Priestess, Mother,

Sacred Sister
Priestess, Mother,
Sacred Sister
RELIGIONS DOMINATED
BY WOMEN

Susan Starr Sered

OXFORD UNIVERSITY PRESS

New York Oxforda
For Yishai

Oxford University Press

Oxford NewYork
Athens Auckland Bangkok Bombay
Calcutta Cape Town Dar es Salaam Delhi
Florence Hong Kong Istanbul Karachi
Kuala Lumpur Madras Madrid Melbourne
Mexico City Nairobi Paris Singapore
Taipei Tokyo Toronto
and associated companies in
Berlin Ibadan

Copyright © 1994 by Susan Starr Sered

First published in 1994 by Oxford University Press, Inc.
198 Madison Avenue, NewYork, NewYork 10016
First issued as an Oxford University Press paperback, 1996
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

Sered, Susan Starr.
Priestess, mother, sacred sister: religions dominated by women / Susan Starr Sered.
p. cm. Includes bibliographical references and index.
ISBN 0-19-508395-4
ISBN 0-19-510467-6 (Pbk.)
i. Women and religion—Cross-cultural studies.
2. Women—Religious life—Cross-cultural studies. I. Tide.
BL 458.545 1994
200'.82—dc20 93-35557

1 09 8 7 6 5 4 3 2 1
Printed in the United States of America
on acid-free paper
PREFACE

The impetus for this study grew out of the fieldwork projects that I have
conducted among Israeli women. During the past decade I have investigated
women's rituals at Jewish and Christian shrines in Bethlehem, observed and
participated in the religious world of elderly Jewish women who came to Israel
from Kurdistan, and listened to the fertility, pregnancy, childbirth, and post-
partum rituals and beliefs of women in a maternity hospital. By choosing to
work with populations and situations in which there is a great deal of sexual
segregation, I tried to maximize my contact with women's own concerns,
ideas, and rituals.
While conducting my own fieldwork, I also endeavored to seek out studies
of other societies in which women have been able to develop some of their
own religious rituals and interpretations. Many of the themes that I found
relevant to Jewish women's religious lives in Israel resonate strongly with
those described in the literature on Christian women in Europe and Latin
America, Hindu women in India, and Islamic women in the Arab world. It
became increasingly clear to me that when women have opportunities to ex-
press their own religious ideas and rites, themes peripheral to men's religious
lives emerge as central to women's religiosity.
My work in Israel led me to wonder if similar themes are important to
women who have even more control over their religious lives—women who
lead or join women's religions (as opposed to women who have some freedom
to maneuver within men's religious structures). This book was written in
order to explore that question.
In the course of researching this book, I learned about women who have
developed or participated in religions that enhance their lives, validate their
concerns, sacralize their bodies, and meet their gender-specific needs. I hope to
share with readers of this book some of the excitement I felt in finding out
about religious situations so different both from the one in which I was raised
and from those I encountered in my fieldwork.

Jerusalem, Israel S. S. S.
August
ACKNOWLEDGMENTS

Writing a book that covers diverse examples and draws upon more than one
discipline is always a risk. Many people helped me reduce that risk by taking
the time to check the accuracy of my work. In particular, I wish to thank the
following people who were kind enough to read sections of this book that deal
with their areas of expertise. I. M. Lewis and Yoram Bilu helped me sort out
my ideas on spirit possession (Chapter 9). Daphna Izraeli explained the funda-
mentals of organizational sociology to me (Chapters n and 12). Kaja Finkler
offered important comments regarding women's illness and healing (Chapter
5). Judith Lorber took the time to read and help me organize the material on the
secular benefits that accrue to women who belong to female-dominated reli-
gions (Chapter 13). Shimon Cooper demonstrated how classical anthropolo-
gists treat social structure and lineality (Chapter 2). And Joel Gereboff offered
me a theologian's insights into immanence and transcendence (Chapter 7).
Special thanks go to the experts who were both generous and broadminded
enough to share their time and expertise with me. The following people helped
me verify the accuracy of my information. Teigo Yoshida and Cornelius
Ouwehand (Ryukyu Islands), Melford Spiro (Burma), Eric Cohen and Gehan
Wijeyewardene (Thailand), Laurel Kendall (Korea), Steve Kaplan and Janice
Boddy (zar cult), Esther Pressel (Afro-Brazilian cults), Starhawk and Julie
Greenberg (Feminist Spirituality Movement), Priscilla Brewer and Stephen
Stein (Shakers), Virginia Kerns (Black Caribs of Belize), Susan Setta (Christian
Science), and Carol MacCormack (Sande).
A number of people read all or parts of the manuscript. In particular, I wish
to thank Ross Kraemer, Carol MacCormack, Rosemary Ruether, and Adele
Reinhartz. And finally, I continue to appreciate the ongoing inspiration that I
receive from Harvey Goldberg and Zvi Werblowsky.
My deepest appreciation to Ariella Zeller, Laini Kavalovski, and Cynthia
Read, all of whom helped me pull together this often unwieldy project.
CONTENTS

Introduction 3
1 The Examples 11
2 The Social Contexts of Women's Religions: Gender Disjunction,
Matrifocality, and a Critique of Deprivation Theory 43
3 Maternity and Meaning 71
4 When Children Die 89
5 Misfortune, Suffering, and Healing 103
6 Rituals and Relationships 119
7 The Sacred in the Profane 145
8 No Father in Heaven: Androgyny and Polydeism 161
9 Summoning the Spirits 181
10 Gender Ideology 195
11 Leaders and Experts 215
12 Women, Sacred Texts, and Religious Organization 243
13 Spiritual Gifts and Secular Benefits 257a
Conclusion 279

v APPENDIX A Alphabetical Summaries of Key Examples 289

APPENDIX B Alphabetical Summaries of Auxiliary Examples 291
REFERENCES
Key Examples 293
General 305
INDEX 317
Priestess, Mother,
Sacred Sister
Introduction

Ethnographic and historical studies of women and religion have thoroughly

documented patterns of women's exclusion from positions of significant reli-
gious leadership. In many societies women have active religious lives, yet
ecclesiastic hierarchies rarely include women, and official or "great tradition"
religious concepts generally reflect men's and not women's priorities and life-
experiences.
Scattered throughout the world and the centuries, however, are instances of
religions dominated by women—in which women have been the leaders, the
majority of participants, and in which women's concerns have been central.
Through analysis of religions dominated by and oriented toward women, this
book explores the questions: What are the cultural, structural, and historical
circumstances most likely to allow women to develop autonomous religious
systems? What are the salient characteristics of women's religions? Does the
fact that women control a religious system mean that there is anything "fe-
male" about the content or structure of the religion? These intriguing questions
have not yet been seriously addressed by either social scientists or historians of
religion.
The key examples developed in this book are ancestral cults among Black
Caribs in contemporary Belize, the indigenous religion of the Ryukyu Islands,
the zdr cult of northern Africa, the Sande secret society of Sierra Leone, ma-
trilineal spirit cults in northern Thailand, Korean shamanism, Christian Sci-
ence, Shakerism, Afro-Brazilian religions, nineteenth-century Spiritualism,
the indigenous nat cultus of Burma, and the Feminist Spirituality Movement in
the twentieth-century United States.
These twelve examples are particularly appropriate for a number of reasons.
First, they are all unquestionably female dominated both in terms of leadership
and membership; that is, the majority of their leaders and members are
women. In none of these religions is there discrimination against women on
any level of leadership or participation. Additionally, in all these examples an
awareness exists on the part of leaders and/or members that this is a women's
religion. And finally, in each case there is some sort of recognition that this
4 Priestess, Mother, Sacred Sister

religious group is independent from a larger, male-dominated institutional

context. All the examples I have chosen to describe are sufficiently well-
documented to serve as the basis for wide-ranging comparisons.
I am indebted to the scholars who have researched and written about each of
these religious situations. In particular, the work of I. M. Lewis, Janice Boddy,
Carol MacCormack, Virginia Kerns, Ann Braude, Carol Christ, William Lebra,
Starhawk, Gehan Wijeyewardene, Melford Spiro, and Laurel Kendall has en-
abled me to ask the sorts of comparative questions that this book addresses.

WOMEN AND RELIGION

Feminist anthropologists have determined that although male dominance is
near universal, the forms taken by patriarchy vary from culture to culture. In
one society women may be excluded from political decision-making yet wield
extensive economic power; in another society women may be barred access to
high prestige roles yet still amass a great deal of authority and autonomy
through the existence of women's organizations. In general, feminist anthro-
pology has treated religion as the ideological foundation of and justification for
patriarchy; through religious doctrine and ritual, women and men are per-
suaded of the "rightness" of male dominance.
However, in many cultures—including highly patriarchal cultures—it is
precisely within the realm of religion that many women find avenues for
sacralizing their domestic lives (e.g., through holiday food preparation), accept-
able outlets for their frustration with subordination (e.g., through rituals of
reversal), or opportunities to celebrate with other women (e.g., at life-cycle
rituals). Perhaps because religion frequently emphasizes such internal experi-
ences as belief, faith, and mysticism (all of which are difficult to legislate or
supervise), in many cultural situations women have found room to maneuver
within the religious sphere.
The study of religion is of profound importance to feminist scholarship
because it is so often via religious rituals and ideologies that women and men
express their deepest concerns, their truest selves, their fears, hopes, and pas-
sions. Studies of women's religious lives—whether utilizing textual, archeo-
logical, ethnographic, historical, sociological, or psychological techniques—
seek to discover how women construct reality.
Although the academic study of women and religion is still relatively new
and lacking in theoretical focus, several patterns have begun to emerge. These
patterns, drawn from studies of male-dominated religions, are briefly re-
viewed below.

THE STUDY OF WOMEN AND MALE-DOMINATED

RELIGIONS
Although women are frequently excluded from certain leadership roles (e.g.,
as priests within the Catholic Church), other sorts of leadership roles tend to be
open to women. In particular, women often serve as shamans and charismatic
 Introduction 5

preachers. Among the factors that encourage women's attainment of authority

roles are emphasis on personality or supernatural powers rather than on hierar-
chy or the training of cult/church leaders, and spiritual identity predominating
over social categories (Haywood 1983; Weinstein and Bell 1982). These factors
are particularly significant in new religious movements, where women often
function as founders/leaders, losing their leadership roles as the religion be-
comes more institutionalized.
Neither the presence of women founders nor devotion to goddesses or female
saints necessarily translate into equal privileges for all women in the religious
cult. And equality of the sexes in relationship to God may co-exist with com-
plete male monopolization of leadership roles, religious law, and authority in
community affairs (as, for example, in traditional Judaism). On the other hand,
even within religious frameworks that exclude women from positions of author-
ity, women may be active participants. Australian women, for example, attend
church more frequently than men (de Vaus and McAllister 1987) and in many
European and North African societies women are the majority of devotees
making pilgrimages to the shrines of saints (Mernissi 1977).
David Moberg (1962) has studied gender and religious participation in
North America. He found that two factors seem to encourage women and
discourage men from joining particular churches. First, churches that offer
leadership opportunities for women draw fewer male members. Second,
churches (such as Pentecostal and Holiness churches) that encourage highly
expressive emotions tend to attract more female than male members. Phrased
differently, his findings suggest that women are particularly attracted to reli-
gions with female leaders and emotional rituals.
There is evidence that religions of the disprivileged classes—in contrast to
aristocratic cults—allow more equality to women. Max Weber (1966 [1922])
has argued that women are more receptive to all religious prophecy except
that which is clearly military or political in orientation. Weber contends that
members of disprivileged classes are particularly drawn to salvation religions
which promise that an individual's true worth will be acknowledged at a
future time. In general, disprivileged groups are drawn to religions that
assure fair compensation—reward for one's own good deeds and punishment
for the bad deeds of others.
Frequently there is greater female participation and influence in religions that
teach brotherly love and love for one's enemies. Ross Kraemer, for example,
suggests that very few women joined the Qumran religious community be-
cause "the entire cosmology, theology, and symbolic universe of Qumran was
so pervasively male that no woman would have found it acceptable, let alone
compelling. Asceticism as a by-product of cataclysmic war may have been
incomprehensible to women. Asceticism derived from less militaristic eschatol-
ogy, such as that of earliest Christianity, or asceticism unconnected to eschatol-
ogy, such as that at Mareotis, may have held considerably more appeal for
women" (igSob, 306-7).
Women's religious lives are often closely linked to their interpersonal con-
cerns. The network of relationships that seems most relevant to the understand-
6 Priestess, Mother, Sacred Sister

ing of women's religiosity is the family. Historical and ethnographic accounts

describe women's religious activity as embedded within, complementary to, en-
riching of, growing out of, and occasionally rebelling against women's familial
involvements. An intense concern with the well-being of their extended families
characterizes the religious lives of many women. Women who have a great deal
invested in interpersonal relationships, and are denied access to venues of formal
power, tend to be associated with religious modes that stress relationship. Wil-
liam Christian, studying women in rural Spain, has commented that "Women
are more likely to fix on personal patrons [saints] than are men. . . . Such a close
relationship carries over into old age when practical requests from the women
still center on the children, now grown up, and when there is a deepening of the
personal, affective ties between the woman and the divine patron" (1972, 133).
Even within male-dominated religious contexts, women "domesticate" reli-
gion by emphasizing rituals and symbols that give spiritual meaning to their
everyday lives. In many cultures food is one of the few resources controlled by
women, and food indeed plays a central religious role for women cross-
culturally. This may take the form of feeding the poor, choosing to give better
or more food to certain categories of people, observing food taboos, abstaining
from food, or preparing holiday, symbolic, or sacramental foods (Bynum
1987). Food rituals imbue with holiness the everyday domestic work of many
women (Sered 1992).
Illness plays a significant role in women's religious paths. Among saints in
medieval Europe, fortitude in illness was the one form of asceticism "in which
women were not merely statistically overrepresented but constituted an abso-
lute majority" (Weinstein and Bell 1982, 235). In a large number of religious
situations women are active in healing rituals. Their involvement may be as
participant, healer, or both participant and healer. Women in the United States,
for example, are disproportionately represented in cults concerned with health
service and therapy (Stark and Bainbridge 1985). Healing tends to blur the
distinction between pragmatic care for family members and religious exper-
tise. Women who seek out and join religious groups frequently do so as a result
of illness—either their own or that of a family member.
Studies of women and religion are notable for emphasizing ritual instead of
theology. As a result, we know a great deal more about what women do than
what women believe. This situation may reflect the preferences of researchers,
or it may imply an actual tendency on the part of women cross-culturally to
invest more energy into ritual than into theological speculations. There is some
evidence from Afro-Baptist churches in the southern United States, for exam-
ple, that men are more interested in theology—the nature of God and the
meaning of life—while women are in charge of rituals and caring for the
church and congregation (Pitts 1989).
Within patriarchal cultures women are not always at liberty to follow their
religious callings. Some religious women (particularly women opting for celi-
bacy) become subject to rejection, ridicule, and even physical harm at the
hands of their families, communities, or other authorities. Intensive religious
involvement for women may be seen as threatening familial strength, commu-
 Introduction 7

nity solidarity, or government control. A recurring motif in the life stories of

Christian women martyrs of the Syrian Orient during the fourth to seventh
centuries, for example, is gruesome torture as punishment for "refusal to
marry, for the common Christian practice of virginity was unsettling in its
social impact, eroding as it did the most fundamental aspect of the social and
economic order, the family" (Brock and Harvey 1987, 16).
Other external factors that have an impact on women's religious options
include elements of the political and economic milieu. Jane Schulenburg
(1989), for example, has shown that in seventh-century Europe women's mon-
asteries flourished, at least in part because this was a time of weak and decentral-
ized royal and ecclesiastical authority. Society was open, fluid, peaceful, and
prosperous, and the Church was not yet highly organized. Threats to women's
monasteries during the eighth through eleventh centuries came both from
political events—the Viking invasions—and from economic factors—noble
families needed their women for strategic alliances.
The themes and patterns outlined in the previous paragraphs grow out of the
existing literature on women and religion, a literature that for the most part
looks at religious situations not dominated by or oriented toward women. The
issues raised by that literature encourage other sorts of inquiries: Are religions in
which women are the majority of participants and leaders less hierarchical or less
centralized than male-dominated religions? Are women especially drawn to
rituals or to altered states of consciousness rather than theological learning? Do
women's religions stress positive interpersonal relationships? How do women's
religions relate to fertility, menstruation, menopause, and celibacy? When
women are in control, are food preparation, illness, or childcare ritual foci? Do
women's religions emphasize female symbols or goddesses? Does the presence
of a religion dominated by women affect women's status in the society at large?
Does the social structure of the society at large affect the occurrence of women's
religions? Does female leadership repel male membership?

THE FINDINGS
A preliminary issue that a book of this sort needs to address is the legitimacy of
talking about gender and religion in a comparative or universal sense. As an
anthropologist I am committed to the idea that religious beliefs and rituals are
not inscribed on x or y chromosomes.1 If that is the case, why would we
expect to find trends or patterns in women's (or men's) religiosity?
The answer to that question lies in the fact that there are cross-culturally
relevant social patterns in women's lives. To begin with, in every known
culture adult women grapple with motherhood. Most women are, have been,
or try to become mothers, or conversely, make efforts—sometimes even life-
threatening efforts—to avoid becoming mothers. Many, if not most women,
are concerned with controlling the number of children whom they bear and
raise, and with determining the way in •which their children are raised. The
diverse implications of motherhood, as I demonstrate throughout this book,
strongly resonate with women's religious beliefs and rituals.
8 Priestess, Mother, Sacred Sister

At this point, it is sufficient to propose that a range of aspects of motherhood

could be envisaged as relevant to women's religions. First, we might expect
that the physiological processes of fertility—pregnancy, childbirth, lactation—
receive attention in women's religions. Second, 1 would anticipate finding the
experience of nurture and childcare—that is, motherhood in the social sense of
ongoing and intimate relationships with children of both sexes—to be ad-
dressed and possibly enhanced in these religions. In addition, I would expect
these religions to in some way exhibit the facility with interpersonal relation-
ships that women develop as a result of being raised by women (Chodorow
1974; Gilligan 1982). And finally, assuming that birthing and raising children
have an effect on how women experience the world, I would expect to see
some form of "maternal thinking" present in women's religions (Ruddick
1982). These ideas will be referred to repeatedly, especially in Chapter 3, and
their validity will be assessed in the Conclusion.
A second pervasive theme in the cross-cultural study of gender concerns
male dominance. A great deal of feminist anthropology has been dedicated to
the question of whether male dominance is universal. Two major schools of
thought have emerged. One school claims that male dominance looks as if it is
universal, and concentrates on figuring out why (e.g., Ortner 1974). The other
school contends that male dominance is not universal, and concentrates on
finding evidence of egalitarian societies (e.g., Sanday and Goodenough 1990).
In recent years, more and more anthropologists have come to the conclusion
that it is unhelpful to rate entire cultures as patriarchal or not patriarchal. Cul-
tures are complex—men may dominate in one realm and women in another; and
even within a given culture different women have different life-experiences—
old women may be much freer than young women, and women of the nobility
are likely to exert greater power than poor women. The evidence of women's
religions certainly strengthens this line of argument—the very existence of these
religions shows that male dominance is not universal or absolute. On the other
hand, none of these religions occurs in matriarchal or even fully egalitarian
societies. Therefore, when I ask what the common elements of women's experi-
ences are—elements that help us understand women's religions—patriarchy is
among them. Yet this does not mean that all women's religions respond to
patriarchy in the same way.
Before continuing, I wish to offer a preview of what is surely the most
important finding of this study. The impact of gender on religion is quite
limited—women's religions are not so very different from men's religions.
Female-dominated religions, like male-dominated religions, worship super-
natural beings, perform rituals of thanks and appeasement, utilize techniques
that induce altered states of consciousness, and provide devotees with persua-
sive explanations for the ultimate conditions of existence. It is necessary to
make this statement precisely because "Woman" in western cultures is so often
treated as inherently "Other"—as incomprehensible and mysterious. Having
said this, I do believe that the data offered in this book show that gender has a
significant—although not absolute or universal—impact on how people image
supernatural beings, on the form and interpretation of the rituals performed,
 Introduction g

on whether and why one seeks altered states of consciousness, and on the
manner in which individuals grapple with the ultimate conditions of existence.
In short, the material in this book demonstrates that although gender has an
important influence on religiosity (and religion has an important influence on
conceptions of gender), an individual's religiosity is not determined by his or
her sex. Indeed, the two near-universal elements of women's experiences dis-
cussed in the previous paragraphs—motherhood and patriarchy—are them-
selves culturally constructed, and thus intersect with religion in disparate and
sometimes contradictory ways. 2

HOW THIS BOOK IS ORGANIZED

Chapter i introduces the key examples and situates each in its historical context
and in relationship to other religious options available at the same time and
place. Chapter 2 looks at the cultural and social structural settings of the
religions, and concludes that women's religions tend to occur in matrifocal
societies. In Chapter 3 I show that women's religions address women primar-
ily as mothers. Chapter 4 focuses on one aspect of motherhood—child death—
as an impetus for women's religious activity. In Chapter 5 I argue that women
(mothers) are dissatisfied with religions that claim that suffering is necessary or
unavoidable. As a result, women's religions tend to offer eclectic and persua-
sive explanations for and solutions to misfortune and illness.
Chapter 6 explores the orientation toward interpersonal relationships found
in the ritual complexes of women's religions. Subsections of Chapter 6 deal
with food rituals, mourning rituals, and initiation rituals.
Chapter 7, really the pivotal chapter of the book, argues that women's
religions are characterized by a this-worldly orientation that does not empha-
size the difference between sacred and profane spheres. Chapter 8 asks what
kinds of deities and spirits interact with humans in women's religions. Consis-
tent with women's this-worldly orientation, almost all the religions posit
highly immanent deities. Chapter 9 concentrates on the most dramatic manifes-
tation of immanence—spirit possession—and explores the connection between
women and possession trance.
Chapter 10 turns to the question of gender ideology, asking whether
women's religions postulate egalitarian, matriarchal, or patriarchal views of
gender. Chapter n investigates the life histories and roles of women who are
leaders of women's religions. Chapter 12 explores the tendency toward non-
centralization, non-standardization, and lack of codification of sacred scripture
in women's religions. Chapter 13 asks whether women's religions serve
women's secular interests.
Finally, the Conclusion returns to the question of whether there is anything
intrinsically "female" about religions dominated by women, and evaluates the
differences between women's religious activities in female- and male-dominated
religions.
Within each chapter, shorter sections zero in on patterns that are relevant to
many or most women's religions. Each section explores in some detail one or
io Priestess, Mother, Sacred Sister

more of the key examples. Since information about each religion is scattered
throughout the thematically organized chapters, readers who wish to find out
about theological, ritual, contextual and structural aspects of a particular
religion are referred to the index. I urge the reader not to be overly concerned
with keeping track of the many examples; my main aim in this book is to
uncover thematic patterns, not to present comprehensive accounts of a lim-
ited number of religions. The Appendixes briefly summarize each religion.
The Bibliography should be helpful to readers wishing to learn more about
particular examples.
In light of the thematic (rather than geographic or historical) organization of
this book, it is in order to make a few comments concerning cross-cultural
research. A serious challenge to any cross-cultural study concerns the nature of
comparisons: To what extent are religions that developed on different conti-
nents during different centuries comparable? Anthropologists have long grap-
pled with this issue, and with few exceptions have concluded that without
comparison anthropology ceases to be the study of human culture and be-
comes merely a technique for collecting exotic customs. While the aim of this
book is unabashedly comparative, I strive throughout to remain sensitive to
cultural context.
I have written this book because I believe that Okinawan and Afro-Brazilian
priestesses, Black Carib grandmothers, Burmese and Sudanese housewives,
Sande initiates, Spiritualist mediums, Shaker celibates, American Spiritual
Feminists, Christian Science practitioners, and Korean shamans do share a
common bond—they are all active in religions that address women's concerns
and encourage women's religious leadership. Tracing the similarities and the
differences among them offers important insights into the intersection of gen-
der and religion.

Notes
1. It is a tenet of anthropology that one always begins by exploring social explana-
tions for differences between groups. Only if all possible social explanations fail is it
legitimate to turn to biological explanations.
2. It has become increasingly fashionable in academia to share with the readers one's
intellectual and personal progression in writing a book. I will take this opportunity to
tell the reader that when I began researching this book I did not have any sort of
hypothesis in mind; I was simply looking for themes and patterns in women's religions.
Once I had finished a first draft, it became clear that motherhood and patriarchy seemed
to explain almost all the interesting patterns. As I continued to ponder my data, I
realized that these two factors are also what legitimate the expectation that there will be
near-universal patterns in women's religious experiences.
 1
The Examples

DEFINITIONS
Women's religions are, from a cross-cultural perspective, anomalous. Despite
my growing excitement as I "discovered" more and more women's religions
buried in the pages of ethnographic and historical tomes, the fact is that most
of the religions of the world are dominated by men. The anomalous status of
women's religions leads us to ask when, why, and how they occur.
In the following pages twelve examples of religions dominated by women
are presented. The descriptions in this chapter should help the reader follow
the thematic analyses in subsequent chapters. (Concise summaries of the main
features of each religion can be found in the Appendixes.) But before we begin
our journey into the worlds of women's religions, it is worthwhile to point out
ways in which these religions differ from one another.
Some of the examples are self-consciously independent religions that exist in
a society where the dominant religion is male dominated (e.g., Feminist Spiri-
tuality, Afro-Brazilian religions). Others are religious streams that co-exist
alongside of, and sometimes intertwined with, male-dominated religions (e.g.,
zar, Spiritualism, Korean shamanism, Burmese nat cultus). Still others are the
major religion of an entire society (e.g., the Ryukyu Islands, the Black Caribs
of Belize). And finally, others are sects of otherwise male-dominated religions
(e.g., Christian Science, Shakerism).
One problem I faced in writing this book concerns terminology: the ex-
perts whose research I rely on use uneven nomenclature to describe their
religions. Among the terms used are "cult," "cultus," "sect," "religion,"
"group," "society," and "movement." In addition, at least one of the exam-
ples (Korean shamanism) is probably most accurately described as a "reli-
gious situation." Rather than repeat the phrase "religions, religious cults,
groups, cultures, sects, societies, and situations that are oriented toward and
dominated numerically and in terms of leadership by women," I have chosen
to gloss all these terms as "women's religions" or "female-dominated reli-
gions." 1 Although my decision may offend some scholars, in light of the fact
12 Priestess, Mother, Sacred Sister
that all of these terms are etic (imposed from the outside by western scholars)
and not emic (the terms used by members of the various religious groups),
my decision is defensible—provided that I show each example in its proper
societal context. I wish to clarify that I use the term "female-dominated
religions" to indicate that women are the majority of participants and leaders,
there is no higher level male authority that ultimately directs these religions,
and that these religions focus on women as ritual actors. By female domi-
nated I do not mean to imply physical dominance or institutionalized power
inequality between men and women. I use the term "women's religions"
interchangeably with "female-dominated religions."
Some examples are all-women religions (Sande is a women's secret soci-
ety), others include some men (Spiritualists do not exclude men, just more
women join), and still others include many men yet limit leadership to
women (on the Ryukyu Islands men attend religious ceremonies, yet the
leaders are all women). The implications of these distinctions are significant,
and it may well be that future studies will narrow down the field of inquiry.
In thinking about the religions described in this book, it is helpful to treat the
diverse examples as points on a continuum, rather than as a strictly homoge-
neous group.
In this chapter I am not comparing "women's religion" to "men's reli-
gion." There is no such thing as an archetypical "women's" or "men's"
religion; I do not believe that religiosity is a biologically determined sexual
characteristic. My goals are far more modest: I am comparing specific reli-
gions to other religion(s) of the same place and time. Among the questions I
ask are: Is this religion the only option available at the given time and place?
If not, how is it different or similar to other religions of its day? What is the
relationship of this religion to other religions of its day (nat religion vs.
Buddhism; Thai spirit cults vs. Buddhism; Korean shamanism vs. Confucian-
ism; Afro-Brazilian cults vs. Catholicism; Feminist Spirituality vs. American
civil religion; African cults of affliction vs. ancestor worship; Spiritualism,
Christian Science, Shakerism vs. Calvinism)? What is the status of this
religion—is it the official state religion, a new religion, a persecuted cult?
What is the historical and cultural context of this religion? In what way is this
religion female dominated?
These religions are not randomly scattered throughout the globe. In fact, all
of them are clustered in three loosely defined culture areas.

EAST AND SOUTHEAST ASIA

Four of the women's religions are located in East or Southeast Asia. The
similarities among these religions will become apparent as we look at each
religion in turn. What I wish to point out here is that unlike most of the
women's religions outside of East and Southeast Asia, all these religions seem
to be indigenous, probably ancient, religions. While they all have undergone
changes in the past century, none of them can be identified as having arisen at a
 The Examples 13

particular point in the historically recorded past; none has a named historical
founder. 2
Japanese scholars speculate that the ancient religion of Japan was dominated
by women, and that vestiges of that domination can still be found in modern
Japan. According to Teigo Yoshida (1989), the continued religious domination
of women in the Ryukyu Islands reflects the failure of Buddhism to have made
much of an impact there. Whether or not Yoshida is correct, it is clear that in
East and Southeast Asian societies in which the indigenous religions are domi-
nated by women, men tend to be involved in newer religions. Thai and Bur-
mese men are active Buddhists—many if not most young men enter Buddhist
monasteries as novitiates for several years, and Korean men are responsible for
Confucianist ancestor worship. In three of the four key East and Southeast
Asian examples, men embrace religions that came in from the outside while
women dominate the indigenous religion.
I would tentatively suggest two possible historical scenarios. First, it may be
that in very ancient times the only religions in East and Southeast Asia were
dominated by women. When new religions (Buddhism, Confucianism) were
introduced into the area, men were especially eager to adopt the new religions
because of their subordinate status in the traditional religions. 3 Alternatively, it
may be that in ancient times both men and women were involved in the
indigenous religion. For known or unknown social reasons, men more than
women were attracted to the new religions, leaving the ancient religion in
female hands, where today it has an ambivalent and often inferior status. It is
significant that the male-dominated "new" religions of East and Southeast
Asia, unlike the female-dominated indigenous religions, preach doctrines of
female pollution or subordination.
It is perhaps not unexpected to find that women's religions have continued to
thrive in Buddhist countries. In contrast to Islam or Catholicism, Buddhism is
relatively tolerant of other religions, and generally allows adherents to partici-
pate both in Buddhist and non-Buddhist religious rituals. In East and Southeast
Asia it is quite common to find that the men of a family are active Buddhists,
whereas the women are dedicated participants in women's religions. Such a
situation would probably not be tolerated in Islamic or Christian societies.

The Religion of the Ryukyu Islands

Religion on the Ryukyu Islands is inseparable from ethnicity or nationality; the
religion has no name, it is simply the religion of the people of the Ryukyu
Islands. The Ryukyu Islands, of which Okinawa is the main island, extend
south and east from the south of Japan. They are a combination of sparsely
populated, well-populated, and totally unpopulated islands and coral reefs.
The Islands have, at various points in their history, been an independent king-
dom, a tributary to China and Japan, incorporated into the Japanese Empire,
and occupied by the United States (after World War II). Since 1972 the Islands
have been part of Japan. Yet because of the isolation provided by the sea, many
14 Priestess, Mother, Sacred Sister

old customs have survived. Ryukyuan culture has been notably unreceptive to
outside beliefs, and has typically reinterpreted foreign elements in light of its
own, indigenous cultural system. 4
Ryukyuan religion is characterized by a highly elaborated ritual calendar and
sacred geography. Myriad events, dates, objects and locations are reasons for
priestesses to perform religious rituals. Most rituals are directed toward appeas-
ing, notifying, or thanking the divine beings known as kami. 5
The most outstanding feature of Ryukyuan culture is the belief in the spiri-
tual predominance of women; on the Ryukyuan Islands women dominate the
religious life of the family, community, and (in the past) state. Only women
can officially mediate between the supernatural and human beings, women are
expected to be much more knowledgeable about religious matters than men,
and men are required to participate in religious rituals led by women. It is of
interest that the Ryukyuans have adopted certain Chinese ancestral rites in
minute detail—with one exception: in Ryukyu unlike China the rites are con-
ducted by women. In general, Ryukyuans are dissatisfied if rituals are con-
ducted by men. Ryukyuan men do not even pray at the household hearth.
Women are so thoroughly associated with the sacred that in the past the rare
man who entered the sacred groves had to practice a form of ceremonial
transvestism—that is, dress like a woman. This was true even for the chief
ministers of state.
While the precise role of priestesses and the exact dates and compositions of
rituals vary somewhat from island to island (Mabuchi 1964), the religious
culture of the Ryukyus is basically homogeneous. 6 Within the household the
sister predominates in spiritual matters, and the same is true within the kin-
group, cult groups (in the southern Ryukyu), village, and state. All public and
almost all private religious rituals and festivals are conducted by women. In
addition, personal problems are often solved by shamans, most of whom are
women.
According to the Ryukyu Islanders, men are spiritually protected by their
sisters. The sister of the kin-group head is in charge of domestic rituals; the
sister of the village headman is the village priestess; the sister of tne king on
Okinawa was the head priestess of the kingdom. Priestesses who marry out
because of virilocal residence rules continue to return to their natal houses to
act as priestesses for their brothers' households.
In contrast to Japan, Korea, and many other cultures in which the main-
stream or official religious practitioners (priests) are men while the marginal
practitioners (shamans) are women, on the Ryukyus both priestesses (noro) and
shamans (yuta) are women. The very few male shamans on Okinawa are men
with major physical disabilities that preclude their functioning in typical male
roles. These men have a low social status and a reputation for being emotion-
ally disturbed.
Almost all the clients who call on yuta are women. Even if the woman is
accompanied by a husband or son, it is the woman who communicates directly
with the yuta. In Ohashi's (1984) study of a middle-sized town in Okinawa,
more than half the women had contacted yutas, with the percentage increasing
 The Examples 15
as women age. Of women in their sixties, 85% had contacted yutas. When the
women were asked in what religion they believe, more than three-quarters said
"ancestor" (i.e., the traditional Okinawan religion), and among the older
women nearly 100% gave that answer.
Douglas Haring noted that the association of women and religion in the
Ryukyu Islands is similar to what we know of early Japan—but in Japan,
Buddhism and Confucianism exterminated that association. "Perhaps Bud-
dhism failed to gain wide acceptance in ... [the Ryukyu Islands] despite centu-
ries of propaganda because its priests were male" (1964, 121). On the other
hand, William Lebra believes that there has been a decline in the power of
priestesses on Okinawa beginning in the sixteenth or seventeenth century as a
result of contact with China and the influence of Confucianism with its stress
on the superiority of males, and the penetration of Japanese Buddhism empha-
sizing the ancestral cult through the male line. "From the sixteenth and seven-
teenth centuries, history reveals a persistent effort by the government to reduce
the powers of the priestesses" (1966, 117).
It must be emphasized that unlike most other women's religions that are one
option in a cultural situation providing two or more religious choices, the
Ryukyu Islands are the only known instance in which the official religion of a
people is dominated by and oriented toward women.

Burmese Nat Religion

Buddhism is the state religion of Burma. Alongside of Buddhism, however,
there is an earlier, indigenous religion that revolves around appeasement of
spirits known as nats. Nats are propitiated to prevent and cure illness, at key
stages in the agricultural cycle, at births, deaths, and marriages, and at Bud-
dhist initiations. Unlike Buddhism, nat rituals are dominated by women.
Melford Spiro, who conducted fieldwork in Burma, gives the name "super-
naturalism" to this non-Buddhist Burmese religion. Since the concept "super-
naturalism" is one that western scholars have imposed from the outside, I
prefer to focus on the central element of the indigenous Burmese religion—
belief in and rituals concerning nats or spirits—and refer to non-Buddhist
Burmese religion as "nat religion." Nat religion is, like Buddhism, an elabo-
rated and articulated system of beliefs, rituals, and practitioners, organized at
the household, the village, and the regional levels. Unlike Buddhism, nat
religion is not a literate tradition—it has no sacred writings—nor is it a world
religion—nat religion is practiced only in Burma. "Resting on a complex
mythological charter, the nat cultus consists of an elaborated ritual system
under the supervision of socially recognized cult leaders and practitioners.
Although, from the Burmese point of view, the nat cultus does not constitute a
religion, the fact remains that it rivals Buddhism in its elaborate cognitive,
ceremonial, and organizational systemization" (Spiro 1967, 40). Buddhism and
nat religion present radically different belief systems, yet there is little conflict
between the two because canonical Buddhism recognizes the existence of spir-
its and demons, and sanctions their propitiation (Spiro 1971, 4).
16 Priestess, Mother, Sacred Sister

Spiro's fieldwork was carried out in 1961—62 in a rural village in Upper

Burma approximately 10 miles from the city of Mandalay. Spiro discovered
that almost half the men but none of the women claimed not to believe in nats.
Men and women agreed that women are more involved in nat propitiation,
that women fear the nats more, and that women perform more nat rituals.
Village nat shrines are almost always tended by women, and village nat ceremo-
nies are attended almost exclusively by women. Shamans, who become pos-
sessed by nats, are almost all women. Women also bow to nats, while men do
not. Villagers explained this difference in terms of the different status of men
and women: within the [Buddhist] thirty-one abodes of existence men occupy
a higher position than nats; women occupy a lower position. Thus women are
more susceptible to attack by nats than men are. At major nat festivals, which
attract thousands of people, women outnumber men by at least twenty to one
(Spiro 1967, 123).
Every household gives offerings to the nat inherited through the mother's
matriline and to the nat inherited through the father's patriline. According to
June Nash (1966), if the parent's nats differ, the mother's nat is more likely to be
inherited than the father's nat.1 Women are more concerned than men about
propitiating nats, and it is women who give all the food offerings to the nats.
Men, who are more involved in Buddhism, tend to either cease believing in
nats or to believe that Buddhism gives them power to overcome the power of
the nats. Typically, a woman assumes the responsibility for nat offerings when
her mother, old and dying, pleads with her to carry on the propitiation or
beware of the nat's wrath.
Although both men and women are involved in both Buddhism and nat
religion, men are more identified with Buddhism and women with nat reli-
gion. Even more to the point, specialists in Buddhism (monks) are exclusively
male, and nat shamans are overwhelmingly (although not exclusively) female.
In Burma, almost all male children enter Buddhist monasteries as novitiates for
a number of years (usually during adolescence). The time spent in the monas-
tery strengthens men's ties to and knowledge of Buddhism, and reinforces
women's perception of Buddhism ,as belonging to men more than to women.
The Buddhist leaders with whom Burmese are acquainted—the monks and
teachers—are men.
Spiro believes that Buddhism is the most important cultural force in the
Burmese village. Buddhist devotions, holy days, and pilgrimages are observed
by almost all villagers. In fact, Buddhism and nat religion are so well inter-
twined that the household nat also receives an offering at the ceremony of
initiation of boys into the Buddhist monastic order, and, when disaster strikes,
a ritual to propitiate the village nat may be combined with rituals performed by
Buddhist monks. During the rice harvest one of the women of the work group
performs a nat propitiation ritual for protection against snakes. Immediately
afterwards, one of the men leads the group in the worship of a particular
Buddha image believed to have the power to cure snake bites (Spiro 1967,
249). In many cases Buddhist and nat shrines are adjacent to one another.
On the other hand, the nats can be understood as symbolizing opposition to
 The Examples 17

authority; when people participate in nat rituals they are, on the unconscious
level, expressing their dissatisfaction with official Buddhism. Burmese often
feel that Buddhist moral requirements are too strict, and that the consequences
of violating Buddhist precepts are too severe. It is significant that nat myths
often involve anti-Buddhist themes: disobeying Buddhist monks, honoring
sensuality.
Spiro has developed a comparative model that highlights some of the central
differences between nat religion and Buddhism. Whereas Buddhism empha-
sizes a moral code (right speech, right conduct, etc.), nats punish those who
offend them regardless of the moral state of the individual, and the charter
myths of the individual nats have nothing to do with moral behavior. Bud-
dhism, deeply ascetic, teaches that attachment (to people, possessions, one's
own body, etc.) is the cause of suffering and that the way to eliminate suffering
is to eliminate attachment. Nat religion is absolutely involved in this world—
with this-worldly passions, this-worldly desires, and this-worldly relation-
ships. Whereas the Buddhist monk renounces sexuality, sexual intercourse
with a nat is the path to becoming a nat shaman. For Buddhists, liberation
means dis-attachment from this world; Buddhism teaches that one should
abandon family, wealth, glory, comfort. Nats, on the other hand, are propiti-
ated solely for worldly ends—health and livelihood; nats and nat rituals are not
concerned with salvation.
At the beginning of the eleventh century King Anawrahta tried to suppress
the nat cult and enforce Buddhism as the state religion. He did not succeed in
eradicating the nat cult, but did institutionalize it by appointing the Thirty-
Seven "official" nats, and establishing a Buddhist nat as their overlord.
Both nat religion and Buddhism are deeply embedded within Burmese cul-
ture. Overlaps occur, but the two religious systems have remained distinct,
and neither shows any signs of dying out. However, the two religions are not
of equal status—Buddhism is consistently seen as more powerful. Observing
Buddhist precepts renders one less vulnerable to harm by the nats. Monks are
venerated, while shamans may be subject to ridicule and critique. The higher
status of Buddhism, coupled with the belief that Buddhism offers magical
protection even stronger than that offered by nat propitiation, leads one to
wonder why nat religion continues to flourish. While I cannot provide a com-
plete answer to that question, I would at least like to raise the possibility that
the tenacity of nat religion lies in the dominant role of women as opposed to
women's secondary role in Buddhist rituals.

Korean Shamanism and Household Religion

The Korean religious milieu includes Confucianism, Buddhism, Christianity,
and numerous new religions—all dominated by men—and the indigenous
household religion—which is the province of women. Women's religion,
often called "superstition" or "shamanism," is sometimes considered to be
everything that is not Confucian, Buddhist, or Christian. Some of its rituals
are led by a shaman, and some are carried out by women alone at home. I
18 Priestess, Mother, Sacred Sister
will refer to this religious complex as shamanism or women's household
religion. Korean shamanism, dominated by women, is a "professional elabo-
ration upon Korean household religion" (Kendall 1983, 166); shamans are the
experts in household religion. The Korean shaman is an accepted and recog-
nized religious professional.
Youngsook Kim Harvey contends that in Korea, "The predominant religion
is shamanism, despite its lack of organization, coherent doctrine, the outcast
status of its practitioners, and a long history of official suppression extending
back at least to the fourteenth century" (1976, 189). Shamanism is in practice
supported by the majority of Koreans. To get a sense of the extent of sha-
manism, Harvey (1976) cites a 1932 source that estimates one shaman per 300
residents in P'yongyan, then the second largest urban center. According to a
1972 survey there was one shaman for every 314 people in Korea. A 1982
survey showed that there was only i Protestant minister per 1000 people and
even fewer Buddhist monks than Protestant ministers (Suh 1989). Despite
modernization, urbanization, the proliferation of new religions, and organized
suppression of shamanism, shamans remain popular in Korea, and there is
evidence that in urban areas their popularity is even increasing.
Women's household religion has received little scholarly attention, yet it is
clearly a mainstream and pervasive phenomenon. Lacking ecclesiastical build-
ings and written doctrines, the religion is centered around the individual per-
son of the shaman. "Scholars of Korean shamanism carefully avoid the term
"religion" in defining the mudang [shaman] phenomenon . . . but . . . despite
its lack of teachings or doctrines, . . . [it] still deserves to be called a reli-
gion. ... It is the basic religious mindset of the Korean people" (Suh 1989, 6—
7). In Korean women's household religion, the senior housewife (and because
most families live neolocally, almost all women become senior housewives)
honors the household gods. Similarly, village women bargain with the gods on
behalf of the entire community. Larger rituals are presided over by shamans.
At an elaborate ritual, kut, the costumed shaman is possessed by a succession of
ghosts and ancestors. The job of the Korean shaman is to seek out the gods,
engage them in conversation, lure them into houses, and bargain with them.
She is an independent practitioner and does not collaborate with any sort of
male religious functionary (Kendall 1985). Among the reasons for consulting
shamans are to communicate with and placate ancestral spirits, to pick auspi-
cious days for weddings and funerals, and to divine causes of illness, misfor-
tune, and family discord.
Almost all Koreans who consult and hire shamans are women, and women
are the most numerous and enthusiastic participants at shaman rituals. At these
rituals most participants are women over the age of forty (Janelli and Janelli
1982). Not only the client but also her neighbors and friends actively partici-
pate in the shaman's kut. The women spend small amounts of money on
divinations, they dance, and they form a concerned chorus (Kendall 1985).
While this is going on, men gather in other rooms to drink and make derisive
jokes about "superstitious women" (Wilson 1983, 124). Korean women some-
 The Examples 19

times hide from their husbands their involvement (particularly financial) with
shamanism.
Male opposition to shamanism is sometimes merely a pose, and when very
ill or suffering from prolonged misfortune, many men are perfectly happy to
have their wives turn to shamans. Kendall stresses that Korean women and
shamans and their rituals are not at odds with men and their goals. Shamans
and housewives accept the values of Confucian society: children should respect
elders, the living should honor ancestors, sons should be born. "This is often
not the case, however, and shamans provide not only explanations but thera-
pies" (1989, 141). It is crucial to understand how closely the various religious
streams are woven together. "The same informant might worship at Buddhist
temples, visit shaman shrines, and set down rice cake for the household gods"
(Kendall 1983, 35). Buddhism, unlike Christianity, Judaism, and Islam, does
not make an official demand of exclusive allegiance from its followers. In
many Buddhist countries, other religions co-exist alongside, and sometimes
intimately intertwined with, Buddhism. Many Korean Buddhists avail them-
selves of shamans, and even visit Christian churches of various denominations
without feeling a conflict of loyalty. According to Kendall, Korean women see
the rituals in the shaman's shrine and the offerings they make at the Buddhist
temple as parallel. The prayers women make at Buddhist temples and sha-
man's shrines are the same: for the children, health, and a peaceful family life
(19833, 84).

Northern Thai Matrilineal Spirit Cults

In Northern Thailand groups of matrilineally related kin—people who are "of
the same spirit"—make ritual food offerings to spirits known as phii puu njaa.
Phii puu njaa are tutelary spirits who have been inherited from matrilineal
ancestors who once served them. According to Cohen and Wijeyewardene,
"We are justified in treating the phii puu njaa cult as a single cluster of institu-
tional activity" (1984, 250). Since the ethnographic literature reports consider-
able variations both in cult group rituals and in Northern Thai social structure,
the following paragraphs should be seen as an attempt to summarize in a very
general way the nature and role of matrilineal spirit cults in the context of
Northern Thai family and village life. For specific villages some of these re-
marks do not hold true.
"The role of women as custodians of the house spirit cult is one aspect of a
more general association between women and domestic spirits. Every woman
possesses a certain mystic essence, sometimes called a spirit (phii) and some-
times a teewadaa, which derives from her house and ultimately from her cult
group spirit" (Davis 1984, 266). Thai matrilineal spirits are believed to reside in
the female body and to succeed from a mother to her daughters. The spirits
lodged in a woman's body would be violated if she were to have sex or any
bodily contact with an outside man. A man giving money to propitiate his
wife's spirits is the crucial element of marriage ceremonies.
2O Priestess, Mother, Sacred Sister
"The [Northern Thai] matrifocal kinship system was legitimized by a belief
in protector (territorial) spirits . . . . This spirit cult demanded one woman per
family to reside in her household until the end of her life to take care of her
domestic ancestral spirits. . . . Through this custom, kinship lineages evolved
around the female members of related families" (Tantiwiramanond and Pandey
1987, 137).
The ritual constellation surrounding the phii puu njaa takes place within the
context of extended family groups, also referred to as cult groups. All the
matrilineal descendants of a founding ancestress constitute a cult group. Men
are formally members of their mothers' groups, but in some families husbands
join their wives' cult groups. The median size of the descent group (cult group)
is four households (with a range of 1-24). The groups never seem to be more
than jsix generations in depth, and the oldest ancestors remembered are typi-
cally a set of sisters. Ideally, the group spirit is lodged in a shrine located in a
house site containing the original house, in which a female member of the
senior generation currently lives. This woman is both descent group head and
ritual officiant.
While in a very few ethnographic accounts men appear as ritual officiants of
cult groups, most ethnographers have reported that older women officiate at
all cult rituals. (A man, however, may fill in as cult leader after the death of the
previous woman leader and until an appropriate young woman of the next
generation comes of age.) Typically, men, having little to do with the cults,
claim that they are women's business.
There has been scholarly disagreement as to the function of the cults: Some
say that the primary function of the cults is to allocate rights to land, and the
ritual aspects are secondary. Others argue that the cults serve(d) as an institu-
tion of social control, and especially of controlling female sexuality. Andrew
Turton contends that descent groups and their cults have important jural and
symbolic, rather than economic, functions. He quotes a village headman:
"Even within the memory of my mother there were no officials, no law . . .
not even a village headman, no single leader, only the old men . . . then the
phii puu yaa (descent group spirits) were the law . . . though not really law, it
was mutual respect" (1976, 214).
In recent times many Northern Thai men and women have moved away
from their villages to urban centers. Ethnographers have found that adult
women who have moved away from their natal villages and from the authority
of their parents remain under the control of their cult group spirits.
In the cities, professional women spirit mediums have begun to replace
female family cult leaders. In many instances these mediums claim to be pos-
sessed by their own ancestral spirits, thereby guaranteeing the continuation of
the traditional cults (Irvine 1984, 316). Gehan Wijeyewardene believes that this
contemporary mediumship "had its institutional anchor in the matrifocal spirit
cults, though there is no reason to suppose that its practice was confined to
these cults" (1986, 153). Although urban mediumship is not an obvious or
spectacular phenomenon, most urban Northern Thai probably know how to
find a spirit medium, and some people are acquainted with several. During the
 The Examples 21
ninth lunar month groups of women (and some men) get together and dance in
public. This festival, according to Wijeyewardene, "establishes the institution of
urban membership as a public, culturally sanctioned one, rather than some
vaguely illicit, private practice" (1986, 223).
Although Thais have traditionally been tolerant of other religions, Bud-
dhism is the official religion of Thailand, and the Thai king is required by the
constitution to be a Buddhist. In Thailand, as in Burma, the Buddhist temple is
the symbolic center of village life, monks are the most prestigious residents of
the village, the annual cycle is structured around Buddhist festivals and ritual
events, and contributions to monks and the temple take a significant amount of
village money.
Buddhism defines women as inherently lower in religious status than men.
Women are excluded from the sangha (community of monks), while many
Thai men become monks, or at least novices, for some portion of their lives.
Whereas a man's major merit-making act (in Buddhist terms) is ordination, a
woman's is giving a son for ordination. The rationale for excluding Thai
women from Buddhist monastic orders is the fear that women would lead holy
men astray. In this system, monks are always meritorious whereas the laity are
always deficient in merit. This compels the laity (especially women) to be
economically active in order to have enough to give food to the monks,
whereas monks are exempt from economic activity. Monks welcome women's
economic support, but are suspicious of their sexual and reproductive powers.
The belief that a son as a monk can accrue special merit for his parents (espe-
cially his mother) means that sons and daughters in a family are treated differ-
ently. The son is excused from household chores while girls are expected to be
active in the house. Sons are given greater consideration. Daughters are taught
to serve and yield to elders and males. Being a mother or wife does not give
one merit. A mother gains supreme merit only when her son is ordained.
Tantiwiramanond and Pandey refer to this as a "merit trap" for women, induc-
ing women to become mothers of sons and depreciating women's economic
contributions (1987, 131). A son could repay his debt to his parents by becom-
ing a monk. A daughter could repay this debt only by being a lifelong care-
taker of her parents.
Because women were deprived of entry into monasteries, they were also
deprived of literacy, which until the twentienth century was acquired only in
Buddhist monasteries. Lack of literacy meant lack of access to knowledge of
medicine, arts, and lack of social mobility and political participation. "Bud-
dhism became a legitimizing agent for the Thai patriarchy to affirm and sanc-
tion the role of women which was limited to reproduction and economic
production" (Tantiwiramanond and Pandey 1987, 132). Although women are
seen as being on a lower plane of karma than men, in many other ways laymen
and women are not greatly differentiated in Theravada Buddhism, and the
distinction between male and female appears to be less important in old age as
sexual activities wane.
Thai women are very involved in economic activity. Thomas Kirsch (1985)
interprets this as reflecting women's lower status in Buddhism; economic
22 Priestess, Mother, Sacred Sister

activity is one more symptom of being "rooted" in the world. Charles Keyes
(1984), on the other hand, argues that popular Buddhist texts portray women
as inherently good, especially as mothers. Through their personal experiences
of the loss of lovers and children, women are more easily able to understand
Buddhist teaching. Kirsch, in return, contends that women's attachment to
their children has a negative valuation in Buddhist ideology. According to
Tantiwiramanond and Pandey, Buddhism has played a crucial role in subordi-
nating women. "Through religion, women internalized a view of themselves
as the subordinate sex in society. A woman not only had to become a mother—
to provide fresh labor for the subsistence agricultural system—but also had to
bear at least one son—in order to be eligible for 'extreme merit' " (1985, 142).
Wijeyewardene (1970; 1977) contrasts the inner shrine in the Buddhist mon-
astery in which monks conduct rituals from which laymen (and all women) are
excluded, with the spirit cult shrine located in the eldest woman's bedroom
from which unrelated males are excluded. He believes that the spirit cults are a
female response to the male domination of the monastic Buddhism and to Thai
ideas about female pollution. However, since there is no reason to believe that
Buddhism predates the matrilineal spirit cults (indeed, the opposite is more
likely), we can equally well argue that the ease with which Buddhism con-
quered Thailand constituted a male response to the female domination of
household religion.8
In a fascinating study, Steven Piker (1972) has looked at how Thai individu-
als deal with the conflicting beliefs of Thai spirit religion and Buddhism. He
found that most people feel that the two sets of beliefs are coherent. One of the
most common ways that Thai villagers reconcile conflicting beliefs is by saying
that they themselves are not learned in Buddhism and that if one wants to
understand these questions one should go to the city and ask the really learned
monks.9

NORTH AMERICA
The second culture area in which women's religions cluster is North America,
specifically during the nineteenth and twentieth centuries. The confluence of a
variety of historical and cultural factors has given rise to this cross-culturally
rather anomalous situation.
To begin with, neolocal living arrangements combined with high geo-
graphic mobility have meant that Americans tend to live outside the sort of
closed and powerful extended family groups that could perpetuate kinship-
based religions. Individual Americans, including individual women, are rela-
tively free to join new religions. In addition, the cultural pluralism of America
has resulted in a situation in •which individuals are aware that other people
believe in different religions than they themselves do. Unlike in isolated vil-
lages or tribes, people know that their own beliefs and rituals are not the only
ones that exist.
The well-documented "feminization of American religion" that took place
in the nineteenth century has meant that American women arc inclined to see
 The Examples 23
religion as their own sphere of action (Welter 1966). Nineteenth-century
women's religions grew out of a century of evangelical revivalism that spoke
of women's superior moral qualities. It bears emphasizing that in contempo-
rary American culture, religion in general is seen as women's sphere. Studies
show that American women are more religious than men both in terms of
church attendance and personal faith and commitment to orthodox beliefs
(Argyle and Beit-Hallahmi 1975, 71-79). While the feminization of religion is
clearly rooted in European religious experience (see Desan 1990), the institu-
tionalization of the process is more characteristic of American religion.
Both in the nineteenth and twentieth centuries the growth of feminism has
parallelled the development of women's religions in the United States. It is not
surprising that feminist political awareness is reflected in feminist religious
awareness.

Christian Science
After decades of struggling with illness and marital difficulties, Mary Baker
Eddy discovered the central tenet of Christian Science: The world as we see it
with our physical senses is illusory. I0 Therefore, sickness and suffering are
illusory. Once the individual understands that this is so, his or her sickness and
suffering will disappear. Women have been the majority of Christian Scientists
since its founding by Eddy during the second half of the nineteenth century.
Christian Science rejected the Calvinist theology of its day, which empha-
sized final judgment, endless punishment of sinners and nonbelievers, predesti-
nation, and the salvation of only a select few. Christian Science sees itself as
part of the Judeo-Christian tradition, and relies heavily on Biblical interpreta-
tion. The God of Christian Science is the God of the Bible, and Christian
revelation is accepted as true.
Christian Scientists believe that people are already perfect; there is no need
for a future salvation. Through Christian Science people are awakened to an
understanding of the perfection that is already theirs as children of God. The
function of Christian Science ritual is to demonstrate that perfection. People
under the illusion that matter is real cannot break with this illusion on their
own; they need a mediator to awaken them from their false belief. Jesus was
one such mediator, but Christ as the ideal of humankind was not confined to
Jesus. Jesus was unusual in that he fully embodied the Christ, but everyone is
capable of attaining what Jesus attained. Christian Science endeavors to be a
very rational religion; its belief system is "proven" through successful spiritual
healing.
In the early years female practitioners (healers) of Christian Science outnum-
bered males by about five to one (Gottschalk 1973). In the United States of
1926, 55.7% of members of all churches were women, yet 75% of Christian
Scientists were women (Stark and Bainbridge 1985, 237). In the United States
during the 19505 87.7% of Christian Science practitioners were women
(mostly married women, Wilson 1961, 198). Through the 19505 Christian
Science church attendance was made up of more than twice as many women as
24 Priestess, Mother, Sacred Sister

men (Wilson 1961, 199). A more recent study found the ratio of female to male
practitioners to be 8:1 (Fox 1989, 100).
Recruitment to Christian Science is through conversion; even those raised in
the faith must make a deliberate decision to accept or reject the religion. There
is a formal membership process in Christian Science, and members cannot
belong to any other denominations. Christian Science realizes that healing is its
best advertisement and its best way of recruiting members. Since the early
twentieth century most members have been middle class. Stephen Gottschalk
has explicated the affinity between Christian Science and American middle-
class beliefs that a person can control his or her own destiny, that it is wrong to
submit to undesirable conditions, and that achievement and upward mobility
are highly desirable.
A recurrent pattern throughout Christian history has been the labeling of
new movements as heretical. Accusations of heresy often coincide with the
condemnation of active participation and leadership of women in these move-
ments (e.g., Montanism). Christian Science, especially in the early days, was
vehemently opposed by Protestant clergy; it was accused by the medical profes-
sion of harming people's health; and Eddy herself was often the target of
personal abuse.
As I will later show, Christian Science differs from almost all other women's
religions in several ways: Its belief system is complex and unconditionally
obligatory, it is based on a large written literature, and the church organization
is highly centralized.

Shakers
In 1742 a group known as the Shakers emerged from the multilayered spiritual
environment of eighteenth-century England. The early Shakers attracted the
local proletariat, particularly from the textile mills of Lancaster. In 1758 Ann
Lee, born into a working-class family in Manchester, joined the Shakers, and
by 1770 assumed a leadership role. In 1774, having received a vision promising
that the millennial church would be established in New York, she took a small
group of followers with her and left England for the New World. Believing
Ann Lee to be the Savior, in 1776 the Shakers founded a community near
Albany. Both in England and in the United States the Shakers were persecuted
for their religious beliefs.
Shakers believed the End of Days to be imminent. They saw proof for this in
the socioeconomic upheavals of the late eighteenth century, and in the revela-
tion of Ann Lee as the second coming of Christ. They emphasized religious
experience instead of doctrine or creed; Shakers were famous for their emotion-
ally stirring and physically active forms of worship (dancing and shaking). In
the early years, Shakers made use of the emotional shock brought about by the
unveiling of personal secrets. Sermons dealt at length with the regions of
darkness in which lost souls suffered in agony.
Shaker communities grew during the late eighteenth and nineteenth centu-
ries, yet Shakerism never became a mass movement. The rural communities
 The Examples 25
served as the ideological, spiritual, and organizational centers of the Shaker
religion, although not all Shaker believers renounced their former lives and
went to live in Shaker communities. Within these communities, Shaker "breth-
ren" and "sisters" were organized into "families" or administrative units of
between thirty and one hundred members, each of which was governed by
two male elders and two female eldresses.
Based on records kept by the American Shakers, we have some picture of
who joined their agricultural communities. "Though they embraced a radical
gospel of millennarian perfectionism, the Shakers were not a dispossessed or
socially deviant constituency. . . . sectarian beliefs could appeal to a representa-
tive cross section of rural New England society" (Marini 1982, 96). From the
perspective of gender, however, it is clear that the Shaker way was not equally
appealing to everyone. Whereas almost equal numbers of men and women
joined the Shakers, far more women than men remained in the communities
for long periods of time (Foster 1981; Stein 1992).11 Thus, the reality was that
Shaker women outnumbered Shaker men by approximately 2:1 (Desroche
1971, 131), and in some communities there were less than half a dozen brethren
between the ages of eighteen and forty-five (Brewer 1992, 630). Shaker com-
munities underwent demographic changes over time, and the unbalanced gen-
der ratio became much more pronounced as the years went on. It is likely that
while Shaker theology was equally attractive to men and women, the neat,
safe, and ordered life-style was particularly compelling for women. In addi-
tion, it may be that the Shakers consciously sought greater female than male
membership. William Sims Bainbridge raises the provocative possibility that
as the Shakers came to realize that many boys and men would defect, they
began to accept fewer and fewer males (1982, 360).
Shaker ideology explicitly addressed issues of gender inequality. They devel-
oped a dual form of leadership, with a man and a woman at every level of the
hierarchy. In addition, they believed that celibacy would contribute to the
breakdown of women's subordination.
Ann Lee was the leader of the Shakers until her death in 1783. After her
death, leadership was taken over by men, James Whittaker from 1784 until
1787, and Joseph Meacham from 1787 until 1796. This pattern—the founding
of a new religion by a charismatic woman and then institutionalized leadership
by men—is a common one in the history of religions. In the case of the
Shakers, however, female leadership was resumed in 1796 by Lucy Wright,
who retained leadership until her death twenty-five years later. Following
Wright, there was no one single leader, either male or female, but rather groups
of men and women who led the Society. At certain times, male leadership was
in ascendancy. From 1876 onwards, women played increasingly prominent
roles in the Society's leadership. During the twentieth century, "Shaker women
began to dominate nearly every aspect of the society's life" (Stein 1992, 256).
The mid- to late nineteenth-century religious climate in the United States
did not encourage large groups of people to seek radical religious alternatives,
and Shaker membership accordingly declined from the mid-nineteenth century
on. Because of the celibate Shaker life-style, Shaker membership was limited
26 Priestess, Mother, Sacred Sister

to adults who had actively chosen to join the Shakers, and to children who
were brought by their parents to Shaker communities. Many of these children
left the Shakers after adulthood (Stein 1992). Shakerism thrived as long as its
religious vision and economic organization appealed to segments of the Ameri-
can and English population. Peak membership was reached in the mid-
nineteenth century at nearly 4000 members (Bainbridge 1982).
It is tempting to compare Shakerism to Christian Science—both are Chris-
tian-derived religions founded by English-speaking women. A closer look,
however, shows some interesting differences. Shakers, unlike Christian Scien-
tists, advocated celibacy and pacifism. Shakers lived in isolated communities
where all property was communally owned; Christian Science encouraged its
members to mix with the world. Shaker communities were in rural areas,
while most Christian Scientists have been urban dwellers. The Shakers at-
tracted mostly (although not exclusively) poor and working-class members,
while Christian Science membership has tended to be middle class. Whereas
Christian Science provides means of overcoming problems met with in this
world, Shakerism prevents its followers from encountering such problems by
living in Utopian communities. Shakers emphasize the millennium while Chris-
tian Science emphasizes healing (see Klein 1979). And finally, the Shakers have
all but disappeared while Christian Science still thrives.

Feminist Spirituality and Womanist Theology

Writing about the contemporary Feminist Spirituality Movement(s) of North
America and Western Europe is an extraordinarily complex task. As a religious
system it is quite new (it began in the 19705) and still evolving. There are no
official sacred texts, no absolute leaders, no required affirmations of faith, no
membership dues, and no undisputed agenda of beliefs and rituals. Feminist
Spirituality encourages and accepts as valid and legitimate the inspirations,
dreams, visions, experiences, and interpretations of individual women. In con-
trast to many other women's religions that lack self-generated literature (as
opposed to studies conducted by outsiders), Feminist Spirituality is an excep-
tionally prolific religious movement; its corpus of literature includes novels,
diaries, descriptions of rituals, sacred histories, and philosophical treatises. Yet
none of these writings is considered canonical. The theological and ritual focus
of Feminist Spirituality is the celebration of womanhood.
Because women active in the Movement are aware that they are creating a
new religion, one of their most important challenges is the search for authentic-
ity. On the one hand, Feminist Spirituality roots its claims for authenticity in
the academy: archeology (statues of ancient goddesses), anthropology (studies
of primitive cultures in which women are less oppressed than in modern
society), history (reports of persecution of women throughout the ages), liter-
ary analysis (of books by women authors), and psychology (Kristeva, Jung,
Melanie Klein, and others). The very eclectic literature of the Feminist Spiritu-
ality Movement has delved into symbols from African religions, African-
American traditions, and the Ancient Near East. On the other hand, Spiritual
 The Examples 27
Feminists focus their search for authenticity on women's dreams, fantasies,
intuitions, and direct revelations and encounters with the sacred.
A second challenge faced by the Feminist Spirituality Movement has been to
negotiate the relationship between religion and politics. The Movement is
clearly a feminist (as opposed to women's) movement, and in many groups po-
litical action, especially around issues of ecology and women's rights, is per-
ceived as having spiritual elements. Simultaneously, celebrating and strengthen-
ing womanhood through rituals is seen as preparing women for political action.
Just as it is difficult to point to a static set of beliefs and rituals and call them
"Feminist Spirituality," it is difficult to point to a particular group of women and
call them "Spiritual Feminists." A large core of Spiritual Feminists identify them-
selves as witches and belong to organized covens. Other Spiritual Feminists
belong to ritual or study groups that do not identify with Wicca (witch) religion.
To further complicate matters, many women who feel a sense of identity with the
Spiritual Feminist Movement have retained affiliation with the Jewish or Chris-
tian group in which they were raised. Such women would be likely to participate
in both Feminist rituals and Jewish or Christian rituals, and to work at effecting
feminist change within mainstream Jewish or Christian denominations. Al-
though no one really knows how many Spiritual Feminists there are, Solovitch
(1990) has estimated that they number about 100,000 in the United States today. I2
The Feminist Spirituality Movement offers women rituals celebrating the
female life cycle: menarche, first orgasm, birth, and menopause. Other rituals
reflect the cycle of nature, and the winter and summer solstices and autumn
and spring equinoxes are celebrated by many Spiritual Feminist groups.
Starhawk's Truth or Dare (1987) includes rituals for building community and
rituals for self-knowledge and conquering fear, a ritual for healing from abuse,
a body praise ritual, and a ritual of preparation for political struggle.
Parallel to the Feminist Spirituality Movement—which has attracted mostly
white and middle-class women—the Womanist Movement has developed
among black women.13 Cheryl Sanders offers the following succinct definition
of Womanism: "The womanist is a black feminist who is audacious, willful and
serious; loves and prefers women, but also may love men; is committed to the
survival and wholeness of entire people, and is universalist, capable, all loving,
and deep" (1989, 86).
Womanist theology draws on secular feminism, Christianity, the writings of
Alice Walker (author of The Color Purple), and African-American folk culture.
Unlike white Feminist Spirituality, which is often separatist and anti-male,
Womanism affirms black women's "historic connection with men through
love and through a shared struggle for survival and for productive quality of
life" (Williams 1989, 182). Womanism gives black women the freedom to
explore their own history and culture, without being constrained by what
white feminists have already identified as women's issues. 14
Both Feminist Spirituality and Womanism differ from contemporary main-
stream Christianity and from American civil religion in several ways. These
women's religions embrace non-materialistic value systems, de-centralized or-
ganizations, on-going revelation, and female images of divinity.
28 Priestess, Mother, Sacred Sister

Spiritualism
Spiritualism, like Feminist Spirituality and Korean household religion, is diffi-
cult to pin down. To begin with, Spiritualists are infamous for being unable or
unwilling to form a permanent organization to which all Spiritualist groups
belong. In addition, a number of somewhat different religious streams in
North and South America and in Europe call themselves "Spiritualists." In this
chapter I focus primarily on the popular and vibrant Spiritualist movement of
nineteenth-century North America. In addition, I draw upon studies of contem-
porary North American and British Spiritualist groups, and of contemporary
Mexican Spiritualism. The core belief of all these groups is that spirits of the
dead can communicate with the living. The central rituals are seances at which
this communication is carried out.
The founding of Spiritualism as a popular movement is usually said to have
occurred in March 1848 when Kate and Margaret Fox, two young sisters who
lived on a New York farm, heard loud rappings emanating from within their
house. They attributed the rappings to the spirit of a murdered traveling
salesman. Although the founding of Spiritualism may be traced to a discrete
event, Spiritualist phenomena spread spontaneously throughout large parts of
the United States and England. In the half century after the Fox sisters heard
their rappings, hundreds of thousands of individuals, mostly women, gathered
in private homes and in public lecture halls, and witnessed the ability of some
human beings to communicate with the spirit world. Since most people who
attended Spiritualist rituals did not belong to any kind of organization, it is not
possible to know how many Spiritualists there were during the mid-nineteenth
century. Estimates ranged from I to 11 million (Moore 1977, 14). When Spiritu-
alism spread to England, it became even more popular than in the United
States. There was intense opposition to Spiritualism from Christian ministers,
doctors, and scientists. Physicians looked as Spiritualism as an illness. Oppo-
nents of Spiritualism were preoccupied both with supposed fraudulence in
Spiritualist performances, and with the evil inherent in summoning the dead.
In the beginning Spiritualism was not a full-fledged religion—it presented
no new theology. In fact, it was seen primarily as a new scientific innovation—
similar to the telegraph. But in response to opposition from Christian denomi-
nations, it began to develop a coherent theology. Spiritualist beliefs boil down
to two key notions: The human personality survives the death of the body, and
it is possible to communicate with the spirits of the dead.
Geoffrey Nelson (1969) has itemized all the forms of spirit manifestations
practiced by Spiritualists in 1860: rappings, spirit writing and drawing, trance
and trance speaking, clairvoyance and clairaudience, luminous phenomena,
spiritual impersonation (behaving with mannerisms of the departed), spirit
music, visible and tactual manifestations, spirit intercourse by means of mirror
or crystal or water, apparitions of the departed, visions and previsions, dreams,
presentiments, spirit influx by which ideas are infused into the mind, speaking
in tongues, and possession. Spiritualists sec these manifestations as empirical
proof of Spiritualist tenets.
 The Examples 29
Unlike most other religious denominations of the nineteenth century, Spiritu-
alism afforded women equal authority and opportunities. Studies conducted at
the time show that the majority of Spiritualist mediums were women, and the
popular press portrayed male mediums as effeminate and confused (Moore
1977, 105). June Macklin (1977) found that in 1975 approximately 70% of
Spiritualist mediums in the United States were women.
Ann Braude (1989) analyzes women's involvement with Spiritualism in light
of the other religious options of the day. Although piety and morality were
viewed as female traits, women were forbidden from preaching and sometimes
even speaking in mainstream churches. Women could see that most people
who attended church were female, but the church leaders were male. In con-
trast, "If, during a seance, a woman became the principal actor, the instigator
and director of the proceedings, it was because she was thought to possess not
only genuine spiritual power but also the right to exercise it. Spiritualism
validated the female authoritative voice and permitted women an active profes-
sional and spiritual role largely denied them elsewhere" (Owen 1981, 6).
During the nineteenth century Spiritualism was popular among the educated
and among religious skeptics, as well as among former political radicals who
had become disenchanted with the possibility of creating Utopia in this world
and began to look for Utopia in the next. Mid-nineteenth-century Spiritualists
tended to support abolition, temperance, women's rights, and social reform.
By the end of the nineteenth century, Spiritualism had ceased functioning as a
reform movement, those who cared for social reform left the movement, and
Spiritualist societies became more and more conservative.
Spiritualist seances typically consist of women mediums helping women cli-
ents communicate with departed family members. Contact with the spirits is
sought in order to request advice concerning earthly problems, and this aspect of
Spiritualism has remained popular today in Great Britain and the United States.
Spiritualism does not demand that its members forgo membership in other
religions, and Spiritualist groups often choose not to meet on Sunday in order to
allow their members to attend church with more conventional Christian denomi-
nations. In modern Spiritualist groups most members have been drawn in after
having experienced some event that the person considers to be psychic.
In contemporary Mexican Spiritualism, women predominate as clients, heal-
ers, and leaders. Although their beliefs are similar to those of American and
British Spiritualism, the ritual focus is more on healing than on conversational
communication with the spirit world (Finkler 1985b. According to the head of
one Spiritualist Temple in Mexico City, Spiritualism is "a total religion, we
have our symbols, our laws, and our liturgy" (Finkler 1986, 629).

AFRICAN AND AFRICAN-AMERICAN

RELIGIONS
The social and historical contexts of the African and African-American exam-
ples are different from those of either the East and Southeast Asian or the
3O Priestess, Mother, Sacred Sister

North American examples. While it would be absurdly reductionist to make

any sort of global claim about all of Africa, it does seem that a recurrent theme
in the ethnographic literature on Africa is gender complementarity—the no-
tion that men and women contribute to the preservation of society in different
but equally important ways. Religiously, this often means that men and
women have their own somewhat independent rituals, cults, or secret soci-
eties. I will present the Sande secret society of Sierra Leone as an example of
this cultural pattern. Another characteristic of traditional African religion is its
embeddedness in kinship organization and family relations. As we will see, this
has rather different implications for men and women.
Knowing that the indigenous religions of Africa often offered opportunities
for religious expression and leadership both to men and to women, it becomes
interesting to follow the effects of various kinds of changes on this more-or-
less egalitarian situation. Throughout much of Africa, Islam has now become
the dominant religion. Parallel to growth of Islam with its almost total exclu-
sion of women from Islamic rituals, in some parts of Islamicized Africa women
have formed new, non-Islamic (albeit "marginal" in the eyes of male authori-
ties) religious groups. Religious practices involving zar possession are the best-
documented, yet far from sole, example of this pattern. A question that cannot
be fully answered is to what extent women's possession cults (like the zar) are
new responses to Islam, and to what extent they are leftover from pre-Islamic
African religion (as in the model for Southeast Asia presented above). Writing
about spirit possession among the Diga of south Kenya, Roger Gomm demon-
strates that the traditional propitiation of nature spirits once involved both men
and women, but is now almost solely a female concern (1975,134). Similarly,
there is good evidence that what is currently the almost all-female bori cult of
the Hausa once was the mainstream religion that involved both men and
women (Lewis et al. 1991). Although anthropologists and historians talk about
the zar cult as having originated in Ethiopia during the past 200 years, spirit
possession has a long history in Africa. What we may say with some assurance
is that while the zar cult in its present institutionalized and predominantly
female form is a new phenomenon, zar-type beliefs and practices are neither
foreign nor innovative.
An equally interesting question concerns what happened to African religion in
the wake of forcible transfer of slaves to America. While all slaves suffered op-
pression, violence, and culture clash, men and women experienced slavery in dif-
ferent ways. For example, women more often than men were victims of sexual
assault. In many parts of America slavery led to the breakup of families, and men
more than women were cut off from any sort of family life, while women and
children were often able to maintain some semblance of family relationships.
Given the embeddedness of traditional African religion in the family, slavery
often meant that women more than men continued to observe religious rituals.
In the post-slavery period African-American men have often found it neces-
sary to leave their families in order to obtain employment either in distant cities
or as migrant laborers. Again, the result has been that women are more in-
volved in both family and religion than are men.
 The Examples 31
Variations of this process took place in Brazil and among the Black Caribs of
Belize. It is significant that in both these societies women are also more in-
volved in Roman Catholic rituals than men. Like white North Americans,
African-Americans perceive religion (whether female dominated or not) as a
female enterprise. In contrast to East and Southeast Asia where new religions
attracted men, thus leaving the indigenous religion in women's hands, eco-
nomic necessity forced African-American men to withdraw from both the
familial and religious realms, thus leaving religion (albeit syncretic rather than
purely indigenous ones) in women's hands.

Sande
The Sande women's secret society exists under various names throughout West
Africa, particularly in Liberia, Sierra Leone, the Ivory Coast, and Guinea.
Among the Mende, Sherbro and Temne tribes of Sierra Leone nearly 95% of
all women undergo initiation into Sande (Margai 1948). Among the Mende
(and most likely among the other tribes in which Sande is present) the whole
realm of the sacred is controlled by the secret societies. "Most observers of
Mende society have been struck by the enormous weight secret societies carry
in Mende life. . . . Like the medieval church, Poro [the men's secret society]
and Sande provide sanctions for nearly every sphere of secular life. They
embody and control supernatural power, lay down rules of conduct, and pro-
vide the major source of propitiation for transgressions of sacred and of secular
law" (Cosentino 1982, 22).
During initiation rituals, Sande women gather for weeks or months in "bush
schools," away from contact with nonmembers and men. Sande initiation is
concerned with the cultural construction of fertility and reproduction. Adoles-
cent girls are taught about childbirth, trained in household tasks, and encour-
aged to cooperate with other women. Sande teaches women what they need to
know to function in their community: spinning, weaving, fishing, net making,
house and mother craft, first aid, and medicinal herbs.
A central element of initiation is ritual clitoridectomy (see Chapter 6). An-
other is the appearance of a masked figure who is said to embody Sande spirit.
"These masks are representations of spiritual and mythological symbols trans-
lated into wood and designed to express a spiritual message so complete that
future generations can do no more than learn from its mysteries" (Richards
1973, 76).
Sande chapters are not unified into any sort of central organization. Each
chapter owns secret knowledge that is passed on to initiates. Sande chapters
also own ritual objects and medicine; that is, "physical substances with effec-
tive pharmacological properties and physical substances which link persons
with sources of power in the universe" (MacCormack 1977, 95).
Girls are typically initiated into the Sande chapter in their mothers' villages,
but, because post-marriage residence is virilocal, transfer to the chapter in the
husband's village. Women try to return to their natal villages to give birth in
the chapter in which they were initiated. "Sande spreads as women migrate,
32 Priestess, Mother, Sacred Sister

following marriage, to live virilocally with their husbands' people." (Mac-

Cormack 1979, 28).
Women continue to participate in Sande activities throughout their lives.
Older women especially may devote a great deal of time to Sande concerns. A
men's secret society known as Poro exists in parallel to Sande, and both Poro
and Sande leaders may have a great deal of power outside the secret society.
The available ethnographic descriptions of Sande suggest that Sande women
do not see themselves as a formal congregation. On the other hand, Carol
MacCormack discovered that Sande women in Lungi, Moyamba District, are
buried in a great mound in the village. "The mound reminds inhabitants of the
power of women to bless and succor, their power existing in unbroken continu-
ity from the living to the ancestors" (1977, 95).
Sande has served as a power base for women both vis-a-vis their own
menfolks who can be punished by Sande for infringement of Sande rules, and
vis-a-vis outside authorities. For example, Sande women have organized to
protest unfair taxes on a number of occasions (see Chapter 13).

Afro-Brazilian Religions
The rich and eclectic twentieth century Afro-Brazilian cults are the most re-
nowned and best-documented examples of religions dominated by women.
Known as Candomble (in Bahia), Umbanda (in large metropolitan areas),
Batuque (in Belem), Xango (in Recife), and Macumba (in the southern parts of
Rio dejaneiro and Sao Paulo), the cults appeal primarily to women. Anthropolo-
gists consider these cults to be syncretic—synthesizing elements of African
tribal religions, Amerindian religions, Catholicism, and Kardecism (French
Spiritism). The various Afro-Brazilian religions differ among themselves, and I
will point out some of the more interesting differences. They are, however,
sufficiently similar to be treated as a group.
The most celebrated feature of these cults are public rituals in which medi-
ums are possessed by supernatural beings from an eclectic retinue of African,
Catholic, and Amerindian deities, spirits, and heros. Possession is attained
through a variety of techniques, most notably dancing, and possessed medi-
ums behave in the manner characteristic of the possessing spirit. What attracts
most adherents to the religions is curing, and most people who attend Afro-
Brazilian religious rituals do so in order to be healed of a variety of illnesses or
misfortunes. There are many supporters or followers who visit cult centers,
but never become mediums. These people approach the gods and saints who
are incorporated in a medium, and ask for advice concerning particular prob-
lems. The petitioner is typically given elaborate instructions about herb baths
and candle lighting, and told to make contributions to the medium or cult
center.
Esther Pressel (1974; 1980) explains the widespread appeal of the Afro-
Brazilian religions in light of the strong historical base of spirit possession
religions in both Africa and the New World, the failure of medical facilities to
keep pace with developments in other spheres of Brazilian society, and the
 The Examples 33
failure of the elitist Catholic Church to meet the needs of people in the rapidly
modernizing Brazil.
In fact, Afro-Brazilian religions do not seem to compete with the Church.
Despite Brazil's official Catholicism, true Catholic belief is rather rare; the real
religion of most people is Afro-Brazilian religion which includes Catholic
influenced practices (Landes 1940). For example, mediums honor the saints
who are adored by their chief encantados (spirits) and place pictures of these
saints on their home shrines (Leacock and Leacock 1972). Over 90% of Brazil-
ians are baptized in the Catholic Church, and over 50% of Brazilians are either
spirit mediums or have attended sessions of spirit groups to obtain spiritual
assistance. Babies are sometimes baptized in both Catholic and Afro-Brazilian
rituals, and marriages are performed by both religions (Pressel 1974). Not
surprisingly, the established Catholic Church is less concerned about the tradi-
tional lower-class Afro-Brazilian religions like Batuque and Candomble (as
they are seen as part of the superstition of the ignorant poor) and more both-
ered by the religions that attract the educated middle class (like Umbanda).
John Burdick (1990) has compared the appeal of Umbanda, Pentecostalism,
and Roman Catholicism for urban Brazilian women. (Most of his comparisons
are true for other Afro-Brazilian religions as well.) Whereas Umbanda and
Pentecostalism attract women seeking help with domestic problems, women
claim that Catholic priests cannot understand their problems because celibate
men do not know anything about domestic life. In addition, Catholic church
groups are based in the local neighborhoods, so if a woman shares her problem
with the Church group all her neighbors will know and gossip about her. The
Church tends to blame people rather than supernatural forces for domestic
conflict, a viewpoint that easily leads to acrimonious relationships and guilt
feelings. The association of poor Brazilian women with the Church is further
weakened by contemporary liberation theology stressing societal rather than
individual problems, thus ignoring the day-to-day needs of most Brazilian
women. Finally, Marian devotion encourages resignation to suffering, an exis-
tential stance many Brazilian women find unappealing.
Pentecostalism and Umbanda, on the other hand, blame supernatural en-
tities for domestic conflict, and provide clear means for coping (expelling the
devil in Pentecostalism, spirit possession in Umbanda). Pentecostal and Um-
banda rituals are gathering places for people with problems—everyone has
problems—so people are not afraid that they will be gossiped about.
But whereas Pentecostal beliefs are absolute and dichotomized (God vs. the
devil), Umbanda is ambiguous and multifaceted. In Umbanda humans can
influence and bargain with the spirits, and the rituals and beliefs are flexible
enough so that one can always find a culprit for one's problems. Whereas
Pentecostalism preaches human weakness and total submission to God, Um-
banda gives a woman magical tools to get revenge against those who have used
the spirits to harm her.
A woman will sometimes ask for help in more than one religion. According
to Burdick she is not simply "hedging her bets" or "shopping around." Rather,
"In each place she can articulate her predicament in a slightly different way,
34 Priestess, Mother, Sacred Sister

emphasizing different aspects of the problem" (1990, 167). Catholicism nur-

tures perseverance, Pentecostalism "letting go," and Umbanda self-help.
The Afro-Brazilian religions documented in the ethnographic literature are
composed primarily of women, and the extent to which men are involved
varies from group to group. In general, in the more conservative (African)
groups men have less of a role, in the newer groups (such as Umbanda) men
are more involved. The first cult centers in Bahia were founded by women
priestesses. And Lerch (1980) has documented that in Porto Alegre in 1974-75,
80-85% of mediums in Umbanda centers were women. 15
A number of factors discourage men from becoming mediums. The strong
Brazilian masculinity complex is not conducive to obedience to spirits and
temple leaders. Women are believed to be "softer" or easier for possessing
spirits to penetrate. In the more traditional Afro-Brazilian cults men are not
supposed to become possessed at all, and the ritual dancing and singing are
seen as feminine activities. There is a popular belief that male mediums are
homosexuals (Pressel, 1974, claims that in fact this is probably not true, but
people believe it anyway.) In addition, more men than women work at jobs
with regular hours and so do not have the time to spend at sessions. When men
do find spare time, they have other places (such as bars) to meet friends,
whereas women are much more confined to their homes. Lastly, in Brazil all
religion is seen as a feminine activity—even the Catholic Church has trouble
recruiting enough priests.

Black Carib (Garifuna) Religion

Old women are the religious leaders of the Black Caribs of Belize. l6 Black
Caribs, who constitute minority communities in Belize, Guatemala, Hondu-
ras, and Nicaragua, are descendants of African slaves and Carib Indians. Black
Carib religion is a composite of African, Amerindian, and Roman Catholic
elements. Middle-aged and old women dominate the religious life of the Black
Caribs of Belize.
The kinship structure of Black Carib communities consists of matrifocal
extended families, and households typically revolve around a woman, her
daughter(s), and their children. Black Carib children are rarely reared exclu-
sively by both biological parents. Fathers are frequently absent working as
migrant wage laborers, and many grandmothers are involved in childcare on a
daily basis. As women are "in charge of" kinship, they are also "in charge of"
religion. According to Virginia Kerns, whose 1983 study is the best source on
Black Carib kinship and religion, informants told her that anyone may take part
in ritual—it is just that older women are "more interested" (1983, 167). Black
Carib religious ritual is directly related to kinship. The greater part of Black
Carib religion consists of rituals aimed at honoring, caring for or appeasing
ancestors. The goodwill of ancestors is deemed as necessary to the well-being of
descendants. "The ancestral cult . . . when considered in terms of its practical
implications, and its role in preserving the traditions of the group, must be
regarded as the core of the Black Carib system of belief" (Coelho 1955, 135).
 The Examples 35

Older women feel that it is their responsibility to perform rituals on behalf

of their ancestors. Because a goal of ancestor ritual is to ensure that ancestors
will not harm descendants, older women—who stand at the center of the
generational chain—are pivotal ceremonial actors. Their ritual responsibility
and expertise guarantee that children and grandchildren will be healthy and
happy. Typical rituals involve singing, dancing, and feasting, and spirit posses-
sion by ancestors. According to Douglas Taylor's 1951 study, at mourning
rituals, "The women . . . at a conservative estimate, outnumber the men in the
proportion of four to one" (117).
Older Black Carib women are most knowledgeable about religious rituals,
and organize and attend many rituals each year. These women have a great deal
of autonomy in deciding how to spend their time and money, and they have
access to other women on whom they may call for assistance in furnishing
rituals. All these factors serve to strengthen the notion that women, and espe-
cially older women, are the most suitable ritual leaders.
Sorcery and magic are very much part of the Black Carib religious mind-set.
Almost every act of everyday living is accompanied by some magical proce-
dure, and "Every Carib uses some form of protection against sorcery" (Coelho
1955, 166). So-called "love magic" involves bodily secretions concealed in
food served by women to men.
Before going on, I wish to clarify that "love magic" is the term used by
outsiders and not necessarily by Black Carib women. As a feminist anthropolo-
gist, I reject this terminology for two important reasons. First, it implies that it
is ontologically distinct from the ancestor religion described above. Yet al-
though different ethnographers have written about the two ritual constella-
tions, it seems to me that they are intrinsically related: both are focused on
interpersonal relationships, utilize food in rituals, and are controlled by
women. Second, in line with old-fashioned Durkheimian notions of magic
versus religion, the words "love magic" imply private, trivial, and possibly
even anti-social acts. However, according to the one ethnographer who has
carefully recorded these practices, "[Love magic] is institutionalized and direct
evidence [for it] is easily found. Practitioners solicit potential clients, the para-
phernalia are occasionally seen, and much of the business of magic is simply
not hidden. Indeed, to be effective it must have some public aspects" (Bullard
1974, 263). Like ancestor rituals, "love magic," according to Bullard, has a
stabilizing influence on marriages and families. On the other hand, Virginia
Kerns notes that her informants never spoke of these two sets of practices as
being equivalent; in fact, her informants related to them as somewhat antitheti-
cal (personal communication 1992). In sum, my own inclination to merge
ancestral rites and "love magic" is also somewhat problematic.
Most Black Caribs are Roman Catholics and receive some Catholic educa-
tion. Knowledge of other spiritual matters is transmitted informally. All rituals
for the dead, including Catholic ones, are perceived by Black Caribs as a
totality—they are all intended to satisfy the dead and protect the living. The
middle-aged and elderly women who are experts at ancestor rituals are also
practicing Roman Catholics, and they do not seem to express conflict between
36 Priestess, Mother, Sacred Sister
Roman Catholic dogma and their own beliefs. "The typical Carib virtues of
flexibility and versatility enabled them to incorporate this non-Christian tradi-
tion into their Catholicism, thus achieving, despite the vigorous denunciations
of the official representatives of the Church, a synthesis that comprises a coher-
ent and unified body of doctrine" (Coelho 1955, 135).
Catholicism and Black Carib religion differ in terms of leadership (male vs.
female), literacy (ancestor worship, unlike Catholicism, does not involve sa-
cred texts), atmosphere at rituals (formal vs. informal), and organizational
structure (Roman Catholicism is a centralized, hierarchical world religion,
whereas Black Carib ancestor worship is a nonhierarchical, decentralized local
religion).

Zar
The zdr cult of North Africa and the Middle East is essentially women's
business. Capricious spirits known as zdr are prone to attack and possess
women, especially married women, who then turn to cult leaders in order to
be cured. Zdr attack takes the form of illness with symptoms ranging from
hysteria to physical disorders. Treatment consists of initiation into the zdr cult,
and the appeasement of the possessing spirit by the presentation of gifts to the
patient. Treatment means reaching accommodation with the spirit—taming
it—not exorcising it. A woman may go through a marriage ceremony with
her zdr, which means a permanent relationship of ritual responsibility. Once
inducted into the cult group, the woman participates in ceremonies that pro-
mote possession-trance experience among its members, all of whom had been
afflicted in the past. Members meet periodically to dance, feast, and incarnate
their spirits. While the relationship with the spirit begins with illness, over the
years the relationship should become one of complementarity and exchange.
In a Northern Sudanese village studied by Janice Boddy, more than 40% of
women ever married and over the age of 15 have been possessed by zdr spirits.
Most afflicted women are between the ages of 35 and 55, two-thirds of the
women in that age group have zdr spirits. Thus, from a gynocentric perspective
the zar cult is far from marginal. Very few men (perhaps 5% of all men) are zar
adepts. Lucie Saunders (1977) found in her study of an Egyptian village that most
women profess to believe in zar spirits, and most men profess not to. Zar posses-
sion is so fully identified with women that among the Egyptian Nubians if the
zar cult patient is a man he is adorned as a female bride (Kennedy 1978).17
Participants in the zdr cult tend to be individuals of secondary status within
their cultural settings. Most scholars agree that the zdr cult originated in central
Ethiopia in the eighteenth century, and then was spread by slaves in Africa and
the Middle East (Natvig 1987). Whether the cult is made up solely of women
(Sudan), or •whether it also includes poor and marginal men (Ethiopia), its
members are people who lack access to most leadership or prestige roles in the
society at large.
For the most part, the zdr cult exists in Islamic cultures. (The zdr also is
popular in Ethiopia where most people are Christian and not Muslim.) It is
 The Examples 37
crucial to bear in mind that all the formal roles of Islamic leadership are limited
to men, and women are excluded from most important public religious rituals.
On the other hand, there are aspects of religious life in which Muslim women
are very active indeed. Lois Beck has argued that in those ritual complexes not
formally required by Islamic law and tradition, "men do not participate with
the intensity, nor in the numbers that do women" (1980, 39). Examples of
arenas in which women are active include life-cycle celebrations, beliefs and
rituals connected with early Islamic figures (such as Hussain and Fatima) and
contemporary saints, pilgrimage to local tombs and shrines, curing and spirit
possession cults, charity, and amulets. According to Jane Smith, "[Women's]
unorthodox practices served both to further isolate women from the formal
rituals of the Islamic community and to give them an arena in which they could
feel comfortable and in control. Despite periodic efforts to 'clean up' such
heterodox practices, they have been and continue to be a powerful part of the
lives of many Muslims, especially women" (1987, 242).
Because women are neither assumed to be acquainted with Muslim liturgy
and doctrines, nor expected to have the moral strength to uphold them,
women tend to be relatively freer to embrace folk beliefs. In many Muslim
societies women have been and continue to be the primary agents in the
relationship of humans to the world of spirits; it is women who are experts at
warding off evil spirits. Ethnographers have documented the existence of cur-
ing and spirit possession cults throughout Muslim societies. Women are active
in many of these quasi-Muslim cults (such as the zar cult, thefiqi cult, and cults
of saints) which are characterized by ecstatic and nonformalized rituals.
Women's zar activities serve as a counterpart to men's involvement in offi-
cial Islamic religious practices. Yet while in certain senses complementary to
Islamic practice, the zar cult is not equal in status to Islam (or Coptic Christian-
ity in the case of Ethiopia). In the Northern Sudan (as elsewhere) the zar cult is
periodically attacked by the male religious and secular elite. During the last 100
years, the zar cult has been described as "un-Islamic innovations," "bad tradi-
tions," "superstition," and "backward customs" (Constantinides 1982, 186).
Janice Boddy's Northern Sudanese informants report that the zar cult became
established in the Sudan at about the same time as Islam thoroughly penetrated
the villages. (Although much of Africa had been formally Islamic for many
centuries, it is only during this century that more than a thin veneer of Islam
has penetrated the villages.) "The [zar] cult gained ground in virtual tandem
with local Islamization" (Boddy 1989, 35). Thus we seem to be seeing a para-
doxical situation in which zar spreads together with its main opponent. Both
zar and orthodox Islam in Africa are responses to twentieth-century social
changes.
While orthodox Islam condemns the zar cult for dealing with devils, opposi-
tion is moderated by the fact that jinn (spirits) are mentioned in the Koran.
Although these jinn can be dispelled by reciting the opening of the Koran,
because they occupy unexpected places individuals may still be attacked and
thus require a zar ritual. Zar lore abounds with stories of government and
religious leaders who attempted to repress the cult until their own womenfolk
38 Priestess, Mother, Sacred Sister
became seriously ill through zdr possession. Then they were forced to agree to
rituals being held in their very houses.18
Similarly, Simon Messing has found that the Coptic Abyssinian priests ex-
press passive resistance to the cult—they profess to condemn it but do little to
counteract it. "This may be because many priests secretly believe in the cosmol-
ogy of the zdr themselves, particularly in spirits that are regarded as Coptic
Christian" (1958, 1121).
The ethnographic literature indicates that there is a great deal of local varia-
tion regarding the status of the zdr cult. Lucie Saunders found that in an
Egyptian village zdr was seen as so compatible with Islam that at the opening
of public zdr ceremonies a male leader would read out loud verses from the
Koran. Cloudsley (1984), on the other hand, discovered that in Omdurman
(Sudan) zdr is thoroughly frowned on by devout Muslims, and considered a
pagan witchcraft ritual used solely by ignorant women. These two conflicting
accounts may possibly be reconciled by Boddy's finding that men see zdr and
Islam as conflicting but women regard them both as part of a "general religious
enterprise" (1989, 142). Men feel that spirits must be exorcised; women feel
they need to be accommodated.
One may be tempted to argue that women's religions develop in situations
in which women are excluded from the mainstream or male-dominated reli-
gion. Zdr certainly seems to bear that out: North African women are ex-
cluded from public ritual roles both in the mosque and in the Coptic Church.
We may recall, on the other hand, that in Brazil the Catholic Church (with
the exception of the priesthood) is perceived as a female domain, and rather
than being excluded from churches, women are typically the vast majority of
participants. I am not convinced that there is any correlation between the
occurrence of women's religions and women's exclusion or inclusion in other
religions of the day.
A related point concerns the use of the terms "marginal" and "peripheral."
The zdr, like many other women's religions, has been considered to be "periph-
eral" to the larger society (Lewis 1975). Yet, as a feminist scholar, I am obli-
gated to ask from whose point of view the zdr is peripheral. It should be clear
that in the village studied by Boddy, where most women participate in zdr
rituals, women would be unlikely to consider zdr rituals as marginal. In an
excellent essay dealing with possession cults in the Swahili coastal area of
Kenya and Tanzania, Linda Giles argues that "spirit possession should be seen
as an integrated part of coastal Islamic belief and practice" (1987, 245). More-
over, she demonstrates that cult performances draw a wide range of spectators
of both sexes and all ages; that even many of those who do not attend the
ceremonies still share much of the cult belief system; that the range of potential
cult members is quite extensive; that there are even more cases of people who
profess that they do not participate or even believe in spirit cult activities yet
who can be found visiting cult mediums in case of illness or the need for other
assistance; and that at least on the Swahili coast the link to societal institutions
was quite explicit in former times when the cults played a central role in
communal rituals (1987, 246-47).
 The Examples 39

The zdr cult is spread over a very large geographical area. The functions and
the form of the cult vary considerably from country to country, and even from
place to place within each country. Ethnographic reports of the zdr cult come
from urban and rural Northern Sudan, Ethiopia, Somaliland, and Egyptian
Nubia. While I have tried to be sensitive to the different manifestations of the
cult reported for the different societies, the main elements of the cult are similar
in all five societies.

RELIGIONS THAT ARE NOT DEALT

WITH IN THIS BOOK
Cross-culturally, relatively few religions are dominated by women. The twelve
key examples dealt with in this book are, to the best of my knowledge,
almost all the female-dominated religions that have been documented in the
ethnographic and historical literature. Thus while these examples are not
comprehensive, neither are they illustrative of a much larger field. An ongo-
ing difficulty that I faced in selecting and researching the examples around
which the book is based concerns the quality and quantity of available infor-
mation. The contemporary examples are better documented than the histori-
cal ones, and several examples are dependent on data gathered by one or two
field-workers.
Since the twelve key examples are not the only women's religions that have
existed, it is appropriate to explain why I chose these particular religions. First,
I preferred not to focus on religious traditions for which evidence is sketchy. I
have chosen not to deal with those ancient Near Eastern, Greek, and Roman
traditions that seem to have been dominated by women, but for which the
only testimony we have is myths and brief inscriptions.19
Another category I avoided includes Gnostic groups and medieval Chris-
tian heresies for which the historical documentation is based primarily on
polemics against the religion by the mainstream Christian Church. Since I
was able to find twelve religions that had been documented by more impar-
tial observers, I chose not to focus on religions for which the primary sources
are groups which opposed the religion, and so can be assumed to present a
biased account. The fact that in most cases the written sources were recorded
by advocates of male-dominated religions makes these sources even more
suspect. For example, the important role of women among the thirteenth-
century Guglielmites led the Church hierarchy to condemn the supposed
sexual orgies held during their rituals, although it is far from likely that these
orgies ever took place (Wessley 1978).
A third category that I, with some reservations, excluded from my list of
key examples are African religions similar but not identical to the African
religions I have included (see Beattic and Middleton 1969; Lewis 1986). In
recent writings scholars have begun to treat the zar cult as part of a complex
that also includes the West African (Nigerian) bori cult (see especially Lewis
1991).20 For the sake of simplicity, in this book I have chosen to look only at
zdr, although much of what I have written about zdr is also true for bori. The
4O Priestess, Mother, Sacred Sister
important role of women in African religions has been well documented (if not
well publicized) and I found it difficult to select the "best" (most representa-
tive, clearly female dominated, and well-documented) examples. In order to
avoid working with an unwieldy number of examples, I decided to focus on
the Sande secret society and the zar possession cult.
I have chosen to exclude religions that were founded by women and/or in
whose early days allowed women leadership roles, but which in the course of
institutionalization began to exclude women from positions of power. Exam-
ples of this are Pentecostalism (Barfoot and Sheppard 1980) and many of the
new religions of Japan (Nakamura 1980) and Africa (Jules-Rosette 1979). I have
also chosen not to focus on Theosophy, both because in many ways it is similar
to nineteenth-century Spiritualism, and because it was founded by both a man
and a woman (Helena P. Blavatsky and Colonel Henry Olcott).
I chose not to work with what could more properly be described as clusters
of rituals within a male-oriented religion (e.g., pilgrimage in Morocco or
childbirth rituals in India) or with streams headed by a woman but which
remained within the rubric of male domination (e.g., Beguines in medieval
Europe and various devotional cults in India). Similarly, I have excluded folk
variations on male-dominated great traditions (e.g., Mexican Spiritism, which
is a form of folk Catholicism). In that my working definition of women's
religions entails some sort of recognition that this religious group is indepen-
dent from a larger, male-dominated institutional context, I have also avoided
looking at groups of women who organize under the rubric of male domina-
tion (e.g., various Catholic celibate orders).
To all these categories of religions I will refer in passing (see Appendix B).
The category I am the most uncomfortable about excluding (and which I
hope to deal with in a subsequent volume) includes various tribal religions in
which both men and women have active roles, and in which each sex predomi-
nates in certain types of rituals. The three main examples I have in mind here
are Australian aboriginal religion as described by Diane Bell (1983), indigenous
Philippine religion as described by Teresita Infante (1975), and traditional Iro-
quois religion as described by Annemarie Shimony (1980). Since the goal of
the present volume is to try to tease out how women "do" religion when they
are in a position of dominance, tribal religions in which men's and women's
religious lives are intermeshed proved difficult to work with. Again, I empha-
size that my entire typology is artificial. Taking a religion like that of the
Ryukyu Islands on the one hand and one like Islam on the other, we can
construct a continuum going from female to male dominance; we cannot
construct absolute categories. Since in all known cultures both men and
women are "religious"—my twelve key examples are not pure types.

