A Reader in the Anthropology of Religion [Second ed.] 9781405136143, 1405136146, 9781405136150, 1405136154

Citation preview

Blackwell Publishing

Praise for the first edition of A Reader in the Anthropology o f Religion

U|A] reader that ambitiously attem pts to represent the full breadth, depth, and complexity o f anthropology’s investigations into religion . . . The masterly general introduction situates this anthology within the long and often difficult anthropological engagement with this m ost mystified and powerful realm o f social action. . . . [AJn excellent text.”

International Social Science Review

Blackwell Anthologies in Social & Cultural Anthropology (ASCA)

Series Editor: Parker Shipton, Boston University Drawing from some of the most significant scholarly work of the 19th and 20th centuries, the Blackwell Anthologies in Social and Cultural Anthropology series offers a comprehensive and unique perspective on the ever-changing field of anthropology. It represents both a collection of classic readers and an exciting challenge to the norms that have shaped this discipline over the past century. Each edited volume is devoted to a traditional subdiscipline of the field such as the anthropology of religion, linguistic anthropology, or medical anthropology; and provides a foundation in the canonical readings of the selected area. Aware that such subdisciplinary definitions are still widely recognized and useful - but increasingly problematic - these volumes are crafted to include a rare and invaluable perspective on social and cultural anthropology at the onset of the 21st century. Each text provides a selection of classic readings together with contemporary works that underscore the artificiality of subdisciplinary definitions and point students, researchers, and general readers in the new directions in which anthropology is moving. Series Board

Fredrik Barth, University of Oslo and Boston University Stephen Gudeman, University of Minnesota Jane Guyer, Northwestern University Caroline Humphrey, University of Cambridge Tim Ingold, University of Aberdeen Emily Martin, Princeton University John Middleton, Yale Emeritus Sally Falk Moore, Harvard Emerita Marshall Sahlins, University of Chicago Emeritus Joan Vincent, Columbia University and Barnard College Emerita Published Volumes:

1. Linguistic Anthropology: A Reader, Second Edition Edited by Alessandro Duranti 2. A Reader in the Anthropology o f Religion, Second Edition Edited by Michael Lambek 3. The Anthropology o f Politics: A Reader in Ethnography , Theory, and Critique Edited by Joan Vincent 4. Kinship and Family: An Anthropological Reader Edited by Robert Parkin and Linda Stone 5. Law and Anthropology: A Reader Edited by Sally Falk Moore 6. The Anthropology o f Development and Globalization: From Classical Political Economy to Contemporary Neoliberalism Edited by Marc Edelman and Angelique Haugerud 7. The Anthropology o f Art: A Reader Edited by Howard Morphy and Morgan Perkins 8. Feminist Anthropology: A Reader Edited by Ellen Lewin 9. Ethnographic Fieldwork: An Anthropological Reader Edited by Antonius C. G. M. Robben and Jeffrey A. Sluka 10. Environmental Anthropology: A Historical Reader Edited by Michael R. Dove and Carol Carpenter 11. Anthropology and Child Development: A Cross-Cultural Reader Edited by Robert A. LeVine and Rebecca S. New

A Reader in the Anthropology of Religion Edited by Michael Lambek

Second Edition

Blackwell Publishing

Ed ito ria l m aterial an d o rg a n iz a tio n © 2 0 0 8 by Blackw ell Publishing L td BLACKWELL PUBLISHING

3 5 0 M ain Street, M alden , M A 0 2 1 4 8 -5 0 2 0 , USA 9 6 0 0 G arsin gton R o ad , O x fo rd O X 4 2 D Q , UK 5 5 0 Sw an ston Street, C arlton , V ictoria 3 0 5 3 , A ustralia The right o f M ichael Lam bek to be identified as the auth or o f the editorial m aterial in rhis w ork has been asserted in accordan ce with the UK C opyrigh t, D esigns, and Patents Act 1988. All rights reserved. N o p art o f this publication m ay be reproduced, stored in a retrieval system , or transm itted, in any form or by any m eans, electronic, m echanical, p h otocopyin g, recording or otherw ise, exccpt a s perm itted by the UK C opyrigh t, D esigns, and Patents Act 1988, w ithout the prior perm ission o f the publisher. D esign ation s used by com pan ies to distinguish their products are often claim ed as trad em arks. All brand nam es and produ ct nam es used in this book are trade nam es, service m ark s, trad em arks, or registered tradem arks o f their respective ow ners. The publisher is not associated with any product or vendor m entioned in this book. T his publication is designed to provide accurate and authoritative inform ation in regard to the subject m atter covered. It is sold on the understanding that the publisher is not engaged in rendering profession al services. If professional advice or other expert assistan ce is required, the services o f a com petent profession al should be sought. First published 2 0 0 2 by Blackwell Publishing Ltd Second edition 2 0 0 8 9

2016

Library o f Congress Cataloging-in-Publication D ata A reader in the an th ropology o f religion / edited by M ichael Lam bek. - 2nd ed. p. cm . - (Blackw ell anthologies in social S i cultural an th ropology ; 12) Includes bibliographical references (p. ) and index. ISBN 9 7 8 -1 -4 0 5 1 -3 6 1 5 -0 (hardcover : alk. paper) - ISBN 978 -1 -4 0 5 1 -3 6 1 4 -3 (pbk. : alk. paper) 1. E th n ology-R eligious aspects. 2. Religion. I. Lam bek, M ichael. B L 2 5 6 .R 4 3 2 0 0 7 3 0 6 .6 -d c 2 2 2007011995 A catalogu e record for this title is available from the British Library. Set in 9 on 11 pt Sabon by S N P Best-set T ypesetter Ltd., H on g K ong Printed in the UK The publish er’s policy is to use perm anent paper from m ills that operate a sustainable forestry policy, and which has been m anufactured from pulp processed using acid-free and elem entary chlorine-free practices. Furtherm ore, the publisher ensures that the text paper and cover bo ard used have met acceptable environm ental accreditation stan dards. For further in form ation on Blackw ell Publishing, visit our w ebsite at w w w .blackw ellpublishing.com

for N adia and Simon

Contents

Preface to Second Edition G eneral Introduction

Part I

xi 1

The Context of Understanding and Debate

19

Opening Fram ew orks Introduction

21 21

1

Religion in Primitive Culture

23

Edw ard Burnett Tylor 2

T h e Elementary Forms o f Religious Life Emile Durkbeim

34

3

T h e Protestant Ethic and the Spirit o f Capitalism M ax Weber

48

4

Religion as a Cultural System Clifford Geertz

57

Skeptical Rejoinders Introduction 5

R em arks on Frazer’s Golden Bough

77 77

79

Ludw ig Wittgenstein 6

Religion, T otem ism and Symbolism W. E. H. Stunner

82

vi

CONTENTS

7

Rem arks on the Verb “ T o Believe” Jean Pouillon

90

8

Christians as Believers

97

Malcolm Ruel 9

T h e Construction o f Religion as an Anthropological Category

110

Fatal A sad P art II

P o iesis: T h e C o m p o s it io n o f R e lig io u s W o rld s

127

Signs and Symbols Introduction

129 129

10

131

The Logic o f Signs and Symbols

Susanne K. Langer 11

T h e Problem o f Symbols £. E. Evans-Pritcbard

139

12

On Key Symbols

151

Sherry B. Ortner 13

T h e Virgin o f G uad alupe: A M exican N ation al Symbol

160

Eric R, Wolf Structure, Function, and Interpretation Introduction

167

14

168

Myth in Primitive Psychology

167

Bronislaw M alinowski 15

Folk Dialectics o f N ature and Culture

176

M arshall Sahlins 16

Land Animals, Pure and Impure

183

Mary D ouglas 17

A Jivaro Version o f Totem an d Taboo

196

Claude Levi-Strauss 18

Text-Building, Epistemology, and Aesthetics in Javan ese Shadow Theatre

2 06

Alton L. Becker M oral Inversions and Spaces o f Disorder Introduction

2 25 225

19

226

T h e Winnebago Trickster Figure

Paul Radin

CONTENTS

20

Witchcraft and Sexual Relations: An Exploration in the Social and Semantic Implications o f the Structure o f Belief

vii

238

Raym ond C. Kelly 21

T h e Politics and Poetics o f T ransgression

253

Peter Stallybrass and Allan White Conceptualizing the C o sm o s Introduction 22

C losure and Multiplication: An Essay on Polynesian C o sm o lo g y and Ritual

265 265 267

Alfred C ell 23

C o sm ological Deixis and Amerindian Perspectivism

280

Eduardo Viveiros de Castro Part III

Praxis: Religious Action

299

The M ovem ent in Ritual: Emergence Introduction

301 301

24

302

T h e Control o f Experience: Symbolic Action

Godfrey Lienhardt 25

Form and M eaning o f M agical Acts Stanley Jeyaraja Tambiab

311

26

Liminality and C om m un itas

326

Victor Turner Gender, Subjectivity, and the Body Introduction

341 341

27

“Jew ish C o m e s U p in Y ou from the R o o ts ” B arbara M yerhoff

342

28

Fate in Relation to the Social Structure

350

Meyer Fortes 29

M e d u s a ’s Hair: An Essay on Personal Symbols and Religious Experience

356

G ananath Obeyesekere 30

Spirits and Selves in Northern Sudan: T h e Cultural Therapeutics o f Possession and T rance

368

Jan ice Boddy 31

T h e Poetics o f T im e in M ayan Divination

Dennis Tedlock

386

CONTENTS

VII i

W hat Ritual Docs: T h e Foundations o f O rder Introduction

397 397

32

398

T h e Disconnection between Power and R an k as a Process

M aurice Bloch 33

Enactments o f M eaning

410

Roy A. R appaport Part IV

Historical Dynamics: Power, Modernity, and Change

429

C apitalism , C olonialism , Christianity, and Conflict Introduction

431 431

34

432

N ew Heaven, N e w Earth

Kenelm Burridge 35

T h e Genesis o f C apitalism am ongst a South American Peasantry: Devil’s L a b o r and the Baptism o f M oney

447

Michael Taussig 36

T h e Colonization o f Consciousness John and Jean C om aroff

37

Convicted by the Holy Spirit: T h e Rhetoric o f Fundamental Baptist Conversion

464

479

Susan F. Flarding 38

O n Being Shege in Kinshasa: Children, the Occult and the Street

495

Filip De Boeck Religious Ethics and Politics in the State, Public Sphere, and T ransnational Scene Introduction

507 507

39

509

Civil Religion in America

Robert N. Bellah 40

Sham anic Practices and the State in Northern Asia: Views from the Center and Periphery

519

Caroline Humphrey 41

“ Using the Past to N egate the Present” : Ritual Ethics and State Rationality in Ancient China

533

M ayfair Mei~bui Yang 42

Passional Preaching, Aural Sensibility, and the Islamic Revival in C airo

Charles Hirschkind

544

CONTENTS

43

M oral Land scap es: Ethical D iscourses a m o n g O rth o d o x and D iasp ora Jain s

ix

560

Anne Vallely 44

C an d om b le in Pink, Green and Black: Re-scripting the Afro-Brazilian Religious Heritage in the Public Sphere o f Salvador, Bahia

573

M attijs van de Port 45

M artyr vs. M artyr: T h e Sacred L an gu age o f Violence

59 0

Galit Hasan-Rokem Afterword

597

46

598

Evidence and Presence, Spectral and Other

Stephan Palmie Part V

Research Tools

611

A Guide to the Literature

613

Bibliography Index

63 0 673

Preface to Second Edition

I have used the opportunity o f an expanded second edition to redress certain om issions from the first. These include the entries by Burridge, Harding, Humphrey, M yerhoff and Pouillon. I have also added som e newer work - essays by De Boeck, H asan -R okem , Palmie, and van de Port. I have tried in particular to strengthen the sections on religious change and the relationship o f religion to politics as well as to add som e coverage o f gender, Ju d a ism , north Asia, the Caribbean, and Latin America. I realize that large ga p s in these and other areas remain but also that there is a limit to the weight o f books students can be expected to shoulder in their k n ap ­ sacks. In making my selections I was assisted by a readers’ survey conducted by Blackwell, by my very able research assistants N ina Nguyen and Shirley Yeung, and by the comments o f a number o f discerning readers, including M aurice Bloch, Fenella Cannell, Rodney N eedham , T o d d Sanders, N icolas Sihle, and doubtless others - although in most cases 1 w as not able to follow their advice. One excellent recommendation w as Igor K o py to ff’s classic essay “ Ancestors as Elders in A frica” (Africa 4 1 , 1 2 9 -4 2 , 1971, followed by the discussion in Man 17, 5 4 6 - 8 , 1982); this is easily accessed on line. Shirley Yeung is largely responsible for the upgraded bibliography and regional and topical indexes and is now co-author o f this section. Unfortunately, given the timing o f the editing and production process, we have not been able to com prehen­ sively cover b ooks published from 2 0 0 5 onw ards. I have made a few minor improve­ ments and updates to the introductions. Jan e Huber initiated the second edition; Emily M artin, Deirdre Ilkson, and others at Blackwell ably guided me through it. Helen G ray w as an efficient and reassuring copy-editor while Shirley Yeung assisted with proofing and updated the index. Jackie Solway once again provided invaluable judgment and a keen eye for the cover. A recent useful overview o f the field is by Rosalind Hackett, “ A nthropo­ logy o f Religion,” in Com panion to the Study o f Religion , edited by Jo h n Hinnells (London: Routledge), pp. 1 4 4 -6 3 , 2 0 0 5 . I have also published a short analysis o f

xii

PREFACE TO SECOND EDITION

“ A nthropology and Religion” in the O xfo rd H andbook o f Religion and Science (2006), edited by Philip Clayton, pp. 4 4 0 - 7 3 . Since the appearance o f the first edition political events and social circumstances have encouraged two trends that remain under-represented in the second edition. The first concerns the politicization both o f Islam and o f the reception o f M uslim s in Europe and N orth America. O ne site o f the latter concerns Vaffair du foulard (the headscarf controversy) and subsequent events in France (Bowen 2 0 0 4 a , b, 2 0 0 6 ; cf. Asad 2 0 0 3 , M ah m o o d 2 0 0 6 ). Secondly, in response to developments in the cognitive and neurosciences there is the grow th o f cognitive approaches to reli­ gion, o f which som e o f the m ost interesting work is Astuti (in press), Bloch (2005), and Whitehouse (2004). Partly in reaction to biblical literalism, perhaps, there is also enthusiasm in some quarters for a related “ neo-D arw inian” school (Atran 2 0 0 4 , Boyer 1994, 200 1 ). See Berliner and Sarro, eds. (2006), for more broadly based discussions on learning religion as well as de Vries, ed. (in press), for contem ­ porary interdisciplinary takes on religion at large.

A ck n o w le d g m e n ts to th e First Ed ition Heartfelt thanks to the following people for specific advice, suggestions, and in some cases course syllabi: Paul Antze, Ellen Badone, Sandra Bam ford, Sarah Caldwell, Fenella Cannell, Hilary Cunningham , Filip De Boeck, Paul H an so n , Keith H art, Bob Hefner, Ja c k Kugelm ass, Joh n Leavitt, Bruce Lincoln, Anne Meneley, Birgit Meyer, Peter Pels, Parker Shipton, Jackie Solway, Jon ath an Spencer, Charles Stafford, Anne Vallely, Andrew Walsh, Nireka Weeratunge, Brad Weiss, Dick Werbner, Aram Yengoyan, and a remaining anon ym ous referee o f the original table o f contents. M y apologies if 1 have missed som eone. Paul Antze, Sam Bam ford, Janice Boddy, Anne Meneley, and Andrew Walsh provided very useful feedback on a penultimate version o f the table o f contents and introductory material. Their com m ents have improved the book. The contributors responded to my request for verification o f abridgm ents with go o d humor and advice (which I could not alw ays follow). Sarah G ould served as an efficient assistant at my end, as did Sarah Colem an at Blackwell. Juan ita Bullough (copy editor), Deirdre R ose (index), Annette Chan, and Audrey Glasbergen provided outstanding assistance. The reader w ould not have been con­ ceived or produced without Jan e H u b er’s steely editorial persistence, so well sheathed in charm and diplomacy. Her advice, enthusiasm, and lunches have been critical. Jackie Solway provided even more advice, encouragement, and meals, and N ad ia and Simon Lam bek urged me enthusiastically to get on with it. M . L.

General Introduction

This anthology dem onstrates the strength and vibrancy o f the anthropological study o f religion across the 20th century. The collection is neither purely historical in orientation, nor a survey o f contem porary w ork, and is not a comprehensive cover­ age o f the broad variety o f topics anthropologists o f religion address. Rather, it views the anthropology o f religion as a live tradition o f intellectual inquiry and attem pts to show broad connections within a body o f work. It charts the recurrence or perdurance o f certain central questions, even if, as Geertz pithily put it, “ Anthro­ pology . . . is a science w hose progress is m arked less by a perfection o f consensus than by a refinement o f debate. What gets better is the precision writh which we vex each other” (1973a: 29). The anthropological tradition in the study o f religion is an “ extended conversa­ tion,” now dating back well over a century. It is difficult to fully understand con­ temporary interventions without tracing back the roots o f debate. Yet the conversation has been long and com plex, and the essays included in the anthology simply provide som e o f the highlights o f that conversation. They are essays signifi­ cant for the originality, elegance, or clarity with which the authors develop their positions or the rich ethnography with which they illuminate particular aspects o f religious practice. M o s t o f them have long been recognized a s such and thus are widely cited. Conversely, much significant work has had to be omitted for consid­ erations o f space. The anthology is addressed to readers both within and outside anthropology w ho seek a direct entry to what anthropologists have (had) to say about religion. It is suitable for students at various levels; the contributions are scholarly and vary in their difficulty. Their sequence in the volume is based on the application o f several criteria: the time at which they were written, the order in which they m ade an impact, and the theme or topic. The contributions can, o f course, be read in any order.

2

GENERAL INTRODUCTION

T h e A n th ro p o lo g ical Tra d itio n Anthropology as a field has long been open to interdisciplinary work and, indeed, may even be conceptualized as a kind o f transdiscipline. As the contents o f this volume reflect, not all o f the most significant contributions have come from people identified as anthropologists. Nevertheless, participants in the anthropological con­ versation on religion can be said to share a certain broad and evolving outlook with respect to the kinds o f questions they find useful or interesting to ask, the methods appropriate to address rhem, and the fram ew orks through which contributions are received. This outlook overlaps with, but is distinct from, that found in alternate approaches to religion characterized as theological, philosophical, sociological, historical, and so on. The specifically anthropological conversation on religion, which, to be sure, has com prised many strands, heated arguments, and rich digressions, can be broadly characterized by the following features (which are often ascribed to the field o f anthropology as a whole). An anthropological approach comprehends arguments that are variously holistic, universalistic, ethnographic, comparative, contextual, historical, dialogical, and critical. By holism is meant that anthropologists see “ religious” facts as parts or dimensions o f larger social and cultural wholes, do not begin by immediately dem arcating a “ religious” sphere from that o f the nonre­ ligious (or one religion from another), and understand the presence or production o f such distinctions as problematic, hence as part o f their subject matter. A holistic approach exam ines the inextricable links a m o n g religion and, for example, the social reproduction o f families, gender hierarchy, political organization, and modernity. Anthropology is universalistic in that it takes the whole range o f human societies, past and present, as its subject matter and attempts not to privilege the western tradition or literate societies. T h us, along with universalistic coverage comes the continuous attem pt to address ethnocentric bias and the need to produce models that make sense both o f diversity and unity. In practice, anthropology’s universalism has meant it w as the only discipline to take seriously the existence o f small-scale societies without traditions o f literacy. A nthropology’s ethnographic perspective means that our analyses are usually developed through the study o f specific societies and even o f smaller units individual communities, congregations, religious experts, particular events, etc. within them. The view is often “ up close and person al,” or draws from data collected by means o f intensive and intimate fieldwork that aims, in the first instance, to understand a local way o f seeing and doing things. The result is often a book-length narrative portrait (an ethnography). Yet anthropology is also resolutely com parative, insofar as the particularistic ethnographic accounts must be m ade to speak to each other and to a developing (and frequently debated) analytic language so that we can think, at least provision­ ally, by m eans o f such concepts as “ deities,” “ the sacred,” “ ritual,” “ sacrifice,” and so on. These terms come to take on particular meanings within the anthropological conversation that they do not necessarily hold outside it. Overall, anthropology must walk a fine line between exoticizing or idealizing, on the one hand, and b an al­

GENERAL INTRODUCTION

3

izing, on the other, the statements, practices, and experiences that exercise those whom we study. The approach is contextual in that anthropological “ fa c ts” are always understood relative to their contexts. The elaboration o f context - linguistic, cultural, social, political, etc. - is a central feature o f anthropological interpretation, yet the relative weight given to specific aspects o f context - e.g., the structural versus the pragmatic - has often been the location o f our most vigorously contested arguments. Increasingly, context has been understood in terms that have been called histori­ cal; in particular, this temporal (and spatial) locatedness has meant understanding personal and collective agency with respect to past and present political and eco­ nomic forces, notably those o f colonialism, capitalism, and emergent globalization. Our approach is also historical insofar as we attend to local (cultural) forms o f historical consciousness and o f making and addressing change; frequently these are couched in religious idioms and practices. The anthropological perspective is dialogical in that our method is rooted in conversation with, and especially in listening to, those whose practices, knowledge, and experience we attem pt to understand. Questions o f how far to privilege which kinds o f voice and how to comprehend diversity, including contextually muted voices, have become increasingly significant. M ore abstractly, anthropology is dialogical insofar as foreign concepts and practices need to be described in terms o f concepts and practices available to the broader community o f anthropologists (including resources from philosophy and other fields). The foreign, in turn, co n ­ tributes to the elaboration and sophistication o f anthropological language. Finally, the anthropological conversation can be characterized as critical, meaning by this both ‘‘ critical” in the sense o f literary critical and “ critical” in the political sense o f concern with power and its subterfuges and abuses. The best anthropology is a lso self-critical; here our concern with overcoming the various and multiple forms o f ethnocentrism and intellectual narrowness remains a characteristic feature o f any contribution that wishes seriously to be taken as anthropological. This is not to assert that we have overcome our biases and constraints, and especially not to assum e that we are som ehow more advanced along the slope o f enlightenment than either our interlocutors or our intellectual forebears. As philosopher David H oy has put it, “ Reflection on the partiality o f past interpretations dem ands reflection on the partiality o f the present” (1978: 167). Anthropologists have also, over time, variously emphasized evolutionist, rational­ ist, functionalist, social structural, structuralist, symbolic, interpretive, political, M arxist, social constructionist, phenomenological, psychoanalytic, poststructural­ ist, cognitive, aesthetic, and ethical approaches or m odes o f understanding. As an intellectual tradition, anthropology is marked less by consensus than by the way such arguments have taken account o f each other.

A n th ro p o lo g y on Religion Questions o f religion have been central to anthropology since its beginnings, whether we place these in the 19th century with thinkers like T ylor or Robertson-Smith, or

4

GENERAL INTRODUCTION

in the 18th century with Vico. Indeed, these are a m o n g an th ro po lo gy ’s hardest and most enduring questions - questions about difference, rationality, community, modernity, symbolization, meaning, relativism, mimesis, projection, mediation, power, order, hierarchy, harm ony, conflict, alienation, love, well-being, dignity, aesthetic coherence, creativity, playfulness, reproduction, fertility, maturation, death, tedium, excitement, motivation, suffering, and redemption. The an th ro p o ­ logy o f religion points to the unique conjunctions o f morality, desire, and power, o f subjection and freedom, o f worldliness and asceticism, o f ideal and violence, o f imagination and embodiment, immanence and transcendence, inwardness and outw ardness, origins and ends, order and chaos, structure and practice, co sm os and history, that have constituted distinctive human worlds. Contem porary anthropology o f religion draws from a number o f sources. From the Americanist (Boasian) tradition it d raw s a sense o f the intimate connection between religion, language, and poetics, and the im portance o f consulting and understanding the experience o f gifted individual practitioners (e.g., the entries by Radin and T edlock, chapters 19 and 31). From Durkheim (chapter 2) com es the understanding o f religious phenomena as social, as well as the intimate connection between function and meaning in religious ritual and representation, and between religion and social order (e.g., the selections by Turner and R ap p a p o rt, chapters 26 and 33). From the evolutionists com es the question o f the rationality o f primitive thought, a question that, while now moor in and o f itself, continues to inform dis­ cussion a b o u t the logic and legitimation o f any particular form o f religious d eclara­ tion or practice (e.g., the essays by D ou glas and T am b iah , chapters 16 and 25). From the evolutionists as well, and variously mediated by phenomenologists and psychoanalytic thinkers, come questions concerning the relative weight o f thought and experience, the intellectual and the practical, the rational and the nonrational, the conscious and the unconscious in the religious sphere (e.g., the essays by Gell and Obeyesekere, chapters 22 and 29). From M a r x com es attention to power, alienation, fetishism, and mystification (e.g., Bloch and T au ssig, chapters 32 and 35). From Weber (chapter 3) com es attention to the com parative aspects o f the text-based religions, to the place o f religion in transitions to modernity, to the links between religion and the wellsprings o f both political and economic action, and finally to the question o f theodicy, the puzzle o f locating meaning in the givenness o f the world, suffering, and death (e.g., Geertz and Hirschkind, chapters 4 and 42). All these questions, in turn, have roots in ancient thought, in Plato’s distinction between mimesis and philosophy, in Aristotle's discussions o f ethics and poetics, and in the portrayals in texts like O edipus and J o b (to draw from the titleo f chapter 28 by Fortes) o f fate and justice. M ost vital o f all has been the stream o f ideas em anating from the variety o f reli­ gious traditions (including those o f the West) that anthropologists have studied. N o t only have they supplied som e o f the key terms with which we have routinely worked - totem, tab oo , m ana, karm a - but our experiences in the field and the lessons we have learned from our interlocutors have shaped the way we ask and answer ques­ tions, and have had enorm ous im pact on the nature o f our reception o f the various theoretical currents mentioned above. The anthropological tradition understands religious w orlds as neither fully objec­ tive nor fully subjective phenomena, but as poised in the mediating space o f culture

GENERAL INTRODUCTION

5

or the social and as participating in a dialectic that both objectifies and subjectifies. G o o d anthropology understands that religious worlds are real, vivid, and significant to those who construct and inhabit them and it tries, as artfully as it can, to render those realities for others, in their sensory richness, philosophic depth, emotional range, and moral com plexity. In acknow ledging the value and power o f such worlds, but also their variety and competition, anthropology must understand them as so many means for acting, asking, shaping, and thinking, rather than as a set o f fixed answers whose validity either can be independently assessed (objectivism) or must be accepted as such (relativism).1 With respect to the ultimate truth o f religion (or specific religious traditions), anthropologists have offered a variety o f answ ers; perhaps the m ost popu lar has been the phenomenological one o f simply bracketing the question as outside o f the scope o f discussion. Another has been to accept the social reality o f religious phe­ nomena for those who adhere to them. A nthropologists d o ask how religious truths become socially realized and confirmed, and also recognize that flourishing symbolic worlds contain the means to reflect back on themselves, to dwell on p a ra d o x , and to enable internal debate, skepticism, and relativity. The anthropology o f religion does not shy aw ay from recognizing the human constitution o f any given religious formation, nor from elucidating the particular mystifications necessary to the construction or enactment o f religion. Anthropology explores the w ays religious practices are embedded in, or complicit with, specific forms o f sociality, regimes o f power, historical struggles, and modes o f production. In sum, the anthropology o f religion engages with politics and history yet refuses to reduce religion to either o f these. Anthropology has som etim es been seen as poised between idealism and material­ ism, although few contributors can be said to fall into either form o f reductionism. Ideas are formed within and therefore with respect to specific material conditions, bur at the sam e time, material conditions can only be grasped and evaluated within or with respect to a specific system o f ideas, semiotic code, or way o f looking at the world. Symbolic and material forms, practices, and contexts change at different rates; no religion worth the name is epiphenomenal. At the sam e time, such dynamic systems and their proponents are often, but not alw ays, in active competition with each other. The nuances o f how all this w orks out in practice or how it is to be understood in theory have been the object o f much debate - hardly any two o f the contributors to this volume take precisely the sam e stand on the issues - but the recognition o f some kind o f dialectical relationship is central. The dialectic may be phrased as base and superstructure, mind and body, structure and practical reason, nature and culture, myth and history, imagination and reality, sacred and profane, ritual and common-sense knowledge. As the w ord “ dialectic” suggests, these terms are not simply oppo sed to one another; we attem pt to work through the successive interrelationships, contradictions, and m ediations to which they give rise. This leads further into the question o f how such relations are understood within any given cultural tradition and how we, in turn, are to interpret that understand­ ing. Are sacred and profane, myth and history, etc., locally distinguished from one another, or is the universe grasped in more holistic terms? T h us Levi-Strauss, for exam ple, argued that totemism is a structure in which can be seen the play o f nature

6

GENERAL INTRODUCTION

and culture, and attem pts to both mediate between them and to think through each domain by means o f metaphors drawn from the other. A persistent theme in the anthropological conversation has concerned the logic o f religious thought and practice. If many o f the 19th-century thinkers saw members o f smaller-scale societies mired in superstition, ignorance, bliss, or folly, the personal connections forged in decent ethnographic fieldwork immediately deprived west­ erners o f any illusions o f intellectual or moral superiority. Concom itantly, theorists began to realize that rather than com pare nonwestern systems o f thought directly to western science, it was much more sensible both to com pare religion, com m on sense, and specialized knowledge a b o u t the world within any given society (and hence how categories and hierarchies o f knowledge are formed and function in the first place), and to com pare western religious practices with nonwestern ones. Thus the problem shifted from explaining the ostensibly irrational religious practices o f others to understanding the nature o f religious practice anywhere. But things are more com plex than this sounds, since the very distinction between religion and science, or for that matter the natural and the supernatural, does not characterize all societies. N o r does it characterize consistently the attitudes o f all members o f any given society. M any anthropologists thus begin with a more holistic picture, or one in which the religious co sm o s provides the place o f the whole or the grounds for truth that a m on g secular thinkers has com e to be replaced by such concepts as “ nature,” “ society,” or “ m ind.” Hence the need to grasp beliefs and practices relative to the universes in which people live, whatever the specific refer­ ences to g o ds, prayer, and the like. It follows from this that the anthropologist’s own co sm os is but one a m on g many; our position is possibly an epistemologically privileged one, but it is nor located in som e neutral space outside o f culture. T hus objectivism, in the sense o f assum ing such a privileged “ A rchim edean” position, must be abandoned. T o leave things here, however, may be to give too much to relativism, since it is clear both that all humans do have much in com m on and that we can learn to understand each other or meet each other halfway. Cultural worlds are open to each other, not closed. M oreover, we can ask how particular worldviews are co n ­ stituted and explore their underlying structure (whether conceived as gram m atical, propositional, projective, tropic, or performative and rhetorical). We can also ask how such world “ views” are com posed , produced, and legitimated; that is, how and in what respects they com e to be realized as “ w o rld s,” taken as real, as “ n atural” rather than socially constructed, as true rather than contingent. Indeed, following R ap p ap o rt (19 9 9 ; chapter 33 below), we ask how certain religious acts and utterances become the ground against which social acts and cultural facts can themselves be established, validated, or, to use a religious w ord, sanctified (as happens, for exam ple, under oath); and how they relate to forms and distributions o f political power, to changing social, material, and economic constraints and opportunities, as well as to psychic processes both rational, nonrational, and irra­ tional. We can also recognize that cultural and religious worlds, including our own, are not necessarily internally consistent but may juxtapose both contradictory and incommensurable concepts, practices, and traditions. The epistemological stance that encom passes this diverse range o f questions and that ap p ro x im ate s the location

GENERAL INTRODUCTION

7

o f many thoughtful contem porary anthropologists might be referred to very broadly as critical hermeneutics . Clifford Geertz, in the 1966 essay abridged as chapter 4 below, com plained that mid-20th-century anthropology o f religion had not been theoretically adventurous enough. Freud, o f course, was in vogue at the time and anthropologists continue to draw on him as well as on the subsequent, if at present som ew hat rocky, career o f psychoanalytic thought. But Geertz had in mind the philosophers. If Geertz drew from Langer (see chapter 10) and Ryle, subsequent writers have found greater inter­ est in the work o f Ryle’s colleague, Austin, on the performative dimension o f speech (see chapters 2 5 and 33). Levi-Strauss (chapter 17) was also on G eertz’s horizon. Levi-Straussian structuralism may be seen as a philosophical transform ation o f Durkheim by way o f Saussurean linguistics. Others have been more influenced by the version o f semiotics initiated by Peirce or by the hermeneutic critique (and perhaps appropriation) o f structuralism elucidated by Ricoeur. M ore recently, some anthropologists have discovered continental phenomenology (Merleau-Ponty), exis­ tentialism (Heidegger), or Lacanian psychoanalysis (chapters 38, 45 ), while others have drawn from the M arx ist tradition, notably from Gramsci and R aym ond Williams, or from Foucault and the poststructuralists (see chapters 9 and 35). Other philosophers have a longer reach; Aristotle is a figure addressed, to quite different ends, in the contributions by Becker and T au ssig (chapters 18 and 35). Since the 1950s, literary criticism has been in dialogue with anth ro po lo ­ gical structuralism; anthropologists then benefited from the new energy released by the discovery o f Bakhtin and Benjamin. The former connection is represented in the selection from Stallybrass and White (chapter 21), the latter in that of Hirschkind (chapter 42). Yet other anthropologists have been increasingly attracted to the insights o f cognitive psychology in the am bitious attem pt to build a universal (objectivist) theory o f human knowledge that would incorporate reli­ gious ideation. In sum, if the parochialism o f which Geertz complained has not been entirely superseded, it is now complemented by attention to a wide range o f theory. C ontem porary anthropology is also engaged with globalism and with the effects o f the expansion o f the West under mercantilism, colonialism, and present-day, “ late” capitalism. With respect to religion this concerns confrontations between so-called “ w o rld ” or universalizing religions like Christianity and more localized forms, as well as the perceived competition a m o n g universalizing tradirions. Some anthropologists have seen small-scale societies gladly replacing their parochialism with broader cosm ologies, while others have emphasized the coercive and hege­ monic effects o f missionary activity or processes o f religious rationalization. T od ay , in many parts o f the world, people are switching from older forms o f missionary Christianity to Pentecostalism and anthropologists are engaged in trying to figure out why (van Dijk 1998, M artin 1990, Meyer 1998, 1999). Given the geopolitical situation and the dem onization o f Islam in the western press, other anthropologists have attempted to present more informed accounts o f M uslim versions o f modernity (Eickelman and Piscatori 1996, Hefner 1998, 2 0 0 0 ). Yet others co m pare seculariz­ ing and religious social movements, exam ine ethnic, regional, or class conflicts that come to be expressed in religious terms, or try to understand the increasing

8

GENERAL INTRODUCTION

ethnicization or hardening o f religious boundaries in places once known for more open, tolerant, holistic or incorporative social Helds (James, ed. 1995). T hus, where anthropological accounts o f religious diversity in South Asia were once shaped by the Redfieldian contrast between “ b ig ” and “ little” traditions (Singer 1972, Fuller 1992), they now attend to political mobilization, nationalism, and collective vio­ lence (T am biah 1992, 1996, Hansen 1999; Van der Veer 1994). Religion has also been understood as a form o f social protest, a s evident in the literature on cargo cults and millenarian movements (Burridge 1969, Worsley 1957; chapter 34 below). Som e anthropologists exam ine the place o f religion in rational­ ized political interventions, such as the Sanctuary movement on the M exican/US border (Cunningham 2 0 0 0 ), Latin American liberation theology, G andhiism , and environmentalism (see chapters 4 3 , 44). Anthropologists have also been concerned with the relationship between state policy and the form ation o f religious institutions and boundaries, with the cult o f political figures, and with the impact o f religious interests and differences on national and transnational politics. The continuing vitality o f religion as a social and political force is amply illustrated by events and activities such as the Iranian revolution, the rise o f the religious right in the United States (chapter 37), the increasing success o f Pentecostalism in Africa and Latin America, the role o f Buddhism in Sri Lankan and Tibetan national struggles, and the growth o f saint cults in Israel. Yet this anthology emphasizes the religious and sym bolic construction o f the world more than the way such constructions form responses to it. This is one o f the things that anthropologists study particularly well and whose elucidation is alm ost exclusively in the hands o f anthropology as a discipline. The anthropological perspective here is neither that o f the theologian, whether as interested insider or outsider, nor that o f the distanced observer who finds any particular local set of practices simply wrong, superficial, or epiphenomenal, whether on epistemological, practical, or political grounds. T o g rasp the attraction and power o f religion, or its motivating force in political and economic affairs, one must, as Weber put it so well, understand the world it proposes and in which it situates its actors. A world in which it can be stated that “ time is m oney” (see chapter 3) is o f a very different order from one where it can be stated with equal matter-of-factness that “ men are p a rro ts” (Crocker 1977). While it would be absurd to deny that politics and econom ics affect religious realities and actions, a view o f religion as merely superstructural is disturbingly close to the dom inant ideology o f bourgeois capitalism. Rather than accepting that view with undue complacency (whether from a liberal or a M arxian perspective), it too must be understood in com plex relationship with the political and economic cir­ cum stances that inform it and that it informs. There is also the question o f how tightly sealed or inevitable one finds such con­ nections. A nthropologists vary from positions o f rather strict determinism, whether o f M arx ia n , Durkheimian, Freudian, or culturalist roots, to positions o f Weberian elective affinities or ones that emphasize contingency, ambiguity, and imagination. And historically, one can note contrasts between societies or sectors o f society that readily gave up particular religious structures and practices in the face o f more powerful social realities, and those that tried to hang on steadfastly.

GENERAL INTRODUCTION

9

Relig io n Itself? Insofar as religion form s the cultural “ gro u n d ” (or “ w orldview” ) o f a society or serves to articulate its m ajo r concerns, it is intimately linked to such matters as human conception, kinship, and the life cycle, to the environment, ideas o f human sociality, affinity, and exchange, conceptions o f and conditions for hum an dignity, justice, creativity, and general well-being. Hence we are led to ask both how subjects are constituted and how agents are motivated. The study o f religion directs us to daily observance o f prayer and dietary and cleanliness taboos, to judicious moral action in the face o f specific events and tow ard the collective go od (practices that Aristotle called phronesis), and to the adventures o f sacrifice, pilgrimage, spirit possession, devotion, study, vision quest, prophecy, and the like. Religion also anticipates failures, om issions, inversions, and distortions; these are imagined in myth, liminality, pollution, sorcery and witchcraft, dem ons and spirits, and som e­ times in the go ds themselves. But just what is religion? The prevalence, significance, and apparent universality o f religion, and hence the scope o f any field devoted to its study, depend on how we define it. Yet such a definition is by no means obvious. Few other than the writers o f introductory textbooks would be satisfied with a concrete and narrow definition (derived from T ylo r’s account o f anim ism , chapter 1 below) such as “ belief in supernatural beings.” This sort o f definition begs many questions: W hat is “ super­ n atu ral” except relative to som eone’s idea o f the “ natural,” a concept surely o f rather recent European vintage? Why a specifically dualist model? Why “ beings” and only “ beings” (or spirit, soul, G o d , or whatever the substantive part o f the definition)? What is meant by “ belief” (chapters 7, 8)? H o w would religious belief differ from ordinary knowledge or other forms o f conviction? Durkheim long a g o moved well beyond “ intellectualist” definitions that depend not only on a dubious concept o f belief but on the assum ption o f an individual rational thinker at the source. He offered instead a definition that is structural, “ relative to sacred things.” The sacred w as defined as that which w as set apart and therefore could vary in substance from one social order to another, the particular content not being especially relevant for understanding the form. D urkheim also dem onstrated the collective rather than the individual basis o f sym bolic classifica­ tion, o f which sacred/profane is a central opposition (chapter 2 below). M ore recently R ap p ap o rt has developed a model o f the sacred as “ a property o f discourse” (1 9 9 9 : 281) and elaborated formal properties o f sanctification (chapter 33 below). “ Belief in supernatural beings” would hardly w arrant Susanne Lan ger’s porten­ tous description o f religion as “ the m ost typical and fundamental edifice o f the hum an m ind” (1948: 33). Indeed, it w as a minimalist definition which enabled certain 19th-century thinkers to look with disdain upon “ primitive societies” whose religions were assum ed to be made up o f little more than primary feelings o f fear or aw e tow ards the spiritual beings in w hom they so bluntly and naively “ believed.” O f course, broader definitions that encom pass the moral have not been lacking either. Valeri, for one, sum s up religion as “ the objectified system o f ideas o f a c o m ­ m unity,” a Hegelian idea found not only in Feuerbach and M a r x , but in

10

GENERAL INTRODUCTION

Robertson-Smith and Durkheim (Valeri 1985: x). Writers such as Geertz, D ou glas, and Levi-Strauss offer variants o f such an approach. In this collection perhaps Stanner (chapter 6) offers the m ost robust substantive depiction o f religion. A p a r­ ticular twist is to be found in the lineage from M au ss to D um ont and Schneider, who note that the separation o f religion from politics or economics or more broadly o f “ co sm os from society” (de C o p p e t and Iteanu, eds., 1995) may be specific to the West. T he most interesting attempts at defining or modeling religion emphasize meaning and order. They thereby both identify religion with other dom ains o f thought or symbolic practice (that is, with culture understood in symbolic, semiotic, or discur­ sive terms) and distinguish its place relative to these domains. T he distinction is on logical, structural, or functional grounds rather than content. Thinkers like Geertz, R ap paport, Bloch, and Asad variously enquire how conditions o f meaning, m ean­ ingfulness, truth, and certainty are produced, guaranteed, and underpinned. T hose mechanisms which most strongly establish, anchor, contextualize, and regulate meaningful order and orderly meaning may be called “ religious.” Em phasis is placed on world construction, entailing an imaginative poiesis by means o f symbols, tropes, and performative rituals, and on locating and orienting people a s subjects and agents within such worlds. M eaning and order are thus to be discovered simultaneously at both the conceptual and m oral levels and questions o f co sm os, eschatology, power, authority, motivation, and discipline are closely connected. Meaningful order becomes all the more clearly marked as anthropologists ad o p t theories that emphasize the fluidity, divisiveness, and creativity o f culture and acknowledge the fragmentation, shallowness, uncertainty, and speed that characterize capital-driven postmodernity. Characteristic o f late-20th-century thought has also been a reaction against pro­ ducing definitions for heuristic purposes and a move tow ard reviewing the genealo­ gies o f terms. Genealogists ask how particular definitions emerge and com e to objectify their referents, how a concept like “ religion” com es to appear as either an ostensibly “ natural” category o f thought or the referent o f a naturally existing “ o rg a n ” o f human society. In doing so it becomes apparent that any definition is relative to the concerns o f its time. Thus, in a trenchant essay, religious studies scholar Jo n a th a n Z. Smith docum ents the shift from a “ C ath o lic” conception o f religion close to ritual to a “ Protestant” one close to piety (1998: 271) and argues moreover that, “ ‘ Religion’ is not a native category . . . It is a category imposed from the outside on some aspect o f native culture” (ibid.: 2 6 9 ) (cf. A sad , chapter 9 this volume). Edw ard Sapir early made the point that “ a very useful distinction can be made between ‘a religion’ and ‘ religion.’ The former appears only in a highly developed society in which religious behavior has been organized by tradition [a Weberian point); the latter is universal” (1 9 5 6 [1928]: 120). Nevertheless, as Smith remarks, the very distinction produces such questions as whether “ the diverse ‘religions’ |form | species o f a generic ‘ religion’ ?” (1998: 275). Smith concludes that M elford Spiro’s (1966: 96) definition o f religion as “ an institution consisting o f culturally patterned interaction with culturally postulated superhum an beings” (a much more sophisticated version o f the intellectualist

GENERAL INTRODUCTION

11

definition I presented above) has gained the widest assent a m on g scholars o f reli­ gion. But he remarks astutely that it “ places human cultural activities or institutions as the summum genus and religion as a subordinate ta x o n ” (1998: 2 8 1 ). Spiro, o f course, is an American cultural anthropologist. If som e early anthropologists saw Christianity as the only “ true” religion, the response that other religions could be equal or equivalent carried its ow n ethnocen­ tric bias - namely the assum ption that religions were to be recognized by their simi­ larities, substantive or analogical, to Christianity, which continued ro serve implicitly as the prototype religion. And yet, o f course, the very nature o f Christianity, defined through its relation to larger cosm ological and social wholes, has changed over time - from its origins as a rebellious millenarian movement in a R om an colony, to the days when it anchored medieval society, to the rise o f Protestant sectarianism , and then the subsumption o f religion within capitalist modernity (com pare Cannell 2006). The conception o f religions as distinct, bounded entities closely follows Christian ideas that religions are mutually exclusive, that one can commit to only one at a time, and that commitment occurs through a form o f inner “ belief” or “ faith” (see the essays by Ruel and A sad, here chapters 8 and 9, respectively). Yet in many parts o f the world religious ideas have diffused and been accepted or rejected in whole or part much as any other elements o f culture; inner conviction is not privileged over outer observance; and conversion is not a salient category. Indeed, it is hard to see how it could be otherwise in contexts where religions are not objectified enti­ ties but simply dimensions o f ways o f life. Syncretism, transformation, and pluralism are all more likely responses. Analysts such as Werbner (1977), R anger (1 9 9 3 a , b), and Wiessner and T u m u (1998) have noted the spread o f precolonial regional cults and in much o f the world today we see the presence o f incommensurable discourses without the idea o f mutual exclusion, competition, or conflict. One may note the multiple forms o f religious practice (Buddhist, Shinto, etc.) in Ja p a n , the copresence o f Confucian and T ao ist temples in Vietnam, or the jostling o f Islam and spirit possession in places as distinct as M orocco, Sudan, and M ad aga sca r. The sam e has been noted for Buddhism and spirit cults in Burma and Thailand and for Catholi­ cism and v oodoo in Haiti. The Indian subcontinent has provided its o w n forms o f open pluralism, despite the emergence and sectarian quarrels o f Buddhists, Jains, and Sikhs, the rise o f “ co m m un alism ,” and the complicity o f Islam, Hinduism , and Buddhism with nationalism. In the United States one can see the conjunction o f Christianity with civil religion, though the two have become perhaps all too closely linked in the time since Bellah wrote his essay (chapter 39). Similarly for Roman Catholicism and nationalism in France, although there the former has had a strong opponent in rationalism (laicite ). T he prevalent conjunctions o f such incommensurable or even contradictory sets o f ideas and practices suggest that the Dumontian unity o f cosm os and society may be too tightly conceived and that structuralist accounts are often a go o d deal tidier than the phenomena they claim to model. Nevertheless, structuralism serves the highly salutary function o f moving us aw ay from the naivete o f simple behaviorism and often reveals alternate forms o f ordering the world and profound levels o f con­ nection and relation which strictly empiricist com parison would certainly miss.

12

GENERAL INTRODUCTION

In sum , we can see that anthropology is caught between, on the one hand, scru­ tinizing, relativizing, and in effect deconstructing the general, dominant, and osten­ sibly universal categories by which it begins its inquiry and, on the other, seeking more general categories by which it can carry out com parison and describe conclu­ sions. Som e o f our best work has been deconstructionist avant la lettre : Steiner’s T aboo (1956), Evans-Pritchard’s Theories o f Primitive Religion (1965), and LeviStrau ss’s Totemism (1963e) spring immediately to mind. Conversely, other signifi­ cant essays have found new ways to generalize across cases: van Gennep, The Rites o f Passage (1 9 6 0 (1908]), Hertz, Death and the Right H and (1 9 6 0 11909 J), M auss, The G ift (1 9 9 0 [1925]), D ouglas, Purity and D anger (1966), Burridge, Neiv Heaven Neiv Earth (1969), Turner, The Ritual Process (1969), and Bloch and Parry, eds., Death and The Regeneration of Life (1982) are a m on g the most notable. T he ethnographic literature invites theoretical comprehension o f such evident similarities as the attention given to ancestors in East Asia and Africa or to men’s houses and sacred flutes in Am azonia and Papua N ew Guinea - not to mention the near universality or family resemblance o f things we can call sacrifice, taboo, prayer, or initiation - without apparently excessive violence to the individual cases. T o be sure, some concepts have proved less durable than others, and the deconstructions o f previously dominant and reified terms such as totemism have been critical. The structuralist movement attempted to replace reified categories with structures o f relations, and N eedh am (1975) usefully drew attention to the idea o f polythetic classification (classes in which each member shares something in com m on with at least one other member but in which there may be nothing shared by all members). These are lessons that anthropology should have internalized and can still draw from productively. What anthropology cannot d o (and has not done) is take an exclusively genealogical or deconstructive position. 1'his is so for two reasons. First, with respect to the ideas and practices o f those we study, we can be skeptical a b o u t specific religious objects or arguments, but not a b ou t the idioms and acts through which they are evidently constructed. And we have no business taking the missionary position, even when the creed we advocate is secular rationalism (or postm odern irrationalism). Second, with respect to our own field o f study, we continue to need a language into which to translate our findings and in which to speak. In any case, this book has been conceived on the edge o f this dilemma, begun with the notion that compiling a set o f articles under the rubric “ religion” w as a valuable exercise, while recognizing equally the danger o f reifying its subject. Defini­ tions, contents, and boundaries should be held in question and the frame should be understood alw ays as only one a m o n g many through which any particular set o f human phenomena can be viewed. What is advocated is not the discovery o f more precise definitions but the acknow ledgm ent o f the contingency, provisionality, and implications o f any definitions we choose to deploy. In particular, we need to be intellectually self-conscious o f how we might talk about “ religion” in those societies that do not have a discrete field that m aps onto the contem porary western notion o f the religious, or that cannot be judged by norms established via the monotheistic Judeo-Christian-Islamic lineage. And we need to be more open and creative in imagining “ religion” in societies that do and can.

GENERAL INTRODUCTION

13

While outsiders look to anthropology for generalizations, anthropologists them­ selves, some notable exceptions aside, remain happier exhibiting differences, e xp lo r­ ing particularities, following connections, tracing patterns, elaborating context, dem onstrating com plexity, illuminating p a rad o x e s, or simply poking holes in the universalizing theories o f other disciplines. O u r craft is one o f recognizing incom ­ mensurabilities and their limits; hence we remain more com fortable with raising questions than with answering them.

In Sum The problems religion raises have been cenrral to anthropological thought from its origins in the 19th century through to the present. As religious belief began to weaken in their own societies, Victorian thinkers focused ever more closely on its presence elsewhere. Early writers wrote a b ou t peoples w ho seemed to have too much belief or not enough, and well into the 20th century they were obsessed with defining religion and discriminating between it and its lesser cousins - magic, super­ stition, etc. Definitional questions also pervaded discussions o f myth and ritual until they were rendered less consequential by the structuralist revolution. But we still scratch our heads over the boundaries o f phenomena - the relationship o f “ limino id ” activities like nightclubbing, d ram a, or ca m p reunions to the “ liminal” phases characteristic o f rites o f passage or pilgrimage; o f purity and pollution activities and the relationships between bodily discipline or healing and religion; o f political ideology or scientific authority and religious d ogm a; o f concepts o f spirit or soul and concepts o f self or mind; even the question o f “ belief” itself and whether it can dem arcate religious from ordinary cognition. Definitive answers to these sorts o f questions are unlikely to be found; indeed, many people would say the only defen­ sible positions are heuristic. But what is clear is that religion, especially that o f “ others,” has proved, in Levi-Strauss’s fam o us phrase, “ g o o d to think.” At the sam e time, the anthropology o f religion has m ade great strides in its enqui­ ries. We are not, on the whole, nearly so ethnocentric as we once were. We have learned to appreciate the intellectual complexity, aesthetic richness, and moral depth o f other peoples’ constructions and practices; learned to understand how certain ideas and practices com e to seem natural and inevitable to those w ho follow them; learned to see the intricate connections between ideology and power, from the grand myths o f state to the minute capillary action o f habitual disciplinary practices; learned to recognize generally recurring means and idioms o f representing (through myth) or enabling (through ritual) creation, transition, transform ation, order, and disorder; learned to appreciate the open-ended, edifying, world-building, lifeenriching, and subversive qualities o f much sym bolic practice. We have a go od general idea o f better and worse ways to go a b o u t trying to understand, com pare, and explain symbolic and social phenomena; and we have learned that the social and symbolic are inextricably connected: all that is social signifies, a n d signifying practices are simultaneously socializing.2 Perhaps, as Levi-Strauss found for myth, it is from music that we can best draw the metaphors for w hat we produce - o des, variations, and so on. A m ong our

14

GENERAL INTRODUCTION

greatest achievements, then, are the ethnographies like Nuer Religion (EvansPritchard 1956), Divinity and Experience (Lienhardt 1961), and Bwiti (Fernandez 1983) - to take only three superb Anglophone accounts o f African religions - that are the closest we com e to symphonies. It is under the rubric o f the “ A nthropology o f R eligion” that fundamental ques­ tions o f human difference have been addressed. These are often posed in dualist form - West versus rest, history versus myth, science versus religion, modern versus traditional, civilized versus primitive, and so on. It is extraordinarily difficult to escape from the pervasive strength o f such dualisms. Some people would argue (dualistically) that the difference between the West and the rest is precisely the for­ mer’s commitment to dualism and the latter’s freedom from it; others have seen dualism as central to human thought. But whatever the case, the history o f anthro­ pological conversation on these topics is largely the increasing sophistication with which these dualism s have been explored, mediated, and partially transcended. On the one hand, we do not want to exoticize and to exaggerate difference; on the other, we d o not want to homogenize, an enterprise in which everyone inevitably com es across as pale versions o f ourselves. The anthropological position is one that precisely straddles this dilemma, not ignoring it, not wishing it aw ay , not reducing the answer to one side or the other.

From T rad itio n to A n th o lo g y Reducing a long, healthy, and com plex intellectual tradition to some 6 0 0 pages entails som e hard choices and a go od deal o f violence. It was all too easy to find g o o d articles to include in this reader. The anthropology o f religion is blessed with riches and the field is a very broad one; structural analyses are appropriate to every d om ain, sym bols and rituals are ubiquitous. If the selection is to be done well there needs to be an overall vision, yet one that is not overly narrow. I shall try briefly to indicate the grounds on which I have made my selections. The primary aim has been to include a substantial body o f significant work o f relatively lasting significance. The temporal scope ranges from the 19th century to the contem porary period. Yet the anthology does not com prise a systematic history o f a nthropological approaches to religion so much as a number o f significant inter­ ventions acro ss that history. At som e moments it captures the density o f intertextual references, the cooperative building and also the agon characteristic o f an intellec­ tual tradition, while at other moments the selections are simply too widespread for direct mutual citation. I try to evoke some o f the density via the order o f presenta­ tion and the accom panying commentary. Thus Part I is constituted by a section setting out several contrasting orientations to religion followed by a section co m pris­ ing skeptical rejoinders. With respect to intellectual breadth, 1 have endeavored to avoid setting obvious boundaries. I take the anthropological tradition to be constituted in dialogue not only with the religious ideas and practices it has encountered, but with the broader western tradition out o f which it has grown and to which (along with postcolonial scholarship) it must continue to speak. Hence I have not hesitated to include pieces

GENERAL INTRODUCTION

15

by authors who are not anthropologists by profession. Likewise I include both work whose theme is “ religion” in a straightforward and obvious sense and w ork that expands the horizons of what we might mean by “ religion,” approaching it sideways, as it were, and thereby linking it to broader questions o f culture and politics. A guiding principle has been to avoid the reification o f both “ religion” and of the various categories and topics - “ prayer,” “ spirits,” “ H in duism ,” etc. - often subsum ed within it. Indeed, a m ajor contribution by anthropologists has been to challenge these kinds o f categories, and I include several essays that take an explic­ itly skeptical and reflexive approach (e.g., Humphrey, chapter 4 0 , on '‘sh am an ism ’ ). I have selected pieces in the first instance on the basis o f their contributions to general intellectual debate about religion in the broad sense, and only secondarily with an eye to systematic coverage o f the range o f topics generally included within religion as a field. However, I have attempted to include a broad range o f approaches as well as a sam ple o f religious practices from diverse social contexts and cultural traditions. Similarly, I have looked for a way to organize the sections and chapters that would not simply reproduce older forms o f classification (myth, ritual, etc.). I have not, of course, either wished or been able to dispense with classification entirely. Parts II and III are distinguished very loosely according to the Aristotelian categories o f making (poiesis) and doing (praxis), which I have found useful both to overcome more stubborn dualities o f reason and unreason (Lam bek 2 0 0 0 a ) and to articulate observations in my own ethnographic w ork (Lam bek 1998, 2 0 0 2 a , b). While each inevitably implies and overlaps with the other, together they form a productive way to conceptualize religion (the production o f w orlds and the activities within them) and to comprehend the range o f emphases that a variety o f anthropologists have brought to the subject. I conclude with Part IV under the rubric o f history, which is simply a matter o f different emphasis rather than o f topic. These essays emphasize the historicity and dynam ism o f religion and the political and economic impinge­ ments on, and entailments of, both poiesis and praxis at specific historical locations, including, o f course, the present.

H om age to th e M issing If the most obvious function o f an anthology is to introduce new readers to a field o f study, it introduces them to old (relatively speaking) authors. This implies that a second function is commemorative. An anthology produces a record o f what has transpired. This record is necessarily selective. It is one o f the tasks o f memory work to forget in a principled way. I have tried to be principled but it has been a cause o f grief that I have had to “ forget” - to put aside, to cut, to disregard - so much. When I began I had the idea that if I consulted widely I could discover som e sort o f consensus as to what should be included, but o f course, the more people I asked, the more suggestions I received. All were interesting and som e I have adopted, but overall consultation m ade the process o f selection harder. I became resigned to the fact that the choice would have to be mine and that any selection would be to some

16

GENERAL INTRODUCTION

degree idiosyncratic. The result is a com bination o f “ classic” essays that have already been inscribed in institutional memory, pieces that I w as taught or that I have used in teaching, w orks that have stayed in mind because o f their place in the development o f my thinking and ethnographic work, and pieces that 1 have discov­ ered in the process o f asking advice and sam pling the literature. The result is the product less o f a systematic search or clearly elaborated plan than a som ew hat tor­ tuous hermeneutic circling around and through its object. In com piling the anthology I have felt that there were many obligations to be met and that in trying to meet them all 1 have not done proper service to any. T o begin with, there is the debt to the ancestors. M uch excellent work written on religion over the course o f the 20th century and before is now too rarely cited or read. It was exciting to rediscover, for exam ple, Lowie and Sapir, but they did not make the cut. Second, is the parental obligation. M y selection is weighted tow ard things that were new when I w as in graduate school and I can only say that the early 1970s were genuinely a high point in the anthropology o f religion in N orth America, vibrant with the work o f Turner, Geertz, and their students and the local reception o f structuralism and hermeneutics. Then there are my peers, to whom I apologize for slighting, and my chronological juniors, where my authority is perhaps weakest. Some o f the other things I thought to balance in the process: theory relative to ethnography; small-scale societies relative to “ world religions;” “ W est” to “ rest;” world-building relative to “ hermeneutics o f suspicion;” thought or reason relative to emotion or perform ance; com plex, multistranded, lengthy interpretation that best exemplifies ethnographic method relative to short, punchy, polemical articulation; historical, topical, and regional representation; m ainstream or canonical relarive to shaking up the consensus with ignored, idiosyncratic, or original pieces. The cuts have been ruthless and the om issions are glaring. N o specialist in the Held (not even the contributors) is likely to be completely satisfied with the selection and it will replicate no instructor’s ideal course. But the anthology can serve as a core text to be supplemented as desired. I have also provided suggestions for further reading, but am all too aw are that even these merely skim the surface. In order to m aximize inclusion I stuck to one principle and gave way on another. Each author receives only one entry. But although I attem pted to select short pieces, I also, after much hesitation, engaged in abridging. The abridgments have been checked with the authors where possible and are clearly marked by ( . . . ] . The individual acknow ledgm ents and som e footnotes have been deleted without nota­ tion. The individual selections are not meant to be “ representative” o f their au thors’ work nor necessarily either their “ best piece” or “ last w o rd .” Rather, each entry w as selected to fit as part o f the emerging whole and thus at times to serve several ends at once. The entries must thus be seen as invitations: invitations to seek the unabridged originals and to read more o f each au th o r’s work, invitations to read more on the topics or themes each entry represents, and invitations, via the bibliographic index, to read on topics, regions, and traditions omitted from the selections. Finally, the quintessence o f anthropological insight is necessarily absent from this anthology; it can only be grasped by reading full m onographs, and ideally more than one, by a given author or on a given society.

GENERAL INTRODUCTION

17

Recent introductory texts on the anthropology o f religion include Bowen (1998b), Bowie (1999), Klass (1 9 9 5 ), and M orris (1987). Wallace (1966) w as particularly original for its rime. Lessa and Vogt, eds. (in various editions, most recently 1979) is the classic anthology and has served as an inspiration for the present work; Dolgin, Kemnitzer, and Schneider, eds. (1977) anthologizes symbolic anthropology. Bowen ed. (1 998a) and Klass and W eisgrau, eds. (1999) provide collections o f recent articles; Glazier, ed. (1999) and Scupin, ed. (2000) offer han dbo o ks. A good advanced introduction to the rationality debate is T am b iah (1990), while Firth (1996) is a lively and accessible collection o f essays on religion by a senior scholar.

NOTES 1 2

See the im portant d iscu ssion in Bernstein 1988. I regret having forgotten the source o f this pithy rem ark.

Part I

The Context of Understanding and Debate

Opening Frameworks

In trod u ction We begin with four classic and profound statements on the nature o f religion which take us from one o f the ablest evolutionists (Tylor), through the two greatest socio ­ logists on the subject (Durkheim and Weber), to the anthropologist w ho has been, if not the most influential, at least the m ost talked about theoretician in the latter decades o f the 20th century (Geertz). The first three figures present what cam e (along with a Freudian model) to be seen as the main alternative approaches to the under­ standing o f religion. T hrough the mid-20th century the choice w as particularly between Durkheim and Tylor. Then, in part through the writing o f Geertz, Weber became o f great interest to anthropology. Tylor and Durkheim both seek the origins or foundations o f religion. For Tylor they lie in individual reasoning, hence his position is often referred to as rationalist o r intellectualist, whereas for Durkheim they lie in the collective consciousness that society has o f itself. D urkheim ’s position is therefore social. Both men seek the clearest exemplification o f religion’s foundations in what they consider to be the simplest known societies. T y lo r’s method is one o f com parison, derived from osten­ sible facts gleaned from a voluminous number o f travelers’ reports o f varying quality, while Durkheim begins with the analysis o f what he takes to be a single case, namely Aboriginal Australia. For T ylor the presence o f religion could be explained as a reasonable, albeit mistaken, attem pt to solve intellectual problems, whereas Durkheim rook a more symbolic approach and argued that religious ideas and rituals both express and regenerate society. W eber’s concerns were not with the origins o f religion per se so much as with the role o f religion in the origins o f modernity. H e w as concerned not with smallscale societies but with states, especially when com parin g Asia with Europe. His approach w as also distinctly historical, in contrast to the synchronic structurefunctionalism to which Durkheimian ideas gave rise. H e w as interested, like Tylor, in what motivated people to think and act as they did, but he understood action in

22

OPENING FRAMEWORKS

rhe context o f collective systems o f meaning. Where T ylor saw religious ideation as the product o f direct and universal human concerns, Weber emphasized how such concerns are themselves shaped by diverse religious traditions. Likewise, where Tylor operated with an ostensibly universal and unequivocally positive idea o f rationality, Weber w as concerned with alternate kinds o f m eans-ends relations, and especially with the emergence o f various forms o f what he called rationalization, which he viewed with som e unease. In Weber the narrative o f unequivocal progress assum ed by the Victorians is replaced by distinct ambivalence. This is not rhe place to rehearse the com plex elements o f either D urkheim ’s or Weber’s analyses o f religion, nor to chart a history o f the respective traditions they founded or the way they twine through anthropological w ork (see O ’T oole 1984 for a g o o d sociological introduction). But it is to note that most contemporary anthropological analyses are informed by both these thinkers, whether in affirm a­ tion or resistance, tacitly or explicitly, and whether leavened by doses o f M a rx and Freud. M a r x and Freud do belong here as well, and would have been represented by excerpts from Capital on com m odity fetishism and from Totem and Taboo. Certain Freudian and M arxist arguments are presented in the selections from Obeyesekere and T aussig, respectively. All these interpretations o f religion are pre­ figured by Feuerbach’s analysis o f religion as a human product (1975). Geertz is present in this section not as a founding ancestor but as one o f the strongest attem pts within modern anthropology to co m po se a definition o f religion and hence a model for subsequent research. M oreover, Geertz attempts a kind o f implicit synthesis o f previous approaches. We can see threads o f both Tylor and Durkheim, while the overriding interest is Weberian. Geertz’s roots are also Boasian, via Ruth Benedict, in that he is interested in portraying distinctive cultural worlds by means o f ethnographic particulars. Few other anthropologists have been able to develop original and powerful defini­ tions of religion or to model its relationships to culture and society. A theoretical synthesis by sociologist Peter Berger, The Sacred Canopy (1 9 6 7 ; see also Berger and Luckmann 1966), has strong parallels with Geertz. R ap p a p o rt (1999) is a magiste­ rial anthropological attempt to understand religion’s place in nature and, like Durkheim, its place at the foundations o f social life.

1

Religion in Primitive Culture Edward Burnett Tylor

"The grow th of religious ideas is envi­ roned w ith such intrinsic difficulties that it may never receive a perfectly satis­ factory exposition. Religion deals so largely w ith the im aginative and em o­ tional nature, and consequently with such uncertain elem ents of knowledge, that all primitive religions are grotesque and to some extent unintelligible." Such was the opinion of Lewis Henry Morgan (1877: 5, as cited by G uenther 1999: 58), a contem porary of Tylor's. W hile Tylor (1832-1917) and M organ are considered the leading figures of 19th-century evo­ lutionary anthropology, and certainly among the very fe w w ho are still read today, Tylor's view of religion could not have departed further from Morgan's statement. Tylor, holder of the first chair in anthropology at the University of Oxford, found prim itive religions nei­ ther grotesque nor unintelligible, and he thought he had developed a model of their developm ent. Nor did he em pha­

size their "im aginative and em otional" side. Tylor was a rationalist and departed from many of his contem poraries in finding this quality in so-called "prim i­ tive man" as w ell. However, he felt equally th at despite the rationality of their thought, the conclusions reached by members of small-scale societies w ere in error. If Tylor is the only 19th-century w riter to be included here, it is not only because he was among the more sensible, but because the core of his definition of reli­ gion as "the belief in Spiritual Beings" remains congenial to many contem po­ rary thinkers and is indeed almost a part of w estern "common sense" on the subject. His characterization of animism remains fruitful and does serve as one means to generalize about religious phenom ena of all kinds. Animism speaks today to reflections on the mind/body problem and conceptualizations of the person, to the relations betw een humans

From Sir Edward Burnett Tylor, Primitive Culture (New York: Harper & Row, 1958 [1871]), pp. 8-19, 8 0 -6 , 4 4 4 -7 . Copyright €> by Harper He Row, Publishers, Inc. Abridged, footnotes removed, paragraph breaks added.

24

EDW ARD BURNETT TYLOR

and other species, especially in hunting societies, and to conceptualizations of death and the centrality of both m ortu­ ary ritual and sacrifice in human societ­ ies. (See the essay by Viveiros de Castro, chapter 23 below.) Tylor's main successor was Sir James Frazer, w ho developed a stronger con­ trast betw een magic, religion, and science as forms of reasoning (1890). It was Lucien Levy-Bruhl (1985 [1910], 1966 [1923]) w ho provided the most direct alternative to rationalism . He is associ­ ated w ith the view th at "natives" lived in some mystical connection w ith the w orld that transcends or evades the rational. It is easy to criticize Levy-Bruhl (and he eventually criticized himself 1949), but his argum ent looks rather more sophisticated if we see it as the com plete inverse of Tylor, nam ely th at if primitives are not rational, neither can they be said to be in error w ith respect to their conclusions. For Levy-Bruhl, at least, the relation to the w orld w as a rich and pow erful one. The criticisms of Tylor have been well rehearsed. The evolutionism entails value judgm ents and assumes a static quality to the thought of small-scale societies, as if they had been frozen in tim e; the com parative m ethod decontextualizes ethnographic facts w hose m eaning is thereby reduced and dis­ torted; the rationalism ignores the em otional side of religion; and the intellectualism ignores the collective, sym­ bolic, representational dimension. The fact that Tylor sees religion grounded in error has appeared to many subsequent thinkers to debase and reduce the subject. M oreover, despite his concern for accurate empirical evidence, Tylor's approach is characterized by his own excessive speculation. Tylor's errors w ere com pounded in Frazer. Both Tylor and Levy-Bruhl present one-sided and therefore impoverished

pictures. Members of small-scale societ­ ies are far more sophisticated thinkers than either gave them credit for. They have far greater know ledge of the human condition and far richer, more variegated, and complex religious lives than Tylor recognized. They are also capable of much greater critical distance and abstract reasoning than L£vy-Bruhl im agined. M oreover, it is a mistake to generalize about small-scale societies; there are great differences among them and over tim e. A com prehensive critique of Tylor, Levy-Bruhl, and other evolu­ tionists is to be found in Evans-Pritchard (1965). Nevertheless, there are kernels of thought here th at subsequent anthro­ pologists have found of interest. There has been a stream of self-conscious "Neo-Tyloreanism" (notably Horton 1993) in response to an overly structurefunctionalist approach to religion. This school draws particular inspiration from the fact that w hatever his ethnocentrism, Tylor begins w ith the rational and questioning nature of all people to try to understand the human situation and our place in the world, and th at reli­ gious ideas are adequate to the worlds they describe and shift as the horizons of those worlds shift, as Horton put it, from microcosm to macrocosm. In reading Tylor today it is striking to see behind the evolutionary language a basic concern w ith universals in human thought and experience and the conti­ nuity of religious tho ug ht between small-scale societies and his own; in this respect Tylor forms a precursor to con­ tem porary cognitivist approaches. M ore­ over, although some of the language used appears highly problematic, Tylor's basic argum ent was against those who saw in smaller-scale societies either a degeneration or a borrowing from large-scale ones. W hat Tylor staunchly defends is th e rationalism and creativity

RELIGION IN PRIMITIVE CULTURE

of all humans. Conversely, authors such as Devisch (1993) and Jackson (1989) (see also Leenhardt 1979 [1947]) sugges­ tively reim agine aspects of mystical participation through ideas o f em bod­ ied know ledge and experience. However,

[ ...| The first requisite in a system atic study o f the religions o f the low er races, is to lay dow n a rudim entary definition o f religion. By requir­ ing in this definition the belief in a suprem e deity or o f judgm ent after d eath , the ad oration o f idols or the practice o f sacrifice, or other partially-diffused doctrines or rites, no doubt m any tribes m ay be excluded from the cate­ gory o f religious. But such n arrow definition has the fault o f identifying religion rather with particular developm ents than with the deeper m otive which underlies them . It seem s best to fall back at once on this essential source, and sim ply to claim , as a m inim um definition o f Religion, the belief in Spiritual Beings. If this stan dard be applied to the descriptions o f low races as to religion, the follow ing results will ap p ear. It can n ot be positively asserted that every existin g tribe recognizes the belief in spiritual beings, for the native condition o f a con siderable num ber is obscure in this respect, and from the rapid change or extinction they are undergoing, m ay ever rem ain so. It w ould be yet m ore unw arranted to set dow n every tribe m entioned in history, or known to us by the discovery o f an tiquarian relics, as necessar­ ily having passed the defined m inim um o f reli­ gion. Cireater still w ould be the unw isdom o f declaring such a rudim entary belief natural or instinctive in all hum an tribes o f all tim es; for no evidence justifies the opinion that m an, know n to be cap able o f so vast an intellectual developm ent, cannot have em erged from a n onreligious condition, previous to that reli­ gio u s condition in which he h appens at present to com e with sufficient clearness within our range o f know ledge. It is desirable, how ever, to take ou r basis o f enquiry in observation rather than from speculation. H ere, so far a s I can judge from the im m ense m ass o f accessible

25

neither mental nor bodily relation­ ships are unm ediated; on both counts Durkheim ian or culturalist understand­ ings o f the w ay society and language shape collective representations and individual experiences are critical.

evidence, we have to adm it that the belief in spiritual beings app ears am on g all low races with w hom we have attained to thoroughly intim ate acquain tan ce; w hereas the assertion o f absence o f such belief m ust apply either to ancient tribes, or to m ore or less im perfectly described m odern ones. The exact bearing o f this state o f things on the problem o f the origin o f religion m ay be thus briefly stated. W ere it distinctly proved that non-religious savages exist o r have existed, these m ight be at least plau sibly claim ed as representatives o f the con­ dition o f M an before he arrived at the religious state o f culture. It is not desirable, how ever, that this argum ent should be put forw ard, for the asserted existence o f the non-religious tribes in question r e s t s . . . on evidence often m istaken and never conclusive. The argum ent for the natural evolution o f religious ideas am o n g m ankind is not invalidated by the rejec­ tion o f an ally to o w eak at present to give effectual help. N on -religious tribes m ay not exist in our day, but the fact bears no m ore decisively on the developm ent o f religion, than the im possibility o f finding a m odern English village w ithout scisso rs or b o o k s or luciferm atches bears on the fact that there w as a time when n o such things existed in the land. I prop o se here, under the nam e o f Anim ism , to investigate the deep-lying doctrine o f Spiri­ tual Beings, which em bodies the very essence o f Spiritualistic as oppo sed to M aterialistic ph ilosoph y. Anim ism is not a new technical term , though now seldom used. From its special relation to the doctrine o f the soul, it will be seen to have a peculiar appropriaten ess to the view here taken o f the m ode in which theological ideas have been developed am on g m ankind. The w ord Spiritualism , though it m ay be, and som etim es is, used in a general sense, has this ob vious defect to us, that it has

26

EDW ARD BURNETT TYLOR

becom e the designation o f a p articular m odern sect, w ho indeed hold extrem e spiritualistic view s, but can n ot be taken as typical represen­ tatives o f these view s in the w orld at large. The sense o f Spiritualism in its w ider acceptation, the genera! belief in spiritual beings, is here given to A nim ism . Anim ism characterizes tribes very low in the scale o f hum anity, and thence ascen ds, deeply m odified in its tran sm ission, but from first to last preserving an unbroken continuity, into the m idst o f high m odern culture. Doctrines adverse to it, so largely held by individuals or schools, are usually due not to early low ness o f civilization, but to later changes in the intel­ lectual co urse, to divergence from , or rejection of, ancestral faiths; and such newer develop­ m ents do not affect the present enquiry a s to the fundam ental religious condition o f m an ­ kind. A nim ism is, in fact, the groun dw ork o f the Philosophy o f Religion, from that o f savages up to that o f civilized men. And although it m ay at first sight seem to afford but a bare and m eagre definition o f a m inim um o f religion, it will be found practically suffi­ cient; for where rhe root is, the branches will generally be produced. It is habitually found that the theory o f Anim ism divides into tw o great d o g m a s, form ing parts o f one consistent doctrine; first, concerning souls o f individual creatures, cap able o f continued existence after the death or destruction o f the body; second, concerning other spirits, upw ard to the rank o f pow erful deities. Spiritual beings are held to affect or control the events o f the m aterial w orld, and m an ’s life here and hereafter; and it being considered that they hold intercourse with m en, an d receive pleasure or displeasure from hum an action s, the belief in their exis­ tence leads naturally, and it might alm ost be said inevitably, sooner or later to active rever­ ence and p ropitiation . T hus Anim ism in its full developm ent, includes the belief in souls and in a future state, in controlling deities and subordin ate spirits, these doctrines practically resulting in som e kind o f active w orship. O ne great element o f religion, that m oral element which am on g the higher nations form s its m ost vital part, is indeed little represented in the religion o f the low er races. It is not that these races have no m oral sense or no m oral

stan d ard , for both are strongly m arked am ong them , if not in form al precept, at least in that traditional con sensus o f society which we call public opin ion , according to which certain action s are held to be go o d or b ad , right or w rong. It is that the conjunction o f ethics and Anim istic philosophy, so intim ate and p ow er­ ful in the higher culture, seem s scarcely yet to have begun in the low er. I propose here hardly to touch upon the purely m oral aspects o f religion, but rather to study the anim ism o f the w orld so far as it constitutes, as unquestion­ ably it does constitute, an ancient and w orld ­ wide philosophy, o f which belief is the theory and w orship is the practice. Endeavouring to shape the m aterials for an enquiry hitherto strangely undervalued and neglected, it will now be my task to bring as clearly as m ay be into view the fundam ental anim ism o f the low er races, and in som e slight and broken outline to trace its course into higher regions o f civilization. Here let me state once for all tw o principal conditions under which the present research is carried on. First, as to the religious doctrines and practices exam in ed, these are treated as belonging to theological system s devised by hum an reason, w ithout supernatural aid or revelation; in other w ords, as being developm ents o f N atu ral Religion. Second, as to the connexion between sim ilar ideas and rites in the religions o f the sav age and the civilized w orld. While dw elling at som e length on doctrines and cerem onies o f the low er races, and som etim es particularizing for special reason s the related doctrines and cerem onies o f the higher nations, it has not seem ed my proper task to w ork out in detail the problem s thus suggested am on g the ph i­ losoph ies and creeds o f Christendom . Such app licatio ns, extending farthest from the direct scope o f a w ork on prim itive culture, are briefly stared in general term s, or touched in slight allusion , or taken for granted w ithout rem ark. Educated readers possess the in fo rm a­ tion required to w ork ou t their general bearing on theology, while m ore technical discussion is left to ph ilosoph ers and theologians spe­ cially occupied with such argum ents. The first branch o f the subject to be co n sid ­ ered is the doctrine o f hum an and other Souls, an exam in ation o f which will occupy the rest

RELIGION IN PRIMITIVE CULTURE

o f rhe present chapter. W hat the doctrine o f the soul is am on g the low er races, m ay be explained in statin g rhe anim istic theory o f its developm ent. It seem s as though thinking m en, as yet at a low level o f culture, were deeply im pressed by tw o group s o f biological p ro b ­ lems. In the first place, w hat is it that m akes the difference between a living body an d a dead one; w hat causes w aking, sleep, trance, d isease, death? In the second place, w hat are those hum an shapes which ap p ear in dream s and visions? Lookin g at these tw o grou p s o f phenom ena, the ancient savage philosophers prob ab ly m ade their first step by the ob vious inference that every m an has tw o things belonging to him, nam ely, a life and a phantom . These tw o are evidently in close connexion with the body, the life as enabling it to feel and think an d act, the phantom as being its im age or second self; both, also, are perceived to be things separable from the body, the life as able to g o aw ay and leave it insensible or dead , the phantom as ap p earin g to people at a distance from it. The second step w ould seem also easy for sav ages to m ake, seeing how extrem ely diffi­ cult civilized men have found it to unm ake. It is merely to com bine the life and the phantom . A s both belong to the body, why should they not a lso belong to one another, and be m an i­ festations o f one and the sam e soul? Let them then be considered as united, and the result is that well-known conception which m ay be described as an ap p arition al-soul, a ghost-soul. T his, at any rate, corresp on d s with the actual conception o f the personal soul or spirit am on g the low er races, which m ay be defined as fol­ low s: It is a thin unsubstantial hum an im age, in its nature a sort o f vap ou r, film, or sh ad ow ; the cau se o f life and thought in the individual it an im ates; independently p ossessin g the per­ sonal con sciousn ess and volition o f its c o rp o ­ real ow ner, past or present; cap able o f leaving the body far behind, to flash swiftly from place to place; m ostly im palpable and invisible, yet a lso m anifesting physical pow er, and esp e­ cially ap pearin g to men w aking or asleep as ph an tasm separate from the body o f which it bears the likeness; continuing to exist and ap p ear to men after the death o f that body; able to enter into, p ossess, an d act in the bodies

27

o f other m en, o f an im als, an d even o f things. T hough this definition is by no m eans o f universal application, it has sufficient general­ ity to be taken as a stan d ard , m odified by m ore or less divergence am o n g any particular people. Far from these w orld-w ide opinions being arbitrary or conventional p rodu cts, it is seldom even justifiable to consider their uniform ity a m o n g distant races a s proving com m unica­ tion o f any sort. They are doctrines answ ering in the m ost forcible way to the plain evidence o f m en’s senses, as interpreted by a fairly consistent and rational prim itive philosophy. So well, indeed, does prim itive anim ism accoun t for the facts o f nature, that it h as held its place into the higher levels o f education. T hough classic and mediaeval philosophy m odified it m uch, and m odern philosophy has handled it yet m ore unsparingly, it has so far retained the traces o f its original character, that heirloom s o f prim itive ages m ay be claim ed in the existing psychology o f the civilized w orld. O ut o f the vast m ass o f evidence, col­ lected am on g the m ost various and distant races o f m ankind, typical details m ay now be selected to display the earlier theory o f the soul, the relation o f the parts o f this theory, and the m anner in which these parts have been aban don ed, m odified, or kept up, along the course o f culture. T o understand the pop u lar conceptions o f the hum an soul or spirit, it is instructive to notice the w ords which have been found suit­ able to express it. The gh o st or ph antasm seen by the dream er or the visionary is an unsub­ stantial form , like a sh adow or reflexion, and thus the fam iliar term o f the shade com es in to express the soul. T hus the T asm an ian w ord for the shadow is also that for the spirit; the A lgonquins describe a m an’s soul as otahchuk , “ his sh a d o w ;” the Q uiche lan guage uses natub for “ shadow , so u l;” the A raw ak ueja means “ sh adow , soul, im age;” the A bipones m ade the one w ord loakal serve for “ sh adow , soul, echo, im age.” The Z u lu s nor only use the w ord tunzi for “ shadow , spirit, g h o st,” but they consider that at death the sh adow o f a man will in som e way depart from the corpse, to becom e an ancestral spirit. The B asutos not only call the spirit rem aining after death the

28

EDW ARD BURNETT TYLOR

seriti o r “ sh a d o w ,” but they think that if a man w alks on the river ban k, a crocodile m ay seize his sh ad ow in the w ater and draw him in; while in O ld C a lab a r there is found the sam e identification o f the spirit with the ukpon o r “ sh a d o w ,” for a m an to lose which is fatal. There are thus found am on g the low er races not only the types o f those fam iliar classic term s, the skia and um bra , but also w hat seem s the fundam ental thought o f the stories o f sh ad ow less men still current in the folklore o f E urope, and fam iliar to m odern readers in C h a m isso ’s tale o f Peter Schlemihl. T h u s the dead in Purgatory knew that D ante w as alive when they saw th at, unlike theirs, his figure cast a sh ad ow on the ground. O ther attribu tes are taken into the notion of soul o r spirit, with especial regard to its being the cause o f life. T h u s the C arib s, connecting rhe pulses with spiritual beings, and especially considering that in the heart dw ells m an’ s chief soul, destined to a future heavenly life, could reason ably use the one w ord touanni for “ sou l, life, h eart.” T he T o n g an s supposed the soul to exist th roughout the w hole extension o f the body, but particularly in the heart. O n one occasion , the natives were declaring to a E u ro­ pean th at a m an buried m onths a g o w as never­ theless still alive. “ And one, endeavourin g to m ake me understand w hat he m eant, took hold o f my hand, an d squeezing it, said , ‘T his will die, but the life that is within you will never die;’ with his other hand pointing to my h eart.” So the B asutos say o f a dead m an that his heart is gone out, and o f one recovering from sick ­ ness that his heart is com ing back. T h is co rre­ sp on d s to the fam iliar O ld W orld view o f the heart a s the prim e m over in life, thought, and passio n . The con nexion o f soul and blood , fam iliar to the Karens and P apuas, a p p ears prom inently in Jew ish and A rabic philosophy. T o educated m oderns the idea o f the M acu si Indians o f Ciuiana m ay seem quain t, that although the body will decay, “ the m an in our eyes” will not die, but w ander abo u t. Yet the associatio n o f personal anim ation with the pupil o f the eye is fam iliar to European fo lk ­ lore, which not unreasonably discerned a sign o f bew itchm ent or ap p roach in g death in the disap p earan ce o f the im age, pupil, or baby, from the dim eyeballs o f the sick m an.

The act o f breathing, so characteristic o f the higher an im als during life, an d coinciding so closely with life in its departure, has been repeatedly and naturally identified with the life or soul itself. L.aura Bridgm an show ed in her instructive w ay the analogy between the effects o f restricted sense and restricted civilization, when one day she m ade the gesture o f taking som ething aw ay from her m outh: “ I d ream ed ,” she explained in w ords, “ that G od took aw ay my breath to heaven.” It is thus that West A ustralian s used one w ord w aug for “ breath, spirit, so u l;” that in the N etela language o f C alifo rn ia, pints m eans “ life, breath, so u l;” that certain G reenlanders reckoned tw o souls to m an, nam ely his sh adow and his breath; that the M alay s say the soul o f the dying man escapes through his nostrils, and in Ja v a use the sam e w ord hawa for “ breath, life, so u l.” H ow the notions o f life, heart, breath, and phantom unite in the one conception o f a soul or spirit, and at the sam e tim e how loose and vague such ideas are am o n g barbaric races, is well brought into view in the answ ers to a religious inquest held in 1528 am on g the natives o f N icaragu a. “ When they die, there com es ou t o f their m outh som ething that resem bles a person, and is called julio [Aztec yuli - to live]. T h is being goes to the place where the m an and w om an are. It is like a person, but does not die, and the body rem ains here.” Question. “ D o those w ho g o up on high keep the sam e body, the sam e face, and the sam e lim bs, a s here b e lo w ?” Answer. “ N o ; there is only the h eart.” Question. “ But since they tear out their hearts [i.e. when a captive w as sacrificed], w hat h appens th en ?” Answer. “ It is not precisely the heart, but that in them which m akes them live, and that quits the body when they d ie .” O r, as stated in another interrogatory, “ It is not their heart that goes up above, but w hat m akes them live, that is to say, the breath that issues from their m outh and is called ju lio.” The conception o f the soul as breath m ay be follow ed up through Sem itic and Aryan ety­ m ology, and thus into the m ain stream s o f the philosophy o f the w orld. H ebrew show s nepbeshy “ breath ,” passin g into all the m ean­ ings o f “ life, sou l, m ind, an im al,” while ruacb and nesbamah m ake the like transition from

RELIGION IN PRIMITIVE CULTURE

“ breath ” to “ sp irit” ; and to these the A rabic

nefs and rub corresp on d . The sam e is the history o f Sanskrit dtman an d prana , o f Greek psyche and pneum a , o f Latin anim us , anim a , spiritus . S o Slavonic duck h as developed the m eaning o f “ breath ” into that o f soul or spirit; an d the dialects o f the G ypsies have this w ord ditk with the m eanings o f “ breath, spirit, g h o st,” whether these pariah s brought the w ord from India a s part o f their inheritance o f A ryan speech, or whether they adop ted it in their m igration a cro ss Slavonic lands. G erm an geist and English ghost, to o , m ay possibly have the sam e original sense o f breath. And if any should think such exp ressio n s due to mere m etaph or, they m ay judge the strength o f the im plied connexion between breath and spirit by cases o f m ost unequivocal significance. A m ong the Sem inoles o f Florida, when a w om an died in childbirth, the infant w as held over her face to receive her parting spirit, and thus acquire strength and know ledge for its future use. T hese Indians could have well understood why at the death-bed o f an ancient R om an , the nearest kinsm an leant over to inhale the last breath o f the departin g (et excipies hanc an im am ore pio). Their state of m ind is kept up to this day am on g T yrolese peasan ts, w ho can still fancy a go o d m an’s soul to issue from his m outh at death like a little white cloud. It will be show n th at men, in their com posite and confused notions o f the soul, have brought into connexion a list o f m an ifestation s o f life and thought even m ore m ultifarious than this. But also, seeking to avoid such perplexity o f com b in ation , they have som etim es endeavored to define and classify m ore closely, especially by the theory that man has a com bination of several kinds o f spirit, soul, or im age, to which different functions belong. Already in the b arb aric w orld such classification has been invented or ad op ted . T h u s the Fijians distin­ guished between m an’s “ dark sp irit” or sh ad ow , which goes to H ad es, and his “ light sp irit” or reflexion in w ater or a m irror, which stays near where he dies. The M alag a sy say that the saina or m ind vanishes at death, the aina or life becom es mere air, but the m atoatoa or gh ost hovers round the tom b. In N orth A m erica, the duality o f the soul is a strongly

29

m arked A lgonquin belief; on e soul goes out and sees dream s while the other rem ains behind; at death one o f the tw o abides with the body, and for this the survivors leave offer­ ings o f fo od , while the other departs to the land o f the dead. A division into three souls is also know n, and the D a k o ta s say that m an has four souls, one rem aining with the co rpse, one stayin g in rhe village, one go in g in the air, and one to the land o f spirits. The K arens distinguish between the “ la ” or “ k elah ,” the personal life-phantom , and the “ th ah ,” the responsible m oral soul. M o re or less under H indu influence, the K h on ds have a fourfold division, as follow s: the first soul is that capable o f beatification or restoration to B oora the G o o d D eity; the second is attached to a Khond tribe on earth and is reborn generation after gen eration, so that at the birth o f each child the priest ask s w ho has returned; the third goes out to hold spiritual in tercourse, leaving the body in a languid state, an d it is this soul which can p a ss for a time into a tiger, and tran sm igrates for punishm ent after death; the fourth dies on the dissolu tion o f the body. Such classification s resem ble those o f higher n a tio n s ,. . . I-.-]

H aving thus surveyed at large the theory of spirits or souls o f ob jects, it rem ains to point out w hat, to general students, m ay seem the m ost im portant con sideration belonging to it, nam ely, its close relation to one o f the m ost influential doctrines o f civilized philosophy. The savage thinker, though occupyin g him self so m uch with the phenom ena o f life, sleep, disease, and death, seem s to have taken for gran ted, as a m atter o f co urse, the ordinary operatio n s o f his ow n m ind. It hardly occurred to him to think ab o u t the m achinery o f think­ ing. M etaph ysics is a study which first assum es clear shape at a com paratively high level of intellectual culture. T he m etaphysical ph iloso­ phy o f thought taught in ou r m odern E u ro­ pean lecture-room s is historically traced back to the speculative psychology o f classic Greece. N o w one doctrine which there com es into view is especially associated with the name o f D em ocritus, the ph ilosoph er o f A bdera, in the fifth century BC. W hen D em ocritus p ro­ pounded rhe great problem o f m etaphysics,

30

EDW ARD BURNETT TYLOR

“ H ow do we perceive external th in gs?” - thus m aking, as Lew es says, an era in the history o f philosophy - he put forth, in answ er to the question, a theory o f thought. He explained the fact o f perception by declaring that things are alw ays throw ing o ff im ages (ei'ScoXa) o f them selves, which im ages, assim ilatin g to them selves the surroundin g air, enter a recipi­ ent soul, and are thus perceived. N ow , su p p o sin g D em ocritus to have been really the o rigin ator o f this fam ed theory o f ideas, how far is he to be considered its inven­ tor? W riters on the history o f philosophy are accustom ed to treat the doctrine as actually m ade by the ph ilosoph ical school which taught it. Yet the evidence here brought forw ard show s it to be really the savage doctrine o f object-souls, turned to a new purpose as a m ethod o f explain in g the phenom ena of thought. N o r is the correspondence a mere coincidence, for at this point o f junction between classic religion and classic philosophy the traces o f historical continuity m ay be still discerned. T o say that D em ocritus w as an ancient Greek is to say that from his childhood he had looked on at the funeral cerem onies o f his country, beholding the funeral sacrifices o f garm ents and jew els and money and fo od and drink, rites which his m other and his nurse could tell him w ere perform ed in order that the ph antasm al im ages o f these objects m ight pass into the p ossession o f form s shadow y like them selves, the sou ls o f dead men. T hus D em o­ critus, seeking a solution o f his great problem o f the nature o f thought, found it by sim ply decanting into his m etaphysics a surviving doctrine o f prim itive savage anim ism . This thought o f the ph an tom s or souls o f things, if sim ply m odified to form a philosophical theory o f perception, w ould then and there becom e his doctrine o f Ideas. N o r does even this fully represent the closeness o f union which con ­ nects the sav age doctrine o f flirting objectsouls with the Epicurean philosophy. Lucretius actually m akes the theory o f film-like im ages o f things (sim ulacra, m em b ran e) accoun t both for the ap p aritio n s which com e to men in dream s, and the im ages which im press their minds in thinking. So unbroken is the con ­ tinuity o f ph ilosoph ic speculation from savage to cultured thought. Such are the debts

which civilized philosophy ow es to prim itive anim ism . The doctrine o f ideas, thus developed in the classic w orld, h as, indeed, by no m eans held its course thenceforth unchanged through m etaphysics, but has undergone transition som ew hat like that o f the doctrine o f the soul itself. Ideas, fined dow n to the abstract form s or species o f m aterial objects, and applied to other than visible qualities, have at last com e merely to denote subjects o f thought. Yet to this day the old theory has not utterly died out, and the retention o f the significant term “ id ea” (I8ea, visible form ) is acco m p a­ nied by a sim ilar retention o f original m eaning. It is still one o f the task s o f the m etaphysician to display and refute the old notion o f ideas as being real im ages, and to replace it by m ore ab stract conceptions. It is a striking instance that D ugald Stew art can cite from the w orks o f Sir Isaac N ew ton the follow ing distinct recognition o f “ sensible species:” “ Is not the sensorium o f an im als, the place where the sentient substance is present; and to which rhe sensible species o f things are brought, through the nerves and brain , that there they m ay be perceived by the m ind present in that p lace ?” A gain, Dr. Reid states the original theory o f ideas, while declaring that he co n ­ ceives it to have no solid foundation, though it has been adopted very generally by philosophers. . . . This notion o f our perceiving external objects, not immediately, but in certain images or species o f them conveyed by the senses, seems to be the most ancient philosophical hypothesis we have on the subject o f percep­ tion, and to have, with small variations, retained its authority to this day. G ranted that D r. Reid exaggerated the extent to which m etaphysicians have kept up the notion o f ideas a s real im ages o f things, few will deny that it does linger much in m odern m inds, and that people w ho talk o f ideas do often, in som e hazy m etaph orical w ay, think o f sensible im ages. O ne o f the shrew dest things ever said ab o u t either ideas or gh osts w as Bishop Berkeley’s retort upon H ailey, w ho bantered him ab o u t his idealism . The bishop claim ed the m athem atician as an idealist also ,

RELIGION IN PRIMITIVE CULTURE

his ultim ate ratios being ghosts o f departed q uan tities, ap p earin g when the term s that p ro ­ duced them vanished. It rem ains to sum up in few w ords the d oc­ trine o f souls, in the various ph ases it has assum ed from first to last am o n g m ankind. In the attem pt to trace its m ain course through the successive grades o f m an ’s intellectual h istory, the evidence seem s to accord best with a theory o f its developm ent, som ew hat to the follow ing effect. At the low est levels o f culture o f which we have clear know ledge, the notion o f a ghost-soul anim ating m an while in rhe body, and ap p earin g in dream and vision out o f the body, is found deeply ingrained. There is no reason to think that this belief w as learnt by sav age tribes from con tact with higher races, nor that it is a relic o f higher culture from which the savage tribes have degener­ ated; for w hat is here treated a s the prim itive anim istic doctrine is thoroughly at home am on g savages, w ho ap p ear to hold it on the very evidence o f their senses, interpreted on the biological principle which seem s to them m ost reason able. We m ay now and then hear the sav age doctrines and practices concerning souls claim ed as relics o f a high religious culture pervading the prim eval race o f m an. They are said to be traces o f rem ote ancestral religion, kept up in scanty and perverted m em ory by tribes degraded from a nobler state. It is easy to see that such an explanation o f som e few facts, sundered from their c o n ­ n exion with the general array, m ay seem p la u ­ sible to certain m inds. But a large view o f the subject can hardly leave such argum ent in p o s­ session. The anim ism o f sav ages stan ds for and by itself; it explains its ow n origin. The anim ism o f civilized men, while m ore ap p rop riate to advanced know ledge, is in great m easure only explicable as a developed product o f the older and ruder system . It is the doctrines and rites o f the low er races which are, accordin g to their ph ilosoph y, results o f point-blank natural evi­ dence and acts o f straightforw ard practical purpose. It is the doctrines and rites o f the higher races which show survival o f the old in the m idst o f the new, m odification o f the old to bring it into conform ity with the new, a b a n ­ donm ent o f the old because it is no longer com p atible with the new.

31

Let us see at a glance in w hat general rela­ tion the doctrine o f souls am o n g savage tribes stan d s to the doctrine o f sou ls am on g barbaric and cultured nations. A m ong races within the lim its o f savagery, rhe general doctrine o f souls is found w orked out with rem arkable breadth and consistency. The souls o f anim als arc rec­ ognized by a natural extension from the theory o f hum an souls; the souls o f trees and plants follow in som e vague partial w ay; and the souls o f inanim ate objects expan d rhe general category to its extrem est boun dary. Thence­ forth, as we explore hum an thought onw ard from savage into barbarian and civilized life, we find a state o f theory m ore conform ed to positive science, but in itself less com plete and consistent. Far on into civilization, men still act a s though in som e half-m eant w ay they believed in souls or ghosts o f ob jects, while nevertheless their know ledge o f physical science is beyond so crude a philosophy. A s to the doctrine o f souls o f plan ts, fragm entary evidence o f the history o f its breaking dow n in A sia has reached us. In our ow n day and country, the notion o f souls o f beasts is to be seen dying out. Anim ism , indeed, seem s to be draw ing in its ou tp osts, and concentrating itself on its first and main position, the doctrine o f the hum an soul. This doctrine has undergone extrem e m odification in rhe course o f culture. It has outlived the alm ost total loss o f one great argum ent attached to it - the objective reality o f app arition al souls or ghosts seen in dream s and visions. The soul has given up its ethereal substan ce, and becom e an im m aterial entity, “ the sh adow o f a sh ad e.” Its theory is becom ing separated from the investigations o f biology and m ental science, which now discuss the phenom ena o f life and thought, the senses and the intellect, the em o­ tions and the will, on a groun dw ork o f pure experience. There has arisen an intellectual product w hose very existence is o f the deepest significance, a “ p sych ology” which has no longer anything to d o with “ so u l.” The sou l’s place in m odern thought is in the m etaphysics o f religion, and its especial office there is rhat o f furnishing an intellectual side to the religious doctrine o f the future life. Such are the alterations which have differ­ enced rhe fundam ental anim istic belief in its

32

EDW ARD BURNETT TYLOR

course through successive periods o f the w o rld ’s culture. Yet it is evident that, n otw ith standing all this profoun d change, the conception o f the hum an soul is, a s to its m ost essential n ature, con tin u ous from the philosophy o f the savage thinker to that o f the m odern professor o f the­ ology. Its definition has rem ained from the first that o f an an im atin g, sep arable, surviving entity, the vehicle o f individual personal e x is­ tence. T he theory o f the soul is one principal part o f a system o f religious philosophy which unites, in an unbroken line o f m ental co n n ex­ ion, the savage fetish-w orshipper and the civi­ lized C h ristian . The divisions which have separated the great religions o f the w orld into intolerant and hostile sects are for the m ost p art superficial in com p arison with the deepest o f all religious schism s, rhat w'hich divides Anim ism from M aterialism . |...| Lastly, a few w ords o f explanation m ay be offered as to the topics which this survey has included and excluded. T o those w ho have been accu stom ed to find theological subjects dealt with on a d ogm atic, em otion al, and ethical, rather than an ethnographic schem e, the present investigation m ay seem m isleading, because one-sided. T his one-sided treatm ent, how ever, has been ad op ted with full co n sid er­ ation. T h u s, though the doctrines here e x a m ­ ined bear not only on the developm ent but the actual truth o f religious system s, I have felt neither able nor w illing to enter into this great argum en t fully and satisfactorily, while e xp eri­ ence h as show n that to dispose o f such q u e s­ tions by an occasion al dictatorial phrase is one o f the m ost serious o f errors. T he scientific value o f description s o f savage and b arb aro u s religions, draw n up by travellers and especially by m ission aries, is often lowered by their controversial tone, and by the affectation o f infallibility with which their relation to the absolutely true is settled. There is som ething pathetic in the sim plicity with which a n arrow student will judge the doctrines o f a foreign religion by their an tagon ism or con form ity to his ow n o rth o d o x y , on points where utter d if­ ference o f opinion exists am on g the m ost learned and enlightened scholars. The system atization o f the low er religions, the reduction o f their m ultifarious details to

the few and sim ple ideas o f prim itive p h iloso­ phy which form the com m on groun dw ork of them all, appeared to me an urgently needed contribution to the science o f religion. This w ork I have carried ou t to the utm ost o f my pow er, and I can now only leave the result in rhe hands o f other students, w hose province it is to deal with such evidence in wider schem es o f argum ent. A gain, the intellectual rather than the em o­ tional side o f religion has here been kept in view. Even in the life o f the rudest savage, religious belief is associated with intense em otion, with aw ful reverence, with ago n iz­ ing terror, with rapt ecstasy when sense and thought utterly transcend rhe com m on level o f daily life. H ow much rhe m ore in faiths where not only does the believer experience such enthusiasm , but where his utm ost feelings of love and hope, o f justice an d mercy, o f for­ titude and tenderness and self-sacrificing devotion, o f unutterable m isery and dazzling happiness, twine and clasp round the fabric of religion. L an gu age, d ro p p in g at tim es from such w ords as soul and spirit their mere p h ilo­ sophic m eaning, can use them in full co n fo r­ mity with this tendency o f the religious m ind, a s phrases to convey a m ystic sense o f tran­ scendent em otion. Yet o f all this religion, the religion o f vision and o f passio n , little indeed has been said in these p ages, and even rhat little rather in incidental touches than with purpose. T h ose to w hom religion m eans above all things religious feeling, m ay say o f my a rg u ­ ment that 1 have w ritten soullessly o f the soul, and unspiritually o f spiritual things. Be it so: I accept the ph rase not as needing an apology, but as exp ressin g a plan. Scientific progress is at tim es m ost furthered by w orking alo ng a distinct intellectual line, w ithout being tem pted to diverge from the m ain object to w hat lies beyond, in how ever intim ate con­ nexion. The an atom ist does well to discuss bodily structure independently o f the w orld o f h appiness an d misery which depends upon it. It w ould be thought a mere im pertinence for a strategist to preface a dissertation on the science o f w ar, by an enquiry how far it is law ful for a C h ristian m an to bear w eapons and serve in the w ars. M y task has been here not to discu ss R eligion in all its bearings, but

RELIGION IN PRIMITIVE CULTURE

to portray in outline the great doctrine o f A nim ism , as found in w h at I conceive to he its earliest stages am o n g the low er races o f m ankind, and to show its tran sm ission alo n g the lines o f religious thought. T he alm ost entire exclusion o f ethical q u es­ tion s from this investigation has m ore than a m ere reason o f arrangem ent. It is due to the very nature o f the subject. T o som e the sta te ­ m ent m ay seem startlin g, yet the evidence seem s to justify it, rhat the relation o f m orality to religion is one rhat only belongs in its rudi­ m ents, or not at all, to rudim entary civiliza­ tion. The com parison o f sav ag e and civilized religions brings into view , by the side o f a deep-lying resem blance in their philosophy, a deep-lying co n trast in their practical action on hum an life. So far as sav age religion can stand a s representing natural religion, the p opu lar idea that the m oral governm ent o f the universe is an essential tenet o f natural religion sim ply falls to the groun d. Savage anim ism is alm ost devoid o f that ethical elem ent which to the educated m odern m ind is the very m ainspring o f practical religion. N o t, a s I have said, that m orality is absen t from the life o f the low er races. W ithout a code o f m orals, the very existence o f the rudest tribe w ould be im possible; and indeed the m oral stan d ard s o f even savage races are to no sm all extent welldefined and praisew orthy. But these ethical law s stand on their ow n groun d o f tradition and public opin ion , com paratively indepen­ dent o f the anim istic belief and rites which exist beside them. The low er anim ism is not im m oral, it is unm oral. For this plain reason, it has seem ed desirable to keep the discussion o f anim ism , as far as m ight be, sep arate from th at o f ethics. The general problem o f the rela­

33

tion o f m orality to religion is difficult, intri­ cate, and requiring an im m ense array of evidence, and m ay be p erh aps m ore profitably discu ssed in connexion with the ethnography o f m orals . . . T he essential connexion o f theology and m orality is a fixed idea in m any m inds. But it is one o f the lessons o f history that subjects m ay m aintain them selves independently for a ges, till the event o f coalescence takes place. In the course o f history, religion h as in various w ay s attached to itself m atters sm all and great ou tside its central schem e, such a s prohibition o f special m eats, observan ce o f special days, regulation o f m arriage as to kinship, division o f society into castes, ordin ance o f social law and civil governm ent. L o o k in g at religion from a political point o f view , a s a practical influ­ ence on hum an society, it is clear that am on g its greatest pow ers have been its divine sanction o f ethical law s, its theological en­ forcem ent o f m orality, its teaching o f m oral governm ent o f rhe universe, its supplantin g the “ continuance-doctrine” o f a future life by the “ retribution-doctrine” supplyin g m oral m otive in the present. But such alliance belongs alm ost or w holly to religions above the savage level, not to the earlier and low er creeds. It will aid us to see how m uch m ore the fruit o f religion belongs to ethical influence than to ph ilosoph i­ cal d o g m a, if we con sider how the introduc­ tion o f rhe m oral element sep arates the religions o f the w orld, united as they are throughout by one anim istic principle, into tw o great classes, those low er system s w hose best result is to supply a crude childlike natural philosophy, and those higher faiths which im plant on this the law o f righteousness and o f holiness, the inspiration o f duty and o f love.

2

The Elementary Forms of Religious Life Emile Durkheim

Em ile Durkheim (1858-1917) was a soci­ ologist of Jewish background concerned primarily w ith questions of social soli­ darity, vitality, and malaise in modernity, especially in his native France. If Tylor ended by arguing that primitive reli­ gions are characterized by their amoral quality, one o f the central aims for Durkheim is to show the intrinsic con­ nection of the moral and the religious. The excerpt here is inevitably composed of small portions of a large and hugely influential body of work. Like Tylor, Durkheim was an evolutionist and, like Tylor, he sought the origins of religion. But he was much clearer than Tylor (or Freud) th at one could not trace social phenom ena to some mom ent of sheer beginning, and so restates the question of origins in a structural m anner as a quest for "the ever-present causes upon w hich the most essential forms of reli­ gious thought and practice depend" (1915 [1912]: 20). His strategy is also

radically d ifferent from Tylor's. W here Tylor progresses by citing a vast sample of material to support his g eneraliza­ tions, Durkheim turns a slow and careful eye on w hat he considered a single case, namely the Aborigines of central Australia, th at he tho ug ht could show the "elem entary forms" o f religion most directly. Durkheim also departs from Tylor by proposing an original w ay in which a nonbeliever can yet understand any and every religion as not being in error. Furtherm ore, the core of his defi­ nition of religion lies not w ith any spe­ cific belief or kind of belief but with a system of classification. By defining the sacred as that which is set apart, Durkheim deftly evades having to give it any substantive content, a strategy th at has enabled subsequent scholars to move beyond trite defini­ tions. This is vastly superior to some­ thing like "belief in the supernatural," w here, as noted in the General Intro­

From Emile Durkheim, The Elementary Forms o f Religious Lifey ed. and trans. Karen Fields {New York: The Free Press, 1995 [1912]), pp. 1-13, 35—44, 41 9 -2 3 . Abridged, most footnotes removed.

THE ELEM ENTARY FORMS OF RELIGIOUS LIFE

duction, both "belief" and "super­ natural" beg a good many questions. (But see Collier and Yanagisako 1989 for a fem inist critique of the sacred/profane dichotomy.) A system of classification is also collective and no longer to be derived from individual psychology or experience or from Kantian innate cate­ gories of understanding. W ith respect to the social, Durkheim makes a number of significant arguments. First is the strong idea that religion is a natural expression of society, society's moment of reflecting on its own transcendent power. Second is the functionalist notion that religion provides a form of social cohesion, the glue of mechanical soli­ darity. These are actually inversions of one another: the first can be captured in the phrase that "the fam ily that stays together prays together" and the second, to quote from a billboard from my youth, that "the fam ily that prays together stays together." Perhaps w hat is most interesting about the latter is less the functionalism than the atten­ tion given to ritual as a form of action. Durkheim's most infamous argum ent is that all religions can be understood as true once it is seen th at w hat they rep­ resent is actually society. His position is thus one that recognizes the essentially symbolic quality of religion. If totems or gods symbolize society, this is not as reductive as it sounds, since Durkheim's understanding of society is itself so high-minded. Durkheim accepted and drew upon the dualism present at the tim e (in Freud as w ell) between the bio­ logical or natural individual and the social and moral collective. For Durkheim society enables hum anity to transcend itself, both to overcome selfish and violent urges and to seek, via the cate­ gories of understanding it provides, higher and ennobling paths. Society seen in this light does in fact approxi­ mate the view of religion as understood by many non-Durkheimians.

Durkheim influenced a subsequent line o f French thinkers, including his n ephew Marcel Mauss, w ho published in th e A n n £ e Sociologique. Am ong the significant essays of the Durkheim school (translated into English by EvansPritchard and his Oxford colleagues) are those by Durkheim and Mauss on sym­ bolic classification (1963 [1903]), Hubert and Mauss on sacrifice (1964 [1898]), Mauss on the gift (1990 [1925]), on bodily habitus (1973 [1935]), and on the concept of the person (1985 [1938]), and Hertz on death and on the right hand (1960 [1909]; cf. Needham, ed. 1973). Im portant collections of Durkheim's own essays (1973, 1974, 1992) include useful modern introductions. L6viStrauss was also interested in problems of symbolic classification but, as it is often said, he turned Durkheim on his head. If there are correspondences betw een society and ideational pat­ terns, for Levi-Strauss this is because both of these stem from the same source, namely the mind, rather than from society. Insofar as Durkheim depicted totemism as the elem entary form of religion, L£vi-Strauss's book deconstructing totemism may also be seen as an attack on his intellectual ancestor. In describing the northwest American potlatch as a "total" phenomenon, Mauss (1990 [1925]) makes the shatter­ ing observation that attem pting to dis­ tinguish the religious from the economic or the political makes little sense in certain kinds of societies, and may provide quite distorted images. These categories - religion, economy, etc. - are conceptual tools emerging from the social experience of modern western societies (based on organic solidarity). The ethnographic facts from other times and places are not tailored to fit them . M oreover, the lesson th at the study of other societies may bring back to us is the arbitrariness of our own systems of

35

36

EMILE DURKHEIM

classification and division into discrete social institutions. And thus mana itself, which w as the essence of religion in the theories of the proponents of animatism (w ho succeeded Tylor and the concept of animism), is revealed by Mauss to mean w ealth or authority as much as sacred power and, indeed, to refer to a w orld in which these are not understood as discrete and autonom ous. Since Durkheim's and Mauss's w ork on the categories of thought, one of the effects of the anthropological study of "religion" has thus been to immensely com plicate the issue by seeing things of "religious" or "symbolic" im port in domains of life that western society has tried to argue are quite distinct and built up entirely on practical, secular, or rational grounds. Conversely, w here reli­ gion is not separated from other social institutions, so it does not stand opposed to them as some morally distinct and distinctively moral realm. (The analogy here is to th e argum ent th at the gift in such societies cannot take on the con­ notations of pure generosity that it has for us. See the im portant discussion by Parry (1986).) A nother elem ent of Mauss's thought th at has had a profound influence on the anthropology of religion and is more fully w orked out in his essay on the person (1985 [1938]) is that the chiefs taking part in the potlatch w ere understood as incarnations of the gods and ancestors (cf. M auze 1994). Indeed, the concept of the "individual" no less than th at of "religion" is revealed as ethnographically and historically spe­ cific. This, in turn, would challenge th e ­ ories like Tylor's that tend to assume a universal individuality. These points have been particularly well developed by Dum ont (1970, 1986). Finally, it may be mentioned th at in his discussion of honor, Mauss makes questions of m orality central. Honor,

dignity, self-worth, and the virtuous com portm ent and action they suppose are as critical to human consciousness as the puzzlem ent, aw e, and fear attrib­ uted to humans by some thinkers or the instrum ental concerns w ith food, sex, or pow er attributed to them by others. Elsew here, in his essay on the body (1973 [1935]), Mauss sets out the notion of the habitus, subsequently developed by Bourdieu (1977), in which moral com­ portm ent is understood as rooted in em bodied habit. In these respects Mauss and Bourdieu draw on Aristotelian con­ ceptions of virtuous disposition and practice (Lam bek 2000a). The other major locus of Durkheim's influence was on the structuralfunctionalism that developed in British anthropology w ith Radcliffe-Brown (1964 [1952]) and produced a number of major studies of religion in specific societies (e.g., W arner 1959, Middleton 1987 [1960]). The British also drew on Durkheim's predecessors Robertson Smith (1894) and Fustel de Coulanges (1956 [1864]). For all these thinkers, society or the social group was under­ stood as primary, and among the British it was the representational and func­ tional sides of Durkheim's approach that w ere developed and elaborated, often making very good sense of aspects of the lineage-based societies of Africa, although not of all aspects (as EvansPritchard, in particular, was quick to note). Durkheim's conception of the sacred and his concerns w ith symbolic classification are most systematically pursued by Douglas, w ho is perhaps the most Durkheim ian of the generation of symbolic anthropologists (see chapter 16 below). Durkheim's emphasis on the moral remains extrem ely significant w hile his chief weakness, as has often been noted, lies w ith the inability to address historical change. W eber is much more attuned to history.

THE ELEM ENTARY FORMS OF RELIGIOUS LIFE

I propose in this book to study the sim plest and m ost prim itive religion that is known at present, to discover its principles and attem pt an explanation o f it. A religious system is said to be the m ost prim itive that is available for ob servation when it m eets the tw o follow ing condition s: First, it m ust be found in societies the sim plicity o f w hose organ ization is nowhere exceed ed ;1 second, it m ust be explain able w ithout the introduction o f any element from a predecessor religion. I will m ake every effort to describe the o rg a ­ nization o f this system with all the care and precision that an ethnographer or a historian w ould bring to the task. Hut my task will nor sto p ar description. Sociology sets itself differ­ ent problem s from those o f history or ethnog­ raphy. It does not seek to becom e acquainted with bygone form s o f civilization for the sole purpose o f being acquain ted with and recon­ structing them. Instead, like any positive science, its purpose above all is to explain a present reality that is near to us and thus cap ab le o f affecting ou r ideas and actions. T h at reality is m an. M ore especially, it is present-day m an, for there is none other that we have a greater interest in know ing well. T herefore, my study o f a very arch aic religion will not be for the sheer pleasure o f recounting the bizarre and the eccentric. I have m ade a very archaic religion the subject o f my research because it seem s better suited than any other to help us com prehend the religious nature of m an, that is, to reveal a fundam ental and perm anent aspect o f hum anity. T his proposition is bound to provoke strong objection s. It m ay be thought strange that, to arrive a t an understanding o f present-day hum anity, we should have to turn aw ay from it so as to travel back to the beginning o f history. In the m atter at h and, that procedure seem s especially u n o rth od o x. R eligions are held to be o f unequal value and stan din g; it is com m only said that not all contain the sam e m easure o f truth. T hu s it w ould seem that rhe higher form s o f religious th ought cannot be com pared with the low er w ithout bringing the higher form s dow n to the low er level. T o grant th at the crude cults o f A ustralian tribes might help us understand C h ristianity, for exam ple, is to assum e - is it n ot? - that Christianity

37

proceeds from the sam e m entality, in other w ord s, that it is m ade up o f the sam e supersti­ tion s and rests on the sam e errors. The theo­ retical im portance som etim es accorded to prim itive religions could therefore be taken as evidence o f a system atic irreligion that invali­ dated the results o f research by prejudging them. I need not g o into the question here whether sch olars can be found w ho were guilty o f this and w ho have m ade history and the eth n ogra­ phy o f religion a m eans o f m aking w ar again st religion. In any event, such could not possibly be a so cio lo g ist’s point o f view. Indeed, it is a fundam ental postu late o f sociology that a hum an institution can n ot resr upon error and falseh ood. If it did, it could not endure. If ir had not been groun ded in the nature o f things, in those very things it w ould have m et resis­ tance that it could not have overcom e. There­ fore, when I ap p roach the study o f prim itive religions, it is with the certainty that rhey are grounded in and express the real. In the course o f the an alyses and discu ssion s that follow , we will see this principle com ing up again and again . W hat I criticize in the sch ools I part co m pan y with is precisely that they have failed to recognize it. N o d o u b t, when all we d o is con sider the form u las literally, these religious beliefs and practices ap p ear disconcerting, and ou r inclination m ight be to write them o ff to som e sort o f inborn aberration . But we m ust know how to reach beneath the sym bol to g ra sp the reality it represents an d that gives the sym bol its true m eaning. The m ost bizarre or b arb aro u s rites and the strangest m yths tran s­ late som e hum an need and som e aspect o f life, w hether social or individual. The reason s the faithful settle for in justifying those rites and m yths m ay be m istaken, and m ost often are; but the true reason s exist nonetheless, and it is the busin ess o f science to uncover them. Fundam entally, then, there are no religions rhat are false. All are true after their own fashion: All fulfill given con ditions o f hum an existence, though in different w ays. G ran ted, it is not im possible to rank them hierarchically. Som e can be said to be superior to oth ers, in rhe sense that they bring higher m ental facul­ ties into play, that they are richer in ideas and feelings, that they contain proportionately

38

EMILE DURKHEIM

m ore concepts than sen sations and im ages, and that they are m ore elaborately system ­ atized. But the greater com plexity and higher ideal content, how ever real, are not sufficient to place the correspon din g religions into sep a­ rate genera. All are equally religious, just as all living beings are equally living beings, from the hum blest plastic! to m an. If 1 ad d ress m yself to primitive religions, then, it is not with any ulterior m otive o f d isp arag in g religion in general: These religions are to be respected no less than the others. They fulfill the sam e needs, play the sam e role, and proceed from the sam e cau ses; therefore, they can serve just as well to elucidate the nature o f religious life and, it follow s, to solve the problem I wish to treat. I - ..] M y research is not solely o f interest to the science o f religions. There is an aspect o f every religion that transcends the realm o f spec­ ifically religious ideas. T hrough it, the study o f religious phenom ena provides a m eans o f revisiting problem s that until now have been debated only am o n g philosophers. It has long been known that the first system s o f representations that man m ade o f the world and him self were o f religious origin. There is no religion that is not both a cosm ology and a speculation ab o u t the divine. If philosophy and the sciences were born in religion, it is because religion itself began by serving as science and philosophy. Further, and less often noted, religion has not merely enriched a hum an intellect already form ed but in fact has helped to form it. M en ow e to religion not only the content o f their know ledge, in signifi­ cant p art, but a lso the form in which that know ledge is elaborated. At the roo t o f ou r judgm ents, there are certain fundam ental notions that dom inate our entire intellectual life. It is these ideas that philosophers, beginning with A ristotle, have called the categories o f understanding: notions o f time, sp ace, num ber, cause, substan ce, per­ sonality. They correspon d to the m ost univer­ sal properties o f things. They are like solid fram es that confine thought. T hought does not seem to be able to break out o f them w ithout destroying itself, since it seem s we cannot think o f ob jects that are not in tim e or space,

that cannot be counted, and so forth. The other ideas are contingent and changing, and we can conceive o f a m an, a society, or an epoch that lacks them ; but these fundam ental notions seem to us as alm ost inseparable from the norm al functioning o f the intellect. They are, as it w ere, the skeleton o f thought. N o w , when one analyzes prim itive religious beliefs m ethodically, one naturally finds the principal categories am o n g them . They are born in and from religion; they are a product o f religious thought. This is a point that 1 will m ake again and again in the course o f this book. Even now that point has a certain interest o f its ow n, but here is w hat gives it its true significance. The general conclusion o f the chapters to follow is that religion is an eminently social thing. R eligious representations are collective representations that express collective realities; rites are w ays o f acting that are born only in rhe m idst o f assem bled groups and w hose purpose is to evoke, m aintain, or re-create certain m ental states o f those groups. But if the categories are o f religious origin, then they m ust participate in w hat is com m on to all religion: T hey, to o , m ust be social things, products o f collective thought. At the very least - since with our present understanding o f these m atters, radical and exclusive theses are to be guarded again st - it is legitim ate to say that they are rich in social elements. T h is, it m ust be ad d ed , is som ething one can begin to see even now for certain o f the ca te g o ­ ries. For exam ple, w hat if one tried to im agine w hat the notion o f time w ould be in the absence o f the m ethods we use to divide, m easure, and express it with objective signs, a time that w as not a succession o f years, m onths, w eeks, days, and hours? It w ould be nearly im possible to conceive of. We can conceive o f time only if we differentiate between m om ents. N o w , w hat is the origin o f that differentiation? U ndoubtedly, states o f consciousness that we have already experienced can be reproduced in us in the sam e order in which they originally occurred; an d, in this w ay, bits o f ou r past becom e im m ediate again , even while spon tan e­ ously distinguishing them selves from the present. But how ever im portant this distin c­ tion might be for ou r private experience, it is

THE ELEM ENTARY FORMS OF RELIGIOUS LIFE

far from sufficient ro constitute the notion or category o f time. The category o f time is not sim ply a partial or com plete com m em oration o f our lived life. It is an ab stract and im per­ sonal fram ew ork that con tain s not only our individual existence bur a lso that o f hum anity. It is like an endless can vas on which all d u ra ­ tion is spread ou t before the m ind’s eye and on which all possible events are located in relation to points o f reference that are fixed and specified. It is not my time that is o rg a ­ nized in this w ay; it is time that is conceived o f objectively by all men o f the sam e civiliza­ tion. This by itself is enough to m ake us begin to see that any such organ ization w ould have to be collective. And indeed, observation establishes that these indispensable poin ts, in reference to which all things are arran ged tem ­ porally, are taken from social life. The division into days, w eeks, m onths, years, etc., co rre­ spon d s to the recurrence o f rites, festivals, and public cerem onies at regular intervals. A calen dar expresses the rhythm o f collective activity while ensuring th at regularity. The sam e applies to space. As H am elin has show n, space is not the vague and indeterm i­ nate m edium that K ant im agined. If purely and absolutely h om ogeneous, it w ould be o f no use and w ould offer nothing for thought to hold on to. Spatial representation essentially co n ­ sists in a prim ary coordin ation o f given sense experience. But this coord in ation w ould be im possible if the parts o f space w ere q u alita­ tively equivalent, if they really w ere m utually interchangeable. T o have a spatial ordering of things is to be able to siruate them differently: to place som e on the right, others on the left, these above, those below , north o r south , east or w est, and so forth, just a s, to arran ge states o f con sciousn ess tem porally, it m ust be p o ssi­ ble to locate them at definite dates. T hat is, space w ould not be itself if, like time, it w as not divided and differentiated. But where do these divisions that are essential to space com e from ? In itself it has no right, no left, no high or low , no north or south , etc. All these d is­ tinctions evidently arise from the fact that dif­ ferent affective colorin gs have been assigned to regions. And since all men o f the sam e civiliza­ tion conceive o f space in the sam e m anner, it is evidently necessary that these affective co lo r­

39

ings and the distinctions that arise from them also be held in com m on - which im plies alm ost necessarily that they are o f social origin. Besides, in som e instances this social ch ar­ acter is m ade m anifest. T here are societies in A ustralia and N orth A m erica in which space is conceived in the form o f an im m ense circle, because the cam p itself is circular; and the spatial circle is divided in exactly the sam e way as the tribal circle and in its im age. As m any regions are distinguished as there are clans in the tribe, and it is the place the clans occupy in the encam pm ent that determ ines the orien­ tation o f the regions. Each region is defined by the totem o f the clan to which it is assigned. A m ong the Zun i, for exam ple, the pueblo is m ade up o f seven sections; each o f these sec­ tions is a grou p o f clan s that h as acquired its ow n unity. In all likelihood, it w as originally a single clan that later subdivided. Space sim i­ larly contain s seven regions, and each o f these seven sections o f the w orld is in intim ate rela­ tionship with a section o f the pueblo, rhat is, with a grou p o f clans. “ T h u s,” say s C ushing, “ one division is considered to be in relation with the north; another represents the w est, another the south, e tc.” Each section o f the pueblo has its distinctive co lo r, which sym bol­ izes it; each region has its ow n color, which is that o f the co rrespon din g section. O ver the course o f history, the num ber o f basic clans has varied, and the num ber o f regions has varied in the sam e w ay. T h u s, spatial o rg an iza­ tion w as m odeled on social organ ization and replicates it. Far from being built into human nature, no idea exists, up to and including the distinction between right and left, that is not, in all probability, the produ ct o f religious, hence collective, representations. A n alogou s dem on strations concerning the notions o f genus, force, personality, and effi­ cacy will be found below . O ne m ight even ask whether the notion o f con tradiction does not also arise from social conditions. W hat tends to m ake this plausible is the fact th at the hold the notion o f contradiction h as had over thought has varied with times and societies. T o d a y the principle o f identity governs scientific thought; but there are vast system s o f representation that have played a m ajor role in the history o f ideas, in which it is com m only ignored: These system s

40

EMILE DURKHEIM

are rhe m ythologies, from rhe cru dest to the m ost soph isticated. M yth ologies deal with beings that have rhe m ost con tradictory attri­ butes at the sam e tim e, that are one and m any, m aterial and sp iritual, and cap ab le o f su bdivid ­ ing them selves indefinitely w ithout losing that which m akes them w hat they are. T hese h istori­ cal variation s o f rhe rule rhat seem s to govern our present logic show rhat, far from being encoded from eternity in the m ental co n stitu­ tion o f m an, the rule depends at least in part upon historical, hence social, factors. We do not know exactly w hat these factors are, but we can presum e that they exist. O nce this hypothesis is accepted, the problem o f know ledge can be fram ed in new term s. I-.-1 R eligious phenom ena fall into tw o basic cate­ gories: beliefs and rites. T he first are states o f opinion and con sist o f represen tation s; the second are p articu lar m odes o f action . Between these tw o categories o f phenom ena lies all that separates thinking from doing. The rites can be distinguished from other hum an practices - for exam p le, m oral p ra c­ tices - only by the special nature o f their object. Like a rite, a m oral rule prescribes w ays of behaving to us, but those w ays o f behaving add ress objects o f a different kind. It is the object o f rhe rite that m ust be characterized, in order to characterize the rire itself. The special nature o f rhat object is expressed in rhe belief. Therefore, only after having defined the belief can we define the rite. W hether sim ple or com plex, all know n reli­ giou s beliefs display a com m on feature: They presuppose a classification o f the real or ideal things that men conceive o f into tw o classes tw o op p o site genera - that are w idely desig­ nated by tw o distinct term s, which the w ord s profane and sacred tran slate fairly well. The division o f the w orld into tw o d o m ain s, one containing all th at is sacred and the other all that is profan e - such is the distinctive trait o f religious thought. Beliefs, m yths, d o g m a s, and legends are either represen tation s or system s o f representations that exp ress the nature o f sacred things, the virtues and pow ers attrib ­ uted to them , their history, and their relation ­ ships with one another as well a s with profan e

things. Sacred things are not sim ply those personal beings that are called g o d s or spirits. A rock, a tree, a sprin g, a pebble, a piece o f w oo d, a house, in a w ord anything, can be sacred. A rite can have sacredness; indeed there is no rite that does not have it to som e degree. T here are w ords, ph rases, and form u las that can be said only by consecrated person ages; there are gestures and m ovem ents that cannot be executed by just anyone. If Vedic sacrifice has had such great efficacy - if, indeed, sacri­ fice w as far from being a m ethod o f gaining rhe g o d s’ favor bu t, accordin g to m ythology, actually generated the g o d s - that is because the virtue it p ossessed w as co m p arable to that o f the m ost sacred beings. T he circle o f sacred objects cannot be fixed once and for all; its scope can vary infinitely from one religion to another. W hat m akes Buddhism a religion is that, in the absence o f g o d s, it accepts rhe existence o f sacred things, nam ely, the Four N ob le T ru th s and the practices that are derived from them. But I have confined m yself thus far to enu­ m erating various sacred things as exam ples: 1 m ust now indicate the general characteristics by which they are distinguished from profan e things. One m ight be tem pted to define sacred things by the rank that is ordinarily assigned to them in the hierarchy o f beings. They tend to be regarded as superior in dignity and pow er to profan e things, and particularly to m an, in no w ay sacred when he is only a man. Indeed, he is portrayed as occupying a rank inferior to and dependent upon them. While that portrayal is certainly not w ithout truth, nothing a b o u t it is truly ch aracteristic o f the sacred. Subordin ation o f one thing to another is not enough to m ake one sacred and the other not. Slaves are subordinate to their m asters, subjects to their king, soldiers to their leaders, low er classes to ruling classes, the m iser to his go ld , and the pow er seeker to the pow er holders. If a m an is som etim es said to have the religion o f beings or things in which he recognizes an em inent value and a kind o f superiority to him, it is ob vious th at, in all such cases, the w ord is taken in a m etaphorical sense, and there is nothing in those relations that is religious in a strict sense.

THE ELEM ENTARY FORMS OF RELIGIOUS LIFE

O n the other hand, we should bear in mind that there are things with which m an feels rela­ tively at ease, even though they are sacred to the highest degree. An am ulet has sacredn ess, and yet there is nothing extraord in ary about the respect it inspires. Even face to face with his g o d s, m an is not alw ays in such a m arked state o f inferiority, for he very often uses phys­ ical coercion on them to get w hat he w ants. He beats the fetish when he is displeased, only to be reconciled with it if, in rhe end, it becom es m ore am enable to the w ishes o f its w orship­ per. T o get rain, stones are throw n into the sprin g or the sacred lake where the god o f the rain is presum ed to reside; it is believed that he is forced by this m eans to com e out and show him self. Furtherm ore, while it is true that m an is a dependent o f his g o d s, this dependence is m utual. The g o d s also need m an ; w ithout offerings an d sacrifices, they w ould die. I will have occasion to show that this dependence o f g o d s on their faithful is found even in the m ost idealistic religions. H ow ever, if the criterion o f a purely hierar­ chical distinction is at once to o general and too im precise, nothing but their heterogeneity is left to define the relation between the sacred and the profan e. But w hat m akes this hetero­ geneity sufficient to characterize that classifica­ tion o f things and to distinguish it from any other is th at it has a very particular feature: It is absolute. In the history o f hum an thought, there is no other exam ple o f tw o categories o f things as profoundly differentiated or as ra d i­ cally op p o sed to one another. The traditional opp o sitio n between go o d an d evil is nothing beside this one: G o o d and evil are tw o opposed species o f the sam e genus, nam ely m orals, just a s health an d illness are nothing m ore than tw o different asp ects o f the sam e order o f facts, life; by con trast, the sacred and the profan e are alw ays and everywhere conceived by the hum an intellect as sep arate genera, as tw o w orlds with nothing in com m on. The energies at play in one are not merely those encountered in the other, but raised to a higher degree; they are different in kind. T h is o p p o si­ tion has been conceived differently in different religions. H ere, localizing the tw o kinds of things in different regions o f the physical uni­ verse has ap p eared sufficient to separate them;

41

there, the sacred is throw n into an ideal and transcendent m ilieu, while the residuum is ab an d on ed as the property o f the m aterial w orld. But while the form s o f the co n trast are variab le, the fact o f it is universal. T h is is not to say that a being can never p ass from one o f these w orlds to the other. But when this p assage occu rs, the m anner in which it o ccu rs dem onstrates the fundam ental duality o f the tw o realm s, for it im plies a true m eta­ m orph osis. Rites o f initiation, which are p rac­ ticed by a great m any peoples, dem onstrate this especially well. Initiation is a long series o f rites to introduce the young m an into reli­ giou s life. For the first tim e, he com es out o f the purely profan e w orld, where he has passed his ch ildh ood, and enters into the circle o f sacred things. This change o f statu s is con­ ceived not a s a mere developm ent o f preexist­ ing seeds but as a tran sform ation totius substantiae.2 At that m om ent, the young man is said to die, and the existence o f the particu­ lar person he w as, to cease - instantaneously to be replaced by another. He is born again in a new form . A ppropriate cerem onies are held to bring ab o u t the death and the rebirth, which are taken not merely in a sym bolic sense but literally. Is this not p ro o f that there is a rupture between the profane being that he w as and the religious being that he becom es? Indeed, this heterogeneity is such that it degenerates into real an tagon ism . The tw o w orlds are conceived o f not only a s separate but a lso as hostile and jealous rivals. Since the con dition o f belonging fully to one is fully to have left the other, m an is exhorted to retire com pletely from the profan e in order to live an exclusively religious life. From thence com es m on asticism , which artificially organ izes a milieu that is ap art from , outside of, and closed to the natural milieu where ordinary men live a secu lar life, and that tends alm ost to be its an tago n ist. From thence a s well com es mystic asceticism , which seeks to up root all that may rem ain o f m an ’s attachm ent to the w orld. Finally, from thence com e all form s o f religious suicide, the crow ning logical step o f this asceti­ cism , since the only m eans o f escapin g profane life fully and finally is escapin g life altogether. The opposition o f these tw o genera is expressed outw ardly by a visible sign that

42

EMILE DURKHEIM

perm its ready recognition o f this very special classification, wherever it exists. The mind experiences deep repugnance ab o u t m ingling, even sim ple con tact, between the co rresp on d ­ ing things, because the notion o f the sacred is alw ays an d everywhere separate from the notion o f the profan e in m an’s m ind, and because we im agine a kind o f logical void between them . The state o f dissociation in which the ideas are found in con sciousn ess is too strongly con tradicted by such m ingling, or even by their being too close to one another. The sacred thing is, p ar excellence, that which the profan e m ust not and cannot touch with im punity. T o be sure, this prohibition cannot go so far as to m ake all com m unication between the tw o w orlds im possible, for if the profane could in no w ay enter into relations with the sacred , the sacred w ould be o f no use. This placing in relationship in itself is alw ays a delicate operation that requires precautions and a m ore or less com plex initiation. Yet such an operation is im possible if the profan e does not lose its specific traits, and if it d oes not becom e sacred itself in som e m easure and to som e degree. The tw o genera can n ot, at the sam e tim e, both com e close to one anoth er and rem ain what they were. N ow we have a first criterion o f religious beliefs. N o d ou b t, within these tw o fundam en­ tal genera, there are secondary species that are them selves m ore or less incom patible with each other. Rut characteristically, the religious phenom enon is such that it alw ays assum es a bipartite division o f the universe, know n and know able, into tw o genera that include all that exists bur radically exclude one another. Sacred things are things protected and isolated by proh ibition s; profan e things are those things to which the prohibitions are applied and that m ust keep at a distance from w hat is sacred. Religious beliefs are those representations that express the nature o f sacred things and the relations they have with other sacred things or with profan e things. Finally, rites are rules o f conduct that prescribe how man m ust conduct him self with sacred things. When a certain num ber o f sacred things have relations o f coordin ation and su b o rd in a­ tion with one another, so as to form a system that has a certain coherence and d oes not

belong to any other system o f the sam e sort, then the beliefs and rites, taken together, co n ­ stitute a religion. By this definition, a religion is not necessarily contained within a single idea and does not derive from a single princi­ ple that m ay vary with rhe circum stances it deals w ith, while rem aining basically the sam e everywhere. Instead, it is a whole form ed o f separate and relatively distinct p arts. F^ach hom ogeneous grou p o f sacred things, or indeed each sacred thing o f any im portance, con sti­ tutes an organ ization al center around which gravitates a set o f beliefs and rites, a cult o f its ow n. There is no religion, however unified it may be, that does not acknow ledge a plurality o f sacred things. Even C hristianity, at least in its C ath olic form , accepts the V irgin, the angels, the sain ts, the souls o f the dead, etc. above and beyond the divine personality (w ho, besides, is both three and one). As a rule, fu r­ therm ore, religion is not merely a single cult either but is m ade up o f a system o f cults that possess a certain autonom y. This autonom y is a lso variable. Som etim es the cults are ranked and subordinated to som e dom inant cult into which they are eventually absorb ed; but som e­ times as well they sim ply exist side by side in confederation. The religion to be studied in this book will provide an exam ple o f this con ­ federate organization. At the sam e tim e, we can explain why groups o f religious phenom ena that belong to no co n ­ stituted religion can exist: because they are not or are no longer integrated into a religious system . If, for specific reason s, one o f those cults just m entioned should m anage to survive while the whole to which it belonged has d is­ ap peared, it will survive only in fragm ents. T h is is w hat has happened to so m any agrarian cults that live on in folklore. In certain cases, w hat persists in th at form is not even a cult, bur a mere cerem ony or a particular rite. A lthough this definition is merely prelim i­ nary, it indicates the term s in which the problem that dom in ates the science o f reli­ gions m ust be posed. If sacred beings are believed to be distinguished from the others solely by the greater intensity o f the pow ers attributed to them , the question o f how men could have im agined them is rather sim ple: N oth ing m ore is needed than to identify those

THE ELEM ENTARY FORMS OF RELIGIOUS LIFE

forces th at, through their exceptional energy, have m anaged to im press the hum an mind forcefully enough to inspire religious feelings. But if, as I have tried to establish , sacred things are different in nature from profan e things, if they are different in their essence, the problem is far m ore com plex. In that case, one m ust ask w hat led m an to see the w orld as tw o hetero­ geneous and in com parable w orlds, even though nothing in sense experience seem s likely to have suggested the idea o f such a radical duality. Even so, this definition is not yet com plete, for it fits equally well tw o orders o f things that m ust be distinguished even though they are akin: m agic and religion. M ag ic, to o, is m ade up o f beliefs and rites. Like religion, it has its ow n m yths and d ogm as, but these are less well developed, probably because, given its pursuit o f technical and utili­ tarian ends, m agic does not w aste time in pure speculation. M agic also has its cerem onies, sacrifices, purifications, prayers, son gs, and dances. T hose beings w hom the m agician invokes and the forces he puts to w ork are not only o f the sam e nature a s the forces addressed by religion but very often are the sam e forces. In the m ost prim itive societies, the souls o f the dead are in essence sacred things and objects o f religious rites, but at the sam e tim e, they have played a m ajor role in m agic. In A ustralia as well a s in M elan esia, in ancient Greece as well as am on g C hristian peoples, the sou ls, bones, and hair o f the dead figure am on g the to ols m ost often used by rhe m agician. D em ons are a lso a com m on instrum ent o f m agical influ­ ence. N o w , dem ons are a lso surrounded by proh ibition s; they too are separated and live in a w orld apart. Indeed, it is often difficult to distinguish them from g od s proper. Besides, even in Christianity, is not the devil a fallen g o d ? And ap art from his origin s, does he not have a religious ch aracter, sim ply because the hell o f which he is the keeper is an indispens­ able part in the m achinery o f the C hristian religion? The m agician can invoke regular and official deities. Som etim es these are g od s o f a foreign people: For exam p le, the G reek m agi­ cian s called upon Egyptian, A ssyrian, or Jew ish gods. Som etim es they are even national gods:

43

H ecate and D iana were ob jects o f a m agic cult. The Virgin, the C h rist, and the saints were used in the sam e m anner by C h ristian m agicians. M u st we therefore say that m agic can n ot be rigorously differentiated from religion - that m agic is full o f religion and religion full o f m agic an d, consequently, that it is im possible to separate them and define the one w ithout the other? W hat m akes that thesis hard to sustain is the m arked repugnance o f religion for m agic and the hostility o f m agic to religion in return. M agic takes a kind o f professional pleasure in profaning holy things, inverting religious cerem onies in its rites. On the other hand, while religion has not alw ays condem ned and prohibited m agic rites, it has generally regarded them with disfavor. As m essieurs H ubert and M au ss point o u t, there is som e­ thing inherently antireligious ab o u t the m aneu­ vers o f the m agician. So it is difficult for these tw o institutions not to o ppo se one another at som e point, w hatever the relations between them. Since my intention is to limit my research to religion and stop where m agic begins, discovering w hat distinguishes them is all the m ore im portant. H ere is how a line o f dem arcation can be draw n between these tw o dom ains. R eligious beliefs proper are alw ays shared by a definite group that professes them and that practices the correspon din g rites. N ot only are they individually accepted by all m em bers o f that grou p, but they also belong to the grou p and unify it. T he individuals w ho com prise the grou p feel joined to one another by the fact o f com m on faith. A society whose m em bers are united because they im agine the sacred w orld and its relations with the profan e w orld in the sam e w ay, and because they translate this com m on representation into identical practices, is w hat is called a Church. In history we d o not find religion w ithout Church. Som etim es the Church is narrow ly national; som etim es it extends beyond frontiers; som etim es it encom passes an entire people (R om e, A thens, the H ebrew s); som etim es it encom passes only a fraction (C hristian denom inations since the com ing o f Protestantism ); som etim es it is led by a body o f priests; som etim es it is m ore or less w ithout any official directing body. But

44

EMILE DURKHEIM

wherever we observe religious life, it h as a definite grou p a s its basis. Even so-called private cults, like the dom estic cult or a c o r­ porate cult, satisfy this condition: They are alw ays celebrated by a gro u p , the fam ily or the co rp oratio n . A nd, furtherm ore, even these private religions often are merely special form s o f a broader religion that em braces the totality o f life. T hese sm all C h urches are in reality only chapels in a larger Church and, because o f this very scope, deserve all the m ore to be called by that name. M agic is an entirely different m atter. G ran ted , m agic beliefs are never w ithout a certain currency. They are often w idespread am on g bro ad strata o f the p op u lation , and there are even peoples where they count no fewer active follow ers than religion proper. But they d o not bind men w ho believe in them to one another and unite them into the sam e grou p , living the sam e life. There is no Church o f magic. Between the m agician and the indi­ viduals w ho con sult him, there are no durable ties that m ake them m em bers o f a single m oral body, co m p arab le to the ties that join the faithful o f the sam e god or the adherents o f the sam e cult. T he m agician has a clientele, not a C hurch, and his clients m ay have no m utual relations, and m ay even be unknow n to one another. Indeed, the relation s they have with him are generally accidental and transient, an alogou s to those o f a sick m an with his doctor. The official an d public ch aracter with which the m agician is som etim es invested m akes no difference. T h at he functions in broad daylight does not join him in a m ore regular and lastin g m anner with those w ho m ake use o f his se r v ic e s.. . . By co n trast, religion is in separable from the idea o f Church. In this first regard, there is already a fundam ental difference between m agic and religion. Furtherm ore, and above all, when m agic societies o f this sort are form ed, they never en com pass all the adher­ ents o f m agic. Far from it. They en com pass only the m agician s. Excluded from them are the laity, a s it were - that is, those for w hose benefit the rites are conducted, which is to say those w ho are the adherents o f regular cults. N o w , the m agician is to m agic w hat the priest is to religion. But a college o f priests is no m ore

a religion than a religious congregation that w orships a certain sain t in the sh adow s o f the cloister is a private cult. A Church is not sim ply a priestly broth erh ood; it is a m oral com m u­ nity m ade up o f all the faithful, both laity and priests. M agic ordin arily has no com m unity o f this sort. But if one includes the notion o f Church in the definition o f religion, does one not by the sam e stroke exclude the individual religions that rhe individual institutes for him self and celebrates for him self alone? There is scarcely any society in which this is not to be found. As will be seen below , every O jibw ay has his personal mamtou th at he ch ooses him self and to which he bears specific religious ob ligation s; the M elanesian o f the Banks Islands has his tam aniu ; rhe R om an h as his genius; the C h ris­ tian has his patron saint and his guardian angel, and so forth. All these cults seem , by definition, to be independent o f the group. And not only are these individual religions very com m on th roughout history, but som e people to d ay pose the question whether such religions are not destined to becom e the d om i­ nant form o f religious life - whether a day will not com e when the only cult will be the one that each person freely practices in his inner­ m ost self. But, let us put aside these speculation s about the future for a m om ent. If we confine our discussion to religions a s they are in the present and as they have been in the past, it becom es o b vious that these individual cults are not d is­ tinct and au ton om ou s religious system s but sim ply asp ects o f the religion com m on to the whole Church o f which the individuals are part. The patron saint o f the Christian is chosen from the official list o f sain ts recog­ nized by the C ath o lic C h urch, and there are can on ical law s that prescribe how each believer m ust con duct this private cult. In the sam e w ay, the idea that every m an necessarily has a protective genie is, in different form s, at the b asis o f a large num ber o f A m erican religions, a s well a s o f R om an religion (to cite only these tw o exam ples). A s will be seen below , that idea is tightly bound up with the idea o f soul, and the idea o f soul is not am o n g those things that can be left entirely to individual choice. In a w ord, it is the Church o f which he is a

THE ELEM ENTARY FORMS OF RELIGIOUS LIFE

m em ber that teaches the individual w hat these personal g od s are, w hat their role is, how he m ust enter into relations with them , and how he m ust honor them . When one analyzes the doctrines o f that Church system atically, sooner or later one com es acro ss the doctrines that concern these special cults. T hus there are not tw o religions o f different types, turned in op p o site directions, but the sam e ideas and principles applied in both cases - here, to cir­ cu m stan ces that concern the grou p a s a w hole, and there, to the life o f the individual. Indeed, this unity is so close that, am o n g certain peoples, the cerem onies during which the believer first enters into com m unication with his protective genie are com bined with rites w hose public ch aracter is incontestable, nam ely, rites o f initiation. W hat rem ains are the present-day a sp ira­ tions tow ard a religion that w ould consist entirely o f interior and subjective states and be freely constructed by each one o f us. But no m atter how real those asp iratio n s, they cannot affect our definition: T his definition can be applied only to real, accom plished facts, not to uncertain possibilities. R eligions can be defined a s they are now or as they have been, not as they m ay be tending m ore or less vaguely to becom e. It is possible that this religious indi­ vidualism is destined to becom e fact; but to be able to say in w hat m easure, we m ust first know w hat religion is, o f w hat elem ents it is m ade, from w hat causes it results, and w h at function it perform s - all question s w hose answ ers cannot be p reordain ed, for we have not crossed the threshold o f research. O nly at the end o f this study will 1 try to look into the future. We arrive thus at the follow ing definition: A

religion is a unified system o f beliefs and prac­ tices relative to sacred things , that is to say , things set apart an d forbidden - beliefs and practices which unite into one single m oral community called a Church , all those who adhere to them. T he second element thus holds a place in my definition th at is no less essential than the first: In show in g that the idea o f reli­ gion is inseparable from the idea o f a Church, it conveys the notion that religion m ust be an eminently collective thing.

1. - . ]

*

45

M o st often, the theorists w ho have set out to express religion in rational term s have regarded it as being, first and forem ost, a system o f ideas that correspond to a definite object. T h at object has been conceived in different w ays nature, the infinite, the un know able, the ideal, an d so forth - but these differences are o f little im portance. In every case, the representations - that is, the beliefs - w ere considered the essential element o f religion. For their part, rites ap peared from this stan d p oin t to be no m ore than an external, contingent, and physi­ cal tran slation o f those inw ard states that alone were deem ed to have intrinsic value. T h is notion is so w idespread that m ost o f the time debates on the topic o f religion turn arou n d and ab o u t on the question o f whether religion can or cannot be reconciled with science - that is, whether there is room alo n g ­ side scientific know ledge for another form o f thought held to be specifically religious. But the believers - the men w ho, living a religious life, have a direct sense o f w hat con ­ stitutes religion - object th at, in term s o f their day-to-day experience, this w ay o f seeing does not ring true. Indeed, they sense that rhe true function o f religion is not to m ake us think, enrich our know ledge, or add representations o f a different sort and source to those we ow e to science. Its true function is to m ake us act an d to help us live. The believer w ho has co m ­ m uned with his god is not sim ply a man who sees new truths that the unbeliever know s not; he is a m an w ho is stronger. W ithin him self, he feels m ore strength to endure the trials o f existence or to overcom e them. H e is as though lifted above the hum an m iseries, because he is lifted above his hum an condition. Fie believes he is delivered from evil - w hatever the form in which he conceives o f evil. T he first article o f any faith is belief in salvation by faith. But it is hard to see how a mere idea could have that pow er, in fact, an idea is but one elem ent o f ourselves. H ow could it confer on us pow ers that are superior to those given us in ou r natural m akeup? As rich in em otive pow er a s an idea m ay be, it can n ot add anything to our natural vitality; it can only release em otive forces that are already w ithin us, neither creat­ ing nor increasing them. From the fact that we im agine an object as w orthy o f being loved and

46

EMILE DURKHEIM

sought after, it d oes not follow that we should feel stronger. Energies greater than those at our disposal m ust com e from the object, and, m ore than rhat, we m ust have som e m eans o f m aking them enter into us and blend into our inner life. T o achieve this, it is not enough that we think about them ; it is indispensable that we place ourselves under their influence, that we turn ourselves in the direction from which we can best feel that influence. In short, we m ust act; and so we m ust repeat the necessary acts as often as is necessary to renew their effects. From this stan d poin t, it becom es apparen t that the set o f regularly repeated action s that m ake up the cult regains all its im portance. In fact, anyone w ho h as truly practiced a religion know s very well that it is the cult that stim u ­ lates the feelings o f joy, inner peace, serenity, and enthusiasm th at, for the faithful, stand as experim ental p ro o f o f their beliefs. The cult is not merely a system o f signs by which the faith is outw ardly exp ressed; it is the sum total o f m eans by which that faith is created and re­ created periodically. W hether the cult con sists o f physical op eratio n s or m ental ones, it is alw ays the cult that is efficacious. T his entire study rests on the postulate that the unanim ous feeling o f believers dow n the ages cannot be m ere illusion. T herefore, like a recent ap ologist o f faith ,3 I accept that reli­ gious belief rests on a definite experience, w hose dem onstrative value is, in a sense, not inferior to that o f scientific experim ents, though it is different. I to o think “ that a tree is known by its fru its,” 4 and rhat its fertility is the best p ro o f o f w hat its roots are w orth. But merely because there exists a “ religious e xp eri­ ence,” if you will, that is grounded in som e m anner (is there, by the w ay, any experience that is n ot?), it by no m eans follow s that the reality which g rou n d s it should conform o b je c­ tively with the idea the believers have o f it. The very fact that the way in which this reality has been conceived h as varied infinitely in different times is enough to prove that none o f these conceptions exp resses it adequately. If the sci­ entist sets it dow n as axiom atic that the sen sa­ tions o f heat an d light that men have correspon d to som e objective cause, he does not thereby conclude that this cause is the sam e a s it ap p ears to the senses. Likew ise, even if the

feelings the faithful have are not im aginary, they still do not constitute privileged intui­ tions; there is no reason w hatever to think that they inform us better ab o u t rhe nature o f their object than ordinary sensations d o ab o u t the nature o f bodies and their properties. T o d is­ cover w hat that object con sists of, then, we m ust apply to those sen sations an analysis sim ilar to the one that has replaced the senses’ representation o f the w orld with a scientific and conceptual one. This is precisely w hat 1 have tried to do. We have seen that this reality - which m ythologies have represented in so m any different form s, but which is rhe objective, universal, and eternal cause o f those sui generis sen sations o f which religious experience is m ade - is society. I have show n w hat m oral forces it develops and how it aw aken s that feeling o f support, safety, and protective guidance which binds the m an o f faith to his cult. It is this reality that m akes him rise above him self. Indeed, this is the reality that m akes him, for w hat m akes m an is that set o f intellectual g o o d s which is civilization, and civilization is the w ork o f society. In this w ay is explained the preem i­ nent role o f rhe cult in all religions, w hatever they are. T h is is so because society cannot m ake its influence felt unless it is in action, and it is in action only if the individuals w ho co m ­ prise it are assem bled and acting in com m on. It is through com m on action that society becom es con scious o f and affirm s itself; society is above all an active cooperation . As I have show n, even collective ideas and feelings are possible only through the overt m ovem ents that sym bolize them . T h u s it is action that dom inates religious life, for the very reason that society is its source. T o all the reason s adduced to justify this conception, a final one can be added that em erges from this book as a whole. A long the w ay, I have established rhat the fundam ental categories o f thought, and thus science itself, have religious origins. T he sam e has been shown to be true o f m agic, and thus o f the various techniques derived from m agic. Besides, it has long been know n th at, until a relatively advanced m om ent in evolution, the rules o f m orality and law w ere not distinct from ritual prescriptions. In short, then, we can say that

THE ELEM ENTARY FORMS OF RELIGIOUS LIFE

nearly all rhe great social institutions were born in religion. For the principal features o f collec­ tive life to have begun as none other than various features o f religious life, it is evident that religious life m ust necessarily have been the eminent form and, as it were, the epitom e o f collective life. If religion gave birth to all that is essential in society, that is so because the idea o f society is the soul o f religion. T h u s religious forces are hum an forces, m oral forces. Probably because collective feel­ ings becom e con scious o f them selves only by settling upon external ob jects, those very forces could not organize them selves w ithout taking som e o f their traits from things. In this w ay, they took on a kind o f physical nature; they cam e to mingle as such with the life o f the physical w orld, and through them it w as thought possible to explain events in that w orld. But when they are considered only from this standpoint and in this role, we see only w hat is m ost superficial about them. In reality, the essential elem ents out o f which they are m ade are borrow ed from consciousness. O rdinarily, they d o not seem to have a hum an character except when they are thought o f in hum an form , but even the m ost im personal and m ost anonym ous are nothing other than objectified feelings.

O nly by seeing religions in this w ay does it becom e possible to detect their real m eaning. If we rely on appearan ces, the rites often seem to be purely m anual operation s - anointings, purifications, m eals. T o consecrate a thing, one places it in contact with a source o f reli­ giou s energy, just as today a body is placed in contact with a sourcc o f heat or electricity in order to heat or electrify it. The procedures used in the tw o cases are not essentially differ­ ent. U nderstood in this w ay, religious tech­ nique seem s to be a kind o f m ystical m echanics. But these physical operation s are but the outer envelope in which m ental operation s lie hidden. In the end, the point is not to exert a kind o f physical constrain t upon blind and, m ore than th at, im aginary forces but to reach, fortify, and discipline con sciousnesses. The low er reli­ gions have som etim es been called m aterialistic. T h at term is incorrect. All religions, even the crudest, are in a sense spiritualistic. The pow ers they bring into play are, above all, spiritual, and their prim ary function is to act upon m oral life. In this w ay, we understand that w hat w as done in the nam e o f religion cannot have been done in vain, for it is necessarily the society o f men, it is hum anity, that has reaped the fruits.

