Women Lives Women's Rituals PDF

Women’s Lives, Women’s
Rituals in the Hindu Tradition

This page intentionally left blank
Women’s Lives,
Women’s Rituals
in the Hindu
Tradition
Edited by
tracy pintchman
1
2007
1
Oxford University Press, Inc., publishes works that further
Oxford University’s objective of excellence
in research, scholarship, and education.
Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Copyright Ó 2007 by Oxford University Press, Inc.
Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016
www.oup.com
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data

Women’s lives, women’s rituals in the Hindu tradition /
edited by Tracy Pintchman.
p. cm.
Includes bibliography and index.
ISBN 978-0-19-517706-0; 978-0-19-517707-7 (pbk.)
1. Hindu women—Religious life. 2. Hinduism—Rituals.
3. Hindu women—Social life and customs. I. Pintchman, Tracy.
BL1237.46.W67 2007
294.5'38082—dc22 2006021014
9 8 7 6 5 4 3 2 1
Printed in the United States of America

on acid-free paper
For my son,
Noah Lawrence French,
born August 4, 2002
Acknowledgments and
Note on Transliteration
I would like to thank all of the contributors for their hard work and
their patience during the years this volume was in process. Thank
you, thank you, thank you! I would like to express my gratitude as
well to the editorial staff at Oxford University Press, especially Theo
Calderara, who cheered us on and never seemed to doubt that some
day we would actually finish this book and get it to press. Many
thanks, too, to Corinne Dempsey for her very helpful comments on
the manuscript.
We have chosen to eliminate all diacritical marks from the book

and to minimize the use of terms from Indian languages in order
to make the chapters more accessible to a nonspecialist audience.
Contents
List of Illustrations, xi
Contributors, xiii
Introduction, 3
Tracy Pintchman
PART I Engaging Domesticity
1. The Cat in the Courtyard: The Performance of Sanskrit

and the Religious Experience of Women, 19
Laurie L. Patton
2. Wandering from ‘‘Hills to Valleys’’ with the Goddess:

Protection and Freedom in the Matamma
Tradition of Andhra, 35
Joyce Burkhalter Flueckiger
3. Lovesick Gopi or Woman’s Best Friend? The Mythic

Sakhi and Ritual Friendships among Women in Benares, 55
Tracy Pintchman
4. Words That Breach Walls: Women’s Rituals in Rajasthan, 65

Lindsey Harlan
x contents
5. Threshold Designs, Forehead Dots, and Menstruation

Rituals: Exploring Time and Space in Tamil Kolams, 85
Vijaya Rettakudi Nagarajan
PART II Beyond Domesticity
6. Domesticity and Difference/Women and Men:

Religious Life in Medieval Tamilnadu, 109
Leslie C. Orr
7. The Anatomy of Devotion: The Life and Poetry

of Karaikkal Ammaiyar, 131
Elaine Craddock
8. The Play of the Mother: Possession and Power

in Hindu Women’s Goddess Rituals, 149
Kathleen M. Erndl
9. Does Tantric Ritual Empower Women? Renunciation

and Domesticity among Female Bengali Tantrikas, 159
June McDaniel
10. Performing Arts, Re-forming Rituals: Women

and Social Change in South India, 177
Vasudha Narayanan
Index, 199
List of Illustrations
Figure 2.1. Govindamma invoking the goddess Gangamma

to possess her. Photo by Joyce Burkhalter
Flueckiger, 47
Figure 5.1. A Tamil girl making a labyrinth kolam,
Mayiladuthurai, Tamilnadu. Photo by Vijaya
Nagarajan, 86
Figure 7.1. Drawing of Karaikkal Ammaiyar by Haeli Colina, 132
Figure 9.1. Sannyasini Gauri Ma and the author in
Bakreshwar, West Bengal. Photo by Jim Denosky, 167
Contributors
Elaine Craddock is an associate professor and chair of the Depart-

ment of Religion and Philosophy at Southwestern University. She
teaches courses in Hinduism, Buddhism, and feminist studies. Her
research interests include Hindu goddesses, tantra in South India,
Tamil bhakti poetry, and religion and the body. Her publications in-
clude ‘‘Reconstructing the Split Goddess as Shakti in a Tamil Village,’’
in Seeking Mahadevi: Constructing the Identities of the Hindu Great
Goddess, edited by Tracy Pintchman (2001).
Kathleen M. Erndl is an associate professor of religion at Florida

State University, where she teaches undergraduate and graduate
courses on Hinduism, South Asian religions, Sanskrit, gender and
religion, and theory and method in the study of religion. She is the
author of Victory to the Mother: The Hindu Goddess of Northwest India
in Myth, Ritual, and Symbol (1993), and with Alf Hiltebeitel she co-
edited the volume Is the Goddess a Feminist? The Politics of South Asian
Goddesses (2000). She has also published numerous essays on god-
dess traditions and women in South Asian religions. She is currently
writing a book entitled Playing with the Mother: Women, Goddess Pos-
session, and Power in Kangra Hinduism. Her other research interests
include gender and religious hybridity in India and the United
States.
Joyce Burkhalter Flueckiger is a professor in the Department of

Religion at Emory University. She specializes in performance studies,
with a particular interest in religion and gender. She has carried
xiv contributors
out extensive fieldwork in India, working with both Hindu and Muslim pop-
ular traditions. Her most recent book, In Amma’s Healing Room: Gender and
Vernacular Islam in South India (2006), addresses questions of religious and
gender identities and boundaries in the healing practice of a female Muslim
folk healer in the city of Hyderabad. Her most recent fieldwork was conducted
in Tirupati with the goddess tradition of Gangamma. She is also the author of
Gender and Genre in the Folklore of Middle India (1996) and has published
numerous articles on South Asian performative traditions.
Lindsey Harlan is a professor of religious studies at Connecticut College. She is

the author of The Goddesses’ Henchmen: Gender in Indian Hero Worship (2003)
and Religion and Rajput Women (1992). With Paul Courtright, she edited the
volume From the Margins of Hindu Marriage (1995). Her most recent work has
focused on cultic heroic traditions in India and traditions of hero memo-
rialization in the United States. She also writes on Hinduism in Trinidad.
June McDaniel is a professor of religious studies at the College of Charleston,

where she teaches classes on world religions, women and religion, the reli-
gions of India, the phenomenology of religion, mysticism and religious ex-
perience, and other topics. Her publications include a number of articles and
book chapters and three books: Offering Flowers, Feeding Skulls: Popular Goddess
Worship in West Bengal (2004); Making Virtuous Daughters and Wives: An In-
troduction to the Brata Rituals of Bengal (2003); and The Madness of the Saints:
Ecstatic Religion in Bengal (1989). Her current research is on Hindu bhakti,
Shakta tantra, world mysticism, and religious ecstasy.
Vijaya Rettakudi Nagarajan is an associate professor of South Asian religions

in the Department of Religious Studies and Environmental Studies at the
University of San Francisco, where she teaches courses on Hinduism, religion
and ecology, religion and nonviolence, and other topics. Her publications in-
clude a number of articles and essays on the kolam and Hinduism and ecology,
as well as a forthcoming book, Drawing Down Desires: An Exploration of the
Kolam: Women, Ritual and Ecology in Southern India. Her current research fo-
cuses on tree temples, sacred groves, and the commons in Tamilnadu, India.
Vasudha Narayanan is a professor of religion at the University of Florida and a

past president of the American Academy of Religion (2001–2002). Her fields
of interest include the Sri Vaishnava tradition; Hindu traditions in India,
Cambodia, and America; and gender issues. She is the author and editor of six
books, including, most recently, Hinduism (2004) and a forthcoming work, A
Hundred Autumns to Live: An Introduction to Hindu Traditions. She is currently
researching Hindu temples and Vaishnava traditions in Cambodia.
contributors xv
Leslie C. Orr is an associate professor in the Department of Religion at Con-

cordia University in Montreal, Canada. Her research is concentrated in two
overlapping areas: first, the roles and activities of women in Hinduism, Bud-
dhism, and Jainism; and, second, the organization of religious life in the
history of South India up to the late medieval period, including the examina-
tion of institutional structures, religious authority and social hierarchy, ritual
forms and ritual performers, and the physical space in which social and ritual
dynamics transpired, with a particular focus on the interactions among reli-
gious communities. She is the author of a number of articles in these fields and
the book Donors, Devotees and Daughters of God: Temple Women in Medieval
Tamilnadu (2000). She has a book in progress on the sociology of the medieval
South Indian temple.
Laurie L. Patton is Winship Distinguished Research Professor in the Human-

ities at Emory University. She has authored two books on early Vedic inter-
pretation of mythology and ritual: Myth as Argument (1996) and Bringing the
Gods to Mind (2005). She has also served as editor or co-editor of five volumes
on myth theory, Vedic hermeneutics, the Indo-Aryan controversy, history of
religions, and gender and early Hinduism. She is currently working on a book
on women and Sanskrit as well as a book on scandal in the study of religion.
She has recently completed a book of poetry, Fire’s Goal: Poems from the Hindu
Year (2003), and a translation of the Bhagavad Gita.
Tracy Pintchman is a professor of religious studies and Hindu studies at Loyola

University Chicago, where she teaches courses on Hindu traditions and women
and religion. Her previous publications include a number of articles and book
chapters; two monographs, The Rise of the Goddess in the Hindu Tradition (1994)
and Guests at God’s Wedding: Celebrating Kartik among the Women of Benares
(2005); and an edited volume, Seeking Mahadevi: Constructing the Identities of the
Hindu Great Goddess (2001). She is currently researching narratives of meaning
surrounding motherhood.
Women’s Lives, Women’s
Rituals in the Hindu Tradition
Introduction
Tracy Pintchman
We give birth to a child, we put oil on the child’s body, and we

massage the child. Men can’t do all this. So ritual worship [puja] is
like this. Men cannot do as much as we do.
The woman who spoke these words, Bhagavanti, was reclining on

the floor of her home in the city of Benares, North India, and watch-
ing her daughter-in-law sort rice as my research associate, Sunita
Singh, and I sat before her, trying to interview her. It was November
1997, and I had come to Bhagavanti’s home to ask about her partic-
ipation in Kartik puja, a tradition of women’s devotional practices
specific to the month of Kartik (October–November). Bhagavanti sat
before us, her elbow resting on a pillow as she responded patiently to
the questions we put before her. She was quite elderly at the time of
our interview and had lost a good deal of her hearing, so Sunita and I
had to shout to make ourselves heard, drawing the attention of many
neighborhood women and children, who came to peer through the
windows to see what was going on. Bhagavanti’s daughter-in-law kept
getting up and milling about, but she also chimed in whenever she
had a point she wanted to make. At this particular moment, I was
asking Bhagavanti why men don’t participate in the particular ritual
tradition, called Kartik puja, about which I was asking her. Like many
other women I interviewed, Bhagavanti attributed men’s absence
to differences between men’s and women’s natures and social
roles in ways that seemed to favor women as the more virtuous and
disciplined gender (cf. Pintchman 2005).
4 introduction
Bhagavanti’s words quoted above bring to mind the common perception

among many Indian Hindus that women have a special aptitude for ritual per-
formance. During the years from 1995 to 1998, when I was conducting field
research in Benares, on numerous occasions both women and men suggested
to me that, in India, women are more religious than men and do the lion’s
share of the day-to-day religious work. Like other women I interviewed, Bha-
gavanti also drew a clear parallel between the work women do in the household
and the work they do in this particular puja tradition, comparing ritual tasks
specifically with maternal tasks and the responsibilities involved in birthing
and caring for children. For Bhagavanti, the domain of ritual practice is not
separated from day-to-day existence but instead involves the kinds of skills and
commitments that appear regularly and without special notice in the mundane
arena of ordinary life.
The chapters in this book explore the relationship between women’s rit-
ual practices and the lives and activities of Hindu women beyond the ritual
sphere. Bhagavanti’s words provide a helpful point of departure for our col-
lective inquiry into the nature of this relationship. Like Bhagavanti, we pre-
sume that Hindu women are deeply engaged and invested in the performance
of religious practice. Rituals that take place in Sanskritic, Brahminical Hindu
environments continue to be instituted and directed largely by Brahmin males,
but women largely control many types of ritual practice that occur outside of
such contexts, including many household, calendrical, and local devotional
practices. Even in environments where Sanskritic traditions maintain a strong
presence, women often sustain active ritual agendas and function as engaged
actors in many types of ritual work. Indeed, in some parts of India, women are
taking leadership roles in Sanskritic ritual performance as well.
Like Bhagavanti, too, we maintain that Hindu women’s religious prac-
tices are not isolated from social, cultural, domestic, or larger religious roles or
frames of meaning but tend to engage realms that transcend individual ritual
contexts. Contemporary scholarship pertaining to both ritual and women’s
religiosity provides support for such a premise. Recent work on ritual practice
emphasizes its nature as a type of performance that is inherently constructive
and strategic, producing specific types of meaning and values through par-
ticular strategies. Because it is constructive and strategic, ritual practice facili-
tates the ability of ritual agents to appropriate or reshape values and ideals that
help to mold social identity (Bell 1997, 82, 73). Elizabeth Collins observes that
thinking of ritual as performance also requires greater sophistication in think-
ing through issues of agency. Collins notes:
The model of performance implies several different agents

and different kinds of agency. There is the agency of the author of
the text, but also the agency of the performers who choose to per-
form a particular ritual or a particular variant of a ritual text and who
introduction 5
may even revise the text or tradition in their performance. There

is [also] the agency of those who participate as audience. (Collins
1997, 183–184)
In different contexts, women may function as ritual authors, performers, au-

dience, or any combination of these, and the chapters in this book collectively
address women’s ritual agency in all of these ways.
People who participate in ritual practices are also embedded in larger
communities that maintain particular social norms and values. When individ-
uals engage in ritual performances as agents, their engagement may function
to help produce, reproduce, transform, resist, or even defy these larger norms.
The constructive nature of ritual in this regard extends to ideologies and prac-
tices that concern gender. Judith Butler has argued that gender is primarily
performative and that gender identity is constituted through a stylized repe-
tition of acts (Butler 1990, 171–180). Religious practice, which certainly entails
a stylized repetition of acts executed in a ritual arena, is, in this regard, clearly
an engendering process. Lesley Northup notes that rituals are, at least in part,
‘‘constitutive of persons,’’ helping to construct, enhance, and enable person-
hood (Northup 1997, 87). Ritual engenders through the performance of ac-
tions that help to produce identity, including gender identity, in relation to
predominant social norms in ways that may be compliant, resistant, or both,
complexly and simultaneously. In her work on Hindu women’s domestic rit-
uals in South India, for example, Mary Hancock explores ways that domestic
ritual practice acts simultaneously as a site for both ‘‘reproduction of and
resistance to hegemonic images of female subjectivity’’ in Sanskritic Hinduism
(Hancock 1999, 32, 137).
Recent work on women’s religiosity further corroborates an emphasis for
women in particular, including Hindu women, on the continuity between reli-
gious and social domains, affirming that women appropriate religion in ways
that tend to engage women’s gender-specific social roles, experiences, and val-
ues. Meena Khandelwal, for example, notes that female renunciants in the
Hindu tradition tend to identify themselves as mothers, emphasizing the moral
and spiritual strengths particular to women in the mothering role and adapting
the term Ma (‘‘mother’’) as a form of address (Khandelwal 2004, 184–185). Susan
Sered (1992, 1994) observes that women often personalize religion, emphasiz-
ing practices and symbols that give spiritual meaning to women’s everyday lives.
In speaking of the relationship between religion and other dimensions of
women’s lives, Marjorie Proctor-Smith distinguishes between what she calls the
emancipatory function of religion (including ritual practices), where religion
may help women to transcend existing social restraints and behave in ways
contrary to social expectation, and a sacralizing function, where religion may
serve to establish women’s traditional roles and experiences as sacred (1993, 25–
28). While the sacralizing function can serve to justify traditions that limit
6 introduction
women’s power and freedom in both public and private spheres, at its best it
may serve to reveal ‘‘the dignity and holiness of women’s work’’ (28) and to en-
hance women’s self-esteem and feelings of self-worth.
In engaging Bhagavanti’s words as our point of departure, it is important
to acknowledge one important way in which the chapters in this book may
depart from the assumptions that Bhagavanti seems to bring to the table: the
chapters do not necessarily affirm a straightforward correlation between wom-
en’s daily work in the domestic arena and women’s religious practices. Instead,
we interrogate the relationship of women’s ritual activities to normative do-
mesticity, exposing and exploring the nuances, complexities, and limits of this
relationship. Here we understand domesticity in terms that are primarily
spatial and relational. We take ‘‘the domestic’’ to refer to a place, a domicile or
home, including the activities that occur therein and the kinship relations
among particular individuals. In many cultural and historical contexts, in-
cluding contemporary India, women’s everyday lives tend to revolve heavily
around domestic and interpersonal concerns, especially care for children, the
home, husbands, and other relatives; hence, women’s religiosity also tends to
emphasize the domestic realm and the relationships most central to women.
Within the study of Hinduism, work by Susan Wadley, Mary McGee, and Anne
Pearson has elucidated the concern for familial relationships and domestic
well-being that permeates some women’s rituals, especially votive rituals
(Wadley 1989; McGee 1987, 1991; Pearson 1996). But these scholars also
remind us that such rituals involve other goals and concerns—spiritual lib-
eration, for example. Furthermore, even the domestic religious activities that
Hindu women perform may not merely replicate or affirm traditionally for-
mulated domestic ideals; rather, these activities may function strategically
to reconfigure, reinterpret, criticize, or even reject such ideals.
The contributions in this book form a collective commentary on the rela-
tionship of Hindu women’s ritual practices to the domestic arena in particular.
Women may challenge normative domesticity by reinterpreting, expanding, or
rejecting prescribed spatial restrictions, domestic practices, ideologies of kin-
ship, or familial expectations. Traditional domesticity intersects with women’s
lives ambiguously, providing freedoms as well as constraints, danger as well
as protection, devalued as well as heightened status; women’s challenges to
normative domesticity may be similarly ambiguous.1 We emphasize female
innovation and agency in constituting and transforming both ritual and the
domestic realm, and we call attention to the limitations of normative domes-
ticity as a category relevant to at least some forms of Hindu women’s religious
practice.
Many chapters in this volume also consider the relationship between Hindu
women’s ritual practices and political, religious, or sociocultural concerns and
values beyond the domestic. Some, for example, explore larger questions of
power and women’s empowerment, asking whether women’s rituals empower
introduction 7
women in Hindu society and, if so, how (e.g., Erndl, McDaniel). Several con-
sider ways that women’s ritual practices engage cultural notions of beauty and
artistic ideals or, conversely, ascetic ideals that are often relegated to the male
sphere (e.g., Nagarajan, Narayanan, Craddock, Orr). Other chapters explore ways
that women’s religious practices may engage ethical, devotional, or social val-
ues that diverge from or even challenge those more central to Brahmanical
traditions and institutions (e.g., Harlan, Flueckiger, Pintchman). And one
chapter (Patton) addresses how Sanskritic knowledge might be transformed by
women from within. No matter what particular issues they address, however,
all of the chapters sustain a focus on the fundamental theme of this volume: the
relationship between specific forms of Hindu women’s religious praxis and the
lives of women outside of these ritual contexts. This is the essential axis around
which every chapter in the book revolves.
Notes on the Scholarly and Historical Context for This Study
In her edited book Roles and Rituals for Hindu Women, Julia Leslie noted the
challenge faced by contemporary scholars of Hinduism in ‘‘hearing the voices
of women’’ in our research on Hindu traditions. Leslie argued that this chal-
lenge means ‘‘not only seeking out the voices of women but also hearing their
own evaluations’’ of the traditions in which they participate (Leslie 1991, 3). In
the early 1990s, this was something of a new challenge; the relatively small
amount of Western scholarship that had been produced about Hindu women
in earlier decades tended to focus more on what authoritative Hindu texts said
about Hindu women than on what Hindu women themselves said and did
within the context of their own religious values and convictions. In the last
decade and a half, however, a number of works have been published that are
perhaps more apt to meet the kind of challenge Leslie put before us in her
pathbreaking book. These include works like Gloria Raheja and Ann Gold’s
Listen to the Heron’s Words (1994), Meena Khandelwal’s Women in Ochre Robes
(2004), Anne Pearson’s ‘‘Because It Gives Me Peace of Mind’’ (1996), and sev-
eral of the essays in Laurie L. Patton’s edited book, Jewels of Authority (2002).
Hearing the voices of women certainly seems to be a great deal easier for
those of us interested in the present than it is for those whose scholarly in-
terests focus on the historical past. Through ethnography, we can observe the
practices of contemporary Hindu women, ask them to talk to us, and solicit
their opinions and interpretations regarding ritual performance. The historical
record gives us rather less to go by, since authoritative Hindu texts are pre-
dominantly male-authored. Yet even the male-authored texts clearly reveal that
Hindu women have always been active, engaged ritual actors, and sophisticated
readings of the historical record can tell us a great deal about their involvement
in religious practice, even when we are unable to actually hear the voices
8 introduction
of the women themselves in these texts. Here I might mention, as examples,

Stephanie Jamison’s Sacrificed Wife/Sacrificer’s Wife (1996), which does a mas-
terful job of reading and interpreting the role of the sacrificer’s wife in the
practice of ancient Vedic ritual, or Mary McGee’s work on ancient mimamsaka
debates concerning women’s eligibility to perform Vedic sacrifices (McGee
2002). Furthermore, historically we do have a limited number of texts that
were authored by women, and these serve as rich resources for those interested
in the religious lives of Hindu women in past centuries, including their par-
ticipation in ritual performance. Eleanor Zelliot (2000), for example, engages
the recorded songs of female saints in medieval Maharashtra to explore ten-
sions that emerge in these songs between the performance of household duties
and the performance of meditation. Women’s agency and active participation
in various forms of Hindu ritual practice is clearly evident in a number of ways
throughout the history of the Indian subcontinent.
The contributions in this book are indebted to the existing body of schol-
arship on the religious lives of Hindu women, past and present. We hope to move
the conversation forward through our specific interrogation of the relationship
between women’s ritual practices and women’s lives outside of ritual practice.
Organization of the Volume
This book is divided into two parts: ‘‘Engaging Domesticity’’ and ‘‘Beyond
Domesticity.’’ The first part consists of five chapters that engage domestic and
interpersonal values in relation to women’s ritual practices that tend to ex-
pand the boundaries of normative domesticity. Laurie L. Patton’s ‘‘The Cat in
the Courtyard: The Performance of Sanskrit and the Religious Experience of
Women,’’ for example, demonstrates how contemporary female Sanskritists
in Maharashtra reconfigure Sanskrit, the ‘‘father language’’ of Brahminical
Hinduism, as a ‘‘grandmother language’’ that women engage in new ways to
imbue the practices and activities of everyday life with religious meaning. Fe-
male Sanskritists unite their stridharma, or ritual duties as women, with the use
of Sanskrit in everyday life, importing Sanskrit into mundane events and
practices like childbirth, daily food preparation, and the feeding of loved ones.
In so doing, these women reconstitute Sanskrit as a domestic language of inter-
personal care and personal transformation; they also reconfigure the ‘‘profane’’
moments of everyday life, imbuing them with religious meaning by sacralizing
them with powerful religious mantras.
In the second chapter, ‘‘Wandering from ‘Hills to Valleys’ with the God-
dess: Protection and Freedom in the Matamma Tradition of Andhra,’’ Joyce
Burkhalter Flueckiger explores the ways that South Indian Hindu women also
expand the boundaries of domesticity, this time through a ritual alliance forged
between women and the goddess Gangamma. In many Indian contexts, mar-
introduction 9
riage is understood to be the quintessential domestic institution, serving most

often to curtail significantly women’s freedom and agency in the public sphere.
Flueckiger explores a form of marriage that is socially liberating for Hindu
women—a form of ritual marriage that women may enact with Gangamma.
When illness strikes in villages around the pilgrimage town of Tirupati in
Andhra Pradesh, particularly illnesses of poxes, rashes, and high fevers associ-
ated with the hot season and particular village goddesses, little girls may be of-
fered to the goddess Gangamma (or one of her sisters) in exchange for the
latter’s protection and healing. When these girls reach puberty, they exchange
talis (wedding necklaces) with the goddess and are considered married to her.
The girls are called matammas in Telugu. Women who have been pledged in
marital alliance to Gangamma serve the goddess first and hence are subject to
male control only secondarily; since their primary marital allegiance is to a di-
vine female, not a human male, these women are not bound by the same social
rules as are other married women. Hence, they are able to move in and through
public space and claim a place in the public sphere much more easily than are
most married women. Their alliance with the goddess, formalized by the rit-
ualized exchange of talis, affords these women protection, freedom of move-
ment, and agency outside the domestic sphere.
Like Flueckiger’s, my chapter, ‘‘Lovesick Gopi or Woman’s Best Friend?
The Mythic Sakhi and Ritual Friendships among Women in Benares,’’ engages
a form of religious practice that functions to establish an ongoing relationship,
one that continues beyond the ritual performance itself. The ritually estab-
lished relationships we both explore, furthermore, share common ground in
that they mimic human marriage but function as lifelong bonds between two
females. In the materials highlighted in my chapter, however, the bond is not
between a woman and a goddess but between human women who establish
ritual friendships. I focus on a tradition undertaken by women living in Be-
nares, Uttar Pradesh (North India), in which the gopis, the famous cowherdes-
ses of Krishna mythology, become the model for ritually based human female
friendships. The gopis are also referred to as the sakhis, or ‘‘female friends,’’ and
in ordinary speech the term sakhi is used to refer commonly to a girl’s or
woman’s female friend. The tradition I explore involves a ritual process of ‘‘be-
coming’’ or ‘‘tying’’ sakhi, which entails pledging lifelong friendship. For many
Benarsi women, the sakhi relationship represents a female-female union that
imitates the marital bond but may surpass blood or marital kinship bonds in
terms of its professed meaningfulness in women’s lives. The appropriation of
the sakhi figure in this ritual tradition also demonstrates how women have
adapted Krishna traditions in ways that engage their own interpersonal con-
cerns and values.
The fourth chapter, Lindsey Harlan’s ‘‘Words That Breach Walls: Women’s
Rituals in Rajasthan,’’ explores women’s ritual practices in Rajasthan, North
India. Harlan echoes Flueckiger in emphasizing ways that ritual functions to
10 introduction
facilitate women’s movement across domestic boundaries and among various

spaces beyond individual domiciles. The materials Harlan engages, however,
move us in a different direction, emphasizing the permeability of domestic
space, the flexibility of the category ‘‘domestic,’’ and the continuity between do-
micile and extra-domestic spaces. Harlan explores Rajput women’s performance
of ratijagas, women’s ‘‘wake’’ rituals, which women perform for weddings and
other important occasions. Ratijagas are domestic ritual performances inasmuch
as they occur in household spaces. But Harlan uses her exploration of ratijagas
to interrogate and render problematic any simple distinction between public and
private domains when it comes to women’s ritual performance. As travelers to
other households for ratijaga performances, for example, women may transform
the environments from which they come and to which they travel; what goes on
in one household may affect processes in other households. Harlan engages the
popular film Chocolat as a comparison, demonstrating that the dynamics she
highlights so beautifully in the materials she explores are not specifically Hindu
or Indian but may be seen as broader insights about women’s potential influence
on religious praxis.
The fifth chapter, Vijaya Rettakudi Nagarajan’s ‘‘Threshold Designs, Fore-
head Dots, and Menstruation Rituals: Exploring Time and Space in Tamil
Kolams,’’ brings us back to South India, this time to Tamilnadu. Like Harlan,
Nagarajan explores the ways that women’s religious practices cross bound-
aries and traverse thresholds, and she also emphasizes continuity and overlap
between private and public domains. The focus of this chapter is the rela-
tionship between kolams, auspicious designs that women create daily at their
domiciles’ thresholds, and pottus, the auspicious red dots that adorn Tamil
women’s foreheads. Nagarajan argues that the kolam and the pottu are parallel
ritual expressions that embody larger Hindu cultural values, especially aus-
piciousness and inauspiciousness, purity and pollution. Both kolam and pottu
mark thresholds, those of the home and the body, and function to mark spatial
and temporal transformations: from auspicious to inauspicious times or pure
to impure ones, as in the erasure of the pottu and the absence of kolam pro-
duction during menstruation and their reappearance following the period of
menses. Pottu and kolam both embody the status of married women as aus-
picious householders, a status that is rooted in their domestic location, but
both send that auspiciousness forth beyond the domestic threshold into the
larger communities in which female Hindu householders are embedded.
Together, these first five chapters demonstrate ways that women exercise
agency in religious performance as ritual authors, performers, and consum-
ers. The practices these chapters explore expand notions of what constitutes
domesticity in Hindu women’s ritual traditions, and they dissolve conceptual
boundaries that serve to divide the sacred from the profane and domestic spaces
and performances from public ones, demonstrating the ways that women’s
ritual practices, and the women who engage in them, cross thresholds,
introduction 11
traverse boundaries, reposition or reinterpret larger cultural formations and

social norms, and provide room for relationships beyond the terms dictated by
predominant patriarchal family norms.
The five chapters in part II, ‘‘Beyond Domesticity,’’ similarly reveal the
many ways that women’s religious performances permeate diverse realms
and breach borders. These chapters collectively take up a somewhat different
challenge, however, exploring women’s ritual practices outside the confines of
strictly domestic contexts and contesting the impulse to link women’s ritual
performance primarily with domestic realms and concerns.
Leslie C. Orr’s chapter, ‘‘Domesticity and Difference/Women and Men:
Religious Life in Medieval Tamilnadu,’’ questions any special emphasis on do-
mestic values in evidence pertaining to women’s ritual practices in medieval
Tamilnadu. Orr compares the practices and goals of women and men in me-
dieval South India, particularly with respect to gift giving and vow taking, as
these are revealed in Tamil inscriptions engraved on the stone walls of Jain and
Hindu temples of the ninth through thirteenth centuries. Women seem to have
been especially involved in the sponsorship of images, notably images of god-
desses, to be worshipped, and they used their gifts to forge links with family
members. Somewhat different patterns are seen in the case of men’s donations,
but they also demonstrate familial involvement in, for example, men’s gifts of
thanksgiving for the births of sons or gifts for the merit of their fathers.
Orr observes, perhaps surprisingly, that ‘‘domestic’’ motives are not in
evidence in the few records we have of women’s vows and self-immolation,
which do not indicate that these acts were performed for the sake of offspring
or a husband. There seems to be nothing particularly feminine about such acts,
which appear to be expressions of values of valor and loyalty shared with men
and akin to vows of political allegiance, whose character as ‘‘religious’’ under-
takings is questionable. Orr argues that men’s and women’s activities recorded
on temple walls did indeed have distinctive colorings, but that the contexts,
roles, and motives for these actions were overlapping and often congruent.
Men expressed ‘‘domestic’’ concerns through their ritual activity, and women,
through theirs, not only participated in the public religious space of the temple,
but also shared in wider social and political worlds that transcended both the
domestic and the religious. Orr concludes her chapter by questioning the con-
ceptual boundaries that divide religion from political and economic realms,
domestic ritual from ‘‘high’’ public ritual, and interpersonal religious aims
from ultimate, transcendent ones in relation to the materials she explores.
Elaine Craddock also questions these kinds of conceptual boundaries
in her chapter, ‘‘The Anatomy of Devotion: The Life and Poetry of Karaikkal
Ammaiyar.’’ Karaikkal Ammaiyar, a.k.a. Punitavati, was one of the Tamil
nayanmar, or Shaivite saints. Before Punitavati became Karaikkal Ammaiyar,
she was married to a merchant, but her ardent devotion to Shiva conflicted
with her ritual duties as a wife. Her husband became frightened by Shiva’s
12 introduction
response to her devotions and released her from marriage; she immediately
made a devotional pilgrimage to the Himalayas, where Shiva granted her wish
to obtain a demon form and to be the eternal witness to his fierce dance in the
cremation ground at Tiruvalankatu. There, she composed 143 verses, which
represent the earliest Tamil poetry to Shiva. Some of these poems describe the
cremation ground as a scene of ghoulish domesticity, relocating the center of
existence in Shiva’s presence: the site of death and burning flesh is reposi-
tioned as the domicile, and ‘‘home’’ is redefined as anyplace where one dwells
in Shiva’s presence. Karaikkal Ammaiyar’s own life embodies a shift in em-
phasis from the performance of wifely domestic rituals, normally the primary
ritual domain of married women, to the understanding of her entire life as a
ritual offering to Shiva. This move from domestic practice to devotional prac-
tice as the central sphere of ritual concern pushes both women and men to see
ordinary household rituals as meaningless and to transcend values like hu-
man beauty, promoting the view that only a life lived entirely as a ritual offering
to Shiva has meaning. Like Orr, Craddock stresses the importance of religious
values other than domesticity in women’s ritual performance; in the materials
Craddock explores, however, there is a self-conscious appropriation and re-
positioning of ‘‘the domestic’’ as a suitable site for women’s religious per-
formance.
The eighth chapter, Kathleen M. Erndl’s ‘‘The Play of the Mother: Pos-
session and Power in Hindu Women’s Goddess Rituals,’’ explores questions
of Hindu women’s power in connection with goddess possession rituals in the
Kangra Valley area of Himachal Pradesh, North India. In Kangra, as in many
other regions of India, it is not uncommon for women to become possessed
by a goddess, to speak with her voice, and to act as healers and mediums in their
communities. Divine possession as a form of religious expression is inter-
connected with such practices as pilgrimage to temples, puja (image worship),
recitation of sacred texts, fasting, and meditation, practices that comprise the
religious complex of Shaktism or goddess worship in the region. Erndl argues
that possession practices can be a source of both religious and social empow-
erment for women, providing opportunities for women to improve their lives
and to help other women. Like Harlan, Erndl stresses the mobility that reli-
gious practice, in this case possession performance, affords, granting house-
holder women in particular opportunities to travel beyond their domiciles and
to form with other women a female community, however temporary, which in
turn may provide women access to advice, support, or even material assistance.
She describes these ritual spaces as ‘‘cracks’’ in a patriarchal system that cannot
be completely controlled by patriarchal norms and that provide sites for wom-
en’s creativity and interconnection.
Erndl notes that goddess-possessed women, or matajis, may be married or
single, householders or renunciants. The same is true of the women whom June
McDaniel explores in her chapter, ‘‘Does Tantric Ritual Empower Women?
introduction 13
Renunciation and Domesticity among Female Bengali Tantrikas.’’ McDaniel,

however, emphasizes renunciation, not domesticity, as the preferred locus of
women’s religious expression in the materials she explores, namely, the ritual
practices of women active in tantric traditions in West Bengal. These female
tantrikas tend to reject the traditional domestic values associated with wom-
en’s householder rituals, emphasizing instead ritual practices allied more
closely with renunciation and its goals. In this context, renunciation is highly
valued in women, as it is in men, while sexuality returns a woman to the
sphere of traditional domesticity, where she takes on the role of supporter and
helper of a man rather than the role of individual seeker. Hence, the world of
female tantric ritual challenges the connection between domesticity and
women’s ritual practices predominant in other contexts. Like Orr and Crad-
dock, McDaniel demonstrates that women’s religious practices may engage
ascetic goals and activities as fully as or even more fully than domestic ones,
sometimes rejecting domesticity or subordinating the domestic realm of
householder life to the goals associated with religious renunciation and celi-
bacy.
The final chapter, by Vasudha Narayanan, ‘‘Performing Arts, Re-forming
Rituals: Women and Social Change in South India,’’ addresses the expression
of Hindu women’s religiosity through music and dance. Narayanan observes
that in contemporary Hinduism the performing arts, which she argues are
essentially forms of religious performance, may serve as vehicles not only for
women’s religious expression, but also for dynamic social commentary and
reform. Narayanan points to dancers like Mallika Sarabhai and Chandralekha,
who use dance to highlight women’s issues and to express themes of anguish
and strength, as examples. As authors, performers, and consumers of the per-
forming arts, women may engage music and dance both to express their own
subjectivities and to help effect social change.
Collins articulates two distinctive approaches to the study of ritual: one
that emphasizes what ritual does to people and another that emphasizes what
people do with ritual (Collins 1997, 17). The first approach elicits a herme-
neutics of suspicion, seeking to elucidate ways that ritual practices affirm and
reproduce larger relations of social power, often without the conscious assent
of ritual actors. The second approach emphasizes instead the ways that people
use ritual forms to pursue their own individual and collective interests, appro-
priating and sometimes modifying rituals when convenient (178). While the
first approach stresses the nature of ritual actors as (often unwitting) recipients
of larger ideological and hegemonic structures, the second stresses their na-
ture as agents who may creatively deploy ritual for their own purposes. Both
approaches certainly have a role to play in shedding light on the nature of
ritual practice. But the chapters in this volume overwhelmingly stress the sec-
ond approach, highlighting what Hindu women do with their rituals to shape
their worlds as agents acting in pursuit of their own desired ends. The women
14 introduction
whose traditions and lives we explore in this book exercise initiative, inge-
nuity, and resolve in creating and sustaining female-centered traditions and
practices that are uniquely meaningful to women’s experience in Hindu
culture. This volume as a whole suggests that those interested in Hindu wom-
en’s ritual practices would do well to attend fully to women’s ritual creativity
and to the religious concerns of women beyond the sphere of conventional
domesticity.
note
1. Many thanks to Corinne Dempsey, whose commentary on and description
of ‘‘domesticity’’ I am borrowing here from her helpful manuscript review.
references
Bell, Catherine. 1997. Ritual: Perspectives and Dimensions. New York: Oxford Uni-
versity Press.
Butler, Judith. 1990. Gender Trouble: Feminism and the Subversion of Identity. New
York: Routledge.
Collins, Elizabeth Fuller. 1997. Pierced by Murugan’s Lance: Ritual, Power, and
Moral Redemption among Malaysian Hindus. DeKalb: Northern Illinois University
Press.
Hancock, Mary Elizabeth. 1999. Womanhood in the Making: Domestic Ritual and
Public Culture in Urban South India. Boulder, CO: Westview.
Jamison, Stephanie. 1996. Sacrificed Wife/Sacrificer’s Wife: Women, Ritual, and
Hospitality in Ancient India. New York: Oxford University Press.
Khandelwal, Meena. 2004. Women in Ochre Robes: Gendering Hindu Renunciation.
Albany: State University of New York Press.
Leslie, Julia, ed. 1991. Roles and Rituals for Hindu Women. Rutherford, NJ: Fairleigh
Dickinson University Press.
McGee, Mary. 1987. ‘‘Feasting and Fasting: The Vrata Tradition and Its Signifi-
cance for Hindu Women.’’ Th.D. diss., Harvard University Divinity School.
———. 1991. ‘‘Desired Fruits: Motive and Intention in the Votive Rites of Hindu
Women.’’ In Roles and Rituals for Hindu Women, ed. Julia Leslie, 71–88.
London: Pinter.
———. 2002. ‘‘Ritual Rights: The Gender Implications of Adhikara.’’ In Jewels of
Authority: Women and Textual Tradition in Hindu India, ed. Laurie Patton, 32–50.
New York: Oxford University Press.
Northup, Lesley A. 1997. Ritualizing Women. Cleveland, OH: Pilgrim.
Patton, Laurie. 2002. Jewels of Authority: Women and Textual Tradition in Hindu India.
Pearson, Anne Mackenzie. 1996. ‘‘Because It Gives Me Peace of Mind’’: Ritual
Fasts in the Religious Lives of Hindu Women. Albany: State University of New York
Press.
introduction 15
Pintchman, Tracy. 2005. Guests at God’s Wedding: Celebrating Kartik among the Women
of Benares. Albany: State University of New York Press.
Proctor-Smith, Marjorie. 1993. ‘‘‘In the Line of the Female’: Shakerism and
Feminism.’’ In Women’s Leadership in Marginal Religions: Explorations Out-
side the Mainstream, ed. Catherine Wessinger, 23–40. Chicago: University of
Illinois Press.
Raheja, Gloria Goodwin, and Ann Grodzins Gold. 1994. Listen to the Heron’s
Words: Reimagining Gender and Kinship in North India. Berkeley: University of
California Press.
Sered, Susan Starr. 1992. Women as Ritual Experts. New York: Oxford University
Press.
———. 1994. Priestess, Mother, Sacred Sister: Religions Dominated by Women. New
York: Oxford University Press.
Wadley, Susan. 1989. ‘‘Hindu Women’s Family and Household Rites in a North
Indian Village.’’ In Unspoken Worlds: Women’s Religious Lives in Non-Western
Cultures, ed. Nancy A. Falk and Rita M. Gross, 94–109. San Francisco: Harper
and Row.
Zelliot, Eleanor. 2000. ‘‘Women Saints in Medieval Maharashtra.’’ In Faces of the
Feminine in Ancient, Medieval, and Modern India, ed. Mandakranta Bose,
192–200. New York: Oxford University Press.
part i
Engaging Domesticity
1
The Cat in the Courtyard:
The Performance of
Sanskrit and the Religious
Experience of Women
Laurie L. Patton
Will the nature of Sanskrit and Hindu religious experience change

as a result of its being increasingly in the hands of women? One
young woman Sanskritist, Pradnya Deshpande, makes the following
comparison:
Sanskrit is difficult, but it is also like prasadam. When
someone offers you prasadam, you have to eat it, even when
there is a stone in it. Modern people reject the stone—they
say, ‘‘There’s a stone in this!’’ And throw it away. It is also
like a cat: When the cat comes into the courtyard, the older
women will call it bhau-ji [brother-in-law]. Even the cat is
bhau-ji, like a brother-in-law, in that house. But the mod-
ern girl will say, ‘‘Go away, cat!’’ In this way, the modern
researchers can accept some doubt about the tradition. They
can be critical. Now these teachers still love Sanskrit, and
they love anyone who is learning Sanskrit.
Background
This young woman’s comments indicate a certain domestic wel-

coming of the opportunity to learn and study Sanskrit, even if it comes
in the less pleasant form of a cat. Before I explore this idea in more
detail, let me begin with some background. What is the situation
of Sanskrit education in contemporary India? In his recent volume
20 engaging domesticity
on the role of the pandita, Alex Michaels (2001) outlines the dual education
system in contemporary India in which traditional teachers find themselves
caught: the university system, based loosely on the English model of govern-
ments, schools, and universities, and the pathshalas and samskrita-vidyapithas,
where Sanskrit is taught according to the traditional methods. There, the guru-
shishya sambandha, or relationship between teacher and student, is the primary
model, where the teacher stands for wisdom, memory, and personal and moral
guidance. Michaels mentions the various ways in which, after the publication
of the report of the government of India’s Sanskrit Commission (1958), various
agencies have been implementing its recommendations—especially the
Rashtriya Sanskrit Sansthan, formed in 1970.
Despite dire predictions, Sanskrit has hung on. According to one report,
almost all of the recommendations of the Sanskrit commission have been im-
plemented; there are now 4,000 pathshalas funded by the Rashtriya Sanskrit
Sansthan, as well as many other vidyapithas and Sanskrit colleges, which are
funded independently or by local communities and temples. The study of San-
skrit in secondary schools has also been a major priority, with more limited
success because of the other options for language study (English, regional lan-
guages, Hindi) recommended by state governments in India. The Bharatiya
Janata Party (BJP) government’s support for Sanskrit study in the 1990s was
stronger than that of previous governments, but there was little improvement in
the overall system of traditional education as India increasingly competed on the
global stage in technology, science, and engineering. In the English university
system, Sanskrit has become a ‘‘humanities’’ subject, with less qualified students,
or students with a lesser need for high income, flocking to the registration desks.
While several other scholars1 have recently commented on the dual edu-
cation system and its effect on the study of Sanskrit in India, one crucial and
overlooked element in this system is gender. Recent academic work is com-
mendable indeed for keeping the study of Sanskrit alive in the scholarly imag-
ination; however, this work has totally ignored the role of women in this sea
change in the study of Sanskrit. In many parts of India, the dual education
system is also clearly a dual gender system.
Let me be more specific. In postcolonial India, Sanskrit has become a marker
of the Hindu religiosity of women as well as men. In certain places, if the trend
continues, it will soon become entirely the prerogative of women. With the
massive entry of men into the fields of science, technology, and engineering, this
change has happened without the help of postcolonial theory or secular femi-
nism, either Indian or Western. It will continue without that help. My larger book
project, Grandmother Language, from which this chapter is derived, is a study of
women Sanskritists through their personal narratives. Its chapters will comprise
an examination of their lives, their religious commitments and practices, and
their understandings of their roles as teachers and scholars. Such change is only
the cat in the courtyard 21
possible with an unlikely amalgamation of factors: traditional Hindu ideologies

of gender combine with a historical emphasis on women’s educational reform in
Maharashtra to create a unique environment for innovation. This combination
makes it possible for women to take on new roles as caretakers of a classical
language which has been prohibited to them for millennia.
Let me begin with some facts about Maharashtra, the area where I did my
research. In Maharashtra, Sanskrit is still alive and well within the educational
system: there are eight major independent research institutes in Pune and
Bombay; six universities that offer degrees up to the master’s and doctoral
levels; thirty-three major Sanskrit manuscript collections; and nine Indological
journals published in the state. The personnel needed to maintain this large
educational tradition is extensive.
The bulk of that personnel is increasingly composed of women. With one
retirement, the University of Pune’s Department of Sanskrit will consist en-
tirely of women. The ratio of male to female students registered for the M.A.
degree in Sanskrit in 2004 was one to six. In the Deccan College Dictionary
Project, Pune, there are seven women and two men on the regular research
staff. Bhandarkar Oriental Research institute has 50 percent women research-
ers on its staff. The number of stri-purohits, or women ritual specialists, is
growing rapidly, and, according to one report by V. L. Manjul, women ritual
specialists now outnumber their male counterparts in certain neighborhoods
in Pune. On a nationwide scale, the chairs of six of the major universities—
Delhi, Madras, Nagpur, Pune, Calcutta, and Hyderabad—are women.2
I have completed eighty oral life narratives of women in the field of Sanskrit
in Maharashtra, using as my starting point (and my starting point only) a
questionnaire with ten questions about family life, educational experience, and
vision for the future. The conversations were long and meandering, lasting
about two to three hours each, and longer if necessary. They were conducted
mostly in my native language of English, about eight in conversational Hindi,
and some with occasional short exchanges in conversational Sanskrit. My col-
laborator in this project, Maitreyee Deshpande, also conducted several interviews
in Marathi. She was able to clarify questions in Marathi during our conversa-
tions as well. All of the conversations except eleven were recorded.
Perhaps a story from my fieldwork best sums up the tensions that we
might see in the coming years, as the dual education system unfolds. One
young woman in my study, who requested not to be named, was part of a team
that hired two young traditional panditas to teach at a major Indian univer-
sity. The students wanted to get exposure to this method of learning, and the
local pathshala was happy to help. One of the young panditas approached her,
red-faced, and asked embarrassedly whether the young women he would be
teaching could absent themselves during their menstrual periods so that he
would be spared the problems of impurity that might result. She replied:
Sir, perhaps I might ask you to consider where our respective

institutions might be in fifty years. I can as much as guarantee you
that my university will be here and standing, and probably teach-
ing Sanskrit then. However, neither I nor you can guarantee that
your pathshala will survive that long. Given that this is the case,
I suggest you comply with the university system, where no such
concerns about monthly cycles apply.
Definitions
So much, then, for the basic background. While there are many other aspects
of this change I will explore in my book, such as the role of caste, region, edu-
cational institution, and postcolonial status of Sanskrit, the purpose of this
chapter is to explore only one small aspect of this rather large change: the more
explicit inclusion of Sanskrit utterance in ‘‘domestic’’ moments of personal
change and transformation. I want to formulate the issue in this way because I
want to nudge the basic categories in which we tend to view Sanskrit—as either
a dead language or an artificial language. In the case of Sanskrit as a dead
language, we tend to hear this from non-Indological colleagues who may not
be aware of the places it is still used, such as at academic conferences, in
villages in Karnataka and Himachal Pradesh, and during revivals of sacrificial
performances. The artificial language definition of Sanskrit is perhaps more
accurate, describing as it does the fact that Sanskrit remains an elite, father
language learned at one’s male teacher’s or father’s feet—the opposite of the
natural language learned at one’s mother’s knee. However, this definition
implies that Sanskrit has nothing to do with the household, with the world of
the domestic sphere, and with personal changes involving the heart as well as
the head.
My preliminary research results show that even such a characterization
of Sanskrit as an ‘‘artificial’’ language tends to be misleading. I would like to
spend the time remaining to me arguing that, in the hands of Brahmin as
well as some non-Brahmin women guardians, Sanskrit’s cultural placement is
shifting. Let me put my two basic points more accurately and specifically: first,
Sanskrit’s domestic usages, which have always been present in Hindu cul-
tures, are now even more pronounced. Second, Sanskrit’s relevance as a lan-
guage which can be used in everyday moments and in moments of personal
transition is increasing.
Let me begin by focusing on the contexts of Sanskrit and the ways in
which we might view Sanskrit as we already do, as a partly formal language,
but also as a partly proverbial language, one which depends upon informal
contexts and particularly defined moments for its utterance. Notice that both
of the characterizations of Sanskrit mentioned above, as either dead or arti-
ficial, are dependent upon performative context for their definitions. In the
case of a dead language, there is assumed to be no performative context, hence
the language’s morbidity. In the case of an artificial language, the performa-
tive context is not ‘‘natural,’’ but learned as a secondary and not a primary skill.
But the emergence of performance studies would insist that we be even more
specific and careful about such definitions, because language is performed
in a much wider variety of contexts than these definitions tend to suggest.
As Charles Briggs (1988: 372) puts it, ‘‘The emergence of contextual and
performance-based studies is crucial, since they point to the status of contex-
tual elements as central elements of the performance.’’3
Moreover, it is crucial to acknowledge, as linguists rarely do, the role of the
local influence of Sanskrit in places such as Maharashtra. As Dandekar has
noted in his excellent, but now somewhat dated, Sanskrit and Maharashtra
(1972),4 Sanskrit influenced and was influenced by local cultural movements
throughout the history of the region. Most notably for our purposes, the political
events of the late colonial and postcolonial periods provided ample opportu-
nity for creative work in Sanskrit. Maharashtrian authors K. V. Chitale and
V. Bagewadikar produced Sanskrit works on colonial figures such as Lokma-
nya Tilak, as well as Sanskrit biographies of Nehru, Gandhi, and the Free-
dom movement. Short story, poetry, and drama competitions have dotted
the region, beginning from the 1930s and onward to this day. Translations of
English works into Sanskrit by Marathi authors include Wordsworth (C. T.
Kenghe), Longfellow (G. B. Palsule), Keats (N. P. Gune), the Sermon on the
Mount (S. N. Tadpatrikar), and Goethe’s Faust (L. V. Deshpande). In 1961, A. R.
Ratnaparakhi composed the Samvadamala, which is a series of thirteen dia-
logues on daily subjects like breakfast, office, shopping, and so on, which
might be the precursor of the Speak Sanskrit movement of Krishnashastri,
called Samskriti Bharati today. While these examples may be well known to
people within scholarly or literary circles, their existence does not seem to have
affected the definition or treatment of Sanskrit in any demonstrable way until
recently.
One exception to this is Jan Houben’s edited volume, Ideology and Status
of Sanskrit (1996).5 There, Madhav Deshpande discusses the changing nature
of priestly ‘‘recited’’ Sanskrit, from the early treatment of changes in mantra
to the present-day usage of Sanskrit in American temples. Saroja Bhate writes
of a ‘‘renaissance’’ in Sanskrit literature in the same period mentioned above,
as well as its current role in public education and scientific research. Victor
van Bijlert comments on the role of Sanskrit in the Bengal intellectual for-
mation of nationalism. And Albrecht Wezler does a first pass through the
‘‘spoken Sanskrit’’ practices and their related grammars of the Middle Ages.
All of these perspectives bode well for the treatment of Sanskrit as a liv-
ing, breathing language with its own organic life, learned and used in a variety
of contexts.
Formal Language within the Household: Women

and the Flexibility of Sanskrit
It is a truism and fairly uninteresting to state the obvious—that Sanskrit is

performed regularly in household puja rituals throughout India, at the time
of major festivals such as Ganeshotsav, Sarasvati Puja, Krishna Jayanti,
Diwali, and many others. What is more interesting is the ways in which wom-
en’s increasing participation and leadership as guardians of mantric recita-
tions at these celebrations and their leadership in Sanskrit instruction within
the home are changing our understanding of why and how they are performed.
The stereotypes have been that Sanskrit japa, or low-toned recitation,
lends a certain inflexibility and even inaccessibility to even domestic ritual
performances. Kumud Pawde (1992), a Dalit Sanskritist at the University of
Nagpur, describes her Brahmin neighbors as ‘‘The Splendid People,’’ whose
performance of Sanskrit in their home rendered them removed from their
Dalit neighbors—cleaner, more learned, and ‘‘shining’’ in a way that she could
never emulate.6 This view is clear in a near-endless number of examples, from
the early sixth-century bhakti poets (see Ramanujan, 1973) to A. K. Murthy’s
now-classic novel Samskara (1978).7 The performance of japa is sometimes
even beyond the performer’s comprehension and is invested with an aura of
elitism and oppression, combined with the formal exactitude of sound.
Yet there is another side to ritual performance—one in which, even in
highly formal situations, actors react to and improvise on their surroundings
in subtle ways. To put it theoretically, performance studies has suggested that
ritual worlds are created worlds governed by roles and instruments, and there-
fore there is a higher likelihood of ritual actors using pragmatic forms of com-
munication and metonymically referring to and identifying with those roles
and instruments. Many performance theorists8 (Tedlock, 1993; Tedlock and
Mannheim, 1995; Gill, 1987; Laderman, 1991; Driver, 1993; Mudimbe, 1997;
Spiziri, 1997; Grimes, 1990, 1995, 2000) have examined the linguistic aspects
of performance in ethnographic contexts similar to the highly structured world
of domestic puja ritual.
Further, as these authors have argued, in these structured ritual contexts,
while the verbal ‘‘text’’ may be fixed, the world constructed around that text, and
the actors’ connection of that text to the world around them, does vary in fas-
cinating ways. Briggs uses the example of the Easter liturgy in Mexicano verbal
art to demonstrate this: while the prayers and the hymns are the same, the
volume, the emphasis, the processions, and the placement of the ritual vary
according to the status, age, and disposition of the person performing the text,
as well as more external factors such as the weather, the current political
climate of the village, and the priest in charge (1988: 289–340). Moreover, as
Briggs also notes, these ‘‘fixed’’ texts are as much about the emotions and
events they evoke as they are about the words contained in them.
Relatedly, in his article ‘‘Contextualizing the Eternal Language: Features of
Priestly Sanskrit,’’ Madhav Deshpande (1996)9 also speaks about the ways in
which ritual situations might change seemingly fixed ritual language. Some of
the changes have to do with the influence of the vernacular language, leading
to the term ‘‘vernacular Sanskrit.’’ Others have to do with the actual ritual sit-
uation in which participants find themselves, such as priests wishing to make
sure that the individual participants are acknowledged in American temples in
a particular way.
Domestic Worlds
My research to date suggests a similar richness and variety in women’s expe-

riences of the domestic performance of Sanskrit in the home—whether it
is through elaborate rituals or simple recitation. First, at a general level, it should
be noted that, in sixty-two out of the eighty interviews, the women reported that
it was the domestic use of Sanskrit in the childhood home that was in part
responsible for their continuing with the study of Sanskrit as a profession.
One woman recounts her childhood experience of learning Sanskrit by
having her mother teach her as she worked on the household chores: ‘‘Every
household item we were cleaning or dusting, she would give me the Sanskrit
name for it. In this way I learned simple Sanskrit words, and I began to
associate in my mind the work my mother did for all of us in the family with
the Sanskrit language. I still feel as [if ] I am doing some of that today.’’ A
younger woman said that her connection with her grandfather and his
kindness was integrally bound up with learning Sanskrit. ‘‘Since I was five we
had to recite chapters from the Gita, and the subhashitas. It was a compulsion
for us—not our wish—but we loved our grandfather because he never pun-
ished us. He died in 1986, and some spirit has carried on in my mind.’’
Second, and relatedly, for sixty-five out of the eighty women I interviewed,
Sanskrit is associated with respect and honor for women as well as for men in
the early family history. As one scholar, Mrs. Manik Thakar, told me, ‘‘Emi-
nent scholars like S. D. Joshi, and so on, came to see us. They were all a keen
group. I recited a few shlokas and felt very much a part of the group. I was
respected as a child, and what is more, my mother was interested in educating
girls.’’ Mrs. Shailaja Bapat, currently the head of the Sanskrit Department in
Pune, commented, ‘‘I learned it from puja only—in my childhood. In the
morning, the priest used to come for worship; he used to tell us how to wor-
ship. Also, with Ramaraksha mantra—our father and grandmothers used to
tell us how to recite everyday.’’ Relatedly, Ranjana Date tells of her teacher:
He used to teach only for boys, but by special tuition he taught

girls on Sundays. First girls’ batch he taught. The amount wasn’t
much—seven or eight rupees per month. But that was an age-old
matter of prestige, not of money for scholarship. He used to teach not
only Sanskrit, but chemistry and French. Our headmistress said,
‘‘OK, we’ll observe the rules of purity, so what harm is there in
learning?’’ He had many different sides to him. [My teacher] Bhide
taught us Panini [the Sanskrit grammarian]. We became familiar
with that, and then we learned Laghukaumudi. We would memorize
certain portions, and as we recited he would explain.
Third, and most important for our purposes, the women report their
memories of domestic Sanskrit rituals as filled with color, light, fascination,
and variation. More than half of those interviewed spoke of fascination for
the ritual use of Sanskrit as part of their ongoing interest in the field, and
forty-two mentioned morning pujas (both positively and negatively). One
very prominent Sanskritist spoke of morning pujas as the thing that ori-
ented her and grounded her as a child. Saroja Bhate commented, ‘‘Sanskrit,
one can do at home, through morning pujas, as well as being a teacher in a
college.’’ One woman reminisced that she became fascinated with Sanskrit
from her weekly and holiday trips to the temple. As she put it, her trips to
the family temple continued back inside the household: ‘‘The chanting from
the temple was so beautiful, and there was so much time we would spend
together with my aunts and uncles and cousins—the whole family. We would
continue the chanting from the temple back into our living room, and
even some of the tunes I would hear from my father in the puja room the next
morning too.’’
Several of the women spoke about negotiating with the priest in learning
how to do the pujas; for each of them, understanding Hindu practices was a
matter of trial and error, learning and perfecting. Mandakini Kinjavadekar had
very powerful memories of such learning:
I remember my father’s teacher, Pandit Khareshastri from Gokarna-
Mahabaleshwar. Ganesh-shastri Khare was the guru of my father.
While coming, he used to bring Konkan products—jackfruit cha-
pati, mango chapati. He was for us as a loving grandpa bringing
all these products. More than one year he lived with us, and he
was observing ‘‘cleanliness.’’ After a bath, nobody would touch
him, not speak, etc. Also we had to obey all these things. When my
father and his teacher were discussing all these things, we were
not supposed to talk or go there. The teacher was observing
no speaking, maunavrata. We felt that day very pure—that some-
thing pure is in our house.
None of these women actually performed the ritual in their childhoods, but it
was clear that Sanskrit was a part of their negotiations in their narratives of
childhood learning.
Most important, following Briggs’s insight above about the emotional
impact of a fixed text, the women reported that Sanskrit recitation was expe-
rienced in different ways in different climates. Forty-three of them mentioned
that, while reciting had no meaning for them initially, they were moved by the
beauty of Sanskrit to create a mood. As Saroja Bhate put it, ‘‘I remember the
Hindu vaidikas of my childhood, where pronunciation was down to the min-
ute details, in the exact and correct way. I was so impressed by the beauty of
the language.’’
Yet many of these same women also reported that Sanskrit verses—
sometimes even the same verses—created a mood of seriousness and respect
when publicly recited outside of the home. Mrs. Asha Gurjar actually reported
that her father did not allow her to recite the Gita because it seemed ‘‘too
big’’ for her, but over time she began to experience it as ‘‘smaller’’ and even-
tually learned how to recite it publicly. Mrs. Menakshi Kodnikar also con-
nected the work of the stripurohits, or women priests, with the work of women
Sanskritists:
There are a large number of women priests now—and they want to
do service, to be of service like we do. And so whatever branch of
knowledge they know, they perform this service for the family line.
Many women, they don’t take a degree. They teach more in the
household, and they perform service for the householders—Shrisukta
and things like that. And there are many ladies now doing this. La-
dies also teach puja-vidhi—the rules about worship.
Sanskrit as the Scriptural Utterance of Transformation
In my research with these women, the second arena of Sanskritic performance