Notes
i. Anthropologists, sociologists, and historians of religion have defined religion in
a multitude of ways. One of the most successful definitions is proposed by Clifford
Geertz (1969): "Religion is a system of symbols which acts to establish powerful,
 The Examples 41

pervasive, and long-lasting moods and motivations in people by formulating concep-

tions of a general order of existence and clothing these conceptions with such an aura of
factuality that the moods and motivations seem uniquely realistic." All the women's
religious situations explored in this book fall well within Geertz's rubric.
2. Laurel Kendall (personal communication) confirms my hunch that there is some
sort of connection among these religious traditions. The same, or a similar, trend
toward female dominance in the religious domain is also probably a characteristic of
certain tribal cultures of the Philippines. I hope to look into this issue further in a future
study.
3. Indian colonization in Burma started in the first century c. E. and in Thailand in
the second century c.E. Buddhism as a dominant religion was introduced into Burma in
the eleventh century and Thailand in the thirteenth. Buddhism was the state religion but
not imposed by force. Missionary monks were responsible for spreading Buddhism.
4. Religious tolerance is characteristic of the Ryukyu Islands, and Confucianism,
Buddhism, and Christianity can be found on the Islands. Upper-class Ryukyuans
learned and practiced Chinese ancestral rituals. Buddhist funeral rites from Japan (but
no other elements of Buddhism) became fairly popular on the Ryukyus, and monks
were looked at with some disdain. Confucian rituals and status symbols became impor-
tant in a superficial way and only among upper class families. Even among Catholic
converts, no real break was made with former religious consciousness (Anzai 1976, 62).
5. Ouwehand tells me that the word should be written kan, that kami is Japanese
(personal communication 1992). However, since all the English-language sources use
the word kami I will, with some misgivings, continue to do so.
6. Haring (1964) believes that the Amami Island, have had longer and deeper cul-
tural contact with Japan and China (through Japan) than Okinawa has had. Mabuchi
(19763) states that belief in the spiritual predominance of the sister survives more in the
southern part of the southern Ryukyus than in the main island Okinawa, and is even
rarer in Miyako Island (northern part of the southern Ryukyus).
7. Spiro, on the other hand, says that it is normally the father's nat who is inherited
(1967, 101).
8. According to Tambiah (1970), it is not only in Thai Buddhism but also in spirit
cults (other than the matrilineal ones treated here) that the ritual specialists are male.
Kirsch (1985, 312), on the other hand, found that while women do support Buddhism
"it is also true that most 'animist' spirit practitioners, displaying a kind of symbolic
'anti-Buddhism,' are also women."
9. In Burma, on the other hand, people do report feeling conflict between Bud-
dhism and nat beliefs.
10. Because Mary Baker Eddy exerted an enormous influence over the development
of Christian Science, and because movements founded by women (and women who
found movements) tend to strike observers as sufficiently unusual to need to be "ex-
plained," many studies of Christian Science have traced Eddy's life story and shown
how the events and psychological traumas of her life caused her to discover Christian
Science. This fits in with an all-too-common tendency in the scholarly literature to treat
women's religion as somehow psychologically abnormal. Perhaps in reaction, I pay
attention primarily to the key theological and existential concepts of Christian Science,
rather than Eddy's life story.
11. In the early years, men actually outnumbered women in several Shaker communi-
ties (Brewer 1986, 215).
12. Starhawk believes that the number is even larger (personal communication
1992).
42 Priestess, Mother, Sacred Sister

13. Many Womanist writers do not wish to connect themselves to the predominantly
white Feminist Spirituality Movement. On the other hand, most Spiritual Feminists
express a strong sense of kinship with Womanism, and generally include articles by
Womanist writers in their anthologies.
14. In the past several years Hispanic feminists have begun to produce a body of
literature reflecting their own, special religious concerns. Since this budding movement
has (at least until now) identified itself with the Christian tradition and the Catholic
Church, I will not be dealing with it in the present volume. Among the many interest-
ing ideas given voice in this growing literature are an acknowledgment of women's
numerical dominance in the Catholic Church in Latin America, respect for non-
Catholic religions (such as Candomble), recognition of the feminization of poverty and
the connections between capitalism and patriarchy, use of traditional female symbols
such as the Virgin Mary, and an explicit addressing of political issues that affect
women's and men's lives (see Tamez 1989 for a sample of this literature).
15. Herskovits (1943) found more equal numbers of men and women in Porto
Alegre (southern Brazil) than studies of northern Brazil have shown.
16. Black Caribs are also known as Garifuna. In this book I look primarily at the
Black Caribs of Belize because that is the group whose religious and kinship structure
has been most thoroughly studied. I have no reason to think that the culture of the
Black Caribs of Guatemala, Honduras, and Nicaragua is very different.
17. This is not the case further south, in the Sudan.
18. According to Boddy, in the Sudan women consider zar to be part of Islam and
not a religion in its own right. Men are more ambivalent: The more educated deny that
it is an Islamic practice, and the less educated are not sure (personal communication
1992).
19. Ross Kraemer's newest book (1992) is a brilliant reconstruction of women's
religious lives in ancient Greece and Rome. However, much of what she describes
seems to be specific festivals rather than full-fledged religions. Even so, much of what
she found resonates with what I will be describing in this book.
20. Scholars of African religion have described many local spirit possession phenom-
ena as being similar to or even part of the zar cult. It seems to me of the utmost
importance to differentiate between spirit possession in which women are possessed by
spirits who are exorcised by a male ritual specialist, and spirit possession in which
possessed women join an ongoing group of fellow-sufferers who gather to share experi-
ences of possession trance. Among the Digo of South Kenya, for example, spirit
possession is officially diagnosed and exorcised by a male practitioner, mganga (Gomm
1975). I would treat this a totally different phenomenon from the zar cult of the Sudan.
There is a tendency among scholars to treat all African spirit possession as if it were one
phenomenon, whereas it is far more reasonable to treat spirit possession as a common
form of religious activity, which in different African societies is used and interpreted in
a variety of ways. The fallacy of treating all African spirit possession as a unified
phenomenon could be compared to someone claiming that all European prayer is one
phenomenon—that the secretive Latin of a Catholic priest, the speaking in tongues of a
Pentecostal woman, and the formal Hebrew daily liturgy of a Jewish man are essentially
the same thing.
 2
The Social Contexts
of Women's Religions
GENDER DISJUNCTION, MATRIFOCALITY,
AND A CRITIQUE OF D E P R I V A T I O N THEORY

Just as there is no one historical scenario that explains why women's religions
occur, there is no single cultural configuration in which we find all the
women's religions, nor a single cultural context in which women's religions
are always present. The ethnographic literature suggests that three sets of
factors tend to be associated with women's religions. The first of these factors
is gender dissonance: situations in which culturally accepted notions of gender
are either highly contradictory and/or rapidly changing.1 The second factor is
matrifocality—a cultural emphasis on the maternal role, often coupled with
either matrilineality and/or matrilocality. And the third is a relatively high
degree of personal, social, or economic autonomy for women. 2

GENDER DISSONANCE
Nineteenth-century United States (Shakerism,
Spiritualism, and Christian Science)
Shakerism, Spiritualism, and Christian Science developed during a period of
industrialization and urbanization. As I have shown previously (Sered 1990),
modernization affects women's religious lives in complex ways. Although
science and new literate religions (such as normative Islam) tend to accuse
women's traditional religions of being superstitious magic, rapid cultural
change may also mean the introduction of new religious options that women
in particular find attractive.
During the nineteenth century (when Shakerism flourished and Spiritualism
and Christian Science began) North America was characterized by highly patri-
archal cultural norms. Victorian gender ideology and the "cult of true woman-
hood" proclaimed female passivity, frailty, and weakness (Welter 1966). In
addition, as I explain in Chapter 4, nineteenth century women were increas-
ingly expected to be able to guarantee their children's health and welfare, an
expectation that most women found impossible to fulfill.
44 Priestess, Mother, Sacred Sister
What I have just described sounds more like a situation of women's distress
than of gender disjunction. And indeed, Alex Owen (1981) argues that at the
end of the nineteenth century and the beginning of the twentieth as women's
status improved (e.g., the 1882 married women's property law gave married
women control over their own property), Spiritualism (and Shakers) declined.
This argument, however, merits closer examination. Owen shows that dur-
ing the mid-nineteenth century respectable employment for women began to
develop—and this is the period in which some famous women mediums
were able to make a living in Spiritualist circles. In other words, Spiritualism
"took off" when new economic options had begun to open up for women. By
the 18905, when Spiritualism started to decline, many thousands of women
had already begun to work for wages in business and commerce. In short,
during the mid- and late nineteenth century there was a cultural climate in
which women aspired to, and occasionally gained, social and legal equality
with men. "The second half of the century was a period of shifting attitudes,
holding out a promise of broader and brighter horizons, but it was also a time
of anxiety and uncertainty as the old order began grudgingly to give way to the
new ... it is no accident that spiritualism, a movement which privileged
women and took them seriously, attracted so many female believers during a
period of gender disjunction and disparity between aspiration and reality"
(Owen 1981, 4).
American social life in the mid-nineteenth century was fluid. Individuals
could change their social roles with remarkable ease. Migrations across the
continent, immigrants flooding into the country, and industrialization created
a spirit of pluralism and change. American religious freedom meant that reli-
gious organizations were voluntary; religious toleration led to the growth of
democracy within many of the churches; and the denominations had no effec-
tive power to discipline their members (Nelson 1969). In addition, the
nineteenth-century Second Great Awakening legitimated public and ecstatic
religious roles for women, even within male-dominated Protestant sects.
While men continued to dominate all leadership roles in mainstream Protes-
tant denominations, by the early nineteenth century American iconographic
representations of Christ showed feminine characteristics predominating.
Louis Kern (1981) has emphasized the dissonance in nineteenth-century Ameri-
can religious culture: On the one hand women were morally "better" than men
(i.e., less materialistic, more pious), but on the other hand they were not
allowed to exercise the power that could be expected to derive from moral
superiority.3 Kern demonstrates that nineteenth-century men had ambivalent
feelings about women. Women were simultaneously the source of salvation
from the materialistic world, and secretive, sexual beings. He concludes that
this period was characterized by intense sexual ambivalence and anxiety caused
by social, cultural, and economic changes.
I would argue that Shakerism, Spiritualism, and Christian Science "took
off" not at a time when oppression of women was at its peak, but during a
time in which sex roles and gender ideology were changing. Nineteenth-
century America (like late eighteenth-century England) was characterized by
 The Social Contexts of Women's Religions 45
an intellectual climate of radical ideas and spiritual turmoil. Many new reli-
gious groups emerged during this period; science and technology became a real
alternative (in the eyes of many people) or threat (in the eyes of many other
people) to religion; and population shifts meant sudden and intense cultural
contacts.
The emergence of Shakerism, Spiritualism, and Christian Science seems to
have less to do with distress or deprivation than with gender disjunction—
disjunction between ideology and practice, combined with a situation in which
cracks had opened up in the old ways—in which new social and intellectual
options began to be available.

West Africa (Sande Secret Society)

West African cultures posit official ideologies of male dominance. On the
other hand, West African women traditionally have had active and sometimes
dominant economic roles. Women have farmed, sold surplus food at local
markets, engaged in trade, and accumulated money and other resources. As
David Rosen has argued, "West African women maintain a separate sphere of
production and exchange which provides them with the ability to carry out a
wide variety of economic transactions independently of male household mem-
bers. . . . the situation in West Africa, unlike elsewhere in the world, is one
in which women as well as men mediate the relationship between the peasant
community and outside society" (1983, 37). While modernization has had
certain negative consequences for African women (Boserup 1970), it has "pro-
vided new opportunities for Kpelle [a West African group among whom
almost all women belong to Sande secret society] women to acquire indepen-
dence and prestige. Where cash cropping is possible or wage labor opportuni-
ties are available, women can remain unmarried because cash from marketing
or from wage-earning lovers allows them to hire farm labor, pay house taxes,
and buy household necessities themselves, a pattern noted as well by studies
in other parts of Africa" (Bledsoe 1980, 3).
Ideologically, West African legal systems give rights in women's reproduc-
tive and productive services to men (fathers, husbands, etc.). On the other
hand, in actuality, many women are independent. For example, among the
Nuer over half the women of marriageable age are not under any man's legal
control (Bledsoe 1980). In parts of West Africa women have a long history of
political activity that even includes serving as chiefs of towns. Among the
Mende and Sherbro tribes, women can be household and descent group heads,
positions of importance and power in managing the group's land and a basis
for community influence. In one Sherbro community, 23 of the 39 household
heads in 1970 were women (Hoffer 1972, 154).
An interesting West African twist on kinship organization is that while men
want as many wives as possible to enhance their political and economic status,
many head wives also want junior wives for companionship and help with the
chores. Sometimes the wife is the one who actually chooses the secondary
wives. The Kpelle believe it is important for co-wives to get along well,
46 Priestess, Mother, Sacred Sister

whereas husbands and wives are not expected to really love each other, just to
cooperate. First wives often enlist their friends or younger sisters to be junior
wives for them (Cosentino 1982).
On the one hand, Sande membership must be seen in the context of cultures
in which women control important economic resources. On the other hand,
West Africa is far from matriarchal. It can more accurately be described as a
situation of gender disjunction—ideology (patriarchal) and reality (fairly egali-
tarian) clash. We will return to West Africa and the Kpelle later on.

MATRIFOCALITY
In a fascinating study of women and religion in the Greco-Roman world, Ross
Kraemer has drawn on the work of anthropologist Mary Douglas to explain
why in certain societies positions of religious activism and leadership are open
to women and in others they are not. Douglas developed a system of classifica-
tion in which human social experience is described in terms of two factors:
group and grid. Group refers to the degree to which individuals feel themselves
to be part of a community, to the degree to which the individual is incorpo-
rated in the group in shared households, work, resources, and leisure-time
activity. Grid refers to the extent to which rules and regulations govern an
individual's activities. Kraemer concludes that in the Greco-Roman world a
strong group and low grid constellation is correlated with increased religious
authority and options for women. When women are well incorporated into
associations of some kind, and when rules and hierarchy are relaxed, women
become more religiously active (1992, 199). To a large extent, Kraemer's obser-
vations are also true outside the Greco-Roman world. As we saw above,
women's religions are often associated with societies that are undergoing rapid
cultural change; that is, low grid societies. As I will show now, women's
religions are also associated with matrilineal and matrilocal societies in which
group—for women—is very strong.
Before I begin it is helpful to clarify the terminology I will be using. A term
that is crucial to my argument is matrifocality. Traditionally, anthropologists
have believed that matrifocal societies are rare, certainly far rarer than our
sample of women's religions would suggest (see, Table i). Exactly how rare
we do not know because, as both Marvin Harris (1985) and Nancy Tanner
(1974) point out, researchers have treated them as aberrations. Anthropologists
have been inclined to assume that normative societies are patrifocal; for a
society to be described as matrifocal there had to be a very dramatic absence of
male presence. Feminist anthropologist Sylvia Yanagisako argues, however,
that matrifocal does not necessarily mean that the husband is absent from the
house; it is sufficient for the mother to be the focus of the household (1979:
178). Yanagisako expands the category of matrifocality to include far more
situations that have hitherto been recognized, and certainly includes practically
all the cultures in which female-dominated religions occur. 4
Nancy Tanner (1974, 131) defines matrifocality as a kinship system in which
the structurally, culturally, and affectively central role of the mother is seen as
 The Social Contexts of Women's Religions 47
Table I. Female-Dominated Religions and Matrilineality, Matrilocality,
or Matrifocality—Summary
Country Summary
Burma: Clearly matrifocal with a matrilineal bias. Some evidence for official
matrilineality in earlier centuries.
Northern Thailand: Matrilocal with a matrilineal bias.
Ryukyu Islands: Bilateral kinship; evidence of matrilocality in former times.
Korea: Matrifocal with some evidence of former matrilocality.
Sudan (Zar): Recent shift toward "budding" matrifocality. Evidence of matrilineality
in former times.
Brazil: Matrifocal.
Belize: Matrifocal and often matrilocal.
United States: Officially neolocal and bilateral, yet evidence of a matrilineal bias.
West Africa: Matrifocal and often de facto matrilocal.

legitimate. In addition, according to Tanner, matrifocal societies are character-

ized by relatively egalitarian relationships between the sexes, and relatively
equal valuations of men's and women's economic and ritual statuses. These
attributes certainly describe the social contexts of most women's religions.
For the purposes of this book, I find it most helpful to define matrifocality
as a complex of traits rather than as a uniform or absolute organizational
type. The cultures addressed here are sufficiently different from one another
to demand the use of somewhat flexible conceptual constructs. The complex
of traits that make up matrifocality include: identifying women primarily as
mothers rather than as wives, absence of adult males from the household,
matrilocality, matrilineality, structural authority vested in mothers, maternal
control of key economic resources, decision-making authority vested primar-
ily in the hands of mothers, and household groups in which members are
defined in terms of their relationship to the mother. For a society to be
considered matrifocal it does not need to exhibit all these traits; it needs to
exhibit many of them. Similarly, societies that are patrifocal may exhibit one
or two of these traits.
Nancy Tanner's insistence on structural and not only affective features pre-
vents expanding the concept of matrifocality to include societies in which the
mother-child bond is the strongest emotional bond, but in which women have
no real structural power. Thus, for example, neither Tanner nor I would con-
sider a society matrifocal in which children are far more attached to their moth-
ers than their fathers but in which women have no economic or social authority.
It is of interest to note that in many societies of that sort, women are more
religiously active than men, but men dominate positions of religious leadership
(e.g., the Catholic Church in Mexico; see Fromm and Maccoby 1970). The
evidence suggests that intense mother-child bonds facilitate women's religious
involvement, but without structural recognition—typically in the form of ma-
48 Priestess, Mother, Sacred Sister

trilocality or matrilineality—these bonds do not promote women's religious

leadership.
Matrilineality refers to a system of reckoning descent through the female
line.5 Matrilineal kinship is, statistically, far more rare than patrilineal kinship.
In George Murdock's (1967) sample of 1179 societies, 558 had patrilineal kin
groups while only 164 had matrilineal kin-groups. In the same sample 71%
percent of the societies were patrilocal or virilocal (the wife lives with the
husband or with the husband and his family). In Marvin Harris's words,
"Patrilocality and patrilineality are the statistically 'normal' mode of domestic
organization" (1985, 281). As we will see, most female-dominated religions
occur in societies that are neither patrilocal nor patrilineal.
In addition, according to Harris, because men are so reluctant to relinquish
control over their sons (and, I may add, their women), matrilineal societies are
more often avunculocal (sons at maturity go to live with their mothers' broth-
ers, thus allowing a core of related male kin to live together) than matrilocal.
Since none of the matrilineal societies in which women's religions occur is
avunculocal, the matrilineal and matrilocal nature of these societies is statisti-
cally even more outstanding.
In the following sections I will briefly summarize aspects of kinship organiza-
tion in cultures in which female-dominated religions are located. I believe that
each of these situations can be construed as matrifocal. In certain examples the
matrifocality receives clear definition through matrilineality and matrilocality.
In other examples matrifocality runs as an undercurrent through an officially
patrilineal and patrilocal society. What I wish to draw attention to is that even
in the latter situations, the female-dominated religion comes into play at the
point of matrifocality.

Northern Thailand (Matrilineal Spirit Cults)

The Northern Thai matrilineal spirit cults in which related women give offer-
ings to their matrilineal ancestors are a prime example of the matrifocal context
of women's religions. Formal Thai ideology—reinforced by Buddhist belief
and practice—posits male superiority (see Davis 1974). Although Thai women
are active in many spheres, there are certain crucial jobs from which women
are excluded. The village headman is always male, as are Buddhist monks and
officials in charge of irrigation.
Still, in comparison to most other known cultures, Northern Thai men's and
women's work is quite interchangeable. Both sexes work in the fields. Women
are particularly involved with communal transplanting and harvesting of rice
and in selling their goods at rural markets. Both men and women perform
household chores, and it is not uncommon for Northern Thai men to cook
their families' meals. Care of small children is carried out by women, girls,
men, and boys (Hanks and Hanks 1963). Children of both sexes are treated
with affection; the same ceremonial procedures are carried out at the birth of a
boy or a girl. In general, Thai culture tends to minimize sex differences.
In Southeast Asia, according to Penny Van Esterik, "Both in the gardens of
 The Social Contexts of Women's Religions 49
shifting cultivators and the wet-rice paddies of lowland agriculturalists, wo-
men's labor has been critically important" (1982, 8). In Thailand women not
only work hard, but also contribute to decision making in rural farming com-
munities. They are involved in market trading and entrepreneurial activity.
History and myth support the idea of powerful female political and legal
figures in Southeast Asia's past. More recently many women have left their
villages to work on road building crews, at factories, in services, and as prosti-
tutes. Women have a great deal of independence and authority, and often are
responsible for managing the financial affairs of their households.
Shulamith Potter (1977) has described Northern Thai social structure as
conceptually female centered, a system in which the significant blood ties are
those between women. While there is broad agreement in the ethnographic
literature regarding Northern Thai matrilocality, not all writers concur that
Northern Thai kinship organization is truly matrilineal. For the purposes of the
present book, it is sufficient to note that there seems to be a matrilineal bias that
emerges particularly clearly in the context of the domestic spirit cults (Wijeye-
wardene, personal communication 1992). Northern Thai matrilineal spirit
cults are an intrinsic part of a matrilocal and matrilineally oriented household
organization.
While there is variation among villages and among families, newly married
couples are typically expected to live with the bride's parents for one year, after
which they move to another house in the same compound as that of the bride's
parents. Female kin rely on each other for assistance with both agricultural and
domestic work. Postmarital living arrangements reinforce the strength of
women's identification with and loyalty to their own families. This has contin-
ued to be the case in urban areas, where married daughters often live near their
mothers (Hanks and Hanks 1963).
Matrilocal residence is both the ideal and reality; in the village studied by
Potter it occurs in 73% of all marriages. Daughters and sons inherit equally but
it is customary for sons to sell out their rights in the land to their sisters. In a
family with several daughters, the youngest daughter and her husband are
expected to continue living with her parents, caring for them in their old age,
and finally inheriting the house upon their deaths. When the family land is not
sufficient to support additional households, married sisters and female cousins
move off as a group with their husbands to new lands. "Thus, male individual-
ism and rivalry is dampened among these households and neighborhoods orga-
nized around co-operating females" (Hanks 1984, 106). Divorce, which can be
initiated by either partner, is frequent and typically it is the husband who goes
away while the wife keeps the land.
Gehan Wijeyewardene speculates that the Northern Thai preference for
matrilocality reflects a previous era in which men were subject to military
conscription and corvee labor. Until about fifty years ago, Northern Thailand
was characterized by small population, malaria, epidemics, warfare, and wild
animals. There was abundant irrigable rice land, yet a shortage of labor to
farm that land. Men were often absent hunting, in feudal services, or in
warfare. Warfare and migration have always made the Northern Thai, espc-
50 Priestess, Mother, Sacred Sister
daily the men, mobile. In traditional Thailand peasant men (the majority of
the population were peasants) owed four months yearly labor to feudal lords.
A recurrent theme in the development of Thai culture is the scarcity of labor
in relation to land. Women were the key asset for the continuity of peasant
society (19843, 288).
Northern Thai explain the ideological underpinnings for their matrilocal
household system by saying that two women with different spirits (women of
different matrilineal cult groups) cannot share the same roof. The implication
is that when a man marries he cannot bring his wife (who has a different spirit)
to live with his mother. Instead, he must go to live with her family. Childrear-
ing practices reflect and reinforce the notion that boys are destined to leave
their natal homes, while girls are expected to remain with their natal families.
When a boy is born, the mother is enjoined to recuperate by lying next to the
fire for 29 days; when a daughter is born she should lie near the fire for 30 or
more days. The shorter period for boys is said to ensure that they will be able
to wander abroad; the longer period for girls is said to ensure that they will
remain at home.
In Chapter i, we saw that women's dominance in economic activities re-
flects a cultural assessment of •women as more rooted in worldly attachments—
an assessment with clearly negative implications in a Buddhist society. Accord-
ing to Thomas Kirsch, "The institutions of uxorilocal [matrilocal] residence,
female ultimogeniture, and gaining access to land through a wife's family
reflect this grounding ('attachment') of women in the home locale and rural
life, a limitation of women's options and mobility compared to the situation of
young men" (1985, 313-314). What Kirsch does not add is that this very
attachment to land and family strengthens women's role in the matrilineal
spirit cults.
Shulamith Potter has argued that although Northern Thai culture is ma-
trilineal and matrilocal, women are not more powerful or influential than men.
Thailand is not a feminist Utopia (witness, for example, the widespread prosti-
tution); gender inequality affects many spheres of life; and gender dissonance
results from Buddhist assessments of women as rooted in this world. On the
other hand, women dominate the kinship system. As we will see, this is a
theme that recurs in many women's religions.

Belize (Black Carib Religion)

The next example is one that even more strongly emphasizes the role of
women in kinship organization, and the interplay between women's domi-
nance in kinship and in religion. Black Carib family and household organiza-
tion is matrifocal; that is, families revolve around a woman and her children.
The mother-child, and especially the mother-daughter link is the most impor-
tant and the most stable in Black Carib society. Mothers look after their daugh-
ters' children, daughters care for aging mothers, old women organize rituals
on behalf of their ancestors. If the focal woman leaves the household perma-
nently, because of death or any other reason, the household is likely to break
 The Social Contexts of Women's Religions 51
up. Other household members come and go with much less effect. In short,
each resident's primary tie is to the focal woman (Kerns 1983, 129; see also
Gonzalez 1969, 1983; Helms 1981).
Fathers are expected to provide financially for their children, yet many are
unable to. In any case, the bottom line is that mothers and mothers' mothers
see to it that children are not neglected. Women, who must care for their
children, can rarely do so single-handedly. The public sphere does not provide
mothers with lucrative employment, nor does the government force fathers to
support children. Women learn at an early age that they must depend on
others—in particular kin. The ritual complex of Black Carib women tests,
strengthens, and dramatizes kin ties.
In both the cities and villages of Belize approximately 50% of children under
the age of fourteen live with both biological parents, 25% with their mothers
but not fathers, and 25% with other relatives, most typically a grandmother or
maternal aunt (Sanford 1974). Even when the biological mother is present in
the household, the child may be cared for and 'mothered' by another relative.
Childcare is the responsibility of kin, not only of the mother, and whoever is
most interested in enacting the social role of mother is likely to do so. The
reason for the absence of the mother is almost always that she is away work-
ing. "The ideal role of the strong and nurturing mother can thus be fulfilled
whether one is a genetrix or not" (Sanford 1974, 398).
In contemporary Belize Black Carib men rarely work together. Women's
work (financial and ritual) is much more cooperative than men's. In Black
Carib villages men are frequently absent for long periods of time, either work-
ing or looking for work in the cities. Women, left to care for children, have
various economic strategies. They expect and often receive financial help from
their husbands and sons. They garden, take in laundry, and sell cakes, candles,
and dried fish in nearby markets.
Most daughters, at some point in their lives, turn to their mothers for
assistance. Daughters return to their mothers' houses for a number of reasons:
financial problems, marital problems, mistreatment, unemployment, the need
to go to the city to find a job (Kerns 1983, 112). Because absent children are
expected to send financial support to their mothers, mature women often serve
to redistribute financial resources among siblings.
Black Carib women are quite independent of men. Their labor and sexuality
are not treated as the property of men, nor are women under the authority of
men. Women choose their own spouses and leave unsatisfactory ones at will.
They control their own money and own their own property. Spouses generally
keep their finances separate. The kinship system is officially bilateral, yet al-
though all kin are expected to help each other, it is primarily women who are
responsible for the networks of reciprocity. Most daughters stay in the same
communities in which they grew up. Most marriages are among residents of
the same village, and women show a marked preference for living with their
own rather than with their spouses' kin (Kerns 1983).
Marriages are either legal or extra-legal, long term or short term, and no
stigma is attached to extra-legal relationships. Most marriages begin with
52 Priestess, Mother, Sacred Sister

emotional and physical attraction, and women do expect fathers to acknowl-

edge paternity. When a marriage dissolves, relationships with in-laws are
generally preserved. Because most people have more than one spouse during
their lives, networks of affmes tend to be large (Kerns 1983, 113-116). The
result is that village Black Carib women rarely interact with anyone who is
not kin (in some sense). Boys and girls play freely with each other, yet from a
fairly young age boys and girls (and men and women) show preference for
congregating with their own sex. Throughout their lives, men develop ties
with friends, while women's bonds are with kin. In short, cultural expecta-
tions for women and men are different. Men have more contact with the
outside world than women do. Women have more intra- and intervillage
contact with other Black Caribs (Kerns 1983, 93). The picture we have
painted here is of villages in which women are the permanent residents, and in
which women—linked together by bonds of kinship and mutual help—raise
and support children.
Black Carib children of both sexes are socialized to be independent, to be
able to take care of themselves, and also to be generous to their kin. Men are
expected to do much of the heavy work of building houses and clearing fields,
yet for reasons of expedience many tasks are performed by both sexes (Kerns
J983, 93). Black Carib society does discriminate against women in certain
ways. Although men and women are expected to enjoy and seek sexual plea-
sure, women are expected to control their sexual urges at various points in the
life-cycle, especially after birth. (Virginity before marriage is neither expected
nor desired.) Women are praised for sexual restraint; men are admired for
sexual conquests and for "breeding" many children with many women. From
a young age girls are dressed more modestly, and their behavior is more
monitored and controlled than is boys' (Kerns 1983, 90—92). Boys have fewer
responsibilities and spend less time at home.
Within Black Carib communities there is little official or authoritarian leader-
ship of any sort. Most property is personal and portable, few people own any
productive land, and inheritance is not considered an important means of
acquiring property (Kerns 1983, 76). We may assume that absence of the sorts
of centralized, male-dominated institutions that characterize many other soci-
eties has facilitated the important role played by women in Black Carib reli-
gion. More important, however, is the kinship system, which encourages
dependence on female kin and autonomy from male dominance.6

Upper Burma (Nat Religion)

Researchers have considered the kinship structure of Burma to be bilateral—
individuals are equally affiliated with the families of both of their parents.
Melford Spiro (whose ethnographic work in Upper Burma is considered au-
thoritative), however, found a definite matrilineal and matrilocal slant. "Al-
though kinship distance is a function of biological distance, there is some
suggestion that within the same degrees of genealogical distance the kinship tie
to females is conceived to be stronger than that to males. There is no
 The Social Contexts of Women's Religions 53
doubt . . . that this is the case at the affective level . . . but even at the cogni-
tive level there is some reason to believe that, because of the importance of the
womb, the (biological) stuff of kinship is believed to be transmitted more
through females than males, and that, consequently, more of it is shared with
females than with males" (1977, 46).
The Burmese term for the biological family is mitha su, literally "mother-
child group." The same term is used for the descent group, which, according
to Spiro, implies that the founding female contributes more than the founding
male to its common kinship. While each person belongs to both the father's
and the mother's descent group, Burmese say that they feel closer to their
maternal kin (1977, 65).
Residence patterns are typically neolocal. However, as in Thailand, it is the
custom for newlyweds to reside for some period of time with the bride's
parents, and the youngest daughter is expected to remain with her parents after
her marriage. For both parents the attachment to daughters is said to be
stronger than to sons (Spiro 1977, 81). Statistically, there are more households
of parents and married daughters than of parents and married sons. While there
is a preference for village endogamy, in the past it was the custom in inter-
village marriages for the husband to move to the wife's village.
Spiro found that fathers are viewed more as figures of authority and mothers
as figures of affection. The tie between mother and child is believed to be
stronger because of pregnancy and nursing. In partial exchange for this
nurturance, in Buddhist initiation the son "worships" his mother before his
father, and the merit he acquires from becoming a novice is transferred to his
mother. " . . . if cultural expressions are to be taken seriously, the mother may
be said to be viewed as the pivotal person in the family" (Spiro 1977, 84). In the
case of divorce, children are almost always given to the mother, not the father.
Mothers and daughters are especially close, and grown daughters and moth-
ers visit each other constantly. Mothers take care of daughters' children. The
mother-daughter bond is the core of the Burmese family. In many ways, the
mother-daughter relationship is seen as threatening to the wife-husband relation-
ship; men feel that women neglect their husbands for their mothers. It may well
be that this feeling is strengthened by lengthy sexual taboos: There is a taboo on
sex after the fifth month of pregnancy and while the woman is nursing. Sons are
demonstrably emotionally attached to their mothers, yet "the women of the
matriline comprise an especially close circle" (Spiro 1977, 89).
As we might expect, gender dissonance is a feature of Burmese culture.
Despite the dominant role of women in Burmese families, ideology accords
men a superior status. The Burmese view of marriage is that the man "owns"
the woman's sexuality; there is no notion that the woman "owns" the man in
the same way (Spiro 1977, 150). Girls—not boys—are expected to be virgins
when they become engaged. Men may have extramarital affairs, but not
women. Polygyny is permissible but rare (each co-wife has her own house).
Burmese women wish to be reborn as men—a higher incarnation and an easier
life (no pain of menstruation, pregnancy, or childbirth, and less housework.)
Men (not women) are believed to possess a sort of psychospiritual essence
54 Priestess, Mother, Sacred Sister
known as hpoun. Hpoun is a man's source of strength and power. Contact
(including indirect contact) with the vagina is seen as a threat to hpoun. The loss
of hpoun means that a man's powers decline, and that he is subordinate to his
wife and children. According to Spiro (1977, 266), men fear women's desire to
control them. They believe that women are temptresses and sorceresses, and
that women use magical techniques to dominate men. Men are attracted by
women but view their sexuality as a snare.
Burmese women are quite free and independent in most ways. Women
control the family economy and most retail trade. Proprietors of shops and
bazaars are preponderantly women, and women are well represented in large
business enterprises and in all professions except engineering. Women partici-
pate in agriculture and receive the same wages as men for the same work.
Legally men and women are equal and inherit equally. Women can own prop-
erty, enter freely into marriage, and initiate divorce. In modern Burma women
vote and are active in politics, and in traditional Burma a daughter could
succeed to the hereditary headmanship. In comparison to China or India, the
status of women is high: there is no footbinding, purdah, veil, child betrothal,
or widow immolation. Women have freedom of action and are involved in
both domestic and public society.
Spiro found that in many and possibly most marriages women are the
dominant partners (1977, 277). Despite the ideology of male superiority,
women make most of the major and minor decisions, and control the family's
purse (the husband turns over his earnings to his wife and she doles it out; men
are believed to be irresponsible with money and to squander it). Women are
expected to be more responsible for the family. In short, Burma presents a clear
picture of matrifocality.
There is some evidence that Burma kinship structure was once truly ma-
trilineal. I. C. S. Furnivall (1911) has argued that the custom of widowed
queens marrying a succession of men who upon marriage become kings is a
vestige of matrilineality. At the beginning of the twentieth century, when
Furnivall conducted his fieldwork, it was still the custom in Burmese villages
for the groom to live with or near the bride's family.

Ryiikyu Islands (Indigenous Religion)

Observers of the Ryukyu Islanders have been impressed by their lack of aggres-
sion and militarism, and their high level of interpersonal cooperation (Mabuchi
1968; Lebra I966).7 They have also noted that the status of women in tradi-
tional Ryukyuan society is high—certainly higher than in China or Japan
(Hopkins 1951; Lebra 1966). For example, division of labor on the basis of sex
is rather flexible and easily crossed.
The kinship system is usually described as patrilineal and patrilocal. A num-
ber of scholars, however, have argued that this organization is rather new (Ota
1989). In the past, brideservice (the groom working for the bride's family) was
commonly practiced, lasting for approximately two years, and matrilocal resi-
dence often continued until the first child was born. According to Takenori
 The Social Contexts of Women's Religions 55
Noguchi, the practice of a husband temporarily staying with his wife's family
is still common on at least some of the Ryukyu Islands (1966, 27). Robert
Spencer reports that in ancient times a woman, having high economic value,
did not go to live with her husband but stayed and worked in her parental
home, with her husband coming from his parental fields to stay with her
nights. According to the Ryukyuan scholar Shimabukuro Genichiro (quoted in
Spencer 1931, 96), this was still the custom in parts of Okinawa as late as the
beginning of the twentieth century, and Toichi Mabuchi (1967) found that in
the southern Ryukyus the kinship system is explicitly bilateral. 8
Although nowadays a woman typically goes to live with her husband's
family, it is not unusual for her brother or her parents to move in order to live
near her. In general, cognatic rather than agnatic kinship is important in peo-
ple's daily lives (Noguchi 1966, 28). Village endogamy is very much urged and
in some areas required. Thus patrilocality is attenuated by the fact that all
residents of a village are related to each other through multilayered kinship
ties. In addition, as we will see in Chapter 10, sisters regularly return to their
natal families to perform rituals. 9
To summarize, despite the inclination on the part of anthropologists to label
the Ryukyus as patrilineal and patrilocal, there is good evidence that until quite
recently the typical kinship pattern was bilateral and matrilocal.

Korea (Women's Household Religion

and Shamanism)
A somewhat parallel situation has been described for Korea. On the one hand,
Korea is a well-known patrilineal and patrilocal society. Confucian ancestor
worship means that male children are especially valued, and the first-born son
inherits the house and most of the property. On the other hand, it is likely that
Korean society was matrilocal until fairly recent times (Kendall 1985).
Korea became a Confucian society rather recently (by the seventeenth cen-
tury). 10 Until then, kinship was not exclusively patrilineal and daughters could
inherit. Husbands lived in the wives' homes for a number of years. A married
woman would eventually establish her own home, or join her mother-in-law's
home—but as a matron and mother, not as a young and powerless daughter-
in-law. Still today, daughters continue their relationship with their natal fami-
lies after marriage—many women give birth at their mother's homes; mothers
may give their married daughters financial help on the sly; daughters attend
shaman's kut rituals at their mothers' houses.
In contemporary Korea most couples have their own homes and within the
household women have a great deal of power (Lee 1984). The mother is in
charge of housekeeping, controlling storage, and supervising the daily life of
the family members (food, clothing, and sleeping). According to Youngsook
Kim Harvey, despite the official "patriarchy" of Korean society, the Korean
man "is an outsider in the heartland of his own household. His formal author-
ity is hardly ever challenged, but his actual authority within the household is
much curtailed, for, like the king who knows only what his ministers choose
56 Priestess, Mother, Sacred Sister

to tell him, he is often ignorant about his domestic affairs" (1976, 190). It seems
that the matrifocal household, rather than the patrilineal family, functions as
the socially and economically significant unity in Korean society.
Household religion and shaman's rituals reinforce women's pivotal position
in the household. "While dramatically acknowledging a woman's affective ties
to her own dead kin, a kut also reaffirms bonds between a woman and her
living kin. Like natal ancestors, living mothers, sisters, brothers, and brothers'
wives may be present when a woman holds a kut in her own home. . . . The
pattern of participation by living and dead . . . reveals a chain of households
linked by out-marrying women" (Kendall 1983, 106).

Brazil (Afro-Brazilian Religions)

Urban Brazil may seem at first glance to contradict my proposed correlation
between women's religions and matrifocality. The introduction of a wage and
market economy and the elimination of artisan jobs as the result of industrializa-
tion has hurt many women. Although in recent years there has been an increase
in education and employment for women, women are still heavily employed
in the service (especially domestic service) fields, with all its attendant prob-
lems. Women tend to work in the informal sector of the economy, where there
is no job security or protection, and employment tends to be low paid, tempo-
rary, and with no chance of advancement. Whereas the informal sector is
transitional for men—men tend to move on to better jobs—it is permanent for
women (Lerch 1980). Brazilian women suffer from a patriarchal ideology, a
sexual double standard, and economic discrimination.
On the other hand, many urban Brazilian women live for many years quite
independently of men. The dissolution of extended families in urban centers
means that fathers and brothers have limited control over daughters and sis-
ters. And the ease with which men leave wives and girlfriends means that
many women are in charge of their own finances, social interactions, and
mobility. "[A]mong poor urban populations the proportion of women who
head their own households without a male breadwinner present is high and
steadily increasing" (Merrick and Schmink 1983, 232)." Let us note that the
female-dominated Afro-Brazilian cults are a predominantly urban phenome-
non. Significantly, in Bahia, which is one of the most important centers of
Afro-Brazilian religions, households are typically headed by the mother, not
the father. Ruth Landes explicitly associates the high status of women in the
Afro-Brazilian cults with the authority Brazilian women have in the home, and
with the history of slavery, which recognized women and not men as family
heads (1947).

Twentieth-century United States (Feminist Spirituality)

The Feminist Spirituality Movement has developed in a demographic climate
of late marriage, frequent divorce, and single-parent families in which the
parent is almost always the mother. Combined with typical American neolocal
 The Social Contexts of Women's Religions 57
household arrangements, this means that many women are the heads of fami-
lies and households (Lopata 1987). Although American women certainly suffer
from economic discrimination, both single and married women are legally
entitled to control their own finances.
American families tend to be organized around the mother. Psychologists
have noted that for American children of both sexes the mother is the most
important figure (Baruch and Barnett 1983, 601). Daughters especially see
their mothers as role models. Bert Adams has observed that, "The close
mother-daughter bond may be extended further to indicate that females play a
more dominant role in the kin network than males. Young wives (in compari-
son to young husbands) tend to express a closer affectional relation to all
degrees of kin, are in slightly more frequent contact with siblings and secon-
dary kin, and are more likely to feel that kin are an important part of their
lives" (1969, 169-170). Jessie Bernard has pointed out that mother-daughter
relationships constitute the major component of old women's lives, "due to the
matriarchal emphasis on kinship relations in America" (1975, 150-151; also see
Yanagisako 1977; Hagestad 1984).12 And finally, Miriam Johnson points out
that the parents of the young middle-aged (the population from which most
Spiritual Feminists are drawn) often consisted of a mother whose family of
origin was of a higher socioeconomic class than that of the father. In this kind
of situation, the father's greater economic contribution could be neutralized by
the mother's class "superiority" and equal education. In light of the general
American tendency toward greater emotional ties between mothers and chil-
dren than between fathers and children, Johnson makes a claim for white
middle-class families leaning toward matrifocality (1988, 241).
A further element in the cultural setting of the Feminist Spirituality Move-
ment is the large number of lesbians among both leaders and participants.
Although demographic information is not available, we do know that many
Movement writings are by women who identify as lesbians. Lesbians are, on
the whole, a population of women who control their own finances and who
run their own households,13 and lesbian families are excellent examples of
matrifocal families.

Another Look at West Africa (Sande)

West Africa is not a clear-cut example of matrifocality, yet I believe that enough
elements of the trait complex are found to justify including it in the discussion
here. "An analysis of Kpelle [a society in which Sande secret societies are situ-
ated] consumption units reveals that nonkin members of a consumption unit
usually have ties to the woman in charge of the cooking pot" (Bledsoe 1980,
113). Women use kin ties to their own advantage. Although women do not have
legal authority over their children, they do have moral authority. Men and
women informants told Bledsoe that they "revered their mothers above every-
one else and felt morally obliged to support them in their old age" (1980, 114).
Among the West African Kpelle, if a woman bears a child out of wedlock the
child is considered hers unless the father has paid compensation to the mother's
58 Priestess, Mother, Sacred Sister
family. There is no stigma on these children and "in a sense children are a form
of property that fathers must pay for and maintain if they are to be considered
the legal 'owners' " (Bledsoe 1980, 91). In other words, patrilineality needs to
be actively created; unless that happens, children are considered part of their
mother's families.
Kpelle husbands have ongoing economic obligations to wives' parents. Al-
though the official ideology is virilocal, in actual fact only wealthy men can
afford to pay brideprice (a payment to the bride's parents in return for her
contributing her productive and reproductive capabilities to him or his family).
In reality, most men perform brideservice (living with or near the wife's parents
and contributing his productive capabilities to their household). Using demo-
graphic data, Bledsoe shows that it is far more often the wife's than the hus-
band's parents (and especially the wife's mother) who lives with a married
couple (1980, 108-109). Classical anthropologists would probably not consider
these last two traits to be indicative of matrifocality. Yet, in light of arguments
made by Tanner and Yanagisako (see above, page 46), it is reasonable to interpret
them as part of the matrifocal complex. It would be unwise to ignore the day-to-
day reality that a mother, her adult daughter, and the daughter's children live
together and most probably form the core household group. In sum, the ethno-
graphic evidence portrays Kpelle society as matrifocal and de facto matrilocal.
Regarding the Mende of Sierra Leone (another society in which Sande is
found), Harris and Sawyerr note the importance of the mother's brother who
"seems to have considerable authority perhaps even more than one's natural
father" (1968, 129). This piece of information is important in light of the
classical anthropological contention that a critical role for the mother's brother
is typically a characteristic of matrilineal societies. In addition, the Mende
believe that the physical parts of a child are derived from the father's semen,
but the spirit derives from the mother. "This accounts for the psychic
connexion which, it is assumed exists between mother and child" (1968, 129).
Harris and Sawyerr have also found that matrilocal residence is common
among the Mende. "In the days before any bride-price was paid, men married
only one wife by working for the parents of the bride. Even after the custom of
paying a bride-price had been established, there was a time when young men,
who could not provide an adequate sum to this end, adopted the earlier prac-
tice of going to live and work for the parents of the bride. . . . The young
man, anxious to marry the girl, was then absorbed into the girl's family as a
son of the house. . . . When the Sande ceremonies were over, the girl's mother
or one of her sisters would inform the man that he might cohabit with the girl,
and they began life together as man and wife. ... In such matrilocal mar-
riages, the wife's family was dominant" (1968, 129).

Haiti (Urban Vodou) and Ancient

Japanese Shamans
We will now look at two additional examples that strengthen the case for the
association between women's religions and matrifocality. In both examples,
 The Social Contexts of Women's Religions 59

we can trace the historical circumstances that accompanied the emergence or

decline of female dominance.
Haitian Vodou has evolved in the direction of female domination. In the
Haitian countryside men are the vast majority of Vodou priests; in the city as
many women as men are in positions of religious leadership. This trend is even
more conspicuous in New York, where Haitian Vodou seems to be almost
entirely a female-led enterprise. Because this development has been so rapid, it
is possible to identify the social structural changes that have accompanied the
modification of religious leadership.
Karen Brown has looked at the alterations in Haitian life resulting from
the move from the countryside to the city. In rural Haiti typical households
consist of extended families headed by an older male. In the cities, in con-
trast, women have better chances than men of finding employment. Urban
young male unemployment may be as high as 60%, and white employers
prefer hiring women for repetitive factory jobs. Moreover, urban women
have continued the market tradition of rural women, and many if not most
are involved in small-scale commerce of some kind. "The urban woman
spun away from the rural extended family frequently ends up not only in
charge of her house and her children—as she might well have been in the
country—but also solely responsible for their financial support" (Brown
1989, 262). What is being described here is a household arrangement charac-
terized by absence of adult males, female control of economic resources and
decision making, and primary ties revolving around women (often two gen-
erations of adult women). Clearly, the shift from male to female dominance
in Haitian religion occurred alongside the shift from patrifocal to matrifocal
households.
A similar historical shift seems to have taken place in Japan. In classical
Japanese mythology great shamanesses (miko) often married or were possessed
by heavenly kami or gods (Hori 1968, 70), and shamanesses had pivotal politi-
cal and ritual roles (1968, 112). Significantly, Hori argues that the role of the
miko changed with the transition from matrilineal to patrilineal society.
Under Chinese influence, the miko lost their important political role, and be-
came part of folk rather than official or great tradition culture (1968, 187).
Instead of practicing as independent religious specialists, they began to func-
tion as assistants to Buddhist mountain ascetics, and as village diviners, dancers
and singers, and reciters of ballads. In a process the reverse of that which
occurred in Haiti, in Japan the shift from female to male religious dominance
occurred alongside the shift from matrilineality to patrilineality.

Luvale of Zambia: Matrilineality

and Virilocality
A last brief example further clarifies the association between kinship arrange-
ments and women's religion. In Zambia, women of the Luvale tribe "partici-
pate in a system of curing that is run for women by women" (Spring 1978,
166). Because of the high infant mortality rate and the high incidence of illness
60 Priestess, Mother, Sacred Sister
14
among Luvale women, mothers undergo spirit possession rituals for their
own illnesses and on behalf of their sick children.
The Luvale are a matrilineal people who practice virilocal residence norms.
In other words, women move away from their natal families at marriage, yet
children belong to their mother's lineage. The ancestral shades or spirits that
afflict women belong to their own lineages and not their husband's. Anita
Spring asserts that "[Sjpirit possession and ritual become the vehicle for bond-
ing women of a matrilineage. Along with membership, ritual expertise is
passed down through matrilineal lines" (1978, 182). We see here how the
women's religion is concerned with matrilineality, in a society in which
women do not live in matrilocal households. Matrilineality is what is "picked
up" in the female dominated religion.
The spirit possession ritual for childbearing, for example, begins with public
all-night drumming, dancing and singing in front of the patient's home. Cult
members and relatives gather to aid the patient by singing and teaching her the
appropriate dances. The ritual leader invokes the "ancestral shades" who are
believed to be located within the patient's body and to have caused the sick-
ness. The shades appear via the possessed patient who dances in standardized
movements associated with the cult. Afterwards the patient continues to take
medicines and observe food and behavioral taboos. Following recovery the
possession ritual is repeated and the patient is initiated into the cult. Through
this ritual, bonds are formed among fellow and former sufferers, and misfor-
tune is concretized and cured. "It may be a unique situation that, in a ma-
trilineal society, women may gain access to ritual power through their own
ancestors, who were themselves women" (Spring 1976, in).
According to Spring, female models of interrelatedness may be intensified in
matrilineal cultures. "Undoubtedly, there is a feminine ideal of community
among . . . the Luvale . . . whose realization is prevented by virilocal resi-
dence. However, female rituals that transcend residential separation involve all
women in a community of concern for themselves, their children, their moth-
ers and grandmothers, and the whole body of female members of Luvale
society, both alive and departed" (1978, 188).
In matrilineal societies it often seems to be the case that even when the
formal or great tradition religion is male dominated, women gather in the
context of female-dominated cult groups or ritual constellations. Another ex-
ample of this phenomenon has been documented for the virilocal, matrilineal
Tulu speakers of India among whom large numbers of people—almost all
women—gather annually to propitiate the foundress-goddess Siri, sing her
legend, and become possessed by her or by members of her family (Claus
1975)-

MATRIFOCALITY AND RELIGION:

SOME CONCLUSIONS
A pattern has begun to emerge: Women's religions are most often found in
societies in which women control important resources, in which families are
focused upon mothers, in which kinship is matrilineal. I5 It bears emphasizing
 The Social Contexts of Women's Religions 61
that these societies are not matriarchal, nor are they even fully egalitarian. In all
the cultures dealt with in this chapter we found ideologies of male dominance,
uneven sexual mores, or other "unfair" beliefs and practices. Given that these
societies are not really egalitarian, the dominant position of women in the
family becomes even more significant.
Melford Spiro has argued that family and religion "are related to one another
in a systematic relationship which holds for no two other sociocultural systems"
(1984, 35). Religious systems, in Spire's view, are metaphorical expressions of
family relations. "[R]eligious symbols often represent the transformation and
elaboration, at the cultural level, of fantasies and cognitions that are found at the
psychological level, which in turn are produced by family relations at the social
level" (1984, 36). The examples presented in this chapter certainly bear out the
association between kinship and religion: Family and household arrangements,
more than any other social structural element, are correlated with the rise and
existence of female-dominated religions. To my mind, this is hardly surprising.
As Clifford Geertz has eloquently shown, religion is how people come to terms
with the "ultimate conditions" of existence, with birth, suffering, and death.
Birth, suffering, and death, of course, also constitute the very essence of kinship
and domestic relations. Thus we find that in societies in which kin ties center on
mothers, mothers have powerful roles in religious activity.
Three sets of factors provide insight into the association between women's
religions and matrifocality or matrilineality. First, in several instances we have
seen that the female-dominated religion is concerned with ancestor worship,
and that the ancestors are solely or primarily matrilineal ancestors. Women's
religions and matrilineality reinforce one another through the institutionaliza-
tion of matrilineal ancestor worship. It must be pointed out, however, that this
is only one possible scenario. In many matrilineal societies there is no female-
dominated religion, and many female-dominated religions do not concern
themselves with matrilineal ancestor worship.
Second, as I will show in greater detail in Chapter 3, women's religions tend
to sacralize the maternal role. This meshes well with the fact that matrilineal
and matrifocal societies also emphasize maternity. On the other hand, sacraliza-
tion or idealization of the maternal role is not unique to female-dominated
religions; it is also a feature of, for example, Catholicism and orthodox Juda-
ism. In short, matrilineal ancestor worship and the sacralization of maternity
provide some but not all of the answer to why women's religions are found in
matrifocal/lineal/local societies.
Third, we note with interest Alice Schlegel's findings that the combination
of matrilineality and matrilocality correlates with women having greater con-
trol over their persons and property, and with the availability to women of
positions of importance outside the home (1972, 98-99). Schlegel concludes
that when women are more independent at home, they are also more impor-
tant in the society at large. And, "Even when female autonomy is low, my
general impression is that it is higher in matrilineal than in patrilineal societies
within any given culture cluster" (Schlegel 1972, 141). Thus matrilineality and
matrilocality can be seen as allowing women the autonomy to develop their
own religions.16
62 Priestess, Mother, Sacred Sister

It may be more useful to turn the question around and instead of asking why
women's religions occur in matrifocal and matrilineal societies, asking why they
rarely occur in patrilineal societies. In patrilineal societies men are extremely
(sometimes even obsessively) concerned with guaranteeing that the children
whom their wives bear are truly their progeny. While it is likely that men in
many societies wish to avoid raising other men's children, in patrilineal societies
the issue is more acute—a man's children are the ones who will continue his
patriline. Mistaken incorporation of a child who is not really his into his patriline
can be a grievous affront to a man's ancestors, and can leave him without "true"
descendants to perform ancestor rituals on his behalf when he dies. When the
basic social link is along the male line, guaranteeing the soundness of that line
becomes a key social problem. As a result, patrilineal societies develop various
means—including religious means—to control women's fertility and thus to
ensure that the patriline is "true." Among the most common means are virginity
tests, modesty rules, and sexual taboos. As I will show in later chapters, these
sorts of beliefs and rituals are rarely found in female-dominated religions. Quite
to the contrary, in women's religions what we often find is that women's control
of their own fertility is somehow sacralized or institutionalized. Apparently,
religions that sacralize or institutionalize women's control of their own fertility
are anathema to patrilineal societies.

THE ZAR CULT AND DEPRIVATION

THEORY
One of our key examples—the zar cult—seems to be an exception to the
patterns laid out above. We now turn our attention to this extraordinarily
fascinating case.
I. M. Lewis has argued that women's possession cults are peripheral religious
movements that function as "thinly disguised protest movements directed
against the dominant sex. They thus play a significant part in the sex-war in
traditional societies and cultures where women lack more obvious and direct
means for forwarding their aims" (1975, 31). Lewis's theory is an example of
deprivation theory—the notion that people who are deprived of satisfaction in
other areas of their lives turn to religion as compensation or as an outlet for their
frustration. While Lewis himself has backed off from deprivation theory in the
new edition of his classic Ecstatic Religion (Second Edition, 1989), deprivation
theory has been, and continues to be, widely used to explain women's religious
participation.17
Deprivation theory suggests that when women suffer from intense subordi-
nation in political, ideological, economic, and social realms, membership in
marginal (female) religions may serve as a safety valve. By periodically partici-
pating in exciting religious rituals, women blow off enough steam to enable
them to continue functioning in an overall cultural context that denies them the
freedom to control their own lives.
Zar is the very example that inspired Lewis's ideas about women and religion.
As I will show in greater detail in Chapter 13, zar beliefs and rituals are situated
in highly patriarchal cultures. Women who participate in the zar cult are un-
 The Social Contexts of Women's Religions 63
doubtedly oppressed in almost every sphere of life. If we accept that zdr beliefs
and rituals attract women because of their subordinate status, we would expect
that when there is a change in women's status for the better zdr would decline,
and for the worse zdr would increase. Pamela Constantinides (1982) provides us
with a test case: In rural Sudanese villages social and kinship ties overlap, and se-
clusion of women is often irrelevant because there are few unrelated males from
whom women need to be secluded. Women engage in gardening work outside
the household, attend life-cycle events, and can band together to pressure their
husbands into protesting against a man who wants to divorce his wife. When a
woman moves to town her range of social activity is more limited, and she loses
the support of female kin. Whereas men who have moved to town canjoin work
associations, sports clubs, Muslim brotherhoods, and political parties, these
options are unavailable to women. And indeed, Constantinides has found that in
the Sudan zdr is particularly popular among women in urban environments.
Thus far everything that I have written is common wisdom concerning both
women and religion, and the zdr cult. Yet I suspect that this is not the entire
story.
Historical documentation for zdr religion is problematic. While some schol-
ars trace its inception to eighteenth-century Ethiopia, it is clear that similar
beliefs and rituals involving spirit possession are not new to Africa. What may
be new is the organization of these beliefs and rituals into an identifiable "cult"
rather than their being an intrinsic and inextricable part of the broader cultural
setting. Given the rapid crystallization and spread of zdr religion in the modern
era, it is crucial to note that zdr tends to grow together with male migration
from villages to cities. Could it be that as more and more men leave the
villages, the social organization and kinship structure in the villages shift toward
matrifocality?18
Here I would like to quote at some length from Janice Boddy's study of the
zdr in a Sudanese village. "Today more men native to Hofriyat live outside the
village for most of the year than live within it. However, adult women are far
less mobile. Whereas the nonresidency rate for males between the ages of
fifteen and forty-nine is 63 percent, the comparable rate for women is only 25
percent" (1989, 39). Boddy describes a typical contemporary family history:
After the wedding the groom goes back to the city to work while the bride
"remains in her natal household until after the births of several children. . . .
The ratio of adult women to adult men in Hofriyat is understandably high:
2.2:1. Yet it is a situation which many women say they prefer" (1989, 40-41,
my emphases). Boddy found that in Hofriyat there "is an explicit preference
for women who are near relatives to live together" (1989, 41). Even women
who live with their husbands' kin generally return to their own mothers'
homes during pregnancy, remaining there for a prolonged period after the
birth, often extending such visits to quasi-permanent residence. Boddy explic-
itly refers to the growing "tendency toward matrilocal residence and ma-
trifocal groupings" (1989, 83) in the village she studied.
Village women manage the household and maintain informal networks of
exchange of food, childcare, support, and information with other women.
Women have a great deal of say regarding whom their children marry because
64 Priestess, Mother, Sacred Sister
in this highly sexually segregated society only women can provide information
about potential brides. Women prefer their sons to marry girls from their own
family, not from their husband's family. And indeed, mother's brother's daugh-
ter marriage is more frequent than father's sister's daughter. This preference
serves to strengthen the bonds among matrilineally related women.
Women spend a great deal of their time socializing and visiting, especially
with sick neighbors and relatives, new mothers, girls and boys after circumci-
sion, and mourners. Women mostly socialize with other women, yet they also
maintain strong social ties with their brothers and sons (relations with hus-
bands tend to be weaker). Rural Northern Sudanese women, according to
Boddy, are far from social isolates, yet they are highly active in zar religion.
I would argue that zar possession both mirrors and enhances the matrifocal
leanings of the Northern Sudanese village.19 Zar affliction is transmitted along
the maternal line, particularly in cases in which the zar has been turned into a
powerful protective spirit (Messing 1958). Thus mothers and daughters or
sisters may share the same spirit; "[Tjhese bonds resemble the alignments of
descent" (Boddy 1989, 241).
In the preceding paragraphs I have tried to show that in the villages zar is
associated with matrifocality. Yet we began this section with reference to the
zar cult among urban women, and with an assumption that because of their
greater social deprivation they have greater need of zar involvement. Now let
us try to rephrase the same bit of information. According to I. M. Lewis's
most recent book, "The implicitly rebellious, feminist tome of Ethiopian zar
is, perhaps, most clearly evident in towns where a high proportion of mar-
ginalised single women, living autonomously as beer sellers and prostitutes,
are heavily involved in zar coteries (my emphases)" (1991, 3). In other words,
zar flourishes among urban women who live in matrifocal households!
Could it be that the same woman whom one researcher sees as lonely and
deprived, another researcher sees as independent and autonomous? In reality,
the same woman may well be both lonely and autonomous. The question is
which side of the coin—deprivation or independence—more convincingly
explains women's religious participation. In light of the other examples of
women's religions that we have seen in this chapter, I unequivocally argue for
independence.
Deprivation theory would lead us to expect that in highly patriarchal soci-
eties women have more of a need to use religion as an emotional outlet. In
relatively egalitarian societies women would either have less need for religious
activism and so be content with lower level membership roles in male-
dominated religions, or the mainstream religion itself would offer women
meaningful roles and so obviate the need for an autonomous women's religion.
While this hypothesis sounds eminently plausible, I believe it to be untrue.
In thoroughly patriarchal societies women are unlikely to have the intellectual,
financial, or social freedom to establish or join women's religions. A far more
common religious strategy in very patriarchal cultures is for women to "domes-
ticate" religion—to reinterpret aspects of the dominant male religion in ways
that they as women find meaningful.
 The Social Contexts of Women's Religions 65
Elderly Kurdish Jewish women in Israel, for example, are deprived by al-
most every measure of social and psychological deprivation: they are poor,
old, sick .and illiterate in a culture that values money, youth, health, and educa-
tion; they are women in a man's world; they are pious members of a religious
system that relegates women to the sidelines ritually and theologically. As I
have shown in an earlier book (Sered 1992), these women do not establish an
alternative, female-dominated religion. Instead, they modify and reinterpret
Jewish rituals, symbols, institutions, and beliefs in directions that give meaning
to their role as mothers and protect the children for whom they, as mothers,
are responsible. Similar religious strategies and emphases have been noted for
Muslim women (Beck 1980; Jamzadeh and Mills 1986; Ladislav 1988), urban
Hindu women (Beech, 1982), rural Hindu women (Thompson 1983; Wadley
1980), Spanish Christian women (Christian 1972), and Christian women in
rural Greece (Danforth 1982).
An even more serious problem with deprivation theory is one that it shares
with almost all functionalist interpretations of religious behavior: It assumes
that people (and, it seems, especially women) behave religiously because of
social and psychological reasons rather than for religious reasons. Leading
deprivation theory proponents Rodney Stark and William Bainbridge (1985,
417) claim that American women more than men join novel religious move-
ments, and in particular deviant religious movements, because of the women's
relative deprivation in career and other public opportunities, and because of
their solitary confinement in the nuclear family which turns women into social
isolates. The essence of their argument is that women join religious groups for
nonreligious motives. My own fieldwork among elderly Kurdish Jewish women
in Israel suggests that this formula glosses over the actuality that many women
do have nonreligious options that would also give them a chance to get out of
the house. The Kurdish women, for example, had access to a number of pro-
grams for senior citizens, yet chose to attend a senior citizens' day center at
which almost all the activities, lectures, and tours deal with religion. Put dif-
ferently, religiously active women should be understood as having chosen to
be religiously active (unless in a specific instance there is clear evidence to the
contrary).
A number of recent studies have begun to challenge deprivation theory in
regard to women's religiosity. Arvind Sharma (1977) has asked how and why
women in ancient India became Buddhist nuns. Out of a total of 68 women for
whom he found textual information, at least 42 indicated that it was the spiri-
tual attraction of Buddhism and not some form of personal, familial, or social
deprivation that led them to become Buddhist nuns. Sharma found that depri-
vation theory explained only a small number of the cases.
Contrary to what deprivation theory would lead one to expect, the West
African example cited at the beginning of this chapter suggests that some
economic independence may be a "good" setting for women's religions.
Nineteenth-century America shows that situations of social change may lend
themselves to the formation of women's religions. All the examples indicate
that gender disjunction or gender dissonance (manifested either as rapidly
66 Priestess, Mother, Sacred Sister
changing gender relationships or as important discrepancies between gender
ideology and practice) is a feature of societies in which there are women's
religions. But the examples that provide the most meaningful paradigm are
those that show matrifocality (and especially matrilineality and matrilocality)
as a significant social structural correlate of women's religions.
Matrifocality, more than any other element of the social structure, seems to
provide women with a sense that their roles are powerful and important, with
the opportunity for groups of related women to congregate on a regular basis,
with an absence of ongoing male supervision of women's activities, and with
access to vital social and economic resources.
The goal of this chapter was to explore in what types of societies women's
religions are found. In looking for the "cause" of religious behavior in social
structure, and especially kinship organization, we have followed along well-
worn conceptual paths. Women's religious behavior has traditionally been
treated as especially rooted in social structure (rather than in abstract thinking).
A close look at the situations described in this book, however, suggests a
dynamic interplay between religion and other aspects of human experience.
The examples developed in this chapter leave us wondering whether female-
dominated religions develop in matrifocal societies, or whether matrifocality is
a structural expression of an ideology of female power.
The significant point, to my mind, is that women's religions not only occur
in matrilineal and matrifocal societies, but that they also emphasize and pre-
serve matrilineality and matrifocality. In officially matrilineal societies such as
Northern Thailand the connection is obvious, but in the other examples it is
perhaps of even greater interest. Sande membership, for instance, ensures that
a married woman retains her ties with her natal family. Sande women return to
their mother's chapters (where they themselves were initiated) to give birth.
Women's religions typically tie into the society at the point of matrifocality or
matrilineality, strengthening and dramatizing that point.
In addition to the features of matrifocality that we have spelled out, Nancy
Tanner has noted two other characteristics of matrifocality: a woman's role as
mother is more important than her role as wife, and the role of mother is
ritually elaborated. In most female-dominated religions, as in most matrifocal
societies, a woman's identity as mother predominates over her identity as wife.
To a large extent, women's religions center around the ritual elaboration of the
maternal role. In the next chapter we will see how this works. We will look more
closely at the multidirectional borrowing of contexts, needs, and meanings
between women's familial lives and women's religions. Women's religions not
only occur in matrifocal societies, but motherhood is what brings many women
to women's religions, and in these religions maternity is a potent symbol.

Notes
I. This does not mean that gender consonance is a cross-culturally "normal" situa-
tion. It simply means that women's religions seem to occur in situations in which the
degree of gender dissonance is greater than usual.
 The Social Contexts of Women's Religions 67
2. The reader may notice that I have avoided using the words "patriarchal" or
"egalitarian" in laying out the social contexts of these religions. Cross-culturally, gender
inequality takes a multitude of forms. Evaluating particular cultures as less or more
patriarchal is a complex and often delicate task. In recent years many feminist anthro-
pologists have concluded that the most interesting question is not "Is this society
sexist?" but rather "What manifestations of patriarchy are most apparent in this cul-
ture?" For the purposes of this book (and with some misgivings), I define as "highly
patriarchal" societies in which women do not exert control over their own time, bodies,
fertility, finances, and mobility. "Egalitarian" (or relatively egalitarian) societies are
ones in which women are free to make important decisions regarding their own lives,
and have input into decisions that affect the society as a whole. The majority, although
not all, of the women's religions are located in relatively egalitarian societies.
3. This is not an unusual scenario. Moral superiority often correlates with structural
inferiority.
4. Some anthropologists, including Nancy Tanner, now believe that the "original"
or "core" human family is the mother-child unit (cf. also Gough 1968) and that other
kinship systems are superstructures that build on or partially negate this matrifocal core.
While I suspend judgment regarding the absolute truth of this claim (I do find it
tempting), such a universalistic notion of matrifocality does not help us understand the
particular societies in which female-dominated religions occur. What is relevant for our
purposes is that in these particular societies matrifocality is explicit.
5. Anthropologists formerly used a much stricter definition of matrilineality and
patrilineality, basing their conceptions on African systems of unilineal descent. We now
know that the African models are not particularly useful outside Africa. In Southeast
Asia, for instance, the notion of unilineal descent is problematic. On the other hand, in
many cultures children are identified as belonging primarily to either the mother's or
the father's kin-group. They inherit rights, obligations, and loyalties almost completely
from one side or the other. It is in this broader, non-African sense that I use the term
matrilineality.
According to Yanagisako, anthropology traditionally labeled household organization
"according to the tie between its most genealogically 'close' members, because it is
presumed that this relationship forms its structural core. If it contains two adult broth-
ers, their wives and children, it is thus labeled a fraternal-joint family or household. . . .
Yet obviously there is more to 'family structure' than genealogical composition" (1979,
185). Yanagisako's observation allows us to de-emphasize the patrilineal ideology of,
for example, Korea, the Ryukyu Islands, and the Northern Sudan, and look instead at
the actual relationships and power distribution in the households. In all three cases,
letting go of an overemphasis on genealogy allows us to see the actual matrifocal
household arrangements.
6. In a recent paper Janet Chernela (1991) has re-examined one of the most notori-
ous elements of Black Carib culture—couvade. Generations of anthropologists have
been fascinated by cultures in which husbands imitate their wives' pregnancies and
labor pains, or observe pregnancy or postpartum taboos along with or instead of their
wives. Chernela explains couvade among Black Caribs of Honduras in terms of their
belief that the father and the infant are inseparably linked because they share the same
soul and because the baby is created out of the father's blood (the mother just nourishes
it). Since the bond between the father and child is so strong, injudicious acts on the part
of the father may injure the infant. Thus men must abstain from strenuous work when
their babies are small. During this extended time, a woman's brothers (rather than her
husband) are providers to her offspring. Couvade, by strengthening the economic ties
68 Priestess, Mother, Sacred Sister
between sisters and brothers and by limiting the role of husbands, serves to underscore
the matrifocal nature of Black Carib society.
7. There is little criminality, suicide, or violence on the Islands. During Sho Shin's
reign (fifteenth-sixteenth century) the army was disbanded, and private use and owner-
ship of arms outlawed.
8. During the past five or six generations there has been a tendency not to record
women in genealogical records; before then women were always recorded. William
Lebra sees this as evidence of the increasing stress on the male line in modern Okinawan
society (1966).
9. A further bit of information that I am not sure how to interpret is Noguchi's
observation that on Ikema Island wives are often older than their husbands (1966, 27).
Since this arrangement is, from a cross-cultural perspective, anomalous (cross-culturally,
husbands are far more often older than wives, and this age difference is typically reflected
and enhanced through norms of male dominance and female subordination), we may
wonder if there is some connection between female religious dominance and familial
dominance that could be linked to wives being older than husbands.
10. Confucian ideology was known long before, and Neo-Confucian philosophy
was adopted as a social blueprint in the fifteenth century, but the transformation of
institutions took quite a bit longer (Kendall, personal communication 1992).
11. More urban than rural Brazilian households are headed by women. Merrick and
Schmink (1983) estimate that between 1960 and 1970 one out of every five urban
households was headed by a woman, and that number is increasing. Households headed
by women show a higher incidence of poverty—44.9% fall below the poverty line,
compared with 27.4% of households headed by men. There are several factors that
contribute to the feminization of poverty in Brazil: in female-headed households it is
unlikely that there is a second earner, women lack access to the formal sector job
market, female-headed households have less access to basic government services such as
the government sponsored health program (Institute Nacional de Previdencia Social),
and working mothers suffer from the lack of day care facilities.
12. The precise cultural constellation I have just described is true for late twentieth-
century America, yet there is historical evidence that American society of the nineteenth
century was also characterized by neolocality and by particularly strong bonds among
women. Carol Smith Rosenberg, in a fascinating essay entitled "The Female World of
Love and Ritual," has demonstrated that in the late eighteenth and nineteenth centuries,
intimate and close female relationships were socially acceptable and normal parts of life.
Close and trusting and loving female relationships included sisters, friends, mothers and
daughters, young children, and mature women. Social networks tended to be single
sex, and women shared in one another's birthings, deaths, weddings, vacations, and
educational experiences. All of this should be seen as part of the background to the
emergence of female-dominated religions in nineteenth-century America.
13. This is not meant to belittle the discrimination lesbians suffer in many areas of
American life.
14. Luvale women, according to Anita Spring, have few live births, many genital
and urinary-tract diseases, barrenness, and frequent miscarriages and stillbirths. Many
women suffer from chronic abdominal pain, dysmenorrhea, and fevers caused by bacte-
ria and parasites. These diseases are seldom fatal, but do cause infertility and weak
health. The women arc particularly susceptible to bilharziasis, a disease that affects the
genitalia, because they enter stagnant pools of water to catch fish with drag baskets.
Almost half the adult women reported symptoms consistent with either gonorrhea or
bilharziasis (see Chapter 5 for more on women and illness).
 The Social Contexts of Women's Religions 69
15. One female-dominated religion that is located in a truly patrilocal and patrilineal
society is the religion of the Tetum of Timor Island. David Hicks describes the complex
series of birth rituals (some performed by men and some by women) that culminates in
a rite in which "[T]he infant's agnates place themselves on one side of the mat facing
their affines, opposite. With the baby between, agnates and affines commence to lam-
poon each other, with the women cupping their hands into nearby pitchers, and fling-
ing water at those on the opposite side of the mat [i.e., their in-laws]" (1984, 48). This
rite seems to indicate a certain discomfort with the patrilineal identity of the newborn
baby.
16. Not all matrilineal societies include female-dominated religions. The Hopi—the
very society studied by Schlegel—is an example of a matrilineal and highly egalitarian
society in which men dominate the religious domain. Women do have active roles in the
religion, but men's roles are more prominent (Schlegel 1977).
17. "Lewis bashing" has become a favorite pastime of some anthropologists. While I
do not agree with everything that Lewis wrote about women and religion, he is one of
the very few scholars who has even tried to develop cross-culturally meaningful models
for looking at women's religious lives. It is worth pointing out that the bulk of his work
on women and religion was completed before the introduction of feminist ideas into the
academy.
18. The link between male migration and women's religion is intriguing and needs
further research. Regarding Mexican Spiritualism, for example, Kaja Finkler reports
that more adherents than non-adherents are unmarried and more are members of
families in which at least one person migrated in search of wage labor at some point
(1986, 633).
19. Boddy cites evidence suggesting that in the past Northern Sudanese society was
not patrilineal. "From well before the Christian millennium until at least the fifteenth
century AD, and perhaps for some centuries thereafter, matrilineality with preferred
endogamy and adelphic succession were predominant structural principles in the re-
gion of Sudan stretching north along the Nile from Khartoum to the Dongola Reach"
(1992, 7)-
 3
Maternity and Meaning

In the previous chapter we have seen that family organization, more than any
other aspect of social structure, sets the stage for women's religions. A range of
identities describe women's familial relationships: daughter, sister, wife, co-
wife, mother, grandmother, aunt, cousin, niece. In different cultural situa-
tions, and throughout the life cycle, various of these identities become most
salient for specific women. The implications of this seemingly obvious state-
ment are important.
According to anthropologist Karen Sacks (1979), in African societies charac-
terized by a patrilineal kin corporate mode of production, young women—
daughters in their natal groups and wives in their affinal groups—have a
uniformly low status. As a woman ages, she becomes a sister in her own group
(her brothers also age and so her key relationship is to her brothers who have
now replaced father as decision makers and resource allocators) and a mother
(of adult children) in her husband's group. Because mothers and sisters (as
opposed to daughters and wives) control their own and other people's labor,
these roles entail higher status. Sacks looked only at African societies, yet it is
likely that the distinction between mother and wife holds true in many other
cultures as well. For example, in Northern Thailand a woman's status as
mother is significantly higher than her status as wife (Mougne 1984; see also
Martin 1990 on Mexican women's use of motherhood as a source of spiritual
and political strength). Miriam Johnson cites studies carried out in the United
States indicating that in young children's fantasy play both boys and girls
depict wives as helpless, yet they depict mothers as nurturant and efficient
(1988, 198).
Religions, like all social institutions, tend to address individuals through
only one or two of their possible identities. In male-dominated religions
women receive attention when they impinge on male space or interests. Not
surprisingly, male-dominated religions—institutions in which adult men are in
charge—deal with and define women primarily as wives. As a result, much of
the discourse regarding women concerns menstrual pollution and taboos as
they affect husbands (a woman's children and even her parents are rarely
72 Priestess, Mother, Sacred Sister
affected by her menstruating). Male-dominated religions also relate to women
as daughters, especially in the sense of daughters whose virginity must be
guarded over.
In female-dominated religions neither wife nor daughter is the most sa-
lient aspect of women's identity (which is not to say that women are never
addressed as wives or daughters). Instead, women receive attention primar-
ily as mothers, grandmothers, or sisters. It is typically in these roles that
women join these religions; women's concerns as mothers, grandmothers,
or sisters are addressed by these religions; women's roles as mothers, grand-
mothers, or sisters are highly elaborated and dramatized in these religions;
and mothers, grandmothers, and sisters have prominent mythic and sym-
bolic roles.
In Chapters 10 and 13 I look more closely at women as sisters. The remain-
der of the present chapter is dedicated to exploring the religious significance of
motherhood in women's religions. In light of the discussion of social context in
the previous chapter, it is clear that the accent on women as mothers in female-
dominated religions is associated with the matrifocal nature of the cultures in
which these religions are located. Matrifocal societies tend toward structural
arrangements that de-emphasize a definition of women as wives and give
prominence to women as mothers.

MOTHERHOOD AND RELIGION

Motherhood is a multifaceted experience, both in terms of how a particu-
lar woman feels and behaves as a mother during the course of her life (preg-
nant women and great-grandmothers experience motherhood in different
ways) and in terms of how different cultures construct motherhood. That
women give birth is biological fact. The ways in which cultures interpret
that fact are infinitely variable. Still, the intense and powerful psychological,
social, and biological experiences of motherhood are central in the lives of
most women. Religions (male dominated and female dominated) address
issues of motherhood. Religions may tell women how many children to
have, when, and with whom; religions may tell women that celibacy is
preferable to fertility; religions may tell women that infertility is a punish-
ment from the gods; and religions may tell women how to raise and edu-
cate their children. Cross-culturally, religions interpret the experiences of
motherhood.
Many male-dominated religions recognize that motherhood gives women
power. This realization leads to beliefs and practices such as subincision (in
which men cut their penises in order to mimic women's natural functions),
belief in virgin mother goddesses or male gods or heroes who create out of
their own bodies (Adam and Eve; Zeus and Athena), couvade, and glorifica-
tion of male supremacy (see Kittay 1983).
Motherhood is a fundamental image, a key ritual focus, and a chief theologi-
cal concern in women's religions. In some women's religions motherhood is
believed to bestow upon women deep spiritual insights. In others motherhood
 Maternity and Meaning 73
is what led women to be dissatisfied with the male-dominated religions prac-
ticed by their fathers or husbands. And in still other female-dominated reli-
gions motherhood is what shapes women's ritual roles. Through ritual and
theology, female-dominated religions enhance, dramatize, and strengthen
women's identities as mothers.
In contrast to male-dominated religions where mothers are typically envi-
sioned from the perspective of children (and especially sons; see Chapter 8 on
mother goddesses as projections of male infantile fantasies), in female-domina-
ted religions the concern with motherhood is from the perspective of the
mothers themselves.
Psychologist of religion Diane Jonte-Pace has used psychoanalytic object
relations theory to develop a multidimensional understanding of the connec-
tion between mothering and religion. According to Jonte-Pace, object rela-
tions theorists have suggested that "the psychological capacity for religious
experience lies in the relational maternal-infant matrix out of which a sense of
selfhood and otherness emerges" (1987, 319). This capacity, according to
theorists, is rooted in the early experience of symbiosis before the infant
discerns the existence of a separate self.
While object relations theory is useful to feminist analysis, it is crucial to
bear in mind that like all psychoanalytic theory, it locates religiosity (together
with almost all other elements of personality) in early stages of psychological
development. Motherhood, however, is an experience of adult women. Alex-
andra Kaplan and Janet Surrey critique most existing theories of psychologi-
cal development on just that point. They suggest that women's capabilities
for relationships should be examined not only in terms of the early precursors
of this relational capacity as developed within the mother-daughter relation-
ship, but also in terms of the development of this relational capacity through-
out the life cycle. They propose a model of psychological growth that under-
scores "development as a dynamic process of growth within relationship"
(1984, 87).
The evidence of women's religions highlights the relevance of developmen-
tal models. In this and the next chapter I present women who, because of their
adult experience of child death, became dissatisfied with the religion in which
they were raised. It is difficult to interpret this sort of process within a model
that situates religion in infancy. Psychological development does not end with
early adulthood, and in the second part of the present chapter I show how
grandmothers differ from mothers in their experience of religion.
Before we continue, let me clarify that I do not claim that the mother-child
bond is intense, loving, and momentous in all cultures; I do claim that this is
the case in the societies in which women's religions are situated. Let me also
clarify that I have not found that in all female-dominated religions women are
seen only as mothers. As Diana Burfield (1983) has pointed out, in Theosophy
maternal feelings seemed not to be emphasized, and several female leaders
were childless. On the other hand, in the vast majority of known women's
religions, motherhood so clearly emerges as the pivotal theme that I feel justi-
fied in organizing my own observations around the notion of motherhood.
74 Priestess, Mother, Sacred Sister

Ritual Control of Fertility—Sande

In women's religions childbirth is not merely a biological event. Motherhood
is vested with important cultural significance, and rituals express and validate
the emotional and symbolic meanings of motherhood. In many women's reli-
gions, theology and ceremony provide women with some measure of control
over fertility.
Sande rituals give social definition to motherhood, and teach that definition
to young initiates. The gender ideology taught by Sande stresses gender distinc-
tion (women and men are clearly different), respect for women's bodies, depen-
dence on fellow women, and preserving women's secret knowledge and
power.
During the Sande initiation period girls are fed abundant quantities of high
quality food. Fatness is seen as linked to beauty, prosperity, health, and fertil-
ity. Carol MacCormack (1982, 125) suggests that in geographical regions (such
as Sierra Leone) where there are periodic food shortages, the deliberate fatten-
ing of young women combined with the belief that full-bodied mature women
are more beautiful has a selective advantage in terms of promoting fertility and
decreasing infant mortality. This is a crucial consideration in a society in which
women give birth to many children (often eight or more), and more than half
are likely to die at a young age. In addition, Sande rules prohibit sexual inter-
course while a woman is lactating, ensuring some spacing of children. Sande
women know that they control a scarce and important resource: offspring.
Sande women also know that they can withhold that resource if their husbands
violate Sande laws.
Initiated Sande women "do not have sexual intercourse in any place, at any
time, with any person they wish, as animals do. Such 'natural' behavior would
be a grievous offense to ancestral spirits, whose wrath might rain disease and
infertility on the people and the land. Successful bearing and rearing of children
is informed by Sande knowledge about hygiene, nutrition, medicine, and
myriad other practical techniques rather than being a careless matter of doing
what conies naturally. . . . Rather than being uncontrolled reproduction ma-
chines, Sande women, with their secret knowledge, public laws, legitimate
sanctions, and hierarchical organization, bring women's biology under the
most careful cultural control" (MacCormack 1977, 94).
According to Carol MacCormack, a central function of Sande is to provide
women and men with a cultural rather than a purely biological understanding
of fertility. Sande beliefs and ceremonies are a striking example of how a
women's religion empowers motherhood.

At the Center of the Web—Korean Mothers

Almost all the women's religions enhance and support women's familial
involvement and provide women with concepts and rituals that interpret
motherhood from a gynocentric perspective. In the following paragraphs the
household-oriented religion of the women of Korea is presented. The crucial
 Maternity and Meaning 75
point, as shown by Laurel Kendall, is that for Korean women it is not the
formal definition of family (in the sense of official patrilineality) but rather a
relationship-based informal conception of family (in the sense of all people
dead and alive with whom the woman has ties that she herself perceives as
familial) that has meaning in the female ritual sphere.
Korean women are marginal in formal Confucianist and Buddhist ritual. A
daughter who marries out of her natal family is no longer part of that family, and
as a wife in her husband's family she has only minor and limited ritual obliga-
tions. From the point of view of children, mother's relatives have no ancestor
rituals. "Some informants even say that a man's natural mother is entitled to
[Confucianist] ancestor rituals only because she is the wife of his father—it does
not matter that she bore him" (Janelli andjanelli 1982, 120). In Korean men's
rituals, family hierarchy and patrilineal kinship relations are critical.
Women's household rituals and shaman's kut, on the other hand, include the
extended bilateral family. Various ancestors and relatives who are irrelevant to
men's rituals are important to kut (dead children, matrilineal relatives, husband's
married sisters, etc.). In women's household religion, mothers have key ritual
and theological roles. A great deal of women's ritual life revolves around domes-
tic rituals aimed at safeguarding the well-being of their families. As a mother
with her own household a woman comes into her own religiously. The house-
hold religion of Korean women has to do, above all, with women's role as moth-
ers. Illness of a child or other family member is often the reason for a Korean
woman to make a food offering to the household gods or to consult a shaman.
Whereas both men's and women's religion in Korea are oriented toward
family, male religion treats family as a hierarchical ladder linking generations
of men to their ancestors. In the female religion family is envisioned as a web
of varied relationships—a web in which a mother/housewife stands at the hub.
Korean men's religious rituals revolve around generations of fathers. Korean
women's rituals revolve around mothers.

Symbolic Motherhood—Shakers,
Black Spiritual Churches, and
Tensho-Kotai-Jingu-Kyu
Moving from concrete motherhood to more symbolic treatments, we find that
the Shakers used imagery that reflected maternity. Founder Ann Lee described
herself as having "labored" until she received the gifts of God and the Gospel.
When Lee experienced her soul's encounter with God, "I felt as sensibly as ever
a woman did a child, when she was delivered of it. Then I felt an unspeakable
joy in God, and my flesh came upon me, like the flesh of an infant" (quoted in
Kern 1981, 73). Later Shakers also used childbirth metaphors in regard to Ann
Lee. Sarah Lucas, for example, described Lee's role in delivering believers
"from the fetal condition of our natural and worldly life" and birthing souls
"out of the world state into the Christ state" (quoted in Kitch 1989, 137).
According to Shaker traditions, Ann Lee embodied the highest ideals of
motherhood. "She was depicted as a concerned, loving, solicitous individual
76 Priestess, Mother, Sacred Sister
who developed deep personal relationships with her followers—her children"
(Stein 1992, 22).1
Images of laboring and pregnancy and nurturing children were important
symbols for Shakers, and many of the early Shaker converts regarded Lee as
their mother. Procter-Smith (1985, 148) quotes a song about Ann Lee written
in the 1820s (in this song "Mother" is Ann Lee):

Mothers love is like an ocean

Mothers love will make me free
Mothers love it is so wholesome
I can skip and dance and play
Mothers love it beautifies me
Mothers love is pure I know
Precious love o how I prize thee
Mothers love come round me flow.

Sally Kitch persuasively argues that for the Shakers the "metaphor of spiri-
tual motherhood is a positive symbol for women because it transforms quali-
ties that have been perceived as liabilities of female gender symbolism into
strengths. . . . The metaphor of spiritual motherhood reveals the cultural
value hidden in traditional female characteristics, personalities, and experi-
ences" (1989, 138). We will return to this theme in Chapter 10, where I show
that many women's religions accept prevalent notions of femininity, while
reinterpreting as strengths traits defined as weaknesses in the society at large.
Female founders of religions are often called Mother, and utilize maternal sym-
bols to explain their religious roles. For example, Mother Catherine, one of the
early leaders of the New Orleans Black Spiritual movement, claimed, "I got all
kinds of children, but I am they mother. Some of'em are saints, some of'em are
conzempts (convicts) andjailbirds; some of'em kills babies in their bodies; some
of'em walks the streets at night—but they's all my children. God got all kinds,
how come I cain't love all of mine" (quoted in Jacobs and Kaslow 1991, 167).
To take one last example, Mrs. Kitamura, the foundress of the new Japanese
religion Tensho-Kotai-Jingu-Kyu, realized her special calling when she became
conscious of a mysterious spiritual force or soul residing in her abdomen. At first
she believed that this strange inner force, which annoyed her and compelled her
to do queer things, was an evil spirit, but it later identified itself to her as the
Absolute God of the Universe. This God is referred to by Mrs. Kitamura and her
followers as "God-in-her-abdomen" (May 1954, 123). In light of our previous
examples, I am not surprised that a woman who "gives birth to" a new religion
does so by discovering the deity residing within her belly.2

Motherhood and Spiritual Intuition—

Spiritualism and Feminist Spirituality
A number of women's religions posit that motherhood enables women to
intuit spiritual truths and reach higher levels of ethical understanding. "Spiritu-
 Maternity and Meaning 77
alists and Women's Rights advocates shared the belief that women's experience
as mothers gave them a heightened sense of values which justified their depar-
ture from tradition. As mothers, they objected to the fate of their children in
the theology preached by men and in the society controlled by men" (Braude
1985, 427).
The contemporary Feminist Spirituality Movement has emphasized the no-
tion that because women are mothers they understand love, relationship, and
spirituality in ways that men do not. One of the most developed feminist
works to deal with motherhood from a religious perspective (or, more pre-
cisely, to deal with religion from the perspective of motherhood) is Kathryn
Rabuzzi's book Motherselfm which she contrasts the mythic, male quest for
selfhood with the "way of the mother." "Through the natural processes of
childbirth women inherently possess means for experiencing the mysteries
associated with the ancient earth mother goddesses such as Demeter. Like the
'dead' seed of grain buried for a season in the ground, the human seed planted
in the woman's body also forms new life. Rather than cause for guilt, this is
reason for joyous wonder. Emergence of life from the womb—whether that of
woman or earth—incorporates the essence of women's mysteries. . . . Child-
birth really means participating in the mysteries of the Goddess" (1988, 200).
Rabuzzi argues that through giving birth women participate in and come to
understand cosmic creation.
Spiritual Feminists explore the relationship between mother goddesses and
real women. (This theme will be examined in more detail in Chapter 8.)
Writings of the Feminist Spirituality Movement claim that human religiosity
originated in the awe felt by early humans contemplating women's powers
of childbirth. "The mysteries of female biology dominated human religious
and artistic thought, as well as social organization, for at least the first
200,000 years of human life on earth. . . . The first human images known to
us ... are magic images of the mysterious power of the female to create life
out of herself, and to sustain it" (Sjoo and Mor 1987, 46). Monica Sjoo and
Barbara Mor imagine what early humans were thinking and feeling when
they designed and celebrated the Goddess. "As we read the powerful magic
signs of the Great Mother's celebration, we can read these first women's
powerful discoveries and celebrations of themselves. The religious beliefs,
the mysteries and rites developed by ancient women, grew organically out of
women's supreme roles as cultural producers, mothers, and prime communi-
cators with the spirit world. The mysteries of creation, transformation, and
recurrence—the primal mysteries of all religions—emerged from women's
direct physical and psychic experiences of these mysteries: in bleeding, in
growing a child, in nursing, in working with fire, in making a pot, in
planting a seed" (1987, 50).
In contrast to white Spiritual Feminists, black Womanist theologians have
emphasized the historical experience of nurturing more than the symbolic
meaning of childbirth. "For the womanist, mothering and nurturing are vitally
important. [Alice] Walker's womanist reality begins with mothers relating to
their children and is characterized by black women (not necessarily bearers of
78 Priestess, Mother, Sacred Sister

children) nurturing great numbers of black people in the liberation struggle"

(Williams 1989, 183).

GRANDMOTHERS—EXPERT
MOTHERS
In many cultures the grandmother role is an extension of the mother role.
Grandmothers, like mothers, nurture and raise the young. This is especially so
in matrifocal societies, where households typically revolve around a senior
woman, her daughters, and their children. While participants in women's
religions are mothers, leaders tend to be grandmothers.
The available literature on women in male-dominated religions suggests two
rather different sorts of reasons why older women are ritually active. Ideologi-
cally, postmenopausal women (unlike younger women) would not be ex-
cluded from sacred places and activities because of menstrual pollution (Gut-
mann 1977). Pragmatically, older women are more likely to have the necessary
free time and autonomy to dedicate to religious leadership (Sered 1992).
Writing about women in Morocco, Susan Davis reports that women are
thought to become expert practitioners of magic as they grow older. Old
women may actively seek contact with the spirit world. Certain old women
are healers and are believed to know magic that helps wives deal with unsatis-
factory husbands. "Not that a woman would ever threaten her husband
openly; but all men know that angry women, and especially angry older
women, are dangerous" (1982, 117). The scenario presented by Davis is a
common one in male-dominated religious cultures—old women, and espe-
cially widows, are suspected of being independent of male control and run the
risk of being accused of witchcraft. In women's religions, on the other hand,
the enhanced religious activity and prestige of older women lies in their supe-
rior knowledge of rituals and theology—erudition that they have accumulated
over the years of their lives. Unlike in Morocco, this female expertise is judged
as beneficent. And unlike in, for example, Jewish societies, this expertise is
institutionalized (see Chapter n).

Ritual Expertise—Black Carib Grandmothers

In Black Carib culture women "plug in" to the religious system as mothers and
grandmothers. "Female responsibility to lineal kin serves as an organizing
principle of Black Carib kinship and ritual, and as the focus of female unity and
collective action. Maternal obligations are primary, broadly defined, and life
long. As mothers, women share common concerns and a valued identity, one
that commands respect (not only from their children). Motherhood connotes
strength, the capacity and duty to protect others. Women act together to
achieve this end" (Kerns 1983, 182).
Black Carib women of all ages exhibit concern for their children's health.
This concern is manifested through ritual and more pragmatic procedures.
Observers have noticed that among the Black Caribs childless women are
 Maternity and Meaning 79

usually less involved in ritual affairs, having fewer resources and less incentive
to take part in ceremonies (Kerns 1983, 77).
For Black Carib women, as for women around the world, infant mortality
has been a major emotional and religious concern (see Chapter 4). Virginia
Kerns reports that some older women in Belize tell of having borne a dozen
children of whom less than half survived to adulthood, and even today many
women lose at least one child (1983, 95). "Their major responsibility as moth-
ers, at any age, is to care for their children and protect them" (Kerns i992b,
97). It is typically grandmothers who have the knowledge and the resources—
both medical and spiritual—to protect children.
Women (as opposed to men, in Black Carib culture) are responsible for
preserving the health of children and grandchildren. Women are in charge of
searching for cures when a child is sick. They are in charge of observing
precautions on behalf of their unborn children while pregnant. A woman's
behavior and her child's well-being are seen as inexorably intertwined. Among
the culturally recognized sources of illness (see Chapter 5) is taboo behavior on
the part of the mother of the afflicted individual. For example, a pregnant or
post-partum woman may endanger her child's health by consorting with a
man other than the father of that child (see Staiano 1981). A woman who does
not fulfill her ritual responsibilities endangers her children's health. Yet (and
here is where we see one of the more distinctive features of Black Carib
culture), female infidelity is not seen as a violation of male property rights, but
rather as a dereliction of maternal duty (Kerns I992b, 99).3
For women, ritual is a matter of maternal and filial duty. From a young age
women teach their children a sense of responsibility and gratitude toward
mothers. In return, Black Carib women represent their children to the ances-
tors at placation rituals. Black Carib culture is a potent example of an ongoing
situation in which motherhood and religion overlap. Yet it is as grandmothers
that the mothering-religion complex emerges most forcefully. Most women
begin to become ritual experts when their own mothers die.
Among the Black Caribs, women of childbearing age must follow certain
restrictions, in particular ones having to do with mobility. The capacity to bear
children carries a social cost. Older women are more mobile and more socially
visible. The loss of fertility is accompanied by the loss of many constraints.
(Kerns 1983, 193). Older women are responsible for enforcing the behavioral
constraints that are incumbent on younger women. In old age women enjoy
increasing autonomy and freedom of movement as they are freed both from
household chores and suspicions of improper sexual conduct. "Immunity from
supernatural harm is said to increase with age" (Kerns i992b, 106). In addition,
older women have access to money sent them by children employed in the
cities. And finally, older women tend to know more about both reproductive
processes and rituals.
Old age is more difficult for men, for whom loss of independence is a
difficult issue. Women have spent their lives developing bonds of interdepen-
dence, and in old age may cash in on some of the rewards of years of nurturing
others. They expect and receive support from grown children and grandchil-
8o Priestess, Mother, Sacred Sister

dren, and many old women manage to accumulate more money than they had
been able to accumulate when younger.
Kerns (1983, 190) found that many older Black Carib women admit that they
once thought little of rituals but have now come to understand their value. The
central ritual constellation of the Black Caribs concerns death and funeral cere-
monies for lineal ancestors. These ceremonies, organized by older women,
protect the health and well-being of their descendants. Older Black Carib
women are the ritual leaders, and they may badger anyone who is reluctant to
sponsor or attend appropriate rituals. It is usually the eldest living woman
relative who organizes the ritual. The organizer collects funds, sets the dates,
informs others, purchases supplies, and oversees preparations. Smaller rituals
are attended only by old women, and old women have informal networks of
reciprocity among themselves in organizing rituals (Kerns 1983, 171-176).
In sum, as Virginia Kerns shows, "Older women, who act together to
enforce . . . rules . . . that support moral and social order, have a perceived
power to protect and preserve, and so to serve the common good" (i992b,
107). It could be said that the religious power of Black Carib women does not
even end with their deaths: Douglas Taylor (1951, 98) notes that the Dominica
Caribs mourn longer for a mother (three years) than for a father (two years).

Family Representatives—East and Southeast Asian

Senior Women
In several female-dominated religions the senior woman of the household or
the lineage is in charge of rituals on behalf of the family. In Okinawan
(Ryukyuan) religion, "Calendrical dates and calendrical fortunes are closely
watched by the senior female in the household, who is usually familiar with the
details concerning each member's personal calendar history" (Lebra 1966, 49).
The senior woman announces forthcoming rites, prepares the ceremonial offer-
ings, places them on the altar, and prays. Similarly, Burmese women make
large ceremonial offerings to their own and their husband's nats (spirits) once a
year. Alternatively, the family may give a contribution to an older female
relative who knows better than they how to make the offerings. The offering
to the nat of the paddy fields is made by the work gang leader, who is usually
the eldest of the women who transplant and harvest the rice (J. Nash 1966).
In Northern Thailand shrines to the phii puu njaa (spirits) are located in the
stem house; that is, in the house where the founding ancestress of the family or
cult group lived. Other branches of the family usually live nearby. The entire
cult group shares a spirit (some ethnographers refer to this as the clan spirit),
which is conceptualized as divided among the individual households of the cult
group. In each generation, one family member—typically the eldest woman—
is responsible for presenting the offerings to the spirits. Within each house-
hold, the eldest woman is responsible for caring for the household spirit. If
members of a cult group move far away from the stem house, they are ex-
pected to build a new shrine at the home of the eldest woman of their group.
 Maternity and Meaning 81
Nature and Nurture
Grandmothers as expert mothers epitomize the theme of nurture rather than
the theme of nature. Grandmothers are experts not because they continue to be
fertile, but because they are specially knowledgeable about how to take care of
children. The Womanist emphasis on the social aspects of motherhood is more
in line with the majority of women's religions than is the Spiritual Feminist
emphasis on the mysteries of the biological process of birthing (see above).
Female-dominated religions offer women conceptions of motherhood that
are not merely biological. Female-dominated rituals spiritualize motherhood
and proclaim that motherhood has social value. It is not only the biological
ability to give birth that empowers women in female-dominated religions, and
birth is rarely glorified or even especially ritualized. Female-dominated reli-
gions honor women less in their role as birthers than in their role as nurturers. I
see it as significant that the one women's religion to ceremonialize the physi-
cality of birthing—Sande—in fact socializes fertility and birth. In women's
religions mothers are powerful beings whose social bonds with their children
give them ritual and spiritual strength. It is instructive that Shakers—a
women's religion that requires celibacy—still use motherhood imagery in talk-
ing about spiritual progress.

THE MEANING AND IMPLICATIONS

OF MOTHERHOOD
In this chapter we have begun to look at the implications of women's identity
as "mother." Unlike in male-dominated religions, in women's religions the
content of that identity is described from an adult and gynocentric perspective.
In subsequent chapters we will see the myriad ways in which motherhood
affects women's religious choices. But before we go on, it is necessary to
clarify what I mean by motherhood. The meaning of motherhood is not as
indisputable as it seems at first glance. Surrogacy cases have surely shown us
that motherhood, like gender, is culturally constructed.4
There are two overlapping yet distinct aspects to motherhood: physiological
and social. The physiological aspects include conception, pregnancy, child-
birth, lactation, and maternal mortality. A central question in this book is
whether these physiological processes receive particular attention in women's
religions. The discussion of motherhood in this chapter does not indicate that
this is the case.s
The social aspects of motherhood consist of the activities, rights, responsibili-
ties, relationships, and social statuses that make up motherhood in particular
cultural contexts. Because the social content of motherhood is infinitely vari-
able, I would expect to find that women's religions relate to an assortment of
social aspects of motherhood. I would not expect to find motherhood treated
identically in all women's religions, and the evidence we have presented thus
far indeed supports that assumption.
On the other hand, I would expect to uncover certain similarities, rooted in
82 Priestess, Mother, Sacred Sister
the fact that social arrangements lead women in most cultures to be more
involved in childcare than men. Nancy Chodorow (1974; 1978) has explored
the social-psychological implications of women's greater responsibility for
childcare. She argues that girls develop their sense of self and gender identity in
the context of ongoing, intimate, daily contact with their mothers. Girls learn
not only what their mothers do, but what their mothers are, how their moth-
ers feel, and how their mothers react to a multitude of situations. Boys are also
raised primarily by mothers. This means that boys develop their sense of self
and gender identity in a situation in which the male role model is often absent.
The growing boy may know what his father does (or he may only have a
vague sense of what his father does)—his father hunts, plows, directs traffic,
assembles machinery parts—but he doesn't know what his father is. Raised by
women, the growing boy at some point realizes that he is not female but he
does not really know what it is to be male. He comes to define maleness as not-
femaleness, and to see femaleness as something to be rejected (inferior, pol-
luted, etc.). Unlike his sister, he develops his sense of self in a negative way.
The ramifications of Chodorow's argument are twofold. First, it explains
why in so many cultures women and female pursuits are considered less good,
less noble, or less pure than men and male pursuits. And second, it explains
why women seem to have an easier time with interpersonal relationships than
men do. In Chapters 6 and 10 I discuss these matters as reflected in women's
religions.
While I believe that Chodorow's theory is on the whole valid, it situates the
roots of "culture" entirely in the preoedipal experiences of sons and daughters. 6
By highlighting the role of motherhood in shaping adult women's religiosity, I
am implicitly critiquing psychological and psychoanalytic theories that situate
religion in infantile experiences. Recent studies have clarified the numerous
ways in which pregnancy and motherhood are growth experiences for women.
Since certain of these findings have important implications for women's reli-
gious lives, I will briefly review two of the more interesting studies.
In a fascinating inquiry into the psychological dimensions of pregnancy,
Myra Leifer has demonstrated that the personality does not become fixed by
adolescence, but continues to change and grow throughout the life cycle.
Leifer followed a group of women throughout their first pregnancies and
post-partum period. "From a psychological perspective . . . the emotional
turbulence of pregnancy may be a positive phenomenon when viewed within
a developmental perspective" (1980, 42). According to Leifer, the intense
emotional changes during pregnancy reflect significant reorganizations in per-
sonality as a woman moves toward the developmental stage of parenthood.
Pregnant women appeared to be more open to their inner experiences than
other women. "Looking back on their pregnancies, almost all the women felt
that despite the difficulties and strains, pregnancy had been a unique and
valuable experience and that they had changed considerably in the course of
these nine months" (Leifer 1980, 25). Although Leifer does not raise the issue,
it seems to me that there are religious ramifications to her discovery of the
 Maternity and Meaning 83
inner orientation of women during pregnancy; pregnancy is a time of concen-
trated emotional and spiritual growth for many women.
Mary Hales (1990) investigated mothers who were parenting second or
subsequent children, and found that second-time parenting brought about
changes in many of the mothers. Some of the more common changes included
more assertiveness in getting their husbands to share in homemaking, in-
creased inner-directedness and greater control over children, an enhanced sense
of competence, achievement and endurance, an increased appreciation of their
own rights and needs, and a deepening appreciation of femaleness. Again,
although Hale did not write about religion, I suspect that in the religious
domain these characteristics translate into leadership ability (see Chapter u), a
sense that one does not have to passively submit to fate (see Chapter 5), and a
recognition of the power inherent in femaleness (see Chapter 10). The evidence
from women's religions strengthens both Leifer's and Hale's understandings of
intra-psychic processes of motherhood.
A useful complement to these ideas can be found in Sara Ruddick's writings
regarding "maternal thinking." Ruddick begins by pointing out that because
most mothers are young women, their adult psyches are strongly shaped by
the maternal practices in which they engage (1983, 236). Ruddick's work gives
specific and explicit consideration to how the experience of mothering affects
the ways in which mothers (that is, most adult women) think. Although her
writings make no pretense at cross-cultural applicability, I have found it useful
to "test" her ideas against what I have learned about women's religions around
the world.
First, according to Ruddick, mothers are concerned with preserving the lives
of their children. The theme of responsibility for the health and lives of chil-
dren is indeed central in women's religions. More specifically, "Faced with the
fragility of the lives it seeks to preserve, maternal thinking . . . implies a
profound sense of the limits of one's actions and of the unpredictability of the
consequences of one's work" (1982, 80-81). Somewhat surprisingly, the theo-
logical implications of this statement are not borne out by the data on women's
religions. Contrary to Ruddick's expectations, women's religions teach that
individuals have a great deal of control over their own and their children's
destinies. On the whole, they do not posit capricious and all-powerful deities.
Instead, they provide many different rituals designed to ensure children's well-
being (See Chapter 8.)
Second, mothers not only preserve life but also foster growth and welcome
change. A mother realizes that her child continuously grows and changes,
and "her realistic appreciation of a person's continuous mental life allows a
mother to expect change, to change with change" (Ruddick 1982, 82). In line
with Ruddick's expectations, we in fact find that women's religions tend to
preach flexible ideologies, ideologies that recognize and sacralize change (see
Chapter 12).
Third, according to Ruddick, mothers try to raise children who will be
acceptable to the next generation. Mothers take on the values of the subcul-
84 Priestess, Mother, Sacred Sister
tures to which they belong and of the men with whom they are allied. One
of the most interesting patterns in women's religions is the tendency to accept
the dominant patriarchal ideology of the culture in which the religion is
located (while reinterpreting that ideology in ways that empower women—
see Chapter 10).
Fourth, women on the whole remain peaceful in situations in which they are
powerful, namely, in battles with their own and other children. Ruddick
writes, "I can think of no other situation in which someone with the resent-
ments of social powerlessness, under enormous pressures of time and anger,
faces a recalcitrant but helpless combatant with so much restraint. It is clear
that violence—techniques of struggle that damage—is by definition inimical to
the interests of maternal work" (1983, 243).' We find that women's religions
indeed praise harmonious interpersonal relationships, and offer women help in
coping with the emotional conflicts created by the situation Ruddick describes.
Many female-dominated religions offer women concrete assistance that amelio-
rates conflict and powerlessness (see Chapter 13). Furthermore, forceful mis-
sionizing is not a component of any known women's religion. This finding
certainly resembles Ruddick's realization that mothers rarely use violent force
against their children, even though the physical possibility (and the tempta-
tion) surely exist.
Fifth, Ruddick believes that women are more attracted to concrete thinking
and men to abstract thinking.8 For Ruddick, abstraction refers to the disposi-
tion to simplify, dissociate, generalize, and sharply define. Its opposite—
concreteness—respects complexity, connection, particularity, and ambiguity.
Ruddick roots women's concreteness in the experience of motherhood. "Con-
creteness can be seen as a mix of interwoven responses to a growing, changing
child. ... A mother attends to a particular child and understands her as best
she can on a given day, tolerating both the ambiguity of the child's actions and
the tentativeness of her own interpretations. She will eschew generalization,
not only because children are very particular beings to whom she attends, but
also because they confound prediction. ... In short, her thinking will be
'holistic,' 'field-dependent,' 'open-ended,' not because of any innate sex differ-
ence, but because that is the kind of thinking her work calls for" (1983, 249).
(We will return to this notion in Chapters 7 and 12.) It does seem to be the case
that the belief systems and moral frameworks of women's religions are contex-
tual and "field dependent." These religions do exhibit a high tolerance for
ambiguity. They do focus ritual and thought on specific individuals rather than
on generalized categories. And they most certainly bear in mind the "moral
and human significance" of any plan or doctrine (1983, 250).
And finally, Ruddick finds that women are more concerned with sharing
responsibility than with claiming rights. One of the clearest patterns to emerge
in this book is that women's religions are consistently more attentive to inter-
personal relationships than to rules, valuing caring more than abstract justice
(see Chapters 6 and 7).
Motherhood has received remarkably little attention from theologians, and
the hypotheses I have laid out in the preceding paragraphs grow out of social
 Maternity and Meaning 85

psychological literature, not out of theological discourse. In a rare article by a

theologian, Bonnie Miller-McLemore (1992) draws attention to one more rele-
vant aspect of motherhood—mothers typically lack the space, time, energy,
money, and solitude to articulate and record their ideas. Realizing this leads us
to surmise that elaborately written theological tracts will not be a feature of
most women's religions. In Chapter 12 we will see that this indeed is the case.
Empathy is another consequence of motherhood considered by Miller-
McLemore. Mothers' enhanced empathy has two somewhat discrete anteced-
ents. First, "Ultimately, to lactate when another thirsts teaches a certain
empathic, connected knowing" (1992, 242). And second, "Somehow, through
the mutual understanding learned and practiced over and over in the intense
moments of attachment with a little, developing person, one who has truly
cared for a child gains new modes of relating and new empathy for others—
parents, other children, one's spouse, the oppressed" (1992, 246). If Miller-
McLemore's impressions are correct, we should expect to find that women's
religions offer institutionalized support for all categories of weak and op-
pressed people, and encourage members to develop empathy. In fact, although
most women's religions encourage members to preserve excellent relations
with their families and communities, few of the religions urge members to
devote time or money to helping unknown unfortunate individuals.
Motherhood—culturally constructed in various shapes—does not impact
upon religiosity in a uniform manner. What is present in all the women's
religions is direct, emphatic, and serious addressing of the diverse implications
of motherhood. Yet I wish to clarify once again that when I speak of mother-
hood as interacting with women's religious lives, I am building on evidence
that points to the social far more than the biological aspects of motherhood.
Recent research by psychologists and sociologists indicates that the parenting
role shapes both men's and women's characters in rather similar ways, regard-
less of men's and women's very different biological contributions to parent-
hood. In a fascinating study, Barbara Risman (1987) compared the personality
traits of single mothers, single fathers, and married parents. She found that
responsibility for childcare was strongly related to such "typically" female
traits as nurturance and sympathy in both men and women. In other words,
single fathers with primary responsibility for childcare were more like mothers
than like married fathers. 9
In this chapter we have looked at a wide range of religious ramifications of
motherhood. We have seen symbolic and practical aspects of motherhood
reflected and dramatized in women's religions. Of all of these many facets,
there is one that so dominates discussions of women's religions as to warrant a
separate and detailed analysis. That facet is child death, which we will look at
in Chapter 4.