NOTES 1

I will call those societies and the men o f those societies prim itive in the sam e sense. T his term certainly lacks precision, but it is hard to avoid; if care is taken to specify

47

2 3 4

its m eaning, how ever, it can safely be used. O f the w hole essence. W illiam Ja m es, The Varieties o f Religious Experience [L on don , Lon gm an s, 1902]. Ibid. (p. 19 o f the French translation)

3

The Protestant Ethic and the Spirit of Capitalism Max Weber

A large question addressed by the anthropology of religion has concerned the em ergence of secular modernity. W hat is to be m eant by modernity, and how does it differ from other social form s? W hat has been the role of reli­ gion in its form ation? Here the most suggestive thinker has been German sociologist Max W eber (1864-1920), whose concepts of rationalization and the disenchantm ent of the w orld, no less than the elective affinity betw een certain forms of religious thought and certain kinds of economic structures and activities, notably between Protestant­ ism and capitalism , have led to rich analyses and enormous debate (e.g., Lehm ann and Roth, eds., 1993). A lthough they w ere first published in 1904-5 I have placed these excerpts from The Protestant Ethic a n d the Spirit o f Capitalism after those of Durkheim for tw o reasons. The first is that W eber's contribution is in a fundam ental sense more contem porary than Durkheim's.

W ith W eber there is no recourse to universalistic evolutionary, functional, or determ inist schemes; he is historical through and through. Furtherm ore, he no longer takes religion as an essence to be uncovered and defined from the bottom up, as it w ere, but is rather con­ cerned w ith the relationships between religious factors (ideas, practices, insti­ tutions, and forms of authority) and economic and political processes. Second, W eber cam e to be appreciated by most anthropologists after Durkheim. If during the first half of the 20th century Durkheim was the central figure, the second half was W eber's. This is pre­ cisely because his central concern lies w ith social change and specifically w ith the transitions to capitalism and m oder­ nity. I pluralize transitions because W eber's approach is always rigorously com parative. As W eber himself put it, he "always underscored those features in the total picture of a religion which have been decisive for the fashioning of

From M ax Weber, The Protestant Ethic and the Spirit o f Capitalism , trans. Talcotr Parsons (New York: Charles Scribner & Sons, 1958 [1904-5]), pp. 4 7 -5 6 , 8 7 -9 2 , 180-3. Reprinted by permission o f Simon &T Schuster, Inc., New York. Abridged, references and most notes removed.

THE PROTESTANT ETHIC AND THE SPIRIT OF CAPITALISM

the practical w ay of life, as w ell as those which distinguish one religion from another" (1946b [1915]: 294). W eber is interested in historical con­ tingency and, specifically, the w ay certain religious form ulations and class or status positions w ithin particular sociopolitical orders have an affinity w ith one another such that their con­ junction forms the basis for transform a­ tive social action. This is a nondeterm inist approach to historical generalization. W here the Durkheim ians look to the relationship between thought and ritual, W eber asks w hat a given religious form ulation establishes as ethical and practical outlooks for its adherents, and conversely, which class is likely to accept and advocate such a view of the world. Ritual is only of interest insofar as a priestly group practices excessive ritual­ ism by contrast to another status group th at em phasizes intellectual rationaliza­ tion, ecstatic experience, contem plation, or w hatever. W eber was particularly interested in the rise of w hat he called w orldly asceticism or a calling, as in the Protestant w ork ethic, and its relation­ ship to processes of economic and politi­ cal rationalization, as well as w hat he mem orably referred to as the "disen­ chantm ent of the w orld." All of this would be enough to justify W eber's im portance, but our selection indicates another reason for Weber's inspirational quality, namely his careful

In the title o f this study is used the som ew hat pretentious phrase, the spirit o f capitalism . W hat is to be understood by it? T he attem pt to give anything like a definition o f it brings out certain difficulties which are in the very nature o f this type o f investigation. If any object can be found to which this term can be applied with any understandable m eaning, it can only be an historical individ­ ual, i.e. a com plex o f elem ents associated in historical reality which we unite into a con cep­

49

and lucid depiction of m eaningful and ethical action specific to a highly particu­ lar ethos. He is a master of the art of interpretation and the elucidation of cultural difference; his use o f Benjamin Franklin as an exemplary "cultural text" anticipates both interpretive anthropo­ logy and cultural studies, and does so both in its method and its unblinking ability to see the cultural basis of his own tim e. W eber's depictions of various types o f religious institutions and in par­ ticu lar his discussion of charisma and its routinization have also been extremely influential. W eber's interest in theodicy is developed in the selection by Geertz (chapter 4). Finally, it should be men­ tioned th at the applications of W eber to econom ically naive and politically conservative versions of m odernization theory run counter to both his relativism and skepticism w ith respect to m oder­ nity and his own careful balancing of political and economic w ith cultural factors. W eber should be understood as com plem entary rather than opposed to Marx. Readers eager for more W eber should turn first to the three m agnifi­ cent essays on religion published in From M a x W eber (Gerth and Mills, eds., 1946). An essay that develops an evolutionary model of several stages from Weber's scheme is Bellah (1964), which was sub­ sequently developed into an insightful textbook on the anthropology of reli­ gion by Peacock and Kirsch (1980).

tual w hole from the stan dpoin t o f their cul­ tural significance. Such an historical concept, how ever, since it refers in its content to a phenom enon signifi­ can t for its unique individuality, cannot be defined accordin g to the form ula genus proximumt differentia $pecificay but it m ust be grad u ally put together out o f the individual parts which are taken from historical reality to m ake it up. T h u s the final and definitive con cept cannot stand at the beginning o f the

50

M AX W EBER

investigation, but m ust com e at the end. We m ust, in other w ord s, w ork out in the course o f the d iscu ssion , a s its m ost im portant result, the best conceptual form ulation o f w hat we here understand by the spirit o f cap italism , that is the best from the point o f view which interests us here. This point o f view (the one o f which we shall speak later) is, further, by no m eans the only possible one from which the historical phenom ena we are investigating can be analyzed. O ther stan dpoin ts w ou ld, for this as for every historical phenom enon, yield other characteristics a s the essential ones. The result is that it is by no m eans necessary to under­ stand by the spirit o f capitalism only w hat it will com e to m ean to us for the p urp oses o f our an alysis. This is a necessary result o f the nature o f historical concepts which attem pt for their m eth odological p urposes not to g rasp historical reality in ab stract general form ulas, but in concrete genetic sets o f relations which are inevitably o f a specifically unique and indi­ vidual character. T h u s, if we try to determine the object, the analysis and historical explanation o f which we are attem pting, it cannot be in the form of a conceptual definition, but at least in the beginning only a provisional description o f w hat is here m eant by the spirit o f capitalism . Such a description is, how ever, indispensable in order clearly to understand the object o f the investigation. For this purpose we turn to a docum ent o f that spirit which contains what we are look in g for in alm ost classical purity, and at the sam e time has the ad van tage o f being free from all direct relationship to reli­ gion, being thus, for our p urposes, free o f preconceptions. Remember, that time is money. He that can earn ten shillings a day by his labour, and goes abroad, or sits idle, one half of that day, though he spends but sixpence during his diversion or idleness, ought not to reckon that the only expense; he has really spent, or rather thrown aw ay, five shillings besides. Remember, that credit is money. If a man lets his money lie in my hands after it is due, he gives me the interest, or so much as I can make of it during that time. This amounts to a considerable sum where a man has good and large credit, and makes good use o f it.

Remember, that money is of the prolific, generating nature. Money can beget money, and its offspring can beget more, and so on. Five shillings turned is six, turned again it is seven and threepence, and so on, till it becomes a hundred pounds. The more there is of it, the more it produces every turning, so that the profits rise quicker and quicker. Fie that kills a breeding-sow, destroys all her offspring to the thousandth generation. He that murders a crown, destroys all that it might have produced, even scores of pounds. Remember this saying, The good paymaster is lord o f another man's purse. He that is known to pay punctually and exactly to the time he promises, may at any time, and on any occasion, raise all the money his friends can spare. This is sometimes o f great use. After industry and frugality, nothing contributes more to the raising of a young man in the world than punctuality and justice in all his dealings; therefore never keep borrowed money an hour beyond the time you pro­ mised, lest a disappointment shut up your friend’s purse for ever. The most trilling actions that affect a m an’s credit are to be regarded. The sound o f your hammer at five in the morning, or eight at night, heard by a creditor, makes him easy six months longer; but if he sees you at a billiardtable, or hears your voice at a tavern, when you should be at work, he sends for his money the next day; demands it, before he can receive it, in a lump. It shows, besides, that you are mindful of what you owe; it makes you appear a careful as well as an honest man, and that still increases your credit. Beware o f thinking all your own that you possess, and o f living accordingly. It is a mistake that many people who have credit fall into. T o prevent this, keep an exact account for some time both of your expenses and your income. If you take the pains at first to mention particulars, it will have this good effect: you will discover how wonderfully small, trifling expenses mount up to large sum s, and will discern what might have been, and may for the future be saved, without occasioning any great inconvenience.

THE PROTESTANT ETHIC AND THE SPIRIT OF CAPITALISM

For six pounds a year you may have the use o f one hundred pounds, provided you arc a man of known prudence and honesty. He that spends a groat a day idly, spends idly above six pounds a year, which is the price for the use of one hundred pounds. He that wastes idly a groat’s worth o f his time per day, one day with another, wastes the privilege of using one hundred pounds each day. He that idly loses five shillings* worth of time, loses five shillings, and might as pru­ dently throw five shillings into the sea. He that loses five shillings, not only loses that sum, but all the advantage that might be made by turning it in dealing, which by the time that a young man becomes old, will amount to a considerable sum of money. It is Benjam in Franklin w ho preaches to us in these sentences, the sam e which Ferdinand Kiirnberger satirizes in his clever and m ali­ cious Picture o f American Culture as the su p ­ posed confession o f faith o f the Yankee. T hat it is the spirit o f capitalism which here sp eaks in ch aracteristic fash ion , no one will doubt, how ever little we m ay wish to claim that every­ thing which could be un derstood as pertaining to that spirit is contained in it. Let us pause a m om ent to consider this p assag e , the ph iloso­ phy o f which Kiirnberger sum s up in the w ord s, “ They m ake tallow out o f cattle and m oney out o f m en.” The peculiarity o f this philosophy o f avarice ap p ears to be the ideal o f the honest m an o f recognized credit, and above all the idea o f a duty o f the individual tow ard the increase o f his cap ital, which is assum ed as an end in itself. Truly w hat is here preached is not sim ply a m eans o f m aking on e’s w ay in the w orld, but a peculiar ethic. T he infraction o f its rules is treated not as foolishness but as forgetfulness o f duty. T h at is the essence o f the m atter. It is not mere bu si­ ness astuteness, that sort o f thing is com m on enough, it is an ethos. This is the quality which interests us. When Ja c o b Fugger, in speaking to a bu si­ ness associate w ho had retired and w ho w anted to persuade him to d o the sam e, since he had m ade enough m oney and should let others have a chance, rejected that a s pusillanim ity and answ ered that “ he [Fugger| thought oth ­

51

erw ise, he w anted to m ake m oney as long as he c o u ld ” , the spirit o f his statem ent is evi­ dently quite different from that o f Franklin. W hat in the form er case w as an expression o f com m ercial daring and a personal inclination m orally neutral, in the latter takes on the ch ar­ acter o f an ethically coloured m axim for the conduct o f life. The concept spirit o f cap ital­ ism is here used in this specific sense, it is the spirit o f m odern capitalism . For that we are here dealing only with W estern European and A m erican capitalism is ob vious from the way in which the problem w as stated. C apitalism existed in C h ina, India, Babylon, in the classic w orld, and in the M iddle Ages. But in all these cases, a s we shall see, this particular ethos w as lacking. N o w , all Franklin’s m oral attitudes are coloured with utilitarianism . H onesty is useful, because it assures credit; so are punctuality, industry, frugality, and that is the reason they are virtues. A logical deducation from this w ould be that where, for instance, the a p p ear­ ance o f honesty serves the sam e purpose, that w ould suffice, and an unnecessary surplus o f this virtue w ould evidently ap pear to Franklin’s eyes as unproductive w aste. And as a m atter o f fact, the story in his autobiograph y o f his conversion to those virtues, or the d is­ cu ssion o f the value o f a strict m aintenance o f the appearan ce o f m odesty, the assid u ou s belittlem ent o f on e’s ow n deserts in order to gain general recognition later, confirm s this im pression. A ccording to Franklin, those virtues, like all others, are only in so far virtues as they are actually useful to the individual, and the surrogate o f mere appearance is alw ays sufficient when it accom plishes the end in view. It is a conclusion which is inevitable for strict utilitarianism . The im pres­ sion o f m any G erm an s that the virtues p ro ­ fessed by A m ericanism are pure hypocrisy seem s to have been confirm ed by this striking case. But in fact the m atter is not by any m eans so sim ple. Benjam in Franklin’s ow n character, as it ap p ears in the really unusual candidness o f his au tobiograp h y, belies that suspicion. The circum stance that he ascribes his recogni­ tion o f the utility o f virtue to a divine revela­ tion which w as intended to lead him in the path o f righteousness, show s that som ething

52

M AX W EBER

m ore than m ere garnishing tor purely egocen­ tric m otives is involved. In fact, the sitmmum bonum o f this ethic, rhe earning o f m ore and m ore m oney, co m ­ bined with the strict avoidance o f all sp on tan e­ ou s enjoym ent o f life, is above all com pletely devoid o f any eudacm on istic, nor to say h edo­ nistic, adm ixture. It is thought o f so purely as an end in itself, that from the point o f view o f the happiness o f, or utility to, the single indi­ vidual, it ap p ears entirely transcendental and absolutely irrational. M an is dom in ated by the m aking o f m oney, by acquisition a s the ulti­ m ate purpose o f his life. Econom ic acquisition is no longer subord in ated to m an as the m eans for the satisfaction o f his m aterial needs. T his reversal o f w hat we should call the natural relationship, so irrational from a naive point o f view, is evidently as definitely a leading principle o f cap italism as it is foreign to all peoples not under capitalistic influence. At the sam e rime it expresses a type o f feeling which is closely connected with certain reli­ giou s ideas. If we thus a sk , why should “ m oney be m ade ou t o f m en” , Benjam in Franklin him self, although he w as a colourless deist, answ ers in his au tobiograp h y with a q uotation from the Bible, which his strict C alvinistic father drum m ed into him again and again in his youth: “ Seest thou a m an diligent in his busin ess? He shall stand before k in gs” (Prov. xxii. 2 9 ). The earning o f m oney within the m odern econom ic order is, so long a s it is done legally, the result and the expression o f virtue and proficiency in a callin g; and this virtue and proficiency are, as it is now not difficult to see, the real Alpha and O m ega o f Franklin’s ethic, a s expressed in the p assage s we have q u o ted , as well a s in all his w orks w ithout exception. And in truth this peculiar idea, so fam iliar to us tod ay, but in reality so little a m atter o f course, o f on e’s duty in a calling, is w hat is m ost ch aracteristic o f the social ethic o f c a p i­ talistic culture, and is in a sense the fundam en ­ tal basis o f it. It is an ob ligation which the individual is su p p osed to feel and does feel to w ards the content o f his profession al activ­ ity, no m atter in w hat it con sists, in p articular no m atter whether it a p p ears on the surface as

a utilization o f his personal pow ers, or only of his m aterial possession s (as capital). O f co urse, this conception has not appeared only under capitalistic conditions. On the con ­ trary, we shall later trace its origins back to a time previous to the advent o f capitalism . Still less, n aturally, d o we m aintain that a con ­ scious acceptance o f these ethical m axim s on the part o f the individuals, entrepreneurs or lab ourers, in m odern capitalistic enterprises, is a condition o f the further existence o f presentday capitalism . The capitalistic econom y o f the present day is an im m ense cosm os into which the individual is born, and which presents itself to him , at least a s an individual, as an unalterable order o f things in which he m ust live. It forces the individual, in so far as he is involved in the system o f m arket relation ­ ships, to con form to capitalistic rules o f action. The m anufacturer w ho in the long run acts counter to these n orm s, will just a s inevitably be elim inated from the econom ic scene as the w orker w ho can n ot or will not ad ap t him self to them will be throw n into the streets w ithout a job. T hus the capitalism o f today, which has com e to dom inate econom ic life, educates and selects the econom ic subjects which it needs through a process o f econom ic survival o f the fittest. But here one can easily see the lim its o f the concept o f selection as a m eans o f h istori­ cal explan ation . In order that a m anner o f life so well ad ap ted to the peculiarities o f cap ital­ ism could be selected at all, i.e. should com e to dom in ate others, it had to originate som e­ where, and not in isolated individuals alone, but as a w ay o f life com m on to w hole grou p s o f men. T h is origin is w hat really needs e x p la ­ nation. C oncernin g the doctrine o f the m ore naive historical m aterialism , that such ideas originate as a reflection or superstructure o f econom ic situ atio ns, we shall speak m ore in detail below . At this point it will suffice for our purpose to call attention to the fact that w ithout d ou b t, in the country o f Benjam in Franklin’s birth (M assach u setts), the spirit o f capitalism (in the sense we have attached to it) w as present before the capitalistic order. There were com plain ts o f a peculiarly calculatin g sort o f profit-seeking in N ew England, as

THE PROTESTANT ETHIC AND THE SPIRIT OF CAPITALISM

distinguished from other parts o f A m erica, as early as 1632. It is further undoubted that capitalism rem ained far less developed in som e o f the neighbouring colon ies, the later South­ ern States o f the United States o f A m erica, in spite o f the fact that these latter were founded by large cap italists for business m otives, while the N ew England colonies were founded by preachers and sem inary g rad u ates with the help o f sm all bourgeois, craftsm en and yeom en, for religious reason s. In this case the causal relation is certainly the reverse o f that su g­ gested by the m aterialistic stan dpoin t. Hut the origin an d history o f such ideas is much m ore com p lex than the theorists o f the superstructure sup p ose. T he spirit o f cap ital­ ism , in the sense in which we are using the term , had to tight its w ay to suprem acy again st a w hole w orld o f hostile forces. A state o f mind such as that expressed in the p assages we have quoted from Franklin, and which called forth the ap p lause o f a w hole people, w ould both in ancient tim es and in the M iddle Ages have been proscribed as the low est sort o f avarice and as an attitude entirely lacking in self-respect. It is, in fact, still regularly thus looked upon by all those social grou p s which are least involved in or ad ap ted to m odern capitalistic con dition s. T h is is not wholly because the instinct o f acquisition w as in those tim es unknow n or undeveloped, a s has often been said. N or because the aim sacra fam es , the greed for g old , w as then, or now , less pow erful outside o f bourgeois capitalism than within its peculiar sphere, as the illusions of m odern rom an ticists are w ont to believe. The difference between the capitalistic an d precapitalistic spirits is not to be found at this point. T he greed o f the Chinese M an d arin , the old R om an aristocrat, or the m odern p easan t, can stand up to any co m p arison . And the aun sacra fames o f a N eap o litan cab-driver or barcaiuolo, and certainly o f A siatic representa­ tives o f sim ilar trad es, a s well a s o f the craftsm en o f southern European or Asiatic coun tries, is, as anyone can find out for him self, very much m ore intense, and especially m ore un scrupulous than that of, say , an Englishm an in sim ilar circum stances.

53

Lu th er's C o n cep tio n of th e C allin g . . . A lthough the R eform ation is unthinkable w ithout Luther’s ow n personal religious devel­ opm en t, and w as spiritually long influenced by his personality, w ithout C alvin ism his w ork could not have had perm anent concrete success. N evertheless, the reason for this com m on repugnance o f C ath o lics and Luther­ an s lies, at least partly, in the ethical peculiari­ ties o f C alvinism . A purely superficial glance show s that there is here quite a different rela­ tionship between the religious life and earthly activity than in either C atholicism or Luther­ anism . FLven in literature m otivated purely by religious factors that is evident. T ak e for instance the end o f the Divine Com edy , where the poet in Paradise stan ds speechless in his passive contem plation o f the secrets o f G o d , and com pare it with the poem which has com e to be called the Divine Comedy o f Puritanism. M ilton closes the last son g o f Paradise Lost after describing the expulsion from paradise as follow s: They, looking back, all the eastern side beheld O f paradise, so late their happy seat, Waved over by that flaming brand; the gate With dreadful faces thronged and fiery arms. Some natural tears they dropped, but wiped them soon: The world was all before them, there to choose Their place of rest, and Providence their guide. And only a little before M ichael had said to A dam : . . . Only add Deeds to thy knowledge answerable; add faith; Add virtue, patience, temperance; add love, By name to come called Charity, the soul O f all the rest: then wilt thou not be loth T o leave this Paradise, hut shall possess A Paradise within thee, happier far. O n e feels at once that this pow erful expres­ sion o f the Puritan’s serious attention to this

54

MAX W EBER

w orld, his acceptance o f his life in the w orld as a task, could not possibly have com e from the pen o f a m ediaeval writer. But it is ju st as uncongenial to Lutheranism , as expressed for instance in Luth er’s and Paul G erh ard ’s ch o­ rales. It is now ou r task to replace this vague feeling by a som ew h at m ore precise logical form ulation, and to investigate the fundam en­ tal basis o f these differences. The appeal ro national ch aracter is generally a mere co n fes­ sion o f ignorance, and in this case it is entirely untenable. T o ascribe a unified national ch ar­ acter to the Englishm en o f the seventeenth century w ould be sim ply to falsify history. Cavaliers and R oun dh eads did not ap p eal to each other sim ply as tw o parties, but as ra d i­ cally distinct species o f m en, and w hoever looks into the m atter carefully m ust agree with them. On the other hand, a difference of character between the English m erchant adven ­ turers and the old H an seatic m erchants is not to be found; nor can any other fundam ental difference between the English and G erm an characters at the end o f the M iddle A ges, which cannot easily be explained by the differ­ ences o f their political history. It w as the pow er o f religious influence, not alone, but m ore than anything else, which created the differences o f which we are co n sciou s to-day. We thus take as our starting-point in the investigation o f the relationship between the old Protestant ethic and the spirit o f capitalism the w orks o f C alvin , o f C alvinism , and the other Puritan sects. But it is not to be under­ stood that we expect to find any o f the founders or representatives o f these religious m ovem ents considering the prom otion o f w hat we have called the spirit o f capitalism as in any sense the end o f his life-work. We can n ot well m aintain that the pursuit o f w orldly g o o d s, conceived as an end in itself, w as to any o f them o f positive ethical value. O nce and for all it m ust be rem em bered that program m es o f ethical reform never were at the centre o f inter­ est for any o f the religious reform ers (am on g w hom , for our p u rp o ses, we m ust include men like M en n o, G eorge Fo x, and W esley). They were not the foun ders o f societies for ethical culture nor the proponen ts o f hum anitarian projects for social reform or cultural ideals. The salvation o f the soul and that alone w as

the centre o f their life and w ork. Their ethical ideals and the practical results o f their d o c­ trines were all based on that alone, and were the consequences o f purely religious m otives. We shall thus have to adm it that the cultural consequences o f the R eform ation were to a great extent, perhaps in the particular aspects with which we are dealing predom inantly, unforeseen an d even unw ished-for results o f the labours o f the reform ers. They were often far rem oved from or even in contradiction to all that they them selves thought to attain. The follow ing study m ay thus perhaps in a m odest w ay form a contribution to the under­ standing o f the m anner in which ideas becom e effective forces in history. In order, how ever, to avoid any m isunderstanding o f the sense in which any such effectiveness o f purely ideal m otives is claim ed at all, I m ay perh aps be perm itted a few rem arks in conclusion to this introductory discussion. In such a study, it m ay a t once be definitely stated, no attem pt is m ade to evaluate the ideas o f the R eform ation in any sense, whether it concern their social or their religious w orth. We have continually to deal with asp ects o f the R eform ation which m ust appear to the truly religious con sciousn ess as incidental and even superficial. For we are merely attem pting to clarify the parr which religious forces have played in form ing the developing web o f our specifically w orldly m odern culture, in the com plex interaction o f innum erable different historical factors. We are thus inquiring only to w hat extent certain characteristic features o f this culture can be im puted to the influence o f the R eform ation. At the sam e time we m ust free ourselves from the idea that it is possible to deduce the R eform ation , as a historically necessary result, from certain econom ic changes. C oun tless historical circum stances, which cannot be reduced to any econom ic law , and are not susceptible o f econom ic e x p lan a­ tion o f any sort, especially purely political p ro ­ cesses, had to concur in order that the newly created Churches should survive at all. On the other hand, how ever, we have no intention w hatever o f m aintaining such a foolish and doctrinaire th esis1 as that the spirit o f capitalism (in the provisional sense o f the term explained above) could only have arisen

THE PROTESTANT ETHIC AND THE SPIRIT OF CAPITALISM

as the result o f certain effects o f the R efo rm a­ tion, or even that capitalism as an econom ic system is a creation o f the R eform ation. In itself, the fact that certain im portant form s o f capitalistic business organ ization are know n to be considerably older than the R eform ation is a sufficient refutation o f such a claim . On the contrary, we only wish to ascertain whether and to w hat extent religious forces have taken part in the qualitative form ation and the q u an ­ titative expan sion o f that spirit over the w orld. Furtherm ore, w hat concrete asp ects o f our capitalistic culture can be traced ro them. In view o f the trem endous confusion o f interde­ pendent influences between the m aterial basis, the form s o f social and political organ ization , and the ideas current in the time o f the R efor­ m ation, we can only proceed by investigating whether and at w hat poin ts certain co rrela­ tions between form s o f religious belief and practical ethics can be w orked out. At the sam e time we shall as far as possible clarify the m anner and the general direction in which, by virtue o f those relationships, the religious m ovem ents have influenced the developm ent o f m aterial culture. Only when this has been determ ined with reason able accuracy can the attem pt be m ade to estim ate to w hat extent the historical developm ent o f m odern culture can be attributed to those religious forces and to w hat extent to others.

[. . . ] O ne o f the fundam ental elem ents o f the spirit o f m odern cap italism , and not only o f that but o f all m odern culture: rational conduct on the basis o f the idea o f the calling, w as born - that is w hat this discussion has sought to dem onstrate - from the spirit o f Christian asceticism . O ne has only to re-read the passage from Franklin, quoted at the beginning o f this essay, in order to see that the essential elements o f the attitude which w as there called the spirit o f capitalism are the sam e as w hat we have just show n to be the content o f the Puritan worldly asceticism , only w ithout the religious basis, which by Franklin’s time had died aw ay. The idea that m odern lab our h as an ascetic ch ar­ acter is o f course not new. Lim itation to spe­ cialized w ork, with a renunciation o f the Faustian universality o f m an which it involves, is a condition o f any valuable w ork in the

55

m odern w orld; hence deeds and renunciation inevitably condition each other today. This fundam entally ascetic trait o f m iddle-class life, if it attem pts to be a w ay o f life at all, and not sim ply the absence o f any, w as w hat Goethe w anted to reach, at the height o f his w isdom , in rhe Wanderjahren> and in the end which he gave to the life o f his Faust. For him the real­ ization m eant a renunciation, a departure from an age o f full and beautiful hum anity, which can no m ore be repeated in the course o f our cultural developm ent than can the flower o f the Athenian culture o f antiquity. The Puritan w anted to w ork in a calling; we are forced to do so. For when asceticism w as carried out o f m onastic cells into everyday life, and began to dom inate w orldly m orality, it did its p art in building the trem endous co sm os o f the m odern econom ic order. T h is order is now bound to the technical and econom ic con di­ tions o f m achine production which today determ ine the lives o f all rhe individuals who are born into this m echanism , not only those directly concerned with econom ic acquisition, with irresistible force. Perhaps it will so deter­ mine them until the last ton o f fossilized coal is burnt. In B axter’s view the care for external g o o d s should only lie on the shoulders o f the “ saint like a light clo ak , which can be thrown aside at any m om ent” . But fate decreed that the cloak should becom e an iron cage. Since asceticism undertook to rem odel the w orld and to w ork out its ideals in the w orld, m aterial g o o d s have gained an increasing and finally an inexorable pow er over the lives of men as at no previous period in history. T o ­ day the spirit o f religious asceticism - whether finally, w ho know s? - has escaped from the cage. But victorious capitalism , since it rests on m echanical foun dations, needs its support no longer. The rosy blush o f its laughing heir, the Fjilightenm ent, seem s a lso to be irretriev­ ably fading, and the idea o f duty in one’ s calling prow ls about in our lives like the ghost o f dead religious beliefs. W here the fulfilment o f the calling cannot directly be related to the highest spiritual and cultural values, or when, on the other hand, it need not be felt sim ply as econom ic com pu lsion, the individual gener­ ally ab an d o n s the attem pt to justify it ar all. In the field o f its highest developm ent, in the

56

M AX W EBER

United States, the pursuit o f w ealth, stripped o f its religious and ethical m eaning, tends to becom e associated with purely m undane p as* sion s, which often actually give it the character o f sport. N o one know s w ho will live in this cage in the future, or whether at the end o f this tre­ m endous developm ent entirely new prophets will arise, or there will be a great rebirth o f old ideas and ideals, or, if neither, m echanized petrification, em bellished with a sort o f co n ­ vulsive self-im portance. For o f the last stage o f this cultural developm ent, it m ight well be truly said : “ Specialists w ithout spirit, sen su al­ ists w ithout heart; this nullity im agines that it has attained a level o f civilization never before ach ieved.” Rut this brings us to the w orld o f judgm ents o f value and o f faith, with which this purely historical d iscu ssion need not be burdened. The next task w ould be rather to show the significance o f ascetic ration alism , which has only been touched in the foregoing sketch, for the content o f practical social ethics, thus for the types o f organ ization and the functions o f social grou p s from the conventicle to the State. Then its relations to hum anistic rationalism , its ideals o f life and cultural influence; further to the developm ent o f philosophical and scien­

NOTES 1

2

In spite o f this and the follow ing rem arks, which in my opinion are clear enough, and have never been ch anged, I have again and again been accused o f this. For the above sketch has deliberately taken up only the relations in which an influence o f religious ideas on the m aterial culture is

tific em piricism , to technical developm ent and to spiritual ideals w ould have to be analyzed. Then its historical developm ent from the m edi­ aeval beginnings o f w orldly asceticism to its dissolution into pure utilitarianism w ould have to be traced out through all the areas o f ascetic religion. Only then could the quantitative cul­ tural significance o f ascetic Protestantism in its relation to rhe other plastic elem ents o f m odern culture be estim ated. H ere we have only attem pted to trace the fact and the direction o f its influence to their m otives in one, though a very im portant point. But it w ould also further be necessary to inves­ tigate how Protestant A sceticism w as in turn influenced in its developm ent and its character by rhe totality o f social condition s, especially econom ic. The m odern m an is in general, even with the best w ill, unable to give religious ideas a significance for culture and national character which they deserve. But it is, of course, not my aim to substitute for a one­ sided m aterialistic an equally one-sided spiri­ tualistic causal interpretation o f culture and o f history. Each is equally p o ssib le,2 but each, if it does not serve as the prep aration , but as the conclusion o f an investigation , accom plishes equally little in the interest o f historical truth.

really beyond doubt. It w ould have been easy to proceed beyond that to a regular construction which logically deduced everything characteristic o f m odern culture from Protestant ration alism . But rhat sort o f thing m ay be left to the type o f dilettante w ho believes in the unity o f the grou p mind and its reduceability to a single fo r m u la .. . .

4

Religion as a Cultural System Clifford Geertz

Clifford G eertz (1926-2006) was an Am erican anthropologist trained at Harvard University w ho, after teaching for several years at the University of Chicago, served as Professor at the Insti­ tute for Advanced Study in Princeton, New Jersey. Geertz's essay is among the most interesting and influential of mid20th-century attem pts to develop a general depiction of religion. If W eber notoriously postponed defining religion, G eertz steps in on his behalf, but in a m anner th at effortlessly incorporates insights from both the Durkheim ian and Boasian traditions. Geertz's approach has even been referred to as neo-Tylorean. In fact, Geertz draws from the synthesis of Durkheim and W eber developed by the sociologist Talcott Parsons and is able to render Parsons' rather ponder­ ous thought in a much more readable language, in part because of his w on­ derful way w ith anecdote and quota­

tion and his finely w rought ethnographic descriptions. G eertz has been the m ajor exponent of a Weber-inspired interpretive anthro­ pology which attempts to understand religion w ithin a broadly cultural/ symbolic domain, but also w ith refer­ ence to public circumstances in all th eir historical messiness. In this essay Geertz describes how religions consti­ tute the worlds their adherents inhabit and provide guides for th eir action w ithin them . Such worlds must provide assurances that they are ultim ately com­ prehensive and com prehensible. W hen the pow er to do so breaks down, new religious movements may arise. Geertz is at his best w hen he does w h a t he advocates and weaves his theo ­ retical insights from ethnographic m ate­ rial, often by means of contrasting cases. Hence, in addition to turning to the unabridged version of this essay, I rec­ ommend readers exam ine a parallel exposition of the argum ent as worked

From Clifford Geertz, “ Religion as a Cultural System ,” in Anthropological Approaches to the Study o f Religion (London: Tavistock, 1966), pp. 1-46. Abridged.

58

CLIFFORD GEERTZ

out in a discussion of Islam Observed (1968, especially chapter 4). Geertz's preferred vehicle is the extended essay. The Interpretation o f Cultures (1973a) collects some of the best, including the one reprinted here. The first three essays of the book provide the theoretical underpinnings, w hile "Person, Time, and Conduct in Bali" (originally 1966) provides a superb exem plification of his approach to culture, and the concluding essay on the Balinese cockfight is w here he most fam ously and explicitly com­ pares cultural analysis to the interpreta­ tion of texts. O ther essays relevant to religion include Negara (1980), on the precolonial Balinese "theater state," Local K n o w le d g e (1983), and A vailab le L ig h t (2000). Despite, or perhaps because of, Geertz's staunch rejection of French structuralism, his w ork has found tre­

Any attem pt to sp eak without sp eakin g any particu lar lan gu age is not m ore hopeless than the attem pt to have a religion that sh all be no religion in p a r tic u la r .. . . Thus every living an d healthy religion has a m arked id iosyn ­ crasy. Its p o w er consists in its sp ecial an d surprising m essage an d m the bias which that revelation gives to life. The vistas it opens an d the m ysteries it p ro p o u n d s are another w orld to live in; an d another w orld to live in whether ive expect ever to p a ss wholly over into it or no - is w hat we m ean by having a religion. S a n ta y a n a

, Reason in Religion

As we are to deal with m eaning, let us begin with a p aradigm : viz., that sacred sym bols function to synthesize a people’ s ethos - the tone, ch aracter, an d quality o f their life, its m oral and aesthetic style and m ood - an d their w orld view - the picture they have o f the way

mendous resonance in the humanities. Equally, it has drawn more than its share of criticism from w ithin anthropology, often due to a false identification on the part of critics of Geertz's herm eneu­ tic methodology w ith philosophical ide­ alism and subjectivism (both of which Geertz is at pains to reject). "Person, Time, and Conduct in Bali" provoked a response from Bloch (1989b [1977]) that nicely illustrates the contrast between a Boasian/W eberian approach, on the one hand, and a Durkheimian/M arxian approach on the other. W hile G eertz is sometimes criticized for painting overly consistent portraits of other peoples' religions, in fact, his first book, The Reli­ g io n o f Java (1960), em phasized the internal diversity of tradition. Scholarly appraisal of Geertz can be found in Ortner, ed. (1999) and Schweder and Good (2005).

things in sheer actuality are, their m ost co m ­ prehensive ideas o f order. In religious belief and practice a g ro u p ’s ethos is rendered intel­ lectually reasonable by being shown to repre­ sent a way o f life ideally adapted to the actual state o f affairs the w orld view describes, while the w orld view is rendered em otionally co n ­ vincing by being presented as an im age o f an actual state o f affairs peculiarly w ell-arranged to accom m odate such a w ay o f life. T his con ­ frontation and m utual confirm ation has tw o fundam ental effects. On the one hand, it objectivizes m oral and aesthetic preferences by depicting them as the im posed conditions o f life im plicit in a w orld with a particular struc­ ture, as mere com m on sense given the unalter­ able shape o f reality. On the other, it su pp orts these received beliefs ab o u t the w orld’s body by invoking deeply felt m oral and aesthetic sentim ents a s experim ental evidence for their truth. R eligious sym bols form ulate a basic congruence between a p articular style o f life and a specific (if, m ost often, implicit) m eta­ physic, and in so doin g sustain each with the borrow ed authority o f the other.

RELIGION AS A CULTURAL SYSTEM

Phrasing asid e, this m uch m ay perhaps be granted. T he notion that religion tunes hum an actions to an envisaged cosm ic order and p ro ­ jects im ages o f cosm ic order on to the plane o f hum an experience is hardly novel. Hut it is hardly investigated either, so that we have very little idea o f how, in em pirical term s, this p a r­ ticular m iracle is accom plished. We just know rhat it is done, annually, weekly, daily, for som e people alm ost hourly; and we have an enorm ous ethnographic literature to dem on­ strate it. But the theoretical fram ew ork which w ould enable us to provide an analytic account o f it, an account o f the sort we can provide for lineage segm entation, political succession, labor exchange, or the socialization o f the child, does not exist. Let us, therefore, reduce our p aradigm to a definition, for, although it is n otorious that definitions establish nothing, in them selves they d o , if they are carefully enough co n ­ structed, provide a useful orientation, or reori­ entation, o f thought, such that an extended unpacking o f them can be an effective way o f developing and controlling a novel line o f inquiry. They have the useful virtue o f explicit­ ness: they com m it them selves in a w ay discu r­ sive p rose, which, in this field especially, is alw ays liable to substitute rhetoric for a rg u ­ m ent, d oes not. W ithout further ad o , then, a religion is: (1) a sy stem o f sy m b o ls w hich a c ts to (2) e sta b lish p o w e rfu l, p e rv a siv e , a n d longla stin g m o o d s a n d m o tiv a tio n s in m en by (3) fo r m u la tin g c o n c ep tio n s o f a g e n e ra l o rd e r o f e x isten c e a n d (4) c lo th in g th ese c o n c ep tio n s w ith su ch a n a u r a o f fa c tu a lity th a t (5) the m oods and m o tiv a tio n s seem u n iq u ely realistic. a system o f sym bols which a c ts to . . .

Such a trem endous weight is being put on the term “ sy m b ol” here that our first m ove m ust be to decide with som e precision w hat we are goin g to m ean by it. T h is is no easy task , for, rather like “ cu lture,” “ sym bol” has been used to refer to a great variety o f things, often a num ber o f them at the sam e time.

59

In som e hands it is used for anything which signifies som ething else to som eone: dark clou ds are rhe sym bolic precursors o f an on ­ com ing rain. In others it is used only for explic­ itly conventional signs o f one sort or another: a red flag is a sym bol o f danger, a white o f su r­ render. In others it is confined to som ething which expresses in an oblique and figurative m anner that which cannot be stated in a direct and literal one, so that there are sym bols in poetry but not in science, and sym bolic logic is m isnam ed. In yet others, how ever, it is used for any object, act, event, quality, or relation which serves a s a vehicle for a conception - the con ­ ception is the sym bol’s “ m eaning” - and rhat is rhe approach I shall follow h ere.1 The num ber 6, w ritten, im agined, laid out a s a row o f stones, or even punched into the program tapes o f a com pu ter, is a sym bol. Hut so a lso is the C ro ss, talked ab o u t, visualized, shaped w orriedly in air or fondly fingered at the neck, the expanse o f painted can vas called “ G u ern ica* or the bit o f painted stone called a churinga, the w ord “ reality,” or even the m orphem e “ -in g.” They are all sym bols, or at least sym bolic elements, because they are tangible form u lations o f n otion s, abstraction s from experience fixed in perceptible form s, concrete em bodim ents o f ideas, attitudes, judgm ents, longings, or beliefs. T o undertake the study o f cultural activity activity in which sym bolism form s the positive content - is thus not to aban don social analysis for a Platonic cave o f sh adow s, to enter into a m entalistic w orld o f introspective psychology or, w orse, speculative philosophy, and w ander there forever in a haze o f “ C o g n itio n s,” “ A ffec­ tio n s,” “ C o n atio n s,” and other elusive entities. C ultural acts, the construction, apprehension, and utilization o f sym bolic form s, are social events like any other; they are a s public as m ar­ riage and as observable as agriculture. They are not, how ever, exactly the sam e thing; or, m ore precisely, the sym bolic dim en­ sion o f social events is, like the psychological, itself theoretically absrractable from those events a s em pirical totalities. There is still, to p araph rase a rem ark o f Kenneth Burke’s, a difference between building a house and draw in g up a plan for building a house, and reading a poem ab o u t having children by m ar­ riage is not quite the sam e thing as having

60

CLIFFORD GEERTZ

children by m arriage.2 Even though the bu ild­ ing o f the house m ay proceed under the g u id ­ ance o f the plan or - a less likely occurrence - the having o f children m ay be m otivated by a reading o f the poem , there is som eth ing to be said for not con fusin g our traffic with sym bols with our traffic with objects or hum an beings, for these latter are not in them selves sym bols, how ever often they m ay function as such .1 N o m atter how deeply interfused the cultural, the social, and the psychological m ay be in the everyday life o f houses, farm s, poem s, and m arriages, it is useful to distinguish them in an alysis, an d , so doin g, to isolate the generic traits o f each again st the norm alized b a ck ­ ground o f the other tw o. So far as culture pattern s, that is, system s or com p lexes o f sym bols, are concerned, the generic trait which is o f first im portance for us here is that they are extrinsic sources o f in for­ m ation. By “ extrin sic,” I m ean only that unlike genes, for exam ple - they lie ou tside the boun daries o f the individual organ ism as such in that inter-subjective w orld o f com m on understandings into which all hum an in dividu­ als are born, in which they pursue their se p a ­ rate careers, and which they leave persisting behind them after they die. By “ sources o f in fo rm ation ,” I m ean only that - like genes they provide a blueprint or tem plate in term s o f which processes external to them selves can be given a definite form . As the order o f bases in a stran d o f D N A form s a coded p rogram , a set o f instructions, or a recipe, for the synthesis o f the structurally com plex proteins which shape organ ic functioning, so culture patterns provide such p ro g ram s for the institution o f the social and psychological processes which shape public behavior. T hough the sort of in form ation an d the m ode o f its tran sm ission are vastly different in rhe tw o cases, this co m ­ parison o f gene and sym bol is m ore than a strained an alogy o f the fam iliar “ social hered­ ity” sort. It is actually a substan tial relation ­ ship, for it is precisely because o f the fact that genetically program m ed p rocesses are so highly generalized in m en, as com pared with low er an im als, that culturally program m ed ones are so im portan t; only because hum an behavior is so loosely determ ined by intrinsic sources o f inform ation that extrinsic sources are so vital.

T o build a dam a beaver needs only an a p p ro ­ priate site and the p roper m aterials - his m ode o f procedure is shaped by his physiology. But m an, w hose genes are silent on the building trades, needs also a conception o f w hat it is to build a d am , a conception he can get only from som e sym bolic source - a blueprint, a tex t­ b o ok , or a string o f speech by som eone w ho already kn ow s how d am s are built - or, of co urse, from m an ipulatin g graphic or linguis­ tic elem ents in such a way as to attain for him self a conception o f w hat d am s are and how they are built. T h is point is som etim es put in rhe form o f an argum ent th at cultural patterns are “ m od els,” that they are sets o f sym bols w hose relations to one another “ m odel” relations am on g entities, processes or w hat-have-you in physical, organ ic, social, or p sych ologi­ cal system s by “ p arallelin g,” “ im itatin g,” or “ sim ulating” them .4 The term “ m odel” has, how ever, tw o senses - an “ o f ” sense and a “ fo r” sense - and though these are but asp ects o f the sam e basic con cept they are very much w orth distinguishing for analytic p urposes. In the first, w hat is stressed is the m anipulation o f sym bol structures so a s to bring them , m ore or less closely, into parallel with the preestablished n onsym bolic system , as when we g rasp how d am s w ork by developing a theory o f hydraulics or constructing a flow chart. The theory or ch art m odels physical relationsh ips in such a w ay - that is, by expressin g their structure in synoptic form - as to render them apprehensible; ir is a m odel of “ reality.” In the second, w hat is stressed is the m anipulation o f the nonsym bolic system s in term s o f the rela­ tionships expressed in rhe sym bolic, as when we construct a dam according to the specifica­ tions im plied in an hydraulic theory or the conclusions draw n from a flow chart. H ere, the theory is a m odel under w hose guidance physical relation sh ips are organized: it is a m odel fo r “ reality .” Eor psychological and social system s, and for cultural m odels that we w ould not ordinarily refer to as “ th eories,” but rather as “ d octrin es,” “ m elodies,” or “ rites,” the case is in no w ay different. Unlike genes, and other nonsym bolic inform ation sources, which are only m odels for » not m odels o f culture patterns have an intrinsic double

RELIGION AS A CULTURAL SYSTEM

aspect: they give m eaning, that is, objective conceptual form , to social and psychological reality both by sh ap in g them selves to it and by sh apin g it to them selves. It is, in fact, this double aspect which sets true sym bols o ff from other sorts o f significa­ tive form s. M od els for are found, as rhe gene exam p le suggests, through the w hole order o f nature; for wherever there is a com m unication o f pattern, such p rogram s are, in sim ple logic, required. A m ong an im als, im print learning is perh aps the m ost striking exam ple, because whar such learning involves is the autom atic presentation o f an ap p ro p riate sequence o f behavior by a m odel anim al in the presence o f a learning anim al which serves, equally au to ­ m atically, to call out and stabilize a certain set o f respon ses genetically built into the learning an im al.5 The com m unicative dance o f tw o bees, one o f which has found nectar and the other o f which seeks it, is an oth er, som ew hat different, m ore com plexly coded, exam ple.* C raik has even suggested that the thin trickle o f w ater which first finds its w ay dow n from a m ountain spring to the sea and sm ooth s a little channel for the greater volum e o f w ater th at follow s after it plays a sort o f model for function.7 But m odels o f - linguistic, graphic, m echanical, n atural, etc., p rocesses which function not to provide sources o f inform ation in term s o f which other processes can be p at­ terned, but to represent those patterned p ro­ cesses as such, to exp ress their structure in an alternative m edium - are m uch rarer and m ay perhaps be confined, am on g living an im als, to m an. The perception o f the structural co n gru­ ence between one set o f processes, activities, relation s, entities, and so on , and another set for which it acts a s a program , so that the program can be taken as a representation, or conception - a sym bol - o f the p rogram m ed, is the essence o f hum an thought. The intertran sposability o f m odels fo r and m odels o f which sym bolic form ulation m akes possible is the distinctive ch aracteristic o f ou r m entality. . . . to establish powerful, pervasive , and long-lasting moods and motivations in men b y . .. So far as religious sym bols and sym bol system s are concerned this intertransposability

61

is clear. The endurance, co u rage, indepen­ dence, perseverance, and p assio n ate w illful­ ness in which the vision q u est practices the Plains Indian are the sam e flam boyant virtues by which he attem pts to live: while achieving a sense o f revelation he stabilizes a sense o f direction .8 The con sciousn ess o f defaulted o b ligation , secreted guilt, an d, when a con fes­ sion is ob tain ed, public sham e in which M an u s’ seance rehearses him are the sam e sentim ents that underlie rhe sort o f duty ethic by which his property-con scious society is m aintained: the gaining o f an absolu tion involves the forging o f a conscience.* A nd the sam e selfdiscipline which rew ards a Ja v an ese mystic starin g fixedly into the flam e o f a lam p with w hat he takes to be an intim ation o f divinity drills him in that rigorous control o f em otional expression which is necessary to a man w ho w ould follow a quietistic style o f life.10 W hether one sees the conception o f a personal guardian spirit, a fam ily tutelary, or an im m anent G o d as synoptic fo rm u la­ tions o f the character o f reality or as tem plates for produ cin g reality seem s largely arbitrary, a m atter o f which asp ect, the model o f or m odel for, one w ants for the m om ent to bring into focus. The concrete sym bols involved - one or another m ythological figure m aterializing in rhe w ilderness, rhe skull o f the deceased household head hanging censori­ ously in the rafters, or a disem bodied “ voice in rhe stilln ess” soundlessly ch anting enigm atic classical poetry - point in either direction. They both express the w o rld ’s clim ate and shape ir. They shape it by inducing in the w orshipper a certain distinctive set o f d isp osition s (tenden­ cies, cap acities, propensities, skills, habits, liabilities, pronenesses) which lend a chro­ nic ch aracter ro the flow o f his activity and the quality o f his experience. A disposition describes not an activity or an occurrence but a probability o f an activity being perform ed or an occurrence occurring in certain circum ­ stances: “ When a cow is said to be a rum inant, or a m an is said to be a cigarette-sm oker, it is not being said that the cow is rum inating now or that rhe m an is sm okin g a cigarette now. T o be a rum inant is to tend to rum inate from rime to tim e, and ro be a cigarette-sm oker is

62

CLIFFORD GEERTZ

to be in the h abit o f sm okin g cigarettes.” 11 Sim ilarly, to be p ious is not to be perform ing som ething we w ould call an act o f piety, but to be liable to perform such acts. So, to o, with the Plains Indian ’s bravura, the M an u s’ com pu n ctiousn ess, or the Jav an e se ’s quietism , which, in their con texts, form the substance o f piety. The virtue o f this sort o f view o f w hat are usually called “ m ental traits” or, if the C artesian ism is unavow ed, “ psychological forces” (both unobjectionable enough term s in them selves) is th at it gets them out o f any dim and inaccessible realm o f private sensation into that sam e well-lit w orld o f ob servables in which reside the brittleness o f g lass, the inflam ­ m ability o f p ap er, an d , ro return to rhe m eta­ phor, the d am pn ess o f England. So far a s religious activities are concerned (and learning a myth by heart is as m uch a religious activity as detaching one’s finger at rhe knuckle), tw o som ew hat different sorts o f disposition are induced by them: m oo d s and m otivations. A m otivation is a persisting tendency, a chronic inclination to perform certain sorts o f acts and experience certain sorts o f feeling in certain sorts o f situ ation s, the “ so rts” being com m only very heterogenous and rather illdefined classes in all three c a s e s .. . . As a m otive, “ flam boyant co u rage” consists in such enduring propensities as to fast in the w ilderness, to con duct solitary raids on enemy cam p s, and to thrill to the thought o f counting coup. “ M oral circum spection ” con sists in such ingrained tendencies as to honor onerous prom ises, to con fess secret sins in the face o f severe public d isap p ro v al, and to feel guilty when vague an d generalized accu sation s are m ade at seances. And “ d isp assio n ate tranquil* ity” con sists in such persistent inclinations as to m aintain on e’ s poise com e hell or high w ater, to experience distaste in the presence of even m oderate em otional d isp lays, and to indulge in contentless contem plations o f fea­ tureless objects. M otives are thus neither acts (that is, intentional behaviors) nor feelings, but liabilities to perform p articular classes o f act or have particular classes o f feeling. And when we say that a m an is religious, that is, m oti­ vated by religion, this is at least part - though only part - o f w hat we mean.

Another part o f w hat we mean is that he has, when properly stim ulated, a susceptibility to fall into certain m oo d s, m oods we som e­ tim es lum p together under such covering term s a s “ reverential,” “ solem n ,” or “ w orsh ip fu l.” Such generalized rubrics actually conceal, how ever, the enorm ous em pirical variousness o f the disposition s involved, an d, in fact, tend to assim ilate them to the unusually grave tone o f m ost o f our ow n religious life. The m oods that sacred sym bols induce, at d if­ ferent tim es and in different places, range from exultation to m elancholy, from self-confidence to self-pity, from an incorrigible playfulness to a bland listlessness - to say nothing o f the erogenous pow er o f so m any o f the w orld’s myths and rituals. N o m ore than there is a single sort o f m otivation one can call piety is there a single sort o f m ood one can call w orshipful. The m ajor difference between m oods and m otivations is that where the latter are, so to speak, vectorial qualities, the form er are merely scalar. M otives have a directional cast, they describe a certain overall course, gravitate tow ard certain, usually tem porary, con sum ­ m ations. Hut m oo ds vary only a s to intensity: they g o nowhere. They spring from certain circum stances but they are responsive to no ends. Like fogs, they just settle and lift; like scents, suffuse and evaporate. When present they are totalistic: if one is sad everything and everybody seem s dreary; if one is gay, every­ thing and everybody seem s splendid. T h u s, though a m an can be vain, brave, w illful, and independent at the sam e tim e, he can ’t very well be playful and listless, or exultant and m elancholy, at the sam e tim e.12 Further, where m otives persist for m ore or less extended periods o f time, m oods merely recur with greater or lesser frequency, com ing and goin g for w hat are often quite un fathom able reasons. But perhaps the m ost im portan t difference, so far as we are concerned, between m oo ds and m otivations is that m otivations are “ m ade m eaningful” with reference to the ends tow ard which they are conceived to conduce, w hereas m oods are “ m ade m ean in gful” with reference to the conditions from which they are co n ­ ceived to spring. We interpret m otives in terms o f their con sum m ation s, but we interpret

RELIGION AS A CULTURAL SYSTEM

m oods in term s o f their sources. We say that a person is industrious because he w ishes to succeed; we say that a person is w orried because he is con scious o f the hanging threat o f nuclear holocaust. And this is no less the case when the interpretations are ultim ate. C harity becom es Christian charity when it is enclosed in a conception o f G o d ’s p urposes; optim ism is Christian optim ism when it is grounded in a p articular conception o f G o d ’s nature. The assiduity o f the N av a h o finds its rationale in a belief that, since “ reality” operates m echanically, it is coercible; their chronic fearfulness finds its ration ale in a co n ­ viction that, how ever “ reality” operates, it is both enorm ously pow erful and terribly d a n g e ro u s.13

. . . by formulating conceptions o f a general order o f existence a n d . . . T h at the sym bols or sym bol system s which induce and define disp osition s we set o ff as religious and those which place those d isp o si­ tions in a cosm ic fram ew ork are the sam e sym bols ought to occasion no surprise. For w hat else do we mean by saying that a particu­ lar m ood o f aw e is religious and not secular, except that it springs from entertaining a con ­ ception o f all-pervading vitality like m ana and not from a visit to the G ran d C an yon ? O r that a particular case o f asceticism is an exam ple o f a religious m otivation, except that it is directed tow ard the achievem ent o f an unconditioned end like nirvana and not a conditioned one like w eight-reduction? If sacred sym bols did not at one and the sam e time induce disp osition s in hum an beings and form ulate, however obliquely, inarticulately, or unsystem atically, general ideas o f order, then the em pirical dif­ ferentia o f religious activity or religious exp eri­ ence w ould not exist. A m an can indeed be said to be “ religious” ab o u t go lf, but not merely if he pursues it with p assion and plays it on Sundays: he m ust also see it as sym bolic o f som e transcendent truths. And rhe pubes­ cent boy gazing soulfully into the eyes o f the pubescent girl in a W illiam Steig cartoon and m urm uring “ There is som ething about you, Ethel, which gives me a sort o f religious feelin g,” is, like m ost adolescen ts, confused. W hat any p articular religion affirm s ab o u t the

63

fundam ental nature o f reality may be obscure, shallow , o r, all too often, perverse; but it m ust, if it is not to con sist o f the m ere collection o f received practices and conventional sentim ents we usually refer to a s m oralism , affirm som e­ thing. If one were to essay a m inim al definition o f religion today, it w ould perhaps not be T y lo r’s fam ou s “ belief in spiritual bein gs,” to which Cioody, w earied o f theoretical subtle­ ties, has lately urged us to return, but rather w hat Salvad or de M ad a ria g a has called “ the relatively m odest dogm a that G od is not m a d .” 14 U sually, o f course, religions affirm very m uch m ore than this: we believe, as Jam es rem arked, all that we can and w ould believe everything if we only c o u ld .15 The thing we seem least able to tolerate is a threat to our pow ers o f conception, a suggestion rhat our ability to create, g rasp , and use sym bols may fail us, for were this to h appen, we w ould be m ore helpless, as I have already pointed out, than the beavers. The extrem e generality, dif­ fuseness, and variability o f m an ’s innate (that is, genetically program m ed) response cap aci­ ties m eans that w ithout the assistance o f cultural patterns he w ould be functionally incom plete, not merely a talented ape who h ad, like som e underprivileged child, unfortu­ nately been prevented from realizing his full potentialities, but a kind o f form less m onster wirh neither sense o f direction nor pow er o f self-control, a ch aos o f sp asm od ic im pulses and vague em otions. M an depends upon sym bols and sym bol system s with a depen­ dence so great as to be decisive for his creatural viability an d, as a result, his sensitivity to even the rem otest indication that they may prove unable to cope with one or another asp ect o f experience raises within him the gravest sort o f anxiety. . . . There are at least three points where ch aos - a tum ult o f events which lack not just inter­ pretation s but interfnetability - threatens to break in upon m an: at the lim its o f his analytic capacities, at the limits o f his pow ers o f endur­ ance, and at the lim its o f his m oral insight. Bafflem ent, suffering, and a sense o f intract­ able ethical p a rad o x are all, if they becom e intense enough or are sustained long enough, radical challenges to the proposition that life

64

CLIFFORD GEERTZ

is com prehensible and that we can , by takin g thought, orient ourselves effectively w ithin it - challenges with which any religion, how ever “ prim itive,” which hopes to persist m ust attem pt som eh ow to cope. O f the three issues, it is the first which has been least investigated by m odern social an th rop ologists (though EvansPritchard’s classic d iscu ssion o f why gran aries fall on som e A zande and not on others, is a notable exception ). . . . 16 Any chronic failure o f on e’s exp lan atory a p p ara tu s, the com plex o f received culture patterns (com m on sense, science, philosophical speculation , myth) one has for m appin g the em pirical w orld, to explain things which cry ou t for exp lan ation tends to lead to a deep disquiet - a tendency rather m ore w idespread and a disquiet rather deeper than we have som etim es supposed since the pseudoscience view o f religious belief w as, quite rightfully, deposed. After all, even that high priest o f heroic atheism , Lord R ussell, once rem arked that although the problem of the existence o f G o d had never bothered him, the am biguity o f certain m athem atical ax io m s had threatened to unhinge his m ind. And Ein­ stein’s profoun d dissatisfaction with quantum m echanics w as based on a - surely religious inability to believe that, as he put it, G o d plays dice with the universe. But this quest for lucidity and the rush of m etaphysical anxiety that occurs when em piri­ cal phenom ena threaten to rem ain intransigently op aq u e is found on m uch hum bler intellectual levels. C ertainly, I w as struck in my ow n w ork, m uch m ore than I had a t all expected to be, by the degree to which my m ore anim istically inclined in form an ts behaved like true T ylorean s. They seem ed to be co n ­ stantly using their beliefs to “ e x p la in ” phe­ nom ena: or, m ore accurately, to convince them seives that the phenom ena were exp lain ­ able within the accepted schem e o f things, for they com m only had only a m inim al attach ­ ment to the p articu lar soul possession , em o ­ tional disequilibrium , ta b o o infringem ent, or bew itchm ent hypothesis they advanced and were all too ready to aban d on it for som e other, in the sam e genre, which struck them as m ore plausible given the facts o f the case. W hat they were not ready to d o w as aban d on

it for no other hypothesis at all; to leave events to them selves. And w hat is m ore, they adop ted this nervous cognitive stance with respect to phenom ena which had no im m ediate practical bearing on their own lives, or for that m atter on anyon e’s. When a peculiarly sh aped, rather large to a d ­ stool grew up in a carpen ter’s house in the short space o f a few days (or, som e said, a few hours), people cam e from m iles around to see it, and everyone had som e sort o f explanation - som e anim ist, som e an im atist, som e not quite either - for it. Yet it w ould be hard to argue that the to ad sto o l had any social value in R adcliffe-B row n ’ s sense, or w as connected in any w ay with anything which did and for which it could have been standing proxy, like the A ndam an cicada. . . . 17 N o r is this to argue that it is only, or even m ainly, sudden eruptions o f extraordin ary events which engender in m an the disquieting sense that his cognitive resources m ay prove unavailing or that this intuition ap p ears only in its acute form . M ore com m only it is a per­ sistent, constantly re-experienced difficulty in g rasp in g certain asp ects o f nature, self, and society, in bringing certain elusive phenom ena within the sphere o f culturally form ulatable fact, which renders m an chronically uneasy and tow ard which a m ore equable flow o f diagn ostic sym bols is consequently directed. It is w hat lies beyond a relatively fixed frontier o f accredited know ledge that, loom ing a s a con stan t backgroun d to rhe daily round o f practical life, sets ordinary hum an experience in a perm anent co n text o f m etaphysical co n ­ cern and raises the dim , back-of-the-m ind suspicions that one m ay be adrift in an absu rd w orld: Another subject which is matter for this char­ acteristic intellectual enquiry [among the Iatmul| is the nature of ripples and waves on the surface of water. It is said secretly that men, pigs, trees, grass - all the objects in the world - are only patterns of waves. . . . On one occasion I took some Iatmul natives down to the coast and found one o f them sitting by himself gazing with rapt attention at the sea. It was a windless day, but a slow swell was breaking on the beach. Among the totemic ancestors o f his clan he counted a personified

RELIGION AS A CULTURAL SYSTEM

slit gong who had floated down the river to the sea and who was believed to cause the waves. He was gazing at the waves which were heaving and breaking when no wind was blowing, demonstrating the truth of his clan myth.1* The second experiential challenge in w hose face the m eaningfulness o f a p articular pattern o f life threatens to dissolve into a ch aos o f thingless nam es an d nam eless things - the problem o f suffering - has been rather m ore investigated, or at least described, mainly because o f the great am oun t o f attention given in w orks on tribal religion to w hat are perhaps its tw o m ain loci: illness and m ourning. Yet for all the fascinated interest in the em otional au ra that surrou n ds these extrem e situations, there h as been, w ith a few exception s such as Lienh ardt’ s recent discussion o f D inka divin­ ing, little conceptual advan ce over the sort o f crude confidence-type theory set forth by M alin ow ski: viz., that religion helps one to endure “ situ ation s o f em otional stress” by “ open [ing| up escap es from such situations and such im passes as offer no em pirical way out except by ritual and belief into the dom ain o f the su p ern atu ral.” 1* The inadequacy o f this “ theology o f o p tim ism ,” as N adel rather dryly called it, is, o f co urse, rad ical.20 O ver its career religion has p robably disturbed men as much a s it has cheered them ; forced them into a head-on, unblinking con fron tation o f the fact th at they are born to trouble as often as it has enabled them to avoid such a con fron tation by projecting them into sort o f infantile fairy-tale w orld s where - M alin ow ski again - “ hope cannot fail nor desire deceive.” 21 With the p o s­ sible exception o f C h ristian Science, there are few if any religious trad ition s, “ g re a t” or “ little,” in which the proposition that life hurts is not strenuously affirm ed, and in som e it is virtually glorified: She w as an old [Ba-lla] woman o f a family with a long genealogy. Leza, “ the Besetting One” , stretched out his hand against the family. He slew her mother and father while she w as yet a child, and in the course of years all connected with her perished. She said to herself, “ Surely 1 shall keep those who sit on my thighs.” But no, even they, the children of

65

her children, were taken from her . . . Then cam e into her heart a desperate resolution to find Ciod and to ask the meaning of it a l l . . . So she began to travel, going through country after country, always with the thought in her mind; “ 1 shall come to where the earth ends and there I shall find a road to Ciod and I shall ask him: ‘What have I done to thee that thou afflictest me in this m anner?’ ” She never found where the earth ends, but though disap­ pointed she did not give up her search, and as she passed through the different countries they asked her, “ What have you come for, old w om an?” And the answer w'ould be, “ I am seeking Leza.” “ Seeking Leza! For w hat?” “ My brothers, you ask me! Here in the nations is there one who suffers as 1 have suffered?” And they would ask again, “ How have you suffered?” “ In this way. 1 am alone. As you see me, a solitary old woman; that is how I a m !” And they answered, “ Yes, we see. That is how you are! Bereaved o f friends and husband? In what do you differ from others? The Besetting-One sits on the back o f every one o f us and we cannot shake him o ff.” She never obtained her desire; she died o f a broken heart.22 A s a religious problem , the problem o f suf­ fering is, parad o x ically , not how to avoid suf­ fering but how to suffer, how to m ake of physical pain, personal lo ss, w ordly defeat, or the helpless contem plation o f oth ers’ agony som ething bearable, su p p o rtab le - som ething, a s we say, sufferable. It w as in this effort that the B a-lla w om an - perh aps necessarily, per­ h aps not - failed an d, literally not know ing how to feel ab o u t w hat had happened to her, how to suffer, perished in con fusion and despair. W here the m ore intellective aspects o f w hat W eber called the Problem o f M eaning are a m atter affirm ing the ultim ate explicability o f experience, the m ore affective aspects are a m atter o f affirm ing its ultim ate sufferable­ ness. A s religion on one side an ch ors the pow er o f our sym bolic resources for form u lating a n a ­ lytic ideas in an authoritative conception o f the overall shape o f reality, so on another side it an ch ors the pow er o f o u r, a lso sym bolic, resources for expressin g em otion s - m oods, sentim ents, p assio n s, affection s, feelings - in a sim ilar conception o f its pervasive tenor, its

66

CLIFFORD GEERTZ

inherent rone an d tem per. For those able to em brace them , an d for so long a s they are able to em brace them , religious sym bols provide a cosm ic guaran tee not only for their ability to com prehend the w orld, but also , com preh end­ ing it, to give a precision to their feeling, a definition to their em otion s which enables them, m orosely o r joyfully, grim ly or ca v a ­ lierly, to endure it. C on sider in this light the well-known N av ah o curing rites usually referred to as “ sin gs.” 21 A sing - the N av ah o have about sixty different on es for different p u rp o ses, but virtually all o f them are dedicated to rem oving som e sort o f physical or m ental illness - is a kind o f religious psych odram a in which there are three m ain acto rs: the “ sin ger” or curer, the patient, an d , as a kind o f antiphonal chorus, the p atien t’s fam ily and friends. The structure o f all the sings, the d ram a’s plot, is quite sim ilar. There are three m ain acts: a puri­ fication o f the patient and audience; a state­ ment, by m eans o f repetitive chants and ritual m an ipulation s, o f the wish to restore w ell­ being (“ h arm o n y” ) in the patient; an identifi­ cation o f the patient with the H oly People and his consequent “ cu re.” The purification rites involve forced sw eating, induced vom iting, and so on, to expel the sickness from the patient physically. The chants, which are num ­ berless, con sist m ainly o f sim ple optative phrases (“ m ay the patient be w ell,” “ 1 am getting better all o v er,” etc.). And, finally, the identification o f the patient with the Holy People, and thus with cosm ic order generally, is accom plished through the agency o f a sand painting depicting the H oly People in one or another ap p ro p riate mythic setting. The singer places the patient on the painting, touching the feet, hands, knees, shoulders, breast, back, and head o f the divine figures and then the co rre­ spon din g parts o f the patient, perform ing thus w hat is essentially a bodily identification o f the hum an and the divine.24 This is the clim ax o f rhe sing: the w hole curing process m ay be likened, Reichard says, to a spiritual osm osis in which the illness in m an and the pow er o f the deity penetrate the cerem onial m em brane in both direction s, the form er being n eutral­ ized by the latter. Sickness seeps ou t in the sw eat, vom it, and other purification rites;

health seeps in as the N av a h o patient touches, through the m edium o f the singer, the sacred sand painting. C learly, the sym bolism o f the sing focuses upon the problem o f hum an su f­ fering and attem pts to cope with it by placing it in a m eaningful con text, providing a m ode o f action through which it can be expressed, being expressed, un derstood, and being under­ sto o d , endured. T he sustaining effect o f the sing (and since the com m onest disease is tuber­ culosis, it can in m ost cases be only sustaining) rests ultim ately on its ability to give the stricken person a vocabulary in term s o f which to g rasp the nature o f his distress an d relate it to the wider w orld. Like a calvary, a recitation o f Buddha’s em ergence from his father’s palace, or a perform ance o f O edipus Tyrannos in other religious traditions, a sing is m ainly co n ­ cerned with the presentation o f a specific and concrete im age o f truly hum an, and so endur­ able, suffering pow erful enough to resist the challenge o f em otional m eaninglessness raised by the existence o f intense and unrem ovable brute pain. The problem o f suffering p asses easily into the problem o f evil, for if suffering is severe enough it usually, though not alw ays, seem s m orally undeserved as w ell, at least to the su f­ ferer. But they are not, how ever, exactly the sam e thing - a fact 1 think W eber, too influ­ enced by the biases o f a m onotheistic tradition in which, a s the various aspects o f hum an experience m ust be conceived to proceed from a single, voluntaristic source, m an ’s pain reflects directly on G o d ’s goo dn ess, did not fully recognize in his generalization o f the dilem m as o f C h ristian theodicy F’astw ard. For where the problem o f suffering is concerned with threats to our ability to put our “ undis­ ciplined sq u ad s o f em o tion ” into som e sort o f soldierly order, the problem o f evil is co n ­ cerned with threats to ou r ability to m ake sound m oral judgm ents. W hat is involved in the problem o f evil is not the adequacy o f our sym bolic resources to govern our affective life, but the adequacy o f those resources to provide a w orkable set o f ethical criteria, norm ative guides to govern ou r action. The vexation here is the g ap between things as they are and as they ought to be if our conceptions o f right and w rong m ake sense, the gap between w hat

RELIGION AS A CULTURAL SYSTEM

we deem various individuals deserve and what we see thar they get - a phenom enon sum m ed up in rhar profound quatrain : The rain falls on the just And on the unjust fella; Hut mainly upon the just, Because the unjust has the just’s umbrella.

i...] N o r is it necessary to be theologically selfcon scious to be religiously soph isticated. The concern with intractable ethical p arad o x , the disquieting sense that on e’s m oral insight is inadequate to on e’s m oral experience, is as alive on the level o f so-called prim itive religion as it is on that o f the so-called civilized. The set o f notions ab o u t “ division in the w orld ” that Lienhardt describes for the D inka is a useful case in point.25 L.ike so m any peoples, the D inka believe that the sky, where “ D ivin­ ity” is located, and earth, where m an dw ells, were at one time con tiguous, the sky lying just above the earth and being connected to it by a rope, so that men could m ove at will between the tw o realm s. There w as no death and the first m an and w om an were perm itted but a single grain o f millet a d ay, which w as all that they at that time required. O ne d ay, the w om an - o f course - decided, ou t o f greed, to plant m ore than the perm itted grain o f millet, and in her avid haste and industry accidentally struck Divinity with the handle o f the hoe. O ffended, he severed the rope, w ithdrew into the distan t sky o f tod ay, and left m an to labor for his fo od , to suffer sickness and death, and to experience separation from the source o f his being, his C reator. Yet the m eaning o f this strangely fam iliar story to the D inka is, as indeed is G enesis to Jew s and C h ristian s, not hom iletic but descriptive: Those [Dinka] who have commented on these stories have sometimes made it clear that their sympathies lie with M an in his plight, and draw attention to the smallness o f the fault for which Divinity withdrew the benefits of his closeness. The image o f striking Divinity with a hoe . . . often evokes a certain amusement, alm ost as though the story were indulgently being treated as too childish to explain the consequences attributed to the event. Hut it is clear that the point o f the story o f Divinity’s

67

withdrawal from men is not to suggest an improving moral judgment on human behav­ iour. It is to represent a total situation know'n to the Dinka today. Men now are - as the first man and woman then became - active, selfassertive, inquiring, acquisitive. Yet they are also subject to suffering and death, ineffective, ignorant and poor. Life is insecure; human calculations often prove erroneous, and men must often learn by experience that the con­ sequences of their acts are quite other than they may have anticipated or consider equita­ ble. Divinity’s withdrawal from M an as the result of a comparatively trifling offence, by human standards, presents the contrast between equitable human judgments and the action o f the Power which are held ultimately to control what happens in Dinka life. . . . To the Dinka, the moral order is ultimately con­ stituted according to principles which often elude men, which experience and tradition in part reveal, and which human action cannot change. . . . The myth o f Divinity’s withdrawal then reflects the facts o f existence as they are known. The Dinka are in a universe which is largely beyond their control, and where events may contradict the most reasonable human expectations.2* T h u s the problem o f evil, or perhaps one should say the problem about evil, is in essence the sam e sort o f problem o f or ab o u t baffle­ ment and the problem o f or ab o u t suffering. The strange opacity o f certain em pirical events, the dum b senselessness o f intense or in exora­ ble pain , and the enigm atic unaccountability o f g ro ss iniquity all raise the uncom fortable suspicion that perhaps the w orld, and hence m an ’s life in the w orld, has no genuine order at all - no em pirical regularity, no em otional form , no m oral coherence. And the religious response to this suspicion is in each case the sam e: the form ulation, by m eans o f sym bols, o f an im age o f such a genuine order o f the w orld which will account for, and even cele­ brate, the perceived am biguities, puzzles, and p arad o x es in hum an experience. The effort is not to deny the undeniable - that there are unexplained events, that life hurts, or that rain falls upon the just - but to deny that there are inexplicable events, that life is unendurable, and that justice is a m irage. The principles

68

CLIFFORD GEERTZ

which constitute the m oral order m ay indeed often elude m en, as Lienhardt puts it, in the sam e w ay a s fully satisfactory exp lan atio n s of an om alou s events or effective form s for the expression o f feeling often elude them. W hat is im portan t, to a religious man at least, is that this elusiveness be accounted for, that it be not the result o f the fact that there are no such principles, exp lan ation s, or form s, that life is absurd and the attem pt to m ake m oral, intel­ lectual, or em otional sense out o f experience is bootless. The D inka can adm it, in fact insist upon, the m oral am biguities an d co n trad ic­ tions o f life a s they live it because these am b i­ guities and co n tradiction s are seen not as ultim ate, but as the “ ra tio n al,” “ n a tu ra l,” “ logical” (one m ay ch oose one’s ow n adjective here, for none o f them is truly adequate) outcom e o f the m oral structure o f reality which the myth o f the w ithdraw n “ D ivinity” depicts, o r as Lienhardt says, “ im ages.” The Problem o f M eaning in each o f its inter­ grad in g asp ects (how these asp ects in fact intergrade in each p articular case, w hat sort o f interplay there is between the sense o f analytic, em otion al, and m oral im potence, seem s to me one o f the ou tstan d in g, and except for W eber untouched, problem s for com p arative research in this w hole field) is a m atter o f affirm ing, or at least recognizing, the inescapability o f ign o­ rance, pain, an d injustice on the hum an plane while sim ultaneously denying th at these irra­ tionalities are ch aracteristic o f the w orld as a w hole. And it is in term s o f religious sym bol­ ism , a sym bolism relating m an’s sphere o f e x is­ tence to a w ider sphere within which it is conceived to rest, that both the affirm ation and the denial are m ad e.27 . . . and clothing those conceptions with such an aura o f factuality th at . . . There arises here, how ever, a m ore profound question: how is it that this denial com es to be believed? H ow is it that the religious man m oves from a troubled perception o f exp eri­ enced disorder to a m ore or less settled convic­ tion o f fundam ental order? Ju st w hat does “ belief” mean in a religious co n text? O f all the problem s surrou ndin g attem pts ro conduct an th ropological an alysis o f religion this is the one that has perh aps been m ost troublesom e

and therefore the m ost often a v o id e d .. . . But the problem will not go aw ay, it is not “ m erely” psychological (nothing social is), and no an th ropological theory o f religion which fails to attack it is w orthy o f the nam e. We have been trying to stage H am let w ithout the Prince quite long enough. It seem s to me th at it is best to begin any ap p roach to this issue with frank recognition that religious belief involves not a Baconian induction from everyday experience - for then we should all be agn o stics - but rather a prior acceptance o f authority which tran sform s that experience. The existence o f bafflem ent, pain, and m oral p a ra d o x - o f The Problem of M eaning - is one o f rhe things that drives men tow ard belief in g o d s, devils, spirits, totem ic principles, or the spiritual efficacy o f can n ib al­ ism (an enfolding sense o f beauty or a dazzling perception o f pow er are others), but it is not the basis upon which those beliefs rest, but rather their m ost im portant field o f application: We point to the state o f the world as illustra­ tive of doctrine, but never as evidence for it. So Belsen illustrates a world o f original sin, but original sin is not an hypothesis to account for happenings like Belsen. We justify a par­ ticular religious belief by showing its place in the total religious conception; we justify a reli­ gious belief as a whole by referring to author­ ity. We accept authority because we discover it at some point in the world at which we worship, at which we accept the lordship of something not ourselves. We do not worship authority, but we accept authority as defining the worshipful. So someone may discover the possibility o f worship in the life o f the Reformed Churches and accept the Bible as authoritative; or in the Roman Church and accept papal authority.2* T his is, o f co u rse, a Christian statem ent o f the m atter; but it is not to be despised on that account. In tribal religions authority lies in the persuasive pow er o f traditional im agery; in m ystical ones in the apodictic force o f super­ sensible experience; in ch arism atic ones in the hypnotic attraction o f an extraordin ary per­ sonality. But the priority o f the acceptance o f an authoritative criterion in religious m atters

RELIGION AS A CULTURAL SYSTEM

over the revelation which is conceived to flow from that acceptance is not less com plete than in scriptural or hieratic ones. The basic axiom underlying w hat we m ay perh aps call “ the reli­ giou s perspective” is everywhere the sam e: he w ho w ould know m ust first believe. But to speak o f “ the religious perspective” is, by im plication, to sp eak o f one perspective a m o n g others. A perspective is a m ode o f seeing, in that extended sense o f “ see” in which it m eans “ d iscern ,” “ ap p reh en d ,” “ under­ sta n d ,” or “ g r a sp .” It is a p articular w ay o f look in g at life, a p articular m anner o f con stru­ ing the w orld, as when we sp eak o f an h istori­ cal perspective, a scientific perspective, an aesthetic perspective, a com m on-sense per­ spective, or even the bizarre perspective em bodied in dream s and in h allucinations. The question then com es dow n to , first, w hat is “ the religious perspective” generically co n sid ­ ered, as differentiated from other perspectives; and second, how d o men com e to ad o p t it. If we place the religious perspective again st the backgroun d o f three o f the other m ajor perspectives in term s o f which men construe the w orld - the com m on-sensical, the scien­ tific, and the aesthetic - its special character em erges m ore sharply. W hat distinguishes com m on sense as a m ode o f “ seein g” is, as Schutz has pointed ou t, a sim ple acceptance o f the w orld, its ob jects, and its processes as being just w hat they seem to be - w hat is som etim es called naive realism - and the p rag ­ m atic m otive, the wish to act upon that w orld so as to bend it to on e’ s practical p urposes, to m aster it, or so far as that proves im possible, to ad ju st to it.2y T he w orld o f everyday life, itself, o f course, a cultural p roduct, for it is fram ed in term s o f the sym bolic conceptions o f “ stubborn fa ct” handed dow n from gen era­ tion to generation, is the established scene and given object o f our action s. Like M t. Everest it is just there, and the thing to d o with it, if one feels the need to d o anything with it at all, is to clim b it. In the scientific perspective it is precisely this givenness which d isa p p ea rs.30 D eliberate d ou bt and system atic inquiry, the suspen sion o f the p ragm atic m otive in favor o f disinterested ob servation , the attem pt to analyze the w orld in term s o f form al concepts w hose relationship to the inform al con cep­

69

tion s o f com m on sense becom e increasingly prob lem atic - there are the h allm arks o f the attem pt to g rasp the w orld scientifically. And a s for the aesthetic perspective, which under the rubric o f “ the aesthetic attitu d e” has been perh aps m ost exquisitely exam in ed, it involves a different sort o f suspen sion o f naive realism and practical interest, in that instead o f q u es­ tioning the credentials o f everyday experience, one merely ignores that experience in favor o f an eager dw elling upon ap pearan ces, an engrossm ent in surfaces, an absorption in things, as we say, “ in them selves” : “ The function o f artistic illusion is not ‘ m akebelieve’ . . . but the very op p o site, disen gage­ m ent from belief - the con tem plation o f sensory qualities w ithout their usual m eanings o f ‘here’s that ch air’ , ‘th at’s my tele­ phone’ . . . etc. The know ledge that w hat is before us h as no practical significance in the w orld is w hat enables us to give attention to its appearan ce as su ch .” 31 And like the com m on sensical and the scientific (or the historical, the philosoph ical, and the artistic), this perspective, this “ w ay o f seein g” is not the produ ct o f som e m ysterious C artesian chem is­ try, but is induced, m ediated, and in fact created by m eans o f curious q u asi objects poem s, d ram as, sculptures, sym phonies which, dissociatin g them selves from the solid w orld o f com m on sense, take on the special sort o f eloquence only sheer ap p earan ces can achieve. T he religious perspective differs from the com m on-sensical in th at, as already pointed ou t, it m oves beyond the realities o f everyday life to w ider ones which correct and com plete them , and its defining concern is not action upon those w ider realities but acceptance o f them , faith in them. It differs from the scien­ tific perspective in that it question s the realities o f everyday life not out o f an institutionalized skepticism which dissolves the w orld’s given­ ness into a swirl o f p rob ab ilistic hypotheses, but in term s o f w hat it takes to be w ider, nonhypothetical truths. R ather than detach ­ m ent, its w atchw ord is com m itm ent; rather than an alysis, encounter. And it differs from art in that instead o f effecting a disengagem ent from the w hole question o f factuality, deliberately m anufacturin g an air o f sem blance

70

CLIFFORD GEERTZ

and illusion, it deepens the concern with fact and seeks to create an aura o f utter actuality. It is this sense o f the “ really real” upon which the religious perspective rests and which the sym bolic activities o f religion as a cultural system are devoted to producing, intensifying, and, so far as possible, rendering inviolable by the discord an t revelations o f secular exp eri­ ence. It is, again , the im buing o f a certain sp e­ cific com plex o f sym bols - o f the m etaphysic they form ulate and the style o f life they recom ­ mend - with a persuasive authority which, from an analytic point o f view, is the essence o f religious action. Which brings us, at length, to ritual. For it is in ritual - that is, consecrated behavior that this conviction that religious conceptions are veridical and that religious directives are sound is som ehow generated. It is in som e sort o f cerem onial form - even if that form be hardly m ore than the recitation o f a m yth, the con sultation o f an oracle, or the decoration o f a grave - that the m oods and m otivations which sacred sym bols induce in men and the general conceptions o f the order o f existence which they form ulate for men meet and rein­ force one another. In a ritual, the w orld as lived and the w orld as im agined, fused under the agency o f a single set o f sym bolic form s, turn ou t to be the sam e w orld, produ cin g thus that idiosyncratic tran sform ation in on e’s sense o f reality to which San tayan a refers in my epigraph. W hatever role divine interven­ tion m ay or m ay not play in the creation o f faith - and it is not the business o f the scientist to pronounce upon such m atters one w ay or the other - it is, prim arily at least, ou t o f the context o f concrete acts o f religious o b ser­ vance that religious conviction em erges on the hum an plane. H ow ever, though any religious r it u a l,. . . in­ volves this sym bolic fusion o f ethos and w orld view, it is m ainly certain m ore elaborate and usually m ore public on es, ones in which a broad range o f m oods and m otivations on the one hand and o f m etaphysical conceptions on the other are caugh t up, which shape the sp iri­ tual con sciousn ess o f a people. F^mploying a useful term introduced by Singer, we may call these full-blow n cerem onies “ cultural perform an ces” and note that they represent

not only the point at which the disposition al and conceptual aspects o f religious life co n ­ verge for the believer, but also the poin t at which the interaction between them can be m ost readily exam in ed by the detached observer. . . ,n O f co urse, all cultural perform ances are not religious perform ances, and the line between those that are and artistic, or even political, ones is often not so easy to draw in practice, for, like social form s, sym bolic form s can serve m ultiple purposes. Rut the point is that, p a r a ­ phrasing slightly, Indians - “ and perh aps all peoples” - seem to think o f their religion “ as encapsulated in these discrete perform ances which they [can] exhibit to visitors and to them selves.” 33 T he m ode o f exhibition is however radically different for the tw o sorts o f w itness, a fact seem ingly overlooked by those w ho w ould argue that “ religion is a form o f hum an a r t.” 14 W here for “ v isitors” religious perform ances can , in the nature o f the case, only be presentations o f a particular religious perspective, and thus aesthetically appreciated or scientifically dissected, for participants they are in addition enactm ents, m aterializations, realizations o f it - not only m odels o f w hat they believe, but a lso m odels for the believing o f it. In these plastic d ram as men attain their faith as they portray it. As a case in point, let me take a sp ectacu ­ larly theatrical cultural perform ance from Bali - that in which a terrible witch called R angda engages in a ritual com b at with an endearing m onster called B aron g.35. . . the dram a co n ­ sists o f a m asked dance in which the witch depicted as a w asted old w idow , prostitute, and eater o f infants - com es to spread plague and death upon the land and is opposed by the m onster - depicted as a kind o f cro ss between a clum sy bear, a silly puppy, and a strutting Chinese d ragon . R an g d a, danced by a single m ale, is a hideous figure. H er eyes bulge from her forehead like sw ollen boils. H er teeth becom e rusks curving up over her cheeks and fan gs protrudin g dow n over her chin. H er yel­ low ed hair falls dow n aroun d her in a m atted tangle. H er breasts are dry and pendulous dugs edged with hair, between which hang, like so many sau sag es, strings o f colored entrails. Her long red tongue is a stream o f fire. And as she

RELIGION AS A CULTURAL SYSTEM

dances she splays her dead-w hite hands, from which protrude ten-inch claw like fingernails, ou t in front o f her and utters unnerving shrieks o f m etallic laughter. Baron g, danced by tw o men fore-and-aft in vaudeville horse fashion, is another m atter. H is shaggy sheepdog co at is hung with gold and m ica orn am en ts that glitter in the half-light. He is adorned with flow ers, sash es, feathers, m irrors, and a com ical beard m ade from hum an hair. And though a dem on too, his eyes a lso p op and he sn ap s his fanged jaw s with seemly fierceness when faced with R an gda or other affron ts to his dignity; the cluster o f tinkling bells which hang from his absurdly arching tail som ehow contrives to take m ost o f the edge o ff his fearfulness. If R an gda is a satanic im age, B aron g is a farcical one, and their clash is a clash (an inconclusive one) between the m alignant and the ludicrous. This odd counterpoint o f im placable malice and low com edy pervades the w hole perfor­ mance. R an gd a, clutching her m agical white cloth, m oves aroun d in a slow stagger, now pausin g im m obile in thought or uncertainty, now lurching suddenly forw ard. The m om ent o f her entry (one sees those terrible long-nailed hands first as she em erges through the split gatew ay at the to p o f a short flight o f stone stairs) is one o f terrific tension when it seem s, to a “ visitor” at least, that everyone is about to break and run in panic. She herself seems insane with fear and hatred as she scream s deprecations at B aron g am id the wild clanging o f the gam elan. She m ay in fact go am ok. I have m yself seen R an g d as hurl them selves headlong into the gam elan or run frantically ab o u t in total con fusion , being subdued and reoriented only by the com bined force o f a half-dozen spectators; and one hears many tales o f am ok R an gd as holding a w hole village in terror for hours and o f im personators becom ing perm anently deranged by their expe­ riences. But Barong, though he is charged with the sam e m ana-like sacred pow er (sakti in Bali­ nese) as R an gd a, and his im personators are also entranced, seem s to have very great diffi­ culty in being serious. H e frolics with his retinue o f dem ons (w ho ad d to the gaiety by indelicate p ranks o f their ow n), lies dow n on a m etallaphone while it is being played or

71

beats on a drum with his legs, m oves in one direction in his front h alf and another in his rear or bends his segm ented body inro foolish con tortion s, brushes flies from his body or sniffs aro m as in the air, and generally prances ab o u t in p aroxysm s o f narcissistic vanity. The con trast is not absolute, for R an gda is som e­ tim es m om entarily com ic as when she pretends to polish the m irrors on B aron g’s co at, and B aron g becom es rather m ore serious after R an gda ap p ears, nervously clacking his jaw s at her and ultim ately attackin g her directly. N o r are the hum orous and the horrible alw ays kept rigidly separated, as in that strange scene in one section o f the cycle in which several m inor w itches (disciples o f R angda) toss the corpse o f a stillborn child arou n d to the wild am usem ent o f the audience; or another, no less stran ge, in which the sight o f a pregnant w om an alternating hysterically between tears and laughter while being knocked ab o u t by a g ro u p o f gravediggers, seem s for som e reason excruciatingly funny. The twin them es o f horror and hilarity find their purest expression in the tw o p rotagon ists and their endless, inde­ cisive struggle for dom inance, but they are woven with deliberate intricacy through the whole texture o f the dram a. They - or rather the relations between them - are w hat it is about. It is unnecessary to attem pt a th orough ­ goin g description o f a R an g d a -B a ro n g p erfor­ m ance here. Such perform ances vary widely in detail, con sist o f several not to o closely inte­ grated p arts, and in any case are so com plex in structure a s to defy easy sum m ary. For our purposes, the m ain point to be stressed is that the dram a is, for the Balinese, not merely a spectacle to be w atched but a ritual to be enacted. There is no aesthetic distance here separatin g actors from audience and placing the depicted events in an unenterable w orld o f illusion, and by the time a full-scale R an g d a B aron g encounter has been concluded a m ajo r­ ity, often nearly all, o f the m em bers o f the gro u p spon sorin g it will have becom e caught up in it not just im aginatively but b o d ily .. . . As a perform ance, the dram a is like a high m ass, not like a presentation o f Murder in the Cathe­ dral: it is a draw ing near, not a standin g back. [...|

72

CLIFFORD GEERTZ

Ir is in the direct encounter with the tw o figures in the co n text o f the actual perform ance that the villager com es to know them a s, so far a s he is concerned, genuine realities. They are, then, not represen tation s o f anything, hut presences. And when the villagers go into trance they becom e - nadi - them selves part o f the realm in which those presences exist. T o ask , as I once did, a m an w ho has been R an gd a whether he thinks she is real is to leave oneself open to the suspicion o f idiocy. The acceptance o f authority that underlies the religious perspective that the ritual e m b o d ­ ies thus flow s from the enactm ent o f the ritual itself. By inducing a set o f m oods and m otiva­ tions - an ethos - and defining an im age o f cosm ic order - a w orld view - by m eans o f a single set o f sym bols, the perform ance m akes the m odel for and m odel o f asp ects o f religious belief mere tran spo sition s o f one another. R an gd a evokes fear (as well as h atred, disgust, cruelty, horror, an d , though I have not been able to treat the sexual asp ects o f the p erfor­ m ance here, lust); but she also depicts it: The fascination which the figure o f the Witch holds for the Balinese imagination can only be explained when it is recognized that the Witch is not only a fear inspiring figure, but that she is Fear. Fler hands with their long menacing finger-nails do not clutch and claw at her victims, although children who play at being witches do curl their hands in such gestures. But the Witch herself spreads her arms with palms out and her finger flexed backw ard, in the gesture the Balinese call kapar , a term which they apply to the sudden startled reac­ tion o f a man who falls from a tree. . . . Only when we see the Witch as herself afraid, as well as frightening, is it possible to explain her appeal, and the pathos which surrounds her as she dances, hairy, forbidding, tusked and alone, giving her occasional high eerie laugh.36 And on his side B aron g not only induces laughter, he incarnates the Balinese version of the com ic spirit - a distinctive com bin ation o f playfulness, exh ibition ism , and extrav agan t love o f elegance, which, alo n g with fear, is perhaps the dom in an t m otive in their life. The constantly recurring struggle o f R an gd a and

B aron g to an inevitable draw is thus - for the believing Balinese - both the form ulation o f a general religious conception and the au th o rita­ tive experience which justifies, even com pels, its acceptance. . . . that the moods and motivations seem

uniquely realistic But no one, not even a sain t, lives in the w orld religious sym bols form ulate all o f the time, and the m ajority o f men live in it only at m om ents. The everyday w orld o f com m onsense objects and practical acts is, as Schutz say s, the param ou n t reality in hum an experi­ ence - param ou n t in the sense that it is the w orld in which we are m ost solidly rooted, w hose inherent actuality we can hardly ques­ tion (how ever m uch we m ay question certain portions o f it), and from w hose pressures and requirem ents we can least escape.37 A m an, even large grou p s o f m en, m ay be aesthetically insensitive, religiously unconcerned, and unequipped to pursue form al scientific an aly­ sis, but he cannot be com pletely lacking in com m on sense and survive. The dispositions which religious rituals induce thus have their m ost im portan t im pact - from a hum an point o f view - outside the boun daries o f the ritual itself as they reflect back to color the individual’s conception o f the established w orld o f bare fact. The peculiar tone that m arks the Plains vision quest, the M an u s co n ­ fession, or the Ja v an ese m ystical exercise per­ vades areas o f the life o f these peoples far beyond the im m ediately religious, im pressing upon them a distinctive style in the sense both o f a dom in an t m ood and a ch aracteristic m ovem ent. The interw eaving o f the m alignant and the com ic, which the R an g d a-B aro n g com bat depicts, an im ates a very w ide range o f everyday Balinese behavior, much of which, like the ritual itself, has an air o f candid fear narrow ly contained by obsessive playful­ ness. R eligion is sociologically interesting not because, as vulgar positivism w ould have it, it describes the social order (which, in so far a s it does, it does not only very obliquely but very incom pletely), but because, like environ­ m ent, political pow er, w ealth, jural obligation , personal affection , and a sense o f beauty, it sh apes it.

RELIGION AS A CULTURAL SYSTEM

The m ovem ent back and forth between the religious perspective and the com m on-sense perspective is actually one o f the m ore obvious em pirical occurrences on the social scene, th ough, again , one o f the m ost neglected by social an th rop ologists, virtually all o f w hom have seen it happen coun tless tim es. Religious belief has usually been presented a s a h om oge­ n eous characteristic o f an individual, like his place o f residence, his occu p ation al role, his kinship p osition , and so on. But religious belief in the m idst o f ritual, w here it engulfs the total p erson , tran sportin g him, so far as he is con ­ cerned, into another m ode o f existence, and religious belief as the pale, rem em bered reflec­ tion o f that experience in the m idst o f everyday life are not precisely the sam e thing, and the failure to realize this h as led to som e con fu­ sion, m ost especially in connection with the so-called prim itive-m entality problem . M uch o f the difficulty between Levy-Bruhl and M alin ow ski on the nature o f “ native th ough t,” for exam ple, arises from a lack o f full recogni­ tion o f this distinction; for where the French ph ilosoph er w as concerned with the view o f reality sav ages ad op ted when takin g a specifi­ cally religious perspective, the Polish-English ethnographer w as concerned with that which they adopted when raking a strictly com m onsense one.™ Both perhaps vaguely sensed that they were not talkin g ab o u t exactly the sam e thing, but where they w ent astray w as in failing to give a specific accoun ting o f the w ay in which these tw o form s o f “ th ough t” - or, a s I w ould rather say , these tw o m odes o f sym bolic form ulation s - interacted, so that w here Levy-Bruhl’s sav ag es tended to live, despite his p ostludial disclaim ers, in a w orld com p osed entirely o f m ystical encounters, M alin o w sk i’s tended to live, despite his stress on rhe functional im portan ce o f religion, in a w orld com p osed entirely o f practical actions. They becam e reductionists (an idealist is as m uch o f a reductionist a s a m aterialist) in spite o f them selves because they failed to see m an a s m oving m ore or less easily, and very fre­ quently, between radically co n trastin g w ays o f look in g at the w orld, w ays which are not co n ­ tin uous with one another but separated by cultural g ap s acro ss which K ierkegaardian leaps m ust be m ade in both directions.

73

[ ...] H avin g ritually “ lept” (the im age is perhaps a bit to o athletic for the actual facts - “ slip p e d ” m ight be m ore accurate) into rhe fram ew ork o f m eaning which religious conceptions define, and the ritual ended, returned again to the com m on-sense w orld, a m an is - unless, as som etim es happens, the experience fails to reg­ ister - changed. And a s he is ch anged, so also is the com m on-sense w orld, for it is now seen as but the partial form o f a w ider reality which corrects and com pletes it. But this correction and com pletion is not, a s som e students o f “ com parative religion” w ould have it, everywhere the sam e in content. The nature o f the bias religion gives to o rd i­ nary life varies with the religion involved, with the particular disposition s induced in the believer by the specific conceptions o f cosm ic order he has com e to accept. O n the level of the “ g re a t” religions, organic distinctiveness is usually recognized, at tim es insisted upon to the point o f zealotry. But even at its sim plest folk and tribal levels - where the individuality o f religious traditions has so often been d is­ solved into such desiccated types a s “ an im ism ,” “ an im atism ,” “ totem ism ,” “ sh am an ism ,” “ an cestor w orsh ip ,” and all the other insipid categories by m eans o f which ethnographers o f religion devitalize their data - the idiosyn­ cratic ch aracter o f how various grou p s o f men behave because o f w hat they believe they have experienced is clear. A tranquil Ja v a n e se w ould be no m ore at hom e in guilt-ridden M an us than an activist C row w ould be in p assion less Ja v a . And for all the w itches and ritual clow ns in the w orld, R an gda and B aron g are not gen­ eralized but thoroughly singular figurations o f fear and gaiety. W hat men believe is a s various as w hat they are - a proposition that holds with equal force when it is inverted. It is this particularity o f the im pact o f reli­ gio u s system s upon social system s (and upon personality system s) which renders general assessm en ts o f the value o f religion in either m oral or functional term s im possible. The sorts o f m oo ds and m otivations which ch arac­ terize a m an w ho has ju st com e from an Aztec hum an sacrifice are rather different from those o f one w ho has just put o ff his K achina m ask. F>en within the sam e society, w hat one

74

CLIFFORD GEERTZ

“ learn s” a b o u t the essential pattern o f life from a sorcery rite and from a com m ensal meal will have rather diverse effects on social and psych ological functioning. O ne o f the m ain m eth odological problem s in writing about religion scientifically is to put aside at once the tone o f the village atheist and th at o f the village preacher, as well a s their m ore sophisticated equivalents, so that the social and psych ological im plications o f p articular religious beliefs can emerge in a clear and neutral light. And when that is don e, overall questions ab o u t whether religion is “ g o o d ” or “ b a d ,” “ fu n ction al” or “ d ysfun ction al,” “ ego strengthening” or “ anxiety p rod u cin g,” d isa p ­ pear like the chim eras they are, and one is left with particular evaluation s, assessm en ts, and diagn oses in particular cases. . . . [••■I The an th ropological study o f religion is therefore a tw o-stage operation : first, an a n a l­ ysis o f the system o f m eanings em bodied in the sym bols which m ake up the religion proper, and, second, the relating o f these system s ro social-structural and psychological processes. M y dissatisfaction with so much o f co n tem p o­ rary social an th ropological w ork in religion is

NOTES 1 2

3

4 5

6

S. L an ger, Philosophy in a New Key , 4th ed. (C am brid ge, M ass., 1960). K. Burke, The Philosophy o f Literary> Form (Baton R ouge, La.: L ouisian a State University Press, 1 9 4 1 ), p. 9. T he reverse m istake, especially com m on am on g neo-K an tian s such as C assirer, o f takin g sym bols to be identical with, or “ constitutive o f ,” their referents is equally pernicious. [C f. E. C assirer,

The

not that it concerns itself with the second stage, but that it neglects the first, and in so doing takes for granted w hat m ost needs to be elucidated. T o discu ss the role o f an cestor w orship in regulating political succession , o f sacrificial feasts in defining kinship o b lig a ­ tions, o f spirit w orship in scheduling agricul­ tural practices, o f divination in reinforcing social con trol, or o f initiation rites in p rop el­ ling personality m aturation , are in no sense unim portant endeavors, and I am not recom ­ m ending they be aban doned for the kind o f jejune cabalism into which sym bolic analysis o f exotic faiths can so easily fall. But to attem pt them with but the m ost general, com m onsense view o f w hat ancestor w orship, anim al sacrifice, spirit w orship, divination, or initia­ tion rites are a s religious patterns seem s to me not particularly prom ising. Only when we have a theoretical an alysis o f sym bolic action com parable in sophistication to w hat we now have for social and psychological action , will we be able to cope effectively with those aspects o f social and psychological life in which religion (or art, or science, or ideology) plays a determ inant role.

Philosophy

o f Symbolic

Forms

(N ew H aven: 1 9 5 3 -7 ), 3 vols.] “ O ne can point to the m oon with on e’s finger,” som e, p rob ab ly w ell-invented, Zen M aster is su p p osed to have said , “ but to take on e’s finger for the m oon is to be a fo o l.” JC C raik , The Nature o f Explanation (C am bridge, 1952). K. Lorenz, King Solom on's Ring (London , 1952).

7 8 9 10 11 12 13

14

15

K. von Erisch, “ D ialects in the Lan guage o f the Bees,” Scientific American , A ugust 1962. C raik , N ature o f Explanation. R. H . Low ie, Primitive Religion (N ew Y ork, 1924). R. E. Fortune, M anus Religion (Philadelphia, 1935). C. G eertz, The Religion o f Ja v a (G lencoe, III., 1960). G . Ryle, The Concept o f Mind (London and N ew Y o rk , 1949). Ibid., p. 99. C. K luckhohn, “ T he Philosophy of the N av a h o In dian s,” in Ideological D iffer­ ences and World O rder , ed. F. S. C. N o rth ro p (N ew H aven, 1 9 4 9 ), pp. 3 5 6 -8 4 . J. Cioody, “ R eligion and R itual: The Definition P rob lem ,” British Jou rn al o f Psychology { 1961): 143-64. W. Ja m e s, The Principles o f Psychology , 2 vols. (N ew Y ork, 1904).

RELIGION AS A CULTURAL SYSTEM

16

17

18 19

20

21 22

E. Evans-Pritchard, Witchcraft, Oracles and M agic Among the Azande (O xford , 1 937). A. R. R adcliffe-Brow n, Structure and Function in Primitive Society (Glencoe, 111., 1952). G. Bateson, Naven , 2nd ed. (Stanford, 1958). G. Lienhardt, Divinity and Experience (O x fo rd , 1961), p. 151 ff; B. M alin ow ski, Magic, Science and Religion (Boston, 1 9 4 8 ), p. 67. S. F N ad el, “ M alin ow ski on M agic and R eligion ,” in Man and Culture , ed. R. Firth (London , 1957), pp. 1 8 9 -2 0 8 . M alin ow ski, M agic , Science and Religion , p. 67. C W. Smith and A. M . D ale, The lla-

28

29

30 31 32

Speaking Peoples o f Northern Rhodesia

23

24 25 26 27

(London , 1 9 2 0 ), p. 197ff.; quoted in P. R adin , Primitive Man as a Philosopher (N ew Y ork, 1957), pp. 1 0 0 -1 . C . Kluckhohn and D. Leighton. The N avaho (C am bridge, M a ss., 1946); G. R eichard, N avaho Religion , 2 vols. (N ew Y ork, 1950). R eichard, N avaho Religion. Lienhardt, Divinity and Experience , pp. 2 8 -5 5 . Ibid. T h is is not , how ever, to say that everyone in every society does this; for as the im m ortal Don M arq u is once rem arked, you d o n ’t have to have a soul unless you really w ant one. The oft-heard generaliza­ tion that religion is a hum an universal em bodies a confusion between the p ro b ­ ably true (though on present evidence unprovable) proposition that there is no hum an society in which cultural patterns that we can, under the present definition or one like it, call religious are totally lacking, and the surely untrue proposition that all men in all societies are, in any m eaningful sense o f the term , religious. But if the an th ropological study o f reli­ giou s com m itm ent is underdeveloped, the an th ropological study o f religious

33

34 35

36 37 38

75

noncom m itm ent is nonexistent. The an th ropology o f religion will have com e o f age when som e m ore subtle M alin ow ski w rites a book called “ Belief and Unbelief (or even ‘ Faith and H ypocrisy’ ) in a Savage Society.” A. M acIntyre, “ The Logical Statu s of R eligious Belief,” in M etaphysical Beliefs, ed. A. M acIntyre (London , 1957), pp. 167—211. A. Schutz, The Problem o f Social Reality , vol. 1 o f Collected Papers (The H ague, 1962). Ibid. S. Langer, Feeling and Form (N ew Y ork, 1953), p. 49. M . Singer, “ The C ultural Pattern o f Indian C ivilization ,” Far Eastern Q uar­ terly 15 (1 955): 2 3 -6 . M . Singer, “ The G reat T radition in a M etropolitan Center: M a d ra s,” in Tradi­ tional India , ed. M . Singer (Philadelphia, 1958), pp. 1 4 0 -8 2 . R. Firth, Elements o f Social Organization (London and N ew Y ork, 1951), p. 250. T he R an g d a-B aro n g com plex has been extensively described and analyzed by a series o f unusually gifted ethnographers an d I shall m ake no attem pt to present it here in m ore than schem atic form . |See, for exam ple, J. Belo, Bali: Rangda and Barong (N ew Y ork, 1 9 4 9 );J. Belo, Trance in Bali (N ew Y ork, 1960); B. D eZoete an d W. Spies, Dance and D ram a in Bali (L o n do n , 1938); G. Bateson and M . M ead , Balinese Character (N ew Y ork, 1942); M . C ovarru b ias, The Island o f Bali (N ew Y ork, 1937).] M uch o f my interpretation o f the com plex rests on personal observations m ade in Bali during 1 9 5 7 -8 . G. Bateson and M . M ead , Balinese Character , p. 36. Schutz, The Problem o f Social Reality , p. 2 26ff. M alin ow ski, Magic, Science and Religion ; L. Levy-Bruhl, How Natives Think (N ew Y o rk , 1926).

Skeptical Rejoinders

In trod u ction “ R eligion” is not a given in the world, nor the referent for a discrete species, bur the product o f a particular kind o f discursive stance rhat distinguishes som e things (e.g., “ religion’') from others (e.g., “ eco nom y ” ), and unites some things with others within the category o f “ religion.” A book like this cannot help doing the sam e, but the essays here do try to problematize the categories w hereof they speak. Although many anthropologists in the first half o f the 20th century gave a g o o d deal o f atten­ tion to definitional problem s concerning religion, ritual, myth, primitive society, and so on, a working skepticism at the resulting objectifications has also alw ays been a part o f the anthropological tradition. C ontem porary skepticism is not the product o f som e radical “ p o stm o d e rn ” break so much as ongoing investment in an intellec­ tual practice intrinsic to the anthropological craft. What is perhaps new is the spe­ cifically Nietzschean strain that sees intellectual discourses as inevitably derivative of, and hence contributory to, larger schemes o f power. In any case, this section includes five strong skeptical responses to the various fram ew orks established or evoked in the previous one. They are, in turn, the product o f diverse standpoints. It is worth draw ing attention to three additional powerful w orks o f skeptical “ deconstruction” written within the anthropological mainstream. The first is that o f Franz Steiner who, in a po sth um o us publication o f a course o f lectures, entitled Taboo (1 9 5 6 ; Adler and Fardon, eds., 1999), systematically dem ol­ ishes the model o f “ primitive religion” that w as built up around the term, querying not only the objectification o f “ t a b o o ” as a kind o f religion but its status a s a general concept, thereby throwing the entire com parative method into question. Steiner showed how authors like Robertson, Smith, and Frazer used the category o f ta b o o to claim that so-called primitive religions lacked adequate concepts o f morality. For more positive accounts o f ta b o o s, see van G ennep (1904), D ou glas (1966), Gell (1979) and Lam bek (1992).

78

INTRODUCTION

The implications o f Steiner’s critique have been expanded by philosopher Alasdair MacIntyre as part o f his program to show the inadequacy o f an objectivist (or what he calls encyclopaedist) approach to moral inquiry (1990). For MacIntyre, the approach o f Frazer et al. w as sym ptom atic o f the incoherence o f liberal modernist culture with respect to its own (and therefore our) moral concerns. The breakdown o f traditional social orders like those o f precolonial H aw aii or the European Christian M iddle Ages simultaneously produces false abstractions like “ ta b o o ” or “ obligation ” and confuses the means by which to actually learn to live a moral life. Steiner, MacIntyre, and D ou glas speak from what is ultimately a conservative perspective that values hierarchy, for the holism it provides. The holism comprises social cohesion embedded in political hierarchy that is in turn encom passed by religion. M acIntyre’s implicit nostalgia for the age o f Aquinas can be contrasted to M a r x ’s forthright critique o f feudalism (1961 [1 8 8 7 |), or the depiction o f medieval Christianity provided by Asad (1993). Levi-Strauss accomplished the sam e feat for and in Totemism (1963e [1962]) that Steiner did for and in Taboo. His argument there is that w hat is relevant about socalled totemic practices (such as anim als serving as sym bols for social groups) is less the intrinsic qualities o f individual species o f anim als than the system o f a n a lo ­ gies they establish. A given Native American group is less interested in the intrinsic properties o f “ b ears” or “ deer” than the contrast that can be set up by means o f them. Totem ism is thus to be understood as a system o f classification or naming rather than as a religious phase or institution. Finally, Evans-Pritchard pursued the question o f rationality through both a su s­ tained critical demolition o f the theories o f the evolutionists and a positive e x p o si­ tion o f the logic o f ostensibly irrational beliefs in an African society. The critiques first appeared in a pair o f powerful essays published in Egypt in 1933 and 1934 and were later incorporated in revised form in Theories o f Primitive Religion (1965). They formed the prolegomena to his masterly Witchcraft, O racles , and M agic am ong the Azande (19 3 7 ; 1976 for a condensed version), which set a new standard for thinking ethnographically a b ou t rationality that few other w orks have met. Evans-Pritchard showed, am ong other things, why it w as reasonable for Azande to continue to believe in witchcraft as an explanation for misfortune and, indeed, how difficult it is to escape from its terms; how witchcraft does not preclude, but rather complements, a materialist explanation for events; and how the practices associated with witchcraft provide a guide for orderly action. His account o f the workings o f a relatively “ clo sed ” system o f thought w as not unlike philosopher T h o m a s Kuhn’s model o f successive paradigm s in the history o f western science (1970), and influ­ enced philosophers (Polanyi 1 9 7 0 ; cf. Gluckm an 1970). What Evans-Pritchard perhaps failed to fully acknow ledge w as how the witchcraft system operated in the interests o f the local system o f pow er (M cL eod 1972). For reasons o f copyright, the excerpt by Wittgenstein that w as found in the first edition o f this Reader has had to be cut back. Readers are encouraged to turn to the original for the full text.

5

Remarks on Frazer's Golden Bough Ludwig Wittgenstein

An Austrian w ho spent much of his professional life in England, Ludwig W ittgenstein (1889-1951) was among the most profound philosophers of the 20th century. His brilliance is evident in these incisive remarks on Frazer. In fact, they skew er not only Frazer but the entire intellectualist tradition that sees ritual and religious practices as prim arily a form of explanation, and a simple-minded and mistaken one. But W ittgenstein's remarks also effectively counter any reductionist account "M an," he says simply, "is a cerem oni­ ous anim al." W ittgenstein anticipates many developm ents in anthropology, which generally proceeded w ithout benefit of these remarks. These insights are evident in works such as those of Lienhardt (chapter 24), Fernandez, or Tam biah (chapter 25), who see ritual shaping experience; in those like Langer (chapter 10) and O rtner (chapter 12), w ho take seriously the symbolic dim en­ sion of culture and religion; and in those w ho explore the nature and effects of language as something at once simpler

and more complex than a direct repre­ sentation of reality (as variously, Becker, Bloch, and Rappaport, chapters 18, 32 and 33 below). But perhaps W ittgen­ stein would go further in sweeping aw ay all preconceptions and refusing to generalize. These aphorisms speak for themselves and resist explication. As W ittgenstein fam ously said, "W hereof w e cannot speak, thereo f w e must be silent." Nevertheless, readers may find the extended comm entary in "Rem arks on W ittgenstein and Ritual" (in Needham 1985) useful. Rodney Needham has been the earliest and strongest champion of W ittgenstein in anthropology (1972, 1975) as w ell as authoring a number of other intriguing essays on religion, notably "Characteristics of Religion" (in Needham 1981). W ittgenstein is also invoked in an intriguing analysis of mor­ tuary ritual (M etcalf 1987), w hile a recent overview of anthropology whose Wittgensteinian influence is evident in its title is The Cerem onial A n im a l (James 2004).

From Ludwig Wittgenstein, Remarks on Frazer's Golden Bought trans. A. C. Miles, rev. and ed. Rush Rhees (Nottingham: Brynmill Press Ltd., 1979 [c.1931]), pp. le-7e. Footnotes deleted.

80

LUDW IG WITTGENSTEIN

Frazer’s accoun t o f the m agical and religious notions o f men is unsatisfactory: it m akes these notions ap p ear as mistakes. W as A ugustine m istaken, then, when he called on Ciod on every page o f the Con­

fessions? Well - one m ight say - if he w as nor m is­ taken, then the Buddhist holy-m an, or som e other, w hose religion expresses quite different n otions, surely w as. But none of them w as m aking a m istake except where he w as putting forw ard a theory. Even the idea o f trying to explain the p ra c­ tice - say the killing o f the priest-king - seem s to me w rong-headed. All that Frazer d oes is to m ake this practice plausible to people w ho think as he does. It is very queer that all these practices are finally presented, so to sp eak , as stupid action s. But it never does becom e plausible that people d o all this ou t o f sheer stupidity. When he exp lain s ro us, for exam ple, that the king m ust be killed in his prim e because, accordin g to the notions o f the sav ages, his soul w ould not be kept fresh otherw ise, we can only say: where that practice and these views go together, rhe practice does not spring from the view , but both o f them are there. It m ay h appen , as it often does to d ay , that som eone will give up a practice when he has seen that som eth ing on which it depended is an error. But this h appens only in cases where you can m ake a m an change his way o f doin g things sim ply by calling his attention to his error. T h is is not how it is in connexion with the religious practices o f a people; and w hat we have here is not an error. Frazer say s it is very difficult to discover the error in m agic and this is why it p ersists for so long - because, for exam ple, a cerem ony which is su p p osed to bring rain is sure to ap p ear effective soon er or later. But then it is queer that people do not notice sooner rhat it d oes rain sooner or later anyw ay. I think one reason why the attem pt to find an exp lan ation is w rong is that we have only to put together in the right w ay w hat we know , w ithout ad d in g anything, and the satisfaction we are trying to get from the explanation com es o f itself.

And here the explan ation is not w hat satis­ fies us anyw ay. When Frazer begins by telling the story o f the K ing o f the W ood at N em i, he does this in a tone which show s that som ething strange and terrible is happening here. And that is rhe answ er to the question “ why is this h ap p en in g?” : Because it is terrible. In other w ords, whar strikes us in this course o f events a s terrible, im pressive, horrible, tragic, & c ., anything but trivial an d insignificant, that is w h at gave birth to them. We can only describe and say, hum an life is like that. C o m p ared with the im pression that w hat is described here m akes on us, the explanation is to o uncertain. Flvery explan ation is an hypothesis. But for som eone broken up by love an explan atory hypothesis w on ’t help m uch. - It will not bring peace. T he crush o f thoughts that do not get out because they all try to push forw ard and are w edged in rhe door. Put that accoun t o f the King o f the W ood at N em i together with the phrase “ the m ajesty o f d eath ” , and you see that they are one. T he life o f the priest-king show s w hat is m eant by that phrase. If som eone is gripped by the m ajesty o f death, then through such a life he can give expression to it. - O f course this is not an explanation : it puts one sym bol in place o f another. O r one cerem ony in place o f another. A religious sym bol does not rest on any

opinion. And error belongs only with opinion. O ne w ould like to say: T h is is w hat to ok place here; laugh, if you can. T he religious action s or the religious life o f the priest-king are not different in kind from any genuinely religious action today, say a con fession o f sins. T h is also can be “ e xp lain ed ” (m ade clear) and cannot be explained. Burning in effigy. K issing the picture o f a loved one. T h is is obviously not based on a belief that it will have a definite effect on the object which the picture represents. It aim s at som e satisfaction an d it achieves it. O r rather, it d oes not aim at anything; we act in this w ay and then feel satisfied.

REMARKS ON FRAZER'S GOLDEN BOUGH

We could alm ost say, m an is a cerem onious anim al. T h is is partly false, partly nonsensical, but there is a lso som ething in it. In other w ords, one m ight begin a book on an th ropology in this w ay: W hen we w atch the life and behaviour o f men all over the earth we see that ap art from w hat we might call anim al activities, takin g food &.C., & c \,

81

men also carry out action s that bear a peculiar character and m ight be called ritualistic. But then it is nonsense if we g o on to say that the characteristic feature o f these actions is that they spring from w rong ideas ab o u t the physics o f things.

I...]

6

Religion, Totemism and Symbolism W. E. H. Stanner

W. E. H. Stanner (1905-81) taught at the Australian National University and was a leading figure in the study of A borigi­ nal religion and the initiation of land rights. W hat W ittgenstein does to the intellectualists, Stanner does to Durkheim . But Stanner's standpoint is very different. He writes as someone w ith extensive ethnographic experience among the people of whom he speaks. Australian Aboriginal culture has played a particularly significant role in the developm ent of anthropological thought. Having suffered trem endous violence and massive appropriation at the hands of w hite Australian settlers, and deploying a fairly simple material technology, the Aborigines w ere thought in the 19th century to be among the simplest societies on the planet. It was for this reason that Durkheim looked to Australia for the fundam ental aspects of religion. However, anthropologists w ho subsequently w orked w ith still viable

communities have been universally impressed by the depth and complexity of Aboriginal thought as well as by the m anner in w hich their philosophy was able to encompass their social structure and w ay of life, providing a uniquely holistic and com prehensive world. Stanner provides a w onderful rejoin­ der to the ethnocentric work that preceded him. He offers a lucid criticism of Frazer and the evolutionists and an equally strong critique of the model of elem entary religion as totemism, which Durkheim had constructed from w ritten material on Australia. Here Stanner's cri­ tique is not on empirical grounds alone. He faults Durkheim and others for ignor­ ing the grounds of Aboriginal ontology and hence for seeing religion as deriva­ tive of the social. There are numerous other critiques o f an overly sociological approach to religion and, as noted, Levi-Strauss's short book, Totemism , is perhaps the most comprehensive

From W. E. H. Stanner, “ Religion, Totemism and Symbolism,” in White Man G ot N o Dream ing, Essays 1938-73 (Canberra: Australian National University Press, 1979 [ 1962])* pp. 111-25. © W. E. H. Stanner, 1979. Abridged.

RELIGION, TOTEMISM AND SYMBOLISM

deconstruction of th at false subject - "a mentity, not an entity," as Stanner pithily puts it (1979 [1962]: 126). But Stanner's critique is simultaneously a superb introduction to the positive features of Aboriginal religion - indeed, he provides a forceful and one might even say "religiously musical" account of it. His observation concerning "the absence from [Aboriginal] religious thought and practice of any lifecompensatory them es" might be com­ pared to Sahlins's analysis of western cosmology (1996). Stanner's essay provides an example of one of the many critiques launched by the generations of solid fieldw orkers against the reified abstractions like animism and totem ism that w ere central to the preceding generations of "arm ­ chair" thinkers w ho speculated about

[...] M odern an th rop ologists criticize their nine­ teenth-century predecessors for m any faults, and the force o f the criticism s is reflected in m any aban don ed p osition s. N o one now su p ­ poses that natural science is free o f m etaphysic, or assum es that if the m ethod o f natural science is follow ed then hum an facts under study have to be treated as if they to o are non-hum an. N o t m any now hold to the idea that historical explan ation is, intrinsically, the only right explan ation o f co-existent hum an facts. The gran diose sw eep o f theory, the “ ineffable p aro ch ialism ” (Low ie) o f definitions and assu m p tion s, and the pejorative use o f ra tio ­ nalistic con cepts, are much less com m on. But a cardin al fault - the invincible ignorance ab o u t A boriginal religion - has not been criti­ cized sufficiently. 1 have already m entioned the superficial causes - unim aginative observation , preconception and bigotry. But there were also a lack o f detachm ent tow ard the intellectual foun dation s o f the new discipline; a certain sycophancy tow ard fashion able vogues in science (especially biology) and philosophy;

83

the religious lives of peoples whom they had never encountered. The im portance of sympathetic, firsthand understanding is clear from Stanner's account. Indeed, although he stays at the general and continental level, Stanner's appreciation is a to u r de force of anthropological interpretation. Readers may appreciate that this essay of 1962 presages several subsequent developments in anthropo­ logical thought, including even an allu­ sion to initiation as a form o f discipline. Readers interested in Stanner should turn to the original text and supplem ent it w ith Stanner (1963). A nother pow er­ ful and influential essay on ontology w ritten about the same tim e as Stanner and equally from the sympathetic per­ spective of a master ethnographer is that by Hallowed (1960) on the Ojibwa of Canada.

and an unw illingness to risk being thought guilty o f w riting devotional exercises. A desire to free study from em otion becam e a fear o f em otion, and a drying ou t o f true sensibility. The lack o f sensibility certainly strikes forcibly a m odern reader o f their w orks. Even Sir Ja m es Frazer, in unconscious p arap h rase o f Ridley h alf a century earlier - and, o f course, o f G oethe long before - cam e very near to saying so in addressin g the Ernest Renan Society: “ . . . w ithout tenderness, w ithout poetry, one cannot understand m an or his creatio n s.” (N eedless to say, he w as not thinking o f the A borigines: as late as 1937 he still referred to them as “ these sav ag e s” .) [...] The sch olars w ho w rote as if religion did not exist, and the men o f religion w ho w orked as if it could not exist, am on g so b arb aro u s a people, were not in even distan t collusion with authorities w ho had no m otive o f credibility to think or act beyond a vague and ill-policed policy o f protection. But the nineteenth-century pages echo with a tacit co m p act to underwrite a negativism to which ugliness and ignobility w ere added in every decade dow n to the 1930s.

84

W . E. H. STANNER

Under the A ustralian law o f charity, sym pathy for racial injustice had com e to vary inversely with nearness to the evil. There m ust have been a score o f cau ses co n ­ tributing to A boriginal m isery. But from the early nineteenth century, none had a m ore dev­ astatin g effect than the pervasive doctrine o f A boriginal w orthlessness. T h at depended to a decisive extent on the specific blindness to which I have referred. Yet, a s R. M . Berndr has rightly said : traditional A boriginal religion w as “ a living faith, som ething quite in sep ara­ ble from the pattern o f everyday life and th ough t” . The connection w as so intim ate that “ there is no sharp dem arcation between secular and sacred life” . In the w ords o f Father F^. A. W orm s, A boriginal religion “ penetrates all facets o f life an d has little to fear from distin c­ tions which are both ab stract and disunitive and which we, with ou r philosophical ed u ca­ tion, often m ak e ” . I shall sketch a s briefly as possible the positive character o f rhe religion as we now understand it. (1) T he A borigines thought the w orld full o f signs ro m en; they tran sform ed the signs into assuran ces o f m ystical providence; and they conceived life’s design as fixed by a founding d ram a. (2) At its best the religion put a high w orth on the hum an person, both as flesh and as spirit. (3) It m agnified rhe value of life by m akin g its conservation and renewal into a cult. (4) It ackn ow ledged the m aterial dom ain a s being under spiritual authority. (5) R eligious practice included a discipline to subdue egotistical m an to a sacred, continuing purpose. (6) R eligious belief expressed a phi­ losophy o f assenr to life’s term s. (7) T he m ajor cults inculcated a sense o f m ystery through the use o f sym bolism s pointing to ultim ate or m etaphysical realities which were know n by their signs. Each p roposition rests on wellesrablished facts, which have often been recorded and left, uninterpreted, as mere “ cu sto m ” . T o deny them w hat seem their plain im plications is now unjustifiable. I shall discu ss each o f the m ain statem ents in turn. It will then be clear, I hope, that w hat prevented their earlier recognition were the unexam ined assu m p tion s o f sch olars w ho were either not interested in religion a s such, or had too

n arrow a conception o f it, or m isunderstood their task. 1. The A borigin es’ positive know ledge has been well ap preciated, especially that involved in their techniques o f subsistence an d in their m anipulation o f the segm entary form s o f social organ ization . In those fields o f life they were m asterfully confident. W har m ay be called their religious confidence h as been left rather understated. They lived a s though sure o f their pow er, through ritual observan ces, to sustain their being in a w orld w hich, though grounded on m ystery, had no real problem o f futurity. The n om adic life o f hunting and foraging m ust have had its fair share, perh aps m ore, o f vicis­ situdes. But their religion had a notably strong theurgic com pon en t which expressed itself everywhere in the continent, at least in all the regions ab o u t which we have go o d know ledge, in the conception o f a great founding dram a. T har dram a w as m arked by a clim ax in which everything - including m an, and his w hole condition o f life - cam e to be as it is. Form , style, and function becam e determ inate. C o n ­ sequently, the types o f tension between past, present, and future that characterize so m any system s o f religion were entirely absen t from theirs. T he given condition o f life w as one in which the typical preoccu pations o f m any other religious faiths could have had no func­ tion. A full understanding o f the A boriginal view o f life an d rhe w orld requires a careful study o f the w hole body o f doctrine ab o u t The Dream Tim e (altjira , bugari ), which is the com m on but not universal w ay o f referring to the time o f the foun din g dram a. It has not yet been ap praised at all adequately. But it repre­ sents an im m ense store o f m eanings, variably draw n on by different cult-groups, yet evi­ dently never fully exp lored or used by any o f them because subtle (and prob ab ly im portant) variation s occurred in different regions. The religious tone w as certainly affected. In C ape Y ork U rsula M cC onn el found that m any m yths dealing with the founding dram a had a quality o f “ self-dedication ” ; in the N orthern Territory the quality seem ed to me rather that o f “ sad finality” . T h o se were not wholly su b ­ jective im pression s. O ne w ould expect that, within a continent o f so m any con trasting environm ents, m any qualitative differences

RELIGION, TOTEMISM AND SYMBOLISM

w ould occur. H ad there been a higher rate o f social change in A boriginal life than w as evi­ dently the case, m any elem ents which were sublim inal w ould p rob ab ly have developed. A foun dation existed for a system atic belief in g o d s and for institutions o f priesth ood, prayer, and sacrifice. O nce observers were able to cease identifying religion with theism , a per­ ception o f those foun d ation s drew them on to m any false attributions. T h at error in turn has to be rectified. T he central problem o f study is to stay within the actual evidence but at the sam e time to draw from it the legitim ate reli­ gio u s im plications. W idely, tw o com plem en­ tary em ph ases sto o d out in the doctrine o f The D ream Tim e: the fixation or instituting o f things in an enduring form , and the sim ultane­ o u s endow m ent o f all things - including m an, an d his condition o f life - with their go o d and/or bad properties. T he central m eaning w as clear. M en were to live alw ays under that foun dation . When the myths ab o u t the d ram a o f The D ream T im e are studied with care it becom es clear that the A borigines had taken, indeed had gone far beyond, the longest and m ost difficult step tow ard the form ation o f a truly religious ou tloo k. They had found in the w orld a b o u t them w hat they took to be signs o f intent tow ard m en, and they had transform ed those signs into assurances o f life under m ysti­ cal nurture. Their sym bolic observan ces to w ard the signs, in rites o f several kinds, were in essence acts o f faith tow ard the ground o f th at assurance. It is not yet possible to bring together under th at principle all the ritualized cults o f which w e have heard, but those th at fit within the trilogy suggested m any years a g o by A. P. Elkin - historical rires, initiation rites, and “ in crease” rites {tain, inticbiuma) intended to m aintain and renew the life o f natural species - app eared in som e sense to recapitulate som e feature or aspect o f the founding d ram a. O ne could dou btless speak o f “ im itatio n ” and thus cast all ritual into a m ould o f “ m ag ic ,” but th at really will not do. T he aetiology is o b vi­ ously too profoun d. If the w ord “ religion ” m ean s, a s its etym ology prob ab ly suggests, tw o disp osition s in m an - to ponder on the fo u n d atio n s o f hum an life in history, and to

85

unite or reconcile oneself with the design in corporated in those foun dation s - then the A borigines were a very religious-m inded people. The m otive o f their stron g sense o f religious duty and the purpose o f their rites becom e m ore understandable if approach ed from that view point. So do the intellectual, em otion al, psych ological, and social co m p o ­ nents o f their religious thought and life. If life has a m ystical foun dation , and if its design w as fixed once-for-all, w hat else should rational men d o but m aintain and renew that design? M o st an th ropologists fam iliar with the A b o rig­ ines w ould testify to their apparen t inability to g rasp that life can have any other ration ale as satisfyin g and conclusive as that on which their religion is founded. If one can judge from con tem porary and recent A boriginal life, w hat m ust have fasci­ nated them - it still does - w as the apparen t evidence o f design in the w orld; design in the sense o f pattern, shape, form , structure; given design that seem ed to them to point to intent. It w ould be tedious to list the facts o f that kind o f which they take sharp note, but the proofs th at they alw ays did so are contained in their lan guage categories. It is to those facts, not to the im aginary phenom ena with which H erbert Spencer, T y lo r, and Frazer m ade so much play, that a theorist o f the origins o f A boriginal religion should turn. Pattern, sh ape, form , and structure, occurring in w hat we call “ n atu re” , con stituted for them a w orld o f signs to men. Part o f their religion seem s to be like a return o f equivalent or com pen satory signs to the m ysterious dom ain w hence they cam e. There can not have been m any prim itive rites which so strongly suggested a con scious attem pt by men to bind them selves to the design in things they saw ab o u t them , and to the enduring plan o f life a s they experienced it. 2. The w orth attached to the person w as show n in a striking m anner by the high ritualization o f the life-cycle o f m ales. A lw ays a p articu lar person, or a very sm all grou p o f equivalent persons, w as thus hon oured; and the com m unity, not a clique or set, paid the h onour. Each individual, at his due tim es, w as brough t to the first place in public life. For d ays or w eeks he w as m ade the focus o f e lab o ­ rate efforts o f the im aginative and m aterial

86

W . E. H. STANNER

arts. The effect w as to dignify and in som e sense sanctify each person so honoured. One is im pelled to conclude that the rites had a plain m eaning: man is o f value in him self and for others. T h e relative value o f initiates at such tim es w as the highest that society could contrive for them . The m eaning “ m an has value” w as a lso im plied by the respect for totem s, totem -places, and insignia and em blem s standing for person s; by the restraints again st the use o f nam es, or other exten sion s o f personality such as sh adow s and track s, in a dan gerou s or disrespectful w ay; by the unde­ m onstrative care o f the sick, blind, halt, and m entally afflicted; and by the dutiful obsequies to the body, the spirit/soul/ghost/sh ade, and the social m em ory o f the dead. Such acts, atti­ tudes, and beliefs are deeply inconsonant with a low valuation o f hum an life and personality. One could not rightly say that in them selves they am ounted to a religious view o f m an. But there w as a further fact rhat, added to them , w arran ts such a conclusion. In several parts o f A boriginal A ustralia one met the fundam ental belief that great guardian -spirits (Baiam e, Kunm anggur) existed - whether as ancestral or as self-subsistent beings - to “ look a fte r” living men. Elsew here, lesser spirits did so. The conception thus deepens: m an is o f value in him self and for others, and there are spirits who care. T h at, by any test, is a religious view o f m an. But the generalization m ust be given its true m easure in the light o f certain negative facts. The religious valuation w as qualified by a secular valuation both w ithin, and especially between, clan s and tribes. The w orth o f infants and the very old w as notoriously held o f small account: in desperate circum stances, both were left to die. O n o ccasion s, individuals acted tow ard oth ers with intense cruelty, disregard, and selfishness within sm all kin-groups and, outside - except in respect o f close cogn ates and affines - w ithout restraint other than that induced by fear o f consequences. A lm ost uni­ versally, the valuation o f wom en w as low in respect o f their personal as distinct from their functional w orth. They were usually held in low regard ritually, to o, but not alw ays in all circum stances. Their blood-m akin g and childgiving pow ers were thought both m ysterious and d an gerou s, but there w as nothing elevated

in their sex or m arriage. It m ay be suggested rhat those negative facts were the products o f p ragm atic, egotistic, and politic con dition s, the concom itan ts o f any religious system in practice. A boriginal religion w as not alone in being infiltrated an d, in som e respects, m ade part-prisoner by expediency, pow er, and vested interest. But all that only qualified A boriginal m an ’s dignity. It flaw ed, but did not destroy, the estate into which he cam e in The D ream Tim e. 3. W hat I have called the “ m agnification o f life” w as show n by the intense, one could alm ost say obsessive, preoccupation with the signs, sym bols, m eans, portents, tokens, and evidences o f vitality. The whole religious corpus vibrated with an expressed aspiration for life, abu ndan t life. Vitality, fertility and grow th; the conservation, production, protection, and rescue o f life: them es such as these seem to have been widely im plicit and, in som e notable regional cases, quite explicit. V italistic things obtruded throughout the m yths and rites w ater, blood, fat, hair, excrem ents; the sex organ s, sem en, sexuality in all its ph ases, the quickening in the w om b; child-spirits, m ystical im pregnation and reincarnation; the develop­ ment o f the body from birth to death; the tran ­ sitions o f the hum an spirit from before organic assum ption until after physical dissolu tion ; apparen tly anim ated phenom ena such as green leaves, rain and the season s, lightning, w hirl­ w inds, shootin g stars and the heavenly bodies; or things o f unexplained origin, unusual appearance and gian t size. Poor descriptions o f rites, and bad or over-literal tran slation s o f m yths, have often left such stresses latent or obscure. But the careful studies by W arner, Elkin, and R. M . and C. H. Berndt in Arnhem Land, by E. A. W orm s m ore w idely, and the skilful linguistic w ork by A. Capell and T . G. H . Strehlow , to ch oose a few exam ples only, m ake clear w hat m ust have been com m only the case. The know n evidence suggests that A b o rig­ inal religion w as probably one o f the least m aterial-m inded, and m ost life-m inded, o f any o f which we have know ledge. It m ay not have “ m agnified g o o d n e ss” , as Bacon said o f C h ris­ tianity, but it did m agnify life. 4. T he overrule o f the m aterial dim ension by spiritual authority w as not com plete. By

RELIGION, TOTEMISM AND SYMBOLISM

“ spiritual au th o rity” I m ean the rule o f all invisible potencies, how ever im agined, that were believed to have effects on men’ s lives, effects not possible by unaided m eans in the hands o f ordinary men. We have evidence that the whole o f m ateriality w as not thought to be influenced in that w ay, so that runaw ay d o c­ trines o f anim atism and anim ism are unjusti­ fied. N o one spirit or potency had authority over all the m ateriality th at w as so influenced. N o t all spirits were thought o f a s m an-like; som e were supposed to have quasi-anim al form s, or even to be indescribable. O f those that were m an-like only som e were thought an cestral; others were considered to be “ self­ finding” (self-existent, self-subsistent). But there were m any things in the environm ent that were just things, them selves only and no m ore, w ithout im port, stan din g for nothing. And the authority o f spirits and other poten ­ cies, as understood by the A borigines, w as only vaguely a m oral-ethical authority. T hose reservations having been stated, one really need point only to two well-know n classes o f fact to justify the m ain prop osition , though o f course m any m ore could be cited. The first is the class o f beliefs concerning the im pregna­ tion o f w om en by pre-existing child-spirits that act under their own volition. T he second is the class o f beliefs concerning the depen­ dence o f men on a potential o f life (for exam ple, o f hum ans, an im als, and plants) pre-existing in totem -places. M en could - should - help the child-spirits to d o their w ork, and could indeed, m ust - ritually facilitate the release o f the potential. But they did not create that store and w ithout it were helpless. The manifesta­ tion o f life on a visible, m aterial plane w as thus a spiritual function. So w as the pow er o f hum ans to subsist on that plane. T hose p o stu ­ lates were fundam ental to A boriginal social existence as a form o f being-as-it-is. (The ques­ tion whether we are dealing with “ m ag ic” or “ religion” does not arise in the case o f the first set o f beliefs an d , in the second, concerns only the m ode o f releasing the potential.) 5. The myths contain much o f the “ humanall-too-h um an ” character o f man. A certain im age o f original m an em erges as though with tw o faces, one well draw n , the other less so. T he first face has on it the m arks o f egotism ,

87

alw ays w ayw ard and self-w illed, som etim es w anton: greed, envy, bad faith, an ger, selfish­ ness, pride, disobedience, and the like are com m on them es in the m yths. T o com plete the features o f the other (let us think o f it for the time being as one face, though we will p ro b a ­ bly find m any when the m atter is studied with care) one has to d o tw o things: elicit the con ­ ventions o f understanding within which the m yths were told and heard, and interpret the clim ax o f each myth. Both are dan gerou s pro­ cedures since it is easy to slip beyond the evi­ dence. M an y m yths, one can n ot say all, had a hom iletic effect; perh aps the A borigines drew a m oral lesson from them ; but to all a p p ear­ ances a stron g, explicit religious ethic w as absen t, prob ab ly for the sam e reason s that a religious creed w as absent. Three vital precon­ ditions were m issing - a tradition o f intellec­ tual detachm ent; a class o f interpreters who had the prerogative or duty to codify principle; and a challenge that w ould have forced m orals and beliefs to find anatom ies. All this m ade the m oral aspect o f the religion rather am o rp h o u s, although w hat w as there w as consistent. But a study o f the ritual prac­ tices now suggests a possible need to m odify that rather unfavourable judgem ent. Until recently, to know how to investigate the problem m ore adequately seem ed peculiarly difficult. But it now ap p ears possible to co m pare fruitfully tw o things that did not seem co m parable - the structural an atom ies o f m yths and rites. M any myths reveal a m ount­ ing o f incidents to a crisis or culm ination that exh ibits a cluster o f m eanings with a distinct m oral quality. The initiatory rites all rose to a tense crisis that brought ab o u t, or w as su p ­ posed to bring ab o u t, a physical-m oralspiritual change in the initiates. The tw o types o f crisis ap p ear to have been sym bolic param orph s. In m yth, an im agined crisis w as dealt with by a spoken im agery. In rite, an actual crisis w as dealt with by a gestural-visual im agery. In such cases the m yths, although a sort o f allegorical poesy, m ay have served as the im plicit m oral “ th eory” o f the rites. How far that app roach will stand up to test, and how far the m orph ological likenesses can be traced through the sym bolic system s, remain to be seen. It is to o soon to say certainly that

88

W . E. H. STANNER

the funerary rites contained the sam e sym bolic pattern, or one co m p arab le, or whether all varian ts o f the trilogy studied so form atively by Elkin did so. But even a partial success in dem on stratin g that that w as the case will rein­ force w hat we already know . All the evidence collected since C o llin s’s time [1798] e stab ­ lishes that the rites o f initiation existed a s dis­ ciplines. They both fashioned uncom pleted m an, and transform ed him into a being o f higher w orth. T he m oral and m ystical content o f the rites varied regionally. It m ay have varied to o over time a s one cult replaced or m ixed with another. But the canon o f the rites w as in vari­ able: to subdue refractory, unfinished p erson ­ alities to a p urpose held to be sacred and tim eless. They put on the body, m entality, and social personality o f initiates ineffaceable signs design ating stages in the socialization o f m an. It is a plau sible hypothesis that the ou tw ard signs were thought o f as having inw ard coun ­ terparts; that the rites were held to put on initi­ ates a m oral-spiritual m ark a s well. The crude vehicles o f that purpose - tooth-avulsion, depilation , scarification , circum cision, subin ci­ sion - have been stum bling-blocks o f E u ro­ pean understanding. It is m ost necessary here ro look beyond the sym bol to the sym bolized. But it is also necessary to take m ore accoun t o f the experiential and creative asp ects o f A boriginal religion. T he convention follow ed for so long that the study o f a religion is to be equated with the study o f its beliefs and actions (m yths and rites) is plainly too restrictive. A boriginal religion drew on a hum an e xp eri­ ence o f life, and had a creative purpose in life. The four categories o f experience, belief, action and purpose were co-ordinate. If any is neglected a study m ay be about religion but not o f it. 6. There were no A boriginal ph ilosoph ers and one can thus speak o f “ ph ilosop h y” only m etaphorically. But there is ground for saying that they lived - and therefore thought - by a x io m s, which were “ objective” in that they related to a su p p osed nature o f m an and con ­ dition o f hum an life. M yths presented the ax io m s in an intuitive-contem plative aspect. R ites presented them in a passionist-activist aspect. N o A boriginal put the ax io m s into

w ords but the existence and efficacy o f any­ thing - including intuitional aw aren esses and insights - d o not depend on som eone’ s form al affirm ation o f them in w ords. M yths w ould not be stories, and rites w ould not have an invariant structure, if ax io m s could not subsist by other than form alized m eans. I shall not try to d o m ore than state w hat I believe to have been the principle o f A boriginal philosophy in the m etaphorical sense. 1 prop ose to call it a principle o f assent to the disclosed term s of life. A n th ropologists w ho have w orked with A borigines com m only note that a supposed p ast - the w hole doctrine o f The D ream Tim e - w as said to , and to all appearan ces did, weigh on the present with overm astering authority. But as far a s one can tell, the hum an response to that situation w as not tragic, pes­ sim istic, fatalistic or even quietistic on the one hand, or rebellious and com plain in g on the other. 1 have rem arked elsewhere that the A borigines seem ed either to have stopped short of, or gone beyond, a true quarrel with the term s o f life. They ap peared to assen t to a reality-as-it-is-and-m ust-be. H ence, 1 suggest, three things: the “ h um an-all-too-hum an” quality p ostu lated as true o f men and life’ s condition in The D ream T im e; the constancy o f rhe ritual m otive to m em orialize the culm i­ nating events o f that m ythical tim e; and the absence from religious thought and practice o f any life-com pensatory them es. But wirhin that larger equipoise they evidently sought to m ake the physical and social life-process o f m an a process o f m oral developm ent as well. Any such con struction m ust take into accoun t a num ber o f facts w hich, though not new, are now com ing into better perspective am on g an th rop ologists. There is no dou b t that cul­ tural influences, including religious influences, com ing from beyond the continent (especially but not only to Arnhem Land - and thence, w ho know s how far?) had pow erful effects before European settlem ent. It is also certain that the dynam ic o f developm ent within A ustralia w as higher, and diffused its produ cts m ore w idely, than w as once sup posed. T hirdly, cults recently an d now under study give som e evidence - as yet indirect; but to my mind very suggestive - o f a process o f religious discovery. Con ceivably, all those things were causally

RELIGION, TOTEMISM AND SYMBOLISM

connected. T h at possibility rem ains to be inves­ tigated. But takin g, as far as one can at this stage, a continental view , it is difficult to resist a conclusion that both the religious and the social cultures were in a dynam ic state when E u rop ean s cam e. We shall undoubtedly learn m uch m ore by deeper an alyses o f the surviving regional cults, no less in their steady phases than in the fervour o f their rise and the possible degenerations o f their fall. O ne m ay end with a question . If the philosophy w as one o f assent, why the creative effort o f new cult? 7. In several respects the know n cults suggest a classification with the mysteryreligions. With rem arkable theatrical skill, they used m ystagogy to inculcate an attitude - an archaist attitude - to things o f this w orld. W hether the cultists taught or learnt anything o f m oral or spiritual significance is perhaps open to doubt. T he fact m ay be, as A ristotle said o f the Greek m ysteries, that “ the initiated d o not learn anything so m uch a s feel certain em otion s and are put in a certain fram e o f m in d” . M o st an th rop ologists w ho have seen the cults practised w ould agree that there were prob ab ly deep effects on both m ind and per­ sonality. But the ritual sym bolism s were also treasured for their ow n sak es. in som e sense, the A borigines m ay have been im prisoned by them through the aesthetic pleasure o f taking part. O ne need not hesitate to speak o f “ m ysti­ cal p articip atio n ” in the sense o f takin g part in evocative d ram as having to d o with m yster­ ies. But any suggestion o f mere traditionalism or m indless autom atism w ould be w rong. Effort, treasure, and enthusiasm were spent far to o freely. C on tem porary study is w eakened by the fact th at there is so much bias in the old printed record. O ne can n ot turn very hopefully to it for test or confirm ation o f new insights. Far to o m uch o f the in form ation w as the product o f m inds caugh t up with special pleadings o f one kind or another. T he Parson T hw ackum s: “ W hen I m ention religion, I mean the C h ris­ tian religion; and not only the C h ristian reli­ gion , but the Protestant religion; and not only the Protestant religion, but the Church of F'ngland.” The F'rnest C raw leys: “ M agic being sim ply the superstitious or religious method as

89

o p p o se d to the scientific.” T he sceptics: all those influenced by the succession from Com te an d Spencer, through T ylor and Frazer to Freud, w ho appeared to su pp ose that a disdain, ironical or hostile, o f all religion should be part o f a scientific attitude to any religion, even if unknow n. A nthropology becam e w hat it is w ithin, and in a definite sense because of, the great historical dissent from religion. Its debt to scepticism is profoun d. T ak e aw ay the influ­ ence o f the men listed above and the discipline w ould be unrecognizable. F'ven the fo rm u la­ tion o f problem s for study retains much o f their ou tloo k. Bur the dissent in the first place w as from particular religion - Christianity over an historical phase - and in respect o f som e only o f its theological form u lation s and institutional practices. T h at did not m ake a w arran t for the critique o f all religion or the contem pt o f any. A gain far to o much o f our old in form ation bears the influence o f men w ho w rote, as it w ere, with their left hands about religion but not o f it w hile, with their right, they were m ystery-m ongering about M ag ic, T otem ism , and an im aginary C h ild­ h ood o f M an . 1- . . I

If any o f the early sch olars had found A boriginal on tology o f interest, the history of study w ould read less dism ally. It is p reposter­ o u s that som ething like a century o f study, because o f ration alism , positivism , and m ate­ rialism , should have produced tw o options: that A boriginal religion is either (to follow D urkheim ) w hat som eone called “ the m irage o f society” or (to follow Freud) the neurosis o f society. T he fault goes deeper than concepts and categories. It is due to the im position o f a philosophy o f understanding. A philosopher of religion m ay feel entitled to m ake such an im position but not, I suggest, an an th rop olo­ gist o f religion. But be all th at a s it m ay. T he prim ary duty now is to avoid m ishandling w hat op p o rtu n i­ ties o f study rem ain. The A boriginal religions m ust be described and analysed as significant in their ow n right, a s exp ressio n s o f hum an experience o f life; as essays o f p assio n , im agi­ n ation , and striving am on g people to whom our historical dissents, clear and blurred, have m e a n t. . . nothing.

7

Remarks on the Verb "To Believe" Jean Pouillon

Jean Pouillon (1916-2002) was one of the great intellectuals of 20th-century France. His work embraces an engage­ ment w ith ethnology, philosophy, li­ terature, politics, and psychoanalysis. Pouillon w rote on literature, film and politics in Les Temps Modernes, begin­ ning with the first volume in 1945 and moving on to anthropological subjects by the mid-1950s. He carried out field ­ work among the Hadjerai of Chad and in Ethiopia, and was editor of L'Hom m e from 1960 to 1996. Some of his best essays are collected in Fetiches sans F6tichisme, Paris, Maspero, 1975, and Le Cru et le Su, Paris, le Seuil, 1993. The essay here, reprinted in virtual entirety, is elegant and to the point, incisively demonstrating the paradoxical and culturally specific nature of the verb "to believe." Together w ith the fo llo w ­ ing essay by Malcolm Ruel it successfully dismantles w hat is often taken as the key term in definitions or descriptions of

religion. A lthough the article was des­ cribed as "structuralist," the approach has strong affinities to the school of ordinary language philosophy associ­ ated w ith Austin and the later portion of W ittgenstein's thought in that it begins by looking at the ways in which words are actually used in everyday speech ("parole" rather than "langue"). As Pouillon implies, this complicates the translation problem; w e have first the difficult task of exploring language use (rather than simply definition in the abstract) in our field settings and second the problem of mapping uses from one linguistic comm unity to another. Pouil­ lon offers the suggestion of tw o kinds of religion, one based on faith and the other on local knowledge. But this in turn might lead us to inquire about the presence of w hat would then be a kind of philosophical skepticism (rather than theological doubt) in the latter context.

From Remarks on the Verb “ To Believe” in Michel Izard and Pierre Smith, eds, John Leavitt, trans. Between Belief and Transgression: Structural Essays in Religion„ History, and Myth (University of Chicago Press, 1982 11979J) pp. 1-8. Slightly abridged.

