(methodology and history in anthropology) robert parkin and anne de sales (editors)-out of the study...

OUT OF THE STUDY AND INTO THE FIELD

Volume 1Marcel Mauss: A Centenary Tribute Edited by Wendy James and N.J. Allen

Volume 2Franz Baerman Steiner: Selected WritingsVolume I: Taboo, Truth and ReligionFranz B. Steiner. Edited by Jeremy Adler andRichard Fardon

Volume 3Franz Baerman Steiner: Selected WritingsVolume II: Orient politik, Value, and Civilisation.Franz B. Steiner. Edited by Jeremy Adler andRichard Fardon

Volume 4The Problem of Context: Perspectives fromSocial Anthropology and ElsewhereEdited by Roy Dilley

Volume 5Religion in English Everyday Life: An Ethnographic ApproachBy Timothy Jenkins

Volume 6Hunting the Gatherers: EthnographicCollectors, Agents and Agency in Melanesia,1870s–1930sEdited by Michael O’Hanlon and Robert L. Welsch

Volume 7Anthropologists in a Wider World: Essays on FieldResearchEdited by Paul Dresch, Wendy James andDavid Parkin

Volume 8Categories and Classifications: MaussianReflections on the SocialBy N.J. Allen

Volume 9Louis Dumont and Hierarchical OppositionBy Robert Parkin

Volume 10Categories of Self: Louis Dumont’s Theoryof the IndividualBy André Celtel

Volume 11Existential Anthropology: Events, Exigenciesand EffectsBy Michael Jackson

Volume 12An Introduction to Two Theories of SocialAnthropology: Descent Groups and MarriageAllianceBy Louis Dumont. Edited and Translated byRobert Parkin

Volume 13Navigating Terrains of War: Youth andSoldiering in Guinea-BissauBy Henrik Vigh

Volume 14The Politics of Egalitarianism: Theory and PracticeEdited by Jacqueline Solway

Volume 15A History of Oxford AnthropologyEdited by Peter Rivière

Volume 16Holistic Anthropology: Emergence andConvergenceEdited by David Parkin and Stanley Ulijaszek

Volume 17Learning Religion: Anthropological ApproachesEdited by David Berliner and Ramon Sarró

Volume 18Ways of Knowing: Anthropological Approachesto Crafting Experience and KnowledgeEdited by Mark Harris

Volume 19Difficult Folk? A Political History ofSocial AnthropologyBy David Mills

Volume 20Human Nature as Capacity: TranscendingDiscourse and ClassificationBy Nigel Rapport

Volume 21The Life of Property: House, Family andInheritance in Béarn, South-West FranceBy Timothy Jenkins

Volume 22Out of the Study and Into the Field:Ethnographic Theory and Practice in FrenchAnthropologyEdited by Robert Parkin and Anne de Sales

Methodology and History in Anthropology

General Editor: David Parkin, Fellow of All Souls College, Oxford

OUT OF THE STUDY AND INTO THE FIELD

Ethnographic Theory and Practice in French Anthropology

Edited by

Robert Parkin and Anne de Sales

Berghahn BooksNew York • Oxford

First published in 2010 by

Berghahn Books

www.berghahnbooks.com

©2010 Robert Parkin and Anne de Sales

All rights reserved. Except for the quotation of short passages

for the purposes of criticism and review, no part of this book

may be reproduced in any form or by any means, electronic or

mechanical, including photocopying, recording, or any information

storage and retrieval system now known or to be invented,

without written permission of the publisher.

Library of Congress Cataloging-in-Publication Data

Out of the study and into the field : ethnographic theory and practice in French

anthropology.

p. cm. -- (Methodology and history in anthropology vol.22)

Includes bibliographical references and index.

ISBN 978-1-84545-695-5 (hardback : alk. paper)

1. Anthropology--France--Philosophy. 2. Anthropology--France--Field work. 3.

Anthropology--France--Methodology. I. Parkin, Robert, 1950- II. Sales, Anne de.

GN585.F8O88 2010

301.010944--dc22

2010018543

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Printed in the United States on acid-free paper

ISBN: 978-1-84545-695-5 (hardback)

Parkin & DeSales text_v4:Layout 2 5/25/10 4:51 PM Page iv

CONTENTS

List of illustrations vii

List of authors discussed in this volume ix

Preface xi

Introduction: ethnographic practice and theory in France 1


1. ‘Keeping your eyes open’: Arnold van Gennep and the autonomy

of the folkloristic 25

Giordana Charuty

2. Canonical ethnography: Hanoteau and Letourneux on Kabyle

communal law 45

Peter Parkes

3. Postcards at the service of the Imaginary: Jean Rouch,

shared anthropology and the ciné-trance 75

Paul Henley

4. Eric de Dampierre and the art of fieldwork 103

Margaret Buckner

5. What sort of anthropologist was Paul Rivet? 125

Laura Rival

6. Alfred Métraux: empiricist and romanticist 151

Peter Rivière

7. Roger Bastide or the ‘darknesses of alterity’ 171

Stefania Capone

8. The art and craft of ethnography: Lucien Bernot, 1919–1993 197

Gérard Toffin

9. André-Georges Haudricourt: a thorough materialist 219

Alban Bensa

vi Contents

10. Louis Dumont: from museology to structuralism via India 235

Robert Parkin

11. Will the real Maurice Leenhardt please stand up? Four anthropologists

in search of an ancestor 255

Jeremy MacClancy

Notes on contributors 273

Subject index 277

Name index 289

LIST OF ILLUSTRATIONS

Figures

1.1. Arnold van Gennep, aged 80, lighting a bonfire on the summer solstice

(21 June 1953), also the saint’s day of St John the Baptist. 26

2.1. Commandant Adolphe Hanoteau at Fort-Napoléon, 1861. 47

2.2. Opening column of qanun rulings transcribed in Arabic with French

translation by Captain Alphonse Meyer, military interpreter at Dellys, 1864. 50

2.3. Dedication of Ait Iraten qanun rulings submitted to Hanoteau by Si Mula,

ca. 1859/60. 52

2.4. Submission of the Kabyle tribes to Marshall Randon in 1857. 64

2.5. Hanoteau and two Tuareg, 1858. 65

3.1. Jean Rouch shooting in a market in the Gold Coast in 1954. 76

4.1. Eric de Dampierre at Madabazouma, thirty kilometres from

Bangassou, Central African Republic. 104

5.1. Paul Rivet. 126

6.1. Alfred Métraux, seated second from right, doing fieldwork among

Chipaya in Bolivia, 1931 or 1932. 152

7.1. Roger Bastide, seated left on bench, Ifanhin, Bénin, 1958. 172

8.1. Lucien Bernot, on the occasion of his being honoured with a Festschrift

at the École des Hautes Études en Sciences Sociales, Paris, 1987. 198

9.1. André-Georges Haudricourt: ‘le maître à la recherche de la petite bête

ou le maître dans l’exercice de ses fonctions’, June 1972. 220

10.1. Louis Dumont, taken by himself, among the Kallar, Tamil Nadu (India),

with his chief informant, Muttusami Tevar, 1949. 236

11.1. Maurice Leenhardt (back row, centre), with Melanesian pastors

during a conference, Nouvelle Calédonie 1916. 256

LIST OF AUTHORS DISCUSSED IN THIS VOLUME

Authors dealt with in this volume, their dates and main fieldworkarea(s) and/or peoples of interest.

Author Dates Main fieldwork area(s)and/or peoples ofinterest

Roger Bastide 1898–1974 Afro-Brazilians (north-east Brazil)

Lucien Bernot 1919–1993 Marma, Cak(Bangladesh, Myanmar)

Eric de Dampierre 1929–1997 Nzakara (Central AfricanRepublic)

Louis Dumont 1911–1998 IndiaAdolphe Hanoteau 1814−1897 Kabyle (Algeria)André-Georges Haudricourt 1911–1996 VietnamAristide Letourneux 1820−1890 Kabyle (Algeria)Maurice Leenhardt 1878–1954 New CaledoniaAlfred Métraux 1902–1963 Argentina and other

South America, Haiti,Easter Island

Paul Rivet 1876–1958 Ecuador, ColombiaJean Rouch 1917–2004 Songhay (Niger)Arnold van Gennep 1873–1957 France, Europe

PREFACE

The present volume originated in a conference, ‘Out of the Study and intothe Field’: Ethnographic Theory and Practice in French Anthropology,held at the Maison Française d’Oxford on 22–24 April 2005. It does not,however, represent the formal proceedings of the conference, since someof the original contributors have not been able, for a variety of reasons,to submit the papers they gave on that occasion for inclusion here.Conversely, the chapters by Buckner, Capone and Parkin were writtensubsequent to the conference especially for this volume.

The chapters by Bensa, Capone and Charuty have been translated fromthe French, the first by Amy Jacobs, the second and third by the editors.Quotations from texts originally written in French have either beentranslated by the authors of those chapters or the editors, or else replacedby the equivalent passage from an existing published English translation.Due to the rarity of some of the original French texts in Rivière’s paper,there the original French texts have been retained in footnotes.

The editors wish to thank the contributors to both the originalconference and the present collection, where these are different, as wellas the staff and management of the Maison Française for providing theconference venue and refreshments. The conference was supported by agrant from the British Academy, which is gratefully acknowledged. Wealso thank the publishers of this collection, Berghahn Books, especiallyMarion Berghahn, as well as Prof. David Parkin, the series editor, for theirsupport of this project. We are also grateful to those contributors whocommented on the introduction to the volume, and to the twopublishers’ reviewers for their very useful reviews of the whole volume,even though we have not felt able to incorporate all their observations.

The editors are also grateful to the following organisations andindividuals for helping them obtain the plates used in this volume:L’Agence photographique de la Réunion des musées nationaux (France)for the portrait of van Gennep; Harold Prins for the portrait of Métraux;Alex Baradel, Fundação Pierre Verger (Brazil), for the portrait of Bastide;Jean-Claude Galey, for the portrait of Dumont; and Christophe Dervieux,Archiviste, Direction des affaires culturelles et coutumières, Service desarchives, Noumea (Nouvelle Calédonie), for the portrait of Leenhardt.

xii Preface

Last but not least, this whole project has proved to be a smooth andconvivial joint venture between British and French anthropology,represented here by the respective editors, who both feel they have areasonable knowledge, understanding and appreciation of each other’snational anthropological traditions and have enjoyed working together.They would therefore like to take this opportunity of thanking each other.

Robert ParkinAnne de Sales

Oxford, October 2009

Introduction

ETHNOGRAPHIC PRACTICE ANDTHEORY IN FRANCE


Introduction

Rather like the nations they represent, there is a sense in which whatpass as the British and French schools of anthropology really are eachother’s Other: on both sides of the Channel, there is a wary awarenessof the other’s alleged achievements and failings, perpetually shaped bya strong feeling of, and for, difference and distinctiveness. Perhaps thissense of respectful rivalry was first expressed aptly back in the latesixteenth century when, in a passage from Astrophel and Stelladescribing what appears to be a joust, a minor but very EnglishElizabethan poet, Sir Philip Sidney, referred to ‘that sweet enemy,France’. Be that as it may, it is clear that British anthropologists have along history of being influenced by their French colleagues in a wholeseries of disciplines, often despite themselves, and often in reaction tothem rather than accepting their teachings wholesale. The list is long:even a partial one would have to include at least Durkheim, Mauss,Lévi-Strauss, Barthes, Althusser, Ricoeur, Dumont, Merleau-Ponty,Bourdieu, Foucault, Lacan, Baudrillard, Derrida, and more recently deCerteau and Latour.

However, as this collection is intended to demonstrate, somepowerful but often distorting stereotypes have been at work here. Thisgaze from across the English Channel has given rise to two commonlinked impressions about French anthropology among the British.1 The

first is that it is dominated by theory based mainly on rationality anddeductive reasoning. Secondly, and conversely, it is commonly said notto be very concerned to derive general principles inductively fromethnographic facts. The latter, of course, is often thought to be thestrength of the British tradition in particular, which also likes to thinkof itself as cultivating a healthy scepticism of theory. Indeed, it is hardto think of a major British contributor to theory who has not been, atsome time or other, a fieldworker too.

This is far less true of the French school, notwithstanding, forexample, Lévi-Strauss’s travels around the Amazon. However, this isless because the French are all theorists than because, au contraire, agood many of them are ethnographers obsessed with the facts anddismissive of theory, to the extent that they might be described as‘ethnographic essentialists’. Accordingly, we argue that there is asharper distinction, and disjunction, between theory and ethnographicpractice in France than in Britain, where, as just noted, manyanthropologists have seen it as their task to contribute to bothsimultaneously.

The British editor of this volume still remembers being struck by thenovelty of this discovery, which came as a revelation after years of hisviewing French anthropology as excessively theoretical and almostanti-empirical, in accordance with the prevailing British stereotype.2i

Indeed, so-called ‘British empiricism’ is frequently trumped by theethnographic essentialism purveyed by many of the figures treated inthis collection. Is not the conventional British view of Frenchanthropology therefore seriously distorted? Are not the grand theorists,who are mostly anyway associated with other disciplines, falsely andperversely seen as being more representative of French anthropologythan those who have pursued their profession in the field as much as inthe study, if not more so? These are the main questions we are asking inthis volume.

We fully acknowledge that this situation has nothing to do with anylack of theoretical awareness or competence generally among Frenchethnographers, as Lucien Bernot showed in his brief but pungentdismissal of structuralism (discussed below). Moreover, the quality oftheir ethnographic work is undoubtedly as high as in other traditions.Nor do we wish to exaggerate this tendency in France, far less claim thatit has been the only approach to fieldwork there, nor indeed suggest thatit is entirely absent outside the country. Dumont, as well as the FrenchMarxist anthropologists – both those who were influenced mainly byAlthusser, such as Emmanuel Terray, Claude Meillassoux and PierrePhilippe Rey, as well as Maurice Godelier, famous for his attempts tocombine Marxism with structuralism – all did fieldwork and had a clear

2 Robert Parkin and Anne de Sales

theoretical framework within which to do so (which, however, was oftenseen by others as directing, rather than reflecting, the search for facts).Similarly, the research team set up by Louis Dumont and later takenover by Daniel de Coppet brought together a number of French andinternational anthropologists who had done fieldwork in different partsof the world and asked them to frame their work in relation to Dumont’stheoretical notions of hierarchy, value and hierarchical opposition. Yeteven Dumont, who perhaps comes closest to what we see as typicalBritish practice, liked to present himself first and foremost as acraftsman or technician (Delacampagne 1981: 4). We therefore arguethat ethnographic essentialism represents a distinct but not exclusivetrend in French anthropology, one based not just on a simple disinterestin theoretical positions but a positive rejection of them. In fact, thistendency seems every bit as characteristic of the French school as thetheory-heavy ruminations of those thinkers we have all learned toknow and, sometimes, even love so well.

What are the reasons for this? Any assessment has to be based on thehistory of fieldwork and of field enquiries generally in Frenchanthropology. In the rest of this introduction, we provide a brief surveyof this history, starting with the early nineteenth century and proceedingto the heyday of structuralism.3 As we shall see, one trajectory ofsignificance here is a series of shifts from learned societies to museums toresearch and training institutes, only finally reaching the universities ata relatively late stage. We then proceed to provide a brief overview of eachchapter before considering what commonalities and differences can bediscerned in the lives, careers and works of these subjects.

Fieldwork in French anthropology: a brief history

An interest in field enquiries in France can be traced back to around1800, when the short-lived Société des Observateurs de l’Hommepromoted the use of anthropological questionnaires by travellers toother parts of the world and issued guidelines for anthropologicalenquiries. This was the era of antiquarian and other learned societies inFrance, as elsewhere in Europe, that is, of amateur intellectuals andcollectors working in an intellectual environment that was only thenbeginning to institutionalise itself. At this early stage, French universitieswere hardly involved directly at all in either teaching or research inanthropology, and it was a museum, the Musée d’Histoire Naturelle, thatestablished the first chair in anthropology in 1855, in cultural as well asphysical anthropology (Gaillard 2004: 85). Later in the nineteenthcentury, however, in 1878, the first anthropological museum was

Introduction 3

founded in France, namely the Musée d’Ethnographie, housed in theTrocadéro, by which name it was commonly known.4 Although initiallyfocused on pre-Columbian New World artefacts − the chief interest of itsfirst curator, Ernest-Théodore Hamy − the expansion of the FrenchEmpire soon encouraged collection elsewhere and, along with it, basicfieldwork yielding highly factual ethnographic monographs. Thispromoted rather than initiated such activities, which were already goingon, for example, in Senegal in the 1850s, where General Faidherbe wasalready busy producing anthropological and linguistic studies of itsindigenous peoples (Gaillard 2004: 86). In addition, many missionarieswere also active in this period in various parts of the world, such as JeanKemlin, who went out to the Bahnar in Vietnam in the same decade,long before French rule had been established there. Apart from a crudecolonial-style evolutionism, none of this work can be consideredtheoretically informed. However, methodologically attempts werealready being made to supplement earlier, purely biological approachesto the study of humankind with a specific perspective on culture(promoted, among others, by Hamy and his colleague in setting up theTrocadéro museum, Armand de Quatrefages), as well as to treat thecollection and display of anthropological objects as scientific, not artistic,in character. Even at this early stage, therefore, a certain separationbetween ethnography and theory can be discerned in France.

Other currents in the nineteenth century can be linked to Franceitself, or at any rate Europe, rather than growing overseas empires. Inearly sociology of the mid-nineteenth century, Frédéric Le Play’ssurveys, made as part of his roving work as a mines inspector, producedinsights into, or at least theories concerning, the nature, evolution andsustainability of family forms. Perhaps of greater influence were studiesinto the folklore of France in this period and later. Though dating backwell into the nineteenth century, like early anthropology, folklorestudies were also stimulated subsequently by the founding of amuseum, this time the Musée des Arts et Traditions Populaires, byGeorges-Henri Rivière and André Leroi-Gourhan in 1937. A goodexample is Louis Dumont’s study La Tarasque, a festival in southernFrance (Dumont was at one time an employee of the aforesaidmuseum). The main figure here, though, is Arnold van Gennep, ahighly active fieldworker whose major work in sheer scale was hismultivolume Manuel de folklore française contemporaine (1938–1958).However, as can be seen from Giordana Charuty’s chapter in thisvolume, van Gennep is really a transitional figure who attempted totransform the folklore of France from a concern with origins andsurvivals to synchronic studies that were more in tune with post-evolutionist trends in anthropology more generally. In doing so, he


resisted the attempts of the French political right to enlist folklore forits own nationalist agenda, as well as becoming almost a structuralistavant la lettre in his most famous work, Les rites de passage (1909; onritual forms in the world in general). For Susan Rogers, this fusion offolklore and anthropology still informs the anthropology of Franceitself, partly because of a desire to challenge sociological studies of thedeath of rural France by stressing the uniqueness and continuedviability of such communities (2001: 490–91). Indeed, some of thefigures treated in this volume took part in studies of Frenchcommunities before moving on to fieldwork in other parts of the world(Bastide, Bernot, Dampierre, Dumont). But also, writers like FrançoiseZonabend and Martine Segalen used a combination (variously) ofmaterial culture, historical documents, oral histories and literatures,and anthropological fieldwork in their histories of the family in differentparts of France – an interest that can be traced back to Le Play’ssurveys. Yet even in Les rites de passage, what we have just called vanGennep’s structuralism was adventitious rather than programmatic,and facts predominate over grand theory in the bulk of his work, apartfrom an interest in the experience of fieldwork itself.

This practice of separating fieldwork and theory persisted into thetwentieth century in France, where anthropology as a distinct disciplinedeveloped differently than it did in Britain and America, especially inturning to professional fieldwork rather later. In the early twentiethcentury, however, fieldwork by amateur missionary and administratorethnographers still continued. One representative figure is LéopoldSabatier, active in producing legalistic coutumiers, or compendia oftribal custom, in the highlands of Vietnam. Work in this part of theFrench Empire was supported by the École Française d’Extrème-Orient,set up in Hanoi in 1898 as a research institute. Perhaps the mostfamous figure here, however, is Maurice Leenhardt (see MacClancy, thisvolume), though he is not entirely typical: in returning to France andteaching anthropology as part of Mauss’s circle between the wars –after living in and writing on New Caledonia for many years – he, atleast, can be said to have made the transition from amateur toprofessional status in his career.5

Nonetheless, in the main, fieldwork by professional academics cameto France later than in Britain or America. One factor here wasobviously the dominance of Durkheimian sociology, which for a longtime was deeply suspicious of the term ‘anthropology’ and anyone oranything to do with it. First, it was seen as having been discredited bythe speculations of the nineteenth-century British intellectualists-cum-evolutionists – for Durkheimians, one of the main examples of wrong-headedness in the social sciences of the time. Secondly, it was too closely

Introduction 5

connected with amateur, antiquarian folklore. This attitude is reflectedin the group’s hostility to van Gennep, who sullied his reputation stillfurther in their eyes by using ethnography to criticise Durkheim’s viewson totemism (van Gennep 1920). It may also be found in the criticismthat Robert Hertz, a leading Durkheimian scholar, faced from his owncolleagues after conducting a brief period of fieldwork on the cult of StBesse in northern Italy in 1911 (Parkin 1996: 12, MacClancy andParkin 1997). Consistently, even in the case of what had already longbeen a central anthropological topic like religion, the Année sociologiquegroup saw their work as sociology, not anthropology, despite theirincreasing use of ethnography.

After his uncle’s death, though, Mauss eventually overcame thesescruples, at least in part. Conscious that French anthropology wasfalling behind British in this regard, he encouraged others to do long-term fieldwork in the 1930s without participating in any himself.6 Thiswas reflected in, and perhaps also reinforced by, Mauss’s and others’activities in teaching the virtues of ethnography to French colonialofficers and trainees for administrative positions. Such activity, oneassumes, would not give emphasis to theory. Mauss taught thesecourses at the Institut d’Ethnologie, which had been set up for thepurpose by his friend Lucien Lévy-Bruhl, and with which a whole rangeof key figures in the history of French anthropology were involved,including Leroi-Gourhan, Paul Rivet and Maurice Delafosse.

Mauss’s Manuel d’ethnographie (1947), which has recently beentranslated (2007), was also linked to these activities (having been usedfor lecturing prior to publication). In fact, a scrutiny of some of his moreprogrammatic statements indicates that he, more than anyone elseexcept perhaps Marcel Griaule, was the probable source of thewidespread focus on the facts and on ethnography in much Frenchanthropology after the First World War. In the Manuel, Mauss callsethnology ‘a science of facts and statistics’, its aim being ‘the knowledgeof social facts’ (2007 [1947]: 7). Further, ‘comparative ethnography’should be ‘based on comparison between facts, not between cultures’(ibid.: 8). Earlier too, in an ‘Intellectual self-portrait’ evidently writtento support his candidature to the Collège de France in 1930 (Mauss1998), he states repeatedly that ‘the facts’, or alternatively ‘description’,have enjoyed the priority in his work over theory. Thus right at the startof this self-evaluation, he describes himself as ‘a positivist, believingonly in facts’, and asserts that ‘descriptive sciences attain greatercertainty than theoretical sciences’ (1998: 29). Similarly, in contrastto some of his other activities, at the Institut d’Ethnologie, ‘I havealways confined my teaching to the purely descriptive’ (ibid.: 32). Themain aim of himself and his collaborators over the past four years has


been ‘to promulgate and often to establish the facts deriving fromunclassified civilizations’ with a view to classifying them (ibid.: 34).7

Finally, ‘the only objective of the discipline to which I have devotedmyself ’ has been to show ‘the place of social life ... in the life ofhumanity’ through ‘sensitive contact with the facts’ (ibid.: 42). Perhapsthe admission that ‘discoveries and novelties were a constant delight’(ibid.: 36), with the hint that processing them further throughclassification and theory were less exciting, had something to do withthe development of this attitude. Certainly, in reading theseformulations from Mauss’s pen, one acquires a distinct feeling thattheory is secondary in his view of his own work and its aims – asurprising realisation, in the light of his long and intimate associationwith one of the supreme sociological theorists, his uncle Durkheim.8

Apart from Mauss’s teaching, another stimulus to anthropology inthis period was the Colonial Exhibition, organised by Marshal Lyautey,a key French Empire-builder, and held at Vincennes outside Paris in1931. A celebration as much as exhibition of the French Empire and itscultural variety, it attracted millions of visitors and stimulated both aninterest in anthropology in the general public and a desire to do morefieldwork among a growing class of professional ethnographers (seeL’Estoile 2003, 2007). Yet, this was also the period of expeditions andethnographic travel at least as much as fieldwork in the Malinowskiansense, the former method sometimes being allied with diffusionism, ashad been the case about a quarter of a century earlier with, forexample, the Torres Straits expedition in Britain. Thus the famousDakar-Djibouti expedition of the early 1930s, led by Marcel Griaule,was soon followed by Lévi-Strauss’s travels around the Amazon later inthe decade, though the latter, of course, were put to the service ofstructuralism. As for Griaule, he did much to popularise anthropologyin France, both before and after the Second World War, partly throughhis own charisma as a teacher and partly through the quite large cohortof his colleagues and students he gathered around him. Many of thesewere significant figures in their own right, such as Michel Leiris (whosoon broke with him), but also Marcel Delafosse, Germaine Dieterlen,Denise Paulme and Jean Rouch (on the latter, see Paul Henley, thisvolume). Although Griaule himself has been accused of exploitinginformants in questionable ways and of indulging in culturalreproduction rather than ethnographic reporting by deliberatelystaging ritual events, he abandoned his early diffusionism in favour ofa focus on the field and a theorising of field methods.9 And underGriaule’s influence, members of this group at least spoke up for thevalidity of indigenous ideas and ways of life, often comparing themfavourably with ‘Western civilisation’.

Introduction 7

Although Griaule’s influence persisted after the Second World War,there was certainly a change of emphasis with the arrival ofstructuralism. This was a method rather than a theory in Lévi-Strauss’sown view, though not one specifically directed towards fieldwork.Nonetheless it rapidly came to be treated as a theoretical tendency, if nota school. Lévi-Strauss’s influences were many and varied, and were notconspicuously dominated exclusively by previous periods ofanthropology in France. Of course, the Année sociologique school,especially Mauss, was a key influence, but so were the structurallinguistics of Ferdinand de Saussure and Roman Jakobsen. In addition,the cultural anthropology, if not entirely the cultural relativism, of theBoas school influenced Lévi-Strauss, who had been exposed to it duringhis wartime exile from France at the New School of Social Research inNew York. In his critiques too, his target was British structural-functionalism more than anything else in anthropology. Above all, hisaim of creating a science of culture on the model of structural linguisticswas explicitly a break with the past. This was also a period in whichanthropology became more rooted in the universities in France, togetherwith research groups in, for example, ORSTOM (Organisation pour laRecherche Scientifique et Technique de l’Outre-Mer)10 and, perhaps mostimportantly, CNRS (Centre National de la Recherche Scientifique).

Lévi-Strauss himself was, of course, at the opposite extreme toethnographic essentialism, using structuralist theory to explainethnographic facts rather than vice versa (he is perhaps the most explicitlydeductive of all major international anthropologists). His influence was suchthat the fieldwork of others and the facts they collected began to be shapedand organised in relation to his theoretical ideas. Key figures here, who alldid proper fieldwork in relation to various theoretical agendas, include Luc deHeusch, Françoise Héritier and Philippe Descola. As already noted above, intandem and, through Maurice Godelier, even overlapping with structuralismwas the work of mostly Althusser-inspired Marxists like Terray, Meillasouxand Rey, chiefly on West Africa. Here too, theory (Marxist this time) was usedto explain ethnographic facts rather than vice versa. With structuralism andMarxism, therefore, French anthropology converged more with practice inother national traditions of anthropology in intimately uniting theory andpractice, and even in subordinating the latter to the former.

However, we should not forget that both structuralism and Marxismco-existed with other intellectual currents: the psychoanalysis ofJacques Lacan; the phenomenology of Maurice Merleau-Ponty; theexistentialism of Jean-Paul Sartre; the historical sociology andphilosophy of Michel Foucault; the contemporary sociologies ofGeorges Gurvitch and Pierre Bourdieu; postmodernism; archaeologyand material culture; the alternative, non-structuralist anthropologies


of Georges Balandier or Eric de Dampierre; the cognitive anthropologyof Dan Sperber and the continuance of ethnographic essentialism insuch figures as Rouch, Lucien Bernot, André-Georges Haudricourt (alltreated in this volume) and Georges Condominas.

French studies of fieldwork in French anthropology

To what extent have these issues − namely the part fieldwork has playedin the history of French anthropology and its relationship to theory −been addressed in France itself? In fact, several important publicationshave recently tackled these issues from various points of view. ThusClaude Blanckaert has produced a historical perspective on thetransformation of the status of the observer in the course of the past threecenturies in a collection of studies of texts, basically French, which enactresearch directives and codify the empirical work of travellers and, afterthem, researchers (Blanckaert 1996). Daniel Céfaï has brought togetherfourteen classic British and American texts, translated into French, onthe subject of the field, participant observation and ethnographicdescription, with an important postface devoted especially to Frenchworks on these questions (Céfaï 2003). Four manuals directed at studentson methods of enquiry have also appeared.11 Moreover, the last ten yearshave seen a revival of studies on the social sciences in colonial situationswhich take the view that colonialism was ‘constitutive’ of these disciplinesrather than ‘disqualifying’ them as legitimate modes of intellectualenquiry. Thus four recent studies deal with the research actors, colonialadministrators, indigenous scholars, official and unofficial researchers,and institutions involved in colonial research.12 In plunging actors intothe heart of colonial realities, the field appears as a crucial experience tobe taken into account in reconstructing the history of the social sciences.

Benoît de L’Estoile in particular (see notes 4 and 12) has focused onthe links in France between anthropological museums, anthropologyas a ‘scientific’ discipline and the politics of empire and, more recently,on global multiculturalism and the place of France within it. His periodtherefore begins with the Colonial Exhibition of 1932 and the creationof the Musée de l’Homme six years later, and ends with the transfer ofthe latter’s collections to the new Musée du Quai Branly in 2005. He isespecially critical of claims that such museums are all about displays ofalterity, pointing out how, instead, they really represent western ideasof the Other rather than the Other itself, and also seeing continuity, nota break, in the transition from the Trocadéro to the Quai Branly. This,of course, is a dilemma for anthropology generally, and it is especiallysignificant in fieldwork, where not only are facts and impressions

Introduction 9

collected, but also the Other is confronted on a human level of mutualcomprehension and incomprehension. For L’Estoile, therefore,museums should be sites for the display of relations between collectorsand collected, and avoid either an explicit focus on the Other or aconcealed focus on western perspectives of the Other.

These works have done something to make good the lack of anyFrench histories of French anthropology, a lack highlighted, forexample, by Jean Jamin in the introduction to a collection (Copans andJamin 1994 [1978]) of very early texts produced under the auspices ofthe Société des Observateurs de l’Homme of the late eighteenth andearly nineteenth centuries. Mention should also be made of the seriesTerre Humaine, published in Paris by Plon over many years, the focus ofwhich was precisely the publication of ethnographies in French.13

Nonetheless, all this is still something of a closed book to the worldoutside France. While we do not engage directly with these texts here,we do seek to supplement them with a wholly English-languageperspective on the particularities of the relationship betweenethnographic practice and theory in French anthropology.

The present collection

The approach adopted in addressing this question was to ask French andBritish anthropologists to compose intellectual biographies of Frenchanthropologists, some of them little known, if at all, to the Anglo-Saxonpublic, yet who offer particular potential in exploring the relationshipbetween ethnography and theory. We chose to focus on actualpractitioners of anthropology rather than on movements or schools,meaning that, in relation to his or her subject, each contributor has hadto make more complex a picture that the ‘international commerce ofideas’ (Cusset 2008 [2003]) tends to simplify, even to caricature. Hencethe eclectic character of this gallery of portraits when compared to eithera manual of ethnographic practice or a history of the discipline. Also,despite Rivet’s involvement with the Trocadéro and the interests of someof those featured here in material objects (especially Bernot andHaudricourt), this is not a volume about French museology.

Thus the present collection is selective rather than comprehensive. Itis unfortunate that there is no chapter on a female Frenchanthropologist. This partly reflects the principle we chose to adopt of notfeaturing any living anthropologists in this collection, which restrictedus in large measure to the middle and early histories of Frenchethnography – and these periods in France appear to have had evenfewer women fieldworkers than the British and American schools. Many


French women ethnographers, now deceased, such as GermaineDieterlen and Denise Paulme, were linked to Griaule, a circle representedhere by Jean Rouch. Outside this circle was Germaine Tillion, a muchdiscussed figure in France itself in recent years for her fieldwork in theAurès area of Algeria and her political activism as a supporter of andmediator for the resistance movement against French rule, as well ashaving been a resistance fighter earlier against the Germans in theSecond World War (see Todorov and Bromberger 2002, Todorov 2007).

A main thrust of these chapters is therefore historical. Is theethnographic essentialism of many of the figures dealt with in thisvolume now similarly historical? In fact, given what has been identifiedas the general tendency for anthropologists to refrain from large-scalecomparisons and theoretical statements today (Gingrich and Fox2002), with a concomitant concentration on the facts of specificethnographic situations, ethnographic essentialism appears rather tobe alive and kicking in at least some quarters. In addition, of course, itcannot be said that the fundamental problems of doing fieldwork havegone away, nor that the basic process itself has changed markedly sincethe time those discussed in this collection were themselves in the field,despite the distinctive attitudes of many of them to fieldwork. The timetherefore seems right to draw attention to this tendency once more inthe context of the past practices of some though not all adherents ofthe French tradition, in the belief that, in a more general way too, theirexperiences and their own telling of them remain very relevant tocontemporary anthropology. A review of the chapters follows, whichare arranged broadly according to the ethnographic areas in whichtheir subjects mainly or wholly worked.

The first chapter in the collection focuses on a key figure in thetransition from folklore to a recognisable anthropology of symbolism andritual, Arnold van Gennep (1873–1957). Charuty shows that vanGennep did not accept his dismissal by the Durkheim school as a ‘merefolklorist’ lying down. Indeed, it produced a reaction in him whichconceded nothing to the theoretical peculiarities of his rivals, whileoutperforming them in relation to his greater ‘feel’ for ethnographicrealities and the problems involved in both eliciting and reporting theseproblems in the field. Being almost entirely armchair anthropologists, hisrivals were especially vulnerable to attacks of this kind. Much of thisreaction was formulated in the Chroniques pages of the Mercure de France,but these pages were not only critical of others, they also put forward aprescription for how fieldwork in a literate or semi-literate society shouldbe carried out. Thus neutral observation should be coupled withinformants’ memories and life histories; as a fieldworker, one shouldmaintain an intellectual distance, while also being exposed fully to the

Introduction 11

exotic world one is examining; and such experiences should be embodiedin one’s own self, in a manner that almost suggests a form ofphenomenology. Also, van Gennep addressed the often problematic statusof fieldwork questions, answers and other methods. For example, intalking about ‘fake rituals’ – that is, performances in the form of festivalsput on to support conservative nationalist agendas in rural France – hecame close to the idea of the ‘invention of tradition’ (for van Gennep, onlythe rituals the people put on for themselves were ‘authentic’).

Charuty points out the centrality of the rite in van Gennep’sapproach to the whole ethnographic project. For him ritual is, amongother things, a manifestation of universal structure, marked not only bythe famous three stages, but also by transition and by the marginalityof the central, liminal stage. It is hard, therefore, to avoid remarking onthe double irony that van Gennep himself represents not onlyintellectual transition in his work, but also marginality in respect of hisown institutional destinies.

In his chapter, Peter Parkes examines the contribution of twocolonial functionaries, Adolphe Hanoteau and Aristide Letourneux, tothe early ethnography of the Kabyle Berbers of Algeria and to thedistinctive genre of what Parkes calls ‘canonical ethnography’(indigenous juridical documentation and its analytical interpretation).This was based on legal ‘canons’ or qawanin, a neglected but valuableform of early ethnographic documentation, and the prototype of lateradministrative ethnographies in sub-Saharan Africa. The work of thesetwo officials, comprising a gazetteer of general information about thearea and its people, together with their legal customs and social systems,was collected through a peculiarly intensive kind of ‘dialogical’fieldwork in the 1860s and published in the early 1870s. Significanthere was their key informant, Si Mula, a Sayyid ‘alim or religious scholarand Hanoteau’s khoja or interpreter-cum-secretary at Fort-Napoléon inKabyle. Si Mula became, in Parkes’s words, ‘at least an equal co-author’with the two Frenchmen, though they do not openly credit him as such.

Parkes describes the ‘canonical ethnography’ of Hanoteau andLetourneux as severely factual or documentary, largely eschewinghistorical contextualisation. Nonetheless Hanoteau, the main author,was well aware of the extent to which French conquest and militaryrule had already disrupted Kabyle society, an account of whosetraditional social organisation he was therefore keen to draw up. Ineffect, therefore, while historical or reconstructive in intent, thetreatment is paradoxically synchronic in presentation, describing anindependent Kabyle society on the eve of its conquest.

Although the two authors’ juridical approach would be displaced byMaussian transactional ethnographies of the inter-war period, not even


Bourdieu was able to escape their influence entirely, despite his surfacecriticism of their ‘legalistic’ prejudice. Indeed, as Parkes finally notes,there is reason to believe that some, at least, of Bourdieu’s fundamentalideas as perhaps the most famous ethnographer of the Kabyles wereoriginally forged in reactive opposition to the rule-based ‘canonicalethnography’ of Hanoteau and Letourneux – a ‘theory of practice’ thatboth complements and contrastively highlights the significance of thejuridical fieldwork they pioneered.

Paul Henley’s chapter deals with a figure who is probably the mostfamous ethnographic film-maker of them all, Jean Rouch (1917–2004).Seen already as somewhat passé in France by the 1980s, it was preciselyat this time that ‘Anglo-Saxon’ anthropology began to discover Rouchas a precursor of post-modernism. As Henley makes clear, however, thisis not entirely what it seems, and in many respects Rouch actuallybelongs to a specifically French tradition of ethnography dating back tothe surrealists as much as to Mauss, but also reflecting the strong thoughnot overwhelming influence of his doctoral supervisor, Marcel Griaule.

Henley discusses the ways in which the experience of working withGriaule did and did not influence Rouch. While Rouch refrained fromdeliberately antagonising informants in the way that Griaule frequentlydid, and stressed their co-authorship with him in what he saw as agenuinely collaborative effort (the source of his later being claimed asa prophet of ‘dialogical anthropology’), he also relied on provocation inthe ethnographic encounter – but only by the camera itself. For Rouch,the fact of it not being possible to hide the camera’s presence wascreative, not disadvantageous, since what it provoked in the informantwas a reaction different from, but at the same time deeper than, normalbehaviour, uncovering the truth underlying the superficiality of theeveryday world.

Henley also shows, though, that Rouch took his ideas about theimpact of the camera a great deal further than the simple claim that itis provocative to the subjects. Filming also allows the film-maker toimmerse him- or herself in the culture. If film can provoke trance in thenatives, as Rouch claimed it actually did in at least one case, the film-maker him- or herself can also be provoked by the act of filming to entera trance. Hence Rouch’s famous ciné-trance, conceived as a metaphor forthe film-maker’s own cultural creativity. At the very least, just as, forthe Songhay, spirit possession changes the medium’s experience of theworld, so for Rouch the film-maker is changed by filming it. In otherwords, in Rouch’s conception, these processes of collaboration betweenauthor and subjects involved a performative element that goes beyondthe merely verbal exchange implicit in the conventional Anglo-Saxonconception of ‘dialogical’ anthropology.

Introduction 13

Margaret Buckner’s chapter on Eric de Dampierre (1928–1997) is oneof a minority in this collection in which the pupil discusses the work andcareer of the master (also Toffin on Bernot). Like Bernot, Bastide andDumont, Dampierre began his career with a study of a French ruralcommunity, in his case as part of a multi-disciplinary social-scienceresearch team. His life-long fieldwork, from 1954 to the late 1980s, wasamong the Nzakara, in what is now the Central African Republic. As aFrench aristocrat, he was clearly comfortable living in a highly stratifiedand class-conscious African society, while recognising that they seemedless able to cope with the consequences of colonialism and modernity thantheir close neighbours the Zande, otherwise a very similar society, madefamous through Evans-Pritchard’s earlier work among them. Dampierre’swork therefore provides us with a little-known but very valuable Frenchcounterpart to Evans-Pritchard’s famous monograph (1937).

Dampierre identified what he called ‘thinking in the singular’ as akey aspect of Nzakara thought, this being perhaps the most original ofhis findings, which he saw as pervading all domains of Nzakara life,from politics to music. It stresses the unique, the incommensurability ofany two beings, so that, for example, one cannot count people, norclassify them, for fear of treating them all the same. Although, in hissophisticated attempt to define this mode of thought, he may haveturned to Greek philosophy, it was still his experiences among theNzakara, his observations of their practices, discourse and materialculture, that had launched his research in the first place.

The Lévi-Straussian flavour of the title of his last work, Une esthétiqueperdue (Dampierre 1995), links Dampierre with that generation ofanthropologists who had the feeling that they were living at the end ofan era, the traces of which they wanted to preserve as lucidly andfaithfully as possible. Not the least of Dampierre’s legacies, however, ishis founding and support of the Department of Ethnology andPrehistory in the University of Paris-X at Nanterre, to the west of thecity, perhaps the major university department dedicated toanthropology and to training anthropologists in the whole of France,where one of the present editors received her own training and withwhich she continues to be associated.

Laura Rival’s chapter on Paul Rivet (1876–1958) discusses a nowneglected figure who was one of the key figures institutionally in theanthropology of France in the inter-war period. His work with theInstitut d’Ethnologie and later at the Trocadéro (including the Musée del’Homme) gave him a pivotal role in the organisation of anthropologyin France between the two world wars, not far behind those of Maussand Lévy-Bruhl, with both of whom he cooperated closely and sharedmany aims for the promotion of French anthropology.


A pupil of literature and philosophy at school who then trained as adoctor, Rivet spent five years in the early twentieth century conductingpolymathic fieldwork and collecting as part of a French geodesicexpedition to Ecuador, with which, together with Colombia, he was tobe associated for the rest of his life. Rivet can hardly be described as anexemplary fieldworker from the point of view of post-Malinowskiananthropology. This was basically because he had no direct contact withthe native population, but used what Rival calls ‘indirect’ methods ofenquiry, interviewing intermediaries who were in the happy but self-deceptive position of ‘knowing’ the natives without having to questionthem about anything. In many respects, Rivet seems to have beenmainly an observer, ‘collecting, classifying and comparing’, in Rival’swords. He rarely if ever asked questions about native meanings or ideas– he had little interest in religion, for instance, except to see in it anexample of the ignorance that was holding the natives back. Here wehave the Third-Republic scientific mind finding fault with Amerindiansociety – especially for its ignorance born of religious mysticism andsuperstition – while at the same time rejecting race as an explanationfor difference in favour of a humanism that unites us all as equal andequivalent. In view of what has been said about the links betweenmodern French identity and a generalised humanity (e.g. Dumont1986), it is perhaps not surprising that we also find a focus in Rivet’swork on the generic human condition rather than the specifics ofdifferent cultures.

Although Alfred Métraux (1902–1963) was born in Switzerland,brought up largely in Argentina and later became an American citizen,he belongs to the French tradition of anthropology primarily by virtueof the institutional side of his training: taught by Mauss and Rivet inthe 1920s, his theses on the TupÍ-GuaranÍ of Brazil were submitted inParis. However, as Peter Rivière notes in his chapter on him, he washardly influenced intellectually by Mauss, nor even by Rivet, whosupported him in his career early on. Instead Métraux fell under thespell of the Swedish ethnologist Nils Erland Nordenskiöld, adoptingespecially the latter’s tracing of trait distributions across one or moreethnographic regions and his theoretically uncontextualised treatmentof ethnographic data. Rivière argues that Métraux saw himselfprimarily as a collector of facts, retaining a strict and almostnineteenth-century demarcation between this activity and the widercomparison or theorising done by others in the library or study. As aresult, there is little or no contextualisation or analysis in his ownwritings, which are rather of the nature of compilations.

This apparent hostility to theory indicates a mind that is notprepared to speculate over what cannot be known concretely. Yet

Introduction 15

Métraux’s attitude to fieldwork and the collection of data through itwas not entirely straightforward. On the one hand, he doubted whetherthe collection of ethnographic facts could ever be truly scientific, mainlybecause he felt that the civilised mind cannot readily grasp them. At thesame time, not only did he frequently complain about local conditionsin the field, he felt that ethnographers – including himself implicitly –were essentially misfits in their own societies. He was clearly somewhatprone to romanticising the people he studied, in a manner which seemsto have been fashionable in French anthropology for a time (Rivièrementions Lévi-Strauss’s Tristes tropiques, and some of Pierre Clastres’work is in a similar vein; cf. Colchester 1982). Like Lévi-Strauss,Métraux praised what he saw as the neolithic in the native SouthAmerican, which he also considered as in some sense the end point ofhuman happiness in human evolution, not least because he saw it asbeing on the verge of disappearing. Perhaps it was this feeling ofwitnessing the disappearance of a way of life he much admired, asmuch as the sense of his having received little recognition for his life’swork, that led him apparently to take his own life in 1963.

Like some other figures dealt with in this book (Dumont, Bernot,Dampierre), Roger Bastide (1898–1974) took part in an early study inFrance itself, this time on Armenian immigrants in the town of Valence.However, being already interested in mysticism, and in 1938 findinghimself a professor at the University of São Paulo in Brazil in successionto Lévi-Strauss, he embarked on a long-term though intermittent studyof candomblé in the northeast of the country. This brought him intocontact with Pierre Verger, who became a life-long friend andcollaborator. Bastide and Verger shared a belief in the importance ofexperience in fieldwork, including the idea that one could notunderstand something like possession without going through it oneself.In addition, they both rejected the standard view of northeast Brazilianculture being an original form born of acculturation and religioussyncretism: Verger’s life-long concern in particular was to prove to Afro-Brazilians the Africanness of their cultic practices. Although it wasmysticism that was the focus of Bastide’s interest, it was ironically thesceptic Verger who went furthest into the candomblé as a religiousexperience: Bastide stopped halfway out of fear for his own sanity if hewere to allow his grip on reality to be loosened by continuing.Nonetheless Bastide felt able to proclaim ‘Africanus sum’, and, as withGriaule’s defence of African religion as represented by the Dogon, hedeveloped a view of Afro-Brazilian religion as being comparable in itssophistication to any of the religions of ‘civilisation’. Moreover, there issomething similar here to the Rouchian ciné-trance described by Henley(this volume): in both cases, the trance state affects the ethnographer as


much as the people he is studying. Bastide’s subsequent return to Parisin 1954 to work with Georges Gurvitch exposed him to yet moreinfluences, though academic this time, including Marxism, a renewedview of Mauss, and Gurvitch’s own ‘depth sociology’. Between them,they became a sort of ‘opposition’ to Lévi-Strauss’s structuralism, whichwas just then taking off in France.

Despite the support given to him at key points in his career by Lévi-Strauss, Lucien Bernot (1919–1993) was no structuralist. Indeed, heonce remarked that, while ethnographic monographs could always betreated structurally, structuralism was quite incapable on its own ofreconstructing the original ethnography. He was also of the view that,in always being available to later generations of anthropologists, theethnographic monograph invariably outlasts theory, which is subjectto changes in intellectual fashion. His main influence was therefore theanti-structuralism and ethnographic essentialism of Leroi-Gourhanand some of his own more exact contemporaries among FrenchSoutheast Asianists, in particular André-Georges Haudricourt, but alsoGeorges Condominas.