Notes
I. Ann Lee's title is "Mother," but, according to Jean Humcz, anecdotes portraying
Lee's relationship to her followers do not present a picture of a warm, nurturing,
86 Priestess, Mother, Sacred Sister
maternal figure. To the contrary, Lee is most commonly portrayed as stern, authorita-
tive (yet devoted), and even bordering "on the sadistic" (1992, 94). Reconciling the title
with the descriptions, Humez argues that, "As a pious, working-class Englishwoman
of her day, the historic Lee would likely have held what to modern minds would seem
an unacceptably severe attitude toward children. The more sentimental ideal of all-
loving motherhood with which we moderns are so familiar had simply not yet been
invented" (1992, 95).
2. Many Japanese believe that the soul area is the abdomen (rather than the chest or
the head as many Westerners believe). I find it interesting that Mrs. Kitamura taught
women to have their babies naturally, that is, without the aid of physicians or midwives
(May 1954, 131). While on the one hand this advice is difficult or even dangerous for
some women, for other women it leads to a strong and positive self-image of them-
selves as mothers.
3. According to Kerns, given the rather fluid nature of heterosexual relationships in
Black Carib society, publicly celebrated female fidelity is a potent means to ensure
paternal acknowledgment, which in turn assures "a child a full set of kin . . . and an
unambiguous social identity" (i992b, 99).
4. In the Baby M case, for example, U.S. courts declared the wife of the biological
father, rather than the biological mother, to be the baby's legal mother (Chesler 1989).
5. Until recent decades in the West (and still today in many parts of the world),
childbirth has been a direct confrontation with death and large percentages of mothers
died giving birth or in the immediate post-partum period. I have not found that
women's religions address this issue head on. On the other hand, the central role of
healing in women's religions (see Chapter 5) may be assumed to reflect concern with
maternal mortality.
6. A theory as all-encompassing as Chodorow's is bound to draw criticism. Some
critics have contended that Chodorow's work is methodologically problematic because
it is based on clinical practice that relies on verbal recall, effectively ruling out the
possibility for the therapist to access memories of the first year of the patient's life. Yet it
is precisely these early months that are critical to Chodorow's theories (Rossi 1981,
495). Moreover, the patients who supply the psychoanalyst's "clinical evidence" differ
in many respects from the general population. Since Chodorow's research deals with
the development of healthy adults and not psychiatric patients, the use of clinical
illustrations for her evidence raises difficulties.
Chodorow does not ask why it is that all known societies have found it useful or at
least not harmful to have men and women "reproduced" in this manner (Laub Coser
1981, 488). Further objections to her ideas concern the historical and cross-cultural
applicability of a model that situates fathers out of the home and mothers as the primary
caretakers of young children. Unfortunately, Chodorow chose not to examine social
structural conditions causing women to become the primary childrearers (Bart 1983,
150). Tavris (1992) has suggested that the traits Chodorow associates with women are in
fact traits shared by all subordinates, whether male or female. A number of critics have
drawn attention to Chodorow's lack of attention to factors of class, race, and ethnicity
(Lorber 1981, 483, Bart 1983, 150-151, Rossi 1981, 493-494).
Another, perhaps crucial, objection is that gender may not be as significant a part of
an individual's identity and sense of self as Chodorow claims. Psychological studies in
cognition, behavior, and even emotion consistently show men and women to be more
similar than different (see Hyde 1990).
Chodorow has tried to correct some of these deficiencies in her later work
(Chodorow 1989). In any case, the quantity of attention (both critical and laudatory)
 Maternity a 87

paid to Chodorow's work is a tribute to the importance of her ideas. In this book, I have
tried to apply Chodorow's ideas cross-culturally, outside of the psychoanalyst's couch,
and have found most of her ideas useful in helping me understand women's religions.
7. In contrast to Ruddick's ideas, Macaulay argues that there are "far too many
studies with no gender differences to support any theory that strong biological factors,
deeply ingrained personality traits, or well-learned gender roles make even a majority
of women reliably or consistently very much less aggressive than men" (1985, 192).
8. I am not convinced that Ruddick is correct in this assertion. It is unproven and
probably unprovable that cross-culturally, or universally, women and men have differ-
ent cognitive styles. Regarding all of Ruddick's ideas, one should bear in mind that she
does not bring empirical, scientific evidence as proof. Ruddick is more of a philosopher
than a scientist, and tends to exaggerate the differences between men and women.
9. The single fathers in her study had custody of their children through circum-
stances beyond their control (e.g., widowhood); they were not a self-selected group of
"feminist" or "feminine" men.
 4
When Children Die

Young children are especially vulnerable to illnesses and accidents, and in many
cultures mothers are not provided with adequate resources to safely birth and
raise children. Cultural decisions regarding allocation of resources may lead to
high maternal and infant mortality rates. And even in the best of circum-
stances, all mothers know that their children's health and well-being are not
guaranteed. (Even we middle-class Westerners never know when a drunken
driver will run over our child while walking home from school.) Biology—
carrying a baby in one's womb, childbirth, and lactation—combines with
social structure—prolonged and intensive childcare, to ensure that most
women love their children and are vitally concerned with their children's
health and well-being.1 Most adult women devote a great amount of time and
energy to childcare, and women's emotional ups and downs tend to be related
primarily to family happenings, while men's tend to be connected to work (cf.
Hagestad 1984, 38, on American women). Not surprisingly, women's reli-
gious lives often reflect occurrences within the family. In situations where
male-dominated religions do not offer women adequate help with their prob-
lems and worries as mothers, female-dominated religions that address the
ramifications of mother-love and child death attract many women.
We Westerners find it comfortable to believe that Third World women do
not love their children as much as we do; that African and Asian women do
not care as much as we do when their children suffer and die (cf. Dally 1982,
esp. pp. 28—29). While anthropology of emotion is a relatively undeveloped
field, I see no evidence in the ethnographic record to support such a belief.
The literature certainly shows that both abortion and child death take an
emotional toll on women regardless of their socioeconomic status. On the
other hand, I would not assert that all mothers feel the same toward their
children. The model of one-on-one nurture we know in the West is far from
universal, and it is likely that in cultures in which the vast majority of
children die before the age of five parents develop different modes of relating
to their children than in cultures in which the vast majority of children
survive. "Different modes" could include both reluctance to becoming overly
90 Priestess, Mother, Sacred Sister

emotionally involved with the baby and heroic measures to ensure the child's
survival.
One might claim that infanticide is evidence that mother-love as we know it
in the West is not universal. Yet I insist that not even infanticide should
automatically be taken as evidence of lack of maternal feeling. To the contrary,
infanticide is often practiced by women who realize that they will not be able
to take care of their other young children if the new baby lives. And I suspect
that in cultures in which women practice infanticide, women's experience of
infanticide is reflected in their religious lives.
In this chapter I focus on the religious implications of child death. My
argument is not that child death leads women to "escape" into religion. My
argument is that child death encourages women to ponder existential and
theological questions. As part of that process, some women change their reli-
gious beliefs and affiliations. 2 In sum, although Western-style mother-love is
not universal, child death is something with which women in all cultures must
grapple, albeit in different ways. Encouraging motherhood, rejecting mother-
hood, and the loss of motherhood are important to women, and all of these
themes surface in women's religions.

MATERNAL RESPONSIBILITY
Women's religions plug into the mother-love and child death dynamic in a
number of ways: They offer consolation to grieving mothers (Spiritualism),
they provide means of controlling fertility (Sande, Shakers), they ensure that
mothers receive support from their families (Thai matrilineal spirit cults, Black
Carib religion), and they encourage women to organize and demand social
CHANGE
begin with several brief examples that show some of the ways in which
women infuse their responsibility for their children with a religious dimension.
Burmese women are bothered by men's skepticism concerning nats, and
urge men to stop making comments that will offend nats and so cause the nats
to harm their families. One informant explained to anthropologist Melford
Spiro that she as a woman needs to be more concerned with nats because she is
responsible for her children—if the nats are angered they will cause her children
to fall ill. "All she asks is that they not harm her children; so let him (her
husband) keep his peace" (1967, 59).
The infant mortality rate among the Luvale tribe of Zambia is especially
high—between 20% and 33.3% of all children die within the first year of life.
The Luvale believe that ancestors send illness and death to children whose
mothers have not behaved properly, either in terms of social behavior or
ancestral devotion. According to Anita Spring (1978), this belief serves to
institutionalize the stronger grief reactions experienced by bereaved mothers as
opposed to bereaved fathers. Fathers never undergo spirit possession on behalf
of their sick children. Since the spirits who afflict Luvale women belong to
their matrilineages, we can see a multigenerational maternal responsibility for
child health and illness.
 When Children Die 91
Kaja Finkler has studied the attraction that Spiritualism holds for lower-class,
rural Mexican women (see Chapter 5 for more on Mexican Spiritualism). One
typical devotee, Lupe, began to suffer from vague chronic illnesses shortly after
her marriage at age fifteen. In interviews with Finkler, she attributed her illness
to the difficult life she had with her husband who beat her, and to her mother-in-
law. "The first attack [of illness] she experienced coincided with the death of her
firstborn infant at four months of age. She held herself responsible for its
death. . . . During one irradiation [Spiritualist treatment], God called Lupe up
to stand before Him. He told her that the baby's death was not her fault but
rather that it was His will, that she was being tested, and that it was her duty to
serve Him in His house" (1985^ 29ff). Lupe's biography is especially signifi-
cant in light of Finkler's finding that the vast majority of Spiritualist "temple
patients—especially regulars—have suffered the death of a child, whereas only
three out of nineteen controls [people from the same village who do not attend a
temple] have known a similar loss" (igSsb, 72). Death of a child is the single
most common attributing event offered by temple patients to explain their
illnesses, but only among women patients. "None of the men [at the Temple]
related their symptoms to the death of a child, whereas over 24% of the women
did, even though in at least one instance a man had lost four infants" (19853, 39).
Finkler explains women's greater adverse reaction to child death in terms of the
fact that a rigid sexual division of labor means that successful mothering is a basis
for women's self-esteem. Both the bereaved mother herself and her husband
hold the women culpable for the tragedy.

Brazil—A Case Study

Matrifocality means that mothers (and not fathers) are seen as ultimately re-
sponsible for children's health and well-being. In urban Brazil the typical
pattern of plural mating—serial common-law marriages in which one man
visits but rarely supports a number of households consisting of women and
children—strengthens the perception that it is women and not men who are
accountable for the survival of children.
Many poor urban mothers lack the resources to adequately care for their
babies, and infant mortality rates are appallingly high. The Catholic Church in
Brazil, dedicated to liberation theology, no longer provides bereaved mothers
with spiritual scripts that help them make sense out of the loss of their children.
"Traditionally, the local Catholic church taught . . . [that] . . . if an infant died
suddenly, it was because a particular saint had claimed the child. . . . The
infant funeral was, in the past, an event celebrated with joy. . . . The new
theology of liberation imagines a kingdom of God on earth based on justice
and equality, a world without hunger, sickness, or childhood mortality. At the
same time, the church has not changed its official position on sexuality and
reproduction, including its sanctions against birth control, abortion, and
sterilization" (Scheper-Hughes 1989, 16).
Worry about their ability to care for their children is often what draws
women as clients and as mediums to the Afro-Brazilian cults. Two of the three
92 Priestess, Mother, Sacred Sister

women mediums interviewed by Esther Pressel reported having lost a child

shortly before joining the cult.4 One of the mediums became involved with the
spirits after her young child died; another after she lost four children. Pressel
records in detail the biography of a medium whose first spiritual experience
was at age eight. She moved from a small town to a city at age seventeen,
married and bore three children. Her older daughter died and her boy was
killed in a car accident the following year. Her husband began to act strangely
and lost his money; he had a mistress. A friend suggested that she go to an
Umbanda center for help. A spirit there told her she needed to be a medium
because her dead mother had a mission to fulfill through her.
This medium's story incorporates a number of common patterns in women's
religious life stories: child death, a history of suffering because of the double
sexual standard of patriarchal culture, the role of her mother in her decision to
become a medium, the "matrilineal" character of her mediumship, and the long
and gradual process of becoming a medium (see Chapter u on leadership).

RESPONSES TO CHILD DEATH—

NINETEENTH-CENTURY UNITED STATES
The United States in the nineteenth century presents an excellent case study of
the connection between motherhood and infant death and women's involve-
ment in religion. It is also one of the best-documented examples of the cultural
construction of mother-love and the intersection of mother-love and religion.
To understand what happened in the nineteenth century, it is necessary to go
back to the colonial period. During the colonial period, friends and kin shared
in childcare, and Divine Providence was the explanation for infant death.
Seventeenth- and early eighteenth-century motherhood meant almost continu-
ous illness and death of children. Parents loved their children yet did not expect
that caring for them properly would cure illness, avert accident, or prevent
death. Children were God's temporary gift to parents. Women's diaries and
letters display feelings of powerlessness over infant and child health and safety
(Dye and Smith 1986).
During the second half of the eighteenth century a move began toward more
affectionate, nuclear families in which children became the focus of indulgent
attention and mothers became defined as the moral and physical guardians of
their children. By the early iSoos the mother was seen as the most important
force shaping and preserving a child's life, replacing God as the one responsible
for a child's welfare, and replacing the network of kin and friends who had
formerly looked after children. During the nineteenth century ministers, educa-
tors, physicians, and social commentators built up the notion that motherhood
is a full-time job for which women are especially fitted by nature. The increas-
ingly private and isolated nature of the family in the nineteenth century encour-
aged very strong mother-infant bonds. Although American women had always
been more involved in daily childcare than had men, in the early nineteenth
century changes in the organization of the labor force tended to increase their
share even more (Dye and Smith 1986).
 When Children Die 93
In the wake of technological advances, American families began to believe in
the responsibility and power of humans to ensure children's health. At the
same time, the infant mortality rate stayed high (it even may have risen due to
urbanization). These circumstances created anxiety for many nineteenth-
century women. Nineteenth-century mothers (unlike eighteenth-century mo-
thers) recorded their children's idiosyncracies and developmental milestones.
"But beneath this delight lay the ever-present fear that their children might die.
In the first days and weeks after giving birth, many mothers appear to have
been frightened by the intensity of their feelings for infants whose lives seemed
so fragile" (Dye and Smith 1986, 340). By the late nineteenth century women
more and more believed that good mothering should and could ensure a baby's
survival, and mothers worried incessantly about their children. Nancy Dye
and Daniel Smith, writing about the nineteenth century argue that, "Living
beyond the time of unquestioned confidence in the providential power of an
omnipotent God and before that of public health advances and reliable medical
help, women experienced motherhood in solitude—sole possessors of all the
delights and fears that come with raising children" (1986, 346).5
"Death literally occurred in woman's sphere. Most people died at home in
bed, attended by female relatives. In New York City in 1853, 49 percent of
those who died were children under five, who presumably had been under
primarily female care. Women, who were expected to focus their lives on the
nurture of family members during life, were also expected to feel losses
through death more deeply than men, who might turn their attention to other
duties. Middle-class women remained at home by empty cribs and unoccupied
seats at the dining room table and produced and purchased a variety of memo-
rial artifacts, from postmortem photographs to jewelry woven from the hair of
the deceased. Etiquette prescribed longer periods of mourning for women than
for men" (Braude 1989, 52).
Nineteenth-century evangelical revivals stressed the importance of early reli-
gious and moral training in attaining salvation. This placed an even heavier
burden on mothers: If a child were damned, it was not solely because of
predestination, but also because his or her mother failed in her duty. 6 Through-
out the eighteenth and nineteenth centuries women's diaries and letters suggest
that at least some women were dissatisfied with the mainstream Calvinist
doctrine that proclaimed that dead babies, like all individuals who have not
undergone an authentic conversion experience, are denied entry into heaven.
Three of the female-dominated religions available to nineteenth-century
American women addressed the issue of infant mortality head on, yet came up
with rather different answers. Spiritualism told women: Your dead children are
happy in a good place, you will certainly be reunited with them when you die,
and you can even communicate with them now through seances. Shakerism
told women: If you don't want to grieve when your children are sick and die, if
you don't want to see your beloved children suffer, become celibate—stop
having children. And Christian Science told women: Matter is illusory, suffer-
ing is illusory, and by joining with us we can heal you and your children of
pain and disease.7
94 Priestess, Mother, Sacred Sister

Spiritualism
Against this background of high child mortality rates, maternal love and re-
sponsibility, and Calvinist theology, we can trace the history of a rather typical
nineteenth century middle-class family who became involved in Spiritualism.
Ann Braude tells about an American woman, Annie Denton Cridge, who was
griefstricken after losing her first child as an infant. Yet her grief was assuaged
by the fact that during his final moments she saw the spirits of her own dead
parents above his bed "waiting to bear his sweet spirit away." She saw his spirit
withdraw from his body and "•with the help of his grandparents assume a
spiritual body. Since then, she held her child in her arms everyday. He weighed
nothing and within a week had recovered from the illness that took his life"
(1989, i).
Alex Owen relates the history of a British Spiritualist family of the same era:
Five children died, some in infancy and some in childhood, and one child was
stillborn. Of eleven pregnancies only four children survived. "Gradually a
trusting relationship developed between the earthly family and its spirit mem-
bers, the spirit children evolved into separate and distinct personalities, and
their communications became increasingly serious and meaningful" (1981, 82).
During seances a child who had died a number of years earlier informed the
family of the welcome that the newly dead baby received in Summerland. This
must have brought much needed comfort to the bereaved parents, and espe-
cially the mother. "Through Louisa the Theobalds learned, to their great joy,
that their babies were safe and continued to grow and develop in the spirit
land. . . . When baby Percival died in 1870 at the age of six-and-a-half months,
the spirit children were anxious to emphasize the suffering the convulsive
'tubercles of the brain' had caused him and the improvement already registered
in his new life" (1981, 84). The spirits recognized that parents felt guilty
because of the death of children (If only I had/had not . . . ), and the spirits
made efforts to tell the parents that they were not to blame. An excellent
example of the manner in which Spiritualism related to child death can be
found in the Funeral Service for Children (Twentieth Century Formulary of Songs
and Forms by W. C. Bowman [1907], cited in Ward [1990]): "Tis vain to bid the
sorrowing heart dismiss its grief; but to those who mourn so deeply and so
sadly o'er this coffin we may offer the sure consolation that 'it is well with the
child.' "
"The more a woman's identity derived from her family relations, the more
she must have been devastated by the irreversible separation from family mem-
bers at death. Although many nineteenth century Christians leaned toward the
Spiritualist position that individuals retained their personal identity after death,
Spiritualism went further than any other religion in promising to prevent the
severance of family relations. Spirit messengers painted rosy pictures of life
after death, in which reunited families picnicked by babbling brooks. . . . The
possibility of communicating with lost children was a great appeal of Spiritual-
ism to bereaved mothers" (Braude 1985, 427). Ann Braude quotes from one
woman comforting another on the death of a baby: "But one thing I know,
 When Children Die 95
that in the other world your baby and mine will know us—their mothers, else
God were not God" (1989, 54)."

Shakers
Shaker leader Ann Lee was born to a working-class family in Manchester,
England, in 1736. She was the second of eight children, and her mother, like
many women of her day, died in childbirth. Ann was sent to work in a cotton
factory as a child, at age fourteen or fifteen she became involved in religion,
and at twenty-two joined a Shaker sect. She was pressured by her family to
marry (against her will and her religious beliefs) and bore four children, all of
whom died in infancy or early childhood. Ann Lee was not so different from
other women of her time: High birth rates and high infant mortality rates in
eighteenth-century urban England (possibly as high as 68.6%, see Dally 31982,
25-26) meant that women spent most of their adult years pregnant and mourn-
ing. Unlike most women of her time, however, Ann Lee interpreted the deaths
of her children as punishment for her having engaged in sexual relations with
her husband. At age thirty-four she had a vision that led her to assume leader-
ship of the Shakers. The essence of the revelation was the necessity for celibacy
in order to follow Christ.
For the Shakers, celibacy was a solution to the pain and suffering women
endure as mothers.9 Shaker celibacy freed women from the hideous threat of
death in birth. "If she [Ann Lee] repudiated marriage because it exploited
women, she had only two logical choices sexually: chastity or promiscuity. In
an era without effective birth control, the choice of the latter would have
condemned women to suffering and death" (Klein 1979, 367). It is highly
significant that although Shaker women in general outnumbered Shaker
men,10 the sex imbalance was particularly large in the childbearing age group.
Women aged twenty to forty-four outnumbered men two to one (versus the
ratio of three to two for other age groups) in nineteenth-century Shaker com-
munities (Kitch 1989, 202). In addition, many Shaker converts were mothers
with small children, suggesting that the Shaker life-style was particularly com-
pelling for women struggling with raising children in a society not overly
conducive to the needs of mothers and children (Kitch 1989, 202).
The Shakers were clear in their wish not to propagate for the sake of provid-
ing factories and armies with new workers and soldiers. In later Shaker writ-
ings a Malthusian theme appears: The unrestrained generation of the human
species will overpopulate the earth; voluntary celibacy is the way to ensure
adequate space and resources for earth's creatures. The Shaker view of history
led them to "conclude that it was necessary to remake society on the basis of a
new type of human relationship . . . since the economic evils were based on
'biological' evils, only a new kind of biological relation would make it possible
for them to refashion society and its economic structure" (Desroche 1971, 39).
Shakerism grew out of the Christian world view in which the sinful act of
the Fall in Genesis was sexual, and marriage was an imperfect solution for the
96 Priestess, Mother, Sacred Sister

basest human instincts. Shaker writings argue that the marital (sexual) relation-
ship violates the laws of God and nature; those who live by the flesh would
suffer in the world to come. The Shakers envisioned a new type of relationship
between men and women: Instead of husband and wife—an unegalitarian
relationship in which women are exploited—Shaker men and women lived as
brothers and sisters. The Shakers themselves, unlike many other millennarian
groups, believed that the millennium had already arrived, that Ann Lee was the
second coming of Christ. Their celibacy was understood as part of the process
of human divinization in Christ (Whitson 1983, 156).
"Shakers are convinced that a free sharing in both temporal and spiritual gifts
is not really possible except among the celibate simply because the married
have a properly exclusive concern for each other and their children. . . . While
other attempts at communitarianism quickly failed (Owenism, Harmonism,
Oneida Perfectionism, Amana, and many more) largely because of conflicts in
family/children values, the Shaker communes survived the better part of two
centuries" (Whitson 1983, 158). Celibacy allowed a fuller commitment to
nonviolence; without an obligation to defend children or spouse, Shakers were
free to avoid violence even in instances where non-Shakers felt forced to use
violence to protect their children. 11
It is of interest to briefly refer to another communal, Christian, celibate,
female dominated religious group that emerged in Texas during the nineteenth
century—the Sanctificationists, or Women's Commonwealth. We need feel no
surprise to learn that Martha McWhirter, the founder of the Sanctificationists,
received a vision instructing her to embrace celibacy shortly after the deaths of
her brother and two of her children (Kitch 1989, 49). Unlike most other
women's religions, however, the Sanctificationists did not employ metaphors
of spiritual motherhood, nor did they encourage special bonds between moth-
ers and children. Members shared in childcare and nurturing responsibilities. It
is likely that at a time when many mothers lost one or more child, diffusing
maternal affection served women's interests well.

DEAD CHILDREN AS TUTELARY

SPIRITS—THE CASE OF CANTONESE
SHAMANISM
Among Cantonese Chinese there are women shamans known as mann seag
phox—"old ladies who speak to spirits." According to anthropologist Jack
Potter's description (1974), these female religious specialists deal solely and
expertly with interpersonal relationships, among the living and between the
living and the dead. They intercede between the villagers and the supernatural
world by sending their own souls to the supernatural world where they com-
municate with deceased members of village families. They also know how to
predict the future and to recapture the kidnapped souls of sick village children.
In particular, they care for the souls of girls who die before marriage, and they
protect the life and health of village children by serving as fictive mothers.
Potter describes one ritual: Speaking through a shaman, a deceased first wife
appeared at a gathering in which the second wife was present. The ghost
 When Children Die 97
assured the second wife that she would not bother her, that she was visiting
because she was lonely. She spoke with several women in the audience, ex-
pressed anxiety about her children, and admonished the second wife to take
good care of them. This anecdote is a good example both of a religious re-
sponse to women's fears that death •will separate them from their children, and
of the sacralization of maternal nurturing: The mother in this anecdote contin-
ues to voice concern for her child even after her own death.
The three mann seag phox whom Potter studied became shamans after severe
crises, including deaths of children. One shaman had lost five daughters and
two sons. The second lost three daughters and one son. Each was visited in her
dreams by her children's spirits who urged her to become a shaman. "Deceased
children, who mediate between their mother and the supernatural world, are
essential to a career as a spirit medium" (Potter 1974, 226). The parallels
between Cantonese shamanism and American and British Spiritualism are
instructive. One of the most interesting aspects of Spiritualism is the role of
dead children as spirit guides (as intermediaries between their mothers and
other spirits). Both in the case of Spiritualism and in the case of Cantonese
shamanism, women's religions provide solace to bereaved mothers and trans-
form the very fact of their bereavement into spiritual power.
Of particular interest is Potter's description of how one of the shamans
works—she always begins by calling the spirits of her dead children "because
she is powerless without their help" (1974, 220). The dead children are the
intermediaries through whom she contacts the more powerful deities on her
altar. Sometimes the children's spirits refuse to enter their mother's body, in
which case she is powerless. The spirits may be very young and prefer to go off
and play. They may also be uncooperative when they feel slighted. The cere-
mony ends when the shaman scatters rice around to feed her tutelary spirits
(her children) and gives her clients rice to take home to the patient. Clients are
typically women who have come on behalf of their children.

KOREAN SHAMANS VERSUS

KOREAN NUNS
Korea is one of the few societies in which women seeking professional reli-
gious paths have more than one available option. Hesung Chun Koh (1984) has
compared the life-stories of five Korean Buddhist nuns to the life-stories of six
Korean shamans. Each of these individuals was free to choose the type of
religious specialist role that appealed to her; family preference was irrelevant.
Both nuns and shamans tend to be the first born or eldest daughter. Shamans
typically came from lower middle-class families who had declined in wealth
during the shaman's teen years; nuns came from middle- or upper-class fami-
lies. Both shamans and nuns experienced role conflict. Nuns tended to enter
monasteries in their early twenties; religion became important to them during
their teens when their families pressed them to make decisions (such as mar-
riage) regarding their adult female roles. Shamans found their callings in their
mid-thirties, when they were already married and had children, and often
when despite their own desires to do what proper women do, their husbands
98 Priestess, Mother, Sacred Sister
did not function as proper husbands (i.e., they did not earn money to support
their households).
Finally, both groups of women reported recent loss of beloved relatives: for
the nuns it was typically a parent, grandparent, or fiance; for the shamans it
was typically a child. 12 I believe that this distinction is a crucial one. While all
religions address "ultimate" issues of life and death (Geertz 1969), women's
religions particularly address maternal grief at the death of children.
I find interesting the difference between the way that dead babies are treated
in Korean women's shamanistic and household rituals, and how they are
treated—or more precisely not treated—in men's Confucianist rituals. As we
saw in Chapter 3, Confucianist ancestral rites only acknowledge patrilineal
ancestors who left male progeny. Women's household rites and shaman's se-
ances, on the other hand, relate to other categories of dead relatives—including
babies and children.

MEN, WOMEN, AND CHILD DEATH

In this chapter I have not tried to show that men do not love their children, that
men do not mourn for their children who die, or that men do not interpret
child death in religious terms. I have suggested that because of a variety of
social (and possibly biological) reasons, women's grief at child death tends to
be longer lasting and especially likely to lead to religious responses.
Several recent studies have compared mothers' and fathers' responses to
child death. Although all these studies have been conducted either in North
America or Western Europe, they do suggest trends that may be applicable in
other cultural situations. First, fathers seem to grieve for shorter periods of
time than mothers, and to express more of a desire to get on with their lives. A
typical mother's reaction is reported by Dyregrov and Matthieson: "I have not
recovered my own self following the death. I am much more anxious for
everything, and I think about illness and death every day" (1987, 9).
Mothers have been found to feel higher anxiety, self-reproach, sadness, and
sleep disturbances and to report more thoughts of the child. Fathers find it
more difficult to talk about the death, and both parents agree that the mothers'
grief is stronger and more intense (Dyregrov and Matthiesen 1987).
Mothers are more likely than fathers to suffer from increased health com-
plaints following child death (Dyregrov 1990, 269). As we will see in Chapter
5, women's religions offer diverse ritual and nonritual means for treating
vague and chronic somatic symptoms—the very sorts of symptoms com-
monly experienced by bereaved mothers.
A variety of theories have been advanced to account for the differences
between bereaved mothers and bereaved fathers. Some writers emphasize gen-
der differences in emotional responses in general, arguing that women report a
larger proportion of interpersonal situations as stimulating emotions than men
do, or arguing that men and women have different styles of coping with stress
(see Dyregrov 1990, esp. p. 275 for an overview of the literature). Differences
in response to child death are then interpreted as one manifestation of this
 When Children Die 99
global difference. I3 Other writers accentuate the different work experiences of
men and women, pointing out that because men are more likely than women
to return to work outside the home after the death, men are forced to "get on
with it" and not continue to brood and mourn (Rando 1986, 26). Women, on
the other hand, are surrounded by objects, activities, and other sensory stimuli
that remind them of the dead child (Schatz 1986).
A third approach has highlighted the special strength of the mother-child
attachment. During pregnancy the baby is physically part of the mother;
breast feeding continues this physical bond; and social arrangements in which
women have primary or exclusive responsibility for childcare reinforce that
connection. Especially during the first year of life, the psychological bound-
aries between the mother and child overlap. Simon Rubin has found that
many bereaved mothers continue to think of their deceased infants as develop-
ing children (he terms it a "phantom child syndrome") and points to the
"permanent presence of the child in the maternal experience" (1984-85, 351).
"When people ask me how many children I have, said Mrs. G., I answer
four; and if they ask further, I explain that three are living and one passed
away" (Rubin 1984-85, 351; see also Pine and Brauer 1986, 71, on "enshrine-
ment" of dead children).
The studies cited here indicate a number of differences between fathers' and
mothers' responses to child death, differences that have implications for
women's religions. That mothers are more affected both emotionally and in
terms of their day-to-day lives is reflected in women's religions' explicit appeal
to maternal sorrow. Of equal interest is the "phantom child syndrome" de-
scribed by Rubin. Can we understand the role of dead children in the rituals of
Spiritualists and Cantonese Shamans as a sacralization of this phenomenon? In
studies conducted by Western psychologists, bereaved mothers are unwilling
to unequivocally part from their children. Women commonly believe that their
children continue to live in some form, in some sphere. I4 It is of interest to us
that most women's religions elaborate on these beliefs—describing pleasant
hereafters, and encouraging communication between this world and the next.
Gehan Wijeyewardene shows that the Northern Thai spirit cult, like Western
Spiritualism, denies death. "In the eschatology of the caw [spirit], it almost
seems that Nirvana has been completely abolished, life is eternal" (1986, 201). I
see it as especially significant that Korean women's religion acknowledges the
spiritual presence of dead children (as restless ghosts and spirits), whereas
Korean's men's religion is concerned solely with adult ancestors who died after
producing male offspring.
The most important piece of information we can learn from psychological
studies of parental response to child death is that mothers, more than fathers,
report never "getting over" the death of the child. For many women, being the
mother of a dead child remains a—or the—core element of personal identity;
memories of the child do not fade; sorrow does not ebb; life does not "go on."
For these reasons, women seem especially likely to seek religious interpreta-
tions of and responses to child death.
The literature suggests that in addition to grief, sadness, and anger, bereaved
100 Priestess, Mother, Sacred Sister
parents try to find meaning in the loss. "Comparative cultural studies demon-
strate that the level of material abundance necessary to sustain life and induce a
sense of well-being is indeed minimal. . . . What most people cannot tolerate
for extended periods is the lack of a satisfying framework through which they
can make sense out of what they perceive to be inequity or other misfortunes"
(Williams 1980, 150). Because the death of a child, particularly of an infant or
young child, seems so incomprehensible, psychologists believe that searching
for meaning is a healthy way to recover from such tragedy. Vanderlyn Pine and
Carolyn Brauer cite studies showing that the existential search for meaning on
the part of bereaved parents may ultimately lead to positive growth outcomes.
Bereaved parents have reported developing a stronger faith, more compassion
and care toward others, and an increased sense of the preciousness of life (1986,
73). Thus I would argue that involvement in women's religions that offer
meaningful interpretations of child death (and techniques for avoiding deaths
of other children) provides more than emotional support or compensation; it
offers expertise in reassessing and coming to terms with the existential "mean-
ing of life". Participation in women's religions provides bereaved mothers
with opportunities for intellectual and spiritual growth. Ann Lee, for instance,
questioned accepted theological doctrines in light of her experiences of child
(and maternal) death. For Ann Lee, that process culminated in religious innova-
tion and leadership.
Infant mortality is one of the most dramatic concerns of women's religions,
but the sorts of theological questions raised by it also arise in regard to other
situations of suffering and illness. We will pursue this theme in Chapter 5.

Notes
1. A cautionary note is in order: It must not be assumed that all women share a
common experience of motherhood. Some women choose not to have children, some
women wish to have children yet are unable to do so. Some women thoroughly enjoy
rearing children, some women enjoy raising children when they have adequate financial
and social support, and some women find childraising a difficult, even agonizing process.
2. In a recent newspaper interview, Israeli author Yehudit Rotem described the
process leading up to her rejecting the ultra-orthodox Jewish world into which she was
born and raised and in which she lived as a married woman for twenty years. In the first
paragraph of the article we read that "Two children died in infancy and another was
stillborn, but pictures of the remaining seven—her greatest assets, she calls them—
cover almost an entire wall of her tiny Ramat Gan office" (Surie Ackerman, "The
Other Side of Sisterhood, "Jerusalem Post Magazine, August 21, 1992, p. 12). Somehow
I am not surprised to find that a woman who has made a "career" out of personal
religious change is also a bereaved mother.
Another example of the religious significance that women allot to child death conies
from contemporary Japan, where rituals for aborted fetuses are commonly performed
by women (Werblowsky 1991; see also Ferichou 1991, 216, on child death and women's
spirit possession cults in Tunisia.)
3. On the Ryukyu Islands the religious response to infant death is notably different
from other women's religions. According to Noguchi (1966, 28), dead infants (up to
 When Children Die 101
three years of age) are "hated and thrown into ravines, without their remains being
accorded 'bone-washing' and a tomb."
4. Pressel points out that we do not have a large sample of life histories of mediums
(personal communication 1992).
5. By the twentieth century, women in clubs and organizations began to act in the
public sphere to turn infant mortality from a private tragedy into a public social and
political issue. Twentieth-century mothers share responsibility for child welfare with
public health officials, the medical profession, and the state.
6. Dally cites several examples of mainstream Protestant interpretations of child
death from this period. "In Massachusetts, when Cotton Mather's daughter fell into the
fire and burned herself badly her father wrote: 'Alas, for my sins the just God throws
my child into the fire,' (1982, 44-45). Other fathers interpreted child death as a test of
faith.
7. Some women who stayed within the mainstream male-dominated denomina-
tions also sought religious means for coming to grips with child death. Debra Campbell
(1989) has shown that the life-stories of nineteenth-century Protestant women indicate a
correlation between women's conversion experiences and the deaths of their children.
8. Not only child death but also adult death led many women to Spiritualism.
Nelson (1969) shows that the popularity of Spiritualism in England in 1916-18 reflected
the large number of women who had been bereaved in World War I and wished to
contact their dead.
9. The question of why celibacy appeals to men necessitates a more extensive
analysis than I am prepared to present in a book dealing with women's religions. In the
case of the Shakers, the millennarian message proclaiming that celibacy was part of the
process of human perfection was probably equally compelling to men and women.
10. The male-female ratio changed over the years. In the early years, the sect was not
nearly as "feminized" as it later became (I thank Priscilla Brewer for pointing this out to
me, personal communication 1992).
11. Unlike Spiritualists, Shakers did not pay particular attention to sustaining rela-
tionships with dead children. On the other hand, Ann Lee's ability to interact with the
spirit world was a matter of great importance to early Shakers. Early Shakers were
concerned with assisting their dead relatives on the path to Redemption. According to
Humez (1992, 93), Lee is frequently remembered as speaking of witnessing the passage
into the resurrection of the souls of dead relatives of living Shakers.
12. In the belief system of Korean shamanism, deceased people remain in the physio-
logical state they were at the time of death. Therefore, dead children never become
adults.
13. Rosenblatt, Salsh, and Jackson suggest that women may not experience a death
more strongly; it may be that women are used (and allow themselves to be used) as the
persons who publicly symbolize the loss that everyone (men and women) have experi-
enced. This may be because it is seen as out of character for high status persons (e.g.
men in patriarchal societies) to self-mutilate, cry, or be self-indulgent in the face of loss
(1976, 27).
14. In light of Rubin's findings, it is not surprising that American women report a
significantly greater belief in the life hereafter than men (Keller, Sherry, and Piotrowski
1984).
 5
Misfortune, Suffering,
and Healing

The most conspicuous similarities among women's religions emerge in the

realm of suffering and healing. Whereas all religions deal in some way with
explanations of and solutions to suffering, women's religions are characterized
by the particular emphasis placed on illness and curing.1 This emphasis is
manifested theologically (elaborate explanations for suffering), ritually (key
rituals are often healing rituals), in terms of membership (members often join
because of illness), and in terms of leadership (a history of illness is typically
part of the path to leadership). Almost all the women's religions devote a great
deal of attention to suffering—especially illness—to the extent that some of the
religions have been "accused" of being not much more than healing cults.
Why do women create or join religions that deal with healing? First, as
primary childcare providers, most women are concerned with children's health
and function informally as domestic healers. In women's religions, this infor-
mal role is enhanced and formalized. Second, as subordinate members of sexist
societies, women more than men seem to suffer from persistent and recurrent
conditions such as headaches and dizziness that are not amenable to treatment
by either folk or modern medicine. In addition, women bear the brunt of
culturally assigned responsibility for infertility and all the physical conse-
quences of pregnancy and childbirth.
Women's religions advance ideological approaches that declare that suffering
is not necessary to the human condition—that it has discernable causes and can
be alleviated and avoided. Reflecting the insistence that suffering can and
should be avoided, these religions offer a variety of healing techniques, includ-
ing spirit possession, and eclectic approaches to suffering and healing that
allow the religion to be used as one of a number of culturally acceptable
solutions for suffering.
Let me say from the outset that I have not found any particular healing
technique to be unique to women's religions. What is striking about women's
religions is the accent on healing and, in most cases, the multiplicity of healing
procedures. In addition, the healing approaches of women's religions tend to
be characterized by a holistic mind-body-spirit approach: 111 bodies reflect
104 Priestess, Mother, Sacred Sister
emotional and spiritual distress, physical healing means that psychic tension
has been allayed. This trait is especially noteworthy in the women's religions
that are situated in Western cultures in which the mainstream world view is
highly dualistic. Both Western Christianity and Western medicine posit models
that split the body from the soul, bestowing care of the body on physicians
who typically pay little attention to the patient's overall emotional or spiritual
state, and bequeathing the soul to priests and ministers who, with the excep-
tion of "fringe" sects, are disinterested in bodily healing. In contrast, in
women's religions curing is typically a combination of healing interpersonal
relationships, healing relations with the gods or spirits, and healing physical
symptoms (Carol MacCormack, personal communication).
Women's religions rarely differentiate between illness and other types of
suffering. For the Korean shaman, for example, misfortune includes all sorts of
bad luck, illness, and financial loss. Also for Christian Scientists illness includes
not only bodily illness, but business problems, problem relationships, and
finding lost articles. "Nonphysical and minor problems constitute the greater
part of the [Christian Scientist] practitioner's work" (Fox 1989, no). In North-
ern Thai cities, women mediums are consulted for such matters as arthritis,
back problems, goiters, social problems, business problems, and care of chil-
dren (Wijeyewardene 1986). And on Okinawa, although the majority of the
people who consult yuta (shamans) do so for health problems, the roles of the
yuta include giving advice regarding plans, business, and wedding dates; inter-
preting unusual experiences such as dreams or accidents in religious terms
(neglect of rituals, etc.); communicating with relatives after death; leading
exorcisms to guide lost spirits to the tomb; leading rituals of thanks to the
house deity; and teaching traditional rituals (Lebra 1966, 79).

WOMEN AND ILLNESS

Women are associated with illness and healing on two levels: As primary
caretakers of children they feel themselves responsible for their children's well-
being, and as women carrying, bearing, and tending children in an imperfect
world, they themselves are often ill. The connection between women and
illness has been investigated from a number of perspectives (cf. Whelehan et al.
1988). As victims of institutionalized oppression in sexist societies, women
may indeed be ill more often than men. In many cultures men have access to
important resources that are denied women through food taboos or notions of
serving men first. Practices such as forcing very young girls to marry, and
prohibitions on birth control and abortion take their toll on women's health.
Furthermore the stigmatizing of infertile women—treating childlessness as a
female illness—causes many otherwise healthy women to be perceived as ill.
Feminist sociologists have argued that patriarchy actually makes women ill.
Women in patriarchal cultures may have no avenue available to them other
than illness for expressing their dissatisfaction with their lives. As Harriet
Lerner (1989) has pointed out regarding American society, both men and other
women are more comfortable with women who feel sick than with women
 Misfortune, Suffering, and Healing 105
who are angry. In general, women receive far more positive responses when
they define their problems in medical terms than in political terms.
It has been well documented that women all over the world utilize health-
care facilities of every type more than men, although men suffer from serious
disabilities and injuries more than women (Nathanson 1979). In the United
States, males in all age groups die at a higher rate than females and suffer higher
death rates from heart disease, cancer, accidents, suicide, homicide, diabetes,
and almost everything else. Yet even when diseases due to reproductive func-
tions are excluded, women suffer more from acute and nonfatal chronic condi-
tions than men. Women report more depression and are treated more for
mental illness. They report more minor ailments such as headaches, dizziness,
and stomach upsets. Women use more prescription and over-the-counter drugs
than men. They restrict their activities due to health problems about 25% more
days per year than men, and spend about 40% more days per year in bed
(Powell 1988).
Women's overwork, burn-out, and frustration often lead to chronic ailments
which do not respond either to modern or herbal medicine. Especially in
cultures that do not differentiate between physical and nonphysical ailments,
these are precisely the sorts of symptoms women seek to alleviate through
ritual means. Thus rural Mexican men and women exhibit similar emotional
states when faced by acute impairments, but different responses when faced
with chronic conditions. The chronic conditions are typically what bring
women to Spiritualist temples for treatment (Finkler 1985a).
In previous chapters we have looked at how the experience of motherhood
shapes women's religious activities and ideas. In this chapter we draw attention
to the connection between motherhood and illness. Judith Hibbard and Clyde
Pope (1987) conducted a study of gender roles and interest in health in the
United States. They found that not only are women in general more likely than
men to engage in health protective behaviors, mothers of young children are
more interested in health than those women with older or no children in the
household.
"What is missing in the understanding of mothering is the extent to which it
is a painful relationship and an exhausting, often thankless, occupation. ... In
the United Kingdom, research has demonstrated that up to 50% of mothers
with small children (under age five) have symptoms of intense emotional
distress on a regular or continual basis. . . . Women are five times more likely
to be diagnosed as mentally ill in the first year after their first child's birth than
at any other time in their lives" (Knowles 1990, 4). An even more interesting
statistic for our purposes comes from the Haj Yousif area of Khartoum (one of
the centers of the zdr cult) where 40.3% of mothers of three to fifteen-year-old
children described themselves as being anxious or depressed, and about one-
third of the women interviewed believed in zdr (Rahim 1991, 138). 2
In many cultures mothers complain of somatic symptoms rather than psycho-
logical ones. Research has shown that stress and emotional distress influence the
functions of the immunological system, and that social and psychological fac-
tors are linked to delayed recovery from infectious diseases. Life crises, and most
106 Priestess, Mother, Sacred Sister

especially loss and bereavement, have been shown to correlate with chronic
illness (Finkler 1985, 50). KajaFinkler suggests that subjectively perceived stress-
ful life events play a crucial role in illness expression "especially when they are
culturally and collectively accepted to be stressful" (1985b, 29). Finkler's insight
into the illness-causing role of bereavement allows us to identify another ele-
ment in the connection between child death and women's religion. Child death
may lead to chronically ill mothers who are incurable through physical means.
This is a population from which women's religions often draw members.
There is also a positive side to the motherhood-illness connection. The
strengths that women acquire through grappling with motherhood may be
incorporated into their repertoire of healing skills. Carol McClain has explored
the role of women as healers cross-culturally and found that in many instances
women who become "ritual or other specialized practitioners of medicine . . .
infuse powerful female symbolism, including the metaphor of maternal nur-
ture, into their healing practices" (1989, 6).
A more direct way in which motherhood figures into the religion-illness
equation concerns the key biological transitions in women's lives: menarche,
pregnancy, birth, lactation, menopause. In many cultures these events are inter-
preted in a health context, and lead women to seek the services of health
practitioners. Simultaneously, these same events may be interpreted in a reli-
gious context, and lead women to seek the services of ritual practitioners. Put
differently, motherhood is often understood as both a fundamental physical
crisis and a fundamental spiritual matter.

WOMEN'S RELIGIONS AS HEALING

CULTS
Zar: Reapportioning Responsibility
for Infertility
Muslim women in North and East Africa spend most of their time in domes-
tic units with other women and children. Looking at Muslim women's reli-
gious activities (Islamic and non-Islamic) in the general context of their lives,
Lois Beck concludes that their various religious responses address their par-
ticular problems as women. Much of the activity around pilgrimage to holy
tombs and saints, for instance, has to do with problems of fertility and child
mortality (1980).
Recruitment to the zdr cult is typically through illness. Although the dra-
matic possession rituals are the most flamboyant element of zdr religion, the
true emphasis in zdr cults is on curing, not ecstasy (Saunders 1977). Women
suffering from illness inflicted by zdr spirits need to join zdr ritual groups to
bring the spirits under some control—in order to alleviate their symptoms.
Possession rituals function as techniques to bring about healing. The Egyptian
Nubia, for example, use the zdr ceremony to cure a range of emotional and
chronic physical disorders. Patients most frequently treated through zdr are
those suffering from wasswassa—a state brought on by the death of a close
relative or by a frightening meeting with a spirit. The symptoms of this state
 Misfortune, Suffering, and Healing 107

are apathy, withdrawal from human company, minimal communication, re-

fusal to work, a desire to die, lack of appetite, and sleeplessness. Other com-
mon complaints are unlocalized pain, weakness, and listlessness. John Kennedy
argues that zdr cult initiation "seems particularly tailored for alleviation of the
hysterias, anxiety-produced problems, and psychosomatic ailments apparently
related to living conditions in Nubia" (1978, 203).
In North and East Africa if a woman fails to conceive, or miscarries, or bears
a stillborn child or a daughter, or loses an infant, it is her ability to procreate and
not her husband's that is called into question. Not surprisingly, many women
who join the zdr cult do so because of fertility and childbirth problems. Accord-
ing to Janice Boddy, zdr possessed women have been pregnant more times than
nonpossessed women, and have lost more children than women who are not
possessed (1989, 172). In the Northern Sudanese village in which she carried
out fieldwork, one of every two possessed woman has had both fertility and
marital difficulties, whereas only one in five nonpossessed reported both types
of problems. In light of Chapter 4*5 discussion of mother-love and child death,
it is significant that a number of women linked the onset of zdr possession to the
death of an adult daughter, usually in childbirth (Boddy 1989, 237).
Zdr possession as an explanation for women's fertility problems removes
responsibility and blame both from the afflicted woman and from Allah (who
is supposedly beneficent); instead, the zdr spirits are deemed to be the guilty
party. Furthermore, the fact that the afflicted woman's husband must pay for
her zdr initiation serves to reallocate the burden of infertility more evenly
between husband and wife.

Mexican Spiritualism:
Patriarchy Sickens Women
Kaja Finkler presents a convincing case for the relationship between women's
subordination and women's illness in rural Mexico. In recent years, high
school education for all children has become the goal of most families, a goal
that creates economic and psychological stresses and an increased workload
for women in particular. "In view of the traditional sexual division of labor,
women continue to carry out the incessant daily household activities, but
now lack the help previously received from their daughters. Moreover, be-
cause of the expanding need for cash, many women attempt to supplement
their spouse's income by petty commerce, or washing laundry for others"
(igSsb, 39). Increased industrialization has led to nuclear families becoming
more common. While the nuclear family does eliminate certain sorts of stress
between women, it also eliminates the possibility for sharing the workload.
More important, in the nuclear family there is greater emphasis on the male-
female dyad, which often means an increase in tension between the spouses.
"Significantly, extreme tension between spouses are reflected in the numerous
illnesses which are attributed by the women to anger and dissension between
spouses. . . . While dissension between males and females is inherently stress-
ful to both sexes, the men can more easily remove themselves from the
108 Priestess, Mother, Sacred Sister
situation by leaving the house and dissipating the emotional strains in the
cantina (liquor serving station) with their cohorts. Rural Mexican women lack
such escapes and outlets. . . . women remain relatively isolated and require
permission from their spouses to leave their households" (Finkler I985b, 40).
Compounding this difficult situation is the high rate of alcohol consumption
by men that results in both a financial drain on the household and increased
likelihood of wife-beating.
Finkler found that 58% of the clientele of a typical Spiritualist temple were
women, 27% children, and only 15% adult males (1985^ 58). Women in
particular complained of suffering from nerves (associated with chronic ill-
ness). 3 Many of the disorders bringing clients to the temple are directly associ-
ated with interpersonal strife, particularly between men and women. "In fact,
the majority of problems presented to Spiritualist healers concern conflicts
between men and women" (1985^ 61). Even more to the point, many of the
temple patients perceive this connection: "Many women tend to link their
symptoms to the onset of their marriage and to abuse by their husbands when
they return home drunk" (198 5 a, 40). Indeed, temple patients were far more
likely than other women to report difficult relationships with their husbands,
and to report that their husbands behave violently when inebriated.
In addition, we saw in Chapter 4 that the high infant mortality rates in rural
Mexico mean that many women spend their adult years grieving for their dead
children. Oppression and bereavement combine to set the stage for women's
religious involvement.

Afro-Brazilian Religions:
Illness as the Path to Recruitment
Illness is typically the path of recruitment to membership and leadership roles
in Afro-Brazilian religions, and healing consultations are the initial source of
contact for most people. Diana Brown (1986) interviewed Umbanda mediums
and found that 62% had originally sought out their present center for aid in
resolving some kind of personal problem, most typically a health problem.
Afro-Brazilian mediums are consulted regarding illness, marital problems,
business decisions, drinking problems, unemployment, interpersonal quarrels,
and insanity. Illnesses for which supernatural curing is sought are typically
chronic illnesses—ailments that did not respond to home remedies. Ritual
efforts to treat all of these sorts of problems are referred to by the same term,
cura (curing). Esther Pressel (1974) quotes one of her informants in Brazil as
telling her that the most common complaint that brings people to Umbanda is
headaches.
We can easily understand the attraction of spiritual curing in isolated and
primitive societies. It is more difficult for us to understand why Brazilian urban
dwellers choose to attend Afro-Brazilian rituals when they are ill. Surely there
are more efficacious medical options available in Rio de Janeiro! The reality,
however, is that in poor, urban neighborhoods in Brazil few people have the
resources to fully utilize Western health care: doctors and medicine are expen-
 Misfortune, Suffering, and Healing 109

sive. In addition, many of the diseases from which people suffer cannot be fully
cured by Western medicine (e.g., malaria, leprosy, tuberculosis). In these cases,
spiritual curing offers hope of miracles. And where the Catholic Church offers
sick people only prayers and vows, Umbanda and Candomble mediums pro-
vide cures that have been ordered specifically for the given situation by the
spirit world.

The Failure of Modern Medicine:

Japan, A Case Study
During the past century there has been a flowering of new religions in Japan,
many of which have been founded by women and almost all of which have
more female than male members. On the whole, these religions have a great
deal to do with healing; most offer explanations and remedies for illness, and
teach that illness is indeed curable. One may wonder why religious healing
should be so attractive in modern Japan where, unlike in Brazilian shanty-
towns, modern medicine (and even socialized medicine) are readily available.
Edward Norbeck (1970) argues that religious healing is encouraged by defi-
ciencies in modern medical practice in Japan. Modern Japanese medicine tends
to ignore psychological aspects of therapy. Relations between urban physicians
and patients are remarkably impersonal. Socialized medicine means that health
care costs less, but it also means that many more people seek medical services
and that poorly paid physicians treat vast numbers of patients. Average Japa-
nese citizens cannot make appointments for medical care, but must wait in
crowded waiting rooms to be treated briefly in an assembly-line fashion by an
overworked physician. In contrast, in the new religions ill people (mostly
women) receive a great deal of personalized attention; the roots of illness are
sought in the specific karma or life experiences of the individual; and healing is
effected through rituals involving notice and concern on behalf of the leader
and the group.

Spiritualism: Women Make the Best Healers

Studies of North American and British Spiritualism show that it is at stressful
periods, and especially periods of illness, that women begin to meet spirits.
"Spiritualists . . . consider themselves beset by a variety of disabilities" (Skul-
tans 1974, 27). June Macklin (1974), writing about the appeal of twentieth-
century Spiritualism, underscores that modern medicine, touted as infallible,
in fact often fails. Mediums in Macklin's study had reason to doubt modern
medicine—they or their kin had suffered from ailments that doctors could not
cure. In addition, many of their physical complaints were interwoven with
difficult social or marital situations. While in the secular world illness made one
dysfunctional, in the Spiritualist world sickness is seen as making a good
medium, and many women became mediums after long and severe illnesses.
"Every pain we suffer helps unfold our medium powers ... all suffering is
friction to the material covering of the soul, that makes the gem shine brighter
110 Priestess, Mother, Sacred Sister
within" (Moore 1977, 121). Spiritualists see the gift of healing as one of the
highest expressions of psychic power, and one that is accessible to everyone.
Women are however, on some level, better suited than men to act as healers,
and the majority of medical mediums have been women (see Chapter n).

Feminist Spirituality: Sexism and Suffering

Two women's religions explicitly acknowledge the connection between sexism
and suffering. The Shakers taught that suffering is caused by continuing
procreation—a process from which women in particular suffer.* Their solution
was a radical one: the formation of celibate communities.
The Feminist Spirituality Movement teaches that suffering, in a global sense,
has been caused by patriarchy that brings with it militarism, rape, conquest,
and disregard for nature. The solution to global suffering is the dismantling of
patriarchy. Rituals focus on healing women from the effects of patriarchy, and
Feminist Spirituality groups have devised rituals to heal women who have
been victims of rape and other forms of male violence.
Although curing is not institutionalized in the Feminist Spirituality Move-
ment, many Spiritual Feminists are involved with alternative healing systems
such as crystals, massage, home births, and aromatherapy. The impetus for
this involvement comes from a number of directions. On a political level,
feminists (including Spiritual Feminists) are unwilling to grant a health-care
monopoly to a capitalist and male-dominated medical establishment. On a
philosophical level, feminists are wary of the Western medical tradition that
tends to see women's bodies as pathological and that cures through attacking
germs or tumors rather than through cooperating with the body. And on a
personal level, many contemporary women suffer from ailments that modern
medicine simply cannot cure. In many instances, these ailments are dismissed
by physicians as trivial and psychosomatic. Spiritual Feminists are inclined to
believe that the physical and nonphysical are intertwined, and that all condi-
tions should be treated on physical, emotional, and spiritual planes.

"THERE ARE CERTAIN THINGS THEY

CAN DO"
In contemporary urban Korean society a variety of biomedical and religious
options are available to a sick individual. Thus it is significant that so many
women turn to shamans when misfortune strikes. The job of the Korean
shaman is to divine, explain, and remove the cause of suffering. Shamans
typically explain that misfortune is caused because the household gods became
affronted by neglect or pollution, and so dropped their defense of the family,
thus allowing ancestors to grow restless and dangerous. Youngsook Kim Har-
vey quotes one shaman talking about how she sees her role: "We give them
hope. . . . They see it's not so bad . . . There are certain things they can do
[my emphasis]" (1976, 196).
Suffering is a core concern of all religions. In various ways, religions en-
 Misfortune, Suffering, and Healing 111
deavor to explain why people suffer, and to offer solutions for suffering.
Differences among religions are manifested in how much attention is given to
the problem of suffering, and in the content of the response to suffering. In the
previous section I argued that because women more than men tend to suffer
from the sorts of chronic illnesses that are unresponsive to medical treatment,
women are especially likely to seek alternative solutions. In addition, in many
cultural situations the male-dominated religion does not offer women ade-
quate responses to suffering. Female-dominated religions, on the other hand,
tend to address head on such specific aspects of women's suffering as infertility
and child death.
In women's religions ideas concerning the causes of illness and misfortune
are highly elaborated. Most of these religions offer clear, persuasive, and
dramatic solutions to human suffering. Most women's religions teach that
suffering does not need to be passively accepted. In contrast, many (not all)
male-dominated religions teach that suffering is the will of the gods, prepara-
tion for bliss in the next world, punishment for acts in this or previous lives,
or the unescapable essence of the human condition.
In his classic sociological study of religion, Max Weber (1966, Chapter 9)
overviews the solutions that religions throughout the world offer to the prob-
lem of suffering. Weber summarizes the four principal responses as follows: (i)
messianic eschatology—a future revolution in this world, (2) heaven and hell
(retribution and reward in the world to come), (3) dualism, and (4) transmigra-
tion of souls. I find it interesting that most women's religions do not propose
any of these solutions. The Shakers, for example, employed messianic theol-
ogy and imagery, but for them the messianic revolution had already happened.
Notions of an afterlife are certainly recurrent in women's religions, but hell—a
place of eternal punishment and suffering—is foreign to the generally positive
views of human nature and the universe offered by women's religions. As we
saw in Chapter 4, hell as the repository of unbaptized babies was particularly
repugnant to nineteenth-century American women. Dualism—a split between
essential good and essential evil, a cosmic war between good and evil—is also
alien to women's religions. As we will see in Chapter 8, most of these religions
posit a multiplicity of good, bad, and indifferent supernatural beings, but none
posits an entity that is thoroughly evil. And finally, while most women's
religions believe in some sort of continued existence of the soul after death,
most seem to lean toward a belief that ancestors retain their identities rather
than a belief in transmigration of the soul.
In the following sections we will see in more detail how women's religions
address the problem of suffering.

Christian Science: Disease Is Not Real

Nineteenth-century North America provides a particularly illuminating case
study of the approach of women's religions to suffering and healing. I quote
here at some length from Ann Braude's wonderful analysis of women, reli-
gion, and medicine in the nineteenth century.
112 Priestess, Mother, Sacred Sister
Orthodox religion and orthodox medicine reinforced a similar worldview in
which human beings in their natural state were seen as flawed from birth and in
need of assistance from officially sanctioned authority figures trained in a special-
ized body of knowledge. While orthodox clergy portrayed the human soul as
inevitably prone to sin, orthodox physicians portrayed the human body, especially
the female body, as inherently prone to disease. Just as ministers traditionally
found a tendency toward sin in woman's moral anatomy because of Eve's instru-
mental role in tempting Adam into disobedience in the Garden of Eden, so doctors
associated woman's physical anatomy with a tendency toward pathology. The
emerging male medical establishment alleged that a disease-prone reproductive
system governed woman's physiology, resulting in inevitable physical frailty that
dictated a severely restricted sphere of action. Regular doctors joined the clergy in
asserting the appropriateness of women remaining within their "sphere," the
clergy basing their arguments on the Bible, the doctors basing theirs on the body.
Doctors and ministers agreed that both physical and spiritual ill health in women
resulted from disobedience. They prescribed obedience to a male authority figure
as a cure for the degenerative tendencies of body and soul. . . . [On the other
hand] Spiritualists [and Christian Scientists] opposed orthodox medicine with the
same fervor with which they opposed orthodox theology, and with some of the
same arguments. Because they viewed each individual as embodying the image of
God and the laws of nature, they viewed health, like godliness, as the natural
condition of human beings, which only misguided human intervention could
destroy (Braude 1989, 143-144).

To my mind, two points here bear emphasis. While Victorian doctors blamed
women's physiology (especially sexual physiology—ovaries and uterus) for all
emotional and physical problems (Smith-Rosenberg 1985^, nineteenth-century
female-dominated religions not only saw women as essentially sound, but
blamed external forces for the problems women suffer. And second, as the
nineteenth century progressed, people more and more believed that doctors
should be able to alleviate illness and suffering. In fact, however, physicians
were not very successful at healing, and many of their treatments were actually
harmful (e.g., blood-letting, or requiring that post-partum women stay in bed
for weeks). The spiritual healing techniques provided by women's religions
were a reaction to the unsuccessful and often intrusive male-dominated medical
establishment.
Of all of the women's religions, Christian Science provides by far the most
intellectually sophisticated explanation for illness and suffering. The first gen-
erations of Christian Scientists expressed disenchantment with the orthodox
Christian idea of resignation to suffering as the will of God in this life. They
felt that it is impossible to reconcile the suffering and evil in this world with a
benevolent and omnipotent deity. Therefore, the suffering and evil must not be
real. According to Christian Science there is no final judgment after "death;"
there is no heaven or hell. If God is good and all-powerful, He does not want
people to suffer.
"True Christianity, [Mary Baker Eddy] taught, must have a practical healing
and redemptive effect. She discouraged passivity in the face of suffering of any
 Misfortune, Suffering, and Healing 113
sort and utterly opposed the belief that man must acknowledge the cycle of
fatality. . . . Indeed, Christian Science itself stood for her as the means
whereby man could rise up and take control over his own destiny. She saw
redemption as including the redemption of the body. . . . According to Chris-
tian Science, one cannot attain final spiritualized consciousness while submit-
ting to any form of suffering" (Gottschalk 1973, 154).
Christian Science does not heal merely in order to lessen discomfort; rather,
because disease proceeds from the illusory belief in the reality of matter, heal-
ing through spiritual power destroys that false belief.5 By overcoming disease
(and eventually death, although not in our times—mankind has not yet at-
tained sufficient spiritual growth), one proves the power of the Spirit to de-
stroy the false beliefs of mortal mind.
Against this background, we can now understand the life history of Mary
Baker Eddy. Eddy was born into a middle-class rural New England family and
as a teenager began to feel dissatisfaction with contemporary Calvinist notions
of original sin and predestination. Her first husband died of yellow fever, and
her second husband felt that her son was too wild and rough for her to take care
of, and so sent him to live with another family. After losing her son, Eddy
became chronically ill. Her second husband left her, her health deteriorated,
she almost died, and then one day, while reading the Bible, she was healed. She
spent the next decades healing and teaching. She died in 1910 at the age of
ninety. For Christian Scientists of both the nineteenth and twentieth centuries,
Mary Baker Eddy serves as a role model of someone who successfully over-
came illness, pain, and suffering (Williams 1980).
Because Christian Science teaches that disease does not really exist, Christian
Scientists have sometimes been blamed for cruelty or disregarding human
suffering. However, the official position of Christian Science is that although
evil and suffering are in fact not real, people do perceive them as real and it is
wrong to ignore suffering. A Christian Science practitioner can only heal,
according to Eddy, if the healing is impelled by true affection. 6
In conclusion, I fully agree with Mary Bednarowski's assessment that, "The
appeal of Christian Science for women lay primarily in its stress on self-help
rather than helplessness, and on the possibility of healing without dependence
on the dictates of doctor or clergy" (1980, 218).

Burmese Buddhism versus Nat Religion

The contrast between nineteenth-century American women's religions and
Calvinism/official medicine is in many ways identical to the contrast between
Burmese nat religion and Buddhism. The differences between male-dominated
Buddhism and female-dominated nat religion in Burma emerge dramatically
when we look at how the two systems deal with suffering. According to
indigenous Burmese belief, nats (together with ghosts, demons, and witches)
are potentially harmful. Human suffering is caused by supernatural beings who
have not been propitiated, whose territory has been invaded by humans who
114 Priestess, Mother, Sacred Sister
did not ask permission, or who sometimes are simply capricious. The solution
to suffering is to perform the proper rituals to appease the nats. Nat religion
absolves humans of responsibility for causing suffering; individuals are the
victims of nats, witches, and ghosts. Humans can, however, learn to alleviate
suffering through religious rituals.
Therevada Buddhism, on the other hand, explains suffering in this life as the
result of karma—the consequences of one's deeds in previous existences. What
one should do in this life is strive for a better situation in the next life. The way
to do this is to follow Buddhist precepts, to transcend the cravings and desires
that produce wrong action. Buddhism teaches that each individual is ulti-
mately responsible for his or her suffering in this world; it was one's own
behavior in previous lives that has caused one to suffer in this life. And al-
though there are certain Buddhist rituals that are apotropaic in aim, the bulk of
Buddhist teaching and ritual does not offer an immediate solution to suffering;
the best one can hope for is less suffering in subsequent incarnations. 7
Buddhist explanations of and solutions to the problem of suffering are not
sufficient in the eyes of many inhabitants of Buddhist lands. Because Bud-
dhism interprets suffering as an unavoidable aspect of this worldly existence,
and locates both the cause and solution in other lives (past and future), individu-
als concerned with alleviating misfortune here and now are not offered much
help. Nat religion, on the other hand, offers both compelling explanations for
and immediate solutions to human suffering. "While the nats do not control
decisions or impose patterns of action, all the exigencies of daily living and the
crises in the life cycle are interpreted in terms of them" (J. Nash 1966, 127).
Indeed, the most common sign of nats' displeasure is illness. When someone
falls ill, Burmese villagers suspect that they have aroused the anger of house-
hold nats. "A stream may flood and drown the people along its bank, a tree
may fall and kill the person walking in its path. The forest is trackless, and it is
easy to lose one's way and die of privation; wild beasts are unpredictable and
may attack a harmless victim. None of these events occurs by chance" (Spiro
1967, 47)-
Steven Piker contends that Buddhism embodies universalistic, categorical,
and all-inclusive explanations of the causes and results of events, whereas
magico-animistic beliefs (spirit cults, nat religion) are particularistic, frag-
mented, and ad hoc in the explanations they offer (1972, 215). In Chapter 7 I
propose that the "particularistic explanations" are what make "magico-animis-
tic beliefs" especially attractive to women.

TECHNIQUES OF HEALING
In women's religions suffering has meaning; illness is more than bad luck or
bad fate. Healing rituals of women's religions are dramatic antidotes to resigna-
tion in the face of suffering. Because healing is so crucial in women's religions,
it should not come as a surprise that many offer a variety of healing techniques.
The message of women's religions is that suffering is not inevitable, that
individuals can and should seek means of alleviating suffering, and that we (the
 Misfortune, Suffering, and Healing 115

shaman, the cult group, the priestess, etc.) offer a range of healing procedures.
Anthropologists have suggested that magical or religious curing lends itself to
the "law of accumulation." Since healing rituals endeavor to be as effective as
possible, healing practitioners accumulate many and diverse rituals in order to
maximize the possibility of efficacy (Bastide 1978, 278).
One of the noteworthy patterns found in women's religions is openness to
the use of healing systems other than those offered by the religion itself (see
Table 2). I would offer two explanations for this finding. First, as I shall show
in Chapter 12, women's religions generally do not teach that "only our way is
true." Aggressive missionizing, for example, is almost never a part of
women's religions. Second, healing is so crucial to women's religions, and the
sorts of illnesses that members suffer from tend to be so difficult to cure, that
prohibiting use of other healing systems might easily backfire and cause
women to forsake membership in order to continue searching for all-too-
elusive cures.
In two fascinating studies of the effects of foreign influences on Okinawa
(Ryukyu Islands), a team of Japanese scholars found that in the realm of curing
there is co-existence—Okinawans are perfectly willing to use Western medi-
cine and yutas (shamans) simultaneously (Ohashi, et al. 1984; Matsui, et al.
1980). The scholars prophesy that Western medicine will not conquer Okinawa
because the biomedical model of disease is limited to answering the question of
how a disease occurs. It is unable to explain why a disease happens to the
specific individual. A typical modern Okinawan pattern is to consult a physi-
cian for treatment and a yuta for explanation.
This pattern is common in women's religions, and gives us insight into why
these religions often pay little attention to actual symptoms. Women's religions
excel at diagnosing and healing the underlying cause of suffering, not its tempo-
ral manifestations. 8

AFFIRMATION AND AUTHORITY

Writers in a number of disciplines (medicine, psychology, anthropology, soci-
ology, history of religions) have found it remarkable that religious healing
systems often do accomplish their therapeutic aims. A thorough examination
of how, in the eyes of Western scholars, these healing rituals "work" lies well
outside of the issues I wish to deal with in this book. Various scholars have
explained the success of healing rituals in terms of the social support the
patient receives, emotional catharsis, the use of evocative symbols that inten-
sify the faith of the participant, trance that frees the patient from social norms
and superego control, and the manipulation of the world of fantasy (the
subconscious).
Two features of healing seem to be particularly salient in women's religions.
First, in these religions a central element of healing rituals is group acknowledg-
ment that one's suffering is real. Typical scenarios begin with a woman whose
family is reluctant to acknowledge that she is in distress; women who come to
female-dominated curing religions are often considered malingerers or hypo-
 Table 2. Eclectic Healing Techniques
In Afro-Brazilian religions misfortune can be a sign that one should be initiated, a punishment
for infringement of religious rules, the actions of an exu (spirit of the dead), or caused by a
sorcerer. Responses to suffering include propitiatory sacrifices, retreats, baths, fumigations, shak-
ings, dusting with protective powders, and divination (Landes I94ob, 262).
Black Carib women understand a variety of reasons that people become ill: germs, sorcery, lack
of practical caution (such as wearing insufficient clothing), angry ancestors who were not prop-
erly treated through ritual, and taboo behavior on the part of the individual's mother. Various
causes of illness legitimate various solutions to illness: Western medicine, bush medicine, and
ancestor rituals.
Black Spiritual people of New Orleans commonly mix healing therapies: physicians, patent medi-
cines, home or folk remedies, and healing rituals at church.
The fertility knowledge owned by Sande women includes a wide variety of empirically tested
herbal medicine, amulets, and more "magical" remedies and prohibitions (MacCormack 1982,
127). The Sande system is inherently flexible; new information is easily absorbed; and Sande
midwives are eager to learn more about hygienic techniques that will reduce maternal and infant
mortality.
Regarding the Northern Sudan, Janice Boddy (1988) describes the following steps likely to be
taken by a sick person: advice from family members, home remedies, patent medicines, Western
doctors, and/efei Islam (male religious specialists) who perform divinations and provide charms.
If spirits are found to be the cause of the distress, thefeki Islam will perform an exorcism.
However, since the zairan (zar spirits) are immune to Islamic ritual techniques, women patients
may then consult a female zar practitioner. Yael Kahana (1985) found that in Ethiopia zar is not
turned to after other cures have failed, but together with other cures, all of which are seen as
complementary.
When an Upper Burman is sick he or she consults many sorts of experts: monks, astrologers,
natkadaws (nat shamans), and doctors (M. Nash 1966). Kunstadter notes that in Southeast Asia
several medical care systems are available; religious eclecticism is the norm. Illness is caused by
various categories of spirits, soul loss, sorcery, sin, germs and more (1978, 187-188).
Spiritualists accept multiple models of illness: social, spiritual, psychological, and organic; and
eclectic approaches to treatment: spirit healing, herbs, psychological insight, and social manipula-
tion of difficult situations. Spiritualism sees itself as complementary to orthodox medicine, not as
a full-fledged alternative to it. Health advice ranges from simple and practical (eat vegetables), to
American Indian remedies, to seances (Skultans 1974).
The Korean shaman functions within a health system in which there are many options. Sick
people typically turn in this order to various practitioners: pharmacist, Chinese-style herbalist,
hospital. The shaman often actually urges the client to also see a Western or Chinese doctor
(Kendall 1985, 94).
The Shakers did not engage in healing per se, yet they were very concerned with health. In
addition to emphasis on neatness and cleanliness, Shakers experimented with various health regi-
mens, including a wide range of medical and manipulative therapies (herbs, hydropathy, bloodlet-
ting, faith healing, Grahamism).
On the Ryukyu Islands reasons for misfortune include insufficient ancestor ritual, priestess or
shaman failing to acknowledge her call to office, improper ritual action, defiling a sacred place,
violating social values, crimes of violence, and improper behavior on the part of one's ancestors
(while they were alive). A wide range of solutions are available to deal with this wide range of
causes
 Misfortune, Suffering, and Healing 117

chondriacs. In the context of the women's religion, on the other hand, the
sufferer receives communal, often public affirmation of her perception of real-
ity. I am arguing that in addition to the emotional support chronically ill
women receive from joining a group, women receive cognitive support. In
patriarchal cultures the dominant model of reality is one that reflects and
expresses men's perceptions. Women often have their own alternative model
of reality, but this model receives far less social confirmation than men's mod-
els, for the simple reason that the dominant social institutions are based on
men's models. Religious healing often offers a model of reality that "fits"
women's experiences and so allows them to move beyond illness to strength.
A second and related feature is the role of the healer; in very few of the
religions is the healer treated as an absolute authority. Women are allowed
space to negotiate explanations and cures that correspond to their own percep-
tions. For example, clients waiting to consult with a Korean shaman behave as
though at an informal social gathering, offering comments and duckings of the
tongue, and the shaman sometimes uses this audience to muster a consensus
support for her advice (Harvey 1976). The shaman brings in a divination
tray—an ordinary tray of the type used for everyday Korean meals. On the
tray is a mound of rice, some coins, and a bell rattle to summon the shaman's
visions. She tosses the rice and coins, and the patterns suggest the client's
concerns. The shaman describes a situation (based on the rice and coin configu-
rations, and on what the gods told her) and asks for confirmation. As the
divination continues, the shaman brings together her own visions and informa-
tion from the client. "They [the friends and relatives present at the divination]
sigh sympathetically for the woman whose divination reveals an adulterous
husband, unruly child, or pitiable ghost. . . . Not for them the confidential
atmosphere of the Western doctor's or analyst's office. The confessional's ano-
nymity is missing here. The women enjoy each other's stories and accept each
other's sympathy" (Kendall 1985, 74). And as David Suh points out, "This is a
religion of dialogue between the mudang [shaman] and the 'congregation'—a
lot of give and take between them" (1989, 22).
Leacock and Leacock have observed similar forms of behavior at an Afro-
Brazilian curing ritual (1972, 253). Haifa dozen clients sat in the medium's
living room. A woman who had come to consult the medium on behalf of her
sick husband knelt in front of the medium and whispered to her about the
problem. Another client leaned closer to hear, and soon all the clients joined
the conversation. The medium (entranced) prescribed certain herbs for the
husband, and the other clients commented on the appropriateness of the herbs.
The contrast between this style of healing and the style of western medical
doctors must be striking to Brazilian or Korean women, who have access to
both [male] Western physicians and [female] religious healers.
I am struck by the social nature of healing rituals in women's religions. As
we will see in the next chapter this sort of communal healing is consistent with
the great emphasis placed on interpersonal relationships in almost all the rituals
of women's religions.
118 Priestess, Mother, Sacred Sister
Notes
1. I am not claiming chat male-dominated religions do not give attention to illness
and healing. Healing is certainly one of the dominant themes in, for example, the life of
Jesus. Dissident religious movements, whether male or female dominated, often use
healing to recruit new members. (I thank Kaja Finkler for pointing this out to me,
personal communication 1992.) And while healing is a dominant theme in most
women's religions, it is not central to all women's religions. The Christian "heresy"
Guglianism (a female-dominated cult that advanced a female pope), for example, inter-
preted suffering as a necessary manifestation of the unredeemed world, and did not
offer followers ritual means for alleviating suffering in this world.
2. It is crucial to bear in mind that a variety of studies have failed to find elevated
rates of psychopathology among participants in religious groups or among nonbio-
medical practitioners (see Finkler 1986, 632, for an overview of the literature).
3. In the part of Mexico in which Finkler carried out fieldwork many people suffer
from parasites that cause vague, chronic symptoms.
4. At the time that Shakerism emerged, the workday in Manchester and similar
cities averaged fourteen hours, and children were hired in the factories and mills from
the age of four. Working conditions were horrible, discipline in the factories was strict,
and workers—including numerous pregnant women and young girls—were forced to
work at a pace that often resulted in broken health.
5. The immediate cause of disease is usually fear (terrifying images or thoughts
impressed on the body), but the basic cause of disease is false belief. Christian Scientists
also believe that certain individuals can harm other people through a process known as
animal magnetism or mental malpractice.
6. Despite Christian Science's well-publicized refusal to turn to other healing sys-
tems in cases in which spiritual healing has not proven effective, Christian Scientists do
use surgeons to set broken bones, acknowledging that surgery is difficult to effect
through spiritual healing.
7. Buddhism (especially "folk Buddhism") does provide rituals aimed at alleviating
present suffering, but it seems to me that the emphasis is far more on alleviating future
suffering in the next life (Spiro 1971, esp. Chapter 6). Indeed, as Spiro shows, Burmese
informants were quite vague concerning how Buddhist apotropaic rituals are even
supposed to work.
8. Kendall makes the same point concerning Korean shamanism. Underlying ill-
ness (especially serious or prolonged illness) are angry gods and restless ancestors.
"While the mansin [shaman] concedes the efficacy of medicine for herself and her clients,
medical expense is [interpreted as] another aspect of household affliction" (Kendall
1985, 94)-
 6
Rituals and Relationships

RITUALS OF THE RYUKYU ISLANDS —

A CASE STUDY
The most striking feature ofRyukyuan religion is the sheer number of rituals performed
by the priestesses. Since neither the kami (gods) nor ancestors are inherently benevo-
lent, humans must work hard in order to be rewarded with health, livelihood, and
fertility; good relations with the kami demand time and attention.
C. Ouwehand (1985) documents forty-one rituals that make up the agricultural cycle.
The rituals dramatize a few basic themes: thanksgiving for the harvest, securing prosper-
ity of crops and of human beings for the next year, warding off evil spirits, and bridging
the preceding to the following agricultural cycle. Priestesses visit ritual sites before the
day of each main ceremony in order to inform the kami that the ceremony will take
place, and to request their attendance.
Within the household, ancestors are informed about matters of importance to the
family: births, deaths, business transactions, and marriages. Ancestral rites are per-
formed regularly by the kin-group for their ancestral spirits.' Household rites center on
the hearth (kamado,), and the hearth spirit is a messenger between the family and
higher kami. It is always the wife who conducts kamado rites. Many rituals are
performed by sisters to safeguard their brothers (see Chapter 10).
Fishing and sea rituals include various rites for ensuring the safety of the fishermen
and the abundance of sea harvest.
Rites of passage include installation and retirement rites for priestesses, and marriage,
birth, naming, and death rites.
In parts of the Ryukyu Islands the priestess (noro) annually entertains a god as lover
in secret rites that precede the harvest festival (Haring 1964, 50). In some communities
there are rituals welcoming deities who have come from mysterious worlds to bring
fertility and prosperity.
Priestesses (noro) are engaged to pray on behalf of sick individuals, or people
facing crises or challenges. They also preside over rites preceding new ventures such as
selecting locations for tombs, building new houses, starting on journeys, and launching
boats.
I2O Priestess, Mother, Sacred Sister

RITUAL, THEOLOGY, AND GENDER

Ritual is the outward face of religion. Through ritual, people give expression
to their beliefs, their myths, and their hopes and fears. Successful and persua-
sive ritual strengthens people's conviction that their religious institutions are
powerful and true. Ritual serves to consolidate disparate individuals into a
group, and ritual serves to link that group to other, supernatural realms. Ritual
is a tool for eliciting altered states of consciousness in participants. And rituals
are among the most widespread means used by human beings who wish to
manipulate their environment, ward off misfortune, and ensure well-being for
themselves and their loved ones.
The existing literature on women and religion enables us to direct our atten-
tion to a number of questions regarding the nature and role of ritual. Are
women's religions ritually rather than theologically oriented? Are women's
rituals particularly expressive emotionally?
Walter Pitts (1989) studied one American Afro-Baptist congregation in
which the membership is 75% female and the clergy is all male. He found a
sexual division of religious labor: men are interested in theology (the nature of
God and life) whereas women are in charge of rituals and caring for the church
and congregation. Pitts has suggested that the association of men and theology
versus women and ritual is in keeping with the traditional African model of
gender and religious activity (personal communication).
I, on the other hand, am inclined to believe that all people, male and female,
think about existential issues (of course some individuals are more "intellectu-
ally inclined" than others). Gender difference, I suspect, lies in proficiency at
talking about or writing about these issues in a language that ethnographers or
historians can understand, and in institutionalized access to acquiring that profi-
ciency. The elderly Kurdish Jewish women among whom I conducted field-
work in Israel are extraordinarily inarticulate. Most do not even know one
language adequately for prosaic purposes, and traditional Jewish teaching
about sacred matters certainly was not available to women of their generation.
Partway through my time in the field however, I began to see that the women
used an eloquent language of hand gestures and food rituals to express complex
theological and philosophical notions. These old women have taught me to be
suspicious of reports of women's religiosity that minimize theology.
Neither all women's religions nor all women are located at the "highly
ritualistic" (as opposed to the "highly theological") end of the religion contin-
uum. Christian Science, for example, stresses study—"science"—rather than
ritual, and their church services are plain and unceremonial. Moving back-
wards by half a millennium, the Beguines, monastic women in medieval Eu-
rope, chose a life-style that emphasized service to the poor over a cloistered life
of prayer and ritual.
When scholars say that women are more concerned with ritual than with
theology, the implication often is that ritual is somehow less noble, im-
portant, or sophisticated than theology. 2 While men sit and ponder and
write about complex metaphysical problems, women jump up and down
 Rituals and Relationships 121

and ask the spirits to cure their children of the flu. This sort of dichoto-
mous and androcentric thinking is of little use in grappling with •women's
religions.
With these words of caution in mind, an overview of women's religions
shows that a strong emphasis on ritual is indeed typical. 3 There are several
women's religions that could essentially be described as constellations of ritu-
als: these religions lack a standardized belief system, formal membership proce-
dures, rules and regulations, and recognized leaders. The clearest example of
this model is American Spiritualism. In fact, an "overemphasis" on ritual has
been one of the criticisms most commonly leveled at nineteenth-century Spiri-
tualism. The principal feature of Spiritualism is the seance—a dramatic ritual in
which mediums establish contact with spirits of the dead. June Macklin has
observed that in the modern Spiritualist movement "the philosophical and
religious aspects of the system are not a primary concern" (1974, 411).
An emphasis on ritual, however, does not mean that women's religions are
simplistic. Human beings have many ways of articulating their beliefs; writ-
ing theological treatises is but one. If we abandon a dichotomy between ritual
(magical, superstitious, ignorant) and theology (philosophical, abstract, un-
selfish, moral), we begin to see that rituals may express very complex belief
systems. Spiritualist seances, for example, explicate and reinforce the notion
that material reality is not the only reality; that all living creatures are eter-
nally connected with one another; that what one person does affects everyone
and everything forever; that people are not specks of dust, disappearing into
nothingness when they die; that love has meaning; that human relationships
are sacred.

WOMEN, RITUALS, RELATIONSHIP,

AND COMMUNITY
Religions dominated by women tend to have an interpersonal rather than an
individualistic orientation. Women's religions rarely include hermits or yogis
dwelling in mystical isolation on mountaintops (although in male-dominated
religions there are women who choose isolated spiritual paths). Far more typi-
cal of female-dominated religions are rituals and belief systems well designed
to strengthen bonds among people. This is consistent with Carol Gilligan's
(1982) contention that women's experiences lead them to understand morality
in terms of webs of relationship and interpersonal caring and responsibility,
whereas men's experiences lead them to understand morality in terms of ab-
stract issues of rights and justice.
In many societies women are especially responsible for caring for children and
for old people. Relationship is often the core element of women's daily experi-
ences. Women's religious lives serve to sacralize those experiences; women's
rituals tend to reflect the interpersonal orientation of women's profane activities.
"[A]ll [Haitian] Vodou healing is the healing of relationships" (Brown 1991,
331). Karen Brown observes that, "In her cures Alourdes [a Vodou priestess]
122 Priestess, Mother, Sacred Sister

puts problematic human relationships into a tangible, external form, where

they can be worked on and ultimately transformed. When a love relationship is
desired, she binds two dolls face to face. When the dissolution of a relationship
is sought, she binds them back to back. For restive, 'hungry' spirits, she
prescribes a meal of their favorite foods. To treat a violent marriage, Alourdes
makes a charm for the wife by filling ajar with ice ('to cool him down') and
molasses ('to make him sweet')" (1991, 348).
Feminist psychologists offer a number of persuasive explanations for wo-
men's interpersonal orientation. One of these, developed by Nancy Chodo-
row, has already been summarized in Chapter 3. Carol Gilligan (1982, 23)
writes that, "The elusive mystery of women's development lies in its recogni-
tion of the continuing importance of attachment in the human life cycle. . . .
The myth of Persephone . . . remindfs] us that narcissism leads to death." A
similar idea has been expressed by Belenky, Clinchy, Goldberger, and Tarule:
"Men, valuing distance and autonomy, are more exclusionary. To them, 'we'
clearly means 'not they'. Women, valuing connection and intimacy, are much
more likely to be inclusionary, finding 'they' and 'we' to be intertwined and
interdependent" (1986, 45).
Women's religions illustrate and clarify these ideas. In Northern Thai
matrilineal spirit cult rituals, for instance, women ask ancestors to guard
over descendants, and descendants fulfill their obligations to the ancestors.
These rituals highlight dependency and interconnectedness. I find it interest-
ing that Northern Thai descent group spirits punish intragroup conflict
through sickness. The offense is brought out into the open, apologies are
made to the spirit, and solidarity is expressed in the communion of food
offerings. The descent group's spirits do not necessarily punish the particu-
lar individual who has committed an offense; instead, another member of
the group may be afflicted by illness or misfortune in a sort of "communal
punishment." The solution, then, is not individual but rather communal
rituals in which all members participate. "The spirits are here identified as a
single source of morality that determines the well-being, health and fortune
of all members of the descent group, irrespective of actual men or women"
(Tanabe 1991, 191).
The importance of harmonious interpersonal relationships in Ryukyuan cul-
ture can be seen in the traditional way of determining who had committed a
crime. In this ritual each member of the village must find a nonrelative to be his
or her close partner. Whoever could not find a partner was assumed to be the
guilty party (Lebra 1966, 131). Like most women's religions, Ryukyuan reli-
gion is essentially communal and the major rituals are carried out in the context
of the family and community. As in North Thailand, supernatural punish-
ments for ritual neglect or misconduct do not always affect the specific individ-
ual who was guilty of the breach in behavior; rather, other members of his or
her family may suffer as well. Relations among the priestesses are particularly
cordial—they believe that discord will harm their relationships with the kami
(gods). And they try to provide a positive example of cooperation for the
people on whose behalf they serve.
 Rituals and Re 123

Similarly, Korean shamans use horoscopes (personal year fate) to link indi-
vidual affliction to the supernatural state of the household. When the gods are
angry and the household's defenses are lowered, the individual who will suffer
is the one whose horoscope indicates particular vulnerability; this is not neces-
sarily the one who committed the improper act. "Religious practice in Korea is
not for the individual but for the family unit" (Lee 1984, 186). Kut (shaman's
ritual) emphasizes the inclusive solidarity of the extended bilateral family. The
goal of Korean shamanistic ritual is to discern which household god was
offended, and to carry out the appropriate ritual action to "patch up the rela-
tions between human and spirit and restore the integrity of the house" (Kendall
1983, 99).
David Suh offers the following analysis of Korean household and shaman's
religion: On the one hand, the living want the dead to be totally dead and go
away. On the other hand, they continue calling the dead back through the
shaman's trance to ask for blessings. "Dead or alive one cannot escape a tight
relationship with family and community" (1989, 14). Suh stresses the commu-
nity centeredness of shaman rituals. "The kut is a family affair, if not an affair of
the whole village. . . . [T]he entranced mudang (shaman) will give the whole
family didactic instruction . . . admonishing them to mutual cooperation, filial
piety, loving each other, sharing the wealth generously among the relatives and
so on. Basically, the mudangs seem to believe that the basic cause of sickness and
misfortune in the family is a lack of giving, care and concern and love among
the members of the family. . . . The mudang kut brings about a renewed spirit
of love among the members of the family" (1989, 18).
An even clearer instance is that of the American Shakers. Shaker religion had
meaning only within the context of Shaker communities where individuals
could rely on the physical and emotional support that developed among sisters
or brothers "in the gospel" (Procter-Smith 1985, 68). According to Stephen
Marini, the Shakers understood that "it was their community—the sharing of
goods, spiritual gifts, and discipline—that had the most impact on New Eng-
landers" (1982, 88). Thus, hospitality was a key feature in Shaker evangelism,
and all visitors were offered food and lodging. 4 Furthermore, "Shaker worship
exceeded all other forms of Christian worship for full and equal congregational
participation" (Procter-Smith 1985, 162).
Spiritualism is characterized by its emphasis on interpersonal bonds that
continue even after death. June Macklin quotes a medium leading a Spiri-
tualist service: "I want each and every one of you, as is my usual custom, to
sit in a little silence for a moment, and ask your spirit loved ones to come
and visit with you this afternoon. You know they like to be invited espe-
cially, just the same as you and I do from day to day by our friends and
loved ones here" (1974, 404). Spirits can atone for sins committed while on
earth, so they frequently apologize for improper conduct toward those left
behind, which serves to reestablish good relationships. Kin spirits transmit
messages of love, and are characteristically concerned and supportive. Spirit
messages not only strengthen ties between the living and the departed, but
also enhance relations among the living who often are told by the spirits to
124 Priestess, Mother, Sacred Sister
mend their relationships with friends and relatives. In Spiritualist groups,
"The client is treated as a significant individual, worthy of the special and
unique attention of the caring powers" (Macklin 1974, 413). Spiritualism
highlights the importance and significance of the individual and that per-
son's relationships. In the Spiritualist view people are not simply cogs in the
wheel; each person can affect his or her own destiny. People and their rela-
tionships matter.
Similarly, the essence of the Feminist Spirituality Movement is interpersonal
relationship. "Earth-based spiritual traditions [such as Spiritual Feminism] are
rooted in community. They are not religions of individual salvation, but of
communal celebration and collective change. Community includes not just the
human but the interdependent plant, animal, and elemental communities of the
natural world" (Starhawk 1987, 23). In recent years a key political concern of
the Feminist Spirituality Movement has been ecology. "With many spiritual
feminists, ecofeminists, ecologists, anti-nuclear activists and others, I share the
conviction that the crisis that threatens the destruction of the earth is not only
social, political, economic, and technological, but is at root spiritual. We have
lost the sense that this earth is our true home, and we fail to recognize our
profound connection with all beings in the web of life" (Christ 1989, 314;
my emphasis).
And to take one final North American example, among the nineteenth-
century Sanctificationist sisters, dreams were the chief form of spiritual inspira-
tion and divine communication. However, dream interpretation was a group
activity which, according to Kitch (1989, 180), fostered group cohesion.
As we have seen in Chapter 5, women's religions devote attention to illness
and healing, and offer persuasive explanations for why people suffer. It is
significant that these religions, for the most part, blame neither the individual
sufferer nor her friends and neighbors for the illness. The zar cult, for example,
offers cures and explanations for illness that are far less socially "hot" than
many other possible explanations (such as sin, sorcery, or the evil eye). Zar
possession does not cast blame on either the sufferer (she did not ask the zar
spirit to bother her) or on neighbors or acquaintances (other people cannot
manipulate zar spirits in order to hurt someone else).
An almost identical argument has been made by Kaja Finkler regarding
Mexican Spiritualism. She demonstrates that Spiritualism differs from other
Mexican models of illness in that it denies the existence of witchcraft and
attributes affliction solely to impersonal spirits. Like most other women's
religions, Spiritualism does not hold the individual liable for her illnesses—
suffering is caused by external forces. Thus, "Spiritualists make an important
statement about human interaction and the shaping of illness. . . . the great
importance Spiritualists attach to amiable social relations is evidenced by their
shifting the onus of an illness from one's neighbors, friends, or relatives to
impersonal spirits. . . . many patients recognize the advantages of an ideology
that stresses positive social interaction" (1985, 52-53).
The remainder of this chapter will explore the theme of relationship in ritual
in the contexts of initiation, mourning, and food rituals.
 Rituals and Relationships 125

INITIATION
Sande Initiation Ceremonies
In Sande secret societies the key ritual complex is initiation into the group. For
Sande women, initiation (rather than marriage, first birth, etc.) is the most
important ritual girls and women participate in. During the prolonged initia-
tion period, young women learn to rely on the group; they learn group secrets,
group myths, group sacred knowledge; and they promise never to speak ill of
other Sande women. Initiates undergo a painful physical procedure that marks
them as members of the group and when it is time to give birth they return to
the group in which they were initiated. Sande is a multifaceted secret society
with various functions and goals, but as a system of myth and ritual it only has
meaning within the context of the group.
Among the Kono of Sierra Leone initiation into Sande is preceded by several
weeks of parading and dancing by groups of female participants known as the
"little Sande"—adult women and their young female children. The "little
Sande" is led by a masked and costumed figure—the spirit Nao—who is the
incarnation of the power of the association.5 The "little Sande" is the main
public organizational means by which women are drawn into the activities of
the Sande. Each day and night women parade and dance, and as the actual
initiation approaches, the crowd of dancers becomes larger and more vocal.
Finally, the "little Sande" leads the initiates into the secret Sande bush encamp-
ment (Rosen 1983).
Initiation classes may consist of as few as three or as many as several hundred
young women, depending on the place and the year. At the start of the initia-
tion period, young women sponsored by their mothers or other kinswomen
leave the village and enter the Sande "bush"—typically a secluded forest clear-
ing. The girls remove their clothing (symbolizing shedding childhood), paint
their bodies with white clay (possibly symbolizing the lochia with which
babies are covered when they are born), and dress alike in short skirts and
beads (the uniform clothing symbolizing group rather than individual iden-
tity). During the time that they are in the grove men may not see them or
approach them sexually.
The initiation process takes a child (the word used by the Mende is a
neutral one in terms of sex) and in ritualized stages turns her into a woman.
Early in the process, when the initiates appear in public they are dressed
androgynously and behave awkwardly. The women conducting the rites be-
rate them for their inabilities, and make a show of instructing them. Later on,
initiands are called by a word meaning "virgins" or "maidens" and they are
the subject of lewd speculation by men. The final time that they are brought
to the village they are called brides and sit in a special building where they are
admired by women and receive gifts from their future husbands and visits
from kin in other villages.
During the time that the initiates are in the bush they learn songs, dances,
and stories. The stories concern both practical matters and mystical ones, and
often end with an ethical or philosophical dilemma that the girls and women
126 Priestess, Mother, Sacred Sister

discuss. The girls learn the philosophies and myths of the group, while also
learning spinning, fishing, cooking, and mothering skills. In recent years,
anatomy, sanitation, and first-aid have been added to the curriculum.
The power of Sande derives both from the secret knowledge and ritual
objects owned by the chapters, and from the founding ancestresses, to whom
offerings are made on ritual occasions. The Sande initiation ritual dramatizes
the notion that all are descended from a common ancestor and so are culturally
one family.

Genital Mutilation
We now turn to a topic I wish I could sweep under the carpet. In Sande—one
of the most "feminist" female-dominated religions (see Chapter 13)—the cen-
tral ritual complex involves clitoridectomy of adolescent women.6
I will begin this discussion with observations and analysis provided by Carol
MacCormack, the foremost feminist scholar who conducted fieldwork among
Sande women. According to MacCormack, the gender ideology taught by
Sande stresses sex distinction (women and men are clearly different), respect
for women's bodies, dependence on fellow women, preserving women's se-
cret knowledge and power, and a cultural rather than a purely biological under-
standing of fertility. Sande training stresses their value as women to society. "In
this institutional setting women dramatically pass on a strong, positive self-
image to other women" (MacCormack 1977, 98).
The central element of Sande initiation is surgery in which the clitoris and
part of the labia minora are cut away. Sande women believe that this procedure
makes women clean and helps women bear many children. Carol Mac-
Cormack (1977, 98) and Donald Cosentino (1982, 24) both interpret this to
mean that by removing the clitoris, which is analogous to the male penis,
Sande women are made totally female; any sexual ambiguity is cut away.
Sande women explain that a woman who has not undergone clitoridectomy
will not be respected; even though she is physiologically mature she will
remain in the status of immature girl. At the end of the initiation period the
girls are washed in a herbal solution. They are now women, "in knowledge-
able control of their own sexuality, eligible for marriage and childbearing"
(MacCormack 1977, 99). Symbolically, the surgery dramatizes the role of
culture—in this case Sande officials—in ensuring reproduction and defining
womanhood. In addition, according to MacCormack, "Shared pain and risk of
death from infection in initiation helps to bond initiates together into a cohe-
sive group" (1979. 32).
MacCormack's comments about Sande initiation reflect a perspective that
anthropologists call emic—an insider's interpretation in which ideas are pre-
sented from the point of view of the members of the culture. In dealing with
clitoridectomy, however, I do not believe that an emic interpretation is suffi-
cient; an etic—from the point of view of the outside researcher—approach is
also called for. I am not convinced that MacCormack's interpretation explains
why Sande women—women who belong to a well-established and powerful
 Rituals and Relationships 127
organization that has scores of means to control fertility at its disposal—opt for a
painful dismembering of women's bodies. From an emic point of view Sande
genital operations empower women and dramatize women's command of fertil-
ity; from an etic perspective clitoridectomy is a culturally condoned procedure
for mutilating women's bodies. While Sande seems, on the face of it, to proclaim
a highly egalitarian philosophy, from a Western feminist perspective, female
genital mutilation is a form of collaboration with patriarchy (Daly 1978).?
I would like to raise several possible (and complementary) interpretations
that may help reconcile these two such different views. First, Sande initiation
serves to reinforce old women's control of young women. Clitoridectomy—
performed by old women on young women—is part of the process by which
old women gain that control (Bledsoe 1980).* Thus, for old women it is indeed
an empowering procedure; for young women it is painful mutilation pure and
simple.
Second, as I suggest below (see pp. 136-38), blood rituals serve to establish
uniquely strong interpersonal bonds in nonmatrilineal societies. Because blood
ties are so enduring, Sande initiation binds women more closely to one another
than to their fathers or husbands. In other words, in societies in which the
primary social organization is along the male line, a dramatic and bloody com-
munal ritual can serve to establish equally compelling connections among
women. This female bonding has important consequences in the patrilineal and
virilocal societies in which Sande is located. In short, Sande clitoridectomy can
be seen as a somewhat feminist response to a patriarchal cultural environment.
Third, given that genital mutilation is so widespread in Africa (Sanderson
estimates that some form of female genital mutilation exists today in thirty-
two out of the forty-eight African countries; 1981, 32), it may be that Sande
initiation—in which the mutilation is carried out in a group (rather than indi-
vidually) and consists of clitoridectomy (rather than the more severe and crip-
pling pharaonic circumcision or infibulation)—is not as oppressive to women
as it seems from the perspective of Western culture. 9
My final suggestion returns to the theme of motherhood. Studying the
Northern Sudan (where infibulation rather than the milder clitoridectomy is
the norm), Janice Boddy has proposed that genital mutilation does not so much
enhance fertility as it "socializes" fertility. Genital mutilation de-emphasizes
sexuality and sexual pleasure, and so emphasizes childbearing. Seen from this
perspective, infibulation is part of a system of curbing women's sexual desire
in order to preserve their chastity in a culture in which the dignity and honor of
a family are defined by the women's sexual conduct (1989, 74). Although
clitoridectomy does not curb sexual desire or preserve chastity in the way that
infibulation does, I would argue that both procedures serve to sever maternal
from sexual functioning, and in so doing emphasize maternity. This line of
reasoning fits in well with the maternal focus of women's religions in general. I
would like here to refer back to Karen Sack's ideas concerning the low status of
wives cross-culturally (see Chapter 3). If Sacks is correct (and I think that she
is), given that women as mothers consistently have a higher status than women
as wives, it may be that women in certain situations opt to downplay their
128 Priestess, Mother, Sacred Sister

identities as wives in order to enhance their identities as mothers, even at the

expense of sexual enjoyment. 10 By splitting off maternity from sexuality,
clitoridectomy serves to raise the status of women through accentuating the
part of women's identities that is most culturally esteemed.
Truthfully, I do not know if any of these interpretations are correct. What I
can say with more certainty is that Sande initiation has meaning because of the
ritual exclusion of men (cf. Jedrej 1976). In the very specific context of Sande,
genital mutilation highlights and dramatizes the fact that this is a women's
secret society and that the physical fact of femaleness has social meaning. It
may well be that Sande genital mutilation is the battle ground on which the
diametrically opposed implications of women's motherhood—male ideologies
of female inferiority and female self-knowledge as powerful and complete—
vie for dominance.