REMARKS ON THE VERB "TO BELIEVE"

The French verb croire (“ to believe” )1 is p a r a ­ doxical in that it expresses dou bt a s well as assuran ce. T o believe (croire) is to state a con ­ viction; it is also to add a nuance to that con ­ viction: “ 1 believe” (je crois) often signifies “ I’m not su re.” T h is am biguity involves the subjective side o f belief (croyance). As regards its object, the situation is no less equivocal, since rhe com plem ent o f the verb can be p ro­ duced in rwo w ays: direct or indirect. W hat is m ore, the indirect construction itself has tw o form s: croire a . . . (“ to believe in ,” “ to think o f” ) is not the sam e thing as croire en . . . (“ to believe in ,” in the sense o f being w illing to rely on), which both differ from croire + direct object or croire que . . . (“ to believe t h a t . . . ” ). Finally, the m eaning o f the verb and the con ­ struction o f the com plem ent can vary depen d­ ing on the nature o f the object: m an, g o d , fact, value, sta te m e n t.. . . T his suggests tw o questions (at least): is it possible to order this diversity o f usages? If so, is this order universal or does it characterize only a certain type o f culture, and in this case w hat is the basis for the w o rd ’s unity? In other w ords: how is it that m ultiple m eanings do not require diverse e x p ressio n s?2 Hut since this is apparently the case, is a tran slation o f the verb in all its senses possible in other lan guages, using a single term? Croire a . . . is to state that som ething exists; croire en . . . is to have confidence; croire que . . . is for som ething to be represented in a certain w ay. A lthough the difference between the tw o indirect con struction s m ay ap p ear slight, it is undeniable, as the follow ing exam ple show s: a person believes in (croire en> “ trusts in ” ) G o d, while one believes in (croire a) the Devil, that is, one recognizes that he exists w ithout, by definition, putting on e’s faith in him: one cannot croire en the Devil. Croyance en (“ belief in, trust in” ) G od does imply croyance a (“ belief in ” ) his existence, but im plication is not identity. O n the other hand, this im plication seem s so ob vious that it often goes unform ulated: a believer believes in (icroire en , “ trusts in ” ) G o d , he feels no need to say that he believes in (croire a) G o d ’s reality; he believes in (croire a) it, one w ould say, im plicitly. But is this certain? In fact, the believer nor only need not say that he believes

91

in (croire a) the existence o f Ciod, but he need not even believe in (croire a) it; precisely because in his eyes there can be no doubt ab o u t it: the existence o f Ciod is not believed in (crue ), but perceived. O n the con trary, to m ake G o d ’s existence an object o f belief, to state this belief, is to open up the possibility o f d ou bt - which begins to clarify the am biguity with which we started. So one could say that it is the unbeliever w ho believes that the believer believes in (croire a) rhe existence o f G o d . O ne could call this a play upon w ords; but these w ords do lend them selves to it, and it is precisely this possibility that m ust be exp lored, if not elucidated, in trying to o rg a ­ nize the field o f their usages. Besides, w hat I have just said will ap p ear much sim pler if we leave the religious dom ain. If I have confidence in a friend, if I believe in (croire en, “ have faith in” ) him, will I say that I believe in (croire a) his existence? Certainly nor; that existence is, sim ply, undeniable. It is only if it were not unquestionable that I w ould have to believe in (croire a) it, and believe in it explicitly. A gain, it will probably be said that this is playing on w ords, this time on the w ord “ existen ce,” for m an ’s existence, by definition, is not on the sam e level as that o f the deity. By definition, yes, but by cultural definition: the distinction between a cultural w orld and a natural w orld, or between a “ this w orld ” and a “ beyon d,” is w idespread, but it is not universal. It is this distinction between tw o m odes o f existence that leads to a distinction between tw o w ays o f apprehending w hat exists: perception and know ledge on one side, belief (croyance) on the other. From this kind o f perspective, the existence o f supernatural beings can only be an object o f belief, and this is why wherever this distinction is m ade the phenom on o f belief as the affirm ation o f existence takes on this am b igu o u s aspect, between the certain and the questionable. T h is is not the only reason. L.et us now consider the relations between croire a . . . (“ to believe in (a fact)) and croire que . . . (“ to believe t h a t . . . ” ). T o believe in (croire a) the existence o f X - “ g o d , tab le, or w ash b asin ” can be expressed in a direct con struction: to believe that (croire que) X exists. But this is a statem ent o f a peculiar type - the existence o f

92

JEAN POUILLON

a god or o f a hundred thalers is not an attri­ bute - and is different from the statem ent that endow s X with certain ch aracteristics which perm it X to be represented to oneself. The representation, the content o f belief, is acco m ­ panied by an affirm ation o f existence but is sep arable from this affirm ation ; the affirm a­ tion can be bracketed - the H usserlian epoche - and this is w h at m akes possible studies o f beliefs a s such: one need not believe in (croire a) w hat one believes in order to analyze it. The “ I believe” (je crois) which precedes so m any statem en ts o f the m ost diverse kinds, is the m ark o f a distan cing and not o f an adhesion. These tw o m ovem ents, which a single verb is able to exp ress, ap p ear radically o p p o se d , or else com pletely unrelated. Belief a s represen ta­ tion, as statem ent, pertains to w hat is also called ideology; there is no isolated belief, every representation is part o f a global system which is m ore or less clearly, m ore or less consciously articulated , a system which m ay be religious but m ay also be ph ilosoph ical, political. . . . Belief as faith (confiance) is the conviction that he to w hom one has given som ething will reciprocate in the form o f sup p ort or protection ; it calls forth a relation ­ ship o f exchan ge, o f which the relationship between the believer an d his god is only a particular case, even if a frequently privileged one. O ne puts on e’s faith (confiance) to the sam e end, whether in an individual, in a party, in an institution. It is significant in this regard that Benveniste, in his Indo-European L an ­ guage and Society (1 9 6 9 ; English edition, Lon d on , 1973), discu sses belief (la croyance) not in the section on “ religion ” but in th at on “ econom ic o b lig a tio n s.” For he sees the o rigi­ nal m eaning o f belief in this credit which has been accorded and should be returned. M u st we then see belief-as-representation as a deriv­ ative m eaning? O r else as a m eaning that has been added o n , and which w ould turn the verb “ to believe” into a con glom erate w ithout unity? T h is derivation is certainly a p ossible one: to believe in ( croire en) som eone, to give him credit, is, am o n g other things, to believe (croire) w hat he says, and in this w ay one goes from the trust to the statem ent that it allow s

one to take a s established fact. T h is is espe­ cially evident when the belief ap p ears in the form o f religious faith: trust in (croyance en) a god is usually the b asis o f w hat we call a credo , a g ro u p o f statem ents which becom e the direct ob ject o f belief. The sam e is true in m any other dom ain s. For political exam ples, there is an over-abundance o f choices. But it is a lso possible (m ore often than is usually . . . believed!) to accept a proposition that is said to be scientific just a s one accepts a d ogm a or even the possibly fan tastic assertion o f a man w ho is judged w orthy o f tru st; I believe it not because I am able to prove it, but because I have faith in those w ho say they have proven it, for exam ple, in Einstein when, follow ing him, I w'rite F' = m e2. But we w ould m iss the essential part o f belief-as-representation if we reduced it to this sole case in which it is based on the argum ent o f authority. The specific trait o f a representation is to ap p ear obviou s, to be self-evident, and rhe fact that it is alw ays p o s­ sible to bracket the judgm ent or the feeling o f ob viousn ess ch an ges nothing: the ob vious is replaced by the arbitrary, but both sim ply mean rhat this form o f belief is based on nothing but itself or the cultural system within which it finds its m eaning. S o it seem s im possible to overcom e the poly­ semy o f the w ord. Its religious usage does allow us to unify rhe verb’s three con stru c­ tions, but it does not elim inate the other u sages; over and above this, it only unites the three constructions in religions o f a certain type. T his ob servation leads us to question its an th ropological usage, now well established and apparently un p ro b lem atic.1 W hat anth ro­ pologist w ould deny th at he seeks to uncover the beliefs o f those w hom he studies, to com pare them with our ow n beliefs or those o f other peoples, a s if this object o f study and its designation presented no a priori problem , a s if it were ob viou s that every hum an being “ believes” (croire) - this being one o f our beliefs - in the sam e w ay, if not, o f course, in the sam e things? T h e dan ger in this situation is not sim ply the well-know n if not alw ays foreseen one o f in appropriately applyin g a ca t­ egory that m ay have m eaning only in our ow n culture; it has to d o with the fact that this category m ay not be a single, unified one at

REMARKS ON THE VERB "T O B ELIEVE"

a ll, even for us, or at the least that it is a sh at­ tered category, w hose fragm en tation is, pre­ cisely, a singular cultural phenom enon. W hat is m ore, an th rop ological usage reduplicates the p a ra d o x 1 em phasized above when I said th at it is the unbeliever w ho believes that the believer believes. If for exam p le 1 say that the D an g aleat4 believe in (croire a) the existence o f margat\ this is because I d o not believe in their existence an d , not believing in it, 1 think that they can only be believing in (croire a) it in the sam e w ay that 1 im agine I co u ld , if I did. But how can one tell whether they believe (croire) and in w hat w ay? W hat question can one ask them , using w hat w ord o f their lan­ g u age , in w hat con text? O r, inversely, how is it possible to translate into French the w ord or w ord s they use to talk ab o u t w hat is to our eyes an object o f belief? In J . Fedry’s Dictionnaire dangaleat , we find the verb abide “ to perform rhe rites faith fully.” It com es from the local A rabic ab ad a , “ to ad ore Ciod,” ad oratio n being understood a s a ritualized activity. It is a m atter o f w orship (du culte ), o f faith in action , and not o f the repre­ sentation o f a being w hose existence m ust also be affirm ed. T his verb is used with a directob ject com plem ent: this being, G o d for con ­ verts to C hristianity or Islam , or the m argai for others. The best w ay to translate it is thus “ to serve,” in the biblical sense o f the term: to w orship (rendre un culte a). N o abday maragi, “ I serve the m argai .” A nother verb, amniye , signifies “ to bestow on e’s trust o n ,” “ to rest o n ,” “ to believe in ” (croire en). It is con ­ structed with an indirect-object com plem ent, introduced by the p reposition ku: no amnay ku m an go , “ I have faith in the m a r g a i” “ 1 give my faith to the m a r g a i this is the verb that C h ristian s use to say “ I believe in (croire en) G o d ,” no am nay ku bungir. In co n trast to the foregoing, this verb is not used exclusively in religious con texts: one can evidently, as in French, put one’ s faith in another person. The first sense given by the d iction ary, besides this, is “ to be used to, fam iliar with . . . a nd one will say, for exam ple: no amniyiy-g pisb , “ I am used to h orses.” T h is to o is a w ord o f A rabic origin w hose Sem itic root has given us the “ am e n ” o f Christian liturgy which, as Fedry poin ts ou t, m arks adhesion to a person more

93

than to a conceptual “ tru th .” A s this author n otes, “ one m ay w onder ab o u t the fact that both o f these verbs com e from A rabic, w hose linguistic influence is very strong in D an galeat, as in other H ad jerai languages. But this should not m ake us dou bt that w hat the D an galeat have taken in has becom e an integral p art o f th em selves.” I will ad d in turn rhat from the lan gu age o f a religion with a credo (an affirm a­ tion o f existence and a set o f statem ents and representations) the D an galeat have taken w h at fits their ow n w ay o f “ believing” (croire): the term s that designate a specific behavior an d m ental attitude - w orshippin g (rendre un culte) and trusting in the addressee o f the w orsh ip - and not term s that are based on definite representations or p roposition s. O ne m ay thus translate our “ believe in ” (croire en) into D an galeat, and the fact thar the H adjerai took the w ord from A rabic suggests th at for them it expresses the essential aspect of belief (and o f religious faith in general, says Fedry, w ho belongs to the Society o f Je su s and should know w hereof he speaks): faith (la con­ fiance). But in this case, how can we translate “ to believe (th at)” (croire que)} T o find out, to know , to know ab o u t som ething, is ibine\ pakktne serves to render: think, su p p o se, figure ou t, foresee. T hese tw o verbs are pure D an g a­ leat. The first will be used to m ark certainty and so tran slates “ to believe” (croire) in cases where the French verb is m ore or less equ iva­ lent with “ to k n o w ,” when for instance Don Ju a n say s to Sganarelie, w ho is questionin g him ab o u t belief, “ I believe that tw o and tw o m ake fo u r.” T he second verb covers the doubtful usages o f our verb, all those in which the speaker takes a certain distance with regard to w hat he is talkin g about. In sum , we can tran slate all aspects o f the verb “ to believe” . . . except the verb itself. W hat we have been able to tran slate has been the French equivalent o f “ to believe” in each o f its particular u sages, but in D an galeat there is no single term rhat serves as the basis o f their unification. In other w ords, we can translate everything except the am biguity. We m ust therefore return to the reasons for this am b ig u ­ ity. A m biguity is nor sim ply polysem y, the fact that a verb som etim es has one m eaning and som etim es an oth er, each o f them unequivocal;

94

JEAN POUILLON

it is, rather, that all o f these m eanings, even the con tradictory ones, are intrinsically linked; that, above all, there is alw ays d ou bt at the heart o f the conviction, and that the affirm a­ tion itself indicates that it could alw ays be suspended. But why condense this p arad o x ical liaison into a single w ord instead o f sortin g out its elem ents, a s the H adjerai d o ? The answ er, “ I believe,” lies in the com p arison between a religion like C hristianity and a religion like that o f the D an galeat. It is not so m uch rhe believer, I w ould say, w ho affirm s his belief as such, it is rather rhe unbeliever w ho reduces to mere believing w hat, for the believer, is m ore like know ing. N evertheless, the Christian cannot avoid expressin g his faith not only as trust in G od (confiance en)9 but also as belief in (croyance a) his existence an d belief that (croyance que) G od possesses such and such attribu tes, that the w orld w as created, and so forth. H e states this as a belief, even though he know s it - but also because he know s that by this very fact it is con testable and contested. A bove all he know s that there are other beliefs, on the one hand because his religion has a history and w as constituted again st the “ fa lse ” g o d s, on the other because this history is not over yet and there are still idols to be elim inated; and there can be other beliefs only because his own belief is one am o n g others. N ex t, he know s - it is even an essential point in his credo - that the object o f his belief is in a “ reality” o f a different order than the realities o f the w orld o f creation, which are the object o f a perm a­ nently revisable scientific know ledge, or o f calculation s, o f predictions that can be proven w rong; and he a lso kn ow s that this possibility o f revision lies in the dem onstrable or verifi­ able character o f the know ledge or the hypoth­ esis, a character w hose legitim acy he challenges in the case o f his belief, but which, inversely, challenges the legitim acy o f his belief. T h u s he m ust sim ultaneously assum e both his affirm a­ tion and the challenge to it, a challenge that belief is, nonetheless, supposed to m ake im p o s­ sible on its ow n level. In other w ord s, the con tradiction is inside his faith, and that is w hat it is “ to believe.” 5 This situation is the result o f the distinction m ade between tw o w orlds: the K ingdom o f

G o d and this w orld. In our culture such a d is­ tinction seem s so ch aracteristic o f religion, to those w ho reject it as much a s to its adherents, rhat religion in general and so-called “ prim i­ tive” religions in particular are usually defined by a belief in supernatural pow ers and by their w orship. There is even a tendency to think rhat rhe extent and im portance o f the supernatural w orld are much greater for “ prim itives” than for “ m od ern s,” that super-nature is not only the dom ain o f g o d s an d spirits but a lso , for exam ple, the dom ain in which the pow er o f the m agician and the sorcerer operates. I cer­ tainly d o not m ean to deny that at any lati­ tudes one can find people w ho believe in (croire a) the supern atu ral, but one equally finds people for w hom such an affirm ation is co m ­ pletely m eaningless - w ithout them being, for all that, areligious - far from it. For here we have a m ajor m isunderstanding: because we have constructed rhe concept o f natural law , we are ready to adm it the supernatural (whether as illusion or as other reality hardly m atters) as a place to put w hatever co n tra­ venes, or seem s to contravene, natural law ; but this is our ow n notion, whether we judge it well grounded or not, and not that o f rhe people to w hom we abusively attribute it. As Evans-Pritchard rem arks, “ m any peoples are convinced that death s are caused by w itch­ craft. T o speak o f w itchcraft being for these peoples a supernatural agency hardly reflects their ow n view o f the m atter, since from their point o f view nothing could be m ore n atu ral.” 6 For his part, C laude Levi-Strauss has stressed the realist, m aterialist character o f m agic, its m onistic, not dualistic, conception o f the w orld. The m argai , these spirits w ho have such an im portant place in the individual and social lives o f the H ad jerai, are invisible, nonhum an pow ers; they act unpredictably, and are the cause o f w hatever distu rbs rhe natural course o f things. But they are no less a part o f the sam e w orld a s hum an beings. The latter believe in (croire a) the existence o f the m argai like they believe in their ow n existence, in that o f an im als things, atm osph eric phe­ nom ena. . . . O r rather they d o not believe in (croire a) it: this existence is sim ply a fact o f experience:8 there is no m ore need to believe

REMARKS ON THE VERB "TO BELIEVE"

in (croire j ) the m argai than to believe that if you throw a stone it will fall. O ne fears and/or trusts in them, one gets to know them , one gets used to them, one perform s the special sacri­ fices that please each m argai , and one is careful to m ake no m istakes, for fear o f getting sick or being affected in som e unpleasant w ay. If we can speak o f a D an galeat religion - another untranslatable expression - it is not in the sense in which the faithful share a single elab o ­ rated body o f beliefs ab o u t supernatural beings, but rather in the etym ological sense, accordin g to Benveniste, o f the L.atin religio :y that o f a m eticulous concern for the proper carrying out o f the cult, w ithout, how ever, being able to define the necessary correctives in advan ce; at every occasion , one takes aim within uncertainty. O ne can only estim ate w hat each m argai desires. The four verbs men­ tioned above define these behaviors equivo­ cally and w ithout con tradiction s: one serves the m argai , one trusts in them (that is, in the m utually fruitful nature o f the exchange in au­ gurated by the sacrifice), one know s from experience that they exist, and one tries to guess their intentions. All this does presuppose a particular representation o f the w orld, but one which excludes the possibility o f its e x p la ­ nation in the form o f “ belief,” o f an assertion that in spite o f itself is d ou btfu l, relativized. The D an galeat certainly know that others think differently, an d it h appens that m any o f them convert to Islam or to Christianity. But this situation can n ot surprise them: one does not believe in (croire a) the m argai ; one experi­ ences them , and this experience is first o f all a local one; such spirits d o not necessarily exist everywhere. While the encounter with other­ ness relativizes C h ristian belief in an other-

w orldly absolute, it confirm s the D angaleat experience o f the w orld, which is relative from the beginning and so can n ot be disturbed by diversity. T his is why religions o f the D an ga­ leat type are w ithout the proselytizing inherent in religions founded on beliefs w hose vulnera­ bility im pels their form idable d yn am ism .10 If rhe D an galeat have no need o f the verb “ to believe” this is not solely because o f their m onism , as opposed to C h ristian dualism . Equally in play is another oppo sitio n , one between the historicism o f the Christian reli­ gion and the em piricism o f D an galeat religion. T h is em piricism assu res everyone o f the pres­ ence o f the m argai, and has no need o f an intercessor. Every m an perform s his own sac­ rifices and will have recourse to the diviner only to know w hat an im al, o f w hat sex and w hat color, he should kill and on w hat day. A religion like C hristianity or Islam is based, on the contrary, on a revelation, testim ony, a transm ission w hose fidelity is guaran teed by a church or specialized experts. T h is revelation is, precisely, thar another w orld exists; the rev­ elation is a unique historical event, its content is constituted by the w ords o f its protagonist, G o d incarnate or prophet. So everything rests on a faith, which is sim ultaneously a trust and a specific credo. All the m eanings o f the verb “ to believe” should then com e together, but this necessity is nothing m ore or less than a cultural necessity. It is only in this perspective, in my opinion, that we can sp eak o f “ religious belief,” and it is only when it is understood that this notion does not have universal value that we can appreciate how difficult the problem o f a general definition o f religion really is; but this m ay also be the point from which we can try to resolve the problem .

NOTES 1

T N : The au th o r’s distinctions am on g d if­ ferent m eanings o f the verbs croire , croire en , croire a , etc., bear on the sem antics o f the w ord, not directly on its m orphology. For this reason I have not follow ed the inflections o f the French verb in my brack­ eted clarifications, but have usually put the infinitive form , w hatever the tense,

95

2

person, etc. o f the verb in rhe text. For exam ple, when the French text say s il croit I have put “ he believes” | croire], ro m ake clear the opposition w ith, say, il croit en , which I translate “ he believes in .” Diverse expression s d o exist, however: credit (creance ), confidence, trust (confi­ ance), faith (foi). . . . But while one might turn to these for the sak e o f precision, they are not required by usage.

96 3

4

5

6 7

JEAN POUILLON

R odney N eedh am has done this ( Belief , Language, an d Experience ), in a perspec­ tive different from my ow n. T he tw o do overlap, how ever: the them es are neces­ sarily the sam e, but they are put together in different w ays. The D an galeat are one o f the g ro u p s called H ad jerai, w ho live in the central region o f the R epublic o f C h ad , D ep art­ ment o f G u era. They w orship (rendent un culte a) w hat one could sum m arily call local spirits: the margai. It w ould be easy to show that today m any “ political believers” find them selves in an a n alo g o u s situation. Hut they are not alw ays as aw are o f it as Saint A ugustine w as when, accord in g to T ertullian, he said: credo quia absurdum. Theories o f Primitive Religion (O x fo rd , 1965), pp. 1 0 9 -1 0 . Levi-Strauss, The Savage M ind , Knglish translation (C h icago, 1966), pp. 2 2 1 - 2 .

8

9

In the sam e w ay, am o n g the N uer the expression Kwoth a thin (“ Ciod is pre­ sent” ) “ does not m ean ‘there is a Ciod.’ T h at w ould be for the N uer a pointless rem ark. G o d ’s existence is taken for granted by everybody. C onsequently, when we say , a s we can do, that all N uer have faith in G o d , the w ord ‘faith ’ m ust be understood in the O ld T estam en t sense o f trust (the N u er Ngath). . . . There is in any case, I think, no w ord in the N uer language which could stand for kI believe’ .” Evans-Pritchard, Nuer Religion (O x fo rd , 1956), p. 9. Benveniste, Indo-E.uropean Language

and Society. 10 I d o not mean to say that som e beliefs are vulnerable and oth ers are not. Any belief, in the fact o f its com m unication, m akes itself, and know s itself to be, vulnerable.

8

Christians as Believers Malcolm Ruel

Malcolm Ruel is a British anthropologist (trained at the University of Oxford, taug ht at the University of Cambridge) w ho has carried out fieldw ork in both W est Africa (1969) and among the Kuria of Kenya, East Africa (1997). The collec­ tion of essays on the Kuria (in which this essay was reprinted as chapter 2) contain some particularly fine reflections on the nature of religion and the means to com prehend it. The thrust of Ruel's argum ent is that one may need to dis­ pense w ith western, Christian-inspired notions of belief and deity and that, indeed, the primary inclinations of Kuria religion may be described w ithout recourse to metaphysical concepts. Ruel concludes th at "religion is possible w itho ut the supernatural and that the supernatural figures only peripherally in Kuria religion" (1997: 237). W hat concerns the Kuria in the conduct of their rituals is "ordered grow th" and "the securing of life."

Ruel's essay is im portant for several reasons. First, it addresses the Christian tradition anthropologically and thereby forces us to become aw are of how much of the anthropological fram ew ork on religion has been compromised by that tradition. Ruel's essay also shows the im portance of historical accounts of key terms whose meaning and use shift over tim e. In effect, he provides us w ith a m ultilayered genealogy of a key term of both Christianity and anthropology. Ruel also shows the connection between the history of the term and the political developm ents w ithin the church. In so doing, h e q u e s tio n sth e v a lu e o f "belief" as a concept for the anthropology of religion, not, like Needham (1972), for reasons of philosophical or psychologi­ cal precision, nor, like Pouillon, for prob­ lems o f com parability, but because of the very d ifferent meanings it has held w ithin even one historical tradition. Ruel offers independent confirm ation

From M alcolm Ruel, “ Christians as Believers,” in John Davis, ed., Religious Organization and Religious Experience (London: Academic Press, 1982), pp. 9 -3 1 . Abridged.

98

M ALCOLM RUEL

of Pouillon's argum ent that in many societies the relations that people hold to the concepts they use or the ideas they express may be radically distorted by the application of the term "belief," thereby leading in turn to a w hole series of false problems such as how such "beliefs" can be justified. For further elaboration readers should turn to Belief, Ritual , a n d the Securing o f Life: Reflexive Essays on a Bantu Reli­ g io n (Ruel 1997). The arguments there

tion of Kuria religion, but in fact have relevance for much of Africa and beyond. Although Ruel is directly and deeply indebted to Evans-Pritchard, it is instruc­ tive to contrast Ruel's conclusions w ith those of Evans-Pritchard w ith respect to the East African Nuer (chapter 11). On the Christian bias underlying anthropo­ logy and the paradoxical consequence that this has obscured the anthropologi­ cal analysis of Christianity in particular, see Cannell (2005, 2006).

are developed by means of an apprecia­

The argum ent o f my paper is sum m ed up in an early observation o f W ilfred Cantw ell Smith: The peculiarity o f the place given to belief in Christian history is a monumental matter, whose importance and relative uniqueness must be appreciated. So characteristic has it been that unsuspecting Westerners have . . . been liable to ask about a religious group other than their own as well, “ What do they believe?” as though this were the primary question, and certainly were a legitimate one. (1978, p. 180. But see also Smith, 1977, 1979) “ U nsuspecting W esterners” m ust o f course include unsuspecting w estern an th rop ologists w ho, as m any texts will show (e.g. EvansPritchard, 1937, p. 21), give prim acy to w hat people “ believe” w ithout fully declaring w hat that w ord m eans, nor recognizing, it w ould seem , ju st how rooted the concept is in our ow n cultural religious tradition, Christian and post-C h ristian , and thus how loaded any statem ent concerning “ belief” easily becom es. T his then is one reason why an an th ro p o lo ­ gist m ay be excused if he m oves so far from his last as to attem pt, however incom petently, to sketch in outline the m onum ental peculiar­ ity o f C h ristian “ belief” . At a time when an th ropology has turned m ore and m ore to give an accoun t o f the cognitive asp ects o f culture it is a s well for us to be aw are o f the com plexity o f the concepts that we draw from

our ow n culture, which have a history and contextual com pulsion of their ow n which often ill-match the ideas and actions they are used to interpret. The need for critical reflec­ tion becom es even greater when, as in the case o f “ believing” , there has been a radical shift in the use o f a term w hilst som ething o f its force has been retained. “ Believing” in the sense o f being com m itted to som e definable set o f values has becom e secularized, detached from Christian believing but not dem oted as a concept, so that in a post-C h ristian , secular culture the phrase “ I believe . . (e.g. in the title o f Forster’s essay, 1939) still gives prom ise o f a personal statem ent o f som e significance, a declaration o f m oral identity. There is here another reason why it is a p p ro ­ priate for an anth ropologist to attem pt the task o f ethnographic placing, for it is part o f the an th rop ologist’s trade to look hard and long at certain key concepts and to explore how use and m eaning, context and idea, are constantly engaged in an interplay in which concepts link situations while situ ations qualify (and thus help to define) concepts. N o w “ belief” is essentially a w ord that relates and defines: it relates people, situations and ideas; but in its turn, a s I shall argue, it is a lso in very im portant w ays defined by the con text o f its use. In this, function and m eaning com e alm ost (but never entirely) to coincide; consider, for exam ple, the ph rase “ the com m unity o f believ­ e rs” that runs like a them atic p assacaglia through H an s K iin g’s The Church (1 9 6 8 ). If

CHRISTIANS AS BELIEVERS

only to keep for som e short time the p h iloso­ phers at bay (to w hom the cognitive prom ise contained in the w ord “ belief” com es, as it were, a s a gift from heaven) let us assert reso ­ lutely, at least for the present, that the (C h ris­ tian) concept o f belief is a s it does and proceed to consider it situationally and behaviourally. T o n arrow som ew hat the vastn ess o f the topic, four periods have been selected from the history o f the church in which to discuss the idea o f belief and how it is involved in any definition, corporate or p erson al, o f Christian identity. They are: (1) the critical, initial phase in which C h ristian s, the N azarene sect, em erged as a distinctive religious m ovem ent, a com m unity o f believers; (2) the im m ediately succeeding period leading to the Council o f N icaea (325) that w itnessed both the develop­ ing form al organ ization o f the Church and the establishm ent o f o rth o d o x creeds, sanctioned by the Church councils; (3) the R eform ation and in particular Luther’s reform ulation o f w hat it m eans to believe (i.e. to have faith); and finally, since we cannot leave ourselves out, (4) the present period, which m ight be characterized in both C h ristian and secular con texts as belief diffused - “ beyond belief” in the phrase o f one (diffusely) believing an th ro­ pologist. In this section I use the w ord “ belief” only with its Christian reference and where at all possible I keep to this one w ord, assum ing a sufficient continuity and overlap in m eaning between “ faith ” and “ belief” to allow “ belief” to do duty for both, except where there is particular need to distinguish them. T h is usage h as the advan tage o f perm itting a single w ord correspondence between belief in English, pistis in Greek and the root ’ mn in H ebrew ; this does not imply that these w ord s have (col­ lectively) the sam e m eaning nor that they have (singly) a con stan t m eaning, only that their range o f m eaning is historically and sem anti­ cally continuous. (On this issue my usage is radically at variance with th at o f Wilfred Cantw ell Smith.) The detailed scholarly w ritings on the term i­ nology o f belief, pistis , in the N ew Testam ent bo o k s m ake it possible to offer a num ber o f sum m ary points. (I rely chiefly on Bultmann and W eiser, 1 9 6 1 ; H atch, 191 7 ; M ichel, 1975;

99

and M oule, unpublished.) In its various form s,

pistis (belief), pisteuo (believe), pistos (faithful, trustw orthy), apistia (unbelief), form a key and m uch used set o f term s in the N ew Testam ent. The m eaning o f the w ord-grou p does not (with som e qualification) depart from its general m eaning, or set o f m eanings, in G reek, but its N ew T estam ent use also carries certain con no­ tations derived from that fact that pisteuo , ro believe, w as rhe term consistently used to tran s­ late the H ebrew h&'min, from the root ’ mn (m eaning to be true, reliable or faithful) in the Septuagint. One needs here to distinguish between the m eanings o f w ords and the reli­ giou s ideas they express, for although the tw o m ay coincide, they do not alw ays d o so , and changes in m eaning follow often from the developm ent o f pre-existing ideas. T hus both the original Greek use o f pisteuo and the H ebrew term ymn express centrally the notion o f trust or confidence. O riginally the Greek w ord-grou p “ denoted con duct that honoured an agreem ent or bond. It had a social orien ta­ tion, and its use indicated m isconduct by im pli­ c a tio n ” (M ichel, 1975, p. 5 9 4 ). In classical G reek literature ptstis m eans the trust that a m an m ay place in other m en, or g o d s; credibil­ ity, credit in business, guaran tee, p ro o f or som ething to be trusted. Sim ilarly, pisteuo m eans to trust som ething or som eone (ibid.). T he word acquired a religious use at an early date, when to “ believe” (pisteuo) the g o d s or an oracle expressed on the one hand confidence in them (their veracity or ability to prom ote welfare) and on the other obedience to them, an acknow ledgem ent o f their pow er to deter­ mine hum an fate. The H ebrew term ’ mn denotes even m ore directly a quality o f rela­ tionship: it w as used o f the reliability or trust­ w orthiness o f a servant, a w itness, m essenger, or a proph et, but it also served to characterize the relationship between G o d and his people, reciprocally trusted and trusting, bound by covenant to each other (M ichel, 1975, pp. 5 9 5 -6 ). In the N ew T estam ent the w ord pistis and its related form s still carry the ideas of trust and confidence. In a citation that rings reverberatively through the theology o f the centuries, Paul refers to the belief (pistis) that A braham had in G o d ’s prom ise that he w ould becom e “ the father o f m any n atio n s” (the story

100

M ALCOLM RUEL

is told in G enesis 15) a s an exem p lar for the kind o f belief (pistis) shared by the early C h ris­ tians. A s the belief (i.e. trust) o f A brah am w as reckoned as righteousness for him , so “ Faith is to be reckoned as righteousness to us a lso , w ho believe in Him W ho raised from the d ead our Lord Je su s C h rist, w ho w as delivered to death for our sins and raised again to secure ou r ju s­ tification” (R o m an s 4: 1 3 -2 5 ). Yet in spite o f this continuity between the H ebrew and the G reek, the O ld an d the N ew T estam en t, the w ord pistis d oes com e to acquire a special tw ist in the ap o sto lic w ritings o f the N ew T estam en t. O ne m ight say th at it acquires a technical use. T h u s the verb pisteuo , to believe, is often used in the sense o f to be converted, to becom e a C h ristian: “ they heard the m essage and believed” is a form ula that repeatedly occurs in the narrative o f the expan din g church in the A cts o f the A postles; Paul w rites o f “ when we were first believed” (R o m an s 1 3 :11) in the sense o f “ when we were first con verted” ; and there are other exam ples. Sim ilarly, the nom inal form “ believers” (either hoi pisteuontes , “ those believing” , or hoi pistoiy “ those o f the belief” ) refers to rhe con ­ verted, the “ bro th ers” or the “ sa in ts” as they are also called. We should note that the w ord “ C h ristian ” is itself rarely used (three tim es in the N ew T estam ent) and then alw ays in the context o f w hat others - the people the C h ris­ tians called unbelievers - were callin g them. Finally, the noun pistis denotes the “ belief” held collectively by the early C h ristian s as a com m on conviction, a shared confidence that both distinguished and united them as a co m ­ munity. Paul lists these identifying features explicitly and succinctly in F'phesians 4 :4 - 5 in which the central elem ents are “ one Lord, one belief, one b ap tism ” . We need to look m ore closely at the su b ­ stance o f this shared belief in the last sense above, for it is in relation to this that the concept gain s added depth and range. Essen­ tially w hat these early converts believed w as w hat th eologian s have com e to call (using another technical term ) the kerygma or p roc­ lam ation o f the C h ristian m essage (Bultm ann and W eiser, 196 1 , p. 6 9 ; M ichel, 197 5 , pp. 6 0 1 , 6 0 5 ; H atch, 1 9 1 7 , pp. 3 3 -4 ). N o w this does not m ean just the teaching o f Je su s, but

rather the teaching ab ou t Je su s, and the crucial fact about Je su s, which sum m ed up all the rest, w as his resurrection: this fact is expressed clearly in the p assage from Paul, quoted above, and throughout rhe epistles (and we should recall that these are the earliest C hristian d o cu ­ m ents that we have; the g osp els were written later). C h ristian belief now begins to part com pany from H ebrew trust . Both refer to relationship - the confidence rhat people have in G o d , and in the case o f the C h ristian s in G o d through C h rist - but for C h ristians there is rhe added confidence or conviction ab o u t an event (the resurrection and all that that signi­ fies) that had actually taken place. The belief is nor just open-ended, oriented to w hat Ciod m ay or can do: it is rooted firmly in w hat G od has done, which to deny is to deny the W ord o f G o d, that is, the action o f G o d in the w orld. (O n this point see especially Bultm ann and W eiser, 1961, p. 82 et seq.) This developm ent w as to have enorm ous consequences for rhe later use o f the concept for it is bur a short step from belief a s accepting as a fact (i.e. the event o f the resurrection) to belief as asserting as a proposition. A distinction m ade frequently today is between “ belief in” (trust in) and “ belief rh at” (proposition al belief). The d is­ tinction m ay clear our m inds today but it con­ fuses history, for the point ab o u t Christian belief, reiterated by th eologians (e.g. Lam pe, 1976; M oule, unpublished), is rhat it w as both at once. The creeds, which we m ust now consider, both reflect an d perpetuate this particular notion o f C h ristian belief, that concerns a com plex person-event, not least in their reiterated verbal form u la: “ I believe in - w ho did - ” : Person + F'vent, the tw o reciprocally defining (Lam pe, 1976). Yet if the kernel o f the creeds is the recognition o f this personevent, their history is one o f grow ing e lab o ra­ tion and fo rm alizatio n , a developm ent that takes place in relation to the developing organ ization o f the Christian body: the shared conviction o f a scattered com m unity o f C h ris­ tians becom es the confirm ed orth o d oxy o f the conciliar church. Brief crcdal ph rases are com m on in the N ew T estam ent and there are occasion al longer sum m aries (as in the p assage by Paul) when

CHRISTIANS AS BELIEVERS

the w riter evidently felt som ething m ore explicit w as required. In either case they serve a s sum m ary statem ents o f the teaching or kerygma concerning C hrist. T he form ulary ph rase, Kurios Iesous , “ Je su s |is] L o rd ” , is com m on and there are m any varian ts. It is clear m oreover that such phrases served as conventional declaration s o f religious alle­ giance o f a sym bolic kind. T h u s Paul: “ If with your m outh you con fess Kurios lesoiis and believe in your heart that Ciod h as raised him from the d ead , you will be sav ed ” (R om an s 10:9) and in another p assage (1 C orin th ian s 12:3) Paul co n trasts this co n fession al form ula o f affirm ation with its op p o site, th at o f denial or denunciation (Kurios Iesous v. Anatheina Iesous , “ C ursed be J e s u s ” ), declaring that only the form er can be spoken by the H oly Spirit, i.e. that a Christian should be unable to deny C h rist. T here is som e evidence that suspected m em bers o f the Christian sect were tested by being ask ed to m ake ju st such a form al denial (L am p e, 1 976, p. 5 4 ; K elly, 197 2 , p. 15) and in the g osp els the story o f Peter’s thrice denial o f C h rist assum es its significance again st the im portance o f thus “ con fessin g C h rist” in the early church (and indeed thereafter). Belief in this con text becom es then a badge, a sym bol, som eth ing that is explicitly affirm ed where the act o f affirm ation has its ow n functional value. Such form ulary statem ents were not however creeds in the usual sense o f extended d eclara­ tion s o f belief. T hese were to develop in the period up to the fourth century. They em erged in the first place in the con text o f baptism and then, it w ould seem , began to be used a s state­ m ents o f the received teaching first for regional co n gregation s and then, in a m ore self­ con sciously developed form , a s the conciliar creeds prepared for and affirm ed by councils representing the whole church or large sections o f it. The fact that the earliest creeds were b aptism al (on this all authorities are agreed) leads us to note the im portan t post-Easter developm ent o f this rite, which acts a s a ritual coun terpart to the “ believing” we have already spoken o f as denoting conversion. Je su s’s reli­ g io u s career w as initiated by baptism from St. Jo h n the B aptist but he him self did not baptize people an d (except in the case o f Je su s) the

101

baptism o f St. Jo h n w as specifically a “ baptism o f repentance” (i.e. a cleansing rite) and not an initiation. (St. Jo h n ’s m essage w as sim ilarly one o f repentance, not o f belief.) From the time o f the early church, how ever, baptism cam e to be used to m ark the transition to m em bership that is so characteristic o f C hristianity and which in this clear-cut, boun dary-m arkin g way is absen t from all other w orld religions. Baptism and belief have parallels in other w ays to o , for baptism re-enacts sym bolically the basic postu late o f the belief: as C hrist died and rose again so too (it is held) the person being baptized dies and rises again “ in C h rist” . Baptism acts therefore not only a s a rite of p assage for individual C h ristian s but also as the act by which the church, identified with the risen C h rist, is perpetually re-constituted. For it should be understood that the church, although pre-figured in C h rist’s life, w as not in fact founded until after his death and resurrec­ tion by those w ho “ believed” , i.e. accepted the kerygma. (On this point see K iing, 1968, pp. 7 0 -9 .) “ Belief” here has theologically and sociologically a critical function in establishing the organ ic relationsh ip between C h rist, the risen L o rd , and the church as the com m unity w ho believe in his resurrection and in this way perpetuate it. Hence the im portance o f conver­ sion (believing) a s a break, a p assage from the old life to the new life, a kind o f resurrection, co m p arab le for the com m unity o f believers to the p a ssa g e through the Red Sea for the Isra­ elites (the O ld T estam en t an alogu e for C h ris­ tian baptism ). All this is im plicit in Paul’s “ one L o rd , one belief, one b ap tism ” . In their relation to baptism the earliest creeds had a dual function: first, and as part o f the ritual, can didates w ere required to respond affirm atively to certain questions ab o u t their belief put to them; secon d, and by exten sion o f the first, statem ents o f belief in the form o f a condensed, con tin u ous d eclara­ tion w ere used for the instruction o f the can ­ didate. (Here and throughout my accoun t o f the creeds I rely heavily on Kelly, 1972.) It w as the latter form that w as to be ad opted by the later con ciliar creeds but it w ould seem that even before th at happened their use w as influ­ enced by the recognition th at the received teaching they em bodied should be uniform ,

102

M ALCOLM RUEL

subject to w hat Irenaeus in the second century defined as a com m on Urule o f truth ” , and such credal declaration s grew in length and elaboraten ess because o f this. Em erging a s “ a by-product o f the C h urch ’s fully developed catechetical system ” (Kelly, 1 9 7 2 , p. 64) such creeds were in fact ancillary to the in terroga­ tory baptism al creeds. T h u s, a fourth-century treatise recalls and com m ents on the questions asked at baptism that date in this form from at least the second century: You were questioned, “ Dost thou believe in Clod the Father alm ighty?” You said, “ I believe” , and were immersed, that is were buried. Again you were asked, “ Dost thou believe in our Lord Jesus Christ and His cro ss?” You said, “ I believe” , and were immersed. Thus you were buried along with Christ; for he who is buried along with Christ rises again with Him. A third time you were asked, “ D ost thou believe also in the Holy Spirit?” You said, “ 1 believe” , and a third time were immersed, so that your threefold confes­ sion wiped out the manifold failings o f your earlier life. {De Sacramentis 2, 7 quoted in Kelly, 1972, p. 37) The triadic structure o f this interrogatory form w as carried over into the declaratory b ap tis­ mal creeds and thence all the later creeds. The baptism al creeds sum m arized the received teaching, but their local use in the widely scattered Christian com m unities, headed as each w as by their bish op, w as subject to variation and reform ulation. By the end o f the third century there is evidence o f certain baptism al creeds being cited to test the acceptability o f the teaching o f a local co m ­ m unity, which is to say the teaching o f the bishop. There w as here a shift in the use o f creeds which w as critical. It w as no longer the catechum en’s belief rhat w as at issue (and thus individual m em bership o f the C h ristian co m ­ munity) but rather the orth od oxy o f the bishop (and thus his and his con gregation ’s valid m em bership o f the Christian body). The m ajor, decisive step w as taken with the Council o f N icaea (3 2 5 ) when the assem bled bishops were asked to accept a statem ent o f teaching, which w as set ou t in the form o f a declaratory creed. (This w as not the “ N icene creed” o f the

prayer bo ok s but it laid the basis for it.) A m ajor preoccu pation in the draftin g o f this statem ent w as to exclude the teaching o f Arius and his follow ers w ho (as it happened) had them selves draw n up a creed-like sum m ary o f their position (Kelly, 1972, p. 2 0 6 ). T w o hundred and eighteen out o f 2 2 0 bishops attending the C ouncil did sign their acceptance and an indication o f the change in function o f this, the first o f the conciliar creeds, is the fact that the docum ent draw n up states not only w hat is the received belief, but also w hat is not: anathema is pronounced on those w ho hold certain propo sitio n s (i.e. those held by Arius and his supporters). “ Belief” now com es to define, not merely the Christian from the non-Christian (rhe believer from the n on ­ believer), but the true Christian from the false (the true believer from the heretic). M oreover, the latter function assum es an organized authority: bish ops in council and not just in their sees. T w o im portan t circum stances are associated with the C ouncil o f N icaea (Kelly, 1972; Ch adw ick, 1967). The first is the patron age o f C onstantine, w ho had recently adopted the Christian cause and w ho w as concerned to bring the scattered C hristian com m unities into som e kind o f com m on organization . N icaea w as the first o f the church councils and its establishm ent o f an overall church authority is evidenced not only by the bish ops’ form al acceptance o f the creed but also by their agreem ent to a num ber o f other liturgical and disciplinary m easures (the latter concerning not least the actions o f bishops). The second circum stance has already been m entioned: the teachings o f A rius w ho, in em phasizing the absolute perfection o f the G o d h ead , w as led to accord a low er, unequal place to C hrist. The Council o f N icaea did not resolve the A rian controversy (nor indeed the underlying issue o f how to interpret a trinitarian Ciod) and the fluctuating fortunes o f the Arian and antiA rian (or N icene) cam p s dom inated the church until the C ouncil and creed o f C on stantin ople (381) form ulated w hat w as to becom e the basic doctrine o f the Trinity. Any account o f this period (e.g. C h ad w ick , 19 6 7 , C h apter 9) m akes it clear th at, w hatever the m erits o f the intellectual issue, com m unity loyalties and

CHRISTIANS AS BELIEVERS

identities were also very closely involved: the m ajor cleavage between G reek E ast and Latin W est; the dom inance o f certain key bishoprics and their sees; the fortunes o f individual bishops w ho were prom oted or ousted acco rd ­ ing to their spiritual (and hum an) loyalties; the relationship o f the church to rhe foundering R om an em pire; all have a part to play in the story. O ut o f this time and ou t o f this general debate also (but not directly out o f A rianism ) emerged the one doctrinal and credal point that separates the R om an from the O rth od ox churches, the West from the E ast. T his co n ­ cerns the phrase filioque , “ from the S o n ” , which found its w ay into the W estern creeds after the C on stan tin opolitan creed and has been consistently rejected by the Eastern churches. The issue bears, like A rianism , on rhe relative statu s o f G o d the Father and the Son: the W estern church, concerned for their equality o f statu s, has com e to hold that the H oly Spirit em anates equally “ from the Father and from the Son ” w hilst the O rth odox churches hold that the phrase both is an inter­ polation and m akes little theological sense (cf. W are, 1964, pp. 5 8 -6 0 ). In all these issues belief as doctrine has becom e em bedded in the authority-strucrure o f the church. The conciliar creeds did not replace the b ap ­ tism al creeds, nor were they intended to do so. The so-called “ A th an asian ” creed (com posed in Latin and unrecognized as a statem ent o f belief in the East) is really a hymn that uses the credal form (and, uniquely, em bodies the anathem as also) to m ake an act o f w orship. The A p ostles’ creed entered the liturgical trad i­ tion by a later and yet different route; 1 return to it below . Such developm ents in the credal form were paralleled by an ever-extending use o f creeds in the liturgy. A lw ays im portant in baptism , creeds were later ad op ted for general use in the eucharist, first in the E ast (from at least the sixth century) and later in the West (form ally, from the early eleventh century; Ju n gm an n , 1959, pp. 2 9 5 -8 ). T he O rth o d ox, R om an , A nglican and (in lesser m easure) N o n ­ con form ist churches all continue to give central place to the singing or saying o f rhe creed in their services. The point to m ake here is that this perform ance o f the creeds is as com plex, sym bolic and condensed an act o f ritual as any

103

other liturgical act and is consequently as much subject to the categories developed, for exam ple, by T urner (1967) for the analysis o f ritual sym bolism . (O n the variable m eaning o f the creed for different person s saying it see the chapter on the creeds in D octrine C om m is­ sion o f the Church o f England, 1976.) ( ...] There is no sim ple form ula to describe what happened at the R eform ation; the processes that are distinguished by that term were already in train before the period usually covered by it and they certainly continued beyond it. In describing the nature and effects o f the R efor­ m ation it becom es necessary to resort to the term “ faith ” which, although by no m eans new, acquired an extra dim ension o f signifi­ cance that gave it (like the pisteuo w ord-group before it) a quasi-technical use. Deriving from the Latin fides (itself cognate with the Greek pistis , which it norm ally translated in the Bible), “ faith ” carries by sem antic origin very m uch the sam e range o f m eanings as did origi­ nally “ belief” /p/sr/s, that is, trust or confidence; an d, as alternative tran slation s o f the Bible show , the tw o w ords in English often serve as synonym s, with “ faith ” becom ing religiously the m ore specialized (see OF'D under “ Belief” ); the fact that “ faith ” has no verbal form has also produced the asym m etrical situation in which the verb “ to believe” is often matched with rhe noun “ faith ” . T his linguistic variation is m ore confusing how ever than the broad sem antic situation w ould seem to w arrant. A ugustine in his treatise on the creed, De Fide et Sym bolo , uses the term faith Ifides precisely in the sense o f orth o d ox belief a s it is expressed in the creeds, “ the C atholic faith ” a s the received teaching o f the church. We m ay co n trast this sense with the m eaning the w ord m ore readily has in m any Protestant w ritings, where w hat is at issue is less the substance o f belief (although that is not unim portant) than how such belief (often expressed as the G ospel o r the W ord) has been subjectively ap p ro p ri­ ated. The difference between “ faith ” in the tw o senses - “ the C ath olic faith ” and a per­ son ’s ow n “ faith ” - is thus less a m atter o f the difference between church authority and indi­ vidual reason (which anyw ay m uch exercised m edieval theologians) than o f the difference

104

M ALCOLM RUEL

between belief as declaration an d belief as com m itm ent. T h is extra dim ension to the notion o f belief had organ ization al im plica­ tions but its im m ediate thrust w as psychological. Luther’s role in helping to effect this shift w as crucial but com plex. Som e w ould see Luther as sim ply re-expressing the personal com m itm ent alread y im plicit in the Pauline explication o f belief (and Paul’s Epistles were o f vital im portance in Luther’s ow n spiritual biograph y); others w ould see his accoun t o f belief as an appeal ultim ately to unreason, a faith that is given by G o d ’s grace, no-one can know how . But how ever one interprets Luther’s life and w ork, a ch aracteristic them e o f both is his stress on the inw ard totality o f Christian belief, the faith o f the believer. T h u s his som om cntous reaction to the sp u rio u s h aw king o f indulgences (the very first o f the Ninety-Five T heses sets the tone by insisting that when “ Je su s C hrist said ‘ R e p e n t. . he m eant that the w hole life o f believers should be one of penitence” ); thus his co n trast between the (external) T h eology o f G lory (that o f the Church to date) and the (inw ardly exp eri­ enced) T heology o f the C ro ss (Luth er’s ow n preferred theology); thus his distinction between law , a m atter o f ou tw ard p erfor­ m ance, and the G o sp el, W ord o r G race, that w orks inw ardly (the point is m ade by one o f the Fleidelberg theses: “ The law say s ‘ D o this’ and it is never done. G race say s ‘ Believe th is’ and everything is d on e” ); and th us, not least, the governing principle o f justification by faith, which while p roffering intellectually an o b jec­ tive view o f G o d ’s grace in fact locates faith in the intensity and totality o f a p erso n ’s exp eri­ ence o f it. (O n these vario u s points see R upp, 1 9 7 5 ; von Loew enich, 1 9 7 6 ; R upp and Drew ery, 1970.) Luther stan ds in history not only as a thinker and w riter but a lso as a p a r a ­ digm o f the person w ho achieves a fully real­ ized belief only after an intense inw ard struggle, who p ossesses belief by being p ossessed by it: such is the “ faith ” that com es from w ithout but signifies a subjective transition from disorganized dou bt to clarity, conviction and a certain kind o f personal freedom . T w o recent representations o f Luther that have som ething o f this parad igm atic aura

provide a useful bridge to the present. F’rikson’s psych oanalytic study (1959) grants all o f Luther’s theological im portance (e.g. p. 2 5 0 ) but focuses especially on the concern Luther show ed in his ow n biographical stru g­ gle with the intensity and expressiveness o f the experience o f belief. T he title o f the key chapter in the bo ok , “ The m eaning o f ‘ m eaning it’ ” , states the issue succinctly: as Erikson w rites, “ T o Luther, the preaching and the praying m an, the m easure in depth o f the perceived presence o f the W ord w as the reaction with a total affect which leaves no dou bt rhat one ‘ m eans it’ ” (1 9 5 9 , p. 2 03). But, Frikson goes on to point ou t, w hat Luther had to fight to secure, com es to us as the easy convention o f our age: [Luther’s) formulations, once revolutionary, are the commonplaces of today’s pulpits. They are the bases o f that most inflated o f all oratorical currency, credal protestations in church and lecture hall, in political propa­ ganda and oral advertisement: the protesta­ tion, made to order of the occasion, that truth is only that which one means with one’s whole being, and lives every moment. We, the heirs of Protestantism, have made convention and pretence out of the very sound o f meaning it. . . . (ibid.) Jo h n O sb o rn e ’s play (1 9 6 1 ), which is based on F’rik son ’s study, points a different but related m essage. If - to p araph rase Erikson we are all sincere believers now , O sb o rn e’s Luther hints that we should bew are o f being to o convinced, to o secure in our righteous co n ­ victions. The point is m ade m ost clearly in the tw o final scenes, m ost notably when “ M artin ” adm its to his lon gstandin g friend and advisor (Staupitz) that his delay under interrogation at the Diet o f W orm s (when his career and even his life were in jeopardy and he is said by trad i­ tion to have declared “ Here I stand. I can do no oth er” ) w as in fact the result o f his uncer­ tainty, o f his doubt. Left alone on the stage, he con fesses “ Oh L o rd , I believe. 1 d o believe. O nly help my unbelief.” In our age we seem then to believe in belief, but with not to o exclusive a conviction. This note o f diffuse belief is struck quite rem ark­ ably by tw o b o o k s that are in their separate

CHRISTIANS AS BELIEVERS

w ay s both revealing and authoritative. The Culture o f Unbelief publishes “ studies and proceedings from the First International Sym ­ posiu m on Belief held in R om e, M arch 2 2 - 2 7 , 1 9 6 9 ” under the sp o n so rsh ip , am o n gst others, o f the V atican Secretariat for N on-believers, which w as set up follow ing V atican II. The Sym posium gathered a highly distinguished gro u p o f social scientists, m ainly sociolo gists, specializin g in the study o f religion, w ithout regard to their personal religious position , together with and including a num ber o f n otable churchm en. It w as not, how ever, the intention o f the Sym posium to produce “ d ia ­ logu e” - we are told that “ in a num ber o f cases the organ izers w ould have been hard put to identify a participant as either a ‘believer’ or a ‘ non-believer’ ” - but rather that o f “ study in prep aration for d ia lo g u e ” (C ap o rale and Cirumelli, 1971, p. ix). O ne m ust break o ff at this point to com m ent on the crucial w ord “ unbelief” . The idea itself is as old as its p o si­ tive coun terpart “ belief” : apistia m eans lack of trust, non-confidence in its general sense and specifically lack o f trust in Ciod or non-belief in C h rist in a religious con text. T hus the rem ark o f O sb o rn e's Luther quoted above recalls the outburst o f the father o f the co n ­ vulsed boy in M ark 9 :2 4 , “ Im m ediately the father o f the child cried out an d said , ‘I believe; help my un belief!’ ” There is then nothing new, or especially con tem p orary, in the idea o f unbelief, although it has no d ou bt a contem ­ p orary tw ist. Pope Paul in his ad d ress to the Sym posium certainly knew w hat he m eant the rejection o f the C h ristian religion (see p. 3 0 2 , op. cit.). Yet one o f the features o f the Sym posium is the very indeterm inate value th at the term s “ belief” and “ unbelief” com e to acquire in it. Berger in his forew ord refers to this “ am biguity in the definition o f the p rob lem ” and notes the “ rather rem ark able” fact that “ the theoretical position p ap ers, each in its ow n w ay, tend to deny the very existence o f the phenom enon under scrutiny (i.e. unbe­ lief]” (ibid., p. xiii). T he rem ark covers papers by both Luckm ann and P arson s but is nowhere so pertinent as in its reference to Bcllah’s p ap er, “ The historical backgroun d o f unbe­ lief” . Starting with an accoun t o f the Greek con cept, the paper sets “ belief” in an institu­

105

tion al, church context and asso ciate s it with “ an effort to m aintain au th o rity ” on the part o f the church, “ p art o f a w hole hierarchical w ay o f thinking ab o u t social co n tro l” (p. 44). Belief (which in Bellah’ s discu ssion has becom e very rapidly intellectualized, a m atter of doctrine and dogm a) is thus distinguished from religion, the form er tied to an institu­ tional (church) structure, the latter m ore dif­ fusely present in shared values. A gainst then this narrow ly defined, historical view o f “ belief” Bellah co n trasts w h at he sees as an “ em ergent religion o f h um anity” (p. 50). N ev­ ertheless, by the end o f the paper he is so carried aw ay in his enthusiasm for the latter that all distinctions have disap p eared and C h ristian s and non-C hristians, belief and reli­ gion , the Church and hum anity are all rolling together in a single g lo rio u s banner: The modern world is as alive with religious possibility as any epoch in human history. It is no longer possible to divide mankind into believers and non-believers. All believe som e­ thing and the lukewarm and those o f little faith are to be found inside as well as outside the ch urches.. . . Christians, along with other men, are called to build the boundaryless community, the body o f man identified with the body o f Christ, although all men are free to symbolize it in their own way. (p. 52) There can be no unbelief in such a w orld and the w ord “ believe” has becom e so generalized a s ro have lost m ost o f its content. Like Forster’s “ W hat I believe” , it is a muffled (but not so m uted) cry. Christian Believing (1 9 7 6 ), a s a b o ok , is an altogeth er less gran diose affair. It is the report o f the D octrine C om m ission o f the Church of En gland, w ho were ask ed by the A rchbishop o f C an terbury to exam ine “ the nature o f the C h ristian faith and its expression in holy scrip­ ture and cre ed s” . The very striking thing about the report is its lack o f d ogm atism , tolerance o f even op p o sed view s, concern to respect both tradition and the right to criticize and re­ evalu ate it. The em ph asis once again is on existential belief, believing as the adventure of faith rather than belief as a body o f doctrine. The m ost singular feature o f the report is, how ever, the fact that over h alf o f it consists

106

MALCOLM RUEL

o f eight individual essays by m em bers o f the C o m m issio n , each in effect outlining his per­ sonal view o f the nature o f rhe Christian faith. The im plication throughout is that each “ believer” m ust find his ow n w ay, respecting traditional truths but respecting also other people’ s right to hold different view s from his ow n. Belief as doctrine has alm ost becom e the honest opinion o f anyone w ho declares him self to be a C hristian. T here is both continuity and change in the notion o f belief that I have sketched above a t four ph ases o f its history: trust becom e conviction ab o u t an event (the “ Christ-event” o f history); becom e an initiatory d eclara­ tion; becom e a corporately declared o rth o ­ d ox y ; becom e an inw ardly organizing experience; becom e values com m on to all men (even though different). Yet th roughout the concept rem ains central to C hristianity, which is clear from the w ay it reflects so much o f the C hurch’ s organ ization al and intellectual history. M oreover m uch o f the w ord ’s m eaning in non-C hristian use can only be draw n from the p articular significance that it has acquired in C hristianity - else why attach any im p o r­ tance at all to having or not having beliefs? In the rem ainder o f this chapter I w rite Belief with a capital letter ro signify this m ulti­ layered, com p lex yet condensed range o f use and m eaning that the concept has acquired in this its long career in C hristianity. N egative dem on stration s are alw ays difficult and usually lengthy and for this reason I argue the com parative case only sum m arily. I find little evidence that there is anything equivalent to Christian Belief in other w orld religions although there are other com p arable o rg an iz­ ing or nodal concepts. The co n trast is greatest with Ju d aism (as one w ould expect, histori­ cally an d sociologically), the sim ilarities (as one w ould a lso expect) closest with Islam . The teaching o f the law , the Torah , stan ds at the centre o f Ju d aism in a w ay functionally co m ­ parable - but with different practical im plica­ tions - to Belief in C h ristianity; and there are com p arable differences in rhe identity-m arkers o f the tw o religions. W ith Islam the parallel is closer: the first o f the “ five p illars” o f Islam w itness to G o d and his prophet - com es close

to being a credo; there is som e concern for orth odoxy o f belief and there are even form al creeds (although the ordinary M uslim is unlikely to know them ). Islam - subm ission to rhe one Ciod - can be identified with having belief, iman\ a M uslim is also a believer, m u m in . Yet, as these w o rd s’ shared root with the H ebrew ’m « testifies, their reference is essentially to the quality o f a relationship, that o f keeping faith, having trust. C o rresp on d ­ ingly, it is less the content o f belief that has becom e elaborated in Islam than the duties o f relationship: the practice o f ritual, the fo llo w ­ ing o f Islam ic cu stom , the observance o f Islam ic law . Ciombrich, w riting ab o u t Buddhism , stru g­ gles aw kw ardly for tw o p ages to find an equiv­ alent to the verb “ to believe” or “ to believe in” before m oving directly to the term ( “ best not translated at a ll” ) dharma (1 9 7 1 , p. 60). For H induism the p arallels are even m ore indirect and fragm ented. The absence o f any self-conscious credal or doctrinal com pon en t form s a com m onplace observation o f m ost, if not all, traditional or com m unity religions. Should one attem pt to distinguish for them , as one can perhaps for other w orld religions, any organizing concept, com parable to that o f Belief in C hristianity? I can answ er only for the tw o cultures I know at first-hand, and in both cases I find the q u es­ tion relevant and revealing. For Banyang the idea o f truth (tetup) has a central significance, in part as an attribute o f G od bur m ore particularly a s a touchstone in people’ s rela­ tionships, where the possibility o f duplicity is obsessively elaborated by Banyang w itchcraft beliefs. On a less cognitive level, the Kuria category o f inyangi , which I am forced to tran slate sum m arily a s “ ritu al” , although it m eans som ething less, and m ore, serves also as touchstone orderin g their own relationships (kin and generational) so a s to accord with w hat K uria see a s the natural principles o f grow th. Both con cepts have content but also operate functionally in the organization and determ ination o f relationships. I---1 Clearly it is not possible, nor even desirable, to limit the w ord “ belief” to its specifically

CHRISTIANS AS BELIEVERS

Ch ristian use. Yet at the sam e time we should be clear that it has a Christian use and that this use m ust affect its co n n otation s in co n ­ texts other than C h ristian . It is surely plain naive to pluck the w ord from the linguistic planisphere (the Ofc’D G uide to the G alax y conveniently at hand) and to use it then as though it were a given, som ething that just happened to be aroun d, which had inciden­ tally been m ade som e use o f by C hristians. Let me be clear: in ordinary speech there are many uses o f the w ord “ believe” that are straigh tfor­ w ard and unam biguous. O n the whole these have a relatively w eak set o f con notation s, im plying usually (of oneself) presupposition or expectation, or (of others) assum ption . There are ad van tages, as N eedh am indicates, in avoidin g “ believe” altogether, but the w ord is current English and in this its w eak sense it is not likely to be m isun derstood. It is when the w ord is given a stron g sense that it m ay well m islead: for exam p le, when it form s part o f a definition or categorization or is used in posin g a problem . Here I w ould argue that it is alm ost im possible not to draw on co n n o ta­ tions from its Christian use. M oreover, these con n otation s, con textually tran sposed , create false assum p tion s that then lead to fallacies. I speak o f these contextually tran sposed assu m p ­ tions as “ shadow fallacies” and for the nonC hristian use o f the term “ belief” identify four: (1) T h at belief is central to all religions the sam e w ay a s it is to Christianity. T h at this is a fallacy is the m ajor argum ent o f my paper and Sections I and II o f the paper are co n ­ cerned with its dem on stration . It is, however, very easy for a W estern w riter to slip from talkin g ab o u t religion to talkin g a b o u t C h ris­ tianity, and back again , w ithout clear distinc­ tion. For exam ple, N eedh am in considering spiritual com m itm ent as a possible criterion o f religious belief (1 9 7 2 , p. 8 6 -9 ) ap p ears to d o ju st this: he m oves from a specifically Christian view o f belief a s com m itm ent to C hrist (ibid., p. 86), then argues that other people can be sim ilarly “ com m itted ” to p articular enterprises or person s (pp. 8 7 -8 ) and then concludes that there is no discrim inable difference between religious and non-religious com m itm ent (ibid.,

107

p. 88). Hut it is not com m itm ent per se that identifies C hristianity but com m itm ent to Christ: Christianity cannot be treated as a type-case o f religious com m itm ent; it is a spe­ cific case o f a particular com m itm ent to a single historical person. The sam e fallacy can be detected I think in a som ew h at different w ay in M artin South w old’s thoughtful paper on “ R eligious belief” . M uch o f South w old’s critical com m entary is highly relevant to any d iscu ssion o f religion bur why focus the discu s­ sion on the nature o f belief? And does not the fram ing o f the question thus itself determine the kind o f answ er that will be obtained? N am ely, that “ basic religious tenets are 1) em pirically indeterm inate 2) ax io m atic 3) sym bolic, and 4) collective” (1 9 7 9 , p. 633). Ch ristian Relief is historically and conceptu­ ally m ore precise in its references than this, but take belief (the shadow idea) to apply to other religions (as one might take Ju d aic torah or a sh adow extension o f it) and one m ay well find rhe correspondence to be indeterm inate and indirect (sym bolic). (2) T h at the belief o f a person or a people form s the ground o f his or their behaviour and can be cited therefore a s a sufficient e x p la n a ­ tion for it. For an exam ple o f this fallacy I w ould draw on my ow n teaching experience. A long no dou bt with m any others, I regularly set m y first-year students in social an th rop o­ logy an essay on Z an de w itchcraft usually in the first few w eeks o f the course. O ne topic I in com m only use runs: since Z an de oracles m ust often give false answ ers, why then do Azande continue to believe in w itchcraft? an d, all being well, the essay that is returned duly rehearses Evans-Pritchard’s situational an aly­ sis o f Z an de reasoning. Hut, not infrequently, all is not well and my (I think now m isguided) w eak use o f “ believe” is turned into a strong use: the evidence in the book for individual Z an d e scepticism is ignored, as is much else, to present A zande with such unalterable firm ­ ness o f conviction a s w ould m ake a Calvinist jealous. N o r does the m atter stop there, for m ake belief fundam ental to the behaviour o f a person or a people and the issue in relation to others is then relativized. T h at is w hat Azande believe: finish: there can be no further discu s­ sion o f the substance o f their belief in term s

108

M ALCOLM RUEL

that d o not bracket it o ff as som ething to do with them (rather than their experience o f the w orld) and thus hinder its discussion in co m ­ parative term s (i.e. w hat we to o experience o f the w orld). T he in sidiousn ess o f this process o f relativization m ust be em phasized. “ We all have our beliefs: all peoples have their beliefs.” It is a w ay o f setting people into cultural com partm ents. (3) T h at belief is fundam entally an interior state, a psych ological condition. The fallacy once again is to tran spo se w hat som e have em phasized as the inw ardness o f C h ristian belief (faith) to the non-C hristian con text and use o f the w ord. I w ould argue that N eedham does just this, to the enorm ous detrim ent o f his d iscu ssion , in ad o p tin g E vans-Pritchard’s assertion concerning N uer religion and gener­ alizing it to all belief. Yet as Alan Ryan points ou t (1 9 7 3 ), over-insistence on the privacy o f w hat goes on in people’ s m inds d istracts from the prim ary task o f con struin g the sense or reality o f w hat it is they believe, and it is one o f the skills o f an th ropology to do precisely this by con textual explication . Is the m eaning o f w ords interior to the w ord s? T h at belief w as a psychological state for Luther does not imply that it m ust be a psychological state for every­ one, or even rhat it w as only that for Luther. It does not m ake much sense to call the belief subscribed to by the C ouncil o f N icaea a psy­ chological state. South w old ’s criticism o f Leach on the “ inner p sy ch ological” aspect o f belief is here m uch to the point (1 9 7 9 , p. 6 3 1 ). We should recall that the notion o f “ belief” gain s much o f its significance in C hristianity from

the first person use o f the verb: I b e lie v e .. . . But an th ropologists necessarily use the verb in the third person: A zande b e lie v e .. . . T o assum e that our presentation o f their belief carries the sam e force as though they said “ We believe . . is to m isunderstand the sem antic con jugation o f the verb and to tran sp o se C h ristian assu m p ­ tions un w arran tably (cf. South w old, 19 7 9 , p. 6 3 0 ; Sm ith, 1977, p. 52 et seq.). (4) T h at the determ ination o f belief is m ore im portant than the determ ination o f the status o f w hat it is th at is the ob ject o f the belief. In C hristianity to be a believer is to acknow ledge an allegiance and to declare an identity: the person does not alw ays have to be clear about the full content o f his belief. The sam e circum ­ stance tran spo sed to non-C hristian religions m akes much less sense. T o say that a people “ believe” in this, that or other ab straction (w itchcraft, Ciod, spirits o f the an cestors, hum anism ) tends to bracket o ff ideas that they hold ab o u t the w orld from the w orld itself, treating their “ beliefs” a s peculiar to them , a badge o f their distinctiveness, and all know l­ edge o f the w orld our privileged m onopoly. T he shadow cast by the Christian respect-forbelief obscures w h at really it is that people see o r think they see. If we are to converse with each other (and I assum e that social an th ro p o ­ logy is a kind o f con versation between cultures) we need fewer such sh ad ow s, cast by the co n ­ textual tran sposition o f in appropriate categ o ­ ries, and a clearer, steadier gaze on to the w orld we share. O r, at least, a clearer ad m is­ sion as to w hat we think, or assum e or presu p­ pose, or understand ab o u t the w orld we share.

REFERENCES

Theologisches Worterbuch zum Neuen Testament. Lon don (A dam and C h arles

Doctrine C om m ission o f the Church o f England, 1976: Christian Believing. London (SPCK ). F’rikson, E. H ., 1959: Young Man Luther. London (Faber and Faber). FvVans-Pritchard, E. F'., 1937: Witchcraft,

Black). C ap o rale, R. an d G rum elli, A. (eds.) 1971: The Culture o f Unbelief. Berkeley (Univer­ sity o f C alifo rn ia Press). C h adw ick , H ., 1967: The Early Church. The Pelican H istory o f the Church V ol. 1. H arm on d sw orth (Penguin Books).

O x fo rd (C larendon). Forster, FL. M ., 1939: What I Believe. H ogarth Sixpenny Pam phlets 1. London (H ogarth Press). G om brich , R. F., 19 7 1 : Precept and Practice. O x fo rd (C laren don ).

Bultm ann, R. an d W eiser, A., 1961: Faith. Bible Key W ords from G . KittePs

Oracles and M agic am ong the Azande.

CHRISTIANS AS BELIEVERS

H atch , W. H. P., 1917: The Pauline Idea o f Faith. H arvard T h eological Studies Vol. 2. C am bridge, M ass. (H arvard University Press). Ju n gm an n , J. A ., 1959: The M ass of the Roman Rite. T ran s, from the G erm an by F. A. Brunner and C . K. Riepe. Lon don (Burns and O ates). Kelly, J . N . I)., 197 2 : Early Christian Creeds. London (Longm an). (3rd edition). K iing, H ., 1968: The Church. T ran slated from rhe G erm an by R. and R. O ckenden. Lx>ndon (Burns and O ates). L am p e, G . W. H ., 1976: The origin s o f the creeds. In Christian Believing , Doctrine C om m ission o f the Church o f England. London (SPCK ). Loew enich, W. von., 197 6 : Luther's Theology o f the Cross. Belfast (C hristian Jo u rn als Ltd). M ichel, O ., 1975: Faith, Persuade, Belief, Unbelief. In The New International Dictionary o f New Testament Theology , ed. C . Brow n. Exeter (Paternoster Press). M ou le, C. F. D ., unpublished: Belief an d trust in the N ew T estam en t vocabu lary. Paper given to the C am b rid ge “ D " Society, m anuscript.

109

N eed h am , R ., 1972: Belief Language and Experience . O x fo rd (Blackw ell). O sb o rn e, J ., 1961: Luther. London (Faber and Faber). R u pp, E. G ., 1975: Luther and the G erm an R eform ation to 1529. In The Reformation (G . R. Elton, ed.). The N ew C am bridge M odern H istory Vol. 2. C am bridge (C am bridge University Press). R u p p , F^. G. and D rew ery, B., 1970: Martin Luther. D ocum ents o f M odern H istory series. L on don (Edw ard A rnold). R yan , A., 1973: By-ways of belief. Review' o f N eedh am (1 9 7 2 ). New Society , 11th Jan u ary . Sm ith, W. C ., 1978 (1 9 6 2 ): The M eaning and End o f Religion. L on don (SPCK). ------, 1977: Belief and History. C harlottesville (University Press Virginia). ------, 1979: Eaith and Belief. Princeton, N . J. (Princeton University Press). South w old, M ., 1979: R eligious belief. Man 14(4) Decem ber. T urn er, V., 1967: The Forest o f Symbols. Ithaca (Cornell University Press). W are, T ., 1964: The O rthodox Church. H arm on dsw orth (Penguin).

9

The Construction of Religion as an Anthropological Category Talal Asad

Talal Asad is an anthropologist trained in the United Kingdom and currently dis­ tinguished professor at the graduate center of the City University of New York. Asad's provocative account, like that of Ruel, grounds its critique of anthropological predecessors in its account of the history of Christianity, but here the critique is even stronger, challenging not only the category of belief but of religion itself. Asad rejects essentialist definitions of religion, arguing th at the very idea of such a d efi­ nition "is itself the historical product of discursive processes," i.e., w ithin th e cul­ tural location of secular modernity. Thus his argum ent is not merely about the use of language but advocates an entirely different conceptual and m ethodologi­ cal fram ew ork from that developed in the essay by G eertz. Asad's account is indicative of a shift aw ay from a sym­ bolic anthropology tow ard a poststruc­ turalist one that is more centrally

concerned with pow er and discipline and w ith the w ay that religious subjects (i.e., practitioners) are formed. Indeed, his essay forms one of the major state­ ments from w ithin the anthropology of religion (there have been many critiques from materialist anthropologists outside the subfield) to offer an alternative to the symbolic approach. The latter, broadly defined, is characteristic not only of Geertz but of many of the authors w ho follow in this anthology. Asad's account also demonstrates the effects a change in perspective can bring. He begins w ith a Muslim assump* tion that religion and power cannot be separated. In addition, he draws ex­ plicitly on Vygotsky and implicitly on Foucault and Bourdieu in this bracing account. But even more interestingly, the argum ent is worked out in part through Asad's ow n historical anthropo­ logical w ork on medieval European Christianity. Unfortunately, for reasons

From “ Anthropological Conceptions of Religion: Reflections on Geertz,” in Man, New Series, Vol. 18, No. 2 (June, 1983), pp. 237-59.

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CATEGORY

of space, I have had to exclude many of Asad's learned footnotes th at help to bring this home. The essay reprinted here is follow ed in his G enealogies o f Religion (Asad 1993) by "Tow ard a Genealogy of the Concept of Ritual," which pursues the m edieval angle more directly. The book also contains a num ber of im portant essays on Islam

In much nineteenth-century evolutionary thought, religion w as considered to be an early hum an condition from which m odern law , science, and politics em erged and becam e d etach ed.1 In this [twentieth] century m ost an th ropologists have ab an d on ed Victorian evolutionary ideas, and m any have challenged the rationalist notion that religion is sim ply a prim itive and therefore outm oded form o f the institutions we now encounter in truer form (law , politics, science) in m odern life. For these twentieth-century an th rop ologists, religion is not an archaic m ode o f scientific thinking, nor o f any other secular endeavor we value today; it is, on the contrary, a distinctive space o f hum an practice and belief which cannot be reduced to any other. From this it seem s to follow that the essence o f religion is not to be confused w ith, say, the essence o f politics, although in m any societies the tw o m ay overlap and be intertwined. In a characteristically subtle p assag e , Louis D um ont has told us that m edieval Christen­ dom w as one such com posite society: I shall take it for granted that a changc in relations entails a change in whatever is related. If throughout our history religion has developed (to a large extent, with some other influences at play) a revolution in social values and has given birth by scissiparity, as it were, to an autonom ous world o f political institu­ tions and speculations, then surely religion itself will have changed in the process. O f some important and visible changes we are all aware, but, I submit, we are not aware of the change in the very nature o f religion as lived by any given individual, say a Catholic. Every­

111

and the contem porary politics of reli­ gion. More recently Asad (2003) has published a series of trenchant essays about secularism as an integral part of modernity and its relation to the politics of Islam both in the M iddle East and in Europe. Recent discussion of his work can be found in Hirschkind and Scott (2006).

one knows that religion was formerly a matter of the group and has become a matter of the individual (in principle, and in practice at least in many environments and situations). But if we go on to assert that this change is corre­ lated with the birth o f the modern State, the proposition is not such a commonplace as the previous one. Let us go a little further: medieval-religion was a great cloak - I am thinking of the Mantle o f O ur Lady o f Mercy. Once it became an individual affair, it lost its allembracing capacity and became one among other apparently equal considerations, of which the political was the first born. Each individual may, of course, and perhaps even will, recognize religion (or philosophy), as the same all-embracing consideration as it used to be socially. Yet on the level o f social consen­ sus or ideology, the same person will switch to a different configuration of values in which autonom ous values (religious, political, etc.) are seemingly juxtaposed, much as individuals are juxtaposed in society. (1971, 32; emphasis in original) A ccording to this view , m edieval religion, per­ vading or encom passing other categories, is nevertheless analytically identifiable. It is this fact that m akes it possible to say that religion has the sam e essence today as it had in the M iddle A ges, although its social extension and function were different in the tw o epochs. Yet the insistence that religion has an au ton om ou s essence - not to be confused with the essence o f science, or o f politics, or o f com m on sense - invites us to define religion (like any essence) as a transhistorical and transcultural phenom ­ enon. It m ay be a happy accident that this effort o f defining religion converges with the

112

TA LAL ASAD

liberal dem and in our time that it be kept quite separate from politics, law , and sc ie n c e -sp a c e s in which varieties o f pow er and reason articu­ late our distinctively m odern life. T h is defini­ tion is at once p art o f a strategy (for secular liberals) o f the confinem ent, an d (for liberal Christians) o f the defense o f religion. Yet this separation o f religion from pow er is a m odern W estern norm , the p rodu ct o f a unique post-R eform ation history. The attem pt to understand M uslim traditions by insisting that in them religion and politics (tw o essences m odern society tries to keep conceptually and practically ap art) are coupled m ust, in my view, lead to failure. At its m ost du b io u s, such attem pts encourage us to take up an a priori position in which religious d iscourse in the political arena is seen as a disguise for political pow er. In w hat follow s I w ant to exam ine the w ays in which the theoretical search for an essence o f religion invites us to separate it conceptually from the d om ain o f pow er. I shall d o this by exploring a universalist definition o f religion offered by an eminent an th rop ologist: C lifford G eertz’s “ R eligion as a C u ltural System ” [reprinted in his widely acclaim ed The Inter­ pretation o f Cultures (19 7 3 )]. I stress that this is not prim arily a critical review o f G eertz’s ideas on religion - if rhat had been my aim I w ould have addressed m yself to the entire corp us o f his w ritings on religion in Indonesia and M orocco. M y intention in this chapter is to try to identify som e o f the historical shifts that have produced our concept o f religion as the concept o f a transhistorical essence - and G eertz’s article is merely my startin g point. It is part o f my basic argum ent that socially identifiable form s, precon dition s, and effects o f w hat w as regarded as religion in the m edi­ eval C hristian epoch were quite different from those so considered in m odern society. 1 w ant to get at this w ell-know n fact while trying to avoid a sim ple n om inalism . W hat we call reli­ giou s pow er w as differently distributed and had a different thrust. There were different w ays in which it created and w orked through legal institutions, different selves th at it shaped and respon ded to, and different categories o f know ledge which it authorized and m ade available. N evertheless, w hat the an th rop olo­

gist is confronted w ith, a s a consequence, is not merely an arbitrary collection o f elements and processes rhat we happen to call “ reli­ g io n .” For the entire phenom enon is to be seen in large m easure in the con text o f Christian attem pts to achieve a coherence in doctrines and practices, rules and regulations, even if that w as a state never fully attained. M y a rg u ­ ment is that there can n ot be a universal defini­ tion o f religion, not only because its constituent elem ents and relation sh ips are historically spe­ cific, but because that definition is itself the historical produ ct o f discursive processes. A universal (i.e., an th ropological) definition is, how ever, precisely w hat Geertz aim s at: A religion , he p ro p o ses, is “ (I) a system of sym bols which act to (2) establish pow erful, pervasive, and long-lasting m oo ds and m otiva­ tions in men by (3) form ulating conceptions of a general order o f existence and (4) clothing these conceptions with such an aura o f factuality that (5) the m o o d s and m otivations seem uniquely realistic” (90). In w hat follow s I shall exam ine this definition, not only in order to test its interlinked assertio ns, but also to flesh ou t the counterclaim that a transhistorical definition o f religion is not viable.

Th e C o n cep t of Sym bol as a C lu e to th e Essence of Religion Geertz sees his first task as the definition o f sym bol: “ any object, act, event, quality, or relation which serves as a vehicle for a con cep­ tio n -r h e conception is the sym bol’s ‘ m eaning’ ” (91). Bur this sim ple, clear statem ent - in which sym bol (any object, etc.) is differenti­ ated from but linked to conception (its m eaning) - is later supplem ented by others not entirely consistent with it, for it turns out that the sym bol is not an object that serves as a vehicle for a con ception, it is itself the concep­ tion. T h u s, in the statem ent “ T he num ber 6, w ritten, im agined, laid out as a row o f stones, or even punched into the program tapes o f a com puter, is a sy m b o l” (91), w hat constitutes all these diverse representations a s versions o f the sam e sym bol (“ the num ber 6 ” ) is o f course

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CA TEG O RY

a conception. Furtherm ore, Geertz som etim es seem s to suggest th at even as a conception a sym bol has an intrinsic connection with em pir­ ical events from which it is merely “ theoreti­ ca lly ” separable: “ the sym bolic dim ension o f social events is, like the psych ological, itself theoretically ab stractab le from these events as em pirical to talities” (9 1 ). At other tim es, how ever, he stresses rhe im portance o f keeping sym bols and em pirical ob jects quite separate: “ there is som ething to be said for not co n fu s­ ing our traffic with sym bols with our traffic with objects or hum an bein gs, for these latter are not in them selves sym bols, however often they m ay function a s su ch ” (92). T h u s, “ sy m b ol” is som etim es an aspect o f reality, som etim es o f its represen tation .2 T hese divergencies are sym ptom s o f the fact th at cognitive question s are m ixed up in this acco u n t with com m unicative ones, and this m ak es it difficult to inquire into the w ays in which discourse and understanding are con ­ nected in social practice. T o begin with we might say , a s a num ber o f w riters have done, rhat a sym bol is not an ob ject or event thar serves to carry a m eaning but a set o f relationships between objects or events uniquely brought together as com plexes or a s co n cep ts,3 having at once an intellectual, instrum ental, and em o­ tional significance. If we define sym bol alo n g these lines,4 a num ber o f question s can be raised ab o u t rhe con dition s that explain how such co m p lexes and concepts com e to be form ed, and in particular how their form ation is related to varieties o f practice. H a lf a century ago , V ygotsky w as able to show how the develop­ ment o f children’s intellect is dependent on the internalization o f social speech. This m eans thar the form ation o f w hat we have here called “ sy m b o ls” (com plexes, concepts) iscondirioned by the social relations in which the grow ing child is involved - by the social activities thar he or she is perm itted or encouraged or obliged to u n d e rta k e -in which other sym bols (speech and significant m ovem ents) are crucial. The co n d i­ tion s (discursive and nondiscursive) th atexp lain how sym bols com e to be con structed, and how som e o f them are establish ed as natural or authoritative as op p o se d to oth ers, then becom e an im portan t object o f an th rop ological inquiry. It m ust be stressed that this is not a m atter o f

113

urging the study o f the origin and function o f sym bols in addition to their m eaning such a distinction is not relevant here. W hat is being argued is thar the authoritative status o f represen tations/discourses is dependent on the ap p ro p riate p roduction o f other represen tation s/discourses; the tw o are intrinsi­ cally and not just tem porally connected. System s o f sym bols, say s G eertz, are also culture patterns, and they constitute “ extrinsic sources o f in fo rm ation ” (92). Extrinsic, because “ they lie outside the b oun daries o f the individ­ ual organ ism a s such in thar inter-subjective w orld o f com m on un derstandings into which all hum an individuals are b o rn ” (92). And sources o f in form ation in the sense that “ they provide a blueprint or tem plate in term s o f which processes external to them selves can be given a definite fo rm ” (92). T h u s, culture p at­ terns, we are told, m ay be thought o f as “ m odels for reality” a s well as “ m odels o f reality.” 5 T h is part o f rhe discu ssion does open up possibilities by speakin g o f m odeling: that is, it allow s for the possibility o f conceptualizing disco urses in the process o f elab o ration , m odi­ fication, testing, and so forth. U nfortunately, Geertz quickly regresses to his earlier position: “ culture patterns have an intrinsic double asp e c t,” he w rites; “ they give m eaning, that is objective conceptual form , to social and psy­ ch ological reality both by sh apin g them selves to it and by sh apin g it to th em selves” (1 9 7 3 , 93). T h is alleged dialectical tendency tow ard isom orp h ism , incidentally, m ak es it difficult to understand how social change can ever occur. The basic problem , how ever, is not with the idea o f m irror im ages as such but with the assu m p tion that there are tw o separate levels - the cu ltural, on the one side (consisting o f sym bols) and the social and psych ological, on the other - which interact. T h is resort to Parson ian theory creates a logical space for defin­ ing the essence o f religion. By adoptin g it, Geertz m oves aw ay from a notion o f sym bols that are intrinsic to signifying and organizing practices, and back to a notion o f sym bols as m eaning-carrying ob jects external to social co n dition s and states o f the self (“ social and psych ological reality” ). T h is is not to say that G eertz d oesn ’t think o f sym bols as “ d o in g ” som ething. In a w ay

114

TALAL ASAD

that recalls older an th ropological ap p roach es to r it u a l/ he states that religious sym bols act “ by inducing in rhe w orshipper a certain d is­ tinctive set o f d isp osition s (tendencies, ca p a c i­ ties, propensities, skills, habits, liabilities, proneness) which lend a chronic ch aracter to the flow o f his activity an d the quality o f his experience” (95). And here again , sym bols are set ap art from m ental states. Rut how plausible are these p rop o sitio n s? C an we, for exam ple, predict the “ distinctive” set o f disp osition s for a C h ristian w orshiper in m odern, industrial society? Alternatively, can we say o f som eone with a “ distinctive” set o f disposition s that he is or is not a C h ristian ? The answ er to both questions m ust surely be no. The reason , o f course, is that it is not sim ply w orship but social, political, and econom ic institutions in general, within which individual biographies are lived o u t, that lend a stable character to the flow o f a C h ristian ’s activity and to the quality o f her experience. R eligious sym bols, Geertz elaborates, produce tw o kinds o f d isposition s, m oods and m otivations : “ m otivations are ‘ m ade m eaningfu r with reference to the ends tow ard s which they are conceived to conduce, w hereas m oo ds are ‘m ade m eaningful’ with reference ro the conditions from which they are conceived to sp rin g” (97). N o w , a Christian m ight say thar this is not their essence, because religious sym bols, even when failing to produce m oods and m otivation s, are still religious (i.e., true) sym bols - thar religious sym bols possess a truth independent o f their effectiveness. Yet surely even a com m itted C h ristian cannot be un con­ cerned at the existence o f truthful sym bols that ap p ear to be largely pow erless in m odern society. H e will rightly w ant to ask : W hat are the con dition s in which religious sym bols can actually produce religious disposition s? O r, as a nonbeliver w ould put it: H ow does (religious) pow er create (religious) truth? T he relation between pow er and truth is an ancient them e, and no one has dealt with it m ore im pressively in Christian thought than Sr. A ugustine. A ugustine developed his view s on the creative religious function o f pow er after his experience with the D on atist heresy, insist­ ing that coercion w as a condition for the real­ ization o f truth, and discipline essential to its m aintenance.

For a Donatist, Augustine’s attitude to coer­ cion w as a blatant denial of Christian teach­ ing: God had m ade men free to choose good or evil; a policy which forced this choice was plainly irreligious. The Donatist writers quoted the same passages from the Bible in favour o f free will, as Pelagius would later quote. In his reply, Augustine already gave them the same answer as he would give to the Pelagians: the final, individual act o f choice must be spontaneous; but this act o f choice could be prepared by a long process, which men did not necessarily choose for themselves, hut which w as often imposed on them, against their will, by God. This was a corrective process o f “ teaching,” eruditio , and warning, admomtio , which might even include fear, constraint, and external inconveniences: “ Let constraint be found outside; it is inside that the will is born.” Augustine had become convinced that men needed such firm handling. He summed up his attitude in one word: disciplina. He thought o f this disciplina, not as many o f his more traditional Rom an contemporaries did, as the static preservation o f a “ Roman way o f life.” For him it was an essentially active process of corrective punishment, “ a softening-up process,” a “ teaching by inconveniences” - a per molestias eruditio. In the Old Testament, God had taught his wayward Chosen People through just such a process of disciplina, checking and punishing their evil tendencies by a whole series o f divinely-ordained disas­ ters. The persecution o f the Donatists was another “ controlled catastrophe” imposed by Ciod, mediated, on this occasion, by the laws of the Christian Hmperors. . . . Augustine’s view o f the Fall of mankind determined his attitude to society. Fallen men had come to need restraint. Even m an’s great­ est achievements had been made possible only by a “ straight-jacket” of unremitting harsh­ ness. Augustine w as a great intellect, with a healthy respect for the achievements of human reason. Yet he w as obsessed by the difficulties of thought, and by the long, coercive pro­ cesses, reaching back into the horrors o f his own schooldays, that had made this intellec­ tual activity possible; so “ ready to lie dow n” was the fallen human mind. He said he would rather die than become a child again. None­ theless, the terrors o f that time had been strictly necessary; for they were part of the

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CATEGORY

awesome discipline o f Ciod, “ from the school­ m asters’ canes to the agonies o f the m artyrs,” by which human heings were recalled, by suf­ fering, from their own disastrous inclinations. (Brown 1967, 236-8) Isn’t G eertz’s form ula to o sim ple to acco m ­ m odate the force o f this religious sym bolism ? N ote that here it is not mere sym bols that im plant true Christian d isp osition s, but pow er - ranging all the way from law s (im perial and ecclesiastical) and other san ctions (hellfire, death, salvation , goo d repute, peace) to the disciplinary activities o f social institutions (fam ily, sch ool, city, church) and o f hum an bodies (fasting, prayer, obedience, penance). A ugustine w as quite clear that pow er, the effect o f an entire netw ork o f m otivated p rac­ tices, assum es a religious form because o f the end to which it is directed, for hum an events are the instrum ents o f G o d . It w as not the mind that m oved spon tan eously to religious truth, but pow er that created the conditions for experiencing that truth. Particular d is­ courses and practices were to be system ati­ cally excluded, forbidden, denounced - m ade as m uch as possible unthinkable; others were to be included, allow ed, p raised, and draw n into the narrative o f sacred truth. The co n ­ figurations o f pow er in this sense have, o f course, varied profoundly in C hristendom from one epoch to another - from A ugustine’s tim e, through the M iddle A ges, to the indus­ trial capitalist W est o f today. The patterns o f religious m oods and m otivation s, the p o ssi­ bilities for religious know ledge and truth, have all varied with them and been con di­ tioned by them. Even A ugustine held that although religious truth w as eternal, the m eans for securing hum an access to it were not.

From Reading Sym b ols to A n a ly zin g Practices O ne consequence o f assu m in g a sym bolic system separate from practices is that im por­ tant distinctions are som etim es obscured, or even explicitly denied. “ T h at the sym bols or sym bol system s which induce and define d is­ p osition s we set o ff as religious and those

115

which place these disposition s in a cosm ic fram ew ork are the sam e sym bols ought to occasion no su rprise” (Geertz, 98). But it does surprise! Let us grant that religious d isp o si­ tions are crucially dependent on certain reli­ giou s sym bols, rhat such sym bols operate in a w ay integral to religious m otivation and reli­ g io u s m ood. Even so , the sym bolic process by which the concepts o f religious m otivation and m ood are placed within “ a cosm ic fram ew ork” is surely quite a different operation , and there­ fore the signs involved are quite different. Put another w ay, theological discourse is not iden­ tical with either m oral attitudes or liturgical discourses - o f which, am on g other things, theology speaks. T houghtful C h ristian s will concede th at, although theology has an essen­ tial function, theological discourse does not necessarily induce religious disposition s and th at, conversely, having religious disposition s does not necessarily depend on a clear-cut con ­ ception o f the cosm ic fram ew ork on the part o f a religious actor. D iscourse involved in prac­ tice is nor the sam e as thar involved in sp eakin g ab o u t practice. It is a m odern idea that a practitioner cannot know how to live religiously w ithout being able to articulate that know ledge. G eertz’s reason for m erging the tw o kinds o f discursive process seem s to spring from a wish to distinguish in general between religious and secu lar d isposition s. The statem ent quoted above is elaborated a s follow s: For what else do we mean by saying thar a particular mood o f awe is religious and not secular, except that it springs from entertain­ ing a conception of all-pervading vitality like mana and not from a visit to the Grand Canyon? O r that a particular case o f asceti­ cism is an example o f a religious motivation except that it is directed toward the achieve­ ment o f an unconditioned end like nirvana and not a conditioned one like weightreduction? If sacred symbols did not at one and the same time induce dispositions in human beings and form ulate. . . general ideas o f order, then the empirical differentia o f reli­ gious activity or religious experience would not exist. (98) The argum ent that a particular disposition is religious partly because it occupies a

116

TA LAL ASAD

conceptual place within a cosm ic fram ew ork ap p ears p lau sib le, but only because it p resu p ­ poses a question that m ust be m ade explicit: how d o authorizing p rocesses represent p rac­ tices, utterances, or d isp osition s so that they can be discursively related to general (cosm ic) ideas o f order? In sh ort, the question pertains to the authorizin g process by which “ religio n ” is created. The w ays in which authorizing d iscourses, presupposin g and expoun din g a cosm ology, system atically redefined religious spaces have been o f profoun d im portance in the history o f W estern society. In the M iddle A ges, such discourses ranged over an enorm ous dom ain , defining and creating religion: rejecting “ p a g a n ” practices or accepting them ;7 authen­ ticating particular m iracles and relics (the tw o confirm ed each other); authorizing shrines; com piling sain ts’ lives, both as a m odel o f and as a m odel for the T ru th ; requiring the regular telling o f sinful thoughts, w ord s, and deeds to a priestly con fessor and giving absolu tion to a penitent; regularizing pop u lar social m ove­ ments into R ule-follow ing O rders (for exam ple, the Fran ciscan s), or denouncing them for heresy or for verging on the heretical (for exam ple, the Beguines). T he m edieval Church did not attem pt to establish absolute un ifor­ mity o f practice; on the con trary, its au th o rita­ tive d iscourse w as alw ays concerned to specify differences, g rad atio n s, exceptions. W hat it sought w as the subjection o f all practice to a unified authority, to a single authentic source that could tell truth from falseh ood. It w as the early C h ristian Fathers w ho established the principle that only a single Church could becom e the source o f authenticating discourse. They knew th at the “ sy m b o ls” em bodied in the practice o f self-confessed C h ristian s are not alw ays identical with the theory o f the “ one true C h u rch ,” that religion requires authorized practice and authorizin g doctrine, and that there is alw ays tension between them - som etim es breakin g into heresy, the subver­ sion o f T ruth - which underlines the creative role o f institutional p ow er.8 The m edieval Church w as alw ays clear ab o u t why there w as a con tin uous need to distinguish know ledge from falsehood (reli­ gion from w hat sought to subvert it), a s well

as the sacred from the profan e (religion from w hat w as ou tside it), distinctions for which the authoritative d iscourses, the teachings and practices o f the C hurch, not the convictions of the practitioner, w ere the final test.v Several tim es before the R eform ation , the boundary between the religious and the secular w as redraw n, but alw ays the form al authority of the Church rem ained preem inent. In later cen­ turies, with the trium phant rise o f m odern science, m odern production , and the m odern state, the churches w ould also be clear about rhe need to distinguish the religious from the secular, shifting, as they did so , the weight of religion m ore and m ore onto the m oo d s and m otivations o f the individual believer. D isci­ pline (intellectual and social) w ould, in this period, gradually aban don religious space, letting “ belief,” “ con scien ce,” and “ sensibil­ ity” take its place. But theory would still be

needed to define religion.

T h e C o n stru ctio n of Religion in E a rly M odern Europe It w as in the seventeenth century, follow ing the fragm entation o f the unity and authority o f the R om an church and the consequent w ars o f religion, which tore European principalities a p art, that the earliest system atic attem pts at producing a universal definition o f religion were m a d e .. . . H erbert produced a su b stan ­ tive definition o f w hat later cam e to be fo rm u ­ lated as N atu ral R eligion - in term s o f beliefs (abou t a suprem e pow er), practices (its ordered w orship), and ethics (a code o f conduct based on rew ards and punishm ents after this life) said to exist in all so cieties.10 This em ph asis on belief m eant that henceforth religion could be conceived a s a set o f proposition s to which believers gave assen t, and which could there­ fore be judged and com pared as between d if­ ferent religions and as again st natural science (H arrison 1990). T he idea o f scripture (a divinely produced/ interpreted text) w as not essential to this “ com m on d en o m in ato r” o f religions partly because C h ristian s had becom e m ore fam iliar, through trade and colonization , with societies

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CATEGORY

117

that lacked w riting. But a m ore im portant reason lies in the shift in attention that occurred in the seventeenth century from G o d ’s w ords to G o d ’s w orks. “ N a tu re ” becam e the real space o f divine w riting, and eventually the in disputable authority for the truth o f all sacred texts written in merely hum an language (the O ld T estam en t and the N e w ) .. . . In this w ay, N atu ral R eligion not only becam e a uni­ versal phenom enon but began to be dem ar­ cated from , and w as a lso supportive o f, a newly em erging dom ain o f natural science. I w ant to em phasize that the idea o f N atural R eligion w as a crucial step in the form ation o f the m odern concept o f religious belief, exp eri­ ence, and practice, and th at it w as an idea developed in response to problem s specific to C h ristian theology at a p articular historical juncture. By 1795, K ant w as able to produce a fully essentialized idea o f religion which could be coun terposed to its phenom enal form s: “ There m ay certainly be different historical confes­ s i o n s he w rote,

generic functions/features, and that it m ust nor be con fused with any o f its p articu lar h istori­ cal or cultural form s, is in fact a view that has a specific C h ristian history. From being a co n ­ crete set o f practical rules attached to specific processes o f pow er and know ledge, religion has com e to be abstracted and universalized. In this m ovem ent we have not merely an increase in religious toleration, certainly not merely a new scientific discovery, but the m utation o f a concept and a range o f social practices which is itself part o f a w ider change in the m odern landscape o f pow er and kn ow l­ edge. T h at change included a new kind o f state, a new kind o f science, a new kind of legal and m oral subject. T o understand this m utation it is essential to keep clearly distinct that which theology tends to obscure: the occurrence o f events (utterances, practices, d is­ position s) and the authorizing processes that give those events m eaning and em body that m eaning in concrete institutions.

although these have nothing to do with reli­ gion itself but only with changes in the means used to further religion, and are thus the prov­ ince of historical research. And there may be just as many religious books (the Zend-Avesta, the Vedas, the Koran, etc.). But there can only be one religion which is valid for all men and at all times. Thus the different confessions can scarcely be more than the vehicles o f religion; these are fortuitous, and may vary with differ­ ences in time or place. (Kant 1991, 1 14)

R eligion as M ean in g and R elig io u s M ean in g s

From here, the classification o f historical con ­ fession s into low er and higher religions becam e an increasingly p op u lar option for ph iloso­ phers, theologian s, m ission aries, and an th ro­ p o lo g ists in the nineteenth and twentieth centuries. A s to whether any p articular tribe h as existed w ithout any form o f religion w h at­ ever w as often raised as a q u e stio n ,11 but this w as recognized as an em pirical m atter not affectin g the essence o f religion itself. T h u s, w hat ap p ears to an th rop ologists today to be self-evident, nam ely that religion is essen­ tially a m atter o f sym bolic m eanings linked to ideas o f general order (expressed through either or both rite and doctrine), that it has

The equation between tw o levels o f discourse (sym bols that induce disp osition s and those that place the idea o f those d isp osition s d iscu r­ sively in a cosm ic fram ew ork) is not the only p roblem atic thing in this part o f G eertz’ s d is­ cu ssion . H e also ap p ears, inadvertently, to be tak in g up the stan dpoin t o f theology. This h appens when he insists on the prim acy of m eaning w ithout regard to the processes by which m eanings are con structed. “ W hat any p articu lar religion affirm s ab o u t the fu n da­ m ental nature o f reality m ay be obscure, sh allow , or, all to o often, perverse,” he w rites, “ but it m ust, if it is not to con sist o f the mere collection o f received practices an d conven­ tion al sentim ents we usually refer to as m oralism , affirm som eth ing” (9 8 -9 ). T he requirem ent o f affirm ation is apparently innocent an d logical, but through it the entire field o f evangelism w as historically opened up, in p articu lar the w ork o f t^uropean m ission ar­ ies in A sia, A frica, and Latin A m erica. The dem and that the received practices m ust affirm

118

TALAL ASAD

something about the fundamental nature o f reality, that it should therefore alw ays be p o s­ sible to state m eanings for them which are not plain nonsense, is the first condition for deter­ m ining whether they belong to “ religion .” The unevangelized com e to be seen typically either as those w ho have practices but affirm nothing, in which case m eaning can be attributed to their practices (thus m aking them vulnerable), or as those w ho do affirm som ething (probably “ obscure, shallow , or perverse” ), an affirm a­ tion that can therefore be dism issed. In the one case, religious theory becom es necessary for a correct reading o f the m ute ritual hieroglyph­ ics o f others, for reducing their practices to texts; in the other, it is essential for judgin g the validity o f their cosm ological utterances. But alw ays, there m ust be som ething that exists beyond the observed practices, the heard utter­ ances, the w ritten w ords, and it is the function o f religious theory to reach into, and to bring out, that backgroun d by giving them m eaning. Geertz is thus right to m ake a connection between religious theory and practice, but w rong to see it as essentially cognitive, as a m eans by which a disem bodied mind can iden­ tify religion from an A rchim edean point. The connection between religious theory and p ra c­ tice is fundam entally a m atter o f intervention - or constructing religion in the w orld (not in the mind) through definitional discourses, interpreting true m eanings, excluding som e utterances and practices and including others. H ence my repeated question: how does th eo­ retical discourse actually define religion? W hat are the historical conditions in which it can act effectively a s a dem and for the im itation, or the proh ibition, or the authentication o f truth ­ ful utterances and practices? H ow does pow er create religion? W hat kinds o f affirm ation, o f m eaning, m ust be identified with practice in order for it to qualify as religion? A ccording to G eertz, it is because all hum an beings have a p rofoun d need for a general order o f existence that reli­ giou s sym bols function to fulfill that need. It follow s that hum an beings have a deep dread o f disorder. “ There are at least three points where ch aos - a tum ult o f events which lack not ju st interpretations but interpretability -

threatens to break in upon m an: at the lim its o f his analytic capabilities, at the lim its o f his pow ers o f endurance, and at the lim its o f his m oral insight” (100). It is the function o f religious sym bols to meet perceived threats to order at each o f these points (intellectual, physical, and m oral): The Problem o f Meaning in each of its inter­ grading aspects . . . is a matter o f affirming, or at least recognizing, the inescapability o f igno­ rance, pain, and injustice on the human plane while simultaneously denying that these irra­ tionalities are characteristic o f the world as a whole. And it is in terms o f religious sym bol­ ism, a symbolism relating man’s sphere of existence to a wider sphere within which it is conceived to rest, that both the affirmation and the denial are made. (108) N otice how the reason in g seem s now to have shifted its ground from the claim that religion m ust affirm som ething specific ab o u t the nature o f reality (how ever obscure, shallow , or perverse) to the bland suggestion that reli­ gion is ultim ately a m atter o f having a positive attitude tow ard the problem o f disorder, o f affirm ing sim ply that in som e sense or other the w orld as a w hole is explicable, justifiable, bearable. This m odest view o f religion (which w ould have horrified the early Christian Fathers or m edieval churchm en)12 is a product o f the only legitim ate space allow ed to C h ris­ tianity by post-Enlightenm ent society, the right to individual belief: the hum an condition is full o f ignorance, pain, and injustice, and religious sym bols are a m eans o f com ing positively to term s with that condition. O ne consequence is that this view w ould in principle render any philosophy that perform s such a function into religion (to the annoyance o f the nineteenth* century ration alist), or alternatively, m ake it possible to think o f religion as a m ore prim i­ tive, a less adult m ode o f com ing to term s with the hum an condition (to the annoyance o f the m odern C hristian). In either case, the su g ges­ tion that religion has a universal function in belief is one indication o f how m arginal reli­ gion h as becom e in m odern industrial society a s the site for produ cin g disciplined know ledge and personal discipline. A s such it com es to resem ble the conception M arx had o f religion

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CATEGORY

as ideology - that is, as a m ode o f co n sciou s­ ness which is other than consciousness o f reality, external to the relations o f production , producing no know ledge, but expressin g at once the anguish o f the oppressed and a sp u ri­ ous con solation . Geertz has m uch m ore to say, how ever, on the elusive question o f religious m eaning: not only d o religious sym bols form ulate con cep­ tions or a general order o f existence, they also clothe those conceptions with an aura o f factuality. T his, we are told, is “ the problem o f belief.” Religious belief alw ays involves “ the prior acceptance o f au th o rity ,” which tran s­ form s experience: The existence of bafflement, pain, and moral paradox - o f the Problem o f Meaning - is one of the things that drives men toward belief in gods, devils, spirits, totemic principles, or the spiritual efficacy o f can n ibalism ,. . . but it is not the basis upon which those beliefs rest, but rather their most important field o f appli­ cation. (109) This seem s to imply that religious belief stan ds independently o f the w orldly conditions that produce bafflem ent, pain, and m oral p arad o x , although rhat belief is prim arily a w ay o f com ing to term s with them. Rut surely this is m istaken, on logical groun d s as well a s h istori­ cal, for changes in the object o f belief change that belief; and as the w orld ch an ges, so do rhe objects o f belief and the specific form s o f b a f­ flement and m oral p a ra d o x that are a part o f that w orld. W hat the Christian believes today ab o u t G o d , life after death, the universe, is not w hat he believed a millennium a g o - nor is the w ay he respon ds to ignorance, pain, and in jus­ tice the sam e now as it w as then. The m edieval valorization o f pain as the m ode o f p articip at­ ing in C h rist’s suffering co n trasts sharply with the m odern C ath olic perception o f pain as an evil to be fought again st and overcom e as C h rist the H ealer did. T h at difference is clearly related to the post-Fnlightenm ent secu lariza­ tion o f W estern society an d to the m oral lan ­ guage which that society now authorizes. Cieertz’s treatm ent o f religious belief, which lies at the core o f his conception o f religion, is a m odern, privatized Christian one because and to the extent that it em phasizes the prior­

119

ity o f belief as a state o f mind rather than as constituting activity in the w orld: “ The basic axiom underlying w hat we m ay perh aps call ‘the religious perspective’ is everywhere the sam e: he w ho w ould know m ust first believe” (110). In m odern society, where know ledge is rooted either in an a-C h ristian everyday life or in an a-religious science, the Christian a p o lo ­ gist tends not to regard belief a s the conclusion to a know ledge process but as its precondition. H ow ever, the know ledge that he prom ises will not p ass (nor, in fairness, d oes he claim that it will pass) for know ledge o f social life, still less for the system atic know ledge o f objects thar natural science provides. H er claim is to a particular state o f m ind, a sense o f conviction, not to a co rp u s o f practical know ledge. Rut the reversal o f belief and know ledge she dem ands w as not a basic axiom to, say, pious learned C h ristians o f the twelfth century, for whom know ledge and belief were not so clearly at odds. On the contrary, Christian belief w ould then have been built on know ledge know ledge o f theological doctrine, o f canon law and Church courts, o f the details o f cleri­ cal liberties, o f the pow ers o f ecclesiastical office (over souls, bodies, properties), o f the preconditions and effects o f confession , o f the rules o f religious orders, o f the locations and virtues o f shrines, o f the lives o f the sain ts, and so forth. Fam iliarity with all such (religious) know ledge w as a precondition for norm al social life, and belief (em bodied in practice and discourse) an orientation for effective activity in it - whether on the part o f the religious clergy, the secular clergy, or the laity. Because o f this, the form and texture and function o f their beliefs w ould have been different from the form and texture and function o f contem ­ porary belief - and so to o o f their dou bts and their disbelief. The assum ption that belief is a distinctive m ental state characteristic o f all religions has been the subject o f discu ssion by con tem po­ rary scholars. T h u s, N eedh am (1972) has interestingly argued that belief is nowhere a distinct m ode o f con sciousn ess, nor a neces­ sary institution for the conduct o f social life. Southw old (1979) takes an alm ost diam etri­ cally op p o sed view, asserting th at questions o f belief do relate to distinctive m ental states and

120

TA LAL ASAD

that they are relevant in any and every society, since “ to believe” alw ays design ates a relation between a believer and a proposition and through it to reality. H arre (1 9 8 1 , 8 2 ), in a criticism o f N eed h am , m akes the m ore p ersu a­ sive case that “ belief is a m ental state, a grounded disp osition , but it is confined to people w ho have certain social institutions and p ractices.” At any rate, I think it is not to o unrea­ sonable to m aintain that “ the basic a x io m ” underlying w h at Geertz calls “ the religious perspective” is not everywhere the sam e. It is preem inently the C h ristian church that has occupied itself with identifying, cultivating, and testing belief as a verbalizable inner co n d i­ tion o f true religion (A sad 1986b).

R eligion as a P ersp ective The phenom enological vocabulary rhat Geertz em ploys raises tw o interesting question s, one regarding its coherence and the other concern­ ing its adequacy to a m odern cognitivist notion o f religion. 1 w an t to suggest that although this vocabulary is theoretically incoherent, it is socially quite com patible with the privatized idea o f religion in m odern society. T h u s, “ the religious perspective,” we are told, is one am o n g several - com m on-sense, scientific, aesthetic - and it differs from these as follow s. It differs from the common-sense perspective, because it “ m oves beyond the realities o f everyday life ro wider ones which correct and com plete them , and [because] its defining concern is not action upon those wider realities but acceptance o f them , faith in them .” It is unlike the scientific perspective, because “ it question s the realities o f everyday life not ou t o f an institutionalized scepticism which d issolv es the w orld ’ s givenness into a swirl o f p robabilistic hypotheses, but in term s o f w hat it takes to be w ider, non-hypothetical tru th s.” And it is distinguished from the aes­ thetic perspective, because “ instead o f effect­ ing a disengagem ent from the whole question o f factuality, deliberately m anufacturing an air o f sem blance and illusion, it deepens the concern with fact and seeks to create an aura o f utter actu ality ” (1 12). In other w ords,

although the religious perspective is not exactly ration al, it is not irrational either. It w ould not be difficult to state on e’s d is­ agreem ent with this sum m ary o f w hat com m on sense, science, and aesthetics are abo u t. Rut my point is that the option al flavor conveyed by the term perspective is surely m isleading when it is applied equally to science and to religion in m odern society: religion is indeed now optional in a w ay that science is not. Scientific practices, techniques, know ledges, perm eate and create the very fibers o f social life in w ays that religion no longer does, in that sense, religion today is a perspective (or an “ atti­ tu d e,” as Geertz som etim es calls it), but science is not. In that sense, to o, science is not to be found in every society, past and present. We shall see in a m om ent the difficulties that G eertz’s perspectivism gets him into, but before that I need to exam ine his analysis o f the m echanics o f reality m aintenance at w ork in religion. C onsistent with previous argum ents about the functions o f religious sym bols is G eertz’s rem ark that “ it is in ritual - that is, con se­ crated behavior - that this conviction that religious con ception s are veridical and thar religious directives are sound is som ehow gen­ erated ” (1 1 2 ). The long p assage from which this is taken sw ings back and forth between arbitrary specu lation s ab o u t w hat goes on in the con sciousn ess o f officiants and unfounded assertions ab o u t ritual a s im printing. At first sight, this seem s a cu riou s com bin ation o f introspectionist psychology with a behaviorist one - but as V ygotsky (1 9 7 8 , 5 8 -9 ) argued long ago , the tw o are by no m eans in con sis­ tent, in sofar as both assum e that psychological phenom ena con sist essentially in the con se­ quence o f various stim ulatin g environm ents. Geertz p ostu lates the function o f rituals in generating religious conviction (“ In these plastic d ram as men attain their faith as they portray it” (1 1 4 ]), but how or why this h appens is nowhere explain ed. Indeed, he concedes that such a religious state is nor alw ays achieved in religious ritual: “ O f co urse, all cultural p erfor­ m ances are not religious perform an ces, and the line between those that are, and artistic, or even political, ones is often not so easy to draw in practice, for, like social form s, sym bolic

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CATEGORY

fo rm s can serve m ultiple p u rp o se s” (11 3 ). Hut the question rem ains: W hat is it that ensures the particip an t’ s takin g the sym bolic form s in the w ay that leads to faith if the line between religious and nonreligious perspectives is not so easy to draw ? M u stn ’t the ability and the will to ad o p t a religious stan dpoin t be present prior to the ritual p erform an ce? T h at is pre­ cisely why a sim ple stim ulus-response model o f how ritual w orks will not do. And if that is the case, then ritual in the sense o f a sacred perform ance can n ot be the place where reli­ giou s faith is attain ed, but the m anner in which it is (literally) played out. If we are to under­ stan d how this h appens, we m ust exam ine not only the sacred perform ance itself but also the entire range o f availab le disciplinary activities, o f institutional form s o f know ledge and p rac­ tice, within which d isp osition s are form ed and sustained and through which the possibilities o f attain in g the truth are m arked ou t - as A ugustine clearly saw . I have noted m ore than once G eertz’s concern to define religious sym bols according to universal, cognitive criteria, to distinguish the religious perspective clearly from nonrelig io u s ones. The sep aratio n o f religion from science, com m on sense, aesthetics, politics, an d so on , allow s him to defend it again st ch arges o f irrationality. If religion has a d is­ tinctive perspective (its ow n truth, as D urkheim w ould have said) and perform s an indispens­ able function, it does not in essence com pete with others and can n ot, therefore, be accused o f generating false con sciousn ess. Yet in a way this defense is equivocal. R eligious sym bols create d isp osition s, G eertz observes, which seem uniquely realistic. Is this the point o f view o f a reasonably confident agent (w ho m ust alw ays operate w ithin the denseness o f historically given probabilities) or that o f a skeptical observer (w ho can see through the representations o f reality to the reality itself)? It is never clear. And it is never clear because this kind o f phenom enological ap p roach doesn ’t m ake it easy to exam ine whether, and if so to w hat extent and in what w ay s, reli­ gio u s experience relates to som ething in the real w orld that believers inhabit. This is partly because religious sym bols are treated, in circu­ lar fashion , as the precondition for religious

121

experience (which, like any experience, m ust, by definition, be genuine), rather than a s one condition for engagin g with life. T o w ard the end o f his essay, G eertz attem pts to connect, instead o f sep aratin g, the religious perspective and the com m on-sense one - and the result reveals an am biguity basic to his entire ap proach . First, invoking Schutz, Geertz states that the everyday w orld o f com m onsense objects and practical acts is com m on to all hum an beings because their survival depen ds on it: “ A m an, even large grou p s o f m en, m ay be aesthetically insensitive, reli­ giously unconcerned, and unequipped to pursue form al scientific an alysis, but he cannot be com pletely lacking in com m on sense and survive” (119). N ex t, he in form s us that indi­ viduals m ove “ back an d forth between the religious perspective and the com m on-sense perspective” (119). T hese perspectives are so utterly different, he declares, that only “ K ierkegaardian lea p s” (120) can cover the cultural g a p s that separate them . T hen, the phenom enological conclusion: Having ritually “ leapt” . . . into the fram e­ work of meaning which religious conceptions define, and the ritual ended, returned again to the common-sense world, man is - unless, as sometimes happens, rhe experience fails to register - changed. And as he is changed, so also is the common-sense world , for it is now seen as but the partial form of a wider reality which corrects and completes it. (122; empha­ sis added) T h is cu riou s account o f shifting perspectives and ch an ging w orlds is puzzling - a s indeed it is in Schutz him self. It is not clear, for exam ple, w hether the religious fram ew ork and the com m on-sense w orld, between which the indi­ vidual m oves, are independent o f him or not. M o st o f w h at G eertz h as said at the beginning o f his essay w ould im ply that they are inde­ pendent (cf. 9 2 ), and his rem ark a b o u t com m on sense being vital to every m an ’s survival also enforces this reading. Yet it is a lso suggested that a s the believer ch anges his perspective, so he him self ch an ges; and th at as he ch an ges, so to o is his com m on-sense w orld changed and corrected. So the latter, at any rate, is not independent o f his m oves. But it w ould appear

122

TALAL ASAD

from the accoun t that the religious w orld is independent, since it is the source o f distinctive experience for the believer, and through that experience, a source o f change in the com m onsense w orld: there is no suggestion anyw here that the religious w orld (or perspective) is ever affected by experience in the com m on-sense w orld. This last point is consistent with the phe­ nom enological ap p roach in which religious sym bols are sui generis, m arking out an inde­ pendent religious dom ain . But in the present context it presents the reader with a p a rad o x : the w orld o f com m on sense is alw ays com m on to all hum an beings, and quite distinct from the religious w orld, which in turn differs from one grou p to another, as one culture differs from another; but experience o f the religious w orld affects the com m on-sense w orld, and so the distinctiveness o f the tw o kinds o f w orld is m odified, and the com m on-sense w orld com es to differ, from one grou p to another, as one culture differs from another. T he p a rad o x results from an am b igu o u s phenom enology in which reality is at once the distance o f an agen t’s social perspective from the truth, m ea­ surable only by the privileged observer, and also the substantive know ledge o f a socially constructed w orld available to both agent and observer, but to the latter only through the fo rm er.11

Conclusion Perhaps we can learn som ething from this p a rad o x which will help us evaluate G eertz’s confident conclusion: “ The anth ropological study o f religion is therefore a tw o-stage op er­ ation: first, an an alysis o f the system o f m ean ­ ings em bodied in the sym bols which m ake up the religion proper , an d , second, the relating o f these system s to social-structural and p sy ­ chological p ro cesses” (1 2 5 ; em ph asis added). H ow sensible this soun ds, yet how m istaken, surely, it is. If religious sym bols are under­ sto o d , on the analogy with w ords, as vehicles for m eaning, can such m eanings be established independently o f the form o f life in which they are used? If religious sym bols are to be taken as the signatures o f a sacred text, can we know

w hat they m ean w ithout regard to the social disciplines by which their correct reading is secured? If religious sym bols are to be thought o f as the concepts by which experiences are organ ized, can we say m uch ab o u t them w ithout considering how they com e to be authorized? Even if it be claim ed that w h at is experienced through religious sym bols is not, in essence, rhe social w orld but the sp iritu a l,14 is it possible ro assert that conditions in the social w orld have nothing to d o with m aking th at kind o f experience accessible? Is the concept o f religious training entirely vacuous? The tw o stages that G eertz proposes are, I w ould suggest, one. R eligious sym bols — whether one thinks o f them in term s o f co m ­ m unication or o f cognition, o f guiding action or o f expressing em otion - cann ot be under­ stood independently o f their historical rela­ tions with nonreligious sym bols or o f their articulation s in and o f social life, in which w ork and pow er arc alw ays crucial. M y a rg u ­ ment, I m ust stress, is not just that religious sym bols are intim ately linked to social life (and so change with it), or that they usually supp ort dom inant political pow er (and o c c a ­ sionally o p p o se it). It is that different kinds o f practice and discourse are intrinsic to the field in which religious representations (like any representation) acquire their identity and their truthfulness. From this it does not follow that the m eanings o f religious practices and utter­ ances are to be sough t in social phenom ena, but only that their possibility and their au th o r­ itative statu s are to be explained as products o f historically distinctive disciplines and forces. The an th ropological student o f particular religions should therefore begin from this point, in a sense unpacking the com prehensive concept which he or she translates as “ reli­ g io n ” into heterogeneous elem ents according to its historical character. A final w ord o f caution. H asty readers m ight conclude that my discussion o f the Christian religion is skew ed tow ards an auth oritarian , centralized, elite perspective, and that co n se­ quently it fails to take into accoun t the reli­ gion s o f h eterodox believers, o f resistant peasantries, o f all those w ho cannot be co m ­ pletely controlled by the orth o dox church. O r,

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CATEGORY

w orse still, that my discussion h as no bearing on nondisciplinarian, volun taristic, localized cults o f noncentralized religions such as Hinduism . But that conclusion w ould be a misunderstanding o f this ch apter, seeing in it an attem pt to ad vocate a better an th ropological definition o f religion than Geertz has done. N oth ing could be farther from my intention. If my effort reads in large p art like a brief sketch o f tran sm utation s in C hristianity from

the M iddle Ages until today, then that is not because I have arbitrarily confined my ethnograph ic exam ples to one religion. M y aim has been to problem atize the idea o f an anthropological definition o f religion by assigning that endeavor to a particular history o f knowledge and pow er (including a particular understanding o f our legitim ate p ast and future) out o f which the m odern w orld h as been constructed.

NOTES 1

2

T h u s, Fustel de C o u lan ges 1873. O rigi­ nally published in French in 1864, this w as an influential w ork in the history o f several overlappin g disciplines - an th ropology, biblical studies, and classics. C om p are Peirce’s m ore rigoro u s account o f representations.

5

6

A representation is an object which stands for another so that an experience of the former affords us a knowledge of the latter. There must be three essential conditions to which every representation must conform. It must in the first place like any other object have qualities inde­ pendent o f its meaning . . . In the 2nd place a representation must have a real causal connection with its object. . . . In the third place, every representation addresses itself to a mind. It is only in so far as it does this that it is a representa­ tion. (Peirce 1986, 62) 7 3

4

V ygotsky (1 9 6 2 ) m akes crucial analytical distinctions in the developm ent o f co n ­ ceptual thought: h eaps, com plexes, pseu­ d ocon cepts, and true concepts. Although, accordin g to V ygotsky, these represent stages in the developm ent o f children’s use o f lan guage, the earlier stages persist into adult life. The argum ent that sym bols organize practice , and consequently the structure o f cognition, is central to V ygotsky’s genetic psychology - see especially “ T ool and Sym bol in Child D evelopm en t,” in V ygotsky 197 8 . A cognitive conception o f

123

8

sym bols has recently been revived by Sperber (1975). A sim ilar view w as taken m uch earlier by Lienhardt (1961). O r, a s K roeber and Kluckhohn (1952, 181) put it much earlier, “ C ulture consists o f patterns, explicit and im plicit, o f and for behaviour acquired and transm itted by sy m b o ls.” If we set aside R adcliffe-Brow n’s wellknow n preoccupation with social cohe­ sion, we m ay recall that he to o w as concerned to specify certain kinds o f psy­ chological states said to be induced by religious sym bols: “ Rites can be seen to be the regulated sym bolic expressio n s o f certain sentim ents (which control the behaviour o f the individual in his relation to others). Rites can therefore be shown to have a specific social function when, and to the extent that, they have for their effect to regulate, m aintain and transm it from one generation to another senti­ m ents on which the constitution o f society depen ds” (1 9 5 2 , 157). The series o f booklets know n a s peniten­ tial m an uals, with the aid o f which C h ris­ tian discipline w as im posed on W estern Europe from roughly the fifth to the tenth centuries, con tains m uch m aterial on pagan practices penalized as unChristian. So, for exam ple, “ The taking o f vow s or releasing from them by springs or trees or lattices, anyw here except in a church, and p artakin g o f fo od or drink in these places sacred to the folk-deities, are offenses condem n ed” (quoted in M cN eill 1933, 4 5 6 ) . . . . The Church alw ays exercised the au th o r­ ity to read Christian practice for its

124

9

10

TA LAL ASAD

religious truth. In this con text, it is inter­ esting that the w ord heresy at first desig­ nated all kinds o f errors, including errors “ un con sciou sly” involved in som e activ­ ity (sim oniaca haersis ), and it acquired its specific m odern m eaning (the verbal fo r­ m ulation o f denial or d ou bt o f any defined doctrine o f the C ath olic church) only in the course o f the m eth odological co n tro­ versies o f the sixteenth century (Chenu 1 9 6 8 ,2 7 6 ). In the early M iddle A ges, m on astic d isci­ pline w as the principal basis o f religiosity. Know les (1 9 6 3 , 3) observes that from roughly the sixth to the tw elfth centuries, “ m on astic life based on the R ule o f St. Benedict w as everywhere the norm and exercised from time to time a p aram ou n t influence on the spiritual, intellectual, liturgical an d ap o sto lic life o f the W estern Church. . . . the only type o f religious life available in the countries concerned w as m on astic, and the only m on astic code w as the Rule o f St. Benedict.” D uring the period the very term religious w as there­ fore reserved for those living in m onastic com m unities; with the later em ergence o f n onm onastic ord ers, the term cam e to be used for all w ho had taken lifelong vow s by which they were set a p art from the ordinary m em bers o f the Church (South­ ern 1 9 7 0 , 2 1 4 ). The extension and sim ul­ tan eous tran sform ation o f the religious disciplines to lay sections o f society from the twelfth century on w ard (Chenu 1968) contributed to the C h urch ’s authority becom ing m ore pervasive, m ore com plex, and m ore con tradictory than before - and so to o the articulation o f the concept and practice o f lay religion. W hen Christian m ission aries found them ­ selves in culturally un fam iliar territory, the problem o f identifying “ religion ” becam e a m atter o f con siderable theoreti­ cal difficulty and practical im portance. For exam ple, The Jesuits in China contended that the reverence for ancestors was a social, not a religious, act, or that if religious, it w as hardly different from Catholic prayers

for the dead. They wished the Chinese to regard Christianity, not as a replacement, not as a new religion, but as the highest fulfillment of their finest aspirations. But to their opponents the Jesuits appeared to be merely lax. In 1631 a Franciscan and a Dominican from the Spanish zone of Manila travelled (illegally, from the Portuguese viewpoint) to Peking and found that to translate the word mass , the Jesuit catechism used the character tsiy which w as the Chinese description of the ceremonies o f aneestor-worship. One night they went in disguise to such a ceremony, observed Chinese Christians participating and were scandalized at what they saw. So began the quarrel of “ the rites,” which plagued the eastern missions for a century and more. (Chad­ wick 1964, 338) 11

12

13

14

For exam ple, by T ylor in the chapter “ A nim ism ” in part 2 o f Primitive Culture [see chapter 11. When the fifth-century bishop o f Ja v o ls spread Christianity into the A uvergne, he found the peasan ts “ celebrating a threeday festival with offerin gs on the edge o f a m a r s h .. . . ‘N ulla est religio in s ta g n o / he said: T here can be no religion in a sw a m p ” (Brow n 1981, 125). For m edi­ eval C h ristian s, religion w as not a univer­ sal phenom enon: religion w as a site on which universal truth w as produced, and it w as clear to them that truth w as not produced universally. In the introduction to his 1983 collection o f essays, G eertz seem s to w ant to aban don this perspectival approach . . . . Cf. the final chapter in Evans-Pritchard 1956, and a lso the conclusion to EvansPritchard 1965.

REFERENCES A sad, T . 1986b. “ M edieval H eresy: An A nthropological V iew .” Social History II, no. 3. Brow n, P. 1967. Augustine o f H ippo. London: Faber and Faber.

THE CONSTRUCTION OF RELIGION AS AN ANTHROPOLOGICAL CA TEG O RY

------. 1981. The Cult o f the Saints: Its Rise and Function in Latin Christianity. London: SCM . B urckh ardt, J. 1 9 5 0 (1 8 6 0 ]. The Civilization o f the Renaissance in Italy. Lon don : Phaidon. Burling, R. 1977. Review o f Political L an ­ guage and O ratory in Traditional Society , by M aurice Bloch. American Anthropolo­ gist 79. Burns, E. 1990. Character: Acting and Being on the Pre-Modern Stage . N ew Y ork: St. M artin ’s. Butler, C . 192 4 . Benedictine Motiasticism. C am brid ge: C am brid ge Univ. Press. Butterfield, H. 1931. The Whig Interpretation o f History1. London: Bell. Bynum , C . W. 1980. “ Did the Tw elfth Century D iscover the In d ividu al?” Jo u rn al o f Eccle­ siastical History 3 1 , no. 1. C aen egem , R. C . van. 1965. “ La preuve dan s le droit du m oyen age occid en tal.” La Preuve , Recueils de la societe Je an Bod in pou r Phistoire com p arative des institutions, vol. 17. Brussels. C h ad w ick , O . 1964. The Reformation. H arm on dsw orth , M iddlesex: Penguin. C h en u, M -D . 1968. Nature, Man, and Society

in the Twelfth Century: Essays on Theologi­ cal Perspectives in the Latin West. C hicago: Univ. o f C h icago Press. C o u lan g es, Fustel de. 1 873. The Ancient City: A Study on the Religion , L aw s , an d Institu­ tions o f Greece and Rome. Boston: Loth rop, Lee and Shepherd. D u m on t, L. 1971. “ R eligion, Politics, and Society in the Individualistic U niverse.” Pro -

ceedings o f the Royal A nthropological Insti­ tute fo r 1970 . Evans-Pritchard, E. E. 1956. Nuer Religion. O x fo rd : Clarendon. ------. 1965. Theories o f Primitive Religion. O x fo rd : Clarendon. G eertz, C . 1973. The Interpretation o f Cul­ tures. N ew York: Basic Books. ------. 1983. L ocal Knowledge: Further Essays in Interpretive Anthropology . N ew York: Basic Books. H arre, R. 1981. “ Psychological Variety.*' In Indigenous Psychologies , edited by P. H eelas and A. Lock. Lon don : A cadem ic Press.

125

H arrison , P. 1990. “ R eligion ” and the Reli­ gions in the English Enlightenment. C a m ­ bridge: C am bridge Univ. Press. K an t, 1. 1991. Kant: Political Writings. Edited by H . R eiss. C am bridge: C am b rid ge Univ. Press. K n ow les, M . D ., ed. 1963. The M onastic O rder in England: 9 4 0-1216. 2d ed. C a m ­ bridge: C am bridge Univ. Press. K roeber, A. L., and C. Kluckhohn. 1952.

Culture: A Critical Review o f Concepts and Definitions. Papers o f the Peabody M useum , vol. 4 7 , no. 1. C am bridge, M a ss.: Peabody M useum . L ea, H . C. 18 9 6 . A History o f Auricular Con­

fession and Indulgences in the Latin Church. 3 vols. Philadelphia: Lea Bros. Leach , E. R. 1954. Political Systems o f H igh­ land Burma. London: Bell. ------. 1973. “ O urselves and O th ers.” Times Literary> Supplement , 6 July. Leavitt, J. 1986. “ Strategies for the Interpreta­ tion o f A ffect.” M anuscript. Leclercq, J. 1957. “ D iscip lin a.” In Dictionnaire de Spiritualite , 3. Paris: Beauchesne. ------. 1966. “ The Intentions o f the Founders o f the C istercian O rd e r.” Cistercian Studies 4. ------. 1971. “ Le cloitre est-il une p riso n ?” Revue d'ascetique et de mystique 47. ------. 1977. The Love o f Learning and the

Desire for C o d : A Study o f M onastic Culture. 2d ed. N ew Y ork: Fordham Univ. Press. —— . 1979. Monks and Love in TwelfthCentury France. O x fo rd : O x fo rd Univ. Press. Leclercq, J ., and G. G artn er. 1965. “ S. Bernard d an s Phistoire de Pobeissance m on astiq u e.”

Annuario De F.studios Medievales 2. Le G o ff, J. 1980. Time , Work, an d Culture in the M iddle Ages. C h icago: Univ. o f C h icago Press. L ek ai, L. J . 1977. The Cistercians: Ideals and Reality. Kent, O h io: Kent State Univ. Press. Lerner, D . 1958. The Passing o f Traditional Society: Modernizing the M iddle Flast. N ew Y ork: Free Press. Levi, A. 1964. French M oralists: The Theory o f the Passions, 1585 to 1649. O x fo rd : C laren don.

126

T A L A l ASAD

Lienhardt, G. 1 961. Divinity and Experience. O x fo rd : Clarendon. M cN eill, J. T . 1933. “ Folk-Paganism in the Penitentials.” Jou rn al o f Religion 13. N eedh am , R. 1972. Belief, Language, and Experience. O x fo rd : Basil Blackwell. Peirce, C. S. 198 6 . Writings o f C. S . Peirce. V ol. 3. B loom ington: Indiana Univ. Press. R adclitfe-Brow n, A. R. 1 9 5 2 11 9 3 9 ], “ T a b o o .” In Structure and Function in Primitive Society. London: C ohen and W est. Southern, R. W. 1970. Western Society and the Church in the Middle Ages. H arm on dsw orth, M idd lesex: Penguin.

South w old, M . 1979. “ R eligious B elief.” M an , n.s. 14. Sperber, 1). 1 9 7 5 . Rethinking Symbolism. C am bridge: C am bridge Univ. Press. T ylor, E. B. 1871. Primitive Culture. London: J. M urray. V ygotsky, L. S. 1962 [1934]. Thought and Language. C am bridge, M ass.: M IT Press. ------. 1978. Mind in Society. C am bridge, M ass.: H arvard Univ. Press.

Part II

Poiesis: The Composition of Religious Worlds

Signs and Symbols

In trod u ction Over the next few chapters we exam ine the way religion can be understood as som ething crafted or m ade. Following Aristotle, we can name this creative process poiesis. Poiesis is a useful concept in part because it does not distinguish the ideal from the material. Indeed, the notion is not all that different from M a r x ’s idea o f productive work and the sense that it is through labor that hum ans express their species being (cf Arendt 1999). Unfortunately, anthropology has long been caught in a duality that distinguishes the realm o f creation o f and by means o f ideas from that o f material creation (physical labor and its tools and products). Flence (despite som e notable exceptions) we have tended to separate and distinguish the study o f religion from the study o f such things as the means o f gaining a livelihood or repro­ ducing family and collective life. Flowever, at the root o f cultural anthropology has been the idea o f the sym bol as both the primary hum an product and the primary hum an tool. Symbols are understood as the building blocks not only o f religion but o f language, culture, and social practice more broadly. As many American anthro­ pologists have argued, sym bols mediate our relations to the world. Linked together by means o f poiesis, creative production, they ultimately shape particular under­ standings o f the world and orientations to it, as Geertz argued in his essay (chapter 4). If we take religious worlds to be crafted or m ade, and if we take as one o f our tasks the comprehension and interpretation o f such cosm ologies (meaningful worlds or worlds o f meaning), then the way to begin is through examining processes o f signification, produced by means o f signs or sym bols, and o f rhetoric, elaborated by m eans o f tropes. Philosopher Susanne Langer’s book is one place in which the argum ent concerning sym bols has been clearly laid out; another is in the work o f Leslie White (1949). These have formed the basis for much American symbolic anthropology.

130

INTRODUCTION

There are several approaches to the production o f meaning: semiotics is inspired by American philosopher Charles Peirce’s extensive discussion o f signification (1960), while structuralism draw s from Swiss linguist Ferdinand de Saussure (1959). Symbolic anthropology, semiotics, and structuralism all see their subject as systems o f signs, meaning, or signification, and hence more broadly than what is generally subsum ed under “ religion.” Conversely, it is questionable whether any o f these approach es are sufficient to fully account for “ religion,” and in particular to c o m ­ prehend ritual, religious experience, or religious action.

10

The Logic of Signs and Symbols Susanne K. Langer

Susanne Langer (1895-1985) was an Am erican philosopher whose book P h i­ losophy in a N e w Key was a particularly incisive and influential account of the turn to meaning. Langer managed to synthesize w ork in anthropology, lin­ guistics, psychology, and aesthetics as well as philosophy. In the chapter before the one presented here, Langer em pha­ sizes the direct origins of religion in human mental functioning. She pro­ vides a vigorous rejection of utilitarian accounts of symbol and ritual, arguing instead that the mind's need to em it a continuous stream of symbolic expres­ sion is as strong as any practical interest: "The fact th at the human brain is con­ stantly carrying on a process of symbolic transform ation of the experiential data that come to it causes it to be a veritable fountain of more or less spontaneous ideas" (1948 [1942]: 34). She also deftly shows the lim itations of an ethological

model (a model drawn from the study of anim al behavior) for understanding humans. Langer distinguishes sign and symbol, providing a simpler model than that of Peirce (in whose term inology, now more w idely used than Langer's, the symbol is one category of sign). Although the debate over m eaning - and over the usefulness of terms like "denotation," "connotation," and the very "m eaning of 'm eaning'" - has continued vigor­ ously since Langer's day, she provides a particularly clear exposition of some of the main issues. Langer emphasizes how symbols are critical for thought. Unlike animals, humans do not simply respond to the world, but must inter­ pret it. In subsequent chapters she distinguishes the effects of discursive forms of symbolism (by means of language) from presentational (nonlinguistic, simultaneous, integral) ones,

Reprinted by permission of the publisher from “ The Logic of Signs and Symbols,” in Philosophy in a New Key: A Study in the Symbolism of Reason, Rite and Art by Susanne K. Langer, (Cambridge, MA: Harvard University Press), pp. 53-67. Copyright © 1942, 1951, 1957 by the President and Fellows of Harvard College, Renewed 1970, 1979 by Susanne K. Langer, 1985 by Leonard C. R. Langer. Abridged.

132

SUSANNE K. LANGER

addressing questions of meaning in art and music. These insights have been developed in a variety of ways by subsequent th in k ­ ers. They lie behind Geertz's profound argum ent concerning the evolutionary dependence of the brain on the process of symbolic transform ation ("culture" in his vocabulary) in order to function (1973c [1962]) as w ell as behind Sahlins's sustained accounts of the m ediation of all human practice by symbolic transfo r­ m ation ("culture" or "structure") (1976). And they anticipate the kind of argu­ m ent m ade by Levi-Strauss concerning the continuous production of "undo ­ m esticated" thought in myth (see chapter 17), w hile recognizing the Freudian account of symbolic transfor­ mation in dreams. Although all three main approaches to m eaning - structuralism, semiotics, and symbolic (interpretive) anthropology share much common philosophical ground, they apportion things d if­ ferently. Structuralism begins w ith a concept of the linguistic sign which, like

. . . M ean in g has both a logical and a p sych o­ logical aspect. Psychologically, any item that is to have m eaning m ust be employed a s a sign or a sym bol; that is to say, it m ust he a sign or a sym bol to som eone. Logically, it m ust be capable o f conveying a m eaning, it m ust be rhe sort o f item that can be thus em ployed. In som e m eaning-relations this logical require­ ment is trivial, and tacitly accepted; in others it is o f the u tm ost im portan ce, and m ay even lead us a m erry ch ase through the labyrinths o f nonsense. T hese tw o asp ects, the logical and the psych ological, are thoroughly confounded by the am b igu o u s verb “ to m ean ” ; for som e­ tim es it is proper to say “ // m ean s,” and som e­ tim es “ / m ean .” O bviously, a w ord - say, “ L o n d o n ” - d oes not “ m ean ” a city in ju st the sam e sense th at a person em ploying the w ord “ m ean s” the place.

Langer's concept of "symbol" (and con­ fusingly, unlike her concept of "sign"), is arbitrary. Structuralists deploy the binary distinction betw een signifier and signified and exam ine sets of signs con­ structed through relations of minimal contrasts, on the model of phonemes. Semiotics, w hich is based more faithfully on Peirce's extensive elaboration, begins w ith a distinction among index, icon, and symbol as the three main categories of signs and em phasizes the key concept of the interpretant. Peirce is deployed to great effect in V alentine Daniel's F lu id Signs: Being a Person the Tamil W ay (1984), w hich includes an unforget­

tab le description of a South Indian pilgrim age. W hat Langer calls the psy­ chological dimension of meaning might today be referred to as the intentional or pragm atic dimension. Semiotics more easily accom m odates pragmatics and so has generally taken precedence in tech­ nical accounts of meaning. Lee (1997) offers an advanced interdisciplinary synthesis.

Both aspects, the logical and the psych ologi­ cal, are alw ays present, and their interplay p ro ­ duces the great variety o f m eaning-relations over which ph ilosoph ers have puzzled and fought for the last Hftv years. The an alysis o f “ m ean in g” has had a peculiarly difficult history; the w ord is used in m any different w ays, and a go o d deal o f controversy has been w asted on the subject o f the correct w ay, the m eaning o f “ m ean in g.” W henever people find several species o f a genus, they look for the prim e form , the archetype that is supposed to be differently disguised in each special case; so , for a long tim e, philosophers hoped to find the true quality o f m eaning by collecting all its various m an ifestations and look in g for a com m on ingredient. They talked m ore and m ore generally ab o u t “ sym bol-situ atio n s,” believing that by generalization they m ight

133

THE LOGIC OF SIGNS AND SYMBOLS

attain to the essential quality which all such situ ation s had in com m on. But generalizing from vague and m uddled special theories can never give us a clear general theory. T he sort o f generalization that merely substi­ tutes “ sym bol-situation ” for “ denotationor-connotation-or-signification-or-associationetc\” is scientifically useless; for the whole pu rp o se o f general con cepts is to m ake the d is­ tinctions between special classes clear, to relate all subspecies to each other in definite w ays; but if such general con cepts are sim ply co m ­ posite p h otograp h s o f all know n types o f m eaning, they can only blur, not clarify, the relation s that obtain am o n g specialized senses o f the w ord. C h arles Peirce, w ho w as p robably the first person to concern him self seriously with sem an tics, began by m akin g an inventory o f all “ sym bol-situ atio n s,” in the hope that when all possible m eanings o f “ m ean in g” were herded together, they w ould show em pirical differentia whereby one could divide the sheep from the goats. But the ob strep erou s flock, instead o f falling neatly into a few classes, each acco rdin g to its kind, divided and subdivided into the m ost terrifying order o f icons, qualisigns, legisigns, sem es, phem es, and delom es, and there is but cold com fort in his assuran ce th at his orginal 5 9 ,0 4 9 types can really be boiled dow n to a mere six ty -six .1 A few further attem pts were m ade to g rasp the essential quality o f m eaning by em pirical m ethods, but the m ore varieties could be foun d , the less did they prom ise to reveal a com m on essence. H u sserl, distinguishing each type o f m eaning a s a special notion, ended w ith as m any theories as there are “ m ean­ in g s.” 2 But we have still the sheep and the g o a ts and all their several relatives, and are still left w ondering why one fam ily nam e, M ean in g, should apply where no fam ily like­ ness can be detected. There is in fact no quality o f m eaning; its essence lies in the realm o f logic, where one d oes not deal with qualities, but only with relation s. It is not fair to say : “ M ean in g is a relatio n ,” for that suggests to o sim ple a busi­ ness. M o st people think o f a relation as a tw o-term ed affair - “ A -in-relation-to-B” ; but m eaning involves several term s, and different

types o f m eaning con sist o f different types and degrees o f relationship. It is better, perh aps, to say: “ M ean in g is not a q u ality, but a function o f a term .” A function is a pattern viewed with reference to one special term round which it centers; this pattern em erges when we look at the given term in its total relation to the other terms about it. The total m ay be quite com pli­ cated. For instance, a m usical chord m ay be treated a s a function o f one note, known as the “ w ritten b a s s,” by w riting this one note and indicating its relation to all the other notes that are to g o above it. In old organ m usic, the chord ^ w o u l d

be written: 6 4

3 which m eans: “ The A-chord with the sixth, the fourth and the third notes above A .” The chord is treated as a pattern surrounding and including A. It is expressed a s a function o f A. T he m eaning o f a term is, likew ise, a func­ tion; it rests on pattern, in which the term itself h olds the key-position. Flven in the sim plest kinds o f m eaning there m ust be at least tw o other things related to the term that “ m ean s” - an object that is “ m ean t,” an d a subject who uses the term ; just a s in a ch ord there m ust be at least tw o notes besides the “ w ritten b a s s ” to determ ine w hat the chord is (one o f these m ay be merely “ u n d ersto o d ” by m usicians, but w ithout it the com bin ation w ould not be a determ inate chord). The sam e m ay be said for a term with a m eaning; the existence o f a subject is often tacitly accepted, but if there is not at least one thing m eant and one mind for which it is m eant, then there is not a com plete m eaning - only a partial pattern which might be com pleted in different w ays. Any term in a pattern m ay be taken a s a key-term to which the oth ers are related. For instance, the chord

m ay be

regarded as a function o f its low est note, and expressed by the description

or it

m ay be treated with reference to the note on which it is built harm onically, which happens

134

SUSANNE K. LANGER

to be D. A m usician analyzing the harm ony w ould call this chord “ the second inversion o f the seventh-chord on the dom in an t, in the key o f Ci.” The “ d om in an t” o f that key is D , not A. He w ould treat the whole pattern as a func­ tion o f D ; that soun ds m ore com plicated than the other treatm ent, which fixed the notes from the A upw ard, but o f course it is not really so , because it com es to just the sam e pattern. Sim ilarly, we m ay view a m eaning-pattern from the point o f view o f any term in it, and our descriptions o f the sam e pattern will differ accordingly. We m ay say that a certain sym bol “ m ean s” an object to a person, or that the person “ m ean s” the object by the sym bol. The first description treats m eaning in the logical sense, the second in the p sych ologi­ cal sense. T he form er takes the sym bol as the key, and the latter the su b ject.1 So, the tw o m ost controversial kinds o f m eaning - the logical and the psychological - are distin ­ guished and at the sam e time related to each other, by the general principle o f viewing m eaning as a function , not a property , o f

terms. In the further an alyses that follow , “ m ean in g” will be taken in the objective sense, unless som e other is specified; that is to say, I shall speak o f term s (such as w ords) as “ m ean in g” som ething, not o f people as “ m eaning” this or that. Later we shall have to distinguish various subjective functions; but at present let us consider the relations o f terms to their objects. W hat relates the term s to their objects is, o f co urse, a subject; that is alw ays to be understood. There are, first o f all, tw o distinct functions o f term s, which have both a perfectly good right to the nam e “ m eaning” : for a significant soun d, gesture, thing, event (e.g. a flash, an im age), m ay be either a sign or a symbol. A sign indicates the existence - past, present, or future - o f a thing, event, or condition. Wet streets are a sign that it has rained. A patter on the roo f is a sign that it is raining. A fall o f the barom eter or a ring round the m oon is a sign that it is goin g to rain. In an unirrigated place, abu n dan t verdure is a sign that it often rains there. A smell o f sm oke signifies the pres­ ence o f fire. A scar is a sign o f a p ast accident.

D aw n is a herald o f sunrise. Sleekness is a sign o f frequent and plentiful food. All the exam p les here adduced are natural signs. A natural sign is a part o f a greater event, or o f a com plex condition, and to an experienced observer it signifies the rest o f that situation o f which it is a n otable feature. It is a symptom o f a state o f affairs. The logical relation between a sign and its object is a very sim ple one: they are associated, som ehow , to form a pair ; that is to say, they stand in a one-to-one correlation. T o each sign there correspon ds one definire item which is its object, the thing (or event, or condition) signi­ fied. All the rest o f that im portant function, signification, involves the third term , the subject, which uses the pair o f item s; and rhe relation o f the subject to the other tw o term s is much m ore interesting than their ow n bare logical coupling. The subject is related, essen­ tially, to the other tw o term s as a pair. W hat characterizes them is the fact that they are paired. T h u s, a w hite bum p on a person ’s arm , as a mere sense-datum , w ould probably not be interesting enough even to have a nam e, but such a datum in its relation to the past is noted and called a “ s c a r.” N o te, how ever, that although the subject’s relation is to the pair o f other term s, he h as also a relation to each one o f them individually, which m akes one o f them the sign and the other the object. W hat is the difference between a sign and its object, by virtue o f which they are not interchangeable? T w o term s merely associated as a pair, like tw o sock s, tw o balances o f a scale, tw o ends o f a stick, etc., could be interchanged w ithout any harm. The difference is, that the subject for which they constitute a pair m ust find one more

interesting than the other, an d the latter more easily available than the former. If we are inter­ ested in to m orrow ’s w eather, the events now present, if coupled with to m orrow ’s weatherphenom ena, are signs for us. A ring round the m oon, or “ m ares’ ta ils” in the sky, are not im portant in them selves; but as visible, present items coupled with som ething im portant but not yet present, they have “ m eaning.” If it were not for the subject, or interpretant , sign and object w ould be interchangeable. T hunder m ay ju st as well be a sign that there has been

THE LOGIC OF SIGNS AND SYMBOLS

lightning, as lightning m ay signify that there will be thunder. In them selves they are merely correlated. It is only where one is perceptible and the other (harder or im possible to per­ ceive) is interesting, that we actually have a case o f signification belonging to term.4 N o w , just as in nature certain events are correlated, so that the less im portant may be taken as signs o f the m ore im portant, so we m ay also produce arbitrary events p u r­ posely correlated with im portant ones that are to be their m eanings. A whistle m eans that the train is ab o u t to start. A gunshot m eans that the sun is just setting. A crepe on the d oo r m eans som eone has just died. These are artificial signs, for they are not part o f a condition o f which they naturally signify the rem ainder or som ething in the rem ainder. Their logical relation to their objects, how ever, is the sam e as that o f natural signs - a one-toone correspondence o f sign and object, by virtue o f which the interpretant, w ho is inter­ ested in the latter an d perceives the form er, m ay apprehend the existence o f the term that interests him. The interpretation o f signs is the basis o f anim al intelligence. A nim als presum ably do not distinguish between natural signs and arti­ ficial or fortuitous signs; but they use both kinds to guide their practical activities. We do the sam e thing all day long. We answ er bells, w atch the clock, obey w arning signals, follow a rrow s, take o ff the kettle when it w histles, com e at the bab y’s cry, close the w indow s when we hear thunder. The logical basis o f all these interpretations, the mere correlation o f trivial events with im portan t ones, is really very sim ple and com m on; so much so that there is no limit to w hat a sign m ay m ean. This is even m ore obviously true o f artificial signs than o f natural ones. A shot m ay mean the beginning o f a race, the rise o f the sun, the sighting o f danger, the com m encem ent o f a parad e. As for bells, the w orld is m ad with their m essages. Som ebody at the front d oor, the back d o o r, the side d o o r, the telephone to ast is ready - typew riter line is ended - school begins, w ork begins, church begins, church is over - street car starts - cash b ox registers knife grinder p asses - time for dinner, time to get up - fire in tow n!

135

Because a sign m ay m ean so m ay things, we are very ap t to m isinterpret it, especially when it is artificial. Bell signals, o f co urse, m ay be either w rongly associated with their ob jects, or the sound o f one bell m ay actually be confused with that o f another. But n atural signs, too, m ay be m isunderstood. W et streets are not a reliable sign o f recent rain if the sprinkler w agon has passed by. The m isinterpretation o f signs is the sim plest form o f mistake. It is the m ost im portant form , for purposes o f p rac­ tical life, and the easiest to detect; for its norm al m anifestation is the experience called

disappointment. W here we find the sim plest form o f error, we m ay expect to find a lso , a s its correlate, the sim plest form o f know ledge. T h is is, indeed, the interpretation o f signs. It is the m ost ele­ m entary and m ost tangible sort o f intellection; the kind o f know ledge that we share with an im als, that we acquire entirely by experi­ ence, th at has obvious biological uses, and equally ob vious criteria o f truth and false­ hood. Its m echanism m ay be conceived as an elabo ration o f the conditioned-reflex arc, with the brain doin g sw itchboard duty, and getting the right or the w rong num ber for the sense organ that called up the m usculature and expects an answ er in term s o f altered sen sa­ tions. It has all those virtues o f sim plicity, compon ability, and intelligibility that recom m end a concept for scientific purp oses. S o it is not surprising that students o f genetic psychology have seized upon sign interpretation as the archetype o f all know ledge, th at they regard signs as the original bearers o f m eaning, and treat all other term s with sem antic properties as subspecies - “ substitute sig n s,” which act as p rox y for their objects and evoke conduct ap p ro p riate to the latter instead o f to th e m se lv es.. . . H ow ever, the ch aracteristics of sym bols in general, and their essential differ­ ence from signs, m ust go on record. A term which is used sym bolically and not signally does not evoke action ap p ro p riate to the presence o f its object. If I say: “ N a p o le o n ,” you d o not bow to the con qu eror o f Europe as though I had introduced him , but merely think o f him. If I m ention a M r. Sm ith o f our com m on acquain tan ce, you m ay be led to tell me som ething ab o u t him “ behind his b a c k ,”

136

SUSANNE K. LANGER

which is just w hat you w ould not d o in his presence. T h u s rhe sym bol for M r. Sm ith - his nam e - m ay very well initiate an act a p p ro p ri­ ate peculiarly to his absence. R aised eyebrow s and a look at the d o o r, interpreted as a sign that he is com in g, w ould stop you in the m idst o f your narrative; that action w ould be directed tow ard M r. Smith in person. Sym bols are not p roxy for their ob jects, but are vehicles for the conception o f objects. T o conceive a thing or a situation is not the sam e thing a s to “ react to w ard it” overtly, or to be aw are o f its presence. In talking about things we have con ception s o f them , not the things them selves; an d it is the conceptions , not the things , that sym bols directly "m ean .” Behav­ ior tow ard con ception s is what w ords n or­ m ally evoke; this is the typical process o f thinking. O f course a w ord m ay be used as a sign, but that is not its prim ary role. Its signific ch arac­ ter has to be indicated by som e special m od i­ fication - by a tone o f voice, a gesture (such as pointing or starin g), or the location o f a placard bearin g the w ord. In itself it is a sym bol, associated with a conception, not directly with a public object or event. The fundam ental difference between signs and sym bols is this difference o f associatio n , and consequently o f their use by the third party to the m eaning function, the subject; signs announce their objects to him, w hereas sym bols lead him to conceive their objects. The fact that the sam e item - say, the little m outhy noise we call a “ w o rd ” - m ay serve in either cap acity, does not obliterate the cardinal distinction between the tw o functions it m ay assum e. The sim plest kind o f sym bolistic m eaning is prob ab ly that which belongs to proper nam es. A personal nam e evokes a conception o f som e­ thing given a s a unit in the subject’s e xp eri­ ence, som ething concrete and therefore easy to recall in im agin ation. Because the nam e belongs to a notion so obviously and unequivocally derived from an individual object, it is often supposed to “ m ean ” that object a s a sign w ould “ m ean ” it. T h is belief is reinforced by the fact that a nam e borne by a living person alw ays is at once a sym bol by which we think o f the p erson , and a call-nam e by which we

signal him. T h rough a con fusion o f these tw o functions, the proper nam e is often deem ed the bridge from anim al sem antic, or sign-using, to hum an lan guage, which is sym bol-using. D ogs, we are to ld , understand nam es - not only their ow n, but their m asters’ . S o they d o , indeed; but they understand them only in the capacity o f call-names. If you say “ Ja m e s ” to a d o g w hose m aster bears th at nam e, the d og will interpret the soun d as a sign, and look for Ja m es. Say it ro a person w ho know s som eone called thus, and he will ask : “ W hat ab o u t Ja m e s ? ” T h at sim ple question is forever beyond the dog; signification is the only m eaning a nam e can have for him - a m eaning which the m aster’s nam e shares with the m aster’s sm ell, with his fo o tfall, and his ch aracteristic ring o f the doorbell. In a hum an being, how ever, the nam e evokes the conception o f a certain m an so called, and prepares the mind for further conceptions in which the notion o f that m an figures; therefore the hum an being naturally ask s: “ W hat ab o u t Ja m e s ? ” There is a fam o u s p assage in the a u to b io ­ graph y o f Helen Keller, in which this rem ark­ able w om an describes the daw n o f Language upon her m ind. O f course she had used signs before, form ed a sso ciatio n s, learned to expect things and identify people or places; but there w as a great day w'hen all sign-m eaning w as eclipsed and dw arfed by the discovery that a certain datum in her lim ited sense-w orld had a denotation , that a particular act o f her fingers constituted a word. T h is event had required a long preparation ; the child had learned m any finger acts, but they were as yet a m eaningless play. T hen, one day, her teacher took her out ro w alk - and rhere the great advent o f L a n ­ guage occurred. “ She brought me my h at,” the m em oir reads, and I knew I w as going out into the warm sunshine. This thought, if a wordless sensa­ tion may be called a thought, made me hop and skip with pleasure. We walked down rhe path to the wellhouse, attracted by the fragrance o f the hon­ eysuckle with which it w as covered. Some one w as drawing water and my teacher placed my hand under the spout. As the cool stream gushed over my hand she spelled into the

THE LOGIC OF SIGNS AND SYMBOLS

other the word water, first slowly, then rapidly. I stood still, my whole attention fixed upon the motion of her fingers. Suddenly I felt a misty consciousness as of something forgotten - a thrill o f returning thought; and somehow the mystery of language was revealed to me. 1 knew then that w-a-t-e-r meant the wonderful cool something that was flowing over my hand. That living word awakened my soul, gave it light, hope, joy, set it free! There were harriers still, it is true, but barriers that in time could be swept away. I left the well-house eager to learn. Every­ thing had a name, and each name gave birth to a new thought. As we returned to the house every object which I touched seemed to quiver with life. That w as because I saw everything with the strange, new sight that had come to me.5 T h is p assage is the best affidavit we could hope to find for the genuine difference between sign and sym bol. T h e sign is som ething to act upon , or a m eans to com m an d action; the sym bol is an instrum ent o f thought. N ote how M iss Keller qualifies the m ental process just preceding her discovery o f w ord s - “ This th ought, if a wordless sensation may be called a thought . ” Real thinking is possible only in the light o f genuine lan guage, no m atter how lim ited, how prim itive; in her case, it becam e possible with the discovery that “ w-a-t-e-r” w as not necessarily a sign that w ater w as w anted or expected, but w as the name o f this su bstan ce, by which it could be m entioned, conceived, rem em bered. Since a nam e, the sim plest type o f sym bol, is directly associated with a conception, and is em ployed by a subject to realize the co n ­ ception, one is easily led to treat a nam e as a “ conceptual sig n ,” an artifical sign which announces the presence o f a certain idea. In a sense this is quite justified; yet it strikes a strained an d unnatural note, which is usually a fair w arning that the attem pted interpreta­ tion m isses the m ost im portan t feature in its m aterial. In the present case, it m isses the rela­ tion o f conceptions to the concrete world , which is so close and so im portan t that ir enters into the very structure o f “ n am es.” A nam e, above all, denotes som ething. “ Ja m e s ” m ay represent a con ception , but it names a

137

certain person. In the case o f proper nouns this relation o f the sym bol to w h at it denotes is so strikin g that denotation has been confused with the direct relation o f sign and object, signification. As a m atter o f fact, “ Ja m e s ” does not, w ithout further ad o , signify a p erson; it denotes him - it is asso ciated with a con cep­ tion which “ fits” the actual person. T he rela­ tion between a sym bol and an object, usually expressed by “ S denotes O ,” is not a sim ple tw o-term ed relation which S has to O ; it is a com plex affair: S is co u p led, for a certain subject, with a conception th at fits O , i.e. with a notion which O satisfies. In an ordinary sign-function, there are three essential term s: subject, sign, and object. In d en o tation , which is the co m m o n est kind o f sym bol-function, there have to be four: subject, sym bol, conception, and ob ject. T he radical difference between sign-m eaning and sym bolm eaning can therefore be logically exhibited, for it rests on a difference o f pattern, it is strictly a different fu n c tio n / D enotation is, then, the co m p lex relation­ ship which a nam e has to an object which bears it; but w hat shall the m ore direct relation o f the nam e, or sym bol, to its associated concept be called? It shall be called by its tra­ ditional nam e, connotation. T he con notation o f a w ord is the conception it conveys. Because the connotation rem ains with the sym bol when the ob ject o f its denotation is neither present nor looked for, we are able to think about the object w ithout reacting to it overtly at all. H ere, then, are the three m ost fam iliar m eanings o f the one w ord, “ m ean in g” : signi­ fication, denotation , and con notation . All three are equally and perfectly legitim ate, but in no possible w av interchangeable. I--.] So m uch, then, for the venerable “ logic o f term s.” It ap p ears a little m ore com plicated than in the m edieval b o ok s, since we m ust add to the long-recognized functions, connotation and den otation , a third on e, signification, which is fundam entally different from the other tw o; and since, m oreover, in discussin g the sem antic functions o f term s we have m ade the rare discovery that they really are functions , not pow ers or m ysterious p ro p ­ erties or w hat-not, and have treated them

138

SUSANNE K. LANGER

accordingly. T he traditional “ logic o f term s” is really a m etaphysic o f m eaning; the new philosophy o f m eaning is first o f all a logic o f term s - o f signs and sym bols - an an alysis o f the relational patterns in which “ m ean in g” may be sought. But a sem antic o f separate sym bols is only a rudim entary foun dation for a m ore interest­ ing aspect o f m eaning. Everything is mere p ro ­ paedeutic until we com e to discourse. It is in discursive thinking that truth and falsehood are born. Before term s are built into p ro p o si­ tions, they assert nothing, preclude nothing; in

fact, although they m ay name things, and convey ideas o f such things, they say nothing. 1 have discussed them at such great length sim ply because m ost logicians have given them such cavalier treatm ent that even so obvious a distinction as that between sign-functions and sym bol-functions passed unnoticed; so that careless ph ilosoph ers have been guilty o f letting am bitious genetic psych ologists argue them from the conditioned reflex to the w isdom o f G. Bernard Shaw , all in one skyrocketing generalization.