had to do with its ability to transform a personal situation—a function I would
identify closely with Charles Briggs’s analyses of proverbial and scriptural
utterances. As Briggs writes of both genres:
The two genres [scriptural and proverbial] each revolve around the
stylistic and ideological intrusion of a third party into the speech
event; whereas the elders of bygone days enter the proverb[ial] per-
formance, God or Christ emerges in the scriptural allusion, of-
ten through the mediation of a sacred text. In both cases, the
quotation addresses the audience and elevates the social interaction.
(1988: 158)
In the case of the scriptural quotation, however, the focus is moved back to the
‘‘earlier times’’ of the scripture itself and becomes a subject in its own right. As
Briggs puts it, ‘‘Scriptural allusions are able to break out of the interactional
setting that gave rise to them and become the raison d’être of the discourse’’
(1988: 169).
While I will discuss particular instances of Sanskrit quotation below, it is
important to note here some overall ideologies about the effect that Sanskrit
recitation has on social situations. Intriguingly, Briggs articulates in his own
anthropological language an idea that the women themselves also articulated,
albeit in a slightly different intellectual vernacular. From the basis of fifty-six
out of the eighty responses, it was quite clear that the women have a general
ideology behind Sanskrit quotation as a practice in its own right. As one
senior woman put it, ‘‘Sanskrit has a purifying effect on the mind. It opens up
a whole world once you recite it.’’ And Asha Gurjar, also a senior, said, ‘‘If
your mind is already accepting samskaras, then you will have better samskaras
in Sanskrit.’’ A younger woman said, ‘‘Whatever conversation is happening in
your mind, you feel as if you have answered the voices if you recite at the
beginning of the day.’’ Menakshi Kodnikar also mentioned the effect that re-
citing a mantra has on cooking. When she recites mantras, the food has a
special quality to it, and it affects everyone whom she feeds in an auspicious
and peaceful way.
Again, at a general level, these women reported frequently that their
Sanskrit ability, particularly the ability to recite, became in its own right a form
of negotiation with their fathers and brothers. A rather striking fifty-two out of
eighty women narrated some version of a story of their father’s negative eval-
uation of them as women who could study and teach on their own right or as
weaker compared to their brothers. One woman’s father studied Sanskrit with
her elder brother, but forbade her to come. Her insistence on continuing on
the path was a means of gaining his respect, if not his approval. She felt she
had done that when he didn’t object to her accepting a scholarship to ‘‘offer’’
the subject of Sanskrit at a women’s college. Another was encouraged to study
Sanskrit because she didn’t need a job that paid well, as her brother did. Others
articulated the same interesting connection between a woman’s status as a wife
and her ability to take the lower-paying job of Sanskrit professor.10
However, for thirty-seven of these eighty women, their ability to recite
Sanskrit somehow resolved the family situation and acted in part as a com-
mentary on the relationship between the women and the men in their families,
whether they be fathers or brothers. Even though I heard no specific story of
Sanskrit recitation suddenly creating peace in the family, many women re-
ported that they were able to prove themselves or, as one woman put it, ‘‘play
a role in the family’’ after they had persevered and succeeded in learning
Sanskrit. Lalita Marathe, a researcher in the Prakrit Dictionary Project, spoke
of this in the following way:
When I was about five, six years old, I recited the twelfth chapter of
Bhagavad Gita, and I got a prize. My parents taught me some stotras,
like shubham karoti kalyanam. In this way, I was acquainted with
Sanskrit in my home only, and my mother, father, uncles,
mama [uncle], and mausi [aunt] all helped me in reciting Sanskrit.
My younger brothers do sandhya [daily recitation]. I also recite
with them.
On a general level, then, in Briggs’s terms, the women’s overall ability and
competence in Sanskrit recitation and study ‘‘addressed the audience and ele-
vated the social interaction.’’
However, specific instances of the kind of scriptural allusion that Briggs
analyzes did arise in our conversations. One older woman spoke of recitation
of the Gita as the ‘‘capstone’’ to her everyday experience. As she put it, ‘‘Only
Gita can be an answer to our quest.’’ Here, she did not mean simply the
substance of the text of the Gita, but its actual recitation at certain moments.
She mentioned that she recited the shlokas of the Gita that involve the word
kartavyam—what is to be done—because they give her a sense of duty and
direction when she is unsure. They also help her to meditate on the purpose
and nature of duty toward family in her life.
There is the idea of pitridev and matridev [mother and father
honored as divine]. You shouldn’t first cut the roots and then try to
join them. And the Gita doesn’t say na kartavyam, it says kartavyam.
[Recites Gita 3.19]:
tasmad asaktah satatam karyam karma samacara
asakto hy acarankarma param apnoti purushah
[My translation: So, without clinging, always perform action to be
done. When one performs actions to be done without clinging, one
attains the highest.]
And I feel that only Gita can be an answer to the quest. My
brother should have done Sanskrit. . . . I tried to carry on the tradition
of my father, even if my brother didn’t. I am not taking his place, but
trying to help society today. Sanskrit is a kind of personality devel-
opment program, as your mind becomes pure when you speak it and
study it. I use it when I teach the Shrisukta and other things.
Here again, we see that in her own experience, reciting such Gita verses
had both social and personal benefits and could act as a kind of commentary
on a situation—both her own situation and those of her family and society.
And the Gita itself, as Briggs also notices, becomes its own raison d’être for the
conversation. God literally became part of the conversation, as Mrs. Gurjar
recited Krishna’s words. Another woman, one of the coordinators for the Sam-
skriti Bharati movement in Mumbai, pushed this idea of speaking Sanskrit as
the ‘‘divinization’’ of an ordinary situation even further. She told me: ‘‘I was
literally cured of my illness after I began to speak Sanskrit. Deva, guru, sam-
skritam—they are all the same for me. They are the same words.’’
Other Sanskrit scriptural allusions used by women tend to push us beyond
Briggs’s definitions and more into the realm of effecting difficult personal tran-
sitions. Dr. Kala Acharya, in the prime of her career as director of the Sanskrit
Institute, spoke of helping her mother as she was dying. As she put it, ‘‘If some
difficulties come in life, mantras will work. First when my mother was ill, I recited
prayer[s] from the Gita. I alternated with other prayers I knew, and then went back
to the Gita. At midnight she expired. I know it helped with the difficulty.’’
So too, a very senior woman who was involved in the Freedom movement
in Pune spoke of her father’s friend, who was a shastri, or learned person,
teaching her an ashirvad, a prayer of hope and aspiration. She told me, ‘‘When
I first began to study Sanskrit, before Pune University was even founded, this
is how I got confidence every day, and got confidence in my heart. When my
husband was ill I also recited it; it gave me psychological satisfaction, and I
know I could face the future without him.’’
She then recited, udvayam tamasaspari pashyanto jyotir uttamam; devam
devatra suryam aganma jyotir uttamam (Rig Veda 1.50.10; Atharva Veda 7.50.7;
Shukla Yajur Veda 20.21, 27.10, 35.14, 38.24; Krishna Yajur Veda 4.17.4; ‘‘Be-
holding a higher light beyond this darkness, we have come to a highest light,
Surya, god among gods’’). She went on, ‘‘Even the use of the verb tense there
expresses that something is about to happen in the immediate future—that
we have almost reached a stage. And that should give us courage.’’ She went
on to explain to me that this mantra is one of the Sandhya mantras—defined
usually as the meeting point between night and day, but also to be recited as
part of the daily canon of mantras. She then concluded, ‘‘But this is truly a
twilight mantra, in that it has helped me in the twilight of my life, to make
it through the sunset period.’’ This woman was able to understand the dou-
ble meaning of sandhya—both as twilight language and as daily recitation
practice—in a transformative way.
Perhaps the most common example of this transformative use of a mantra
is the Ramarakshastotram—the mantra for protection. Most of the interviewees
in the study learned this mantra as a child and would recite it for visitors and
family. A short version of it (and it has many variations) goes as follows:
Om rama candraya vidmahe raghunandaya dhimahi
tan no vishnuh pracodayat
Almost all of the women in Maharashtra spoke of this shloka as a kind of
talisman, to be used in many different circumstances—some of them very
gendered indeed. The one above is modeled on the Gayatri mantra and has
the same all-purpose function. As one young Sanskrit schoolteacher from
Mumbai put it, that mantra got her through both labor and delivery. ‘‘I recited
the Ramarakshastotram all the way to the hospital, and then in the delivery
room too.’’ She told me this as she spoke Sanskrit with her mother in the
kitchen, both of them laughing as they tried various Sanskrit phrases for ‘‘The
baby is crying. You go take care of her!’’ It may be perhaps unremarkable, but
at another level quite remarkable, that Sanskrit would find itself happily in
kitchens and hospital labor rooms. It would be hard to imagine more do-
mestic situations in which the transformation of Sanskrit recitation might
take place.
Concluding Thoughts
We return, then, to the cat in the courtyard. The story itself suggests a number
of different aspects to the study of Sanskrit and its religious and ritual
meanings—aspects which are borne out in this chapter. First, Sanskrit has
underscored its presence in the home, and women are there to greet it in a new
way. Second, there is a flexibility of response to Sanskrit—and women are part
of that flexibility. The preliminary data from my interviews suggest that in their
own experiences of recited Sanskrit, as well as in their use of it in their daily
lives, women will give it yet another new place in postcolonial society. Contra
the traditional definitions, Sanskrit is alive in a new way. Many of the women
in this study were able to use shlokas to help them through moments of tran-
sition, crisis, or grief, whether it was to remove difficulty in childbirth in the
Mumbai hospital, or to learn to face a new future as a widow. In a certain sense,
based on these women’s experiences, the cultural meaning of Sanskrit will
have as much to do with the use of Sanskrit to negotiate domestic life and
personal transformation as it does with the more public ritual life of the temple
or the sacrifice. This change may be a gentler form of innovation, where the cat
is welcomed, especially in the courtyard.
notes
1. See in particular Saroja Bhate (1996: 383–400); Victor van Bijlert (1996:
347–366); Crostiann Van Der Burg (1996: 367–382); Harry Falk (1993: 103–120);
Pierre Sylvain Filliozat (2000); K. K. Mishra (1997); Ralph Marc Steinmann (1986).
2. Personal communication, V. L. Manjul and Devi Tai, Upasani Kanya
Kunari Sthan, August 2004. Also see Manjul (2003: 64–68; 1996: 38–39); Damle
(1997); and anonymous (2004; 2002).
3. Charles Briggs’s analysis is very astute here; in a performative tradition
with very strict rules like Sanskrit, there is a motivation to vary the perfor-
mance according to subtle contextual cues shared by the audience. As the women
described their participation in Sanskrit elocution and drama contests, the bulk of
their stories was made up of some contextual variation (a different stage, a sudden
noise) that varied the otherwise quite strict performative script.
4. R. N. Dandekar himself was always interested in cultural variations and

local transmutations of Sanskrit practices. In a personal conversation (1997), he
told me that he was sad that the ‘‘village’’ Sanskritists were fewer and fewer these
days, because they provided the most interesting studies of ‘‘Sanskrit dialect.’’
5. On the ‘‘ideology’’ of Sanskrit, one of the larger points of my study will be
that there is far more variation in ideology than is presumed by Western scholars.
Partly because of a lack of enrollment in classes and partly because of a Brahminical
exodus to technology and science, the study of Sanskrit in the major universities is
now well populated with non-Brahmin students. In addition, there are several activ-
ist women Sanskritists on the Left of the political spectrum who are left out of the
discussion altogether in the West—if Western scholars pay any attention to these
women at all.
6. Kumud Pawde’s influence is felt keenly among activist women on the Left
in Maharashtra and in other parts of the feminist movement in India.
7. The scene which is most ironic on this score depicts the acharya, bursting
with disease, leaving his temple and planning to cleanse and heal the town in
all kinds of ways. The author goes on: ‘‘For quite some time he muttered to him-
self like a chant, ‘The Idiots, The Idiots!’. . .’’
8. I am grateful to my colleague Joyce Flueckiger for insight into this material
as well as long conversations about my various case studies.
9. The question of ‘‘American Sanskrit’’ has just begun to be studied; a larger
view of the Hindu education system in America is a desideratum at this point.
Arthi Devarajan at Emory has just begun this important work, focusing on educa-
tional practices and aesthetics. The current generation of graduate students (both
Hindu American and non-Hindu) have a wide field open to them.
10. I treat this subject at length in my book project.
references
Anonymous. 2004. ‘‘Maidens Mastering Man’s Mantra-Veda.’’ Sunday Herald
Spectrum 11.24, June 4.
———. 2002. ‘‘Starting Vedic Studies.’’ Hinduism Today (October–December): 59.
Bhate, Saroja. 1996. ‘‘Position of Sanskrit in Public Education and Scientific
Research in Modern India.’’ In Ideology and Status of Sanskrit, ed. Jan Houben,
383–400. Leiden: Brill.
Briggs, Charles. 1988. Competence in Performance: The Creativity of Tradition in
Mexicano Verbal Art. Philadelphia: University of Pennsylvania Press.
Damle, Manjiri. 1997. ‘‘Women Priests Win Hearts for Dedication.’’ Times of India,
November 29.
Dandekar, R. N. 1972. Sanskrit and Maharashtra: A Symposium. Pune, India:
University of Pune.
Deshpande, Madhav. 1996. ‘‘Contextualizing the Eternal: Features of Priestly
Sanskrit.’’ In Ideology and Status of Sanskrit: Contributions to the History of the
Sanskrit Language, ed. Jan E. M. Houben, 401–436. Leiden: Brill.
Driver, Tom Faw. 1993. Life in Performance: Reflections on Ritual, Religion, and
Social Value. Washington, DC: Society for Values in Higher Education.
Falk, Harry. 1993. ‘‘Goodies for India: Literacy, Orality, and Vedic Culture.’’ In
Erscheinungsformen cultureller Prozesse, ed. Wolfgang Raible, 103–120. Tubingen:
Gunter Narr.
Filliozat, Pierre Sylvain. 2000. The Sanskrit Language: An Overview. Varanasi, India:
Indica.
Gill, Sam D. 1987. Native American Religious Action: A Performance Approach to
Religion. Columbia: University of South Carolina Press.
Grimes, Ronald. 2000. Deeply into the Bone: Reinventing Rites of Passage. Berkeley:
University of California Press.
———. 1995. Beginnings in Ritual Studies. Columbia: University of South Carolina
Press.
———. 1990. Ritual Criticism: Case Studies in Its Practice, Essays in Its Theories.
Columbia: University of South Carolina Press.
Hastings, Adi. 2003. ‘‘Simplifying Sanskrit.’’ Pragmatics 13(3/4): 499–513.
———. ‘‘From Ritual to Grammar: Sacrifice, Homology, Metalanguage.’’ Language &
Communication 23(3/4): 275–285.
Houben, Jan, ed. 1996. Ideology and Status of Sanskrit: Contributions to the History
of the Sanskrit Language. Leiden: Brill.
Laderman, Carol. 1991. Taming the Wind of Desire: Psychology, Medicine, and
Aesthetics in Malay Shamanistic Practice. Berkeley: University of California
Press.
Manjul, V. L. 2003. ‘‘Shirdi-Sakuri Don Upasana Kshetre.’’ Sri Sadguru Sairkrpa,
Oct.–Nov. Dassara-Diwali Issue, 64–68.
———. 1996. ‘‘A Hitherto Forbidden Realm.’’ Manushi 97 (November–December):
37–39.
Michaels, Alex. 2001. The Pandit: Traditional Scholarship in India. New Delhi: Manohar.
Mishra, K. K., ed. 1997. Sanskrit Studies in India. New Delhi: Rasthriya Sanskrit
Sansthan.
Mudimbe, V. Y. 1997. Tales of Faith: Religion as Political Performance in Central Africa.
London and Atlantic Highlands, NJ: Athlone.
Murthy, A. K. 1978. Samskara. Translated by A. K. Ramanujan. Delhi: Oxford
University Press.
Patton, Laurie. 2002. Jewels of Authority: Women and Textual Tradition in Hindu India.
———. 2001. ‘‘The Prostitute’s Gold: Women and Sanskrit in One Corner of India.’’
In Postcolonialism, Feminism, and Religious Discourse, ed. Kok Pui-Lan and
Laura Donaldson, 123–141. New York and London: Routledge.
Pawde, Kumud. 1992. ‘‘The Story of My Sanskrit.’’ In Poisoned Bread: Translations
from Modern Marathi Dalit Literature, ed. Arjun Dangle, 96–107. Delhi: Orient
Longman.
Ramanujan, A. K. 1973. Speaking of Siva. Harmondsworth, England: Penguin.
Ramirez, Olga. 1997. La Fiesta de los Tastoanes: Critical Encounters in Mexican
Festival Performance. Albuquerque: University of New Mexico Press.
Spiziri, Frank S. 1997. A Cobbler’s Universe: Religion, Poetry, and Performance in the
Life of a South Italian Immigrant. Edited and translated by Catherine L. Albanese.
New York: Continuum.
Steinman, Ralph Marc. 1986. Guru Shishya Sambandha: Das Mesiter-Schuler-Verhaltnis

im traditionallen und modernen Indien. Wiesbaden, Germany: Steinder, 1986.
Tedlock, Dennis. 1993. Breath on the Mirror: Mythic Voices and Visions of the Living
Maya. San Francisco: Harper.
———, and Mannheim, Bruce, eds. 1995. The Dialogic Emergence of Culture. Urbana:
University of Illinois Press.
van Bijlert, Victor. 1996. ‘‘Sanskrit and Hindu National Identity in Nineteenth
Century Bengal.’’ In Ideology and Status of Sanskrit, ed. Jan Houben, 347–366.
Leiden: Brill.
Van Der Burg, Crostiann. 1996. ‘‘The Place of Sanskrit in Neo-Hindu Ideologies:
From Religious Reform to National Awakening.’’ In Ideology and Status of
Sanskrit, ed. Jan Houben, 367–382. Leiden: Brill.
Wezler, Albrecht. 1996. ‘‘‘Do You Speak Sanskrit?’ On a Class of Sanskrit Texts
Composed in the Late Middle Ages.’’ In Ideology and Status of Sanskrit, ed.
Jan Houben, 327–346. Leiden: Brill.
2
Wandering from ‘‘Hills
to Valleys’’ with the Goddess:
Protection and Freedom in the
Matamma Tradition of Andhra
Joyce Burkhalter Flueckiger
The goddess used to . . . take me from hill to valley [konda-kona],

village to village. That’s why I go wherever [the goddesses] like; I
go and play and come back. This is my story.
—Govindamma, 2000
If I want to go to my native place and stay there for a few days, there’s
no one there for me. I have no one. Whenever I feel sad, I come
here to the temple and stay. She [the goddess] is my mother and
father, and my mother-in-law, too.
—Veshalamma, 1999
Govindamma and Veshalamma are women who have exchanged

talis (gold disk pendants most often associated with marriage) with
one of the many forms of the South Indian village goddess.1 The
tali binds these women to the goddess but, just as important, signals
her protection. Women who have formed such a tali relationship
with the goddess are called matammas; they are freed from the con-
straints of human marriage, but may form nonmarital sexual rela-
tionships with men and have children. Like the women who wear
her talis, the goddess herself wears a tali, has no husband, but may
have children. This chapter explores the relationship between the
goddess Gangamma and matammas who are protected by her and
thus given freedom to ‘‘wander.’’ While the tali of human marriage
has analytic connotations of binding a woman, restricting her
movement, holding her in both physical and social ‘‘place,’’ this chapter sug-
gests that the tali of the goddess has given Veshalamma and Govindamma
freedom to move across the traditional social and spatial boundaries observed
by most Hindu women in similar stages of life.
Gangamma is one of the seven village goddess sisters who traditionally live
on the boundaries of villages. (These seven sisters are not the same as the
Sanskritic saptamatrikas, having different characters, images, and narratives
altogether.) Depending on the context, the sisters may be conflated or individu-
ated, although it is rare that all seven are named individually when someone
is listing them—and the sets of names vary from village to village. The South
Indian sisters are guardians of village welfare, protecting humans from disease
(particularly poxes and fevers associated with the hot season) and ensuring fer-
tility and the health of crops and animals. However, they may also cause disease
if they are left unsatisfied/hungry/heated and if their ugram (potentially destruc-
tive power) overflows its boundaries; much of the worship of these village god-
desses consists of calibrating the heat/desire/hunger of the goddess and feeding
her. Gangamma and her sisters are characterized in both narrative and ritual as
moving/fluid goddesses; they traditionally reside in open-air sites under trees or
on boundaries between village settlements and paddy fields and often actively
resist human suggestions for enclosed shrines. The seven sisters, although they
may have children and are called ‘‘mother/Amma,’’ are not householders. For
their annual festivals ( jataras), they are called to village centers, temporary struc-
tures are often built around them, and village boundaries are literally ‘‘bound’’
shut so that the goddesses will remain stable long enough through these ritual
periods so that they may be served, fed, satiated.
In the countryside around the South Indian temple town of Tirupati,
Andhra Pradesh, young girls or babies may be dedicated to Gangamma or one
of her sisters when illness or drought strikes the village or the girl herself be-
comes ill with one of the diseases associated with the goddess. Upon reaching
puberty, these girls ritually exchange talis with the goddess (sometimes this
is called ‘‘marrying the goddess’’) and are thereafter called matammas. Tradi-
tionally, matammas do not enter human marriages with men. In some villages,
matammas dance at the festivals of the goddess and otherwise serve her at her
shrines; other matammas live more independently of temples or shrines and
serve the goddess in less institutional contexts (including by becoming regu-
larly possessed by the goddess in their homes and serving as her medium). The
matamma tradition is shifting in the contemporary contexts of reform move-
ments, middle-class sensibilities, and modernization; however, it is still prac-
ticed under the radar, so to speak.
Several middle-class Tirupati residents whom I initially asked about the
term matamma explained, ‘‘You know, like devadasis,’’ assuming I would know
that more linguistically widespread word (for a tradition that is no longer prac-
ticed in the large South Indian temples with which it used to be associated).
wandering from ‘‘hills to valleys’’ with the goddess 37
While there are significant differences between the traditions of matammas, who
form relationships with local village goddesses, and devadasis, who were temple
dancers ‘‘married to the god’’ of more Sanskritic temples, both classes of women
are or were under obligation to the deity and were freed from obligations of
human marriage.2 Further, they are ritual specialists whose ritual acts avert
inauspiciousness and bring auspiciousness (Kersenboom 1987: 67, 192, 205).
They themselves are never widowed since they are married to the deity and are,
therefore, called nityasumangali, ever-auspicious women (Kersenboom 1987: xv).
Nevertheless, today there is a societal ambivalence about these women who do
not marry humans, even though the social institutions with which they are or
were associated may have once been conceptually and ritually acceptable and
even necessary. This ambivalence may be created, in part, by the perceived threat
to the social order posed by a woman not ‘‘tied down,’’ who moves across, and
thereby challenges, traditional social, gendered boundaries. As Katherine Ewing
suggests in writing about the threat of the wandering sadhu/faqir to the colonial
order, their very movement itself ‘‘exposed contradictions and challenged the
naturalness of the [colonial] order’’ (1997: 63).
Supported by legislation outlawing the practice of dedicating women to
temples (particularly the Devadasi Act of 1947) and accompanying reform
movements, several nongovernmental organizations (NGOs) based in Tirupati
work with matammas. Specifically, their mission is to encourage matammas to
marry the men with whom they may be living (the persuasion is actually most
frequently directed at the men), to run schools for their children, and more
generally to encourage an end to a tradition that is perceived, from a middle-
class sensibility, to be exploitative of women. The assumptions of the NGOs
seem to be that matammas are being taken advantage of sexually, either being
forced into relationships against their wills or choosing to enter these relation-
ships without being given the protections of the institution of marriage.
Because of these contemporary social and political contexts, the institution
of matammas is rarely openly talked about to outsiders by those who participate
in the village goddess traditions and festivals with which the institution is still
associated. I attended three annual jataras and lived in Tirupati for nearly four
months before I heard about the tradition; but, conducting my fieldwork pri-
marily in women’s ritual environments, once I did become aware of ma-
tammas, I heard and saw multiple signs of the tradition and ultimately met the
two women whose personal narratives are the basis of this chapter.
Perhaps because of the contemporary ambivalence toward the tradition
and the fact that Veshalamma and Govindamma had both married human
males, neither woman self-identified as a matamma. But they both had ex-
changed talis with the goddess, and the temple flower sellers at one of Gang-
amma’s Tirupati temples referred to them as matammas.3 I think it is not a
coincidence that the matammas whom I personally met had married humans.
Govindamma and Veshalamma have a certain social ‘‘respectability’’ because of
their married status, whatever else they may also be. This status may also have
freed them to speak more openly to me than they would have had they re-
mained unmarried. They are also unusual in the fact that they live in the town
of Tirupati and are thus associated with the powerful Tirupati goddess rather
than one of her (arguably) less-powerful village sisters. One flower seller told
me that the more powerful the goddess, the more protection a matamma has—
that if a woman has exchanged talis with Tirupati Gangamma, no one will dare
to harm her for fear of the wrath of this powerful goddess.
Govindamma is a sixty-five-year-old widow, and Veshalamma is a thirty- to
thirty-five-year-old mother of three young children.4 Upon first meetings, both
women stood out as unique and unusually independent. Their life stories
share themes of protection by the goddess and the resulting freedom to move
and act in public spheres, although each narrative addresses some possible
different aspects of a tali relationship with the goddess at different life stages of
a woman.
I first met Veshalamma at Gangamma’s largest Tirupati temple during
the fall festival days of Navratri (Nine Nights of the Goddess) in 1999. For
each evening of the nine nights, a different pan-Indian form of the god-
dess was installed in the courtyard in front of Gangamma,5 and each evening
Veshalamma swept the courtyard and drew a ritual rice-flour design in front
of the goddesses before the nightly rituals began. Veshalamma was not ac-
companied to the temple by female relatives or friends (as are most women
who come to the temple), although sometimes she brought along her young
daughter and baby son. She told me she walked about a half hour to and from
the temple every day and, even on nonfestival days, spent hours sitting in
front of the goddess, sweeping the temple, or sleeping in the shade of the
temple gate or courtyard trees. She was marked as unique by her freedom of
movement, without accompaniment by family members, and her presence in
public space with no particular agenda, except to be in the presence of the
goddess.
I first met Govindamma in her home, so I did not notice her freedom of
physical movement until I heard her extended personal narrative. However,
she was visibly marked by the goddess in that she wore a large, pronounced
red vermilion forehead marking (bottu) and had matted hair. When I regis-
tered surprise upon learning that she was a widow but still wore the bottu and
tali associated with married women, she answered, ‘‘One hundred and one
Gangammas gave me my bottus; why should I take them off when my hus-
band dies?’’ (One hundred and one is being used here as an auspicious
number of power; literally, there are only seven sisters, although they may
have different sets of names in different villages.) At another point in the con-
versational narrative, she similarly asserted, ‘‘Because 101 Gangammas gave
me the pasupu/kumkum [auspicious vermilion-turmeric marking], I haven’t
taken it off [upon becoming a widow].’’
We do not have historical evidence as to how the institution of matammas

actually worked, whether or not these women were sexually exploited in pre-
modern and contemporary India (as reform movements suggest), or whether
the institution provided some women with agency that they may not have oth-
erwise had. However, the personal narratives of Veshalamma and Govin-
damma provide traces of the protection, agency, and freedom of movement in
public and ‘‘jungle’’ spaces6 that the tradition may have offered matammas
generally and these two women specifically, one a widow and the other an
orphan and abused wife.
The freedom provided by the tali in this context raises questions about the
significance of the tali in premodern South India, before it became so closely
tied to marriage. Is the auspiciousness attributed to a married woman given
by marriage itself (i.e., a husband), or is it inherent to a fertile woman? Here, I
first examine several implications of the tali for both the goddess and non-
matamma women, and then look at the implications of the tali for matammas
as suggested through the oral narratives of two matammas (who are, albeit,
atypical since they also married men).
Talis of Women
Talis in contemporary South India are most often associated with human mar-
riage; a woman’s tali is tied by her husband during the wedding ritual se-
quence. However, it may not have always been associated with marriage per se,
but more specifically with a woman’s inherent auspiciousness, her embodied
creative power, and fertility potential. In a chapter about the goddess in Kerala,
M. J. Gentes describes a pre-twentieth-century tali ceremony that used to be
performed for pubescent girls (1992: 317). While the girl was joined with a male
partner, it was not as ‘‘wife,’’ but as ‘‘one who has the power to create and
withhold life’’ (318). Gentes suggests that the tali ceremony
may facilitate a young woman’s claiming of her shakti (power)
and role and her control of her own chastity (in the South Indian
sense) without expressing it as dissolution of her life into that of a
man’s. . . . As a necessary ceremony it marks and values the young
females of the lineage and of society as a whole. . . . From the cere-
mony onward, the girl was considered a reproducer of the matrilin-
eage. (1992: 318)
The young woman was auspicious because she was a potentially fertile woman,
marked by the tali, rather than because she was a potentially fertile married
woman. Vasudha Narayanan describes a female puberty ritual among the Pillai
community of Tamilnadu in which a young girl is given the authority to light
the oil lamp on the family altar. On this occasion, she is given a gold necklace
called a nava tali (lit. necklace of nine), strung with nine gold beads and corals.
She wears this tali until she gets married, at which time it is replaced with that
given by her husband (Narayanan forthcoming, 220). Other South Indian tra-
ditional puberty rituals similarly approximate some elements of marriage cele-
brations, the pubescent girl being dressed in silks and new jewelry (Reynolds
1991: 42).
Another trace of earlier associations of the tali may be found in the custom
observed in some Telugu communities in which the bride is given a tali by her
mother during pre-wedding rituals, before her husband ties one around her
neck during the wedding itself. This ‘‘mother’s tali’’ is called puttininti tali (lit. of
the house of birth) and should be worn for at least forty days after the wedding.7
It is sometimes then strung on the same chain/thread with the tali tied by the
husband. I suggest that these rituals suggest the association of tali and a
woman’s auspiciousness/fertility independent of marriage, although for most
women marriage is the context within which their fertility comes to fruition.
Later, in Kerala and elsewhere in South India, the tali began to be given by
a woman’s husband during the marriage ritual and has come to be strongly
associated with marriage itself. (Leslie Orr, in her chapter in this volume, citing
Jayadev, suggests this association began to be made in Tamilnadu as early as
the eleventh century.) Holly Reynolds interprets the tali in contemporary South
Indian contexts as follows:
When a man ties a tali around the neck of a woman, he binds her
to him with a symbol of all his culturally and socially derived iden-
tities, makes that woman a cumankali [sumangali], and entrusts to her
the well-being of himself and his lineage, an act that paradoxically
makes the wife the protector of the husband. . . . In owning the tali,
the husband controls the auspiciousness of his wife: he confers
cumankali [sumangali] status upon her at marriage and deprives her
of it at his death. (1991: 45–46)
In marriage, the tali serves as a sign of the husband’s protection, but also
often restricts the movement of the bride/wife to staying within the confines
of ‘‘women’s space.’’ Reynolds continues:
The tali delimits boundaries, sets up barriers, confines [a] woman to

a specified domain, that of her husband, a situation captured in one
of the words for ‘‘wife,’’ taram, limit, boundary. It declares that sex-
ual relations are permissible only with the owner of the tali, for
at issue here is control over and possession of female generative
power. (1991: 46)
The tali’s earlier association with female auspiciousness/fertility, which has

now been subsumed within the institution of marriage, suggests an alternative
interpretation/ideology of the tali and marriage: marriage (the husband, the

tali) does not give a woman her power (shakti), but it marks, and gives a context
for, its expression. The difference is subtle but important.
Examining the exchange of talis between matammas and the goddess
makes more explicit what is only suggested in the puberty rite and marriage
contexts of the tali. Receiving a tali from the goddess, instead of from a man,
openly questions the dominant societal need for marriage as a context for the
expression of a woman’s shakti. More specifically, it highlights the control
implied by the marriage tali tied by a man on his bride, in contrast to the
freedom of movement permitted by the tali of the goddess. Govindamma nar-
rates a very different mandate than that of marriage to a man when she tells of
the goddess ordering her to wander from village to village, ‘‘I keep going
around every uru. I don’t know when she’ll stop me and where she’ll allow me
to build [a permanent place].’’
The Goddess’s Tali
While, traditionally, women need the context of marriage within which to express
their auspicious powers (of fertility), the goddess is not bound by this human
context; she wears the tali and has children without a husband. The talis of many
forms of the goddess Gangamma and her village sisters are usually very visible,
in contrast to married women’s talis, whose chains or threads are visible at the
back of the neck, but whose pendants are usually modestly tucked under their
saris, not to be put on public display. Some of Gangamma’s and her sisters’
forms are just a head, with no torso or body, but around the neck is still a tali.
Gangamma takes two primary forms in Tirupati, as an older and younger
sister (Pedda Gangamma and Cinna Gangamma, respectively). The story is told
of the older sister having children whom she tried to protect from the envy of
the younger, childless sister by covering the children with a basket when Cinna
Gangamma came to visit her. In anger, the younger sister turned the children
into chicks. Pedda Gangamma then offered to give some of her children to her
sister if Cinna Gangamma would turn the chicks back into children. This story
was told when I asked the temple attendants who were the small rock forms
that were lined up on each side of Cinna Gangamma. It then occurred to me, if
the goddess had children and was wearing a tali, who was her husband? When
I asked this question during the 1992 jatara, the flower sellers and temple
attendants said she had no husband. When I returned to Tirupati for a year in
1999–2000, these same women answered without any hesitation that Siva was
her husband. The conceptual presence of a husband may be a recent devel-
opment, a way of bringing Gangamma into a more Brahminic world view, as
there is no mythology of her and Siva being married, and he is iconographically
and ritually absent from her temples and shrines.
In her description of river goddesses in Maharashtra, Ann Feldhaus de-

scribes a similar phenomenon of goddesses who wear the auspicious signs of
a bride (bangles and the wedding necklace [mangalsutra]) but who have no
husbands; she calls them ‘‘husbandless wives.’’ A husband may be implied,
but ‘‘no one seems particularly bothered about the question of his identity’’
(1995: 53, 55). One woman told Feldhaus:
‘‘There is no husband. She’s Krsnabai, isn’t she? And Krsnabai has

no husband. Krsnabai is a river. She’s understood to be a river, isn’t
she? Then, doesn’t this river finally join the ocean? So that makes the
ocean her husband. That’s how you should understand it.’’ And then
she added, ‘‘If there has to be a husband!’’ (1995: 54)
When I asked Veshalamma who tied Gangamma’s tali, that is, who is her
husband, she answered:
It’s actually a vow [mokku]; people make a vow that if they get
married, have children, or if something good happens to them,
they’ll give her a tali.8 It doesn’t mean that she [the goddess] got
married. She’s Adi Shakti [the primordial goddess]. Who would
be able to bear her by getting married to her? Even my own hus-
band keeps telling me that after getting married to me, he’s not able
to bear me.
Veshalamma: The Tali as Protection of the Goddess
Veshalamma’s independence and mobility for a married woman in her thirties

were immediately striking, especially given her advanced state of pregnancy
(about seven months) at the time of the Navratri festival when I first met her.
While there were other women who worked on the temple grounds as flower
sellers, sweepers, and attendants to the goddess, Veshalamma rarely engaged
in conversation with them or other devotees. When I asked the flower sellers
who she was—if she was associated with the temple—they answered that she
simply performed various tasks as service (seva) to the goddess. Later, there
were some whisperings about her having exchanged talis with the goddess. I
learned only gradually over the next few months about the institution of ma-
tammas and then realized that Veshalamma was so identified.
While others may have whispered about this identity, the relationship
with the goddess was not one about which Veshalamma was ashamed. When
I met with her to ask more about her relationship with the goddess and elicit
her life story, we (and the tape recorder) were immediately surrounded by a
group of curious onlookers. When I suggested that we move out of the central
temple courtyard to a shaded spot under one of the temple gates, she asserted
boldly, ‘‘Whatever I have to say is by the grace of the goddess. I’ll speak here.’’
(After only a few minutes, however, when the children became too noisy even
for Veshalamma, we moved from the covered courtyard to sit under one of the
gopurams.)
The flower sellers had told me that Veshalamma had spent the previous
week sleeping at the temple. When I asked her why, she answered, ‘‘I was feel-
ing depressed [lit. my mind was not good], so I slept here. At my place, I felt
like I was being crushed [suffocated] but now I’m feeling reinvigorated [ush-
aruga; lit. active].’’ And then she began her story, which she characterized as
one full of ‘‘troubles, troubles, and more troubles.’’
I was born at Manchala Vidhi [Main Street] and matured [reached
puberty, while living in the neighborhood] behind Jyoti Theatre. Then
they took me to Madras. My parents thought I wouldn’t learn any
housework, so they took me to Madras and put me in a hostel, but
I refused to stay there. Why should I be put in a hostel when I have
parents? They themselves took me to the hostel. They called me an
adopted child; they called me ‘‘sister’s daughter.’’ I cried, but they
still took me there. . . .
When I was small, I had to face so many troubles, so they
brought me to Veshalamma Gangamma [whose name Veshalamma
the woman has taken as her own] temple and married me to her.
They said, ‘‘We’re going to give her to you [the goddess],’’ so they got
me married [to her]. They did this in a respectable way [with suffi-
cient grandness]. After that I had to sell coconuts and dry fish; I
did all the household work. I sold idlis to the hospital. I went to
Nagiri, Nagalapuram, Esalapuram to sell pickles. . . . I worked hard
and looked after them [my parents], but still they didn’t look af-
ter me. So I have only her [pointing to the goddess]; I don’t have
any parents. I’ve experienced only troubles, troubles, and more
troubles.
Veshalamma later elaborated some of her early troubles: she was adopted
and never knew who her birth parents were; and as a baby, she had difficulty
drinking and her parents thought she would die. Because of this, she said, her
parents offered her to the goddess; but, she continued, her adoptive mother
never took care of her and cared only for the son to whom she herself had
given birth. At another point in her narrative, in the middle of a description
of her mother dressing her up like the goddess when she was a little girl,
Veshalamma suddenly shifted topic, ‘‘They [her parents] actually showed
some interest in me before I reached puberty, but once I reached puberty
they began to trouble me.’’ Veshalamma bemoaned that she never found out
who her birth parents were: ‘‘If I had been told, I would have gone to look
for them, to see who they were. When anyone [new] like you comes to the
temple, I always hope that they might be my parents. I want to see and talk to
them.’’
Veshalamma said that her parents first wanted her to get married to an
old man (seventy years old), when she was just sixteen years old. She rebelled,
‘‘How could I get married to such an old man! I wouldn’t marry him. So I
loved and got married to my [current] husband [a marriage of choice rather
than an arranged one].’’ When I asked her how a man feels when he gets
married to a woman who is already married to the goddess—was he afraid of
her in some way?—she answered:
Yes, he’s afraid of me; he doesn’t even smoke in front of me. . . . He’s
afraid because I might become possessed by the goddess. Twice,
when I was possessed, I beat him; that’s why he’s afraid of me.
[When I’m possessed] he’s very respectful of me. But when he’s been
given trouble outside, he comes home and beats me. He asks me to
leave the house or chases the children out.
On another occasion, Veshalamma said that she did not like to become pos-
sessed unless she knew that there would be someone there to serve the goddess
(which she implied may not happen at home): ‘‘Someone should give me
turmeric water and camphor. Who will do that for me? When she comes into
my body, she asks for neem leaves, turmeric water, camphor, and whatever else
she likes. Who will give all this? No one; everyone will just stand there and look
at me.’’
Her husband first became physically abusive after she gave birth to a baby
girl: ‘‘He hit me because I gave birth to a daughter, and I didn’t give birth to a
son.’’ She later described the abuse in more detail, saying that her husband left
wounds on her and broke her hand. She says that this is when she began to
come to the temple regularly, telling the goddess, ‘‘You are my mother.’’ At
another point in her narrative, when she told of the death of her mother,
the rejection by her own brother, and the harassment of her mother-in-law, she
rather matter-of-factly stated, ‘‘I have no one. If I want to go to my native place
[i.e., maternal home] and stay for a few days, no one is there for me. I have no
one. Whenever I feel sad, I come here to the temple and stay. She [the goddess]
is my mother and father, and my mother-in-law, too.’’ Although she was
married to the goddess at a young age, it seems to have taken time and the
experience of multiple troubles before Veshalamma learned to turn to the
goddess for her dependable protection. The goddess conceptually protects and
cares for Veshalamma, but the temple is also a physical refuge. She often sleeps
at the temple, sometimes for days at a time, when she is physically sick, when
her husband is abusing her, or when she is simply sad. No one at the temple
asks why she is there, no one bothers her, and no one from her home comes
looking for her.
In the absence of a caring, protective mother, Veshalamma finds refuge

only in the goddess, and the exchange of talis has created and signifies this pro-
tection. Ironically, Veshalamma says that her mother took for herself the tali and
silk sari that had been given to Veshalamma at the time of her marriage to the
goddess; but, Veshalamma maintains, for this transgression, her mother’s hand
and leg were paralyzed (presumably caused by the goddess).
Veshalamma told me that the same day that she first told me her life
history, Gangamma herself came to her in the guise of a Brahmin woman (a
woman possessed by the goddess):
She [Gangamma possessing the Brahmin woman] gave me a
blouse piece and said, ‘‘You don’t have to be afraid; I am with
you. . . . If you hadn’t come to the temple, you would have died. I
am protecting you.’’ . . . The goddess came to me earlier, too. I had
cooked everything for her, dried fish and eggs, and fed her. Gang-
amma told me she was satisfied that day, ‘‘That day, you filled my
stomach, so I’ll see to it that your stomach, too, is full. I’ll not harm
you. Don’t cry.’’ She called my husband and warned him, ‘‘If you
hurt her, I’ll kill you.’’ My husband trembled. [The goddess said,] ‘‘If
she as much as cries, I won’t allow you to live any more.’’
Veshalamma says she never just ‘‘sat and ate,’’ but worked hard as a little
girl, going from place to place in Tirupati neighborhoods selling pickles. This is
the primary reference in Veshalamma’s narrative to the mobility enabled/re-
quired by the goddess. But the motif is strongest in her own observable actions:
she walks to Cinna Gangamma’s temple almost daily, sometimes staying there
for days at a time (leaving her children at home or bringing them with her) and
periodically also makes the rounds to other Gangamma temples all around
Tirupati. One morning, she offered to take my fieldwork assistant and me on
a round of temples and shrines on the outskirts of the north side of Tirupati.
Her pace was brisk and deliberate as we walked for several hours under the
heated April sun, but she did not tire and after several hours even seemed
invigorated.
As mentioned earlier, both Veshalamma and Govindamma did marry
human males, and in this way they may not be typical of many matammas.
Govindamma exchanged talis with the goddess after she had married her
husband and had children, while Veshalamma underwent a more formal mar-
riage ritual to the goddess in the temple upon reaching puberty and before
her human marriage. It is significant that both women chose their own hus-
bands, rather than entering arranged marriages. Govindamma was already
independent from her family (and had moved around a lot) when she met
her future husband while working at a construction site. Veshalamma simply
states that she chose her own husband, without giving further details.
After Veshalamma had gotten married to her husband, her family called
her a prostitute, saying that she was already married [to the goddess], and
presumably a second marriage was adulterous. She described their opposition:
‘‘My brother pushed me out of the house, saying ‘we don’t have any connec-
tion with you any more, so leave.’ They troubled me a lot. If I think of it now, I
feel sad and feel like crying. When my mother was alive, she didn’t take care
of me; now, too, my stepmother doesn’t. My brother, too, isn’t in a position to
care for me.’’ As mentioned earlier, Veshalamma’s husband, too, provides no
protection and abuses her: ‘‘After getting married, I haven’t been happy. I
don’t get along with my in-laws. Only if I work and bring in money will my
husband talk to me. Otherwise, he beats the children and me. Because I can’t
face these problems, I come to Amma [the goddess].’’
Veshalamma’s story of troubles with parents, husband, and in-laws may be
commonly shared by other women and are not unique to matammas. And many
of these women, too, outside of a tali relationship with the goddess, may come to
her for protection from these troubles. But what makes a matamma different is
that this protection/relationship is formalized, and with it, she assumes the
freedom to move, to be independent. It is this freedom of movement that is the
dominant motif in Govindamma’s life story. Not only is she free to move, but
she feels she must move, at the command of the goddess.
Govindamma: Wandering with the Goddess
As a widowed mother of two grown sons, Govindamma is at a very different

stage of life than is Veshalamma, and they also have different kinds of rela-
tionships with the goddess. Unlike Veshalamma, who exchanged talis with the
goddess at puberty and only after that married a man, Govindamma got mar-
ried to a man first—a marriage that she says the goddess herself initiated.
Further, Govindamma’s narrative actively negotiates her unusual position of
being married to a man and the goddess; she directly addresses the issue and
justifies her unusual position.
Govindamma’s life story is most striking for its repeated images of mov-
ing from place to place at the direction of the goddess, a restlessness that does
not permit her to settle down in any traditional way. In response to the ques-
tion of where she was born, she said:
In Kolar [Karnataka]. I was there until I was five. My mother’s
place was Ambur [near Vellore]. From there [Kolar] they brought me
to Ambur. When I was twelve [presumably after reaching puberty],
the goddess took me all around the world; she didn’t allow me to stay
in one place. Even now, I keep moving around. Even now, she makes
me go to old temples, to powerful goddess stones, shakti stones [shakti
rayi]. Wherever people are having problems, she takes me there. . . .
figure 2.1. Govindamma invoking the goddess Gangamma to possess her. Photo
by Joyce Burkhalter Flueckiger.
47
The goddess keeps taking me from hill to hill, valley to valley,

village to village. . . . I go wherever she calls me. Even if I go to Hy-
derabad or Vijayavada, wherever I go, she doesn’t allow me to stay
more than ten days. She says, ‘‘Get up,’’ and she brings me back. . . . I
keep going around from village to village. I don’t know when she’ll
stop me and where she’ll allow me to build and what she will do.
In another conversation when I asked Govindamma if she were afraid of

the potentially destructive power (ugram) of the goddess, Govindamma used
the same image of wandering/moving that had somehow empowered her,
‘‘No, no, I’m not. It’s because I’ve been wandering from hill to hill, valley to
valley, since I was young; so I’m not afraid.’’
Govindamma did not want to get married to a man, but said that
‘‘the goddess married me off. Then the children came.’’ She then suddenly
shifted her reference for the agent of the marriage from the goddess to the
god Vishnu, who came to her in a vision, giving her reasons that she should
get married:
One Friday at twelve o’clock, I sat for puja [worship of the deity]. I
thought that I don’t need a husband, and so I came [to the temple
of the goddess]. That day he [the god] came lying on a five-headed
serpent, and the wall became a sea. He said, ‘‘If you don’t get mar-
ried, you won’t achieve release, liberation [moksha]. You should get
married, have children, and experience both suffering and happiness.
You can’t avoid this. For the good of the world [loka], the drama of
marriage cannot be avoided. For twenty-five years, he [her husband]
will beat you, cut you, poke you, hit you, and cause you to bleed. You
have to experience all of this. This is inevitable. This is the way of
the world [samsaram], and you have to overcome it.’’ He [the god] gave
me this boon [varam] and went off. . . . I sat down at twelve o’clock
for the puja, and he gave me this vision [darshan] at 3 p.m.
I used to do puja at Cinna Gangamma’s temple, and I used to
sleep there before I was married. But now, after marrying, she asked
me to sleep somewhere else: ‘‘You’re now living in the [social] world
[samsaram], so you should live separately and serve me [lit. do
kollavu]. . . . ’’ Next to this temple, there’s Nerelamma’s temple. She
came at one o’clock at night, along with Gudiamma and all the others
[sister goddesses]. I thought to myself, ‘‘How should I talk to them
[the goddesses]?’’ I came here thinking that I shouldn’t get married,
but even then, he [the god] brought me a match [potential husband].
I said to the goddess, ‘‘If you like the match, then get me married.’’
And the flower garland came down from the [image of ] the goddess.
They [others in the temple] asked, ‘‘What did you ask Amma, for
which she’s given you flowers as an answer?’’ I only answered, ‘‘Give

me the flowers. Amma knows what I asked for.’’ And I brought the
flowers back home. I came back home at one o’clock at night.
It is noteworthy that in this account it is the god Vishnu, and not the god-
dess, who tells Govindamma that marriage (and its accompanying suffering)
is necessary for her to achieve liberation. His presence, particularly in his
cosmic form, lying on the serpent in the ocean (rather than through one of
his incarnations on earth, such as Venkatesvara), represents a more textual,
Brahminic tradition than that of the village goddesses and is a tangible mani-
festation of the multiple discourses (Sanskritic/textual and nontextual) that
Govindamma is negotiating through her narrative and vision. The concept of
liberation, moksha, itself is a Brahminic concept not commonly articulated in
relationship to Gangamma traditions. While Govindamma had moved about
from place to place as a pubescent girl—with and at the order of the goddess—
only after she was married and had children did the goddesses give her the tali
and pasupu/kumkum that mark her special relationship with and obligation to
Gangamma. The goddess then told her that she had been with her husband
long enough; now she should move about with her:
One hundred and one Ammavarus [goddesses] came to me. They

took me to the Gangamma temple. They bathed me, applied pasupu/
kumkum, and put bangles on me. They told me, ‘‘All these days you
were with your husband; now you’re our own. Wherever we are,
you should come there.’’ This was like twenty-five years ago, after all
the children. This is because I asked Ammavaru to make me like
Avvayar [an old female devotee who served the god Ganesh]. I don’t
need my exterior form [rupam]; I don’t need my beauty; make me
Avvayar.’’ . . . I haven’t removed my bottu and kumkum even though
my husband is dead. Because 101 Ammavarus put on my kumkum/
pasupu, I haven’t taken it off.
In another later conversation, Govindamma elaborated that she was

twenty-five years old when she got married to the goddess, after her three
children were born, ‘‘It was to keep me as a kanya pilla [virgin girl; i.e., she did
not have sex with her husband after receiving the talis of the goddesses]. The
goddesses tied the talis. Now if the goddess comes on me and speaks, every-
thing she says will happen.’’ Govindamma’s narrative expresses some of the
tension among ideologies of marriage, gender, and women’s auspicious-
ness that may be present in the traditions of matammas and of middle-class
Brahminic rhetoric and practice.
While, for Govindamma, her marriages to a husband and the goddesses
were sequential, in the case of her daughter, there was a struggle between
human and goddess marriages. As a young girl (virgin; kanya), Govindamma’s

daughter had lived with her two maternal, childless aunts; she was thought to
have brought fertility into the house, and both women had their own children.
Thereafter, her relatives wanted to marry the ten-year-old girl to Govindamma’s
brother (an acceptable marriage partner in conventional South Indian cross-
cousin marriage). Govindamma tried to resist this marriage, however, because
of the age difference; and the groom, too, was not interested in marrying
someone who would ‘‘become possessed just like her mother.’’ The goddess
Ammavaru concurred, ‘‘No, let her remain a kanya. When her mother becomes
old, we’ll come to her.’’ However, the relatives prevailed and took the girl
‘‘uphill’’ to Tirumala (site of the temple of the god Venkatesvara) and dressed
both her and her groom in yellow wedding clothes and garlanded them (it
becomes clear later that this was a ‘‘mock’’ wedding). However, as they were
coming down the hill, a monkey tore the girl’s sari, grabbed the flowers, and
ran off. When the family returned home, they found that the picture of the
god on the hill, Venkatesvara, and his wife, Padmavati, had fallen down and
broken. And the daughter lost her sight at the same time.
The narrative sequence is not clear, but it seems that the girl had some kind
of pox even at this time. The goddess told Govindamma that if the daughter did
get married, not only one pox (ammavaru, a manifestation of the goddess) would
come, but seven different poxes would come. The goddess was literally fighting
for the girl in dramatic ways, but the relatives proceeded with the mock mar-
riage. Govindamma narrates, ‘‘The goddess [who is also the pox] peeled off the
skin and ate it. The girl’s hair came off like a wig comes off. The skin from her
hands and legs came off just like that. And the goddess ate it.’’ The girl died af-
ter a twenty-five-day battle between the poxes (the goddess) and the humans who
wanted her to marry their brother. The goddess prevailed. This is the powerful
goddess who gave Govindamma her tali and who orders her to move with her
from village to village, hill to valley. Govindamma responded to the comment
that it was unusual that she could ‘‘bear’’ the demands of a relationship with
such a goddess, when most people were so afraid of her:
You must be born for this in order to bear [Telugu, bharincu] her.
Most people can’t even bear pictures of her. Ask them to keep these
pictures and bear her. Let’s see if they can. They’ll get scared and take
them to a temple. I was born for this, and I’ve faced all these diffi-
culties, without any house, without any stable home village [uru].
I wander around from village to village, between tree and snake hole.
If it was anyone else, they would have left the goddess and gone off.
Because I was born for this, even after having faced all these problems,
I’m not worried that I don’t have a house, that I don’t have this,
that I don’t have that. Whatever will happen, let it happen. Why
should I have what the goddess herself doesn’t have? I’m not both-
ered that I don’t have jewels or property. Let it happen as it should. I

was supposed to face difficulties along with the children, and I’ve
experienced all of it. I’ve faced things with my husband, too. I’m
serving the public. People tease me, ‘‘Ankalamma [one of the seven
sisters] hasn’t given you a house or a plot of land.’’ They’ve teased me
like this. I don’t let it bother me. They say, ‘‘This talli [mother] is
making you roam about like this.’’ But I don’t let this bother me. If
I’m born like this, this is what should happen.
Conclusion: Responsibilities to and Manifestations

of the Goddess
A tali relationship with the goddess does not come freely, but comes with re-
sponsibilities toward the goddess and other humans who must interact with her.
The relationship with the goddess is not simply a private one; both Veshalamma
and Govindamma also serve as intermediaries between the goddess and the
human world by regularly getting possessed by Gangamma on Tuesdays, Fri-
days, and/or Sundays. The goddess speaks directly to people who come to her
with questions and to have their futures discerned. Further, both matammas are
regularly called upon to visit persons in whom the goddess has manifested as
poxes (measles, chickenpox, rashes, and in earlier times, smallpox) and to per-
form rituals for/to them. Govindamma concludes one of her narratives:
Now I’ve come to the end of my life. I don’t own a house. Up un-
til now, I haven’t asked Amma for this or that. I just keep going
around and coming back. I don’t go and earn money; I don’t engage
in fraud and earn money that way. When people come [to the god-
dess, through Govindamma’s possession], she heals them. People
who are childless, she gives them children. Those who are unmarried,
she gets them married. This is the drama of my life.
Govindamma has also taken up an unusual ritual responsibility: to take
garega (a bamboo temple-shaped structure tied to the top of her head) to var-
ious goddess temples in order to bring them prosperity. Govindamma says
that her presence with the garega blesses run-down temples, and they ‘‘rise
up’’ and become active and prosperous. Note here that Govindamma speaks
of permanent structures for the goddess, unlike the traditional locales at the
edge of villages where the goddess resides under open skies—another sign of
shifting village traditions in urban contexts such as Tirupati.
The oral narratives that we have been considering in this chapter were told
by two (albeit atypical) matammas in a contemporary context that is multilay-
ered and rapidly shifting. The narratives reflect and negotiate some of the pres-
sures on the tradition and the individual women participating in it, pressures
created, in part, by shifting patronage, early twentieth-century and contem-

porary reform movements, growing middle-class sensibilities, and the ‘‘upward’’
mobility of the Gangamma tradition in Tirupati.9 However, the narratives also
provide us with hints of what matamma tradition may have provided (and po-
tentially may still provide) to women: protection, freedom to move, and agency.
Each context in which the matamma tradition is represented provides a
different interpretive frame for the social/ritual institution, and to address all of
these social-historical perspectives is beyond the scope of this brief chapter. Here
I have focused on rituals and narratives of the goddess and the narrative voices of
the women themselves, voices frequently lost in the dominant discourse about
matammas in the press, in institutions such as universities and development
organizations, and in middle-class conversational discourse. The narratives of
Govindamma and Veshalamma suggest that the tali relationship with the god-
dess has provided them a social-ritual institution of refuge, in which they are
protected by the goddess. It is also one that enables them to be independent from
reliance on husbands and sons. These narratives suggest that the tali of the
matamma may mark her female agency and auspiciousness in ways similar to
that of the tali of the goddess. The matamma tradition offers an alternative
context to that of marriage in which a woman’s powers and auspiciousness may
be marked and expressed. Govindamma’s and Veshalamma’s narratives also
make clear, however, that a matamma’s agency and auspiciousness do not nec-
essarily translate into a trouble-free life. But as they negotiate these troubles,
their tali relationships with the goddess give matammas freedom (even com-
pulsion) to move beyond the social and physical constraints often imposed by
talis given by husbands and give them a site of refuge and protection.
notes
1. The fieldwork upon which this chapter is based was conducted in Tirupati,
Andhra Pradesh, between September 1999 and June 2000. I gratefully acknowl-
edge the support of an American Institute of Indian Studies senior research grant
for this research.
2. See Vasudha Narayanan’s chapter in this volume for more on temple dancer
devadasis, associated with Sanskritic male deities.
3. These women sell flowers and coconuts to worshippers to offer to the god-
dess. They have been a constant presence over the years that I have been visiting
Tattayagunta Gangamma temple.
4. I use the ethnographic present narrative tense; however, the ethnographic
research upon which this chapter is based took place in 1999–2000.
5. Navratri is a newly introduced festival to Gangamma’s temple; that is, it is
not a festival traditionally associated with this level of village goddess. One could
interpret its introduction as a ‘‘Venkatesvarization’’ of the temple, indicating its
move ‘‘up’’ into the range of Brahminic rituals associated with the god Sri Venka-
tesvara, whose Tirupati temple is the wealthiest in India and a site to which thousands
of pilgrims come each day. However, I would suggest that in this case, the pan-Indian
goddesses were, in the minds of Gangamma’s traditional worshippers, simply an
expansion of the set of seven sisters—that is, they were ‘‘Gangamma-ized.’’
6. In Indian languages, ‘‘jungle’’ connotes uninhabited space, not necessarily
forested space. The phrase konda-kona, while literally meaning ‘‘hills-valleys’’ also
connotes this kind of jungli, uninhabited space.
7. I was able to observe this ritual at the Telugu wedding of an Emory Univer-
sity women’s studies graduate student, Yamini Atmavilas, in September 2004 in
Atlanta. Yamini’s mother tied the tali just before the bride performed Gauri puja, in
an antechamber to the public stage upon which the groom and the bride’s father
were performing Ganesh puja. Yamini’s mother told us that her own mother-in-law
had become very upset when, upon the death of her husband, she was expected to
stop wearing kumkum and jewelry and to start dressing simply. She angrily de-
clared that ‘‘all this’’ had not come to her simply because she was married; after all,
‘‘didn’t my mother tie the puttininti tali before my husband did?’’
8. Leslie Orr’s chapter in this volume documents similar vow making and giv-
ing the goddess ornaments, including talis, as early as 1000 c.e. A key difference
between offering the goddess a tali as an act of devotion or fulfillment of a vow
(an offering still common in contemporary Hindu practice in South India and in
South Indian temples in the United States) and offering a tali as a matamma is that
in the latter case, the human woman also receives a tali from the goddess.
9. In 1992, the traditional female attendant to Gangamma in her largest Tirupati
temple was replaced by Brahmin male priests, who began to recite Sanskrit verses to
the goddess, introduced several Brahminic rituals, and strongly discouraged animal
sacrifice.
references
Egnor, Margaret. 1991. ‘‘On the Meaning of Sakti to Women in Tamil Nadu.’’ In
The Powers of Tamil Women, ed. Susan S. Wadley, 1–34. Syracuse, NY: Maxwell
School of Citizenship and Public Affairs, Syracuse University.
Ewing, Katherine Pratt. 1997. Arguing Sainthood: Modernity, Psychoanalysis, and Islam.
Durham: Duke University Press.
Feldhaus, Anne. 1995. Water & Womanhood. New York: Oxford University Press.
Gentes, M. J. 1992. ‘‘Scandalizing the Goddess at Kodungallur.’’ Asian Folklore
Studies 51: 295–322.
Handelman, Don. 1995. ‘‘The Guises of the Goddess and the Transformation
of the Male: Gangamma’s Visit to Tirupati and the Continuum of Gender.’’
In Syllables of Sky, ed. David Shulman, 283–337. New Delhi: Oxford University
Press.
Kersenboom, Saskia C. 1987 [rpt. 1997]. Nityasumangali: Devadasi Tradition in South
India. Delhi: Motilal Banarsidass.
Marglin, Frederique Apffel. 1985. Wives of the God King: The Rituals of the Devadasis
of Puri. Delhi: Oxford University Press.
Narayanan, Vasudha. Forthcoming. The Hindu Traditions: An Introduction. New York:
Oxford University Press.
Reynolds, Holly Baker. 1991. ‘‘The Auspicious Married Woman.’’ In The Powers of
Tamil Women, ed. Susan S. Wadley, 35–60. Syracuse, NY: Maxwell School of
Citizenship and Public Affairs, Syracuse University.
Wadley, Susan, ed. 1991 [1980]. The Powers of Tamil Women. Syracuse, NY: Maxwell
School of Citizenship and Public Affairs, Syracuse University.
3
Lovesick Gopi or Woman’s
Best Friend? The Mythic Sakhi
and Ritual Friendships among
Women in Benares
Tracy Pintchman
Studies of Hindu women’s social bonds in North India tend to

focus on women’s place in the domestic sphere and the relation-
ships women share with natal and conjugal kin. By way of contrast,
scholars have tended to relegate to the margins women’s extra-
domestic ties with other women, including friendships forged in
both natal and affinal households. There are good reasons for this
in many cases. Kirin Narayan, for example, has observed that in
Kangra, North India, where she does her research, close female-
female friendship tends to be confined largely to youth, especially
adolescence; once women marry, family ties become the all-
consuming focus of women’s emotional energy. She notes that
many women feel the intimacy friends share is not even accessible to
married women (1986, 66).1 Joyce B. Flueckiger, however, has re-
corded the existence in Chhatisgarh (middle India) of ritualized
friendships among both unmarried girls and married women
(Flueckiger 1996, 40; cf. Jay 1973). My own research among
women living in the city of Benares in North India suggests that at
least some Hindu women living in this part of North India also
cultivate close, lifelong friendships with other women—friendships
that are ritually sealed, highly valued, and self-consciously main-
tained, even after marriage.
My observations are based on research that I undertook in the
city of Benares in 1995, 1997, and 1998. I did not set out to in-
vestigate women’s friendships. My work during these years
focused instead on women’s devotions to Krishna during the month of