Zar Initiation
On the face of it, Sande is the only women's religion in our sample that provides
a group initiation ritual for most local women. It is tempting to say that what
differentiates Sande initiation from, say, zdr initiation is that the former occurs
fairly automatically at a certain stage in the life cycle, whereas the latter is a ritual
solution to the physical, social, or psychological disturbances suffered by particu-
lar ill-adjusted women. However, studies suggest that at least in some communi-
ties almost all women are initiated into the zdr cult during their childbearing
years. Therefore, I am led to speculate that zdr initiation may not be so different
from Sande initiation. Anita Spring's observation regarding women's reproduc-
tive rituals among the Luvale of Zambia seems to hold true for zdr initiation as
well: These rituals do not occur randomly or in response to individual sickness,
but are coordinated with life and reproductive cycles (1976, 112).
Once a woman is diagnosed as afflicted by spirits, she is believed to be in
danger until the specific possessing zdr is identified. Therefore, neighbors will
not leave the afflicted woman alone. They sing, dance, and drum in front of
her, promising to bring her whatever she desires in an effort to make her react.
As in several other women's religions, healing comes about through joining
the group.
At the beginning of the zdr initiation ritual, the initiate is confined to a dark
room for several days. Her husband may not see her and she may not wash.
Later, she washes and is reintegrated into the community. The ritual includes
singing, drumming, dancing, and incense (techniques to invoke trance),
fortune-telling and prescribing cures for the audience, animal sacrifice, and a
communal meal. During the initiation the patient is referred to as the bride of
the zdr, and the entire ceremony has ritual, symbolic, and linguistic parallels to
weddings. The core of the zdr audience is women who have been initiated in
the past, and so are obligated to continue attending rituals to appease their
spirits. The result of initiation is the establishment of a community of women
who are or were fellow sufferers.
 Rituals and Relationships 129

Female Initiation: Cross-Cultural Perspectives

Anthropologists and historians of religion have written a great deal about
boys' initiation, but very little about girls' initiation. In his classic Rites and
Symbols of Initiation, Mircea Eliade concluded that girls' initiation ceremonies
are less widespread, less developed, and less dramatic than boys'. Typically,
each girl is initiated at the time of the individual, discrete physiological event of
first menstruation, whereas boys are initiated together with the other boys of
the same age group (1958, 4iff). The implication is that boys' initiation has to
do with creating "male bonding" whereas girls' initiation is concerned with
more personal matters.
In some cultures initiated women form a female secret society (like the
Sande). Sometimes these secret societies terrori/e, threaten, or strike men
whom they encounter, just as men's secret societies often terrorize women.
According to Eliade, this is because women's mysteries might be endangered
by the presence of men. "The tension is always between two different kinds of
sacrality, which are the foundations of two different and polar world views—
masculine and feminine" (1958, 80). Although I greatly admire Eliade's work
(and even see my own work as growing out of his approach) I find it hard to
accept his assumption that men and women have sacred modes that are "polar
opposites" of one another. All recent psychological studies of sex differences
show that men and women are far more similar than different from each other
(Hyde 1990). And indeed, Eliade's phenomenological approach is not popular
among contemporary anthropologists who object to broad generalizations not
grounded in specific cultural contexts.
Bruce Lincoln has also looked at female initiation from a comparative per-
spective. He asks the critically important question: Who is in charge of initiat-
ing girls: men or women? "It is society as a whole that acts, and thus the
initiand experiences both the repressive force of men (who may cut her, rape
her, or simply force her to run, work, or stay up all night) and the support of
her fellow women (who may dress and adorn her, bring food to her, or keep
vigil with her)" (1981, 93). Although Lincoln does not make the point explicit,
the implication seems clear enough: Women's initiation creates or perpetuates
networks of women linked on the one hand by mutual social support and
shared experience of sacrality, and on the other hand by a sense that men are the
"other."
Initiatory segregation of girls (i.e., elaborate initiation rituals) is most com-
mon in cultures where women's status is relatively high, and the rituals often
express joy at menarche. Bruce Lincoln has reached the same conclusion: "My
general sense is that the presence of women's initiation in a given culture is a
mark of the importance of women within the culture and of the culture's
willingness to recognize this publicly and institutionally" (1981, 91).
Another well-known cross-cultural study of female initiation (or, more prop-
erly, female puberty rites) was carried out by Judith Brown (1963). Brown
reached two conclusions with important implications for our present inquiry.
First, female initiation rites occur in societies in which women do not leave
130 Priestess, Mother, Sacred Sister
their natal homes after marriage, because, according to Brown, in these in-
stances there is more of a need to proclaim their changed status at adulthood. If
the bride stays at home instead of moving away, ritual encourages family
members to shift their view of her from girl to woman. Second, female initia-
tion rites occur in societies in which women make a notable contribution to
subsistence activity, because, Brown says, their future importance to the life of
the society requires that girls be given special assurance of their competence. I
am impressed by Brown's findings, but not by her interpretations. In light of
Chapter 2's discussion of the contexts in which women's religions occur, I
would argue that the occurrence of female initiation rites in matrilocal soci-
eties reflects the presence of an ongoing female culture that is celebrated by
women who control substantial economic means.11

FUNERAL RITUALS
Cross-culturally, women are active in mourning rituals (see, e.g., Seremetakis
1991 on Greece). The goals of mourning rituals are typically threefold: to
provide social support for the mourners, to ease the passage of the departed,
and to symbolically explore and explain the existential meaning of death and
separation. Death rituals are often highly emotionally charged. Laurel Kendall
describes the part of Korean kut known as chosang-gori, the ancestors' sequence,
when the ancestors of the house appear in the person of the possessed shaman.
"Manifesting a sobbing ancestor, the shaman clutches at the shoulders of a
child, grandchild, sibling, or spouse and laments death and separation. Tears
course down the cheeks of both shaman and client" (1983, 97).
In most women's religions there is a conscious awareness that the most
significant and permanent human relationships are those that radiate from
mothers. This is the meaning of the matrifocality we discovered in Chapter 2.
In cultures in which female-dominated religions occur women do the "work"
of nurturing relationships, and that "work" is lauded, spiritualized, and often
institutionalized. Women, whose interpersonal bonds are deemed culturally,
spiritually, and personally significant, tend to be preeminent mourners. In
death rituals, bonds with ancestors and descendants are pondered, mytholo-
gized, dramatized, and ultimately either sundered or amplified.
Mourning rituals have to do with entanglement and separation. In the fol-
lowing paragraphs we will see how two women's religions use diametrically
opposed strategies to express and work out these emotions.

Tensho-Kotai-Jingu-Kyu:
Cutting the Ropes that Bind
Our first example is extraordinary in that it highlights severing relations with
the dead, rather than strengthening them. Tensho-Kotai-Jingu-Kyu, a new
Japanese religion, teaches that ancestors are capable of harming their descen-
dants. Avoidance of such harm is not through propitiation or commemoration,
but through symbolic severing of the ties binding ancestors to their desccn-
 Rituals and Relationships 131
dants. Members of Tensho relate that they were attracted to the religion be-
cause it promises alleviation of suffering (particularly ill health), family trou-
bles, and economic problems. Suffering is believed to be the consequence of a
variety of negative supernatural interferences, including malignant ancestors.
All ancestors who have not been redeemed through Tensho practice are be-
lieved to suffer intense agony, and to cause illness and misfortune for their
descendants because of "their attachment to living people" (Kerner 1976, 212).
Tensho teaches that persons in this life are linked by karma both with their
ancestors and other household members (including dead fetuses) and with
other compatriots from their own past lives. "The curing of illness caused by
spirit possession is accomplished by breaking the ties between suffering indi-
viduals and the afflicting spirits" (Kerner 1976, 213)—through "cutting the
karma" of the afflicted person. Mrs. Kitamura, the founder of Tensho, was
believed to possess a special capacity to release individuals from their karmic
bonds. Devout members of the sect have the ability to a lesser degree, and all
members can perform this severing themselves through ritual, prayer, and
dance.
Central to Tensho belief is the destruction and abandonment of the tradi-
tional ancestral soul tablets and the household "god shelf" on which they are
normally placed in Japanese households. Ancestors are no longer considered to
participate in the ongoing life of the family. Once redeemed (and that is what
one must do in order to avoid personal suffering), ancestors no longer are
linked to their descendants.
Parallel to the severing of ties with ancestors is severing of ties with living
friends and relatives who are not members of Tensho. Members are warned
not to interact with unconverted family members. They are told that this is
necessary for their health and well-being. Sect marriages are often arranged
between Japanese members and members living abroad, which means that one
of the spouses is forced to leave his or her family. Children of members are not
infrequently given in adoption to totally unrelated members. Members living
abroad often adopt Japanese children. The newly adopted children are expected
to leave their natural parents free of sentimental attachment to them. "Tensho
therefore redefines the nature of social ties" (Kerner 1976, 215).
In sum, even though the content of Tensho's relationship with ancestors is
different from that of other women's religions, the extent of its concern is
most typical.

Black Carib Mourning Rituals:

Binding the Ropes that Were Sundered
Relationships among Black Caribs continue long after death. The recently
dead rely on the religious practices of their descendants to allow them to make
progress on the road to the highest heavens. Spirits may stay on earth if their
relatives fail to perform the appropriate rites. "The road to the world of the
dead is long and arduous, and the spirits will ask for more baths to refresh
them and offerings of food to restore their forces before they arrive there"
132 Priestess, Mother, Sacred Sister

(Coelho 1955, 145). Along the way, the spirit stops frequently to greet friends
who died earlier, but who have been unable to travel as rapidly. The living
family is kept informed of the journey's progress through omens and dreams
until the spirit finally joins the "blessed souls." The family dead who have been
admitted to the category of "blessed souls" are under obligation to advance the
earthly affairs of their descendants (Coelho 1955, 140-146).
Black Carib women mourn kin for six months or a year (depending on the
relationship). Older women who outlive many kin spend a significant portion
of their adult lives in mourning. Mourning includes refraining from drinking
alcoholic beverages, dancing, quarreling, and wearing bright colored clothes.
Men and children do not mourn formally (Kerns 1983, 156).
Among the Black Caribs, death and sickness are not suffered alone. Large
crowds are required at rituals for the dead, and ancestors who are dissatis-
fied because of a small turnout may demand that the ritual be repeated.
Virginia Kerns has noted that a large crowd makes both the organizer and
the deceased feel appreciated (1983, 173). Rituals for the dead are designed
to meet the needs of the dead and thus protect the living from illness, mis-
fortune, or death at the hands of neglected and resentful spirits. Offerings
and prayers—centered on remembrance—prove to the dead that the living
care about them.
"Ritual is so common as to be almost an aspect of daily life in Black Carib
villages" (Kerns 1983, 147). In one village Kerns found that over a one-year
period, a collective death-related ritual of some kind was held on 102 days.
(This does not include the Roman Catholic holidays or other communal
calendrical holidays.) Ancestor rituals are carried out both on predetermined
anniversaries and in response to a death or sickness thought to be caused by the
spirit of a dead person. Only a few rituals are absolutely obligatory.
Although Black Carib death rituals have a solemn purpose, there is a festive
ambience, which includes song and dance. Virginia Kerns' informants told her
that they enjoy the rituals for the dead (1983, 170). The rituals involve a great
deal of preparatory work—mostly involving food which is shared among
female kin and neighbors. Black Carib mothers expect to be thanked for food
by young children—and to be financially supported by grown children. Ances-
tors have the same expectations. Rituals express gratitude to the ancestors and
provide them with offerings of food and rum.
Kerns (1983, 154-156) describes the preparations for the ninth-night wake.
Early in the morning women grate coconut and express the liquid to make
bread. Other women mix flour, coconut milk, and yeast, and knead it. Addi-
tional women prepare dried coconut husks and firewood to bake the bread. It
takes most of the morning to make the large quantity necessary for the cere-
mony. Here is how Taylor describes the food at an ancestral ritual: "The cooks,
relaying one another from the hall, have been busy all night preparing the
ancestors' favorite dishes . . . and by now [the second day of the ceremony]
the various contributions have been set out and tastefully arranged on the
offering tables. . . . Besides the crabs and other sea food caught by the
adugahatiu, there is meat from the cows that have been sacrificed and from the
 Rituals and Relationships 133

pigs that have been slaughtered, together with corn-cakes, rice, chickpeas, and
quantities of cassavas (areba) and manioc beer" (1951, 125).
At the dtigti ritual the ancestral spirit is the guest of honor at a feast that lasts
for three days and nights of drinking, eating, and dancing in the temple.
Preparations may last for as long as a year. Enormous quantities of food for
sacrifice and eating are purchased. All food and drink must be offered to the
dead, or shared and distributed among the living; none may be retained by the
sponsors. If the sponsors are "greedy" or "stingy" they will fall ill—victims of
ancestors.
At rituals there is also food (bread, rum, and so forth) intended to be eaten
by participants. The sharing of food and strong drink with the living and dead
is a central ritual act. "It provides the quintessential expression of trust and
kinship, and it tangibly demonstrates generosity to the living and gratitude to
the dead" (Kerns 1983, 149).

FOOD RITUALS
Emphasis on food and food preparation is one of the clearest and most com-
mon themes in the ritual systems of women's religions. Cross-culturally, food
is an especially sacred symbol because it is ingested—incorporated into the
body of the believer.
On the Ryukyu Islands, most rituals include offerings of food and incense to
the kami (gods) or to the ancestors. Alcoholic beverages distilled from rice,
millet, or potatoes are also offered. Formerly, one of the ceremonial tasks of
priestesses was to prepare the alcoholic beverages; fermentation was induced
by their chewing the grain and allowing saliva to act as a catalyst.
Food offered to the ancestral spirits and kami is similar to, yet more varied
than, food eaten by mortals. Arrayed at shrines are dozens of individual bowls
containing small amounts of various delicacies prepared by the priestesses.
Ouwehand (1985, 132-134) lists the following items (I have only cited a few):
bowls of holy water from a shrine well (each bowl with different herbs and
leaves), incense, rice cakes, dried fish, rice water prepared in a special way, a
mixture of garlic, herbs, vinegar, bean sprouts, and sesame seeds, crab legs,
salted crabmeat, rice wine, fish soup, and a soup made with bean curd. After
large ceremonies, priestesses and their entourage spend the afternoon feasting,
drinking, and dancing. Food offerings are never wasted and leftovers are
brought home. Within the home, women make offerings to the ancestors and
the hearth kami at least twice a month. These offerings coincide with the
morning meal and are consumed by the family.
Turning to a very different example, for the Shakers, "The only Eucharist
was a collective consumption of what had been collectively produced; and for
them it was these economic practices which day by day fashioned and refash-
ioned the Family body in which the blood of the Spirit circulated. Thus it was
that the Shaker meal transformed communism into a religion. It was their
custom to eat in silence and to kneel in silent prayer before and after every
meal" (Desroche 1971, 210).
134 Priestess, Mother, Sacred Sister

Many Korean household rituals involve food offerings. The most basic
ritual kosa consists simply of leaving offerings in particular places, bowing
stiffly, rubbing the hands together, and making a personal petition. The types
of foods offered as kosa are, in the eyes of the women, snacks or delicacies. "As
a tasteful accompaniment to their wine, she [the housewife] gives them each
[each god] a dried fish" (Kendall 1985, 115). Kosa is offered after bad dreams,
when a child is ill, and at calendrical rituals. Certain gods, or more important
occasions, merit more elaborate dishes. After the food has sat out for a half
hour or so, it is distributed to family and neighbors. For Korean women's
household rituals, according to Lee (1984), the foods must be prepared person-
ally by the housewife. The fact that in the home the senior woman offers kosa
shows that the house is women's space and women represent the household
before the household gods (Kendall 1985, 124). Men do not make offerings to
household gods, and their offerings to ancestors are often prepared by
women.12 Moving from domestic ceremonies to more public ones, ethno-
graphic descriptions of the shaman's kut show an abundance of food and drink.
Among Burmese women, the basic obligation to nats is to provide them
with food offerings; such offerings are made at every life crisis. Precise rules
govern which foods to offer at which occasions. When a child is seven years
old thirty-seven small fish are fried and offered to the Thirty-Seven Nats. The
various domestic nats are known to have food preferences. For example, the
Taungbyon brothers like betel nuts, while Ma Negale likes duck eggs. At
wedding ceremonies bananas and pineapples are offered to the nats of the
bride's and groom's families. Sometimes guests eat the offerings. At regional
nat rituals all the natkadaws (shamans) bring nat images which are lined up and
the nats are offered bits of food (M. Nash 1966).
Also in Sande ceremonies, libations are poured to the ancestresses, and food
offerings are prepared for them. In the Black Spiritual churches of New Or-
leans feasts are prepared in honor of saints, and the saint's altar is decorated
with, (e.g., in the case of St. Joseph's feast) rich cakes, cross-shaped Italian
breads, baskets of fruit, bottles of wine, seafood (including jumbo shrimp,
soft-shelled crabs, crawfish, stuffed oysters, and crabs), and other delicacies
(Jacobs and Kaslow 1991, 118). Preparing feasts necessitates major cooperation
from other church members, and an organizer will request and cajole dona-
tions from friends and relatives.
A rather different use of food has been described among Black Carib
women. According to Kenyon Bullard (1974), Black Carib men and women
have opposite financial agendas: she wants him to give her money for the
household and children; he wants to spend money on alcohol, other women,
and gambling. Therefore, women use ritual means to ensure a man's continu-
ing presence in and economic contributions to the home. A woman's "love
magic" involves secretly putting some of her bodily secretions into his food.
Knowing that women can and do perform such rituals creates an ambivalent
attitude toward food. Men love to eat and are dependent on women for food
(women control food preparation and distribution), but they are scared of
 Rituals and R 135

what women have put into the food. Highlighting the ways in which these
food rituals enhance interpersonal relationships, Bullard emphasizes that "love
magic" encourages men to follow through with co-residence and financial
support after a casual sexual relationship.

Northern Thailand: A Case Study

In Northern Thailand food offerings to the phii puu njaa (spirits) are made on a
variety of occasions. At regular intervals, the spirits are asked to forgive any
offenses committed against them and to continue to protect the family they are
associated with. When a daughter of the household marries, offerings are made
to gain the approval of the spirits of the bride and groom. The essential
element of the marriage ceremony is offerings made to the spirits to ensure that
the spirits of the bride's cult group cause no harm to members of her group
because a man from another cult group has joined the household. Spirits are
propitiated when members of the family have fought among themselves, mis-
treated the shrine, or failed to perform the proper offerings to the spirits.
Ongoing loyalty to the family/cult group is encouraged by the belief that if
the spirits are abandoned, members of the family will turn into witches. Loy-
alty to the spirits means loyalty to the group itself. At certain vital rituals it is
required that all member households are represented, or at least that all mem-
bers who live within walking distance attend. After the offerings to the spirits
are made and accepted, the cult members share the leftover food. The commu-
nal meals in which both cult members and spirits participate contribute to the
sense of group solidarity.
The most important annual event for the group is an offering made soon
after plowing has begun in the ninth month. At this time of year demands for
labor are made on close kinsmen. It is also the time of year when the health of
the villagers tends to be poorest, since rice is in short supply and the beginning
of the wet season gives rise to various fevers. Kin group solidarity is a
priority—it is believed that conflicts may lead to sickness that is interpreted as
punishment by descent group spirits for failure to cooperate.
This is how Andrew Turton describes the annual offering: "The sacrifice
[offering] is made in the stem house and is attended by all women of the group
and some or all of their unmarried children of both sexes. Affmes and married
consanguineal males may attend but seldom do. Each member household
contributes a chicken (in some cases pork), rice, and the basic ritual offering,
common to nearly all types of sacrifice, of a small banana leaf cone containing
flowers, popped rice, candles and possibly incense. The food is cooked, offered
by the ritual officiant to the spirit and then eaten by those present. If a member
household cannot be represented at the sacrifice a small parcel of the cooked
food will be sent to it afterwards" (1984, 278). During the annual offering
(which lasts for several hours), the house and housesite are ritually closed by a
white cotton thread and gates and shutters are closed, signifying the group
identity of the participants.
136 Priestess, Mother, Sacred Sister

Animal Sacrifice versus Communal Meals

In most cultures, women prepare food. Thus it is not surprising that food
rituals are important to women. Food is a resource that women control, and
food rituals sacralize women's everyday activities of cooking and serving. By
sharing food with the gods, profane work becomes elevated to sacred ritual.
Even in male-dominated religions women are often responsible for preparing
ritual foods (although the public food ceremony is typically conducted by men;
e.g., Jewish women cook for the Sabbath and men recite the kiddush).11
Food rituals in women's religions share several significant characteristics. To
begin with, the foods used in female-dominated religions are not served in
small symbolic portions. No Catholic Church style Eucharist ceremonies have
been reported here. Instead, we have seen large quantities of elaborately pre-
pared food. The food rituals are public and communal. They involve both
supernatural and natural domains—both gods and people eat. Food rituals
create bridges between this world and other worlds. Sharing food emphasizes
good relationships with the living and the dead and the spirits, an emphasis in
line with the other sorts of rituals described in this chapter.
The classic scholarly discussion of food rituals is found in W. Robertson
Smith's study of ancient Semitic religion. Smith concluded that animal sacri-
fice was the earliest way that humans conversed with the gods; cereal offerings
were simply an accompaniment to animal sacrifice.14 In ancient Israel all animal
slaughter was seen as ritual sacrifice, and meat was never eaten except as a
religious act. The sacrificial meal was an act of communion, a social act in
which people and God ate together. Eating and drinking collectively was a
symbol and confirmation of fellowship and mutual social obligations. Kin are
people with whom one eats, and who are sealed together in a bond of mutual
obligation. When an animal is slaughtered and kin assemble to eat it, the
communal eating is de facto a ritual of solidarity. Why is this so? Smith sug-
gests that kinfolk, God, and the sacrificial victim were believed to share the
same blood. By ingesting some of the life blood of the sacrificial animal, the
blood bond among people is strengthened. At a more advanced stage of cul-
tural development, the part of the animal that goes to God came to be burnt.
As people acquired a more transcendent notion of the nature of the divine, the
gods came to be understood as inhaling the smoke rather than eating the meat.
In light of Smith's emphasis on animal sacrifice in ritual meals, it is signifi-
cant that in women's religions food rituals rarely include animal sacrifice; in
those cases where there is animal sacrifice it is a rather minor part of a larger
ritual that includes many other elements; and in no cases have we seen burnt
offerings sent up to the heavens. Instead, in women's religions cooked and
mostly vegetable foods are shared by members of the group. 15
Male-dominated religions in patrilineal societies wrestle with the frightening
knowledge that the most significant societal bonds (kinship) are artificial
(patrilineal). Why? Because in patrilineal cultures the blood ties among kin are
hypothetical: Men postulate "common blood," but uncertainty remains. 16 This
is why (and here I base my ideas on those of Robertson Smith) sharing the
 Rituals and Relationships 137
blood of the sacrificial animal (originally the totem—which was believed to be
a relative of all members of the clan) has important ritualistic connotations—it
dramatizes, indeed it forges, the blood ties between the members of the group.
By ritually killing an animal and then sharing in its consumption, groups of
men ingest common blood. In many male-dominated religions, kinship is
"created" by sharing the blood of animals.17
A similar argument has been made by Nancy Jay who reviews a range of
African societies and concludes that there is indeed an affinity between blood
sacrificial religion and patrilineal social organization. Sacrifice serves as evi-
dence of patrilineal descent and functions to constitute and maintain patrilineal
descent systems (1985, 285). She brings examples of societies such as the Nuer,
Dahomey, and Tallensi among whom the actual word for patrilineage trans-
lates as "people who sacrifice together." Among the cases Jay cites is the West
African Yako who organize themselves into both patrilineal and matrilineal
descent groups. Only the patrilineages practice sacrifice. I quote here from
Jay at some length:

Paternity never has the same certainty [as matern ], and birth by itself cannot be
the sole criterion for patrilineage membership. Nor can any enduring social struc-
ture be built only upon the shifting sands of that uncertain relation, biological
paternity. Jural paternity (paternity in terms of rights and obligations) and biologi-
cal paternity may, and often do, coincide^ but it is jural paternity that determines
patrilineage membership. Some sacrificing societies, such as the Romans or the
Nuer, distinguish between biological and jural paternity in their vocabulary, for
example, the Latin distinction between genitor and pater. It was the pater who was
significant sacrificially. (1985, 290-291, my emphases).

In women's religions, which as I have shown in Chapter 2 tend to occur in

matrifocal or matrilineal societies, the bonds between people are already empiri-
cally "in the blood" and animal blood is not needed to strengthen, prove, or
dramatize that bond. Most women's religions do not confront the sorts of
problems that can be solved by animal sacrifice. Significantly, in Northern
Thailand the annual sacrifice occurs at a point in the agricultural cycle in which
blood (kinship) ties are insufficient: the labor of outmarrying males is sorely
needed at this time. Even so, in the Northern Thai case the emphasis seems to
be on the sharing of a mostly vegetable meal, not on the spilling of blood.
Similarly, among the Black Caribs animal sacrifice seems to be a rather periph-
eral part of ancestor rituals; much more emphasis is given to the preparation
and serving of the ancestor's favorite foods.
In two women's religions, however, there are spectacular blood rituals.
First, at zar cult initiations animal sacrifice is followed by a communal meal.
Second, in the Sande secret society the initiate sheds her own blood—through
clitoridectomy—in the communal initiation ritual. The only two women's
religions to perform blood rituals carry out these rituals both in the context
of initiation and in the context of patrilineal and patrilocal societies. Sande
and zar women, like men in Robertson Smith's ancient Near East, create
138 Priestess, Mother, Sacred Sister
interpersonal bonds through blood rituals. The case of Sande is particularly
instructive: Initiated women become Sande "sisters."
Sande and zdr are the only two instances of women's religions (from
among our twelve key examples) that occur in patrilineal, patrilocal societies.
Nineteenth-century America and England were bilateral and neolocal; most
other women's religions occur in matrilineal or matrifocal societies. Thus,
kinship (blood) would not be problematic because kinship ties are self-
evident. In most women's religions creating or strengthening blood ties is
not an issue. What food rituals do dramatize in women's religions is social
ties, and that is why elaborately cooked food is so important. Offering deli-
cious and attractively prepared delicacies to the ancestors or gods is a way of
socializing or domesticating the deity.18
We may ask why the Shakers who self-consciously and articulately created
synthetic "families" of brothers and sisters did not use some sort of blood ritual
or animal sacrifice. I believe that the answer is given by the Shakers them-
selves: They not only deny the importance of blood ties (biological family) but
see in blood ties much of the root of suffering in this world. Shakers wanted
neither biological families nor blood sacrifice. It is by transcending biological
family that Shakers find spiritual and material contentment. As we saw above,
Shaker food rituals contribute to that process, but not through ritual blood
sacrifice, and not even through the common Christian symbolic sacrifice of the
Eucharist.19
The one question that remains is why in women's religions food for the
deity is eaten by human participants and not sent up as burnt offerings. I
believe that this is consistent with the interpersonal emphases of women's
religions. In the elaborate food rituals of women's religions the accent is on
socializing with the gods. Immanent deities are invited to partake of tasty
treats; smoke is not sent away to transcendent deities. In women's religions
neither people (shamans) nor offerings (food) ascend; rather, gods, spirits, and
ancestors descend and join in the communal, human experience. This pattern is
most clearly articulated in rituals of spirit possession, which will be explored in
Chapter 8.

LIFE-CYCLE RITUALS
Anthropologists and historians of religion categorize rituals into three major
groupings: rites of solidarity, rituals of affliction, and rites of passage. In this
chapter I have shown that almost all the rituals of women's religions fall within
the rubric of rites of solidarity: dramatizing and strengthening interpersonal
bonds is an important ritual focus of women's religions. In the previous chap-
ter I demonstrated that rituals of healing that highlight and resolve suffering
and affliction have a momentous role in women's religions. Turning to the
third category of rituals—rites of passage—we are confronted with something
of a puzzle. Rites of passage are relatively rare and unelaborated in women's
religions. Maternity is consequential and respected, yet birth rituals are close
to non-existent. To take one example, among the Northern Thai pregnancy
 Rituals and R 139

and birth are enveloped in ceremony, but these ceremonies seem to have noth-
ing to do with the female-dominated matrilineal spirit cult (see Hanks 1963). I
have found no discussion of birth rituals in the ethnographic literature on most
of the other women's religions, and given the excellence of the ethnographies
available, I do not believe that this absence reflects some sort of anti-natality
bias on the part of the ethnographers. Birth is simply not a focus of ritual in
most women's religions. Maturation rituals also receive little attention in
women's religions. Ironically, contemporary Jewish and Christian feminists
are constructing rituals to celebrate women's physiological passages, yet most
female-dominated religions ignore menarche, menstruation, and menopause.
Marriage ceremonies also play a very minor role in female-dominated reli-
gions. And finally, despite women's fame as mourners cross-culturally, with
one or two exceptions (especially Black Caribs) we do not find that female-
dominated religions particularly emphasize death or funeral rituals.
A number of factors taken together can explain the remarkable absence of
life-cycle rituals. To begin with, women's religions do not seem to pay very
much attention to women's bodies in any context. As I argued in Chapter 3,
although motherhood is consequential in these religions, what receives notice
is the social aspects (nurturing) rather than the physiological aspects (lactation,
pregnancy) of maternity. Nurture, a lifelong process, does not lend itself to
rites of passage. Similarly, as I argue in Chapter u, old women are often
leaders in women's religions, but the definition of "old" is a social one rather
than a physiological one: Women who are freed of household responsibility
and who have accumulated religious wisdom are leaders, menopause does not
seem to enter the picture.
Could it be that the emphasis on women's bodies in male-dominated reli-
gions has more to do with men's concerns than with women's? All the so-
called "unique" or "exceptional" female physiological processes are neither
unique nor exceptional from the perspective of women. It is men who would
find menstruation anomalous; women would more likely take it for granted. I
am reminded here of an interview that I conducted with a post-partum woman
at an Israeli hospital in the context of a research project on the cultural construc-
tion of childbirth. We had been discussing whether birth is a miracle when my
informant quite severely put me in my place, telling me: "Well the baby is in
there, isn't it? It has to come out somehow. What's the big miracle about that!"
(see Sered 1991). Women's religions are characterized by concern with social
ties, not with biological facts. Not surprisingly, other women whom I inter-
viewed in the course of the same project stressed that the miracle of birth was
not the physiological process but the opportunity to get to know and take care
of a new human being. To quote one more informant, "A miracle, yes, when I
saw how he looks around, how he was a person already."
The absence of birth ceremonies in most women's religions stands in con-
trast to for example, rural India, where birth is highly ritualized (see Wadley
1980). I suspect that the de-emphasis of blood taboos in female-dominated
religions (see Chapter 10) partially accounts for the de-emphasis of birth ritu-
als. In many male-dominated religions birth rituals focus on distancing spirits
140 Priestess, Mother, Sacred Sister

attracted to childbirth blood, or disposing of the "polluted" afterbirth. Let me

add that an alternative explanation for the lack of elaboration of childbirth
rituals is Starhawk's suggestion that "birth itself is such a powerful experience
that it doesn't seem to require a lot of ritual . . . when it's happening—it just
takes over, creating its own ritual" (personal communication 1992).
The arguments I have just made explain why the biological events in
women's lives may not receive much ritual notice in women's religions. It does
not explain why wedding ceremonies are marginal in women's religions. Mar-
riage is the social tie par excellence, and as such we would expect it to receive
ritual recognition in women's religions. It is useful here to turn to the excellent
theoretical work on reproductive rituals carried out by Karen Paige and Jeffrey
Paige. Paige and Paige argue that the elaboration of reproductive rituals has to
do with men's interest in controlling women's reproductive power. Rituals
accompany the pivotal times in which women's reproductive potential is trans-
ferred from one group of men to another.
Women's religions do not ritualize the transfer of rights in women's
bodies from one group of men to another. This is not to say that there is no
marriage in societies in which female-dominated religions occur; with the
exception of the Shakers all women's religions are made up primarily of mar-
ried women. However, female-dominated religions do not, for the most part,
feature rituals that legitimate men's control of women's bodies. (We will have
more to say about this in Chapter 13.) I find it significant that one of the very
few women's religions to even offer a wedding ceremony is Spiritualism, in
which the wedding ceremony is modest, unembellished, and totally reciprocal.
Rights over women are not reallocated among men in Spiritualist weddings
(see Ward 1990).
Two women's religions do include dramatic life-cycle rituals, and we have
already dealt with both in some detail in this chapter. Among the Sande pu-
berty and childbirth are hubs of ritual activity. I have already suggested an
explanation for this somewhat unusual (in terms of women's religions) situa-
tion: Sande rituals create blood ties among women in a patrilineal society. In
other words, among the Sande there are exceptionally compelling social rea-
sons for life-cycle rituals.
Among the Black Caribs the dead lie at the heart of the ritual world. To my
mind, the most interesting feature of the Black Carib mourning rites is their
duration. These mourning rites are not rites of passage: they do not take place
immediately after death, nor do they ritualize the passage of the deceased into
the next world. Quite to the contrary, these rituals continue for years after the
death of a family member, and highlight the lack of passage of the deceased.
These rituals are aimed at retaining the deceased as an ongoing presence among
the living.
The absence of rituals that underscore change fits in with the proliferation of
rituals that underscore continuity in women's religions. This is consistent with
Nancy Chodorow's claim that female psychological development in general is
more continuous than male development. In line with Chodorow's ideas, I
suggest that rites of passage may better reflect male spiritual and emotional
 Rituals and Relationships 141

patterns than female ones. We will return to these ideas in greater detail in the
next chapter.

Notes
1. Family tombs traditionally were the focal point of the social unit; upper-class
tombs were among the most prominent features of the landscape. Many of these were
shaped in domes, said by some to resemble a turtle's back and by others to represent the
womb from which all men must come (Kerr 1958, 218).
2. Max Weber bears a large share of the responsibility for the idea (or for "prov-
ing" the idea) that highly ritualistic religions are not as "good" as other religions. He
claimed that pure ritualism is not very different from magic, and may even lag behind
magic on an evolutionary scale. His argument is that rituals have no lasting effect
upon the participant—it is an "essentially ephemeral subjective state [that] is striven
after" (1966, 152).
3. Not all women's religions are characterized by ecstatic and exciting rituals.
Okinawan ritual is highly stylized and may strike the observer as purely perfunctory.
According to Lebra (1966), Okinawan rituals emphasize proper performance and deco-
rum, rather than sincerity and emotionalism. Typically, prayer consists of chanting,
offerings, and softly rubbing the hands together.
4. This was true both in the permanent Shaker settlements and when Shaker leaders
were away from home on evangelical tours. "By supplying hospitality and continuous
informal evangelism, the Shakers added another dimension to the evangelistic ef-
fort. . . . [Ijnstead of relying on traditional notions of church discipline to organize
converts, they simply incorporated families of new Believers into the traveling commu-
nity on the spot" (Marini 1982, 93).
5. During certain ceremonies, Sande ancestresses appear in the form of masked
figures. The masks are highly sacred, and members and nonmembers alike express
reverence and awe toward them. If a man molests a Sande girl or infringes Sande rules,
a masked Sande member arrests him. While wearing the mask of the Sande she personal-
izes the spirit or protector of the Society and everyone obeys her. Although everyone
knows that "regular" people are underneath the masks, it is forbidden to watch the
Sande "ancestress" unmask. Sande masks are representations of complex spiritual and
mythological messages (Richards 1973).
Okinawan priestesses also wear (or wore) masks at certain ritual occasions. At annual
rites in which deities visit the village, the deities are sometimes represented by villagers
wearing costumes and masks.
The previous generation of anthropologists believed that throughout the world men
use masks as parts of rituals which act out why women have been rightly deprived of
the authority that resides in the ownership of masks. According to this theory, men
created secret societies that used masks as aggressive means to take power from early
matriarchies. These men's secret societies terrorize women by using masks representing
spirits. Recently Elizabeth Tonkin (1983) and Henry Fernet (1982) have made the argu-
ment that women's role in men's mask rituals is not as passive as ethnographers have
assumed; women's screaming in fright on seeing the masks is ritualized and symbolic,
not real fright. Neither Tonkin nor Fernet, however, describe rituals in which women
don masks. Further research is needed to uncover the meaning of women's masking
rites.
6. I am limiting my comments here to clitoridectomy, but many of the questions
142 Priestess, Mother, Sacred Sister

that I raise are even more relevant regarding infibulation, a far more severe procedure
that often causes permanent pain and dysfunction.
7. Feminist scholar Mary Daly has looked at African genital mutilation in a cross-
cultural context of atrocities committed against women's bodies. In common with
Indian suttee, Chinese footbinding, and European witchburnings, genital mutilation
betrays an obsession with female purity, a pattern of proliferation from an elite to the
upwardly aspiring lower echelons of society, a ceremonial setting that misfocuses atten-
tion away from the actual cutting, an ideology that says that for a woman to be
normative she must suffer the slicing away of part of her anatomy, and the legitimation
of the whole procedure by outside scholars who treat it as a "custom." Daly indicates
two other characteristics that are particularly relevant here: the total absence of men at
the execution of the mutilation, and "the use of women as token torturers" (1978, 163).
Yet, as Daly point out, the women themselves say that they carry out the mutilation
because no man will marry a woman whose genitals have been left intact (also see
Lightfoot-Klein 1989). Daly leads us to question the facile observation that women
perpetuate genital mutilation; she suggest that the women themselves realize that the
real perpetrators are the men.
8. Kennedy sees excision (genital mutilation) among the Nubia as reflecting and
strengthening social principles of sex separation, male dominance, and age-generation
dominance. He argues that the mixture of social support and pain and fear instilled by
these procedures are "ideal for intimidating a child, suppressing sexual and aggressive
Oedipal desires, and inculcating docility" (1978, 162 my emphasis). According to Ken-
nedy, "Sexual fears, anxieties concerning fertility responsibility, and a clear-cut impres-
sion of her social subordination to men are forcefully stamped into a young girl's
consciousness" (1978, 165).
9. Sanderson (1981) presents a number of possible explanations for genital mutila-
tion, none of which is really provable but many of which are interesting, (i) Female
excision may have been a substitute for human sacrifice at one time. (2) Female excision
may be a kind of "tribal badge" when people were naked. It distinguished them ethni-
cally from other groups and demonstrates status. (3) A severe test of loyalty or bravery
for a neophyte. (4) To please men who dislike the sight and feel of unexcised female
genitalia, or increase the man's sexual pleasure because the vaginal aperture is minus-
cule. (5) An exaggerated form of cleanliness or purification. (6) To control female
sexuality.
10. The issue of the symbolic separation between women's sexuality and fertility in
many cultures warrants more attention. In Africa we see that sexuality and fertility are
ritually detached, and preference is given to fertility. I suggest that in the United States
this separation has taken the unusual form of preference for sexuality and a general
devaluation of fertility. The American penchant for both pornography and hysterecto-
mies is the mirror image of Sande clitoridectomy.
11. Brown also suggests that genital mutilation and other painful puberty rites for
boys and girls occur in societies with patrilocal residence and an exclusive mother-infant
sleeping arrangement. She interprets these rites as having to do with sex identity
conflict. I would tend to see both as having to do with breaking up the strong mother-
child bond in order to allow the patrilocal (patriarchal) culture to repeat itself in the next
generation (Brown acfally hints at this kind of interpretation).
12. Men can also of er kosa—for example, when a new office building is opened.
(Kendall 1985).
13. I refer the reader to Caroline Bynum's fascinating discussion of food rituals
among medieval Christian women (1987, esp. pp. 189-194).
 Rituals and Relationships 143

14. This is because sacrificial worship was standardized while the Semites were still
nomads who did not have agriculture and so did not have cereal as part of their
ordinary diet (Smith 1972 [1889]).
15. There is a problem in the ethnographic literature in that some scholars use the
word "sacrifice" for vegetable offerings followed by a communal meal (see the preced-
ing section on Thailand for an example of this sort of terminological muddle). I do not
consider vegetable offerings or communal meals to be sacrifice. I treat sacrifice, as did
Smith, more narrowly as rituals that focus on the ritual killing of an animal.
16. The problematic nature of kinship in patrilineal societies may also be solved by
religious injunctions concerning the control of women. By ensuring that girls remain
virgins until they marry, and that wives do not have contact with other men, husbands
can have some level of assurance that the children to whom their wives give birth are
really their own. Another solution to the "problem" of patrilineality is boys' initiation
rituals, which symbolize a return to the womb and rebirth as a member of the male
community. Please note that my analysis of patrilineality as inherently problematic is
almost exactly opposite to William Divale's (1984) argument that matrilocality is inher-
ently temporary and problematic because it does not allow fraternal interest groups to
function.
17. I am particularly impressed with David Hick's description of Tetum religion, a
religion almost totally dominated by women. Among the very few roles for men are a
sacrifice performed by a male rainmaker (1984, 85). Priestesses, as far as I can ascertain
from Hick's description, serve a great deal of cooked food, but they do not perform
ritual sacrifices. Hicks also describes the wedding ceremony in this patrilineal, patrilocal
society. As part of the ceremony the father of the bride makes reference to two kinds of
trees that "exude a reddish secretion villagers liken to blood" (1984, 58). This "blood" is
an allusion to blood covenants that unite individuals, clans, and princedoms in this
patrilineal society.
18. The association between men and blood rituals and women and food rituals turns
out to be far broader than I imagined when I first wrote this chapter. For example, in a
comparison of Turkic (including Kazakh, western Turkman, and Karakalpak) versus
Tajik (including Tajik, Uzbek, and Uabek) shamanism in Central Asia, Russian anthro-
pologist V. Basilov notes that among the former only men become shamans, while
among the latter almost all shamans are women. For our purposes, what is most
fascinating is that although the two types of shamans are similar in many ways, only the
former engage in bloody rituals such as jumping barefoot on sabers, hammering a
dagger into the flesh of the shaman, pricking the patient with a knife, and striking their
heads and chests. Bloody rituals are wholly absent from Basilov's portrayal of Tajik
(female) shamanism. In Tajik (female) shamans' rituals, flour and foods made of flour
are used as ritual snacks of which the spirits also partake (1976, 151-152). Skipping half
a continent, we find that Taiwanese male shamans, who in most other ways resemble
Korean and Japanese female shamans, engage in frenzied, bloody rituals such as raking
their foreheads and backs with long nails, and piercing their mouths with needles.
Significantly, their Korean and Japanese female counterparts do not perform similar acts
of bloodshed (Suzuki 1976). And, while in Mitsuo Suzuki's report of Taiwanese male
shamanism there is no mention of food rituals, food plays a crucial role in Korean
shamanistic rituals.
19. The lack of blood sacrifice among the Shakers really needs no explanation;
blood sacrifice was simply not part of nineteenth-century American religion. On the
other hand, the absence of blood sacrifice metaphors and symbolism does warrant
interpretation.
 7

The Sacred in the Profane

THIS-WORLDLY VERSUS
OTHER-WORLDLY RELIGIONS
Looking in from the outside, academicians try to understand the basic orienta-
tions of religious movements. One of the standard categories used by histori-
ans of religion to evaluate religions is "this-worldly" versus "other-worldly."
This-worldly religions emphasize life in the here and now, relationships be-
tween people, and the alleviation of suffering in this world during this lifetime.
Other-worldly religions focus attention on life after death, future redemption,
and mystical truth. Although no religion is totally this-worldly or totally
other-worldly, many scholars find that this terminology is useful in showing
us where particular religions invest the most energy, thought, and time.' The
orientation of a religion is most clearly reflected in its rituals. Other-worldly
religions may be less ritually rich, or may direct their rituals toward self-
perfection and future salvation. This-worldly religions concern themselves
with the elaboration of a rich ritual repertoire focused on concrete problems.
This does not mean, however, that this-worldly religions do not posit the
existence of another, supernatural plane.2 This-worldly religions either under-
stand the supernatural plane to be integrated with the natural world, or they
pay little attention to the too-distant supernatural plane.
Women's religions, attentive to alleviating suffering in the here and now, are
characterized by a this-worldly orientation. Indeed, in these religions the focus
is so much on the here and now that there is almost no elaboration of creation
stories or end-of-the-days scenarios. In some women's religions the this-
worldly orientation is explicit; in some it is disguised by other-worldly jargon.
In all the women's religions devotees learn to sacralize profane experience, to
enhance the quality of their current lives, to comprehend the supernatural
already present within the natural world, and to invite the divine into their
lives or even into their bodies.
146 Priestess, Mother, Sacred Sister
Buddhism versus Women's Religions
To begin, it is of interest to compare Buddhism to the women's religions that
are situated in Buddhist societies. As we saw in Chapters i and 5, Buddhism
makes a radical distinction between worldly (which is mere illusion) and other-
worldly (which is reality). Attachment to the world means being involved
with all sorts of impurities that prevent salvation. Salvation, in Buddhism, is
salvation from the world. Buddhism, in Melford Spiro's words, "is a religion,
par excellence, of other-worldly asceticism" (1971, 9). 3
Women's religions in Buddhist societies challenge the other-worldly inclina-
tion of Buddhism. The goal of the Burmese nat cult, for example, is to avoid
suffering while remaining in the world; one must accept the world and strive
for its rewards. Similarly, the Northern Thai matrilineal spirit cults are entirely
focused on this world and its problems and relationships. Ryukyuan religion is
also characterized by an emphasis on ritual rather than on metaphysical specula-
tion. Almost the only myths are etiological ones explaining the origin of
particular customs. "The range of speculative interest in nature and the super-
natural seems pragmatically limited to those areas directly related to man's
survival and well-being" (Lebra 1966, 42).
Korean shamanism does not describe heaven or hell—except for occasional
borrowing from Buddhist or Christian depictions. There is no good or bad
place to go after death, there is no reward or punishment for souls based on
what they did in this life, and there is no notion of rebirth of the soul. David
Suh argues that Korean shamanism is totally this-worldly in its affirmation of
the importance of the living and life in this world. Even though this world is
filled with pain and suffering, life is precious. "Religious salvation in the
mudang [shaman's] religion is not a release from the life of this world. The
mudang does not pray for a better life in another world, but for this life in this
world" (1989, 21).
The new Japanese religion Tensho-Kotai-Jingu-Kyu was founded by a
woman and attracts mostly women followers. In one of the first English-
language field studies of this religion, anthropologist L. Carlyle May wrote that
the founder "Mrs. Kitamura has not disclosed intricate doctrine concerning life
after death, heaven, and hell mainly because she is primarily interested in build-
ing God's kingdom on earth for the benefit of living people" (1954, 127).

The Individual and the Community: Afro-Brazilian Religions

and Feminist Spirituality
The this-worldly orientation of women's religions is also true outside Asia. In
her study of the Afro-Brazilian cult in Bahia, Ruth Landes found that people
believed that only life (and not death) is good, and that the best future for a soul
is to be reincarnated soon in a living baby. Notions of an afterlife are barely
developed. In general, the Afro-Brazilian religions are devoted to curing ill-
nesses and solving financial and social problems. Rituals have a pragmatic and
instrumental focus (1947, 217).
 The Sacred in the Profane 147
In Afro-Brazilian religions the this-worldly focus concerns the individual
and improving his or her immediate life circumstances. Feminist Spirituality,
on the other hand, is a this-worldly religion in which concerns are communal
and rituals are directed toward broad-based social change. In the writings of
the Feminist Spirituality Movement a recurring theme is the empowering of
women here and now. Movement writings make almost no mention of
heaven, hell, redemption, or the afterlife. The thrust of Feminist Spirituality is
that experiences of the sacred are available immediately, that the spiritual task
of human beings is to make this physical world a peaceful, harmonious, clean,
and safe place for all creatures. The Movement's involvement with ecological
issues is perhaps the clearest indication of the emphasis placed on this world as
the locus of human action. One Spiritual Feminist writer put it this way:
"Meditating individually and collectively, we build a vision that helps us create
the kind of world we want to live in. ... Simultaneously, we become the kind
of people we want to see living in our world" (Iglehart 1982, 303).

This- Worldly Religions in

Other-Worldly Disguise
Both Christian Science and Shakerism devalue life on this earth, yet both
believe that it should be as pleasant as possible. On the one hand, Shakers
withdrew from the world to their rural communities. On the other hand, for
the Shakers heaven was already in this world—in the peaceful, neat Shaker
villages. "The spiritual and the earthly were not two mutually exclusive
spheres, but were a unified whole, with the earthly life containing glimpses of
heaven even in its most mundane aspect [such as cleaning]" (Procter-Smith
1985, 201-202). For the Shakers the redemption had already occurred, here in
this world. It is of interest that although Shakerism demanded celibacy, asceti-
cism was not part of the Shaker platform: Believers were well fed and clothed,
and enjoyed the society of co-believers. According to Sally Kitch, "Shak-
ers . . . extolled the importance of physical and emotional comfort to a spiri-
tual quest" (1979, 80). In the first important Shaker written work, Testimonies
of the Life, Character, Revelations and Doctrines of Our Ever Blessed Mother Ann
Lee, and the Elders with Her (published 1816, quoted in Stein 1992, 27), the
editors explain that Ann Lee and the other Shaker founders had made it a point
to instruct their followers in temporal matters because "such affairs impinged
directly on spiritual development."
When I began reading about Christian Science I felt as if there were two
different religions going by the name of Christian Science. One religion was
paradigmatically other-worldly, teaching that matter is illusory. The other
religion was extraordinarily this-worldly, dedicated to alleviating personal suf-
fering and illness in this world through ritual healing. Christian Science has a
paradoxical relationship to this-worldly concerns. Since the world is unreal,
Christian Scientists do not need to separate themselves from the world, but
rather to learn to see it as an illusion. Physical matter is unreal; death, sin,
suffering and disease are unreal; only the Spirit exists and man is wholly Spirit.
148 Priestess, Mother, Sacred Sister
Ironically, however, Christian Science draws attention to the physical
through spiritual cures of bodily ailments. Thus Stephen Gottschalk (1973)
argues that Christian Science is essentially a pragmatic religion: Its theory is
related to practice, the truth of an idea is to be tested by the actual consequences
of believing in it. Christian Scientists are not ascetic; they believe that individu-
als can and should avoid physical suffering.
Christian Science has a vague and unemphasized eschatology; the spiritual
goal of Christian Science remains in this world—not to prepare people for
heaven. Although Christian Science interprets social problems in spiritual
terms, it does not teach that people must wait for an apocalypse before social
problems can be dealt with. The greatest this-worldly achievement of Chris-
tian Science is the Christian Science Monitor. The newspaper concerns itself
almost solely with such this-worldly matters as politics, war, and economics.
Christian Science believes that social problems, like health problems, are
caused by mental processes and can be healed through prayer. Unlike true
other-worldly religions, Christian Science is concerned with solving social and
health problems, and Christian Scientists give substantial funds for direct relief
in disaster situations (such as floods and fires).
Turning to a less-known example, in the 19205 a black woman named Leafy
Anderson founded the black Spiritual religion of New Orleans. Leafy Ander-
son's spirit guides, spirits who are still popular today, included Father John
(a healer), the Biblical Queen Esther, and Black Hawk (an eighteenth- to
nineteenth-century native American leader who in recent years has become
associated with Martin Luther King). According to Claude Jacobs, "Researchers
have characterized black churches such as ... [black Spiritual churches] ... as
'other worldly' in orientation and largely composed of women who have 'given
up hope that the socioeconomic problems characterizing their daily lives can be
solved by any effort on their part' (Thompson 1974:159). My data indicate that
some reassessment of this position is needed. In the New Orleans black Spiritual
churches, spirit guides such as Father John and especially Queen Esther and
Black Hawk are symbols of protest and empowerment for the largely female
membership who must confront problems that range from personal illness
to ... an oppressed groups's concern for righting 'a long history of wrongs'
(1989, 66). If I may take the liberty to interpret Claude Jacob's fascinating
findings, spirit possession in the black Spiritual churches of New Orleans is not
about escaping into the world of the spirits, but rather about inviting the spirits
into this world in order to empower church members.
Christian Science, Shakerism, and black Spiritual churches are examples of
this-worldly religions disguised by other-worldly costumes or rhetoric. Spiritu-
alism is another religion that on the surface seems obsessed with the next
world. Yet, according to Laurence Moore, "Spiritualism, despite the interest it
generated in a life after death, was not necessarily otherworldly in its outlook.
It too closely defined the spheres of the afterlife in terms of what went on here.
The people who contacted the spirits of the dead were in no particular hurry to
join them" (1977, 86). Spiritualism taught that one should enjoy this life while
in it, but not to fear death. Death is just a state through which one must pass to
 The Sacred in the Profane 149
reach the next existence. In Spiritualism the "other-world" is an extension of
"this-world." Even messages from the spirits tend to deal with this-worldly
and personal rather than theological or philosophical matters. Spiritualists can
be said to have rejected any distinction between natural and supernatural; spirit
is a tangible presence in the everyday world. It is interesting that critics of
Spiritualism (of which there were many, see Chapter 12) censured Spiritualism
for camouflaging its this-worldly focus inside other-worldly trappings. These
critics, according to Moore, "failed to see anything spiritual in the average
seance, and this was a complaint raised within the movement as well. Too
often sitters ordered the heavens down to earth, never encouraging their own
souls to soar any higher than the furniture flying around the room" (1977, 17).

A Question of Gender?
Given that female-dominated religions address women primarily as mothers, it
makes sense to begin interpreting these examples by situating the link between
women and this-worldly religion in the mother-child relationship and "mater-
nal thinking" (see Chapter 3). Grounded in particular relationships with particu-
lar children, mothers are unlikely to find a wholly other-worldly religion
appealing. To borrow Carol Gilligan's language (see below, page 155), people
whose "moral voice" speaks of care and responsibility would necessarily seek
to bring the spiritual down to earth, rather than seek to escape earthly interper-
sonal obligations and gratifications.
While I disagree with most of Erik Erikson's ideas about gender, certain of
his thoughts regarding women and religion provide food for thought. Specifi-
cally, he compares religious rituals that seek to form links with the Ultimate
with the rituals of interaction that all mothers exercise in relation to newborn
children. According to Erikson, rituals make the goodness and presence of the
infinite within the finite relations and facts of the everyday world, actual and
immediate to both child and religious follower (1968; also see Doyle 1974). To
phrase his ideas somewhat differently, the ritual task of both mother and of
women's religions is to insert the sacred into the profane and draw attention to
the unity of the two. In both cases, the ritual process itself is centered on
interpersonal interaction (see Chapter 6).

SACRED AND PROFANE

One of the most fascinating features of the Northern Thai matrilineal spirit
cults is their embeddedness in everyday life. One ethnographer who spent two
years in a Northern Thai village remarked that the cults were so inconspicuous
that "I might have spent my entire period in the village without being aware of
their existence" (Mougne 1984, 300). As Gehan Wijeyewardene has pointed
out, in normal circumstances the spirit cult tends to be so invisible that a
number of well-respected anthropologists have not even mentioned them, or
simply noted them briefly, commenting that they suspected people were trying
to keep them away from these rituals (personal communication 1992).* The
150 Priestess, Mother, Sacred Sister
rituals have to do with warding-off misfortune and illness, with practical
matters like ploughing and intragroup relationships, and with food and food
preparation. Similarly, Wijeyewardene found that the shrines of urban Thai
mediums are typically the front rooms of their houses and the mediums tend to
be consulted regarding everyday matters. "[T]he ritual activity of the me-
dium is concerned with this world" (1986, 163).>
In this-worldly religions the supernatural is immanent (see Chapter 8);
devotees do not need to leave the profane world in order to experience the
supernatural.
The clearest implication of the this-worldly orientation of women's religions
is the use of commonplace—"profane"—locations, individuals, concerns, and
instruments in "sacred" rituals. As I demonstrate in the next few examples, the
physical sites, implements, and goals of women's religions tend to be indistin-
guishable from daily life.

Sacred Geography
According to William Lebra, in Okinawan religion "all things, animate and in-
animate, are conceived as possessed of indwelling spirits. Consequently, a rigid
dichotomy of sacred and profane does not exist, such distinction being merely
one of degree" (1966, 45). Each village has a sacred grove that males are not
allowed to enter.6 The hearth of each house is also believed to be sacred. Various
trees, beaches, caves, and piles of rocks are ritual sites. Although sacred geogra-
phy is a fundamental element of Ryukyuan religion and there are myriad holy
places on the Ryukyu Islands, ritual sites do not look very different from ordi-
nary structures or natural formations. The sacred grove is a common-looking
clump of trees with a censer. The traditional style for constructing shrines was
no different from other buildings, and shrines contain no special decoration. The
community hearth and the household hearths are made up of three stones and a
simple shelf for offerings. There are no religious carvings or statuary. House-
hold rites use no special ritual paraphernalia and household rituals are carried out
by the senior female member rather than by a trained priestess. In the yard out-
side the house are other lesser ritual sites where prayers and incense are offered.
In short, the domestic quarters are both profane and sacred space.7
Similarly, the ritual objects and symbols of Korean women's religion are so
fully embedded within the home as to be almost invisible to the eye of the
observer. Jars of grain in particular places, for example, are offerings to gods
and ancestors. Household gods live inside the very structure of the house (the
roof beam, etc.). Every house, rich or poor, has its own gods. The Korean
shaman's inner room in which she performs divination rituals looks like the
main room of any prosperous village home. In the room are clothes, dishes,
cosmetics, toys, a television, and so on. The domestic appearance of the sha-
man's ritual space makes sense in light of Kendall's observation found that, "A
woman's participation in shaman ceremonies is an extension of her ritual re-
sponsibilities in the home" (1983, 99).
 vThe Sacred in the Profane 151

In England, "Leading spiritualists had always emphasized the importance of

the domestic base as the backbone of the movement, and the home circle was
revered as the focus for spiritual values and family harmony. The family repre-
sented the perfect milieu for uplifting manifestations" (Owen 1981, 75). In
these domestic seances, home furnishings played a central role. Across Amer-
ica people gathered around their dining room tables and tried to communicate
with spirits. The home, as opposed to the church, was seen by Spiritualists as
the true locus of religion. This ideology put religion squarely in women's
sphere.
The cult house of the Afro-Brazilian religions also has no special architec-
tural features, and most groups simply use space in the home of the cult leader.
A marked feature of these cults is the lack of absolute distinction between
sacred and profane. Consultations take place in the kitchen or living room;
food offerings for the gods are prepared in the same pots in which the cult
leader cooks for her family; a television table is converted into an altar at the
beginning of the ritual. Ruth Landes (1947) has described a cult house in Bahia,
Brazil. The chief priestess (also called "Mother") and lesser priestesses live
there, so it is in fact a sort of neighborhood center, where people make both
social and religious calls. The mother and priestesses live in the house
(candomble) in order to be in the company of the gods they tend and to serve
clients who need their intercession with the gods. Unlike Catholic priests who
embody the other-worldly, these priestesses only rarely withdraw from com-
mon life.

Sacred Work
Within religions that are dominated by men, women tend to be ritually active
in the domestic domain: educating young children in religious values, prepar-
ing holiday meals, laundering surplices. This is consistent with Michelle
Rosaldo's (1974) notion that gender inequality grows out of and reflects the
association of women with the domestic sphere and men with the public. In
female-dominated religions—located in societies that are not egalitarian—
women are also associated with domestic work. The difference is that domes-
tic work is interpreted as sacred and powerful.
Shaker communities, for instance, sacralized all of life, and everyday work
became metaphors for sacred work. In particular, the rituals and songs of the
period of Mother Ann's Work raised the everyday activities of women to the
level of spiritual activity. In purification rituals Shakers spent the night scrub-
bing and cleaning the village. Marjorie Procter-Smith cites one of the songs
sung during these rituals (1985, 188):

Bow down low, bow down low,

Wash, wash, clean, clean, clean, clean.
Scour and scrub, scour and scrub
From this floor the stains of sin.
152 Priestess, Mother, Sacred Sister

During and after the period of Mother's Work, "[O]ne implication of the
luxuriant growth of ritual mysticism was the universal sacramentality of the
world in which anything and everything was potentially capable of manifest-
ing Divine Reality . . . the possibility of consecrating the world, transforming
it to share the New Creation" (Whitson 1983, 20).

Sacred People

In daily life the priestesses of Hateruma [Ryukyu Islands] are hardworking peas-
ants who as sisters, mothers, and grandmothers fulfil an important social role in
which they do not differ from other women. The difference is only observable
when they perform their religious functions. In performing the rituals—the invo-
cations of the uyan (gods), the wishes to the uyan, the soft rubbing of the hands,
the offering procedures, the incantations—these women who are so normal in
profane life undergo an extraordinary metamorphosis. The natural authority and
the genuine seriousness and certainty with which the ritual acts are carried through
never fail to make a deep impression on the outsider and even on those who have
long been familiar with them (Ouwehand 1985, 127).

The overlap of the sacred and profane, and of the public and domestic, is a
recurring theme in depictions of leaders of women's religions. Sande leaders,
for example, teach domestic skills in a setting that is most explicitly sacred and
non-domestic—the bush. Afro-Brazilian religious leaders begin as women suf-
fering from personal problems, and as mediums they are involved with indi-
viduals coming to the temple seeking assistance with domestic worries (bad
husbands, sick children, etc.), but as priestesses who head Centers they carry
out a sacred function in the public domain. Most villages in Burma have at
least one shaman who is responsible for the village nat, participates in regional
festivals, and is hired to perform private rites (often to divine why someone is
ill). Village shaman, however, is a very part-time role—shamans are also house-
wives and farmers, indistinguishable from other village women. 8 Similarly,
Margery Fox (1989) found that there is usually a considerable overlap of friend-
ship ties between Christian Science patients and practitioners, and counselling
is apt to include rather pragmatic advice. On the other hand, in order to be able
to help the patient deny reality—and that is the essence of Christian Science
healing—the practitioner must keep well "prayed up;" that is, she must spend
a great deal of time alone reading and meditating. In short, the practitioner is a
combination physician, holy person, and friend, a role description typical of
most the women's religions.

CONTINUITY, NOT TRANSITION

A pattern has emerged in this chapter: Women's religions are embedded in
everyday life, sacred and profane are intertwined. The focus of women's reli-
gions is this-worldly rather than other-worldly; religion straddles the public
and domestic domains. The most obvious explanation for this pattern is the
 The Sacred in the Profane 153
social reality that constrains women more than men, and prevents them from
engaging in the violent ruptures and conversions that men describe and experi-
ence. Women—and especially mothers—cannot get up and walk away from
regular life in the way that men can.
It seems to me that additional factors come into play here, and in order to
expand our discussion I will return to the topic of initiation (see Chapter 6).
According to Mircea Eliade, girls' initiation rituals have to do with the myster-
ies that are natural to women: menstruation, fecundity, fertility. Boys' initia-
tion rituals, in contrast, are concerned with the revelation of a Divine Being,
sacred object, origin myth, or an event that took place at the beginning of the
tribe's sacred history. "For boys initiation represents an introduction to a world
that is not immediate—the world of spirit and culture. For girls, on the con-
trary, initiation involves a series of revelations concerning the secret meaning
of a phenomenon that is apparently natural—the visible sign of their sexual
maturity" (1958, 47). Put differently, girls' initiation has to do with the gradual
revelation of sacrality within the natural world; boys' initiation involves revela-
tion of sacrality outside of the natural world.
With some modification, Eliade's insight into the difference between boys'
and girls' initiation ceremonies is of the utmost relevance to the thesis devel-
oped in this book. Where I disagree with Eliade is in treating the foci of boys'
and girls' rituals (transcendent and immanent) as "polar opposites" [his term].
The entire point of Sande initiation, for example, is to show girls the transcen-
dent that is immediately accessible within the immanent, not to show them
that the transcendent and the immanent are two absolutely different experien-
tial modes.
Bruce Lincoln has reached similar conclusions in his recent work entitled
Emerging from the Chrysalis: Studies in Rituals of Women's Initiation. Lincoln
argues that women's initiation bestows cosmic significance on being a woman.
"In this way, ritual makes it possible for people to derive profound emotional
and intellectual satisfaction from otherwise pedestrian affairs, because it points
to something cosmic, transcendent, or sacred concealed within the tedium of
mundane existence." (1981, 107, my emphasis)
Lincoln asks whether Arnold van Gennep's well-known morphology of
rites of passage (separation from the former status, liminality, reincorporation
into the group as a person with a new status) accurately describes what
happens at female initiations. This question is of interest to us because van
Gennep's model implies that the actual change or "passage" occurs during the
period of liminality—when the initiate is outside of his or her "real world."
Significantly, Lincoln concludes that this model does not work for girls'
initiations. There is no separation (there may be isolation in her own house,
but not separation) so there is no return. The passage does not occur during
an extraordinary liminal stage, but while the initiand is embedded in her
everyday surroundings.
In critique of anthropologist Victor Turner's (and van Gennep's) model of
religious experience as liminal (i.e., outside of the ordinary, "betwixt and
between"), historian of religion Caroline Bynum writes that "the themes [in
154 Priestess, Mother, Sacred Sister

the life stories of medieval Christian women] are less climax, conversion,
reintegration and triumph, the liminality of reversal or elevation, than continu-
ity. Moreover, women's images and symbols—which, according to [Victor]
Turner's model, should reflect either inversion ... or elevation ... do not
quite do either. They rather continue or enhance in image (e.g., bride, sick
person) what the woman's ordinary experience is, so that one either has to see
the woman's religious stance as permanently liminal or as never quite becom-
ing so" (1984, 108). Whereas medieval men describe themselves as women
(role reversal) to assert humility or spiritual prowess, women do not describe
themselves as men but as either androgynous or as female. Margery Kempe (an
English fifteenth-century saint) saw herself as mother to the baby Jesus and
bride to the human Christ "carrying such images to heights of literalism by
actually feeling Jesus's toes in her hands. ... In her own eyes, Margery
achieves spiritual growth not by reversing what she is but by being more fully
herself with Christ" (Bynum 1984, 113). Drawing on Nancy Chodorow's
ideas (see Chapter 3), Bynum suggests that "Girls' more continuous self-
development, involving no fundamental need to develop a concept of'other,'
may help explain women's avoidance of dichotomous imagery and their ten-
dency to elaborate as symbols aspects of life closer to ordinary experience
(eating, suffering, lactating)" (1987, 293). 9
Other scholars have reached parallel conclusions regarding the internal conti-
nuity of women's religious lives. Janice Boddy shows that Sudanese men
afflicted by zdr spirits tend to follow a pattern of a single spirit entering their
lives at a crucial point, helping with a specific problem, and conferring a
lifelong benefit; whereas for women zdr possession is much more of an ongo-
ing dynamic relationship (1989, 261). Laurence Moore, writing about Spiritual-
ist mediums, notes that "almost all who began their careers as adults had felt
that the first signs appeared when they were younger."
Bynum maintains that in comparison to Christian holy men's lives, the
pattern of Christian holy women's lives shows few ruptures; "[Ijnstead,
there is gradually dawning vocation, voiced earlier and consolidated far
more slowly." Moreover, women are less likely to use images of gender-
reversal or to experience life-decisions as sharp ruptures because women are
raised by women and develop a continuous self whereas boys must undergo
the basic and intense reversal from wanting to be their mothers to accepting
being fathers. If Bynum (and Nancy Chodorow) are correct in their argu-
ment (and I believe that in many cultural contexts they are), the profound
sense of continuity in women's experiences would logically be deepened in
matrifocal and matrilocal societies, where young women do not even un-
dergo the rupture of leaving their natal homes to live with their husbands.
Ironically, this is where the parallels between Bynum's study of women in
male-dominated religions and my study of female-dominated religions ends.
The medieval women mystics described by Bynum belong to a male-
dominated religion and a patrifocal culture; they at least partially reject
the female role (through adopting celibacy), and they ultimately affirm a
transcendent religion. In contrast, women in most female-dominated reli-
 The Sacred in the Profane 155
gions embrace the female role, and preach doctrines and perform rituals
of immanence.10

MORALITY, PARTICULARISM,
AND SITUATIONAL ETHICS
According to Carol Gilligan, as a result of men's and women's different life
experiences, men and women often speak in different "moral voices." Whereas
men tend to stress universal rights and justice in making moral decisions,
women tend to see morality as embedded in webs of specific interpersonal
relationships. In my previous work among elderly Kurdish Jews in Israel, I
indeed discovered that "good" religious men were expected to pray, study, and
properly perform appropriate rituals, whereas "good" religious women were
expected to feed the poor, care for the sick and elderly, and help their neigh-
bors. Based on Gilligan's work and my own field experience, I expected to find
that female-dominated religions emphasize correct moral behavior, and that
they define moral behavior in terms of being kind and helpful.
I was startled when I discovered that my expectations were totally un-
founded! Over and over I read descriptions of women's religions in which the
ethnographer or historian wrote:

The actions that the [Afro-Brazilian] encantados [spirits] are thought to punish are
primarily connected with obligations or the failure to follow the law of the cult.
Except in unusual circumstances, the Batuque spirits are expected to be indifferent
to infractions of the moral code of the larger society. Robbery, theft, murder,
assault, and incest are regarded as major crimes that merit imprisonment or other
retributions by civil authorities, and that will also be punished by the Christian
supernaturals. Malicious falsehood, dishonesty, quarrelsomeness, disrespect to par-
ents, neglect of children, friendship with bad companions, excessive consumption
of alcohol—all are regarded as serious personal shortcomings. . . . However,
none of these activities is thought to bring punishment by the encantados (Leacock
and Leacock 1972, 74).
Supernatural rewards [on Okinawa] . . . are contingent, first and foremost, upon
proper ritual action and, to a lesser degree, upon proper conduct in specific social
roles and situations (Lebra 1966, 32).
This [Korean women's] religion teaches people that they are not responsible for
their suffering, and so they do not need to do anything to eradicate suffering. It
does not teach a sense of moral responsibility (Suh 1989, 19).

At first glance it indeed seems that women's religions do not emphasize

what we call moral behavior, and especially not that brand of moral behavior
defined in terms of helping other people. A closer look, however, suggests that
the moral frameworks of women's religions so differ from those of the male-
dominated world religions most of us are familiar with that we are deceived
into misrecognition. The dissimilarities in the moral frameworks are not only
on the level of content (even if in our own culture behaving rudely to mentally
156 Priestess, Mother, Sacred Sister

confused elderly people is not considered immoral, we can easily recognize a

rule of "be polite to all elderly people, even senile ones" as a "moral" rule when
we come across it in another culture). The dissimilarities are also on the level of
structure. I would argue that the "looser" structure of moral expectations in
women's religions seduces Western researchers into describing many of these
religions as focused on ritual (which appear to them to be clearly spelled out)
rather than moral behavior (which appear to them not clearly spelled out),
even when in actual fact proper moral behavior is treated with reverence.11
Karen Brown illustrates this point in regard to Haitian Vodou. "Vodou
morality is not a morality of rule or law but a contextual one. It is tailored not
only to the situation but also to the specific person or group involved. A moral
person, in Vodou, is one who lives in tune with his or her character, a character
defined by the spirits said to love that person" (1991, 240). To an observer
raised in the "Thou Shalts" and "Thou Shalt Nots" of the Ten Command-
ments, this sort of morality may indeed be unrecognizable.
Morality in women's religions tends to be a matter of appropriate behavior
in very specific contexts, rather than a matter of official or universal laws and
commandments. While many male-dominated religions codify or even reify
notions of how people should treat each other (Do not kill, Do not steal),
women's religions tend to preach rather vague ideologies of avoiding strife or
getting along with one's kinfolk. As we saw in Chapter 6, most of the
women's religions accentuate messages of mutual responsibility and peaceful
relations; however, the content of mutual responsibility and peaceful relations
is assumed to be situational rather than universal. Thus North American Spiri-
tualists accepted the Golden Rule as the highest statement of ethical principles,
yet rejected group-imposed standards that would limit individual freedom and
responsibility (Jacobs and Kaslow 1991, 78).
This is indeed consistent with Carol Gilligan's observation that in official
tests of moral development, tests that reflect male moral modes, women's
answers are often characterized by a lack of decisiveness or clarity. Women are
often more likely to think in terms of process, interdependence, and conflict
resolution, rather than justice or law.

Feminist Spirituality: A Case Study

Spiritual Feminist Starhawk has written an eloquent and thoughtful response
to those who criticize Goddess religions (Feminist Spirituality) for lacking a
conception of justice or system of ethics. Since most of her ideas apply to other
women's religions as well, I will quote her ideas at some length.
Starhawk argues that the "conceptions of justice in the western, patriarchal
religions are based on a worldview which locates deity outside the world. Of
course, within each tradition there are exceptions, but in the broad view of
Christianity, Judaism, and Islam, God is transcendent, and his laws are abso-
lutes, which can be considered in a context removed both from the reality of
human needs and desires and the reality of their actual effects" (1985, 194).
 The Sacred in the Profane 157

Starhawk illustrates this through the example of the attitude of the Catholic
Church to artificial birth control. Although Catholics do not believe that God
wants people to be poor and hungry, the Church is opposed to birth control
even in instances in which the result of unchecked procreation is poverty and
starvation. According to Starhawk, "such is the effect of this absolute morality.
Because when we believe that what is sacred—and, therefore, most highly
valued—is not what we see and sense and experience, we maintain an inherent
split in consciousness that allows us to quite comfortably cause pain and suffer-
ing in pursuit of an unmanifest good" (1985, 194).I2
In Goddess religions, on the other hand, the divine is regarded as immanent,
as manifest in nature and in the needs and desires of human beings. Thus,
"justice is not based on an external Absolute who imposes a set of laws upon
chaotic nature, but on recognition of the ordering principles inherent in na-
ture" (1985, 194). What discourages us from breaking "natural law" is not
hellfire or Judgment at the end of days, but the fact that the consequences are
also inherent in the structure of this world. "The Goddess is manifest not just
in human life, but in the interwoven chain of relationships that link all forms of
life" (1985, 195). Therefore, according to Starhawk, "Diversity is highly val-
ued [in goddess religions]—as it is in nature—in a polytheistic worldview
which allows for many powers, many images of divinity. In ecological sys-
tems, the greater the diversity of a community, the greater is its power of
resilience, of adaptation in the face of change. . . . Diversity is also valued in
human endeavors and creations" (1985, 195). As a result, in Spiritual Femi-
nism, "Ethics are concerned with fostering diversity rather than sameness, and
they are not concerned with enforcing a dogma or party line. Individual
conscience—itself a manifestation of the Goddess—is the final court of ap-
peals, above codified laws or hierarchical proclamations" (1985, 195).' 3 Isn't
there danger of anarchy in this sort of idiosyncratic morality? No, answers
Starhawk, because in Goddess religion the individual self is never seen as
separate from the entire human and biological community. "The Goddess is
manifest in the self—but also in every other human self, and the biological
world" (1985, 196).
As we saw in Chapter 6, many of the women's religions teach that the one
who suffers is not necessarily the one who committed an immoral or improper
act. Similarly, Spiritual Feminism proclaims that "Consequences [of immoral
behavior] may not be distributed fairly on an individual level; it is not the
owner of the chemical company who will give birth to a defective child, nor
are we comforted by the belief that he will burn eternally in an afterlife.
Consequences are suffered collectively, because it is our collective responsibil-
ity as a society to change those practices that destroy the lives of individu-
als. . . . No external God, Goddess, angel, or convoy of visitors from another
planet will do this for us; we must create justice and ecological and social
balance; this is the prime concern, the bottom line, the nitty gritty of ethics in a
worldview that sees deity as immanent in human life and the world we live in"
(1985, 199).
158 Priestess, Mother, Sacred Sister
Women and Morality
The absence of dogmatic and firm moral doctrines in many of the women's
religions is coupled with a nontranscendent view of divinity and a primary
concern with people rather than rules. While there is no necessary link among
these characteristics, there certainly is an internal logic. Concern with the here
and now not only has a metaphysical dimension (immanence) but also a moral
dimension (specific people rather than people in the abstract).
What I would like to ask here is whether this constellation of concerns is
more associated with women's experiences than men's. Philosopher Nel Nod-
dings argues that women's affinity to an ethic of caring is rooted in the experi-
ence of motherhood. Mothers quickly learn that absolutes are useless in dealing
with the vagaries of intimate human relationships (1984).
Historian of religion Sally Purvis (1991) has described one of the attributes
of mother-love that sets it apart from other types of relationships: It is indepen-
dent of the specific characteristics of the particular child—the mother loves that
child without knowing what kind of person that child will turn out to be, and
her love is neither dependent on nor canceled by the behavior of the child. To
put it differently, mother-love is a matter of relationship: I love this child
because I am tied to this child in incalculable bonds of nurture and responsibil-
ity, not because he or she followed the rules.
While I basically agree with both Nodding's and Purvis's accounts, I find
them a bit romanticized. In most cultures women do care deeply for their
children, yet the content of that caring varies, and in all cultures some mothers
have a difficult time loving all or certain of their children. What I would stress
in the maternal relationship is particularism. By virtue of becoming a mother
most women do not suddenly love all children, or love children as an abstract
category. Rather, they love particular children—their own children. Mother-
hood may well make one more sensitive to the sufferings of other children, but
it may equally well make one determined to protect one's own children even at
the expense of other children.I4
Particularism is where I see the notions of morality that characterize female-
dominated religions as fitting in with women's life experiences as mothers.
Although the matter is far from black and white, I do believe that it is accurate
to say that women's religions tend to shy away from general moral codes;
rather, they leave it up to the individual to behave properly in specific situa-
tions. In the next chapter we will see that most female-dominated religions also
preach particularism in the supernatural sphere.

Notes
i. Ever since Max Weber's Sociology of Religion, this terminology has been used by
students of religion. It has recently fallen into disfavor among some historians of religion
and has been refined by others (in particular, to differentiate between temporal and spacial
axes). But I do believe that it continues to have use in directing attention to the major
orientation of particular religions. It is, however, crucial to clarify that classical scholarly
assessments of religions as other-worldly versus this-worldly generally have looked at
 The Sacred in the Profane 159

religion from the point of view of the literate elite. It is far from clear that distinctions
between this and other worlds ever have much meaning for laity—male or female.
2. "Supernatural" is another concept that some contemporary theologians and histo-
rians of religion find problematic. For the limited purposes of the present chapter,
however, I do find it to be a helpful term.
3. Both Spiro (1971) and Wijeyewardene (personal communication 1992) draw
attention to the fact that Buddhism incorporates certain this-worldly elements. How-
ever, I do believe it is accurate to describe the thrust of Buddhism as other-worldly,
versus the thrust of the various women's religions that are this-worldly.
4. This episode in ethnographic ignorance naturally leads one to wonder whether
other women's religions have also gone unnoticed by anthropologists. The well-
documented tendency to "mute" women in ethnographic study, combined with the
inclination to assume that the domestic sphere is cross-culturally uniform and thus
uninteresting, may well have resulted in the invisibility of other women's traditions.
5. On the other hand, Wijeywardene cautions against setting up a strict polarity of
spirit cults as this-worldly and Buddhism as other-worldly. In a fascinating analysis of
the rhetoric of spirit mediums and Buddhist monks, he found that the monks expressed
concern with the social and political stability of this world, while the spirit mediums
were concerned with the fate of spirits and deities in the other world (1984b,331).
6. Teigo Yoshida has kindly shared with me the following anecdote: "When I
visited Kudaka Island several years ago, the village headman of Kudaka guided me
around the island and the village. We came to the most sacred place covered by thick
forest, the headman said 'Don't go inside. Men are prohibited to enter.' When I was
anxious to see inside, there was a woman in our group. Pointing to her, the headman
said, 'You can go in' " (personal communication 1992).
7. The kaminchu (priestesses), on the other hand, wear special white robes, a ceremo-
nial headdress and necklace, and carry a special box containing the cups and incense
used in the offerings (see page 233).
8. In the cities there are some full-time shamans, who earn quite a bit of money
serving a wide clientele.
9. In line with Bynum's ideas, Hanks concludes that life-cycle ceremonies in North-
ern Thailand show that "Women achieved compassion and maturity through their
sexual role, while men achieved these qualitites by denying their masculinity [becoming
monks]" (1963, 81).
10. I am indebted to R. J. Z Werblowsky for this insight.
11. Please note that the Shakers and Christian Science are exceptions to almost all my
general observations regarding morality and women's religions. The Shakers in particu-
lar developed detailed codes of behavior. (See Chapter 12.)
12. Starhawk's depiction of absolute moral codes in male-dominated religions is
probably a bit exaggerated. Neither liberal Protestant denominations nor liberal Jewish
denominations are quite so extremely rule oriented.
13. Not all feminist ethicists would agree with Starhawk's privileging of the individ-
ual conscience.
14. In cultures in which the one-on-one type of mothering familiar to Westerners
does not exist—typically because the responsibility for childcare is split with other
women in the kin-group—particularism has somewhat different connotations. In cul-
tures in which a person's identity as an is individual is subsumed to his or her identity as
a member of a community or kin-group, particularism has more to do with care and
responsibility than with Western notions of the individual as an autonomous social unit
and as a unique personality.
 8

No Father in Heaven
ANDROGYNY AND POLYDEISM

GODS, SPIRITS, AND ANCESTORS

IN WOMEN'S RELIGIONS
Notions of the supernatural are what differentiate religious from secular belief
systems. Given the importance that religious devotees accord their deities, we
will look at the gods and spirits of each of the key examples of women's
religions, even those for which data are sparse. In the previous chapters I
argued that the rituals of women's religions are concerned with interpersonal
relationships and reflect an essentially this-worldly, immanent orientation.
Concern with relationships and this world also characterize the supernatural
beings who are worshipped and addressed in women's religions. With the
exception of that one statement, the variations among the supernaturals are
more noteworthy than their similarities.

Burmese Nats
There is no one simple definition of nats—the term includes a wide variety of
entities. The two main categories of nats are nature nats and the Thirty-Seven
Nats who were human beings who died violent deaths. 1 Only a small number
of the nats are known by an average villager. "Burmese conceptions of the nats
(like many of their other conceptions) are marked by inconsistency, contradic-
tion, and by what might be called cognitive looseness" (Spiro 1967, 41).
Melford Spiro found that a particular nat might be called by several names, and
that some people treat a name as specific to a certain nat while others treat it as a
generic name.
June Nash explains that "The landscape surrounding the village and culti-
vated fields is alive with nats. . . . Nat propitiation enters into every significant
phase of the villagers' life" (1966, 118). Nature nats have power over specific
sections of the potentially dangerous world—wild and cultivated—in which
humans live out their lives. The nature nats are often quite amorphous—not all
are named and different people recognize different ones. Individuals perform
162 Priestess, Mother, Sacred Sister

many minor rituals to avoid immediate and personal problems caused by nats.
Often, misfortune is caused by nature nats who are angered by humans trespass-
ing on their turf or by humans who neglected to perform offerings (typically
food offerings). The nature nats are both easily angered and potentially protec-
tive on condition of humans propitiating them. In general, Burmese tend to be
more concerned with the dangerous side of the nats than with the protective
side; most rituals are oriented toward placation.
The most important nats are the "Thirty-Seven Nats." Each of these nats has
a name and a myth. Official lists of the thirty-seven have varied throughout the
centuries, and even the number thirty-seven seems to be a classificatory cate-
gory rather than a precise counting.

The best-known of the Thirty-Seven Nats is Min Mahagiri who is believed to

guard the house. In most Burmese houses a coconut hung from the southeast pillar
represents Min Mahagiri. The coconut is replaced every few months, at which
times offerings of bananas, rice, tea, and other foods are made to Min Mahagiri.
Min Mahagiri's myth goes as follows: A long time ago, Mahagiri lived as a human
blacksmith, and he was the most powerful man in the kingdom. The king, afraid
the blacksmith would raise a rebellion, took Mahagiri's eldest sister as his queen.
One day the king asked his queen to summon her brother to the palace. When he
arrived, the king ordered him to be thrown into a fire burning under a jasmine
tree. The queen then threw herself on the flames with her brother, and they both
died. The brother and sister, having become nats living in thejasmine tree, took to
punishing anyone who came into the shade of their tree. After more adventures, a
different king honored the two nats and ordered everyone in Burma to hang a
coconut inside the house in their honor. Nowadays, Min Mahagiri is propitiated
when people use anything made of metal (because he had been a blacksmith).
Members of the household are perpetually aware of the proximity of the Nat, who
dislikes the presence of birth or death in the house. When a family member is
seriously ill, the coconut is removed in order not to offend Min Mahagiri. Min
Mahagiri also dislikes sexual intercourse and fire, and does not like offerings of
cooked food.

"Nats are petty and irascible tyrants, quick to take offense or to feel jealousy
if any member of the household or the other household nats receive special
attention. Nats require constant attention—some housewives even fan them
during the hot season as well as keep their water offering and flowers fresh" (J.
Nash 1966, 122). Spiro describes the relationship between people and nats as
similar to the one between lord and subjects—a despotic relationship. "Being
despotic they evoke the same sentiments evoked by government. These senti-
ments can be easily expressed. Since they cause trouble, avoid them; if they
cannot be avoided, placate them; if their assistance is desired, bribe (propitiate)
them" (1967, 138).
Village nats guard the village against strange nats and human invaders, just as
the house nat serves as a warning to anyone who wishes to harm members of
the household. Personalized nats seem to receive more attention than ones that
do not have names and legends. The villagers' relationship to the nats is paral-
 No Father in Heaven 163

lelled by their other social relations. Women have similar types of relationships
with husbands and with nats—both must be cared for and deferred to. Adults
treat children similarly to nats—both must be appeased and humored. People
treat each other similarly to the way in which they treat nats—avoiding conflict
and exhibiting consideration. While people may consider nats in general as
irascible and harmful, many individuals have a personally chosen nat whom
they feel to be a constant and helpful companion (Sarah Bekker, in a personal
communication quoted in Spiro 1967, 53).

Spiritualist Spirits
Spiritualism, like other religions, begins with the God idea. Spiritualism, how-
ever, does repudiate the belief in the idea of a personal God dwelling some place in
the heavens, looking down upon His children below, and waiting for them to end
their earthly existence in order that they may receive their rewards and punish-
ments. . . . God is not an overriding personality, but an indwelling animation of
all life, manifesting in illimitable ways, reached from within, rather than
brought to us, from external sources [my emphasis]" (National Spiritualist Associa-
tion Yearbook 1953, quoted in Jacobs and Kaslow 1991, 77).

Spiritualists have professed to a variety of theological beliefs. One of the

most widespread has been in the existence of an androgynous deity (see be-
low). In actual practice, however, Spiritualists were and are much more con-
cerned with the heavily populated spirit world than with the one God.
Communication with the spirits is frequent—through seances, possession,
and dreams. Spirits talk, play music, move furniture, give advice, and make
requests. In general, Spiritualist spirits are not regarded as awesome. Most
spirits are helpful, and most are friends or kin of members. According to Vieda
Skultans (1974, 33), the link between the spirit and physical worlds is believed
to be one of mutual interdependence, and help flows both ways.
American Indian spirits have been particularly admired by Spiritualists, who
see Indians as closer to natural forces and less spoiled by civilization. Two well-
known theories could help account for this admiration. First, I. M. Lewis
(1975) has submitted that there is a connection between women's marginal
status in sexist societies and women's interest in peripheral (including foreign
and nature) spirits. Spiritualist women may have been attracted to Indian
spirits because both they and the spirits were "peripheral" in white, male-
dominated society. Second, Sherry Ortner (1974) has argued that women uni-
versally are seen as closer to nature, while men are associated with culture (see
Chapter 10 for a critique of her theory). In line with Ortner's ideas, we would
expect to find that nature spirits are prevalent in a religion dominated by
women.
While both Lewis's and Ortner's points have merit, I would prefer to situate
the discussion of Indian spirits in the specific historical context of nineteenth-
century America. Although Spiritualist women were of European ancestry,
American Spiritualists lived on American soil. American Indian spirits, spirits
164 Priestess, Mother, Sacred Sister
of people who had lived in the American countryside and hunted and gathered
in the American forests, were perceived as eminently here, as fundamentally
immanent. By embracing American Indian spirits, nineteenth-century Spiritu-
alists were making a theological statement rejecting transcendence (even the
very concrete transcendence of "across the ocean") and were reinforcing their
conception of the spirit world as co-existent with the material world.
In addition, I would venture to say that an added allure of unspoiled Indian
spirits for nineteenth-century Spiritualist women lay in the (perhaps subcon-
scious) realization that white-male-dominated technology had already begun
to irrevocably ruin the planet and decimate entire populations. Spiritualism, as
an American movement, took off in the wake of the Civil War. It seems likely
that many Americans saw, for the first time, how machines could be used as
tools of destruction on a far greater scale than had been possible in a pre-
industrial age.2 As we will see in Chapter 10, this same idea has been central to
the political and spiritual consciousness of the twentieth century Feminist Spiri-
tuality Movement, which has been critical of modern technology.

Ryukyuan Kami
Among Spiritualists, spirits are highly individualized, bearing clear personali-
ties, names, and relationships to the people communicating with them in
ritual. The Burmese nats also have names and stories, even if people are a bit
vague regarding which nat is which. Unlike both the spirits of Spiritualism and
the nats of Burma, Ryukyuan kami (loosely translated as gods) generally do not
have myths associated with them, nor are they highly differentiated among
themselves.
The people of the Islands believe that the entire universe, both animate and
inanimate, is occupied by indwelling spirits or kami. Kami have been described
as a sort of impersonal, mystical power resident in gods, trees, people, nature,
animals, storms, swords, tools, and sacred places (Haring 1964). The kami are
associated with a wide variety of natural elements, places, occupations, ances-
tral spirits, and living human beings. On many of the islands, the hearth kami
serves as a link between the household or community and the higher kami.
The kami are superhuman, nonpersonalized beings who mete out rewards
and punishments. The kami are not omniscient or omnipotent, good or bad,
yet they do have the power to supervise and influence events of this world.
The function of ritual is to harness the power of the kami. As in Burma, lay
people are more interested in avoiding trouble with the kami than with actively
enlisting their support.

Afro-Brazilian Encantados
The various Afro-Brazilian religions have somewhat different theological
approaches and posit somewhat different pantheons. The similarities are, how-
ever, sufficient to warrant treating them as a group. Afro-Brazilian cult mem-
bers believe in God, Mary, and Jesus who live far away with the saints and
 No Father in Heaven 165
angels. Closer to humans, physically and spiritually, are many thousands of
other spirits. These spirits (encantados) can live on earth, and so contact with
them is unavoidable. The spirits have characters similar to people—both good
and bad traits.
Afro-Brazilian possession rituals teach that communication between humans
and gods is possible. This communication takes place through human medi-
ums who are in an ecstatic altered state of consciousness or trance. Most human
misfortune and suffering can be explained and cured by contact with the super-
natural world. Devotees believe that gods and spirits like and need to possess
humans, and must be given regular opportunities to do so. Thus, possession is
the supreme expression of worship.
The large pantheon of gods and spirits may include (depending on the
particular cult) African gods and goddesses, Catholic saints, Indian gods and
goddesses, dead slaves, dead Indians, and an assortment of demons. In
Umbanda, for example, there are four major types of possessing spirits: Indi-
ans (who are stern and aloof and give advice in matters requiring quick and
decisive action), Old Blacks (bent over from slavery, gentle and easy to ap-
proach, good at long and intricate personal problems, knowledgeable about
herbal remedies), children (who died young, and cause the medium to be
playful and innocent), and Exus (evil spirits that may make the medium engage
vin antisocial acts).3 Also, each Umbanda member has an orixa (African deity
who is syncretized to a Catholic saint) who protects him or her.
In certain cults each priestess is a votary of one god; in other cults mediums
become possessed by a number of gods. In those cults in which a medium has
a special spirit who consistently possesses her, this relationship is described in
familial terms: the spirit is the father or mother and the medium the son or
daughter, "and the spirit is expected to keep as close a watch over his devotee
as a good human parent does over his offspring, as long as the devotee keeps
his part of the contract and fulfills his ritual obligations" (Leacock and Lea-
cock 1972, 60). These obligations may include food and sex taboos, ceremo-
nial costumes, and tending a household shrine. The spirits are credited with
maintaining the economic well-being of the household in the face of frequent
financial crisis. Spirits can make other people behave in a way that will
benefit their devotees. The spirits are also credited with maintaining their
devotees and their devotees' families in "reasonably good health" (Leacock
and Leacock 1972, 65).

North African Zar

Zar spirits (zayran) differ from nats, kami, and Afro-Brazilian spirits in many
important ways. One similarity lies in their sheer numbers. The zar pantheon
varies from society to society, even from ritual group to ritual group. Each
zar (unlike kami but not unlike nats) has an individual identity. Zar are divided
according to basic human categories of age, sex, social class, education, reli-
gion, and ethnicity. In Ethiopia, for example, they are organized in a social
hierarchy similar to the Amharic social hierarchy; some are powerful lords
166 Priestess, Mother, Sacred Sister

and others are servants. Some zars are intellectual, and some are benevolent
healers. People tend to be possessed by zar spirits from the same social class
as themselves.
The goal of zdr rituals is to negotiate with a zar who is causing problems,
and to transform him into a protective zar. Zar spirits seek opportunities to
have fun in the human world. The zar spirits are capricious and selfish; they do
not bring illness and infertility in order to harm the woman, but to convince
her to supply their needs. Although zars can be troublesome, most zars are not
exorcised; rather, they are accommodated. In some situations the zdr spirits are
seen as evil. Since the typical zdr cult member had previously tried other
healing methods—all of which had failed—turning to the zdr ceremony can be
an acknowledgment that the evil entities have won.
Z,dr spirits are typically perceived as having originated outside the commu-
nity. Northern Sudanese, for example, recognize spirits of Muslim saints,
Turkish administrators, Europeans, Hindus, Chinese, Ethiopian kings, slaves,
sorcerers, prostitutes, Syrian gypsies, West Africans, nomadic Arabs, Western-
ers (including an archeologist), military officers and doctors of the Ottoman
empire, and Southern Sudanese—in short, everyone with whom villagers
have had contact during the past 150 years—everyone, that is, who is not one
of them!
Unlike Allah the zdr spirits are not eternal. They are born, age, die, and are
gendered; they marry and have children and have relatives and friends and
occupations and specific habitations. "Like humans, zayran are social beings"
(Boddy 1989, 274).

Northern Thai Phii Puu Njaa

In Northern Thai villages many kinds spirits populate the paths, gardens,
fields, and houses. One kind of spirit—matrilineal clan (or descent group)
spirits—are envisioned as the summation of all the spirits of houses in which
the eldest women are of the same clan. These spirits are, simultaneously,
independent beings who are related through female ties in the same way as
their votaries are. In addition, these spirits are, in a vague manner, the nonspe-
cific spirits of former male and female members of the matriline, and the spirits
who were worshipped by former members of the matriline.4 Typically, they
occupy a corner of the bedroom next to the eldest female's bed, where they are
continually present in their disembodied way. The spirits are not named (no
matriline records are kept) and are divisible, but the spirit of each clan is
distinct from the spirit of other clans. This entire belief system seems not to be
very elaborated.
"The spirits offer a conditional protection to the people under their care:
they protect if they are politely treated, receive offerings, and are formally
informed of every important family event or change of status of a family
member. . . . Like living relatives, they are offended if they discover that they
were the last to know anything of interest or importance" (Potter 1977, 116-
117). They then withdraw their protection and cause family members to fall
 No Father in Heaven 167

ill. The spirits are benevolent and protective, and serve as useful disciplinarians
who punish with sickness but not death. (However, some Northern Thai claim
that clan spirits can get other spirits to give out harsher punishments, Turton
1984, 280).
The matrilineal descent group spirits differ from other Thai spirits in two
ways that are significant for our inquiry. First, Thai spirits—except matriline
spirits—are generally seen as frightening and malevolent. In this regard the
matrilineal spirits are more similar to most of the other gods and spirits of
women's religions; the supernatural beings of women's religions are often
capricious or demanding of attention, but not evil or terrifying. And second,
while most of the spirits in the broad Thai pantheon are connected to land and
objects, the descent group spirits are not particularly concerned with property,
a fact Turton (1976, 246) interprets as reflecting the concern of the descent
group with people and not land. Again, this interpersonal focus is similar to
most of the other supernatural entities we have met in women's religions. One
way in which Thai matrilineal spirits differ from many of the other spirits
found in female-dominated religions concerns their ethnic identity. Spiritualist,
zar, and Afro-Brazilian spirits are often foreign (in Lewis's terms, peripheral).
The Thai spirits, on the other hand, are the sum of the matrilineal ancestors
(rural cults) or figures from Thai history (urban cults), in both cases, local.
The caw (spirits) of urban Thai spirit mediums are highly elaborated and
associated with complex and specific epic legends (Wijeyewardene 1986). It
strikes me as noteworthy that these spirits are inordinately concerned with
revealing their individual identities to people. Rituals typically focus on figur-
ing out which caw is responsible for the client's misfortune or the medium's
possession.

Korean Household Gods and Restless Ghosts

In Korean women's kut ritual, possessing ancestors are vocal, weepy, angry,
excited, and full of emotion. Women grieve for their dead ancestors, and also
bicker and negotiate with them. In contrast, in male Confucian rites the ances-
tors are silent and awesome, and the men are solemn.
Possessing gods, spirits, and ancestors who appear in kut have clear identities
and names. The gods of Korean shamans are not distant beings. Like petty
officials, gods are fallible, greedy, and open to bribes and flattery. Janelli and
Janelli (1982) describe how the gods (speaking through the possessed shaman)
act like corrupt and rapacious officials, trying to get better offerings, while the
villagers refuse to be intimidated and cajole, negotiate, and match wits with the
deities. Korean kut is one of the clearest examples of direct interaction between
deities and humans.
In the official Confucianist system only the ancestors of the husband's family
are worshipped. Women's rituals, on the other hand, relate to ancestors from
both the maternal and paternal lines. Laurel Kendall (1983) found that ghosts
and ancestors from the Korean woman's family are not felt to be any more
malevolent than ghosts and ancestors from the husband's family.
168 Priestess, Mother, Sacred Sister

During the kut ritual, ancestors who appear do not identify themselves, nor
do they appear in any particular order. The participants must figure out which
ancestor is speaking. Therefore it is usually close kin who appear, otherwise no
one would be able to recognize the identity of the particular ancestor. By
identifying ancestors and listening to their complaints, participants gain infor-
mation that helps them determine the cause of misfortune. But the ancestors
never give enough information to unequivocally fix the cause of misfortune.
"Even after a kut is finished, participants come away with different conclu-
sions" (Janelli andjanelli 1982, 153).
The dead need the same things as the living (money, food, sex). Equally
important, in Korean women's religious view, the dead need to be related to—
they continue to have psychological and emotional needs. At men's Confucian
ancestor rituals the dead, although benevolent, do not have strong personali-
ties. At women's rites the dead are not so benevolent, but they have stronger
personalities. From the perspective of women, ancestors "are still enmeshed in
their former social relations" (Janelli andjanelli 1982, 174). Because different
people have different social relationships with any particular person, they have
different images of that person as a supernatural being.

Black Carib Ancestors

Black Carib prayers are addressed to God, the Virgin Mary, and deceased
ancestors. Most of the ritual activity is directed toward ancestors. The song
sung in the crucial dance of propitiation at the dtigti ritual begins and ends with
the refrain "Oh my grandmother, we are placating you," whether the actual
ancestor is male or female (Kerns 1983, 163; Taylor 1951, 121).
Ancestors care a great deal about the interests of their earthly kin. When
displeased, they withdraw protection and cease serving as mediators between
their kin and the supreme authorities of the universe. Displeasure results when
the living neglect religious duties, ignore the moral code, or deviate from
traditional ways of life. A lower class of spirits, hiuruha, function as mediators
between diviners and ancestral spirits. Catholic saints are also worshipped,
although they are believed to be somewhat more remote from human affairs.
Black Caribs also recognize a large number of wicked or frightening super-
natural entities. The agaiuma appears as a crocodile, crab, or green-haired
woman. Thefaialanda is an eerie light seen by fishermen. The umeu are small
sea creatures who harm children. The ogoreu is a malignant supernatural being,
usually seen as a blue lizard, which attaches itself to women and is transmitted
in the female line. Many sorts of fantastic animals roam the outskirts of the
villages. Ruy Galvao de Andrade Coelho's (1955) description gives the impres-
sion that many but not all the evil supernatural entities are female.
Ideas concerning the human soul are also well elaborated by the Black
Caribs. The human soul is understood to consist of three parts: the anigi, which
is a kind of physical vital force located in the heart, the iuani whose seat is the
head, and the dfurugu—an astral body that reproduces the material shape of a
person but is composed of a substance similar to that of supernatural entities
 No Father in Heaven 169

(Coelho 1955, 137-138). This astral body is the intermediary between the
supernatural and everyday realms of reality.