Toffin describes Bernot as an acute fieldworker when it came tometiculous observation of what people do. Bernot advocated a focus onsmall-scale communities of 200–300 people, since he felt that in thesecases the ethnographer could come to know everyone within them. Hismain focus was on technology and its relation to society, and later onethnobotany (reflecting Haudricourt’s influence). This factualconcentration in his work recalls Rivet and is similarly diffusionist in itsmethods, if not explicit theoretical orientation. This aspect is perhapsreflected mainly in the ethnolinguistic atlases Bernot created, which tracedthe distribution of key words across vast swathes of Southeast Asia, butalso in his use of written sources for purposes of historical reconstructionand his frequent citation of diffusionist geographers. Fundamentally,though, he was what Toffin describes as a ‘ruralist’ by both upbringing andprofessional interest, that is, a specialist in rural, agricultural communities,which, the world over, had similarities that link them and distinguish themfrom urban society: thus the people of Nouville (northeast France) havemore in common with Burmese peasants than with Parisians – one respectin which he disagreed with his friend Haudricourt’s stress on the differencesamong rural communities in the world at large.

André-Georges Haudricourt (1911–1996) was nonetheless another‘ruralist’, a country-born child who, because of ill health, was educatedfirst by his mother and subsequently by himself. Based on observationalhabits learned during his upbringing, combined with the experience ofearly fieldwork in Vietnam, Haudricourt developed not only an extremefocus on the facts, but what Bensa calls a ‘hyperrealist’ view of facts as

Introduction 17

being restricted to what can be known through the senses alone. Heaccordingly accepted no psychological, sociological or intellectualinterference with our own direct exposure to the world and experienceof it, and he rejected notions such as the autonomy of representationsand ideas (Durkheim), the social being projected on to nature(Durkheim and Mauss 1963 [1903]) or the symbolic transformationof nature by culture (Lévi-Strauss). The structuralist’s dualism ofnature and culture is replaced by a close symbiosis between them inwhich they often imitate each other, though the latter is always rootedin the former, not vice versa.

Haudricourt’s extended comparison between the Middle East andFar East in part relies on a distinction between the culturalpredominance given respectively to plant and animal breeding, butnonetheless it is the different plant and animal ecologies of both areasthat are ultimately the bases of the distinction. Thus in the Middle East,animal herds and wheat both originated outside human environmentsand had to be subdued and controlled by humans, whereas in the FarEast (actually in this example Melanesia) there was a situation in whichhumans, plants and animals started out living symbiotically in thesame environment. From this distinction, Haudricourt derives differentideas of religion, social authority and hierarchy: thus in the Middle Eastthe gods are remote, but in Melanesia they are all around one. Bensauses the term ‘functional historicism’ to characterise Haudricourt’sfocus on origins and history, by which is meant both the biologicalhistory of particular species and the histories of distinct humanpopulations in distinct environments. And, as with some otherethnographers discussed in this introduction, such as Bernot and Rivet,the focus on the facts stresses the particular over the general, theethnographically specific over the universal.

In contrast to many of the other anthropologists featured in thisvolume, Louis Dumont (1911–1998), an exact contemporary ofHaudricourt, discussed here by Robert Parkin, is known for histheoretical contributions and more literature-based writings at least asmuch as for his fieldwork. Nonetheless his fieldwork in south Indiaformed a significant part of his own intellectual development and led toone of the classic ethnographies of the region. Dumont’s subsequentsojourn in Oxford under Evans-Pritchard influenced his anthropologyquite profoundly, and in many ways he is the most ‘Anglo-Saxon’ of thefigures treated in this collection. Yet the earlier influence of Maussremained strong, while the Tamils, whom he regarded as ‘bornsociologists’, influenced him in developing his view that a form ofstructuralism was the key to understanding Indian society and culture.His use of pure/impure as a key ‘hierarchical’ opposition in the values


of the caste system ultimately replaced the simple binary oppositionsof Lévi-Strauss, being focused on values as more important thansymbolism, and recognising the significance of social action while stillsubordinating it to ideology and structure. These ideas were enshrinedespecially in his most famous work, Homo hierarchicus, on the Indiancaste system (Dumont 1966, 1980).

After India and Oxford, Dumont returned to Paris to pursuecomparisons between India and the West, which also involved contrastsbetween hierarchy and egalitarianism, holism and individualism, andindeed two sorts of individual, the individual-outside-the-world and theindividual-within-the-world. This move was also a shift from fieldworkto writings, from observation to ideas, and in its approach it reflectedthe influence of Mauss in the latter’s writings on such themes as thegift and the person, where world history was the framework withinwhich both the topic and the related arguments were set. Finally, in hislast major work on German ‘ideology’ (Dumont 1994 [1991]) or, as wemight say today, ‘identity’, he demonstrated that even in the Westindividualism was not all of a type: in particular, the German stress onpersonal self-development being subordinated to a holistic state isopposed to the ‘individual-against-the-state’ model of Anglo-Saxon andFrench libertarian philosophies.

Jeremy MacClancy’s chapter on Maurice Leenhardt (1878–1954)is a little different from some of the others in this volume, since itdiscusses this quasi-iconic, early, pre-Malinowskian fieldworkerthrough the eyes of his later commentators. Born at Montauban in1878 into a French Protestant family – the latter circumstance heshares with Roger Bastide – Leenhardt wrote an early thesis on the‘Ethiopian’ church movement in southern Africa. However, he spentmost of his career until well into the 1920s as a Protestant missionaryin New Caledonia.

Leenhardt’s interests were many, but they included especiallyMelanesian languages in and around New Caledonia and – what he ismost famous for – his very striking and imaginative analyses ofpersonhood and myth. As MacClancy shows, he has been claimedsuccessively as a post-structuralist in the manner of Clifford andMarcus, a Jungian phenomenologist, a Heideggerian existentialist anda Strathernian advocate of the decentred nature of personhood inMelanesia – the first and last, at least, very much ‘before his time’. Morelikely though, as MacClancy himself suggests, he was basically just aman of his own time. One can argue that his patent sympathy for theindigenes was romantically inclined towards his primitivist vision oftheir way of life, rather than concerned with their progress as such (cf.Métraux or Clastres), while his intellectual perspective was fundamentally

Introduction 19

evolutionist, despite his awareness of Maussian holism. If he seemsanti-structuralist, therefore, it is because of his already outmodedintellectual position.14

Unity in diversity?

What common features emerge from a comparative reading of thechapters in this collection? First, as already noted, it is striking howmany of these figures can be considered anti-theoretical fact-gatherersand compilers, at least in their own view of their activities. CertainlyHanoteau, Rouch, Rivet, Métraux, Bernot and Haudricourt, in theirvery different ways, exemplify this tendency. Yet theory is notnecessarily so very far away, even in these cases. For example, giventheir interests in the distribution of words in particular, Rivet andBernot can be seen as being informed by diffusionist methods andassumptions in their handling of the facts they collected. Moreover, thevery emphasis on ethnographic essentialism can be regarded as atheoretical or at least philosophical position in itself, as it clearly wasfor Haudricourt. As we have remarked already, van Gennep, with hisproject of converting folklore into anthropology; Dampierre, whosenon-structuralist approach was informed at least in part by hisbackground in sociology; and Dumont, with his revisioniststructuralism, all had their own particular theoretical focuses.

It is also remarkable how many of these fact-gatherers seem to havehad rather limited abilities as fieldworkers: thus Hanoteau, Rouch,Rivet, Métraux, Bastide and Bernot had to rely largely on interpreters,Rivet and Bastide on local intellectuals and other sorts of intermediarytoo, while Métraux seemed to spend a lot of time complaining aboutactual fieldwork conditions. Nonetheless most of the figures treatedhere spent long periods of their lives in the field, though Dumont andBastide perhaps least of all. Moreover, arising out of this dedication tothe collection of facts are also a number of real commitments on thepart of many of these figures to the peoples they encountered, to thelatter’s contemporary circumstances and conditions, and to theirrelations with them. There is a whole range of attitudes here, from therelatively passive and neutral to genuine if selective political activism.At one end of the scale is Dumont, whose commitment wasfundamentally restricted to achieving ethnographic understandingwith the aid of particular theoretical frameworks within an overallethos of intellectual neutrality. For example, in defending this principlein relation to phenomena that may shock western sensibilities in fieldsituations elsewhere in the world, Dumont frequently argued that to


seek to understand, say, the caste system in India or female circumcisionin parts of Africa did not necessarily mean that one approved of them(e.g. Dumont 1979). Although Bernot and Rouch clearly developedclose and mutually supportive personal relationships with theirprincipal subjects and greatly admired their cultural traditions, neitherevinced any deep political commitment to their respective peoples.Métraux hardly goes beyond a nostalgia for the neolithic, whichAmerindians represented for him, though as Peter Rivière points out(personal communication), later in life he became somewhat moresympathetic to the peoples he encountered through his activities inassessing war damage in Germany and his involvement with UNESCO.Conversely, while Dampierre merely seems to record the changesassociated with colonialism among the Nzakara, albeit with a tinge ofnostalgia, others – like Hanoteau among the Kabyle and Leenhardt inNew Caledonia – tried to protect the native population from the worstconsequences of colonialism.

At the other end of the scale is Rivet, the only figure here actually tobecome a politician – not in South America against colonialism, but inFrance in the 1930s, against fascism. Otherwise his self-appointed rolewas to affirm the positive in the practice and status of métissage and torecord and discuss the conditions of the Amerindians he encounteredfrom the ‘scientific’ perspective of a social scientist of the ThirdRepublic, even though his direct personal contact with them, becauseof his habitual use of intermediaries, was minimal. Finally, both Rouchand Bastide hailed the experience of the (ciné-)trance as a fulfilment ofthe ethnographic experience that was almost mystical for them; yet thefulfilment they sought was strictly their own, rather than intended to beof any use to those whose cults they were taking part in and recording.Exposure to the field, and even one’s personal bodily experience of it,was also important to van Gennep, though as a tool of ethnographicenquiry rather than a means of personal discovery.

Nonetheless it was perhaps this more personal and/or politicalcommitments that replaced theory as a goal of fieldwork in the mindsof some of these figures. At all events, we argue that, while some Frenchethnographers are scarcely any different from their colleagueselsewhere when it comes to relating facts to theory, very many othershave dedicated themselves to the former to the exclusion, in whole or inpart, of the latter. There can be no question, of course, of thetremendous contribution of French intellectuals in many disciplines tothe enrichment of anthropological theory and model-buildingworldwide. Yet ethnographic practice informs anthropology in Francetoo, often overshadowed by the theorists or neglected entirely, especiallyabroad, but involving a variety of genuine commitments to data

Introduction 21

collection, exotic cultures, ethnographic subjects as fellow humanbeings, one’s relations with them or just the personal experience offieldwork. Amongst other things, this makes the study of fieldwork aperfectly valid and highly productive way of approaching the history ofFrench anthropology generally. That is because France is distinct notjust for its theories and model-building but because, in explicitopposition to them, many a practical fieldworker has theorised awaytheory itself so that the facts of the ethnography can shine forth in alltheir splendour.

Notes

1. See also Cusset (2008 [2003]) on the invention of ‘French theory’ in America.2. His epiphany has already been hinted at in print (Parkin 2005), where an attempt at

a potted history of the whole of French anthropology can also be found. Theseoriginated in lectures given at the official opening of the Max Planck Institute ofSocial Anthropology in Halle, Germany, in June 2002.

3. For a more extended account of these events, see Parkin (2005), in which keyreferences can also be found. More recently, see also Sibeud (2008).

4. In 1938 the Trocadéro was transformed into the Musée de l’Homme by Paul Rivetand Georges-Henri Rivière. Its collections have since been transferred to the newMusée du Quai Branly (see l’Estoile 2003, 2007).

5. Jacques Dournes, sometimes known under his Sre name of Dam Bo, made a similarshift somewhat later (the Sre are located in the Vietnamese highlands).

6. We stress the long-term: Mauss did undertake one brief field trip to witness dances inMorocco.

7. Allen describes this as ‘a longstanding preoccupation that originated in part with thequestion of how to organise the Année sociologique’ (2007: 2), the house journal of theDurkheim group, in terms of the rubrics into which it should be divided.

8. It is hard to be sure whether, in talking about the facts, Mauss necessarily has in mindhis uncle’s idea of the ‘social fact’ as defined quite narrowly (though also discussed atsome length) in Chapter 1 of the Rules of sociological method (Durkheim 1982 [1895]).Nor is it clear to what extent Mauss was concerned with the construction of ‘facts’ inthe epistemological sense. Mauss’s usage often seems to be purely normative in thesepassages.

9. The more questionable aspects of Griaule’s methods were the main reason for Leirisbreaking with him; see Leiris (1934). A good account of Griaule in the field is Clifford(1983).

10. Now l’Institut pour la Recherche et le Développement (IRD).11. In order of appearance, these include Laplantine (1996), Beaud and Weber (1997),

Copans (1998) and Berger (2004).12. In Revue d’Histoire des Sciences humaines, No. 10, 2004. For an innovative analysis,

from a similar perspective, of the genesis of different ‘national anthropologies’ inEurope, the Americas and South Africa, and the linkage between them, see L’Estoileet al. (2005).

13. The second book in this series, which was founded by Jean Malaurie in 1955, wasLévi-Strauss’s Tristes tropiques (1955) – not a conventional ethnography, any morethan its author was an ethnographic essentialist, let alone a willing ethnographer;


more typical, perhaps, of the genre is Georges Condominas’s L’exotique est quotidien:Sar Luk, Vietnam (1965). On this important series, see Aurégan (2001).

14. Laura Rival adds the information that ‘post-structuralist Amazonianist anthropologymakes much of Leenhardt, especially the oft-quoted anecdote about the missionariesbringing to the Canaques not the soul but the body’ (personal communication).

References

Allen, N.J. 2007. Introduction, in M. Mauss, Manual of ethnography, New York and Oxford:Berghahn.

Aurégan, P. 2001. Des récits et des hommes: Terre humaine – un autre regard sur les sciencesde l’homme, Paris: Nathan.

Beaud, S. and F. Weber. 1997. Guide de l’enquête de terrain, Paris: La Découverte.Belmont, N. 1979. Arnold van Gennep: the creator of French ethnography, Chicago and

London: The University of Chicago Press.Berger, L. 2004. Les nouvelles ethnologies, Paris: Nathan.Blanckaert, C. (ed.). 1996. Le terrain des sciences humaines: instructions et enquêtes (XVIII–

XXème siècle), Paris: L’Harmattan.Céfaï, D. (ed.). 2003. L’enquête de terrain, Paris: La Découverte.Clifford, J. 1983. Power and dialogue in ethnography: Marcel Griaule’s initiation, in G.W.

Stocking (ed.), Observers observed: essays on ethnographic fieldwork, Madison: TheUniversity of Wisconsin Press.

Colchester, M. 1982. Les Yanomami, sont-ils libres? Les utopies amazoniennes, unecritique: a look at French anarchist anthropology, Journal of the Anthropological Societyof Oxford, 13(2): 147–64.

Condominas, G. 1965. L’exotque est quotidien: Sar Luk, Vietnam, Paris: Plon.Copans, J. 1998. L’enquête ethnologique de terrain, Paris: Nathan.Copans, J. and J. Jamin. (eds). 1994 [1978]. Aux origines de l‘anthropologie française: les

mémoires de la Société des Observateurs de l’Homme en l’An VIII, Paris: Jean-Michel Place.Cusset, F. 2008 [2003]. French theory: How Foucault, Derrida, Deleuze and co. Transformed

the intellectual life of the United States [French theory: Foucault, Derrida, Deleuze et Cie etles mutations de la vie intellectuelle aux Etats-Unis] (tr. J. Fort), Minneapolis: Universityof Minnesota Press.

Dampierre, E. de. 1995. Une esthétique perdue, Paris: Presses de l’ENS.Delacampagne, C. 1981. Louis Dumont and the Indian mirror, Royal Anthropological

Institute News, 43: 4–7.Dumont, L. 1966. Homo hierarchicus: essai sur le système des castes, Paris: Gallimard

(English trans. 1972, London: Paladin).——— 1979. The anthropological community and ideology, Social Science Information,

18: 785–817.——— 1980. Homo hierarchichus: the caste system and its implications (2nd ed.), Chicago:

The University of Chicago Press.——— 1986. Essays on individualism: modern ideology in anthropological perspective,

Chicago and London: The University of Chicago Press.——— 1994 [1991]. German ideology: from France to Germany and back, Chicago and

London: The University of Chicago Press.Durkheim, É. 1982 [1895]. The rules of sociological method, London: Macmillan. Durkheim, É. and M. Mauss. 1963 [1903]. Primitive classification, London: Cohen & West.Evans-Pritchard, E.E. 1937. Witchcraft, oracles and magic among the Azande, Oxford:

Clarendon Press.

Introduction 23

Gaillard, G. 2004. The Routledge dictionary of anthropologists, London and New York:Routledge.

Gingrich, A. and R.G. Fox (eds). 2002. Anthropology, by comparison, London and NewYork: Routledge.

Laplantine, F. 1996. La description ethnographique, Paris: Nathan.Leiris, M. 1934. L’Afrique fantôme, Paris: Gallimard.L’Estoile, B. de. 2003. From the Colonial Exhibition to the Museum of Man: an alternative

genealogy of French anthropology, Social Anthropology, 11(3): 341–61.——— 2007. Le goût des autres: de l’Exposition colonial aux Arts premiers, Paris:

Flammarion.L’Estoile, B., F. Neiburg and L. Sigaud (eds). 2005. Empires, nations and natives:

anthropology and state-making, Durham: Duke University Press. Lévi-Strauss, C. 1955. Tristes tropiques, Paris: Plon.MacClancy, J. and R. Parkin. 1997. Revitalization or continuity in European ritual? The

case of San Bessu, Journal of the Royal Anthropological Institute, 3(1): 61–78.Mauss, M. 1998. An intellectual self-portrait, in W. James and N.J. Allen (eds), Marcel

Mauss: a centenary tribute, New York and Oxford: Berghahn.——— 2007 [1947]. Manual of Ethnography (tr. D. Lussier), New York and Oxford:

Berghahn.Parkin, R. 1996. The dark side of humanity: the work of Robert Hertz and its legacy,

Amsterdam: Harwood Academic Publishers.——— 2005. The French-speaking countries, in F. Barth et al., One discipline, four ways:

British, German, French, and American anthropology, Chicago: The University ofChicago Press.

Rogers, S. 2001. The anthropology of France, Annual Review of Anthropology, 30: 481–504.

Sibeud, E. 2008. The metamorphosis of ethnology in France, 1839–1930, in H. Kuklick(ed.), A new history of anthropology, Oxford: Blackwell.

Todorov, T. (ed.). 2007. Le siècle de Germaine Tillion, Paris: Seuil.Todorov, T. and C. Bromberger (eds). 2002. Germaine Tillion: une ethnologue dans le siècle,

Paris: Actes Sud.van Gennep, A. 1909. Les rites de passage, Paris: Nourry.——— 1920. L’état actuel du problème totémique, Paris: Ernest Letroux.——— 1938–58. Manuel de folklore français contemporain, Paris: Picard, 9 vols.


Chapter 1

‘KEEPING YOUR EYES OPEN’:ARNOLD VAN GENNEP AND THE AUTONOMY

OF THE FOLKLORISTIC

Giordana Charuty

Introduction

The intention of the exhibition and its catalogue, Hier pour demain, held atthe Grand-Palais in Paris between June and September 1980, was to makethe general public aware of the French ethnographic heritage on theprecise occasion of l’Année du Patrimoine or the Year of the Patrimony(Cuisenier 1980). In the exhibition, a chronology of ethnographicprecursors identified a succession of moments, from the mid-eighteenthcentury of L’Encyclopédie up until 1937, marked by two ‘monuments’. Onewas the Musée des Arts et Traditions Populaires. The other was thebeginning of the publication of the Manuel de folklore français contemporainby Arnold van Gennep (1873–1957), an author in whom one recognisesa double status: the last of the folklorists in the manner of the nineteenthcentury, and the first of contemporary ethnographers.1

However, certain other readings, recent and not so recent, haverestored a greater degree of complexity with respect to the academictraining, initial theoretical interests and intellectual sites that permittedvan Gennep to work without respite for the recognition of a disciplinaryfield in the first half of the twentieth century (see Belmont 1974, Chiva1987, Fabre 1992, Velay-Valentin 1999). After some schooling awayfrom Paris, he received training in linguistics and the history of religionsat the Ecole des Langues Orientales and the Ecole Pratique des HautesEtudes, which he attended at the same time as Marcel Mauss. With LéonMarillier in particular, he discovered there the ethnographies of remote

Figure 1.1. Arnold van Gennep, aged 80, lighting a bonfire on the summer solstice (21June 1953), also the saint’s day of St John the Baptist. Taken by Pierre Soulier. Muséedes civilisations de l’Europe et de la Méditerranée.

worlds and learned the art of rigorous criticism of ethnographicdocuments. However, while he shared the same intellectual interestsand the same knowledge as British anthropologists, his relations withthose who collaborated in L’Année sociologique were more conflictual.In the journal, his Mythes et legendes d’Australie of 1906, which hepresented as a ‘study in ethnography and sociology’, was subjected toa critique, signed by Mauss, that was at least as severe as that of the solework of van Gennep’s that British anthropology was to retain, namelyLes rites de passage of 1909 (Mauss 1908, 1909). In response, theMercure de France and the Revue de l’histoire des religions provided himwith a platform to object to the use that Emile Durkheim had made ofAustralian ethnographic data in the latter’s Les formes élémentaires dela vie religieuse (1912):

I fear that M. Durkheim, despite his apparent care with ethnographic facts,only possesses a metaphysical sense, or, even more so, a scholastic sense; heaccords a genuine reality to both concepts and words. Not having any sense

26 Giordana Charuty

of life, that is, any biological or ethnographic sense, he turns livingphenomena and living beings into plants that have been dried scientifically,as in a herbarium. (2001: 94 )2

He repeated this criticism in his Etat actuel du problème totémique of1920, to which Lévi-Strauss paid homage in his La pensée sauvage,saluting simultaneously his ‘innovatory audacity’ and his limitationsfor having remained, in his turn, the prisoner of ‘a traditional carvingup’ of social institutions (1962: 213–15).

Nonetheless, in the same period Mauss and van Gennep conductedsimilar diagnoses of the stagnation, in France, of the ethnography of theFrench domain or of remote worlds: fieldwork, museums, archives, teaching– none were worthy of the name. However, while the former entered the FifthSection of the Ecole Pratique des Hautes Etudes at a very young age, the latternever obtained any university post apart from a few brief years, between1912 and 1915, when he occupied the chair of ethnography at Neuchâtel.Here he reorganised the museum and arranged the first ethnographiccongress in 1914, before being expelled for his lack of political discretion bycasting doubt publicly on Switzerland’s neutrality during the war.

What does one live off when one has no university position? In1896, aged 23, van Gennep left France for four years to teach French inPoland. On his return to Paris in 1901, he entered the Ministry ofAgriculture as head of translations – he mastered a dozen or solanguages – and he lived materially from his contributions to journalsand his translations. This did not prevent him from conducting a five-month enquiry in Algeria between 1911 and 1914 to make aninventory of the techniques and styles of Kabyle pottery, nor frompublishing several theoretical essays, including the future, muchcelebrated Rites de passage (1909). Between 1904 and 1914, and againon the eve of the Second World War, he ran several journals ‘ofethnography and sociology’, ‘living folklore’ and ‘popular traditions’,as well as providing contributions to a large number of specialistjournals on his own account, which still remain to be studied.

Prefacing his 1967 English translation of Les demi-savants (vanGennep 1911, Needham 1967), Rodney Needham was the first tosuggest that the publication of this brief pamphlet in 1911, a parody ofthe academic world, may not have been unconnected with this lack ofinstitutional recognition. The hypothesis was taken up again in 1992 byAlan Dundes (1992: 4), who identified at least one amateur folkloristamong these ‘semi-scholars’. Since then, other explanations have beensuggested, which may be summarised quite briefly.

The struggle that arose in France after 1925 for ethnology to be fullyrecognised as a university discipline produced a division into two

‘Keeping your eyes open’ 27

28 Giordana Charuty

camps. On the one hand, there was a conservative anthropology, theheritage of Le Play and de Broca, promoted by Louis Marin throughseveral learned societies and two journals, L’Anthropologie andL’Ethnographie. Opposed to this was the Institut d’Ethnologie, withLucien Lévy-Bruhl, Mauss, Marcel Cohen and Paul Rivet, whichprovided training for future researchers, travellers and colonial officers,while seeking to take over the university. This enterprise succeededthrough the creation of degree certificates at the Sorbonne, the namingof Paul Rivet as Professor of Anthropology at the Muséum Nationald’Histoire Naturelle and Director of the Musée d’Ethnographie duTrocadéro, and Mauss’s nomination to the Collège de France in 1931.

A stranger to the Durkheim circle, van Gennep also remained on themargins of an analogous circle of ethnologists of France. Although hehad played an essential role in the Congress of Popular Art organised inPrague in 1928 by the Société des Nations and the Institut Internationalde Coopération Intellectuelle, the following year he was kept away fromthe creation of the Société de Folklore Français and its journal of thesame name, and in particular from the management of the Musée deTrocadéro, whose restoration Paul Rivet entrusted to Georges HenriRivière. This activity prefigured the creation of the Musée des Arts etTraditions Populaires and from 1938 was continued by the teaching ofthe history of popular arts and traditions at the Ecole du Louvre.

Building on the evidence already provided by Nicole Belmont (1974)and Isaac Chiva (1987), Daniel Fabre argues (1992) that van Gennep’sremoval should lead one to reject the convergence, accepted up to thatpoint, between the Musée des Arts et Traditions Populaires and vanGennep’s Manuel de folklore français contemporain, which instead heproposes to read as an ‘inverted monument’ with a highly polemicalrelationship with the principles that governed the ethnography andmuseography of the Musée. This reversal of perspective is supported bythe discovery of another intellectual site where van Gennep publishedwith great regularity between 1905 and 1949, namely his Chroniquesin the journal the Mercure de France (see below). These have just becomethe object of a very useful if partial republication, which permits us topursue alternative readings (van Gennep 2001).

The work he produced in Mercure de France reveals the demands,difficulties and ambiguities of what would later be called ‘ethnology athome’. It was the author Rémy de Gourmont who in 1904 introduced vanGennep, then thirty years old, to the director of this cultural review, whoseaim was to oppose all academic theories. The publication of van Gennep’sthesis, Tabou et totémisme à Madagascar (1904), had attracted the editor’sattention, and his bi-monthly Chroniques were to become one of the sites,on the margins of the university, of the definition and recognition of a


discipline – which he continued, after many hesitations, to call ‘folklore’ –in its relations with the ethnography of non-western societies, the historyof religions, prehistory and physical anthropology. These were not justreviews of recent publications but also, as Jean Marie Privat has stressed(2001), a weapon in the struggle to win autonomy with respect to other,better established scientific disciplines in order to train practitioners anddistinguish good and bad academic practices in this discipline.

There is also, like a watermark, a kind of intellectual biographywhich can be read through the memories of experiences, encounters,readings, research projects and hesitations over the methods andinstruments to be preferred. This reveals, in the course of time,difficulties in reconciling a multiplicity of positions, which are notnecessarily as opposed to one another as has been said, to the limits andimpasses of ethnography in the Musée des Arts et Traditions Populaires.

The ethnographic position

Van Gennep never ceased to repeat: ‘It therefore remains the case thatethnography, folklore, and popular traditions are no longer ethnic orpolitical in kind, but designate the study of the mores and customs of allpeoples, ancient and modern, and of all forms of civilisation’ (2001[1947]: 220). In fact, in the early years, the Chroniques in Mercure deFrance put forward this universalist aim by making ethnographies of theAfrican, Australian, Amerindian and European worlds converge, since,wherever the enquiry takes place, providing the conceptual categories ofa rigorous description forms part of a general theory of culture. Butalthough the reader is invited to approach the popular life of a Frenchprovince and the life of the Nandi or the Masai with the same curiosity,the conversion of gaze required of the western observer in order toreconstitute the diversity of rural societies in Europe in a comparativemanner can in no way be taken for granted: the requirement ofdescriptive objectification presupposes systematic submission to a uniqueform of lived experience in order to be able to produce alterity at home.

‘Keeping your eyes open’

Dispersed among these Chroniques can be found several narratives ofthe genesis of an ethnographic posture in van Gennep’s childhood. Tobegin with, there is his passion for collecting objects whose significancemust be brought out as if they were the traces of previous periods andremote cultures, later known as prehistoric objects. On the other hand,when van Gennep was fifteen his father, a laryngologist, entrusted himwith microscopic specimens; and even though the young van Gennep

30 Giordana Charuty

did not gravitate towards medical studies as his father had wished, hereported having retained his passion for biology. Finally, while a pupil atthe lycée at Nice, he spent most of his time ‘doing bush school’, that is,actually experiencing a different way of conveying knowledge throughelective affinities between boys of different ages. But the passage toethnography – one of the numerous paradoxes in this itinerary – cameafter attempting the theoretical conceptualisation of totemism andabove all of ritual activity, of which he was to say repeatedly that itconstituted a sort of ‘internal illumination’ which permanentlytransformed his relationship with daily life.

To see the most familiar things differently involves a doublemovement of removing oneself from the common sense of ordinary lifeand incorporating strictly localised ways of living and speaking bydistancing oneself from them. Initially it was in Savoy, during hisholidays as a child and an adolescent, that van Gennep busied himselfwith this exercise. Moreover, during the period of the Manuel, leadingevery informant through his memories and his juvenile experiences wasto be erected into a principle of ethnographic enquiry. For the time being,‘keeping one’s eyes open’ designated the first step, which he described ina Chronique of 16 October 1909, when he had just passed the summerdoing fieldwork, having installed himself in the little town of Bonneville:

And, with the thought of being useful to others who were busying themselveswith regional ethnography, I will indicate here the method required: I situatemyself with regard to the Savoyards as if they were savages and their landwere located in the heart of Africa. I assume that nothing is known aboutthem, and that I myself in particular am entirely ignorant of their language,houses, legends, etc. This amounts to keeping one’s eyes open all the time, towonder at everything, to wish everything to be new and to take note of everyobservation. It will then suffice to check one’s personal observations againstthose of others and to let go of any useless ballast. In brief, it is necessary toput oneself in the right frame of mind, use the methods of an explorerthrown right into the middle of black or yellow populations, and to arrangeone’s observations rigorously in series. (2001: 68)

The second step, that of apprenticeship, is objectivised in the sameChronique of 16 October 1909:

As a temporary citizen of Bonneville, Haute-Savoie, I found myself shakingwith laughter [je m’y suis gondolé] repeatedly that summer, as wasappropriate. For you must realise that Bonneville is built in triangularfashion around a square which, according to the locals, has the form of agondola [gondole]. At the centre of the square is an alley with beautiful planetrees and an old fountain; surrounding the plane trees, a triangle of widepaths, cleanly maintained; and making a tour of the square untiringly for


some hours, going right down the middle of these paths, is to shake withlaughter [se gondoler]: oh, the joys of small towns! (ibid.: 67)

In summary, it is a matter of testing very physically, in his body, hisemotions, his thoughts, the impact of a landscape or a piece ofarchitecture, of other ways of living in a place and the categories ofthought borne by another language (a patois), as well as the attribution ofintentionality to other beings than just humans. This is a position thatconnects the insistent interrogation of literary or aesthetic avant-gardes– ‘Where is one primitive? When is one primitive?’ – to the existence ofother psychic states and other ways in which social life may functionwhich are no longer linked to another place or a remote past, but identifiedvery near to us, within ourselves. The claim of the specificity of thisknowledge, derived from a double movement – establishing a distancewhile at the same time reincorporating local micro-societies’ own usages– was to be a constant feature of all these Chroniques. Reviewing theManuel de folklore of P. Saintyves on 1 April 1937, he again wrote:

It must be admitted that everyone has the right to consider the facts fromthe angle that pleases him the most; it would be ungracious of me – justbecause my temperament pushes me to take facts and people in hand, tomassage them, and to have a horror for the fluid phraseology or the uncoiledmetaphysics – to reproach someone else for preferring them … [But] Iconsider it inauspicious for a science of direct observation like ours toimmure it in a verbal system and a study, when this requires fresh air, thebottle of white wine, disdain for what one is told, the diffusion of oneselfamong the mass, yet also the maintenance of the self as individualised aspossible, without contempt for anything whatever, and without the idea thatbookish instruction represents a superior human value. (ibid.: 189)

Apart from his criticism of de Saintyves there are other disagreements,those that opposed van Gennep to the Durkheimian sociologists fromthe outset. Saluting some years earlier, in 1931, the reissue of some ofRobert Hertz’s papers by L’Année sociologique in Mélanges de sociologiereligieuse et folklore (1928), van Gennep recalled the part he had playedin the genesis of Hertz’s enquiry into the cult of St Besse:

… one day I said to Hertz that to write [faire] a thirteenth book on the basis of[avec] twelve others was unworthy of him and that, reserving Savoy to myself,I advised him to ‘do’ Piedmont and go and live with the Alpine peasants beforereconstructing the psychology of the Australians! […] Whoever wishes tolearn what sociology could become in the hands of Robert Hertz, freed from thedogmas of the school, must read this volume. (2001 [June 1931]: 310)3

32 Giordana Charuty

Off the beaten track

Roaming around the countryside, either chancing encounters or shiftingmethodically from village to village while knowing ‘by heart’ which itemsto document; sharing in the informal sociability of rural life while riddingoneself of bourgeois value hierarchies; paying attention to thedetermining role of highly individualised actors, as well as the sexualdivision of competences and knowledge; submitting oneself to a technicalapprenticeship in the arts one intends to describe – these pieces of advice,distributed throughout the Chroniques and systematised in the Manuel, areinsufficient to characterise a good ethnography. Whether one’sinterlocutor is literate or not, the account of an enquiry must favour thesame lifting of internalised forms of social distance and intellectualcensure to reach, as in psychoanalysis, the unconscious cultural memory:

In folkloristic practice, this means that it is not necessary to submit witnessesto a methodical interrogation as a judge would do. However, it is necessaryto let them take short cuts and yield themselves up to reminiscences, whichmay seem only digressions, but which have the value of mnemotechnicalprompts through associations of ideas. (van Gennep 1943, I: 60)

One can assess how innovative this method, which seems so familiar tous today, was then when one recalls the inquisitorial form of questioning,which, fifteen years earlier, Marcel Griaule had recommended in hisinstructions for researchers on the Dakar-Djibouti expedition.4

For van Gennep, as for modern anthropology, this mode of interactionis directly governed by the particular regime of thought which governsthe facts of folklore, that is, ‘mores and customs’, a regime which he doesnot yet describe in terms of a symbolic logic but for which he acceptspermanence and universality, while rejecting any theory of survivals:

In folklore especially, it is necessary to take care not to presuppose scales ofvalue, nor that participation is prior to logic. In reality, always andeverywhere, people have used these two ways of thinking, and they continueto use them sometimes in certain moments or circumstances, sometimes inothers. The two ways of reasoning and concluding, and as a consequencethe two modes of action, are elements that are equally constitutive of andnormal in the thought of the whole human species. (van Gennep 1943: 97)

Training the collectors

The Chroniques respond to a pedagogical need on several levels. Thereaders of the Mercure de France are invited to become informantstemporarily. But in addition, lacking the power to train universityresearchers, van Gennep seems to have dreamt of a regionalisednational organisation, and to begin with he sets himself the task of


stimulating literate amateurs to adopt rigorous methods of enquiry, aswell as to identify learned societies and scholars who are already engagedin the practice of collecting, so as to transmit to them his requirementsand experience. To both, the same advice returns like a leitmotiv: distrustexhaustive questionnaires, pay attention to details and beware ofgeneralities that have no scientific utility. In addition, citing his ownexperience while reviewing the Ethnographic Congress of Neuchâtel:

After trials, I have come to prefer to draw up questionnaires that arerestricted, for example, just to the house, or to birth and death ceremonies,or to fairies, revenants and sorcerers, etc. It is better to do them several times,but then in great detail. (2001 [1 August 1914]: 105–9)

Beyond this attempt, there is a cartographic concern: being able totransfer all the ritual variants, hamlet by hamlet, on to a map of a scaleof eighty to a thousand in order to draw up ‘a great descriptive work ofrural, popular France’. But this attention to detail, this concern to keepone’s eyes open, also proceeds via an encouragement to multiply thedescriptive instruments – notably drawing and photography – withprecise instructions for the visual documentation of rituals, thusacknowledging that the image has descriptive powers beyond those ofwriting. Likewise, the quality of illustrations in publications – drawings,engravings, water colours – are the object of very careful remarks inhis reviews, for, undoubtedly more than writing, they constantly riskbeing pervaded by stereotypes of rurality that screen out the actualdiversity of modes of life.

However, does not training amateur collectors, for want of trainingthe professionals, amount to confirming a division of labour that partlycontradicts the principle of the personal experience of defamiliarisationby reintroducing a cleavage between the point at which data arecollected and that of their treatment?

The rejection of regionalism

The rejection of regionalism and the emblematic use made of culturaltraits leads van Gennep to stimulate all Mercure de France’s readers todevelop in their turn a different way of looking at and listening toordinary life and its significant details as a mode of opposition to the formsof celebration of ‘traditionalism’. Reviewing a book entitled La Bretagnedes druides, des bardes et des légendes in 1931, he affirms quite simply:

This is probably a little book of propaganda like hundreds of others alreadyin the literature called ‘regionalist’, which does more harm than good tofolklore. For this is usually just a retarded folklore as adulterated as thechemical aperitifs we have these days. (2001: 311)

34 Giordana Charuty

And the brochure ‘Art populaire et loisirs ouvriers’, presented by theInstitut International de Coopération Intellectuelle for the 1937Exhibition, incurs, retrospectively, a double condemnation for itsincompetence and reactionary ideology:

Little by little, exhibitions of regional costumes, more or less faked, have beenorganised; and each year now, one sees filing through different towns, andright up to the cinema in Bourg-la-Reine, troupes of actors in costumerepresenting the ‘French provinces’. These troupes sing arranged popularsongs, invariably with accompaniment, which deprives these songs of theirstrictly vocal character; they dance rounds, farandoles, carols, rigaudons,bourrées and God knows what, on the boards, in a closed room, or on thePromenade des Anglais, without the prior stimulus of the work ofhaymaking or of harvesting, in the bawdy atmosphere of the wedding day,without the iridescent light of the barns hazy with dust, or the smoke-blackened light of the rooms below. In brief, as the common people say, it’sa carnival, all right, but not so much fun as the proper one. (2001: 372)

Thus a militant position is affirmed, supported by its adherence to thenon-Marxist left, in order to dissociate a culturally authentic popularheritage from conservative values which encourage the performanceof rural customs that conform to clerical morality. But why, then, theregret that the leisure time activities of the peasants and workers shouldhenceforward be directed towards the songs of the café or the radio?

Disciplinary frontiers

To make people aware of ethnographic knowledge as a scientificdiscipline was all the more necessary, given that in the 1930s ruralsocieties were made the objects of collective enquiries into social history,sociology and human geography, which van Gennep could not ignore.A concern for cartography formed a part of this project as a visual toolproviding a demonstration of the autonomy of the ‘folkloristic’ thatsocial anthropology would later recognise as the autonomy of the‘symbolic’. Reviewing his own Folklore du Dauphiné (van Gennep 1932–33) on 15 August 1933, he identified a general property:

… those collective phenomena that are called folkloristic evolve according toan autonomous plan that is independent of geography, politicalorganisation, diocesan organisation, economic differentiation or dialect,which obey laws that one might summarily call sociological, althoughuniquely nuanced. […] I have used ten or so methods simultaneously(experimental, statistical, cartographic, psychological, comparative etc.),and the image one thus obtains of a group like that of the peasants of Isère


differs enormously from that which geographers, historians or novelists areable to give us, since the geographer subordinates man to the land, thehistorian the present day to the past, the romantic the normal to theabnormal. (2001 [15 August 1933]: 372)

This method has a demonstrative value when it allows the limited rangeof variants of a ritual designation or performance to be reconstructed.However, what had been a heuristic tool at a point in the foundation ofthe discipline would suddenly cease to be so once it had become an end initself, precisely because of its failure to lead to a conceptualisation of therelational character of symbolic thought. The multiplication of points ofenquiry simply led to a confirmation of the chance dividing up ofdifferences and the superposition of internal boundaries. This method,born in Switzerland, did not lead to the preparation of a nationalethnographic atlas in France. However, through spatial projection, itprofoundly influenced the treatment of ritual or technical facts, atreatment which ought to have deconstructed the national territory infavour of a social description of the ‘regions’, a genuine measure of theunderstanding of the differences which are the object of ethnography.

What always seemed relevant, on the other hand, was theexamination of relations with another field of knowledge, namely literarystudies. Van Gennep’s ties with philologists and specialists in the Romancelanguages determined his concern to insist on normalised forms ofdescription in taking account of oral narrative materials. By constantlyputting collectors of stories on their guard against any literarytransposition, which could only provide ‘fake’ documents, it wasespecially the typological and philological concept of the catalogue, suchas would impose itself in the 1950s, that he promoted, in opposition to thequest for a literary form of writing or the reassertion of regionallanguages. Nevertheless, alongside the identification of genres andtypical plots, attention became more focused on less formal narrativediscourses, performances and social institutions, in which were inscribedthe words of the storytellers or singers, all objects that ethnographerswould only later place at the centre of their analyses. Finally, hisreflections on the relations between ethnography and literature weremore complex than just the concern to establish ‘reliable documents’.Certainly van Gennep busied himself in decoding novels and short storiesas a source for historical ethnography, within a logic of the extraction ofdocuments. For example, this led him to unpack in meticulous detail the‘ethnographic illusion’ of novels set in the countryside, notably those ofGeorge Sand. But he is also attentive to identifying the subtle interactionsin dialect and literary writings by means of unexpected comparisons. Forexample, his Chronique of 15 February 1935, ‘A precursor of Stendhal: B.Chaix, a statistician from the Hautes-Alpes’, begins as follows:

36 Giordana Charuty

There is, in Stendhal’s prose, the Stendhal of Henri Brulard and the Letters,a rhythm and rubbing together of words, which I have always experiencedas pertaining to the Dauphinois patois when translated, but which Remy deGourmont, a Norman critic, commentators on other provinces and above allprofessors of literature have regarded as a matter of ‘style’, that is, as anintentional form of expression, not at all spontaneous … The texture ofDauphinois when heard, but not necessarily read, is quite different fromSavoyard or Provençal, its neighbours. It is more pounding, drier, and in itssyntax readily suppresses all redundancy.

Gourmont smiled at what he judged to be only a theory, and I do not knowto what extent Paul Léautaud or L. Royer, who, however, lives in Grenoble,would take seriously my affirmation that the real Stendhal – not tidied up forthe Paris salons, nor for the literary esteem of his time and afterwards (let ussay, with him, around 1880) – represents the Dauphinois patois preservedsince his childhood and imposed on the dulled French of good company.

However, the chances of my folkloristic researches have brought me anunexpected proof. I do not know whether or not it has escaped Stendhal’sfollowers, and I do not have the leisure to find out. But I doubt whether anyof them are far-seeing enough to find literature in reading any Dauphinoisstatistical treatise. (2001 [15 February 1935]: 346–47)

There follows an account of this treatise, Préoccupations statistiques,géographiques, pittoresques et synoptiques des Hautes-Alpes of 1845, and itsauthor, who was sub-prefect of Briançon from 1800 to 1815. Then, citinglong extracts from passages, van Gennep comments on the jerky writing,the compressed turns of phrase, the rhythmic variations, the ‘verbalcascades’ which juxtapose incidents without repeating grammaticalsubjects, which substantivise participles: ‘this is genuine Stendhal, but alsogenuine Dauphinois patois’. He concludes that the Baron would have donebetter to write short stories, even a novel, on the lives of these Dauphinois,whom he knew so well because he could speak their tongue.

Thus he was quite ready to admit specifically that certain novelistspossessed an ethnographic gaze that had escaped contemporaryfolklorists. When novelistic writing articulates the disparate elements ofthe social, not in a folkloristic reconstruction but in a biographicalexperience, it might even constitute the best means of accessingethnographic knowledge. These statements are not reserved for thereaders of the Mercure de France: one finds similar remarks in the firstvolume of the Manuel, where van Gennep states that infantile customs– in particular those of one’s ‘second childhood’ – are described betterby novelists than by ethnographers, and he cites in support of thisassessment the autobiographical accounts of Renan, Vallès, Mistral andPergaud (1943, I: 169, 174).


From folkloristic to symbolic autonomy

The rejection of folklore by social anthropology during the 1960s cameabout through a re-centring of analysis on the principles ofreproduction and the exercise of power within localised societies. But itwas also done by passing from ‘folkloristic’ autonomy in van Gennep’ssense – ceremonial customs, rites, ways of speaking and believing – tosymbolic autonomy in Mauss’s sense, as reinterpreted by Lévi-Strauss.

In passing from the ethnography of remote worlds to that of nearbysocieties, van Gennep always maintained as central the generalquestion of ritual, which he revived on the theoretical level before evenundertaking that vast description of rural France, which was nevercompleted. We know that the notion of a ‘ceremonial sequence’ governsthe idea that rites of separation, liminality and reincorporation projectthe moments of passage into space by dramatising the change for theindividual ‘who is passing’, as well as for the social group being affectedby this change. First in Italy, then in France, another interpretativemove was created by refusing to reduce this conceptualisation simply tofunction and form, and by abandoning the interpretative categories ofa Frazer or a Lévy-Bruhl that are preserved in the Manuel in order toassimilate ritual efficacy to the processes of magical action. But, in bothcases, the debt to the Manuel is evident on the part of researchersconfronted for the first time with the recurring question of the placeand legitimacy of an ethnology of Europe within general anthropology.

The readings of de Martino

We owe an initial metamorphosis of van Gennep’s conceptualisationsnot to a highly Durkheimian French anthropology, but to Italianreligious anthropology as relaunched by Ernesto de Martino on the eveof the Second World War. De Martino did not hesitate to turn theManuel de folklore français into a tool with which to oppose an indigenousapproach to the British intervention in ‘the Mediterranean’, as well asto American-inspired studies in applied anthropology. In particular, themethodological reflections of the French ethnographer served to guidede Martino – who was trained in the same fashion in the history ofreligions and was also a critic of Durkheim – in passing from knowledgeconstructed through an exclusive familiarity with ancient texts and theworks of missionaries to the ethnographic observation of the religiouspractices of southern Italy.

The preparatory notes of all de Martino’s initial enquiries in Lucaniaat the beginning of the 1950s, undertaken to revive strictly economicways of understanding the southern question in Italy, document anattentive reading of van Gennep’s work in order to define the proper

38 Giordana Charuty

method of enquiry: the necessity of distinguishing informantsaccording to whether they belong to a literate or a popular culture; theneed not only to collect accounts, but also to witness gestures in action;methods of criticising and classifying ethnographic documents; andthe requirement for comparativism. The first collective enquiry usedseveral questionnaires from the Manuel, on life-cycle rites, popularsongs and dances, magic, the church and the clergy. The use made ofthese is all the more surprising because, ten years later, de Martino wasto have two favourite expressions for deriding the ‘folkloristic’ position:‘the people sing’, and ‘from cradle to tomb’ (see Gallini 1995: 52).

But running through different unpublished versions made inpreparation for the editing of Sud e magia (de Martino 1959), one seesthe category of the rite of passage being progressively abandoned infavour of an existential type of interrogation of the historicity of theperson and another definition of the critical moments of individualexistence – no longer the passage from one social state to another, butthe confrontation of the individual with the historical development ofhis or her society. As for the notion of ‘magic’, this will be restrictedsolely to the therapeutic techniques that use ritual gestures andmythical accounts, while excluding all the customary prescriptions ofwhich life-cycle rites consist. However, the concern to historicise theseusages and to restore cultural flows between learned cultures andpeasant cultures from a sociological perspective systematises an aimthat is present everywhere in the Manuel. A large number of practices,pieces of knowledge and ways of speaking grasped while the enquiry istaking place can only be integrated into sets of significant relations bytaking into account the temporal depth evidenced by the traces andindications that are present in a great variety of documentarycollections. These include archives of administration and power,dictionaries, encyclopaedias, ecclesiastical and medical enquiries. Forthese collections, it is appropriate, depending on the case, to explain thenormative constraints that have informed the description.

