a text like an epic story in the written
tradition of the Rāmāyaṇa seems to allow endless variations. Hundreds of
versions exist, written, sung, danced and sculpted in South and Southeast
Asian languages. Though I would insist that each of these many tellings
should be treated as a separate (often fixed) text, it is still remarkable that the
orally transmitted Vedas should be remarkably fixed and the written
Rāmāyaṇas should take such liberties with the story and should be almost as
fluid as an oral folktale. The contrast will become clear when we compare
the great Indian epics with the great Western texts. Imagine a Shakespeare
play or Homers Odyssey having as many widely differing versions in
different ages and languages. We cannot jump from this to the paradox that
in India the oral is invariably fixed and the written is what is fluid. The fixed
and the fluid, or what should be called fixed-phrase and free-phrase forms,
exist in both written and spoken texts.
Language like other communication systems, depends on both fixed or
invariant forms, and free or variant ones. Without the one the system would
not be stable; without the other it would not be capable of change,
adaptation, creativity. Our ordinary language is full of fixed forms, not only
in terms of underlying structures at every level, but even in lexical
combinations. To give just one example, idioms like ‘he kicked the bucket’
cannot be changed for tense, article, or number. Any variation such as ‘he is
kicking the bucket; he is kicking a bucket; they are kicking buckets; he has
been kicking the bucket for a week now’ would all be ungrammatical, mean
other things, and be seen as funny. In language, as we move from phonology
to syntax, we become freer and freer in combinatory possibilities. Still, some
things are not subject to variation and not open to innovation. Not even
Shakespeare or Kalidasa, acknowledged masters and not servants of their
languages, can make a new pronoun or add a tense to the language. When
writers like Joyce try to take such liberties, they achieve such specialised
effects that they require glossaries and notes, and explication quickly
becomes a cult and a cottage industry.
In discourse too different genres allow different degrees of fixity and
freedom. Where the written form is only a mnemonic, a score to be
performed orally, it is used freely for improvisation. The texts of a Yakṣagāna
performance or a Kathakali performance are hardly a few pages long, but an
actual performance may take a whole night. The text of a song may be only a
few lines long, but when sung may take an hour, and usually does. On the
other hand, orally transmitted texts have fixed components, formulae,
refrains, obligatory descriptive passages, and traditionally defined motifs
and narrative structures. Different genres have different proportions of
these; for example, a proverb is an entirely fixed-phrase form within a
speech community. One can play on its fixity to produce new effects as wits
like Oscar Wilde, did: ‘Nothing succeeds like excess’, or ‘Work is the curse of
the drinking classes’, or my favourite, ‘All’s well that ends’. In a joke,
everything may be free, but the punchline may be fixed—to garble it would
be to muff the joke. A folksong would have practically every word fixed,
except performance elements like the number of repetitions, or the way a
phrase is broken to accord with the musical phrase. A folktale told by a
grandmother in the kitchen may have nothing at all fixed in the phrasing,
only the design of the story and the sequence of motifs. Yet it may have fixed
phrases, like ‘Open Sesame’ in the story of Ali Bābā—a phrase that his
brother treats as a free phrase, with disastrous results. The Vedas are an
extreme case of a 4,000-hymn cycle fixed in oral transmission, as if it were
inscribed (as secret codes are in spy stories) in the transmitter’s memory.
 Furthermore, oral and written forms in a culture often wish to be like
each other, like the two sexes, male and female, each envying what the other
has. Yet each defines and marries the other. In the oral forms, in folklore,
many devices such as refrains, formulae and memory-training exist to give
the relative permanence of writing. From time to time, in writing traditions,
writers wish to return to the freshness of speech and imitate it, as in modern
Indian (and other) poetry. Flaubert, master of the written word who waited
for days, for the mot juste, is the exemplar of the opposite end of the oral
arts, where to hesitate is to be lost. Yet it was Flaubert who said that style
should be adjusted to the rhythms of respiration.
In all cultures, and especially in the Indian, the oral and the written are
deeply intermeshed in another way [emphasis added]. If we distinguish
composition and transmission, as Ruth Finnegan10 reminds us we should,
we find that in the history of a text, oral and written means may alternate. A
work may be composed orally but transmitted in writing, as Vyāsa said he
did with Gaṉeśa as his scribe. Or it may be composed in writing, as
Kumāravyāsa (Vyāsa junior) said he did in Kannada, but the text kept alive
by gamakis or reciters who know it by heart and chant it aloud. There are of
course texts, such as proverbs and tales, that are usually composed orally
and orally transmitted, many of which never get written down. And texts,
like newspapers—written, printed and silently scanned or read—may never
go through an oral phase. Thus, over a long history, a story may go through
many phases. An oral story gets written up or written down in the Jātakas or
the Pañcatantra. Then (as W. Norman Brown tried to show in a famous
paper) the written text may reach other audiences who pick up the story and
retell it orally, maybe in other languages, and then it gets written down
somewhere else, perhaps starting another cycle of transmissions. That is one
kind of cycle; another may be entirely oral and may run parallel to the oral-
written complex. Many of the differences in our classsical texts like the
Mahābhārata recensions, may be due to the way the texts do not simply go
from one written form to another but get reworked through oral cycles that
surround the written word. Western critical methods, based entirely on an
examination and reconstruction of written texts, made the critical editions
of Indian texts possible. But they may not be suitable for a reconstruction of
the Mahābhārata at all. For methods of Western textual criticism aim at
making tree-diagrams, relating one written version to another,
demonstrating that one came directly from another, reaching back to a
single Ur-text. Texts like the Mahābhārata may not have a reconstructable
Ur-text at all, enmeshed as they were in oral traditions at various stages of
their composition and transmission.
In a folktale told about Aristotle in Europe and about a philosopher in
India, the philosopher meets a village carpenter who has a beautiful old
knife, and asks him, ‘How long have you had this knife?’ The carpenter
answers, ‘Oh, this knife has been in our family for generations. We have
changed the handle a few times and the blade a few times, but it is the same
knife.’ Similarly, the structure of relations may remain constant, while all the
cultural details change, as in a folktale that goes on changing from teller to
teller. Any fixity, any reconstructed archetype, is a fiction, a label, a
convenience.

ORAL TRADITIONS: THE DIFFERENCE THEY MAKE

Thus anyone concerned with written texts has to reckon with the oral
materials that surround it. This contrasts strikingly with modern America,
where the end of any formal oral communication is a written text. You speak
in Congress so that your speech may be read into the Congressional Record;
everything anybody says in a court is typed up; and at the end of what is
supposed to be spontaneous conversation on a TV talk show, you get the
message, ‘Send three dollars to such and such and you can get the transcript
of this show.’ And finally the most popular TV game show, Wheel of Fortune,
has to do with spelling words and phrases. Every letter is cashed into dollars,
every phrase into furniture and a trip to Hawai’i. In a culture like the Indian,
however, and certainly in villages and certain communities to this day,
writing lives within the context of oral traditions. Even newspapers are read
aloud. If you have been near any primary school in a small town or even in
Madras, you would hear the pupils a mile away, for the classes recite their
lessons in a loud chorus. Not only the alphabet and the multiplication tables,
but every major religious or literary text like the Rāmāyaṇa is memorised
and chanted aloud. As Philip Lutgendorf11 has shown in a Chicago doctoral
dissertation, Tulsidas’s Rāmcaritmānas is the focus of cults, festivals, formal
and informal recitations, tableaus and oral forays into interpretations of the
most wide-ranging and ingenious kinds. The author and the text themselves
are the subject of innumerable tales. Every text like that creates a textual
community held together by oral traditions as well as written ones. Scholars
are just now realising that this interweaving of the oral and written is true of
the Quran and the Bible as well.12 But the Indian examples have needed no
pointing out, except of course to scholars like ourselves. As a proverb in
Kannada says, ‘Why do we need a mirror to see a blister on our hands?’ Yet,
we seem to, for we believe in the mirror of writing, or even better, the mirror
of print.
Oral traditions thus enlarge the range and they complicate and balance the
texts we know [emphasis added]. Yet we ignore the oral. Take mythology, for
instance. At present, in all our anthologies of Hindu mythology there is not
one folk myth. Every text is from the Sanskrit, though myths occur in Tamil
and Bengali and every other language. They even occur in scores of written
texts like the sthalapurāṇas, which David Shulman has studied,13 or the ma
galakāvyas which Edward Dimock14 has written about. In the oral tradition,
that literature without letters (eluta eluttu), there are hundreds more. As Alf
Hiltebeital’s work on Draupadi eloquently demonstrates,15 they complement
the Sanskritic myths and epics in important ways. Oral traditions give us
alternative conceptions of deities that balance and complete, and therefore
illuminate the textual conceptions. For instance, the goddesses of pan-
Indian mythologies, like Lakśmi and Saraswatῑ, rise out of the sea churned
by the gods and the antigods; Pārvatῑ is the daughter of the King of
Mountains. They are consort-goddesses; their shrines are subordinate to
those of their spouses, Viṣṇu or Śiva. Their images are carefully sculpted to
the fingertips. They are usually saumya or mild and docile. They preside over
the normal auspicious cycles of life, especially marriage, prosperity and
such.
But look at the village goddesses and see how different they are. Their
myths tell us of ordinary human women who were cheated into marrying
untouchables, or raped by a local villain, or killed and buried by cruel
brothers. Out of such desecrations they rise in fury, grow in stature to
become figures that span heaven and earth, with powers of destruction that
terrify the village into submission, sacrifice and worship. Theirs are not
myths of descent or avatāra, but of ascent from the human into divine
forms. They become Boundary Goddesses of the village, give it their name,
or take their names from the village. While the Sanskritic Breast Goddesses
(as I call them because they give us their breasts) receive vegetarian offerings
of fruit and flowers, these village goddesses require animal sacrifices and a
sprinkling of blood of their devotees. The Tooth Goddesses represent the
other side of the mother (as stepmothers do, in folktales), who punish, afflict
people with plague and pox, and when propitiated heal the afflicted. They
are goddesses of the disrupted lifecycle, deities of crisis; they preside over
famine, plague, death, and madness. Their images are often pots and pans,
faceless stones, sometimes only a severed head. They dwell outside the
village boundaries and are brought in only for special worship, often in
times of crisis. Without them, life is not complete, nor is the Hindu view of
the divine.
The goddess Kāli, as the sanskrit texts present her, is a sankritised version
of hundreds of village goddesses all over the country and certainly partakes
of their fierce aspects. Yet, in the Sanskrit Purāṇas (encyclopaedias of Hindu
myths) and myths based on all of them, Kāli is created by the gods pooling
their weapons and powers and let loose on the Buffalo Demon whom the
male deities cannot destroy. The emphases, details, and major themes of the
village mythologies are quite different. The village Mariyamman goddess
arises out of human deception and tragedy. If the Breast Goddesses are
consorts to their male spouses, the Tooth Goddess is often a virgin and, if
married, she tears her villainous male consort to pieces. He is later
symbolically offered as a buffalo or goat sacrifice to her images. The Consort
Goddesses are auspicious, consecrated. The village goddesses are
ambivalent, they afflict and heal.16
Such a conception of divinity is not confined only to female deities.
Consider the village gods, such as Muttuppattan. He is a brahman who falls
in love with a cobbler chieftain’s daughters, marries them, skins and tans
cowhides, eats cow’s flesh, dies in the battle defending his village against
robbers, and becomes a god to whom his community of cobblers makes
offerings (kodai) of gigantic leather sandals. It is one of the most moving
long poems of south India. Until recently no record or translation of this
tragic story was available. Now Stuart Blackburn has made an effective
translation of it.17
I use the word tragic advisedly. It is customary to speak of Indian
literature as having no genre of tragedy. In the Sanskritic tradition (by which
I mean both works in sanskrit and sanskritised works in our regional
languages), it is true there are no tragedies in the Greek or shakespearean
sense, though some plays of Bhāsa may be exceptions. It is significant, I
think, that his plays were unearthed in south India in areas where dance
dramas like Kathakali developed, dramas that do not flinch from gory
scenes, and where also the more tragic aspects of the Mahābhārata are fully
enacted. Our sense of our literature and its possibilities would change if we
included oral epics like the Tamil villuppattus and the Tulu paddanas18 in
our studies. (Fortunately, a book of essays on Indian oral epics has just been
published.)19 Oral epics embody a theory of emotion different from that of
rasa, explore ranges in the emotional spectrum like shame, terror, fury and
disgust that are not usually explored in the Sanskrit poems and plays
[emphasis added]. And how can we, mere mortals, do without them?
The oral traditions offer us also a different view of the female from the
views found in the written texts. When the Rāmāyaṇa is sung by the
Tamburi Dāsayyas of Mysore, the centre of attention is Sῑtā, her birth,
marriage, exile, sufferings, and final disappearance into Mother Earth. In the
Tamil story of Mayili Rāvaṇan, set in a time after Rāma has defeated the ten-
headed Rāvaṇa, a new thousand-headed Rāvaṇa arises to threaten the gods,
and this time Rāma cannot handle it. It is Sῑtā who goes to war and
demolishes the impossible demon.20
In the Upanisadic creation myth, the Primordial Person or Puruśa is
alone, needs a companion, and splits into male and female, for he is
originally the same size as a man and a woman put together. Then the male
pursues the female and unites with her, creating mankind. She runs from
him, saying, ‘I was born out of you, I cannot unite with you’, and becomes a
cow. He becomes a bull and unites with her, creating cattle. Then she
becomes a she-goat, he a he-goat; they unite and create goats. And so on
down to the ants.
 But see what happens in an oral, folk purāṇa sung ceremonially on
Mādeśvara hill (Karnataka) every year by several bardic groups during the
festival devoted to this hero/saint/god called Mādeśvara.21 The purāṇa
begins with a creation myth.
The Primordial Goddess is born three days before everything else. She
grows up very quickly, attains puberty, and wants a man to satisfy her.
Finding no one around, she creates out of herself Brahmā, the eldest of
the gods, and asks him to grow up quickly and sleep with her. But as he
grows up and she urges him on, Brahmā says, ‘You are my mother. How
can I sleep with you?’ She gets angry, calls him a eunuch, and burns
him down to a heap of ash with the eye of fire in the palm of her hand.
The next day, she creates Viṣṇu, who is very handsome. She cannot wait
for him to grow up and satisfy her. But he too will not sleep with his
mother. So, in a rage, she burns him down to a heap of ash. On the
third day, she creates Śiva, and urges him to grow up and become her
lover. He too has misgivings until she says, ‘Look around and see what
happened to your brothers who refused me.’ He turns around and sees
the two heaps of ash that were once his brothers. He sizes up the
situation and says to his mother, ‘All right, I will do as you say. You
want me to be your husband, don’t you? Do you not want your
husband to be at least equal to you? Don’t you want to teach him all
your skills and give him your powers?’ The Mother Goddess,
Ammavaru, is delighted and says, ‘Of course, I want you to have
everything’, and teaches him all her magic arts and bestows on him all
her powers. Then Śiva, now grown up, says, ‘Let us dance. You must do
whatever I do. Let us see who is better.’ They whirl around in a fantastic
cosmic dance together, each mirroring the other, until suddenly, Śiva
puts his hand on his head in a dance movement. His mother, following
him, puts her hand on her own head and the eye of fire in her palm
begins to burn her. As she burns, she curses Śiva, ‘You, you refused a
woman. May one half of your body become female, may you never get
rid of her!’ That is how Śiva came to be the lord whose one half is
woman. Then as his mother burned down and became a heap of ash,
the eye of fire that lived in her hand came to Śiva and said it had
 nowhere to go. So he took it and slapped it on his forehead. That’s how
he got his third eye.
After his mother had gone up in flames, Śiva looked around and
found the two heaps of ash that were once his brothers. With his newly
learned powers, he revived them. Now the three gods, Brahmā, Viṣṇu,
and Śiva, said to each other, ‘There is work to do. We must create the
worlds.’ One of them said, ‘How can we create without women?’ Then
Śiva sees the third heap of ash that was once their mother, divides it
into three smaller heaps, and gives them life. Out of these portions of
their mother’s ash, come Lakśmi, Saraswatῑ, and Pārvatῑ, the three
consorts of the Hindu trinity, who then marry them. Creation begins.
In the Sanskritic myth, the male gods create the goddess and give her
their powers. In the foregoing myth it is exactly in reverse. She gives Śiva his
powers. In the Sanskritic myth it is the father figures that lust after the
daughters. Here the female too has her share of sexual desire, made explicit.
She is cheated out of her powers by the male god who uses them to destroy
her. Furthermore, her sons still end up marrying portions of their mother—
both Jung and Freud would be interested in that. But the male gods marry
her only after fragmenting and domesticating her into a nice tame
threesome—feminists would be interested in that. This is a way of looking at
male/female power relations very different from anything we know from the
better-known written texts.
I could go on to talk about alternative views of the gods, karma and
chastity, as well as why tales themselves are told. Since I have talked about
them elsewhere, I shall content myself with giving you some short examples.
The gods in the Purāṇas and the heroes in the epics have bodies without
bodily functions: they are not supposed to sweat, urinate, defecate or pass
wind. They do not blink their eyes nor do their feet touch the ground. But in
folk traditions, they have bodies, they are embodied, localised,
domesticated. In the place legend of Gokarṇa (which I heard from Girish
Karnad), Rāvaṇa prays to Śiva and receives from him the boon that Śiva,
with all his goblin attendants, should go with him to Laṇkā. Śiva gives him
the boon, but does not really wish to go. He tells Rāvaṇa that he can carry
him as a linga all the way, but that he should not put it down anywhere until
he reaches Laṇkā. Rāvaṇa agrees. When he gets to Gokarṇa, he must answer
the call of nature. He cannot hold the sacred linga in his hands while he
takes a crap, can he? So he puts it down, and the linga begins to grow
downwards and take root. Rāvaṇa hurries back and tries to twist it out of the
earth, but he is not able to. That is how Gokarṇa has a linga and they say
that, if you dig under it, you will find that it is twisted. Aldous Huxley once
complained that, even for a realistic novelist like Tolstoy, the heroines never
go to the bathroom nor do they menstruate. In the village oral traditions,
they do. Gods like Gaṇeśa, heroes like Bhῑma, demons like Rāvaṇa, or even
poets like Vyāsa cannot help going to the bathroom, and goddesses like
Ganga and Gauri menstruate. As the bhakti poem says:
Bodied, one will hunger.
Bodied, one will lie.
O you, don’t you rib and taunt me again for having a body:
Body Thyself for once like me and see what happens,
O Kamanatha!
Devara Dāsimayya22
Folklore that is in many ways close to bhakti traditions, gives to them and
takes from them, sharing genres, motifs and attitudes, and seems not only to
ask the gods to embody themselves, but actually envisions them as having
bodies with all the needs and ills that flesh is heir to.
Folk renditions of the pan-Indian epics and myths not only bring the gods
home, making the daily world mythic, they also contemporise them. In
village enactments of the Rāmāyaṇa, when Sῑtā has to choose her
bridegroom, princes from all over the universe appear as suitors. In a north
Indian folk version, an Englishman with a pith helmet, a sola topi, and a
hunting rifle regularly appears as one of the suitors of Sῑtā. After all, since
the eighteenth century the English have been a powerful presence in India
and ought to have a place in any epic ‘bridegroom choice’ or svayamvara.
In a Karnataka performance,
Rāma is exiled, and as he takes the little boat on the river Sarayu to go
to the jungle, all of Ayodhyā follows him in tears. He bids them farewell
from his boat, making a short speech: ‘O brothers and sisters, please go
 home now. I take leave of you now, but I will be back in fourteen years.’
Then he leaves, and wanders through the forests. Sῑtā is abducted by
Rāvaṇa, Rāma gathers the monkey army, kills Rāvaṇa, and returns
victorious with Sῑtā. When he arrives at the spot where he had bid his
people farewell fourteen years earlier, he sees a group standing there,
their hair grown grey, their nails long and uncut, their feet rooted to
the banks of the Sarayu. He asks them who they are. They say, ‘O Rāma,
you forgot us when you took leave. You bade farewell only to the men
and women, calling them brothers and sisters. We are the eunuchs of
Ayodhyā. We have waited for you here all these fourteen years.’ Rāma is
very touched by their devotion and, feeling guilty at his negligence,
gives them a boon: ‘O eunuchs of Ayodhyā, may you be reborn in India
again and rule the country as the next Congress party!’23
I can go on forever, detailing what happens to karma or chastity in the
oral tales, retelling the bawdy tales of the villages about clever women who
cheat on their husbands and get away with it, unlike all the chaste women of
the epics who never cheat or the unchaste ones who are chastened by their
infidelity like Ahalyā. But I think I have said enough to argue the essential
relevance of folklore to Indian studies and the alternative view and systems
folklore carries. Folk materials also comment continually on official and
orthodox views and practices in India. So I wish to end with a satiric tale
about kings, gurus and disciples, the legal process, belief in rebirth, and the
very logic of karma that looks for causes in infinite regress. I shall tell it
without any further comment than that here, if we listen, we can hear the
voice of what is fashionably called the subaltern—the woman, the peasant,
the non-literate, those who are marginal to the courts of kings and offices of
the bureaucrats, the centres of powers.24