NOTES

4

1

5

2

3

From tw o letters to Lady W elby, 1904 and 1908 respectively, first cited by O gden and R ich ard s in The Meaning o f Meaning (A pp. D , pp. 4 3 5 —44), and now published in The Collected Papers o f Charles S. Peirce (1 9 3 2 ), II, 33 0 . Edm und H u sserl, Logische Untersuchungen , 2 vols. (1 9 1 3 and 1 9 2 1 ), vol. II, part I, passim . Where the ob ject is taken a s the key, the resulting description begins with the “ know ledge-content” postulated in som e epistem ologies.

6

Cf. W hitehead, Symbolism [1 927], pp. 9 -1 3 . Helen Keller, The Story o f My Life (1 9 3 6 ; 1st ed. 1902), pp. 2 3 - 4 . If a sym bol could be said norm ally to “ signify” anything, its object w ould be the occurrence o f an act o f conception. But such a function o f a sym bol is casu al, and crosses with its use as a symbol. In the latter function it is not the act o f con cep­ tion, but what is conceived , that enters into the m eaning-pattern. W e shall avoid m uch confusion and q uib blin g by recognizing that signification does not figure in sym bol­ ization at all.

11

The Problem of Symbols E. E. Evans-Pritchard

E. E. Evans-Pritchard (1902-73), or "E.-P.," as he was know n, was arguably the most im portant anthropologist of his generation, certainly in the United Kingdom (he was professor at the University of Oxford). He was a superb fieldw orker and theoretician and his theory is often implicit in his ethnographic exposition. Trained by M alinowski and influenced by RaddiffeBrown, he was an original and indepen­ dent thinker w ho shifted from a prim arily structure-functionalist to a more historical and hum anistic posi­ tion. W hile many anthropologists have taken religious symbols as complex vari­ ants of ordinary symbols as defined by Langer, thus as "packets" of meaning that could be unpacked (see chapters 12 and 13, respectively, as w ell as Turner 1967, for notable unpackings), EvansPritchard is unusual in follow ing more clearly the point made by Langer that

the relevant question is one of predica­ tion. In the selection from N u er Religion reprinted here, Evans-Pritchard looks less at the content of the symbol (for which see the profound analysis in his preceding chapter on the refractional nature of Nuer deity, Kw oth) than on the meaning of the verb that establishes it as a predicate. He is concerned, then, w ith w hat it means for Nuer to say that tw ins are birds or w hat it means w hen Nuer replace cattle w ith cucumbers in their sacrifices. The question is thus not the m eaning of twins, birds, oxen, or cucumbers qua symbols but w hat the Nuer mean w hen they say and do certain things about and w ith twins, birds, oxen, and cucumbers. In this very subtle essay Evans-Pritchard thus takes us to the complexities of religious meaning and the intricate gradations betw een literal and m etaphorical predication. It is sometimes remarked th at EvansPritchard portrays Nuer religion in a

From E. E. Evans-Pritchard, “ The Problem of Sym bols,” in N uer Religion (Oxford: Clarendon Press, 1956), pp. 123-43. Reprinted by permission o f Oxford University Press. Abridged.

140

E. E. EVANS-PRITCHARD

m anner not entirely dissimilar to Roman Catholicism, or rather that the argu­ ments he makes to explain Nuer refrac­ tions of spirit or the m aterial symbolism of spirit are not altogether d ifferent from the ways in w hich Catholic th eo lo ­ gians m ight defend the concept of the Trinity or the Com m union. This is not necessarily a criticism; some people have gone on to suggest th at anthropologists w ho practice religion can have a better appreciation of the religion of their sub­ jects than the agnostic majority. W hile this may be true in the case of EvansPritchard, I doubt the general validity of the argum ent. M oreover, w hile a reli­ gious sensibility is certainly evident in this essay, more to the point is EvansPritchard's philosophical acuity and eth­

In the last chapter I discussed how the N uer conception o f Spirit is figured in different w ays to different person s and categories and grou p s. In this chapter I consider the m aterial form s in which Spirit m anifests itself or is represented. G o d is, properly speaking, not figured in any m aterial representations, nor are alm o st all the spirits o f the above, though both Clod and his supra-terrestrial refractions m ay reveal them selves in signs. But the spirits o f the below are represented in creatures and things. O ur problem chiefly concerns these spirits o f the below . It can be sim ply stated by the question: W hat m eaning are we to attach to N u er state­ ments that such-and-such a thing is kw oth , spirit? T he answ er is not so sim ple. There are several w ays in which w hat we w ould render as “ is” is indicated in the N uer language. The one which concerns us here is the particle e. It is used to tell the listener that som ething belongs to a certain c lass or cate­ gory and hence ab o u t som e ch aracter or quality it has, a s “ e d it” > “ it is a bird ” , “ gat nath e car ” , “ the N u er is b la ck ” , and uD uob e ram me g o a g h '\ “ D u ob is a go o d m an .’ T he q u e s­ tion we are ask in g is w hat m eaning or m ean ­ ings it has for N uer when they say o f som ething

nographic precision. I w ould, however, make the broader herm eneutic point that any anthropological interpretation will be inform ed by the position of the anthropologist, not only by the ethnog­ rapher's own religious form ation but by the context of the debate over religious, philosophical, or political issues in which the anthropologist is immersed w ithin his or her own tim e and tradition. EvansPritchard's professional debate, as w e glimpse tow ard the end of his essay, was w ith Tylor, Levy-Bruhl, and others w ho made argum ents th at denigrated either the conclusions or the very rationality of "prim itive tho ug ht." As discussed above, Evans-Pritchard's oeuvre on religion is a masterly refutation of these arguments.

“ e kw oth ” , “ it is S p irit” (in the sense either o f Ciod or o f a divine refraction). N uer d o not claim to see G o d , nor d o they think that anyone can know w hat he is like in him self. When they speak about his nature they d o so by adjectives which refer to attri­ butes, such as “ g r e a t” and “ g o o d ” , or in m etaph ors taken from the w orld aroun d them , likening his invisibility and ubiquity to wind and air, his greatn ess to the universe he has created, and his gran deu r to an o x with w ide­ spread horns. They are no m ore than m eta­ ph ors for N u er, w ho do not say that any o f these things is Ciod, but only that he is like (cere) them. They express in these poetic im ages a s best they can what they think m ust be som e o f his attributes. N evertheless, certain things are said, or m ay be said , “ to b e” Ciod - rain, lightning, and various other natural - in the N uer w ay o f speech, created - things which are o f com m on interest. There is here an am biguity, or an obscurity, to be elucidated, for N u er are not now saying that G o d or Spirit is like this or that, but that this or that “ is” Ciod or Spirit. Elucidation here does not, how ever, present great difficulties.

THE PROBLEM OF SYMBOLS

G o d being conceived o f a s in the sky, those celestial phenom ena which are o f particular significance for N u er, rain and lightning, are said , in a sense we have to determ ine, to be him. There is no noun denoting either phe­ nom enon and they can only be spoken o f by verbs indicating a function o f the sky, as *ce nhiai deam ” , “ the sky rain ed ” , and “ ce nbial m ar” , “ the sky th undered” . A lso pestilences, m urrain s, death, and indeed alm ost any natural phenom enon significant fo r men are co m ­ m only regarded by N uer as m anifestations from above, activities o f divine being. Even the earthly totem s are conceived o f as a relation ­ ship deriving from som e sin gu lar intervention o f Spirit from above in hum an affairs. It is chiefly by these signs that N uer have know l­ edge o f G o d . It m ight be held, therefore, that the N uer conception o f G o d is a con ceptual­ ization o f events w hich, on account o f their stran gen ess or variability a s well as on account o f their potentiality for fortune or m isfortune, arc said to be his activities or his activities in one or other o f his h yp ostases or refractions. S u p p ort for such a view m ight be found in the w ay N uer som etim es sp eak o f one or other o f these effects. They m ay say o f rain or lightning or pestilence ue kw oth ” , “ it is G o d ” , and in storm s they pray to G od to com e to earth gently and not in fury - to com e gently, it will be noted, not to m ake the rain com e gently. I d o not d iscu ss this on tological question here beyond saying that were we to suppose th at such phenom ena are in them selves re­ gard ed as G od we w ould m isunderstand and m isrepresent N uer religious thought, which is pre-em inently dualistic. It is true that for them there is no abstract duality o f natural and supern atu ral, but there is such a duality between kwoth , Spirit, which is im m aterial rather than supern atu ral, and cak , creation, the m aterial w orld know n to the senses. Rain and lightning and pestilences and m urrains belong to this created w orld and are referred to by N u er as nyin kw oth , instrum ents of God. N evertheless, they and other effects o f sig­ nificance fo r men are SioarjHta, signs or m an i­ festation s o f divine activity; and since N uer apprehend divine activity in these signs, in G o d ’s revelation o f him self to them in m aterial

141

form s, the signs are, in a low er m edium , w hat they signify, so that N uer m ay say o f them “ e kw oth ” , “ it is G o d ” . Rain and pestilence com e from G o d and are therefore m an ifestation s of him , and in this sense rain and pestilence are G o d , in the sense that he reveals him self in their falling. But though one can say o f rain or pestilence that it is G o d one can n ot say o f G od that he is rain or pestilence. T h is w ould m ake n o sense for a num ber o f reason s. In the first place, the situation could scarcely arise, G od n ot being an observable ob ject, in which N uer w ould require or desire to say ab o u t him that he is anything. In the second place, the w ord kwoth does not here refer to a particular refraction o f Spirit, a spirit, but to Spirit in its oneness, G o d , and he could not be in any way identified with any one o f his m anifestations to the exclusion o f all the others. A third, and the m ost cogen t, reason is th at rain is w ater which falls from the sky and pestilence is a bodily condition and they are therefore in their nature m aterial things and not Spirit. Indeed, a s a rule, rain is only thought o f in connexion w ith Spirit, and is therefore only said to be Spirit, when it does not fall in due season or falls too much or to o violently with storm and lightning - w hen, that is, the rain has som e special significance fo r hum an affairs. This gives us a clue to w h at is m eant when N uer say o f som ething that it is G o d or that it is a spirit o f the air, as thunder m ay be said to be the spirit win or a prophet o f the spirit deng m ay be said to be deng ~ especially as N uer readily exp an d such statem ents by ad d in g that thunder, rain, and pestilence are all instru­ m ents (nyin) o f G od or that they are sent by (jak) Ciod, and that the spirit deng has filled (gwang) the prophet through w hom it sp eaks. In the statem ent here that som ething is Spirit or a spirit the particle e , which we translate “ is” , can not therefore have the m eaning of identity in substantial sense. Indeed, it is because Spirit is conceived o f in itself, as the creator and the one, and quite ap art from any o f its m aterial m an ifestation s, that phenom ena can be said to be sent by it or to be its instru­ m ents. When N uer say o f rain or lightning that it is G o d they are m aking an elliptical state­ ment. W hat is understood is not that the thing in itself is Spirit but that it is w h at we w ould

142

E. E. EVANS-PRITCHARD

call a m edium or m anifestation or sign o f divine activity in relation to men and o f sig­ nificance for them . W hat precisely is posited by the hearer o f any such elliptical statem ent depends on the nature o f the situation by re­ ference to which it is m ade. A vulture is not thought o f as being in itself Spirit; it is a bird. But if it perches on rhe crow n o f a byre or hut N uer m ay say “ e kwoth ” , “ it is S p irit” , m eaning rhat its d oin g so is a spiritual signal presaging disaster. A lion is nor thought o f as being in itself Spirit; it is a beast. But it m ay, on account o f som e event which brings it into a peculiar relation to m an, such a s being born, as N uer think som etim es happens, as twin to a hum an child, be regarded as a revelation o f Spirit for a particular fam ily and lineage. Like­ wise, d iseases, or rather their sym ptom s, are not thought o f as being in them selves Spirit, but their appearan ce in individuals m ay be regarded as m an ifestation s o f Spirit for those individuals. Spirit acts, and thereby reveals itself, through these creatures. This distinction between the nature o f a thing and w hat it m ay signify in certain situations or for certain persons is very evident in totem ic relation ­ ships. A crocodile is Spirit for certain person s, but it is not thought to be in its nature Spirit, for others kill and eat it. It is because N uer separate, and quite explicitly when questioned ab o u t the m atter, spiritual conceptions from such m aterial things as m ay nevertheless be said “ to be” the con ception s, that they are able to m aintain the unity and auton om y o f Spirit in spite o f a great diversity o f accidents and are able to sp eak o f Spirit without reference to any o f its m aterial m anifestations. So far I have been m ostly speaking o f the conception o f G od and o f those o f his refrac­ tions which belong to the category o f the sky or o f the above. With tw o possible excep tio n s,1 we can n ot say that the things said “ to be” these spirits are m aterial sym bols or represen­ tations o f them ; at any rate not in the sam e sense as we can speak o f things being sym bols o f those lesser refractions o f Spirit N uer call spirits o f the earth or o f the below , in which G od stan ds in a special relationship to lineages and individuals - such diverse things as beasts, birds, reptiles, trees, phosphorescent objects, and pieces o f w oo d . These lesser refraction s o f

Spirit, regarded as distinct spirits in relation to each other, cann ot, unlike the spirits o f the air, easily be thought o f except in relation to the things by reference to which they derive their individuality, and which are said “ to b e” them. W hen, therefore, N uer say that the pied crow is the spirit buk or that a snake is Spirit, the w ord “ is” has a different sense from w hat it has in the statem ent that rain is Spirit. The difference does not merely lie in the fact that kwoth has here a m ore restricted con notation , being spoken o f in reference to a particular and exclusive refraction - a spirit - rather than com prehensively a s G od or Spirit in its oneness. It lies a lso in the relation understood in the statem ent between its subject (snake or crow ) and its predicate (Spirit or a spirit). The snake in itself is not divine activity w hereas rain and lightning are. The story accounting for a totem ic relationship m ay present it as arising from a revelation o f divine activity, but once it has becom e an established relationship between a lineage an d a natural species, the species is a representation or sym bol o f Spirit to the lineage. W hat then is here m eant when it is said that the pied crow “ is” buk or that a snake “ is” Spirit: that the sym bol “ is” w hat it sym bolizes? C learly N uer do not m ean that the crow is the sam e a s buk , for buk is also co n ­ ceived o f as being in the sky and a lso in rivers, which the pied crow certainly is not; nor that a snake is the sam e as som e spiritual refrac­ tion, for they say that the snake just craw ls on the earth while the spirit it is said to be is in the sky. W hat then is being predicated abo u t the crow or snak e in the statem ent that either is Spirit or a spirit? It will be sim pler to discuss this question in the first place in relation to a totem ic relation­ ship. When a N uer says o f a creature ue nyang ” , “ it is a cro cod ile” , he is saying that it is a crocodile and not som e other creature, but when he say s, to explain why a person behaves in an unusual m anner tow ards crocodiles ‘V kwothdien ” , “ it (the crocodile) is their sp irit” , he is obviously m akin g a different sort o f state­ ment. H e is not sayin g w hat kind o f creature it is (for it is understood that he is referring to the crocodile) but that w hat he refers to is Spirit for certain people. But he is also not

THE PROBLEM OF SYMBOLS

saying that the crocodile is Spirit - it is nor so for him - but that certain people so regard it. Therefore a N uer w ould not m ake a general statem ent that “ nyang e kw otbny ‘‘crocodile is Sp irit” , but w ould only say, in referring to the crocodile, “ e kwoth ” , “ it is S p irit” , the d istin c­ tion between the tw o statem ents being that the first w ould m ean that the crocodile is Spirit for everyone w hereas the second, being m ade in a special context o f situ ation , m eans that it is Spirit for certain person s w ho are being d is­ cu ssed, or are understood, in that context. Likew ise, whilst it can be said o f the crocodile that it is Spirit, it cannot be said o f Spirit that it is the crocodile, or rather, if a statem ent is fram ed in this form it can only be m ade when the w ord kwoth has a pronom inal suffix which gives it the m eaning o f “ his sp irit” , “ their sp irit” , and so forth; in other w ord s, where the statem ent m akes it clear that w hat is being spoken o f is Spirit conceived o f in relation to particular persons only. We still have to ask , how ever, in w hat sense the crocodile is Spirit for these persons. Since it is difficult to discu ss a statem ent that som ething which can be observed, crocodile, is som ething m ore than w hat is ap p ears to be when this som ething m ore, Spirit, cannot be observed, it is helpful first to consider tw o exam ples o f N uer statem ents that things are som ething m ore than they ap p ear to be when both the subject term and the predicate term refer to observable phenom ena. When a cucum ber is used as a sacrificial victim N uer speak o f it as an o x . In doing so they are assertin g som ething rather m ore than that it takes the place o f an o x . They do not, o f course, say that cucum bers are oxen , and in sp eakin g o f a particular cucum ber as an ox in a sacrificial situation they are only indicating that it m ay be thought o f as an o x in that particular situation; and they act accordingly by perform ing the sacrificial rites as closely as possible to w hat happens when the victim is an o x . The resem blance is con ceptual, not per­ ceptual. The uis” rests on qualitative analogy. And the expression is asym m etrical, a cucum ­ ber is an o x , but an o x is not a cucum ber. A rather different exam ple o f this way o f speakin g is the N uer assertion that tw ins are one person and that they are b ird s.2 When they

143

say “ tw ins are not tw o persons, they are one p e rso n ” they are not saying that they are one individual but that they have a single p erson al­ ity. It is significant that in speaking o f the unity o f tw ins they only use the w ord ran , which, like ou r w ord “ p erson ” , leaves sex, age, and other distinguishing qualities o f individuals undefined. They w ould not say that tw ins o f the sam e sex were one dhol , boy, or one nyal, girl, but they d o say, whether they are o f the sam e sex or not, that they are one ran , person. T heir single social personality is som ething over and above their physical duality, a duality which is evident to the senses and is indicated by the plural form used when speaking o f tw ins and by their treatm ent in all respects in ordinary social life as tw o quite distinct indi­ viduals. It is only in certain ritual situations, and sym bolically, that the unity o f tw ins is expressed, particularly in cerem onies co n ­ nected with m arriage and death, in which the personality undergoes a change. T h u s, when the senior o f m ale tw ins m arries, the junior acts with him in the ritual acts he has to perform ; fem ale twins ought to be m arried on the sam e day; and no m ortuary cerem onies are held for tw ins because, for one reason , one o f them cann ot be cut o ff from the living w ithout the other. A w om an w hose twin brother had died som e time before said to M iss Soule, to w hom I am indebted for the in form ation , “ Is not his soul still living? I am alive, and we are really children o f C iod.” There is no m ortuary cerem ony even when the second twin dies, and I w as told that twins do nor attend the m ortuary cerem onies held for their dead kin sfolk, nor m ourn them, because a twin is a ran nhial , a person o f the sky or o f the above. He is also spoken o f as gat kwoth , a child o f Ciod. T hese dioscuric descriptions o f tw ins are com m on to many peoples, but the N uer are peculiar in holding also that they are birds. They say “ a twin is not a person (ran), he is a bird (dit)n, although, as we have just seen, they assert, in another sense, that tw ins are one person (ran). Here they are using the w ord ran in the sense o f a hum an being as distinct from any other crea­ ture. The dogm a is expressed in various w ays. Very often a twin is given the p roper nam e D/r, bird, Gwong, guineafow l, or N gec , fran colin .1

144

E. E. EVANS-PRITCHARD

All N u er consider it sham eful, at any rate for adults, to eat any sort o f bird or its eggs, but w ere a twin to d o this it w ould be much m ore than sham eful. It w ould be nueer, a grave sin, for tw ins respect (tbek) birds, because, N uer say, birds are a lso tw ins, and they avoid any sort o f con tact with them . The equivalence o f tw ins and birds is expressed particularly in connexion with death. When an infant twin dies people say *c e p a r \ “ he has flown a w a y ” , using the w ord denoting the flight o f birds. Infant tw ins w ho die, as so often h ap p en s, are not buried, a s other infants are, but are covered in a reed basket or w innow ing-tray and placed in the fork o f a tree, because birds rest in trees. I w as told that birds which feed on carrion w ould not m olest the bodies but w ould look at their dead kinsm en - tw ins and birds are a lso said to be kin, though the usage m ay be regarded as m etaph orical - and fly aw ay again. When I asked a N uer whether adult tw ins w ould be buried like other people he replied “ no, o f course not, they are birds and their souls g o up into the a ir ” . A p latform , not used in the norm al m ode o f burial, is erected in the grave and a hide placed over it. The body is laid on this hide and covered with a second hide. Earth is then carefully patted over the upper hide in stead o f being shovelled in quickly, a s in the burial o f an ordin ary person. I w as told that the co rp se is covered with earth lest a hyena eat it an d afterw ard s drink at a p oo l, for men m ight drink at the sam e pool and die from con tam in ation (nueer). It is understandable that N uer draw an an alogy between the m ultiple hatching o f eggs and the dual birth o f tw ins. T he an alogy is explicit, an d, through an extension o f it, the flesh o f crocodiles and turtles is a lso forbidden to tw ins on the groun d thar these creatures to o, like birds, lay eggs. M iss Soule once had a girl twin in her household w ho refused fish for the sam e reason - the only case o f its kind know n to either o f us. But the an alogy between m ultiple births in birds and men d oes not adequately explain why it is with birds that hum an tw ins are equated when there are m any other creatures which habitually bear several young at the sam e rime and in a m anner m ore closely resem bling hum an p arturition . It can ­ not be just m ultiple birth which leads N u er to

say that tw ins are birds, fo r these other crea­ tures are not respected by tw ins on that account. The proh ibition on eating eggs is clearly secondary, and it is extended to include crocodiles and turtles - and by M iss Soule’s girl fish a lso - not because they lay eggs but because their laying eggs m akes them like birds. M oreover, it is difficult to understand why a resem blance o f rhe kind should in any case be m ade so much of. The m ultiple hatch­ ing of chicks is d ou b tless a resem blance which greatly strengthens the idea o f tw ins being birds, but it is only part o f a m ore com plex an alogical representation which requires to be explained in m ore general term s o f N uer reli­ giou s thought. A tw in, on account o f his pecu­ liar manner o f conception is, though not Spirit him self, a special creation, an d, therefore, m anifestation o f Spirit; and when he dies his sou! goes into the air, to which things asso c i­ ated with Spirit belong. He is a ran tibtal, a person o f the a b o v e, w hereas an ordin ary person is a ran piny , a person o f the below . A bird, though also not in itself Spirit, belongs by nature to the abo ve and is a lso w hat N uer call, using “ p e rso n ” m etaphorically, a ran nhial, a person o f the ab o ve, and being such is therefore a lso asso ciated with Spirit. It can n ot, o f course, be determ ined for certain whether a twin is said to be a person o f the above because he is a bird or whether he is said to be a bird because he is a person o f the above, but rhe connexion in thought between tw ins and birds is certainly not sim ply derived from the m ulti­ ple birth sim ilitude but also , and in my view prim arily, from both birds and twins being classed by N uer a s gaat kwoth , children o f G o d . Birds are children o f Ciod on account o f their being in the air, and tw ins belong to the air on accoun t o f their being children o f Ciod by the m anner o f their conception and birth. It seem s o d d , if not ab su rd , to a E uropean when he is told that a twin is a bird as though it were an o b v io u s fact, for N uer are not saying that a twin is like a bird but that he is a bird. There seem s to be a com plete contradiction in the statem ent; and it w as precisely on state­ m ents o f this kind recorded by observers o f prim itive peoples that Levy-Bruhl based his theory o f the prelogical m entality o f these peoples, its chief characteristic being, in his

THE PROBLEM OF SYMBOLS

view , that it perm its such evident con trad ic­ tion s - that a thing can be w hat it is and at the sam e tim e som ething altogether different. But, in fact, no con tradiction is involved in the statem ent, w hich, on the con trary, ap p ears quite sensible, and even true, to one w ho pre­ sents the idea to him self in the N u er language and within their system o f religious thought. H e does not then take their statem ents ab o u t tw ins any m ore literally than they m ake and understand them them selves. They are not sayin g that a twin has a beak, feathers, and so forth. N o r in their everyday relations with tw ins d o N uer speak o f them as birds or act to w ard s them as though they were birds. They treat them as w hat they are, men and w om en. But in addition to being men and w om en they are o f a tw in-birth, and a tw in-birth is a special revelation o f Spirit; and N uer exp ress this special character o f tw ins in the “ tw ins are b ird s” form ula because tw ins and birds, though for different reason s, are both asso ci­ ated with Spirit and this m akes tw ins, like bird s, “ people o f the a b o v e ” and “ children o f G o d ” , and hence a bird a suitable sym bol in which to exp ress the special relationship in w hich a twin stan ds to G o d . W hen, therefore, N uer say that a twin is a bird they are not sp eak in g o f either as it ap p ears in the flesh. They are sp eakin g o f the anim a o f the tw in, w h at they call his tie, a concept which includes both w hat we call the person ality and the sou l; and they are speaking o f the associatio n birds have with Spirit through their ability to enter the realm to which Spirit is likened in m eta­ ph or and where N uer think it chiefly is, or m ay be. The form ula does not exp ress a dyadic relationship between tw ins and birds but a triadic relationship between tw ins, birds, and G o d . In respect to G od tw ins and birds have a sim ilar character. It is because N uer d o not m ake, or take, the statem ent that tw ins are birds in any ordinary sense that they are fully aw are that in ritual relating to tw ins the action s are a kind o f m im ing. T his is show n in their treatm ent o f the co rp se o f a tw in, for, acco rd in g to w hat they them selves say, w hat is a bird, the tie or an im a , has gone up into the air and w hat is left an d treated - in the case o f ad ults platform burial being a convenient alternative to d is­

145

p osal in trees - as though it m ight be a bird is only the ring , the flesh. It is show n a lso in the convention that should one o f a pair o f tw ins die, the child w ho com es after them takes his place, coun tin g as one o f them in the various cerem onies tw ins have to perform and respect­ ing birds a s rigorously as if he were him self a tw in, w hich he is not. The cerem onies have to be perform ed for the benefit o f the living twin and their structure and purpose are such that there have to be tw o persons to perform them , so a brother or sister acts in the place o f the dead. T h is discussion o f w hat is m eant by the statem ent that a twin is a bird is not so far aw ay from the subject o f totem ism as it m ight seem to be, for the stock explan ation am ong the N u er o f a totem ic relationship is that the an cestor o f a lineage and a m em ber o f a n atu­ ral species were born twins. The relationship o f lineage to species is thereby m ade to derive not only from the closest o f all p ossible rela­ tionships but also from a special act o f divine revelation; and since the link between a lineage and its totem is the tutelary spirit o f the lineage asso ciated with the totem it is ap p rop riate that the relationship should be thought o f as having com e ab o u t by an event which is a direct m ani­ festation o f Spirit. H ow ever, an exam ination o f the N uer d o g m a that tw ins are birds w as m ade not on accoun t o f totem ic relationships com m only being explained in term s o f tw inship but because it w as hoped that it w ould be easier to understand, in the light o f any conclusions reached ab o u t w hat is m eant by the statem ent that a twin is a bird, w hat N u er m ean when they say th at som e totem ic creature, such as the crocodile, is Spirit. C ertainly there is here neither the sort o f m etaphor nor the sort o f ellipsis we found in earlier statem ents. N o r can N u er be un derstood to mean that the creature is identical with Spirit, or even with a spirit, Spirit conceived o f in a particular totem ic refraction. They say quite definitely them selves that it is not; and it is a lso evident, for N uer a s well as fo r us, that a m aterial sym bol of Spirit can not, by its very nature, be that which it sym bolizes. N evertheless, though crocodile and Spirit are quite different and unconnected ideas, when the crocodile is for a certain

146

E. E. EVANS-PRITCHARD

lineage a sym bol o f their special relationship to Ciod, then in the con text o f that relationship sym bol and w hat it sym bolizes are fused. As in the case o f the “ tw ins are b ird s” form u la, the relation is a triadic one, between a lineage and a natural species and G od. There are obvious and significant differences between the creature-Spirit exp ression and the cucum ber-ox and bird-twin expression s. C ucum ber, o x , m an, and bird are all things which can be known by the senses; but where Spirit is experienced other than in thought it is only in its effects or through m aterial repre­ sentations o f it. We can , therefore, easily see how N uer regard it as being in, or behind, the crocodile. The subject and predicate term s o f the statem ent that som ething is Spirit are here no longer held ap art by tw o sets o f visible properties. C onsequently, while N uer say that totem ic spirits and totem s are not the sam e they som etim es not only speak of, but act to w ards, a totem as if the spirit were in it. Thus they give som e m eat o f sacrifice to the lion-spirit to lions, and when they sacrifice to the durra-bird-spirit they ad d ress a lso the birds them selves and tell them that the victim is for them. N evertheless, they m ake it clear in talking about their totem s that w hat respect they show for them is on accoun t o f their re­ presenting the spirits associated with them and not for their own sake. Another difference is that w hereas in the cases o f the cucum ber-ox and tw in-bird exp res­ sions the equivalence rests on an alogies which are quite obviou s even to us once they are pointed ou t - the cucum ber being treated in the ritual o f sacrifice a s an o x is, and tw ins and birds both being “ children o f G o d ” and also m ultiple births - an alogy is lacking in the cre a­ ture-Spirit expression . There is no resem blance between the idea o f Spirit and that o f cro co ­ dile. There is nothing in the nature o f cro co ­ diles which evokes the idea o f Spirit for N uer, and even for those w ho respect crocodiles the idea o f Spirit is evoked by these creatures because the crocodiles are a representation o f Spirit in relation to their lineage and not because there is anything crocodile-like ab o u t Spirit or Spirit-like ab o u t crocodiles. We have passed from observation o f resem blances to thought by m eans o f sym bols in the sort o f

w ay that the crocodile is used a s a sym bol for Spirit. We are here faced with the sam e problem we have been considering earlier, but in w hat, in the absence o f an alogical guidance to help us, is a m ore difficult form . The difficulty is increased by N uer sym bols being taken from an environm ent unfam iliar to us and one which, even w hen we fam iliarize ourselves with it, we experience and evaluate differently. We find it hard to think in term s o f crocodiles, snakes, and fig-trees. But reflection show s us that this problem is com m on to all religious thought, including our ow n; that a religious sym bol has alw ays an intim ate association with w hat it represents, that which brings to the mind with w hat it brings to the m ind. N uer know that w hat they see is a crocodile, but since it represents Spirit to som e o f them it is for those people, when thought o f in that w ay, also w hat it stan ds for. The relationship o f m em bers o f a N uer lineage to Spirit is repre­ sented by a m aterial sym bol by which it can be thought o f concretely, and therefore a s a relationship distinct from the relationships o f other lineages to Spirit. W hat the sym bols stand for is the sam e thing. It is they, and not w hat they stand fo r, which differentiate the relationships. There results, when w hat acts as a sym bol is regarded in this w ay, a fusion between Spirit, as so represented, and its m ate­ rial representation. I w ould say that then N uer regard Spirit as being in som e w ay in, or behind, the creature in which in a sense it is beholden. The problem is even m ore difficult and com plex than I have stated it, because we might say that w hat are fused are not so much the idea o f Spirit and its m aterial represen ta­ tion as the idea o f Spirit and the idea o f its m aterial representation. It is rather the idea o f crocodile than the saurian creatures them selves which stan ds for Spirit to a lineage. If a N uer cannot see Spirit he likewise in som e cases seldom , if ever, sees his totem ; so that it is no longer a question o f a m aterial object sym bol­ izing an idea but o f one idea sym bolizing another. 1 d ou b t whether those w h o respect m onorchid bulls or w aterbuck often see a m em ber o f the class or species, and children in these and other cases m ust often be told ab o u t

THE PROBLEM OF SYMBOLS

their totem ic attachm ents before they have seen their totem s. There m ust also be N uer w ho respect dom palm s w ho live in parts o f N uerlan d to the east o f the N ile where this tree d oes not g row .4 Indeed, I feel confident that one totem , the lou serpent, a kind o f Loch N ess m onster, d oes not exist, an d if this is so , a totem can be purely im aginary. As this point has som e theoretical im portance for a study o f totem ism 1 draw attention to a further significant fact. N uer d o not speak o f the spirit o f crocod iles, lions, tam arindtrees, and so on, but alw ays o f the spirit o f crocodile, lion, and tam arind-tree, and they w ould never say that crocodiles, lions, and tam arind-trees were som eb od y’s spirit but alw ays that crocodile, lion, and tam arind-tree, w as his spirit. The difference in m eaning between the plural and singular usage is not, perh aps, very ob vious in English but it is both clear an d vital in N uer. It is the difference between crocodiles thought o f as they are seen in rivers and crocodiles thought o f as crocodile or as the crocodile, as a type o f creature, cro c­ odile a s a conception. The point I am m aking is exem plified by the story already recorded (p. 65) o f a m an w ho gave up respecting lions because they killed his cattle. He still regarded lion-spirit, Spirit in the representation o f lion, as a spirit connected with his fam ily. But if a totem ic relationship m ay be an ideal one, and has alw ays som ething o f the ideal in it, I w ould still say that N uer regard Spirit as being in som e w ay in, or behind, totem ic creatures when they think o f them a s representations o f Spirit. [•••I

. . . T he m ore Spirit is thought to be bound to visible form s the less it is thought o f as Spirit and the m ore it is thought o f in term s o f w hat it is bound to. In other w ords, there are g ra d a ­ tions o f the conception o f Spirit from pure unattached Spirit to Spirit associated with h um an, anim al, and lifeless objects and m ore and m ore closely bound to w hat it is a sso c i­ ated with the farther dow n the scale one goes. This scale o f Spirit, as I have explained earlier, is related to segm entation o f the social order and is represented by N uer by levels o f space a s well as by levels and degrees o f im m anence. So when N uer say o f som ething that it is Spirit

147

we have to consider not only w h at “ is” m eans but also w hat “ Spirit” m eans. N evertheless, though the sense o f “ kw oth ” varies with the context, the word refers alw ays to som ething o f the sam e essence; and w hat is being said, directly or indirectly, in the statem ents is alw ays the sam e, that som ething is that essence. We can m ake som e contribution tow ards a solution o f the problem in the light o f this discussion. When N uer say o f som ething ‘V kw oth ” , “ it is Spirit” , or give it a nam e o f which it can be further said “ th at is Sp irit” , the “ is” does not in all instances have the sam e con notation. It m ay be an elliptical statem ent, signifying that the thing referred to is a m ani­ festation o f Spirit in the sense o f G od revealing him self in instrum ents or effects. O r it m ay be a sym bolical statem ent, signifying that w hat in itself is not Spirit but represents Spirit to certain persons is for these persons Spirit in such con texts as direct attention to the sym ­ bolic character o f an object to the exclusion o f w hatever other qualities it m ay p ossess. O r it m ay be a statem ent signifying som ething closer to identity o f the thing spoken o f with w hat it is said to be, Spirit. The statem ents never, how ever, signify com plete identity o f anything with Spirit, because N uer think o f Spirit as som ething m ore than any o f its m odes, signs, effects, representations, and so forth, and also as som ething o f a different nature from the created things which they are. They are not able to define w hat it is, but when it acts within the phenom enal w orld they say it has com e from above, where it is conceived to be and whence it is thought to descend. C o n se­ quently Spirit in any form can be detached in the mind from the things said to be it, even if they cannot alw ays be so easily detached from the idea o f Spirit. I can take the analysis no farther; but if it is inconclusive it at least sh ow s, if it is correct, how w ide o f the m ark have been an th ropologi­ cal attem pts to explain the kind o f statem ents we have been considering. A nthropological explan ation s display tw o m ain errors. The first, best exem plified in the w ritings o f LevyBruhl, is that when a people say that som e­ thing is som ething else which is different they are contravening the Law o f C ontradiction

148

E. E. EVANS-PRITCHARD

and substitu tin g for it a law o f their ow n prelogical w ay o f thinking, that o f m ystical p a r­ ticip atio n .s I hope at least to have show n thar N uer d o not a ssert identity between the tw o things. They m ay say that one is the other and in certain situ ations act tow ard s it a s though it were thar other, or som ething like it, but they are aw are, no d ou bt with varying degrees o f aw aren ess, and readily say, though with varying degrees o f clarity and em ph asis, rhat the tw o things are different. M oreover, it will have been noted that in the seem ingly e q u i­ vocal statem ents we have considered, with perh aps one exception , the term s can n ot be reversed. T he exception is rhe statem ent that twins are birds, because it can also be said that birds are twins. T h at a hatch o f birds are tw ins is a statem ent, ro which we also can give assen t, which does not derive logically from the statem ent that tw ins are birds bur from a perception independent o f that p ro p o sitio n ; so it does not concern our problem . R ain m ay be said to be Ciod but Ciod cannot be said ro be rain; a cucum ber m ay be called an o x bur an o x can n ot be called a cucum ber; and the cro c­ odile m ay be said to be Spirit but Spirit can n ot be said to be the crocodile. Consequently these are not statem ents o f identity. They are sta te ­ m ents not that som eth ing is other than it is but that in a certain sense and in p articular co n ­ texts som eth in g has som e extra quality which does not belong to it in its ow n nature; and this quality is not contrary to, or incom patible with, its nature but som ething added to it which does not alter w hat it w as but m ak es it som ething m ore, in respect to this quality, than it w as. C onsequently, no con trad iction , it seem s to m e, is involved in the statem ents. W hether the predicate refers to a conception or ro a visible object the addition m ak es the subject equivalent to it in respect to the quality which both now have in com m on in such con texts a s focus the attention on that quality alone. T he things referred to are not the sam e as each other but they are the sam e in th at one respect, an d the equivalence, denoted by the co p u la, is not one o f substance but o f quality. C on sequen tly we cannot speak here, a s LevyBruhl d oes, o f m ystical p articipation , or a t any rate not in his sense o f the w ords, because the tw o things are not thought to be linked by a

m ystical bond but sim ply by a sym bolic nexus. T herefore, w hat is done to birds is not thought to affect tw ins, an d if a totem is harm ed the spirit o f that totem m ay be offended but it is not harm ed by the harm done to the totem ic creature. T h at the relation between the thing said to be som ething else an d that som ething else it is said to be is an ideal one is indeed obvious, but an th ropological e xp lan ation s o f m odes o f prim itive thought a s wide ap art as those o f T ylor, M a x M uller, and Levy-Bruhl, are based on the assum ption that though for us the rela­ tion is an ideal one prim itive peoples m istake it for a real one; and those an th ropologists w ho spon sor psych ological explan ation s often m ake the sam e assum ption . This is the second error. If my interpretation is correct, N uer know very well when they say that a crocodile is Spirit that it is only Spirit in the sense that Spirit is represented to som e people by that sym bol just a s they know very well that a cucum ber is only an o x in the sense that they treat it as one in sacrifice. T h at they d o not m istake idea relation s for real ones is show n by m any exam p les in this b o ok : the identifica­ tion o f a sacrificial spear with that o f the ancestor . . the identification o f m an with o x in sacrifice . . . , the identification o f a m an’s herd with that o f the an cestor o f his clan . . ., the identification o f sickness and sin in a sa c ­ rificial c o n te x t. . ., and the identification o f the left hand with death and evil. . . . It is show n also in the sym bolism o f m any o f their rites, where their purpose is expressed in m im icry .. . . 1 think rhat one reason why it w as not readily perceived that statem ents that som e* thing is som ething else should not be taken as m atter-of-fact statem ents is that it w as not re­ cognized that they are m ade in relation to a third term not m entioned in them but under­ stood. They are statem en ts, as far as the N uer are concerned, not that A is B , but that A and B have som ething in com m on in relation to C. T h is is evident when we give som e thought to the m atter. A cucum ber is equivalent to an ox in respect to Ciod w ho accepts it in the place o f an o x . A crocodile is equivalent to Spirit only when conceived o f as a representation o f Ciod to a lineage. C on sequen tly, though N uer

THE PROBLEM OF SYMBOLS

d o not m istake ideal relation s for real ones, an ideal equivalence is none the less true for them , because within their system o f religious thought things are not ju st w hat they a p p ear to be but a s they are conceived o f in relation to Ciod. T h is im plies experience on an im aginative level o f thought where the m ind m oves in figures, sym bols, m etaph ors, an alo g ies, and m any an elaboration o f poetic fancy and lan­ g u age; and another reason why there has been m isun derstan din g is that the poetic sense o f prim itive peoples has not been sufficiently allow ed for, so that it has not been appreciated that w hat they say is often to be understood in that sense and not in any ordin ary sense. T h is is certainly the case with the N uer, as we see in this chapter and in m any places else­ where in this b o ok , for exam p le, in their hym ns. In all their poem s an d so n g s a lso they play on w ords and im ages to such an extent th at no European can tran slate them w ithout com m entary from N uer, and even N uer them ­ selves cannot alw ays say w h at m eaning they had for their authors. It is the sam e with their cattle- and dance-nam es, which are chosen both for euphony and to exp ress an alogies. H o w N uer delight in playing with w ords is a lso seen in the fun they have in m aking up tongue-tw isters, sentences which are difficult to pronounce w ithout a m istake, and slips o f the tongue, usually slips in the presence of m others-in-law , which turn quite ordinary

rem arks into obscenities. L ack in g plastic and visual arts, the im agination o f this sensitive people finds its sole expression in ideas, im ages, and w ords. In this and the last chapter I have attem pted to lay bare som e features o f the N uer concep­ tion o f Spirit. We are not ask in g w hat Spirit is but w hat is the N u er conception o f kwoth , which we tran slate “ Spirit” . Since it is a con ­ ception that we are inquiring into, our inquiry is an exploration o f ideas. In the course o f it we have found that w hilst N u er conceive o f Spirit as creator and father in the heavens they a lso think o f it in m any different representa­ tions (w hat I have called refractions o f Spirit) in relation to social g ro u p s, categories, and person s. The conception o f Spirit h as, we found, a social dim ension (we can also say, since the statem ent can be reversed, that the social structure has a spiritual dim ension). We found a lso that Spirit, in the N u er conception o f it, is experienced in signs, m edia, and sym bols through which it is m anifested to the senses. Fundam entally, how ever, this is not a relation o f Spirit to things but a relation of Spirit to persons through things, so that, here a gain , we are ultim ately concerned with the relation o f G od and m an , and we have to consider not only w hat is the Ciod-to-m an side o f the relationship, to which attention has so far m ostly been given, but a lso the man-toCiod side o f it, to which I now turn.

NOTES 1

2

3

T he spear win m ay be said to stan d for the spirit win , and the pied crow m ay be said to stand for the spirit buk which is the m ost terrestrially conceived o f am on g the greater spirits. I have given a m ore detailed accoun t in “ C u stom s and Beliefs R elatin g to T w ins am o n g the N ilotic N u e r” , Uganda Jou rn al , 1936. T h at the nam es, at least all those I have heard, are taken from birds low est in the scale o f N u er reckoning requires com m ent, especially in view o f the argum en t I later develop. It m ay be due to the N uer habit

149

4

o f sp eakin g o f their relation to Ciod - the birth o f tw ins constitutes such a context by co m p arin g them selves w ith lowly things. O n the other hand, it m ay be sim ply in keeping with the logic o f the analogy. T w in s belong to the class o f the above but are below ; just as guineafow l and francolin belong to the class o f birds, which as a class is in the category o f the abo ve, but are alm ost earthbound. D r. Lienhardt tells me that a num ber o f lin­ eages in w estern D inkalan d respect crea­ tures which no longer exist there. A Dinka w ho travelled with him to other parts o f the Southern Sudan w as aston ish ed when he first saw his totem , an elephant. N an a

150

E. E. EVANS-PRITCHARD

K obin a N k etsia IV o f Sekondi perm its me to say that the first time he saw his totem , the b u ffalo, w as last year in a film at O x ford . Professor I. Schapera tells me that the ruling fam ily o f the senior tribe in the Bechuanaland P rotectorate, the K w ena, have been living for a hundred years in a region where their totem , the crocodile, is unknow n (see also w hat he says in The Tsw ana (Interna­ tional A frican Institute), 1 9 5 3 , p. 3 5 , and H ugh A shton, The Basuto , 1952, p. 14). O ther exam ples could be cited. It m ay help us to appreciate the point better if we co n ­ sider the nearest parallels in our own

5

country. W hen we think o f the lion as our national sym bol we do not think o f the m angy creatures o f the A frican bush or in zoos. N o r does it incom m ode us that there are no unicorns and never have been any. I refer to his earlier w ritings, in particular

Les Fonctions mentales dans les societes inferieures (1 9 1 0 ) and L a mentalite prim i­ tive (1922). T he second part o f his last b o ok , L'F.xperience mystique et les symboles chez les pnm itifs (1 938), which took account o f m odern research, is a brilliant discourse on the problem s we have been discussing.

12

On Key Symbols Sherry B. Ortner

Sherry Ortner is an Am erican anthro­ pologist, w ho trained at Chicago and taug ht at Michigan and Columbia before moving to UCLA. She is renowned for the clarity of her synthetic appraisals of theory, of w hich this is an early one (see also 1984f 1995), her contributions to gender theory and fem inist anthro­ pology (1974, 1996a; O rtner and W hitehead 1981), and her ethnogra­ phies of Sherpa Buddhism (1978, 1989, 1999b) as w ell as class in the USA. O rtner was a student of Geertz w ho, despite her interest in seeking broader and more "m uscular" forms of social expla­ nation than he provides (Dirks, Eley, and Ortner, eds., 1994), has remained one of the most articulate spokespersons for the significance of culture in social theory. If Langer and W hite established the symbolic ground of culture, there was a trend in Am erican anthropology to go further and seek the symbols that could

sum up the particular focus, ethos, char­ acter, or w orldview of a given cultural system. In this essay O rtner captures a particularly exciting m om ent in the discipline as symbolic anthropology emerged into prominence, and usefully articulates the various w ays anthropolo­ gists have conceived, derived, or applied dom inant symbols and ho w the symbols operate. Ortner suggests both the ways systems of meaning are organized and the ways different kinds of symbols work to condense, produce, or invite m eaning. W hether or not such symbols work effectively in the ethnographic portraits constructed by anthropolo­ gists, they often do appear to serve such ends for the groups in w hich they are found. W hat Ortner offers is a series of W eberian "ideal types," not species of symbols. This essay represents the state of the art in the early 1970s; Ortner has since moved w ell beyond it in an attem pt to fully historicize cultural accounts. For

From Sherry B. Ortner, “ On Key Symbols,” American Anthropologist 75 (1973): 1338-46. Reprinted by permission of the American Anthropological Association. Notes deleted.

152

SHERRY B. ORTNER

a later essay th at links m eaning w ith power, see O rtner (1999a). Victor Turner (1967) provides several great essays on religious symbols from a Durkheim ian perspective. Douglas (1975) offers a classic analysis of a central symbol, th e Lele pangolin, w h ile De Boeck (1994) gives a m ore recent intricate symbolic interpretation from the same Central African culture area as Turner and Douglas. A general account of symbols is to be found in Firth (1973), w hile M unn (1973; cf. 1986, 1990) is an excellent discussion of the transform ative properties of ritual symbols. Schneider moves aw ay from individual symbols to account for the

This paper reviews the use of the notion o f “ key symbol ” in anthropological analysis. It analyzes phenomena which have been or might be accorded the status o f key symbol in cultural analy­ ses , categorizing them according to their primary modes o f operating on thought and action.

It is by no m eans a novel idea that each culture has certain key elem ents which, in an ill-defined w ay, are crucial to its distinctive organ ization . Since the publication o f Benedict's Patterns o f Culture in 1 9 3 4 , the notion o f such key ele­ m ents has persisted in A m erican an th ropology under a variety o f rubrics: “ them es” (e.g., O pler 1945; C ohen 1 9 4 8 ), “ focal valu es” (A lbert 195 6 ), “ dom in an t valu es” (D uB ois 1955), “ integrative co n cep ts” (D uB ois 1936), “ d o m i­ nant o rien tatio n s” (F. Kluckhohn 1950), an d so forth. We can a lso find this idea sneaking namelessly into British social an th rop ological w riting; the best exam ple o f this is L ien h ard t’s (19 6 1 ) d iscu ssion o f cattle in D inka culture (and I say culture rather than society advisedly). Even Evans-Pritchard has said, “ a s every e x p e ­ rienced field-w orker kn ow s, the m ost difficult task in social an th rop ological field w ork is to determ ine the m eanings o f a few key w ord s,

symbolic constitution of persons and the conceptions underlying such cultural domains as kinship and religion (1977 [1969], 1980 [1968]). The notion of cul­ tural scenarios is w ell developed in Schieffelin (1976). Both Pepper (1942) and W hite (1973) offer suggestive accounts of the pow er of underlying metaphors. In addition to the essay by Tam biah (chapter 25, below), the best anthropological w ork on m etaphor is by Fernandez (1986b; ed. 1991) and Sapir and Crocker, eds. (1977). For a more directly linguistic account illustrat­ ing the prevalence of m etaphor in ordi­ nary language, see Lakoff and Johnson (1980).

upon an understanding o f which the success of the w hole investigation d epen d s” (1 9 6 2 :8 0 ). Recently, a s the focus in the study o f m eaning system s has shifted to the sym bolic units which form ulate m eaning, the interest in these key elem ents o f cultures has becom e specified as the interest in key sym bols. Schneider (1968) calls them “ core sy m b o ls” in his study o f A m erican kinship; T urner (1 9 6 7 ) calls them “ dom inant sy m b ols” in his study o f N dem bu ritual; I called them “ key sy m b o ls” in my study o f Sherpa social relations (O rtner 1970). The prim ary question o f course is w hat do we m ean by “ key” ? But I will postpon e con ­ sidering this problem until I have discussed the various usages o f the notion o f key sym bols in the literature o f sym bolic analysis. T w o m eth odological ap p roach es to estab ­ lishing certain sym bols as “ co re ” or “ key” to a cultural system have been em ployed. The first ap p roach , less com m only used, involves analyzing the system (or dom ain s thereof) for its underlying elem ents - cognitive distin c­ tions, value orien tation s, etc. - then lookin g a b o u t in the culture for som e figure or im age which seem s to form ulate, in relatively pure form , the underlying orien tation s exposed in the an alysis. T he best exam ple o f this app roach in the current literature is D avid Schneider’ s (1968) an alysis o f A m erican kinship; Schneider

ON KEY SYMBOLS

first analyzes the kinship system for its basic com pon en ts - nature and law - and then decides that conjugal sexual intercourse is the form w hich, given its m eaning in the culture, expresses this opposition m ost succinctly and m eaningfully. Schneider exp resses his debt to Ruth Benedict, and this debt turns out to be quite specific, since the other m ajor w ork which em bodies this m ethod is Benedict’s The Chrysanthemum and the Sw ord (19 6 7 ). The sw ord and the chrysanthem um were chosen by Benedict from the repertoire o f Jap an ese sym bols as m ost succinctly, or perhaps m ost poetically, representing the tension in the Ja p a n e se value system which she p ostulated. She did not arrive at this tension through an an alysis o f the m eanings o f chrysanthem um s and sw ord s in the culture; she first established the tension in Ja p a n e se culture through an aly­ sis o f various sym bolic system s, then chose these tw o item s from the repertoire o f Jap an ese sym bols to sum up the op p osition . In the second, m ore com m only em ployed ap p ro ach , the investigator observes som ething which seem s to be an object o f cultural inter­ est, and analyzes it for its m eanings. The observation that som e sym bol is a focus o f cultural interest need not be very m ysterious o r intuitive. I offer here five reason ably reliable in dicators o f cultural interest, and there are p rob ab ly m ore. M o st key sym bols, I venture to suggest, will be signaled by m ore than one o f these indicators: (1) (2)

(3)

(4)

The natives tell us that X is culturally im portant. The natives seem positively or negatively aroused about X, rather than indifferent. X com es up in m any different contexts. These con texts m ay be behavioral or sys­ tem ic: X com es up in m any different kinds o f action situation or conversation, or X com es up in m any different sym bolic d om ain s (m yth, ritual, art, form al rheto­ ric, etc.). There is greater cultural elabo ration su r­ rounding X , e.g., elaboration o f v o cab u ­ lary, or elaboration o f details o f X ’s nature, com pared with sim ilar phenom ­ ena in the culture.

(5)

153

There are greater cultural restrictions sur­ rounding X , either in sheer num ber of rules, or severity o f sanctions regarding its m isuse.

A s I said , there m ay be m ore in dicators even than these o f the key statu s o f a sym bol in a culture, but any o f these should be enough to point even the m ost insensitive fieldw orker in the right direction. I should a lso add that 1 am not assu m in g that there is only one key sym bol to every culture; cultures are o f course a produ ct o f the interplay o f m any basic orienta­ tion s, som e quite conflicting. But all o f them will be expressed som ew here in the public system , because the public sym bol system is ultim ately the only source from which the natives them selves discover, rediscover, and tran sform their ow n culture, generation after generation. It rem ains for us now to sort ou t the bew il­ dering array o f phenom ena to w hich various in vestigators have been led to assign im plicitly o r explicitly the status o f key cultural sym bol. A nything by definition can be a sym bol, i.e., a vehicle for cultural m eaning, and it seem s from a survey o f the literature that alm ost anything can be key. O m itting the sym bols established by the first ap proach cited ab o ve, which have a different epistem ological statu s, we can cite from the an th ropological literature such things a s cattle am ong the D inka and N u er, the N aven ritual o f the Iatm ul, the A ustralian churinga, the slam etan o f the Ja v an ese, the potlatch of the northw est co ast, the forked stick o f N dem bu rituals, and from my ow n research, the wheelim age in T ibet and food am o n g the Sherpas. We could a lso add such intuitive exam ples as the cro ss o f C hristianity, the A m erican flag, the m otorcycle for the H ell’s A ngels, “ w o rk ” in the Protestant ethic, and so on. The list is a jum ble - things and a b stra c­ tions, nouns and verbs, single item s an d w hole events. I should like to p rop o se a w ay o f su b­ dividing and ordering the set, in term s o f the w ays in which the sym bols operate in relation to cultural thought and action. The first m ajor breakdow n am o n g the vario u s types o f sym bols is alo n g a continuum w hose tw o ends I call “ su m m arizin g’’ vs. “ e la b o ratin g .” I stress that it is a continuum ,

154

SHERRY B. ORTNER

but I w ork with the ideal types at the tw o ends. Sum m arizing sym bols, first, are those sym bols which are seen as sum m ing up, expressing, representing for the particip an ts in an em otionally pow erful and relatively undif­ ferentiated w ay, w hat the system m ean s to them. T his category is essentially the category o f sacred sym bols in the broadest sense, and includes all those item s which are objects o f reverence an d/or catalysts o f em otion - the flag, the cross, the churinga, the forked stick, the m otorcycle, etc. The Am erican flag, for exam ple, for certain A m ericans, stan d s for som ething called “ the A m erican w ay ,” a co n ­ glom erate o f ideas and feelings including (the­ oretically) dem ocracy, free enterprise, hard w ork, com petition, p rogress, national superi­ ority, freedom , etc. And it stan ds for them all at once. It does not encourage reflection on the logical relations am on g these ideas, nor on the logical consequences o f them as they are played out in social actuality, over tim e and history. On the con trary, the flag encourages a sort o f all-or-nothing allegiance to the w hole pack age, best sum m ed up on a billboard I saw recently: “ O ur flag, love it or leave.” And this is the point about sum m arizing sym bols in general - they operate to com poun d and synthesize a com plex system o f ideas, to “ sum m arize” them under a unitary form which, in an oldfashioned w ay, “ stan ds fo r” the system as a whole. Elaborating sym bols, on the other hand, w ork in the o p p o site direction, providing vehi­ cles for sortin g out com plex and undiffer­ entiated feelings and ideas, m aking them com prehensible to oneself, com m unicable to others, and tran slatable into orderly action. Elaborating sym bols are accorded central statu s in the culture on the basis o f their c a p a c ­ ity to order experience; they are essentially analytic. R arely are these sym bols sacred in the conventional sense o f being ob jects o f respect or foci o f em otion; their key statu s is indicated prim arily by their recurrence in cultural beh av­ ior or cultural sym bolic system s. Sym bols can be seen as having elaboratin g pow er in tw o m odes. They may have prim arily conceptual elabo ratin g pow er, that is, they are valued as a source o f categories for con ceptu­

alizing the order o f the w orld. O r they m ay have prim arily action elaboratin g pow er; that is, they are valued a s im plying m echanism s for successful social action. These tw o m odes reflect w hat I see as the tw o basic and o f course interrelated functions o f culture in general: to provide for its m em bers “ orien tation s,” i.e., cognitive and affective categories; and “ strate­ g ie s,” i.e., p rogram s for orderly social action in relation to culturally defined goals. Sym bols with great conceptual elaboratin g pow er are w hat Stephen Pepper (1942) has called “ roo t m etap h o rs,” and indeed in this realm the basic m echanism is the m etaphor. It is felt in the culture that m any aspects o f exp e­ rience can be likened to, and illum inated by the co m parison with, the sym bol itself. In Pepper’s term s, the sym bol provides a set o f categories for conceptualizing other aspects o f experi­ ence, or, if this point is stared too uni-directionally for som e tastes, we m ay say that the root m etaphor form ulates the unity o f cultural o ri­ entation underlying m any aspects o f exp eri­ ence, by virtue o f the fact that those m any aspects o f experience can be likened to it. O ne o f the best exam ples o f a cultural root m etaphor in the an th ropological literature is found in G odfrey Lienh ardt’s discussion o f the role o f cattle in D inka thought. C o w s provide for rhe D inka an alm ost endless set o f cate g o ­ ries for conceptualizing and responding to the subtleties o f experience. For exam ple: “ The D in k as’ very perception o f co lo u r, light, and shade in the w orld aroun d them i s . . . inextri­ cably connected with their recognition o f colour-configurations in their cattle. If rheir cattle-colour vocabu lary were taken aw ay, they w ould have scarcely any way o f describ­ ing visual experience in term s o f colour, light and d ark n ess” (1 9 6 1 :1 3 ). M ore im portant for Lienhardt’s thesis is the D inka con ceptualiza­ tion o f the structure o f their own society on analogy with the physical structure o f the bull. “ ‘The people are put together, a s a bull is put together,’ said a D inka chief on one o c casio n ” (ibid.: 2 3 ), and indeed the form ally prescribed division o f the m eat o f a sacrificed bull is a m ost graphic representation o f the statu ses, functions, and interrelationships o f the m ajor social categories o f D inka society, as the Dinka them selves represent the situation.

ON KEY SYMBOLS

In fact, as M ary D o u g las points out, the living organism in one form or another func­ tions as a root m etaphor in m any cultures, as a source o f categories for conceptualizing social phenom ena (1 9 6 6 ). In m echanized society, on the other hand, one root m etaphor for the social process is the m achine, an d in recent times the com puter represents a crucial m odification upon this ro o t m etaphor. But the social is not the only aspect o f experience which root-m etaphor type sym bols are used to illum inate; for exam ple, much o f greater IndoT ibetan cosm ology - the form s and processes o f life, space, and time - is developed on an alogy with the quite sim ple im age o f the wheel (O rtner 1966). A roo t m etaphor, then, is one type o f key sym bol in the elaboratin g m ode, i.e., a sym bol which operates to sort out experience, to place it in cultural categories, and to help us think ab o u t how it all h an gs together. They are sym bols which are “ go o d to th ink,” not exactly in the Levi-Straussian sense, but in that one can conceptualize the interrelation­ ships am o n g phenom ena by an alogy to the interrelations am on g the parts o f the roo t m etaphor. The other m ajor type o f elaboratin g sym bol is valued prim arily because it im plies clear-cut m odes o f action ap p rop riate to correct and successful living in the culture. Every culture, o f course, em bodies som e vision o f success, or the go o d life, but the cultural variation occurs in how success is defined, an d , given th at, w h at are considered the best w ays o f achieving it. “ Key scen ario s,” as I call the type o f key sym bol in this category, are culturally valued in that they form ulate the culture’s basic m eans-ends relationships in actable form s. An exam ple o f a key scenario from A m erican culture w ould be the H o ratio Alger myth. The scenario runs: p oo r boy o f low statu s, but with total faith in the Am erican system , w orks very hard and ultim ately becom es rich and pow erful. The myth form u­ lates both the A m erican conception o f success - w ealth and pow er - and suggests that there is a sim ple (but not easy) w ay o f achieving them - singlem inded hard w ork. T his scenario m ay be contrasted with ones from other

155

cultures which present other actions a s the m ost effective m eans o f achieving w ealth and pow er, or which form ulate w ealth and pow er as ap p rop riate go als only for certain segm ents o f the society, or, o f course, those which do not define cultural success in term s o f w ealth and pow er at all. In any case, the point is that every culture has a num ber o f such key scenarios which both form ulate ap p ro p riate g o a ls and suggest effective action for achieving them; w hich form ulate, in other w ords, key cultural strategies. T h is category o f key sym bols m ay also include rituals; Singer seem s to be m aking the point o f rituals as scenarios when he w rites o f “ cultural perform an ces” (1 9 5 8 ), in which both valued end states and effective m eans for achieving them are dram atized for all to see. T hus this category w ould include naven, the slam etan, the potlatch, and others. The cate­ gory could also include individual elem ents o f rituals - objects, roles, action sequences in sofar as they refer to or epitom ize the ritual as a w hole, which is why one can have ac­ tions, objects, and whole events in the sam e category. Further, scen arios as key sym bols may include not only form al, usually nam ed events, but also all those cultural sequences o f action which we can observe enacted and reenacted accordin g to unarticulated form ulae in the norm al course o f daily life. An exam ple o f such a scenario from Sherpa culture w ould be the hospitality scenario, in which any individ­ ual in the role o f host feeds a guest and thereby renders him voluntarily cooperative vis-a-vis oneself. T he scenario form u lates both the ideally valued (though infrequently attained) m ode o f social relations in the culture - vol­ untary cooperation - an d, given certain cul­ tural assum p tion s ab o u t the effects o f fo od on people, the m ost effective w ay o f establishing those kinds o f relations. O nce again then, the scenario is culturally valued - indicated in this case by the fact that it is played and replayed in the m ost diverse sorts o f social contexts because it suggests a clear-cut strategy for arriving at culturally defined success. I have been discussing the category o f key sym bols which I called “ e lab o ratin g ” sym bols, sym bols valued for their con tribution to the

156

SHERRY B. ORTNER

sortin g out o f experience. T his class includes both ro o t m etaph ors which provide categories for the ordering o f conceptual experience, and key scen arios which provide strategies for organ izin g action experience. While for p u r­ poses o f this discu ssion I have been led by the data to sep arate thought from action , 1 m ust hasten to put the pieces back together again . For my view is that ultim ately both kinds of sym bols have both types o f referents. R o o t m etaph ors, by establish in g a certain view o f the w orld, im plicitly suggest certain valid and effective w ays o f actin g upon it; key scen a­ rios, by prescribing certain culturally effective co u rses o f action , em body and rest upon certain assu m p tion s ab o u t the nature o f reality. Even sum m arizing sym bols, while prim arily functioning to co m p o un d rather than sort out experience, are seen as both form ulating basic orien tation s and im plying, though much less system atically than scen arios, certain m odes o f action. O ne question which m ight be raised at this point is how we are to understand the logical relation sh ips am o n g the types o f key sy m ­ bols I have distinguished. As the schem e stands now , it has the follow ing unbalanced structure: sum m arizing

vs.

elaborating

root m etaphor

key scenario

I w ould argue that this asym m etry follow s from the content o f the types: the m eaningcontent o f sum m arizing or sacred sym bols is by definition clustered, condensed, relatively undifferentiated, “ th ick,” while the m eaningcontent o f elabo ratin g sym bols is by definition relatively clear, orderly, differentiated, articu ­ late. T h u s it is possible to m ake distin ction s am o n g the different ordering functions o f elabo ratin g sym bols, while the denseness o f m eaning o f sum m arizing sym bols renders them relatively resistan t to subdivision and orderin g by types. N on eth eless, in the interest o f sys­ tem atic an aly sis, we m ay raise the question of whether such subdivision s are p ossible, an d in particu lar whether the thought/action distin c­ tion which subdivides elaboratin g sym bols (into root m etaph ors and key scenarios)

a lso crosscuts and subdivides sum m arizing sym bols. The im portan t m ode o f operation o f su m ­ m arizing sym bols, it will be recalled, is its focusing pow er, its draw ing-together, intensi­ fying, catalyzin g im pact upon the respondent. T h u s we m ust ask w hether som e sum m arizing sym bols prim arily op erate to catalyze thought or in any case internal states o f the acto r, while others prim arily operate to catalyze overt action on the part o f the acto r. N ow it does seem possible, for exam p le, to see the cross or som e other religious sym bol as prim arily focusing and intensifying inner attitude, with no p articular im plied public action , while the flag or som e other political sym bol is prim arily geared to focusin g and catalyzing overt action in the public w orld. Yet, intuitively at least, this distinction seem s relatively w eak and unconvincing com pared to the easily form u­ lated and grasped distinction between the tw o types o f elabo ratin g sym bols: static form a! im ages serving m etaph or functions for thought {root m etaph o rs), and dram atic, phased action sequences serving scenario func­ tions for action (key scenarios). O f co urse, as I said, root m etaph ors m ay imply particular m odes of, or at least a restricted set o f possible m odes of, action ; and key scen arios presup­ pose certain orderly assu m p tion s o f thought. Hut the distinction - the form er geared prim ar­ ily to thought, the latter to action - rem ains sharp. Sum m arizing sym bols, on the other hand, speak prim arily to attitu des, to a crystalliza­ tion o f com m itm ent. A nd, in the m ode o f co m ­ m itm ent, the th ought/action distinction is not particularly relevant. T here m ay certainly be consequences for thought and action a s a result o f a crystallized com m itm ent, but co m ­ m itm ent itself is neither thought nor action. The point perh aps illum inates the generally sacred statu s o f sum m arizing sym bols, for they are speakin g to a m ore diffuse m ode o f orien­ tation in the acto r, a bro ader con text o f atti­ tude within which p articu lar m odes o f thinking and acting are form ulated. T h is is not to say that nothing analytic m ay be said ab o u t sum m arizing sym bols beyond the fact that they catalyze feeling; there are a num ber o f p ossible w ays o f subdividing the

ON KEY SYMBOLS

ca ta lo g o f sacred sym bols in the w orld, som e n o d o u b t m ore useful or illum inated than others. My point is m erely that rhe p articular factor which subdivides elaboratin g sym bols - the thought/action distinction - does not serve very pow erfully to subdivide the category o f sum m arizing sym bols, since rhe sum m ariz­ ing sym bol is sp eakin g ro a different level o f respon se, the level o f attitude and com m itm ent. We are now in a position to return ro the question o f “ key” or central statu s. Why are w e justified in calling a p articular sym bol “ key” ? The in dicators provided earlier for at least provision ally regardin g certain sym bols a s key to a p articular culture were all based on the assum ption rhat keyness has public (though not necessarily con scious) m an ifesta­ tion in the culture itself, availab le to the observer in the field, or at least availab le when one reflects upon on e’s ob servation s. But the fact o f public cultural concern o r focus o f interest is not why a sym bol is key; it is only a signal that the sym bol is playing som e key role in relation to other elem ents o f the cul­ tural system o f thought. The issue o f keyness, in sh o rt, has to d o with the internal o rg an iza­ tion o f the system o f cultural m eaning, as that system functions for acto rs leading their lives in the culture. Broadly speaking, the tw o types o f key sym bols distinguished ab o v e, defined in term s o f how they act upon or are m an ipulated by cultural acto rs, also indicate the tw o broad m odes o f “ keyness” from a system ic poin t of view , defined in term s o f the role such sym bols are playing in the system ; that is, a given sum m arizing sym bol is “ key” to the system in sofar as the m eanings which it form ulates are logi­ cally o r affectively prior to other m eanings o f the system . By “ logically o r affectively p rio r” 1 m ean sim ply that m any other cultural ideas an d attitu d es p resuppose, and m ake sense only in the co n text o f, those m eanings form ulated by the sym bol. The key role o f an elaboratin g sym bol, by co n trast, derives not so m uch from the statu s o f its p articular substantive m ean ­ ings, but from its form al or organ ization al role in relation to the system ; th at is, we say such a sym bol is “ key” to the system in sofar as it extensively and system atically form ulates

157

relationsh ips - parallels, isom orp h ism s, com ­ plem entarities, and so forth - between a wride range o f diverse cultural elem ents. T h is co n trast between the tw o m odes o f “ keyn ess” m ay be sum m ed up in various w ays, all o f which oversim plify to som e extent, but which nonetheless give perspective on the point. (1) “ C ontent versus fo rm ” : The keyness o f a sum m arizing sym bol derives from its particular substantive m eanings (content) and their logical priority in relation to other m ean­ ings o f the system . The keyness o f an e lab o rat­ ing sym bol derives from its form al properties, and their culturally p ostu lated pow er to for­ m ulate widely applicab le m odes o f organizing cultural phenom ena. (2) “ Q uality versus q u an ­ tity” : T he keyness o f a sum m arizing sym bol derives from the relative fu n d a m e n ta lly (or ultim acy) o f the m eanings which it form ulates, relative to other m eanings o f the system . The keyness o f an elaboratin g sym bol derives from the bro ad n ess o f its scope, the extent to which it system atically draw s relation sh ips between a w ide range o f diverse cultural elem ents. (3) “ V ertical versus lateral” : T h e keyness o f a sum m arizing sym bol derives from its ability to relate low er-order m eanings to higher-order assu m p tion s, or to “ g ro u n d ” m ore surfacelevel m eanings to their deeper b ases. (The issue here is degree o f generality o f m eaning. W hether m ore general m eanings are termed “ higher” or “ d eeper,” “ ultim ate” or “ fu nda­ m en tal,” by a particular cultural an alyst seem s a m atter o f personal preference.) T he key ness o f an elaboratin g sym bol by contrast derives from its ability to interconnect disp arate ele­ m ents at essentially the sam e level, by virtue of its ability to m anifest (or bring into relief) their form al sim ilarities. All o f these term inological co n trasts - form / content, quantity/quality, lateral/vertical - are really perspectives upon the sam e basic co n trast, for which we have no m ore general term ; that is, when we say a sum m arizing sym bol is “ key” to rhe system , we m ean that its substan tive m eanings have certain kinds o f priority relative to other m eanings o f the system . When we say an elaboratin g sym bol is key to the system , we refer to the pow er o f its form al or organ ization al role in relation ro the system .

158

SHERRY B. ORTNER

Rut at this point we m ust stop short o f reify­ ing the distinctions, for, in practice, the con ­ trast between the tw o broad types o f key sym bols and the tw o m odes o f “ keyness” m ay break dow n. It seem s em pirically to be the case that an elaboratin g sym bol which is accorded w ide-ranging applicability in the culture played in m any con texts, or applied to m any different sorts o f form s - is generally not only form ally ap t but also substantively referential to high-level values, id eas, cognitive assertio n s, and so forth. Indeed, insofar as such high level fo rm u lation s are m ade, a key elaboratin g sym bol o f a culture m ay m ove into the sacred m ode and op erate in much the sam e w ay as d oes a sum m arizing sym bol. A nd, on the other hand, som e sum m arizing sym bols may play im portan t ordering functions, as when they relate the respondent not merely to a cluster o f high-level assum p tion s and values, but to a p articular scenario which may be replayed in ongoin g life. (One may think, for exam p le, o f the Christian cross evoking, am o n g other things, not only a general sense o f G o d ’s purpose and su p p ort, but a lso the p articular scenario o f C h rist’s m artyrdom .) T h u s we are brought to an im portant point, nam ely, that we are distinguishing not only types o f sym bols, but types o f sym bolic func­ tions. These functions m ay be perform ed by any given sym bol - at different tim es, or in different co n texts, or even sim ultaneously by different “ levels” o f its m eaning. While there are m any exam p les o f sum m arizing and e la b o ­ rating sym bols in their relatively pure form s, the kinds o f functions or operation s these sym bols perform m ay also be seen a s asp ects o f any given sym bols. T o sum m arize the original schem e briefly, key sym bols m ay be discovered by virtue o f a num ber o f reliable indicators which poin t to cultural focus o f interest. They are o f tw o b ro ad types - sum m arizing and elaboratin g. Sum m arizing sym bols are prim arily ob jects o f attention and cultural respect; they synthesize o r “ co llap se ” com plex experience, and relate

the respondent to the groun ds o f the system as a w hole. They include m ost im portantly sacred sym bols in the traditional sense. Elaboratin g sym bols, on the other hand, are sym bols valued for their contribution to the ordering or “ sorting o u t” o f experience. Within this are sym bols valued prim arily for the ordering o f conceptual experience, i.e., for providing cultural “ orien tation s,” an d those valued pri­ m arily for the orderin g o f action , i.e., for p ro ­ viding cultural “ strategies.” The form er includes w hat Pepper calls “ root m etap h o rs,” the latter includes key scen arios, or elements o f scenarios which are crucial to the m eansend relationship postu lated in the com plete scenario. This schem e also suggests, at least by the choices o f term s, the m odes o f sym bolic an aly­ sis relevant to the different types o f key sym bols. The first type (sum m arizing sym bols) suggests a range o f question s pertaining to the cultural conversion o f com p lex ideas into various kinds o f relatively undifferentiated com m itm ent - patrio tism , for exam ple, or faith. T he second type (root m etaphors) su g­ gests questions applicab le to the analysis o f m etaphor in the broadest sense - questions o f how thought proceeds and organizes itself through an alogies, m odels, im ages, and so forth. And the third type (key scenarios) su g­ gests dram atistic m odes o f an alysis, in which one raises question s concerning the restructur­ ing o f attitudes an d relationships as a result o f enacting particular culturally provided sequences o f stylized actions. This article has been frankly program m atic; I am in the process o f im plem enting som e o f its ideas in a m on ograph on Sherpa social and religious relations. H ere I have sim ply been concerned to show th at, although a m ethod o f cultural analysis via key sym bols has been for the m ost part unarticulated, there is at least incipiently m ethod in such analysis. It is w orth our while to try to system atize this m ethod, for it m ay be our m ost pow erful entree to the distinctiveness and variability o f hum an cultures.

ON KEY SYM BOLS

REFERENCES CITED A lbert, Ethel. 1 9 5 6 . The C lassification o f Values: A M ethod and Illustration. Ameri­ can Anthropologist 58: 2 2 1 -4 8 . Benedict, Ruth. 1934. Patterns o f Culture. Boston: H oughton-M ifflin. ------.1 9 6 7 . The Chrysanthemum and the Sword. Cleveland and N ew Y ork: W orld. Cohen, A. K . 1948. O n the Place o f “ T h em es” and Kindred C on cepts in Social Theory. American Anthropologist 50: 4 3 6 -4 3 . D o u glas, M ary. 196 6 . Purity and D anger . N ew York: Praeger. D uB ois, C o ra. 193 6 . The W ealth C oncept as an Integrative Factor in T olow a-T ututn i Culture. In Essays in Anthropology Pre­ sented to A. L. Kroeber. R obert Low ie, Ed. Berkeley: University o f C alifo rn ia Press. ------.1 9 5 5 . The D om in an t Value Profile o f Am erican Culture. American Anthropolo­ gist 57: 1 2 3 2 -9 . Evans-Pritchard, E. E. 1962. Social Anthropol­ ogy and Other Essays. N ew Y ork: Free Press.

159

K luckhohn, Florence. 1950. D om inan t and Substitute Profiles o f C u ltural O rientation. Social Forces 28: 3 7 6 -9 3 . Lienhardt, G odfrey. 1961. Divinity and Expe­ rience. O x fo rd : C larendon Press. O pler, M orris E. 1945. Them es as D ynam ic Forces in Culture. American Jou rn al o f Soci­ ology 51: 1 9 8 -2 0 6 . O rtner, Sherry B. (Sherry O . Paul). 1966. T ibetan Circles. M .A . thesis, University o f C h icago. ------.1 9 7 0 . Food for T h ough t: A Key Sym bol in Sherpa Culture. Ph.D. thesis, University o f C h icago. Pepper, Stephen. 1942. World Hypotheses. Berkeley and L os Angeles: University o f C alifo rnia Press. Schneider, D avid M . 1968. American Kinship. Englew ood C liffs, N .J.: Prentice-H all. Singer, M ilton. 1958. The G reat T raditio n in a M etropolitan Center: M ad ra s. In Tradi­ tional India: Structure and Change. M ilton Singer, Ed. Philadelphia: Am erican Folklore Society. T urn er, Victor. 1967. The Forest o f Symbols. Ithaca: Cornell University Press.

13

The Virgin of Guadalupe: A Mexican National Symbol Eric R. Wolf

Eric W olf (1923-99) was one of the most significant Am erican anthropologists of the 20th century, serving as a leading advocate and clear expositor of a schol­ arly, politically accountable historical m aterialist approach and advancing the study o f peasant or agrarian societies. He conducted fieldw ork in both Latin Am erica and Europe. W olf gives central place to pow er (1999) and to historicizing (rather than abstracting or idealizing)culture;hisbookontheconsequences for "the people w ithout history" of European expansion (1982) has been particularly influential. W olf's approach is evident already in this early interpre­ tation o f a master symbol (in Ortner's terms, a sum m arizing key symbol) that gained national im portance in Mexico. The essay illum inates both the nature of such a symbol and the process of syncre­ tism (cultural mixing). W olf suggests how th e imposition of Christianity was "localized" to provide continuity w ith

pre-Hispanic (Aztec) religion, yet in such a m anner as to conceal the continuity from the conquerors. He thus draws close links betw een religious practices, social change, resistance, and state pow er - them es to be pursued in Part IV. W olf also provides an insightful analy­ sis of the Virgin as a mother-figure, thus im plicitly illustrating th at symbols can become especially powerful w hen they draw on both deep psychological and social structural dimensions (a point explicitly articulated in Turner 1967). Interestingly, W olf draws on tw o kinds of mother-images which figure longing and rebellion, respectively. The G u ad al­ upe symbol links the psychological sources w ith the struggle for class and national deliverance and dignity. For a powerful psychoanalytic interpretation of th e Virgin in a different cultural context (Italy), see Parsons (1969), as well as subsequent w ork by Carroll (1986).

Eric R. Wolf, “The Virgin of Guadalupe: A Mexican National Symbol,” Journal of American Folklore LXX1 (1958): 34-9. Reprinted by permission of the American Anthropological Association.

THE VIRGIN OF G UADALUPE: A MEXICAN NATIONAL SYM BOL

There is much w ork from a theologi­ cal and missionary perspective on w hat is called "inculturation", i.e., the adap­ tation of Christianity to local under­ standings. Anthropologists are more likely to look at the w ay symbols and symbolic practices provide foci and means of struggle and resistance to the hegem ony of colonizers and other dom­ inant groups, or play a role in address­ ing moral problems produced by rapid

O ccasion ally, we encounter a sym bol which seem s to enshrine the m ajor hopes and a sp ira ­ tion s o f an entire society. Such a m aster sym bol is represented by the Virgin o f G u ad alu p e, M ex ico ’ s patron saint. D u rin g the M exican W ar o f Independence again st Spain , her im age preceded the insurgents into b a ttle .1 Em iliano Z a p a ta and his agrarian rebels fought under her em blem in the G reat R evolution o f 1910.2 T o d a y , her im age adorn s house fronts and inte­ riors, churches and hom e a ltars, bull rings and gam blin g dens, tax is and buses, restaurants an d houses o f ill repute. She is celebrated in p o p u lar son g and verse. H er shrine at T epeyac, im m ediately north o f M ex ico C ity, is visited each year by hundreds o f th ousan d s o f pil­ grim s, ranging from the in h abitants o f far-off Indian villages to the m em bers o f socialist trade union locals. “ N oth in g to be seen in C an ad a or E u ro p e,” say s F. S. C . N orth rop , “ equals it in the volum e o r the vitality o f its m oving quality or in the depth o f its spirit o f religious d evotio n .” 3 In this p ap er, I should like to d iscu ss this M exican m aster sym bol, and the ideology which surrou nds it. In m ak in g use o f the term “ m aster sy m b o l,” I do not wish to im ply that belief in the sym bol is com m on to all M ex i­ can s. We are not dealin g here with an element o f a putative nation al ch aracter, defined as a com m on denom in ator o f all M exican nation­ als. It is no longer legitim ate to assum e “ that any m em ber o f the [national] g ro u p will exhibit certain regularities o f behavior which are com m on in high degree am o n g the other m em bers o f the society.” 4 N a tio n s, like other

161

social change, abjection, class form a­ tion, and divergent economic opportu­ nities. Hunt (1977) offers a structural analysis of pre-Colum bian M esoamerican reli­ gion; for Mexican Christianity see Ingham (1986) and Eiss (2002). A thoughtful collection on syncretism is Stew art and Shaw (1994) and see also G ellner (2001). Finally, W olf has an excel­ lent essay on Santa Claus (1964).

co m plex societies, m ust, how ever, “ p ossess cultural form s or m echanism s which grou ps involved in the sam e over-all w eb o f relation ­ ships can use in their form al and inform al dealin gs with each o th er.” 5 Such form s develop historically, hand in hand with other processes which lead to the form ation o f n ations, and social g ro u p s which are caught up in these processes m ust becom e “ accu ltu rated” to their u sage.6 O nly where such form s exist, can com m u nication and coordin ated behavior be establish ed am o n g the constituent g ro u p s o f such a society. They provide the cultural idiom o f behavior an d ideal representations through which different g rou p s o f the sam e society can pursue and m anipulate their different fates within a coordin ated fram ew ork. T h is paper, then, deals with one such cultural form , oper­ atin g on the sym bolic level. The study o f this sym bol seem s particularly rew arding, since it is not restricted to one set o f social ties, but refers to a very w ide range o f social relationships. T he im age o f the G u ad alu p e and her shrine at T ep eyac are surrounded by an origin m yth.7 A ccording to this m yth, the Virgin M ary appeared to Ju a n D iego, a C hristianized Indian o f com m oner statu s, and addressed him in N ah u atl. The encounter took place on the Hill o f T epeyac in the year 1531, ten years after the Spanish C o n qu est o f T enochtitlan. The Virgin com m an ded Ju a n D iego to seek ou t the arch ­ bish op o f M ex ico and to inform him o f her desire to see a church built in her honor on T epeyac Hill. After Ju a n D iego w as twice unsuccessful in his efforts to carry out her

162

ERIC R. W OLF

order, the Virgin w rought a m iracle. She bade Ju an D iego pick roses in a sterile sp ot where norm ally only desert plan ts could grow , g ath ­ ered the roses into the Indian’s clo ak , and told him to present clo ak and roses to the incredu­ lous arch bish op. When Ju an D iego unfolded his cloak before the bishop, the im age of the Virgin w as m iraculously stam ped upon it. The bishop ackn ow led ged the m iracle, and ordered a shrine built where M ary had ap p eared to her humble servant. The shrine, rebuilt several tim es in centuries to follow , is to d ay a basilica, the third highest kind o f church in W estern C hristendom . A bove the central altar hangs Ju a n D iego’s cloak with the m iraculous im age. It show s a young w om an w ithout child, her head low ered demurely in her shaw l. She w ears an open crown and flow ing gow n, and stan d s upon a h alf m oon sym bolizing the Im m aculate C onception. The shrine o f G u ad alu p e w as, how ever, not the first religious structure built on Tepeyac; nor w as G u ad alu p e the first fem ale sup ern atu­ ral associated with the hill. In pre-H ispanic tim es, T epeyac had housed a tem ple to the earth and fertility god d ess T on antzin , O ur Lady M oth er, w ho - like the G u ad alu p e - w as associated with the m oon. T em ple, like b asil­ ica, w as the center o f large-scale pilgrim ages. T h at the veneration accorded the G u adalu pe drew inspiration from the earlier w orship o f T onantzin is attested by several Spanish friars. F. Bernardino de Sah agu n , w riting fifty years after the C o n qu est, says: Now that the Church of Our Lady o f G uada­ lupe has been built there, they call her Tonantzin too. . . . The term refers . . . to that ancient Tonantzin and this state o f affairs should be remedied, because the proper name of the M other of God is not Tonantzin, but Dios and Nantzin. It seems to be a satanic device to m ask idolatry . . . and they come from far away to visit that Tonantzin, as much as before; a devotion which is also suspect because there are many churches o f Our Lady everywhere and they do not go to them; and they come from faraway lands to this Tonantzin as o f old.8 F. M artin de Leon w rote in a sim ilar vein:

On the hill where Our Lady of Guadalupe is they adored the idol o f a goddess they called Tonantzin, which means Our Mother, and this is also the name they give Our Lady and they always say they are going to Tonantzin or they are celebrating Tonantzin and many o f them understand this in the old way and not in the modern w a y .. . .9 The syncretism w as still alive in the seven­ teenth century. F. Ja c in to de la Serna, in d is­ cussing the pilgrim ages to the G u ad alu p e at T epeyac, noted: . . it is the purpose o f the wicked to [w orship] the goddess and not the M ost H oly V irgin, or both togeth er.” 10 Increasingly p o p u lar during the sixteenth century, the G u ad alu p e cult gathered em o­ tional im petus during the seventeenth. D uring this century ap p ear the first know n pictorial representations o f the G u ad alu pe, ap art from the m iraculous origin al; the first poem s are written in her hon or; and the first serm ons announce the transcendental im plications o f her supernatural appearance in M exico and am on g M e x ic a n s." H istorian s have long tended to neglect the seventeenth century which seem ed “ a kind o f D ark Age in M exico. ” Yet “ this quiet tim e w as o f the utm ost im por­ tance in the developm ent o f M exican Society.” 12 During this century, the institution o f the haci­ enda com es to dom inate M exican life.11 D uring this century, also , “ N ew Spain is ceasing to be ‘new ’ and to be ‘S p ain .’ ” 14 These new exp eri­ ences require a new cultural idiom , and in the G u ad alu p e cult, the com ponent segm ents o f M exican colonial society encountered cultural form s in which they could express their p a ra l­ lel interests and longings. The prim ary purp o se o f this paper is not, how ever, to trace the history o f the G u adalu pe sym bol. It is concerned rather with its func­ tional asp ects, its roots and reference to the m ajor social relation sh ips of M exican society. The first set o f relationships which I w ould like to single out for consideration are the ties o f kinship, and the em otions generated in the play o f relationships within fam ilies. I w ant to suggest that som e o f the m eanings o f the Virgin sym bol in general, and o f the G u a d a l­ upe sym bol in particular, derive from these em otions. I say “ som e m eanings” and I use the

THE VIRGIN OF GUADALUPE: A MEXICAN NATIONAL SYMBOL

term “ derive” rather than “ origin ate,” because the form and function o f the fam ily in any given society are them selves determ ined by other social factors: technology, econom y, residence, political pow er. The fam ily is but one relay in the circuit within which sym bols are generated in co m p lex societies. A lso, I used the plural “ fam ilies” rather than “ fam ily,” because there are d em on strably m ore than one kind o f fam ily in M e x ic o .151 shall sim plify the available inform ation on M exican fam ily life, and discuss the m aterial in term s o f tw o m ajor types o f fam ilies.16 The first kind o f fam ily is congruent with the closed and static life o f the Indian village. It m ay be called the Indian family. In this kind o f fam ily, the husband is ideally dom inant, but in reality labor and authority are shared equally am on g both m ar­ riage partners. E xploitation o f one sex by the other is atypical; sexual feats d o not add to a person’s status in the eyes o f others. Physical punishm ent and authoritarian treatm ent o f children are rare. The second kind o f family is congruent with the much m ore open, m obile, m anipulative life in com m unities which are actively geared to the life o f the nation, a life in which pow er relationsh ips between indi­ viduals and grou p s are o f great m om ent. T his kind of fam ily m ay be called the M exican fam ily. H ere, the father’s authority is unques­ tioned on both the real and the ideal plane. D ouble sex stan d ards prevail, and m ale sexu­ ality is charged with a desire to exercise dom i­ nation. Children are ruled with a heavy hand; physical punishm ent is frequent. The Indian fam ily pattern is consistent with the behaviour tow ard s the G u ad alu p e noted by Jo h n Bushnell in the M atlazin ca-speakin g com m unity o f San Ju a n A tzingo in the Valley o f T o lu c a .17 There, the im age o f the Virgin is addressed in passio n ate term s as a source o f warm th and love, and the pulque or century plant beer drunk on cerem onial occasion s is identified with her milk. Bushnell postulates that here the G u ad alu p e is identified with the m other as a source o f early satisfaction s, never again experienced after separation from the m other and emergence into social adulthood. A s such, the G u ad alu p e em bodies a longing to return to the pristine state in which hunger and unsatisfactory social relations are minimized.

163

The second fam ily pattern is a lso consistent with a sym bolic identification o f Virgin and m other, yet this time within a con text o f adult m ale dom inance and sexual assertion , dis­ charged again st subm issive fem ales and chil­ dren. In this second context, the G u adalupe sym bol is charged with the energy o f rebellion again st the father. H er im age is the em b odi­ ment o f hope in a victorious outcom e o f the struggle between generations. T h is struggle leads to a further extension o f the sym bolism . Successful rebellion again st pow er figures is equated with the prom ise o f life; defeat with the prom ise o f death. As John A. M ack ay has suggested, there thus takes place a further sym bolic identification o f the Virgin with life; o f defeat and death with the crucified Christ. In M exican artistic tradition, as in H ispanic artistic tradition in general, Ch rist is never depicted as an adu lt m an, but alw ays either as a helpless child, or m ore often as a figure beaten, tortured, defeated and killed. In this sym bolic equation we are touching upon som e o f the roots both o f the passio n ate affir­ m ation o f faith in the Virgin, and o f the fascin a­ tion with death which characterizes Baroque C hristianity in general, and M exican C ath o li­ cism in particular. The G u adalu p e stan ds for life, for hope, for health; C hrist on the cross, for despair and for death. Supernatural m other and natural m other are thus equated sym bolically, as are earthly and otherw orldly hopes and desires. These hopes center on the provision o f food and em otional w arm th in rhe first case, in the successful w aging o f the O edipal struggle in the other. Fam ily relations are, however, only one element in the form ation o f the G u adalupe sym bol. Their analysis does little to explain the G u ad alu p e a s such. They merely illum inate the fem ale and m aternal attributes o f the more w idespread Virgin sym bol. The G u ad alu p e is im portant to M exican s not only because she is a supernatural m other, but a lso because she em bodies their m ajor political and religious aspiration s. T o the Indian g ro u p s, the sym bol is more than an em bodim ent o f life and hope; it restores to them the hopes o f salvation. We m ust not forget that the Spanish C on qu est sig­ nified not only m ilitary defeat, but the defeat

164

ERIC R. W OLF

a lso o f the old go d s and the decline o f the old ritual. T he ap p aritio n o f the G u ad alu p e to an Indian com m oner thus represents on one level the return o f T on antzin . As T an n enbaum has well said, “ The Church . . . gave the Indian an opportunity not merely to save his life, but also to save his faith in his ow n g o d s .” 1* O n another level, the myth o f the ap p aritio n served as a sym bolic testim ony that the Indian, as much as the S p an iard , w as cap able o f being saved, cap able o f receiving C hristianity. T his m ust be understood again st the background o f the bitter theological and political argum ent which follow ed the C o n qu est and divided churchm en, officials, and con qu erors into those w ho held that the Indian w as incapable o f conversion, thus inhum an, and therefore a fit subject o f political and econom ic exp lo ita­ tion; and those w ho held that the Indian w as hum an, cap ab le o f conversion and that this exploitation had to be tem pered by the dem ands o f the C atholic faith and o f orderly civil processes o f governm ent.20 The myth of the G u ad alu p e thus validates the Indian’s right to legal defense, orderly governm ent, to citi­ zenship; to supern atu ral salvation , but a lso to salvation from random oppression . Hut if the G u ad alu p e guaran teed a rightful place to the Indians in the new social system o f N ew Spain, the myth a lso held appeal to the large g ro u p o f disinherited w ho arose in N ew Spain as illegitim ate o ffsp rin g o f Spanish fathers and Indian m others, or through im pov­ erishm ent, acculturation or loss o f status within the Indian or Span ish g ro u p .21 For such people, there w as for a long tim e no proper place in the social order. T heir very right to exist w as questioned in their inability to com m and the full rights o f citizenship and legal protection. W here Span iard and Indian stood squarely within the law , they inhabited the interstices and m argins o f constituted society. These grou p s acquired influence and w ealth in the seventeenth and eighteenth centuries, but were yet barred from social recognition and pow er by the prevailing econom ic, social and political ord er.22 T o them , the G u ad alu p e myth cam e to represent not merely the guarantee o f their assured place in heaven, but the guarantee o f their place in society here and now . On the political plane, the wish for a return to a

paradise o f early satisfaction s o f food and w arm th, a life w ithout defeat, sickness or death, gave rise to a political wish fo ra M exican p arad ise, in which the illegitim ate sons w ould possess the coun try, and the irresponsible Spanish ov erlords, w ho never acknow ledged the social responsibilities o f their paternity, w ould be driven from the land. In the w ritings o f seventeenth century eccle­ siastics, the G u ad alu p e becom es the harbinger o f this new order. In the book by M iguel Sanchez, published in 1648, the Spanish C o n ­ quest o f N ew Spain is justified solely on the grou n d s that it allow ed the Virgin to becom e m anifest in her chosen country, and to found in M exico a new p arad ise. Ju st as Israel had been chosen to produce C h rist, so M exico had been chosen to produce G u ad alu p e. Sanchez equates her with the apocalyptic w om an o f the R evela­ tion o f Jo h n ( 1 2 :1), “ arrayed with the sun, and the m oon under her feet, and upon her head a crow n o f twelve sta rs” w ho is to realize the prophecy o f D euteronom y 8 :7 -1 0 and lead the M exican s into the Prom ised Land. C olon ial M exico thus becom es the desert o f Sinai; Inde­ pendent M ex ico the land o f milk and honey. F. Francisco de Florencia, w riting in 1688, coined the slogan which m ade M exico not merely another chosen nation, but the C hosen N atio n : non fecit taliter omni nationi,2i w ords which still adorn the p o rtals o f the basilica, and shine forth in electric light bulbs at night. And on the eve o f M exican independence, Servando T eresa de M ier elab o rates still further the G u ad alu p an myth by claim in g that M exico had been co n ­ verted to Christianity long before the Spanish C on qu est. The apostle Saint T h o m as had brought the im age o f G uadalupe-T on an tzin to the N ew W orld a s a sym bol o f his m ission, just as Saint Ja m e s had converted Spain with the im age o f the Virgin o f the Pillar. The Spanish C o n qu est w as therefore historically unneces­ sary, and should be erased from the an n als o f h istory.24 In this perspective, the M exican W ar o f Independence m ark s the final realization o f the apocalyp tic prom ise. The banner o f the G u ad alu p e leads the insurgents; and their cause is referred to as “ her law .” 25 In this ulti­ m ate extension o f the sym bol, the prom ise o f life held ou t by the supernatural m other has becom e the prom ise o f an independent M ex ico,

THE VIRGIN OF G UADALUPE: A MEXICAN NATIONAL SYMBOL

165

liberated from the irrational authority o f the Spanish fath er-oppressors and restored to the C hosen N ation w hose election had been m anifest in the ap p aritio n o f the Virgin on T epeyac. The land o f the supern atu ral m other is finally possessed by her rightful heirs. The sym bolic circuit is closed. M oth er; fo o d , hope, health, life; supern atural salvation and salv a­ tion from op p ressio n ; C hosen People and national independence - all find expression in a single m aster sym bol.

The G u ad alu p e sym bol thus links together fam ily, politics and religion; colonial past and independent present; Indian and M exican. It reflects the salient social relationships o f M exican life, and em bodies the em otions w hich they generate. It provides a cultural idiom through which the tenor and em otions o f these relationsh ips can be expressed. It is, ultim ately, a w ay o f talking a b o u t M exico: a “ collective represen tation ” o f M exican society.

NOTES

13

1 2

3

4

5

6 7 8

9

10

11

12

N iceto de Z a m a co is, H istoria de Mexico (B arcelon a-M exico, 1 8 7 8 -8 2 ), VI, 253. A ntonio Pom pa y P om pa, Album del IV centenario guadalupano (M exico , 1 938), p. 173. F. S. C. N orth rop , The Meeting o f East and West (N ew Y ork, 1946), p. 25. D avid G . M an delb au n i, “ O n the Study o f N ation al C h a ra cte r," American Anthro­ pologist., LV (1 9 5 3 ), 185. Eric R. W olf, “ A spects o f G ro u p R ela­ tions in a C o m p lex Society: M ex ic o ,” American Anthropologist , l.VII (1 9 5 6 ), 1 0 65-78. Erie R. W olf, “ La fo rm acion de la n acion ,” Ciencias Sociales , IV, 5 0 -1 . Ernest G ruening, M exico and Its Heritage (N ew Y ork, 1928), p. 2 3 5 . Bernardino de Sah agu n , H istoria general de las cosas de nueva espana (M exico, 1938), I, lib. 6. Q uoted in C arlo s A. Echanove T rujillo, Sociologia mexicana (M exico , 1948), p. 105. Q uoted in Je su s A m ay a, L a madre

14 15

16

17

18 19

de D ios: genesis e historia de nuestra senora de G uadalupe (M ex ico , 1931), p.

20

230. Francisco de la M a z a , El guadalupism o mexicano (M exico , 1 9 53), pp. 1 2 -1 4 , 143, 3 0 , 3 3 , 82. Lesley B. Sim pson, “ M e x ic o ’s Forgotten C en tu ry,” Pacific H istorical Review , X X II ( 1 9 5 3 )’ 115, 114.

21

Fran cois Chevalier, La form ation des grands domaines au M exique (Paris, 1952), p. xii. de la M a z a , p. 41. M arfa Elvira Berm udez, L a vida familiar del m exicano (M exico , 1 9 55), chapters 2 and 3. For relevant m aterial, see: Berm udez; Jo h n G illin, “ Ethos and C u ltural A spects o f Person ality,” and R obert Redfield and Sol T a x , “ G eneral C h aracteristics o f Present-D ay M esoam erican Indian Society,” in Sol T a x , ed., Heritage o f Conquest (G lencoe, 1952), pp. 1 9 3 -2 1 2 , 3 1 - 9 ; G o rd on W. H ew es, “ M exican s in Search o f the ‘ M ex ican ’, ” American jo u rn al o f Economics and Sociology , XIII ( 19 5 4 ), 2 0 9 - 2 3 ; O ctavio Paz, El laberinto de la soledad (M exico, 1 9 4 7 ), pp. 7 1 -8 9 . Jo h n Bushnell, “ L a Virgen de G u adalu p e as Surrogate M oth er in San J uan A tzin go,” paper read before the 54th Annual M eetin g o f the A m erican A nthropological A ssociation , 18 N ovem ber 1955. Jo h n A. M ack ay , The O ther Spanish Christ (N ew Y ork, 1933), pp. 1 1 0 -1 7 . Frank T an n enbaum , Peace by Revolution (N ew Y o rk, 1 9 33), p. 39. Silvio Z a v a la , L a filosofia en la conquista de America (M exico, 1947). N ico las Leon, L as castas del M exico colo­ nial o Nueva Espana (M exico, 1924); C. t:. M arsh all, “ T he Birth o f the M estizo in N ew S p ain ,” Hispanic American H istori­ cal Review , X IX (1 9 3 9 ), 1 6 1 -8 4 ; W olf, “ La fo rm acion de la n acio n ,” pp. 1 0 3 -6 .

166 22

23

ERIC R. W OLF

G regorio T orres Q uin tero, M exico hacia el fin del virreinato espanol (M exico, 1 921); Kric R. W olf, “ The M exican B ajio in the Eighteenth C en tu ry,” Middle Amer­ ican Research Institute Publication XV II (1 9 5 5 ), 1 8 0 - 9 9 ; W olf, “ A spects o f G ro u p R elations in a C o m p lex Society: M ex ic o .” de la M a z a , pp. 3 9 - 4 0 , 4 3 - 9 , 64.

24

25

Luis V illoro, L os grandes momentos del indigenismo en M exico (M exico, 1 9 50), pp. 1 3 1 -3 8 . Luis G on zalez y G onzalez, “ El optim ism o nacionalista co m o factor en la independencia de M e x ic o ,” Estudios de historiografia am ericana (M exico, 1948), p. 194.

Structure, Function, and Interpretation

Introd uction It is one thing to recognize sym bols as the building blocks o f religious w orlds and another thing to know how to go about analyzing symbolic constructions. Anthro­ pologists have been in som e disagreement over rhe correct manner to investigate such symbolic phenomena as myths, ideology, systems o f classification, ritual injunc­ tions, and aesthetic productions and perform ances. Should one focus on structure or content, surface or depth, figure or ground? Should explanation lie with prag­ matic interests, social function, meaningful texture, unconscious conflicts, or cogni­ tive processes? This section highlights som e o f the forms o f explanation or interpretation to which anthropologists have turned, and debates am on g them. This is a large topic and a rich field. I have selected a few forceful statements by wellknown figures.

14

Myth in Primitive Psychology Bronislaw Malinowski

Bronislaw M alinowski (1884-1942) is one of the founders of modern anthro­ pology. In a series of rem arkable works w ritten largely in the 1920s, he dem on­ strated the im portance of rich ethno­ graphic observation. Born in Poland, he spent most of his teaching career at the London School of Economics, w here he influenced a large cohort of students, many of w hom made their own im por­ tan t contributions to the ethnographic corpus on religion (notably Firth 1940, 1967, and Richards 1956, am ong many others). From this entry readers may glimpse how M alinowski was able to capture the im agination of earlier generations of audiences, both professional and public. M alinowski invites us to share his rom antic and pleasurable field odyssey in the Trobriand Islands of M elanesia w hile m aking no bones about the supe­ riority of his approach over preceding ones. This essay is fam ous for develop­

ing the argum ent of myth as "charter" but it can be seen, I think, that M alinowski escapes the narrow ly func­ tionalist interpretations that are often placed on his ideas. Among other things, he emphasizes the immediacy and "living reality" of myth as w ell as its discursive and pragm atic (hence dialogi­ cal) qualities (that is, the recitation of myths as speech events). W hile it would be wrong to suggest either that all myths have political functions or that the interest or value of any given myth can be reduced to its instrumental political function, M alinowski was undoubtedly correct to look at the place of myth in legitim ating particular forms of social organization and loci of power or interest and the contestation this inevitably brings. Myth becomes a lan­ guage of legal argum ent. Useful developm ents and exem plifica­ tions of M alinowski's approach include Leach's discussion of "myth as a justifica-

h'rom Bronislaw Malinowski, “ Myth in Primitive Psychology,” in Magic * Science and Religion and Other Essays (Garden City, NY: Doubleday &£ Co., Inc., 1954 |1926]), pp. 100-26, 145. Reprinted by permission of Macmillan Publishers, Inc. Abridged.

MYTH IN PRIMITIVE PSYCHOLO GY

tion for faction and social change" in highland Burma (1964: Chapter IX) and Andriolo's essay on genealogy in the Old Testam ent (1973). Leach him self later turned to structuralist (Levi-Straussian) analyses of the Old Testam ent (1969). W hereas M alinowski claims the meaning

M yth as it exists in a sav age com m unity, th at is, in its living prim itive form , is not m erely a story told but a reality lived. It is not o f the nature o f fiction, such as we read today in a novel, but it is a living reality, believed to have once happened in prim eval tim es, and continuing ever since to influence the w orld an d hum an destinies. T h is myth is to the sav age w hat, to a fully believing C h ristian , is the Biblical story o f C reation , o f the Fall, o f the R edem ption by C h rist’s Sacrifice on the C ro ss. A s our sacred story lives in ou r ritual, in ou r m orality, as it governs ou r faith and con trols ou r con duct, even so d oes his myth for the savage. The lim itation o f the study o f myth to the mere exam in ation o f texts has been fatal to a p rop er understanding o f its nature. The form s o f myth which com e to us from classical an ti­ quity and from the ancient sacred b o o k s o f the E ast an d other sim ilar sources have com e dow n to us w ithout the con text o f living faith, w ithout the possibility o f obtain in g com m ents from true believers, w ithout the concom itant know ledge o f their social organ ization , their practiced m orals, and their p op u lar cu stom s a t least w ithout the full in form ation which the m odern fieldw orker can easily obtain . M o re ­ over, there is no d ou bt th at in their present literary form these tales have suffered a very con siderable tran sform ation at the hands o f scribes, com m en tators, learned priests, and th eologian s. It is necessary to go back to prim i­ tive m ythology in order to learn the secret o f its life in the study o f a myth which is still alive - before, m um m ified in priestly w isdom , it has been enshrined in the indestructible but lifeless repository o f dead religions. Studied alive, m yth, as we shall see, is not sym bolic, but a direct expression o f its subject

169

of myth is on the surface, many writers w ould disagree. Marx and Freud are both noted for arguing th at m eaning is concealed and needs to be recovered, w hile structuralists see the issue as one of revealing the codes or gram m ar by w hich m eaning is produced.

m atter; it is not an exp lanation in satisfaction o f a scientific interest, bur a narrative resurrec­ tion o f a prim eval reality, told in satisfaction o f deep religious w an ts, m oral cravings, social su b m issio n s, assertion s, even practical require­ m ents. M yth fulfills in prim itive culture an in dispen sable function: it expresses, enhances, and codifies belief; it safeg u ard s and enforces m orality; it vouches for the efficiency o f ritual and co n tain s practical rules for the guidance o f m an. M yth is thus a vital ingredient o f hum an civilization; it is not an idle tale, but a hard-w orked active force; it is not an intellec­ tual e xp lan ation or an artistic im agery, but a p ragm atic charter o f prim itive faith and m oral w isdom . [ ...] In the subsequent ch apters o f this bo ok we will exam ine a num ber o f m yths in detail, but for the m om ent let us glance at the subjects o f som e typical m yths [from the T ro brian d Islands). T ak e, for instance, the annual feast o f the return o f the dead. E lab orate arran ge­ m ents are m ade for it, especially an enorm ous d isplay o f food. When this feast app roach es, tales are told o f how death began to chastise m an, an d how the pow er o f eternal rejuvena­ tion w as lost. It is told why the spirits have to leave the village and d o not rem ain at the fire­ side, finally why they return once in a year. A gain, at certain season s in p reparation for an o v erseas expedition, can oes are overhauled and new ones built to the accom pan im ent o f a special m agic. In this there are m ythological allu sion s in the spells, and even the sacred acts con tain elem ents which are only com prehen­ sible when the story o f the flying canoe, its ritual, and its m agic are told. In connection with cerem onial trading, the rules, the m agic, even the geograph ical routes are associated with correspon din g m ythology. There is no

170

BRONISLAW MALINOWSKI

im portan t m agic, no cerem ony, no ritual w ithout belief; and the belief is spun out into accoun ts o f concrete precedent. T he union is very intim ate, for myth is not only looked upon as a com m entary o f addition al in form a­ tion, but it is a w arran t, a charter, and often even a practical guide to the activities with which it is connected. O n the other hand the rituals, cerem onies, cu stom s, and social o rg a ­ nization con tain at tim es direct references to myth, and they are regarded as the results o f mythical event. T he cultural fact is a m on u­ ment in which the myth is em bodied; while the myth is believed to be the real cause which has brought ab o u t the m oral rule, the social g ro u p ­ ing, the rite, or the custom . T h u s these stories form an integral p art o f culture. Their e xis­ tence an d influence not merely transcend the act o f telling the n arrative, not only d o they draw their substance from life and its interests - they govern and control m any cultural fea­ tures, they form the d ogm atic backbon e o f prim itive civilization. This is p erh ap s the m ost im portant poin t o f rhe thesis which I am urging: I m aintain that there exists a special class o f stories, regarded as sacred , em bodied in ritual, m orals, and social organ izatio n , and which form an integral and active part o f prim itive culture. These stories live not by idle interest, not as fictitious or even as true narratives; but are to the natives a statem ent o f a prim eval, greater, and m ore relevant reality, by which the present life, fates, and activities o f m ankind are deter­ mined, the know ledge o f which supplies man with the m otive for ritual and m oral action s, as well a s with indications as to how to perform them. In order to m ake the point at issue quite clear, let us once m ore com pare our con clu­ sions with the current views o f m odern an th ro ­ pology, not in order idly to criticize other opin ion s, but so that we m ay link our results to the present state o f know ledge, give due acknow ledgm ent for w hat we have received, and state where we have to differ clearly and precisely. It will be best to quote a condensed and authoritative statem ent, and I shall ch oose for this p u rp o se o f definition an an alysis given in Notes an d Queries on Anthropology , by the

late M iss C. S. Burne and Professor J. L. M yres. Under the heading “ Stories, Sayings, and S o n g s,” we are inform ed that “ this section includes m any intellectual efforts o f p eop les” which “ represent the earliest attem pt to exer­ cise reason , im agination, an d m em ory.” With som e apprehension we ask where is left the em otion, the interest, and am bition, the social role o f all the stories, and the deep connection with cultural values o f the m ore serious ones? A fter a brief classification o f stories in the usual m anner we read ab o u t the sacred tales: “ Myths are stories which, how ever m arvelous and im probable to us, are nevertheless related in all go o d faith, because they are intended, or believed by the teller, to explain by m eans o f som ething concrete and intelligible an abstract idea or such vague and difficult conceptions as C reation , D eath, distinctions o f race or anim al species, the different occupation s o f men and w om en; the origins o f rites and cu stom s, or striking natural objects or prehistoric m on u­ m ents; the m eaning o f the nam es o f persons or places. Such stories are som etim es described as etiological , because their purpose is to explain why som ething exists or h appen s.” 1 H ere we have in a nutshell all that m odern science at its best h as to say upon the subject. W ould our M elan esian s agree, how ever, with this opin ion ? Certainly not. They do not w ant to “ e x p la in ,” to m ake “ intelligible” anything which h appens in their m yths - above all not an ab stract idea. O f that there can be found to my know ledge no instance either in M elanesia or in any other savage com m unity. The few ab stract ideas which rhe natives possess carry their concrete com m entary in the very w ord which expresses them . When being is described by verbs to lie, to sit, to stan d, when cause and effect are expressed by w ords signifying foundation and the p ast standing upon it, when various concrete nouns tend to w ard s the m eaning o f space, the w ord and the relation to concrete reality m ake the ab stract idea su f­ ficiently “ intelligible.” N o r w ould a T robriander or any other native agree with the view that “ C reatio n , D eath , distinctions o f race or anim al species, the different occupation s o f men and w om en ” are “ vague and difficult co n cep tion s.” N oth ing is m ore fam iliar to the native than the different occupations o f the

MYTH IN PRIMITIVE PSYCHOLOGY

171

malt* and fem ale sex; there is nothing to be

subsequently by the writers in Notes and

explained ab o u t it. Hut though fam iliar, such

Queries on Anthropology.

differences are at tim es irksom e, unpleasant, or at least lim iting, and there is the need to justify them , to vouch for their antiquity and reality, in short to buttress their validity. Death, alas, is not vague, or ab stract, or diffi­ cult to g rasp for any hum an being. It is only too hauntingly real, too concrete, too easy to com prehend for anyone w ho has had an exp e­ rience affecting his near relatives or a personal foreboding. If it were vague or unreal, man w ould have no desire so m uch as to mention it; but the idea o f death is fraugh t with horror, with a desire to rem ove its threat, with the vague hope that it m ay be, not explained, but rather explained aw ay, m ade unreal, and actu ­ ally denied. M yth , w arranting the belief in im m ortality, in eternal youth, in a life beyond the grave, is not an intellectual reaction upon a puzzle, but an explicit act o f faith born from the innerm ost instinctive and em otional reac­ tion to the m ost form idable and haunting idea. N o r are the stories ab o u t “ the origins o f rites and cu sto m s” told in mere explanation o f them. They never explain in any sense o f the w ord; they alw ays state a precedent which constitutes an ideal and a w arran t for its con ­ tinuance, and som etim es practical directions for the procedure. W e have, therefore, to disagree on every point with this excellent though concise statem ent o f present-day m ythological opinion. T h is definition w ould create an im aginary, non-existent class o f narrative, the etiological m yth, correspon din g to a non-existent desire to explain , leading a futile existence as an “ intellectual e ffo rt,” and rem aining outside native culture and social organ ization with their pragm atic interests. The w hole treatm ent ap p ears to us faulty, because m yths are treated as mere stories, because they are regarded a s a primitive intellectual arm chair occu p a­ tion, because they are torn ou t o f their life context, and studied from w h at they look like on p aper, and not from w h at they do in life. Such a definition w ould m ake it im possible either to see clearly the nature o f myth or to reach a satisfactory classification o f folk tales. In fact we w ould a lso have to disagree with the definition o f legend and o f fairy tale given

Hut abo ve all, this point o f view w ould be fatal to efficient field w ork, for it w ould m ake the observer satisfied with the mere w riting dow n o f narratives. The intellectual nature o f a story is exh austed with its text, but the func­ tional, cultural, and pragm atic aspect o f any native tale is m anifested as much in its enact­ m ent, em bodim ent, and contextual relations as in the text. It is easier to write dow n the story than to observe the diffuse, com plex w ays in which it enters into life, or to study its function by the observation o f the vast social and cultural realities into which it enters. And this is the reason why we have so m any texts and why we know so little about the very nature o f myth. We m ay, therefore, learn an im portant lesson from the T robrian ders, and to them let us now return. We will survey som e o f their myths in detail, so that we can confirm our con clusion s inductively, yet precisely.

M yths o f O rig in We m ay best start with the beginning o f things, and exam ine som e o f the m yths o f origin. The w orld, say the natives, w as originally peopled from underground. H um anity had there led an existence sim ilar in all respects to the present life on earth. U nderground, men were organ ized in villages, clan s, districts; they had distinctions o f rank, they knew privileges and had claim s, they ow ned property, and were versed in m agic lore. Endow ed with all this, they em erged, establishing by this very act certain rights in land and citizenship, in eco­ nom ic prerogative an d m agical pursuit. They brought with them all their culture to continue it upon this earth. There are a num ber o f special sp o ts - g ro t­ toes, clum ps o f trees, stone heaps, coral o u t­ cro p s, sprin gs, heads o f creeks - called “ h oles” or “ h ou ses” by the natives. From such “ h oles” the first couples (a sister as the head o f the fam ily and the brother as her guardian ) cam e and took p ossession o f the lands, and gave the totem ic, industrial, m agical, and sociolo gi­ cal ch aracter to the com m unities thus begun.

172

BRONISLAW MALINOWSKI

The problem o f rank which plays a great role in their sociology w as settled by the em er­ gence from one special hole, called O b u ku la, near the village o f L a b a ’ i. T h is event w as notable in th at, contrary to the usual course (which is: one original “ h ole,” one lineage), from this hole o f L a b a ’i there em erged repre­ sentatives o f the four m ain clans one after the other. T heir arrival, m oreover, w as follow ed by an apparently trivial but, in m ythical reality, a m ost im portan t event. First there cam e the Kaylavasi (igu an a), the anim al o f the Lukulabuta clan, w hich scratched its w ay through the earth as iguan as d o, then clim bed a tree, and rem ained there as a mere on looker, follow ing subsequent events. Soon there cam e out the D og, totem o f the L u ku b a clan, w ho originally had the highest rank. A s a third cam e the Pig, representative o f the M alasi clan, which now holds the highest rank. L ast cam e the Lukw asisiga totem , represented in som e versions by the C rocodile, in others by the Snake, in others by the O p o ssu m , and som etim es com pletely ignored. The D o g and Pig ran roun d, and the D og, seeing the fruit o f the noku plan t, nosed it, then ate it. Said the Pig: “ T h ou eatest noku , thou eatest dirt; thou art a low -bred, a co m ­ m oner; the chief, the guya'u, shall be I.” And ever since, the highest subclan o f the M alasi clan, the T a b a lu , have been the real chiefs. In order to understand this m yth, it is not enough to follow the dialogue between the D o g and the Pig which m ight ap p ear pointless or even trivial. O nce you know the native soci­ ology, the extrem e im portance o f rank, the fact that food and its lim itations (the tab o o s o f rank and clan) are the m ain index o f m an’s social nature, and finally the psych ology o f totem ic identification - you begin to under­ stand how this incident, happening as it did when hum anity w as in statu nascendi , settled once for all the relation between the tw o rival clans. T o understand this myth you m ust have a goo d know ledge o f their sociology, religion, custom s, and ou tloo k . Then, and only then, can you appreciate w hat this story m eans to the natives and how it can live in their life. If you stayed am o n g them and learned the lan­ guage you w ou ld constantly find it active in discussion an d squ ab bles in reference to the relative superiority o f the vario u s clan s, and in

the d iscu ssion s ab o u t the various food tab o o s which frequently raise fine questions of casuistry. A bove all, if you were brought into con tact with com m unities where the historical p rocess o f the spread o f influence o f the M alasi clan is still in evolution, you w ould be brought face to face with this myth as an active force. R em arkably enough the first and last anim als to com e ou t, the iguana and the Lu kw asisiga totem , have been from the beginning left in the cold: thus the num erical principle and the logic o f events is not very strictly observed in the reason in g o f the myth. If the main myth o f L a b a ’ i ab o u t the relative superiority o f the four clans is very often alluded to th rou ghout the tribe, the m inor local m yths are not less alive and active, each in its ow n com m unity. When a party arrives at som e distan t village they will be told not only the legendary historical tales, but above all the m ythological charter o f that co m m u ­ nity, its m agical proficiencies, its occupation al ch aracter, its rank and place in totem ic organ ization . Should there arise land q uarrels, encroachm ent in m agical m atters, fishing rights, or other privileges the testim ony o f myth w ould be referred to. Let me show concretely the w ay in which a typical myth o f local origins w ould be retailed in the norm al run o f native life. Let us w atch a party o f visitors arriving in one or the other o f the T ro b rian d villages. They w ould seat them selves in front o f the h eadm an ’s house, in the central place o f the locality. A s likely as not the spot o f origin s is nearby, m arked by a coral ou tcro p or a heap o f stones. T h is spot w ould be pointed o u t, the nam es o f the brother and sister an cestors m entioned, and perh aps it w ould be said that the m an built his house on the spot o f the present headm an’s dw elling. The native listeners w ould know , o f course, that the sister lived in a different house nearby, for she could never reside within the sam e w alls as her brother. As addition al in form ation , the visitors might be told that the an cestors had brought with them the substan ces and paraph ernalia and m ethods o f local industry. In the village o f Y alak a, for instance, it w ould be the processes for burning lime from shells. In O k o b o b o , O bw eria, and O b o w ad a the an cestors brought

MYTH IN PRIMITIVE PSYCHOLO GY

the know ledge and the im plem ents for polish ­ ing hard stone. In Hwoytalu the carver’s tool, the hafted shark tooth , and the know ledge o f the art cam e out from underground with the origin al an cestors. In m ost places the econom ic m on opolies are thus traced to the au toch th o­ nous em ergence. In villages o f higher rank the insignia o f hereditary dignity were brought; in others som e anim al asso ciated with the local subclan cam e out. Som e com m unities started on their political career o f stan din g hostility to one anoth er from the very beginning. The m ost im portan t gift to this w orld carried from the one below is alw ays m agic; but this will have to be treated later on and m ore fully. If a European bystander were there and heard nothing but the in form ation given from one native to the other, it w ould mean very little to him. In fact, it m ight lead him into serious m isunderstandings. T h u s the sim ulta­ neous em ergence o f brother and sister might m ake him su sp icio u s either o f a m ythological allusion to incest, or else w ould m ake him look for the original m atrim onial pair and inquire ab o u t the sister’s husband. T he first suspicion w ould be entirely erron eous, and w ould shed a false light over the specific relation between brother an d sister, in which the form er is the indispensable guard ian , and the second, equally indispensable, is respon sible for the tran sm ission o f the line. O nly a full know ledge o f the m atrilineal ideas and institutions gives body and m eaning to the bare m ention o f the tw o ancestral nam es, so significant to a native listener. If the European were to inquire who w as the sister’s husband an d how she cam e to have children, he w ould soon find him self once m ore confronted by an entirely foreign set o f ideas - the sociological irrelevance o f the father, the absence o f any ideas ab o u t physio­ logical p rocreation , and the strange and co m ­ plicated system o f m arriage, m atrilineal and patrilocal at the sam e time. The sociological relevance o f these accounts o f origins w ould becom e clear only to a E u ro­ pean inquirer w ho had grasp ed the native legal ideas ab o u t local citizenship and the hereditary rights to territory, fishing g rou n d s, and local pursuits. For accord in g to rhe legal principles o f the tribe all such rights are the m on opolies o f the local com m unity, and only people

173

descendent in the fem ale line from the original an cestress are entitled to them. If the European were told further th at, besides the first place o f em ergence, there are several other “ h oles” in the sam e village, he w ould becom e still m ore baffled until, by a careful study o f concrete details and the principles o f native sociology, he becam e acquain ted with the idea o f co m ­ pound village com m unities, i.e., com m unities in which several subclan s have m erged. It is clear, then, that the myth conveys much m ore to the native than is contained in the m ere story ; that the story gives only the really relevant concrete local differences; that the real m eaning, in fact the full accoun t, is contained in the traditional foun dation s o f social o rg an i­ zation ; and that this the native learns, not by listening to the fragm entary m ythical stories, but by living within the social texture o f his tribe. In other w ords, it is the con text o f social life, it is the gradu al realization by the native o f how everything which he is told to d o has its precedent and pattern in bygone times, which brings home to him the full account and the full m eaning o f his m yths o f origin. For an observer, therefore, it is necessary to becom e fully acquain ted with the social o rg a ­ n ization o f the natives if he w ants really to g rasp its traditional aspect. T he short accoun ts, such a s those which are given ab o u t local o rigin s, will then becom e perfectly plain to him. H e will a lso clearly see that each o f them is only a part, and a rather insignificant one, o f a m uch bigger story, which can n ot be read except from native life. W hat really m atters ab o u t such a story is its social function. It conveys, expresses, and strengthens the fu n da­ m ental fact o f the local unity an d o f the kinship unity o f the grou p o f people descendent from a com m on ancestress. C om bin ed with the conviction that only com m on descent and em ergence from the soil give full rights to it, the story o f origin literally con tain s the legal ch arter o f the com m unity. T h u s, even when the people o f a vanquished com m unity were driven from their groun ds by a hostile neigh­ bor their territory alw ays rem ained intact for them ; and they were alw ays, after a lapse o f tim e and when their peace cerem ony had been con clu ded, allow ed to return to the origi­ nal site, rebuild their village, and cultivate

174

BRONISLAW MALINOWSKI

their gard en s once m ore. The traditional feeling o f a real and intim ate connection with the land; the concrete reality o f seeing the actual sp o t o f emergence in the m iddle o f the scenes o f daily life; the historical continuity o f privileges, occup ation s, and distinctive ch arac­ ters running hack into the m ythological first beginnings - all this obviously m akes for coh e­ sion, for local patriotism , for a feeling o f union and kinship in the com m unity. But although the narrative o f original emergence integrates and w elds together the historical tradition, the legal principles, and the various cu stom s, it m ust a lso be clearly kept in m ind that the original myth is but a sm all part o f the whole com plex o f traditional ideas. T h u s on the one hand the reality o f myth lies in its social func­ tion; on the other hand, once we begin to study the social function o f m yth, and so to recon ­ struct its full m eaning, we are gradually led to build up the full theory o f native social organ ization . O ne o f the m ost interesting phenom ena co n ­ nected with traditional precedent and charter is the adju stm en t o f myth and m ythological principle to cases in which the very foundation o f such m ythology is flagrantly violated. This violation alw ays takes place when the local claim s o f an autoch th on ous clan, i.e., a clan which has em erged on the sp ot, are overridden by an im m igrant clan. Then a conflict o f prin­ ciples is created, for obviously the principle that land and authority belong to those w ho are literally born out o f it does not leave room for any new com ers. On the other hand, m em bers o f a subclan o f high rank w ho choose to settle dow n in a new locality cannot very well be resisted by the autochthons - using this w ord again in the literal native m ythological sense. T h e result is that there com e into existence a special class o f m ythological stories which justify and accoun t for the an om alou s state o f a ffairs. The strength o f the various m ythological and legal principles is m anifested in that the m yths o f justification still contain the an tago n istic and logically irreconcilable facts and poin ts o f view , and only try to cover them by facile reconciliatory incident, o b vi­ ously m anufactured ad hoc. The study o f such stories is extrem ely interesting, both because it gives us a deep insight into the native p sych o­

logy o f tradition, and because it tem pts us to reconstruct the p ast history o f the tribe, though we m ust yield to the tem ptation with due caution and scepticism . [.-•I

A s far as the sociological theory o f these legends goes the historical reconstruction is irrelevant. W hatever the hidden reality o f their unrecorded p ast m ay be, myths serve to cover certain inconsistencies created by historical events, rather than to record these events exactly. T he myths associated with the spread o f the pow erful subclan s show on certain points a fidelity to life in that they record facts inconsistent with one another. The incidents by which this inconsistency is obliterated, if not hidden, are m ost likely fictitious; we have seen certain m yths vary according to the local­ ity in which they are told. In other cases the incidents bolster up non-existent claim s and rights. T he historical consideration o f myth is inter­ esting, therefore, in that it show s that myth, taken a s a w hole, cannot be sober d isp assio n ­ ate history, since it is alw ays m ade ad hoc to fulfill a certain sociological function, to glorify a certain group, or to justify an an om alou s statu s. T hese con sideration s show us also that to the native mind im m ediate history, semihistoric legend, and unm ixed myth flow into one another, form a continuous sequence, and fulfill really the sam e sociological function. And this brings us once m ore to our original contention that the really im portant thing ab o u t the myth is its character o f a retrospec­ tive, ever-present, live actuality. It is to a native neither a fictitious story, nor an accoun t o f a dead past; it is a statem ent o f a bigger reality still partially alive. It is alive in that its prece­ dent, its law , its m oral, still rule the social life o f the natives. It is clear that myth functions especially where there is a sociological strain, such as in m atters o f great difference in rank and pow er, m atters o f precedence and su b o r­ dination, and unquestionably where profoun d historical changes have taken place. So much can be asserted as a fact, though it m ust alw ays rem ain doubtful how far we can carry out historical reconstruction from the myth. We can certainly discard all explanatory as well as all sym bolic interpretations o f these

MYTH IN PRIMITIVE PSYCHOLOGY

myths o f origin. The person ages and beings which we find in them are w hat they ap p ear to be on the surface, and not sym bols o f hidden realities. As to any exp lan atory function o f these m yths, there is no problem which they cover, no curiosity which they satisfy, no theorv which they contain. [ ...] ' The science o f myth in living higher cultures, such a s the present civilization o f India, Ja p a n , C h ina, and last but not least, ou r ow n, m ight well be inspired by the com parative study o f prim itive folklore; and in its turn civilized culture could furnish im portan t addition s and exp lan ation s to savage m ythology. T h is subject is very m uch beyond the scope o f the present

NOTE 1

from N otes and Queries Anthropology , pp. 2 1 0 and 21 1 .

Q uo ted

on

175

study. I do, how ever, w ant to em phasize the fact that an th ropology should be not only the study o f savage custom in the light o f our m entality and our culture, but also the study o f o u r ow n m entality in the distant perspective borrow ed from Stone Age m an. By dw elling m entally for som e time am o n g people o f a m uch sim pler culture than our ow n, we m ay be ab le to see ourselves from a distance, we m ay be able to gain a new sense o f p ro p o r­ tion w ith regard to our ow n institutions, beliefs, and custom s. If an th ropology could thus inspire us with som e sense o f proportion , and supply us with a finer sense o f hum or, it m ight justly claim to be a very great science.

I S

Folk Dialectics of Nature and Culture Marshall Sahlins

M arshall Sahlins is one of the most intel­ lectually im aginative and acute of North Am erican anthropologists, know n for his sharp w it and powerful theoretical argum ents. He has carried out fieldw ork on Fiji and engaged in the study of precolonial Hawaii. Sahlins is Professor Em eritus at the University of Chicago. In this brief analysis Sahlins takes us to the heart of some of the West's most cherished discourses in political philoso­ phy and biological science concerning human nature and the individual. This entry follow s perfectly from M alinowski because, w hile Sahlins illustrates M alinowski's idea of myth as charter w ith respect to the relations of biology and capitalism , he also inverts much of M alinowski's logic. For Sahlins, myth, religion, or culture are not simply prod­ ucts of interest or practice but shape or m ediate practice. In this essay he speaks usefully of a dialectic. W hile the nature/ culture dialectic he describes is particu­

lar to the modern (capitalist) W est, the broader argum ent might fit all societies - t h e w ay in which social logic is autho­ rized by som ething th at is supposedly outside its reach but whose content (as it is understood locally) is actually shaped by th at very logic. The use of "nature" in this dialectic provides echoes both of Marx's analysis of comm odity fetishism in Capital (1961 [1887]) and Levi-Strauss's reworking of "totem ism " in The Savage M in d (1966 [1962]; (see also chapters 23 and 43, below, for alternate cultural approaches to nature). This dialectic is mystified to those w ho are caught up in it. A lthough Sahlins speaks of sociobio­ logy as ideology, contem porary readers m ight prefer to apply Gramsci's concep­ tion of hegem ony (1971, Com aroff and Com aroff 1991), since w hat Sahlins describes has been so largely taken for granted as authoritative rather than rem arked and contested. There is an im portant m ethodological lesson in the

From Marshall Sahlins, “ Folk Dialectics of Nature and Culture,” in The Use and Abuse of Biology: An Anthropological Critique of Sociobiology (Ann Arbor: University of Michigan Press, 1976), pp. 93-107.

FOLK DIALECTICS OF NATURE AND CULTURE

fact th at the "m yth" in this analysis is hardly w hat th e natives w ould classify as such. Sahlins has a superb essay on western cosmology th at surveys the entire history of western philosophy (1996). Among many other w orks, particularly instruc­ tive are his discussions of the relation of structure to practice and event, both in

.So that in the first place , I put for a general!

inclination of all mankind, a perpetuall and restlesse desire o f Power after power , that ceaseth onely in Death. T

h o m as

H

o bbes

, Leviathan

T o discover the lineam ents o f the larger society in the concepts o f its biology is not altogether a “ M odern Syn th esis.” In Euro-A m erican society this integration has been goin g on in a particu lar dialectic w ay since the seventeenth century. Since H o b b es, at least, the com peti­ tive and acquisitive ch aracteristics o f W estern m an have been con fou n ded with N atu re, and the N atu re thus fashioned in the hum an im age h as been in turn reapplied to the explanation o f W estern m an. T he effect o f this dialectic has been to anchor the properties o f hum an social action , as we conceive them , in N atu re, and the law s o f N atu re in ou r conceptions o f hum an social action. H um an society is n atural, and natural societies are curiously hum an. A dam Smith produ ces a social version o f T h o m a s H o b b es, C h arles D arw in a n atural­ ized version o f A dam Sm ith; W illiam G rah am Sum ner thereupon reinvents D arw in a s society, and Edw ard O . W ilson reinvents Sum ner as nature. Since D arw in , the m ovem ent o f the con ceptual pendulum has accelerated. Every decade, it seem s, we are presented with a m ore refined notion o f m an as species, and a m ore refined species o f “ natural selec­ tio n ” a s man. In the opening ch apters o f Leviathan there is presented a picture o f m an as a self-m oving and self-directing m achine. C . B. M acp h erson ,

177

the abstract (1976) and in specific rela­ tion to the history of H aw aii and the H aw aiian reception of Captain Cook (1985; cf. the controversy raised in Obeyesekere 1992, and Sahlins 1995). A com prehensive collection of his essays appeared in 2000 w hile th e most recent is 2004.

w hose reading o f H o bb es an d explication o f “ possessive individualism ” I here follow very closely, describes the H o b b esian natural man a s an “ autom ated m achine,” having built into it “ equipm ent by which it alters its m otion in response to differences in the m aterial it uses, and to the im pact and even the expected im pact o f other m atter on it” (1 9 6 2 , p. 31). The m achine is part o f the in form ational system o f the w orld in which it m oves, a s nothing is present to its mind that w as not first present to its senses - “ there is no conception in a m an ’s m ind, which hath not at first, totally, or by p arts, been begotten by the o rg an s o f sense” (H ob b es, p art 1, chap. 1; all citatio n s o f Levia­ than are from the Everym an p aperb ack edition [1 9 5 0 ]). Lan guage introduces the potentiality o f error into this sensory epistem ology, as also a greater capacity for right m ovem ents, but it can n ot transcend the intrinsic values o f sensory experience. In ch apters 5 through 11, the general direction o f the m achine is indicated. “ Felicity o f this life,” H o b b es say s, “ consisteth not in the repose o f a mind s a tis fie d .. . . N o r can a m an any m ore live, w h ose D esires are at an e n d .. . . Felicity is a con tinuall progresse o f the desire” (chap. 11). T h e m achine acts to continue its ow n m orion by ap p ro ach in g things th at sustain that m otion and avoidin g things inim ical. M otion tow ard is “ desire” (or “ a p p e ­ tite” ) and its objects are “ g o o d .” M otion aw ay is “ aversion ” and its ob jects are “ evil.” Each hum an m achine “ endeavoureth to secure him­ self again st the evill he feares, an d procure the go o d he desireth ” (chap. 12). As the abstract positive and negative o f h um an action , these tw o m otion s are com prehensive. They exh aust all p articu lar m otivations which are ju st so

178

MARSHALL SAHLINS

many circum stantial m odalities o f m otion tow ard or m otion aw ay. Appetite with the opinion it will be satisfied is “ h op e” ; w ithout this opinion “ d e sp a ir.” Aversion with the anticipation o f hurt from the object is “ fear” ; with the hope o f resisting hurt, it is “ c o u ra g e .” And so for an ger, confidence, diffidence, indignation, benevolence, covetousn ess, p u sil­ lanim ity and m agnanim ity, liberality and p a r­ sim ony, kindness, lust or jealousy - they are products o f a single-m inded concern for on e’s own goo d . In the eighth chapter, how ever, H obbes states the relativity o f the calculus o f goo d . Insofar as it is social, it is a differential goo d . H obbes argu es th at the go o d men value is determ ined by w hatever other men already have. Virtue and w orth are only realizable as a differential success, as preem inence, and “ consisteth in co m p arison . For if all things were equally in all m en, nothing w ould be prized” (chap. 8). The success o f men in secur­ ing their ow n go o d thus depends on the strength o f their desires and their respective abilities. But then, the pursuit o f on e’s own good cannot rem ain at the level o f independent production. For the pow er o f one m an to obtain his ow n g o o d is op p o sed by the pow ers o f others. “ The pow er o f one m an resisteth and hindereth the effects o f the pow er o f an oth er” (cf. M acph erson 1 962, pp. 3 5 -6 ). There is an o p p o sitio n o f pow ers. And in the end, success turn s on the com petitive a p p ro ­ priation o f the pow ers o f others. A m an secures his ow n good to the extent he can harness the pow ers o f other men. There is a net transfer o f pow ers. T h e m eans are all such things as riches, repu tation , love, and fear. Riches joyned with liberality, is Power; because it procureth friends, and servants. . . . Reputa­ tion of Power, is Power; because it draweth with it the adhearence o f those that need pro tection .. . . Also, what quality soever maketh a man beloved, or feared o f many, or the repu­ tation o f such quality, is Power; because it is the means to have the assistance, and service of many. (chap. 10) M acph erson notes that in H o b b es’s schem e, men actually enter into a m arket for the exchange o f pow ers. M en find their w orth as

the price others will pay for the use o f their pow ers. It is in this m ode, as acquisition , that H o bb es put as the “ generall inclination o f all m ankind, a perpetuall and restlesse desire o f Power after pow er, that ceaseth onely in D eath ” (chap. 11). As all men are so inclined, no one m an can rest secure in his ow n pow ers w ithout engagin g “ by force, or w iles, to m aster the persons o f all m en he can , so long, till he see no other pow er great enough to endanger h im ” (chap. 13). Hence the fam ou s struggle am on g men in a state o f nature, the “ W arre” o f every m an again st every m an, enduring so long as they d o not agree to surrender their force to a C om m on Power (the State) that will “ keep them all in a w e .” W riting in an era o f transition to a devel­ oped m arket society, H o b b es reproduces the historical sequence a s a logic o f hum an nature. The expropriation o f m an by m an at which H obb es arrives in the end is, as M acpherson explain s, the theory o f action in a fully co m ­ petitive econom y. It differs from a mere stru g­ gle for preem inence, as w ould occur in transitional ph ases o f sim ple com m odity p ro ­ duction, because in the m odel o f the latter each m an has access to his ow n m eans o f livelihood and need not convey his pow ers to other men. Producers m ay m axim ize their ow n position in m arket exchange; they rem ain, how ever, independent proprietors and their lab or pow er as such is not a com m odity. The full m arket system also differs from exploitative structures such as feudalism and slavery, since in the latter con dition s, the rights to pow er, although they m ay yield a net transfer, are relatively fixed am on g the classes. N o one is free to convey his pow ers a s he will, for none can escape his definition a s a social being, defini­ tion that p resupposes his position in the circu­ lation o f pow ers. M en are slaves and serfs, others are lords and m asters, but the system is not com petitive such that it w ould be neces­ sary to struggle after m ore pow er just to con ­ serve the am ount one h as, or else lose out to those stronger in desire o r capacity. The full m arket system refers to the historical time when men d o becom e free to alienate their pow ers for a price, as som e are com pelled to d o because they lack the productive m eans to independently realize their ow n goo d . T h is is

FOLK DIALECTICS OF NATURE AND CULTURE

a very distinctive type o f society as well as a particular period o f history. It is m arked by w hat M acph erson styles “ p ossessive individu­ a lism .” Possessive individualism entails the unique notion - coun terpart to the liberation from feudal relation s - that men ow n their ow n bodies, the use o f which they have both the freedom and necessity to sell to those w ho control their ow n capital. (It w as M arx , o f course, w ho penetrated the inequities o f this exchange, that is, the net tran sfer, since the value produced by lab or pow er is greater than its price.) In such a con dition , every man con ­ fronts every m an a s an ow ner. Indeed, society itself is generated through the acts o f exchange by which each seeks the greatest possible ben­ efits in oth ers’ pow ers at the least possible cost to his own. It w as, M acp h erson explain s, a conception o f the individual as essentially the proprietor of his own person or capacities, owing nothing to society for them. The indi­ vidual was seen neither as a moral whole, nor as part o f a larger social whole, but as owner of himself. The relation of ownership, having become for more and more men the critically important relation in determining their actual freedom and actual prospect of realizing their full potentialities, w as read back into the nature of the individual. . . . Society becomes a lot of free individuals related to each other as proprietors o f their own capacities and of what they have acquired by their exercisc.

Society consists o f relations o f exchange between proprietors. (1962, p. 3; italics added) Social scientists will recognize in this descrip­ tion the “ utilitarian ism ” that has beset their ow n disciplines since Spencer and before (cf. P arson s 1968; Sahlins 1976). It is precisely a perspective in which the individual is seen “ neither a s a m oral w hole, nor a s part o f a larger social w hole, but a s ow ner o f him self.” In the social sciences, as in sociobiology, the hom ebred econom izing o f the m arket place is then all to o easily tran spo sed from the analysis o f cap italist society to the explication o f society tout court . The analytic place thus left to the social fact has been well described by Louis D um ont:

179

In modern society . . . the Human Being is regarded as the indivisible, “ elementary” man, both a biological being and a thinking subject. Each particular man in a sense incarnates the whole of mankind. He is the measure o f all things (in a full and novel sense). The kingdom of ends coincides with cach m an’s legitimate ends, so the values are turned upside down. What is still called “ society” is the means, the life of each man is the end. Ontoiogically, the society no longer exists, it is no more than an irreducible datum, which must in no way thwart the demands o f liberty and equality. O f course, the above is a description o f values, a view o f m in d .. . . A society as conceived by individualism has never existed anywhere for the reason we have given, namely, that the individual lives on social ideas. (1970, pp. 9-10) I underscore D u m on t’s ob servation s on the indivisibility o f the hum an being in the per­ spective o f the sociological utilitarianism : man a s a thinking subject is a lso the sam e m an as a biological being. H ence society m ay be derived from the rational action o f individuals seeking to satisfy their needs - a project in which “ th ough t” serves m erely as the m eans and the representation o f inherent ends. S ocio­ biology operates on exactly the sam e prem ise. H o bb es provided the original basis for this subordin ation o f the sym bolic to the natural by situating the society he knew in the state o f nature. M an w as seen as a w o lf to m an. Again one can say that the objective o f socio b io lo ­ gists is very sim ilar so far as it concerns human society. But it goes further. Since they w ould now extend the sam e folk conception o f c a p i­ talism to the anim al kingdom as a w hole, for sociob iologists it is a lso true that the w olf is a m an to other w olves. A ctually, how ever, I com press a long cycle o f reciprocal interpreta­ tions o f nature and culture th at has been ch ar­ acteristic o f the W estern co n sciousn ess, both as science and as ideology. I can briefly describe this cycle by m aking tw o further points. First, it is clear that the H o b b esian vision o f man in a natural state is the origin myth o f W estern capitalism . In m odern social practice, the story o f G enesis pales by com parison . Yet it is a lso clear that in this co m p ariso n , and indeed in com parison with the origin m yths of

180

MARSHALL SAHLINS

all other societies, the H o bb esian myth has a very peculiar structure, one that continues to attend ou r un derstandin gs o f ourselves. So far a s I am aw are, we are the only society on earth that thinks o f itself as having risen from sa v ­ agery, identified with a ruthless nature. Every­ one else believes they are descended from god s. Even if these g o d s have natural representa­ tions, they nonetheless have supern atural a t­ tributes. Ju d g in g from social behavior, this co n trast m ay well be a fair statem ent o f the differences between ourselves and the rest o f the w orld. In any case we m ake both a folklore and a science o f ou r brutish origin s, som etim es with precious little to distinguish between them . And just a s H o bb es believed that the institution o f society or the C om m on w ealth did not abolish the nature o f m an as w o lf to other men but m erely perm itted its expression in relative safety, so we continue to believe in the savage within us - o f which we are slightly ash am ed. At an earlier period it w as H om o economicus , with a natural propensity to truck and barter, an idea that rationalized the devel­ opin g cap italist society to itself. It took but tw o centuries to evolve another species, H om o belltcosus , or so one m ight classify that co n ­ tentious ape p opu larized by Ardrey and other recent w riters. N o w com es sociobiology, and with it apparen tly a reversion to econom ic type, program m ed in the natural propensity o f I)N A to m axim ize itself at the expense o f whom it m ay concern. Hence the respon se by men o f the Left becom es intelligible, a s does the interest o f the public at large. W hat is inscribed in the theory o f sociobiology is the entrenched ideology o f W estern society: the assuran ce o f its n atu ral­ ness, and the claim o f its inevitability. The second poin t concerns the ideological dialectic to w'hich I previously alluded. Since the seventeenth century we seem to have been caugh t up in this vicious cycle, alternately applyin g the m odel o f cap italist society to the anim al kin gdom , then reapplying this bourgeoisfied anim al kingdom to the interpretation o f hum an society. M y intent in ad o p tin g the M acp h erson read in g o f H o bb es w as just to im ply that m ost o f the elem ents and stages o f the biological theory o f natural selection from differential success to the com petitive

struggle to reproduce on e’s stock and the transfer o f pow ers - already existed in the Leviathan. As a critic o f this capitalist con cep­ tion, it w as left to M arx to discern its realiza­ tion in D arw inian theory. In a letter to Engels, M arx w rote: It is remarkable how Darwin recognizes among beasts and plants his English society with its division o f labour [read, diversifica­ tion], competition, opening up o f new markets [niches], “ inventions* [variations], and the M althusian “ struggle for existence.” It is H obbes’s “ helium omnium contra om nes,” and one is reminded o f Hegel’s Phenomeno­ logy where civil society is described as a “ spir­ itual animal kingdom ,” while in Darwin the animal kingdom figures as civil society. (M arx in Schmidt 1971, p. 46) T he sam e point w as to be m ade later by H ofstadter: A parallel can be drawn between the patterns o f natural selection and classical economics, suggesting that Darwinism involved an addi­ tion to the vocabulary rather than to the sub­ stance o f conventional economic theory. Both assumed the fundamentally self-interested animal pursuing, in the classical pattern, plea­ sure or, in the Darwinian pattern, survival. Both assumed the normality of competition in the exercise o f the hedonistic, or survival, impulse; and in both it w as the “ fittest,” usually in a eulogistic sense, who survived or prospered - either the organism most satisfac­ torily adapted to his environment, or the most efficient and economic producer, the most frugal and temperate worker. (1959, p. 144) In a letter to L avro v, Engels described the ensuing dialectical return, the representation o f culture to itself in the form o f a capitalist nature: The whole Darwinist teaching o f the struggle for existence is simply a transference from society to living nature o f H obbes’s doctrine o f “ helium omnium contra om nes” and o f the bourgeois-economic doctrine of competition together with M althus’s theory of population. When this conjurer’s trick has been per­ formed . . . the same theories are transferred back again from organic nature into history and now it is claimed that their validity as

FOLK DIALECTICS OF NATURE AND CULTURE

eternal laws of human society has been proved. (Engels in Schmidt 1971, p. 47) It m ight be noted that D arw in w as not a lto ­ gether happy with this reciprocal reflection of the anim al kingdom as his ow n English society. “ 1 have received in a M anchester new spaper rather a good sq u ib ,” he w rote to Sir C h arles Lyell, “ show ing that I have proved 'm ight is right,' and therefore that N ap o leo n is right, and every cheating tradesm an is also righ t” (cited in H ofstadter 1959, p. 85). But no such reserve w ould inhibit W illiam G rah am Sum ner - to take the ou tstan din g A m erican exam ple - from transferring the D arw in ian teaching back to its original social source. “ The truth is that the social order is fixed by law s o f nature precisely an alogou s to those o f the physical o rd e r” (Sum ner 1934, vol. 2, p. 107). H o fstad ter succinctly su m m a­ rizes Sum n er’s inspiration: In the Spencerian intellectual atmosphere of the 1870s and 1880s it w as natural for con­ servatives to see the economic contest in com ­ petitive society as a reflection o f the struggle in the animal world. It w as easy to argue by analogy from natural selection of fitter organ­ isms to social selection o f fitter men, from organic forms with superior adaptability to citizens with a greater store of economic virtues. . . . The progress o f civilization, according to Sumner, depends on the selection process; and that in turn depends upon the workings of unrestricted competition. C om ­ petition is a law of nature which “ can no more be done away with than gravitation,” and which men can ignore only to their sorrow. (Hofstadter 1959, p. 57) O ne aspect o f Sum ner’s biologism deserves special com m ent. It concerns the m otivation which Sum ner frequently alleged for the accu ­ m ulation o f wealth in a ruthless com petitive struggle. This is exactly the sam e m otivation adduced by sociobiology for the parallel stru g­ gle in nature - “ inheritance” (by the offsp rin g o f the fittest). T he double service o f the term is not unusual. From the late M iddle A ges o n w ard , W estern society has gone to con sider­ able effort to encode its econom ic activity within a pervasive m etaphor o f im provem ent o f the stock. T erm s for anim al reproduction

181

have been ap prop riated for econom ic catego­ ries and vice versa, at first figuratively, but then so consistently that m etaph or dies and it becom es im possible to distinguish the original reference from the derived. T h e peculiarity of a native category that refers interchangeably to the social reproduction o f econom ic g o o d s and the natural reproduction o f anim ate beings then g o e s unnoticed, banished from co n sciou s­ ness a s well a s m em ory. On the con trary, the category becom es a basis for scientific or po p u lar reflections on the essential identity o f the tw o processes. T hese reflections acco rd­ ingly take the form o f a folk etym ology. They recapitulate, for exam ple, the derivation o f the English term s “ c a p ita l” and “ ch attel” from an older “ cattle,” which precisely as the m ovable and increasable “ livestock” w as distinguished from the dead stock o f fixed farm equipm ent. (Indeed the com m on origin o f the concepts o f tran sactable w ealth and cattle in the IndoEuropean peku, together with the appearance o f a cogn ate category o f pasu viru in A vestan including men and their dom estic an im als, su g­ gests a prim itive integration o f the econom ic, the social and the natural; m odern usage w ould merely represent cognitive h om ology [of. Benveniste 1969; and relevant entries o f the O ED \.) It is the sam e with “ inh eritance,” which ini­ tially referred to the continuity o f g o o d s over generations o f people, only to denote at a later date the continuity o f the generational “ sto c k ” itself. W. Ci. Sum ner w as thus em pow ered by the folk w isdom to find cau se for the econom ic com petition over resources in a genetic tran s­ m ission - just as O . W ilson w ould later describe the natural process o f genetic tran s­ m ission a s a struggle for resources: The socialist assails particularly the institution o f bequest or hereditary p roperty.. . . The right o f bequest rests on no other grounds than those o f expediency. The love of children is the strongest motive to frugality and to the accumulation o f capital. The state guarantees the power of bequest only because it thereby encourages the accumulation of capital on which the welfare of society depends . . . he­ reditary wealth transmitted from generation to generation is the strongest instrument by which we keep up a steadily advancing civili­ zation. (Sumner 1934, vol. 2, pp. 112-13)

182

MARSHALL SAHLINS

We seem unable to escape from this p erp et­ ual m ovem ent, back and forth between the culturalization o f nature and the n atu raliza­ tion o f culture. It frustrates our understanding at once o f society and o f the organic w orld. In the social sciences we exh aust our own sym ­ bolic cap acities in an endless reproduction o f utilitarian theorizing, som e o f it econom ic, som e ecologic. In the natural sciences, it is the vulgar and scientific sociobiologies. All these efforts taken together represent the m odern encom passm ent o f the sciences, both o f culture and o f life, by the dom inant ideology o f p o s­ sessive individualism . The net effect is a curious form o f totem ism o f which scientific sociobiology is the latest incarnation. For if totem ism is, as L,evi-Strauss says, the explication o f differences between hum an grou p s by reference to the distinctions between natural species, such that clan A is related to and distinct from clan B as the eagle hawk is to the crow , then sociobiology m erits classification a s the highest form o f the totem ic philosophy. For its sophistication and advance over the prim itive varieties, both in the W est and a b ro ad , it does seem to merit a special nam e, one in keeping with its ow n synthetic pretensions as the latest branch o f the sciences and the principal hope o f civilization. G ive it its due: sociobiology is a Scientific T otem ism .

But with all respects to the pensee sauvage , this reliance on the deep structure o f W estern thought, with its assim ilation o f the reproduc­ tion o f people to the reproduction o f g o o d s as a kinship o f substan ce, can n ot do for the science to which we now aspire. The confusion o f categories is to o im m oderate. It puts us all, biological and social scientists alike, in the state know'n all to o well to the practitioners o f totem ism : o f m ess and “ d irt,” as M ary D ouglas has taught us, o f pollution and tabu. Beyond all the politics, it is o f course this descent into the kingdom o f tabu that ultim ately m akes so cio ­ biology so fascinating. But we pay a heavy penalty in know ledge for the distinctions we are forced to surrender. “ The m ost serious harm to science that I see in the present fashion o f applying ethnological term s to an im als,” Susanne Lan ger w rites, “ is that - odd as it m ay seem - it is really based on the assum ption that the tw o studies, ethnology and w hat is called ‘eth ology’ . . . will never becom e true integral parts o f biological science. If they should ever do so, the use o f w ords literally in one context and figuratively in another w ould cause h av oc” (1 9 7 1 , p. 328). Yet we stan d to lose even m ore than our science. We should have to aban don all understanding o f the hum an w orld as m ean­ ingfully constituted, and so the one best hope o f know ing ourselves.

REFERENCES

and W. S. D illon. W ashington, D C : Sm ith so­ nian Institution Press, pp. 3 1 4 -3 2 . M acph erson, C . B. 1962. The political theory of possessive individualism. London: O x fo rd University Press. Parsons, T alcott. 1 9 6 8 . The structure o f social action . 2 vols. N ew Y ork: The Free Press. Sahlins, M arsh all. 1976. Culture and practical reason. C h icago: University o f C h icago Press. Schm idt, Alfred. 1971. The concept o f nature in Marx. London: N LB . Sum ner, W illiam G rah am . 1934. E ssays o f William Graham Sumner. 2 vols. Ed. A. G. Keller, and M . R. D avie. N ew H aven: Yale University Press.

Benveniste, F’mile. 1 969. Le vocahulaire des

institutions indo-europeennes; vol. 1: Economiey parentey societe. Paris: F2ditions de M inuit. D um ont, Louis. 1970. H om o Hierarchicus. C h icago: University o f C h icago Press. H o b b es, T h o m as. 1950 [1651]. Leviathan. N ew Y ork: E. P. Dutton. H o fstad ter, R ichard. 1959. Social Darwinism in American thought. Revised edition. N ew Y ork: Braziller. Langer, Susanne K. 1971. The great shift: Instinct to intuition. In Man and beast: C om ­ parative social behavior , ed. J . F. E isenberg,

16

Land Animals, Pure and Impure Mary Douglas

Mary Douglas (1921-2007) was a most distinguished British anthropologist w ho trained at the University of Oxford and taught at University College London and elsew here. She is the author of many works, o f which the key ones w ith respect to religion include Purity a n d D an ger (1966), N atu ral Symbols (1970), and the essays in Im plicit M eaning s (1975, cf. 1996). She conducted research among the Lele of Central Africa, on contem porary British and Am erican society, and on the Old Testam ent (1993, 1999). Her w ork has been the subject of an extensive study by Fardon (1999). The present essay is taken from a recent interpretation of the biblical book of Leviticus. As Douglas explains, its origins lie in her much earlier analysis of "The Abom inations of Leviticus" in Purity an d Danger. That w ork, in turn, owed its im m ediate inspiration to Stein­ er's essay on taboo (1956), which in some respects paralleled Durkheim on

the sacred and Levi-Strauss on totemism in showing its subject as relationally rather than substantively constituted. In an enormously influential analysis Douglas argued that systems of classifi­ cation and the anom alies they inevita­ bly produce are central objects of religious concern in any society. "Dirt," she famously stated, is simply "m atter out of place" and hence alw ays relative to a particular system of classification. Rather than explaining systems of purity and taboos in terms of biological germ theory (as though, som ehow, the ancient Israelites had known about th e dangers of trichinosis in uncooked pork), Douglas reverses the prevalent argum ent and sees modern concerns w ith hygiene as another instance of symbolic ordering. Purity an d D anger thus helped develop a radical shift in the w ay th at symbolic systems of small- and large-scale societ­ ies could be compared to one another, and was also inspirational for establish-

From Mary Douglas, “ Preface” and “ Land Animals, Pure and Impure,” in Leviticus as Literature (Oxford: Oxford University Press, 1999), pp. v-viii, 134-51. Reprinted by permission of Oxford University Press. Abridged.