There is a more hidden affinity, perhaps, that links the two authors:the importance accorded to the regimes of temporality of peasantcultures that are modelled on Christianity. With van Gennep, theintroduction of the notion of a ‘ceremonial cycle’ in the 1920s served todistinguish, alongside biographical time, the festive cycles based onseasonal variations, the official solar calendar, the Christian calendar ofthe saints, and the rhythms of agricultural and pastoral activities. Unlikehuman geography or Marxist-inspired social history, this involved notseparating material culture from the symbolic elaboration that gives itmeaning by recognising the uniqueness of the social construction oftime – the existence of a cyclical time produced by the ritual, unlike


events or history – which was particularly pregnant in the rural societiesof the nineteenth century, and right up to the Second World War.

Having become a sort of commonplace, this notion of a ‘ceremonialcycle’ became fixed within an unsurprising functional reading. DeMartino, on the other hand, revived this question by treating Christianityas an idea of time imbued with a tension between two models, one linear,the other cyclical, which anticipated more recent analyses ofChristianity as a ‘religion that has left religion behind’. But one recallsthat van Gennep’s work served above all as an ‘eye-opener’ to thecomplexity of the cultural history of southern Italy and of perceivingthe procedures of mythico-ritual symbolism that gave meaning to thecultural idiom of Apulian tarentism (de Martino 2005 [1961]).5

The French heritage

In France at the end of the 1960s, two opposed attitudes marked the re-launching of the discipline. One was to reject the ethnography of van Gennepas that of a folklorist, which would return the rural world to assimilation intothe domain of superstition and mental retardation. Such was the aim ofJeanne Favret-Saada, who, analysing the logic of sorcery in the NormanBocage, initially isolated a domain of social activity – rituals to remove sorcery– as an expression of an indigenous theory of magical efficacy that theanthropologist could not make his or her own (Favret-Saada 1981). The otherapproach was to use regional monographs and the Manuelas a sort of culturalmemory to adopt other forms of seeing and listening to contemporary societiesthat at first sight have none of the ceremonial richness of earlier rural societies.This was, to begin with, the aim of Yvonne Verdier in a research team led byLévi-Strauss, on Minot in the Bourgogne. Alongside Françoise Zonabend, TinaJolas and Marie Claire Pingaud, she worked to link an ethnography in thepresent with van Gennep’s theoretical enquiry regarding rites of passage andthe impressive cultural materials collected by ethnographers since the lastthird of the nineteenth century. The study she devoted to the cycle ofexemplary lives – the washerwoman, the seamstress, the cook – which leadsto an encounter with other village lives at the most crucial moments in theirexistences in order to ‘make custom’ (Verdier 1979), represents a profoundtransformation of van Gennep’s model on the basis of a doublemethodological choice. One is to adopt the point of view of the women, theirknowledge, their techniques and the world of prescriptions and prohibitionsthat govern representations of feminine physiology by conditioning fertilityor sterility, in order to reveal ‘the lives of young girls’, with their rules, rights andduties. The other is to identify semantic codes, in the Lévi-Straussian sense,that construct the symbolic logic of ritual action.

One example will suffice to illustrate the renewal of the analysis thatderives from this. Van Gennep devoted almost an entire volume of his

40 Giordana Charuty

Manuel to marriage rites (Vol. 1, part 2, 1946), marked simultaneouslyby a succession of highly stable sequences and a considerableproliferation of local variants. One of these enigmatic moments, therôtie,6 particularly caught his attention. Van Gennep rightly recalls thegreat historical depth of this culinary rite, attested since the medievalperiod, its generalisation and its resistance in the face of all attempts toban it, since it regularly became the object of disorder, a desire for itsabolition and the condemnation of its scandalous nature. He also notesthat the principal actors in this case are the young people, assisted byclose relatives, godfathers and godmothers. And he comments at lengthon the transformation of culinary materials and objects in order toconclude that the scatological form which became dominant during the1930s stresses an aspect of parody that was absent from usages attestedearlier. Finally, the cartography of the variants gives way here to aninterrogation of the rite’s meaning. According to him, the indigenousexegesis bears witness to the permanence of ‘very primitive ideas’regarding fertility. He rejects any interpretation along the lines of asimple decoding of symbols, but concludes, in a manner that is verylikely to disappoint today, with a ‘rite of the socialisation of marriage’.

For Verdier (1979), adopting a structuralist position consists inmaking explicit all the semantic relations, which, in the contemporaryor 1970s form of the rite, underlie what van Gennep traced back to theremote past of hypothetical primitive ideas about fertility. The rite usesa culinary code, that of sugar and spices, for an action which isequivalent to a ‘seasoning’ of the bride that is equivalent, from themasculine point of view, to a sexual act, public and shared, and onewith procreative value. On the other hand, however, the exploration ofthe vocabulary illuminates the rules of the transmission of procreativepowers between female generations. Thus, one might add, by imposinga language that is virile or scatological, the young people appear to bediverting the female cook’s role of guide [passeuse] to the extent that thevalue of marriage itself has changed, namely to perpetuate not a houseany longer, but a couple for whom the language of desire prevails (seeFabre’s review, 1980).

The work carried out in the 1980s and 1990s at the Centred’Anthropologie de Toulouse (EHESS-CNRS-Toulouse-Le Mirail),directed by Daniel Fabre, expanded this perspective by departing fromthe framework of monographic enquiry to pursue a re-description ofthe social institutions and symbolic logics characteristic of Christiansocieties. To begin with, it is to the observer sensitive to linguisticdifferences between ‘countries’ that we owe the attention paid tosituations of diglossia as a recognition of cultural differences. Bycontrast to monographs that are blind to linguistic differences and the


social relations that inform them, it is a matter of restoring thecoherence of semantic logics modelled by the diverse dialects in whichthese are actually thought. Fabre has revived the analysis of socialinstitutions so well identified by van Gennep – such as masculine youth,with its principles of organisation, its rights and duties – by taking intoaccount its confrontations for the maintenance of a prerogative, thesocial control of ‘houses’. From this point on, and in opposition to theiratemporal definition, the conflictual relations of village societies withcertain categories of ritual, notably the political stakes of the charivarior hullabaloo, have been made evident.7 Claudine Fabre-Vassas’sexploration of all forms of expression of a ban on the consumption ofpork for Christians and of a popular antisemitism made systematic useof the cultural materials collected in the Manuel: for example, theethnography of a twelve-day cycle between Christmas and Epiphany,which reveals the metaphysical issue involved in cooking one part ofthe pig, namely the blood (Fabre-Vassas 1997 [1993]). Likewise thestudy of spiritual kinship by Agnes Fine (1994) made considerable useof customary usages ordered by the succession of life-cycle rites.

To describe a ‘Christian custom’ – the discontinuous ensemble ofusages, prescriptions and ritual or ceremonial activities that linkbiographical time with the cyclical time of a localised society modelled ona unique biography, namely the life of Christ – it is certainly necessary tocall into question the categories of medicine and popular religion thatorganise van Gennep’s ethnography.8 Making an ethnography of twocategories of disorder that are invariably thought of as belonging together,namely hysteria and epilepsy, I have myself been led to describe not anethnopsychiatry but a metaphysics in action, that is, certain ways throughwhich the most abstract theological notions may become the object of anexperience through the senses (Charuty 1997). In Europe as elsewhere,the ‘person’ is produced by a work of modelling the body and by socialinteractions that link the different ages of infancy, adolescence and ‘youth’to the gradation of the ritual operations of the clergy and customaryofficiants. Parallel with this, socialised trials within age groups ensure thebiographical inscription of dogmatic utterances transmitted by thecatechism in the fashion of impersonal knowledge. This modelling givesplace to a profusion of discourse on ritual faults in the fulfilment ofcustomary liturgical rites and prescriptions with regard to the relations tobe maintained with categories of beings – souls, the deceased, the Virgin,saints – which are simultaneously deprived of corporality and creditedwith intentionality. And the anthropologist then discovers the significanceof a category – the ‘sickness of the saint’ – properly identified by vanGennep as designating all sorts of somatic, psychic and social disorderssanctioning these transgressions, in his concern to treat Christianity as ‘a

42 Giordana Charuty

magico-religious activity’.9 Only the analysis of rituals of atonementpermits the etiological thought that unifies them to be explained, namelydiversified forms of dissociation of the Christian person. Thus alloperations of measurement and of the manufacture of doubles, iconicand aniconic, of sick bodies to ‘revive’ them suggest a literal reading ofmetaphors which, in ethical discourses and devotional texts, oppose theheaviness of the ‘flesh’ to the lightness of the ‘soul’.

Concluding remarks

Van Gennep’s work guaranteed a transition between the history of religionsand ethnology in Italy, and between positivist ethnography and theanthropology of the symbolic in France. It is not only a turning point inthe history of the discipline, fixed within a limited period – it has had thevalue of a ‘passage’ in the disciplinary conversion of Europeanistethnologists of my generation, who, most often, come from otherdisciplines, such as literature, philosophy and history, and whose universitytraining made them read Durkheim, Lévi-Strauss and Malinowski, but whofailed to realise that an anthropology of themselves was conceivable.

However, our understanding of van Gennep is influenced by areading of de Martino and Lévi-Strauss, reintroducing the dimension ofpower to the very heart of the logic of meaning, to explore conflictualsituations born of confrontation between hierarchised cultural codes,without, nonetheless, reducing the symbolic to an emblematic orexpressive function of social divisions. Thus, although local societies inwhich our first ethnographic experiences are inscribed very often seemeddechristianised, in fact we have been led to recognise the heterogeneityof social practices capable of taking charge of religious representations.We clearly see that the treatment of the cultural materials collected byvan Gennep – who became an ‘institution’ in himself – was only madepossible by resuming the dialogue with historians and sociologists, whilenonetheless maintaining the specificity of the questions of a generalanthropology in the face of these disciplines.

Notes

1. Arnold Van Gennep, Manuel de folklore français contemporain, Paris, Picard (1938–1958). Van Gennep had begun by publishing Volumes III and IV: Questionnaire.Provinces et pays: Bibliographie méthodique (1937) and Bibliographie méthodique (fin)(1938), then Volume I in six parts: Introduction générale. Du Berceau à la tombe:naissance, baptême, enfance, adolescence, fiançailles (1943); Du berceau à la tombe (fin):


mariage, funérailles (1946); Cérémonies périodiques cycliques: 1. Carnaval, Carême, Pâques(1947); Cérémonies périodiques cycliques: 2. Cycle de mai, la Saint-Jean (1949); LesCérémonies agricoles et pastorales de l’été (1951) and Les Cérémonies agricoles et pastoralesde l’automne (1953). Cycle des Douze Jours (1958) was published posthumously. Theintroduction to each volume focuses on a particular theoretical problem.

2. Appeared originally in Mercure de France, 16 January 1913.3. This somewhat cryptic passage contains an implicit criticism of the ‘armchair

anthropology’ of the Durkheimian school, including, at this time, Hertz.4. For Michel Leiris’s criticism of this relationship of suspicion, see Jamin (1996: 38–39).5. This work, originally published in Italian in 1961, has recently been translated into

English, with notes by Dorothy Louise Zinn and a preface by Vincent Crapanzano (deMartino 2005 [1961]).

6. Literally ‘roasted’, this refers to a marriage rite in which drink and food is brought tothe newly married couple during the wedding night. The basis of this culinarypreparation, which varied according to local traditions, was for a long time breadsoaked in broth or spiced and sugared wine. Practically everywhere after the FirstWorld War, this soup was replaced by a mixture of fizzy wine and chocolate carriedin a chamber pot.

7. Fabre (1986). In the same manner, Natalie Z. Davis has acknowledged her debt tovan Gennep’s Manuel in analysing the practices of the charivari in the sixteenthcentury (see Davis 1975).

8. For a presentation of these works, see Charuty (2001).9. Van Gennep devotes a separate rubric to this category in his bibliography of popular

medicine by noting, correctly, that the relationship of Christians with saints is highlyambivalent, involving a power that is now maleficent, now beneficent. Contemporaryanthropology shows the relevance of this remark, which has been forgotten by thehistorians and ethnographers of ‘popular religion’.

References

Belmont, N. 1974. Arnold Van Gennep: le créateur de l’ethnographie française, Paris: Payot.Charuty, G. 1997. Folie, mariage et mort: pratiques chrétiennes de la folie en Europe occidentale,

Paris: Le Seuil.——— 2001. Du catholicisme méridional à l’anthropologie des sociétés chrétiennes, in D.

Albera, A. Blok and C. Bromberger (eds), L’anthropologie de la Méditerranée/Anthropologyof the Mediterranean, Paris and Aix-en-Provence: Maisonneuve and Larose/MMSH.

Chiva, I. 1987. Entre livre et musée: emergence d’une ethnologie de la France, inEthnologies en miroir: La France et les pays de langue allemande, essais réunis par I. Chivaet U. Jeggle, Paris, MSH.

Cuisenier, J. (ed.). 1980. Hier pour demain: arts, traditions, patrimoine, Paris, Editions de laRMN.

Davis, N.Z. 1975. Society and culture in early modern France, Stanford: Stanford UniversityPress.

de Martino, E. 1959. Sud e magia, Milan: Feltrinelli.——— 2005 [1961]. The land of remorse: a study of southern Italian tarentism [La terra del

rimorso], London: Free Association Books.Dundes, A. (ed.). 1992. The evil eye: a casebook, Madison: The University of Wisconsin

Press.Durkheim, É. 1912. Les formes élémentaires de la vie religieuse, Paris: Alcan.Fabre, D. 1980. Passeuse aux gués du destin, Critique, 402 (November): 1075–99.

44 Giordana Charuty

——— 1986. Le privé contre la coutume, in P. Ariès and R. Chartier (eds), Histoire de lavie privée, Paris: Le Seuil, vol. 3.

——— 1992. Le Manuel de folklore français d’Arnold Van Gennep, in P. Nora (ed.), LesLieux de mémoire, Les France, vol. 2: traditions, Paris: Gallimard.

Fabre-Vassas, C. 1997 [1993]. The singular beast: Jews, Christians and the pig [La bêtesingulière: les juifs, les chrétiens et le cochon] (tr. C. Volk), New York: Columbia UniversityPress.

Favret-Saada, J. 1981 [1977]. Deadly words: witchcraft in the Bocage [Les mots, la mort, lessorts: la sorcellerie dans le Bocage] (tr. C. Cullen), Cambridge: Cambridge UniversityPress.

Gallini, C. 1995. La ricerca, la scrittura, in E. De Martino, Note di campo: Spedizione inLucania, 30 sett.–31 ott. 1952 (ed. C. Gallini), Lecce: Argo.

Hertz, R. 1928. Mélanges de sociologie religieuse et folklore, Paris: Alcan.Jamin, J. 1996. Introduction to Michel Leiris, Miroir de l’Afrique, Paris: Gallimard.Lévi-Strauss, C. 1962. La pensée sauvage, Paris: Plon.Mauss, M. 1908. Review of Arnold van Gennep, Mythes et legendes d’Australie: études

d’ethnographie et de sociologie, Année Sociologique, 10: 226–28.——— 1909. Review of Arnold van Gennep, Les rites de passage, Année Sociologique, 11:

200–2.Needham, R. 1967. Introduction to Arnold van Gennep, The semi-scholars, London:

Routledge & Kegan Paul.Privat, J.-M. 2001. Preface to Chroniques de folklore d’Arnold Van Gennep: recueil de textes

parus dans le Mercure de France 1905–1949, Paris: Editions du Comité des TravauxHistoriques et Scientifiques.

van Gennep, A. 1904. Tabou et totémisme à Madagascar: étude descriptive et théorique, Paris:Leroux.

——— 1906. Mythes et legendes d’Australie: études d’ethnographie et de sociologie, Paris:Guilmoto.

——— 1909. Les rites de passage, Paris: Nourry.——— 1911. Les demi-savants, Paris: Mercure de France.——— 1920. État actuel du problème totémique, Paris: Leroux.——— 1932–33. Folklore du Dauphiné (Isère): étude descriptive et comparée de psychologie

populaire, Paris: Maisonneuve (2 vols).——— 1938–58. Manuel de folklore français contemporain, Paris: Picard (5 vols).——— 1943. Manuel de folklore français contemporain, vol. 1, Paris: Picard.——— 2001. Chroniques de folklore d’Arnold Van Gennep: recueil de textes parus dans le

Mercure de France 1905–1949 (ed. J.-M. Privat), Paris: Editions du Comité des TravauxHistoriques et Scientifiques.

Velay-Valentin, C. 1999. Le 1er Congrès International de Folklore de 1937, AnnalesHistoire, Sciences Sociales, 2 (March–April): 481–506.

Verdier, Y. 1979. Façons de dire, façons de faire: la laveuse, la couturière et la cuisinière, Paris:Gallimard.

Chapter 2

CANONICAL ETHNOGRAPHY:HANOTEAU AND LETOURNEUX ON KABYLE

COMMUNAL LAW

Peter Parkes

Perhaps never was a system of self-government more radicallyimplemented, never has an administration relied on fewer functionaries,nor imposed less on those governed. The ideal of liberal and effectivegovernance – whose formula our philosophers forever seek in a thousandutopias – was a living reality for centuries in the Kabyle highlands.(Hanoteau and Letourneux 1873 [1893] II: 1)

La Kabylie et les coutumes kabyles by Adolphe Hanoteau and AristideLetourneux is a unique monument of early legal ethnography. Threelarge volumes, amounting to fifteen hundred pages, reported a decadeof intensive investigation among the Kabyle Berbers of Algeria in the1860s. It is recognised to be a definitive account of their autonomoussocial organisation by a distinguished line of Maghribian ethnographers,from Émile Masqueray and Robert Montagne to Jacques Berque andJeanne Favret. Its rare archival documentation is being redeemed bycurrent anthropological historians of the Kabyles such as Alain Mahéand Tilman Hannemann. Yet it is generally unknown to anthropologybeyond an intimate circle of Berber specialists.

This neglect is readily understandable, for the monograph has theforbidding appearance of an overblown gazetteer. That was its intendedpurpose, compiled by an army officer and an imperial legal councillorjust after the French conquest and pacification of Kabylia in 1857.1 Itsencyclopaedic documentation is barely leavened by cultural or historicalexegesis. The first volume is an exhaustive compendium of thetopography, geology, flora and fauna of the Jurjura massif, followed by

tabulated population statistics and detailed synopses of traditionalmedicine, agriculture, crafts and industries. Its major anthropologicalinterest lies in the remaining parts, but these appear equally daunting.A vivid but summary account of indigenous social and politicalorganisation prefaces a vast analytical inventory of Kabyle communallaws – exactly formatted in accordance with the Napoleonic Civil Code– followed by translations or extracts of some fifty written ‘canons’(qanun, pl. qawanin) or itemised lists of village regulations, assiduouslycollected from all the main tribes of Kabylia. This would never be relaxedreading for an armchair anthropologist: ‘a work of erudition, exclusivelytechnical, reserved for specialists’ (M. Hanoteau 1923: 143). As anofficially sanctioned manual for military administration it is nowadayseasily dismissed by postcolonial sensibilities as a monstrous artifice ofoccidental arrogation. Yet it remains one of the most meticulouslydocumented accounts of communal institutions of indigenous justiceand moral order available to anthropology. It is, indeed, the canonicalethnography of early French colonial social science.2

Conceived within a decade of Morgan’s League of the Iroquois, itbelongs to a foundational era of incipient field anthropology, combiningclassical methods of textual-philological scholarship with the inductiveobservational sciences. Informed by engaged and prolonged fieldexperiences, Hanoteau and Letourneux also derived their majoranalytical insights from the collaborative commentaries of one primaryinformant – a Berber counterpart to Ely Parker among the Iroquois(Tooker 1983, Trautmann 1987: 43–50) – who was well placed tointerpret their administrative and legal interests as a maraboutmediator and Muslim jurist among Kabyles.

This essay aspires to reconstruct their triadic field collaborationduring the 1860s – assessing the respective contributions of a soldier,a magistrate and a marabout, who conjointly established an originaljuridical ethnography of communal law and consensual self-government. It also considers the applied intentions and ambitions ofthese authors, in an era of precarious military rule and intrepiddevelopment planning in Kabylia, shortly before its fateful civiliancolonisation under the Third Republic.

Collaborative fieldwork

Over four years we neglected no available means of investigation: the studyof qanun laws, reading the reports of communal deliberations (jama‘a) andthe decisions of clerics (‘ulama), with daily examination of public and privatepractices, together with information taken from those actively involved inthese affairs prior to the French occupation [of 1857]. (I: v)3

46 Peter Parkes

Figure 2.1. Commandant Adolphe Hanoteau at Fort-Napoléon, 1861. Source:Bernard (1930: 325).

A one-page preface to La Kabylie et les coutumes kabyles scarcelyadvertises its peculiar conditions of collaborative fieldwork. However,a more detailed memoir was published fifty years later by MauriceHanoteau (1923), who witnessed its last moments of compilation as aschoolboy of twelve, visiting his father in Kabylia during the summerholidays of 1868. Hanoteau’s family also assisted two subsequentscholars of Berber customary law, Augustin Bernard and Louis Milliot(1933), whose access to his home archives enabled them to reconstructhis longstanding project.4

This project began within a decade of Hanoteau’s posting as amilitary engineer to Algiers in 1845. A graduate of the ÉcolePolytechnique, he was earmarked for service within the Bureau Politiquedes Affaires Arabes (or bureaux arabes), the elite military department ofindigenous administration in Algeria.5 Its director was Colonel Eugène

Canonical ethnography 47

48 Peter Parkes

Daumas, a soldier-scholar who had earlier campaigned in Kabylia andwho was then completing its first regional history (Daumas and Fabar1847). Hanoteau also saw military action in Kabylia during the 1850s,but most of this decade was devoted to his linguistic training in Algiers.Under Daumas’s patronage, he became associated with the distinguishedArabist William de Slane, translator of Ibn Khaldun’s Histories of theBerbers, and with other soldier-scholars engaged in writing up early fieldreports on the Kabyles (e.g. MacCarthy 1847–48, Carette 1848).

Hanoteau soon published a pioneering series of philological studies:original grammars of Kabyle and Tuareg Berber languages (1858a,1860), and a remarkable anthology of Kabyle oral poetry (1867).6 Hisenvisioning of a similar compilation of customary law stems from thislinguistic scholarship, for Hanoteau had already transcribed a qanun‘village charter’ as a speciment text of Kabyle dialect (1858a: 324–38,1858b). The existence of these written rulings – inscribed in Arabic bymarabout clerics at village assemblies – had only just been discovered.Apart from their historical value, these rare village records offered aninvaluable means of examining underlying principles of communalgovernment, necessary for indirect military administration, since anofficial policy of non-interference with traditional social organisationhad been proclaimed by Marshal Randon on the defeat of insurgent AitIraten Kabyles in 1857. There was therefore widespread interest inhaving further qanun rulings collected throughout Kabylia.7

In January 1859 Hanoteau was appointed to command a bureauarabe outpost at Dra el-Mizan (western Kabylia), and in the followingyear he was posted to Fort-Napoléon, in the tribal heartland of the AitIraten. As Maurice Hanoteau recalled:

Commandants enjoyed widely extended powers over indigenes. Issues thatarose were numerous and diverse – political, administrative, judicial – andtheir resolution was always a delicate matter, since France had determinedto leave intact the traditional organization of the country. To actconscientiously, to judge fairly, one needed to know in detail the people oneadministered and the local laws by which they governed themselves …Commandant Hanoteau scarcely obtained these details without seeing themfor himself in the villages … Such information was also available at Dra el-Mizan, where the commandant’s office was open to all. Some came to pleadan injustice or to defend themselves from a criminal accusation; some toaccuse fellow tribesmen, or even a French colonist; others to make appealsfor themselves, their family or their village. It would not take long to surmisefrom all these claimants and plaintiffs the full scope of Kabyle legal codes.(M. Hanoteau 1923: 138–40)

But despite Hanoteau’s linguistic abilities, his knowledge of theintricacies of Kabyle customary law would depend upon a khoja


interpreter-secretary, the native scribe attached to each outpost. It wasthe khoja’s responsibility to register the public acts of communalassemblies, to prepare briefs of appeals, and to transcribe theseproceedings for monthly reports. At Fort-Napoléon, this was the duty ofSi Mula n Ait u ‘Amar, who is paid a handsome if cursory tribute in thepreface to the monograph:

We found, above all, a precious assistant in Si Moula Naït Ameur; hiseducation, as advanced as that of any Kabyle marabout might be, embracedMuslim law as well as customary law; and his word – in accord with hisrenown and the influence of his family – was respected in the councils ofhis tribe. (I: v)

Maurice Hanoteau again fills in the gaps in his later memoir, for on hisfather’s arrival at Fort-Napoléon in November 1860, Si Mula wasimmediately employed to collect qanun rulings throughout the district:

Si Moula had problems knowing what was expected of him at first. But oncehe understood, he became engrossed, marshalling every effort to cooperatetowards the end pursued. He would reflect on administrative and juridicalquestions, and if he knew of any fact – be it the judgement of a jama‘aassembly or an article of qanun law – which either confirmed or qualified agiven opinion, he would readily convey it. He was one of many assistants;but he was the main informant, the most useful, and the most precious. (M.Hanoteau 1923: 142)

We shall see that Si Mula would become at least an equal co-author ofLa Kabylie et les coutumes kabyles as Letourneux. He was a dignitary ofAit ‘Amar marabouts among Ait Iraten, resident at their village ofTamazirt near Fort-Napoléon. He was also a religious cleric (‘alim) ofSayyid descent, whose ancestors had emigrated from Turkey in theseventeenth century, renowned as teachers of Arabic literacy andKoranic education among Ait Iraten.8 In the learned tradition of hisforefathers, Si Mula had studied Maliki fiqh jurisprudence at a localreligious college. With a politically influential elder brother, Si Lunis,he may have assisted in the capitulation of the Ait Iraten confederationto Marshal Randon in 1857, after participating in their insurrection,for a fine two-storey house was then built for the two brothers by Frencharmy engineers at Tamazirt. Si Lunis and Si Mula would remainsteadfast supporters of French officers at Fort-Napoléon, gaining thehighest appointments of political and judicial authority among AitIraten. But in 1860, Si Mula was only beginning to acquaint himselfwith the collaborative potential of Hanoteau’s ethnographic ambitions.

In 1862, after four years in the field, Hanoteau was recalled to ametropolitan posting in Algiers. His reports detailing the intricacies of

Figure 2.2. Canoun des Beni Ouaguennoun. Opening column of qanun rulingstranscribed in Arabic with French translation by Captain Alphonse Meyer, militaryinterpreter at Dellys, 1864. Source: Bernard and Milliot (1933: Pl. IX).

Kabyle social institutions, yet also ‘the admirable simplicity of theircustomary laws’, had gained widespread attention among his militaryand civil superiors.9 These included Louis-Adrien Berbrugger, presidentof the Historical Society of Algeria and founder of its Revue africaine, aswell as Ismail Urbain, the famous half-Guyanese convert to Islam andSaint-Simonian advocate of pluralist democracy, who was then aninfluential adviser on colonial policy to Napoleon III.10 Concerted planswere made for Hanoteau to publish his field notes on Kabyle customarylaw as a series of articles in the Revue africaine. But these plans weresuperseded by a far more ambitious project: ‘The idea took shape oftreating the Kabyle question on a truly scientific footing, within acomprehensive work embracing everything known about theorganization of that society’ (M. Hanoteau 1923: 143).11 Relieved fromother duties to concentrate on this programme, instructions wereposted by Hanoteau to bureaux arabes commandants throughoutKabylia. They were to submit to him ‘as soon as possible … copies of allthe main qanun laws of Kabyle villages’, ensuring that each was ‘ascomplete as possible, sending translations with transcriptions of theoriginal Arabic texts’ (letter of 4 April 1864, in Bernard and Milliot1933: 4). Within a few weeks, Hanoteau received numeroustranscribed copies of qanun rulings (see Fig. 2.2).It was at this juncture that Hanoteau by chance encountered AristideLetourneux, a magistrate at the Imperial Court of Appeal in Algiers,

50 Peter Parkes

also familiar with the Jurjura highlands of Kabylia as a keenexpeditionary field botanist. Hanoteau was already encounteringdifficulties in tackling Kabyle customary law without training incomparative jurisprudence. Letourneux, versed in French national lawas well as Muslim shari‘a, would be an ideal collaborator, and he readilyagreed to become co-author of the envisaged monograph.12

After two years relief from regular duties, however, Hanoteau wasordered to return to his old field command at Fort-Napoléon inFebruary 1866. This was a back-posting he may have requested inorder to re-establish his vital collaborative links with Si Mula. ForLetourneux had meanwhile become distracted from the project byrenewed passions for field botany that jeopardised his authorialcommitment. Only during the summer legal vacation in 1868 was hepersuaded to stay on at Fort-Napoléon:

M. Letourneux, recovering his original fervour, then drafted the last chapterson Kabyle law … Every morning, the two authors worked together, almostalways with Si Moula, but sometimes with other Kabyles too … Afternoonswere devoted to copy-editing, mostly done by M. Letourneux, whooccasionally verified a point in doubt by listening to Kabyles. This daily workonly ended in the evening after dinner; and it would resume early nextmorning in the commandant’s office until it was all brought to completion.(M. Hanoteau 1923: 146f.)

The manuscript was thus ready in late September 1868 to be submittedto Marshal MacMahon, who had commissioned its publication by theImprimerie impériale. After delays of scrutiny by a beleagured war officein Paris, its first volume was printed in 1872, the second two in 1873,acclaimed in a rapturous notice by Ernest Renan (1873).

Participant recollection

Fifty years later, Maurice Hanoteau’s memoir was vigorously written toprotect his father’s legacy from a damaging defamation. A colonial judgein Algiers had intimated that the monograph was a mere ‘codifiedcompilation’ of Kabyle customary laws that had been collusivelyorchestrated by Si Mula and Si Lunis, and it was even suggested that thetwo brothers had subsidised expenses of the research, adopting Hanoteauand Letourneux as their tribal guests at Tamazirt (Luc 1917: 54f.).Although this aspersion was easily rebutted, the subsequent archivalenquiries of Bernard and Milliot do indicate Si Mula’s extensiveinvolvement in every stage of the project. They even reproduced a longletter from Si Mula in 1885,13 responding to Hanoteau’s earnest enquiries


Figure 2.3. To Whom we owe obedience – Serene and Perfect Knight, Upholder of the Realm,Glory of Warriors, Commandant of the Circle of Dra el-Mizan – I render these customs in useamong Beni Iraten… Dedication of Ait Iraten qanun rulings submitted to Hanoteau by SiMula, ca. 1859/60. Source: Bernard and Milliot (1933: Pl. V).

about a surely improbable circumstance of Kabyle customary law: itsprovisions for the inheritance of property by a hermaphrodite (Bernardand Milliot 1933: 27–30, cf. Hannemann 2002: 90–93). This curiouslyrevealing document may give us a candid impression of Hanoteau’srelentless techniques of field interrogation, here exploiting Si Mula’sexpertise in Maliki Muslim jurisprudence, which he is asked to comparewith Kabyle custom. It also identifies Si Mula’s fine Arabic handwriting,found to recur on all copies of qanun codes issued from Fort-Napoléon –including a presentation copy of Ait Iraten customs inscribed in elegantAndalusian calligraphy, dedicated to Hanoteau while he was still at Drael-Mizan (see Fig. 2.3). Bernard and Milliot’s inventories of other qanundocuments in Hanoteau’s archives further demonstrate that almost allwere commissioned by him in the early 1860s.

52 Peter Parkes

Hanoteau was probably not a hyperactive fieldworker or homme deterrain in the later proclaimed Malinowskian mode. Rooted in the field,he was a studious philological ethnographer, a homme de cabinet, perhapsrarely operating off the verandah of the commandant’s office at Fort-Napoléon, where he would be reliant on Si Mula’s expert assistance.Despite Maurice Hanoteau’s protestations of his father’s many variedexperiences of Kabyle village life, these are not mentioned in themonograph. A bureau arabe commandant would certainly visit hisconstituencies on regular district circuits by horseback; but theseoccasions, under armed escort, could have afforded few opportunitiesto observe the intimacies of Kabyle village affairs that are relayed in theethnography. In any event, Hanoteau was determined to describe Kabylesocial institutions retrospectively, in the years just prior to theircapitulation to French authorities in 1857, for he was acutely aware ofthe disruptive effects of the military presence he embodied. Theethnography is thus reconstructive, yet also scrupulously synchronic. Itrejects all historical speculation about a barely documented Kabyle past,reassembling its autonomous social and political circumstances withina bare decade of its composition. Every illustrative case is therefore anindirect report of some recently recalled incident, doubtless mainlyrelayed through Si Mula. Yet its substantiation always returns to thoseqanun documents, the monographs’s so-called pièces justificatives.

Qanun rulings were certainly being written in Kabylia beforeHanoteau’s arrival there, even if his collection seems to have beenlargely created in response to his own orders, for several documentswere copies of prior manuscripts dated at least a decade earlier (III: 351,447–50). A classic qanun decree would be the proclamation of anagreement about some new ruling or readjustment of customary lawthat had been decided by a communal jama‘a (or tajmaat) assembly. Atleast one such proclamation dated from the mid-eighteenth century: aconfederational treaty that suspended women’s rights to inheritproperty in the interests of maintaining inter-tribal peace (III: 451–54,cf. Patorni 1895, Mahé 2001: 68ff.). But most qanun rulings weresimpler tariffs of fines imposed for village offences, such as thefts andbreaches of the peace, or cursory regulations about marital affairs andcontrols of sumptuary expenses. Such rulings might be collated as listsamounting to a hundred or more items. As Hanoteau realised, theqanun deeds of any community recorded only a fraction of itsrecognised customs (‘ada, pl. ‘awa‘id), which were otherwise committedto oral memory by illiterate village elders (II: 101).14

The genius of Hanoteau’s project was to realise that these unpromisingscraps of village law might serve as reliable epigraphic foundations forreconstructing the entirety of Kabyle traditional society. Assembled in


54 Peter Parkes

sufficient quantities, one could discern recurrent themes and resonantmotifs that adumbrated their underlying principles as well as their regionalvariants. Enigmatic qanun rulings also served as perfect elicitatory cues formore intensive field interrogations: they were undeniable agreements ofcommunal law, whose subsequent justification and explication by enlistedinformants would unravel their tacit social circumstances. These includedintimate details of the vulnerable moral reputation (hurma) of a community,otherwise scarcely accessible to outside enquiry. Hanoteau would also becrucially assisted in this delicate unravelling of village morality by Si Mula,an experienced marabout mediator, who may have witnessed many qanunrulings that he later collected and copied for Hanoteau.

Their colloquies at Fort-Napoléon evidently informed themonograph’s introductory sketch of Kabyle society (II: 1–134). Thisincludes vivid portrayals of a typical village (taddart) and its sof politicalfactions; its jama‘a public assemblies, amin president and electedcouncil; its festivals and redistributive welfare; its complex conditionsof ‘anaya protection over dependants; and its tribal or confederationalleadership in peacetime and in war. All of these finely evoked details arepresented just as they might have been observed by Hanoteau – until,in an abrupt concluding chapter, one is disarmingly apprised that theentire tableau vivant was artfully reconstructed. For everything had beentransformed under French military rule:

The autonomy of the village as a corporate body no longer exists; thepolitical powers of its jama‘a assembly have no meaning any more, and theyhave all disappeared … Its administrative powers are nominally intact, butthey are nonetheless subject to our governing control … The roles of theamin [assembly leader], without being officially altered, have necessarilychanged along with those of the jama‘a assembly, now accountable to theFrench authority for all that happens in the village, being the executiveagent of its orders … These, in brief, are just some of the modifications ofKabyle organization since the conquest that might be mentioned. (II: 133f.)

Hanoteau may have overstated these modifications, just as he understatedhis own field observations. As commandant, he would be besieged by Kabyleplaintiffs demanding his adjudication of intransigent disputes. He wastherefore an engaged participant auditor of Kabyle juridical procedures,even if his own experiences were, alas, unelaborated as case histories.15


Codified compilation

The ethnographic introduction to the second volume was only intendedto give a congenial overview to its substantive documentation: thoseassembled qanun rulings, whose principles of social organisation aredistilled in its massive Second Section, Droit Civil. Here one would expecta major contribution from the jurist Aristide Letourneux, for thetreatment of Kabyle familial institutions is all exactingly framedthrough the thematic grid of the Napoleonic Code civil. Even the layoutof the account conforms with its codified prototype: by Title andChapter, with subset paragraphs of terse judicial prose, as if these werethe deliberated decrees of some Grand Kabyle legislative assembly.

This format is easily deprecated as a preposterous imposition of a whollyalien scheme of modern legislation, un pur plaquage artificiel (Bousquet1950c: 448). But on inspection, its imposed matrix was perhaps notinappropriate for organising the disparate qanun fragments and verbalcommentaries that Hanoteau and Si Mula had assembled. Each Title thussets out purportedly rational discriminations of natural law beforeconsidering their variable recognition in Kabyle custom; each Chapter thenelaborates particular circumstances of Kabyle village institutions, withfootnotes on qanun rulings indicating their concordance or divergence withRoman private law or Muslim shari‘a, indexing their equivalent numberedarticles in French national law. This familiar format would facilitate anenvisaged administrative application of the monograph as a field manual foradjudicating officers. But it was also a possibly felicitous framework forcomprehending Kabyle customary law in this period; for it required its qanundecrees to be regarded from a relatively unprejudiced rational perspective –in the enlightened spirit of Montesquieu – more or less untainted by thespeculative historicism then characteristic of ethnological jurisprudence.

Some historicist speculation is unquestionably present in themonograph, notably in the introduction to its Third Section on PenalLaw (III: 53–59). This is presented as having a peculiarly progressivecivic morality and social clemency, overlaying an archaic bedrock ofpersonal law condoning violent retribution. Copious footnotes citecorrespondences of Kabyle legal concepts with those attested in AncientHebraic, Indian, Greek, Roman and early Germanic law, againintimating an evolved but arrested development of Kabyle communalmorality and criminal justice.16 These erudite annotations were surelycontributed by Aristide Letourneux, an acknowledged antiquarian whowould be familiar with the comparative historical jurisprudenceemerging from Savigny’s Rechtsgeschichte. Yet Hanoteau’s firmlyimposed grid of the Code civil otherwise demanded a strictlycontemporaneous assessment of Kabyle legal discriminations that were

56 Peter Parkes

evident in the qanun documentation. Such sophisticated institutions ascivil contract, trust law and commercial partnerships are therefore allgiven due analytical scrutiny, even if they had but faintcorrespondences in Kabyle customary usage.17

This rigorous empirical discipline was a hallmark of Hanoteau’sphilological positivism, equally characterising his earlier linguisticscholarship, and doubtless instilled by his training at the ÉcolePolytechnique (where Auguste Comte once lectured to his corps ofengineers). Its principles of unadorned descriptive precision needed to bebraced against the more romantic historical imagination of Letourneux:

Colonel Hanoteau held definitely, with a determined will, to what might bemaintained as established facts and uncontestable realities, irrespective ofthe past or the historical origins of the peoples he was studying. But M.Letourneux, by contrast, begged release from this harnessing of his brilliantimagination, for he would eagerly construct elaborate theories and systems.Yet his collaborator would have none of that, preferring to disencumberhimself at once of all wild conjectures, whose duration tends to beephemeral and whose scientific value would always be contestable. (M.Hanoteau 1923: 148–49)

In the last year of their collaboration, Letourneux was thus resigned torenouncing historicism. In a letter of February 1868 he declared being‘done with all those theories that in Kabyle law are never easy todisentangle, let alone to express clearly; from now on, it shall be enoughfor me simply to state the positive facts as these are actually supportedby the [qanun] texts’ (Bernard and Milliot 1933: 11).

In fact, the idea of using the 1804 Civil Code as a translational gridfor the ethnography was already being pursued by Hanoteau evenbefore his collaboration with Letourneux. Maurice Hanoteau recalledhis father’s studious examinations of French national law as a meansof specifying and classifying Kabyle folk notions of legal trust, contract,fraud, tort and so on (1923: 143f.). Comparison of Kabyle customarylaw with Maliki fiqh jurisprudence – as rendered in the Mukhtasarcommentaries of the medieval Egyptian jurist Khalil ibn Ishaq (see1848–52 edition) – was also being pursued by Hanoteau with Si Mulabefore Letourneux was co-opted into the project (Hannemann 2002:90f., 170ff.). Its anthropological peculiarity primarily consists in thispersistent dialogical triangulation of the divergent yet often congruentperspectives of modern western legislation and of classical Islamicjurisprudence with respect to Kabyle tribal custom.

Yet Letourneux assuredly made original contributions of his own,particularly in composing the Third Section on Criminal Law, whichwas his acknowledged responsibility in the final year of their


collaboration. Here the Code pénal proved far less serviceable than hadthe Code civil for family law. The plan of this section (III: 59) wastherefore more closely adapted to indigenous schemes of qanun penaltariffs. Of outstanding originality is the cumulative elucidation of apolyvalent Kabyle notion: that of ‘anaya or extended ‘protection’ againstviolations of physical or moral integrity (III: 77–83, 107–11, cf.Daumas and Fabar 1847: 70–75, Mahé 2001: 106–17). Hanoteau’sethnographic sketch had indicated a shifting range of such ‘protections’extended to vulnerable individuals by notable leaders and their kingroups, by sof factions, or by entire tribes and confederacies (II: 61ff.).Breaches of proclaimed ‘anaya protection were considered heinousoffences, which justified righteous homicide, for its respected value wasa precise index of the moral credit of its collective defenders. Hanoteauhad even eulogised the heroic qualities of self-sacrifice it demanded:

However it may operate, one can scarcely deny the moral grandeur of thisinstitution of ‘anaya protection; it is an original form of mutual assistance,even stretched to the point of self-abnegation; and the heroic acts it inspiresconstitute the greatest honour of Kabyles – even if the necessity of suchdevotion signifies an ill-developed societal state, where the individual isobliged to step in on behalf of the law in protecting persons … (II: 63)

Treatment of anaya in the final part of the monograph, however, showsthat this ‘protection’ was by no means restricted to personal orinterpersonal associations. Rather, all rights of personal redress wereultimately sanctioned by the overarching ‘anaya protection assumed bya village community through the popular sovereignty of its jama‘aassembly. Any infringement of its protective public order was thus amoral affront to the community’s corporate renown (hurma), whosescales of violation are precisely gauged in qanun penal tariffs. Inelucidating this complex moral notion, Letourneux acutely draws on aseries of case studies, otherwise all too rarely employed as evidence. Asone might expect, many cases of violent raqba feud or disputes overhurma honour were derived from the Ait Iraten around Fort-Napoléon(III: 104f., 110), and they are often ascribed to the finely narratedtestimony of Si Mula n Ait u ‘Amar (III: 191f., 303f.).

Critical counterpoints

We see that La Kabylie et les coutumes kabyles was a curious chimaera oftextual philology and field anthropology: an epigraphic ethnography. Itwas also an applied ethnography, even a work of indigenous advocacy; forwhile it archived a vanished era of former political autonomy in Kabylia,

58 Peter Parkes

it was ultimately intended to inform its future devolved administration.Hanoteau’s introduction to the second edition of the monograph evenrecalled his original hopes that it might underpin a constitutionalcodification of pan-Kabyle customary law (I: viii).18 The monographtherefore had prescriptive as well as descriptive pretensions, which mayexplain its carefully cantilevered evaluations: its critical counterpoints.

A recurrent antithesis is thus established between an admired ‘spiritof solidarity’ invigorating Kabyle communities and the ‘selfish egoisms’of modern Europe (II: 1, III: 296). But this congenial contrast is alwaysquickly qualified by contrary recognition of the violent concomitants ofungoverned democracy, where ‘a spirit of equality, taken to extremes,incites insatiable envies …’ (II: 2). Admirable solidarities of patriarchalkinship are similarly counterposed with an equally rhetoricaldenunciation of the miserable marital status of Kabyle bartered brides(II: 148f.). Hanoteau here declared his concern to correct overindulgentportrayals of Kabylia as a primitive paradise of republican virtues: ‘Asto the status of Kabyle women … one must be disabused of errorsrelayed by the brilliant paradoxes of our eminent writers’, whodoubtless included some tactfully unnamed colleagues.19

Disparagement of tribal disrespect for women’s Koranic rights ofendowment, however, surely echoed the marabout opinion of Si Mula.

A copious correspondence between Hanoteau and Letourneuxshows that both authors acutely apprehended the polemical dangers ofoverly romanticised depictions of Kabyle cultural affinities withsentimental values of French popular republicanism: what Charles-Robert Ageron would later characterise as le mythe kabyle (1960, 1968:268–77).20 Their sober perspectives rather exemplified the pluralist andpaternalist policies of Marshal Randon’s original programme of indirectrule in Kabylia. This had endorsed its traditions of village governmentwith a view to anticipated assimilation of Kabyles to voluntary Frenchcitizenship. Qanun rulings might then be expected to converge with civiljurisdiction, cultivated by a mission civilisatrice of appropriate educationand technically assisted development. Hanoteau, for example, longcampaigned for a training college in industrial arts and crafts (to beinstructed by officers seconded from his own corps of engineers), whichhe eventually established at Fort-Napoléon in 1866.21

This broader developmental programme provides a necessary contextfor evaluating the monograph. Its ideological underpinnings lay in theindustrial-socialist doctrines of Saint-Simon, instilled among manybureaux arabes officers trained at the École Polytechnique, but propagatedin Algiers by the redoubtable Ismail Urbain, the imperial adviser onindigenous affairs under the Second Empire.22 For Urbain, who eagerlypromoted Hanoteau’s project, recognition of Kabyle customary law


promised to vindicate his own boldly envisaged plans for a culturally andlegally pluralistic assimilation of all Muslim indigenes, L’Algérie pour lesAlgériens (1860). Hanoteau and Letourneux may have been quizzical ofUrbain’s passionate indigenophilia; but they shared his liberalpreoccupation with indigenous advocacy, pitted against the rapaciousencroachments of colon settlers then lobbying for civilian administration.23

The monograph’s treatment of Kabyle Muslim religion is alsoindicative of this truly applied anthropologie positive. A concise statementin the first volume establishes that all Kabyles were unquestionablydevout and orthodox Muslims (I: 380–84). Hanoteau was determined toquash popular misconceptions that Islam was but a thin veneer ofassumed faith imposed on secular or pagan cultural roots, renderingKabylia susceptible to easy laicisation or Christian evangelisation.24 Inthe second volume, however, more critical perspectives are advanced (II:83–105). These concern the role of marabout religious specialists(imrabden), insinuated as elementary teachers and adjudicators amongKabyles, whose spiritual influence was waning under Frenchadministration. Far more dangerous, therefore, were expansive religiousconfraternities (ikhwan) that had begun to absorb or replace thisdisaffected local clerisy. Such was the Rahmaniyya Sufi order in Kabylia,a pietist movement of moral reform that galvanised tribal insurgenciesby declaring jihad or holy war against infidel foreign occupation.25

Hanoteau’s premonitions of impending discontent among suchreligious leaders again attests to the vital collaborative contribution ofSi Mula, whose reports on his own Ait ‘Amar marabout lineage are citedas apt illustrations. They indicate a hereditary line of conservativereligious scholars, scornful of mountbank marabouts, but more waryof the reformist pretensions of Rahmaniyya spiritual devotees, whothreatened their own established position as tribal intermediaries withexternal powers. Formerly aligned with Turkish Beylik rulers, theclerical family of Si Mula had made overtures to the French militaryauthorities in the 1840s, ultimately rewarded by Si Mula’s appointmentas khoja interpreter at Fort-Napoléon. Unlike devotees of Rahmaniyyamoral reform – who would advocate the replacement of tribal customsby shari‘a law – traditional marabout dignitaries such as Si Mula weremore inclined to condone customary usages in the interests ofmaintaining tribal peace, albeit also deliberating on their acceptable orunacceptable divergence from orthodox Maliki Sunni Muslim law. Suchtolerant conservatism conveniently dovetailed with the gradualistdevelopmental programme envisaged by Hanoteau.