Kartik (October–November). Krishna is widely worshipped in contemporary
India in his forms as both divine child and playful lover. He is said to have
been raised in a community of cowherds, and stories about his youth em-
phasize his special relationship with the many cowherdesses, called gopis, of
Vrindavan, the region where he grew up. In his child form, it is said that
Krishna is irresistible to the gopis, who take great delight in his boyish charm.
As he matures into adolescence, Krishna becomes an irresistible young man,
and the cowherdesses fall deeply in love with him, eager to enjoy his erotic
embrace. In this role, Krishna is the divine lover of women. One of the most
famous images of Krishna situates him in the rasa-lila, the famous ‘‘circle
dance,’’ where Krishna, surrounded by a circle of gopis in the forest, not only
dances with them, but also multiplies himself many times over so that he
can make love to each and every one of them. While early texts focus on
Krishna’s dalliance with all of the gopis, subsequent traditions pair Krishna
especially with one consort, Radha, and portray Radha as the chief gopi and
Krishna’s primary partner in love. Krishna eventually leaves Vrindavan to take
up residence in Dvaraka, where he leaves Radha and the cowherdesses be-
hind, acquiring a vast number of wives and assuming the rulership of the
Yadava clan.
During Kartik, many Benarsi Hindu women perform a collective daily puja,
a form of ritual worship, in which they raise Krishna from infancy to adult-
hood, culminating in his marriage to the plant-goddess Tulsi toward the end
of the month.2 During this puja, women assume the devotional stance of the
gopis. Just as the gopis are believed to have gathered around Krishna in a circle
in the original circle dance, so Kartik puja participants gather in a circle
around icons of Krishna and other deities; and just as the gopis of long ago
adored Krishna with song and dance, puja participants worship him with song
and devotional offerings. Popular Krishna traditions equate the gopis with the
faithful female servants, known collectively as the sakhis, who accompany and
serve the divine couple Radha-Krishna. Sakhi means ‘‘female friend,’’ and in
Kartik puja circles, women are enjoined to refer to themselves and to each other
only with this term.
In the course of discussing the meaning of the term sakhi with Kartik puja
participants, I found that participants tended to define it first in terms of hu-
man friendship, bringing up the term’s connection to Krishna mythology only
at my prompting. When I probed further, I was told also of a ritual of be-
coming (banana) or tying (bandhana) sakhi, in which women exchange vows
of lifelong friendship. Once I learned of the existence of this ritual, I began
to ask informants about it explicitly. About twenty of the thirty-six women I
formally interviewed in the course of my research confirmed its existence
and described for me both the act of ritually becoming sakhi and the bond
that is thereby established, including the meaning of the relationship and the
lovesick gopi or woman’s best friend? 57
obligations that it entails. Almost all of these women claimed the authority of
experience, contending that they themselves had established ritually sealed
sakhi relationships that they value and strive to maintain.
I have never witnessed the ritual that women described to me. The re-
ports tended to be consistent, however. Furthermore, Edward Jay recorded
the existence of ritual friendships among women, including ritual sakhi re-
lationships, in Chhatisgarh more than thirty years ago (Jay 1973), indicating
that the tradition exists beyond the confines of Benares and has been around
for a while. Several of my informants lamented that the practice of creating
ritual friendships between female friends was a tradition in decline and that
girls and younger women were not preserving the practice to the same degree
as the previous generation. This may be true, although I did talk to some young
women who had entered into ritual sakhi friendships. I suspect that in their
conversations with me, informants tended to idealize the sakhi relation-
ship and gloss over ways in which their own sakhi ties may deviate from the
ideal, so one certainly ought to exercise caution regarding informants’ reports
about what has actually transpired in their own sakhi relationships.3 What
especially interests me in this context, however, is the way that ritual
friendships performed among the women I interviewed seem to appropriate
the mythic sakhi of Krishnaite sacred history as a model for extra-domestic
human female-female social and emotional bonds that women choose for
themselves.
Six informants explicitly described the sakhi bond among women as one
that imitates or replicates divine models. All six invoked the relationship be-
tween Radha and her sakhis, or the relationships that Radha’s sakhis shared
with one another, as the root (mul ) or role model (adarsh) for the sakhi bond. One
woman, for example, proclaimed, ‘‘My relationship with my sakhi is like the
relationship of Radha and her sakhis. And we hope that we will be together in
the same way for our whole lives.’’ Another invoked as a model the relationship
shared by Radha and Krishna as well, noting, ‘‘The way Krishna used to love
Radha and the kind of deep affection that the sakhis have for each other, similarly
we also become sakhi.’’
This seeming collapse of an erotic, male-female relationship—that be-
tween Radha and Krishna—with the relationships of deep friendship attrib-
uted to Radha and her female friends adumbrates a larger issue surrounding
sakhi relationships as they were described to me: the sakhi bond in many ways
imitates or echoes some of the social and emotional aspects of the marital
bond. Like marriage, the sakhi relationship is considered unique, deeply inti-
mate, and entailing specific rules and obligations. I would argue that the sakhi
bond that informants in Benares described to me deploys religious and marital
imagery in ways that sacralize ongoing relationships among female friends,
according them social and even religious legitimacy and establishing a socially
valid place for them in women’s lives. Although these relationships exist only
at the margins of patriarchal social discourse, which defines women largely in

terms of male-centered kinship relationships, they are reported to be of great
importance to many of the women who enter into them.
All but one of the women who spoke with me about the sakhi bond af-
firmed the existence of a ritual whereby the bond is sealed. The essential
elements of this ritual practice include an exchange of gifts and food, the
swearing of an oath, and the presence of a deity, who acts as a witness.4 This is
how one informant, Gita,5 described the process of becoming sakhi:
You buy bangles, bindi, hair ribbons, clothes, and some ornaments—
like earrings—to give. By giving these, this is tying sakhi. If the
girls are unmarried, then they give each other these gifts and
go to the Sakshi Vinayak temple and say, ‘‘Considering you as a
witness, we will remain sakhi.’’ And they take an oath that ‘‘we will
remain friends with each other, participate in each other’s auspi-
cious and inauspicious functions, in birth and death, marriage, and
so forth. And at the time of death, I will be with you.’’
The gifts exchanged most often include those like the ones Gita described:
clothes, makeup, jewelry, bindis (the decorative dots that Indian women place
on their foreheads), and sindur, a bright red or orange powder that married
women place in the part of their hair. Two informants described the items
exchanged as ‘‘stuff for marital auspiciousness’’ (suhag ka saman). Other women
stressed the exchange of food, especially sweets and pan,6 as crucial to the
sealing of the bond. Sakhis not only feed one another, but they also self-
consciously exchange with one another food polluted by their saliva. One elderly
informant described the process as follows:
In the ritual of becoming sakhi, there are some puja things—sweets,
yogurt, pan—and the sakhis feed each other these things. . . . For ex-
ample, if you and I were becoming sakhi, then I would feed you
sweets, you would feed me sweets, and then you would bite off some
pan and I would chew it; and I would bite off some pan, and you
would chew it. And they say to each other, ‘‘Everyone may leave
us, whether it is husband, or mother, or father, or brother, but we
will never leave each other!’’
The exchange of such polluted food signals both intimacy and lack of hier-
archy, indicating that both parties are willing to accept the other’s pollution
into their bodies. Equality is an essential component of the sakhi relationship.
Jay stresses the nature of ritual friendships as a means to transcending social
difference; he notes that the ritual friendships he observed in Chhattisgarh
never occurred within the same caste in the same community, claiming that
the very hallmark of ritual friendships is their ability to legitimize close per-
sonal relationships across caste.7 Similarly, the women I interviewed insisted
that there should be no hierarchy between sakhis, even though one’s sakhi may
come from another caste, religion, or nationality. One informant, for example,
proclaimed, ‘‘Sometimes we even make sakhi with a Harijan. . . . we are never
aware of caste after making sakhi.’’ It is difficult to know to what extent such
claims are true: none of the women I interviewed claimed to have a Dalit or
Muslim sakhi, although one informant did claim that her sakhi is Christian.
Nevertheless, the rhetoric of equality is very important in women’s descriptions
of their sakhi relationships. Other informants described the sakhi bond as one
of choice based solely on strong mutual affection. As one put it, ‘‘Sakhi means
that you should have true love [saci priti].’’
Regarding the types of foods exchanged in the sakhi bonding ritual, Ved
Prakash and Sylvia Vatuk have noted that sweets play an important role in all
kinds of ritual activity and in cementing social ties between individuals and
groups; they are associated with pleasure, celebration, and rejoicing (Vatuk
and Vatuk 1979, 182). Bride and groom also exchange sweets and yogurt—
which some informants mentioned explicitly as items exchanged in the sakhi
bonding ritual—at the time of marriage (usually as yogurt mixed with brown
sugar, gur), suggesting parallels between the sakhi relationship and the mar-
riage relationship. Such parallels are also evident in naming conventions.
About half the women who spoke with me about the sakhi bond insisted that
one should never refer to one’s sakhi by name, claiming that this is an essential
rule (niyam) governing the relationship. One informant, Lilavati, drew an ex-
plicit parallel to the marriage relationship in this regard, noting, ‘‘The way we
don’t call our husband by his name, similarly, we don’t call our sakhi by her
name. We call her ‘sakhi’ or ‘mother of so-and-so.’ ’’ Whereas husband and wife
are unequal partners, however, sakhis are not; hence one informant described
the importance of using the term sakhi as reflecting the special closeness and
feelings of equality that are characteristic of the relationship, signaled in the
mutual feeding that takes place in the sakhi bonding ritual.
Like marriage, too, the sakhi relationship is also a bond that informants
generally described as lifelong and unbreakable. In her research on this topic,
Flueckiger notes that married women who enter into ritual friendships as-
sume ritual obligations similar to those of kin (1996, 40).8 This also seems to
be the case among the women I interviewed. Here is how one informant, Hem
Kumari, responded to my questions about her sakhi relationship:
My sakhi and I have always been friends. We were brought up

together. When I was a child, we used to play together, run
around the village, and bathe together in the village tank. . . . We used
to work in the fields together. So that’s how our friendship started.
Then when we got older, we decided we should become sakhi. I
became sakhi with her on the day of Makkar Samkranti—women
often make sakhi on that day. So that is how we became sakhi, and
then we started giving each other gifts. When I got married, she gave
me a gift; when she got married, I gave a gift to her. And since we are
sakhi, if she is in trouble, I can help her, and if I am in trouble, she
can come help me. Even though we are now married and living in
our husbands’ place, we send each other letters, and we keep in
touch. My sakhi’s husband died. She has two daughters and one son;
so I occasionally call her here. Whenever there is a marriage in
my family she will come over here (to Benares) because she still lives
in my (natal) village. And whenever I have a chance to go and
visit her, I also go and visit her. Whatever we have to feed each
other, we feed each other. This is what we do.
Hem Kumari’s description points to the obligations that several informants

described as intrinsic to the relationship: attending one another’s important
family functions, making an effort to spend time together even if one has moved
away, exchanging gifts, feeding each other, and sharing resources. Of course,
the ability to fulfill such obligations is contingent on the cooperation of par-
ents, husbands, or in-laws. But when permission is granted, sakhis are obli-
gated to maintain the tie through such specific behaviors.
Another characteristic of the relationship that informants stressed was
complete honesty and trust, especially in keeping one another’s secrets. One of
my informants, Ramavati, noted:
When two sakhis sit down with one another, they will obvi-
ously talk about their sorrows and pleasures [dukh-sukh]. . . . And
whatever one sakhi tells the other, she must keep it secret. They must
not tell one another’s secrets to anyone else because the rules of
the sakhi relationship are very strict. You must always keep to your-
self all the things you hear from your sakhi.9 There are lots of things
about which sakhis only tell their sakhis. There are lots of things a
woman cannot say to her husband but can say to her sakhi.
Narayan has also noted the importance among female friends of sharing
secrets. She quotes one informant named Veena, who claims, ‘‘With a female
friend [saheli] you can share those things that can’t be shared with others, you
can say things that you shouldn’t say to our husband’s sister [nanad] or to the
sister [bahen] made there’’ (meaning, in the husband’s village) (Narayan 1986,
66). Another of Narayan’s informants, Kamal, is quoted on the same page:
‘‘Only a saheli can be counted on to keep secrets. A woman you know later
might tell anyone.’’
Among the women I interviewed, some also described the special bond of
honesty and trust that exists between sakhis as surpassing that of a woman
and her husband or kin. Lilavati, for example, noted, ‘‘When we become sakhi,
then we will not hide anything from one another. . . . you never lie to your sakhi.
You may lie to your husband sometimes, but you never lie to your sakhi. This
is how pure [shuddh] the relationship with the sakhi is.’’ Another woman vol-
unteered, ‘‘After the joining of sakhis, no one is as trustworthy as your sakhi,
not even your brother, sister, anyone from the in-laws’ house, like mother-in-
law or father-in-law; no one is as trustworthy as your sakhi.’’ Yet another
informant explained, ‘‘In Hindi, there is a phrase: ‘relatives are left, but one
doesn’t leave one’s friends [hit chut jate hain, lekin mitr nahin chute].’ Friends
are always friends. Women have their sakhis like this; if you have made a
sakhi, it is your duty [dharma] that you should never leave your sakhi.’’
Informants repeatedly stressed the need to keep up the sakhi relationship
and refrain from violating the rules and obligations perceived to be intrinsic to it.
Most agreed that this is difficult, and for this reason several informants cautioned
against taking more than one or two sakhis. Seven of the women I interviewed
urged that there should be only one sakhi, usually taken during childhood and
maintained throughout one’s lifetime. Four women allowed for two sakhis, one
in the natal home and one in the home into which one marries. Others claimed
there is no restriction on how many sakhis one may have, but one must maintain
each and every sakhi relationship and fulfill all the obligations that are entailed in
the relationship—and for that reason, it is important to limit the number of
one’s sakhis.
The existence of ritually bonded sakhi friendships among the women I
interviewed raises two important issues I wish to stress here. First, the sakhi
bond affirms Susan Sered’s observation that even in religious traditions that
accord institutional authority primarily to men, women may appropriate and
reshape religious traditions in ways that are uniquely meaningful to them
(Sered 1992, 1994). As noted, many women consider the bonds of friendship
between human sakhis to be modeled on the intimacy enjoyed among Radha’s
sakhis, between Radha and her sakhis, or between Radha and Krishna. The
sakhi relationship integrates aspects of Krishna mythology in ways that support
and strengthen women’s social bonds with one another rather than with dei-
ties or human males. The love between Krishna and the numerous gopis/sakhis
is often interpreted as sexual and hence transgressive, transcending earthly,
human morality. For many Benarsi women, however, the sakhi represents an
earthly female-female bond characterized by ties of mutual trust and caring,
and it may imitate or even surpass blood or marital kinship bonds in terms of
its professed emotional valuation in women’s lives.
Second, the sakhi bond provides an alternative to predominant social con-
structions that locate women’s important social ties solidly within the domestic
sphere, especially within the conjugal home. Gloria Raheja has argued that
Indian women’s songs may question the discourse of patriliny, challenging its
claim to exclusive authority and constructing alternative readings of kinship
(Raheja and Gold 1994, 105). The sakhi bond offers another kind of alternative
construction. It mimics the husband-wife bond in significant ways (ritually
sealing the bond, not using each other’s names, feeding one another, entailing
lifelong obligation, and so forth). Yet it is a tie in which the two parties are
equal, and both affection and obligation are understood to be mutual. It is also
a relationship over which women have control. Generally, Hindu women living
in this part of India do not freely choose their husbands; but they can, and do,
choose their sakhis.
Susan Seymour notes that, in Indian contexts, love tends to be experienced
as a ‘‘deep sense of emotional connectedness,’’ which she calls relational love.
Seymour maintains that feelings of relational love may be extended to non-
family members as well, including friends. With respect to her own experience
among Indian women, she remarks that even married women expect and value
feelings of love among friends. She writes that her friends Mita and Sita
‘‘frequently spoke to me of their affection for and friendship with me and their
fear that I would one day go away and forget them. They wanted to build into our
relationship some sense of dharma—some agreement that I would take the
friendship seriously and, after leaving India, would continue to communicate
with them’’ (Seymour 1999, 85). I had a similar experience when I was preparing
to leave Benares in 1998 after spending many months over the course of four
years conducting research among a group of women. I had grown very close to
a few women, and it was hard for all of us to think about saying goodbye
without knowing when I might be returning to the city. This is what I recorded
in my journal during the final days of my stay:
I went to do some last follow-up interviews today with Krishna Devi
and Kusumlata, who loaded me down with gifts of necklaces. I
took pictures of Kusumlata’s whole family and promised to bring
them by on Friday. She wanted to tie sakhi with me. When I went
there today, as she was giving me all the necklaces, she said, ‘‘This
is for us to become sakhi.’’ She was saying that this was her way
of tying sakhi with me, but I didn’t have anything to give her. I felt
so bad. As she put me on the Rickshaw, she said that ‘‘it is as if
half my body is leaving and going to America.’’
As I look back on that moment, difficult for both Kusumlata and me, it
now seems obvious that the image of friendship Kusumlata invoked in ex-
pressing her feelings about my departure—that it was like half of her body
leaving for America—evokes the image of jori, meaning something like ‘‘united
couple’’ or two persons joined together in a harmonious oneness, a single
being embodying two persons. The ideal of the jori is captured in the image of
Ardhanarishvara, Shiva in his form as half-male, half-female, god and goddess
fused together in the same body. Sudhir Kakar contends that the ‘‘wished-for
oneness of the divine couple’’ is especially important to Indian Hindu women
and represents their idealized image of marriage (Kakar 1990, 83–84).
Narayan observes, however, that in Kangra, North India, where she has
conducted field research, the same term, jori, is used for the relationships
between unmarried girlfriends and between bride and groom. Narayan con-
tends that the shared use of this term might indicate that a husband is expected
to psychologically replace a group of girlfriends (Narayan 1986, 68). While
this may be true, it is also possible that the image of two beings sharing the
same body points to an underlying conception common both to deep
friendship and to marriage of an ideal, unbreakable, and transformative bond
between persons, forged of intimacy and affection, that transcends mere so-
cial convention in its significance and claims to mutual obligation. This may
be the longing that Seymour notes for ‘‘some sense of dharma’’ in female
friendships. And the longing for a sense of dharma is precisely what is ad-
dressed in the sakhi relationship—through the deployment of religious and
marital symbolism, ritualization, and the elaboration of rules and obligations
entailed in forming and maintaining the bond.
notes
1. Ursula Sharma, another researcher who has worked in Kangra and whom
Narayan cites (Narayan 1986, 66–67), also concludes that married women’s rela-
tionships with other women tend to revolve primarily around female kin living in
the husband’s household. Sharma also notes that the term for friend (saheli) ‘‘is
not much used among rural women except to express the relationship among unmar-
ried girls of the same village’’ (Sharma 1980, 185). Instead, the married women
Sharma studied tend to use fictive kinship terms to describe all nonfamily women,
including those with whom women share a close emotional relationship (1980,
185–190).
2. For more on women’s performance of Kartik puja, see Pintchman 2003,
2005a, and 2005b.
3. Many thanks to Kirin Narayan for bringing to my attention the need to clarify
this point.
4. Three informants insisted that the ritual must take place in front of a Tulsi
plant, but other women cited other divine witnesses, including Satyanarayan,
Ganesh, the Ganges, and Shiva. See Flueckiger 1996 and Jay 1973 for their descrip-
tions of friendship-bonding rituals in Chhattisgarh.
5. I have changed the names of all informants to conceal their identities.
6. Pan is a mixture of betel nut, spices, and other additives rolled up in a betel
leaf and chewed for enjoyment.
7. Jay notes (1973, 154), ‘‘Ceremonial friendships are a means of bridging the
gap between castes when two individuals wish to establish a dyadic relationship
other than the normal one characteristic of members of different castes.’’
8. Jay (1973) describes ritual friendships of all sorts as essentially fictive kin
relationships.
9. Literally, ‘‘you must always digest’’ all the things you hear from your sakhi.
references
Flueckiger, Joyce B. 1996. Gender and Genre in the Folklore of Middle India. Ithaca,
NY: Cornell University Press.
Jay, Edward. 1973. ‘‘Bridging the Gap between Castes: Ceremonial Friendship in
Chhattisgarh.’’ Contributions to Indian Sociology 7: 144–158.
Kakar, Sudhir. 1990. Intimate Relations: Exploring Indian Sexuality. New Delhi:
Penguin.
Narayan, Kirin. 1986. ‘‘Birds on a Branch: Girlfriends and Wedding Songs in Kangra.’’
Ethos 14, no. 1 (Spring): 47–75.
Pintchman, Tracy. 1999. ‘‘Karttik as a Vaisnava Mahotsav: Mythic Themes and the
Ocean of Milk.’’ Journal of Vaisnava Studies 7, no. 2: 65–92.
———. 2003. ‘‘The Month of Kartik and Women Ritual Devotions to Krishna in
Benares.’’ In The Blackwell Companion to Hinduism, ed. Gavin Flood, 327–342.
Oxford: Blackwell.
———. 2005a. Guests at God’s Wedding: Celebrating Kartik among the Women of
Benares. Albany: State University of New York Press.
———. 2005b. ‘‘Domesticating Krishna: Friendship, Marriage, and Women’s
Experience in a Hindu Women’s Ritual Tradition.’’ In ‘Alternative’ Krishna
Traditions: Krishna in Folk Religion and Vernacular Literature, ed. Guy Beck, 43–63.
New York: State University of New York Press.
Raheja, Gloria Goodwin, and Anne Grodzins Gold. 1994. Listen to the Heron’s
Words: Reimagining Gender and Kinship in North India. Berkeley: University of
California Press.
Sered, Susan Starr. 1992. Women as Ritual Experts. New York: Oxford University
Press.
———. 1994. Priestess, Mother, Sacred Sister: Religions Dominated by Women. New
Seymour, Susan C. 1999. Women, Family, and Child Care in India: A World in
Transition. Cambridge: Cambridge University Press.
Sharma, Ursula. 1980. Women, Work, and Property in North West India. London:
Tavistock.
Vatuk, Ved Prakash, and Sylvia Jane Vatuk. 1979. ‘‘Chatorpan: A Culturally Defined
Form of Addiction in North India.’’ In Studies in Indian Folk Traditions, ed.
Ved Prakash Vatuk, 177–189. Delhi: Manohar.
4
Words That Breach Walls:
Women’s Rituals in Rajasthan
Lindsey Harlan
This chapter contemplates some ways that Hindu women’s rituals

expand, dissolve, interrogate, reconfigure, and challenge the concep-
tual space and normative value of domesticity. It focuses on Rajas-
thani women’s performance of ratijagas, celebratory night wakes
that women perform in conjunction with important rites of passage,
especially births and marriages, and refutes the premise that the
impact of women’s rituals performed at home is contained and cir-
cumscribed by location. Investigating ways in which women’s ritual
utterances exert influence, it agues that women’s domestic ritual
performances influence praxis in other domiciles, which constitute
realms exterior to them, but also in nondomestic public space,
often conceptualized in India, as elsewhere, as primarily the sphere
of men.1 Attended by women traveling between households and
audited by familial men who inhabit public spaces more conven-
tionally and routinely than do women, women’s ratijagas, like other
women’s rituals, breach walls and influence social life in realms
beyond their domiciles.
Women and Domiciles: On Location and Orientation
Because many of the rituals that women perform are performed at

home, we tend to think of them, and to encode them, as ‘‘domestic’’
rituals. These include daily puja (worship) of household deities,
vrats (weekly, monthly, annual, or ad hoc ‘‘vows’’ that please deities
and generally entail some kind of fasting), other calendrical rituals
(such as Navratri, the semiannual celebration of ‘‘Nine Nights’’ of devotion to

the goddess Durga), and rites of passage (birth rites, weddings, etc.). Women’s
performance of these rituals in homes is too easily identified with the space of
enactment. Conceptualizing the rituals as essentially locative conveys various
troublesome assumptions, including the notion that their efficacy is limited to
the home and that they are expressions of women’s desires that are norma-
tively bounded or defined by domestic space. In this simplistic equation of
space and purpose, women’s rituals are deemed to be designed to promote the
welfare of the family members, including men, women, and children living
within the homes in which women live and so sharing walls that separate
immediate family from others: other relatives and nonrelatives.
It would be an error to dismiss the locative dimension of many women’s
rituals: the place in which women perform them, the place that women in-
habit, is important. It literally grounds identity. Various scholars have argued
that locus, whatever its dimensions, is a significant and vital ingredient in the
formation and expression of identity. Valentine Daniels has shown the extent
to which location, with its own soil and water, is key to these processes.2 Joyce
B. Flueckiger and Peter Gottschalk, inter alia, have argued that identity is in-
fluenced not only by one’s village, but also by larger areas, whether regions or
village nexuses, which incorporate nearby locations or settlements into one’s
understanding of self and social location.3 Identity is, then, at least partially
constructed by notions of residence, which are tangibly and multiply refer-
enced by indices such as the water brought into and drunk at home and the
soil that becomes the mud and dust traipsed into homes, later to be swept up
by women performing domestic duties.
For many women in India, the domicile is not simply a place of residence
or grounding, but also a location that they leave or enter as brides and that
designates some degree of confinement. For women in different parts of India
and from different social locations in the same part of India, there are varying
degrees and modes of restriction that bind women to their natal or conju-
gal homes and discourage or prohibit their participation in public life beyond
the domicile. I say ‘‘public life beyond the domicile,’’ because there are public
activities in which women engage within the domicile, and, as we shall see,
this fact should prevent any facile contradistinction between domestic and
public, or between ‘‘female’’ and ‘‘male’’ realms.
Among Rajput women, about whom I have written over the years, spatial
restrictions have been among the strictest in India. In the past few decades,
however, the institution of parda has become increasingly elastic, at least
for those women living in and near towns and cities. In and around Udaipur,
where I have conducted much of my research, many of the young women
whom I came to know well in the 1980s are now mothers-in-law who have
decided to loosen or dispense with some or all of the strictures of parda for
their daughters-in-law. It is now not uncommon to see Rajput women
words that breach walls 67
shopping in town. Moreover, many women no longer veil at home in front

of senior family members and visitors.4 That said, women who do not observe
parda in the cities often do observe it when they return to their country resi-
dences in the erstwhile thikanas (estates) that their families once ruled.5
Their family status among villagers remains tied to a code that is largely an-
tiquated in town, especially a town like Udaipur, which has become a trendy
destination for tourists, who often rent rooms in guesthouses owned by
Rajputs.
It would be easy, but naive, to imagine that in the ‘‘olden days’’ of parda,
when women’s movement was highly restrained, the rituals performed by
women would have been faithfully passed down from mothers-in-law to
daughters-in-law and therefore would have stayed strictly within domestic pe-
rimeters and so remained uninfluenced by what they learned in their natal
homes. Such a segregated imaginative construct would unrealistically limit
the influence and impact of rituals to the domicile where they were practiced
at one particular place and time.6 It is important to emphasize that women’s
rituals have always been subject to change, which should not be linked simply
with modernity. The so-called traditional has always changed, and so ‘‘tradi-
tional’’ ritual has changed with every performance. It has been filtered through
the sieves of performers’ diverse improvisational strategies and inspirations. It
has also been apprehended and interpreted variously by those who participate
in it, even if ‘‘only’’ as attendees or auditors, whose presence, expression, bear-
ing, and commentary are, in any case, noted by and reacted to by performers,
whose actions are not only influenced by audience but also, at least partially,
motivated by a desire to impress those who attend. This is so even if perform-
ers’ utmost desire is limited to impressing upon people, including children,
the importance of the particular (albeit ephemeral) incarnation of ritual that
they enact.
Apart from this objection, which insists on the diachronic nonfixity or flu-
idity of traditional ritual praxis, there is the objection about the limitation of
women’s rituals to domicile. However constrained women have been to des-
ignated parameters, they have regularly traveled between domiciles—as brides
or as wedding guests—and so observed variant practices. Fertile opportunities
for encountering variation are provided by women’s ratijagas, for which women
invite relatives and friends, as they do all the deities worshipped by members
of the hostesses’ households. Women’s movement across household borders
for ratijagas has introduced visiting individuals to novel constellations of dei-
ties, some of whom various women may not have worshipped in their homes.
Occasioning discussion of worship, blessings, and miracles, night-long wakes
have allowed women to exchange information about bhakti (dedication to dei-
ties) and enabled women to take away with them new impressions of practice
and even, perhaps, new commitments to self-chosen deities (isthadevtas).7 Thus
the performance of ratijagas for marriages and other rites of passage requires
and promotes a kind of transportation of women and ideas that functions in-
evitably as a source of information, inspiration, and creativity.
The close connection between travel and transformation is not, of course,
limited to the case of Rajput women attending rituals. It is, for example, a
dominant theme in Fatema Mernissi’s book Scheherazade Goes West, which
argues that travel is a source of power for women. Discussing this book about
harem life in Morocco with some students, I recalled a conversation I once had
with an elderly Rajput woman in Udaipur. She told me that when she was
growing up, her father prohibited all females in the home from reading news-
papers.8 Although as a child she must have overheard men’s conversations
about important political events, she, along with others observing parda in
that household, would have been relatively sheltered from politics and so
relatively uninformed, for example, about conditions under which women (and
men) live elsewhere. Traveling to different homes, however, she and other
women sheltered by household codes and strictures would be able to sample
the experiences of women living beyond their domestic borders by observing
what goes on, absorbing directions and narratives, and learning from other
women about what they had learned from other travels and other sources, in-
cluding newspapers.
That sampling must have always been the source of great enrichment and
the impellent cause of some alternative thinking and doing. Here it should be
noted that in arid and rocky Rajasthan, where villages, towns, and kingdoms
have typically had long distances between them, even populations living in
relative proximity have varied in surprising ways: popular wisdom has it that
the language (boli) in Rajasthan changes every few miles.9 With the change
of language comes differences in idiom, expression, and so on. For many
women married into Udaipur families, the Mewari tongue that their husbands
have spoken since birth is a dialect to be learned. In-marrying brides in Mewari
homes have created an insurgence of language and of notions that language
constructs.
Thus marriage has been followed by a period of transition with words and
ideas surely being exchanged or loaned, perhaps even on a long-term or per-
manent basis. Add to this circumstance the tradition of high social status
families making alliances with other important families from erstwhile king-
doms, which are inevitably distant (both inside and outside of Rajasthan), and
there arises the opportunity for fluid, if perhaps at times slowly seeping, ac-
commodation, both to new and old ways, among cohabiting women. When these
women stream into and out of the homes of other women, who have their own
diverse natal dialects/languages and upbringings, there is often much to talk
about. Exchanging news, gossip, and recipes and perhaps discussing films and
television shows, women learn about lands and customs far beyond the bor-
ders of domicile, and even of region.10
On a macro level of analysis, the movement suggests that women’s

practice of rituals in their individual homes has surely, if often subtly, altered
the religious terrain inside and outside their dwellings. Emblematic of this
change in Mewar is the legendary bhakti poet Mira Bai, who upset Mewar’s
ruling family when she arrived as a bride with an unshakable commitment to
her isthadevta, Krishna.11 Treasuring her worship and considering it more
important than her in-laws’ familial goddess worship, she affronted members
of her sasural by according Krishna priority. Eventually she abandoned parda,
domicile, and kingdom to wander the roads and forests while praising Krishna
before ultimately taking refuge in his temple at far-away Dvaraka. There,
many devotees believe, she miraculously merged with his icon when her natal
family tried to retrieve her to Mewar.
Whatever the impact that Mira’s legend of rebellious devotion may have had
on women throughout the centuries, it is clear that she is understood today as
having had tremendous clout: there is a well-visited Mira temple at the great
fortress of Chittor, where her devotion is celebrated. Residents and visitors
know of patronage of the ‘‘Mira Bai temple’’ (as the Vaishnava temple has
become known) by the royal family, which originally rebuked her for her
insolent worship and departure, then ultimately and futilely demanded her
return. This shows a tacit, even implicit, recognition of a prominent instance
of a bride bringing with her a devotional agenda.
Mira is patently exceptional in her rebelliousness: a brides expects, and is
expected, to tour and pay respect (dhok) at the prominent temples in her hus-
band’s village or estate (thikana) and to honor all deities installed there, as well
as the other deities worshipped in her sasural’s shrine or shrines. Brides
typically conform to expectation but are not, however, asked to forswear all
prior theological commitments and ways of worshipping. Moreover, women’s
narratives about ancestral women’s importation of their natal family’s god-
desses parallels and complements men’s narratives of goddess procurement
through icon seizure, a feature of the spoils of war. Victorious kings often
demanded brides and icons. That the two might come together and so directly
import ritual at times seems reasonable and is supported by indigenous nar-
rative.12
In short, brides arriving at their sasural bring with them a new wind that
may stir the course of religious thought and praxis. Women singing at rati-
jagas for weddings and other occasions receive exposure to new religious acts
and ideas, as well as tidings of divine blessings and miracles at various peo-
ple’s domestic shrines or temples, to which pilgrimage may later be made
on subsequent visits that also, of course, allow for recreational visiting. Thus
women, who are often represented as conservators of tradition, are also agents
of change, who usher in new influences that may alter praxis at home and
beyond.
Before offering some specific ideas about how this takes place, I pause to
frame this discussion with a non-Indian allusion. Briefly thinking through
some of the dynamics in the film Chocolat will help to connect the erosive and
etching winds of change in India with those effected by women elsewhere.
Realizing that not all readers will be seasoned Indologists, I offer the example
to demonstrate that some myth and ritual can be construed as revealing the
influence of women on religious praxis and the power of women not only to
bring others into their domestic realm, and so render and exert nonprivate,
nonfamilial influence, but also to influence what goes on beyond their space,
in the homes of others and in the institutional practice of religion outside all
domestic spaces.
Chocolat
In the opening scene appears a sunny, medieval-looking French village. Almost

immediately, however, the sunshine vanishes and a snowstorm hits. Blown
into this town by the blustery north wind are two female travelers in Red Riding
Hoodish red capes, which offer the only visual relief from the village’s stony
architecture. With their vibrant attire, so evocative of fairy tale, they foreshadow
the infusion of enlivening change—of novel and alien customs, heretical ideas,
and transgressive joie de vivre—that their residence is soon to effect. The fore-
shadowing is not subtle. As soon as the hooded travelers arrive in this town,
which has been rendered lifeless and deserted by the unanimous church atten-
dance of the town’s Catholic denizens, the church doors blow open with a bang
and disrupt the solemnity of the service, which, we are to understand by the
wriggling and dozing of the attendees in their pews, is uninspiring.
Most visibly upset by the wind’s disruption is the Comte de Reynaud, re-
presented in the film as emblematic of solemn religious devotion and as stalwart
in upholding and enforcing Catholic morality, with its sharply gendered division
of space and duties. After learning of the arrival of the newcomers, mother and
daughter who have just rented a defunct and dusty patisserie, he greets them and
invites them to attend church. The mother, Vianne Rocher, declines. Visibly
dismayed by her unabashed announcement that they do not attend church, and
by her revelation that she is an unwed mother, the count departs abruptly.
Further provoking the count is Vianne’s opening of a chocolaterie during
Lent. Outraged, he forbids churchgoers to patronize the shop, with its vibrant
Mayan decor. He informs Vianne that one of his ancestors, a previous count,
whose grimacing statue guards the square outside the church gates, routed
the Huguenots and that, as his heir, he should have little trouble shutting her
down and driving her out.
The count is wrong. He is no match for Vianne, who verily incarnates the
winds of change. Raised in Central America and belonging to a group of
nomadic healers, her mother married a French apothecary on his mission to

seek new cures. Having returned with him to France and given birth to Vianne,
she ultimately eschewed settled life with him and took off with Vianne to travel
and to vend cacao remedies. Vianne, in turn, had a daughter and took up a
peripatetic life selling curative and exquisite chocolates, laced with exotic chili
pepper.
Although forbidden by the count and by Catholic Lenten tradition to buy
Vianne’s chocolates, which the count images as the work of the devil, some of
the villagers most afflicted by stagnant social norms and unyielding expec-
tations of conformity to rules about proper gardening, husband maintenance,
and courtship succumb to the allure of Vianne’s inviting establishment. Ad-
miring the luscious chocolates in her window, they heed Vianne’s plea to enter
and accept, albeit at first tentatively, her overtures of hospitality and friend-
ship. Gifting samples, she initiates the process of rescuing them from lone-
liness, shyness, timidity, sexual lethargy, and domestic abuse. Consuming
Vianne’s carefully formulated and soothing sweets, the socially alienated vil-
lagers find themselves altered and strengthened in their sense of worth and
purpose.
At Vianne’s urging, these deviating townsfolk enter, and thrive in, the build-
ing housing both Vianne’s shop and her living quarters: the chocolaterie is lit-
erally ‘‘where she lives,’’ a slang expression economically conveying the
connection between domicile and identity. Having rented the room above
her shop, she lives where and as she heals. Moreover, by the end of the film,
her patients have become apprentices occupying Vianne’s inner sanctum, her
kitchen, where she exerts nurturing power. As a much maligned and besieged
Vianne ponders whether to leave them to wander away with the wind, they staff
her kitchen and wield the tools of her trade as they concoct chocolates to take
into town for a planned chocolate festival. Having learned her craft, they take
Vianne’s ways and their influence to the streets, where they can visit change on,
and bring happiness to, the previously uninitiated.
Whether or not one is willing to see Vianne and her followers as practicing
a ‘‘religion of chocolate,’’ one should easily recognize that there are aspects of
their practice that are evocative of, or at least similar to, the religious. The count
represents her praxis as heretical. Moreover, she deploys ritual divination: she
asks customers to gaze at a spinning painted disk and then to tell her what they
see. According to their responses, she discerns ‘‘their favorites’’ and fills their
divined prescription with healing morsels. Moreover, the chocolate festival,
heralding the arrival of spring and employing street performers resembling
those abounding at contemporary May Day celebrations, implies benign, if
paganesque, transgression. Thus the chocolates created in Vianne’s kitchen by
her disciples ultimately make their way into the town square, just outside
the church gates and directly under the nose of the memorialized defeater of
Huguenots. What began in the kitchen of the exotic woman and daughter
who blew into town proceeds outward and onward into the town square, the
quintessentially public place.
Among the attendees at the festival is the vanquished count. Having starved
himself during Lent, he becomes possessed by an uncontrollable craving for
chocolate. That his transformation, his acceptance of desire, is not ephemeral
is suggested by his decision to date his secretary now that he realizes that
his wife has completely abandoned him, and also by his withdrawal of his
unwelcome influence over the town’s young priest, who now preaches to pa-
rishioners tolerance and acceptance of some nomads, ‘‘river rats’’ anchored in
the river bordering the town. The breezes blown from Vianne’s kitchen thus
continue to move outward and so affect not just the center, but the periphery,
of the community.
The change visited on the town is sufficiently transformative that the town
now exerts a boomeranging influence on Vianne: it tempts her to abandon her
perpetual roaming and abide in one location. Because her actions have trans-
formed the place where she lives by altering its perspectives and conventions,
she is now free and able to live there and to resist the urge to wander. Having
changed the town, she can, as her travel-weary daughter puts it, live like a
normal person.
What are we to take away from this exotic story in order to understand better
the wind blown by, and as, women in India? There are many possible things one
could take away, but for the purpose of this chapter, let me focus on three.
1. As a traveler, a woman can be understood as incarnating, as well as
experiencing, change. In the context of ritual, as in life generally, she serves as
an agent shaping to some extent her environment even as she adjusts to it.
This agency may or may not take the form of what is generally called ‘‘every-
day resistance.’’13 A woman may or may not understand the implementation
of change as subversive even though it does, in effect, transform the house-
hold, which is so often too easily conceptualized as tranquilly reproducing
generations in a stable and conservative patriarchal culture.
In Chocolat, the circumstances of the radical catalyst Vianne, who is, after
all, a fictional figure, are atypical: they little resemble those of postwar French
women, and they certainly diverge from those of Rajput women in whatever
time period. Vianne has no man. This heroine is a ‘‘loose woman,’’ unsuper-
vised, and so unbound by male authority. It is therefore unsurprising that she
will transgress the town’s mores by having sexual relations with a nomadic
boatman. She is blown by the wind, but she also is the wind, blowing away
what is stagnant even if by the end, also adjusting, that is to say, being calmed
and soothed by the rhythm of settled life in a small town.
2. The domestic location of women is not inherently private. Access may
be ‘‘by invitation only,’’ but women incorporate outsiders into the bodies of
their households and exert their influence on others in these bodies in a way
that renders their apparently private worlds at least intermittently public. At the
same time, of course, their exposure to outsiders influences them and so mod-
ifies their domestic social body and praxis. Familial and nonfamilial guests
come and go, but their influence, like the influence of incoming brides, leaves
non-ephemeral impressions.
In the film, Vianne makes chocolates that lure others into her cafe/home
and, through the influence she wields within the walls of her living space,
manages to alter radically the lives of those she feeds, who ultimately learn to
make her recipes in her kitchen. Influencing the behavior of consumers, who
eat her food but also take in her sparse but powerful advice, she effects changes
that produce both familial cohesion and discord. Consumed by the coarse and
unappealing husband of one consumer/patient, the chocolates cause him to
desire his wife, even as she cleans their toilet: in later scenes, he appears to be
attentive and supportive of her as she attends a transgressive chocolate soiree
catered by Vianne for her landlady. After eating Vianne’s chocolates, Josephine,
a key character whose nervousness and kleptomania portend imminent break-
down, is emboldened to leave her physically and mentally abusive husband,
Serge. Thus what comes from Vianne’s kitchen is radically transformative. It
strengthens, creates, or, in certain circumstances, destroys domestic social
bonds in other domestic spaces.
3. The influence of women’s praxis overflows the bounds of domesticity
and seeps into, or even floods, quintessentially public dynamics and institu-
tions. This is true, of course, in realms other than religion. Note an example
from economics: European women’s utilization of sugar in their recipes once
supported the institution of slavery in Caribbean islands.14
In Chocolat, this point is best represented in Vianne’s radical transfor-
mation of the count, who is also the mayor. When he attends, along with the
priest, the paganesque chocolate festival, we know that the townspeople are
now liberated from his dogged interference, which he justifies as enforcing
Catholic morality. By the end of the film, the town is no longer bound by the
moribund ethos it calls tranquillite.
With these three points in mind, let us turn to ratijagas to see how they belie
domestic boundaries and wield influence on other places and institutions.
Ratijagas and Their Spheres of Influence
We begin with point one. As a traveler, a woman incarnates and experiences

change: she is an agent shaping her environment even as she adjusts or as-
similates to it. As we have seen, women coming into a household, as guests or
brides, are influenced by what they encounter, while the women in the house-
hold into which they come are influenced by them and will also be further
exposed to alternate ways and things when they visit other women in their
domiciles. This occurs when women perform ratijaga rituals. Playing for
hostesses and guests, the musicians (harmonium players and drummers)

typically sing songs that facilitate identification, and even substitution, of one
deity for another. Identification can, of course, effect temporary ritual homog-
enization, which may counteract or balance, rather than effect, differentiation.
But familial deities, whether kuldevis (familial goddesses) or locative, cultic dei-
ties, are worshipped and conceptualized in situ—in their (household or tem-
ple) shrines—as discrete deities with local and cosmic identities.
Let us take the example of kuldevis. Singing songs to invite and honor the
familial kuldevi, a goddess with whose identity many singers may not be fa-
miliar, musicians strategically employ generic names, such as Mata, Bhavani,
and Amba. In one such song, there is a common and recurrent invocation of
the goddess Amba, yet identified with her, in the very first verse, is the well-
known Mewar cultic deity Avadi Mata (Avadi Mother), invoked also simply as
Avada and as Avadi Rani (Queen Avadi):15
Reveal yourself again and again, O Amba.
Reveal yourself again and again.
Bhavani, reveal yourself again and again.
When you reveal yourself, I am happy, Avada.
When you reveal yourself, I am happy.
Amba, reveal yourself again and again.
The next two verses identify Amba also with Yogmaya, a well-known epithet
with a pan-Indian range of recognition. But in the verse following them, this
goddess is identified specifically with the Sisodiya Rajput kuldevi, Bayan Mata
(Bayan Mother), who is invoked here as Bayan Rani (Queen Bayan):
Amba, the necklace on your breast is beautiful.
Your earrings are fully inlaid, your earrings, Avadi Rani.
Your earrings are inlaid, inlaid, Oh Bayan Rani.
Reveal yourself again and again.
Oh Sukh Devi, reveal yourself again and again.
This verse invoking the family kuldevi, who specifically watches over fam-
ily members and protects them against outsiders, also summons Sukh Devi,
whose shrine in Bedla thikana (a tributary estate in the erstwhile Mewar king-
dom), is well known and much visited by people from various nearby places and
caste backgrounds. Given the invocation of well-known or transfamilial deities,
as well as the one kuldevi, one should note that the generic epithet Ambaji surely
also, and more specifically, references for those living in the erstwhile Sisodiya-
ruled kingdom of Mewar, the Ambaji temple in Udaipur, which is patronized by
the royal family.
Thus, this song identifies all deities and so allows everyone to relate to an
inclusive deity belonging to no one particular place or no particular set of
persons. Nevertheless, while centrifugally stating that these goddesses are ‘‘the
same,’’ the verses also centripetally retain discrete names and epithets dif-
ferentiating all goddesses with their sets of devotees. I have written at length
about the homogenization and differentiation of goddesses traveling between
residences with brides.16 In the context of ratijagas, it should be emphasized
that exposure to the kuldevis worshipped by women in different families is
made through the lyrics of ratijaga songs. The same principle holds for locative,
cultic goddesses, such as Mewar’s Avadi Mata.
Moreover, in songs such as this one, musicians invoking and pleasing
particular goddesses and identifying them with other deities map a constel-
lation of divinities with complex relations, including identification, but also
homology and other forms of association. Together with songs sung for other
divinities, such as Bheruji, Satimata, heroes ( jhumjharjis, bhomiyajis, and an-
cestors ( pitrs or purbaj), the ratijaga singers convey significant local knowl-
edge, which expands the horizons of any traveling women’s information and
allows for the diverse construction and absorption of various divine identities,
with their geographical (dwelling and shrine) referents and their associations
with outside (hostesses’) occasions.17 These associations infiltrate/inform the-
ology and ritual performance in different domiciles.
Women, then, inform their own and others’ understandings of divinity
with ritual experiences and discussions in their homes. They inform their in-
laws’ conventions when they arrive as brides. Traveling to one another’s homes
to participate in ratijagas, they see familiarity in the ratijaga observance of
others, while learning about ritual practices and divinities who may be im-
ported to help them back in their homes.
By this time, it should be clear that, as point two contends, the home in
which a ratijaga is performed is not simply a family’s private space and that
the realm in which women perform the ritual is not simply private or domestic,
as opposed to public, space. Women living at such a house temporarily, but
repeatedly, render their home public when they exert the power of invitation.
Whereas Vianne attracts guests and changes/is changed by guests with the
lure of chocolate, Rajasthani women attract ratijaga participants—outsiders,
whether human or divine—with promises of hospitality. This sets the terms
of mutual influence and, perhaps, endearment.
In the days before printed invitations, these promises were typically made
with grains of rice, in other words, food, one of the most basic elements of
hospitality and, in this context, a metonym. In the Rajasthani epic Pabuji, the
songs sung by performers as Rajput women’s ratijaga songs at the weddings
of the heroes Pabuji and Gogaji (who has an independent cult but who also
appears in Pabuji’s epic) describe the invitation of various familial deities via
rice grains delivered to each god by the goddess Amba. Thus with women
singing praises for deities and the goddess also summoning all divine par-
ticipants, the point is driven home: females, divine and human, are respon-
sible for drawing outsiders inside. They are represented as bringing into the
household others, travelers whose evanescent presence as guests blesses the

family and reflects/produces its social capital.
Thus we see the conjoint power of women and goddesses to attract and to
bring together guests from other realms, including the divine. One might ask,
of course, what sort of power is this power of divinities, whose existence may
be challenged? Attributing to women the power to attract and make use of
divine power in this connection might be deemed to be simply crediting them
with the illusion, rather than the exercise, of power. And yet, few would doubt
the power of Brahmins, whose status and influence are mostly unquestioned
because of their perceived control over powers unseen. In the case of both Brah-
mins and women, power is claimed, but also perceived, by others: men,
children, and women.18 Women are expected to perform rituals that benefit
not just themselves, but others, including men and children. Furthermore, in
matters religious, women’s knowledge is often considered authoritative. I have
noted that women typically defer to men when it comes to matters they deem
to be historical (itihasik), but men often defer to women when it comes to
matters of theology and ritual.19
Women’s power to invite outsiders is not absolute—men often wield
substantial influence when it comes to determining which men will be invited
to visit during the ratijaga and so whose wives must accompany them.
Moreover, like women, men worship isthadevtas, chosen deities, whom women
are expected to please for their own good and for the benefit of the family.
Despite some sharing of power, it is primarily women’s responsibility
to organize and to invite other women to attend ratijagas, and women trav-
eling to ratijagas are understood as having been invited by the women who host
them. During this time in which women host women, hostesses effectively
construct an overwhelmingly female space by excluding household men
(though not children) from their vicinity and so claim authority for the divi-
sion of social space.20 Barring men, women assume control and define their
physical and personal boundaries.
Denying men access when outsiders visit is an exertion of moral authority
that effectively removes men and relegates them to the position of blind au-
dience. Men outside the women’s space can hear the songs women sing but
cannot access or influence immediately the women who sing. Thus, for exam-
ple, during a ratijaga I attended, when a middle-aged man from the hostesses’
household peeked around the corner for a moment to see what was up, he
was quickly banished by women, albeit gently and with much gaiety.
The power of women to shape the social world through their activities
as hostesses inviting, arranging, and managing guests is much observed by
writers with diverse agendas, including, for example, Stephanie Jamison,
writing about Vedic ritual, but also Virginia Woolf, whose fictional hostesses
make it clear that even a nonreligious party is an expression of self, truth,
artistry, and power, particularly the power to create and guard social order.21
Through invitation and hospitality, women construct a social sphere for them-
selves and their families: they build, break, and renegotiate alliances by in-
cluding and honoring some while excluding and degrading others.
In this ephemeral world of women, which encompasses worlds exterior
to it, ratijaga ritual enacts the passage of time in such a way as to shape the
future for the holders of the ratijaga, by securing blessings but also by in-
structing others, including the household’s daughters, who are often termed
‘‘guests’’ in, and who will marry out of, their natal households. Inviting both
deities and other women, often accompanied by their children, women map
and configure their social world, which exists both inside and outside their
domiciles. In mapping, which both reflects and shapes or charts social in-
teraction, women’s rituals encompass and exclude certain people and deities
and so define both social and theological realms.
The mapping of social relations is an explicit and common feature of
ratijaga songs. Let us look, for example, at one ratijaga pamphlet’s song for the
hero Tejaji, whose name or history is not even mentioned in the song, except
for the indexical, printed title, ‘‘Tejaji.’’ This song lists various relations, such
as sister and sister-in-law. Each is said to be a ‘‘big flower,’’ though the role of
conjugal flowers takes precedence over the natal flowers. Thus, the sister might
give one a blouse, but a blouse’s sleeve rips easily. By contrast, a sister-in-law
gives one a blessing, and a blessing ‘‘lasts forever.’’ The song goes on to map
other familial relations and, while affirming the importance of natal ones,
maintains the superiority of conjugal ones.22
Another song, which I heard at a ratijaga at the home of a girl who de-
scribed her family as ‘‘little brother Rajput’’ (i.e., a family tracing its line from
a younger brother of a ruler or nobleman and so not inheriting parental
property according to the principle of primogeniture), clearly maps the social
world represented and created by the performance of ratijaga ritual.23 Naming
participants and describing their participation, women count themselves pres-
ent and render themselves responsible for the ritual that so many divinities
are attending. The song reads as a veritable register of sociability in which
everyone works together for the mutual procurement of blessings:
Bhagvat Kanwar put the wick in the lamp.
Mahavir Singh’s wife filled it with butter.
Light the lamp for the four watches of the night.
In this song, as in the ratijaga ritual, women are featured as agents with a
shared purpose and an espoused cohesion that bonds family to family in
terms of hospitality and responsibility. The song represents women as coa-
lescing with other women to effect good fortune for the host family and also
for the participants, who share in the benefits of ritual placation.24 Like Amish
men together building a barn for a recently married couple, these women
conspire to give beneficiaries what they need for a bright future.
This social mapping could be construed to be inherently conservative. It
could be seen as enforcing a code of cooperation that preserves patriarchal
norms and so the status quo. And yet, by invoking names, it inscribes indi-
vidual people onto the social map, and these individual people are not simply
substitutable units. Named persons with discrete identities and histories, they
are agents who presumably would not always agree with each other nor accept
each other’s lives uncritically. Furthermore, the song includes various people
who perform the same function, whatever their relations to each other. Thus
sisters and sisters-in-law are joined in purpose, even as social code, including
code conveyed in the lyrics of ratijaga songs, may put these in tension.
Having argued that the world of women performing ratijaga ritual is far
from private and that what goes on in one household has a charge or force
that affects and helps to define other households, I turn to the final point:
women’s ritual praxis saturates the boundaries of domesticity, including mul-
tiple domiciles, and penetrates overtly public dynamics and institutions. In
Chocolat, one woman making chocolates and selling them out of her home-
cum-shop transforms the mayor, the church, and ultimately the entire town,
whose members congregate joyously in the town square. Recalling that this is
dramatic fiction, we might ask, are there any parallels?
There are many. Think about the fact that the epic Pabuji, which is per-
formed publicly by Nayaks (often by a male-female pair), incorporates repre-
sentations of Rajput women’s songs into its episodes (parvaros) treating Pabuji’s
and Gogaji’s marriages. Performing this epic in public places for diverse au-
diences, Nayak performers represent Rajput women as singing songs that
summon the gods, as ratijaga songs do.25 True, the Nayak singers may be viewed
as co-opting the voices of Rajput ratijaga singers, but co-optation is often an
act of power recognition and appropriation. Nayak performance of these
songs represents Rajput women as having power, even as singers deploy power
by representing them according to their performance agendas in songs that
stand for, but in their dense narrativity little resemble, the ratijaga songs that
Rajput women sing in their homes.26
Another example is the exportation of women’s songs into various cultic
contexts. In hero shrines, women, typically musicians but also women from
diverse communities, sing ratijaga songs for Rajput heroes. One female mu-
sician, whom I watched perform many such songs in Udaipur, regularly sings
ratijaga songs for the murdered prince Surtan Singh in a small, satellite tem-
ple.27 During the annual festival for Surtan Singh in his imposing, main
temple in Udaipur, other drumming and harmonium-playing women musi-
cians come to sing songs for him and summon him to the festival, which
culminates in a public ratijaga, attended by both men and women.
A similar situation is found at shrines for Bheru Singh, who was mur-
dered by his relatives because he insulted them when he wore too fancy a coat
for his status. Bheru Singh’s memorial pavilion (chhatri) is to be found at
Mewar’s royal cremation ground in Ahar. There is also a temple for him in
the mansion (haveli) once owned by Bheru Singh’s family. Both shrines for
this hero, who, like Surtan Singh, was murdered by order of the royal family,
draws women singing songs that Rajput women sing in ratijagas.28 Having
taped and transcribed the songs, I showed the lyrics to a Rajput woman, who
immediately recognized them and noted that they were ‘‘sexy.’’ Like the lyrics
sung for many other heroes, they describe the hero as exceedingly alluring.29
The singing in a cultic context of songs that are sung by descendants of Rajput
heroes at home surely represents Rajput women’s ritual tradition as author-
itative and invests Rajput women with the power to enact and claim authen-
ticity in a culture that is heavily influenced by Rajput norms and values.30
Before leaving this discussion of lyrics sung by women in the homes of
Rajput heroes’ descendants but also performed in cultic shrines, let me note
that the participants in such cults are of exceedingly diverse origins and so the
dispersion of ratijaga songs is wide.31 In cultic worship, people form impres-
sions of Rajput heroic history and authority that are filtered through women’s
songs of praise and adoration.32
To Bheru Singh’s Ahar shrine come men and women from various com-
munities, including some Muslims and many Jains, among them the main
bhopa (medium), who is regularly possessed by Bheru Singh. As Lawrence
Babb has observed, Jain families have adopted martial heroes as emblematic of
valorous conquest, even as they have transformed the image of these heroes
from violent to virtuous warrior by investing them with Jain values, particularly,
non-injury.33 At the cult of Bheruji, this process occurs among Jain women, who
worship at the shrine of a Hindu hero while espousing Jain identity and norms.
Equally intriguing is the veneration at Bheruji’s shrine by a young Muslim
woman with whom I struck up a conversation while doing fieldwork. I asked
her how she squared her veneration of a Hindu hero with her practice of Islam.
She said that she was forbidden by Islam to worship an image and that she would
not bring an image of Bheru Singh into her home, but that she can, and does,
light a devotional diya (oil lamp) for him at home. This ritual act, she said, does
not violate Islam or trouble her in-laws, whose welfare she aspires to promote
through ritual veneration. In effect, she could perform her aniconic worship
for the cultic hero at home, having taken from the cult her mental impression,
based on his icon, the bhopa’s possession, and depictions in ratijaga songs.
Thus we see that the songs that women sing for their ancestral heroes at
home are integrated into cultic praxis and help to represent, through verbal
icons and descriptions of veneration, Rajput heroism in diverse cultic milieus.
The impressions that women attending cultic rituals take away from their
communal worship at shrines and bring back into their homes are a creative
admixture formed from elements of ratijaga songs, other songs (such as bhajans
and aratis), public worship of adorned images, and possession manifestations.
Together, these impressions influence both cultic theology as construed by
devotees (note the Muslim informant’s theologizing) and also the form and
nature of their individual domestic practice (in the Muslim woman’s case, her
deployment of a diya in lieu of an icon).
To conclude: women’s ratijaga rituals are not simply a family affair. They
are scenarios enacting public praxis. When guests are invited in, as they are for
ratijagas, the putative division between private and public space is suspended.
Granted, not all of the public is allowed in, but that is true for many other
public spaces, including businesses operating in public space but regarded as
‘‘private enterprises.’’ Moreover, traveling between households, women wit-
ness ritual devotion and trade experiences: they exchange news, influence one
another’s views, and so color the lenses through which travelers will appre-
hend their own praxis and social interaction when they return to their own
domestic spaces. Finally, the views and actions of women at home inform and
influence what goes on outside the home. The words of women singing rati-
jaga songs in their homes are expressed, even as they are adapted, in the very
public veneration of their culture’s heroes, who in various ways reflect and
shape cultural ethos. The words of women singing ratijaga songs at home
breach walls to permeate public culture, with its fluid assumptions about and
assessments of divinity and value.
notes
1. On this division elsewhere, see Buchanan 1996.
2. Daniels 1984.
3. Gottschalk 2000 and Flueckiger 1996.
4. For further reflections on the elasticity of parda see Harlan 1992.
5. My trip in 2002 provided many opportunities to discuss parda and its morph-
ing into an increasingly multiform phenomenon, with individual families mak-
ing increasingly liberal decisions about women’s confinement, concealment, and
mobility.
6. On the problematics of ‘‘tradition,’’ see Appadurai et al. 1991.
7. On women’s importation of goddesses, see Harlan 1992 and 2001.
8. See Harlan 2003, 41.
9. I have also heard the expression used with kilometers.
10. On the impact of television, see Steindorph 2004.
11. There are many variations of this legend. The account I employ here is
one familiar in many of its elements to women living in Udaipur. For a discussion
of variants of the Mira legend, see Harlan 1992, Mukta 1997, and Martin
(forthcoming).
12. See Harlan 2001 and 1992.