IN THE KINGDOM OF FOOLISHNESS25

In the kingdom of foolishness, both the king and the minister were idiots.
They did not want to run things like other kings. So they decided to change
night into day and day into night. They ordered everyone to be awake at
night, till their fields and run their businesses only after dark; and they
should all go to bed as soon as the sun came up. If anyone disobeyed, he
would be punished with death. The people did as they were told for fear of
death. The king and the minister were delighted at the success of their
project.
A guru and a disciple arrived in the city. It was a beautiful city, it was
broad daylight, but there was no one about. Everyone was asleep, not a
mouse stirring. Even the cattle had been taught to sleep. The two strangers
were amazed by what they saw and wandered around till evening, when
suddenly the whole town woke up and went about its daily business.
The two men were hungry. Now that the shops were open, they went to
buy some groceries. To their astonishment, they found that everything cost
the same, a single duddu (a small coin)—whether they bought a measure of
rice or a bunch of bananas, it cost a duddu. The guru and his disciple were
delighted. They had never heard of anything like this. They could buy all the
food they wanted for a rupee.
When they had cooked the food and eaten, the guru realised that this was
a kingdom of fools and it would not be a good idea for them to stay here.
‘This is no place for us. Let us go’, he said to his disciple. But the disciple did
not want to leave the place. Everything was cheap here. All he wanted was
good cheap food. The guru said, ‘They are all fools. This will not last very
long and one cannot tell what they will do to you next.’
But the disciple would not listen to the guru’s wisdom. He wanted to stay.
The guru finally gave in and said, ‘Do what you want. I am going’, and he left.
The disciple stayed on, ate his fill everyday, bananas and ghee and rice and
wheat, and grew fat as a streetside sacred bull.
One bright day a thief broke into a rich merchant’s house. He had made a
hole in the wall, sneaked in, and as he was carrying out his loot, the wall of
the old house collapsed on his head and killed him on the spot. His brother
ran to the king and complained: ‘Your Highness, when my brother was
pursuing his ancient trade, wall fell on him and killed him. This merchant is
to blame. He should have built a good strong wall. You must punish the
wrong-doer and compensate the family for this injustice.’
 The king said, ‘Justice will be done. Do not worry’, and at once summoned
the owner of the house.
When the merchant arrived, the king asked him questions
‘What is your name?’
‘Such and such, Your Highness.’
‘Were you at home when the dead man burgled you house?’
‘Yes, my lord. He broke in and the wall was weak. It fell on him.’
‘The accused pleads guilty. Your wall killed this man’s brother. You have
murdered a man. We have to punish you.’
‘Lord’, said the helpless merchant, ‘I did not put up the wall. It is really the
fault of the man who built the wall. He did not build it right. You should
punish him.’
‘Who is that?’
‘My lord, this wall was built in my father’s time. I know the man. He is an
old man now. He lives nearby. ’
The king sent out messengers to bring in the bricklayer who had built the
wall. They brought him tied hand and foot.
‘Yes there, did you build this man’s wall in his father’s time?’
‘Yes, my lord. I did.’
‘What kind of wall is this that you built? It has fallen on a poor man and
killed him. You have murdered him. We have to punish you by death.’
Before the king could order the execution, the poor bricklayer pleaded,
‘Please listen to me before you give your orders. It is true I built this wall and
it was no good. But that was because my mind was not on it. I remember
very well a harlot who was going up and down that street all day with her
anklets jingling and I could not keep my eyes or my mind on the wall I was
building. You must get that harlot. I know where she lives.’
‘You are right. The case deepens. We must look into it. It is not easy to
judge such complicated cases. Let us get that harlot wherever she is.’
 The harlot, now an old woman, came trembling to the court.
‘Did you walk up and down that street many years ago, while this poor
man was building this wall? Did you see him?’
‘Yes, my lord. I remember it very well.’
‘So you did walk up and down, with your anklets jingling. You were
young and you tempted him. So he built a bad wall. It has fallen on a poor
burglar and killed him. You have killed an innocent man. You will have to be
punished.’
She thought for a minute and said, ‘My lord, wait, I know now why I was
walking up and down that street. I had given some gold to the goldsmith to
make some jewellery for me. He was a lazy scoundrel. He made so many
excuses, said he would give it now and he would give it then and so on all
day. He made me walk up and down to his house a dozen times. That was
when this bricklayer fellow saw me. It is not my fault, my lord, it is that
damned goldsmith’s.’
‘Poor thing. She is absolutely right’, thought the king, weighing the
evidence, ‘We have got the real culprit at last. Get the goldsmith wherever he
is hiding, at once!’
The king’s bailiffs searched for the goldsmith who was hiding in a corner
of his shop. When he heard the accusation against him, he had his own story
to tell.
‘My lord’, he said, ‘I’m a poor goldsmith. It is true I made this harlot
woman come many times to my door. I gave her excuses because I could not
finish making her jewellery before I finished the rich merchant’s orders.
They had a wedding coming, and they would not wait. You know how
impatient rich men are!’
‘Who is this rich merchant who kept you from finishing this poor
woman’s jewellery, made her walk up and down, which distracted this
bricklayer which made a mess of his wall, which has now fallen on an
innocent man and killed him? Can you name him?’
The goldsmith named the merchant and he was none other than the
original owner of the house where the wall had fallen. Now justice had come
full circle, thought the king, back to the merchant. When he was rudely
summoned back to the court, he arrived crying. ‘It is not me, but my father
who ordered the jewellery! He is dead! I am innocent!’
But the king consulted his minister and ruled decisively, ‘It is true your
father is the true murderer. He is dead but somebody must be punished in
his place. You have inherited everything from that criminal father of yours,
his riches as well as his sins. I knew at once, even when I set eyes on you that
you were at the root of this horrible crime. You must die.’
And he ordered a new stake to be made ready for the execution. As the
servants sharpened the stake and got it ready for final impaling of the
criminal, it occurred to the minister that the rich merchant was somehow
too thin to be properly executed by the stake. He appealed to the king’s
common sense. The king too worried about it.
‘What shall we do?’ he said, when suddenly it struck him that all they
needed to do was to get a man fat enough to fit the stake. The servants were
immediately all over town looking for a man who would fit the stake, and
their eyes fell on the disciple who had fattened himself for months on
bananas and rice and wheat and ghee.
‘What have I done wrong? I am innocent. I am a saṃnyāsi! he cried.
‘That may be true. But it is the royal decree that we should find a man fat
enough to fit the stake’, they said, and carried him to the place of execution.
He remembered his wise guru’s words: ‘This is a city of fools. You do not
know what they will do next.’ While he was waiting for death, he prayed to
his guru in his heart, asking him to hear his cry wherever he was. The guru
saw everything in a vision. He had magical powers; he could see far and he
could see the future as he could see the present and the past. He arrived at
once to save his disciple who had gotten himself into a scrape again through
love of food.
As soon as he arrived, he scolded the disciple and told him something in a
whisper. Then he went to the king and addressed him
‘O wisest of kings, who is greater? The guru or the disciple?’
‘Of course the guru. No doubt about it. Why do you ask?’
 ‘Then put me to the stake first. Put my disciple to death after me.’
When the disciple heard this, he caught on and began to clamour.
‘Me first! You brought me here first! Put me to death first, not him!’
The guru and the disciple now got in a fight about who should go first.
The king was puzzled by this behaviour. He asked the guru, ‘Why do you
want to die? We chose him because we needed a fat man for the stake.’
‘You should not ask me such questions. Put me to death first.’
‘Why? There is some mystery here. As a wise man you must make me
understand.’
‘Will you promise to put me to death, if I tell you?’ said the guru. The king
gave him his solemn word. The guru took him aside, out of the servant’s
earshot, and whispered to him, ‘Do you know why we want to die right now,
the two of us? We have been all over the world but we have never found a
city like this or a king like you. That stake is the stake of the god of justice. It
is new, it has never had a criminal on it. Whoever dies on it first will be
reborn as the king of this country. And whoever goes next will be the future
minister of this country. We are sick of living the ascetic life. It would be nice
to enjoy ourselves as king and minister for a while. Now keep your word, my
lord, and put us to death. Me first remember.’
The king was now thrown into deep thought. He did not want to lose the
kingdom to someone else in the next round of life. He needed time. So he
ordered the execution postponed till the next day and talked in secret with
his minister. ‘It is not right for us to give the kingdom to others in the next
life. Let us go up the stake ourselves and we will be reborn as king and
minister again. Holy men do not tell lies’, he said, and the minister agreed.
So he told the executions, ‘We will send the criminals tonight. When the
first man comes to you, put him first to death. Then do the same to the
second man. Those are orders. Do not make any mistakes.’
That night, they went secretly to the prison, released the guru and
disciple, disguised themselves as the two and, as arranged beforehand with
their loyal servant, were taken to the stake and promptly executed.
 When the bodies were taken down to be thrown to crows and vultures the
people panicked. They saw before them the dead bodies of the king and the
minister. The city was in confusion.
All night they mourned and discussed the future of the kingdom. Some
people suddenly thought of the guru and the disciple and caught up with
them as they were preparing to leave the town unnoticed. We people need a
king and a minister, said someone. Others agreed. They begged the guru and
the disciple to become their king and their minister. It did not take many
arguments to persuade the disciple, but it took long to persuade the guru.
They finally agreed to rule the kingdom of the foolish king and the silly
minister, on the condition that they would change all the old laws. From
then on, night would again be night and day would again be day, and you
could get nothing for a duddu. It became like any other place.

* This essay appeared in 1990.—ed.

 Glossary of Select Terms
All terms are in Sanskrit except where specified.

abhinaya histrionic representation

Agama a scripture, especially Shaiva

ahimsa non-injury to living beings

Akam (Tamil) early Tamil poems of love

akavunar (Tamil) declaimer

anukriti likeness, imitation

Avadana stories containing Buddhist teachings in Sanskrit

from circa the middle of the first millennium CE

atukalam (Tamil) ‘dance field’ or place for the public performance

of song and dance

bali ritual offerings

bhava emotion

bhikkhu Buddhist monk

bhikkhuni Buddhist nun

Bodhisattva an enlightened, messianic being worshipped by

Buddhists

Brihatsamhita a fifth century ce Sanskrit text on astrology and

mathematics

canror ceyyul poetry of heroes

(Tamil)

charana itinerant painter

chaitya Buddhist shrine usually housing a stupa

chandala an outcaste

chauri a whisk borne by bearers fanning the central

image in a sculptural relief or painting

chhayastambha memorial pillar

chittagara (Pali) picture gallery

Dakshinapatha southern trade and travel route running from

Pratishthana on the Godavari further south

dana gift, donation

dhamma (Prakrit) the creed preached by the Buddha, the ethico-

spiritual message of King Ashoka

dhamma- officials appointed by King Ashoka to spread

mahamatta (Prakrit) dhamma

Dharmachakra the wheel of law/piety said to have been

promulgated by the Buddha

dharana a type of gold coin in the Chalukyan realms

dhvaja flag, column

Dashakumaracharita an eighth century ce long tale (katha) of

adventure in Sanskrit by Dandin

demiurge (Greek) artist

desi local, regional

drishyakavya ‘visual poetry’, i.e., drama

gadyana punch-marked gold coin found in the Chalukyan

realm

gahapati a rich landed person

gana mythical dwarf-followers of deities like Shiva and

Parvati

garbhagriha literally ‘womb’, sanctum sanctorum of a temple

ghatapallava a decorative motif consisting of a pot filled with

overflowing flora

gotra a social grouping traced to a common, mythical

ancestor

gudhamandapa enclosed hall of a temple

Hitopadesha a collection of popular Sanskrit fables in prose

 and verse composed around the beginning of the
second millennium CE

Jatakas birth stories of the 550 births of the Buddha

before that of Prince Siddhartha, composed in
Pali, circa second century bce

jayastambha victory column

kala art, craft

kalasa an auspicious motif of a water pot

karma fruit of action, theory of transmigration based on

it

Kathasaritasagara a twelfth century ce compendium of popular

tales in Sanskrit attributed to Somadeva

kavya highly aesthetic poetry, drama, tale and

biography mainly in Sanskrit and Prakrits

kshaya-vriddhi the principle of foreshortening in painting

kumara a Mauryan prince, governor

kuttar (Tamil) male and female folk dancer-actors

lepya karma terracotta-making

mahamatta (Prakrit) a Mauryan minister

mahatmya a local text extolling a tirtha or place of

pilgrimage

mahasamanta local chief

Mahavastu a text compiled in the first few centuries ce in
hybrid Sanskrit presenting stories from the life of
the Buddha and Boddhisattvas

marga ‘high road’, refers to the universal or classical

modes in Sanskrit literature as compared to the
local (desi)

mattar a measure of land

mithuna a common sculptural theme depicting an

amorous couple

nagaraka a culturally sophisticated urban dweller

nagas a common sculptural motif of snake-deities

nata/nati stage actor/actress

natya drama

nishidi a monumental commemorative stone bearing

(Kannada/Sanskrit) inscriptions by pilgrims

Panchatantra a collection of fables in Sanskrit in verse and

prose from about the middle of the first
millennium CE

panar (Tamil) singers who sang to the accompaniment of a lute

palliyam (Tamil) orchestra

Paramita the six Buddhist perfections or virtues like

generosity and patience

pan (Tamil) melody type

panar (Tamil) singer

panchamu khalinga a Shiva linga or phallus with five heads of the

deity carved or painted around it

Parinibbana (Pali) the passing away of the Buddha

pattana market town

Prakrits eight different dialects found in ancient Indian

texts and epigraphs

pratoli doorway

prekshaka spectator

Puram (Tamil) early Tamil poems of war

pulavar (Tamil) wise men, learned poets

purunar (Tamil) war bard

puttabhedana commercial centre

rajuka a Mauryan revenue official

rasa abstracted essence of feelings evoked through

works of art, nine in number including love,
anger, humour, fear and hatred.

shalabhanjika a fertility symbol consisting of the sculpture of a

woman standing near a tree and holding a branch

Sangam literature early Tamil poems said to be composed at three

legendary assemblies of poets

sangha the Buddhist monastic organisation

Saptamatrika the ‘seven mothers’, female deities widely
worshipped

shilpa art, sculpture

sthapati architect

sreshthi merchant, magnate

stupa a Buddhist funerary mound and object of

veneration

svastika an ancient auspicious ensign carved extensively

on monuments

Therigatha (Pali) collection of verses by early Buddhist nuns, part

of the Tipittaka

tirtha literally ‘ford’, a place of pilgrimage

Tipitaka (Pali) literally ‘Three Baskets’, Buddhist canonical

literature that began to be compiled circa third
century bce

Uttarapatha northern trade and travel route that stretched

from Tamralipti in Bengal to Takshashila in
Pakistan

upasaka dharma Buddhist code for the laity

vajralepa hard plaster applied to the surface before painting

varttana modelling/shading

Vastushastra science of architecture

vidushaka comic sidekick of the hero in Sanskrit plays

vidyadharas semi-divine aerial beings depicted in painting

and sculpture

vihara Buddhist monastery

vimana temple spire

vinoda entertainment

viraliyar (Tamil) female dancers and singers

yakshas personified benevolent nature spirits who are

believed to be custodians of treasures and people

Yakshagana a folk theatre form of Karnataka from medieval

times

yamapata painted scroll displayed by itinerant storytellers

yupa Vedic sacrificial pole

 Notes on the Contributors
Uma Chakravarti is a feminist historian of ancient India and a civil rights
activist. She taught at Miranda House, University of Delhi.
Ananda K. Coomaraswamy was a philosopher and art historian. He served
as curator of the Boston Museum and authored a large number of influential
monographs on traditional South Asian culture and thought.
Vidya Dehejia is Barbara Stoler Miller Professor of Indian and South Asian
Art at Columbia University and the author of several important studies on
Indian iconography and aesthetics.
Devangana Desai is an acclaimed art historian and author, and is vice
president of the Asiatic Society of Mumbai.
K. Kailasapathy was professor of Tamil at Jaffna University and also its first
president. This followed a successful stint in journalism as editor of a Sri
Lankan daily.
Shonaleeka Kaul is a cultural historian who specialises in classical Sanskrit
literature. She is assistant professor in the Department of History, University
of Delhi, and has also been the Dinakar Singh Lecturer in South Asian
Studies at Yale University.
Jonathan M. Kenoyer is an archaeologist and professor of anthropology at
the University of Wisconsin-Madison. He is a leading expert on the
protohistory of the subcontinent.
Stella Kramrisch was an art historian and curator. She taught at the
University of Calcutta, and later at the University of Pennsylvania and the
Philadelphia Museum of Art, and authored a number of pioneering studies
on early Indian cultural forms.
Jaya Mehta is an Odissi dancer with training in the fine arts and ancient
Indian history from College of Art, Delhi, and Jawaharlal Nehru University
respectively. She conducts art-related lecture-demonstration workshops.
Erwin Neumayer is an independent archaeologist and ethnologist best
known for his extensive study of Indian prehistoric art.
A. K. Ramanujan was a poet, linguist, and folklorist who worked in
Kannada, Tamil and Sanskrit. He was professor in the Department of South
Asian Languages at the University of Chicago.
Himanshu Prabha Ray is chairperson of the National Monuments
Authority, Government of India, and former professor at the Centre for
Historical Studies, Jawaharlal Nehru University.
Upinder Singh is professor at the Department of History, University of
Delhi, and author of wide-ranging works on ancient Indian history.
Notes
Introduction
1. In ancient Greece, too, the ordinary words for artist were ‘demiurge’ and ‘technician’, and under

these headings Plato included poets, painters, musicians, archers, weavers, potters, carpenters, etc. See
p. 31 in this volume.
2. The main exception would be where craftsmen were also donors, as at the stupas in Sanchi,

Sannathi and Bharhut. Names of artisans inscribed on their work of art as autograph exist but are rare.
For some examples see Carol Radcliffe Bolon, ‘Evidence for Artists of the Early Chalukya Period’ in
Making Things in South Asia: The Role of Artist and Craftsman, ed. Michael W. Meister, Philadelphia:
University of Pennsylvania, Department of South Asian Studies, 1988, pp. 52–66.
3. For the author’s critical role in Sanskrit literature, see Shonaleeka Kaul, Imagining the Urban:

Sanskrit and the City in Early India, Delhi: Permanent Black, 2010, pp. 1–38. For Tamil literature, see
David Shulman, ‘Patrons and Poets in Tamil Literature and Literary Legend’ in The Powers of Art:
Patronage in Indian Culture, ed. Barbara Stoler Miller, Delhi: Oxford University Press, 1992, pp. 89–
119.
4. See p. 30 in this volume.

5. For a discussion, see Kaul, Imagining the Urban, Introduction.

6. Coomaraswamy, ‘Ornament’ in Figures of Speech or Figures of Thought: Collected Essays on the

Traditional or ‘Normal’ View of Art, London: Luzac & Co., 1946, p. 87.
7. See p. 73 in this volume.

8. See p. 115 in this volume.

9. See p. 136–40 in this volume.

10. See p. 140 in this volume. Similarly, the Heliodorus Pillar at Besnagar, originally a Vaishnava pillar,

came to be worshipped as Khamba Baba by local tribals who were oblivious of the contents of the
second century BCE inscription on the pillar.
11. ‘Interpretive community’ is a notion drawn from reader-response literary theory as formulated by

Stanley Fish, and cited in Richard H. Davis, Lives of Indian Images, Delhi: Motilal Banarsidass, 1999, p.
8.
12. This dichotomy was presented in Niharranjan Ray’s seminal thesis, Maurya and Post-Maurya Art:

A Study in Social and Formal Contrasts, Delhi: Indian Council of Historical Research, 1975.
13. See p. 145 in this volume.

14. This association is repeated in pp. 147, 152 of this volume.

15. For a recent set of essays, see Akira Shimada and Jason Hawkes (eds.), Buddhist Stupas in South

Asia: Recent Archaeological, Art-Historical, and Historical Perspectives, New Delhi: Oxford University
Press, 2009.
16. A recent work, Michael Rabe’s The Great Penance at Mamallapuram: Deciphering a Visual Text,

Chennai: Institute of Asian Studies, 2001 suggests that early South Asian artistic and literary modes of
representation, or visual and literary cultures, share a great deal in common, down to the employment
of rhetorical techniques like shlesha or pun, which could be used in both fine literature like kavya and
monumental sculpture like the monolithic relief at Mahabalipuram. This had already been suggested
by Devangana Desai in her ‘Puns and Intentional Language at Khajuraho’ in Kusumanjali—New
Interpretation of Art and Culture: C Sivaramamurti Commemoration Volume, ed. M. S. Nagaraja Rao,
Delhi: Agam Kala Prakashan, 1987. pp. 383–87.
17. The fact that the aim was certainly to aid the practice of viewing is supported by the additional

evidence of inscriptional labels on a large number of friezes at Bharhut and Sannathi, though not at
Sanchi, which named the story or character depicted. See Vidya Dehejia, ‘Questioning Narrativity and
Inscribed Labels: Buddhist Bharhut, Sannati, and Borobudur’ in Sacred Landscapes in Asia: Shared
Traditions, Multiple Histories, 2007, ed. H. P. Ray, Delhi: India International Centre and Manohar, pp.
285–307. It is also supported by the occurrence of analyptic insertions of ‘flashback’ scenes in the very
midst of a different but connected line of narration. Also see Dehejia’s essay in this volume.
18. Vidya Dehejia, Preface in Discourse in Early Buddhist Art: Visual Narratives in India, Delhi:

Munshiram Manoharlal, 1997.

19. H. P. Ray, ‘Creating Religious Identity: Archaeology of Early Temples in the Malaprabha Valley’, in

Archaeology and Text: The Temple in South Asia, ed. H. P. Ray, Delhi: Oxford University Press, 2009,
pp. 25–28.
20. For example, the Paharpur copper plate inscription refers to the donation of a piece of land by a

brahmana, Nathasharman, and his wife to a Jaina vihara of the panchamahastupanikaya founded by
Nirgrantha shramanacharya of Varanasi, for the worship of the arahats. Quoted by A. K. Narain,
‘Religious Policy and Toleration in Ancient India’ in Essays on Gupta Culture, ed. B. L. Smith, Delhi:
Motilal Banarsidass, 1993, p. 43. Similarly, a brahman woman, married to a devout brahman, gifted a
Buddhist chaitya hall at Kuda. Quoted by Dehejia, ‘The Popular and Collective Basis of Early Buddhist
Patronage’, in The Powers of Art: Patronage in Indian Culture, ed. Barbara Stoler Miller, Delhi: OUP,
1992, p. 40. The Satavahana rulers and their wives made gifts to Buddhist viharas even as they
described themselves as ‘the sole brahmanas’ (ekabahmana) and the performers of Vedic sacrifices.
Then, within the one dynasty, some Kushana kings were Shaiva, some Vaishnava, and others
promoters of the Buddhist creed. Similarly, was Harshavardhana a Shaiva or a Buddhist? Narain
concluded that ‘in ancient India, religion was more personal than institutionalized’ in ibid, p. 49.
21. John E. Cort believes that communities express different sense of identities by allegiance to

different deities. See Cort, ‘Communities, Temples, Identities: Art Histories and Social Histories in
western India’ in Ethnography and Personhood—Notes from the Field, ed., Michael Meister, Jaipur:
Rawat Publications, 2000, chapter 5.
22. Devangana Desai, Erotic Sculpture of India: A Socio-cultural Study, New Delhi: Tata McGraw-Hill

Publishing Company, 1975, p. 30. A. K. Ramanujan’s unmatched essay, ‘Is There an Indian Way of
Thinking?: An Informal Essay’ discusses several principles and behaviour which, he argues,
characterise Indians, in The Collected Essays of A. K. Ramanujan, ed. Vinay Dharwadker, Delhi:
Oxford University Press, 1999, pp. 34–51.
23. G. Buhler observed, with reference to the multi-faith site of Mathura, that a sectarian distinction

among the artisans cannot be established and it is possible that the same set of artists were
commissioned to create the stone work related to different sects—Buddhist, Jaina and Hindu. Cited in
R. N. Mishra, Ancient Artists and Art Activity, Simla: Indian Institute of Advanced Study, 1975, p. 15,
note 61. Mishra reiterated the continuity regarding Brahmanical and Buddhist textual traditions on
architecture. See ibid., pp. 8, 11–12.
24. Desai, Erotic Sculpture of India, p. 3.