The whole project of redacting village decrees as approved tribalcanons may have been Si Mula’s ambition all along, initiated in hiscalligraphic gift to Hanoteau (see Fig. 2.3) ten years earlier. It is

60 Peter Parkes

therefore tempting to interpret many of the equilibrated opinions of themonograph as further ventrilocutions of Si Mula: a scholarly Arabicvoice resounding through two French authors, who may have onlypartially comprehended its contrapuntal techniques of dialecticalreasoning and justification for Kabyle customary usage, employingclassic rhetorical methods of Islamic legal science (‘ilm i-qalam). Theseare apparent in Si Mula’s subsequent letter to Hanoteau (Bernard andMilliot 1933: 27–30), which also displays an adept deployment of citedreferences to the medieval legal commentaries of Khalil ibn Ishaq(Hannemann 2002: 90–93). But such informed juristic commentariesare evident throughout the monograph: their analytical yield would becumulatively harvested by subsequent scholars of Berber customarylaw throughout the Maghrib.26

Colonial transposition

The colonial deployment of the monograph would not be what any ofits authors could have wished. For its composition during the 1860scoincided with the last decade of protective military administration inKabylia, which collapsed with the fall of the Second Empire in 1870.Defeat at Sedan and a Republican colonist uprising in Algiers, matchingthe Paris Commune of 1871, then catalysed a final Kabyle insurrection,followed by punitive colonial reprisals and civilian reforms. These arerecorded in painful detail by General Hanoteau in a long appendix tothe second edition of La Kabylie (III: 455–514) written after hisretirement to France:

In truth, the ancient institutions of Kabylia were shattered when our armycolumns triumphed over the insurrection of 1871 … In a few years, theentire civil edifice of traditional liberties, which had resisted armies ofconquerors over thousands of years, simply collapsed. Its ruin is nowcomplete, and it is not without regret for a past era that lacked neithergrandeur nor glory, that I inscribe an epitaph adapted from the history ofanother great lost nation: finis Kabiliae! (III: 462)

As Hanoteau had gloomily predicted, fears of impending civilianadministration in Algeria incited revolts in its eastern provinces, takenup as a last-ditch tribal jihad by the Rahmaniyya order in Kabylia.27

Hanoteau was obliged to assist in a merciless suppression of Kabyleinsurgents in the early summer of 1871, when hundreds of Rahmaniyyadevotees were massacred in hopeless assaults on Fort-Napoléon. SiMula and Si Lunis vainly tried to secure a treaty among Ait Iraten toabstain from this insurrection; but the two brothers were besieged with


French officers in the fort, accompanied by pupils they had saved fromHanoteau’s new college of arts and crafts, which was set ablaze byangry rebels, as was their house at Tamazirt. Their loyalty was brieflyrewarded: Si Lunis was later appointed tribal chief (amin al umana) ofAit Iraten, while Si Mula was confirmed as tribal judge (qadi) over anewly created administrative district of Fort-Napoléon, now renamedFort-National.28 But a punitive war reparation of ten million francs wasimposed on the district, allowing civilian colonists to sequester hugetracts of its indigenous territory (III: 333, Ageron 1968: 24–36, Mahé2001: 212ff.).

The new Governor-General, Admiral de Gueydon, appears to havebeen persuaded by Hanoteau and Letourneux to reverse an immediateplan to revoke all of Randon’s immunities of indirect rule in Kabylia. Forwhen the monograph was published in 1873, de Gueydon abandonedhis original intention to apply full French national law throughoutJurjura, conceding the benefits of retaining its village jama‘a assemblies,whose customary laws appeared as opportune bulwarks against furtherantagonistic Islamisation (Ageron 1968: 282). Letourneux was evenasked to devise a simplified qanun codification for civilian administrators(a commission he wryly declined). The new district of Fort-Nationalwould be temporarily retained as an ‘indigenous commune’ undercontinuing military protection. But by 1880 almost all of Kabylia hadbecome civilian territory under the jurisdiction of French civilmagistrates (juges de paix), who struggled to adjudicate its customarylaws with often baffled reference to the monograph (Ageron 1968: 284).Their courts, which denied normal appeal to shari‘a law, would be largelyignored by Kabyles, whose village affairs continued to be regulated byclandestine jama‘a assemblies well into the twentieth century.29 By thenthe imperial dream of indigenous assimilation had retrenched into thenotorious colonial doctrine of separatist association (Betts 1961),reinforced by racism that tarnished and deformed all extantethnography (Lorcin 1995).

Canonical ethnography

La Kabylie et les coutumes kabyles was the original prototype of whatwould shortly become a prolific genre of colonial ethnographicdocumentation: the tribal coutumier. By the end of the nineteenthcentury, there would be many comparable collations of customary lawcompiled by colonial officers operating on the tribal frontiers of BritishIndia, Imperial Russia and the Dutch East Indies, as by all European powers in sub-Saharan Africa.30 These dense ethnographies,

62 Peter Parkes

which underpinned a burgeoning new discipline of ethnologicaljurisprudence, are scarcely appreciated by anthropologists nowadays.Professional criticism, voiced ever since Malinowski (1926), highlightstheir neglect of quotidian disputation as a negotiable matrix of socialorder distinct from stipulated legislation. Colonial compilations ofcustomary law are thus distrusted or dismissed as artefactual‘fabrications’ of alien administrative convenience, if not darklydisparaged as collusive and divisive tools of imperial hegemonies.31

It seems easy to address such criticism to the foundational work ofHanoteau and Letourneux. Bernard and Milliot (1933: 22–25) alreadycast worrying doubts about the distorting effects of Hanoteau’scommissioning of qanun transcripts, which may have solicited aselective submission of ‘normalised’ regional exemplars, incipientlycodified even before their analytical synthesis within the monograph(cf. Milliot 1932: 141ff.). Georges-Henri Bousquet aimed a moreflamboyant diatribe, expressed with jocular pugnacity, at what hepilloried as ‘A cult to be destroyed: the adoration of Hanoteau andLetourneux’ (1950c). Pitched against the colonial canonisation of themonograph, Bousquet noted its stifling impact on subsequentethnography, when literate Kabyle informants relied on the dead wordof Hanoteau and Letourneux, even if their living practice contradictedits sanctified account. Bousquet (1950b) thence elaborated an originaltheory of habituated jural transactions, counterposed with thenormative regulations emphasised by Hanoteau and Letourneux,which would be famously expanded by Pierre Bourdieu. Yet both Milliotand Bousquet ultimately acknowledged their profound debts to themonograph, which crucially guided their own pioneering scholarshipon Islamic juridical practice and customary law in the Maghrib.

The broader anthropological legacy of the monograph remainsdifficult to assess, although a line of inspired descent to classicethnographies of the Maghrib is easy to trace.32 Hanoteau’s project ofencouraging a constitutional recognition of customary law would beadopted more concertedly (and controversially) in the neighbouringprotectorate of Morocco, stimulating further compilations of Berber legalcustom there.33 Their monograph inspired similar collections ofcustomary law in French West Africa and Indo-China; and it wasevidently known to Snouck Hurgronje and Cornelis van Vollenhoven,feasibly instigating the Adatrecht school of Leiden and Batavia in theDutch East Indies.34 It would be discussed in almost all major works ofethnological jurisprudence at the end of the nineteenth century,featuring prominently in socialistic literature on agrariancommunalism.35 Yet its analytical insights and empirical documentationhave only recently attracted the serious scrutiny they long deserved:notably in two magisterial surveys of Kabyle communality and


customary jurisprudence over a span of two centuries (Mahé 2001;Hannemann 2002), and in current ethnography (Scheele 2008, 2009)that traces both continuities and transformations in Kabyle villageorganisation and communal legislation down to presently troubled times.

A more occluded legacy of such early ethnographies lies in covertreactions to their former authority in later academic anthropology. Formodern fieldwork was effectively ‘reinvented’ from the 1920s – bydisciples of Malinowski and Mauss – in upstart denial of its establishedprecedents (Blanckaert 1996). This subversive reaction is well displayedin subsequent ethnographies of Kabylia, such as the avowedlyMaussian analyses of René Maunier (1927, 1935) and Jeanne Favret(1968), and especially of Pierre Bourdieu (1972), who all adroitlyreworked the rich heritage of Hanoteau and Letourneux while subtlydemoting its legitimacy.

The apotheosis of this last ‘canonical transposition’ of themonograph occurs in Bourdieu’s renowned theory of practice. Thisdenied almost any regulatory role for the normative order ofconsensual legislation that Hanoteau and Letourneux haddocumented, let alone its nuanced dialogues with Maliki Muslim lawnoted by Si Mula. Kabyle social order seems to emerge as a spontaneousoutplay of competitive transactions, arising from a dispositionalcalculus of patrimonial power, whose Hobbesian brutality is barelymollified in its ritual idiom of honour and prestige. The reactive impetusof this canonoclastic vision of anomic sociality, within the congestedterrain of Kabyle ethnography, is evident:

When I began working as an ethnologist, I wanted to react against what Icalled legalism, against the tendency among ethnologists to describe thesocial world in the language of rules … I managed to show that in the caseof Kabylia the most codified, namely customary law, is only the recording ofsuccessively produced verdicts in relation to individual transgressions, basedon the principle of the habitus. (Bourdieu 1990: 76f. [1987: 94f.])

Bourdieu’s anticanonical struggle with Hanoteau and Letourneux hasbeen of great provocative value to social theory, even if one may queryhis bold pretension to have deciphered a deep transformationalgrammar of Kabyle customary legislation: viz. ‘one can re-generate allthe concrete acts of jurisprudence that are recorded in customary lawson the basis of a small number of simple principles’ (ibid., cf. Bourdieu1977: 16f. [1972: 209f.]). Such principles were what Hanoteau andLetourneux endeavoured to delineate: yet scarcely as ‘mythico-ritual’dichotomies (night/day, inside/outside, male/female), but rather asmorally coherent rulings about communal order, underpinned by thecomplex jural notion of ‘anaya protection assumed by the popular

Figure 2.4. Submission of the Kabyle tribes to Marshall Randon in 1857 (ViolletCollection).

sovereignty of jama‘a village assemblies. These were instituted toproclaim qanun ordinances regulating social conduct according toknown customary usages (‘urf, ‘ada), in variably defensibleaccommodation with Maliki Muslim jurisprudence deliberated bymarabout clerics such as Si Mula (II: 136ff.). The habitual socialpractice and tactical manipulation of such usages were surelysignificant (and indeed not disregarded by Hanoteau and Letourneux),but so were the regulatory effects of their consensual legislation – theircanonical rules, no less.36

This essay has simply narrated an extraordinary collaborativeventure of early French ethnographic fieldwork, reconstructing theconsonant interests and skills of three oddly combined contributors –a soldier, a magistrate and a marabout – as well as the peculiarcircumstances of their cooperation in Kabylia during the last decade ofthe Second Empire. For all its imperial arrogance and its unwantedcolonial consequence, La Kabylie et les coutumes kabyles should surelybe redeemed by anthropology as an exemplary ethnography ofcustomary law-making and communal regulation, evincing a trulyreflexive conscience commune. Despite its methodological reticence, it

64 Peter Parkes

Figure 2.5. Hanoteau and two Tuareg, 1858. From Jacques Frémeaux, Les bureauxarabes dans l’Algérie de la conquête, Paris, Denoël 1993.

pioneered a painstaking hermeneutics of concerted textual and oralexegeses of tribal jurisprudence. Although defeated by adverse forcesof colonial administration, its positive plans to enable a devolution ofindigenous jurisdiction were not ignoble (nor impertinent to currentnational imperatives in Algeria that still have to accommodate regionalregimes of self-government). Its dedicated project of archivaldocumentation and elicited explication – informed by an intimate andengaged participation of indigenous experts – still begs to be emulatedand elaborated today.37


66 Peter Parkes

Notes

Acknowledgements

I am indebted to Judith Scheele for generous encouragement, comments and corrections.This essay also relies substantially on the classic work of Charles-Robert Ageron for thecolonial context, and on the specialist current scholarship of Alain Mahé and TilmanHannemann, as cited.

1. The Jurjura mountains of Kabylia form a steep massif behind the coastline ofnortheast Algeria. Tribal insurgencies in these highlands culminated in their militaryoccupation in 1857 after fierce campaigns under Marshal Randon. See Julien (1964)and Ageron (1966).

i2. Kabyle qanun is derived from Greek kanon ‘rule, law’ (via Arabic and Turkish), cognatewith English (and French) canon – hence canonical ethnography in several sensesexplored here. On the Arabic term, see Linant de Bellefonds (1978: 566f.). On thecanonical status of the monograph, see especially Berque (1956: 305ff.).

i3. Translated passages are from the second (1893) edition of La Kabylie et les coutumesKabyles, referenced by volume and page number alone in this chapter. I have simplifiedHanoteau’s francophone orthography of Kabyle Berber (taqbaylit) and Arabic terms.

4. For a general biography of Hanoteau’s career, see Poussereau (1931). On thefollowing narrative account, compare Hannemann (2002: 80–93, 2003: li–lix).

5. On the bureaux arabes, see Ageron (1960, 1966), Julien (1964: 333–41), Perkins(1981), and Frémaux (1993); also Lorcin (1995: 79–85, 130–40). A broaderbackground of their policies is summarised by Ageron (1991: Ch. 3).

6. On Hanoteau’s Poésies populaires de la Kabylie du Djurjura (1867), see Goodman(2002a, b). This foundational work of ethnopoetics also appears to have been largelycompiled and transcribed by Hanoteau’s primary informant, Si Mula n Ait u ‘Amar(Hanoteau 1867: xii).

7. On Randon’s ‘indirect rule’ or l’organisation Kabyle, see Ageron (1968: 277f.) andFrémaux (1993: 52f.). Qanun rulings, noted by Daumas (1853: 227f.), were alsopublished by Féraud (1862: 276, 1863: 67) and Aucapitaine (1863, 1864: 71–76).

8. These details are briefly mentioned in the monograph (II: 92). The family history ofSi Mula is related in a long letter sent to Maurice Hanoteau by Si Mula’s son, Si Sultanben Si Mula, reproduced in Bernard and Milliot (1933: 5–7).

9. The quote is from an 1860 report by Hanoteau (Hannemann 2002: 62). Earlierreports were publicised as extracts in the Revue Algérienne et Coloniale (Anonymous1859), foreshadowing the ethnographic synopsis of the monograph (II: 1–134).

10. On Ismail Urbain and his circle, see below (n. 22), Ageron (1968: 397–414) andLorcin (1995: 88–92).

11. The monograph would thereby crown thirty-nine volumes of the Explorationscientifique de l’Algérie, commissioned by a founder of the bureaux arabes, Pellisier deReynaud, in the 1840s. Its first volume summarises their exploratory surveys ofKabylia, especially Carette (1848).

12. On Letourneux, see the affectionate sketch by M. Hanoteau (1923: 144f.); alsoBernard and Milliot (1933: 10–12). A field botanist and malacologist of worldrenown, Letourneux’s discoveries of many new species in Kabylia are itemised (withKabyle and Arabic nomenclature) in the first volume of the monograph (I: 49–234).A keen epigrapher of Romano-Berber inscriptions, Letourneux succeeded Berbruggeras president of the Historical Society of Algeria in 1873–76.


13. Bernard and Milliot (1933: 33, Pl. I). This letter may have been intended to providesupplementary material for the second edition of the monograph, although it is notused there. Hannemann (2002: 89) surmised that its date in Arabic might be a scribalerror for January 1868, but this seems unlikely since Hanoteau would then have beenresident at Fort-Napléon together with Si Mula.

14. On the composition of qanun rulings, see Masqueray (1886: 57–72) and Hannemann(2002: Ch. 5). The form of oral qanun rulings in the Kabyle language is uncertain:transcripts made for Hanoteau (1858a: 324–38, Bernard and Milliot 1933: Pl. II)may be back-translations from Arabic transcribed by his interpreter at Dra el-Mizan,El Haj Sa’id u ‘Ali (Bernard and Milliot 1933: 8). The style of these decrees, itemisingfines for offences, resembles those documented in rural communities throughout theMediterranean: from Spain (Behar 1986: Ch. 7) to Albania (Durham 1928: 64–92,Hasluck 1954: 261–74, Gjeçov 1989) and the Caucasus (Leontovich 1882, Kemper2004). See also Scheele (forthcoming).

15. Cases of Kabyle disputation are, however, summarised in reports of the bureaux arabesat Fort-Napoléon (Gouvernement Géneral de l’Algerie n.d.), which might be collatedwith the monograph. Cf. Hannemann (2003: xxxv–l), Perkins (1981: 67–75).

16. An often cited source on early Germanic law is Jules Michelet’s Origines du droitfrançais (1837). Such comparisons had already been suggested in Hanoteau (1858b)and Aucapitaine (1863); they would be elaborated by Masqueray (1886).

17. Kabyle regulations of commerce were pertinent to their extensive engagement inpetty trade, outlined in the first volume of the monograph (I: 498–508). Cf. Mahé(2001: 29–39).

18. Letourneux also wrote excitedly to Hanoteau of this prospect: ‘a unified and codifiedKabyle coutume will be accepted by a million Berbers, even by French magistrates; ouridea shall make its way …’ (letter of 1869 in Bernard and Milliot 1933: 25).

19. E.g. Daumas (1853) and Aucapitaine (1864), who had eulogised Kabyle matriarchy.Cf. also Hanoteau (1867: 287–94). The position of Kabyle women has always beencontested (Lorcin 1995: 64–67): their brideprice would be similarly deprecated byHacoun-Campredon (1921), Morand (1927) and Lefèvre (1939).

20. On this Kabyle myth, contrasting assimilable Berbers with disparaged Arabs, cf. Lazreg(1983), Lorcin (1995) and Mahé (2001: 147–57).

21. Ageron (1968: 323f.). On other technical and agricultural innovations introduced bybureaux arabes officers, see Perkins (1981: 131–48) and Lorcin (1995: 83). Hanoteau alsoencouraged Si Mula and Si Lunis to establish the first secular primary school in Kabylia,which opened at Tamazirt in 1873 (Ageron 1968: 333). His elaborate plans for a creditassociation to support Kabyle enterprise are detailed by Poussereau (1931: 80–82).

22. On Saint-Simonism in Algeria, see Emerit (1941) and Lorcin (1995: Ch. 5). On thecolourful career of Ismail Urbain and his campaigns for indigenous Muslim rights ofcitizenship, see Ageron (1968: 397–414) and Levallois (1989, 2001).

23. On this conflict between civilian and military authorities, see Ageron (1960, 1966,1968) and Lorcin (1995: Ch. 4). Hanoteau’s anti-colon sentiments were stronglystated: ‘What our settlers dream of is a bourgeois feudalism, in which they will be thelords and the natives their serfs’ (Ageron 1991: 39f.).

24. On Hanoteau’s prolonged battles against Christian missionaries in Kabylia, notablyArchbishop Lavigerie, see Ageron (1968: 273f., 279f.) and Mahé (2001: 180–82).

25. On these religious orders, Hanoteau relied on an earlier study of de Neveu (1845),although his disparaging account of Rahmaniyya education was based on a visit totheir ma‘mara college (II: 91–95). On the Rahmaniyya order in Kabylia, see Clancy-Smith (1994: 39–45) and Mahé (2001: 46–54, 193–99) citing Salhi (1979).

26. Notably Milliot (1932), Marcy (1939, 1949), Berque (1953, 1955) and Bousquet(1950b, 1956). See now Hannemann (2002, 2005).

68 Peter Parkes

27. The Moqrani insurrection in Kabylia was documented by Rinn (1891) and Robin(1902). Cf. Julien (1964: 453–500), Ageron (1968: 3–7) and Mahé (2001: 190–99). The following details on Fort-Napoléon are mainly derived from the account ofSi Mula’s son, Si Sultan ben Si Mula (Bernard and Milliot 1933: 6f.).

28. They were soon divested of these posts under Camille Sabatier’s maladroit regime atFort-National in the early 1880s (Ageron 1968: 287 n.1, Mahé 2001: 250f.). Si Mulathen transferred his services as a qadi judge to a neighbouring district. He died inAlgiers in 1890 (Bernard and Milliot 1933: 7), as did Letourneux. Close relationspersisted thereafter between the lycée-educated sons of Si Mula and General MauriceHanoteau, who followed his father’s career as a military engineer in Algeria.

29. On administrative reforms in the 1880s, including bungled attempts to restore amodified form of indirect rule around Fort-National by Camille Sabatier, see Ageron(1968: 285–92) and Mahé (2001: 245–51). Qanun codes up to the 1920s (somewritten in French) were documented by Milliot (1926) and Bousquet (1936, 1949).

30. Similar compendia of tribal-Muslim adat laws were C.L. Tupper’s Punjab CustomaryLaws (1881) and F.I. Leontovich’s Adaty Kavkazskikh Gortsev (1882–3).

31. On colonial ‘fabrications’ of customary law, see Snyder (1981), Chanock (1985),Moore (1986) and Salemink (1991); for darker Foucauldian suspicions, see Mamdani(1996), Burns (2004) and Le Cour Grandmaison (2006). On comparativecodifications of customary law a still useful collection is Gilissen (1962).

32. Via Masqueray (1886) to Montagne (1930) and Berque (1955). On Masqueray’sethnography, following respectfully in the footsteps of Hanoteau and Letourneux, seeColonna (1983) and Ould-Braham (1996). On this distinct ethnographic lineage, seeBerque (1956: 305ff.), Favret (1968: 18f.), Roberts (2002), and Mahé (2003: ii).

33. See Burke (1973: 189–99), also treating the controversial dahir berbère decree of1930. General Lyautey’s programme of collating Berber izerf customary law in theMiddle Atlas from 1913 was exactly modelled on Hanoteau’s earlier project (Surdon1938: 105ff., cf. Bousquet 1950a and Ageron 1971). See now Kraus (2005) andBurke (2007).

34. On Kabyle models in French West African judicial policy, see Christelow (1982: 14ff.)and compare Sibeud (2002) and Shinar (2006). On coutumiers in Indo-China, seeSalemink (1991) and Parkin (2005: 201); on the Leiden Adatrecht school, see TerHaar (1948) and Holleman (1981).

35. La Kabylie is cited in classic compendia on comparative customary law by A.H. Post,J. Kohler and S.R. Steinmetz. For socialistic references, see Laveleye (1874: Ch. 5, withnote 3), Kovalevsky (1879: 200ff.) and Kropotkin (1902: Ch. 4, notes 33–35). Muchhas been made of its citation (with Masqueray 1886) in Durkheim’s De la division dutravail social (Ch. 6 with note 6, see Gellner 1985); but there is little evidence that itwas seriously studied by Durkheim (see Roberts 2002: 117ff.).

36. On Bourdieu’s antinomian problems with such rules, see Just (2005); for his deepfamiliarity with the monograph, compare Bourdieu (1958: Ch. 2). The delegitimationof Kabyle village assemblies at the end of the nineteenth century had discernibleconsequences of anomic anarchy and notorious banditry (Mahé 2001: 215ff.). Seealso Mahé (2000) and Scheele (2008, 2009) on recent struggles to revive villagelegislative powers in Kabylia.

37. An urgent archival ethnography of customary law throughout the MuslimMediterranean has long been advocated by Frank Stewart (1987, 1989–90; cf. alsoDresch 2007. Recent collections (e.g. Dostal and Kraus 2005, Kemper and Reinowski2005) point to a resurgent anthropology of legal pluralism in Islam, which Hanoteauand Si Mula pioneered.


References

Ageron, C.-R. 1960. La France a-t-elle eu une politique kabyle? Revue historique, 226:311–51.

——— 1966. La politique kabyle sous le Second Empire, Revue française d’histoire d’Outre-Mer, 53: 67–105.

——— 1968. Les Algériens musulmans et la France (1871–1919), Paris: PressesUniversitaires de France (2 vols).

——— 1971. La politique berbère du protectorat marocain, Revue d’histoire moderne etcontemporaine, 18: 50–90.

——— 1991 [French edn. 1964]. Modern Algeria: a history from 1830 to the present (tr.M. Brett), London: Hurst.

Anonymous [after A. Hanoteau] 1859. Constitution sociale de la Kabylie: extrait d’unrapport du général commandant la division territoriale d’Algér, Revue Algérienne etColoniale, 1: 1–20.

Aucapitaine, H. 1863. Kanoun du village de Thaourirt Amokran chez les Aith Iraten(Kabilie), Revue africaine, 7: 279–85.

——— 1864. Études sur le passé et l’avenir des Kabyles: les Kabyles et la colonisation del’Algérie, Paris: Challamel.

Behar, R. 1986. Santa María del Monte: the presence of the past in a Spanish village,Princeton: Princeton University Press.

Bernard, A. 1930. Histoire des colonies françaises et de l’expansion de la France dans le monde,in G. Hanotaux and A. Martineau (eds), 2: L’Algérie, Paris: Plon.

Bernard, A. and L. Milliot. 1933. Les qânoûns kabyles dans l’ouvrage de Hanoteau etLetourneux, Revue des Études Islamiques, 7: 1–44.

Berque, J. 1953. Problèmes initiaux de la sociologie juridique en Afrique du Nord, StudiaIslamica, 1: 137–62.

——— 1955. Structures sociales du Haut Atlas, Paris: Presses Universitaires de France.——— 1956. Cent vingt-cinq ans de sociologie maghrébine, Annales ESC, 11: 299–331.Betts, R.F. 1961. Assimilation and association in French colonial theory, 1890–1914, New

York: Columbia University Press.Blanckaert, C. 1996. Histoires du terrain: entre savoirs et savoir-faire, in Le Terrain des

sciences humaines (XVIIIe–XXe siècle), Paris: L’Harmattan.Bourdieu, P. 1958. Sociologie de l’Algérie, Paris: Presses Universitaires de France.——— 1977 [1972]. Outline of a Theory of Practice, tr. R. Nice, Cambridge: Cambridge

University Press. [Esquisse d’une théorie de la pratique, Geneva: Librairie Droz.]——— 1990 [1987]. In other words, tr. M. Adamson, Cambridge: Polity Press. [La

codification, In Choses dites. Paris: Les Éditions de Minuit.]Bousquet, G.H. 1936. Un qanoun kabyle contemporain, Revue africaine, 79: 867–72. ——— 1949. Documents contemporains curieux relatifs au droit en Kabylie, Revue

algérienne, tunisienne et marocaine de législation et de jurisprudence, 1: 93–96.——— 1950a. Islamic law and customary law in French North Africa, Journal of the

Society for Comparative Legislation, 32: 57–65.——— 1950b. Justice française et coutumes kabyles, Algiers: Imprimerie Nord Africaine.——— 1950c. Un culte à détruire: l’adoration de Hanoteau et Letourneux, Revue de la

Méditerranée, 8–9: 441–54.——— 1956. Le droit coutumier de Aït Haddidou, des Assif Melaoul et des Isselaten

(Conféderation des Aït Yafelmane): notes et reflections, Annales de l’institut d’étudesorientales de l’université d’Alger, 14: 113–230.

Burke, E. III. 1973. The image of the Moroccan state in French ethnological literature: anew look at the origins of Lyautey’s Berber policy, in E. Gellner and C. Micaud (eds),Arabs and Berbers, London: Duckworth.

70 Peter Parkes

——— 2007. The creation of the Moroccan colonial archives, 1880–1930, History andAnthropology, 18: 1–9.

Burns, P. 2004. The Leiden legacy: concepts of law in Indonesia, Leiden: KITLV.Carette, E. 1848. Études sur la Kabilie proprement dite, Paris: Imprimerie Royale (2 vols).Chanock, M. 1985. Law, custom and social order, Cambridge: Cambridge University Press.Christelow, A. 1982. The Muslim judge and municipal politics in colonial Algeria and

Senegal, Comparative Studies in Society and History, 24: 3–24.Clancy-Smith, J.A. 1994. Rebel and saint: Muslim notables, populist protest, colonial

encounters, Berkeley: University of California Press.Colonna, F. 1983. Présentation: introduction to the reprint of É. Masqueray [1886]

Formation des cités chez les populations sédentaires de l’Algérie: Kabyles du Djurdjura,Chaouïa de l’Aurâs, Beni Mazab, Aix-en-Provence: Édisud.

Daumas, E. 1853. Moeurs et coutumes de l’Algerie, Paris: Hachette.Daumas, E. and P. Fabar 1847. La Grande Kabylie, études historiques, Paris: Hachette.Dostal, W. and W. Kraus (eds). 2005. Shattering tradition: custom, law and the individual in

the Muslim Mediterranean, London: I.B. Tauris.Dresch, P. 2007. The Rules of Barat: tribal documents from Yemen, Sanaa: Centre Français

d’Archaeologie et de Sciences Sociales.Durham, M.E. 1928. Some tribal origins, laws and customs of the Balkans, London: George

Allen and Unwin.Emerit, M. 1941. Les Saints Simoniens en Algérie, Paris: Belles Lettres.Favret, J. 1968. Relations de dépendance et manipulation de la violence en Kabylie,

L’Homme, 8: 18–44.Féraud, L.-C. 1862, 1863. Moeurs et coutumes kabyles, Revue africaine, 6: 272–83, 429–

41; 7: 67–84.Frémaux, J. 1993. Les bureaux arabes dans l’Algérie de la conquête, Paris: Denoël.Gellner, E. 1985. The roots of cohesion, Man (n.s.) 20: 142–55.Gilissen, J. (ed.). 1962. La rédaction des coutumes dans le passé et dans le présent, Brussels:

Université Libre de Bruxelles, Études d’histoire et d’ethnologie juridique 3.Gjeçov, S. 1989. Kanuni i Lekë Dukagjinit (tr. L. Fox), New York: Gjonlekaj.Goodman, J.E. 2002a. The half-life of texts: poetry, politics and ethnography in Kabylia,

Algeria, Journal of Linguistic Anthropology, 12: 157–88.——— 2002b. Writing empire, underwriting nation: discursive histories of Kabyle Berber

oral texts, American Ethnologist, 29: 86–122.Gouvernement Géneral de l’Algérie. n.d. Série I, Bureaux arabes, cercle de Fort-Napoléon

puis Fort-National, rapports, 1857–1880. Ss. 42I 1–7, 43I 1–7. 71 MIOM 288–91,294. Archives d’Outre-mer d’Aix-en-Provence, France.

Hacoun-Campredon, P. 1921. Etude sur l’évolution des coutumes kabyles, Algiers: Carbonel.Hannemann, T. 2002. Recht und Religion in der Grossen Kabylei (18/19 Jahrhundert):

zu rechtskulturellen Wandlungsprozessen des tribalen Gewohnsheitsrechts, PhDdissertation, University of Bremen (internet publication 2004).

——— 2003. La mise en place du droit kabyle dans l’Algérie coloniale (1857–1868).Introduction to the reprint of Hanoteau and Letourneux (1893), Paris: ÉditionsBouchène.

——— 2005. Gewohnheitsrecht in einer islamischen Rechtsumgebung: theoretischeVergleichperspectiven aus der Grossen Kabylei, in M. Kemper and M. Reinkowski(eds), Rechtspluralismus in der islamischen Welt: Gewohnheitsrecht zwischen Staat undGesellschaft, Berlin: de Gruyter.

Hanoteau, A. 1858a. Essai de grammaire kabyle …, Algiers and Paris: Bastide/Challamel.——— [unsigned] 1858b. Une charte kabile, Revue africaine, 3: 75–80.——— 1860. Essai de grammaire de la langue tamachek …, Paris: Imprimerie Impériale.——— 1867. Poésies populaires de la Kabylie du Djurdjura, Paris: Imprimerie Impériale.


Hanoteau, A. and A. Letourneux. 1872–73 [1893]. La Kabylie et les coutumes kabyles.Paris: Imprimerie Nationale (2nd ed.) 1893.

Hanoteau, M. 1923. Quelques souvenirs sur les collaborateurs de ‘La Kabylie et lescoutumes kabyles’, Revue africaine, 64: 134–49.

Hasluck, M. 1954. The unwritten law in Albania, Cambridge: Cambridge University Press.Holleman, J.F. 1981. Van Vollenhoven on Indonesian adat law, The Hague: Martinus Nijhoff.Julien, C.-A. 1964. Histoire de l’Algérie contemporaine: la conquête et les débuts de la

colonisation, Paris: Presses Universitaires de France.Just, R. 2005. In defence of rules: Pierre Bourdieu en Grèce, Journal of Mediterranean

Studies, 15: 1–24.Kemper, M. 2004. Communal agreements (ittifaqat) and ‘adat-books from Daghestani

villages and confederacies (18th–19th centuries), Der Islam, 81: 115–51.Kemper, M. and M. Reinkowski (eds). 2005. Rechtspluralismus in der islamischen Welt:

Gewohnheitsrecht zwischen Staat and Gesellschaft, Berlin: Walter de Gruyter.Khalil ibn Ishaq, al Jundi [Sidi Khalil]. 1848–52. Précis de jurisprudence musulmane, ou

Principes de législation musulmane civile et religieuse, selon le rite Malékite par Khalil,traduit de l’Arabe par M. [Nicolas] Perron, Paris: Exploration scientifique de l’Algérie,sciences, histoires et geographies, vols 10–15.

Kovalevsky, M.M. 1879. Obshchinnoye zemlevladenie, Part 1, Moscow: F.B. Miller.Kraus, W. 2005. Tribal law in the Moroccan High Atlas: pre-colonial legal practice and its

transformations, in W. Dostal and W. Kraus (eds), Shattering tradition, London: I.B. Tauris.Kropotkin, P. 1902. Mutual aid: a factor in evolution, London: Heinemann. Laveleye, E. de. 1874. De la propriété et de ses formes primitives, Paris: Librairie G. Baillière.Lazreg, M. 1983. The reproduction of colonial ideology: the case of the Kabyle Berbers,

Arab Studies Quarterly, 5: 380–95.Le Cour Grandmaison, O. 2006. The exception and the rule: on French colonial law,

Diogenes, 53(4): 34–53. Lefèvre [Bousquet], L. 1939. La femme kabyle, Paris: Sirey.Leontovich, F.I. 1882–3. Adaty kavkazskikh gortsev, Odessa: G. Ulrikha (2 vols).Levallois, M. 1989. Ismayl Urbain: éléments pour une biographie, in M. Morsy (ed.), Les

Saint-Simoniens et l’Orient: vers la modernité, Aix-en-Provence: Édisud.——— 2001. Ismaÿl Urbain: une autre conquête de l’Algérie, Paris: Maisonneuve.Linant de Bellefonds, Y. 1978. Kanun, Encyclopaedia of Islam (2nd ed.), 4: 556–57.Lorcin, P.M.E. 1995. Imperial identities: stereotyping, prejudice, and race in colonial Algeria,

London and New York: Tauris.Luc, B. 1917. Le droit kabyle, Paris: Challamel (2nd ed.). MacCarthy, O. 1847–48. La Kabylie et les Kabyles: études économiques et

ethnographiques, Revue de l’Orient et de l’Algérie 1: 345ff., 2: 28ff., 137ff.Mahé, A. 2000. Les assemblés villageoises dans la Kabylie contemporaine: traditionalisme

par excès de modernité ou modernisme par excès de tradition? Études rurales, 155–156: 179–212.

——— 2001. Histoire de la Grande Kabylie, XIXe–XXe. Siècles: anthropologie historique dulien social dans les communautés villageoises, Paris: Éditions Bouchène.

——— 2003. Entres les moeurs et le droit: les coutumes. Remarques introductives à LaKabylie et les coutumes kabyles, Introduction to the reprint of Hanoteau andLetourneux (1893), Paris: Éditions Bouchène.

Malinowski, B. 1926. Crime and custom in savage society, London: Kegan Paul.Mamdani, M. 1926. Customary law: the theory of decentralized despotism, in Citizen and

subject: contemporary Africa and the legacy of late colonialism, Princeton: PrincetonUniversity Press.

Marcy, G. 1939. Le problème du droit coutumier berbère, La France méditerranéenne etafricaine, 2: 7–70.

72 Peter Parkes

——— 1949. Le droit coutumier zemmour, Algiers: Carbonel.Masqueray, É. 1886. Formation des cités chez les populations sédentaires de l’Algérie: Kabyles

du Djurdjura, Chaouïa de l’Aurâs, Beni Mazab, Paris: Ernest Leroux. Repr. (1983 Aix-en-Provence: Édisud.)

Maunier, R. 1927. Recherches sur les échanges rituels en Afrique du Nord, L’AnnéeSociologique (n.s.), 2: 11–97.

——— 1935. Coutumes algériennes, Paris: Domat-Montchretien.Michelet, J. 1837. Origines du droit français cherchées dans les symboles et formulas du droit

universel, Paris: Hachette.Milliot, L. 1926. Les nouveaux qanoun kabyles, Hespéris, 6: 365–418.——— 1932. Les institutions kabyles, Revue de Études Islamiques, 6: 127–74.Montagne, R. 1930. Les Berbères et le Makhzen dans le sud du Maroc, Paris: Alcan.Moore, S.F. 1986. Social facts and fabrications: ‘customary’ law on Kilimanjaro 1880–1980,

Cambridge: Cambridge University Press.Morand, M. 1927. Le statut de la femme kabyle et la réforme des coutumes berbères, Revue

des Études Islamiques, 1: 47–94. Neveu, E. de. 1845. Les khouan: ordres religieux chez les musulmans d’Algérie, Paris: Guyot.Ould-Braham, O. 1996. Émile Masqueray en Kabylie (printemps 1873 et 1874), Études

et documents berbères, 14: 5–74.Parkin, R. 2005. The French-speaking countries, in F. Barth et al., One discipline, four

ways: British, German, French, and American anthropology, Chicago: The University ofChicago Press.

Patorni, F. 1895. Délibération de l’année 1749 dans la Grande Kabylie, Revue africaine, 39:315–20.

Perkins, K.J. 1981. Qaids, captains, and colons: French military administration in the colonialMaghrib, 1844–1934, New York: Africana.

Poussereau, L.-M. 1931. La carrière d’un officier nivernais en Algérie: le général A.Hanoteau (1814–1897), in A. J. Parès (et al.), Mémoires sur l’histoire de l’Algérie auXIXe siècle, Paris: Rieder.

Renan, E. 1873. La société berbère: exploration scientifique de l’Algérie [review ofHanoteau and Letourneux 1872–73], Revue des deux mondes 107: 138–57.

Rinn, L. 1891. Histoire de l’insurrection de 1871, Algiers: Jourdan.Roberts, H. 2002. Perspectives on Berber politics: on Gellner and Masqueray, or

Durkheim’s mistake, Journal of the Royal Anthropological Institute, 8: 107–26.Robin, J.N. 1902. L’insurrection de la Grande Kabylie en 1871, Paris: Lavauzelle.Salemink, O. 1991. Mois and Maquis: the invention and appropriation of Vietnam’s

Montagnards from Sabatier to the CIA, in G.W. Stocking (ed.), Colonial situations: thecontextualisation of ethnographic knowledge, Wisconsin: University of Wisconsin Press.

Salhi, M.B. 1979. Étude d’une confrérie religieuse algérienne: la Rahmania à la fin duXIXe siècle et dans la première moitié du XXe siècle. PhD dissertation, Paris: EHESS.

Scheele, J. 2008. ‘A kind of savage Switzerland’: local law-giving in Kabylia (Algeria), inComparative Studies in Society and History, 50(4).

2009. Village matters: knowledge, politics and community in Kabylia, Oxford: James Currey.Shinar, P. 2006. A major link between France’s Berber policy in Morocco and its ‘policy

of races’ in French West Africa: Commandant Paul Marty (1882–1938), Islamic Lawand Society, 13: 33–62.

Sibeud, E. 2002. Une science impériale pour l’Afrique? La construction des savoirs africanistesen France, 1878–1930, Paris: EHESS.

Snyder, F. 1981. Colonialism and legal form: the creation of customary law in Senegal,Journal of Legal Pluralism, 19: 49–90.

Stewart, F. 1987. Tribal law in the Arab world: a review of the literature, InternationalJournal of Middle Eastern Studies, 19: 473–90.


——— 1988–90. Texts in Sinai Bedouin Law, Wiesbaden: Otto Harrasowitz (2 vols).Surdon, G. 1938. Institutions et coutumes des Berbères du Maghreb, Tangiers: Editions

Internationales.Ter Haar, B. 1948. Adat Law in Indonesia, New York: Institute of Pacific Relations.Tooker, E. 1983. The structure of the Iroquois League: Lewis H. Morgan’s research and

observations, Ethnohistory, 30: 141–54.Trautmann, T.R. 1987. Lewis Henry Morgan and the invention of kinship, Berkeley:

University of California Press.Tupper, C.L. 1881. Punjab customary laws, Calcutta: Office of the Superintendant of

Government Printing (3 vols).Urbain, I. [under pseudonym G. Voisin] 1860. L’Algérie pour les Algériens, Paris: M. Levy.

Chapter 3

POSTCARDS AT THE SERVICE OF THE IMAGINARY:

JEAN ROUCH, SHARED ANTHROPOLOGY AND THE CINÉ-TRANCE

Paul Henley

Introduction

The very nature of ethnographic cinema – how it is practised, how it istalked about, where its limits are deemed to lie – has been profoundlyshaped by the work of the late Jean Rouch, who died tragically in a roadaccident near Tahoua, Niger in February 2004. In the course of a sixty-year career, beginning with his first tentative ethnographic reportspublished in a French colonial journal in the early 1940s and endingwith his last film, poignantly entitled Le rêve plus fort que la mort andreleased in 2002, Rouch produced over a hundred completed films andalmost as many published texts. While a handful of the films have beenwidely distributed, reaching far beyond the confines of academicanthropology, the great majority remain little known and extremelydifficult to see, particularly in the English-speaking world.

In France, Rouch’s reputation was probably at its peak in the early1960s. He was not only well known among anthropologists, but he wasalso a national figure in French cinema and his films were regularlyreviewed in the avant-garde screen studies journal, Cahiers du Cinéma(Prédal 1996). In contrast, at this time his work was virtually unknownin what the French like to call the ‘Anglo-Saxon’ world.1 From the early1970s, Rouch’s star gradually began to wane in France and by the early

Figure 3.1 Jean Rouch shooting in a market in the Gold Coast in 1954. Bibliothèquenationale de France, Département des manuscrits, ref. 28464.

1980s, his profile was significantly diminished, both in anthropologicaland cinema circles. Ironically, it was precisely at this time that he cameto be discovered by Anglo-Saxon visual anthropologists, since his viewsabout anthropology and cinema, and the relationship of both toempirical reality, struck a chord with the postmodernist tendencies thatwere then sweeping through Anglo-Saxon anthropology, particularlyon the other side of the Atlantic.2 But although Rouch may have beenhailed as a prophet of postmodernism in Anglo-Saxon anthropology –much to his surprise and amusement – his approach was deeply rootedin a distinctively French intellectual tradition.

The surreal encounter

Rouch first became attracted to anthropology through a priorengagement with surrealism while he was still a teenager. There is astory that he liked to tell about this first encounter, which involved, itseems, something of a Damascene conversion, complete with blindinglight. It happened one spring afternoon in 1934, when the seventeen-year-old Rouch passed a bookshop in the Montparnasse quarter ofParis. There, in the window, in a pool of light cast by the setting sun,

76 Paul Henley

was a display of two juxtaposed double-paged spreads from recenteditions of Minotaure, a journal which published the work both of theleading surrealists of the day and anthropologists. One of these spreads,from the second edition, showed some funerary masks worn by theDogon of the Bandiagara Cliffs in what is now eastern Mali, but whichwas then the colony of the French Soudan. This image was but one ofmany illustrations accompanying the principal subject matter of thatedition of the journal, namely a special report on the celebrated Dakar-Djibouti Expedition of 1931–1933, written by its leader, Marcel Griaule.The other spread included a reproduction of a painting of two figuresin a dream-like landscape by Giorgio de Chirico, a painter muchadmired by the surrealists. In the mind of the young Rouch, the twoimages became inextricably associated and the Bandiagara Cliffs tookon the character of a fabled landscape to which he dreamed that oneday he too would be able to travel (Rouch 1995c: 410).

The many connections between ethnology (as the study of social orcultural anthropology was then known in France), surrealism and l’artnègre – the latter embracing everything from traditional African masks,such as those featured in Minoature, to jazz, the dancer Josephine Bakerand black American boxers – have been extensively commented upon byJames Clifford, Christopher Thompson and others, as well as by Rouchhimself on a number of occasions.3 The distinguished historian ofanthropology in France, Jean Jamin, has questioned the true extent ofthe connections between ethnology and surrealism, suggesting that itwas more a question of two activities occupying adjacent intellectualspaces rather than being involved in a genuine exchange. While theethnologists were committed to the detached observation and rigorousanalysis of cultural phenomena, the surrealists sought a highly subjectiveimmersion in other cultural realities, hoping to tap into the primitivecreative life forces which they imagined to be inherent in such cultures,particularly those of Africa. According to Jamin, although there mayhave been certain ‘complicities and affinities’ between ethnology andsurrealism, there was no long-term or systematic transfer of methodsand concepts (Jamin 1991: 84). But while this may have been generallytrue, it certainly does not apply to Rouch, whose film-making methodscontinued to be informed by the precepts of surrealism right until the endof his film-making career, long after these ideas had not only lostwhatever currency they might have once had in ethnology, but had alsofallen out of fashion in the worlds of the plastic arts and poetry.

The oeuvre of Jean Rouch as a film-maker is not only remarkably largebut also remarkably eclectic. Although the great majority of his filmswere shot in West Africa, or failing that in Paris, they include both full-scale feature-length documentaries and short documentation films.

Postcards at the service of the Imaginary 77

78 Paul Henley

Films about large-scale African ritual events alternate with intimateportraits of Rouch’s friends, and promotional films for West Africandevelopment agencies sit side by side with ‘ciné-poems’. A major strandof this oeuvre consists of fictional feature films, though true to Rouch’ssurrealist leanings, these did not generally involve any kind of formalscript but relied instead on spontaneous improvisation around a seriesof ideas that Rouch and his protagonists made up as they went along.Initially these fiction films were anchored in Rouch’s ethnographicresearch, but they became progressively more fantastical and surrealistas his career developed. In an allusion to the ethnographic origins of hisearlier work in this mode, Rouch himself referred playfully to his fictionfilms as ‘science fiction’, though third-party commentators on Rouch’swork have tended to prefer the somewhat debatable term ‘ethnofiction’.

Rouch was a man of great energy, but even he could not have beenso productive a film-maker had it not been for his particular professionalcircumstances. In 1947, while still engaged in doctoral research on thereligious life of the Songhay of the middle reaches of the Niger river, hewas admitted to the CNRS. Apart from a brief interlude in 1951–53,when he was temporarily expelled for not having completed his thesis –largely due to competing film-making activities – this research positionenabled him to go to West Africa almost every year for the rest of hiscareer and shoot a number of films, unencumbered by any heavyteaching obligations. Not every one of the films that he shot on thesealmost annual fieldwork ‘missions’ was a masterpiece. Many wereunpretentious descriptive accounts of religious ceremonies, notably ofspirit possession rites in which he had a particular interest and which hehimself estimated to feature in some fifty of his films (Taylor 2003:140). Quite a number of his films remained unfinished and to this dayexist only as rushes or unmarried prints. Some were experiments thatfailed. But even if all these minor, incomplete or unsuccessful works arediscounted, it remains the case that Rouch was and is by far the mostproductive film-maker, living or dead, to have made ethnographic films.The films that he made in the early part of his career, especially those hemade in a particularly creative period between 1950 and 1960,established a new standard in the history of ethnographic film-makingthat continues to inspire ethnographic film-makers the world over.


The joking relationship

Although Rouch was entirely self-taught as a film-maker, he did receivea formal anthropological training. In the academic year 1940–1941,with Paris already under the German occupation, Rouch enrolled onan extramural course given by Marcel Griaule in the basement of theMusée de l’Homme. Germaine Dieterlen, Griaule’s partner in bothpersonal and professional life, supported the lectures with a series of‘magic lantern’ slide shows. At the time, Rouch had no formalinvolvement with either anthropology or film-making. In fact, he was inhis final year as an engineering student at the elite grande école, Ponts etChausées, but the memory of the episode in front of the Montparnassebookshop evidently still burned brightly in his mind.