13. For classic works on ‘‘everyday resistance,’’ see Haynes and Prakash 1991,
1–22; and Scott 1985.
14. See Sussman 1994, 48–69; and 2000.
15. Her name also appears in English as ‘‘Avara’’ or Avari.
16. Harlan 2000 and 1992.
17. On the associations among these various deities in song, see Harlan
2003, ch. 5.
18. On the ritual complementarity of and the tension between women and
Brahmins, see Harlan 1992 and forthcoming (b).
19. On this phenomenon and itihas, which often blends history with myth,
see Harlan 2003.
20. In effect, they temporarily reestablish the zanana, the women’s quarters
of a traditional Rajput home in centuries past.
21. See, for example, Mrs. Dalloway; ‘‘The New Dress,’’ esp. 64; and To the
Lighthouse, esp. 94–97. I thank Julie Rivkin for instructive comments on Woolf’s
parties as ‘‘aesthetic moments’’ countering disorder, February 2002.
22. Ratijaga Ka Git, n.d.
23. Some of the Rajputs said they were reluctant to accept the claim of ‘‘lit-
tle brother Rajput’’ without genealogical proof and presumed what they call ‘‘Daroga’’
provenance, that is to say, descent tracing back to a union between a Rajput man
and a woman of another caste. See Harlan 2003.
24. For further reflection on this song and others, see Harlan 2003 and 1995.
25. For detailed analysis, see Harlan 2003.
26. On this appropriation, see Harlan 2000.
27. This woman refers to herself as a rajgahak, ‘‘royal singer,’’ to dispense
with any impression that she was ever merely a musician for hire by just anyone.
Others refer to her as a Dholhin, a member of a ‘‘drummer caste’’; she contests
this designation.
28. On heroes, death, and agency, see Harlan 2003.
29. On lyrics, see Harlan 2003, ch. 5.
30. Harlan 2003, ch. 2.
31. It is not necessary to establish where such songs were composed originally;
what matters, in terms of authority and as in the case of the songs in Pabuji, is
the perception that the songs are ancestral songs.
32. Extensive analysis of the cults of Surtan Singh, Bheru Singh, and other
Rajput heroes is to be found in Harlan forthcoming (a).
33. Babb 1996.
references
Appadurai, Arjun, Frank J. Korom, and Margaret A. Mills. 1991. ‘‘Introduction.’’ In
Gender, Genre, and Power in South Asian Expressive Traditions, ed. Arjun
Appadurai, Frank J. Korom, and Margaret A. Mills, 3–20. Philadelphia:
University of Philadelphia Press.
Babb, Lawrence A. 1996. Absent Lord: Ascetics and Kings in a Jain Ritual Culture.
Berkeley: University of California Press.
Buchanan, Constance. 1996. Choosing to Lead: Women and the Crisis of American
Values. Boston: Beacon.
Chocolat. 2001. Directed by Lasse Halstrom. Miramax.
Daniels, Valentine. 1984. Fluid Signs: Being a Person the Tamil Way. Berkeley: Uni-
versity of California Press.
Flueckiger, Joyce Burkhalter. 1996. Gender and Genre in the Folklore of Middle India.
Ithaca, NY: Cornell University Press.
Gottschalk, Peter. 2000. Beyond Hindu and Muslim: Multiple Identity in Narratives
from Village India. New York: Oxford University Press.
Harlan, Lindsey. Forthcoming (a). Lasting Impressions: Representing Heroism in Con-
temporary Hero Cults. Manuscript under preparation.
———. Forthcoming (b). ‘‘Nala and Damayanti’s Reversal of Fortune: Reflections
on When a Woman Should Know Better.’’ In Nala and Damayanti, ed. Susan S.
Wadley and Joyce Flueckiger. New Delhi: Chronicle Books.
———. 2003. The Goddesses’ Henchmen: Gender in Indian Hero Worship. New York:
———. 2001. ‘‘Battles, Brides, and Sacrifice: Rajput Kuldevis in Rajasthan.’’ In Is
the Goddess a Feminist? The Politics of South Asian Goddesses, ed. Kathleen Erndl
and Alf Hiltebeitel, 69–90. Sheffield, England: Sheffield Academic Press.
———. 2000. ‘‘Heroes Alone and Heroes at Home: Gender and Intertextuality in
Two Narratives.’’ In Invented Identities: The Interplay of Gender, Religion, and
Politics in India, ed. Julia Leslie and Mary McGee, 231–251. Delhi: Oxford Univer-
sity Press.
———. 1995. ‘‘Women’s Songs for Auspicious Occasions.’’ In Religions of India in
Practice, ed. Donald Lopez, 269–280. Princeton, NJ: Princeton University Press.
———. 1992. Religion and Rajput Women: The Ethic of Protection in Contemporary
Narratives. Berkeley: University of California Press.
Haynes, Douglas, and Gyan Prakash. 1991. ‘‘Introduction: The Entanglement of
Power and Resistance.’’ In Contesting Power: Resistance and Everyday Social
Relations in South Asia, ed. Douglas Haynes and Gyan Prakash, 1–22. Berkeley:
Martin, Nancy. Forthcoming. Mirabai Manifest: The Many Faces of a Woman Poet-Saint
in India. New York: Oxford University Press.
Mernissi, Fatema. 2001. Sheherazade Goes West. New York: Washington Square Press.
Mukta, Parita. 1997. Upholding the Common Life: The Community of Mirabai. Delhi:
Ratijaga Ka Git. n.d. Edited by Sarasvati Devi Bhensali and Jnandevi Teli. Ajmer:
Sarasvati Prakasan.
Scott, James. 1985. Weapons of the Weak: Everyday Forms of Peasant Resistance. New
Haven, CT: Yale University Press.
Steindorph, Sally. 2004. ‘‘Missed Messages: TV Producers and Rural Viewers Dis-
cuss the Representation of Villages on Indian Television.’’ Paper delivered at the
18th European Conference on Modern South Asian Studies, Lund, Sweden, 6–9
July 2004.
Sussman, Charlotte. 2000. Consuming Anxieties: Consumer Protest, Gender and Brit-
ish Slavery 1713–1833. Palo Alto, CA: Stanford University Press.
———. 1994. ‘‘Women and the Politics of Sugar, 1792.’’ Representations, no. 48
(Autumn): 48–69.
Woolf, Virginia. 1973. ‘‘The New Dress.’’ In Mrs. Dalloway’s Party, 61–73. New York:
Harcourt.
———. 1927. To the Lighthouse. New York: Harcourt.
———. 1925. Mrs. Dalloway. New York: Harcourt.
5
Threshold Designs, Forehead
Dots, and Menstruation
Rituals: Exploring
Time and Space
in Tamil Kolams
Vijaya Rettakudi Nagarajan
We make the kolam to indicate auspiciousness and to prevent rit-

ual pollution.
—Janakimami, Tirunagar village
The Kolam on the Front Stoop and the Red Dot

on the Forehead
Throughout the Hindu cultural and religious world of Tamilnadu,

India, women’s ritual drawings called kolam—ground rice-flour
designs in white—mark space, including the thresholds of homes,
the edges of streets, and trees. These women’s ritual designs
also mark time: dawn and dusk; the month of the winter solstice,
or Markali; and the abundant rice harvest festival called Pongal.
Sometimes outlined in red clay, and sometimes a matrix of dots
circled by one curvaceous line, these kolams are ubiquitous, drawn
daily in public places by millions of Tamil female hands, and visi-
ble to everyone (figure 5.1).1 Also abounding in everyday life is
the visible ritual marking of the red dot floating in the middle of
Tamil women’s foreheads, the pottu as it is called in Tamil, or
bindi in Hindi or Sanskrit. The design is usually a filled circle of
red, or shades of burnt sienna, a reflection of its mercury oxide
figure 5.1. A Tamil girl making a labyrinth kolam, Mayiladuthurai, Tamilnadu.

Photo by Vijaya Nagarajan.
nature. Alongside these more traditional shades of dark red, pottu colors
nowadays vary from parrot green to magenta, bright yellow, or colors that
match the sari or the outfit one is wearing.2
It took me many years to realize that these two kinds of ritual designs are
more intimately connected than I would have first imagined. They form a kind
of kinship with each other; they mirror and echo each other as parallel ritual
expressions of complicated and nuanced concepts, such as auspiciousness and
inauspiciousness, purity and pollution (Carmen and Marglin 1985; Marglin
1985). These keywords are central to the understanding of the inherent ambiv-
alence of expressive ritual power in Tamil women (Williams 1983).
As Mangalapatti from the village of Rengalachetty once said to me, ‘‘The
kolam is the pottu of the house; the pottu is the kolam of the house. Do you
understand how that is so?’’ Lakoff and Johnson have observed, ‘‘The essence
of metaphor is understanding and experiencing one kind of thing in terms of
another’’ (1980: 5). The visual presence of the pottu, kolam, and other forms of
ritual traces of worship indicates that a space has been initiated for ritual pur-
poses. A comparison of the kolam with the pottu can deepen our understanding
of the way visual signs function metaphorically and spatially in Tamilnadu.
This chapter is an exploration of how ritual marks on the thresholds of
homes and bodies mean what they mean. In my many conversations with Tamil
threshold designs, forehead dots, menstruation rituals 87
women over the years, one of the leitmotifs was that the kolam on the front stoop
‘‘indicates auspiciousness and prevents ritual pollution.’’ This binary encoding of
meaning, presence indicating one state of being and absence another, is at the
core of these domestic rituals. Their presence and absence enable kolams to carry
meaning. The pottu on the forehead, when present, similarly communicates a
state of auspiciousness. The absence of a pottu, on the other hand, may reflect a
state of being which is soaked through with ritual pollution. I see these ritual
marks now as a form of communication, a way of letting what is or what is not
happening on the inside of the house or the body be visible to the outside world,
akin to a silent but very visible announcement. Sometimes these binaries be-
come indicative of the deeply ambivalent valences of women’s ritual power,
sometimes alluding to a woman’s responsibility for death, primarily the death of
her husband. These ritual markers act as a binder of sorts, bringing together a
multitude of strands of local vernacular thinking of what it means to be a Tamil
house, a Tamil woman householder, or a Tamil woman (Hart 1973, 1975; Daniels
1984; Wadley 1980).
This chapter explores two Hindu women’s rituals that mark time and
space in Tamilnadu. They are, in one case, a household ritual and, in the other,
a ritual performed on the body. I argue here that these two public rituals em-
body Tamil Hindu women’s interior states of being auspicious householders
and, simultaneously, function as active vectors for sending forth positive in-
tentionalities or activated blessings for the day, on the surface of both the home
and the woman’s body. This chapter expands the realm of the ‘‘domestic’’ to in-
clude the householder’s physical body and the community beyond the thresh-
old of the domicile by mapping correlations and correspondences between
domicile and householder. The kolam and the dot become vehicles of com-
munication beyond the threshold and have implications across the interiority
and exteriority of thresholds by articulating in ritual time and space moments
of emotional and physical transformation. This chapter knits together theories
of auspiciousness/inauspiciousness and ritual purity/pollution with the ap-
pearance and the disappearance of the kolam and pottu.
How do the kolam and pottu define and articulate these differing axes of
value, that is, auspiciousness/inauspiciousness and purity/pollution (Carmen
and Marglin 1985; Marglin 1985)? The kolam functions as a key semiotic indi-
cator on the Tamil cultural landscape, not only mapping the individual’s passage
from degrees of ritual purity and auspiciousness to those of ritual pollution, but
also articulating the nonlinearly punctuated contours of a cosmology soaked
through with auspiciousness and ritual pollution in ritual time and space; that is,
sometimes one object may be auspicious, and another next to it inauspicious,
and so on.3 The kolam is linked to these categories in three important ways. First,
its presence or absence prescribes social relationships, determining the bound-
aries of appropriate interactions between auspicious and inauspicious people,
places, and objects. Second, its location marks several types of thresholds, both
spatial and temporal, indicating the boundaries between auspicious and inaus-
picious worlds. And finally, discourse related to the kolam conjoins the notions of
mangala (auspiciousness) and amangala (inauspiciousness), mati (ritual purity),
and turam (particularly menstrual pollution of ritual distance) or teetu (gener-
alized ritual pollution).4 In this chapter, I focus on the active, fluid, and porous
folk notions of auspiciousness and ritual pollution as they shape and constrain
the domestic interactions of Tamil women.
Presence and Absence of the Kolam on the Front Stoop
The presence or absence of the kolam on the front stoop is equivalent to

announcing that a household is open or closed to the world, a cultural category
that sends off a capacious sense of hospitableness or hospitality, specifically,
the willingness and ability to feed a stranger. By looking immediately at the
front stoop, if I see a kolam, then I can imagine that a woman is functioning at
a high level of open hospitality. If I were a wandering sadhu (an ochre-dressed
holy man or woman), a minstrel, or a beggar, I could perhaps hope for a meal
or some uncooked or cooked rice in my begging bowl. I could hope for and
imagine a possible site of hospitality, an ‘‘open’’ household—healthy, func-
tioning, perhaps holding a surplus of food, which can overflow outward into
the community. Telegraphing receptivity and hospitality, the presence of the
kolam indicates the ability of the household to serve as a welcoming, ‘‘feeding’’
host to strangers, visitors, and guests.
The presence of the kolam also indicates the sense that women’s power to
create a sphere of ‘‘positive intentionalities’’ moves in two directions: outward,
to the world beyond the threshold, and inward, to the household. The aus-
picious power travels from women’s hands through the kolam and upward
into the bodies of those passing through its energy field, as they step over,
around, or through the kolam. This capacity is especially significant during
moments of women’s ritual life-cycle ceremonies, articulating how and why
the household may be seen as open. At particularly important life-cycle cere-
monies, an overflow of generosity is imagined, required, and enacted. It is
through the capacity of generosity that auspiciousness is generated.
Yet another way that the kolam communicates its presence is to infer the
‘‘good, auspicious’’ news from the inside of a woman’s body, layered on the
outside ‘‘skin’’ of the house itself. As one Tamil Brahmin woman, Padma, from
the village of Aiyappur said:
Listen, Vijaya, in the time before phones and telegraphs, the

kolam was the way we found out what happened in the house dur-
ing the night. If there was a huge, vishesham [special] kolam, and
we knew that there was no wedding planned that day, then,
we may guess that the girl in the house has come of age or a
baby has been born.
It is clear from this that these huge kolams, reflecting either the birth of a child
or the first menstruation of a girl—events that are both auspicious and ritually
polluting––are, then, celebrated and announced publicly.5
At the most visible level, the presence of a kolam on the threshold both
signifies and invites auspiciousness while keeping ritual pollution at bay. The
kolam, though announcing auspicious yet polluting events, also has the power
simultaneously to alleviate the pollution caused by the event. It indicates that
the woman of the house is ‘‘in’’ and that the household is successfully main-
taining its everyday rhythm. In the words of an eighty-three-year-old Tamil
Brahmin Iyengar woman, Padma, in Srivilliputtur: ‘‘If someone dies, then, one
should not put the kolam for one year. Otherwise, you have to put the kolam
every day. If you do not put the kolam, it is akkiyanam [spiritual ignorance].
That means someone will die in your house. You should not not do the kolam.
You have to put the kolam.’’ So, if the presence of the kolam is a generous
invitation for the outside world to come in, its absence indicates a closed
household––a site of suffering, death, or other ritually polluted or polluting
states of being. Usually during menstruation, after childbirth, if she is ill, or
when there has been a death in the family, a woman is not expected to make
the kolam. All of these ritual practices and their rules are not universally
applied but may shift and change, depending on the caste, class, and degree of
modernity adapted by the household. In some communities, childbirth is not
necessarily announced by a kolam, but in others it is. And so on.
In Tamil culture, suffering is accepted as an inevitable part of everyday life,
and the kolam helps to structure the experience of suffering. As it was put
succinctly by one Tamil woman, whose family had migrated generations ago
to Kerala and still amazingly drew the kolam every day, ‘‘The kolam is done to
prevent future suffering, to be able to manage our current suffering, so, we
know when it is happening around us, to reduce tension, and a kind of med-
itation.’’ The lack of a kolam announces a household’s suffering to the commu-
nity as a whole. The continued absence of the kolam signifies the gradual
ebbing of suffering in the family; the reappearance marks visually the end of
the period of suffering and ritual pollution. Veena Das’s understanding of ex-
ternally oriented suffering, that is, suffering that originates from the outside, is
relevant here:
[The external orientation] holds suffering to be accidental . . .
holds existence to be blameworthy but points to the capriciousness
of the gods, the inexplicability of the world, and the contingency of
life as the reasons for suffering. It does not make the sufferer inter-
nalize her suffering, nor does it posit a meaningful world or a just
god or a comprehensive scientific discourse within which suffering
can be made comprehensible. Dare one say that it gives irresponsi-
bility a positive sense? (1995: 139–140)
As part of the larger context of the experience of suffering in Tamilnadu,

the presence of the kolam, too, affirms a normal, well-functioning existence
and affirms the suffering person by its absence. In a world where both happi-
ness and suffering are equally inexplicable, the ephemeral kolam represents
the capriciousness of the gods and the contingencies of life, and it helps to
attract neighbors and strangers to assist those in need. Besides indicating death,
a lack of a kolam on the stoop could indicate a menstruating woman under-
going a period of incapacity and, therefore, the possibility of a woman expe-
riencing pain and suffering during her menstruation.
In conclusion, the presence or absence of a kolam on a threshold visually
cues people about what to expect and how to behave; for example, the absence
of the kolam may catalyze looking in on the household and bringing gifts of
food, clothing, or comfort. By its absence, the kolam prescribes a supportive
response to suffering, engendering a responsibility to enter the ritually polluted
household in an empathetic emotional state. In this way, the kolam alleviates
suffering by structuring the community’s response in situations of grief, pain,
and loss. The Hindu concepts of auspiciousness and ritual pollution and their
reflection within the kolam on the stoop make possible the recognition and
celebration of joy and the detection of and support for suffering. It might be
said that the visual sign of the kolam affirms life by its presence and affirms
suffering by its absence. As Clifford Geertz observes, ‘‘[T]he problem of suf-
fering is, paradoxically, not how to avoid suffering, but how to suffer, how to
make of physical pain, personal loss, physical defeat, or the helpless contem-
plation of others’ agony something bearable, supportable—something, as we
say sufferable’’ (1973: 104). Therefore, one can say that the kolam may be a
sign of a community’s perceptive control over chaos, pollution, and suffering
(Hart 1973, 1975). By providing an orientation to the emotional state of the
household, the kolam imposes a pattern on the community landscape that re-
veals the contours of satisfaction, happiness, and suffering in each individual
household.
The Pottu, or Dot on the Forehead
Commonly consisting of one or two red dots and occasionally a black line, the
pottu is actually a multitude of signs that telegraph relevant cultural infor-
mation to informed viewers. The pottu varies in frequency, style, and color,
according to caste, religion, class, and marital status. For example, the pottu
reveals whether a woman is old or young, maritally available or unavailable.
The pottu’s communications function to guide people’s behavior in relation to
a woman, letting them know whether they can tease her as a sister or must bow
down with respect.
When the Dot on the Forehead Is Present

If a girl wears a red pottu, this signifies that she is past her first menstruation
and is therefore eligible for betrothal and eventual marriage. The red color of
the pottu has both a literal and metaphorical significance. In Tamil culture,
red is usually worn after a girl has passed the stage of ‘‘blood magic,’’ or
puberty (Buckley and Gottlieb 1988). One of the key colors of the sari worn
during the wedding ceremony, red also symbolizes the potential and actual
power of sexuality. Among certain subcastes, if a woman is wearing two red
pottus—one in the center of her forehead and the other at her hairline—this
signifies that she is married. Stella Kramrisch has observed a similar signif-
icance for the color red in the kolam, too, pointing out that the threshold zone
is ‘‘protected by the design traced on the floor in an unbroken line forming
loops and enclosures, each marked by a dot in its center, the dot being a
symbol of the seed, the source of life. Sometimes the threshold is dressed in
red dots, similar to the red dot commonly seen on an Indian woman’s fore-
head. The dot is a symbol of blood, the source of life’’ (1985: 105; 1983).
The woman of the household is well aware that her power stems from her
ability to create auspiciousness, and she wears indicators to acknowledge this
fact, such as the pottu (or bindi), tali (wedding necklace), toe rings, or henna
on her hands and feet (Reynolds 1980). Each morning, after her ‘‘purification’’
bath and before she puts the new pottu on her forehead, a woman may offer a
prayer for the longevity of her present or future husband. Indeed, some
Hindu women believe that placing the pottu on their forehead has a positive
impact on the length of life of their husband, present or future. A woman is
considered to have the ability to keep her husband alive, and therein lies some of
her power as the bearer and container of auspiciousness. After her bath, she is
also considered to be in a nonpolluting state, capable of personal and household
generosity to the community. Red dots may be placed on the foreheads of both
men and women to signal a recent visit to a household or temple shrine. Red
powder may also be smudged onto new clothes in gratitude to the goddess
Lakshmi, for her generosity. It is important, however, to note that these signi-
fiers have less importance in contemporary times as these kinds of markings
have shifted from an emphasis on ritual meanings to an emphasis on beauty.
Therefore, these signifiers have come, increasingly, to transcend caste, class,
and religious affiliation.
When the Dot on the Forehead Is Absent

A black line on the forehead generally indicates a prepubescent girl, who is
still considered a child. This status implies that she is ‘‘tease-able’’ and able to
receive gifts from anyone in the community with no special meaning at-
tached. If an adult Hindu woman is not wearing a pottu, it is likely that she is
a widow. However, if other signs do not corroborate this (such as wearing a
beige or white sari and no jewelry), then the absence of a pottu could indicate
that a woman is menstruating, in a state of mourning, or perhaps just too
busy working to put on a pottu. Christian or Muslim women do not usually
wear the pottu, although some do so as a fashion statement.
Two Memories of Absence

I have a personal interest in the concept of ritual pollution, which stems from
a lifelong awareness of its multifaceted nature within Indian culture. I am es-
pecially intrigued by practices associated with menstrual pollution. To some,
women’s isolation during the state of menstrual ritual pollution may seem like
an inconvenience or insult, a perspective with which I partially agree. But I
would suggest that this isolation can also be simultaneously and paradoxically a
period of welcome rest and conviviality. The positive or ambivalent aspects of
practices associated with ritual pollution are rarely reflected in fieldwork accounts
of childbirth, menstruation, or death. I will share two personal stories that hint
at the underlying emotional paradoxes of rituals informed by this concept.
One of my earliest encounters with the idea of ritual pollution took place
in our ancestral village, Rettakudi, when I was nine years old in 1970. Our
family had just returned to India from a four-year hiatus in America and had
quickly resettled in Rettakudi for the summer. I adapted easily to the Tamil
language and culture, which were integral to our home life in both India and
America. A few weeks after we arrived, however, a strange event happened to
me: I lost my mother. Whenever I asked my father or grandparents where
she was, they would look away as though I had asked a very embarrassing
question. Curious and sad, I thought my elders were hiding something from
me. I searched everywhere but couldn’t find her, so I concluded that my mother
must have either run away or died.
The next day, I decided to follow my grandmother all around the house, at
a discreet distance. After we had eaten in the afternoon, and all the family
members were moving to their respective corners for napping, I noticed my
grandmother putting together a simple meal of rice and vegetables, wrapping
it carefully in banana leaves. She walked out the back of the house, through
the cowshed that circled the house, and on toward the front. I followed her,
wondering whom she was on her way to feed so surreptitiously. Standing
behind the door of the cowshed, I watched her walk across the dusty wayside
path and place the small bundle before the threshold of an adobe hut that I
had never noticed before. Her gestures were distant, as if she were making an
offering to a goddess or an Untouchable.
To my surprise, she called out my mother’s name and announced, ‘‘The
food is here! Is everything going well?’’ My mother’s voice, restful and con-
templative, answered softly with a faint ring, as if from a distance, ‘‘I’m fine,
amma, thanks. How is everything out there?’’ My grandmother replied in an
amused tone, ‘‘Your eldest daughter, Vijaya, is giving us trouble. She keeps
thinking you have disappeared. She doesn’t at all understand what is going
on. Don’t you teach her anything useful in America?’’ I was deliriously happy
to hear my mother’s voice at last. ‘‘She is alive after all!’’ I thought. ‘‘I hear her
voice, and now I must see her. Or maybe she is just a ghost from the dead.’’
As soon as my grandmother returned to the main house, I saw my
mother’s bangled hand reach tentatively across the threshold to pick up her
food. The shadow of her form appeared and disappeared so quickly that I
wondered if I had only imagined it. Leaping across the threshold of the hut, I
felt as if I were on a dangerous mission to rescue my mother and return her to
the main house. I pushed open the heavy wooden door into the very dark room
and was about to dash into her lap, crying out with joy, laughing. But, barely
seeing my mother’s form lying in a shadow-laden, dusty corner through the
light of the kerosene lamp, I hesitated, disturbed by what I saw. My mother was
frowning and seemed angry with me, and I noticed she had no pottu on her
forehead. She looked different, with her hair unbound and in disarray, and
relaxed, away from the prying eyes of a dense household community.
On seeing me, she said sharply, ‘‘Shoo! Get out, you silly monkey child!
You shouldn’t be here.’’
‘‘But, amma, I thought you had died!’’ I blurted out, jumping happily into
her lap. ‘‘Why are you here? Is this where you have been all this time? Nobody
would tell me where you were. I am so happy you are still alive! Can I take you
back to the main house now?’’
‘‘No,’’ she replied:
Don’t touch me. I hope no one saw you. I am turam [distant]; no

one can touch me now. Listen: don’t tell your grandmother you came
here. She would be very upset if she knew. But it is probably all right;
you’re still a child so it shouldn’t affect you. But don’t tell any-
one you have seen me. I am turam. . . . later I will tell you what that
means. I promise to be back tomorrow. Just think that your mother
didn’t have to work for three days, and she’s having some quiet
resting time. So, run off my foolish little monkey—and be sure to
wash your hands and body wherever it touched mine. But don’t go in
through the front door of the house. Enter the backyard through the
cowshed at the side of the house, and wash at the water pump. Then
wait awhile before you touch your grandmother. She is madi [pure],
you know.
Puzzled and disturbed by her behavior and her stern tone of voice, un-
usual in my mother’s normally indulgent self, I rose out of my mother’s lap,
walked backward out of the mud hut, and washed as she had instructed. Then
I ran into the main house and immediately demanded of my grandmother,
‘‘What is my amma doing in that other house? Why have you put her there?’’
She replied with the mysterious word turam again. She added, ‘‘For three
days your amma will be in the other house, and then she will come out and
join us on the fourth day, after she’s washed her hair. Then she can eat with
us again.’’
From that day onward, I began to notice the sudden disappearance and
reappearance of other women to and from the small huts scattered throughout
the village, set apart from the houses. The neighboring women would bring
food to these huts for each other, and a few days later the mothers, wives, and
daughters would return with glistening wet hair, once again considered ‘‘touch-
able.’’ And then they would go back to work hard in the domestic chores of an
extended household.
As I spent my childhood traveling between India and North America, two

overlapping, yet distinct, codes of ethics, attitudes, and world views trailed me
from one dislocation to another. I was in one place, then the other, each place
continuing to mark me even when I was not there. Then, when I was a teen-
ager, we settled down in suburban Maryland. Whenever American friends
crossed the threshold to our home, their first comments were invariably about
the kolams that my mother created on the threshold each morning. ‘‘What is
that?’’ ‘‘Can I step on it?’’ they would always ask. Of the pottu I wore most days
on my forehead, my friends would ask ‘‘whether it was blood,’’ with an echo of
disgust and disbelief. Or, if I were discovered to be indeed wearing a red dot
with intentionality, the questions would inevitably come: ‘‘Why was I wearing
it? Did I have to put it on every single day? What does it mean?’’ and so forth. I
struggled as best as I could to answer their questions. I kept on wearing it, as
I could not imagine doing otherwise.
If one of the women in our house was veetil illai, or ‘‘not in the house’’
(menstruating), or if we all were ‘‘not in the house,’’ my astonished and puz-
zled friends would witness my father walking briskly back and forth between
the kitchen and the living room serving all of us tea, sweets, and the cooked
food he had prepared. He would discreetly set the food down at a suitable
distance from himself; we would wait until he moved away, and only then
would we proceed to eat. We had become turam.
Whenever I think of menstrual time, an amusing image comes to mind: my

father backing away from us when we stretched playfully our hands to him over
the threshold of our ritually polluted spaces and encroached upon his ritually
pure space. He would retreat, beseeching us to stop, saying with fear and
annoyance, ‘‘Do not touch me, do not come closer!’’ as if even the intersection of
the ritually pure and impure would cause him enormous pain or destroy him in
some way. The look of terror in his eyes was not a look we saw often. As a
teenager, it made me feel powerful to think that I could make my father fear me
because I was a woman who had become turam. That ‘‘touch’’ was itself dan-
gerous somehow intrigued me; yet I also questioned daily the very basis of the
distinction between pure and impure. So what if women were menstruating;
why couldn’t they still be priests in the temple? What had menstruation to do
with purity? I wondered to myself and argued with others, including our local
Hindu temple priests, often. Why was menstruation celebrated and honored the
very first time it occurred, then hidden underground, like a golden, auspicious
spring which was named temporarily ‘‘untouchable’’?
Since menstruating women are not allowed near spaces where gods and
goddesses are housed (in temples and kitchens), my menstruating sisters
and I could not touch anything in the kitchen and consequently could not do
much housework. I remember the times spent ‘‘not in the house’’ as restful,
unruly, indulgent, and playful—periods when we were permitted to be lazy
and to reflect on our lives in America. All three of us were treated particularly
well; we were served tea, food, and sweets and our wishes were fulfilled. When
the fourth day arrived, we would take our head baths and become ritually pure
again, losing our special space of quiet and rest.
Reflecting on this experience, it is not difficult to see that this kind of
bodily experience of ritual pollution has its mix of positive and negative va-
lences in terms of women’s sacredness and ritual power. The emotional and
cultural ambiguity and paradox of women’s ritual domestic power plays itself
out through these kinds of lived experiential and bodily narratives, a subtle
weaving of power and powerlessness, a valorizing of female auspiciousness,
on the borderlines of temporary ‘‘untouchability’’ and ‘‘touchability,’’ in the
Indias and Americas of my own and other Indian women’s pasts, resurrected
only in memories, a series of nodes and experiences, in tandem and counter-
point to sealed theories and understandings.
In conclusion, we can draw the following analogy: the pottu is to the body
as the kolam is to the house. So, the house:kolam::body:pottu. Just as the kolam
marks the threshold between the interior of the household and the community
outside, the pottu on the forehead marks the threshold between the internal
body/soul and the external world. The same visual metaphor can be observed
in the red henna applied to the hands and feet as ritual markers of ceremonial
time and space, as in marriage ceremonies. The significance of the literal and
metaphorical edges of the body is elaborated in folktales, proverbs, notions of

hospitality, and the many stories about the inauspicious consequences if a
woman does not put a kolam on her threshold or a pottu on her forehead.
Marking Thresholds of Space and Time
In addition to its presence and absence, another way the kolam articulates
concepts of auspiciousness/inauspiciousness and purity/pollution is through
its location in space and time. The thresholds on which the kolam appears are
both spatial and temporal, indicating the boundaries between auspicious and
inauspicious realms and periods.
Spatial Thresholds
The spatial threshold is a powerful metaphor in Indian secular and ceremo-
nial life, a charged location between ritually pure and impure or between aus-
picious and inauspicious places. The kolam is created at three types of spatial
thresholds: (1) the household shrine in the kitchen, (2) the entrance to the
main house, and (3) the entrance to village temples.
The kolam at the household shrine in the kitchen marks the women’s do-
main, creating a threshold between the kitchen activities and the separate and
sacred space of the divinities that inhabit the shrine. The gods face the wor-
shipper from the east (the most highly valued sacred direction), and the wor-
shipper prays toward the glancing gods. In relation to the rest of the house, the
kitchen is considered to be the abode of the gods and goddesses in the secular
world of the householder. The place where the food is prepared is the center
where all the family’s ritual activities are commenced, maintained, and com-
pleted. In fact, the entire women’s cooking area is the literal and metaphorical
hearth—the spiritual and psychological core of the household and the most
valued site for the production of auspiciousness. Architecturally, the kitchen is
also the most protected part of the household. One of the few fully walled and
bounded rooms, it is the most distant from the outside world (Blier 1995).
The second type of kolam is created on and beyond the threshold of the main
body of the house, facing the village path and dividing inside and outside, known
and unknown, safe and unsafe worlds. As we have seen, the key site for a kolam
is the front entrance to the house, distinguishing inside from outside, household
from commons, and private from public. Kolams on household shrines and
interior thresholds mark the movement of family members from the inner
sanctum through all the interior doorways to the front door—the place where
the house meets the exterior world. Proceeding down the village street, one can
observe that each household in an auspicious state of being is marked with a
kolam. Between sacred and profane, auspicious and inauspicious, controllable
and uncontrollable, the threshold guards the house from the chaos of the outside
world. The kolam on the front of the house demarcates the ritually polluted
commons from the private, domesticated household space.
Shulman has eloquently referred to the kolam as a protective, invisible,
three-dimensional screen in front of the house, which is seen through its
visible, two-dimensional form. He points to the threshold area as the ‘‘point at
which it [the kolam] emerges into form—a complex form at that, carefully
planned and executed, a reflection of some inner labyrinth externalized here
at the boundary, the line dividing the inner and the outer, the pure from the
chaotic’’ (1985: 3). This type of kolam may also be seen as a visual metaphor for
the division of the commons, the shared public civil space, from the controlled
and contained space of the home and, therefore, metonymically, the woman
of the house.
The third type of kolam is made at the village temple, where it carefully
delineates each temple threshold from the preceding one and from the world
outside. Approaching a South Indian temple, one observes that a kolam marks
each threshold for the advancing worshipper, from the outer entrance to the
innermost ‘‘womb chamber’’ of the divinities. This sequence of kolams marks
the passage of worshippers as they travel to the interior shrine to visit the gods
and goddesses and to receive darshan, or blessing (Eck 1985). At the edge of
the village community, the temple may be conceptualized as a giant, three-
dimensional kolam marking the threshold where the village ends and the
‘‘outside’’ world begins. Here again, the kolam maps the journey from a ritually
polluted space to a ritually pure space.
The designs of kolams drawn in the kitchens, doorways, and temples re-
flect symbolic tirthas (crossings)—that is, spaces to be crossed with a con-
sciousness of the sacred (Eck 1981). When a space has been sacralized, this
affects people’s behavior in ways that are characterized by ritual hospitality
and auspiciousness. Kolams at the front entrance to the house are made as an
offering and blessing, to be stepped on and erased slowly under passing feet.
In the words of one Tamil woman, ‘‘Stepping on the kolam is like stepping
into the Ganges River.’’
Another woman said emphatically, ‘‘Stepping on the kolam is akin to
taking a bath in the sacred Ganges River, an act that purifies body and spirit.’’
But, it is important to point out that not everyone steps on the kolam; many
walk around it, so they do not step directly on it and smudge it, revealing its
increasing presence as more and more a symbol of beauty, rather than its
many ritual aspects. On the other hand, a kolam should not be stepped on if
it is located at the center of a sacred space, such as a household shrine or a
temple before a deity, because it is believed to be highly charged with divine
energy. Kolams made at household shrines function as porous boundaries
between earthly and divine realms, while those made at temples are part of a
continuum of sacrality.
Temporal Thresholds
It is crucial that the kolam be created before the sun rises every day, when
darkness is transformed into light, at a time of betwixt and between. Dawn
exercises a particular sort of ritual imagination. The time between four and
five-thirty in the morning is called brahma-murti or brahma-muhurtam, ‘‘the
time of God’s face,’’ when the deities turn their faces toward humans. The
kolam is a visual, aesthetic signal designed to attract the gaze of the divinities.
Thus the kolam marks the temporal threshold between night and day.
Perhaps the kolam could be considered a parallel version of the male yogic
positions of the surya asana, ‘‘the worship of the sun.’’ In fact, the ritual prac-
tice of drawing a symbol on the ground to worship the sun is mentioned in the
Rig Veda. According to the art historian Stella Kramrisch:
The most ancient Sanskrit treatise on Indian painting pre-

scribes the worship of the sun god through an eight-petaled
lotus flower drawn on the ground. Several Puranas speak of
the art of drawing the sun on the ground and that the sun was
worshipped in a circle in early days. However, this practice was
not sanctioned by the Vedas; it belonged to those outside the
Vedic pale. The drawing of a magic diagram on the floor, how-
ever, became essential in building a Hindu temple. (1983:
105–107)
Madan, a social anthropologist, points out one of the congruencies be-

tween auspiciousness and temporal location:
There are many auspicious and inauspicious moments in

one’s life, just as there are in a day. The most auspicious mo-
ment of the day is the rising of the sun. It fills the earth, the sky,
and the heavens with light and brings with it the promise of good
works and wisdom for men. . . . Sunrise manifests the glory of
God, enlivens our intelligence, and purifies the whole earth.
(1987: 48)
In addition to demarcating the boundary between night and day, the kolam
serves another important temporal function. The month of the winter solstice
from mid-December to mid-January (Markali) is the time of year when the
sun travels at its lowest point in the sky. At this time, kolam making marks
both the sun’s nadir and the zenith of the Tamil agricultural cycle. Markali
is considered to be the month that spans one day in the life of a divinity.
In other words, if one year in human experience is equivalent to one day in
the community of gods and goddesses, Markali is the beginning of the new
divine day.
At the end of the Markali month and the beginning of the Thai month is a
threshold period that is celebrated with the Pongal festival, the most popular
festival across castes and classes. This is a highly auspicious time for the
community, signifying the abundance of the fields during the harvest. In the
village Krishnagudi, Chellamma, a young dalit (‘‘Untouchable’’) woman who
lived in a lovely home with beautiful squash plants gracing the side and roof
of the house, expressed the sense of joyful anticipation at this time of year:
The ammans [goddesses] are coming. . . . We feel this is when

everything good is coming. The happiness . . . is coming. Food is
coming. Children are coming. Wealth is coming. Beauty is where the
divine comes. Rice is turned in. The fields are brimming over with
the harvests. It is when we are at the wealthiest time of our year and
most hopeful of financial security.
Because of its beauty, the kolam is the focalized site where the divinities are
hosted by the woman of the household.
Kolam making reaches its annual peak during Markali and its monthly
peak during the Pongal festival, when the best and most elaborate kolams are
created. Drawing the kolam in rice flour becomes the event of the day, creating
in the end a sense of fine white lace cloth draping over every surface imagin-
able. The month of Markali is an auspicious time for honoring divinities but
an inauspicious time for human celebrations. As one elderly Tamil woman
put it:
The reason we create such elaborate kolams during the month

of Markali is because the threshold, the doorway between heaven
and earth, is the most open during this time. It is the time to com-
municate with the gods and goddesses. That is why we go on pil-
grimage during that time. It is also a great time to die, because
you automatically go to heaven. It is a bit like dying on the banks
of the Ganga.
On the other hand, Markali is considered to be a highly inappropriate time for

getting married because one’s energies should be devoted to spiritual rather
than material matters.
The kolam also signals the temporal rhythms of families, communities,
castes, regions, and religions. Certain castes and religious traditions may have
family or lineage commitments to particular astrological calendars, and the elab-
oration and density of the kolam patterns indicate special days of celebration. For
example, Christian families might make substantially larger and denser kolams
on Christmas Day. Or, Hindu orthodox castes may mark the period between
mid-July and mid-August with dense kolams, and other families or castes may
mark the festivals of individual saints or animal deities such as snakes.
Mapping Auspiciousness and Ritual Pollution
While the degree of ritual pollution can be inversely related to the degree of
auspiciousness, the demarcations between ritually polluted and auspicious
states are complex. As mentioned earlier, menstruation and childbirth are
auspicious, life-affirming conditions, though paradoxically they are situated
in a ritually polluted state of being (Marglin 1985; Das 1995; Madan 1987;
Carmen and Marglin 1985). Additionally, death is usually considered to be
inauspicious and ritually polluting—although even this state is ambiguous
because it is linked to the reproduction of the family lineage. For example, days
of worship that commemorate the dates of ancestors’ deaths are considered to
be ritually polluted and auspicious, since these occasions focus on both the loss
of family members and the continuation of blood kinship ties. In most cases,
the kolam reflects the conjoining of states of being that are auspicious and
‘‘pure,’’ that is ritually nonpolluting. That is why the kolam is not made when
women are menstruating or when there is a death in the family, to show that
the household is presently not hospitable. There are two exceptions, however.
First, when a girl attains her first menstruation, a huge feast is made, a giant
kolam is created, food is served, and so forth. And second, when a child is born,
then, too, a huge kolam is made, though in this case food is brought in by neigh-
bors, because childbirth is considered to be a state of incapacity and closed
hospitality.6
The Role of Domesticity in Hindu Everyday Life
One of the most critical ways to understand the role that ritual pollution plays
in Tamil Hindu women’s social spaces is to understand the role and ideology
of the householder in Hindu traditional life. Female householders may in
some circumstances hold an equivalent, complementary, ritual importance to
male householders in Hindu everyday life. Especially when we look at them
from the perspective of women’s narratives of their own rituals, these women’s
rituals become charged with an intense celebration of women’s sense of their
own female importance and vitality. It is also important to note the ideology of
the householder and its hold on women’s ritual lives.
The Ideology of the Householder

The state of the household, the pragmatics of everyday life, and the concern
for material existence are all part of the making of the kolam, the affixing of the
pottu, and the performance of many of the rituals that signify the life of the
householder. The importance of the householder has been eclipsed by the as-
cetic in scholarly treatises. In the West, there has long been a fascination with the
sadhu, the wandering monk who renounces the world in search of pure truth
and understanding. By contrast, the householder, a worldly character, is a much
more realistic and popular expression of Hindu personal conduct and moral life.
Madan’s detailed and wonderful ethnography, a landmark study of non-
renunciation in the community of Kashmiri Brahmin men, may apply here. He
notes:
The figure in the centre of the stage is a rather homely character,
namely, the householder [grhastha]. If not exactly cast in a heroic
mould, he is not the ‘‘phantom-like’’ man either that Dumont
(1966: 48) considers him to be. It is the ideal of his life to ‘‘live
in the world’’ but to do so in the light of the renouncer’s philosophy
(see Dumont 1970: 12, 41, et passim). Translated into the house-
holder’s idiom, renunciation becomes the twin ideals of self-
possession and detachment in the midst of worldly involvements,
which are not considered by him evil in themselves. What he seeks to
resist is being enslaved by such involvements. He hopes to medi-
ate between total indulgence and total renunciation. It is, indeed, all a
matter of relations. (1987: 2–3)
In the idiom of the nonrenounced realm of the householder, the kolam
ritual can be seen as an ‘‘affirmation of a disciplined this-worldly life as the
good life’’ (Madan 1987: 3). It expresses everyday concerns, hopes, and de-
sires, such as good health and prosperity within the family. Many women with
whom I spoke articulated the view that the kolam brings not only auspicious-
ness and goodness, but also the orientation of disciplining earthly desires. The
householder must always be aware of binding her or his desires, constantly
incorporating ascetic values into daily life. This relates to Madan’s interpre-
tation of the religious ideology of the householder as one who ‘‘acknowledges
the sovereignty of good: the desired must be brought under the regime of, and
encompassed by, the preferred’’ (3). The continuity of domestic life is at the
heart of the notion of auspiciousness.
Lakshmi is the goddess of auspiciousness, or mangala, which includes good
luck, wealth, wakefulness, alertness, quickness, and abundance. When Lakshmi
is invited in by the woman of the household, a portion of the divine auspicious-
ness is transferred to the earthbound realm of the woman householder. Indeed,
the woman of the household is often referred to as the Lakshmi of the house.
Like the goddess, the woman has the power to attract wealth and prosperity
into the household and to prevent poverty from crossing the threshold. Since
the householder is seen as the creator of mangala, when domestic life is inter-
rupted, the flow of auspiciousness also comes to a halt. This interruption in the
flow of auspiciousness from the goddess Lakshmi to the woman of the house-
hold, and its further effect, to the household itself, is usually attributed to her
generally unbeloved sister, the goddess Mudevi, or Jyestha. Mudevi is the god-
dess of sleep, restfulness, laziness, ill luck, poverty, and scarcity (see Leslie 1991;
Nagarajan 1993).
Within the domain of the kolam, domesticity reigns in the model of the
woman householder as the source of the flow of auspiciousness throughout
the community. Creating the kolam is an active way for Tamil women to ar-
ticulate their desires daily. It is a form of prayer in which the women of the
household directly communicate their intentions to Lakshmi, the goddess of
prosperity, beauty, and good fortune. Madan discusses the nature of this type
of intentionality:
The distinction between the state of auspiciousness and the crea-
tive agent . . . is most important as is the relation between the
two. . . . The point to note about these usages and similar others
is that it is not the person himself or herself who is auspicious
but rather his or her intentions, actions, or even merely the pres-
ence (and witnessing the same), which are so and are, therefore,
expected to have happy consequences. The ultimate source of aus-
piciousness is, of course, the divinity. (Madan 1987: 53–54)
Therefore, the notion of auspiciousness is itself bound with positive inten-
tions, actions that are themselves expected to be seething with goodness, and
is likely to spread all around (Nagarajan 2000: 565–566).
The Moral and Good Life
In conclusion, the kolam then is a sign that proclaims a Tamil woman

householder’s moral status, separating the woman of the house from those
who have chosen not to take the householder path. It helps to give us a clue
perhaps to the puzzle of why a Hindu woman does not need to do any puri-
fying rituals before she makes the kolam, as she would in most other rituals.
The kolam celebrates the female body in its potential or actualized sexual state,
rising directly from the marital—or, perhaps, premarital—bed, and it an-
nounces this bodily state directly through its material form. The lack of need to
purify the body to make the auspicious kolam makes sense, for it is not purity
that the kolam embodies, but rather the auspicious, sexually satisfied or po-
tentially satisfied woman who is within the moral and cultural social bounds of
householder life. As one municipal sweeper woman in the medieval town of
Thanjavur shouted out to me as she swept up the dirt of the day before, ‘‘Hey, I
hear you are asking everyone why we are doing the kolam. I will tell you why. It
shows you slept in your own house that night, that you woke up there.’’7 She
grinned and laughed, with great merriment. And all of the women sitting
around me on a porch also laughed. When I looked up to find her, to continue
our conversation once I had gathered my work supplies, she had swirled out of
my line of sight. And I could not find her again, though her voice rang in my
head for years to come. What did she mean by that statement?8
The ritual principle of auspiciousness is embedded in cultural notions of
morality, value, and the meaning of life. Madan observes that the notion of
auspiciousness is bound with the life worth living: ‘‘For the common Pandit, the
life of the man-in-the-world—epitomized in the role of the householder—
though arduous, is the moral and good life. It is a life worth living’’ (1987: 47).
Tamil female householders equally substantiate their desires for a ‘‘moral and
good life’’ through the daily practices of banishing laziness and attracting status,
wealth, material possessions, health, children, good fortune, and other forms of
auspiciousness. The kolam and its ritual parallel, the pottu, or red dot, according
to many Tamil women with whom I spoke, are visual and aesthetic statements
that ‘‘we are living in the world and experiencing life fully; we want to be free
from poverty and ritual pollution and have a life worth living.’’ The assumption
is that weaving thoughts and words into designs has the power to shape reality.
notes
An earlier version of this chapter was presented at the Annual Meeting of the
American Academy of Religion in Kansas City in 1991. I want to thank my mother,
Pichammal Nagarajan; Alan Dundes; Elizabeth Collins; the Fulbright-Hays Disser-
tation Research Award (1992–1994); Frederique Apffel Marglin; and Harvard Uni-
versity’s Women’s Studies in Religion Fellowship (2001–2002) Program and all
the wonderful fellows. I want especially to thank the editor of this volume, Tracy
Pintchman, for inviting me to present this paper on the panel ‘‘Women’s Rituals,
Women’s Lives’’ in 2002 at the University of Wisconsin South Asia Conference and
for inviting me to contribute to this book, and I thank the careful and thoughtful
anonymous reviewer at Oxford University Press for making a dense web of wonder-
ful suggestions, some of which I took and all of which made me think hard about
what I was trying to do, though I alone am responsible for any errors.
Most of all, I thank my husband and companion for these many years, Lee
Swenson, and our two children, Jaya and Uma, who have been eagerly pressing for the
kolam work to come forth.
1. For a broader view of the kolam and an exploration of other related aspects,
see Nagarajan 1993, 1998a, 1998b, 2000, and 2001, among others; and Kramrisch
1983.
2. The pottu has counterparts for men, although they take different forms and
reveal different kinds of knowledge.
3. For example, a large kolam outlined in red kavi (from a reddish soft powder)
signifies the highest degree of auspiciousness (first menstruation, marriage, child-
birth, and so forth) in an upper-caste household.
4. These terms mostly apply to upper-caste households; further research is

needed to understand how these categories would work from the position of multiple-
caste households.
5. Marglin 1985, Carmen and Marglin 1985, and others have argued that
events and bodies may be both auspicious and polluting, including acts of menstru-
ation, sexuality, childbirth, and certain kinds of auspicious deaths.
6. For some other critically important texts on the complex valences with
which to read menstrual and ritual pollution practices around the world, see Buckley
and Gottlieb 1988; Douglas 1984.
7. See Nagarajan 1993 for a fuller explanation.
8. There are numerous stories of the powers of chaste Tamil women. The clas-
sical Tamil texts of the Shilapadikaram and Manimekalai come to mind.
references
Blier, Suzanne Preston. 1995. African Vodun: Art, Psychology and Power. Chicago:
University of Chicago Press.
Buckley, Thomas, and Alma Gottlieb, eds. 1988. Blood Magic: The Anthropology of
Menstruation. Berkeley: University of California Press.
Carmen, John, and Frederique Apffel Marglin, eds. 1985. Purity and Auspiciousness in
Indian Society. Leiden: Brill.
Daniels, Valentine. 1984. Fluid Signs: Being a Person the Tamil Way. Berkeley: Uni-
versity of California Press.
Das, Veena. 1995. Critical Events: An Anthropological Perspective on Contemporary India.
Douglas, Mary. 1984 [1966]. Purity and Danger: An Analysis of the Concepts of Pollution
and Taboo. London: Ark.
Dumont, Louis. 1966. Homo Hierarcchicus: Essai sur le systeme des castes. Paris: Gal-
limard.
Eck, Diana. 1981. India’s Tirthas: ‘‘Crossings’’ in Sacred Geography. History of Religions
20, no. 4: 323–344.
———. 1985. Darsan: Seeing the Divine Image in India. Chambersburg, PA: Anima.
Geertz, Clifford. 1973. The Interpretation of Cultures. New York: Basic.
Hart, George L. 1973. ‘‘Woman and the Sacred in Ancient Tamilnad.’’ Journal of Asian
Studies 32: 233–250.
———. 1975. The Poems of Ancient Tamil: Their Milieu and Their Sanskrit Counterparts.
Berkeley: University of California Press.
Kramrisch, Stella. 1983 [1968]. ‘‘Unknown India: Ritual Art in Tribe and Village.’’
In Exploring India’s Sacred Art, ed. Barbara Stoler Miller, 85–120. Philadelphia:
University of Pennsylvania Press.
———. 1985. ‘‘The Ritual Arts of India.’’ In Aditi: The Living Arts of India, ed. Robert
Adams and Rajeev Sethi, 247–270. Washington DC: Smithsonian Institution Press.
Lakoff, George. 1987. Women, Fire and Dangerous Things: What Categories Reveal
about the Mind. Chicago: University of Chicago Press.
Lakoff, George, and Mark Johnson. 1980. Metaphors We Live By. Chicago: University
of Chicago Press.
Leslie, Julia. 1991. ‘‘Sri and Jyestha: Ambivalent Role Models for Women.’’ In Roles
and Rituals for Hindu Women, ed. Julia Leslie, 107–127. New Delhi: Motilal
Banarsidass.
Madan, T. N. 1987. Non-Renunciation: Themes and Interpretations of Hindu Culture.
Delhi: Oxford University Press.
Marglin, Frederique Apffel. 1985. Wives of the God-King: The Rituals of the Devadasis
of Puri. Delhi: Oxford University Press.
Nagarajan, Vijaya. 1991. ‘‘The Kolam and Ritual Aesthetics: Gender, Ritual
Pollution and Rice Paintings in Tamil Nadu.’’ Paper presented at the Annual
Meeting of the American Academy of Religion, November 23–26, Kansas City,
Missouri.
———. 1993. ‘‘Hosting the Divine: The Kolam in Tamil Nadu.’’ In Mud, Mirror
and Thread: Folk Traditions of Rural India, ed. Nora Fisher, 192–204. Middletown,
NJ: Grantha Corporation, and Santa Fe: Museum of New Mexico Press.
———. 1998a. ‘‘The Earth as Goddess Bhudevi: Towards a Theory of Embedded
Ecologies in Folk Hinduism.’’ In Purifying the Earthly Body of God: Religion and
Ecology in Hindu India, ed. Lance Nelson, 269–298. Albany: State University
of New York Press.
———. 1998b. ‘‘Hosting the Divine: The Kolam as Embedded Ritual, Aesthetic and
Ecology in Tamil Nadu, India.’’ Ph.D. diss., University of California, Berkeley.
———. 2000. ‘‘Rituals of Embedded Ecologies: Drawing Kolams, Marrying Trees
and Generating Auspiciousness.’’ In Hinduism and Ecology: The Intersection of
Earth, Sky, and Water, ed. Christopher Chapple and Mary Evelyn Tucker,
453–468. Cambridge, MA: Harvard University Press.
———. 2001. ‘‘(In)Corporating Threshold Art: Kolam Competitions, Patronage and
Colgate.’’ In Religions/Globalization: Theories and Cases, ed. Lois Lorentzen,
Dwight Hopkins, David Batstone, and Eduardo Mandieta, 161–186. Durham,
NC: Duke University Press.
Tamil Women, ed. Susan Wadley, 35–60. New Delhi: Manohar.
Shulman, David. 1985. The King and the Clown in South Indian Myth and Poetry.
Princeton, NJ: Princeton University Press.
Wadley, Susan, ed. 1980. The Powers of Tamil Women. New Delhi: Manohar.
Williams, Raymond. 1983. Keywords: A Vocabulary of Culture and Society. New York:
part ii
Beyond Domesticity
6
Domesticity and Difference/
Women and Men: Religious
Life in Medieval Tamilnadu
Leslie C. Orr
In recent scholarly examinations of women’s ritual activity, there has

been an increasing appreciation and valorization of the ‘‘domestic.’’
This analytic category, which has been utilized with particular
success by Susan Starr Sered in her research on elderly Jewish
women’s religious lives, has afforded investigators the opportunity
to explore new contexts, to abandon an exclusive focus on formal
and public expressions of religiosity, and to consider also the
religious activities that take place outside of these frameworks and
in arenas where women are more likely to be found (Sered 1992,
139–140). But the conception of ‘‘domesticity’’ implies not only
a rethinking of the location of religion; even more important is
the fact that it prompts us to pay attention to the interpretation of
religious activities by those who engage in them. As Sered says,
domesticity ‘‘is not an inherent characteristic of any particular ritual,
place, or event,’’ but is above all a matter of intent, in which ‘‘the
ultimate concerns of life, suffering, and death are personalized—
domestic religion has to do with the lives, sufferings, and deaths of
particular, usually well-loved, individuals’’ (Sered 1992, 32). The
acknowledgment of the validity and significance of such an ori-
entation seems particularly helpful in the study of women’s religious
lives within the Hindu tradition, whose institutional structures
generally exclude women from publicly recognized roles as re-
nunciants or ritual specialists and whose textual traditions focus
largely on men as the central religious actors and on transcend-
ing attachment as the primary goal of religious activity.
110 beyond domesticit y
I propose in the present chapter to explore whether a domestic religious

orientation, engaged with the personal and the particular, can be discerned in the
context of precolonial South India. My focus is on the period of the ninth to
thirteenth centuries in that part of India today known as Tamilnadu, and I draw
on the resources provided by the thousands of inscriptions composed in the
Tamil language and engraved in stone on the walls of Hindu and Jain temples
during this period. These inscriptions record actions, particularly the making of
gifts to temples, that were undertaken by a wide variety of people. These people—
who sponsored building projects and gave land, money, livestock, ornaments,
and images to the temple—included kings and queens, merchants and shep-
herds, Brahmins and temple women, and local ‘‘lords’’ and their wives and daugh-
ters. There is, of course, a great deal that the inscriptions do not tell us; religious
activities carried out in the home and in other contexts apart from the temple
were rarely documented. Further, the inscriptions’ accounts are restricted to the
undertakings of people who had both the means to commission the engraving of
such records and the desire that their actions should be known to posterity. But
despite these limitations, the inscriptional corpus represents a discursive and
social space in which both men and women participated, and it gives voice to
particular individuals whose goals and motivations in undertaking various reli-
gious activities can be glimpsed in the records they have left.
Vows and Self-Offering
Perhaps the quintessentially domestic religious activity with which contem-