25. See M. Somathilake, ‘Understanding the Painted Jatakas in Ancient India and Sri Lanka’ in Sacred

Landscapes in Asia: Shared Traditions, Multiple Histories, ed. H. P. Ray, Delhi: India International
Centre and Manohar, 2007, pp. 309–25.
26. Kapila Vatsyayana has spoken eloquently of the underlying unity of the Indian arts. See her

‘Aesthetic Theories underlying Asian Performing Arts’ in Sacred Landscapes in Asia: Shared Traditions,
Multiple Histories, ed. H. P. Ray, Delhi: India International Centre and Manohar, 2007, pp. 22–23.
27. Natyashastra XXVII.57, 61, cited in Kaul, Imagining the Urban, p. 24, note 57.
28. Many Ramayanas: The Diversity of a Narrative Tradition in South Asia, ed. Paula Richman,

Berkeley: University of California Press, 1991. Inherent to the multiplicity of tellings is the act of
questioning, which occurs both within the narrative at the hands of characters, and as external
listeners or readers respond to the Ramakatha. See Richman, ‘Questioning and Multiplicity within the
Ramayana Tradition’ in Questioning Ramayanas: A South Asian Tradition, ed. Richman, Berkeley:
University of California Press, 2001, pp. 1–21.
29. K. Kailasapathy, Tamil Heroic Poetry, Oxford: Clarendon Press, 1968, pp. 135–86.

30. Kailasapathy, Tamil Heroic Poetry, pp. 11, 13.

31. Ibid., pp. 258–59.

32. Kailasapathy actually spells out three types of bards. The third is an intermediate category of sorts,

the akavunar, who were declaimers and narrators. See p. in this volume.
33. Kailasapthy, Tamil Heroic Poetry, pp. 63–69.

34. A. K. Ramanujan, ‘Towards a Counter-system: Women’s Tales’ in The Collected Essays of A. K.

Ramanujan, p. 429.
35. Stuart H. Blackburn and A. K. Ramanujan, Introduction in Another Harmony: New Essays on the

Folklore of India, eds. Stuart H. Blackburn and A. K. Ramanujan, Delhi: Oxford University Press, 1986,
p. 30.
36. Ramanujan, ‘Two Realms of Kannada Folklore’, in The Collected Essays of A. K. Ramanujan, p. 490.

37. Susan Wadley, ‘The Katha of Sakat: Two Tellings’ in Another Harmony: New Essays on the Folklore

of India, eds. Stuart H. Blackburn and A. K. Ramanujan, Delhi: Oxford university Press, p. 196.
38. Blackburn and Ramanujan, Introduction, Another Harmony, p. 11.

39. Ramanujan, ‘Where Mirrors are Windows: Toward an Anthology of Reflections’ in The Collected

Essays of A. K. Ramanujan, p. 31. The balancing could occur in the form of a control/release dialectic
between the classical and the folk. See Another Harmony, pp. 15–21.
40. Stuart Blackburn and Alan Dundes, Introduction to section on folklore in The Collected Essays of

A. K. Ramanujan, p. 349.
Chapter 1
1. Quintilian IX. 4. 117. Cf. Plato, Republic 601 B.

2. Cited with approval by Herbert Read, Art and Society, 1937, p. 84 from A. N. Whitehead, Religion in

the Making.
3. St. Thomas Aquinas, Summa Theolologica, I–II. 57. 3 c (art is an intellectual virtue); 5. 4. AD I
(beauty pertains to the cognitive, not the appetitive faculty).
4. Charles S. Baldwin, Medieval Rhetoric and Poetic, 1928, p. 3. ‘A real art of speaking, which does not

lay hold upon the truth, does not exist and never will’ (Phaedrus 261 A); cf. Gorgias 463–65, 513 D,
517A, 527 C, Laws 937 E, etc.
5. E. F. Rothschild, The Meaning of Unintelligibility in Modern Art, 1934, p. 98. A ‘purple passage’!

6. St. Thomas Aquinas, Summa Theolologica, I. 45. 6 c, I. 14. 8 c; I. 45.7 c.

7. Gorgias 503 E.

8. Symposium 205 C.

9. Statesman 259 E, Phaedrus 260 E, Laws 938 A, etc. Whenever Plato disparages the mechanical arts it

is with reference to the kinds of work that provide for the well-being of the body only, and not at the
same time spiritual food; he does not connect culture with idleness.
10. Republic 342 B, C. What is made by art is correctly made (Alcibiades 108 B). It will follow that

those who are in possession of and governed by their art and not by their own irrational impulses
which yearn for innovations, will operate in the same way (Republic, 349, 350, Laws 660 B): ‘Art has
fixed ends and ascertained means of operation’ (St. Thomas Aquinas, Summa Theologica, I–II. 47. 4 ad
2 and 49. 5 ad 2). It is in the same way that an oracle, speaking ex cathedra, is infallible, but not so the
man when speaking for himself. Similarly, in the case of the guru.
11. Republic 369 B, C, Statesman 279 C. D., etc.

12. Republic 398, 401 B. f., 605–07, Laws 656 C, etc.

13. Deuteronomy VIII. 3, Luke IV. 4.

14. Republic 376 E, 410, 521 E, F., Laws 673 A, etc.

15. St. Thomas Aquinas, Summa Theologica, I–II. 57. 3 ad 2 (based on Plato’s view of ‘justice’, which

assigns to every man the work for which he is naturally fitted). None of the arts pursue their own
good, but the patron’s (Republic 342 B, 347 A), which lies in the excellence of the product.
16. Timaeus 47 D, E. Cf. Laws 659 E, on the chant.

17. Timaeus 80 B, echoed in Quintillian IX. 117. Cf. Timaeus 47, 90 D.

18. SāhityaDarpana III. 2–3. Cf. Ananda K. Coomaraswamy Transformation of Nature in Art, 1935,

pp. 48–51
19. Sophist226, Phaedrus 227, 243 A, B, Phaedo 66, 67, 82 B, Republic 399 E, Aristotle, Poetics VI. 2,

etc.
20. Laws 840 C.

21. Phaedo 67 D, E.

22. Phaedrus 279 C.

23. Republic 389–98.

24. The aesthetic man is ‘one who is too weak to stand up against pleasure and pain’ (Republic 556 C).

If we think of impassibility (not what we mean by ‘apathy’, but a being superior to the pulls of pleasure
and pain, cf. Bhagavad Gītā II. 56) with horror, it is because we should be ‘unwilling to live without
hunger and thirst, etc., if we could not also suffer the natural consequences of these passions’, the
pleasures of eating and drinking and enjoying fine colours and sounds (Philebus 54 E, 55 B). Our
attitude to pleasures and pains is always passive, if not indeed masochistic.
It is very clear from Republic 606 that the enjoyment of an emotional storm is just what Plato does
not mean by a katharsis; such an indulgence merely fosters the very feelings that we are trying to
suppress!
A perfect parallel is found in the Milinda Pañha; it is asked, of tears, shed for the death of a mother
or shed for love of the Truth, which can be called a ‘cure’ (bhesajjam), i.e. for man’s mortality, and it is
pointed out that the former are fevered, the latter cool, and that it is what cools that cures.
25. Jaiminīya UpaniṣadBrāhmaṇa III. 30. 2 and 39. 2, Bṛhadāraṇyaka Upaniṣad III, 7.3 f., Chāndogya

Upaniṣad VIII. 13, Svetāśvatara Upaniṣad V. 14. Cf. Phaedo 65–69.

26. Statesman 288 C.

27. Philebus 67.

28. Republic 399–404. Cf. Laws 656 E, 660, 797, 798, etc.
29. Laws 659 E.

30. Republic 399 E, Cf. Dante, Paradiso I. 19.

31. Aristotle, Physics 11.2 (194 a. 20).

32. Art is iconography, the making of images or copies of some model, whether visible (presented) or

invisible (contemplated), (Republic 373 B, 377 E, 392–397, 402, Laws 667,669, Statesman 306 D,
Cratylus 439 A, Timaeus 28 A, B, 52 B, C, Sophist 234 C, 236 C, cf, Aristotle, Poetics I. 1 f.). In the same
way Indian works of art are called counterfeits or commensurations (anukṛti, tadākārata, pratikṛti,
pratibimba, pratimāna) and likeness (sārūpya, sādrśya) is demanded. This does not mean that it is a
likeness in all respects that is needed to evoke the original, but an equality as to the whichness and
whatness, or form and force of the archetype; it is this ‘real equality’ or ‘adequacy’ that is the truth and
the beauty of the work (Laws 667, 668, Timaeus 28 A, B, Phaedo 74 f.). We have shown elsewhere that
the Indian sādṛśya does not imply an illusion but only a real equivalence. It is clear from Timaeus 29,
29 that by ‘equality’ and ‘likeness’ Plato also means a real kinship and analogy, and that it is these
qualities that make it possible for an image to ‘interpret’ or ‘deduce’ (Skr. ānī) its archetype.
33. ‘The mathematician’s patterns, like the painter’s or the poet’s, must be beautiful.’ (G. H. Hardy, A

Mathematician’s Apology, 1931). See Ananda K. Coomaraswamy, Why Exhibit Works of Art? 1943, ch.
IX.
34. Philebus 5.1 C.

35. Laws 657 C, 665 C, 700 C, etc.

36. Laws 670 E; St. Thomas Aquinas, Summa Theologica, 1. 91. 3 and I–II. 57.3 ad 2.

37. Pro Quinctio XXV, 78.

38. Republic 395–401, especially 395 C, cf. 401 B. f., 605–607 and Laws 656 C.

39. Republic 400 A and 598 B, cf. Timaeus 29 c.

40. Republic 500 E.

41. Plotinus, Enneads V. 9. 11, like Plato, Timaeus 28 A, B.

42. Exodus XXV, 40.

43. Aitareya Brāhmaṇa VI. 27

44. Śāṅkhāyana Āraṇyaka VIII, 9.

45. Śatapatha Brāhmaṇa VII. 2, 1. 4, cf. III. 3. 3. 16, XIV. 1, 3. 26, etc. and Taittiriya Saṁhitā V. 5. 4. 4.

Whenever the sacrificers are at a loss, they are required to contemplate (cetayadhvam), and the
required form thus seen becomes their model. Cf. Philo, Moses II, 74–76.
46. Republic 377, 402, Laws 667, 668, Timaeus 28 A, B, Phaedrus 243 A, B, Republic 328 C (misuse of

words is a symptom of sickness in the soul).

47. Republic 601, etc. Porphyry tells us that Plotinus refused to have his portrait painted, objecting,

‘Must I consent to leave, as a desirable spectacle for posterity, an image of an image?’ Cf. Asterius,
Bishop of Amasea, circa 340 ce ‘Paint not Christ: for the one humility of his incarnation suffices him’
(Migne, Patrologia Graeca XI. 167). The real basis, of the Semitic objection to ‘graven images’, and of
all other iconoclasm is not an objection to art (adequate symbolism), but an objection to a realism
that implies an essentially idolatrous worship of nature. The figuration of the Ark ‘according to the
pattern that was seen upon the Mount’ (Exodus XXV, 40) is not ‘that kind of imagery with reference
to which the prohibition was given’ (Tertullian, Contra Marcionem II, 22).
48. Timaeus 28 A, B, cf. Note 33. The symbols that are rightly sanctioned by a hieratic art are not

conventionally but naturally correct (Laws 657 A). One distinguishes, accordingly, between le
symbolisme qui sait and le symbolisme qui cherche. It is the former that the iconographer can and must
understand; which he will hardly be able to do unless he is himself accustomed to thinking in these
precise terms.
49. The realities are seen ‘by the eye of the soul’ (Republic 533 D), ‘the soul alone and by itself ’

(Theatetus, 186, 187), ‘gazing ever on what is authentic’ (Timaeus 28 A, cf. Phaedrus 253A), and thus
‘by inwit (intuition) of what really is’ (Philebus 59). Just as in India, where it is only when the senses
have been withdrawn from their objects, only when the eye has been turned round (āvṛtta cakṣus),
and with the eye of gnosis (jñāna cakṣus), that the reality can be apprehended.
50. The contemplative actus primus (Skr. dhĩ, dhyāna) and operative actus secundus (Skr. karma) of the

Scholastic philosophers.
51. ‘One man is able to beget the productions of art, but the ability to judge of their utility or

harmfulness to their users belongs to another’ (Phaedrus 2744 E). These two men are united in the
whole man and complete connoisseur, as they are in the Divine Architect whose ‘judgments’ are
recorded in Genesis I. 25.31.
52. Laws 667; for a need as first and last cause, Republic 369 B. As to ‘wholesomeness’ cf. Richard

Bernheimer in Brynmawr Notes and Monographs IX, 1, pp, 28, 29, ‘There should be a deep ethical
purpose in all of art, of which the classical aesthetic was fully aware . . . . To have forgotten this
purpose before the mirage of absolute patterns and designs is perhaps the fundamental fallacy of the
abstract movement in art.’ The modern abstractionist forgets that the Neolithic formalist was not an
interior decorator but a metaphysical man who had to live by his wits.
53. Laws 668 E, 669 A, B, 670 E.

54. Laws 860 B.

55. Philebus 61 B f.

56. Republic 376 E, 410–12, 521 E f.

57. Republic 519, 520, 539 E, Laws 644, 803, 807. Cf. Bhagavad Gītā III. 1–25, Cf. Ananda K.

Coomaraswamy, ‘Lila’, in Journal of African and Oriental Studies, XXVIII, 1941 and ‘Play and
Seriousness’, Journal of Philosophy, XXXIX, 1942.
58. Deuteronomy VIII. 3, Luke IV. 4, John VI. 58.

59. R. R. Schmidt, Dawn of the Humans Mind [Der Geist der Vorzeit (Translated by R. A. S.

Macalister)], London, 1936, p. 167.

60. Republic 495 E. cf. 522 B, 6II D. C. Theatetus 173 A, B. That ‘industry without art is brutality’ is

hardly flattering to those whose admiration of the industrial system is equal to their interest in it.
Aristotle defines as ‘slaves’ those who have nothing but their bodies to offer (Politicus 1254 b 18). It is
on the work of such ‘slaves’, or literally ‘prostitutes’ that the industrial system of production for profit
ultimately rests: their political freedom does not make of chain-belt workers and other ‘base
mechanics’ what Plato means by ‘free men’.
61. Republic 395 B, 500 D, Cf. Philo.

62. Republic 433 B, 443 C.

63. Republic 370 C. cf. 374 B, C. 347 E, 406 C, etc. Professor Paul Shorey had the naiveté to see in

Plato’s conception of a vocational society ‘an anticipation of Adam Smith’s division of labour’ (Loeb
Library Republic, p. 151, note B). Actually, no two conceptions could be more contrary. In Plato’s
division of labour it is taken for granted that the artist is not a special kind of man, but every man a
special kind of artist; his specialisation is for the good of all concerned, producer and consumer alike;
Adam Smith’s benefits no one but the ‘manufacturer’ and salesman. Plato who detested any
‘fractioning of human faculty’ (Republic 395 B) could hardly have seen in our division of labour a type
of justice (see note 17). Modern research has rediscovered that ‘workers are not governed primarily by
economic motives’ (see Stuart Chase, ‘What Makes the Worker Like to Work?’, Reader's Digest,
February 1941, p. 19)
64. Chuang Tzu as cited by Arthur Waley, Three Ways of Thought in Ancient China, 1939, p. 62. It is

not true to say that ‘the artist is a mercenary living by the sale of his own work’ (F. J. Mather,
Concerning Beauty, 1935, p. 240). He is not working in order to make money, but accepts money (or
its equivalent) in order to be able to go on working at his living; and I say ‘working at his living’
because the man is what he does.
65. ‘A man attains perfection by devotion to his own work . . . by his own work praising Him who

wove this all . . . . Whoever does the work appointed by his own nature incurs no sin.’ Bhagavad Gītā
XVIII. 45 f.
66. Republic 395 C.

67. Republic 377 E, Symposium 197 A.

68. H. J. Rose in Greek Mythology (Second Edition), 1933, p. 11. Mr Greenberg, in the Nation for April

1941, tells us that ‘the modern artist derives his inspiration from the very physical materials he works
with’. Both forget the customary distinction of spirit from matter. What their statements actually mean
is that the modern artist may be excited, but is not inspired.
69. Laws 892 C.

70. Odyssey XIX, 138.

71. Theogony 31.

72. John VIII. 28, cf. V. 19, 31 and VII. 16, 18 (‘He that speaketh of himself seeketh his own glory’). A

writer in Parnassus, for May 1941, speaks of the female nude as Maillol’s ‘exclusive inspiration’. That is
mere hot air: Renoir was not afraid to call spade a spade when he said with what brush he painted.
73. Purgatorio XXIV, 51–54.

74. Phaedrus 260 E.

75. Symposium 201 C.

76. Timaeus 60 C. 90 A.

77. Greater Hippias 288 D.

78. Symposium 197 A.

79. Ambrose on I Corinthians xii. 3. cited in St. Thomas Aquinas, Summa Theologica, I–II. 109, 1.

80. Plato’s doctrine of inspiration is not ‘mechanical’ but ‘dynamic’: in a later theology it became a

matter for debate in which of these two ways the Spirit actuates the interpreter.
81. ‘The madness that comes of God is superior to the sanity which is of human origin’ (Phaedrus 244

D, 245 A). Cr. Timaeus 71 D–72 B, Laws 719 C and Maitri Upaniṣhad VI 34–-7. ‘When one attains to
mindlessness, that is the last step.’ The subject needs a longer explanation: briefly, the supralogical is
superior to the logical, the logical to the illogical.
82. Ion 534, 535.

83. ‘What we call “chants”, which are evidently in reality “incantation” seriously designed to produce

in souls that harmony of which we have been speaking’ (Laws 659 E, cf. 665 C, 656 E, 660 B, 668, 669,
812 C, Republic 399, 424, Theatetus, etc.). Such incantations are what are called in Sanskrit mantras.
84. The whole purpose of contemplation and yoga is to reach that state of being in which there is no

longer any distinction of knower from known, or being from knowing. It is just from this point of
view that while all the arts are imitative, it matters so much what is imitated, a reality or an effect; for
we become like what we think most about; ‘one comes to be of just such stuff as that on which the
mind is set’ (Maitri Upaniṣad VI, 34).
85. ‘To become like God so far as that is possible is to “escape”’ (Theatetus 176 9); ‘We all, with

unveiled face, reflecting as in a mirror the glory of the Lord, are changed into the same image . . . .
Looking not at the things which are seen and temporal, but at the thing which are not seen and are
eternal’ (II Corinthians III, 18 IV. 18), ‘This likeness begins now again to be formed in us’ (St.
Augustine, De Spiritu et Littera 37). Cf. Ananda K. Coomaraswamy, ‘Traditional Conception of Ideal
Portraiture’, Why Exhibit Works of Art?
86. Timaeus 90 D.

87. Aitareya Brāhmaṇa VI. 27.

88. Republic 409, 410.

89. See ‘Ornament’, in the Ananda K. Coomaraswamy, Figures of Speech or Figures of Thought, London:

Luzac and Co., 1946, chapter III.

90. Walter Andrae, Die ionische säule, 1933, p. 65. Cf. Timaeus 28 A, B.
91. Śatapatha Brāhmaṇa VIII. I. 2. 8, Aitareya Brāhmaṇa V. 23, Taittiriya Saṁhitā II. 5. 11. 5,

Jaiminiya Upaniṣad Brāhmaṇa I. 33. 4 (karoty eva vācā . . . gamayati manasā).

Vac is the Muse, and as the Muses are the daughters of Zeus, so is she the daughter of the
Progenitor, of Intellect (manas), i.e., intellectus vel spiritus, ‘the habit of first principles’. As Sarasvatī
she bears the lute and is seated on the sun-bird as vehicle.
92. ‘This the “Beatitude” (ānanda) of Brahma, that by means of Intellect (manas), his highest form, he

betakes himself to “the Woman” (Vāc); a son like himself is born of her’ (Bṛhadāraṇyaka Upaniṣad, IV.
1. 6). The son is Agni, bṛhad uktha, Logos.
93. Rgveda Saṁhitā X, 32. 2 (śreyānsaṁ dakṣaṁ manasā jagṛbhyāt); BṛhadDevatā II. 84. The

governing authority is always masculine, the power feminine.

94. Aitareya Brāhmaṇa V. 33, etc. Śri as brahmavādinī is ‘Theologia’. Śatapatha Brāhmaṇa III. 2, 4, 11,

cf. ‘the Asura’s gibberish’ (III, 2. 1. 23). It is because of the dual possibility of an application of the
Voice to the statement of truth or falsehood that she is called the ‘double-faced’, i.e ‘two-tongued’ (III,
2. 4. 16). These two possibilities correspond to Plato’s distinction of the Uranian from the Pandemic
and Disordered Aphrodite, one the mother of the Uranian or Cosmic Eros, the other, the ‘Queen of
Various Song’ and mother of the Pandemic Eros (Symposium 180, 187, Laws 840 C).
95. Republic 399 E.

96. Jaiminiya Brāhmaṇa, II. 69, 70, 73.

97. Śatapatha Brāhmaṇa, III. 2. 4, 1–6 and 16–22, cf. III. 2. 1, 19–23.

98. Saṁkhyānam is ‘reckoning’ or ‘calculation’. We have seen that accuracy is the first requirement for

good art, and that this amounts to saying that art is essentially iconography, and to be distinguished
by its logic from merely emotional and instinctive expression. It is precisely the precision of ‘classical’
and ‘canonical’ art that modern feeling most resents: we demand organic forms adapted to an ‘in-
feeling’ rather than the measured forms that require ‘in-sight’.
99. Aitareya Āraṇyaka III. 2. 6, sa chandobhir ātmānaṁ samadadhāt, Aitareya Brāhmaṇa VI. 27

chandomavam . . . utmānaṁ saṁskurute.