184

M ARY DOUGLAS

ing the significance of the m etaphoric properties of th e body and food w ithin cultural systems. Douglas provides a forthright critique of the al!-too preva­ lent perspective she calls (after W illiam James) "m edical m aterialism ," which is not distinct from the ideology that Sahlins critiques in the preceding essay. This form of argum ent w hich reduces symbol to function is evident in Harris's popular explanations for the refusal to eat pigs in th e M iddle East or cows in South Asia (1985), and subject to lively refutation in Sahlins (1976). However, Douglas differs from Sahlins in that w hile she has embraced new directions in structuralism and phenom enology, she remains tru e to her Durkheim ian roots in seeing symbolic practice as em erging from and addressing social boundaries and relations. In Leviticus as Literature Douglas revisits the question of symbolic classifi­ cation in th e Bible. Her account is no longer m erely a very clever piece of analysis but an interpretation th at is grounded in a deep and em ergent

. . . T o study the book o f Leviticus a s an an th rop ologist h as been a project very dear to my heart. It seem ed far beyond my reach. Yet not to d o it w ould be to leave dan glin g a num ber o f th reads from early w ork. Let me explain som e things ab o u t my training which have influenced my attitude to the Bible. Y oun g an th rop ologists in O x fo rd in the late 1940s and 1 9 5 0 s were heirs to an old debate about hum an ration ality, a debate p rovoked by the experience o f science and biased by the e xp eri­ ence o f em pire. N ineteenth-century ration alists centred on w h at they thought o f a s the n atives’ intellectual p roblem s. G ro ss superstition s, naive m agic, and im m oral g o d s, were explained by reference to m oral evolutionism . The mind o f the prim itive in aeon s past had been h am ­ pered by illogical m ental habits and proneness to letting em otion s govern reason , an d the sam e h an d icaps were thought to afflict present-

understanding of th e religion of ancient Judaism and the styles of reasoning and writing used by the authors of the d if­ ferent books of the Torah (Pentateuch). Remaining rigorously opposed to piece­ meal m oralizing and m aterialist expla­ nations of individual food taboos, in this chapter she explains the rules of im pu­ rity in terms of the logic of the whole. The argum ent now attends to the posi­ tive meaning of ritual purity and the rules of sacrifice rather than the osten­ sible disgust or puzzlem ent at anom a­ lies. It also draws on the model developed elsew here in the book of a system of analogies betw een M ount Sinai, the tabernacle, anim al offerings, and the bodies of the people of Israel. Some­ w hat to Douglas's own puzzlem ent, and much to her credit, the interpretation moves beyond her inclination to seek direct correspondences betw een the social and symbolic orders. For other w ork on symbolic classification see Needham, ed. (1973); for an additional anthropological foray into the Bible, Leach (1969).

day back w ard peoples. H ow ever, in reaction, for the students o f my generation the m ain text w as Evans-Pritchard’s Witchcraft, Oracles and M agic (1 937). From this we learnt that people from alien traditio n s, trusting in their g od s and an cestors and fearing their w itches, were every bit as logical as we (or ju st as illogical). It is actually no m ore “ lo gical” to believe in a divinely created m oral universe than to believe in an am o ral self-generating universe. F o u n d a­ tional beliefs stan d beyond the operation s o f logic. O ur researches were fram ed by an inter­ est in the m oral con stru ction o f the universe and the nature o f belief. In those d ays it w as ax io m atic for an th ro­ p ologists th at, how ever peculiar they m ight seem to us, the strange beliefs o f a foreign tradition m ake sense. E xp lan atio n s o f other m inds based on m ystery, m ystique, native cre­ dulity or m ysticism , were out. M oral evolution

LAND ANIMALS, PURE AND IMPURE

w as replaced by a dow n-to-earth ap p ro ach to alternative w ays o f living an d dying. We took on a hardy skepticism and a nuts-and-bolts dem an d for evidence. The point ab o u t doin g fieldw ork w as to learn how a w orld-view w as ad ap ted to w hat the people were trying to achieve, especially to w hat they were d oin g to w ard s living together in society. Hence our attention to ritual and sym bolism . R ain rites, for exam p le, w ould be a collective act o f affir­ m ation. The rite did not attem pt to prove the p riests’ control o f m eteorology, it w as done to affirm publicly the m oral asp ect o f the natural order. Spectacular cerem onials to ap p ease the g o d s w ere also perform ed for the sake o f influ­ encing each other’s m inds. I w ould never have felt im pelled to attem pt an an th rop ological reading o f Leviticus if during A frican fieldw ork I had not been co n ­ fronted by local dietary rules, and so thought o f lookin g up the passage in chapter 11 on the forbidden an im als. I actually cited Leviticus and the parallel p assage in D euteronom y in my “ A nim als in Lele religious sy m b olism ” (1 9 5 7 ). W hat 1 w rote ten years later ab o u t uncleanness and pollution in Purity and D anger (1 9 6 6 ) w as driven by fieldw ork experience, stiffened by train ing in O x fo rd an th ropology and enriched with som e reading ab o u t the psychology o f perception. But before lookin g up those b a f­ fling ch apters, 1 had never read the Bible, either at school or at university or subsequently. When I cam e eventually to read the scholarly com m en taries on the M o saic dietary law s 1 w as surprised to find so much disagreem ent on such an im portan t subject. T h ough with som e m inor variation s sch olars alm o st unanim ously a sso c i­ ated the forbidden an im als with unpleasant ch aracteristics, there w as no agreem ent an d no satisfacto ry explanation either in the b o o k or ou tsid e it ab o u t why each particular species should have been selected an d not others which m ight equally be abom in ated. Reflecting on these an im als I w as draw n to focus on the class o f unclassifiable things. T he forbidden land anim als were certainly described a s such a class, and I extended it with som e confidence to w ater creatures and speculatively to those in the air that could not be identified. I prop o sed a theory o f an om aly, a universal feeling o f disquiet (even o f disgust) on c o n ­

185

frontation with unclassifiables. T ak in g the Levitical classification system a s it revealed itself, the said abom in able species failed to show the taxon o m ic requirem ents o f inhabit­ ants o f the three environm ental classes, land, air, w ater, and the abom in ability o f species that “ go upon the belly” in all environm ents w ent by the sam e rule: the forbidden anim als were species that escaped being classified. C onsistently with the m ain th rust o f social an th ropology o f my period, the argum ent explained abom in ability, but denied m agicality and favoured the rationality o f the M osaic dietary code. It w as gratifyin g to find that som e Bible sch olars accepted the idea that the puzzles o f the abom in able an im als in Leviticus and D euteronom y could be laid to rest, the proh ibitions being p art o f the process o f tidying up the classification s o f the environm ent (see Levines FPS Commentary , Leviticus (1 9 8 9 ), 2 4 3 ). But a puzzle rem ained. The central argum ent o f Purity and Danger w as that classification s are not otiose. They do som ething, they are necessary in organization . The pollution theory th at I have seen develop over the last thirty years sh ow s th at where lines o f abom inability are draw n heavy stakes are at issue. The classification o f the universe is part and parcel o f social organ izatio n , and the cate­ gories are useful in defining w ho can be adm it­ ted w here, and w ho com es first and w ho com es second or nowhere at all. T h is w orks so effec­ tively elsew here that I w as im plicitly w aiting for it to be found true o f biblical pollution (see my “ Sacred co n tagio n ” (1 9 9 6 )). It applies well enough, in fairly ob viou s w ay s, for the cult o f the tabernacle and the dignity o f the priest­ h ood, but for the organ ization o f society the doctrine o f pollution did nothing except draw a boun dary round the people o f Israel again st ou tsiders. N oth ing h appen s at the level of action to explain the selection o f forbidden anim als. A gainst everything I believe, the co g ­ nitive schem e which left these creatures unclas­ sified hung in the air uselessly. If chapter 11 of Leviticus w as a case for pollution theory the classifyin g o f the anim als should correspon d to som e im portant classifying for the internal organ ization o f society. But the m ore that p o l­ lution theory developed, an d the m ore that pollution w as seen as the vehicle o f accusation s

186

M ARY DOUGLAS

and d ow n gradin gs, the m ore I w as bound to acknow ledge that it does not apply to the m ost fam ou s instance o f the W estern trad itio n , the Pentateuch. All o f this volum e is an attem pt to explain why. General pollution theory still stan d s, but its application to the Bible is limited. The forbidden an im als turn out to have a much m ore interesting role than ever 1 im agined. [.■■] Behold, I establish my covenant with you and your descendants after you, and with every living creature that is with you, the birds, the cattle, and every beast o f the earth with you, as many as came out of the ark. (Gen 9: 9-10) For thou lovest all things that exist, and hast loathing for none of the things which thou hast made, for thou wouldst not have made anything if thou hadst hated it. (Wisdom 11:24) G od met his people on M oun t Sinai and con ­ tinues to meet them in the tabernacle. In sac­ rifice the body o f the sacrificial anim al becom es another m icrocosm in its ow n right, co rre­ spon din g to the tabernacle and the holy m ou n ­ tain. Then the sequence o f cultic law s is interrupted by the narrative in ch apters 8 -1 0 . W hen the law -giving is resum ed it develops a different bodily m icrocosm . This time the body o f the w orshipper is m ade an alogou s to the sanctuary and the altar. W hatever will render the altar im pure will d o the sam e for the Isra­ elite’s body. The law s o f im purity sketch out the parallel in m eticulous detail over ch apters 1 1 -1 5 . The anim al that is taken into the body by eating corresp on d s to that which is offered on the altar by fire; w hat is disallow ed for the one is disallow ed for the other; w hat harm s the one harm s the other. O ne thing th at the book never say s is that it is bad for the health o f the body to eat any o f the forbidden anim als.

Land A n im als u n der th e C o ven a n t C h apter 11 is probably the best know n in Leviticus because it deals with the M osaic

dietary law s. It has been taken to imply that the forbidden anim al m eats are abom in able, detestable, or unedifying in one way or another. T ak in g account o f the full context, which is the rest o f the Pentateuch, it w ould be difficult to overlook one biblical principle: G o d is co m ­ passionate for all living things; not only to the hum ans, he is go o d to all his creatures (Psalm 145: 8 -9 ). S o if he him self does not detest them , why should he tell hum ans to detest any o f his an im als? This is a serious and central doctrinal problem . T w o kinds o f covenant are the basis o f chapter 11. First the covenant with N o ah and his descendants in which G o d said he w ould never again punish the land and the living things on it for the evil things done by m ankind, and m ade the rain bow its sign. It is em ph ati­ cally also a covenant with the anim als: “ This is the sign o f the covenant which I m ake between me and you and every living creature that is with you, for all future gen erations” (Gen 9: 1 2 ) . . . I will rem em ber my covenant which is between me and you and every living creature o f all flesh . . .” (Gen 9: 15, repeated in vv. 16 and 17). A few verses earlier in the sam e chapter G od has required a reckoning for the life-blood o f hum ans. “ Only you shall not eat flesh with its life, that is, its blood. For your life-blood I will surely require a reckoning; o f every beast I will require it and o f m an ” (Gen 9: 4,5 ). In G enesis G od gave man dom inion over anim als. R obert M u rray 1 has argued persuasively that dom inion for G enesis alw ays entails responsibility. Leviticus presents the further im plications o f hum an dom inion over anim als. A one-sided pledge from G o d cannot quite be called a covenant: the anim als are not bound by any counter-obligation s, unless by a stretch o f the im agin ation the com m and to them at the creation to g o forth and m ultiply coun ts a s such. Later, the covenant with A braham is a prom ise o f fertility to his descendants. It does not m ention the an im als, but extravagan tly it say s that his descendants will be a s innum era­ ble as the dust (Gen 13: 16). It echoes the blessing o f G enesis to N o a h and his sons, “ Be fruitful and m ultiply, and fill the earth ” (Gen 9: 1), and “ Be fruitful an d m ultiply, bring forth abundantly on the earth and m ultiply in

LAND ANIMALS, PURE AND IMPURE

it” (Gen 9: 7), the very w ords used for his blessing on the creatures o f the w ater and the air after they had been created (Gen 1: 22). There is no d ou bt that this G o d is concerned with fertility, and that his prom ise is linked with their obedience. The covenant with M oses on Sinai is the explicit assertion o f G o d ’ s overlordsh ip over the people o f Israel and their livestock. It spe­ cifically includes the servan ts and the cattle in the sab b ath observance (E xod 2 0 : 8). From h ouseholder to children, to servants, to cattle, the an im als com e under the lines o f authority draw n by the Sinai covenant. Sab bath o b ser­ vance only affects the w ork an im als, the ox that treads the husks o ff the grain , that draw s the cart, that turns the w ater-w heel.2 E xod u s also m akes the point strongly by requiring that the m ale first-born o f the dom estic anim als be offered to the Lord just as the first-born o f hum ans. “ C on secrate to me all the first-born; w hatever is the first to open the w om b am ong the people o f Israel, both o f m an and o f beast, is m ine” (E xod 13: 2). “ Y ou shall set ap art to the Lord all that first open s the w om b. All the firstlings o f your cattle that are m ales shall be the L o rd ’s ” (F'xod 13: 12). The rule for land anim als which alw ays soun d s so com plicated is quite sim ple when the coven an t is seen to be its guiding principle. G od is the feudal Lord. From this it follow s that no one is allow ed to harm G o d ’s people or use G o d ’s things, nor m ust his follow ers harm each other, or harm the other living beings on his territory w ithout his express perm ission. T h is he gives for the killing o f herd anim als in sacrifice, and use o f their carcasses. The question o f whether they d o or do not com e under the covenant is p aram ou n t. Leviti­ cu s divides land an im als into tw o categories, first, the herds and flocks which share the lives o f their ow ners, travel with them , an d provide their sustenance, and secon d, all the rest. The pure an im als com e under the term s o f the co v ­ enant o f their m asters, an d their treatm ent is strictly regulated. T he feudal relationship extends from G o d to his people and to their livestock. T he teaching ab o u t the sanctity o f blood derives from this feudal relationship. G od p ro ­ tects the people o f Israel, his rites give them

187

covering, sacrifice is the m eans he has given to them for expiation. Sacrifice protects them from the consequences o f their ow n behaviour, even from his just anger. They are never, ever, allow ed to eat blood, but he has given them the right to consecrate the lives o f their herd an im als, to use their blood to m ake atonem ent to him for their sins, and to eat the blood-free flesh for their ow n nourishm ent (Lev 17: 11). T h is solem n injunction teaches the sanctity o f life (the life is in the blood). In religious term s, the m osaic dietary code is an invitation to Israel to join in the divine w ork o f creation by living a life that honours the w ay G o d m ade the w orld and the covenants Ciod has m ade with his people.

Th e T w o Te xts The tw o texts in Leviticus and D euteronom y start by running in close parallel. They give a perfectly logical classification which echoes the opening ch apters 1-7 on sacrifice with a description o f the dom esticated rum inants o f their herds, cattle, sheep, g o a ts, which m ay be consecrated for offering on the altar. Then follow s in both texts a careful set o f rules to discrim inate near-m isses, can didates for entry into the class o f dom estic rum inants which fail because they show one but not both the required criteria. M arch in g in step, the tw o texts say that the unclean anim als are only “ unclean for y o u ” . Because the dietary rules ab o u t land anim als derive from the covenant, they only apply to the people o f Israel. D euteronom y seem s to say by its prefatory rem ark that abom in able things are the things forbidden as unclean in 14: 7 -8 . The equation o f unclean with abom in able in D euteronom y is the source o f the idea that the forbidden anim als have som e detestable characteristic, the focus o f so m uch scholarly ingenuity. But in Leviticus the unclean anim als are not abom inable. T he m icrocosm is based on the body o f anyone o f the congregation o f Israel ab o u t to take nourishm ent; the body is equivalent to the altar and so is his hospitable bo ard round which he gath ers his fam ily and friends. Way back in the time o f the Leviticus w riter the

188 Table 1

M ARY DOUGLAS

Land animals, pure and impure, Deuteronomy 14: 3-8 and Leviticus 11: 2-8

Leviticus

Deuteronomy

2. These are the living things which you may eat among all the beasts that are on the earth. 3. Whatever parts the hoof and is clovenfooted and chews the cud, among the animals, you may eat. 4. Nevertheless among those that chew the cud or part the hoof, you shall not eat these: The camel, because it chews the cud but does not part the hoof, is unclean to you. 5. And the rock badger, because it chews the cud but does not part the hoof, is unclean to you. 6. And the hare, because it chews the cud but docs not part the hoof, is unclean to you. 7. And the swine, because it parts the hoof and is cloven-footed but does not chew the cud, is unclean to you. 8. Of their flesh you shall not eat and their carcasses you shall not touch.

3. You shall not eat any abominable thing. 4. These are the animals you may eat: the ox, the sheep, the goat, 5. the hart, the gazelle, the roebuck, the wild goat, the ibex, the antelope, and the mountain-sheep. 6. Every animal that parts the hoof and has the hoof cloven in two, and chews the cud, among the animals, you may eat. 7. Yet of those that chew the cud or have the foot cloven you shall not eat these: The camel, the hare and the rock badger, because they chew the cud but do not part the hoof, are unclean for you. 8. And the swine, because it parts the hoof but docs not chew the cud, is unclean for you. Their flesh you shall not eat and their carcasses you shall not touch.

body w as already the an alogue o f the altar. N o t a secular an alogu e, for in a total religious system (such a s that o f Leviticus) the w ord secular d oes not have much m eaning. The table, and all w ho eat at it, and everything that has been co ok ed for them to eat, are under the sam e law o f h olin ess.3 Body for altar, altar for body, the rules which protect the purity o f the tabernacle are paralleled by rules which protect the w orshipper. W hat he can eat w ithout con ­ tracting im purity and w hat can be offered to G od in sacrifice are the sam e. An interesting difference between the texts o f D euteronom y 14 and Leviticus 11 is that the Leviticus open in g, with its reference to living beings on the earth (Lev 11: 2 ), recalls the accoun t o f the creation in G en esis, “ Let the earth bring forth living creatures accord in g to their kinds: cattle and creeping things and beasts o f the earth, accordin g to their k in d s” (Gen 1: 24). T he opening verses o f Leviticus II are only the beginning o f a larger survey o f land an im als (the beasts o f the earth) which is not featured in D euteronom y. W hen Leviticus has listed the different types o f land an im als that are im pure and m ust not be eaten, look round and see w hat is left -

nothing! All these land an im als are either clean o r unclean, pure or im pure. Now' that verse 26 is taken to refer to an im als with solid h oofs, we can use the follow ing com prehensive typo­ logy o f the land anim als: (i) (ii) (iii) (iv) (v) (vi)

an im als o f the flocks and herds, rum i­ nants, split hooves; rum inants w ithout split h oofs, e.g. cam el, rock badger, hare; non-rum inant with split h oofs, pig; solid hooves, e.g. a sses, horses; p aw s, e.g. lion, civet cat, dog, hyena; list o f eight land an im als that g o on their belly: the m ole, the m ouse, the great lizards, the gecko, the land crocodile, the lizard, the san d lizard, and the cham eleon.

The zoological criteria are good enough to m ake an exh austive list o f land anim als. By the end o f the ch apter everything living on the land has been included: goin g on h oofs, goin g on p aw s, gliding on the belly - w hat else is there? Everything h as been accounted f o r . . . . The first set are under the covenant, and clean. All other land an im als, excludin g the first set, are unclean or im pure, and their dead

189

LAND ANIMALS, PURE AND IMPURE

Table 2

Living beings on rhe earth humans (bipeds)

anim als (quadrupeds)

.I

I

rum inant

(i) split hoofs (1-4)

(iii) split hoofs (7)

.

non-rum inant

(ii) no split hoofs (4, 5, 6)

(iv) solid hoofs (26)

bodies not ro be touched or earen. Deuteronom y m akes a point o f allow in g the killing and earing o f wild co u n terp arts o f dom estic herds (D eut 13: 5), o f which Leviticus d oes not speak. W ere Leviticus explicitly to permit secular slaughter, it w ould underm ine the co v­ enant basis o f the Levitical rules constraining hum ans from eating an im als that they have nor reared. A sim ilar silence covers w hat is to be done with blem ished an im als from their flocks and herds which are not allow ed to be sacri­ ficed. Presum ably they are quietly eaten. M an y four-footed an im als are not specifi­ cally m entioned in the catalogu e; for instance, rodents such as squirrel, rab b it, rat, cat, do nor need to be listed since they are covered by the rules for anim als thar go on their paw s. Finally, not on the chart ab o v e, there are other landdw ellers indicated a s “ sw arm in g things that g o on the belly, g o on all fo urs, or with m any feet” (Lev 11: 4 2 ), including land-dw elling insects, snakes, w orm s, spiders, centipedes. Leviticus calls these by a w ord that is tran s­ lated as abo m in ab le, but Leviticus uses a d if­ ferent w ord from thar used by D euteronom y an d also tran slated rhe sam e w ay. Leviticus a lso con fuses rhe issues by ap p lyin g a general defilem ent term for land-sw arm ers: rhe people m ust not “ defile” them selves by con tact with their carcasses (Lev 11: 4 3 -4 ). The sanction for infringing the rule is not severe. T hough uncleanness is very con tam in atin g, the san c­ tion is just to w ait till sundow n (Lev 11: 31, 39). If som eone has gone further than just touching an unclean corpse, the sanction is still m ild, he m ust w ash his clothes and rem ain

(v) paws (27)

(vi) land-swarmers (29-38)

unclean until evening; and anyone w ho carries its carcass shall w ash his h ands and rem ain unclean until evening (Lev 11: 4 0 ). We should not exaggerate rhe penalties or the severity of the rules. T he rule o f uncleanness only affects contact with the dead carcass. In this respect contact with the carcasses o f land an im als are accorded a sim ilar, though lesser, im purity as contact with hum an corpses. The people o f Israel are enjoined to have to w ards their livestock som e o f the responsibilities o f a feudal lord to his follow ers. Both hum ans an d livestock are called to be separate and pure in the interior circle o f a w orld o f unclean n ation s or unclean an im als. M ilgrom says that the effect o f the criteria for edible q u ad ru p ed s (Lev 11: 3) is to limit Israel to three dom estic species: sheep, g o a ts, and c a ttle / But there is m ore to it than just not eating. N otice som e o f the tacit restric­ tions that follow from these rules. All land an im als have been classified w ithin the system . They can all be touched alive, but the only ones th at can be touched after death are the classified rum inants. T his m eans that only the latter can be killed for sacrifice. While they are alive cam els and asse s can be harnessed, load ed , ridden, d ogs can be beaten, cats can be kicked, mice can be trap p ed , w ithout incur­ ring im purity, but once rhey are dead rhey convey uncleanness. In effect rhe rule again st touching a dead anim al protects it in its lifetim e. Since its c arcass can n ot be skinned or dism em bered, m ost o f the w ays in which it could be exploited are ruled ou t, so it is nor w orth breeding,

190

M ARY DOUGLAS

hunting, or trap p in g. These unclean anim als are safe from the secular as also from the sacred kitchen. T he rule is a com prehensive com m an d to respect the dead body o f every land anim al. If anyone were to take it seriously it w ould be very restrictive. The verb to touch has also the idea o f harm ing, dam agin g, laying hands upon as if to steal or strike. An exam ple is in G enesis when Abimelech com m an d s his follow ers not to harm Isaac: “ W hoever touches this m an or his wife shall be put to d e ath ” (Cien 26: 11), an d says to Isaac: “ We have not touched you and have done to you nothing but g o o d ” (Gen 2 6 : 29). The rule o f not touching the co rp se m akes the skins useless for fur coats or fur blankets, no leather w aistcoats or bags, no shoe leather or w ine-skins. Their bones and teeth cannot be carved for co m b s, buttons, containers, dice, jew ellery, utensils. T heir gut cannot be used for stringed instrum ents, or their stom ach s or bladders for b ags, or their sinew s for sew ing. In practice the penalty is so light that the rule w ould hardly prevent a ta x i­ derm ist or tanner from pursuing his trade, so long as he purified his clothes and him self before ap p ro ach in g the tabernacle. N on eth e­ less it is still unequivocally forbidden to touch these creatures when dead. The trem endous dom estic com p lication s entailed by the high degree o f con tagiou sn ess m ight deter the furrier, and it w ould be aw kw ard for the w earer o f a mink co at to have to keep w ashing, and no one else could so much as shake hands with the w earer w ithout afterw ards p erform ­ ing the sam e ablution s. T o be classified unclean ought to be an ad van tage for the survival o f the species.

In terp reta tio n s of U n cle an n e ss/lm p u rity T hough M o ses adm onished A aron and his sons to distin guish between the holy and the com m on an d between the clean and the unclean (Lev 10), he did not explain w hat is m eant by unclean, or holy. The sages did not m ake much sense o f it at all. Why pig is counted unclean in the Bible has been the subject o f m uch speculation. C h an gin g the w ord to “ im p u re” does not really help. Why

ever should a non-rum inant with cloven hoofs be counted im pure? Som e sch olars favo u r the idea that pig has to d o with cults o f the dead in Egypt or C an aan , but this does not explain the unclean­ ness o f the other three an im als nam ed with it. When the classification is so com prehensive it cannot be sound to take the anim als a s if they were separate item s in a catalogue w ithout headings or subh eadin gs. If som e o f the rules have the effect o f banning p redators, or blood, or carrion-eaters, it is not a com prehensive explan ation . It is a lso to o H ellenistic, to o o ri­ ented to feelings, for this book . T he sam e for the idea o f Philo, the first-century CF. Jew ish p h ilo so p h er/ that the forbidden species each signifies vice or virtue. In a long, ram bling hom ily he takes each forbidden anim al se p a ­ rately and explain s its prohibition in term s o f sym bols. H is fanciful allegories are not rooted anyw here in the Bible text, only in the im agi­ nation o f the ph ilosoph er, and so inevitably his m oral preoccu pations dom in ate the reading. He derives from Leviticus 11 a lesson to control gluttony, p assio n s, and desire. The anim als that go on the belly are forbidden so as to teach the people not to pay attention to their bellies. It is not a n aturalist explanation , he does not consider the forbidden anim als to be bad in them selves, rather the contrary: All the animals o f land, sea or air whose flesh is the finest and fattest, thus titillating and exciting the malignant for pleasure, he sternly forbade them to eat, knowing that they set a trap for the most slavish of the senses, the taste, and produce gluttony, an evil very dan­ gerous both to soul and body. Philo w as not w orking from any tradition that w as close to the w riting o f the text, for he does not draw on the rest o f Leviticus or D euter­ onom y to con struct his serm ons. Som e m oralizing interpreters have regarded the law s as obliqu e com m an ds that will restrain human “ om n ivorou sn ess and ferocity” ;'’ others again as arbitrary com m an ds to test o b ed i­ ence.7 O thers give up on interpreting at all, treating the rules as inexplicable, though deriv­ ing from an ancient time when they presum ­ ably once m ade som e kind o f sense. Som e treat the Levitical schem e as a relic o f a p astoral w ay

LAND ANIMALS, PURE AND IMPURE

o f life.* For this to be serious there w ould need to be a theory o f why som e relics rem ain strong when their supportin g con text has passed aw ay, while others are forgotten. In default o f such a theory it can still be argued that the Leviticus writer could not betray all the p a sto ­ ral tradition. H is com m itm ent to the idea o f the ancient coven an t w ould preserve rules about herd anim als. H ow ever, the general d is­ array am o n g rival interpretations testifies to the lost tradition. N o t surprisingly the general public is ready to believe that there is som e­ thing abhorrent ab o u t the creatures which the book tells them to abh or. T o this day it is com m on to hear distinguished sch olars explain ritual purity by natural reactions: Many people wince at having to pick up a dead animal; most people (except twoyear-olds) try to avoid touching defecation; corpses inspire a natural feeling of awe, and we hesitate to touch them; washing off semen and blood is almost natural, and certainly not hard to remember. Even gnat-impurity, which sounds picky, is nor hard to understand. Who wants a fly in one’s soup?9 M o st o f the discussion is based on rhe D euter­ onom y form ula, not on Leviticus. The first m istake in this quotation is to have used lines from D euteronom y as if they cam e from Levit­ icus and as if they all m eant the sam e thing. D euteronom y says thar w inged insects are unclean, but not Leviticus. H um an s are co n ­ stantly under their attack , flies feast on babies’ eyes, they breed their m aggots in the larder, w alk contem ptuously over fo o d , they suck blood and sting to frenzy. It is plausible that invasive insects and creepy-craw lies might be disliked universally. Another m istake is to use supposedly natural or “ alm ost n atu ral” reac­ tions to justify all the uncleanness rules. The nuisance value o f insects m akes this exp lan a­ tion plausible, but the rules are not m ostly abo u t insects. Why revile shy an im als like hares and useful an im als like cam els by classi­ fying them with the naturally dislikeable? The n aturalist explanation m ust be w rong for a b o ok so sophisticated as Leviticus, and for an th rop ologists it is alw ays w ron g to take natural as a universal category, forgetting that nature is culturally defined.

191

A gain, the text itself specifically say s that rhese are rules m ade for rhe people o f Israel; w hat is designated as unclean for them is not unclean for the w hole o f hum anity. T h u s the rules o f im purity are not a w ay o f prom oting a universal hygienic principle or pronouncing a general health w arning. The only e x p lan a­ tion will be in the rest o f the rule system . M any civilizations have been built on cam el m eat, or p ork , and though hyrax is hard to get, there is nothing bad for you in adding hyrax or hare to your diet; som e people habitually ear blood, and the dietary value o f suet fat can be under­ estim ated. O ne popu lar explan ation for the banning o f the w ater-sw arm ers (which in Leviticus are not unclean) is that they are scavengers: pigs will eat carrion; shrim ps and crabs feed upon dead fish; so dirty feeders are forbidden. T his explan ation is w eak because a lot o f anim als w ould op po rtun isti­ cally consum e carrion if they found it, and anyw ay the text says nothing ab o u t carrionfeeding anim als. T he concept o f dirtiness has contam inated the conceptual field; the idea o f disgust at eating unclean things dom in ates interpreta­ tion. The kitchen, m edical, and bathroom senses intrude. Leviticus certainly plays upon d isgust at bodily exu dation s in its long d isqui­ sition on uncleanness o f bleeding and leprosy in ch apters 1 2 -1 5 . If im pure w as not originally a term o f vilification it certainly has becom e one. A ppeals to m edical and aesthetic princi­ ples are not the w ay to interpret an enigm atic law in Leviticus: the only safe path is to trace the co n trast sets and parallels the book itself develops. The im purity o f an anim al kind is part o f the technical m eaning o f ritual purity. In itself the idea o f im purity is not difficult to translate. The w ord is well chosen from secular con texts where unclean, defiled, im pure, dirty correspond to a situation which calls for an act o f cancellation. Hut w ashing, polishing, burnishing, are too superficial to carry all the m eanings o f purification. In C h ris­ tianity im pure is used for the defilem ent o f sin, with frequent reference to the parallel wirh a soiled garm ent, it is raught that the repentant soul requires a cleansing rite. But the listed uncleannesses o f Leviticus arc not sin in general, they are a separate set o f sins, they

192

M ARY DOUGLAS

depend on physical con tact only, and the central principle is that the con tam in ated body has con tagiou s pow er, which entails that all its future physical co n tacts convey con tam in a­ tion. The rules prescribe how the object sp re ad ­ ing defilem ent m ust be w ashed, destroyed, or som ehow stop p ed , accordin g to the gravity o f the defilement. The w ord for im pure, tame, is w orked very heavily in Leviticus and used sparsely else­ where in the B ib le.10 We m ay ask why it becam e such a favourite w ord for the priestly writer, but the first question is how it relates to holiness. O nce again , the m ost illum inating passage to explain w hat it is ab o u t is the w arning that G o d gave to M oses ab o u t the sanctity o f M ou n t Sinai. In E xodu s 19: 1 0 -2 4 he tells M o ses to m ake a fence round the m ountain and to prevent the people from app roach in g the m ountain, not even to touch the edge o f it. He tells M o ses to tell them to purify them selves, to w ash their clothes and be ready for the day when he will ap p ear to them, but to w ait until they are sum m oned by trum pet: “ G o dow n and w arn the people, lest they break through to the Lord to gaze, and m any o f them perish. And a lso let the priests w ho com e near to the Lord consecrate them ­ selves, lest the Lord break out upon th em ” (Exod 19: 2 1 - 2 ). The danger is tw o-edged: the people might break through or the Lord might break ou t, and in either case, people will die. T his is the effect o f holiness. The holy thing that is not correctly guarded and fenced will break out and kill, and the im pure person not correctly prepared for con tact with the holy will be killed. Furtherm ore, a person w ho has had the m isfortune to ‘'co n tract”” holiness, to use M ilgrom ’s term , m ay inadvertently co n ­ tam inate other unprotected things or persons, merely by c o n ta ct.11 The nearest usage in European lan guages for the idea o f con tagion is in the discourse of honour, especially with reference to the virtue o f wom en or the honour o f a knight. The taint o f dish on our gives a fair idea o f im purity and violation. In M editerran ean cultures a w om an ’s honour m ust be protected at all co sts; if she is defiled her violator m ust be killed; if her father or brothers fail to cancel the offence they will be dishonoured to o, an d the w hole fam ily. It

is not a m etaph or, it is a concept a b o u t behav­ iour that has practical consequences: none o f her sisters will be able to m arry, no respectable person will d o business on equal term s with the m enfolk, they will not be able to hold up their heads at a m eeting, the contam inated fam ily is ruined . l~ Israel had a p atronal society in which the patron-client relation is expressed by the cli­ ent’s respect for the honour o f the patron. At m eetings between lord and v assals the latter bring specified gifts o f food to be cerem oni­ ously shared. Leviticus says that the cereal offering m ust expressly be given with the “ salt o f the co ven an t” (Lev 2 : 13), which suggests that the term inology and values o f covenant w ould have been current and easily interpretable for the people for whom the book w as written. Defilem ent as a violation o f holiness is a particularly ap t expression for an attack on the honour o f G o d perceived as a feudal lord. The w ord for holy has the sense o f “ co n ­ secrated ” , “ p led g ed ” , “ betroth ed” , as “ sa c ro ­ san ct” in m odern English, som ething forbidden for o th ers,11 not to be encroached upon, diluted, or attack ed . A key text for under­ stan din g im purity in Leviticus w ould be: “ For you are people holy to the Lord your G o d ; the Lord your G o d has chosen you to be a people for his own possession , out o f all the peoples that are on the face o f the earth ” (Deut 7: 6). T h is is follow ed by a reference to the principle o f requital on which the covenant rests (Deut 7: 9 -1 0 ). A few verses later the sam e text g oes on to say w hat being holy or reserved to the Lord entails in term s o f behaviour. It corresp on ds to the requirem ents o f ch ast­ ity and fidelity in the discourse o f honour and betrothal, which is sim ilar to , or rather, m odelled upon the discourse o f alliance and covenant. “ Y ou shall therefore be careful to do the com m andm en t, and the statu tes, and the ordinan ces, w hich 1 com m and you this d a y ” (Deut 7: 11). In addition to obedience the covenant with the overlord requires protecting his honour, or abstain in g from insult. H is pow er protects his people or his things and places, and to insult any o f them is an insult to his honour. The parallel with the discourse o f honour explain s why sins cau se uncleanness to adhere to the

LAND ANIMALS, PURE AND IMPURE

sanctuary and to the altar. Ja c o b M ilgrom is cu riou s to know why the altar should need atonem ent when the altar has not sinned. He develops a convincing theory o f con tagion from sin clustering on and aroun d the holy places until it is w ashed o ff by the rite o f ato n em en t.14 H is an alysis o f co n tagiou s im pu­ rity is im peccable, but one can notice that the lan guage o f dirt an d ablution is unnecessarily m aterialist. In the co u rts o f chivalry a w arrior w ould recognize th at his arm our is dish on ­ oured if he him self is im peached: as well as his children, and father and m other, his helmet, his co at o f arm s, his house, all are tainted and m ade w orthless by the co n tagiou s dish onor. Blood w ash es o ff the m ajor taint, a noble gift cancels a m inor fault. In the sam e w ay, bring­ ing uncleanness into the Lord G o d ’ s sanctuary m ak es it im pure since the place shares in the insult to G o d . Leviticus has first described the pure anim als a s rum inant hoof-cleavers, and then has gone on to exclude “ ru m in an ts” which d o not cleave the h oof an d the one non-rum inant species which does (the pig). T h is order o f listing gives the im pression o f excluded anim als trying to get into the privileged enclosure so th at they to o could be con secrated and share in the L o rd ’s cult. T here w ould have been pres­ sure from enterprising c o o k s seeking to allevi­ ate the m onotony o f the m enu. The sense o f pressure to be included ad d s to the m eaning o f the an im als excluded for having only one but not both defining features. In the midrash the image o f a reclining pig stretching out its cloven hoofs and saying: “ Look, I’m pure,” while concealing the fact that it does not chew the cud, is used to char­ acterize the hypocrisy o f the Rom an empire, which posed as being dedicated to law and justice, while actually oppressing the peoples it ruled.15 Frivolously one can ask why pig or any other anim al w ould seek to be accoun ted pure when the pure an im als are destined for early death and the fire o f the altar. O n a secular view, having one but not both the criteria for purity w ould be a saving blessing, but the con text is religious.

193

T he m eaning o f purity depen ds on the sense o f G o d ’s aw ful m ajesty, m anifest in his creation. E x o d u s describes it in a narrative o f volcan ic explosio n , thunder, fire. D euteronom y describes it with w ords a b o u t G o d ’s pow er, and with verbal w arnings o f disaster. Leviticus conveys it by dou ble, triple, m ultiple m icro­ co sm s. The people, with their children and their servants and their dom estic anim als too, benefit from his covenant. As vassals o f G od their unw orthiness is im m easurable, but yet they are invited to eat at his tab le, and m ay eat the fo od that is offered to him . Sacrifice is a com m unal feast. T heoretically the people o f Israel never eat m eat except in G o d ’s com pan y, in his house and with his blessing. They have been singled out for the h on our o f being con ­ secrated to Ciod, to be his people. T he height and the depth o f this h onour is inexpressible. At another level it is a parallel honour for their flocks and herds, the cloven-hoofed rum inants, to be singled ou t o f all an im al kinds to be consecrated to G o d. T h is p arad igm turns the covenant anim als into v a ssa ls in relation to the people o f Israel, as are the people o f Israel the vassals o f G od.

Sacred C o n tag io n We still can ask w hat interest Leviticus could have had in elaboratin g the con cept o f holiness an d im purity in these w ays. T he full answ er m ust relate to the fact that belief in the m alefi­ cent pow er o f dem ons has been dem olished. The theodicy has to be ch anged: his friends will no longer be able to tell a sick m an that he h as been seized by a leprosy dem on or a w om an that her child has died because a fem ale dem on took it. Sufferin g and sorrow still rem ain, and death. T he priests are expected to explain , give co m fort, and help. T h is is w hat the doctrine o f purity does. If you fall sick, it could be that Ciod h as broken ou t on you because you unknow ingly incurred holi­ ness o r im purity. T h is is a close parallel to the superseded idea that a dem on m ight have caugh t you. A sacrifice will put it right, or a w ash and w aiting till evening, accordin g to the gravity o f the tran sgression . The w ord

194

M ARY DOUGLAS

“ unclean” is particularly apt for relating the Held o f dem on ological medicine to the new regim e, it a ffo rd s a theory o f pain and su ffer­ ing free o f dem on s and affo rd s an alternative explanation for bodily afflictions. So why should touching unclean anim als provoke a dan gerou s breaking-out o f this kind? The insult to G o d is to have com e into his sacred place after profan e and co n tagious contact with the co rp se o f one o f his creatures. T aken together the food purity rules and the touch purity rules are p art o f a unified doctrine in which co rp se pollution, bloodsh ed, and unsanctified death are classed as breaches o f covenant. It has been a puzzlem ent to Christian readers that Leviticus puts unclean con tact into the sam e bracket as breaches o f the m oral code. H ow ever, there is nothing puzzling ab o u t both kinds o f disobedience to the L o rd ’s com m and being treated together. T o touch an unclean thing an d then to ap p roach the tab er­ nacle puts the person in need o f atonem ent. Leviticus in ch apter 5 begins the topic o f uncleanness: O r if anyone touches an unclean thing, whether the carcass of an unclean beast or a carcass of unclean cattle, or a carcass of unclean swarm ing things, and it is hidden from him, and he has become unclean, he shall become guilty (Lev 5: 2). When a man is guilty in any o f these, he shall con­ fess the sin he has committed and he shall bring his guilt offering to the Lord for the sin he has committed . . . and the priest shall make atonement for him for his sin. (Lev 5: 5-6) And again in the sum m ing up, chapter 7 em phasizes the co n tagion s principle: Flesh that touches any unclean thing shall not be eaten; it shall be burned with fire. All who are clean may eat flesh, but the person who eats o f the flesh of the sacrifice o f the Lord’s peace offerings while an uncleanness is upon him, that person shall be cut off from his people. And if anyone touches an unclean thing, whether the uncleanness of man or an unclean beast or any unclean abomination, and then eats o f the flesh of the sacrifice o f the

Lord’s peace offerings, that person shall be cut off from his people. (Lev 7: 19-21) T his is very em phatic lan guage, repetitive, classificatory, and redolent o f m ythopoetic an alogy. Such a statem ent from an archaic thought style can not be decoded into m odern term s. T he interpreter m ust not read em otional quality into lan guage which is prim arily cast in a spatio-tem poral m ode. The contact has been forbidden, and the person w ho has becom e co n tagiou s shall not carry the co n ta ­ gion sacrilegiously to defile the holy place or to eat the flesh o f the L o rd ’s peace offerings; he will be punished. A dom estic rum inant is the designated m edium o f atonem ent and the priest follow ing rhe instructions for a sin offering in ch apters 4 and 5 will m ake atonem ent for a sinner an d he will be fo r­ given. H e can live his ordinary life in this co n tagious state, but because o f the con tagion he and other person s he m ay contact will com m it sacrilege if they take part in the cult o f the tabernacle. He can expect to be criti­ cized by his believing fellow s and be m ade to take the blam e for a com m unity-w ide disaster, and possibly expelled, like Jo n ah by the sailors. Unclean is not a term o f psychological horror and d isgust, it is a technical term for the cult, as com m en tators have often pointed out. T o im port feelings into the translation falsifies, and creates m ore puzzles. The tech­ nique o f delayed com pletion postpon es the m eanings until chapter 17. At that point Levit­ icus com m an ds the people not to eat blood, not to eat an anim al that has died an uncon­ secrated death, that is, an anim al that has died o f itself, o r an anim al torn by beasts, presum ­ ably with its blood still in it (Lev 17: 8 -1 6 ; see a lso D eut 14: 21). The dietary law s thus support the law again st unconsecrated killing. T he Leviticus w riter’s reverential attitude to life, anim al and hum an, explain s the anim al corpse pollution rules. “ T h o u shalt not stand upon [profit from ] an oth er’s b lo o d ” (Lev 19: 16). The case o f the an im al’s blood and the case o f the h um an’s blood are parallel. Ritual im purity im poses G o d ’s order on his creation.

LAND ANIMALS, PURE AND IMPURE

NOTES 1 2

3

4 5 6 7 8 9 10

11

12 13 14 15

M u rray 1992. T he ass that carries lo ad s and person s is not m entioned here. A half-w ay category, it is given a half-w ay treatm ent when it com es to offering the first-born to the tabernacle: “ Every firstling o f an a ss you shall redeem with a lam b . . . ” (E xod 13: 13). A fter the destruction o f the tem ple when the M ishnah substituted the cleanness o f the w orshipper’s body and food for that o f the altar and sacrifice, they already had a strongly developed precedent in Leviticus, chapter 11. N eusn er 1977. M ilgrom 1989. Philo 1 939: 9 9 -1 0 2 . K a ss 1994: 12. M aim on ides 1881. H o uston 1993. San ders 1990: 145. tam e , im pure, occurs 89 times in the Bible; 4 7 tim es in Leviticus; 8 tim es in D euteronom y, and not at all in E xod u s. M ilgrom 1991: 4 4 3 - 5 6 argu es that in Leviticus holy things d o not transm it holi­ ness to persons, but con tact kills them , w hereas Ezekiel follow ing an older trad i­ tion taught that holy things could tran s­ mit holiness to persons as well a s to objects. C am pbell 1964. T igay 1996: 86. M ilgrom 1983. T igay 1996: 139.

REFERENCES C am pbell, Jo h n , H onour, Family and Patron­ age (O xford University Press, 1964).

195

D o u glas, M ., “ A nim als in Lele religous sym ­ b o lism ” , Africa 2 7 (1 957). ------ “ Sacred C o n tag io n ” , in Jo h n Saw yer (ed.), Leviticus (Sheffield A cadem ic Press, 1996), 8 6 -1 0 6 . ------ Purity an d Danger: An Analysis o f the Concepts o f Pollution an d T aboo (London, R outledge &C Kegan Paul, 1966). F^vans-Pritchard, E. E., Witchcraft, Oracles, an d M agic am ong the Azande (Clarendon Press, 1937). H o u ston , W alter, Purity an d Monotheism:

Clean and Unclean Animals in Biblical Law (JS O T S 140, Sheffield A cadem ic Press, 1993). K a ss, L. R ., The Hungry Soul, Eating and the Perfection o f O ur N ature (Free Press, 1994). Levine, Baruch, The FPS Commentary , Leviti­ cus (Jew ish Publication Society, 1989). M aim on ides, M o ses, Guide for the Perplexed (London , 1881). M ilgrom , Ja c o b , “ Israel’s sanctuary: the priestly ‘ Picture o f D orian G ra y ’ ” , in Studies in Cultic Theology and Terminology , Studies in Ju d aism in Late A ntiquity, ed. Jacob N eusn er, 36 (Brill, 1983), 3 9 0 -4 0 0 . ------ “ R ation ale for cultic law : the case o f im purity” , Semeia 45 (1 9 8 9 ), 1 0 3 -9 . ------ Leviticus 1 -16, The A nchor Bible (D o u­ bleday, 1991). M urray , R obert, The Cosm ic Covenant , H eythrop M on ograph s, 7 (Sheed and W ard, 1992). N eusner, Ja c o b , “ H istory and Structure, the C ase o f the M ish n ah ” FA A R 45 /2 (1 977), 1 6 1 -9 2 . Philo, with English tran slatio n , ed. F. H. C o lso n , viii: The Special Law s (H arvard University Press, 1939). Sanders, E. P., Jewish Law from Jesus to the Mishnah, Five Studies (SC M Press, 1990). T igay , )., Deuteronomy, the People's Torah (Jew ish Publication Society, 1996).

27

A Jivaro Version of Totem

and Taboo Claude Levi-Strauss

Professor em eritus at the College de France, Paris, L£vi-Strauss has been a tow ering figure in the intellectual debates of the 20th century. A part from his reputation as a founder of structur­ alism, he has made profound contribu­ tions to anthropology w ith respect to kinship, symbolic classification, myth, and human thought more generally. L£vi-Strauss conducted fieldw ork in Brazil (1970 [1955]) and it is from the rich m ythology of Brazilian Am erindians th a t his analyses and explorations of mythical thought spiral outw ard to em brace the Am ericas as a w hole. This essay is a relatively late piece in Levi-Strauss's long and distinguished career but it provides a succinct exem ­ plification of his approach to myth, dem onstrating both his ability to gener­ ate his own im aginative chain of con­ nections and his sharply polemical style, here directed against Freudian inter­ pretation. As usual, his intelligence.

wit, and encyclopedic erudition are on display. The deliberate pun in the title La Pens£e Sauvage (1966 [1962]), "the wild pansy/' poorly translated as The Savage M in d, provides a clue to Levi-Strauss's pursuit. By "savage" he means undo­ mesticated or relatively spontaneous thought (which is much the same as "unrationalized" in a W eberian fram e­ work). In that w ork Levi-Strauss makes a strong case for the systematic observa­ tional know ledge w ithin Am erindian societies, but he is less interested in the practical (m aterial) consequences than in how signs draw n from the natural world are spontaneously taken up and reapplied to produce a rich mytho­ logical superstructure. For Levi-Strauss, unlike Freud, it is never the intrinsic meaning of any given symbol or myth that is at issue; draw ing from Saussure's account of the structure of phonemes (1959), as w ell as Jakobson's application

From Claude Levi-Strauss, “ A Jivaro Version o f Totem and T aboo,” in The Jealous Potter, trans. Benedicte Chorier (Chicago: University of Chicago Press, 1988 [1985]), pp. 185-206. Abridged.

A JIVARO VERSION OF TOTEM AND TABOO

of Saussurian principles of syntagmatic and paradigm atic relations to poetics (1960), Levi-Strauss understands signs as having m eaning only in term s of their contrastive relations to other signs. Signs draw n from one sem antic domain, or code, as he refers to it here, are in turn contrasted w ith sets of signs drawn from other codes to produce chains, analo ­ gies, and "double twists" (M aranda 2001). He argues that through myth prim ary oppositions (like life and death) are mediated by oppositions from other codes (e.g., day and night), and shows how neighboring myths play off against each other in complex patterns of m eta­ phor, metonymy, repetition, inversion, and puns that, in another analogy, can best be com pared to those of music. These patterns are not produced to any singular purpose or m eaning (as music is not), but express the mind's continu­ ous signifying activity. Levi-Strauss has had a trem endous if som ewhat idiosyncratic influence, showing more vividly than anyone the richness of human creativity w hile rigor­ ously evading subjectivism or hum an­ ism and challenging anthropologists to become philosophically and aestheti­ cally more sophisticated in their ap­

Freud gave the follow in g subtitle to Totem and T ab o o : “ Som e Points o f Agreem ent between the M ental Lives o f Sav ages and N eu ro tics.” In the preceding p ages I have set out to show instead that there are poin ts on which the m ental lives o f sav ag e s and psych oan alysts coincide. At alm ost every step we have encoun­ tered perfectly explicit notions and categories - such a s oral ch aracter an d an al character that psych oan alysts will no longer be able to claim they have discovered. All they have done is to rediscover them. Better yet: it is Totem an d T aboo in its entirety th at, well ahead o f Freud, the Jiv a ro Indians anticipated in the myth that for them p lays the part o f a G enesis: societies arose

197

proach to cultural phenom ena. However, w hile some attem pt to tu rn the argu­ ments concerning human tho ug ht into m athem atical form ulae (M aranda 2001), others have grown im patient w ith the endless transform ational play of signifi­ e s and w ant to know w h a t precisely people in given contexts are thinking about or trying to say, and w hich of their symbolic productions become relatively "fixed," entextualized, and authorized in dom inant regimes of m eaning. The herm eneutic critique has been best expressed in the debate betw een Rtcoeur and L6vi-Strauss (1970), by G eertz (1973e [1967]), and by Douglas in her response to one of Levi-Strauss's best short pieces of myth analysis, known as Asdiw al (in Leach, ed. 1967). The essays in Structural A n th ro p o lo g y I (1963a) form a good introduction to Levi-Strauss before tackling the fourvolum e M y th o lo g iq u e s (1964-71). The form er w ork also contains a pair of out­ standing essays on sham anic cure that are rather more sym pathetic to psycho­ analysis (1963a, b). The secondary litera­ tu re on Levi-Strauss is enorm ous. Leach, in particular, provided introductions to an Anglophone public (1970, 1976), and Culler (1975) is also very good.

when the prim itive horde sp lit into hostile clans after the m urder o f the father w hose wife had com m itted incest with their son. From a psych ological point o f view the Jiv a ro myth offers an even richer and m ore subtle plot than Totem and T aboo . Let us outline the plot again . While his father, U nushi, w as aw ay on a long trip. A him bi, the Snake, slept w ith his m other, M ik a , the C lay Pot. It is as if these tw o off­ enders - the snake and the vase - sym bolized, respectively, the m ale and the fem ale genitals, naturally destined to unite, notw ithstanding the social rules th at w ould restrain their freedom . And the patriarch - their father and gran dfath er - actually banished them. They

198

CLAUDE LEVI-STRAUSS

rem ained vagran ts and had m any children. When the deceived husband returned and d is­ covered his m isfortune, his w rath w as directed not again st the offenders but again st his ow n m other, w hom he accused o f having encour­ aged their crim e. It w ould be tem pting to say that he held her respon sible for his ow n inces­ tuous desire for her and that his son ’s crim e w as the enactm ent o f his own secret w ish. The offsp rin g o f the incestuous couple w anted to avenge their gran dm oth er, so they beheaded their m other’s h usban d in Totem and Taboo style. This triggered a series o f conflicts. M ika killed her children, w ho had m urdered her h usban d; her in cestuous son then sided again st her, an d , from that point on, the three cam ps - the father’s, the m other’s, and the so n ’s engaged in a m erciless fight. T h is is how Society cam e ab o u t. Psychoanalytical theory cannot be credited with uncovering the latent m eaning o f m yths. M yths were its precursors in this. The Jiv a ro Indians’ theory on the origin o f society may well be sim ilar to Freud’s - indeed, they did not w ait for him to announce it. H ow wise are the A m ericans in calling psych oan alysts “ h eadshrin kers,” thus spontaneously a sso c ia t­ ing them with the Jiv a ro ! T herefore, Freud cannot be credited with know ing w hat m yths say better than the myths them selves d o. M yth s d on ’t need any help when it com es to reasoning like a p sy ch oan a­ lyst. Freud’s m erit lies elsew here; it is o f the sam e order as the merit I have recognized in M ax M uller’s achievem ents (see The N aked M an , p. 4 4 ). Each o f these great m inds deci­ phered one o f the codes - M uller, rhe code o f astron om y; Freud, the psycho-organic code thar myths have alw ays known how to use. But each o f them m ade tw o m istakes. First, they tried to decipher myths by m eans o f a single and exclusive code, while a myth will alw ays put several codes in play, and it is from this layering o f codes, one on to p o f another, that rules o f interpretation derive. The signification o f a myth is alw ays g lo b al; it cannot be reduced to the interpretation p ro ­ vided by one p articular code. N o language astron om ical, sex u al, or other - conveys the “ better,” m eaning. As I pointed out in The Raw and the C ooked (p. 2 4 0 ): “ T he truth o f

the myth does not lie in any special content. It con sists in logical relation s which are devoid o f content or, m ore precisely, w hose invariant properties exh aust their operative value, since com parable relations can be established am on g the elements o f a larger num ber o f different con ten ts.” There is no m ore truth in one code than in any other. T he essence o f the myth (or its m essage, if one w an ts to call it that) is founded on the property inherent in all codes: that o f being m utually convertible. The second m istake lies in the belief that, am on g all the codes available to m yths, one particular code is obligatorily em ployed. A myth alw ays uses several codes, but it does not follow that all con ceivable codes, or all the codes identified by com parative an alysis, are sim ultaneously at w ork in all myths. One could certainly draw up a list o f all the codes rhat mythic thought uses - or could use - and such a list w ould be helpful to m ythologists in the sam e w ay a s the periodic table o f elements is helpful to chem ists. But each myth or fam ily o f myths m akes a choice am o n g all these codes. The few at w ork in a specific myth are in no way representative o f all o f the inventoried codes and are not necessarily the sam e ones as another myth or fam ily o f myths w ould have selected for its ow n particular use. In this book I have concentrated on one fam ily o f m yths in which the psycho-organic code - the sexual code, if you will (but I will com e back to that) - is pressed into service, along with others: the technological, the z o o ­ logical, the co sm ological, etc. O ne w ould be w rong to assum e from this that the psychoorganic code will have rhe sam e operational value in any other myth or fam ily o f m yths, which m ay use entirely different codes. The follow ing problem w as raised by Freud in Lecture X o f his General Introduction to Psychoanalysis: “ Even if there were no dream censorship, we should still find it difficult to interpret dream s, for we should then be confronted with the task o f translating rhe sym bolic language o f dream s into the language o f w aking life” (Freud 19 3 5 : 150). In other w ords, the essence o f dream s lies in the fact that they are coded. But how can it be that we have access to this code, th at “ we arrive . . . at constant substitu tion s for a series o f dream -

A JIVARO VERSION OF TOTEM AND TABOO

elem ents, just as in p o p u lar books on dream s we find tran slation s for everything that occurs in d re am s,” even though, “ when we em ploy the m ethod o f free associatio n , such con stan t substitution s for dream -elem ents never m ake their ap p earan ce ” ? (ibid., p. 134). Fifteen years later, in his New Introductory Lectures on Psycho-Analysis , Freud, still preoccupied by the problem , form ulated it in the sam e term s: the contents o f dream s “ are to be taken as sym bols for som ething else. . . . Since we know how to translate these sym bols, while the dream er d oes n o t . . . , the sense o f the dream is im m ediately clear to us . . . , while the dream er him self is still puzzled by it” (Freud 1933: 23). Here psych oan alysis and structural analysis diverge on an essential point. T hroughout his w orks, Freud oscillates - and in fact never succeeded in ch oosin g - between tw o concep­ tions o f the sym bol: realist and relative. A realist conception w ould attribute one and only one signification to each sym bol. These significations could all be listed in a dictionary, which, as Freud suggested, w ould differ little from a “ dream b o o k ,” except for its greater size. The second conception adm its that the signification o f a sym bol varies with each p a r­ ticular case, and, to discover the signification, it has recourse to the m ethod o f free a sso c ia ­ tion. So, in an elem entary and unsophisticated fashion , it recognizes that the sym bol draw s its signification from the con text, from the w ay it relates to other sym bols, which them selves, in turn, find their m eaning only in relation to it. T h is second conception can yield positive results, provided the sim plistic method o f free associatio n finds its due place within a global attem pt at understanding the individual by reconstructing his personal history and the history o f his fam ily, his social environm ent, his culture, and so on. O ne w ould thus seek to understand an individual in the w ay an ethnographer seeks to understand a society. So, though Freud had taken the first steps in this direction, he went no further; instead, hopin g to discover an absolu te signification for sym bols, he seem s to have turned more and m ore to everyday language, etym ology, an d philology (at rimes building on som e significant m istakes m ade in these fields, as

199

Benveniste has pointed out). Freud’s purpose in this w as identical to Ju n g ’s; they differed only in that Ju n g proceeded with great haste, while Freud lingered in the backw aters o f scholarly research and the ard u o u s pursuit o f w hat he called “ the original m yth” : Consequently [in a search for the absolute meaning of symbols] I hold that the surface versions of myths cannot be used uncritically for comparison with our psychoanalytical findings. We must find our way back to their latent, original forms by a comparative method that eliminates the distortions they have undergone in the course o f their history. (Freud and Jung 1974: 472) Freud is quite right in o p p o sin g his m ethod to Ju n g ’s, for, “ in his recent m ythological studies, [Ju n g . . .| uses any m ythological m aterial w hatsoever . . . w ithout selection. . . . N o w , m ythological m aterial can be used in this way only when it ap p ears in its original form and not in its derivatives” (N unberg and Federn 1 9 6 2 -7 4 : III, 335). T h is criticism is quite pertinent - and intrigu­ ing, to o, for it can also be applied to its author. Under the pretense o f goin g back to the origi­ nal m yth, all Freud did - all he ever did - w as to produce a m odern version even m ore recent than the ones Ju n g used, which he condem ned for their inauthenticity. Psychoanalysis has never been able to prove that its interpreta­ tion s re-create myths in their original form if only for the sim ple reason th at the original form (provided this notion m eans anything) is and rem ains forever elusive. H ow ever far back we m ay g o , a myth is know n only as som e­ thing that has been heard and repeated. In startin g o ff on a search for the original form , and in believing that he has found it, the psych oanalyst finds in the myth only w hat he him self has introduced into it, a s Freud him self has candidly confessed: “ T he m aterial has been transm itted to us in a state that does not perm it us to m ake use o f it for the solution o f our problem s. On the contrary, it m ust first be sub­ jected to psychoanalytic elu cidation ” (N unberg and Federn 1 9 6 2 -7 4 : 111, 3 35). In a letter to Ju n g ab o u t the w riting o f Totem and Taboo and his difficulties, he sadly adm itted, “ Besides, my interest is dim inished by the conviction that

200

CLAUDE LEVI-STRAUSS

1 am already in possession o f the truths I am trying to p rov e” (Freud and Ju n g 1 9 7 4 : 4 7 2 ). One could hardly say it better. In a w ay, th ough , Freud is unfair to him self. H is greatn ess lies partly in a gift he p ossesses in the highest degree: he can think the w ay m yths do. C on sid erin g that the snake can take on a m ale or fem ale con n otation , he w rote: “ This does not, how ever, m ean that the sym bol has tw o signification s; it is sim ply em ployed in the inverse sen se” (N un berg and Federn 1 9 6 2 74: III, 335). He also w rote that in d ream s we often find “ the procedure o f reversal, o f turning into the op p o site, o f inverting relation sh ip s” (Freud 1964: X X II, 188). Still, w orking in an indirect w ay, Freud here reaches a key notion: that o f tran sform atio n , which is at the ro o t o f all his analyses. A ccording to him, in order to understand the biblical myth o f Genesis o r the Greek myth o f Prom etheus, one needs to invert them. Fve becom es the m other w ho gives birth to A dam , an d m an, rather than w om an , fecun­ dates his sp ou se by giving her seeds (a p om e­ gran ate) to eat. The myth o f Prom etheus also becom es clearer if the fennel stalk con tain in g fire becom es, through inversion, a penis, that is, a tube carryin g w ater (urine), allow ing men to destroy fire (instead o f obtaining it). Like­ w ise, by an inversion o f container into content, the actual them e o f the legend o f the Labyrinth turns ou t to be an an al birth: the w inding path s o f the m aze represent the intestines, and A riadne’s thread is the um bilical cord. T hese are excellent variants. The w ay they relate to the m yths they are based on is quite sim ilar to the relationship the ethnologist observes between the m yths o f one p opulation and those o f an oth er, which, in borrow in g the m yths, has inverted the term s or has tran s­ posed them into a new code. One can easily im agine the neighbors o f the ancient H ebrew s telling the myth o f A dam and Fve in the Freud­ ian fashion , o r H esio d ’s Boeotian co n tem p o­ raries sim ilarly giving their ow n version o f the Prom ethean myth. The trouble is that they did not. But it is in p art th anks to Freud that these m yths are still present in our spiritual heritage. The O edipus m yth, to cite but one exam ple, has retained its vividness, still has an im pact on us, because o f Freud’s new interpretations o f it and the interest they have aroused am o n g

all grou p s and levels o f our society. T h at is why, as 1 said thirty years a g o , we m ust not hesitate to place Freud after Soph ocles am on g our sources for the O ed ipu s myth. The vari­ an ts elaborated by Freud obey the law s o f mythic th ought; they respect the sam e con ­ straints and apply the sam e transform ational rules. Freud him self w as aw are o f this affinity between mythic thought and his ow n. I m en­ tioned above (pp. 1 8 7 -8 ) that th roughout his life he w as haunted by the sam e question: “ H ow do we p rofess to arrive at the m eaning o f these dream -svm bols, ab o u t which the dream er him self can give us little or no infor­ m atio n ?” (Freud 1935: 141). If such a diffi­ culty exists, it com es from Freud’s strange conception o f the w ay this in form ation reaches the an alyst: “ Ju st as in prim itive, gram m arless speech, only the raw m aterial o f thought is expressed, and the ab stract is m erged again in the concrete from which it sp ra n g ” (Freud 1933: 3 2 -3 ). The ethnologist and linguist will certainly be startled by the notion that prim i­ tive lan guages are gram m arless; but, leaving that aside, we can see that Freud actually touched on the crux o f the problem when he w rote: “ T he dream seem s to be an abridged extract from the asso c iatio n s, which has been put together in accordan ce with rules which we have not yet u n d ersto o d ” (ibid., p. 22). These rules are precisely those o f a gram m ar he considered from the start to be nonexistent, a s we have just seen. In order to avoid this dead end, Freud m akes a strategic move: We derive our knowledge |o f the meaning of dream-symbolsj from widely different sources: from fairy tales and myths, jokes and witti­ cisms, from folklore, i.e., from what we know o f the manners and custom s, sayings and songs, o f different peoples, and from poetic and colloquial usage o f language. Everywhere in these various fields the same symbolism occurs, and in many of them we can under­ stand it without being taught anything about it. (Freud 1935: 141)' It is true that in all lan guages there are m ore or less exactly m atching expressio n s, based on sim ilarities or co n trasts that m ight be thought

A JIVARO VERSION OF TOTEM AND TABOO

to issue from asson an ces or hom oph onies that are specific ro each lan guage hut in fact are the em an ation , in p op u lar lan guage, o f thoughts th at d raw their substan ce from the very roots o f the mind. Freud could have propped up his theory better by q uotin g from ch apter three of R o u sse a u ’s Essay on the Origin o f Languages: “ Figurative language cam e first; literal m eaning w as discovered last. M en first sp oke only in poetry; it w as a long time before they invented reaso n in g .” But if we accept Freud’ s solution , can we consider psych oan alysis as anything m ore than a branch o f com parative an th rop ology applied to the study o f individual m inds? Freud him self acknow ledged m ore than once the dependence o f psych oan alysis on the social sciences and the hum anities: “ The province o f sym bolism is extraordin arily wide: dream -sym bolism is only a sm all part o f it. . . . Psycho-analvtic w ork is so closely intertwined with so m any branches o f science, the investigation o f which gives prom ise o f the m ost valuable conclusions: with m ythology, philology, folk-lore, folk psych o­ logy, and the study o f religion ” (Freud 1935: 1 4 9 -5 0 ). H ow ever, this recognition proved so em b arrassin g that he hastened ro add: “ In its relation with all these other subjects, psych o­ an alysis has in the first instance given rather than received” (ibid., p. 150) - a claim su p ­ ported only by rhe assertion that “ rhe mental life o f the hum an individual yields, under psychoanalytical investigation, exp lan ation s which solve m any a riddle in the life o f the m asses o f m ankind or at any rate can show these problem s in their true light” (ibid.). But the w hole lecture from which these q u o tatio n s are draw n rests on the very o p p o site o f this prem ise, nam ely, that vario u s facts that are relevant to the m ental life o f the individual facts for which he him self can find no ex p lan a­ tion - can be understood only by relating them to “ the life o f rhe m asses o f m an k in d .” Precisely. H is New Introductory Lectures , written later than the General Introduction , show m ore caution on this issue. They cast a cloud over the w hole debate, sim ply sayin g that “ any confirm ation we could get from other sources, from philology, folklore, m ythology or ritual, w as particularly w elcom e” and that “ very

201

often pictures and situ ation s ap p ear in the m anifest content o f rhe dream which remind one o f well-know n them es from fairy stories, legends and m yth s” (Freud 1933: 3 8 -9 ). But precedence is no longer an issue. In The Raw and the Cooked (p. 338) I m yself pointed our that the interpretation o f myths from distant regions, m yths that ap p ear extrem ely obscure at first, is som etim es sim ilar to the very obviou s an alogies we m ake in our native ton gue, w hatever it m ay be. But to understand the phenom enon we need to define sym bolism a s m ore than m ere com p arison . N either figurative language nor its m ost com m on m eans o f exp ressio n , m etaphor, can be reduced to a transfer o f m eaning from one term to another. For these term s do not start ou t jum bled together in an indiscrim inate m ass; they are not contained in a com m on pool from which one could draw , at will, just any term and associate it or o p p o se it to just any other. M ean in g is transferred not from term to term but from code to code - that is, from a category or class o f term s to another category or class. It w ould be especially w rong to assum e that one o f these classes or cate g o ­ ries naturally pertains to literal m eaning, the other to figurative m eaning; for these functions are interchangeable and relative to each other. As in the sex life o f snails, the function o f each class, literal or figurative, starts ou t as unde­ term ined; then, accordin g to the role that it will be called upon to play in a glo bal structure o f signification, it induces the opposite func­ tion in the other class. In order to show that dream w ork tran slates a b stract ideas into visual im ages, Freud cites an observation m ade by Silberer: “ I think thar I intend to sm ooth out an uneven p a ssa g e in an essay I am w riting. Visual im age: I see m yself planing a piece o f w o o d ” (Freud 1933: 37). H ow ever, the im age o f a w riter sitting at his desk and bending over his m anuscript to cross ou t a w ord w ould be no less visual than the im age of the carpenter. T h is exam ple is rem ark­ able (note that it bears no trace o f repression or sexuality), not because o f the p assage from the ab stract to the concrete, but because an expression that in w akin g life is used in a figu­ rative w ay is m etaphorically tran sposed by the

202

CLAUDE L£VI-STRAUSS

discourse o f the dream into its literal m eaning. It might he objected that the adjective “ uneven,” properly speaking, can refer only to a m aterial surface, but in everyday life none o f us thinks in the categories o f the gram m arian . T o the w riter, the w ork o f the carpenter is an im age o f his ow n w ork, just as the w riter’s w ork m ight remind a carpenter o f his own activity. A m eta­ phor alw ays w orks both w ays; if I m ay use a rough sim ile, it is like a tw o-w ay street. In sw itching term s that belong to different codes, the m etaphor rests on an intuition that these term s connote the sam e sem antic field when seen from a m ore global perspective. The m eta­ phor restores this sem antic field, n otw ith stan d­ ing the efforts m ade by analytic thought to subdivide it. In Silberer’s dream the m etaphor does not replace an abstract element with a concrete one. Like all m etaphors, it restores the full m eaning o f a notion that, whether used in its literal or its figurative m eaning, is bound to be im poverished in everyday language. In other w ords, the m etaphorical process is a regression effected by the savage m ind, a m om entary su p ­ pression o f the synecdoches that are the o p e ra ­ tive m ode o f the dom esticated m ind. V ico, and R ousseau after him , were well aw are o f this. They cam e short in only one thing: they saw figurative or m etaphorical language as issuing directly from p assio n s and feelings. (V oltaire held the sam e belief, saying that m etaphor, when it com es naturally, pertains to p assio n s, w hereas co m p arison s pertain solely to the mind.) They failed to see that it is, rather, the prim itive apprehension o f a global structure o f signification - and that is an act o f the understanding. V ico, R ousseau, and V oltaire were on the w rong track, and Freud follow ed them in claim ing that, for dream sym bols, there is an unlimited num ber o f signifiers, while the signifieds rem ain alw ays the sam e - m atters co n ­ cerning sexuality. W hat w as Freud’s real stand on this issue? There is no d ou bt that any true disciple could provide a brilliant d em o n stra­ tion that, in his New Introductory Lectures , Freud w as not con tradictin g him self when he repudiated a few form u las “ which we have never put fo rw ard , such as the thesis that all dream s are o f a sexual n ature” (Freud 1933:

17), and then declared, only a few p ages later, “ O ur w ork o f interpretation uncovers w hat one m ight call the raw m aterial, which often enough m ay be regarded as sex u al” (ibid., p. 39); o r, again , when he broadened his co n ­ ception o f sexuality, defining it as an “ uncon­ scious im p u lse ,. . . the real m otive force o f the dream ” (ibid., p. 3 5 ), and when he reproached Pfister for disputin g “ the splitting up o f the sex instinct into its com ponent p a rts” (Freud 1963: 6 2 ); for, as he had clearly stated in the New Introductory Lectures , “ It is one o f the task s o f psycho-analysis to lift the veil o f am nesia which shrouds the earliest years o f childhood and to bring the expression s o f infantile sexual life which are hidden behind it into con scious m em ory” (Freud 1933: 4 4 ), addin g that “ all im perishable and unrealizable desires that provide the energy for the form ation o f dream s throughout one’s w hole life are bound up with the sam e childish experiences” (ibid.) - these childish experiences being o f a sexual nature, as he has just stated. Besides, dream form ation is not the only thing at stake: “ The w orld o f myths and fairy tales first becam e intelligible through the understanding o f children’s sexual l i f e. . . . T h at has been achieved as a beneficial by-product o f psychoanalytical stu dies” (Freud 1950: 60). Such statem ents, oscillatin g between explic­ itness and am biguity, leave one puzzled. It is not that sexuality is shocking to a m ythographer: the tales told by A m erican Indians and other peoples have put him into the sw ing o f things, so to sp eak. But isn’t it becom ing increasingly clear that, even though dream s that can be interpreted a s em anating from repressed sexual desires do indeed constitute a real, even an im portant, category o f dream s, they rem ain just that: one special category am on g oth ers? T he dream er uses a much m ore com plex m aterial in elaboratin g his dream . He doubtless draw s on conscious or repressed desires; but he also vaguely perceives noises around him, his m ovem ents m ay be restrained by the presence o f a foreign object in the bed, he may be physically indisposed, he m ay be w orried ab o u t his w ork or his career, etc. Freud agrees that “ the condition o f repose w ithout s t i mu l i . . . is threatened . . . in a chance fashion by external stim uli during

A JIVARO VERSION OF TOTEM AND TABOO

sleep, by interests o f the day before which have not yet abated , a n d . . . by the unsatisfied repressed im pulses, which are ready to seize on any opportun ity for exp ressio n ” (Freud 1933: 28). For him, how ever, these stim uli and inter­ ests constitute raw m aterials used by the repressed im pulses to code a m essage that rem ains their property from start to finish. C o u ld n ’t one rather say that all these d isparate elem ents are offered to the dream er’s subcon ­ scious a s the scattered pieces o f a puzzle and that, since their heterogeneity is intellectually discom forting, the subcon scious will be obliged in the dream (dream -w ork also being a form o f “ bricolage” ), by piecing them together into a syntagm atic sequence, to give them , if not coherence (certainly not all dream s are coher­ ent), at least som e sort o f organized fram e­ w ork? A ccording to Freud, “ the real motive force o f the dream alw ays finds its outlet in a w ish-fulfillm ent” (Freud 1933: 35). But wishfulfillment p resup p oses desire, which is one o f the m ost obscure notions in the w hole field o f psychology, and there are other m otives at play, upstream from desire, so to speak: ap p e ­ tites and needs; and the universal need m oti­ vating dream -w ork is, con trary to w hat Freud som etim es appeared to think (see above), a need to im pose a gram m atical order on a m ass o f random elements. It is not my p urpose here to replace sexual sym bolism with a sym bolism o f a linguistic or philosophical nature; that w ould bring us d a n ­ gerously close to Ju n g , w h o, as Freud rightly pointed out, “ attem pted to give to the facts o f an alysis a fresh interpretation o f an abstract, im personal, and ah istorical ch aracter” (Freud 1948: 96). We will not attem pt to find the “ tru e” signification o f m yths or dream s. M yth s, and perh aps also d ream s, bring a variety o f sym bols into play, none o f which signifies anything by itself. They acquire a sig­ nification only to the degree that relations are established am o n g them. Their signification is not absolute; it hinges on their position. T o m ake a sim plistic co m p arison , the an aly­ sis o f m yths is faced with som ething rem inis­ cent o f Jap an e se script or, rather, scripts. Ja p a n e se uses tw o syllabaries, which differ only in the way they are w ritten; it a lso uses, in ad dition , a set o f ideogram s derived from

203

Chinese. T hese scripts are not independent but com plem entary. Each o f the tw o kana sylla­ baries gives an un am bigu ous phonetic rendi­ tion o f w ords but an am b igu o u s sem antic one because o f the great num ber o f hom onym s in Ja p a n e se ; for exam ple, the w ords kan , kd , and sho are each given no less than fifteen hom onym s in a dictionary o f everyday language. The Chinese ch aracters, or kanji , w ork in the opposite w ay: m ost include a key or root that indicates the sem antic field to which the transcribed w ord or w ords belong, whether in their noun or verbal form s. In Jap an e se one ch aracter can refer to m ore than ten w ords that arc sem antically related but som etim es com pletely different phonetically. The sound o f the w ord is indi­ cated by one or m ore kana, written above or next to the kanji ; the m eaning is m ainly pro­ vided by the kanji. Ja p a n e se w riting thus uses tw o codes at once (and even three, though n ow adays one o f the tw o syllabaries tends to be reserved for the transcription o f foreign w ord s not yet integrated into the language). The m eaning o f a text can n ot be draw n from one or the other code, since each o f them, taken alone, leaves am b iguities; it is the co m ­ bination o f the tw o that provides com plete understanding. M yths w ork in a sim ilar w ay, except that a greater num ber o f codes are brought into play. [--•I

1 will perhaps be charged with reducing the life o f the mind to an ab stract gam e, replacing the hum an soul and its p assio n s with a clinical form ula. I d o not contest the existence o f im pulses, em otions, or the tum ultuous realm o f affectivity, but I do not accord prim acy to these torrential forces; they irrupt upon a structure already in place, form ed by the archi­ tecture o f the m ind. If we w ere to ignore these mental con strain ts, we w ould regress to the illusions o f a naive em piricism , with one dif­ ference: the mind w ould a p p ear passive before internal rather than external stim uli, a tabula rasa tran sposed from the realm o f cognition to that o f em otional life. A prim itive schem atism is alw ays there to im pose a form on the turm oil o f em otions. In its m ost spon tan eo u s im pulses, affectivity tries to break through ob stacles that

204

CLAUDE L^VI-STRAUSS

a lso act a s lan d m ark s: these m ental ob stacles restrain affectivity while leading it alo n g a limited num ber o f possible path s, each with required halting-places. [ ■■•I

We know that the m eaning o f a w ord is doubly determ ined: by the w ords that precede o r follow it in the sentence and by the w ords that could be substituted for it to convey the sam e idea. Sequences o f the first type are called syntagm atic ch ain s by linguists; they are artic­ ulated in time. T h e second type are called p a r a ­ digm atic sets; they are m ade up o f w ords that could be m obilized at the m om ent a speaker ch ooses one in preference to others that he m ight also have used. N o w , w hat are the p rocesses involved in defining a w ord, in shifting it into a figurative sense, and in ch oo sin g a sym bol to represent rhe notion it stan d s for? Defining a w ord is replacing it with anoth er w ord or phrase draw n from the sam e p arad igm atic set. U sing a m etaphor is tak in g a w ord or phrase from one syn tagm atic chain and placin g it in another syntagm atic chain. The sym bol, for its part, is an entity that entertains within a given con cep ­ tual realm the sam e syntagm atic relations with its con text a s, w ithin a different conceptual realm , the thing sym bolized has with its context. Sym bolic thought thus brings together into the sam e p arad igm atic set h om ologou s term s each o f which belongs to its ow n syn­ tagm atic chain. But the signification, or added signification, that one is aim in g at does not belong per se to the new w ord , the new syntagm atic chain, or

REFERENCES Benveniste, E. 1 9 6 6 . Problemes de hnguistique generate. 2 vo ls. Paris: G allim ard. Freud, S. 1 9 3 3 . N ew Introductory Lectures on Psycho-Analysis . T ran slate d by W. J . H. Sprott. N ew Y ork: N orto n . ------. 1935. A G eneral Introduction to PsychoAnalysis. A uthorized Flnglish tran slation of the revised edition by Jo a n Riviere. N ew Y ork: Liveright. ------. 1948. An Autobiographical Study. T ran slated by Ja m e s Strachey. Lon don :

the new p arad igm atic set. Signification is the produ ct o f the relations established between them and the other w ord, ch ain, or set, which rhey supplem ent rather than replace, so that they will enrich or nuance the sem antic field to which they belong or will define its limits m ore precisely. Signifying is nothing but e stab ­ lishing a relation between term s. Even rigorous lexicograph ers, aw are o f the dan gers o f circu­ lar definitions, know th at, in their efforts to avoid them , they often d o no m ore than w iden the circle. D efinitions are bound to be circular: w ords are defined by other w ords that are ultim ately defined by the very w ords they were defining. The vocabu lary o f a given language m ay be m ade up o f tens or hundreds o f th ou­ san d s o f w ords; nevertheless, ideally at least, at a given point in time it constitutes a closed system . T h u s the reciprocity o f perspectives that I have seen a s the specific ch aracter o f mythic thought can claim a m uch w ider range o f ap plications. It is inherent in the w orkings o f the mind every tim e it tries to delve into m eaning. The only difference lies in the dim en­ sion s o f the sem antic units to which the mind applies itself. Free from the concern o f an ch or­ ing itself to an ou tside, absolu te reference, independent o f all co n text, mythic thought should not thereby be op p o se d to analytical reason. With an authority that can n ot be denied, it arises from the depth s o f tim e, setting before us a m agnifying m irror that reflects, in the m assive form o f concrete im ages, certain m echanism s by which the exercise o f thought is ruled.

H ogarth Press and the Institute for Psycho-Analysis. — . 1950. The Question o f Lay Analysis. T ran slated by N ancy Prater-G reg. N ew Y ork: N o rto n . — . 1962. Totem and Taboo. T ran slated by Ja m e s Strachey. N ew Y o rk : N orto n . — . 1963. Psychoanalysis and Faith: The

Letters o f Sigmund Freud and O skar Pfister. Edited by Heinrich M en g and Ernst L. Freud. T ran slated by F'ric M osbach er. London: H ogarth Press an d the Institute for Psycho-Analysis.

A JIVARO VERSION OF TOTEM AND TABOO

------. 196 4 . “ The A cquisition and C on trol o f Fire.” T ran slated by Ja n ie s Strachey. Pp. 1 8 7 -9 3 in The Standard Edition o f the

Com plete Psychological Works o f Sigmund Freudy vol. 2 2 . London: H ogarth Press an d the Institute for Psycho-Analysis. Freud, S., and Ju n g , C. G . 1974. The Freud-

Ju n g Letters: The Correspondence between Sigm und Freud and C. G . Jung. Edited by W illiam M cG uire. T ran slated by R alph M anheim and R. F. C . H ull. Bollingen Series X C IV . Princeton: Princeton University Press. Levi-Strauss, C. 1969. The Raw and the Cooked. Vol. 1 o f Introduction to a Science

205

o f Mythology. [Mythologiques.] T ran slated by Jo h n and Doreen W eightm an. C h icago: University o f C h icago Press. ------. 1981. The N aked Man. Vol. 4 o f Introduction to a Science o f Mythology. [Mythologiques.] T ran slated by Jo h n and D oreen W eightm an. N ew Y ork: H arper 6c Row . N u n berg, H ., and Federn, E ., eds. 1 9 6 2 -7 4 .

M inutes o f the Vienna Psychoanalytic Society. 4 vols. T ran slated by M . N u n ­ berg. N ew Y ork: International Universities Press.

18

Text-Building, Epistemology, and Aesthetics in Javanese Shadow Theatre Alton L. Becker

A lton (Pete) Becker taught in Linguistics and the Centre for Southeast Asian Studies at the University of M ichigan, w here he also played in the gam elan orchestra conducted by his w ife, ethnomusicoiogist Judith Becker. He is renow ned for his unique, culturally inform ed approach to linguistic texts. Becker approaches ritual and religion from the direction of language, practic­ ing w hat he calls philology (1995). His interests lie in the ways in w hich lan­ guages build worlds, and conversely, the w orld that can be seen contained w ithin a single sentence or, in this case, a text. Becker invites us to see many subtle things about the relationship of language to religion and aesthetics in Southeast Asia and the differences betw een Javanese and western concep­ tions of textual coherence. Becker elaborates an exem plary m ul­ tileveled method for the interpretation of cultural texts. The Javanese shadow

theater performances can be under­ stood to be drawn from literal texts of the Ram ayana, but Becker's method applies to a far broader understanding of cultural texts. Ricoeur (1971, 1976) provided a basic model for understand­ ing the w ay speech or action becomes text and made a strong argum ent for interpretation w ith respect to changing context. Geertz (1973a) applied the textual m etaphor to cultural interpretation. Geertz's critics often took this too literally; in fact, it can be fruitfully and flexibly applied to such open-ended, never w ritten phe­ nomena as spirit possession (Lambek 1981, Boddy 1989). The approach may be called structural - hermeneutic insofar as it draws from both structural linguistics and poetics and seeks both interpretive m eaning and cultural grammar. For a more extensive discus­ sion of shadow theater in its cultural context see Keeler (1987).

From A. I.. Becker, “ Text-Building, Epistemology, and Aesthetics in Javanese Shadow Theatre,” in A. E. Becker and Aram A. Yengoyan, eds., The Imagination o f Reality: Essays in Southeast Asian Coherence Systems (Norw ood, N J: Ablex Publishing C orp., 1979), pp. 2 1 1 -4 3 . Abridged.

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

If aesthetics is ever to be more than a specula­ tive play, o f the genus philosophical, it will have to get down to the very arduous business o f studying the concrete process o f artistic production and appreciation. E

dw ard

S

a pir

W ords an d phrases m ust be described, often in great detail, not merely m apped on to a foreign term . T his description traditionally takes the form o f m asses o f footn otes which explain the con textual relations o f w ords, phrases, sentences, and larger units o f the text. These relation s ideally include the follow ing: 1.

Introduction: Speaking the Past and Speaking the Present In this essay I w ould like to describe som e o f the con strain ts on text-building in a language quite different from our ow n. The language is Jav an ese, the kind o f text the Javan ese shadow play, wayang kulit , as I learned to perform it from an E ast Ja v an ese puppeteer, or dalang , Ki Soedjathi D jath ikoesoem o, in daily lessons and in w atching perform an ces and discussing them together over a period o f tw o years, 1 9 6 9 -7 1 . M y goal there w as not to becom e a d alan g m yself - though th at w as necessary in order to discover w hat not to d o - but to discover how to build a text in Jav an ese, to explore w hat text-building revealed ab o u t Javan ese epistem ology, an d to learn how to respond aesthetically to a very different artistic m edium . I have studied these things and shall describe them within a p articular, evolving set o f assum ption s ab o u t w hat a text is and how it can be said to be m eaningful. T hese a ssu m p ­ tion s have their roo ts in traditional philology, m odified and expan ded by the insights o f m od em linguistics, ethnography, psychology, and Jav an e se aesthetic theory itself into w hat m ight be called a modern philology. These assu m p tion s form a partial ep istem o grap h y1 a specification o f what it is im portant to write ab o u t concerning Ja v an ese sh ad ow theater, and how one achieves coherence and co m ­ pleteness in w riting ab o u t it. As an intellectual discipline, philology can be defined as the text-centered study o f lan ­ guage. Philologists have traditionally set them ­ selves the task o f m aking ancient and foreign texts readable. O nly part o f this task is sim ple tran slation , since any careful philologist know s that few foreign w ords have translations.

207

2.

3.

4.

T he relations o f textual units to each other within the text, which establish es hierar­ chy and coherence in the text. T he relations o f textual units to other texts, since part o f the context o f any text is, m ore or less, all previous texts in a particular culture, especially texts co n sid ­ ered to be in the sam e genre; readable literature is structurally coherent with its ow n ancestors. The relations o f the units in the text to the intention o f the creators o f the text, with intention defined as the relations o f the creator to the content o f the text, the m edium , and to the hearers or readers. The relation o f textual units to nonliterary events with which units in the text estab ­ lish relations o f the sort usually called

reference. The meaning o f a text, then, is a set o f rela­ tions, by no m eans all o f which are listed above. The in form ation necessary to describe the kinds o f relations just listed m ust be know n, discovered, or reconstructed before one can know the essential m eaning o f a text, any text. For contem porary F'nglish w orks - except for the m ost esoteric or specialized literature contextual relations have been presum ed not to require philological explication for Englishspeakin g readers. H ow ever, texts w hose con ­ texts (or epistem ologies) are distant from the best-trained readers require ph ilological notes as an essential foundation for interpretation. In a m ulti-cultured w orld, a w orld o f m ultiple epistem ologies, there is need for a new ph ilolo­ gist - a specialist in con textual relations - in all areas o f know ledge in which text-building (written or oral) is a central activity: literature, history, law , m usic, politics, psych ology, trade, even w ar and peace. The specific activity o f the philologist is con textualizing conceptually distan t texts. For m any ph ilologists in the p ast that w as the only

208

ALTON L. BECKER

g o a l, an an n o tated edition o f a w ritten or oral text. Som e p h ilologists, how ever, in the course o f this activity and based upon it, have sought generalizations ab o u t the m ajor con strain ts on text-building itself, the law s o f gram m ar, poetics, n arration , etc. N o w a d ay s philology has been partition ed and distributed am on g various specialists. In the study o f literature, there has developed a gu lf between those w ho study particular texts (especially written texts) and those w ho study con strain ts on the activ­ ity o f creatin g texts: the form er is usually part o f the hum anities (literary sch olarsh ip), the latter a science (linguistics). In the study o f texts, how ever, these tw o activities correct each other, since any m ean ­ ingful activity is a conjunction o f preexisting con strain ts (or rules, or structures, or law s, or myths) with the present, the unpredictable, particular now. in this w ay a text alw ays - but to varying degrees - con textualizes the present in rhe p a st.2 O ne can roughly specify for any language activity the degree to which the speaker/w riter is speakin g the p ast or the present. R epeatin g is alm ost entirely sp eakin g the p ast, whether it be repeating som ething said a m om ent a g o , or written a m illennium a g o - a repeated rem ark, a prayer, a son g. Yet in these activities there is alw ays som eth ing o f the present, som e vari­ able o f the com m unicative act which is free to exp ress the now , be it only the voice quality o f the sp eaker, the variation s o f tem po and pitch and resonance that exp ress the repeater’s attitude ab o u t w hat he is repeating. Further­ m ore, each repetition o f a text (or bit o f a text) is in a new co n text and takes new m eaning from its con text. O ne can never wholly speak the p ast. Even in those ritual repetitions when we sp eak the p ast a s intently as possible in a kind o f tem porary trance, there is alw ays som ething o f the present com m unicated. Likew ise, one can never w holly sp eak the present. Even everyday language is highly conventional, far m ore constrained than we norm ally recognize. C on sider how sm all talk varies from lan guage to lan guage in both content and form . M o st con versation s begin with repeated conventional content which is not m eant to be discussed truthfully (i.e., in the present): H o w are you? (FLnglish); Where

are you goin g? (Javan ese); H ave you eaten yet? (Burm ese), etc. At this point in a conversation relation sh ips arc being establish ed, between speaker and hearer first, then between speaker and other - the people or things referred to. Som e con versation s never get beyond this stage, and the pace at which one m oves con ­ versationally from the conventional, predict­ able past to the present varies w idely from language to language. N otice that lan guage, in these instances and alw ays, com m unicates on at least tw o levels, the actual surface content o f the m essage (the proposition being asserted, requested, q u es­ tioned, etc.) and the relational statem ents that are conveyed sim ultaneously, m ore often by intonation, postu re, facial expression , and the like, than by direct statem ent. This relational com m unication has been called metacommumcation by G regory B ateson and others com m unication ab o u t relationsh ips, ab o u t the con text o f rhe m essage. Hence in sp eakin g the p ast, in prayer or sm all talk, to o, we are co m ­ m unicating ou r relations to the hearer and the people or things referred to in the lexically expressed m essage. Ritual language speaks the p ast on the surface, but conveys the present at the m etalinguistic level. [...] AH lan guage activity, including literature, involves, then, variation between spontaneity (present) and repetition (past) and com m uni­ cates on at least tw o levels, the lexically expressed m essage (L) and the relational m essage (Lm ). O f course the lexically expressed m essage m ay be ab o u t the relationship, in which case a new relational com m unication is conveyed, leading, if repeated, to the sorts o f linguistic involutions exp o sed by R. L). Lain g (1970). T o sum m arize, then, the analysis o f a text requires, m inim ally, that the m odern philologist describe several kinds o f relations in order to re-create a conceptually distan t context. A m inim al set o f these relations includes: 1.

The relation o f w ords, phrases, sentences, and larger units o f the text to each other (i.e., rhe coherence o f the text),

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

2.

3.

4.

The relation o f this text to other texts; the extent that it is repetition or new (sp eak ­ ing the present or the p ast), The relation o f the auth or to both the text and the hearers/readers o f the text seen from the point o f view o f the author or from the poin ts o f view o f rhe hearers/ readers (i.e., the intent of the text-builder), T he relation o f units in the text to nonliterary events (i.e., reference).

C o n text, then, includes coherence, degree o f repetition/spontaneity, intent, and reference. Sorting out the sources o f con strain ts on all these relations is a further task for the m odern p h ilologist: to w h at extent are the constraints on these relations hum an (i.e., universal to all texts)? O r are they operative only within a single lan guage fam ily or cultural tradition, or w ithin a single lan guage, or only in a specific genre, or only in the w orks o f one auth or? Any w ork is con strain ed at all these levels. The m ethodology o f this essay will be to describe, in the order just stated, the various so rts o f relations a p articular kind o f text, the Ja v an ese shadow p lay, has with its con text. I have been able to isolate som e o f the generic con strain ts on con textual relation s, and som e o f these abo ve the generic, particularly at the level o f the Jav an e se lan guage itself. By im pli­ cation , to o , I reveal som ething o f that area o f variation con strain ed only by the individual perform er (the d alan g) in a particular place an d time (A. Becker, 197 4 ; Y oun g, Becker, and Pike, 1971).

Textual Coherence in a Wayang: Plot as Symbolic Action T extu al coherence can be exam ined at any level o f structure in the hierarchy o f structures th at m ake up the text. O ne m ight exam ine the structure and categories o f w ord s in a w ayang, isolatin g the special vocabu lary and distinctive ph on ology o f the language o f the puppeteer (b asa p ad alan gan ). At the level o f sentences, and acro ss sentences, there are kinds o f coher­ ence unexploited in m ost W estern lan guages,

209

coherence based not on tense (which is the b asis o f W estern narrative coherence) but upon a system o f person (in its gram m atical sense) elaborated far beyond sim ilar system s in other lan guages. I have described this system else­ w here in relation to O ld Ja v a n e se (Kawi) literature (A. Becker an d O k a , 1976; Z u rbuch en , 1976). T hese are con strain ts which are used in building m any other kinds o f Ja v a ­ nese texts a s well. H ere I w ould like to focus on a higher-level system o f co n strain ts, a level interm ediate between the usual sentential (i.e., sentence-based) concerns o f the linguist and the glo b al concerns o f the literary sch olar, the level o f plot. T he plot o f a story or a play is a set o f con­ strain ts on the selection an d sequencing o f d ram atic episodes or m otifs. T hese con strain ts are like the rules o f a gam e, say tennis, which con strain the selection o f p ossible acts in the g am e (i.e., defining illegal acts) and the arran ge­ m ent o f acts in the gam e (i.e., defining w hat m ay not be done at certain tim es within the co n text o f the gam e). Plots, like tennis rules, d o not allow one to predict - except in very genera! term s - w hat will h appen in a play. R ather, plots tell us w hat cann ot be done appropriately . They a lso , like scientific theo­ ries, tell us one other im portan t thing; w hat the relevant variables are in the things one can d o in the play. There is no rule in tennis again st scratching my head a s m uch a s I w ant to in the course o f a gam e. There is a rule, how ever, again st serving with my feet acro ss the base line. H ead-scratch in g is, by im plication, an irrelevant variable, but foot-faultin g is co n ­ strain t on position: it tells me where 1 m ay not stan d, not where I m ust stan d. Likew ise, a set o f con strain ts on a plot specifies w hat areas o f variation are particularly relevant and what are insignificant. If I m ay borrow from a closely related m edium , m usic, we m ay note that m elodic variation is highly relevant to som e kinds o f W estern m usic, but rather insig­ nificant in som e kinds o f Ja v a n e se m usic. An A m erican w ho is lookin g for m elodic variation in gam elan m usic will be b o red ; a Javan ese look in g for dense m usical texture in a sym ­ phony will a lso be bored. Likew ise in dram a, an A m erican w ho seeks ch aracter developm ent in w ay an g is going to be disapp oin ted in all

210

ALTON L. BECKER

but a very few w ay an g stories, and a Javan ese w ho seeks com p lex coincidences in all but a few A m erican m ovies (those few being com e­ dies, like the M a r x Brothers’ Animal C rackers) is also goin g to be d isappoin ted. Plot (i.e., con strain ts on the selection and arrangem ent o f dram atic episodes) includes con strain ts on the kinds o f variation that are relevant. For the m ost p art, in m ost cultures, kn ow l­ edge o f plot con strain ts is unstated background know ledge, like the know ledge o f gram m ar and syntax. It is learned indirectly, first through fairy tales and nursery rhym es (and their equivalents in other cultures), and then from the various m edia that have access to children. Som e G reeks, how ever, were self-conscious about plots. A ristotle’s Poetics includes a description o f plot which still holds for m ost W estern d ram a and narrative. A ristotle calls plot fable: “ The im itation o f the action is the fa b le ,” he writes. “ By fable I now m ean the contexture o f incidents, or the p lo t.” He lists the six m ajor variables in a dram a: 1. 2. 3. 4. 5. 6.

fable or p lot, m anners or ch aracter, diction o r m etrical com position , sentim ents or speeches, decoration , m usic.

A ristotle continues, “ O f all these parts the m ost im portant is the com bin ation o f incidents or the fab le .” Fable or plot is m ost im portant because it im itates w hat A ristotle held w as the m ost im portan t referential content o f the d ram a, action (or im itation o f action). A m ong the con strain ts on plot which A ristotle lists are the follow ing. N ote that they are all phrased negatively - i.e., as constraints. 1.

A proper fable m ust not be incom plete: “ The poet w ho w ould construct his fable properly is not at liberty to begin and end where he pleases. . . . ” A fable should, he exp lain s, su p p ose nothing to precede it, and require nothing to follow it (book 2, chapter IV). C om pleteness here is co m ­ pleteness o f linear (i.e., tem poral) c a u sa l­ ity, a pow erful con strain t on selection and arrangem ent.

2.

3.

4.

Coincidences are to be avoided. Sequences should follow as prob ab le and necessary consequences. N oth in g im probable should be adm itted, o r, if necessary, it should arise out o f the fable. Perhaps A ristotle’s m ost fam ous com m ent on p lo t m akes just this point: “ Im possibilities, rendered prob ab le, are preferable to things im probable, though p o ssib le ” (book 2, chapter VI; book 4 , chapter VI). N o part o f a proper fable m ay be tran s­ posed or om itted w ithout destroying the w hole. Anything th at can be left out, should be (book 2 , chapter V). A gain there is em ph asis on linear (tem poral-causal) sequence. The time in the text should not be m ore than a single day (book 1, chapter IX).

These basic constrain ts all have to d o with unity and causality, above all with tem poral unity and linear causality - tw o aspects o f the sam e thing. All o f them are rooted in the sim ple fact that intersentence coherence in Indo-European lan guages is achieved p rim ar­ ily by tense. C larity and coherence means to speakers o f these languages linear tem poral/ causal sequencing. Tense is seen as iconic: that is, past, present, and future are taken as facts ab o u t the w orld, rather than facts ab o u t language. Tense is not iconic in all Ianguagecultures and hence tem poral-causal linearity is not the m ajor con straint on textual coherence in all lan gu ages.3 The linearity o f A ristotle’s constraints can be stated in another w ay. If m eaning com es from tem poral-causal sequences, then epistem ologies d o not, and can n ot, change from episode to episode, or, a s stated in a recent study o f plot, “ Sem antically stan dard universes alw ays have consistency in the interpretation o f several connected am b igu o u s e p isod es” (H ahn, 1973:8). T h at is, Ja y G atsb y, G odzilla, A gam em non, Jo h n W ayne, and Charlie C haplin d o not and m ay not ap p ear in the sam e plot. W hat em erges in the episodes o f W estern serious dram a are the disam b igu atin g causes o f actions. T hese cau ses are at base represented as character defects, often m inor ones. The

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

episodes lead to a catastroph e and a clim ax, a reversal o f expectations, all o f which leads on to the end o f the causal chain. N early all these con strain ts are violated by w ayan g plot structure. It is not that w ayan g plots m ay not have tem poral unity, causal linear sequences, catastrop h es, reversals and all the rest. These do ap p ear in w ayang plots, particularly in those plots m ost adm ired by W estern view ers, such as the plot o f Dewa Ruci , a linear search for the w ater o f im m ortality, or the plot o f the sim plified and shortened versions o f the R am ay an a, a search for a stolen wife. These A ristotelian con strain ts, how ever, are not necessary to a goo d w ayang plot, and to focus on, for instance, causal sequences and character devel­ opm ent is to m iss the area o f relevant variation in w ay an g theater and to m iss the subtlety and depth o f good w ayang. W ayang plots are built prim arily around coincidence, a w ord which we in the W est use to exp lain aw ay things o f no m eaning. “ A mere coincidence” can n ot, in the W est, sustain prolonged scrutiny and analysis. In w ayan g theater coincidence m otivates actions. There is no cau sal reason that A rjuna, the frail w ayang hero, m eets C ak il, a sm all dem on, in the forest, as he (or a counterpart) does in each w ayang. It is a coincidence; it h appens (jadi), and because they are w ho they are, they fight and C akil dies, but not forever; he will be killed over an d over again in each w ayang. When A rjuna and C akil m eet, tw o w orlds, tw o epistem ologies coincide for a m om ent, C ak il is purely physical. He attack s A rjuna because A rjuna m akes him un com fortable. A rjun a’s m editation has raised the heat o f rhe forest higher than the creatures w ho live there can bear. C akil responds instinctively to this therm al pollution. O n the other hand, A rjuna attack s C akil because he recognizes him a s evil (i.e., other), not because o f anything he has done, but because he know s - by thought, nor instinct - that it is his duty (dharm a) to com b at evil. H e kills coolly, disp assion ately, the passio n ate C ak il, w ho is defending his forest hom e again st the intruder. A rjuna con trols nature by killing it, but it renews itself again and again . There arc other interpretations o f this m otif. N ot every observer o f w ayang will

2 11

agree with this interpretation o f it. It does seem evident, how ever, that A rjuna and C akil live in different conceptual w orlds and that their m eeting is not caused but is rather an accident, a coincidence o f these w orlds. N oth ing in the prior events o f the text nor in the succeeding events m ade it a necessary part o f the plot in A ristotle’s term s. Yet this m otif is neces­ sary (obligatory) within rhe con strain ts o f w ayan g plot. T h is is but one coincidence, one intersection in the interw oven, cyclic action s that inform a w ayan g plot - unm otivated, unresolved, m ean­ ingless within a chain o f cau ses and effects, but sym bolically very rich. The nam e for a w ayan g plot derives from the root laku “ step or acr” plus the suffix -an , which norm alizes the roo t, giving us, by vowel sandhi, lakon “ an action, a w ay, an event, a p lo t.” A lakon includes three m ajor divisions, within each o f which a certain range o f voice pitches and a valuing o f particular pitches is m aintained, and within each o f which there is a prescribed internal structure. These divi­ sion s, called pathet , include com bination s o f scenes called jejer (audience scenes, before a ruler or holy m an), adegan (scenes outside the audience hall, e.g., adegan wana “ forest a d e g a n ,” adegan gapuran “ gate ad eg an ,” adegan gara-gara “ turm oil a d e g a n ,” etc.), and perang (battle). The m eanings o f these nam es for the p arts o f a w ayang are richly m eta­ phoric. U nderstanding them as w ords helps to contextualize them within the Ja v an ese sem an­ tic w orld. Jejer also m eans “ w hat e x ists,” “ the subject o f a sentence,” and “ the handle o f a k ris,” as well as “ an audience before the K in g.” Adegan , the scenes outside the audience hall, m eans a lso , “ propped u p ,” “ stan d in g,” “ d o o r­ fram e” and “ the punctuation o f a sentence” (tw o vertical, parallel lines). The linguistic m etaphors suggest a parad igm atic, associational link between sentences and plays, within w hose structures experience is shaped and expressed. A w ayan g plot, then, is built hierarchically in structures m ade up o f three basic units. A lakon (event) is divided into three pathetan or acts, each with the sam e internal structure. Each pathet is m ade o f three basic scenes: (1) the jejer (static audience in a court or a

212

ALTON L. BECKER

herm itage where a problem arises and a plan is form ed); follow ed by (2) tw o or m ore adegan derived from that audience, and alw ays involv­ ing a journey aw ay from the audience place; (3) a perang (battle) at the end o f the journey.

Each scene, in turn, has three basic co m p o ­ nents: (1) description o f a situation (either janturan “ description o f a place” or carios “ description o f prior a ctio n ” ); (2) dialogue (ginem ); (3) action (sabetan). The minimal structure o f a play - or an event is:

LAKON (EVENT)

Pathe Jejer /

I \

12

3

?m

Adegan /

I \

12

3

Pathet Sanga Adegan

/ I\

12

3

Perang

/ I\

12 3

Pathet Manjura Jejer

/ I\

1 2 3

Adegan

/ I\

1 2 3

Peranq

/ I\

1 2 3

(Key: 1 = description, 2 = dialogue, 3 = action or motion)

Any given w ayan g allow s for three basic o p e r­ ation s on this m inim al structure: perm utation (reordering units below the level o f pathet ), conjunction (repeating units below pathet ), and em beddin g (putting units within units below the level o f pathet). Pathet structure w as fixed in the tradition I studied. I...] As night goes on, different parts o f the scenes are foregrou nded, that is, som e parts are shortened, others prolonged. In the first act {pathet item) description usually takes m ore time than d ialogue and action com bin ed. In the second act, d ialogue - m ostly jo kes, but also very heavy spiritual instruction from a holy m an - is foregrounded. In the third and final act (pathet manyura)> action - usually battle - predom inates. When one p art o f a scene is d om in an t, how ever, the other tw o alw ays ap p ear, albeit often briefly. O ne m ay notice that in describing the stru c­ ture o f a d ram atic event, the w ords used are all Ja v an ese not Sanskrit. T hough stories are often im ported into w ayan g, chiefly from Sanskrit epics, p lots ap p ear to be uniquely Jav an ese. H avin g seen som ething o f the sequencing o f events or m otifs in a w ayan g play, let us turn now to the parad igm o f events them selves and the kinds o f coherence that ap p ear w ithin the structure that has been described. It is often very difficult for the viewer, foreign or Ja v a n e se , to know just where he is in the story

being presented, i.e., in know ing that polarities between p rotagon ists an d an tagon ists are being established. O ne alw ays know s, how ever, where one is in the plot - the structure defined earlier. The story m ay be very obscure, much o f the action m ay take place o ff the screen or be assum ed by the dalan g to be well know n, and there m ay be all sorts o f loose ends left after the plot cycle has finished. It is prim arily the clow ns w ho try to tell the audience w hat is happening. Certainly little o f the motivation for action ap p ears in the plot. The clow ns, using m odern lan guage, m odern ideas, and m odern behavior, step am o n g the heroes and dem ons and g o d s like w ideaw ake men in a dream w orld. They bring the present into the story (i.e., they alw ays speak the present), and with the p a rad o x o f forethough t, contextualize the present w ithin the tradition, ch anging both, as usually seem s to happen when epistem ologies are allow ed to coincide. In the coincidence o f epistem ologies, a s just noted, the real subtlety o f w ayang app ears. The m ajor epistem ologies are (1) that o f the dem on s, the direct sensual epistem ology o f raw nature, (2) that o f the an cestor heroes, the stratified, feudal epistem ology o f traditional Ja v a , (3) that o f the ancient g o d s, a distan t cosm ological epistem ology o f pure pow er, (4) that o f the clow ns, a m odern, pragm atic epistem ology o f personal survival. All these epistem ologies coexist in a single w ayang, and others m ay be added (m ost usually the epistem ology o f the

TEXT-BUILDING, EPISTEM O LOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

Islam ic sain ts, that o f the m odern m ilitary, or th at o f som e strange foreign land where one o f the clow ns goes to be king, like G ulliver am on g the Lilliputians). Between each o f these episte­ m ologies there m ay be - an d usually is - a con ­ frontation and a perang , a battle. N o one ever wins conclusively, but rather a proper balance is restored. Each epistem ology, each category o f being, exists within a different concept o f tim e, and all the tim es occur sim ultaneously. T h at is, nature tim e, an cestor tim e, god time, and the present are all equally relevant in an event, though for each the scope o f an event is different. T h rough out the w ayan g, each is kept distinct, even in language (which will be discu ssed later). T he con strain ts on w ayan g plot sustain the notion o f m ultiple time and m ultiple epistem ology. The differences with the A ristotelian notion o f plot should now be ap p aren t. W hat in the w ayan g plot are significant coincidences, in the A ristotelian plot are crudities, violations o f the basic notions o f unity and causality. In w ayan g, we m ight say that G atsb y , G o d zilla, A gam em non, Jo h n W ayne, and Charlie C h aplin - or their coun terparts - d o appear in the sam e plot, and that is w hat causes the excitem ent; that clash o f conceptual universes is w hat im pels the action. A s far as I kn ow , the w ayan g tradition h as no A ristotle, no one w ho has attem pted to articulate the set o f con strain ts which underlie the tradition. I can n ot, as an o u t­ sider, d o this with any depth or hope o f a d e ­ quacy. I am not even sure that in Javan ese eyes it is w orth doin g, but the sym m etry o f this essay, the plot we are caugh t within at this m om ent, seem s to dem and it. A w ayang plot, then, seem s to be con strain ed in the fo l­ low ing w ays, all stated by contrast to A risto­ telian constraints. 1.

A w ayan g plot can begin at any point in a story. It has no tem poral beginning, m iddle, or end. Indeed, a w ayan g plot is very sim ilar to a piece o f traditional Ja v a ­ nese m usic, in which a m usical pattern is expan ded from w ithin, producin g layer upon layer o f pattern m oving at different times. A w ayan g p lot, how ever, m ust begin and end in certain places; it can n ot begin

2.

3.

213

and end anyw here, though it can begin and end anytim e. It m ust a lso p a ss through a certain place in the m iddle. T h u s w ayang plot has a sp atial, rather than tem poral, beginning, m iddle, and end. It m ust begin and end in a court, the first the court o f the an tago n ists, the last the court o f the p ro tag o n ists (to use the G reek agon term i­ n ology, which seem s ap p ro p riate here). T he m iddle section m ust be in nature, usually in the forest on a m ountain, but som etim es, to o, in or beside the sea. It is m ovem ent out an d back, a trip. This struc­ ture m ay well reflect the origin o f w ayang a s an instrum ent o f com m unication with the dead via trance (R assers, 1959). Like an A ristotelian plot, a w ayan g m ust not be incom plete, but incom plete­ ness is not tem poral or ca u sa l, but rather sp atial. C oincidences, far from being avoided, im pel action, for they induce cognitive puzzles or p arad o x es. Coincidences are the w ay things h appen, and the way co m ­ m unication between unlikes occurs. In Ja v a n e se and Indonesian, the w ord used to describe w hat we call a coincidence (a causeless interaction) is kebetulan (or kebenaran ), literally a “ tru th ” (an ab stract noun derived from the adjective betull benar m eaning “ tru e” ). T here are m any related term s (e.g., dadi “ happen, b eco m e,” cocok “ com e together, fit” ) which m ake up a sem antic set used to describe events none o f which imply linear causality. Likew ise, a piece o f m usic is structured by the coincidence o f gon gs occurring together, and a holy day by the coincidence o f sim ultan eous calendrical cycles. Any scene in a w ayan g plot m ay be tran s­ posed or om itted, except for the con strain t that the plot begin in a court, have its center in nature, and return to the court. T ran sp o sitio n s and om ission s o f story m aterial d o not destroy or even change the w hole. A lm ost anything can be left ou t or brough t in. When som ething is brought in, how ever, it m ust follow the p arad igm atic and syntagm atic con strain ts o f the lakon structure described a b o v e . . . .

214

4.

ALTON L. BECKER

There are further con strain ts on the sets o f ch aracters (dem ons, heroes, g o d s, clow ns) in relation to one another (e.g., how they speak and how they m ove) which will be described later. All o f this m akes up w hat m ight be called the gram m ar o f a w ayan g plot. A ristotle su ggests that the time o f a serious d ram a should nor be m ore than a single day. He m eant the time enacted within the plot on stage, not the w hole story. H ere is his m ost stringent constraint on tem poral unity, one not alw ays follow ed by W estern playw rights but rather held as an ideal, even by such m odern Am erican d ram atists as O ’N eill, M iller, or Albee. Indeed, it m ay be one o f the reason s for identifying these as g o o d , serious playw rights in the W estern tradition. The time enacted within w ay an g is unconstrained, except that it m ust be m ul­ tiple. Coincidences are tim eless. But, the perform ance time o f a w ayan g is sym boli­ cally a single day. It is necessary to explain this rather strange phenom enon. T he divi­ sion o f scenes is m arked by a large im age o f a tree (or a m ountain) called a kayon (or gunungan). D uring the play, which is usually perform ed at night, the kayon m arks the im aginary progression o f the sun from east to w est by the angle at which it is set again st the screen (which is properly set up on an east-w est a x is, or if necessary, north-south, in which case north su b sti­ tutes for east). T he kayon is a dram atic clock which m arks only the p rogression , o f the p lot , not the tim es in the story or the time on the w ristw atches o f the viewers.

These are but a few o f the features which define the coherence o f a w ayan g plot, p articu ­ larly those few which con trast m ost sharply with A ristotle, w hose w ritings about plot well define the unconscious con strain ts on plot that m ost o f us in the W est have ab sorb ed since childhood. I now turn from discussion o f the structure or coherence o f a w ayan g to con ­ sideration o f the relation s o f the text with its con text, from inner to outer relations, with a full aw aren ess that there is m uch m ore to be said , particularly at m ore technically linguistic

levels o f focus, ab o u t Javan ese textual coh er­ ence in general, and w ayan g coherence in particular.

Text within Text: The Javanese Art of Invention The distinction between story and plot is very im portant in studying the structure and devel­ opm ent o f a w ayan g text. The plot has been defined a s a set o f con straints on the selecting and ordering o f episodes or m otifs. The story is a prior text, fictitious or factual or both, which is the source o f these episodes or m otifs; it is a prior text to som e degree know n by the audience. Literature, in this sense, is m ostly ab o u t prior literature. For exam ple, in our ow n tradition any cow b oy m ovie tells the story o f the past m ore in the sense that it repeats episodes and ch aracters o f previous cow boy m ovies and novels than that it recounts “ real” events th at occurred in the A m erican W est. The “ tru th ” o f a cow boy m ovie is m uch m ore a m atter o f its correspondence with a m ythology (a body o f prior literature) than with any events recognizable by nonfiction cow boys in their ow n experiences. W ayang has reference to a m ythology accessible to us in O ld Javan ese or Sanskrit lit­ erature, prim arily the tw o great epics, the R am ayan a and the M ah ab h arata. Jav an ese, o f course, have access to this m ythology in m any less literary w ays: in nam es o f people and places, in other theatrical perform ances and oral literature, in com ics, in the very language itself (R esink, 1 9 7 5 ; A nderson, 1965; Emmerson, forthcom ing). A w ayan g plot, how ever, need not draw on this m ythology, though it alm ost alw ays does. T h at is, a dalan g m ay well turn to Islam ic or Christian or autocthon ous Javan ese m ythology, wholly or in part, as a source o f the m otifs and ch aracters for his per­ form ance, and he can do so w ithout violating any o f the plot con strain ts discussed earlier. The story, w hatever its source, provides content and context for the plot. T o introduce A rjuna, the hero o f the M ah ab h arata, as a character into a particular plot establishes as a context for that particular plot all the prior texts (m ythology), oral or written, related to

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

A rjuna. A rjuna has done certain things, relates in certain w ays to other ch aracters, and is associated with m any details o f appearance, dress, behavior, speech, etc., which have been established in prior texts (A nderson, 1965; H ard jo w irogo, 196 8 ; K ats, 1923). W hat h appens to A rjuna in a p articular plot m ay either repeat episodes from prior texts or it m ay be new, although consistent with prior texts. The new creation fills in m ore details o f the grow in g text or m ythology related to A rjuna, new episodes in his life, only hinted at previously, or a return to the w orld by Arjuna across tim e, into, for instance, an ancient Ja v a ­ nese court. The A rjuna m ythology (or R am a m ythology, or H an um an m ythology, etc.) is a living expan din g text in Ja v a . T w o exam ples o f this sort o f text expan sion m ay help to m ake this process o f invention clearer. D uring the Indonesian national elections in 197 1 , one d alan g w ho supported the incum ­ bent m ilitary governm ent created a w ayang in which K rish na, when he realizes that he m ust direct the Pandaw a arm ies (the arm ies o f A rjuna an d his brothers Y udistira, Biam, N ak u la, Sad ew a, and their allies), in the great w ar o f the B h arata, seeks out the old clownservant Sem ar. K rishna ask s Sem ar w hat he should d o and how he should behave a s a m ilitary leader. T hen, in the center o f the play, in the forest, Sem ar instructs Krishna in his duty, the com m on m an in an era o f dem ocracy instructing the ruler. The text for these instruc­ tions w as the Sapta M arga , the official C ode o f Behavior for m odern Indonesian soldiers. T h is brilliant new story, Bagaw an Ismojo Sandi , conceived by Ki H ari Puribadi, very deftly con textualizes p ast in present and present in p ast sim ultaneously; the Sapta M arg a is sanctified as a m odern Bhagavad G ita, and the ancient m ythology is given rich current relevance. Another kind o f invention involves no overt innovation at all, but rather lets the audience infer the connection with current events. This second exam ple w as perform ed in 1 9 7 1 , too, but this time by a dalan g op p o sed to the m ili­ tary governm ent, a supporter o f rhe PN I, the political party associated with form er Presi­ dent Soekarno. This d alan g perform ed the old text nearly w ithout change, except that the

215

clow ns did say they were volunteer w orkers at the PNI parry headquarters and m ade several jokes ab o u t cam paign activities on a day-today level. The story w as Kangsa Adu Ja g o y a traditional Sanskrit story o f a pow erful villain (K an gsa/K am sa) w ho usurps K rish n a’s kingdom and drives Krishna into the forest. Krishna seeks the aid o f his cousin s, the Pand a w as, particularly Bim a, in defeating and driving out the pow erful K an gsa. N o one m issed the political statem ent. It is interesting to note that in those n ation al elections the m ost pow erful public statem ents again st the governm ent were m ade by d alan g s, using just rhis technique. Kvery other m edium o f com m unication, including other form s o f theater, w as noncrirical. It is also interesting thar tw o sides, the governm ent and the PNI, recognized the sam e m ythological con text; the difference lay in whether Krishna represented the m odern K satria or the deposed king. O n e o f the m ost im portant differences between traditional artistic expression and m odern individualistic artistic expression is that in a traditional m edium the artist is con­ sciously expan din g a prior text, an open corpus o f literature, art, or m usic, w hereas an artist w hose intent is self-expression creates and develops his ow n text, his ow n m ythology, so far a s he can and still com m unicate. When an artist can no longer w ork w ithin the inherited m ythology and plot con straints, he seeks new m ythology and constrain ts, often from his ow n im agination, and he w orks in alienation from his ow n society. T h is sam e distinction a p p ears to have been m ade by Levi-Strauss, this time in distinguishing the sham an and the psychoanalyst: . . . the shamanistic cure seems to he the exact counterpart to the psychoanalytic cure, but with an inversion o f all the elements. Both cures aim at inducing an experience, and both succeed by recreating a myth which the patient has to live or relive. But in one case, the patient constructs an individual myth with elements drawn from his past; in the other case, the patient receives from the outside a social myth which docs not correspond to a former personal state. T o prepare for the abreaction, which then becomes an “ adreae-

216

ALTON L. BECKER

tion,” the psychoanalyst listens, whereas the shaman speaks. Better still: When a transfer­ ence is established, the patient puts words into the mouth o f the psychoanalyst by attributing to him alleged feelings and intentions; in the incantation, the sham an speaks for his patient. {Levi-Strauss, 1963) N o d alan g is in this sense a m odern artist (or psych oan alyst). It is as if he were perform ­ ing a new act o f H am let, or relating a new episode from the G o sp els, w orkin g on an expan din g text which extends through space and time far beyond his ow n im agination. In this kind o f traditional creation, the skill o f the dalan g is revealed in his ability to re-create the past, which he m ust d o at the beginning of each w ay an g an d at certain points throughout the perform an ce, m ost particularly in singing short descriptive p assage s from O ld Javan ese (Kaw i) texts. H ere he sp eaks directly to the past o f his ow n culture in w ords alm ost entirely unintelligible to the d alan g or his audience. !• • -I The art o f invention for the d alan g, w orking within the plot con strain ts o f his m edium , involves selection o f m otifs and ch aracters from the body o f m ythology he believes in. This is not unlike the A ristotelian art o f inven­ tion, which w as prim arily the selection o f q u o ­ tation s an d ideas from the classics - a kind of inform ation retrieval - in order to interpret the present (Y o un g and A. Becker, 1966). A political change in Indonesia can be reflected in w ay an g as a change in m ythology, as it has been described in the penetrating studies o f D on ald Kmmerson (forthcom ing), and Benedict A nderson (1 9 6 5 ), and Ci. J. Resink (1 9 7 5 ). O ne generation o f heroes m ay replace an oth er, or one set o f g o d s m ay replace another, a s w as the case in a village w ay an g I saw in L o m b ok in which the H indu g o d s were the villains w ho were defeated by M oslem heroes. T h is is, how ever, essentially new wine in old bottles, or w hat we m ight call surface change. D eep change, in term s o f this essay, w ould be change in the plot, change in the con strain ts on selecting and orderin g the ch ar­ acters an d m otifs. Deep change w ould be change in the Ja v an ese conception o f tim e and event, change o f epistem ology.4

Intentionality in a Text: The Uses of Texture O ne o f the first things a d alan g learns is that not everyone will respond to a w ayang in the sam e w ay. There is no assum ption that every* one will be interested in the sam e things at the sam e time; som eone will alw ays be dozing. The setting for a w ayan g is noncom pulsive, m ore like a W estern sports event than serious theater. It is not sham eful or em b arrassin g to sleep through w hat som eone else is enjoying. Jo k e s, ph ilosoph y, action, poetic lan guage, each has different appeal to different people, depending on their ow n m ental m akeup, which is often described in a w ay parallel to the Indian theory o f rasa and guna (C o om arasw am y, 1957), a theory parallel, in turn, to the arch aic theory o f hum ors in the W est. O ne responds accordin g to his m akeup. There can be no single, intended correct response to a play, no one com plete interpretation. This m ultiplicity o f events and perspectives builds the kind o f thick texture that Javan ese favor. As an old m an responded when ask ed why he liked w ayan g, “ A salnya ra m a i!” (“ A bove all because it is bustling/com plex/busy/beauti­ fu l!” ) Ram ai < O ld Ja v an ese ramia < Sanskrit ramya “ pleasin g, beau tifu l.” N otice the sem an­ tic change in Ja v a from “ beautifu l” to “ beauti­ ful because bustling and co m p lex .” Sanskrit w ords, like Sanskrit stories, are recontextualized in Ja v a . W ithin the variety o f responses - to o thick to be untangled here - there are alw ays tw o separate audiences at every w ayang, an essen­ tial audience, w ithout w hom the play is poin t­ less, an d a nonessential audience, w ho m ay or m ay not be present and w ho in som e sense overh ear m uch o f the dram a. It is the nones­ sential audience that we have described so far, the various people w ho have various responses to a noncom pulsive event, which is noncom ­ pulsive precisely because they are the nones­ sential audience. T he essential audience o f a w ay an g is n or­ m ally unseen: spirits, dem ons and creatures, g o d s, an d an cestors. T o w hom docs the dalan g speak in O ld Jav an e se and Sanskrit if not to those w ho understand these lan guages, which

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

are unintelligible to the nonessential audience? A rchaic language is not merely em bellishm ent or m ystification, else it w ould have been lost long ago . R ather it is essential language addressed to the essential audience, the ancients, the dead. All d ram a, as we have noted, sp eak s from the p ast, the unseen sources o f pow er which are the w idest co n text o f the play. The first w ords o f a w ay an g - prior even to the lakon itself - are uttered softly to unseen hearers, “ p rayers” or m antra to the sources o f pow er. Before the puppeteer arrives at the place o f perform ance he establish es relations with this w ider spiritual con text, including his ow n , nonhum an brothers (kan da em pat) w ho gu ard and extend his senses and provide buffers in an un predictable, often hostile envi­ ronm ent (H o o y k aas, 1 9 7 4 ) . . . . T h is initial ph rase o f a w ayan g text is called the m anggala in O ld Javan ese (K aw i) w ritten literature. A m anggala is anything - w ord , g o d , or person - which h as the pow er to sup p ort the poet. The m anggala is invoked, p raised , and then relied upon to sustain the poet/dalan g in his effort. H ere is a point o f choice, then, for the puppeteer, w ho is likely to turn his mind to several sources o f sup p ort. For the dalan g, unlike the poet in the kakaw in (poetic litera­ ture o f the O ld Ja v an ese period), it is a private act, invoking the w idest context o f the shadow play, the earth, the light, the w ind, the m ountains. The lan guage o f the m an ggala-prayers is usually a single expan ded sentence which includes a descriptive subject and an im pera­ tive predicate. The sentence is preceded by the original syllable, O ra, which establishes the param eters o f all language sou n d s, in Sanskrit linguistics. In structure, the m anggala is very sim ilar to a Vedic hymn: O m . (V (insert nam e o f the m anggala and ph rases describing him/her/it) + im perative predicate. For exam ple, as he a d ju sts the lam p (kero­ sene or oil), the d alan g m ay softly say: “ O m . Be there no hindrance. Ciod o f spirit, cenrer o f ail, G o d o f light - let the flam e o f this lam p illum ine the w o rld .” The ph rases o f the prayer linguistically are parallel. All prayers follow the general pattern just given, except that the

217

phase “ Be there no h in drance” is not alw ays stated. The language is a blend o f Sanskrit and Jav an e se , the subject in San sk rit (the language o f the g o d s, the rem ote p ast), and the predicate in m odern Ja v an ese (the lan guage o f the dalan g him self, the im m ediate present). T he w ords bridge past and present, and m ust be uttered with full attention. Perhaps here is the place to note an e x trao r­ dinary fact ab o u t the language o f the w ayang, a fact o f great im portance in understanding w hat is h appening at any given m om ent. A w ayan g includes within it, in each p erfor­ m ance, the entire history o f the literary lan­ g u age, from O ld Ja v a n e se , pre-H indu incantation an d m ythology to the era o f the San skrit g o d s and their lan guage, blending with Ja v an ese in the w orks o f ancient poets (the sulu ks), addin g A rabic an d C olon ial ele­ m ents, changing with the pow er o f Ja v a to new location s and dialects, up to the present Bahasa Indonesia and even a bit o f A m erican Flnglish (in which one clow n often instructs another). 1 do not just m ean here w hat m ight be said o f F^nglish, that it reflects its history in v o cab u ­ lary, syn tax, and phonological variation. T h at is a lso true o f m odern Javan ese. The difference is that in the shadow play, the lan guage o f each o f these different eras is separate in func­ tion from the others; certain voices speak only one o r the other o f these lan gu ages and d ia­ lects, and they are continually kept alm ost entirely separate from each other. O ne could even say that the content o f the w ay an g is the lan gu ages o f the past an d the present, a m eans for con textualizin g the past in the present, and the present in the past, hence preserving the exp an d in g text that is the culture. I shall point o u t these different kinds o f lan guage a s they ap p ear, though we have already seen that the prayers (m antra) to the g o d s and other sources o f pow er use San skrit and m odern Jav an ese, the suluk use O ld Ja v an ese (K aw i), and the clow ns use all m odern lan gu ages, Jav an ese, Indonesian, D utch, F'nglish, Ja p a n e se , French, neatly reflecting the context o f m odern Indo­ nesia. C low n s speak the older lan guages only to m ock them. Like the m an ggala-prayers, the suluk sp eak to the ancients (not the g o d s but the Ja v a n e se an cestors) in their ow n language at

218

ALTON L. BECKER

the beginning o f each scene. In m any cases rhe chanted suluk are addressed to the individual ch aracters represented by the puppets in the w ayang. Like Vedic hymns they invoke the character in his ow n language by a kind o f w ord m agic, in which to state a thing properly and effectively, even w ithout intent (as in a casual Brahm in’s curse, which can n ot be revoked), is to effect pow er in the w orld, bridging time and space. It is here that w ayan g becom es an education in pow er. W ayan g teaches men ab o u t their w idest, m ost com plete con text, and it is itself the m ost effective w ay to learn ab o u t that context. There has been much written about the m ystical com m unication in w ayan g, and its details are best left to Javan ese them selves to w rite ab o u t. For us in the W est it m ight be called trance-communication. The d alan g is above all a m an w ho can be “ en tered,” a “ m edium ,” though to use our ow n term inol­ ogy is to invoke all the w rong associatio n s. Trance speaking can be defined as com m u nica­ tion in which one o f the variables o f the speech act (I am speakin g to you ab o u t a: at time y in place z with intent a) is denied, m ost fre­ quently the variable I is p arad o xically both speaking and not speaking, or speakin g in vol­ untarily or nonintentionally. T ran ce is a kind o f incongruence between statem ent and intent (I/not I am sp eakin g to you/not you. . . . ) , and covers a w ide spectrum o f linguistic exp eri­ ences, from the m inor trance o f singing the national anthem - or any son g you believe - to the m ajor trance o f hypnosis and schizophre­ nia (H aley, 1963). In any case, it is a s trance com m unication as a m eans o f relationship with an unseen, essen ­ tial audience that w ayan g can be linked to the Barong d ram a o f Bali, the autoch th on ous trance ritual o f the other islands (e.g., the m a’ bugi in Sulaw esi), and the use o f puppets and dolls as spirit m edia throughout Southeast Asia. W hat is the use o f com m unication with the ancients, besides preserving the text o f the culture, which is p rob ab ly not a prim ary goal but a con stan t effect o f this com m unication ? T w o uses are im plied in the instruction book s for the d alan g: to exorcise danger or potential danger, and to contextualize the present in the

past. There are m any well-know n m yths ab o u t the origin o f w ay an g a s a w ay o f subduin g or at least calm ing dow n dan gerou s pow er, the pow er o f Siva am uck or the pow er o f his dem on son K ala (time) w ho form erly dealt out death indiscrim inately. H ow does w ay an g control pow er gone am uck, m adn ess, dem on s, disease, and stu pid­ ity? By nature all these are sources o f chainreacting, linear pow er, which accelerates by repeating m ore and m ore o f the sam e. Som eone w ho is am uck kills and kills without intent until he in turn is killed. Likew ise disease and m adness feeds upon itself. The closest answ er to the question o f how w ayan g subdues pow er gone am uck cam e to me from a Balinese friend, w ho answ ered, “ Y ou know , it’s like the d oo rs in B ali.” ( N O T E : an entrance in Bali and tra d i­ tional Ja v a is backed by a flat wall or screen (Javanese wrana) a few feet behind the entrance g ap in the outer w all, so that one cannot g o straight in but m ust p ass right or left. D em ons and people p ossessed or am uck m ove in straight lines, not in curves like norm al hum an beings.) M y friend continued, after I looked m ystified, “ The dem ons can ’t get in. The m usic and shadow play m ove round and round and keep the dem ons o u t.” Then he paused and laughed heartily, and added, “ As you m ight say, dem ons think in straight lin es!” Clearly, from this point o f view, it is not the story or the archaic w ords or the puppets but the whole thing, the texture itself, the m aze o f relations, that is m ost im portant. The structure o f the m edium itself subdues pow er gone am uck, inducing p a rad o x and coincidence, anathem a to those w ho think in straight lines. In sum m ary, then, the dalan g speaks as him self and as the p ast through him self to an unseen, essential audience and to the im m edi­ ate, nonessential audience, each containing a wide variety o f perspectives on the action being perform ed. And he is playing with fire. If, for the im m ediate audience, the event is noncom pulsive, for him, it is pow erfully co m ­ pulsive. O nce begun, he m ay not for any reason (illness, storm , violence, pow er failure) stop until the play has finished. H ence, he m ust be careful not to begin anything he cann ot end.5

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

R eference: O n L an g u a g e and Th in g s o f T h is W orld [■■■I In the dom inant W estern notion o f reference (the one assum ed in introductory and pop u lar bo ok s ab o u t linguistics), there are three cate­ gories which can be labeled roughly words (language), thoughts (or con cepts), and things (objects in the sensible w orld). These are assum ed to be separable (though slightly over­ lapping) categories o f being, since concepts ap p ear to be stateable in different languages, and there appear to be different, unrelated nam es for the sam e things in different lan­ guages. The relations o f language to concepts and things are therefore felt to be fundam en­ tally arbitrary. If anything, natural language gets in the w ay o f clearly seeing things as they are (B acon ’s “ idols o f the m ark et” ), and gets in the w ay o f clear, logical thought (based as it is for us now on m easurable identities and differences). Thinkers in the W est tend to give priority to concepts or things and treat lan­ guage as a “ to o l” to be shaped to our ends, or discarded and replaced. N o t for m any centu­ ries in the W est (until recently in the w orks o f Foucault and Lacan and with the developm ent o f m odern linguistics) has language itself been given priority as a source o f highly valued know ledge. O p p osed to this notion o f the arbitrary n ature o f reference is one fam iliar in Am erican thought in the w ork o f Em erson, particularly in his essay “ L a n g u a g e .” In this earlier view, the relation o f w ords, thoughts, and things is n ot arbitrary, though it has been confused by the m ultiplicity o f languages. The law s o f N atu re govern thoughts, w ords, and things alike. Em erson could , therefore, m ake his essay “ L an g u age ” a subsection o f his larger w ork, Nature. Signified an d signifier are con ­ strained by the sam e law s. T o know is to inter­ pret either w ords or things or concepts. All three - signifier (w ords), signified (things), and the relations between them (concepts) - offer them selves to men to be deciphered in order to discover the “ te x t” o f the w orld. As Em erson w rote, “ The w orld is em blem atic. Parts o f speech are m etaph ors, because the w hole o f

219

nature is a m etaphor o f the hum an m in d” (Em erson, 1 9 4 8 :1 8 ; A. Becker, 1975). A favored form o f discourse in this episte­ m ology is the com m entary or the essay, a deci­ pherm ent or interpretation o f lan guage and nature. In com m entary, etym ology is an im por­ tant strategy, not as an attem pt to discover the original m eaning o f w ords, but rather as an attem pt to discover the “ intrinsic ‘properties’ o f the letters, syllables, an d, finally, whole w o rd s.” O ne o f the things th at strikes us about the text o f a Javan ese sh adow play is the per­ vasiveness o f etym ologizing as an explanatory strategy. Javan ese call this etym ologizing djarw a dhosok or “ fo rced ” (im posed) inter­ pretation. M y ow n first im pulse w as to dism iss etym ological com m entary in w ayan g as “ fo lk ” linguistics, rooted in ignorance ab o u t the true history o f the w ords explain ed, for m any o f which I knew the San skrit etym ons. I dism issed etym ologizing, in spite o f its frequency and o b viou s im portance as a text-building strat­ egy, since it did not give the “ tru e” origin o f w ords. Even m ore, it app eared to me as an e m b arrassing and silly aspect o f w ayang. W hat I failed to see then w as th at, since the m eanings o f w ords constantly change, etym ologies must be reform ulated (like gen ealogies), based upon w hat one now , in the present, sees as the “ intrinsic” m eaning o f the w ord under co n sid ­ eration. A brief exam ple: etym ology A o f the w ord history traces it to French histoire , then to Latin historia “ a narrative o f p ast events” to G reek istoria “ Learning by inquiry” and back to tstor “ arbiter, ju d ge” and hence back in time to a possible Indo-European root. F^tym ology B o f the sam e w ord divides it into “ h is” and “ sto ry ,” and interprets the elements o f the w ord in the present. “ H is-sto ry” is also an accoun t o f p ast events, but an account relating prim arily to men, with w om en in a secondary role. Which etym ology is correct? It is im possible to answ er, for the question is w ron g in insisting that we reject one or the other conceptual strategy, etym ology A or ety­ m ology B. Certainly etym ology B tells us m ore that is relevant and true to current thought than etym ology A. In traditional Jav an e se d is­ course, including w ayang, but also including history and com m entary, the strategy we have called etymology B is held to be serious and

220

ALTON L. BECKER

an im portan t p art o f a text, a basic w ay o f deciphering this w orld. E tym ologizing o f this second sort is know n to us, in part, a s explicating , and the object to be explicated is usually a text clearly recog­ nized as literary or religious or legal, an d we have specialists w ho explicate each o f these kinds o f text. W hat they d o is relate the w ords o f the text (and the ph rases and sentences, etc.) to the current con text. Precisely in this sense, though less specialized, the d alan g relates the old w ord s to the current context. W hat differ­ entiates the d alan g from the exp licators of texts in our society is that he explicates pri­ m arily p roper nouns, nam es for things, w hereas we tend to feel that nam es are the m ost a rb i­ trary w ords o f all, given to people and places before they really “ a re .” Etym ologizing about nam es is not unknow n in our culture, o f co urse, but it is not particularly highly valued as a w ay o f understanding people and places. W hat can we know by explain in g the nam e “ D etro it,” via etym ological strategy A or B? T here are tw o structural points in a lakon when etym ologizing, as a text-building stra t­ egy, is ap p rop riate: in description or dialogue. (It never occurs in suluks , where it is m ost needed.) Etym ologizing is the descriptive part o f a scene (either janturan , “ description o f a p lace,” or carios , “ description o f prior a ctio n ” ) is done by the d alan g directly and it is serious. Etym ologizing in a dialogue is done by one o f the ch aracters, and m ay be serious and “ a c a ­ dem ic,” if spoken by Krishna or A b iasa, or only half-serious, if spoken by a clow n. A m ajor skill in puppetry is the ability to ety­ m ologize in all these w ays. Let us exam in e a few instances. After the m antra and a set m usical interlude, the d alan g brings ou t the puppets for the first scene, and begins the description o f the first scene o f the first pathet o f the lak on , using fixed phrases: Once there w as a land. Many are god ’s crea­ tures that walk the earth or fly the air or swim in the water. M any are the beauties of the world. Yet none can equal those of this land, M anikm antaka (here the name o f the particu­ lar place in the particular story is inserted). Among a hundred there are not two, among a thousand not ten like M anikm antaka. . . .

Then the d alan g describes the kingdom fo llow ­ ing the strategy o f m oving from w idest physi­ cal context to n arrow est, from the place o f the kingdom am o n g all kin gdom s, the m ountains aroun d it, the sea, the tow n itself, the houses, the people, n arrow in g to a specific person, the king and those ab o u t him. All this is set lan ­ gu age, though ph rases can be left out or reor­ dered slightly. In these p assag e s, the skill o f the dalan g in con trolling the rhythm and pitch con tou rs o f his voice in relation to the gam elan is established (or n o t ) / In speaking the past alm ost entirely his legitim acy as a d alan g is being proved in one area. At som e poin t in the description, usually as a transition from the description o f the kingdom to the description o f the king, the d alan g begins his first etym ol­ ogy, either on the nam e o f the country or on the nam e o f the king, or both. Here another skill is brough t to the foregrou nd, for the ety­ m ologies are not set, although one m ay borrow them from w ayan g p ro m p tb o o k s called pakem (at the risk o f being know n, condescendingly, a s a “ b o o k ” dalan g). T he dalan g displays his skill at exp lication ; he m ust be authoritative and inform ative. H e does not, how ever, explain w ords by consultin g a dictionary o f Sanskrit roots, but interprets the elem ents o f the w ords as Ja v an ese w ords: The king who ruled this land is called M aha Prabu Niw ata Kaw aca. And his name means “ one who wears armor that may never be pierced” which, in our time, means “ one who could not be defeated,” for he and all his people believed that, and acted as if that were true. His name is made o f three words: N i, W ata, and Kaw aca. Ni or nir is from the word nirwana. N irw ana means freedom from desires, freedom from the past, freedom from the future, something which cannot be likened to anything. In other words, the Great God. Wata means blind, without vision. Kawata comes from Kaca, which means mirror. Hence, his name, N iw atakaw aca, means a mirror that is broken, a mirror which has lost its ability to reflect the truth, rhe Great God. When he was young he w as called Nirbito, which comes from Nir and bito. Nir is, as was said, from Nirwana. Bito means afraid. For although all feared him, he w as himself a coward and turned away from the Great Ciod.

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

T h is is a version o f the first etym ology from the story, A rjuna W iw aha, as 1 learned it. N otice that the nam e is explicated m ore than once, and that the m eaning a s a w hole (“ one w ho could not be d efeated ” ) is not the sam e a s the m eaning o f the parts (“ a m irror blind to n irw an a” ). Both are true an d both, alo n g with the childhood nam e o f the king, tell us a b o u t him. If he had other nam es they w ould be interpreted here, too. C learly, w ord s here are not arbitrarily related to people and things. 1 d o not intend here to g o into the ritual and m agical potentialities o f this lan guage, chiefly because I only very dim ly understand them. It is enough to say that the sh ad ow play is a text nonarbitrarily related to the w orld outside the play, and that explication o f the language is a m eans to cut through the hidden nature o f things. The d alan g is a skilled exp licator, w ho dem on strates that com plexity and obscurity can be unm asked and, hence, provides a model for understanding the w orld. O thers (A nderson, R esink, Hmmerson) have described how the present w orld look s within this m odel. Events in Indonesia reallv are inter­ preted by som e Ja v an ese a s lak on , the lakon plot does have psych ological reality as a kind o f m editation, nam es o f political leaders are taken a s revealing ch aracter and role, changes in stories or m ythologies from which m otifs are draw n to parallel social and religious ch an ges. T h at is, each w ay th at the text relates to its con text (see the first section) is em blem ­ atic o f the w orld and defines a w ay o f inter­ preting the w orld, once one believes, know s, o r pretends that reference is nonarbitrary.

C o n clu sio n : T o w a rd an A e sth e tic U n d ersta n d in g of Com m u n icatio n T he m ethodology o f this essay has been to describe the various sorts o f relations a text (or a p art o f a text, a w ord, a sentence, a p assage, an episode) has with its con text. Parts o f a text relate to the w hole under the con strain ts o f w hat we called plot coherence. T he m otifs or ep isod es o f a text relate to their source in a cultural m ythology under the con strain ts o f

221

invention. T he text and its parts relate to the participan ts in the linguistic act (direct or indi­ rect speaker, direct or indirect hearer, direct or indirect beneficiary, etc.) under the con straints o f intentionality. The text and its p arts relate to the nontext w orld under the con strain ts o f w hat we have called reference (either nam ing or m etaph oric reference). In the previous sec­ tions o f this com m entary, these relations have been exam in ed, not in term s o f their specific content, but at the m ore general level o f con ­ strain ts on specific content. I ...I T he g oal o f the ph ilologist is to guide o u tsid ­ ers (here non-Javanese) to w hat m ight be called an aesthetic understanding o f a text. T o achieve an aesthetic understanding it seem s reason able to say that in interpreting a text, the outsider m ust be aw are o f his ow n differences - p ar­ ticularly those m ost “ n atu ral” to him - and m ust learn to use new conventions o f coher­ ence, invention, intentionality, and reference. For an aesthetic response to be possible, a text m ust a p p ear to be m ore or less coherent; the m ythology it draw s upon and presupposes m ust be m ore or less know n; the conventional intent o f the creator or speaker o f the text in relation to on e’s ow n role as hearer/reader/ interpreter m ust be relatively well understood; even the m ore basic assu m ption s a b o u t how w ords relate to thoughts and the things o f the w orld need to be m ore or less shared. If any o f these kinds o f m eaning is not understood, then on e’s responses to w ayan g are either incom plete or contradictory. N ever fully to understand and constantly to m isunderstand are linguistic path ologies that characterize a w ide range o f phenom ena from the strategic understanding o f the schizophrenic to the per­ sistent con fusion and uneasiness o f one w ho is learning to use a foreign lan guage; all these p ath ologies subject one to a w orld in which lan guage and m etalanguage are incoherent, w here, to take an extrem e case, people say “ I love y o u ” and at the sam e time reveal co n tra­ dictory m essages, even “ I hate y o u ,” in a look o r a slap. T h e universal source o f language path ology is th at people ap p ear to say one thing and “ m ean ” another. It drives people m ad (the closer it gets to hom e). An aesthetic response

222

ALTON L. BECKER

is quite sim ply the op p o site o f this p ath ology. It is op p o site in the sense that the sam e co n ­ straints are relevant to both, but there is one difference. T h at is, op p o sites are things which are in the sam e class but differ in one feature (H ale, 1974). Schizophrenia, foreign lan guage learning, and artistic expression in lan guage all operate under the sam e set o f linguistic vari­ ables, con strain ts on coherence, invention, intentionality, and reference. The difference is that in m adness (and in the tem porary m adness o f learning a new language or a new text) these

con strain ts are m isun derstood and often appear contradictory; w hereas in an aesthetic response they are understood a s a coherent integrated w hole. Sh adow theater, like any live art, presents a vision o f the w orld and on e’s place in it which is whole and hale, where m eaning is possible. The integration o f co m ­ m unication (art) is, hence, a s essential to a sane com m unity a s clean air, go o d fo od , and, to cure errors, m edicine. In all its m ultiplicity o f m eaning, a w ell-perform ed w ayan g is a vision o f san ity.7

NOTES

5

1 2

3

4

I ow e this term , and m uch o f my under­ stan din g o f it, to Vern C arroll. The notion o f speaking the present and speakin g the p ast cam e to me from M aurice Bloch. Speaking the past is a particular kind o f speech act or m ode o f com m u n ica­ tion, which Bloch defines for the M erina o f M ad a g asca r, w ho them selves describe certain ritual speech m aking a s “ speaking the w ords o f the a n cesto rs” (Bloch, 1974). Bloch is w ron g, 1 think, in con trastin g fo r­ m alized speech acts and everyday speech acts, on a scale o f m ost to least form alized language. Everyday speech acts are also highly form alized. 1 feel that the poles o f this scale range from repetition (m ost fo rm al, speakin g the past) to im agination or internal discourse (least form al, sp e a k ­ ing the present), and I argue that neither pole is ultim ately attainable. For an early view o f w ayan g as “ speakin g the w ord s o f the an cesto rs,” see W. H. R assers (1 9 5 9 ). The notion o f iconicity is derived from Kenneth Boulding (1 9 6 1 ), a basic text in the study o f com parative epistem ography. The centrality o f tense in establishing textual coherence in English narrative is dem on strated in W illiam L ab o v , T ra n sfo r­ m ation o f experience in narrative syn tax (197 2 ). Ironically, m ost attem pts to “ preserve” tra­ ditional d ram a require deep change. This “ iron y” is discussed in Becker (1 9 7 4 ).

6

7

M y teacher tells the story o f his first w ayang perform ance in which he insisted on per­ form ing a story which w as too “ h eavy” (berat) for him. All the oil lam ps died midperform ance, so his grandfather pushed him aside and continued the perform ance, and all the lights cam e on again. O n M oun t K aw i, near M alan g, a w ayan g perform ance goes on every day and every night, n on stop year round, perform ing for the essential audience and preserving the spiritual texture, the ruwatan. See G regory Bateson, for further exam ples (chiefly Balinese) o f the role o f skill a s a basic element in aesthetics: “ Style, grace, and inform ation in prim itive a r t.” Bateson w rites, “ Only the violinist w ho can control the quality o f his notes can use variation s o f that quality for m usical p u rp o se s” (1 9 7 2 :1 4 8 ). T his essay, taken together with Ju dith Becker’s Time and Tune in Ja v a (in the volum e from which this extract com es, pp. 1 9 7 -2 1 0 ], suggests the possibility o f a single set o f constraints running through the w hole o f the traditional Javan ese ep is­ tem ology, in m usic, calendars, texts, rituals, and social relations. O f course, this unity m ay be in p art the oversim plification o f an ou tsider, but if true, this unity is probably a rather rare situation in a culture, a s it is in a person. In both it has great pow er. The com plex o f ch an ges we call modernization necessarily fragm en ts this unity. Social change alters one by one, and in no p a r­

TEXT-BUILDING, EPISTEM OLOGY, AND AESTHETICS IN JAVANESE SHADOW THEATRE

ticular order, it seem s, those relations o f a text to its context which constitute its m eaning. M od ern single time (Greenwich M ean T im e, m anifest in the m odern neces­ sity o f life, the w ristw atch) thus strongly affects plot coherence by devaluing m ulti­ ple time. If m ultiple time is devalued, coin ­ cidence ceases to be “ truth ” (kebetulan), and is replaced, usually, by narrative/causal “ tru th .” . . .

REFERENCES A ristotle. Politics and poetics (B. Jo w ett and T. T w ining, tran s.). N ew Y ork: V iking Press, 1969. A nderson, Benedict R. O ’ G. Mythology and the tolerance o f the Javanese. D ata Paper N o. 2 7 . Ithaca: Cornell University Southeast A sia Program , 1965. Bateson, G regory. Steps to an ecology o f mind. N ew Y ork: Ballantine, 1972. Becker, A. I.. The journey through the night: Som e reflections on Burm ese traditional theatre. In M oh d. T aib O sm an (F d .), Tradi­

tional dram a and music o f southeast Asia. K uala Lum pur: D ew an B ah asa dan Pustaka, 1974. (A lso in The dram a review , W inter 1970.) Becker, Alton L. A linguistic im age o f nature: The Burm ese num erative classifier system .

International Jou rn al o f the Sociology o f Language. 197 5 , 5 , 1 0 9 -2 1 . Becker, Alton and I G usti N gu rah O k a. Person in Kaw i: E xploration o f an elementary sem antic dim ension. Oceanic linguistics, 1 9 7 6 , 13 , 2 2 9 -5 5 . Bloch, M aurice. Sym bols, son g, dance, and features o f articulation. European journal o f sociology , 1974, X V , 58. B oulding, Kenneth. The image. Ann A rbor: University o f M ichigan Press, 1961. C o om arasw ain y, A. K. H indu view o f art: Theory o f beauty. In The dance o f Shiva. N ew Y ork: N o o n d ay Press, 1957. Em erson, R. W. Nature. N ew Y ork: Liberal Arts Press, 1948.

223

Em m erson, D onald. The R am ayan a syndrom e. Forthcom ing. H ahn, Edw ard. Finite-state m odels o f plot com plexity. Poetics: International review fo r the theory o f literature , 19 7 3 , 8. H ale, Kenneth. A note on a W albiri tradition o f autonym y. In D. D. Steinberg and L. A. Ja k o b o v its (F^ds.), Semantics: An interdisci­ plinary reader in philosophy , linguistics , and psychology. N ew Y ork: C am bridge Univer­ sity Press, 1974. H aley, Ja y . Strategies o f psychotherapy. N ew Y ork: G rove and Stratton, 1963. H ard jo w irogo. Sedjarah Wajang Purwa. D ja ­ karta: Balai Pustaka, 1968. H o o y k aa s, C . Cosm ology and creation in the Balinese tradition. The H ague: M . N ijh off, 1974. K ats, J. Het Javansche Jooneel / Wayang Poerwa. W eltrveden, 1923. L ab ov, W illiam . Language in the inner city. Philadelphia: University o f Pennsylvania Press, 1972. L acan , Jacq u e s. The language o f self (A. W ilden, tran s.). Baltim ore: Jo h n s H opkins Press, 1973. Lain g, R. D . Knots . N ew Y ork: V intage, 1970. Levi-Strauss, C laude. Structural anthropology. N ew Y ork: Basic B o ok s, 1963. R assers, W. H . On the origin o f the Javan ese theatre. In Panji , the culture hero. The H ague: M o u to n , 1959. R esink, G . J . From the old M ah ab h arata- to the new R am ayan a-order. Bijdragen tot de TaalLan d en Volkenkunde> D L 131 II/III, 1975, 2 1 4 -3 5 . Y oung, R. E., and Becker, A. L. T o w ard a m odern theory o f rhetoric: A tagm em ic contribution. In J. Em ig, J. Flem ing, and H . M . Popp (Eds.). Language an d learning. N ew Y ork: H arcou rt, Brace and W orld, 1966. Y oung, R. E., Becker, A. L., an d Pike, K. L. Rhetoric , discovery and change. N ew Y ork: H arcou rt, Brace and W orld, 1971. Zurbuchen , M ary. K aw i discourse structure: C ycle, event, and evaluation . Rackham L it­ erary Studies , W inter 1976, 4 5 - 6 0 .

Moral Inversions and Spaces of Disorder

In trod u ction As D ou glas argued, any structure will leave g a p s and anomalies. In addition, any system o f order will generate images o f its inversion. Religion has to address d isor­ der as well as order, anom aly as well as consistency, ambivalence as well as c o m ­ mitment, evil as well as go o d - and often does so by attem pting to assimilate them. Indeed, religion d raw s on the rich imagery o f the m argins in order to m ake higherlevel generalizations a b o u t the order it proposes. Often this happens in rituals o f initiation (see the essay by Turner, chapter 26). In this section we d raw on three different sources - myth, witchcraft, and carnival. These essays explore the preva­ lence o f both the am oral and the immoral in the cultural imagination and in social practice. For further provocative ethnography that explores the social and historical consequences o f a politics o f the imaginary see, respectively, the essay by De Boeck (chapter 38) and van de Port (1998).

19

The Winnebago Trickster Figure Paul Radin

Paul Radin (1883-1959) was an out­ standing Am erican anthropologist of the Boasian school noted for his close work w ith the W innebago. As the present essay shows, Radin had trem en ­ dous appreciation for the insight con­ veyed in Am erindian narrative. This excerpt from Radin's interpre­ tive essay on the trickster myth o f the W innebago Indians is best read in con­ junction w ith his record of the entire cycle set out earlier in his book. The trickster is an amoral yet essentially good-natured creature whose adven­ tures as both duper and dupe express a kind of ontogeny, a coming into being of consciousness and differentiation that may be understood at both psycho­ logical and cosmological levels. The trickster cycle exemplifies the role of chaos and am biguity in creation myth. Radin also points to the centrality of the comic. As Diamond (1972) indicates, the trickster personifies a frank recogni­

tion and acceptance of human ambivalence. Trickster figures are known from many parts of the w orld. Individual trickster gods are found w ithin the pantheon in polytheistic religions like those in ancient Greece or in West Africa. Trick­ sters are central to the storytelling of many Native Am erican groups; I recom­ mend Howard Norman's exquisite ren­ derings of Swampy Cree versions (1976). A recent w ork on Southern African Bushmen emphasizes the role of trick­ ster figures there and makes a number of useful connections w ith the literature on shamanism (Guenther 1999). Comedy and irony are also central to spirit pos­ session and widespread in religious tra ­ ditions, albeit rather absent from the sustained seriousness characteristic of most public or "official" rituals and rep­ resentations of Christianity and Islam. As a Boasian, Radin worked closely with gifted raconteurs in order to record

From Paul Radin, “ The Nature and the M eaning o f the M yth,” in The Trickster: A Study in American Indian Mythology (reproduced by kind permission o f the Philosophical Library, New York, 1972 [1956]), pp. 132-54.