porary Hindu women are engaged, and which is attracting increasing scholarly
attention, is the ‘‘vow’’ or, in Sanskrit, vrata (Reynolds 1980; Robinson 1985;
Peterson 1988; McGee 1991; Tewari 1991; Pearson 1996; McDaniel 2003).
Women’s observance of vratas typically involves worship, the creation of ritual
designs, fasting or other austerities, and the recitation of stories concerning the
origin and power of the vow. These rituals are usually undertaken annually, on
days sacred to the particular deity whose blessings are sought, and almost
invariably have as their overt aim the well-being of children, brothers, or hus-
bands. Hindu women throughout India observe such vows, and in contem-
porary Tamilnadu, they are referred to by the term nonpu. It is therefore a
matter of great interest to discover this word in the medieval Tamil inscriptions
in the context of women’s religious activities. Three stones set up in front of the
temple dedicated to the goddess Mariyamman in Kandachipuram (a small
place in South Arcot district), with short inscriptions in tenth-century char-
acters, record the observance (nol) of nonpu by three women, all of them iden-
tified with reference to their fathers (ARE 57, 58, 59 of 1935–1936); another
stone found in a field in the nearby village of Ariyur documents the nonpu in
the Durga temple of a woman who is nameless but is identified as someone’s
domesticity and difference/women and men 111
wife (ARE 234 of 1936–1937).1 Another four stones from Gangayanur, in the
same area and evidently inscribed in the same period, bear very similar in-
scriptions; here the term parani is used instead of nonpu, but the verb nol,
which is related to nonpu and means ‘‘to endure, suffer, do penance,’’ is applied
to the action of the four women—each of whom is identified both as a daughter
and as a wife (ARE 458, 459, 460, 461 of 1937–1938).2 As these eight inscrip-
tions are so few and so terse, we may find it useful to turn to Tamil literary
sources for a greater understanding of the meaning of the term nonpu and the
possible contexts for women’s votive practices a thousand years ago.
An obvious starting point for such an examination is with two ninth-
century devotional works, Tiruppavai and Tiruvempavai, the first composed by
the female poet-saint Andal, a devotee of Vishnu, and the second by Man-
ikkavachakar, whose poems praise Lord Shiva. Both of these hymns frame their
expressions of praise and self-dedication within the context of a women’s ritual
in which a group of young women wake up early in the morning during the
month of Markali in the cold season and go together to bathe in a pond or river,
with the object of being granted a good husband. Andal’s Tiruppavai is par-
ticularly detailed and informs us of the terms of the vow: the girls refrain from
consuming ghee and milk, they bathe in the cold water and keep themselves
unadorned, and they distribute alms. Following this period of self-denial, the
girls dress up, feast on rich milk-rice, and take part in a procession featuring
music, drums, singing, lamps, flags, and banners (Tiruppavai 2, 26, 27).
The pavai rituals outlined in these two ninth-century devotional hymns
may possibly be linked to the temple festivals in the month of Markali that are
described in inscriptions of the eleventh to thirteenth centuries. Although the
inscriptions do not mention any special observances undertaken by women,
or austerities as part of the proceedings, there are processions and feasting
and even, at the temple of Tirupampuram (Tanjavur district), arrangements
made for the singing of Tiruvempavai.3 It is difficult, however, to see any
connections between the pavai observances of the poems and the practices
commemorated by the tenth-century nonpu and parani stones. For one thing,
Andal and Manikkavachakar do not use the term nonpu to refer to the girls’
ritual, which is called simply pavai or markali niratal, ‘‘Markali bathing,’’ in the
poems. When we do encounter this term in the Tamil devotional literature—
for example, in the hymns praising Shiva (composed in the sixth to ninth
centuries) that are collected in Tevaram—it consistently refers to severe ascetic
practices. In fact, those who most often perform nonpu in these texts are Jain
monks, who are castigated by the Shaiva poets for their excesses (e.g., Tevaram
2:121.10; 3:103.10; 5:32.9). In the sixth-century Buddhist text Manimekalai, the
word also refers to austerities, which, again, are presented in negative terms, as
selfish and vain, when undertaken by Jains (3.75, 90, 120; 21.98), although the
text displays admiration for others who practice nonpu—including the Buddha
himself (3.60; 5.99) and other sages and ascetics.4 Twice in the Manimekalai,
nonpu refers specifically to fasting, and in one case this is a fast to death; neither
of these fasts is disparaged, nor identified as a Jain observance (14.95; 17.42).5
In light of the literary evidence, the tenth-century nonpu/parani stones
would seem to commemorate the achievement of women who had carried out
rather extraordinary acts of self-discipline—more along the lines of the nonpu
of renunciants than the votive ritual of the young women in the pavai poems.
In contrast to those who observed the pavai ritual, hoping to attain the aus-
picious state of marriage, the women who undertook severe ascetic practices
would have done so with intentions similar to those of renunciants, seeking
detachment from the realm of relationship and desire—a manifestly nondo-
mestic goal. The inscribed stones, perhaps set up by their fathers or their
husbands, may have honored them for having completed an especially lengthy
and rigorous fast, of the sort that is undertaken by Jain women today and
celebrated by their families at its conclusion (Reynell 1987).6 Or perhaps these
women died in consequence of their austerities—since this is what is usually
implied by the existence of a commemorative stone, in medieval South India
as in contemporary North America.
For example, what is called in Tamil a nicitikai memorial was established
in honor of Jains, both laypeople and renunciants, who undertook a fast to
death (sallekhana). In contrast to the many such inscriptions from the medi-
eval period that one finds to the west, in what is today the state of Karnataka
(Settar and Korisettar 1982; Orr 1999), there are only a handful of these in
Tamilnadu. Of the Tamil records, there is only one (from Coimbatore district,
in the western part of Tamilnadu) that commemorates the fast to death of a
woman, and her name, Pullappai, indicates that she was originally from
Karnataka (ARE 597 of 1905). It appears that the practice of fasting to death as
a means of purification—avoiding harmful behavior and the accumulation of
bad karma, striving for transcendence of one’s physical state and liberation
from this world—was less common, or in any case less commonly commem-
orated, by Jains in Tamilnadu as compared to their cousins to the west. But in
the Tamil country, these observances may not have been exclusive to the Jain
community: it is possible that such practices and such motives are precisely
those celebrated in the nonpu/parani stones.7
Another type of memorial that, again, is more commonly found in Kar-
nataka (Settar 1982, 196)—and, even farther to the west, in Maharashtra
(Sontheimer 1982, 277–281)—is the sati stone. Indeed I know of no inscribed
sati stones in Tamilnadu (cf. Srinivasan 1960, 6–7). There are, however, at
least four inscriptions of the medieval period that have been engraved on
temple walls in various parts of the Tamil country that record women’s self-
immolation: a tenth-century inscription from Allur, in Tiruchirappalli district,
that records the gift of gold to the temple by a woman named Gangamadeviyar
‘‘who was entering the fire’’ (SII 8.690); an inscription from Dharmapuri
district, dated a.d. 1017, in which a wife is said to have ‘‘entered the fire’’
following her husband’s death (Avanam 12, 21);8 the grant of land, recorded at
Cheydunganallur in Tirunelveli district in the mid-twelfth century, for the
merit of Puricanti, who ‘‘entered the fire’’ (ARE 363 of 1959–1960); and the gift
of lamps by Vikrama Kampan for the merit of two persons who had died as a
result of his attack—a warrior whose home had been ambushed and his wife,
Vampu, who had subsequently ‘‘entered the fire’’—the donor having been
ordered by the elders of the community to make this gift to the temple at
Tirukalakkunram (Chingleput district) so that the brother of the dead man
should desist from further vengeance (ARE 162 of 1932–1933).9 What is in-
teresting is that two of these four inscriptions make no reference to the wom-
an’s status as a wife or a widow. In the absence of the mention of a husband,
we may question whether these acts of self-immolation were satis at all, and I
suggest that these women’s self-sacrifice was not motivated by the desire to be
reborn again with the bond of marriage preserved and to maintain the web of
personal and particular familial ties even after death. That the first of these
inscriptions records the woman’s own gift to the temple, rather than arrange-
ments made by others to honor her or make expiation for her death, indicates
that the relationship being solemnized by her renunciation of life was that with
God rather than with man. Thus this act of self-giving appears to be inspired by
an impulse to transcend one’s specific human and social condition.
In the Tamil country, and especially in the northwest parts of this region,
the type of memorial we encounter most frequently is the hero stone. It is clear
from the Tamil ‘‘Sangam’’ literature of the early centuries of the first millen-
nium that the setting up of hero stones for men who had fallen in battle or
in cattle raids was a long-standing tradition (Srinivasan 1960, 3–6; Soundara
Rajan 1982, 59–75; Settar 1982, 184–187). The hero stones from the period we
are considering—the ninth to thirteenth centuries—generally bear an image of
an armed man and a Shiva linga, and about half of them are inscribed with a
short statement telling how the hero met with death. Often he was engaged in
combat on behalf of a lord and is referred to as his servant (cevakan); sometimes
the person or group who erected the stone will also be mentioned (see, e.g., CN
56). As is the case for nicitikai and sati memorials, hero stones are much more
abundant in Karnataka than in Tamilnadu, and there they are adorned with
more elaborate relief sculptures, including images of the hero ascending heav-
enward in the embrace of celestial maidens (Settar 1982; Rajasekhara 1982).
Although the hero stones found in Tamilnadu lack such narrative depictions
of the hero’s postmortem destiny, the sculpting of the Shiva linga on many of
these memorials suggests that the hero in death is consecrated to Lord Shiva,
and will attain his divine abode—or, perhaps, that the hero’s courageous pro-
tection of his community and loyalty to his chief are equated with the acts of
self-sacrificing devotion performed by the worshipper.10
Another kind of self-offering recorded in the medieval Tamil inscriptions—
and one with which women as well as men were involved—is the vow of the
servant (fem. velaikkari, masc. velaikkaran) not to survive her or his master.
There is a group of more than thirty records of such oaths of fealty by both
women and men inscribed on the temple walls of Arakandanallur, in South
Arcot district—a place very near to the sites where the nonpu/parani stones
have been found (ARE 122–126, 136–150, 153–160, 162, 187, 188 of 1934–
1935). Another such record is found in the temple at nearby Elvanasur (SII
22.156): here we read of the vow made by the velaikkari Tevapperumal that she
die together with her lord. Although these vows express the same values of
faithfulness and self-dedication that are manifest in the hero stone inscrip-
tions, they lack any explicit reference to religious motivations and expecta-
tions, devotion to a deity or anticipation of a heavenly reward; these records
resemble other oaths engraved on temple walls in this part of South Arcot
district, which solemnize political alliances among clans and agreements of
mutual defense (see Orr 1998b).
Although we do not hear of any cases where the velaikkari or velaikkaran’s
promise of self-sacrifice in allegiance to a human lord was actually effected,
devotion to the cause of a divine lord did result in acts of self-immolation.
There is an inscription of the twelfth century from Punjai, to the south, in the
Kaveri delta region, where we see a group of servants (velaikkarar) loyal to the
trident of Shiva, who gave up their lives—by entering the fire—in support of
the temple’s contention of ownership of certain properties. The rival claimants
to the land had to concede to the temple and were required by the local as-
sembly to set up metal images of those who had died and to make a donation to
provide for worship of these images (ARE 188 of 1925). At Paiyanur in Chin-
gleput district, an inscription of uncertain date records another land dispute,
evidently between the temple and the Brahmin assembly, in which two ascetics
gave up their lives (ARE 108 of 1932–1933). Finally, we might add to these
examples of self-sacrifice a couple of cases in which the interests of the temple
deity were furthered by the offering of one’s head. At the temples of Jambai
and Arakandanallur—in precisely the same area of South Arcot district where
we find the nonpu/parani stones and the velaikkarar oaths of fealty discussed
above—inscriptions of the thirteenth and fourteenth centuries record land
grants to the families of men who had cut off their heads in order that the
temple mandapa might be completed (SII 12.178; ARE 197 of 1934–1935). It
may be significant that, in both cases, these men were the relatives of temple
women, but such extreme acts of self-dedication do not seem to have been
undertaken by temple women themselves. It is unclear how precisely such acts
were efficacious in advancing temple building projects; perhaps the self-sac-
rifice exerted moral pressure on the people of the community who were in a
position to provide financial support to the temple, in the same way that the
deaths of Shiva’s servants at Punjai forced local landowners to submit to the
temple’s demands. A further possibility, however, is suggested by the wording
of another memorial of a head offering: this is a stone, with a representation of
a man having decapitated himself with a sword, which bears a tenth-century

inscription recording the man’s offering of his head to the goddess and the
granting of land to his relative, in recognition of this superior asceticism
(me tavam) (SII 12.106).11 The use of the term tavam, equivalent to the San-
skrit tapas, suggests that this act is equivalent to the austerities of sages and
renunciants—performers of nonpu—and generative of a spiritual energy ca-
pable of bringing about a desired goal.
In the case of the nonpu/parani stones and several of the records of women
entering the fire, the aims of the performers remain obscure, yet all of the self-
sacrificing acts we have considered here seem to have a similar quality: the
virtues displayed by women cannot be distinguished from those of the men
whose deaths are commemorated in medieval Tamil inscriptions. These ex-
traordinary acts and the fashion in which they were recognized and cele-
brated reflect the value placed on valor and self-sacrifice, whether that of
women or of men. Within this group of activities, the undertaking of aus-
terities is apparently the focus for some of the observances, while elsewhere
loyalty and allegiance are stressed. In most cases, both aspects are present, as
are both domestic and transcendent elements. Ironically, perhaps, women’s
acts of self-discipline and self-offering, documented in the records of nonpus
and satis, point less consistently toward a domestic orientation, with their lack
of reference to beneficiaries of the observance and their suggestion that it is
the woman herself who attains her religious goal by renouncing her connec-
tion with the world. The self-sacrifices represented in the hero stones, in the
velaikkarar oaths of fealty, and in the self-immolations and self-decapitations
performed for the sake of the temple deity are motivated by this-worldly goals
and are highly personalized, and thus domestic. The sense of relationship
with and service to one’s community, lord, or local deity—concern for ‘‘par-
ticular, usually well-loved, individuals,’’ in Sered’s words—is much more viv-
idly presented in these cases, which refer for the most part to men’s actions,
than in the records of nonpus and satis carried out by women. But even here,
in the absence of an explicit reference to a divine power or to a future exis-
tence after death, a transcendent dimension is invoked—by the celebration
of the devotion, mastery of the self, and fearlessness that have allowed those
whose acts are commemorated to rise above the ordinary sphere of human
conduct.
Gifts to Gods and Goddesses
The vast majority of medieval inscriptions concern much more mundane

activities, but they do have another type of connection to the transcendent
inasmuch as they record gifts to the divine beings enshrined in the tem-
ple. Overall, the kinds of gifts that women and men made were identical in
substance and in purpose. But closer examination reveals subtle differences, in

terms of preferences and emphases in donative activity. To get a more precise
sense of what these differences are, I surveyed a group of more than 2,000
inscriptions, of the ninth to thirteenth centuries, from six study areas in dif-
ferent parts of the Tamil country.12 About 1,200 of these inscriptions are the
records of gifts from individuals where it is possible to determine the sex of the
donor: 161, or a seventh of these, concern the gifts of women. Most of the gifts
of both men and women were for lamps to be burned in the temple (42 percent
of women’s gifts; 35 percent of men’s) or for various worship services and
festival observances (16 percent of women’s gifts; 25 percent of men’s).13 But a
number of the inscriptions record the setting up of images of deities—and this
is a type of endowment that was apparently of considerable interest to women.
While overall only one-seventh of the inscriptions recording gifts by individ-
uals present women as donors, nearly a quarter of the 100 or so inscriptions
documenting the establishment of new images in the temple identify women
as their sponsors. Of the 25 goddess images newly consecrated, a third had
been set up by women. This pattern seems to indicate a special affinity of
women for the patronage of goddesses, and this is borne out in other ways: in
the case of the 40 inscriptions recording arrangements for worship and special
gifts for goddesses already established in the temple, a quarter identify women
as sponsors, a proportion that is about twice what one would expect given the
number of female donors. Women’s visibility in the establishment of images
and the patronage of goddesses is all the more impressive when we consider
that in inscriptions of the ninth to thirteenth centuries, half of all the images
were sponsored and half of all the goddess-related donations were made during
the thirteenth century—a period when the proportion of all gifts that were
made by women had dropped to its lowest point, just over 10 percent (see Orr
2000b).
Apart from the quantitative patterns that this survey of study areas reveals,
we find in inscriptions recording women’s donations from all parts of Ta-
milnadu certain predilections with respect to the kinds of gifts they chose to offer
goddesses. Very often they donated jewelry to adorn the goddesses’ images. We
see this even in the case of a Jain religious woman, identified as the disciple of
a male sage, who presented gold ornaments to a Jain goddess at Chitaral in
Kanniyakumari district at the end of the ninth century (TAS 1.194–195). Around
the year 1000, there are many records of gifts of ornaments made by women of
the royal court to goddesses enshrined in the temples of the Kaveri delta (see
Venkataraman 1976, 52–58, 66–71, 130–136). Among these donations, it is in-
teresting to find in several instances the gift of an adornment referred to as a tali.
Judging from the Tamil literature of the period, the word tali could simply mean
a particular type of necklace, but beginning in the eleventh century, it began also
to designate the emblem of marriage worn by women so long as their husbands
remained alive (Jayadev 1960, 50–51). In a.d. 981, a tali was presented by the
Chola queen Cempiyan Mahadevi to the goddess Uma at Vrddhachalam (SII

19.302); a decade later, a palace woman in the service of one of Rajaraja’s queens
gave to the goddess Uma at Tiruvidaimarudur a gold tali set with a double row of
gems and a pearl necklace (SII 23.278); and, in 1015 at Tiruvisalur, the daughter
of a chief from northern Tamilnadu who had married the Pandya king of the
south similarly presented the goddess with a jewel-encrusted gold tali and other
valuable ornaments (SII 23.46). Later, toward the end of the thirteenth century,
we find the wife of a temple Brahmin giving her own tali to the goddess of the
temple at Tiruvattatturai in South Arcot district (ARE 227 of 1928–1929). The
gift of this special ornament, in the case of at least some of these women, would
seem to express a particularly feminine concern that the goddess—in all of these
cases, the consort of Shiva—be appropriately adorned as a wife. Further, the
auspiciousness associated with the tali would be greatly enhanced when worn by
the goddess (whose husband is immortal), and she who had presented the tali
would have shared in that heightened auspiciousness, although she would not
have received a tali in return, as do the matammas of Tirupati described by Joyce
Flueckiger elsewhere in this volume. On the other hand, the Tamil inscriptions
rarely record male gifts to the goddess of jewelry of any kind, and I have come
across only one instance where a tali was offered by someone other than an
individual woman; this is in an inscription of the late tenth century which re-
cords the granting of a tali to the goddess by the Brahmin assembly at Tiruvallam
in North Arcot district (ARE 210 of 1921).
Another dimension of the feminine solidarity expressed through wom-
en’s adornment of female deities emerges in inscriptions that indicate that a
woman’s gift to the goddess was a means of linking her with her female kin
and of connecting them with the goddess. For example, a gift for the merit of
her mother was made by a woman of the Malaiyaman chief’s family, who, in
1133, built a shrine and installed an image of the ‘‘bedroom goddess’’ (tir-
upalliyarai nacciyar), the form of the goddess whom the god Shiva would join
every evening, at the temple of Siddhalingamadam in South Arcot district (SII
26.422). An inscription of the late twelfth century from Viravanallur in the far
south of Tamilnadu (Tirunelveli district) records that a woman serving in the
palace of the Pandya kings at Madurai set up an image of the goddess, in the
name of her daughter and named after her daughter, to which she presented
jewels and other gifts to support worship (ARE 720 of 1916). And a hundred
years later, in 1300, a dancing woman (cantikkutti) of the temple at Tiruvanaik-
kaval built a goddess shrine in the town of Valliyur, again in Tirunelveli dis-
trict and far to the south of her hometown, where she set up images of the
goddess and of her granddaughter (ARE 364 of 1929–1930).14
Sometimes the network of female relations and feminine concerns is
expressed in terms of a familial relationship between female donors and the
goddesses themselves. In two inscriptions of the late tenth century from
Tanjavur district, we find women referring to the goddess Uma, whom they
had endowed with land, as their daughter (SII 19.404; Varalaru 1, 33–34).15
But in another inscription of precisely the same period, from a temple farther
up the Kaveri River, it is a male donor who claims the goddess Uma as his
daughter, provides her with land to support daily worship and offerings,
and gives her in marriage to the lord of the temple (ARE 151 of 1936–1937).
And several other inscriptions, somewhat later and from farther north, record
land grants to goddesses given—by individual men or local assemblies—as
stridhana (dowry) on the occasion of the goddess’s marriage, again expressing
a parental role and demonstrating that it was not only women who adopted a
highly personal approach to temple patronage.16
There are two ways in which this concept of connection with a deity
resonates with the Tamil devotional literature of the centuries immediately
preceding the time in which the inscriptions were engraved. First, the sense of
intimacy is present in both contexts, within the framework of a variety of
possible close relationships, including—especially in the cases of the poems
of the Alvars dedicated to Vishnu and in the inscriptions at which we have just
looked—that of the role of parent to the divine child.17 Even where a familial
connection is not explicitly evoked, the poems and the inscriptions express—
each in their own fashion—a familiar personal relationship between the
devotee and the divine. Second, the particularity of place is emphasized: the
deity being praised by the poet-saints or granted gifts by the worshipper is not
represented primarily as an abstract and universal power—although his or her
transcendent nature is of course acknowledged—but is recognized as the lord
or mistress of a particular locality. The deity’s distinctive site-specific per-
sonality is far more manifest in the inscriptions than is his or her sectarian
or iconographic identity. These qualities of intimacy and local particularity
found among worshippers in medieval South India—and the perspective in
which the goddess is regarded as a daughter—prompt a comparison with the
ritual activities of women in the contemporary celebrations of Durga puja in
Bengal analyzed by Sandra Robinson. As an example of women’s ritual ac-
tivities that ‘‘are separate from but coordinate with brahmanic festivals,’’
Robinson describes what takes place after the priest’s conclusion of the formal
worship:
[Women] approach the image of the goddess Durga to place food

on her lips as a gesture of farewell before the image is taken away for
immersion in a pond or tank. A psychodrama of reluctant departure
accompanies this activity, inasmuch as the goddess has come only
to go home again; there are wailing laments which explicitly repli-
cate and anticipate the farewells of young brides as they leave their
own families to return to their husbands’ family homes. (Robinson
1985, 196–198)
In Robinson’s analysis of the temple setting of modern Bengal, there is a

definite division between the official observances of the male priest, mandated
by the Brahminical tradition, and the activities of female worshippers, which
are clearly marked as domestic because of their personalized character and
which are treated as marginal or supplementary rituals, however essential
they seem to the women who perform them. There are two aspects of this
dichotomy—the split between formal male ritual and domestic female ritual,
and that between the role of the specially qualified ritualist and the role of the
‘‘lay’’ worshipper—and, in each case, the first of the two activities or roles is
acknowledged as fundamental, and the second is generally viewed as auxiliary
or derivative. In the context of the medieval South Indian temple, on the other
hand, at least from the perspective offered to us by the inscriptions, I would
argue that there was a recognition of the value of the domestic orientation as
expressed through the public religious activities of both women and men, and
that lay religious activity—particularly temple patronage—by both women and
men had an authority and impact that was not overshadowed by the prestige
and expertise of priests and other temple servants.
If we consider the corps of temple personnel as the main actors, and their
ritual duties as the crucial functions, we will perceive in the medieval South
Indian temple a situation that is more complex but not dissimilar to the picture
of the temple in modern Bengal that Robinson has sketched for us, with
respect to the recognition of the relative value of male and female roles. While
the temple in medieval Tamilnadu presented more possibilities for female
ritual participation than is the case today, women’s presence and activities in
temple ritual were regarded as optional and incidental (Orr 1993, 2000a). Tem-
ple women were vastly outnumbered by their male counterparts and were
entirely excluded from many of the roles that men fulfilled, including that of
priest. The increasing visibility, in the thirteenth century, of temple women as
singers and dancers at festivals—displacing, to some extent, the male per-
formers who had earlier had these roles—seems to have been significant more
as a manifestation of the privileges accorded to temple women (often in con-
sequence of their gifts to the temple), than as the provision of a necessary
service to the temple.18 But if we shift our gaze from the temple servants to the
temple’s patrons, we see that women were far from marginal and that the
religious activity of gift giving had a profound impact on the shaping of ritual
life in the temple, giving form to new services and establishing new deities to
be worshipped.
Donors were, of course, not only motivated by the desire to make their
mark—both by effecting changes in the temple through their gifts and by
documenting their piety and generosity for posterity on the temple walls—but
had other objects in view. The expression of purpose that is most commonly
encountered in the inscriptions is related to the creation and transfer of the
merit that accompanies the making of a religious gift. In the medieval Tamil
inscriptions, we do not find the explicit mention of punya or other terms
denoting ‘‘merit,’’ but the transfer of merit is indicated by the statement that a
gift was made ‘‘for’’ another or was ‘‘connected to’’ (cartti) the recipient of the
benefit of the donation. Although both men and women transferred merit at
the same rate—in about 5 percent of their gifts—there were clear differences
in the identities of the beneficiaries.19 Merit produced by the gifts of women
was almost invariably transferred to relatives, while this was the pattern for
men’s gifts only about a third of the time; the merit more often went to men
unrelated to the donor, including the king or local notables. Of the relatives
to whom merit was transferred, by both women and men, male kinsfolk
predominated—but the relative most likely to be mentioned by a male donor,
his father, was scarcely ever mentioned by women. Here we see women lo-
cating themselves through their gift giving within a familial network in a
pattern that is familiar from studies of contemporary Hindu women’s votive
behavior—where vows are undertaken for the benefit of husband and children
(and particularly, perhaps, sons)—and it seems that this is more characteristic
of medieval women’s donative behavior than it is of their male counterparts’.
But it cannot be said that men’s purposes in making gifts were radically dif-
ferent: they too had domestic motives and used temple patronage as a means of
expressing connection with their kinsmen—as well as their mothers, sisters,
and wives. That men may, in some ways, have had a greater stake than women
in domestic goals, at least as these were made public in temple inscriptions, is
suggested by the fact that several of the donations by men to temples are said in
the inscriptions to be gifts of thanksgiving for having been granted a son (SII
26.516; ARE 366 of 1959–1960). There is nothing equivalent among female
donors.
Apart from such donations, made in gratitude, what kinds of future conse-
quences were anticipated as the result of making gifts on behalf of oneself or
another? The inscriptions tell us almost nothing about the benefits for the living.
But a hint about the goal of gifts made for the merit of those who had left this
world appears in an early twelfth-century inscription that records the gift of land
for one who was deceased, ‘‘praying [that he attain] Shivaloka’’ (civalokaprarttam)
(SII 8.460). Other possibilities are suggested by the pallipatai, or ‘‘sepulcher
temple,’’ erected in medieval Tamilnadu over the remains of a prominent person
and named after him or her (Srinivasan 1960, 12–13; Sethuraman 1991). Amid
the obscurity that surrounds the meaning of these shrines, several authors
maintain that their primary function was to legitimate the claim to rule by Chola
kings descended from the deceased, who was glorified as a hero and king (Ra-
ghotham 1995; Ogura 1999). But the fact that more than half of those interred
within the pallipatais were women seems to suggest rather different motives.20
Meanwhile, in religious terms, it is not clear whether gifts for the merit of the
departed, or the setting up of nonpu stones, hero stones, or pallipatais, had to do
with fame and the honoring—or even ‘‘deifying’’—of a person of the past, or
whether instead they were a matter of supporting the departed in an ongoing
existence in the afterlife, where benefits would flow from the continuing wor-
ship carried out in his or her name.
Some further clues come from the inscriptions that record the arrange-
ments made after someone had met a violent death. We have already had
occasion to consider the inscription (ARE 162 of 1932–1933) that describes the
gift made by Vikrama Kampan in expiation of his murderous attack on a man
and the subsequent act of sati by that man’s wife: in this case, we learn that the
two lamps were given for the merit of the two dead and as a means of averting
the vengeance of the living. There are a few other such records where violence
was purposefully done, either in battle or in rage—there is, for example, the
case of the Brahmin who had beaten a man to death and who was required to
give a perpetual lamp in expiation of this crime (ARE 528 of 1937–1938). But,
according to the inscriptions, most gifts of expiation were made after having
inadvertently caused another’s death. There were a large number of hunting
accidents—described in inscriptions from Chingleput, North Arcot, and South
Arcot districts—that prompted such gifts. In one such case, the motive for
making the gift seems to have been to placate the spirit of the dead: the local
council determined that the man who had accidentally killed another while
hunting should donate a perpetual lamp to the temple in order to ‘‘remove
enmity’’ (pakai ara) (SII 7.85). This is also suggested by the case in which an
intruder was stabbed to death by a merchant as the intruder attempted to rape
the merchant’s concubine; both the merchant and a relative of the deceased
were together held responsible for donating gold for a lamp in the temple.
Here the resolution does not so much reflect culpability or the need for the
merchant to atone for wrongdoing but rather the necessity for there to be a
memorial for the deceased (SII 22.77).
Women as well as men figure in these records, both as victims and as
perpetrators. We have two inscriptions that order a gift of expiation for hav-
ing caused the death of a woman—one of whom took poison because her
husband was marrying another woman (SII 17.389), and another who killed
herself after she had been put through an ordeal to get her to pay taxes she
claimed she did not owe (SII 22.80). A chilling inscription of the twelfth cen-
tury describes an act of violence carried out by a woman: Koccattan Kaman’s
wife threw a stick at her daughter, which accidentally hit another girl; the
girl died twenty days later. It was agreed that a lamp should be donated to
the temple by Koccattan Kaman, the husband of the murderous mother (SII
22.148). The inscription provides neither the name of his wife, nor that of the
dead girl, who is identified simply as the daughter of Tappi Mintan Kaman. It
is unlikely, therefore, that the lamp was being offered for the merit of the
nameless girl. In fact, although women are the main actors in this unhappy
drama, they seem to be entirely off-stage in the sequel. The gift of expiation
was not made by the perpetrator of the violent act, but rather by her husband,
who seems to bear responsibility for his wife’s action.
If religious activity is regarded as domestic when it is undertaken for
personal and individualized motives and for the benefit of family members
or other well-loved individuals, it is difficult to make the case that medieval
South Indian women’s engagement with temple patronage indicates such an
orientation to a greater degree than that of their male counterparts. It is true
that women were more likely than men to transfer the merit arising from their
gift giving to their kinfolk, especially their sons and husbands. But women
refrained from making gifts for expiation or thanksgiving, which were the
types of gifts that frequently expressed involvement on the part of men in
a familial network and that made men the representatives for their family’s
culpability and its interests. Donations were also the means by which indi-
viduals forged links with the divinity enshrined in the temple. On the one
hand, women’s attention to and special gifts of jewelry for consort goddesses
suggest a connection to these deities made more intimate by the sharing of the
status of wife. On the other, records of men’s gifts to the goddess as daughter
express the closeness and tenderness of their relationship with the deity whom
they worshipped. And both men and women, through their donations to gods
and goddesses alike, participated in a realm of religious activity that was—in
terms of its expression (and constitution) of a personal relationship with a deity
and its focus on the identity of the particular god of a specific place—highly
domestic in character.
Concluding Reflections
The category of the domestic is clearly a useful one in expanding our sense of
what can properly and seriously be regarded as a religious context, a religious
role, or a religious motive. Religion is not just what yogis and priests do. But
the characterization of women’s engagement with religion as preeminently
domestic—personal, particular, and familial—and distinct from men’s frame-
works of religious undertaking cannot be maintained in the face of the evi-
dence from medieval Tamilnadu. Women’s and men’s religious lives were of
course dissimilar in many ways. But the spheres of activity in which they found
themselves were in large part congruent. If we want to differentiate male and
female religious behaviors and purposes, it seems that we can best do this by
thinking in terms of different colorings and shadings, rather than different
positions or different perspectives on the meanings of their religious activities.
The inscriptions show us that renunciatory and self-sacrificial observances,
undertaken by both women and men, had significance both in terms of ultimate,
transcendent aims and as expressions of highly specific relationships and this-
worldly goals. The devotional activities that were a part of temple life, including
making gifts to gods and goddesses, similarly had both feminine and masculine
manifestations and simultaneously engaged abstract conceptions of devotion
and divinity, on the one hand, and personal connections with the local and par-
ticular, on the other. Finally, our exploration of the contexts and roles in which
medieval South Indian women functioned suggests that the category of the
‘‘religious’’ itself is in the end less helpful than we might have expected in an
analysis of these acts of renunciation and devotion. The inscriptions demon-
strate again and again that the religious act has antecedents, meanings, and
consequences that spill over into other realms. The records of women’s gifts
portray them as participants in networks of property transactions, as sponsors of
land improvement projects, and as parties to contracts and compacts. The epi-
graphs, although engraved on the walls of temples, speak to issues of identity,
position, and power—and not only mark the special gestures and achievements
of individuals but are themselves constitutive of the shifts in status that ac-
company these acts. The boundary between the religious and the political, or the
religious and the economic, is as indistinct as that between women’s domestic
rituals and men’s ‘‘high’’ religion. In other words, what yogis and priests—and
women—do is not just religion.
notes
I am very grateful to Tracy Pintchman for her patience and perseverance as editor
of this volume, to Norma Baumel Joseph for much conversation about and insight
into women’s ritual lives, to S. Swaminathan for invaluable aid in my work with
the inscriptional material, to Padma Kaimal for delightful discussions about kings,
queens, goddesses, and temples, to Anne Monius for sound advice on Tamil literary
matters, to Katherine Young with whom I first read the pavai poems, to Uma Narayan
for questions that helped clarify the issues at stake in this chapter, and to Michelle
Bakker for her bibliographic assistance.
1. That three of the nonpu stones are placed in proximity to a goddess temple, and
that one explicitly mentions the observance taking place in a goddess temple also
suggests a possible link to contemporary women’s votive ritual, which is often dedi-
cated to female deities (Reynolds 1980, 50; McDaniel 2003, 111–112).
2. One of the four parani stones records the fact that a pillar (tari—evidently
the stone itself which bears the inscription) has been erected in recognition of the
observance of the parani. In general, in inscriptional usage, the term parani refers
to a division of time (a nakshatra or asterism) which occurs every month. That the
practice of austerities, or the observance of a vow, should have a specific calendrical
referent is not surprising, although I know of no other examples where the nakshatra
parani has these associations.
3. A temple woman (tevaratiyal) was granted the privilege of singing part of
the hymn at Tirupampuram’s Markali festival in exchange for her gift of several
images to the temple, according to an inscription of the early thirteenth century
(NK 139). Other references to Markali festivals come from Chidambaram (SII
4.223), from Kudumiyamalai, in Tiruchirappalli district (IPS 291 and 301), and
from Tirumogur, in Madurai district, where it was the occasion for the celebration
of the marriage of Vishnu to the goddess (ARE 334 of 1918). See Orr 2004 for fur-
ther discussion of festival observances.
4. Monius (forthcoming) suggests that the Manimekalai’s emphasis on nonpu
indicates that this text is affiliated especially with Hinayana or Theravada Buddhist
teachings. Not surprisingly, in Cilappatikaram, a Jain text written at around the
same time as the Manimekalai, we find the word nonpu utilized with quite positive
connotations, generally with reference to the austerities of Jain renunciants and lay
people (10.24, 47; 15.153, 164; 16.18; 26.226), but also applied to the rites of mar-
riage observed by the hero (1.53).
5. We also find nonpu referring to the fast to death in Cilappatikaram (17.83).
6. The fasts carried out by contemporary Jain women have different aims
and different effects from those undertaken by Hindu women as an aspect of the
observance of votive rituals. Jain women’s austerities bring about a withdrawal from
the world and are a means of self-purification; the power generated by fasting (tapas)
is understood as ‘‘cooling’’ female sexuality (Reynell 1987, 343–351). On the other
hand, while the ascetic component of Hindu women’s vratas equally generates power,
this is utilized in the context of engagement with the world, allowing women to
influence the welfare of their families; their tapas, viewed as ‘‘heat,’’ has a benefi-
cent effect because it is channeled toward such ends (Reynolds 1980, 46–50;
McDaniel 2003, 108–109; cf. Pearson 1996, 211–217).
7. Three of the eight women whose observance of nonpu/parani was commem-
orated are described only as daughters and not as wives. These may have been very
young women carrying out this practice, or it is possible that they were renunciant
women, similar to the Jain ‘‘religious women’’ identified in medieval Tamil inscrip-
tions as teachers (Orr 1998a). There are very few inscriptional references to ascet-
ics among women whom we would classify as ‘‘Hindu’’—terms like tapasyar are
almost invariably applied to men rather than women—but those few that exist
(e.g., SII 8.225; ARE 120 of 1912), as well as the intriguing references to ‘‘pilgrim
mothers’’ (paradeshi ammaimar—e.g., SII 5.748; ARE 271 of 1927–28), point to the
potential for alternative ways of life for women.
8. I am indebted to S. Swaminathan for calling my attention to this inscrip-
tion, and for providing me with the text, published in the Tamil journal Avanam.
9. None of these four records commemorates the death of a royal woman.
There is an inscription at the temple of Brahmadesam, in North Arcot district, dat-
ing from the middle of the eleventh century, that records the gift of a water shed
by the brother of the Chola queen Viramahadeviyar, to quench the thirst of his sis-
ter who, having arisen to Shiva’s heaven and joined the feet of Brahma, was interred
in the pallipatai (sepulchre temple) of her deceased husband Rajendra I (ARE 260
of 1915). This inscription has often been considered evidence of the Chola queen’s
act of sati, and it is dated in the same year as Rajendra’s death, but, unlike the four
inscriptions we have just considered, there is no direct reference to Viramahadeviyar
taking her own life. A more explicit indication of such an act by a woman of the
Chola royal family—although it is found in a record dating a half century later than
the event itself—is found in the lengthy prashasti of Rajendra I that prefaces the
Tiruvalangadu copper plate inscription (SII 3.205). Here, in verses 65–66, the queen
Vanavanmahadevi is depicted as following her husband Sundara Chola (Rajendra’s
grandfather) to heaven, jealous of the attentions that the celestial maidens would
bestow upon him.
10. Padma Kaimal (1999, 132) points out that the portrait sculptures of donors
in early Chola period temples are stylistically similar to the images carved on hero
stones, and suggests that this reflects a similarity in the valuation of the worthiness
of the acts that are commemorated in the two contexts—both of which involve loyalty,
submission, and offering of the self. It is interesting to note that after AD 970,
portrait sculptures of female donors virtually disappear (despite the major activities
of women as temple patrons), while they were more in evidence in the preceding
hundred-year period (Kaimal 2000, 179).
11. On head-offerings to goddesses in South Indian sculpture and literature,
see Vogel (1930–32), Srinivasan (1960, 29–30), and Filliozat (1967).
12. These six study areas are—from the northern to the southern part of
Tamilnadu—Chingleput taluk in Chingleput district, Cheyyar taluk in North Arcot
district, Cuddalore taluk in South Arcot district, Mayavaram taluk in Tanjavur dis-
trict, Tiruppattur taluk in Ramnad district, and Ambasamudram taluk in Tirunelveli
district. For this survey, I relied on the abstracts of the published and unpublished
inscriptions (drawn for the most part from the ARE publications) in the volumes of
A Topographical List of the Inscriptions in the Tamil Nadu and Kerala States (Mahalingam
1985–).
13. Women were much less likely than men to sponsor arrangements for the
offering of food to Brahmins, ascetics, and devotees in the temple. This may be
because the institutionalization of such feeding eclipsed a sphere of religious activity
that had belonged to ‘‘the housemistress at the door’’ (Findly 2002; see also Balbir
1994). The giving of alms to mendicants probably continued to be largely a private,
informal, individual, and home-based affair, rather than taking place in the public,
permanent, and formal contexts described in the inscriptions, especially the matha
(‘‘mutt’’), which was predominantly a feeding-house in this period. But as feeding
came to be a means through which the temple, or the sectarian community, could
confer honor and recognition for patronage, service, and leadership, there may
have been a diminution of women’s importance with respect to this religious
activity.
14. Women’s gifts to the male deity enshrined in the temple, in support of
various offerings and services, could equally serve to create links with female kin—
or, for that matter, with the men of their families.
15. Again, I thank S. Swaminathan for alerting me to the existence of the in-
scription published in the Tamil journal Varalaru, and for kindly supplying me
with the text.
16. In AD 1037, a gift of land for a garden was given by a Brahmin man as
stridhana to the goddess Sita who was to be married to Lord Rama, at Vadamadurai
in Chingleput district (ARE 262 of 1952–53); in 1160, at the Shiva temple in Brah-
madesam in South Arcot district, the Brahmin assembly (mahasabhaiyar) provided
land as stridhana for the goddess who had been established in the temple by a
local man (ARE 192 of 1918); and in 1137, the village assembly (urar) gave land as
stridhana to the ‘‘bedroom goddess’’ at the Shiva temple of Seyyur, in Chingleput

district (SII 8.30). On stridhana in medieval Tamilnadu—which meant not only the
transfer of property to a daughter, but could involve the transfer from a man to his
son-in-law or a woman to her daughter-in-law—see Orr 1998b and Orr 2000a 72–73,
77–78, 226.
17. Despite the theological precedent in Tamil devotional literature, worship-
pers do not take a parental role in the inscriptions recording their gifts to male deities.
The god enshrined in the temple is not addressed as ‘‘son’’; instead we find the
reverse—that records issued in the name of the Lord refer to devotees as ‘‘our chil-
dren.’’ Perhaps this discrepancy can be explained by the fact that the majority of
temples providing inscriptions in our period were dedicated to Shiva, a deity who
is rarely if ever envisioned as a child—in contrast to his son Murugan, his consort
Uma, and various manifestations of Vishnu (see Richman 1997).
18. See note 3 above for an example of a temple woman’s ‘‘deal,’’ in which
her right to participate in ritual resulted from a donation. I have discussed at length
temple women’s activities—both as temple servants and as temple patrons—in
Orr 2000a.
19. This discussion is based on Orr 2000b.
20. In the seven inscriptions where there is an explicit indication of the character
of the temple as a pallipatai, and a clear identification of the person in whose
honor it was built, we find mention of four pallipatais constructed for men (of
whom three are Chola kings), and six for women (including five Chola queens). In
chronological order, from the ninth to the thirteenth centuries, these inscriptions are:
EI, 192ff; SII 8.529; SII 3.15; ARE 271 of 1927; ARE 260 of 1915; EI 41.10B; and
ARE 124 of 1936–37. Sethuraman (1991) and Ogura (1999) emphasize the role of
Pashupata Shaiva preceptors as officiants at the pallipatais, and Sethuraman considers
that those interred in the pallipatais would have been initiated into the Pashupata
tradition. The tradition of building royal memorial temples, widespread in the
Deccan during this same period, was evidently more deeply-rooted and of a different
character from the practice found in the Tamil country, to the south and east
(see Wagoner 1996).
references
Tamil Inscriptions
Annual Reports on Indian Epigraphy (ARE). 1905–1978. Delhi: Manager of Publica-
tions. Transcripts of the inscriptions abstracted in the ARE were graciously
made available to me at the Office of the Chief Epigraphist, Archaeological
Survey of India, Mysore.
Cenkam Natukarkal (CN). 1972. Ed. Ira Nakacami. Cennai: Tamilnatu Aracu
Tolporul Ayvutturai.
Epigraphia Indica (EI). 1892–. Calcutta/Delhi: Director General, Archaeological
Survey of India.
Inscriptions (Texts) of the Pudukkottai State Arranged according to Dynasties (IPS).
1929. Pudukkottai.
Nannilam Kalvettukal (NK). 1979–1980. 3 vols. Ed. A. Patmavati. Cennai: Tamilnatu

Aracu Tolporul Ayvutturai.
South Indian Inscriptions (SII). 1891–1990. Vols. 2–26. Delhi: Director General,
Archaeological Survey of India.
Travancore Archaeological Series (TAS). 1910–1938. Vols. 1–8. Trivandrum.
Tamil Literature
Cilappatikaram, Ilanko Atikal. Tamil text. Tancavur: Tamilp Palkalaik Kalakam,
1985.
Cilappatikaram, translation, with an introduction and postscript, by R. Partha-
sarathy, The Cilappatikaram of Ilanko Atikal: An Epic of South India. New York:
Columbia University Press, 1993.
Manimekalai. Tamil text with commentary by Po. Ve. Comacuntaranar. Tirunelveli:
South India Saiva Siddhanta Works Publishing Society, 1971.
Manimekhalai (The Dancer with the Magic Bowl) by Merchant-Prince Shattan, transla-
tion by Alain Daniélou. New York: New Directions, 1989.
Tevaram, Tamil text. Edited by T. V. Gopal Iyer. 3 vols. Pondichéry: Institut français
de Indologie, 1984–1991.
Tevaram, translations by V. M. Subramanya Aiyar and Jean-Luc Chevillard, in
Digital Tevaram, courtesy Jean-Luc Chevillard.
Tiruppavai, Tamil text and English translation by Norman Cutler in his Consider Our
Vow. Madurai: Muttu Patippakam, 1979.
Tiruvempavai, Tamil text and English translation by G. U. Pope in The Tiruvacagam
or ‘‘Sacred Utterances’’ of the Tamil Poet, Saint, and Sage Manikka-vacagar,
103–116. Oxford: Clarendon Press, 1900 (reprinted New Delhi: Asian Education
Services, 1995).
Secondary Sources
Balbir, Nalini. 1994. ‘‘Jainism.’’ In Religion and Women, ed. Arvind Sharma, 121–138.
Filliozat, J. 1967. ‘‘L’Abandon de la vie par le sage et les suicides du criminel et du
héros dans la tradition indienne.’’ Arts Asiatiques 15: 65–88.
Findly, Ellison Banks. 2002. ‘‘The Housemistress at the Door: Vedic and Buddhist
Perspectives on the Mendicant Encounter.’’ In Jewels of Authority: Women, Text,
and the Hindu Tradition, ed. Laurie Patton, 13–31. New York: Oxford University
Press.
Jayadev, C. J. 1960. ‘‘Literary and Ethnographic References to the Tali and the
Tali Rite. ’’ In Archaeological Society of South India: Transactions for the Year
1959–60, 43–71. Madras: Archaeological Society of South India, Madras
Museum.
Kaimal, Padma. 1999. ‘‘The Problem of Portraiture in South India, circa 870–970
a.d.’’ Artibus Asiae 59, nos. 1–2: 59–133.
———. 2000. ‘‘The Problem of Portraiture in South India, circa 970–1000 a.d.’’
Artibus Asiae 60, no. 1: 139–179.
Mahalingam, T. V. 1985–. A Topographical List of the Inscriptions in the Tamil Nadu

and Kerala States. 8 volumes. New Delhi: Indian Council of Historical
Research.
McDaniel, June. 2003. Making Virtuous Daughters and Wives: An Introduction to
Women’s Brata Rituals in Bengali Folk Religion. Albany: State University of New
York Press.
McGee, Mary. 1991. ‘‘Desired Fruits: Motive and Intention in the Votive Rites of
Hindu Women.’’ In Roles and Rituals for Hindu Women, ed. Julia Leslie, 71–88.
Teaneck, NJ: Fairleigh Dickinson University Press.
Monius, Anne. Forthcoming. ‘‘The Manimekalai’s Buddhist Audience.’’ In Proceed-
ings of the Eighth International Conference-Seminar on Tamil Studies. Madras:
International Association of Tamil Research.
Ogura, Yasushi. 1999. ‘‘The Changing Concept of Kingship in the Cola Period: Royal
Temple Constructions, c. A.D. 850–1279.’’ In Kingship in Indian History, ed.
Noboru Karashima, 119–141. New Delhi: Manohar.
Orr, Leslie C. 1993. ‘‘Women of Medieval South India in Hindu Temple Ritual: Text
and Practice.’’ Annual Review of Women in World Religions 3: 107–141.
———. 1998a. ‘‘Jain and Hindu ‘Religious Women’ in Early Medieval Tamilnadu.’’
In Open Boundaries: Jain Communities and Cultures in Indian History, ed.
John E. Cort, 187–212. Albany: State University of New York Press.
———. 1998b. ‘‘Tracing Women’s Lives in Medieval Tamil Inscriptions.’’ Paper
presented at the Annual Meeting of the American Academy of Religion,
November 21–24, 1998, Orlando, Florida.
———. 1999. ‘‘Jain Worship in Medieval Tamilnadu.’’ In Approaches to Jaina Studies:
Philosophy, Logic, Rituals and Symbols, ed. N. K. Wagle and Olle Qvarnstrom,
250–274. Toronto: Centre for South Asian Studies, University of Toronto.
———. 2000a. Donors, Devotees and Daughters of God: Temple Women in Medieval
Tamilnadu. New York: Oxford University Press.
———. 2000b. ‘‘Women’s Wealth and Worship: Female Patronage of Hinduism,
Jainism, and Buddhism in Medieval Tamilnadu.’’ In Faces of the Feminine in
Ancient, Medieval, and Modern India, ed. Mandakranta Bose, 124–147. New
———. 2004. ‘‘Processions in the Medieval South Indian Temple: Sociology,
Sovereignty and Soteriology.’’ In South Indian Horizons: Felicitation Volume for
François Gros on the Occasion of His 70th Birthday, ed. Jean-Luc Chevillard and
Eva Wilden, 437–470. Pondichéry: Institut français de Pondichéry/Ecole
française d’Extrême-orient.
Pearson, Anne MacKenzie. 1996. Because It Gives Me Peace of Mind: Ritual Fasts
in the Religious Lives of Hindu Women. Albany: State University of New
York Press.
Peterson, I. V. 1988. ‘‘The Tie That Binds: Brothers and Sisters in North and South
India.’’ South Asian Social Scientist 4, no. 1: 25–52.
Raghotham, Venkata. 1995. ‘‘Kinship, Politics and Memory in Early Medieval Tamil
Country: A Study of the Funerary Shrines of the Cholas.’’ In Sri Nagabhinanda-
nam: Dr. M. S. Nagaraja Rao Festschrift, ed. L. K. Srinivasan and S. Nagaraju, 593–
608. Bangalore: Dr. M. S. Nagaraja Rao Felicitation Committee.
Rajasekhara, S. 1982. ‘‘Rastrakuta Hero-stones: A Study.’’ In Memorial Stones: A

Study of Their Origin, Significance and Variety, ed. S. Settar and Günther D.
Sontheimer, 227–230. Dharwar: Karnatak University.
Reynell, Josephine. 1987. ‘‘Prestige, Honour and the Family: Laywomen’s Religiosity
amongst the Svatambar Murtipujak Jains in Jaipur.’’ Bulletin d’Etudes Indiennes 5:
313–359.
Tamil Women, ed. Susan S. Wadley, 35–60. Syracuse, NY: Maxwell School of
Citizenship and Public Affairs, Syracuse University.
Richman, Paula. 1997. Extraordinary Child: Poems from a South Indian Devotional
Genre. Honolulu: University of Hawai’i Press.
Robinson, Sandra P. 1985. ‘‘Hindu Paradigms of Women: Images and Values.’’ In
Women, Religion, and Social Change, ed. Y. Y. Haddad and E. B. Findly, 181–215.
Sered, Susan Starr. 1992. Women as Ritual Experts: The Religious Lives of Elderly
Jewish Women in Jerusalem. New York: Oxford University Press.
Sethuraman, N. 1991. ‘‘Crusade against Guhai (Monastery).’’ Journal of the Epigraph-
ical Society of India (Purabhilekha Patrika) 17: 29–37.
Settar, S. 1982. ‘‘Memorial Stones in South India.’’ In Memorial Stones: A Study of Their
Origin, Significance and Variety, ed. S. Settar and Günther D. Sontheimer, 183–
197. Dharwar: Karnatak University.
Settar, S., and Ravi K. Korisettar. 1982. ‘‘Nisidhis in Karnataka: A Survey.’’ In Memorial
Stones: A Study of Their Origin, Significance and Variety, ed. S. Settar and Günther
D. Sontheimer, 283–293. Dharwar: Karnatak University.
Sontheimer, Günther D. 1982. ‘‘Hero and Sati-stones of Maharashtra.’’ In Memorial
Stones: A Study of Their Origin, Significance and Variety, ed. S. Settar and Günther
D. Sontheimer, 261–281. Dharwar: Karnatak University.
Soundara Rajan, K. V. 1982. ‘‘Origin and Spread of Memorial Stones in Tamil Nadu.’’
In Memorial Stones: A Study of Their Origin, Significance and Variety, ed. S. Settar
and Günther D. Sontheimer, 59–76. Dharwar: Karnatak University.
Srinivasan, K. R. 1960. Some Aspects of Religion as Revealed by Early Monuments and
Literature of the South. Madras: University of Madras.
Tewari, Laxmi G. 1991. A Splendor of Worship: Women’s Fasts, Rituals, Stories and
Art. New Delhi: Manohar.
Venkataraman, B. 1976. Temple Art under the Chola Queens. Faridabad: Thomson.
Vogel, J. P. 1930–1932. ‘‘The Head-offering to the Goddess in Pallava Sculpture.’’
Bulletin of the School of Oriental and African Studies 6: 539–543.
Wagoner, Phillip B. 1996. ‘‘From ‘Pampa’s Crossing’ to ‘The Place of Lord
Virupaksha’: Architecture, Cult and Patronage at Hampi before the Founding
of Vijayanagara.’’ In Vijayanagara: Progress of Research 1988–91, ed. D. V. Devaraj
and C. S. Patil, 141–174. Mysore: Directorate of Archaeology and Museums.
7
The Anatomy of Devotion:
The Life and Poetry
of Karaikkal Ammaiyar
Elaine Craddock
In the southernmost Indian state of Tamilnadu, Shaiva Siddhanta

developed over many centuries to become the dominant philosophi-
cal, theological, and ritual system associated with the god Shiva.
The tradition was systematized between the twelfth and fourteenth
centuries but draws its devotional perspectives from the stories and
hymns of the Nayanars, or ‘‘Leaders,’’ the sixty-three devotees of
Shiva who were canonized as saints in Cekkilar’s twelfth-century
hagiography, the Periya Puranam. Seven of these saints wrote poems
to Shiva between the sixth and ninth centuries. Along with the Alvars,
who sang to Vishnu, these poets were part of the bhakti, or devotional,
movements that began in South India and spread the emotional
worship of a personal god throughout the Indian subcontinent.
Karaikkal Ammaiyar, the ‘‘Mother from Karaikkal,’’ was the
first poet to write hymns to the god Shiva in Tamil, in the mid-
sixth century, when the boundaries between Shiva’s devotees and
competing groups were just starting to be articulated in a self-
conscious way. Speaking to God in one’s mother tongue, rather
than Sanskrit, was pivotal to the triumph of Hindu devotionalism
over the religions of Jainism and Buddhism, which reached the
apex of their popularity in South India during the fifth and sixth
centuries.1 Her powerful poetry is what Indira Peterson calls a
‘‘rhetoric of immediacy,’’ as it speaks to a particular community
defining itself in a context of competing religious allegiances
(1999, 165). Along with the hymns of the later saints, Karaikkal
Ammaiyar’s 143 poems envision a world where devotees can dwell
in perpetual bliss with Shiva, ridicule those who cannot see that Shiva
132 beyond domesticity
is the only Truth, and include elements of the sophisticated philosophy that
would be systematized as Shaiva Siddhanta centuries later.2
Her story remains popular in Tamilnadu,3 and it vividly encapsulates
notions of gender and devotion that are embedded in Tamil culture while
problematizing the connection between women’s ritual activity and the do-
mestic realm. Before Karaikkal Ammaiyar became a poet, she was the model
of a dutiful Hindu wife; her devotion to Shiva forms part of the continuum of
her domestic life, in which she faithfully serves both her husband and her
god. Yet it is her unswerving devotion to Shiva that ultimately ruptures her
orderly domestic world and drives her to restructure her life outside the or-
figure 7.1. Drawing of Karaikkal Ammaiyar by Haeli Colina.

the anatomy of devotion 133
dinary domestic realm. The ascetic path she embodies and praises forms a
critique of her previous life as a devoted wife. Karaikkal Ammaiyar’s story and
poetry portray a life lived as a ritual offering to Shiva as the only true life,
wherein rituals are not performed to achieve specific goals, as in the domestic
sphere, but where goal and ritual merge in perpetual devotion to Shiva alone:
text is practice, and practice is text. Her praise poems are the central ritual
activity of her life pursuing personal salvation, but they are also a record that
transcends her individual path and communicates her knowledge of Shiva to
the members of a newly emerging devotional community.
The devotional movements contained elements of social as well as reli-
gious reform, protesting Brahminical orthodoxy along with the heterodox faiths
of Buddhism and Jainism. But this revivalist Hinduism was rooted in the tem-
ple, which depended on royal patronage and the evolving sociopolitical alliance
between Brahmins and agriculturalists.4 So, although the devotional ideology
undercut caste and gender hierarchies in principle, in practical terms the pa-
triarchal boundaries remained. Statistically, women are not very visible among
the Tamil devotional movements: Andal is the only woman Vaishnava saint,
and out of the sixty-three Shaiva Nayanars, only three are women (Ramaswamy
1997, 120–121). However, the life and poetry of Karaikkal Ammaiyar, the only
woman poet among the Nayanars, reveals a fascinating portrait of the locali-
zation of a pan-Indian god and the potential space for women in this emerging
tradition.
Her Story
Karaikkal Ammaiyar was born in the sixth century into a well-to-do trading
family in the coastal town of Karaikkal; she was originally named Punitavati.5
In the well-known story about her, she was a beautiful girl who was married to
the rich merchant Paramatattan, to whom she was a faithful wife, although
this role proved to be at odds with her ardent devotion to Shiva. One day,
Paramatattan’s customer gave him two mangoes, which he told his wife to
serve him later for his midday meal. But before he returned home for lunch, a
Shaiva holy man came to the door for alms, so Punitavati gave him one of the
mangoes and some rice. When her husband came home, she gave him his
meal along with the remaining mango. He thought the mango was delicious
and asked for the other one. Punitavati went to the kitchen to pray to Shiva for
help; another mango appeared, which she served to her husband. This one was
so much more delicious than the first that her husband was suspicious and
asked his wife where she’d gotten it. She reluctantly told him, but he doubted
her story and asked her to repeat the miracle in his presence. Again, Punitavati
prayed to Shiva, and another mango appeared; her husband was terrified of her
power and fled without releasing her from her wifely duties.
He set up another household in another city, while Punitavati continued to

keep up his house and her appearance in anticipation of his return. Eventually,
her parents found out where he was and took their daughter to him. He and
his second wife and daughter, named Punitavati, fell at her feet in worship,
calling her a goddess. When Punitavati learned that her husband didn’t want
her as a wife any more, she begged Shiva to take away the beauty she no
longer needed and to give her a demon form. He granted her wish; she then
made a pilgrimage to the Himalayas, walking on her hands so as not to defile
god’s heavenly abode with her feet. Shiva was so moved by her devotion he
called her ‘‘Ammai,’’ or mother, and allowed her to join his troupe of ghouls,
his ganas, and to perpetually witness his dance at Tiruvalankatu, where she
lived as his adoring slave.6
This story upholds the traditional notion that a woman’s religious duty is
to be devoted to her husband and her home: Punitavati does not forsake her
wifely role until her husband has officially renounced her.7 Punitavati’s hus-
band acknowledges her as a goddess, but cannot accept her as a wife; Puni-
tavati’s dramatic role reversal begins with her husband falling at her feet.
Punitavati’s gift for sincere devotion is, paradoxically, what disrupts the house-
hold’s harmony; the boundaries of the domestic realm prove to be porous. Puni-
tavati can finally indulge the true focus of her unswerving devotion: the god
Shiva.
Ritualization and Asceticism
The story and the sculptures of Punitavati before her transformation stress
that she is beautiful (Dehejia 1988, 135–137), but as soon as her husband re-
leases her from her wifely role, she asks Shiva to take away her beauty, her
femininity and sexuality, and give her the demonic form she considers wor-
thy for worshipping him.8 Ammaiyar is also called ‘‘Pey,’’ or Demon; she iden-
tifies herself as the demon or ghoul from Karaikkal in several stanzas among
her four works.9
A female ghoul with withered breasts, bulging veins, hollow eyes,

white teeth and two fangs,
shriveled stomach, red hair, bony ankles, and elongated shins,
Stays in this cemetery, howling angrily.
This place where my Lord dances in the fire with a cool body,
His streaming hair flying in the eight directions,
is Tiruvalankatu. (Tiruvalankattu Mutta Tirupatikam, 1.1)10
In another poem she says:

The One who has kept another eye on His forehead,

Has made me understand a little of Him.
I am one of the ghouls among His good ganas.
Whether or not this grace lasts,
I don’t want anything else. (Arputattiruvantati, 86)
And in the last stanza of her longest work, she says:

By speaking this garland of antati venpa verses,
The words uttered by the Demon of Karaikkal, melting
with love,
Those who worship with unending love will reach Shiva,
And worship Him with great love. (Arputattiruvantati, 101)11
Many of the Nayanars performed dramatic feats of self-sacrifice out of

devotion to Shiva, but Karaikkal Ammaiyar’s transformation from ideal wife
to ideal demon devotee is particularly transgressive and serves to highlight the
rupture between the domestic world of ordinary rituals and a life lived entirely
as a ritual offering to Shiva. Karaikkal Ammaiyar’s renunciation of domestic
life to live in the cremation ground, praising Shiva, is an example of ‘‘ritual-
ization,’’ a term used by several theorists and defined by Catherine Bell as
ritual as lived practice, as a way of acting that uses diverse strategies to dif-
ferentiate meaningful, powerful, or sacred action from ordinary behavior (Bell
1992, 88–93). Ritualization creates a spatial/temporal environment in which
an individual embodies and enacts structures of personal and social meaning
within a perceived field of possibilities. Bell writes:
Ritualization always aligns one within a series of relationship[s]
linked to the ultimate sources of power. Whether ritual empowers or
disempowers one in some political sense, it always suggests the
ultimate coherence of a cosmos in which one takes a particular place.
This cosmos is experienced as a chain of states or an order of exis-
tence that places one securely in a field of action and in align-
ment with the ultimate goals of all action. (Bell 1992, 141)
Through her poetry, Karaikkal Ammaiyar delineates the only realm of ac-
tion that has ultimate meaning: sublimating herself as one of Shiva’s adoring,
ghoulish attendants. Her poetry expresses in literary Tamil a life of perpetual,
spontaneous worship of Shiva in which all thought and action fuse in a ritual
offering of pure awareness of God. Unlike female devotional poets who relate
to God as their beloved, such as Andal relates to Krishna, Karaikkal Ammaiyar
does not violate rules regarding chastity. In one sense, her life has moved along
a continuum of devotion to others, with the others simply changing in im-
portance.12 But ultimately, Karaikkal Ammaiyar rejects the entire social and
domestic world of rules and obligations to pursue personal salvation, relocat-

ing her sphere of activity on the periphery of the social world (Mahalakshmi
2000, 33). This is not the classic renunciation called sannyasa that is typically
undertaken at the end of life after fulfilling one’s obligation to a family; the
severe asceticism she undertakes in the prime of her life implicitly critiques
the location of women in a domestic space of family relationships.13 Karaikkal
Ammaiyar’s poetry dismantles the paradigm of human order and duty rooted
in the household not by focusing on gender roles, but by extolling devotion to
Shiva in a community of devotees in which gender is irrelevant.
Localizing God
Karaikkal Ammaiyar descends from her vision of Shiva and Parvati on Mt.
Kailash to spend the rest of her life singing to Shiva as he dances in the cre-
mation ground at Tiruvalankatu, or ‘‘Sacred Banyan Tree Forest,’’ where he per-
forms his fierce dance called kalikatandava or urdhvatandava, in which he defeats
Kali by lifting his leg up to the sky (in this case, his left leg).14 The Tiruvalankatu
temple, north of Chennai, must have originally been at the base of a banyan tree
(Dorai Rangaswamy 1990, 825). But now, in the Tiruvalankatu temple, Shiva
dances in the Ratnasabhai, one of the five sabhais, or temple halls, in Tamilnadu
associated with Shiva Nataraj, Shiva as Lord of the Dance.15 Karaikkal Ammaiyar
is shown at the feet of Shiva, playing cymbals and singing her praises to him.
Her closed samadhi or shrine is behind this main image, not far from the giant
banyan that is the temple tree (sthala vriksham); some of her verses are inscribed
inside the temple. There is a separate shrine where Kali is seen dancing.
In the cemetery where you hear crackling noises
and the white pearls fall out of the tall bamboo,
The ghouls with frizzy hair and drooping bodies,
shouting with wide-open mouths,
Come together and feast on the corpses.
In the big, threatening cremation ground,
When The Lord dances,
The Daughter of the Mountain watches Him,
In astonishment. (Tiruvalankattu Mutta
Tirupatikam, 2.8)
Karaikkal Ammaiyar’s poetry reveals an early stage in the process of the
transcendent, pan-Indian deity Shiva taking up residence in a landscape dense
with local, and sometimes competing, religious and cultural meaning. The po-
etry is filled with vivid images of Shiva as the heroic god whose grace rescues
his devotees from the sorrows of the world, sometimes conveyed through
detailed descriptions and at other times through metonymic references that
imply that Karaikkal Ammaiyar’s audience was at least minimally acquainted

with Shiva’s stories. These rhetorical strategies help to create and define a
nascent community dedicated to the worship of Shiva.16
I thought of only One.
I was focused on only One.
I kept only One inside my heart.
Look at this One!
It is He who has Ganga on His head,
A moonbeam in His hair,
A radiant flame in His beautiful hand.
I have become His slave. (Arputattiruvantati, 11)
Her poetry conveys to her audience a conception of Shiva drawn from the
Sanskritic culture that began to permeate South India in the early centuries of
the common era. She refers to several of Shiva’s most famous deeds and man-
ifestations: his heroic destruction of the Three Cities of the Demons; his burn-
ing of Kama; his crushing Ravana with his big toe when Ravana tries to lift Mt.
Kailash; his swallowing the poison during the churning of the ocean so that his
throat is blue; his rescuing of Markandeya from death; his killing of the ele-
phant demon; his manifestation as the fiery linga; his terrifying Bhairava form,
forced to wander as a beggar in penance for Brahmanicide; and most centrally
for Karaikkal Ammaiyar, dancing with Kali in the cremation ground. She refers
to Shiva as the Knower of the Vedas, indeed as the Vedas.17 She thus builds a
detailed iconography that unites the emerging Shaiva community and links
it to the broader Indian traditions surrounding Shiva. It seems clear that even
at this early date Karaikkal Ammaiyar and the emerging community of wor-
shippers were seeing iconographic images of Shiva in temples, whether those
temples were actual structures or open-air shrines.18 However, unlike the later
poets, she does not talk about temple worship and in one verse even criticizes
those who expect to see Shiva through empty ritual.19 Karaikkal Ammaiyar
does not praise the Tamil landscape nor the Tamil language as the later poets
do. Karaikkal Ammaiyar praises Shiva in the cremation ground at Tiruvalan-
katu. Through her powerful poetry, Karaikkal Ammaiyar reveals that the hor-
rific cremation ground is really the cosmos, and the terrifying form of Shiva
performing his dance of destruction is really the most sublime and blissful
experience of the lord. She makes the terrible beautiful (Dorai Rangaswamy
1990, 441, 387) and leads the devotee beyond the limits of ordinary awareness
into a transcendent knowledge of Shiva as Truth, as the dancer who dances the
world and into the heart of his adoring devotee.20
The Tamil word katu means forest, jungle, or desert, but it also means
cremation or burial ground, as well as boundary or limit.21 In the life and po-
etry of Karaikkal Ammaiyar, the cremation ground encompasses the notion of
the forest as the opposite of civilization (Tamil natu, city), in addition to the
pollution, fear of death, and ghoulish forces traditionally associated with a

burning ground. Her poetry is connected to the Shaiva tradition that develops
after her, but it also resonates with the earlier Cankam puram poetry that praises
heroes in war and provides detailed descriptions of the battlefield, gruesome
places of death and sacrifice, where demons feast on the corpses lying on the
field and dance, garlanded with intestines. Korravai, the Tamil goddess of vic-
tory, is described in Cankam poetry as surrounded by demonesses, eating
flesh and dancing with dead bodies on the battlefield (Hart 1975, 31–41; Ra-
maswamy 1997, 129–130; Mahalakshmi 2000, 24–27). Indeed, Karaikkal Am-
maiyar uses a trope of the heroic warrior to describe Shiva as the conqueror of
death for the devotee and as the ultimate sovereign of the universe:
Ghouls with flaming mouths and rolling, fiery eyes,
Going around, doing the tunankai dance,
Running and dancing in the terrifying forest,
Draw out a burning corpse from the fire and eat the flesh.
The place where our Lord raises His leg in the vattanai posture
with the kalal jangling and the anklets tinkling,
Dancing so that the fire in His hand spreads everywhere and
His hair whips around,
Is Tiruvalankatu. (Tiruvalankattu Mutta Tirupatikam, 1.7)
Here, Shiva is dancing with Kali and takes the vattanai posture with his
leg in the air—his special posture at Tiruvalankatu—to defeat her. The poem
uses two sets of ankle bracelets to signify the divine competition: Kali’s an-
klets jingle as she dances, and Shiva’s kalal, or hero’s victory anklets, jangle as
he dances so dynamically that his hair whips around in his frenzy, dancing as
the god of destruction, but also as the conqueror of death (Shulman 1980,
213–221). The tunankai dance is associated with Korravai; it was also danced at
festivals,22 and in fact can mean ‘‘festival.’’ It is a type of village dance in which
the arms are bent and struck against the sides of the body. Tunankai is also
a kind of pey, or demon.23 The Tamil imagery that Karaikkal Ammaiyar uses
in her poetry serves to localize the god in a familiar landscape and to connect
local sacred forces to the Shaiva pantheon. During the post-Cankam period
(approximately 300–600 c.e.), Korravai is increasingly associated with Durga
and Kali, and therefore with Shiva. It seems that Karaikkal Ammaiyar is
consciously associating the powerful forces of the demons occupying Tiruval-
ankatu with the transcendent god dancing in the cremation ground. By be-
coming one of Shiva’s ganas, or ghouls, she assumes a kind of power familiar
in the early Tamil world and connects the indigenous demon tradition with
one of the central myths of Shiva (Mahalakshmi 2000, 29). Karaikkal Am-
maiyar is also linked closely to Tiruvalankatu because of Nili, the ancient and
fierce goddess located in the nearby town of Palaiyanur.24 Despite the strong
goddess tradition in the area, however, Karaikkal Ammaiyar remains focused
on Shiva; Parvati appears in several of her verses, but Ammaiyar’s single-

minded devotion never wavers from the lord.25
Defining a Path
The cremation ground is the stage for Shiva’s dance of life and death and
salvation. It is also the space in the heart of the devotee where the ego is burned
up as she surrenders to him.26 Through vivid and jarring imagery, Karaikkal
Ammaiyar reveals that the terrifying place of death is really the beautiful and
blissful abode of the lord and the sacred grove of liberation from this world.
The picacu, wearing a white skull garland tied tightly,

Swallowed up the congealed fat.
Having named its child Kali,
Bringing her up with comfort,
She wiped the dust off the child, suckled it, then went away.
The child, not seeing the mother returning, cried itself to sleep.
The place where our Lord dances in the cemetery is
Tiruvalankatu. (Tiruvalankattu Mutta Tirupatikam, 1.5)
Karaikkal Ammaiyar’s close attention to mundane activities confronts the

devotee’s habitual understanding of the cremation ground as a threatening, mar-
ginal, ‘‘other’’ place. Here, the picacu ghoul is a gentle mother, caring for a child.
The intimate scene plays out in the same space as Shiva’s dance, breaking down
the barrier between domestic space and the place of death. Karaikkal Ammaiyar
attempts to shatter the illusions of ordinary awareness and show that ultimately,
the cremation ground is a state of mind, where the true devotees who meditate on
Shiva overcome their fear of death and experience him as the beautiful lord, with
his upraised dancing foot pointing to liberation through him.
If, with wisdom,

And without ridiculing His body that is garlanded with bones,
They praise the One who wears a moon on His long matted
hair, and
Who conquered the three great cities of the powerful asuras
who did not respect Him;
They will not be born here in this world in a body with bones.
(Arputattiruvantati, 37)
As we see in this poem, bones are a central liberation motif in Karaikkal

Ammaiyar’s poetry. Shiva ornaments himself with a garland of bones he finds
in the cremation ground, the bones of everyone and anyone; he does not dis-
criminate, just as he responds to all devotees.27 But devotees must get beyond
terror and revulsion at his horrific form to see that he is the conqueror of death.
If they are able to reach this point, Shiva will liberate them from life in this world
‘‘in a body with bones.’’ Karaikkal Ammaiyar is an emaciated skeleton of a figure,
an assemblage of bones at home in the cremation ground but liberated by it.
The other major symbol of liberation is fire. The cremation ground burns
with funeral fires, signifying the end for all those whose ignorance clouds
their vision of the lord. Shiva’s terrifying dance takes place in fire, and he holds
fire in his hand. Whether that fire is terrifying, signaling death and destruc-
tion, or illuminating depends on the devotee’s consciousness. The center of
this crowded, gruesome scene is Shiva dancing in the fire of destruction, yet
his yogi body is cool, not affected by the fire he creates. Karaikkal Ammaiyar
promotes a vision of Shiva as the beautiful embodiment of the rhythm of life,
burning away our illusions with the fire in his hand.
If you consider the One who has the complexion of the red
rays of the setting sun, and of a smoldering fire,
And whose matted hair hangs down,
You would say that to those who have surrendered to Him,
He shines like a golden flame;
But to those who move away without taking refuge in Him,
He has the nature of leaping fire. (Arputattiruvantati, 82)
Creating Community
The devotional movement denies caste or gender privileges, but ultimately the
only real egalitarianism is the spiritual equality of the worshippers of Shiva,
in contrast to other religious groups, including devotees who follow a different
path to Shiva (Zvelebil 1973, 194). Karaikkal Ammaiyar is pursuing her own
path to salvation, but at the same time she is working to create a community
of devotees who also understand that Shiva is the ultimate Truth. In addition to
sharing a Vedic, mythic understanding of Shiva, the Tamil Shaiva community
was forged partly by the harsh rhetoric of the Tamil Shaiva saints against the
Buddhists and Jains in particular. Peterson argues that ‘‘the negative represen-
tation of Jains was an important part of a process of self-definition and consoli-
dation of power for the Tamil Shaiva sect. . . . Jains were not only a threatening
rival group, but a very useful foil against which to establish the superiority of the
Shaiva religion’’ (1999, 164). Karaikkal Ammaiyar does not refer to any group by
name, lumping together as ‘‘others’’ the people against whom she is defining her
spiritual path.28 But it is clear that she is referring to non-Vedic groups. She says:
Look!
Having become a slave to the beautiful feet of the
One whose red matted hair has the waves of Ganga,
We have realized Him through scriptures,

We have become suitable for this life and for the other world.
Why do others gossip about us behind our backs?
Understand us. (Arputattiruvantati, 91)
In other poems, she is even more scathing:
Ignorant mind,
Worship the feet of the devotees, again and again
Focusing on them, and praising them with words.
Leave that group of people who do not think about
The One who wears a moon as a small garland,
Which no one else wears. (Arputattiruvantati, 40)
Oh! You pitiable people
Who are without wisdom.
It is an easy way to live,
Thinking of our Lord all the time,
Our Father with the gleaming throat,
Who wanders around,
Wearing a snake. (Arputattiruvantati, 46)
His greatness is such that it is not known by others.
[But] [o]thers know he is the great consciousness.
Our Lord, wearing the bones of others,
Happily dances along with the strong ghouls
In the fire at night. (Arputattiruvantati, 30)
In this poem, the first ‘‘others’’ refers again to the usual suspects who do
not understand him. The second ‘‘others’’ refers to his devotees, who do
realize who he is. The third, ‘‘the bones of others,’’ refers to ‘‘just anyone,’’
conveying that Shiva does not discriminate against anyone concerning whose
bones he wears; he treats all equally, is God to all equally, and is open to all
devotees. The many references to these ‘‘others’’ who do not understand Shiva
reveal how influential these heterodox communities were. The lumping to-
gether of all these groups probably reveals that the Jains were not the only
powerful voice that the Shaivas had to resist. But, perhaps most important at
this early period, Karaikkal Ammaiyar’s refusal to specify any one community
conveys the key point: everyone who doesn’t follow the right path to Shiva is
equally ignorant. She urges followers to ‘‘worship the feet of the devotees,’’
requiring that worthy individuals be identified as belonging to the emerging
community. Serving Shiva by serving his devotees remains an important part
of the tradition (Peterson 1989, 41–47).
All those other people who do not understand that He

is the real truth, Have seen only His ghoul form:
His lotus-like body smeared with ash and garlanded with bones.
See that they ridicule Him? (Arputattiruvantati, 29)
Although Karaikkal Ammaiyar doesn’t transgress the boundaries of chastity

for a woman, she does transgress other boundaries, evoking a tantric orientation
to the world.29 Karaikkal Ammaiyar’s transgressive behavior does not go to the
extremes of some Shaiva worshippers of her milieu, but she engages in behavior
that turns ordinary categories of personal and social perception upside down in
order to force a transcendent spiritual awareness, including an understanding of
the self as rooted in divine power. This perspective is intensified by the Tamil
view of the world as pervaded by powers, including the forces present in places of
death, such as the cremation ground. Karaikkal Ammaiyar’s extreme ascetic
emphasis reveals the early, broad roots of the Tamil Shaiva tradition before it is
systematized into a temple-based orthodoxy. This asceticism is rooted in a milieu
of multiple traditions, drawing particularly from the earliest Shaiva sects, often
grouped together as the Pashupatas, whose followers imitated Shiva in his ter-
rible Bhairava form, often in the cremation ground. These devotees enacted
Shiva’s penance for Brahmanicide, which he incurred by cutting off Brahma’s
fifth head when he tried to sleep with his own daughter, and because of which
Shiva is known as the beggar Bhiksayatana and as Kapalin the skull-bearer.30
Later poets refer to Pashupata ascetics worshipping at Shiva’s shrines, but the
emerging orthodox tradition rejected their devotional mode. This is probably
one major reason that Karaikkal Ammaiyar’s poetry is not regularly sung in
Shiva temples today, in contrast to the Tevaram poets, whose hymns are regu-
larly sung. Perhaps because the nearby town of Tiruvorriyur is known to have
had a significant Pashupata presence at this time, Karaikkal Ammaiyar was
drawn to a more extreme asceticism than the later Nayanars embodied (Krishna
Murthy 1985). In one poem, she takes the point of view of a male ascetic:
My heart!
Give up your bondage, your wife and children.
Saying that you take refuge here at His feet,
Think of Him and worship. (Tiruvirattai Manimalai, 13)
And elsewhere she says:
Not following a false path ruled by the five senses,

We have achieved merits,
Because of the Lord’s love for His slaves. (Tiruvirattai
Manimalai, 16)
Now we have been elevated;
We have reached God’s feet.
Now we do not have any troubles.
O my heart!
You will see that we have now crossed over the turbulent
sea of inescapable births
That causes an ocean of karma. (Arputattiruvantati, 16)
The intimacy of Ammaiyar’s relationship to Shiva combines with her living

beyond social norms to create a powerful model of the devotional path. A wife is
expected to serve her husband and sacrifice herself for his welfare; Karaikkal
Ammaiyar relocates this behavior in the cremation ground, where she serves
Shiva as her lord. Through Karaikkal Ammaiyar’s life and poetry, a new form of
Shiva takes shape (Ramaswamy 1997, 129). Karaikkal Ammaiyar’s poetry does
not describe the ecstatic fusing with the divine seen in other devotional poetry
(Hardy 1983, 306–307). The goal of Shaivism is to become liberated from the
cycle of rebirth by becoming like Shiva or, at least, more like him.31 Karaikkal
Ammaiyar exhorts devotees to focus all of their energies in constant meditation
on Shiva as the discipline that leads to salvation. Her poetry is intellectual,
focusing on the experiential understanding of Shiva as knowledge and the
ultimate truth gained through meditation, an important perspective in the
development of the orthodox Shaiva Siddhanta tradition.
He is the one who knows.

He is the one who makes us know.
He is the one who knows because He is knowledge itself.
He is the truth that is to be known.
He is the sun and moon, the earth, sky,
and all the other elements. (Arputatiruvantati, 20)
In the ritualization of her life, Karaikkal Ammaiyar renounces her life in the
domestic sphere, becomes one of Shiva’s ghoul attendants, and dwells in per-
petual bliss with him. Her radical transformation from a lovely, dutiful wife into
an emaciated, frightening demon reveals that knowing Shiva requires the dev-
otee to transcend ordinary human awareness and to see that the terrifying cre-
mation ground is really the beautiful place of liberation. Her poetry urges people
to give up a life rooted in family relationships and bounded by conventional
rituals and goals, and instead to live their lives as ritual offerings to Shiva.
Through vivid descriptions of the beautiful lord Shiva, multiple mythic refer-
ences to his deeds, and regular references to a host of ignorant ‘‘others,’’ Ka-
raikkal Ammaiyar delineates a spiritual path and creates a community that
centers on a self-conscious understanding of Shiva as the ultimate Truth and the
only path to liberation.
notes
1. On the interplay of multiple religions in medieval South India, see in addi-
tion, Sastri 1975, 3, 130, 135, 137; Davis 1999; Peterson 1999; and Monius 2001.
The great Tamil Jain epic Cilappatikaram and the Tamil Buddhist epic Manimekalai
were composed during this period.
2. See Monius 2001, especially ch. 1, for a discussion of imagined and imagi-
nary community.
3. In addition to written versions, her story is told in a Tamil film, Karaikkal
Ammaiyar.
4. See Stein 1980; Davis 1999; Peterson 1999; Mahalakshmi 2000, 17–19;
Ramaswamy 1997, 118–121. On the development of the Shaiva Siddhanta tradition,
which is beyond the scope of this chapter, see Davis 2000 and Prentiss 1999.
Shaiva Siddhanta was originally a pan-Indian tradition that for a number of reasons
survived and thrived in Tamilnadu. On the bhakti movements in South India, see
Peterson 1989, Hardy 1983, and Champakalakshmi 2004.
5. The story was originally told in Cekkilar’s Periya Puranam, which is retold
in many versions. I am also drawing here on the Tiruvalankattu Talavaralarum, the
temple history book sold at Tiruvalankatu.
6. Shiva at Tiruvalankatu is called Vatanaroshvara, and the temple is called
Vatanaroshvara Swamy Tirukoyil. On the back cover of the temple’s official history
is an upside-down picture of Mt. Kailash, which is the view Karaikkal Ammaiyar
had as she walked down from Mt. Kailash to Tiruvalankatu on her hands. The last
part of the temple book contains the story of her life from Cekkilar’s hagiography
and a few verses of her poetry. There is a sculpture in the temple of Karaikkal Am-
maiyar climbing on her hands to see Shiva and Parvati. This is the only instance
when Shiva called anyone ‘‘mother’’ (Periya Puranam, 539).
Karaikkal Ammaiyar is most closely associated with Tiruvalankatu, but there is
a yearly mango festival in her honor during the month of Ani (June–July) in her
hometown of Karaikkal; see Ramaswamy 1997, 132.
7. There is a large literature on women’s roles in Hindu life; see, for instance,
Harlan and Courtright 1995.
8. Mahalakshmi (2000) discusses the notion, raised by other scholars, that Kar-
aikkal Ammaiyar renounced her body as a reaction to her husband abandoning her. She
maintains that Karaikkal Ammaiyar’s evocation of the demonic figure ‘‘suggests a de-
liberate evocation of certain symbols, and the denial of still others’’ (33).
9. These four works are Arputattiruvantati with 101 venpa verses; Tiruvirattai
Manimalai with 20 stanzas in venpa and kattalaik kalitturai; and the two patikams
called Tiruvalankattu Mutta Tirupatikankal with 11 verses each. She probably
wrote the first prabandha literature, and invented the antati form. See especially
Zvelebil 1975, 136–137. Zvelebil states that Karaikkal Ammaiyar likely introduced
the kattalaik kalitturai form. Accounts of her life suggest she wrote the first two
works before going to Mt. Kailash, then she wrote the Tiruvalankatu poems
when she arrived there; see Dorai Rangaswamy 1990, 972–973.
10. All translations are my own, which I made in partnership with
Dr. R. Vijayalakshmy, professor emeritus at the International Institute of Tamil
Studies. This text is from Tiruvalankattu Talavaralarum Tiruppatikankalum.
11. The antati is a poetic structure in which the last syllable or word of a verse
is repeated as the first syllable or word of the next verse, forming a kind of chain.
Venpa is a poetic meter that consists of two, three, or four lines per verse. The last
line of the verse contains three feet (Tamil cir); the other lines contain four feet.
See Zvelebil 1975, 135–137, 278–280; and Hart 1975, 197–199.
12. Caroline Walker Bynum’s works on medieval European Christianity have
addressed women’s spiritual lives as continuous with their domestic lives and their
explorations of the connections between femaleness and physicality. See, for example,
Bynum 1987.
13. See Denton 1992 on women’s asceticism.
14. For a discussion of whether the urdhvatandava pose requires the right leg,
see Dorai Rangaswamy 1990, 456.
15. The others, represented and labeled at the Tiruvalankatu temple, are
Chidambaram, Madurai, Tirunelveli, and Kutralam. Dorai Rangaswamy lists them
as Kanchipuram, Tiruceenkattangudi, Tenkasi, and Taramangala (1990, 452).
16. Several scholars have written about how a text can construct community.
For South India in particular, see Monius 2001; Richman 1988.
17. Such as in Arputattiruvantati, 15.
18. Dorai Rangaswamy 1990, 3–18; Hardy 1983, 202–213; Prentiss 1999, 51;
Ayyar 1974, 211; Krishna Murthy 1985, 6–14, 42–47; Peterson 1982, 72.
19. See Arputattiruvantati, 17.
20. See Narayanan 2003, especially 500–501.
21. Tamil Lexicon 1982, 855. See also Hart and Heifetz 1999, 362–363.
22. Hart 1975, 23, 45, 142; Hart and Heifetz 1999, 166.
23. Tamil Lexicon 1982, 1963.
24. The story of Nili is told in the epic Cilappatikaram, Canto 23. Nili’s story is
included in the temple’s history, the Tiruvalankatu Talavaralaru, 178–196; her story
continues to be important in the area. See Mahalakshmi 2000, 33–40; Shulman
1980, 194–197, 213–221; and Peterson 1989, 203.
25. This pure focus on Shiva is contrasted in the film about Karaikkal Ammai-
yar in which Parvati tells her she must first worship the goddess in order to get to
Shiva.
26. See Coomaraswamy 1999, 89.
27. See also Arputattiruvantati, 30.
28. In Shaiva poetry, ‘‘others’’ are sometimes referred to as ‘‘the six faiths’’ (aka-
camayam), which can refer to six sects of Shaiva Siddhanta, or six sectarian groups with
different cultic deities. Tamil Lexicon 1982, 8; Peterson 1989, 132–133, n.52.
29. Tantra is a vast subject that cannot be taken up in detail here. There are
many perspectives on and definitions of tantra, as the scholarship shows, including
elements of yoga and meditation. See Denton 1992, 225–227; Brooks 1990, 3–6;
Padoux 2002. Among other things, Padoux addresses the porous boundary between
tantra and bhakti. The later Shaiva Siddhanta tradition uses the tantric texts called
Agamas as the ritual manuals in temple worship; see Davis 2000; Prentiss 1999;
Filliozat 1983.
30. Flood 1996, 154–173; Bhandarkar 1983, 145ff.; Dorai Rangaswamy 1990,
392–393, 400, 1265.
31. Davis 1999, 221.
references
Ayyar, C. V. Narayana. 1974. Origin and Early History of Saivism in South India.
Madras: University of Madras.
Bell, Catherine. 1992. Ritual Theory, Ritual Practice. New York: Oxford University Press.
Bhandarkar, Ramkrishna Gopal. 1983. Vaisnavism, Saivism and Minor Religious
Systems. New Delhi: Asian Educational Services.
Brooks, Douglas Renfrew. 1990. The Secret of the Three Cities: An Introduction to
Hindu Sakta Tantrism. Chicago: University of Chicago Press.
Bynum, Caroline Walker. 1987. Holy Feast, Holy Fast: The Religious Significance of
Food to Medieval Women. Berkeley: University of California Press.
Champakalashmi, R. 2004. ‘‘From Devotion and Dissent to Dominance: The
Bhakti of the Tamil Alvars and Nayanars.’’ In Religious Movements in South Asia
600–1800, ed. David N. Lorenzen, 47–80. New Delhi: Oxford University Press.
Cilappatikaram. 1997. Trans. V. R. Ramachandra Dikshitar. Chennai: International
Institute of Tamil Studies.
Coomaraswamy, Ananda. 1999 [1918]. The Dance of Shiva. New Delhi: Munshiram
Manoharlal.
Davis, Richard H. 1999. ‘‘The Story of the Disappearing Jains.’’ In Open Boundaries:
Jain Communities and Cultures in Indian History, ed. John E. Cort, 213–224. Delhi:
Sri Satguru.
———. 2000. Worshiping Siva in Medieval India: Ritual in an Oscillating Universe.
Delhi: Motilal Banarsidass.
Dehejia, Vidya. 1988. Slaves of the Lord: The Path of the Tamil Saints. New Delhi:
Munshiram Manoharlal.
Denton, Lynn Teskey. 1992. ‘‘Varieties of Hindu Female Asceticism.’’ In Roles
and Rituals for Hindu Women, ed. Julia Leslie, 211–231. Delhi: Motilal
Banarsidass.
Dorai Rangaswamy, M. A. 1990 [1958–1959]. The Religion and Philosophy of Tevaram.
Madras: University of Madras.
Filliozat, Jean. 1983. ‘‘The Role of the Saivagamas in the Saiva Ritual System.’’ In Ex-
periencing Siva: Encounters with a Hindu Deity, ed. Fred W. Clothey and
J. Bruce Long, 81–101. Columbia, MO: South Asia Books.
Flood, Gavin. 1996. An Introduction to Hinduism. Cambridge: Cambridge University
Press.
Hardy, Friedhelm. 1983. Viraha-Bhakti: The Early History of Krsna Devotion in South
India. New Delhi: Oxford University Press.
Harlan, Lindsey, and Paul B. Courtright. 1995. ‘‘Introduction: On Hindu Marriage
and Its Margins.’’ In From the Margins of Hindu Marriage, ed. Lindsey Harlan
and Paul B. Courtright, 3–18. New York: Oxford University Press.
Hart, George L. 1975. The Poems of Ancient Tamil. Berkeley: University of California
Press.
Hart, George L., and Hank Heifetz, trans. and eds. 1999. The Four Hundred Songs
of War and Wisdom. New York: Columbia University Press.
Karaikkal Ammaiyar. 1973. Directed by A. P. Nagarajan. EVR Films.
Karaikkalammaiyar Pirapantankal. 1961. Commentary by Shri Arumukattampiran.
Tiruvavatuturai: Tiruvavatuturai Math.
Krishna Murthy, C. 1985. Saiva Art and Architecture in South India. Delhi: Sundeep
Prakashan.
Lorenzen, David N. 1972. The Kapalikas and Kalamukhas: Two Lost Saivite Sects. Delhi:
Motilal Banarsidass.
———. 2002. ‘‘Early Evidence for Tantric Religion.’’ In The Roots of Tantra, ed.
Katherine Anne Harper and Robert L. Brown, 25–36. Albany: State University
of New York Press.
Mahalakshmi, R. 2000. ‘‘Outside the Norm, within the Tradition: Karaikkal Am-
maiyar and the Ideology of Tamil Bhakti.’’ Studies in History 16, no. 1, n.s.: 17–40.
New Delhi: Sage.
Monius, Anne E. 2001. Imagining a Place for Buddhism. Oxford: Oxford University
Press.
Narayanan, Vasudha. 2003. ‘‘Embodied Cosmologies: Sights of Piety, Sites of Power.’’
Journal of the American Academy of Religion 71, no. 3 (September): 495–520.
Padoux, Andre. 2002. ‘‘What Do We Mean by Tantrism?’’ In The Roots of Tantra, ed.
Katherine Anne Harper and Robert L. Brown, 17–24. Albany: State University
of New York Press.
Periya Puranam: A Tamil Classic on the Great Saiva Saints of South India, by Sekkizhaar.
1985. Trans. G. Vanmikathan. Madras: Sri Ramakrishna Math.
Peterson, Indira V. 1982. ‘‘Singing of a Place: Pilgrimage as Metaphor and Motif
in the Tevaram Songs of the Tamil Saivite Saints.’’ Journal of the American
Oriental Society 102, no. 1: 69–90.
———. 1989. Poems to Siva. Princeton, NJ: Princeton University Press.
———. 1999. ‘‘Sramanas against the Tamil Way.’’ In Open Boundaries: Jain Commu-
nities and Cultures in Indian History, ed. John E. Cort, 163–185. Delhi: Sri Satguru.
Prentiss, Karen Pechilis. 1999. The Embodiment of Bhakti. New York: Oxford
University Press.
Ramaswamy, Vijaya. 1997. Walking Naked: Women, Society, Spirituality in South India.
Shimla: Indian Institute of Advanced Study.
Richman, Paula. 1988. Women, Branch Stories, and Religious Rhetoric in a Tamil
Buddhist Text. Syracuse, NY: Maxwell School of Citizenship and Public Affairs,
Syracuse University.
Sastri, K. A. Nilakanta. 1975. A History of South India. New Delhi: Oxford University
Press.
Shulman, David Dean. 1980. Tamil Temple Myths: Sacrifice and Divine Marriage in
the South Indian Saiva Tradition. Princeton, NJ: Princeton University Press.
Stein, Burton. 1980. Peasant State and Society in Medieval South India. New Delhi:
Tamil Lexicon. 1982. Madras: University of Madras.
Tiruvalankattu Talavaralarum Tiruppatikankalum. 1998. Tiruttani: Cuppiramaniya
Swamy Tirukoyil.
Tiruvalankatu Talavaralaru: Alankattil Antamura Nimirntatum Maniyampalakkuttan.
n.d. Chennai: Kavuniyan.
Zvelebil, Kamil. 1973. The Smile of Murugan: On Tamil Literature in South India.
Leiden: Brill.
———. 1975. Tamil Literature. Leiden: Brill.
8
The Play of the Mother:
Possession and Power
in Hindu Women’s
Goddess Rituals
Kathleen M. Erndl
Context: Shakti ‘‘Plays’’
This chapter explores Hindu women’s religious empowerment in

connection with goddess possession rituals in the Kangra Valley
area of Himachal Pradesh. In Kangra, as in many other regions of
India, it is not uncommon for a goddess to ‘‘play’’ in the body of a
woman, or as it is also stated idiomatically, for a woman to ‘‘play,’’
that is, to embody the shakti (power, creative energy, consciousness)
of a goddess, to speak with her voice, and to heal and prophesy in
her community. Those women who experience this shakti deeply
and on a continuous and sustained basis may attain the status of a
mataji, that is, a respected mother, a religious leader, a holy woman,
a goddess herself. Divine possession as a form of religious expres-
sion is interconnected with such practices as pilgrimage to temples,
puja (image worship), recitation of sacred texts, fasting, and medita-
tion, which comprise the bhakti- (devotion) and tantra- (esotericism)
oriented religious complex of Shaktism or goddess worship in the
region. In the Kangra area, there are several famous goddess temples,
such as Jvala Mukhi and Camunda Devi (Erndl 1993), and hun-
dreds of smaller, more localized goddess shrines. Although the wor-
ship of male deities such as Krishna, Shiva, and Baba Balak Nath is
also prevalent, the worship of goddesses is so pervasive that in the
Kangra context, Shaktism can be considered practically congruent
with general popular Hinduism rather than a sectarian designation.
The material in this chapter is taken from a larger study, based on fieldwork
conducted in Kangra with holy women, known as matajis, who become pos-
sessed, and their devotees, both female and male.
In the Hindi, Panjabi, and Kangri languages in which this research was
conducted, ‘‘playing’’ (khelna, khedna) refers to the creative, dynamic, and awe-
some activity of both the goddess and the woman who is possessed by her.
Similarly, words meaning ‘‘mother’’ (ma, mata, mataji) refer not only to ordi-
nary mothers but also both to the goddess and to the woman she possesses. In
other words, the woman both plays and the goddess plays in her; the woman is
both the agent and instrument of the playing. Thus, the title of this chapter,
‘‘The Play of the Mother,’’ is a play on words suggestive of the ambiguous flu-
idity between human and divine identities and powers in the phenomenon of
Hindu goddess possession.
The questions informing my inquiry into this phenomenon include the fol-
lowing: to what extent do women’s ritual activities, especially those connected
with goddess possession, articulate a discourse that reproduces, legitimates,
and validates the social order (i.e., the elite Brahminical ideology of women’s sub-
ordination), and to what extent do they articulate a discourse that challenges,
alters, and transforms the social order? To what extent do these rituals reflect
women’s roles in the domestic sphere, and to what extent do they transgress
these boundaries? How does goddess possession transform women’s identities
and socioreligious roles? In what sense are matajis and their devotees power-
ful? In this brief chapter, I will not address all of these issues explicitly or in
detail, but they do form the background of my inquiry. Here, I present women’s
goddess possession rituals as traditional cultural resources which create an
arena for women’s empowerment in the varied and rapidly changing context of
contemporary India.
I wish to state at the outset that I do not see possession as a pathological
condition either of the individual person or of the society, nor do I see it as an
exotic remnant of the distant past, an anachronism which will (or should) soon
disappear as people give up ‘‘superstition’’ or adopt a more ‘‘rational’’ outlook. I
would like to think that such notions are outdated, if not in popular understand-
ing, then in scholarly discourse, but such is not the case. Such views are eth-
nocentric in that they privilege the culturally constructed post-Enlightenment
European notion of a separate, individual, impenetrable, and inviolable self,
what Alan Watts called the ‘‘skin-encapsulated ego’’ and Gregory Bateson de-
cried as ‘‘the epistemological error of Occidental civilization’’ (Macy 1990, 53).
Nor do I see goddess possession as a less important or inferior form of religious
expression, as compared with the more male-dominated, textual, and elite San-
skritic rituals. In general, I am in agreement with such feminist scholars as
Janice Boddy (1989) and Susan Starr Sered (1994), who view possession trance
as part of the normal range of human experience, even a talent, which tends
to appear more often among women, not because of any essential quality of
the play of the mother 151
women, but because their life experiences and culturally constructed roles and
identities predispose them to it. Considering spirit possession cross-culturally
and seeing it as a prime example of the this-worldly orientation and imma-
nence of the divine characteristics of female-dominated religions, Susan Sered
has eloquently written:
If we take stock of . . . explanations of women and spirit possession,
an interesting pattern becomes evident. All the theories . . . (social
deprivation, sexual deprivation, calcium deprivation, and over-
determination of gender) start from the assumption that possession
trance is an abnormal phenomenon. Therefore, the explanation for
women’s involvement in spirit possession necessarily lies in some
form of divergence from normal, healthy human experience. I would
like to raise a different possibility. Is it possible that possession trance
is one of a range of normal human abilities or talents, in much
the same way that musical ability or athletic ability is? Could it be that
in many cultures male socialization prevents most men from devel-
oping the ability to embrace the enriching, exciting, normal experi-
ence of spirit possession? Is it perhaps the case that the vast majority
of men, for a variety of psychosocial reasons, are so preoccupied
with guarding their ego boundaries from the threat of ‘‘invasion’’ that
they reject, or refuse to recognize, a religious experience that involves
melding one’s being with another entity? As Janice Boddy writes, ‘‘It
is imperative to ask why so many Western scholars . . . are committed
to viewing possession as a consequence of women’s deprivation
rather than their privilege, or perhaps their inclination’’ (1989, 140).
The answer to her question, it seems to me, lies in the double-
barreled intellectual weakness of ethnocentrism and androcentrism.
(Sered 1994, 190–191)
I have come to understand the possession experience, both for the woman
possessed and for her devotees who may be observing her, as continuous with
ordinary lived experience, so that women’s experiences in possession rituals
both influence and are influenced by their everyday lives. Moreover, I largely
reject the so-called deprivation theory, proposed by anthropologist I. M. Lewis
(2003) and other scholars, which holds that women and other low-status
people resort to possession and ecstatic religious expressions in order to com-
pensate for their relative lack of power in secular life. Besides noting the fact
that many high-status women, as well as some men, are involved in ecstatic
religious practices, I suggest not only that such religious or spiritual power is
valid in its own right, not inferior to, derivative of, or a substitute for economic
or social power, but also that ‘‘religious’’ and ‘‘secular’’ power are simply con-
venient analytical categories that often obscure more than they illumi-
nate. Furthermore, goddess possession should not be viewed as an isolated
phenomenon but in the Hindu context must be viewed as a religious expres-

sion coexisting and intertwined with many other ritual, devotional, yogic,
and contemplative practices associated with Shakta traditions.
Matajis: Shakti-filled Women
Possession by many types of deities and spirits, benevolent and malevolent, is

widespread, particularly among women throughout South Asia (Schoembu-
cher 1993). Divine possession by goddesses, again primarily among women, is
prevalent in the northwest part of India, which includes Kangra, and in other
areas of India, such as Bengal (McDaniel 2004).1 The goddess—in Kangra
worshipped by such names as Vaishno Devi, Jvala Mukhi, Durga, and Kali—
manifests herself in many different personalities and modes, including icons
(murti), stone pillars ( pindi), and flame ( jot or jyoti), but possession is the most
dramatic of these manifestations, for it brings the goddess in face-to-face and
voice-to-voice contact with her devotees. The shakti, or divine power, of the
goddess enters into, or plays with, the body and temporarily blots out the con-
sciousness of the human medium. While in this possession trance, the me-
dium speaks with the voice of the goddess. Afterward, she returns to her ‘‘normal’’
state of consciousness and may or may not remember what was said in trance.
By word of mouth, such women typically begin to attract followers who come to
them for help with problems and address them as mataji, a term which not
only means mother in the ordinary sense, but also a respected holy woman or
even a goddess herself. A transformation takes place through repeated pos-
session, so that the woman becomes more and more ‘‘goddess-like,’’ more and
more divinized, even in her nonpossessed state. Matajis may be married or
single, householders or renunciants, high caste or low caste, and locally, re-
gionally, or even nationally known. I give brief snapshots here of three matajis
to give a sense of the spectrum of women who play.
One such woman was Tara Devi, a barely literate housewife with several
children who lived in a village near Baijnath. She was transformed through a
process of repeated possession and ‘‘negotiation’’ with the goddess Jvala Mukhi
into a respected healer with a home temple and a large clientele. In an earlier
publication (Erndl 1997), I focused on her life and activities, suggesting that Tara
Devi’s identification with the goddess was empowering for her and for other
women in her family and village. I also suggested that because of the prominence
of goddess worship in Kangra and the cultural acceptance of women as her
legitimate vehicle, Tara Devi had tapped into (or was led into, depending on
where one places the agency) a traditional source of power for women.
Tara Devi had to work within the constraints of her marriage and motherly
duties to live her life dedicated to serving the goddess, but other matajis operate
more independently, rejecting marriage, often at the behest of the goddess
herself. Such women are self-supporting through their healing and ritual
activities and are often well respected and sought after in their communities
(Erndl 1993, 113–134). Divine possession by the goddess is one of the few cul-
turally accepted forms of avoiding marriage in traditional Indian society, which
allows few avenues of self-expression or economic support for women outside
marriage. However, while such unmarried matajis are viewed in their com-
munities as extraordinary women to whom the normal rules do not apply, they
do exercise influence concerning women’s roles.
An example is Passu Mataji, an unmarried healer with her own Vaishno
Devi temple in a village near Dharamsala in Kangra district. Interestingly, I
first heard of Passu Mataji some years before meeting her through a Dhar-
amsala psychiatrist, who told me that he sometimes referred clients to her.
This psychiatrist explained his views on possession to me in terms of classical
Freudian psychoanalytic theory, but it was clear that he respected the tradi-
tional healers and mediums and considered that their practices could help at
least some patients. He also told me some tantalizing details of Passu Mataji’s
life, which she confirmed and expanded upon when I later spent several months
visiting her and her devotees. She was born around the time of Indian inde-
pendence (1947), the youngest of six children to an impoverished family of the
Giraths, a low-caste cultivating community of Kangra. She had no formal ed-
ucation and never learned to read or write. As a child, she never played with the
other village children but only with another little girl who was always with her,
but whom only she could see. When she told her mother about this little girl,
her mother became concerned and brought her to a healer, a Harijan (Un-
touchable caste) who made ‘‘tantra-mantra’’ (a type of magical incantation or
sorcery) on her and gave her holy water and cardamom. After that, she ‘‘went
crazy,’’ and a wandering holy man came by her house, saying that the goddess
wanted to come there, but that it was contaminated. Her parents arranged
a marriage for her when she was very young and tried to send her off to her
husband’s home at about age twelve. When she was placed in the palanquin,
the goddess entered into her and said, ‘‘Don’t go there; stay here and build a
temple for me.’’ She followed the goddess’s order, though it is unclear how
much family resistance she faced. Nevertheless, she did build a temple at her
parents’ home, and some years later, again ordered by the goddess, moved her
temple and residence to its present location in another village. She never mar-
ried again and characterizes herself as ‘‘not a householder,’’ though she does
maintain a household and has brought up several relatives’ children. She is
very active in the local Mahila Mandal (Women’s Organization), has been ap-
pointed to the board of a nearby free medical clinic, and supports educational,
social, and economic opportunities for women. More radically, while respect-
ing the status of the householder and stressing interdependence and an ethic
of caring among men, women, and children in the family, she believes that
marriage is not necessarily a sacred duty for women. She told me that women
should think long and hard before agreeing to marriage and that no one should
be forced into marriage against her will. Interestingly, this attitude assumes
that women have some agency over the events of their lives and are not merely
passive recipients of others’ decisions or of a social system beyond their con-
trol. In response to a question of mine, she also said that some men worship
the goddess but mistreat their wives and other women, because they have a
problem with inflated egos (‘‘they think they are big’’). Such ideas she attri-
butes to the influence of the goddess in her life, for she is illiterate and has had
virtually no exposure to Western ideas.
Some matajis have taken on a renunciant lifestyle and have become gurus
in a broader sense, teaching meditation to their devotees, building ashrams and
in some cases free clinics and schools, and giving discourses on bhakti (devo-
tion) and other spiritual paths. An example is Usha Mataji, about whom I have
previously written as Usha Bahn or Sister Usha (Erndl 1993). When I first met
her in 1982, Usha, a West Panjab émigré based in Agra, had a growing fol-
lowing due to her personal charisma and her standing as a vehicle of Vaishno
Devi and Kali. In the intervening years, she has expanded her movement to
include a large temple and ashram in Delhi and an ashram in Nurpur, Hi-
machal Pradesh, adjacent to Kangra, as well as hospitals and clinics. While still
experiencing possession trance, she also practices and teaches meditation tech-
niques and mantras to her disciples. She now identifies herself as a renunciant
(sannyasini) and initiates women and men as both lay disciples and renunci-
ants. By her own account, some years ago, she was granted the title Mahaman-
daleshwar, the highest renunciant designation, by an assembly of sannyasis in
Hardwar. She leads mass pilgrimages every year to Vaishno Devi and other
goddess temples, travels around India, and has even visited her devotees in the
United States. Some of the women I met who are devotees of local matajis in
Kangra have also taken initiation from Usha Mataji, thus showing continuity
between village-style healer matajis and international tantric-style gurus. I met
Usha Mataji again in 1997 in Nurpur and at her ashram in Delhi, where she
answered my questions about her experience of shakti, her religious teachings,
and her understanding of renunciation. In contrast to some renunciants who
have become involved in Hindu nationalist activities, she emphasized to me
that she was totally opposed to their activities and had absolutely nothing to do
with them.
Possession Rituals
What happens during a possession ritual? There is considerable variation.2 The

possession may take place at a temple, in a woman’s home, in an outdoor area,
or in a small shrine which a woman has built in her home compound. The
possession may be unpredictable and spontaneous, or it may be a more regular,

predictable occurrence. Most of the matajis I have encountered set aside par-
ticular times and days of the week for devotees to come and consult them about
their problems. They typically will sit in front of a small shrine with goddess
images, such as those of Durga, Vaishno Devi, Jvala Mukhi, or Kali. At the be-
ginning of the session, before the mataji goes into trance, devotees begin to
gather outside, and after removing their shoes and washing their hands, they
enter and bow to the mataji and to the goddess image. They may make offerings
of food, flowers, incense, or cash and may also exchange a few words of greeting
with the mataji. If they plan to ask a question, they will place a personal item
such as a watch, ring, article of clothing, or clay from their house in front of
her. Then they will seat themselves and join the singing of devotional songs to
the goddess.
After some time, the goddess will enter into, or play, in the mataji. The
mataji begins to sway, her eyes glaze over, and her head may begin to rotate in a
circular fashion, causing her hair to loosen and fly about. While the mataji is in
this trance, a voice speaks, identifying herself as a particular goddess or sig-
naling her presence by a phrase like Jay Mata di (Victory to the Mother). She
picks up each object in turn, asking whose it is. She talks to the devotee, asking
questions, stating the problem, giving a diagnosis, and prescribing a remedy.
Sometimes, the devotee asks questions and engages in a conversation with the
goddess. In this way, the mataji moves from one item to the next. Usually, she
will pick up all the items and answer all of the questions before departing, but
sometimes the goddess will depart abruptly. After the trance is over, devotees
will line up to receive prasad (usually sweets or fruits that have been previously
offered to the goddess), the blessings of the mataji, and any blessed items such
as holy water, cardamom, or colored threads which have been prescribed dur-
ing the trance.
The questions that devotees present to the goddess concern the entire
array of personal problems, including infertility, illness of a family member or
domestic animal, possession by a harmful spirit, family or marital disharmony,
depression, legal cases, employment troubles, or an upcoming examination.
Highly personal and painful tragedies are discussed openly within earshot of
all participants. The personal objects that the devotees place before her often
have some direct connection with the problem: job application papers, a sick
child’s shirt, a family photograph. Once, I even saw the owner’s manual for a
motorcycle. In some cases, the mataji tells them that the outcome will be good
and that they shouldn’t worry. In others, she gives a more specific diagnosis
which may fall into one or more of the following categories: (1) sorcery, in
which some evil wisher has brought harm by feeding the devotee something
unwholesome or burying it in her house, (2) an inauspicious astrological in-
fluence, such as Saturn or the sun, or (3) some action or omission on the part of
the devotee or the devotee’s family member, such as quarrelsome behavior or

failure to perform a prescribed ritual. Remedies prescribed include sacred ash,
holy water, and cardamom; wearing a multicolored thread; performance of a
ritual such as a havan (fire sacrifice); making an offering to a deity; and ad-
hering to certain dietary restrictions. Devotees who have been helped or cured
return later to give a thanks offering, and many of them become regular visi-
tors, participating in devotional singing, helping out with ritual chores such as
distributing holy water, and just hanging around to talk with the mataji or
other devotees. Healing takes place within a context of religious devotion to the
goddess and a shared community, albeit a temporary one, of devotees who have
joined together to consult the mataji.
Women’s Space
The importance of the ‘‘hanging out’’ time before and after the actual trance
session should not be minimized, as it creates a space and opportunity for
women of different castes and backgrounds to come together in ways that they
might not otherwise and to form a community. In rural Kangra, there are few
such legitimate public spaces for women. Women may visit the bazaar and move
about publicly, but only for some specific errand. Men may chat and smoke in
the bazaars with their friends and wile away the hours gossiping in tea stalls,
but women seldom do. They are expected to conduct their business and leave,
not to linger and socialize. In contrast, the courtyard or temple of a mataji is a
place where women can legitimately and safely spend leisure time and where
they can also turn to other women for solace and even practical assistance.
For example, on one occasion, I was sitting with a group of about ten
women on Passu Mataji’s veranda, waiting for her to return from a doctor’s ap-
pointment. Most of the women were from relatively high-status, prosperous,
peasant families, and a few were office workers or teachers. Among us that day
was a low-caste teenage woman with a young daughter from a neighboring vil-
lage. Most of us knew her slightly, as she had come to see the mataji a few
times before. I remembered her especially well, because unlike the other
women and girls I had met, she knew no Hindi or Panjabi and could speak
only the local Kangri dialect. That day, she was crying and despondent, telling
us that her mother-in-law beat her and that her husband wanted to divorce her.
Her husband had offered her a rather large sum of money if she would take her
daughter and leave. She was reluctant to do this, as she felt it was her duty to
stay in her husband’s home and accept her fate. Also, she was afraid about
what would become of her, as her natal family was of no help. The other
women rallied around and provided support. No one blamed her for her plight
or encouraged her to stay with her husband. In fact, the general consensus was
that she should take the money, put it into a fixed deposit for her daughter’s
future, and get a job. One of the women said that she would help her at the
bank, and another said that she would get her a job as a maintenance worker in
an office. This story illustrates the fact that, while oppression of women is very
real, there are ways to deal with it, both spiritual and practical. Not all of the
cards were stacked against this young woman since she had the support of this
multicaste group of women from outside her family. While the mataji was not
even present during this exchange, she provided the space for it to take place. It
was their devotion to the mataji and their desire to take part in her rituals that
brought these women together. Such spaces may be seen as ‘‘cracks’’ in a pa-
triarchal system, that is, spaces that exist at least in part because of patriarchy
but which provide sites for women’s creativity and interconnection, sites for
thoughts and activities that can never be completely controlled by patriarchal
norms and which furthermore have the potential to resist, transcend, or trans-
gress patriarchy.3 A cross-cultural example comes to mind. When the Taliban
took control of Afghanistan, hard-won women’s rights were eroded one by one,
as women were deprived of jobs, education, freedom of movement, and civil
liberties. But the fatal blow came when the Taliban closed down the hamam, the
traditional bath houses for women, the one place where women could congre-
gate and communicate away from the watchful eyes of men. Perhaps the
Taliban saw the hamam as not only a place where women could experience
(hitherto legitimately) bodily sensual pleasure, but also as a site of potential
rebellion against the regime.
Possession rituals are a traditional cultural resource that can be a source of
empowerment for women. In no way am I suggesting that possession rituals
are necessarily liberating nor that they are ‘‘the answer’’ to women’s problems.
But I do believe that women’s lives are enriched—spiritually, socially, and
emotionally—in ways they would not be if they did not have these practices and
that possession rituals give them a sense of confidence that there is something
they can do about their problems and that they are not helpless victims. Pos-
session is an embodied phenomenon in which women experience the divine
either within their own bodies, as in the case of the matajis themselves, or as
part of a collective, participatory process, as in the case of the matajis’ clients
and devotees. In either case, women gain access to power, many kinds of power,
including the power to improve their lives, mitigate suffering, help others, and
achieve spiritual insight.
notes
1. I do not want to give the impression that either possession in general or even
possession by goddesses is an exclusively female activity. In Kangra, for example, men
are occasionally possessed by goddesses, though it is generally a female-dominated
activity. See Erndl 2001 for my discussion of a male mataji who takes on the guise
of a woman while possessed by his goddess. Also, in some regions of India in
some ritual contexts, possession is primarily a male activity as, for example, in the
Bhadrakali performance traditions of Kerala. Sarah Caldwell (2001) explains why this
is so in Kerala, while hypothesizing that tribal hill women were the original mediums
for the goddess Bhadrakali.
2. See Erndl 2000 for the transcript of a particular session at Passu Mataji’s
temple.
3. Janet Chawla (2002, 165) points out that childbirth, like possession, is also
a liminal embodied phenomenon and thus can never be completely controlled by
patriarchy.
references
Boddy, Janice. 1989. Wombs and Alien Spirits: Women, Men, and the Zar Cult in
Northern Sudan. Madison: University of Wisconsin Press.
Caldwell, Sarah. 2001. Oh Terrifying Mother: Sexuality, Violence, and Worship of the
Goddess Kali. New York: Oxford University Press.
Chawla, Janet. 2002. ‘‘Negotiating Narak and Writing Destiny: The Theology of
Bemata in Dais’ Handling of Birth.’’ In Invoking Goddesses: Gender Politics in
Indian Religion, ed. Nilima Chitkopekar, 165–199. New Delhi: Shakti.
Erndl, Kathleen M. 1993. Victory to the Mother: The Hindu Goddess of Northwest India
in Myth, Ritual and Symbol. Oxford and New York: Oxford University Press.
———. 1997. ‘‘The Goddess and Women’s Power: A Hindu Case Study.’’ In Women
and Goddess Traditions in Antiquity and Today, ed. Karen L. King, 17–38.
Minneapolis, MN: Fortress.
———. 2000. ‘‘A Trance Healing Session with Mataji.’’ In Tantra in Practice, ed.
David G. White, 97–115. Princeton, NJ: Princeton University Press.
———. 2001. ‘‘Goddesses and the Goddess in Hindu Religious Experience: Con-
structing the Goddess through Personal Experience.’’ In In Search of Mahadevi:
Constructing the Identities of the Hindu Great Goddess, ed. Tracy Pintchman, 199–
212. Albany: State University of New York Press.
Lewis, I. M. 2003 [1971]. Ecstatic Religion: A Study of Spirit Possession and
Shamanism (3d ed.). London: Routledge.
Macy, Joanna. 1990. ‘‘The Greening of the Self.’’ In Dharma Gaia: A Harvest of
Essays in Buddhism and Ecology, ed. Alan Hunt Badiner, 53–63. Berkeley, CA:
Parallax.
McDaniel, June. 2004. Offering Flowers, Feeding Skulls: Popular Goddess Worship in
West Bengal. Oxford and New York: Oxford University Press.
Ram, Kalpana. 2001. ‘‘The Female Body of Possession.’’ In Mental Health from a
Gender Perspective, ed. Bhargavi V. Davar, 181–216. New Delhi: Sage.
Schoembucher, Elisabeth. 1993. ‘‘Gods, Ghosts, and Demons: Possession in South
Asia.’’ In Flags of Fame: Studies in South Asian Folk Culture, ed. Heidrun Brüchner,
Lothar Lutze, and Aditya Malik, 239–267. New Delhi: Manohar.
Sered, Susan Starr. 1994. Priestess, Mother, Sacred Sister: Religions Dominated by
Women. New York: Oxford University Press.
9
Does Tantric Ritual Empower
Women? Renunciation and
Domesticity among Female
Bengali Tantrikas
June McDaniel
There is a variety of roles for women in Indian religion, ranging

from the female ascetic who renounces the world to the wife and
mother who practices religious ritual as part of her household obli-
gations. In this chapter, we shall explore Hindu understandings
about female Shakta tantric practitioners in West Bengal. Tantra
allows women a broad range of responses to domestic norms, from
conformity to rejection of domesticity as an important arena of fe-
male religious expression. But women who are tantric celibates
tend to have higher social status in Bengali culture, while women
who perform tantric sexual rituals tend to have lower status, sug-
gesting a larger religious valorization of asceticism and renuncia-
tion over domesticity and householdership.
This chapter will contrast the roles for women as written in
tantric texts and as lived out in women’s practice. These are in
many ways quite different. Tantric texts describe three major roles
for women: they can be incarnations of the goddess, or consorts
for sexual ritual, or female gurus. These are the roles most often
mentioned in the texts, though other roles are occasionally noted.
However, women’s lived roles in tantric practice are much broader
than these three roles. We see female tantrikas who are renunciant
practitioners, holy women of various types: the woman who has re-
nounced worldly life (sannyasini); the woman who is dedicated to
celibacy, service, and obedience to a tradition (brahmacarini); the
woman who practices yoga, especially kundalini yoga (yogini); the
woman who is married but has left her husband to pursue a spiritual life (grihi
sadhika). A woman may be a devotee of a tantric deity and worship with tantric
mantras, or she may get possessed as a vocation (thus becoming a ‘‘bhar
lady’’). The female tantrika may also be a wife who practices tantric sexual
rituals as part of her marriage, or a professional ritual partner in tantric sexual
practices outside of marriage. She may be a female teacher (stri-guru), usually
celibate and head of a group of devotees or an ashram. She may also be a
widow or celibate wife, whose practice involves ritual tantric worship (puja), a
mixture of devotional love of a deity, service to that deity, and tantric ritual
meditation.
Woman and Text
The descriptions of women in the tantric texts tend to follow an exaggerated

style—they are strongly sexual, or attractive at a distance, or both. One does not
often get the sense of the physical women used as models for them. Rather,
these women are ‘‘perfect’’ in various senses (beautiful, graceful, happy, quiet,
obedient), or they are imaginary women, also beautiful and graceful but dwell-
ing on lotuses wearing silk and jewels, sometimes having a frightening de-
meanor. Women in tantric texts are described primarily in terms of their looks
and actions: the male tantrika should find a suitable woman (according to a
long list of qualifications) and then perform rituals with her. The Kularnava
Tantra states that the woman must be beautiful, young, pious, devoted to her
guru and god, always smiling, pleasing, and without jealousy, among other
qualities.1 The female tantrika cannot be unattractive or old or sleepy, and she
cannot feel desire or argue with her partner—these disqualify her from tantric
practice, even if she has been initiated.2
Ritual Incarnations of the Goddess
The attitude toward women in tantric texts is generally positive. Many texts say
that women can be tantric gurus, and a male practitioner’s mother is the best
guru possible. A woman may be knowledgeable as a tantric consort without
regard to her social status: she may be a courtesan or laundrywoman or dancer
or fisherwoman, a woman who sells meat or one who works with leather. Some
tantras encourage the worship of goddesses within living women and girls, for
women may incarnate Shakti. Some tantras say that one must never harm a
woman nor look down upon her nor even hit her with a flower. In the Kali Tan-
tra, women are respected, especially the kula woman (a female tantric practi-
tioner of the kula marga):
does tantric ritual empower women? 161
5. [The practitioner] should imagine the whole world as female, and he

should also think of himself as female.
6. A wise person should consider drink, food to be chewed or sucked,
all edible things, the household, himself and everything else, as a
young woman. When he sees a kula woman, he should bow [to her]
with reverence.
7. If by good fortune he should encounter a kula woman, he should men-
tally worship her.
8. He should bow respectfully before a young girl, a teenaged girl, an
old woman or a young woman, even if she is nasty, ugly or bad.
9. Women should never be beaten, insulted, or cheated, and should
never be treated badly. If a person does treat a woman badly, he will
be unable to attain success [in his practice].
10. Women are deities, women are life [ prana], women are beauty.3
In texts that speak of the woman as a ritual incarnation of the goddess, a

suitable woman is found and offered worship. Sometimes a young girl is
worshipped (kumari puja), and sometimes a mature woman is the object of
worship (stri puja or shakti puja). Kumari puja is often performed during the
nine-day festival of Durga Puja, and it is believed to bring great blessings. The
Kubjika Tantra details the worship of young virgins in kumari puja (primarily
for girls ranging in age from five to twelve years, though the Mahacina mode
includes worshipping girls ranging from one to sixteen years), and girls from
the ages of six to nine years old are especially to be desired, as such worship
grants the devotee all wishes. This tantra also includes worshipping both
one’s own wife and the wives of others as goddesses, repeating mantras 108
times and seeing the woman as the symbolic form in which the goddess dwells.4
As well as being within the woman, the goddess may be located in food, wine,
fish, red cloth, red flowers, and a red sun. The Guptasadhana Tantra speaks of
the nine types of virgin girls who may incarnate the goddess, including the
actress, prostitute, Brahmin woman, low-caste woman, wives of washermen
and barbers, and daughters of a kapalika (skull-carrying) ascetic, cowherd, or
garland maker.5
In the Kulacudamani Tantra, the goddess describes the worship of eight
women, who represent the eight shaktis, or mothers. They are sometimes called
the consorts or powers of the Vedic gods (as one may deduce from their names,
such as Brahmani, Maheshvari, and Indrani). However, they are usually wor-
shipped independently of the gods, and they are said to grant supernatural
powers, good karma, and the removal of obstacles to the desires of the tantrika.
The text gives directions, suggesting that the tantrika bring the women to a
deserted place, such as a river bank, a crossroads, a burning ground, or the foot
of a bilva tree. He brings the women sanctified water and looks closely at them.
28. By observing the differences in the appearance, mood and behav-

iour of the women, they are given the names of the eight shaktis
beginning with Brahmani, etcetera.
29. First offering them a seat and welcoming them with a mantra,
[he should present] blessed water for drinking, water for the feet,
plain water and an offering of milk and honey.
30. He should bathe and dress the hair with scents and flowers and
after censing the hair, he should offer silken garments [to the shak-
tis].
31. Then spreading out a seat in a different place and having led the
shaktis there, when he has given [them] a pair of sandals [and]
adorned [them] with jewels and ornaments,
32. He should offer ointments, scents, and garlands. And having invoked
the shakti of each of them, he should place [the designated shakti]
on the head of each of the women.6
The practitioner gives each woman the name of one of the mothers, and
he chants a brief hymn to each woman. The goddess explains that without these
hymns, the tantrika will lose the fruits of his worship. He should also give them
betel nuts and sweet seeds, as well as a garland with sandalwood paste and
perfume. In the interests of inclusiveness, the goddess adds:
52. . . . Oh Bhairava! If there is a man there who knows the kula teach-
ings, he [too] is deserving of worship!
The worship of the human woman as the goddess is generally a temporary
phenomenon; the goddess does not remain in her body after the ritual is over.
The human woman is much like a statue in this ritual, used as a temporary
home for the goddess who comes to visit. Goddesses may ritually dwell almost
anywhere—from trees to corpses. However, because the human female has
her own feminine power, or shakti, she is an especially appropriate place for a
goddess to dwell.
Female Ritual Consorts
The second and most publicized role for a woman in the tantric texts is the role
of consort in sexual rituals. The Niruttara Tantra suggests worship of the vesya
(the term traditionally means prostitute), including those who come from a
tantric family, those who are independent of family, those who join (the pro-
fession) voluntarily, those who are married to male tantrikas, and those who
have been ritually united with the deity.7 In this usage, the term vesya does not
refer specifically to a prostitute, but rather to a woman who roams about as
freely as a prostitute may, and enjoys herself like Kali. She has sex accompa-
nied by the chanting of mantras, and meditates upon the union of Mahakala
and Kalika. While such an image may initially give an impression of a free
woman in the modern sense, this is not the case—her freedom is limited by
the roles defined in the Niruttara Tantra. She is not a tantric vesya if she
becomes involved with a man other than her husband; as the text phrases it, if
she worships a Shiva other than her own Bhairava, she will live in the fierce hells
until the destruction of the universe. If she gets involved with other male
practitioners due to passion, desire for money, or other temptations, she will go
to hell. She is then called an animalistic prostitute (pasu-vesya). Any man in-
volved with her will suffer disease, sorrow, and loss of money. The proper vesya
must be chaste and pious, doing rituals with her own partner.8 She cannot be
respected and take on a different partner, and thus she cannot instruct other
male partners by ritual.
Sexual practice within a single couple is called lata sadhana, the spiritual
exercise in which the woman is like a vine (lata) growing around the man. Lata
sadhana is individual practice, with a single couple alone practicing mantras,
breath control, and other forms of meditation in a ritual context. The kula cakra
is a group practice, where men and women sit in a circle (cakra or chakra) in
couples, and perform the ritual of the pancatattva, taking the five forbidden
things, of which one is intercourse. The tantric texts tend to be rather evasive as
to the details of lata sadhana; these should come from the guru. However, a
good deal of ritual worship is involved, as the Maya Tantra states:
4. Bring a woman while she is menstruating, and at midnight worship

your ishtadevata [personal deity] within her genitals.
5. After that, the practitioner must chant 336 mantras daily [the text does
not specify which mantras], for three days. By means of this, he can gain
the fruit of one thousand corpse rituals. There is no doubt about it.
6. Here is another type [of ritual], please listen carefully. First, to gain
perfection in the four paths, [the practitioner] should chant mantras
108 times. Then he should worship his personal deity in the genitals
of a woman who is not his own. Then he should worship Mahamaya
108 times, using menstrual blood.9
7. After that, he should offer a burnt offering, and chant mantras 108
times. He should become devoted to the practice, and be continually
absorbed in Mahamaya.
8. If he does this daily for sixteen days, he will become rich, powerful, an
orator and a poet, and loved by all. There is no doubt about it.10
The goal here is the enrichment of the male tantrikas; the woman merely
brings along the goddess within her. We see the same problem in the ritual
cakra, as the Kamakhya Tantra states:
35. [The practitioner] will bring an initiated woman, and establish a ritual
circle [kula cakra].
36. Then the practitioner will joyfully worship the goddess, [especially]
her genitals. Then he will sing hymns sweetly and chant mantras
continuously, while looking at his partner.
37. He who chants mantras continuously while in this state, will be
lord of all supernatural powers in the Kali Yuga.11
Here the male tantrika is involved with a woman, with worship and mantras as
ritual. The Kamakhya Tantra states that the man will gain siddhis, or supernat-
ural powers. The effect of this practice on the female tantrika is not mentioned.
The most vivid description of the female role in a more orgiastic style of
cakra is probably found in the Kularnava Tantra. It is an interesting description,
as it violates the notions that practitioners should not be desirous, and the text
allows them to have other partners. However, the passions are justified because
the participants are in an altered state (ullasa), and because everything in the
cakra is transformed (eating becomes fire sacrifice, sight is meditation, sleeping
is worship, and union with one’s partner is liberation):
67. Intoxicated by passion, the women take shelter with other men,
treating them as their own. Each man also takes a new woman
[shakti] and treats her as his own, when in the state of advanced
ecstatic joy.
68. Seized by delusion, the men embrace other men. . . .
71. O Shambhavi! The yogis take the food from each other’s plates and
dance with their drinking pots on their heads. . . .
73. The women who are not in their normal senses clap and sing songs
whose words are unclear, and they stagger while dancing.
74. Yogis who are intoxicated with alcohol fall upon the women, and
the intoxicated women fall upon the men, O Kulanayika! They are
induced to perform such actions, to fulfill their mutual desires.
75. When this state of ecstasy is not accompanied by corrupt thoughts,
the bull among yogis reaches the state of godhood [devata-bhava].12
The effect upon the cow among yogis is not clear, but one assumes that the
women would have similar ecstatic experiences.
Female Gurus
A third female role is that of the guru. In some tantras, we see female gurus
idealized. For instance, the Guptasadhana Tantra gives a visualization of the fe-
male guru: she is located in the Sahasrara lotus above the head, and her eyes
look like lotus petals. She has high breasts and a slender waist, and she is shin-
ing like a ruby. She wears red clothes and jeweled ornaments. She is seated at
the left of her husband, and her hands show mudras giving boons and freedom
from fear.13 She is graceful, delicate, and beautiful. Such an image is quite
different from the reality of the physical female tantric gurus, who tend to be
older, unmarried, sometimes bald nuns, often toughened from ascetic and out-
door life, looking strong and sometimes grizzled. They generally do not wear
jewelry, seeking to avoid the dangers of sexual attractiveness. The last thing
they want to be is beautiful and delicate, while sleeping alone on temple floors
or wandering on pilgrimage; they often travel alone and need to defend them-
selves. Their emphasis is on independence and attaining liberation rather than
seductiveness.
As we can see, actual female tantrikas do not exactly fit these idealized
textual roles. Instead, the female Shakta tantrikas whom I interviewed and
who were described by informants during fieldwork in West Bengal tended to
fall into five roles:
1. Celibate tantric yoginis, whose status was the highest among the
women interviewed, were lifelong celibates. Many were gurus with
disciples, and some headed temples, ashrams, or tantric study cir-
cles. Some emphasized the importance of devotion toward the god-
dess or guru; others were believed by their disciples to be partial
incarnations of the goddess. Tantra for them was a dedicated prac-
tice involving mantras, visualizations, austerities, and ritual actions
(kriya). The goal of tantra was to gain liberation and also shakti,
both as the goddess and as power. There was little evidence of do-
mesticity, except occasional cooking for disciples and festivals, and
service to visitors.
2. Holy women, called grihi sadhikas, had been married but left their
husbands and families to follow a religious calling. They had lower
status than the lifelong celibates, but some had disciples. Often they
would wander, practicing tantric meditation and worship, and live
at temples or ashrams. Some would go into states of possession by
the goddess (Kali bhava) or other deities, induced by chanting tantric
bija mantras or singing hymns to the devi. Tantra for them was de-
votion and possession, usually in response to a call by the goddess.
The goal of tantra was to follow the goddess’s will in an ascetic set-
ting. Here, we have traditional domesticity followed by renunciation.
3. Tantric wives performed tantric ritual sex and worship as part of
their devotion toward their husbands and gurus. The woman was
often initiated by the same guru as her husband and followed his
teachings. Tantra was a form of service, involving obedience to hus-
band and guru and following women’s marital obligations. The goal
of tantra was to fulfill dharma and social obligations. Thus, tantra
as practiced within marriage was equivalent to domesticity and
followed traditional social norms.
4. Professional consorts were women who performed ritual sex and

worship as a way to make a living. The consort and her children
were generally supported by the man who was her ritual partner.
The woman may have moved from one sadhu to another, depending
on who would shelter and support her. Tantra here was profes-
sional sexual practice, a career choice. The goal of tantra was to help
the male tantrika in his practice, make money, and possibly get a
permanent home and a male protector. Perhaps such practices
might be understood as serial domesticity.
5. Celibate wives and widows were householders who incorporated
tantric practice as an aspect of worship. Tantra was a form of devo-
tion, especially in combination with bhakti yoga. The goal of tantra
was to please the goddess and gain blessings, without sexual prac-
tices. Here, renunciation and domesticity were combined in a fash-
ion that was socially respectable and fairly widespread (according to
informants).
Let us examine some of these roles in greater detail. As background, I inter-

viewed practicing female tantrikas in West Bengal in 1983–1984 and 1993–
1994. Some were ascetics and heads of ashrams; some were householders;
some made no secret of their practice; some practiced underground, unknown
to family members, coworkers, and neighbors. Most female tantrikas inter-
viewed were not only celibate, but insistently so. Several said that tantric
meditation involves purity and concentration and that desire would be a dis-
traction and would cause them to fail.
Celibate Tantric Yoginis and Gurus