The relationships that he formed with Griaule and Dieterlen throughthis course would be of crucial importance in the shaping of his futurecareer as both anthropologist and film-maker. Shortly afterwards, inorder to fulfil his dream of travelling to West Africa as well as to escapefrom wartime France, Rouch took a job as a road-building engineer inNiamey, capital of the French colony of Niger. Here he came acrossspirit possession first-hand among his labourers and began his firstethnographic research of the phenomenon, aided by a questionnaireprepared for him by Dieterlen and sent out to him by a roundaboutroute. Later, in 1944–45, after a couple of years combining engineeringwork and private study in the Institut français d’Afrique Noire in Dakar,he joined the Free French forces in Africa and participated in theliberation of France and invasion of Germany. After the war, hereturned to France and enrolled at the Sorbonne to study for a doctoratein anthropology under the supervision of Griaule.

In the immediate post-war period, Griaule was under something of acloud. Unlike most leading anthropologists, many of whom had takenrefuge abroad during the war, Griaule had chosen to remain behind andco-operate with the Vichy government. He not only accepted a Chair atthe Sorbonne but also, having been an aviator in his youth, acommission as a colonel in the air force. However, despite his personalaversion to everything associated with the Vichy regime, Rouch electedto study under Griaule because, he claimed, Griaule and his group simply‘had more fun’ than the other leading Africanists with whom he mighthave worked.4 There were probably some more pragmatic reasons too:Griaule was the leading French authority on the middle Niger whereRouch wanted to work and, with Dieterlen, he had supported Rouch’sown first amateur ethnographic research during the war years. It couldalso have been important that Griaule was sympathetic to film-making,as he had made two films among the Dogon in the 1930s.5

80 Paul Henley

Yet although there may have been a number of good reasons forRouch to study with Griaule, there always remained a certain ambiguityin Rouch’s attitudes towards his mentor, involving a curious mixture ofdisdain and respect. Rouch liked to present this as an extrapolation of thetraditional joking relationship between the cliff-dwelling Dogon, whomGriaule had studied, and the Songhay and the other peoples of thelowland fluvial plains of the Niger whom he himself had worked with.But it seems very likely that it was also the result of a certain dissonancein their political views, not only in relation to collaboration during thewar years, but also with regard to the French colonial project in Africa.6

Griaule passed on to Rouch his particular take on the intellectualinheritance that he had received from his own mentor, Marcel Mauss.As Clifford has described, from a methodological point of view, Mauss’sapproach involved a clear differentiation between the process ofethnographic description and the process of theoretical explanation. Inthe Maussian methodology, the first stage in any research project shouldconsist of the systematic accumulation of large numbers of‘documents’, namely bodies of ethnographic data. These ‘documents’could be culled from a broad variety of sources, both historical andanthropological, textual as well as verbal. Indeed, everything andanything could be grist to the ethnographer’s mill, for, as GermaineDieterlen once remarked, the most clumsy design scratched on a wallwith a fingernail could provide a clue to ideas about the structure of theuniverse (Dieterlen 1988: 252). This process of accumulating‘documents’ should be as objective as possible and free from a prioriexplanatory concerns. The ‘documents’ could then be subjected torigorous scholarly exegesis within the framework of indigenous conceptsand linguistic categories. But the elaboration of exogenous theoreticalexplanations or arguments in terms of comparative ethnography wereprocesses that should happen later, a posteriori, rather than in the processof accumulating the ‘documents’ in the first place.

Although some of the theoretical conclusions that Mauss drew fromthe minute analysis of ethnographic ‘documents’ have been the sourceof great inspiration to subsequent generations of anthropologists,contemporary accounts suggest that, in his lectures, Mauss often gotso immersed in the ethnographic detail that he never quite arrived atthe elucidation of the theoretical conclusions (Clifford 1988a: 123–25). Rouch’s recollection of Griaule’s lectures as a series ofdisaggregated ethnographic titbits suggests that they too may havesuffered from the same shortcomings (Rouch 2003b: 103–4). Moregenerally, in this particular anthropological school, it seems that therewas a distinct tendency for detailed ethnographic description to beprioritised and appreciated for its own sake, while theoretical


explanation and generalisation were treated as matters of a secondorder of elaboration that in the ideal case should eventually follow, butin actual practice might not do so.

Certainly this was true of Rouch’s textual anthropology. His doctoraldissertation on the Songhay was submitted in 1952 and later publishedunder the title La Religion et la magie songhay, first in 1960, and then ina substantially annotated second edition in 1989. The originaldissertation was completed shortly after the publication of Dieterlen’sclassic work, Essai sur la religion bambara, and it was clearly heavilyinfluenced by this model.7 Not only the general approach, but even thestructure and layout of Rouch’s work follow those of Dieterlen’s workvery closely. That is, Rouch provides a highly detailed but entirelydescriptive account of Songhay beliefs in which each element or aspectof Songhay traditional religion is described sequentially and in isolation– the general cosmology, the myths of origin associated with particularcult activities, the various roles or offices involved, the ‘texts’ chanted,the forms of dance, the types of musical instrument and so on. The onlyexplanations that are offered for these practices are in terms of locallegends or beliefs, often quoted verbatim. At the end of the book, in a‘conclusion’ of less than three pages, Rouch makes no attempt toidentify any general theoretical consequences of his study. He declinesto present Songhay religious ideas and practices as ideologically relatedto particular social or political structures. Analyses in terms of eitherthe comparative ethnography or history of West Africa as a whole arenot merely eschewed but ridiculed, albeit humorously. Instead, Rouchchooses to celebrate – in defiance of considerable evidence to thecontrary, to which he himself even alludes – the original and uniquecharacter of Songhay religion.8

But if Rouch’s general intellectual formation can be traced backultimately to Mauss, his ideas about the actual practice of anthropologyin the field were more directly influenced by the methodology of Griaulehimself. For although Mauss actively advocated fieldwork, his owninvestigations were entirely bibliographic. Griaule, in contrast, washighly committed to fieldwork in practice as well as in principle. Hisideas about how to conduct fieldwork are laid out very explicitly in hisMéthode de l‘ethnographie. This slim handbook was not published until1957, the year after his death, but it draws on his experiences inworking with the Dogon since the 1930s. The approach that Griauleproposes here is very different from that developed around the sametime by Anglo-Saxon anthropologists, notably by BronislawMalinowski, which, since the 1960s, has become the orthodoxy insocial and cultural anthropology generally, even in France. WhereasMalinowski proposed that the fieldworker should ‘plunge into the life

82 Paul Henley

of the natives’, usually alone, becoming a relatively unobtrusive‘participant-observer’ engaged in the day-to-day life of a community,Griaule advocated the formation of teams of fieldworkers, organisedalong quasi-military lines who would not only maximise the collectionof data within any given time period but at the same time triangulatethe results that they were obtaining. Far from the discreet observationof life as it is lived with minimal interference, as in the Malinowskianapproach, Griaule’s method was highly pro-active, involving intensiveinterrogatory interviews based on systematic questionnaires. Ratherthan observing the subjects interacting among themselves, Griaulepreferred to work with a select group of elite informants, using bilingualintermediaries rather than the native language.9

Rouch gives a rather droll account of the fieldwork routines ofGriaule and Dieterlen as he observed them while visiting their camp atSanga, at the foot of the Bandiagara Cliffs, in 1950. First thing in themorning, Griaule would give all members of the team their tasks for theday. While he and Dieterlen worked through questionnaires with theirestablished informants and other researchers in the team weredispatched elsewhere, groups of traditional musicians would besummoned to perform so that they could be filmed by Rouch and hisfilm-making partner and sound-recordist on this particular expedition,Roger Rosfelder, also a student of Griaule. Alternatively, Rouch andRosfelder would be sent off to film daily routines in a nearby Dogonvillage. At noon, the whole team would meet up in the company of theDogon informants and interpreters and exchange the informationgathered in the morning. On the basis of these discussions – in which,Rouch emphasises, the Dogon played an active part – Dieterlen wouldtypically develop an inspired series of further hypotheses which Griaulewould then order into a new series of questionnaires to be used in theafternoon. With perhaps just a touch of irony, Rouch compares thisapproach to the Socratic method of successive dialogicalapproximations to philosophical truth.10

The fieldwork approach that Rouch himself would develop as both afilm-making and a text-making anthropologist shared certain similaritieswith that of his mentor. Like Griaule, Rouch returned faithfully to thesame field sites in West Africa over a prolonged period. Indeed, Rouchliked to quote Griaule and Dieterlen’s view that one needed at least twentyyears of first-hand experience of a given society before one could begin toachieve a ‘deep knowledge’ of its systems of thought (e.g. Rouch 2003b:111). Like Griaule, Rouch tended to rely on a key group of informantsand worked largely through the medium of French, perhaps because, ashe himself confessed, he was ‘not very good at languages’.11 In his‘straight’ anthropological fieldwork, he often used formal questionnaires,


both during his doctoral research and in the migration studies that hecarried out in West Africa in the 1950s (Rouch 1956: 37). Also likeGriaule, Rouch tended to focus his attention on what might be called thepublic cultural rhetoric of the groups whom he studied ethnographically.That is, the great majority of his films are about public ceremonialperformances of one kind or another, and there is very little emphasis onprivate domestic life and the routines of the everyday. Partly for thisreason, Rouch’s films mainly concern the public world of men, the moredomestic world of women being relatively neglected.12

Although Rouch never used interviews of any kind in his films, onecan also perhaps detect, as Clifford has done, a certain continuitybetween, on the one hand, what he calls Griaule’s ‘dialogical method’,in which interrogatory questions were aimed at provoking the subjectsinto revealing answers, and on the other, Rouch’s idea that the cameracould act as a catalyst to provoke his subjects into revelatoryperformances (Clifford 1988b: 77). Indeed, it is tempting to argue that,in the same way that Griaule’s pro-active methods contrasted with themore passive participant observational methods of Anglo-Saxonanthropology, so too did Rouch’s pro-active cinematographic methodscontrast with the more low-key methods of direct or observationalcinema as practised by his Anglo-Saxon film-making contemporaries.13

Shared anthropology

If there were certain similarities between Griaule’s and Rouch’sfieldwork methods, though, there were also profound differences, mainlyrelated to their very different attitudes towards their subjects. ForGriaule’s ‘dialogical method’, notwithstanding the positive connotationsof this way of describing it, was essentially antagonistic, being based onthe initial working assumption, stated repeatedly in the methodologicalhandbook, that the informant was lying. In an extended legal analogy,Griaule suggests that the informant should be considered the equivalentto the ‘guilty party’ in a court of law, while the remainder of the societyshould be considered his ‘accomplices’. In order to combat aninformant’s congenital tendency to lie, Griaule recommended that theresearcher – compared variously to a prosecution lawyer, judge and evena bloodhound – should use whatever trick or stratagem was necessary tocircumvent the informant’s defences. Although Griaule may havedeveloped a profound respect for African culture, coming to regardDogon cosmology as the equal of that of ancient Greece, hismethodological recommendations suggest that he had no respect for theAfricans themselves as individuals.14

84 Paul Henley

Griaule’s aggressive and unscrupulous attitude, shocking to a modernsensibility and undoubtedly the product of a colonial mentality wherebyall indigenous knowledge was fair game, could not be further from Rouch’sown attitudes. Shortly after his arrival in Niger to take up the post of roadsengineer in December 1941, and in defiance of the Vichy governor’sdisapproval of familiarity with Africans, Rouch became friendly with ayoung local man, Damouré Zika, and appointed him as his assistant.Damouré was a member of the Sorko sub-group of the Songhay, whosespecialist metier was fishing the waters of the Niger. It was he who firstintroduced Rouch to spirit possession through his grandmother, Kalia, apriestess of one of the local spirit possession cults. Damouré was the firstand most significant of a group of Africans whom Rouch subsequentlygathered around himself and who accompanied him whenever he went toAfrica. Later additions to this group included Lam Ibrahim Dia, a Fulanicattle-herder, Illo Gaoudel, also a Sorko fisherman, and TallouMouzourane, a Bella cattle-herder and general ‘go-fer’. Somewhat later,Moussa Hamidou, who was from the Zerma, a group closely related to theSonghay, also joined the ‘gang’. These men helped Rouch in a variety ofdifferent ways: they conducted surveys for his migration studies, crewedon his documentaries and took a leading part as actors in his ethnofictions.They also drove his Land Rover, carried his equipment and generally actedas his local fixers. In return, Rouch not only paid them salaries while theyworked for him but shared the profits of his films on a fifty-fifty basis. Healso supported them in many other ways: he arranged for Damouré to betrained as a medical auxiliary and later as a pharmacist, which allowedhim, in local terms, to achieve great wealth and status; Lam became aprofessional driver after having learnt to drive with Rouch and used hispart of the income from the films to buy himself vehicles; through hiscinema work with Rouch, Moussa was able to pay for all his sons to beeducated as professionals; when Rouch met Tallou, he was an orphansuffering from leprosy, so Rouch arranged for him to be cured and thentook him under his wing, supporting him for the rest of his life.15 WhenRouch died in the tragic road accident in February 2004, travelling in thesame car with him, though fortunately not seriously hurt, was Damouré,still accompanying Rouch some sixty-two years after they first met.

These attitudes of respect for and engagement with his subjects weremade manifest in various ways in Rouch’s works, both textual andvisual. Thus, in the introduction to his doctoral thesis, Rouch advisesthe reader that he has omitted some of the secret knowledge to whichhe felt privileged to have been made party since he had promised theSonghay that he would not divulge it. ‘The ethnographer is not apoliceman who extorts matters about which there is a desire neither totell him nor show him’ (Rouch 1989: 17–18). This is certainly a far cry


from Griaule’s analogy of the ethnographer as a bloodhound of thesocial fact, desperate to break down his informants’ resistance bywhatever means necessary.

Similar attitudes are evident in Rouch’s film-making methods. In theearly 1950s, he began making a point of regularly screening his filmsto their subjects. Rouch liked to trace this ‘feedback’ practice back to theexample set by Robert Flaherty, the so-called ‘father’ of ethnographicdocumentary, who, during the making of Nanook of the North in theearly 1920s, had screened his rushes to his subjects in order to decidewhat they should film the next day. But Rouch went very much furtherthan this, giving his African collaborators a much greater role indevising the content of his films than Flaherty ever gave to Nanook andhis companions. Flaherty asked the Inuit to adjust their houseconstructions, subsistence activities, their costumes and even theirpersonal identities to the requirements of his film. In contrast, Rouchwas reluctant to ask his subjects to dress up or behave in any special way.Instead he would simply ask them to improvise along whatever lines theythemselves thought fit (Colleyn 1992: 46–47, Rouch 1995a: 88).

Rouch’s feedback procedures were also much more elaborate thanthose of Flaherty. He did not merely screen his rushes to his subjects inorder to plan the next day’s shooting: instead, feedback screeningsbecame the very cornerstone of his way of working. Often he wouldreturn, months or years later, not with the rushes, but with thecompleted film and screen that to his subjects. Like Flaherty, Rouchappreciated the pragmatic advantages that could arise from suchscreenings. When he first began his ethnographic research, Rouch hadtried giving his written works to the Songhay, but had quickly discoveredthat they had no use for them, even when they were read out loud by thevillage school-master. On the other hand, when he started screeninghis films, not only did the Songay understand his objectives moreclearly, they became his active collaborators (Rouch 1995b: 224).

At the simplest level, this collaboration merely consisted ofcommenting on the ethnographic content of the films. This provedparticularly valuable many years later, in the late 1960s and early1970s, when Rouch and Dieterlen came to make their series of filmsabout the Dogon Sigui festival and had very little idea about what wasgoing on in the actual moment of shooting. By listening to thecomments of the subjects during the feedback screenings, as well as tothose of a ritual specialist, Amadigné Dolo, whom they took back to Paristo work with them in the editing suite, they learnt a great deal about thesymbolic significance of particular forms of dancing, the many items ofritual paraphernalia, the reasons for particular sequences of events andso on, none of which they would otherwise have understood.

86 Paul Henley

However, although Rouch readily acknowledged the pragmaticadvantages to be gained from screening his films, he also thought ofthem in ethical terms, believing that ethnographic film-makers had anabsolute duty to screen their films to their subjects. He liked to think ofthese screenings as a form of ‘audiovisual countergift’ – a veryMaussian term – offered in exchange for the support he had receivedfrom the subjects during the production process. In this way, hesuggested, ethnographic film-makers could avoid acting as if they were‘entomologists’ capturing exotic specimens.16 Instead, their work couldform the basis for promoting mutual understanding and respectbetween observer and observed:

This is the start of what some of us are already calling ‘sharedanthropology’. The observer is finally coming down from his ivory tower;his camera, tape recorder, and his projector have led him – by way of astrange initiation path – to the very heart of knowledge and, for the firsttime, his work is not being judged by a thesis committee but by the verypeople whom he came to observe (Rouch 1995a: 96).

However, these feedback screenings were only the start of a longer-termprocess since the ‘shared anthropology’ that Rouch practised involvednot merely the ‘audiovisual countergift’ of screening the films to thesubjects, but also their direct engagement in the process of making thefilms themselves. Much more important than the feedback per se – arelatively passive activity – was the highly active collaboration thatfollowed thereafter. For Rouch discovered that, at the end of a feedbackscreening, one or more members of the audience would often come upto him and suggest an idea for a new film. These could be people whohad been directly involved in the first film, or other members of theaudience who had concluded that a film about their activities would beeven more interesting than the film that Rouch had just shown. In thisway, the screening of one film could lead to the making of another inwhich the subjects who proposed the idea were not merely protagonistsbut, as we might say today, ‘stakeholders’ in the making of a new film.

Postcards of the imaginary

This commitment to the idea of a ‘shared anthropology’ as far back asthe 1950s (even if he did not give it precisely this name until the early1970s) anticipated by more than two decades the ‘dialogicalanthropology’ that, under the influence of postmodernism, becamefashionable in Anglo-Saxon anthropology from the late 1970sonwards. But although Rouch’s methodology was certainly marked by


this and a number of other apparently postmodern traits – andalthough he firmly rejected the great twentieth-century meta-narratives of Marxism, Freudianism and structuralism – he had arrivedat these positions not through antipathy to modernism as such but bya series of quite different routes.17 As a result, although there werecertainly some similarities between Rouch’s practices and thoseadvocated by Anglo-Saxon postmodernists, there were also somesignificant differences.

As far as the technology of film-making itself was concerned, Rouchwas actually very modernist in his ideas, believing enthusiastically inthe potential of technological advance to transform human experiencefor the better. In the early part of his career, drawing on his ownengineering background, he collaborated actively with camera andsound-recording design engineers to develop a system of mobile,lightweight cameras and portable audio tape-recorders that couldoperate in tandem with one another in such a way as to produce imagesin which actions and sounds would be perfectly synchronised.Nowadays, such synchronised images are entirely commonplace sincethey are achieved automatically by video technology. But in the late1950s and early 1960s, the achievement of synchronised images usingportable equipment became a sort of holy grail for manydocumentarists, and many technical experiments took place, not onlyin France among Rouch and his associates but also in North America,notably among the so-called Direct Cinema group of documentaristsheaded by Robert Drew and Richard Leacock, as well as among a groupof film-makers working for the National Film Board of Canada (Rohmerand Marcorelles 1963: 16–22, Mamber 1974).

Prior to the development of this new technology, the only way torealise full synchronisation of sound and image – particularly ofspeech, which was by far the most difficult form to achieve – was bymeans of equipment that was far too heavy to be portable. Therefore, ifa film-maker wanted ‘synch sound’, he or she had to bring the subjectsto wherever the equipment was located and ask them to perform there,with obvious negative consequences for the spontaneity, authenticityand range of the behaviours that could be filmed. Following thedevelopment of portable synch sound systems – achieved more or lesssimultaneously on both sides of the Atlantic in the early 1960s, albeitusing slightly different technologies – documentary film-makers wereable to follow their subjects as they moved around their particularprivate social worlds, allowing them to decide for themselves exactlywhat they should say or do in front of the camera.

In North America, the documentarists of the Direct Cinema groupsought to use this new portable technology to maximise their own

88 Paul Henley

effacement while shooting, interfering as little as possible in thebehaviour of their subjects. In this way, they hoped that they would beable to film their subjects behaving just as they would have behaved hadthe camera not been there. Rouch’s attitude to the new technology wasvery different. Although he appreciated the greater indexicality that fullysynchronous sound brought to his images and also the greater possibilityfor spontaneity on the part of the subjects that the new technologyallowed, he certainly did not aspire to self-effacement, nor hope to filmhis subjects behaving as they would have been had his camera not beenthere. On the contrary, Rouch believed that the presence of the camerawould inevitably affect the performance of the subjects, howeverdiscretely it was operated. But far from devaluing the quality of thematerial that was filmed, he thought that this provocation ofextraordinary behaviour increased its value. This was because, in puttingon a special performance for the camera, the subjects would reveal moreabout themselves, and particularly about the inner thoughts, dreamsand fantasies of their imaginaries. ‘What has always seemed verystrange to me’, he commented in an interview he gave in 1964, ‘ is thatcontrary to what one might think, when people are being recorded, thereactions that they have are always infinitely more sincere than thosethey have when they are not being recorded’ (Blue 1996: 268–69).

This idea that, beneath the surface of social behaviour, there isanother and more significant reality to be discovered in the minds ofthe subjects is consistent with a long tradition of thought in Frenchanthropology. This intellectualist strand can be traced back throughGriaule and Dieterlen to Mauss, but it is also expressed in the prioritygiven to langue over parole in structuralist thinking. But while thistradition might certainly have influenced Rouch, the primaryinspiration for this idea appears to have been his youthful encounterwith surrealism. For, paradoxically, although the new technologicaladvances greatly increased the fidelity of the copy of the world offeredby the cinematographic apparatus, it was its ability to bring to thesurface what was normally hidden that was most appreciated by Rouch.As he put it in a 1967 interview:

For me, cinema, making a film, is like surrealist painting: the use of the mostreal processes of reproduction, the most photographic, but at the service ofthe unreal, of the bringing into being of elements of the irrational (as inMagritte, Dalí). The postcard at the service of the imaginary. (Fieschi andTéchiné 1967: 19)

However, it was not only its capacity to provoke a spontaneousperformance on the part of the subject that Rouch appreciated aboutthe new technology: he also appreciated the fact that, by its very


mobility, it allowed a much greater subjective immersion on the part ofthe film-maker in the world of the subject. In this connection, Rouchenthusiastically endorsed the analogy drawn by Edgar Morin, the co-director of Chronicle of a summer, who suggested that, with the aid of thenew technology, Rouch could become a sort of ‘filmmaker-diver’ who,unencumbered by equipment, could ‘plunge into real-life situations’.Here, he could discover ‘virgin territory, a life that possesses aestheticsecrets within itself ’.18

However, for Rouch this subjective immersion in the world of thesubjects was not just a straightforward technical-operational strategy:it also involved a transformation in the state of mind of the film-maker.If film-making had the power to provoke the subjects into entering theworld of their imaginaries and revealing their innermost secrets, itcould also lead the film-maker to immerse himself in the world of theimaginary and produce a performance of his own. The key to Rouch’sviews on this matter was his notion of the ciné-trance.

The ciné-trance

Rouch’s most systematic discussion in print of the notion of the ciné-trance is in an article that was first prepared for a celebrated CNRSconference, La Notion de personne en Afrique noire, that took place in 1971.Here he discusses a series of ideas that had come to him earlier that yearin the course of making a short film, Les Tambours d’avant: Tourou et Bitti.19

This film was shot in a Zerma village about fifty miles north of Niamey,capital of Niger. It consists almost entirely of a single ‘sequence shot’ – anunbroken take of approximately eleven minutes lasting the full durationof a 16mm film magazine. The subject of the film is a spirit possessionceremony in which the villagers seek the aid of the spirits in preventinglocusts from destroying their new millet crop.20 Arriving at the village onwhat was already the fourth day of the ceremony, Rouch and his sound-recordist Moussa Hamidou discovered that the mediums had not beenable to go into trance. This was despite the strenuous efforts of themusicians to attract the spirits by playing their signature music on themonochord violin and various types of percussion instrument. The latterincluded the traditional drums alluded to in the title of the film, the tourouand the bitti, which the spirits were known to favour particularly. By fourpm, as the light was beginning to fade, Moussa suggested to Rouch thatthey should at least take the opportunity to film the tourou and the bittisince these drums were played with increasing rarity.

After a couple of preliminary shots outside the village, the sequenceshot begins on the sun and then pans down to enter the village, passing

90 Paul Henley

a herd of tethered sacrificial goats on the left and, on the right, adisconsolate male medium still awaiting inspiration, before crossing thesmall earthen plaza and approaching the orchestra. The musiciansredouble their efforts as the camera glides over them, revealing therange of their different instruments, one by one. At this point, the musicbegins to peter out and the camera starts to withdraw, when suddenlythere is a sudden cry of ‘Meat!’ and the medium goes into trance as heis possessed by the spirit of Kure the Hyena. The priests of the cult, thezima, then approach and engage Kure in a bantering dialogue, offeringhim ‘meat’ in the form of sacrificial animals, in exchange for ‘grass’, agood harvest. At this point, with the camera still turning, an old womanhops across the plaza, shivering all over because she has been possessedby the spirit of Hadyo the Fulani slave. The zima continue theirnegotiations with Kure, who is now threatening to leave unless he gets‘blood’. But, as it is nearing the end of the magazine, the camerawithdraws to the edge of the plaza. From here, it ends on a wide shotshowing the young people looking on, before finally panning up againto the now-setting sun.

A number of different elements of Rouch’s film-making praxis cometogether in this short film. Although he began to shoot with merelydescriptive ethnographic objectives, the action develops into somethingmuch more interesting as a result of the presence of the camera. For,Rouch claimed afterwards, it was the fact that he was shooting a film thatserved to provoke the adepts to go into a trance state. 21 As such, the filmrepresented a good example of the positive benefits that can arise from thechange in reality brought about by the presence of the camera. However,the film also exemplifies Rouch’s understanding of the play betweensubjectivity and objectivity that is involved in making a film in hisparticular way. Close to the beginning, over one of the preliminary shotsoutside the village, he explains on the commentary track that the film is‘an attempt to practise ethnographic cinema in the first person’. This isfollowed by a cut to black with the title ‘un film de Jean Rouch’ discreetlydisplayed in one corner. Only then does the sequence shot proper begin,starting off in travelling-shot mode with Rouch commenting over it, ‘Toenter into a film is to plunge into reality, and to be, at once, both presentand invisible’. Thus the film is presented as an unexpurgated slice of timeinvolving a ‘plunge’ into reality, but at the same time as a view of thisreality that is both intensively subjective (‘ethnographic cinema in the firstperson’) and authored (‘un film de Jean Rouch’).

It was directly as a result of shooting Tourou et Bitti that Rouch firstbegan to develop his concept of the ciné-trance. He later described how,when he and Moussa Hamidou put down their equipment at the end ofthe take, they were trembling with exhaustion, aware that they had just


been through a powerful experience. However, Rouch did not attributethis to the physical exhaustion that might reasonably follow fromconcentrating intensively on shooting a ten-minute sequence shot.Rather, he thought that it was due to the fact that the engaging rhythmof the tourou and bitti drums had not merely sent the two adepts intotrance but the two film-makers as well. While the adepts had beenpossessed by the spirits Kure and Hadyo, he and Moussa had gone intoa ciné-trance, possessed by what he would later describe as a sort of‘enthusiasm which cannot be defined but which is essential to poeticcreativity’. This was comparable, he suggested, to the German conceptof Stimmung, a term which literally means ‘humour’, in the sense of ‘aframe of mind’, or ‘a tuning’ as of a musical instrument, but which,Rouch claimed, defies translation in this more poetic sense.22

On other occasions, he referred to this state as a form of ‘grace’, acondition that he associated not with any Christian notions but withNietzsche’s concept of the Dionysian – namely creative activity that isspontaneous and intuitive rather than rational. This condition cannotbe guaranteed, he suggested: in fact, it is relatively rare. If it is notpresent, Rouch believed, one might as well give up filming becausenothing truly significant is going to be revealed (Fulchignoni 1981: 8–9, also Fulchignoni 2003: 150). But on the basis of his own experience,Rouch claimed that when one entered this state, one was liberated fromthe weight of anthropological and cinematographic theory and becamefree to rediscover what he called ‘la barbarie de l’invention’ – a phrase thatalso defies any simple, literal translation but which one might renderhere as ‘raw creativity’ (Rouch 1997: 226).

In effect, then, in Rouch’s view, film-making could only be trulysuccessful if it involved performances on both sides of the lens: while thesubjects were provoked by the circumstances of film-making into aperformance revealing the contents of their imaginaries, the film-makerentered a trance-like state in which his most elemental creative abilitieswould be released. Moreover, these performances had to be in harmonywith one another. Over the years, Rouch used many different analogiesto describe this harmonisation of performances between film-maker andsubject. Sometimes he compared it to a ballet, at other times to a matadorimprovising his passes before the bull (Rouch 1995a: 89–90). In the idealcase, the conjunction of these performances produced something greaterthan the sum of their parts, similar, Rouch suggested – in a characteristicclutch of metaphors – to the transporting moments of a jam sessionbetween Duke Ellington and Louis Armstrong or, perhaps even closer tohis heart, the flashes of revelation that could arise from the electrifyingeffects of an encounter between strangers, as described by the surrealistpoet, André Breton (Fulchignoni 1981: 31, also Fulchignoni 2003: 186).

92 Paul Henley

Privileged truths

There was also another, very different ingredient to Rouch’s notion ofthe ciné-trance. The fact that he refers to this trance-like state not justas a trance, but as a ciné-trance is a sign of the influence on his thinkingof the ideas about the nature of cinematographic reality associatedwith the Polish-Russian film-maker, Dziga Vertov. Best known for his1929 film Man with a movie camera, Vertov’s work appears to have beenan enthusiasm that Rouch first took up in the course of makingChronicle of a summer with Edgar Morin in 1960–1961.23 Rouch seemsto have found in Vertov’s work an endorsement of his own view thatthe cinematographic apparatus offered a new and privileged way ofrepresenting the world:

Dziga Vertov … understood that the cinematographic way of looking washighly distinctive, employing a new organ of perception, the camera, whichbore little relation to the human eye, and which he called the ‘ciné-eye’.Later, with the appearance of sound, he identified a ‘radio-ear’ in the sameway, as an organ specific to recorded sound … Taken as a whole, he calledthis discipline cinéma-vérité (cinema-truth), which is an ambiguousexpression since, fundamentally, cinema cuts up, speeds up, slows down,thereby distorting the truth. For me, however, ‘cinema-truth’ has a specificmeaning in the same way that ‘ciné-eye’ does, designating not pure truth,but the truth particular to recorded images and sounds: ‘ciné-truth’. (Rouch1997: 224)

The term cinéma-vérité has a chequered history in the literature ondocumentary film-making. For a period, in North America particularly,it was understood to denote a documentary-making practice thataspired to reveal an entirely objective truth about the world. As such, itcame to be used to refer to the work of the Direct Cinema film-makers,who, as described above, aspired to use the new synchronous soundtechnology to reduce their effect on the action that they were filming toa minimum, so as to provide their audiences with an account of theworld that was as objective as possible. What exactly was meant by theterm cinéma-vérité when applied to this North American group continuesto be a matter of debate, but as far as Rouch is concerned, the passagequoted above makes it quite clear that, for him at least, the denotatumof this term was never some chimerical objective or absolute truth butrather a distinctive form of truth that was particular to the cinema.

But while Rouch and Vertov may have shared this view about thenature of cinematographic reality at a very general theoretical level, atthe level of actual practice, there seems to be very little in commonbetween their respective approaches to film-making. Where the visual


aesthetic of Rouch’s films was generally realist and, once thetechnology allowed, based on the long take and a ‘normal’, progressivechronology, Vertov’s approach, at least as represented by Man with amovie camera, was based on the flamboyant use of montage and acomplete disregard for any conception of realism or a ‘normal’chronology. Another fundamental difference concerned thecircumstances of shooting. Both Rouch and Vertov laid great emphasison recording life sur le vif, that is directly as it is lived. But whereas thereis an almost voyeuristic quality to Vertov’s praxis, with the camera oftenintruding clandestinely on the privacy of its subjects, in Rouch’s case,the process of filming normally took place within a well-establishedrelationship with his subjects.

But the difference between Rouch and Vertov that is perhaps themost significant in this context concerns the precise nature of the truthmade possible by the ‘ciné-eye’. For Vertov, the term cinéma-véritéreferred primarily to the process of perceiving the world: the ciné-eyecould go anywhere and see anywhere. It could fly in the air withaeroplanes, watch from beneath as a train thundered overhead,pry intoa lady’s boudoir. In the editing suite, these images captured by the ciné-eye could then be transformed in all manner of ways: they could be cutup, speeded up or slowed down. In this way, humanity’s vision of theworld was transformed. For Rouch, on the other hand, it was not somuch the perception of the world but the world itself that wastransformed by the cinematographic process as the presence of thecamera provoked the subjects into revelatory performances that weredifferent from their normal forms of behaviour.

Rouch seemed either not to notice these fundamental differences, orto believe that they were not as significant as they might appear. For, inthe article presented at the CNRS conference, he theorises his notion ofthe ciné-trance through an extraordinary fusion of Vertovian ideasabout cinematographic reality with Songhay-Zerma ideas about soul-matter and the transformations that this undergoes when a person ispossessed by a spirit. Rouch explains that the Songhay-Zerma believethat every individual has a quality known as bia, variously glossed by theSonghay themselves as ‘reflection’, ‘shadow’ or even ‘soul’. Rouch, onthe other hand, refers to it as a ‘double’, a term often used in theanthropological literature of West Africa to describe this phenomenonwhich, in a variety of different forms, appears to be a common feature ofthe religious belief systems of the region.24 In death the double, which isimmortal, leaves the body, but even in life it can take off on its own whileits owner is dreaming or under certain other circumstances. In thecourse of possession, the adept’s double is displaced or submerged by thedouble of the spirit. While possessed, adepts are no longer themselves

94 Paul Henley

but become like the spirits that have possessed them, dancing in waysthat are suggestive of particular attributes or behaviours that areconventionally associated with those spirits. In effect then, they becomethe physical incarnation, literally, of the double of a spirit.

Rouch suggests that there is an analogy here between the condition, onthe one hand, of adepts submerged by the double of the possessing spiritand, on the other, of film-makers who carry the ciné-trance to its ultimateconclusion, becoming completely immersed in the reality they are filmingand thereby entering their own trance of creativity. In the same way thatthe Songhay adepts possessed by a spirit imagine themselves to be enteringa world that is different to that of everyday experience, so too do ‘possessed’film-makers enter a different reality when turning on the camera.Whereas the adept’s ‘double’ is taken over by the ‘double’ of a spirit, thefilm-maker is taken over by Stimmung, poetic creativity. It is this analogythat Rouch is alluding to when he refers to cinema as the ‘art of thedouble’, suggesting that, just as in the case of spirit possession, it similarlyinvolves a transition from the world of the real to the world of theimaginary (Fulchignoni 1981: 28–29, also Fulchignoni 2003: 185).

It is also here that Rouch discerns a connection with the Vertoviannotion of cinéma-vérité. He suggests that this alternative reality, this domainof poetic creativity into which the ‘possessed’ film-maker enters when inthe ciné-trance, is none other than the world of cinema truth. When thefilm-maker is in the ciné-trance, everything he or she does is determinedby the particular qualities of this distinctive world. For this reason, indescribing his own actions while in a state of ciné-trance, Rouch attachesVertovian prefixes to all the verbs. Thus when he films he ‘ciné-looks’, whenhe records sound he ‘ciné-listens’ and while editing he ‘ciné-thinks’ as he‘ciné-cuts’. In fact, he becomes totally identified with this ciné-persona:

With a ciné-eye and a ciné-ear, I am ciné-Rouch in a state of ciné-tranceengaged in ciné-filming ... That then is ciné-pleasure, the joy of filming …(Fulchignoni 1981: 8, also Fulchignoni 2003: 150).

Moreover, as this ideal state of ‘grace’ can only be achieved if there areeffective performances on both sides of the lens, his film subjects tooshould become involved in this world. Rouch claimed that since hissubjects understood perfectly well what he was doing as a result of hismany feedback screenings, they reacted to his film-making as theywould do to those who are possessed by spirits, namely by lendingthemselves to the performance on its own terms. Thus, as he ‘ciné-observes’, they allow themselves to be ‘ciné-observed’ (Rouch 1997:224–25). In the most extreme case – as he suggests may have happenedin the filming of Tourou et Bitti – in response to the film-makers’ ciné-trance, the subjects may go into their own kind of trance.


Shared anthropology and the ciné-trance reconsidered

As we have seen in this chapter, Jean Rouch’s practice of ‘sharedanthropology’ took two forms: one, relatively passive, involved ‘feedbackscreenings’ at which the subjects would provide ethnographic commentand, ideally, suggest ideas for new films: the other, more active, involveddirect collaboration in the making of the films and, in the ideal case,the orchestration of performances between film-maker and subject soas to bring to the surface of experience a privileged domain of truth.

There is certainly a common ground between the collaborativeapproaches to anthropology generally advocated by Anglo-Saxonpostmodernists from the 1980s onwards and Rouch’s idea of ‘sharedanthropology’ when considered in its relatively passive form, namely thedialogical exchanges around the feedback screenings. But in its activeform there is an extra, performative dimension to Rouch’s idea of ‘sharedanthropology’ that is absent from most Anglo-Saxon understandings of‘dialogical anthropology’. For, in the conventional Anglo-Saxonformulation, ‘dialogical anthropology’ consists of a process in whichresearcher and subjects engage in a verbal dialogue in order to arrive atsome commonly agreed construction of the subjects’ social world inwhich the political, intellectual or cultural biases that either or bothparties might have brought to the fieldwork encounter have in somesense been negotiated. In the Rouchian formulation, on the other hand,both parties are conceived as undergoing a transformation as each putson an almost theatrical performance for the other, thereby creating aparticular kind of knowledge that is a direct result of the encounter itself.Moreover, for Rouch, this orchestration of performances could occurwhether or not the researcher was armed with a camera:

Once in the field, the simple observer undergoes a change. When he is atwork, he is no longer the person who greeted the Elders at the entrance tothe village. To resort to the Vertovian terminology again, he ‘ethno-looks’,he ‘ethno-observes’, he ‘ethno-thinks’, while those before him undergo asimilar change once they have come to trust this strange visitor who keepsreturning: they ‘ethno-show’, they ‘ethno-speak’, they might even ‘ethno-think’ ... (Rouch 1997: 227)

In Rouch’s view, the knowledge that arises in such situations – far frombeing dismissed as false because it is an artifice of the encounter, as itmight be by Anglo-Saxons of a classically empiricist persuasion – shouldbe considered more privileged and more valuable than any form ofobjective, detached observation that only reveals the surface of things.

This extra dimension to the active form of Rouch’s understanding of‘shared anthropology’, to which the ciné-trance is central, can be traced,

96 Paul Henley

I would suggest, to Rouch’s early encounter with surrealism. Foralthough he may have chosen a Vertovian vocabulary to describe theciné-trance, the idea that in order to gain access to an otherwise hidden,privileged domain of truth, it is necessary to free oneself from intellectualpreoccupations, put one’s trust in chance inspiration and reconnect withsome more deep-seated, primitive well-spring of creativity has clearechoes with a number of key surrealist ideas and methods, of whichautomatic writing is merely the best known. But whether we consider thepedigree of the ciné-trance to be surrealist or Vertovian, in either case itis an idea that stems from the 1920s. Does it then represent a methodthat is still relevant to anthropology today, some eighty years later?

Any experienced documentary film-maker will probably recognise,at least to some degree, the state of mind that Rouch describes by theterm ‘ciné-trance’. Some film-makers may even be able to identify withthe joyful sense of unbridled creativity – perhaps of Stimmung orDionysian ‘grace’ – which Rouch associates with this condition. I amcertainly aware from my own experience of a similar state of mind thatcan arise when one has been filming an event over a prolonged period,particularly a long and repetitious ritual event, in a situation in whichone knows the protagonists well and feels confident of one’s role as afilm-maker. Under these circumstances, banal considerations of bothtechnique and theory can fall away, and everything can seem almostmiraculously to ‘work’, including not just one’s own handling of theequipment but also the movements and reactions of the subjects withinthe frame. A certain complicity is established between film-maker andsubjects, giving rise to a sense that both parties are conspiring toproduce a sequence that is not only well-executed, but also revealscertain things about the ideas or attitudes of the subjects that hadpreviously remained hidden.

But even though one might freely admit that the notion of the ciné-trance refers to an entirely recognisable condition, the idea that one canarrive at a privileged truth by this means is a proposition that is verydifficult for any anthropologist formed in the Anglo-Saxon tradition,myself included, to accept. The ciné-trance may indeed generateperformances on both sides of the lens that give rise to revelatory snap-shots or ‘postcards’ of the inner state of mind of the subjects. But, incommon with many other Anglo-Saxon anthropologists, I would arguethat in order to be truly valuable to anthropology these revelations cannotbe taken at face value but rather must be located and interpreted – notimmediately perhaps, but eventually – as ideological manifestations of aparticular network of social relations and cultural understandings.25

One should also be wary of uncritically saluting the politicallyprogressive nature of Rouch’s idea of ‘shared anthropology’, be it in the


active or the passive form. Some African film-makers associated withRouch, even some of those whom he supported early in their careers,have been critical of his work, considering it irredeemably colonialist,even if in a largely benign paternalist manner (Haffner 1996,Hennebelle 1996). Clearly, these views have to be understood within thecomplex entanglements of the last period of the French colonialpresence in Africa. But they should make one wary of laying too great aburden of expectation on the notion of ‘shared anthropology’. At theend of the day, although his collaborators may have played a highlyactive part in creating the films, Rouch remained the overall author: theymay have received screen credits, but the films still bore the legend ‘unfilm de Jean Rouch’. Although Rouch always preferred the term ‘film-maker’, so as to avoid the title ‘director’, with all that this implied interms of directing the activities of a team, in actual practice hiscollaborators ended up being the protagonists or the technicians, whilehe remained the director in the sense that it was he who organised thefilms, gave shape to them and distributed them afterwards. As Jean-PaulColleyn commented in his obituary of Rouch, within the inequalities ofNorth–South relationships the idea of an entirely ‘shared anthropology’,based on a genuinely collective authorship, was always going to besomething of a fiction and continues to be so, even under presentcircumstances, almost half a century after the formal end of Europeancolonialism in West Africa (Colleyn 2004: 540, also Colleyn 2005).

But even while we should recognise this in more sober moments, weshould be careful not to be too presentist in our assessment of Rouch’swork. We should not underestimate the hurdles of both a cultural andpolitical nature that Rouch had to overcome to make collaborative filmswith Africans of relatively marginal status in the still-colonial era ofthe 1940s and 1950s. Nor should we forget that the idea ofsurrendering any degree of authorship to the subjects of study was farahead of the practice of the majority of even the most progressive ofhis anthropological contemporaries, in both France and the Anglo-Saxon world. In short, even if Rouch’s ability to practise a fully sharedanthropology was limited by the particular conjuncture of politicalconditions under which he was working, this does not diminish thechallenge that his example still poses to the film-makers of today.

98 Paul Henley

Notes

1. See Forbes (1996). By ‘Anglo-Saxon’, the French generally mean those who are bothEnglish-speaking and of European extraction, i.e. primarily people from the US, Britain,southern Africa and Australasia. This classification does not take into account thepersonal, cultural or ethnic affiliation of individuals, so someone who is, say, a ScottishJew would also be ‘Anglo-Saxon’. In the context of academic anthropology, even thePolish-born Bronisław Malinowski can be placed in the category ‘Anglo-Saxon’ becausehe was European, mostly published in English and spent most of his career in English-speaking countries. I shall retain the term here in that sense.

2. Mick Eaton’s early edited collection of articles played an important role in introducingRouch to English-speaking audiences (1979) as did the special editions of the journalsStudies in visual communication vol. 11, no. 1 (1985) and of Visual anthropology vol. 2,nos. 3–4 (1989), edited by Steven Feld and Jay Ruby respectively, and, somewhat later,Paul Stoller’s monograph, The cinematic griot (1992).

3. See, for example, Clifford (1988a, 1991), Price and Jamin (1988), Richardson(1993), Thompson (1995), Douglas (1995), de Huesch (1995), Bate (2007) and alsoRouch (1995c).

4. See Rouch (2003b: 110). Rouch often referred to the Griaule’s ‘smile’ and oncecompared studying ethnography with him as being like a game of blind-man’s-buffthat made him laugh so much, it almost made him ill (Rouch 1989: 10).

5. Piault (2000: 114–16). Griaule was also a very talented photographer, as evidencedby the magnificent photographs in the appendix to his major work, Masques dogons. Hehad also played an active part in the development of aerial photography (Bate 2007).

6. In contrast to the ambiguity in his attitudes to Griaule, Rouch always retained thehighest regard for Germaine Dieterlen. When they were both in Paris, they spent agreat deal of time in one another’s company, particularly after Rouch’s first wife, Jane,died in1987. When Dieterlen herself died in 1999 at the age of 95, Rouch’s closestassociates report that he was cast into a deep depression and never quite recapturedhis celebrated joie de vivre again.

7. Dieterlen (1988). This book was first published in 1950 and was based on her doctoralthesis.

8. Rouch (1989: 320–1). These ideas concerning the primary importance of collectingethnographic ‘documents’ in as objective and rigorous a fashion as possible wouldlater come to influence an important strand of Rouch’s film-making methodology,particularly when he worked with Dieterlen in the late 1960s and early 1970s in theproduction of a series of films about the Sigui, the major ritual cycle of the Dogon(see Henley 2007).

9. See Griaule (1957). The locus classicus for Malinowski’s fieldwork method is theIntroduction to his Argonauts of the Western Pacific, first published in 1922.

10. Rouch (2003b: 112). See also the first few pages of Dieu d’eau (1988, originallypublished in 1948), in which Griaule describes the scene as he and his threecolleagues pursue their interrogatory investigations in the immediate vicinity of theirfield station, first thing in the morning on the day after their arrival.

11. Rouch (1995b: 228); see also Taylor (2003: 140–41).12. When challenged about the relative absence of women in his films, Rouch explained

that he had found it quite impossible for a European man to film African women, sincethis would not be permitted by the local people (Georgakas et al. 2003: 217).

13. For a discussion of ‘direct cinema’, see Mamber (1974) and Winston (1988, 1995);for a discussion of ‘observational cinema’, see Young (1995) and Henley (2004).

14. See particularly Griaule (1957: 59). Clifford (1988b: 80ff.) suggests that, having beeninitiated into the arcane parole claire in the latter stages of his work among the Dogon,


Griaule underwent some sort of conversion, becoming more respectful of the Dogonand considering them to be sage ‘doctors’ rather than base liars. However, Griaule’shandbook was published some ten years after his supposedly transformativeencounter with the blind hunter Ogotommêli.

15. See Jutra (1960: 40) on Rouch’s support of Tallou. Otherwise, most of thisinformation comes from the two excellent films on Rouch’s African associates: Rouch’sgang (1993) by Steef Meyknecht, Dirk Nijland and Joost Verhey; and Friends, fools,family: Rouch’s collaborators in Niger (2005) by Berit Madsen and Anne MetteJørgensen. In addition to this immediate ‘gang’, Rouch also helped the careers ofmany other African associates of his: Oumarou Ganda, Moustapha Alassane and SafiFaye, all of whom became significant figures in West African cinema, were not onlyfirst introduced to cinema by acting in Rouch’s ethnofictions but were alsoencouraged by him in their later careers to become film-makers in their own right.

16. Rouch (1995a: 96). The reference in this passage to ethnographic film-makers actingas ‘entomologists’ is probably an allusion to a celebrated remark by SembèneOusmane, the Senegalese feature film director, who, in a debate that took place in1965, claimed that Africanists such as Rouch ‘look at us like insects’ (see Cervoni1996).

17. Rouch disdained both Marx and Freud on the grounds that they were thinkers whoexploited other peoples’ dreams rather than being dreamers themselves (Taylor 2003:132; see also Rouch 1995c: 411–12).