THE W INNEBAGO TRICKSTER FIGURE

cultural texts, and paid careful attention to w hat participants themselves said about the m aterial and to the individual variants. In his classic accounts of Prim i­ tive M a n as P hilosoph er (1957a [1927]) and Prim itive Religion (1957b [1937]), Radin argued powerfully against the Levy-Bruhlian view of a prelogical m en­ tality and, indeed, against the entire evolutionist approach, the "fundam en­ tally misleading doctrine that primitive peoples represent an early stage in the history of the evolution of culture" (1957a [1927]: x). For Radin there are always individuals holding more or less religious perspectives or w ho are more or less inclined to engage in philosophi­ cal speculation. Radin's portrait of intel­

T he W innebago w ord for trickster is w akd­ junkaga , which m eans the tricky one . The co r­ respon din g term for him in Ponca is ishtinike , in the kindred O sag e, itsike and in D akotaS iou x, ikto-mi. T he m eaning o f the Ponca and O sage w ord s is unknow n, that o f the D akota is spider. Since all these three stem s are clearly related etym ologically, the question arises as to w hether the W innebago rendering, the tricky one, does not really m ean sim ply onewho-acts-like-W akdjunkaga, and is thus sec­ on dary. In no other Siouan language is the stem for tricky rem otely like wakdjnnkaga. It seem s best, then, to regard the real etym ology o f w akdjnnkaga as unknow n. The sim ilarity o f the exp loits attributed to W ak d ju n k aga and all other trickster-heroes in N orth Am erica is quite astoundin g. The only possible inference to be draw n is that this myth-cycle is an old cultural possession o f all the A m erican Indians, which has rem ained, as far as the general plot is concerned, relatively unchanged. Ju st because o f this fact the spe­ cific differences between the W innebago mythcycle and the others assum e special im portance and dem and explan ation . T o do this a d ­ equately it will first be necessary to sum m arize the plot o f the W akdjnnkaga myth-cycle in con siderable d e ta il.’

227

lectual and reflective speculation is far broader and richer than th a t portrayed by Tylor. Just as he demonstrates, in the chapter im m ediately follow ing the ones presented here, the satirical dimension of the Trickster tales, so he locates "skepticism and critique" am ong certain thinkers in any "prim itive" society (1957a: Chapter XIX). O ther significant members of the Boasian school include Sapir (1956 [1928]), Lowie (1935, 1948), and Benedict (1934). They frequently em phasized the em otional dimension in religious expression and its relation to personality. For a lively rendition of the Am ericanist language-based tradition by contem porary practitioners see V alentine and Darnell (1999).

The cycle begins with an incident found in no other version, nam ely W ak dju n k aga pic­ tured as the chief o f the tribe, giving a warbundle feast on four different days. H e, although host and consequently ob ligated to stay to the very end, is described as leaving the cerem ony in order to cohabit with a w om an, an act which is absolutely forbidden for those partici­ pating in a w arbundle feast. O n the fourth day he stays to the end and invites all the partici­ pan ts in the feast to acco m pan y him by boat. H ardly has he left the shore when he returns and destroys his bo at as useless. At this piece o f stupidity som e o f his co m p an io n s leave him. He then starts on fo ot, but after a short time destroys both his w arbundle and his arrow bundle and finds him self eventually deserted by everyone and alone; alone, that is, as far as hum an beings and society are concerned. With the w orld o f nature he is still in close contact. H e calls all objects, so our text tells us, younger brothers. He understands them ; they under­ stand him. T his is clearly an introduction and its purpose is m anifest. W ak dju n k aga is to be desocialized, to be represented as breaking all his ties with m an and society. Why our racon ­ teur began the cycle in this particular fashion it is im possible to say, but it is best to assum e

228

PAUL RADIN

that it is a literary device. Presum ably he has decided that W ak d ju n k aga is to be depicted as com pletely unconnected with the w orld ot' m an and a s g rad u ally evolving from an a m o r­ ph ous, instinctual and unintegrated being into one with the lineam ents o f m an and one fo re­ shadow ing m an ’s psychical traits. H e h as, in short, like m any another epic w riter, begun in m edias res . W hat he seem s to be say in g is: “ Here is W ak d ju n k aga pretending to be th o r­ oughly socialized and ab o u t to em bark on a w arparty. But let me tell you w h at he really is: an utter fool, a breaker o f the m ost holy ta b o o s, a destroyer o f the m ost sacred o b je c ts!” And then he proceeds, in kaleid oscop ic fashion, to reduce W ak d ju n k aga to his prim itive self. T he exp loits that follow tell us precisely w ho W ak dju n k aga is. (See incidents 4 - 1 0 in text.) In the first he treach erously lures an old b u ffalo to destruction , kills him in m ost cruel fashion and butchers him. N o ethical values exist for him. And how d oes he kill and butcher the bu ffalo? W ith only one hand, his right. The next incident show s why only one hand has been used. He is still living in his un con scious, m entally a child, and this is here sym bolized by the struggle between his right and his left hands in which his left hand is badly cut up. H e him self is hardly aw are o f why this has occurred. He can only ejaculate, “ W hy have 1 done th is?” In contradistinction to W ak djun k­ a g a , the w orld o f nature is represented as con ­ scious, and the birds, in a lan guage he cannot understand, exclaim , “ L o o k , look ! There is W ak d ju n k aga. T here he g o e s!” In the next incident he is still W ak d ju n k aga the undifferentiated and instinctual. H e com es upon a being with four little children w ho m ust be fed in a certain m anner and at a certain time lest they die. In sh ort, the principle o f order m ust be recognized. But he k n ow s no such principle. T he father w arn s him that if the children die because W ak d ju n k aga has failed to follow his instructions, he will kill him. Yet W ak d ju n k aga, because o f his ow n hunger, d isob ey s the instructions given him and the children die. Im m ediately the father is upon him. W ak d ju n k aga is pursued around the island w orld, that is the universe, and only by jum ping into the ocean surrou ndin g it does he escape death.

As he sw im s aim lessly in the w ater, not know ing where the shore is, if, indeed, one exists, com pletely w ithout bearin gs, he ask s fish after fish where he can find land. N on e o f them know s. Finally he is told that he has been sw im m ing alo n g the shoreline all the time. H e has barely lan ded, that is, he has barely got his bearings, when he attem pts to catch som e fish. But all he can obtain is the w ater through which som e fish have passed. O u t o f this he enthusiastically prepares a so u p , and fills him self to his utm ost capacity. As he lies there, practically in cap acitated, his stom ach shining from being distended, a dead fish drifts by. He seizes it, but he cann ot eat any m ore and he buries it. Here we find W ak djun k aga com pletely unanchored. He is not only isolated from man and society but - tem porarily at least - from the w orld o f nature an d from the universe as well. Sm all w onder, then, that he is described a s thoroughly frightened and as saying to him self, “ T h at such a thing should happen to W ak d ju n k aga, the w arrior! Why I alm ost cam e to g rie f.” W hat the auth or intended here - the enraged father, the pursuit, the headlong flight and the subm erging in the ocean - m ay well have been m eant as a description o f w hat can happen to anyone w ho leads the life instinctual. H ow ever, there is also another point involved here. Being frightened is, in W innebago sym bolism , generally the in dica­ tion o f an aw ak en in g con sciousn ess and sense o f reality, indeed, the beginning o f a co n ­ science. And that seem s to be borne out by the next incident (11 in text), where W ak djun k aga is represented as im itating w hat he takes to be a m an poin ting at him but which turns out to be a tree stum p w ith a protrudin g branch. The im portan t point here is his reaction to his blunder and stupidity. “ Yes, in deed,” so he say s, “ if is on this account that the people

call me W akdjunkaga, the foolish one! They are r i g h t H e h as one o f the necessary traits o f an individualized being now , a nam e. In W innebago society a child had no legal e x is­ tence, no statu s, until he received a nam e. T he episode which follow s (12 in text) is know n th rou gh out N orth A m erica in a practically identical form . It describes how

THE W INNEBAGO TRICKSTER FIGURE

W akdjun kaga persuades som e ducks to dance for him with eyes closed an d how he w rings their necks as they dance, though m ost o f them succeed in escaping. He ro a sts the few he has killed an d, exh austed from his encounter, goes to sleep after instructing his an us to keep w atch. H is an us does its best to aw aken him w hen foxes ap p ear, but to no avail, and W akdjun kaga aw aken s to find the ducks have been eaten. In anger he punishes his an us by burning it, and when he can endure the pain n o longer, he exclaim s, “ O uch! This is too m uch! . . . Is it not fo r such things that I am called W akdjunkaga ? Indeed, they have talked

me into doing this, just as if I had been doing something w rong!” Im portan t for ou r purpose is this exclam ation , and a lso the one in inci­ dent 14, where W ak dju n k aga discovers that he has been devouring parts o f his ow n intestines and com m enting upon how delicious they taste: “ Correctly indeed am I named Wakd-

junkaga, the foolish one! By being called thus I have actually been turned into a wakdjunk­ aga, a foolish on e!” With these incidents (12, 13, 14 in text) we have reached a new stage in W ak d ju n k aga’s developm ent. The em ph asis is now upon defin­ ing him m ore precisely, psychically and physi­ cally. He is now to be show n em erging out of his com plete isolation and lack o f all identity, and as becom ing aw are o f him self and the w orld aroun d him. He has learned that both right and left hands belong to him, that both are to be used and that his an us is part o f h im self and cannot be treated as som ething independent o f him. H e realizes, to o, that he is being singled ou t, even if only to be ridi­ culed, and he has begun to understand why he is called W ak djun k aga. Hut he does not a s yet accept responsibility for his action s. In fact, he holds other people, the w orld outside o f him self, as com pelling him to behave as he does. It is only at this point that we are told an y­ thing specific ab o u t W ak d ju n k ag a’s a p p e a r­ ance. Every W innebago, o f co urse, knew w hat it w as. Why then are we inform ed ab o u t it just here? The answ er seem s to be that his original appearan ce is now to be altered. He is now to be given the intestines and an u s o f the size and shape which m an is to have.

229

T h a t this episode has not been placed here ju st through accident is proved by the fact that in the episode which im m ediately follow s we have the first m ention o f his penis, o f its size and o f his m anner o f carrying it in a box on his back. And for the first tim e are we m ade aw are o f his sexuality. In all other specifically trickster m yths lust is his prim ary ch aracteris­ tic; in these all his adven tures reek with sex. If in the W innebago W ak dju n k aga cycle it is not m entioned until now , th is is because the au th o r or au th o rs w ho gav e this cycle its present shape w ished to give us not a series o f T rick ster’s adventures a s such but the evolu­ tion o f a T rickster from an undefined being to one with the physiognom y o f m an, from a being psychically undeveloped and a prey to his instincts, to an individual w ho is at least co n sciou s o f w hat he does an d w ho attem pts to becom e socialized. Sex is treated prim arily in its relation to W ak d ju n k ag a’s evolution. Sexual escap ad es d o not really seem to interest ou r raconteu rs as such. It is not strange then th at the first sexual episode related o f W ak d ju n k aga should consist o f his w akin g from his sleep to find him self w ithout a blanket. He sees it floating above him, and only gradu ally recogn izes that it is resting on his huge penis erectus. H ere we are brough t back again to the W ak d ju n k aga w hose right hand fights with his left, w ho burns his an us and eats his ow n intestines, w ho endow s the p arts o f his body with independent e x is­ tence and w ho does not realize their proper functions, where everything takes place o f its ow n acco rd , w ithout his volition. wT h at is alw ays happening to m e,” he tells his penis. It is not an accident th at this episode is placed just here. It belon gs here for it is to serve as an in troduction to giving W ak djun k aga an un derstandin g o f w hat sex is. Q uite properly, we first have the sym bol for m asculine sexuality and an exam ple o f how it is thought o f socially, nam ely as an alo g o u s to the banner raised by the chief when the tribal feast is given, and that then there follow s an exam ple (incident 16 in text) o f how it is used concretely an d properly. Incident 16, the sending o f the penis across the w ater so that W ak dju n k aga can have inter­ course with the chief’s daugh ter, is as well

230

PAUL RADIN

know n in N orth Am erica as rhat o f the h oo d ­ winked ducks. In m ost o f the trickster cycles it is im m aterial where it is placed. H ere, clearly, this is not true. It belongs here, for it is to be used to indicate how m eaningless and undif­ ferentiated W ak d ju n k ag a’s sex drive still is inherently; indeed, to show how m eaningless it is for all those involved. Penis, cohabitation are only sym bols here; no sense o f concrete reality is attached to them . T h at W akdjun kaga has as yet developed no sense o f true sex d if­ ferentiation is m ade still clearer by the episode where he tran sform s him self into a w om an (incident 2 0 in text). Im m ediately follow ing incident 16 we find the w ell-know n theme o f how he begs the turkey-buzzard to carry him on his back and fly with him. W hatever m ay be its larger p sy ­ chological im plication s, this incident seem s to play no role in the dram a o f W ak d ju n k ag a’s developm ent and m ust be regarded as an inter­ lude. H is rescue by w om en, after turkeybuzzard has treacherously dropped him into a hollow tree, is part o f the secondary satire on m an and society that perm eates the w hole cycle and ab o u t which we will have m ore to say in the next section. We have now reached the crucial episode where W ak d jun k aga changes his sex and m arries the ch ief’s son. T he overt reason given for his doin g this is that he and his com pan ion s have been overtaken by winter and are starv ­ ing and that the chief and his son have plenty. This episode like the preceding ones is well know n; no trickster cycle om its it. T he reason generally given is that T rickster does it to avenge som e insult. The change o f sex is a trick played on an oversexed individual in order to show to w hat lengths such a person will go, w hat sacred things he will give up and sacrifice to satisfy his desires. Such is its role in one o f the m ost fam ou s o f all N orth A m erican Indian trickster cycles, that o f Wisaka o f the Fox tribe.2 But here in the W innebago cycle it is not to avenge an insult but ostensibly to obtain food that the tran sform ation o f sex has occurred. T aken in conjunction with the sex episodes which have preceded and the tw o incidents that follow , its m eaning becom es clear. It is part o f W ak d ju n k ag a’s sex education . This

m ust begin by sharply differentiating the tw o sexes. It is a s if W ak dju n k aga were being told: this is the m ale; this, the penis; this is co h ab ita­ tion; this is the fem ale o rgan ; this is pregnancy; this is how w om en bring children into the w orld. Yet how can W ak d ju n k aga, with his generalized sexual o rg an s, arran ged in the w rong order an d still living distinct from him in a receptacle on to p o f his body, how can he be expected to understand such m atters? For that reason W ak d ju n k ag a’s sex life, indeed, his w hole physical life, is for him still so m e­ thing o f a wild ph an tasm ago ria. T h is p h an tas­ m agoria reaches its culm ination point in incidents 2 0 and 21 o f the text. Satire, R ab e ­ laisian hum our and grotesquen ess are co m ­ bined in these p a ssa g e s with am azin g effect. Thrice, within very short intervals before the visit to the ch ief’s son , the m an-w om an, W ak djun k aga, is m ade pregnant; she, a w om an , does her ow n courtin g; the m anw om an becom es pregnant again. W hose are the children he brings forth ? We are purposely left in ignorance in order to stress the fact that it m akes no difference. Parenthood is im m ate­ rial, for they are born o f a m an-w om an. We have here reached a point where o rd i­ nary w ords and term s are indeed com pletely inadequate. Only sym bols, only m etaph ors, can convey the m eaning properly. As soon as the last child is born he begins to cry and nothing can stop him. A specialist at pacifying children, an old w om an w ho has passed her clim acteric, that is, one w ho is beyond sex, is called, but she is helpless. Finally the infant cries out, “ I f I could but play with a piece o f white cloud ” T o translate this into m eaning then becom es the task o f a special sham an. So it is with the child’s other requests. They seem all unreasonable and unseasonable. W hat else can we expect in this p h an tasm ago ria? Yet these requests are, at the sam e tim e, reason ­ able and have concrete non-sym bolic signifi­ cance. N ot for the child, how ever, but for W ak dju n k aga, w ho is w aiting for spring to com e and for the time when he can obtain his food him self. Be it rem em bered: at no time is W akdjun kaga represented a s becom ing a victim o f this ph an tasm ago ria. H e alw ays rem ains his old prim ordial self. He has as yet not learned very m uch and has forgotten even less.

THE W INNEBAGO TRICKSTER FIGURE

The denouem ent arrives when W akdjun kaga is ch ased aroun d the fireplace by his motherin-law , when his vulva d ro p s from him and he is revealed as his true self. O rdinarily on such an occasion in the W akdjun kaga cycle he is represented a s laughing at the discom fi­ ture o f those on w hom he has played a trick. But here he runs aw ay. The reason is clear; the situation is fraught with to o m any difficulties. T o o m any tab o o s have been broken, the sen­ sibilities o f too m any people have been ou t­ raged, to o m any individuals have been hum iliated. It is serious enough for a chief’s son to be indulging in w h at turns out to be h om osexu al practices, but far m ore serious is the situation in which the chief’s w ife, W akdju n k aga’s “ m other-in-law ” , finds herself. A m ong the W innebago the m other-in-law tab o o w as very strict, yet here she is openly associatin g with one w ho could have m arried her daugh ter and becom e her son-in-law and thus a person with w hom she is not allow ed to speak and with w hom no joking is possible. The right to joke with and to tease an indi­ vidual im plies a very special relationship. It can only take place between a very restricted num ber o f blood-relatives and a less restricted num ber o f relatives-by-m arriage. Jo k in g between a m other-in-law and son-in-law is sim ply unthinkable. A pparently it w as even unthinkable in this Walpurgisnacht a tm o ­ sphere, for the n arrator does not use the term daughter-in-law when he sp eak s o f the chief’s w ife teasing W ak d ju n k aga, but the term hiciga> brother’s son ’s wife. T he fact that W akdjun k­ aga when functioning a s the daughter-in-law could not possibly be hiciga a W innebago audience, o f course, w ould know , but, under the circum stances, any term w as better than to call him daughter-in-law . I'he shock o f all these revelations to those participatin g in this com ic-tragic dram a is clear, and our raconteur has expressed this shock by bringing his narrative to a full stop. H e apparently feels that one m ust get out o f this insane atm osph ere quickly. I think he has done this very astutely. He has W akdjun kaga not only run aw ay but suddenly com e to som e realization o f w hat he w as doing. Suddenly, and for the first time in the cycle, he is pictured as a norm al m an with a w ife to w hom he is

231

legally m arried and a son for w hom it is still necessary to provide. In short, he is suddenly represented as a go o d citizen, as a thoroughly socialized individual. And so he returns to his hom e, is received there with joy and stays with his fam ily until his child is well able to take care o f him self. The only indication that it is W ak djun k aga with w hom we are here dealing is found in the last three sentences o f this episode. “ 1 will now g o around and visit people for I am tired o f staying here. 1 used to w ander around the w orld in peace but here I am just giving m yself a lot o f tro u b le.” In these w ords we have his protest again st dom estication and society with all its ob ligation s. D oubtless this also voices the protest o f all W innebago again st the sam e things. The biological education o f W akdjun kaga is now to be resum ed. T he next adventure is a utilization o f a strictly R ab elaisian theme found throughout aborigin al A m erica, the talking laxative bulb (incidents 23 and 24 in text). Although he now possesses intestines o f norm al hum an size he know s nothing about them. He com es upon a bulb which tells him that w hoever chews it will defecate. N ature has never taunted him in this fashion before. So he takes the bulb an d chew s it to find that he does not defecate but only breaks wind. T his expulsion o f g as increases in intensity progressively. He sits on a log, but is propelled into the air with the log on to p o f him; he pulls up trees to which he clings, by their roots. In his helplessness he has the inhabitants o f a village pile all their possession s upon him, their lodges, their d o g s, and then they them ­ selves clim b upon him, for he tells them that a large w arpartv is ab o u t to attack them. And so the w hole w orld o f m an is now on W akdju n k a g a ’s back. With a terrific expulsion o f gas he scatters the people and all their possession s to the four quarters o f the earth. And there, we are told, he stood laughing until his sides ached. A part from the grotesque hum our and the o b viou s satire, is there anything else involved here? Yes. Broadly speaking, a W innebago w ould say this is an illustration o f w hat h appens when one defies nature even in a m inor fashion , that this is w hat h appens when m an clim bs on W ak d ju n k aga’ s back.

232

PAUL RADIN

But this w orld to which he has fled to escape from society, the w orld where he could w ander aroun d in peace, has not finished its test with him. He now begins to defecate. The earth is covered with excrem ent. T o escape it he takes refuge in a tree, but to no avail, and he falls into m ountain s o f his ow n excrem ent. Blinded by the filth clinging to him he grop es helplessly for a path to w ater. The trees w hom he ask s for in form ation m ock and m islead him. Finally he reaches the w ater and can cleanse him self. H ow ever, despite this rem inder o f ign o­ rance, know ledge concerning him self and the outside w orld co m es to him slow ly. N o sooner has he cleansed him self com pletely than he m istakes the reflection in the w ater o f plum s grow ing on a tree on the shore for the plum s them selves.3 There now follow s a series o f incidents (2 7 -4 6 in text) that have little bearing on the education o f W ak d jun k aga. A part from their m anifest satiric im plications they are m ore or less the typical adventures o f all N orth A m eri­ can tricksters. They exem plify all the traits custom arily attribu ted to him, the m eaningless cruelty he inflicts upon others in ord er to obtain fo o d , and how , at the last m om ent, he is alw ays frustrated and cheated, cheated in fact, not only by others but by him self (see incidents 3 0 , 3 1 ); how he com es to grief by trying to im itate others (incidents 3 2 , 3 3 , 4 1 4); and how occasion ally he turns the tables on his torm en tors (incidents 3 4 , 4 5 , 4 6 ). From a literary and psych ological point o f view our myth-cycle b reaks dow n after incident 2 6 , where W ak d jun k aga is knocked unconscious by diving after the reflection o f plum s in the w ater, although som e o f the th reads are pulled together, albeit not too well, after inci­ dent 38. W hat should have follow ed incident 2 6 , I feel, is the episode where W ak d jun k aga, rhrough the instrum entality o f ch ipm un k, is taught where his genitals should be placed on his body and the p roper order o f penis and testicles. (See incidents 38 and 39 in text.) The w ords o f W ak d ju n k aga in the dialogue between him and chipm unk are w orth noting. They are m eant to point out that W ak d jun k­ aga is at last to becom e aw are concretely o f his sex. “ Is it not your penis you are carrying on

your b a c k ?” chipm unk shouts at him, and W ak djun k aga an sw ers, “ W hat an evil person it is w ho m entions th at! H e seem s to have full know ledge o f w hat I am carrying on my b a c k .” A gain chipm unk shouts at him, “ Y our testi­ cles to geth er!” and W ak djun k aga an sw ers, “ W hy, this being m ust have been w atching me closely.” T h ro u gh o u t W ak dju n k aga acts bew ildered and em b arrassed. At first he behaves purely passively, although he follow s the instruction. He becom es angry only when chipm unk finally shouts at him his last injunc­ tion - “ Put the head o f the penis on to p , put it on to p !” It is then, when his genitals are in their right place and correctly arran ged, when he has really becom e aw are o f his sex and his m asculinity, it is only then, that he pursues his torm entor. He attack s chipm unk with his penis, not, ostensibly, in order to co h ab it with him but to punish an d destroy him for m aking him aw are o f his genitals and o f his sex. It is his final protest at becom ing a m ature m ale. Be it rem em bered that his penis is still o f tre­ m endous length. T he farther he penetrates the hole in which chipm unk has sought refuge, the m ore o f his penis the latter bites o ff until it finally has been reduced to hum an size. In such fashion does W ak dju n k aga becom e a m ale and attain sex consciousn ess. A very im portan t addendum now follow s. In con trast to the m anner in which he d isposes o f the slough ed-off portion s o f his intestines, nam ely, by eating them him self, the parts o f the penis which chipm unk has bitten o ff are throw n into the w'ater an d transform ed into food plan ts for m an. W ak d ju n k aga’s resistance to attaining sexual m aturity h as innum erable larger p sy ­ ch ological and psychoanalytical im plications the explan ation o f w hich, how ever, I m ust leave to others. W hat I w ould like to stress here are tw o question s: first, the fact that he cannot him self reduce his large and am o rp h o u s genitals to their norm al hum an size, arran ge them in their p rop er order o r place them p ro p ­ erly. T h is m ust be accom plished through som e outside agency. Yet, on the other hand, he him self is represented as responsible for reduc­ ing the size o f his intestines. Second, it might be ask ed , whether there is involved in the final act, where chipm unk in his hole bites o ff large

THE W INNEBAGO TRICKSTER FIGURE

p arts o f W ak d ju n k aga’s penis, som e form o f em asculation or som e form o f coh ab itation . M y ow n belief is that neither is involved, but th at we are still dealing with W ak d ju n k ag a’s biological evolution and that w hat is being im plied here sym bolically is his transition from a generalized natural and procreative force to a concrete heroic hum an being. T h is, 1 feel, is expressly stated in his e xclam atio n , “ O f w hat a w onderful organ have I been deprived! But why should 1 say this? I can m ake useful objects o f all these pieces o f my penis for hum an b e in g s!” T h u s from being an uncon­ scious benefactor he has now becom e a con ­ scious benefactor not only o f m ankind but o f nature as well. H avin g attained biological m aturity one w ould have im agined that the narrative w ould then indicate how he attain s full psychical and social-ethical m aturity. But the incidents that follow show this very inadequately and incon­ sistently, if at all. It w as p erh ap s actually an im possible thing to do, con siderin g W akd­ ju n k a g a ’s traditional associatio n s* O ne o f the reason s for this failure, at least from a literarypsych ological point o f view , lay probably in the fact that one basic exploit or rather, series o f exp loits, connected with W ak d ju n k aga and w ithout which in the m inds o f the W innebago the W akdjun kaga cycle w as unthinkable, had still to be included, namely his visits to various an im als, the m anner in which he w as enter­ tained by them and the m anner in which he attem pts, quite unsuccessfully, to reciprocate their hospitality. (See incidents 4 1 —4.) But these episodes could only with the greatest o f difficulty be used to illustrate any progressive developm ent in W ak d ju n k aga’s ch aracter. An attem pt, how ever, seem s clearly to have been m ade, at least in one direction, nam ely, to show him as developing som e sense o f social an d m oral responsibility. In the incident im m ediately follow ing the tran sform ation o f the gn aw ed -off pieces o f his penis we are told o f W ak d ju n k ag a’s meeting with coyote and his attem p t to com pete with him as a keen scenter. Its only significance in o u r cycle is to serve as an introduction to the them e o f his visits to the m uskrat, snipe, w oo d ­ pecker and polecat (incidents 4 1 - 4 ) , and to m otivate his turning the tables on coyote (inci­

233

dent 4 6 ). W hat we have in the coyote episode is a very abbreviated form o f a com petition between W ak djun k aga and coyote which plays a much greater role in trickster m yths in other parts o f N orth Am erica. T he m ost that our raconteur can do with the episode o f the visits to the various anim als is to credit W ak djun k aga with w ishing to provide his fam ily with food, to present him a s a h arm ­ less, vain glorious blunderer and fool, and as one w ho succeeds in finally ob tain ing revenge on those w ho have hum iliated him or desire to d o so, like m ink and coyote. (See incidents 40 and 4 6 .) T h is is all part o f his socialization. T h u s, for exam ple, after polecat visits him and kills innum erable deer for his fam ily we have the follow ing idyllic scene. It is really best to quote it: “ Well, wife, it is about time for us to go back to the village. Perhaps our relatives are lone­ some for us especially for the children.” wl w as thinking of that m yself,” replied his wife. . . . Then they packed their possessions and began to carry them away. . . . After a while they got near their home and all the people in the village came out to greet him and help him with the packs. The people o f the village were delighted. “ Kunu, firstborn, is back,” they shouted. The chief lived in the middle o f the village and alongside o f him they built a long lodge for W akdjunkaga. There the young men would gather at night and he would entertain them for he was a good-natured fellow. The prodigal son has made good and returned! T h is reads alm ost like an accoun t o f the return o f a successful w arleader, or at least a great hunter. Yet som ething o f his old unregenerate self still adheres to him, as is seen in the delight he takes in hum iliating mink and coyote. H ow ever, a W innebago audience w ould have sym pathized with this hum iliation o f m ink and coyote. They w ould have agreed that W ak djun k aga w as a very good-natured person, a blundering fool, it is true, but m ore sinned again st than sinning, one w ho really m eant well but w hose good intentions alw ays went am iss. It is in this light that we m ust interpret the tw o episodes (incidents 4 7 and 48) which follow , W ak d ju n k aga’s rem oval o f

234

PAUL RADIN

natural ob stacles in the M ississip p i River that w ould interfere with the free m ovem ent o f hum an beings. But before proceeding to the discussion o f these, a few w ords ab o u t the im plication s o f one o f the points in the fourth o f his visits, that to the polecat, seem in point. In that delightful and R abelaisian episode polecat kills deer by shooting them with wind he exp els from his anus. He “ lo a d s” W ak d jun k aga with four such shots to take home with him. W ak d jun k­ aga is now faced with a new situation. In the case o f his visits to m uskrat, snipe and w oodpecker (4 1 -3 ), all he had to d o to get him self into difficulties and inflict pain upon him self w as to im itate them. But now , p ro ­ vided with the m eans for really accom plishing w hat his host, polecat, had don e, how w as he to fail, for fail he m ust? The problem is sim ply solved: he m ust w aste these p rovision s. So, w ithout any reason , he persuades him self that polecat has deceived him, and he sh o ots at four objects in succession blow ing them to pieces - at a knoll, at a tree, at an enorm ous rock and at a rocky precipitous hill, the last the sym bol o f a sacred precinct. It is his last act o f defiance again st the w orld o f nature with which he h ad, until recently, been on such intim ate term s. It is C aliban protesting again st the civilization which had been forced upon him. It w ould be quite erroneous to think that the author-racon teurs o f our cycle were trying in incidents 4 7 and 48 to exhibit to us a W ak d­ ju n kaga w ho had now becom e a wholly benef­ icent being, a semi-deity in fact. W hat we have here is a purely secondary addition with no actual connection with w hat h as preceded. It represents largely the influence o f the m ost sacred o f all W innebago narratives, the Origin Myth o f the Medicine Rite. T here, after Earthm aker has created the universe and all its inhabitants, anim al and hum an, he discovers that evil beings4 are ab o u t to exterm inate m an. In order to help them he sends W ak dju n k aga, the first being com p arable to man he has created, dow n to earth. T his is w hat is m eant when we are told that W akdjun kaga suddenly rem em bered the purpose for which he had been sent to the earth. In the Origin Myth o f the Medicine Rite W akdjun kaga is described

as failing com pletely. N o t even Earthm aker apparently could properly “ rehabilitate” him. But on earth W akdjunkaga could accom plish nothing. As the myth phrases it, “ Every variety o f sm all evil an im als began to play pranks on him and plague him and he finally sat him self dow n and adm itted to him self that he w as incapable o f d oin g anyth ing.” 5 Yet in spite o f all his trickster antecedents he has here, for a m om ent, been elevated to the rank o f a true culture-hero, although the specific role he is being given belongs properly to an entirely different hero, or rather heroes, the Twins. As 1 have indicated above, I think that a large part o f this tran sform ation o f the ch ar­ acter o f W ak djun k aga is due to the role he plays or w as intended to play in the founding o f the M edicine Rite. H ow ever, to judge from the fact that there seem s to have been a differ­ ence o f opinion am o n g the W innebago tw o generations a g o , and one which w as definitely not o f recent origin, as to how he w as to be evaluated, it m ay very well be that people alw ays interpreted him and his activities in tw o w ays. But to this we will return in the follow ing section when we deal with the W in­ n ebago attitude tow ard W akdjun kaga in the first decade o f this century. In the last scene (incident 49) we get still another picture o f him. We see him as a deity, an aspect o f his nature com pletely neglected in our cycle, an d as the elemental trickster, an ageing trickster, indeed alm ost a dem iurg, taking his last m eal on earth. H e is pictured sitting on to p o f a rock with his stone kettle, eating. He perpetuates this last meal for all time, leaving in the rock the im print o f his kettle, o f his buttocks and his testicles. He then departs and, since he is the sym bol for the procreating pow er as such and the sym bol for m an in his relation to the w hole universe, he first dives into the ocean and ascends to that island-w orld over which he presides, that lying im m ediately under the w orld o f E a rth m ak e r.. . . The above sum m ary should give the reader som e idea o f the com posite nature o f the W in­ n ebago trickster cycle and the degree to which the various episodes com posin g it have been welded together into a new whole. T o obtain a better conception o f the success the W in­

THE W INNEBAGO TRICKSTER FIGURE

n ebago achieved in this regard one m ust read the trickster cycles o f other Am erican Indian tribes. Then it will becom e clear to w hat an extent in the W innebago W akdjun kaga epi­ sod es, incidents, them es and m otifs have been integrated, and then the consum m ate literary ability with which this has been done will stand out sharply. T h at this literary rem odel­ ling an d reinterpretation is secondary there can be no question. It is apparen tly due to special circum stances in W innebago history and to the existence o f a special literary tra d i­ tion there. T o form som e idea o f w hat the W ak d jun k aga cycle w as originally we m ust, how ever, divest our version o f all those fea­ tures which have m ade it an aborigin al literary m asterpiece. This we shall attem pt to d o in the concluding section o f this introduction when we deal with the N orth Am erican Indian trickster-cycle in general.

T h e A ttitu d e of the W in n e b a g o to w a rd W a k d ju n k a g a M uch o f the an alysis given in (the previous] section is the analysis o f an outsider, o f a white m an, an d it goes w ithout saying that such an an alysis has its dangers and p itfalls, no m atter how well such an outsider thinks he know s an aboriginal culture. It is alw ays best to let m em bers o f the culture them selves sp eak , and I shall, therefore, attem pt to present now in a few w ords w hat were the ideas and evalu ­ ation s o f contem porary W innebago - I am sp eakin g o f 1 9 0 8 -1 8 - in regard to W ak d ­ ju n kaga and how he w as pictured in W in­ n eb ago literature. In those years when the new Peyote religion w as spreadin g throughout the tribe and m any W innebago began to m ake evalu ation s and re-evaluations o f their culture, W ak d jun k aga found both defenders and an ­ tagon ists. Let me com m ence with the state­ ment o f an old conservative which he prefaced to a myth not included in the cycle being given here: “ The person we call W akdjunkaga,” so he said, “ was created by Earthmaker, and he was a genial and good-natured person. Earth­

235

maker created him in this manner. He was likewise a chief. He went on innumerable adventures. It is true that he committed many sins. Some people have, for that reason, insisted that he really w as the devil/ Yet, actu­ ally, when you come to think o f it, he never committed any sin at all. Through him it was fulfilled that the earth w as to retain for ever its present shape, to him is due the fact that nothing today interferes with its proper func­ tioning. True it is that because o f him men die, that because of him men steal, that because of him men abuse women, that they lie and are lazy and unreliable. Yes, he is responsible for all this. Yet one thing he never did: he never went on the warpath, he never waged war. “ W akdjunkaga roamed about this world and loved all things. He called them all broth­ ers and yet they all abused him. Never could he get the better o f anyone. Everyone played tricks on him.” W hat this particular W innebago is undoubt­ edly trying to say is that W ak djun k aga repre­ sented the reality o f things, that he w as a positive force, a builder, not a destroyer. The reference to his not having gone on the w arpath is very illum inating. It indicates th at, for this p articular individual, W ak d ju n k ag a’s failure to help m ankind by destroying those w ho were plaguing it w as not a reprehensible thing because it w ould have m eant violence, meant w aging w ar. If W ak djun k aga w as thus useless after he had prepared the earth for m an that is quite intelligible. T h at men do not under­ stand him, that they m isinterpret and laugh at his activities, this to o is intelligible. He does not belong in the w orld o f men but to a much older w orld. In con trast to this sym pathetic attitude we have that o f the m em bers o f the Peyote rite. They used W akdjun kaga and his cycle to point a m oral. It w ould be quite erron eous to imagine that this w as an entirely new attitude; it existed long before the Peyote rite cam e into existence. The attitude o f the Peyote people is best illus­ trated in the follow ing hom ily: The older people often spoke to us of W akd­ junkaga. However, we never knew what they meant.7 They told us how, on one occasion, he wrapped a racoon-skin blanket around himself and went to a place where there were

236

PAUL RADIN

many people dancing. There he danced until evening and then he stopped and turned around. There w as no one to be seen any­ where, and then he realized that he had m is­ taken for people dancing the noise made by the wind blowing through the reeds. So do we W innebago act. We dance and make a lot o f noise but in the end we accom ­ plish nothing. Once as W akdjunkaga was going tow ard a creek he saw a man standing on the other side, dressed in a black suit and pointing his finger at him. He spoke to the man but the latter would not answer. Then he spoke again and again but without receiving any reply. Finally he got angry and said: wSee here! 1 can do that to o.” So he put on a black coat and pointed his finger across the creek. Thus both o f them stood all day. Tow ard evening, when he looked around again, he noticed that the man across the creek who had been pointing his finger at him w as really a tree stump. “ O my! What have I been doing all this time? Why did I not look before I began? N o wonder the people call me the Foolish-O ne!” W akdjunkaga w as walking around with a pack on his back. As he walked along someone called to him. “ Say, we want to sing.” “ All right,” said he. “ I am carrying songs in my pack and if you wish to dance, build a large lodge for me with a small hole at the end for an entrance.” When it was fin­ ished they all went in and W akdjunkaga fol­ lowed them. Those who had spoken to him were birds. He told them that, while they were dancing, they were not to open their eyes for if they did their eyes would become red. Whenever a fat bird passed W akdjunkaga would choke it to death, and if the bird squeaked he would say, “ T hat’s it! T hat’s it! Ciive a w h oop!” After a while one o f the birds got somew'hat suspicious and opened its eyes just the least little bit. He saw that W akdjunkaga w as choking all the birds he caught to death, and he cried out, “ Let all those who can run save themselves for he is killing u s!” Then this bird flew out through the top o f the house. W akd­ junkaga took the birds he had killed and roasted them. But he did not get a chance to eat them for they were taken away from him.

So are we Winnebago. We like all that is forbidden. We say that we like the Medicine Rite; we say that it is good and yet we keep it secret and forbid people to witness it. We tell members o f the society not to speak about it until the world comes to an end. They are, in consequence, afraid to speak of it. We, the Winnebago, are the birds and W akdjunkaga is Satan. Once as W akdjunkaga w as going along the road someone spoke to him. He listened and he heard this person saying, “ If anyone eats me, faeces will come out of him.” Then W ak­ djunkaga went up to the object that was talking and said, “ What is your nam e?” “ My name is Blows-himself-away.” W akdjunkaga would not believe it and so he ate this object. (It wras a shrub.) After a while he blew himself away. He laughed. “ O , pshaw! I suppose this is what it m eant.” As he went along it grew worse and worse, and it w as really only after the greatest hardship that he succeeded in returning home. So are we Winnebago. We travel on this earth all our lives and then, when one of us tastes something that m akes him unconsci­ o u s / we look upon this very thing with sus­ picion upon regaining consciousness . . . Here we have W ak dju n k aga as both rhe g lo ­ rified im age o f m an and a s the tem pter. The W innebago term used for Satan here is Hereshguina. The latter is the great evil spirit w ho is believed to have existed from the beginning o f tim e, w ho is a s old a s Farth m aker and alw ays negating w hat Farth m aker creates. A ccording ro one etym ology his nam e m eans “ he-of-w hose-existence-one-is-doubtful” .T h a t W akdjun kaga should be equated with him by the m em bers o f the sem i-C hristian Peyote rite is not strange. T h is insistence on W ak d ju n k aga’s purely negative side is very old attitude. We find it am on g the D ak o ta-S iou x and the Ponca. But equally old is the interpretation o f his ch arac­ ter and o f his positive activities to which I have referred before and which finds its best exp res­ sion in a very old m yth, The Two Boys.9 In this myth he is represented as actively helping H are in his endeavours to secure the pow ers that are eventually to help one o f the great spirits to victory over his enem y. In this myth

THE W INNEBAGO TRICKSTER FIGURE

purpose for which you were created. What happened to you after you reached the earth that you brought upon yourself alone. It is because o f your own actions and activities that you became the butt of everyone’s jest, that everyone took advantage o f you, even the smallest of insects. How is it then that now you are presenting as a model to be followed that very individual, Hare, who did do what I told him to? You, although you were given the greatest o f powers, made light o f my cre­ ation. It w as not anything I told you to do. It is therefore your own fault if people call you the Foolish-One. I created you to do what your friend Hare actually did. I did not create you to injure my crcation.

W ak d ju n k aga is represented as addressin g E arthm aker as follow s: Father, it is well. That which we desired, this you have given us precisely as we wished it and without any hesitation. It is my friend Flare who is to see that our purpose is attained. He is the only one who can accomplish it. All the spirits in the lodge from which we have come listen and obey what he says, for his are good thoughts. It is he who helped the human beings before, and he will do this for them too. T o this speech o f W ak dju n k aga Farth m aker replies in the follow ing fashion: Firstborn, you are the oldest o f all those I have created. I created you good natured: I made you a sacred person. I sent you to the earth to remain there so that human beings would listen to you, honour you and obey you and that you might teach them by what means they could secure a happy life. This w as the

T h is ap p aren t bew ilderm ent o f the W in­ n eb ago suprem e deity concerning the reason s for W ak d ju n k ag a's action s and this disavow al o f responsibility for them , it will be im portant to keep in mind. I ...I

NOTES 1

2

3 4 5

T he story is presented a s “ T he W innebago T rickster C ycle” in the volum e from which this extract com es, pp. 3 -6 0 . W. Jo n e s, Fox T exts , Publications o f the Am erican Ethnological Society, Leyden, 1 9 0 7 , pp. 315 ff. T h is incident is prob ab ly o f European origin. T hese are not represented as having been created by him. Another version o f the sam e myth states, “ He w as like a sm all child craw ling

237

6 7

8 9

ab o u t. . . . All one saw o f him w as his anus. H e accom plished no go o d and in fact injured E arth m aker’s creatio n .” H e is referring to the follow ers o f the Peyote rite. T h at is, they did not understand the sig­ nificance o f W ak d ju n k aga’s actions. The follow ing episode is not found in our version o f the myth. H e is referring to the eating o f the peyote. C f. Special Publications o f Bollingen Foun ­ datio n , N o . 3, Basel, 1954.

20

Witchcraft and Sexual Relations: An Exploration in the Social and Semantic Implications of the Structure of Belief Raymond C. Kelly Raymond Kelly, w ho taught at the Uni­ versity of M ichigan, is recognized for the precision w ith which he builds theo ­ retical argum ents from ethnographic m aterial. Kelly's account of Etoro w itch­ craft is exem plary for the w ay it situates w itchcraft as an integral part of a broader, coherent cultural cosmology. The dem onstration draws on structural­ ist m ethod w ith its emphasis on concep­ tual oppositions and analogies. For the Etoro of Papua New G uinea, these operate not simply in the realm o f the im agination but are part of a consistent logic of reproduction embracing eating practices, nurturance, and sexual rela­ tions. They generate specific anxieties that lead surely to the uncovering of witches, and even the killing of plump infants assumed to be witches. W hile much of the earlier literature saw witch

beliefs as a secondary symbolic expres­ sion of the social structure or as operat­ ing to m aintain social cohesion or regulate conflict, Kelly shows the far deeper roots of w itchcraft as an integral dimension of Etoro society. For further developm ent o f Kelly's ideas see his magisterial m onograph on the Etoro (1993), especially the elaboration of the present essay in Chapter 3. Knauft provides a fine account of sorcery in a neighboring group (1985) and a synthe­ sis of sexual culture in the region (1993). W itchcraft can be broadly understood as an im aginative inversion of dom inant norms and values, especially those asso­ ciated w ith nurture, kindness, and reciprocity characteristic of kinship. In a famous phrase, Monica Wilson charac­ terized witch-beliefs as the "standard-

From Raymond C. Kelly, “ Witchcraft and Sexual Relations: An Exploration in the Social and Semantic Implications o f the Structure of Belief,” in Paula Brown and Georgeda Buchbinder, eds., Man and Woman in the New Guinea Highlands (Washington, DC: American Anthropological Association, 1976), pp. 36 -5 3 . Abridged.

W ITCHCRAFT AND SEXUAL RELATIONS

ized nightm are of a group" (1970 [1951]: 263). Although the best early account, Evans-Pritchard's Witchcraft, Oracles , a n d M a g ic am ong the A za n d e (1937, 1976), emphasized the logic of practice and the self-sustaining quality of a belief system, most mid-century studies con­ cerned themselves with the purported functions, either psychological or social, of w itchcraft accusations. Historical accounts attem pted to explain epidem ­ ics of accusations and w hy certain cate­ gories of people are targeted. M arwick (1970) is the standard collection from this period. Kapferer (1997) interprets the elaborate anti-sorcery rites of Sri Lanka through a phenom enological account of the experiences of suffering from sorcery and being released from it, w hile Schneider (1990) is a stim ulating broader analysis of the decline in w itch­ craft in Europe. Recent Africanist work addresses w itchcraft in relation to

This paper is directed to the general objectives o f delineating the place o f w itchcraft within a larger system o f belief and - m ore im portantly - exp lorin g the significance o f interrelation­ ships between spheres or dom ain s o f belief within such w ider contexts. Both areas o f inquiry are susceptible to quite extensive exp osition and 1 here attem pt to probe only a restricted segm ent o f each. The specific o b je c­ tives o f the present effort are to analyze the relationship (within a N ew G uinea culture) between tw o discrete but interconnected d om ain s o f belief pertaining to w itchcraft and sexual relations, respectively, and to elucidate the sem antic interdependence between these d om ain s which is derived from both their m utual em beddedness within a larger system o f ideas and their an alo g o u s position s within this system . E toro thought and co sm ology are ordered by a central and fundam ental concept: that life and death are com plem entary and reciprocal aspects o f a larger process w hereby a spiritual “ life-force” is transm itted from one hum an being to another. Acts o f w itchcraft

239

m odernity, severe economic instability, rapid social change, and the excesses of the state (Com aroff and Com aroff, eds., 1993, Geschiere 1997, Weiss 1998). See also chapters 35 and 38 below by Taussig and De Boeck, respectively. Students are often confused by the diverse referents of the term "w itch­ craft" (Wicca, the European and Am erican w itch trials, w estern folklore and media images of figures on broom­ sticks, and the various ethnographic examples), and there is no reason w hy they should all be lumped to­ gether. Nevertheless, contem porary moral panics in North Am erica and northern Europe concerning such matters as child abuse can best be under­ stood by drawing on an anthropological perspective informed by studies of w itchcraft (La Fontaine 1998). On prac­ ticing, self-described "w itches" in England see Luhrmann (1989).

and acts o f sexual intercourse are the two m odes o f interaction through which this transm ission is effected. M oreover, “ life” and “ d eath ” are the reciprocal term s o f the tran s­ action in each case. R elation s between sexual partners are com parable to those between witch and victim by virtue o f this analogic conjunction, and the idiom s in which each sphere o f interaction is (respectively) grounded therefore stand in a m etaphoric relation to each other. The ju xtap osition o f w itchcraft and sexual relations enriches the cultural m eaning o f each and also provides a m echa­ nism for projecting the ch aracteristics o f actors in one dom ain upon their an alogic counter­ parts in the other, thereby contributing to the cultural delineation o f social position s and the relationships between social positions. These sem antic and social definitions are not directly specified by identifications (i.e., that x is y), but are products o f the structure o f the belief system itself, viewed in a w ider con text, in term s o f m etaphoric relations between dom ains.

240

RAYM OND C. KELLY

The E toro, w ho num ber about 4 0 0 , inhabit the southern slopes o f M t. Sisa which libs along the southern edge o f the central Cordil­ lera o f N ew G uinea facing on to the G reat Papuan Plateau. The residential unit in E toro society is a longhouse com m unity (of 3 5 to 4 0 people) which is ideally com p osed o f tw o patrilineage segm ents linked by sister exchange. The classificatory term inological system im poses this idealized organ ization on diverse em pirical arrangem ents such that every co m ­ m unity is constituted as tw o sets o f “ sib lin gs” w ho have m arried each other’s “ sisters” and “ b ro th ers.” T he longhouse itself . . . is divided into a com m unal section (tow ard the front) and separately enclosed m en’s and w om en ’s sleeping quarters in the rear. W idow s m ay sit only alo n g the outer w alls o f the com m unal section and m ay enter the dw elling only by side d o o rs. T he bach elors’ sleeping q uarters transect the m en’s and w om en's sections at the rear o f the longhouse. This spatial a rran g e­ ment o f social categories . . . is con cord an t with the sem antic delineation o f their interre­ lation sh ips discu ssed herein. The lineages which enter into local o rg an iza­ tion are sm all and o f limited genealogical sp an , seldom en com passin g agn ates m ore distantly related than FEBSS. “ Brother” relation sh ips between these patrilines are based on m atrilateral sibling-ship rather than descent (cf. Kelly 1973: 1 3 9 -9 2 ). Each patriline p o ssesses a num ber o f associated Sigisato spirits w h o are connected to the descent g ro u p through coow nership o f a territory from which both the spirit g ro u p and agn atic grou p draw their su s­ tenance. (The Sigisatos occupy the bo d ies o f casso w aries by day and subsist on the fruits and nuts o f lineage lands.) The souls o f m em bers o f the agnatic grou p are im planted in them by their associated Sigisatos , and the latter also a ssist an d protect lineage m em bers in a num ber o f w ay s, o f which several relating to w itchcraft m ay be m entioned. Acting through a m edium , a Sigisato cures illness resulting from w itch­ craft, confirm s the identity o f the witch resp o n ­ sible for a death , and m ay also decide the w itch’s fate by proclaim ing, “ 1 give you this person to kill.” (For am plification o f this very brief outline o f selected aspects o f social o r g a ­ nization relevant to my present concerns see

Kelly 1973; discu ssion o f the role o f w itchcraft in social processes such as lineage fission and local grou p form ation m ay also be found there.) The E toro believe every hum an being to be possessed o f tw o discrete spiritual aspects: the ausulubo , an im m aterial spirit double, and the hame , a life-force or an im atin g principle. Both are im parted to a child (at and before birth, respectively) by a Sigisato spirit associated with his or her lineage, an d both are (jointly) im m ortal, persisting beyond the grave in the form o f a Kesam e or spirit o f the dead. These spiritual aspects are significant to the present analysis in sofar a s they are affected by w itchcraft and sexual relations, and will be exam ined from that perspective. The ausulubo replicates the p o ssesso r’s physical body in form but lacks corporeal su b ­ stance. The sam e term is em ployed with reference to an individual’s sh adow , reflected im age, and the echo o f his voice, and these m ay be conceived a s visible (or aural) m an ifes­ tations o f the ausulubo. All three are ch arac­ terized by qualities o f duplication and separability which are essential attributes o f this spiritual aspect. The ausulubo is norm ally contained within the ow ner’s physical body but possesses a capacity for co n sciou s independent activity under certain circum stances. The average individual’s spirit dou ble m ay w ander only in dream s; its m ovem ents tend to be restricted to the im m ediate vicinity, and it does not n or­ m ally venture into the spirit realm s frequented by m edium s. In co n trast, the ausulubo o f a m edium m ay g o on extended excursion s to a variety o f spirit w orlds while he sleeps, during a seance, or even while his body carries out a norm al routine o f daily activities (although in a som ew hat distracted m anner). A witch (mugwabe) is likew ise cap ab le o f leading a double life, and acts o f w itchcraft are generally perpetrated by an ausulubo operatin g ap art from the body of its ow ner. The spirit double is invisible to an individual in his norm al state, and the witch can therefore carry ou t his n efar­ ious activities w ithout being seen or identified by his victim or oth ers present at the time. H ow ever, objects em ployed by the ausulubo o f a witch (e.g., an ax) can be seen, and the

W ITCHCRAFT AND SEXU A L RELATIONS

soun d o f tw igs sn ap p in g a s the invisible spirit dou ble ap p roach es through the forest are also audible. In add ition , the w itch’s ausulubo (which differs from that o f norm al individuals) gives o ff a pale glow , or m oropa , that betrays the presence o f a witch w ithout revealing his (or her) identity. The com m ission o f an act o f w itchcraft entails the infliction o f som e injury upon the spirit double o f the victim. In the typical case, som e foreign object is thrust or im pelled into the ausulubo o f an individual, cau sin g illness. The witch m ay then proceed ro dism em ber the ausulubo o f the w eakened victim , lim b by lim b, on successive nights. T hese parts are either consum ed by the witch o r hidden for later consum ption. R em oval o f the heart and liver (of the ausulubo) cau ses the final dem ise (and this m ay occur w ithout the intervening dism em berm ent). When the spirit double is injured, illness afflicts the correspon d in g p art o f the corporeal body; this provision applies to witch as well as victim. T h u s a witch w hose ausulubo has been burned with a torch or firebrand in the course o f an act o f w itchcraft can later be identified by the an alogou s burns on his physical body. An arrow shot into the heart o f rhe w itch’ s spirit double will likew ise kill the witch himself. Events such as these occur only in m yths and legends, but are nevertheless indicative o f the perceived relationship between rhe soundness o f the ausulubo and that o f its co rp oreal co u n ­ terpart. It is also im portan t to note that acts o f w itchcraft are perpetrated upon the ausu­ lubo o f the victim by that o f the witch, such that the entire tran saction takes place on w hat m ay be called the ausulubo “ plane o f existen ce” . The hame is the second m ajo r com ponent o f an individual’s spiritual con stitution . The bam e em bodies the an im atin g principle and vital energy o f hum an existence. It is form less, like the w ind, and is m anifested in breath (des­ ignated by the sam e term) in m uch the sam e w ay that the ausulubo is m anifested in a p erson ’s shadow or reflection. Srrength, vigor, an d vitality em anate from the hame. H ow ever, it lacks con sciousn ess as well as form , and is incapable o f op eratin g as an independent entity a p art from the co rp oreal body and

241

ausulubo which are m utually anim ated by it. T h is dual role is evident in accoun ts o f w itch­ craft. W hen a witch dism em bers the lim bs o f his victim ’s spirit double, the co rporeal lim bs are w eakened by the loss o f both their essence and their vitality (corresponding to elem ents o f the ausulubo and hame , respectively). A lthough the hame is thereby dim inished, it nevertheless continues to provide a reduced quantity o f vital energy to the corporeal flesh which rem ains - for it is im m inent in both the physical and spirit body. A sigisato spirit an im ates a fetus by im plant­ ing a nascent hame within it (and, at birth, im parts an ausulubo to the child). At the m om ent o f death , the residual hame is exhaled with the last breath and is subsequently merged with the ausulubo to form a single entity which persists a s a Kesam e , or spirit o f the dead. The ausulubo - which h as previously been m aim ed, dism em bered, and consum ed by a witch - is som eh ow reconstituted and united with the an alogou sly traum atized hame through a m ys­ terious supernatural process described by a special verb (keketosa ), elucidated by the p a ra ­ ph rase “ m any things becom e o n e .” H ow ever, the m anner in which this is effected is said to be beyond hum an understanding (perhaps because it is contradictory to other articles of belief). The evil spiritual aspect o f a witch that distinguishes him (or her) from norm al indi­ viduals is segregated in this process and ascends to the heavens to becom e a star, glow in g with the pale light o f the m oropa. T h u s purified of evil, the w itch’s hame-ausulubo descends to the river and associated underw orld which is the d om ain o f the Kesames. T he degree o f strength and vitality which em an ates from rhe hame varies during the course o f an individual’s lifetime. The hame m ay be conceived, in this respect, as a reservoir o f life-force which can be augm ented or dim inished. T his occurs in tw o w ays - through w itchcraft and through sexual intercourse. When a witch rem oves a lim b from an individ­ u al’s ausulubo an d con sum es it, he in corpo­ rates a portion o f the victim ’s life-force, and thereby augm en ts the strength o f his hame , while depleting that o f his victim . The witch grow s uncom m only large and acquires added physical strength and vigor through this

242

RAYM OND C. KELLY

ap p rop riation . T he children he (or she) begets “ as a w itch” are a lso physically large and are them selves identifiable as kagom ano or witchchildren by this quality. (This is one o f tw o w ays w itchcraft is acquired, a s is discussed further below .)1 The correspon d in g dim inution o f the victim ’s life-force is evident from the w eakened condition accom pan yin g illness and is also registered by labored breathing, shortw indedness, coughing, chest p ain s, an d the like. Breath is a physical m anifestation o f the hame which reflects its spiritual condition, and respiratory difficulties are taken to be indica­ tive o f depletion o f the hame* s reservoir o f life-force. Labored breathing an d short-windedness are on om atopoetically referred to as hame hah hah , and this condition is generally recognized as the outw ard sign o f a co n com i­ tant spiritual disability. It is, m oreover, a con ­ sequence o f sexual intercourse as well as a result o f bewitchment. Every adult m ale p ossesses a limited q u an ­ tity o f life-force which resides in his body as a w hole, but is especially concentrated in his semen. A portion o f this is expended in each act o f sexual intercourse so that a m an ’s reservoir o f life-force is grad u ally depleted over the course o f his lifetime. T h e vital energy which em an ates from the hame is co rresp on d ­ ingly dim inished, rendering the subject ener­ vated and enfeebled. An elderly m an - who w as once stron g and vigorous - is thus shortwinded and suffers hame hah hah when clim b­ ing hills or engagin g in other strenuous activity. The Etoro also point ou t that this self-sam e shortness o f breath is experienced upon co m ­ pletion o f the sexual act itself; the tem porary condition becom es general a s sem en is progressively lost. The life-force which a m an expen ds in sexual intercourse is transferred, through the process o f conception, to the children he begets (or, in the absence o f conception, is lost). T h is serves to augm ent the nascent hame im planted in each child by the Sigisato spirits, and co n trib­ utes to their early grow th. Pre-pubescent boys are insem inated by their elders so as to provide them with semen (which they lack), and also to supply the life-force or vital energy they require in order to grow , m ature, and develop

manly strength. The hame o f the youth is a u g ­ mented by this acquisition (while that o f his insem inator is depicted). The grow th-inducing properties attributed to semen are clearly expressed in these beliefs. It is evident here that w itchcraft and sexual relations are m utually em bedded in a single conceptual system insofar as they constitute the tw o m odes o f interaction through which life-force is transm itted from one hum an being to another. The interconnection is particularly w ell-illustrated by the aforem entioned belief that the child o f a witch will be physically large. This follow s from the fact that the parent has appropriated the life-force o f others through acts o f w itchcraft, and is thus capable o f transm itting an enriched infusion o f this to his offspring in the process o f conception. The witch-child grow s large in the w om b as a con ­ sequence o f this. T o explain why a large infant is the child o f a witch clearly requires an understanding o f the ideology o f sexual rela­ tions as well as the ideology o f w itchcraft, for the tw o interpenetrate. H ow ever, the relation goes deeper than this, insofar as acts o f inter­ course are in som e respects equivalent to acts o f w itchcraft, such that each stan ds in a m eta­ phoric relation to the other. The sem antic im plications o f this will be elucidated after beliefs concerning sexual relations are m ore fully explored. The associatio n o f sexual intercourse with im pairm ent o f the hame is expressed in a num ber o f restrictions which govern malefem ale relations and circum scribe the postcopulatory interaction o f a m an with his fellow s. A m an w ho has had intercourse with a w om an m ust refrain from offering food or to b acco sm oke (from his b am b o o pipe) to another m an throughout the rem ainder o f the day. Should an individual consum e such food or inhale sm oke from the pipe o f a co pu lator, he will experience hame hah hah and, in the latter case, a coughing spell as well. W om en are at all tim es enjoined from step­ ping over men’ s personal possession s, pipes and to b acco, and especially food and the items em ployed in its cookin g and consum ption. T hus a w om an m ust never step over split fire­ w ood or sit on the w oodpiles at the front o f the longhouse. If a m an consum es a m orsel o f

WITCHCRAFT AND SEXUAL RELATIONS

fo od cooked by such firew ood (or sm okes to b acco which has likew ise been subjected to this negative influence), he will suffer hame hah hah and a general lack o f strength while pursuing his daily activities. Should a w om an step over his a x , it will becom e dull. A m an m ust avoid seeing or com ing into contact with a new born child for 17 days. If he fails to do so , he will suffer especially severe hame hah hah and m ay die as a consequence. In addition , the father o f the child should not offer food or pipe to any m ale for seven days. The grow th o f a youth w ould be forestalled by con sum ption o f such food or sm oke; an adult m an w ould experience hame hah hah. In all these above instances, the enjoined behavior is tobi , i.e., stringently forbidden or tab oo. The E toro do not esp ouse any beliefs con ­ cerning fem ale pollution o r the contam ination o f objects through the addition o f im pure substan ces or essences which w ould account for the negative effects attributed to violation o f these regulation s, and I w ould suggest that they are best interpreted a s extensions o f the basic concept o f depletion discussed above. When a w om an steps over an object, it “ p asses between her th igh s,” and this is a sym bolic equivalent o f intercourse. An object such as an a x thus loses its cutting edge, or essential p ro p ­ erty, in the sam e way that a m an loses the essential property o f his m asculinity in co p u la­ tion, i.e., his sem en. The pipe sm oke o f a m an w ho has recently co p u lated , and fo od , fire­ w oo d , etc., subjected to sym bolic intercourse all have the sam e effect on a m an w ho internal­ izes them as intercourse itself; they are capable o f transm itting the depletion signified by hame hah hah. These objects are, in som e sense, negatively charged with respect to m asculinity, and they have the capacity to drain or negate the life-force o f a m ale w ho consum es them. A new born child h as the m ost pow erful an ti­ thetical effect as a consequence o f having lain within the w om b for such as extended period. It is an especially potent carrier o f depletion which can be transm itted to m ales by the mere sight o f it. Retrospective analysis suggests an im p or­ tant corollary to these effects o f the gestation period upon the child. T he E toro note that m ales attain physical m aturity later than

243

fem ales, and are particularly concerned to prom ote and insure m ale grow th and develop­ m ent through insem ination. A lthough I did not g rasp the associatio n during my fieldw ork, recorded conversation s strongly suggest that the perceived cause o f this retarded m ale devel­ opm ent which so concerns the Etoro is the birth o f men from w om en; late m aturation is a consequence o f the adverse effects upon m ales o f close con tact with the depleting pow ers o f femininity in the w om b - adverse effects which the early-m aturing fem ale o ff­ spring never experience. T he fact that food received from the father o f a new born child forestalls the grow th o f boys (but not girls) is consistent with this notion. T he cum ulative depletion o f the father is m anifested in the event o f birth itself, since a child possesses life-force a s a direct consequence o f his father’s loss o f it. The state o f a m ale parent after birth is co m parable in nature to th at o f a m an after copulation (although intensified in degree). Although w om en have a w eakening influ­ ence upon men (directly or indirectly) under certain specific conditions o f actual and sym ­ bolic intercourse, they are not a source o f p o l­ lution and therefore do not pose a continuous threat to m ale w ell-being. T h e absence o f any beliefs concerning m enstrual pollution is sig­ nificant in this respect. W om en d o not move to physically separate q u arters during m en­ struation , although they d o avoid the com m u­ nal section o f the longhouse (and generally rem ain in confinem ent w ithin the w om en’s section for several d ays). M en d o not perform protective m agic to coun teract fem ale co n tam ­ ination, and there are no g ro u p (or individual) purification rites for m ales such as those reported for other areas o f M elan esia, particu­ larly the N ew G uinea H igh lan ds (cf. M eggitt 1964; Allen 1967). While intercourse is depleting, the fem ale persona is not con tam inating, and men there­ fore need not avoid contact with w om en per se. It follow s that the general tenor o f malefem ale relations is neither constrained nor hedged with anxiety. M en and w om en mingle and interact freely in garden in g, in sago w orking, and in the com m unal portion o f the longhouse during the course o f daily activities. W om en are careful to observe the prohibitions

244

RAYM OND C. KELLY

on steppin g over certain item s (noted above), but this does not restrict or con strain their m ovem ents, since fo od is norm ally kept o u t o f the w ay in net b ag s hung from w alls or house posts, and m en’s p o ssessio n s are sim ilarly placed or are stored in the men’s q uarters. If a w om an does accidentally step acro ss food, men w ould refrain from eating it but w ould not be exercised by the incident (except perh aps in the case o f highly valued gam e or pork). Should a m an ’s pipe be affected, he w ould set it aside for several d ays (and borrow another) but w ould be unlikely to discard it unless p a r­ ticularly concerned ab o u t his health at the time. A lthough the ow ner o f the pipe m ight exp ress mild irritation , other men tend to regard the situ ation a s a hum orous one. Je sts m ade in such co n texts stem from the prevalent view that a m an w ho is truly stron g (keloi) can w ithstand the w eakening influence o f w om en w ithout noticeable ill effects. T h is view is also relevant to the m ale attitude tow ard m arriage and polygyny. A m an w ho has tw o wives and nevertheless rem ains vigorou s in spite o f the resultant double d ose o f depletion (through intercourse) takes pride in his dem onstrated strength and vitality, and is adm ired for this by others. The m ale attitude tow ard heterosexual intercourse itself is ch arged with am bivalence and am biguity. M en privately nurture variable degrees o f anxiety concerning the debilitating effects o f cop ulation . H ow ever, they can quite effectively alleviate such anxiety by engaging in the very act that they fear and not e xp eri­ encing the negative consequences believed to follow from it. A m an thereby develops selfconfidence in his residual vitality which is at once co m fortin g and highly p recarious. His attitude will shift rapidly to one o f deep concern if he feels enervated or develops a cough or other respiratory sign o f depletion. The op p o site poles o f m ale am bivalence are w ell-illustrated by the events o f an ad u lterou s liaison (as related to me by an inform ant). A young m an in his early tw enties w as secretly trysting with his FBW. W hen this w as d isco v­ ered, he tem porarily fled to another co m m u ­ nity to avoid his uncle’s w rath, returning after several d ays when com pen sation (o f one m other-of-pearl shell) had been arran ged by

his father. At the time this w as publicly paid, the cuckolded h usband m ade a speech in which he vividly described the depleting effects o f intercourse to his BS and other youn g men present at the time. T o w ard the conclusion o f this haran gue, he dram atically lifted his w ife’s skirt and exp o sed her genitals for all to behold. Several young men (including my inform ant) retched forthw ith and the adulterous youth him self becam e queasy and visibly discom fited. O lder men turned their faces aside with e xp res­ sions o f d isgust, while the chastened wife wept from sham e and repudiation . The point o f this tale is that the sp on tan eo u s vom iting clearly indicates a very real an d deep-seated anxiety, w hile the adulterous liaison docum ents the capacity to disregard it, at least tem porarily. (The parties to the tran sgression continued to co-reside w ithout further incident.) M en are protected from undue depletion (and anxiety pertaining to sam e) by prolonged p eriods o f enjoined abstinence. H eterosexual relations are narrow ly circum scribed with respect to perm itted tim es (and places) o f occurrence. In all, co pu lation is prohibited (tobi) for an estim ated 20 5 to 2 6 0 days a year. T he specific p roscrip tion s and the im puted results o f their violation are: (1)

(2)

(3)

(4)

(5)

T h ro u gh o u t the period from the co m ­ m encem ent o f a new garden until the trees have been felled (at the end o f the fourth m onth). If this tab o o is violated, the cro ps will not m ature properly, yields will be p o o r, and p igs will ravage the garden. From the time a sa g o palm is cut until processin g is com pleted (lest yields be poor). W henever deadfall trap s or snares have been set for wild pigs, m arsu pials, or casso w aries (else little o r nothing will be caught). D uring any period when sag o grubs are m aturing in an unprocessed palm cut for this purpose. If this restriction is not observed, the grow th o f any youth w ho con sum es the grubs will be stunted. When a new longhouse is under co n stru c­ tion (else the project will be beset with difficulties).

W ITCHCRAFT AND SEXUAL RELATIONS

(6)

While a tradin g party from the longhouse is carrying out an expedition and for four days prior to their departure (or little o f w hat is sought will be obtain ed).

These prohibitions are linked to virtually all m ajo r asp ects o f the productive econom y garden in g, sag o processin g, an d trappin g are nearly tan tam ou n t to subsistence. Exchange and construction are not om itted. M oreover, these activities tend to be segregated in time so th at there is relatively little overlap between one proscribed period an d another. T his im poses very substan tial lim itations on the frequency o f heterosexual relation s and there is, m oreover, indirect evidence o f general ad h er­ ence to the prohibitions. Eight o f ten births which occurred w ithin the tribe during a 15m onth period to ok place over a short span o f 3 1/2 m onths. Although the econom ic preoccu­ patio n s o f all ten pairs o f paren ts at the tim es o f conception are not precisely know n, the d is­ tribution o f births is predictable in term s o f the season al cycle o f subsistence activity. (The birth rate itself is also rather low .) H eterosexual intercourse should take place only in the forest, never w ithin a garden , in a garden dw elling, in the lon gh ouse, or the general vicinity o f the longhouse. (Even in the forest one can n ot be entirely at ease, for the E toro m aintain th at death adders are offended by the n oxio u s o d o r o f intercourse an d are particularly likely to strike a couple thus engaged.) V iolation o f the prohibition on co p u lation in and aroun d the longhouse is a serious offense which m ay p rovoke public rebuke and expu lsion from the com m unity. T he follow ing incident is a case in point. A w idow and an older bach elor (aged 28) h ad been carrying on an a ffa ir, sp oradically, for over a year. T h is w as a subject o f gossip an d private d isap p ro v al, but no public action. It is not uncom m on for younger w idow s to solicit the attentions o f m arriageable bache­ lors; indeed, this is frequently a prelude to rem arriage. H ow ever, when the couple w as discovered violating the location al proh ibi­ tion s by co p u latin g at the sprin g near the long­ h ouse, and subsequently w ithin the men’ s section (!) o f the lon gh ouse, the com m unity w as ou traged . The close kin o f the tran sgres­

245

sors were im pelled by public opin ion to put a sto p to this im m oral behavior. W hen the co m ­ m unity foregathered for the usual afternoon meal (several days after the latest offense), the w idow and bachelor were set upon and pummeled by three close kinsm en to the acco m p a­ nim ent o f a ch orus o f recrim inations and verbal abu se from the rem ain ing m em bers o f the longhouse. Both ran o ff to take refuge (separately) with kinsm en a t other locations. The bachelor returned a b o u t three d ays later and resum ed his residence at the com m unity after a public statem ent o f contrition; the w idow rem ained elsewhere. T he prohibition o f heterosexual relations in the environs o f the longhouse is also enforced upon the canine population. O n several differ­ ent o ccasio n s, the m em bers o f the longhouse com m unity where I resided spent the better part o f the day throw ing ston es at a bitch in heat and her coterie o f m ale adm irers in order to drive them into the forest. T he w om en were especially vigilant in this effort to ensure proper behavior on the p art o f the d ogs. In the first incident described abo ve, the w om en were also m ost ad am an t in d em an din g that strong m easures be taken to reestablish m oral behav­ ior. T hese d ata are significant in that they connote fem ale acceptance o f and adherence to the norm s described herein. These location al restrictions em phasize the fact that heterosexuality h as no place within the com m unity (or in inhabited areas), and is properly conducted only in the wild. It is fun­ dam entally antisocial beh avior in the strict sense o f the term . The post-partum and postco p u latory ta b o o s noted ab o ve accentuate this them e inasm uch as they proscribe the cu sto m ­ ary sh arin g o f food and to b acc o which is one o f the central m oral values o f com m unity life. H eterosexu ality segregates a m an and places him outside o f society an d, especially, outside the social com m unity o f m ales. The im puted effects which are said to follow from violation o f the (activity-related) tem po­ ral proh ibitions are also instructive. C o p u la ­ tion dim inishes the yields o f garden s, sag o palm s, trap s, and tradin g expedition s, just as it dim inishes a m an ’s life-force and depletes his ham e . T he central belief which is (once again) expressed here is a lso explicitly enunciated

246

RAYM OND C. KELLY

with respect to the undesirable effects said to result from co p u lation in garden s. The Etoro m aintain th at heterosexual intercourse in a garden will cau se the cro p s to wither an d die. The general consequences o f heterosexuality are death and depletion. H ow ever, this is only one com pon en t o f a com plem entary o p p o si­ tion between heterosexuality and h o m o sexu al­ ity, for the E toro also m aintain that hom osexual relations in a garden will cause the cro p s to grow , flourish, and yield bountifully. O ne o f the conceptual cornerstones o f E toro cosm ology is the view that accretion at one point in the system entails depletion elsewhere. Life cannot be created ex nihilo , and the birth (and grow th) o f one generation are in extrica­ bly linked to the senescence and death o f its predecessor. Life and death are com plem en ­ tary and reciprocal asp ects o f a larger process. In the con text o f this general concept, the gradu al depletion o f life-force which men experience in heterosexual intercourse is, at the sam e tim e, a precondition for the perpetu­ ation o f life through birth. The transference o f life-force is sim ilarly a precondition for the grow th and m aturation o f boys into men. Boys differ m ost im portantly from men in that they com pletely lack the m ost critical and essential attribute o f m an h ood, i.e., semen. The E toro believe, m oreover, that semen does not occur naturally in boys and m ust be “ p lan ted ” in them. If one does not plant sweet p o tato vines, then surely no sw eet p otatoes will com e up in the garden an d , sim ilarly, semen m ust be planted in boys if they are to possess it as men. M oreover, all asp ects o f m anliness are seen as consequences o f this acquisition. A youth is continually insem inated from a b o u t age ten until he reaches his early to m id-tw enties. T h is period is also m arked by rapid grow th in stature, increased physical strength and endurance, the sproutin g o f facial and body hair, and the developm ent o f m as­ culine skills and characteristics such a s hunting ability and co urageou sn ess in w ar. These em pirically ob servab le changes are uniform ly regarded as the direct results o f insem ination. (This is accom plished orally. The boy m an ip u­ lates the m an to the point o f ejaculation and consum es the sem en. The above effects are only realized through ingestion, and therefore

are not app licab le to heterosexual relations: w om en do not acquire strength, etc., in this way.) The recipient o f sem en in h om osexual inter­ course experiences beneficial effects which are the exact converse o f the negative effects su f­ fered as a consequence o f loss o f sem en in heterosexual intercourse. W hile the hame o f a m an is w eakened and depleted by copulation , the hame o f a youth is correspondingly strengthened by insem ination. Prior to this, boys are short-w inded an d tire readily; after­ w ards they p o ssess strength and endurance. T hese effects are thought to be cum ulatively m anifested over the m aturation period but are also apparen t on specific occasion s. For exam ple, young men say th at they are able to fell trees in a new garden for hours on end w ithout tiring after they have consum ed semen. These . . . beliefs . . . are all relatable to a general equation w hereby receiving semen : life, grow th, and vitality :: losing sem en : w eakness, senescence, and death. H ow ever, receiving semen is also culturally associated with h om osexu ality, and the loss o f it with heterosexuality such that the form er is to life as the latter is to death. This opp osition is expressed in m yth, and a lso in the previously cited beliefs that h eterosexual relations in a garden will cause the cro p s to wither and die, while h om osexu al relations will cause them to flourish and yield bountifully. A lthough in for­ m ants report that loss o f sem en in both types o f intercourse is equally enervating, the nega­ tive effects adhere only to heterosexual rela­ tions in form al ideology. T hus there are no rules which prohibit a m an from offering food or sm oke to others after hom osexual inter­ course, and there are no prohibitions which circum scribe such relation s with respect to time and place. M en and boys m ay properly engage in sexual relation s in the men’s section o f the longhouse (and in garden s) on any day o f the year. Indeed, this is essential if the youths are to grow and attain m an h oo d.2 The personal ch aracteristics which a youth develops as he m atures are believed to co rre­ spond to those o f his insem inator. If a m an is strong (keloi ), vigorou s in his advanced years, a proficient hunter and trap per, an d/or a co u ­

WITCHCRAFT AND SEXUAL RELATIONS

rageou s w arrior, then his protege will p ossess identical qualities and abilities upon attaining m an h ood. W itchcraft is also transm itted in this m anner. T he sem en o f a witch is said to contain m inute fro gs and w orm s which are the seed o f w itchcraft. A cquisition o f these predis­ p oses the ausulubo o f a witch to develop a m utant evil spiritual aspect: the tohorora. This distinguishes a witch from norm al individuals and endow s one with the extraordin ary c a p a ­ bilities that enable him (or her) to perform acts o f w itchcraft. (T ran sm ission o f w itch­ craft between wom en is discussed below .) The tohorora is located or concentrated in the heart and liver, and gives o ff a cold, p h osph o­ rescent light (the moropa) that is som etim es visible when a witch is active at night. T his can also be seen when the heart o f an executed witch is cut out in order to confirm his (or her) guilt. The organ (which is displayed on a stake) is said to glow like an em ber with the light o f the m oropa. The fro gs and w orm s th at are the seed o f w itchcraft are visibly present in the auricles and ventricles respectively. H ow ever, the acquisition o f these frogs and w orm s is not sufficient, in itself, to engender the develop­ ment o f the tohorora ; the seed o f w itchcraft m ust also be nurtured by the m alice and ill-will in a p erson ’s heart. The individual thus bears the ultim ate responsibility for the m utation o f his soul th at m akes him a w itch.3 Youths are initiated into m anhood in their iate teens or early tw enties, when they are physically m ature (although not fully bearded). At intervals o f about three years, all young men w ho have reached this stage o f develop­ ment go into seclusion at a lodge which is especially constructed for this purpose in an isolated area near the m argin o f the prim ary forest on the upper slopes o f M t. Sisa . . . A generalized insem ination o f the youths by older men . . . takes place at the seclusion lodge. T h is is im portant in that it m akes it im possible to retrospectively determine w ho m ay have transm itted the seed o f w itchcraft to w hom . If a young m an is nam ed as a witch, his principal insem inator is not necessarily im plicated. Sim ilarly, any attem pt to identify a particular individual a s the source o f w itch­ craft in a specific instance im plicates many young m en, som e o f w hom are likely to be

247

close kinsm en o f the individual draw in g the inference. A ttem pts to trace lines o f transm is­ sion are thus avoided by everyone. Confirm ed w itches are excluded from the seclusion lodge proceedings, and those w ho participate are presum ed to be free o f w itchcraft. Y oun g men w ho are subsequently accused are thought to have secretly consorted with w itches outside this context. Little inform ation is available concerning the beliefs and activities o f w om en, since it is im possible, in the context o f E to ro culture, for a m ale an th ropologist to develop an inform ant relationsh ip with a fem ale. M oreover, each sex is su pposed to be totally ignorant o f the private activities o f the other, and men generally respond to inquiries by saying that “ only the w om en know w hat the w om en d o .” H ow ever, w om en are thought to engage in som e form o f h om osexu al activities and to transm it the seed o f w itchcraft in this w ay. M en app ear to be genuinely ignorant concerning the p rocesses o f fem ale m aturation , and d o not know whether this occurs naturally or as a consequence o f tran sm ission o f substance. Som e inform ants thought it plausible that girls obtained m en­ strual blood from m ature w om en in the sam e w ay that boys obtained sem en from men; oth ers pointed out that fem ales possess blood from birth (even though they d o not m enstru­ ate until later), while boys com pletely lack sem en. The Sigisato spirits determ ine the sex o f a child by im planting a m ale or fem ale ausu­ lubo within it. Sex is therefore an aspect o f the soul, and m ales and fem ales are spiritually dif­ ferentiated. The fem ale hame is considered to be w eak and underdeveloped, so that wom en are alw ays short-w inded in com parison with men. In the preceding discussion , I have attem pted to establish and docum ent the point that an array o f E toro beliefs concerning sexual rela­ tions m ay be interrelated through a general equ ation whereby receiving se m e n : life, grow th , and vitality :: losing sem en : w eakness, senescence, and death. This form ulation also con stitutes the specific ideational content o f a m ore general and fundam ental conceptual ori­ entation o f E toro cosm ology, viz • the concept that the total system is closed and bounded

248

RAYM OND C. KELLY

such that accretion at one node necessarily entails a co rresp on d in g depletion at another (and vice versa). T h is general concept is also expressed in the d om ain o f w itchcraft belief as is evident from m aterial adduced earlier. A witch con sum es portion s o f the ausulubo and hame o f his victim , and thus grow s unusually large and vigo ro u s, while the victim is w eak ­ ened an d enfeebled by the resultant illness. The hame o f the victim is depleted (as evidenced by respiratory distress), while that o f the witch is augm ented such that he fathers children o f exception al size. After the victim ’s dem ise, the witch is thought to return to the burial p la t­ form by night to fatten him self on the flesh o f the corpse. An executed witch (and external enem ies killed in w arfare) are subject to ca n ­ nibalism , alth ough other m em bers o f the E toro tribe are not. T he flesh o f the witch is co n ­ sum ed ju st as he con sum es the flesh o f his deceased victim s. (H ow ever, no spiritual ele­ m ents or person al attributes are transferred in can n ib alism , and this view is understandable in sofar as the soul o f a witch is inherently evil and his ch aracteristics totally undesirable.)4 T here is a further correspondence between these tw o d om ain s o f belief which follow s from the structurally an alo g o u s position o f w itch­ craft and sexual relations as alternate m odes through which life-force is transm itted. This brings the tw o sets o f belief into an im m ediate relation such th at each is partially defined with reference to the other. T his is evident when the equ ation noted abo ve is reform ulated at a m ore general level, viz* augm en tation o f the hame : life, grow th , and vitality :: depletion o f the hame : w eakness, senescence, and death. H om osexual relations (from the stan d poin t o f a youth), acts o f w itchcraft (from the perspec­ tive o f the w itch), and conception (vis-a-vis the child) are all interchangeable in term s o f a u g ­ m enting the ham e , and are equally productive o f life, grow th , and vitality for the youth, witch, and child, respectively. Sim ilarly, hom osexual and heterosexual relations (for an adult man) and acts o f w itchcraft (from the perspective o f the victim) are interchangeable with respect to depletion o f the hame and the effects which follow from this. There is, m oreover, a threefold identification o f youth, w itch, and w om an as agents o f depletion in ad d ition to the

identification o f child, youth, and witch as beneficiaries thereof. The logical derivatives o f these identities are the opposition al equations: child : father protege : inseminator (youth) woman : man (wife) (husband)

::

witch : victim

All these invidious co m p arison s are applicable to E toro thought and behavior in som e degree, although each is elaborated in a som ew hat different w ay, and is thus m odified in its final ram ifications. T h is elaboration takes the form o f internal differentiation by con trasts (and further sim ilarities) am o n g the four term s which are here identified (i.e., child, youth, w om an, and witch). A witch is the epitom e o f m aliciousness and antisocial selfishness, for he (or she) feeds on the souls and bodies o f others out o f spite, and with full intent to cause harm ; he sates his gluttony as well as his hate and grow s large and vigorou s at oth ers’ expense. A w om an , on the other hand, does not augm ent her hame through heterosexual relations. The loss o f life-force a m an suffers in this context con trib­ utes to the conception and early grow th o f his children. A w om an is thus an agent o f deple­ tion but not a beneficiary and, conversely, a child is beneficiary but not agent. N either m anifests the intentional m aliciousness o f the w itch, and the dim inution o f a m an ’s hame is here viewed as p art o f the essential tragedy o f hum an existence th at senescence and death are precondition s for the perpetuation o f life through birth. H ow ever, excessive copulation and sexual relations which are not productive o f offsprin g deplete a m an to no end. A w om an w ho encou rages, entices, or dem an ds her husband to engage in needless copulation - from which he alone will suffer - thereby app roach es the purely negative role o f witch. She sates her sexual appetite selfishly, know ingly causing harm , and perh aps with m alicious intent. (This con notation applies with double force to a w om an w ho co n so rts with an im m ature youth w hose grow th and developm ent will be p erm a­ nently arrested by loss o f sem en.) The identifi­ cation o f w om an and witch therefore refers in

W ITCHCRAFT AND SEXU A L RELATIONS

only a restricted sense to an inherent quality o f w om an hood while being fully applicable to the potential ch aracteristics o f aggressive and dem anding feminity and unregulated fem i­ nine sexuality. T he associatio n thus con trib­ utes to a co n strain in g delineation o f the fem ale role and is not merely a static negative characterization. W itchcraft also has its prim eval origin in the epitom e o f unregulated sexuality - the act o f incest. M oreover, the myth which recounts the event em ph asizes the responsibility o f the w om an for releasing this agent o f death upon m ankind. At a longhouse community of the distant past there lived a brother and sister (of about age 8 to 10). One day they went to the forest together and copulated secretly there. The girl became pregnant and grew large with the child. “ What have you been doing?” an elder man asked here. But she didn’t answer: she turned her head aside and said nothing. In due course she gave birth to a boy. Secretly, she suckled it at her breast. The elders did not know o f this at first, but one day a man came upon her as she was nursing the child. As he drew near, she tried unsuccessfully to conceal the baby from him. They had become witches, these three, and the elders decided they must be killed. The men caught the brother and sister and struck them and they cried out in pain as they died. Then, for the first time, the men heard the (now characteristic) whistle of the witch from the other side o f the longhouse. “ Tua, tua” shrilled the witch-child as he slipped away. “ N ow we have witches and men will die,” an old man lamented. “ Before today there were none, but these two became witches and henceforth we will die.” A ccording to E toro exegesis, the precocious siblin gs were tran sform ed into w itches by their in cestuous conjunction itself, and the first w itch-child (or kagom ano) w as the fruit of their union. T hus incest begat w itchcraft, and introduced sickness and death into the w orld. Before this tim e, men were im m ortal. It is im portan t to note that the agent o f death is “ o f w om an b o rn .” T he girl refuses to co n fess her evil act when queried by her elders, an d later conceals the child from them . W orst

249

o f all, she suckles the w itch-child rather than subm ittin g it to infanticide (as w ould be expected under the circum stances). By not only bringing death into the w orld but nurtur­ ing it at her breast, she thus tran sfers to w om en the ultim ate responsibility fo r m ankind’s lost im m ortality. It is a lso significant that the evil o f unregulated sexuality (here represented in its m ost extrem e form ) is ontologically prior to the evil o f w itchcraft. T h e w idow and children o f a deceased man have both contributed to his dem ise (as agent an d beneficiary o f his depletion) and are subject to addition al food restrictions a s a result. A bout 7 7 percent (65/83) o f the edible native ta x a o f gam e an im als m ay not be consum ed by a w idow , and 71 percent (59/83) are like­ w ise prohibited to the son s and daugh ters of rhe deceased below the ages o f 17 and 12, respectively. Only ap p rox im ately 35 percent o f the total anim al protein in the diet is av ail­ able to these tw o m ost heavily restricted social categories o f individuals, w hile the w ives and children o f living men have access to 9 7 percent and 74 percent, respectively (cf. Kelly 1973: 5 6 - 9 ). T hese restrictions are not integral to the three-year m ourning period for w idow s as they extend beyond it. They are m aintained until a w om an rem arries, or indefinitely if she does not (with the exception o f one tab o o which is rem oved when she p asses ch ildbear­ ing age). T he children are relieved o f their ta b o o s at ab o u t the time they reach puberty. T hese regulations are significant in that they follow from the joint responsibility o f w idow and child for a h usban d/fath er’s death, and a lso im pose disabilities con sequent upon this responsibility. T o be “ eaten ” is a euphem ism for sexual intercourse (of either variety), and a m an is indeed spiritually con sum ed by it. It is therefore ap p ro p riate thar the con sum ption o f those w ho have partaken o f him be restricted after his dem ise. The fo o d ta b o o s also dem on­ strate the im pact o f the ideological features under con sideration upon behavior outside their im m ediate dom ain. T he position o f the uninitiated youth differs from that o f a w om an or child in that lie is both agent and beneficiary o f his in sem in ator’s depletion. H ow ever, frequent hom osexual intercourse within the co n text o f this relation ­

250

RAYM OND C. KELLY

ship is not negatively regarded, since a youth grow s in size, strength, and prow ess in direct proportion to his consum ption o f life-force (and is also thought to require a heavy dosage in order to attain full m aturity). H ere again we encounter the tragic necessities o f hum an e x is­ tence. The full associatio n o f youth with witch turns on unregulated sexuality in this instance as well, but concerns in appropriate liaisons (and not excess within sanctioned relation ­ ships). It is the youth w ho co n so rts with others in his age class - and enhances his ow n grow th and vitality at their expense - that replicates the behavior o f the witch. The im m ature young men from w hom the culprit draw s life-force will suffer arrested or retarded developm ent at the least, and m ay well die young as a result. The youth w ho garn ers this addition al lifeforce m anifests precocious, preternatural m at­ uration, and will not only reach puberty at the early age o f 13 or 14 but will also possess a fairly w ell-developed beard shortly thereafter (m ore specifically, the beard o f a m an in his early to m id-twenties). Any individual w ho possesses this trait (i.e., early beardedness) is thought to have taken advan tage o f other im m ature youn g m en, and is included within the cultural category sa:go which design ates such tran sgressors. Sargos are not only like w itches but are universally deem ed to be w itches, and especially vicious ones at that. As in the case o f w om en, the associatio n o f unini­ tiated youth with witch is fully ap plicab le only with respect to a potential characteristic, and thus serves to regulate and bound sexual behavior. T he category sa:go is a particularly im p or­ tant one inasm uch as it represents an intersec­ tion o f the ideologies o f w itchcraft and sex u al relations. A sa:go is deem ed a witch as a co n ­ sequence o f his sexual tran sgression s. While all s a ^ o s are thought to be w itches, the converse does not hold, and the category thus co n trib­ utes in a unique way to the cultural specifica­ tion o f the defining qualities o f the witch. A s a ^ o is the m ore despicable o f the tw o. His essential depravity turns on the fact that he intervenes in and perverts the norm al tran s­ m ission o f life-force prescribed by the “ n atu ral” order o f the system , exp rop riatin g for him self that which w ould otherw ise provide for the

grow th and m aturation o f others. Such beh av­ ior is not only evil (by virtue o f its harm ful effects upon the depleted victim) but also su b ­ versive; it is a crim e again st both “ n atu re” and hum an society. A witch (other than a s a ^ o ) also intervenes in natural process by ap p rop riatin g life-force that w ould otherw ise be bestow ed upon the next generation, but acquires only a portion o f this, as his victim s have typically reproduced them selves to som e extent. H e foreshortens the life-span o f individuals w ho w ould otherw ise expire from sexual depletion at an advanced age, but does not so severely threaten the perpetuation o f life itself a s does the sa ^ o . The sargo em bodies and m ost fully expresses the full range o f dim ension s which com prise the Etoro conception o f evil. W itchcraft and sexual relations occupy a n a lo ­ gou s structural position s within a larger con ­ ceptual system - a system in which life and death are com plem entary and reciprocal aspects o f the tran sm ission o f life-force. The an alogic relation engenders a sem antic inter­ play between the tw o constellation s o f beliefs in which w itchcraft and sexual intercourse are ideologically groun ded. T h e sem antic dim en­ sions o f each set o f beliefs is enriched and expan ded by their ju x tap o sitio n ; the cultural m eaning o f w itchcraft is inform ed by the idiom o f sexual relations (and vice versa). M oreover, this ju x tap o sitio n establish es a m etaphoric relation through which the attributes o f actors in one dom ain are projected upon their co u n ­ terparts in the other, such th at social position s are a lso invested with added dim ensions o f m eaning. M ore specifically, the an alogic co rre­ spondence between acts o f w itchcraft and acts o f sexual relation s connotes a like relation between the ch aracteristics o f the (respective) actors. At the sam e tim e, both the acto rs and their interaction are delineated - in the dom ain o f sexual relation s - by social roles. T he social position s o f w ife, protege, and (to a lesser extent) child thereby acquire a penum bra o f negative attributes (derived from the ch aracter­ ization o f the witch) that sharply dem arcate the boun daries o f acceptable behavior, particularly in the sexual sphere. Behavior which exceeds these boun ds is not merely erroneous, ungram ­

WITCHCRAFT AND SEXUAL RELATIONS

m atical, or tran sgressive; it replicates the behavior o f the witch and is categorically “ evil.” The strongest negative sanction - the w itchcraft accu sation - m ay be m arshalled again st those w ho violate such n o rm s.' It is instructive here to co n trast the respective cul­ tural evaluation s o f a m an and w ife w ho engage in excessive sexual relations. He is, at w orst, foolhardy in his beneficent generosity with the gift o f life-force, while his w ife - party to the self-sam e acts - is thought to be a witch. T hese cultural definitions o f certain p aram ­ eters o f a set o f social position s arc the direct product o f a specific structural arrangem ent within the larger belief system , viz. a m eta­

NOTES The d ata on which this paper is based were collected during a fieldw ork period extending from April 1968 to Ju ly 1969. 1

2

The Etoro m aintain that the child o f a witch and non-witch will not norm ally p o ssess this characteristic. T he offsp rin g o f tw o witches are thought likely to be k ago­ m ano , but are not necessarily so. These contingencies turn on the belief that a witch m ust copulate “ a s a w itch” in order to beget a kagom ano, and only a pair o f them are likely to d o so (perhaps because the lone witch w ould reveal his true iden­ tity to his “ n atu ral” spouse by such behav­ ior). In addition , tw o natural parents m ay have a witch-child born to them if the w om an is unknow ingly raped by the ausulubo o f a witch while she sleeps. The birth o f a kagom ano therefore does not neces­ sarily im plicate either parent. All physi­ cally large infants are presum ed to be kagom ano, and are subm itted to infanti­ cide. (Thirty percent - 3 /1 0 - o f all children born during the 15 m onth period o f my fieldw ork met this fate.) O nly w itches are thought to preserve such offspring. Both types o f sexual relations are uniform ly regulated by kinship. W ives and young m ale proteges should be selected from a m o n g those individuals w ho are within

251

phoric relation between tw o dom ain s o f belief which are discrete an d largely au ton om ou s at the con scious level. T his structural relation is, m oreover, a m echanism for the production o f an elem entary system o f inequality based on age and sex. (All m ature w om en occupy the social position “ w ife,” and all young men are likew ise “ proteges” .) R elation s o f inequality between adult m ale and adult fem ale, young m an and child, are delineated by the relation­ ship o f these categories to the encom passin g system o f life-force transm ission and by the structural relation between the idiom s o f w itchcraft and sexual relation s within this belief system .

the kinship category au a (m .s.); husbands and insem inators are found w ithin the reciprocal category naua (m .s. and w .s.). (See Kelly 1973: 305 -1 1 for discussion o f this kin category.) The ideal insem inator (am on g those eligible by kinship) is a boy’ s true sister’s husband or her betrothed; brother and sister will then receive semen from the sam e m an (ideally a FM BS) and be, in a sense, co -spo u ses to him. The young m an is insem inated by his Z H from ab o u t age ten until he is fully m ature - as evidenced by a m anly beard - in his early to m id-twenties. U pon reaching m aturity he will in turn becom e the insem inator o f his WB (or the brother o f his wife-to-be). It is prohibited for a m an to be both donor an d recipient o f semen during the sam e period o f his life, and m arriage therefore n ecessitates a transition to insem inator, although this m ay occur anytim e after full m aturity is attained (irrespective o f m arital statu s). T hus when the ego o f our exam ple is a b o u t 4 0 , his WB m akes this transition, an d h om osexu al relations between them necessarily cease. At this tim e, ego’s kinship ob ligation s to act a s insem inator are ful­ filled. Subsequent relationsh ips m ay be con tracted in the event o f polygyny, or if ego is especially vigorou s and therefore ask ed to serve in this connection (He will, in any event, continue to serve as an insem ­ inator in the con text o f initiation cerenio-

252

3

4

RAYM OND C. KELLY

nies.) It should be noted that a lather arran ges his d augh ter’s m arriage (when she is ab o u t five years old) and thereby n om i­ nates his so n ’s insem inator, since these are ideally one and the sam e. The entire tran s­ action involves the exchange o f w om en for w om en at one level, the exchange o f boys for boys at an oth er, the exchange o f w om en for life-force at still another level, and finally, the exchange o f life and death (for youth and insem inator, respectively). D uring the critical period o f a boy’s grow th and developm ent (in his early teens), he will reside in the sam e com m unity a s his hom osexu al partner. M en related as WF and DFI frequently co-reside, and the b o y ’s residence with Z H therefore does not n or­ m ally entail sep aratio n from his father (F and Z H being related to each other as WF and DFI). Residence patterns are exten ­ sively discu ssed in Kelly (1 973). A p redisposition tow ard developm ent o f the desirable ch aracteristics o f on e’s insem ­ inator is a lso transm itted in the sem en itself (rather than by tutelage). Som e young men thus thought it conceivable that they m ight acquire the capabilities possessed by person s o f W estern culture (and hence the associated m aterial culture) by ingesting their sem en, and they asked me to m ake arran gem en ts for them to go to the co ast as con tract lab orers so that this experim ent m ight be tried. The basic concept o f a circular closed system entailing a reciprocal relation o f life and death is m anifested in other asp ects o f cosm ology which are outside the scop e o f this paper but m ay be briefly m entioned.

5

The spirits o f the dead occupy the bodies o f a certain fish (ab a so ) and reside in the rivers. The Sigisatos are descended from a prim ordial crocodile (and culture hero), and at tim es take this form (although they characteristically reside in the bodies o f casso w aries by day). T he crocodiles feed on the ab aso , and this sym bolically closes the circle o f life an d death inasm uch as the Sigtsatos a lso im plant sou ls in men. In other w ords, the Sigisatos sym bolically consum e “ d eath ” an d a lso bring forth life. A witch nam ed as respon sible for one death is likely to be nam ed again , and thrice-nam ed w itches are characteristically executed. T he full range o f social d isab ili­ ties im posed upon a witch are to o exten­ sive to be enum erated here, although it m ay be noted th at these also include o stra ­ cism for p eriods o f up to a year.

REFERENCES CITED Allen, W. R. 1 9 6 7 . Male Cults and Secret Initiations in M elanesia. M elbourne: M elbourne University Press. Kelly, R. C 1973. F^toro Social Structure: A Study in Structural C on tradiction . Ph.D. dissertation . University o f M ichigan. Ann A rbor: University o f M ichigan Press (1977). M eggitt, M . J . 1964. M ale-Fem ale R elation ­ ships in the H igh lan ds o f N ew G uinea. In N ew G uin ea: T he C en tral H ighlands. J. B. W atson, ed. American Anthropologist 6 6 (4 , Pt. 2): 2 0 4 -2 4 .

21

The Politics and Poetics of Transgression Peter Stallybrass and Allon White

A llon W hite (1951-88) was an original literary and cultural critic (see W hite 1993); Peter Stallybrass is professor in the hum anities at the University of Pennsylvania. Am ong Stallybrass's sub­ sequent w ork is an elegant essay on "M arx's coat" found in the intriguing collection B order Fetishisms (Spyer 1998). Stallybrass and W hite illustrate the exciting conjunction of literary and his­ torical studies w ith anthropology. They bridge the w ork of Bakhtin (1968) w ith anthropologists like Gluckm an (1965)# M oore and M yerhoff (1977), Babcock, ed. (1978), and Douglas (1966), on car­

A m ongst the m any rem arkable things to be found in Ernst R obert C u rtiu s’s F.uropean Literature and the l.ater Middle Ages is an acco u n t o f how the idea o f “ the C lassic au th o r” w as originally derived from ancient taxatio n categories. In a chapter d iscu ssin g the idea o f m odel auth ors and the atten dan t notion o f

nival and forms of structural inversion and add a strong mix of Marx and Freud. Here they review th e literature on rituals of rebellion - events th at permit, encour­ age, or dem and the inversion of social norms - and ask w hether such carnivalesque spaces and images serve to energize or challenge dom inant and hegem onic social and cultural orders. The issue is central to th e study of popular culture as w ell as religion, and forces us to acknow ledge th e pervasive­ ness of the aesthetic and social coding of "high" and "low " (refinem ent and vulgarity) and the prevalence of bodily metaphor.

can on -form ation , C u rtiu s e xp lain s how taxban ds, a social division o f citizens according to property qualification s under the constitu­ tions o f Servius, were ad o p ted by A ulus G ellius a s a w ay o f designating the prestige and rank o f w riters. Citizens o f the first tax atio n category, the to p rank, cam e to be know n as

From Peter Stallybrass and Allon White, “ Introduction,” in The Politics and Poetics o f Transgression (Ithaca, N Y and London: Cornell University Press and Methuen Books, 1986), pp. 1-20. Abridged.

254

PETER STALLYBRASS AND ALLON WHITE

“ classici ” . T h is developm ent in the generic ter­ m inology o f antiquity (Gellius fl. c. 1 2 3 -c. 165) subsequently had an enduring influence on rhe European system o f hierarchizing auth ors and w orks. It separated out a distinct elite set (the classici) from the com m onality (rhe proletariats ) and used this as a m odel for literary discrim inations. C urtius rem arks: But it was not until very late, and then only in a single instance, that the name classicus appears: in Aulus Gellius (N odes Atticae, X IX , 8, 15) . . . The thing to do is to follow the usage o f a model author: “ e cohorte ilia dum ataxat antiquiore vel oratorum aliquis vel poetarum, id est classicus adsiduusque aliquis scriptor, non proletarius” ; “ some one of the orators or poets, who at least belongs to the older band, that is, a first class taxpaying author, not a proletarian” . . . T he proletaries , whom Gellius mentions by way o f com pari­ son, belongs to no tax class. When SainteBeuve, in 1850, discussed the question What is a Classic?, he paraphrased this passage in Gellius: “ un ecrivain de valeur et de marque, un ecrivain qui compte, qui a du bien au soleil, et qui n’est pas confondu dans la foule des proletaires.” (Curtius 1953: 249-50) And C urtius ad d s wryly “ W hat a titbit for a M arx ist sociology o f literature!” It w as not C u rtiu s’s inclination to follow up his ow n rem ark and indeed we can detect a certain am used astonishm ent when he realizes where his investigations have finally led him. From the first it seem s that the ranking o f types o f au th o r w as m odelled upon social rank accordin g to property classifications an d this interrelation w as still being actively invoked in the nineteenth century. In recent tim es we have been inclined to forget this ancient and en du r­ ing link between social rank and the o rg an iz­ ing o f auth ors and w orks, including literary genres, although for the m ajor p art o f F’ uropean history it w as a natural assum ption for readers an d w riters alike. Ian Ja c k noted that precisely this habit o f ranking “ kinds in a hierarchy an alo g o u s to that o f the sta te ” has led to a con tem porary distrust, particularly o f R enaissance categories: Just as the social hierarchy was traced from the prince through the nobility down to the

common people, so the realm of Poetry had its own “ degrees” , from Epic, the Prince of all the kinds, down to the lowest species o f all, “ from Homer to the Anthologia , from Virgil to M artial and Owen’s Epigrams . . . that is from the top to the bottom of all poetry” . (Jack 1942: 4) It is the contention o f the present book that cultural categories o f high and low , social and aesthetic, like those m entioned above but also those o f the physical body and geograph ical space, are never entirely separable. The ranking o f literary genres or auth ors in a hierarchy a n alogou s to social classes is a particularly clear exam ple o f a much broader an d m ore com plex cultural process whereby the hum an body, psychic form s, geograph ical space and the social form ation are all constructed within interrelating and dependent hierarchies o f high and low . This book is an attem pt to m ap som e o f these interlinked hierarchies on the terrain o f literary and cultural history. M ore particu­ larly it attends both to the form ation o f these hierarchies and to the processes through which the low troubles the high. The high/low opposition in each o f ou r four sym bolic dom ain s - psychic fo rm s, the hum an body, geographical space and the social order - is a fundam ental basis to m echanism s o f ordering and sense-m aking in European cultures. D ivi­ sions and discrim inations in one dom ain are continually structured, legitim ated and d is­ solved by reference to the vertical sym bolic hierarchy which operates in the other three dom ains. C ultures “ think them selves” in the m ost im m ediate and affective w ays through the com bined sym bolism s o f these four h ierar­ chies. Furtherm ore (and this is where the title o f the book com es in) tran sgressin g the rules o f hierarchy and order in any one o f the dom ains m ay have m ajo r consequences in the others. Although there are all so rts o f subtle degrees and g rad ation s in a culture it is striking that the extrem es o f high and low have a special and often pow erful sym bolic charge. T h u s, in the exam ple given above from C urtius, Gellius (as a lso C icero and A rnobius) im m ediately fixes upon the to p and bottom , the classicus and rhe proletarius , even though the system

THE POLITICS AND POETICS OF TRANSGRESSION

has five different grad es w ithin it. T his does not necessarily m ilitate again st subtlety since “ a b o v e ” and “ below ” m ay be inscribed within a minutely discrim inatory system o f classifica­ tion, but it does foster a sim plifying binaryism o f high an d low within which further classifi­ cation will be m ade. In other w ords the verti­ cal extrem ities fram e all further discursive elaboration s. If we can g rasp the system o f extrem es which encode the body, the social order, psychic form and sp atial location, we thereby lay bare a m ajor fram ew ork o f d is­ course within which any further “ redress o f balan ce” or judicious qualification m ust take place. In our study therefore we have focused upon the sym bolic extrem ities o f the exalted and the base. We have follow ed the instruction o f Boethius - “ Look to the highest o f the heights o f heaven” - and we have also plum bed the depth s o f social classification , the low er bodily stratum , the sew ers, the underw orld - w hat one might call the rock bottom o f sym bolic form . We have tried to see how high discourses, with their lofty style, exalted aim s and sublim e ends, are structured in relation to the d ebase­ m ents and d egradatio n s o f low discourse. We have tried to see how each extrem ity structures the other, depends upon and invades the other in certain historical m om ents, to carry political charge through aesthetic and m oral polarities. Indeed, the o p p o sitio n s, interpenetrations and tran sgression s o f high and low bear such an enorm ous weight o f cultural organ ization that one m arvels at the sheer lab our o f tran scodin g, displacem ent and partition involved in the elaborate netw orks o f super- and sub- in our cultural history. It w ould be w rong to imply that “ high” and “ lo w ” in this con text are equal and sym m etri­ cal term s. When we talk o f high discourses literature, ph ilosoph y, statecraft, the lan guages o f the Church and the University - and con ­ trast them to the low discourses o f a peasantry, the urban p oor, subcultures, m argin als, the lum penproletariat, colonized peoples, we already have tw o “ high” and tw o “ lo w s” . H istory seen from above and history seen from below are irreducibly different and they con ­ sequently im pose radically different perspec­ tives on the question o f hierarchy.

255

Indeed they m ay and often d o p o ssess quite different sym bolic hierarchies but because the higher discourses are norm ally associated with the m ost pow erful socioeconom ic grou p s exist­ ing at the centre o f cultural pow er, it is they which generally gain the authority to designate w hat is to be taken as high and low in the society. This is w hat R aym ond W illiam s calls the “ inherent dom inative m o d e ” and it has the prestige and the access to pow er which enables it to create the dom inant definitions o f supe­ rior and inferior. O f course the “ lo w ” (defined a s such by the high precisely to confirm itself a s “ h igh” ) m ay well see things differently and attem pt to im pose a counter-view through an inverted hierarchy. There is a grow ing body o f research devoted to rhe topic o f hierarchy inversion, o f “ w orld upside d o w n ” (W U D), and we have m uch to say ab o u t this in the follow ing pages. H ow ever the politics o f hierarchy inversion as a ritual strategy on the part o f subordinate grou p s is not our principal theme. We have chosen to concentrate rather on the contradictory nature o f sym bolic hierarchies w ithin the dom inant con struction s o f literature, the body and the social form ation. The prim ary site o f co n tradiction , the site o f conflicting desires and m utually in com pati­ ble representation, is undoubtedly the “ lo w ” . Again and again we find a striking am bivalence to the representations o f the low er strata (of the body, o f literature, o f society, o f place) in which they are both reviled and desired. R epug­ nance and fascination are the twin poles o f the process in which a political im perative to reject and elim inate the debasin g “ lo w ” conflicts pow erfully and unpredictably with a desire for this O ther. Edw ard Said in his w ork on Orien­ talism - the myth o f rhe M iddle East con ­ structed by Europe to legitim ate its own authority - has convincingly show n this o p e ra ­ tive am bivalence in action. In political term s O rientalism “ depends for its strategy on [a] flexible positional superiority, which puts the W esterner in a w hole series o f possible rela­ tionships with the O rient w ithout ever losing him the upper h an d ” . But at the sam e time Said notices that “ European culture gained in strength and identity by setting itself o ff again st the O rient a s a sort o f . . . underground self”

256

PETER STALLYBRASS AND ALLON WHITE

(Said 1979: 3, 7). “ An underground self with the upper h an d ” . T h is cu riou s, alm ost oxym oronie form ulation cap tu res a nexus o f pow er and desire which regularly reap p ears in the ideological con struction o f the low -O ther. It is nor only a phenom enon o f colonial and neo­ colonial representation. We find the sam e co n ­ stitutive am bivalence arou n d the slum and the dom estic servant in the nineteenth century; aroun d the d isp o sal o f “ w aste ” p rod u cts in the city (though not in pre-R enaissance rural culture); aroun d the carnival festivity o f p op u lar culture; aroun d the sym bolically base and abject an im als like the pig and the rat. T hese are the subjects o f specific ch apters which follow in which we explore the con tradictory and unstable representation o f low -O thers. A recurrent pattern em erges: the “ to p ” a t­ tem pts to reject and elim inate the “ b o tto m ” for reason s o f prestige and statu s, only to d is­ cover, not only that it is in som e w ay fre­ quently dependent upon that low -O ther (in the classic w ay that Hegel describes in the m aste rslave section o f the Phenomenology ), but also that the to p includes that low sym bolically, a s a prim ary eroticized constituent o f its ow n fantasy life. The result is a m obile, conflictual fusion o f pow er, fear and desire in the construction o f subjectivity: a psychological dependence upon precisely those O thers which are being rigorously op p o sed and excluded at the social level. It is for this reason that w hat is socially peripheral is so frequently sym boli­ cally central (like lon g hair in the 1960s). The low -O ther is despised and denied at the level o f political organ ization and social being whilst it is instrum entally constitutive o f the shared im aginary repertoires o f the dom inant culture. T his is evidenced by the history o f the representation o f “ lo w ” entertainm ent an d the carn ivalesque, to which we now turn.

From C a rn ival to T ran sg ressio n The new historian, the genealogist, will know what to make o f this m asquerade. He will not be too serious to enjoy it; on the contrary, he will push the m asquerade to its limits and prepare the great carnival o f time where

m asks arc constantly reappearing. Genealogy is history in the form o f a concerted carnival. (Foucault 1977: 160-1) In the world o f carnival the awareness of the people’s immortality combined with the reali­ sation that established authority and truth are relative. (Bakhtin 1968: 10) There is now a large and increasing body of w riting which sees carnival not sim ply as a ritual feature o f E u ropean culture but as a mode o f understanding , a positivity, a cultural analytic. H ow is it that a festive ritual now virtually elim inated from m ost o f the pop u lar culture o f Europe h as gained such prom inence as an epistem ological category? Is there a con ­ nection between the fact o f its elim ination as a physical practice an d its self-conscious em er­ gence in the artistic an d academ ic discourses o f our tim e? For both M ichel Foucault in the p assage cited abo ve and for M ikhail Bakhtin in his sem inal study Rabelais and his World , the N ietzscheian study o f history leads to the ideal o f carnival. Everyw here in literary and cultural studies to d ay we see carnival em erg­ ing as a m odel, as an ideal and as an analytic category in a w ay th at, at first sight, seem s puzzling. U ndoubtedly it w as the translation o f M ikhail Bakhtin’ s m onum ental study o f R ab elais and the carnivalesque which initially catalyzed the interest o f W estern sch olars (albeit slow ly - the book w as only tran slated into English in 1968) aroun d the notion o f carn ival, m arking it out a s a site o f special interest for the an alysis o f literature and sym ­ bolic practices. Since the 1970s there has been an increasing num ber o f literary an d historical studies devoted to the topic. In 1978 K rystyna P om orska could w rite with every justification that “ M ikh ail Bakhtin is today one o f the m ost p op u lar, if not the m ost popu lar, figures in the dom ain o f hum anistic stu d ies” (P om orska 1978: 3 7 9 ). M ore recently T on y Bennett averred that Bakhtin’s study o f R ab elais should hold an exem plary place in m aterialist cultural criticism (Bennett 1979: 9 0 -2 ). T h is is surely correct: Rabelais and his World is ostensibly a scholarly study o f R ab e ­ lais’s pop u lar sources in carnivalesque folkculture which sh ow s how indebted R ab elais is

THE POLITICS AND POETICS OF TRANSGRESSION

to the p op u lar, non-literary, " lo w ” folk hum or o f the French R enaissance. H is intention in the study w as self-consciously iconoclastic. “ N o d o g m a, no au th o ritarian ism , no n arrow ­ m inded seriousn ess can co exist with R ab elai­ sian im ages; these im ages are opposed to all th at is finished and polish ed, to all p om posity, to every ready-m ade solution in the sphere of thought and w orld o u tlo o k ” (Bakhtin 1968: 3). N atu rally this reading o f R ab elais has not gone unchallenged by conventionally learned sch olars (Screech 1979: 1 -1 4 , 4 7 9 ; also 1984: 1 1 -1 3 , but in this latter article, “ H om age to R ab e la is” , Screech is m uch closer in spirit to Bakhtin than in the earlier b o ok ). But although Bakhtin is deeply concerned to elucidate the sources o f R ab elais’s w ork, the m ain im p or­ tance o f his study is its bro ad developm ent o f the “ carn ivalesq u e” into a potent, popu list, critical inversion o f all official w ords and hierarchies in a w ay that has im plications far beyond the specific realm o f R ab elais studies. C arn iv al, for Bakhtin, is both a populist utopian vision o f the w orld seen from below an d a festive critique, through the inversion o f hierarchy, o f the “ h igh” culture: As opposed to the official feast, one might say that carnival celebrates temporary liberation from the prevailing truth o f the established order; it marks the suspension o f all hierarchi­ cal rank, privileges, norms and prohibitions. Carnival w as the true feast of rime, rhe feast o f becoming, change and renewal. It was hostile to all that w as immortalized and com ­ plete. (Bakhtin 1968: 109) C arn ival in its w idest, m ost general sense em braced ritual spectacles such as fairs, p o p u lar feasts and w ak es, procession s and com petition s (Burke 1978: 1 7 8 -2 0 4 ), com ic sh ow s, m um m ery and dancing, open-air am usem en t with costum es and m asks, gian ts, d w arfs, m onsters, trained anim als and so forth ; it included com ic verbal com position s (oral and w ritten) such as p arod ies, travesties an d vulgar farce; and it included various genres o f “ Billingsgate” , by which Bakhtin designated cu rses, o ath s, slang, hum our, p op u lar tricks an d jo kes, scatalogical form s, in fact all the “ lo w ” and “ d irty” sorts o f folk hum or. C arn ival is presented by Bakhtin as a w orld

257

o f topsy-turvy, o f heteroglot exuberance, o f ceaseless overrunning and excess where all is m ixed, hybrid, ritually degraded and defiled. If there is a principle to this hotch-potch it resides in rhe spirit o f carn ivalesque laugh ­ ter itself, to which Bakhtin ascribes great im portance: Let us say a few initial w ords about the com plex nature o f carnivalesque laughter. It is, first o f all, a festive laughter. Therefore it is not an individual reaction to some isolated “ com ic” event. Carnival laughter is the laugh­ ter of all the people. Second, it is universal in scope; it is directed at all and everyone, including the carnival’s participants. The entire world is seen in its droll aspect, in its gay relativity. Third, this laughter is ambiva­ lent: it is gay, triumphant, and at the same time mocking, deriding. It asserts and denies, it buries and revives. Such is the laughter of the carnival. (Bakhtin 1968: 11-12) C arn ival laughter, then, has a vu lgar, “ earthy” quality to it. With its oath s and profan ities, its abu sive lan guage and its m ocking w ords it w as profoun dly am bivalent. W hilst it hum iliated and m ortified it also revived an d renew ed. For Bakhtin ritual defilem ents w ent alo ng with reinvigoration such that “ it w as precisely this am bivalent abu se which determ ined the genre o f speech in carnival in tercourse” (Bakhtin 1968: 16). T he “ c o a rse ” and fam iliar speech o f the fair and the m arketplace provided a co m plex vital repertoire o f speech patterns excluded from official discourse which could be used for parody, subversive hum our and inversion. “ Laughter degrades and m ateri­ a lise s” (Bakhtin 1968: 20). Fundam ental to the co rp oreal, collective nature o f carnival laughter is w hat Bakhtin term s “ grotesque realism ” . G rotesque realism uses the m aterial body-flesh conceptualized as corpulent excess — to represent cosm ic, social, topograph ical and linguistic elem ents o f the w orld. T hus already in Bakhtin there is the germ inal notion o f transcodings and displacements effected between the high/low im age o f the physical body and other social dom ain s. G rotesque realism im ages the hum an body a s m ultiple, bulging, over- or under-sized, p rotub eran t and incom plete. The openings and orifices o f this

258

PETER STALLYBRASS AND ALLON WHITE

carnival body are em phasized, not its closure and finish. It is an im age o f im pure corporeal bulk with its orifices (m outh, flared nostrils, anus) yaw ning w ide and its low er regions (belly, legs, feet, buttocks and genitals) given priority over its upper regions (head, “ sp irit” , reason). Bakhtin is self-consciously utopian and lyrical ab o u t carnival and grotesque realism . “ The leading them es o f these im ages o f bodily life are fertility, grow th and a brim m ing-over abundance. M an ifestation s o f this life refer not to the isolated biological individual, not to the private, egoistic, ‘econom ic m an ’ but to the collective ancestral body o f all the peop le” (Bakhtin 1968: 19). T o com plete the im age o f grotesque realism one m ust add that it is alw ays in process, it is alw ays becoming , it is a m obile and hybrid creature, d isp ro p o r­ tionate, exo rb itan t, outgrow ing all lim its, obscenely decentred and off-balance, a figural and sym bolic resource for parodic e x a g g era ­ tion and inversion. All these grotesque q u a li­ ties have a positive force in Bakhtin. It w as only after the R enaissance, accordin g to Bakhtin, th at the principles o f grotesque realism were subjected to a m on ologic reading. Stigm atized as the vulgar practices o f a superstitious and crude populace, the ca r­ nivalesque w as prettified, incorporated into com m ercial or civic display or regarded as a purely negative phenom enon. Bakhtin’s o p ti­ m istic populism is at its m ost insistent (and problem atic) in those p assages where he em phasizes the positivity o f the grotesque bodily element. The grotesque body w as traditionally pre­ sented, Bakhtin argues, not in a private, egotistic form, severed from the other spheres o f life, but as something universal, representing all the people. As such it is opposed to the severance from the m ate­ rial and bodily roots of the world; it makes no pretense to renunciation of the earthy, or independence of the earthy and body. We repeat: the body and bodily life have here a cosmic and at the same time an all-people’s character; this is not the body and its physi­ ology in the modern sense of these words, because it is not individualised. The material

bodily principle is contained not in the bio­ logical individual, not in the bourgeois ego, but in the people, a people who are continu­ ally growing and renewed. This is why all that is bodily becomes grandiose, exaggerated, immeasurable. (Bakhtin 1968: 19) It is difficult to disentangle the generous but willed idealism from the descriptively accurate in p assage s like these. Bakhtin constantly shifts between prescriptive and descriptive categories in his w ork. In this p assage the cosm ic p o p u ­ lism , which seem s to us rather w ishful and finally unusable as an analytic to ol, assorts with an acute perception ab o u t the historically variable nature o f the body-im age. . . . In Bakhtin’s schem a grotesque realism in pre-capitalist Europe fulfilled three functions at once: it provided an im age-ideal o f and for popu lar com m unity as an heterogeneous and boundless totality; it provided an im aginary repertoire o f festive and com ic elements which stood over again st the serious and oppressive languages o f the official culture; and it p ro ­ vided a thoroughly m aterialist m etaphysics w hereby the grotesque “ bodied fo rth ” the co sm os, the social form ation and language itself. Even linguistic rules are played up by w hat Bakhtin calls a gram m atica jocosa w hereby gram m atical order is transgressed to reveal erotic and obscene or merely m aterially satisfying counter-m eaning. i...r W hilst alm ost every reader o f Bakhtin adm ires his com prehensive and engaged gen ­ erosity, his com bin ation o f festive populism and deep learning, and whilst few w ould deny the im m ediate app eal and the vitality o f the notion o f carn ival, various w riters have been sceptical o f Bakhtin’s overall project. Terry Eagleton thinks that the w eakness o f Bakhtin’s positive em brace o f carnival is transparent: Indeed carnival is so vivaciously celebrated that the necessary political criticism is almost too obvious to make. Carnival, after all, is a licensed affair in every sense, a permissible rupture o f hegemony, a contained popular blow-off as disturbing and relatively inef­ fectual as a revolutionary work o f art. As Shakespeare’s Olivia remarks, there is no

THE POLITICS AND POETICS OF TRANSGRESSION

slander in an allowed fool. (Eagleton 1981: 148} M o st politically thoughtful com m en tators w onder, like Eagleton, whether the “ licensed release” o f carnival is not sim ply a form o f social control o f the low by the high and there­ fore serves the interests o f that very official culture which it apparen tly opposes. The classic form ulation o f this is in M a x Gluckm an ’s now som ew hat dated O rder and Rebel­ lion in Tribal Africa (1 9 6 3 ) and Custom and Conflict (1 9 6 5 ), in which he asserted that while these “ rites o f reversal obviously include a protest again st the established order . . . they are intended to preserve and strengthen the established o rd er” (G luckm an 196 5 : 109). R oger Sales am plifies both on this process o f containm ent and its am bivalence: There were two reasons why the fizzy, dizzy carnival spirit did not necessarily undermine authority. First of all, it w as licensed or sanc­ tioned by the authorities themselves. They removed the stopper to stop the bottle being smashed altogether. The release of emotions and grievances made them easier to police in the long term. Second, although the world might appear to be turned upside down during the carnival season, the fact that Kings and Queens were chosen and crowned actually reaffirmed the status quo. Carnival was, however, Janus-faced. Falstaff is both the merry old mimic of Eastcheap and the old corruptible who tries to undermine the author­ ity, or rule, o f the Lord Chief Justice. The carnival spirit, in early-nineteenth century England as well as in sixteenth century France, could therefore be a vehicle for social protest and the method for disciplining that protest. (Sales 1983: 169) As G eorges Balandier puts it succinctly in Political Anthropology: “ The suprem e ruse o f pow er is to allow itself to be contested ritually in order to con solid ate itself m ore effectively.” It actually m akes little sense to fight out the issue o f w hether or not carn ivals are intrinsi­ cally radical or conservative, for to d o so au to ­ m atically involves rhe false essentializing o f carn ivalesque tran sgression (W hite 1982: 60). The m ost that can be said in the ab stract is

259

that for long periods carnival m ay be a stable and cyclical ritual with no noticeable politi­ cally tran sform ative effects but that, given the presence o f sharpened political an tagon ism , it m ay often act as catalyst and site o f actual and

symbolic struggle. It is in fact striking how frequently violent social clashes apparently “ co in cid ed ” with carnival. Le R oy Ladurie’s Carnival in Romans (1 9 8 1 ) has popularized one such incident when the 1 580 festival at R om an s in eastern France w as turned into arm ed conflict and m assacre. O ther social historians have docum ented sim ilar occurrences (D avis 1975; Burke 1978; T h o m p son 1972). H ow ever to call it a “ coin ­ cidence” o f social revolt and carnival is deeply m isleading, for as Peter Burke has pointed out, it w as only in the late eighteenth and early nineteenth centuries ~ and then only in certain areas - that one can reasonably talk o f popu lar politics dissociated from the carn ivalesque at all. Jo h n Brewer has described English politics in the eighteenth century as “ essentially a calendricai m arket” , by which he designates a deliberate com m ingling o f holiday and political events (in this case organized by the H an overian s for conservative m otives): Far too little attention had [sic] been paid to the emergence during the eighteenth century o f a Hanoverian political calendar, designed to inculcate loyal values in the populace, and to emphasize and encourage the growth of a national political consensus. Nearly every English market town celebrated the dates which were considered the important political landm arks o f the nation. They can he found in m ost alm anacs of the period, barely distin­ guishable from the time-honoured dates of M ay Day, Plough M onday, Twelfth Night, Shrove Tuesday and the like . . . In the early eighteenth century, these dates, together with the occasion o f the Pretender’s birthday, were occasions o f conflict. The year o f the Jacobite Rebellion, 1715, was especially contentious, with Hanoverian M ug House clubs fighting it out in the streets with Jacobite apprentices and artisans. On October 30, frequenters of a Jacobite alehouse on Ludgate Hill were beaten up by members of the Loyal Society who were celebrating the birthday o f the Prince of W ales, the future George II. A Jacobite attempt

260

PETER STALLYBRASS AND ALLON WHITE

to burn William III in effigy on November 4 was thwarted by the same Whig clubmen who the next day tried to cremate effigies o f rhe Pretender and his supporters. On 17 N ovem ­ ber further clashes ensued and two Jacobites were shot dead. (Brewer et al. 1983: 247) Again this should act as a w arning again st the current tendency to essenrialize carnival and politics. O n the one hand carnival w as a spe­ cific calen drical ritual: carnival prop er, for instance, occurred aroun d February each year, ineluctably follow ed by Lenten fasting and abstinence bound tightly to law s, structures and institutions which had briefly been denied during its reign. O n the other hand carnival also refers to a m obile set o f sym bolic p rac­ tices, im ages and discourses which were em ployed th roughout social revolts and co n ­ flicts before the nineteenth century. . . . In recent social histories o f England there has been a con siderable d ebate over the inter­ relationsh ip between p op u lar culture and class conflict (Y eo and Y eo 1981; Bushaw ay 198 2 ; W alvin 1 9 7 8 ; C unningham 1 9 8 0 ; T h om pson 1972; M alcolm so n 1973; Stedm an Jo n e s 1983). M o st o f these studies unearth evidence o f a long battle (with occasion al truces) w aged by the State, ecclesiastical and bo u rgeo is authorities again st p op u lar custom . It is a batrle that goes back well beyond the R en ais­ sance but which, from the R en aissan ce on, produced local festivities as sites o f resistance to the extension o f pow er by the propertied and the State. Bushaw ay rem arks: “ C ustom and cerem ony becam e a battleground in the struggle between the labourin g p oo r an d the increasingly w ealthy landow ners and p rop ri­ etors over the defence o f pop u lar rights and the protection o f a norm ative view o f the struc­ ture o f the com m unity held by the labourin g p o o r” (B ush aw ay 1982: 2 1 -2 ). T his seem s an altogether m ore accu rate w ay o f conceiving the relationship. C arn iv als, fairs, pop u lar gam es and festivals were very swiftly “ politi­ cized ” by the very attem pts m ade on the part o f local authorities to elim inate them. The dialectic o f an tagon ism frequently turned rituals into resistance at the m om ent o f inter­ vention by the higher pow ers, even when no overt op p o sitio n al element had been present

b e fo r e .. . . H ere we are only underscoring the banal but often ignored truth that the politics o f carnival cannot be resolved outside o f a close historical exam in ation o f p articular con ­ junctures: there is no a priori revolutionary vector to carnival and transgression. In his research on the carnivalesque Bakhtin had substantially anticipated by som e thirty years m ain lines o f developm ent in sym ­ bolic an th ropology. In his exploration o f the relational nature o f festivity, its structural inversion o f, and am bivalent dependence upon, “ official cu lture” , Bakhtin set out a m odel o f culture in which a high/low binarism had a fundam ental place. Bakhtin’s use o f carnival centres the concept upon its “ dou ble­ ness . . . there is no unofficial expression w ithout a prior official one or its possibility. H ence, in Bakhtin’ s analysis o f carn ival, the official and unofficial are locked togeth er” (W ilson 1983: 3 20). Sym bolic polarities o f high an d low', official and popu lar, grotesque and classical are m utually constructed and deform ed in carnival. T w o o f rhe best general syn opses o f Bakhtin’s w ork correctly perceive this to be the m ost significant aspect o f Rabelais and bis World. Ivanov (1976) links Bakhtin’s discovery o f the im portance o f binary op p o sitio n s with the w ork o f Levi-Strauss: the books by Bakhtin and Levi-Strauss have much in common in their treatment of the functioning o f oppositions in the ritual or the carnival which can be traced back historically to ritual performance. For Levi-Strauss the chief purpose o f the ritual and the myth is the discovery o f an intermediate link between the members o f a binary opposition: a process known as mediation. The structural analysis o f the ambivalence inherent in the “ m arket­ place w ord” and its corresponding imagery led Bakhtin to the conclusion (made indepen­ dently from and prior to structural myth­ ology) that the “ carnival image strives to embrace and unite in itself both terminal points of the process of becoming or both members o f the antithesis: birth-death, youthage, top-bottom , face-lower bodily stratum, praise-abuse” [Bakhtin 1968: 238]. From this standpoint, Bakhtin scrutinized various forms o f inverted relations between top and bottom

THE POLITICS AND POETICS OF TRANSGRESSION

“ a reversal o f the hierarchy o f top and bottom ” [Bakhtin 1968: 81] which takes place during carnival. {Ivanov 1976: 35) The convergence o f Bakhtin’s thinking and th at o f current sym bolic an th rop ology is highly significant. W here Ivanov poin ts to the kinship Bakhtin shares with Levi-Strauss and Edm und Leach (particularly L each ’s essay on carn ival, “ T im e and false n o ses” , 1961), M a sa o Y am aguchi suggests that Bakhtin’s w ork signifi­ cantly p arallels that o f V ictor T urner, Barbara B abcock and M ary D o u g las in their shared interest in cultural n egation s and sym bolic inversions (Y am aguchi 1983). We m ay note, for instance, the sim ilarity o f Bakhtin’s concept o f carnivalesque high/low inversion to the co n ­ cepts developed in The Reversible World , a collection o f essays on an th rop ology and lit­ erature edited by B arb ara B abcock . Although apparen tly unaw are o f Bakhtin’s study she assem b les a range o f w riting on “ sym bolic inversion and cultural n egation ” which puts carnival into a much w ider perspective. She w rites: “ ‘Sym bolic inversion’ m ay be broadly defined as any act o f expressive behaviour which inverts, co n tradicts, ab ro gate s, or in som e fashion presents an alternative to co m ­ m only held cultural cod es, values and norm s be they linguistic, literary or artistic, religious, social and p olitical” (B abcock 1978: 14). T h is is w hat we refer to in this book as “ tran sgressio n ” (though there is an oth er, m ore co m p lex use o f the term which arises in con ­ nection with extrem ist practices o f m odern art and ph ilosoph y; these designate not just the infraction o f binary structures, but m ovem ent into an absolutely negative space

beyond the structure o f significance itself). For the m om ent it is enough to suggest th at, in our view , the current w idespread ad op tion o f the idea o f carnival a s an analytic category can only be fruitful if it is displaced into the b ro ad er concept o f sym bolic inversion and tran sgression . T h is is not to deny the usefulness o f the carnivalesque a s a sort o f “ m odellin g” , at once utopian and counter-hegem onic, whereby it is view ed, in R oberto d a M a rta ’s w ords, as a privileged locus o f inversion. In his attem pt to go beyond Bakhtin’s n ostalgic and over-

261

optim istic view o f carn ival, M atta ackn ow l­ edges the degree to which festivity is licensed release, but he also praises its deep m odelling o f a different, pleasu rable and com m unal ideal “ o f the p eo p le” , even if that ideal cannot im m ediately be acted upon. V ictor T urn er has sim ilarly argued with respect to role reversal th at carnival is “ a m om ent when those being m oved in accordance to a cultural script were liberated from norm ative dem an ds, where they w ere . . . betw ixt and between successive lodge­ m ents in jural political sy stem s” . C arnival in this view has been defended a s having a persistent demystifying potential (Jones 1983; A rthur 1982; Stam m 1 9 8 2 ; D avis 1975). Even T erry Eagleton w ants to salvage Bakhtin’s ca r­ n ivalesque by seeing it as a utopian m odelling yoked to a glim pse through the ideological constructs o f dom inance, a “ kind o f fiction” , a “ tem porary retextualizing o f the social for­ m ation that exp o ses its ‘fictive’ fo u n d atio n s” (Eagleton 1981: 149). In this perspective the carn ivalesque becom es a resource o f action s, im ages and roles which m ay be invoked both to m odel and legitim ate desire and to “ degrade all that is spiritual and a b stra c t” . “ T he cheerful vulgarity o f the p ow ­ erless is used as a w eapon again st the pretence and hypocrisy o f the pow erful” (Stam m 1982: 4 7 ). In a m ost engaging description o f this utopian/critical role o f carn ival Stam m continues: On the positive side, carnival suggests the joyful affirmation o f becoming. It is ecstatic collectivity, the superseding o f the individuat­ ing principle in what Nietzsche called “ the glowing life of Dionysian revellers” . . . On the negative, critical side, the carnivalesque suggests a demystificatory instrument for everything in the social formation which renders such collectivity difficult of access: class hierarchy, political m anipulation, sexual repression, dogm atism and paranoia. Carni­ val in this sense implies an attitude of creative disrespect, a radical opposition to the illegiti­ mately powerful, to the m orose and monological. (Stamm 1982: 55) R efreshingly icon oclastic, this nevertheless resolves none o f the problem s raised so far

262

PETER STALLYBRASS AND ALLON WHITE

concerning the politics o f carnival: its n ostal­ gia; its uncritical populism (carnival often vio­ lently ab u ses and dem onizes weaker , not stronger, social grou p s - w om en, ethnic and religious m inorities, those w ho “ d o n ’t belon g” - in a process o f displaced abjection ); its failure to d o aw ay with the official dom in an t culture, its licensed com plicity. In fact those w riters and critics w ho rem ain purely within the celebratory term s o f B akh tin ’s form ulation are unable to resolve these key dilem m as. It is only by com pletely shifting the groun ds o f the debate, by tran sform in g the “ prob lem atic” o f carn ival, that these issues can be solved. It is precisely such an interven­ tion in the current surge o f Bakhtin-inspired studies which we have attem pted in this b o o k .. . . We have chosen therefore to con ­ sider carnival a s one instance o f a generalized econom y o f tran sgression and o f the recoding o f high/low relations across the w hole social structure. The sym bolic categories o f grotesque realism which Bakhtin located can be rediscov­ ered as a governing dynam ic o f the body, the household, the city, the nation-state - indeed a vast range o f interconnected d om ain s. . . . By track in g the “ grotesque b o d y ” and rhe “ low -O th er” through different sym bolic dom ain s o f bourgeois society since the R en ais­

sance we can attain an unusual perspective upon its inner dyn am ics, the inner com plicity o f disgust and desire which fuels its crises o f value. For the classificatory body o f a culture is alw ays dou ble, alw ays structured in relation to its negation, its inverse. “ All sym bolic inver­ sion s define a culture’s lineam ents at the sam e time a s they question the usefulness and the absoluten ess o f its orderin g” (Babcock 1978: 29). Indeed by attending to the low and the m arginal we vindicate, on the terrain o f E u ro­ pean literary and cultural history, the m ore general an th rop ological assertion that the process o f sym bolic inversion, “ far from being a residual category o f experience, is its very opposite. W hat is socially peripheral is often sym bolically central, and if we ignore or m ini­ mize inversion and other form s o f cultural negation, we often fail to understand the dynam ics o f sym bolic processes generally” (Babcock 1978: 32). This is a scrupulously accurate and in dis­ pensable form ulation. The carnival, the circus, the gypsy, the lum penproletariat, play a sym ­ bolic role in bourgeois culture ou t o f all p ro ­ portion to their actu al social im portance. The dom inant features o f the psycho-sym bolic dom ain can n ot be m apped one-to-one on to the social form ation.

BIBLIOGRAPHY

Brewer, J ., M cK en drick, N . and Plum b, J . H. (1983) The Birth o f a Consumer Society:

\1 M B = M ikhail Mikhailovich Bakhtin: Hts Circle , His Influence. Papers presented at the International C olloquium , Q ueen ’s University, K ingston, O n tario, 7 -9 O ctob er 1983. Arthur, K. (1 9 8 2 ) “ Bakhtin, K risteva and carn ival” , unpublished d issertation, M elbourne. Babcock, B. (1 9 7 8 ) The Reversible World: Symbolic Inversion in Art and Society , Ithaca, Cornell University Press. Bakhtin, M . M . (1 9 6 8 ) Rabelais and his World (tr. H . Isw olsky), C am brid ge, M a ss., M IT Press. Bennett, T . (1 9 7 9 ) Formalism and M arxism , Lon don , M ethuen.

The Commercialization o f Eighteenthcentury England , L on don , H utchinson. Burke, P. (1 9 7 8 ) Popular Culture in Early Modern E.urope , L on don , T em ple Smith. Bushaw ay, B. (1982) By Rite: Custom , Cere­ mony and Community in England 170 0 1880 , Studies in Popular C ulture, L on don , Junction B ooks. C unningham , H . (1980) Leisure in the Indus­ trial Revolution , Lon don , C'room Helm. C u rtiu s, E. R. (1 9 5 3 ) European Literature and the Latin M iddle Ages , London , R outledge 6c K egan Paul, 1979. D avis, N . Z . (1 9 7 5 ) Society an d Culture in F]arly Modern France , Stan ford, Stanford University Press.

THE POLITICS AND POETICS OF TRANSGRESSION

Eagleton, T . (1 9 8 1 ) Walter Benjamin: Tow ards a Revolutionary Criticism , Lon d on , Verso. Foucault, M . (1 9 7 7 ) iMnguage/CountermemoryfPractice (ed. D . F. Bouchard, tr. D. F. Bouchard and S. Sim on), Ithaca, Cornell University Press. G luckm an , M . (1 9 6 3 ) O rder and Rebellion

in Tribal Africa: Collected Essays with an A utobiographical Introduction , London, Cohen. G luckm an , M . (1 9 6 5 ) Custom and Conflict in A frica , O x fo rd , Blackw ell. Ivanov, V. V. (1 9 7 6 ) “ T he significance o f Bakhtin’s ideas on sign, utterance and d ia ­ logue for m odern sem io tics” , in Papers on Poetics and Semiotics 4 , Tel Aviv, The Israeli Institute for Poetics and Sem iotics, Tel Aviv University. Ja c k , I. (1942) Augustan Satire: Intention and Idiom in English Poetry 1 6 6 0 -1 75 0 , O x fo rd , C larendon Press. Jo n e s, A. R. (1 9 8 3 ) “ Inside the outsider: N ash e ’s ‘U nfortunate traveller’ and Bakhtin’s polyphonic novel” , English Literary History , 5 0 ,' I, 6 1 -8 2 . Le Roy Ladurie, E. (1 9 8 1 ) Carnival in Romans (tr. M . Feeney), Flarm ondsw orth, Penguin. Leach , F. (1 9 6 1 ) “ T im e and false n o ses” , in Leach, E. (ed.) Rethinking Anthropology , M on ograph /Social A nthropology 2 2 , L o n ­ don , Athlone Press, 1 3 2 -6 . M alcolm so n , R. W. (1 9 7 3 ) Popular Recre­ ations in English Society 1 7 0 0 -1 8 5 0 , C a m ­ bridge University Press. P om o rsk a, K. (1 9 7 8 ) “ M ikh ail Bakhtin and

263

his verbal universe” , Poetics an d Theory o f Literature (PTL), 3, 3 7 9 -8 6 . Said, E. (1979) O rientalism , N ew Y ork, V intage Books. Said, E. (1 9 8 4 ) The World , the Text and the Critic, London , Faber 6c Faber. Sales, R. (1 9 8 3 ) English Literature in History 1 780-1830: Pastoral and Politics , London, Flutchinson. Screech, M . A. (1979) R abelais , London, Duckw orth. Screech, M . A. (1 9 8 4 ) “ H o m age to R ab e la is” , London Review o f Books, 6, 17. Stam m , R. (1982) “ On the carn ivalesq u e” , Wedge, 1 ,4 7 - 5 5 . Stedm an Jo n e s, G . (1 9 8 3 ) Languages o f Class, C am bridge, C am bridge University Press. T h o m p son , E. P. (1 9 7 2 ) “ ‘ R ough m usic’ : le charivari an g lais” , Annales E SC , 2 7 .2 , 2 8 5 -3 1 2 . W alvin, J. (1978) Leisure an d Society 18301950 , Lon don , Longm an. W hite, A. (1982) “ Pigs and pierrots: politics o f transgression in m odern fiction” , Raritan , II, 2 (Fall), 5 1 -7 0 . N ew Brunsw ick, N ew Jersey. W ilson, R. (1983) “ C arn ival and p lay ” , M M B (C oll.), K ingston, O n tario , Q ueen ’s Univer­ sity, 3 1 8 -2 1 . Y am agu ch i, M . (1983) “ Bakhtin and sym bolic an th ro p o lo g y ” , M M B (C oll.), 3 2 3 -3 9 . K ingston, O n tario , Q ueen’s University. Y eo, E. and Y eo, S. (eds) (1981) Popular Culture and Class Conflict , Brighton, H arvester.

Conceptualizing the Cosmos

Introd uction By contrast to m ost o f the previous essays, which have interpreted single parts, segments, o r dimensions o f cultural worlds, these essays on cosm ology attempt to portray such worlds as wholes. They thus move to levels o f greater abstraction, offering larger pictures in terms o f which the specific details o f Polynesian carving or tattooing, or o f Amerindian myth or sham anism , make sense. Such analytic portraits are surely one o f the goals o f anthropology and usually only realized in full-length m onographs. It will be seen that both Gell and Viveiros de C a stro raise co m plex issues and draw , to a degree, on western philosophical literature in order to more clearly locate and refine their ethnographic descriptions. One o f the values o f such broad cosm ological arguments is that they radically throw into question the “ native” (western) cosm ology from which anthropologists normally begin their analyses and which forms the common-sense grou nd for most readers. M o re force­ fully than any other approach , they relativize o ur assum ptions and m ake us see ourselves, i.e., “ our culture,” as simply “ another other a m on g others.” T o this end, I would have liked to include Sahlins (1996) on the cosm ology o f the West. C o sm ology is som etim es phrased in terms o f the conjunction o f space and time. Eliade (1959) is a fam ous if som ew hat idealized picture o f different constructions o f co sm o s and history, while Daniel (1996) provides a subtle account o f different conceptions o f history a m o n g Sinhalese and T am il speakers (Buddhists and Hindus, respectively) in contem porary Sri Lanka. Evans-Pritchard (1940) and Leach (1961) are classic loci on time (cf. Gell 1992), while B asso (1996) offers a profound medi­ tation on place. As B o u rd ie ifs concept o f habitus (1977) and Bakhtin’s concept o f ch ron otope (1981) suggest, space and time are inextricably connected. Lam bek and Walsh (1999) and Lam bek (2002) offer ethnographic exam ples. Questions o f c o s­ mology are often tied to accounts o f transform ations to modernity, itself implicitly a cosm ology although increasingly come to be seen as either unraveling or intensify­ ing and contracting time and space (Appadurai 1 9 9 6 , Jam e so n 1991).

266

INTRODUCTION

The beauty o f unitary cosm ologies and the pow er o f holistic analysis (R ap p ap o rt 1999) notwithstanding, cosm ological accounts should be supplemented by those that emphasize local diversity, whether through the social organization o f knowledge (James 1988, Lam bek 1993a), “ p o p u la r” religion contrasted to orth o do xy (Badone, ed. 1990, Gellner 2 0 0 1 ), religious debate within societies (Fischer 1980, Fischer and Abedi 1990), gender divisions (Boddy 1989, Gillison 1993), conversion (Hefner, ed. 1993), or pluralism and contestation (M asquelier 2001 ).

22

Closure and Multiplication: An Essay on Polynesian Cosmology and Ritual Alfred Gell

Alfred Gell (1945-97) was an outstand­ ing iconoclastic thinker w ho taught at the London School of Economics and wrote on tim e and art, among many other subjects. Gell's essay is a powerful account of Polynesian cosmology that begins w ith the startling assertion that the Polynesian w orld is one of imma­ nence rather than transcendence and that such im m anence brings problems in its w ake. Rather than the carefree life that Polynesians are often portrayed as having, Gell paints a picture of anxiety in which you had to "w atch your back," not so much from personal enem ies as from the pervasive effects of the sacred. Gell's aesthetic sensibility is evident in the acute analysis of Polynesian carving

The m ain features o f Polynesian cosm ological beliefs show a high degree o f consistency and have becom e reason ably well know n, through the efforts o f a host o f nineteenth-century w riters, m ostly m ission aries or governm ent

and tattooing, supplem ented by his own drawings and those of earlier w estern visitors to Polynesia (only some of which are reproduced here). See also his longer study W rapping in Images (1993). The anthropology of art is a rich field overlapping that of religion. Gell's analysis is based entirely on sec­ ondary sources; his own fieldw ork was carried out in Melanesia (1975) and tribal India. He is the author of a par­ ticularly fine essay on taboo (1979). O ther essays on Oceanian cosmologies are to be found in de Coppet and Iteanu (1995); see also Strathern (1988) and Bam ford ed. (1998), as well as many excellent ethnographies.

officials, and subsequent re-w orking by Polynesia specialists. T he C entral-Eastern Polynesian culture area, com prising the Society Islan ds (Tahiti), the M arq u e sa s, N ew Z ealan d, and H aw aii to nam e only the larger su b ­

Erom Alfred Gell, “ Closure and Multiplication: An Essay on Polynesian Cosm ology and R itual,” in Daniel de Coppet and Andre Iteanu, eds, Cosm os and Society in Oceania (Oxford: Berg Publishers, 1995), pp. 21 -5 6 . Abridged.

268

ALFRED GELL

regional system s, all m aintained broadly sim ilar system s o f ritual and belief, which were founded on the co sm ological schem e in which the creation o f the co sm o s cam e ab o u t through conflict and sep aratio n (H an son 1 9 8 2 ). The co sm o s w as originally one but this affo rd ed no scope o f action to the creator god (T a'aro ay Tane, etc.) w hose initial act w as to bring about the sep aratio n o f sky and earth, night and day, thus creatin g the “ bi-cam eral w o rld ” o f the Polynesians which w as divided into the p o , the other-w orld, the w orld o f night (po), darkn ess, the original g o d s, the dead , etc. and the ao, the w orld o f light, day (ao) life, hum an activity, and so forth. T h is original separation w as su b ­ sequently elabo rated differently in different places, and m ight becom e form idably co m p li­ cated, as is suggested by Figure 1, which show s a diagram o f the co sm o s created by a T uam otuan inform ant for an inquisitive m issionary in the late nineteenth century. The co sm o s a s a

fi f e

.

V-a -y a j* *

Figure 1

w hole takes the form o f a nested arrangem ent o f “ shells” ('ap u ), each associated with a p ar­ ticular cosm ogo n ic episode o f separatio n ; and path w ays or ladders are indicated by m eans o f which intercourse is possible between different co sm ological levels. Polynesian thought ab o u t the universe dif­ fered from Judaeo-C'hristian “ creatio n ist” thought in that it w as predicated, not on the creation o f the universe ex nihilo , by Ciod, but on the initial existence o f everything in an allem bracing plenum or tightly-bound contin ­ uum. The creative epoch occurred a s a process o f “ differen tiation ” within this pre-existing plenum , undertaken by a Clod w ho m ade “ c u ts” . The T ah itian creation chant, part o f which 1 am ab o u t to cite, is really w rongly so called, because nothing w hatsoever is “ created ” in it. Instead, w hat the Ciod does is to articu­ late, or differentiate, the w orld into its distinct com pon ents and qualities, but the substan ce o f

& *

* - i 'f

* f r tfM .

The Tuamotuati conception o f the Cosm os by Paiore, c. 1820

,o.

AN ESSAY ON POLYNESIAN COSM OLO GY AND RITUAL

the new ly-articulated co sm o s rem ains w hat it alw ays w as, nothing other than the Ciod him self. The name borne by T a'aro a , the T a h i­ tian creator g o d , actually m eans “ the sever­ e r” . T a'aroa is called thus because he m ade the initial severance between the po and the do, an d he created the various natural phenom ena by rearranging his ow n body: So he overturned his shell and raised it up to form the sky . . . and he slipped out o f another shell which covered him which he took for rock and sand. But his anger was not yet appeased, so he took his spine for a mountain range, his ribs for mountain slopes, his vitals for the broad fleeting clouds, his flesh for the fatness o f the earth, his arm s and legs for the strength o f the earth, his finger-nails and toe­ nails for scales and shells for the fishes, his feathers for trees, shrubs and creepers, to clothe the earth, his intestines for lobsters, shrimps and eels for the rivers and seas, and the blood o f T a’aroa got heated and drifted away for redness for the sky and the rain­ bows . . . Ta'aroa had crusts, that is, shells, so everything has a shell. The sky is a shell, that is, endless space in which the God placed the sun, the m oon, the sporades [the scattered or individual stars], and the constellations o f the gods. The earth is a shell to the stones, the water, the plants which spring from it. M an’s shell is woman, because it is by her that he comes into the world, and wom an’s shell is woman, because she is born of woman. One cannot enumerate the shells o f all the things the world produces. {Henry 1928: 339-40) T h is projection o f the body o f the Ciod into the phenom enal w orld, though developed in an exceedingly poetic w ay in the texts assem bled by T euira H enry, w as not in itself an unortho­ d o x conception o f cosm ogen esis. But it had profoun d consequences for the subsequent developm ent and elaboration o f indigenous religious thought. Because the po and the ao were originally em ergent p arts o f a continuum , their separation had, for the Polynesians, only a provision al character, and they continually threatened to m erge together again . Indeed the very instrum ents that kept them ap art, kept them together and in com m unication . T h is w as

269

brough t ou t m ost perspicuously in relation to the four p ro p s (/o ’o) which T a ’aro a positioned in order to keep the sky up an d open a space (atea) for the a o , the w orld o f light and o f hum anity. The p rop s that separated the po and the a o also , by definition, joined them together, and thus becam e the exem plary prototypes o f sacred objects, endow ed with the pow er to com m unicate the divine presence. Society Islands politics revolved aroun d rival claim s ro con trol certain ultra-sacred w ooden billets, called to'o, which were coun terparts o f the “ p ro p s” o f the co sm os as a w hole. But, being so sacred, these w ooden billets had them selves to be separated from the w orld, w hose contin­ ued existence/differentiation they guaran teed; so they were confined to an elaborate ark, and were m assively bound aroun d with sennit co rd ag e , and other w rap pin gs, and were only uncovered on the m ost sacred and propitious o ccasio n s, requiring m any hum an sacrifices to allay the danger rhey presented to m ortals, how ever sanctified by birth and ritual p rep ara­ tions (Henry 1928; O liver 1974). T he cosm ogon ic schem e o f differentiation rather than creation ex nihilo an d o f im m anent rather than transcendent divinity w as, I think, the source o f certain on tological anxieties that played an enorm ous part in Polynesian life. For us, the im m anent deus sive natura of Spin oza represents an optim istic rather than a pessim istic deism , a blessed relief from the angry and punishing Alm ighty G o d o f trad i­ tional C h ristianity, set a p art from H is creation and ju dging it harshly. But that is because the idea o f an im m anent Ciod w as never really n aturalized in C hristian Eu rope, how ever much eighteenth-century intellectuals may have hankered after one. In Polynesia the situ­ ation w as the precise op p o site; the im m anence o f the G o d s w as rhe source o f continuous anxiety (the proverbial hedonism o f the South Sea Islanders w as founded on a sense o f acute and abidin g hysteria) and the rapidity and enthusiasm with which the Polynesians acco m ­ plished their conversion to Christianity stem m ed from their untold relief upon d isco v­ ering that CJod w as, after all, transcendent, nor part o f this w orld. In this essay, the theme I w ant to ad d ress is Polynesian anxiety ab o u t im m anence, ab o u t

270

ALFRED GELL

differentiation, and som e o f the w ays in which this fundam ental anxiety shaped the patterns o f their cultural life. T h at is to say, I w ant to explore som e o f the w ays in which the under­ lying conception o f the creation o f the w orld through a process o f differentiation (splitting, separation ) gave rise to countervailing pres­ sure to w ard s fusion, absorp tion , co llap se, etc., which had to be continually held in check through ritual action. In particular ! w ant to identify the sym bolic strategies through which social individuals sough t to keep im m anence at bay, while, at the sam e tim e, ackn ow led ging the fundam ental fact that im m anence pervaded the w orld and constituted it. T he essential strategy o f counter-im m anence (averting the danger o f being ab sorb ed , de-differentiated, etc.) w as the recapitulation , in one form or another, o f the process o f separation on which both the co sm o s in general, and the being o f the in dividual, were founded. In other w ords, the co sm ogo n ic activities o iT a 'a ro a the severer had to be recapitulated at the level o f the indi­ vidual subject; not ou t o f a vain glorious desire to em ulate the G o d , but certainly for the sam e reason s that the creative go d s engaged in the original splitting, i.e. in order to open a space, to establish a difference. And just as the origi­ nal co sm ogo n ic act w as the opening up o f a dom ain , within the plenum , which w as rela­ tively non-sacred (the j o , the w orld o f light) so the m eans o f ritually securing the subject w as the opening o f a space between the subject and the sacred dom ain whence he originated. M o st im portan t Polynesian ritual operated in precisely the inverse sense to C h ristian co m ­ m union, i.e. the intenrion w as to cause the divinity to leave (som e part of) the w orld, rather than to induce the divinity to enter (som e part of) it. There were no rituals through which person s (e.g. kings) were consecrated, because they were already all-too-sacred, but action in the w orld w as im possible for persons o f high intrinsic sanctity, because they did not truly belong to it, belonging, instead, to the p o . T h u s, for instance, the non-executive sacred ruler o f T o n g a, the T u ’i T o n g a, w as not m ourned when he died, and his funeral m ound w as covered with excrem ent rather than offer­ ings, because he had never lived (Bott 1982). Instead o f being sanctified, Polynesians o f high

intrinsic (i.e. genealogically-based) rank had to be elaborately de-sanctified. T h u s, the m ain passage-rites o f the Society Islands were the am o'a or “ head-releasing” rituals through which children, w h o were considered to be “ little g o d s ” (and highly tapu) were m ade “ secu lar” by being offered blood draw n from the co n gregation, which replaced their sacred substance and freed them for ordinary w orldly social interactions. We can begin to consider the consequences o f the co sm ology o f im m anence for personal ontology by dw elling for a m om ent on the general Polynesian w ord for “ g o d ” , atu a . This w ord is based on the m orphem e tua, which m eans ‘‘ b ack ” , or the far, invisible, side o f any object. The particularly sacred parts o f a hum an being were the back (the spine) and the head - i.e. precisely those parts o f the body that cannot norm ally be seen by their p o s­ sessor. The atua (spiritual element) o f the person w as the tua (back) o f the person. Tua a lso refers to elders and ancestors, the “ b a ck ” portions o f time, and these were also sacred. Imm anent divinity w as not a contingent pres­ ence, but a categorical, dim ensional, feature o f objects, spaces, tim es. O rdinary hum an inter­ course w as front to front, expressed in w estern Polynesia (and Fiji) in the fundam ental social m orality o f “ facing one an oth er” (fa’agaga, cf. Shore 1982), while it w as alw ays strictly fo r­ bidden to m ove a b o u t behind som e im portant person, invisibly to them. The back w as the “ individual” as o p p o sed to the cosm ological p o , the inaccessible, threatening, but ever­ present unseen. But it w as possible a lso to intervene so a s to protect this aspect o f the person. I think one can identify tw o basic strategies through which the integrity o f the person could be m aintained again st cosm ological c o l­ lapse, to which I have attached the labels o f “ closu re” and “ m ultiplication” . “ C lo su re ” is the provision o f extra reinforcem ent, h arden­ ing the target o f spiritual danger, while “ m ulti­ plication” is the strategy o f reduplicating the person in m yriad form s. We can see both stra t­ egies at w ork in the cosm ological m yths referred to above: thus T a'aroa m ultiplies him self by opening up the ao, assum in g his m yriad im m anent form s; but a lso he closes it

AN ESSAY ON POLYNESIAN COSM OLO GY AND RITUAL

o ff with the “ shells” (secured by rhe “ p ro p s” ) so thar it cannot fall in again . In fact, the entire co sm os consisted o f such “ shells” , as Figure 1 show s. But, as the chant a lso specifies, people were a lso “ shells” , and so were their kingrou p s. (In the eighteenth century, ’ap u , “ shell” w as the stan dard T ah itian expression for the bilateral kindred, the basic unit o f social organ ization , cf. M ao ri: hapu.) But rather than continue with these general o b ser­ vations, let me turn to consider som e m ore specific im agery o f closure and m ultiplication in the con text o f Polynesian art traditions. Polynesian carving is particularly rich in Jan ifo rm im ages (Figure 2), which, though they occur elsewhere (particularly in India), nowhere assum e the im portance they have in this part o f the globe. Figure 2, which we m ay take as a typical instance, is one o f a sm all num ber o f im ages o f Siam ese twin goddesses carved from w hales' teeth, all o f which were actually collected in Fiji, where they were regarded as the con sorts o f N den g’ei , the main god ; but they are known to have been carved in F la’apai (part o f the T on gan group o f islands), and they prob ab ly originally repre­ sented the T on gan Siam ese twin culture hero­ ines N afanua and Tokupulu. T he T on gan Siam ese twin god d esses (C o lo cott 19 2 1 ; Reiter 1907), w ho play an im portant role in the T on gan creation myth (and were responsible for assurin g plentiful catches for T on gan fish­ erm en), have Sam oan coun terparts in the form o f the tw ins Taema and T ilafaega , a lso known as N afanua the godd esses o f tattooin g and w arfare respectively (K raem er 1902). I d o not w ant ro expan d on the m ythological detail, but I w ould like to offer som e suggestion s as to why Jan ifo rm im ages o f this type had so much ritual significance in the area. The Siam ese twin archetype com bines closure and m ultiplication in their m ost ele­ m entary form s. The m ost ou tstan din g feature o f a Siam ese twin is sym m etry about the verti­ cal a x is, not only in one plane (left/right) but a lso ab o u t the other vertical plane (front/back). In other w ord s, a Siam ese twin has no “ b a c k ” , only a m iddle and a periphery. Such a being is protected, or en com passin g, in a way in which no ordinary m ortal can be. For merely onedim ensionally sym m etric beings the back

271

Tongan twin deities carved in whale ivory. Collected in Fiji by Cordon (Fladdoti Museum)

Figure 2

alw ays rem ains the m ost vulnerable a r e a .. . . Siam ese tw ins are im m une from such dangers, and have therefore a kind o f com pleteness which justifies their being considered am ong the im m ortals. Indeed, they seem to adumt>rate creation (as splitting) itself, since their im age seem s to capture the m om ent at which the god created the w orld by splitting ap art the plenum , which w as him self. Siam ese twins obviously qualify a s “ m ultiplied” beings, but at the sam e time they are closed in that their tw ofold sym metry closes o ff that part o f the ordinary being which is open to assau lt from the back.