I interviewed in depth three women who were female gurus, initiated into
Shakta tantric lineages. Two of them practiced tantric rituals themselves, and
the third was of the Shakta universalist perspective, and she knew many prac-
ticing women tantrikas. All of these women were highly respected, and all of
them were celibate. All of the highest-status tantrikas whom I met, male or
female, were celibate.
For Gauri Ma, head of an ashram in Bakreshwar, tantric ritual revealed a
person’s ‘‘inner history,’’ giving the power to ‘‘see inside,’’ to watch the inner
life of the spirit. The goal is to ‘‘gain’’ Shakti (shakti labh kara), to have her dwell
in the heart. It is Shakti who enlightens you, who brings you to the highest
states. Shiva is as useless as a corpse, and that is why he is portrayed as one in
the iconography. In her practice of kundalini yoga, the male and female aspects
of the person were united, and there was no necessity for any union of physical
males and females. In the pancatattva ritual, or circle with five forbidden things,
the actions are symbolic, with drinking wine representing control of the breath,
figure 9.1. Sannyasini Gauri Ma and the author in Bakreshwar, West Bengal. Photo
by Jim Denosky.
and ritual sex representing the sort of union seen in yogic meditation. She
made a special point of saying that no outward practice of these is necessary for
a strong and disciplined tantrika and that tantric rituals are symbolic of inner
transformations.14
For Jayashri Ma, the female guru of a group of devotees, tantric ritual is
a way of getting a fused identity with Shakti, which lasts over a lifetime.
Jayashri was initiated by her tantric guru while they sat on matched sets of
skulls, and with the mantra came the direct entrance of the goddess Adya
Shakti into her heart. The mantras, hand positions, trances, and rituals were
ways of preparing her body for Shakti’s entrance. Union with Adya Shakti is
the highest state possible, for she is identical with Brahman, and mother of
the universe. Jayashri came from a tantric family, and her guru was a tantrika
who practiced secretly. Jayashri too practices in an underground fashion, for
she said that religious practitioners (especially tantrikas) were persecuted by
the communist authorities in her area. She has many health problems, but the
ascent of the goddess out of her heart and into full consciousness blots these
away, and she becomes aware of nothing but the goddess’s love and power.
She is a celibate tantric guru, who said that she no longer needs to perform
rituals because the goddess has already taken up permanent residence in her
heart.15
For Archanapuri Ma, tantric practice has a strong component of devotion.

She is the head of an ashram near Calcutta and a celibate member of a Rama-
krishna lineage. Her Shakta guru, Satyananda, began with a Vedanta perspec-
tive, but later had a revelation from the goddess Kali Bhavatarini that he must
worship her. He performed tantric meditations and offered blood to Kali, and
he taught Archanapuri Ma many meditative and ascetic techniques. Her un-
derstanding of tantra is heavily infused with devotion, toward both guru and
goddess.
In the nineteenth century, the most well-known case of a female tantric
guru was probably that of Bhairavi Brahmani or Yogeshvari, a woman tantrika
who came to see the Bengali saint Ramakrishna Paramahamsa and stayed with
him for about three years. She is described in Ramakrishna’s biographies as
almost forty years old and attractive, and she validated Ramakrishna’s spiritual
status. She had him sit on skull seats, chant mantras, eat fish and human flesh
from a skull, and perform many practices described in the major tantras. She
brought him women with whom Ramakrishna could perform some of the
rituals, but the biographies are unclear about his practice with them; Rama-
krishna claimed to have fallen into trance and been unaware of performing
anything.16 There is no direct evidence that Yogeshvari at any point lost her
celibate status.
Ramakrishna seems to have been quite ambivalent toward these practices,
partly accepting her as a guru and partly rejecting tantric practices generally.
His followers were also ambiguous about the Bhairavi. Some found her teach-
ing acceptable, as she was high caste, and even sexual practices were allowable
if a guru were present (rather like having a chaperone). Others found her a bad
influence and were glad when she left.
Tantric Holy Women

In the second role, we have tantric holy women who have been married but left
their homes due to a religious call. They have lower status, though their reli-
gious dedication and newfound celibacy still gain them respect. The holy woman
(grihi sadhika) who has gone out on her own, only returning occasionally to
visit her husband and children, does not have an easy life. Often she has been
initiated by the family’s Shakta household priest (kulaguru) and heard a pro-
phetic call in a dream or vision from a deity, who asked her not to oil her hair,
not to eat certain foods, and to go on a pilgrimage. She leaves the household
and may survive by begging, telling fortunes, or being possessed and gaining
donations from observers. She gains social status when she starts to gain
devotees, and sometimes she may have a special set of supernatural powers
given by the goddess (especially healing or materializing food). If she under-
goes possession, she is usually possessed by the goddess Kali, though she may
be possessed by other deities as well.
Such calls often begin while the girl is very young. As an example, the grihi
sadhika Lakshmi Ma was a devotee of Kali and Tara, both traditionally tantric
goddesses in West Bengal, and she used to see Kali and play with her when
she was a child. After she married, she continued to see Kali, who would com-
plain if she did not get sufficient offerings. Lakshmi Ma told her family of
her visions, and they thought that she was possessed. They bound her with
ropes and had her undergo an exorcism. It was unsuccessful, though the ex-
orcist burned her and bound her in iron chains. The family calmed down
when her husband had a dream of Kali, telling him to build an altar for her,
and they began to worship there. She later lived separately from her husband,
performing rituals at Tarapitha and other holy places. She dressed in classic
tantric fashion—red clothing, matted hair, heavy rudraksha garlands—and she
often carried a large trident.17
Tantric Wives
For tantric wives who remain in the household, the religious goals tend to be
devotion and obedience to husband and guru and desire for union with Shakti.
I spoke several years ago with a Sahajiya, or tantric, Vaishnava couple, and the
perspectives of the man and the woman in the couple were very different. The
man emphasized adventure, pleasure (he claimed that male tantrikas could
have sex for four hours), and increased attractiveness (for tantra worked as a
sort of birth control and fountain of youth, allowing women not to get pregnant
and not to turn into wrinkled old hags by the age of twenty-four years, which he
said was what happened without it). Tantra was fun, exciting, and a way to es-
cape the routine. He felt that it made Indian men superior to Western men in
endurance.
His wife’s perspective was quite different. Tantric practice for her was obe-
dience to guru and god and a way to help her husband and please him. Tantra
was not rebellious but rather following stri-dharma, for it was the wish of her
husband and guru. Tantra was a way to serve them. She was unwilling to give
details of her practice unless I was initiated, but she did say that her guru’s face
was present at all times within her mind during the ritual.
This couple lived in a large joint family which farmed land in rural West
Bengal. They would leave the house late at night to practice, after everyone else
was asleep, and return in time to get some sleep before the day’s chores began.
Nobody in the family knew that they were practicing tantrikas. Many house-
holder tantrikas seem to practice this way, where either the family does not know
of their practice, or the family is of tantric lineage and they know, but the neigh-
bors do not.18
Such practice among couples is often highly secretive, known only to other
religious practitioners. Archanapuri Ma knew some women who were in tantric
arranged marriages. She described them as married women who were helpers
to their husbands. She said that these women were not used and thrown away,
as most people believe:
Tantra is like this: India has always tried to elevate all traits of
the human character through religion. The mind is the eternal
playground of sensual desire. While Western psychology has
understood this carnal tendency of mankind as his original or
root [mula] inclination, Hindu religion has tried to transform it
through the path of spiritual discipline, to divinize all tendencies
of the person. This is the basis of tantric viracara practice, and
it is very difficult, because some rituals involve taking a woman
companion [bhairavi]. Though these paths appear difficult to us,
they are very potent and useful for the deserving aspirants, and
have been revealed by liberated people. This practice centers
around strength, and requires a powerful mind and great concen-
tration.
We must also pay attention to the female tantrika [bhairavi].
Is her part only mechanical, required only by male aspirants to
prove their mental and spiritual strength? Is her life useless once
the above purpose is served? No, for there are many female tan-
trikas who can legitimately be called equal travelers of the spiritual
path and equal sharers in its benefits. Some have attained to great
religious heights, but most prefer to remain inconspicuous, and
people do not know about them.
In some cases both husband and wife take part together in
the tantric cakra ritual and practice together while leading an ac-
tive householder’s life. I have met a few such female tantrikas
who are engaged in spiritual practices, despite the responsibility
of bringing up children. They only belong to one cakra, for a fe-
male tantrika who participates in a cakra conducted by one male
tantrika does not join any other cakra conducted by some other male
tantrika. In some cakras, the mental strength of the tantrika is put
to severe tests. A young sixteen-year-old girl is brought to the ritual,
who has all of the prescribed auspicious signs, and is pure and
holy in both her character and her mind. Her company will lead
the tantrika across the difficult path to perfection.19
Tantric householder wives are rare, and they tend to identify themselves as
basically traditional wives who are following religious teachings. This is be-
cause the sexual rituals and their accompanying yogic practices are performed
under the instructions of the guru and following her husband’s wishes. Some
practitioners have said that it is much better to have one’s wife as a partner, for
otherwise the woman may not be respected.
Professional Consorts
The fourth role includes female tantrikas who are professional ritual assistants,
who practice sexual rituals with male Shakta tantrikas. Such a role is a very low-
status one in Bengali culture, and nobody that I interviewed was willing or able
to tell me anything of such women. They were understood as having a specialty
within the profession of prostitution, as some women have skills in dominating
men (like Kali). It was rather like an addendum to the courtesan’s traditional
sixty-four arts, an extra set of skills which professional women could gain. I was
told that most of such women were low caste and wanted to gain extra money
for the household, or else they were widows (especially child widows) who had
no other way to make a living. Some informants actively condemned them, but
most pitied them.
This analysis was supported by Bholanath Bhattacharjee’s article ‘‘Some
Aspects of the Esoteric Cults of Consort Worship in Bengal: A Field Survey
Report.’’20 He interviewed forty-eight women who were professional ritual con-
sorts and gave detailed case histories for four of them. These women were called
either bhairavis (those involved in Shakta rituals) or sadhikas (those involved in
Vaishnava rituals). The term sadhika is a general term for holy woman or fe-
male religious practitioner, while the term bhairavi refers specifically to the con-
sort of a bhairava, a male Shaivite tantrika. A majority of the women followed
this profession as a family occupation; it was almost a caste, as the job was
handed down within the family and was hereditary. A minority were converted
to the profession by people whom they met. Almost all were initiated and given
new names.
Their tantric gurus taught the women breathing techniques and mantras to
lessen passion, as well as positions and muscular contractions to control the pace
of intercourse. These techniques allow the woman to be qualified to practice with
either one man for an extended period, or with a variety of men over time. For
those women following their hereditary role, these practices were understood to
be both service, helping the male tantrika to gain awareness of Brahman, and
following their own caste obligations, which brings spiritual advancement.
Many of these women learned this profession while they were young girls,
orphaned or without a father, from much older men. Bhattacharjee’s article
described four case histories. In the first case, an orphaned girl of seventeen
years met an elderly man who taught her these practices; she later practiced
with at least fifty other male practitioners in Birbhum. When her main bhairava
did not practice properly and she became pregnant, he left her. However, he later
came back, saying that if his guru liked practicing with her, he would take her
back and also pay her rent and support their baby. She stated that she was
forced by poverty to agree to this. The guru appreciated her abilities and asked
her to become his own consort and leave the other man (he also was willing to
support her and her baby). She stayed with him until his death and then lived
in a sexual relationship with a bhairava who came for shelter to her hut. She
trained her daughter as a consort, and the daughter learned the sexual rituals
from the bhairava with whom they were living.
In the second case described by Bhattacharjee, a widow was forced by
hunger to become a prostitute. She later learned tantric sexual skills from a
guru. She found living with the guru to be preferable to prostitution, as she did
not have to entertain many men, though she still called it hard work. Her guru
was almost twice her age. In the third case, the woman was born due to an
accident in her mother’s ritual practice with a bhairava. She felt that she was
under obligation to follow in her mother’s footsteps and also became a bhairavi.
She and her own guru initiate couples and teach them techniques of sexual
ritual in the same bed (though she seemed more concerned that they practice
‘‘even by day, if necessary’’). They are taught various rhythms and to think of
ocean waves and lions and tigers rather than snakes (which bring loss of con-
trol). She said that, in cases of problems in the practice (usually pregnancy), the
woman returns to prostitution.
In the fourth case, the consort was the daughter of a cook who was very
poor. The cook asked a bhairava (who was a relative of the household for which
she worked) to find a match for her daughter. He was much older than the girl,
and he decided to initiate her as his bhairavi. At the time of the article, she had
lived with him for nine years. The bhairava had also had ritual sex with the
girl’s mother.
These women have come to accept their roles as consorts as their lot in life.
Several were forced by poverty and hunger into these roles, and they needed
money to care for their children. Often, they were bound by considerations of
dharmic obligation; since their mothers were sadhikas, they too must follow
that profession. Many of these relationships are semi-incestuous, where the
man with whom the mother is sleeping is also sleeping with her daughter.
However, because it is placed in a ritual context, the father figure becomes the
guru, and the relationship is understood as a sort of religious apprenticeship.
While the consort role is often idealized in the West, representing freedom
and liberated sexuality, there are clearly problems for women in Hindu society
who are forced into the role and would themselves prefer a more traditional
life.
Celibate Wives and Widows

The fifth role is that of the celibate wife, who uses tantric mantras and visual-
ization for worship. The celibate wife remains in the household, has already
had children and wants no more, or has been celibate for the entire marriage
(as in the famous case of the saint Anandamayi Ma). The wife becomes a dev-
otee and leads a privately ascetic life. She spends most of her time in the
worship room before the deity’s image, while the husband acquiesces and stays
celibate as best he can. Sometimes, the wife may become a worship leader
( pujarini) for a group of other women, or the leader of a kirtan singing group.
In such cases, she gains a reputation as a holy woman, while the husband stays
in the background. Many husbands are quite amenable to their wives becoming
celibate devotees in later years.
We may also see tantric and devotional practices within the home as per-
formed by widows. The householder widow who spends her life in religious
ritual and pilgrimage may be respected or disparaged. I have seen Shakta re-
ligious widow matriarchs, who are called holy women by members of the ex-
tended family, who dominate both their households and the Brahmin priests
called in to perform rituals. They hold the keys to the household and the money
box, and thus have financial control—despite their renunciation. On the other
hand, I have also seen non-Shakta widows ignored, alone and unwanted, where
even their gurus look down on them.21
In one case in Calcutta, a priest was called in to perform rituals for Durga
puja by the matriarch of a large joint family. She had been initiated into one of
the Ramakrishna lineages, but continued to live in the family home. She was a
holy woman who lived a householder life, and she performed daily meditation
(involving tantric bija mantras but not tantric sexual rituals) and frequent pil-
grimage. The priest was ostensibly in charge of the rituals, but the householder
matriarch kept correcting him on his Sanskrit pronunciation and his actions,
and he acquiesced to her claims of proper ritual behavior. Thus, we have a
Brahmin male and a semi-householder female with different types of control
over ritual practice. I spoke later with some of her sons and grandsons, who all
agreed that she knew more about ritual than most of the priests in the area.
For all of the female tantric gurus interviewed, sexual ritual was under-
stood to be peripheral rather than central. It is a practice done for people who
cannot control their instincts, by people (generally male) who do not have the
necessary yogic discipline for real practice and need to take a few extra courses
to qualify for advanced practice. The issue is not that the sexually active tan-
trikas are sinful, but rather that they are weak and spending their time at the
lower end of practice rather than the higher end. Some female tantrikas im-
plied that men were generally weaker than women in this area and more com-
pulsive about needing sexual interaction. As Gauri Ma stated,
Most women have no need of sexual ritual [lata sadhana]. It is for
men, who are bound by lust, and need to overcome it. Then they take
a consort. In women, lust is not so strong. Tantric practice is medi-
tation [kriya] and worship [ puja].22
None said that lata sadhana was evil, or sinful, or scandalous. They did not
appear to be hiding their own secret practices. They simply said it was rare and
unnecessary. Some female tantrikas were more outspoken, saying that no man
was going to take away the power they had gained by hard austerities and long
recitation of mantras (several female tantrikas mentioned that sexual ritual
involved the loss of their power).
Does tantric ritual empower women in Hindu culture? It depends upon
the rituals performed and local understandings of these rituals. Women are
empowered by tantric practices that emphasize renunciation and asceticism,
especially meditation and the use of tantric mantras. Female tantric gurus have
positions of ritual authority usually possessed by males, including running ash-
rams, taking offerings, and advising householders on life decisions. Celibate
tantric holy women often have disciples who support them and whom they
bless. Celibate wives and widows may become respected worship leaders in the
community.
Female tantrikas who are not celibate are viewed with greater hesitation,
and often sexual ritual may act to disempower them. If they are married and
faithful to their husbands, tantra becomes a part of their wifely duty, helping
the husband in his spiritual development. Tantric ritual is then a part of do-
mestic female obligations. If the female tantrika is acting outside of marriage,
though, and performing sexual rituals with more than one man, mainstream
society is likely to look down upon her. She is generally pitied as a desperate
woman; she is not seen as a wise or compassionate guru. She may be under-
stood to be a prostitute, or a woman deserted by her husband who must feed
herself and her children, and she may be called either immoral or unfortunate.
In this case, tantric ritual disempowers her. The notion of tantric ritual as ‘‘free
love,’’ independent of marriage or commitment, is not appropriate to the
Bengali Shakta tantric tradition. In West Bengali society, tantric spiritual prac-
tices may sacralize a woman’s life and actions, or cause her to be rejected by the
community, depending on the type of ritual involved.
It is accepted in the tantric texts that women have a special aptitude for both
devotion and ritual practice. Both living in the home and leaving home can be
sacred acts for women. But women’s roles as renunciants are in direct conflict
with their obligations as childbearers; it is the classic opposition between dharma
and moksha (ordinary responsibility to the social order and the extraordinary state
of liberation), which produces a tension in the lives of all religious people in the
Hindu tradition, male or female. While ‘‘domestic tantra’’ exists, it is rare, and
women do not generally admit to it (even to family members). For most women,
domestic life and renunciant life are opposed, and one cannot simultaneously be
the domestic mother of the household and the tantric Mother of the Universe.
notes
1. Kularnava Tantra 7.47–49.
3. Kali Tantra 8.5–10.
4. Banerji 1988, 222–223.

5. Banerji 1988, 183.
6. Finn 1986, 95–96.
7. Banerji 1988, 261.
8. Banerji 1988, 262–263.
9. Literally navapuspa, or new flower.
10. Maya Tantra 12.1–8.
11. Kamakhya Tantra 4.35–37.
13. Banerji 1988, 184.
14. Interview, Gauri Ma, Bakreshwar, 1994.
15. Interview, Jayashri Ma, Birbhum, 1994.
16. Saradananda 1978, 220–227.
17. For further details of her life, see McDaniel 1989, 215–220.
18. Interview, Sahajiya informants, Birbhum, 1983.
19. Interview, Archanapuri Ma, Calcutta, 1994.
20. Bhattacharjee 1977, 385–397.
21. I have only seen this attitude of scorn toward widows among Vaishnava
groups, however. The Shakta widows I have personally seen were strong women and
respected by both families and teachers.
22. Interview, Gauri Ma, Bakreshwar, 1994.
references
Banerji, S. C. 1988. A Brief History of Tantra Literature. Calcutta: Naya Prokash.
Bhattacharjee, Bholanath. 1977. ‘‘Some Aspects of the Esoteric Cults of Consort
Worship in Bengal: A Field Survey Report.’’ Folklore 10 (December): 385–397.
Das, Upendrakumar, ed. 1976. Kularnava Tantra: Mula, Tika, O Banganubadsaha.
Calcutta: Nababharat.
Dasa, Jyotirlal, ed. 1978a. Kamakhya Tantram. Calcutta: Nababharat.
———. 1978b. Maya Tantra (Mula Samskrta O Banganubad Samet). Calcutta:
Nababharat.
Finn, Louise M., trans. 1986. The Kulacudamani Tantra and the Vamakesvara Tantra,
with the Jayaratha Commentary. Wiesbaden, Germany: Otto Harrassowitz.
McDaniel, June. 1989. The Madness of the Saints: Ecstatic Religion in Bengal. Chicago:
University of Chicago Press.
Saradananda, Swami. 1978. Sri Ramarkrishna: The Great Master. Translated by
Swami Jagadananda. Mylapore: Sri Ramakrishna Math.
Smrititirtha, Pandit Srinityananda, ed. 1981. Kali Tantram: Mula, Tippani O
Banganubadsaha. Calcutta: Nababharat.
10
Performing Arts, Re-forming
Rituals: Women and Social
Change in South India
Vasudha Narayanan
On Sunday mornings, at a time quite popular for television view-

ing, Jaya TV, a popular television channel in South India, presents
a show called Thaka dhimi tha. The title of the show refers to the
syllables used for rhythmic steps in Bharata Natyam, a form of clas-
sical South Indian dance. In the course of this well-watched show,
girls and women of varying ages perform classical Bharata Natyam
dance in an easily recognizable game show format and are jud-
ged in various competitive categories. In a typical show (e.g., 27
February 2005), three young girls around eight to ten years of age
were introduced by a teenaged girl, who seemed to be a dancer herself.
They then participated in various competitions and were judged—
based on the innovative and imaginative nature of their choreogra-
phy as well as for showcasing traditional dances—for a grand prize.
The dances were enacted against a background of a large floor-to-
ceiling illustration of a dancing Shiva known as Nataraja, the Lord
of the Dance. The dances included praise and adoration to various
deities, including Ganesha. There were short segments where the
young girls choreographed popular film music in the Bharata
Natyam style in the game show—something of which many purists
would not approve but which evidently make the ratings soar. At
the end of the program, the judge, a noted dancer herself, performed
a short piece to a song composed by Mira, the well-known sixteenth-
century woman poet. This song expressed Mira’s love and passion
for lord Krishna, and in the refrain she sang about the chiming of
her anklets to the music and dance. As the dancer emoted the poet’s
passion and swirled around the stage, the viewers were drawn into
the drama of the love among Mira the poet, Lakshmi the dancer, and Krishna;
the music and the lyrics, hauntingly familiar, recreated that arena of devotion
in a television program sponsored by corporate giants.
The program showcases a justifiable pride in classical dancing and Hindu
culture. It was conceived by Radhika Shurajit, a dancer and choreographer
who is making this classical Bharata Natyam form popular among children
and a large television audience, and the show was still running when this
book went to press in 2006. And yet it was only in the 1930s that this form of
dance was introduced from temple and private settings onto the ‘‘secular’’
stage and into community halls and auditoriums. Through the regularizing of
shows like Thaka dhimi tha, the dance is now reaching tens of thousands of
people, not just in South India, but in many parts of the world through various
satellite providers, such as Asianet.
While young boys are taking to this dance occasionally, it is primarily
women who are exponents of Bharata Natyam. According to some informal esti-
mates, about 5 percent of the performers are male; the most generous estimates
put it at 10 percent.1 The show, presented as a cultural program, takes for
granted the popularization of this form of dance and the accompanying music in
the mid-twentieth century. It takes place on a secular stage with multinational
corporations sponsoring the program; the decor of the stage, the lyrics, the music,
and the devotional themes of the dance intertwine with fusion music and some
popular film music to present an unusual combination of religion and art.
Are the program and the dances it showcases to be seen and experienced
as culture, religion, art, or secular entertainment? Dancers would probably say
it is all of these, and many academics consider the categories to be fluid with
porous boundaries. Within the Hindu traditions, religion, art, and culture often
merge seamlessly, and certainly programs such as these make us question the
artificial lines between disciplines in Western academia. But more important,
what we are looking at here is the transition of dance from temples, courts, and
private homes to the public sphere and the transition of what women consider
to be authority in the re-formation and the reconfiguration of rituals. When, in
the early twentieth century, there were massive efforts to reclaim or, as others
would say, appropriate classical Bharata Natyam dance from the exclusive prov-
ince of the temple dancers and some classes of women and make it available
for all people, there were deliberate moves to take it away from the temple milieu
to the secular stage. When that was accomplished, ironically, the secular stage
became transformed into a shrine replete with icons of the dancing Shiva or
Krishna, lighted lamps, incense, and flowers. The dance itself became an act of
ritual religious offering to the deity.
This chapter looks at ritual as performance and performance as ritual. It
explores the performing arts as ritual acts of worship and social resistance in
which women, as ritual agents, ‘‘reshape values and ideals that help mold
social identity’’ (Bell 1997, 82, 73). It is through being re-formers that women
performing arts, re-forming rituals 179
became performers in public forums; and women of various classes and cat-
egories ritualized their performances as worship. By revising and recreating
texts, these women use the performing arts as agents of social reform.
Although the move from temples and private courts in the houses of pa-
trons to the secular stage is significant in the history of twentieth-century
classical dance in India, the association of music and dance with temples (al-
beit somewhat romanticized) is still very strong in the Hindu consciousness.
In classical dance performances in the United States, one specifically sees a
nostalgia for the temple as a backdrop and as the frame for the dances. In 2001,
for instance, there was a scintillating dance performance, Saptapadi (‘‘seven
steps,’’ a term that ordinarily refers to the seven sacramental steps taken by the
bride and groom in a wedding ceremony), in which several prominent dancers
of the metropolitan Atlanta area participated. The choreographer of this pro-
gram, Dr. Seshu Sharma, a local physician with considerable involvement in
the performing arts, envisioned the dances as a ritual offering in the context of
Indian temples and wanted them to be framed with slides of Hindu temples
and architecture from India. Each dance was preceded by slides of one relevant
temple in India where that particular style of dance flourished or where the
deity depicted in the dance was worshipped. Then, as the priests from the local
Atlanta temple recited Sanskrit verses to the deity who would be praised in the
dance, the dancer entered the stage against the backdrop of a slide of the Indian
temple. It is this framing of classical Indian dance today by a romanticized
temple culture that is striking. This format was repeated in December 2004 for
a similar mega dance program called Sivoham Sivoham2 (‘‘I am Shiva, I am
Shiva’’). Both dance programs were organized by the Hindu Temple of Atlanta
but performed in the sumptuous auditorium of Georgia Tech University. They
portrayed the dances as a significant expression of Hindu religiosity and larger
culture and specifically portrayed dance and music as ritual expressions of
worship. While all of the dancers in Saptapadi were women, there were one or
two male dancers in Sivoham Sivoham.
This chapter explores the expression of women’s religiosity through clas-
sical music and dance in public forums in South India. Straddling the fuzzy
boundaries among religion, culture, and entertainment, we will look at women
who serve as exemplars of devotion—devotion for the arts and devotion for the
deity—as well as at what they considered to be authority in the course of this re-
formation of the classical performing arts such that it became accessible to
women of many castes and classes. We will also look at those women who now
use these art forms as vehicles to raise social consciousness. Parts of this chap-
ter are, in a sense, a flashback which will explain how the dance milieu de-
scribed in the TV game show Thaka dhimi tha and the temple framework for
Saptapadi and Sivoham Sivoham are recalled and reconfigured from the past and
reframed for dancers in the present. We will initially discuss the importance
of music and dance in the larger South Indian Hindu culture and, against
this background, look more closely at three topics: the role of a ninth-century
woman poet-saint in inspiring public expressions of women’s religiosity in the
twentieth century; the efforts of three women in mainstreaming classical mu-
sic and dance in South Indian culture; and finally the movement by some
dancers, like Mallika Sarabhai, in using this dance form to raise people’s
awareness of social issues, including the status of women.
The inclusion of the fields of music and dance underscores more than
creating the opportunity for the entry of women into the realm of the arts; it
acknowledges the efforts of women who made it possible for others to publicly
express their devotion and spiritual longing in ways that were not available to
them earlier. But while this freedom to publicly portray devotion through the
performing arts is seen as laudable, we also raise a question in the end: did this
opening up of the performing arts come at the expense of taking it away from
those women, the devadasis, the servants of god, who had safeguarded it for
centuries? Was it a creation of new ritual space for women or an appropriation
of dance forms by high-class women from the courtesans who came to be in-
creasingly marginalized during the colonial era? Before we discuss these is-
sues, let us look first at the position of music and dance in India.
Social Values and Perceptions of Performing Arts
Hindu women—as women in many other cultures—have been both empow-

ered and subjugated by religious traditions over the centuries. Some of the
dancers loosely gathered under the umbrella term devadasi had agency and
control over their property, on the one hand, but were also objects of ‘‘enjoy-
ment’’ (bhogam), on the other. The contrast drawn in literature—even in texts
such as the Kama Sutra and the Tamil work Silappatikaram—is between the
decorum of wives, who were dependent on their husbands, and the activities of
courtesans, who did not marry and were independent agents in many ways.
The Brahminical Hindu tradition has been marked by the curtailing of free-
dom to women in the spaces connected with dharmic roles. And yet, it may
also be argued by some that women in Hindu traditions—either because of
or in spite of gender, caste, and class restrictions—have continuously found
opportunities to create new spaces for artistic expression and paths to salva-
tion. One can also note that the curtailment they faced in many social spheres
was paralleled by a lack of similar opportunities for men; in other words,
gender was only one factor among many in determining how one could have
the life one wanted to live. It is in this context that we can understand the
attitudes toward the performance of classical music and dance by women.
A popular Sanskrit saying ascribed to various sources expresses clearly that
the term sangita, usually translated as ‘‘music’’ in English, really includes vocal
and instrumental music as well as dance (gitam vadyam tatha nrityam trayam
sanitam uchyate). Music and dance are intertwined in the Hindu traditions, and
in this chapter, we will deal with both.
The place of music and dance and of the musicians and dancers in preco-
lonial and colonial India is a very complicated topic. There were, of course, many
kinds of classical and popular music and dance performed at homes and in
temples. The received narrative tradition today is that classical dance was safe-
guarded by the devadasis, or temple dancers, and that the temple dancers who
practiced erotic arts as well as the dance form itself were ‘‘emancipated’’ by the
British overlords and several social reform movements. The story from the
viewpoint of the devadasis, ritual specialists at the temples and in some domestic
events as well, is different; they claim that they did not need any emancipation
and that their arts were appropriated, indeed, hijacked by the social elite.
Music and dance were performed in many forums in precolonial and co-
lonial India by women from many classes. While classical vocal music was
probably part of domestic rituals for upper-class women, playing instruments
and performing classical dances in public were within the province of deva-
dasis. Devadasi is a shorthand term for a variety of practitioners, not all of whom
were either dancers or courtesans or performed in temples. Some of them were
dedicated to the temple, some served at the local ruler’s pleasure, some were
simply born into matrilineal families associated with these arts. Dancing was
associated with courts, temples, and halls of rich patrons; in some areas, there
were different classes of dancers for each of these arenas, while in other re-
gions, the same dancers participated in all three. There were also many cate-
gories of lifestyles within these streams, and women associated with the temples
were ritual specialists with very specific roles to play in daily, periodic, and an-
nual events. They were paid largely through endowments made to the temples.
A few women were long-term companions of patrons. In the colonial era,
however, they came to be generically portrayed as women of loose morals, or
even as prostitutes. The devadasis sang songs of love and praise to both the deity
and the patron. The music and dance frequently contained double entendres—
the words could refer to the king or the deity, and the emotions could be con-
strued as kama/sringara (physical, sensual love) or bhakti (devotion).
Since many of the colonizing forces in India had dominant anti-dance
ideologies, the attitudes toward the dancers and dance in the nineteenth and
twentieth centuries have been lamentably negative. The movement to make
classical dance available to girls of all castes was not just a struggle against high-
caste male resistance or a culture where there were some patriarchal power
structures in place, but also against the dominant colonial ideology. In the early
part of the twentieth century, the dance known as dasi attam (the dance of
temple servants or courtesans) and sadir (from the Urdu word sadara, meaning
dance) was renamed Bharata Natyam, the dance according to the sage Bharata.
Since the indigenous name for India was the ‘‘land of Bharata’’ (after another
legendary Bharata), the renaming itself had nationalistic nuances.
Acting, music, and dance have been considered to be some of the optional
ways to salvation within Hinduism. The treatise on theater and dance, the Natya
Sastra, written by a legendary person called Bharata, is considered to be the fifth
Veda or scripture. Dance is said to involve a total control of the body, a control
central to the physical discipline of yoga. The Natya Sastra is supposed to have
originated with a divine being called Brahma and was then passed on to the
sage Bharata. Brahma is said to have taken the reading text from the Rig Veda,
the earliest and one of the most sacred texts in the history of Hindu literature,
music from the Sama Veda, gestures from the Yajur Veda, and rasa (lit. essence
or juice), that is, the aesthetic element, from the Atharava Veda. Combining
the essence of the four Vedas, he compiled a fifth Veda that depicts dance as a
way to salvation. In oral tradition, the very name Bharata is said to incorporate
the main elements of Bharata Natyam: Bha- stands for bhava, a state of mind,
an attitude; -ra for raga; and -ta for tala, or rhythm. In addition to this, music and
dance were known as the Gandharva Veda.
One can get liberation either through dancing or by being overwhelmed by
the joy which comes by witnessing the dance. This could be any dance; theo-
retically speaking, all dance is divine. Thus, in classical Bharata Natyam, there
are ‘‘pure’’ dances or dances without any particular lyrics and emotions attached
to them. Many of the dances, however, are devotional in tone. At times, the
dancer expresses devotion through his/her body and soul to get liberation. This
is particularly seen in the Bharata Natyam style of South Indian dancing,
where the dancer may express her love for the Lord in explicitly erotic termi-
nology. The pining of a human soul for union with the Lord is expressed
through passionate longing and a desire to belong to him. The audience is also
granted salvation through participation in the divine joy of movement. Either
by dancing or by beholding the beauty of the divine dance—whether it is that of
Krishna with his cowherd friends, or Shiva, known as Nataraja, or the King of
the Dance—one obtains liberation.
Sanskrit texts on dance such as the Natya Sastra usually make a distinction
between classical and folk dances. These categories have not been immutable;
sometimes, the boundaries have been fluid, and both forms have derived inspi-
ration from the other. People from various communities would perform those
dances that were part of their traditions both at home and in public—in village
streets, on festival days, for example—but classical dancing was regulated. A
striking example of public singing and dancing (though originally even these
were contained within homes) is the garbha dance and worship in the state of
Gujarat. This is frequently classified as folk dancing, but as in any other tradi-
tion, one cannot draw hard lines between classical and popular dance forms: they
are in dialectical relationship with each other. Garbha or garbhi is ‘‘womb,’’ the
source of all creative energy; it is the Mother Goddess who is present in the lamp
inside the clay pot that is called garbha. Women and young girls dance around
the garbha all night, celebrating the goddess. It seems that when the focus of dis-
cussion is moksha (liberation), rather than dharma (issues of righteousness) in

this world, a greater freedom is seen in women’s participation in public spaces.
Androcentric regulation of women’s modesty or actions in public, which were
incumbent on some castes and classes, is simply bypassed in cases where the
tradition focuses on the salvation potential of all human beings.
The ostensible reason for the ‘‘revelation’’ or composition of the treatise on
dance is said to be that of making accessible the difficult statements of the
Vedas to all human beings. However, it is interesting to note that at least after
the fourteenth and fifteenth centuries (or possibly earlier), while we have doc-
umentation of male singers (of many castes) and dancers, the only women who
sang and danced in public seem to have been courtesans. This was apparently
not the case in earlier centuries; even as late as the twelfth century, we see
sculptures of women dancers adorning the niches of the Belur temple in
Karnataka. Shantala Devi, a Hoysala queen, is said to have been responsible for
building this temple. It is also said that one of the dancing figures within the
temple is a figure of the queen herself. According to one popular account, her
dance was admired by the king and the courtiers, and ‘‘the king as desired
by the courtiers gave her publicly the titles of ’Natyasarasvati‘ [the sarasvati
or goddess of dance] and ‘Sakala samaya rakshamani’ (Imperial protector of
all faiths)’’ (Nagappachar 1960, 7).3 The human body is celebrated in these
sculptures, and the later restrictions against women from ‘‘decent families’’
dancing in public may have come from a chronologically evolving coalescence
of conservative attitudes explicit in many Hindu dharmic texts, Islamic mores
in India, and the puritanical perspectives of European missionaries.
It is possible that with new attitudes toward the human body (attitudes
that reinforce some of the negative statements made in some Dharma Shastras
about a woman’s uncontrollable sexual propensity), perspectives on the overtly
sensual presentations involved in the performing arts also changed. Many
devotional lyrics, after all, use the rapturous love metaphor in constructing the
relationship between the human being and God. The singer or dancer dis-
played these sensual bhavas, or expressions, either with the face or with the en-
tire body in the case of the dancer, and through voice modulation in the case of
the singer. While the allegorical meaning of these poems clearly indicated the
love to be spiritual, metaphors pertaining to the body were prominently used.
Upper-class men would have thought that it was inappropriate for women
to display their bodies in public. Women did continue to dance and sing, but
it seems to have been within the confines of the home, without a choice of
whether they wanted to do this in public. It is possible that women did par-
ticipate in devotional singing through the streets in some medieval traditions,
especially those associated with the movement of Caitanya, but even this is
debated and was probably more of an exception.
Thus, while song and dance have been considered to be significant ways
to salvation, until the mid-twentieth century, ordinarily only men and female
courtesans performed in public. Although music and dance were theoreti-

cally considered to be sacred, in time their beauty came to be associated with
physical sensuality and the performance of courtesans. It was considered inap-
propriate for women from decent families and the so-called higher castes to
perform in public. We see this illustrated in the life of Sister Subbalakshmi
(1886–1969), a child widow (not to be confused with M. S. Subbulakshmi,
1916–2004, the famous Carnatic music singer of South India), who worked
hard for the rights of young widows to be educated in schools and universities.
Monica Felton, her biographer, records that when young, Subbalakshmi was
taught to play the violin. This was an act that brought about the ire of her
tradition-bound grandmother. Felton writes:
It was not, and he knew it, the extravagance of buying such a thing
which shocked Grandmother. She had been far from pleased by the
decision to give Subbalakshmi an education of a kind that was re-
ally only suitable for boys. . . . These things had been bad enough.
The violin was really the last straw. Ladies might sometimes sing,
though only in the privacy of their own homes. Even that was very
unusual, and no lady would ever, in any circumstances whatever,
learn to play a musical instrument. Music was for dancing girls, deva-
dasis, as the temple prostitutes were called. Surely Subramania, who
came from such a good family, could not really intend to engage a
teacher and allow his widowed daughter, of all people, to take lessons?
(Felton 1967, 38)
Subbalakshmi’s parents taught her to play not only the violin, but also the veena,
a traditional South Indian instrument. Subbalakshmi eventually became an
educator and also provided the infrastructure (a boardinghouse and regu-
lar meals cooked by Subbalakshmi’s aunt, herself a widow) to facilitate young
widows getting a solid public education.4 It was due in part to the efforts of
Subbalakshmi’s family and the work of other women like Bangalore Nagar-
atnammal, Balasaraswati, and Rukmini Devi Arundale (whom we will en-
counter shortly) that the fields of education and performing arts opened up for
Hindu women in the twentieth century.
In southern India, the dynamic creativity in the production and perfor-
mance of classical song and dance was largely seen in the lives of men be-
longing to several castes, including Brahmins, and devadasi families. The annals
of the Tanjore court in the eighteenth and nineteenth centuries are filled with
accounts of prominent male singers/musicians and of courtesans (Seetha
1981, 35, 98, 122–126).5 Devadasis performed these arts in temples and were
also ritual specialists. There were also other courtesans not necessarily dedi-
cated to temples, patronized by royalty and nobility, who were involved with the
performing arts. Courtesans were generally the only women, apart from roy-
alty, to have access to textual scholarship and the arts of singing and dancing
for many centuries. They also held the rights to inherit property and to adopt
children. But devadasis have never served as role models except within their
own communities in South Indian society. Although they expressed their bhakti
through performing arts, the dominant Kalakshetra school of dance and sev-
eral other dance schools do not look upon them as sources of authority; rather,
they appeal directly to texts such as the Natya Sastra to derive ‘‘authenticity’’ for
their dances.
In the twentieth century, the coming together of nationalistic concerns, an
increased consciousness of premodern women poets, a new pride in Indian
classical performing arts, and the new public spaces created for expressions of
devotion has created a veritable boom in classical music and dance.
One stream that feeds into this appropriation/expression of music and
dance is the growing awareness of women poets who lived more than
a thousand years ago. While there are records of dozens of such poets all over
India—women who composed in local languages such as Tamil—most of
them are not known outside a small region. The poets express a passionate
desire to unite physically and spiritually with the deity, and this passion is not
unlike the songs sung by the devadasis.6 However, the women’s groups that
sing devotional songs look to the women poets, not the devadasis, as role
models. In other words, while the passion for the deity is the same in both
sets of songs—those composed by the women poets and those sung by the
devadasis—women in the twentieth century look to the poets as the authority,
not the devadasis. And although literally hundreds of classical dancers today
choreograph and dance Andal’s songs, the authority for the performances
becomes Andal, on the one hand, and the remote Natya Sastra, on the other,
and not the devadasis from whom the techniques of the dance style are directly
derived. The exception, to some extent, is the school founded by Balasar-
aswati.
Let us now turn to the question of authority and how women have en-
ergized classical music and dance and re-formulated them so they can be ac-
cessible to women of many castes. We will discuss the impact of the poet Andal
as a role model in Tamilnadu and also explore the contributions of three other
women, Bangalore Nagaratnammal, Balasaraswati, and Rukmini Devi Ar-
undale, who made a difference in the public performance of music and dance.
In the twentieth century, these three women were instrumental in bringing
South Indian classical Carnatic music and Bharata Natyam to the public stage
and making the performing arts accessible—for performance, audience par-
ticipation, and viewing by the general public. They thus helped to transform
the ritual form of the performing arts. Ironically, Bangalore Nagaratnammal
is said to have been the daughter of a courtesan, and Balasaraswati also hailed
from a devadasi family. Rukmini Devi Arundale, on the other hand, learned
dance from a devadasi but worked to bring dance out into the public and secular
sphere.
The Goda Mandali, or the Circle of Andal
Since the 1970s, many groups of women in the suburbs of Madras, a city
in South India, have formed circles (mandali) to sing the poems of Andal
and other Vaishnavaite Tamil saints. These groups look directly to Andal, an
eighth-century poet, as the authority for their bhakti. The groups in Chennai
have interesting names: Goda Mandali (circle of Andal; Goda is another name
for Andal), Sreyas Mandali (circle [leading] to the higher path), Ranganatha
Paduka Mandali (circle that venerates the sandals of Ranganatha [Vishnu]),
Subhasri Mandali (very auspicious circle), and Bhaktanjali (worshipful circle
of adoration). The groups meet once or twice a week to learn devotional songs
from a teacher and also perform in music fests (like the Tamil Music Asso-
ciation festival in December) and temples. Sometimes, they raise money for
the upkeep of various shrines through their singing. In the last few years, the
Goda Mandali located in Thyagarayanagar, Madras, has used the medium of
singing to reappropriate and participate in the passion of Andal/Goda.
Goda Mandali, appropriately named after one of the earliest woman poets
dedicated to Vishnu, was first formed in 1970 and reorganized in 1982. It has
stopped meeting regularly since about the year 2000 after the tragic murder of
one of the members, but there are periodic moves to revive it. The women of this
group frequently sang for events connected with temples, and occasionally, they
sang for a secular audience. The piety of Goda Mandali has also achieved con-
siderable visibility through television and radio, leading to the formation of more
circles, through which the passion of Andal lives again. All practice sessions and
performances begin with a unique invocation. In the Sri Vaishnava community,
recitation of the songs of the Tamil saints generally begins with the line ‘‘I take
refuge with the sacred feet of Nammalvar [a male poet-saint who lived in the
ninth century] and Emperumanar [Ramanuja, a male theologian of the twelfth
century c.e.].’’ Goda Mandali, however, begins with the line ‘‘I take refuge with
the sacred feet of Nammalvar, Andal, and Ramanuja; I take refuge with the sacred
feet of Andal.’’ This special emphasis on Andal is unique to this circle of her
companions in Goda Mandali and in some of the other women’s groups. Goda
Mandali also has a unique mangalam, or last song. In almost all classical South
Indian (Carnatic) music concerts, the last ‘‘auspicious’’ song focuses (by tradition)
on the male god Rama, and either a song of Tyagaraja (eighteenth century) or
Ramdas (seventeenth century) is sung. Goda Mandali, however, sings a song
praising the goddess Lakshmi in her manifestation in the temple of Srirangam.
There are also dozens of dancers who choreograph and dance Andal’s songs.
Srinidhi Rangarajan, a physician and noted Bharata Natyam dancer, in Madras,
Tamilnadu, rearranged the verses of Tiruppavai (a song composed by Andal) to
reflect her experience of a spiritually progressive sequence and choreographed it.
This performance was held in December 1994 to inaugurate the Festival of
Tiruppavai. Following the performance of Dr. Rangarajan, students and teachers

from a large school in Madras went around the streets in a lighted float, singing
the verses of Andal. This float was subsequently taken out at dawn for several
days in the month of Markali (December 15–January 13, a month that is con-
sidered to be sacred to Andal) and traveled through the city of Madras. These new
rituals highlighting women’s piety and public role in religious expression are
becoming increasingly common; their popularity is attested to by the fact that the
Festival of the Tiruppavai, the dance by Dr. Rangarajan, and the floats were all
sponsored by the Indian Bank, a leading commercial institution in India.
And who is this Andal, the role model and authority for these women sing-
ers and dancers? She was an eighth-century poet-devotee who sought the deity
Vishnu as her beloved.
Andal, an Eighth-Century Devotee, and the Performing Arts
Normative texts and oral tradition in some Hindu communities held that
revering one’s husband as a god was enough to attain liberation. While there
are several narratives to exemplify this ideal, there is no indication that many
women took this notion literally. Andal did not want to get married and con-
sider her husband as a god; she wanted God as her husband. Andal is the only
woman poet-saint venerated by the Sri Vaishnava community of South India.
In two poems, the Tiruppavai (‘‘Sacred Lady’’) and the Nachiyar Tirumoli (‘‘The
Sacred Words of the Lady’’), she expresses her passionate desire to marry lord
Vishnu, whom she sees in the forms of Krishna or Rama (incarnations of
Vishnu), or as the resident deity in the temple. In the Tiruppavai, she imagines
herself as a young gopi, or cowherd girl, in the village where Krishna grew up.
She wakes up her friends very early in the morning in the month of Markali,
and they all go to wake up Krishna and give him their petition: their longing to
be with him and to serve him for all time. Andal asks her friends to come with
her to ‘‘bathe’’; the Tamil word niratal or bathing, was frequently used to in-
dicate a sexual union in Tamil literature. Since Vishnu is compared to a lotus
pond frequently, Andal’s intentions would be clear to those who recite the
poem, and if they are not, the commentators articulate this concept quite well.
Andal’s life offers not just a theological model for all human beings, but a
model of love that women may use in approaching God. Women see in her a
person who attained salvation not by worshipping her husband as a deity (as
Manu would have it), but by approaching the deity directly and seeking union
with him. What Andal and other women poets did by living the way they did
was to negotiate a space within a marriage-dominated society and force at least
some sections of the society to make room for them. The typology of women
that we see in the law books (Dharma Sastras) are young unmarried girls,
married women, and widows: single women who either reject marriage or who
walk out on their husbands are not even recognized. But by the power of their
love and the consolidation of their aspirations, the women poets carve out a
different space for themselves.
Andal’s songs have been recited daily in domestic and temple liturgies,
and her lyrics have been set to ragas and performed on sacred and secular
stages. Her songs have been part of the repertoire for all classical Bharata
Natyam dancers since the 1970s. More important, Andal, Mira, and other
women poets have become the focus of inspiration and the authority for bhakti
for many women’s groups in India.
How do these women who dance and sing Andal’s songs relate to Andal?
In Sri Vaishnava theology, Andal is clearly a paradigmatic devotee and by
identifying with or imitating her love, it may be possible to reach Vishnu.
Parasara Bhattar, a twelfth-century theologian, is quoted as advising a male
disciple that one ought to recite the entire Tiruppavai every morning and
experience its emotions. If it is not possible to recite at least one verse, he is
advised to just think of the way he has experienced the joy of the verses. In
other words, the experiencing of the path taken by Andal, however vicariously,
is highly recommended for both men and women. Commentaries say that by
the practice of some rituals, the cowherd girls got Krishna; by the imitation of
those rituals, Andal reached Krishna. By ritually making the words of Andal
her words, the devotee is extolled to be like Andal, imitating her passion,
emulating and appropriating her devotion (Periyavaccan Pillai 1974, 202–203).
In other words, one seeks union with Vishnu through the words of Andal and
by sharing her passion and her power.
The women of Goda Mandali see Andal’s words and the words of the other
saints as portraying their own emotions and spiritual longing and as ritually
giving them direct access to the Lord. Chitra Raghunathan, one of the mem-
bers, says that singing with this group is the most fulfilling aspect of her re-
ligious life: in fact, she added, this is her only real religious life; this is her
direct prayer, her ritual. In temples, special groups (like the adhyapakas and
araiyars, the traditional male cantors) have the religious and often the legal
right to recite and pray; in household rituals, a man may officiate; but through
singing and dancing, the women communicate directly with the deity. In this
context, two factors are important: the companionship of the group, which is
like the collective prayer of Andal and her friends, and second, the opportunity
to sing the sacred words in private and in public forums. The growing pop-
ularity of these women’s groups, which sing the prayers in sacred and secular
forums, is leading to an increase in the transformation of ritual patterns in
some Hindu communities.
Hundreds of women’s groups like Goda Mandali have become popular
all over the world largely due to the pioneering efforts of Bangalore Nagar-
atnammal, who helped to create a public space where women could partici-
pate in singing classical music. Classical Carnatic music in South India is almost
completely devotional in tone, and the lyrics explicitly express one’s bhakti to a
deity.
Bangalore Nagaratnammal
Bangalore Nagaratnammal (1878–1952) was the daughter of a courtesan-singer

in the royal court of Mysore. Coming from a community that was low in social
status from certain Brahminical perspectives, it was nevertheless one that
accorded young women the privilege of education of every kind. A celebrated
singer and musician, Nagaratnammal was responsible for instituting a music
fest in honor of the male composer Tyagaraja (1767–1847) in the town of
Tiruvaiyaru. This festival was instituted to encourage all singers—male, fe-
male, famous, and unknown—to participate equally, with equal time and
opportunity for all to celebrate the piety of Tyagaraja’s music.7 This was more
than a music festival; the music itself is considered to be a form of worship and
is also accompanied by religious rituals, extolling the musical composer and,
now, Bangalore Nagaratnammal.
Nagaratnammal’s authority for institutionalizing this festival of music was
not derived from the devadasi community; rather, she appealed to the authority
of dreams. She was apparently led by dreams to Tiruvaiyaru, a small village
in Tamilnadu where the composer Tyagaraja had lived. She cleared the land
where he had been buried, tirelessly collected funds to build a temple nearby,
and then instituted the music festival. She befriended and encouraged younger
women to sing and perform, singing to celebrate the deity and to give full ex-
pression to their talents and fervor. Bangalore Nagaratnammal is to be remem-
bered for creating the public space for women and men to participate in a
festival where one can sing the devotional lyrics of Tyagaraja and participate
both in a choral and individual expression of reverence.
The Tyagaraja Utsava (Festival of Tyagaraja) was and remains a grand
vision and celebration, and it has grown to be one of the best-known music
festivals in India and for Hindus in the diaspora. Tyagaraja Utsavas are cele-
brated with great eclat not just in Tiruvaiyaru, but now in every major city in the
United States and Canada with the most prominent one being in Cleveland,
Ohio. The lead singers in many of these festivals in India and abroad are now
women musicians. The establishment of this public, devotional forum for the
display of talent—of both men and women—is due to the efforts of Bangalore
Nagaratnammal. While during the early part of Bangalore Nagaratnammal’s
life, Carnatic music was not sung in public by high-caste women, they are now
conspicuous by their presence.
The opening up of classical dance and music to men and women of the
‘‘higher castes’’ is an important development in the twentieth century. While
this is particularly true of Carnatic music, it is also true of the performing arts
all over India. It was only in the twentieth century that the pioneering efforts of
Balasaraswati and Rukmini Devi Arundale resulted in making Bharata Natyam
accessible to women of all castes.
Balasaraswati and Rukmini Devi Arundale
While Vallathol in Kerala and Rabindranath Tagore—two men—worked for the

revival of classical dance forms in Kerala and in Bengal, the classical Bharata
Natyam dance became popular in Tamilnadu with the efforts of both a man
(E. Krishna Iyer) and two women, Balasaraswati (1918–1984) and Rukmini Devi
(1904–1986). Balasarawati’s impact on the field of dance was tremendous:
No dancer captured the public imagination as Balasaraswati did
in the thirties and forties. Bharatanatyam to the public until then
was an esoteric art practised for the pleasure of a few connois-
seurs and dilettantes. Outside the pale of the temple and court
there were hardly any public performances. Balasaraswati made
the public aware of Bharatanatyam, not by deliberate efforts as a
reformer, but by the beauty and the eloquence of her dancing. It
was left to others to fight prejudices and stupidity, do research,
delve into the past, give the dancer’s profession respectability and
so on.8
Narayana Menon’s account of Balasaraswati makes an important point—that
she made this classical dance form popular, ‘‘not by deliberate efforts as a
reformer, but by the beauty and the eloquence of her dancing.’’ This is, per-
haps, the primary difference between her efforts and those of Rukmini Devi.
Balasaraswati came from a family of courtesans and Rukmini Devi from
a Brahmin family. Balasaraswati learned dance from the time she could walk
and had her stage debut (arangetram) when she was seven years old; Rukmini
Devi learned Western dance first, was a friend of Anna Pavlova and learned
Bharata Natyam when she was much older from Gowri Ammal, a devadasi, and
had her stage debut when she was in her mid-thirties. Balasaraswati sought
the authority for her style of dance and the traditional sequence of Bharata
Natyam recitals from her own family heritage; Rukmini Devi from Sanskrit,
pan-Indian sources.
Biographies of Balasaraswati frequently speak of her ancestors and her
place in a lineage of people who had dedicated themselves to the arts. She was
proud of this heritage and fought to preserve and transmit this authenticity of
the received tradition. For her, this tradition was local and the authority came
from her foremothers. Rukmini Devi’s school of dance, on the other hand,
emphasized the pan-Indian heritage of this dance form with a serious study
of the Natya Sastra composed by Bharata and also the Abhinaya Darpana
(ca. fourth through sixth centuries), a Sanskrit text attributed to Nandikeswara