18. Morin (2003: 230–231, 264 n. 3). There is an intriguing echo here of Malinowski’sfamous observation in his methodological preface to Argonauts of the Western Pacificthat it was through his ‘plunges into the life of the natives’ that he discovered that‘the behaviour, their manner of being ... became more transparent and easilyunderstandable than it had been before’ (1932: 22).

19. The article was first published in the proceedings of the conference in 1971.Convenient republications are to be found in the second edition of Rouch’s majorwork on Songhay religion (Rouch 1989: 337–49) and in the more recent collectionof his ethnographic essays (Rouch 1997). An abbreviated version is added to hisinterview with Enrico Fulchignoni, published in 1981 (Fulchignoni 1981). Both theoriginal article and the Fulchignoni interview have been translated and republishedby Steven Feld (see Fulchignoni 2003: 182–85, Rouch 2003a).

20. In the film commentary, Rouch refers to sauterelles, grasshoppers, as causing theproblem, though in describing the film elsewhere he says that it was chenillesprocessionnaires, army caterpillars (Rouch 1989: 185–86n). It is only too likely thatthe unfortunate villagers were suffering from both.

21. Rouch (2003a: 101). But see also his somewhat more sceptical discussion in Colleyn(1992: 41–42).

22. Rouch (1989: 186n). Rouch explains that this term was used by such diverse artisticfigures as the early nineteenth-century German Romantic poet Friedrich Hölderlin,the philosopher Friedrich Nietzsche and the early twentieth-century surrealist artistGiorgio de Chirico, creator of the image that had been part of Rouch’s ownDamascene encounter with anthropology. See also Fulchignoni (2003: 186).

23. Rouch (1968, 1995a: 82–83). By the 1960s, Vertov was a largely forgotten figure,even though he had died only in 1954. However, the publicity given to his ideas by theFrench cinema historian Georges Sadoul served to revivify an interest in his work (seeSadoul 1963).

24. See Rouch (1989, especially pp. 38–39), also Stoller (1995, passim).25. Although I have attributed these sceptical views, obviously stereotypically, to Anglo-

Saxons, it was evident at the conference out of which this volume arose that many ofthe French participants had similar reservations about this aspect of the Rouchian

100 Paul Henley

approach. Equally, there are certain ‘Anglo-Saxon’ anthropologists who mightdisagree with the view that I propose here: see, for example, Paul Stoller’s argumentsabout Rouch as a radical empiricist (1992: 202–18).

References

Bate, D. 2007. Everyday madness: surrealism, ethnography and the photographic image,in J. ten Brink (ed.), Building bridges: the cinema of Jean Rouch, London: Wallflower Press.

Blue, J. 1996. Jean Rouch: interviewed by James Blue, in K. Macdonald and M. Cousins(eds), Imagining reality: the Faber book of documentary, London and Boston: Faber andFaber. Originally appeared in Film Comment 2 (2), 1964

Cervoni, A. 1996. Une confrontation historique en 1965 entre Jean Rouch et SembèneOusmane, in R. Prédal (ed.), Jean Rouch ou le ciné-plaisir: CinémAction, 81: 104–6.Éditions Corlet.

Clifford, J. 1988a. On ethnographic surrealism, in J. Clifford, The predicament of culture: twentieth-century ethnography, literature, and art, Cambridge, MA: Harvard University Press.

——— 1988b. Power and dialogue in ethnography: Marcel Griaule’s initiation, in J.Clifford, The predicament of culture: twentieth century ethnography, literature, art,Cambridge, MA: Harvard University Press.

——— 1991. Documents: a decomposition, Visual Anthropology Review, 71: 62–83.Colleyn, J.-P. 1992. Jean Rouch, 54 ans sans trépied, CinémAction, 64: 40–50.——— 2004. Jean Rouch, presque un homme-siècle, L’Homme, 171–172: 537–42.——— 2005. Jean Rouch: an anthropologist ahead of his time, American Anthropologist,

1071: 112–15.de Heusch, L. 1995. Pierre Mabille, Michel Leiris, anthropologies, in C.W. Thompson (ed.),

L’Autre et le Sacré: surréalisme, cinéma, ethnologie, Paris: L’Harmattan.Dieterlen, G. 1988. Essai sur la religion bambara (2nd ed), Brussels: Éditions de la Université

de Bruxelles.Douglas, M. 1995. Réflexions sur le renard pâle et deux anthropologies: à propos du

surréalisme et de l’anthropologie française, in C.W. Thompson (ed.), L’Autre et le Sacré:surréalisme, cinéma, ethnologie, Paris: L’Harmattan.

Eaton, M. (ed.). 1979. Anthropology, reality, cinema: the films of Jean Rouch, London: BritishFilm Institute.

Fieschi, J.-A. and A. Téchiné. 1967. Jean Rouch: ‘Jaguar’, Cahiers du Cinéma, 195: 17–20.Forbes, J. 1996. Jean Rouch et la Grande-Bretagne, in R. Prédal (ed.), Jean Rouch ou le ciné-

plaisir: CinémAction, 81: 136–37. Éditions Corlet.Fulchignoni, E. 1981. Entretien de Jean Rouch, in P.-E. Gallet (ed.), Jean Rouch: une

rétrospective, Paris: Ministère des Relations Extérieures and Centre National de laRecherche Scientifique.

——— 2003. Jean Rouch with Enrico Fulchignoni: ciné-anthropology, in S. Feld (ed.),Ciné-Ethnography, Minneapolis and London: University of Minnesota Press.

Georgakas, D., U. Gupta and J. Janda. 2003. The politics of visual anthropology, in S. Feld(ed.), Ciné-Ethnography, Minneapolis and London: University of Minnesota Press.Reprinted from Cinéaste, 8: 4 (1978).

Griaule, M. 1957. Méthode de l’ethnographie, Paris: Presses Universitaires de France.——— 1985. Dieu d’eau: entretiens avec Ogotemmêli, Paris: Fayard. First published in

1948.Haffner, P. 1996. Les avis de cinq cinéastes d’Afrique noire, in R. Prédal (ed.), Jean Rouch

ou le ciné-plaisir: CinémAction, 814: 89–103. Éditions Corlet.


Henley, P. 2004. Putting film to work: observational cinema as practical ethnography, inS. Pink, L. Kurti and A.I. Afonso (eds), Working images: methods and media inethnographic research, New York and London: Routledge.

——— 2007. Jean Rouch and the legacy of the ‘pale master’: filming the Sigui, 1931–2003, in J. ten Brink (ed.), Building bridges: the cinema of Jean Rouch, London:Wallflower Press.

Hennebelle, G. 1996. Jean Rouch et l’éthique du cinéma ethnographique, in R. Prédal(ed.), Jean Rouch ou le ciné-plaisir: CinémAction, 81: 76–79. Éditions Corlet.

Jamin, J. 1991. Anxious science: ethnography as a devil’s dictionary, Visual AnthropologyReview, 7(1): 84–91.

Jutra, C. 1960. En courant derrière Rouch I, Cahiers du Cinéma, 113: 32–43.Malinowski, B. 1932. Argonauts of the western Pacific: an account of native enterprise and

adventure in the archipelagos of Melanesian New Guinea, London: Routledge and KeganPaul.

Mamber, S. 1974. Cinema verite in America: studies in uncontrolled documentary, Cambridge,MA and London: MIT Press.

Morin, E. 2003. Chronicle of a film, in S. Feld (ed. and tr.), Ciné-Ethnography, Minneapolisand London: University of Minnesota Press.

Piault, M.-H. 2000. Anthropologie et cinéma: passage à l’image, passage par l’image, Paris:Éditions Nathan.

Prédal, R. 1996. Bibliographie, in R. Prédal (ed.), Jean Rouch ou le ciné-plaisir. CinémAction,81: 227–36.

Price, S. and J. Jamin. 1988. A conversation with Michel Leiris, Current Anthropology,29(1): 157–74.

Richardson, M. 1993. An encounter of wise men and cyclops women, Critique ofAnthropology, 13(1): 57–75.

Rohmer, E. and L. Marcorelles. 1963. Entretien avec Jean Rouch, Cahiers du Cinéma, 144:1–22.

Rouch, J. 1956. Migrations au Ghana (Gold Coast): enquête 1953–1955, Journal de laSociété des Africanistes, 26: 33–196.

——— 1968. Le film ethnographique, in J. Poirier (ed.), Encyclopédie de la Pléiade:ethnologie générale, Paris: Gallimard.

——— 1989. La Religion et la magie songhay (2nd ed.), Brussels: Éditions de la Universitéde Bruxelles.

——— 1995a. The camera and man, in P. Hockings (ed.), Principles of visual anthropology(2nd ed.), Berlin and New York: Mouton de Gruyter.

——— 1995b. Our totemic ancestors and crazed masters, in P. Hockings (ed.), Principlesof visual anthropology (2nd ed.), Berlin and New York: Mouton de Gruyter.

——— 1995c. L’autre et le sacré: jeu sacré, jeu politique, in C.W. Thompson (ed.), L’Autreet le Sacré: surréalisme, cinéma, ethnologie, Paris: L’Harmattan.

——— 1997. Essai sur les avatars de la personne du possédé, du magicien, du sorcier, ducinéaste et de l’ethnographe, in J. Rouch, Les Hommes et les dieux du fleuve: essai ethnographiquesur les populations songhay du moyen Niger 1941–1983, Paris: Éditions Artcom.

——— 2003a. On the vicissitudes of the self: the possessed dancer, the magician, thesorcerer, the filmmaker, and the ethnographer, in S. Feld (ed.), Ciné-Ethnography,Minneapolis and London: University of Minnesota Press.

——— 2003b. The mad fox and the pale master, in S. Feld (ed.), Ciné-Ethnography,Minneapolis and London: University of Minnesota Press. Originally published inFrench in 1978.

Sadoul, G. 1963. Actualité de Dziga Vertov, Cahiers du Cinéma, 144: 23–34.Stoller, P. 1992. The cinematic griot: the ethnography of Jean Rouch, Chicago and London:

The University of Chicago Press.

102 Paul Henley

——— 1995. Embodying colonial memories: spirit possession, power, and the Hauka in WestAfrica, New York and London: Routledge.

Taylor, L. 2003. A life on the edge of film and anthropology, in S. Feld (ed.), Ciné-Ethnography, Minneapolis and London: University of Minnesota Press. Originallypublished in Visual Anthropology Review, 71: 92–102 (1991).

Thompson, C. 1995. De Buñuel à Rouch: les surréalistes devant le documentaire et lefilm ethnographique, in C.W. Thompson (ed.), L’Autre et le Sacré: surréalisme, cinéma,ethnologie, Paris: L’Harmattan.

Winston, B. 1988. Direct cinema: the third decade, in A. Rosenthal (ed.), New challengesfor documentary, Berkeley and London: University of California Press.

——— 1995. Claiming the real: the documentary film revisited, London: British FilmInstitute.

Young, C. 1995. Observational cinema, in P. Hockings (ed.), Principles of visualanthropology (2nd ed.), Berlin and New York: Mouton de Gruyter.

Chapter 4

ERIC DE DAMPIERRE AND THE ARTOF FIELDWORK

Margaret Buckner

The aim of this chapter is not to circumscribe the work of Eric deDampierre – that would be a task too daunting for this author – but toshine a light on some of its various aspects, especially those that arerelated to his fieldwork in Africa. In order to limit the danger of reducinghis thought, I will use his words – albeit translated into English – to lethim speak for himself. My personal acquaintance with Dampierre beganin 1982, when I was a Peace Corps volunteer stationed in Bangassou,Central African Republic, the town where Dampierre carried out hisfieldwork. By the time my three-year Peace Corps stint was up, I hadenrolled at the University of Paris-X (Nanterre) to pursue graduatestudies in ethnologie under his direction.

Early life

Eric de Dampierre was born on 4 July 1928 to an aristocratic family.His father was French, his mother Belgian, and he had an older sister.He must have been very gifted as a student, for he graduated fromsecondary school as a bachelier in philosophy when he was sixteen yearsold, received his license ès lettres (the equivalent of an AmericanBachelor of Arts) at age eighteen and a second license en droit at agenineteen, and then graduated from the l’Institut d’études politiques ofParis (‘Sciences Po Paris’) as a twenty-year-old. He was a prolific reader,not only in French, but also in English, German, Spanish, Italian, Latinand Greek. He read literature, the classics, philosophy, history, sociology,

Figure 4.1. Eric de Dampierre at Madabazouma, thirty kilometres from Bangassou,Central African Republic. Date unknown. Courtesy of Laboratoire de l’ethnologie et dela sociologie comparative (UMR 7186, CNRS), Université Paris-Ouest.

political science and anthropology, among others. In 1948–49 hecompleted his compulsory military service, first in Casablanca, then inVillacoublay (France).

At twenty years of age, he published his first article, ‘Sociométrie:note étymologique’ (Dampierre 1948). The paper explored the originof the word ‘sociometry’, which a certain Dr Moreno claimed to havecoined in 1943 (Moreno 1943). Dampierre, however, traced it to an

104 Margaret Buckner

Austrian, F.X. de Neumann-Spallart, who used the term during asession of the International Institute of Statistics in Rome in 1887.Dampierre then cited August Chirac, who said he had invented the wordfirst, and who developed his ideas in a published article (Chirac 1897).This first publication, at age twenty, showed Dampierre’s extremelyconscientious use of terms, and also his meticulous care in finding andcritiquing the original sources. He continued to trace words andconcepts even – or rather, especially – after he went to Africa. Forexample, in 1984 he wrote an extremely detailed study of the wordnguinza, now meaning ‘money’, concluding it was probably brought toCentral Africa by Senegalese riflemen (Note de recherche n° 17).1

By 1949, he had become involved with UNESCO, probably thanks tohis acquaintance with Alfred Métraux, who at that time was director ofthe Department of Social Sciences at UNESCO. Dampierre participatedin an interdisciplinary study of a French commune in the Paris suburbsand he wrote the report in 1949; it was published in 1956 under the title‘Malvire-sur-Desle: Une commune aux franges de la région parisienne’(Dampierre 1956). The study resulted from a conference held at theRoyaumont Abbey in September of 1948 on the comparative method insocial sciences, and a follow-up meeting in May 1949. The researcherswere trained in history, sociology, philosophy, psychology and socialpsychology. The report is a classic example of community studies carriedout at the time. It describes ‘Malvire-sur-Desle’ (a fictional name) in allits social, economic and political complexity, and shows how the variousfactions in the community interacted – or not. In the methods section,Dampierre explains that the research team basically moved into thevillage, frequented cafes and bars, attended religious services, went to themovies and dances, and helped organise local festivals. Two months later,when the study officially began, residents were accustomed to seeing theresearchers, and were inviting them to their homes for meals. Thesociologists worked both with written documents and by carrying outethnographic fieldwork. Dampierre, in this report and in others, does nottreat sociology any differently from anthropology, and shows that goodfieldwork can and should be carried out in both fields.2

In 1950, at age twenty-two, he left for the United States as anExchange Fellow at the University of Chicago, where he was a memberof the Committee on Social Thought. There he interacted with scholarssuch as Leo Strauss, Robert Redfield and John Nef. He also rubbedshoulders with the anthropologists there and was impressed by the four-field approach of American anthropology, as opposed to the divisions inFrance between ethnology, (physical) anthropology and prehistory.

One of his earliest manuscripts, dated September 1951, perhapswritten while he was in Chicago, is entitled ‘Sur deux différents typesd’hérétiques’ (On two different kinds of heretic). In stark contrast to all

Eric de Dampierre and the art of fieldwork 105

of his other papers, this one has not a single footnote, citation orreference. Thus Dampierre writes:

Man, the first animal to want to discover his place in the universe and tosearch relentlessly for the meaning of his existence, uses in that search twotypes (Ideal-Typus) of thinking: dogmatic thinking and scientific thinking. ...In dogmatic thinking, truth is already there at the start, before it isdiscovered by man. It could be a revelation from God, ... a rational essence,... or the meaning of history. In all these cases, man needs the key to thetreasure; it is either given to him, or he must find it, or he must make it. Inscientific thinking, there is no dogma. To continue with the metaphor, manmust seek not the key to the treasure, but the treasure itself, though it neverappears to him immediately. He constructs it himself, by abstracting it fromreality and mentally organizing it. That is scientific theory. While dogma isthe truth that is given to me, science is the truth that I verify, and thus thatI create. ... Dogma is by its very essence unchangeable, perfect, and finite ...The dogmatic heretic is burned at the stake. [Progression is possible, but itis progression in the revelation of the truth.] ... Science is, on the contrary,imperfect, cumulative and infinite. The scientist does not seek to translatereality, he abstracts it to master it. ... He invents concepts, and the richer hisimagination, the stronger his power of abstraction, the better his theory. ...Progress is inevitable. ... The scientific heretic inaugurates new theory. ...The heretic of yesterday is the doctor of tomorrow. (1951, passim)

Thus, well before Robin Horton and others, Dampierre described in his ownway how a traditional worldview differs from modern scientific thinking.

Back in France in 1952, he resumed his studies and research underthe direction of the sociologist Raymond Aron. Aron ‘defines the aimproper to sociology as the combination and reunion of the study of thepart with the study of the whole’ (Aron 1968: 10). It was this aim thatDampierre followed. He was interested in how each society organisesitself, based on its own principles, and in how societies hold together, howall the different participants play their respective roles. His goal was tounderstand Society by discovering how societies work, how individualsform a society, what holds the group together and what keeps it going.

In 1952, Dampierre became a researcher for the CNRS (Centrenational de recherche scientifique) and was assigned to the Centred’études sociologiques in Paris. In that same year, he launched andedited for Plon, a well respected publishing house in Paris, the seriesRecherches en sciences humaines. Over the next twenty years or so, a totalof thirty-three books were published in the series under Dampierre’sdirection, which included the first French translations of such scholarsas Max Weber and Leo Strauss.

And so, by the time he made his first trip to Africa in 1954 at agetwenty-six, he was extremely well read in several languages and in


several disciplines, had earned several university degrees, had publishedarticles and edited journals and books, had hobnobbed with leadingsocial scientists in Chicago, had worked closely with sociologists andAfricanists in Paris (for example, Michel Leiris, André Schaeffner andDenise Paulme) and had carried out interdisciplinary projects in the field.

Fieldwork in Central Africa

In the preface to Un Ancien Royaume Bandia (1967), Dampierre writesabout his first mission to Bangassou, a town in the east of what is nowthe Central African Republic:

In 1954, a French government agency asked myself and a colleague to gofind out why the Nzakara, who were thought to have been dying out over thepast half-century, were having so few children. It was a time whengovernment administrators, who often had difficulty posing the rightquestions, still didn’t know the answer. It was also a time when sociologists,unaware of their limits, wouldn’t think of turning down an opportunity towork, no matter how uncertain the resources of their discipline.

Nzakaraland had never known researchers before us. It is no longer verycommon to be the first researchers to arrive anywhere in Africa. We had toscout things out before we could build a research project. That first year, wenever took our boots off. And we still had to answer the question. It is not toofar-fetched to suggest that there is no greater, more difficult or complex problemto address, whose meaning escapes us so mercilessly, than the problem thattouches the meaning that humans attach to giving life. At the end of our firstperiod of fieldwork (1954–1955), we were not able to satisfy those who hadsent us, and, besides, their interests had already shifted to other matters.

Whether the difficulties were resolvable or not, we learned much byaddressing them. Moreover, I developed respect for the way this old Zandesociety worked and affection for its political warriors, its intrepid poets, itswitches, its noble diviners. (1967: 11)

Dampierre was hooked from the start, and returned to Bangassou in1957–1958. He surmised that the Nzakara were in no way practisingcollective suicide by refusing to have children, as some had suggested.He asked that a physician go to Bangassou to explore medical reasonsfor the drop in natality. That physician was Dr Anne Laurentin, who in1960–61 discovered that venereal diseases were probably causinginfertility. She also took up ethnographic studies of her own.

That first trip to Bangassou, commissioned by ORSTOM (Office de laRecherche Scientifique et Technique d’Outre-Mer), was the maiden



mission of the MSHO (Mission sociologique du Haut-Oubangui orSociological Mission of the Upper Ubangui). The EPHE (Ecole pratiquedes hautes études) and then the CNRS financed successive periods offieldwork for the MSHO in 1957–1958 and 1964–1965, then annuallyfrom 1966 to 1979, and again annually from 1981 to 1987. Dampierrealso went to Bangassou in 1960–1961. The MSHO had an office in thebasement of the Musée de l’Homme in Paris, next to the office of MichelLeiris and other Africanists. In the early 1980s, the MSHO joined theLaboratoire d’ethnologie et de sociologie comparative on the campus ofthe University of Paris X-Nanterre. The MSHO had a post office box(number 98) at the Bangassou post office at least until the mid-1990s.

Dampierre established a research station on the outskirts of thetown of Bangassou, at the home of a former plantation owner, Godeste,which also became the name of the station. The house, which sat in aclearing surrounded by forest, was made of stone, with a thatched roof;it had two large rooms and a small annex for washing. About twentyyards from the house, there was a smaller, round house that had a sparebedroom and a large shady porch that served as dining area. The stationbecame Dampierre’s second home, and he returned regularly forperiods of several months until the late 1980s. He more or less adoptedan extended Nzakara family to help run the research station, includinga housekeeper, a cook, a driver, a mechanic, a groundskeeper and a fewothers. He would stop in at their village on the way to Bangassou, loadthem all into his Land Rover and drive them to Godeste, where theywould make themselves at home for the season. An aristocrat, he feltat home among the class-conscious Nzakara, a people who shared hisinterest in making living an art.

With Godeste as his home base, Dampierre travelled the length andbreadth of Nzakaraland, stopping in at villages and getting to knowespecially the elders who lived there. He also made excursions intoZandeland, going at least as far east as Mboki. On a few occasions, hewould bring elders and musicians to the Godeste research station. Onesuch festival of music and poetry took place on 19 November 1971; hesupplied transport, room, board and drink to several renowned Nzakaraharpist-poets so they could all relive the music, language and courtlyambience of yore.

Dampierre involved as many researchers as possible from otherdisciplines and backgrounds, inviting them to spend time at Godeste.The list of guests and colleagues includes botanists, a geologist, amusicologist and a linguist, as well as other anthropologists. Butfieldwork was not enough. He had to join in people’s lives and buildrelationships; people were not just objects of study but collaborators,friends and family, and he felt quite comfortable among them. In an


earlier publication (Dampierre 1956), he had listed getting to know thelocals and participating in village life as a technique to gain betterknowledge about them. In Africa, it no longer seemed that establishinga warm relationship with one’s fellows was a means to an end; itbecame, instead, an end in itself. He was already ready to help, bothmonetarily and logistically; in emergencies his Land Rover often servedas the local ambulance. He taught sporadically at the lycée atBangassou; he encouraged students there to pursue studies in sociologyand ethnology, and sought financial backing for them to continue theirstudies in Bangui and eventually Paris.

Fieldwork philosophy

For Dampierre, ethnology and sociology are really one and the same. Hestudied French villages and Nzakara villages using the same techniquesand methods: a combination of historical documents, interviews andconversations with local people about their past and present, anddetailed observation of behaviour, practices, and institutions. All three(historical documents, oral history, ethnography) reinforce each other.

He had always been interested in social dynamics, in how and whysocieties change over time. To understand social and cultural processesand dynamics, a historical perspective is essential. Many, if not most, ofhis descriptions of Nzakara society of the nineteenth and early twentiethcentury are based on historical documents. His major work, Un AncienRoyaume Bandia, has a 70–page review of historical sources (Étude critiquedes sources). But he combined historical accounts with an understandinggained from living among the Nzakara, learning their language, listeningto their poets, learning their proverbs and observing their customs andtraditions. In his dissertation defence, Dampierre explains:

The method I used in this work is perhaps not completely recommended.Our British colleagues, who rightly insist that the ethnologist should observebehavior rather than listen to what people say or read old texts, warn us notto read the past into the present or read the present into the past. But that isexactly what I have done, while taking special precautions. The first, andthe most important, is to do fieldwork before reading historical documents.One is often surprised to find, after five or ten years, new meaning indocuments that at first seemed absurd, wrong, or crazy. One must, of course,make the necessary transpositions, and, from the very beginning,understand how such or such behavior would have appeared to theinnocent explorer or administrator. To that first effort I have added a second:to treat the European context and the African context in counterpoint. Thecontrasting interpretations of the treaty between [King] Bangassou and


Vangele [Belgian explorer] provide a good example. Furthermore, one mustspend a long time in a place in order to grasp the African reality behind theadministrative accounts (1968: 1–2).

He wove together written history and oral history to piece together thepast, to sketch out past events that led to current social organisation. Hehad faith that oral history, properly gathered and interpreted, could bemore reliable than second-, third-, and fourth-hand accounts recordedby European travellers and administrators. In a very slim book, Desennemis, des arabes, des histoires (1983), Dampierre refutes the generallyaccepted historical account that the Arab slave-trader Rabih invadedthe Nzakara kingdom twice, vanquished the royal army and pillagedthe territory. He demonstrates that skilfully gathering oral traditionsyields better results than consulting the frequently erroneous accountsof European explorers. His concluding paragraph summarises the roleof the ethnographer or oral historian:

I have attempted here, after critically reviewing the sources, as all goodhistorians should, to reconstruct the collective experience of partialtestimonies, scattered in space and time, and to understand that experiencethrough its reflection in the mirror of the outsider. In a society withoutwriting, asking piecemeal questions in privacy gets only useless informationor answers that most please the interrogator. That is why continuouslyquestioning the elders can only be useful over a lifetime and done in public.One must learn to get old. Contrary to what one often feels obliged to write,bringing that experience to the surface has nothing to do with tradition. Thatvery action, for the society that wants or accepts it, can actually preservetradition. We need to know how to use tradition to uncover what refutes it.Such is the work of the mandrels, those modest intermediaries. (1983: 41)

Like Nzakara poets, Dampierre uses a metaphor to describe his role inthe process: that of a mandrel. A mandrel is a cylindrical, rotating shaftthat serves as an axis for a larger rotating part. He saw himself as a toolallowing the various partial memories of Nzakara experience to take shapein a coherent history, thus enabling the Nzakara to solidify their tradition.

Dampierre emphasised his point about the necessity of long-termfieldwork by offering a counterexample, in the form of an epigraph, onthe same page: ‘We were so successful that at the end of two hours, thePygmy had been sketched, measured, feasted, showered with gifts andsubmitted to a detailed interrogation’ (Schweinfurth 1875: 113).

Using documents and texts from individual perspectives is somethinghe had long thought about. In an early publication, ‘Le sociologue etl’analyse des documents personnels’ (1957), Dampierre proposed thatusing ‘personal documents [autobiographies, personal letters, diaries,drawings, unguided interviews faithfully transcribed], for the same time


and effort, provide much richer material than most other techniquesused in social science’ (1957: 444). One must remember, he continued,that personal documents may not be sincere, may not be pertinent andmay be affected by direct stimulation or solicitation. They also representonly one point of view. But they can be useful to ethnologists who aretrying to reconstruct an indigenous culture from the inside. In the samepaper, Dampierre offers instructions on how to use such documentsfruitfully, while avoiding the traps. In the field, he collected many kindsof personal documentation, both written (for example, by school-ageZande refugees living at the refugee camp in Mboki) and tape-recorded(especially life histories and historical narratives).

Dampierre learned the language – one of the first Frenchethnographers to do so – and he learned it well. People told me that hespoke Nzakara like an elder, which was a compliment indeed. Dampierrewas versed in phonology and linguistic theory, and he set aboutcompiling a Nzakara dictionary. His understanding of the languageallowed him not only to question and converse with the Nzakara butalso to pay attention to people’s unsolicited, spontaneous comments,to how they formulate their ideas and to their choice of phrases andfigurative expressions. Many of his findings are based on what he heardpeople say, as well as what they never said. For example, whendiscussing musical instruments, he states, ‘To have been made by achild is a way for an object not to exist’ (1995: 68); in other words,saying that something was made by a child is saying that it is irrelevant.

Besides being a means to an end, Dampierre also considered thelanguage as an end in itself. The language serves as a window into theculture. He used features of the language to support his ideas about theNzakara mode of thought and aesthetic. In a paper written in 1983(Note de recherche n°12), he exposed the Nzakara catégories del’entendement in a structured table similar to studies of Greek categories.We would spend hours discussing possible English, French and othertranslations for Zande words, always flitting through the pages of theLalande Vocabulaire technique et critique de la philosophie.

Dampierre was drawn to the poetry and music of the Nzakara andZande harpists. He travelled the length and breadth of Nzakara countryto record the best harpists, lugging the Nagra, and later the Uher, taperecorders, along with microphones, batteries, cables, extra reels and soon. He often encountered obstacles, such as technical malfunction,poor weather, illness, absences and funerals. He learned Nzakara poetryinside out, the ‘double speech’, the allusions, the idiomatic expressions,the figures of speech. Many of the examples in the dictionary he wascompiling come from the poems. Throughout his books and articles, herefers to snatches of song that, through allusions and proverbs, reveal


the inner workings of Nzakara society. Dampierre himself usedlanguage poetically. It is evident in his translations of ‘oral literature’and texts, as well as in his writings. He published two volumes ofNzakara poetry (1963, 1987),3 which he rendered into French poetry.

As anyone who worked with him knew, Dampierre was extremelyobservant, down to the finest details. Nothing escaped his interest. Hisobservations were eclectic, all-inclusive, involving every facet of discourseand behaviour, from greetings to building houses to playing the harp tofighting battles. No person, practice, technique or word was irrelevant.He talked with all members of society, commoners and nobles, young andold, men and women; studying them in isolation is pointless, for theirinteraction is what holds a society together. Though he had tremendousadmiration for Evans-Pritchard, Dampierre bemoaned the fact that hehad not been able to include women in his fieldwork.4

He sought guiding principles that held through different practices andinstitutions. For example, in Note de recherche no. 1 (1974), he arguesthat playing kisoro, a Zande and Nzakara board game, actually re-enactsthe strategy and tactics of Zande armies as described by Evans-Pritchard(1971). The game board consists of four rows of eight holes; eachadversary has two rows and thirty-two ‘men’. Among the principles:territory is never conquered by force. Victory belongs to the side thatweakens the enemy to the point of having no more warriors. Both sidesplay at the same time; there is no handicap at the start of a battle, andeach side has an equal chance to win. The manoeuvres are parallel, buteach adversary moves his men independently of the other’s movements.Captured men are immediately incorporated into the captor’s army.Strategy involves taking advantage of the imbalances of the opposingarmy in order to capture the most men, while at the same time notexposing one’s own army to attacks of the adversary. The best tactic is tomove the most men quickly, which tends to re-establish the startingpositions by redistributing one’s army, and to take the most men toincrease the size of one’s army. Thus, studying the way the Nzakaraplayed kisoro also involved studying Nzakara and Zande military tactics.

Similarly, he saw an analogy between the keys on a sanza (‘thumbpiano’, an idiophone) and kinship (1982). As he watched a sanza-playerwork, he asked him questions and listened thoughtfully to his responses.Through these technical conversations there gradually emerged arepresentation of the sounds and the scale. The series of keys is calledthe lineage. The bridge (chevalet) is called ‘the mother-in-law carryingchildren’. The six keys on one side are all slightly lower than the six keyson the other side: the elder and younger lineages. The keys are named:the fathers (one on each side), the mothers (three on each side,including the favourite, the head wife, the ‘big wife’ who works for the


head wife, and two lesser wives; then four children, in birth order.Gilbert Rouget (1982) then studied the intervals between the key’snotes, and found that key number 7, the ‘big wife’, the one who alwaysdoes whatever she likes, also sticks out musically, being asymmetrical.

Political organisation of the Bandia kingdoms

Not surprisingly, since he was a graduate of the Institute of PoliticalScience, Dampierre’s early research focused on the politicalorganisation of the Bandia kingdoms in the upper Ubangui and Uelebasins. His main question was the social foundations of politicalauthority. He was well acquainted with Max Weber’s three types oflegitimate rule: legal or rational authority, charismatic authority andtraditional authority (Weber 1958 [1922]). He seemed to use theBandia kingdom in Nzakaraland as a living example of Weber’straditional authority. He was especially interested in how the Bandiaclan, foreigners in Nzakaraland, established their political power, and inhow they made the Nzakara need them. He addressed this question inhis dissertation defence:

Every once in a while, we see appear in history what historians call a militaryautocracy. Not long ago, P. de Vaux (1967) described the secret of theHorites in Genesis: ‘Once they infiltrated Palestine they seized power in theprincipal Palestinian cities and, without imposing their language or theircustoms, they quickly assimilated into the native populations.’ Those arepeople who resemble our Bandia. But how did it happen? By whatmechanisms, by what needs, by what liberties? Can one truly explain powerwithout analysing dependence? For power is in part violence, and theexercise of violence, like the exercise of war, is not an easy object ofsociological study. Looking through the other end of the telescope is, I feel,more fruitful. How are the bonds of dependence in a given society woven,organized and hierarchized? That indeed makes a good object of study.(1968: 3)

Evans-Pritchard (1971) had studied the political organisation of theVungara dynasties in central and eastern Zandeland. Dampierreanalysed the political organisation of the three Bandia dynasties inwestern Zandeland and Nzakaraland, and, in particular, the kingdomof Bangassou. Out of ten or so Zande kingdoms, Evans-Pritchardstudied the one farthest east (that of King Gbudwe, Vungara) andDampierre studied the one farthest west (that of King Bangassou,Bandia); between them, they thus covered the two ends of thegeographical spectrum.


Ethnically, the Zande and Nzakara are very closely related; the twogroups are so close that early explorers called the Nzakara ‘the WesternZande’. They have virtually identical kinship systems, social andpolitical organisation, and belief systems (witchcraft, oracles, magicand diviners). The Nzakara and Zande languages are still close enoughto be mutually comprehensible for some native speakers; it is estimatedthat they diverged no more than 500 years ago. Between the Nzakara-speaking kingdom of Bangassou and the western-most Zande-speakingkingdom, Rafai, the language boundary is fuzzy, with many bilinguals;there is also much intermarriage.

Though the Zande and Nzakara shared a similar social and politicalorganisation, there was a crucial structural difference between the tworuling dynasties. The Vungara were a native Zande clan who grew todominate their own people and then expanded eastward to incorporateand Zande-ise foreign peoples. Conversely, the Bandia were foreignNgbandi-speakers who came north and adopted the Nzakara and Zandelanguage and customs even as they established political domination.In a nutshell, the Vungara moved out, the Bandia moved in. Thisinversion led to further distinctions between the two dynasties. Forexample, the Vungara kingdoms were very unstable; a twenty-yearperiod saw a new set of kingdoms. The Bandia, on the other hand, hadthree very stable kingdoms. The Vungara princes, especially in thenewer, easternmost regions, were each others’ worst enemies, whileamong the Bandia there was much less royal fratricide. While the Zandekingdoms (especially the eastern ones) were made up of diverse peoples,the Nzakara and western Zande were more homogenous.

In Azande History and Political Institutions (1971), Evans-Pritchardargued for the classic progression from hunters and gatherers toagriculture, which produced a surplus; the surplus was used by theVungara for political advantage. The Vungara kings and princes, byusing permanent battalions of young warriors and the temporarylabour of adult men to work their fields, and also by receiving tributefrom the surrounding area, were able to control very large amounts offood, which they then redistributed in a way that strengthened theirauthority. The Vungara courts also assured stability, military protectionand justice for peoples who until then had been small-scale,autonomous groups. Food was given generously to feed the courtiers,the battalions and their leaders, and the people who came to the courtfor redress of wrongs or with requests of the king. Evans-Pritchardstates that the king gave bridewealth (in the form of marriage spears)to anyone who asked. He also gave wives to loyal governors, militaryleaders and others who had shown him great service or loyalty. Thenumber of subjects of a given king was directly related to the king’s


hospitality, military strength and justice. Wars were fought to acquiresubjects, who would contribute to the king’s stores and his labour andmilitary pools.

Dampierre, however, in Un ancien royaume bandia (1967) andanother publication (1971), describes a very different scenario for theBandia conquest. In ancient Nzakara (and Zande) society, lineages wereequal and wives were exchanged between lineages. No lineage was anybetter than any other. The circulation of women was strictly limited tomarriage transactions. Families exchanged sisters and became allies.Each lineage was simultaneously wife-giver and wife-taker. In thissystem, a régime de la parentèle, allies were assured (1971: 267). Whenthe Bandia arrived, they adopted the Nzakara and Zande system ofkinship and alliance, but they co-opted the system and used it to theiradvantage. The Bandia were foreigners and needed to get into the goodgraces of the local Nzakara and Zande population. They did this bysupplying wives not to their relatives but to their clients (subjects).Women went from being exchanged by lineage elders to beingdistributed by Bandia rulers. No longer was equality at the heart of theexchange. By controlling the circulation of women, the Bandiadeveloped clienteles at the expense of the traditional, egalitarian lineagesystem. ‘A surplus of women and their distribution by the dominantclan are the keys of the new system, which, though it creates allies, ismuch better equipped to create subjects’ (1967: 294–95). Gradually,allegiance replaced alliance. Residence was no longer based on kingroupings, but on client groupings.

To continue to be givers of wives, the Bandia needed a surplus ofwomen. Annual wars were fought not to expand territory or toincorporate more subjects into the kingdoms, but to bring back womenand girls to give away as wives to subjects as the Bandia pleased.Dampierre was explicit about the reasons for the wars: ‘The maximumacquisition of women became the means of government and renewedthe symbolic pomp of power’ (1967: 273). When the Europeansarrived, the well dried up. There were no more wars to capture womento distribute. And because the Europeans upheld the right of women tobe married by compensation only, the Bandia were no longer able todistribute women as needed to maintain their authority; they lost theirclients, without whom they ceased to be patrons.

A later paper, ‘Les idées-forces de la politique des Bandia à travers lespropos de leurs souverains (1870–1917)’ (1998), further contributesto our understanding of the Bandia kingdoms. In it, Dampierreexamines the kings’ own words to see how they themselves regardedtheir power. For example, King Djabir said to Commandant Francqui, ‘Icannot yet tell you which of my two sons will be designated by my people


to succeed me; certainly the best will be chosen, and whatever mypeople decide will be for the best’ (1998: 7). In his commentary,Dampierre explains that the king is chosen by a royal council, approvedby the royal family and acclaimed by an assembly of adult men. Second,King Bangassou said he is ‘the master of people, not the guardian ofborders’ (ibid.). Dampierre comments that the notion of borders istotally foreign to African political life. In a third example, KingBangassou said to Bonnel de Mézières: ‘You see the Kengu [Mbomu]River? It is big because the other streams flow into it. It is the same withmy chiefs: if they didn’t need my gifts, they would no longer come to meand I would be nothing’ (ibid.: 8). Dampierre explains that the powerof a Bandia king only becomes authority when he renounces violenceand sets about meeting the needs of those who have sworn himallegiance. In other words, the king commands only because heredistributes food, goods and especially wives.

Dampierre was as interested in the demise of the Bandia kingdoms asin their origin, and he also traced the breaking up of traditional Nzakarasociety. He once observed that the Zande, because they adopted newpractices so readily, ‘bent’ as they adapted to the modern world, whereasthe Nzakara, intent on defending their traditions, resisted and ‘broke’. Hewas saddened by the rupture he observed taking place between Nzakaraelders and Nzakara youth, especially those who went to school.

In his article ‘Coton noir, café blanc’ (1960), he describes in detailhow the introduction of the plantation system was apparently the mostimmediate source of conflicts and of the breaking up of traditionalsociety. The paysannat system (used for coffee cultivation) brought aboutimportant changes in cultural practices and modifications in thenetwork of daily social relations. Traditional grouped fields werereplaced by the strip plantation system. The new system accentuatedthe ‘injustices’ of the gendered division of labour: men had nothing leftto do, since, traditionally, women did the work in the fields. The newplantation system also upset the time frame for rotating fields:traditionally plots were planted for three or four years, then left fallowuntil the seventeenth year. Now, new plots would be cleared and plantedeach year, and would not be used again for seventeen years. Finally,Dampierre commented on a more subtle aspect of the change: apaysannat system implies the existences of paysans, or peasants, in otherwords, farmers who grow a surplus that will be sold in a market. Butthere can be no peasants until there are citizens (residents of a city);there can be no countryside without a city, since they are indissolublytied together by a double flow of exchanges. The countryside cannot dowithout the city, and the city cannot do without the countryside.Populations in the Mbomu did not yet have needs. The Nzakara did not


produce more than they needed for their own subsistence, theirtraditional obligations and their taxes. Moreover, they often left someof their crops unharvested. The products of the city reached them onlyin the form of cloth and aluminium basins. Why would they want todouble their income?

In this paper and others (for example, the chapter in Un ancienroyaume bandia entitled ‘A model pillage economy’), Dampierre analysesin detail the motives and practices of the colonial powers and the tradingcompanies, in particular the Société des Sultanats, which sucked theland dry. He shows once again that a process or situation cannot beunderstood without considering the historical context and all the actorsinvolved. In fact, in his dissertation defence (1968), he even considersthe reasons for colonisation in the first place, and especially thespecificity of the French colonial context, by citing philosophers such asRenan and colonial administrators such as Jules Ferry. The Frenchcolonist, according to Ferry (1892), ‘believes he is carrying out an act ofcivic virtue by leaving the land of his birth, and sees his motherland lessas a benefactor than as having an obligation’ (cited in Dampierre 1968:5). On the other hand, as Aron (1951: 70) says, the characteristic thatall imperialist policies have in common is that they find their origin in the‘political ambitions that chancelleries camouflaged (or rationalised) byinvoking realistic motives’ (cited in Dampierre 1968: 6).

Nzakara poetry

As mentioned earlier, Dampierre was drawn to the musical poetry ofthe Nzakara and Zande. That society’s music and oral art became asecond research theme. At first, he collected texts of the poems – sungby harpists as they played – to learn the language better and work on aNzakara glossary, but the poetry appealed to him in its own right. Eachsong is a unique event, improvised on the spot, without recognisablebeginnings or endings. The poets were often minstrels at royal courts;their social and political commentary was keen. The poems are full ofword-play, humour, irony, satire and stinging criticism veiled inmetaphor. They also express the complaints and the desires of everydaylife. Finally, they are a chronicle of court life. He published PoètesNzakara (1963) after spending many long months perfecting thetranslations of the harpists’ songs with the help of his Nzakaracollaborator Robert Bangbanzi; it was the first collection of texts to bepublished in the Nzakara language. The translations were all the moredifficult in that French and Nzakara are very different languages, and inaddition the texts were poems.


Besides the words of the songs, Dampierre bent his interest to themusic itself, and to the instrument. He spent several years trackingdown Zande harps that had found their ways into European museums.He corresponded with lute-makers, art historians and curators aroundthe world. His two last books, Harpes zande (1992) and Une esthétiqueperdue (1995), are dedicated to harps, harp music and harpists, andhave received enthusiastic reviews from international specialists.

‘Thinking in the singular’

A truly overarching theme that seemed to anchor Dampierre’s fieldworkis the Nzakara-Zande way of seeing and thinking the world in terms ofthe singular, in terms of more or less. In his 1984 book, Penser au singulier,he proposed that the Nzakara ‘pensent au singulier’ – think in thesingular. ‘Everything on earth has a singular existence. Nothing isidentical, or equal, to anything else. Each thing or being is viewed in itsdifference’ (1984: 11). The Nzakara language cannot express identity. Inother words, A and B cannot be identical, or equal, though they can besimilar. Thus, two shadows made by the same person are aberrant andsigns of disorder. No two people are identical, or equal. Among otherthings, this explains why the birth of twins – two ‘identical’ beings – issuch a disruptive event. It also explains why counting human beings isrude. It implies they are interchangeable, that each has the same valueand characteristics as another. King Bangassou knew how manybattalions he had, but not how many men fought in them. Lengths,distances, volumes and periods of time are not measured using abstractmeasurements. Distances are described in terms of days of walking, ornumber of streams crossed. Time is described by using points (sunrise,noon, sunset and so on). Quantities and surfaces are never divided intoequal parts, for there must always be a remainder. Symmetry is avoided.Besides being different, or as a result of being different, each thing or beingis ranked or ordered. Notes on a musical scale, lineages, brothers, wives... each occupies a place on an ordered scale and is thought of in thatorder. He observed this way of seeing the world in everyday life: womenselling palm oil at the market (the ‘remainder’ was their profit), dividinga piece of food, building a roof or teaching mathematics at the lycée.

Dampierre observes that ‘thinking in the singular’ permeates theZande-Nzakara aesthetic in rhythms, voice, musical scales, sculptureand performance. Harpists, singers and sculptors look for asymmetry,for individual, ‘singular’ performances, for ‘remainders’ rather than forsymmetry and regular rhythms, intervals or features. And, perhapsespecially, no two harps or performances should ever be identical.


In Une esthétique perdue (1995), one of his last publications, he callsthe Nzakara (and Zande) a ‘society of irreducible individualists’. Heemphasises that this same way of thinking in the singular influencesall Nzakara (and Zande) thought, discourse and practices. He asks: ‘isn’tan aesthetic, whatever its source, necessarily totalizing (totalisant)?’ Heseeks to describe ‘an aesthetic of the singular, which I think is at workin several areas: rhetoric, sculpture, music’ (1995: 14).

From morning to night, all Nzakara thinking heads think in terms of moreand less, of excess and deficiency (... and also elder and younger, father andson, head wife and favorite), just as pre-Socratics who would haveunderstood why Plato replaced the One and the Infinite of Pythagorus withthe One and the Dyad of the Greater and the Lesser (1995: 16).

In ‘Accord entre deux harpes, accord entre deux voix en Afriqueéquatoriale’ (Note de recherche n° 29, 1994), he relates this principle tothe voice, especially in chantefables, the cycle of Trickster tales thatalternate sung parts and spoken narrative:

Not only does the singing voice oppose the speaking voice, but it is sung atan octave of the normal speaking voice. [...] The ‘head voice’, sometimes afalsetto (throat voice, voix de faucet), is opposed to the chest voice. [...] Inchantefables, this head voice designates the intervention of a supernaturaloperator, either to transpose the action from the everyday world to a specialworld of marvels, or to bring the action back to the everyday world. [...] Headvoice and falsetto could be considered variants of an octave voice, though itis more complex. An ‘octave voice’ alone would be considered ‘the same’ asa normal voice.

I would propose the following hypothesis: with regard to Nzakara courtmusic, the conscious discussion among musicians keeps coming back to anantinomy between the ‘impossible’ unison and the ‘obvious’ antiphony(octave consonance), between the impossible same and the Other that passesfor the same. To escape this unsolvable antinomy, the best solution is theDyad of more and less, of Excess and Deficiency. (1994: 2–3)

A footnote ties the Nzakara to the Greeks (something he did often); hecites Aristotle’s comment that ‘mixed is always more agreeable thanhomogenous’.

He also tied ‘thinking in the singular’ to rhythms: there is alwayssomething left over; rhythms are staggered, they always have a gap, alag, an irregular interval.

For the Nzakara and Zande, the distinction is there, in the fine analysis of therepetitions. But distinction is not difference, distinction is not a relationship,because all true relations imply the analysis of the particular and would


reveal some level of participation of the being. We remain faced with theDyad to create formulas, fleeing all strictly equivalent relations. That notion,described in the mind of a [Greek] philosopher, is found again, in thesocieties of the Upper-Ubangui, shaping the daily, lived experience, startingwith that of musicians. (ibid.: 18)

Then, in one of his last papers, ‘Le reste épimore’ (Note de recherche n°33, revised version, 1996), he makes perhaps his clearest descriptionof ‘thinking in the singular’:

I continue searching for the basis of the practices described in Penser ausingulier and Une esthéthique perdue concerning the rejection of equal sizes,the rejection of symmetrical areas and the rejection of commensuratedurations. The rejection correlates with the emphasis put on ‘remainders’,which are not truly remainders since no exhaustive procedure to reach alimit was sought. These ‘remainders’, which make calculations troublesome,are conceived of by Zande and Nzakara as a privileged property of naturethat only human will can, in certain cases and under certain conditions,get rid of.