100. For what Plato means by wings see Phaedrus 246–56 and Ion 534; ‘It is as a bird that the Sacrificer

reaches the world of heaven’ (Paṇcaviinsanśa Brahmana V. 3. 5). Phaedrus 247 C corresponds to Pa
caviinśa Brāahmaṇa XIV. 1. 12, 13 ‘Those who reach the top of the great tree, how do they fare
thereafter? Those who have wings fly forth, those that are wingless fall down’; the former are the ‘wise’,
the latter the ‘foolish’, cf. Phaedrus 249 C. ‘It is only the philosopher’s discriminating mind that is
winged.’ For the Gandharva (Eros) as a winged ‘maker’ and as such the archetype of human poets cf.
Ṛgveda Saṁhitā X. 177. 2 and Jaiminiya Brāhmaṇa III. 36. For ‘metrical wings’, Pa cavimśa Brāhmaṇa
X, 4. 5, XIX. 11. 3, Jaiminīya Upaniṣad Brāhmaṇa III. 13. 10, Atharva Veda Saṁhitā VIII. 9. 12. The
metres are ‘birds’ (T. S. VI. 1.6. 1; P. B. XIX. 11. 8).
101. Śāṅkhāyana Āraṇyaka VIII. 10.

102. Bhagavad Gītā II. 50 yogaḥ karmasu kauśalam. If yoga is also the ‘renunciation’ (sannyāsa) of

works (BG. V. 1, VI. 2), this is only another way of saying the same thing, since this renunciation is
essentially the abandonment of the notion ‘I am the doer’ and a reference of the works to their real
author whose skill is infallible: ‘The Father who dwelleth in me, he doeth the works’. John XIV, 10.
103. Śāṅkhāyana Āranyaka VII. 5. 7; cf. Phaedo 61 A, B.

104. What is meant by vidyā s opposed to avidyā is explicit in Phaedrus 247. ‘All true knowledge is

concerned with what is colourless, formless and intangible (in Skr., avarṇa, arūpa, agrahya) . . . not
such knowledge as has a beginning and varies as it is associated with one or another of the things that
we now call realities, but that which is really real’ (Skr. satyasya satyam). Cf. Philebus 58 A.
105. Sānkhāyana Āraṇyaka XIV

106. Chāndogya Upaniṣad 1. 6. 7 f. Cf. Ananda K. Coomaraswamy, ‘Sunkiss’ in Journal of African and

Oriental Studies, 60, 1940, p. 49, note 11.

107. For all the statements in this paragraph, Chāndogya Upaniṣad I. 6–9, Saṁgita Darpaṇa I. 4–6 and

Daśarūpa I. 12–14.
108. Timaeus 22 B, C.

109. The Nation, 10 December 1938.

Chapter 2
1. The only find of a portable artifact bearing a similar design, which came to my notice, was

discovered in a Mesolithic assemblage at the site of Chandravati near Abu in Rajasthan. This artifact is
a microblade core about 50 mm long, with a rhomboid design engraved on it. The core was found by
V. A. Sonavane during an exploration conducted by the University of Baroda. (Personal
communication, Dr V. A. Sonavane, Dept. of Archaeology, University of Baroda.) Incidentally, these
intricate design patterns are almost like a signum of Mesolithic art and are highly uniform in all
paintings of this period on the Indian subcontinent.
Design patterns in southern Deccan Mesolithic art are so similar to those found in central India
that one is surprised to find this uniformity spanning such a great distance. I have found similar
designs in the Henanegala Galge of Sri Lanka. I believe that this reflects a specific mental condition of
society common to all those whose patterns are similar. From this uniformity I gather that the socio-
economic background of the Mesolithic hunter-gatherers on the Indian subcontinent was, except for
differences due to specific ecological situations, quite the same.
2. Exceptions to this observation are the Mesolithic paintings in the area around Badami where eyes

on animals as well as human beings are very prominently depicted. Also, in the Mesolithic art of the
Vindhya region there are a few exceptions to this general observation. In Kathotia there are two
extremely fine drawings of rhinos, in which both their eyes are in a twisted perspective, that is, side by
side. In Pachmarhi there are a few historic depictions of people in which both eyes are shown,
although the head is painted in profile. These depictions closely resemble the leather shadow puppets
of Andhra Pradesh.
3. The horned boar is also found in the frescoes of Chatal Hujuk in Anatolia, and as in Indian

Mesolithic paintings, he is surrounded by comparatively small human figures.

4. The Mesolithic figures which I found recently near Badami portray females and males in a uniform

manner. The only difference between the sexes is that the breasts of the women have been indicated.
Both men and women have a line composed of multiple minute lines running down the middle of
their equally slender bodies. This line cannot be mistaken for the spine for there are similar designs on
the legs as well. The Mesolithic animal depictions in the same area show an X-ray style and body
patterns similar to those we know of in central India.
5. At least thirty species of trees and plants in the forest (of Bhimbetka) have edible flowers, fruit and

seeds, and several plants have edible roots and tubers. Among the former the more important are:

Mahua (Bassia latifolia) Ber (Zizyphus jujuba)

 Achar (Buchanania latifolia) Imli (Tamarindus indica)

Tendu (Diosphyros tomentosa) Ghator (Zizyphus xylopera)

Bel (Aegle marmelos) Papda (Gardenia latifolia)

Bhandara (Erytherina indica) Khajur (Phoenix sylvestris)

Gular (Ficus glomerata) Sitaphal (Anona squamosa)

Aonla (Phyllanthus emblica) Kakhera or kankher and Menar

Among tubers, the more common are: Semal Kand, Safed Musli, Potia Kand, Karhari Kand and
Khalula.
V. N. Misra, ‘The Acheulian Industry of Rock Shelter III, F-23 at Bhimbetka, Central India: A
Preliminary Study’, Puratattva, no. 8, 1975–76.
6. Of course, within the context of prehistory ‘very rapidly’ does not mean from one day to the next.

This is evident from the fact that, in living memory, tribals have opposed the introduction of certain
types of technology because they actually harmed their mental equilibrium. For instance, Red Indians
opposed the felling of trees or the introduction of the iron plough—for they held that these wounded
the earth which they considered their mother.
Chapter 3
1. H. C. Beck, Classification and Nomenclature of Beads and Pendants, York, PA: George Shumway

Publ., 1973 (first published in Archeologia, LXXVII, 1928).

2. E. J. H. Mackay, Further Excavations at Mohenjodaro, New Delhi: Government of India, 1938 and S.

J. Marshall, Mohenjo-daro and the Indus Civilization, London: A. Probsthain.

3. G. F. Dales and J. M. Kenoyer, ‘Preliminary Report on the Third Season of Work at Harappa,

Pakistan’ in Pakistan Archaeology, 24, 1990, pp. 68–176 and J. F. Jarrige, ‘Excavations at Mehrgarh-
Nausharo’, Pakistan Archaeology, 10–22, 1986, pp. 62-131.
4. R. S. Bisht, ‘Further Excavations at Banawali: 1983–84’ in Archaeology and History, eds. B. M. Pande

and B. D. Chattopadhyaya, Delhi: Agam Kala Prakashan, pp. 135–56 and ‘A New Model of the
Harappan Town Planning as revealed at Dholovira in Kutch: A Surface Study of its Plan and
Architecture’ in History and Archaeology, ed. B. Chatterjee, Delhi: Ramanand Vidhya Bhawan, 397–
408., and K. T. Hedge, V. H. Sonawane, D. R. Shah, K. K. Bhan, K. K. Ajitprasad and S. P. Chardnan,
‘Excavation at Nagwada—1986 and 1987: A Preliminary Report’, Man and Environment, 12, 1988, pp.
55–65.
5. J. M. Kenoyer, M. Vidale and K. K. Bhan, ‘Contemporary Stone Bead Making in Khambhat India:

Patterns of Craft Specialization and Organization of Production as Reflected in the Archaeological

Record’, World Archaeology, 23,1, 1991, pp. 44–63.
6. J. M. Kenoyer, ‘The Indus Tradition of Pakistan and Western India’, Journal of World Prehistory, 5(4),

1991: 331–85.
7. M. R. Mughal, ‘Further Evidence of the Early Harappan Culture in the Greater Indus Valley: 1971–

90’, South Asian Studies, 6, 1990, pp. 175–200.

8. J. M. Kenoyer, ‘The Indus Tradition of Pakistan and Western India’.

9. K. N. Dikshit, ‘Late Harappa in Northern India’, in Frontiers of the Indus Civilization, eds. B. B. Lal

and S. P. Gupta, New Delhi: Books and Books, 1984, 253–70, Kenoyer, ‘The Indus Tradition’, and J. G.
Shaffer, ‘The Indus Valley, Baluchistan and Helmand Traditions: Neolithic Through Bronze Age’ in
Chronologies in Old World Archaeology (Third Edition), ed. R. Ehrich, Chicago: University of Chicago
Press, 1988, pp. 441–64.
10. Shaffer, ‘The Indus Valley, Baluchistan and Helmand Traditions’.
11. J. M. Kenoyer, ‘Shell Working Industries of the Indus Civilization: An Archaeological and

Ethnographic Perspective’, Ph.D. dissertation, University of California, Berkeley, 1983.

12. For more discussion, see A. Rubin, Art as Technology, Beverly Hills: Hillcrest Press, 1989.

13. For e.g., P. R. S. Moorey, ‘Materials and Manufacture in Ancient Mesopotamia: The evidence of

archaeology and art’, BAR Int. Ser 237, Oxford, 1985.

14. A. Lucas, Ancient Egyptian Materials and Industries, 4th edition, rev. J. R. Harris, London: Edward

Arnold Ltd, 1962.

15. Beck, Classification and Nomenclature of Beads and Pendants.

16. J. M. Kenoyer, ‘Shell Working Industries of the Indus Civilization: A Summary’, Paléorient, 10, 1,

1985, pp. 49–63.

17. J. F. Jarrige, ‘Economy and Society in the Early Chalcolithic/Bronze Age of Baluchistan: New

perspective from Recent Excavations at Mehrgarh’, South Asian Archaeology 1979, ed. H. Hartel,
Berlin: Dietrich Reimer, pp. 93–114; J. F. Jarrige and R. Meadow, ‘The Antecedents of Civilization in
the Indus Valley’, Scientific American, 243, 2, 1980, pp. 122–33, and B. Barthelemy de Saizieu, ‘Le
cimetiere neolithique de Mehrgarh (Balouchistan Pakistanais). Apport de l’analyse factorielle’,
Paléorient 16, 1, 1990, pp. 23–43.
18. Mackay, Further Excavations at Mohenjodaro; Marshall, Mohenjo-daro and the Indus Civilization,

and M. Vidale, Some Observations and Conjectures on a Group of Steatite-Debitage Concentrations on

the Surface of Mohenjo-Daro, IsMEO, 1989.
19. Mackay, Further Excavations at Mohenjodaro and Marshall, Mohenjo-daro and he Indus

civilization.
20. B. McCarthy and P. Vandiver, ‘Ancient High-strength Ceramics: Faience Manufacture at Harappa

(Pakistan), ca. 2300–1800 BC’, in Materials Issues in Art and Archaeology 2, eds. P. Vandiver, J. Druzik
and G. Wheeler, Pittsburg: Materials Research Society, 1990, p. 2.
21. M. A. Halim and M. Vidal, ‘Kilns, Bangles and Coated Vessels: Ceramic Production in Closed

Containers at Mohenjodaro’, in Interim Reports, vol. I, eds. M. Jansen and G. Urban, Aachen: RWTH-
IsMEO, 1984, pp. 63–97.
22. Dales and Kenoyer, ‘Preliminary Report on the Third Season of Work at Harappa, Pakistan’.
23. J. F. Jarrige, ‘Excavations at Mehrgarh: Their Significance for Understanding the Background of the

Harappan Civilization’, in Harappan Civilization, ed. G. L. Possehl, 1982, New Delhi: Oxford and IBH
Publishing Co., pp. 79–84.
24. J. F. Jarrige, ‘Les Cites oubliees de l’Indus’, Introduction in Les Cites oubliees de l'Indus, in ed. J. F.

Jarrige, 1988, Paris: Musee National des Arts Asiatiques Guimet, pp. 13–37.
25. Ibid.

26. Ibid., p. 35.

27. C. Jarrige, ‘Terracotta Figurines from Nindowari’ in South Asian Archaeology 1981, ed. B. Allchin,

Cambridge: Cambridge University Press, 1984, pp. 129–35 and ‘Les figurines humaines du
Baluchistan’ in Les Cites oubliees de l'Indus, ed. J. F. Jarrige, 1988, Paris: Musee National des Arts
Asiatiques Guimet, pp. 65–70.
28. Mackay, Further Excavations at Mohenjodaro, pl. LXXXV, pp. 5, 10.

29. J. F. Jarrige, ‘Les Cites oubliees de l’Indus: Introduction’ in Les Cites oubliees de l'Indus, ed. J.-F.

Jarrige, Paris: Musee National des Arts Asiatiques Guimet, 1988.

30. Mackay, Further Excavations at Mohenjodaro, pl. LXXII, p. 10 and pl. C.

31. Ibid., C, f.

32. Ibid., p. 335, pl. LXXXVII, p. 222.

33. Dales and Kenoyer, ‘Preliminary Report on the Third Season of Work at Harappa, Pakistan’.

34. R. E. M. Wheeler, ‘Harappa 1946: The Defenses and Cemetery R-37’, Ancient India 3, 1947, pp. 58–

130.
35. M. R. Mughal, ‘Harappa—1966 (Cemetery R 37)’, Pakistan Archaeology, 5, 1968, pp. 63–68.

36. Dales and Kenoyer, ‘Preliminary Report on the Third Season of Work at Harappa, Pakistan’.

37. A. Samzun, ‘La chalcolithique ancien a Mehrgarh’ in Les Cites oubliees de l’Indus, Paris: Musee

National des Arts Asiatiques Guimet, 1988, pp. 53–55.

38. J. F. Jarrige, ‘Les Cites oubliees de l’Indus’

39. C. Jarrige, ‘Terracotta Figurines from Nindowari’ and ‘Les figurines humaines du Baluchistan’.
40. Marshall, Mohenjo-daro and the Indus Civilization, pls. CXLVII, A, CXLIX, CLI, A, pp. 519–24;

Mackay, Further Excavations at Mohenjodaro, pl. CXXXV, pp. 19, 22 and pl. CXLI, p. 61, and M. S.
Vats, Excavations at Harappa, Delhi: Govt. of India Press, 1940.
41. Marshall, Mohenjo-daro and the Indus Civilization, vol. III, pl. CXLIX, p. 4.

42. Vats, Excavations at Harappa, p. 64, pl. CXXXVII, pp. 8–13.

43. Marshall, Mohenjo-daro and the Indus Civilization, vol. III, pl. CLI, A.

44. Ibid., vol. I, p. 34.

45. Walter A. Fairservis, personal communication.

46. Marshall, Mohenjo-daro and the Indus Civilization and Mackay, Further Excavations at

Mohenjodaro and Chanhu-Daro Excavations 1935–36, New Haven, CN: American Oriental Society,
1943.
47. Kenoyer, ‘Shell Working Industries of the Indus Civilization: An Archaeological and Ethnographic

Perspective’.
Chapter 4
1. As there is a great deal of literature on various aspects of Aśoka’s inscriptions and his dhamma, this

article will refer only to those writings that are directly relevant to the arguments being made.
2. The Delhi-Meerut pillar is not in situ and has had a rather eventful history which includes

dismemberment and reconstitution. Nevertheless as it stands today, its inscriptions cover an area
roughly between 18 and 21 feet above the base. I have not seen all of Aśoka’s pillar and rock
inscriptions and can only go by those that are situated in Delhi—the Delhi-Topra and Delhi-Meerut
pillars, the Rock Edicts at Bahapur or Srinivaspuri, and the replica of the Girnar rock at the National
Museum. My colleague, Mr P. S. Dwivedi tells me that he has seen the Kalsi rock and that he did not
find it possible to read the top-most lines of the set of edicts while standing at the base. He adds that
the average ancient Indian could not have been significantly taller than he is.
3. This is a point that has been recognised by several scholars including Gerard Fussman, ‘Central and

Provincial Administration in Ancient India: The Problem of the Mauryan Empire’, The Indian
Historical Review, vol. XIV, nos. 1–2, July 1987 and January 1988, p. 59. We are talking here about
modes of transmission, not routes of transmission of the edicts, about which scholars such as
Fussman, K. R. Norman, and U. Schneider have put forward their hypotheses.
4. E. Hultzsch (ed.), Corpus Inscriptionum Indicarum, vol. I: Inscriptions of Aśoka, (New Edition)

Oxford, 1925, p. 94, II. 17–18; p. 112, I. 9.

5. Ibid., p. 98, II. 10–11; p. 117, I. 15. The three Caturmāsīs and the Tiṣya tithis are referred to in Pillar

Edict V in connection with the specific occasions on which Aśoka declared a ban on the killing and
selling of fish, on the killing of animals in the elephant- park and in the fishermen’s preserves, on the
castration of bulls, he-rams, boars and other animals, and the branding of horses and bullocks (ibid.,
pp. 126. 128; II. 11–19). Hultzsch points out that the Caturmāsīs were certain full moon days of each
of the three seasons (summer, rains, and winter) (ibid., p. 128, fn.1). D. C. Sircar concurs, elucidating
that the Caturmāsīs refer to the full moon day of the months of Aṣāḍha, Kārttika, and Phālguna,
which were regarded as auspicious days. He also suggests that the Tiṣya constellation was perhaps
considered auspicious by Aśoka because he was born under it (D. C. Sircar, Inscriptions of Aśoka,
second revised edn., Delhi, 1967, pp. 80, 85).
6. sotaviya
7. The priority of the Major Rock Edicts over the Pillar Edicts is clear from the chronological

references in the inscriptions. While the Major Rock Edicts contain chronological references ranging
from the eighth to the thirteenth years after the abhiṣeka, Pillar Edicts I, IV, V, and VI state that they
were inscribed twenty-six years, and Pillar Edict VII twenty-seven years, after the consecration. The
Greek-Aramaic Minor Rock Edict found at two places near Kandahar in Afghanistan is dated ten
years after the abhiṣeka. This lends credence to the contention that the Minor Rock Edicts are among
the earliest of Aśoka’s inscriptions. As for the inscriptions in the Barabar hills, the one in the Sudama
Cave, which records the dedication of the Nyagrodha cave to the Ājīvikas is dated twelve years after
the abhiṣeka, while the inscription in the Karna Chaupar Cave refers to the dedication of this cave for
the stay of ascetics when nineteen years had elapsed since the abhiṣeka (op. cit., p. 65). The Nigali
Sagar and Rummindei Pillar inscriptions refer to events that occurred fourteen and twenty years
respectively after the abhiṣeka. All these dates are to be understood as references to expired, not
current, years. For a comprehensive catalogue of the various Aśokan inscriptions and their contents,
see F. R. Allchin and K. R. Norman, ‘Guide to the Aśokan inscriptions’, South Asian Studies, vol. I,
1985, pp. 43–49.
8. V. S. Agrawala has discussed in detail the rich and many-faceted symbolism of the various parts of

the Aśokan pillars, with particular focus on the Sarnath pillar (The Wheel Flag of India: Cakra-dhvaja
[Being a History and Exposition of the Meaning of the Dharma-Chakra and the Sarnath Lion Capital],
Varanasi, 1964). John Irwin has also explored various aspects of the symbolism of the Aśokan pillars
in a four-part series of articles ‘Aśokan Pillars: a reassessment of the evidence’, Burlington Magazine,
vol. CXV, 1973, pp. 706–28; vol. CXVI, 1974, pp. 712–27; vol. CXVII, 1975, pp. 631–43, and vol.
CXVlIl, 1976, pp. 734–53. See also John Irwin, ‘Buddhism and the Cosmic Pillar’, in Orientalia losephi
Tucci Moriae dictata. Edenda curaverunt, eds. Gnoli and L. Lanciotti, Serie Orientale Roma, vol. II,
Rome, 1987, pp. 635–60.
9. Whether some of the pillars are pre-Aśokan, as lrwin has argued, or were installed by him in the

earlier part of his reign, is a matter of debate.

10. While the first three of these can be assigned to the north-western part of the Indian subcontinent,

there is some debate regarding the Risṭikas and Pitiṇikas, who perhaps can be placed in the southern
region between the Godavari and Krishna rivers (see B. M. Barua, Aśoka and his Times [Second
Edition], Calcutta, 1955, pp. 83-87).
11. For the various differences in details between the Aśokan and Persepolitan pillars, see Niharranjan

Ray, Maurya and Post-Maurya Art, New Delhi, 1975, pp. 24–26.
12. I am indebted here to the ideas that Margaret Cool Root has put forward regarding Achaemenid

art (The King and Kingship in Achaemenid Art: Essays on the Creation of an Iconography of Empire,
Leiden, 1979).
13. S. P. Gupta (The Roots of Indian Art, Delhi, 1980) has dealt in great detail with various elements of

Mauryan art. While he accepts West Asian influence on certain motifs employed in Mauryan art, he is
at pains to emphasize how these motifs were ‘lndianized’. What is being suggested in this paper is not
an ‘Indianization’ of individual artistic elements but the uniqueness of the artistic and epigraphic
whole. Also, I do not agree with what Gupta has to say about the nature of the difference between
Achaemenid and Aśokan kingship.
14. Agrawala, The Wheel Flag of India, p. 54.

15. Gupta, The Roots of Indian Art, pp. 311–12. We may take note of the peacock engraved below

ground-level on the Lauriya-Nandangarh pillar and the peacocks and fish executed at a similar level
on the Rampurva pillar (Ibid., pp. 23–24); but we will not include these motifs in our discussion as
they were not meant to be seen.
16. Gupta, The Roots of Indian Art, pp. 126–27.

17. Agrawala, The Wheel Flag of India, pp. 29–53.

18. See Vibha Tripathi and Ajeet K. Srivastava, The Indus Terracattas, Delhi, 1994, pp. 130–40. Also,

Sir Mortimer Wheeler, The Indus Civilization (Third Edition), Cambridge, 1968, pp. 101–04.
19. Gupta, The Roots of Indian Art, p. 5

20. For details, see ibid., pp. 81 ff., 166, 181.

21. Michael Mitchiner, The Origins of Indian Coinage, London, 1973.

22. Irwin, ‘Aśokan Pillars: A reassessment of the evidence—iv: Symbolism, Burlington Magazine, vol.

CXVIII, no. 884, November 1976, p. 747.

23. Radhakumud Mookerji, Aśoka (Third Edition, Revised and Enlarged), Delhi, 1962, pp. 62, 90.

24. Ibid., p. 76.

25. A. Foucher, The Beginnings of Buddhist Art and other essays in Indian and Central-Asian

Archaeology, Paris, London, 1917, p. 21. In the same place, while discussing symbols on coins,
Foucher suggests that the lotus symbolises the birth of the Buddha, representing the lotuses that,
according to tradition, sprang up beneath his first seven steps.
26. Gupta, The Roots of Indian Art, p. 43.