Sande Ancestresses
Writing about Sande in a chapter on theology is a difficult task. Like Black
Carib religion, it is ritually oriented. Unlike Black Carib religion, it is a secret
religion and the uninitiated do not know what Sande women are taught during
their time in the bush. What can be ascertained from the ethnographic record is
that Sande teaches girls how to behave, and enforces that behavior through the
threat of natural and supernatural sanctions. The natural sanctions are carried
out by Sande leaders, and the supernatural sanctions are carried out by the
founding ancestresses of the Sande chapter, who are sometimes represented by
Sande women wearing masks. Each Sande chapter has a different founding
ancestress who oversees human behavior. 5

WOMEN AND POLYDEISM

As we have seen, most women's religions relate to many spirits, gods, or
ancestors. In general, these beings are not all good or all bad. There are no
great benevolent lords or thoroughly evil devils. Most are somewhat "fair to
middling" sorts of divinities—depending on how they are treated.
The weight of the evidence brought here indicates that women who belong
to female-dominated religions are attracted to religions that offer interaction
with multiple supernatural entities, even though many of these women live in
societies in which monotheistic beliefs are known. I have chosen to gloss as
"polydeism" a range of beliefs in more than one supernatural entity; polydeism
characterizes almost all women's religions. In some women's religions the
supernatural entities are close to what Westerners think of as gods, and scholars
have indeed translated the native term for the supernatural entity as "god." In
other women's religions spirits or ancestors direct the fortunes of human be-
ings, but, again, I stress that a great deal of the terminological distinction
reflects how Western scholars have elected to translate native terms. What
seems to me significant is that even in situations in which members of the
women's religion are also Catholics or Muslims (and so officially monotheis-
tic), in the framework of the female-dominated religion, rituals address many
supernatural beings. Similarly, ethnographers have been struck by the predomi-
nance of Jewish, Christian, and Muslim women at shrines of saints. Thus, we
have reason to suspect that women in male-dominated religions also tend to be
attracted to polydeism.
Ronald Stover and Christine Hope (1984) conducted a statistical analysis of
312 societies around the world to determine whether gender status affects reli-
gious beliefs or whether religious beliefs affect gender status. For the purposes of
their study they focused on one aspect of religious belief, monotheism—which
is almost always the belief in one male god. Significantly, they found a clear
relationship between monotheism and male dominance. 6
170 Priestess, Mother, Sacred Sister
In a very different sort of study, Judith Ochshorn has shown an interrelation-
ship between the rise of monotheism and gender inequality in ancient Near
Eastern religions. Comparing Sumer, Elam, and Athens, she discovered that
"where there was a polytheistic pantheon and cultic practices directed toward
its propitiation, the presence of both sexes in the priesthood, in cult, and in
popular religion was almost universal" (1981, 108; see also Lerner 1986).
And finally, writing about the Greco-Roman world, Ross Kraemer suggests
that "there is conceivably some relationship between monotheism and the
exclusion of women from Jewish and Christian priesthood, an exclusion that
carries over to monotheistic Islam as well. When divinity is perceived to be
one, and the gender of that divinity effectively presented as masculine in lan-
guage, imagery, and so forth, perhaps only the sex which shares that gender is
perceived as able to perform priestly functions. Conversely, among the Greeks
and Romans both the gods and their clergy came in two genders" (1992, 197).
In short, it seems that monotheism is not an attractive religious option for
some women. In making that claim, I am not saying that men and women are
born with inherently different theologies. Rather, I am suggesting that
women's life experiences may lead them to prefer more than one deity. (Given
the assortment of supernatural beings recognized in women's religions, I use
the word deity in a very loose sense.)
We could propose a variety of means of interpreting women's affinity for
polydeism. Freudian psychoanalysts could probably explain to us that the
resolution of little boys' Oedipal complex leads to the development of a super-
ego (i.e., monotheism) while girls' infantile experiences do not. Some psy-
chologists have asserted that women's cognitive styles are more diffuse than
men's, allowing us to postulate that women's theologies are also more diffuse
(i.e., polydeistic). Other psychologists would tell us that women as mothers
know that they are not all-powerful, so that even if as children they had
fantasized an all-powerful parent/deity, as adults they know that reality is far
more complex. Men, on the other hand, might still harbor a fantasy of
mother/god as omnipotent.
Sociologists and anthropologists argue that in most cultures women's place
in the social structure is in small, domestic units—of which there are many,
rather than in the overarching institutions that tie together the individual do-
mestic units. Men who run state level institutions could be presumed to imag-
ine single, all-powerful deities, while women in their domestic units would
tend to see many, smaller deities. Or, as David Suh claims regarding Korean
society, women and other helpless and oppressed people have no other place to
turn "except to the invisible spirits of their ancestors. After all, parents and
relatives are the only reliable support, dead or alive, in this world of political
suppression and economic exploitation" (1989, 17). In other words, the rich
and powerful (and by implication we may also say men) enjoy more "public"
support in the form of governments, armies, and courts, whereas women and
the poor have only familial support. Translated to the spiritual sphere, the rich
and men have transcendent "public" deities, while the poor and women have
personal spirits.
 No Father in Heaven 171

Historians might contend that monotheism is part of a political ideology

linked to the notion of divine kingship and leading to totalitarianism. Max
Weber connected political processes (the rise of a world empire in China, the
extension of the power of the Brahmin caste in India, and development of the
Persian and Roman empires) to the rise of universalism and monotheism (1966,
23). It is suggestive that none of the women's religions are linked to either
divine kingship or totalitarian social structures.'
One could show that monotheism is connected to a transcendent view of the
divine. As I have demonstrated in previous chapters, women have an affinity
for—or more precisely, an aptitude in—immanence, thus, female-dominated
religions do not need transcendent monotheism. One could elaborate upon this
idea by demonstrating that in the "great" monotheistic religions—Judaism,
Christianity and Islam—"the mother's domination over individual birth and
child care fades in significance with the great cosmic birth and transcendent
care that men in connection to the great male God are envisioned to give"
(Combs-Schilling 1989, 264). Or finally, one could join feminist philosopher
Nel Noddings who writes that "It seems to me quite natural that men, many of
whom are separated from the intimacy of caring, should create gods and seek
security and love in worship. But what ethical need have women [mothers] for
God? . . . What I mean to suggest is that women have no need of a conceptual-
ized God, one wrought in the image of man. All the love and goodness
commanded by such a God can be generated from the love and goodness found
in the warmest and best human relations [i.e., motherhood]" (1984, 97).
All these hypotheses are food for thought, and many are likely to be true. In
light of what we know about other aspects of female-dominated religions,
however, I am inclined to look for the connection between women and
polydeism in a somewhat different direction. In Chapter 5 I showed that
women's religions often carve out a niche of expertise in addressing the prob-
lem of suffering. I demonstrated that most of these religions offer highly
convincing explanations for why people suffer, and varied and eclectic solu-
tions to suffering. I also advanced suggestions as to why women in particular
are attracted to religions that elaborate this-worldly responses to suffering.
Women in sexist societies are indeed sick a great deal, and women as mothers
who love their children are anxious to ensure that their children survive. It is
highly significant that in many situations described in this book, women who
do have other religious options have turned to polydeistic and female-
dominated religions because of illness, misfortune, or death of their children. It
seems to me that monotheism may be too final, too absolute, to answer the
needs of religions whose raison d'etre is the alleviation of suffering in this
world. Monotheism leaves little room for negotiation, and it offers fewer
healing options than religions with myriad supernatural entities whose neglect
may mean illness and whose appeasement can mean recovery. Polydeism, in
contrast, offers endless possibilities for explaining, dealing with, and (hope-
fully) healing suffering and illness.
A final remark is in order here. So far we have avoided looking at three out
of the four North American women's religions. Two of these religions, Shak-
172 Priestess, Mother, Sacred Sister
ers and Christian Science, do not fit my portrayal of women's religions as
polydeistic. Not surprisingly, both religions are situated in a broader cultural
setting that considers monotheism to be more spiritually and intellectually
advanced than polytheism. What does characterize the Shakers and Christian
Science is the rejection of a purely masculine monotheism. We turn to the
gender of the deities in the next section.

IN SEARCH OF THE GODDESS

Stover and Hope's study suggests that it is not simply the belief in one god,
but the belief in one male god that women do not find compelling. The
second half of this chapter explores the gender of the deities worshipped in
female-dominated religions. In many male-dominated religions (including Ju-
daism, Christianity, and Islam) the deity is known by masculine names and
described as possessing masculine attributes (see Ruether 1974). Feminist schol-
ars have shown how male deities both reflect and strengthen patriarchy. Mary
Daly has explained, "The symbol of the Father God, spawned in the human
imagination and sustained as plausible by patriarchy, has in turn rendered
service to this type of society [patriarchy] by making its mechanisms for the
oppression of women appear right and fitting. If God in 'his' heaven is a
father ruling 'his' people, then it is in the 'nature' of things and according to
divine plan and the order of the universe that society be male-dominated"
(1973, 13)-
In some male-dominated religions goddesses are worshipped alongside gods.
Western scholarship, a product of religious traditions that lack goddesses, has
tended to look at goddess worship from a psychoanalytic perspective—
implying, perhaps, that goddesses are psychological anomalies in need of scien-
tific explanation (see, e.g., Fromm and Maccoby 1970). Freud, for example,
posited that devotion to goddesses reflects an infantile desire for reunification
with the all-powerful mother.8
Fortunately, recent scholarly studies of goddess worship have begun to
move beyond psychoanalytic approaches. Sifting carefully through ethno-
graphic and textual evidence, James Preston, in a wonderful book entitled
Mother Worship, has identified a number of common characteristics of god-
desses around the world. These include antiquity (fertility goddesses were the
first deities formally worshipped by human beings), a connection to mother-
hood, ambivalence (encompassing such opposites as love and anger or for-
giveness and vengeance), a protective function vis-a-vis their devotees, an
association with the resolution of human problems—especially illness, and
virginity or virgin motherhood. Since Preston's study looked at goddesses in
male-dominated religions, one of my goals in the present section is to verify
whether these attributes are also true of the supernatural entities of female-
dominated religions.
As I said earlier, feminist scholars have argued that male gods in patriar-
chal religions reflect and reinforce patriarchal social structures. On the whole
 No Father in Heaven 173

this seems to be true, yet there are occasional situations in which male gods
have been found to empower women, or goddesses to restrict women. "The
existence and power of a goddess ... is no indication or guarantee of a high
status for human women" (Frymer-Kensky 1992, 80). I will limit my com-
ments here to two brief examples. Mormonism is so patriarchal that it is
inconceivable for God to perform "female" tasks. Thus, Mormon theology
requires a Goddess—Mother in Heaven—to perform women's specific func-
tions (Heeren, Linsey, and Mason 1984). The Mormon Goddess clearly is
not an empowerer of women. In a very different situation, Julia Esquivel
makes an impassioned plea to poor Guatemalan women to encounter the
"God of the poor"—the God for whom all people are equal, and who
accepts "poor women, peasants, Indians, servants, and factory workers . . .
as [His] daughters" (26). Once women open their eyes to this God and to
their own human dignity, women will be empowered to struggle for social
justice. Esquivel's male deity, unlike the Mormon female one, does em-
power women. 9
Several sociological and psychological studies of American men and women
have called into doubt simplistic models of the relationship between gender
roles and gods and goddesses. Tamayo and Dugas (1977) found that among
both male and female college students, God (who is referred to by masculine
names: Lord, King) is described in terms more similar to those in which the
students describe mothers than those in which they describe fathers. Similarly,
Nelsen, Cheek, and Au found that American men and women choose the
word "father" over the word "mother" to describe God, but their description
of God is supportive (maternal) rather than punishing (paternal). We now turn
to several examples of how women in female-dominated religions portray the
gender of the deity.

The Goddess Within

For me the divine/Goddess/God/Earth/Life/It symbolizes the whole of which we
are a part. ... To poison rivers and seas and the ground on which we stand so that
we can have televisions and air-conditioning, to engage in wars of conquest in
order to exploit other people's labor and take the resources of their land, is to
forget that we are all connected to the web of life" (Christ 1989, 320-2).
"God is inside of you and inside of everybody else. You come into the world with
God. But only them that search for it inside find it. And sometimes it just manifest
itself even when you not looking, or don't know what you looking for" (Walker
1982, 177).

Despite convincing evidence that there is no necessary or simplistic relation-

ship between women's status and the gender of deities, I was astonished to find
that very few of the women's religions worship goddesses. The one exception
is Feminist Spirituality.
Feminist Spirituality draws on various ancient civilizations as sources for
174 Priestess, Mother, Sacred Sister
Goddess imagery and myths, yet "these [ancient] traditions are filtered through
modern women's experiences. . . . Ancient traditions are tapped selectively
and eclectically. . . . The Goddess symbol has emerged spontaneously in the
dreams, fantasies, and thoughts of many women around the country in the past
several years" (Christ 1979, 276).
"I found God in myself and loved her fiercely"10 is perhaps the clearest
declaration of the Goddess within the Feminist Spirituality Movement. The
Goddess, unlike the Gods of Judaism, Christianity and Islam, is not "out
there." Rather, goddess nature and human nature overlap, or according to
some theologians, are identical. Goddess is immanent in the extreme—she is
the spiritual expression of sisterhood. For some Spiritual Feminists, Goddess is
a metaphor for women's spiritual power. Others believe in a more traditional
female deity, albeit one whose names and traits are drawn from diverse reli-
gious traditions." For all Spiritual Feminists, a female Goddess consciously
enhances and reflects their own identities as women.
Carol Christ (1979, 2 7 7 f f . ) elaborates on the meanings of the Goddess: (I)
the acknowledgment of the legitimacy and beauty of female power as a benefi-
cent and independent power; (2) an affirmation of the female body and cycles;
(3) the personification or manifestation of energy that flows between living
beings and the natural world; (4) the embodiment of women's heritage, of
sisterhood. Naomi Goldenberg (1979), in addition, stresses the use of the
goddess concept to give women positive self-images in all stages of life: as girl
(maiden), mother, and old woman (crone).

Mother-Father Deity
Several Western women's religions have modified traditional Christian male
god imagery, replacing it with a male-female deity. Androgynous deities in
women's religions, unlike the androgynes in Indian tradition, are not associ-
ated with androgynous physical characteristics (of. O'Flaherty 1980). It is their
moral and spiritual characteristics that make them androgynous.
In Christian Science God is androgynous, and referred to as "Father-
Mother." Christian Scientists believe in the equality of the sexes, both in this
world and in Spirit. Mary Baker Eddy, the founder of Christian Science, felt
that her own sex was essential to the nature of her mission—it had been given
to her to reveal the Motherhood of God. "In divine Science we have not as
much authority for considering God masculine, as we have for considering
Him feminine, for Love imparts the clearest idea of Deity" (Science and Health
1903, 517)- in Eddy's own words, "God could not be less loving than my
mother" (Peel 1966, 23). The Christian Science God is wholly good—Eddy
rejected the classic Christian explanation of evil, claiming instead that it is
impossible that the good God would have made man capable of sin, then
punished him for sinning (Gottschalk 1973, 65). Christian Science opposes any
anthropomorphic view of God; therefore God is described through language
of negations (not this and not that).
 No Father in Heaven 175
Like most Protestant theologies, Christian Science understands God as dis-
tinct from His creation—God and man are not the same. But whereas most
Protestants see man as not only separate in existence but also in nature from
God (our senses tell us that man, unlike God, is finite and imperfect), Christian
Science says that our senses deceive us—God is the Principle of all real being,
and that "in Science" man and the universe are, like God, perfect, spiritual and
eternal (Gottschalk 1973, 58).
Susan Setta (1977) has shown how Christian Science theology offered an
alternative both to societal views of women and accepted religious views of
God as male. According to the nineteenth-century view of women, a normal
woman was passive and dependent, and Calvinism offered few options to
women seeking transcendence. Women were not even allowed to speak or
pray out loud in church. The Christian Science Father-Mother God, who
incorporated both such "masculine" qualities as strength and such "feminine"
qualities as tenderness, could allow Mary Baker Eddy and other women to
become active and strong. By positing a Father-Mother God, Eddy could
proclaim that masculine and feminine, men and women, are equally impor-
tant within society. Just as God has masculine and feminine properties, men
and women should have a balanced combination of masculine and feminine
attributes.
Like Christian Scientists, Shakers attributed theological significance to the
gender of their leader. Shakers believed that Ann Lee's femaleness was neces-
sary to complement the maleness of Jesus. And like Christian Scientists,
Shakers believed in a Mother-Father deity. The source of this image seems to
lie in two rather different directions. Among the Shakers the notion of dual
male-female leadership (see Chapter II) developed before the notion of the
Mother-Father deity (Procter-Smith 1985, 6). This is a case in which a rela-
tively egalitarian social structure encouraged the development of a relatively
egalitarian theology. I2 In addition, the characteristic traits of the female aspect
of the Shaker God—Holy Mother Wisdom—seem to have been influenced
by nineteenth-century notions of femininity—motherhood, passivity, love,
forbearance, chastity, meekness, and virtue. However, Sally Kitch makes the
interesting point that in Shaker theology gender traits are not consistently
associated with male or female aspects of Divinity; for example, wisdom is
sometimes associated with Holy Mother and sometimes with Father (1989,
173). Let us note that the dual gender composition of Shaker divinity should
not be confused with divinity that is bifurcated in terms of good and evil.
The Shaker God was held to be wholly good and merciful to human beings,
and to enlighten everyone without distinction of sex or class.
During the mid-nineteenth century, Shaker communities experienced an
intense and ecstatic spiritual flourishing known as Mother Ann's Work. In this
period the female aspect of the Deity—Holy Mother Wisdom—was the most
fully developed. When Holy Mother Wisdom communicated, a woman was
needed to be her instrument or medium. This served to reinforce and legiti-
mate the role of women in Shaker worship. "As women became increasingly
176 Priestess, Mother, Sacred Sister
prominent as instruments [mediums], female symbols became increasingly
prominent in the symbol-system of the sect" (Procter-Smith 1985, 196). Yet
Procter-Smith has suggested that the prominence given to the female deity did
not really serve women's interests. "Although the prominence of a powerful
female aspect of God undoubtedly was an empowering experience for many
Shaker women, especially the instruments [mediums], it had the effect of
reinforcing the strict division of labor and sexual stratification which made
women peripheral in the first place" (1985, 208).
Rosemary Ruether, on the other hand, has argued that although female
divine figures do not automatically further the status of women, in religious
groups founded by women—such as the Shakers and Christian Science—they
do promote the equality of women. In fact, we know of so few examples of
women's religions with either goddesses or androgynous deities that both
Procter-Smith's and Ruether's hypotheses remain unprovable.
It is important to ask why androgynous deities were attractive to many
nineteenth-century North American women (and some men). I see the answer
as lying in two directions. First, in nineteenth-century North America religion
was increasingly deemed to be a female concern. Even male ministers were
suspect for entering a profession that was "too feminine" (Welter 1974). Dur-
ing the nineteenth century even the mainstream Protestant denominations ex-
hibited a tendency to emphasize the more "maternal" aspects of Jesus. It may
be that an all-male deity was incompatible with a cultural assessment of reli-
gion as feminine. On the other hand, femininity in nineteenth-century Amer-
ica was defined as weak and passive. It also may be that an all-female deity was
incompatible with a cultural assessment of women as powerless. In such a
situation, an androgynous deity was a perfect solution. It could incorporate the
"desirable" aspects of culturally construed femininity with the "desirable" as-
pects of culturally construed masculinity, while remaining faithful to Christian
monotheism.
Ross Kraemer suggests that the association between women religious leaders
and androgynous deities may have a long history in the Christian world.
Drawing on what is known of Gnostic groups from the first centuries of the
Christian era, Kraemer finds that "there seems to be some correlation between
the perception of God as androgynous and the view that women could exercise
Christian office" (1992, 197).I3
The three nineteenth-century North American women's religions (Christian
Science, Shakers, Spiritualism) related to benevolent, androgynous deities who
embodied all the best qualities of mortal parents (of. Becker 1990 on Christian
Science and Tenrikyo). This vision had important implications. "Nineteenth-
century women refused to believe that a benevolent deity would cause precious
sons and daughters to be born knowing all along that he would condemn
some, if not most, of them to eternal punishment in hell" (Braude 1989, 39). I
see it as far from coincidental that all three of these religions offered essentially
positive visions of both this life and of the next world; such visions were fitting
for a parental deity.
 No Father in Heaven 177

THE GENDER OF THE DEITY

Table 3. The Gender of the Deity
In Burma there are both male and female nats, but more male ones. There is good evidence that
goddesses were preeminent in Southeast Asia before Buddhism penetrated the area (of. Ferguson
1982).
On the Ryukyu Islands the hearth kami is female. Whoever wields kami is kami, by Ryukyuan
thinking, so that priestesses and even wives and sisters are seen in some way as divine. Accord-
ing to Ryukyuan cosmogony, humans were created when a brother-sister divine couple de-
scended to earth and produced offspring. The emphasis on brother-sister pairs is common in
women's religions (see Chapter 10), and we may conjecture that brother-sister deities are a varia-
tion of androgynous deities.
In Afro-Brazilian cults there seem to be fairly even numbers of male and female possessing
spirits.
Male and female zar spirits behave differently. According to Yael Kahana (1985) hysteria is
usually believed to be caused by possession by female zar spirits who are silly and irresponsible
and cause the woman to behave in an infantile manner. Male zar spirits cause apathy and cata-
tonic states. It seems that possessing spirits are more often male than female.
Korean gods and ancestors include both males and females, yet from the reports of kut that I
have read in the ethnographic literature I get the sense that possession by female ancestors may
be more common than by male ones.
Spiritualist spirits are male and female (possibly more male spirits); the deity is androgynous.
Sande ancestresses are female.
Black Carib ancestors are male and female, but are addressed in ritual singing as "grandmother."
Northern Thai matrilineal spirits are not personalized, but may possibly be thought of as female
(I cannot be certain from the available ethnography). The spirits (caw) that possess urban medi-
ums seem to be mostly male.
Shakers imaged a Mother-Father deity.
Christian Science has a Mother-Father deity.
Feminist Spiritualists relate to a variety of Goddesses.

We will now briefly return to the polydeistic women's religions to review the
gender of the deities (Table 3). We immediately see that the gender of deities in
polydeistic women's religions is split fairly evenly; there is no overall prefer-
ence for female or for male supernatural entities.

CONCLUSION
What we cannot say about the deities of women's religions is possibly more
interesting than what we can say. None of the women's religions worships a
single, omnipotent male deity. In light of the insights of Mary Daly and other
feminist scholars regarding the connection between patriarchy and the "Great
King in Heaven," I do not find this surprising.
On the other hand, none of the women's religions worships a single great
goddess. 14 Even the Goddess "created" by Spiritual Feminists is far from being
178 Priestess, Mother, Sacred Sister

an all-powerful sovereign. Nowhere have we found the type of great mother

goddess described by Preston (see page 172 above). In light of the emphasis on
mortal motherhood in women's religions, this finding is certainly noteworthy.
My guess is that mortal mothers realize that maternity does not make them
omnipotent. The realities of motherhood lead women to diverse sorts of deities
that "fit" the diversity of mothering experiences. Motherhood is not a matter of
absolutes, but of particular sorrows, joys, decisions, and personalities. The
same factors that encourage the development of immanent deities in women's
religions discourage the development of either a male omnipotent god or a
female great goddess. Grounded in the here-and-now, in particular relation-
ships, the deities of women's religions tend to be as ambiguous as life itself.
A last point bears notice here. In most women's religions, supernatural
entities—both male and female—appear at rituals in the bodies of their almost
exclusively female spirit mediums. Since the deities who possess human beings
look like the humans they possess, even male spirits, gods, and ancestors look
female when the mediums are women. As we will see in the next chapter, it
would be a grave mistake to underestimate either the psychological or theologi-
cal implications of spirit possession.

Notes
1. In addition, Buddhist devas are sometimes referred to as nats, although they are
not part of the nat cultus.
2. This does not negate that other aspects of science had a great attraction for
Spiritualists.
3. There are other spirit types besides these four. These are the four major types.
4. If this description seems unclear, it is because the descriptions given by the native
informants and recorded by the ethnographers are also unclear.
5. According to Harris and Sawyerr (1968), the Mende have a belief in some kind of
one high god. In addition, they believe in and relate to a large assortment of ancestral
and nature spirits.
6. They were not able to determine which causes which; the relationship seems to
be reciprocal and subsistence patterns affect both.
7. On the other hand, some historians might simply claim that monotheism is so
rare in the world (after all, there are only three truly monotheistic religions) that
explaining why female-dominated religions are polytheistic is not even an interesting
question; most religions are polytheistic. Other historians of religion disagree with the
claim that monotheism is rare. Commonly cited examples include primitive monothe-
ism (high gods in African religions) and certain streams in Greek and Hindu religion.
8. I recommend David Wulff's paper "Prolegomenon to a Psychology of the God-
dess" (1984) for an acute critique of the Freudian view of goddesses.
9. A deeper inquiry into goddesses should take into account the effects of specific
social conditions. Jane Harrison (1955, 260-261), for example, sees a link between
matrilineality and independent goddesses, and between patrilineality and subordinate
goddesses. Other scholars have argued that there is some correlation between women's
status and the presence of goddesses, yet are vague about exactly how that correlation
works. For example, according to Tikva Frymer-Kensky, an expert in ancient Near
Eastern religion, "The eclipse of the goddesses was undoubtedly part of the same
 No Father in Heaven 179

process that witnessed a decline in the public role of women, with both reflective of
fundamental changes in society that we cannot yet specify" (1992, 80).
10. This is a line from Ntosake Shange's play For Colored Girls Who Have Considered
Suicide When the Rainbow Is Enough.
II. In The Grandmother of Time, Z. Budapest explores and meditates on a different
goddess for each month of the year. The goddesses she chose reflect the range of
traditions from which Feminist Spirituality self-consciously draws its symbols: Greek
goddesses, Swedish goddess, Japanese goddess, Celtic goddess, Egyptian goddess,
and more.
12. Neither Shaker belief nor social structure were wholly egalitarian (see the discus-
sion of Shaker gender ideology in Chapter 10, and Brewer 1992, 612).
13. Moving to Japan, we learn that the female founder of the new religion Tensho-
kotai-jingu-kyo proclaimed her own body to be the dwelling place of both a goddess
and a god who unite with her to form an androgynous trinity (Nakamura 1980, 140).
14. The closest we come to this is the Tetum earth mother (see Appendix), out of
whose vagina humanity climbed and into whose womb the dead return. Yet even
among the Tetum there is a passive, masculine deity who resides in the sky. Unlike the
earth mother, he is uninterested in the affairs of humans, and humans likewise are
devoid of interest in him (Hicks 1984, 6).
 9
Summoning the Spirits

The possessed woman herself was unaware of everything. She was taken by her
god against her will, and so when the god first started to ride her, she bucked like a
wild horse. Her face set rigidly, her body jerked wantonly, she lost her balance,
and she conveyed a general atmosphere of tension and pain. . . . But soon the
horse [medium] was tamed and began to execute superbly, in deep trance, the
dance of her god. And now, animated by the deity, she was the deity, and the ekedi
[priestess] led her off to dress her in the luxurious clothes of divinity. For the next
twelve hours or so, ... the woman lived in a trance, harnessed by the injunction
to dance for the drums, forbidden to eat, drink, or relieve other wants. I never saw
one cough or scratch, yawn or stretch, or go to the closet. The entranced creatures
responded only to signals from the mother [chief priestess] and the drums (Landes
1947, 54)-
Alourdes [a Haitian Vodou priestess] invariably diagnoses the origins of her cli-
ents' problems as disturbances in relationships of one sort or another. . . . Most
problems are diagnosed as supernatural, and because they fall into the realm of the
spirits, something can be done about them. ... [In Alourdes' words] "You read
the Bible, you read about God. But Spirit you don't read about. The spirit come in
people' head [possession], and you see them, you talk to them. They help you"
[BROWN (1991). 346-347].

WOMEN, TRANCE, AND POSSESSION

Many religious traditions embrace the belief that spirits or gods can enter the
bodies of individuals who are in states of trance and then engage in activities typ-
ical of the gods' personae. Possession trance is a pivotal component in the major-
ity of women's religions. In nine out of our twelve key examples, possession
trance occurs either frequently or occasionally in the context of ritual. Yet lest
we are tempted to exaggerate the import of that statement, we must point out
that the ethnographic record reports the presence of trance states in most
cultures (52% in Bourguignon's 1976 sample). Trance is like prayer or
sacrifice—a type of ritual behavior that occurs in many or possibly most
religions and so is not in need of special explanation in women's religions.
Both male-dominated and female-dominated religions utilize techniques of
dancing, swaying, spinning, drumming, meditation, and singing to induce
trance.
182 Priestess, Mother, Sacred Sister
Erika Bourguignon does, however, comment on one interesting gender
difference. Trance—an altered state of consciousness induced by a number of
well-recognized techniques and not involving possession—is reported more
often for men. Possession—the belief that supernatural beings can enter human
beings, that humans can temporarily become gods, spirits, demons, or
ancestors—is reported more often for women (1976, 1983). Put differently,
what is more common among women is not a dissociative state known as
trance, but a cultural interpretation known as spirit possession.
While spirit possession is an essential element in the ritual complexes of many
women's religions, it by no means occurs in all our examples (Table 4). At
various ritual occasions Sande members wear masks and costumes representing
Sande ancestresses. Spirit possession, if it occurs, is rare during the rituals. 1 In
the Feminist Spirituality Movement techniques such as meditation, dance, and
chanting are used to induce altered states of consciousness, and rituals that
dramatize the goddess-nature of mortal women are common. Spirit possession,
however, has not been reported in the vast ritual literature of the Movement.
And spirit possession is most definitely not a part of Christian Science ritual,
where the emphasis is on rational discourse. In short, it is legitimate to argue that
women have some sort of affinity for spirit possession (or that spirits have some
sort of affinity for women) but that this affinity is neither absolute nor universal.

THEORIES OF WOMEN AND SPIRIT

POSSESSION: CONCEPTUALIZING
GENDER
In many cultures women are believed to be particularly skilled at, or prone to,
possession trance. Anthropologists have offered a number of explanations for
this propensity. As early as 1966, I. M. Lewis argued that where women are a
disadvantaged and subordinate group (and that is the case in most societies),
possession trance is a tool for them to obtain material, emotional, or social
benefits that would otherwise be unavailable. Possession allows them to pro-
test their lowly position as women in sexist society. Women are frequently
possessed by male spirits that may allow women, at least temporarily, some of
the advantages of being a man. Lewis has refined this argument in his 1986
book Religion in Context, but not substantially changed it.
Other writers have stressed the sexual element of possession trance: The
possessed women, often with abandoned and ecstatic facial expressions and
body movements, is said to have sexual intercourse with the possessing god or

Table 4. Women's Religions That Embrace Spirit Possession

Burma Nat Cult Ryukyu Island Religion
Zar Religion Korean Shamanism
Afro-Brazilian Religions Northern Thai Spirit Cults (urban)
Spiritualism Black Carib Religion
Shakers
 Summoning the Spirits 183
spirit. In cultures in which a sexual double-standard means that many women
are sexually deprived for large portions of their lives, possession is a way for
some women to seek sexual satisfaction (Spiro 1967).
Both the sexual deprivation explanation and the social deprivation explana-
tion embody all the problems with deprivation theory in general that I out-
lined in Chapter 2 (I will not repeat them here.) Still other anthropologists
have claimed that women's propensity for possession trance is a biochemical
matter; that trance symptoms are in fact the symptoms of calcium depriva-
tion suffered by pregnant and lactating women in societies in which women
have limited access to calcium-rich foods (Kehoe and Giletti 1981). Bourguig-
non, Bellisari, and McCabe utterly discredit this hypothesis, and show that
"although impressionistic accounts suggest that women often predominate in
possession trance cults, the simple equation women = possession trance, is
not justified by the available data" (1983, 414). I would add that the voluntary
nature of possession trance—women attend rituals so as to induce trance and
call down the spirits—argues against physiological explanations that assume
spirit possession is involuntary. 2
Another school of thought focuses upon cultural notions of gender-appro-
priate behavior. Kevin Neuhouser (1989) for example, argues that spirit posses-
sion cults attract more women than men in Brazil because it is not "masculine"
to be submissive to a dominant spirit. Similarly, Melford Spiro notes that
Burmese men have other alternatives that may be functionally equivalent to
being a shaman—for example, monasticism (1967, 224). On the whole, I also
lean toward those sorts of interpretations that carefully situate spirit possession
in particular cultural contexts, and that presuppose intelligent and religious
motives on the part of women involved in spirit possession.
Given the predilection of women for spirit possession, the high incidence of
spirit possession in women's religions, and the rather rigid cultural conceptions
of gender in many of the societies in which spirit possession occurs (see also
Chapter 10), we note with interest that women are often possessed by male
spirits. In the Afro-Brazilian cults, for example, the typically female medium
becomes the sometimes female, sometimes male spirit. According to Seth and
Ruth Leacock (1972), Brazilian Batuque women mediums often engage in
"male" behavior such as smoking cigarettes or cigars, drinking alcoholic bever-
ages, and shouting vulgarities. (The entranced medium is unaware of all this
and so cannot possibly be enjoying herself!)
Vieda Skultans cites a 1886 study showing a preponderance of Spiritualist
mediums to be female, but a majority of possessing spirits to be male. "In
other words, the typical spiritualist experience involves a female medium and a
male spirit or control" (1983, 17). It may well be that some female mediums
enjoyed the opportunity to take on a male role during trance. Favorite control-
ling spirits included swearing, highly sexed, formally educated, and physically
strong men, who gave the mediums a chance to break out of the conventional
and often stifling nineteenth-century female role. Laurence Moore argues that
part of the satisfaction felt by women mediums derived from "their assump-
tion during the trance state of an otherwise forbidden male social role. Time
184 Priestess, Mother, Sacred Sister
and again under the influence of their spirit controls, they turned into swearing
sailors, strong Indian braves, or oversexed male suitors" (1977, 111).
Janice Boddy has developed a subtle and fascinating interpretive analysis of
role play in zdr spirit possession. She argues that zar, when viewed as an
aesthetic genre, serves to open thought, to free it from limitations of prior
associations, to pose challenging problems, and encourage reflection on the
everyday. "Symbolic inversion ... or negative metaphor ... is its essence:
during possession rites women become men; villagers become Ethiopian, Brit-
ish, Chinese; the powerless and impoverished become powerful and affluent.
Essentially irreversible processes—genderization, aging—become reversible;
established categories are undermined. Hierarchical orderings are telescoped
and undone when Islamic holy men and pagan prostitutes possess the same
[Sudanese] woman" (1989, 306).
The entranced woman ceases being herself; instead, she experiences the
world the way her possessing spirit does. Boddy quotes one of her informants:
"When it descends into you . . . you see through the eyes of the European. Or
you see through the eyes of the West African, whichever spirit it is. You see
then as a European sees—you see other Europeans, radios, Pepsis, televisions,
refrigerators, automobiles, a table set with food. You forget who you are, your
village, your family, you know nothing from your life. You see with the eyes
of the spirit until the drumming stops" (1989, 350).
In the course of a zar ritual, and over years of participation in zar rituals, a
woman both observes and experiences trance and so both observes and experi-
ences other cognitive and behavioral possibilities. "Zar (as both possession and
performance) is a powerful medium for unchaining thought from the fetters of
hegemonic cultural constructs" (1988, 23). When a woman is possessed she
forgets who she is, she temporarily steps outside her usual world, she sheds her
culturally constructed personhood. "In proportion to her subjective experience
of otherness, her everyday reality is made to appear as one of many [possible
realities]—less naturalized, less unquestionable, indeed, less subjectively real"
(1988, 19).
Why is this process particularly attractive to women? Boddy contends that
in the cultures where zdr occurs, gender, for women, is overdetermined. A
woman is never "just" a person; the social facts of gender determine all her
daily activities, life events, and interpersonal interactions. Thus women, more
than men, enjoy stepping outside their social identities to experience other
human possibilities.
In an insightful review of Boddy's book, I. M. Lewis has pointed out that
although her theory is attractive, she "presents no evidence to show that, as a
result of the intellectual re-focusing or re-framing . . . possessed women do
think and feel differently" (1990, 590). I accept Lewis's critique, and sincerely
hope that future studies will rectify this deficiency. On the other hand, I believe
that Boddy is justified in her claim that spirit possession has the potential to
teach that "gender is not a natural attribute but a cultural construct" (Boddy
1988, 21). At the very least, in contrast to Islamic and Christian rituals that
tend to present unambiguous cultural norms and hierarchies, possession rituals
 Summoning the Spirits 185
allow participants and audience to engage in "the dialectic expression of an
entire range of often contradictory elements which form the socio-cultural
universe of the society" (Giles 1987, 250).

SPIRIT POSSESSION IN FEMALE-

DOMINATED RELIGIONS

Women's religions offer two chief models of spirit possession: specialist posses-
sion and lay possession. In Korea, Burma, the Ryukyu Islands, Thailand,
Shakerism^, Black Carib religion4 and Spiritualism there are women who are
experts or professionals at contacting the spirits or gods, and who use their
expertise to divine, heal, solve problems, or contact the ancestors or gods on
behalf of other people. In zdr cults and Afro-Brazilian religions, possession
trance occurs at group rituals in which some, most, or all participants become
possessed. These two models do not constitute an absolute dichotomy. In the
Afro-Brazilian cults and the zdr cult there are women who are "lay" mediums,
and other women who are expert priestesses or "Mothers". Even within the
specialist model we can identify a variety of levels of expertise, ranging from the
spontaneous possession of young Shaker women during the period of Mother
Ann's Work, to the cultivated and highly controlled possession of urban Thai
spirit mediums. Still, I do think it is useful to acknowledge the difference be-
tween mediums who become possessed in order to teach or heal other people,
and mediums who become possessed as part of a group religious experience.
Two other types of spirit possession do not occur in women's religions. In
some male-dominated religions, women become possessed by evil spirits who
must be exorcised by male specialists—the entire process underscoring female
inferiority and male power (cf. Bilu 1987 on Eastern European dybbuks). In
certain other male-dominated religions, possessed woman serve as oracles
whose words are interpreted by men (e.g., Delphi). It is significant that these
models are not found in female-dominated religions.
Spirit possession in women's religions incorporates both sacred and medi-
cal dimensions (Csordas 1987). Possession by the spirits offers a culturally
recognized and meaningful interpretation of and solution to chronic misfor-
tune. The healing function of spirit possession should not be taken to mean
that female-dominated religions are havens for maladjusted, neurotic, or psy-
chotic women. In women's religions (and in some male-dominated religions)
overtly wild behavior during rituals is seen as a sign that the person is drunk
or insane, not possessed. Because a spirit must make sense to those whom it
encounters, "successful negotiation of the possession context requires the
patient to have or develop considerable cultural awareness. It is thus inapplica-
ble to those who suffer severe psychological disturbance" (Boddy 1989, 146).
A final comment about spirit possession in women's religions is one that
may be self evident, but nonetheless critical. The audience at spirit possession
events is predominantly female. Spirit possession may fruitfully be inter-
preted as a ritual drama with women—not men—as the audience. Possession
186 Priestess, Mother, Sacred Sister

in female-dominated religions has a great deal to do with relationships among

women. This may include competition for status or resources (cf. Wilson's
1967 argument that spirit possession is correlated with social situations which
give rise to tension between members of the same sex), enhancement of social
or kinship bonds that are not always recognized by the dominant male culture,
and healing of the many physical and emotional hurts from which women qua
women suffer.
Janice Boddy has reached a similar conclusion regarding the zar. "[Ojnce the
idiom of possession has been invoked, people are freer to communicate in
ways antithetic to the harmony-preserving tactics of everyday discourse. Close
kin, spouses, affmes, might indirectly discuss issues which otherwise could not
be broached without injuring their relationship. And this is possible because of
the potential for obfuscation inherent in the possession idiom: the distinction
between human self and zar self is not rigidly drawn, even during possession
trance. Responsibility can be assigned to zayran, to humans, to both. It is this
latent confusion . . . that permits vitally important messages of a sort not
normally countenanced to be transmitted without permanent rupture to the
social fabric" (Boddy 1989, 236).

SHAMANS AND MEDIUMS

Throughout this book I have been repeating the nomenclature for religious
leaders used by the experts upon whose work I draw. With the exception of
Melford Spiro who directly transliterates the native word natkadaw, most other
scholars seem to use the terms "medium" and "shaman" in a somewhat vague
manner.
In descriptions of shamans in diverse cultures, one of the common features
(although not, as I. M. Lewis has informed me, a necessary feature) is an
ecstatic flight of the soul to a different world (plane, dimension, etc.) in which
the individual undergoes unusual experiences and learns portentous informa-
tion that she or he brings back to this world, and which allows him or her to be
a healer.5 The pervasiveness of ecstatic flight makes it crucial to point out that it
most definitely is not a feature of religious specialist roles in women's religions.
In none of the women's religions described in this book does the specialist
experience an ecstatic flight in which the soul travels to other realms, even if
the western ethnographer used the word "shaman" to describe the religion.
Instead, what the so-called shamans in women's religions seem to do is
incorporate into their own bodies other personae (gods, spirits, ancestors,
nats, etc.). This process is what in the anthropological and religious studies
literature is known as being a medium—a channel for a supernatural being
who wishes to communicate with beings in this world. Cross-culturally,
there does seem to be compelling evidence to associate women with spirit
possession (mediumship) and men with ecstatic flight of the soul (sha-
manism). For example, in Janelli and Janelli's overview of spirit communica-
tion in East Asia, we learn that in Korea, Japan, and parts of China female
 Summoning the Spirits 187
specialists become possessed, whereas in Taiwan male specialists travel to the
other world to speak to ancestors (1982).
In his recent book, I. M. Lewis proposes a new way of looking at spirit
possession and ecstatic flight of the soul. "All these features, which others have
seen as separate self-sustaining styles of religiosity, are in reality constituent
elements in the composite shamanistic complex" (1986, 85). In other words,
they are serial phases in the assumption of the shamanistic career. Lewis cites
several examples from the ethnographic literature demonstrating that spirit
possession and astral voyages often overlap. Even more important, he argues
that the shamanistic career typically begins with unsolicited and uncontrolled
possession, yet with practice and experience the possessed individual learns to
"master" the spirits and become a true shaman. Lewis concludes that both
historically and in particular situations, "It does not matter how the balance
between possession and soul-flight is pitched" (1986, 92).6
While I am tempted to agree with Lewis on this, I am not sure that I can, in
good conscience, assent to his entire thesis. The essence of Lewis's argument
seems to be that there is no real distinction between descent of spirits and
ascent of the soul. However, the very examples that he cites to prove his case
seem to suggest that this distinction does in fact exist, and is at least to some
extent linked to gender. For example, quoting Carmen Slacker's study of
Japanese shamanism, Lewis tells us that in feudal Japan "it was common to
find ... an ascetic [what Lewis calls shaman] husband married to a female
medium" (quoted in Lewis 1986, 87).
Rather than extinguishing the ecstatic flight versus possession controversy,
Lewis's analysis sharpens it. If, as Lewis claims, the same individuals are
capable of both possession and ecstatic flight, why is it that religious special-
ists in female-dominated religions consistently opt for the former and not
the latter?
I do believe that there is a significant difference between flight and in-
dwelling, between what in other chapters I label transcendence and imma-
nence. Whether that difference is ontologically pure or whether it is merely a
matter of accent is beyond the scope of our present inquiry. In sum, despite my
inclination to be swayed by Lewis's recent work, I am unwilling to gloss
leaving one's body with sharing one's body. The implications of this distinc-
tion will become apparent in the following section.

GENDER, EGO FORMATION,

AND SPIRIT POSSESSION
Laurence Moore has observed regarding Spiritualists, "The success of spirit
communication depended on the ability of mediums to give up their own
identity to become the instruments of others" (1977, 106). The essential trait of
the medium (and not only in Spiritualism) is the ability to loosen her (or his)
ego boundaries and literally share her or himself with another (or many other)
beings. It seems to me that an individual must be extraordinarily comfortable
188 Priestess, Mother, Sacred Sister
with interpersonal relationships in order to do so. This realization brings us
back to the argument I have been making throughout this book concerning
gender and modes of relating. Yet before I continue, I wish to reiterate a point
that frequently gets lost in discussions of gender and psychological develop-
ment: Theorists such as Nancy Chodorow and Carol Gilligan do not claim that
men and women are essentially different; rather, they assert that in specific
(albeit very common) cultural circumstances, men and women are socialized
into different behavioral modes, different moral voices, and different emo-
tional proclivities. Thus, what follows should not be taken as a discussion of
sex (biologically given) but as a discussion of gender (culturally constructed).
Sharon Conarton and Linda Silverman argue that "The developmental cycle
of women must be viewed with the awareness that women's primary striving
is for relatedness and connection. . . . Ego formation is different for a male
because he is more aware of his boundaries. The boundaries of the feminine
ego are very thin [my emphasis]. After experiencing herself as an extension of
the mother, it is hard for the girl child to know where she stops and the other
begins" (1988, 46). Drawing on Nancy Chodorow's writings and her own
clinical practice, Lillian Rubin postulates that the normal development of gen-
der identity in a girl requires no wrenching breaks with the past. As a woman
she will have ego boundaries that are less rigid, more permeable [Rubin's
words] than a man's, because she does not need defenses against feeling and
attachment. Girls never need to separate themselves as completely as boys do.
Girls experience themselves as more continuous with others. The capacity for
participating in another's inner life is more developed among women. For the
same reasons, a girl develops a more complex inner life than a boy (1983, 58-
59). All of these ideas were influenced by David Gutmann's definition of two
types of ego: the allocentric masculine ego that tends to objectify and experi-
ence its own separateness from others, and the autocentric feminine ego that is
characterized by more permeable boundaries between self and others and be-
tween self and environment (1965). "Indeed, naturalistic observation might
indicate that female domestic and maternal competence requires those very
qualities of impulsivity and personalization which violate our criteria for ego
adequacy. For example, careful observation of the mother-child relationship
might show that the tendency to personalize the world, to abolish self-other
boundaries, is essential to good mothering. . . . Indeed, when women main-
tain firm ego boundaries and treat their children as men treat the impersonal
objects of the geometric-technical world, we call them either rejecting or
schizophrenogenic" (Gutmann 1965, 239).
The developmental process described by Gutmann, Chodorow, Rubin,
Conarton, and Silverman is highly relevant to understanding spirit possession.
Unlike in non-possession trance or in classic shamanism, the essence of spirit
possession is the willingness to share one's body with another being. In spirit
possession, as Janice Boddy has observed, "[Although the identities of host
and spirit are distinct and even here remain functionally independent, it appears
that they coalesce in possession trance. Both host and spirit are present in the
host's body to varying degrees" (1989, 151). The classic (typically male) sha-
 Summoning the Spirits 189
man, on the other hand, transcends his body—a role Nancy Chodorow and
others would surely see as suitable for individuals who are raised to develop
their gender identity and sense of self as negation rather than connection.
The etic (scientific) argument I have outlined above is strengthened by the
various emic (native) explanations for why women are possessed more fre-
quently than men. In many cultures in which women are believed to be particu-
larly skilled at, or prone to, possession trance, the indigenous interpretation is
that women are softer, easier to penetrate, and that is why gods or spirits
choose women as their vehicles ("horses"). This is the case in Brazil, for
example. Writing about Thailand, Stanley Tambiah notes that although men
fill almost all ritual specialist roles, there is an opening for women in roles
involving possession because women are seen as more prone to possession;
spirits possess them because they are soft and vulnerable and therefore effective
hosts (1970, 283). In the Northern Thai matrilineal spirit cults women predomi-
nate as mediums, and women are believed to have an inferior sort of "soft
soul" that makes them more susceptible to the effects of spirits (Tanabe 1991,
189). Similarly, in certain Japanese religions, "Women are more fitted to be
mediums than men, because they have stronger feelings, better intuition, and
more sensitive emotions" (Offner and Van Straelen 1963, 124). I am struck by
the similarity between the emic explanations of women's preponderance in
spirit possession, and psychological explanations of gender differences in ego
formation.
In the Burmese case the difference between the female natkadaw and male
exorcists is instructive. The woman religious specialist works by propitiating
nats, soliciting their help, asking them to possess her, and acquiescing to their
power. The male exorcist combats and attempts to drive away nats, using
Buddhist power to do so. The natkadaw is a person through whom the nats
speak, while the exorcist is one with whom the nats speak. The natkadaw shares
her body with the nat; the exorcist remains separate—confronting the nat
(1967, 243).
One could argue that the possessed medium does not "relate" to the spirit;
instead, her own identity disappears and is replaced by the spirit. However, a
careful reading of ethnographic accounts of spirit possession suggests that it is
only inexperienced and unskilled mediums who totally disappear. More skilled
mediums, in the words of one Black Spiritual minister, "entertain" the spirit.
"When I'm in the spirit, if I'm dancing in the spirit, I'm asking [the spirit to
intercede on my behalf]. I'm working with my spirit. I know what to say,
what to work with. See, when I finish dancing, I know what's going to
happen" (Jacobs and Kaslow 1989, 132). Writing about the bori cult, Murray
Last makes a similar observation: "What is being taught by this training in
possession is how to put on, like a suit of clothes, a further social self, while not
putting in jeopardy one's own identity" (1991, 55).
I simply cannot close this section without reference to the obvious parallels
between pregnancy and spirit possession. Hilary Graham has argued that the
two phenomena are indeed homologous. I cite several of the analogies identi-
fied by Graham: "In both, the individual's body is seen to be invaded by and to
190 Priestess, Mother, Sacred Sister
interact with an alien being, by virtue of which both actor and being occupy
marginal, indeterminate social positions. ... In both, the presence of this alien
provides a sufficient explanation of the actor's behavior. ... In both, society
typically manages instances of possession not by expelling the spirit (aborting
the foetus) but by domesticating it, by channelling its demands into socially
expected forms" (1976, 296). In short, I speculate that the same socialization
process that constructs adult women who seek to share their bodies with their
babies, also sometimes constructs adult women who willingly share their bod-
ies with spirits, ancestors, or gods. The process of pregnancy—of sharing
one's body—may be a suitable training process for spirit mediumship. Let us
recall that in several women's religious situations (Spiritualism, Cantonese
shamanism) the spirit guides are actually the dead children of the women
mediums.
Ross Kraemer has suggested that it is useful to view spirit possession as a
gendered, embodied metaphor rooted in women's physical experiences of het-
erosexual intercourse and pregnancy. That is, women seem to experience pos-
session as an analogue of heterosexual intercourse, pregnancy, and more
broadly, marriage and relationships with men: "It's not inherently bad, it has
its advantages and disadvantages, and you can usually find some way to accom-
modate it!" (personal communication, 1993). If Kraemer's argument regarding
the inherently gendered nature of spirit possession is correct, it could well
explain why men often treat spirit possession as demonic, and why possessed
men are often concerned with exorcism rather than accommodation; men
would experience possession as "unnatural." Since my own inclination is to
avoid essentialist, biologically based arguments, I find it more helpful to em-
phasize the socialization process that produces women who are willing (and
often desperately eager) to accommodate "possession" by men, by babies, and
sometimes by spirits.

MEETING THE DEITIES

If we take stock of the explanations of women and spirit possession that we
have seen in this chapter, an interesting pattern becomes evident. All the theo-
ries I summarized in the beginning of the chapter (social deprivation, sexual
deprivation, calcium deprivation, and overdetermination of gender) start from
the assumption that possession trance is an abnormal phenomenon. Therefore,
the explanation for women's involvement with 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 way that
musical ability or athletic ability is? Could it be that in many cultures male
socialization prevents most men from developing the ability to embrace the
enriching, exciting, normal experience 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 or their sense of self from the
threat of "invasion" that they reject, or refuse to recognize, a religious experi-
 Summoning the Spirits 191

ence that involves melding one's being with another entity? 7 As Janice Boddy
writes, "It is imperative to ask why so many Western scholars . . . are commit-
ted 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 weaknesses of
ethnocentrism and androcentrism.
I have chosen to look at spirit possession in detail for a number of reasons.
First, I am struck by the uses to which possession trance is put in women's
religions: to build and strengthen interpersonal relationships 8 , to deconstruct
gender, and to heal. Spirit possession encapsulates the particularistic, this-
worldly orientation of female-dominated religions.
A second reason has to do with the role of the audience at spirit possession
rituals. From the perspective of the audience (mostly female in women's reli-
gions), the essence of spirit possession is the willingness to interact with one's
deities face to face (see Finkler 1986, 638). Such willingness reflects a high level
of ease with interpersonal relationships, and a facility for perceiving the sacred
in the midst of everyday reality.

Michael Lambek has studied spirit possession in Mayotte, an island in the Indian
Ocean off the coast of Africa, between Madagascar and Mozambique. 9 On
Mayotte more than one third of adult women are possessed, while only one tenth
of adult men are possessed, and more non-possessed women than men assist at
curing ceremonies and are spectators in public ceremonies. Lambek emphasizes
the social nature of relationships with spirits. A host and spirit can have a relation-
ship that lasts for many years. According to Lambek, "Spirits rise precisely in
order to interact with people" (1981, 3, my emphases). The point of spirit posses-
sion is not worship—Mayotte women do not pray to the spirits. The point of
possession is social interaction between humans and spirits. On Mayotte spirit
possession occurs not only in public ceremonies, but also at home in informal
settings. When a woman enters into a relationship with a spirit, her husband also
becomes involved with the spirit. In time, a spirit may rise on its own volition
outside of ritual contexts, or the husband may decide to call it up. "[Mayotte]
spirits are not merely classified, they are socialized; they develop particular identity
relationships with particular people, and they are taught the norms of social inter-
course" (1981, 149-150).

More than any other kind of ritual, spirit possession dramatizes an ideol-
ogy of the immanence of the divine: In possession trance the god, spirit, or
ancestor is actually present on earth—incarnate and able to engage in dialogue
with devotees. Recalling that mediums typically add a bit of their own person-
alities to the spirits whom they enact, what we are describing is an interactive
model of the supernatural—the medium and the spirit reflect one another's
personalities.
Thomas Csordas suggests that spirit possession "can be seen as a pure form
of ritual drama, where the parts of deities are not played by humans, but where
the deities in effect play themselves" (Csordas 1987, 6). With this in mind, we
turn here to one last example.
192 Priestess, Mother, Sacred Sister

A noteworthy aspect of Ryukyuan belief is the pattern of treating certain

individuals as if they were gods, even to the point of calling them by the god's
name (see Chapter n). Yoshinobu Ota (1989) cites the traditional oaroari rite to
transform a woman into a priestess: During the rite the spirit of the deity
possesses a woman who has three grains of rice on her head, while she chants
the deity's name three times. According to Ota, the essence of becoming a
priestess is the act of personifying the deity. The implications of spirit posses-
sion among Ryukyuan priestesses are vividly recounted by Shinobu Orikuchi
(quoted in Kamata 1966, 61), "Since there is no clear-cut distinction in
Okinawa religion between the gods and the priestesses and female functionar-
ies, they are often directly referred to with the divine names."
I quote here at some length from Hisako Kamata's description of ritual
specialist women known as uy igami—women of around fifty years of age who
inherit their position from mothers or mothers-in-law.

[T]he uyigami have come to possess a sacred nature in the consciousness of the
islanders; they are gods during the period of the feast, at least in their imagina-
tion. Thus, they call them uyigami, ancestor-gods. The old women, in their turn,
regard themselves as more than human beings, and if they see a human being
approaching them, they utter words, "uru-uru-uru," and scare him away. They
are in a state of divine possession, wearing a crown made from a vine" (1966, 65,
my emphases).

I have cited these passages to underscore an issue raised earlier: From the
point of view of the audience, the possessed individual is the deity. Deities are
seen to be physically present on earth, and seen to look like—to wear the
bodies of—women.

Notes
1. According to Carol MacCormack, Sande masked ancestresses do not become
possessed, but occasionally a member (usually a younger woman) goes into a state in
which she communicates with the watery underworld of the ancestresses, sometimes
bring up a new mask or other object (personal communication 1993).
2. In most women's religions (e.g., the zar cult) a woman is not diagnosed as
possessed because she became entranced; rather, she initially seeks trance to find out
who is possessing her. Later on, trance gives her spirits a chance to play around in the
human world.
3. I am referring here specifically to the period of Mother Ann's Work when medi-
ums possessed by Holy Mother Wisdom (and other spirits) transmitted messages,
prayers, and rituals to the larger group. During other periods many if not most Shakers
experienced various forms of religious ecstasy, but I would not consider these other
forms to fall within the accepted definition of spirit possession.
4. Possession by ancestors is part of Black Carib rituals, especially important rituals
such as dugii. Ancestors temporarily use the bodies of the living to enjoy themselves and
to speak. More women than men are possessed. According to Taylor (1951), most
women who become possessed are between the ages of eighteen and twenty-five. There
also are shamans who contact the spirits to determine why someone is sick or suffering
 Summoning the Spirits 193

misfortune, and why the ancestors are unhappy. The shaman enters some type of trance
during the ritual. Today all practicing shamans in Belize are women (Kerns 1983). In the
past, however, it seems likely that more shamans were male (Taylor 1951).
5. Anthropologists and historians of religion have devoted a great deal of attention
to defining what is a shaman and determining whether the category "shaman" is cross-
culturally applicable or whether it describes a very specific type of religious specialist
found in and near Siberia.
6. Lewis's book (1986) includes a careful and interesting discussion of shamans
versus mediums. He begins by summarizing and critiquing Mircea Eliade's view that
possession by spirits is a later and diluted form of the authentic and archaic shamanistic
celestial ascent. He also cites and disproves Pater Schmidt's contention that spirit posses-
sion (what he called "black shamanism") was associated with early matriarchy, while
superior "white" or true shamanism developed under later patriarchy.
7. Additionally, one might argue that external social institutions prevent men from
experiencing possession trance. For example, Lambek explains the predominance of
women in spirit possession in Mayotte in terms of men's involvement with Islam,
which restricts their participation in spirit possession (1981).
8. To take one short example, according to Jacobs and Kaslow's study of the Black
Spiritual churches of New Orleans, "the bonding which develops in the group is
reinforced by the flurry of activity with which members come to one another's aid
either to prevent injuries or fan and comfort an individual in the aftermath of posses-
sion" (1981, 108).
9. Lambek sees Mayotte spirit possession as part of the same general phenomenon
as the zar cult (1981, 35 and 194 n. 3). Mayotte fits in well with our general picture of
women's religions being associated with matrifocal societies in which women's status is
relatively high. On Mayotte, "women gain autonomy though their status as village
owners, their rights to productive land, ownership of houses and household goods, and
equal control with the husbands over subsistence crops. ... A woman should remain a
virgin until her first marriage, and indeed it is in her best interests to do so, but from
this point she has as much control over her sexuality as a man has over his. The
wedding is a celebration of the bride's emergence into womanhood as much as it is
anything else, and ease of divorce means that women are not pawns in an exchange
controlled by men but can act as independent agents on their own behalf. . . . By
means of judicious marriages and child fosterage over time a woman can develop a
coterie of followers. ... In addition, women are politically organized among them-
selves, have certain rights and responsibilities in village affairs, and select their own
leaders" (Lambek 1981, 23-24). Because a bride typically receives a gift of a house from
her parents, initial residence of newly married couples is likely to be in the wife's
village. On Mayotte, barren women and women in polygynous marriages (i.e., de-
prived women) are not possessed by spirits more than other women.
 10
Gender Ideology

Religions in which men are structurally dominant have developed elaborate

ideologies that endeavor to explain, justify, cause, or apologize for women's
subordinate status. Examples of such ideologies include claims that women are
more prone to sin—especially sexual sin, that God chose to become incarnate
as a male, that woman (Eve) brought sin into the world, and that women's
souls or intelligences are inferior to men's. These ideologies both reflect and
give form to patriarchal social structures.
Can we expect to find that women's religions offer ideologies that explain
why men are inferior? Or should we expect that women's religions do not
challenge broad cultural conceptions of male superiority? Late twentieth-
century American experience has shown that the fact that women create or
control an organization does not necessarily mean the organization will advo-
cate female dominance or even gender equality. Women of the New Right, for
example, have formed organizations with explicitly anti-feminist agendas.'
During the past two decades, as feminist anthropologists have searched the
ethnographic record for universals concerning gender role and ideology, the
most ubiquitous finding is that no matter what the actual content of the traits
that a society attributes to men and women, the level of affect concerning
those traits is singularly high. For example, whether a particular culture re-
gards men as strong and women as weak, or men as weak and women as
strong, one thing remains the same—members of that society are ardent,
earnest, and zealous regarding that belief. Thai scholar Nancy Eberhardt has
pointed out that "what makes gender beliefs, images, and expectations so
compelling to the people who live with them is their privileged position as part
of the society's moral order; violations of the gender system are experienced
not as harmless eccentricities but as disturbing transgressions, an invitation to
chaos and evil" (1988, 6).
Given the high emotional charge elicited by ideas about gender, we can
expect to find that members of women's religions care very much indeed about
issues of gender, that gender ideologies of women's religions are not isolated
from the gender beliefs of the society as a whole, and that their gender ideolo-
196 Priestess, Mother, Sacred Sister

gies are frequently the object of attack by groups and individuals outside the
female-dominated religion.
Female-dominated religions often occur in cultural settings in which women
as a group are believed to be more religious than men. Brazilian women
predominate numerically not only in the Afro-Brazilian cults, but also in Pente-
costal and Catholic churches. In Brazil all religion is seen as a female activity—
most worshippers at Catholic churches are women, priests are treated with
some suspicion (why would a man opt for celibacy?), and the Church has
difficulties recruiting enough men to be priests. In nineteenth-century North
America religion became notoriously "feminized," and in twentieth-century
North America women continue to attend church more often and profess
greater religious faith than do men. Among the Black Caribs women predomi-
nate not only in ancestral rites, but in Catholic religious associations as well.
According to Coelho, 80% of the members of these associations are women
(1955; 159)- in Southeast Asia women excel at Buddhist merit-making acts.
On the Ryukyu Islands women and the sacred are so thoroughly associated
that certain Buddhist sects that normally have only male clergy have female
clergy on Okinawa. 2 And even in the regions in which the zar cult is located
women are more involved than men in all sorts of non- and quasi-Islamic
religious activities. In all these settings, women's religions fit into rather than
challenge conventional notions of sex roles. Women's religions are not making
any extraordinary claims in stressing women's affinity for matters of the spirit.
In Chapter 3 I proffered several hypotheses regarding gender ideology. In
light of Sara Ruddick's contention that maternal thinking tends to be "conven-
tional," I suggested that female-dominated religions would probably not en-
courage women to raise children who will not conform to culturally accepted
gender roles; that is, women's religions would not really challenge normative
gender ideologies. This hypothesis indeed proves to be correct. Sande initia-
tion for example, inculcates in women a strong sense of female strength and
solidarity, but its main goal is to prepare women for heterosexual marriage.
According to Nancy Chodorow, boys who are raised by women eventually
repudiate femaleness to develop their male gender identity. Chodorow's
theory explains why ideologies of female inferiority, impurity, and subordina-
tion occur in so many cultures throughout the world. Struggling with forming
a sense of self in the absence of a realistic role model (because fathers are absent
so much of the day), boys come to define being male as not-being-female. This
leads boys to wish to repress whatever in themselves is female, to polarize
masculinity and femininity, and to accord higher status to whatever they label
as male. What does this process have to do with women's religions? On the
one hand, in female-dominated religions gender ideology is created and elabo-
rated by women. Therefore, female-inferior ideology, which according to
Chodorow is created by men, should not occur. Similarly, female-superior
ideology should not occur because growing girls do not need to reject maleness
in order to develop a positive gender identity and sense of self. In women's
religions, ideologies of gender dominance and subordination should not pros-
per. As we will see below, this hypothesis proves to be correct.
 Gender Ideology 197

I also suggested that because mothers bear and raise children of both sexes
from infancy, women are likely to know that boys and girls are not really very
different (cf. Johnson 1988, 86-87). Mothers, more than fathers, will realize that
babies of both sexes kick in the womb, emerge through the cervix, breastfeed,
urinate, defecate, smile, sneeze, clap hands, and crawl. Therefore, women's
religions should tend to play down gender as a cultural category. This hypothe-
sis proves to be utterly incorrect. In fact, all the female-dominated religions
teach that men and women are essentially different. The Shakers, for example,
went so far as to say that Ann Lee's femaleness was an inherent and necessary
precondition for her role in redemption. In some women's religions the differ-
ence between men and women is seen as complementary and in others this
difference leads to conflict (see below, "Ritual Tugs of War").
Female-dominated religions do not arise in cultural vacuums, and we
might expect that ideologies of male superiority either infiltrate into women's
religions or that the women's religions invest a great deal of energy in chal-
lenging ideologies of male superiority. One of the most consistent patterns
regarding gender beliefs in female-dominated religions is apparent acceptance
of widespread sexist ideas concerning women's nature and role. These ideas
are, however, reinterpreted as evidence of women's greater interest in, or
talent for, religious activity. As we will see in the following section, for
example, nineteenth-century Spiritualists embraced contemporary stereotypes
of women as passive and weak, yet claimed that passivity and weakness are
ideal traits for powerful mediums!

Spiritualism: A Case Study

Nineteenth-century North American popular, religious, and scientific ideol-
ogy identified men with reason, strength, and the public sphere. Women were
seen as weak, impressionable, and domestically oriented. "The requirements
of successful mediumship are more typical of what is or was taken to be a
woman's character than of a man's—passivity and submissiveness being
among the most important" (Skultans 1983, 16). And, "Mediums were weak
in what were considered to be the masculine qualities of will and reason and
strong in what were considered to be the female qualities of intuition and
nervousness. They were impressionable (i.e., responsive to outside influence)
and extremely sensitive. Above all, they were passive. After all, it was queried,
what spirit could manifest anything through a medium whose own personality
was strongly assertive?" (Moore 1977, 196).
The naive female mediums stand in contrast to the highly educated, wealthy
(male) men of science—the researchers of psychic phenomena—with whom
they sometimes teamed up in order to "prove" the existence of spirits. "It has
been the habit to say that great intellect stands in the way of personal psychic
experiences. The clean slate is certainly most apt for the writing of a message"
(Doyle 1926, 2—quoted in Skultans 1983, 16).
According to Ann Braude, "Spiritualism made the delicate constitution and
nervous excitability commonly attributed to femininity a qualification for reli-
198 Priestess, Mother, Sacred Sister

gious leadership. If women had special spiritual sensitivities, then it followed

that they could sense spirits, which is precisely what mediums did" (Braude
1985, 422). In different words, "Victorian stereotypes of femininity . . . bear a
remarkable resemblance to the conception of the ideal medium. The following
adjectives can equally well describe the ideal woman as the ideal medium:
unsophisticated, innocent, passive, young, tender, feeling, intuitive" (Skultans
1983, 23).
In the context of Spiritualism this was a double-edged sword—the innate
feminine qualities that made women good mediums (and so offered them
power, mobility, and some economic independence) also bound them to the
prevalent sexist ideology. The good medium, like the good woman, had to
renounce her self. Thus, for Spiritualists, passivity and power converged.
"And here lay the crux of the dilemma. For the very quality which supposedly
made women such excellent mediums was equally construed as undermining
their ability to function in the outside world. Female passivity, the leit-motif of
powerful mediumship, also positioned women as individuals without social
power" (Owen 1981, 10).
Female mediums, for the most part, accepted the Victorian stereotypes of
women as chronically ill and long suffering. Mediums were proud of their
willingness to sacrifice their own well-being for the spiritual benefit of others.
"Spiritualism allowed women to discard Victorian limitations on women's
role without questioning Victorian ideas about woman's nature" (Braude
1985, 422, my emphases). According to Owen, on the other hand, "What the
seance promised was the ritualized violation of cultural norms" (1981, 203).
Women's illness and passivity, in Spiritualist seances, came to mean power, not
powerlessness. Significantly, Moore shows that women mediums took obvi-
ous joy in conquering male adversaries and scoffers (1977, 113).
Although Spiritualists seem to have accepted prevalent evaluations of
women, women's rights has been part of the Spiritualist platform and Spiritual-
ists were among the most ardent advocates of women's suffrage. Spiritualists
believed that the advent of spirit communication heralded a new era and a new
social order that included women's rights, and a number of leading feminists,
including Susan B. Anthony, were interested in Spiritualism. "Spiritualism
became a major—if not the major—vehicle for the spread of woman's rights
ideas in mid-century America" (Braude 1989, 57).

NATURE AND CULTURE

Feminist anthropologist Sherry Ortner (1974) has argued that culture is univer-
sally valued over nature because culture is what subdues or conquers nature.
Women, because of childbirth and lactation, are regarded as closer to nature.
Therefore, cross-culturally men are valued over women. The root of male
dominance and female subordination lies in the universal tendency to associate
women with nature and men with what dominates nature—that is, culture. 3
Male-dominated religions throughout the world utilize this chain of dichoto-
mies to justify gender inequality: Woman is in essence different from man, and
 Gender Ideology 199

this difference is analogous to the difference between nature and culture, be-
tween self-control and chaos, between sinners and saints. Ortner's typology
rings true to those of us familiar with Christian teachings that equate women
with flesh and matter and the masculine principle with logos, the word (cf.
Holden 1983, 6).
This scheme does not hold true for the gender ideologies of women's
religions. In the religious ideology of the Ryukyu Islands nature and culture
are integrated: Natural sites and materials and culturally constructed sites and
materials are equally sacralized in ritual. Afro-Brazilian religions see men (and
not women) as so rooted in the natural world (of lust, alcohol, etc.) that they
cannot open up to spirits. Korean men's ancestor religion is limited to strictly
biological ties of kinship (nature), whereas women's household religion in-
volves both broader social definitions of kinship, and worship of gods associ-
ated with the house (culture).
The nineteenth-century examples are of particular importance because in the
American cultural climate of the time "the female domain was perceived as
natural" (Kitch 41). Christian Science, on the other hand, does not posit that
either men or women are closer to nature or culture; in fact, nature is illusory
whether one is male or female. If anything, for Mary Baker Eddy the female
principle is equivalent to spiritual perfection while the male principle is associ-
ated with materialism. Another nineteenth-century women's religion, Theoso-
phy (see Appendix B) identified male with matter (in Ortner's terminology—
nature) and female with spirit, and some Theosophists asserted not only the
superiority of the feminine principle (spirit over matter) but of the female sex. To
take another example, according to Sally Kitch, "In contrast to mainstream
beliefs in an opposition between nature and culture, Shaker writing in the late
nineteenth century considered both realms to be human creations that stood in
opposition to true or divine nature. They classified reproductive humanity in the
realm of human culture that is necessarily opposed to divine nature" (1989, 50).
In several female-dominated religions, Ortner's women and nature versus
men and culture dichotomy holds true, but with a twist: Women and nature are
seen as superior to men and culture! Feminist Spiritualists believe that men and
culture have violently destroyed our planet, and ecological activism is crucial to
the Spiritual Feminist platform. In the Spiritual Feminist vision, women who
identify with nature—with the web of being—can save our planet from destruc-
tion at the hands of male culture. American historian Catherine Albanese (1990)
explicitly relates contemporary Feminist Spirituality to nineteenth-century "na-
ture religions" such as Christian Science. I see contemporary Spiritual Femi-
nism's affinity for American Indian rituals and lore as an extension of
nineteenth-century Spiritualists' partiality for messages from American Indians.
In both instances, American Indians are seen as embodying the "natural" (see
Chapter 8).+ The Spiritualist attitude towards nature is eloquently expressed in
the following graveside funeral service: "To Nature, the source of all, we now
surrender him (or her) who has passed on before us. May all the sweet and
thrilling influences of fragrant fields, of flowering plants, of bursting buds and
blossoming vines, of silvery streams and genial showers, of setting suns, of
2oo Priestess, Mother, Sacred Sister
jeweled nights and dawning days, melodious with the songs of birds and with all
the wondrous harmonies of Nature, be with our dead" (from the Twentieth
Century Formulary of Songs and Forms by W. C. Bowman, 1907, pp. 153-154,
cited in Ward 1990). I barely need to point out how this conception of death
differs from the highly transcendent vision found in Judaism, Christianity, Is-
lam, or Buddhism: rather than leaving the transient and inferior natural world
for a better and higher place, the Spiritualist dead is immersed in splendid and
vigorous Nature.
The Tres Personas Solo Dios religious community in the Philippines is a
fascinating example of an indigenous, quasi-Christian women's religion. 5 Ac-
cording to Rosemary Ruether, "Some time in the 1960's this community de-
cided to have only women as priests, based on their view that males were not
able to remain celibate. The founder and his son both married, but the commu-
nity believes that celibacy is necessary for priesthood. . . . The women priests,
who are called padre, celebrate a mass on the holy days" (n.d., 2-3). This highly
politicized group is dedicated to "preserving the natural environment against
exploitative development" (Ruethern.d., i), and Filipino women as well as men
are seen as associated with nature, in contrast to foreign colonial powers who
disregard and exploit nature. Not wishing to read too much into Rosemary
Ruether's brief but engrossing account of this community, what does seem to be
the case is that men are associated both with culture and with the "darker" side of
nature—the inability to control one's body in order to develop spiritually, while
women are associated with a more benevolent natural mode.
In a similar vein, Northern Thai spirit cults sacralize women's ties to the
land. Writing about Southeast Asia, Penny Van Esterik points to "a dualism
linking women with fertility, nurturance, and attachment, and men with su-
pramundane power and detachment" (1982, 5). Men do not own land, they
leave their natal families, and they are involved with Buddhism, which con-
demns "earthly" attachments. Consistent with the most prominent pattern of
gender ideology in female-dominated religions, Northern Thai matrilineal
spirit cults accept the prevalent assessment of women as "worldly" while
interpreting attachment to this world (land and family) as good and sacred.6
"Where religious visions of reality allocate at least some sanctity and dura-
bility to the natural world, what is striking is the plethora of females and
female elements in their sacrifice myths. Where religious visions deny ulti-
mate sanctity and durability to the natural world, as in Christianity, Judaism,
and Islam, women fall out of the communal quests for ultimate connection,
the mythic searches for communal gods that establish not only who is capable
of reaching for the heavens but who is capable of ruling on earth" (Combs-
Schilling 1989, 262).

The Body
Women's religions relate to bodies in distinctive ways. Unlike in many (if not
most or even all) male-dominated religions, women's bodies are not seen as
particularly polluted or polluting. Spiritual Feminists, for example, honor
 Gender Ideology 201
women's bodies as especially reflective of the essentially good cycles of nature.
I see it as interesting that in the two somewhat ascetic religions—Christian
Science and Shakers—women's bodies are not more "worldly" or more "sex-
ual" than men's.
In many male-dominated religions, ideas concerning menstrual and child-
birth pollution are used to explain women's inferiority and/or to exclude
women from religious rituals. Can we expect to find that menstrual and child-
birth taboos do not exist in women's religions? The answer is far from simple.
The Foundress of the new Japanese religion Tenryko called menstruation
"flowering" rather than "pollution" (as it is called in many other Japanese
religions; see Tenryko: Its History and Teachings).'' To take another example,
among the African Mende (a society in which Sande is present) a menstruating
woman is described as "busy" and as "having seen the moon." According to
Harris and Sawyerr, this terminology is significant: "The wife is virtually
committed to another claim, supernatural indeed, with whom the husband
must not compete" (1968, 96).
Among the Tetum, at birth the umbilical cord and placenta are treated as
sacred; they are tied to the house ritual pillar and later ceremoniously trans-
ferred to the house of the father's clan (Hicks 1976). This sort of treatment of
the effluvia of birth stands in sharp contrast to the many cultures, such as
Bangladesh (see McConville 1988), where the afterbirth and bloodied clothing
are buried or burned because of the extreme pollution they are believed to
carry. Tetum women are so thoroughly associated with the sacred and men
with the secular that all spirits are treated as feminine, even male ancestral
ghosts. Women priestesses preside at almost all rituals (including wedding
rituals and agricultural rites), most shamans are women, and sacred objects are
kept in the "female" room of the house.8
Thailand, on the other hand, is characterized by extensive beliefs regarding
menstrual pollution. According to Wijeyewardene, menstruation is used as a
rationale for excluding women from many places and activities. The female
mediums of Thai urban spirit cults turn themselves into men in possession
(they are possessed by male spirits), and in their ordinary lives take on some of
the characteristics of men—specifically the susceptibility to pollution through
contact with menstrual blood (1986, 196). Similarly, zdr spirits do not possess
menstruating women, and a menstruating woman ties a knot in her braids to
warn the zayran not to enter her (Boddy 1989, 275).
During their menstrual periods Black Carib women must not eat any of the
products of the hunt, nor handle the hunters' weapons or dogs, lest they spoil
the luck of the hunters (Taylor 1951, 94). On the other hand, women may use
their own menstrual blood to fabricate charms or potions that give them
power over others. "One purportedly successful way of establishing complete
domination over a man is for a woman to prepare a bundle of herbs . . . and
place it inside her vagina; then for a number of nights she will make her mate
smell it" (Coelho 1955, 169—170). While men also make charms, their charms
are deemed as less powerful than women's charms (Taylor 1951, 134). Beliefs
in menstrual "pollution" in several female-dominated religions seem actually
202 Priestess, Mother, Sacred Sister
to be beliefs in menstrual "power." One wonders whether in other cultural
situations similar beliefs have been misinterpreted by anthropologists who
interviewed primarily men (cf. Buckley & Gottlieb 1983).
On the Ryukyu Islands ethnographers have found evidence of the existence
of beliefs in menstrual and childbirth pollution (Lebra 1966). We have reason to
believe, however, that these reports misinterpret Ryukyuan attitudes. Accord-
ing to Teigo Yoshida (1989), in traditional Okinawan culture menstruation was
seen as an expression of divinity. It is only under Buddhist influence that
pollution beliefs developed. Even today, there does not seem to be as strong an
ideology of female pollution in the Ryukyus as in Japan. Yuta, priestesses, and
high priestesses (noro) may perform ceremonies while menstruating. 9
In Ryukyuan tradition male and female are paired and equal, but deemed
linked to separate spheres: men to the political and women to the spiritual.
This binary division is reflected in almost every aspect of Ryukyuan culture,
even, as Toichi Mabuchi (1968) has shown, in the architecture of the houses. In
a mirror image of beliefs in female pollution, on Hateruma Island men are not
allowed to enter shrines, and during days in which priestesses stay in retreat in
the shrine (usually before major ceremonies) the priestesses even avoid sewing
or mending men's and boy's garments (weaving and sewing are major occupa-
tions for the priestesses during retreats). Similarly, rain chants are such impor-
tant ritual events that they must not be witnessed by men (Ouwehand 1985,
258). The evidence suggests that on the Ryukyu Islands maleness is perceived
as more polluting than femaleness.

Men and Women and Life and Death

Another manifestation of the identification of women with nature is the
association between women and death in Western religious and psychoana-
lytic traditions. This is a multifaceted association, including, for example, the
idea that woman (Eve) brought death into the world, and that the womb is
analogous to the tomb. Diane Jonte-Pace demonstrates that both religious and
psychoanalytic theories have constructed chains of associations in which fe-
male becomes a metonym for absence and death, and male becomes a
metonym for presence and life. (I will not repeat her evidence here.) Accord-
ing to Jonte-Pace, these associations contribute "to a deeply embedded per-
sonal and cultural misogyny" (1990, 18).
I find Jonte-Pace's exposure of the link between the metonymic association
of women and death on the one hand and misogyny on the other to be highly
convincing, and wish to ask whether in female-dominated religions this chain
breaks down. Are women associated with death in female-dominated reli-
gions? As I explained in Chapter 6, funeral rituals receive a great deal of
attention in several women's religions. Yet before we treat this as corrobora-
tion of the association of women with death, we must distinguish between the
symbolic view of women as death (as is the case in western religious traditions
and psychoanalysis) and the involvement of real women in ancestor rituals (as
is the case in several female-dominated religions). I contend that the ancestor
 Gender Ideology 203

rituals of, say, the Black Caribs, have far more to do with life than with death.
One could even say that ancestor rituals serve to keep the dead alive; they
certainly serve to enlist the assistance of the dead in keeping the living alive.
In female-dominated religions food is associated with life, fertility, abun-
dance, and women. Let us recall that the elaborate food rituals of women's
religions include almost no animal sacrifice (i.e., death symbolism; see Chapter
6). This point comes across dramatically in Shigeharu Tanabe's (1991) descrip-
tion of an elaborate ritual of a large Northern Thai matrilineal descent group.
The space inside the sanctuary where the ritual takes place is divided into
female and male sides. The polluted (in Thai terms) animal sacrifice takes place
on the male side, while the female side remains pure.
On the Ryukyu Islands feminine is associated with fertility, and masculine
with hunger and scarcity. Priestesses are prohibited from attending funerals
or coming into contact with the dead.10 Buddhism on Okinawa is basically a
cult of the dead, and men are much more involved in Buddhist funeral rituals
than they are in other sorts of rituals, and even predominate at certain death
rites. Women (daughters and wives), however, wash the bones of the corpse
after the seventh anniversary of the death. "People say that it is always
women who act intermediary roles between this world and other worlds"
(Yoshida, personal communication 1992). In this case, the distinction seems
to be between contact with death (which men do) and contact with the
supernatural (which women do).
An even more striking instance is provided by the Tetum. As I said earlier,
among the Tetum women preside at almost all religious rituals. The one signifi-
cant exception is death rituals at which men are more active than women. The
most dramatic figure in death rituals is the "lord of death" who is a senior male
of the dead person's hamlet. This is Hicks' description of the role of the "lord
of death": "[A] tremendous cry erupts from somewhere in the depths of the
death house, and, brandishing his own sword high in the air, the lord of death,
galvanized into action, leaps onto the frontal veranda, in a ritualized attempt to
keep the corpse among its kin" (Hicks 1984, 120-121).
To take a few more quick examples, Janice Boddy has observed that in the
Northern Sudan (home to the zdr) women preside over births but are not
allowed to attend burials (1989, 70). And according to Gehan Wijeyewardene
(1986), urban Thai mediums (almost all of whom are women) are also subject to
taboos regarding death—many will not even attend funerals; male monks, on
the other hand, do attend funerals."
The sorts of binary images found in women's religions are not what Sherry
Ortner's theory (or Jonte-Pace's analysis) would lead us to expect. At the risk
of both reductionism and overgeneralization, I suggest that binary equations in
female-dominated religions include the following pairs, some of which associ-
ate men with nature and some of which associate women with nature.

female (is to) male (as)

life (is to) death (and)
fertility (is to) hunger (and)
2O4 Priestess, Mother, Sacred Sister
harmony (is to) war (and)
spiritual (is to) earthly (and)
sacred (is to) profane (and)
intuitive (is to) rational (and)
ritual expertise (is to) administrative expertise

RITUAL TUGS OF WAR

Although my intellectual instinct is to reject, as overly universalistic and sim-
plistic, the sort of binary scheme I have laid out above, the available informa-
tion on the cultural contexts of women's religions does seem to accentuate
some form of gender opposition.12 Northern Sudanese women, for example,
demand that their value be socially recognized not through trying to be more
like men, but by emphasizing their difference.
In several women's religions gender opposition is dramatized in ritual tugs
of war. Throughout the Ryukyu Islands there are annual ritual tugs of war
between two groups representing the male and female principles. The female
side should win in order to ensure fertility. Here is a description of one such tug
of war:

The puru festival continues for three days in Hatoma Island. A tug of war is
played on the third day between the east and west teams. . . . From the east
teams there steps out a man dressed like a warrior with a long sword and rice
whisky, and from the west team a woman dressed like a peasant woman with a
basket containing the heads of both rice and millet in her left hand, and with a
sickle in her right hand. . . . After this the tug-of-war starts, and every year the
west side wins, as it is believed that when the west wins the coming year is a
bumper year (Ito 1966, 48).

A similar ritual (evidence comes from Sosu Village on Okinawa) consists of

a sumo wrestling match in which a masked adult male and an old woman fight
each other. It is believed that a rich harvest can be expected when the old
woman wins, and so she always wins (Ito 1966, 45).
Among the Northern Thai there is an annual ritual tug of war in which men
and women compete. As in other ritual tugs of war in female-dominated
religions, the women—together with the spirits—always win (McMorran
1984). Among the Northern Thai, in situations in which women hold ritual
authority men are likely to display aggression toward the women and their
spirits, and incidents of men damaging shrines have been reported by a number
of researchers.13 Similarly, women possessed by the spirits during cult rituals
sometimes display aggressive behavior toward men.
Many West African societies are fairly egalitarian in terms of economic
rights, yet men and women are not identical in terms of roles. "In virtually
every aspect of Kpelle [a society in which Sande is active] life, ranging from
secret society activities to the most mundane interactions, sexual differences
are expressed" (Bellman 1979, 40). The rather fierce structural opposition of
 Gender Ideology 205

men and women in Mende society (another one of the societies in which Sande
is located) is made manifest in a ritual tug of war between men and women that
the women win by pulling the men out of the women's ritual site. According
to M. C. Jedrej (1976), the separation of women from men is like the separa-
tion of sacred from profane. The sacred is associated with the forest (where
girls' initiation takes place), fertility, potency, danger, and the supernatural—
and so men must be excluded.
Recalling Nancy Chodorow's contention that exclusive female mothering
leads men to develop ideologies of gender antagonism, how can we explain the
friction or even enmity embodied in these ritual tugs of war in women's
religions? I believe that Chodorow herself gives us the critical clue. Chodorow
has observed that especially sexist ideologies tend to arise in societies in which
the father is absent, in other words, in societies in which the most salient
emotional bonds are between mothers and children rather than between moth-
ers and fathers. In such societies boys spend their earlier years exclusively or
predominantly with women. In societies characterized by "father-absence" or
"low father-salience," boys and men engage in what Chodorow calls "compul-
sive assertions of masculinity" (1971, 280). At this point, it is essential to bear
in mind our discovery in Chapter 2 that women's religions tend to occur in
matrifocal societies—that is, in societies with very "low father-salience."
Women's religions—with ideologies and structural manifestations of fe-
male strength, tend to co-exist with ideologies and structural manifestations
of extreme "violent behavior, male narcissism, pride, and phobia toward
mature women" (1971, 280). This is amplified by the fact that girls in ma-
trifocal societies also grow up with "low father-salience," perhaps leading
them to feel that males are drastically "other." What we seem to be describing
are cultural situations characterized by deeply rooted gender conflict or, return-
ing to the terminology we used in Chapter 2, gender dissonance. Periodic
ritual tugs of war are certainly one way of expressing (and temporarily resolv-
ing) gender conflict.14
In the following sections we look more closely at the ideological strategies of
several women's religions. By far the most common strategy is to reinterpret
rather than challenge the prevailing view of women. This is not to say that
women's religions do not seek to correct structural manifestations of gender
inequality. As we will see in Chapter 13, female-dominated religions defend
and further women's individual or collective rights. What I am saying is that
female-dominated religions tend to accept (exploit?) non-egalitarian views of
women.

EXAMPLES
Pattern 1: Condemnation of Patriarchy—Feminist
Spirituality
Spiritual Feminists have elaborated a sacred history in which gender conflict is
the moving force. In brief, the sacred history of the Feminist Spirituality
206 Priestess, Mother, Sacred Sister
Movement runs as follows: A long time ago, all over the world, people lived in
harmony with nature, with their fellow human beings, and with the cosmos.
They expressed their image of the beneficence, the generosity, the mystical and
magical fertility of the universe through the natural symbol of the goddess-
woman who brings forth and nourishes life. During this era, gender hierarchy
either did not exist at all, or human orientation was matriarchal. Proof for this
reconstruction can be found in archeology, ancient myths, and contemporary
ethnography (see Spretnak 1982; Stone 1976; Gimbutas 1974).
At some point, for some reason (population pressures, technological and
economic changes, invasions by small groups of militant men, and/or discov-
ery of male role in reproduction) this idyllic world was turned topsy-turvy.
"Earth is the bountiful female, the ever-giving Mother, Who sends forth
food on Her surface in cyclical rhythms and receives our dead back into Her
womb. Rituals in Her honor took place in womb-like caves, often with vulva-
like entrances and long, slippery corridors. The elemental power of the female
was the cultural focus as far back as we can trace. At the moment this awe
turned to envy, resentment, and fear, patriarchy was born. Why or how we do
not know. . . . The objective of patriarchy was and is to prevent women from
achieving, or even supposing, our potential. . . . They [patriarchy] almost
succeeded" (Spretnak 1982, xii).
The apex of patriarchal oppression of women was reached during the Euro-
pean witch hunts of the late middle ages, in which it is estimated that millions
of women were killed as witches. Women's resistance throughout the ages,
culminating in the twentieth-century feminist movement, is what has pre-
vented patriarchy from successfully erasing all vestige and memory of non-
patriarchal culture.
Whether or not this historical scenario is absolutely "true" is not an issue for
Spiritual Feminists. A sentiment expressed repeatedly in writings of the Move-
ment is that "history" as an objective "something out there" does not exist;
rather, history exists in the eyes, the organizing strategies, and the interpreta-
tions of particular individuals. Given the necessarily subjective nature of his-
tory, it is legitimate and even imperative for feminists to select and create
herstories that empower them as women. "There was a time when you were
not a slave, remember that. You walked alone, full of laughter, you bathed
bare-bellied. You say you have lost all recollection of it, remember . . . you
say there are no words to describe it, you say it does not exist. But remember.
Make an effort to remember. Or, failing that, invent" (Witting 1971, 47).
Spiritual Feminists are far from unanimous in their opinions regarding gen-
der differences. Some Spiritual Feminists stress the basic goodness, the
Goddess-nature, of all living creatures—male and female; and some Feminist
Spirituality groups include both male and female members. Others believe
men have so distorted both their own beings and the purity of nature that
women must separate themselves from men to find their own female sources
and styles of power.
Although the gender ideology of the Feminist Spirituality Movement is
more explicit than the gender ideologies of other women's religions, two
 Gender Ideology 207

similarities emerge. First, like most women's religions, Feminist Spirituality

acknowledges the tension between the sexes. And second, Feminist Spiritual-
ity, in a roundabout sort of way, accepts and reinterprets the wider societal
views of women's nature. In the writings of Spiritual Feminists, women are
typically portrayed as less aggressive, more emotional, and more intuitive than
men. This description of women does not differ from the highly stereotyped
images of women prevalent in contemporary American culture. Where the
difference does lie is in how these attributes are valued. Unlike in mainstream
America, these attributes are seen as honorable and essential for the preserva-
tion of life on Earth.

Pattern 2: Gender Complementarity

Notions of gender complementarity (not gender sameness) occur in several
women's religions. Ironically, the first example I bring is one that has already
appeared in the section on ritual tugs of war. Carol MacCormack offers a very
different analysis of West African gender relations than that developed by
M. C. Jedrej (see above, page 205). "The Sherbro [one of the societies in which
Sande is active] folk model presents the analogy that nature is to culture as
children are to initiated and married adults. It stresses gender interdependency
rather than male domination" (1980:95). According to MacCormack, secret
societies function to transform (natural) children into initiated adults who
understand and vow to live by ancestral laws (culture). Sande both reflects and
reinforces the cultural value of gender interdependence. A men's secret society,
Poro, is complementary to Sande. Officials of the two societies are responsible
for different matters of importance to the community at large. Poro officials
see that wells are kept clean and that disputes do not turn into full-scale fights;
Sande officials treat illnesses and enforce sexual prohibitions. Men stay out of
the way while Sande performs certain rituals; women stay out of the way while
Poro performs other rituals. "Poro and Sande vertically polarize all of Mende
society; they balance Mende men and Mende women within a complex equilib-
rium" (Cosentino 1982, 22).
The evidence does not allow us to determine whether Jedrej's or Mac-
Cormack's interpretation is "more" correct. Instead, I would suggest that in
cultural situations characterized by highly dichotomized views of gender roles,
male and female simultaneously conflict and complement. If men's and
women's spheres are radically different, both are necessary, yet each tends to
oppose the other. Sometimes the complementary mode will be more promi-
nent, and sometimes the conflict mode will take over. Perhaps the most impor-
tant lesson to be learned from Jedrej's and MacCormack's analyses is that the
identical gender ideology may be perceived as complementary or conflicting
depending on the point of view of the observer.
Our next example is more clear-cut. From its inception, Theosophy sought
to form a nucleus of spiritually enlightened individuals without distinction of
race, creed, sex, caste, or color. Women were well represented both as leaders
and as members. Theosophists believed that the body is the temporary vehicle
208 Priestess, Mother, Sacred Sister

for an eternal spirit, which in its evolutionary progress passes through all
material forms from the mineral to the angelic, and reincarnates innumerable
times in both male and female guises. The dynamic of the cosmic process
requires the interaction of yin and yang, male and female. "Thus Theosophy
provides a theoretical legitimation at the highest cosmological level for mun-
dane notions of equality between the sexes" (Burfield 1983, 35-36).
This sort of gender ideology seems to have been especially appealing to
nineteenth-century Americans. Spiritualists also believe that male and female
are polar opposites that are united in the Divine. And Susan Setta argues that
one of the reasons for Christian Science's success lies in the fact that Mary
Baker Eddy "replaced the discriminatory separation of male and female with
the liberating union of masculine and feminine" (1977, 295).''
On the Ryukyu Islands gender complementarity is expressed through ideas
regarding the interdependence of brothers and sisters. People believe that the
spirit of the sister, either dead or living, protects her brother from malevolent
influences. According to myth, the creator couple was a sister and brother.
Brothers and sisters have a special relationship; he is expected to safeguard her
well-being in the profane world (even after she has married), and she is ex-
pected to safeguard his well-being on the spiritual plane. The eldest sister is
called by her brother "guardian spirit" or "sister-feawi." The role of the sister is
highlighted during times of danger. When the brother sets out on a journey, for
example, his sister gives him a piece of cloth that she has woven, or a lock of
her hair, in order to protect him. On the Yaeyama Islands (a subgroup of the
Ryukyus) the sister plays a critical role in family rituals, and is entitled to a
portion of the first fruits of the harvest, so that "her spirit will be pleased and
will favor a good crop the following year" (Mabuchi 1964, 82). Married-out
sisters return to their natal families annually to play the key role in family and
agrarian rituals. "People say that, owing to the beneficent spirit of the sister,
one remains healthy enough to work and has a good crop" (Mabuchi 1964, 82).
The spiritual relationship between brother and sister also has more prosaic
implications: parents instruct brothers to treat their sisters well, so that sister's
spirit will not become angry and cause him misfortune. Although women
marry outside their natal families and become somewhat incorporated into
their husbands' families, brothers—especially eldest brothers—are expected to
help their sisters in times of need. Married-out sisters often return to help their
brothers in the field at harvest time. People say that whereas wife and husband
divorce, brother and sister are eternally connected. A happy marriage may be
described as "They are happy as brother and sister" (Haring 1964, 45).
In a creation myth similar to the Ryukyuan one, Northern Thais tell that the
first two human beings were an elder brother and younger sister who married.
It is believed auspicious for twins who were separated at birth to marry, and
brother-sister incest does not incur serious ritual sanctions. Northern Thai
parents, concerned about their daughters' futures, make them take care of
younger brothers so that these brothers when grown up will return the favor
and help their sisters. "Thai family arrangements hinge a good deal upon
warmth between brothers and between sisters" (Hanks and Hanks 1963, 434).
 Gender Ideology 209
Myths of the first couple being brother and sister, and expectations of coopera-
tion among brothers and sisters, seem to be common throughout continental
and island Southeast Asia. These myths are often associated with social struc-
tures in which male and female are complementary rather than hierarchical.
The Shakers developed a gender ideology that was far more egalitarian
than other gender ideologies pervasive in nineteenth-century American soci-
ety. Shakers seized on Darwin's hypothesis of an androgenous ancestor, and
accentuated the contention of some physiologists that each sex includes, in
undeveloped form, the organs of the other sex. Shakers believed that men's
exploitation of women for thousands of years had been one of the main
causes of failure in human history, and understood that men's exploitation of
women had been supported by the image of a male god.16 They believed it
necessary to remake society on the basis of a new type of human relationship:
since economic evils are based on biological evils, a new kind of biological
relationship would make it possible to improve society and its economic
structure. The leadership structure of the Shakers is self-consciously egalitar-
ian, with a man and a woman at each level of the hierarchy.17
In light of what we have just seen about the Shakers, it is somewhat surpris-
ing to find that they believed men and women to possess different natures—
women more spiritual, nurturing, emotional, and affectionate, and men more
worldly, physical, rational, and intellectual (Bednarowski 1980, 211; Brewer
1992, 621; Kern 1981). Shakers believed that all of nature is made up of male
and female forms, and that the distinction of sex is eternal and inherent in the
soul itself. Male and female are diametrically opposed forces that find perfec-
tion only in an all-encompassing unity, a male-female godhead.18 Also on the
human level, both male and female are necessary, an ideology that was re-
flected in, for example, the insistence that men and women live under the same
roof in every house in every Shaker community (Kitch 1989, 85).
The Shakers acknowledged many of the same constitutional differences be-
tween the sexes as did the society at large, yet insisted that gender differences
and sexual separation do not mean that women should have lesser rights or
privileges (Kitch 1989, I34). 19 The Shakers did not challenge contemporary
notions of female and male nature. Instead, they insisted that female nature and
female roles are as good as male nature and roles.20

THE IMPORTANCE OF GENDER

DIFFERENCES
As a feminist, disappointment was my first reaction on organizing the material
for this chapter. I wanted to find out that women's religions around the world
proclaim egalitarian gender ideologies. Instead, I found that women's religions
buy into prevalent unegalitarian notions of gender and gender roles. As an
anthropologist, however, I was delighted to discover a discernable pattern.
That pleasure turned into intellectual excitement when I read Nancy Eber-
hardt's overview of gender relations among hill tribes in Thailand.
Eberhardt shows that among a hill tribe known as the Akha, gender is a
210 Priestess, Mother, Sacred Sister

crucial cultural category—just about every object, event, activity, and symbol
is considered either male or female. Among the Shan and Yunnanese tribes, on
the other hand, gender is a rather secondary cultural category. Comparing the
status of women among these groups, Eberhardt found that "in the process of
subordinating gender as a cultural category, women are also devalued"
(1988, 8). In a similar vein, Tikva Frymer-Kensky has amassed evidence show-
ing that in Mesopotamian and Old Babylonian mythology, and in Hellenistic
mythology and philosophy, men and women are portrayed as essentially, onto-
logically different. In the Bible, on the other hand, men and women are essen-
tially similar, the only difference being that of genitalia (1992). Although this is
not the point that Frymer-Kensky was trying to make, her findings make it
clear that the "egalitarian" gender philosophy of the Bible co-existed with a
religious system in which women had a far smaller role than in the "non-
egalitarian" religious philosophies to which she compares the Bible.
The writings of most feminist anthropologists carry either an implicit or
explicit message that the blurring of gender categories is what will lead to the
demise of patriarchy (cf. Rosaldo 1974; Chodorow 1974). The findings I have
brought here suggest a different scenario. The very few women's religions of
which we have knowledge consistently stress rather than play down gender
differences. It seems to me that women's religions have in fact selected the
most efficacious strategy possible considering the limitations within which
they operate—none of these religions is located in a truly non-sexist cultural
context. In other words, if these religions were to claim that men and women
are the same, no one would believe them anyway! Gender differences seem to
be so obviously "true," that enhancing and reinterpreting those differences
offers women an ideology that is both believable and somewhat empowering
(see Chapter 13). In an imperfect world, it may well serve women's interests to
own a clearly bounded portion of the cultural map.
In almost all the cultural situations in which women's religions are found,
women can choose between the female-dominated religion and a male-
dominated religion (nat cult vs. Buddhism, Afro-Brazilian religion vs. Ca-
tholicism and Pentecostalism, Shakers vs. mainstream Protestantism). It is
important to note that in each of these cases the female-dominated religion
teaches a gender ideology that is less patriarchal than that of the parallel male-
dominated religion. In other words, when women are in a position in which
they can choose between a more and a less sexist gender ideology, many do
opt for the less sexist one. Female-dominated religions may teach that
women are passive or soft; they do not teach that women are evil or stupid.
Although most female-dominated religion, like most male-dominated reli-
gions, deem different spheres of activity to be appropriate for men and for
women, in female-dominated religions women's sphere is considered as
good (if not better) than the male sphere, and women fully control the
female sphere. By way of contrast, in modern Israel where a great deal of
women's religious activity involves cooking, recipes are "custom" (minhag)
rather than "law" (halacha). As a result, the status of cooking as a religious ritual
is lower than the status of men's prayer or Torah study. Moreover, the one
 Gender Ideology 211

aspect of cooking that falls into the realm of law is kashrut (food taboos and
regulations), yet for pious women the male rabbi is the ultimate authority
concerning kashrut; if an orthodox Jewish woman is not sure whether a chicken
is kosher, she is expected to consult a rabbi and accept his decision. My argu-
ment here is not that the existence of a "female sphere" characterizes female-
dominated religions; most known cultures have some sort of sexual division of
labor. What does characterize female-dominated religions is an ideological as-
sertion that the female sphere is as good or better than the male sphere, and the
institutionalization of autonomy for women operating in the female sphere.

Notes
1. New Right female-dominated organizations oppose the Equal Rights Amend-
ment, abortion rights and gay rights, and urge women to return to the home and family
as caretakers and wives (Klatch 1987, 45).
2. Haring (1964) reports that in a Buddhist temple in the city of Naze (Ryukyu) he
found a woman officiating as a full priest of the Shin sect. While this would not have
been permitted anywhere else, on the Ryukyu Islands, "There was no innovation about
a woman priest—this was something eminently proper and fitting that called for no
comment. Perhaps Buddhism failed to gain wide acceptance in Amami despite centuries
of propaganda because its priests were male" (1964, 121).
3. Many feminist scholars have offered critiques of Ortner's theory. The fact that
women are able to bear children does not mean that doing so is essential to women's
existential being or sense of self. If nothing else, birth control means that the ability to
give birth can be suppressed. And the contention that men are further from nature
seems ridiculous when we bear in mind that men, like women, eat, breathe, excrete,
sleep, and die.
4. The Shakers also had a penchant for native Indian spirits (Stein 1992, 176).
5. This is a syncretic religion including elements of both Catholicism and indige-
nous Filipino beliefs. A central theme in this religion is the liberation of the Philippines
from Western imperialistic rule. Rosemary Ruether believes that the prominence of
women in this and other similar communities is at least partially an outgrowth of the
indigenous Filipino tradition of female priests (n.d., 4).
6. On the Ryukyu Islands women are equated with the spiritual and men with the
material, whereas in East and Southeast Asia men are equated with the spiritual and
women with the material. I do not know how to interpret this difference, but it is
certainly worth noting. The answer may lie in the role of Buddhism and Confucianism
as "great traditions" in East and Southeast Asia, as opposed to their marginal role on the
Ryukyu Islands.
7. Tenryko claims to believe in equality of the sexes. While they say that there is no
real difference between the sexes, they also say that it is women's primary duty to give
birth and raise children because men are by nature incapable of this job. Women are
passive by nature and men are active. Few of the new Japanese religions preach egalitar-
ian ideologies. For example, the True-Light Supra-Religious Organization (Sukyo
Mahikari), founded by a man, teaches that the "spiritual pattern" of the male was
created 20,000 years before that of the female, and is naturally associated with the
general superiority of the yang principle. Winston David quotes one of their most
prominent teachers as proclaiming that "if women's liberation succeeds in Japan, the
country will be doomed" (1980, 199).
212 Priestess, Mother, Sacred Sister

8. One of the more unusual traits of Tetum religion is that although the woman is
the household's sacred head, the household itself is patrilineal and patrilocal. The wife
performs rituals for her husband's ancestors, not for her own. When she marries she
symbolically leaves behind her own ancestors (see Appendix B).
9. Anthropologist Mary Douglas has argued that menstrual taboos are more empha-
sized or more severe in cultures where there is an unclear message regarding gender—
where male dominance is not absolute. Yoshida claims (and I agree) that gender is an
extraordinarily unclear cultural category on Okinawa and so, according to Douglas,
should stress menstrual taboos. However, menstrual taboos are quite unelaborated on
Okinawa (1989).
10. Yoshida points out that in some cases a priestess may attend the funeral of her
mother or father, but in the role of relative and not in the role of priestess (personal
communication 1992).
11. In her study of a village in central Thailand, Jane Hanks found that men's and
women's magical practices had different objectives. Men wield magic to gain power
over spirits, people, elements, and forces. Their magic could be beneficial or malevo-
lent. "Men had a nurturing (Hang) role, but only for non-human beings, e.g., buffalo,
pet birds, and guardian spirits of house and land. Subverting this role for hostile ends,
they obtained possession of the ghost of a recent corpse by gathering the drop of oil at
the chin. . . . For men, nourishing something gave control" (1963, 79, my empha-
ses). Women's magic, on the other hand, was linked to the female task. It did not give
power over people and it could not be destructive. There were no female sorceresses in
the village, and women did not fear each other's magic. "A woman's magic was
devoted solely to promoting the processes of growth and life" (1963, 79).
12. Whether this reflects the intellectual biases of the scholars who document them
rather than of the religions is impossible to determine.
13. Richard Davis argues that the northern Thai are characterized by structural domi-
nance of women but an ideological dominance of men. Men are associated with auspi-
cious directions (north and right) and women with less auspicious. Davis builds a set of
oppositions: north, east, male, senior, right, settlements, high things versus south,
west, female, junior, left, forest, low things. "Agrarian rites are only performed by men
and the aim of the rites is to give male cultivators control over the female rice deity. The
phallicism evoked by erecting the 'great hawk's eye' during first-threshing suggests a
triumph of male efforts to domesticate and suppress the female forces of the wild"
(1974, 18).
14. Rituals that express gender conflict are also found outside female-dominated
religions. Among the Greeks of Thrace and the Bulgars the Day of Babo was celebrated
by married women who "while pretending to honour the midwife, worshipped the
Phallus behind closed doors, introducing newly-wed young girls to the society of
women; the participants, disguised as soldiers or policemen, behaved with inordinate
aggressivity towards the men they met" (Anastassiadou 1976, 101).
15. A number of minor women's cults preach similar visions of the union of male
and female in the age to come. Beit-Hallahmi has described a small female-dominated
messianic cult in Israel. Most of the beliefs of this cult are similar to those of any
number of male-dominated revitalization movements in contemporary Israel. One of
the few distinguishing features of this cult is their belief that in the New Age the two
sexes will become one (1992, 7off).
16. On the other hand, Ann Lee herself stated that "man is first, and the woman . . .
second" in the "order of nature" (quoted in Brewer 1992, 612).
17. The parallel system of leadership with a man and woman at each level was most
 Gender Ideology 213

likely introduced by a male leader, Joseph Meachum. Similarly, another male leader,
Benjamin S. Youngs was responsible for developing the theology of Holy Mother
Wisdom (Humez 1992, 86). It is important to bear in mind that parallel titles for leaders
did not necessarily mean equal authority, and particularly in the early years the distribu-
tion of power among male and female Shakers was not fully egalitarian (I thank Priscilla
Brewer for drawing my attention to this point, personal communication 1992).
18. When Ann Lee was alive she was seen as the "Queen of Heaven" by her follow-
ers. After her death the sect's theologians de-emphasized that role, and in nineteenth-
century Shaker writings Mother Ann is no longer described as superior to Jesus, but as
his completion (cited in Brewer 1992, 613).
19. Priscilla Brewer, on the other hand, argues that "Most Believers felt that men
and women should exercise unequal power because of their distinctive traits. Wherever
possible, economic and executive responsibilities were assigned to brethren. The femini-
zation of the Society's leadership beginning in the last half of the nineteenth century
came about only because of a worsening gender imbalance. The Shakers were never
committed to the complete equality of the sexes" (1992, 635).
20. While much of what I have described in terms of Shaker gender ideology is also
true of Christian Science, the Shakers addressed issues of gender far more explicitly. In
various places in her writing Eddy contends that gender differences really do not exist,
that men and women are complementary, and that women are superior.
 11
Leaders and Experts

WOMEN AND LEADERSHIP IN

FEMALE-DOMINATED RELIGIONS
Looking at Shaker history, Marjorie Procter-Smith has concluded that gender
ideology and gender of leaders are firmly interlocked. She shows that "the
system of dual [male and female] leadership was in place in the [Shaker] soci-
eties several years prior to the appearance of Youngs' Testimony, where the
doctrine of a dual deity was first expressed" (1985, vii). And, during the period
of Mother Ann's Work (the internal revival during the mid-nineteenth century
in which messages were received from deceased Shaker leaders, God, and Holy
Mother Wisdom), "As women became increasingly prominent as instruments
[mediums], female symbols became increasingly prominent in the symbol-
system of the sect" (1985, 196). The pattern that emerges is of women stressing
female symbols after they become prominent in leadership; we can assume that
this functions to validate their authority in much the same way that masculine
symbols validate the authority of male leaders in other religious traditions
(Catherine Wessinger, personal communication 1993).
A feature of all the religions treated in this book is that both in theory and in
practice leadership roles are open to women, and women predominate in most
positions of authority and prestige. Although few of the female-dominated
religions claim that women are "better" than men, quite a few assert that
women are more suited (or uniquely suited) to fill leadership roles. For the
most part, leaders in women's religions embrace rather traditional (albeit ex-
panded and empowered) female roles and traits. For example, when Virginia
Kerns asked the Black Caribs why women are more prominent in ritual, she
was told that actually anyone can participate in ritual (except children) but
women are more interested, more attuned to the supernatural as they age, have
better memories than men and so remember the intricate rituals, and feel more
gratitude and duty toward the ancestors. As we saw in earlier chapters, many
of the rituals performed by the Black Carib women involve cooking and other
activities that are common parts of women's lives.
216 Priestess, Mother, Sacred Sister
When Patricia Lerch asked her Brazilian informants why more women than
men are Umbanda mediums, the qualities most often named in describing
mediums were those also used by informants in discussing the attributes of a
good mother or wife. Women's life experiences were said to prepare them for
the role demands of spirit medium. "Men are perceived to be weak creatures
unable to endure the spiritual tests and the rigors of development. Where do
women get their spiritual strength? Strength comes from dealing with every-
day life" (Lerch 1982, 248). Suffering trains one for mediumship, and in the
words of one informant, "Women suffer more than men, both on the spiritual
level and on the material level. This history of suffering makes women better
suited to be mediums. Women suffer because of their husbands. . . . Women
learn to suffer with patience and resignation" (1982, 249). Women's suffering
typically stems from domestic problems—in particular, unfaithful husbands.
Male infidelity means that men do not give enough money to their wives, so
that the wives cannot educate their children and manage their households
properly. This situation reflects both the economic marginality of most Brazil-
ian women and the infamous sexual double standard. Yet the outcome is that
women are deemed to be the preferable religious leaders in Afro-Brazilian
cults. In addition, the actual work performed by mediums is more similar to
Brazilian women's work than Brazilian men's work: According to Ruth
Landes (1947), women are seen as more suited to the types of duties involved in
religious ritual-—cooking sacred foods and caring for the altars—because that
is what they do at home.
Suffering is a theme that emerges repeatedly in the life stories of women
leaders, and is used by many of the women's religions to explain why more
women than men are leaders. In a study of women Spiritist healers in Puerto
Rico, Koss-Chioino commented that pain seems to be more central to
women's than to men's life experiences.1 Women in sexist societies suffer from
the denial of their talents and worth, in addition to the physical and psychologi-
cal distress that all women undergo throughout their reproductive cycles. I
would argue that the feminization of suffering is used to justify women's
religious leadership in societies in which leadership is perceived to be a
male prerogative. Pain—a female prerogative—explains why women can be
religious leaders.