These practices in the Upper-Ubangui, which lead to the explicitformulation of ‘thinking in the singular’, render vain – illusory, evenscandalous by nature – all relations of identity, whatever they may be.(1996, passim)

Finally, in ‘Les idées-forces de la politique des Bandia à travers les proposde leurs souverains (1870–1917)’ (1998), Dampierre shows how thesingular had been enacted politically and how it came to an end by apolitical act. Here he summarises the important change that wasbrought about consciously by Bandia rulers:

In the 1880s, Bangassou declared by an oath before the shrine Bendo – andnot at the ancestors’ shrine – that from now on he would reign over animmense people, made up of foreigners from all over. It was a historicmoment: the oath before Bendo is the equivalent of a reform of Clisthenes.The sovereign solemnly renounces the ancestral foundations of hisauthority. He renounces treating people in the singular organized by descentand alliance, kin ties and clienteles. Lineage affiliation will no longer be theorganizing principle of his people. [...]

Bendo is the refusal of natural determinism, the subversion ofsuperdetermined order. It is the inauguration of civil society. The lineagesrebelled. The king’s own sister turned the Bendo shrines upside-down, ‘tosave the throne of her father, Mbali’.

This inversion of accepted values could only have been carried out by theking. The decision created identical subjects, and, in particular, treatedArabs, Whites, and Blacks on the same level. Bendo, the friend of women,the protector of harvests, had presided over the identification of everyone


with everyone. A universalist rule had appeared. The singular, which isignorant of the distinction between the particular and the universal, hadlived politically. (1998: 13–14)

It is this idea of ‘thinking in the singular’ that guided Dampierre’sresearch and writing since the early 1980s. The idea was born of andborne out by his observations and experiences during long, repeatedperiods of fieldwork among the Nzakara.

Dampierre’s legacy

First and foremost, thanks to his skilled, intense, long-term fieldwork,Dampierre helped preserve the history, language, knowledge and musicof the Nzakara for the Nzakara themselves. He showed that he wasconscious of that contribution in the introduction to Un ancien royaumebandia (1967):

Akabati, Zangandu, Nukusa, Kaali, Gbesende, Vugba, Sayo and others whooffered me hospitality have now died without knowing that, in talking withme, they were also writing their people’s history. Their sons, all too consciousof their past because they want to be someone else, are reclaiming thathistory out of fear of never knowing it. If, nevertheless, this book, by somedreadful trick of history, could transmit to the sons the knowledge of theirfathers, it would take its place among the uncertain fruits of those few, veryrare years in human history: those few years in which our commoncivilization – impoverished because become one, but infinitely rich in ahistory it endlessly recreates while at the same time making a project of itsfuture – encounters and immediately but impertinently relates the complexsplendour of societies that live in the present, content with their origins, butdiscovering, for the first and the last time, the face of the outsider in thehearts of their children. (1967: 12)

Dampierre laid the groundwork for us to follow. In the preface of Uneesthétique perdue, he challenges us to continue to search for what makesthe Nzakara and Zande society so distinct:

We hope that these [...] projects will be completed. As of now, it clearly seemsthat the ideas elaborated elsewhere to explain the sculpture and music ofcentral Africa do not allow us to understand the particularity/uniquenessof Zande society. That should not be surprising: different analyses fordifferent societies. It must be said that, after a century of work, the deepestworkings of this society of irreducible individualists still escape us. [...] Wemust one day undertake head-on a study of Nzakara and Zande rhetoric. Itgoes without saying that what is said must be qualified by the discourse


situation and context. But we must also ask ourselves a less obviousquestion: can an entire society wish daily for the same and its contrary inorder to escape the vicissitudes of human life? [...] This edited collection,preceded and followed by other works, is only a milestone on a very difficultroad, on a wild goose chase whose first entrant was E. E. Evans-Pritchard.[...] We know that others will have to join us, others that the generous andrigorous analysis of the ways of African aesthetics also sometimes keep fromsleeping. (1995: 10–11)

The office of the Mission Sociologique du Haut-Oubangui, currentlyattached to the Laboratorie d’ethnologie et de sociologie comparativeat the University of Paris-X (Nanterre), is one of Dampierre’s mostimpressive legacies and is a monument to his fieldwork. He collectedpublished and unpublished works dealing with the general area of theUpper Ubangui as well as neighbouring peoples. He deposited there forthe use of interested researchers miles of recorded tapes (both musicand narrative, in Nzakara, Zande and, more rarely, Sango), hundreds ofphotographs (both historical and those he himself took in the field), thecomplete genealogical records of the various lineages of the Bandia clangoing back some fourteen generations, a map collection, a plantcollection, a French-Nzakara dictionary and dozens of indexed fieldnotebooks. Because he used historical sources so painstakingly yetsuccessfully, he was very much aware of the importance of leavingproper records of his own observations and experiences; his fieldnotebooks – all fifty or so of them – are numbered, paginated, indexedand cross-referenced, furnishing evidence and examples for hispublished works and unpublished papers.

Finally, Eric de Dampierre will be remembered for and by his studentsand colleagues at both the Department of Ethnology and the Laboratoired’ethnologie et de sociologie at the University of Paris X. As director ofthose institutions, and then as an ‘elder’, he instilled in all of us theimportance of thorough, careful fieldwork. Long, repeated periods in thefield were crucial, as was learning the language. He was sceptical ofethnographers who went only once to the field, did not stay long, anddid not learn the language well. His principles were passed on throughhis graduate seminars and supervision of dissertations. The emphasison fieldwork that he instilled in ethnology at Nanterre lives on.

As the Nzakara would insist, he was a man hors par – without equal.


Notes

1. Dampierre wrote a total of thirty-eight notes de recherche (‘research notes’) of varyinglengths and on a variety of subjects. Six were later revised and published. They arelocated in the office of the Mission sociologique du Haut Oubangui, at the Universityof Paris-X (Nanterre).

2. Dampierre continued working on UNESCO projects even after he began carrying outfieldwork in Africa. In 1959, he was named programme specialist at UNESCO andwas responsible for the section on human rights and the struggle against racialdiscrimination. In 1960, he travelled to Jerusalem and to the Negev for UNESCO toundertake sociological research on irrigation in arid zones.

3. Poetes nzakara (2 volumes, 1963, ms.; Poetes nzakara II is a finished manuscript butwas not published); also Satires de Lamadani (1987, text and cassette).

4. Dampierre had only compliments for Evans-Pritchard’s ethnography of ‘the Zande society,which has been so magnificently studied since 1927 by E.E. Evans-Pritchard, whom Icould not sufficiently praise for his scrupulous exactitude’ (1967: 247). He readeverything Evans-Pritchard had written about the Zande. He sent Evans-Pritchard a draftof at least some chapters of his dissertation, and he visited him on at least a few occasions.

References

Aron, R. 1951. Les guerres en chaîne, Paris: Gallimard.——— 1968. Main currents in sociological thought I, Garden City, NY: Doubleday and

Company. Translation of first half of Les étapes de la pensée sociologique, Paris:Gallimard, 1967.

Chirac, A. 1897. Sociométrie, Revue Socialiste, 6(34) (October).Dampierre, E. de. 1948. Sociométrie: note étymologique, Echanges sociologiques, 2: 63–66.——— 1951. Sur deux different types d’hérétiques, unpublished ms., 5 pp.——— 1956. Malvire-sur-Desle: une commune aux franges de al région parisienne,

L’Information géographique, 20: 68–73.——— 1957. Le sociologue et l’analyse des documents personnels, Annales, 12: 442–

54.——— 1960. Coton noir, café blanc, Cahiers d’études africaines, 2: 128–47.——— 1963. Poètes nzakara, vol. 1, Paris: Julliard.——— 1967. Un ancien royaume Bandia du Haut-Oubangui, Paris: Plon. ——— 1968. Présentation de theses soumises à la faculté des letters de l’Université de

Paris en vue du grade de docteur ès letters, unpublished manuscript.——— 1971. Elders and youngers in Nzakara kingdom, in Kinship and culture, F.L.K. Hsu

(ed.), Chicago: Aldine Publishing Company, 246–70.——— 1982. Sons aînés, sons cadets, Revue de musicologie, 68: 325–9.——— 1983. Des ennemis, des arabes, des histoires, Paris: Société d’Ethnographie.——— 1984. Penser au singulier, Paris: Société d’Ethnographie.——— 1987. Satires de Lamadani, Paris: Armand Colin (2 vols, with cassette).——— 1992. Harpes zandé, Paris: Klincksieck.——— 1994. Accord entre deux harpes, accord entre deux voix en Afrique équatoriale,

Note de recherche, no. 29, 1–4.——— 1995. Une esthétique perdue, Paris: Presses de l’ENS.——— 1996. Le reste épimore, Note de recherche no 33, 1–3.——— 1998. Les idées-forces de la politique des Bandia à travers les propos de leurs

souverains (1870–1917), Africa, 53(1): 1–16.


Evans-Pritchard, E.E. 1971. Azande, History and Political Institutions, Oxford: Clarendon Press.Ferry, Jules 1892. Rapport sur l’organisation et les attributions du gouverneur general de

l’Algérie, Paris: Imprimerie nationale.Moreno, Dr. 1943. Sociometry and the cultural order, Sociometric Monograph, 2: 318

(cited in Dampierre 1948).Rouget, G. 1982. Note sur l’accord des sanza d’Ebézagui, Revue de musicologie, 68: 330–44.Schweinfurth, G. 1875. Au coeur de l’Afrique, 1868–1871: voyages et découvertes dans les

régions inexplorées de l’Afrique centrale, vol. 2, Paris: Hachette.Vaux, P. de. 1967. CR [Compte rendu?], Academie inscr. et belles-lettres (cited in Dampierre

1968).Weber, M. 1958 [1922]. The three types of legitimate rule, Berkeley Publications in Society

and Institutions 4(1): 1–11.

Chapter 5

WHAT SORT OF ANTHROPOLOGISTWAS PAUL RIVET?

Laura Rival

Few anthropologists today know who Paul Rivet was. Even in France,where he played a central role in shaping the discipline during theinterwar years, the name of Paul Rivet evokes only vague memories:‘Rivet, the Director of the Museum of Mankind?’ ‘Rivet, theAmericanist?’ ‘Didn’t he write that controversial book on the origins ofAmerican Indians?’ The name is known, but no one seems to rememberRivet’s theoretical contribution or teaching.

Rivet was a medical doctor, military officer, field naturalist, collector,physical anthropologist, ethnologist, linguist, a builder of academicinstitutions and a politician – indeed, a success in all these professions.He became an anthropologist while working in the field in theEcuadorian Andes. His five years of fieldwork little resembled the classicethnographic fieldwork that Malinowski was to undertake in theTrobriands ten years later, but they nevertheless determined the rangeof issues, methods and theoretical questions he was to explorethroughout his long career. Even though he may have been practicallyerased from the discipline’s collective memory today, Rivet’s workshaped and influenced the development of post-Second World Warsocial anthropology in France, including Lévi-Strauss’s structuralism.Moreover, his holistic and humanist vision of anthropology as thescience of humankind, as well as his political commitment to educatingthe public about the value of cultures other than their own, aresurprisingly relevant today. Indeed, as I suggest in the conclusion, theyare perhaps even more relevant today than at any time since his deathin 1958.

Figure 5.1. Paul Rivet. From obituary in Journal de la Société des Americanistes, 1959.

Paul Rivet’s multi-stranded career

The second of six children, Paul Adolphe Rivet was born on 7 May1876 in Wasigny, a small village in Lorraine.1 He started school nearNancy, in the village of Blénod-les-Toul, where his father, a soldier whohad lost his right arm during the Franco-Prussian war, worked as a taxcollector. From him, Rivet learnt to be a fervent patriot deeplycommitted to equality of citizenship and human rights (Dussán deReichel 1984: 70). His primary education completed, Rivet enrolled inthe grammar school at Nancy, where he excelled in literature and poetryand passed the baccalaureate in philosophy with distinction(d’Harcourt 1958, Pineda Camacho 1985, Zerilli 1991–93). Despite

126 Laura Rival

his obvious talent for literature and philosophy, this son of a modestprovincial family decided against preparing the entry exam for the EcoleNormale Supérieure (Laurière 1999: 109). Although he was notparticularly attracted to medicine and had no ambition to serve in theFrench army as his grandfather, uncle and father had done before him,he chose, mainly for financial reasons, to become a military doctor(Chevasse 1958: 106). Rivet’s unusual twin training in Frenchhumanism and medical science gave him the encyclopaedic breadth hewas to deploy so fruitfully throughout his long academic career. Onecan also assume that his employment as medical warrant officer in theParis cavalry from 1898 to 1901 helped him develop the discipline andorganisational skills for which he later became so famous, not least asDirector of the Musée de l’Homme.

As is often the case in anthropological careers, serendipity played alarge part in Rivet’s choice of Ecuador as the country in which he wouldconduct field research. By pure chance he was given an opportunity toaccompany the second French geodesic mission, charged with re-measuring the equatorial meridian. As the mission involvedcollaboration with the Académie des Sciences de Paris, Rivet was ableto receive some scientific training at the Muséum National d’HistoireNaturelle (MNHN) before departing, especially in anthropometricmethods. Far from being unusual, Rivet’s expertise in comparativeanatomy and his passion for natural history were shared by othermembers of the geodesic mission, and, indeed, by many early twentieth-century anthropologists (Stocking 1992b: 17–32). León (1958: 307)nicknames them ‘science missionaries’.

In Ecuador, where he was to spend the next five years of his life, Rivettravelled indefatigably, attending patients and collecting numerous anddiverse materials. Free medical services were obviously very popularand were consciously used by the geodesic mission as an efficient meansof enlisting the locals’ goodwill. A zealous field naturalist, Rivetcollected plant and animal specimens, archaeological finds (bones,pottery and so on), indigenous artefacts, oral traditions andvocabularies of many of the languages spoken in Ecuador at the time,and took anthropomorphic measurements.

In the first two years of his stay, Rivet mainly collected species forthe MNHN, contributing equally to entomology and botany. Thespecimens were sent for identification to leading scientists based inFrance, elsewhere in Europe or the United States. The breadth of hiscollections is reflected in the fact that more than thirty animal specieswere named after him.2 His botanical collections also contained manynew species and varieties. Very soon, the anthropological collections hesent back to Paris roused the interest of scholars at MNHN. However,

What sort of anthropologist was Paul Rivet? 127

Rivet had to fight hard not to have them treat him as a simple fieldsurveyor and collector. He had personally funded the expeditions, andthe collections were the fruit of his personal efforts. He aspired to thestatus of full researcher, and was determined to be the one who wouldclassify, study and analyse the objects now stored at the MNHN, which,as far as he was concerned, belonged to him (Laurière 2006: 236–39).

In July 1906, having won the battle, he was officially detached fromthe Army’s geographical services to work at the MNHN as anindependent scholar under the supervision of René Verneau andErnest-Théodore Hamy. He soon became a member of the Société desAméricanistes de Paris (in 1907), even serving as its general secretaryfor several months the following year, a post to which he was to beformally elected in 1922 and which he was to occupy for more thanthirty years, until his death in 1958. He received much praise andscientific recognition in 1908 for his collections, exhibited in thezoology galleries of the MNHN. In 1909, when Verneau took up theChair of Anthropology at the MNHN in succession to Hamy, Rivetbecame his research assistant.

Soon after, Rivet published one of his most important scientificpapers,3 which offers a systematic refutation of the theories andmethodologies that defined physical anthropology at the time. While inthe field, he had collected complete anthropometric measurements for300 Indians (mainly adult men) and had also measured at least 60skeletons in cemeteries. His ancient bone collection comprised 350skulls and about 500 bones, including 400 long ones. Back in Paris, hisintention had been to write a substantial anthropological study of theAmerican Indian race with his MNHN colleague, Raul Anthony. Thepaper was to explain the scope, origin and history of the internaldiversity of the race. However, the systematic study of the empiricaldata he had gathered led him instead to reject the premises ofanthropometry. Skull measurements, he concluded, lead to thearbitrary classification of humans in entirely fictitious groupings. Thehuman types so determined have nothing real in common, only thearbitrary trait used as a basis for their classification.

In 1910, ideological disagreements between the MNHN and theSociété d’Anthropologie de Paris over anthropometric measurementsand the scope of anthropological research reached crisis point, whichprompted Rivet and Verneau to resign. A year later, Rivet created a newlearned society based at the MNHN, the Institut Françaisd’Anthropologie, of which he became the archivist-librarian. Durkheim,Hertz, Mauss, Lévy-Bruhl and many others immediately joined the newInstitut, which sought to redefine anthropology ‘in its widest sense’. Nolonger a mere synonym for physical anthropology, anthropology was to

128 Laura Rival

be based on a federative integration of all the disciplines involved inexplaining ‘the human phenomenon’, chiefly linguistics andethnography (Dias 1991: 246–50). The Institut’s main function in theeyes of its members was to fight against sterile specialism and to facilitatethe circulation of new ideas and knowledge in all the domains thatinterested broad-minded anthropologists (Laurière 1999: 111).

Published in 1912, L’Ethnographie ancienne de l’Équateur,4 ostensiblyco-authored by Rivet and Verneau, but in reality very much Rivet’swork, was awarded several prizes. Based on an abundance of sources –which are not simply listed but also commented on and integrated fullyinto the text – the book provides a comparative analysis of the largearchaeological collections that Rivet brought back from Ecuador. It isstill widely regarded as a masterpiece of rigour in the methodicaldescription of pre-Colombian material culture.

Rivet’s subsequent publication plans were curtailed by the outbreakof the First World War.5 Throughout the war, he worked with the sameenergy, method, discipline and sense of organisation, this time to reformthe medical care of wounded soldiers and, later on, the running of amilitary hospital in Salonika, where he also managed to find the time toorganise scientific expeditions and prepare naturalist collections for theMNHN. After the war,6 Rivet almost resigned from the Société desAméricanistes de Paris over political issues. A number of its Frenchmembers wanted to see German-speaking scholars banished from theSociété’s membership list. Rivet, who strongly believed in internationalscientific cooperation and knew how much America owed to Germanand Austrian scientists, fought hard against the motion and won.7

In the 1920s, Rivet increasingly turned his attention to linguistics.He had already authored or co-authored 22 communications on SouthAmerican indigenous languages between 1907 and 1919, but the 21works published between 1920 and 1925 are more comparative andmore ambitious in scope, especially the chapter on American languageshe wrote for Meillet and Cohen’s Langues du monde (Rivet 1924).

On 6 August 1925, after months of preparatory work and lobbyingby Lévy-Bruhl, the French ministry for colonial affairs agreed to fundthe creation of an Institut d’Ethnologie within the University of Paris.Rather like the British government, the French governmentinstitutionalised and funded modern anthropology when it becameconvinced that anthropology could play a positive role in theadministration of the colonies. Lévy-Bruhl invited Marcel Mauss andRivet to become the institute’s general secretaries. The Institut’s rolewas to train ‘professional ethnologists’,8 introduce anthropologicalfindings to civil servants assigned to the colonies and educate the publicabout the invaluable contribution to civilisation made by the


130 Laura Rival

populations, societies and cultures found in the French colonies andbeyond. According to Laurière (1999: 114), Lévy-Bruhl and Maussdelegated to Rivet the task of organising and administering the Institut.The close collaboration between the three men, who shared the samesocialist political ideals and humanist values (Pineda Camacho 1985:8, Jamin 1989: 282), led to the institutional association of threeParisian research and teaching centres: the Sorbonne (under Lévy-Bruhl), the Ecole Pratique des Hautes Etudes (under Mauss), and theMNHN (under Rivet). Through their association, the three scholarshoped to facilitate the federation of three complementary theoreticalapproaches to the study of mankind, or sciences de l’homme as they arestill referred to in France: philosophy (under Lévy-Bruhl), sociology(under Mauss) and anthropology (under Rivet). This they saw as anecessary condition for strengthening a modern Frenchanthropological project firmly anchored in a common methodology,namely ethnographic fieldwork (Allen 2000).9

In 1928, Rivet was elected to the MNHN Chair in Anthropology, thetitle of which he immediately attempted to change to that of the Chairof the Ethnology of Modern and Fossil Men. Rivet’s appointmentrevived the muted, yet persistent intellectual dissensions between thosewho favoured a broad, holistic anthropology programme and those whowere eager to maintain physical anthropology as a separate, if no longerhegemonic discipline. The following year Rivet, who had obtainedagreement that the Trocadéro be attached to his chair, became thedirector of the museum, where he created a library of ethnology. Facedwith the challenge of holding so many different officessimultaneously,10 Rivet chose to concentrate on the MNHN and theTrocadéro, leaving Mauss and Lévy-Bruhl fully in charge of the Institutd’Ethnologie. By then, and perhaps as a result of this unintendeddivision, it was decided that the University of Paris would offer twodifferent qualifications in anthropology, a certificate in ethnologyawarded by the Science Faculty (Faculté de Sciences), and anotherawarded by the Faculty of Arts and Humanities (Faculté de Lettres).11

Rivet, who continued to advocate ethnology as a total science givingequal importance to physical anthropology, linguistics andethnography (defined as the study of past and present material culture),became increasingly convinced that such a project required theupgrading of the old Trocadéro Museum, which he finally achieved in1937 with the creation of the Musée de l’Homme.

As Jamin (1989) and Laurière (2006: 518), who refer to Rivet as a‘scholar and a politician’, both note, the various aspects of Rivet’s multi-faceted career and his different projects – to develop Americanist studieson an international scale, participate actively in progressive politics and


popularise anthropology as a science in the service of universalcivilisation, that is of a true humanism – became increasingly merged fromthe mid-1930s onwards. Having responded favourably to Lévy-Bruhl’soffer to join the Socialist Party (SFIO), Rivet soon involved himself informal politics. In 1934, he founded the Comité de Vigilance desIntellectuels Antifascistes with the physicist Langevin and the philosopherAlain, and acted as its first president. A year later, in 1935, Rivet becamethe first politician to be elected on a Popular Front list as councillor for theCity of Paris and the Regional Council of the Seine (Lottman 1982: 78),inaugurating a long and distinguished political career.

Rivet also continued to contribute to the expansion of the Sociétédes Américanistes. Under his presidency, various collaborative projectswith European and North American Americanists were planned orundertaken, and links with researchers, informants andcorrespondents multiplied throughout South America, where hetravelled and taught with increasing regularity. In her biography ofJulian Steward, Kerns (2003: 224) mentions that Rivet was involvedwith Lowie as early as the late 1920s in an encyclopaedia project onAmerica’s cultures and societies.12 By the time US funding becameavailable in the early 1940s, France could no longer, for obviousreasons, participate in the project, which became exclusively associatedwith its sponsor, the Smithsonian Institution. If Julian Steward becamethe encyclopaedia’s general editor, it was a former student of Rivet,Alfred Métraux, who wrote many of the ethnological entries.

When the Vichy regime fired him from the directorship of the Muséede l’Homme in 1941, Rivet accepted an invitation from the Colombianpresident, allowing him to escape from the Gestapo just in time.Between 1941 and 1943, he helped develop the National Institute ofEthnology (Instituto Etnológico Nacional),13 the Ethnological Institutesof the Universities of Cauca, Antioquia and the Atlantico, and theColombian Society of Ethnology (Sociedad Colombiana de Etnología).He also trained the first Colombian social anthropologists (includingGerardo Reichel-Dolmatoff), participated in the foundation of Bogotá’sethnographic museum, mounted several fieldwork expeditions (deFriedemann 1984) and worked closely with Paulo Duarte, the leadingBrazilian archaeologist, who had been his student.

Rivet’s best-known book, Les origines de l’homme américain, waspublished in 1943, and his major Bibliographie des langues aymará etkicua completed in 1954, four years before his death. The Musée del’Homme, which he considered to be his most important scientificachievement, survived him for nearly five decades, until the creation ofthe Musée du Quai Branly in 2006. Rivet was in many ways stimulatedby the Victorian ambition of using anthropological knowledge to

132 Laura Rival

reform society (Stocking 1987: 324–29), and the museum he directedconstituted, in his eyes, a perfect synthesis between research, educationand politics. Aimed at popularising ethnological science in a way thatwould valorise the material cultures of so-called primitive societies andelevate them to the rank of civilisations, the collections of the Muséede l’Homme were designed on the one hand to educate the Frenchpublic against racist ideologies and prejudice, and on the other hand tosignal to officials and elites that the duties and responsibilities of goodgovernment equally apply to all citizens.

Paul Rivet’s fieldwork

In line with much of the anthropological writing of his time, Rivet’searly work deals essentially with the questions of origins and historicalchange. The almost complete absence of concern for issues of structure,function or meaning is striking. Yet even though he uses hisobservations of contemporary indigenous cultures more as a source offacts to extrapolate about the ‘aboriginal’ past than as a basis fromwhich to examine particular forms of social organisation, theethnographic interest in the diversity of human ways of life is neverentirely absent from his writings. An examination of his first field studyof indigenous people (Rivet 1903), his essay on the Jibaros (Rivet 1907–8), and his 1906 programmatic report will illustrate both hisanthropological interests and his methodological approach.

The Quechua Indians of Riobamba (1903)

A medical doctor14 doubling as a field naturalist, Rivet came intocontact with many patients during his two first years in Ecuador. As amatter of policy, he treated not only the geodesic mission staff but alsomany influential Ecuadorians, local scholars and, on occasion,indigenous people and mestizos from the rural communities he visitedduring his collecting trips (Dussán de Reichel 1984: 70–71). He talkedto many people, but never directly to the Indians themselves. Althoughhe did interact with indigenous contracted labourers and patients, mostof his ethnography comes from observing the Indians at a distancethrough layers of stereotypes. What kind of ethnographic observationsdid he make, then, in the highlands of Ecuador? An examination of hisfirst field study with indigenous people, the Quechua Indians ofRiobamba (Rivet 1903), gives us some clues.

The article starts with a detailed geographical description of theRiobamba region, followed by a discussion of whether ‘the Indians ofthe Riobamaba region’ actually are an adequate unit for anthropological


analysis. What differentiates these Indians from their neighbours? Rivetargues that, although a full study of their customs and physical make-up (through anthropometric measurements) would be needed toconclude with certainty that they indeed form a distinguishablebiocultural unit, they seem to possess enough traits in common to betreated as a separate group. ‘Group’ here is not used in the social orpolitical acceptation of the term, but rather to refer to a population madeup of individuals sharing a common physical appearance, due partly totheir shared biology and partly to their shared material culture.

Rivet proceeds to describe systematically and with clinical accuracythe physical appearance of men and women (clothing, adornment,hairstyle), dwellings (including furniture, tools and sleepingarrangements) and the yards around them, where children play anddomesticated animals live. He observes that compounds are fenced andgives the scientific names for all the plant species making up the hedges.The different species of animals found are also listed. He then goes onto describe the people’s food preferences, the ways in which food isprepared and consumed, and women’s work, in particular thepreparation of maize beer. The detailed description is followed by adiagnosis: these Indians live an extremely impoverished way of life,which has been forced upon them by history. Their amenities are basicand rustic, the pervasive filth and lack of hygiene a constant source ofillness, but their diet is frugal and healthy. In this early ethnographicexperience, Rivet’s training as a medical doctor clearly shapes andchannels his field observations, guiding his assessments of local healthconditions, remedies and attitudes to western medicine. He shows agreat interest in and appreciation of indigenous cultural achievements:their physical endurance, their capacity for hard work, their localknowledge of plant remedies. If the Indians of Riobamba are notresistant to malaria or smallpox and die at a young age, it is becausethey refuse to be vaccinated, live far from medical centres and do notlook after themselves properly. They use harmful treatments such asthrowing themselves into icy cold rivers when they have a high fever, orrubbing guinea pigs onto their bodies for purposes of ritual cleansing.Their witch doctors are charlatans who exploit their ignorance and partthem from the little cash they are able to earn. Here ends the ‘individualethnography’, as he calls it (Rivet 1903: 64), of the Indian ofRiobamba, who, despite the abject poverty in which he is forced to live,manages – just about – to retain some human dignity.

The second section of the article deals with the regionalconfiguration created by the two co-existing, but not mixing,biocultural populations found in the Riobamba valley. Before examiningthe social networks that unite the Indians, Rivet examines the reasons

134 Laura Rival

for the profound chasm that exists between the Indians and the whites(1903: 64–65). What amazes him most is that, while the Incassucceeded in imposing their language and customs on the localindigenous population in less than fifty years, Spanish culture, evenafter three centuries or more, has not fully penetrated, much lessreplaced, the Indian cultural heritage. The two ‘races’ live side by side,but without mixing and with minimal communication, the whitesconcentrated in the urban centres, the Indians dispersed throughoutthe countryside. The Indians may have adopted Catholicism, but theiranimistic way of practising the Old World religion has very little to dowith what the priests and missionaries intended to teach them.

He then outlines the different types of political organisation found inthe valley, compares hacienda-bonded and free communities, anddiscusses the power and authority of self-appointed chiefs (he names afew). Here, the reader is left to wonder how much his own childhood,spent on the historically disputed border between France and Germany,which is rife with cultural and political divisions, and his militarytraining have influenced his ethnographic understanding. Theunfamiliar landscape and the human settlements are looked atstrategically, and political alliances and divisions mapped out. At severalpoints in the text, Riobamba indigenous customs and institutions arecompared with Arab practices.

Rivet goes on to examine the patterns of authority within the familybefore describing family life (1903: 65–67), for instance the affectionateties between husband and wife. Pre-marital sexual life, marriagenegotiations, wedding rites and various family relations are thendescribed succinctly. Stress is put on the common human desire tofound a family, and cultural differences are played down. This passagedemonstrates again Rivet’s sharp observational skills (for instance, hispenetrating description of how women care for their drunkenhusbands), but also his almost total lack of interest in any ‘why’question. Although exotic practices are accurately noted and exposedwithout a trace of moral judgement, they never become a source ofwonder. It is the generically human experience that retains Rivet’sattention throughout.

The style in which the third section is written is entirely different. Itis as a socialist that Rivet analyses the category ‘Indian’ as a politicalconstruction inherited from the colonial order. The analysis is aimed atuncovering the particular political and economic relationships, notbetween humans pertaining to different ‘races’ but rather betweenlandowners and labourers. Three categories of Indians are differentiatedand analysed, and the Spanish terms (concierto, apegado and suelto orlibre) defined. Concierto Indians, the most exploited (Rivet actually says


‘unfortunate’), are debt-bonded to large hacienda landowners. The debtcontract is described, as are the working conditions. Although legallyfreer than concierto Indians, apegado Indians in fact live almost asprecariously and inhumanly as their debt-bonded brothers, for they haveno way of themselves producing or selling the products of their labouron the market. Suelto Indians sell not only their labour force but alsofarm products (e.g. crops, eggs, cheese), products they collect (e.g.firewood or grasses, whose scientific names are given) and the artefactsthey make (e.g. ropes made with the fibre of certain plants, whosescientific names are also listed). If they are not much better offeconomically than the Indians in the two other categories, it is becausesuelto men end up wasting all the cash they earn.

The last section is dedicated to indigenous religiosity. Rivet starts withthe assertion that, in their religious practices as in their linguisticbehaviour, the Indians, although a vanquished race, have successfullyresisted the victors (1903: 74). The sketchy mention of religious beliefs,superstitions, funerary rituals, the cult of the dead and life-cycle ritualsthat follows is aimed at supporting the general argument that, despitemissionisation and colonisation, the Indians are essentially pagans whostill believe in mountain spirits, assimilate the figure of Christ to thehacienda landlord and let the priests exploit and oppress them shamelessly(ibid.: 74–75). Unsurprisingly, it is in the treatment of religious beliefs,where the native point of view and the indigenous symbolic meaningsmust form the central focus of the inquiry, that Rivet’s lack ofethnographic empathy is most disturbingly evident. The problem is not somuch that Rivet did not talk directly with Indian informants about theirreligious ideas and practices (he had no means of establishing thenecessary relations of trust without total immersion) but rather that heis incapable of imagining that the Indians of the Riobamba valley mayhave their own reasons to believe what they believe in, or to act the waythey act. To him, they are simply superstitious, ignorant people living ina backward and profoundly racist country. With his naïve and simplisticconclusion that religion is a powerful ideology used to keep the Indianmasses in servility, Rivet fully reveals himself as the Third Republicthinker he is, totally imbued with the superiority of his scientificaspirations to rationality, progress and enlightenment. Scientificknowledge is a weapon against not only religious mystification andignorance but also poverty and inequality. The destitution and socialimmobility in which the Indians live, he concludes, should in no way beattributed to racial inferiority, for they result from three centuries of harshtreatment and enslavement (ibid.: 78).

Rivet’s primary interest in the human condition, wherever hehappened to find it, comes across vividly in this first study, perhaps even

136 Laura Rival

more so in the absence of an interest in cultural difference. Rivet mayhave lacked cultural empathy, but certainly not social empathy. He didnot choose to research a complex and depressing situation, but simplytried to analyse it as a witness, without exoticising the very poor Indianshe knew.15 This first study also shows Rivet’s unusual interest in thecomplex history of mixed-race and acculturated populations, or whattoday would be called in France les métissages culturels, or the ‘mestizomind’ (Gruzinski 2002, Castelain et al. 2006).

The Jibaro Indians (1907–8)

Rivet organised expeditions apart from those of the geodesic missionand travelled to the eastern and western slopes of the Andes, where hemet and studied independent and isolated Indian populations such asthe Tsachilas (Rivet 1905). Like any other traveller, he becamefascinated with the famous ‘Jibaros’ and planned to spend at least ninemonths in their territory, a project he ultimately could not carry out(Laurière 2006: 192). He nevertheless wrote a substantial essay onJibaro culture, mainly based on secondary sources, and on the fewdirect observations he had made himself while measuring andinterviewing travelling Jibaros.

The study, which compiles approximately thirty references in variouslanguages and includes a number of personal observations, as well asanswers to questionnaires Rivet sent to missionaries and traders, isjustified on the ground that, even though the Jibaros are universallyknown among travellers as a ‘fiercely independent, pure and originalIndian race’, no ethnographic synthesis of their culture exists as yet(Rivet 1907: 338, 349). After a brief exposition of the facts justifyingthe treatment of Jibaro culture as a separate, distinct and homogeneousbiocultural unit, Rivet presents the highly detailed data he compiled innine different sections: historical background, geographicaldistribution, census data, physical anthropology, material culture,family life, social life, religious life and psychic life. Material life issubdivided into the house, agriculture and husbandry, food, weapons,musical instruments, daily life, and hunting and fishing; family life intogender relations, children’s education, marriage, birth and death; sociallife into social organisation, commerce and warfare; religious life intotraditions, divinities, witchcraft, afterlife, totemic beliefs, celebrations,superstitions and medicinal recipes; and finally, psychic life into generalknowledge, counting, arts, moral life, cultural personality andreflections on the race’s future.

The historical section is based almost entirely on Federico GónzalezSuárez’s Historia General de la República del Ecuador.16 In Rivet’scharacteristic ethnographic style, direct observations are complemented


with descriptions of collected materials and annotated summaries ofsecondary sources. Archaeological data are used to complement orillustrate historical works, and direct field observations juxtaposed withresponses to questionnaires sent to ‘intermediary informants’. Often,questionnaires are used to expand the ethnographic inventory bycovering facts on which no direct observation was possible, instead ofbeing used to offer a complementary perspective on the same object ofanalysis. This leads to a problematic division of labour between directand mediated observation: whereas the ethnographer (Rivet) specialisesin the direct observation and collection of concrete material items, theintermediary informant, being deemed more intimately familiar withthe Indian, is asked to contribute the sociological, intellectual, moraland symbolic data.

The stereotypical generalisations offered in the sections on thefamily, social institutions, religious beliefs and Jibaro personality aredrawn from what people who ‘know’ the Jibaros told Rivet. Except for afew corrective comments, where Rivet uses his direct knowledge tocorrect stereotypes that he finds erroneous or exaggerated (1907: 608–9), there is no questioning of the sources. Moreover, and perhaps moredisturbingly, wonderful ethnographic facts are noted in the course ofdescriptions of material culture or daily life – and just left at that. Thesefacts, so fascinating for the modern ethnographer, are nevercommented upon or analysed. There is absolutely no attempt on Rivet’spart to raise a question or to call upon the native point of view;ethnographic facts do not awaken his imagination. For instance, hestates in passing (ibid.: 601) that Jibaro men put their left fist in theirmouths each time they tell myths or tribal war stories. A beautiful, verygraphic drawing by one of his colleagues from the geodesic missioneven illustrates the scene (ibid.: 600). The same indifference greetsother facts, such as the magical practices surrounding the sale of a gunto a white trader (ibid.: 602) or the high rate of female suicide amongthe Aguaruna (1908: 239).

In contrast with his manifest disinterest in cultural aspects thatcannot be catalogued, Rivet is unstoppable when it comes to giving thescientific names of the plants used, describing artefacts or praisingtechniques, technologies, bodies of practical knowledge and other typesof indigenous material achievement.17 The same goes for his expertappreciation of the Jibaro’s diet, physical beauty, strong constitution,economic self-sufficiency and preventive measures against smallpoxepidemics. He has, of course, no time or sympathy for remediesgrounded in superstition or beliefs in witchcraft.

Zerilli (1991–93: 390) sees in this essay the richest and mostethnographic of all Rivet’s works. Laurière does not share this point of

138 Laura Rival

view (2006: 193), nor do I. What this essay illustrates so well, however, isthe ways in which the ‘natural scientist-cum-anthropologist’ (Stocking1992b: 20) used ‘indirect informants’ in the collection of ethnographicdata (Rochereau 1958, Pineda Camacho 1996). While still employed bythe geodesic mission, Rivet had no choice but to rely on the information hecould obtain from priests, military officers, local scholars or traders. Whatstarted as a practical way of gathering data on places and people he wascurious about but could not visit himself became a way of working, evena methodology. Like his predecessors (cf. Dias 1991: 82–83), Rivet thoughtthat good ethnography did not depend on field professionals, and that goodquestionnaires filled in by knowledgeable correspondents were sufficient.Observation and classification could remain two separate activities as longas the anthropologist had a vast network of secondary informants withwhom he could correspond regularly. Rivet’s pragmatic methodologicalapproach was particularly successful for researching Amerindianlanguages. He not only amassed great quantities of linguistic data (mainlyvocabulary lists) but also co-authored scores of publications with indirectinformants and co-researchers formally trained in linguistics.

As we know today, what works for the collection of material itemsand factual information may be totally inappropriate for both ‘thesociological study of systems of action’ (Leach 1957: 119) and thereconstruction of psychic life ‘fixed in language, art, myth and religion’(Stocking 1992b: 37). However, Rivet was not studying indigenoussocial classifications: his task, as he saw it, was to survey the field bygathering basic ethnographic descriptions that could be mapped ontothe South American continent and methodically classified. That Rivetremained uninterested in the general structure of society and the nativepoint of view until the end of his life explains why his 1941–1942 fieldresearch in Colombia was similar in almost every respect to thefieldwork he had carried in Ecuador at the beginning of hisanthropological career. Accompanied by his Colombian students, hesailed through various remote regions to collect anthropometricmeasurements, blood samples, archaeological artefacts, numerousitems of material culture and vocabulary lists (Pineda Camacho 1996,Laurière 2006: 817–18). Here too, fieldwork was aimed at constitutingthe material archives of disappearing cultures.

Origins and migrations of the ‘American Man’

Rivet’s early publications, authored in his capacity as a ‘medical doctorattached to the French geodesic mission’, well illustrate both hisanthropological interests and his methodological approach. They touchon many burgeoning domains (from studies of prehistoric skeletalremains, to studies of diseases, indigenous languages, religious beliefs,


artefact use and economic activities, and physical features orbiomedical conditions), and combine field data reports or syntheses ofpreviously published work (often in Spanish) with comparativeanalyses. Of the nineteen articles written between 1901 and 1908 (aslisted by León 1958), eleven deal with contemporary groups andcustoms, three concern pre-Columbian cultures and five involve generalcomparative discussions. These early publications give a good indicationof how Rivet decided, under Mons. Gónzalez Suárez’s influence, to stopsurveying the natural environment and start studying the origins ofSouth America’s aboriginal populations, as well as the trajectory of pre-Columbian civilisations, which involved researching the continent’sarchaeology, physical anthropology and folklore. As Uribe (1996: 52)remarks, Rivet thus shifted his interests from natural history to humanculture, but without changing his basic methodology: collecting,classifying and comparing.

His foremost interest in origins and migrations is clearly stated inthe 1906 programmatic report (Rivet 1906: 236), which starts withseveral pages dedicated to the geography and history of the Andeanregion, where the geodesic mission had taken the majority of itsmeasurements. In fact, twice as much space is dedicated to history as togeography, while just a few pages towards the end provide someethnographic information. Why so much importance given to history?Because, replies Rivet (ibid.: 232), the traveller can easily reconstruct,beneath the cultural manifestations of Inca civilisation, the traces ofanterior and original local civilisations, as the bewildering diversity ofancient burial arrangements existing in the Ecuadorian Andes testifies.Spanish empire-building, exactly like Inca empire-building, took placein the inter-Andean valley, where, as a result, the population is raciallymixed, in contrast with the lowlands, where ‘racial purity is almostabsolute’ (ibid.: 233, my translation). The ethnological problem,concludes Rivet, is therefore far more complex (and more interesting)on the high plateaus.

The ethnological research summarised in the rest of the articleconsists in numerous archaeological excavations and anthropometricmeasurements – more than two hundred subjects of both sexes and allages. These two different modes of direct empirical data collection – oneaimed at reconstructing the past of human and cultural diversity, theother at understanding the nature of contemporary diversity – arecomplemented by two types of secondary sources: the published workof historians, and interviews with outsiders in daily contact with theIndians, essentially priests, military officers and traders. The ‘AmericanMan’ was, from the beginning, Rivet’s most systematic ethnologicalconcern. He asks in the 1906 report the questions he will answer in Les

140 Laura Rival

origines de l’homme américain: What can today’s diversity of physicalappearance, language and material culture tell us about the origins ofindigenous cultures? Where did the American Indians come fromoriginally? Although he cannot address these questions fully as yet, hetries to account for what he is already calling métissage. For him,hostility between the two races (conquerors and vanquished) explainswhy Andean Indians have not lost their identity, despite centuries ofHispanic influence. The ‘races’ have mixed biologically, but notculturally. Civilised by the Incas and still speaking their Quechualanguage, the mixed-blood Indians continue to resist hispanisationstubbornly (1906: 233).

The continuities in Rivet’s intellectual development are noteworthy.Even earlier, in the journal he kept during the transatlantic cruise fromFrance to Ecuador (Zerilli 1991–93), Rivet revealed his curiosity aboutthe mechanics of intercultural communication. The facts he observed,described and meticulously recorded during stopovers in the Caribbeanillustrate his profound interest in biological and cultural hybridity, as wellas their social and political consequences. In Martinique, for example, hereflected on the condition of Christianised blacks, who have remainedfetishist at heart (‘les nègres catholiques restés fétichistes’). And until theend of his life, for instance in the teaching he delivered in Colombia (acountry even more racist and conservative than Ecuador) on theremarkable civilisational achievements of indigenous cultures, hecontinued to develop with passion and eloquence the themes of universalhumanism and racial equality (Dussán de Reichel 1984), two ideals hesaw as inseparable from an anthropological reflection on métissage.

That Rivet was a diffusionist is clearly revealed by his fascination fororigins, his interest in historical migrations, miscegenation and linguisticdiversity, and his (over-) use of historico-comparative methodology.Whereas most diffusionists, especially those associated with the ViennaSchool of ethnology (Haekel 1956), emphasised a people’s historicitywith reference to their spatial distribution and the spatial and temporaldiffusion of their material culture, Rivet chose to classify people in termsof linguistic distribution, which he saw as scientifically more accurateand more rigorous. His analysis of South American languages wasmodelled on the philological work produced by the linguists who hadreconstructed the Indo-European family. Boas, who had adopted thesame model at the start of his research on North American languages,soon departed from it, as Stocking explains: ‘By 1920 [Boas’] position ...had changed drastically, and he was inclined to believe that diffusion ofmorphological traits could modify the fundamental structuralcharacteristics of a language’ (Stocking 1992a: 86).


The legacies of Paul Rivet’s vision for anthropology

As I have tried to show in the previous section, Rivet was not simply a‘frenetic empiricist’, as Claude Lévi-Strauss called ethnologists fromprevious generations to highlight the unique nature of his ownconception of fieldwork as a revelatory experience of ‘radicaldisplacement’ (Johnson 2003: 9, 169). Nor was his way of doingfieldwork simply the result of circumstantial constraints or lack ofmaturity, as his last spell of fieldwork in Colombia amply demonstrates.Rivet was bent on collecting a certain kind of empirical data in the field,as a result of both his training and his own understanding ofanthropology. A careful reading of Dias (1991) amply supports a viewof Rivet’s work as being entirely in line with previous attempts inFrance: firstly, to define anthropology as broadly as possible; secondly,to oppose racist rankings of human cultures by showing that languagesare better guides to the study of lasting cultural differences thanphysical traits are; thirdly, to have museum collections accepted asmajor research tools and to treat material culture as the embodiment ofa people’s cultural creativity and technological achievements; andfourthly, to demonstrate through a range of scholarly studies that greatcivilisations had developed in the New World, a continent of which solittle was known.

Like Boas, Rivet fully embraced physical anthropology and masteredanthropometric measurement techniques, only to use them against theanalyses and conclusions reached by phrenologists bent on proving thegenetic existence of separate human races and their hierarchical ordering(cf. Pineda Camacho 1996: 65). The relationship between Rivet’sEcuadorian field experience and his thesis that persisting differencesbetween human groups are cultural and linguistic rather than biological,and therefore that ‘race’ is a vacuous concept, has been examined by bothZerilli (1991–93, 1998) and Laurière (2006). The two authors may givetoo much importance to the construction of scientific discourse and notenough to Rivet’s own life experience. A medical doctor with a passionfor bettering human health through a greater knowledge of anatomy,biology and epidemiology, a keen observer of all things natural andhuman, and a firm believer in a universal, enlightened civilisation, Rivetsoon connected facts collected in the field or learnt in libraries in a novelway, which led him to oppose firmly the notion of racial inferiority andpropose instead the theory of métissage.

What struck Rivet upon arriving in Ecuador was the bewilderingphenotypic diversity found in the country, both within and betweenethnic groups. The co-occurrence of intra-ethnic phenotypic diversityand inter-ethnic linguistic diversity fascinated him. After four centuries

142 Laura Rival

of Quechua and Spanish imposition, and despite the intermingling ofraces, Ecuador had remained an ethnic and linguistic mosaic. Humanlanguages and human biology, he concluded, do not change at the samerate, nor in the same direction. Whereas human bodies are prone tomixing, linguistic differences persist over time. Moreover, theintermingling of races, far from being a cause of generation, as so manyof his contemporaries believed, was a source of biological vitality andcultural progress. European and American societies were both raciallymixed. The purpose of physical and biological anthropology, includinganthropometric measurements and the study of blood groups, wastherefore to measure the historical process of métissage (Pineda Camacho1996: 65). Finally, it is by sharing the lives of Ecuadorian indigenouspeoples that Rivet could fully measure the gap existing between theirintelligence and cultural creativity, and the racist stereotypes held byLatin American elites and authorities of the ‘savages’. The fact that allhuman societies contributed equally to the general development ofhumankind could be demonstrated through the study of indigenousmaterial culture and technology, for instance pre-Columbian metallurgyand gold smelting. It is in Ecuador, a country of entrenched paternalismand deep racial prejudice, that Rivet learnt to feel the human dimensionof his indigenous informants, to lift the barrier between ‘us’ and ‘them’once and for all, and to reach a profoundly anti-racist and anti-ethnocentric vision of humanity (Dussán de Reichel 1984: 71).