272

ALFRED GELL

It w ould therefore be natural to expect that one o f the strategies through which ordin ary m ortals m ight seek to recapitulate, in them ­ selves original co sm ogo n ic splitting, and thereby assure the integrity o f their person s, might be to im itate the tw ins, and attach a secondary person to their backs. We are able to know that this possibility occurred to the m inds o f the inhabitants o f W estern Polynesia, from where the statuettes o f the Siam ese twin god d esses com e, not because they practised artificial tw inning, but because they were specifically disallow ed from doing so. G eorge T urn er, a nineteenth-century m issionary eth ­ nographer, recorded that in Sam oa it w as fo r­ bidden for m em bers o f the fam ily to sit leaning again st one another back-to-back, on pain o f im m ediate fatal, divine retribution (Turner 1884). I believe that in con tem porary C h ris­ tian Sam oa it w ould still be thought very im proper to ad o p t this posture. Turner a lso noted that all fruit or vegetables that grew together to form a fused pair (as happens som etim es with b an an as, taro, yam s, etc.) were im m ediately to be offered to the g o d s and might on no accoun t be eaten, f ie m akes both these ob servation s in the sam e p arag rap h , and leaves one in no dou bt th at they were co n ­ nected; but he offers no m ore specific e x p la n a ­ tion. In the light o f w hat has been said so far, I think one can reason ably infer that the prin­ ciple o f the “ sacredn ess o f d o u b les” and the prohibition on back-to-back sitting relate to the im age o f divinities as protected, encom ­ p assin g back-less beings w ho escape from the fundam ental on tological deficit o f ordin ary m ortals, i.e. an asym m etrical relationship with tua , the back. T o becom e such a double being w as to usurp a privilege afford ed to the g o d s (or god desses) alone. It w as sacrilege, in other w ords. But although overt doubling-up w as not a feasible sym bolic strategy, since it infringed to o much on divine privilege (like eatin g “ d o u b le” fruit or vegetables) there were other, m ore indirect m eans o f achieving a sim ilar result. At this point, m ultiplication and closure begin to diverge. Let me take m ultiplication first. R eturning to the Siam ese twin im age, one possible reading o f it is th at it represents a m utual and sym m etrical parturition (note that

the W estern Polynesian mythic tw ins are all fem ale). There is a direct connection between the idea o f d ou b lin g and birth. M om entarily, at least, any w om an giving birth is a Siam ese tw in, and indeed in the M arq u esas (if not any­ where else, so far as I know ) the w ord for “ birth ” (fan au ) a lso referred to fused-double objects, such as double b an an as, etc. (l)ordillon 1 9 3 4 -5 ). It goes w ithout saying that there is a stron g relationship between cosm ogony and birth, where co sm ogo n y is specifically not conceptualized as creation ex nihilo: the G od o f the O ld T estam en t does not noticeably “ give birth” to the co sm o s w hereas the corre­ spon din g Polynesian texts, as has frequently been noticed, are perm eated with procreative sym bolism . Birth am o n g m ortals w as a lso a process o f co sm ological significance recapitu­ lating creation, in that children (in som e places only the children o f elite kindreds) were believed to com e from the po and to be, in fact g o d s; exceptionally sacred, ranking higher than their parents (O liver 1974). Because wom en were cap ab le o f recapitulating divine cosm ogon ic acts and were the m eans through which divinities entered the a o y it follow s that wom en were not just m ore sacred, but also m ore protected again st “ defilem ent” (which I construe as de-differentiation) than men. W om en were “ n atu rally” noa (m e’ie , etc.: vario u s synonym s), i.e. “ clear” , non-sacred in the sense o f being not endangered by other person s, though very capable o f harm ing oth ers, especially m ales, by infringing their tapu (H anson 1982). Tapu (T ah itian ra'a) w as the general quality o f person al, sacred distinctiveness, which w as continually subject to dispersion. It is usually glo ssed as “ sacred n ess” , and so much has been written on the subject, that I hesitate to em bark on a discussion o f it here. But, as briefly as possible, here are the salient points. O bjects and person s were tapu in sofar as they were held to be invested with po-derived, atuaderived attribu tes and qualities. Because these sacred properties were im m anent, they were liable to dispersion: any object, or person invested with tapu quality rem ained so only in sofar a s that ob ject, or person did not lapse back or becom e m erged, with the generality of things. Im m anent sacredn ess can be localized

AN ESSAY ON POLYNESIAN COSM OLO GY AND RITUAL

and preserved only via isolation from nonsacred things: it is quite different from our kind o f sacredn ess, which is transcendental in which the sacred object is secular until it has been consecrated by ritual action . With im m a­ nent sacredn ess it is the ob ject, or the person which is sacred sim ply by virtue o f being that object or person because, in som e w ay, it is itself an atu a. Broadly speakin g, things and person s contained tapu in p rop ortion to the extent to which their existence recapitulated co sm ogo n ic acts. Persons (especially o f the elite) were sacred because their com ing-intobeing recapitulated the birth o f god s. Creative activities, such as carving, canoe-building or tattoo in g were sacred because the activity of craftsm en is godlike, they cut things, m ake lines, hollow things our, and so on. Another w ay o f putting this is to say thar tapu is “ d if­ ference” in the sense that any thing or person that is non-substitutable, w hose being there an d then m akes a difference, is ch aracteristi­ cally tapu, and m ust be protected from indis­ crim inate con tacts that dim inish this difference. T he effect o f the philosophical attitude o f im m anence, just a s is the case in India, is to place the w hole weight o f the religious system on the preservation o f distinctions and boun d ­ aries (D um ont 1970). W here there is a god in everything, the need to keep things ap art is overw helm ing, because only this separation preserves essences, and essence precedes exis­ tence. W here boun daries are tran sgressed, annihilation follow s. T o return to the question o f the differential “ sacred n ess” o f m ales and fem ales. M ale tapu qualities were intrinsically vulnerable, though they a lso had their m eans o f defence, which I will consider shortly. But w om en, though no less im bued wirh tapu than m ales (som etim es m ore) were not vulnerable to quite the sam e extent, because a w om an is, so to sp eak, a com plete system o f differences in herself, a m icrocosm , w hereas m ales were irreducibly incom plete beings, dependent on w om en for their presence in the w orld (and their d ep ar­ ture from it). W om en w ere con tain ers and con d u ctors o f tapu quality, w hereas men sim ply possessed tapu like a static electrical ch arge, and could be much m ore easily deprived o f it. As the T ah itian chant says,

273

“ w om an is the shell o f w om an , because she is born o f w o m an ” . O n rhe surface this indicates a transitive relation w om an (1) —> w om an (2) —» (3) —» (n)

in which one w om an gives birth to another. But the phraseology suggests at the sam e time a recursive, self-referential relationship w om an (1) —» w om an (1) —> (1) —» . . .

i.e. a w om an giving birth recapitulates her ow n birth, ad infinitum. W hereas the schem es, a lso given in the chant, w om an —» man

ad m its o f no such interpretation. H ence, I w ould argu e, the procreative pow ers o f w om en m ake them temporally “ sym m etrical” beings (beings with no past and no future) in a way precisely an alo g o u s to the spatial sym m etry of Siam ese tw ins alo n g the spatial a x is front/ back. They are thus, like the Siam ese tw ins, both m ultiple and closed, their tapu not lessened by the fact that it is in no danger from w ithout. For m ales this solution does not exist: they faced an acute problem in becom ing suffi­ ciently noalme'ie in order to carry on a norm al existence, while not dispersin g their tapu quality in such a fashion as to lose their d is­ tinctiveness. They had essentially tw o p o ssi­ bilities: they could find other m eans o f im itating rhe natural characteristics o f Siam ese tw ins, w om en, etc., which enabled them to contain their sacredn ess so that it w as not in danger from w ithout, or they could disperse their sacredn ess under “ con trolled” con ditions. I will con sider instances o f both o f these. O n e m eans that men could and did em ploy in order to convert them selves from m ortals, with the fundam ental ontological deficit that m ortality im plies, into “ com plete” beings, recapitu lating cosm ogon y , w as tattooin g, which w as w idely practised in Polynesia, espe­ cially by men (alm ost exclusively by men in W estern Polynesia). T atto o in g w as an o b lig a­ tory p a ssa g e rite o f late adolescence for m ales, particularly in S am o a, which produced tatto o

274

ALFRED GELL

artists for the T on gan elite as well. As w as noted above, tattoo in g w as thought to have originated from the Siam ese twin god d esses, Tilafaenga and Team a (who brought it from Fiji). T his m ythical associatio n o f tattooing with Siam ese tw ins underlines the latent equ iv­ alence between the strategy o f m ultiplicationtw inning and the strategy o f tattooin g which w as predicated m ainly on “ clo su re” . O ne is able to know this by m aking a study o f ta tto o ­ ing ritual and the nom enclature o f tattoo designs. Figures 3, 4 , and 5 show tatto o designs, tw o from W estern Polynesia (Sam oa), the other from Eastern Polynesia (the M arq u e ­

Figure 3 Kramer°

Sam oan

male

tattooing

(rear),

sas). O f the three, the Sam oan exam ples are rhe sim pler, so I will discuss it first. Sam oan tattooin g w as a necessary p rep a ra ­ tion for w ar and for sexual life; untattooed m ales m ight not engage in either. Both o f these were dan gerou s activities: it has already been m ade clear that w om en, by virtue o f their p o si­ tion a s m ediators between the po and the i/o, were a potent source o f spiritual danger. W arfare w as as well, indirectly, in that death in battle w as never merely a m atter o f m ilitary incom petence, but w ould alw ays be traced to spiritual debility. T attoo in g strengthened young m ales and prepared them for w arlike and am atory exploits. But how w as tattooin g conceptualized? T o begin with, let us note that the area that w as first tattooed w as the low er part o f the back, i.e. tua , and that over this vulnerable p art were placed designs o f a sp e­ cifically protective nature (Buck 1930; Kraem er 1902; Stair 1897). The back-design a s a w hole w as called p e'a , “ flying fo x ” , alluding, am on g other things, to the m otto o f the king-m aking descent-group o f Sava'i , the To?iumatpe>a , descendants o f N afanu a , which w as “ salvation com es from the flying fo x ” . M ore generally, rhe flying fox w as a well-protected anim al;

by Figure 4

Sam oan female tattooing

AN ESSAY ON POLYNESIAN COSM OLO GY AND RITUAL

275

KAMI HANAUPE

\/^

1 hope vehine

2

t i'i pepehipu mil

a malahoaia f i mho pea»a 3 li’i pepehipu mil jl

y

papua

a type o f spirit possession . In H o friyat, a village on the N ile som e kilom eters dow n stream from K h artou m , I found that in different years, 4 2 percent (1 9 7 7 ) and 4 7 percent (1 9 8 4 ) o f w om en ever m arried and over the age o f 15 have succum bed to this affliction .1 M arital statu s is a significant factor in possession illness. Z airan (zar spirits), H ofriyati assert, rarely trouble them selves with the unwed with w om en w hose fertility has yet to be acti­ vated. M o st affected are those between the ages o f 35 and 5 5 , am o n g w hom tw o-thirds ackn ow led ge them selves to be p ossessed . The latter proportion is due to a cum ulative effect: once possessed , a w om an is alw ays possessed thereafter. T h is paper offers an interpretation, grounded in the H ofriyati w orld, o f the zar cult and w om en ’s participation in it. M y concern is to avoid view ing p ossession phenom ena in term s th at, though ou r culture finds them accessible, are foreign to H ofriyati - whether biochem ical reactions to nutritional deficiency (cf. K ehoe and G iletti 1981), or w om en’s instrum ental efforts to assu ag e their su b o rd i­ nate statu s by acquirin g g o o d s or garn er­ ing attention (cf. Lew is 197 1 , 1 9 8 6 ). Such ap p roach es m ay prove fruitful in assessin g an d tran slatin g specific cases o f possession illness, but since they neither accoun t for possession form s, nor adequately credit the

369

collective practice w hereby spirits are provided w ith the spaces in which to perform. Possession thus provides a context in w hich contem porary experi­ ence can be actively m ediated by past myth models, and vice versa. Second, it provides an instance in which the collec­ tive and personal clearly interpenetrate. Collective form s are internalized by indi­ viduals and, as Boddy shows, become self-transforming.

taken-for-grantedness o f spirits in the everyday lives o f the p ossessed , ultim ately they distort and im poverish w hat they propose to under­ stan d. If the aim o f the enterprise is to com prehend the scope o f possession phenom ­ ena, to situate them in their cultural con ­ texts, ethnographers m ust attend to their in form an ts’ experiences o f possession and not seek merely to explain them aw ay a s som e­ thing at once less dram atic and m ore clinical than they appear. In w hat follow s, I observe th at, as others (for exam p le, Bastide 1958; C rap an zan o 1973, 1977, 1980; Lam bek 1981, n .d.; K apferer 1983; Kleinm an 1980; M etrau x 1 9 5 9 ; N elson 1971; O beyesekere 1981) w ho have investi­ gated such cults have foun d, spirit possession in Sudan is concerned with fundam ental ques­ tions o f identity and selfhood. Further, the preponderance o f w om en am o n g the possessed can be seen to derive from an interplay o f factors, salient am on g them w om en’s repro­ ductive function as it is culturally constructed in H ofriyat. A m ajor issue addressed by the zar is, I suggest, a problem o f socialization: the cultural overdeterm ination o f w om en’ s selves. A nd, although I agree that possession trance and ritual are legitim ate psychotherapy, I w ould not sto p short, as have others who pursue this app roach (Bourguignon 1979: 2 9 0 - 1 ; Kennedy 1967: 191; Prince 1964: 115), o f acknow ledging their potential to foster insightful reflection in the possessed . The p ro v ­ ince o f zar is m eaning, and it is best addressed in that light.

370

JA N IC E B O D D Y

The Cultural Context: Female Selves H ofriyat is a village o f som e 5 0 0 A rabicspeaking M uslim residents and is sim ilar in size, ch aracter, and com position to dozens o f other farm ing settlem ents that cling to the N ile in the desert north o f K hartoum . H ofriyati are loosely organized into a num ber o f patrilineal, variably co rp o rate, and putatively endogam ous descent grou p s. Ideally, m arriage takes place between patrilateral parallel cousins but in practice the m ajority o f couples are close cognatic kin w hose natal fam ilies reside in H ofriyat. T he m ost im portant criterion o f m arriageability is that the fam ilies o f p ro sp ec­ tive spouses be boun d, prior to their w edding, by a thick net o f m oral ob ligation , the legacy o f p ast interm arriages and generations o f ram ifying kinship. In H ofriyat, m arriage is inform ed by an idiom or sym bolic orientation I have elsew here term ed “ inferiority” (1 9 8 2 ; 1989): an inw ard focus, a concern for lim iting social and physical openings, which co n sti­ tutes a general organizing p aradigm and aesthetic stan d ard o f H ofriyat culture. “ Inferiority” expresses the defensive orien ta­ tion o f the village, an orientation sustained, in part, by a history o f invasion, colonization, and exp loitatio n from w ithout, and sim ilar to the stance o f other peoples living in precarious social and physical environm ents (cf. Fans 1972; Isbell 1978). The idiom o f inferiority or defense underw rites a range o f other cu stom s (dddt), significant am o n g them the genital operation perform ed on fem ale children know n as pharaonic circum cision (excision o f the labia and clitoris follow ed by alm ost com plete infibulation: intentional occlusion o f the vulva entailing obliteration o f the vaginal m eatus). A ccording to villagers, this operation co m p le­ m ents the procedure o f rem oving the penile prepuce for H ofriyati boys and, like it, is per­ form ed between the ages o f 5 and 10. Prepubescent circum cision accom plishes the social definition o f a ch ild's sex by rem oving physical traits deem ed ap p rop riate to his or her o p p o ­ site: external genitalia in the case o f fem ales, the covering o r “ veil” o f the penis in the case

o f m ales (cf. a lso A ssaad 1980: 4). The socially salient reproductive organ o f the m ale is e sta b ­ lished as such when exposed, and that o f the fem ale - the w om b - when covered and enclosed. G enital surgery ritually initiates the process o f genderization, and im plicitly identi­ fies neophytes with their gender-appropriate spheres o f operation as adults: the interiors o f h ouseyards enclosed by high m ud w alls in the case o f fem ales; the outside w orld o f farm ­ lands, m arkets, other villages, and cities in the case o f m ales. Fem ales are associated with enclosure, with the m aintenance o f life within the village; m ales are associated with the precarious ou tside w orld, with political and econom ic life. Follow ing their operation s children begin to perform gender-specific tasks and are increasingly segregated from their counterparts. I...] A lthough circum cision restrains fem ale sex ­ uality, this is not, [women] say, its purpose. T he surgery, in that it is “ h o t” or “ pain fu l” (hdrri), prepares a girl for w om an h ood; m akes her body clean, sm ooth , and pure; renders her m arriageable; con fers on her the right to bear children; and invests her with fertility (cf. Boddy 1982). Fertility and sexuality are, o f course, tw o sides o f the sam e coin, yet each sex em phasizes one m ore than the other. Both point to a fundam ental concern in H ofriyat with hum an reproduction, the responsibility for which rests principally with w om en. Only when physically transform ed and shaped to the im age o f hum an m orality can they be entrusted to reproduce. W om en’s susceptibility to possession is, I subm it, closely bound up with the im plications o f ph araonic circum cision for a fem ale ch ild’s developing self-perception and is, I will argue below , indicative o f w om en’s problem atic, socially overdeterm ined selfhood. I suggest that through this operation and ancillary p ro ­ cedures involving heat or pain - the tw o are equated here - app ropriate feminine d isp o si­ tions are being inculcated in young girls, d is­ position s w hich, follow ing Bourdieu (1 9 7 7 : 15), are em bedded in their bodies not only physically, but also cognitively and em o ­ tionally, in the form o f m ental inclinations, “ schem es o f perception and th ough t.” The

SPIRITS A N D SELVES IN N ORTHERN SUDAN

traum a o f p h araon ic circum cision alone is insufficient to cultivate a prescriptively fem i­ nine self: such acts m ust also be m eaningful to those w ho undergo and reproduce them. For villagers, m eaning is carefully built up through m etaphors and associatio n s that operate implicitly and overtly to establish an identification o f circum cised wom en with m orally ap p rop riate fertility, thence to orient them subjectively tow ard their all-im portant generative and tran sform ative roles in Hofriyati society. I-I Both m en’s and w om en’ s everyday experi­ ence is inform ed by the inw ard orientation o f H o friyat; yet the surgically altered bodies o f w om en actively sym bolize this cultural logic and their society’s principal values. In this exist som e addition al im plications for w om en’s identity: (1) H ofriyati w om en bear the onus o f m aintaining those values, a responsibility am plified by increasing m ale labor em igration (cf. H ale 1 9 8 5 ); (2) w om en, w ho are identified with the inside, are not only protected but also dependent; and (3) intrusions of pow erful forces from w ithout are m ore likely to be reg­ istered in fem ale bodies than in m ale ones. T ak in g these points in order, not only does a village w om an ’s self-im age rest squarely with her procreative ability, but also her sense o f w orth and social value. For only by giving birth, legitim ately, to son s, m ight she achieve a position o f respect and inform al authority a n d , correspondingly, m ight her husband fulfill the ideal m ale role as founder o f a lineage section. But m ore than this, the equation o f femininity with socialized fertility and o f the fem ale m arital role with reproduction m eans that responsibility for ensuring successful procreation - for giving men fitting, properly raised descendants and continuing society belongs alm ost entirely to w om en. So, in the first stage o f this process, if a w om an fails to conceive within a year or tw o o f her m arriage, if she m iscarries, bears a stillborn child, p ro ­ duces a daughter with her first pregnancy, or loses an unw eaned infant, it is her ability to procreate that is called into question, not her h u sban d’s, and it is she w ho suffers a s a result, through divorce or co-wifery. Such occurrences illum inate a fundam ental co n tra­

371

diction for w om en in H ofriyat: they are collectively indispensable to society while indi­ vidually dispensable to men. At the very least, fertility problem s place intense strain on a w o m an ’s self-im age and often fragile m arital relations. A s noted earlier, the sexes are segregated. W om en are expected to spend m ost o f their time within fam ily co m p o u n ds bounded by high m ud-brick w alls; m en, on the other hand, have con siderable m obility. The sexes rarely eat together, and never do so in public or before guests. In traditional households - the m ajority - they also sleep in separate quarters. In m ost respects there is a significant rift between m en’s and w om en’s w orlds, which is w idened by labor em igration, entailing pro­ longed separation s o f h usbands from their w ives (cf. Kennedy 1978: 10 for Egyptian N ub ian s). W ithin this polarized w orld, the sexes are conceived to be econom ically and productively interdependent: h usbands and adult son s bear prim ary responsibility for production - they provide for their fam ilies through w age labor and/or farm ing - w hereas wives and adult daughters consum e and tran sform m aterials into item s for consum ption. R elated to this, the essential dialectic between h usband and wife, set in m otion at their w edding, is that between producer and social reproducer. The dual role o f a m arried w om an , that o f co n ­ sum er and reproducer, is aptly expressed in sym bols and m etaph ors that identify her with pigeons and livestock: dom estic an im als kept within fam ily houseyards, fed on grain and fodder supplied by m ale lab or, and valued apart from their use as food - for their c a p a c ­ ity to beget offspring. In other m atters, asym m etric gender co m ­ plem entarity extends to religion and, until recently, education. M en participate fully in the public rites o f Islam and m ost, having learned the Q u ’ran, are functionally literate. They belong with rare exception to one o f the tw o religious fraternities in the area and regu­ larly attend their cerem onies, called zikrs. On the other hand, w om en over the age o f 25 are largely illiterate. They perform their daily prayers in private, if at all, and only the elderly am o n g them attend m osque. W om en are

372

JA N ICE B O D D Y

som etim es perm itted to w atch a zikr hut are alw ays barred from active participation , and whether at m osque or a zikry they are segre­ gated from men. As these poin ts dem on strate, neither sexual com plem entarity nor the sym bolic value o f femininity im plies gender equivalence. A w om an ’s life is subject to notable con strain ts in H ofriyat: she is forever a jural m inor, m orally and financially governed by her m ale kin, subordin ate also to her husband. M o re ­ over, since personal integrity, dignity, and em otional control are highly valued in both sexes, and d isp lays o f em otion considered vulgar - especially so in m ixed com pan y - a w om an m ay be hard pressed to dispute her husband or brothers should she feel w ronged. H ow ever constrained their interactions, and how ever weighted the system w ould seem in favor o f m ales, it m ust be rem em bered that both sexes participate in and reproduce H ofriyati culture. Both are subject to its pres­ sures, though in different w ays; for both there are ad van tages and d isad van tages, if men are better able to cope w'ith problem s in a m ar­ riage by takin g a second w ife or divorcing the first, they are still under considerable o b lig a ­ tion to conform to kinship values, m asculine ideals, and the tenets o f Islam . If w om en are m ore restricted than men in other w ay s, they are less con strain ed with regard to religion: they are neither expected to becom e fam iliar with m atters o f liturgy and doctrine, nor to have sufficient m oral strength on their ow n to uphold them . Phrased positively, this m eans that w om en are relatively freer to em brace w hat men con sider folk beliefs, those h av­ ing so-called p agan elem ents and in w hose com pan y they place the z a r . . . . Although men publicly scorn the zar , pri­ vately they are nor so intractable. For they, to o, recognize the superior pow ers o f zairan 9 and, fem inine associatio n s notw ithstanding, a few told me they believe them selves to be p o s­ sessed but w ould not openly adm it to rhe affliction for fear o f losing face. So, despite app earan ces o f o p p o sitio n , men tolerate w om en’s involvem ent in the cult and are generally w illing to provide for the sp irits’ dem an ds. Few are inclined to d ou bt its efficacy.

Zairan and Zar Possession In order to g rasp w hat is involved in w om en’ s p ossession , we need to know som ething m ore ab o u t the spirits them selves. Zairan belong to a class o f beings know n as jinn, w hose e xis­ tence is substan tiated in several verses o f the Q u ’ran. A ccording to H ofriyati, jinn are the physical com plem entaries o f hum ans in a holistic, q u adripartite creation: where hum ans are com posed o f earth and w ater (A dam , the first hum an, is said to have been fashioned by Allah from m oist clay), jinn are m ade up o f fire and w ind, or air. H u m ans are visible, su bstan tial, and diurnal; jinn are norm ally invisible (though able to take hum an shape), form less, and nocturnal. T h u s, jinn are natural beings - they are born, eventually they die bur their nature is such that they can n ot wholly be confined by physical barriers like those (bodies, w alls, an d so forth) that contain hum an beings. They are therefore able to infil­ trate hum ans and take possession o f rhem at will. O nce in p ossession , a jinn can influence rhe health and beh avior o f its hum an host; it does so either from w ithin the body or by takin g up a position above the head. M o st jinn are assim ilated into three categ o ­ ries, coded by co lo r. White jinn are benign; p osession by one is not serious and in fact m ay g o unnoticed. Black jinn or devils (shawdttn) bring grave d isease and intractable m ental illness; possession by one is a dire m atter and cu rable, if at all, only by violent exorcism . Sickness caused by a black jinn m ight well result in death. L ast, there are red jinn or zairan , w hose co lo r points to a characteristic a ssociatio n with blood and hum an fertility. T hese are pleasure-seeking, cap riciou s, am b iv­ alent beings thar bring m ilder form s o f illness which, though initially distressful, never result in death or severe m ental dysfunction. Should som eone w ho is zar possessed fall critically ill, then natural cau ses or sentient agents other than zairan are im plicated. Zairan m ust be placated an d d o not respond kindly to attem pted exorcism . Z ar possession is a lifelong, fundam entally incurable con dition that is, how ever, m an age­ able. Indeed, after its initial stages, it m ay be

SPIRITS A N D SELV ES IN N O RTH ERN SUDAN

transform ed out o f all resem blance to w hat we m ight consider illness. The possessed can hope to gain som e control over her sym ptom s first, by accepting - both subjectively and publicly - the p ossession d iagn osis, then by undergoing a curing cerem ony during which she enters, via tran ce, into a con tractual relation sh ip with the spirit(s) responsible for her lapse in health. D u rin g the cerem ony ch an ts are drum m ed invoking the schedule o f nam ed zairan. When the spirit that plagues her is sum m on ed, the patient ideally enters tran ce; now identified, the intruder m anifests itself through her body and m akes know n its d em an ds, in return for which it should agree to restore, and refrain from further jeopardizing, her w ell-being. Yet the successful con clu sion o f a w om an ’s curing cerem ony by no m eans ends her a sso ­ ciation - in or out o f trance - with her zar. T h e spirit is said to rem ain above her, ready to enter her body at will. From a spirit’s per­ spective, con tracts with hum ans are infinitely r e n e g o tia te : if the p ossessed w ishes to allay further attack from her zar she m ust mollify it continually. T his requires her regular atten­ dan ce at the zar cerem onies o f others and per­ form ance o f certain ritual acts on the spirit’s beh alf (for exam p le, staining her hands or feet with henna in a p articu lar [spirit] design). Should spirit or hum an neglect to uphold their agreem ent at any tim e, the latter m ay suffer relapse o f her form er illness. Yet the “ cure” h as opened com m u nication s between the tw o and any future difficulties can be dealt with expeditiously. If all goes w ell, w hat begins as an uneasy truce between a w illful spirit and its reluctant host m ight g rad u ate to positive sym ­ biosis as their relation sh ip stabilizes and m atures. O ne thing that the hedonistic zar hopes to gain by p ossessin g a hum an being is a venue for access to the hum an w orld in cerem onial co n texts where it can frolic an d be entertained. T h u s, when a spirit is ritually sum m oned by its chant, it sim ultaneously infiltrates each o f its entranced hosts so as to interact with the hum an assem bly through their bodies. Such rituals are alw ays fraught with tension and surprise, for at any m om ent a w om an might be “ seized” by a spirit that H ofriyati did not before know existed , or she did not know she

373

had. O ne by one, th roughout an evening’s drum m ing, the spirits “ descen d” (nazal) into their hosts in the order that their chants are played. A spirit enters its host when called upon an d, if well behaved, relinquishes her body when its chant is over an d the rhythm shifts to that o f another zar. Since an adept m ay be p ossessed by several different spirits at once, a w om an m ight be in and out o f trance all night a s her various spirits descend and m anifest them selves for up to 2 0 m inutes each. When not entranced she participates in the drum m in g and chanting while observing other w om en w ho are. T h u s, in Flofriyat a w om an is not co n ­ sidered to be p ossessed because she becom es entranced; rather, she becom es entranced because she is possessed. A ccording to villag­ ers, since p ossession trance fulfills their p art o f a b argain with the spirit w orld to restore and m aintain hum an health, it is not path ological but therapeutic. Y et ju st how it m ight be th era­ peutic rem ains unresolved, and is an issue I will shortly address. O n leaving trance w om en say they “ see things differently” and they “ feel w ell,” statem ents that m ight be construed as affirm ing that the state itself has intrinsic rem e­ dial pow ers. But, even should this be the case, to lim it the therapeutic value o f zar to that o f trance w ould be artificial and in adequate. The zar rite is a cultural therapy; its curative pow ers derive less from a virtual experience o f trance than from the entire p ossession con text that renders it, and countless other experiences, m eaningful. Indeed, an individual’s experience o f trance is largely constructed by its context: in H o friy at, possession provides her with a m odel o f w hat trance is and should be like. Still, if trance is an integral p art o f p o sses­ sion therapy and relapse prop h y lax is, it is only one m anifestation o f p ossession , not co n sis­ tently evinced by the possessed during cerem o­ nies an d , when evinced, variable in apparen t depth and duration from one individual to the next. N o t all w ho enter trance d o so for im m e­ diate therapeutic m otives; m any w ho m anifest their spirits d o not feel sick at the time, though they are classed a s ill. A zar rite is m ore than ju st a cure; it is a lso referred to a s a “ p a rty ” (bafla) (cf. Sau nders 1 9 7 7 on zar in Egypt). D espite its solem n aim to alleviate suffering, it

374

JA N ICE B O D D Y

can be a great deal o f fun, m ixing com edy, satire, and intellectual challenge in a heady atm osphere w here nothing is quite as it seem s. Possession trance is, 1 contend, therapeutic, but its therapeutic potential is b ro ad , not co n ­ fined to achieving a m edical or psychological cure, and lies as m uch with observing it in others a s experiencing it oneself. Even the experience o f it is subordin ate to an earlier acceptance o f a p ossession d iagn osis. In short, the relevant issue in the case o f the zar is not trance per se, but trance firmly situated in a m eaningful cultural con text - possession having m edical, social, psych ological, and often profoun d aesthetic im plications. Before considering these, the nature o f the illness that signals p ossession rem ains to be addressed.

Possession as Illness Here I will begin to pull together the threads o f my argum en t, indicating how p ossession is linked to w hat I consider is H ofriyati w om en’s problem atic selfh ood. But on this issue, espe­ cially, I w alk the eth n ograph er’s unsteady line between keeping faith with in form an ts’ experi­ ence o f their w orld and rendering that experi­ ence intelligible in the m ore fam iliar parlance o f W estern culture. In the attem pt I m ust surely transgress realities on either side. For the zar in H ofriyat is a holistic phenom enon; it pene­ trates every facet o f hum an existence. C o n se­ quently, it defies analytic reduction to a single constituent dim ension: p sych ological, m edical, or social, with which m em bers o f W estern cu l­ tures m ight feel m ore at hom e (cf. C rap an zan o 1977: 11). I have no dou bt that in partially subjecting the p ossession idiom to an alien culture’s con stru cts, som e o f its texture, rich­ ness, and cultural integrity will be lost. But sim ilarly, in ad ap tin g concepts that have accepted m eanings in ou r culture to asp ects o f another in which they d o not exactly apply, the concepts - such a s “ illn ess,” “ p e rso n ,” or “ self” - m ay be altered. The result is a partial distortion from either perspective. Yet, as the zar itself p ro p o se s, distortion s can som etim es be instructive. By now it is com m onplace to say that a sociocultural system provides the stresses that

trigger the illnesses to which its m em bers succum b, directs those m em bers’ copin g re­ spon ses, and furnishes an interpretive fram e­ w ork which renders their experiences m eaningful, enables them to be exp ressed, and suggests possible therapeutic resolutions. Earlier I described som e stresses and c o n ­ straints to which H ofriyati wom en are e xp o sed , and noted the stron g identification they are im plicitly and m aterially subjected, in the Foucaultian sense (Foucault 1980: 97), to feel with their fertility. When a w om an ’s fertility m andate is im paired - for w hatever reason her self-im age, social position, and ultimately general health are threatened. W om en un der­ goin g severe m arital and/or fertility crises tend to phrase their experiences as illness or, less directly, to co-locate their difficulties with the onset o f apparently unrelated physical sym p­ tom s. In thus com plain in g o f illness, a w om an avails herself o f a culturally sanctioned m edium for articulating her dysph oria (C onstantinides 1977: 6 5 ; cf. Kleinm an 1980: p assim ; with specific reference to com plaints o f p ossession illness, cf. C rap an zan o 1977, Firth 1967, Lam bek 1981, O beyesekere 1970). Once this is done she can act upon her problem s, where before she could not, by setting out to find a cure. T he w om an w ho claim s to be ill but does not ap p ear diseased does not feign sickness; her pain is real and m ay be attributed to w hat, from her perspective, are natural agents:

zairan. Flofriyati them selves link fertility problem s and illness, although m ore subtly and obliquely than I have just proposed. A w om an w ho is an xio u s or depressed is considered a prim e target for zairan seeking entry to the hum an w orld; should such a spirit descend upon her it m akes her feel “ unw ell.” But, tau tologically, zairan are able to create the very circum stances that m ake a w om an an xiou s and prone to spirit assau lt. Their m ost com m on tactic is to “ seize,” “ h o ld ,” or “ steal” offspring, bringing about m iscarriage, stillbirth, am enorrh ea, or other problem s affectin g w om en’s blood. T hus spirits can hold a w om an ’s future and that o f her h usband for ran som , forcing both to acknow ledge their presence and accede to their dem ands. Z airan have considerable leverage over hum an reproduction, and for this reason

SPIRITS A N D SELV ES IN N O RTH ERN SUDAN

also arc draw n ro m arried w om en and unlikely to beset the unwed. H ow ever, not all p o sses­ sion incidents can be traced to uncertain fertility. . . . [--•I

'

. . . The dem ands o f possessive spirits that their hosts eat “ clean ” fo od s; abstain from traditional m ourning behavior such a s sleeping on the groun d; bathe with im ported (Lux) so a p ; use henna and cologn e; w ear gold and clean clothing; and avoid overw ork, anger, and frustration with kin, all have to do with the m aintenance o f fem inine ideals. C om bined with the link between possession and fertility, the above points suggest that for H ofriyati w om en, p ossession is intim ately bound up with feminine self-im age and stan d ard s o f conduct. It is a condition perpetrated by cap ri­ cious external agents, which dem and preserva­ tion o f a w om an ’s ideal self, but m ay create or seize upon a situation in which these ideals are jeopardized in order to take control o f her body, thus tran sform in g it into a vessel o f otherness. This they d o so as ultimately to enjoy the earthly pleasure that upholding such ideals provides. But, m ore convoluted still, the spirits that so adm ire H ofriyati w om anhood are them selves exem plars o f its antithesis. This is really the crux o f the possession phenom e­ non, and requires further explication.

Discussion: Spirits and Female Selves l-.-l

Zairan inhabit a w orld parallel to our ow n and con tiguous with it, yet im perceptible to hum ans except under certain conditions. Like us, they are divided into ethnic grou p s, fam i­ lies, occup ation s, and religions. They are either m ale or fem ale (principally the form er), and children, adults, or elders. They are neither wholly go o d nor wholly bad but, like hum ans, som ething o f both. Their salient characteristics are am bivalence, am orality, and caprice, in all o f which they differ from hum ans only by excess. A m ong their num bers are the spirit an alogu es o f M uslim sain ts, Turkish adm inis­ trato rs, “ E u ro p ean s” (including N orth

375

A m ericans, H indus, and Chinese), Ethiopians, Syrian gypsies, W est A fricans, nom adic A rabs, and Southern Sudanese, in short, o f all human groups with whom H ofriyati have had contact over the past 150 years or m ore. Each spirit within these categories is nam ed and has an individual history, typical behaviors when ap pearin g in the hum an w orld, and relation­ ships with other zairan that crosscut ethnic and religious lines. Significantly, the spirits that p o ssess H ofri­ yati belong only to foreign societies: villagers have no zairan that m atch them selves. Indeed, zairan epitom ize all that is not H ofriyati in an integral, sentient universe; they are quintes­ sential “ oth ers,” at once physically, socially, and culturally alien to hum an villagers. This is especially true for H ofriyati w om en, since zairan represent extrem e exteriority: they are pow erful outsiders p ar excellence. Even M uslim zairan , exhibiting the positive traits of piety and self-control, ultim ately subvert local values by being wed exo gam ou sly , to non-kin adherents o f other faiths. Unlike circum cised, socialized w om en, w ho represent a victory o f aqel over their naturally dom inant nafs - o f social concerns over self-interest - zairan are beings in w hom nafs is perm itted free rein. C onsidering the qualities o f zairan and vil­ lagers’ cultural logic, it is unsurprising that w om en should be thought m ore vulnerable than men to spirit assau lt. Clearly this has bro ader im plications than for d iagn osin g indi­ vidual illness. T h is is so because for H ofriyati, w ho have a keen sense o f w ho they are and how they differ from outsiders (cf. also Kennedy 1978), it is not only fertility but cul­ tural identity that is vested in w om en’s selves. In local thought the body is a virtual m icro­ cosm o f village society (cf. D o u g las 1966, 1973). Like village boun daries, body orifices are am b ig u o u s, however necessary and inevi­ table. They are prone to a litany o f dan gers spirit intrusion not least - and regulated by com plex ritual procedures. They are co n sid ­ ered best, both defensively and aesthetically, if kept as small as possible. The m ost am b iguous and problem atic body opening is the vaginal m eatu s, for it is through w om en’s bodies and uterine blood that village society can be renewed ap propriately, from within itself,

376

JA N IC E B O D D Y

or in appropriately opened up to potentially destructive influences from w ithout. So a w om an ’s fertility m ust be defended, reserved and safeguard ed first by ph araon ic circum ci­ sion , then by en d o g am o u s m arriage. This brings us back to the issue o f feminine self-im age. I have discussed how through circum cision a w om an ’s body is tran sform ed into a living vessel o f her culture’s m oral values, and she is thereafter exh orted to conduct herself accordingly. Fem ale circum ci­ sion, plus the entire com plex o f asso ciatio n s and practices th at exp ress and realize fem inin­ ity, strongly su p p o rt the identification o f the individual with her role - both social and sym ­ bolic - in H ofriyati cu lture.2 T h is is the point: so tangibly socialized are w om en to this view o f them selves th at, for m any, to experience the w orld otherw ise is to experience it, quite liter­ ally, as a non-H ofriyati. T h u s reform ulated, a central problem that possession ad d resses, in hundreds o f idiosyn ­ cratic w ays, is the cultural overdeterm ination o f w om en’s selfh ood. T o review the various concepts o f “ se lf” and “ p erson ” found in the literature is beyond the scope o f this p ap er (see C arrithers, C ollin s, and Lukes 19 8 5 ); how ever, let me clarify my analytical usage o f these term s before giving closer con sideration to the H o fri­ yati case. D raw in g upon Burridge's argum ent in Someone, N o One: An E ssay on Individual­ ity (1 9 7 9 ), an d , in a parallel vein, from K egan ’s constructive-developm ental p sych ol­ ogy (1 9 8 2 ), the self as a theoretical construct can be provision ally conceptualized, not as an entity, but a s a creative energy or process, which actively engages rhe w orld, integrating the hum an biological organism with its physi­ cal and sociocultural environm ents, con tin u­ ally m oving, becom ing, m aturing, m aking and organ izin g m eaning (Burridge 1979: 5 ff., 2 1 ; K egan 1982: 2 - 1 5 ; cf. Elster 1986). Burridge w rites, “ the fact o f integration - som e sort o f coherence or coord in ation o f the parts o f con ­ stituents o f being - does not detach the in tegra­ tive energy or self from its con stituen ts, but still m akes it m ore than the sum o f the parts and, in th at sense, conceptually and em piri­ cally d istin ct” [1 9 7 9 : 5]. In these term s, a self that is integrating in conform ity with others realizes the “ p erson ” : “ [one) w ho, in repro­

ducing in w ord an d deed the norm s o f a given traditional order, m anifests the relations o f that trad itio n ” (1 9 7 9 : 5). In H o friyat, the extrem e identification o f wom en with the cul­ tural im age o f w om an h ood precipitates such a com pression o f the subjective self into an ob jec­ tive fem ale person: a norm ative set o f given roles and statu ses, an entity in w hom exp eri­ ence is continuously subordin ated to cultural categories (Burridge 1979: 2 8 , passim ), a p u b ­ licly confirm ed social representation (La F o n ­ taine 1985: 124). T he projections o f significant others - just a s determ ined, perh aps, by fem i­ nine im ages and ideals - dom inate her internal dialogue. She is com pelled to personhood. W hat the H ofriyati w om an does not becom e or, better, is not at this stage given scope to becom e, is, in Burridge’s term s, an “ individ­ u a l” : a “ m oral critic w ho envisages another kind o f social or m oral ord er” (1 9 7 9 : 5). Becoming aware o f a gap between the person’s reproductions and the truth o f things by seizing on or being seized by peculiarly signifi­ cant events, the self is moved to a transcendance of the traditional categories, to a reintegration of the event in a new rationaliza­ tion assigning new meaning and relevance. In this transcendance and reintegration, manifest in the new rationalization, the self realizes the individual. [1979; 7) Burridge su ggests that m ost people oscillate between these tw o integrative m om ents (p. 5); yet, given the close identification wom en are encouraged to feel with their fertility, the q u o ­ tidian H ofriyati con text actively conspires again st such m ovem ent and the realization o f “ individuality” in their case. It effectively denies them the possibility to grow in selfaw aren ess, to m ature, to reflect on the categ o ­ ries o f early socialization in which selfh ood is enm eshed. W om en’s selfhood is, I subm it, cul­ turally overdeterm ined. Why overdeterm ined? Because, p a ra d o x i­ cally, the m oral self-im age w om en are enjoined to assum e can n ot alw ays be sustained by e x p e ­ rience. In this lies an am biguity inherent to m orality - between w hat is and w hat ough t to be - which their continuous socialization m ay eventually fail to overcom e: the w om an w ho is, by definition, “ m orally app rop riate fertil­

SPIRITS A N D S ELV ES IN N O RTH ERN SUDAN

ity ,” m ay experience infertility or som e other significant contravention o f her feminine self-im age. The tension that the aw aren ess o f such untow ard experience creates in her is first co n ­ strued as a problem o f internal d isorder and registered a s illness: in H o friyat, body and person are not distin guished; hence, threats to on e’s sense o f self and to the w orld in which on e’s self is located autom atically jeopardize physical w ell-being. W hen this is coupled with the restriction on exp ressin g em otion, the som atization o f her d ysp h oria seem s a n orm al, culturally ap p ro p riate response (cf. Kleinm an 1980). It has the added effect o f directing attention tow ard the one im periled and aw ay from a precipitating con text likely to involve close superordin ate kin, w hom w om en, espe­ cially, feel pow erless to rebuke o r entreat. T his leads to an im portan t point: the feminine self in H ofriyat is constituted n ot only, a s i have suggested, ideally, but a lso (am on g other ways) relationallv. It is a virtual im possibility for a villager to think o f him self o r herself except in relation to kin. T h u s illness, a s an idiom for the expression o f threatened selfh ood , actually defends certain constituents o f the self, notably, its form ative relationships. A great m any villagers som atize negative affect, but not all g o on to con sider their illness a sym ptom o f p ossession . Such adm ission sign als a p rofoun d tran sform atio n o f context, from one narrow ly described by idiom s o f inferiority that govern the com m onsense w orld, to a broader one th at places this orien­ tation in relation to its converse: dysphoria for the p ossessed originates not within the self and its constituents, but ou tside them , indeed, ou tside the hum an w orld o f H ofriyat. T his, on the one h an d, ration alizes the untow ard event in a w ay which vehemently defends the social­ ized self, for the self’s experience is again su b ­ ordinated to “ n atu ral” categories - zairan - how ever extraordin ary they m ight seem. From an observer’s perspective, zairan sym ­ bolize and render concrete a w om an ’s exp eri­ ences o f the w orld that conflict with her con sen su ally validated view o f w hat that exp e­ rience should be like. T h u s, with acceptance o f a p ossession d iag­ n osis com es d isassociatio n o f the experience

377

from her self. Yet this not only su p p orts her self-im age, but also plants the seeds o f its m odification . First, by shifting the context o f her illness from one o f internal con trad ic­ tion to external con fron tation , o f self or selfand-village-other to self-plus-alien-spirit, the potential for a negotiated resolution becom es ap p aren t. M ore than this, zairan are represen­ tatives o f non-H ofriyati cultures an d by virtue o f their extraordin arin ess, o f their abilities and pow ers, o f their failure to con form with local norm s and rules even a s they intervene in the course o f village life, p ossession as a hum an condition allow s for the possibility o f am b igu­ ity an d otherness, otherw ise lacking in the gender socialization process for w om en in H o friyat. The con text o f the p o ssesse d ’s situ­ ation is now w idened to incorporate the actions o f beings from an alien yet parallel w orld. Roth the entropy o f well-being and location o f its source in p ossession open up pathw ays for self-renew al, perm itting a limited an d , in this co n text, functional disson ance between person and self. In such cases illness itself may be therapeutic. W hat I am describing is, I think, rather dif­ ferent from w hat ap p ears to happen in em o­ tional disord ers com m on to W estern cultures. If the self is truly a social construct and individual selves are constructed (cf. Berger and Luckm ann 1967) or integrated (Burridge 1979) in the course o f social interaction, the con stitu en ts and param eters o f selfh ood can be expected to vary from society to society. So, while parallels can be found between hysteri­ cal neurosis and H ofriyati p ossession , in that both con dition s involve dissociation and m ay present initially a s som atic co m plain ts, these m ay be m ore ob viou s than real (cf. W ard 19 8 2 : 4 1 6 ). The tw o “ illnesses” are groun ded in dis­ parate cultural co n texts, based on rather different con ceptualization s o f the self. At the risk o f sim plification, perh aps one could ch ar­ acterize certain neuroses in W estern cultures, where “ self” is conceived as a boun ded, indi­ viduated entity (cf. Geertz 1983: 5 9 ), as an overdeterm ination o f selfhood w hose sym p­ tom s are excessive subjectivity - a w eakening o f the ability to take the role o f the “ o th er” relative to on e’s self. In H ofriyat, where the essential fem inine self is highly idealized, the

378

JA N ICE B O D D Y

problem seem s to be one o f excessive ob jecti­ fication: self is firmly identified with village “ oth er” an d identity em otionally realized in cultural sym bolism to the point where any event perceived to negate that tenuous e q u a ­ tion negates the w om an ’ s self. T h u s, the m ost striking sim ilarities occur between rhe n orm a­ tive p rocess o f curing or accom m od atin g p o s­ session illness (that is, disengaging the self from its context) and the aberran t one o f developing a neurosis - resulting in m any an unfortunate lay observation that ad ep ts are chronic hysterics. D espite thorough d isassociation o f the untow ard event from the H ofriyati w om an ’s self, she does not, like rhe textb oo k hysteric, unconsciously deny her experiences o f otherness so much as em brace them , while consciously recognizing them a s aspects o f her being over which she has limited if potentially increasing control. If we see spirits a s sym bolic o f sym ptom s (cf. O beyesekere 1981: 3 4 -5 ), and sym ptom s as id iom atic o f spirit intrusion (cf. K apferer 9 83: 8 7 ; I.am bek 1981: 53), we d o not stray far from H ofriyati logic. But if we view spirits and sym ptom s as dissociated facets o f the p o s­ sessed w om an ’s self, as our ow n psychology might direct us to do (cf. LaB arre 1975: 4 1 ; B ourguignon 1979: 2 8 6 ), we violate villagers’ reality. We tacitly dism iss zairan as facts o f H ofriyati existence and m istakenly em ploy a highly individualistic and com partm entalized concept o f the self which has no basis in village culture. T h is, in turn, leads to an individualis­ tic orientation to illness w hich, because it rarely ad d resses social context, m isses the point o f m ost illness in H o friyat, including p ossession . M oreover, even the assum ption that spirits represent projections o f intolerable feelings is, a s C rap an zan o (1 9 7 7 : 12) notes, a debatable one: spirits, like the illnesses they cau se, originate outside the hum an self, not within it. Unlike W estern psychotherapy, which encourages the patient to accept and integrate previously dissociated feelings a s part o f herself, zar therapy w orks by convincing her to recognize them as separated from herself in the first place. C learly, any attem pt to merge such feelings and experiences with the H o fri­ yati w om an ’s self - which I have described as idealized and relation al, but is, after all, her

self - w ould be ethnopsychiatrically in ap p ro ­ priate. It could only deny her the validity o f that self and potentially do m ore harm than goo d. A gain from an observer’s perspective, one way to m ake sense o f all this is by reference to the concept o f fram ing. Follow ing Elster (1 9 8 6 : 2 7 ), the H ofriyati w om an ’s illness is “ refram ed ” by a diagn osis o f possession in such a w ay that the precipitating behavior or event - for exam ple, infertility - becom es co m ­ patible with her self-im age: she is fertile, for spirits have seen fit to usurp this m ost valuable asset. She generalizes that future untow ard experiences d o not underm ine the equation o f w om an h ood with fertility and all the rest; they signify the action s o f zairan , w ho, unlike hum ans, are cap riciou s and unpredictable. But for this there is a remedy. T hu s, although d is­ sociation m ay be psychologically adaptive for both W estern neurotics and H ofriyati p o s­ sessed, only for the form er m ay it be sym ptom ­ atic o f path ology. For H ofriyati it is therapeutic. M ost o f those w ho acknow ledge p ossession are com petent, m entally healthy w om en w ho have responded in a culturally ap p ropriate way to a stressful situation (cf. C rap an zan o 1977: 14). When a village w om an w ho feels unwell but has identified no organ ic or m ystical source for her com plain t accepts that she is possessed, she can begin to recover. H er possessive spirit or spirits, soon to be revealed, gradu ally take shape as a part o f her being which is not, so to speak, a part o f her person, her H ofriyati self. It is during possession trance that the identity and ch aracteristics o f this non-kin, non-H ofriyati, non-hum an, but above all non­ self-existent are publicly established, both for the w om an and those w ho observe her. Once establish ed, this veritable non-self is linked, inextricably, to her self; it is not, how ever, integrated with her person, a situation that possession rituals stress an d seek to m aintain. Still, though rarely m anifest in her body, the spirit constituents o f her non-self are in co n ­ stan t attendance, influencing her decisions and perceptions to the point where som e o f my inform ants spoke o f them selves as if they were pluralities, substitutin g “ w e” (nehna) for “ I” (ana). Furtherm ore, since a spirit might p ossess

SPIRITS A N D SELV ES IN N ORTHERN SUDAN

any num ber o f H ofriyati sim ultaneously, a w om an ’s non-self, like her self, is unlikely to be individualistic: p ossession by a com m on spirit binds her to other H ofriyati selves and zar non-selves, yet in w'ays other than those specified by kinship, providing new w ays to think about hum an relationships. The felt presence o f a “ non -self” enhances, by opposition , a w om an ’s sense o f personhood, continuously affirm ing the integrity o f w hat once might have been problem atic. T hough exogen ou s to her self, the non-self is, and becom es increasingly, essential to the self’s com prehension (cf. Y oun g 1 9 7 5 :5 7 8 ). C o n ­ versely, how ever, a w om an ’s sense o f self p ro­ vides a negative ground by which to apprehend the param eters o f her spirits. T hese tw o aspects o f her being are m aintained in contra-position th roughout her life, neither reducing to its opp o site, each becom ing enriched in sym p a­ thy with the other, shifting, expan din g, or contracting a s their m utual situation changes over time. In this way p ossession enables a w om an to evolve, to recontextualize her experiences from a broadened perspective (cf. K egan 1982). Zairan posit alternative sets o f m oral d is­ crim inations that are realized and displayed through her body and oth ers’ during trance. As derailed below , the ob servation and enact­ m ent o f such episodes provides the possibility, by no m eans the assu ran ce, that the integrating self will be “ seized with a con trary or critical perception ” (Rurridge 197 9 :2 8 ) and em pow ­ ered to alter her condition in g, to transcend the categories that have con strain ed her, to recog­ nize rhem for w hat they are - cultural con ­ structs, not im m utable truths. The p arad o x o f H ofriyati possession is that it defends the person while also enabling the self: it is at once a self-enhancing and self-m aintaining condition.

Trance It is through possession trance th at the exis­ tence o f her non-self becom es subjectively real to the possessed, or “ in trojected” (cf. C rapan zano 1 9 7 7 :1 3 ), and rhe culturally overdeter­ m ined self m ay be felicitously repositioned,

379

perhaps transcended. T he experience o f trance and its observation in others is the locus o f po ssessio n ’s creativity, for in trance a w om an becom es, legitim ately if tem porarily, a nonH ofriyati. In doin g so she is indirectly cau ­ tioned that she and the sym bolic constructs that define her sex are separate and distinct, how ever much the latter inform her im age o f self. T rance provides her the possibility o f insight, o f m aturing, yet it does so obliquely, in a w ay that does not dem and she take respon ­ sibility for her conclusions. T o begin with the notion o f introjection, or the patien t’s subjective realization o f a spirit’s attachm ent and influence, the self o f the p o s­ sessed is not merely absent or repressed during trance in deference to that o f the spirit, but, accordin g to H ofriyati, actively engaged. T hough she m ay be unaw are o f w hat her zar is doin g while m anifest to others, she is, villag­ ers say, still aw are, for when she and her spirit coalesce in her body they exchange experien­ tial dom ain s. With this the possessed tran ­ scends the visible w orld and “ sees through the eyes o f the sp irit” into the norm ally invisible parallel universe. T h u s she has, not a m ystical experience, but an eminently social one. If p o s­ sessed by a European spirit she experiences as a “ E u ro pean ” w ould for the duration o f its chant. M y inform ants tell me they see and interact with other Europeans in European w ays and perceive them selves to be surrounded by the trapp in gs o f European culture.3 For that brief period, they say, a w om an forgets who she is, her village, and fam ily; she “ know s nothing from her life.” In having such a vision o r, to use villagers’ description, in briefly step­ ping outside the H ofriyati w orld and into another, a w om an also briefly divests her self o f its person h ood, o f its norm ative contents and constraints. In proportion to her subjec­ tive experience o f otherness, her everyday reality is m ade to appear as one o f m any - less n aturalized, less unquestionable, indeed, less subjectively real. The experience o f trance is a lso, o f co urse, a cultural one. Spirits are rec­ ognized entities; their social m ilieus are know n, if incom pletely understood. Yet despite this cultural patterning, or, in fact, because o f it, trance is a lim inal excursion. Ry the p o ssesse d ’s ow n adm ission the experience is one o f tem ­

380

JA N IC E B O D D Y

porary isolation , o f alienation from her H o fri­ yati w orld. T here is, I think, a subtle difference between this situation and w hat K apferer (1 9 8 3 , 1 9 86) suggests takes place during the com p lex Sin h a­ lese exorcism cerem ony, which it is instructive to explore. T here the dem onic victim com es to the ritual already in “ an existential state of solitude in the w o rld ” (1 9 8 6 :1 8 5 ) and it is the purpose of the cerem ony to reintegrate her with society. T h is is accom plished in the stru c­ ture o f the perform ance where “ the culturally understood subjective w orld o f the patient finds external fo rm ” (1 9 8 6 :1 9 9 ); the dem onic, in all its ch aos an d terror, becom es tem porarily m anifest and dom in an t in the hum an w orld. H aving draw n the victim ’ s fam ily and friends - non-possessed participants - to experience w hat is construed to be the subjective state o f the p ossessed , an d thus linked their percep­ tions, the ritual then proceeds, via a su bse­ quent com edic ep isod e, to reassert cultural order and bring both patient and audience back into the w orld o f shared understandings (1 9 8 6 :2 0 1 ). In H ofriyat it is not so m uch that shared understandings are precipitately underm ined by a w om an ’s un tow ard experience, but that the rigidity o f these understandings an d her em otional identification with them prevents them from being underm ined, prevents her from being able to appreciate the distinction between ideals and the exigencies o f concrete situations. If anything, she is to o firmly grounded in the social w orld. In the villagers’ view, spirits are attem pting to subvert the order o f that w orld but she resists their influ­ ence - as she is exorted by Islam to do. Yet as curers and ad ep ts rightly observe, it is only when she lets them in, when she loosens her hold on her reality and enters an “ existential state o f so litu d e,” that she can start to recover. R ecovery entails rhe experience and o b se r­ vation o f trance in others. Like other such ritual m om ents, the occasion is rife with am b i­ guities and potential am bivalences.4 T he q u ali­ ties o f spirits when ju x tap o sed to those o f hum ans give play to the im agination. It is during such episodes th at, as Burridge argu es, the w ould-be individual “ perceives a hidden

m essage and accepts the invitation to e x p lo re ” (1 9 7 9 :1 4 5 ). Possession trance encourages reflection, a lim ited dism an tling o f the takenfor-granted w orld, enabling the p ossessed , in its afterm ath, to see her life in a very different light. T o take a m undane exam ple: an orange or a piece o f bread, when eaten with knife and fork by a W esterner zar during trance (which norm ally H ofriyati w ould not d o ), becom es som ething other than villagers’ fo od , or a m etaphor for H ofriyati gender dialectics: interior/exterior, fluid/substance, and so on. Its “ n atural” a sso c ia tio n s are stripped aw ay, deconstructed. A m ore protracted exam ple further illus­ trates my point: it relates to the H ofriyati w edding. T he clim ax o f that cerem ony com es near daw n o f the third day when the virgin bride is led out o f seclusion. A bridal shaw l o f red and gold silk, used also to cover girls at their circum cisions and w om en during child­ birth, is draped over her head, concealing all o f her body but her legs. She is positioned on a red m at in the center o f the co u rty ard, where she stands, barefoo t and im m obile, until her husband steps on to the m at and rem oves the shaw l. N ow unveiled, she is seen in all her finery and her fam ily’s gold, elaborately hennaed hands covering her face in a gesture o f timidity. G ently the groom releases her arm s and she begins the exactin g bridal dance - eyes tightly shut, arm s extended to the sides, back arched, feet m oving in m incing rhythm ic steps that barely leave the m at. T ow ard the end of each son g she b reaks o ff her dance and shyly recovers her face, then recom m ences with the gro o m ’s sign al, a s before, repeating the sequence until she has had enough and her kinsw om en lead her aw ay. At no time ought the bride to have seen her husband or the gathering for w hom her dance w as the focus o f rapt attention and long anticipation. The w edding dance is a poetic, crystalline d em o n ­ stration o f fem ininity in H ofriyat. By co n trast, in the zar realm , Luliya is a fem ale Ethiopian prostitute spirit which dem an ds that its hum an host ob tain Sudanese w edding incense, jew elry, and a bridal shawl for its use during p ossession cerem onies. When Luliya ap p ears during a ritual, a bridal m at is spread and the spirit, in the body o f its host,

SPIRITS A N D S ELV ES IN N O RTH ERN SUDAN

dan ces a s a H ofriyati bride. When the silken veil is rem oved L u liya’s h ost’s hands cover her face; when these are pulled aw ay she starts to dance with eyes closed, though in less inhibited m anner than the bride an d with obvious pretense at shyness. W hat is happening here? On one level, a w an ton , uncircum cised, nom inally Christian alien presum es to dance as a ch aste, circum ­ cised, M uslim village w om an. In the attem pt the spirit tries to sup p ress its libertine d isp o si­ tion but overcom pen sates, exaggeratin g the controlled steps o f a bride to the point where sim ulated H ofriyati d ram a becom es a spirit farce. Luliya is not by nature bash ful; its tim id­ ity m ust be feigned. T he sp irit’s real p erson al­ ity show s through the facade it erects with the aid o f its h ost, illum inating the enacted H o fri­ yati beh aviors again st a backgroun d o f patently non-H ofriyati traits. But this is not all. W hat the audience actu­ ally ob serves is a norm ally restrained, circum ­ cised H ofriyati w om an in the role o f a w anton, uncircum cised alien w ho in turn “ p lay s” a village w om an w ho is the epitom e o f restraint an d self-control. In lookin g at the “ oth er,” H ofriyati see the other lookin g at them , while in lookin g at the w om an entranced, they see them selves look in g at the other lookin g at them . The m ultiple reflection is dram atically sustain ed . . . then suddenly shatters a s Luliya peeks furtively over the hands o f its host, giving itself aw ay to the u p ro ario u s laughter o f its hum an audience. These an d other densely convoluted epi­ sodes constitute m ore than a com ic discourse on the am biguities o f gender and sexuality, fo r, a s such, they raise them as issues in them ­ selves and point to the som ew h at subversive ob servation (in H ofriyat) th at gender is not a n atural attribute but a cultural con struct and th u s, perh aps, m odifiable. H ence, categories th at are largely unquestioned in the course o f daily life becom e problem atic in the zar. T hrough the accum ulation o f such episodes trance a ffo rd s the participant an opportunity to m ature: to grow , a s T urn er say s, through antistructure (1 9 8 2 :1 1 4 ), to g rasp not only her con text, but the con text o f her con text (cf. Bateson 1972). Possession is as m uch an a e s­ thetic, a m eans to perceive new and rew arding

381

or possibly distu rbin g significances in what w as form erly taken for gran ted, a s it is therapy, a m ean s to correct faulty perceptions, to cure. Y et m any w ho consider p ossession trance and ritual legitim ate psychotherapy nonethe­ less disclaim their capacity to prom ote insight­ ful reflection am o n g the possessed (Bourguignon 1979: 2 9 0 - 9 1 ; Kennedy 1967: 191; Prince 1964: 115). Such “ fo lk ” therapies are gener­ ally considered to be effective in repatterning idiosyncratic conflicts and defenses in cultur­ ally ap p ro p riate w ays, and furnishing a correc­ tive em otion al experience, the sanctioned release o f negative affect (Bourguignon 1979: 2 7 4 ; D evereaux 1980: 1 7 -1 8 ; Kennedy 1967: 189; Kleinm an 1 9 8 0 :1 6 9 -7 0 ). Here the patien t’s condition m ay be rem edied, not cured, though her acknow ledged vulnerability to relapse m ay be m itigated if she is in corpo­ rated into a cult providin g gro u p support for a healthful reorientation (Bourguignon 1 9 7 9 :2 9 1 ; Kennedy 1967: 1 9 1 -2 ; Lew is 1971: passim ; M essin g 1958: 1 125). Yet, despite the success o f “ fo lk ” psych otherapies in securing sym ptom rem ission, and regardless o f how culturally ap p rop riate such techniques m ay be, they are often dism issed as inadequate when co m pared to W estern psych oan alysis (Kiev 1964; for exam ple, Derret 1979: 291 and O ztu rk 1964: 361). They are judged deficient because apparen tly unable to provide the patient an opportunity for m ature reflection, which constitutes the b asis for a psychiatric cure (D evereaux 1980: 1 7 -1 8 ). N o n e o f these view s d oes justice to the rich­ ness o f the possession experience. In H ofriyat the co n text o f p ossession carries within it the potential for insightful self-exam ination, how ever differently conceived from th at o f W estern psych oan alysis. For w hat constitutes insight into the self is surely described by the cultural construction o f an in dividual’s self­ h ood. In cases o f neurosis in W estern cultures, psychotherapy provides a con text in which rhe patient can learn to objectivate him self through co n versation , to gain distance from an e x a g ­ gerated “ I.” In Flofriyat, as we have seen, p o s­ session trance provides a con text in which the patient is encouraged to achieve distance from her cultural con text, the source o f her over­

382

JAN ICE B O D D Y

objectification. Both therapies aim at replen­ ishing the culturally specific constitution o f the self by exp lorin g and transcending form er pitfalls: a s individuals acquire insight into the p rocess o f self-construction, healthy, more ap p ro p riate disposition s o f selfh ood are suggested . In the zar , dialogue takes place between a w om an an d her spirit(s) - her non-self - inter­ nally, through visions and dream s, and exter­ nally, through the reports o f fellow villagers about her sp irit’s actions during trance (cf. Lam bek 1980). T hrough such oblique d is­ course the p ossessed m ight w ork through her problem s to achieve a greater understanding o f herself and her society (C rap an zan o 1977: 26). She is now given occasion to achieve a degree o f detachm ent from the gender con ­ structs th at have so com pletely shaped her being, thus to establish a basis for rhe n egotia­ tion o f her subordin ation . Possession , like an th rop ology, is a reflexive discourse: through it H ofriyati w om en can step outside their everyday w orld and gain perspective on their lives. T rance is a significant factor in this p rocess, whether experienced or observed. Ir h as been defined a s a tem porary, subjectively felt change in an in dividual’s reality orientation acco m p a ­ nied by a fadin g to abeyance o f reflective, criti­ cal aw aren ess (D eikm an 1969: 4 5 ; Ludw ig 1968; Shor 1969: 2 4 6 ; Van der W alde 1968). If we accept this view, the w om an w ho sees into the spirit w orld during trance is in a state o f heightened receptivity; she becom es, like the N dem bu initiand, a virtual tab ula rasa (Turner 1969: 103). If her trance is deep enough and involving enough, she is thus presented with pure experience, vivid, unedited, em otionally real, and not just once, but several tim es, as throughout the ritual she takes on a sequence o f other selves. It m ay not be during trance that she deepens her understanding o f herself but afterw ard s, in rem em bering her trance experiences (cf. K apferer 1986: 198) a s she is expected to d o. Such insights a s are gleaned com e indirectly, through w itnessing several dim ensions o f w hat her self is not. Yet none o f this is certain. The therapeutic efficacy o f p ossession trance resists objective m easure­ m ent, as d oes any aesthetic experience. For

som e w ho say they see things differently, this may signal a real change in ou tloo k , som atic disp osition , and em otional balance. For others, it m ay not. When d iscu ssin g the therapeutics o f p o sses­ sion trance, its potential effect on an unen­ tranced audience is rarely considered. But possession trance is only part experience; it is a lso part perform ance (Leiris 1958). When the spirits m anifest them selves in the bodies o f their hosts a catalo g o f otherness com es to life. The spirits behave in w ays ap p rop riate to their respective ethnic g ro u p s, social roles, religions, and sex. They m ay be w anton and undignified, take on superior airs, beg piteously, dance a b o u t w ildly, speak in brash or coy tones, exhibit any con duct fitting to their type. Spirits may be kings or slaves, prostitutes, nuns, male h om osexu als, m erchants, C optic priests, or fierce tribal w arriors. When a spirit’s chant is sung and all w hom it p ossesses ideally enter trance, it m anifests itself in each sim ultane­ ously, som etim es presenting different aspects o f its character, but alw ays interacting with the audience in strange and som etim es terrify­ ing w ays. T o observe possession trance in another is to w itness a p a rad o x : w om an w ho is not w ho she is - not hum an, not H ofriyati, not even, in m ost cases, fem ale. A lthough the identities o f the p ossessed and her intrusive zar are distinct, and it is the aim o f the cerem ony to cultivate aw aren ess o f their distinction in the possessed , for observers this separation o f entities is not alw ays easy to m aintain. D uring trance the tw o are brought into intim ate and often per­ plexing asso ciatio n , and those describing the episode often refer to the w om an an d her spirit interchangeably. Yet this risk o f confusion , this am biguity, is, I think, key to the aesthetics and therapeutics o f zar in Flofriyat. Ju st as when one sees a play, the interpretation o f a trance event is never w holly given in the event but m ust, in part, be constructed anew by each observer, w ho brings to the m om ent her own past experiences, present concerns, and critical aw aren ess. But unlike the audience at a play, the H ofriyati observer o f possession trance is utterly com m itted to the literal reality o f w hat she sees. And w hat she sees is som eone at once essential to her ow n construction o f self and a

SPIRITS AN D SELV ES IN N ORTHERN SUDAN

383

sym bol o f it, w ho is also her ow n sheer antith­ esis. This thorough p a ra d o x , taken with the various properties o f the entranced and her spirit, their individual traits and biographies, and the relations o f p arod y , travesty, and inversion am on g elem ents o f the episode, m akes possible any num ber o f interpretations, destructuring naturalized associatio n s and tem porarily freeing ideation from its m oorings in the everyday w orld. In the course o f a cere­ m ony the possessed alternately observes and experiences trance; thus, for an entire evening she is given to see herself an d those around her as in a hall o f m irrors, the prop o rtio n s o f her selfhood shifting from m om ent to m om ent, context to context, now fam iliar, now alien,

now frightening, now bizarre. In the course o f her long associatio n with the zar there are m any such occasion s, each affo rd in g her the possibility o f new insights, refined understand­ ings, and continued grow th. Z ar (as both possession and perform ance) is a pow erful m edium for unchaining thought from the fetters o f hegem onic cultural co n ­ structs and, to p arap h rase R icoeur (1 9 7 6 ), for opening it up in different and possibly illum i­ n ating directions. In the p ossession context, o f which trance is an integral part, the self becom es a pure issue, a subject for con tem pla­ tion, negotiation, an d, perh aps, felicitous regeneration.

NOTES

Research, and Prospects for Change. Studies in Family Planning 11 (1): 3 -1 6 . Bastide, R oger. 1958. Le Candom ble de Bahia. Paris: M outon . Bateson, G regory. 1972. Steps to an Ecology o f Mind, N ew Y ork: Ballantine Books. Berger, Peter L , and T h o m as Luckm ann. 1967. The Social Construction o f Reality. G arden C ity, N Y : D oubleday. Boddy, Jan ice. 1982. W om b a s O asis: The Sym bolic C o n text o f Pharaonic C ircum ci­ sion in R ural N orthern Sudan. American Ethnologist 9(4): 6 8 2 -9 8 . ------. 1989. Wombs and Alien Spirits: Women , Men , and the Z a r Cult in Northern Sudan. M ad iso n , WI: University o f W isconsin Press. Bourdieu, Pierre. 1977. Outline o f a Theory o f Practice. Richard N ice, trans. London: C am bridge University Press. Bourguignon, Erika. 1976. Possession. San Francisco: C handler and Sharp. ------. 1979. Psychological Anthropology. N ew Y ork: H olt, R inehart and W inston. Burridge, Kenelm. 1979. Som eone , N o One: An Essay on Individuality. Princeton, N J: Princeton University Press. C arrithers, M ichael, Steven C ollin s, and Steven Lukes, eds. 1985. The Category o f the Person. C am bridge: C am bridge University Press. C onstan tin ides, Pam ela. 1977. “ Ill at Ease and Sick at H e a rt.” Sym bolic Behavior in a

1

2

3

4

In 1977: N = 129, 54 p ossessed and 75 n on-possessed; in 1984: N = 135, 63 p o s­ sessed and 72 non-possessed. . . . An irony here is that fem ale circum ci­ sion has a deleterious effect on fertility from a physiological stan dpoin t. See the Sudan Fertility Survey 1979 (D em ocratic R epublic o f Sudan 1 9 82:56). As W allace (1 9 5 9 ) has so ably dem on­ strated, how one experiences an altered state o f con sciousn ess is strongly influenced by cultural m eanings and expectations. Spirits are am bivalent figures, represent­ ing exaggeration s o f village values (for exam ple, piety, dignity, or fertility), while a lso dem onstrating their op p o sites: exogam ous m arriage or p rom iscu ou s sexuality. Z airan in their biograph ies are liminal: they “ m ediate between alternative and op p o sin g con texts and are thus im portant in bringing ab o u t their tran sfo rm atio n ” (Turner 198 2 : 1 13). In the case o f zar, tran sform ation takes place in the co n ­ sciousness o f the possessed.

REFERENCES CITED A ssa ad , M arie Bassili. 1980. Fem ale C ircum ci­ sion in Egypt. Social Im plications, Current

384

JA N IC E B O D D Y

Sudanese H ealin g C ult. In Symbols and Sen­

timents: C ross Cultural Studies in Sym bol­ ism. I. M . Lew is, ed. pp. 6 1 —83. N ew Y ork: A cadem ic Press. C rap an zan o , Vincent. 1973. The H am adsha:

A Study in M oroccan Ethnopsychiatry. Berkeley: University o f C aliforn ia Press. ------. 1977. Introduction. In Case Studies in Spirit Possession. Vincent C ra p a n za n o and Vivian G arrison , eds. pp. 1 -3 9 . N ew York: Jo h n Wiley. ----- . 1980 Tuhami: Portrait o f a M oroccan. C h icago: University o f C h icag o Press. D eikm an, A rthur J. 1969. D eautom atization and the M ystic Experience. In Altered States o f Consciousness. C h arles T . T art, ed. pp. 2 5 - 4 6 . N ew Y ork: A nchor Books. D em ocratic R epublic o f Sudan , M inistry o f N atio n al Planning. 1982. The Sudan Fertil­ ity Survey, 1979. Vol. 1. K h artoum : D ep art­ m ent o f Statistics. D erret, J. D uncan. 1979. Spirit Possession and the G erasen e D em oniac. Man 14(2): 2 8 6 -9 3 . D evereaux, G eorge. 1980. Basic Problems o f Ethnopsychiatry. Basia M iller G u lati and G eorge D evereaux, trans. C h icago: Univer­ sity o f C h icago Press. D o u glas, M ary. 1966. Purity and Danger: An

Analysis o f Concepts o f Pollution and Taboo. London: R outledge & K egan Paul. ------. 1973. N atural Sym bols . H arm on dsw orth: Penguin. Elster, Jo n . 1986. Introduction. In The M ulti­ ple Self. Jo n Elster, ed. London: C am bridge University Press. Fans, Jam es. 1972. Cat Harbour. N ew fo u n d ­ land Social and Econom ic Studies N o . 3. St. Jo h n ’s: Institute o f Social and Econom ic R esearch, M em orial University of N ew foun dlan d. Firth, R aym ond. 1967. Tikopia Ritual and Belief. Boston: Beacon. Foucault, M ichel. 1980. Power/Knowledge. C olin G o rd on , ed. C olin G o rd on et al., trans. N ew Y ork: Pantheon. G eertz, C lifford. 1983. L ocal Knowledge:

Further F.ssays in Interpretive Anthropol­ ogy. N ew Y ork: Basic Books. H ale, Son dra. 1985. W om en, W ork, and Islam : Sudanese W omen in C risis. Paper

presented at the Annual M eeting o f the Am erican A n th ropological A ssociation , 3 -8 Decem ber. W ash ington, D C . Isbell, Billie-Jean. 1978. T o Defend Ourselves: Ecology and Ritual in an Andean Village . Prospect H eights, II.: W aveland. K apferer, Bruce. 1983. A Celebration o f

Demons: Exorcism and the Aesthetics o f Healing in Sri Lanka. Bloom ington: Indiana University Press. ------. 1986. Perform ance and the Structure o f M eaning and Experience. In The Anthropology o f Experience. V. W. T urner and E. M . Bruner, eds. pp. 1 8 8 -2 0 6 . C h icago: Univer­ sity o f Illinois Press. K egan, R obert. 1982. The Evolving Self: Problem and Process in Human Develop ment. C am b rid ge, M A : H arvard University Press. Kehoe, Alice B., and D ody H . Gilletti. 1981. W om en’s Preponderance in Possession C ults: The Calcium -D eficiency H ypothesis Extended. American Anthropologist 83(3): 5 4 9 -6 1 . Kennedy, Jo h n G. 1967. N u b ian Z ar C erem o ­ nies as Psychotherapy. Human O rganiza­ tion 2 6 : 1 8 5 -9 4 . Kennedy, Jo h n G ., ed. 1978. Nubian Ceremo­

nial Life: Studies in Islamic Syncretism and Culture Change. L o s Angeles: University o f C alifornia Press. Kiev, Ari. 1964. T he Study o f Folk Psychiatry. In M agic , Faith , and Healing. Ari Kiev, ed. pp. 3 -3 5 . N ew Y ork: Free Press. Kleinm an, A rthur. 1980. Patients and Healers in the Context o f Culture. Berkeley: U niver­ sity o f C alifo rn ia Press. La Barre, W eston. 1975. A nthropological Per­ spectives on H allucination and H allucin o­ gens. In H allucination: Behavior , Experience^ and Theory. R. K. Siegel and L. J. W est, eds. N ew Y ork: Jo h n Wiley. La Fontaine, J. S. 1985. Person and Individual: Som e A n th ropological Reflections. In The Category o f the Person. M . C arrith ers, S. C ollins, and S. Lukes, eds. C am bridge: C am bridge University Press. Lam bek, M ichael. 1980. Spirits and Spouses: Possession a s a System o f C om m unication am on g the M alag a sy Speakers o f M ayotte. American Flthnologist 7(2): 3 1 8 -3 1 .

SPIRITS A N D SELV ES IN N O RTH ERN SUDAN

------. 1981. Human Spirits: A Cultural Account o f Trance in Mayotte. N ew Y ork: C a m ­ bridge University Press. ------. n.d. From D isease to Discourse:

Rem arks Trance.

on

the

Conceptualization

of

Leiris, M ichel. 1958. La p ossession et ses aspects th eatraux chez les Ethiopiens de G o n d ar. L’homme: Cahiers d'ethnologic , de geographic et de linguistique. Paris: Plon. Lew is, 1. M . 1971. Ecstatic Religion: An

Anthropological Study o f Possession and Shamanism. H arm on d sw orth : Penguin. ------. 1986 Religion in Context: Cults and Charisma. C am bridge: C am brid ge Univer­ sity Press. Ludw ig, A rnold M . 1968. Altered States o f C on sciou sn ess. In Trance and Possession States. R aym ond Prince, ed. pp. 6 9 -9 5 . M on treal: R. M . Bucke M em orial Society. M essin g, Sim on. 1958. G ro u p T herapy and Social Statu s in the Z ar C ult o f Ethiopia. American Anthropologist 60: 1 1 2 2 -6 . M etrau x, A lfred. 1959. Voodoo in Haiti. N ew Y ork: O x fo rd University Press. N elso n , Cynthia. 1971. Self, Spirit Possession, and W orld View: An Illustration from Egypt. International Jo u rn al o f Psychiatry 17: 1 9 4 -2 0 9 . O beyesekere, G an an ath . 1970. The Idiom o f P ossession. Social Science and Medicine 4: 9 7 -1 1 1 . ------. 1981. M edusa's Hair. C h icago: Univer­ sity o f C h icago Press. O ztu rk , O rhan M . 196 4 . Folk T reatm ent of M ental Illness in T urkey. In M agic, Faith , and Healing . Ari Kiev, ed. pp. 3 4 3 - 6 3 . N ew York: Free Press.

385

Prince, R aym ond. 1964. Indigenous Y oruba Psychiatry. In M agic , Faith , an d H ealing . Ari Kiev, ed. pp. 8 4 -1 2 0 . N ew Y ork: Free Press. R icoeur, Paul. 1976. Interpretation Theory: D iscourse an d the Surplus o f Meaning. Fort W orth: T e x a s C h ristian University Press. Sau n ders, Lucy W ood. 1977. V arian ts in Z ar Experience in an Egyptian V illage. In Case Studies in Spirit Possession. Vincent Crap an zan o and Vivian G arriso n , eds. pp. 1 7 7 9 1 . N ew Y ork: Jo h n Wiley. Sh or, R on ald E. 1969. H ypnosis and the C on cept o f G eneralized Reality O rientation. In Altered States o f Consciousness. C h arles T . T art, ed. pp. 2 3 9 - 6 7 . N ew Y ork: Anchor B ooks. T urner, V ictor. 1969. The Ritual Process. C h icag o : Aldine. ------. 1982. From Ritual to Theatre: The H um an Seriousness o f Play. N ew Y ork: Per­ form ing A rts Jo u rn al Publications. Van der W alde, Peter H . 1968. T ran ce States an d F-go Psychology. In Trance and Posses­ sion States. R aym ond Prince, ed. pp. 5 7 - 6 8 . M on treal: R. M . Bucke M em orial Society. W allace, Anthony. 1959. C ultural D eterm i­ n ants o f R esponse to H allucinatory E xperi­ ences. A.M. A. Archives o f General Psychiatry 1 :5 8 -6 9 . W ard, Cohleen. 1982. A T ran scultu ral Per­ spective on W om en and M ad n ess: T he C ase o f the M ystical Affliction. Women's Interna­ tional Jo u rn al Forum 5(5): 4 1 1 - 1 8 . Y o u n g, A llan. 19 7 5 . Why A m hara Get K ureynya: Sickness and Possession in an Ethiopian Z ar Cult. American Ethnologist 2(3): 5 6 7 -8 4 .

31

The Poetics of Time in Mayan Divination Dennis Tedlock

Dennis Tedlock is professor of English and anthropology at the State U niver­ sity of New York at Buffalo, and is renowned for his translations of Zuni and Mayan religious texts and discus­ sions of issues of translation and tran ­ scription (1983). Tedlock's essay illustrates a double turn to language. Double because it is attentive both to the lan­ guage of religion - as it is expressed in such forms as prayer, sermons, divina­ tion, and the marvelous Mayan calendricat system - and attentive to the language of anthropologists w ho make a sustained effort to evoke the richness, subtlety, and complexity of religious worlds and the wholeness and polyva­ lence of experience. W e see here w hat a difference attention to under­ standing and conveying speech forms can make. Tedlock also provides the portrait of

a religious practitioner at w ork. He describes a real event at which he was a central participant, ostensibly an appren­ tice being tested to perform as a diviner. The result is a beautiful evocation of a perform ance of divination, including a twist in the tail. Tedlock's essay thus also illustrates the approach th at draws on anthropologists' own mimetic ex­ perience to understand that of their subjects. Too often naively ethnocentric, occasionally the anthropologist manages to pull it off. Here Tedlock's authority as diviner is evoked through the tw itching of his own body. The essay also speaks indirectly to issues of sorcery and the allocation of responsibility. For further work on the M ayan calendar and daykeepers see Colby and Colby (1981), B. Tedlock (1992), and W right (1990); on bodily experience see Csordas (ed. 1994).

From Dennis Tedlock, “ The Poetics o f Time in M ayan Divination,” in John Leavitt, ed., Poetry and Prophecy: The Anthropology o f Inspiration (Ann Arbor: University of M ichigan Press, 1997), pp. 77 -9 2 .

TH E PO ETICS OF TIM E IN M A Y A N DIVIN ATION

The successive impressions o f discourse, which strike a redoubled blow , produce a different feeling from that o f the continuous presence o f the same object, which can be taken in at a single glance. J

f a n -J a c q u e s

R o v ssfa o

M ay an s seem devoted to poetry in the sense G asto n Bachelard gave that term , which is to say that they have long been skilled at w orking again st “ the sim ple continuity o f sequential tim e,” having relatively little interest in w hat he called “ time as duration that shatters ech oes” (1988: 173). It is a s if they w ould not allow time to m ove forw ard or backw ard w ithout first listening for reverberations, and when it does m ove on they listen for the rhythms o f its duration. A m usical m etaphor is an apt one, and so is the m etaphor o f a brocaded textile that com plicates large p at­ terns with variable details, the kind o f textile the Q uiche M ay a produce in G u atem ala (Tedlock and T edlock 1985). N either o f these m etaph ors describes a circle, and they give us a w ay o f thinking our w ay out o f the naive dichotom y between lineal tim e, o f which we in the W est im agine ourselves to be the m asters, and cyclical rime, which keeps the Other outside o f history a s we define it - never mind the “ w heels o f p ro g ress.” When M ayan n arrators run through a wellplotted action sequence, their language m oves forw ard with very little reverberation, and indeed their stories are very short and fast m oving when com pared with those told by the indigenous peoples o f the A m azon, for exam ple, or those o f southw estern or north­ w estern N orth A m erica. But when M ayan s converse, pray, or m ake speeches, they tune in on tim e’s echoes again and again . Their prin­ cipal m eans is verse, parallel verse, in which the patterns o f soun d, syn tax, and m eaning w ork together to produce a paradigm atic pattern o f sam eness and difference, usually proceeding by pairs o f ph rases but som etim es by a grou p o f three or four (D. T edlock 1983: ch ap. 8). Verse p assages within a longer d is­ course are usually introduced by w ords or ph rases that stand alone, as if to locate small-

387

scale rhythm s within a larger system o f widely spaced singular lines, im plicitly parallel, that m ay or m ay nor becom e m anifest within a particular utterance. In the case o f the Quiche M ay a , these verse patterns arc canonical, having been established, in the ancient book known as the Popol Vuh, by the g od s them ­ selves (IX T edlock 1985: 78). T he anim als o f the present-day w orld were dem oted to their present statu s when they failed to learn the language lesson the g od s gave them , a lesson that places a value not on the com pletion o f a sentence but on the com pletion o f verse p at­ terns. H ad the anim als been able to learn, w hat the g o d s w ould have heard from them w as the opening o f a prayer:

Juraqan , ch'ipa kaqulja , raxa kaqulja , uk'ux kaj, uk'ux uleWy tz’aqol, b'itol, alom , k'ajolom One o f a Kind, Newborn Thunderbolt, Raw Thunderbolt, Heart o f Sky, Heart o f Earth, M aker, Modeler, Bearer, Begetter After a singular opening, nam ing the only god w ho is not a m em ber o f a pair o f g o d s, com e four parallel couplets, each o f them consisting o f a gram m atical paradigm in which one element is held con stan t while another is changed. In both the text and tran slation , the first pair o f couplets is constructed at the lexical level, while the second is constructed at the m orphological level. Shifting from the patterns o f the signs them selves to their rela­ tionship with their objects, the first tw o co u ­ plets are actually reverberations o f rhe opening singular line, giving tw o pairs o f epithets for the singular god. The second pair o f couplets shifts from the vertical lightning a x is already established to the horizontal surface o f the prim ordial sea, w hose g od s are som etim es called uk'ux cho, uk'ux plo “ H eart o f Lake, H eart o f S e a ” . These g od s have no definite num ber save that they are all in pairs, each consisting o f a preparer o f m aterials (M aker) and a shaper o f those m aterials (M odeler), and/or a fem ale (Bearer) and m ale (Begetter).

388

DENN IS T E D L O C K

Fhe point o f interest here is that the w ords o f the lan guage lesson d o not have an iso m o r­ phic, one-to-one relationship with a series o f individual g o d s. The singular thunderbolt god gets a total o f five nam es or epithets, while the other g od s outnum ber the w ords that happen to be add ressed ro them. Parallel verse has the effect o f constantly shattering the notion that there is an essential connection between any particular w ord or phrase and any p articular object in the w orld - a notion lon g d ear to the W estern m ind, if only as a lost or as yet u n at­ tained ideal. N o t even M ayan proper nam es can foster such an illusion, since there is scarcely a single god or hum an or place p o s­ sessed o f just one nam e. The language lesson could have given both the singular and dual g od s m ore nam es and epithets or fewer w ithout changing the basic rhythm s o f its soun d s and m eanings and w ithout any need for an a u x il­ iary lesson on the difference between objects and their representations. For the Q uiche, there is a level at which language d oes reflect the w orld, but this occurs at the level o f structure rather than thar o f w ords. A clear and accurate d iscourse, right with itself an d right with the w orld, is called lemo tzij , cholo tzij, ajilan tzij “ m irror w ords, ordered [lined up) w ords, counted w o rd s,” and there is textual evidence for a sim ilar con ­ ception am o n g the ancient low land M aya (D. T edlock 1988: 93). The lan guage lesson “ reflects” rhe g od s by m eans o f an iconiciry thar is m ore diagram m atic than im agistic, m atching their singularity and duality with sin gu lar and dual w ord s and ph rases and p o si­ tioning the singular g od with respect to the co sm o s and the others with respect to each other. A ctual m irrors a lso reflect relationships am o n g ob jects, but rhat is not the quality of m irrors that firsr occurs to the W estern mind. Bakhtin characterized poetry as m onological and prose (prose fiction in p articular) as dialogical (or polyphonic), but he d idn ’t think ab o u t the poetics o f parallelism . He w rites o f a force he calls “ d ialo g ism ,” w'hich operates even at the level o f individual w ords, since a given w ord exists in an environm ent o f other w ord s that could have been used with refer­ ence to the sam e object, w ord s that m ay com e to the m ind o f the hearer or reader (Bakhtin

1981: 3 2 6 - 3 0 , 4 2 6 ). W hat happens in parallel verse is th at one or m ore o f these other w ords is actually given voice. Once the dialogic is on the surface o f the poetry rather than hidden beneath it, the im plication that still other w ords could have been used becom es ob vious and verbatim reproduction from one occasion to another becom es a pointless essentialism . Som e o f the sam e ph rases m ay be carried over, whether from one public inscription to another in the ancient tem ples at Palenque or from one speech to another by a contem porary Q uiche prayer m aker, but even when the ob jects o f the signs rem ain the sam e we are unlikely to read or hear the exact reproduction o f an entire sentence. At larger scales, there is now here an attem pt to present or represent a com plete d is­ course to which nothing should or could be added. When M ay an s use poetry as a m eans o f elucidating events that are hidden from the present time o f sp eak ers and hearers, whether lost in the p ast or belonging to an uncertain future, they carry the dialogic beyond the ju x ­ taposition o f alternative w ords and phrases and let m ore than one person be heard - or at least w h at sou n d s like m ore than one person. When prophetic m an uscripts are given public readings in a Y ucatec village in Q uin tana R o o , tw o speakers take turns reading p assages, im provising e lab o ration s at the scale o f p a ral­ lel verse as they g o (Burns 1983: 2 2 -3 ). A m ong the Q uiche M a y a , it is said rhat when both m em bers o f a m arried couple are trained as diviners and they each seek an answ er to a client’s question at the sam e rime, their co m ­ bined pronouncem ents will shed m ore light than will those o f a diviner w orkin g alone. Solo divinatory perform ances are far m ore com m on, but even then the client hears ph rases rhat com e from m ore than one source, spoken by w hat soun ds like m ore than one voice. T he voices in which the diviner addresses the g od s and tran sm its their answ ers are poetic, highly versified, and thick with figures o f speech. In co n trast, the voice in which the diviner sp eaks directly to the client is relatively prosaic, taking shape a s sim ple question s at first and later offering interpretations o f divinatory poetry rhat fill g ap s in the client’s narrative under­ stan ding o f the p ast or future.

TH E PO ETICS OF TIM E IN M A Y A N D IVIN ATION

It is alo n g the path o f narrative that I will ap p ro ach Q uiche divination here, letting the poetry be heard where it brings the story to a stop. D espite the developm ent o f the eth n ogra­ phy o f speaking over the last twenty years, with all its attendant em p h asis on placing texts in their sociocultural co n texts, narrative acco u n ts o f particular speech events occurring a t particular tim es and places rem ain su rp ris­ ingly rare. The ethnography o f sp eakin g has in fact been w ritten, for the m ost part, a s an eth­ nology o f speakin g, which is to say that it rep ­ resents a p articular occurrence as if it typified an entire genre, recurring in a situation typical for that genre, and a s if no outside observer had been present when it occurred. It is not only historical time that d isap p ears in this process, but the echoes and reverberations that connect a particular event with others o f the sam e kind - or, by w ay o f a farther-reaching intertextualitv, with events o f different kin ds, reach ing even into the p ast experiences o f the fieldw orker. A narrative accoun t has room for a fuller an d richer tem porality, one in which particip an t observation is not only a p ast m eth­ od o lo gy for rhe gathering o f data for the w riting o f a present ethnology but continues to reverberate a s the w riter engages in “ the o b se r­ vation o f p articip atio n ” (B. T edlock 1991). O ne o f the earliest narrative eth nograph ies o f sp eakin g w as Z o ra N eale H u rston ’s Mules and Men (1 9 3 5 ), published long before a rg u ­ m ents o f the sort rehearsed here were form u­ lated. She w rote an accoun t o f fieldw ork am o n g Florida and Louisian a b lacks, with m ultiple instances o f a full range o f the verbal arts, including sayings, o rato ry , prayer, and son g lyrics, all o f them perform ed in the p res­ ence o f H urston herself. O ne o f her con tem ­ poraries a s a student o f Franz B oas and Ruth Benedict at C olu m b ia University w as Ruth M . U nderhill, w ho published accoun ts o f the verbal arts o f the P ap ago o f A rizona. U nder­ hill, like H urston , w as a skilled w riter, but in her Singing fo r Power (1 9 3 8 ) she proceeds genre by genre, describing w hat a P ap ago w ould typically sing when gath ering salt, hunting for deer, and on other generic o cca­ sions. The difference between the tw o w riters is only partly accoun table to the fact that H u rston w as a northern urban black returning

389

to her ow n southern rural roo ts, w hereas Underhill had to contend with rhe differences between urban white and rural Indian and between the F^nglish and P ap ago languages. H u rston becam e a w riter a s such, w hereas Underhill finished her d octorate in an th rop ol­ ogy an d pursued an academ ic career. It has taken us a long time to reach the point where an th rop ology , no longer denying itself the full resources o f w riting, m ight no longer be a career separate from that o f a w riter. In this new an th rop ology, the verbal arts o f the O th ers might at last play a full an d central role in eth nograph y in general rath er than being the ob ject o f an ancillary subcategory o f ethnography. Between H u rston ’s bo ok an d the 1980s there are very few narrative eth nograph ies o f sp eak ­ ing to be foun d, but G eorges C o n d o m in a s’s We H ave Eaten the Eorest (1 9 7 7 ), first published in French in 1957, provides an ou tstan din g exception. In this accoun t o f a M on tagn ard village in V ietnam , rhe eth n ograph ic exp eri­ ence, the representation o f village life, and the poetic voices o f the villagers are all w oven into a single narrative fabric. M ore recent b ook s o f this kind include W illiam K. Pow ers’s Yuwipi (1 9 8 2 ), an account o f a new religion am on g the O glala L a k o ta ; Steven Feld’ s Sound and Sentiment (1 9 8 2 ), on poetry and son g in the lives o f the Kaluli o f P apua N ew G uinea; Lila A bu -Lu gh od’s Veiled Sentiments (1 9 8 6 ), on the com position and recitation o f lyric poetry am o n g Bedouins in Fgypt; Yaqui Deer Songs/M aso Bw ikam , jointly authored by Larry F'vers and Felipe S. M olin a (1 9 8 7 ); and H aa Tuw unaagu Yis/Eor H ealing O ur Spirit , an account o f Tlingit o rato ry by N o ra M ark s D auenh auer and R ichard D auenh auer (1990). These sam e au th ors m ay, in varying degrees, attem pt such synchronic rasks a s delineating genres and subgenres or categorizin g tropes, but the poetry in their b o o k s rem ains con ­ nected with particular events an d their rever­ berations in the lives o f particular persons. Like som e o f the au th ors ju st m entioned, I have found m yself expected to learn, however im perfectly, som e o f the skills I w as observing. T h is happened when B arb ara T ed lock an d I reached a point in our question s ab o u t divinatory practices where the only w orkab le answ er

390

DENN IS TED LO C K

w as an offer to teach us those practices (see B. T ed lock [1 9 8 2 ) 1992: ch aps. 1,3). In the episode presented here, 1 found m yself called upon to put w hat I had learned into action and to d o so not for a layperson but for another diviner, far from any possibility o f prom pting from my teacher. T his happened w ithout any w arning and w ithout the m aking o f a tape recording, so that much o f w hat I have written is a reconstruction rather than a direct tra n s­ position o f field records. W hat I had to w ork from , in addition to my personal m em ory o f that occasion and parallel occasion s, w as a general know ledge o f divinatorv procedure, a sketchy n otebook entry that includes the date, and several p h otograp h s. The m ost useful im ages show the disposition o f my divinatorv paraph ernalia at tw o different m om ents, revealing the results o f the sortilege I per­ form ed. T his inform ation perm itted me to rerun the divination in the very p rocess o f w riting it. T he result is certainly not an exact representation o f w hat “ really h ap p en ed ” on a m ou ntain top near Santa Cruz Q uiche, G u a ­ tem ala, on the afternoon o f Eight Bird in the year Three T h ough t - or Sunday, D ecem ber 5, 1976. There m ay even be a flat-out error o f fact in the accoun t, though I have rem ained strictly factual abo u t every fact 1 im agine I still possess. M y m ain effort has been to w rite an accoun t w hose ow n internal tem porality retains not only the sequences but a lso the resonances o f w hat particular persons said and did on a p articular occasion - or w hat it seem s, from here, they m ust have done. As the scene open s, we are already on the m ountain top, in the com pan y o f a Q uiche ajq'ij, or “ daykeep er” (diviner), nam ed M ateo Uz A baj and tw o o f his children. Like the man w ho taught us daykeeping, D on M ate o is chucbkajaw, or “ m otherfath er,” the head of his patrilineage. The date W ajxaqib ’ T z ’ikin , or “ Eight B ird,” belongs to the 260-day M ayan calender that serves as the basis for divinatorv investigations o f tim e, consisting o f all the per­ m utations o f 13 day num bers and 2 0 day nam es. Eight is g oo d for visiting m ountaintop shrines, while Bird is good for prayers, esp e­ cially prayers for wealth in general and m oney in p articular. We have already finished setting out offerin gs and saying prayers, and w e have

already eaten the lunch we brought and sam pled the bootleg liquor we used to anoint the shrine. N ow we are sim ply relaxin g before the shrine, an o u td o o r hearth created by re ar­ ranging the rubble o f an ancient temple. The voice w ritten in italics is low and prayer­ ful, loud enough to be heard by the hum ans present but not addressed to them. Another and low er voice, in s m a l l c a p s , gives the divinatory m eanings o f the sm all m uscle sp asm s or tw itches that occur, unseen, in various p arts o f the diviner’s body. Phrases that are ovcrseored are unspoken but alw ays on the tip o f a divin­ er’s tongue, ready for use whenever needed. By this time the incense on the shrine was down to a mound o f charcoal, sending up traces o f white smoke. All that remained of the sticky tallow candles, given to the Holy World beneath our feet, w as a large spot of grease, embedded with curled black lines of charcoal that had once been wicks. The smooth white candles, the tall w ax tapers for the celestial gods, still had a long time to burn, but one of them was crooked now, bent by the heat while the incense w as ablaze. Still laid out on one of the flat slabs of basalt in front o f rhe hearth were our divining bundles, the large and lumpy one belonging to Don M ateo and the neat, fist-sized ones belonging to the ethnographers. Don M ateo looked at my bundle and then at me, and he said, “ 1 want to ask a question.” “ A question?” “ Yes, I want you to count days for m e.” “ But where? There’s no table here.” “ T hat’s all right. You can do it on that rock.” And he reached out to tap the largest of slabs in front o f the shrine, the one I was seated right next to. “ But there’s no cloth to spread on it.” Diviners always spread bright brocaded cotton cloth on their tables. “ That’s all right, just do it right on the stone.” Diviners cannot refuse to answer a ques­ tion, so long as rhe day on which they’re asked has been kept. Today was a good time to ask, since the four o f us with divining bundles wouldn’t have come up here on this mountain in the first place unless we’d been keeping this day, unless we’d all been planning to abstain from sex and from anger for the whole day.

TH E PO ETICS OF TIM E IN M A Y A N DIVIN ATION

I felt awkward divining for a man who had been a diviner for who knows how many years, and speaking my lines with an accent, but I had no way out. My bundle was on a rock beside Don M ateo, but I had to go and get it myself, since 110 one should handle a bundle that belongs to someone else. I sat back down on the ground, scrunching up as close as 1 could get to the large slab and setting my bundle down at the center of it, with the mouth, which was tied shut, pointing away from me. For a moment my fingers came in contact with the stone, and I could alm ost see the soft cloth I wished I had with me. Don M ateo moved closer for a better view, one knee on the ground and one folded up against his breast. With a table he would’ve been sitting across from me, but instead he w as to the left of my work surface, and when 1 w asn’t turning to address him 1 was facing the shrine. It w as as if the Holy World were asking the question, or Don M ateo and the Holy World were asking it between them. The motherfather who taught Barbara and myself how to divine often began by crossing himself and doing an Our Father and a Hail M ary and sometimes adding The Apostles' Creed, all in Spanish, before he prayed in Quiche and made his first move toward the bundle at the center o f the table. I skipped all o f this, which Don M ateo didn’t mind in the slightest, and started in on the Quiche prayer I’d memorized - yes, memorized - from a text our teacher had dictated to us. He had com ­ posed it as he dictated, but then we studied it the only way we knew how, which was to commit it to memory word for word, like memorizing verse in school. Except that we'd made som e small changes at the beginning, putting in a few words from the Popol Vuh. Pardon my trespass, so go the opening words, but then, instead o f saying T iox, which is to say Dios, which is to say G od, what I say is Heart o f Sky, and then pardon my trespass, Heart o f Earth , instead o f M undo, or World, and pardon my trespass, grandmotherfathers, all the human beings who once lived here on the face o f the earth, just as I w as taught, and I hold out my right hand with the cutting edge down, touching my bundle as if to divide into left and right halves, and then touch it again at right angles to make top and bottom halves,

3 91

the four quarters o f the World. I can feci the small, hard things inside the cloth bag as I do this.

Pm taking hold o f the yellow mixture, white mixture, and here I take up the bundle in my right hand, dark beans, light beans, dark seeds, light seeds, undo the bow in the thin string that keeps the bag shut, cloudy crystals, clear crystals, unwind the turns in the string, yellow necklace, white necklace, gently pour the red seeds and white crystals into a pile on the black basalt, before you, Skyearth , and 1 set the empty bag and its string to my left, near Don M ateo. While I mix and spread the pile, running the palm of my right hand over the seeds and crystals in circles, he finds a ten-cent piece in his pocket and places it on the bag, and the seeds and crystals click and clink as ! stroke them. Pm borrowing, just for a moment, your

breath, and also what we have as our moun­ tain , which is to say our body, and here I begin picking out the larger crystals from am ong the others, I'm asking for the blessing, the favor , place the largest crystal, not quite the length of a finger joint, on the far side of the pile, pointing away from me, this one is the mayor, the M am , Old Man who bears the years, for just this one clear light, flank the mayor with a pair o f aldermen, for the sake of my work, my burden, just one scribe, clearer than the other crystals, to the right of the right-hand alderman, and one treasurer, to the left, my mixing, my pointing, a pair of staff bearers flanking the left-hand alderman and the scribe, and another pair o f staff bearers flanking them, and one more to the left. With all ten presiding officials lined up in a tight row at the far side o f my work surface, I depart from my set piece to address the longest crystal, the M am , Come hither. Old Man Deer, Old Man Wind, the powerful lords o f the days that bring in dangerous years, and then the next two crystals, Old Man Tooth, Old Man Knowhow , the lesser lords o f the days that begin the calmer years, then Secre­ tary Net, lord o f the day of written records, Treasurer Bird , lord o f the day o f money, do me a favor, and then I pick up where I left off, and also , dark clouds, light clouds, the mist,

the cold, the wind, walking over mountains, over plains. I’m running my hand over the

392

DENNIS T E D LO C K

seeds and small crystals again, and also , the targe bills, little hills, large flats, little flats, large volcanoes, little volcanoes , large plains, little plains , come out and talk! They'll speak

for just a moment, for just this one light, for just this one question, in the chill, in the wind, and here I look at Don M ateo, they're speak­ ing on this holy and precious day , /// the chill, in the wind, they're speaking for a moment, the clicking and clinking stops, for just this one question, on the day Lord Eight Bird. “ What is your question?” I ask Don Mateo. “ It’s my pig. I want to know who killed my P»g-” We heard him praying about this pig just a little while ago, asking that the person who killed it be delivered into the hands o f the apostles and the Holy Quaternity, and to the mountains with gaping stones on top and the lightning archangels. “ What day did it die?” “ It was three days a g o .” “ Very well then, today is Eight Bird, so that would be, let’s see, Seven Jaguar, Six Cane, Five Tooth. It happened on Five T ooth ,” I conclude, and go back to running my hand over the seeds and crystals. Come hither. Lord Five Tooth , speak for a

moment, for }ust this one light, just this one question, for Mateo Uz Abaj , by name. Then I hesitate, wondering just how to frame this question. H e’s already certain someone killed his pig, rather than wondering whether the pig just died, but I start off with the question I’ve been taught to ask.

Does his pain have an owner? Perhaps some neighbor, some companion, piled up words behind his legs, behind his arms , and again the clicking and clicking, I'm taking hold o f the yellow mixture , white mixture, dark beans, light beansy the seeds and crystals that can move a question into a tinted light and then, for just a moment, into bright clarity, and also, the grandmothers, the grand­ fathers, they talk, they speak , for a moment, for a while, concerning this one question , and I stop for a moment. My eye is caught by the glint o f the crystal scribe, I’ve alm ost forgot­ ten to call the sheet lightning, the lightning that moves over distant horizons: I'm borrowing the yellow lightning, white

lightning that moves over the large lakes, the

small lakes , from the place where the sun comes out, to the place where the sun goes down, four-cornered sky, four-cornered earth , may the lightning move over the horizon of my body, may I catch its reflection in my blood. The hand stops moving. 1 raise it, cupped, to my mouth. We're doing this right now, into my hand I blow a short breath, pardon my trespass , 1 grab a fistful o f seeds and crystals, as many as 1 can tightly hold, and put them in a separate pile to the right. Then I push w hat’s left o f the original pile, the better part of it, off to the left, leaving a work space about the size o f a large open book. From the right-hand pile. I take the seeds and crystals, in whatever com ­ binations come to hand, and arrange them in clusters of four, making rows and files like those of a calendar, starting at upper left. If Don M ateo were doing this right now he’d be praying the whole time, but I’m just about out o f words. At the end of the third row, 1 have one seed left over, which counts as a cluster but means the answer to the question will not be as certain as it would’ve been with an even number. Leaning back a little, I look the whole arrangement over. The mixing is done and it’s time for the pointing. Come hither. Lord Ewe Tooth , do me a favor , and I’m ready to count the clusters, starting at the upper left and moving across the rows, pointing with my first two fingers held together, but before 1 can even get started I feel lightning in my blood, a twitch moving up just under the skin o f my left calf, and Don M ateo sees me withdraw my hand and glance down at the back o f my pant leg. “ ‘th e woman is c o m in g ,’ it say s,” I say in the low but urgent voice of the blood, to no one in particular, and then in my normal voice, to Don M ateo, “ There's a woman involved in this.” N o sooner do I turn back to the counting o f days than the lightning moves across the base o f the big roe on my right foot, and Don M ateo sees me glance at the toe of my shoe. “ ‘ he ’ s i i x ,’ it say s,” 1 say in the blood voice again, and then to Don M ateo, “ Some old man, or som e important man, is ill, or will be ill, hut I don’t know whether he’ll die,” and even as I speak the lightning darts into the crease behind my right knee, and this time

TH E PO ETICS O F TIM E IN M A Y A N D IVIN ATIO N

Don M ateo sees me move that leg a little, cocking my head to glance at the gathering of the pant leg behind my knee. “ ‘T h e to m b ,’, it sa y s,” I say, giving voice to my blood once again, and then, looking over at Don M ateo, “ He’s on his way to the grave, this old m an.” Again 1 look back to the rows of seeds, point at the upper left cluster, and this time the lightning holds back, the blood falls silent to hear the count. All the days come with auguries, but 1 speak only their names out loud:

Five Tooth, the good road, the had road, Six Cane in the house, Seven jaguar befo re the Holy—World, h'jght Bird praytttg , pbradtng, gold, stiver, Nine Sinner before G od, Ten Knowhow good thotrghts,— bad —thoughts, FJeven Knife lips that tremble, Twelve Rain filling the bowl, setting the table, Thirteen Marksntan - the lightning moves on the front o f my right thigh - not in the middle, but off to one side - and Don M ateo sees me glancing there. “ ‘ T h e s h r i n e , ’ it say s,” the blood says, a n d the count stops. To Don M ateo I say, “ It’s the shrine o f your lineage, something’s off center in your family. Or on your land. The table is not balanced on all four legs.” Then 1 look back at the cluster I w as pointing at when the lightning moved, repeating the name of the day, but this time 1 give voice to the augury that comes with it: “ This w as Thirteen M arksm an. ‘ Before the grandm otherfathers,’ it say s,” and 1 add, looking at Don M ateo, “ It’s got something to do with the ancestors, and its been a long time.” Ju st a trace o f displeasure crosses his face. He serves as the motherfather for his lineage, and it’s his responsibility to feed and pray to his ancestors for everyone else, but he says nothing. On with the count.

One Lefthanded going mad, Two Wind getting—angry. Three Foredawn opening, blaming, Four Net a burden, a debt. Five Snake the enemy, Six Death bring good,Seven Deer the motherfather, Eight Yellow ripening, Nine Thunder paying, suffering, Ten Dogprab ottsy, uncertainty, FJeven Monkey spinning thread , and here I reach the final, single seed, and announce the day in a louder voice, and the augury: “ Twelve Tooth. ‘The good road, the bad road ,’ it says. Things are on the w rong track

393

because o f what happened.” There’s more to this than just the incident of rhe pig. Don M ateo says nothing, and I continue with a second count clear through from the top, starting with the next day:

Thirteen Cane, One Jaguar, Two Bird, Three Sinner, Four Knowhow, Five Knife, Six Ram , Seven Marksman, Eight Lefthanded , the lightning moves under the edge of the heel of my right foot: “ ‘ h e ’ s b e e n s t e p p e d o n , ’ it say s,” and then I tell Don M ateo why he’s seen me glance at my shoe: “ Whoever did this deed, you’ve alm ost got him, youve just - ” the lightning interrupts me, ripples the blood at the edge of the palm o f my right hand: “ ‘h e ’s been g r a b b e d ,’ it says. He’s alm ost caught. Ju st this much is lacking,” and I show Don M ateo my thumb and forefinger just a finger’s breadth apart. Then I look at the cluster of seeds where I stopped counting: “ Eight Lefthanded. ‘He’s going m ad,’ it says. Whoever this is, he was crazy to kill your pig, bur it’s craziness that’s going to grab him .” And so on w'ith the days: Nine Wind, Ten Foredawn^ the blood moves on the back of my left thigh: “ ‘ i t a l r e a d y h a p p e n e d , ’ it says. The woman w ho’s in this, 1 don’t know, som e­ thing happened between her and the m an’s lineage, some time ago. So says the blood.” Don M ateo brightens, and I go on to take the day into account: “ Ten Eoredawn. ‘ Blaming,’ it says. W hat­ ever happened, the man blames som eone.” Don M ateo gives just a trace o f a nod. On with the count: Eleven Net , Twelve Snake , Thirteen Death,

One Deer, Two Yellow, Three Thunder, Four D og , Five Monkey, Six Tooth , the lightning m oves in my right testicle, I look straight down, and the only voice the blood has for this one is a chuckle. “ T h is w om an had con tact w ith a m an ,” I tell D on M ateo , and then I reach the single seed at the end: “ Seven Cane. ‘ In the house, in the home,’ it says. The man who killed the pig is in your own household. Or at least you're related to him in some w ay.” Don M ateo looks excited now:

394

DENNIS T E D LO C K

“ Raim undo Sucac! Ask if it was Raimundo Sucac Aw el!” That would be the father of his eldest son’s wife. M aybe that man didn’t really like giving his daughter in m arriage to Don M ateo’s son. But it’s not mine to ask the ques­ tioner why he wants to test this name. I make a single pile o f all the seeds and crystals again (except for the ones that serve as officials), mix again, grab a handful again, make clus­ ters of four items each, posing the new question:

Raimundo Sucac Awel: is he the owner of this, the pain o f my neighbor, my companion heref Don M ateo watches closely as I com ­ plete the last row, coming out to exactly four in the final cluster. A perfectly clear answer we’ll get this time, I’m thinking, hut as I reach to start the count, Don M ateo puts his hand on my arm: “ It's clear! Very clear! There’s no need to count. The answer is yes. It’s him .” “ N o need to count the days?” “ N o. We already know very well who it is.” So I give in, though I w as taught otherwise. Don M ateo himself, who’s not one to be fussy about the minute details o f ritual, always dis­ penses with counting when all he wants is a yes/no answer. An odd number o f seeds is no; an even number is yes. Four seeds at the end is an even better yes than two. This time I bring the seeds and all the crys­ tals together (officials included), rub them a few times, pick them up by handfuls, and funnel them into the bag, down to the last clicks and clinks. Don M ateo points out a seed that slipped off the slab at some point, prop­ erly leaving it to me to pick it up, and that goes in the bag, too, along with his ten-cent coin. When I’ve got the string wound tight and tied, I rub the whole bag between my palms a few times - it sounds like a rattle, but muffled - and then press it to my lips, putting it away in the big breast pocket of my wool shirt. Nothing more was said about the pig that day, nor about Raim undo Sucac. But we later

heard that the man had died two weeks after our visit to the mountain. That would have been on Nine Thunder fraying, suffering, which is to say Nine To;, once the day o f the god Tojil, the very god whose temple lies in ruins on that mountain. When I heard the news, it came home to me that Don M ateo had tricked me into being a sorcerer’s apprentice. 1 should have known it at the time. Whatever is said at a shrine, so long as incense is still smoking or candles are still burning, goes on the record with that mountain and with the Floly World, goes over the telegraph line that connects the mountains together, reaches Secretary Net, is written in the First Book, in the Second Book , on the First Shelf on the Second shelf. Don M ateo had put the pig and its unnamed killer on record with his own prayers on the day Eight Bird , praying, pleading asking that the killer fall into the hands of the powers that be. Then he had set me in search of the m an’s identity - or else put me to work confirming a suspi­ cion he already had. In the end, the name of the late Raimundo Sucac Awel had been uttered by both o f us in the presence o f the smoke and candlelight. Don M ateo, as we already knew at the time, is the kind of person who knows how to ask for justice , as the matter is delicately put. That’s the way he would put the matter, but there are other people who would call it by other names.

Itzb ’al. Brujeria. Witchcraft. But he uses no scraps of clothing, no nail clippings, no locks of hair, nor toads, nor poisonous brews. Candles, incense, words, and names work well enough for him. And he gets a little help from a saint he has at home, the darkly clothed figure he keeps in the same room with his brightly painted Apostle John, but on the opposite side. It’s the saint they call St. Simon, Simon Judas.

NOTE

Mythic Voices an d Visions o f the Living M aya (San Francisco: Flarper, 1993) and

Portions o f this contribution have appeared in Dennis T ed lock , Breath on the M irror:

are reprinted here with perm ission o f the author.

TH E PO ETICS OF TIM E IN M A Y A N DIVIN ATION

REFERENCES A bu-Lugh od, Lila. 1986. Veiled Sentiments:

H onor and Poetry in a Bedouin Society. Berkeley: University o f C alifo rn ia Press. Bachelard, G aston. 1988. The Right to Dream. J. A. U nderw ood, trans. D allas: D allas Institute. Bakhtin, M . M . 1981. The D ialogic Im agina­ tion. C . Em erson and M . H o lq u ist, trans. A ustin: University o f T e x a s Press. Burns, Alan F. 198 3 . An Epoch o f Miracles: O ral Literature o f the Yucatec M aya. Austin: University o f T ex a s Press. C o n d om in as, G eorges. 1977. We Have Eaten

the Forest: The Story o f a M ontagnard Village in the Central H ighlands o f Vietnam. Adrienne Foulke, trans. N ew Y ork: Flill and W ang. D auenhauer, N ora M ark s, and R ichard D auen ­ hauer. 1990. H aa Tuw unaagu Yts/For Healing O ur Spirit: Thngit O ratory. Seattle: University o f W ashington Press. Evers, Larry, and Felipe S. M olin a. 1987.

Yaqui Deer Songs/M aso Bwikam: A Native American Poetry. T ucson : University o f A rizona Press. Feld, Steven. 1982. Sound an d Sentiment: Birds, Weeping, Poetics , and Song in Kaiuli Expression. Philadelphia: University o f Pennsylvania Press.

395

Flurston, Z o ra N eale. 1935. Mules and Men. Philadelphia: Lippincott. Pow ers, W illiam K. 1982. Yuwipi: Vision and Experience in O glala Ritual. Lincoln: Uni­ versity o f N eb rask a Press. T ed lock , B arbara. 11982] 1992. Time and the Highland M aya. Rev. ed. A lbuquerque: Uni­ versity o f N ew M exico Press. ------. 1991. From Participant O bservation to the O b servation o f Participation: The Em er­ gence o f N arrative Ethnography. Jo u rn al o f Anthropological Research 4 7 : 6 9 -9 4 . ------. and Dennis T edlock. 1985. T ext and T extile: Language and T echnology in the A rts o f the Quiche M ay a. Jou rn al o f Anthro­ pological Research 4 1 : 121—46. T edlock , D ennis. 1983. The Spoken Word and the Work o f Interpretation. Philadelphia: University o f Pennsylvania Press. ------. 1985. Popol Vuh: The M ayan Book o f the Dawn o f Life. N ew Y ork: Sim on and Schuster. ------. 1988. M ayan Linguistic Ideology. In Paul V. K roskrity, ed., On the Ethnography o f Communication: The Legacy o f Sapir , 5 5 -1 0 8 . L os Angeles: University o f C a lifo r­ nia at Los Angeles, D epartm ent o f A nthropology. Underhill, Ruth M . 1938. Singing fo r Power:

The Song M agic o f the Papago Indians o f Southern Arizona. Berkeley: University o f C alifornia Press.

What Ritual Does: The Foundations of Order

introdu ction Here we move to accounts that analyze ritual less in terms o f the sym bols it puts into play than in terms o f its general effects. The structure o f ritual is seen to enable a particular kind o f action, one which carries significant consequences. Following Austin’s analysis o f speech acts (19 6 2 ; introduced in T a m b ia h ’s essay, chapter 25), ritual is recognized not only to say som ething meaningful to those w ho participate in it, or to actively persuade them o f something, but to institute change through its very performance, as when the correct enactment o f a wedding ritual transform s two unmarried people into a married couple. H o w does this w ork, and why do people accept it? If the consequences are that people com e to subscribe to larger orders o f meaning and morality, ones that frequently entail their own subordination, w hat does this say for the relationship between ritual and religion, on the one hand, and society and politics, on the other? The essays here are selections from two o f the most original and profound thinkers on these critical subjects. They should be rounded out with T a m b ia h ’s essay on performativity (1979).

32

The Disconnection between Power and Rank as a Process Maurice Bloch

M aurice Bloch was professor of anthro­ pology at the London School of Eco­ nomics and visiting professor at the College de France. He is w idely recog­ nized for his incisive and provocative interjections in anthropological debate. His essay addresses the central issue of ideology or the relationship of th e sym­ bolic to the m aterial, and hence of "rank" to "pow er." Refusing reduction­ ist lines of either the idealist or the m aterialist variety, he develops a pene­ trating account of the w ay in which the production of rank through ritual has played a significant role in the evo­ lution of political systems in highland M adagascar. In the full version Bloch develops a model for the rise of M erina and Betsileo kingdoms in central M adagascar and their organization into ranked demes (endogamous, landhold­ ing descent groups). Bloch's essay also touches on the ques­ tion of "sacred kingship/' a topic of

interest since Frazer and reinforced by the kinds of political systems anthro­ pologists have described in Africa, Poly­ nesia, and elsew here. "Sacred kingship" conjoins the "religious" and the "politi­ cal," and Bloch shows how this can come about. W hile Bloch's model does not account for all sides of ritual, he pro­ vides a com pelling argum ent that attempts to address w hat is certainly a common feature of both political hier­ archy and religion, namely mystifica­ tion. It may be argued that in positing a transcendental order or sacred author­ ity and a particular view of the world any religious system must w ork by con­ cealing both the elem ents of its own devices and viable alternatives. This is not to suggest that any particular reli­ gious system is false, but rather that each must conceal something. W hile some might follow Marx to see this as a general critique of all religion, others (such as Rappaport in the essay that

From M aurice Bloch, “ The Disconnection between Power and Rank as a Process: An Outline ot rhe Devel­ opment of Kingdoms in Central M adagascar,” in Ritual, History and Power: Selected Papers in Anthropo­ logy {O xford, Berg Publishers, 1989 [1979]), pp. 4 6 -8 8 . Abridged.

TH E D ISCO N N ECTIO N B ET W EEN PO W ER AN D RA N K A S A PRO CESS

follows) argue that some mystification is necessary for the developm ent of sacred authority th a t can stand as the guarantor of truth. A particularly intriguing look at Marxist dem ystifica­ tion and the religious roots of the concept of ideology is Jarvis (2000). Bloch has pursued the ideas devel­ oped in this early essay in a num ber of interesting directions, puzzling always over the distinction betw een everyday

In trod u ction The com plex ranking system s o f south Asia and Polynesia offer a special challenge and a special trap to rhe an th rop ologist or historian attem pting to explain their developm ent. He will inevitably trace rhe political and econom ic developm ent o f the state and link those to the app earan ce o f new types o f pow er grou p s and new structures o f stratification. The trap lies in identifying and correlatin g directly the a cto rs’ concept o f these g ro u p s, rhe distin c­ tions and associatio n s m ade in the emic system , with interest g rou p s identified by analyses o f the political or econom ic structures. T his may be done consciously or by im plication by tran slatin g a cto rs’ category term s by w ords such a s “ nobility” , “ ruling c la ss” , “ bo u rgeo i­ sie ” . T his is likely to lead to gross ethnocentric assum p tion s and obscure the specific nature o f the ethnographic exam ple. The danger o f such an ap p roach has been explained well by D um ont and Pocock (1 9 5 8 ; D um ont, 1966) for the m ost com plex and the m ost fam ou s o f such hierarchical system s. They poin t out how the Indian caste system does not fit either in structural form or in id eologi­ cal content m odels borrow ed from feudal o r class system s. Indeed such com p arison s obscure rhe essentially religious aspect o f H indu ideology contained in rhe scheme. Sim ilar criticism s w ould a lso be pertinent to the w ay the hierarchical system s o f central M ad a g asc a r have been an alysed , and for very sim ilar reasons.

399

common sense or practical logic and the cultural/religious worlds established through ritual. See in particular his response to Geertz (Bloch 1989b [1977]), his attem pt to follo w M erina circumci­ sion ritual historically (Bloch 1986), and his more general theory of sacrifice and the transcendent order (Bloch and Parry, eds., 1982, Bloch 1992). Most recently he has taken a cognitive approach to the problem (1998, 2005).

If the trap offered by the study o f such system s has been brilliantly identified by Pocock and D um ont, the challenge on the other hand has been refused. The study o f caste produced by these w riters, how ever illu­ m inating it is o f the Indian case in separatin g pow er and religion and in pointing ou t that the latter can n ot be reduced to the form er, goes no further. H ow ever much the religious prin­ ciple o f purity is in practice reinterpreted in particular situ ations, it rem ains for them unal­ tered. Im m aculately conceived in Indian theol­ ogy, it has survived equally unsullied. There is, how ever, another legitim ate and m ore am b i­ tiou s way o f handling such system s. It is the w ay suggested in part by Leach (1 9 5 4 ; 1960; 1968). While D um ont sees the econom ic and political base as affected by the ideology o f caste, the ideology itself is for them unaffected by the tran sform ation s in the base. By contrast with this ap p ro ach , the case o f central M a d a ­ g ascar will be exam ined here as a tw o-w ay relation w here the econom ic base is not just being affected by the ideology but the ideology is being indirectly created by the base. Neither is reduced to the other since both transform them selves within their ow n logic, which because o f their different nature is a different logic. The disconnection between rank or statu s and pow er is seen a s a result o f the evolution o f states in central M a d a g a sc a r, not a s a reason to give ideology an existence and an origin o f its ow n, or, in other w ords, to put it beyond explan ation in m aterial term s.

400

M A U R IC E BLOCH

Hasina Underlying both the notion o f hierarchy and rank, a s well a s the notion o f continuity and o f the “ n atu ral” quality o f pow er, is the concept o f hasina. T he best discu ssion s o f this for the M crina are to be found in the w ork o f A. Delivre (1 9 6 7 : 1 7 7 -8 8 ). . . . The w ord has proved to be one o f the m ost difficult to tran s­ late from M alag asy . The first tran slation o f hasina and o f its adjectival form m asina w as “ h oly” ; thus the earlier m ission aries tran slated “ H oly S p irit” by fanahy (spirit) masina. All n otions o f spirituality, o f superiority a s o f essence, are describabie as masina. H ow ever, the w ord also h as a political asp ect where it m eans legitim ate or traditional authority. T h u s, the hero kings o f M a d a g a sc a r are thought to be the epitom e o f the quality o f hasina. The sites o f their tom bs and o f their cap itals are them selves masina. In this sense the w ord m eans pow er, vigour, fertility, effi­ cacy or even sain th ood . It is the essence o f royalty and the essence o f superiority o f one person over an oth er through “ virtue” in the old sense. It is closely associated with the notion o f pow er and virility, mahery, w hich is in fact the fetishized virility created by the cir­ cum cision cerem ony and which is associated with descent and repetition o f life channelled through the to m bs (Delivre, 1967: 188). H asina is linked with the m ystical pow er of nature, especially the pow er o f reproduction, both in its hum an aspect and its asp ect in rela­ tion to crops. In its hum an aspect the p o sse s­ sion o f hasina by the rulers and the denies is w hat ensures the p assin g on o f life from gener­ ation to generation and the transcendence o f death which can be seen as the kernel o f M alag asy thought (Bloch, 1971: 2 2 2 ). In its agricultural asp ect it is the presence o f hasina which ensures fertility and clim ate; thus we see that those with hasina are able to bless cro p s in tim es o f d rough t (C o p alle, 19 7 0 : 5 2 ; Delivre, 196 7 : 182). The sam e can be said o f the Betsileo notion o f hasina. D u bois defines it as a “ vertue inherente a un etre” and Edholm goes on to say that it is “ given at birth - not acquired by political leadership. H asina is quite widely p ossessed , although carefully m ain­ tained within the sam e group a s a result o f

stringent rules o f exo gam y . . . ” (Edholm , 1971: 144). D u bo is say s that a m an possessin g hasina had a “ nature superieure et une authorite sac re ” (D u bo is, 1938: 99). H ere again , there­ fore, hasina is inherent to certain people and its presence assu res fertility. T he pre-em inence o f the ruler com es from his suprem e possession of hasina: “ He stan d s for the perm anence and rightness o f justice and authority and spiritual superiority” (Edholm , 1971: 153). H asina is a supra-hum an quality and therefore although contained by certain domes in differing degrees it is not the result o f their achievem ent but is given in their nature. T he concern o f the hasina holders should be to preserve it; creating hasina is out o f the question . The concern o f those w ho hold less or no hasina should be to pre­ serve it in those w ho have it because they too benefit from it. T hus the authority o f the ruler is seen a s a m anifestation o f som ething which has alw ays existed, which is given in nature by the order o f the w orld. The ideological picture could not be further rem oved from the short-lived, brutishly extortionate reality o f traditional M erina and Betsileo pow er. T h is sort o f notion is found associated with traditional authority in m any parts o f the w orld (Bloch, 1961; C haney, 1970; de H eusch, 1966; Balandier, 1970). . . . W e find the sam e con cepts again and again : thus Southall referring to Ker am o n g the A lur sees this notion as denotin g pow er, beneficial d o m i­ nation an d, in B alan dier’s w ords, an “ o rg an is­ ing and fertilising fo rce” (B alan dier, 1 9 7 0 :1 0 4 ; Southall, 1956). Balandier, follow ing M au ss, a lso points o u t another essential aspect o f such concepts which is particularly relevant here: how the concepts o f innate authority transcend and are a w ay o f vanquish ing the instability o f real pow er. These con cepts can , therefore, becom e agen ts o f tran sform atio n in a con cep­ tual alchem y where the achieved and therefore unstable turn s into the perm anent and there­ fore cosm ic (B alandier, 1970: 11 0 -1 3 ). J . Pitt-Rivers goes further in asso ciatin g such concepts especially in bringing together the notion o f tnana , which in its political m an ifes­ tation for the M ao ri is sim ilar to that o f hasina for the M erina and Betsileo (Salm on d, 1975), and the E uropean concept o f honour; in doing so he show s up the essential and necessary

TH E D ISCO N N ECTIO N B ET W EEN P O W ER A N D RA N K AS A PRO CESS

am biguity o f such notions. O n the one hand honour and authority is unchallenged, given and religious; on the other, it is dependent on the action s o f others since they m ust show their acceptance o f the sup eriors’ honour or mana. Thus the ab stract w ord “ h on ou r” is dependent on the verb “ to h on ou r” - the action s o f inferiors (Pitt-Rivers, 1974). In the sam e w ay we find that there is another side to the w ord hasitta than the one o f innate reli­ giou s superiority with which we have been concerned so far. T his side o f hasitta turns out not to be a state o f superiors but an action o f inferiors. For the sake o f clarity 1 shall call that part o f the sem antic held which corresp on d s to the form er idea hasina m ark I, while thar part o f the sem antic Held which corresp on d s to the latter hasina m ark II. H asina m ark II m anifests itself in the render­ ing o f hom age by som ebody with lesser hasina . Fie gives hasina to the superior in the sam e w ay as rhe inferior renders honour ro the hon­ ou rab le. T his recognition on the part o f som e­ body with lesser rank tak es the form o f the giving o f gifts o f respect an d honour o f a special kind, the use o f certain greetings and rhe giving o f precedence a t such times as form al assem blies. O f p articular relevance is the fact that the m ost significant gift which m arks differential rank is the gift o f an uncut silver coin ; the coin is itself called hasina (C allet, 1908: 2 9 1 ; Delivre, 1967: 186). This coin is the M aria-T h eresa T haler which w as in circulation throughout M ad a g asca r for the period during which we are concerned. Its sym bolical significance com es in part from the fact that it w as utsy v a k y '\ unbroken, while m ost thalers in circulation w ere cut to obtain sm aller denom ination. The coin therefore w as the suitable gift to give to a superior to recog­ nize his authority. At the im portant state rituals the differences in rank were m arked by the giving o f hasina m ark II by subjects o f d if­ fering denies to the ruler an d by inferiors to superiors o f any kind right dow n to fathers an d son s (Ellis, 1838: 4 2 7 ). T he apparen t con tradiction between hasina in its m ark I aspects and its m ark II aspects is, how ever, not clear to the acto rs w orking within their ow n cultural system and mystified by its representation o f pow er. T o the an alyst o f the

401

system the contradiction turns ou t to be o f crucial significance tor linking up the system o f pow er described in the [om itted| section and the system o f rank described in this one. W hat rhe am biguity o f the term hasina produ ces is the possibility o f the representation o f hasina as an innate quality possessed by su periors, which is a benefit and a blessing, to in feriors, while in fact this illusion is created by an act which is its o p p o site, the giving o f a benefit, hasina m ark II, by the inferior to the superior. H asina m ark II is thus doubly the op p o site o f hasina m ark I: while the form er is n atural, the latter is su p ern atu ral; while the latter flow s dow n w ards, the form er flow s upw ards. The difference is represented in Figure 1 which show s a ranked hierarchy o f denies in the light o f hasina m ark I and m ark II. In the light o f the form er, hasina is an essence which flow s in the form o f fertility from the superior to the inferiors; in the light o f the secon d, hasina is gifts goin g the other way. T h u s, by the am biguity o f the term hasina , the supern atu ral gift o f the superiors turns out to rest, a s we know it m ust (hence the general­ ity o f this am biguity) on the natural act o f the inferior w ho, instead o f receiving the blessing o f hasina, creates it. Indeed, this reversal is still to be seen in fam ily relations am o n g the M erina and Betsileo to this day in the institution o f the tsodrano (Bloch, 1971: 163). T h is w ord m eans the blow ing o f w ater and it is perform ed by elders for their juniors a s a kind o f blessing for fertility and success. The elder stan d s those he w an ts to bless in front o f him and sprays them with w ater from a saucer he holds before his lips. H ere we have the notion o f hasina m ark 1 tran sferrin g a blessing o f fertility on the juniors because o f the pow er o f the holiness o f the elders. H ow ever, in order that the elder m ay be able to d o this he m ust be given coins, preferably the sam e coins as those used in the royal hasina- giving. T he coin is placed in the w ater that is sprayed on those receiving the blessing, bur it is pocketed by the elder as is the case for hasina m ark II.

W hen Hasina is m anifest So far I have stressed the disconnection between pow er and rank and how a turbulent state

402

M A U R IC E BLOCH

Hasina Mark I

Hasina Mark Higher demes

Blessings of fertility and efficacy

Gifts

Lower demes

Figure 1

based on unscrupulous exploitation by a sm all m inority o f rulers o f a large m ajority o f ruled is m isrepresented as an orderly harm onious system o f fine gradation o f rank which co n ­ tains no sh arp social breaks. We have seen how the representation o f differential rank is seen in term s o f a supernaturally given essence held in greater degree by those with high rank but which benefits all, an essence which not only com es from beyond men but has effect beyond m en, since it accords with and stim u­ lates the natural processes o f reproduction. T his cultural m ystification is, how ever, m ore than just a veil over the eyes o f the acto rs; by hiding the reality o f exploitation and tran s­ form ing it into an ideology which stresses the beneficent effect o f the presence o f the ruler, it serves to preserve the pow er o f the rule, to facilitate its acceptance and thereby to m ain­ tain it. By draw ing the line between rulers and ruled in the w ay that is done by the ranking system , the possibility o f class con sciousn ess on the part o f the ruled is m ade m uch less likely and so here to o the ideology preserves and indeed is an essential element o f e x p lo ita ­ tion. The non-correspondence we have noted is, therefore, m ore than an ethnographic or a historical puzzle: it is one o f the innerm ost m echanism s o f the reproduction o f the politi­ cal system . T his view o f ideology is o f course fam iliar in the social sciences through the

insight o f M arx a s well as several preceding and subsequent w riters. M ost recently one o f its m ost elegant form ulations has com e in M . G odelier’s an alysis o f the Inca state, where he show s how the transfer o f surplus from infer­ ior classes to superior classes is represented as a sm all return for the divine gift o f fertility from the g o d s m ediated by the lnca king and his co llab o rato rs (G odelier, 1973: 83). If ou r task could be limited sim ply to dem ­ on strating the function o f ideology as a m echa­ nism for the reproduction o f social form ation s, this paper could end here. H o w e v e r,. . . if the ideology o f M erina and Betsileo states obscures reality so as to reproduce it, the question o f how this representation can have com e about becom es o f prim e interest. The sim plest answ er, and one perh aps im plied in G odelier’s article, is that the false representation is a device on the part o f the ruling class to m aintain that which is plainly to their advan tage. T his, how ever, con jures up a totally unreal situation and is incom patible with the fact that all the evidence suggests that rulers were a s mystified a s ruled. The an th ropologists m ust therefore do m ore and produce theories which account for the rise o f ideologies m aintaining certain social system s, w ithout falling into the trap o f thinking that these ideologies are either, on the one han d, actu al reflections o f reality, or on rhe other, con scious M achiavellian devices

TH E D ISCO N N ECTIO N BETW EEN P O W ER AN D RA N K A S A PRO CESS

thought up by rulers to fool their subjects. The second h alf o f this paper is an attem pt to outline tentatively how a theory which avoids both these pitfalls can be constructed. The first step in this case is to ask a question which should follow from the preceding section. If hasina is prim arily a false mystical essence, w hat are the observable m an ifesta­ tions o f this illusion an d, secondly, when do they occur? These m an ifestation s are o f tw o kinds: first, m inor and frequent m an ifestation s an d, sec­ ondly, m ajor infrequent m anifestations. O f the m inor m an ifestation s the m ost im portant ones are greetings. People o f differing rank should be greeted by different types o f greeting, so andrianas [nobles), in the wide sense, in Imerina and to this day should be greeted by the phrase Tsara va Tom poko while other phrases should be used for non-andrianas. In a sense, andrianas have their rank at least at the beginning o f con versation s. A part from this, in the p ast in both Betsileo and M erina kingdom s different ranks had rights to wear certain types o f clothing, to sp eak at public assem blies in a certain order, and to be exem pt from a few law s, none o f great significance. For exam ple, high demes in Imerina had the privilege o f not having their blood spilt. So when they were condem ned to death, instead o f being killed in the usual w ays they were either strangled or drow ned. A lso o f signifi­ cance were rights relating to tom bs. Higher demes could have their tom bs within the village w alls, while low er demes could not, and the very highest demes had special shaped to m bs, the top m ost p art o f which were called trano m asina : holy houses. ( M asina in this case is the adjectival form o f hasina (C allet, 1908: 3 0 1 -7 4 )). A part from these m inor m anifestations o f differential hasina the dem on stration o f royal hasina from which all others spran g w as focused on four m ajor national rituals o f o u t­ standing im portance: the circum cision ritual, the coron ation cerem ony, the cerem ony o f the royal bath, and royal funerals. O nly a very brief outline o f these can be given here. The circum cision cerem ony o f the M erina and Betsileo is sim ilar and it is the only one o f the four which is still p ractised to d ay, though

403

in a changed form . F’very seven years or so , the ruler w ould declare the time for the circum ci­ sion com e and for the follow ing week or perh aps longer a series o f rituals are perform ed. First, the w hole kingdom is cleaned, both liter­ ally, and m ystically in the case o f w itches, who are killed if they are discovered through the adm inistering o f a poison oracle to rhe whole population . Then hasina (m ark II) is offered by all g ro u p s in ranking order to the ruler and also probably within the demes and between demes equally in order o f rank. Then to m ark the ritual state o f the popu lation , all plait their hair in a characteristic and com plex w ay. The king a s well as all the heads o f demes g o to the tom bs o f their predecessors and ancestors (assum ed to be the sam e) and m ake offerings and prayers which in the case o f the ruler are clim axed by the phrase “ H o m asina an ie” , “ Let me be m asin a ” . Then follow s the cere­ m ony o f the fetching o f the ratio m asina , holy w ater. T his w ater is fetched from several fixed places and its significance is that it increases the fertility o f hum ans; thus through the agency o f these with hasina the forces o f nature are harnessed to be passed on in the increased fertility o f the next generation which is the principal aim o f the cerem ony. Then the hasina- giving is repeated and all take part in a co m plex dance called soratra which is said to **hampandroso hasina'*, to “ force hasina to en ter” , but which at the sam e tim e acts out the differential hasina o f the various demes o f the kingdom (C allet, 1908: 7 3 -8 2 ). T he circum cision cerem ony stresses very explicitly certain them es which we shall find again in the other cerem onies: 1) The king is associated by the cerem ony with his royal pre­ decessors and through con tact with them he indirectly p asses on fertility to the w hole p o p u ­ lation. 2) The king is associated by the cere­ mony with the reproductive force o f nature and because o f his hasina he can canalize this to the rest o f rhe popu lation . 3) T he king’s and the dem es’ hasina is created by the giving of hasina m ark II as well as by the dance; at the sam e time the relative am oun t o f hasina is defined. T he second cerem ony where the hasina o f the ruler is stressed is the accession cerem ony.

404

M A U R IC E BLO CH

T his takes place at a sp o t called M ah am asin a (rhe place which renders m asina ; the presentday football stadium o f T an an arive), where ap art from receiving the hasina from his su b ­ jects in ranked order the king absorb ed the hasina from his predecessors by drinking w ater containing earth draw n from the tom bs o f the previous kings which he claim ed to be descended from (Ellis, 1838: vol. 1, p. 4 2 0 ). T his ritual clearly dem on strates the in co rp ora­ tion by the present king o f the hasina o f his predecessors, which is seen as reproducing itself again and again from generation to gen ­ eration channelled in one unchanging line. O f course, this type o f m aterial descent, contact with the co rp ses o f predecessors, which is typical o f central M a d a g a sc a r (Bloch, 1971) is a lso highly m an ipulable, since it is not descent through any notion o f filiation but descent through in corporation o f predecessors which can be done by an ybody irrespective o f w ho his parents w ere.

|...] Finally, and perh aps m ost interestingly, we have the yearly cerem ony o f the royal bath, the fandroana , . . . The best study availab le . . . is to be found in A. D elivre’s Interpretation d'une Tradition O rale (1 9 6 7 ). Delivre stresses how the royal cerem ony, involving much singing and dancing and other ritual activity and lasting for anything up to a m onth, show s a clear pattern em erging: on the one hand the p assage o f the old year and the com in g o f the new, and on the other the repetition o f king­ ship, not only within a reign, but beyond it from reign to reign. The m ain rituals o f the F an d roan a are in order: 1. 2.

3.

A m oratorium on funerals. O pening up o f the tom bs follow ed by a tidying up o f the to m b o f his predecessors by the king echoed in all demes by sim ilar tidying up in deme tom bs. The ritual o f the royal bath which gives the cerem ony its nam e. The ruler, after receiv­ ing hasina in the form o f the coin from his subjects in order o f rank, takes a bath with much ritual. He then blesses the subjects by sprayin g them with the w ater o f the bath. T h is increases their fertility and that

4.

o f the cro ps. O n steppin g out o f the bath the ruler say s: “ H o m asina an ie” , “ M ay I be m asin a,\ Fie then receives the gift o f hasina again . T h is is echoed by gift-giving th roughout the w hole kingdom as inferiors give to su p e rio rs.1 The night preceding the actual bath is m arked by the w hole p o p u ­ lation takin g part in a torchlight proces­ sion and in the past by a period o f sexual licence. T he cause o f both these is as much the royal bath a s the fact that the royal bath coincides with the new year, the p assage from the month o f A lahotsy a ss o ­ ciated with death to the m onth o f A lah am ady associated with fertility. T h is is a lso the m id-point between harvest and the beginning o f sow in g the new crop o f rice. Fo rw ard-looking cerem onies at the royal tom bs and closin g o f the tom bs (echoed at deme tom bs).

. . . The cycle o f the year is linked with the action s o f kinship. Then and by inevitable extension the fertility o f the crops is linked with the hasina o f the king and m ade depen­ dent upon it. The giving o f rank by the subjects to rhe king is linked to the giving o f rank by all inferiors to all superiors, whether superior and inferior dem es , or fathers and son s, or elder brothers and younger brothers, etc. In other w ord s, the receiving o f hasina by the king is seen a s the final result o f a w ave o f hastna giving th rou ghout the society. The recognition o f rank by all is m erged with the continu ation o f all, sym bolized by the con quest over death , o f which tom bs are the sym bol, a s well as the fertility-giving blessing o f the bath-w ater. T h is conquest over death for each deme and each fam ily is seen as dependent on the co n qu est over death o f the kingship. For this the present incum bent m ust be the rightful successor o f other kings w hose to m bs are his t o m b s .. . . M aterial and essen­ tial succession is m erged with a process which is in fact m an ipulable but which, when looked back upon, has all the appearance o f biological descent, since it involves the transfer o f m ater­ ial substan ce between generations o f rulers. T he actual bathing a lso re-echoes central them es o f both the circum cision cerem ony and the royal funerals. The w ater for the bath

TH E D ISCO N N ECTIO N B ET W EEN P O W ER A N D RA N K A S A PRO CESS

should be obtained from the sam e places as the w ater tor the ratio m asina , the holy w ater o f the circum cision cerem ony; by con tact with the body o f those with hasitta it tran sform s n atural fertility into social fertility, which is then used to bless the assem bled m ultitude in the sam e w ay that the ratio m asina is poured by the elders on the penis o f the child. T his use o f w ater also re-echoes the tsocirano discussed above. In all these cerem onies then we see the asso ciatio n o f the pow er o f nature and the pow er o f men inextricably linked to become hasina. The substance which gives rank its legitim acy is as much o f nature a s o f men; it is, therefore, beyond question and infinitely desirable since on it relies the life o f all. The p a ssa g e from one year to the next and the prom ise o f the future rice harvest is asso c i­ ated with the p assage o f kings from year to year through tim e, a p a ssa g e which is con tin u­ o u s since the present king is the representative o f the corp ses inside the to m bs where his pre­ decessors have been buried which will contain him and his successors. T he ritual associates the king and his bath with the renewal o f hum an fertility through the w ater o f the bath w hich, as in other cerem onies, sym bolizes the continuity o f hum an life. T h e w hole cerem ony is focused on the ritual at the palace but is re­ echoed by m any observan ces inside every M erina home. Its centre is m arked by the bath, preceded by a night o f unbridled sexual licence which has won it much fam e am o n g European w riters but which often m ark s the turningpoint in cerem onies o f continuing fertility in M ad a g asc a r. The whole cerem ony establish es, recognizes and differentiates between degrees o f hasina. As the king com es out o f his bath he utters the phrase: “ H o m asina an ie” , “ that I m ay be m asina” , an d accom pan yin g this, both im m ediately preceding an d follow ing the bath, he is given hasina m ark II by the other high-ranking m em bers o f his entourage; a gesture repeated throughout the kingdom by those o f lesser rank giving to those o f higher rank, repeated inside every fam ily by sons giving hasina to their fathers, younger brothers giving hasina to their older brothers, etc. At the very m om ent when the king m anifests his suprem e hasina , associatin g it with the recur­ ring cycles o f nature and reproduction, the

405

w hole hierarchical rank system o f the kingdom is, so to sp eak , activated in order to m ask the actu al pow er situation. The m askin g process tak es place in the tran sposition o f political p ow er, achieved by force again st subjects and predecessors, to an aspect o f the natural lifegiving processes o f nature and hum an sex u al­ ity; in other w ords, pow er is represented as beneficient nature - it is the tran sposition of unique achievem ent into repetitive and there­ fore infinite events and the tran sposition o f a sh arp break between rulers an d ruled into a continuum o f differential hasina. T h e . . . royal rituals we have looked at so far carry the sam e m essage, an am azing m essage in the light o f the reality o f M erina an d Betsileo history. T he m essage is thar politi­ cal pow er is an aspect o f a m ystical pow er distributed throughout the popu lation in dif­ fering degrees. T his pow er is not prim arily political but is an aspect o f nature. Like nature it is repetitive but unending. Like nature it is life-giving. Like nature (by definition) it is not created by m an but exists independently o f m an. T he present ruler if he has hasina is, therefore, nothing but the continuation o f his predecessors, the vessel o f the hasina which they had. T he life o f the subjects and rhe produce o f the earth is dependent on the co m ­ m union with the religious forces o f the uni­ verse which is achieved through him. The an alysis o f the . . . royal rituals which h as been outlined abo ve show s when ir is that the illusion o f hasina and the rank are created. A part from the m inor m anifestation o f hasina and rank discussed above, it is during rituals thar pow er is tran sform ed into sain th ood , thar hasina m ark II takes on the illusory a p p ea r­ ance o f hasina m ark 1. For periods o f varying length, norm al activity is suspended by rituals w here rhe experience o f exploitatio n and intim idation, im plied by political rule m ost o f the tim e, is transform ed into a totally different picture, deperson alizing the relation sh ips o f oppressio n to relationships o f protection by rulers w ho are vessels o f universal beneficence in the form o f fertility. The fact that hasina and rank are illusions o f certain tim es but not all tim es is w hat I believe can furnish us with an answ er to the question posed above: how can a m isrepresen-

406

M A U R IC E BLOCH

tarion o f political and econom ic reality com e abo u t w ithout anybody w orking the system consciously so as to m islead oth ers? The answ er will com e by lookin g at w hat is special abo u t those tim es when hasina and rank ap p ear to have reality, in other w ords by utiliz­ ing a theory o f ritual. The reason why a theory o f ritual is relevant to the understanding o f the relation between rank and pow er com es from rhe fact that when we are lookin g at pow er we are lookin g at unform alized (unritualized) action using ord i­ nary language and physical force for its co m ­ m unication. We are looking at activities for which the m anner in which things are done is secondary to w h at is done. When we are looking at rank, by con trast, we turn to rituals, whether religious or secular, whether largescale extended affairs like the cerem ony o f the bath or very sm all-scale affairs like greetings. These riruals use a type o f com m unication different from other com m unication: fo rm al­ ized language, son g, form alized bodily ges­ tures, dance and com m unication by m aterial sym bols. W hen we are looking at pow er and rank we are therefore looking at different activities and different types o f com m unication involving the sam e actors. From time to time life is “ lifted u p ” , so to sp eak, to an oth er plane and we m ust understand the significance o f this “ lifting” . It is im portant to stress that the difference between pow er and rank refers to different types o f activities in which the actor is involved and not a move from the ob server’s to the actor’s category (I believe this is a fairly com m on m istake in an th ropology). Indeed, the perception o f exaction in its unadorned form is just as much part o f the experience o f the actors as the experience o f rank in rituals, but it occurs at different tim es and in a different m ode o f com m unication. The M erina peasant is som etim es faced by arm ed superiors who take aw ay part o f his cro p s but a few m onths later he is a singer and a dancer in the cere­ m onies o f the bath.