HIERARCHICAL RELIGIONS
Scholars have asserted that women's organizational style is less hierarchical
than men's (e.g., Barfoot and Sheppard 1980, on Pentecostal churches.) Ac-
cording to Michelle Rosaldo (1974, 29), well-articulated systems of rank are a
male rather than female mode. Emblems and insignia of rank, such as those
found in armies, are rarely created by women. A range of theories have been
offered to explain this gender difference. Drawing on Nancy Chodorow's
ideas, Iris Marion Young (1983) argues that because boys are mothered by
women, they come to relate to others in an oppositional (dichotomous) way.
This leads to a mode of conceptualization that emphasizes mutually exclusive
 Leaders and Experts 217
dualities (body/mind; self/other), which results in hierarchies and oppression
of all kinds. Carol Gilligan (1982) has argued that men's thinking is more
competitive, more concerned with winning and losing, whereas women's
thinking centers more on cooperation and care (see also Johnson 1988, 94-95).
We indeed find that most women's religions preach "egalitarian" ideologies—
that every human being regardless of sex, class, or age has the potential for
communicating with the spirits. However, in actual fact, in all the women's
religions certain categories of people are more attracted by, or attractive to, the
gods or spirits. In contrast to sociological theories that show women opting for
nonhierarchical organizational styles, in all these religions there is some sort of
hierarchy based on spiritual development. 2 Although women's religions tend to
provide equal opportunities to potential leaders with little regard to class and
education, rank can be achieved through acquiring skills and followers. Unlike
in many male-dominated organizations, neither wealth nor schooling deter-
mines who acquires skills and followers: ritual skills and clients often accrue to
the poorest, least educated, and most unfortunate women. Still, and let me
emphasize this finding, women's religious organizations do not seem to be any
less hierarchical than men's. Organizational sociologists know the "iron law of
oligarchy"—only the very smallest groups do not have hierarchy. Thus the fact
that women's religions are hierarchical is not totally unexpected, because all
groups other than very small ones have hierarchy. Patricia Martin's study of
feminist organizations in the United States strengthens this observation: despite
a myth of nonhierarchical relations, no particular internal structure seems to be
a defining characteristic of feminist organizations (1990).

Sande: A Case Study

Sande initiation rituals in which all initiates are covered in white clay seem to
dramatize the innate equality of all members. At the same time, Sande chap-
ters are hierarchically organized. The route to leadership in Sande is through
acquiring more knowledge. Increased adeptness is acknowledged by rites that
raise some women to higher ranks. Leaders or headwomen have special
skills—they are responsible for the initiates' training, and function as mid-
wives and advisers on gynecological problems. Some women are so adept that
they are known over a wide area and travel across linguistic and ethnic bound-
aries. The Sande leadership structure is as follows: The Mojo is the head-
woman; the Sowei are in charge of teaching the initiates; the Ligba initiate the
girls; the Ndoli are dancers (some of whom wear masks); the Klawa are counsel-
lors for the initiates; the Sowei Ligba are in charge of providing protection for
members of the Society.
This system of ranking has implications aside from ceremony. Caroline
Bledsoe argues that Sande serves the interests of older women while contribut-
ing to the oppression of younger women. In that sense, Sande actually intensi-
fies power differences among women. Sande leaders' power over female initi-
ates' sexual and domestic services can be used to gain the loyalty of young men
or those from outside lineages whom elders—male and female—seek to con-
218 Priestess, Mother, Sacred Sister
trol. Bledsoe interprets what initiates learn in bush school as "ideological dress-
ing on a rigid gerontocratic hierarchy" (1980, 69).
Bledsoe has observed that the skills Sande initiates learn are ones they al-
ready know—that the learning is symbolic. But unlike Carol MacCormack
who explains that what the girls really learn is cooperation (see Chapter 6) or
Bruce Lincoln who argues that initiates learn to see the inner cosmic signifi-
cance and sacrality in everyday activities (see Chapter 7), Bledsoe claims that
what initiates really learn is "absolute obedience to Sande leaders, both while
they are in the bush school and in later life. Disobedience or disrespect may be
threatened with infertility or even death" (1980, 68). The leaders can convinc-
ingly threaten infertility and death because they are the midwives. Both men
and women fear and respect midwives whose power stems from their exclu-
sive knowledge of obstetrics and gynecology. And despite myths to the con-
trary, Sande leaders actually hide more than they teach about reproduction and
midwifery, and even try to keep the knowledge of childbearing secret. Women
believe themselves to be utterly dependent on the midwives, and the midwives
are able to use women's dependence on them during birth to extract secrets
that further enhance their social power.

Other Examples
The manifestations of hierarchy vary among women's religions. The Afro-
Brazilian religions are characterized by elaborate hierarchies based on spiritual
talents rather than on knowledge (as in Sande). Diana Brown found that in
Umbanda centers, mediums who are sought out for their consulting abilities
gain great prestige and advance rapidly through the ritual hierarchy of the
center. In contrast to our expectations of supportive and nonhierarchical rela-
tionships among women, most cult leaders do not want their mediums to
practice outside the public sessions held at Centers.3 (The leaders say that most
mediums are not sufficiently developed spiritually to handle private sessions;
the mediums claim that the cult leaders are jealous.) On the other hand, medi-
ums wear simple white outfits and there are no class indicators such as jewelry
to show socioeconomic status. In many cases poor and black mediums may
outrank rich and white ones, and women may outrank men.
In Burma all the nat wives (mediums) attached to a particular nat have an
internal hierarchy, led by those who are full-time devotees. Devotees feel a
certain rivalry among themselves regarding whom the nat visits in dreams and
to whom he reveals his presence. Although there is no official centralized
organization of mediums associated with particular nats, there is an informal
yet clearly recognized hierarchy based on spiritual development.
Unlike Sande leaders, the zar leader is a fellow sufferer, someone who deals
with the same sorts of problems as her followers. Her authority does not lie in
orders that the patient must obey, but in her personal charisma and her skill at
the process of negotiating with the spirits. On the other hand, zar leaders are
addressed as shaykha (a title of respect), and have followers (or "daughters")
ranging in number from fifty to several hundred who attend annual rituals
 Leaders and Experts 219

provided by the leaders for their followers. Not only is there a difference in
rank between leaders and followers, but also between initiated and non-
initiated followers. At zar rituals non-initiated women attend, but have a lower
status; for example, they may not sit on the mattresses, or be served coffee.
Ironically, despite the existence of the Mother Church (see Chapter 12),
Christian Science congregations are more democratic than most other wom-
en's religions. There is no clergy and officers are elected from among members
of the congregation. On the other hand, and this is the essential point, Mary
Baker Eddy insisted on being acknowledged and obeyed as the sole discoverer
and founder of Christian Science.
The Shakers described an explicit chain of command, beginning with Al-
mighty God and Holy Mother Wisdom, then Jesus and Ann Lee, and then the
first church elders. Each Shaker community had a fully articulated hierarchy of
religious and secular leaders: deacons, trustees, and elders directing the spiri-
tual and working lives of members. Members were divided into three classes
of families: novitiates, junior or second class who still owned their own prop-
erty but ceased using it to join the production-consumption Shaker coopera-
tive, and senior or third class who had renounced their property and signed a
binding covenant with the Shaker community.
On the Ryukyu Islands laypeople have rather minor ritual roles; specialists are
responsible for the majority of ritual action. In the past the noro priestesses lived
in semi-seclusion, apart from their fellow villagers, often in a special dwelling
near the sacred grove. Douglas Haring quotes an informant from one of the
smaller Ryukyuan Islands, "They [noro] are women who live apart, wear ki-
mono twice the usual length, with double-length combs stuck in their hair. By
this costume they convince simple folk that they have come down from heaven
and have alighted on a mountain top" (1953, 112). On the main island of
Okinawa during the fifteenth and sixteenth centuries a strict religious hierarchy
was established by the government. The hierarchy was composed of a high
priestess, village priestesses, kin-group priestesses, and household ritual ex-
perts. 4 These female ritual experts were arranged into a precise, elaborated, and
strict hierarchy. Today there are four different names for priestesses (the mean-
ings of these names are not always clear) and a significant distinction between
noro (high priestesses who are responsible for leading and preparing public
rituals) and ordinary priestesses who are housewives or grandmothers.5

HISTORY AND PATH TO LEADERSHIP

What kinds of women become religious leaders or experts? Do they exhibit
from a young age leadership abilities that are nurtured by the previous genera-
tion of leaders? Are they women who are misfits—never quite able to fulfill the
role marked out for women in their particular cultures? Are they women with
heightened spirituality—women who are chosen by the gods? Or are they
women with an eye to personal gain and profit? 6
In the following section we will see that the two most common themes in
the life histories of women leaders are illness and an initial resistance to taking
22O Priestess, Mother, Sacred Sister

up the leader role. But since the life histories of most of these leaders are
reconstructed through the leader's own memories, I wish to begin by making
some general comments about autobiography. Typically, when people tell their
life histories, they relate stories with plots, themes, and character develop-
ment, rather than a random series of unrelated incidents. Although life in fact is
made up of millions of random and unrelated incidents, when people are asked
about their lives, they tend to try to make some kind of sense or order out of
what they have experienced. In the case of religious autobiography this is even
more true. In constructing religious autobiography—in answering the ques-
tion "How did you come to be active in this religion?"—individuals sort
through their life experiences and select those incidents and events that to their
mind show spiritual development. The material presented below should not be
treated as "objective" or factual accountings of people's lives, but rather as
religious interpretations of events deemed significant by specific individuals for
very specific reasons.
All of what I have just said regarding religious autobiography in general is
even more true in the case of women leaders. Sidonie Smith explains that
women who tell their stories "understand that a statement or a story will
receive a different ideological interpretation if attributed to a man or to a
woman. As a result, the [female] autobiographer . . . approaches her 'fictive'
reader as if'he' were the representative of the dominant order, the arbiter of the
ideology of gender and its stories of selfhood" (1987, 48). In other words,
when women religious leaders tell their life-stories they are well aware that
their stories will be judged by people who, to say the least, are uncomfortable
with the very idea of a woman leader. A similar point has been made by Marie-
Francoise Chanfrault-Duchet: "[I]n women's life stories, the social self does
not merely occupy a place within the social order; rather, its place is overdeter-
mined by the status of woman. This means that women's life-stories, unlike
men's, deal not only with the relation between the self and the social sphere,
but also, and above all, with woman's condition and with the collective repre-
sentations of woman as they have been shaped by the society with which the
woman being interviewed must deal" (1991, 78).
In sum, I ask the reader to take the life histories of women religious leaders
with a triple grain of salt: as autobiographies, as religious autobiographies,
and as autobiographies of women religious leaders.

Illness, Visions, and Being Called by the Spirits

In Chapter 5 we saw that illness is often what leads women to become involved
in women's religions. In several religions prolonged illness is a mark that a
woman is destined to be a leader. For example, in Brazil signs of incipient
mediumship include excessive crying, protracted illness, unexplained events,
unsolvable problems, and unusual occurrences. (Psychological testing of
women mediums in Brazil demonstrate that mediums tend to be reasonably
well adjusted; cf. Leacock and Leacock 1972, 326-327). The path to becoming a
medium usually involves a stage of sickness or other misfortune that a cult leader
 Leaders and Experts 221

interprets as caused by undeveloped mediumship. Umbandists believe that the

ability to be a medium is innate, inheritable, and God-given. Resistance is
dangerous—it can lead to further misfortune or even death. Husbands of future
mediums may object to or disbelieve in the mediumidade of their wives. Most
women mediums and cult leaders believe that a married woman should have her
husband's consent, but that if he won't give it, the spiritual order should take
precedence over his resistance, otherwise her illness will worsen.
On some of the Ryukyu Islands, priestesses are selected from certain families
or lineages. On the other hand, the woman is only considered qualified to be a
priestess upon experiencing kamidari—the mark of the divine in the form of
visions, hallucinations, dreams, and abnormal behavior. Shamans (yuta) judge
if the illness is kamidari or not. Only then will the candidate learn the dances
and rituals and stories of now.7 Ryukyuan yutas, like Afro-Brazilian mediums,
show life histories of marital discord, intrafamily conflict, chronic illnesses, and
hallucinatory experiences.
Historians and anthropologists stress that many Spiritualist mediums had
lost their fathers at a young age, suffered from insufficient parental affection as
children, or had unfortunate experiences with husbands or male friends.8 How-
ever, since most women in most societies have suffered from insufficient paren-
tal affection or from bad relationships with husbands or male friends, I am
reluctant to pay much heed to this supposed attribute. Of greater interest is the
claim by mediums themselves that they did not consciously choose to become
mediums; rather, the spirits chose them. It was often at adolescence and mar-
riage that potential mediums displayed the alarming symptoms and illnesses
that would be interpreted as the call of the spirits. Still, according to Laurence
Moore, "There was no consistent marital pattern among Spiritualist mediums:
some were married, some had been divorced and remarried, and some re-
mained single" (1977, 107).
Alex Owen contends that it is significant that Spiritualist mediums often
experienced illness at key stages in the development of sexual maturity and
adult femininity. Illness at these times offered attention and recognition, but
also provided an arena for expressing conflict. "In other words, spiritualist
mediumship was both expressive of an inner struggle with the problem of
femininity and instrumental in reconciling that tension" (1981, 209). Again, I
am not surprised to hear that women in sexist societies feel conflict at periods
in their lives in which they are brought face to face with the ways in which
women's freedom is restricted.
Burmese women also do not choose to become shamans. Instead, the nat
falls in love with a woman and wishes to marry her. If she resists, she will
suffer. It is said that nats love women who have beautiful souls. The nat makes
itself manifest through dreams or possession. Typically, another shaman identi-
fies the nat in a new devotee. Being loved by a nat is not sufficient to make one
a shaman or "nat wife"; what is needed is a formal marriage ceremony. The
marriage ceremony is quite expensive—costumes, orchestra, fees for shamans,
and food for guests. It may take years to accumulate enough funds to carry out
the ceremony.
222 Priestess, Mother, Sacred Sister
Finally, Gehan Wijeyewardene observes that an urban Thai medium typi-
cally considers herself "called" to her role, and describes having initially tried
to escape the caw (spirit), who continued to persecute her until she agreed to be
his "horse."
Before going on, I wish to reiterate that the stories of illness I have just cited
do not necessarily mean that women religious leaders have been ill more than
other people; these stories do mean that women religious leaders tend to
attribute greater significance to their illnesses. I am struck by Joan Koss-
Chioino's observation regarding women Spiritist healers in Puerto Rico: these
women describe having felt deep concern for the illnesses of their family
members from an early age, and that they are especially aware of the illnesses
they themselves experienced as children (1992, 18). If I understand Koss-
Chioino correctly, women healers in Puerto Rico are characterized by cogni-
zance of illness, and not necessarily by actual sickness. My suspicion is that this
pattern may be true in other religions as well.

Training
One path to leadership that is rarely (although very occasionally) found in female-
dominated religions is that of a woman consciously choosing a leadership role,
and then undergoing lengthy, formal training to learn the role. The absence of
formal training is consistent with the aversion to sacred texts and official doc-
trines that characterize most of the women's religions (see Chapter 12).
The path to Spiritualist mediumship, for example, does not include any sort
of prescribed training or apprenticeship. Indeed, lack of education was seen as
conducive to successful mediumship. Many sympathetic contemporary ac-
counts of mediums stress their naivety and lack of worldly knowledge.
A very short formal training path is found among Christian Science practitio-
ners who must take a two-week course that allows them to be officially recog-
nized as healers. The career of a practitioner begins when she has shown
success at healing family and friends. Then, through word of mouth her abili-
ties gradually become known in her own congregation. As the practitioner
becomes better known, patients give testimony to her healing prowess at
church during the Wednesday meetings, and the practitioner begins to be seen
as a source of information concerning Christian Science beliefs. Finally, after
she has healed a number of serious illnesses, the practitioner applies for formal
recognition from the Mother Church as a healer. According to Margery Fox,
"Formal recognition actually means in social terms that a number of people in
the branch congregation have made the decision to depend on this person for
friendship, comfort, health, and religious instruction and advice. They have
entrusted her with the most important elements of their lives" (1989, 106).
A somewhat exceptional pattern is exemplified by the Afro-Brazilian reli-
gions in which mediumistic abilities require formal training and developing.
Once an individual decides to become a Umbanda medium she passes through
approximately four years of preparations and preparatory roles (Lerch 1982).
In Bahia Brazil, initiation requires time and money—three months of absolute
 Leaders and Experts 223
seclusion in the cult house, abstinence from sex, rich foods, and amusement.
There is a large fee for initiation, which is usually paid off gradually in service
to the cult center.

Mothers and Daughters

In many of the female-dominated religions, the women who become religious
leaders are women whose mothers were religious leaders. For instance, when a
Korean shaman dies her spirit usually possesses her daughter or daughter-in-
law (Harvey 1979, 127). Among Spiritualists in South Wales today, mediums
are often daughters of mediums (introduction to Spiritualism takes place in
childhood, yet regular involvement does not usually happen until after mar-
riage.) Mayotte women are much more likely to be possessed by spirits if their
mothers or sisters are; this is not true for Mayotte men (see page 191). And on
Okinawa it was formerly the custom that a noro was succeeded by her daugh-
ter.' And still today, in the life stories of priestesses collected by Yoshinobu
Ota, we find one who was "called by the deity to succeed her mother ... as a
priestess" (1989, 117).
In Burma certain nats are particularly popular because they are served by
natkadaws (nat wives or shamans) who go from town to town offering prayers
and dances to the images of their nats. Typically, these women are daughters of
other natkadaws. Two shamans interviewed by June Nash reported that they
were asked by their dying mothers to become natkadaws, refused, became seri-
ously ill, were advised by other natkadaws to join their ranks, and recovered.
According to Janice Boddy, a Sudanese woman who is interested in becom-
ing a shayka (zar leader) learns by apprenticing, often with a close maternal
kinswoman. The proclivity to be a shayka tends to be handed down in the
maternal line. The leader learns how to call the spirits, bargain with them, and
recognize their individual characteristics. She does not declare herself a shayka;
others have to attribute this status to her as her reputation grows. Similarly,
Ethiopian women zar leaders claim their status through transfer of power and
knowledge from their mothers. Men, who do not inherit zar spirits from their
mothers, may claim that zar spirits had kidnapped them during childhood
(Messing 1958).
The propensity for daughters of leaders to follow in their mothers' footsteps
is consistent with the emphasis on family relationships and the pattern of
matrifocality in women's religions.

Korean Shamanism: A Case Study

Korean women called by the spirits to become shamans try to avoid taking on
the shaman role, primarily because shamans and their families are regarded
somewhat as deviants and outcasts. The path to becoming a shaman is not an
easy one, and no one claims to have sought or selected this career. Potential
shamans are tormented by gods and ancestors until they feel compelled to
accept. The only other choice would be to deny the calling and die insane. The
224 Priestess, Mother, Sacred Sister

central event indicating the shamanistic calling is sinbyong or possession sick-

ness. Once a woman accepts the role of shaman, the same spirits who had
tormented her become her allies.
The spirits tend to call a potential shaman in middle life, at a time when she
is suffering from a specific misfortune or more general discontent. Symptoms
of sinbyong include a wide range of physical distresses (dizziness, headaches,
digestive disorders), hallucinations and strange dreams, and inappropriate be-
havior (such as speaking openly about matters that are normally not dis-
cussed). The behavioral symptoms, disruptive to the family and community,
are what confirms the diagnosis of possession sickness.
Youngsook Kim Harvey traces the typical spiritual development of a Korean
shaman: (I) The victim feels conflict between her own needs and the housewife's
role, in addition to conflict with her husband or in-laws. (2) She tries unsuccess-
fully to cope with these conflicts but feels trapped by them. (3) She becomes ill
with vague physical symptoms, and is given a period of time to retreat from
household responsibilities. (4) The symptoms persist, the woman is exempted
from household duties, and her family tries to find a cure for her illness. (5) The
victim recovers. (6) Old patterns of interaction within the family are resumed.
(7) She becomes sick once more. This process may be repeated a number of
times, until finally an experienced shaman makes a diagnosis of sinbyong (posses-
sion sickness), and initiates and trains the novice shaman.
Harvey documents one case in which the process continued during twenty-
eight years. That particular shaman's life story, not atypical, included being
forced into an arranged marriage, refusing to consummate the marriage and
being raped by her husband, trying hard but unsuccessfully to accommodate
herself to the housewife role, deaths of her children, illness, re-marriage to a
man who became involved in extra marital affairs and who brought financial
ruin on their family, abandoning her infant daughter because she did not have
the financial resources to care for her properly, and—finally—a diagnosis of
possession sickness.
Brian Wilson details the life story of another shaman: She was raised by a
mother who supported the children because the father was ill. Her older sister
was given away to another family. As a young girl she wanted to be a boy, and
she resented having to obey an older brother. Her parents sent her to Seoul as a
housemaid. She was married to a man not of her own choosing. Her adult life
included many hardships, including the war and being financially responsible
for her baby while her husband was away. One of her children died, her
husband cheated on her with another woman, and she quarreled with him.
Finally, she became possessed by her dead mother-in-law's spirit, which gave
her a new power over her husband.
Although victims of sinbyong are likely to behave in ways that are both
socially unacceptable and extremely disruptive to family life (e.g., they may
stop performing their household duties), they are not personally blamed for
their behavior because Koreans believe that spirits like to possess people who
have been mistreated by their families. By allowing the victim to take on the
role of shaman, her family may disprove suspicions of mistreatment.
 Leaders and Experts 225

Traits of Leaders
The life stories of Korean shamans are especially important because the shamans
themselves make use of autobiographical details to demonstrate to their clients
the powers of the gods and the dangers of ritual neglect. Earlier in this chapter I
suggested that religious autobiographies are not objective histories. Instead,
they are selections and interpretations of events that seem to the teller, at the time
of telling, to explain why her life is the way it now is. I believe that illness is
significant to female religious specialists not because they are ill more than other
people, but because they themselves see their illnesses as meaningful to their
religious roles. In light of the discussion of sickness and healing in Chapter 5, I
do not find this to be at all surprising—women's religions excel at interpreting,
sacralizing, and curing illness. The importance of a history of illness for a reli-
gious leader is not the illness per se (except insofar as it makes her more under-
standing of her clients' illness and suffering 10 ), but rather her triumph over
illness. Female religious leaders with a history of illness dramatize and personify
the existential claim that suffering is not inherent to the human condition.
The reported reluctance of the leaders to take up the mantle of leadership
demands a different sort of explanation. As I demonstrate later on in this
chapter, the status of female religious leaders tends to be ambiguous because
none of these religions is located in a truly nonsexist society. The fact that in
many situations the potential leader is chosen (or even harassed) by the gods
and does not herself choose the specialist role is a solution for female leaders
who do not wish to openly challenge prevalent sexist norms. This is consistent
with what we have said about gender ideologies in the previous chapter. I find
it interesting that even in those women's religions in which leaders are highly
respected, they rarely take credit for their own skills or achievements. Esther
Pressel quotes one Umbanda medium as telling her that she does not know
how to heal on her own, "I do things my spirits tell me, but I myself do not
know how to perform them" (1974, 152). Mexican Spiritualist healers also
deny credit for their healing, although they do demand that patients obey their
healing instructions.
Finally, a few words are in order concerning the types of women who
become religious leaders. In Table 5, I have summarized twelve of the most
common character traits of leaders in women's religions. Several of these traits
warrant further comment. First, many religious leaders are especially gifted
musicians, orators, or actresses. These talents are necessary for religious special-
ists who excel at possession trance; they must be adept at portraying the
various spirits and gods, many of whom are male or in other ways very
different from the possessed woman herself.
Second, the leaders seem to be both especially intelligent and especially intu-
itive and empathetic. I find it interesting that psychologists have noted that these
two attributes tend to go together, especially in girls. Sharon Conarton and
Linda Silverman, in the context of a book on feminist psychotherapies, contend
that: "Bright children in particular have high degrees of emotional sensitivity,
and are likely to show compassion for others even as toddlers. . . . The gifted
226 Priestess, Mother, Sacred Sister

Table 5. Traits of Women Religious Leaders

Empathy: A zar leader is thought to be one who "really understands" (Kahana 1985, 133).
Korean shamans are typically highly sensitive to intuitive cues of others, and excel at interper-
sonal skills.
Rapport: The most important attribute of a Christian Science practitioner is rapport with her
patients. Despite the fact that practitioners send bills to their patients, the affective ties between
the two are extremely important (Fox 1989).
Kindness: Afro-Brazilian mediums must be loving, kind, patient, motherly, and pure (Lerch
1980; 1982).
Intelligence: Korean shamans tend to be unusually intelligent (Harvey 1979).
Age: On the Ryukyu Islands younger women typically show less interest in their role as
kaminchu than older women. Older priestesses seem to have their thoughts and attention on
spiritual matters, and behave and speak differently from other people.
Eloquence: Common characteristics of Korean shamans include intelligence, ability to impro-
vise, verbal fluency and persuasiveness, goal orientation, sensitivity and intuition, calculating and
manipulative interpersonal skills, a sense of justice, artistic talent, and physical attractiveness
(Harvey 1979). A characteristic shared by most Christian Science practitioners is eloquence (Fox
1979). Christian Science healing is carried out through verbal exchange. Richards (1973, 72) has
found that Sande leaders are women who are gifted in the traditions of the group and are great
religious and moral leaders who employ myths and mysteries in explaining phenomena.
Acting ability: The Spiritualist medium role necessitated the ability to represent or speak for
spirits. Many mediums were fine actresses, able to enact the personae of a large number of spirit
controls—both male and female.
Musical ability: Zar leaders tend to be particularly talented at drumming and remembering
special songs for summoning spirits, and excel at entertaining, surprising, and thrilling their
audiences (Kennedy 1978).
Poise: Afro-Brazilian priestesses are known for their poise. "Under her guidance there flourishes
a realm of peace and security" (Lande 1935, 388). Zar leaders also tend to be self-confident people
who have the ability to inspire confidence and hope in others (Kennedy 1978).
Courage: The prestige of the Afro-Brazilian medium stems from the fact that it is difficult to
become and continue being a medium; it requires fortitude, courage, and the ability to withstand
pain and suffering.
Authority: Senior priestesses in Afro-Brazilian religions are characterized by their sense of inde-
pendence and power. Among senior priestesses possession was actually discouraged as they liked
to think of themselves "as mastering the deities instead of serving them" (Landes 1947, 53-54).
The Afro-Brazilian religions teach that the gods use misfortune to punish people, and priestesses
know how to manipulate the gods to cure. Gods are considered whimsical and unreliable (Landes
1940, 267). Priestesses ("mothers" in Bahia) are considered far more reliable, and know how to
control the gods.
Competitiveness: Modern Spiritualist mediums tend to be competitive, exhibitionistic, and
manipulative (Macklin 1977).

female usually excels at taking care of others. Her antennae may be ... [espe-
cially] attuned to the needs of others" (1988, 45-47). Given that many of the
women religious leaders excel at spirit possession (see Chapter 9), it is interest-
ing to note that David Gutmann has found that men whose TAT (Thematic
Apperception Test) scores indicate strong ego boundaries also tend to score high
 Leaders and Experts 227
on other measures of morale, social effectiveness, and mental energy. For
women, on the other hand, he found a trend in the opposite direction: Women
whose TAT scores showed weaker ego boundaries (please recall that in Chapter
9 I connected weak ego boundaries to the ability to become possessed by spirits)
"often achieved top scores on interview rating of contentment and effectiveness,
and they achieved notably higher scores than women with a more 'masculine'
style—that is, women who approached the TAT in a reasonable, delaying and
boundary maintaining fashion" (1965, 231). Put differently, women who excel
at spirit possession might very well be women who are generally competent and
strong—in other words, good leaders.
Third, many women religious leaders take up the mantle of leadership in old
age. Regarding new Japanese religions, Kyoko Motomochi Nakamura makes
the point that mature foundresses "have experienced the whole life cycle of a
woman, having lived as a girl, a wife, a mother, and sometimes a divorcee or
widow. . . . Nine times out of ten the women founders have lived dramatic
lives of material as well as spiritual oppression; their own suffering allows them
to attain a most penetrating insight into the sufferings of others" (1980, 141). It is
far from coincidental that most Japanese foundresses began to preach after their
children were grown and their husbands' (and, I may add, children's) demands
on their time had lessened (cf. Sered 1987; Sered 19913). As I suggested in
Chapter 7, few wives and mothers are able to (choose to?) abandon their family
responsibilities to pursue extraordinary religious paths. 11 Moreover, religious
leadership roles (and especially healing roles) allow women to continue to use
mothering skills in a postmaternal period of life (Koss-Chioino 1992, 20).
And finally, almost all of the traits I have summarized are indicative of
knowledge or talent rather than of authority. In many women's religions what
we seem to find is more of a specialist or expert role than a true leadership role.
With the glaring exception of Sande, very few of the "leaders" of women's
religions exercise much direct control over the lives of other participants. This
is consistent with the growing literature on the more cooperative leadership
style of American women ministers and rabbis (cf. Simon, Scanlan, & Nadell
1993). I would especially note Edward Lehman's study of mainstream Protes-
tant ministers in which he found that men were more willing than women to
use coercive power over the congregation whereas more women ministers
sought to empower their congregations (1993).

LEADERS AND GAIN

A familiar theme in the literature on women and religion is that women
gain—materially or emotionally—by becoming leaders. A great deal of this
literature focuses on psychological issues—women use religious leadership to
compensate for deprivation experienced in other aspects of life. (I have ar-
gued against this model in Chapter 2.) Some writers have proposed even
narrower compensation models: Becoming a religious leader allows a small
number of women to reap economic gains. I have not found this explanation
228 Priestess, Mother, Sacred Sister
convincing. Quite to the contrary, few leaders in women's religions seem to
accumulate real financial benefits.
For example, Spiritualist mediumship was one of the few career opportuni-
ties open to women in the nineteenth century, and opponents of Spiritualism
accused mediums of tricking clients out of money. However, according to
Laurence Moore (1977, 108), few mediums became rich. Not all mediums
even demanded payment in return for their services (although most were
willing to accept it). Spiritualist believers were unlikely to give regular finan-
cial support to mediums, and the role of permanent minister was generally
unavailable. In fact, many mediums ended their careers as paupers, alone and
destitute.
Like Spiritualism, Christian Science offered women leadership opportunities
at a time when almost no other religious denominations did so. Christian
Science practitioners theoretically can make a living from their healing work,
yet a recent study suggests that most make very little money from their healing
practices (Fox 1989).
Neither do Burmese nat shamans make sufficient money from conducting
rituals for this to be seen as a reason for a woman to become a shaman. Korean
shamans, on the other hand, do charge for their services, and the money earned
by a shaman is often the difference between abject poverty and some degree of
financial independence. However, studies do not indicate that Korean shamans
ever become wealthy.
More meaningful benefits seem to lie in the realm of relationships. June
Macklin (1977) argues that being a Spiritualist medium allows one to redefine
and reorganize events in a way that ensures some control over important
relationships with family members. Becoming a professional medium pro-
vides an acceptable escape route to some women who are unhappy with their
husbands and domestic life. Laurence Moore concludes that mediumship offers
a deliverance from boredom, marital abuse, and loneliness. "Mediumship was
an occupation not often pursued by women who enjoyed physical well-being,
economic security, a happy family and social life, and sexual fulfillment" (1977,
129; for my critique of sexual frustration theories see "Conclusion," p. 279).
Spiritualist culture gave women possibilities for attention, mobility, and status
denied elsewhere, and possibilities for circumventing rigid gender norms. The
professional medium had far greater opportunities for travel and adventure
than other American women.
Patricia Lerch (1982) found that Umbanda spirit mediums reported satisfac-
tion with their role for a number of reasons: listening to other peoples' prob-
lems teaches how to deal better with one's own, pleasure from being able to
help troubled people, having a purpose in life, recovery from lingering illness,
and a steadying influence that their possessing spirits had brought to their own
lives. According to Esther Pressel, "the mastery of [their] personal difficulties
is a type of self-reward" (1974, 205). Only for a very few people is the role
financially rewarding, and even they do not become rich from it. The rewards
are more subtle: freedom to leave the house and socialize, respect for unself-
ishly helping others, a social life at the Umbanda Center, the attention one
 Leaders and Experts 229
receives while possessed, and the fun and drama of the rituals. Other advan-
tages of being an Umbanda medium include prestige (clients think of them as
having access to super-human wisdom), power (by being at the center of a
client network a medium can offer clients practical help through reallocating
resources, goods, and services), and modest economic benefits that can supple-
ment, although not usually replace employment (Lerch 1982).
These examples indicate that three of the most important gains experienced
by leaders of women's religions are enhanced spiritual talents and insights,
increased success at dealing with relationships, and the possibility of acquir-
ing some of the social advantages normally available only to men.
Becoming a shaman in Korea leads to a reversal of power within the family.
"Such a radical role reversal is possible only where players temporarily suspend
the conventional order of things in favor of a new order of reality imposed by a
power transcending the man-made social order. The agent of this transcenden-
tal power is the mudang [shaman]" (Wilson 1983, 125). Once the diagnosis of
possession sickness has been accepted, a number of changes take place in the
family dynamics. First, the victim's in-laws no longer accuse the victim of
faking illness to avoid work. And, once the in-laws accept her calling, the
victim no longer accuses her in-laws of causing her illness. Since both the
woman and her family are now seen as fellow victims of supernatural actions,
they bond together to begin to deal with and benefit from the changed situa-
tion. The former dyadic relationship between the in-marrying victim and her
in-laws is transformed into a triadic relationship involving the spirits who
must be consulted about any actions involving the victim. The "victim," of
course, is the only one who has direct access to the spirits, so her position in the
family is enhanced. Further strengthening her position is the fact that the
possessing spirits are often ancestral ghosts from her husband's lineage.
When a Korean woman takes on the role of shaman, other members of her
family take over her former household responsibilities. In some cases the house-
work and childcare are assumed by female relatives; in other cases the husband
needs to become involved in household chores and childcare. Typically, once a
woman becomes a shaman—communicating with the spirits, financially inde-
pendent, newly self-confident—roles within the family are transposed, with
the shaman functioning as the household head. A Korean shaman experiences
the freedom and power normally found in the male domain. According to
Wilson, the shaman may wish that she could have been born a man, but "Not
penis envy, she wanted power and has acquired it now as a healer" (Wilson
1983, 124). He quotes one Korean shaman: "When I die, I want to be a male
spirit. I want to be a great general like the Spirit General so I can lead hundreds
of thousands of people" (1983, 122).

STATUS
Leaders operate within cultural contexts that are, to a greater or lesser extent,
patriarchal. Women religious leaders—women who are recognized as being in
some way more powerful than most people—threaten patriarchal structures
230 Priestess, Mother, Sacred Sister
and ideologies. Thus, for example, opponents of Christian Science, both in the
nineteenth and twentieth centuries, focused their attacks on Mary Baker Eddy
(and especially Eddy as a woman) more than on the teachings. On the other
hand, within the religion, a great deal of respect accrues to the successful
practitioner, and Christian Science practitioners are well esteemed in their
church communities.

Korea: A Case Study

The ambiguous position of women leaders in female-dominated religions lo-
cated in patriarchal societies is illustrated by the Korean shaman. There are two
words describing this role: mansin and mudang, one polite and one derogatory.
The two words reflect a cultural situation in which the power of the shaman is
scorned by men, but understood and utilized by women. Wilson makes the con-
nection between the systematized oppression of women and the image of sha-
mans in Korea. "The stereotype of mudang as ignorant, irrational, perverse crea-
tures is but an extension of the Confucian stereotype for all Korean women. The
mudang dares put her 'ignorance' and 'irrationality' on display, in dramatic form,
for all to see and hear" (1983, 126). On the other hand, at the Korean kut ritual,

Garbed in the red robes of an antique general or wielding the Spirit Warrior's
broadsword as she drives malevolent forces from her path, the Korean mansin
[shaman] claims an imposing presence. Even in everyday dress and sprawling
comfortably on the heated floor of her own home, she speaks with authority. By
virtue of the powerful gods who possess her, she can summon up divination
visions and probe the source of a client's misfortunes, exorcise the sick and chroni-
cally unlucky, remove ill humors from those who have difficulty finding mates,
and coax a reluctant birth spirit into an infertile womb. The professional shaman
makes the gods and ancestors a vivid presence in the home; she spots them in her
visions and gives them voice in trance (Kendall 1989, 138).

In ancient Korea there were both male and female shamans. Although an-
cient rulers most likely combined the roles of politician and shaman, these two
functions were gradually differentiated. By the historical period female sha-
mans outnumbered male shamans and political leaders were almost all male.
Traditionally, female shamans have had three major functions: as priestesses
they presided in national ceremonies, both agriculturally oriented ones and
rites offered for the well-being and blessing of the royal family; as exorcists
they were invited to drive out evil spirits and solicit the favor of the gods when
there was a sickness in the family; and as diviners they foretold the future of the
nation and of individuals.
A number of factors explain the overwhelming preponderance of female
shamans in contemporary Korea. First, shamans were systematically removed
from positions of national importance and driven underground and degraded.
This led to the profession's becoming increasingly dominated by women; men
had other more powerful and more appealing career options open to them.
 Leaders and Experts 231

Korean shamans have been able to obtain limited material and social benefits
from spirit possession. Shamanism allows some women to escape from oppres-
sive and stressful family situations, to form lucrative and satisfying careers, and
to use their skills for socially recognized beneficial purposes. Women have few
other ways of obtaining these (or similar) goals in Korean society. During the
Yi Dynasty (1392-1910), whose founding fathers were neo-Confucianist schol-
ars, Korean shamans were attacked for their "lewd" and deceitful practices. In
the twentieth century both the Japanese colonial administration and the inde-
pendent Korean government persecuted shamans for "superstitious" practices
that are counter to modernity. More recently, the government once again has
allowed shamanism, but shamans and their families are considered as having
outcast status. Shamans, singing and dancing in public, have been disliked by
Confucians, among other reasons because they do not behave in the proper,
modest female manner. Officials have labeled shaman's activities as fraudulent,
and endeavored to protect the public from being "duped" (Kendall, personal
communication 1992).
A second reason for the preponderance of female shamans stems from the
fact that within the home Korean women have been responsible for dealing
with the same sorts of ghosts and spirits that shamans call on. Women—
housewives and shamans—deal with the restless and potentially dangerous
dead.
Despite centuries of persecution, shamans have continued to play an impor-
tant role in national ceremonies. Kendall notes that in the village that she
studied shamans are hired to make an annual offering to the community tute-
lary god, to exorcise malevolent forces, and (before water pumps became
available) to petition the rain dragon during droughts (1985, 31). Her descrip-
tion of the activities of Korean shaman also includes numerous small-scale,
institutionalized, domestic rituals (such as lunar calendrical rituals) on behalf of
children and other family members. In short, Korean shamans are at one and
the same time community priestesses and community outcasts.
In the following pages I present two models that describe the statuses of
women religious leaders. These two models are not absolutes, however, and
many leaders experience both models, either sequentially or simultaneously
(one at the hands of their adherents and one at the hands of the society at large).
What I see as especially interesting in both models is the role of the leader's sex
in accounting for the status accorded her.

Model 1—Persecution
Social ostracism was often the fate of Spiritualist mediums, and many medi-
ums were repudiated by their families. Professional mediums were criticized
for abandoning the traditional female domestic role, and (justifiably or not) for
promiscuity. The history of the latter half of the nineteenth century shows
growing legal restrictions placed on mediums. Mediums were sometimes ar-
rested (on various charges), and the increasing institutionalization of the medi-
cal profession cut into the demand for medium healers.
232 Priestess, Mother, Sacred Sister

The life of a Spiritualist medium was not easy. "It was difficult at best to
maintain professional status on traits universally recognized as qualities of
physical and intellectual weakness, even if they did imply moral superiority
[see Chapter 10]. . . . Seance goers often treated a private medium, because of
her passiveness, as an unimportant intermediary, to be praised if things went
well, but only for her strange gifts rather than for her trained skills. A good
sitting might save the medium a scolding, but not necessarily the humiliation
of being bound, gagged, and searched to insure proper 'test conditions'
(Moore 1977, 119). The ritualistic testing of Spiritualist mediums for fraudu-
lence should be interpreted both in terms of sex and class. When working-class
women mediums were bound with leather straps and locked in cabinets by
upper-class men "the motif of male mastery surfaced. . . . Female powerless-
ness was especially evident in these bondage rituals and it is possible that male
spirit aggression [by female mediums] was a partial response to this enforced
denigration" (Owen 1981, 231).
The case of Spiritualist mediums is not unusual. According to Procter-
Smith, Shaker founder Mother Ann Lee was persecuted as a female religious
leader. She was accused of being a witch and a destroyer of families, both of
which are common charges against powerful women (1985, 19; cf. Barstow
1986 on Joan of Arc; Wessley 1978 on the founder of the Guglielmites). Lee was
also accused of harlotry, drunkenness, and an assortment of other crimes. She
and her followers were beaten and driven out of their homes in a number of
towns. Finally, Mother Ann was tied by the heels behind a wagon and dragged
for several miles over an icy road. She died from these and other injuries in
1784. It is significant that much of the public criticism leveled against the
Shakers focused on derogatory depictions of Ann Lee as a woman. "People
were outraged by what they thought of as the popery of celibacy, which
threatened to break up families. But they were even more incensed at the idea
of full equality of women with men, which clearly struck at the roots of family
and society, neither of which recognized anything approaching a legal auton-
omy of women" (Whitson 1983, 15). Thus, in different ways, Lee's gender
was relevant both to her followers and their detractors. For her followers, Lee
was the female completion to the male Jesus (see Chapter 10); for her detractors
she was the epitome of a woman who rejected the subordinate role deemed as
"natural" by the society of her day.12
In Burma, female nat shamans are sometimes believed to be sexually promis-
cuous. Being loved by a nat means having [metaphorical] sexual relations with
him, and shamans dance in public—possessed, wildly, and with abandon. In
the Burmese village studied by Melford Spiro, 60% of the male population
believe that all shamans are dishonest and 30% believe that most are dishonest.
Women, on the other hand, believe that shamans are honest, but still have a
low opinion of them. Mothers have been known to beat their daughters to try
to prevent their marrying nats. Shamans themselves seem to share these cul-
tural attitudes. Again, let me stress that in the charges leveled against these
shamans, the fact they that are female is central—they are accused of sexual
promiscuity.
 Leaders and Experts 233

Model 2—Respect and Honor

In contrast to the previous examples in which leaders were accused of witch-
craft, promiscuous sexuality, and breaking up families, the status of Sande
leaders is consistently high. They are believed to possess powers greater than
those of regular mortals. These very independent women usually have hus-
bands and children in the outside world, yet disappear for weeks, months, or
even years into the Sande bush schools, where their power is near absolute.
When they leave the Sande bush, their elevated status carries over into village
life.
Similarly, Gehan Wijeyewardene relates that nineteenth century female Thai
mediums came from high-status groups, upper-class elites, and royalty.
Wijeyewardene cites a description of Princess Ooboon Lawana (sister-in-law
to the chief), who "was called upon to question the spirits whenever any
difficulty occurred either in public or private affairs" (1986, 154). This same
woman was also a large-scale trader. In modern times, urban "mediums and
clients, particularly regular clients, describe the relationship as a total one,
almost like that of devotee to deity" (1986, 163). Male monks in Thailand are
highly respected, and on the face of it, female mediums do not seem to com-
pete with or contest monks. Yet "in their own view, they seem covertly to be
challenging the status of monks" (Wijeyewardene 1986, 184). Some mediums
suggest that the rituals of monks are inadequate, or similar to their own rituals.
The medium may even see herself as filling a role that is yet more demanding
than that of the monk. A monk can leave the order, but a caw [spirit] may not
allow his horse [medium] to leave.
Also on the Ryukyu Islands the status of the noro (priestess) is high—they are
a sort of peasant aristocracy. Villagers stand in awe of the noro, and treat them
with respect. Douglas Haring conducted fieldwork on Amami Island where
priestesses wear white robes and head cloths, silver and feathered ornaments,
and a sacred necklace of rock crystal beads interspersed with curved semipre-
cious stones believed to be magically potent. One of Haring's informants
recalled that in his childhood the noro would ring bells on the mountains all
night "and the people thought the gods had come down" (1953, 113). Priest-
esses on Okinawa, according to William Lebra, are always seen as working for
the benefit of the community and not in their own self-interest (1966).
On Okinawa the impressively dressed chief priestess was spiritual head of
the island, and perceived as the earthly abode of one of the most important
gods. The chief priestess was also involved in secular affairs of state. Lebra cites
accounts by Chinese and Japanese visitors to Okinawa in which the chief
priestess is described as being above the king in rank, and in which priestesses
judge and punish crimes. 13 A high level of theological sophistication character-
izes many priestesses: speculation about the nature of the cosmos "seem often
to be objects of discussion among the priestesses who are in charge of ritual
affairs" (Mabuchi 1968, 135).
During the time of the Okinawan Kingdom, the noro were a female theoc-
racy with enormous influence at court. With privatization of land under Japa-
234 Priestess, Mother, Sacred Sister

nese rule during the last century, however, many noro lost their traditional
agricultural land and economic independence. The land reforms of 1899—1903
under Japanese rule placed their land in the hands of male relatives. The evi-
dence indicates that it was the Japanese, not the local Ryukyuans, who
chipped away at the priestesses' power base.
According to Ruth Landes, in Bahia, Brazilian people may complain about
the high fees of the priestesses, but have no doubt about their abilities. Clients
and mediums evince a great deal of respect for the priestesses, begging bless-
ings from them, averting their gaze, and speaking very softly in their presence.
"Trained to rule independently, she [the priestess] has developed into a type of
matriarch that is not only unique in modern times but is anachronistic in
patriarchal Brazil" (1940, 268). Priestesses call themselves by the names of their
gods rather than by the names of their husbands to show their personal
independence—that they belong to the god or goddess and not their husbands.
In the Brazilian case it is crucial to take into consideration issues of class.
Diana Brown found that in Umbanda, lower-class women frequently become
mediums and leaders and can achieve positions of considerable local influence.
Lower-class men, in contrast, are relatively more powerless in Umbanda. In
the middle-class sectors, leadership roles for women are rarer and men tend to
predominate. This finding leads us to consider the intersection of gender and
class in women's religions. In Sande, on Okinawa, and in nineteenth-century
Thailand women religious leaders came from the elite classes and were treated
with extraordinary respect; in Brazil, contemporary Korea, and the zar region
women religious leaders come from lower socioeconomic classes and are
treated ambivalently. This contention is strengthened by Margery Wolf's analy-
sis of a low-status Taiwanese village woman who began to display shamanistic
behavior; although she had many of the attributes of a successful shaman, she
was eventually labeled crazy "because of her marginal status in the community
and in the male ideology" (1990, 419).

MEN AND LEADERSHIP

Given that the theme of this book is women's religions, it is somewhat of a
tautology to point out that most leaders are women. A more interesting ques-
tion is whether there any men who are leaders, and if so, what are their roles.
We can discern five patterns of specialist roles for men.

Pattern i: Gender Reversal

On the Ryukyu Islands there are male ritualists called uranai who are diviners
and fortune tellers. When performing rituals, they dress and talk like women.
In modern Korea almost all professional shamans are women; the few men are
marginal and, until recently, dressed in women's clothing. Melford Spiro
(1967) found that in Burma many male shamans were homosexuals. Similarly,
Edison Carneiro has discovered that Afro-Brazilian priests in Bahia try to
assimilate the ideal type of priestesses, acting in a feminine manner and being
 Leaders and Experts 235
homosexual (1940). And Gehan Wijeyewardene notes that urban Thai male
mediums are often homosexual or transvestites. Some are blatant in their
adoption of women's clothing and few are married to women (1986).
In women's religions, where (as we saw in the Chapter 10) ideologies of
gender difference tend to be highly elaborated, leadership is so totally associ-
ated with female characteristics that a man must "become" female in order to
take on leadership roles. This pattern is particularly significant because it is not
a common one (in reverse, of course) in male-dominated religions. While
celibacy for women is encouraged in certain male-dominated religions, I know
of no instances in which women who are religious leaders are expected to dress
like men or become lesbians. In Judaism, for example, women are explicitly
forbidden to wear men's clothing, and women who become religious leaders
are typically women who excel at female (not male) tasks. 14

Pattern 2: Shared Leadership

Mother Ann Lee shared leadership with her brother William and her adopted
son James, and Shaker organization insists on a woman and a man at each level
of leadership. At Christian Science Sunday services two readers—almost al-
ways a man and a woman—lead the service. Again, this pattern is not a
common one in male-dominated religions, where women may have roles as
assistants but rarely as equal partners in leadership. In Catholicism, for exam-
ple, nuns are not the structural equivalents of priests, and there is no woman
sharing the papal throne at the Vatican. And to take an even more extreme
example, in Islam there are no formal leadership roles for women whatsoever.

Pattern 3: Secular Administrators and Scribes

In a number of the women's religions men hold important secular positions as
administrators. Men in Afro-Brazilian religions serve as ogans (providers) who
pay for ceremonies, keep the cult house in repair, and, if necessary, defend the
cult before the police. Women are the spirit mediums, but men are in charge of
the finances and the organization. On the Ryukyu Islands there are men who
represent the village in dealing with the priestesses—these men raise money
from villagers to support the cult, and take care of the cult's finances. In
Christian Science women predominate as practitioners and men predominate
in the higher-paying administrative jobs in the organization. A variation on
this pattern is men who specialize in writing books and religious tracts. (See
Chapter 12 for a discussion of the role of sacred writings.) In particular, among
Spiritualists and in the Afro-Brazilian religions, men seem to be tremendously
prolific writers.
This third pattern is also uncommon in male-dominated religions, -where
women are typically excluded from both administrative and spiritual leadership
positions. To take an example from modern Israel, women have been prohibited
from serving on municipal religious councils—bodies that perform purely secu-
lar functions of allotting and overseeing finances for religious institutions.
236 Priestess, Mother, Sacred Sister

It could be argued that if men govern the financial and bureaucratic aspects
of the religion, it is not really female dominated. Therefore, I have been careful
to include in this book religions with a male/female-administrative/spiritual
mode of leadership only when the administrative side is not believed to be
more important than the spiritual side and when the male administrators do
not have the power to control the actions or beliefs of the female practitioners.
The absence of patterns 2 and 3 (in reverse) in male-dominated religions
reflects the reality that male-dominated religions occur in cultures in which
gender inequality is consistently a matter of male dominance and female subor-
dination. Male leadership is perceived as natural, right, and complete; there is
no need to share leadership with women. In contrast, women's religions occur
in cultures in which there is at least some degree of male dominance and female
subordination. Thus female leadership is treated somewhat ambivalently, and
certain sorts of roles are reserved for men.
A somewhat unusual case is that of the Black Spiritual churches in the
United States. According to Hans Baer, "Because of the disproportionate num-
ber of female members [women may outnumber men by a ratio of as high as
43:7 women in Spiritual churches generally make concerted efforts to encour-
age males to belong, often by granting them religious offices. Spiritual women
generally are not interested in establishing matriarchal congregations, but
rather sexually egalitarian ones. In order to induce males, often their spouses,
sons, nephews, and brothers, to join, they may make certain concessions that
may eventually culminate in males attempting to assert dominance (Baer 1993,
77). Again, we would not expect to find the reverse situation in male-
dominated religions for the simple reason that in patriarchal societies men
would be unlikely to make concessions in order to include women in their
religious organizations. Women's presence would not be necessary to legiti-
mate the religion; quite to the contrary, the exclusion of women (who are, by
definition in patriarchal societies, of secondary status) can be used as a symbol
of the high status of a religious group.

Pattern 4: Assistants
Men serve as holders of priestesses' umbrellas in Okinawa. In the Afro-
Brazilian cults men drum while female mediums dance and enter trance.15 This
pattern—men as specialist assistants to women leaders—is the only one with
widespread parallels in male-dominated religions. To women who have
cleaned the altars in Catholic churches, arranged the flowers in Protestant
churches, and prepared the cakes and cookies for kiddush at Jewish synagogues,
this pattern is easily recognized. In women's religions the men who become
religious specialists are typically men of low status, and their status is not
enhanced by becoming assistants to the priestesses or shamans. Given the
generally sexist orientations of the societies in which these religions are situ-
ated, assistant to a leader in a female-dominated religion is not a high-prestige
job for men.
 Leaders and Experts 237

Pattern 5: Animal Sacrifi

The oddest pattern that I have discovered is that in the small number of
women's religions that practice animal sacrifice, men perform the actual kill-
ing. This is true in Thailand, in certain Afro-Brazilian cults, and in zdr initia-
tion rituals. For example, Boddy points out that in zdr sacrifice it is the zdr
leader's son rather than the leader herself who actually sacrifices the animal
(1989, 129). As I argued in Chapter 10, this most likely reflects the fact that in
women's religions women are so thoroughly associated with life that it is
appropriate for men to deal with death.

SOME COMPARISONS
Analysis of female religious leadership takes on meaning in the context of three
types of comparisons: (I) women religious leaders versus men religious lead-
ers, (2) women religious leaders versus women secular leaders (3) women
religious leaders in female-dominated religions versus women religious leaders
in male-dominated religions. We will now look at each of these in turn.

Women Religious Leaders versus

Male Religious Leaders
Although the differences between female religious leaders and male religious
leaders are far from absolute, several trends merit comment. Women religious
leaders, more than men, seem to straddle the public and the domestic spheres
(Sered 1991). They use their homes as ritual spaces, they help other women
deal with children and children's illnesses, and they officiate at rituals at which
cooking and food are central elements.
Richard Hutch has summed up several other differences between male and
female religious leaders. Women typically become leaders through a slow pro-
cess of self-recognition, whereas men more often report a "once and for all"
achievement during a single, identifiable stage of life. We have indeed seen that
the life histories of many women leaders involve years of illness, visions, and
struggling with the call of the spirits—a process that often begins in early
childhood and only ends in old age (see Chapter 7). Hutch notes that women
leaders are often concerned "with the internal processing of personal experi-
ence as a source of religious authority in itself" (1984, 159). For male leaders,
the source of authority is more often external—education, ordination, central-
ized hierarchies. The material we have seen strengthens Hutch's observation.
Particularly instructive is the case of the Korean shamans who use their own
life stories to explicate the ritual troubles of clients.
Another difference between male and female religious leaders has been
noted by Brita Gill. Writing about her own experiences as a Protestant minis-
ter in the United States, she refers to her "ministry of presence—a ministry
that is oriented to individuals, not just to problems" (1985, 90). This certainly
is in line with what we have described in this book. Male leaders, implies
238 Priestess, Mother, Sacred Sister
Gill, are more concerned with rules, symbols, and knowledge. The associa-
tion of women religious leaders with personal relationships rather than with
theology or organization is significant. We will return to this theme in the
next chapter.
A last way in which female religious leaders differ from male religious
leaders concerns their status in the society at large. Women who wish to
become religious leaders challenge patriarchal notions of women's place (subor-
dinate) and women's role (familial). Not surprisingly, the model labeled above
as "persecution model" is far less common for male religious leaders (except
for founders of new religions).

Women Religious Leaders versus

Women Secular Leaders
Women religious leaders resemble women secular leaders in many ways. Both
sorts of leaders tend to attribute their success to outside factors: spirits, ances-
tors, easy exams, or lack of other qualified candidates, rather than to their own
innate ability or hard-earned achievements (Powell 1988, 107). This fits well
with the socialization of women to be subordinate.
For a woman to achieve a leadership position, whether religious or secular,
she needs to be known personally. Studies of gender and management have
shown that when strangers evaluate one another they consistently see men as
dominant, but when group members know one another, gender differences
tend to disappear (Powell 1988, 109). In women's religions leaders are person-
ally known to their followers: the model of a priest who is sent or a rabbi or
minister who is invited from the outside to an unfamiliar congregation is not
one we have seen in women's religions. Female religious leaders tend to slowly
build up a clientele through word-of-mouth recognition. The example of
Christian Science practitioners illustrates this well.

Women Leaders in Female-Dominated Religions versus

Women Leaders in Male-Dominated Religions
The literature on women leaders within male-dominated religions shows both
similarities and contrasts to women leaders within female-dominated religions.
One common trait is an emphasis on healing. Illness and cure is a common
theme in the lives of women religious leaders both in female-dominated and
male-dominated religions (cf. Weinstein and Bell 1982 on Christian women
saints). Since in many cultures women are especially concerned with healing,
and healing and religion are often intertwined (see Chapter 5), this is not
surprising.
On the other hand, women leaders in male-dominated religions are, on the
whole, excluded from public leadership roles—and especially roles as ritual
officiants. The types of roles open to women in male-dominated religions tend
to be as teachers of children, assistants, nurses, and caretakers. 16 In female-
dominated religions, public ritual officiant roles are available to women leaders.
 Leaders and Experts 239

Studies show that women leaders in male-dominated religions often buy

into the anti-female ideology of that religion. Weinstein and Bell, for instance,
show that female Catholic saints "seem to have internalized the denigration of
their sex's spirituality" (1982, 227). Female-dominated religions, on the other
hand, typically celebrate or even exalt femininity (see Chapter 10).
Audrey Brown has conducted a particularly fascinating study of women and
ritual authority in [male-dominated] African-American Baptist churches of
rural Florida. In common with leaders of many female-dominated religions,
the women leaders in the Florida churches are, for the most part, elderly,
experts in fertility and childbirth, and involved in food rituals. On the other
hand, she found that "Even though women are often better educated than
males and frequently economically independent, it is their nurturing and sus-
taining qualities that are glorified in church life. Female assertive behavior or
aggressive leadership is discouraged and censured" (1988, 2). This is in dra-
matic contrast to female-dominated religions where women leaders are typi-
cally strong and articulate.
In addition, Brown found that many Baptist women leaders were aging
wives of pastors. These women often had eight or more children. This clearly
differs from leaders in many female-dominated religions, who more typically
have lost some or even all of their children (see Chapter 4). It does seem that
African-American Baptist women leaders express greater satisfaction with tra-
ditional female roles than leaders of female-dominated religions do. And
whereas the African-American women seem to have found Baptist theology
and philosophy sufficient for interpreting their life experiences (e.g., marriage
and many children), leaders of female-dominated religions do not find male-
dominated theology and philosophy helpful for making sense out of their life
experiences (e.g., child death).
In contrast to female-dominated religions where marital status is irrelevant,
"Regardless of a woman's Christian virtue, she cannot become a [Baptist] dea-
coness unless she is married to a deacon" (Brown 1988, 3). Here we note one of
the most significant differences between the two categories of women religious
leaders. Women leaders in female-dominated religions frequently inherit their
roles from their mothers, often in the form of inheritance of the deceased
mother's guardian or shamanic spirit. Women leaders in male-dominated reli-
gions often acquire their roles by virtue of their husband's status: sometimes as
the wife of a male leader and sometimes as the widow of a male leader. This
distinction is particularly important. The leader who inherits her position
from her mother symbolizes both an independent, female chain of command
and the centrality of motherhood in religious life. As I argued in Chapter 3,
this is consistent with the trend in female-dominated religious to address women
primarily through their identities as mothers. The leader who acquires her role
through her husband symbolizes the tangential—even incidental—position
of women on the male chain of command. The spouse-of-the-leader model
strengthens the bias toward addressing women through their identity as wife.
As I argued in Chapter 3, the role of "wife" tends to be less autonomous and less
empowering for women than the role of "mother."
240 Priestess, Mother, Sacred Sister

Notes
1. Spiritism is a mixture of Spiritualism, Kardecism, folk Catholicism, and tradi-
tional folk healing.
2. The closest to nonhierarchical religions are those of the Black Caribs and the
Feminist Spirituality Movement. Yet the former does seem to have a cadre of elderly
women who are in charge of the rituals, and certain branches of the latter have priest-
esses and even high priestesses. It is of interest to note Karen Sacks's findings that
among the North American Cheyenne, men's hunting societies were not hierarchical,
while women's quilling and rawhide painting societies were hierarchical (1979, 80).
3. In the more traditional (more African) cults, communication with the supernatu-
ral is often through divination, which only the chief priestess knows how to do. The
chief priestess divines by throwing cowrie shells into stated patterns that allow her to
learn the will of the gods. Although the manner of interpretation of the cowrie shells is
highly traditional, individual priestesses may prescribe different remedies.
4. The latter two roles were traditional; the first two were imposed by the govern-
ment.
5. I thank Teigo Yoshida for helping me clarify the rather confusing terminology
regarding contemporary priestesses (personal communication 1992).
6. A preliminary question here is which women in female-dominated religions can
be considered leaders. In some religions the answer is clear: an explicit hierarchy exists
and women at the top are the leaders (Shakers, Sande). A much thornier situation arises
in the religions involving spirit possession. I would consider the Spiritualist medium to
be a leader in the sense of a specialist with extraordinary skills and talents, although, as
we see in this chapter, her status is not high and she does not organize a chapter or
church. I would not consider women who dance at zar ceremonies to be leaders,
because in many parts of Africa most women are initiated into the zar cult. For the zar,
the leaders are women who are addressed by others as "leader" (shaykha). The most
difficult decision concerns the Afro-Brazilian cults. In these cults there are certain
women who are priestesses or in charge of cult centers. They are clearly leaders. Yet I
am inclined to also treat the mediums as leaders, albeit on a lower level, both because
(unlike in the zar cult) most people do not become mediums, and because mediums are
consulted for healing advice.
7. Western scholars claim that the difference between priestesses and yuta in the
Ryukyu Islands is that the former concentrate on public rituals and the latter on private
rituals. It appears, however, that there may be far more overlap between Ryukyuan
priestesses and shamans than meets the casual eye. For example, all Ryukyuans belong
to kin-groups, each of which has a woman kaminchu who attains office by virtue of
being born on a high spiritual level. Her identity is detected by the older women of the
kin-group and confirmed by a shaman. On the Ryukyus it is quite common for a
woman to begin as an ukudi or family priestess and then also become a yuta. The
priestess and shaman systems on Okinawa tend to interpenetrate in the person of the
kin-group priestess, who frequently is also a shaman.
Hisako Kamata convincingly argues that the distinction between yuta and noro may
be a late development. She notes that the practice of closing the eyes and shaking the
head up and down is common to both yuta and noro. Even of greater interest is her
discovery that because divine possession is evidence of eligibility for the noro role, it
sometimes happens that the same women are priestesses presiding over village rituals
and festivals in one village, yet are regarded as yittas in other villages. Evidence is also
cited that suggests that in the past both men and women were yutas, and that the status
 Leaders and Experts 241

of yutas was much higher than at present. They worked for the government as custodi-
ans of oral history and possessors of knowledge (1966, 62).
Similarly, Yoshinobu Ota has studied the southern Ryukyu Islands and found that
certain women experience series of traumatic illnesses that lead them to consult a
shaman. The shaman may interpret the illness as a sign that the woman was born with
"high shiji"—a sort of enhanced spirituality. "This shaman will tell her that, in order to
put an end to the recurring sickness, she must become either a priestess or else a
shaman" (1989, 115). Ota relates life stories of three priestesses and one shaman, and all
the stories are rather similar in terms of illness and initial resistance to the calling.
8. Skultans (1983, 18), on the other hand, suggests that the connection between the
death of one's father and mediumship may well have been financial—young women
deprived of the economic support of their fathers were more in need of pursuing paying
careers. Few other paying careers were open to women in the mid-nineteenth century.
9. It then became customary for a noro to be succeeded by her brother's daughter. In
more recent times the position is succeeded by the son's wife (Shimabukuro Gen'ichiro
quoted in Kamata 1966, 59).
10. For example, Korean shamans explain that their own personal deprivations as
women allow them to identify with their clients—clients' problems are similar to ones
they themselves have already worked through (see Wilson 1983).
11. I. M. Lewis interprets the high involvement of older and infertile women in
possession cults in terms of their cultural status of "half-men" (since, non-fertile, they
are no longer fully female). According to Lewis, this is what underlies the androgynous
character frequently attributed to the leaders of peripheral cults (1971, 95). While I agree
with most of Lewis's ideas, in this case I am far from convinced by his interpretation. It
seems to me more likely that women without small children are able to become leaders
because they have greater control over their time. I am also not convinced that women
religious leaders are androgynous in any very deep sense. Rather, because in most
cultures leadership is defined as a male trait, when a woman becomes a leader she looks
as if she has adopted a male role. However, the data that I present in this book demon-
strate that women religious leaders do not copy male modes of leadership, male identi-
ties, or male character traits.
12. Humez argues that even among Lee's followers gender was a problematic issue.
"Shaker manuscript records for the first thirty years of the nineteenth century abound in
evidence of continuing male rebellion against female headship" (1992, 89).
13. "It was apparently this state of affairs that led a Chinese traveler in 1683 to assume
that there were no courts of law, since he observed no male judges" (Lebra 1966, 102).
14. Catholic nuns do in a sense abandon their femaleness, but they do not ac-
quire male gender roles. Among the Bimin-Koskusmin of New Guinea, some older
women ritually become men (or really androgynous) in order to be ritual leaders
(Poole 1981).
15. According to Esther Pressel, the few male Umbanda mediums tend to be young,
and as they age either drop out or become center directors looking after the financial
affairs of the center. In some newer Afro-Brazilian groups there are men who are in
charge ("fathers") who have a very low status both socially and religiously and who do
not command the respect that the "mothers" command. In centers run by a man,
women still retain important functions. It is telling that criticism about "fathers" is
always more venomous than about "mothers" and priests are often accused of being
insincere, dishonest, or evil (see Carneiro 1940).
16. In certain male-dominated religions there are other roles—for example, ascetic,
temple dancer—that are sometimes available to small numbers of exceptional women.
 12
Women, Sacred Texts,
and Religious Organization

UNCENTRALIZED AND
UNSTANDARDIZED
None of the women's religions is nonhierarchical, and, as we have already
seen, many have well-articulated systems of internal ranking. On the other
hand, very few of the religions are organized around a central authority or even
a roof organization that incorporates all chapters, temples, branches, or
groups. Related to the absence of centralized authority, there is frequently an
absence of obligatory rituals, sacred scripture, and uniform dogma. I will
argue that interest in preserving and enforcing compliance to written doctrines
is linked to a political agenda (missionizing) and a theological vision (transcen-
dence), neither of which is meaningful to women's religions, and both of
which are often associated with patriarchal rule.
Women's religions, on the whole, do not emphasize members' exclusive com-
pliance to one specific set of doctrines. A number of the religions dealt with in
this book are one alternative within a culture in which it is common for individu-
als to participate in the rituals of, and believe the theologies of, more than one reli-
gion (Korean shamanism, Burmese nat religion, Spiritualism). In addition, many
of these religions are internally eclectic, easily and consciously absorbing new
ideas or deities (Feminist Spirituality, Spiritualism, Afro-Brazilian religions).
A case in point is North American Spiritualism. The Fox sisters, who are
commonly referred to as the "founders" of Spiritualism, were not particularly
involved in organizing the movement, nor were there other people who orga-
nized on their behalf or in their name. Nineteenth-century Spiritualism spread
like wildfire with no institutional backing. Ironically, it seems likely that the
efforts of journalists and Christian clergy to persuade the public not to partici-
pate in Spiritualist rituals actually served to publicize Spiritualism and entice
people to try their hand at contacting the dead in many communities where there
were no ongoing Spiritualist groups or high-powered mediums. "Not only did
244 Priestess, Mother, Sacred Sister

the [Spiritualist] movement have no orthodox doctrine; it had no membership

because it had nothing for adherents to belong to. It had no official leadership
because it had no offices for leaders to hold. Mediums received no training and
no ordination" (Braude 1989, 8). The one attempt at national organization
during the 18505—the Society for the Diffusion of Spiritual Knowledge—did
not succeed. According to Alex Owen (1981), Spiritualists were noted for their
commitment to individualism and anti-authoritarianism. They were never inter-
ested in or successful at establishing central organizations.1
Scholars assert that Spiritualism never developed a genuine theological creed
(cf. Macklin 1974). Spiritualists back away from official dogmas and compul-
sory laws; they stress the individual conscience and individual experience, and
try to avoid authority over member's beliefs and behavior. Spiritualist theol-
ogy is vague, nondogmatic, and constantly revealed and developing. The
spirits themselves profess varying beliefs, and teach varying doctrines to differ-
ent individuals and groups.
Within Spiritualist groups it has been common to add to resolutions a proviso
that the resolution itself is not binding on anyone who does not like it. "Lacking
coherence, most gatherings that claimed to explore the truths of spiritualism
simply made diversity a virtue" (Moore 1977, 14). The history of Spiritualism
underscores the connection among respect for the needs of the individual, ab-
sence of official doctrine, and lack of central authority or organization.
Afro-Brazilian cults are also characterized by a lack of centralized structures,
and attempts to unify the movement have been weak and unsuccessful. It is of
interest that men predominate in the external leadership positions within the
various federations governing Umbanda activities in the cities and states. Ac-
cording to Patricia Lerch (1982), Umbanda's flexible organizational style al-
lows local cult leaders and spirit mediums to adapt rituals to the specific needs
of their cult followers. Among Umbanda leaders, there is frequently animosity
and even attacks on each others' doctrines and rituals. 2 The real functioning
units of structure in Umbanda are the individual ritual centers or temples.
Again, we note that the absence of centralized organizational structures is
associated with lack of mandatory rituals and beliefs. According to Diana
Brown, "The ambiguities and complexities of Umbanda's cosmology, which
form part of the ongoing dialectical process of its development, give it enor-
mous flexibility of interpretation and permit different sectors with different
interests to identify with it in different ways" (1986, 76).
The Afro-Brazilian cults are nonliterate religious systems (even if particular
members and leaders know how to read), and each priestess constructs her ritual
repertoire according to her own experiences, ideas, and inspirations. "The
[Afro-Brazilian] Batuque today is far from being a completely systematized
body of beliefs, and, at the rate it is still changing, this condition seems likely to
continue. Batuque members seem rarely to be troubled by the lack of a highly
rationalized theology, since their interest is usually pragmatic rather than philo-
sophical. The lack of philosophical coherence may limit the Batuque's appeal to
individuals with intellectual pretensions, but of course the very looseness and
 Women, Sacred Texts, and Religious Organization 245
lack of rigid systematization of its theology provide one explanation for the ease
with which new themes are absorbed" (Leacock and Leacock 1972, 320).
We will now look briefly at a few other examples that illustrate the lack of
centralization which characterizes women's religions. In Northern Thai ma-
trilineal spirit cults, fission is common when descent groups reach a large size
(ten or more households). The new group has its own spirit, shrine, officiant,
and its members are no longer "of the same spirit" as the old group. To ensure
this separation, the new group holds its annual sacrifice on the same day as the
original group, thus preventing double attendance. Parenthetically, we may
add that by preventing double attendance, any kind of larger organizational
system is avoided. "There are no nesting, hierarchical, or segmentary struc-
tures, nor any superordinate genealogical structures" (Turton 1976, 221). Let
us recall that even the descriptions of the spirits in this religion are vague and
contradictory (see Chapter 8).
Among the Black Caribs no one tries to compel conformity in beliefs about
the dead. There is no formally organized priestly hierarchy, nor an officially
defined dogma. Although ritual obligations are enforced, there is a great deal
of variability in the manner of performance of rituals for the dead. Details vary
according to the wishes of the ancestor who requested the ceremony and the
resources of the sponsors. Dream experiences, for instance, are highly valued
by the Black Caribs as a source of prophetic knowledge and a means of commu-
nicating with the dead. Inspiration for religious songs and all sorts of esoteric
and non-esoteric knowledge comes from ancestral spirits through dreams
(Coelho 1955, 139-140).
In a similar vein, zar ceremonies vary considerably according to the idiosyn-
crasies of the leader and the type of illness being treated. There is no central
organization of any kind (Kennedy 1978). According to one observer, zar
adherents and leaders emphasize that there are a variety of zar traditions,
beliefs, and rituals (Kenyon 1991, 101).
This sort of self-conscious non-institutionalization also characterizes Korean
women's religion. Whereas Confucianist ritual manuals have enabled male
ancestor worship to become quite standardized 3 , Korean women's rituals and
shaman lore are learned through observation and oral transmission, and varia-
tion occurs from place to place. Korean shamans are initiated by and trained by
more experienced shamans, but there is no official (or even unofficial) roof
organization of shamans.
Each Sande chapter is autonomous, there is no centralized Sande organiza-
tion, and cooperation among chapter leaders is informal. (Because local offi-
cials participate in rituals in other chapters, there is some informal integration.)
The noncentralized organization of Sande is even more interesting when we
learn that one of the most striking differences between Sande and Poro (the
men's secret society parallel to Sande in almost all ways) lies in structure.
According to Donald Cosentino, the structure of Sande is apt to be more
pluralistic than that of Poro. A medium-sized town is likely to have several
Sande chiefs, each conducting her own initiation classes, often at different
246 Priestess, Mother, Sacred Sister
times of the year. Poro structure, on the other hand, is far more centralized
(1982, 23). This example is particularly important because it defuses the poten-
tial counter-argument that women's religions are noncentralized because so
many of them are situated in societies in which all religion is noncentralized. If
we review the examples I have presented, we see that a number of the noncen-
tralized women's religions are situated in societies in which men's religions are
highly centralized. The clearest examples are the women's religions that co-
exist with Catholicism or Buddhism.
Skipping both geographically and chronologically, we discover that "The
beguines, the only [Christian] movement created by women for women before
the modern period . . . [were characterized by] lack of leaders, rules, detailed
prescriptions for the routine of the day or for self-regulation, lack of any over-
arching governmental structures" (Bynum 1984, 116). Please note the contrast
to many other medieval monastic movements, with their detailed codes of
behavior, vows, and organizational hierarchy.
Among Spiritual Feminists there is no central organization; small, even inti-
mate, circles are the basic units of the Movement. The following remark by a
Spiritual Feminist writer well sums up what I have demonstrated thus far: "In
order to keep our theology supple and responsive to individual differences and
changes, we must remember that all theology—all thinking about deities and
godly powers—is done by individual people in particular situations. Human
beings of different sexes, ages, races and environments have different experi-
ences of spirituality and religious phenomena" (Goldenberg 1979, 115). Among
Spiritual Feminists the organizational structure is, to say the least, loose. Accord-
ing to Naomi Goldenberg (1979, 140), "[P]eople probably do not have to en-
force a standardized set of religious images on everyone in order to feel a sense of
community." This preference for noncentralization is also true, for the most
part, of the secular feminist movement. As Patricia Martin explains, feminist
organizations tend to be local, and many discussions of feminist organizations
suggest that only small organizations can be truly feminist. Face-to-face interper-
sonal exchange—one of the key ideological components of feminism—is more
possible in organizations with fewer members (1990, 199).
In Chapter 6 I demonstrated that women's religions tend to be characterized
by an accent on ritual rather than official doctrine. So far in this chapter I have
shown that even the ritual complexes of women's religions are nonstan-
dardized and idiosyncratic. The pivotal role of spirit possession in women's
religions further contributes to the focus on individual experience instead of
organization. I am impressed with the countless descriptions of spirit posses-
sion in which the observer writes that although the possessed woman (or man)
enacts a standardized character, the individual injects a bit of her or himself to
the role. For example, in the less traditionally African of the Afro-Brazilian
religions, each medium adds personal touches (behaviors, bits of costumes,
etc.) to the spirit or deity by whom she or he is possessed. Similarly, although
each zar spirit has its own attributes (its own beat on the drum and sometimes
its own costume), each host brings something unique to her performance, "For
a spirit reveals different aspects of itself in various women" (Boddy 1989, 303).
 Women, Sacred Texts, and Religious Organization 247

Case Study. The Shakers

We will look at the Shakers at some length because they present a more com-
plex case than the previous examples. Shaker history—which is far better
documented than the histories of any of the other women's religions—exhibits
a certain tension between centralized authority and local autonomy. Unlike the
other examples I have cited, the Shakers did aspire to a centralized organiza-
tional structure; the community at New Lebanon had the status of "Mother
Church;" and a great deal of time, energy, and resources were dedicated to
travel and communication between Shaker villages (Stein 1992). Moreover,
early Shakers (although not later Shakers) did not demonstrate the tolerance
toward other beliefs and groups that characterizes most other women's reli-
gions. For example, Ann Lee's incarceration in Manchester followed her ada-
mant condemnation of Anglicanism and her attempts to disrupt Anglican
services. Two other points should be noted: During their early American pe-
riod Shakers engaged in a great deal of missionizing, and Shakers have gener-
ated enormous amounts of written material.
Internal standardization of belief has not, however, always been a priority
for the Shakers. Clarke Garrett has commented on one of the most important
differences between the Shakers and earlier ecstatic Christian groups. "Other
groups . . . were as dedicated to following the guidance of the Spirit as the
Shakers were, but all of them . . . were constrained by a tacit biblicism that
not even their prophets and visionaries overstepped. . . . The Shakers, on the
other hand, had from the outset believed that the messages sent by the Holy
Spirit could be in any form and could entirely supersede anything that Scrip-
ture had commanded for earlier generations. The fact that their techniques of
inducing ecstacy in their worship services affected everyone present meant that
everyone present had access to the prophetic 'gifts' that the Lord rained upon
them" (1987, 195-196).
With the death of Ann Lee the Shakers experienced a shift from mixed
female and male leadership (Ann Lee, William Lee, and James Whittaker) to
male leadership—James Whittaker became her successor, together with a new
emphasis on legalism and organization and the construction of a formal meet-
inghouse.4 Writing about Whittaker's leadership, Garrett explains that "While
he [Whittaker] accepted the power of the spirit that operated in the night-long
meetings and fully believed that Ann Lee was divinely inspired, he was also
concerned with creating a framework of order and discipline in which the
Spirit would operate" (1987, 211). As Marjorie Procter-Smith shows, this had
portentous implications for women. "Early Shaker worship was a matter of
intimate and spontaneous contact with the Divine, manifested through various
signs such as dancing and jumping. Shaker worship gradually became more
institutionalized. In the beginning there was no preaching, sermon, little read-
ing, no public prayer—just the Holy Spirit. Then, a sermon—only given by
men—was added. Then a tune sung in unison was added. Then dancing in
unison. Then a hymnbook. All this happened together with increasingly strict
sexual segregation during worship. . . . With the establishment of ordered
248 Priestess, Mother, Sacred Sister
worship came a corresponding reduction of women's public roles in Shaker
worship" (1985, 139, my emphases). 5
After Whittaker's death, Joseph Meacham took over Shaker leadership. Ac-
cording to Stephen Marini, "Consciously designed Shaker institutions were
the creation of Joseph Meacham" (1982, 127). Under Whittaker's rule the
Shakers were organized into an elaborate and all-inclusive hierarchy of mem-
bers based on spiritual merit. "Father Joseph elaborated Mother Ann's practices
into formal norms" (Marini 1982, 129). And Stephen Stein argues that
"Meacham played the principal role in formulating the rules and regulations
that governed membership in the society and the activities of the gathered
Shakers" (1992, 45). In short, the evidence seems to suggest that standardiza-
tion of Shaker ritual was implemented by men, to the detriment of women. 6
Still, the Shakers never developed the strict uniformity of worship familiar
to most Christians and Jews. The first collection of Shaker hymns, Millennial
Praises, published in 1813, included in the Preface a note of warning that these
songs should not be regarded as eternally useful, "for no gift or order of God
can be binding on Believers for a longer term of time than it can be profitable
to their travel in the gospel" (quoted in Stein 1992, 104).
Lucy Wright, who succeeded Meacham to Shaker leadership, "stubbornly
refused" to gather "scattered rules into one set of regulations" (Stein 1992, 95).
(This does not mean, however, that Wright was averse to issuing rules. Accord-
ing to Priscilla Brewer, Wright "continually issued new and revised directives,
even from her death bed," personal communication 1992). Less than six
months following her death in 1821, the new central ministry (which consisted
of two men and two women) began circulating manuscript copies of the
"Orders and Rules of the Church at New Lebanon." In the years following
Wright's death, males increasingly dominated the Shaker central ministry
(Stein 1992, 133).'
Yet even with the publishing of hymn books and rules, "Shakerism did not
produce mindless conformity. Believers were able to establish a healthy indi-
vidualism within the society" (Stein 1992, 149). Shakers were among those
who embraced Spiritualism in the mid and late nineteenth century, and Shaker
spiritualists "welcomed insight wherever it was found, whether in Eastern
religious traditions, Christian Science, or harmonial philosophies" (Stein 1992,
323). Significantly, it seems that the period in which Shakers most clearly
embraced eclectic beliefs coincides with the period in which "women were
increasingly prominent in all aspects of the Society's life and leadership" (Ste-
phen Stein, personal communication 1993).
Am I arguing that the greatest thrusts toward institutionalization in Shaker
history occurred during the relatively brief periods in which men were the
leaders? I am reluctant to make such a bold statement. The fact is that in the
history of new religious movements it is common to find that the first genera-
tion of leaders makes use of personal charisma, leaving it to the second genera-
tion to standardize and codify (see Weber's 1966 [1922] discussion of charisma).
Thus, one could argue that Whittaker's and Meacham's sex is irrelevant; rather,
 Women, Sacred Texts, and Religious Organization 249
their place in Shaker chronology is what prompted them to work toward institu-
tionalization of worship, beliefs, and interpersonal relationships.
On the other hand, I see it as extremely significant that although Lucy Wright
sanctioned the collection of Shaker laws, she did not sanction their publica-
tion. While this difference may sound trivial, I insist that the importance of
written codes should not be underestimated. Once laws are written down, they
take on a life of their own; they become eternal, immutable, universal, and
(often) more consequential than the individuals who live by and with them.
Moreover, as we saw earlier (page 248), regarding the nineteenth century when
the Shakers became much more of a truly female-dominated religion than they
had been in earlier periods, historians have noted both a deceleration of the
institutionalizing process and a new openness to other religious beliefs.

Men and Centralization in Women's Religions

The historical record shows over and over again situations in which the impe-
tus for centralization and standardization in women's religions comes from the
(often small number of) men. This was true for North American Spiritualism.
It was also true in the Black Spiritual churches of New Orleans, where in the
early days each congregation—almost always headed by a woman—was au-
tonomous. In the early days Black Spiritual religion had been almost exclu-
sively a woman's religion under women's leadership. When male leaders
emerged in the 1930s, they (the men) formed national organizations in which
men dominated. Today women still outnumber men at the level of local minis-
ters, but the national organization is headed by a male archbishop and men
outnumber women on the Executive Board (Jacobs and Kaslow 1991, 184).
Jacobs and Kaslow attribute men's rise to organizational power in terms of
their literacy at a time when formal education was unavailable to most women
(1991, 185). According to Hans Baer, "It appears that men have headed the
largest and most institutionalized associations in the Spiritual movement. . . .
My ethnographic data suggest that women tend to serve as the heads of rela-
tively small Spiritual associations, consisting of several congregations or inde-
pendent congregations" (1993, 79).
Turning to an example we have not yet met, in the Harrist Movement in
Ghana, "From the beginning, it was Nackabah [the male leader], rather than
Madame Tani [the female leader], who was fascinated with the modes of
organization, symbols of material progress, and activities of teaching and
preaching" (Breidenbach 1979, 111). Moreover, Paul Breidenbach notes that
"Literacy is essential to the preaching master, while, in fact, I did not encounter
a single healing prophetess who was literate" (1979, 111).We will return to the
issue of texts and gender later on.
Two women's religions do not fit neatly into the pattern of noncentralized
authority: Okinawan religion and Christian Science.8 Until the Japanese an-
nexation in 1879, the entire Okinawan kingdom was organized for religious as
well as political ends, with an official, centralized hierarchy of hereditary priest-
250 Priestess, Mother, Sacred Sister

esses (noro) in the villages, districts, and regions, and the chief priestess who
was a close relative of the ruler and virtually his equal in rank.
This centralized hierarchy began with the unification of the Okinawan king-
dom in the fourteenth century. Until that time, the religion was kin based and
noncentralized. Based on the available historical sources, I speculate that the
noro hierarchy was established as a means for implementing political centraliza-
tion of the Okinawan kingdom. Evidence shows that the move toward central-
ization did not come from the priestesses but from the male, secular leaders.
It is suggestive that despite centuries of official centralization, there continues
to be a great deal of variation regarding the exact order to the principle rituals.
Even within each island—often from house to house—there is diversity in
terms of the precise enactment of the ritual cycle. Certain ceremonies only are
observed in particular villages, and it is rare that the annual cycles of two
villages completely coincide. Each village is believed by Okinawans to be
unique, with its own god and customs. And Garland Hopkins (1951) has noted
that Okinawans easily appropriate parts of new religions and rituals that meet
their fancy.

SACRED WRITINGS
An Exception
The other centralized women's religion is Christian Science. Christian Science
churches around the world belong to an organization that is controlled by the
Mother Church in Boston, Massachusetts. Mary Baker Eddy founded the
Mother Church so that Christian Science would be centralized, and she herself
presided over the transformation from a charismatic into a bureaucratic
church. Unlike most other women's religions, Christian Science strongly dis-
courages experimentation with, participation in, or even study of other reli-
gions and healing paths. Christian Science is the final revelation and that is all
people need to know.
Christian Science's centralized organizational structure is associated with an
emphasis on sacred writings. The Mother Church publishes selected texts from
Mary Baker Eddy's Science and Health and from the Bible, and these selections
are read at all local churches on specific Sundays. 9 Church members are ex-
pected to study these texts during the preceding week—a set part of the text
each day. Thus, not only do all Christian Scientists hear the same texts at the
same time, but there is an expectation that individual members will devote
time to developing a high-level understanding of Christian Science beliefs.
Christian Science healing is often effected through study of Science and Health,
and Christian Science Reading Rooms are located in many cities and towns.
As we said earlier, Mary Baker Eddy herself organized Christian Science
around the central Mother Church. On the other hand, the current administra-
tive leadership of Christian Science is almost all male. Margery Fox shows that
while women predominate in the therapeutic roles of practitioner and nurse,
men predominate as both administrators and lecturers (1989, 100). We may
speculate that many of the differences between Christian Science and other
 Women, Sacred Texts, and Religious Organization 251
women's religions stem from the fact that in Christian Science, unlike in most
other women's religions, men are in charge of both administration and formal
teaching.

Aversion to Sacred Texts

With the exception of Christian Science, women's religions are characterized
both by an aversion to centralization and a lack of authoritative sacred texts.
Let us bear in mind that despite the lack of emphasis on texts, almost all these
religions exist in societies in which many (or most) people are literate. North-
ern Thai matrilineal cults, for example, have no written records of any kind,
although literacy and papermaking technologies have been known in Thailand
for many generations.
Similarly, Spiritualism arose in a highly literate culture, yet Spiritualists have
no sacred scripture. Religious beliefs and practices are learned from more devel-
oped Spiritualist mediums. "In fact, there is a stigma attached to learning about
spiritualism, especially mediumship and healing, from published material"
(Skultans 1974, 3). In light of our discussion of how women's religions tend to
reinterpret rather than challenge prevalent notions of gender (Chapter 10), it is
of interest to point out that among nineteenth-century Spiritualists lack of educa-
tion was treated as a "female" characteristic that made women especially suitable
for mediumship. While it was certainly the case that formal education was
unavailable to many if not most nineteenth-century women, it was also the case
that women's ignorance was turned into a virtue by Spiritualists. An "empty
slate" was considered the finest instrument for spirit communication.
The Ryukyu Islands provide a particularly interesting case study here. Wil-
liam Lebra (1966) notes that small pieces of paper bearing written characters
obtained from the male Buddhist priest and ashes from the hearth of the chief
priestess are used as charms. Put differently, male Buddhist ritual specialists
ward off misfortune through sacred writing; female indigenous ritual special-
ists ward off misfortune through cooking accouterments. 10
Feminist Spirituality, on the other hand, is a highly literate religious move-
ment. Each year dozens of new books—both novels and nonfiction—are pub-
lished by feminist and mainstream presses. For many women who identify
with Spiritual Feminism yet live in small towns where there is no Spiritual
Feminist group, books and magazine articles serve as the sole means of partici-
pation in the Movement. Thus it is especially significant that there is no official
sacred scripture or a required litany of rituals that must be performed in a
standardized fashion. The prolificacy of Spiritual Feminism is not part of a
campaign to create uniformity of belief.
As we already have seen, in their early days the Shakers did not have any
printed prayers, songs, or regulations, because they believed that a static tradi-
tion would prevent the continuing inspiration of the Spirit. Shakers felt them-
selves to be in the process of developing important new doctrines—even their
theology was treated by them as inherently changeable. They believed that
during their worship services people received direct messages from the spirit
252 Priestess, Mother, Sacred Sister

world. The Bible was not their only source of revelation, and Biblical texts
were not used to prove their doctrines. Shakerism is centered on religious
experience and not doctrine or creed, and it is of theological significance to
Shakers that founder Ann Lee was illiterate. According to Sally Kitch, "Lee's
illiteracy was sometimes seen as a possible link to Jesus; like her, Jesus left no
written expressions of his message. To some believers, Lee's illiteracy only
enhanced the legitimacy of her role: an illiterate who cannot study religion
must receive her faith directly from God" (1979, 78). On the other hand,
Shakers have produced a great deal of writings about their beliefs, and records
of their lives, poems, testimonies, journals, letters, and tracts.

Literacy and Orality: Theoretical Perspectives

Anthropologists and sociologists have argued that literacy actually affects how
people think. W. J. Ong, for example, has written that "More than any other
single invention, writing has transformed human consciousness" (1982, 78).
Jack Goody specifically connects writing to the transition from concrete (par-
ticular?) to abstract thinking (1977). These ideas are certainly consistent with
my findings in Chapter 7 regarding particularism and morality in women's
religions. Yet I am tempted to phrase the process somewhat differently from
Goody's formulation: Women's religions, characterized by "concrete" think-
ing (Goody's word), choose to avoid literacy. Given that most of these reli-
gions are situated in societies in which literacy is known, I suggest that the lack
of emphasis on sacred texts is at least to some extent volitional, and not only
shapes but also reflects a variety of aspects of these religions. I concur with
Ruth Finnegan that one cannot assume the uses of literacy to be the same in all
situations, "For printing can be—and has been—used for enlightenment and
for mystification; for self-expression or rebellion and for repression; for system-
atic analysis and the development of knowledge on the one hand and for
obfuscation, dogma and the propagation of prejudice and intolerance on the
other" (1988, 163; see also Schousboe and Larsen 1989). Finnegan's important
insight allows us to give even greater weight to the avoidance of sacred texts
(as opposed to optional books and literature) in most women's religions.
Shmuel Eisenstadt has shown that highly articulate theological traditions
generally develop within literate cultures that are interested in producing and
preserving sacred texts, commentaries, and legal and philosophical treatises.
Catholicism and Judaism exemplify this trend: in both religious traditions elite
[male] groups produced complex, written theologies (1982). Literate, central-
ized, standardized, transcendent religious cultures are often the religious
cultures that are the most male dominated. The history of Islamization in
Africa during the past century, for example, shows that in many instances
growing emphasis on religious literacy coincides with the lowering of the
status and participation of women in religious life (see also McGilvray 1988 on
Sri Lanka).
There is indeed a logical correlation between sacred texts and centralization—
they are complementary techniques for ensuring uniformity of religious belief
 Women, Sacred Texts, and Religious Organization 253
and practice. As I have argued in Chapter 5, such uniformity is anathema to
women's religions that use diverse means to address and alleviate problems of
real, immediate, and emotionally charged suffering in this world. A pragmatic,
particularistic, and eclectic approach to human misfortune is inconsistent
with either a central authority or a sacred text—because central authorities
and sacred texts typically declare that "only our way is legitimate" and thus
work to enforce conformity. Jack Goody has made a similar point in his discus-
sion of the effects of literacy: According to Goody, literacy aids the development
of large-scale, impersonal organizations and states with universalistic rather
than local (particular?) outlooks. Orality, on the other hand, according to Ruth
Finnegan, is more linked to the context in which it takes place, and thus tends to
encourage pragmatic face to face interaction (1988, 165).

RANK VERSUS CENTRALIZATION

At first glance, it would seem that rank and centralization should go hand in
hand because both reflect status-conscious, classificatory, and dichotomous
thinking (Young 1983). Yet in women's religions we have discovered highly
articulated internal systems of rank co-existing with an almost total lack of
centralized authorities.
I suspect that this finding can be explained in terms of what we have already
learned about women's religions. On the one hand, these religions are charac-
terized by a this-worldly orientation and concern with interpersonal relation-
ships. It is consistent with both these characteristics that human organization
in the religion is elaborate and consequential. On the other hand, we have seen
that women's religions are characterized by an absence of transcendent mono-
theism and a de-emphasis of sacred texts. It is consistent with these characteris-
tics that suprahuman organization is unelaborated. In women's religions, the
nearby, the known, the particular, and the personal are the foci of interest.
It is striking that in the same religions in which no one is quite sure which
deity does exactly what or how, everyone is quite sure which human individual
does exactly what and to whom. A similar contrast can be made between
internal rank and centralized authority. Internal rank has to do with relation-
ships among specific and known people; organizational centralization is the
profane parallel to monotheism—a declaration that the "transcendent" is some-
how better than the particular, a declaration that only one way is correct and
everybody must adhere to it. I am impressed by William Lebra's observation
that although Ryukyuan deities are not highly differentiated, among people in
the religious organization there is a precise specification of roles "down to the
level of a horse holder or a ladler" (1966, 205). Lebra comments that "This
rigid differentiation of roles surely reflects a basic concern with human relation-
ships" (1966, 205).
Feminist sociologists and psychologists have devoted a great deal of thinking
to questions of organization, gender, and power, yet the currently available
data are far from clear-cut. In her classic book on men and women in organiza-
tions, Rosabeth Kanter cites studies showing that women do not handle power
254 Priestess, Mother, Sacred Sister

differently from men: men are not more competitive, dominant, or instrumen-
tal, and women are not more naturally cooperative. According to Kanter, there
is no evidence of sex difference in leadership style (1977, 199). What differs are
men's and women's opportunities for access to leadership, not their inherent
leadership techniques.11
On the other hand, Stamm and Ryff (1984) have looked at women's exercise
of power in a variety of cultural situations, and concluded that women's power
operates differently from men's. It is more often personal than positional, it
tends to be situationally oriented, and is frequently exercised outside the tradi-
tional authority structures of a society.12
The contrast between Kanter's and Stamm and Ryff's conclusions are dis-
turbing to those of us who like neat answers. Yet I think that both studies shed
light on the organization of women's religions. On the one hand, as Kanter
would expect, they are hierarchical—they have leaders whose styles do not
seem to differ appreciably from men's (see Chapter II). On the other hand, as
Stamm and Ryff would suppose, they consistently display an aversion to
centralization—there is a leaning toward personal rather than positional
power, leadership is highly situational, and the religions generally exist outside
the dominant power structures of the society.
A common pattern in the history of religions is a new religion or religious sect
inspired by a charismatic woman, which then becomes institutionalized through
the efforts of men (Hutch 1984). This pattern has been documented in Africa,
Japan, and the United States (Jules-Rosette 1979; Hardacre 1986; Ruether 1974).
The literature on these religions generally claims that the transition in leadership
occurred as the religion became institutionalized. Yet the definition of institution-
alization is typically vague, encompassing both the development of internal
rank and the development of centralization. A closer look at the evidence proves
that women's leadership often persists long past the point at which hierarchy
emerges. Where women's power declines seems to be precisely at the point in
which individual chapters or groups form into a centralized association.
To explain why that is the case, I would like to propose two lines of reason-
ing. To begin with, as I have described throughout this book, women's reli-
gions do not occur in feminist Utopias; they occur in societies which are, to
some extent, patriarchal. Thus, I would argue that the lack of centralization is
due to the fact that in male-dominated societies women's access to the public
domain is restricted. Leadership of a small group is far less problematic than
leadership of a nationwide or international corporation. We may conjecture
that noncentralized organizations are the only conditions under which women
can get and keep power in a patriarchal society. Once a religious movement
moves toward centralization, men take over.
The second line of reasoning I wish to propose starts from the opposite
direction. It may be that women's religions are noncentralized not because
centralized organizations exclude women from leadership, but because women
themselves have no need or desire for a centralized organizational structure. If
we look at male-dominated religions, we can ask what function a central
organization serves. The answer is clear: Central organizations are vehicles for
 Women, Sacred Texts, and Religious Organization 255
power and conquest, for bringing in new members, and for ensuring that old
members do not leave. Yet, as I showed earlier in this chapter, women's
religions tend not to preach exclusivist theologies, and they are not missionary
religions.13 Eisenstadt (1982) and others have demonstrated that the centraliza-
tion of religious functionaries seems to be connected to the rise of political
centralization—the development of the state. In many cases, including that of
the Ryukyu Islands during the Okinawan kingdom, religious centralization
was actually implemented by political authorities in order to facilitate govern-
mental centralization. There is convincing evidence that the rise of states is a
process in which women generally have very little part. The development of a
state-level political system (whether an indigenous monarchy or through for-
eign imperialism) has been shown to have a detrimental affect upon women's
religious status (cf. Frymer-Kensky 1992 on the ancient Near East; Silverblatt
1980 on the Andes).
To my mind, what we are dealing with here are complex interactions of
theology and organization. On the one hand, the rather flexible theologies of
women's religions follow and justify their noncentralized organizational struc-
tures. On the other hand, nondogmatic, supple theologies do not demand a
centralized organization to spread or enforce them. Codification can be seen as
a response to the need for maintaining power and domination—and this may
be where an important difference between men and women lies. Whether that
difference is rooted in maternal thinking (as Sara Ruddick claims) or in social
structure (as Rosabeth Kanter claims), I leave up to the intellectual inclinations
of the reader.
Issues of organization have important implications for the ways in which
female-dominated religions do or do not serve women's interests. Lack of
centralization decreases the potential impact these religions can have on the
society as a whole, both because each chapter preaches a slightly different set of
beliefs and because adherents to the religion cannot easily be mobilized to work
together for a unified cause. In the next chapter we will turn to the secular
consequences of women's religion.