27. See Jean Przyluski, ‘Le Symbolisme du Pilier de Sarnath’, Linossier, vol. II, 1932, pp. 481–98.

28. Gupta, The Roots of Indian Art, p. 26.

29. On the other hand, one could simply see the lion as an imperial symbol used by Aśoka, whether he

was addressing his subjects in general or the Buddhist saṅgha in particular.

30. Sir Alexander Cunningham, The Stūpa of Bharhut (First Edition), London, 1879, p. 121.

31. See Sir John Marshall, Alfred Foucher and N.G. Majumdar, The Monuments of Sanchi, 3 vols, 1st

edn, London, 1940; reprint edn, 1983, plates 12a and 12b, 74, 3a and 5a, 95. The convention of this
sort of pillar apparently continued into the Gupta period, because the pillar of that age at Sanchi,
while stylistically different, is similar in terms of composition (shaft, inverted lotus, addorsed lions,
dharmacakra) to the pillars cited above. It may be added that the pillars depicted in the Sanchi reliefs
are frequently flanked by individuals in a pose of veneration, indicating that they were among the
important symbols of the Buddhist dhamma at the time.
32. Marshall, Foucher and Majumdar, The Monuments of Sanchi, vol. I, pp. 48–49.

33. For a hypothesis on the rationale that underlay Aśoka’s dhamma and the relationship between

dhamma and the Mauryan state, see Romila Thapar, Aśoka and the Decline of the Mauryas, Oxford,
1963, chapter V.
34. This is Minor Edict No. IV in D. C. Sircar’s classification.

35. This is Minor Rock Edict No. II in D. C. Sircar’s classification.

36. Bühler cited in Corpus Inscriptionum Indicarum, ed. Hultzsch, p. 17, fn. 2.

37. D. C. Sircar, Inscriptions of Aśoka, p. 42. Lines 10–11 of the Ahraura edict read as follows: ‘. . . esa

sāvane vivuthe[na] [du]ve sapaṁna-lati sati aṁ maṁ[c]e Budhasa salīle alodhe [ti] . . . ’
38. Hultzsch, Corpus Inscriptionum Indicarum, p. 50.

39. Ibid., pp. 91–92; p. 92. fn. 1.

40. Ibid., p. 27, fn. 2.

41. It may be noted that the elephant appears in two other contexts that may be assigned to the

Mauryan period. The first is at Dhauli, where the forepart of an elephant is sculpted out of rock. The
second is on the bas relief carved on the recessed portion of the arch of the doorway of the Lomas
Rishi cave in the Barabar hills, which depicts four pairs of elephants advancing towards stūpas. These
caves were associated with the Ājīvika sect.
42. An exception is A. Ghosh’s paper ‘The Pillar of Aśoka: Their Purpose’, East and West, New Series,

vol. XVII, nos. 3–4, pp. 273–75. Ghosh argues that although the evidence is not conclusive, the
Aśokan pillars seem to have marked sites of Buddhist stūpas or shrines constructed under the
direction of Aśoka.
43. Thapar, Aśoka and the Decline of the Mauryas.

44. The shift from rock to pillar seems to have happened by the twenty-first year after the abhiṣeka,

which is when the Rummindei pillar was erected.

45. James Legge, The Travels of Fa-hsien: Fa-hsien’s Record of Buddhistic Kingdoms (Reprint Edition),

Delhi, 1971, pp. 47–50.

46. Samuel Beal, Chinese Accounts of India: Translated from the Chinese of Hiuen Tsang (Reprint

Edition), vol. 2, Calcutta, 1957, p. 277.

47. Ibid., p. 274.

48. Ibid., p. 283.

49. Ibid., p. 292.

50. Ibid.

51. F. Kielhorn, ‘Delhi Siwalik Pillar Inscriptions of Vīsaladeva; the Vikrama year 1220, Indian

Antiquary, vol. XIX, 1890, pp. 215–19.

52. H.M. Elliot and J. Dowson, The History of India as Told by Its Own Historians:Tarikh-I Firoz Shahi

of Shams-I Siraj ‘Afif (Reprint Edition), Calcutta, 1953, pp. 91–95.

53. Pushpa Prasad, Sanskrit Inscriptions of Delhi Sultanate, 1191–1526, Bombay, 1990, pp. 40–41.

54. The text of the short inscriptions is given by Alexander Cunningham in Vol. 5 of the

Archaeological Survey Reports (‘Report for the year 1872–73’), pp. 143–44.
55. Prasad, Sanskrit Inscriptions of Delhi Sultanate, pp. 19–20, 32, 41–42.

56. These details are given in Cunningham’s Tour Reports Made During the Years 1862–65, vol. 1, pp.

309–11.
57. The details that follow are based on Cunningham’s account in ibid., pp. 298–301.
58. Ibid., p. 300.

59. Ibid., pp. 73–74.

60. Harry Falk, ‘A Neglected Pillar of Aśoka’, South Asian Archaeology, vol. I, 1995, pp. 429–38.

61. A. C. L. Carlleyle, ‘Report of Tours in Gorakhpur, Saran, and Ghazipur in 1877–78–79 and 80’,

Archaeological Survey of India Report, vol. XXII, p. 51.

62. Cunningham, Tour Reports Made During the Years 1862–65, pp. 61, 74. That such nomenclature of

the Aśokan pillars continues into the present time is suggested, for instance, by the notice of a
sandstone pillar (probably Aśokan) at Bhagalpur, which is locally known as ‘Bhīm-kā-Chharī’ (Dilip
K. Chakrabarti, R. N. Singh and Krishnanand Tripathi, ‘Notes on the Archaeology of the Sarayupar
Plain, Eastern Uttar Pradesh’, South Asian Studies, vol. XIII, 1997, p. 290).
63. Cunningham, Tour Reports Made During the Years 1862–65, p. 67.

64. A few other examples of what may be parts of Aśokan pillars being worshipped in present times as

Śiva lingas are given in S.P. Gupta, The Roots of Indian Tradition, p. 27.
Chapter 5
1. See also S. P. Gupta, ‘Sociological Interpretation of Ancient Indian Terracottas’, Proceedings of the

4th Annual Conference of the Indian Archaeological Society, Nagpur. 1970.

2. Three terracotta animal figures have been found from the late phase of Painted Grey Ware (PGW)

level of Hastinapur. (Ancient India, nos. 10–11, 1954–55 pl. XLII, p. 86). The chronology of the late
level of the PGW has now been revised to circa 800–400 bce. The late levels of PGW and early levels
of Northern Black Polished Ware (NBPW) exist side by side at several sites. (D. P. Agarwal, The
Copper Bronze Age in India, New Delhi: Munshiram Manoharlal, 1971, pp. 84ff, 102). If this is the
case, the animal figurines of Hastinapur fall within the early phase of the period under discussion.
3. A. Ghosh, The City in Early Historical India, Shimla, 1973, p. 71.

4. Moti Chandra, ‘Architectural Data in Jain Canonical Literature’, Journal of the Bombay Branch of the

Royal Asiatic Society, vol. 26, 1951, p. 71 and U. N. Ray, Pracina Bharatiya Nagara tatha Nagara-
Sannivesa’ (Hindi) Gopinath Kaviraj Abhinandana Grantha 1967, pp. 447ff.
5. M. M. Singh, Life in North-Eastern India in Pre-Mauryan Times, Delhi, 1967, pp. 236ff.

6. R. S. Sharma, Sūdras in Ancient India, Delhi, 1958, pp. 180–81 and A. Bose, Social and Rural

Economy of Northern India (cir. 600 BC–200 AD), Calcutta: Motilal Banarsidass, 1945, vol. II, pp. 206–
07.
7. Agarwal, The Copper Bronze Age in India, pp. 102, 108.

8. Indian Archaeology-—A Review (henceforth IAR), 1963–64, p. 8, pl. V-A.

9. IAR, 1955–56, P. 22, pl. XXXII-A and B. P. Sinha and L. A. Narain, Pataliputra Excavation, 1955–56,

p. 10, 41, pls. X, XVI.

10. IAR, 1961–62, p. 7.

11. B. P. Sinha in Purātattva, no. 6, 1972–73, p. 71, pl. IV.

12. IAR, 1959–60, p. 14; IAR 1963–64, p. 8.

13. IAR, 1963–64, pl. V–A, p. 8.

14. IAR, 1960–61, pl. LIX. The report does not indicate the date of the figurines.
15. Sinha and Narain, Pataliputra Excavation, p. 41, pl. x.

16. IAR, 1960–61, p. 5, IAR 1968–69, p. 6 and Dilip Chakrabarti in the Journal of Economic and Social

History of the Orient, vol. 15, 1972, pp. 216–18.

17. Niharranjan Ray, Maurya and Suṅga Art, Calcutta: University of Calcutta, 1945.

18. Niharranjan Ray in The Age of Imperial Unity, Bombay: Bharatiya Vidya Bhavan, p. 533.

19. T. Webster, Hellenistic Art, London, 1966, pls. 13, 18.

20. V. S. Agarwala, Pāṇinikālīn Bhāratavarṣa (Hindi), p. 233.

21. R. S. Sharma, Sūdras in Ancient India, p. 89.

22. IAR, 1961–62, pl., IV-A, pp. 4–5.

23. IAR, 1963–64, pl., V–B, pp. 8–9.

24. Devangana Desai, Erotic Sculpture of India, A Socio–Cultural Study, New Delhi: Tata McGraw-Hill,

pp. 10–12.
25. IAR, 1954–55, p. 15, pl. XXVI-A.

26. Ancient India, no. 4, 1947–48, pl. XXXI–A, pp. 106–07.

27. Ancient India, no. 10–11, 1954–55, pl. XXXVI–4.

28. IAR, 1967–68, pl. XXIII.

29. B. P. Sinha and S. C. Roy, Vaisali Excavations 1958–62, Patna: Directorate of Archaeology, 1969, pl.

XL–2 & 4.
30. IAR, 1964–65, pl. XXXVI 1& 15. pp. 42–43.

31. C. R. Sharma, Excavation at Kausambi (1957–59), Allahabad: Institute of Archaeology, pl. 44–53

and S. C. Kala, Bhāratīya Mṛttakakalā (Hindi), Allahabad, fig. 4.

32. Mortimer Wheeler, Charsada-A Metropolis of the Northwest Frontier, Oxford, 1962, pp. 104ff.

33. S. C. Kala, Terracotta Figurines from Kauśāmbī, Allahabad, 1950, p. 13.

34. R. S. Sharma, Sudras in Ancient India, p. 180.

35. Ibid., p. 181.

36. IAR, 1962–63, pl. XI.

37. Devangana Desai, Erotic Sculpture of India, pp. 90 ff.

38. Ibid., p. 15., pl. XII.

39. Archaeological Survey of India, Epigraphia Indica, vol. X, no. 1137.

40. G. Yazdani in the Annals of the Bhandarkar Oriental Research Institute, Poona, 1941, pl. XII, XIII a

& c.
41. H. D. Sankalia, ‘The Nude Goddess or “Shameless Woman” in Western Asia, India and South-

Eastern Asia’, Artibus Asiae, XXIII, 1960.

42. Archaeological Survey of India, Epigraphia Indica, vol. XXIX, pp. 137ff.

43. W. Van Ingen, Figurines from Selcucia, University of Michigan, 1939, p. 27.

44. V. S. Agrawala in Ancient India, no.4, pp. 125ff, pls. XXXVIII, XXXIX and B. B. Lal in Ancient

India, nos. 10-11, pl. XLII.

45. Niharranjan Ray in The Age of Imperial Unity, p. 533.

46. R. S. Sharma, Light on Early Indian Society and Economy, Bombay: Manaktalas, 1966, p. 78.

47. R. S. Sharma, ‘Decay of Gangetic Towns in Gupta and Post-Gupta Times’, Indian History Congress,

33rd session, Muzaffarpur, 1972, p. 93.

48. R. S. Sharma, Indian Feudalism, Calcutta, 1965, pp. 4ff .

49. S. K. Srivastava’s article in Chavi, Bharat Kala Bhavan, Banaras, Golden Jubilee Volume.

50. Niharranjan Ray in The Classical Age, Bombay: Bharatiya Vidya Bhavan, p. 556.

51. Moti Chandra in the Bulletin of the Prince of Wales Museum, no.7, 1959–62.

52. R. S. Sharma, Light on Early Indian Society and Economy, p. 86.

53. R. S. Sharma in The Indian Historical Review, vol. I, 1974, p. 5.

54. R. S. Sharma, Indian Feudalism, pp. 127ff.

Chapter 6
1. Roland Barthes, ‘introduction to the Structural Analysis of Narratives’, in Image, Music, Text,

London: Fontana Press, 1977, p. 79.

2. The 1988 symposium on the date of the Buddha indicates that it is necessary to set aside the most

frequently quoted date of 483 bce. See H. Bechert (ed.), The Dating of the Historical Buddha, pt. 1,
Gottinger, 1991.
3. The exact number is undetermined although 550 is frequently cited.

4. The number of paramitas varies in different sources. The most usual number is six or ten.

5. Kurty Weitzmann, ‘Narration in Early Christendom’ in Narration in Ancient Art, Special Issue of

the American Journal of Archaeology, LXI, 1957, p. 91.

6. Richard Brilliant, Visual Narratives: Story-telling in Roman and Etruscan Art, Ithaca, 1984, p. 32.

7. See for instance ch. 5 of Wallace Martin, ‘Narrative Structure: A comparison of Methods’ in Recent

Theories of Narrative, Ithaca, 1986, pp. 107–29.

8. Gregory Schopen has eloquently and convincingly argued against the primacy of textual versus

archaeological material in the study of the Buddhist religion. See his ‘Archaeology and Protestant
Suppositions in the Study of Indian Buddhism’ in Koichi Shinohara and Gregory Schopen (eds.),
From Benaras to Beijing: Essays on Buddhism and Chinese Religion, Oakville: Mosaic Press, 1991, pp.
1–23.
9. For the uncertainty regarding the date of this text, see Gregory Schopen, ‘On Avoiding Ghosts and

Social Censure: Monastic Funeral in the Mulesavastivada Vinaya’, Journal of Indian Philosophy, 20,
1992, pp. 1–39, p. 36, n. 69.
10. See Angela F. Howard, ‘in Support of a New Chronology for the Kizil Murd Paintings’, Archives of

Asian Art, XLIV, 1991, pp. 68–83.

11. A. K. Coomaraswamy, ‘Picture Showmen’, Indian Historical Quarterly, V, 1929, pp. 82–187.

12. Ibid.

13. Eve Ray, ‘Documentation for Paithan Paintings’, Artibus Asiae, XI, 1978, pp. 239–82.
14. See Victor Mair, Painting and Performance, Chinese Picture Recitation and Its Indian Genesis,

Honolulu, 1988 for a survey of the tradition of picture-showmen.

15. F. L. Woodward and C. A. F. Rhys Davids (eds.), The Book of Kindred Sayings Samyutta Nikaya, pt.

3, Bristol: Pali Text Society, 1917–30, p. 128.

16. For details, see chs. 5 and 7 of Vidya Dehejia, Discourse in Early Buddhist Art: Visual Narratives of

India, Delhi: Munshiram Manoharlal, 1997.

17. E. B. Cowell (ed.), The Jataka or Stories of the Buddha’s Former Births, vol. 5, bk. 18, Jataka 528,

London: Luzac and Co., 1957, pp. 116–26.

18. Ibid., p. 119.

19. For a discussion on the nature of early patronage see Vidya Dehejia, ‘ The Collective and Popular

Basis of Early Buddhist Patronage’ in Barbara Stoler Miller (ed.), The Powers of Art: Patronage in
Indian Culture, Delhi: Oxford University Press, 1992, pp. 35–45. See also Gregory Schopen, ‘Monks,
Nuns, and ‘Vulgar’ Practices: The Introduction of the Image Cult into Indian Buddhism’, Artibus
Asiae, 49, 1988–89, pp.153–68.
20. See Dehejia, Discourses in Early Buddhist Art, Appendix 1.

21. While I would agree with Schopen (‘An Old Inscription from Amaravati and the Cult of the Local

Monastic Dead in Indian Buddhist Monasteries’, 313f) that monks at several sites supervised
construction, we shall see that inscriptional and thematic evidence from Bharhut and Sanchi Schopen
indicates the absence of a narrative programme.
22. Two more architraves at Sanchi are devoted to the narration of the Chaddanta Jataka; while one of

them repeats the monoscenic mode, though less imaginatively, the other uses the mode of continuous
narration to present four episodes of the story.
23. Meyer Schapiro, Words and Pictures: On the Literal and Symbolic in the Interpretation of a Text,

The Hague, 1973.

24. Otto Pacht, The Rise of Pictorial Narrative in Twelfth Century England, Oxford, 1962, p. 8.

25. Susan Huntington in, ‘Aniconism and the Multivalence of Emblems: Another Look’, Ars Orientalis,

22, 1992, pp. 111–56 prefers to identify this architrave as a later celebration in honour of horse
Kanthaka. In such an interpretation, there is no indexical sign of the Buddha’s presence, and the
parasol merely honours horse Kanthaka himself.
26. Gerard Genette, Narrative Discourse: An Essay in Method (Translated by Jane Levin), Ithaca:

University of Ithaca, 1980, p. 40.

27. Exceptions to this are generally restricted to relatively small spans of narrative such as those that

extend 3 to 4 feet in length.

28. Information from Dr Robert Knox of the British Museum.

29. Joanna Williams, ‘From the Fifth to the Twentieth Century and Back’, College Art Journal, 1990,

pp. 363–69.
30. In my preliminary article on the subject (‘On Modes of Narration in Early Buddhist Art’, The Art

Bulletin, 72, 3, 1990, pp. 374–91) I described this mode as linear; it seems to me, however, that the
term sequential is preferable. The mode has sometimes been referred to as cyclical narrative.
31. This is not an incomplete life cycle; rather it is the life–story as narrated in the Lalitavistara text

which commences in the Tushita heavens and ends with the first sermon.
32. I use the word synoptic in a wider sense than A. M. Snodgrass, Narration and Allusion in Archaic

Greek Art, London, 1982 (The Eleventh J. L. Myres Memorial Lecture), who excludes from his
synoptic category those scenes with a repetition of the protagonist.
33. Brilliant, Visual Narratives, p. 18.

34. Dieter Schlingloff, Studies in Ajanta Paintings, New Delhi, 1988, pp. 408–12.

35. See Dieter Schlingloff ’s plan of the Vessantara wall in Ajanta Handbuch der Wandmalerien, vols.

1–2, Wiesbaden: Harrassowitz, 2000

36. I had earlier thought that scenes in the front veranda, to the upper left of the main doorway, were

vignettes from the initial stages of the Vessantara story, and wondered if their placement in the
veranda was intended to draw worshippers into the cave to view the complete denouement of the tale
(V. Dehejia, ‘On Modes of Visual Narration in Early Buddhist Art’ and ‘Narrative Modes in Ajanta
Cave 17’, South Asian Studies, 7, pp. 45–57). However, Dieter Schlingloff has recently identified these
scenes as pertaining to the story of Kalodayin, one of the messengers sent by king Suddhodana to
request the Buddha to visit his home town, as contained in the Mulasravastivadin Vinaya (personal
communication). Schlingloff ’s Ajanta-Handbuch der Wandmalereien will contain identifications of all
the murals, together with line drawings, and notes on the textual sources drawn upon by sculptors.
37. John D. Smith, The Epic of Pabuji: A Study, Transcription and Translation, Cambridge: Cambridge

University Press, 1991, p. 57.

38. A pigeon king who is a minor participant in the main story narrates story I; the mouse relates

story 2 but is himself interrupted by one of story 2’s characters narrating story 3. Having found our
way back to the mouse, the story moves on but is put on hold by the mouse who tells story 4. The
main story inches along; the tortoise relates story 5; the mouse brings in story 6; and finally the main
legend comes to a conclusion.
39. Remarking that a particular situation had occurred previously, the Buddha repeatedly narrates

stories of previous lives. Thus at the Great Departure, the Buddha narrates the legend of the courtesan
Syama to illustrate that he had left Yashodhara on previous occasions (The Mahavastu, translated by J.
J. Jones, Sacred Books of the Buddhists, 3 vols., London: Luzac and Co., 1949–56, 161 ff.). When king
Suddhodhana refuses to believe that his son has died of excessive penance, the Buddha narrates the
Syama Jataka to illustrate that on a previous occasion too, Suddhodhana had refused to believe his son
dead (ibid., 199 ff.). To illustrate that he had previously fallen into the hands of Mara but had been
able to escape, the Buddha narrates a tortoise Jataka (ibid., 232ff.).
40. See E. B. Cowell, The Jataka or Stories of the Buddha’s Former Births. For instance, the Maha-

panada Jataka (number 264) is interrupted by the story of Surici which is also given independently as
Jataka 489. Similarly the Mittavinda Jataka (number 369) is in fact a fragment of Jataka 41. Other
examples exist.
41. See Wolfgang Iser, ‘The Reading Process: A Phenomenological Approach’, in Reader-Response

Criticism, ed. J. Tompkins, Baltimore, 1980, pp. 50–69. I have applied Iser’s comments on reader-
response to literature, to viewer-response to art.
42. I am grateful to Gregory Schopen for drawing my attention to J. Przyluski, ‘La Roue de la Vie a

Ajanta’, Journal Asiatique, 11e Serie, 16, 1920, p. 319.

43. Brilliant, Visual Narratives, p. 63.

44. See Dehejia, Discourse in Early Buddhist Art, ch. 10.

45. ibid.

46. This is suggested by a donation to a chaitya hall at Kuda by a brahmana woman (H. Luders, ‘A List

of Brahmi inscriptions from the Earliest Times to ad 400’, Epigraphia Indica, vol. X, 1912 and a Nasik
inscription in Vihara 3 of Queen Ealasri, mother of the ruler Gautamiputra Satakarni whose donatory
inscription speaks of her son as a peerless brahmana and as one who prevented the intermingling of
the four castes (V. V. Mirashi, The History of the Satavahanas and Western Kshatrapas, Bombay, 1981,
pp. 41–49).
47. I am indebted to V. Narayana Rao (verbal communication) for this observation.
Chapter 7
1. H. Cousens, The Chalukya Architecture of the Kanarese Districts, Calcutta: Government of India,

1926, p. 21.
2. Michael W. Meister and M. A. Dhaky (eds.), Encyclopaedia of Indian Temple Architecture, ‘South

India: Upper Dravidadesa’, New Delhi: Oxford University Press and American Institute of Indian
Studies, 1986, pp. 28–31.
3. Himanshu Prabha Ray (ed.), Sacred Landscapes in Asia: Shared Traditions, Multiple Histories, New

Delhi: IIC-Manohar, 2007.