C o m m u n icatio n in Ritual O nly a very brief outline o f a theory o f ritual can be given here. The argum ent is presented

in an article entitled “ Sym bols, son g, dance and features o f articulation: Is religion an extrem e form o f traditional au th o rity ?” . . . (Bloch, 1974, reprinted a s chapter 2 o f the volum e from which this extract com es]. The starting point o f the argum ent is the o b se rv a­ tion that ritual, whether religious or not, is characterized by certain special (extrao rd i­ nary) types o f com m unication. These are mainly form alized speech and son g, form alized bodily m ovem ent and dance and m aterial sym bols which replace w ords. All these m edia o f com m unication share features in com m on in contradistinction to m ore ordinary co m m u ­ nication, especially ordinary language. The first and m ost fundam ental aspect o f ritual com m unication is that although it ap p ears richer in aesthetic and em otional values, it is poorer a s to its sem antic potential and its ability to carry a logical argum ent. T his is because, as has been show n by generative sem antics, the logical potential o f com m u n ica­ tion is dependent on aspects o f syn tax, and these are w eakened or destroyed in these m edia. O rdinary language depends for its ability to say new things within a given m ould on the articulation s which unite a continuous unit o f com m unication. In ordinary language one bit o f com m unication can be follow ed by any other bit. T h is is because syntax is a system which can link m eaningfully practically any tw o proposition s. In ordinary language one bit can be follow ed by a near infinity o f other bits within a fram ew ork which gives this sequence m eaning. In the case o f form alized lan guage, and espe­ cially son g and such non-linguistic com m uni­ cation as dance, this high flexibility in the articulation o f com m unication is either greatly reduced or alm ost absen t, and so the potential for saying new things within a given structure is equally reduced or absent. The articulation o f com m unication becom es, so to sp eak , arthritic. While in ordinary language the su p ­ pleness o f com m unication m eans that unit A can within the rule be follow ed by an infinity o f units B, in ritual language or son g A can, within the rule, be follow ed by very few or even only one B. T h u s, while for ordin ary lan ­ guage the only reason which links A and B is the nature and force o f the argum ent, in ritual

TH E D ISCO N N ECTIO N BETW EEN PO W ER AN D RA N K AS A PRO CESS

the link between A and B is partly given by the nature o f the com m unication. T w o points follow from this: 1) C om m un ication is not easily or at all ad ju stab le to the reality of a particular place or time because o f its arthritic nature. 2) R easoned con tradiction or argu ­ ment, which im plies a supple form o f link-up with previous utterances, is equally reduced or ruled out. As a result the com m unication o f ritual is protected by its form from challenge, rapid m odification, or evaluation again st other statem ents or em pirical d ata. Secondly, not only does ritual com m unication restrict the articulations between p arts o f the m essage, it a lso dram atically limits the range o f w ords o f larger units o f com m unication which are ap propriate. Particularly it reduces the range o f possible illustrations which are thought suitable parallels to m atter at hand. O ften it specifies that any illustration m ust be drawn from a scriptural co rp u s or from a single his­ torical tradition. The effect o f this restriction on com m unication is particularly interesting: it m eans that any single ritual event is, in the perspective o f ritual com m unication , going to be seen sim ply as a m an ifestation o f an event which recurs or which is cyclic. Ritual co m ­ m unication m akes the social w orld appear organized in a fixed order which recurs without beginning and w ithout end. A s a result the social is like the n atural, even a part o f nature, and so ritual com m unication projects the political, the social, the discontin uous, the cul­ tural and the arbitrary into the im age and the realm o f repetitive nature. Since ritual cannot “ com e to g rip s” with reality this is transform ed into som ething quite different. The theory is relevant to w hat we have been looking at so far in the follow in g way: it explain s why the ritual representation o f rank is apparently unchallenged by very different experiences o f pow er. It a lso exp lain s why rank and hasina are represented as cyclic and therefore perm anent. It a lso m akes clear why statem ents o f pow er which are “ given ” in the universe (since they cannot be challenged) and repetitive, inevitably link up with the processes o f nature and especially its beneficial cyclical aspects, fertility and reproduction. Putting statem ents into ritual language puts them outside society and into nature since the state­

407

m ents ap p ear to have transcended the actions o f men and are experienced as though from outside the individual (pace D urkheim ). T he theory o f ritual under discussion , how ever, has som e m ore directly evolutionary im plications. We should think o f rituals less as given, m ore as part o f rhe process o f leader­ ship. Political leaders have been divided between those with traditional authority and those with charism atic authority. It is clear that it is those with traditional authority who use rituals. T his w as brilliantly analysed by W eber w ho stressed the processual side o f this. T hose who acquire pow er institutionalize it to m ake it less vulnerable from the attack s o f rivals, rhey put this pow er “ in rhe bank o f ritu al” . They do this by creating an office o f which they are the legitim ate holders. But which has reality beyond them. T his is done by g radu al ritualization o f the pow er-holder’s com m unication with the rest o f the w orld, especially his infe­ riors. A s this ritualization process proceeds, com m unication loses the appearan ce o f a cre­ ation on the part o f the speaker, and ap pears like repeats o f set roles specified by the office which ap p ears to hold him. Reality is thus reversed and the creation o f the pow er-holder ap p ears to create him. T h is process m eans tw o things: pow er having becom e authority ap pears less and less challengeable, but at the sam e tim e it becom es less rhe pow er-holder’s own. This results in the fact, also em phasized by W eber, that traditional authority is eminently transferable. T h is is because as the leader puts his control “ in the b a n k ” he rem oves it from his ow n achievem ents by m aking it unrelated to history and part o f nature (see above). Ritualizing pow er m eans that the achieved pow er is tran s­ form ed into ritual rank by rituals, but it also m eans that in the process rank becom es sepa­ rated from its origin, i.e. the acts o f pow er, and any direct feedback is ruled ou t by the very process o f rituals. In this w ay rank will inevi­ tably becom e disconnected from pow er. Power produ ces rank through ritual but this gives rank independence, a “ realism ” which results in greater time persistence. Then pow er will change but rhe rank it has produced will stay.

408

M A U R IC E BLO CH

The fact that ritualization rem oves authority from the person o f pow er-holders has anoth er evolutionary im plication which we m ust touch on briefly, to return to it below . T h is rem oval is a problem for the ruler him self, in that it not only rem oves events from place and time and com pels others by placing them in a repetitive and apparen tly external system , but it also deperson alizes him , rem oving from him the possibility o f m anipulation. If a ruler totally ad o p ts the m ode o f com m unication o f ritual for carryin g ou t his w ill, he at the sam e time loses his ability to affect events for his ow n personal ends as op p o sed to the ends o f his office. He m ust therefore continually switch from the m ode o f com m unication o f ritual to a secular m ode o f com m unication , using the form er for statem ents concerning his unchal­ lenged role a s “ father o f his p eop le” , the latter in order that he m ay play the astute politician (Bloch, 1975). In evolutionary term s one can see this whole process a s one w hereby rulers as they becom e accepted ritualize their c o m ­ m unication with their subjects to protect them ­ selves from short-term challenges, but as they do this they begin to need tw o alternative cod es, playing one after the other, which co r­ respond to tw o roles which they m ay well alternate, that o f priest and politician, th at o f orato r and p atron . The superior w ho tries to avoid sep aratin g the con tradictory roles o f father o f his people and politician is a su ccess­ ful ruler, but he is at all tim es facing the d an ger o f being kicked upstairs a s a priest or kicked aside by a successful rival w ho takes over the role he has created.

|...|

Conclusion D u m on t’s characterization o f the Indian caste system and his insistence on the disconnection between statu s and pow er gave us our startin g point. While fully accepting the strictures on the d an ger o f reducing one o f these system s to the other I have tried to show that in an a n a lo ­ g o u s case it is still possible to see the ideology o f hierarchy a s a p rodu ct o f political econom ic history. O f co urse, M alag asy history d o e s not explain the specific form o f the Indian system .

T o use it in this w ay w ould be to fall again into the reductionism and ethnocentricism D um ont criticizes. W hat the study presented here does show is a system with striking sim i­ larities to the Indian caste system which can be accounted for. T h is is especially so since the sim ilarities between the tw o system s, the gradu al an d all-encom passin g nature o f the hierarchy, the disconnection between religious and secular pow er, the continuity in the reli­ giou s interpretation o f hum an society and o f nature are precisely those features o f the Indian system particularly stressed by D um ont. T he m ethod here has been to outline the existence o f tw o system s. T he first is the system o f political econom y evolving accordin g to its ow n logic and affected by such things a s the ecology o f central M a d a g a sc a r and the tech­ nology o f w hat J . Cioody h as called the “ m eans o f destruction ” (J. Cioody, 1971).2 Then I have considered another system , the system o f ideas held by the M erina and the Betsileo, and 1 have outlined the system as having tw o different kinds o f relations to the first system . T he system o f ideas m ust, on the one hand, be in p art congruent with the eco­ nom ic and political system , since it m ust offer a practical system o f com m unication to operate it, in the sam e w ay a s it m ust offer a practical system o f com m unication to exploit nature and obtain a living from it. O n the other hand, the system o f ideas is in certain dom ain s sep a­ rated from the reality o f life and feeds on itself to create another pseudo-reality. T his is p o ssi­ ble only because o f the sem antic im plications o f ritual com m unication which protect ideas from challenge by em pirically generated know ledge, and which links up ideas in a w ay which is sub-logical. The system o f ideas is therefore m ade up o f tw o kinds o f system s o f com m unication which can never interact since they use different m edia. It is because o f the existence o f this second system o f ideas that a disconnection occurs between ideas and reality, rank and pow er. O nce this disconnection has occurred an d the system o f ideas in som e m atters gain s a life o f its ow n, its relation to such events a s the grow th and death o f king­ dom s becom es indirect, evolving within its ow n m ystical rationality an d creating further disconnections with the base.

409

TH E D ISCO N N ECTIO N BETW EEN P O W ER A N D RA N K AS A PRO CESS

It is because o f this disconnection that m ost evolutionary theories, whether p ast or present, ap p ea r sim ple-m inded or sim ply w rong. They all try to establish a direct and unique relation between the politico-econom ic base and ideol­ o gy , while any study o f a p articular case show s

this not to be valid. O n the other hand, if we have to use much m ore com plex theories in the future I hope I have show n, w hatever the value o f this particular case, that we need not give up.

NOTES

------

1

De H eusch , L. 1966. Pour une dialectique de la sacralite du pouvoir. Le Pouvoir et le Sacre. A nnales du Centre d’ fitudes des R eli­ gion s. Brussels. D u b o is, H. M . 1938. M onographie des Betsileo, T ra v au x et M em oires de Plnstitut d ’Ethnologie, vol. 3 4, Paris. Q uoted in F. Fdholm 1971. D u m on t, L. 1966. H om o Hierarchus. Paris. ------and D. Pocock 1958. A. M . H o cart on caste: religion and pow er. Contributions to Indian Sociology, vol. 2. Fvdholm, F. 1971. R oyal funerary rituals am ong the Betsileo o f M ad a g asca r. U npublished thesis, London School o f Econom ics. Ellis, W. 1838. History o f M adagascar , vols 1 and 2. London. G odelier, M. 1973. Horizons, Trajets M arxistes en Anthropologie. Paris: M asp ero . G o o d y , J. 1971. Technology , Tradition and the State in Africa. L on don : O x fo rd University Press. Leach , E. R. 1954. Political Systems of High­ land Burma. Lon don : Bell. ------I 9 6 0 . T he frontiers o f Burm a. C om para­

2

It should be noted that gift-giving in M ad a g asca r is alw ays a sign o f inequality, while dem an ds are a sign o f equality. See full version.

BIBLIOGRAPHY Balandier, G . 1970. Political Anthropology. T ran slated from the French. London. Bloch, M arc. 1961. Les Rois Thaumaturges:

Etude sur le Caractere Surnaturel attribue a la Puissance Royale particulierement en Prance et en Angleterre. Paris. Bloch, M aurice. 1971. Placing the Dead: Tom bs, Ancestral Villages and Kinship Organisation am ong the Merina o f M adagascar. Lo n d on , N ew Y ork: Sem inar Press. ------ 1974. Sym bols, son g, dance and features o f articulation. European Jou rn al o f Sociology, vol. 15, pp. 5 5 -8 1 . ------ 1975. Introduction to M . Bloch (ed.),

Political Language and O ratory in Tradi­ tional Society. L on d on , N ew Y ork: A ca­ dem ic Press. C allet, R. P. 1 9 0 8. Tantaran ny Andriana eto M adagascar 2nd edn, vol. 1, T an an arive: A cadem ie M algache. C haney, W. A. 1970. The Cult o f Kingship in Anglo-Saxon England. M anchester. C o p alle, A. 1 9 7 0 . Voyage a la Capitale du Roi Radam a 1 8 2 5 -1 8 2 6 . D ocum ents anciens sur M ad a g asc a r, vol. 1, A ssociation M algach e d ’ A rcheologie. Delivre, A. 1967. Interpretation d ’ une tradi­ tion orale. U npublished thesis, Paris. N ow published a s Delivre 1974.

1974. Interpretation d'une Tradition O rale. Paris: Klincksieck.

tive

Studies

in

Society

and

History,

vol. 3. ------ 1968. Introduction to E. R. Leach (ed.), Dialectic in Practical Religion. C am bridge: C am bridge University Press. Pitt-Rivers, J . 1974. M ana. The London School o f Fxonom ics and Political Science. S alm on d, A. 1975. M an n ers m aketh m an. In M . Bloch (ed.), Political Language, O ratory and Traditional Society. Lo n don , N ew Y ork: A cadem ic Press. S outh all, A. 1956. A im Society. C am bridge.

33 Enactments of Meaning Roy A. Rappaport

Roy Rappaport (1926-97) was professor of anthropology at the University of Michigan and one of the most distin­ guished theoretical and applied anth ro ­ pologists of his generation. His w ork on the ritual regulation of a society in Papua New Guinea (1984) was a master­ piece of ecological analysis and he always retained an interest in under­ standing the w hole. W hereas Bloch em phasizes ritual's separation from - and mystification of earthly power, Rappaport, like Durk­ heim, understands ritual lying at the root of all social connection. It is through ritual, Rappaport argues, that the bases of comm itm ent and trust are forged. Although he does not mention it here, this reverses the preoccupation of earlier thinkers w ith the role of belief, as though it happened prior to action. For Rappa­ port, the act of performing ritual both brings into being moral states of affairs and entails comm itment to th e ir terms and to the order of w hich the ritual

forms a part. To engage in a marriage ceremony w hen one is eligible to do so is to get married and become obligated to live by its conditions - w hatever one's personal feelings in the matter. W hether or not the parties intend to keep their commitments, the point about ritual is that it produces them . This is an excerpt from a lengthy th eo ­ retical analysis and profound reflection on the nature o f religion and its place in the life of the human species that deserves to be read in its entirety. Rappaport's discussion of the sacred and processes of sanctification (1999: Chap­ ters 9 and 10) is especially important, and the trenchant last chapter on con­ tem porary pathologies of religion is reminiscent of the final passages of W eber (1958). A prelim inary version of the book's argum ent is to be found in Rappaport (1979, especially the essay "The Obvious Aspects of Ritual"), w hile Messer and Lambek (2001) offers some contextualization.

From Roy A. Rappaport, “ Enactments o f M eaning,” in Ritual and Religion in the Making o f Humanity (Cambridge: Cam bridge University Press, 1999), pp. 104-38. Abridged.

EN A C TM EN TS O F M EA N IN G

|...]

The Physical and th e M ean in g fu l Both rhe occurrence o f ritual and ritual’s contents form and tran sform that upon which they are im posed, b u t . . . not prim arily by force o f energy or expenditure o f m atter. W hat is often called their “ p ow er” rests upon other m eans or principles. We m ay discern in nature tw o general classes o f efficacy - the physical and the m eaningful. T he efficacy o f w hat Leach calls “ technique,” achieving as it does its results through the deploym ent o f m atter and energy in accordan ce with the law s o f physics and chem istry is, largely if not entirely, physical, but prayer is not, nor is ritual, nor are w ords. T heir efficacy is grounded in principles o f com m unication. Bateson has noted som e general differences between the tw o classes: when you enter the w orld o f com m unication, o f organ ization , etc., you leave behind the w hole w orld in which effects are brought abo u t by forces and im pacts and energy exchange. Y ou enter a w orld in which “ effects” - and I am not sure one should still use the sam e w ord - are brought abo u t by differences [bits o f inform ation). . . . The whole energy relation is different. In the world of mind nothing - that which is not can be a cause. In the hard sciences, we ask for causes and we expect them to exist and be “ real.” But remember that zero is different from one, and because zero is different from one, zero can be a cause in . . . the world of communication. The letter which you do not write can get an angry reply; the income tax form that you do not fill in can trigger the Internal Revenue boys into energetic action, because they, too, have had their breakfast, lunch, tea and dinner and can react with energy which they derive from their m etabo­ lism. The letter which never existed is no source of energy . . . what we mean by infor­ mation, the elementary unit of information, is a difference which makes a difference, and it is able to make a difference because the neural pathways along which it travels and is con­ tinually transformed are themselves provided with energy. . . .

411

Because, in possible disagreem ent with Bateson, I take inform ation to be no m ore than one form o f m eaning and because I do not think that all form s o f m eaning can be reduced to inform ation in the strict sense, I designate the class into which ritual falls as that o f “ m ean­ ingful (rather than inform ational) a c ts .” T o distinguish the m eaningful an d the physi­ cal as tw o distinct classes o f efficacy, is not to propo se that they are separate or separable in nature. T h at ritual achieves its effects through the com m unication o f m eanings d oes not im ply that it does not both consum e and m ob i­ lize energy and m aterial, nor that technique proceeds in som e m indless w ay w ithout the guidance o f m eaning. M atter-energy devoid o f inform ation is necessarily devoid o f life: in for­ m ation and other form s o f m eaning separated from m atter-energy could be conceived as m athem atical or form al ab straction , pure spirit or Platonic ideal, but the act o f conceiving it as such, being a biological process like all other acts o f conception, could not occur in the absence o f m atter and energy. Although there seem s to be no direct relationship between the m eaningfulness o f m essages and the am ount o f energy required to transm it them , the co m ­ m unication o f m eaning, both inform ational and o f higher order, alw ays requires energy and som etim es m atter as well. Even speech is propelled by energy, and inform ation is inscribed on p aper, engraved on stone, co m ­ bined in D N A and encoded on m agnetized tape. Conversely, any change detected by an organism in the energy flux to which it is su b ­ jected, or any change it detects in its m aterial environm ent conveys inform ation to it, for inform ation is, in one o f its asp ects, detected, or at least detectable, difference. M oreover, as Bateson ob serves, m atter-energy and in form a­ tional processes frequently can n ot be separated from each other because the receipt o f infor­ m ation leads organ ism s to expend their own energy to bring abo u t effects that m ay include both m aterial and in form ation al elements. M atter-energy processes and inform ational processes, although they can be distinguished, are in separable in nature, and it is o f interest that the m yths o f m any peoples, including both A ustralian aborigin es and ancient H ebrew s, describe creation, either in w hole or

412

R O Y A . RA PPA PO RT

in p art, as an act or set o f acts, im posin g form upon an already existent but inchoate p rim or­ dial m atter (Bateson 1 9 7 2 a: xxiii ff.). C reation , this is ro say, is conceived as the inform ing o f substance an d the substan tiation o f form . H igher-order m eaning com es later. Inform ation and m atter-energy processes m ay be inseparable in nature, but they, the ob jects they affect, and the w ays in which they achieve their effects m ay be distinguished. M atter and energy, it seem s clear, operate m ost effectively, that is with m ost predictable results, upon inert m aterials. The efficacy o f inform ation an d other form s o f m eaning, on the other hand, rests not only upon the ability o f senders to encode and transm it in form ation , bur upon the ability o f others to receive those m essages, that is to recognize, com prehend and take account o f them.

|...| T h at there are differences in w hat m em bers o f different hum an g rou p s take to be m eaning­ ful is, o f course, rhe raison d'etre o f co m ­ parative an th ropology. Be this as it m ay, it is ob viou s that few A m ericans w ould be inform ed o f much by a speech in M o n go lian , and the sam e m ay be said o f som e o f the non-verbal as well as verbal m essages transm itted in rituals. M o st A m erican youths m ight be deeply im pressed by being subin cised, but none, it is prob ab ly safe to say, w ould be hastened on the road to sociological m anhood by such an o p er­ ation, a s w ould youn g W albiri or A runta. The ritual form m ay be universal, but all hum an rituals include signs specific to the society, church, or co n gregation in which they are per­ form ed, and the arrangem ent o f sign elem ents into liturgical orders is a lso in som e degree socially and culturally specific. If rhe reception o f the m essages encoded in liturgical orders requires that they be in som e sense “ under­ sto o d ” (which is not to say that they m ay nor at the sam e tim e be m ysterious), then it is necessary that they be in som e w ay learned, because the un derstandin g o f signs only co n ­ ventionally related to their referents could not possibly be specified genetically. T h is is to say that particip an ts m ust be trained, in doctri­ nated or otherw ise prepared to receive the m essages rituals transm it. The w ork of C am pbell (1 9 5 9 ), Erikson (1 9 6 6 ), T urner

(1 969) W allace (1 9 6 6 ) and G ooden ou gh (1 990) suggests that the ability to be inform ed by ritual is itself establish ed in rhe individual, in part in the course o f a series o f rituals sta rt­ ing in early infancy and proceeding to m atu­ rity. Erikson (1 966) has referred to this process o f p reparation as “ ritu alization ,” and we shall return to ir in a later chapter. Here it is neces­ sary to note only that the inform ative capacity o f ritual, its ability to form and transform , rests not only upon its special m ode o f tran s­ m ission but also upon its reception by specially prepared receivers. The effects that can be achieved by the deploym ent o f m otion and energy on the one hand and m eaning on the other are a lso o b v i­ ously different. Physical effects - w eights lifted, ditches d ug, billiard balls knocked into pockets, acids neutralized, m etals sm elted, plants culti­ vated - m ust rely upon physical processes for their achievem ent. C onventional effects, on the other hand - princes transform ed into kings and w ords into prom ises, the profan e m ade sacred and truces declared - can only be achieved by m eaningful acts. W hatever energy m ay have been required by a young m an to fulfill rhe responsibilities o f knighthood, or even required ro com plete his tran sform ation to knighthood, did not flow into his shoulder from the sw ord with which he w as dubbed, nor from rhe voice o f w hom ever ir w as w ho dubbed him. It cam e, to recall Bateson, from his breakfast and dinner. D ubbin g tran sform ed him into a knight nor by the force o f its blow bur by inform ing him o f his knighthood, or better, inform ing him with knighthood. (In fact, accordin g ro the O x fo rd U nabridged D ic­ tionary, a m an is dubbed to knighthood, rhe tran sform ative asp ect o f being inform ed being em ph asized.) A lthough certain physical acts were perform ed with certain ob jects in dubbing, w hatever tran sform ation to o k place in the p assage ro knighthood w as neither a physical alteration o f the young m an, nor w as it effected by a process that can be co m p re­ hended as physical, chem ical or biological. The tran sform atio n w as, and could only be, effected by com m u nicatin g m eaning to w h om ­ ever w as the locus o f the tran sform atio n , and to other concerned person s, in accordance with the conventions o f the society in which it

EN A C TM EN TS OF M EA N IN G

w as occurring. The sam e m ay he said o f the tran sform atio n s in grou p m em bership and belligerency statu s effected by the planting o f rumbim am o n g the M arin g, and even o f those rituals in which the initiate is subjected to severe pain and by which he or she m ay be left physically m arked for life: scarified, subin­ cised, circum cized, canines rem oved, septum pierced, fingers lopped off. The significant tran sform atio n s produced by such operation s are obviously not physical, nor are they o u t­ com es o f the physical changes per se w rought by such op eration s but o f the m eanings those changes carry. Ritual acts, such as body m utilation, plan t­ ing rumbim and dubbin g are, to use Van N eu m an n ’s term , “ m ark ers,” that is “ ob serv­ able bundles, units or changes o f m atter-energy w hose patterning bears or conveys the infor­ m ational sym bols from the ensem ble” (Jam es M iller 196 5 : 164). It is, o f co urse, a m atter o f great interest that hum ans, w ho have a range o f codes an d m arkers to ch oose from , em ploy precise, subtle, energetically and m aterially inexpensive speech for the transm ission o f som e m essages and com paratively crude, expensive an d som etim es painful physical acts for the transm ission o f others. I---]

Speech A cts There are im portan t differences am on g the occurrences that have been rather casually included here in the class o f m eaningful acts. Events, such as dubbin gs, rumbim plantings an d supercision m ust be distinguished from m essages which sim ply inform receivers o f condition s in their social or physical environm ents. First, such rituals are m ore likely to inform the p articip an ts o f changes in them selves or perh ap s it w ould be better to say with changes in them selves, than they are o f changes in their surrou ndin gs. Secondly, w hereas sim ple m essages con­ cerning environm ental con dition s leave the respon ses o f receivers to their ow n devices, rituals specify their respon ses, often very pre­ cisely. W hen, in m edieval E u rope, a youth w as

413

d ub bed he w as not stripped o f his statu s to becom e anything desire or im agin ation sug­ gested to him. H e becam e a knight and nothing but a knight. When an alien g rasp s the rumbim o f a M arin g local grou p he becom es a de jure m em ber o f that local grou p , and nothing else. W hen they uproot rumbim , M arin g h usban d­ men becom e potential w arriors. When they are supercised it is t'aure'are'a , and nothing else, th at T ah itian boys becom e. T hirdly, while a m essage concerning the state o f the social or physical environm ent m ay lead us to undertake an action which will tran sform ourselves or rhe condition s su r­ roundin g us, at least som e rituals them selves com plete the tran sform ation s with which they are concerned. D ubbin g, we have observed, does not tell a youth to be a knight, nor does it tell him how to be a knight. It m akes him a knight. R itual is full o f conventional utterances and acts which achieve conventional effects. “ I dub thee to kn igh th ood,” ‘T nam e this ship the Q ueen E lizabeth ,” “ I sw ear to tell the tru th ,” “ 1 prom ise to sup p ort you in w arfare ,” “ We find the defendant guilty.” The im portance o f such utterances in the conduct o f hum an affairs is so paten t as to obviate the need to argue it, but ph ilosoph ers, in the last few decades espe­ cially, have given con siderable attention to their peculiar ch aracteristics. J. L. Austin (1 9 6 2 ) has called them “ perform ative utter­ an ces” and “ illocutionary a c ts,” J. R. Searle (1 9 6 9 ) includes them am on g w h at he calls “ speech a c ts,” F. O 'D oh erty (1 9 7 3 ) refers to an im portan t sub-class as “ factitive” acts or utterances, J. Skorupski (1 9 7 6 ) uses the term “ operative a c ts” for a class resem bling them closely. It is im portant to m ake clear that the force o f w h at 1 shall call “ perform ativ es,” follow ing A ustin ’s earlier and sim pler term inology, does not depend in any sim ple and direct w ay upon the effect o f these acts and utterances upon the m inds and hearts o f those exp o sed to them. W hether or not he has reservations, planting rumbim joins a man to the g ro u p with whom he plan ts it. R egardless o f w hat they m ay think or feel ab o u t it those w ho are excom m un i­ cated, outlaw ed, found guilty or dem oted in rituals properly conducted by authorized

414

R O Y A . RA PPA PO RT

persons are thereby excom m unicated, o u t­ law ed, m ade felonious or degraded. If a u th o ­ rized persons declare peace in a proper m anner, peace is declared whether or not the a n ta g o ­ nists are persuad ed to act accordingly. T h is is not to say that acts and utterances which are perform ative m ay not be persuasive, threaten­ ing, inspiring or otherw ise affect the receiver in w ay s inducing him to act in p articular w ays. In the language o f speech act theory they m ay have “ perlocution ary” as well as “ illocution ­ a ry ” fo rce.1 It is to say that an action having a conventional effect is com pleted in the gesture or utterance itself. Perform atives differ in the scope o f the action they com plete. If an authorized p erson, follow ing a proper procedure, nam es a ship the Q ueen Elizabeth, the ship is so nam ed. O thers m ay, if they like, call it “ H o rte n se ,” but its nam e h appens to be Queen Elizabeth, and th at’s really all there is to it. On the other hand, if a m an has danced at another g ro u p ’s kaiko, thereby prom ising to help his h o sts in w arfare, that is not all there is to it, for it rem ains for him to fulfill his pledge and he m ay fail to d o so. The nam ing, which not only constitutes an action but actually brings into being the state o f affairs with which it is con ­ cerned, is o f the class o f perform atives that we m ay call “ factive.” W hereas m any action s com pleted in ritual - dubbings, declaration s o f peace, m arriages, purification - are factive, it is ob viou s that all are not. Som e - am on g which are those that Austin called “ co m m is­ sives” (1 9 6 2 : 150ff.) - d o not bring into being the stares o f affairs with which they are con ­ cerned, but merely bring into being the com m itm ent o f those perform ing them to do so som etim e in the future.2

Th e Special R elatio n sh ip b e tw e e n Rituals an d P erfo rm ativen ess While m any liturgies are perform ative, where som e sort o f perform ative act is the m ain point o f the perform ance, tran sform in g w ar into peace, restoring purity to that which h as been polluted, joining men and w om en in w edlock,

perform ativeness is not confined to ritual. There is no advan tage to be gain ed, for instance, in takin g the pu b lican ’s utterance “ The bar is clo sed ” to be ritual, but when he says “ the bar is clo sed ” it is thereby closed, and you are not likely to get another drink. Perform atives are not confined to ritual, but there is a special relationsh ip between ritual and perform ativeness. First, the form al characteristics o f ritual enhance the chances o f success o f the p erfor­ m atives they include. Like any other acts per­ form atives can fail. If, for instance, I were to dub one o f my junior colleagues Knight o f the G arter he w ould not thereby becom e a Knight o f the G arter, even if the conduct o f the ritual were letter perfect. C onversely, if Queen Eliza­ beth dubbed Princess A nne’s horse to knight­ hood it probably w ou ldn ’t m ake him a knight, even granted the w ell-know n English tolerance o f eccentricity. And if a befuddled cleric recited the funeral liturgy rather than the m arriage service I doubt if the couple standing before him w ould thereby becom e objects o f m ourn­ ing (Austin 1962: passim ). All o f these instances o f faulty perform atives are o f ritual p erform a­ tives and ritual perform atives can misfire. The ludicrous nature o f these instances suggests, how ever, that they are less likely to do so than are other non-ritualized perform atives because the form ality o f liturgical orders helps to insure that w hatever perform atives they incorporate are perform ed by authorized people with respect to eligible person s or entities under proper circum stances in accordance with proper procedures. M oreover, the form ality o f ritual m akes very clear and explicit w hat it is that is being done. For instance, if one M arin g casually said to another w hom he happened to be visiting, “ I’ ll help you when next you g o to w ar” it w ould not be clear whether this w as to be taken as a vague statem ent o f intent, as a prediction o f w hat he w ould be likely to do, or as a prom ise, nor w ould it necessarily be clear w hat m ight be m eant by “ help.” T o dance this m essage in a ritual, how ever, m akes it clear to all concerned that a pledge to help is undertaken, and it is conventionally under­ stood that that help entails fighting. R itual, this is to say, not only ensures the correctness o f the perform ative enactm ent, but also m akes

EN A CTM EN TS OF M EA N IN G

the perform atives it carries explicit. It gener­ ally m akes them w eighty a s well. If a m essage concerning the current states o f participants is com m unicated by p articipation in ritual it will not be vague, and the form ality, solem nity and decorum o f ritual infuses w hatever perform a­ tives the ritual in corporates with a gravity that rhey otherw ise m ight not possess. In sum , sim ply by m aking their perform atives explicit rituals m ake clear to their p articipants just w hat it is that they are d oin g and thus they specify, at least tacitly, w hat w ould constitute an abuse or violation o f w hatever ob ligation s those perform atives entail. C lear definition itself may reduce the likelihood o f abuses and violations sim ply by leading people to “ think tw ice” before acting. T o put this a little differ­ ently, clear definition, which is intrinsic to the form ality o f ritual, itself p ossesses perlocutionary force, and so do the gravity, solem nity and decorum characteristic o f m any rituals. Reflexively, the perlocutionary force inhering in the form ality o f a ritual su p p orts whatever perform atives are enacted in that ritual. There are tw o other closely related reasons for considering the perform ativeness o f ritual. First, the associatio n o f the sacred and occult with perform atives in m agical and religious rituals m ay m ystify their conventional nature, and this m ay enhance their chances o f success. T o take the state o f affairs established by a king’s enthronem ent to derive from the sacra­ m ental virtue o f crow n and chrism may be m ore effective with respect to the m aintenance o f the social order over which the king reigns than w ould be the recognition o f enthrone­ ment as a naked perform ative, a mere conven­ tional act, the effects o f which could, for instance, in response to the short-run d isplea­ sure o f his subjects, easily be reversed. Second, as Ruth Finnegan (1 9 6 9 : 50) has suggested, albeit rather unspecifically, the “ truth lying behind” assu m p tion s concerning w hat is often called “ the m agical pow er o f w o rd s” m ay be related to their illocutionary force or perform ativeness. It m ay be proposed, rather m ore specifically, that the m agical pow er o f som e o f the w ords and acts form ing parts o f liturgies derives from the relationships between them and the conventional states o f a ffairs with which they are concerned. As we

415

have already observed, the relationship o f per­ form atives to the states o f affairs with which they are concerned is the inverse o f that o f statem ents. The facts, events or situations to which a statem ent refers presum ably exist independent o f and previous to the statem ent referring to them , and a statem ent is assessed true if it accords in som e sufficient degree to those previously existin g and independent states o f affairs. Since perform atives bring ab o u t the facts, events, situations, etc., with which they are concerned, these facts are su b ­ sequent to and contingent upon them. Perfor­ m atives, and m ost un am biguous factives, are self-fulfilling: they m ake them selves true in the sense o f stan din g in a relationship o f co n fo r­ m ity to the states o f affairs with which they are concerned. In light o f this it m ay be suggested that the perform ativeness, and more especially the factiveness o f ritual acts and utterances provide a basis for occult efficacy in general, including the m agical pow er o f w ords in particular. R itu al’s w ords d o, after all, bring conventional states o f affairs, or “ institutional fa c ts” into being, and having been brought into being they are a s real as “ brute fa cts” (Searle 1969: chapter 2 , passim ). It m ay also be that m agical pow er is attributed to other w ords by extension o f the principle o f factive­ ness beyond the dom ain o f the m eaningful, in which it is clearly effective, into the physical, in which it is not, but we must be very careful ab o u t stipulatin g the lim its o f the effects ritual can accom plish. We have already argued in som ew h at different term s that their illocution­ ary force m ay be augm ented by perlocutionary force. The effectiveness o f persuasion, threat, cajolery, inspiration and ecstasy m ay well thrust beyond the purely conventional, and beyond discursive con sciousness into the organ ic, as in the cures o f healing rituals and in the injuries o f ensorcellm ent.

R itual's First Fu n d am en tal O ffice Perform atives are not confined to ritual, and there seem s to be m ore to som e or even all liturgies than the perform atives that they

416

R O Y A . R A PPA PO RT

incorporate. Indeed, som e liturgies m ay not seem to include perform atives in any sim ple sense at all. M an y religious rituals d o nor seem to be directed tow ard achieving sim ple co n ­ ventional effects through conventional p ro ce­ dures. If, how ever, sim ple perform atives is not criteria! o f ritual, som ething like it, but o f higher order, m ay be. A lthough not all rituals are obviously and sim ply perform ative, perform ativeness irself m ay be m ade possible by ritual. We ap p roach here the conjunction o f form ality and perform ance | . . . ) . We com e, this is to say, to w hat is intrinsic to the act o f performing a liturgical order, and thus to the heart o f the relation sh ip between the selfreferential and can on ical. T he ch aracteristic o f liturgical orders salient here is the sim ple fact o f their perform ance; that they m ust be perform ed. W ithout p erfo r­ m ance, there is no ritual, no liturgical order. R ecords or description s o f liturgies perform ed in Ur and T hebes survive but they are merely about liturgies, not them selves liturgies. They are rem ains o f the d ead , for the liturgical orders they recall are no longer given life and voice by the bodies and breath o f men. Perfor­ m ance is not merely one w ay to present or exp ress liturgical ord ers but is itself a crucial aspect or com pon en t o f the m essages those orders carry. The follow ing m ay seem invo­ luted. The involution is intrinsic to the phenom enon, and not to my accoun t o f it. A liturgical order is a sequence o f form al acts and utterances, and as such it is realized m ade real, m ade into a res - only when those acts are perform ed an d those utterances voiced. T his relationship o f the act o f perform ance to that which is being perform ed - that it brings it into being - can n ot help but specify a s well the relationship o f the perform er to that which he is perform ing. H e is not merely transm itting m essages he finds encoded in the liturgy. He is participatin g in - that is, becoming part o f the order to which his own body and breath give life. T o perform a liturgical order, which is by definition a m ore or less invariant sequence o f form al acts and utterances encoded by someone other than the perform er him self, is necessarily to conform to it. A uthority or directives, therefore, seem intrinsic to liturgical order (see

M aurice Bloch 1973). The account just offered suggests, how ever, som ething m ore intim ate and perhaps m ore binding than w hatever is connoted by term s like “ au th o rity” and “ con­ form ity.” The general notion o f com m u n ica­ tion m inim ally im plies transm itters, receivers, m essages, and channels through which m es­ sages are carried from transm itters to receiv­ ers. Som etim es, furtherm ore, as in the case of canonical m essages, which ritual’s perform ers find already inscribed in prayer b o ok s or pre­ scribed by tradition , transm itters should be distinguished from encoders, w hose identities m ay be lost in time and w hose dicta are, in part for that very reason , tim eless. We earlier noted a peculiarity o f ritual com m unication, namely that in ritual the transm itter and receiver are often one and the sam e. At least the transm itter is alw ays am o n g the m ost im portant receivers. Now' we note an oth er o f ritual’s peculiarities. To say that performers

participate in or become parts o f the orders they are realizing is to say that transmitterreceivers become fused with the m essages they are transmitting an d receiving . In conforming to the orders that their performances bring into being , and that come alive in their perfor­ mance, performers become indistinguishable from those orders , parts o f them, for the time being. Since this is the case, for performers to reject liturgical orders being realized by their own participation in them as they are partici­ pating in them is self-contradictory , and thus impossible. Therefore, by performing liturgical order the participants accept , and indicate to themselves and to others that they accept whatever is encoded in the canon o f that order. T his act o f acceptance is the first o f ritu al’s fundam ental offices. The self-referential and the canonical are united in the acceptance o f the canon. A cceptance is the self-referential m essage intrinsic to all liturgical perform ances, the indexical m essage w ithout which liturgical orders and the canonical m essages they encode w ould be w ithout consequence, non-existent, or vacuous. It is not a trivial m essage because hum ans are not bound to acceptance o f p a r­ ticular conventional orders by their genotypes. They are often free nor to participate in rituals if they do not care to, and refusal to participate

EN A C TM EN TS OF M EA N IN G

is alw ays a possibility, at least logically co n ­ ceivable, by potential acto rs. Participation, and thus acceptance, alw ays rests in logic and in som e degree in fact, upon choice. Such choices m ay som etim es be extrem ely costly, but are alw ays possible.

A ccep tan ce, Belief, and C o n fo rm ity The assertion that acceptance is intrinsic to liturgical perform ance m ay still seem to be either d u b io u s or in dubitable. It therefore requires som e elaboration an d clarification in this section and the next. First, acceptance is not belief. The concept o f belief is difficult to define and the occu r­ rence o f belief difficult to establish (see R. N eed h am 1972). Let us say that the term “ belief” at least suggests a m ental state co n ­ cerning, or arising out o f, the relationsh ip between the cognitive processes o f individuals and representations presented to them as p o s ­ sible can did ates for the statu s o f true. A s such, “ belief” is a second-order process, that is, one concerned with the relationsh ip between a first order process and external reality. By this accoun t, belief is an inw ard state, know able subjectively if at all, and it w ould be entirely unw arranted either for us or for participants or w itnesses to assum e that participation in a ritual w ould necessarily indicate such a state. Acceptance , in contrast , is not a private state , but a public act , visible both to w itnesses an d to perform ers them selves. People m ay accept because they believe, but acceptance nor only is not itself belief; it d oesn ’t even im ply belief. R itual perform ance often possesses per­ locution ary force, and the private p rocesses o f individuals m ay often be persuaded by their ritual participation to com e into conform ity with their public acts, but this is not alw ays the case. Belief is a cogent reason , but far from the only reason , for acceptance. Conversely, belief can provide groun d s for refusals to accept. R eform ers and heretics, for the very reason th at they believe deeply in certain p o s­ tu lates concerning the divine, m ay refuse to particip ate in the rituals o f religious institu­

417

tions they take to have fallen into error or corruption . T h is accoun t suggests that although partici­ pation in liturgical perform ance m ay be highly visible it is not very p ro fo u n d , for it neither indicates nor does it necessarily produce an inw ard state conform in g to it. Such a view is widely held by critics o f religion w ho are inclined to take ritual p articipation to be nothing m ore than em pty or even hypocritical form alism , a view reflected in one o f the term ’s com m on m odern m eanings, form al behaviour devoid o f substan ce or consequence. But, p a ra ­ doxically, it m ay be, and it surely has been im plied by religion’s defenders, that the acceptance indicated by liturgical perform ance, being independent o f belief can be m ore pro­ found than conviction or sense o f certainty, for it m akes it possible for the perform er to tran­ scend his or her ow n d o u b t by accepting in defiance o f it. Even the m ost devout, indeed especially the m ost devout, som etim es harbour dou b ts or even voice scepticism concerning propo sitio n s expressed in liturgies to which they scrupulously con form , and acceptance in this deep sense has much in com m on with certain C h ristian notions o f faith. Fehean O ’ D oherty, a C ath olic priest, w rites “ faith is neither subjective conviction nor experienced certitude, but m ay be at its best where doubt e x ists” (1 9 7 3 : 9), and Paul Tillich has said that faith necessarily includes an elem ent o f uncer­ tainty or dou bt (1 9 5 7 : 16 ff.). It is also o f interest in this regard that Ju d a ism does not require the devout to believe, for belief is not subject to com m and. It does, how ever, dem and o f them that they accept rhe law , and this acceptance is signalled by, an d is intrinsic to, conform ity to the ritual observan ces that pervade all o f life. Be this as it m ay, there m ay well be, and often are, disparities between the act o f accep­ tance and the inw ard state asso ciated with it. O ne can accept publicly not only that which one d o u b ts bur that which one privately despises or secretly denies. But if acceptance is intrinsic to perform ance it is not vitiated by secret denial. T o recognize th at secret denial m ay hide beneath the acceptance inhering in the act o f perform ance is to recognize that the grou n d s o f acceptance m ay vary w idely, that

418

R O Y A . RA PPA PO RT

acceptance is not necessarily founded upon belief, and that it does not even necessarily imply the subjective state term ed “ a p p ro v a l.” A cceptance, then, can be unconvinced and “ insincere,” but insincerity does not nullify acceptance. In w hat ap p ears to be a flaw o f sufficient seriousn ess to vitiate its m ean in gful­ ness lies the very virtue o f acceptance through liturgical perform ance. Its social efficacy lies in its very lack o f profundity, in the possibility o f disparity between the outw ard act an d the inw ard state. T he distinction between belief and acceptan ce corresp on d s to the distinction m ade in the third chapter between the public and private. Participation in ritual dem arcates a boun dary, so to sp eak , between private and public processes. Liturgical orders, even those perform ed in solitude, are public orders and participation in them constitutes an acceptance o f a public order regardless o f the private state o f belief o f the perform er. We m ay cite here A ustin’s view s on a great range o f p erfo rm a­ tives - prom ising, sw earing, repudiating, co m ­ m ending, assessin g - am on g which accepting is to be included: we must not su p p o se . . . that what is needed in addition to the saying of the w ords in such cases is the performance of some internal spiritual act, o f which the words are then the report. It’s very easy to slip into this view at least in difficult portentous cases . . . In the case o f promising - for example, “ I promise to be there tom orrow ” - it’s very easy to think that the utterance is simply the outward and visible . . . sign of the performance of some inward spiritual act of promising, and this view has been expressed in many classic places. There is the case o f Euripedes’ Hippolytus who said “ My tongue swore to, but my heart did not” - perhaps it should be “ m ind” or “ spirit” rather than “ heart,” but at any rate some kind of backstage artiste. (1970: 236) It is gratifying to observe in this very exam p le [that o f H ippolytus) how excess o f profundity, or rather solem nity, at once paves the w ay for im m orality. For one w ho says “ p rom isin g is not merely a m atter o f uttering w ords! It is an inw ard and spiritual a c t!” is ap t to a p p e a r as a solid m oralist stan din g out again st a gen era­

tion o f theorizers: w e see him a s he sees him self, surveying the invisible depth s o f ethical space, with all the distinction o f a specialist in the sui generis. Yet he provides H ippolytus with a let-out, the bigam ist with an excuse for his “ I d o ” and the w elsher with a defence for his “ I bet.” A ccuracy and m orality alike are on the side o f the plain saying that our w ord is our bond (Austin 1962: 10). A cceptance in, or through, liturgical p erfor­ mance m ay reflect an inw ard state o f convic­ tion; it m ay also encourage “ the m ind,” “ the heart” an d “ the sp irit” into agreem ent with itself. It does not necessarily d o either, how ever, and therefore it does not elim inate all o f the shenanigans o f which the m ind, the heart, the spirit, and other “ b ack stage a rtistes” m ay be capable, but my argu m ent, based on A ustin’s, prop o ses that although liturgical perform ance does not elim inate insincerity, it renders it publicly im potent. It is the visible, explicit, public act o f acceptance, an d not the invisible, am b iguo us, private sentim ent, which is socially and m orally binding. Because public and private processes are (and m ust be) related, but only loosely related, a range o f w hat Austin (1 9 6 2 : 9 5ff., passim) called “ infelicities” - insincerities and the like - are possible. But if, som eh ow , public orders could be required to depend upon the continu­ ing belief, sincerity, goodw ill, conviction or enthusiasm o f those subject to them , the p o s­ sibility o f insincerity or deceit w ould surely be replaced by the high probability o f non-order or disorder because o f the near im possibility o f m eeting such a stan dard. This is nor to say that the private processes may not be im por­ tant in the dynam ics o f ritual. In a later chapter we shall take up belief and religious experi­ ence. It is sim ply to recognize that the private states o f others are in their nature unknow able and even one’s ow n attitu des m ay not alw ays be easy to ascertain , for w e are inclined to be am bivalent ab o u t m atters o f im portance, like the conventions to which we are subordinate, and private states are likely to be volatile. “ C om m on belief” cannot in itself provide a sufficiently firm groun d upon which to e stab ­ lish public orders, even in very sim ple societies. We cannot know if a belief is com m on, for one thing, and w hereas belief is vexed by

EN A CTM EN TS O F M EA N IN G

am bivalence and clouded by am biguity, accep ­ tance is not. Liturgical orders are public, and p articipation in them constitutes a public acceptance o f a public order, regardless o f the private state o f belief. A cceptance is not only public but clear. One either participates in a liturgy or one does not; the choice is binary and a s such it is form ally free o f am biguity. While ritual participation m ay not transform the private state o f the perform er from one o f “ disbelief” to “ belief,” ou r argum ent is that in it the am biguity, am bivalence and volatility o f the private processes are subordin ated to a sim ple and un am biguous public act, sensible both to the perform ers them selves and to w it­ nesses as well. Liturgical perform ance is, thus, a fundam ental social act, for the acceptance intrinsic to it form s a basis for public orders which unknow able and volatile belief or con ­ viction cannot. T h at a liturgical order is accepted in its perfor­ m ance d oes not, furtherm ore, guaran tee that the perform er will abide by w hatever rules or norm s that order encodes. We all know that a m an m ay participate in a liturgy in which co m ­ m andm ents again st adultery and thievery are pron oun ced, then pilfer from the p o o r box on his w ay ou t o f church, or depart from com m u­ nion to tryst with his neighbor’s wife. T o recog­ nize such sordid realities is not to agree that liturgical acceptance is hypocritical, trivial or m eaningless (D ouglas 1973: 3 0 ), nor is it to dism iss claim s for the social efficacy o f accep ­ tance through liturgical perform ance. It is, in fact, to affirm them , for such violations do not nullify acceptance, nor render it trivial. It is in such instances that the im portance o f ritual acceptance is m ost dram atically dem onstrated. The prim ary function or m etafunction o f litur­ gical perform ances is not to control behavior directly, but rather to establish conventional understandings, rules and norm s in accordance with which everyday behaviour is supposed to proceed. Participation in a ritual in which a prohibition again st adultery is enunciated by, am on g others, him self m ay not prevent a m an from com m itting adultery, but it does establish for him the prohibition o f adultery a s a rule that he him self has both enlivened and accepted. Whether or not he abides by that rule , he has

419

obligated him self to do so. If he d oes not, he has violated an obligation that he him self has avow ed. The assertion here is sim ilar to those o f A ustin (see above) and o f the philosopher Jo h n Searle, w ho has argued that when one enters an institutional activity by invoking the rules o f that institution one nec­ essarily commits oneself in such and such ways, regardless o f whether one approves or disapproves o f the institution. In the ease of linguistic institutions like promising [and accepting] the serious utterance o f words commits one in ways which are determined by the meaning of the words. In certain first person utterances the utterance is the under­ taking o f an obligation. (1969: 189) Searle later notes that the notion o f o b liga­ tion is closely related to those o f accepting, recognizing, acknow ledging. T h is suggests that there is no ob ligation w ithout acceptance, and perh aps that m orality begins with accep­ tance. We m ay also note that while the accep­ tance o f conventional undertakings, rules and procedures is possible outside o f ritual, the form al and public nature o f liturgical perfor­ m ance m akes it very clear that an act o f accep­ tance is taking place, that the acceptance is serious, and w hat it is that is being accepted. In A ustin’ s term s (1962: passim ) it is “ explic­ itly p erform ativ e.” In sum , it is not ritual's

office to ensure compliance but to establish obligation .

P erfo rm ativen ess, M etap e rfo rm a tive n ess, an d th e Estab lish m en t of C o n ven tio n We m ay now return to the assertion that although all ritual m ay not include sim ple per­ form atives - conventional procedures for achieving conventional effects - som ething fo r­ m ally sim ilar to sim ple perform ativeness, but o f higher order, is intrinsic to ritual’s form , and that this characteristic o f ritual m akes per­ form atives possible. Austin (1 9 6 2 : 26ff.) listed six conditions that m ust be fulfilled if perform atives are to be

420

R O Y A . R A PPA PO RT

successful (see a lso Searle 1969: passim ). These include a num ber o f ob vious stipu lation s already noted at least tacitly - that they be perform ed by properly authorized person s under proper circum stances, and that they be executed correctly and com pletely. We have observed that the form ality o f ritual goes a long w ay to assu re that those con dition s are met, but ritual’s contribution to perform ative­ ness is not limited to its service as a p rotocol, conform ity to which assu res full and correct perform ance. Its significance is m uch m ore fundam ental. A ustin states the first and m ost basic co n d i­ tion for perform ative success, that which he labels A .l, as follow s: “ There m ust exist an accepted conventional procedure having a certain conventional effect, the procedure to include the uttering o f certain w ord s [or the perform ance o f certain sym bolic acts] by certain person s in certain circum stan ces” (19 6 2 : 14) (em phasis mine). Conventional effects can n ot be achieved w ithout conven­ tions for achieving them. If young men are to be tran sform ed into knights there m ust be a procedure for d oin g so , and this procedure m ust be acceptable to the relevant public. We m ay also note, although A ustin does not, that the acceptan ce o f a procedure for dubbin g knights tacitly but obviously entails an acceptan ce o f the convention o f knighthood itself. Yet further, if the youn g man is “ arm ed as a knight for the service o f C h rist” by priests or bish ops, an acceptance o f C h rist’s divinity is also entailed (M arc Bloch 1961: ch. 33). A ustin’s basic con dition seem s o b viou s, but it is not trivial because its violation is possible. A perform ative attem pt could misfire because, for instance, no one but the perform er recog­ nized the procedure which it em ployed o r the state o f affairs it sough t to achieve. An attem pt by an A m erican citizen living in rhe United States to win a divorce by repeating to his wife “ I divorce y o u ” three times w ould not rid him o f m atrim on y’s burdens, nor w ould any divorce procedure w hatsoever succeed in a society not recognizing divorce at all (Austin 1962: 27). C onventional procedures and con ­ ventional states, or even entire conventional cod es, m ay be accepted by som e and not by

others and there are surely ch anges in the co n ­ ventions o f any society through time, with the scope o f their applicability expan din g or co n ­ tracting, som e d isap p earin g altogether while others ap p ear. C on ven tion s can cease to exist because they are no longer accepted (Austin 1962: 3 0 ), as, for instance, in the case o f the code o f honor o f which duelling w as a part. It is unlikely, to say the least, that in contem po­ rary United States o r G reat Britain a slap o f a glove across a cheek w ould lead to a duel. The conventions o f which this ritual act w as an element are no longer accepted. They are, as we say, “ h istory .” Austin stipulated a s requisite to the effec­ tiveness o f perform atives th at relevant conven­ tions exist and be accepted, but he gave only scant attention to the w ays in which this pre­ requisite m ight be fulfilled. The argum ent being presented here is that ritual m ay fulfill it. T o establish a convention - a general public understanding, a regular procedure, an institu­ tion - is both to ascribe existence to it and to accept it. The tw o are hardly distinct, as Austin (1962: 26) understood, for the existence o f a convention, given rhe m eaning o f the w ord, is a function o f its acceptance (see Bateson 1951: 2 1 2 ff.). T o perform a liturgy is at one and the sam e time to conform to its order and to realize it or m ake it substan tial. Liturgical per­

form ance not only recognizes the authority o f the conventions it represents , it gives them their very existence. In the absence o f p erfor­ m ance liturgical orders are dead letters inscribed in cu riou s volum es, or insubstantial form s evap oratin g into the forgotten. A ritual perform ance is an instance o f the conventional order o f which it con form s. Conversely, a ritual perform ance realizes the order o f which it is an instance. Participan ts enliven the order that they are perform ing with the energy o f their ow n bodies, and their ow n voices m ake it articulate. They thereby establish the e x is­ tence o f that order in this w orld o f m atter and energy; they substantiate the order as it informs them. If perform atives are to be understood as conventional procedures for achieving conven­ tional effects, rituals are, by this account, m ore than sim ple perform atives. We have already noted that a dubbin g did m ore than transform

EN A CTM EN TS O F M EA N IN G

a particular young m an into a knight; it also repeatedly established (accepted the existence of) a conventional procedure for tran sform in g youn g men into knights. It further established an d re-established the conventions o f knight­ hood itself, and o f the divinity o f the god in w hose nam e and service knights were dubbed. The m ass, in con trast, establish es a more general conventional understanding o f the relationship o f hum ans to the divine. The act o f acceptance intrinsic to ritual perform ance is not sim ply perform ative, as are specific conventional acts occurrin g within rituals crow ning, m arrying, d ubbin g, purifying - but meta-performative. R itu als d o m ore than achieve conventional effects through conven­ tional procedures. They establish the conven­ tions in term s o f which those effects are achieved. T he establishm ent o f convention is the second o f ritual’s fundam ental offices. It is fun­ dam ental because all o f ritu al’s sim ple perfor­ m ative functions are founded upon, o r presum e it, and so m ay conventional p rocedures outside o f ritual itself. It is fundam ental, this is to say, because the establishm ent o f convention is w h at might be called a “ m etafun ction ” m aking possible the fulfillment o f p articular functions by the particular conventions established. It is fundam ental in a secon d , form al sense, because the establishm ent as convention o f w hatever is encoded in can on is intrinsic to the form o f ritual, that is, to the perform ance o f m ore or less invariant sequences o f form al acts an d utterances not encoded by the perform er. We observe here the profoun d im portance o f invariance and form ality. These are the fea­ tures that m aintain co n stan t that which is accepted. In the absence o f such constancy that w hich is accepted w ould not be conventional. Indeed, acceptance w ould be inconsequential, m eaningless, or even logically im possible if the can on were m ade up afresh by each partici­ p an t for each perform ance. We note in p assin g that as one o f ‘‘ the reali­ ties” lying behind notions o f the m agical pow er o f w ords m ay be sim ple perform ativeness or factiveness m ystified, so m ay w idespread n otion s concerning the creative pow er o f The Word rest upon m eta-perform ativeness or m eta-factiveness m ystified, upon the realiza­

421

tion o f conventions through participation in invariant liturgical orders.

R itual an d D aily Practice in th e Esta b lish m en t o f C o n ven tio n T h a t the ob ligation s clearly and explicitly accepred in liturgical perform ance are nullified by neither disbelief nor violation has a signifi­ cance transcending the problem s o f insincerity and deceitfulness. We approach here a m atter o f profo u n d im portance, the relationship of convention to behavior an d, m ore particularly, difficulties in establish ing convention through ordin ary usage. I have argued that one o f rit­ u al’s fundam ental offices is the establishm ent o f convention, bur no claim w as m ade thar convention is established only in ritual. It therefore m ay seem in this regard that ritual is no different from usage or practice in general. C on vention , be ir noted, may a lso be e stab ­ lished by decree. There are im portan t differ­ ences, how ever, between liturgical perform ance and other m eans for establish ing convention. First, we m ay con trast liturgical perfor­ m ance with quotidian practice. As Bateson (1 9 5 1 : 214) long ago rem arked, “ every state­ m ent in a given codification is an affirm ation o f rhat codification and is therefore in som e degree m etacom m unicarive (when I say i see the c a t’ I am im plicitly affirm ing the p ro p o si­ tion that the w ord ‘c a t’ stands for w hat I see” ). The core m eaning o f the term “ co d e ” is lin­ guistic, in som e u sages it denotes vocabu laries an d the rules for com bining their elem ents into larger m eaningful units without reference to or

restriction upon what can or may be said. W hile linguistic conventions m ay be taken to be p arad igm atic o f those established in o rd i­ nary practice, it should be kept in mind that the concept o f code has been extended beyond language by som e cognitive an th rop ologists, w ho tend to see cultures generally as com plex codes m ade up o f “ shared finite cognitive set|s| o f rules for the socially ap propriate co n stru c­ tion an d interpretation o f m essages and beh av­ io u r” (Kernan 1972: 3 3 3 , cf. Frake 1 9 64). But a liturgy is not a code in this wide an d sem an ti­ cally unspecifying sense. It is a m ore or less

422

RO Y A . RA PPA PO RT

fixed sequence o f stereotyped action s and utterances and as such w hat can be expressed in it is n arrow ly circum scribed. Permissible variation s in som e aspects o f perform ance do allow or require the participants to encode indexical m essages, but the content o f the canon, the invariant aspect o f the liturgy in respect to which the indexical m essages m ay vary, is fixed, and therefore the range o f index­ ical m essages that m ay be transm itted in any liturgy is restricted. M oreover, a s we have noted, the p articipants d o not encode but only transm it the invariant m essages the canon em bodies. Since this is the case, the term “ litur­ gical order ” seem s m ore ap p rop riate than “ liturgical code.” It follow s that the accep­ tance o f an ord er, because it is in its nature highly restrictive, is therefore m ore socially consequential and significant than the affirm a­ tion o f a m ore or less unrestrictive code. High valuation o f the qualities o f the consequential and the restrictive, as previous discussion su g ­ gests, invites the application o f liturgical order, for liturgy tends to m ake explicit precisely w hat is being stipu lated, it is in its very form constricting, and further, liturgical form and decorum tends to m ake its substan ce seem grave. It w ould, o f course, be m istaken to im pose the sim ple dich otom ous distinction o f “ co d es” versus “ o rd e rs” upon the conventions organ iz­ ing social life. Linguistic codes and the conven­ tional d ogm atic understandings em bodied in som e liturgies stan d at o p p o site ends o f a co n ­ tinuum o f con strain t. Between them are stretched the conventions which organize not merely w hat people say to or ab o u t each other or the w orld , but w hat they d o to and with each other an d the w orld aroun d them. Little is know n a b o u t the order o f this continuum , whether certain dom ain s o f culture are more likely to be subject to narrow er or m ore rig o r­ ou s con strain ts than others, or w hether v aria­ tion from one society to the next is wide. We shall return to this general question in a later chapter when we discuss sanctification. For now it m ay be noted that com paratively high degrees o f con strain t seem entailed by the invariant nature o f liturgy, and it m ay further be suggested that the m ore highly m otivated people are to violate a convention or the m ore

consequential its violation is deem ed to be, the m ore likely it is to be established in liturgy than in daily practice, or the m ore closely and strongly will it be associated with conventional understandings that are so represented (see the discussion o f sanctification in chapter 10 below). Be this as it m ay, certain conventions, for instance, those o f speech, emerge ou t o f ordinary usage and are m aintained by o rd i­ nary usage in sufficient stability to allow m eaningful and orderly social interaction. In such cases, “ the norm is identical with the statistical av erag e” (Leach 1972: 320). It m ay be suggested that variation with respect to such conventions can be com fortably tolerated and day-to-day usage m ay be allow ed to e stab ­ lish, m aintain or change them. But ordinary practice or usage is not in itself sufficient to establish all conventions, nor are statistical averages arisin g ou t o f behavior alw ays c o e x ­ tensive with conventions. First, it is im possible for ordinary usage to establish conventions to which no ordinary usage corresp on ds. Such conventions include m ost im portantly, and perhaps exclusively, the understandings upon which religions are founded, d o g m as and m ysteries concerning g ods and the like w hich, being typically w ithout m aterial referents and alw ays being taken to be extraordin ary, can n ot grow out o f ordinary usage. The fundam ental im portance o f these conventions will be discussed in later chapters. Secondly, statistical averages arising out o f usage represent no m ore than com m on p rac­ tice, sum m ation s o f behavior, and the utility o f sum m ations o f behavior for guiding or assessin g the behavior o f which they are sum ­ m ations is lim ited, at best. They are p articu ­ larly in adequ ate, first, in the case o f conventions with m oral im port, for they tend to reduce the notion o f im m orality to deviation from a sta ­ tistically average behavioral range - to that which “ is not d o n e ,” and the m oral to that which “ is d o n e .” Such a notion o f m orality and im m orality is, on the one hand, descrip ­ tively erroneous - no society operates with such a conception o f m orality - and, on the other hand, operatively inadequate. N o society could so operate, at least for lon g, because it w ould be w ithout any m eans for assessin g

EN A CTM EN TS OF M EA N IN G

com m on practice itself, and a com m on p rac­ tice m ay, even in term s o f the m oral code o f the society in which it occurs, be vicious, an ti­ social or self-destructive. C om m on practice, statistical average, ordinary usage will have difficulty establishing conventions concerning asp ects o f social life th at are obviously re­ strictive, obviously arbitrary, highly charged em otionally, especially dan gerou s, or require obedience conventions, this is to say, that dem and o f individuals that they subordinate their self-interests to the com m on goo d . Behav­ ioral variation may be less tolerable with respect to these m atters than with respect to linguistic usage, and uncertainty as to the precise nature o f the conventions them selves, a different m atter, m ay be even less tolerable than variation s in the practices which they pre­ sum ably direct. O rdinary usage alw ays varies, and in ordinary usage rules and conventions are frequently violated. Leach w as generally pointing in the right direction but did not go far enough when he suggested that “ if anarchy is ro be avoided , the individuals w ho m ake up a society m ust from time ro time be reminded o f the underlying order that is supposed to guide their social activities. Ritual perfor­ m ances have this function for the grou p as a w hole. They m om entarily m ake explicit w hat is otherw ise fiction” (1 9 5 4 : 16). Although usage m ay not be faithful to it, that which is represented in a liturgical order is not a fiction, (except in the sense o f being “ m ade u p ,” rather than given by non-hum an nature, see Geertz 1973: 15), and the perform ance does m ore than rem ind individuals o f an underlying order. it is well to m ake explicit an assum ption tacit in ou r general argum ent. The orders o f societies, like the order o f the universe in general, tend to degenerate into disorder. Their m aterial elem ents disintegrate or decay into non-functioning fragm ents if they are not m aintained, and their m eaningful elements, including conventional understandings and rules, dissolve into error, nonsense, am biguity, vagueness, hypocrisy and m eaninglessness unless continually clarified, corrected and re­ establish ed. Far from clarifying and reasserting conventions, the vagaries o f practice m ay tend to erode them. It is therefore necessary to

423

estab lish at least som e conventions in a m anner which protects them from dissolu tion in the variation s o f day-to-day behavior and the vio­ lation s in which history abo u n ds. Liturgy does not sim ply remind people o f the orders which usage - behavior and history - violates and dissolves. It establishes and ever again reestab­ lishes those orders. Liturgy preserves the con­

ventions it encodes inviolate in defiance o f the vagaries o f ordinary practice , thereby provid­ ing them with existence independent o f and insulated against , the statistical averages which characterize behavior. T h at “ everyone docs ir” exon erates no one. For people o f the Book adultery w ould rem ain a sin even if every m arried person indulged in it. It should be noted, how ever, th at the v iola­ tion o f convention is not alw ays sim ply a m atter o f enrropy, ch aos or anarchy asserting itself again st an ideal but not fully realized order. The constitution o f som e societies is such th at the violation o f som e conventions is not only frequent but system atic, and yet the convention has a vital part to play in the life o f the society. G luckm an long a g o (1954) con ­ sidered certain A frican rituals in this light, and we m ay note that am on g the M arin g and other H ighland N ew Guinea peoples, a strong patrilocal patrilineal ideology prevails. It is putatively patrilineal clans that hold territories and putatively patrilineal sub-clans that claim sm aller tracts (R ap p ap o rt 19 6 8 : ch. 2). R ituals, addressed largely to patrilineal an cestors, are conducted by these grou p s at special places on the land which they and their deceased ances­ tors are said to occupy together. But the exi­ gencies o f life and death are such that the dem ograph ic fortunes o f these sm all groups (sub-clans am on g the T sem b aga ran, in 1963, from alm o st none to ab o u t thirty-five persons, clans from sixteen to seventy persons), fluctu­ ate w idely, and fluctuations m ay lead to the violation o f the patrilineal patrilocal ideal. G ro u p s m ust m aintain their strength vis-d-vis their neighbors, and when their num bers are low the m em bers o f a group will attem pt to attract outsiders to settle am o n g them. Their kinship term inology, which is Iroquois on e go ’s generation, but generational on all descending generations, as well as on the second ascending generation and above, seem s

424

R O Y A . RA PPA PO R T

well suited to the assim ilation o f strangers, obliteratin g a s it d oes distinctions between agn ates, other co gn ates and affines in tw o gen­ eration s, and rumbim planting can be seen as the beginning o f a process by which cogn ates are tran sform ed into agn ates. This ritual tran s­ form ation o f non-agn ates into agn ates is able to preserve the conventions o f patrilocality and patrilineality, if not patrifiliation, inviolate in the face o f continual violation in usage (see LiPum a 1990). T his is o f con siderable a d a p ­ tive im portance. D ensities som etim es becom e high in the N ew G uinea highlands and there­ fore it som etim es becom es necessary or d esir­ able to exclude people w ho w ould like to im m igrate (M eggitt 1965). A gnatic rules, nec­ essarily violated by practice but preserved by ritual, provide a basis for such exclusion when there is need. The form ality definitive o f ritual and distin ­ guishing it from ordinary behavior is clearly o f im portance in preserving the conventions it encodes from the errors and trespasses o f daily practice. Because preservation is virtually entailed by ritual’s form ality, and because the acceptance as well as the precise stipulation o f convention is intrinsic to ritual’s form , ritual may well be w ithout functional, or m etafunc­ tional, equivalents. R itual, to be sure, is not altogether unique in establish in g conventions at the sam e time that it insulates them from the variation s and violations o f behavior. C onventions m ay also be prom ulgated by decree and m aintained by force. But the acceptance o f those subject to a decree is not intrinsic to rhe prom ulgation o f that decree. In co n trast, it is one and the sam e ritual act th at both realizes and accepts a litur­ gical order. Furtherm ore, the act o f acceptance establishes an ob ligation with respect to the convention accepted, an ob ligation that is not specifically undertaken and m ay not be felt by those subject to decrees. It is o f interest in this regard that even in those instances in which conventions are self­ con sciously prom ulgated by kings or p arlia­ m ents the act o f p rom ulgation and those particip atin g in it are surrounded by ritual. K ings are crow n ed, public officers sw orn into office, m eetings o f parliam en ts are cerem oni­ ally open ed an d closed, and their deliberations

set w ithin a m ore or less invariant procedure. M oreover, their decrees m ay be accepted, albeit indirectly and non-specifically, by those subject to them in such ritual acts as pledges o f allegiance to the entities, or sym bols thereof, from which the prom u lgators derive their authority. T o accept an order is to ascribe legitim acy to its term s. T o ascribe legitim acy to its term s is to oblige oneself to abide by them , or to put it a little differently, to agree to their application as a set o f stan dards again st which the accepto r’ s ow n actions are properly judged. Insofar as participation in a liturgical order is an acceptance o f that order, it legiti­ mizes that o r d e r .. . .

Th e M o ra lity Intrinsic to R itu al's Structure The perform ance o f a liturgy not only brings conventions into being but invests them with m orality. M oral dicta are not explicit in all liturgies, but m orality, like social con tract, is im plicit in ritual’s structure. We have, follow ing Searle, noted that o b li­ gation is entailed by the acceptance intrinsic to participation in ritual. Breach o f obligation , it could be argu ed, is one o f the few acts, if not, indeed, the only act that is alw ays and every­ w here held to be im m oral. H om icide, for instance, is not. There are con dition s, so com m on as to require no illustration, under which killing hum ans is laudable or even m an ­ datory. W hat is im m oral is, o f course, killing som eone w hom there is an obligation , at least tacit, not to kill. A sim ilar point can be m ade ab o u t m ost or possibly even all other specific acts generally taken to be im m oral. Breach o f ob ligation is o f a higher order o f generality than any such specific breach as m urder, rape or robbery and it m ay be suggested that it is breach o f ob ligation that tran sform s otherw ise m orally positive, neutral or empty acts into crim es such a s m urder or robbery. Breach of ob ligation m ay, then, be the fundam ental im m oral act, the element in the absence of which an act can n ot be construed to be im m oral, in the presence o f which it is ipso facto im m oral. The topic is a difficult one, and surely can n ot be settled here. I will only

EN A C TM EN TS OF M EA N IN G

em phasize that failure to abide by the term s o f an ob ligation is universally stigm atized as im m oral. T o the extent, then, that ob ligation is entailed by the acceptance intrinsic to the perform ance o f a liturgical order, ritual e stab ­ lishes m orality as it establish es convention. T he establishm ent o f a convention and the establishm ent o f its m orality are inextricable, if they are not, in fact, one and the sam e. We m ay refer again in this regard to the relation sh ip o f perform atives to the states o f affairs with which they are concerned. Austin initially tried to say that perform atives differ from statem ents in that perform atives are neither true nor false (1 9 7 0 : 2 3 3 ff.) w hereas statem en ts are either true or false. Later he found this view to be q uestion able because certain perform atives, notably verdictives, are su p p osed to stand in a relationsh ip to states o f affairs sim ilar to that o f true statem ents to states o f affairs. Later he found this not alw ays to be the case. Perform atives d o, how ever, differ from statem ents in a related w ay which he did not note but which does have to d o with truth, and with the foun dation s o f m orality. In discu ssin g their indexical nature, we have observed th at the relationship o f perform atives to the states o f affairs with which they are concerned is the inverse o f that o f statem ents o r descriptions. Statem ents report a u to n o ­ m ously existing states o f affairs. Perform ative acts realize states o f affairs. T he inverse nature o f these relationships has ob vious im plications for assessm en t. The ad eq uacy o f a descriptive statem ent is assessed by the degree to which it co n form s to the state o f affairs that it purports to describe. If it is in sufficient conform ity we say th at it is true, accurate or correct. If it is not we say that it is false, erron eous, inaccu­ rate or lying. The state o f affairs is the criterion by which the truth , accuracy or adequacy o f a statem ent is assessed. In the case o f p erform a­ tives there is an inversion. If, for instance, a m an is properly dubbed to knighthood and then proceeds to violate all o f the can on s o f chivalry, or if peace is declared in a properly conducted ritual but soon after one o f the parties to the declaration attack s the other, we d o not say that the dubbin g or rhe peace dec­ laration were faulty, but th at the subsequent states o f affairs are faulty. We judge the state

425

o f affairs by the degree to which it conforms to the stipulations o f the perform ative act. Liturgical orders provide criteria in term s o f which events - behavior and history - m ay be judged. As such, liturgical o rd ers are intrinsi­ cally correct or m oral. M orality is inherent in the structure o f liturgical perform ance prior to w hatever its can on s explicitly assert about m orality in general or w hatever in particular m ay be taken to be m oral. M orality derives ultim ately not from statem en ts ab o u t w hat m ay be right and w rong but from w hat liturgy establish es as right or w ron g. T o put it a little differently, to establish a convention indepen­ dent o f usage is to establish an “ o u gh t” against which the “ is” o f beh avior m ay be judged. The establishm ent o f m orality is clearest in the case o f sim ple factiveness an d the com m is­ sive im plications thereof. It is patently im m oral to act incom patibly with the term s o f a con­ ventional state o f affairs th at one has ritually participated in bringing into being. M y argu ­ m ent im plies, how ever, that m orality is also intrinsic to the m etafactiveness o f ritual, that is, to the establishm ent o f particular conven­ tions and conventional orders. O ne w ho vio­ lates not merely the term s o f a conventional state o f a ffairs, but o f the conventional order defining such states o f affa irs is not guilty o f a sim ple im m oral act, but o f ap o stasy . It is o f interest here that in Z o ro a stria n Persia and Vedic India states o f affairs th at departed from the proper liturgically establish ed order were designated by term s that also seem to have m eant “ lie,” druj in Persia an d anrta in India (D uchesne-Guillem in 1966: 2 6 ff., N . Brown 1972: 2 5 2 ff., O rlin 1976). W hat m ay be called “ Vedic lies” or “ Z o roastrian lies” (R ap p ap o rt 1 9 7 9 a), states o f affairs that their perpetrators are aw are d o not conform to prevailing liturgi­ cally established orders, are the inverse o f “ vulgar lies,” statem ents that their tran sm it­ ters believe m isrepresent the states o f affairs which they purport to report.

Ritual as th e Basic Social A ct T o sum m arize, the existence o f a conventional order is contingent upon its acceptan ce; in fact

426

R O Y A . R A PPA PO RT

a rule or understanding cannot be said to be a convention unless it is accepted. In ritual, how ever, acceptance and existence entail each other, for a liturgical order is perforce accepted in its realization , in, that is to say, the p erfor­ mance which gives it substance. Since o b lig a ­ tion is entailed by acceptance, and the breaking o f ob ligation is per se im m oral, the existence, acceptance, and m orality o f conventions are joined together indissolubly in rituals; they are, in fact, virtually one and the sam e. The sam e can n ot be said o f principles, rules, p ro ­ cedures o r understandings established by p roc­ lam ation , or legislation on the one hand o r by daily practice on the other. This is to say that there is a logically necessary relationship between the form which is ritual, the p erfor­ m ance o f m ore or less invariant sequences o f form al acts and utterances not encoded by the perform ers, and the m essages rituals contain concerning both w hat is perform ed and the

relationship o f the perform er to w hat he per­ form s. R itual is not merely another w ay to “ say th in gs” or “ d o things” that can be said or done as well or better in other w ays. The form which is ritual is surely w ithout com m unicational equivalents and thus, possibly, w ithout functional or m etafunctional equ iva­ lents. T h at ritual’s abilities are intrinsic to its form and in indissoluble association only with its form , goes far to account for its ubiquity. In attending to ritual’s form we m ust not lose sight o f the fundam ental nature o f w hat it is that ritual does as a logically necessary outcom e o f its form . In enunciating, accepting and m aking conventions m oral, ritual contains within itself not sim ply a sym bolic representa­ tion o f social con tract, but tacit social contract itself. As such, ritual, which also establishes, gu ard s, and bridges boun daries between public system s and private processes, is the basic social act.

NOTES 1

2

An act or utterance is said ro have illocu­ tionary force if it achieves its effect in its very utterance or enactm ent. An effect is intrinsic to an illocutionary act or p erfor­ m ance; the concept o f illocutionary force is a secular equivalent o f the doctrine o f ex opera operato , explicitly specifying the effi­ cacy o f the sacram ents o f R om an C ath o li­ cism , but im plicit in the thought and practice o f other religious traditions. In co n trast, an act or utterance is said to have perlocutionary force if it achieves its result through its effect upon receivers. Its effect is not intrinsic to the act or utterance itself, but is realized only if the act or utterance p ersuades, threatens, cajoles (or whatever) som e party into takin g action. (Austin 1 9 6 2 , esp. Lecture VIII). The identification o f tw'o classes o f p erfor­ m ative, factives and com m issives, is suffi­ cient for our p urposes, but it does not con form to A ustin’s taxon om y. H e identi­ fied five classes (1 9 6 2 : 150ff.). 1

Verdictives, “ typified by the giving o f a verd ict,” but including also less form al

2

3 4

5

judgem ents, e.g., assessin g, gradin g, ranking. E.xercitives, “ the exercising o f pow ers, rights, or influence,” e.g., appoin tin g, voting, ordering. Commissives , “ typified by prom ising; they com m it you to do som eth in g.” Behabitives , “ a very m iscellaneous grou p . . . |having] to do with attitudes and social behavior . . . apologizin g, co n gratu lating, com m ending, co n d ol­ ing, cursing and challenging” (em pha­ sis in original). Expositives , “ m ake plain how our utterances fit into an argum ent or co n ­ versation, how we are using w ords or are, in general, expository . . . I argue, I concede, I illustrate . . . I p o stu late .”

T hese categories are obviously not m utu­ ally exclusive. Austin also notes that the distinction between perform atives and statem ents, reports, and descriptions (the class o f expression s he calls “ co n statives” (1 9 6 2 : 3)) is not alw ays sharp and it is clear that illocutionary acts often possess perlocutionary force as well. Austin did not identify factives at all, but it is further

EN A C TM EN TS OF M EA N IN G

clear that they cross-cut at least tw o o f his categories, and that som e (e.g., “ we declare p eace” ) have com m issive entailm ents. All in all, it is better to regard A ustin’s categ o ­ ries, and the term “ factive” as well, a s des­ ign ation s o f illocutionary functions rather than o f distinct classes o f expressions.

REFERENCES A ustin, J . L, 1962. H ow to do Things with Words. O x ford : O x fo rd University Press. ------. 1970. “ Perform ative utterances,” in J . U rm son and G . W arnock (eds.), Philo­ sophical Papers o f J. L. Austin. 2nd ed. O x fo rd : O xford University Press. B ateson , G regory. 1951. “ C onventions o f com m unication: where validity depends upon belief,” in J. Ruesch and G. Bateson,

Communication: The Social M atrix o f Psy­ chiatry. N ew York: N orton . ------. 1 9 7 2 . Steps to an Ecology o f Mind. N ew Y ork: Ballantine. ------. 197 2 a. “ The science o f mind and o rd e r,” in B ateson 1972. ------. 1 9 72b. “ Form , subsistence, and differ­ ence,” in Bateson 1972. First published in G eneral Semantics Bulletin no. 37 (197 0 ). Bloch, M arc. 1961. Feudal Society. V ols. 1 &c 2. T ran s. L. M anvon. C h icago: University o f C h icag o Press. Bloch, M aurice. 1973. “ Sym bols, son g, dance and features o f articu lation ,” Fluropean Jo u rn al o f Sociology 15: 5 5 -8 1 . Brow n, N orm an W. 1972. “ Duty as tru th ,”

Proceedings o f the American Philosophical Society 116 (3): 2 5 2 -6 8 . C am pbell, Jo sep h . 1959. The M asks o f God. vol. 1, Primitive Mythology. N ew Y ork: The Viking Press. D o u g las, M ary.

197 3 . N atural Symbols: Explorations in Cosmology. H arm on ds-

w orth: Penguin. First published by Barrie and R ocklift, 1970. D uchesne-Guillem in, Jacq u e s. 1966. Symbols an d Values in Zoroastrianism . N ew Y ork: H arp er and Row . Erikson, Erik. 1966. “ O ntogeny o f ritualization in m an ,” in J. H uxley (convenor), A

427

Discussion o f the Ritualisation o f Behaviour in Animals and Man. Philosophical T ra n sac­ tions o f the R oyal Society o f London. Series B. Biological Sciences 251 (772). Finnegan, Ruth. 1969. “ H o w to do things with w ords: perform ative utterances am ong the Lim ba o f Sierra L eo n e,” Man 4: 5 3 7 -5 1 . Frake, C. 1964. “ A structural description o f Subanun ‘religious beh avior’, ” in W. H. G oodenou gh (ed.), Explorations in Cultural

Anthropology: Flssays in H onor o f George Peter Murdock. N ew Y ork: M cG raw -H ill. G eertz, C lifford. 1973. The Interpretation o f Cultures . N ew Y ork: Basic Books. G luckm an , M ax . 1954. Rituals o f Rebellion in Southeast Africa. The Frazer Lecture, 1952. M anchester: M anchester University Press. G ooden ou gh , Erwin R am sdell. 1990. Good-

enough on the Beginning o f Christianity. Ed. A. T . K raabel. A tlanta: Sch olar’s Press. Leach, Edm und R. 1954. Political Systems o f

Highland Burma. A Study o f Kachin Social Structure. Boston: Beacon Press. ------. 1972. “ The influence o f cultural context on non-verbal com m unication in M a n ,” in R. H inds (ed.), Non-verbal communication. C am bridge: C am bridge University Press. LiPum a, Edw ard. 1990. “ T he term s o f change: Linguistic m ediation and reaffiliation am ong the M arin g ,” Journ al o f the Polynesian Society 99: 9 3 -1 2 1 . M eggitt, M . J. 1965b. The Lineage System o f the M ae Enga o f New Guinea. Edinburgh: O liver and Boyd. M iller, Jam es. 1965. “ Living system s: Basic co n cep ts,” Behavioral Science 10: 1 9 3 -2 5 7 . N eedh am , Rodney. 1972. B elief Language , and Experience. O x ford : Basil Blackwell. O ’D oherty, F-. Fehean. 1973. Ritual as a second order language. Paper prepared for Burg-W artenstein C onference, no. 59: Ritual and Reconciliation. M . C . Bateson and M . M ead (conveners). N ew Y ork: W enner-Gren Foundation. O rlin, Louis L. 1976. “ A thens and Persia ca. 5 0 7 B .C .: a neglected perspective,” in L. O rlin (ed.), Michigan O riental Studies in H onor o f George G. Cameron Ann A rbor,

428

R O Y A . R A PPA PO RT

M I: D ept, o f N ear Eastern Studies, U niver­ sity o f M ichigan. R ap p a p o rt, R oy A. 1968. Pigs for the Ances­

Skoru psk i, Jo h n . 1976. Sym bol and Theory: A

tors: Ritual in the Ecology o f a New Guinea People. N ew H aven, C T : Y ale University

C am bridge University Press. Tillich, Paul. 1957. The Dynamics o f Faith. N ew Y ork: H arper and R ow . Turner, Victor. 1969. The Ritual Process. C h icago: Aldine. W allace, Anthony F. C. 1966. Religion: An Anthropological View. N ew Y ork: R an dom H ouse.

Press. ----- . 1979. Ecology , M eaning , an d Religion. Richm ond C A : N orth Atlantic Books. ------. 1979a. “ On cognized m od els,” in R ap p a p o rt 197 9 . ----- . 1979b. “ Sanctity and lies in e volu tion ,” in R ap p a p o rt 1979. Searle, J. 1969. Speech Acts. C am bridge: C am b rid ge University Press.

Philosophical Study o f Theories o f Religion in Social Anthropology. C am bridge:

Part IV Historical Dynamics: Power, Modernity, and Change

Capitalism, Colonialism, Christianity, and Conflict

In trod u ction From the 1960s anthropology became more politically self-conscious. In response to American engagement in the w ar in Vietnam and in observing the stark inade­ quacy o f modernization theory as propounded by economists and other social sci­ entists, anthropologists began to pay much greater attention to economic inequality and political oppression. There w as a historical turn as the field began not only to exam ine the historical roots o f particular structures o f inequality but to recognize the fundamental historicity o f all social formations, religion included. T hus began a lively discussion about the role o f religion in the waves o f colonial and capitalist expansion, the part played by missionaries, the transformations o f the religious systems o f colonized or rapidly changing societies, and the sometimes highly original religious responses they took. O f course, a s the essay by W olf (chapter 13) indicates, this interest had precedents. In particular one can point to discussions o f revitaliza­ tion and millenarian movements, o f which the finest contributions include Burridge (1 9 6 0 , 1969), Waliace (1972), and Worsley (1957). M uch o f this tradition owes a more or less direct debt to M a r x , but varies with respect to the role given to eco­ nomic and politica forces. M a r x ’s passages on com m odity fetishism (19 6 1 , 1977) have been particularly useful for m aking sense o f the culture o f capitalism.

34

New Heaven, New Earth Kenelm Burridge

Kenelm Burridge, professor em eritus at th e University of British Colum bia, was trained at Oxford and the Australian National University (PhD 1953) and sub­ sequently taug ht at the Universities of Baghdad, Oxford, and W estern A ustra­ lia, before settling at UBC in 1968. Bur­ ridge is the author of highly acclaimed w ork on M elanesian myth and religion, notably M am bu: A M elan esian M ille ­ nium (1960, reprinted 1995 and 2004) and Tangu Traditions (I969) and pio­ neered the study of indigenous forms of Christianity and of missionaries (1991). N e w Heaven , N e w Earth (1969), from w hich these passages are taken, w as a highly original look at w hat are som e­ tim es called m illenarian movements and w h a t in the Pacific w ere described under th e som ew hat condescending label of "cargo cults" (Worsley 1957, Lindstrom 1993). The book addresses the source and course of radical religious change. The argum ent has a surprisingly con­

tem porary ring. Burridge singles out redem ption - in his definition, the means by which religious practitioners return the debts incurred in realizing their potential (thereby combining a notion of salvation w ith ethical goals) as w ell as concerns w ith "the truth about pow er." M illenarian movements - new religions in the making - arise w hen assumptions about power are challenged, notably under colonial oppression, and w hen new avenues are needed to seek redem ption, that is, to lead a dignified, responsible life. Bur­ ridge thus links religious perturbations to moral disquiet and political and eco­ nomic processes. Here he was prescient in focussing on the significance of money in transform ing systems of value and thus the conditions for social prestige, self-respect, and moral integrity. In the book Burridge compares m ille­ narian movements from a w ide variety of sources; in addition to the Native

From Kenelm Burridge, New Heaven, New Earth: A Study of Millenarian Activities (Oxford: Blackwell, 1969), pp. 3-14, 41-46, 75-83. Abridged.

N EW H EA V EN , N EW EARTH

Am erican case described here, these include European, South Asian, M elane­ sian and Polynesian examples. If Burridge's thinking owes much to Chris­ tianity, it could be argued th a t early

T o dream a dream and m ake it com e true; to realize the shape o f w hat can be seen only in the m ind’s eye; to feel com pelled to bring ab o u t the seem ingly im possible - these are the prerogatives o f m an. Ja m e s N ay lo r w as plough ­ ing his fields when, in a blinding and tim eless m om ent, he knew why he had been born. Like Saint Paul, w ho never w avered in his adher­ ence to a truth revealed to him in a vision, Ja m e s N ay lo r, despite the cruelties o f p arlia­ m ent and a bigoted religious orth od oxy, rem ained stead fast in w hat his vision had revealed to him. Driven by her voices, a French peasant girl put new life into a dispirited army an d routed the alien invader. She w as burned at the stake as a heretic and w itch; she w as a lso canonized Saint Jo a n . Ann Lee claim ed to be the new C h rist; Jo se p h Smith had a vision o f heavenly bliss to be realized in an earthly com m unity life; and th o u san d s o f m iles and m om ents aw ay a M ao ri, a P apuan , an African, an Indian - each is im pelled to tell his good new s o f a new w ay o f life. W hether as fool, frau d , saint, respectable b o u rgeo is, farm er or tycoon , the pain o f the m illennium belongs only to m an. It is why he is m an, w hy, when the rime com es, he has to m ake a new man. Som e fifty years ago H ad d o n w ro te:1 “ An aw aken in g o f religious activity is a frequent characteristic o f period s o f social unrest. The w eakening or disruption o f the old social order m ay stim ulate new and often bizarre ideals, an d these m ay give rise to religious m ovem ents that strive to sanction social an d political aspiration s. Com m unities that feel them selves oppressed anticipate the em ergence o f a hero w ho will restore their p rosperity and prestige. And when the people are im bued with reli­ giou s fervour the expected hero will be regarded as a M essiah . Phenom ena o f this kind are well know n in history, and are not unknown at the present day am o n g peoples in all stages o f civilization .”

433

Christianity, arising in a colonized prov­ ince of the Roman em pire, itself offers an elegant exam ple of a m illenarian movement.

A forceful an d succinct enough statem ent. In the space o f a p aragraph H addon describes the kind o f activity we are setting ou t to exam ine. It is w orthw hile going over it to m ake sure we know w h at we m ean by som e o f the w ords H ad d o n uses.

R eligion an d Redem ptio n W hat we m ean by “ religious activity” is clearly o f key im portance. W riting at the tim e he did H ad d o n m ay have had in mind T y lo r’s m inimal definition o f religion: “ the belief in spiritual b ein gs” .2 O r, since it is m ore usual and puts m ore generally much the sam e point a s T ylor m ade m ore specifically, it m ay be that H ad d on th ought o f “ religious activity” as essentially defined by a belief in the supernatural. Hut neither o f these definitions is o f much so cio ­ logical value a s they stan d. For though we can observe rites and rituals and infer their sym ­ bolic references with som e accuracy, the problem o f belief begs the question , we cannot know w hat a spiritual being is w ithout further q u alification , and it is to o often tem pting to define a belief in the supernatural in term s that w ould scarcely apply to an ybody anyw here.3 We need a broader view , one th at subsum es the variety o f activities that m ay be religious. M editatin g on the infinite m ay be a religious activity, so m ay w riting a cheque, eating co rp se s, copu latin g, listening to a thum ping serm on on hell fire, exam in in g o n e’s co n ­ science, painting a picture, grow in g a beard, licking leprous sores, tying the body into knots, a dogged faith in hum an rationality - there is no hum an activity which can n ot assum e reli­ giou s significance. When it does so it has over­ riding im portance. It points to that which perm eates and inform s a w hole w ay o f life, an d, m ore crucially, it indicates sources or principles o f pow er which are regarded as

434

K EN ELM BURRID GE

particularly creative or destructive.4 Indeed, all religions are basically concerned with pow er. They are concerned with the discovery, identi­ fication, m oral relevance and orderin g o f different kinds o f pow er whether these m an i­ fest them selves as thunder, or lightning, a to ­ mic fission, untram m elled desire, arrogan ce, im pulse, ap p aritio n s, visions, or persuasive w ords. W ithin these term s a spiritual being, whether thought o f a s a deity or gh o st or hum an being or angel or goblin or fairy, becom es a nam ed and identified source or principle o f pow er with particular and often m easurable attributes and ranges o f p o w e r / And all that is m eant by a belief in the su p er­ natural is the belief that there do exist kinds o f pow er w hose m an ifestation s and effects are observable, but w hose natures are not yet fully com prehended. R eligions, let us say, are concerned with the system atic ordering o f different kinds o f pow er, particularly those seen a s significantly benefi­ cial or dan gerous. T his entails a specific fram ew ork o f rules. But because a religion is concerned with the truth o f things, and reaches out to discover and identify those so rts o f pow er w'hich, though sensed and affective, are currently not w holly com prehended, its rules about the use and control o f different kinds o f pow ers are groun ded in an interplay between experience, w orking assu m p tion s, an d those m ore rooted assu m p tion s we call faith. As experience w idens and deepens, som e o f the rules and assu m p tion s will be qualified, and others aban don ed altogether - a developm en­ tal process in which received truths or a ssu m p ­ tions give w ay to new truths, and in which the new truths becom e in their turn the received assum ption s o f future generations. These assum ption s are com m unity truths, truths which com m and a consensus. From them are derived the sets o f m oral im peratives, o b lig a ­ tions, and rules o f conduct to which men, because they live in com m unity, subject them ­ selves. Yet though m an governs his condition with explicit and articulate rules o f this kind, they are rarely interconsistent. C oncrete situ a­ tions often involve selecting one rule at the expense o f anoth er, and individuals, whether selfishly or otherw ise m otivated, accord d iffer­ ing priorities to som e ob ligation s at the expense

o f others. And here to o we touch on the essence o f religious activity. For, given a con text deter­ mined by current assu m ption s ab o u t pow er, the process whereby individuals attem pt to discharge their o b ligation s in relation to the m oral im peratives o f the com m unity is no less than a “ redem ptive p ro cess” . From the pen o f a social scientist, “ redem p­ tive p ro cess” seem s a cu riou s phrase. But it is useful. The hum an condition ap p ears as one o f general indebtedness: a feature which we acknow ledge in variation s o f the aphorism “ paying ou r debt to society” . For whether the capacities o f a hum an being are given him by Ciod and/or a particular com bination o f genes, his potential can only be realized after a long process o f feeding, nurturing, teaching and training by parents and others. Society, m ore­ over, prescribes the attitudes and activities by which its m em bers can pay back or redeem the debt incurred in being nurtured, m ade m orally aw'are, and enabled to exert and realize their potential. While these prescribed activities m ay be thought o f a s “ redem ptive m ed ia” , the m edia through which the debt is repaid or redeem ed, the process o f engaging in the activi­ ties - activities which are ordered in term s o f particular kinds o f ob ligation s - is, in our idiom , the redem ptive process, a process which leads on to redem ption itself. But this, the paym ent o f the debt in full, can only be real­ ized when a hum an being becom es in him self com pletely unobliged, w ithout any obligation w hatsoever - a free-m over in heaven, enjoying nirvana, or joined with the ancestors. For since existence in com m unity, a m oral order, neces­ sarily entails existence within a netw ork o f obligation s, redem ption itself can only be real­ ized at or after that app ropriate death which brings to an end an app rop riate m ode o f d is­ charging one’s obligation s. We m ay now m ove to w ards a w orking defi­ nition o f religion and religious activity. Let us say that they refer to

The redemptive process indicated by the activ­ ities, moral rules, and assumptions about power which, pertinent to the moral order and taken on faith, not only enable a people to perceive the truth o f things, but guarantee that they are indeed perceiving the truth o f things/

435

N EW H EA V EN , N EW EARTH

This definition has several advan tages. Instead o f tucking religion into an obscure and even alm ost irrelevant com partm en t o f social life, we give it the overriding im portance it actually seem s to have. For not only are reli­ gion s concerned with the truth ab o u t pow er, but the reverse a lso holds: a concern with the truth about pow er is a religious activity. O p er­ ationally, this concern is expressed in m ain­ taining or challenging the rules which govern the use and control o f pow er. And these rules assum e the form o f a set o f m oral discrim ina­ tions which, in constricting anim al m an, a lso provide him with opportun ities for realizing his m oral nature and potential. The definition im plies, and therefore leads us to expect despite the conservatism o f particular religious o rth o d oxies - that religious activities will change when the assu m p tion s ab o u t the nature o f pow er, and hence the rules which govern its use and control, can no longer guarantee the truth of things. The use o f this definition spares us unneces­ sary w rangling with distinctions between reli­ gion and m agic. We are not led into the im passe o f calling the religions o f other peoples bundles o f superstitions. We are insulated from the prejudgem ents contained in the dichotom ies rational/irrational and secular/religious. If a general rationality and order am o n g human beings and their affairs are not assum ed from the start, sociological an alysis m ust founder. The label “ irration al” rends to becom e a port­ m anteau o f ethnocentric prejudices, for faith is faith whether thought o f as religious or secular. M ore positively, with this definition we can identify activities and m ovem ents o f a generally m illenarian type where the w ords “ G o d ” , “ d eity” , or “ spiritual being” or their synonym s in other lan guages are not in evi­ dence. Because politics, to o , are concerned with pow er, it becom es clear and explicit that no religious m ovem ent lacks a political ideol­ ogy. And, accepting the political significance, we are forced to look at the w ays in which w ealth is distributed, and we m ust take account o f w hat pow ers are dependent on various kinds o f wealth. If it is not to becom e overly ethnocentric, anth ropology or co m p arativ e sociology m ust, initially, use broad concepts capable o f con­

taining the varied arrangem ents offered by different cultures. T he “ redem ptive p ro c ess” is just such a concept. N o t necessarily hedged with m ystery, it is som ething we can observe, ask ab o u t, talk about. The rules which govern the use o f pow er can be determ ined. Both emerge from the w ays in which individuals disch arge or evade their ob ligation s, w hat they do to counter or meet the consequences o f evasion , how rhey cope with a pledge redeem ed, w hat they say the consequences will or might be. We can , to o , identify prelim inary or tem ­ porary states o f red em p tion / Finally, we can accept provisionally the crude form ula, Salva­ tion = Redem ption = U nobligedness, or release from all obligation s. For, by exam in in g the kinds o f redem ption or releases from o b lig a­ tion that are offered by particular kinds o f m illenarian activity, we might be able to see m ore accurately w hat assu m ption s and rules are currently not revealing the truth o f things, w hat kinds o f redem ptive process w ould be m ore in tune with an actual or desired distri­ bution o f pow er.

N ew Ideals It will be clear from w hat has been said above that “ periods o f social unrest” and the “ w eak­ ening or disruption o f the old social o rd er” refer to situ ations where the relevant a ssu m p ­ tions ab o u t pow er are w eakening and no longer enable individuals to perceive the truth o f things. They cannot project a satisfactory redem ptive process. Hence the “ new and often bizarre id eals” . These m ay be seen a s attem pts to reform ulate assum p tion s a b o u t pow er so that they m ay account for the w idening experi­ ences o f everyday life and provide the basis for a new m ode o f redem ption. N ew ideals, new a ssu m p tion s certainly. But whether or not they are bizarre is entirely subjective. N o one, it m ay be assu m ed, does seriously w hat he him self thinks is bizarre. Ju st a s other kinds o f seem ing strange and esoteric activities in foreign cultures have yielded their m ystery to investigation, so again and again the a p p a r­ ently bizarre in m illenarian m ovem ents has been show n to be unexceptionable in the cir­ cum stan ces, given the prem ises. The hypothe­

436

K E N E LM BU RRID G E

sis that m illenary activities predicate a new culture or social order com in g into being which is w hat H ad d on im plies when he sp eak s o f these m ovem ents a s sanctioning “ social and political asp iratio n s” - is a fair one. Certainly it is m ore scientific to regard these activities as new -cultures-in-the-m aking, or as attem p ts to m ake a new kind o f society or m oral co m m u ­ nity, rather than as oddities, diseases in the body social, or troublesom e nuisances to effi­ cient adm in istration - though o f course they may be all these as well. Finally, o f course, a m illenarian m ovem ent is a new religion in the m aking. N ew a ssu m p ­ tions are being ordered into w hat m ay becom e a new orth o d oxy.

O p p ressio n H ad d on w rote in the heyday o f British Im pe­ rial an d colonial pow er. D espite the p u rp o se ­ ful w idening o f the problem in the last sentence - which is often left out when auth ors quote him - it is fair to say that m ost an th rop ologists have un derstood by “ o p p re sse d ” sim ply the effects o f the m achinery o f colonial exp an sion and governm ent. The vast bulk o f an th ro p o ­ logical evidence h as been draw n from the c o lo ­ nial situ ation: the effects o f the activities o f m ission aries, traders, settlers, com m ercial enterprises and adm inistrative bodies upon subject peoples; the w hole business o f a so p h is­ ticated, often greedy, and highly technological civilization im posin g its rules and conventions and experience upon those w ho, hitherto, had led a relatively sim ple life in sm all com m unities based upon hunting, fishing, agriculture and handicrafts. Such evidence m ay well give a distorted im pression, and it is useful to bear in mind that “ o p p re ssio n ” does not necessarily com e from ou tside a p articular cultural b o u n d ­ ary. A people or grou p m ay op p ress them ­ selves and m ay only afterw ards direct their fru stration s at outsiders. H ence perhaps H a d d o n ’s telling ph rase: “ C om m un ities that feel [my italics) them selves op p ressed . . Further, it should not be thought that the w ord “ o p p re sse d ” could refer to anything m ore than w hat we have identified already: traditional assu m p tion s w eakening, a m oral order d ecay ­

ing, a positive and active if not alw ays a p p a r­ ently sober will to participate in w ider or different categories o f understanding whether these are phrased in econom ic or political or m ore m ystical term s. “ Feeling them selves op p re sse d ” by current assu m ption s about pow er, participants in m illenarian activities set them selves the task o f reform ulating their assu m ption s so as to create, or accoun t for and explain , a new or ch an ging m aterial and m oral environm ent within which a m ore satisfactory form o f redem ption will be obtained.

H eroes The “ em ergence o f a h ero” touches the wellsprings o f an abu ndan t literature. We can afford to be brief. If he is to be accepted as such, a hero or prophet or m essiah m ust m ake his presence know n to the com m unity that is expecting him; and in order to be recognized and enabled to com m unicate with his fo l­ low ers he m ust con form in som e way to the popu lar im age o f a hero or prophet or m essiah. Since existing authorities are quick to scent a challenge, it is only prudent to gain the sup port o f respected leaders o f the com m unity w ithout app earin g a s a rival to them ; and this an em erging hero can only do if w hat he has to say expresses and articulates just those q u es­ tions which the com m unity feels disturbed and an xio u s abo ut. Further, he should provide clear and acceptable answ ers or solution s to the questions being asked. Is acceptance really dependent on H a d d o n ’s rather vague “ reli­ giou s fervour” ? Perhaps. But even if observers o f the scene w ere com petent to analyse effec­ tively the hysteria and sim ilar em otional d is­ turbances which accom pan y m any a m illenarian m ovem ent, within the term s o f our view of religion we can see that m any other m ore accessible and a s relevant features enter into the developing situation: differences in eco­ nom ic and political circum stances; variation s in capacities and opportun ities as between m il­ lenarian grou p and others; im passes o f co m ­ m unication and understanding. N on eth eless, providing that we understand by “ religious fervour” not sim ply an em otional exhibition, but a state-of-being having so cio ­

N EW H E A V E N , N EW EARTH

logical relevance, it is an im portan t phrase. N ew assu m p tion s which predicate the creation o f a new m an, a new culture, society or co n d i­ tion o f being are being w rought. And these relate directly to “ prosperity and prestige” though not necessarily to a restoration o f tra­ ditional kinds o f prosperity and prestige. T he new assu m p tion s tend to meet in the hero and his (divine) revelation, and alm o st alw ays refer to a “ prosperity and prestige” that are co n sis­ tent w ith, and even define, the new conditions o f being, the new m an. As we shall see, an ad eq u ate or m ore satisfacto ry w ay o f gaining prestige, o f defining the criteria by which the content o f m anhood is to be m easured, stan ds at the very heart o f a m illenarian or m essianic m ovem ent. And these criteria relate on the one hand to gaining or retaining self-respect, statu s, an d that integrity which is im plied in the ap proved retention o f a p articular statu s; and on the other hand to an acknow ledged process w hereby redem ption m ay be won. Lo gically, a m essianic m ovem ent is one that requires a m essiah. And so accu stom ed are we to a “ m ovem ent” having a leader or hero or p roph et that we tend to forget that som e m il­ lenarian activities m ay take place w ithout one. There m ust have been m any episodes or occu r­ rences in history, a s there certainly have been in O cean ia in recent tim es,8 w hich, lacking one w ho could be identified as a leader, still evoke new assu m p tion s and are otherw ise o f the sam e genre as those m ore coherently organized activities in which som e kind o f hero or proph et o r m essiah has been identified. In such cases, one m ay say, the new assum p tion s are im plicit rather than explicit. The m essiah or hero or prophet is but an em erging idea, unre­ alized a s yet in the flesh.

Prophets There is no need to say m uch ab o u t the w ord s “ m essiah ” , “ h ero” , “ p ro p h et” , “ ch iliastic” , “ m illen arian ” , “ m essianic” , “ cu lt” , “ m ove­ m en t” , “ activities” . T o think that each term refers to a distinct person or situation obscures rather than clarifies, closes rather than opens the sociological problem . O nly when the activ­ ities have been m ade to yield to a general co n ­

437

ceptual fram ew ork m ay we g o on to distinguish different types. T h u s although outside the Judaeo-C h ristian -Islam ic traditio ns there can in the strict sense be no proph ets, m essiahs or m essianic traditions or expectatio n s - though there m ay well be traditions o f a m essianic type, and m any w hom we could say were heroes - w e shall use the w ord prophet to refer to the leader, prim e m over, star or central personality in the kinds o f activities we are discussing. C h iliastic, m illenarian and m essi­ anic will be regarded as syn on ym ous term s; and whether rhe activities constitute a cult or a m ovem ent the reader m ay decide for him self.* Such usage does not prejudge any useful d is­ tinctions there might be between these term s, but it does avoid argum ent a s to whether a person w as really a hero or m essiah or false or true prophet before we are in a position to m ake the argum ent w orthw hile. Being ou r­ selves the children o f a long-established tradition o f m essianic expectatio ns there are a d van tages in using fam iliar term s, oppo rtu n i­ ties only for confusion in a d d in g to the pleth­ ora o f nom enclatures invented over recent years. O n the other hand, tem ptation s to seize on sim ilarities at the expense o f differences should be resisted. As we shall see, there are pertinent differences between one kind o f m illenarian activity and another. There are a lso different kinds o f prophet - a half-crazed w om an ob sessed with her visions, for exam ple, or a sickly lad w ho dream s a dream upon which oth ers act, but w ho him self is heard o f no m ore; a shrew d m an o f stature with organizing ability; a visionary or seer given to trances; or a saintly m an misled by his ow n piety. Further, at least in the prelim inary stages before a m ovem ent has cohered, there m ay be no proph et at all - just a series o f apparently im pulsive activities accom pan ied by inchoate ideas, feelings and em otions. O nly when the inchoate ideas have begun to cohere into new assu m p tion s m ay a proph et em erge to arti­ cu late them , show them fo rth , m ake them explicit. T h is is the revelation w hich, thought o f as divinely inspired, provides the basis for a new departure, a new m ode o f redem ption. Every m illenarian m ovem ent for which we have evidence com es to us as a story, as a

438

K E N E LM BURRIDGE

narration o f historical or quasi-historical events. O ften, to o , we find either a d hoc or system atic attem pts to relate the events o f the story to events and features o f social relation ­ ship which lie outside the confines o f the story itself. T his, in essence, is w hat is m eant by “ extractin g the sociology from the h istory” . The issue is how satisfactorily, com pletely and system atically it m ay best be done. W hat we have attem pted in developing each brief e x cu r­ sus above in term s o f the content o f oth ers is a m ovem ent to w ard s defining particular kinds o f social relations in term s o f the content o f other kinds o f social relations. We have rejected religion a s sim ply a belief in spiritual beings; oppression as necessarily connoting the c o lo ­ nial situation; the prophet as a p articu lar p er­ sonality type. Instead, we have tried to m ake som e interrelated statem ents about m illenarian activities which can be filled out with em pirical fieldw ork m aterial. T o sum m arize: (a)

(b)

(c)

(d)

We are discussin g assum ption s ab o u t pow er which, w hether or not investigator o r people concerned think o f them a s per­ taining to spiritual beings, predicate or entail a p articular redem ptive process. While the redem ptive process is disco ver­ able in disch arges and evasion s o f o b lig a ­ tion, redem ption itself refers to com plete release from obligation and is roughly equivalent to salvation. The redem ptive process, and so redem p­ tion, bears significantly on the politicoeconom ic process, particularly the prestige system . A prestige system is based upon p articular m easurem ents o f m anhood which relate to gaining or retaining self-respect and integrity, and which refer back to the politico-econom ic process, the redem ptive p rocess, and assum p tion s ab o u t pow er.

A ccepting these points as a baseline, m ille­ narian m ovem ents involve the ad o p tio n o f new assu m p tion s, a new redem ptive p rocess, a new politico-econom ic fram ew ork, a new m ode o f m easuring the m an, a new integrity, a new com m unity: in short, a new m an. A precondition o f this regeneration is a d issatis­ faction with the current system . But if we

describe this precondition as “ feeling o p p re sse d ” , the “ o p p re ssio n ” does not neces­ sarily derive from an external political control: it m ay be rooted in internal dissatisfaction s with present assu m p tion s, rules and m odes o f redem ption. Either w ay, we are interested in the patterns described when faith belies experi­ ence, when given assu m ption s ab o u t pow er no longer reveal the truth o f things, when the redem ptive process m ust change to accord with the new assum ption s born o f new experi­ ence. W hen, as m ay happen , the new a ssu m p ­ tions and rules are im plicit and unorganized, m illenarian activities m ay take place w ithout a prophet being identifiable. A prophet is he or she w ho organ izes the new assum ption s and articulates them ; w ho is listened to and found acceptable; w hose revelation is accorded authority for how ever brief a period. But a prophet cannot identify him self in term s o f the com m unity as it is: he identifies him self in an im age o f w hat m ight or should be.

M oney M illenarian activities, we have suggested, indi­ cate com peting sets o f assum ption s which, in turn, point to com petin g m oral system s or social orders. T o o m uch pow er on one side, how ever, m akes com petition im possible. While there m ay be attem pts at m illenarian activity, if the m ore pow erful g ro u p actually exerts its pow er the developm ent o f m illenarian thoughts into activities tends to be inhibited or snubbed out altogether. O n the other hand, when rhe com peting m oral system s have som e com m on ground in certain vital assu m ption s, then a m illenarian m ovem ent is likely. O r, to put it in another w ay, m illenarian activities are likely where, sharing or w anting to share a com m on set o f assum ption s which point to a particular redem ptive process, one o f the grou p s co n ­ cerned is debarred from access to the rew ards o f the assum ption s and cann ot earn the kind o f redem ption im plied. Fem ale prophets are usefully appreciated in this light. For besides being prophets in the ordinary sense they also participate in an infrastructure o f com petition and privilege, th at o f men versus w om en in a

N EW H EA V EN , N EW EARTH

w orld where men are privileged. Given this particular kind o f com petitive am bience, the relevant point o f con tract - through which we are enabled to recognize the existence o f co m ­ peting m oral system s - lies in ou r ability to distinguish different w ays o f m easuring the m an, different kinds o f redem ptive process. In this the presence o r absence o f m oney is vital. M oney goes alo n g with p articular m oral and social relations. M ak in g a radical distinction between social orders which use m oney as a basic m easure o f m an, an d those which d o not, is crucial. In the first case, a “ com plex econ o m y” , we have a highly differentiated division o f labour; m oney; free exchange o f g o o d s and services through m oney and m arkets; generalized full-time spe­ cialization ; conservation o f w ealth through the gen erations; and a basic m easure o f prestige and status by reference to m oney, though o f course a variety o f other qualities and cap aci­ ties also enter rhe situation. In the second case, a “ subsistence econ om y” , we have a sim ple division o f labour, a set o f relations in which the task s to be perform ed are basically deter­ mined by age and sex ; a prescribed and specific use o f treasure articles; specific and prescribed types o f exchan ge; no m oney; part-tim e spe­ cialization ; and a basic m easure o f prestige and statu s by reference to the subsistence activities which m ost or all undertake in com m on though again , a variety o f other capacities also enter into the situation. It is true that in som e subsistence com m unities there are one or tw o m agical, ritual, or even technological full-time specialists. Bur where these exist they are alm ost alw ays found to have a prestige o f their ow n; they are insulated from the requirem ents o f the m ain prestige system . Further, while there are num erous exam ples o f cash circulat­ ing in subsistence econom ies, whether or not this m oney has succeeded in becom ing a basic m easure o f prestige and statu s is precisely the issue. We are not co m p arin g stan d ard s o f living, for m any subsistence com m unities are richer and have higher stan d ard s o f living than particular parts o f a com plex econom y. We are distinguishing between tw o sets o f social rela­ tion s which seem , in principle, to be m utually exclusive. We are draw in g the divide between one kind o f m oral system and another.

439

In any com m unity the basic m easure o f pres­ tige and status m ust refer either to the activities which all or m ost undertake in com m on, or to those assets to which all or m ost in the co m ­ munity have a com m on access, or both. In addition to these basic m easures other more specific qualities and capacities are usually taken into account. In a subsistence com m u­ nity relative abilities in the m ain subsistence activities emerge as the b asic m easures. The possession , exchange and relative turnover o f treasure articles reflect or m easure industry, efficiency, cunning, shrew dness, foresight and a num ber o f other qualities; further prestige m ay be gained or lost by the dem onstration o f virtues and vices - such as virility, courage, w arrior capacities, envy, secretiveness - which m ay be, but are not necessarily, directly related to capacities in the subsistence activities. In the com plex econom y, on the other hand, where with a highly differentiated division o f labour there can be no one set o f activities engaged in com m on, the basic criterion for m easuring farm er, sm ith, carpenter, w heelright, teacher and others against each other m ust be that to which all have access, and by virtue o f which they are enabled to specialize: m oney. And though one or m ore o f a series o f highly dif­ ferentiated qualities and capacities are also usually taken into account, m oney rem ains the basic m easure. Succeeding to a large extent in m easuring locally preferred and selected m oral qualities, the treasure articles o f a subsistence econom y are, how ever, only w ealth in them selves so far a s, indicating particular qualities and cap aci­ ties, they can be used to gain credit. For in order to get m eaningful possession o f such an article a m an has to have been seen to do well, and in the prescribed w ay, the w ork that is required o f him. M oney m ay, but does not necessarily, m easure the m oral qualities. It keeps, m ay be banked for future use, is w ealth in a w ay that the treasure articles o f a su b sis­ tence econom y are not. For whether gotten inside or outside the law , by g o o d or evil acts, preferred or condem ned kinds o f procedures, m oney in a com plex econom y can com m and or buy lab ou r, things, hearts, souls and even credit. As concrete as any could w ish, suscep­ tible to touch and handling, pieces o f money

440

K E N E LM BU RRID G E

yet refer to an entirely ab stract system o f q u an ­ titative and factorial relations which endure and rem ain w hat they are through all so rts o f social vicissitudes. Ban ks, and the keeping qualities o f m oney, ow e their perm anence m ore to the m ath em atical system to which they refer than to their m aterial construction. W hereas the relative values o f treasure articles depend on intrinsic qualities as well as social and historical asso c iatio n s, unless an old shil­ ling is to becom e a treasure article it is w orth a s m any pennies a s a new one. M ore subtly, m oney is a factorial m easure o f m an which entails q uan tifyin g his different capacities. It sep arates and differentiates qualities and cap acities, and in d oin g so gives each a referent on a quantitative scale. Further, the handling o f m oney is an exercise in unitary and factorial relations. But q ualities can only be identified within a system o f binary op p o sites (good /b ad , stron g/w eak, skilled/unskilled, etc.); an d the factorial nature o f m oney introduces a hierar­ chy o f q uan titative values none o f which can be said to have an opposite. The tw o schem es, binary op p o sites on the one hand, and w hat we m ay call the one-and-the-m any on rhe other, are in principle m utually op p o sed . A co rresp on d in g antithesis is encountered where the w orth an d stature o f m an are m ea­ sured by m oney and clearly differentiated capacities on the one hand, and the quality o f his perform ance in an agglutinated “ package d e al” o f prescribed activities on the other. The problem is how to reconcile the antitheses. T hough in practice the m oneyed com m unities o f a com plex econom y also use q ualitative cri­ teria, and m ost subsistence com m unities use som e kind o f quantitative criteria, far from detracting from the point at issue the antithesis is em phasized. A cursory glance at the m ille­ narian activities which have taken place within the traditions o f the great m oneyed econom ies show that they turn significantly on m oney: on how access to m oney, and the uses to which m oney m ay be put, m ay be so prescribed and defined that this factorial and quantitative m easure can accurately assess selected m oral qualities. T h e p o o r m an w hose skills and honesty g o unrecognized is as wretched as the richer w hose acum en and thrift earn him nothing. O ften obscured in the colonial situ a­

tion by the m ore idealistic and hum ane rele­ vances o f different kinds o f paternalism , this issue o f quantifying and differentiating the qualities by reference to m oney is often ignored. M inted by the com plex econom y, by peoples w ho organize them selves into large, loose and highly differentiated entities, w ho engage in a m ultitude o f specialized task s and occupation s, w hose prestige system s, form s o f organ ization and division o f lab our depend upon m oney, money quite naturally ch ases the stuffs its m akers w ant, finds its w ay back into the pockets o f those w ho know how to use it and m ake it w ork to advan tage. Despite attem pts on the parts o f colonial adm in istration s to get m oney circulating am on gst an indigenous people, they have alw ays found difficulties in penetrating an indigenous prestige system based upon q u a li­ tative criteria, and then ad ap tin g or converting it to one based upon m oney. The indigenous qualitative criteria tend to persist. Prestige continues to depend upon com petence in the subsistence activities, on turnover o f exchanges, on the p assage and exchan ge o f treasure arti­ cles. The returned labourer with m oney in his pocket m ay enjoy a brief notoriety but, w ithout a continuing access to m oney m ust needs put his hand to cro o k , g o a d , m attock, axe or adze. Such m oney as he m ay have tends to be buried, h oarded, used in ad hoc paym ents or as trea­ sure articles. If m oney is to circulate significantly, if an indigenous com m unity is to have a satisfactory access to m oney, then the com m unity m ust so a d ap t and alter its prestige system that m oney becom es a basic m easure o f w orth. Unless and until m oney begins accurately to assess those qualities which a people or com m unity finds m eaningful, so long will they not be able to use m oney a s it w as m eant to be used, so long will they not have a satisfacto ry access. Belong­ ing to , and connoting, the com plex social order o f those w ho have minted it, use it, and bring it to foreign shores, m oney, particularly when o f the m ore valuable or pow erful curren­ cies, dem an ds acceptance o f the kind o f social ordering ad op ted by those w ho m ake it. N evertheless, corrupted by a covetou sn ess that is hum an, lured on by w hat seem s to be a finer or m ore satisfacto ry m eans o f realizing their

N EW H EA V EN , N EW EARTH

potential, yet reluctant to ab an d on w hat trad i­ tion has h allow ed, the dilem m a is how to replace a current prestige system by another; how to further differentiate, factorialize and quantify the qualities; how to redefine the qualities so thar they m ay be m ore satisfacto ­ rily quantified and differentiated. And both for those in a subsistence econom y w ho w ant to m ake m oney a relevant m easure o f m an, as well a s for like-m inded individuals within a co m p lex econom y w ho w ant to m ake money m ore accurately define the m oral qualities, a m illenarian m ovem ent is one w ay o f resolving the dilem m a. T o becom e it is first ne.cessary to belong; and belonging m ak es it possible to define just w ho o r w hat on e is. W hether for participants or bystan ders, m illenarian activities provide the opportun ity for becom ing som eone distinctive an d w orthw hile. A nd, because attitudes to m oney can so readily find ou t the m an, define w hat is m eant by integrity, millenial a sp ira ­ tions often find their focus in money, l -. -l

A sp ects o f In teg rity All cultures and com m unities have particular criteria whereby m em bers m ay m easure rela­ tive prestige am on g them selves. As we have seen very briefly, am on g Polynesians a variety o f criteria meet in rhe m eaning o f mana, com m an d over the lab ou r and allegiance of others, an d com petence in w ar. O ther kinds o f ability certainly contribute to the accu m u la­ tion o f m an a : know ledge o f protocol and m yths, sexual com petence, physical strength, handsom e features, w ell-m oulded m uscular prop o rtio n s, orato ry, and skills in dancing, sp orts, sailin g, fishing, hunting, agricultural activities and the m anufacture o f artifacts, to nam e but a few. But all these qualities tend to be subsum ed in the required qualities o f the political leader and w arrior. And the basis o f both these com petences w as control over land an d its resources. A landless fellow , a man w ithout any control over its resources, w as necessarily a man w ithout kin, or one whom oth ers w ould not recognize a s kin: scarcely a m an at all. When Te Ua claim ed to hold the

441

prom ise o f a “ redem ption for his people, who had becom e forgetful, d esolate and in d o u b t” his m eaning in an objective sense m ust be related to the land question and to w arfare. As accessible and useful lands began to shrink, and a s traditional w arfare began to becom e difficult or im possible w ithin the con text o f a Pax Britannica , so the criteria o f relative pres­ tige becam e less and less clearly defined. N o t only that. For a s rhe criteria o f prestige becom e loose and ill-defined, so d o the criteria o f integrity becom e doubtful. The kinds o f pow ers attached to a particular office or status becom e uncertain. There is no con sen su s as to the q ualities required o f those in position s o f authority and responsibility. Further, trad i­ tional criteria o f integrity tend to seem unim ­ portan t when com pared with the apparently superior cap acities o f other peoples - in this case the European settlers. W hether we choose to regard rhe Pai-marire m ovem ent as one that attem pted to capture a new integrity; or a s one th at attem pted to recapture a traditional integ­ rity; o r as one that attem pted to define an integrity com pounded o f both traditional and C h ristian or European features; the question o f integrity em erges as param ou n t. T h is itself is bound up with the redem ptive p rocess, the program m e o f particular kinds o f w ays in which individuals are required to meet their obligation s. In a m illenarian context both redem ption and integrity are part and parcel o f an overall developm ental process - a m ovem ent from the m ore sim ple to the m ore co m p lex , an increasing differentiation o f the pow ers, statu ses, q u ali­ ties, jo b s and techniques required to sustain com m unity life. But we still have to show how varying aspects o f integrity and redem ption are related to different kinds o f developm ental process. Som e exam ples from N orth A m eri­ ca . . . will dem onstrate my m eaning.

Th e Plains In dian s For som eth ing over tw o hundred years until their final defeat by the white m an , the cultures o f the Plains Indians o f N orth Am erica had been based on p ossession o f the horse, fight­ ing, an d hunting wild bu ffalo. Before the

442

K EN ELM BU RRID G E

Spanish brough t their horses to N orth A m erica, the indigenous peoples o f the region had lived m ainly on the edges o f the plains, and co m ­ prised sm all hunting ban ds w ho follow ed the m igrating herds o f bu ffalo in the season on foot. They also practised som e horticulture. But when they acquired horses they were able to take to the p lains m ore or less perm anently, relying for subsistence alm ost wholly on the buffalo. T he popu lation increased; the size o f the effective political grou p increased. From being a m otley collection o f sm all hunting bands w ho could only have put up a token resistance to the white pioneers, they becam e the w ell-organized and form idable an tago n ists o f an in dustrialized nation. They had neither artillery nor the m eans to m anufacture guns and bullets. But they acquired gun s and am m u­ nition an d fought for their plains and their buffalo until only a few rem nants rem ained. Lookin g back, we can see that their fight again st white men, w ho w anted the plains for farm land and range cattle, w as doom ed to failure. Yet they had fought according to the dem ands o f their culture. And those w ho died bravely m ay be said to have earned a trad i­ tional redem ption: they asked only that they should show them selves true to the p rescrip­ tions o f their social order. But w hat o f those who survived? O f the m any changes that possession o f the horse initiated, the m ost crucial related to integrity and relative prestige. H orses becam e the new w ealth. G ain in g access to horses by taking them from enem ies cither by stealth or in open w ar becam e an im portant criterion o f prestige. G etting horses by breeding them w as tan tam ou n t to forgery. W arrior values becam e param ou n t. The dash in g fighter w ho could get horses, count co u p s, gallo p in am on gst a herd o f bew ildered b u ffalo, and face extrem es o f physical pain in stoic silence, w as well on the w ay to a p osition o f high success. Add to these abilities qualities o f w isdom , m agnanim ity, cunning, know ledge o f the w orld and his people, and a certain ruthlessness, and the picture o f the Plains Indian chief is com plete in its outlines. At the other extrem e, a m ale w ho could not show som e com petence in these m atters w as regarded as hardly a m an at all. H is alternatives were either to becom e a

sh am an, when again he w ould have to show him self able to face w ithout flinching extrem es o f physical pain an d deprivation; or becom e like a w om an - dress in w om an ’s clothes, d o w om en’s w ork, becom e a disposable part o f a real m an’s household. H ighly specialized, the prestige system w as based on an intim ate rela­ tionship between w ar, bu ffalo and horses. D eath in w ar, fighting and dying bravely w ithout ask in g for, or giving, quarter w as thought a highly satisfactory redem ption. T o die peacefully having survived innum erable co m b ats with a fully feathered w ar-bonnet and co u p s p ast counting; to have hunted success­ fully, fed dependan ts, and disposed o f plenty o f horses; to have fingers w hose joints had been severed in attem pts to gain visions o f, and rapport with, a protector spirit; to bare a to rso scarred with the h ooks and skew ers o f volu n ­ tary torture - all these were tokens o f a perhaps higher redem ption. W hat then w as left for those rem nants w ho lived through to the end o f the Indian w ars in reservations w ithout horses, w ithout bu ffalo, w ithout w ar, dependent on the goodw ill o f an official agent, a trader with his store, and perhaps a m ission ary? Even as we look back with the ad van tage o f hindsight it is not easy to suggest a realistic com prom ise for this brave but defeated people. T o them defeat w as a disgrace. But they had to live on in defeat and they were denied the opportun ity o f living and dying in the only fit w ays they knew. In their ow n eyes rather less than men, despite a ssu r­ ances to the contrary their conquerors treated them with undisguised contum ely. H ow and in w hat w ay could they overcom e the irony o f a glo riou s efflorescence based on the horse, a European im port, and their ultim ate defeat by Eu ropean s? H ow and in w hat w ay could they find a new integrity, a new redem ption? In a sense there w as no solution. Plains Indian culture w as dead, could not develop into anything else. If in later years descendants o f these folk could point to the fact that all the w orld plays at C o w b o ys and Indians, this sim ply reiterates the problem for the im m edi­ ate survivors. Bereft o f a traditional integrity, the scattered rem nants o f a once proud people could d o little else but look back on their glories an d, with the aid o f som e new sym bols,

N EW H EA V EN , N EW EARTH

screw hope and w ishful thinking to the point where bu ffalo and horses m ight seem to be on the point o f returning. For the S iou x, who were to suffer so dreadfully at the battle o f W ounded Knee, one version o f the m any rev­ elations contributed by num erous prophets to that com plex o f activities generally subsum ed under the nam e “ G h ost D ance R eligion” 10 posited a general catastrop h e as heralding the m illennium . After a series o f earthquakes, landslides, storm s, w hirlw inds and floods in which all the w hites w ould be destroyed, the Indians w ould survive to see boundless prairies covered with wild g rass and filled with great herds o f bu ffalo and other gam e. O ther proph ets held that after the great catastroph e all race distinctions were to be obliterated, and that in the millennium that follow ed both w hites and Indians w ould live am icably together. “ Y ou m ust not fight. D o not harm anyone. D o right alw a y s,” 11 said the prophet W ovoka. A longer text from W ovoka’ s reaching reads as follow s: I found my children were bad, so I went back to heaven and left them. I told them that in so many hundred years I would come back to see my children. At the end o f this time I was sent back to try to teach them. My father told me the earth w as getting old and worn out and the people getting bad, and that I w as to

renew everything as it used to he and make it better. He also told us that all our dead were to be resurrected; that they were all to come back to earth, and that, as the earth was too small for them and us, he would do away with heaven and make the earth itself large enough to contain us all; that we must tell all the people we met about these things. He spoke to us about fighting, and said that was bad and we must keep from it; that the earth was to be all good hereafter, and we must all be friends with one another. He said that in the fall of the year the youth of all good people would be renewed, so that nobody would be more than forty years old, and that if they behaved themselves well after this the youth o f everyone would be renewed in the spring. He said if we were all good he would send people am ong us who could heal all our

443

wounds and sickness by mere touch and that we would live forever. He told us not to quarrel or fight or strike each other, or shoot one another; that the whites and Indians were to be all one people. Fie said if any man dis­ obeyed what he ordered his tribe would be wiped from the face o f the earth ; that we must believe everything he said, and we must not doubt him or say he lied; that if we did, he would know it; that he would know our

thoughts and actions in no matter what part o f the world we might be.n H ere plainly is the attem pt to excise the purely existential and apparently purposeless nature o f the present by searching into the past in order to posit a viable future. The G h ost D ance R eligion spread from com m unity to com m unity in the plains area in much the sam e way as carg o cult activities have in m ore recent years spread from village to village in N ew G uinea. And in spite o f M o o n ey ’s insistence on w hat w as the ‘true’ or ‘real’ doctrine o f the G h ost D ance Religion, it is prob ab ly w iser to accept that here (as in N ew G uinea and elsewhere) there were differ­ ent versions; differently projected assum ption s and a sp iratio n s turning on the central theme o f a g lo rio u s p ast and a m iserably uncertain future. On the other hand the rites o f the G h o st D ance itself, which included singing, chanting, falling into trance, stripping naked, and the w earing o f special shirts which w ould render the w earers invulnerable to the bullets o f the white m an, seem to have varied only slightly. A lso, the overt p u rpo ses o f the rites seem to have been much the sam e: attaining union with the glorious dead, enjoying now in trance w hat in the past had been fully realized. M oreover, if we look closely at the G host Dance we can discern certain them es - such as trance, self-torture, the excision o f discordant ideas, the renewal and re-em phasis o f co m ­ m unity values, the reconciliation o f disputants - which it had in com m on with the great Sun D ance: that vital political occasion when nom adic ban ds o f Indians gathered together to dance, feast, resolve their disputes, com pete in gam es and ritual exercises o f self-torture, and, despite or because o f their sectional exclusivism s, renew their com m on allegiance to an overall political o rd e r.11 T hus the G h o st Dance

444

K EN ELM B U RRID G E

w as heir to the Sun D ance; the traditional gave authenticity to the new. Here is Short Bull, with a m essage to the Sioux shortly before the battle o f W ounded Knee: My friends and relations: I will soon start this thing [the millennium] in running order. I have told you that this would come to pass in two seasons, but since the whites are interfer­ ing so much, I will advance the time from what my father above told me to do, so the time will be shorter. Therefore, you must not be afraid o f anything. Some o f my relations have no cars, so I will have them blown away. N ow there will be a tree sprout up, and there all the members o f our religion and the tribe must gather together. That will be the place where we will see our dead relations. But before this time the earth will shiver very hard. Whenever this thing occurs, I will start the wind to blow. We are the ones who will then see our fathers, mothers, and everybody. We, the tribe o f Indians, are the ones who are living a sacred life. G od, our father himself, has told and commanded and shown me to do these things. My father has shown me these things, therefore we must continue this dance [The Ghost Dance]. If the soldiers surround you four deep, three of you, on whom I have put holy shirts [supposedly bullet-proof], will sing a song, which I have taught you, around them, when some o f them will drop dead. Then the rest will start to run, but their horses w ill sink into the earth. The riders will jump from their horses, but they will sink into the earth also. Then you can do as you desire with them. N ow , you must know this, that all the soldiers and that race [the whites] will be dead. There will be only five thousand o f them living on the earth. . . . N ow , we must gather at Pass Creek where the tree is sprouting. There we will go among our dead relations. You must not take any earthly things with you. Then the men must

take off all their clothing and the women must do the same. N o one shall be ashamed of exposing their persons. My father above has told us to do this, and we must do as he says. You must not be afraid of anything. The guns arc the only things we are afraid of, but they

belong to our father in heaven. He will see that they do no harm. Whatever white men may tell you, do not listen to them, my rela­ tions. That is all. 1 will now raise my hand up to my father and close what he has said to you through m e.14 Is there not a note o f desperation in that con cludin g “ w hatever white men m ay tell you, d o not listen to them , my relation s” ? Here were people the reverse o f “ forgetfu l” but “ d esolate and in d o u b t” indeed - yet takin g com fort in w hat a prophet had to say, and w ho to ok action because o f w hat a prophet had to say. Perhaps one m ay be allow ed to see the battle o f W ounded Knee not a s the horror that in fact it w as, but rather as a final attem pt to gain a traditional kind o f redem ption. For the battle w as nor overtly intended by either side. Startin g a s an inaugural for peace, during which the Sioux were to hand over their arm s, an excited young Indian, losing control and com posu re in the tension accom pan ying the surrender, drew a rifle from under his blanket and fired at the white soldiers w ho were present in large n u m bers.M At this the whites to o lost con trol, and three hundred and seventy Indian m en, w om en and children were killed in the m assacre that follow ed. D ead men can be replaced. But here it w as traditional Plains Indian culture itself which had been destroyed. The survivors had to find quite new principles and assum ption s on which to base com m unity life. M ere individuals with com m on traditio n s, they had to reintegrate them selves a s com m unities based on quite dif­ ferent assu m p tion s from those they had know n in the past. But, unlike the m ore usual desert island “ startin g a g a in ” , there w as little m ate­ rial on which they could exercise their im agi­ nations an d traditional technical skills. First they had to becom e an organized com m unity. A prophet could say, m ore positively, “ D o right a lw a y s” . But w hat kinds o f activities could be thought o f a s “ righ t” now or in the future? T raditio n ally, the “ righ t” thing to do alw ays w as to stan d on e’s grou n d, defend one’s h onour and integrity, and fight. Yet they were being told not to d o precisely this. W hat w as a “ righ t” thing to d o - w hat the w hites told them to d o ? Should they w ait patiently in

N EW H E A V E N , N EW EARTH

lodge or ripi doin g no harm to anyone until a nervous Indian agent, afraid that his com m u­ nity m ight starve to death , handed out supplies from the trade store? Certainly they could plant a little to bacco. But few o f these survi­ vors knew how to grow corn or vegetables, or how to rear chickens an d pigs so that there w ould be a “ righ t” w ay o f doin g it which, through the productive and distributive process, w ould d em on strate ability, prestige and integrity. W age lab ou r in the tow ns? T h at m eant being controlled and treated as a “ n atu­ rally ” inferior sort o f person. W ith a future th at could only be phrased in vague and general ab straction s, it is hardly m atter for surprise th at such ab straction s as m ight be realized, were in fact realized in term s o f traditional activities. T h ose injunctions that could not be realized in term s o f p ast activities could hardly be realized at all. As it w as, those traditional activities which could be taken into the future - trance states, self-torture, the seeking o f a gu ard ian spirit by m eans o f techniques calcu­ lated to induce a release from ordinary w orldly an d social cares, com m unication with the an cesto rs and so continuity with the cultural heritage - were contained within the rituals of

rhe G h o st D ance itself. As in so m any other instances over the w orld, the activity o f the com m unal dance - in which particip an ts might speak as with tongues, have visions, escape from the prisons o f the body and form al m odes o f know ledge, gain inspiration - seem s to have been the seed o f such organ izatio n al form s as were to develop in the years that follow ed. Faced with organizing them selves from vir­ tually nothing into quite new kinds o f co m ­ m unities, the Plains Indians could only realize new w ays o f behaving by using traditional sources o f authority. T h at these traditional sources were sought through the com m unal dance and its accom pan yin g trance-like states, takes the Plains Indians ou t o f their purely cultural co n text and places them in a generally hum an settin g.1'' For not only does the co m ­ m unal trance-dance app ear a s a generally hum an device for seeking inspiration and fresh initiatives, it is in itself the com m unal expres­ sion and an alogue o f just th at experience w hereby a proph et, a lone individual, com es to be inspired with the revelation that others afterw ard s discover they w ould like to m ake their ow n.

NOTES 1 Chinnery and H ad d on , p. 4 5 5 . 2 T ylor, p. 4 2 4 . 3 If a “ spiritual bein g” is expressly not hum an, it still m ust involve question s as to m ass, visibility, an d attributes o f bilocation. N ot to “ believe in” phenom ­ ena such as trances, stig m ata, possession , levitation, w alking on hot co als w ithout being burned, or skew ering the cheeks w ithout leaving a w ound - which are all above or beyond the n atural, not found in nature - is surely equivalent to being a “ flat-earther” . 4 When we say o f a m an that “ art is his religion ” we m ean that he gives overrid­ ing im portance to art, that he is p articu­ larly concerned with nurturing and developing his sources o f inspiration, and gu ard in g him self again st those influences

445

5

6 7

which m ight endanger his inspiration, betray his integrity as an artist, or nullify his ideas on w hat Art should be or do. T he sam e applies to those o f w hom we m ight say “ science is his religion ” , or “ socialism is his religion” , or “ an th rop ol­ ogy is his religion ” . T h is holds even though it m ight be said th at a “ spiritual bein g” is but rhe ratio ­ nalized projection o f som e internal im pulse. C f. Lienhardt, pp. 3 2 7 - 9 ; Yinger, pp. 9, 7 1 - 2 ; V ernon, pp. 4 6 - 5 7 ; H orton. See, for exam ple, Burridge (1 9 6 5 ), pp. 2 2 6 - 9 , where a N ew G uinea people, after a scries o f reciprocal exchanges or d is­ ch arges o f ob ligation , attain to a state know n as mngwotngwotiki , a w ord which connotes a particular field o f relations in which the individuals concerned are tem ­ porarily unobliged to each other.

446 8 9

10 11 12 13 14 15 16

K EN ELM BU RRID G E

See Burridge ( I 9 6 0 ), pp. 1-4. While “ activity” is a general descriptive term , one m ay suggest that a cult, co n ­ cerned with a p articular source o f pow er, is already organ ized, firm and static, w hereas a m ovem ent seeks to im pose its ideas and present organ ization upon a w ider field. M ooney. M oon ey, p. 19. M oon ey, pp. 2 6 - 7 (my italics). See H oehel (1 9 4 1 ); (1 9 6 0 ), pp. 1 1 -1 6 ; and Ew ers, pp. 1 7 4 -8 4 , 2 9 8 ff. M oon ev, p. 31 (mv italics. C f. M am b u , p. 6 5 ) .' M oon ey, p. 118. Cf. Burridge (1 9 6 0 ), pp. 1-14.

REFERENCES CITED Burridge, Kenelm (K .O .L .). M ambu: A M ela­ nesian Millennium. Lon don , 1960. ----- . T an g u , N orthern M ad an g D istrict, in G ods, G hosts and Men in Melanesia (eds. P. Law rence and M . J. M eggirt), pp. 2 2 4 -4 9 . O x fo rd University Press, M elbourne, 1965. Chinnery, E. W. P. and H ad d on , A. C . Five N ew R eligious C ults in British N ew G uin ea, The Hibbert Jou rn al , vol. X V , no. 3, 1917: 4 4 8 -6 3 . F!kka, Philip. The Tana Bhagats: A Study in Social Change (O xford ). U npublished D. Phil, thesis, 1966.

Ew ers, Jo h n C. The Blackfeet. University o f O klah om a Press, 1961. Fuchs, Stephen. Rebellious Prophets. Bom bay, 1965. H oebel, E. A. The C om anche Sun D ance and M essianic O utbreak o f 1873, American A nthropologist , vol. XLII1, 1941: 3 0 1 -3 . ------. The Cheyennes. N ew Y ork, 1960. H orton , R obin. A Definition o f Religion, and its U ses, Jo u rn al o f the Royal Anthro­ pological Institute , vol. 90, Pt 2: 2 0 1 26. Lienhardt, R. G. R eligion, in M an , Culture and Society (ed. H . L. Shapiro). N ew Y ork, 1956, pp. 3 1 0 -2 9 . M ooney, Jam es. The G host Dance Religion and the Sioux Outbreak o f 1890 (ed. Anthony F. W allace). University o f C h icago Press, Lon don , 1965 (1896). R oy, S. C . A N ew R eligious M ovem ent A m ong the O rao n s, Man in India , vol. 1, no. 4 , 1921: 2 6 7 -3 2 4 . ------. O raoti Religion and Customs (pp. 3 1 2 4 1 0 ). R anchi and C alcu tta, 1928. Scholem , G ersholm G. On the K abbalah and its Symbolism , L on don , 1965. Serrin, O . The B h agat M ovem ent in C h ota N ag p u r, Indian Academy. Kurseny, 1917. T ylor, Sir F,dward Burnett. Primitive Culture (2 vols.). Lon don , 1891. V ernon, Glenn M . Sociology o f Religion. M cG raw Hill, N ew Y ork, 1962. Yinger, M ilton. Religion , Society and the Indi­ vidual. N ew Y ork, 1957.

35

The Genesis of Capitalism amongst a South American Peasantry: Devil's Labor and the Baptism of Money Michael Taussig

An Australian, trained first in medicine and then at the London School of Eco­ nomics, Michael Taussig is currently pro­ fessor at Colum bia University, New York. He engages in a w onderful dialogue with Marx, W eber, and even Aristotle in this piece, w hich shows how certain "m agical" practices found among plan­ tation workers in Colom bia em erge not only in response to capitalist exploita­ tion but serve as an enlightened moral critique of its consequences. This might be seen as a late phase o f the rationality debates, in w hich the tables are turned such that the rationality of the West comes under question from its victims and from the postcolonial w orld. Taussig argues that th e plantation workers have an understanding of comm odity fetish­ ism akin to Marx. Commodity fetishism is a kind of inversion or transform ation of totem ism in which attention is placed on human products rather than natural

species, but in which they are under­ stood to have life, vitality, reproductive potential, and possibly even intentional­ l y , w hile disguising the human labor entailed in their production. Taussig's argum ent has stimulated much discussion; lively criticism as well as com parative m aterial on the im agina­ tion of money can be found in Parry and Bloch (1989); cf. Burridge (chapter 34), De Boeck (1999). The local critique of forms of accumulation has form ed the basis for many anthropological accounts, often w ith respect to w itchcraft (Fisiy and Geschiere 1991, Solway 1998, Weiss 1998). This essay was one of th e first in a rem arkable series of texts by Taussig th at explore the creative responses to the horrendous impact of Europeans on Latin Am erica and th at serve, as in this essay, also as mirrors for m odernity and m editations on m ateriality. The argu-

From Michael Taussig, uThe Genesis of Capitalism amongst a South American Peasantry: Devil’s Labor and the Baptism of Money,” Comparative Studies in Society and History 19(2) (1977): 130-55. Abridged.

448

M IC H A EL TAUSSIG

ment is expanded in his 1980 book, w hich adds an analysis of the practices of Bolivian tin miners w ell described by Nash (1979). Taussig then delved into surrealism and Frankfurt School theory to produce rich and stylistically experi­ mental books on shamanism and vio­ lence (1987), colonial mimesis (1993; cf. Kram er 1993, Stoller 1995), and the

W hat does w age lab or and capital m ean to a peasantry that is subjected to rapid rural p ro­ letarianization and w hat is the basis o f that m eaning? I wish to d iscu ss an aspect o f this question in the light o f certain ideological reactions m anifested by a South Am erican low land p eas­ antry as exp an d in g su gar plan tation s a b so rb their lands and p easan ts are converted into landless w age laborers. In the southern extrem ­ ities o f the C au ca V alley, C o lo m b ia, it is co m ­ m only thought th at m ale plantation w orkers can increase their ou tp u t, and hence their w age, through entering into a secret contract with the devil. H ow ever, the local p e asan ts, no m atter how needy they m ay be, never m ake such a con tract when w orking their ow n plots or those o f their p easan t neighbors for w ages. It is also thought that by illicitly baptizing m oney instead o f a child in the C atholic church, that m oney can becom e interest bearing cap ital, while the child will be deprived o f its rightful chance o f entering heaven. A nalysis o f these beliefs in their social and historical co n text indicates that the low er classes’ im plicit understanding o f the new m ode o f production is inherently critical and an tagon istic and that the axiom atic b asis o f this an tagon ism rests on their con scious o p p o ­ sition o f “ use v alu e s” to “ exchange v alu e s” the o p p o sitio n o f the satisfaction o f natural w an ts, on the one side, to the lim itless search for profits and cap ital accum ulation on the other. In m aking this distinction these people share a close affinity with the econom ic th eo­ rizing o f the Schoolm en o f the E uropean M iddle A ges, and with the econom ic p h ilo so ­ phy o f A ristotle, w hose insights on this m atter

state (1997). A powerful auto-critique of the functionalist inclinations of the orig­ inal article - and th at should be read in conjunction w ith it - is an essay that emphasizes the seductive qualities of transgression, danger, and sheer excess (1995, cf. 1999). Taussig (2006) is a recent collection of his essays.

were often quoted favo rab ly by Karl M arx him self, since the op p o sitio n o f use value to exchange value w as basic to his entire system o f analysis. Further elucidation o f this distinction leads to a discussion o f the m etaphysical and m oral bases underlying the low er classes’ un derstand­ ing o f cap italist relations o f production and exchange, in which their folk m ysticism is con ­ trasted with that form o f cap italist m ystifica­ tion to which M arx gave the nam e o f “ com m odity fetish ism .” E valuation o f this co n trast is enhanced by an an alysis o f the m ode o f reasoning utilized by the sup porters o f the use value econom y. T h is reason in g ap p ears to derive from a con cept o f the uni­ verse as an interrelated organ ism which is understood through the co n sciou s application o f anim istic an alogies, rather than by m eans o f the atom istic causal paradigm which has gained ascendan cy in the social sciences since the rise o f N ew ton ian m echanics and the birth o f the industrial revolution in rhe W est.

A ttitu d e s to w a rd s W a g e Labor A theme o f co n stan t interest to historians o f the industrial revolution in F'urope, as well as to sociolo gists concerned with socio-econom ic developm ent in the T hird W orld, is the atti­ tude o f w orkers w ho are new to the m odern w age labor situ ations. T h e first reaction of peasants and artisan s w ho becom e w age w orkers in m odern business enterprises is fre­ quently if not universally one o f indifference to w age incentives. R egarded as a failure to

TH E G EN ESIS O F C A P ITA LISM A M O N G S T A SO U TH A M ER IC A N P EA SA N TR Y

m axim ize m arker opportun ities, a lack o f labor com m itm ent, or a peculiar irrationality about com plyin g with the p ostu lates em bodied in homo oeconom icus , this respon se has persis­ tently frustrated cap italist entrepreneurs the w orld over. M a x W eber referred to this respon se as “ prim itive trad itio n alism ,” and m uch o f his research w as an attem pt to explain its transcendence by the cap italist spirit and the cap italist w ork ethic. This traditionalism survives far down into the present; only a human lifetime in the past it w as futile to double the wages o f an agricul­ tural laborer in Silesia who mowed a certain tract o f land on a contract, in the hope of inducing him to increase his exertion. He would simply have reduced by half the work expended because with this half he would have been able to earn as much as before. (Weber, J927: 335) In a recent w ork, an an th rop ologist sum s up som e o f his findings on this subject in term s o f the persistence o f use value practices rather than in term s o f “ prim itive trad itio n alism ” : Recruited as plantation hands, they frequently showed themselves unwilling to work steadily. Induced to raise a cash crop, they would not react “ appropriately” to market changes: as they were interested mainly in acquiring spe­ cific items of consumption, they produced that much less when crop prices rose, and that much more when prices fell off. And the intro­ duction o f new tools or plants that increased the productivity o f indigenous labor might only then shorten the period o f necessary work, the gains absorbed rather by an expan­ sion o f rest than o f output. All these and similar responses express an enduring quality o f traditional domestic production, that it is production o f use values, definite in its aim, so discontinuous in its activity. (Sahlins, 1972: 86) The exotic attitudes and beliefs which socalled traditional cultures exhibit tow ard s the early stages o f penetration by the m odern cap i­ talist econom y often seem bizarre and irratio­ nal to W estern eyes. Luther’s identification o f the devil with capitalism (Brow n, 1959: 2 1 8 19; T aw n ev, 1954: 7 2 -9 1 ), som e M elanesian

449

ca rg o cults, certain m essianic m ovem ents, W eber’s exam ples o f “ prim itive trad itio n al­ ism ,” a s well as the beliefs we consider below , are but som e o f the num erous instances o f such attitudes and concepts. R ather than dis­ m issing these respon ses as “ trad itio n al” or irration al, the approach ad opted in this essay is that it w ould seem to be m ore true to the facts a s well as m ore enlightening to consider these reactions as ou tcom es o f a clash between a use value orientation and an exchange value orientation, thus view ing them as the begin­ ning o f a potential critique o f capitalism . They provide us with insights into the irrational b asis o f our ow n econom y and stereotype o f homo oeconomicus , and can be usefully con ­ sidered as illustrative o f form o f “ prim itive M a rx ism .” T h is “ prim itive M a rx ism ” w as undoubtedly inherent in the ou tlook o f the European pro­ letariat in the early stages o f the birth o f the cap italist system , but has since been largely superseded by a new w orld view which regards the w age con tract system , m arket pricing, and the in stitutionalization o f profit and greed a s natural and ethically com m endable. In the light o f this historical am n esia, which a f­ flicts all social classes in a developed m arker econom y, it is all the m ore im portan t to dwell on the critique offered us by those neophytic proletarian s in the Third W orld today, w ho are ju st entering rhe capitalist system with their g o o d s and labor and w ho often ap p ear to regard that system as anything but natural and g o o d . In the C au ca Valley the sense given to the devil and his role in con tracting w age labor is like the definition o f rhe early Christian fathers as “ he w ho resists the cosm ic p ro c ess,” which in this context com es close to the idea o f forcing things in the interest o f private gain w ithout regard to w hat are seen a s their intrin­ sic principles (cf. N eedh am 19 5 6 : 6 9 -7 1 ). The destruction o f the pre-capitalist m eta­ physics o f production and exchange w as considered by at least tw o influential social theorists a s m andatory for the successful e sta­ blishm ent o f m odern cap italism . M ax W eber regarded the m agical superstition s associated with production and trade to be one o f the greatest ob stacles to the ration alization o f eco­ nom ic life (1 9 2 7 :3 5 5 ), and in his essay on The

450

M IC H A EL TAUSSIG

Protestant Ethic an d the Spirit o f Capitalism often reiterated the point that Labor m u st. . . be performed as if it were an absolute end in itself, a calling. But such an attitude is by no means a product o f nature. It cannot be evoked by low wages or high ones alone, but can only be the product of a long and arduous process o f education. Today, capitalism once in the saddle, can recruit its laboring force in all industrial countries with comparative ease. In the past this was in every case an extremely difficult problem. (Weber, 1958: 62, my emphasis) And as K arl M arx observed, the transition to the cap italist m ode o f production is only com pleted when direct force and external eco­ nom ic con dition s, although still used, are only em ployed exceptionally. An entirely new set o f traditions and habits have to be developed am on g the w orkin g class, to the point where com m on sense regards the new con dition s as natural. It is not enough that the conditions of labor are concentrated in a m ass, in the shape of capital, at the one pole o f society, while at the other are grouped m asses of men who have nothing to sell but their labor power. Neither is it enough that they are compelled to sell it voluntarily. The advance of capitalist produc­ tion develops a working class, which by edu­ cation, tradition, habit, looks upon the conditions o f that mode o f production as selfevident laivs o f Nature. (M arx, 1967, 1:737, my emphasis) H ow ever, the behavior and certain beliefs o f the low er classes on whom this article focuses suggest strongly that the characteristics o f the capitalist m ode o f production are there viewed neither as go o d nor as self-evident law s o f N atu re; in fact they are regarded as unnatural and even evil. The low er classes are o f A fro-Am erican stock, and until 1955 or thereabouts m ost o f them were sm allh olders ow ning parcels o f land w ithout legal title in the southern extrem ­ ities o f the C au ca V alley, an extrem ely fertile region in the southw est o f C o lo m b ia. Since ab o u t 1955 the m ajority o f these peasan ts have had little choice but to w ork full- o r parttime a s w age laborers for the surrounding

sugar plantations which have expropriated much o f the p easan t’s lan d, often through direct physical force. H istorical sources am ply dem onstrate that ever since the abolition o f slavery in 1851 the peasan ts have show n a m arked aversion tow ards w age labor on the large estates o f the rural elite. Until the early decades o f the tw en­ tieth century they were able to retain their position as independent sm allholders and form ed a pow erful political grou p capable o f thw arting their previous ow ners w ho desper­ ately needed lab or to m aintain the flagging estates and gold mines. H ow ever, with the opening o f the valley to the international sea lanes in 19 1 4 , the influx o f vast am ounts o f foreign capital at about the sam e time, and a sharp natural increase in local popu lation , the large estate ow ners have been able to a p p ro ­ priate peasant lan ds and establish a flourishing plantation econom y, a process accelerated by the Violencia o f the early 1 9 5 0 s.1 T oday no m ore than one-fifth o f the p o p u la­ tion in the south o f the valley lives in the countryside, while the rem ainder live in the rural slum tow ns, little m ore than im pover­ ished b arracks for plantation w orkers, bereft o f clean drinking w ater an d adequate sew er­ age. Few o f the tow nspeople ow n land, and m any o f the peasants w ho still occupy farm land also w ork on the plantations. By any stan dards the bulk o f the population is p o o r, and m alnutrition is rife. [■-■] The w orkers them selves . . . feel that they are being constantly cheated, underpaid, and hum iliated. They distrust and hate the p la n ta ­ tions and generally resent the fact that the crops produced on the large estates are d es­ tined for export and not for local co n su m p­ tion. With their trade unions in ruin since the early 1 9 6 0 s, sab o tag e o f mill m achinery and tractors is com m on, and the plantation ow ners and m anagers travel the area in constant fear o f assau lt and kidnapping. Police and soldiers are continually stationed in the mill co m ­ poun ds, and w orkers have been shot during scuffles on paydays. The mill ow ners alw ays travel with an escort o f arm ed g u ard s, and all high-ranking personnel are equipped with tw o-w ay rad ios in com m unication with the

TH E G EN ESIS O F CA PITA LISM A M O N G ST A SO U TH A M ER IC A N PEA SA N TR Y

Defensa Civil and arm y located in the nearest city.

The Devil and Proletarian Labor In the southern C au ca Valley today, it is co m ­ m only believed by local people - tow nspeople and country people, p roletarian s and peasants - that the m ale plantation w orkers som etim es m ake individual and secret con tracts with the devil in order to increase individual ou tpu t and hence their w age. G enerally speaking, it is felt that output can be increased greatly by the individual w orker only if he does m ake such a contract. The landless w age w orkers are widely su p p osed to sell their souls to the devil in order to m aintain, or m ore com m only increase, p ro ­ ductivity. Indian tin m iners in highland Bolivia show sim ilar patterns o f behavior (N ash , 1972). But both grou p s o f people d o not d o this as peasants w orking their ow n land. It is only when they becom e proletarianized that the devil enters the scene in rhis w7ay. While the im agery o f Ciod and g o o d , or the spirits o f N atu re and o f an cestors, dom inate the ethos o f labor in the peasant m ode o f production, the devil and evil perm eate the local m etaphys­ ics associated with the capitalist m ode o f production. It is com m only thought that such a contract can increase a plantation w orker’s yield by tw o to three times. As a result o f such a co n ­ tract, how ever, the individual concerned will die prem aturely and in agon y, and while alive is soulless and a mere puppet in the hands o f the devil. Furtherm ore, certain conditions attach to the m oney thus gained. T his money can n ot serve as productive “ ca p ital” but has to be spent im m ediately on w hat are co n sid ­ ered to be luxury consum er item s such as fine clothes, liquor, butter, and so on. T o invest this money so as to produce m ore m oney i.e., to use it as capital - is to invite ruin. For exam ple, if one buys or rents land, that land will not produce, an d if one buys a piglet to fatten for m arket, the anim al will sicken or die. It is also believed that plan tation su gar cane thus cut will not regrow. The ratoon or root will die, and the plantation

451

land will not produce until plow ed over and replanted. It is crucial to realize that the local peasantry d o not m ake contracts with the devil in order to augm ent or m aintain productivity levels on their ow n plots. N either d o they m ake such a contract when w orking for w ages on the plots o f other peasants. People poin t ou t that such a practice w ould d am age the peasant plot and that the m oney gained could not be used to m aintain equipm ent or land. D espite the poverty that afflicts the m ajority o f the local peasan ts, they are not therefore tem pted by the m onetary rew ards that result from the devil con tract, since those rew ards are thought to dim inish or curtail natural fertility. The notion o f extractin g m ore m oney from nature in the above w ay applies only to p ro letarian s.2 The only m agic allegedly used on the peasant plots is “ g o o d ” m agic w orked through the C ath olic saints and the dnimas - the souls o f the an cestors - and such m agic is aim ed at curing the plot o f m align influences in general and a s a proph ylaxis again st thieving in p a r­ ticular. Petty retailers and m iddlem en a lso use m agic to ensure success, but once again this is considered to be “ g o o d ” m agic and is w orked through the m edium o f the Saints and the

dnim as?

Baptism of Money and the Secret of Interestbearing Capital A ccording to the belief in el bautizo del billete (baptism o f the bill), rhe G odparen t-to-be con ­ ceals a peso note in his or her hand during the baptism o f the child by the C ath o lic priest. The peso bill is thus believed to receive baptism instead o f the child. When such a baptized bill enters into general m onetary circulation it is believed that it will continually return to its ow ner with interest, enriching the ow ner and im poverishing the other parties to the deals transacted by the ow ner o f the bill. The ow ner is now the G odp aren t o f the peso bill. The child rem ains unbaptized, a cause o f great concern since the child’ s soul is denied super­ n atural legitim acy and has no chance o f esca p ­

452

M IC H A EL TA U SSIG

ing from Lim bo or Purgatory, depending on when it dies. T h is practice is heavily penalized by the Church. T he baptized bill receives the nam e - the “ Christian n am e” as we say in English - that the baptism al ritual w as m eant to bestow on the child and is now referred to by that name. It is then set to w ork as follow s. T he G o d p a r­ ent pays rhe bill over as part o f a routine m onetary tran sactio n , as when one p ays for g o o d s in a store. T he G o d p aren t m utters the follow ing type o f refrain: Jose £te vas o te quedas?