Notes
1. As we saw in the previous chapter, there were unofficial leaders. And Spiritual-
ists did attempt numerous times to build enduring national organizations, with women
among the leaders of these attempts. For example, in the 1870s Victoria Hull was three
times elected president of a national spiritualist association.
2. I do not know if infighting is characteristic of other Afro-Brazilian religions or if
it is connected to the higher presence of male leadership in Umbanda.
3. In men's ancestor worship the emphasis is on correct performance. There are all
kinds of ritual manuals and etiquette books with step-by-step instructions.
4. Lee had wanted Lucy Wright to succeed her, and she eventually did.
5. Garrett has pointed out another difference between Lee's and Whittaker's leader-
ship. "Ann Lee had always manifested in herself and elicited in others a great range of
ecstatic experiences and possessed behaviors, but most of the time the prevailing mood
seems to have been one of celebration, of rejoicing in the presence of the divine. Under
256 Priestess, Mother, Sacred Sister

Whittaker's leadership, the gifts of the Spirit were generally those of mortification,
forcing believers to confront and overcome their own sins, especially those of sexual-
ity" (1987, 212).
6. The argument I am making here is far from undisputed. Priscilla Brewer, the
expert in Shaker history who has been the most helpful to me in writing this book, feels
that I have overstated the case and that the sex of the sect's early leaders was not as relevant
as I claim. I choose to go ahead and make the argument anyway, largely in hope of
inspiring experts in other religions to take a new look at gender and institutionalization.
7. Women continued to play important roles. Brewer particularly points to the
power exercised by Ruth Landon and Asenath Clark (personal communication 1992).
8. Since many of the new religions of Japan are highly centralized (McFarland 1967,
84ff.), it is important to stress that these religions, after the first generation, are typically
dominated by male leadership.
9. Christian Scientists claim that God authored Science and Health—Eddy was but
the scribe. Cynthia Read has told me that Christian Science scriptures are printed in a
standard size so as to fit into the standard book racks with which their churches are
outfitted (personal communication, 1993).
10. Buddhism is a literate tradition; the indigenous religion is not (although people
know how to read). On Okinawa there are male ritual specialists who deal with
divination and geomancy and derive their knowledge from books.
11. Ranter's ideas are based on patterns of women's and men's leadership in "mascu-
line" institutions (American corporations). Thus, her work is not a test of "authentic"
female leadership styles, and may have only marginal relevance to women's religions
that are "feminine" institutions.
12. Of course, all of these may be traits of the leadership styles of subordinate groups
rather than specifically women's leadership traits.
13. For example, Cosentino has noticed that although Sande and Poro are very
powerful, the Mende showed a high level of tolerance to new religions that have been
introduced into the area (1982, 31).
 13
Spiritual Gifts and Secular Benefits

RELIGION AND SOCIAL STRUCTURE

Religion is never isolated from other elements of the cultural system. Max
Weber has claimed that religion is the form in which people worship society
by creating a supernatural world that mirrors human social structure, includ-
ing gender arrangements. Other scholars have shown that male-dominated
religions not only mirror gender discrimination but actually cause the subordi-
nation of women in the economic sphere (cf. Silverblatt 1980; VanEecke
1989). Can we then expect to find that female-dominated religions not only
reflect women's identities (e.g., as mothers), but also actively serve their
interests?
Before we begin to address this question, I wish to underscore that I do not
believe women participate in religion primarily in order to reap social, psy-
chological, economic, or political profit. Women, like men, join religious
groups and perform religious rituals for fundamentally religious reasons—in
Clifford Geertz's terms, in order to grapple with the ultimate conditions of
their existence. On the other hand, since religion is always integrated with
other aspects of the social structure, even the loftiest and most mystical
matters of spirit are clothed in the prosaic garb of rules, codes of behavior,
and organization. The factors that draw particular people to particular reli-
gions, and the advantages that accrue to them through that membership, are
multifaceted and include both sacred and profane elements. Put differently,
even though women become active in female-dominated religions for essen-
tially "religious" reasons, this does not negate the "secular" results and func-
tions of these religions.
In most contexts in which female-dominated religions occur, individual
women have some amount of free choice. Women's religions are rarely the only
available religious option. As we saw in Chapter 4, for example, Korean women
who wish to become religious specialists can choose between becoming sha-
mans or Buddhist nuns. This point is even clearer in North America, where
women have many religions—including many Protestant denominations—to
258 Priestess, Mother, Sacred Sister
choose from. Membership in women's religions must be treated as volitional
and members must be assumed to have some awareness of the advantages of
membership.
What happens once women join female-dominated religions? Do these reli-
gions offer ecstatic rituals that allow women to temporarily forget their suffer-
ing? Is religion their opiate? Do religious leaders tell poor women, women
whose children are hungry, that they will eat "pie in the sky when they die"?
The answer to these questions is—No! Social deprivation and personal crises
undoubtedly lead some individuals to seek a "better" or more convincing belief
system—one that more adequately explains why things are the way they are.
Spiritual solutions do not, however, preclude action designed to change an
unsatisfactory status quo. In Chapter 5 we saw that theological explanations
and ritual solutions for the problem of suffering in this world are the foci of
most women's religions. In the following pages we will see that many of the
women's religions offer, in addition, more concrete help with suffering. Given
the this-worldly orientation of women's religions, spiritual and earthly bene-
fits tend to be intertwined.
Many kinds of nonreligious organizations can meet women's needs for so-
cial interaction, group support, networking, empathy, even healing. Yet it does
seem that religious organizations have a unique role—even in terms of meeting
women's secular needs and furthering women's secular interests. This connec-
tion has been made explicit by Spiritual Feminists. A cornerstone of the Femi-
nist Spirituality Movement is that intellectual understandings of feminist princi-
ples are not sufficient to bring about real changes in consciousness. As Mary
Daly has written, "we have seen that patriarchy is designed not only to possess
women, but to prepossess/preoccupy us, that is, to inspire women with false
selves which anesthetize the Self" (1978, 322). Having grown out of the secular
feminist movement, Feminist Spirituality has transferred political concerns to
the spiritual sphere. The motto "The personal is political and the political is
personal" is reflected in almost all Movement writings. For many Spiritual
Feminists, religious ritual plays a key role in assisting women to liberate them-
selves from the thought patterns of patriarchy. "Magic . . . can be called the art
of evoking power-from-within" (Starhawk 1987, 24). And indeed, sociologist
Nancy Finley has documented the links between religious beliefs, a sense of
personal efficacy, and political activism among Dianic Wiccans—one branch of
the Feminist Spirituality Movement. Rather than functioning as an "opiate,"
religious involvement may empower women to work for social change in this
world.1
It is important to distinguish between the proclaimed ideology of the reli-
gion and the actual advantages that accrue to members. As we saw in Chapter
10, almost all women's religions preach nonpatriarchal gender ideologies.
Given the more or less "feminist" credos, it is of interest that certain religions
have provided only short-term relief for specific women, whereas others have
also resulted in long-term, structural benefits for women. We will look at these
patterns in detail below.
 Spiritual Gifts and Secular Benefits 259
In addition to spiritual satisfaction, enjoyable rituals, and deeper existential
understanding, female-dominated religions offer women concrete, secular
benefits in the here-and-now. As Michelle Rosaldo has shown, extra-domestic
ties with other women are an important source of power and value for women
in societies that hold to a firm division between male and female roles (1974,
39). And indeed, all the women's religions provide women with extra-
domestic ties and with some sort of improved personal circumstances. What I
wish to focus on here is why, if all (or almost all) female-dominated religions
provide transient help for specific women, certain religions also offer (or at
least work toward) long-term advantages to women as a group. In the remain-
der of this chapter we will look more closely at these two patterns, and try to
understand the structural conditions that allow short-term benefits to lead to
long-term advantages. I believe that attention to this question can provide
important insights into a variety of issues that impact on women's secular and
religious status.
Let me say from the outset that these two models are not mutually exclusive;
they should be visualized as more of a continuum. I have placed Christian
Science and Spiritualism in Model I (only short-term benefits) even though
both have been vocal supporters of women's suffrage (a long-term benefit)2,
because it seems to me that the emphasis in both religions has been on tempo-
rary relief. Once women's suffrage had become reality, neither of these reli-
gions remained at the forefront of other feminist struggles. In contrast, the
other two North American religions—Feminist Spirituality and the Shakers—
have invariably supported women's rights and have also provided long-term
advantageous alternatives for women as a group. It seems reasonable to argue
that even in the case of religions that ostensibly provide only transient benefits
for specific women, there is likely to be some sort of wider carryover. Thus,
while I claim that the zdr cult, for instance, provides benefits to the individual
possessed woman and not to women as a group, I find it hard to believe that
the knowledge that all women are potentially possessible has no generalized
impact on gender relations in the society as a whole.

MODEL 1: MODEL 2:
Only or Primarily Temporary Temporary Relief for Individual Women and
Relief for Individual Women Long-term Benefits for Women as a Group
Christian Science Northern Thai Spirit Cults
Spiritualism Sande
zar Shakers
Afro-Brazilian Cults Feminist Spirituality
Korean Shamanism Ryukyu Island Religion
Burmese Shamanism
Black Carib Religion3
260 Priestess, Mother, Sacred Sister

MODEL 1—TEMPORARY RELIEF FOR

INDIVIDUAL WOMEN
Zar

In the zar cult of Ethiopia most members are married women "who feel neglected
in a man's world in which they serve as hewers of wood and haulers of water . . .
Married women in the predominantly rural culture are often lonely for the
warmth of kinship relations for typical residence is in an exogamous patrilocal
hamlet" (Messing 1958, 1120-1121).

Researchers explain women's participation in zar cults in light of women's

social and cultural disadvantages, and their virtual exclusion from formal Is-
lamic rituals. In Islamic societies a female is placed under the care and authority
of males throughout her life. Legal practices, institutional mechanisms, belief
systems and social codes separate females and males in Muslim societies.
Women's access to public places and public roles is restricted, and women have
no formal role in political or economic decision making, although they may
have informal influence. (See also Chapter 2 for a discussion of the social
context of the zar.)
Women's participation in the zar cult has been interpreted by anthropolo-
gists as a function of all these factors. Other conditions scholars have cited as
laying the groundwork for women's attraction to zar include: cultural assess-
ments of women as weak, treacherous, inferior beings (Lewis 1969); genital
mutilation (Boddy 1988); patrilineal and patrilocal marriage that causes women
to be isolated from their natal families, coupled with marital insecurity—easy
divorce and polygyny (Kennedy 1978); and socialization that trains girls to be
submissive to males (Kennedy 1978). Still other authors have argued that zar
possession is a response to women's sexual deprivation, and human situations
that attract zars include opportunities for the spirit to have sexual intercourse
with a human victim of the opposite sex (Messing 1958).
How does zar cult participation redress any of these problems? "Lacking
full-status economic and political roles in the society at large, many women [in
Islamic societies], often with unconscious intent, struggle for control and
power through a 'manipulation of the supernatural' " (Beck 1980, 51-52). In
addition, women's religious activities create links between and among women
that sometimes override kin and status group barriers. While patriarchal,
patrilineal, and patrilocal social patterns restrict women to the domestic
sphere, zar cult participation gives women a set of activities that are not re-
stricted to the domestic sphere, and that allow women some measure of con-
trol over their time, activities, and lives.
The zar cult has multiple curing, religious, and social functions, and differ-
ent women utilize it for different reasons. Some women attend zar rituals to
induce possession trance, while others attend just for fun. Lucie Saunders
(1977) has found that socioeconomic class tends to determine what kind of
relationship women have to zar. Wealthy women, who are more restricted to
their houses for reasons of modesty and family honor, use zar rituals as oppor-
 Spiritual Gifts and Secular Benefits 261

tunities for socializing with other women. Poor women, who may go out
more freely, are less involved with the zar cult. Women who have migrated to
the cities and so suffer from the loss of their traditional village and kin-based
female support networks actively use the zar network to form friendship and
patron-client relationships, and to offer and gain services (Constantinides
1982). Rural women, on the other hand, have plenty of other opportunities to
get together with and receive support from other women.''
A woman possessed by a zar spirit becomes sick, increasingly unable to
function in the home. Eventually her husband grudgingly agrees to pay for a
zar ceremony. The path to healing, according to I. M. Lewis, is paved with
gifts provided by the patient's husband or other relatives. Gifts include per-
fume, silks, clothing, and gourmet foods. Lewis has analyzed the zar cult in
terms of gender hostility. Somali women have a strong sense of sexual solidar-
ity, coupled with antagonism toward men. Men see women's possession as a
strategy for attaining their goals at men's expense5 and explain that women are
more likely to be possessed because of their inherent moral weakness. The
relative secrecy of the cult arouses suspicion and interest among the men. They
suspect that immodesty and heresy are part of the rituals, yet they are afraid
that if they do not pay, their wives will remain ill and unable to carry out
household tasks. "Thus men are highly skeptical of women's sprite [zar] afflic-
tions which they regard at best as malingering, and at worst as a pernicious
extortion racket through which men are led to indulge their wives' insatiable
demands for new clothes and delicate foods" (Lewis 1969, 208). Although men
ridicule the zar cult, they pay substantial bills for their womenfolk's cures by
the cult.
Lewis, Saunders, and Constantinides have stressed the social and economic
benefits that accrue to women through zar participation. John Kennedy, on the
other hand, points to the role of zar ritual in cathartic discharge of emotions,
and the relief of sexual and aggressive tensions generated by the stressful
conditions of Nubian women's lives. Unlike in Somalia (where Lewis did his
fieldwork), the Nubian women of Upper Egypt studied by Kennedy do not
obtain much material gain from zar possession; instead, they gain attention. In
a similar vein, Yael Kahana (1985) argues that zar patients and leaders can "get
away" with deviant behavior because no one ever knows if it is the person or
the zar acting that way. Zar cults allow women to put their emotions first and
their duties second. During rituals, household work and childbearing and
rearing become secondary duties, while individualism becomes most impor-
tant. In addition, a husband or kinsman's willingness to pay for a woman's zar
activities is a public affirmation of her importance to him.
Most scholars feel that despite elements of gender antagonism, the zar cult
serves to maintain the status quo rather than to change it. "The zar cult is not
a deviant cult, its significance in maintaining the status quo in society has
traditionally been greater than improvement of social status. Zar membership
does not serve to raise or enhance the social status of its participants" (Mess-
ing 1958, 1125). Entranced women may be able to take on male prerogatives
(such as drinking alcohol and smoking) during the zar ceremony yet "[T]he
262 Priestess, Mother, Sacred Sister

new social relationships formed [through zar cult associations] do not in any
substantial way threaten the basic premises and mores o f . . . society. These
are the dominance by men of the formal political, economic, and religious
sphere of social organization, the segregation of the sexes, and the seclusion
of women" (Constantinides 1982, 201). The zar cult confronts individual
men and not patriarchy.

Other Examples of Model 1

We now turn to several other examples of women's religions that offer assis-
tance to individual women without seriously challenging patriarchal social
structures. In a series of case studies Patricia Lerch (1980) illustrates the multi-
faceted assistance that women receive from the Afro-Brazilian Umbanda cult.
A young mother who had been deserted by her husband came to a Umbanda
center for advice. The medium told her that in her past life she had been a man
who had also deserted his wife and family, and her punishment was to be
reincarnated as a woman in the same situation. Umbanda provided this
woman with an interpretation of her suffering that made the suffering explica-
ble and so easier to bear. In another case a woman's husband was cheating on
her. The medium (or actually the medium's spirit guide) warned her in ad-
vance so that she could confront him and tell him to leave before he had a
chance to desert her. A third case concerned a woman who was deserted by her
husband and became so ill and upset that she lost her job. The medium's spirit
guide told her to go the center secretary who would give her some money to
tide her over. In a final incident the medium hooked up an unemployed
woman with another client who needed a domestic worker.
Another way in which women benefit from Umbanda participation has
been noted by Esther Pressel: "Cecilia [an informant] once observed that her
religion had made her a 'free woman.' By this, she explained, she meant that
she could go out without her husband to private spirit sessions and afterward
to restaurants with Umbandists" (1980, 168-169). According to Pressel, Afro-
Brazilian rituals may come into play when tenuous male—female relationships
break down. For example, a single woman may be anxious that her lover is not
planning to marry her so she threatens him with black magic. Or a wife may
find it difficult to act out her role as a faithful spouse and turn to Umbanda for
psychological solace.6
Afro-Brazilian cults generally ignore the usual prejudices of Brazilian soci-
ety, and the cult centers protect criminals, homosexuals, prostitutes, and single
women who are no longer virgins (Carneiro 1940). In words almost identical
to those used to analyze the social impact of the zar cult, Leacock and Leacock
argue that "From the perspective of the members, the appeal of the Batuque
is not that it promises to change the world in which the members live, but
that it will help them to survive in that world" (1972, 326).
A similar theme emerges when we look at the effect that twentieth-century
Spiritualism has on women's lives. In Spiritualist circles in contemporary
Wales, emotional problems are worked out within the confines imposed by the
 Spiritual Gifts and Secular Benefits 263

traditional female role. "Indeed, this is where the contribution of spiritualism

lies. For the weekly repetition of healing activities and the exchange of mes-
sages 'from spirit' constitute a ritual of reconciliation to a situation which does
not permit any radical alternatives to itself. . . . [It] helps women to accept a
traditionally feminine role, which they frequently find frustrating and diffi-
cult" (Skultans 1974, 4-5). The messages of these spirits reinforce traditional
values and family arrangements. Vieda Skultans interprets Spiritualist ritual as
rites of reversal (cf. Gluckman 1963)—rites that appear to be a protest against
the established order but actually reflect an overall acceptance of that order.
Rites of reversal allow individuals to let out frustration and ultimately submit
to and accept the social order. It seems to me that Skultan's analysis holds
equally true for both the zdr cult and the Afro-Brazilian religions.
An even clearer example is Mexican Spiritualism, which, according to Kaja
Finkler, "promotes well-being on the individual level, but tends to perpetuate a
societal status quo that is rejected by the actors themselves and may also be illness
producing" (1986, 628). Religious rules and temple leaders explicitly discourage
political activism. The Mexican Spiritualist God recognizes that His followers
are poor, but tells them that if He were to give them material benefits they would
abandon Him. On the most immediate level, temple attendance often results in
loss of economic productivity—healers sometimes spend two days a week work-
ing (without pay) at the temple. On the other hand, Mexican Spiritualism does
not tell adherents to passively accept suffering—rituals do aim to alleviate indi-
vidual illness (although not the underlying social causes of illness).
Mexican Spiritualism provides certain benefits to the wives of male adher-
ents. Spiritualist men tend to abstain from alcohol and behave in a less "ma-
cho" manner. In Finkler's sample, 33% of male adherents had been alcoholics
when they first became involved with Spiritualism. "In becoming adherents,
men relinquish their drinking habits and thereby become unlike the majority of
Mexican men whose drinking patterns define them as men" (1986, 633). In
contrast to the dominant local gender ideology in which heavy drinking, pro-
miscuity, and wife-beating are what define maleness, Spiritualist men avoid
drinking, exhibit few macho traits, and are more likely to spend their leisure
time with their families. Clearly, individual women benefit from their hus-
bands' spiritualist participation.
The most extreme instance of Model I is the new Japanese religion Tensho-
kotai-jingu-kyo. The foundress of Tensho taught that women (like herself)
who have known much suffering are very close to the Kingdom of God, and
that in the coming Age of God women will play a more significant role than
they have in the past. In the present man-centered age power rules the earth,
while in the better forthcoming age women will be in the vanguard. On the
other hand, she encouraged women to continue to obey their husbands and
elders and she preached highly conventional, sex-linked social roles (Naka-
mura 1980). Since Tensho deals with healing and many Tensho members have
recovered from illness or other forms of suffering, it is undeniable that individ-
ual women do receive certain benefits from Tensho membership, although
Tensho in no way challenges patriarchy.
264 Priestess, Mother, Sacred Sister

MODEL 2—SHORT-TERM AND LONG-

TERM BENEFITS FOR WOMEN
Sande
Sande is a "social and religious association that would promote an egalitarian
position for women and ensure for them a more favorable status and an active
participation in the socio-religious life. ... In addition, women wanted consid-
erable freedom for their sex, and the creation or adoption of a powerful secret
society with a wide range of powers was an effective way to guarantee this
right" (Richards 1973, 71).
In Chapter 6 we explained that Sande initiation teaches girls domestic skills
they will need as adults. These tasks are not new to the girls, who have already
helped their mothers care for the house and younger siblings. What they really
learn in Sande are attitudes toward work: in particular, cooperation with other
women.
Both the fun and the pain of initiation serve to bind the girls together into a
united society. At the end of the initiation period they swear an oath never to
reveal any fault in another Sande woman. The emphasis on cooperation en-
ables women to deal with the many problems they are likely to encounter in
the polygynous households they marry into. "This solidarity training helps to
mitigate co-wife rivalry and the potential divide-and-rule powers of poly-
gynous husbands" (MacCormack 1977, 99). In many societies, virilocal resi-
dence means that a woman is cut off from her own kin and isolated from any
group that could empower or support her, yet Sande provides a woman with
group support throughout her life. Sande rules state that respect must be
shown to women. Sande organization safeguards the rights and position of
women, and punishes men who mistreat their wives.
Sande women about to give birth prefer to return to the chapters in which
they were initiated. Carol MacCormack explains that individual women link
their mother's and mother's-in-law Sande chapters. "Contrary to the Levi-
Straussian model of women as passive objects transferred between groups of
men, in Sande, women link corporate groups composed exclusively of
women" (1977, 96). The same official who performed the clitoridectomy will
serve as midwife; the same women who supported her during the clitori-
dectomy will support her during childbirth. Women are also buried by the
chapters in which they were initiated, and so become Sande ancestresses.
Earlier in this chapter I made the claim that almost all women's religions
benefit specific women on an immediate level. On the face of it, Sande does
not seem to fit into this pattern—clitoridectomy can hardly be said to serve
women's secular interests. On the other hand, the collective effect of Sande
membership clearly strengthens women's economic, political, and family situa-
tions. While I am not fully satisfied with this argument (see Chapter 6), it is
what the anthropologists who have studied Sande report.
During the initiation rituals, sponsoring mothers and Sande officials may
stay in the grove for weeks or months. Sande provides women an alternative
space not controlled by men. Sande women have a special vocabulary that they
 Spiritual Gifts and Secular Benefits 265

use for talking about women's business. This vocabulary contributes to the
mystique of power and secrecy surrounding Sande. Throughout the year
Sande women meet to air common grievances, which certainly enhances their
political position in society.
Belonging to Sande gives women a great deal of power, both in the commu-
nity and within the marital relationship. Sande women know that they control
a scarce and important resource: offspring. Sande women can withhold that
resource if their husbands violate Sande laws. And men know that Sande
women have knowledge of medicines that can cause impotence. A constella-
tion of Sande laws seem designed to protect women's privacy: Men are not
allowed near women's bathing areas, men travelling alone must announce their
presence by means of an appropriate noise. Sande rules also serve to control
male violence against women: a man who strikes his wife in the presence of her
mother is obligated to placate both her and her Sande group (Harris and
Sawyerr 1968, 105).
Sande ritual may not be contravened by any man, and even a chief who
attempted to do so would be punished in a show of authority by Sande
women. Sande laws affect the entire community, and the sanctioning power of
Sande is recognized by the community. When at the end of the initiation season
masked ancestral figures appear in the village, one of the "ancestresses" pub-
licly shakes a bundle of switches at men and women who have violated Sande
rules. All who watch are reminded of the power of Sande officials to catch and
punish transgressors.
For women who have migrated into cities, Sande furnishes neighborhood
chapters that offer women companionship and social support in life-crisis situa-
tions. In the nineteenth century, women in Sierra Leone took advantage of
Freetown's growing population to become traders. Membership in the wom-
en's secret society provided business connections and personal protection in the
absence of kinship ties. Sande sisters could certify that stranger-traders could
be trusted in commercial relations and Sande held the traders to certain stan-
dards of behavior that were locally acceptable.
The group solidarity encouraged by Sande has had broader economic and
political implications. West African men and women normally engage in differ-
ent economic pursuits. However, when other sources of income are not avail-
able to West African men, they compete with women for agricultural re-
sources. This is problematic for women because men are at an automatic
advantage in that they have jural rights to the land. Sande gives women an
organized forum for protest in this and other situations. MacCormack (1979,
35) argues that, "Because Sande women are organized into effective corporate
groups and enjoy female autonomy in all matters pertaining to Sande's sphere
of interests, their political position in the larger society is enhanced."
According to David Rosen, the Women's War of 1929—a tax rebellion led
by peasant women against colonial authorities—developed from women's
Sande activities. Again in 1971 Sande women protested against men who
neglected their families in favor of personal economic gain, and against the
abuse and corruption of local authorities. Rosen documents how in 1971, in the
266 Priestess, Mother, Sacred Sister

wake of economic changes threatening the agricultural, marketing, and eco-

nomic autonomy of women, "Women began parading down the streets, not
bearing the symbols and signs of the Sande, but dressed up in potato leaves,
and other representations of agricultural activities. Women claimed that the
donning of agricultural symbols would bring them good luck. As the dancing
and parading became more widespread, women began to accost and hurl
abuses at men of authority [who backed up the jural and economic advantages
enjoyed by men]. . . . two subchiefs were dragged from their porches and
beaten. Even the arrest of several of the demonstrators failed to quell the
demonstrations. Women claimed that their activities were protected under the
ritual umbrella of the Sande association" (1983, 39).
Despite the culturally acknowledged interdependence of men and women,
the societies in which Sande women live are not egalitarian. But because a
woman's allegiance to Sande—to other women—is seen as equally or more
binding than her allegiance to her husband, women are guaranteed a measure
of independence lacking in many other cultural systems. Sande societies endow
women as a group with long-term, structural benefits, and women as individu-
als with short-term pragmatic benefits.

Shakers
Unlike Sande, which is more or less an automatic part of life for women in
many parts of West Africa, Shaker membership was much more a matter of
choice for individual women. Economic distress, among other factors, moti-
vated many individuals to join the Shakers. Facing poverty or bankruptcy, men
and women saw in Shaker communities a financially safe life-style centered
around values of pacifism, work, thrift, celibacy, and cleanliness for both
sexes.
Undoubtedly, part of the appeal of Shakerism—especially to women—was
the neat and ordered houses in the quiet countryside, away from the noise and
economic and sexual exploitation of industrializing cities. Organized into
"families" of celibate men and women, Shaker villages offered a standard of
living more comfortable than many members had known in their former lives.
Shakers invented numerous technical improvements to simplify household
chores (pot handles, clothespins, machines to cut fruit and shell peas) and
economize agricultural work. Less popular tasks (such as cooking) were ro-
tated. Working conditions for both men and women were humane and pleas-
ant. "Life in a Shaker society offered [women] economic security, meaningful
activity, and a supportive sisterhood; for a few women, it offered the opportu-
nity to exercise gifts of leadership which would hardly have been tolerated in
the outside world or in any other religious institution" (Procter-Smith 1985,
220). The neat and clean Shaker life-style had clear benefits—Shakers were
healthier and lived longer than the general population (Bainbridge 1982, 359).
The religious experiences of Shaker men and women were equally re-
spected, both men and women wrote articles in the Shaker newspapers, and
women predominated in spiritual gifts such as testimonies and writing hymns
and poems. Particularly when Shakerism expanded into the West in the early
 Spiritual Gifts and Secular Benefits 267

nineteenth century, increased leadership and travel opportunities opened up for

Shaker women (Desroche 1971).
Shaker society legislated extreme sexual segregation. This had both positive
and negative implications for women. In the economic sphere Shaker women,
like non-Shaker women, cleaned, cooked, sewed, and engaged in small manu-
facture, while men worked in agricultural production. In general, men were in
charge of business and temporal matters. Shaker women cleaned the men's
rooms and cared for their clothing (but not in the presence of men).
Interactions between the sexes were for the most part forbidden. Relations
between women and men were limited to co-presence at meals and meetings.
Men and women did not shake hands or touch, they worked in separate
workshops, ate at separate tables, prayed and danced in separate groups, and
rarely spoke to one another. Even Shaker architecture reflected their notion of
the sexes as equal, distinct, and separate: buildings often had two outside doors
and two sets of staircases.
Unlike in non-Shaker society, both men and women participated in the care
and education of children (men cared for boys and women for girls). In light of
the emphasis that feminist scholars have placed on the sexist repercussions of
women's predominance in childrearing (see Chapters 3 and 10), the signifi-
cance of this aspect of Shaker life should not be underestimated.
In trying to assess how Shaker sexual segregation affected women, it is
crucial to bear in mind the following points. On the one hand, it limited
women's options in terms of work. On the other hand, it protected women
from the dangers of childbirth and allowed women to exercise leadership over
their own activities. It seems to me that in comparison to the lot of poor and
working women in eighteenth- and nineteenth-century America, the lot of
Shaker women—particularly in terms of economic and sexual security—was
enviable. Especially after 1860, as male membership declined, the Shakers
moved toward more complete gender equality (Brewer 1992, 611). Although
Shaker communities never attracted more than several thousand women, for
those women Shakerism did offer ongoing, collective benefits.

Sanctificationists
The nineteenth-century Sanctificationists—an all-female Christian communal
sect—were critical of the sexist economic implications of traditional relation-
ships in which the husband controlled the family finances, including money
and property that the wife had brought into the marriage or earned on her
own. In protest against discriminatory family arrangements, these women
formed their own cooperative businesses and living arrangements. The Sanctifi-
cationist sisters established a system of communal housework and child-
rearing that alleviated the isolation and boredom of women's work in main-
stream society. And according to Sally Kitch, there is evidence that "some of
the sisters sought refuge from the dominance of abusive or drunken husbands
who, in mainstream society, were still deemed women's superiors, even when
inebriated or violent" (1989, 66). It is of interest to note that although most of
the Sanctificationist sisters left middle-class families to join a community in
268 Priestess, Mother, Sacred Sister
which they worked at unskilled, working-class labor, "The sisters reported
new levels of self-esteem from the discipline and integrity of their menial but
independent work" (Kitch 1989, 152). As Kitch shows, they transformed the
typical tedious household labor of women "from an endless round of ill-
defined tasks to a four-hour workday in which tasks were defined, professional-
ized, and rotated. Outside of those specific hours, a woman's time became her
own. Their collective efforts enabled them to travel extensively" (1989, 152).
In addition, Sanctificationist sisters supported women's suffrage and other
struggles for women's rights.
Although the Sanctificationists began as a religious group, and only created
their communal work and life-style in order to survive once their celibacy
had estranged them from their husbands, they eventually became primarily a
secular community. According to Kitch, "having served its purpose, religion
faded away. Celibacy rose in symbolic importance over the religious beliefs
that had spawned it. Sanctified spiritual power was reinforced by female
secular efficacy. Any group of women who could succeed and prosper as they
had done must be in tune with divine will, the sisters reasoned" (1989, 157).
Like the Shakers, Sanctificationist sisters constructed a new, less patriarchal,
total life-style.

Northern Thai Matrilineal Spirit Cults

The phii puu njaa cult is a ritual complex possessed by, oriented toward, and
controlled by women. The cults arbitrate moral and interpersonal behavior,
determine inheritance and household membership, and allocate land re-
sources. The cults sacralize and legitimate women's land ownership. All fam-
ily members—male and female—are obligated to participate in the cult offer-
ings and behave in accordance with cult expectations. Richard Davis argues
that male power, which derives from knowledge of Buddhist texts, is
strongly balanced by female power deriving from the tutelage of domestic
spirits (1984, 267).
While some ethnographers have argued that the main function of the cults is
to limit the sexual behavior of female cult members, Ann Hale believes that the
fact that men must pay fines for sexual transgressions means that the cults
furnish women with a way to restrain male sexuality—a means that women in
many other cultural situations would surely envy (1979, 147).
The cults encourage women to live in close proximity with their own kin,
and offer women protection in the case of divorce and male mobility—both
very common in Northern Thailand. To summarize, Northern Thai Matrilineal
Spirit Cults provide institutionalized economic, sexual, social and religious
independence for women.

A COMPARISON OF THE TWO MODELS

Women reap spiritual, intellectual, personal, social, economic, and political
benefits from women's religions. These benefits include nonpatriarchal gender
 Spiritual Gifts and Secular Benefits 269
ideologies (Chapter 10) and rituals that address women's gravest existential
concerns (Chapter 5). In this chapter we have seen two models of material
advantages that accrue to women. Model i, as exemplified by the zar cult,
helps individual women overcome isolation and assists them in negotiating
with their husbands. On the other hand, only minor long-lasting collective
benefits for women can be attributed to participation in zar, Spiritualism, or
Afro-Brazilian cults. On the whole, these minor benefits are the result of the
cumulative affect of short-term benefits rather than any real structural improve-
ment. In Model i religions what we typically find is that short-term advan-
tages actually end up disempowering women as a group because the religion
reconciles individual women to powerlessness.7 Model 2, as exemplified by
Sande, Shakerism, and Northern Thai Matrilineal Spirit Cults, not only pro-
vides short-term assistance for women, but also gives them a permanent
power base for protecting their collective interests. In these religions individual
benefits lead to group empowerment. Five sets of factors enable us to under-
stand why these very different models occur.

Sisterhood Is Powerful
Whereas Model i consists of women who meet periodically to form a ritual
group, four of the five examples of Model 2 are comprised of women who
either live together (Northern Thailand, Shakers) or who have a lifetime asso-
ciation with one another (Sande, Ryukyu Islands). 8 It seems that women who
have ongoing, multifaceted relationships with one another reap more extensive
benefits from female-dominated religions.
I find it interesting that in all the Model 2 religions sisters are crucial ritual
actresses (see Chapter 10). In Northern Thai matrilineages, for example,
groups of sisters (and their children) constitute the cult group. In other Model
2 religions fictitious "sisterhood" relations are created. The Shakers forged a
new community of "brethren" and "sisters," and conjugal love was seen as far
inferior to sibling love (Kitch 1979, 51). Feminists, including Spiritual Femi-
nists, praise the universal sisterhood of women, declaring that "Sisterhood Is
Powerful!" Sande women refer to one another as Sande sisters, and adopt an
attitude of mutual support among themselves. And on the Ryukyu Islands,
unlike in almost any other known religion, women gain their spiritual power
through their role as sisters. Sisters, not mothers or wives, perform the most
critical annual rituals. It is useful here to recollect the discussion of Karen
Sack's work (Chapter 3) in which the argument is made that for African
women the role of sister is far more empowering than that of wife. In women's
religions "sisterhood" is a symbolic expression of ongoing unity. While moth-
erhood is what draws women to female-dominated religions, sisterhood is
what empowers them to translate religious participation into secular benefits.
Perhaps the greatest evidence for the importance of sisterhood comes from
two religions in which there is an urban and a rural variety. Concerning both
the zar cult and the Northern Thai spirit cults, I would make the case that the
rural cults—in which members are part of an ongoing, multifaceted
270 Priestess, Mother, Sacred Sister
community—serve women's interests as a group (at least to some extent),
whereas in the city the cults are much more a matter of short-term assistance
for individual, unrelated women. Regarding the zdr, this is precisely the
argument made by Janice Boddy. Boddy has tried to understand why I. M.
Lewis's findings concerning the secular implications of the zdr cult are so
different from her own findings. She concludes that Lewis "extrapolates from
urban cult to rural situation without paying adequate heed to the specific
social and economic configurations of these locales within Sudan" (1992, 3).
Her central observation is that unlike in urban Sudan where women are
indeed socially isolated, in rural Sudan "women daily walk, unchaperoned in
village lanes and visit freely in each others' homes. Indeed, they are morally
enjoined to witness the life transitions and crises of neighbors and kin, even
when the latter reside in distant villages or towns" (1992, 4). Boddy shows
that in Sudanese villages there is a considerable overlap between kin ties and
zdr cult ties. And whereas official kinship ideology is patrilineal, because of
the typical pattern of intravillage marriage (endogamy), individuals tend to be
related to each other both through patrilineal and matrilineal ties. The secular
effect of rural zdr cult involvement, in which related females compose the
membership, is to provide "an embodied counterpoint to officially articulated
kinship; it activates matri-group understandings, mobilizes matri-kin sup-
port" (1992, 7). In short, what Boddy shows is that in a rural context—
where cult members are related to one another through kinship ties ("sister-
hood") zdr religion involvement serves to strengthen matrilineal bonds and
ideology. This is very different from the scenario painted by Lewis and others
regarding urban Sudan.

Economic Independence
In the societies in which women have little or no economic power (Islamic
Africa, urban nineteenth-century North America, and contemporary Brazil)
women's religions serve only temporary, individual women's interests.
The interrelationship between economic organization and women's status
has been investigated at length, particularly by Marxist-feminist anthropolo-
gists. The conclusions are clear: Where women are excluded from economic
roles that allow them to produce and control items of exchange value (rather
than items for private use), women's status is low (see, e.g., Sacks 1974). We
expect to find, and indeed find, that Model 2 religions occur in societies in
which women not only work, but in which they produce items that receive
social value. The important role of women in the markets of Thailand and
West Africa has been noted by many anthropologists (see Chapter 2), and is
certainly relevant to the argument I am making here. In Shaker communities
women, like men, produced items for sale. And in the contemporary United
States women who identify with the Feminist Spirituality movement tend to
work at salaried, socially valued jobs. In contrast, when we look at the Model I
religions we see that Brazilian women tend to be limited to domestic work for
 Spiritual Gifts and Secular Benefits 271

private families; Korean women (particularly in urban areas) are increasingly

seen as "mere consumers" (Kendall, personal communication 1992); Black
Carib women are dependent on their out-migrating children (especially sons)
to send them money; and nineteenth—century Christian Science and Spiritualist
women had few economic opportunities open to them.
That economic independence has an impact on women's religions is not
surprising. As I said at the beginning of this chapter, religion is never isolated
from social structure. In intensely patriarchal societies women do not have
opportunities to form lasting, public alliances that empower them to effect
permanent benefits for women as a group. In fact, in societies in which
women's economic role is extremely marginal, we do not find that female-
dominated religions occur at all. A certain measure of economic freedom seems
to be a precondition for women to form autonomous religious organizations.
Even the most transcendent religious beliefs necessitate some sort of expres-
sion on the earthly plane; otherwise, there can be no community of believers.
My argument is simply that earthly expressions require material resources.
Without access to those resources, women cannot develop or maintain female
dominated religions.
The interesting question is why a higher degree of economic independence—
and specifically the opportunity for women to produce items that have market
value—correlates with the occurrence of female-dominated religions that serve
women's communal interests (Model 2). It seems to me that the answer most
probably has to do with whether women's lives have an important public com-
ponent, or whether their activities are limited to domestic realms. As Rosaldo
(1974) has argued, women's status tends to be the lowest in societies in which
there is a sharp division between the domestic and the public, and in which
women are associated solely with the domestic sphere. In Model 2 societies
women either move freely in and out of the domestic and the public realms (West
Africa, Northern Thailand, contemporary United States), or there is no radical
distinction between the two realms (Shakers9). I suggest that the social location
of women in the public realm is one of the preconditions for numerous kinds of
organizations and ideologies—such as female-dominated religions—that offer
structural advantages to women.
The model I am offering here is circular: economic independence leads to
religions that offer women long-term benefits including greater economic inde-
pendence. Thus, one of the roles of female-dominated religions in the Model 2
scenario is to protect and legitimize women's economic freedom. In the North-
ern Thai cults this is clear—the cults sacralize women's control of land. The
Shakers also present an unambiguous case in point: Shakerism actually con-
structed the female economic realm, and ensured that women (because of
sexual segregation) had some measure of autonomy in running it.
A last comment that is in order here concerns Peggy Sanday's observation
that a correlation exists between the presence of female deities and female
contribution to subsistence.10 While further research is needed to substantiate
this claim, it may well be that when women have an important subsistence
272 Priestess, Mother, Sacred Sister
role, women are perceived as "cosmically" powerful. On the symbolic level
that power is imaged as goddesses; on the structural level that power leads to
women's religions that enhance women's communal status.

Control of Sexuality and Fertility

A third factor that is relevant to understanding the two models concerns fertil-
ity and sexuality. Model 2 religions empower women to control their own
fertility, whereas Model i religions do not.
To begin with Model i, we see that the zdr is located in societies in which
women's sexuality and reproductive powers are perceived as extremely impor-
tant and a great deal of emphasis is placed on men's control of those powers.
Women are carefully watched to ensure that they do not damage their families'
honor. Sexual segregation, the authority of fathers and brothers, and rigid
norms of female modesty all help to ensure that women will have a minimum
of contact with unrelated males. Virginity at marriage is vital, and its lack is a
cause for divorce and disgrace. Zdr rituals do not challenge this cultural ideol-
ogy, nor do they give women control of their own bodies.
Similarly, Brazilian women are expected to be virgins at marriage, and tew
men will marry a nonvirgin. A married woman's behavior is restricted to
ensure that she does nothing that her husband can interpret as inviting the
attentions of other men. There are cultural sanctions against a woman who is
unfaithful to her husband, and her husband may beat or even kill her and her
lover. Legal birth control and abortion are unavailable in Brazil. Until 1977
divorce was not permitted, and in a marriage that broke up a woman had no
chance for remarrying, although the man was expected to take a concubine.
For men, early and frequent sex is regarded as healthy and masculine, and
marriage does not restrict a man's sexual activities. Participation in Afro-
Brazilian cults does not challenge this system; it does not give women control
over their own sexual or reproductive processes.
In short, membership in zar and Afro-Brazilian cults helps an individual
woman negotiate with the particular men who have power over her fertility
and sexuality; it does not shift the balance of power in any real way.
Northern Thai women, on the other hand, have traditionally been in control
of their own fertility. They embraced modern birth control methods in the
19605, and until then knew and approved of folk methods." The spirit cults
ensure that men outside the family do not trespass on the sexuality or fertility
of cult members, but, in contrast to North Africa and Brazil, there is no sense
that women's bodies belong to the men of their family. Similarly, Sande takes
fertility out of the sphere of nature, and places its control squarely in the hands
of the female group. Among the Mende a woman who has given birth is
expected by Sande rules to stay away from her husband until the child is
weaned, a period that often extends to three years (Harris and Sawyerr 1968,
96). In societies with high infant mortality rates, extended post-partum sexual
taboos allow mothers to avoid becoming pregnant during the time that the
baby needs her milk. A prolonged sexual separation also prevents her body
 Spiritual Gifts and Secular Benefits 273
from becoming depleted through constant pregnancy. Sanctions of various
sorts are exercised against those who doubt that Sande gives women jurisdic-
tion over sexuality and reproduction.
To take a last example, political activism of contemporary feminists, includ-
ing Spiritual Feminists, has centered around issues directly related to women's
control of their bodies (e.g., rape crisis centers, battered women's shelters,
pro-choice organizations, and self-help health groups). Spiritual Feminism
grew out of a cultural climate in which women were fighting to control their
own bodies, and it sacralized that control.
Control of fertility and sexuality reaches its apex in voluntary celibacy.
Cross-culturally, celibacy is a two-edged sword: it can be used to promote an
image of women as carnal and polluting, or it can offer liberation from the
bonds of marriage, motherhood, and dependence on men. In the few women's
religions that preach celibacy, the second of these rationales emerges. In
female-dominated religions, celibacy has served women's interests by giving
women authority over fertility (cf. Kraemer 1980). Shakerism, for instance,
provided women with dominion over their fertility at a time in which mother-
hood was a physically and emotionally devastating experience for many
women (Kitch 1989).
In the past, high-ranking Okinawan noro (priestesses) could not marry or
have sexual relations (Mabuchi 1964). Some villagers explain that noro, as the
potential wives of male deities, must keep away from pollution and mortal
men.I2 The celibacy of Okinawan noro makes sense in light of the tradition of
vnoro owning their own land: Celibacy and land ownership were two aspects of
the institutionalization of independence for noro. Unmarried and economically
secure, priestesses were in full control of their own bodies. Descriptions of the
ritual roles of Okinawan priestesses display a great deal of time spent traveling
from holy site to holy site. Children were undoubtedly a hindrance to the
mobility of priestesses. Yet even more important, absolute control of their
own fertility ensured a level of independence and power to Ryukyuan priest-
esses that has rarely been equalled in any known society.13
I would like to refer here to Ruth Bleier's analysis of sexuality as the "king-
pin in the patriarchal formations that serve to oppress women" (1984, 164).
Bleier argues that underlying all forms of patriarchal oppression—political,
economic, religious—"are the assumptions of the institution of heterosexuality
or heterosexism: specifically, the assumptions that men own and have the right
to control the bodies, labor, and minds of women" (1984, 164). She contends
that sex, partly because of its association with intimacy and love, "is by its very
physical nature the most seductive, private, intrusive, direct, and possessing
way to exert power and control" (1984, 181). Bleier shows how Western
religion (especially Catholicism) and Western medicine and psychoanalysis
have served to master women's sexuality in the interests of male dominance.
Both sets of institutions establish, express, embody, or enforce "a set of power
relationships between the participants: one, the authority who compels, ques-
tions, prescribes, judges, punishes, forgives, deciphers, and interprets; and the
subject who sins, speaks, and receives interpreted truth" (1984, 177).
274 Priestess, Mother, Sacred Sister

There are three aspects of male control of women's sexuality that Bleier sees
as especially significant. First, from a young age girls are taught to be appealing
to men. Second, from a young age girls are also taught to fear rape. And third,
violent sex is used as a means of punishing women who "get out of line"
sexually or who otherwise disobey or displease men. In light of these three
components, we can see again how the two models of women's religions
differ. To take the Model 2 religions, in the Northern Thai Spirit Cults the
spirits punish a man who transgresses on a woman's body. In Sande, the
ancestors and sisters punish a man who trespasses on a woman's body. Femi-
nists, including Spiritual Feminists, have made a point of not only demanding
stricter enforcement of anti-rape legislation and expanding legal definitions of
rape to include marital and date rape, but of legitimizing nonheterosexual
sexual expression. And the Shakers, as we saw earlier, broke the bonds of
heterosexuality by demanding celibacy for both sexes. These kinds of themes
are absent from Model I religions.
Not all female-dominated religions give women control over their own
fertility or sexuality. On the other hand, no female-dominated religions ritu-
ally bestow on men institutionalized control over women's fertility or sexual-
ity. Since in so many cultures the role of religion is to legitimize and enforce
male control of women's sexuality and fertility, I am struck by the absence of
wedding ceremonies or other ceremonies in which ownership of a woman's
body is transferred from one man (e.g., father) to another man (e.g., husband)
in female-dominated religions.14 I believe it is fair to say that all the key
examples of women's religions implicitly (if not always explicitly) encourage
some measure of independence and power for women regarding fertility. The
degree of that independence varies, and is reflected in the ability of the religion
to, in the long-term, promote women's secular interests.

Spirit Possession
In Model I religions spirit possession plays a very central role. Certain Model 2
religions include spirit possession, but its role is very much a minor one.
An eloquent discussion of the inability of spirit possession to substantially
affect women's status has been presented by Roger Gomm in his study of the
Digo of south Kenya. First, according to Gomm, spirit possession—in that it
is an involuntary affliction—rarely leads a woman's kinfolk to grant her any
real, meaningful favors. The most that she can hope for are token conciliatory
gestures. In the context of the possession ritual, women may challenge current
sex roles and norms. However, not only is there no direct carryover to the
"outside" world, but this challenge occurs while the woman is in trance and so
in fact renouncing personal responsibility (cf. Skultans 1974 on Spiritualism).
In short, the possessed woman is not acting from a position of power.
Second, according to Gomm, the concept of spirit possession mystifies the
true sources of tension and discontent in the society, and those who "suffer from
possession are set in a system of rewards and punishment which enables them to
be manipulated to confirm a theory. The privileges permitted and the gifts
 Spiritual Gifts and Secular Benefits 275
granted to possessed women are part of this matrix. . . . [SJpirit possession
attacks may redress the disadvantages of a lowly and deprived status, but at the
same time the way in which this amelioration is obtained ensures that deviants
confirm a mystifying theory which is an essential element in the control system
which maintains them as lowly and deprived persons" (1975, i35)- 1 5
I am not sure that I agree with all of Gomm's ideas. In particular, he seems to
ignore the distinction we made in Chapter 9 between lay and professional
possession. His description fits well the experiences of a lay housewife who
becomes possessed in the wake of illness or marital problems, but it sheds no
light whatsoever on professional shamans who voluntarily and expertly
beckon and manipulate the spirits throughout a long and socially recognized
career. Indeed, Gomm defines spirit possession as an "involuntary affliction," a
description that hardly fits the religious work of an experienced Afro-Brazilian
"Mother."
Regarding lay spirit possession I believe that his points are well taken. The
very essence of the role conveys a sense of weakness: the woman herself did not
challenge or demand, she was merely a vehicle or "horse" of the spirits. On the
other hand, I beg the reader to bear in mind the argument I made in Chapters 8
and 9 concerning the cognitive impact of seeing one's gods, spirits and ancestors
consistently embodied in female form (through spirit possession).

Illness
In Model i religions women almost always join the religion because of illness.
Model 2 religions have little, if anything, to do with illness.
Anthropologist Mary Douglas has proposed that bodies serve as social maps
onto which cultural categories and boundaries are projected. If we concur with
Douglas (as most contemporary anthropologists do), how are we to interpret a
model that shows that in order to join a religion a woman must be ill? And
even more to the point, a lay member simply joins through illness, but a leader
typically has suffered and then dramatically overcome long, painful, debilitat-
ing diseases and misfortune (see Chapter 11). Drawing on Douglas's ideas, I
would argue that in Model i religions the cultural categories that are "mapped"
onto women's bodies are weakness, suffering, and powerlessness. In Model 2
religions, the social "map" reads very differently. Illness is, at most, peripheral;
sisterhood, economic independence, and control of fertility are central.
The connection between illness and women's oppression has been well dem-
onstrated in Barbara Ehrenreich and Deirdre English's expose of the sexual
politics of sickness in the United States. Ehrenreich and English show how
nineteenth-century culture fostered among middle-class women a "morbid
cult of hypochondria" together with fervent dependence on medical practitio-
ners (1973, 17). The medical view identified all female reproductive functions
as inherently pathological, a view that surely weakened women's structural
status. Sickness became synonymous with femininity, female disorders were
traced to women's sexual, athletic, or mental exertions, and women were
counseled to rest and refrain from public activities.
276 Priestess, Mother, Sacred Sister

I quote here from Ehrenreich and English: "[L]ate nineteenth century medi-
cal treatment of women made very little sense as medicine, but it was undoubt-
edly effective at keeping certain women ... in their place. As we have seen,
surgery was often performed with the explicit goal of 'taming' a high-strung
woman. . . . The more the doctors 'treated,' the more they lured women into
seeing themselves as sick" (1973, 35-37).
Ehrenreich and English do point out that women were sometimes able to
turn the sick role to their own advantage, both as a form of birth control and as
a way of gaining attention within the family. This is consistent with what we
have seen in female-dominated religions. Illness often allows women a certain
measure of respite from an intolerable family situation (e.g., Korean shamans)
or an opportunity to leave the house for short periods of time and find comfort
in the company of other women (Afro-Brazilian religions).
On the other hand, in contrast to sisterhood (symbolic or biological) that
cornmunalizes women and thus allows women to achieve collective benefits,
illness individualizes women's problems and women's solutions. A sick woman
is treated as a private invalid, not as a victim of structural oppression. As a result,
healing—even in the context of a female-dominated religion—tends to cure the
individual woman and not the societal ills that may well have been responsible
for her illness to begin with.

CONCLUSION
Having laid out and analyzed these two patterns of secular benefits, it is worth
noting that the opposite patterns do not seem to occur: There are no female-
dominated religions in which individual women suffer immediate disadvan-
tages yet women as a group benefit in the long-term16, and there are no female-
dominated religions in which women are harmed both in the long term and the
short term. The absence of the second pattern should not be surprising: As I
said in Chapter 10, women's religions preach relatively egalitarian gender
ideologies; it would be odd to find that egalitarian gender ideologies go to-
gether with entirely disadvantageous effects on women. The absence of the
first pattern warrants more thought. We could easily imagine a religion in
which individual women choose to sacrifice themselves for the greater good of
womankind; martyrdom is certainly a well-known theme in religious history. I
can only guess why this pattern does not occur. It seems to me that an ideal of
individual martyrdom is inconsistent with the immanent character of women's
religions (see Chapter 7). Women's religions tend to focus on the here-and-
now and not on the hereafter, regardless of whether that hereafter is in the next
world or in this one.
In this chapter I have argued that while it is incorrect to assume that women
join female-dominated religions solely because of social and psychological
deprivation, it is clear that participation in these religions serves women's
secular interests in a variety of ways. On the whole, female-dominated reli-
gions provide women with short-term assistance. In certain circumstances,
that short-term assistance either leads to, or is complemented by, long-term
 Spiritual Gifts and Secular Benefits 277

benefits. Those circumstances all involve power and autonomy. Economic

independence (cf. Sacks 1974), control of sexuality and fertility (cf. Bleier
1984), and ongoing extra-domestic ties (cf. Rosaldo 1974) all set the stage for
long-term advantages. Lack of power as embodied either by spirit possession
(the spirit rather than the woman is the real power) and illness tends to limit
secular advantages to the short term.
It is crucial to realize that both models I have presented are circular rather
than linear in nature: Sisterhood and autonomy are prerequisites for the devel-
opment of Model 2 religions, but when we look at the ways in which Model 2
religions benefit women, we see that autonomy and an ongoing collective
power base are among the chief benefits. Similarly, concerning Model i reli-
gions we see that healing and spirit possession provide individual women with
immediate assistance, but in that they individualize and mystify the true source
of women's suffering, they undermine the ability of the religion to offer collec-
tive empowerment.
We have come full circle. We have seen in what cultural contexts women's
religions arise, the content of women's religions, and how women's religions
serve women's interests. I will now bring together the themes that have
emerged from this study and present some more general conclusions.

Notes
1. Women's religions do not have a monopoly on serving women's interests. An
apparently sexist ideology can sometimes encourage women to take political action that
serves their interests as people and as women. This is certainly the case in Latin America
where many liberation movements are grounded in Catholic ideology.
2. Mary Baker Eddy publicly supported suffrage and legal equality for women, and
treated these issues as both political and theological ones. Christian Science in the late
nineteenth and early twentieth century saw in the women's rights movement a reflection
at the human level of the concept of divine womanhood explicated by Mary Baker Eddy.
Also many leaders of the movement for women's suffrage were active Spiritualists.
3. I moved Black Carib religion back and forth between the two columns before I
finally decided that it belongs with Model I. The ancestor rituals of Black Carib women
help individual women, for example, grandmothers, to get their sons and absent daugh-
ters to send them money. The so-called "love magic" rituals aid specific younger
women to negotiate their relationships with husbands and lovers. It is difficult to see
how either set of rituals affects women as a group.
4. "In the [Sudanese] village women do not need a 'support group' that is supplied
by belonging to the cult. Just about every woman in-the village is possessed, and even if
not they still attend ceremonies. Every woman is closely related to every other; they
live in each others' pockets, as they say. Support—moral, financial—is not hard to find"
(Boddy, personal communication 1992).
5. Some men claim that women deliberately seek to waste their husband's financial
resources to prevent his taking a second wife.
6. Ruth Landes has mentioned another way in which some women benefit from
participation in the cults: One purpose of the cults in Bahia was to give the women a
striking stage setting in which they look beautiful and exciting. The temple women had
no lack of male admirers, "It was a duty of the men to look on and admire" (1947, 144).
278 Priestess, Mother, Sacred Sister
7. I thank Judith Lorber for this insight.
8. Feminist Spirituality is the exception to this pattern. The North American mid-
dle class of the second half of the twentieth century is characterized by unusually high
geographic mobility. It would be rare indeed to find Spiritual Feminists who have
lifelong associations with one another.
9. Since the Shakers chose to live in isolated communities, it is legitimate to con-
sider the Shaker village as both the result of and the background to Shaker beliefs.
10. Sanday's findings are of a great deal of interest, yet because women's religions do
not tend to particularly worship female deities, more work needs to be done on the
triple connection between goddesses, women's religious status, and women's economic
status.
11. Peter Kunstadter does not make the connection explicit but it is there in his
writings: The other ethnic groups in Northern Thailand (Lua', Karen) do not have
matrilineal spirit cults run by women. In these groups men perform various sorts of
rituals and healing and divination. And, according to Kunstadter, the Northern Thai (as
opposed to the Lua' and Karen) have lower fertility rates, more women in prominent
public roles, and more women than men visiting the birth control clinic (1978, 209).
12. Okinawan men believed that a husband from the same village as the highest
priestesses (noro) would not live long. (Lower level priestesses are housewives or grand-
mothers and do marry.) Until recently, the marriage of noro was illegal and the children
illegitimate. A noro who wished to marry would continue living with or near her
brother, and her husband would come to "visit." During the last century noro have been
permitted to marry, and almost all do so. Still, before major festivals they sleep apart
from their husbands.
13. In recent times their economic independence has been eroded by loss of land, and
celibacy has declined.
14. An example here is a Spiritualist marriage ceremony found in the Twentieth
Century Formulary of Songs and Forms by W.C. Bowman (1907), cited in Ward (1990).
The ceremony carries no hint of men's control of women's sexuality or fertility, and the
introduction to the ceremony actually acknowledges that marriage is a product of
present political and social conditions, and "the mere fact of its having been made
venerable thus far by law, custom and religion is no conclusive evidence either of its
perfection or its perpetuity" (p. 130). The ceremony itself addresses the couple in these
extraordinarily egalitarian terms: "And recognizing equal rights and equal dignity on
the part of man and woman, the marriage vows you plight to each other are not
separate and distinct, but mutual and equal" (p. 132).
15. Digo spirit possession, as described by Gomm, differs considerably from zar.
Among the Digo, women who court chronic spirit possession are accorded low status
and considered to be mad. The zar cult, on the other hand, is made up of women who
initially were possessed involuntarily, but then chose to pursue further possession
experiences.
16. Field reports indicate that for Sande women the individual benefits outweigh the
pain of clitoridectomy.
Conclusion

CRITIQUE OF PREVIOUS
SCHOLARSHIP
I am indebted to the ethnographers and historians who collected and published
the information the examples in this book are based upon. I wish to reiterate
my thanks, before I continue with a necessary, feminist critique of a great deal
of what I have read.
I have already discussed how deprivation theory—one of the most common
theories used to "explain" women's religiosity—both reduces women's vibrant
existential quests to such banal factors as sexual frustration (cf. Kern 1981) and
fails to do justice to the rich variety of forms of women's spiritual expression. It
is particularly important to address the issue of sexuality, because over and over
in the literature I find scholars using it to rationalize women's religiosity. Erik
Erikson, for example, claims that there are deep existential implications to the
anatomical fact that in sexual intercourse the male penis penetrates the female
vagina.' It seems to me however, that with the exception of rape (and that is not
what Erikson is talking about), the similarities in the ways that men and women
experience sexual relations (and religion) are surely far more significant than the
differences. I find it absurd to interpret such patterns as the belief that suffering
is not necessary to the human condition and can and should be alleviated (see
Chapter 5) in terms of the anatomical organization of heterosexual intercourse.
And indeed, scholars who use physiological and anatomical models rarely (if
ever) look deeply at women's theological and philosophical stances.2
In light of the data presented in this book and the (to my mind unfortunate)
attraction of anatomical and physiological models, it is ironic that few scholars
interpret women's religiosity in terms of motherhood. Accounts of women's
religions, my fieldwork among Kurdish women, and my own personal experi-
ences as a mother all lead me to conclude that the loves, sorrows, and responsi-
bilities of mothering have a much profounder impact than the rather brief sex
act does on one's religious life. Evaluating the lack of attention given to mother-
hood in contrast to the surplus of attention given to sexual intercourse, I
280 Priestess, Mother, Sacred Sister

suspect that we have uncovered a lurking phallocentric obsession in western

scholarship.
It strikes me as significant that so many scholars have seen women's religios-
ity as standing in special need of psychological explanation. It also strikes me as
significant that so many scholars have seen goddesses as standing in special
need of psychological explanation. Does a pattern emerge? I am reminded of
the revolting joke told among gynecologists: Why are there so many gynecolo-
gists in the world? Because the very fact of being female is an illness! What
seems to underlie both the scholarly literature and the joke is a perception of
female as exceptional, other, not normal, and so in need of explanation.
To take one brief example, according to Alex Owen, "Spiritualist medi-
umship was both expressive of an inner struggle with the problem of feminin-
ity and instrumental in reconciling that tension" (1981, 209). I would like to
raise the question of why women's religious participation is so often explained
even by feminist scholars as a function of cultural definitions of womanhood,
while men's religious participation is almost never explained as a function of
cultural definitions of manhood!
It seems to me that previous scholars have devoted disproportionate atten-
tion to the personal histories of the leaders of women's religions. Indeed, for
several religions I easily found volumes purporting to "analyze" the foundress,
but I had trouble finding any detailed descriptions of theology or rituals. In
particular, scholars have been impressed by the apparent central role of illness
in the life stories of leaders. Yet, as I pointed out in Chapter n, it seems far
more likely that these stories, a form of literary device, are constructed to
"prove" the key beliefs of the religion—that suffering is not inescapable in this
life. One wonders whether so many modern scholars would explain Islamic
monotheism in terms of Mohammed's (hypothetical) illnesses, relationship
with his parents, or sexual frustrations.
In Chapter 7 I suggested that the morality of women's religions tends to be
invisible to Western scholars. When I read that Okinawans have little interest in
religious and philosophic speculation (Kerr 1958, 217), I am led to ask whether
the truth was that many of the scholars who reported about Okinawan culture
could not believe that mere women did in fact think about such things. At least
in the Okinawan case, given that men are not allowed to enter the sacred
groves, we can also assume that the majority of ethnographers who studied
Okinawan religion were denied access to key information. Yet we find no
mention in their published treatises of the possibility that they might have
missed out on essential elements of women's religious beliefs and rituals.
In a similar vein, in Chapter i I critiqued the use of the term "love magic" to
describe certain religious rituals performed by Black Carib women in order to
preserve family relationships. When I first began looking for women's reli-
gions, I found very few—because so few were called "religions" by western
scholars. Healing cults, magic, superstition—but not religion. Do we, as schol-
ars, get lost in our own terminology?
Feminist thinker Carol Christ has critiqued Mircea Eliade's "conception of
religion as providing release from the 'chaotic and dangerous flux of things'
 Conclusion 281
(I991, 91)- Eliade, Christ explains, valorizes the spiritual progress shown by
detachment from the "immediate" and the "concrete" (his words). As Christ
asks: Who says that the transcendent is "better" than the immanent? An issue
that must be addressed by feminist scholarship is the tendency among academics
to treat interest in the other world as more noble, virtuous, and estimable than
interest in this world. The models for "pure" spirituality whom many of us read
about in college were, for the most part, mystics who dramatically devalued the
concerns of this world. Once we fall into the trap of judging transcendent
religiosity to be "better" than immanent religiosity, it is but a short step to
believe—as have critics of Spiritualism—that women's religions are this-
worldly because women are less intelligent or less capable of abstract thinking
than men. (And indeed this is an easy trap because the so-called "great world
religions" are both male-dominated and other-worldly.) A feminist critique
must begin with a broad proclamation that recent studies of sex differences show
no gap between men's and women's intelligence (see Hyde 1990). Women's
religions are not characterized by a this-worldly orientation because women are
just not quite as smart as men. Whether or not my suggestion of interpreting the
this-worldly orientation of female-dominated religions in terms of women's
emotional and day-to-day experiences of mothering is correct, I am convinced
that this sort of socio-cultural interpretation is a step in the right direction.
As a feminist anthropologist I am uncomfortable writing a book about
women and religion until I see a few books that are explicitly about men—that
vuse uniquely male experiences to interpret men's religious lives.3 My fear in
writing this book is that readers will join me in exploring these extraordinary
women's religions, yet will abandon me when I ask how religion looks when
we treat the female as normative. For example, could it be that possession
trance is a normal and healthy part of human experience, but men have trouble
with it because they have a problem with relationships?

A LACK OF CONCLUSIONS
I opened this book by asking whether women's religions are in any way
"womanly"; whether there is anything distinctively feminine in the theologi-
cal, symbolic, or ritual content of female-dominated religions. I argued that I
would not expect religiosity to be an automatic function of sex: ritual prefer-
ences are not carried on the x or y chromosomes, and the ethnographic record
does not warrant our making many generalizations about women and men.
To recapitulate some of the non-universals we have seen: In most female-
dominated religions leaders are old women, except in Spiritualism where they
are young. Women's religions do not engage in missionizing, except for the
Shakers who dedicated a great deal of time and resources to seeking converts.
In most women's religions food rituals are very important and elaborated,
except in Christian Science. Animal sacrifice is anathema to women's religions,
except for the zdr cult. On the Ryukyu Islands almost all rituals are performed
by priestesses, the ritual role of laity is marginal, and rituals are highly standard-
ized and unemotional; among Spiritual Feminists all participants have crucial
282 Priestess, Mother, Sacred Sister

ritual roles, ceremonies tend to be innovative and personal, and demonstra-

tions of emotion are respected. Women's religions do not usually itemize and
codify rules of moral behavior, except for the Shakers who spelled out what
one should do and how one should do it every hour of the day and night.
When we turn our attention to the supernatural entities addressed in
women's religions, the differences among the religions leap off the page. In
some of the women's religions the deities insist that people behave morally to
other people; in other religions the deities are concerned primarily with ritual
proficiency. In some the deities are associated with famous myths and easily
identifiable personalities; in others they are not (and in the Northern Thai
matrilineal spirit cults it is near impossible to discern who or what the spirits
even are!). In some of the religions the deities are foreigners or peripheral; in
others they embody the very essence of the community and its native land.
Throughout this book I have demonstrated how women's religions enhance
and glorify interpersonal and familial bonds, and provide support and nurtur-
ance for ill and suffering women. Yet Tensho-Kotai-Jingu-Kyu teaches that the
most common cause of illness is innen—a Karma chain, fate, or bondage that is
transmitted from one individual to another through family links. In Tensho,
healing is effected by breaking chains of innenl Unlike in mainstream Japanese
culture in which a sick person is closely cared for and seen as especially righ-
teous, in Tensho the sick role loses all legitimacy. Members are told to stay
away from sick people so as not to catch the spirit causing the illness. In
contrast to most other women's religions, Tensho teaches that strong interper-
sonal bonds are not desirable, that sick people should not be nurtured, and that
illness does not lead to righteousness.4 Indeed, one of the first acts of the
foundress was to order her followers to throw away their ancestors' tablets and
break off with the Buddhist temples whose cemeteries held their family tombs
(Nakamura 1980, 182).5
The specific historical and cultural contexts of the religions go a great way
toward explaining their differences. Beginning with the global statement that
women's religions are rare, we can identify a variety of circumstances in which
they are more like to occur. First, when a new and powerful male-dominated reli-
gion conquers a tribal or village level society, women may end up in charge of
what used to be the religious system in which both men and women participat-
ed. This model is true, for example, of Korean shamanism. A second model
concerns male migration and resulting matrifocality. When men are absent a
great deal of the time (for a variety of possible reasons including war and
economic necessity) women become more independent in terms of religious
choice. This model is relevant to the zar cult and the religion of the Black
Caribs. In other cultural situations the mainstream, male-dominated religion
does not provide adequate explanations for and solutions to problems that are
particularly acute for women. If the society tolerates alternate religious
choices, women may become involved with "fringe" cults. For example, we
saw how some scholars explain women's dominance in nineteenth-century
North American Spiritualism as an outgrowth of women's dissatisfaction with
Calvinist belief that their dead babies were going to hell rather than heaven.
 Conclusion 283
None of the women's religions originated in a cultural vacuum. All devel-
oped within contexts of sexist cultures, and most emerged out of (or in reac-
tion to) non-female-dominated religions. The Afro-Brazilian religions are
syncretic cults in which Catholicism (clearly not dominated by women) is a
major component; Christian Science and Shakers identify with the Christian
tradition; Korean shamanism was not dominated by women in the past. It
could be argued that each of the women's religions reviewed in this book has
more in common with men's religions of the same time and place than with
women's religions located on different continents.
The dissimilar histories of the women's religions, together with their very
diverse cultural contexts, partially account for the lack of universal patterns.
But again, I wish to stress that I do not really think that the absence of universal
patterns needs explanation: We would hardly expect all men's religions to be
similar, so why should we expect all women's religions to be?
A final factor to consider is social roles versus sex as a shaper of religiosity.
We saw in Chapter 3 that recent studies have linked sympathy and nurturance
to the social role of caring for children. Since most of the themes we have
discovered in women's religions seem to be functions of social role (primary
responsibility for children), we should not be surprised to find that just as
patterns of parenting vary from culture to culture, so do patterns of women's
religiosity. In a similar vein, in an important study of spirit possession in
contemporary Egypt, Soheir Morsy found that "when other dimensions [be-
sides sex] of persons' identities are taken into account by reference to status
within the household, and the developmental cycle of the family, the signifi-
cance of power differentials beyond those associated with gender identity are
brought into focus. Thus, the higher frequency of UZY [spirit possession]
among the male category of brother of household head contrasts with the low
frequency of the illness among relatively powerful females, notably the moth-
ers of married sons" (1991, 205). In brief, I am left wondering which of the
patterns we have discovered in this book are gender linked and which are role
or status linked.

SUMMARY
In the Introduction I proposed two conditions sufficiently widespread to ac-
count for the patterns or motifs that we might uncover in women's religions
situated in dissimilar cultural contexts. These two conditions are patriarchy
and motherhood. Very few of the patterns we have uncovered can be inter-
preted primarily in terms of women's oppression in patriarchal cultures. As
I argued in Chapter 2, women's religions tend to occur in societies character-
ized by a relatively high level of autonomy for women. However, 1 wish to
emphasize that with the exception of Ryukyuan Island religion and Sande,
none of the religions discussed in this book is the official, mainstream, and/or
state-supported religion within its specific cultural context. This critical fact
surely reflects the subordinate status of women within these societies.
Most of the patterns that I have described in this book can be understood as
284 Priestess, Mother, Sacred Sister

functions of women's social role as mothers in societies in which that role is

granted both esteem and structurally recognized authority. Cross-culturally,
women give birth to children and are primary caretakers of children. This
cross-cultural reality leaves its stamp on all women's religions. Motherhood is
a central point of self-definition for women in the cultures with which I am
dealing in this book. In all these cultures, girls are socialized from a young age
to be mothers; in all these cultures motherhood shapes the lives of almost all
adult women. Even celibate Shaker and infertile Luvale women were part of
cultural contexts in which motherhood fashioned women's life experiences. If
there are women's religions in which motherhood is not a pivotal motif, I have
not been able to discover them in the available ethnographic or historical
literature.6
Motherhood does not have automatic consequences for women's religi-
osity—that is one of the themes that has repeatedly surfaced in this study—and
that in part explains why each of these religions is different from the others in
many significant ways. On the other hand, if one wishes to understand why
there are certain traits that characterize almost all women's religions, if one
wants to write about women and religion from a cross-cultural perspective, it is
helpful to look at motherhood, which is, after all, the human experience that
most clearly impacts on women differently than on men. My focus on mother-
hood is neither a literary device nor an ideological stance; rather, it is an accurate
and pervasive reference point for the study of women's religions.
The first trend that has emerged is that women's religions are more likely to
occur in societies that are matrilineal, matrilocal, or matrifocal than in
societies that are not. A number of factors account for this trend. To begin
with, matrifocality in the context of an otherwise patriarchal culture leads to
a situation of gender disjunction in which "cracks" may open up into which
women's religions can then fit. In matrifocal societies women's identity as
mother (an active identity) is emphasized more than their identities as wives
or daughters (passive identities). In matrilocal societies women tend to be
freer to meet together with other women. In matrilineal societies women are
more likely to control their own fertility. Not surprisingly, women's religions
tend to further enhance women's roles as mothers. This is the place to clarify
that Nancy Chodorow's ideas do not explain why women's religions arise;
they simply help us interpret the content of these religions. However, it does
seem relevant that women's religions cluster in matrifocal, matrilocal, and
matrilineal societies—that is, in situations in which all the implications of
women's mothering (as explained by Chodorow) are especially potent.
Drawing on Chodorow's theses, we have suggested that women—who
themselves were raised by women—develop an ease with and affinity for
interpersonal relationships. This is manifested in the ritual constellations em-
phasized in women's religions. Food rituals, ancestor rituals, and initiation
rituals all function to strengthen communal bonds.
Ongoing, intense, and intimate relationships between mothers and children
lead many women to seek religions that offer persuasive interpretations of
child death and provide effective and eclectic means of alleviating illness and
 Conclusion 285

suffering. Women's religions excel at both these tasks, often filling a gap left by
the male-dominated religions of their day.
Perhaps the most important implication of motherhood involves particu-
larism. Women as mothers love and nurture specific children, and they seek
out the divine within the profane "real" world of relationships and nature. Very
few women are able to simply pack up and leave their families to pursue ascetic
or other extraordinary religious paths.
Comfortable with other people, women are willing to meet their gods and
goddesses face to face and even share their bodies with divinities (in spirit
possession). Transcendent, monotheistic male deities have little meaning for
mothers who daily confront existential issues of birth, suffering, and death.
One of the clearest patterns to emerge from this study is really a "non-pattern":
none of these religions worships a single, all-powerful, male deity.
None of the women's religions teaches that women are inherently inferior.
Many of these religions explain why women are more suited for religious
leadership (they understand suffering better, their souls are softer, etc.). These
religions teach that men and women are essentially different, and that while
both male and female are necessary in this world, the two are often in tension.
In many cultures in which we find female-dominated religions women have
primary responsibility for raising children, yet are dependent on (all too often
erratic) economic contributions of men in order to raise children properly. A
philosophy of gender tension fits in well with women's actual experiences.
Focused on people, women's religions tend to be internally hierarchical
within the individual chapters or congregations—they care a great deal about
who does what to whom. Uninterested in transcendence and neglected by the
powers-that-be, women's religions display an aversion to centralization and
institutionalized doctrine. A lack of codified moral laws, however, does not
mean a lack of morality. Moral behavior in women's religions is situational,
rooted in particular relationships, yet perceived as having implications for the
entire community.
Women's religions serve women's secular interests. In situations in which
women enjoy economic independence, control over their own bodies, and
ongoing relationships with their fellow members, female-dominated religions
can obtain permanent, collective benefits for women. In situations in which
women are economically and socially isolated, female-dominated religions pro-
vide specific women with short-term support vis-a-vis specific men.
A last pattern, and the sole pattern that does seem to be truly ubiquitous in
women's religions, concerns militarism. None of the religions engages in any
sort of jihad or holy war. Although most of the religions welcome new mem-
bers, none carries out any sort of forcible conversion.7 The epitome of this
pattern is the Shakers. Despite their interest in drawing in new members, even
being raised in a Shaker community did not automatically translate into Shaker
membership. Moreover, the Shakers were pacifists who refused to fight in
America's wars. Can we deduce from this that women are more tolerant of
alternative beliefs, that women are less violent than men? Sara Ruddick and
many other feminist writers certainly think that is indeed the case. Personally, I
286 Priestess, Mother, Sacred Sister
am reluctant to make such grandiose claims. On the other hand, it may well be
that through the experiences of raising children women come to grips with
issues of power and lack of power in ways that most men do not. The dilemma
of mothering—that we possess far more power than we dare to use but that all
our power cannot protect or mold our children exactly as we wish—could
very well translate into an aversion to forcible proselytizing. Women, as subor-
dinate members of sexist societies, are unlikely to have access to, be socialized
into, or attracted by military power. Most women know all too well what it
means to be victims of violent coercion.
As we review these patterns and once again highlight the implications of
motherhood, it is critical to reiterate that the social rather than the biological
aspects of motherhood have emerged as most salient to women's religions.
Ornate fertility rituals, myths of mother goddesses who gave birth to the
world, and ceremonies that extol the wonders of lactation are almost totally
absent from women's religions. What does receive attention and elaboration in
these religions is women's social roles as nurturers and healers, women's rights
and responsibilities as primary childcare providers, women's emotional experi-
ences of pain at the illness and death of children, women's social ties with other
mothers, matrifocality, and women's proclivity for discovering the sacred
which is immanent in this everyday world of care and relationship.
Most readers of this book grew up in male-dominated religious traditions,
and probably have been tempted to compare their own religious traditions to
those presented here. I want to clarify that the patterns outlined above are not
unique to women's religions; they simply are easier to see in women's religions.
I surmise that many of these patterns can also be found, to some extent, in
women's religious activities within male-dominated religions.8 The difference
between the two types of situations is that in female-dominated religions these
patterns are institutionalized and esteemed. In male-dominated religions, on the
other hand, these patterns tend to be subsumed under the categories of "folk-
lore," "superstition," "syncretism," "heresy," or simply "ladies' auxiliary."

Notes
1. Erikson, like Freud, tries to define the "intrinsically feminine" and "intrinsically
masculine" essence of women and men in terms of the anatomical differences of their
genital organs. According to Erikson, men and women organize space according to the
morphology of their bodies, and this affects psychosocial development and orientation
to "inner" and "outer" space. Men and women follow the spacial morphology of their
bodies in their psychic, social, cultural, and religious lives. Women's womb is an "inner
productive space" that men do not have. Women tend to a fear of being left empty or
deprived of treasures or unfulfilled or drying up. This is manifested in their spiritual
experience.
2. Spirit possession and ecstatic mysticism are generally the favored phenomena
explained by proponents of physiological and anatomical models.
3. I am not referring to books implicitly about men. Almost all available studies of
religion are in fact about men and religion; it is just that the authors do not make that
fact clear and do not use maleness as an analytic category.
 Conclusion 287
4. A second "exception" are the nineteenth-century Sanctificationists, who strictly
regulated relationships among the all-female membership. According to Sally Kitch,
group by-laws forbid members to become especially close to particular co-members,
and even overly intensive mother-child ties were discouraged (1989,108).
5. According to Nakamura, however, maternal ties are not condemned in the way
that other familial ties are (1980).This fits in more closely with other women's religions.
6. The closest I can come to a women's religion that does not emphasize mother-
hood is the religion of the Ryukyu Islands—sisterhood is at least equally if not more
fundamental to Ryukyuan women's religious perceptions and roles.
7. Most women's religions even allow members to belong to other religions concur-
rently (see Chapters 5 and 12).
8. The one pattern rarely if ever found in male-dominated religions is strategies to
serve women's secular interests, particularly what I have called "Model 2"—long-term
structural benefits for women as a group.
APPENDIX A
Alphabetical Summaries
of Key Examples

vAfro-Brazilian Religions Syncretic religions combining elements of Afri-

can tribal religions, Amerindian religions, Catholicism, and Kardecism
(French Spiritism). Known by such names as Candomble, Umbanda,
Batuque, Xango, and Macumba. Two main features are curing and public
rituals in which mediums are possessed by spirits. Most people who attend
Afro-Brazilian rituals do so in order to be healed of illness or misfortune.
vBlack Carib Religion Main feature is numerous mourning and ancestora
rituals organized by old women. Black Caribs live in Central America. The
religious life of the Black Caribs of Belize is especially well documented.
Burmese Nat Religion The indigenous Burmese religion centers on the
appeasement of nats (spirits/gods). Nats are propitiated to prevent and cure
illness, at key stages in the agricultural cycle, at births, deaths, marriages,
vand Buddhist initiations. Elaborate myths are associated with nats. At large
festivals female nat shamans become possessed by their nat "husbands."
Smaller nat rituals are carried out by housewives.
Christian Science Founded in the United States in the nineteenth century by
Mary Baker Eddy. Main tenet: matter is illusory, therefore illness is illusory.
Healing comes about through study and prayer, which lead one to truly
understand the nonreality of matter.
Feminist Spirituality Movement A contemporary American religion that
has grown out of the secular feminist movement. Spiritual Feminists draw
on myths and symbols of many cultures, emphasize the spirituality of female
energy and bodies, relate to goddesses, and perform rituals that empower
women. Womanism is the name used by African-American feminists for
their own women's spirituality movement.
Korean Household Religion Korean women make food offerings to the
gods of their households and consult female shamans to divine why misfor-
tune has struck their households. At the elaborate kut ritual, shamans be-
come possessed by various gods, spirits, and ancestors.
Northern Thai Matrilineal Spirit Cults Northern Thai kinship organiza-
tion is matrilineal and matrilocal. Spirits that are understood as belonging to
the descent (or cult) group are propitiated at various ritual occasions. All the
matrilineal descendants of a founding ancestress constitute a cult group.
Ryukyuan Island Religion The official, mainstream religion of the Ryu-
290 Appendix A

kyu Islands (the main island is Okinawa) is totally dominated by women.

Priestesses (noro) at the household, family, and village levels (and formerly at
the state level) perform multitudinous rituals that seek the assistance of the
katni (gods). Shamans (yuta), who are also women, perform personal healing
and divination rituals.
Sande Secret Society In much of West Africa adolescent girls are initiated
into the women's Sande secret society (also known as Bundu). Sande initia-
tion teaches about childbirth and other skills and attitudes women need to
know. Sande chapters protect their members by punishing men and women
who infringe on Sande rules.
Shakers Ann Lee led a group of eighteenth-century English Shakers to Amer-
ica where they founded rural communes dedicated to agriculture, celibacy,
and worship. Shakers believed that Ann Lee had come to finish the work of
Jesus Christ, and that the Messianic era had already begun. Their worship
was characterized by ecstatic dancing and "shaking."
Spiritualism Beginning in the nineteenth-century United States, Spiritual-
ists believed that individuals continue to exist after death, and that the living
can communicate with the dead at seances. Mediums (mostly women) were
channels for this communication. Currently, Spiritualist groups exist in the
United States, Great Britain, and Mexico (where the emphasis is on healing
rather than communication with the spirit world).
Zar In parts of Africa and the Middle East, individuals (mostly women)
suffering from a variety of ailments are diagnosed as possessed by zar spirits.
In order to be healed and resume functioning, the individual must undergo
initiation into the zar cult. The zar is not exorcised; rather, the woman learns
to accommodate the zar. Members meet periodically at possession trance
rituals.
APPENDIX B
Alphabetical Summaries
of Auxiliary Examples

Beguines The movement of celibate Christian women known as Beguines

originated during the thirteenth century in the Low countries and the Rhine
Valley. These women lived in small communities and dedicated their lives to
charitable acts. For a variety of reasons, including the fact that they lived
outside of monasteries, they did not receive papal approval.
Black Spiritual Churches New Orleans Spiritual people trace their origins
to a woman named Leafy Anderson, who came from Chicago to New
Orleans in the 19205. Spiritual churches are linked to numerous older reli-
gious traditions, including Protestantism, Folk Catholicism, Pentecostal
Movement, nineteenth-century American Spiritualism, and Vodou practices
of the African diaspora. Spiritual people have a widespread eclectic belief
system. During church services, many Spiritual people may enter into trance
or other ecstatic states that include periods of dancing, spinning, uttering
"coos," violent seizures, writhing on the floor, etc. Recent studies show fifty
Spiritual churches in New Orleans today.
Guglielmites This thirteenth-century Christian "heresy" considered wom-
en to be the only hope for the salvation of humanity. They announced that
the third person of the trinity had been incarnated in the female Guglielma of
Milan and would establish a female-ruled church with female cardinals.
Churchmen viewed the Guglielmites as fools and Guglielma was con-
demned as a heretic.
Luvale of Zambia Women's ritual expertise is passed down through ma-
trilineal lines. The Luvale believe that female ancestors afflict their out-
marrying female descendants with illness and infertility to remind them of
their matrilineal obligations. Involvement in women's rituals and participa-
tion in the women's healing cult cures women afflicted by ancestral spirits.
Sanctificationists Also known as the Woman's Commonwealth and as the
Sanctified Sisters. This all-female group became an official community in
1890 in Belton, Texas. The women broke off from the Methodist church,
primarily because of their belief in celibacy and ecumenism. Although they
began as a worship group, so many of the women became estranged from
their husbands (because of their religious beliefs) that they found it necessary
to form a commune to ensure their economic survival. The group never had
more than 100 members. According to Sally Kitch, "Although they left no
292 Appendix B
record of a belief in either a female God or a female Christ, court testimo-
nies, interviews, and the group's constitution reveal a feminist approach not
only to the economic and social independence of women from men, and to
the personal autonomy of individual women, but also to a female-identified
source of spiritual authority" (1989, 14). Kitch notes that the group experi-
enced violent resistance to their celibate communism (1989, 63). Over the
years the religious focus of the group declined, and the economic focus took
precedence.
Tenrikyo This Japanese religion was founded in 1838 when God revealed
himself to the foundress Miki Hakayama. Tenrikyo teaches that sickness
originates from mental attitudes. God is referred to in Tenrikyo as "God the
Parent" and associated with parental love. Diseases are seen as an opportu-
nity to improve oneself because God informs humans of his true intentions
through disease. In recent years approximately half of the priests are
women, and two-thirds of the members are women.
Tensho-Kotai-Jingu-Kyu (also spelled Tensho-Kotai-Jingukyo) This
new Japanese religion was founded in 1945 when Mrs. Sayo Kitamura dis-
covered that she carried God in her abdomen. She became the spokeswoman
of a kami [deity], known as the Tensho Kotaijin (Absolute Almighty God of
the Universe). The Tensho cult draws on Buddhist, Shinto, Confucianist,
and Christian concepts. There is a worldwide membership of approximately
360,000 people, most of whom are Japanese.
Tetum Religion In this Indonesian society women are associated with the
sacred and men with the secular. Women preside over most rituals and most
ghosts are female. The clan shrine is attended by a priestess. Birth symbol-
ism is highly developed in Tetum religion; the first ancestors emerged from
the earth womb, and birth rituals are complex and lengthy.
Theosophy The Theosophical Society was founded in New York in 1875,
and in the i88os spread to London and Madras. Theosophists taught that the
human body is a temporal home for the eternal spirit and this spirit goes
through evolutionary stages and reincarnates itself both in male and female
guises. This evolutionary process involves the interaction of both matter and
spirit, yin and yang, male and female.
Vodou The leaders of Haitian Vodou are mediums who lend their bodies and
voices to a large and variegated pantheon of mostly African spirits. In rural
areas of Haiti most leaders were male; in contemporary urban America most
are female. The essence of Vodou is healing: healing between people and
healing between the living and the spirits. Rituals involve drumming and
possession trance.
REFERENCES KEY EXAMPLES

Afro-Brazilian Religions
Bastide, Roger. 1978 [1960], The African Religions of Brazil. Baltimore: Johns Hopkins
University Press.
Brown, Diana DeG. 1986. Umbanda Religion and Politics in Urban Brazil. Ann Arbor,
Michigan: UMI Reseach Press.
Brumana, Fernando Giobellina, and Elda Gonzales Martinez. 1989. Spirits from the
Margin: Umbanda in Sao Paulo. Upsala, Sweden: Acta Universitatis Upsaliensis.
Burdick, John. 1990. Gossip and Secrecy: Women's Articulation of Domestic Conflict
in Three Religions of Urban Brazil. Sociological Analysis 50(2): 153-170.
Carneiro, Edison. 1940. The Structure of African Cults in Bahia. Journal of American
Folklore 53:271-278.
Herskovits, Melville. 1943. The Southernmost Outposts of New World Africanisms.
American Anthropologist 45(4):495-5io.
Landes, Ruth. 19403. A Cult Matriarchate and Male Homosexuality. American Anthro-
pologist 35:386-397.
. I94ob. Fetish Worship in Brazil. Journal of American Folklore 53:261-270.
. 1947. City of Women. New York: Macmillan .
Leacock, Seth, and Ruth Leacock. 1972. Spirits of the Deep: A Study of an Afro-Brazilian
Cult. New York: Doubleday Natural History Press.
Lerch, Patricia. 1980. Spirit Mediums in Umbanda Evangelizada of Porto Alegre,
Brazil: Dimensions of Power and Authority. In A World of Women, ed. Erika
Bourguignon, 129-159. New York: Praeger.
. 1982. An Explanation for the Predominance of Women in the Umbanda Cults
of Porto Alegre, Brazil. Urban Anthropology n(2):237-26i.
Merrick, Thomas, and Marianne Schmink. 1983. Households Headed by Women and
Urban Poverty in Brazil. In Women and Poverty in the Third World, ed. Mayra
Buvinic, Margaret Lycette, and William McGreevey, 244-271. Baltimore: Johns
Hopkins University Press.
Neuhouser, Kevin. 1989. Sources of Women's Power and Status among the Urban Poor
in Contemporary Brazil. Signs i4(3):685~7O2.
Pressel, Esther. 1974. Umbanda Trance and Possession in Sao Paulo, Brazil. In Trance,
Healing, and Hallucination, ed. Felicitas Goodman, Jeannette Henney, and Esther
Pressel, 113-221. New York: John Wyley and Sons.
. 1980. Spirit Magic in the Social Relations between Men and Women (Sao
Paulo, Brazil). In A World of Women, ed. Erika Bourguignon, 107-127. New
York: Praeger.
294 References—Key Examples

Scheper-Hughes, Nancy. 1989. Death Without Weeping. Natural History 10:8-16.

Simpson, George E. 1978. Black Religions in the New World. New York: Columbia
University Press.
Willems, Emilio. 1953. The Structure of the Brazilian Family. Social Forces 31:339-345.

Black Caribs
Holland, O. Nigel. 1977. The Formation of a Colonial Society: Belize, from Conquest to
Crown Colony. Baltimore: Johns Hopkins University Press.
Bullard, M. Kenyon. 1974. Hide and Secrete: Women's Sexual Magic in Belize. The
Journal of Sex Research io(4):259-265.
Chernela, Janet M. 1991. Symbolic Inaction in Rituals of Gender and Procreation
among the Garifuna (Black Caribs) of Honduras. Ethos I9(i):$2-6j.
Coelho, Ruy Galvao de Andrade. 1955. The Black Carib of Honduras: A Study in Accul-
turation. Diss. Northwestern University.
Gonzalez, Nancie. 1983. Changing Sex Roles Among the Garifuna (Black Carib) and
their Implications for the Family. Jounal of Comparitive Family Studies 14(2):203-
213-
. 1969. Black Carib Household Structure. Seattle: University of Washington
Press.
Helms, Mary W. 1981. Black Carib Domestic Organization in Historical Perspective:
Traditional Origins of Contemporary Patterns. Ethnology 20(1)177-86.
Kerns, Virginia. 1980. Black Carib Women and Rites of Death. In Unspoken Worlds:
Women's Religious Lives in Non-Western Cultures, ed. Nancy A. Falk and Rita M.
Gross, 127-140. San Francisco: Harper and Row.
. 1983. Women and the Ancestors: Black Carib Kinship and Ritual. Urbana: Univer-
sity of Illinois Press.
. 19923. Preventing Violence Against Women: A Central American Case. In
Sanctions and Sanctuary: Cultural Perspectives on the Beating of Wives, ed. Dorothy
Ayers Counts, Judith K. Brown, andjacquelyn C. Campbell, 125-138. Boulder:
Westview Press.
. I992b. Female Control of Sexuality: Garifuna Women at Middle Age. In In Her
Prime: New Views of Middle-Aged Women, ed. Virginia Kerns and Judith K.
Brown, 94-111. Urbana: University of Illinois Press.
Munroe, Robert. 1980. Male Transvestism and the Couvade: A Psycho-Cultural Analy-
sis. Ethos 8(i):49-59.
Sanford, Margaret. 1974. A Socialization in Ambiguity: Child-Lending in a British
West Indian Society. Ethnology I3(4):393~4OO.
Staiano, Kathryn. 1981. Alternative Therapeutic Systems in Belize: A Semiotic Frame-
work. Social Science and Medicine 15^317-332.
Taylor, Douglas MacRae. 1951. The Black Carib of British Honduras. Viking Fund Publi-
cations in Anthropology, no. 17. New York: Wenner-Gren.

Burmese Nat Religion

Furnivall, I. C. S. 1911. Matriarchal Vestiges in Burma. Journal of the Burma Research
Society 1:15-30.
Goswami, M. C., and H. Kamkhenthang. 1972. Mother's Brother in Paite Society.
Man in India 52(i):2l-38.
Khaing, Mi Mi. 1962. Burmese Family. Bloomington: Indiana University Press.
 References—Key Examples 295

Nash, June C. 1966. Living with Nats: An Analysis of Animism in Burman Village
Social Relations. South East Asia Studies, Yale University. In Anthropological
Studies in Theravada Buddhism, 117—136. Cultural Report Scries, vol. 13. New
Haven: Yale University.
Nash, Manning. 1965. The Golden Road to Modernity. New York: Wiley.
. 1966. Ritual and Ceremonial Cycle in Upper Burma. South East Asian Studies,
Yale University. In Anthropological Studies in Theravada Buddhism, 97-115. Cul-
tural Report Series, vol. 13. New Haven: Yale University.
Spiro, Melford. 1967. Burmese Supernaturalism. Englewood Cliffs, N.J.: Prentice-Hall.
. 1971. Buddhism and Society. London: George Allen and Unwin.
. 1977. Kinship and Marriage in Burma. Berkeley: University of California Press.

Christian Science

Becker, Carl B. 1990. Religious Healing in ipth Century 'New Religions': The Cases of
Tenrikyo and Christian Science. Religion 20:199-215.
Eddy, Mary Baker. 1903. Science and Health, with a Key to the Scriptures. Boston: Pub-
lished by the Trustees under the will of Mary Baker Eddy.
Fox, Margery. 1989. The Socioreligious Role of the Christian Science Practitioner. In
Women as Healers: Cross-Cultural Perspectives, ed. Carol Shepherd McClain, 98-
114. New Brunswick: Rutgers University Press.
Gottschalk, Stephen. 1973. The Emergence of Christian Science in American Religious Life.
Berkeley: University of California Press.
Peel, Robert. 1966. Mary Baker Eddy. New York: Hold, Rinehart and Winston.
Setta, Susan. 1977. Denial of the Female—Affirmation of the Feminine: The Father-
Mother God of Mary Baker Eddy. In Beyond Androcentrism: New Essays on Women
and Religion, ed. Rita M. Gross. Missoula, Montana: Scholars Press.
Stark, Rodney, William Sims Bainbridge, and Lori Kent. 1981. Cult Membership in the
Roaring Twenties: Assessing Local Receptivity. Sociological Analysis 4(2):i37—
162.
Trevett, Christine. 1984. Women, God and Mary Baker Eddy. Religion 14:143-153.
Williams, Peter W. 1980. Popular Religion in America. Englewood Cliffs, New Jersey:
Prentice-Hall Inc.
Wilson, Bryan R. 1961. Sects and Society. Berkeley: University of California Press.

Feminist Spirituality

Adams, Bert N. 1969. Kinship in an Urban Setting. Chicago: Markham Publishing

Company.
Adler, Margot. 1979. Drawing Down the Moon. Boston: Beacon Press.
Baruch, Grace, and Rosalind Barnett. 1983. Adult Daughters' Relationships with their
Mothers. Journal of Marriage and the Family 45:601-606.
Bernard, Jessie. 1975. Women, Wives, Mothers: Values and Options. Chicago: Aldine
Publishing Company.
Budapest, Zsuzsanna E. 1989. The Grandmother of Time. San Francisco: Harper and
Row.
Christ, Carol. 1979. Why Women Need the Goddess: Phenomenological, Psychologi-
cal, and Political Reflections. In Womanspirit Rising, ed. Carol P. Christ and Judith
Plaskow, 273-287. San Francisco: Harper and Row.
296 References—Key Examples

. 1989. Rethinking Theology and Nature. In Weaving the Visions, ed. Judith
Plaskow and Carol Christ, 314-325. San Francisco: Harper and Row.
. 1990. Symbols of Goddess and God in Feminist Theology. In The Book of the
Goddess Past and Present, ed. Carl Olson, 231-251. New York: Crossroad.
Daly, Mary. 1978. Gyn/Ecology: The Metaethics of Radical Feminism. Boston: Beacon
Press.
Finley, Nancy J. 1991. Political Activism and Feminist Spirituality. Sociological Analysis
52(4):349-362.
Gimbutas, Marija. 1974. The Cods and Goddesses of Old Europe, 7000-3500 B.C.: Myths,
Legends and Cult Images. London: Thames and Hudson.
Iglehart, Hallie. 1982. Expanding Personal Power Through Meditation. In The Politics
of Women's Spirituality, ed. Charlene Spretnak, 294-304. Garden City, New
York: Anchor.
Morton, Nelle. 1989. The Goddess as Metaphoric Image. In Weaving the Visions: New
Patterns in Feminist Spirituality, ed. Judith Plaskow and Carol P. Christ, 111-118.
New York: Harper and Row.
Porterfield, Amanda. 1987. Feminist Theology as a Revitalization Movement. Sociologi-
cal Analysis 48(3):234~244.
Rabuzzi, Katheryn Allen. 1988. Motherself: A Mythic Analysis of Motherhood. Blooming-
ton and Indianapolis: Indiana University Press.
Sanders, Cheryl J. 1989. Christian Ethics and Theology in Womanist Perspective. Jour-
nal of Feminist Studies in Religion 5(2):83—91.
Sjoo, Monica, and Barbara Mor. 1987. The Great Cosmic Mother: Rediscovering the
Religion of the Earth. San Francisco: Harper and Row.
Solovitch, Sara. 1990. Witchful Thinking. Philadelphia Inquirer, 28 October, 29-32.
Spretnak, Charlene, Editor. 1982. Politics of Women's Spirituality. Garden City, New
York: Doubleday.
Starhawk. 1985. Ethics and Justice in Goddess Religion. In Women's Consciousness,
Women's Conscience: A Reader in Feminist Ethics, ed. Barbara Hilkert Andolsen,
Christine E. Gudorf, and Mary D. Pellauer, 193-200. San Francisco: Harper
and Row.
. 1987. Truth or Dare. New York: Harper and Row.
Stone, Merlin. 1976. When God Was a Woman. New York: Dial Press.
Walker, Alice. 1982. The Color Purple. New York: Washington Square Press.
Williams, Delores S. 1989. Womanist Theology. In Weaving the Visions: New Patterns in
Feminist Spirituality, ed. Judith Plaskow and Carol P. Christ, 179-186. New
York: Harper and Row.

Korean Shamanism and Household Religion

Harvey, Youngsook Kim. 1976. The Korean Mudang as a Household Therapist. In
Culture-Bound Syndromes, Ethnopsychiatry and Alternative Therapies, ed. William P.
Lebra, 189-198. Honolulu: University of Hawaii Press.
. 1979. Six Korean Women: The Socialization of Shamans. American Ethnological
Society Monographs. Vol. 65. St. Paul: West Publishing Company.
. 1980. Possession Sickness and Women Shamans in Korea. In Unspoken Worlds:
Women's Religious Lives in Non-Western Cultures, ed. Nancy A. Falk and Rita M.
Gross, 41-52. San Francisco: Harper and Row.
Janelli, Roger, and Dawnhee Yim Janelli. 1982. Ancestor Worship and Korean Society.
Stanford: Stanford University Press.
 References—Key Examples 297
Kendall, Laurel. 1983. Korean Ancestors: From the Woman's Side. In Korean Women,
View from the Inner Room, ed. Laurel Kendall and Mark Peterson, 97-112. New
Haven: East Rock Press.
. 1984. Korean Shamanism: Women's Rites and a Chinese Comparison. In Reli-
gion and the Family in East Asia, ed. George A. De Vos and Takao Sofue, 57-73.
Berkeley: University of California Press.
. 1985. Shamans, Housewives, and Other Restless Spirits: Women in Korean Ritual
Life. Honolulu: University of Hawaii Press.
. 1989. Old Ghosts and Ungrateful Children: A Korean Shaman's Story. In
Women as Healers: Cross Cultural Perspectives, ed. Carol Shepherd McClain, 138-
156. New Brunswick: Rutgers University Press.
Kim, Yung-Chung, Editor. 1982. Women of Korea: A History from Ancient Times to 1945.
Seoul, Korea: Ewha Women's Press.
Koh, Hesung Chun. 1984. Religion and Socialization of Women in Korea. In Religion
and the Family in East Asia, ed. George A. De Vos and Takao Sofue, 237-257.
Berkeley: University of California Press.
Lee, Kwang Kyu. 1984. Family and Religion in Traditional and Contemporary Korea.
In Religion and the Family in East Asia, ed. George A. De Vos and Takao Sofue,
185-199. Berkeley: University of California Press.
Mattielli, Sandra, Editor. 1977. Virtues in Conflict: Tradition and the Korean Woman Today.
Seoul: Royal Asiatic Society.
Suh, David Kwang-sun. 1989. Shamanism: The Religion of Han. Paper presented at the
First Interfaith Dialogue; Asian Women's Resource Center for Culture and Theol-
ogy. Kuala Lumpur, Malaysia.
Wilson, Brian. 1983. The Korean Shaman: Image and Reality. In Korean Women, View
from the Inner Room, ed. Laurel Kendall and Mark Peterson, 113-128. New
Haven: East Rock Press.

Ryukyuan Island Religion

Anzai, Shin. 1976. Newly-Adopted Religions and Social Change on the Ryukyu Islands
(Japan). Social Compass 23(i):57-7O.
Haring, Douglas G. 1953. The Noro Cult of Amami Oshima: Divine Priestesses of the
Ryukyu Islands. Sociologus 3(2): 108-121.
. 1964. Chinese and Japanese Influences. In Ryukyuan Culture and Society, ed.
Allan H. Smith, 39-55. Honolulu: University of Hawaii Press.
Hopkins, Garland Evans. 1951. Okinawa Religion. International Review of Missions
40:179-184.
Ito, Mikiharu. 1966. Rice Rites in Japan Proper and the Ryukyus: A Comparative
Study. In Folk Cultures of Japan and East Asia, 37-55. Monumenta Nipponica
Monographs, no. 25. Tokyo: Sophia University Press.
Kamata, Hisako. 1966. Daughters of the Gods: Shaman Priestesses in Japan and
Okinawa. In Folk Cultures of Japan and East Asia, 56-73. Monumenta Nipponica
Monographs, no. 25. Tokyo: Sophia University Press.
Kerr, George H. 1958. Okinawa: The History of an Island People. Rutland, Vermont:
Charles E. Tuttle Company.
Kreiner, Josef. 1968. Some Problems of Folk-Religion in the Southwest Islands
(Ryukyu). In Folk Religion and the Worldview in the Southwestern Pacific, ed. N.
Matsumoto and T. Mabuchi, 101-118. Tokyo: Keio University.
Lebra, William P. 1966. Okinawan Religion. Honolulu: University of Hawaii Press.
298 References—Key Examples

Mabuchi, Toichi. 1964. Spiritual Predominance of the Sister. In Ryukyuan Culture and
Society, ed. Allan H. Smith, 79-91. Honolulu: University of Hawaii Press.
. 1968. Toward the Reconstruction ofRyukyuan Cosmology. In Folk Religion
and the Worldview in the Southwestern Pacific, ed. N. Matsumoto and T. Mabuchi,
119-140. Tokyo: Keio University.
. 19763. Optional Cult Group Affiliation Among the Puyama and the Miyako
Islanders. In Ancestors, ed. William H. Newell, 91-103. Hague: Mouton.
. I976b. A Note on Ancestor Worship in 'Cognatic' Societies. In Ancestors, ed.
William H. Newell, 105-117. Hague: Mouton.
Matsui, Hiroko, Kazuya Horike, and Hideshi Ohashi. 1980. Rorschach Responses of
Okinawan Shamans 'Yuta'. Tohuku Psychologica Folia 39(i-4):6i~78.
Matsuzono, Makio. 1976. A Note on the Enshrinement of Ancestral Tablets at Zamami
Island, Okinawa. In Ancestors, ed. William H. Newell, 231-240. Hague: Mouton.
McCume, Shannon. 1975. The Ryiikyu Islands. Harrisburg: Stackpole Books.
Noguchi, Takenori. 1966. The Japanese Kinship System. In Folk Cultures of Japan and
East Asia, 16-36. Monumenta Nipponica Monographs, no. 25. Tokyo: Sophia
University Press.
Ohashi, Hideshi, Shinsuke Sakumichi, and Kazuya Horike. 1984. A Social Psychologi-
cal Study of Okinawan Shamanism (i)—Approach and Some Findings. Tohoku
Psychologica Folia 43(i-4):66-79.
Ota, Yoshinobu. 1989. Creating the Experience of a Sacred Elite: An Examination of
the Concept of Shiji in Southern Ryukyuan Religion. Contributions to Southeast
Asian Ethnography 8:111-125.
Ouwehand, C. 1985. Hateruma: Socio-Religious Aspects of a South-Ryukyuan Island Cul-
ture. Leiden: E. J. Brill.
Sasaki, Kokan. 1984. Spirit Possession as an Indigenous Religion in Japan and Okinawa.
In Religion and the Family in East Asia, ed. George A. De Vos and Takao Sofue,
75-84. Berkeley: University of California Press.
Spencer, Robert Steward. 1931. The Noro, or Priestesses of Loo Choo. Transactions of
the Asiatic Society of Japan 8:94-112.
Yoshida, Teigo. 1989. Lecture. International Cultural Center for Youth. Jerusalem, 23
March.
. 1990. The Feminine in Japanese Folk Religion: Polluted or Divine? In Unwrap-
ping Japan, ed. Eyal Ben-Ari, Brian Moeran, and James Valentine, 58-77. Hono-
lulu: University of Hawaii Press.

Sande Secret Society

Bellman, Beryl L. 1979. The Social Organization of Knowledge in Kpelle Ritual. In
New Religions of Africa, ed. Bennetta Jules-Rosette, 39-56. Norwood, New Jer-
sey: Ablex Publishing Corporation.
Bledsoe, Caroline H. 1980. Women and Marriage in Kpelle Society. Stanford, California:
Stanford University Press.
Cosentino, Donald. 1982. Defiant Maids and Stubborn Farmers: Tradition and Invention in
Mende Story Performance. Cambridge: Cambridge University Press.
Harris, W T., and Harry Sawyerr. 1968. The Springs of Mende Relief and Conduct.
Freetown: Sieera Leone University Press.
Hoffer, Carol P. [MacCormack]. 1972. Mende and Sherbro Women in High Office.
Canadian Journal of African Studies 4(2):i5i-i64.
. 1974. Madam Yoko: Ruler of the Kpa Mende Confederacy. In Women, Culture,
 References—Key Examples 299

and Society, ed. Michelle Z. Rosaldo and Louise Lamphere, 173-187. Stanford:
Stanford University Press.
Jedrej, M.C. 1976. Structural Aspects of a West African Secret Society. Journal of Anthro-
pological Research 32:234-245.
Lamp, Frederick. 1988. Heavenly Bodies: Menses, Moon, and Rituals of License
among the Temne of Sierra Leone. In Blood Magic, ed. Thomas Buckley and
Alma Gottlieb, 210-231. Berkeley: University of California Press.
MacCormack, Carol. 1982. Health, Fertility and Birth in Moyamba District, Sierra
Leone. In Ethnography of Fertility and Birth, ed. Carol MacCormack, 115-139.
London: Academic Press.
. 1977. Biological Events and Cultural Control. Signs 3:93-100.
. 1979. Sande: The Public Face of a Secret Society. In The New Religions of Africa,
ed. Bennetta Jules-Rosette, 27-37. Norwood, New Jersey: Ablex Publishing
Corporation.
Margai, M. A. S. 1948. Welfare Work in a Secret Society. African Affairs 47:227-230.
Richards, J. V. O. 1973. The Sande and Some of the Forces that Inspired its Creation or
Adoption with Some References to the Poro. Journal of Asian and African Studies
8(1-2)169-77.
Rosen, David M. 1981. Dangerous Women: "Ideology," "Knowledge" and Ritual
Among the Kono of Eastern Sierra Leone. Dialectical Anthropology 6:151-163.
. 1983. The Peasant Context of Feminist Revolt in West Africa. Anthropological
Quarterly 56:35-43.
Tonkin, Elizabeth. 1983. Women Excluded? Masking and Masquerading in West Af-
rica. In Women's Religious Experience, ed. Pat Holden, 163-174. London: Croom
Helm.