4. James Fergusson, History of Indian and Eastern Architecture, vols. I and II, revised and edited by

James Burgess (Indian Architecture) and R. Phene Spiers (Eastern Architecture), London: J. Murray,
1910, pp. 39–40, 302.
5. Himanshu Prabha Ray and Sudha Ravindranath, ‘The Archaeology of Ritual Spaces: Satellite

Images and Early Chalukyan Temples’, Man and Environment (forthcoming).

6. Journal of the Bombay Branch of the Royal Asiatic Society, IV, 1853, p. 460.

7. Meadows Taylor and James Fergusson, Architecture in Dharwar and Mysore, Madras: Asian

Educational Services, 1989 (first published 1866).

8. James Burgess, Report of the First Season ’s Operations in the Belgam and Kaladgi districts, New

Delhi: Archaeological Survey of India, 1997, p. 15 [first published 1874].

9. Ibid., p. 27.

10. James Burgess, ‘The Dolmens at Konur and Aihole’, Indian Antiquary 3, 1874, p. 308.

11. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, p. 147.

12. Cousens, Chalukyan Architecture of the Kanarese Districts, p. 19.

13. Ibid., p. 23.

14. Ibid., p. 26.

15. Cousens, ,Chalukyan Architecture of the Kanarese Districts, p. 28.

16. K. V. Soundara Rajan, Early Temple Architecture in Karnataka and its Ramifications, Dharwar, 1969

and ‘Temples of Aihole and Pattadakal’, in The Chalukyas of Badami, ed. M. S. Nagaraja Rao,
Bangalore, 1978.
17. M. Rama Rao, Early Calukyan Temples of Andhra Desa, AP Government Archaeological Series, 20,

Hyderabad: Andhra Pradesh Government, 1965.

18. K. R. Srinivasan, Temples of South India, New Delhi: National Book Trust, 1971.

19. Gary M. Tartakov, The Durga Temple at Aihole: A Historiographical Study, New Delhi: Oxford

University Press, 1997, p. 6.

20. R.S. Gupte, The Art and Architecture of Aihole, Bombay: D. B. Taraporevala, 1967.

21. Aschwin Lippe, ‘Some Sculptural Motifs on Early Chalukyan Temples’, Artibus Asiae, XXIX, no. 1,

1967, pp. 5–24; Aschwin Lippe, ‘Early Chalukya Icons’, Artibus Asiae, XXXIV, no. 4, 1972, pp. 273–
330. Carol R. Bolon, Early Chalukyan Sculpture, Ph.D. Thesis, university of pennsylvania, 1981.
22. George Michell, Early Western Calukyan Temples, London: AARP, 1975; George Michell, ‘Temples

of the Early Chalukyas’, Marg, December 1978.

23. S. Rajasekhara, Early Chalukya Art at Aihole, New Delhi: Vikas Publishing House, 1985.

24. Adam Hardy, Indian Temple Architecture: Form and Transformation, New Delhi: IGNCA, 1995, p.

4.
25. The idea of the temple as a symbol of manifestation was put forward by Stella Kramrisch, The

Hindu Temple, Calcutta: University of Calcutta, 1946, pp. 165, 176.

26. Adam Hardy, Indian Temple Architecture: Form and Transformation, New Delhi: IGNCA, 1995, p.

4.
27. S. R. Rao and M. S. Nagaraja Rao (eds.), The Chalukyas of Badami (seminar papers). Bangalore:

The Mythic Society, 1978, pp. 272–75.

28. D. Kennet and J. V. P. Rao, ‘The Early Historic Brick Temples at Chandor’, South Asian Studies,

XVII, 2001, pp. 97–107.

29. A. Ghosh (ed.), Encyclopaedia of Indian Archaeology, vol. II, New Delhi: Munshiram Manoharlal,

1989, p. 456.
30. Indian Archaeology–A Review,, 1970–71, p. 29.
31. D. Kennet and J. V. P. Rao, ‘Two Early Historic Brick Temples at Paithan in Maharashtra’, South

Asian Studies, XIX, 2003, pp. 113–23.

32. Vasant Shinde, P. P. Joglekar, Ravi Jadhav and V. Naiknavare, ‘A Satavahana Period Shrine of

Laxmi-Parvati at Siddhapur, District Solapur, Maharashtra’, Man and Environment, XXIX, no. 2, 2004,
p. 117.
33. I. K. Sarma, The Development of Early Saiva Art and Architecture, New Delhi: Sundeep Prakashan,

1982, p. 18.
34. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, p. 22.

35. Ibid., p. 42.

36. Himanshu Prabha Ray, ‘Introduction: The Archaeology of Sacred Space’, in Archaeology as History

in Early South Asia, eds. Himanshu Prabha Ray and Carla Sinopoli, New Delhi: Indian Council for
Historical Research and Aryan Books International, 2004, pp. 350–75.
37. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, p. 64.

38. J. F. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, V, 1876, no. XIII, pp. 67–76.

39. S. Rajasekhara, Early Chalukya Art at Aihole, New Delhi: Vikas Publishing House, 1985, p. 119.

40. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, p. 22.

41. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, p. 92.

42. S. Rajasekhara, Early Chalukya Art at Aihole, p. 109.

43. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, pp. 49–52.

44. ‘De-urbanisation was therefore a feature of the first or classic phase of feudalism marked by a

subject peasantry and a dominant class of landlords in conditions of languishing trade and
predominantly agrarian economy. It inaugurated an era of closed economy in which the needs of
landed intermediaries were met locally without the effective intervention of traders whose functions
were reduced to the minimum.’ R. S. Sharma, Urban Decay in India, New Delhi: Munshiram
Manoharlal, 1987, p. 184.
45. R. N. Nandi, Social Roots of Religion in Ancient India, Calcutta: K. P. Bagchi and Co., 1986.

46. ‘Several temples were built outside the main Gupta-Vakataka domain, e.g., in Gujarat, Bengal,

Assam and Punjab. Building of temples away from the main centres and land grants to brahmanas in
uncleared territories led to dissemination of knowledge of agriculture, calendar and technology and
boosted agrarian expansion. It accelerated the Sanskritisation process in tribal areas and remote
villages. The Ramayana and Mahabharata appeared for the first time in temple art at Nachna,
Deogadh, Gaddhwa, Paunar and other sites’, D. Desai, ‘Social Dimensions of Art in Early India’,
Proceedings of the Indian History Congress, Gorakhpur, 1989–90, p. 31.
47. R. S. Sharma, Indian Feudalism c. AD 300–1200, New Delhi: Macmillan, 1965.

48. ‘The rapid growth in the number and networks of temple centres, whose origins certainly date to

pre-Gupta times, become understandable when we begin to appreciate how closely they were linked,
as were gifts and land grants to Brahmins (brahmadeyas and agraharas) with the formation of
subregional and regional kingdoms and their legitimation, consolidation of their resource bases, and
the forging of linkages for social integration across communities.’ B. D. Chattopadyaya,
‘Historiography, History and Religious Centers’, Gods, Guardians and Lovers, eds. Vishakha N. Desai
and Darielle Mason, New York, Ahmedabad: The Asia Society Galleries in association with Mapin,
1993, p. 42.
49. Vijay Nath, Puranas and Acculturation—A Historico-Anthropological Perspective, New Delhi:

Munshiram Manoharlal, 2001.

50. This ruling house is distinct from the later Chalukyas at Vengi and Chalukyas of Kalyana (tenth-

twelfth centuries CE).

51. R. S. Panchamukhi, ‘Badami Inscription of ChalikyaVallabhesvara: Saka 465’, Epigraphia Indica

XXVII, 1947–48, pp. 4–9.

52. K. V. Ramesh, Chalukyas of Vatapi, Delhi: Agam Kala Prakashan, 1984, p. 41.

53. J. F. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, XIX, 1890, pp. 7–20.

54. Tartakov, The Durga Temple at Aihole, p. 103

55. Cousens, The Chalukyan Architecture of the Kanarese Districts, p. 22.

56. J. F. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, VI, 1877, pp. 72–75.

57. J. F. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, XIX, 1890, p. 10.

58. J. F. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, X, 1881, p. 167.

59. J. F. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, VIII, 1879, pp. 237–45.

60. Ibid., pp. 284–5.

61. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, p. 78.

62. J. F. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, 1881, pp. 162–165, nos. XCIX and C.

63. Ibid., p. 166, nos. CII and CIII.

64. Ibid., pp. 166–67, no. CIV

65. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, 1879, p. 245.

66. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, VIII, 1879, pp. 287–88.

67. Meera Abraham, Two Medieval Merchant Guilds of South India, New Delhi: Manohar, 1988, p. 43.

68. Ibid., p. 44.

69. V. V. Mirashi, ‘The Inscriptions of the Silaharas’, Corpus Inscriptionum Indicarum, 6, New Delhi:

Archaeological Survey of India, 1977, p. 185.

70. M. Mitchiner, The Coinage and History of South India, Part 1 Karnataka-Andhra, Sanderstead:

Hawkins Publications, 1998.

71. Gary M. Tartakov, ‘The Beginning of Dravidian Temple Architecture in Stone’, Artibus Asiae, XLII,

1980, p. 70.
72. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, X, 1881, p. 62, no. LXXXV.

73. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, X, 1881, p. 62, no. LXXXVI.

74. Carol Radcliffe Bolon, ‘Reconstructing Galaganatha’ in Kusumanjali, ed. M. S. Nagaraja Rao,

Delhi: Agam Kala Prakashan, 1987, pp. 229–32.

75. Meister and Dhaky, Encyclopaedia of Indian Temple Architecture, p. 78.

76. Carol Radcliffe Bolon, ‘Two Chalukya Queens and their commemorative temples’in Royal Patrons

and Great Temple Art, ed. V. Dehejia, Mumbai: Marg Publications, 1988, p. 62.
77. Gary M. Tartakov, ‘Interpreting the Inscriptions of Mahakuta Art Historically’ in Indian

Epigraphy: Its Bearing on the History of Art, eds. Frederick M. Asher and G. S. Gai, New Delhi: Oxford
& IBH, 1985, pp. 139–40.
78. Fleet, ‘Sanskrit and Old Canarese Inscriptions’, Indian Antiquary, X, 1881, pp. 102–04.

79. Cousens, The Chalukyan Architecture of the Kanarese Districts, p. 36.

80. Lippe, ‘Early Chalukya Icons’, p. 274.

81. Ibid., p. 276.

Chapter 8
1. P. Mitter, Much Maligned Monsters: History of European Reactions to Indian Art, Oxford: Clarendon

Press, 1977.
2. V. Dehejia and P. Pal (eds.), From Merchants to Emperors: British Artists and India, 1757–1930,

Ithaca: Cornell University Press, 1986.

3. K. V. Soundara Rajan, Cave Temples of the Deccan, Archaeological Survey of India, 1981.

4. P. Chandra, On the Study of Indian Art, Cambridge: Harvard University Press, 1983, p. 69.

5. Desai points out that the Jaina and the Hindu temples of Khajuraho have a cognate architectural

style and have sculptures of celestial beings and decorative designs that are common to both. D. Desai,
The Religious Imagery of Khajuraho, Mumbai: Franco-Indian Research Pvt. Ltd., 1996, p. 13.
6. K. Deva, Temples of India, Delhi: Aryan Books International, 1995.

7. K. V. S. Rajan, Cave Temples of the Deccan, Archeological Survey of India, 1981.

8. J. G. Williams, The Art of Gupta India, New Delhi: Heritage Publishers 1983, p. 3.

9. Chandra points out that in Karnataka, where there is plenty of inscriptional evidence of hundreds

of temples erected in the Hoysala kingdom during the reign of the Hoysalas, not more than half a
dozen owe their origin to the ruling monarchs. P. Chandra, On The Study of Indian Art, p. 71.
10. M. Meister, ‘Ethnography, Art History and the Life of Temples’, in Ethnography and Personhood—

Notes from the Field, ed. M. Meister, Jaipur: Rawat Publications, 2000.
11. J. H. van Linschoten, who came to India in 1583, comments on Salsette’s frightening impression:

‘By the town of Bassaym . . . there Iyeth an Island called Salsette. There are two of the most reknowned
Pagodes, . . . the images there in cut out of the [very] rockes of the same hill, with the most fearfull
[formes and] shapes . . . all the chambers . . . are all full of carved Pagodes, of so fearefull, horrible &
devilish formes [and shapes] that it is [an abomination to see].’ A. C. Burnell, The Voyage of J. H. van
Linschoten, London, 1885, p. 289.
12. The Italian Gemelli Careri (1700) was the first traveller to make a serious study of Kanheri’s

dimensions. Similarly, Thevenot, in his limited hours, gathered approximate measurements of the
main building and sculptures at Ellora. He also managed to supply a detailed account of the general
design of Kailasa temple, by analysing each of its three main architectural parts. S. N. Sen, Indian
Travels of Thevenot and Careri, New Delhi, 1949, p. 171. Dr G. E. Gemelli Careri, Giro del Mondo, III,
Naples, 1700.
13. Colebrooke’s articles on Hindu philosophy in Transactions of the Royal Asiatic Society I, 1827 and

Transactions of the Royal Asiastic Society II, 1830.

14. Gasparo Balbi, Viaggio dell, Venice: Indie-Orientali, 1590.

15. Joao do Castro took the Shiva Ardhanarisvara image with a single breast to be the Amazon of

classical mythology and the identification remained firmly embedded in eighteenth century literature
for a long time. Dom Joao de Castro, Primeiro Roteiro da Costa da India desde Goa ate Dio. (Segundo
MS. Autographo, ed. Diogo Kopke, Porto, 1843, pp. 65–81).
16. J. Burgess and J. Fergusson, Cave Temples of India, Delhi: Oriental Book Reprint Corporation,

1880, p. 55.
17. Ibid., p. 368.

18. Ibid., p. 343.

19. According to Burgess, the Jaina caves except those near Kailasha were not remarkable for beauty

or magnificence.
20. Ibid., p. 21.

21. Ibid, Preface, p. xiii.

22. By 1873, Burgess had compiled an extensive inventory of principal monuments in Bombay, Sind,

Berar, Central Provinces and Hyderabad, a monograph on Elephanta, and three portfolios of
photography dealing with the Satrunjaya temples, the monuments of Somnath, Junagadh and Girnar
and the ancient architecture of Gujarat and Ahmedabad. Ancient India, vol. 9, Bulletin of the
Archaeological Survey of India, 1953, p. 21.
23. The tremendous amount of documentation undertaken by Burgess is evident in his list of

monographs: Report of the First Season's Operations in Belgaon and Kaladgi Districts (1874); Report on
the Antiquities of Kathiawad and Kachh, being the Result of the Second Season’s Operations of the
Archeaological Survey of Western India 1874–75 (1876); Report on the Antiquities in the Bidar and
Aurangabad Districts, in the Territories of His Highness the Nizam of Hyderabad, being the Result of the
Third season’s Operations of the Archaeological Survey of Western India 1875–76 (1878); Report on the
Buddhist Cave Temples and Their Inscriptions, being part of the Result of the Fourth, Fifth and Sixth
seasons Operations of the Archaeological Survey of Western India 1876–77, 1877–78, 1878–1879 (1883);
On the Muhammedan Architecture of Broach, Cambay, Dholka, Champaner and Mahmudabad in
Gujarat (1896); Memorandum on the Antiquities at Dabhoi, Ahmedabad, Than, Junagadh, Girnar and
Dhank (1875); Memorandum on the Remains at Gumli Gop and in Kachh, etc. (1875); Provisional Lists
of Architectural and Other Archaeological Remains in Western India, including the Bombay Presidency,
Sindh, Berar, central Provinces and Hyderabad (1875); Notes on the Antiquities of the Talukas of Parner,
Sanganner, Ankole and Kopargaum, with Revised Lists of Remains in the Ahmadnagar, Nasik, Puna,
Thana, and Kaladgi Zilas (1877); Notes on the Bauddha Rock-Temples of Ajanta, their Paintings and
Sculptures, and of the Paintings of the Ragh Caves, Modern Buddha Mythology, etc. (1879); The Cave
Temples of India (1888, co-author James Fergusson); Inscriptions from tile Cave Temples of Western
India, with Descriptive Notes, etc. (1881, co-author Bhagwanlal Indraji); Lists of the Antiquarian
Remains in the Bombay Presidency, with an Appendix of Inscriptions from Gujarat, Compiled from
Information Supplied by the Revenue, Educational and Other Government Offices (1885), and The
Architectural Antiquities of Northern Gujarat, More Especially of the Districts Included in the Baroda
State (1903, co-author H. Cousens).
24. According to Burgess, ‘the new trend in Europe was to go by architectural features, and it was

allowed to supersede all evidence in ascertaining the age of medieval/classical building . . .’, Burgess
and Fergusson, Cave Temples of India, p. 182.
25. Ibid., p. 184.

26. Ibid. p. 182.

27. Ibid., p. 166.

28. Ibid., p. 483.

29. Burgess traces the evolution of the Buddhist caitya façade from Bhaja, Karle, Ajanta to Ellora.

Ellora Cave 10, i.e., the Visvakarma caitya is pointed by Burgess as showing the development of the
evolution of the caitya from the wooden fronted ones at Bhaja, to the stone fronted ones at Karle, to
the elaborate façade at Ajanta. In Ellora the caitya loses all these external features in its exterior as the
great horseshoe-shaped window of earlier times is contracted into the triple Venetian window. J.
Burgess, Report on the Elura Caves and Brahmanical and Jaina Caves Temples in Western India,
Archaeological Survey of Western India 5, reprint, Delhi, 1970, Ch. 3, p. 9.
30. Burgess and Fergusson, Cave Temples of India, p. 450.

31. J. Burgess, ‘Sketch of Archaeological Research in India During Half a Century’, Journal of the

Bombay Branch of the Royal Asiatic Society, 1905, p. 133.

32. J. Burgess and H. Cousens, Architectural Antiquities of North Gujarat, London, 1903, p. vi.

33. With Havell’s books titled, Indian Sculpture and Painting (1908) and The Ideals of Indian Art

(1911), the New Orientalist discourse steered clear of the Eurocentric point of view that played up the
factor of Hellenic influence and dismissed all Hindu iconography as barbaric.
34. According to Deepak Kannal, Burgess’s study of Ellora represented a pioneering work whose

identification of reliefs seldom goes wrong or the dates prove absurd. He states that if factual data were
criteria of judgment, even a century later we have not added, anything very significant to this report
barring few amendments. D. H. Kannal, Ellora: All Enigma in Structural Styles, Delhi: Books and
Books Publishers and Distributors, 1996, p. 9.
35. W. Spink, ‘Ellora’s Earliest Phase’, Bulletin of the American Academy of Banaras, I, 1975, pp. 11–22.

36. K. V. Soundara Rajan, Keynote Address in R. Parimoo, D. Kannal and S. Panikkar (eds.), ElIora

Caves: Sculpture and Architecture, Delhi: Books and Books Publishers and Distributors, 1988, p. 37.
37. Ibid., p. 33.

38. For examining Ellora, Malandra turned to the Ajanta inscription between caves 26 and 27. Instead

of focussing on the style of the inscriptions, i.e., its triangular headmarks, she deciphered important
names which helped compare with records found further east in Vidarbha. This was unlike the
connections made by historians with north, west and south. In this way it was possible to question
tradition regional distinctions regarding Ellora and Ajanta. G. H. Malandra, ‘Ajanta to Ellora: Back
Again’ in The Art of Ajanta—New Perspectives, eds. R. Parimoo, D. Kannal, S. Panikkar, J. Poduval and
I. Sharma, New Delhi: Books and Books Publishers, 1991, pp. 440–50.
39. Burgess points that at Ellora, unlike at Kanheri and Ajanta, devis, and saktis figure more

frequently. At both Ajanta and Kanheri the Buddha is represented without any supporters or only
with two; at Ellora he is represented with six, eight, or ten of the Bodhisattvas, whilst devis figure
much more prominently here. He links this kind of representation to the mythology of the Yogacharya
school of Buddhism. J. Burgess, Report on the Elura Caves and the Brahmanical and Jaina Caves
Temples in Western India, p. 3.
40. M. Soar, ‘The Tirtha at Ellora’ in Ellora Caves: Sculpture and Architecture, eds. R. Parimoo, D.

Kannal and S. Panikkar, Delhi: Books and Books Publishers and Distributors, pp. 82–84.
41. Cort points out that a community’s religious associations might be in terms of allegiance to varied

deities or sacred centres of worship. The members of a caste could express allegiance to a mythic
kuladevi (associated with their home village) and to a different, goddess for marriage. J. E. Cort,
‘Communities, Temples, Identities: Art Histories and Social Histories in western India’ in Ethnography
and Personhood-Notes from the Field, ed. Michael Meister, Jaipur: Rawat Publications, 2000, ch. 5.
42. In his description of Brahmanical Cave 27 at Ellora, Burgess states ‘. . . the arrangement of

separating the hall from the verandah by a wall pierced by three doors and two windows is more in
keeping with the plan of a Bauddha cave than any of the other Brahmanical caves at Ellora’. Burgess
and Fergusson, Cave Temples of India, p. 41.
43. T. V. Pathy, ‘Architectural Interactions Amongst the Buddhist, Hindu and Jain Cave Temples at

Ellora’ in Ellora Caves: Sculpture and Architecture, eds. R. Parimoo, D. Kannal and S. Panikkar, Delhi:
Books and Books Publishers and Distributors, p. 364.
44. K. V. Soundara Rajan suggests that the imagery of Shiva as Lakulisa, where he sits on a lotus with

two rows of adhopadma and urdhuapadma petals, the stalk of which is supported by a pair of nagins,
is usually depicted in Buddhist representations owing to Buddha’s relationship with nagas. The stance
of Lakulisa in the yogasana pose also corresponds to that of Buddha. K. V. Soundara Rajan, Cave
Temples of the Deccan, p. 119.