(O’Shea 1998).
Both Balasaraswati and Rukmini Devi agreed firmly that the Bharata
Natyam was a religious art; we find this over and over again in their speeches
and talks. Even when a piece was dedicated to a human patron, Balasaraswati
explained that it ultimately addressed the supreme being:
The composer of a Sabdam or a Varnam might have dedicated it
to a prince or a noble man. But as far as the dancer is concerned,
the hero can only be the King of Kings, the Lord of the wide world.
It is impossible for her to dedicate her art, which has sanctified her
body and has made her heart sacred, to a mere mortal. She can
experience and communicate the sacred in what appears to be sec-
ular. After all, our composers have been steeped in the tradition of
bhakti.9
While both dancers spoke of dance as a religious exercise, Balasaraswati is
credited with the saying that sringara is the highest form of expression in
Bharata Natyam (O’Shea 1998). Sringara is romantic love; some translate it as
passion or sexual love. Balasaraswati firmly thought—as we can see from the
following passage—that this was not carnal. However, Rukmini Devi seems to
have thought that the erotic elements should be removed from Bharata Natyam
and that passion should be sublimated and expressed as bhakti. Balasaraswati
did not think that Bharata Natyam had to be ‘‘purified’’ in any way:
It is this stream of sringara that swells into the mighty river of the
lover-beloved songs of the Vaishnava and Saiva saints, the Ashtapadi
of Jayadeva and the compositions of Kshetragna. In Bharatanatyam,
too when it comes to abhinaya, sringara has been the dominant mood.
NOT CARNAL
I emphasize all this because of some who seek to ‘‘purify’’ Bhar-
atanatyam by replacing the traditional lyrics, which express sringara
with devotional songs. I respectfully submit to such protagonists that
there is nothing in Bharatanatyam which can be purified afresh; it is
divine as it is and innately so. The sringara we experience in Bhara-
tanatyam is never carnal; never, never. For those who have yielded
themselves to its discipline line with total dedication, dance, like music
is the practice of the Presence. It cannot be merely the body’s rapture.
Bharatanatyam is an art, which consecrates the body, which is
considered to be in itself of no value. The yogi, by controlling his
breath and by modifying his body, acquires the halo of sanctity. Even
so, the dancer, who dissolves her identity in rhythm and music,
makes her body an instrument, at least for the duration of the dance,
for the experience and expression of the spirit.10
Even though Balasaraswati emphasized the religious and spiritual nature of

dance, her critique of the attempts to purify—that is, get rid of—the sringara, or
the passion in the dance, made her school distinct from that of Rukmini Devi.
It was by the beauty and the aesthetic nature of her dance and by dancing
in public that she popularized it. Rukmini Devi, on the other hand, started
a school with a regular curriculum. As noted earlier, she appealed to the au-
thority of Natya Sastra and other texts to frame the dance form as classical and
all-Indian.
Rukmini Devi Arundale, dancer, educator, and nominee to be president of
India in 1977, was born in a Brahmin family and is best known for her es-
tablishment of a school of classical music and dance. Here, the dancers not
only learn from teachers, but see their art as having a continuous connection
with the Sanskrit texts on dance composed centuries ago. Rukmini Devi’s par-
ents belonged to the progressive theosophist movement and were closely as-
sociated with Annie Besant, the Theosophy leader. Rukmini Devi herself broke
with orthodox tradition in many ways. When she was sixteen, she married an
Englishman. She later learned ballet from Western teachers and Bharata Na-
tyam from Mylapore Gowri, an eminent dancer from a traditional family of
courtesans. In 1935, Rukmini Devi gave a public recital of dancing ‘‘in the
midst of dissent and fury’’ (Rajagopalan 1990a, 256). The storm of protest was
evidently against the idea of a ‘‘high-caste’’ woman giving a public dance per-
formance. Rukmini Devi persisted in her efforts and started the Arena of Art
(Kalakshetra), an international academy of the arts in 1936. Classical music
and many kinds of South Indian classical dance were taught there for all those
who desired to find fulfillment through these artistic forms. It is hard to imag-
ine that Bharata Natyam, a field that is dominated by Brahmin women and
girls today, was completely forbidden to them until late into the twentieth
century.
Social Change and Dance
The performing arts now serve as vehicles for further dynamic reform in a way
that is unparalleled within the Hindu tradition. The fight to render erotic/
spiritual longing in artistic ritual form in public forums was a major landmark
of the twentieth century. The performing arts are now used with skill by cele-
brated dancers like Mallika Sarabhai and Chandralekha to express themes of
anguish and strength that pertain to women’s issues. What is striking is that
unlike some of the women mentioned above, who looked at earlier paradig-
matic poets, dreams, or Sanskrit texts for inspiration and as a validation of their
innovation, people like Dr. Sarabhai do not necessarily retrieve authority fig-
ures or texts from the past to mount their challenges.
Mallika Sarabhai’s new school of dance, dedicated to her mother, is called

Natarani, or Queen of Dance. This is a deliberate change from the name
Nataraja, or King of the Dance, a popular name of lord Shiva, also known as
the Lord of the Dance. Mallika Sarabhai has utilized the forms of mime and
dance to underscore some of the injustices done to women. She initially set the
feminist agenda through ‘‘Shakti: The Power of Women’’ and has contin-
ued explorations of feminist issues through her recital ‘‘Sita’s Daughters.’’
‘‘Shakti’’ is a solo performance in which Dr. Sarabhai portrays women whose
power has been negated or neutralized by male society.
‘‘Sita’s Daughters’’ starts with a focus on an episode from the Ramayana,
an epic which is probably the best-known story in the Hindu tradition. At the
end of the story, Prince Rama, who is regarded as an incarnation of lord
Vishnu, questions his wife, Sita, about her chastity. In the last segments of the
epic, Rama banishes the pregnant Sita. Several years later, when he meets his
twin sons and Sita, he says he will accept her if she can give him some ‘‘proof’’
that she is chaste. Sita proves this by an ‘‘act of truth.’’ This is an act by which a
person swears that, if something is true (for example, her love for a particular
person), then a miracle will happen. Sita now swears and says that, if it is true
that she has always been faithful to Rama, Mother Earth will open up and
swallow her. The earth opens up and accepts Sita. Sita proves to Rama that she
is indeed chaste, but Rama does not get his wife back.
Dr. Sarabhai sees these instances of Rama’s questioning and seeking of
proof of chastity as a paradigmatic instance of injustice to women. This initial
episode frames the later sequences of the dance, when she portrays instances
in which some hegemonic segments of the religious culture sanction the in-
stances of gender inequity. She also portrays the current struggles of women in
the northern Doon Valley who work for ecological preservation and try to
prevent trees from being cut down. This movement, called Chipko, is a protest
against the developers who seek to level the land in the name of development
without heed for the future. Using very few props, Dr. Sarabhai focuses on
women who refused to give in to the pressures around them, women who did
not accept, but who questioned and chafed. These women are called the ‘‘Sitas
who refuse ever again to submit to the tests and trials of weak and doubting
men.’’11 In ‘‘Sita’s Daughters’’ and other performances, Dr. Sarabhai uses
dance as a ritual vehicle to publicly portray social causes.
Courtesans like Bangalore Nagaratnammal and Brahmin women like
Balasaraswati and Rukmini Devi Arundale did not see themselves as radicals
or feminists—they were driven by an ardor to share their joy of learning, music,
or dance with others and not have it restricted to certain segments of society.
One may ask, however, if these efforts through which all women had the right
to learn and perform music and classical Bharata Natyam dance in public
should be seen from another perspective—that of the devadasi. Was this a
process of opening up the arts or one by which they were wrested, indeed
hijacked by some upper-class women?
The answers are complicated because these efforts came at a time when
there were legal and social measures to make the devadasi system illegal. One
may also note that Balasaraswati, herself from a devadasi family, worked hard
to change the conservative—and negative—perceptions about Bharata Natyam
by performing all over India and then in other countries as well, attracting not
just connoisseurs and admirers but also a devoted group of students.
The nationalistic fervor of the 1930s and 1940s formed a backdrop to and
helped Rukmini Devi’s efforts in framing this dance as a national heritage of
India. From her perspective—and the hundreds of students in her school—the
transformation of ritual music and dance was an attempt to make opportu-
nities available to women by breaking down social barriers based not only on
gender, but also on social caste, economic class, and colonial dominion.
Ironically, in Tamilnadu, in an attempt to improve the status of the
devadasis, the courtesans who danced in temples, several legal measures were
enacted to ban not just the dedication of girls to temples, but any dancing by
any woman in temples. Part 3, section 3, of the Devadasi Act of 1947 reads,
‘‘Dancing by a woman . . . in a temple or religious institution . . . or procession
or at any festival or ceremony . . . is declared unlawful.’’12 This law, which is
supposed to avoid treating women as commodities just used to entertain, also
denies them the freedom of expression to use music and dance as a ritual of
worship. Happily, in contemporary India, this law is ignored by everyone,
including the government of India. Several dance festivals have been initiated
in temples, and women do sing and dance in processions. Sometimes the
temples are used as dramatic, aesthetic backdrops, as in the Mahabalipuram
dance festival in January and the Khajuraho festival in early spring. Almost
every major temple complex—Konarak, Pattadakal, Khajuraho, Modhera—has
an annual dance festival, which is seen as a tourist attraction and cultural
entertainment. In many other temples however, like Chidambaram, the dance
festival (called Natyanjali, or Worship through Dance) coincides with the holy
day of Maha Sivaratri, the Great Night of Shiva, and is seen explicitly as a devo-
tional exercise. Accounts like the following are seen on Web sites:
The Natyanjali festival dedicated to Lord Shiva is celebrated every year

during February–March for five days in the temple premises. This is an
opportunity for all dancers, from all over India, to perform and to pay
their tribute to Lord Nataraja. It begins on the auspicious occasion
of Maha Shivaratri. During this time leading dancers from all parts of
India congregate and dance in the temple as an offering to Nataraja.
Many dancers think it is a blessing to be able to perform their ‘‘ara-
ngetram’’ (first stage performance) in the vicinity of the sanctum sanc-
torum of Lord Nataraja in Chidambaram. The festival lasts for 5 days.
Natyanjali festival is jointly organised by The Department of

Tourism, Government of Tamilnadu, The Ministry of Tourism, Gov-
ernment of India and The Natyanjali Trust, Chidambaram. It is
designed to promote a universal message of ‘‘Unity in Diversity’’
conveyed in the universal language of music and dance.13
Several factors, therefore, have come together in the use of performing arts
to showcase new spaces in women’s agencies. In this process, there have been
appeals to multiple sources of authority, including personal experiences as ar-
ticulated in dreams, the life histories of renowned women saints, and the re-
vived clout of Sanskrit texts as powerful storehouses of wisdom and practice.
Two important factors contributing to the success of these measures are the
internal paradigms of granting spaces to women along the path to liberation
and the overall significance of custom and practice in the many Hindu tradi-
tions. The latter allows for flexibility, revival and re-formation of paradigms,
appropriations, and the co-opting of concepts and rituals. But in the process of
people from various castes and classes reclaiming this heritage, did the art of
those women, the many kinds of courtesans, the many devadasi communities,
who had performed for centuries get taken away from them and commer-
cialized in a different way? The answers, as we saw, are complex.
Time and space have also been significant in these rituals of devotional
offering. These changes can only be seen in the context of the times. The early
twentieth century was a time when the ardor of nationalistic fervor, the desire
for an independent India free of colonial rule, proud in its history and heritage,
and united in culture was widespread. The upper-class women who began
showcasing the temple arts, this proud heritage, not in the temple but on the
secular stage, were a symbol of this new Bharata. This was the dance of India,
the dance according to the legendary Bharata, the Bharata Natyam. The nar-
rative of women’s ritual roles began to extend beyond the home altar and tem-
ples to newer, hybrid, public spaces. These hybrid spaces were—and remain—
the great community halls and stages where classical music and dances take
place. And now, it has come full circle and come back to the temples. For many
women, these performances have become a professional art; for others, their
performances were and continue to be acts of worship. In temples in India and
the United States, women now perform music and dance as part of their de-
votional offerings. But whereas women in the mid-twentieth century worked to
bring dance from the temple into the public sphere, now, in shows like Thaka
dhimi tha, the public stage and the television platform have become the new
hybrid spaces—extensions of the temple space and domestic altars in secular
forums. This popularization has led to dance and music becoming entertain-
ment for some individuals; for others, it is the language, form, and content of
religious ritual, as well as an agent of social change, leading them to liberation
on earth and then to heaven.
notes
1. Personal communication from Professor Davesh Soneji.
2. I was asked by Dr. Sharma to do the narration and show the slides for this
program and was involved in the presentation of the dance as part of temple culture.
3. Almost all of the official guidebooks to the area point to the image inside the
temple (marked as number 40 now by the government of India authorities and the
Archaeological Survey of India) as representing Shantala Devi. There do not seem to
be any primary sources that make this identification. The restaurant in the official
Indian Tourism Development Corporation hotel at Hassan has named its restaurant
Shantala. The general Indian stereotype of wifely virtue is to perceive a woman as the
server of food, and it is interesting that of all places, a restaurant is named after this
queen, who contributed so much to the fine arts and architecture.
4. I have discussed Sister Subbalakshmi’s contributions in my chapter, ‘‘Brim-
ming with Bhakti, Embodiments of Sakti: Deities, Devotees, Performers, Reformers
and Other Women of Power in the Hindu Tradition,’’ in Feminism in World Religions,
edited by Arvind Sharma and Katherine Young (Albany: State University of New York
Press, 1999).
5. In chapter 3, Seetha discusses the contributions of composers and musicians
between the seventeenth and nineteenth centuries in the Tanjore area, and it is in-
teresting to note that all sixty-two musicians/composers on whom she focuses are
men. It is evident from other sources also (e.g., Rajagopalan 1990b and 1992b, which
lists all the important musicians in South India) that prior to this century almost all
musicians who were well known and who gained public approbation were men.
6. It must, however, be pointed out that some of the devadasi songs are more
explicitly erotic in tone and have overt sexual imagery attached to them. See for
instance, Soneji 2004; Ramanujan, Narayana Rao, and Shulman 1994.
7. Adapted from a booklet containing the biography of Bangalore Nagar-
atnammal, written in Tamil by her disciple Banni Bai.
8. From Narayana Menon, Balasaraswati. Inter-National Culture Center, New
Delhi. Available at http://www.tamilnation.org/hundredtamils/balasaraswati.htm.
9. Balasaraswati, Presidential Address, Tamil Isai Sangam. Available at http://
www.carnatica.net/dance/bharatanatyam1.htm.
10. T. Balasaraswati, Presidential Address, Tamil Isai Sangam. Available at http://
www.carnatica.net/dance/bharatanatyam1.htm.
11. This is from a publicity pamphlet for ‘‘Sita’s Daughters’’ by Darpana Academy
of Dance.
12. This is quoted in ‘‘Devadasis, Part IV,’’ Hinduism Today. Available at http://
www.spiritweb.org/Hinduism Today/94–01-Devadasis_Part_IV.html.
13. Available at http://www.webindia123.com/festival/dance/chidam.htm.
references
Allen, Matthew Harp. 1997. ‘‘Rewriting the Script for South Indian Dance.’’ The
Drama Review 41, no. 3: 63–100.
Andal. 1971. ‘‘Nacciyar Tirumoli.’’ In Nalayira tivviyap pirapantam, ed. P. B. An-
nangaracariyar, 88–109. Kanchi: V. N. Tevanatan.
Bell, Catherine. 1997. Ritual: Perspectives and Dimensions. New York: Oxford University
Press.
Coorlawala, Uttara Asha. 2005. ‘‘The Birth of Bharatnatyam and the Sanskritized
Body.’’ In Rukmini Devi Arundale (1904–1986): A Visionary Architect of Indian
Culture and Performing Arts, ed. Avanthi Meduri, 173–194. New Delhi: Motilal
Banarsidass.
Falk, Nancy Auer, and Rita M. Gross. 1989. ‘‘Women’s Power: Mythical Models and
Sacred Sources.’’ In Unspoken Worlds: Women’s Religious Lives, ed. Nancy Auer Falk
and Rita M. Gross, 233–234. 2nd ed. Belmont, CA: Wadsworth.
Felton, Monica. 1967. A Child Widow’s Story. New York: Harcourt, Brace and World.
Ganguly, Rita, and Vidya Hydari. 1986. ‘‘From Mother to Daughter: Women in
Classical Music.’’ Manushi 6, nos. 3 and 33: 10–14.
Kishwar, Madhu, and Ruth Vanita. 1983a. ‘‘A Grand Old Lady of Music: An Interview
with Asghari Begum Sagarwali.’’ Manushi 3, no. 3: 2–10.
———. 1983b. ‘‘‘Bhairav se Sohni’: Geetika’s All-Women Classical Music Festival.’’
Manushi 3, no. 3: 8–10.
Meduri, Avanti. 1988. ‘‘Bharata Natyam, What Are You?’’ Asian Theatre Journal 5,
no. 1: 6.
Mukherjee, Sharbari. ‘‘Women and Traditional Indian Music.’’ In a brochure
published by Geetika, a music organization for women.
Nagappachar, B. 1960. The Belur Temple. Saklaspur, Karnataka: Sri Krishna Power.
Narayanan, Vasudha. 1991. ‘‘Hindu Perceptions of Auspiciousness, and Sexuality.’’ In
Women, Religion, and Sexuality: Studies on the Impact of Religious Teachings on
Women, ed. Jeanne Becher, 64–92. Geneva: World Council of Churches
Publications.
———. 1999. ‘‘Brimming with Bhakti, Embodiments of Shakti.’’ In Feminism in World
Religions, ed. Arvind Sharma and Katherine K. Young, 25–77. Albany: State
University of New York Press.
O’Shea, Janet. 1998. ‘‘‘Traditional’ Indian Dance and the Making of Interpretive
Communities.’’ Asian Theatre Journal 15, no. 1: 45–63.
Peryavacchan Pillai. 1974. Turuppavai Muvayirappati Vyakyanam, ed. P. B. Annan-
garachariar. Kanchi: Editor.
Rajagopalan, N. 1990a. ‘‘Rukmini Devi Arundale.’’ In A Garland: Biographical Dictio-
nary of Carnatic Composers and Musicians (vol. 1), 255–257. Bombay: Bharatiya
Vidya Bhavan.
Rajagopalan, N. 1990b. A Garland: Biographical Dictionary of Carnatic Composers and
Musicians (vol. 1). Bombay: Bharatiya Vidya Bhavan.
———. 1992a. ‘‘Balasaraswati.’’ In Another Garland: Biographical Dictionary of
Carnatic Composers and Musicians (vol. 2), 104–105. Bombay: Bharatiya Vidya
Bhavan.
———. 1992b. Another Garland: Biographical Dictionary of Carnatic Composers and
Musicians (vol. 2). Bombay: Bharatiya Vidya Bhavan.
Ramanujan, A. K., Velcheru Narayana Rao, and David Shulman, eds. and trans. 1994.
When God Is a Customer: Telegu Courtesan Songs by Ksetrayya and Others. Berkeley:
Rangacharya, Adya, trans. 1998. Natyasastra. Delhi: Munshiram Manoharlal.
Rao, V. Narayana, and D. Shulman. 1989. ‘‘History, Biography and Poetry at the
Tanjavur Nayaka Court.’’ In Identity, Consciousness and the Past: The South Asian
Scene, ed. H. L. Seneviratne. Social Analysis: Journal of Cultural and Social Practice,
no. 25 (September): 115–130.
Seetha, S. 1981. Tanjore as a Seat of Music. Madras: University of Madras.
Soneji, Davesh. 2004. ‘‘Living History, Performing Memory: Devadasi Women in
Telugu-Speaking South India.’’ Dance Research Journal 36, no. 2: 30–49.
Wagoner, Phillip B. 1996. ‘‘‘Sultan among Hindu Kings’: Dress, Titles, and the
Islamicization of Hindu Culture at Vijayanagara.’’ Journal of Asian Studies 55, no.
4: 851–880.
Index
Page numbers in boldface indicate authors of selections. Italicized page numbers refer to
illustrations
Abhinaya Darpana, 190–91 and matammas, 37–42, 49, 52

Acharya, Kala, 30 and medieval Tamil inscriptions,
agency, 4–5, 6, 8, 10, 13 112, 117
and goddess possession rituals, 152, 154 and performing arts, 186
and matammas, 9, 39, 52 and sakhis (‘‘female friends’’), 58
Alvars, 118, 131 austerities
Amba, 74–75 and medieval Tamil inscriptions,
‘‘Ammai’’ (mother), 134, 144n6 110–12, 115, 123n2, 124nn4,6
Ammal, Gowri, 190 and tantric rituals, 165, 174
Ammavaru(s), 49, 50 Avadi Mata, 74–75, 81n15
Anandamayi Ma, 172 Avanan journal, 124n8
Andal, 111, 133, 135, 185–89 Avvayar, 49
animal deities, 99
animal sacrifice, 53n9 Baba Balak Nath, 149
Ani month, 144n6 Babb, Lawrence, 79
antati form, 144–45nn9,11 Bagewadikar, V., 23
Archanapuri Ma, 169–70 Balasaraswati, 184–85, 190–94
Arena of Art (Kalakshetra), 192 banyan tree, 136
ascetics, 7 Bapat, Shailaja, Mrs., 25
Karaikkal Ammaiyar as, 133–36, 142 Bateson, Gregory, 150
and kolams/pottus, 101 Bayan Mata, 74
and medieval Tamil inscriptions, beauty, 7
111–12, 114–15, 124nn6,7, 125n13 and Karaikkal Ammaiyar, 12, 133–34,
and tantric rituals, 13, 159, 161, 165, 137, 139
172, 174 and kolams/pottus, 91, 97, 99, 102
ashrams, 154, 160, 165–66, 174 and matammas, 49
Atmavilas, Yamini, 53n7 ‘‘Because It Gives Me Peace of Mind’’
auspiciousness (Pearson), 7
and kolams/pottus, 10, 86–91, 95–103, Bell, Catherine, 135
103n3, 104nn4,5 Benares (Uttar Pradesh), 9, 55–63
200 index
Benarsi women, 9, 65–80 caste

Besant, Annie, 192 and goddess possession rituals, 152–53,
Bhagavad Gita, 25, 27, 29–30 156–57
Bhagavanti, 3–4, 6 and Karaikkal Ammaiyar, 133, 140
Bhairavi Brahmani, 168 and kolams/pottus, 89, 91, 99, 104n4
bhairavis/bhairavas, 168, 170–72 and performing arts, 179, 181, 183–85,
Bhaktanjali, 186 189–90, 192–95
bhakti (dedication to deities) and ratijagas (‘‘wake’’ rituals), 74, 81nn23,27
and goddess possession rituals, 149, 154 and sakhis (‘‘female friends’’), 58–59, 63n7
and performing arts, 181, 185, 188–89, and tantric rituals, 161, 168, 171
191 cat in courtyard, 19, 31
and ratijagas (‘‘wake’’ rituals), 67 Cekkilar, 131, 144nn5,6
and tantric rituals, 145n29, 166 celibacy, 13, 159–60, 165–66, 168, 172–74
bhakti poets, 24, 69, 131, 144n4 Cempiyan Mahadevi, 117
Bhandarkar Oriental Research Institute, 21 Chandralekha, 13, 192
Bharata, 181–82, 190 Chawla, Janet, 158n3
Bharata Natyam, 177–78, 181–82, 185–86, 188, childbirth, 3–4
190–95 and goddess possession rituals, 158n3
Bharatiya Janata Party (BJP), 20 and kolams/pottus, 89, 92, 100
Bhate, Saroja, 23, 26–27 and matammas, 44
Bhattacharjee, Bholanath, 171–72 and medieval Tamil inscriptions, 11
Bhattar, Parasara, 188 and ratijagas (‘‘wake’’ rituals), 65–66
bindis, 58, 85, 91 and sakhis (‘‘female friends’’), 58
Boddy, Janice, 150–51 and Sanskrit language, 8, 31
bottus, 38, 49. See also pottus Chipko, 193
Brahma, 124–25n9, 137, 142, 182 Chitale, K. V., 23
Brahman, 167, 171 Chittor, fortress of, 69
Brahmani, 162, 168 Chocolat (film), 10, 70–73, 78
Brahmanicide, 137, 142 Chola kings/queens, 116–17, 120, 124–25nn9,
Brahmins, 7 10, 126n20
and goddess possession rituals, 150 Christians, 59, 92, 99
and Karaikkal Ammaiyar, 133 Cilappatikaram, 124nn4,5, 143–44n1,
and kolams/pottus, 88–89, 101 145n24
and matammas, 41, 45, 49, 52–53nn5,9 Cinna Gangamma, 41, 45, 48
and medieval Tamil inscriptions, 110, 114, Colina, Haeli, 132
117–19, 121, 125–26nn13,16 Collins, Elizabeth, 4–5, 13
and performing arts, 180, 184, 189–90, colonial period, 23, 37, 180–81, 183, 194–95
192–94 colored threads, 155–56
and ratijagas (‘‘wake’’ rituals), 76, 81n18 community of devotees
and Sanskrit language, 4, 8, 22, 24, 32n5 and goddess possession rituals, 156–57
and tantric rituals, 161, 173 and Karaikkal Ammaiyar, 137–38, 140–43
Briggs, Charles, 23, 24–25, 27–30, conjugal homes
31–32n3 and ratijagas (‘‘wake’’ rituals), 66, 77
Buddha, 111 and sakhis (‘‘female friends’’), 55, 61
Buddhism, 131, 133, 140, 143–44n1 consorts, female
Butler, Judith, 5 and medieval Tamil inscriptions, 117, 122,
Bynum, Caroline Walker, 145n12 126n17
and tantric rituals, 159–60, 162–64, 166,
Caitanya, 183 171–73
cakras, 163–64, 170 ‘‘Contextualizing the Eternal Language’’
Caldwell, Sarah, 157–58n1 (Deshpande), 25
Camunda Devi, 149 courtesans, 180–81, 183–85, 189–90, 193,
Cankam puram poetry, 138 195
cardamom, 155–56 cowherdesses (gopis), 9, 56, 61, 182, 187–88
Carnatic music, 184–86, 188–89 Craddock, Elaine, 7, 11–12, 131–47
index 201
cremation grounds fasting

and Karaikkal Ammaiyar, 12, 135–43 and goddess possession rituals, 12, 149
and ratijagas (‘‘wake’’ rituals), 79 and medieval Tamil inscriptions, 110, 112,
cross-cousin marriages, 50 124nn5-7
cultic rituals, 78–79, 81nn32,32 and ratijagas (‘‘wake’’ rituals), 65
Faust (Goethe), 23
dance. See performing arts fealty oaths of servants, 114–15
Dandekar, R. N., 23, 32n4 Feldhaus, Ann, 42
Daniels, Valentine, 66 Felton, Monica, 184
Das, Veena, 89–90 female-female friendship bonds. See sakhis
Date, Ranjana, 25–26 (‘‘female friends’’)
death feminist movement, 32nn5,6, 193
and Karaikkal Ammaiyar, 12 fertility, 36, 39–41, 50
and kolams/pottus, 87, 89–90, 92, 100 Festival of Tiruppavai, 186–87
and matammas, 44, 53n7 Festival of Tyagaraja, 189
and sakhis (‘‘female friends’’), 58 festivals
Deccan College Dictionary Project (Pune), and Karaikkal Ammaiyar, 138, 144n6
21 and matammas, 36, 37, 38, 41–42, 52–53n5
demonic form, 12, 134–35, 138, 143, and medieval Tamil inscriptions, 111,
144n8 123–24n3
Dempsey, Corinne, 14n1 and performing arts, 182, 186–87, 189, 194–95
deprivation theory, 151 and Sanskrit language, 24
Deshpande, L. V., 23 and tantric rituals, 165
Deshpande, Madhav, 23, 25 films, 68, 144n3, 145n25
Deshpande, Maitreyee, 21 Flueckiger, Joyce Burkhalter, 7, 8–9, 32n8,
Deshpande, Pradnya, 19 35–54, 55, 59, 66, 117
Devadasi Act (1947), 37, 194 folk dance, 182–83
devadasis Freedom movement, 23, 30
and matammas, 36–37, 52n2 freedom of movement, 8–9, 35–41, 45–52
and performing arts, 180–81, 184–85, Freudian psychoanalytic theory, 153
189–90, 193–95, 196n6
Devarajan, Arthi, 32n9 ganas (ghouls), 134–36, 138–39, 143
dharma (duty) Gandhi, Mahatma, 23
and performing arts, 183 Ganesh/Ganesha, 49, 53n7, 63n4, 177
and sakhis (‘‘female friends’’), 61–63 Gangamadeviyar, 112
and Sanskrit language, 8 Gangamma (village goddess), 8–9, 35–52, 47,
and tantric rituals, 165, 169, 172, 174 52–53nn3,5,8,9
Dharma Sastras, 187 Ganges River, 63n4, 97, 99
Dharma Shastras, 183 garbha dance, 182–83
Durga, 66, 110, 118–19, 138, 152, 155, 161, Gauri Ma, 166–67, 167, 173
173 Gauri puja, 53n7
Dvaraka, 56, 69 Gayatri mantra, 30
Geertz, Clifford, 90
Easter liturgy, 24 gender differences
education system, dual, 20–22 and Karaikkal Ammaiyar, 133
ego, 139, 150–51, 154 and matammas, 37, 40
English language, 20, 21 and medieval Tamil inscriptions, 109–23
equality gift giving, 115–22, 126nn13,14
and Karaikkal Ammaiyar, 140–41 vow taking/self-offering, 110–15
and sakhis (‘‘female friends’’), 58–59, 62 and performing arts, 180, 184, 193, 196n5
Erndl, Kathleen M., 7, 12, 149–58 and Sanskrit language, 20–22
‘‘everyday resistance,’’ 72 gender identity, 5
Ewing, Katherine, 37 Gentes, M. J., 39
exorcism, 169 gift giving in medieval Tamilnadu, 11, 115–23,
expiation gifts, 121–22 125–26nn13-18
202 index
Gita, 25, 27, 29–30 Islam, 79, 183

Goda Mandali (circle of Andal), 186–88 isthadevtas (self-chosen deities), 67, 69, 76
goddess images, 11, 48, 116–18, 155 Iyer, E. Krishna, 190
goddess possession rituals, 12, 149–57
in context of ‘‘play,’’ 149–52 Jains
description of rituals, 154–56 and Karaikkal Ammaiyar, 131, 133, 140–41,
and matajis, 152–54, 157–58n1 143–44n1
and women’s space, 156–57, 158n3 and medieval Tamil inscriptions, 11, 110–12,
goddess worship. See Shaktism (goddess 116, 124nn4,6,7
worship) and ratijagas (‘‘wake’’ rituals), 79
Goethe, 23 Jamison, Stephanie, 8, 76
Gold, Ann, 7 jataras. See festivals
gopis (cowherdesses), 9, 56, 61, 182, 187–88 Jay, Edward, 57–58, 63nn7,8
Gottschalk, Peter, 66 Jayashri Ma, 167
Govindamma, 35–38, 41, 46–51, 47, 52 Jaya TV, 177
Gowri, Mylapore, 192 Jewels of Authority (Patton), 7
Grandmother Language (Patton), 20, 22 Johnson, Mark, 86
grihi sadhikas (holy women), 160, 165, jori, 62–63
168–69 Joshi, S. D., 25
Gune, N. P., 23 jungli (uninhabited space), 39, 53n6
Guptasadhana Tantra, 161, 164 Jvala Mukhi, 149, 152, 155
Gurjar, Asha, 27, 28, 29–30 Jyestha. See Mudevi
gurus, female, 154, 159–60, 163, 164–74
celibate wives/widows, 166, 172–74, 175n21 Kaimal, Padma, 125n10
professional consorts, 166, 171–73 Kakar, Sudhir, 62
tantric holy women, 160, 165, 168–69 Kali, 136–39, 152, 154–55, 162, 168–69, 171
tantric wives, 160, 165, 169–70 Kali bhava, 165
yoginis, 159, 165–68, 167 Kalika, 163
Kali Tantra, 160–61
hamam, 157 Kamakhya Tantra, 163–64
Hancock, Mary, 5 Kaman, Koccattan, 121
Harijans, 153 Kaman, Tappi Mintan, 121
Harlan, Lindsey, 7, 9–10, 65–83 Kama Sutra, 180
head offerings, 114–15, 125n11 Kampan, Vikrama, 113, 121
healers, 12, 152–53, 168 Kangra (North India)
heroic warriors, 138 and goddess possession rituals, 12, 149–57,
hero stones, 113–15, 120, 125n10 157–58n1
Hindu Temple (Atlanta, Ga.), 179 Karaikkal Ammaiyar, 11–12, 131–43, 132
holy women, tantric, 160, 165, 168–69 creating community, 140–43
hospitality defining path, 139–40, 143
and kolams/pottus, 88, 96–97, 100 and localization of pan-Indian god, 133,
and ratijagas (‘‘wake’’ rituals), 75–77 136–39, 145nn16,24
Houben, Jan, 23 and ritualization/asceticism, 134–36
householders, female story of, 133–34, 144nn3,5,6,8
and goddess possession rituals, 12, 152–53 Karaikkal Ammaiyar (film), 144n3, 145n25
and kolams/pottus, 10, 87, 96, 100–103 karma, 112, 161
and matammas, 36 Karnataka, 112–13
and Sanskrit language, 27 Kartik month, 3–4, 56
and tantric rituals, 13, 159, 166, 169–70, 173 Kartik puja, 3–4, 56
Keats, John, 23
icon seizure, 69 Kenghe, C. T., 23
Ideology and Status of Sanskrit (Houben), 23 Kerala, 39–40, 89, 157–58n1
Indian Bank, 187 Khajuraho festival, 194
inscriptions, Tamil, 11, 110–23, 123–24nn2, Khandelwal, Meena, 5, 7
3,7,8, 124–25nn9,12,13,15, 126nn17,20 Kinjavadekar, Mandakini, 26
index 203
kirtan singing group, 173 Mahamandaleshwar, 154

Kodnikar, Menakshi, 27–28 Mahamaya, 163
kolams, 10, 85–91, 86, 94–103, 103nn1,3 Maharashtra, 8, 19–31, 32n6, 42
and householder ideology, 100–102 Maha Sivaratri, 194
mapping of, 100 Mahila Mandal (Women’s Organization), 153
and moral status, 102–3, 104n8 Makkar Samkranti, 59
presence/absence of, 88–90 mangalam (last song), 186
as spatial thresholds, 96–97 mangoes, 133, 144n6
as temporal thresholds, 98–99 Manikkavachakar, 111
Korravai, 138 Manimekalai, 104n8, 111, 124n4, 143–44n1
Kramrisch, Stella, 91, 98 Manjul, V. L., 21
Krishna mantras
and Andal, 135, 187–88 and goddess possession rituals, 153–54
and goddess possession rituals, 149 and Sanskrit language, 8, 23, 25, 28, 30
and Karaikkal Ammaiyar, 135 and tantric rituals, 160–65, 167–68, 171,
and Mira Bai, 69, 80n11, 177–78 173–74
and performing arts, 178, 182, 187–88 Manu, 187
and ratijagas (‘‘wake’’ rituals), 69 mapping
and sakhis (‘‘female friends’’), 9, 56–57, 61 of auspiciousness/ritual pollution,
Krishnashastri, 23 100, 104n6
Krsnabai, 42 of social relations, 77–78
Kubjika Tantra, 161 Marathe, Lalita, 28–29
kula cakra, 163 Marathi language, 21
Kulacudamani Tantra, 161–62 Mariyamman, 110
Kularnava Tantra, 160, 164 Markali month, 85, 98–99, 111, 123–24n3, 187
kula women, 160–62 matajis, 12, 149–50, 152–57, 157–58n1
kuldevis (familial goddesses), 74–75 matammas, 9, 35–52, 47, 53n8, 117
Kumari, Hem, 59–60 Maya Tantra, 163
kumari puja, 161 McDaniel, June, 7, 12–13, 159–75
kundalini yoga (yogini), 159, 165–68 McGee, Mary, 6, 8
meditation, 8
Lakoff, George, 86 and goddess possession rituals, 12, 149, 154
Lakshmi, 91, 101–2, 178, 186 and Karaikkal Ammaiyar, 143
Lakshmi Ma, 169 and kolams, 89
land grants, 114, 118 and tantric rituals, 145n29, 160, 162–65,
lata sadhana (sexual ritual), 13, 159–60, 173–74
162–74 mediums
Leslie, Julia, 7 and goddess possession rituals, 12, 152–53,
Lewis, I. M., 151 157–58n1
liberation, 6 matammas as, 36
and Karaikkal Ammaiyar, 139–40, 143 and ratijagas (‘‘wake’’ rituals), 79
and matammas, 48–49 Menon, Narayana, 190
and medieval Tamil inscriptions, 112 menstruation
and performing arts, 182–83, 187, 195 and kolams/pottus, 10, 88–95, 100, 103n3,
and tantric rituals, 164–65, 170, 172, 174 104nn5,6
Lilavati, 59, 60–61 and Sanskrit language, 21–22
Listen to the Heron’s Words (Raheja and and tantric rituals, 163
Gold), 7 merit, 120–22
‘‘little brother Rajput,’’ 77, 81n23 Mernissi, Fatema, 68
Longfellow, Henry Wadsworth, 23 metaphors, 86, 91, 95–97, 183
Mewar, 68–69, 74–75, 79
Madan, T. N., 98, 101–3 Michaels, Alex, 20
Mahabalipurum dance festival, 194 Mira Bai, 69, 80n11, 177–78, 188
Mahakala, 162 Monius, Anne, 124n4
Mahalakshmi, R., 144n8 ‘‘mother’s tali,’’ 40, 53n7
204 index
Mudevi, 102 parda, 66–69, 80nn4,5

Murthy, A. K., 24, 32n7 Parvati, 136, 139, 144n6, 145n25
Murugan, 126n17 Pashupata Shaiva, 126n20, 142
music and dance. See performing arts Passu Mataji, 153, 156, 158n2
Muslims pasupu/kumkum (auspicious vermilion-
and pottus, 92 tumeric marking), 38, 49
and ratijagas (‘‘wake’’ rituals), 79–80 pathshalas, 20–22
and sakhis (‘‘female friends’’), 59 patriarchal norms, 11
and goddess possession rituals, 12, 157,
Nachiyar Tirumoli (Andal), 187 158n3
Nagarajan, Vijaya Rettakudi, 7, 10, 85–105 and Karaikkal Ammaiyar, 133
Nagaratnammal, Bangalore, 184–85, 188–90, and performing arts, 181, 183
193 and ratijagas (‘‘wake’’ rituals), 78
Nammalvar, 186 and sakhis (‘‘female friends’’), 57–58,
Nandikeswara, 191 61–62
Narayan, Kirin, 55, 60, 63, 63n3 patronage
Narayanan, Vasudha, 7, 13, 39–40, 52n2, and Karaikkal Ammaiyar, 133
177–98 and matammas, 52
natal homes and medieval Tamil inscriptions, 116–20,
and ratijagas (‘‘wake’’ rituals), 66–69, 77 122, 126n18
and sakhis (‘‘female friends’’), 55, 60–61 and performing arts, 181, 184, 191
Nataraja, the Lord of the Dance, 177, 182, and ratijagas (‘‘wake’’ rituals), 69, 74
193–94 Patton, Laurie L., 7–8, 19–34
Natarani (Queen of Dance), 193 pavai rituals, 111–12
nationalism, 23, 181, 185, 194–95 Pavlova, Anna, 190
Natyanjali (Worship through Dance), 194–95 Pawde, Kumud, 24, 32n6
Natyasaravati, 183 Pearson, Anne, 6–7
Natya Sastra (Bharata), 182, 185, 190, 192 Pedda Gangamma, 41
nava tali, 40 performative context of language, 23–25,
Navratri festival, 38, 42, 52–53n5, 66 31–32n3, 32nn6-9
Nayak performance, 78 performing arts, 13, 177–95
Nayanars, 11–12, 131, 133, 142 and Andal, 186–89
Nehru, Jawaharlal, 23 and Balasaraswati, 184–85, 190–94
newspapers, 68 and Bangalore Nagaratnammal, 184–85,
NGOs (nongovernmental organizations), 37 188–90
nicitikai memorial, 112–13 and Goda Mandali (circle of Andal), 186–88
Nili, 138, 145n24 and Rukmini Devi Arundale, 184–85,
Niruttara Tantra, 162–63 190–94
nonpu stones, 110–12, 114–15, 120, 123n1, and social change, 192–95
124nn4,5,7 social values/perceptions of, 180–85
Northup, Lesley, 5 Periya Puranam (Cekkilar), 131, 144n5
personalization of religion, 5–6
Ogura, Yasushi, 126n20 and medieval Tamil inscriptions, 11,
Orr, Leslie C., 7, 11, 40, 53n8, 109–29 109–10, 113, 115, 118–19, 122–23
‘‘others,’’ 140–41, 143, 145n28 and ratijagas (‘‘wake’’ rituals), 74–75
and Sanskrit language, 27–31
Pabuji (Rajasthani epic), 75, 78, 81n31 Peterson, Indira, 131, 140
Padoux, Andre, 145n29 pey (demon), 134, 138. See also demonic form
pallipatais (‘‘sepulcher temples’’), 120, 126n20 pilgrimages
Palsule, G. B., 23 and goddess possession rituals, 12, 149, 154
pancatattva ritual, 163, 166–67 and Karaikkal Ammaiyar, 12, 134
panditas, 20–22 and kolams/pottus, 99
Pandya kings, 117 and ratijagas (‘‘wake’’ rituals), 69
Paramatattan, 133 and tantric rituals, 165, 168, 173
parani stones, 111–12, 114–15, 123n2, 124n7 Pintchman, Tracy, 3–15, 55–64
index 205
‘‘playing,’’ 149–50, 155 and ratijagas (‘‘wake’’ rituals), 65

Pongal festival, 85, 99 and sakhis (‘‘female friends’’), 58–60
possession and Sanskrit language, 24–27
and goddess rituals, 12, 149–57 stri/shakti puja, 161
in context of ‘‘play,’’ 149–52 and tantric rituals, 160–61, 173
description of rituals, 154–56 Pullappai, 112
and matajis, 152–54, 157–58n1 Punitavati, 11–12, 133–34
and women’s space, 156–57, 158n3 puram poetry, 138
and matammas, 36, 44, 47, 51 Puricanti, 113
and ratijagas (‘‘wake’’ rituals), 79–80 purity
and tantric rituals, 160, 165, 168–69 and kolams/pottus, 10, 86–88, 91,
pottus, 10, 85–87, 90–96, 100–101, 103, 103n2 95–98, 102
and householder ideology, 100–101 and medieval Tamil inscriptions, 112
and moral status, 103 and Sanskrit language, 21–22, 26, 28
presence/absence of, 91–96 puttininti tali (‘‘mother’s tali’’), 40, 53n7
power, 13
and goddess possession rituals, 149–50, Radha-Krishna, 56–57, 61
152 radio, 186
and kolams/pottus, 87–88, 91, 95, 103 Raghunathan, Chitra, 188
and medieval Tamil inscriptions, 124n6 Raheja, Gloria, 7, 61
and performing arts, 193 Rajasthan (North India), 9–10, 65–80
and ratijagas (‘‘wake’’ rituals), 75–79, 81n18 Rajendra I, 124–25n9
and talis of women, 39, 41 Rama, 125–26n16, 186–87, 193
and tantric rituals, 165 Ramakrishna Paramahamsa, 168
prabandha literature, 144n9 Ramanuja, 186
praise poems, 133 Ramarakshastotram, 25, 30–31
Prakash, Ved, 59 Ramayana, 193
Prakrit Dictionary Project, 28 Ramdas, 186
pregnancy, 42, 171–72, 193 Ranganatha Paduka Mandali, 186
presence/absence Rangarajan, Srinidhi, 186–87
and kolams/pottus, 87–92 Rangaswamy, Dorai, 145nn14,15
memories of, 92–96 rasa-lila (‘‘circle dance’’), 56
primogeniture, 77, 81n23 Rashtriya Sanskrit Sansthan, 20
Proctor-Smith, Marjorie, 5 ratijagas (‘‘wake’’ rituals), 10, 65–80
prostitution and Chocolat (film), 70–73
and matammas, 46 locative dimension of, 65–70
and performing arts, 181, 184 and spheres of influence, 73–80
and tantric rituals, 162, 171–72, 174 Ratnaparakhi, A. R., 23
puberty red dots. See bindis; bottus; pottus
and matammas, 9, 36, 39–41, 43, 45–46 reform movements
and pottus, 91 and Karaikkal Ammaiyar, 133
public spaces and matammas, 36–37, 39, 52
and goddess possession rituals, 156–57 and performing arts, 13, 178–81, 192–95
and kolams, 96–97 and Sanskrit language, 21–22
and performing arts, 182–83, 185, relational love, 62
188–89, 195 religious life in medieval Tamilnadu, 11,
and ratijagas (‘‘wake’’ rituals), 65–66, 71–72, 109–23
75, 80 gift giving, 11, 115–23, 125–26nn13-18
pujarini, 173 vow taking/self-offering, 11, 110–15,
puja rituals 123nn1,2, 124nn4-7, 124–25nn9-11
Durga puja, 118–19, 161, 173 renunciants
and goddess possession rituals, 12, 149 and goddess possession rituals, 12, 152, 154
Kartik puja, 3–4, 56 and kolams/pottus, 101
kumari puja, 161 and medieval Tamil inscriptions, 109,
and matammas, 48, 53n7 112–13, 115, 124nn4,7
206 index
renunciants (continued) and performing arts, 179, 180–82,

and tantric rituals, 12–13, 159, 165–66, 190–92, 195
173–74 and tantric rituals, 173
Rettakudi, 92 Sanskrit and Maharashtra (Dandekar), 23
Reynolds, Holly, 40 Sanskrit Commission (1958), 20
Rig Veda, 98 Sanskrit Institute, 30
ritual friendships, 9, 55–63, 63nn4-9 saptamatrikas, 36
ritualization, 134–36 Saptapadi (dance performance), 179, 196n2
ritual marriage, 9, 35–52 Sarabhai, Mallika, 13, 180, 192–93
ritual offerings to Shiva, 11–12, 133–36 saravatis, 183
ritual performances, 4–13. See also entries satis, 112–13, 115, 121, 124–25n9
beginning with ritual Satyanarayan, 63n4
and ratijagas (‘‘wake’’ rituals), 9–10, 65–80 Scheherazade Goes West (Mernissi), 68
and Sanskrit language, 24–27, 32nn6-9 scriptural utterance, transformation of, 27–31
ritual pollution secrets, 60–61, 63n9
and kolams/pottus, 10, 86–97, 100, 103, Seetha, S., 184, 196n5
104nn4-6 self-sacrifice, 11, 112–15, 124–25nn9-11, 135
and sakhis (‘‘female friends’’), 58 Sered, Susan Starr, 5, 61, 109, 115, 150–51
Rivkin, Julie, 81n21 Sermon on the Mount, 23
Robinson, Sandra, 118–19 Sethuraman, N., 126n20
Roles and Rituals for Hindu Women sexuality
(Leslie), 7 and Karaikkal Ammaiyar, 134
Rukmini Devi Arundale, 184–85, 190–94 and kolams/pottus, 91, 102, 104n5
and medieval Tamil inscriptions, 124n6
sacralizing function, 5–6 and performing arts, 182–85, 187–88,
and kolams/pottus, 97 191–92, 196n6
of sakhis (‘‘female friends’’), 57–58, 63n4 and tantric rituals, 13, 159–60, 162–74
of Sanskrit language, 8, 19–31 Seymour, Susan, 62–63
and tantric rituals, 174 Shaiva Siddhanta, 131–32, 137–38, 140–43,
Sacrificed Wife/Sacrificer’s Wife (Jamison), 8 144n4, 145nn28,29
sadhikas, 171–72 shakti puja, 161
sadhu/faqir (wandering monk), 37, 101, 166 Shaktism (goddess worship)
sakhis (‘‘female friends’’), 9, 55–63, 63nn4-9 and goddess possession rituals, 12, 149–54
salvation and tantric rituals, 159–74, 175n21
and Karaikkal Ammaiyar, 133, 136, 139–40, ‘‘Shakti: The Power of Women’’ (dance
143 performance), 193
and performing arts, 180, 182–83, 187 Shantala Devi, 183, 196n3
Samskara (Murthy), 24, 32n7 Sharma, Seshu, 179, 196n2
samskaras, 28 Sharma, Ursula, 63n1
Samskriti Bharati, 23, 29 Shiva
Samvadamala (Ratnaparakhi), 23 and goddess possession rituals, 149
Sandhya mantras, 30 and Karaikkal Ammaiyar, 11–12, 131–43,
‘‘Sangam’’ literature, 113 144n6, 145nn15,25
Sanskrit, 4–5 and matammas, 41
as domestic language, 7, 8, 19–31 and medieval Tamil inscriptions, 111,
and dead/artificial language, 22–23 113–14, 117, 124–25n9, 125–26nn16,17
flexibility of, 24–25, 31 and performing arts, 177–79, 182, 193–94
and ideology, 23, 32nn5,6 and sakhis (‘‘female friends’’), 62, 63n4
performative context of, 23–25, 31–32n3, and tantric rituals, 166
32nn6-9 shlokas, 25, 29–31
as scriptural utterance of transformation, Shulman, David, 97
27–31 Shurajit, Radhika, 178
and goddess possession rituals, 150 Silappatikaram, 180
and Karaikkal Ammaiyar, 131, 137 Singh, Bheru, 79, 81n32
and matammas, 53n9 Singh, Sunita, 3
index 207
Singh, Surtan, 78–79, 81nn27,32 ‘‘tantra-mantra,’’ 153

Sister Subbalakshmi, 184, 196n4 tantric rituals, 13, 159–74
Sister Usha, 154 and female consorts, 162–64
Sita, 125–26n16, 193 and female gurus, 159–60, 163, 164–74
‘‘Sita’s Daughters’’ (dance performance), 193 celibate wives/widows, 172–74, 175n21
Sivoham Sivoham (dance performance), 179 professional consorts, 166, 171–73
social consciousness, 179–80, 185 tantric holy women, 160, 165, 168–69
social reform. See reform movements tantric wives, 160, 165, 169–70
South Indian Hindu women, 8–9, 35–52, 47, yoginis, 159, 165–68, 167
53nn7,8 and goddess possession rituals, 154
space, thresholds of, 85–88, 96–97 and incarnations of the goddess, 160–62
Speak Sanskrit movement, 23, 29 and Karaikkal Ammaiyar, 142, 145n29
‘‘Splendid People, The,’’ 24 and tantric texts, 145n29, 159–64, 174
‘‘spoken Sanskrit’’ practices, 23 Tara, 169
Sreyas Mandali, 186 Tara Devi, 152–53
sringara (romantic love), 181, 191–92 Tejaji, 77
stridhana (dowry), 118, 125–26n16 television, 68, 177–79, 186, 195
stri-dharma, 8, 169 Telugu communities, 9, 40, 53n7
stri puja, 161 temple walls (in Tamilnadu). See inscriptions,
Subbalakshmi, Sister, 184, 196n4 Tamil
Subbulakshmi, M. S., 184 temple women
Subhasri Mandali, 186 and matammas, 37–38, 41–43, 52n3, 53n9
suffering, 89–90 and medieval Tamil inscriptions, 110, 114,
Sukh Devi, 74 117, 119, 126n18
Sundara Chola, 124–25n9 and performing arts, 178, 180–81
surya asana, 98 Tevaram, 111
Swaminathan, S., 124n8, 125n15 Tevaram poets, 142
sweets Thai month, 99
and goddess possession rituals, 155 Thaka dhimi tha (television show),
and sakhis (‘‘female friends’’), 58–59, 63n5 177–79, 195
and tantric rituals, 162 Thakar, Manik, Mrs., 25
theosophist movement, 192
Tadpatrikar, S. N., 23 threshold designs. See kolams; pottus
Tagore, Rabindranath, 190 Tilak, Lokmanya, 23
Taliban, 157 time and space, thresholds of, 85–88, 95–99
talis Tirupati (Andhra Pradesh), 9, 35–52, 52nn1,3,
and female householders, 91 52–53nn5,8,9
and matammas, 9, 39–46, 49–52, 53nn7,8 Tiruppavai (Andal), 111, 186–88
goddess’s tali, 41–42 Tiruvalankatu, 12, 134, 136–39, 144nn5,6,9,
tali as protection of the goddess, 145n15, 145nn15,24
42–46, 117 Tiruvempavai (Manikkavachakar), 111
talis of women, 39–41, 53n7 transcendent aims
and medieval Tamil inscriptions, 116–17 and goddess possession rituals, 157
Tamil devotional literature, 111–12, 116, 118, and Karaikkal Ammaiyar, 142–43
126n17, 133, 143–44n1 and medieval Tamil inscriptions, 11, 112, 115,
Tamil language, 110, 185 118, 122
Tamil Music Association, 186 transformation
Tamilnadu (South India), 10, 39–40, 85–103 and goddess possession rituals, 152
and Karaikkal Ammaiyar, 131–43, 132, 144n4 and performing arts, 185, 188, 194
and medieval Tamil inscriptions, 11, 40, of scriptural utterance, 27–31
109–23, 125n12 and tantric rituals, 167
and performing arts, 185–95 and travel, 68, 72–76, 80
Tamil poetry, 12, 133–43, 144–45nn9-11, Tulsi (plant-goddess), 56, 63n4
145n28, 186–89 tunankai dance, 138
tantra (esotericism), 149 turam, 88, 93–95
208 index
Tyagaraja, 186, 189 and medieval Tamil inscriptions, 111, 118,

Tyagaraja Utsava, 189 123–24n3, 126n17
and performing arts, 186–88, 193
ugram, 36, 48 vow taking, 6
Uma, 117–18, 126n17 and matammas, 53n8
university system, 20–22, 32n5 and medieval Tamil inscriptions, 11, 110–15,
Untouchable caste, 93, 95, 99, 153 120, 123nn1,2, 124nn4-7
urdhvatandava pose, 136, 145n14 vratas (vows), 65, 110, 124n6. See also vow
Usha Mataji, 154 taking
Vrindavan, 56
Vaishnava tradition
and Karaikkal Ammaiyar, 133 Wadley, Susan, 6
and performing arts, 186–91 wakes. See ratijagas (‘‘wake’’ rituals)
and ratijagas (‘‘wake’’ rituals), 69 Watts, Alan, 150
and tantric rituals, 169, 171, 175n21 West Bengal, 13, 159–74, 167
Vaishno Devi, 152–55 Wezler, Albrecht, 23
Vallathol, 190 widows
Vampu, 113 and matammas, 37–39, 46, 53n7
van Bijlert, Victor, 23 and medieval Tamil inscriptions, 113
Varalaru journal, 125n15 and performing arts, 184, 187
Varanasi. See Benares and pottus, 92
Vatanaroshvara, 144n6 and Sanskrit language, 30–31
vattanai posture, 138 and tantric rituals, 160, 166, 171–74,
Vatuk, Sylvia, 59 175n21
Vedas, 8, 98, 137, 140, 161, 182–83 winter solstice, 85, 98–99
veena, 184 Women in Ochre Robes (Khandelwal), 7
veiling, 67 Woolf, Virginia, 76–77, 81n21
velaikkarar oaths, 114–15 Wordsworth, William, 23
Venkatesvara, 49–50, 52–53n5
venpa verses, 144–45nn9,11 Yogeshvari, 168
vernacular Sanskrit, 25 yoginis, 159, 165–68
Veshalamma, 35–38, 42–46, 51–52 yogis
vesya, 162–63 and goddess possession rituals, 152
Vijayalakshmy, R., 144n10 and Karaikkal Ammaiyar, 140
village goddesses, 8–9, 35–52, 47, 52n3, and kolams/pottus, 98
52–53nn5,8,9 and medieval Tamil inscriptions,
viracara practice, 170 122–23
Viramahadeviyar (Chola queen), 124–25n9 and performing arts, 182, 191
virgins, 49–50, 161 and tantric rituals, 145n29, 164, 170, 173
Vishnu
and Karaikkal Ammaiyar, 131 Zelliot, Eleanor, 8
and matammas, 48–49 Zvelebil, Kamil, 144n9

Women Lives Women's Rituals PDF

Uploaded by

Copyright:

Available Formats

Women Lives Women's Rituals PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Women Lives Women's Rituals PDF

Uploaded by

Copyright:

Available Formats

Women’s Lives, Women’s

Rituals in the Hindu Tradition

Oxford New York

Copyright Ó 2007 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.

All rights reserved. No part of this publication may be reproduced,

Library of Congress Cataloging-in-Publication Data

Printed in the United States of America

We have chosen to eliminate all diacritical marks from the book

PART I Engaging Domesticity

1. The Cat in the Courtyard: The Performance of Sanskrit

2. Wandering from ‘‘Hills to Valleys’’ with the Goddess:

3. Lovesick Gopi or Woman’s Best Friend? The Mythic

4. Words That Breach Walls: Women’s Rituals in Rajasthan, 65

5. Threshold Designs, Forehead Dots, and Menstruation

PART II Beyond Domesticity

6. Domesticity and Difference/Women and Men:

7. The Anatomy of Devotion: The Life and Poetry

8. The Play of the Mother: Possession and Power

9. Does Tantric Ritual Empower Women? Renunciation

10. Performing Arts, Re-forming Rituals: Women

Figure 2.1. Govindamma invoking the goddess Gangamma

Elaine Craddock is an associate professor and chair of the Depart-

Kathleen M. Erndl is an associate professor of religion at Florida

Joyce Burkhalter Flueckiger is a professor in the Department of

Lindsey Harlan is a professor of religious studies at Connecticut College. She is

June McDaniel is a professor of religious studies at the College of Charleston,

Vijaya Rettakudi Nagarajan is an associate professor of South Asian religions

Vasudha Narayanan is a professor of religion at the University of Florida and a

Leslie C. Orr is an associate professor in the Department of Religion at Con-

Laurie L. Patton is Winship Distinguished Research Professor in the Human-

Tracy Pintchman is a professor of religious studies and Hindu studies at Loyola

We give birth to a child, we put oil on the child’s body, and we

The woman who spoke these words, Bhagavanti, was reclining on

Bhagavanti’s words quoted above bring to mind the common perception

The model of performance implies several different agents

may even revise the text or tradition in their performance. There

In different contexts, women may function as ritual authors, performers, au-

Notes on the Scholarly and Historical Context for This Study

of the women themselves in these texts. Here I might mention, as examples,

Organization of the Volume

riage is understood to be the quintessential domestic institution, serving most

facilitate women’s movement across domestic boundaries and among various

traverse boundaries, reposition or reinterpret larger cultural formations and

Renunciation and Domesticity among Female Bengali Tantrikas.’’ McDaniel,

Will the nature of Sanskrit and Hindu religious experience change

This young woman’s comments indicate a certain domestic wel-

possible with an unlikely amalgamation of factors: traditional Hindu ideologies

Sir, perhaps I might ask you to consider where our respective

Formal Language within the Household: Women

It is a truism and fairly uninteresting to state the obvious—that Sanskrit is

My research to date suggests a similar richness and variety in women’s expe-

He used to teach only for boys, but by special tuition he taught

Sanskrit as the Scriptural Utterance of Transformation

In my research with these women, the second arena of Sanskritic performance

4. R. N. Dandekar himself was always interested in cultural variations and

Steinman, Ralph Marc. 1986. Guru Shishya Sambandha: Das Mesiter-Schuler-Verhaltnis

The goddess used to . . . take me from hill to valley [konda-kona],

Govindamma and Veshalamma are women who have exchanged