Shakers
Bainbridge, William Sims. 1982. Shaker Demographics, 1840-1900: An Example of the
Use of U.S. Census Enumeration Schedules. Journal for the Scientific Study of
Religion 2i(4):352-365.
Bednarowski, Mary Farrell. 1980. Outside the Mainstream: Women's Religion and
Women Religious Leaders in Nineteenth-Century America. Journal of the Ameri-
can Academy of Religion 48(2)^07-231.
Brewer, PriscillaJ. 1986. Shaker Communities, Shaker Lives. Hanover: University Press
of New England.
. 1992. "Tho" of the Weaker Sex": A Reassessment of Gender Equality among
the Shakers. Signs i7(3):6o9-635.
Campion, Nardi Reeder. 1990 [1976], Mother Ann Lee: Morning Star of the Shakers.
Hanover: University Press of New England.
Desroche, Henri. 1971. The American Shakers: From Neo-Christianity to Presocialism.
1955. Amherst: University of Massachusetts Press.
Foster, Lawrence. 1981. Religion and Sexuality. New York: Oxford University Press.
Garrett, Clarke. 1987. Spirit Possession and Popular Religion: From the Camisards to the
Shakers. Baltimore: Johns Hopkins University Press.
Humez, Jean M. 1992. "Ye Are My Espistles": The Construction of Ann Lee Imagery in
Early Shaker Sacred Literature. Journal of Feminist Studies in Religion 8(i):83 —104.
Kern, Louis J. 1981. An Ordered Love: Sex Roles and Sexuality in Victorian Utopias—the
Shakers, the Mormons, and the Oneida Community. Chapel Hill: University of
North Carolina Press.
3oo References—Key Examples

Kitch, Sally L. 1989. Chaste Liberation: Celibacy and Female Cultural Status. Urbana and
Chicago: University of Illinois Press.
Klein, Janice. 1979. Ann Lee and Mary Baker Eddy: The Parenting of New Religions.
Journal of Psychohistory 6(3):36i~375.
Marini, Stephen A. 1982. Radical Sects of Revolutionary New England. Cambridge: Har-
vard University Press.
Procter-Smith. 1985. Women in Shaker Community and Worship: A Feminist Analysis of the
Uses of Religious Symbolism. Lewiston, New York: Edwin Mellen.
Stein, Stephen J. 1992. The Shaker Experience in America. New Haven: Yale University
Press.
Whitson, Robley Edward, Editor. 1983. The Shakers: Two Centuries of Spiritual Reflec-
tion. New York: Paulist Press.

Spiritualism

Braude, Ann. 1985. Spirits Defend the Rights of Women: Spiritualism and Changing
Sex Roles in Nineteenth-Century America. In Women, Religion and Social
Change, ed. Yvonne Yazbeck Haddad and Ellison Banks Findley, 419-432. Al-
bany: State University of New York Press.
. 1989. Radical Spirits: Spiritualism and Women's Rights in Nineteenth-Century Amer-
ica. Boston: Beacon Press.
Dye, Nancy S., and Daniel B. Smith. 1986. Mother Love and Infant Death, 1750-1920.
Journal of American History 73:329-353.
Finkler, Kaja. 19853. Symptomatic Differences Between the Sexes in Rural Mexico.
Culture, Medicine and Psychiatry 9:27-57.
. I985b. Spiritualist Healers in Mexico. South Hadley, Massachusetts: Bergin and
Garvey.
. 1986. The Social Consequence of Wellness: A View of Healing Outcomes from
Micro and Macro Perspectives. International Journal of Health Services i6(4):62j-
642.
Macklin, June. 1974. Belief, Ritual, and Healing: New England Spiritualism and
Mexican-American Spiritism Compared. In Religious Movements in Contemporary
America, ed. Irving Zaretsky and Mark Leone, 383-417. Princeton: Princeton
University Press.
. 1977. A Connecticut Yankee in Summerland. In Case Studies in Spirit Possession,
ed. Vincent Crapanzano and Vivian Garrison, 41-85. New York: John Wiley.
Moore, R. Laurence. 1977. In Search of White Crows: Spiritualism, Parapsychology, and
American Culture. New York: Oxford University Press.
Nelson, Geoffrey K. 1969. Spiritualism and Society. London: Routledge and Kegan Paul.
Owen, Alex. 1981. The Darkened Room. London: Virago Press.
Skultans, Vieda. 1974. Intimacy and Ritual: A Study of Spiritualism, Mediums and Groups.
London: Routledge and Kegan Paul.
. 1983. Mediums, Controls and Eminent Men. In Women's Religious Experience,
ed. Pat Holden, 15-26. London: Croom Helm.
Smith-Rosenberg, Carroll. 19853. The Female World of Love and Ritual. In Disorderly
Conduct, 53-76. New York: Oxford University Press.
. I985b. The Cross and the Pedestal: Women, Anti-Ritualism, and the Emer-
gence of the American Bourgeoisie. In Disorderly Conduct, 167-181. New York:
Oxford University Press.
 References—Key Examples 301

. 19850. Puberty to Menopause: The Cycle of Femininity in Nineteenth-Century

America. In Disorerly Conduct, 182-196. New York: Oxford University Press.
Ward, Gary L., Editor. 1990. Spiritualism II: The Movement. New York: Garland
Publishing.

Northern Thai Matrilineal Spirit Cults

Cohen, Paul T. 1984. Are the Spirits Cults of Northern Thailand Descent Groups?
Mankind i4(4):293-299.
, and Gehan Wijeyewardene. 1984. Introduction. Mankind i4(4):249-262.
Cook, Nerida. 1981. The Position of Women in Thai Buddhism: The Parameters of Religious
Recognition. Master's Thesis. Australian National University.
Davis, Richard. 1974. Tolerance and Intolerance of Ambiguity in Northern Thai Myth
and Ritual. Ethnology 13:1-24.
. 1984. Muang Matrifocality. Mankind I4(4):263-27I.
Eberhardt, Nancy. 1988. Introduction. In Gender, Power, and the Construction of the Moral
Order: Studies from the Thai Periphery, ed. Nancy Eberhardt, 3-12. Madison,
Wisconsin: Center for Southeast Asian Studies.
Ferguson, John P. 1982. The Great Goddess Today in Burma and Thailand: An Explora-
tion of Her Symbolic Relevence to Monastic and Female Roles. In Mother Wor-
ship, ed. James J. Preston, 283-303. Chapel Hill: University of North Carolina
Press.
Hale, Ann. 1979. A Reassessment of Northern Thai Matrilineages. Mankind 12(2)1138-
150.
. 1984. The Search for Jural Rule: Women in Southeast Asia—the Northern Thai
Cults in Perspective. Mankind i4(4):33O-338.
Hanks, Jane Richardson. 1963. Maternity and its Rituals in Bang Chan. Cornell Thailand
Project Interim Reports Series, vol. 6. Ithaca, New York: Department of Asian
Studies, Cornell University.
Hanks, Lucien M. 1983. The Yuan or Northern Thai. In Highlanders of Thailand, ed.
John McKinnon and Wanat Bhruksasri, 101-112. Kuala Lumpur: Oxford Uni-
versity Press.
, and Jane Richardson Hanks. 1963. Thailand: Equality between the Sexes. In
Women in the New Asia, ed. Barbara E. Ward, 424-451. Amsterdam: UNESCO.
Heinze, Ruth-Inge. 1988. Trance and Healing in Southeast Asia Today. Bangkok: White
Lotus Co.
Irvine, Walter. 1984. Decline of Village Spirit Cults and Growth of Urban Spirit Medi-
umship: The Persistence of Spirit Beliefs, the Position of Women and Moderniza-
tion. Mankind 14(4)^ 15-324.
Keyes, Charles F. 1984. Mother or Mistress but Never a Monk: Buddhist Notions of
Female Gender in Rural Thailand. American Ethnologist n(2):223-24i.
Kirsch, A. Thomas. 1982. Buddhism, Sex-Roles and the Thai Economy. Occasional
Paper, Center for Southeast Asian Studies, ed. Penny Van Esterik. Northern
Illinois University.
. 1985. Text and Context: Buddhist Sex Roles/ Culture of Gender Revisited.
American Ethnologist 12(2^302-320.
Kunstadter, Peter. 1978. Do Cultural Differences Make Any Difference: Choice Points
in Medical Systems Available in Northwestern Thailand. In Culture and Healing
in Asian Societies, ed. Arthur Kleinman, Peter Kunstadter et al., 185—217. Cam-
bridge, Massachusetts: Schenkman Publishing Company.
3O2 References—Key Examples

McMorran: M. V. 1984. Northern Thai Ancestral Cults: Authority and Aggression.

Mankind I4(4):3o8-3i4.
Mougne, Christine. 19843. Women, Fertility and Power in Northern Thailand. Paper
presented at the International Conference on Thai Studies. Bangkok, 22-24
August.
. 198413. Spirit Cults and Matrifocality in Northern Thailand: Demographic
Perspective. Mankind i4(4):3OO-3oy.
Piker, Steven. 1972. The Problem of Consistency in Thai Religion. Journal for the
Scientific Study of Religion 11:211-229.
Potter, Shulamith. 1977. Family Life in a Northern Thai Village: A Study in the Structural
Significance of Women. Berkeley: University of California Press.
Rabibhadana, Akin. 1984. Kinship Marriage and the Thai Social System. Paper Pre-
sented at the Conference on Marriage Determinants and Consequences. Pattaya
City, Thailand: Institute for Population and Social Research, Mahidol Univer-
sity, 30 May-3 June.
Tambiah, S. J. 1970. Buddhism and the Spirit Cults in North-East Thailand. Cambridge:
Cambridge University Press.
Tanabe, Shigeharu. 1991. Spirits, Power, and the Discourse of Female Gender: The Phi
Meng Cult in Northern Thailand. In Thai Constructions of Knowledge, ed. Manas
Chitakasem and Andrew Turton, 183—212. London: School of Oriental and
African Studies.
Tantiwiramanond, Darunee and Shashi Pandey. 1987. The Status and Role of Thai
Women in the Pre-Modern Period: A Historical and Cultural Perspective. So-
journ 2(i):i25-i49.
Turton, Andrew. 1976. A Case Study of Jurat and Political Structures at the Village Level and
their Twentieth Century Transformations. Diss. University of London.
. 1984. People of the Same Spirit: Local Matrikin Groups and their Cults. Man-
kind i4(4):272-28s.
Van Esterik, Penny. 1982. Interpreting a Cosmology: Guardian Spirits in Thai Bud-
dhism. Anthropos 77:1-15.
Wijeyewardene, Gehan. 1970. The Still Point and the Turning Point: Towards the
Structure of Northern Thai Religion. Mankind j(4):24j-2^<,.
. 1977. Matriclans or Female Cults: A Problem in Northern Thai Ethnography.
Mankind n(i):i9-25.
. 19843. Northern Thai Succession and the Search for Matriliny. Mankind
i4(4):286-292.
. I984b. Talking about Merit and Karma. In Honouring E.F.C. Ludowyk: Felicita-
tion Essays, ed. Percy Colin-Thome and Ashley Halpe, 314-333. Sri Lanka:
Tisara Prakasakayo Ltd.
. 1986. Place and Emotion in Northern Thai Ritual Behaviour. Thailand: Pandora.

Zar
Boddy, Janice. 1988. Spirits and Selves in Northern Sudan: The Cultural Therapeutics
of Possession and Trance. American Ethnologist I5(i):4—27.
. 1989. Wombs and Alien Spirits: Women, Men, and the Zar Cult in Northern
Sudan. Madison: University of Wisconsin Press.
. 1992. Subversive Kinship: The Role of Spirit Possession in Negotiating Social
Place in Rural Northern Sudan. Paper presented at the Annual Meeting of the
Amercian Anthropological Association. San Francisco, 3 December.
 References—Key Examples 303

Constantinides, Pamela. 1982. Women's Spirit Possession and Urban Adaptation in the
Muslim Northern Sudan. In Women United, Women Divided, cd. Patricia Caplan
and Janet Bujra, 185—205. Bloomington: Indiana University Press.
. 1985. Women Heal Women: Spirit Possession and Sexual Segregation in a
Muslim Society. Social Science and Medicine 2i(6):685-692.
Ferichou, Sophie. 1991. The Possession Cults of Tunisia: A Religious System Function-
ing as a System of Reference and a Social Field for Performing Actions. In
Women's Medicine: The Zar-Bori Cult in Africa and Beyond, ed. I. M. Lewis,
Ahmed Al-Safi, and Sayyid Hurreiz, 209—218. Edinburgh: Edinburgh Univer-
sity Press.
Kahana, Yael. 1985. The Zar Spirits, a Category of Magic in the System of Mental
Health Care in Ethiopia. International Journal of Social Psychiatry 3i(2):i25—142.
Kennedy, John G. 1978. Nubian Ceremonial Life. Berkeley: University of California
Press.
Kenyon, Susan M. 1991. The Story of a Tin Box: Zar in the Sudanese Town of Sennar.
In Women's Medicine: The Zar-Bori Cult in Africa and Beyond, ed. I. M. Lewis,
Ahmed Al-Safi, and Sayyid Hurreiz, 100-117. Edinburgh: Edinburgh Univer-
sity Press.
Last, Murray. 1991. Spirit Possession as Therapy: Bori among non-Muslims in Nigeria.
In Women's Medicine: The Zar-Bori Cult in Africa and Beyond, ed. I. M. Lewis,
Ahmed Al-Safi, and Sayyid Hurreiz, 49—63. Edinburgh: Edinburgh University
Press.
Lewis, I. M. 1969. Spirit Possession in Northern Somaliland. In Spirit Mediumship and
Society in Africa, ed. John Beattie and John Middleton, 188-219. London: Rout-
ledge and Kegan Paul.
. 1990. Spirits at the House of Childbirth. Review of Women and Alien Spirits, by
Janice Boddy. Times Literary Supplement, June 1—7, 590.
. 1991. Introduction: Zar in Context: The Past, the Present and Future of an
African Healing Cult. In Women's Medicine: The Zar-Bori Cult in Africa and Be-
yond, ed. I. M. Lewis, Ahmed Al-Safi, and Sayyid Hurreiz, 1-16. Edinburgh:
Edinburgh University Press.
Messing, Simon D. 1958. Group Therapy and Social Status in the Zar Cult of Ethiopia.
American Anthropologist 60:1120-1126.
Morsy, Soheir A. 1991. Spirit Possession in Egyptian Ethnomedicine: Origins, Com-
parison, and Historical Specificity. In Women's Medicine: The Zar-Bori Cult in
Africa and Beyond, ed. I. M. Lewis, Ahmed Al-Safi, and Sayyid Hurreiz, 189-
208. Edinburgh: Edinburgh University Press.
Natvig, Richard. 1987. Oromos, Slaves, and the Zar Spirits: A Contribution to the His-
tory of the Zar Cult. International Journal of African Historical Studies 2o(4):669-689.
. 1988. Liminal Rights and Female Symbolism in the Egyptian Zar Possession
Cult. Numen 35(i):57-68.
Rahim, S.I. 1991. Zar among Middle-Aged Female Psychiatric Patients in the Sudan. In
Women's Medicine: The Zar-Bori Cult in Africa and Beyond, ed. I. M. Lewis,
Ahmed Al-Safi, and Sayyid Hurreiz, 137-146. Edinburgh: Edinburgh Univer-
sity Press.
Saunders, Lucic Wood. 1977. Variants in 7<dr Experience in an Egyptian Village. In Case
Studies in Spirit Possession, ed. Vincent Crapanzano and Vivian Garrison, 177-
192. New York: John Wiley and Sons.
REFERENCES GENERAL

Albanese, Catherine. 1990. Nature Religion in America: From the Algonkian Indians to the
New Age. Chicago: University of Chicago Press.
Anastassiadou, Iphigenie. 1976. Two Ceremonies of Travesty in Thrace: The Day of
Babo and the Caloyeri. L'homme i6(2-3):69-ioi.
Argyle, Michael, and Benjamin Beit-Hallahmi. 1975. The Social Psychology of Religion.
London: Routledge and Kegan Paul.
Baer, Hans A. 1993. The Limited Empowerment of Women in Black Spiritual
Churches: An Alternative Vehicle to Religious Leadership. Sociology of Religion
S4(i):6s-82.
Barfoot, Charles, and Gerald Sheppard. 1980. Prophetic vs. Priestly Religion: The
Changing Role of Women Clergy in Classical Pentecostal Churches. Review of
Religious Research 22(i):2-iy.
Barstow, Anne Llewellyn. 1986. Joan of Arc: Heretic, Mystic, Shaman. Studies in Women
and Religion, vol. 17. Lewiston, New York: Edwin Mellen Press.
Bart, Pauline. 1983. Review of Chodorow's The Reproduction of Mothering. In Mothering:
Essays in Feminist Theory, ed. Joyce Trebilcot, 147-152. Savage, Maryland: Row-
man and Littlefield.
Basilov, V. 1976. Shamanism in Central Asia. In The Realm of the Extra-Human, ed.
Agehananda Bharati, 149-157. Hague: Mouton.
Beattie, John, and John Middleton. Editors. 1969. Spirit Mediumship and Society in
Africa. London: Routledge and Kegan Paul.
Beck, Lois. 1980. The Religious Lives of Muslim Women. In Women in Contemporary
Muslim Societies, ed. Jane I. Smith, 27—60. London: Associated University Presses.
Beech, Mary H. 1982. The Domestic Realm in the Lives of Hindu Women in Calcutta.
In Separate Worlds: Studies of Purdah in South Asia, ed. Hanna Papanek and Gail
Minault, 110-138. Columbia, Mo.: South Asian Books.
Beit-Hallahmi, Benjamin. 1992. Despair and Deliverance: Private Salvation in Contempo-
rary Israel. Albany: State University of New York Press.
Belenky, Mary Field, Blythe McVicker Clinchy, Nancy Rule Goldberger, and Jill
Mattuck Tarule. 1986. Women's Ways of Knowing: The uevflopment of Self, Voice,
and Mind. New York: Basic Books, Inc.
Bell, Diane. 1983. Daughters of the Dreaming. Melbourne: McPhee Gribble/ George
Allen and Unwin.
Bennett, Lynn. 1983. Dangerous Wives and Sacred Sisters: Social and Symbolic Roles of
High-Caste Women in Nepal. New York: Columbia University Press.
Berger, Iris. 1976. Rebels or Status-Seekers? Women as Spirit Mediums in East Africa.
306 References—General
In Women in Africa, ed. Nancy Hafkin and Edna Bay, 157-181. Stanford: Stan-
ford University Press.
Bilu, Yoram. 1987. Dybbuk Possession and Mechanisms of Internalization and External-
ization: A Case Study. In Projection, Identification, Projectile Identification, ed. Jo-
seph Sandier, 163-178. Madison: International Universities Press.
Bleier, Ruth. 1984. Science and Gender. New York: Pergamon.
Boserup, Esther. 1970. Women's Role in Economic Development. London: George Allen
and Unwin.
Bourguignon, Erika. 1976. Possession and Trance in Cross-Cultural Studies of Mental
Health. In Culture-Bound Syndromes, Ethnopsychiatry, and Alternative Therapies,
ed. William P. Lebra, 47-55. Mental Health Research in Asia and the Pacific, vol.
4. Honolulu: University of Hawaii Press.
, Anna Bellisari, and Susan McCabe. 1983. Women, Possession Trance Cults,
and the Extended Nutrient-Deficiency Hypothesis. American Anthropologist 45:
413-416.
Breidenbach, Paul. 1979. The Woman on the Beach and the Man in the Bush: Leader-
ship and Adepthood in the Twelve Apostles Movement of Ghana. In The New
Religions of Africa, ed. Bennetta Jules-Rosette, 99-126. Norwood, New Jersey:
Ablex Publishing Company.
Briffault, Robert. 1931. The Mothers. New York: Macmillan Company.
Brock, Sebastian B., and Susan Ashbrook Harvey. 1987. Holy Women of the Syrian
Orient. Berkeley: University of California Press.
Brown, Audrey L. 1988. Women and Ritual Authority in Afro-American Baptist
Churches of Rural Florida. Anthropology and Humanism Quarterly I3(i):2-io.
Brown, Judith K. 1963. A Cross-Cultural Study of Female Initiation Rites. American
Anthropologist 65:837-853.
Brown, Karen McCarthy. 19893. Afro-Caribbean Spirituality: A Haitian Case Study. In
Healing and Restoring: Health and Medicine in the World's Religious Traditions, ed.
Lawrence E. Sullivan, 255-285. London: Macmillan.
. I989b. Women's Leadership in Haitian Vodou. In Weaving the Visions, ed. Judith
Plaskow and Carol P. Christ, 226-234. San Francisco: Harper and Row.
. 1991. Mama Lola: A Vodou Priestess in Brooklyn. Berkeley: University of Califor-
nia Press.
Buckley, Thomas, and Alma Gottlieb. Editors. 1988. Blood Magic. Berkeley: University
of California Press.
Burfield, Diana. 1983. Theosophy and Feminism: Some Explorations in Nineteenth
Century Biography. In Women's Religious Experience, ed. Pat Holden, 27-56.
London: Croom Helm.
Bynum, Caroline Walker. 1984. Women's Stories, Women's Symbols: A Critique of
Victor Turner's Theory of Liminality. In Anthropology and the Study of Religion,
ed. Frank Reynolds and Robert Moore, 105-125. Chicago: Center for the Scien-
tific Study of Religion.
. 1987. Holy Feast and Holy Fast. Berkeley: University of California Press.
Campbell, Debra. 1989. Hannah Whitall Smith (1832-1911): Theology of the Mother-
hearted God. Signs I5(i):79-ioi.
Chanfrault-Duchet, Marie-Francoise. 1991. Narrative Structures, Social Models, and
Symbolic Representations in the Life-Story. In Women's Words: The Feminist
Practice of Oral History, ed. Sherna Berger Gluck and Daphne Patai, 77-92. New
York: Routledge.
Chesler, Phyllis. 1989. Sacred Bond: The Legacy of Baby M. New York: Vintage Books.
 References—General 307

Chodorow, Nancy. 1971. Being and Doing: A Cross-Cultural Examination of the

Socialization of Males and Females. In Woman in Sexist Society, ed. Vivian
Gornick and Barbara Moran, 259-291. New York: Basic Books.
. 1974. Family Structure and Feminine Personality. In Women, Culture, and Soci-
ety, ed. Michelle Z. Rosaldo and Louise Lamphere, 43-66. Stanford: Stanford
University Press.
. 1978. The Reproduction of Mothering. Berkeley: University of California Press.
. 1981. Reply by Nancy Chodorow. Signs 6(3):5OO~5i4.
. 1989. Feminism and Psychoanalytic Theory. New Haven: Yale University Press.
Christ, Carol P. 1991. Mircea Eliade and the Feminist Paradigm Shift. Journal of Feminist
Studies in Religion 7(2)^5-94.
Christian, William. 1972. Person and God in a Spanish Valley. New York: Seminar Press.
Clarke, Edith. 1966. My Mother Who Fathered Me: A Study of the Family in Three Selected
Communities in Jamaica. London: George Allen and Unwin.
Claus, Peter J. 1975. The Siri Myth and Ritual: A Mass Possession Cult of South India.
Ethnology I4(i):47~58.
Cloudsley, Ann. 1984. Women ofOmdurman: Life, Love and the Cult of Virginity. London:
Ethnographica.
Colson, Elizabeth. 1969. Spirit Possession among the Tonga of Zambia. In Spirit Medi-
umship and Society in Africa, ed. John Beattie and John Middleton, 69-103. Lon-
don: Routledge and Kegan Paul.
Combs-Schilling, M. E. 1989. Sacred Performances: Islam, Sexuality, and Sacrifice. New
York: Columbia University Press.
Conarton, Sharon, and Linda Kreger Silverman. 1988. Feminine Development Through
the Life Cycle. In Feminist Psychotherapies: Integration of Therapeutic and Feminist
Systems, ed. Mary Ann Dutton-Douglas and LenoreE. Walker, 37-67. Norwood,
New Jersey: Ablex Publishing Corporation.
Crapanzano, Vincent. 1987. Spirit Possession. In Encyclopedia of Religion, vol. 14, ed.
Mircea Eliade, 12-19. New York: Macmillan.
Csordas, Thomas J. 1987. Health and the Holy in African and Afro-American Spirit
Possession. Social Science and Medicine 24(1):i-n.
Cucchiari, Salvatore. 1988. 'Adapted for Heaven': Conversion and Culture in Western
Sicily. American Ethnologist I5(3):4i7-44i.
Dally, Ann. 1982. Inventing Motherhood: The Consequence of an Ideal. London: Burnett
Press.
Danforth, Loring M. 1982. Death Rituals of Rural Greece. Princeton: Princeton Univer-
sity Press.
Davis, Susan S. 1982. Patience and Power: Women's Lives in a Moroccan Village. Cam-
bridge, Massachusetts: Schenkman Publishing Company.
Davis, Winston. 1980. Dojo: Magic and Excorcism in Modern Japan. Stanford: Stanford
University Press.
de Vaus, David, and Ian McAllister. 1987. Gender Differences in Religion: A Test of the
Structural Location Theory. American Sociological Review 52:472-481.
Desan, Suzanne. 1990. Reclaiming the Sacred: Lay Religion and Popular Politics in Revolu-
tionary France. Ithaca: Cornell University Press.
Divale, William. 1984 [1974]. Matrilocal Residence in Pre-literate Societies. Ann Arbor:
UMI Research Press.
Doyle, Patricia Martin. 1974. Women and Religion: Psychological and Cultural Implica-
tions. In Religion and Sexism, ed. Rosemary Radford Ruether, 15—40. New York:
Simon and Schuster.
308 References—General
Dyregrov, Atle. 1990. Parental Reactions to the Loss of an Infant Child: A Review.
Scandinavian Journal of Psychology 31:266—280.
, and Stig Berge Matthieson. 1987. Similarities and Differences in Mothers' and
Fathers' Grief Following the Death of an Infant. Scandinavian Journal of Psychology
28(i):i-i5.
Ebaugh, Helen Rose Fuchs, and Sharron Lee Vaughn. 1984. Ideology and Recruitment
in Religious Groups. Review of Religious Research 26(2): 148-157.
Ehrenreich, Barbara, and Deirdre English. 1973. Complaints and Disorders: The Sexual
Politics of Sickness. New York: Feminist Press.
Eisenstadt, Shmuel. 1982. The Axial Age: The Emergence of Transcendental Visions
and the Rise of Clerics. Archives Europeennes de Sociologie 23(i):294-3i4.
Eliade, Mircea. 1958. Rites and Symbols of Initiation. New York: Harper and Row.
Erikson, Erik. 1968. The Development of Ritualization. In The Religious Situation, ed.
Donald Cutler, 711-733. Boston: Beacon Press.
Esquivel, Julia. 1987. Christian Women and the Struggle for Justice in Central America.
In Speaking of Faith, ed. Diana L. Eck and Devaki Jain, 22-32. Philadelphia: New
Society Publishers.
Finnegan, Ruth. 1988. Literacy and Orality: Studies in the Technology of Communication.
Oxford: Basil Blackwell.
Frazer, J.G. 1911-1915. The Golden Bough: A Study in Magic and Religion. London:
Macmillan.
Fromm, Erich, and Michael Maccoby. 1970. Social Character in a Mexican Village: A
Socio-Psychoanalytic Study. Englewood Cliffs: Prentice-Hall.
Frymer-Kensky, Tikva. 1992. In the Wake of the Goddess: Women, Culture, and the Biblical
Transformation of Pagan Myth. New York: Free Press.
Geertz, Clifford. 1969. Religion as a Cultural System. In Anthropological Approaches to
the Study of Religion, ed. Michael Banton. London: Tavistock Publications.
Giles, Linda L. 1987. Possession Cults on the Swahili Coast: A Re-examination of
Theories of Marginality. Africa 57(2):234-257.
Gill, Brita. 1985. A Ministry of Presence. In Women Ministers, ed. Judith L. Weidman,
89-106. San Francisco: Harper and Row.
Gilligan, Carol. 1982. In a Different Voice. Cambridge: Harvard University Press.
Gluckman, Max. 1963. Order and Rebellion in Tribal Africa. London: Cohen and West.
Goldenberg, Naomi R. 1979. Changing of the Gods. Boston: Beacon Press.
Gomm, Roger. 1975. Bargaining from Weakness: Spirit Possession on the South Kenya
Coast. Man (N.S.) 10:530-545.
Goody, Jack. 1977. The Domestication of the Savage Mind. Cambridge: Cambridge Uni-
versity Press.
Gough, Kathleen E. 1968. The Nayars and the Definition of Marriage. In Marriage,
Family, and Residence, ed. P. Bohannan and J. Middleton, 49-71. Garden City,
New York: Natural History Press.
Graham, Hilary. 1976. The Social Image of Pregnancy: Pregnancy as Spirit Possession.
The Sociological Review 24(2):29i-3o8.
Gutmann, David. 1965. Women and the Conception of Ego Strength. Merrill-Palmer
Quarterly of Behavior and Development n(3):229-240.
. 1977. The Cross-Cultural Perspective: Notes Toward a Comparative Psychol-
ogy of Aging. In Handbook of the Psychology of Aging, ed. James Birren and K.
Schaie, 302-326. New York: Van Nostrand Reinhold.
Hagestad, Grunhild O. 1984. Women in Intergenerational Patterns of Power and Influ-
ence. In Social Power and Influence of Women, ed. Liesa Stamm and Carol D. Ryff,
 References—General 309

37-55. A A AS Selected Symposium, no. 96. Boulder, Colorado: Westview

Press.
Hales, Mary A. 1990. Motherhood and Sex Role Development. In Woman-Defined
Motherhood, ed. Jane Price Knowles and Ellen Cole, 227-243. New York: Har-
rington Park Press.
Hardacre, Helen. 1986. Kurozumikyo and the New Religions of Japan. Princeton: Prince-
ton University Press.
Harris, Marvin. 1985. Culture, People, Nature: An Introduction to General Anthropology.
4th Ed. New York: Harper and Row.
Harrison, Jane. 1955. Prolegomena to the Study of Greek Religion. New York: Meridian
Books.
Hayes, Rose Oldfield. 1975. Female Genital Mutilation, Fertility Control, Women's
Role, and the Patrilineage in Modern Sudan: A Functional Analysis. American
Ethnologist 2:617-633.
Haywood, Carol Lois. 1983. The Authority and Empowerment of Women among
Spiritualist Groups. Journal for the Scientific Study of Religion 22(2)11^7-166.
Heeren, John, Donald Linsey, and Marylee Mason. 1984. The Mormon Concept of
Mother in Heaven: A Sociological Account of its Origin and Development.
Journal for the Scientific Study of Religion 23(4):396~4ii.
Hibbard, Judith H., and Clyde R. Pope. 1987. Women's Roles, Interest in Health and
Health Behavior. Women and Health i2(2):6j-&4.
Hicks, David. 1976. Tetum Ghosts and Kin: Fieldwork in an Indonesian Community. Palo
Alto: Mayfield Publishing Company.
. 1984. A Maternal Religion: The Role of Women in Tetum Myth and Ritual. Mono-
graph Series on Southeast Asia. Northern Illinois University: Center for South-
east Asian Studies.
Holden, Pat. 1983. Introduction. In Women's Religious Experience, ed. Pat Holden, 1-14.
London: Croom Helm.
Hori, Ichiro, 1968. Folk Religion in Japan. Chicago: University of Chicago Press.
Horton, Robin. 1969. Types of Spirit Possession in Kalabari Religion. In Spirit Possession
and Society in Africa, ed. John Beattie and John Middleton, 14-49. London:
Routledge and Kegan Paul.
Houtart, Francois. 1977. Therevada Buddhism and Political Power—Construction and
Deconstruction of its Ideological Function. Social Compass 24(2-3):2O7-246.
Hutch, Richard A. 1984. Types of Women Religious Leaders. Religion 14:155-173.
Hyde, Janet Shibley. 1990. Meta-analysis and the Psychology of Gender Differences.
Signs i6(i):55-73.
Infante, Teresita R. 1975. The Woman in Early Philippines and Among the Cultural Minori-
ties. Manila: Unitas Publications.
Jacobs, Claude F. 1989. Spirit Guides and Possession in the New Orleans Black Spiritual
Churches. Journal of American Folklore 102:45-67.
, and Andrew J. Kaslow. 1991. The Spiritual Churches of New Orleans. Knoxville:
University of Tennessee Press.
Jamzadeh, Laal, and Margaret Mills. 1986. Iranian Sofreh: From Collective to Female
Ritual. In Gender and Religion: On the Complexity of Symbols, ed. Caroline Walker
Bynum, Steven Harrell and Paula Richman, 23-65. Boston: Beacon Press.
Jay, Nancy. 1985. Sacrifice as Remedy for Having Been Born of Woman. In Immacu-
late and Powerful, ed. Clarissa Atkinson, Constance Buchanan and Margaret
Miles, 283-309. Harvard Women's Studies in Religion Series. Boston: Beacon
Press.
3io References—General
Johnson, Miriam. 1988. Strong Mothers, Weak Wives: The Search for Gender Equality.
Berkeley: University of California Press.
Jonte-Pace, Diane. 1987. Object Relations Theory, Mothering, and Religion: Toward a
Feminist Psychology of Religion. Horizons 14(25:310-327.
. 1992. Situating Kristeva Differently: Psychoanalytic Readings of Women and
Religion. In Body-texts in Julia Kristeva: Religion, Woman, Psychoanalysis, ed. Da-
vid Crownfield. Albany: State University of New York Press.
Jules-Rosette, Bennetta. 1979. Changing Aspects of Women's Initiation in Southern
Africa: An Exploratory Study. Canadian Journal of African Studies I3(3):389~4O5.
Kanter, RosabethMoss. 1977. Men and Women of the Corporation. New York: Basic Books.
Kaplan, Alexandra, and Janet Surrey. 1984. The Relational Self in Women: Developmen-
tal Theory and Public Policy. In Women and Mental Health Policy, ed. Lenore
Walker, 79-94. Sage Yearbooks in Women's Policy Studies, vol. 9. Beverly Hills:
Sage.
Kehoe, Alice, and Dody Giletti. 1981. Women's Preponderance in Possession Cults:
The Calcium-Deficiency Hypothesis Expanded. American Anthropologist 83(3):
549-550.
Keller, John W., Dave Sherry, and Chris Piotrowski. 1984. Perspectives on Death: A
Developmental Study. Journal of Psychology 116:137-142.
Kerner, Karen. 1976. The Malevolent Ancestor: Ancestral Influence in a Japanese Reli-
gious Sect. In Ancestors, ed. William H. Newell, 205-217. Hague: Mouton.
Kittay, Eva Feder. 1983. Womb Envy: An Explanatory Concept. In Mothering: Essays in
Feminist Theory, ed. Joyce Trebilcot, 94-128. Savage, Maryland: Rowman and
Littlefield.
Klatch, Rebecca E. 1987. Women and the New Right. Philadelphia: Temple University
Press.
Knowles, Jane Price. 1990. Woman-Defined Motherhood. In Woman-Defined Mother-
hood, ed. Jane Price Knowles and Ellen Cole, 1-7. New York: Harrington Park
Press.
Koss-Chioino, Joan. 1992. Women as Patients, Women as Healers: Mental Health Care and
Traditional Healing in Puerto Rico. Boulder: Westview Press.
Kraemer, Ross S. 19803. Ecstasy and Possession: Women of Ancient Greece and the Cult
of Dionysus. In Unspoken Worlds: Women's Religious Lives in Non-Western Cultures,
ed. Nancy A. Falk and Rita M. Gross, 53-69. San Francisco: Harper and Row.
. igSob. The Conversion of Women to Ascetic Forms of Christianity. Signs
6(2):298-307.
. 1989. Monastic Jewish Women in Greco-Roman Egypt: Philo Judaeus on the
Therapeutrides. Signs I4(2):342-37O.
. 1992. Her Share of the Blessings: Women's Religions Among Pagans, Jews, and
Christians in the Greco-Roman World. New York: Oxford University Press.
Ladislav, Holy. 1988. Gender and Ritual in an Islamic Society: The Berti of Darfur. Man
(N.S.) 23:469-487.
Lambek, Michael. 1981. Human Spirits: A Cultural Account of Trance in Mayotte. Cam-
bridge: Cambridge University Press.
Laub Coser, Rose. 1981. On The Reproduction of Mothering: A Methodological Debate.
Signs 6(3):487-492.
Lebra, Takie Sugiyama. 1970. Religious Conversion as a Breakthrough for Trans-
culturation: A Japanese Sect in Hawaii. Journal for the Scientific Study of Religion
9(3):i8i-i96.
 References—General 311

. 1972. Religious Conversion and Elimination of the Sick Role: A Japanese Sect
in Hawaii. In Transcultural Research in Mental Health, ed. William P. Lebra, 282-
292. Mental Health Research in Asia and the Pacific, vol. 2. Honolulu: Univer-
sity of Hawaii Press.
Lehman, Edward C., Jr. 1993. Gender and Ministry Style: Things Not What They
Seem. Sociology of Religion 54(1): 1-12.
Leiderman, P. Herbert, and Gloria F. Leiderman. 1974. Affective and Cognitive Conse-
quences of Polymatric Infant Care in the East Africa Highlands. In Minnesota
Symposia on Child Psychology, vol. 8, ed. Ann Pick, 81-110. Minneapolis: Univer-
sity of Minnesota Press.
Leifer, Myra. 1980. Psychological Effects of Motherhood: A Study of First Pregnancy. New
York: Praeger Publishers.
Lerner, Gerda. 1986. The Creation of Patriarchy. New York: Oxford University Press.
Lerner, Harriet G. 1989. Women in Therapy. New York: Harper and Row.
Lewis, I.M. 1975 [1971]. Ecstatic Religion. Harmondsworth, Middlesex, England: Pen-
guin Books.
—i . 1986. Religion in Context: Cults and Charisma. Cambridge: Cambridge Univer-
sity Press.
. 1989. Ecstatic Religion. 2nd Ed. London: Routledge.
Lightfoot-Klein, Hanny. 1989. Prisoners of Ritual. New York: Harrington Park Press.
Lincoln, Bruce. 1981. Emerging from the Chrysalis: Studies in Rituals of Women's Initiation.
Cambridge: Harvard University Press.
Lopata, Helen. 1987. Women's Family Roles in Life Course Perspective. In Analyzing
Gender, ed. Beth Hess and Myra Marx, 381—407. Newbury Park, California:
Sage Publications.
Lorber, Judith. 1981. On The Reproduction of Mothering: A Methodological Debate.
Signs 6(3):482-486.
Macaulay, Jacqueline. 1985. Adding Gender to Aggression Research: Incremental or
Revolutionary Change? In Women, Gender, and Social Psychology, ed. Virginia
O'Leary, Rhoda Kesler Unger, and Barbara Strudler Wallston, 191-224. Hill-
side, New Jersey: Lawrence Erlbaum Associates.
Martin, Joann. 1990. Motherhood and Power: The Production of a Women's Culture of
Politics in a Mexican Community. American Ethnologist I7(3):47O-49O.
Martin, Patricia Yancey. 1990. Rethinking Feminist Organizations. Gender and Society
4(2):i82-2o6.
May, L. Carlyle. 1954. The Dancing Religion: A Japanese Messianic Sect. Southwestern
Journal of Anthropology 10:119-137.
McClain, Carol Shepherd, Ed. 1989. Reinterpreting Women in Healing Roles. In
Women as Healers: Cross-Cultural Perspectives, 1-19. New Brunswick: Rutgers
University Press.
McConville, Frances. 1988. The Birth Attendant in Bangladesh. In The Midwife Chal-
lenge, ed. Sheila Kitzinger, 134-154. London: Pandora Press.
McFarland, Neil. 1967. Rush Hour of the Gods. San Francisco: Harper and Row.
McGilvray, Dennis B. 1988. Sex, Repression, and Sanskritization in Sri Lanka. Ethos
i6(2):99-i27.
Meckel, Richard. 1989. Educating A Ministry of Mothers: Evangelical Maternal Asso-
ciations, 1815-1860. In Major Problems in American Women's History, ed. Mary
Beth Norton, 139-145. Lexington, Massachusetts: Heath and Company.
Mernissi, Fatima. 1977. Women, Saints, and Sanctuaries. Signs 3(i):101-112.
312 References—General

Miller-McLemore, Bonnie J. 1992. Epistemology or Bust: A Maternal Feminist Knowl-

edge of Knowing. Journal of Religion 72(2)1229—247.
Moberg, David. 1962. The Church as a Social Institution: The Sociology of American
Religion. Englewood Cliffs, New Jersey: Prentice-Hall.
Murdock, George P. 1949. Social Structure. New York: Macmillan.
. 1967. Ethnographic Atlas. Pittsburgh: University of Pittsburgh Press.
Nakamura, Kyoko Motomochi. 1980. No Women's Liberation: The Heritage of a
Woman Prophet in Modern Japan. In Unspoken Worlds: Women's Religious Lives in
Non-Western Cultures, ed. Nancy Auer Falk and Rita M. Gross, 174-190. San
Francisco: Harper and Row.
Nathanson, Constance. 1979. Sex, Illness and Medical Care. In Health, Illness and Medi-
cine, ed. Gary L. Albrecht and Paul C. Higgins, 16-40. Chicago: Rand McNally.
Neel, Carol. 1989. The Origins of the Beguines. Signs I4(2):32i~34i.
Nelsen, Hart M., Neil Cheek Jr., and Paul Au. 1985. Gender Differences in Images of
God. Journal for the Scientific Study of Religion 24(4) ^96-402.
Neumann, Erich. 1963. The Great Mother: An Analysis of the Archetype. Princeton:
Princeton University Press.
Noddings, Nel. 1984. Caring: A Feminine Approach to Ethics and Moral Education. Berke-
ley: University of California Press.
Norbeck, Edward. 1970. Religion and Society in Modern Japan. Houston: Rice University.
O'Flaherty, Wendy Doniger. 1980. Women, Androgynes, and Other Mythical Beasts. Chi-
cago: University of Chicago Press.
Ochshorn, Judith. 1981. The Female Experience and the Nature of the Divine. Blooming-
ton: University of Indiana Press.
Offner, Clark, and Henry Van Straelen. 1963. Modern Japanese Religions. Leiden: E. J.
Brill.
Ong, W. J. 1982. Orality and Literacy: The Technologizing of the Word. London and New
York: Methuen.
Ortner, Sherry. 1974. Is Female to Male as Nature Is to Culture? In Women, Culture and
Society, ed. Michelle Zimbalist Rosaldo and Louise Lamphere, 67-88. Stanford:
Stanford Univeristy Press.
Paige, Karen Ericksen, and Jeffrey M. Paige. 1981. The Politics of Reproductive Ritual.
Berkeley: University of California Press.
Fernet, Henry. 1982. Masks and Women: Toward a Reappraisal. History of Religions
22(i):45-59.
Pine, Vanderlyn R., and Carolyn Brauer. 1986. Parental Grief: A Synthesis of Theory,
Research, and Intervention. In Parental Loss of a Child, ed. Therese A. Rando,
59-96. Champaign, Illinois: Research Press Company.
Pitts, Walter. 1989. If You Caint Get the Boat, Take a Log: Cultural Reinterpretation in
the Afro-Baptist Ritual. American Ethnologist i6(2):279-293.
Poole, Fitzjohn Porter. 1981. Transforming 'Natural' Woman: Female Ritual Leaders and
Gender Ideology among Bimin-Kugkusmin. In Sexual Meanings, ed. Sherry Ort-
ner and Harriet Whitehead, 80-115. Cambridge: Cambridge University Press.
Potter, Jack M. 1974. Cantonese Shamanism. In Religion and Ritual in Chinese Society,
ed. Arthur P. Wolf, 207-231. Stanford: Stanford University Press.
Powell, Gary N. 1988. Women and Men in Management. Newbury Park, California: Sage
Publications.
Preston, James J. 1982. Conclusion: New Perspectives on Mother Worship. In Mother
Worship, ed. James J. Preston, 325-343. Chapel Hill: University of North Caro-
lina Press.
 References—General 313

Purvis, Sally. 1991. Mothers, Neighbors and Strangers: Another Look at Agape. Journal
of Feminist Studies in Religion 7(1): 19-34.
Rando, Therese A. 1986. The Unique Issues and Impact of the Death of a Child. In
Parental Loss of a Child, ed. Therese A. Rando, 5-44. Champaign, Illinois: Re-
search Press Company.
Redfield, Robert. 1956. Peasant Society and Culture. Chicago: University of Chicago
Press.
Rich, Adrienne. 1986 [1976]. Of Woman Born: Motherhood as Experience and Institution.
New York: W. W. Norton and Company.
Risman, Barbara J. 1987. Intimate Relationships from a Microstructural Perspective:
Men Who Mother. Gender and Society 1:6-32.
Rosaldo, Michelle Zimbalist. 1974. Women, Culture, and Society: A Theoretical Over-
view. In Women, Culture, and Society, ed. Michelle Zimbalist Rosaldo and Louise
Lamphere, 17-42. Stanford: Stanford University Press.
Rosaldo, Renato. 1989. Culture and Truth: The Remaking of Social Analysis. Boston:
Beacon Press.
Rosenblatt, Paul, Patricia Walsh, and Douglas Jackson. 1976. Grief and Mourning in
Cross-Cultural Perspective. New Haven: HRAF Press.
Rossi, Alice S. 1981. On The Reproduction of Mothering: A Methodological Debate. Signs
6(3):492-500.
Rubin, Lillian B. 1983. Intimate Strangers: Men and Women Together. New York: Harper
and Row.
Rubin, Simon Shimshon. 1984-1985. Maternal Attachment and Child Death: On Ad-
justment, Relationship, and Resolution. Omega i5(4):347-352.
Ruddick, Sara. 1982. Maternal Thinking. In Rethinking the Family: Some Feminist Ques-
tions, ed. Barrie Thome and Marilyn Yalom, 76-94. New York: Longman.
. 1983. Preservative Love and Military Destruction: Some Reflections on Moth-
ering and Peace. In Mothering: Essays in Feminist Theory, ed. Joyce Trebilcot, 231 —
262. Savage, Maryland: Rowman and Littlefield.
Ruether, Rosemary. 1974. Religion and Sexism: Images of Woman in the Jewish and Chris-
tian Traditions. New York: Simon and Schuster.
. n.d. In the Mountains of Banahaw Padre Is a Woman. Unpublished essay.
, and Rosemary Skinner Keller. Editors. 1981. Women and Religion in America.
Vol. i. San Francisco: Harper and Row.
, and Rosemary Skinner Keller. Editors. 1983. Women and Religion in America.
Vol. 2. San Francisco: Harper and Row.
Sacks, Karen. 1974. Engels Revisited: Women, the Organization of Production, and
Private Property. In Women, Culture and Society, ed. Michelle Zimbalist Rosaldo
and Louise Lamphere, 207-222. Stanford: Stanford University Press.
. 1979. Sisters and Wives. Westport, Connecticut: Greenwood Press.
Sanday, Peggy Reeves. 1981. Female Power and Male Dominance: On the Origins of Sexual
Inequality. Cambridge: Cambridge University Press.
, and Ruth Gallagher Goodenough, Editors. 1990. Beyond the Second Sex: New
Directions in the Anthropology of Gender. Philadelphia: University of Pennsylvania
Press.
Sanderson, Lilian Passmore. 1981. Against the Mutilation of Women: The Struggle to End
Unnecessary Suffering. London: Ithaca Press.
Schatz, Barbara D. 1986. Grief of Mothers. In Parental Loss of a Child, ed. Therese A.
Rando, 303-314. Champaign, Illinois: Research Press Company.
Schlegel, Alice. 1972. Male Dominance and Female Autonomy. New Haven: HRAF Press.
314 References—General

. 1977. Male and Female in Hopi Thought and Action. In Sexual Stratification: A
Cross-Cultural View, ed. Alice Schlegcl, 245-269. New York: Columbia Univer-
sity Press.
Schoenfeld, Eugen, and Stjepan Mestrovic. 1991. With Justice and Mercy: Instrumental-
Masculine and Expressive-Feminine Elements in Religion. Journal for the Scientific
Study of Religion 3o(4):363-38o.
Schousboe, Karen, and Mogens Trolle Larsen, Editors. 1989. Literacy and Society. Co-
penhagen: Akademisk Forlag, Center for Research in the Humanities, Copenha-
gen University.
Schulenburg, Jane T. 1989. Women's Monastic Communities, 500-1100: Patterns of
Expansion and Decline. Signs i4(2):26i-292.
Sered, Susan Starr. 1987. The Liberation of Widowhood. Journal of Cross-Cultural Geron-
tology 2:139-150.
. 1988. The Domestication of Religion: The Spiritual Guardianship of Elderly
Jewish Women. Man (N.S.) 23:506-521.
. 1990. Women, Religion, and Modernization: Tradition and Transformation
among Elderly Jews in Israel. American Anthropologist 92:306-318.
. 19913. Conflict, Complement, and Control: Family and Religion among Mid-
dle Eastern Jewish Women in Jerusalem. Gender and Society 5(i):io-29.
. i99ib. Childbirth as a Religious Experience? Voices from an Israeli Hospital.
Journal of Feminist Studies in Religion 7(2):j~ 18.
1992. Women as Ritual Experts: The Religious Lives of Elderly Jewish Women in
Jerusalem. New York: Oxford University Press.
Seremetakis, C. Nadia. 1991. The Last Word: Women, Death and Divination in Inner Mani.
Chicago: University of Chicago Press.
Sharma, Arvind. 1977. How and Why Did the Women in Ancient India Become
Buddhist Nuns? Sociological Analysis 38(3):329-35i.
Shimony, Armemarie. 1980. Women of Influence and Prestige Among the Native Ameri-
can Iroquois. In Unspoken Worlds: Women's Religious Lives in Non-Western Cultures,
ed. Nancy A. Falk and Rita M. Gross, 243—259. San Francisco: Harper and Row.
Silverblatt, Irene. 1980. Andean Women under Spanish Rule. In Women and Coloniza-
tion: Anthropological Perspectives, ed. Mona Etienne and Eleanor Leacock, 149-
185. New York: J. F. Bergin.
Singer, Merrill. 1985. The Concept of Justice in the Religions of the Caribbean. Phylon
46:296-299.
Smith, Jane I. 1987. Islam. In Women in World Religions, ed. Arvind Sharma, 235-250.
Albany: State University of New York Press.
Smith, Sidonie. 1987. A Poetics of Women's Autobiography. Bloomington: Indiana Univer-
sity Press.
Smith, W. Robertson. 1972 [1889]. The Religion of the Semites: The Fundamental Institu-
tions. New York: Schocken.
Spiro, Melford E. 1984. Some Reflections on Family and Religion in East Asia. In
Religion and the Family in East Asia, ed. George A. De Vos and Takao Sofue, 35-
54. Berkeley: University of California Press.
Spring, Anita. 1976. An Indigenous Therapeutic Style and Its Consequences for Na-
tality: The Luvale of Zambia. In Culture, Natality, and Family Planning, ed. John
F. Marshall and Steven Polgar, 99-125. Chapel Hill: University of North Caro-
lina, Caroline Population Center.
. 1978. Epidemiology of Spirit Possession among the Luvale of Zambia. In
 References—General 315a

Women in Ritual and Symbolic Roles, ed. Judith Hoch-Smith and Anita Spring,
165-190. New York: Plenum Press.
Stamm, Liesa, and Carol D. Ryff. 1984. Introduction. In Social Power and Influence of
Women, ed. Liesa Stamm and Carol D. Ryff, i-n. AAAS Selected Symposium,
no. 96. Boulder, Colorado: Westview Press.
Stark, Rodney, and William Bainbridge. 1985. The Future of Religion. Berkeley: Univer-
sity of California Press.
Stover, Ronald, and Christine Hope. 1984. Monotheism and Gender Status: A Cross-
Societal Study. Social Forces 63(2):335~348.
Suzuki, Mitsuo. 1976. The Shamanistic Element in Taiwanese Folk Religion. In The
Realm of the Extra-Human, ed. Agehananda Bharati, 253-260. Hague: Mouton.
Tamayo, Alvaro, and Albert Dugas. 1977. Conceptual Representation of Mother, Fa-
ther, and God According to Sex and Field of Study. Journal of Psychology 97-79-
84.
Tamez, Elsa, Editor. 1989. Through Her Eyes: Women's Theology from Latin America.
Maryknoll, New York: Orbis Books.
Tanner, Nancy. 1974. Matrifocality in Indonesia and Africa and Among Black Ameri-
cans. In Women, Culture, and Society, ed. Michelle Z. Rosaldo and Louise
Lamphere, 129-156. Stanford: Stanford University Press.
Tavris, Carol. 1992. The Mismeasure of Women. New York: Simon and Schuster.
Tenrikyo: Its History and Teachings, n.d. ed. Tenrikyo Overseas Mission Department.
Thompson, Catherine. 1983. Women, Fertility and the Worship of Gods in a Hindu
Village. In Women's Religious Experience, ed. Pat Holden, 113-131. London:
Groom Helm.
Thompson, Daniel C. 1974. Sociology of the Black Experience. Contributions in Sociol-
ogy, vol. 14. Westport, Connecticut: Greenwood Press.
Van Binsbergen, Wim M. J. 1981. Religious Change in Zambia. London: Kegan Paul.
Van Esterik, Penny. 1982. Introduction. In Women of Southeast Asia. Occasional Paper,
Center for Southeast Asian Studies, ed. Penny Van Esterik, 1-15. Dekalb: North-
ern Illinois University.
VanEecke, Catherine. 1989. From Pasture to Purdah: The Transformation of Women's
Roles and Identity Among the Adamawa Fulbe. Ethnology 28(i):53-73.
Wadley, Susan S. 1980. 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, 73-93. San Francisco: Harper and Row.
Walker, Sheila. 1972. Ceremonial Spirit Possession in Africa and Afro-America. Lieden: E. J.
Brill.
Ward, Colleen. 1982. A Transcultural Perspective on Women and Madness: The Case of
the Mystical Affliction. Women's Studies International Forum 5(s):4i 1-418.
Weber, Max. 1966. The Sociology of Religion. Boston: Beacon Press.
Weinstein, Donald, and Rudolph Bell, M. 1982. Saints and Society: The Two Worlds of
Western Christendom, 1000-1700. Chicago: University of Chicago Press.
Welter, Barbara. 1966. The Cult of True Womanhood: 1820-1860. American Quarterly
18:151-174.
. 1974. The Feminization of American Religion, 1800-1860. In Clio's Conscious-
ness Raised: New Perspectives on the History of Women, ed. Mary Hartman and Lois
Banner. New York: Harper and Row.
Werblowsky, R. J. Zwi. 1991. Mizuko kuyo: Notulae on the Most Important "New
Religion" of Japan. Japanese Journal of Religious Studies 18:295—354.
316 References—General
Wessley, Stephen E. 1978. The Thirteenth-Century Guglielmites: Salvation Through
Women. In Medieval Women, ed. Derek Baker. Oxford: Basil Blackwell.
Whelehan, Patricia, and Contributors. 1988. Women and Health: Cross-Cultural Perspec-
tives. Granby, Massachusetts: Bergin and Garvey.
Wilson, Peter. 1967. Status Ambiguity and Spirit Possession. Man 2:366-378.
Wittig, Monique. 1971. Les Guerilleres. Trans. David LeVay. New York: Avon Books.
Wolf, Margery. 1990. The Woman Who Didn't Become a Shaman. American Ethnologist
I7(3):4I9-430.
Wulff, David. 1984. Prolegomenon to a Psychology of the Goddess. In The Divine
Consort: Radha and the Goddesses of India, ed. John S. Hawley and Donna M.
Wulff, 283-297. Delhi: Motilal Banarsidass.
Yanagisako, Sylvia Junko. 1977. Women-Centered Kin Networks in Urban Bilateral
Kinship. American Ethnologist 4(z):2Oj-226.
- . 1979. Family and Household: The Analysis of Domestic Groups. Annual Re-
view of Anthropology 8:161-205.
Young, Iris Marion. 1983. Is Male Gender Identity the Cause of Male Dominance? In
Mothering: Essays in Feminist Theory, ed. Joyce Trebilcot, 129-146. Savage, Mary-
land: Rowman and Littlefield.
INDEX

Aboriginal religion, 40 Ancestor rituals

Aborted fetuses, rituals, 100n2 Black Caribs, 35—36, 80, 131-33, 168—69
Absolute morality, 157—58 function of, 3 5
Abstract thinking, 84, 281 Korean household religion, 167-68
Acting ability, 226 life versus death involvement, 202-3
Adam and Eve, 72 versus male Confucian rites, 167-68
Administrative positions, 235—36 and matrilineality, 61
African-American Baptist women leaders, 239 responsibility of older women, 80
African religions, 29—39 Tensho practice, 130-31
Afro-Brazilian religions. See also Batuque cult; Ancestresses, 169
Umbanda cult Anderson, Leafy, 148
benefits, 262 Androgyny
characteristics, 32—34, 289 in deities, 174—76
and child death, 91-92 in religious leaders, 241n11
encantados (spirits) in, 164-65 Anigi, 168
gender ideology, 199 Animal sacrifice
gender of deity, 177 blood bond strengthening function,
healing techniques, 116-17 136-38
hierarchical organization, 218, 240n6 versus communal meals, 136—38
illness as recruitment path, 108—9 men's role, 237
leadership roles for women, 216 in patrilineal societies, 136—38
men's control of sexuality, 272 Anthropological interpretation, 170
priestesses character traits, 226 Apocalypse, 148
sacred and profane in mediums, 152 Asceticism, 5, 146
sacred geography, 151 Assistants, men as, 236
social context, 56 Autobiographical histories, 220, 225
this-worldly nature of, 146-47 Autonomy, 277, 283
training for mediumship, 222—23 Avunculocality, 48
uncentralized organization, 244—45
as urban phenomenon, 56 Baby M case, 81, 86n4
Afterlife, 146—48. See also Transcendence Baer, Hans, 249
Ajumgu, 168 Bahia. See also Afro-Brazilian religions
Aged women. See Older women male leaders, 234—35
Aggression, 84, 87n7 mother-headed households, 56
Akha tribe, 209—10 respect for priestesses, 234
Albanese, Catherine, 199 Baptist women leaders, 239
Amami Island, 41n6 Batuque cult, 32-33
American Indians, 199, 211n5 appeal of, 262
spirits of, 163-64 characteristics, 33
American slavery, 30—31 and morality, 155
American Spiritualism. See Spiritualism uncentralized organization, 244-45
Anatomical models, 279, 286n1—2 Beguines, 120, 246, 291
318 Index
Belize society, 50-52. See also Black Caribs of Brazil, 56. See also Afro-Brazilian religions
Belize Brewer, Priscilla, 256n6
Benefits, secular of women's religions, 257-78 Brideprice, 58
Bereaved fathers, 98—100, 101n13 Brideservice, 58
Bereaved mothers, 89-101 British Spiritualism, 97
in Afro-Brazilian cults, 91-92 and child death, 94, 97
versus bereaved fathers, 98-100, 101n13 sacred geography, 151
Spiritualism appeal, 94—95 women as healers, 109-10
Biblical writings, gender portrayals, 210 Brother-sister relationship, 208-9
Bilharziasis, 68n14 Brown, Audrey, 239
Bilu, Yoram, 185 Brown, Judith, 129, 130, 142n11
Bimin-Koskusmin, 241n14 Brown, Karen, 156
Birth control, 157, 272-74, 278n11 Buddhism
Birth rituals, 138-40. See also Childbirth history, 13, 41n3
Black Caribs of Belize nat religion differences, 15-17, 113-14, 146
ancestor ritual, 35-36, 168-69 as "other-worldly" religion, 146, 159n3,
blood sacrifice de-emphasis, 137 I59n5
and Catholic Church, 35-36 suffering explanation, 113-14
characteristics, 34—36, 289 tolerance of other religions, 13, 19
food rituals, 134-35 transcendent conception of death, 200
gender of deity, 177 women's status in, 21—22
grandmother's role, 78-80 Buddhist nuns, 65
healing techniques, 116 Bundu. See Sande secret society
kinship structure, 34 Burdick, John, 33—34
menstruation beliefs, 201 Burmese culture. See also Nat religion
mourning rites, 131-33, 140 gender dissonance, 53-54
ritual experts in, 79 matrilocality, 52-54
social context, 50-52 older women in, 80
spirit categories, 168—69 Burnt offerings, 136-38
spirit possession, 185, 192n4 Bynum, Caroline, 142n13,153, 154
uncentralized organization, 245
Black Hawk, 148 Calcium deprivation hypothesis, 183
Black Spiritual movement Calvinist doctrine, 93, 101n6, 113
characteristics, 291 Candomble cult, 32-33, 109, 151
food rites, 134 Cantonese shamanism, 96-97
healing techniques, 116 Catharsis, 261
males in leadership positions, 236 Catholic Church
motherhood symbols, 76 absolute morality, 157
spirit possession, bonding function, 193n8 and Afro-Brazilian religions, 33-34
this-worldly nature of, 148 birth control view, 157
Black women, 27. See also Womanism and Black Caribs, 35-36
Bledsoe, Caroline, 57, 58, 218 lack of appeal to Brazilian women, 33
Bleier, Ruth, 273, 274 new theology of liberation effect, 91
"Blessed souls," 132 shared leadership in, 235
Blood rituals written theology, 252
Central Asian shamans, 143n18 Catholic nuns, 241n14
versus communal meals, 136—38 Caw, 167
and interpersonal bonds, 127 Celibacy, 273
patrilineal society practice, 136-38, 140 and child death, 95-96
Taiwanese male shamans, 143n18 medieval women mystics, 154
Blood ties, 136 in patriarchal cultures, 6-7
Boddy, Janice, 36-37, 38, 63-64, 69n19, 107, Sanctificationists, 268
184-86, 188, 191, 203, 270 Shaker belief, 25-26
Bodily pollution beliefs, 200-202 and women's control of fertility, 273
Bondage rituals, 232 Centralized organization, 243-55
Bori cult, 30, 39 men's role, 249-50, 254-55
Braude, Ann, 94, 111, 112, 197, 198 monotheism parallel, 253
 Index 319
versus rank, 253-55 Christian Science Monitor, 148
in Shakers, 247—49 Christianity
women's aversion to, 243—46, 285 conception of justice, 156-57
and written theology, 252-53, 255 masculine deity, 172
Chanfrault-Duchet, Marie-Francoise, 220 monotheism interpretations, 170-71
Charismatic women leaders, 254 transcendent conception of death, 200
Charms, 201 Chronic illness. See also Illness
Chernela, Janet, 67n6 incidence in women, 111
Chief priestess, 233, 24On3 motherhood connection, 105-6
Child death, 89-101, 284. See also Infant Clark, Asenath, 256
mortality Clitoridectomy. See also Genital mutilation;
and Afro-Brazilian cults, 91-92 Infibulation
child as tutelary spirit, 96-97 emic and etic explanations, 126-27
denial of, 99 interpretations of, 126-28, 142n7-11
father versus mother grief response, 98-99, in Sande initiation ritual, 31, 126-28
101n13 Codified laws, 249, 285
Korean shamans link, 97-98 Cognitive styles
as maternal responsibility, 90-91 gender differences, 84, 87n8
in nineteenth century United States, 92-96 and polydeistic beliefs, 170
search for meaning in, 100 Cohen, Paul, 19
and Shakerism, 95-96 Communal interests, 271
and Spiritualism, 91, 94-97, 101n8 Communal meals, 136-38
zar cult function, 106—7 Communal religions, 96
Childbirth Communal rituals, 122
absence of ritual, 138-40 Competitive behavior, 217, 254
metaphors of, 75-76 Conarton, Sharon, 188, 225-26
pollution beliefs, 201—2 Concrete thinking
and male-dominated religions, 139-40 female disposition, 84
and spiritual intuition, 77 and literacy, 252
Childcare Confucianism, 55
emotional/spiritual effects, 83-85 and infant death, 98
Shaker'practices, 267 versus Korean kut ritual, 167-68
women's greater responsibility, 82 in Korean society, 55
Childless women, 78-79 marginality of women, 75
Children's health Constantinides, Pamela, 63, 261-62
Black Carib women's concern, 78-79 Cooperative behavior, 217, 254
and mother's emotional development, 83 Coptic Christianity, 37
Chodorow, Nancy, 8, 82, 86n6, 140, 154, 188, Cosentino, Donald, 126, 256n13
189, 196, 205, 284 Couvade, 67n6
Chosang-gori, 130 Creation myths, 208
Christ, Carol, 124, 173-74, 280, 281 Cults. See Northern Thai matrilineal spirit
Christ icons, 44 cults
Christian holy women, 154
Christian Science. See also Eddy, Mary Baker Daly, Mary, 142n7, 172, 258
androgynous deity, 174-75 Daughters
characteristics, 23—24, 289 leadership inheritance from mother, 223
gender dissonance, 43-45 in patrilineal societies, 71—72
gender ideology, 199 status in Buddhism, 21
illness interpretation, 111—13 Davis, Richard, 212n13, 268
masculine monotheism rejection, 172 Davis, Susan, 78
organizational hierarchy, 219, 250—51 Death. See also Child death
other-worldly nature of, 147—48 and appeal of religion, 100, 101n8
qualities of leaders, 226 association with women, 202—3
ritual in, 120 Death beliefs, 148-49, 200
sacred writings, 250—51 Death rituals, 130-33, 139
Shakerism comparison, 26 Denial of death, 99
social context, 43-45 Depression, 105
320 Index
Deprivation theory, 62-66, 258 English, Deirdre, 275—76
critique, 62—66 Erikson, Erik, 149, 279, 286n1
versus independence theory, 64-66 Esquivel, Julia, 173
and religious leaders, 227-29 Eternal life belief, 99
of women's spirit possession, 183, 190—91 Ethical principles, 156-58
and the zar cult, 62-66 Ethnography, 149, 159n4
Devas, 178n1 Eucharist, 133, 136, 138
Dianic Wiccans, 258 Eve symbolism, 72, 202
Digo of South Kenya, 42n20, 274—75, Exorcists, 189-90
278n15 Exus, 165
Divale, William, 143nI6
Divination tray, 117 Family relations. See also Interpersonal
Divine kingship, 171 relationships
Divorce, 49, 272 focus of Korean household religion, 123
Doctrine, 244-46, 285 Korean shamans, 229
Dogma, 244-45 religious systems as metaphors of, 61
Domestic work, 151, 271 Tensho religion attitudes, 282
Dominica Carib, 80 and women's religiosity, 6
Douglas, Mary, 46, 212n9, 275 Father-absent societies, 205
Dualism, 104, 111 Father God, 172
Dugu ritual, 133, 168 Father John, 148
Father-Mother deity, 174—76
Earth mother, 179n14 Fatherhood, and personality, 85
East Asian religions Fathers, bereavement response, 98-99, 101n13
historical context, 12-13, 41n2-3 Father's death, 221, 241n8
older women in, 80 Feki Islam, 116
Eberhardt, Nancy, 209-10 Female deities, 271-72. See also Goddesses
Ecological perspective, 124, 147, 157 Female inferior/superior ideology, 196
Economic benefits, 227-28 Feminine principle, 199
Economic independence, 270-72. See also Feminine qualities
Matrifocality in androgynous deities, 175-76
communal interests correlation, 271 and gender ideology, Spiritualism, 197-98
and female deities, 271-72 Feminist Spirituality Movement. See also
and women's status, 270—72 Womanism
Economic milieu, 7 characteristics, 26-27, 289
Eddy, Mary Baker, 23, 113, 174, 219 gender ideology, 199-200, 205-7
Education, 251 Goddesses in, 173-74
Egalitarian ideology, 217 interpersonal relationships as essence of, 124
Egalitarian societies and matrifocality, 56-57
definition, 67n2 morality in, 156-57
and egalitarian religion, 175 mother's spiritual intuition belief, 76-78
Ego boundaries political concerns, 27, 258
in leaders, 227 rituals in, 27
men versus women, 226-27 sacred text aversion, 251
and spirit possession, 187-90 search for authenticity in, 26-27
in women, 187 sexuality/fertility control, 273-74
Ehrenreich, Barbara, 275-76 this-worldly nature of, 146-47
Eliade, Mircea, 129, 153, 193, 280, 281 uncentralized organization, 246
Emotional rituals, 5 Fertility, 272-74, 278n11
Emotions control in Sande ritual, 74, 128, 272
motherhood effect on, 84—85 as natural function in ritual, 153
pregnancy effect on, 82—83 separation from sexuality, 128, 142n10
Empathy women's control of, 62, 272-74
leadership quality, 225—26 Fidelity, 79, 86n3
and motherhood, 85 Field dependence, 84
Encantados, 33, 155 Financial gain, religious leaders, 227—28
characteristics of, 164-65 Finkler, Kaja, 91, 106-8, 124, 263
 Index 321
Finnegan, Ruth, 252 Good mothering, 188
Fiqi cult, 37 Goodenough, Ruth, 8
First pregnancy, 82—83 Goody, Jack, 252-53
Food rituals, 6, 133-38 Graham, Hilary, 189-90
animal sacrifice versus communal meals in, Grandmothers, 78-81
136-38 Black Carib society, 78-80
in Northern Thai matrilineal spirit cults, 135 leadership role, 78
social emphasis, 133-38 nurturing role, 81
Fox sisters, 28 ritual responsibility, 80
Freudian theory, 170, 172, 178n8, 286n1 Greco-Roman religion, 170
Frymer-Kensky, Tikva, 178n9, 210 Grid factors, 46
Funeral rituals, 130—33 Grief. See Bereaved mothers; Child death
Black Carib practices, 80, 131-33, 140 Group factors, 46
de-emphasis, 139 Group ritual, 122, 124, 126—28, 135
interpersonal bond emphasis, 130—33 Guglielmites, 39, 118n1, 291
Tensho practices, 130—31 Gutmann, David, 188, 226, 227

Garifuna. See Black Caribs of Belize Haitian urban Vodou, 58—59

Garrett, Clarke, 247 interpersonal focus of healing, 121-22
Gender differences, 129, 209—11 morality in, 156
Gender dissonance Haj Yousif area of Khartoum, 105
Burmese culture, 53—54 Hale, Ann, 268
Korean culture, 55-56 Hale, Mary, 83
nineteenth century U.S. religions, 43-45 Hanks, Jane, 212n11
Northern Thai culture, 48-50 Haring, Douglas, 15, 233
West African culture, 45-46 Harrison, Jane, 178n9
women's religion correlate, 65—66, 285 Harrist Movement, 249
Gender identity, 196 Harvey, Youngsook Kim, 18, 224
Gender ideology, 195—213 Healers
in Feminist Spirituality Movement, 205-7 financial benefits, 228
and gender differences, 209—11 training, 222
and gender identity, 196 Healing, 103-18
nature and culture in, 198-200 Afro-Brazilian religions, 32-33, 108-9
in ritual tugs of war, 204—5 Christian Science emphasis, 23
Gender of deities, 177 holistic approach, 103-4
Gender roles. See Sex roles in male dominated religions, 1 l8n1
Genital mutilation maternal nurturing metaphor in, 106
interpretations, 126-28, 142n7-11 mechanisms of, 115, 117, 124
Sande secret society, 126-28, 142n7—9 modern medicine failure, Japan, 109
Gill, Brita, 237-38 nat religion versus Buddhism, 113-14
Gilligan, Carol, 8, 84, 121, 122, 155, 156, 188, polydeism explanation, 171
217 social nature of, 115-17
Gnostic groups, 176 spirit possession function, 185
"God-in-her-abdomen," 76 and Spiritualism, 109—10
Goddess religions. See Feminist Spirituality techniques, 114-17
Movement training for, 222
Goddesses, 172-76, 280. See also Polydeism Health problems, 105. Sec also Illness
common characteristics, 172, 177—78 Heaven and hell belief, 147
and economic conditions, 271-72 Heresy, Christian Science as, 24
in Feminist Spirituality Movement, 173-74 Heterosexism, 273
Freudian view, 172, 178n8 Hicks, David, 143nI7, 203
and motherhood, 77, 178 Hierarchical religions, 216-19
this-worldliness of, 177—78 Hindu women, 65
and women's empowerment, 173 Historians of religion, 171
Gods, 161-79 Hiuruha, 168
Golden Rule, 156 Holiness church, 5
Gomm, Roger, 274-75 Holy Mother Wisdom, 175-76
322 Index
Homosexuality Intelligence
male leaders, 234—35 leadership trait, 225—26
and male mediums, 34 sex differences, studies of, 281
Hope, Christine, 169, 172 Interpersonal relationships
Hopi society, 69n16 blood rituals function, 127
Household religion. See Korean shamanism and decentralized religion, 253
and household religion focus in women's religions, 84, 121-22, 284
Hutch, Richard, 237 leadership benefits, 228-29
Hypochondria, 275 in men versus women, 122
Hysterectomy, 142n10 and motherhood, 82, 84
and ritual, 119-43
Illiteracy, 252-53 Intuition
Illness, 104—6. See also Healing leadership quality, 225—26
Afro-Brazilian religion recruitment path, and motherhood, 76-78
108-9 Spiritualism gender ideology, 197—98
and child death, 91 Iroquois religion, 40
Christian Science interpretation, 111-13 Islam
impersonal forces as determinants, 124 masculine deity, 172
in life history of mediums, 219—22 monotheism interpretations, 170—71
motherhood connection, 105-6 and possession cults, 30
nat religion versus buddhism explanation, transcendent conception of death, 200
113-14 treatment of women, 38, 65, 260
and oppression of women, 275—76 zar cult relationship, 36-38
patriarchy as cause of, 104—5, 107—8, luani, 168
275-76
Immanence. See also "This-worldly" religions Japan, 13, 109
female aptitude for, 171 Jay, Nancy, 137
Goddess characteristic, 174 Jedrej, M.C., 207
of spirit possession, 191 Jinn, 37
in this-worldly religions, 1 50 Joan of Arc, 232
versus transcendence, shamans, 187 Jonte-Pace, Diane, 73, 202-3
value of, 281 Judaism
Individual conscience, 157, 159n13 conception of justice, 156—57
Indwelling spirits, 164 masculine god, 172
Infant mortality, 89—101. See also Child death monotheism interpretation, 170-71
Black Carib of Belize, 79 transcendent conception of death, 200
Luvale tribe of Zambia, 90 written theology, 252
as maternal responsibility, 90-91 Jural paternity, 137
in nineteenth century United States, 93
Infanticide, 90 Kahana, Yael, 261
Infertility, 106—7. See also Fertility Kamado rites, 119
Infibulation, 127 Kamata, Hisako, 192, 240n7
Infidelity, 79, 86n3 Kami, 14, 41n5, 119, 133
Initiation rituals gender of, 177
boys versus girls, 153 as indwelling spirits, 164
cross-cultural perspective, 129—30 Kamidari, 221
group emphasis, 125—28 Kaminchu, 159
internal continuity emphasis, women, Kanter, Rosabeth, 253, 254, 255, 256n11
153-54 Kaplan, Alexander, 73
liminality critique, 153-54 Karma, 114, 131
sacred and profane in, 153—54 Kashrut, 210-11
Sande secret societies, 125—28, 264—65 Kendall, Laurel, 18, 19, 41n2, 55, 56, 75, 230,
zar cult, 128 231
Innen, 282 Kennedy, John, 261
Inner-directedness, 83 Kerns, Virginia, 34, 51, 52, 78, 79, 80, 132,
Institutionalized religion, 285 133
pattern of, 254 Keyes, Charles, 22
Shaker trend, 248-49, 256n6 Kirsch, Thomas, 21, 22, 50
 Index 323
Kitch, Sally, 76, 175, 268 hierarchical structure, 216-19
Kono of Sierra Leone, 125 illness as common theme, 219—22
Korean Buddhist nuns, 97-98 life histories of women, 219-27
Korean men's religion, 75. See also in male-dominated churches, 238-39
Confucianism and men, 234-37
Korean shamanism and household religion and motherhood, 83
versus Buddhism, 146 and persecution, 231—32
characteristics, 17-19, 289 secular versus religious, 238
and child death, 97-98 situational nature of, 254
versus Confucian rites, 167-68 and status, 229-34
family focus of ritual, 123 suffering as preparation, 216
food rituals, 134 women versus men, 237-38
gender of deity, 177 Lebra, William, 15, 68n8, 146, 150, 253
gods and ghosts in, 167-68 Lee, Ann, 212n16, 213n18
healing techniques, 116-17, 118n8 child death meaning, 95
versus Korean men's religion, 75 as female symbol, 175
matrifocal context, 55—56 founding of Shakers, 24-25
motherhood interpretation, 74—75 illiteracy, 252
response to suffering, no as motherhood symbol, 75—76, 85n1
ritual in, 123, 167-68 persecution of, 232
sacred geography, 150 and Shaker organization, 247, 255n5
uncentralized organization, 245 Lehman, Edward, 227
Korean shamans, 97-98, 223-24 Leifer, Myra, 82-83
character traits, 226 Lerch, Patricia, 216, 228, 229, 262
financial gain, 228 Lesbians, 57
life histories, 223-24 Lewis, I.M., 39, 62, 69017, 163, 182, 184, 186,
social role benefits, 229 187, 193, 241n8, 261, 270
status in society, 230—31 Life-cycle rituals, 139-41
Kosa, 134 Life hereafter belief, 99, 146-47
Koss-Chioino, Joan, 222 Life stress, 105-6
Kpelle society, 45—46 Liminality, 153—54
gender roles in, 204 Lincoln, Bruce, 129, 153, 218
matrifocality, 57-58 Literacy, 252-53
Kraemer, Ross, 5, 42n19, 46, 170, 176, 190 Logos, 199
Kunstadter, Peter, 278n11 "Lord of death," 203
Kurdish Jewish women "Love magic," 35, 134-35, 280
and deprivation theory, 65 Lu vale of Zambia, 68n14, 284, 291
eloquent language of ritual, 120 maternal responsibility for infant mortality,
Kut ritual, 18 90
family focus, 123 matrilineality and virilocality, 59-60
gods and ghosts in, 167-68 reproductive rituals, 128
women's kinship reinforcement, 56, 75
MacCormack, Carol, 31, 32, 74, 126, I92n1,
Lambek, Michael, 191, I93n9 207, 218, 264, 265
Land ownership, 273 Macklin, June, 123, 124, 228
Landon, Ruth, 256 Macumba cult, 32
Lay spirit possession, 185, 274-75 Magic
Leadership, 215-41. 'See also Mediums; Black Carib practices, 35
Shamans men's versus women's practices, 212n11
and age, 227 "Male bonding," 129
benefits, 227-29 Male dominance, 8
and centralized organization, 254-55 Male-dominated religions
character traits, 225—27 birth rituals, 139-40
in female-dominated religions, 215-16 blood sacrifices in, 136-38
financial gain theory, 227—28 centralization in, 254—55
gender differences in style, 253-54, 256n11- leadership positions, 235-36
12 monotheism in, 169-72
and grandmothers, 78 morality in, 155-57
324 Index
Male-dominated religions (continued) in Afro-Brazilian religions, 32—34, 165
motherhood in, 72 as bereaved mothers, 91-92
suffering interpretation, 111 character traits, 225-27
women leaders in, 238-39 ecstatic flight versus possession controversy,
written theology in, 252—53 186-87
Male gods, 172—73 ego boundaries in, 187-90
Male leaders, 234—38 financial gain theory, 227-28
Male migration, 63, 69n18, 282 and gender ideology, 197-98
Male spirits, 183-84 healing function, 109-10
Mann seag phox, 96-97 high status of, 233-34
Mansin, 118n8, 230. See also Korean shamans illness as theme in life histories, 219-22
Marital conflict, 107—8 leadership role, 240n6
Marital patterns, mediums, 221 life histories, 219-27
Marital status, leaders, 239 men as, 241n15
Marriage ceremonies, 274 passivity association, 197-98
minor role of, 139-40 persecution of, 231-32
in Spiritualism, 278n14 resistance in, 219, 221
Martin, Patricia, 217 shaman comparison, 186—87
Martyrdom, 276 social benefits, 228-29
Masculine god, 172—73 status in Korea, 230—31
Masculine principle, 199 testing of, 232
"Masculine" qualities, 175 training, 222—23
Masking rites, 141n5 women as, 29, 197
Maternal role. See Motherhood Menarche, 129, 139
"Maternal thinking," 83-84 Mende tribe, 31, 45. See also Sande secret
Matrifocality, 46—48, 61—63, 284 society
Belize society, 50-52 ancestresses, 178n5
continuity of women's experience in, initiation ceremonies, 125
154-55 matrifocality, 58
definition, 46—48, 67n4 menstruation beliefs, 201
in Haiti and Ancient Japan, 58—59 sexuality/fertility control, 272
Korean society, 55—56 Menopause, 139
maternal role emphasis, 66, 72 Menstruation
in Northern Sudan, 63-64 absence of ritual, 139
Northern Thailand, 48-50 as natural function in ritual, 153
United States, 56-57 pollution beliefs, 201-2
Upper Burmese society, 52—54 taboos, 71-72
West Africa, 57-58 Mental illness, 105
women's religion association, 61-63, 65—66, Mexican Spiritualism, 29
284 benefits of, 263
Matrilineality. See also Northern Thailand and child death, 91
matrilineal spirit cults illness models, 124
definition, 48, 67n5 patriarchy link, 107-8
and independent goddesses, 178n9 Mid wives, 218
Luvale of Zambia, 59—60 Miko, 59
Northern Thailand, 48—50 Militarism, 285-86
women's religion association, 60-62, 284 Miller-McLemore, Bonnie, 85
Matrilocality, 47-48, 61-62, 284 Min Mahagri, 162
Maturation rituals, 139 Ministers, leadership style, 227
Mayotte spirit possession, 191, 193n7, 193n9 Missionizing, 281, 286
McWhirter, Martha, 96 Mitha su, 53
Meacham, Joseph, 25, 213n17, 248 Miyako Island, 41n6
Medical establishment, 112 Moberg, David, 5
Medical treatment, 275-76 Modern medicine, 109, 112
Medieval women mystics, 154 Monasteries, 7
Mediumidade, 221 Monotheism, 285
Mediums, 186-87. See also Leadership centralization parallel, 253
 Index 325
gender inequality in, 170 rituals in, 130-33, 139-40
interpretations of, 170—72 Mrs. Kitamura, 76, 86n2, 131
and male dominance, 169 Mudang, 18, 123, 146, 229-30. See also Korean
male god in, 169 shamans
Moore, Laurence, 183, 187, 197, 198, 228 Musical ability, 226
Morality, 155-58, 280 Muslim women. See Islam
in Feminist Spirituality, 156—57
motherhood influence, 158 Nao, 125
particularism in, 158, 285 Nash, June, 161, 162
this-worldly nature of, 157 Nat religion. See also Burmese culture
Mormon Goddess, 173 Buddhism differences, 17, 113—14, 146
Moroccan women, 78 characteristics, 15-17, 161-63, 289
Mother Ann's Work, 151—52, 175—76, 185, and children's health, 90
I92n3, 215 food rituals, 134
Mother Catherine, 76 gender of deity, 177
Mother-child bonds, 47, 284 healing techniques, 116
and child death, 99, 284 hierarchical organization, 218
nineteenth century U.S. families, 92—93 nat characteristics, 161-63, 177
object relations theory, 73 older women in, 80
Mother Church, 250 sacred and profane in shamans, 152
Mother-daughter relationship shaman qualities, 221
Belize society, 51 social context, 52-54
Burmese family, 53 suffering explanation, 113-14
and leadership inheritance, 223 women mediums in, 189
United States society, 57 Natkadaws, 189, 223
Mother-Father deity, 174-76 Nature nats, 161-63
Mother goddesses, 77 Nature spirits, 163-64
Motherhood, 81-85 Neuhouser, Kevin, 183
Black Caribs of Belize, 78-79 New Orleans movement. See Black Spiritual
developmental model, 73 movement
effect on women's thinking, 83-84, 282—83 New Right women, 195, 211n1
and emotional/spiritual development, 82-85 Noddings, Nel, 158, 171
and gender ideology, 197 Noro. See also Priestesses
genital mutilation meaning, 127—28 fertility/sexuality control, 273, 278n12
illness connection, 105 hierarchical leadership role, 219, 250
in Korean household religion, 74-75 mother-daughter succession, 223
and leadership inheritance, 239 in Ryukyuan Islands religion, 14, 119, 221,
in matrifocal societies, 66, 72 24On7
meaning of, 81-85, 283-85 status of, 233
and morality, 158 yuta distinction, 240n7
and particularism, 284-85 North American religions, 22-29
personality effects, 82—83, 85 Northern Sudan, 63-64. See also Zar cult
power dilemma, 285—86 Northern Thai matrilineal spirit cults. See also
psychoanalytic view, 73 phii puu hjaa cult
sacralization in women's religions, 61—62 benefits, 268
in Sande belief, 74 blood sacrifice de-emphasis, 137
in Shaker religion, 75-76, 81, 284 versus Buddhism, 21-22, 146, I59n5
and spiritual intuition, 76 characteristics, 19—22, 289
symbols of, 75—76 economic independence, 271
this-worldly focus, 149 food rituals, 135, 137
valuation in Buddhist theology, 22 function, 20
versus wife, status, 71 gender ideology, 200, 204, 212n13
women's religion correlate, 66, 72-73, 283-85 group emphasis in ritual, 122, 135
Motherself(Rabuzzi), 77 and matrifocality, 48-50
Mourning. See also Child death matriline spirits in, 166-67
in Dominica Carib, 80 medium characteristics, 189, 222
nineteenth century U.S. women, 93 menstrual pollution beliefs, 201
326 Index
Northern Thai matrilineal spirit cults Passivity, 197-98
(continued) Patriarchal cultures, 8
older women in, 80 as cause of women's illness, 104—5, 107-8
ritual tugs of war, 204 definition, 67n2
sexuality/fertility control, 272—74 economics, 271
sisterhood in, 269—70 and genital mutilation, 127, 142n7
social context, 48-50 male deity function, 172-73
spirits in, 166-67 Spiritual Feminists condemnation of, 205—7
status of mediums, 233 Patrifocal cultures, 154
this-worldly nature of, 149-50 Patrilineal societies, 48, 69n15
uncentralized organization, 245 blood sacrifice in, 136-38, 143n18
Nubian women, 106-7, 261. See also Zar cult control of women's fertility in, 62
Nuclear families Korea, 75
motherhood focus, 92-93 problematic nature of kinship, 136—38,
women's stress in, 107—8 143n16
Nurturing and subordinate goddesses, 178n9
biological role comparison, 81 and women's religion, 64
grandmother's role, 81 Pentecostalism, 5
in single fathers, 85 Persecution, 231-32
in Womanist Movement, 77-78 Personality
pregnancy effect on, 82-83
Oaroari rite, 192
in single fathers, 85
Object relations theory, 73
"Phantom child syndrome," 99
Ochshorn, Judith, 170
Phii puu njaa cult. See also Northern Thai
Oedipus complex, 170
matrilineal spirit cults
Ogam, 235
benefits for women, 268
Okinawa, 13-15, 41n6. See also Ryukyu
characteristics, 19-20, 166-67
Islands religion
ritual, 135
foreign influences, 115
shrine for, 80
high status ofnoro, 233-34
Philippines, 200, 211n6
menstrual taboos, 212n9
Physiological models, 279, 286n1-2
religious hierarchy, 219, 249-50
Piker, Steven, 22
sacred geography, 150
Politics
stylized ritual, 141n3
and Feminist Spirituality Movement, 27
Old Blacks, 165
and women's religious options, 7
Older women
Polydeism, 169-72
in Black Carib society, 79-80
and gender equality, 170
cultural contexts, 78
gender of deities, 177
leadership qualities, 226-27, 241n11
interpretations of women's attraction to,
Oligarchy, 217
170-72
Orality, 252-53
and response to suffering, 171
Organization, 243-56
women's religion characteristic, 169-72
lack of centralized control, 243-46
Polytheistic world view, 157
rank versus centralization in, 253-55
Pornography, 142n10
Shakers, 247-49
Poro, 31-32, 207, 245-46
theology interaction, 252-53, 255
Possession trance. See Spirit possession
Orixa, 165
Postmenopausal women, 78
Ortner, Sherry, 8, 163, 198-99, 203, 211n3
Potter, Jack, 96, 97
Ota, Yoshinobu, 241n7
Power
"Other-worldly" religions, .145—49, 158n1,281
Korean shamans, 229
Ouwehand, Cornelius, 119, 133, 152, 202
and motherhood, 72, 285—86
Owen, Alex, 44, 198, 221, 280
Sande midwives, 217-18
Pacifism, 285-86 women's religion advantage, 277
Pantheons, 164-66 Pragmatism, 253
Parenting. See Motherhood; Nurturing Pregnancy
Particularism, 252 spirit possession parallels, 189-90
and maternal role, 158, 284-85 and women's spiritual/emotional growth,
sacred text incompatibility, 252—53 82-83
 Index 327
Pressel, Esther, 32, 262 death rituals, 203
Preston, James, 172, 178 female clergy, 196, 211n2
Priestesses. See also Noro food rituals, 133
character traits, 226 gender ideology, 199, 204, 208, 211n7
high status of, 233-34 gender of kami, 177
qualifications, 221 healing techniques, 115-16
in Ryukyuan Islands religion, 14, 221, hierarchical organization, 219
240n7, 273, 278n12 high status of priestesses, 233—34
yuta distinction, 240n7 interpersonal focus of ritual, 122
Proctor-Smith, Marjorie, 175, 176, 215, 247 kami (gods) in, 164, 177
Promiscuity, 232 menstrual pollution beliefs, 202
Proselytizing, 286 motherhood in, 287n6
Protestant doctrine, 93, 101n6 older women in, 80
Psychoanalytic theory, 273 response to infant death, 100n3
of goddess worship, 172 ritual tugs of war, 204
of motherhood, 73 rituals, 119, 122
and polydeistic beliefs, 170 sacred text aversion, 251
women and death association, 202 Ryukyu Islands society
Psychosomatic ailments, 105-7 kinship system, 54-55
Puberty rites. See Initiation rituals sacred geography, 150
Purification rituals, 151
Puru festival, 204 Sacks, Karen, 71, 269
Purvis, Sally, 158 Sacred geography, 150—51
Sacred texts, 250-53
Queen Esther, 148 and centralization, 252-53, 255
Qumran religious community, 5 de-emphasis of, 250-53
repression relationship, 252
Rabbinical authority, 211 Sacrificial meal, 136
Rabbis, 227 Sanctificationists, 291
Rank, 253—55 benefits, 267-68
Rape, 274 and child death meaning, 96
Religious autobiographies, 220, 225 group ritual, 124
Reproductive rituals, 140, 154 regulation of human relationships, 286n4
Resistance, in mediums, 219, 221 Sanday, Peggy, 8, 271, 278n10
Rights over women's bodies, 140 Sande secret society
Rites of passage, 138-41, 153-54 ancestresses, 169
Rites of reversal, 263 benefits, 264-66
Rites of solidarity, 138 blood ritual, patrilineal context, 137—38, 140
Ritual, 119-43. See also Food rituals; Funeral characteristics, 31-32, 290
rituals; Initiation rituals clitoridectomy in, 126—28, 137, 147n7-9
continuity emphasis, 140 food rituals, 134
gender opposition in, 204—5 gender ideology, 204-5, 207
interpersonal focus, 119-43 gender of ancestresses, 177
sacred and profane in, 153-54 healing techniques, 116
this-worldly nature of, 149—50 hierarchical power structure, 217-18
versus theology, women's religion, 6, 120— initiation rituals, 31, 125-28
21, 141n2 matrifocal context, 57-58
value of, 141n2 motherhood empowerment, 74
Ritual tugs of war, 204-5 ritual control of fertility, 74
Rosaldo, Michelle, 259, 216, 271 sexuality/fertility control, 272—73
Rubin, Lillian, 188 social context, 45—46
Ruddick, Sara, 83-84, 87n7-8, 196, 255, 285 uncentralized organization, 245-46
Ruethcr, Rosemary, 176, 200, 211n6 Sanderson, Lilian, 142n9
Ryff Carol, 254 Saunders, Lucie, 260, 261
Ryukyu Islands. See Okinawa Schmidt, Pater, 193n6
Ryukyu Islands religion, 13-15 Science, 164, 178n2
versus Buddhism, 146 Science and Health, 250, 256n9
characteristics of, 14-15, 289-90 Scribes, 235-36
328 Index
Seances, 29, 121 character traits, 225—27
Second-time parenting, 83 common features, 186-87
Secular benefits, 257—78 ecstatic flight versus possession controversy,
Self-help, 113 186-87
Sered, Susan, 65, 139, 227 ego boundaries, 188-89
Setta, Susan, 175, 208 financial gain theory, 227—28
Sex differences, 129, 209-11 high social status, 233-34
Sex roles life histories, 219-21, 223-24
nineteenth century America, 44—45 mixture of sacred and profane, 152
and spirit possession, 183—85 persecution of, 231—32
Sexism in Ryukyuan Islands religion, 14
and femininity conflicts, 221 social benefits, 228-29
reinterpretation in women's religions, 197 status in Korean society, 230-31
and sexuality, 273-74 Shan tribe, 210
suffering connection, 110 Shared leadership, 235
Sexual deprivation hypothesis, 182-83, 261, Shayka, 223, 240n6
279 Sherbro tribe, 45, 207
Sexual development conflicts, 221 Sick role, 276
Sexual pomiscuity, 232 Silverman, Linda, 188, 225-26
Sexual segregation, 267 Sinbyong, 224
Sexuality, 272—74 Single fathers, 85
Burmese view, 53-54 Sisterhood, 71, 269-70
control of, 272-74 Situation ethics, 155-58, 285
deprivation theory, 182-83, 261, 279 Skultans, Vieda, 183, 197, 241n8, 263
genital mutilation ritual effect, 127—28 Slavery, 30—31
in patriarchies, 273—74 Smith, Sidonie, 220
separation from fertility, 128, 142n10 Smith, W. Robertson, 136
Shaker village, 278n9 Social class, leaders, 234
Shakers. See also Lee, Ann Social context, 43-69
androgynous deity, 175-76 Social relationships. Sec Interpersonal
benefits to women, 266-67 relationships
blood ties de-emphasis, 138, I43nI9 Social roles, 283
celibacy role, 95-96, 273-74 Social structure, 257—59
characteristics, 24-26, 290 Social support, and healing, 115-17
and child death, 95-96 Sociological theory, 170
Christian Science comparison, 26 Somali women, 261. See also Zar cult
community nature of, 123 Sorcery, 35
economic independence, 270-71 Soul, location of, 86n2
food rituals, 133, 138 Southeast Asian religions
and gender dissonance, 43-45 historical context, 12-13, 41n2—3
gender ideology, 199, 209, 212n17, 213n18—20 older women in, 80
healing techniques, 116 Spirit cults. See Northern Thai matrilineal
hierarchical chain of command, 219 spirit cults
masculine monotheism rejection, 172 Spirit possession, 181-93. See also Mediums
motherhood symbols, 75—76, 81 Afro-Brazilian religion, 165
organizational structure, 247—49 audience characteristics, 185, 191
sacred text aversion, 251-52 characteristics, 181—82
sacred work, 151-52 deprivation theory, 183, 190—91
sexual segregation effect, 267 differences among religions, 42n20
sexuality/fertility control, 273—74 emic explanations, 189
spirit possession, 185, 192n3 gender differences, 182
spiritual beliefs, 147-48 and gender role behavior, 183-85
trend toward institutionalized rules, 248—49, healing function, 185
256n6 immanence of, 191-92
Shamans, 186-87. See also Korean shamans; individualistic aspects, 246
Yutas as involuntary affliction, 274-75
in Ancient Japan, 58-59 male spirits in, 183-84
 Index 329
passivity association, 197-98 Suh, David, 123, 146, 170
physiological explanations, 183 Superego, 170
pregnancy parallels, 189-90 Supernatural entities. See also Goddesses;
sexual aspects, 182-83 Polydeism
in shamans and mediums, 186—87 differences among religions, 282
social function, 182, 186, 191-92 in this-worldly religions, 145, 149-50
and women's ego formation, 187-90, Support group benefits, 261, 277n4
226-27 Surrey, Jane, 73
in women's religions, 185-86
and women's status, 274—75 Taiwanese male shamans, 143n18
Spiritist healers, 216, 222, 240n1 Tajic shamanism, 143n18
Spiritual Feminists. See Feminist Spirituality Tanabe, Shigeharu, 203
Movement Tanner, Nancy, 46-47, 67n4
Spiritualism. See also British Spiritualism; Technology, 164
Mediums; Mexican Spiritualism Teewadaa, 19
benefits for leaders, 227—29 Tenryko, 201, 211n8, 292
characteristics, 28-29, 290 Tensho-Kotai-Jingu-Kyu religion, 292
and child death, 91, 94-95 ancestor tie severance, 130-31
ethical principles, 156 androgynous deity, 176
gender dissonance, 43—45 benefits, 263
gender ideology, 197-98, 208-9 versus Buddhism, 146
healing techniques, 116 family links attitude, 282
individualized benefits of, 262-63 motherhood symbols, 76
interpersonal focus, 123—24 Tetum religion, 143n17, 292
and mother's spiritual intuition, 76—78 childbirth practices, 201
opposition to, 28 death ritual, 203
rites of reversal, 263 earth mother, 179
ritual in, 121 male roles, 143n17
sacred text aversion, 251 patrilineality, 69n17
social context, 43-45 Thailand, 41n3. See also Northern Thai
spirits in, 163-64 matrilineal spirit cults
this-worldly nature of, 148—49, 163—64 Thematic Apperception Test, 226—27
uncentralized organization, 243—44 Theology, 149
wedding ceremony, 140 and centralized organization, 252-53, 255
women as healers, 109—10 versus ritual, 6, 120-21
Spiritualist mediums. See Mediums in women's religions, 84-85
Spiro, Melford, 15, 16, 52, 53, 54, 61, 183 Theosophy, 40, 292
Spousal conflict, 107-8 gender ideology, 199, 207-8
Spretnak, Charlene, 206 motherhood in, 73
Spring, Anita, 59—60 Therevada Buddhism, 114
St. Joseph's feast, 134 Thirty Seven Nats, 161-63
Stamm, Liesa, 254 "This-worldly" religions, 145-49, 158n1, 281;
Starhawk, 140, 156—57, 159n12 See also Immanence
Stein, Stephen, 248 Tolerance for ambiguity, 84
Stover, Ronald, 169, 172 Totalitarianism, 171
Stress, and illness, 105—6 Totem, 137
Sudan, 63-64, 270. See also Zar cult Trance, 181-82. See also Spirit possession
Suffering, 103-18. See also Illness characteristics, 181-82
Christian Science interpretations, 111-13 gender differences, 182
explanation in impersonal forces, 124 Transcendence, 164
and medical mediums, 109-10 centralized religion association, 253
nat religion versus Buddhist explanation, versus immanence, shamans, 187
113-14 and literacy, 252
polydeism explanation, 171 monotheism link, 171
as preparation for leadership, 216 versus natural conceptions of death, 200
sexism connection, no remoteness from motherhood, 285
varying religious interpretations, 110—11a valuation of in men's religions, 281
330 Index
Transvestites, 235 Witches, 27
Tres Personas Solo Dios community, 200, Wives
211n6 inheritance of leadership, 239
True-Light Supra-Religious Organization, versus mothers, social status, 71
211n8 Womanism
Tugs of war, 204—5 characteristics, 27, 42n13
Tulu speakers of India, 60 nurturing interpretation, 77-78
Turkic shamanism, 143n18 Woman's intuition, 76-78
Turner, Victor, 153 Womb, tomb analogy, 202
Tutelary spirits, 96-97 Women's Commonwealth, 96
Wright, Lucy, 25, 248-49
Umbanda cult, 32. See also Afro-Brazilian Written codes, 249, 285
religions Wulff, David, 178
appeal to women, 33—34
benefits, 262 Xango cult, 32
categories of spirits, 165
characteristics, 32-34 Yaeyama Islands, 208
hierarchical organization, 218 Yako, 137
illness as recruitment path, 108—9 Yoshida, Teigo, 202
social benefits of leadership, 228-29 Young, Iris, 216
uncentralized organization, 244 Youngs, Benjamin, 213n17
Uranai, 234 Yunnanese tribe, 210
Uyigami, 192 Yutas. See also Shamans
Uzr, 283 healing techniques, 115
life histories, 221
priestess distinction, 240n7
van Genneps, Arnold, 153 Ryukyuan Islands religion, 14—15, 240n7
Victorian stereotypes, 198
Violence, 84, 87n7 Zar cult, 36-39
Virginity, 7, 272 benefits to women, 260-62
Virilocality, 59-60 blood ritual, 137-38
Vodou. See Haitian urban Vodou central role in society, 38
Voluntary celibacy, 273. See also Celibacy characteristics, 36-39, 290
Votaries, 165 and deprivation theory, 62—66
gender of deity, 177
Walker, Alice, 27 healing techniques, 116
Wasstvassa, 106-7 hierarchical organization, 218-19
Weber, Max, 5, 111, 141n2, 158n1, 171, 248 history, 63
Wedding ceremonies, 139—40, 274 human qualities of spirits, 166
Wessinger, Catherine, 215 illness explanation, 124
West Africa. See also Sande secret society initiation ritual, 128
matrifocality, 57 Islam relationship, 36-38
social context, 45—46, 57—58 and matrifocality, 63—64
Whittaker, James, 25, 247-48, 255n5 men's control of sexuality, 272
Wicca religion, 27. See also Feminist menstruation beliefs, 201
Spirituality Movement origins, 30
Widows, 78 role play in spirit possession, 184
Wijeyewardene, Gehan, 20, 21, 22, 49, 149, sisterhood in, 269-70
150, I59n3, I59n5, 222, 233 spirit possession, 154, 184-86
Wilson, Brian, 230 spirits (zayran) in, 165—66
Wilson, Peter, 186 uncentralized organization, 245
Snake goddess or priestess from Crete, seventeenth century B.C. (Courtesy of
The Museum of Fine Arts, Boston. Gift of Mrs. S. Scott Fitz)
Ryukyuan priestess praying at shrine. (Courtesy of C. Ouwehand)

Ryukyuan priestesses holding food offerings. (Courtesy of C. Ouwehand)

Ryukyuan priestesses arranging ritual food offerings. (Courtesy of C. Ouwehand)
 Sande initiates dancing.
(Courtesy of
Caroline Bledsoe)

Sande drummers with a pair

of drums: a "male" and a
"female." (Courtesy of
Caroline Bledsoe)
Sande dancer in submissive
pose, bringing tips from
spectators to older Sande
leaders. (Courtesy of
Caroline Bledsoe)

Korean Shaman. (Courtesy of

Laurel Kendall)
Shaker sisters and brothers. (From David R. Lamson, Two Year's Experience Among the
Shakers, 1848)

Shaker ecstatic dancing. (From David R. Lamson, Two Year's Experience Among the
Shakers, 1848)
Afro-Brazilian medium in costume. (From Leacock and Leacock, Spirits
of the Deep [Doubleday 1972]. Courtesy of Seth and Ruth Leacock and
Doubleday Publishers)
 Afro-Brazilian shrine. (From Leacock and Leacock, Spirits of
the Deep [Doubleday 1972]. Courtesy of Seth and Ruth
Leacock and Doubleday Publishers)

Afro-Brazilian religious participants. (From Leacock and Leacock,

Spirits of the Deep [Doubleday 1972]. Courtesy of Seth and Ruth
Leacock and Doubleday Publishers)
 Afro-Brazilian dancing. (From Leacock and Leacock,
Spirits of the Deep [Doubleday 1972]. Courtesy of Seth
and Ruth Leacock and Doubleday Publishers)

Entranced Afro-Brazilian medium. (From Leacock and

Leacock, Spirits of the Deep [Doubleday 1972]. Courtesy of
Seth and Ruth Leacock and Doubleday Publishers)
The Fox sisters: Margaret, Kate, and Leah. Nineteenth-century American Spiritual-
ists, they claimed to communicate with the supernatural using table rapping.
(Courtesy of the Department of Rare Books and Special Collections, University of
Rochester Library)
Woodcut of nineteenth-century Spiritualist seance with table rapping. (Courtesy of
the Bettmann Archive)

Priestess Selena Fox leading Mother Earth ritual, on Earth Day 1991. (Photograph
by Michael L. Abramson/Time magazine)
 Contemporary spiritual feminist ritual—birth symbolism—on the Jewish women's
new moon celebration. (Photo by Ilene Perlman from Miriam's Well: Rituals for
Jewish Women Around the Year by PeninaV. Adelman [Biblio Press, 1990]. Courtesy of
Penina Adelman and Biblio Press)