Ratan Parimoo also points to the similarity between the yogesvara Shiva
in the Kailasha mandapa and that of the meditating Buddha in Cave 12. R.
Parimoo, ‘Some Problems of Ellora from the Point of View of Buddhist
Caves’ in Ellora Caves: Sculpture and Architecture, eds. R. Parimoo, D.
Kannal and S. Panikkar, 1988, ch. 9.
45. Deepak H. Kannal points out that though the vocabulary of the Kailasha is south Indian, space is

treated in a Deccan fashion with its temple on a high plinth like most north Indian structures. Its
inner space of mandapa and garbhagriha also differ from the Kailasanatha and Virupaksha temples
that it is frequently compared with. The Calukyan and Pallava idioms are also evident in both Kailasa
and in structures prior to Kailasa. D. H. Kannal, ‘The Regional Lineages and Possible Masters at
Kailasanatha Temple of Ellora’ in Ellora Caves: Sculpture and Architecture, eds. R. Parimoo, D. Kannal
and S. Panikkar, 1988, p. 260.
46. G. Tartakov, ‘The Pratima Pratishthapanam in the Kailasa', Kusumanjali–New Interpretation of

Indian Art and Culture, ed. M. S. Nagaraja Rao, Delhi: Agam Kala Prakashan, 1987.
47. Amongst these individual donors are a Chakresvara of 1235 CE, who is noted under the
Parsavanath figure at Jagannath sabha cave. In the Jain Indra Sabha Cave an inscription under the
Santinatha figure refers to Sohila, a brahmacarin as its patron. Another statue in the cave is referred to
as made by Sri Nagavarma. J. Burgess, Report on the Elura Caves and the Brahmanical and Jains Caves
Temples in Western India, pp. 50–51.
48. M. N. Tiwari and K. Giri, ‘Images of Bahubali in Ellora’ in Ellora Caves: Sculpture and Architecture,

eds. R. Parimoo, D. Kannal and S. Panikkar, Delhi: Books and Books Publishers and Distributors,
1988, pp. 335–39.
49. It is evident from the Suttanipata Gatha of the fifth century that this route was very popular. The

disciples of the Buddhist monk Bavari, reaching Asaka, situated in mid-Godavri valley, proceeded to
Pratisthana and from there to Mahishmati and Ujjayini, and reached Vidisa. M. Chandra, Trade and
Trade Routes in Ancient India, New Delhi: Abhinav Publications, 1977, p. 24.
50. Malandra, 'Ajanta to Ellora: Back Again’.

51. G. Tartakov and V. Dehejia, ‘Sharing, Intrusion and lnfluence’, Artibus Asiae, 45, 1984, pp. 267–

345.
52. The three new terms suggested are: (1) Sharing: When two works draw upon a common or shared

artistic tradition. (2) Influence: When one cultural form changes other cultural forms. (3) Intrusion:
When something not previously present enters into local patterns, through local intervention, or
exogenous borrowing. Ibid.
53. D. Desai, The Religious Imagery of Khajuraho.

54. P. V. Ranade, ‘Echoes of Ellora in Early Marathi literature’ in Ellora Caves: Sculpture and

Architecture, eds. R. Parimoo, D. Kannal and S. Panikkar, Delhi: Books and Books Publishers and
Distributors, ch. 4.
55. Ranade refers to the contemporary folk tale that circulates amongst villagers living around Ajanta

and Ellora. The genesis of the monuments is described as follows: Long, long ago on a full moon
night, devis, devanganas, apsaras, gandharvas and kinnaras visited Sahyadri Mountain. Some of these
who came to Ajanta and Ellora were so enchanted by the scenic beauty that they played and danced
there the whole night. They forgot Indra’s instruction to be home before sunrise. When day broke they
were still playing on the cliffs and ridges of Ajanta and Ellora. Indra’s curse and divine folly resulted in
the transformation of these celestials into stone images. Ibid., p. 108.
56. Ibid., p. 108.

57. Ibid., p. 119.

58. These were the western Indian temples of Sacciya Mata and Mahavir in Osian, Dadhimati Mata in

Goth-Manglod, and Ranchorray in Khed, J. E. Cort, ‘Patronage, Authority, Proprietary Rights, and
History’ in Ethmography and Personhood—Notes from the Field, ed. M. Meister, Jaipur: Rawat
Publications, 2000, p. 166.
Chapter 9
1. Part 3, 46, 1–19; page references are to Venkatesvara Press edition (Bombay, 1923).

2. Part 3, 85, 65–79.

3. Part 3, 85, 63–64.

4. E. Washburn Hopkins, Great Epic of India, New York: C. Scribner’s, 1901, pp. 58–84.

5. F. E. Pargiter, Dynasties of the Kali Age, Oxford, 1913, Introduction, pp. XIII, XVI.

6. Jātaka, vol. 6, pp. 159, 221.

7. Vinaya Piṭaka, vol. 4, Pāchittiya, no. 41, p. 289.

8. Mahā-Umagga-Jātaka, p. 212.

9. Sārattha-Pakāsinῑ, Siamese edition, part 2, p. 398, Cf. B. M. Barua’s, History of Pre-Buddhistic Indian

Philosophy, p. 110; ‘Maskari-Gósala’s Early Life’, Calcutta Review, June 1927, pp. 364–66.
10. Benares ed., pp. 32, 44.

11. Mūdrārāklaṣa, Act 1, Cf. Harṣa-Carita, Nirṇaya Sāgar: Press Edition, V, p. 153, quoted and

discussed by Barua in the Calcutta Review, June 1927, pp. 370–71.

12. ŚRa 46, 143–45.

13. VDhP 1.129, 1–19.

14. Citralakṣaṇa, translated by Berthold Laufer, Dokumente der indischen Kunst, Leipzig: Otto

Harrassowitz, 1913, pp. 129–36.

15. Cf. A. K. Maitra, ‘Aims and Methods of Painting in Ancient India’, Rupam, 1923.

16. Translated by S. Subbarao (Madras, 1917), p. 48.

17. Edited by Sarat Chandra Das (Calcutta, 1901); cf. Laufer, Dokumente, p. 186.

18. ŚRa 4.46. 145–46, Greek tradition, agreeable only to an appreciation of the naturalistic aspect of

art, has it that painting began with the outlining of a man’s shadow (Pliny the Elder, Natural History,
35–45).
19. Journal of the Royal Asiatic Society, vol. 7 (1875). The BṛS (Translated by H. Kern), pp. 93–97.

(2) By Jupiter being in its power will he born (the personage

denominated)Hāṁsa by Saturn, the man Śaśa; by Mars, the Ruchaka; by
Mercury, the Bhadra; and by Venus, the Mālavya.
(7) The length and stretch of the Haṁsa is of 96 digits. The personages
going by the names of Śaśa, Ruchaka, Bhadra and Mālavya, are each taller
than the preceding one by three digits.
(10) The Mālavya will be marked by arms resembling an elephant’s trunk,
and by hands reaching to the knees. His members and joints are fleshy, he
has a well-proportioned and neat frame, and a slender waist. His face, of
oblong form, measures 13 digits, the transverse measure between the ears
being three digits less. He has fiery eyes, comely cheeks, even and white
teeth and not too thick lips.
(11) Having by his valour obtained wealth, he will, residing in the recesses
of Mount Pāriyātra, reign as a wise king over Mālava, Broach, Surāṣṭra, Lāṭa,
Sindh, and so forth.
(12) This Mālavya will at the age of seventy years piously depart from life
at a place of pilgrimage. Having in due form indicated the characteristics of
this man, I now proceed to mention those of the others.
(13) The man Bhadra is marked by having the arms thick, equal, round
and long, his length is equal to the stretch of his arms from one side to the
other; his cheeks are covered with soft, small and dense hairs.
In his constitution, skin and sperm are predominant; his breast is broad
and thick; his prevailing quality is goodness. He has a tiger-like face, is
steadfast, forbearing, virtuous, grateful; he has the pace of an elephant, and
knows many sciences.
(15) He is sagacious, handsome, clever in the arts, constant, and adept in
ascetic philosophy; has the forehead and temples well-shaped; the loins
likewise, the hands and feet lined like the lotus calix, the nose fine, the
eyebrows even and well-knit.
(16) His person smells like earth when moist from fresh rain, or cassia-
leaf, saffron, frontal juice of elephants, agallochum. The hair of his head is
black, curled, and such that each single hair has its own pore.
(18) Should his length come to 84 digits and his weight to one bhāra, then
he will be lord over the Middle country; but if he have the full measure
implied in the words ‘taller by three digits’ he will he emperor of the whole
country.
(19) After dutifully ruling the country he acquired by his bravery, the
Bhadra at eighty years of age, will depart from life at a place of pilgrimage
and go to heaven.
(20) The Śaśa will have somewhat projecting, otherwise fine teeth, fine
nails, blubber eyes; a swift pace; he takes delight in science, mining and
trade; has full checks, is false, a good general; fond of love’s sport and partial
to other men’s wives: restless, valorous, obedient to his mother, and attached
to woods, hills, rivers and wildernesses.
(21) The same Śaśa is suspicious, and a keen observer of another’s weak
points, He is 92 digits in length, and, not being very heavy, has a soft step.
The chief constituent of his body is marrow.
(22) His waist is slender.
(23) This Śaśa will be a border chieftain or provincial governor . . . . He
will, seventy years old, reach Yama’s home.
(24) The marks of the Yama’s are: the mouth red, the face gold-coloured,
and, showing thick cheeks and an elevated nose; the head round; the eyes
honey-like, the nails wholly red—etc.
(25) He delights in water . . . . His length according to the statement of the
Sages will be 96 digits.
(26) The Haṁsa will possess the country of Khasa, Sūrasena, Gāndhāra,
and the land between the Ganges and Yamunā; after exrecising the royal
power for 90 years, he will meet death within a wood.
(27) The worthy Ruchaka by name will have good eyebrows and hairs; a
red tinged dusky colour, a neck marked with three folds like a shell; an
oblong face. He is brave, cruel, an egregrious counselor, a chief of robbers,
and a practised soldier.
 (28) The measure of Ruchaka’s face, in length, being taken four times,
gives the measure of his middle. His skin is thin.
(29) . . . His length is a hundred digits.
(30) He is an adept in charms and spells, and has thin knees and legs.
When this Ruchaka has reigned as king over the Vindhya, Sahyagiri and
Ujjain, he will on reaching seventy years, find his death by sword or fire.
(31) There are five other men, who will be the attendants of the fore-
mentioned monarchs.
(105) The number of digits which make the measure of men’s height is, for
the tallest, 108; for those of middle height, 96; for the shortest, 84.
20. Cf. ŚRa 64.60–110.

21. Memoirs of the Archacological Survey of India 5(1): 22–25.

22. A photo taken before restoration of one of the female figures painted on the rock of Sigiriya shows

the various manners of shading.

23. Ms. 12ct., Mysore library, B. M. Barua draws our attention to an important passage in the

Saṁyutta-Nikāya, part 3, p. 152, where the Buddha has incidentally mentioned the following
vegetable substance used by the dyers or painters for preparing dyes or colours: (I) resin (rajanā),
lac(lākhā), turmeric (haliddā), indigo (nῑlῑ), and madder (ma jeṭṭhi).
24. ŚRa 46.119–30.

25. Cf. K. P. Jayaswāl, ‘A Hindu Text on Painting’, Modern Review, 33: 734.

26. For the preparation of vajralepa, see BṛS 57, Abhilaśitārtha Cintāmani 86 ff., and ŚRa 131–33.

27. For the preparation of the wall underneath the vajralepa cover. Cf. VDhP 3.40.1ff. and ŚRa 41–42.

28. Abhilaśitārtha Cintāmani, p. 60.

29. VDhP 3–27. Cf. Nāṭyaśāstra.

30. VDhP 3–42.

31. VDhP 3.62.1–2.

32. ŚRa 145–47.

33. Nāṭyaśāstra 6.42–43.

34. VDhP 52.10–12.

35. The Viṣṇudharmottara declares the rules for painting as valid also for sculpture, which either may

be hollow or massive (chapter 43); worth noticing as an example of hollow sculpture is the use of skin,
coated with clay and painted over. Hollow figures must have stood among other places also on the
stage, where images of gods, demons, yakṣas, elephants, horses, deer, and birds were to be made of
clay, wool, cloth, leather, or iron (27, 3).

The commentary to the Kāmasūtra by Yaśodhara, when speaking about

the Ṣaḍaṇga, the six limbs of painting, i.e. rūpa-bheda, pramānam, bhāva,
lāvaṇya-yojanam, sādṛśya, and varṇika-bhaṅga, refers to the categories of
types, proportion, rendering, of mood (rasa), embodiment of grace, point of
view with reference to sthanam, and to the preparation of colours . . .
Chapter 10
1. Margaret Cone and Richard F. Gombrich, Perfect Generosity of Prince Vessantara, Oxford, 1977, p.

xv; The Jataka, ed. E. B. Cowell, London, 1957 (henceforth Jataka), p. viii.
2. Jataka, I, p. ix.

3. Ibid.

4. Sukumar Sen, Origin and Development of the Ramayana Legend, Calcutta, 1977, p. 16.

5. Jataka, I, p. X.

6. Uma Chakravarti, Social Dimensions of Early Buddhism, Delhi, 1987, p. 98.

7. Uma Chakravarti, ‘Renouncer and Householder in Early Buddhism’, Social Analysis, 13, 1983, pp.

70–83.
8. Natalie Davis, Society and Culture in Early Modern France, Stanford, 1975, p. 122.

9. Jataka, I, p. ix.

10. Peter Burke, Popular Culture in Early Modern Europe, London, 1978, p. 70.

11. K. Meenakshi, ‘Old Indo-Aryan to Middle Indo-Aryan’ in Aryan’ in Situating Indian History, eds.

S. Bhattacharya and R. Thapar, Delhi, 1986, p. 346.

12. Jataka, I, p. viii, n. 4.

13. Ibid., p. x.

14. M. L. Feer, Study of the Jatakas, Calcutta, 1963, p. 19.

15. Ibid.

16. Ibid., p. 33; A. L. Basham, Wonder That was India, Calcutta, 1963, pp. 274–75.

17. Feer, Study of the Jatakas, p. 42.

18. Stuart Blackburn and A. K. Ramanujan (eds.), Another Harmony, Delhi, 1986, p. 11.

19. Roger Chartier, ‘Intellectual History or Socio-cultural History? The French Trajectories’ in Modern

European Intellectual History, eds. D. La Capra and S. L. Kaplan, New York, 1982, pp. 34–37.
20. Michel de Certeau, quoted in ibid., p. 37.

21. Carlo Ginzburg, The Cheese and the Worms, London, 1980.

22. Feer, Study of the Jatakas, p. 27.

23. Brenda E. F. Beck, ‘Social Dyads in Indic Folktales’ in Blackburn and Ramanujan, Another

Harmony, pp. 81–82.

24. Ibid., p. 83.

25. Ibid., pp. 86–98.

26. Jataka, I, p. viii.

27. Chakravarti, Social Dimensions of Early Buddhism, p. 26.

28. Jataka, p. 42.

29. Ibid., I, p. 87.

30. Ibid., I, p. 302.

31. Ibid.

32. Ibid., II, pp. 26–28.

33. Ibid.

34. Ibid., II, pp. 4–7.

35. Ibid., II, pp. 7–9.

36. Ibid., II, pp. 199–201.

37. Ibid., II, p. 13.

38. Dialogues of the Buddha, III, translated as Dialogues of the Buddha by T. W. Rhys Davids, London,

1977, p. 211.
39. Jataka, I, pp. 773–75.

40. Ibid., I, pp. 284, 158–60.

41. Ibid., I, p. 189.

42. Ibid., I, pp. 309, 158–60; II, pp. 117–18.

43. Ibid., II, p. 81; I, p. 264.

44. Uma Chakravarti, ‘Conceptualizing Brahmanical Patriarchy in Early India: Gender, Caste, Class

and State’, Economic and Political Weekly, vol. XXVIII, no. 14, 3 April 1993.
45. Ananda K. Coomaraswamy, Buddha and the Gospel of Buddhism, New York, 1916, p. 154.

46. Jataka, II, p. 81.

47. Ibid., I, p. 284.

48. Ibid., II, p. 193.

49. Ibid., I, p. 264.

50. Ibid., I, p. 309.

51. Ibid., I, p. 251; III, p. 267.

52. Ibid., III, p. 267.

53. Ibid., III, p. 267; I, p. 312.

54. Ibid., II, p. 63, 292.

55. Ibid., II, p. 286.

56. Ibid., I, p. 98.

57. Ibid., I, p. 286.

58. Chakravarti, Social Dimensions of Early Buddhism, p. 180.

59. Jataka, I, p. 225.

60. Ibid., I, p. 272.

61. Ibid., I, p. 253.

62. Ibid., II, pp. 291–92.

63. Dilip Menon, ‘The Moral Community of the Teyyattam: popular culture in late colonial Malabar’,

Studies in History, 9, 2, 1993, pp. 187–219.

64. Jataka, I, p. 313.

65. Ibid., III, pp. 266–71.

66. Hans Medick, discussion at a seminar in the Department of History, University of Delhi,

November 1987.
67. Jataka, I, p. 156.

68. Ibid., I, p. 177.

69. Ibid., II, p. 4.

70. Ibid., I, pp. 275–77.

71. Ibid., I, pp. 280–81.

72. Ibid., I, pp. 98–99.

73. Chakravarti, Social Dimensions of Early Buddhism, pp. 98–100.

74. Jataka, I, pp. 101–02, 109–13.

75. Ibid., I, pp. 108–09

76. Ibid., III, pp. 154–55.

77. Therigatha, translated by C. A. F. Rhys Davids as Psalms of the Sisters, London, 1945, pp. 116–19.

78. I use common sense here not in the Gramscian sense but in its lay meaning. Punna’s ability to cut

through the layers of knowledge is an indication of her resistance to the world vision of the brahmana,
preventing it from eventually becoming embodied as common sense in the Gramscian sense. See
Joseph V. Femia, Gramsci's Political Thought: Hegemony, Consciousness and the Revolutionary Process,
Oxford, 1987, p. 132.
Chapter 11
1. A. K. Warder dates the Nāṭyaśāstra (henceforth NS) to circa second century ce and the

kāvyālamkāra to the fourth or fifth century ce. Indian Kāvya Literature, vol. I.: Literary Criticism,
Delhi: Motilal Banarsidass, 1989, pp. 21, 82.
2. Warder, Indian Kāvya literature, pp. 128–136.

3. Romila Thapar, Sakuntala: Texts, Readings, Histories, New Delhi: Kali for Women, 1999, pp. 10, 16;

D. D. Kosambi and V. V. Gokhale (eds.) The Subhāṣitaratnakoṣa, vol. II, 1957, Harvard Oriental
Series, vol. 42, Harvard: Harvard University Press, p. xlvi.
4. Thapar, Sakuntala, p. 8.

5. The reference to festivals as occasions for the performance of plays is from NS IV 269. This is also

the opinion of J. A. B. van Buitenen, (van Buitenen, transl., Two Plays of Ancient India, Delhi: Motilal
Banarsidass, 1971, p. 26); Warder (Indian Kāvya Literature, pp. 13, 123); Manmohan Ghosh (Ghosh,
transl., The Nāṭyaśāstra ascribed to Bharata muni, vols. I–II, Revised Second Edition, Calcutta:
Manisha Granthalaya, 1967, vol. I. pp. xlix, li); and A. B. Keith (Keith, The Sanskrit Drama in Its
Origin, Development, Theory and Practice, London: Oxford University Press, 1924, pp. 370–71). In the
Kuṭṭanῑmataṃ, 792 Samarabhata witnesses a musical play at the temple of Lord Śiva in Varanasi. Cited
in Warder, Indian Kāvya Literature, vol. IV, p. 575.
6. Mālatῑ. I. Prologue, p. 2. Bhavabhūti’s Mālatῑmādhava (Text and Translation by M. R. Kale, Third

Edition), Delhi: Motilal Banarsidass, 1967.

7. Kaumudῑmahotsava I, Prologue, p. 3, Kaumudῑ-Mahotsava Nāṭaka, Text and Translation (into

Hindi) by Devadatt Shastri, Delhi: Sahni Prakashan, 1953; Mālavikāgnimitram. I. Prologue, p. 3.

Works of Kālidāsa, vol. 1 (Reprint of First Edition, Edited and Translated by C. R. Devadhar), Delhi:
Motilal Banarsidass, 2002.
8. HC IV, pp. 142–48. The Haṣācarita of Bāna (Second Edition, Translated by E. B. Cowell and F. W.

Thomas), Delhi: Motilal Banarsidass, 1968.

9. NS I. 12. Nāṭyaśāstra, Sanskrit text with the commentary of Abhinavagupta, ed. M. Ramakrishna

Kavi, vols. I–IV, Baroda: Oriental Institute, 1934. The Nāṭyaśāstra ascribed to Bharata muni.
10. For example, Edward C. Dimock, Jr. (in Dimock Jr., Edwin Gerow, C. M. Naim, A.K. Ramanujan,

Gordon Roadarmel, J.A.B. van Buitenen, The Literatures of India: An Introduction Chicago/London:
University of Chicago Press, 1974, p. 45 writes: ‘. . . the theatre deliberately limits its audience to the
educated and refined, . . . [and] rouse[s] in the audience a symphony of moods beyond the
uneducated sensibilities of the uncultured’. For another instance, see Thapar, Sakuntala, p. 46; van
Buitenen in Dimock, Gerow et al, The literatures of India, p. 93. van Buitenen also writes: ‘The
prevalence of Sanskrit and the variety of Prakrits that were used demanded of the audience an
enormous erudition, and it was at the erudite and cultured spectator that the play was aimed’ (ibid., p.
113). Please note that this is an understanding largely based on the concept of sahṛdaya that occurs
only in late works on rhetoric. It indicated a spectator ‘with heart’, i.e. emotionally refined—at best, the
ideal spectator at whom the rasa theory of dramatics was aimed. In contrast to these scholars, taking a
minority stand, Warder observes in connection with the social milieu of the Kavya as described in the
NS and the Kamasutra that: ‘“classical” literature is not opposed to “popular” and has usually sought a
mass audience’ (in A. L. Basham (ed.), A Cultural History of India, New Delhi: Oxford University
Press, 1975, p. 175).
11. NS XXVII 57, 61; XXVI 126.

12. Ibid., XXVII 50–56; 63–64.

13. Ibid., XXVII 59, stated earlier as well in XXVI 126.

14. Mālavikāgnimitra. I. 4, p. 11.

15. NS VI 23; XXIV 3.

16. NS XV 2. A reading of the NS reveals how costumes and make-up were far from mere props in

early India’s understanding of histrionic representation. Chapter XXIII tells us the great deal about
characterisation and mood and plot that was conveyed by carefully calibrated conventions regarding
colour of clothing and face-paint, ornaments, hairstyle, masks, etc. Chapters VIII to XIII make exactly
the same point vis-à-vis gestures and expressions, ranging from gaits to glances. Thus, the NS XXIII 3
comments: ‘Different types of dramatis personae indicated first by their costumes and make up,
accomplish representation without much effort by means of gestures and the like’ [emphasis added].
17. T. Burrow in Basham, A Cultural History of India, p. 162.

18. Burrow in Basham, A Cultural history of India, p. 165; Thapar, Sakuntala, p. 78; Hans Henrich

Hock, and Rajeshwari Pandharipande cited in Jan E. M. Houben, ed., Status and ideology of Sanskrit,
Leiden/New York: E. J. Brill, 1996, p. 160; Daud Ali, Courtly Culture and Political Life in Early
Medieval India, First south Asian edition, New Delhi: Foundation Books, 2006, p. 81.
19. NS XVIII 49.
20. NS XVIII.

21. Keith, The Sanskrit Drama in Its Origin, Development, Theory and Practice, p. 74; Sheldon Pollock

in Sheldon Pollock (ed.), Literary Cultures in History: Reconstructions from South Asia, New Delhi:
Oxford University Press, 2004, pp. 62–63; Dimock in Dimock, et al., The Literatures of India, p. 12.
22. Houben cites several pieces of evidence in favour of Sanskrit as a language which was spoken and

not only written, Houben, Status and Ideology of Sanskrit, pp. 166–73. However, while the jury is out,
this is probably not what the majority of scholars believe. Pollock (in Houben, Status and Ideology of
Sanskrit, p. 198), for example, is very clear that Sanskrit was never the language of the home or even
the court, and Dimock in Dimock, et al., The Literatures of India, pp. 7, 11 would agree (‘it was always
a second language even for those who belonged to the rarified group of highly educated knowers of
Sanskrit’).
23. Kāvyādarśa I, p. 3. This verse does not actually name Sanskrit and Prakrit (and Apabhramsha) but

refers to ‘the three kinds of speech in which the world’s affairs are conducted’. This has commonly
been interpreted as referring to the three languages mentioned.
24. Ibid. I, p. 32; Pollock, Literary Cultures in history: Reconstructions from South Asia, p. 61.

25. Kāvyādarśa I, p. 33.

26. Ibid. I, p. 37 says that the sargabandha or mahākāvya is in Sanskrit, the skandhaka etc. (sic) in

Prakrit, and the osara etc. (sic) in Apabhramsha, while drama is mixed (miśrakaṃ). The skandhaka
and osara were distinguished by the metre they used, and are not common among the texts that have
come down to us.
27. The way these have been defined or distinguished in traditional works on poetics suggests to me

that Sanskrit and Prakrit may in fact have been not different languages as much as different forms of
linguistic or speech practice (vācaṃ, bhāṣā): Sanskrit, that which is shown to be rigorously worked out
and systematised (anuśiṣṭa, anvākhyātā); Prakrit, that which is also disciplined (śiṣṭa), derived from
Sanskrit with or without changes (tadbhav, tatsaṃ), and assimilated partially from local speech (deśῑ).
Of course, I am no linguist and a provocative statement such as this would call for detailed
investigation and argumentation relating to the structure of these languages, which is beyond my
brief. But this is my impression of Kāvyādarśa I, pp. 3, 33 and NS XVIII 3. The NS (XVIII 30) also
speaks of Sanskrit and Prakrit as the two ways (dvividham) of recitation (pāthya) in the ‘common
language’ (Ghosh’s translation of the word jātibhāṣa) which relates to the four castes
(jātibhāṣāasrayam caturvarṇyasamāśrayaṃ). For a discussion of Sanskrit as the hyper- and Prakrit as
the hypo-varieties within the same language, known as diglossia, see Houben in Houben, Status and
Ideology of Sanskrit, pp. 158–80.
28. NS XVII 121.

29. See Shonaleeka Kaul, Imagining the Urban: Sanskrit and the City in Early India, Delhi: Permanent

Black, 2010, especially chapter five.

30. Perhaps the most sophisticated understanding so far of Kāvyas as courtly literature is Daud Ali’s

recent Courtly Culture and Political Life in Early Medieval India (2006). Departing from the tendency
to reduce literature, or art more generally, to a legitimising force for the ruling class, the book sees
Kāvya aesthetics as documenting practices constitutive of relationships between the king, the harem,
and courtiers, and as ‘assisting the education of elites in a mannered system of emotions and
dispositions central for the maintenance of formal relationships at court’ (p. 23). His presentation and
comparison of the material is fresh and persuasive, but the fact remains that Kāvyas are still being
viewed in terms of fulfilling an overwhelmingly political function/agenda. In its exclusively court-
centric perspective, Ali sometimes over-reads the presence of the court in the themes and
terminology of the texts, going so far as to characterize the Nāṭyaśāstra as a book of ‘courtly aesthetics’
(p. 171). His work has the merit of bringing in the urban context of the cultural dispensation he talks
about—only to entirely subordinate the city to the court: in refined matters social and cultural, ‘the
ways of the town mirrored those of the court’ (p. 61); the ways of the latter formed ‘an acculturative
mechanism’ (p. 22) and ‘seeped into the city’ (p. 23). Clearly, the court is seen as the source of cultural
ideals associated with the city. No matter that the ideas and practices in a text like the Kāmasūtra, so
central to Ali’s discussion, are designated by the text as ‘urban’ (nāgaraka) rather than anything else!
Of course, Ali would argue that urban and urbane were synonymous with courtly, but to so privilege
the royal world may be to foreclose exploration of the myriad institutions and influences early
urbanism represented.
Chapter 12
1. K. G. Sesha Aiyar, Cera Kings of the Sangam Period, London, 1937, p. 147.

2. Cf. H. M. Chadwick and N. K. Chadwick, The Growth of Literature (henceforth GL).

3. Purānanuru (henceforth PN) 282: 10–11.

4. Malaipatukatām (henceforth MPK) 77.

5. Cf. HT, p. 16.

6. T. Burrow and M. B. Emeneu, A Dravidian Etymological Dictionary (henceforth DED), Oxford,

1961, s.v. 3351, p. 272.

7. Cirupāṇārrupapatai (henceforth CPA) 34–35.

8. Perumpānārrupapatai (henceforth PPA) 462.

9. Porunarārrupapaçai (henceforth PAP) 64; Perumpānārrupapatai (henceforth PPA) 20;

Maturaikkānci (henceforth MK) 576; MPK 54; PN 370: I l.

10. PN 212: 6–8.

11. Ibid., 400: 16–17.

12. Akanānuru (henceforth AN) 216: l.

13. Ibid., 126: 9–12.

14. Cf. J. R. Marr, ‘The Eight Tamil Anthologies with special reference to Purananuru and

Patirruppattu’, thesis approved for the degree of Doctor of Philosophy in the University of London,
1958, pp. 439 f.
15. Patirruppattu (henceforth PrP) 40: 22–26; 41: 6.

16. Tamil Lexicon (henceforth MTL), Madras University s.v.v., p. 2934.

17. AN 36: 19.

18. DED, s.v. 3708, p. 299.

19. PN 386: 19.

20. DED, s.v. 3716, p. 300.

21. Ainkurunuru (henceforth AKN) 1: 2, 2: 2, 10: 2.

22. Paripātal (henceforth Pari.) 10: 85–86.

23. PN 373: 27–29.

24. Such songs have survived to this day. Any folk-song collection in Tamil or Malayalam would have

polippattu, ‘song in praise of abundance’. Cf. HMV LP Record, Folk Songs of South India (ECLP 2259).
25. Purapporul Veṇpāmālaṅ (henceforth PPVM) v. 159.

26. PN 394: 1–7.

27. Ibid., 397: 1–10, 384: 1–10, 396: 101–14.

28. PAP 1–3.

29. PAP 3–24.

30. Ibid., 100–01.

31. DED, s.v. 1570, p. 129.

32. DED, s.v. 1570, p. 129.

33. I. Shekhar, Sanskrit Drama: Its Origin and Decline, Leiden, 1960 p. 88.

34. The heroic poems contain stray references to the costume and make up of dancers. Cf. PN 29: 22–

23.
35. PN 336: 6; CPA 125; Kurincippattu (henceforth KP) 193; AN 138: 9, 301: 20, 382: 4; MPK 268.

Three poems in the anthologies—PN 64, Kuruntokai (henceforth KT) 178 and 203—are attributed to
persons whose names seem to have some connexion with an orchestra: e.g., Netumpalliyattan, ‘he of
the big orchestra’. Obviously it is not a proper noun, but an appellative noun.
36. ‘History of Tamil Drama’, in Bharatiya Natya Sahitya, Govinda Das Commemoration Volume, 1956,

pp. 421–22.
37. An interesting instance of the survival of muḻavu among the dancers in Malabar is pointed out by

Dr. C. Achyuta Menon, ‘The Histrionic Art of Malabar’, Annals of Oriental Research of the Madras
University iv, pt. i, 1939, p. 15.
38. PPA 55–56.

39. PN 104: 12–13; MPK 2–3, 143; MK 605; PrP 41: 1–5, 56: 2.

40. PrP 41: 1–5.

41. PN 103: 1–2.

42. A small drum.

43. Another kind of small drum.

44. PN 152.

45. MPK 11–12.

46. N. Chengalvarayan, ‘Music and Musical Instruments of the Ancient Tamils’, QJMS, n.s. xxvi

(1935), 85 f.
47. Cilappatikāram (henceforth Cil) 3: 14 and comm.

48. PN 152: 20, and comm.

49. MPK 359, 470.

50. Narriṇai (henceforth NT) 212: 3; AN 301: 21, 309: 9.

51. The word connotes either an open–air ground or some common house or building. In any case

they served many purposes: cattle grazing ground, village theatre, village assembly hall, women’s
gathering place, etc. Cf. Studies, pp. 224 f.
52. AN 301: 23–25. Cf. PN 29: 24.

53. Connected with it is āṭumakaḷ , a woman of the dancing community. Cf. NT 95: 1–2.

54. DED, s.v. 1160, p. 98.

55. KP 192–5.

56. NT 95: 1–2.

57. KT 31: 1–2; MK 591–6.

58. PrP 56: 1–2.

59. Ibid., 30: 18–20.

60. AN 141: 5–11.

61. S. Vithiananthan, ‘The Pattuppattu—a historical, social and linguistic study’, thesis approved for

the degree of Doctor of Philosophy in the University of London, 1950, p. 194; Studies, pp. 290 f.
62. PN 33: 16–17.

63. PrP 41: 3–6.

64. PN 139: 1–2.

65. Purapporul Veṇpāmālai v. 219.

66. Tirumurukārruppati (henceforth TMA) 47–57.

67. PPVM v. 19.

68. MPK 534–9.

69. PN 103: 1–4, 64: 4–5.

70. Ibid., 60: 4–5.

71. Ibid., 64.

72. Ibid., 64: 1.

73. Ibid., 103: 1–2.

74. MK 219.

75. PN 139: 4.

76. NT 170: 1–5; PN 89: 1–2.

77. Cf. Shekhar, Sanskrit Drama, p. 89.

78. Cf. M. S. H. Thompson, ‘The Auvai of the Sangam Anthologies’, Bulletin of the School of Oriental

and African Studies xii. 400 f.

79. DED, s.v. 1824, p. 146. The early Tamils classified their musical instruments into three main

categories: (1) stringed instruments, (2) wind instruments, and (3) percussion instruments. For details
and subdivisions, see P. Sambamoorthy, ‘Catalogue of the Musical Instruments Exhibited in the
Government Museum, Madras’, Bulletin of the Madras Government Museum, New Series iii, pt. 3,
1931.
80. TMA 243–6.

81. DED, s.v. 4514, p. 378.

82. TMA 119–24.

83. MP 92; MK 195.

84. AN 378: 8.

85. Ibid., 328: 1–2.

86. Ibid., 155: 13.

87. MTL, s.v. ii, p. 696.

88. Tamil-English Dictionary, s.v. p. 184.

89. MK 518, Nacc. comm., p. 397.

90. Cil. 5: 30.

91. MPK 50, Nacc. comm., p. 610.

92. PrP 29: 8–9, 23: 3–4.

93. Panipatal (henceforth Pari.) 10: 130, and comm., p. 113.

94. DED. s.v. 11. p. 3.

95. MK 223–4, p. 362.

96. Cf. Manusmrti, x. 11; also N. K. Sidhanta, London, 1929 The Heroic Age of India, pp. 63 f.

97. Marr, ‘The Eight Tamil Anthologies’, p. 470.

98. MTL, s.v. ii, p. 1194.

99. PrP 43: 27–28.

100. AN 97: 9–11.

101. Ibid., 152: 4, 208: 3.

102. PrP 43: 27 and comm., p. 119.

103. PN 152 comm.

104. It is interesting to note that in both Tamil and Greek the staff of the king and the minstrel’s rod

are denoted by the same word or synonyms. Ta. kol, cenkol; Gk. rhabdos, kerykeion. Cf. J. E. Harrison,
Prolegomena to the Study of Greek Religion, Cambridge, 1903, pp. 44–47.
105. DED, s.v. 3563, p. 288.

106. TMA 261, 268, and 280.

107. PN 72: 13–16.

108. Vithiananthan, ‘The Pattupattu’, p. 2.

109. Ibid.

110. PN 27:7–9.

111. Cf. G. Murray, A History of Ancient Greek Literature, p. 3 and London, 1902 A. S. Mackenzie, The

Evolution of Literature; GL 3, p. 748, and passim.

112. PN 206:4.

113. Ibid., 47: 6.

114. PrP 56: 8, 57: 4.

115. MPK 534–9.

116. PN 47: 7–8.

117. The many dialogue songs in the Kaliltokai abounding in dramatic qualities might be said to

represent a transitional form between songs and plays.

118. K. A. Nilakanta Sastri, History of South India, 1955, p. 72 f.; S. V. Pullai, History of Tamil Language

and Literature (henceforth HTLL), pp. 62 f.

119. Chadwick and Chadwick, The Growth of Literature, I, pp. 631–2.

120. PN 162.

121. PN 208: 7–8.

122. Tolkāppiyam, Porulatikāram, Ceyyuliyal, sutra (henceforth Tol. Porul. Ceyyul., cū) 393.

123. Cf. A. C. Chettiar, Advanced Studies in Tamil Prosody (Third Edition), Annanalai Nagar, 1957, p.

43; HTLL, p. 44.

124. Cf. The ‘Tiruvacagam’ (Translated by G. U. Pope), p. lxxxix.

125. Vithiananthan, ‘The Pattupattu’, p. 128. Cf. M. Winternitz, History of Indian Literature, ii. 475,

where he says that it was among the merchant class in particular that the Jains found their most loyal
lay adherents; also D. D. Kosambi, An Introduction to the Study of Indian History, p. 158.
126. DED, s.v. 850, p. 75.

127. Kural 14. I.

128. Meenakshisundaram, Collected Papers, p. 19.

129. Ibid., p. 33.

130. Cf. HTLL, pp. 65 ff.

131. DED, s.v. 1677, p. 136; also the word kilavi, ‘word, speech, language, utterance’ (ibid.). The verb

kila, ‘to speak’ (i.e., ‘to be heard’). In both the bardic corpus and the Tol. kiḷavi means ‘poetical
utterances’, a fact which points to the oral and auditory nature of the literature of the early period. On
the ‘concept’ of utterance in oral poetry see A. B. Lord, The Singer of Tales, pp. 25 f.
132. PN 53: 12.

133. SED, s.v. p. 1027.

134. PrP 64: 4–5; PPA 315. Cf. PN 26: 12.

135. v. 163.

136. PN 15: 16–21, 26: 12–15.

137. Studies, p. 5; K. A. N. Sastri, Studies in Chola History and Administration, Madras, 1932, p. 9.

138. PN 99. 2.

139. W. Logan, Malabar Manual, ii. 7.

140. MK 61.

141. PN 9: 10.

142. Ibid., 201: 10–12.

143. For a recent discussion of the problem, see A. S. Gnanasambanthan, in Patirruppattuc

Corpolivkual, pp. 125–53. Cf. P. T. S. Iyengar, History of the Tamils, (henceforth HT), pp. 495 f.
144. Marr, ‘The Eight Tamil Anthologies’, pp. 285, 292.

145. SCHA, pp. 37 f.

146. HTLL, p. 11.

147. GL 1, p. 2.70.

148. M. I. Finley, ‘Myth, Memory, and History’, History and Theory, iv, no. 3 1965, p. 283.

149. PPA 454.

150. Coccinea indica.

151. HT, pp. 398; 401 ; Vithiananthan, ‘The Pattupattu’, p. 74.

152. PPA 452–54.

153. PPA 29–31.

154. Cf. Gazetteer of the Salem District, i, pt. i, 46 f.

155. GL 1, p. 276.

156. e.g. Iṉṉa nārpatu and Juiyavainārpatu are actually verse catalogues of painful and unpleasant,

and delightful and pleasant things and deeds respectively: see K. Zvelebil, Archiv Orientalniri xxv
(1957), 56–82; xxvi (1958), 385–426.
157. PN 158:1–6.

158. CPA 84–133.

159. Tol. Porul Ceyyul., cu. 437, and comm., p. 388.

160. KP 61–95.

161. HT, p. 546.

162. M. Varadarajan, The Treatment of Nature in Sangam Literature, p. 62.

163. Vithiananthan, ‘The Pattupattu’, p. 20.

164. Mani. 3:160–6.

165. AN 44: 7–9.

166. AN 36: 15–19.

167. Ibid., 135: 12.

168. Ibid., 125: 21.

169. MK 55.

170. Cf. PN 19: 17.76: 12–13.

171. TMA 257–76. This factor must have contributed to the poem becoming a liturgy.

172. Vithiananthan, ‘The Pattupattu’, p. 192.

173. The notable part played by Jain monks in the literary and grammatical developments in Tamil

has been noted by every writer on the subject. One need only recall the important historical events of
the establishment of a Dravida Sangha at the capital of the Pantiya country by the Jain monk Vajra
Nandi in 470 CE. See Studies, pp. 21 f.; HTLL, pp. 14, 58, 61 ff.; Winternitz, History of Indian Literature,
pp. 428, 595.
174. The Tamil word palli means, among other things, a Jaina temple or hermitage, and a school. This

might reflect the early connections between the two.

175. The obvious and perhaps the only comparable parallel to the Tolkāppuyam, both in its breadth

and abiding influence is the work of Aristotle.

176. Murray, The Rise of the Greek Epic, p. 196. Cf. Arokiaswami, The Early History of the Vellar Basin,

p. 15.
Chapter 13
1. Alan Dundes and Carl R. Pagter, Work Hard and You Shall be Rewarded: Urban Folklore from the

Paperwork Empire, Bloomington: Indiana University Press, 1978.

2. Milton singer, When a Great Tradition Modernizes: An Anthropological Approach to Indian

Civilization, Chicago: University of Chicago Press, 1972.

3. R. G. Collingwood, An Essay on Philosophical Method, Oxford: The Clarendon Press, 1983.

(Originally published in 1933.)

4. A. K. Ramanujan, ‘Where mirrors are windows: Toward an anthology of reflections’ in History of

Religions 28(3), 1989, pp. 187–216.

5. For a fuller development of this idea, see ibid.

6. J. F. Stall, Namburi Veda Recitation, vol. 5 of Disputationes Reno-Trajectinae, ed. J. Gonda, Hague:

Mouton, 1961.
7. Archer Taylor, English Riddles from Oral Tradition, Berkeley: University of California Press, 1961.

8. Georges Dumezil, Mythe et Epopee, Paris: Gallimard, 1968.

9. A.K. Ramanujan, ‘The Indian Oedipus’ in Oedipus: A Folklore Casebook, ed. Alan Dundes and

Lowell Edmunds, New York: Garland Press, pp. 234–64, 1983.

10. Ruth Finnegan, Oral Poetry: Its Nature, Significance and Social Context, Cambridge: Cambridge

University Press, 1911.

11. Philip Lutgendorf, ‘The life of the text: Tulasidas’s Rāmacaritamānas in performance’, Ph.D.

dissertation, University of Chicago, 1987.

12. William A. Graham, Beyond the Written Word: Oral Aspects of Scripture in the History of Religion,

Cambridge: Cambridge University Press, 1987.

13. David Shulman, Tamil Temple Myths: Sacrifice and Divine Marriage in the South Indian Saiva

Tradition, Princeton: Princeton University Press, 1980.

14. Edward C. Dimock, The Sound of Silent Guns and Other Essays, New Delhi: Oxford University

Press, 1989.
15. Alf Hiltebeital, The Cult of Draupadi: Mythologies from Gingee to Kuruksetra, Chicago: University

of Chicago Press, 1988.

16. Richard Lee Brubaker, ‘The ambivalent mistress: A study of south Indian village goddesses and

their religious meaning’, Ph.D. dissertation, University of Chicago, 1978.

17. Stuart Blackburn, Singing of Birth and Death: Texts in Performance, Philadelphia: University of

Pennsylvania Press, 1988.

18. For e.g., Peter Claus, ‘Behind the Text: Performances and Ideology in a Tulu Oral Tradition’ in

Oral Epics in India, eds. Stuart Blackburn, et al., Berkeley: University of California Press, 1989.
19. Stuart Blackburn and A. K. Ramanujan, eds., Another Harmony: New Essays on the Folklore of

India, Berkeley: University of California Press, 1986. See also Brenda E. F. Beck, Folktales of India,
Chicago: University of Chicago Press, 1982 and Gene H. Roghair, The Epic of Palnadu: A Study and
Translation of Palnati Vinula Katha, a Telugu Oral Tradition from Andhra Pradesh, India, New York:
Oxford University Press, 1982.
20. David Shulman, ‘Battle as metaphor in Tamil folk and classical traditions’ in Another Harmony:

New Essays on the Folklore of India, eds. Blackburn and Ramanujan, 1986.
21. A. K. Ramanujan, ‘On Folk Puranas’, Conference on Puranas, University of Madison, Wisconsin.

22. A. K. Ramanujan, Speaking of Śiva, Harmondsworth: Penguin Books, p. 101.

23. A. K. Ramanujan, ‘Two Realms of Kannada Folklore’, 1986.

24. I have said little about Indian oral tales, though I end this paper with an example. See Beck,

Folktales of India for a recent, wide selection with anthropological notes, and Kirin Narayan,
Storytellers, Saints, and Scoundrels: Folk Narrative in Hindu Religious Teaching, Philadelphia:
University of Pennsylvania Press, 1989 for a fresh contextual study of tales in religious teaching.
25. This tale, reproduced here in translation from my forthcoming book of Kannada folktales, is also

told in many other regions and languages of India. The Stith Thompsen index of international tale
types (1961) identifies it as 1534 An Innocent Man Chosen to Fit the Stake. This tale has so far been
recorded only for India, and twenty-one variants have been recorded in Kashmiri, Kannada, Tamil,
Marathi, Hindi, Garhwali, and so on.