Although Rivet’s linguistic studies were made to serve his diffusionistthesis, they nevertheless contributed to a better knowledge of SouthAmerican languages. Not only did he pioneer a vast new field of research,he also helped improve the classification of the numerous languages inthis region (Pineda Camacho 1996: 59), even if, in the urgency of datacollection and comparison, his attempt to reduce the number of isolatesled to incorrect affiliations (Campbell 1997: 80–81). Sometimes herushed to conclusions too quickly and his methodology was notsufficiently rigorous, but many of his bold and brilliant intuitions wereconfirmed by later research, and some of his hypotheses are still guidingcurrent research (Landaburu 1996). A number of Andean specialistsstill consider his Bibliographie des langues aymará et kicua (1951–54) animportant work of reference.18 More controversial was his application toSouth American languages of genetic approaches specifically developedby philologists to reconstruct Indo-European languages. Boas, whotended to explain language similarities in diffusional rather than geneticterms (Stocking 1992a: 74, 86), was critical of Rivet’s linguistic studies,particularly his 1924 Langues américaines.19

Rivet’s diffusionist search for correlations between the movements ofmaterial objects and languages led him to put forward a range of


hypotheses, some more insightful or audacious than others, but allhotly debated among Americanists. For example, his proposition thatthe cultural influence of Amazonian civilisations on Andean societiescould be demonstrated on both archaeological and linguistic groundsliterally enthused Nordenskiöld in 1913. Rivet’s main thesis, firstformulated in 1924 and perfected in 1943, that the Americancontinent had received not only Asian migrants entering through theBering Straits but also Malayo-Polynesian and Australian migrantsarriving by sea at different times during pre-Columbian history wasreceived with more circumspection, especially in Anglo-Saxon circles(Pineda Camacho 1996: 62).20 However, the 1943 publication of Lesorigines de l’homme américain in Spanish, which came out just a fewmonths after the French Canadian edition, was received with greatenthusiasm in Latin America. Based on the biological hybridisationhypothesis and the sociological law of substitution,21 Rivet’s métissagetheory helped Creole intellectuals valorise their mestizo heritage. Thanksto Rivet, they could now form a positive image of their multi-strandednational identities, rooted in a long history of mixed ethnic origins andcultural achievements, and see these as contributing to the universalcultural heritage of humankind (Rival ms). It is worth noting thatClaude Lévi-Strauss immediately wrote a very favourable review of Lesorigines de l’homme américain for Renaissance22 in which he lauds Rivet’serudition, lucidity and ‘positively audacious will to speculate’, as well ashis critical stand against ‘timorousness and orthodoxy’. PinedaCamacho (1996: 60) remarks that Rivet’s thesis on the multi-ethnicpeopling of the American continent, far from being an eccentric flightof fancy, was a logical diffusionist hypothesis that had first beenformulated by the Vienna school. Interestingly, various authors (forinstance, Bellwood 2005, Hornborg 2005) are currently working ongrand syntheses of the kind proposed by Rivet and putting forward newhypotheses about cultural identity and migratory movements, whichRivet would doubtlessly recognise as akin to his own theoretical efforts.

The legacies of Rivet’s work are diffuse and varied. As noted in theintroduction to this chapter, the most fascinating and striking fact isthat, whereas Rivet is treated as an anthropological ancestor inColombia and remembered with much respect, admiration andgratitude in various other Latin American countries, he has beenalmost entirely erased from the collective memory of the Frenchanthropological community today, despite the fact that he was Frenchanthropology’s chief guiding spirit for more than three decades.Whether trained in Paris, Ecuador or Colombia, Rivet’s former LatinAmerican students have left a wealth of testimonies, which speak insurprisingly similar terms about his teaching style, theories and vision

144 Laura Rival

of anthropology.23 Although more research is needed on this point,what these testimonies seem to suggest is that Rivet’s anthropology, ashe taught and expounded it in his academic writing and as it informedhis politics, resonated profoundly with the ideals and values of LatinAmerican Creole intellectuals of the time. And if there is such a matchbetween their mestizo consciousness and Rivet’s anthropology, it is notsimply because they broadly shared the same political culture but also,and more importantly, because Rivet’s anthropology was in many waystheir anthropology as well (Rival ms). Rivet’s humanism resonated withthe humanism of the friends he made in the field. With them, hedebated not only the history of Ecuador but also the concepts ofprogress, welfare and commonwealth, the good society, the humancondition, and many other issues that engaged anthropology publiclyas a new form of humanism. Rivet could take part in a fruitful two-wayanthropological dialogue with his ‘secondary informants’ and friends,for there was no obvious colonial impediment to such conversations.Progressive Creoles and mestizos were as interested as he was in whathumans have in common, their common needs or common psychicunity, independently of their social condition or geographical location.For them as for him, the aim of anthropology was to teach the historyof mankind, that is the ways in which humans had moved around theworld, exchanged ideas and goods, learnt from each other andintermarried to create better societies. In this sense, Rivet’s humanismalso represented a Latin American aspiration to humanism.24 Rivet’sobjective studies of the material aspects of human life demonstrated tohis Latin American audience that their heritage contributed directly tohumankind’s global heritage. Put differently, Rivet’s humanism andthat of his progressive Latin American intellectual friends wereinseparable from a ‘civilisation politics’, a politics deeply rooted inmodernist humanistic values, which led him (like Boas) to embark on anempirical and moral critique of racism.

Rivet, whose broad vision of anthropology gave equal importanceto material culture (contemporary and past), linguistics and what wewould today call physical and biological anthropology,25 would haveagreed with Lévi-Strauss’s definition of ethnology as ‘neither a separatescience, nor a new one. It is the most ancient, most general form ofwhat we designate by the name of humanism’ (Lévi-Strauss 1978:272). As Johnson (2003) reminds us,26 structural anthropology wasborn from Lévi-Strauss’ epistemological battle to define the nature andscope of anthropology and its relationship to other academic sciences.However, as they both ignore the debates that shook Frenchanthropology before Lévi-Strauss’s ascent, they cannot appreciate thefact that the latter’s battle had been preceded by Rivet’s battle, nor that


both men’s battles were directly linked to their respective claims thatanthropology was a new kind of humanism, as the change of namefrom anthropology to ethnology and back to anthropology (althoughstructural this time) amply reveals (see endnotes 8, 13).

It is not difficult to find lines of continuity between the two Frenchscholars, both Americanists and admirers of Boas, both equallyextolling the value of linguistic studies and the importance of artefacts,and both equally concerned with the relationship between history andthe production of cultural difference. Whereas one provided Frenchanthropology with the institutional framework it needed to startexisting as an independent, nationally and internationally recognisedfield of investigation, the other gave it the coherent and rigoroustheoretical framework it lacked. Much more, of course, will need to besaid on the matter. In a way, structural anthropology would not haveexisted without Rivet’s ethnology, for Rivet asked the questions thatLévi-Strauss tried to answer: Where do the natural sciences end and thecultural sciences begin? How best to unify natural and culturaldeterminisms methodologically? To what extent can the methodologicalapproaches used in the natural sciences be applied to the social sciences,and vice versa? What do we all share as members of the same humanspecies, and what makes us culturally different?

It is also not difficult to see what impelled Lévi-Strauss to create adistance between his ‘field philosophy’ and Rivet’s twin concern withthe natural history of humankind and the history of societies. As wesaw earlier, Rivet’s fieldwork was emphatically not sociological: hepreferred to classify, order and organise, rather than take stock of nativesignificances. The most important task for (structural) anthropology,as Lévi-Strauss envisaged it, was to make historians, philosophers andthe public at large accept that there is more than one way of conceivinghumanity and its relationship to the world. As Dias (1991: 242) puts it,anthropology has always been a ‘domain of research torn apart’ (‘undomaine de recherche écartelé’). And as anthropology finds itself yet againat a crossroads, unsure of its epistemology, its field researchmethodology, its intellectual mission or the kind of humanism it shouldbe defending (Piña Cabral 2005, 2006; Bloch 2005), there is much tobe learnt, as I hope to have shown in this essay, from Paul Rivet’shumanist positionings and visions for the common ground.

146 Laura Rival

Notes

1. This reconstruction of Rivet’s life owes much to Christine Laurière’s fine intellectualbiography (Laurière 2006), which I read after having completed my own research. Iwould also like to thank all the French and Ecuadorian colleagues who shared withme their memories of Rivet’s life and their knowledge of his work.

2. See León (1958: 312–14) for an exhaustive list of scientific names incorporating theword ‘riveti’. Rivet’s first collections included parasites dangerous to humans anddomesticated animals, and insects involved in propagating a range of diseases andplagues (ibid.: 309). These collections allowed him to produce the first entomologicalstudies ever realised in Ecuador (Aráuz 1958: 75–76).

3. ‘Recherches sur le prognathisme’, published in L’Anthropologie 21 (Rivet 1909).4. The term ‘ancient ethnography’, coined by Rivet, was only used by himself and a few

associates (Laurière 2006: 146, 155).5. He was never to publish, for example, his study of the remarkable collection of 800

pre-Columbian ceramics he had brought back from Ecuador. This collection, currentlystored in the Musée du Quai Branly, has never been studied or exhibited. However,Rivet continued to write about pre-Columbian gold smelting after the war.

6. Rivet received an impressive number of military decorations for his services to thenation during the First World War (Araúz 1958: 31–32), and almost left academiaafter it (Laurière 2006: 454–79).

7. From 1909 to 1941, Rivet wrote in French to Boas, and Boas replied in English. Theletters cover a wide range of topics, from linguistics, politics, the development ofinternational Americanism, anthropology and racism, the rise of fascism in Germany,nationalism and internationalism, fund-raising for publishing, and more. In afascinating exchange of letters written during 1919, we learn that Rivet invited Boasto contribute an article to the Journal de la Société des Américanistes. Boas replied thathe would, but in German. Rivet wrote back saying that he could not accept an articlein German under present circumstances, as every one knew that Boas spoke andwrote fluently in both English and Spanish. It would just be too provocative, and stirup hatred among those whose motion he had just defeated. See archived letters in theMusée de l’Homme: 14/08/1919 [FB to PR]; 04/09/1919 [PR to FB]; 09/10/1919[FB to PR].

8. See Dias (1991) and Laurière (2006) for a discussion of the changing meanings of‘anthropology’, ‘ethnology’, and ‘ethnography’ in France in the pre-Lévi-Straussianera. Johnson (2003: 12), who does not seem to be aware of this most central debatein the history of French anthropology, attributes Lévi-Strauss’s reworking of the threemodes into three different stages of anthropological enquiry to the latter’s effort toreposition anthropology in relation to other human sciences, in particular historyand philosophy.

9. The series Travaux et mémoires de l’institut d’ethnologie (Laurière 1999: 114) wasfounded the following year, in 1926. See Dias (1991: 71–72) for an outline of whatwas taught by Mauss and others at the turn of the century. Very little is known aboutthe professional relationships or intellectual affinities between Rivet, Mauss and Lévy-Bruhl, or about the discussions they must have had while designing the first teachingprogramme. Laurière (2006: 496–97) found letters of students comparing theesoteric, hard-to-follow teaching of Mauss with the limpid, structured quality ofRivet’s lectures. There are no records of Lévy-Bruhl’s teaching.

10. On 3 December 1928, Rivet wrote to Boas (on official MNHN headed paper): ‘I feelemasculated by the reorganisation of the lab at the MNHN and that of the Trocadero.Add to this the directorship of the Institute of Ethnology and that of the Journal of


the Americanist Society of Paris, and you’ll understand why my days fill so quickly’(my translation).

11. This arrangement was to last until the great university reforms of 1968. The Muséede l’Homme, successor to the Trocadéro, became a research unit (‘laboratory’) of theMNHN, which continued to have two anthropology chairs, one in prehistory and onein ethnology, the latter being held by the director of the Musée (Michel Izard, personalcommunication, December 2004).

12. I found a letter from Rivet to Boas sent on official Société des Américanistes-headedpaper dated 23 February 1924, which might be the first mention of a joint Franco-North American encyclopaedia project. After congratulating Boas on the Handbookof American Indian Languages, which he judges ‘absolutely essential reading’, andexpressing the hope that the Smithsonian Institution will fund the publication, despitecurrent economic difficulties, Rivet adds: ‘I would also like to take advantage of thenext Americanist congress to ask the Smithsonian delegates whether they wouldconsider favourably the proposal to produce a handbook of Central and SouthAmerican Indians, very much in the same style as that produced by Hodge for NorthAmerica. I am convinced that many European scholars (Nordenskiöld, Koch-Grünberg, Lehmann, Preuss, Krickeberg, Karsten, myself) would readily and happilycollaborate. We could also ask a few South American colleagues to contribute. Whatdo you think?’ (my translation).

13. Renamed the National Institute of Anthropology after Rivet’s death. 14. Rivet wrote his early publications in his capacity as ‘a medical doctor attached to the

French geodesic mission’.15. Laurière (2006: 101) quotes an unpublished document in which Rivet powerfully

summarises his view: ‘rags are not picturesque’ (‘le haillon n’est pas pittoresque’).Pineda Camacho (1996: 61) mentions letters Rivet wrote to the Colombiangovernment, urging officials to act against the dire poverty in which the country’sindigenous population was living, a remedy without which no truly nationalintegration or democracy could develop. Analytical tensions between ‘culture’ and‘poverty’ still pervade much contemporary work on Amerindians (see, for instance,Hall and Patrinos 2006, Kalt et al. 2008).

16. Rivet’s medical profession opened many doors to him (Zerilli 1991–93: 358), inparticular that of Mons. Gónzalez Suárez, Archbishop of Ibarra, a colonial city northof Quito. This scholar-priest, who had a remarkable knowledge of Ecuador’s historyand prehistory, became Rivet’s friend (Aráuz 1958: 77–78) and informal teacher,advising him on practically all his wide-ranging interests: linguistics, physicalanthropology and pre-Columbian material culture. Rivet publicly acknowledged hisintellectual debt to Gónzalez Suárez (ibid.: 15–17), including an article for the Journalde la Société des Américanistes in 1919. I doubt that Rivet would have agreed withLeón’s assertion (1958: 316) that he (Rivet) was the founder of anthropology inEcuador. He would have demanded that the title be shared with Mons. GónzalezSuárez and with another of his Ecuadorian friends, the historian Jacinto JijónCaamaño.

17. For example, he enthusiastically describes fire-making and other techniques to lighthouses at night (1907: 588–89), giving the impression that, in addition to havingwitnessed these techniques, he tried them out himself.

18. This, inter alia, is the opinion of Olivia Harris and of Carmen Bernand (personalcommunications, July 2005, December 2006), the latter adding: ‘his four volumeson the Quechua language are remarkable. No one has done better since’ (mytranslation).

19. Rivet was much more admiring of Boas, whom he treated as a master (Rivet 1958),than Boas was of Rivet, whose work Boas hardly referred to.

148 Laura Rival

20. In a hand-written letter sent by Rivet to Boas from the MNHN anthropology lab on 13February 1925, Rivet expresses his regret that Boas had not accepted his conclusionsregarding the genetic links between Malayo-Polynesian, Australian and Amerindianlanguages: ‘Of all those who saw the evidence I have marshalled to support my thesis,you are the only one who is raising doubts […] I still hope to convince you in the nearfuture with the publication of more detailed studies, especially on the Yuma group’(my translation). Although Rivet (1932) was invited to give the Frazer Lecture inOxford in 1930 on Les Océaniens, no one in Britain was convinced by hisdemonstration that the Australian Aborigines, Tasmanians, Melanesians,Polynesians, Micronesians, Indonesians, Munda and Khmer formed a single ethniccomplex sharing a common linguistic stock.

21. Cultures do not merge into or mix with each other, but one replaces, that is takes over,the other. See Jamin (1989: 288).

22. Reproduced in its entirety in Laurière (1999: 127).23. See, for instance, Araúz (1958: 39), Pineda Camacho (1985: 11–12), Larrea (in

Araúz 1958), Dussán de Reichel (1984), Duque Gómez (1958), Valera (in Araúz1958), Santiana (in Araúz 1958) and Chevasse (in Araúz 1958). Carmen Bernand(December 2006), another Latin American citizen and anthropologist, mentions that:‘It is while reading The origins of the American man as a young student in Buenos Airesthat I decided to become an anthropologist’ (my translation).

24. Wilder (2005), of course, was to look at Rivet’s project in terms of his defence of‘tempered colonialism’ in Africa and other regions of the Third World.

25. In a letter he sent to Boas on 14 February 1936, he said that he was preparing avolume on ethnology for a French encyclopaedia, adding: ‘that is, all the sciences ofmankind, anthropology, ethnography, sociology and linguistics’ (my translation).

26. Perhaps as a result of being under the spell of the received wisdom that there is ‘abefore and an after Claude Lévi-Strauss’, as Michel Izard told me in an interview, thusstressing the epistemological break, or radical discontinuity, initiated by structuralism.Bertholet (2003), like Johnson, and almost certainly for similar reasons, presents anahistorical version of history in which Lévi-Strauss appears as a total outsider toFrench anthropology, someone who learnt the trade in the USA. For Bertholet, Lévi-Strauss – who was first recognised professionally not in France but in the USA – is atrue heir to Boas. More anecdotal, yet revealing, is Bertholet’s (2003: 148–50, 173)narration of the famous New York dinner during which Boas died. Told in a way thatstresses the direct lineage between Lévi-Strauss and Boas, the story hardly mentionsRivet at all. Having similarly heard many French colleagues tell me, overwhelmed bythe symbolic power of the image, that Boas died in Lévi-Strauss’ arms, I did not payattention to Bertholet’s version until I re-read De près et de loin, where Lévi-Strauss(1988: 57–58) clearly states that the dinner had been organised in Paul Rivet’shonour and that, when Boas felt unwell, he was attended by Rivet, who, after all, wasa medical doctor. The power of myth is indeed … overpowering.

References

Allen, N.J. 2000. Categories and classifications: Maussian reflections on the social, Oxford:Berghahn.

Aráuz, J. (ed.). 1958. Homenaje a Paul Rivet, Special issue of the Boletin de las SeccionesCientíficas de la Casa de la Cultura Ecuatoriana, 10 (86).

Bellwood, P. 2005. First farmers: the origins of agricultural societies, Oxford: Blackwell.Bertholet, D. 2003. Claude Lévi-Strauss, Paris: Plon.


Bloch, M. 2005. Essays on cultural transmission, Oxford: Berg.Campbell, L. 1997. American Indian languages: the historical linguistics of Native America,

New York and Oxford: Oxford University Press.Castelain, J.-P., S. Gruzinski and C. Salazar-Soler (eds). 2006. De l’ethnographie à l’histoire.

Paris-Madrid-Buenos Aires: Les mondes de Carmen Bernand, Paris: L’Harmattan.Chevasse, P. 1958. Las grandes realizaciones francesas: Paul Rivet y el museo del hombre,

in J. Araúz (ed.), Homenaje a Paul Rivet, Special issue of the Boletin de las SeccionesCientíficas de la Casa de la Cultura Ecuatoriana, 10(86): 105–8.

d’Harcourt, R. 1958. Paul Rivet, 1876–1958, Journal de la société des américanistes, 47: 1–20.de Friedemann, N.S. 1984. Etica y politica del antropologo: compromiso professional, in

J. Arocho and N.S. de Friedemann (eds), Un siglo de investigación social antropologicaen Colombia, Bogota: Etno.

Dias, N. 1991. Le musée d’ethnographie du Trocadéro (1878–1908): Anthropologie etmuséologie en France, Paris: Editions du CNRS.

Dussán de Reichel, A. 1984. Paul Rivet y su epoca, Correo de los Andes (Bogota, Mayo-Junio), 26: 70–76.

Gónzalez Suárez, F. 1969 [1890–1903]. Historia General de la República del Ecuador, Quito:Edit. Casa de la Cultura Ecuatoriana.

Gruzinski, S. 2002. The mestizo mind: the intellectual dynamics of colonization andglobalization (tr. D. Dusinberre), New York and London: Routledge.

Haekel, J. (ed.). 1956. Die Wiener Schule der Völkerkunde, Vienna: F. Berger.Hall, G. and H.A. Patrinos (eds). 2006. Indigenous peoples, poverty and human development

in Latin America, Basingstoke and New York: Palgrave (Macmillan). Hornborg, A. 2005. Ethnogenesis, regional integration, and ecology in prehistoric

Amazonia: toward a system perspective, Current Anthropology, 46(4): 589–620.Jamin, J. 1989. Le savant et le politique: Paul Rivet (1876–1958), in C. Blanckaert, A.

Ducros and J.-J. Hublin (eds), Histoire de l’anthropologie: hommes, idées, moments,Bulletins et Mémoires de la Société d’Anthropologie de Paris, 1(3–4): 277–94.

Johnson, C. 2003. Claude Lévi-Strauss: the formative years, Cambridge: CambridgeUniversity Press.

Kalt, J.P. (and other members of the Harvard Project on American Indian EconomicDevelopment) 2008. The state of the native nations: conditions under US policies of self-determination, New York: Oxford University Press.

Kerns, V. 2003. Scenes from the high desert: Julian Steward’s life and theory, Urbana andChicago: University of Illinois Press.

Landaburu, J. (ed.). 1996. Documentos sobre lenguas aborígenes de Colombia del archivo dePaul Rivet, Volumes I and II, Santa Fe de Bogotá: Ediciones Uniandes.

Laurière, C. 1999. Paul Rivet, vie et oeuvre, Gradhiva, 26: 109–28.——— 2006. Paul Rivet (1876–1958): le savant et le politique, unpublished doctoral

thesis, Paris, EHESS.Leach, E. 1957. The epistemological background to Malinowski’s empiricism, in R. Firth

(ed.), Man and culture: an evaluation of the work of Bronislaw Malinowski, London:Routledge and Kegan Paul.

León, L. 1958. Contribución del doctor Paul Rivet al conocimiento científico de larepública del Ecuador, Miscellana: Acts of the 31st International Congress ofAmericanists, Vol. 1, Mexico City: Universidad Autónoma de México, 305–21.

Lévi-Strauss, C. 1978. Structural anthropology II, Harmondsworth: Penguin.——— 1988. De près et de loin: entretiens avec Didier Eribon, Paris: Odile Jacob. Lottman, H. 1982. The left bank: writers in Paris from Popular Front to Cold War,

Heinemann: London.Piña-Cabral, J. 2005. The future of social anthropology, Social Anthropology, 13(2): 119–28.

150 Laura Rival

——— 2006. ‘Anthropology’ challenged: notes for a debate, Journal of the RoyalAnthropological Institute, 12(3): 663–73.

Pineda Camacho, R. 1985. Paul Rivet y el Americanismo, Texto y Contexto (Universidadde los Andes, Bogotá), 5: 7–19.

——— 1996. Paul Rivet: un legado que aún nos interpela, in J. Landaburu (ed.),Documentos sobre lenguas aborígenes de Colombia del archivo de Paul Rivet, Santa Fe deBogotá: Ediciones Uniandes, vol. 2: 53–74.

Rival, L., ms. Land and people in the Ecuadorian Chocó: three political visions.Rivet, P. 1903. Etude sur les Indiens de la région de Riobamba, Journal de la Société des

Américanistes (n.s.), 1: 58–80.——— 1905. Les Indiens Colorado: récit de voyage et étude ethnologique, Journal de la

Société des Américanistes (n.s.), 2(2): 177–208.——— 1906. Cinq ans d’études anthropologiques dans la république de l‘Equateur:

résumé préliminaire, Journal de la Société des Américanistes, 3: 229–37.——— 1907. Les indiens Jibaros: étude géographique, historique et ethnographique,

L’Anthropologie, 18: 333–68 (1st part), and 583–618 (2nd part).——— 1908. Les indiens Jibaros: étude géographique, historique et ethnographique,

L’Anthropologie, 19: 235–59 (3rd part).——— 1909. ‘Recherches sur le prognathisme’, L’Anthropologie, 21: 505–18, 637–69.——— 1912. L’Ethnographie ancienne de l’Equateur: Mission du service géographique de

l’armée en Amérique du Sud, Paris: Gauthier-Villars.——— 1924. Langues américaines, in A. Meillet and M. Cohen (eds), Les langues du

monde, Paris: Edouard Champion, pp. 597–712.——— 1932. Les Océaniens, in W. Dawson (ed.), The Frazer Lectures (1922–1932),

London: Macmillan.——— 1943. Les origines de l’homme américain, Montréal: Les Editions de l’Arbre.——— 1951–54. Bibliographie des langues aymará et kicua, Paris: Institut d’Ethnologie.——— 1958. Tribute to Franz Boas, International Journal of American Linguistics, 24(4):

251–52. Rochereau, H.J. 1958. El professor Rivet y sus corresponsales, in L. Duque Gómez (ed.),

Homenaje al Profesor Paul Rivet, Bogotá: Editorial ABC, 2–7.Stocking, G. 1987. Victorian anthropology, New York: Macmillan Press.——— 1992a. The Boas plan for the study of American Indian languages, in G. Stocking

(ed.), The ethnographer’s magic and other essays in the history of anthropology, Madison:The University of Wisconsin Press.

——— 1992b. The ethnographer’s magic: fieldwork in British anthropology from Tylorto Malinowski, in G. Stocking (ed.), The ethnographer’s magic and other essays in thehistory of anthropology, Madison: The University of Wisconsin Press.

Uribe T. and A. Carlos. 1996. Entre el amor y el desamor: Paul Rivet en Colombia, in J.Landaburu (ed.), Documentos sobre lenguas aborígenes de Colombia del archivo de PaulRivet, Santa Fe de Bogotá: Ediciones Uniandes, vol. 1: 52–73.

Wilder, G. 2005. The French imperial nation-state: negritude and colonial humanism betweenthe two world wars, Chicago: The University of Chicago Press.

Zerilli, F. 1991–93. Il terreno ecuadoriano di Paul Rivet: antropologia, linguistica,ethnografia, Annali della facoltà di Lettere e Filosofia 2: Studi Storico-Antropologici, 29–30: [n.s. 15–16, 1991/92–1992/93]: 353––96.

——— 1998. Il lato oscuro dell’etnologia: il contributo dell’ antropologia naturalista al processodi instituzionalizzazione degli studi etnologici in Francia (D.E.A. 10), Rome: CISU.

Chapter 6

ALFRED MÉTRAUX:EMPIRICIST AND ROMANTICIST

Peter Rivière

Although Alfred Métraux was of Swiss nationality, I think it is correctto say that most people think of him as a French anthropologist andethnographer. He was born in Lausanne in 1902 but spent much of hischildhood in Mendoza, Argentina, where his father, also Alfred,practised as a surgeon from 1907 to 1954. He received all his secondaryeducation in Europe, firstly in Switzerland and then in France. In 1921he enrolled at both l’École des Chartes and l’École Pratique des HautesÉtudes, Section des Sciences Religieuses. In 1922, aged twenty, heobtained eight months’ leave of absence and returned to Argentina toundertake his first fieldwork. He spent the time studying the Calchaquí,a subgroup of the Diaguita, in northwest Argentina, as well astravelling in Peru, Bolivia and Chile. Back in France by late 1922, inJanuary or February 1923, he admitted to Georges Bataille his intentionof becoming an ethnologist specialising in South America (Le Bouler1992: 136). In 1925 he obtained a degree at l’École des LanguesOrientales, in the Section des Langues Africaines, and two years later hewas back at l’École Pratique des Hautes Études, where he studied withMarcel Mauss and Paul Rivet. At this time he also went to study inSweden, at Gothenburg, under one of the leading Americanists of thetime, Baron Erland von Nordenskiöld, who seems to have had a farstronger influence on him than either of the French anthropologists.The thesis which he submitted for his degree at l’École Pratique desHautes Études was La civilisation matérielle des tribus tupi-guarani(1928a). He obtained his Docteur ès-Lettres from the Sorbonne in 1928with a thesis on La religion de Tupinamba et ses rapports avec celle des autrestribus tupi-guarani (1928b).

152 Peter Rivière

In 1928 he was invited to the National University of Tucumán inArgentina, where he founded the Instituto de Etnología and served asits director until 1934. During those years he was able to carry outextensive fieldwork among the peoples of the Gran Chaco, theChiriguano, Mataco and Toba, extending his work in 1932 to the Uru-Chipaya of the Bolivian Altiplano. At the beginning Métraux was happyat Tucumán and in December 1930 was able to write: ‘My institute andmy journal1 are the sole living organisms in this university, and myactivity is regarded as a good element of propaganda’.2 But only sevenmonths later he complained: ‘My situation is far from being as brilliantas last year because of the frightful economic disaster in Argentina’.3 Hefinally gave up in 1934, writing the following year: ‘I have left Argentina,where, as a consequence of the idiosyncratic nature of the country, myactivity risks becoming completely paralysed. I have escaped the mostappalling stagnation’4 (Auroi and Monnier 1996: 14–17).

In 1934–35 he participated, at the invitation of Mauss and Rivet,in a Franco-Belgian expedition to Easter Island. On his return journey,he visited various Pacific islands, including Hawaii. Between 1936 and1938 he was back in Hawaii with an appointment at the Bernice BishopMuseum, Honolulu. In 1938 he was a Visiting Professor at theUniversity of California and during 1939–1941 he was based at Yale,returning for fieldwork to Argentina and Bolivia, as well as beinginvolved with the Human Relations Area Files. In 1941 he joined the

Figure 6.1. Alfred Métraux, seated second from right, doing fieldwork among Chipayain Bolivia, 1931 or 1932. Courtesy of Harold Prins.

staff of the Bureau of American Ethnology of the SmithsonianInstitution in Washington and for the next four years was busy helpingwith the production of the Handbook of South American Indians(hereafter HSAI). During this period he was also Visiting Professor atthe Escuela Nacional de Antropología in Mexico City in 1943 and madetwo visits to Haiti (1941 and 1942). At the end of the Second WorldWar (he had become a United States citizen before the country enteredthe war), he went to Europe as a member of the United States BombingSurvey. It may have been his experiences on this that turned hisattention to applying anthropology to intercultural and interracialunderstanding. He joined the United Nations Department of SocialAffairs in New York in 1946, being seconded to UNESCO in Paris thefollowing year and joining that organisation in 1950.5 He remainedwith UNESCO until his retirement in 1962. During this time he travelledwidely, to Africa and India as well as the New World. He was engaged invarious research projects, including the Hylean Amazon Project in1947–48, research into setting up basic rural education in Haiti from1948 to 1950, into race relations in Brazil in 1950–51, and in 1954,in conjunction with the International Labour Office, into Aymara andQuechua Indian migrations in Bolivia and Peru. He was appointedProfessor of the Anthropology of South American Indians at L’ÉcolePratique des Hautes Études in Paris in 1959. He died, by his own hand,near Paris, on 12 April 1963.

It is worth considering here whose ideas most influenced Métraux’swork. It would be difficult to argue that Mauss left any lasting impressionon it. I have found no reference to Mauss’s work in any of Métraux’s, andhis method is very far from being Maussian. Although La religion deTupinamba is dedicated to Mauss, it is difficult to imagine the latter puttingso much work into a topic and producing no more than a compilationfrom literary sources. Perhaps one brief Maussian moment, an echo of Ledon, is a very short notice (1929b) drawing attention to two little-knownhistorical texts on the potlatch in Florida and the Gran Chaco.

Except very early in Métraux’s career, Paul Rivet does not seem tohave had much influence on him either. Rivet held that South Americawas peopled by Malayo-Polynesian and Australian peoples in relativelyrecent times, and this may well explain an early article by Métraux, ‘Lebâton de rythme: Contributions à l’étude de la distributiongéographique des éléments de culture d’origine mélanésienne enAmérique du Sud’ (1927a). Rivet clearly thought highly of Métrauxand wrote a glowing reference for him when he went to the universityat Tucumán. He was Mauss’s and Rivet’s choice to go on the Franco-Belgian expedition to Easter Island, and Rivet, who edited Volume 7 ofEncyclopédie Française in the early 1930s, chose Métraux to contribute

Alfred Métraux 153

three articles. It would appear, however, that they later fell out, forMétraux was to write to Lowie in 1940 that, ‘Whatever my pastconflicts with him [Rivet] have been, I thought it my duty to do my verybest to help him leave France and come here’ (Auroi and Monnier 1996:46). It is not clear what these conflicts were about, but in 1938 Métrauxhad published an article, ‘The Proto-Indian script and the Easter Islandtablets. (A critical study)’, in which he had firmly refuted an argumentput forward a decade earlier by Guillaume de Hevezy that there is a closeconnection between the script found during excavations in the IndusValley and signs on Easter Island tablets. Various other authors,including Rivet, had supported Hevezy’s position, which was effectively,and with a degree of mockery, dismissed by Métraux.

According to Wagley (1964: 604), Father John Cooper also greatlyinfluenced Métraux’s anthropological career. One of Cooper’s mainconcerns was with questions of cultural distribution and historicalreconstruction, and in his 1942 article, ‘Areal and temporal aspects ofaboriginal South American culture’, he was the first to divide thecultures of the area into three: marginal, silval and sierral. Métrauxdiscusses this schema in his ‘La civilisation Guyano-Amazonienne etses provinces culturelles’ (1946b) and, while pointing out the difficultyof fitting many tribes into it, later referred to Cooper’s article as ‘hismost important theoretical contribution’ (1950: 43). Cooper’s schemawas adopted as the structure of the HSAI, in whose preparationMétraux was closely involved, as Marginal, Tropical Forest and Andean,to which was added a fourth type, the Circum-Caribbean.

The figure from Métraux’s formative years who probably had themost powerful influence on him was Nordenskiöld. As has already beenmentioned, Métraux had worked with him in Gothenburg while still astudent in Paris in the 1920s. The influence that Nordenskiöld exertedwas not so much because he had worked among the peoples of the GranChaco, who were to form the focus of Métraux’s earliest research, butbecause of his method of reconstructing cultural history. This isexemplified by his ten-volume series, Comparative ethnographical studies(1919–1938), in which, using his own data and literary sources, heplotted a huge range of Amerindian culture traits. For example, inVolume 1 of the series, entitled An ethno-geographical analysis of thematerial culture of two Indian tribes in the Gran Chaco, his method involvesexamining where cultural items, such as dwelling places, beds,cultivated plants, agricultural tools, hunting weapons, fishing tackleand many others, found among the two Mataco groups in question, theChoroti and Ashluslay (or Nivacle), are distributed through SouthAmerica. The text is supported by 44 distribution maps covering suchobjects as wooden spades, bowstrings made from animal material and

154 Peter Rivière

Alfred Métraux 155

bird arrows and slings. Although not employed to the same extent,identically constructed maps are found in many of Métraux’s earlyworks. It is also clear that when Métraux uses the terms ‘analytic’ and‘comparative’, it is very much in the sense employed by Nordenskiöld.This is well illustrated by his 1937a article, ‘Easter Island sanctuaries(analytic and comparative study)’, in which he describes variations inthe form of ritual stone platforms and examines their distributionelsewhere in Polynesia.6

Métraux’s publications are numerous, and looked at chronologicallythey reflect his contemporary research interests, with, for example, thefocus on the Gran Chaco, Easter Island and Haiti predominating atcertain periods. His main declared interest in myth, magic and religionis well borne out in the list, but he also wrote about a wide range of othertopics.7 Whereas most of his articles appeared in academic journals, healso occasionally published in more popular publications, such as LaRevue de Paris, Paris-Soir and Natural History.8

The main bulk of Métraux’s publications falls naturally into twomain divisions: those pieces based on his own field research, and thoseon literary sources. This division is well illustrated by his collection ofessays entitled Religions et magies indiennes d’Amérique du Sud. Althoughpublished posthumously in 1967, he had started work on it before hisdeath, had chosen six of the nine chapters and had undertaken someeditorial work on four of them. The blurb quite explicitly states that thebook exhibits two essential aspects of Métraux’s work, those based onliterary sources, and those deriving from his own field data. I have apersonal interest here, for it so happens that this collection was one ofthe first books that I reviewed for The Times Literary Supplement (Rivière1967). I still have the notes I made when preparing the review. On thelibrary-based essays, I quoted the author himself, who wrote: ‘the aimis not to provide new facts, but to bring together scattered data andpresent them in an order that facilitates understanding’. I noted thatthe pieces in the collection based on Métraux’s own field researchconsisted of little more than ethnographic reportage, almost devoid ofany theoretical or interpretive framework. Conclusions, where theyexist, are little more than a summary of the facts that have gone before.

It is worth looking at these two aspects of Métraux’s production in alittle more detail. For many South Americanists, Métraux is best knownfor his publications compiled from the existing historical andethnographic literature. He himself rated such work highly and, in anobituary for Father Cooper, he wrote:

The particular conditions of South American ethnography are such that aresearcher who has never been in the field can, nevertheless, undertakeoriginal research likely to advance the science, if he has a taste for

156 Peter Rivière

scholarship and a critical sense. There exists in fact, on the extinct Indiancultures and on those which still subsist, an enormous body of informationoften of the highest order, which is scattered through travel narratives, therelations of missionaries, and historical documents. In patiently collectingthese texts and in interpreting them in the light of modern ethnography, itis possible to penetrate very deep into the American past and to resolveextremely important problems. Furthermore the accumulation ofdocuments on indigenous cultures is justified only if it provides material forwider and wider synthesis. (1950: 39–40)

His most quoted works appear to be his numerous contributions to theHSAI. That he played a central role in its production is acknowledged bythe editor, Julian Steward, who, at the end of the introduction toVolume 1: The marginal tribes, wrote:

A special word of gratitude, however, is due to Dr. Alfred Métraux. Theextent of his contribution is by no means indicated by the large number ofarticles appearing under his name.9 With an unsurpassed knowledge ofSouth American ethnology and ever generous with his time, his advice andhelp to the editor and contributors alike have been a major factor in thesuccessful completion of the work. (1946: 9)

On the other hand, probably the most influential of such publicationswas one of his earliest, his thesis on Tupinamba religion (1928b). It is thechapter in this on Tupinamba cannibalism that started a line of interestand argument that can be traced via Florestan Fernandes (1963) andmany other authors to Eduardo Viveiros de Castro’s notion of thesymbolic economy of predation, later revised to alterity, andperspectivism – the latest position in Amazonian ethnography. I foundMétraux’s ethno-historical study of Tupí-Guaraní migrations (1927b)extremely useful when conducting a literature survey of Guianeseethnography in 1962, and Pierre Grenand credits it with opening ‘thecontemporary path to research on the lowlands’10 in his ethno-historicalstudy of the Wa api (1982: 45). In the 1940s, there were a whole seriesof works of synthesis in which Métraux put together a great deal ofinformation on specific topics. These included Auraucanian shamanism(1942), tropical South American shamanism (1944b), the cause andmagical treatment of sickness among Tropical Forest people (1944a),shamanism in the Gran Chaco (1945), supreme gods and culture heroes(1946d), mourning rites and burial forms (1947), and the distribution ofcertain mythic motifs (1948a). These have proved rather less successfulinsofar as, unlike his work on Tupinamba cannibalism, they have gonevirtually unnoticed by subsequent writers on these subjects. Perhaps thereason for this is that these topics are – unlike Tupinamba cannibalism,which no longer exists – still there to be studied, and are studied.11

Alfred Métraux 157

Just before we turn to consider Métraux’s field research, it might be worthnoting that in his obituary of Métraux, Charles Wagley remarks that whilehe, Métraux, ‘valued field ethnography more than theory’, in Wagley’sopinion, ‘His great strength was ... synthesis and historical research’, and herefers to his two works on the Tupinamba as ‘classics’ (1964: 606).

Métraux’s fieldwork and the publications based on it will now bedealt with in greater detail, for that is the focus of the present volume.As already mentioned, Métraux undertook a great deal of field researchin a wide range of different places. There are two points about his fieldresearch which might be made here. First, he rarely worked alone; moreoften than not he had at least one companion, and often more. Secondly,nowhere did he undertake any period of extended fieldwork, five or sixmonths being the longest. In fact, Wagley ‘felt that he was too restlessand too eager to be on his way to produce detailed and lengthy fieldreports’ (1964: 606). Some field research consisted of little more thanshort visits, but even so he published, at least briefly, on most of them.12

For the purposes of this assessment of Métraux as ethnographer, I shallconcentrate on three areas: his work in the Gran Chaco andneighbouring regions, his visit to Easter Island, and his study of Haiti.

His earliest fieldwork in Argentina, as already mentioned, was in 1922among the Calchaquí, and the information obtained on that expeditionwas published in the first volume (1929a) of his Revista, where alsoappeared the main accounts resulting from the field research undertakenamong the Chiriguano, Mataco, Pilaga and Toba of the Gran Chacoduring his years at Tucumán. These included ‘Études sur la civilisationdes Indiens Chiriguano’ (1930) and ‘Les hommes-dieux chez lesChiriguano et dans l’Amérique du Sud’ (1931), both based on fieldworkcarried out between February and June 1929. In the southern summerof 1930–31, he visited the Chipaya, an Uro-speaking people of theBolivian Altiplano, and consequently published ‘La religión secreta y lamitologia de los Indios Uro-Chipaya de Carangas (Bolivia)’ (1935c) and‘Civilización material de los Indios Uro-Chipaya de Carangas (Bolivia)’(1935a). His publications from this period of his life were by no meansrestricted to Revista. He published regularly in Journal de la Société desAméricanistes de Paris, ‘Contribution au folk-lore andin’ (1934) being anexample; and in other Argentinian journals such as Revista del Museo deLa Plata, where ‘Mitos y cuentos de los Indios Chiriguano’ (1932)appeared. He undertook further fieldwork in 1932 among the Toba-Pilaga, which gave rise to ‘Études d’ethnographie Toba-Pilaga (GranChaco)’ (1937b). He also made collections of myths, publishing 123Mataco myths (1939), and 106 Toba and Pilaga myths (1946a).

It should be emphasised that, although these are journal articles, theyoften approximate in size to an ethnographic monograph. ‘Études sur la

158 Peter Rivière

civilisation des Indiens Chiriguano’ runs to 189 pages plus 141 figures and93 whole plates. However, ‘La civilisation’ of the title in his earliest works isrestricted almost exclusively to subsistence and material culture. Only thelast of the fourteen chapters into which it is divided, that on the treatmentof illness, does not fit this pattern. A sample of the titles of other chaptersgives a flavour of the work: ‘L’agriculture’, ‘La cuisine et l’alimentation’,‘Outils des Chiriguano’, ‘Les calebasses’, ‘Le tissage’ and so on. Many of theitems and techniques described are fully illustrated. There is nothing aboutkinship, residence, social organisation or any other sociological matter. Theonly deviation from straight description is to be found in the chapter onbasketwork where there is a discussion of the distribution of a certaintechnique in a manner reminiscent of Nordenskiöld’s work. The lattermethod also occurs in his article on the Calchaquí (1929a) with referenceto the distribution of subterranean and semi-subterranean huts in SouthAmerica. There is a brief reference to Frazer’s ideas in his ‘Les hommes-dieux’ (1931: 63), but this is a unique exception. The overall result ismeticulously detailed but rather old-fashioned ethnography.

Within a few years, however, a change is noticeable and theemphasis is increasingly on social institutions. His ‘Les Indiens Uro-Chipaya de Carangas’ (1935d) contains material on social organisation(post-marital residence, filiation, etc.) and even a table of kin terms,although without any attempt to see whether they formed any sort ofpattern. His ethnography of the Toba-Pilaga (1937b) covers such topicsas religion, dreams, puberty, pregnancy and birth, funerary rites, feasts,marriages, kin terms, authority, war, property and games.

Métraux’s next field site was Easter Island (1934–35), and besides anumber of journal articles – including his first piece (1936b) to appear inEnglish (on Easter Island numerals) – the main work that emanated fromthis research was the substantial monograph, Ethnology of Easter Island(1940). Once again this is a work of straightforward ethnographicreportage with no attempt at analysis or interpretation, although thereare a few brief attempts to compare Easter Island practices with those ofwider Polynesia. The largest proportion of the text, some 150 pages outof 412, is given over to a description of subsistence activities and materialculture, including, of course, stone-working. The section on sociologicaltopics, which is given relative prominence and fifty-three pages, includesaccounts of the life-cycle, social organisation, property rights, war andcannibalism. The account of religion occupies thirty-three pages mainlytaken up with lists of gods, whereas the discussion of religious ritualreceives just two pages. There are twenty-seven pages of ‘Tales’.

As well as this detailed ethnographic report on Easter Island, he alsoproduced a much more popular and readable work, explicitly ‘notaddressed either to archaeologists or anthropologists’. The first edition of

Alfred Métraux 159

this, L’île de Pâques, was published in Paris in 1941;13 a revised andexpanded edition appeared in 1951, with an English translation of afurther expanded version, entitled Easter Island, in 1957. The mainreason for this expansion was to take account of Thor Heyerdahl’s Kon-Tiki voyage and his theory that Easter Island and Polynesia were peopledfrom the Americas, a position with which Métraux firmly disagreed.14

Finally, there is the research he conducted on Haiti. This can reallybe divided into his work connected with the UNESCO scheme for theintroduction of basic education into rural areas and that on voodoo. Iam mainly going to ignore the former here, which gave rise to a rangeof publications from Haïti: la terre, les hommes et les dieux (1957b),15

little more than a well-illustrated travelogue, to articles such as ‘Étudesur l’agriculture paysanne dans une vallée haïtienne’ (1948b) and‘Droit et coutume en matière successorale dans la paysanneriehaïtienne’ (1951a). These are once again straightforward ethnographicreporting although the former, which is based mainly on informationsupplied by the agronomist, Edouard Berrouet, concludes with eightrecommendations for development in the area.

When we turn to voodoo, there is a whole series of articles inacademic journals covering many aspects of the topic, from the conceptof the soul (1946c) through animistic beliefs in voodoo (1952) tovoodoo mystical marriage (1956).16 Once again these are a matter ofplain ethnographic description. There is also a monograph, Le Vaudouhaïtien (1958), published in English as Voodo in Haiti (1959). This workcovers most of the topics dealt with in earlier journal articles, althoughthe style and form of presentation make it more readable than them.17

We have now briefly examined a representative cross-section ofMétraux’s publications based on fieldwork across a period of thirty years.Strikingly little has changed in that time. There is from beginning to endthe production of ethnographic description, but no analysis of thematerial of the sort that was becoming increasingly characteristicamong his European contemporaries. There are, however, some changesof emphasis. The earlier concentration on material objects with copiousdescription and illustration is gradually replaced by more attention beingpaid to topics such as social organisation and religion. What I findastonishing about his work is the almost total lack of reference to orconcern with any of the theoretical works on these subjects other thanthe occasional foray into cultural history. His publications are amplybacked up by references to numerous historical and ethnographic works,but with very few exceptions, these are used for the factual evidence theycontain rather than for any ideas their authors might have.

It would be difficult to believe that when he started his main periodof fieldwork in the Gran Chaco in the late 1920s, he was not familiar

160 Peter Rivière

with the most influential works of L’année sociologique. He seems,however, to have found them and any other theoretical formulationsirrelevant to his endeavours. This is evidence, perhaps, to supportWagley’s contention that he valued ethnography over theory; in fact, onoccasion one is left with the impression that he consciously ignoredtheory. One such occasion is his contribution to Encyclopédie Françaiseon ‘La structure sociale’ (1936a), which, in 1936, one might haveexpected to have had some theoretical content. In fact it has none. Thearticle consists of factual accounts of such topics, as ‘Family andmarriage’, ‘The clan’, ‘Property’, ‘Government’, ‘Law and justice’ and‘Economy’. Although the accompanying ‘Further readings’ includesMalinowski (Crime and custom), Bouglé (on caste), Radcliffe-Brown (onAustralian kinship), Lowie (Primitive society), Maine (Ancient law), Tylor(Researches into the early history of man) and van Gennep (on totemism),there is no discussion of the ideas contained in these works.

Nor, during the span of Métraux’s career, is there any lack of moreanalytical and interpretive studies based on fieldwork by otherfieldworkers which might have acted as models for him. In particular,just to limit the examples to France, there are the various works on theDogon, such as Marcel Griaule’s Masques Dogons (1938) and DenisePaulme’s Organisation sociale des Dogon (1940), or, from the Pacific,Maurice Leenhardt’s Do Kamo (1947). These three authors directlyengage, through their field data, with the ideas propounded bynumerous French, American and British anthropologists. Métrauxrefused to go down that path, and until the end of his career hecontinued to report the ethnographic facts as he saw them and recordedthem, without any reference to any wider theoretical notions. Forexample, as late as 1960, he published a collection of Kayapó myths(1960b) without a single word of introduction, interpretation orexplanation of how and when he had collected them.18

On the other hand, it might be said that Métraux’s work was notentirely in vain, for one of those Kayapó myths, that on ‘the origin offire’, is M8 in Lévi-Strauss’s The raw and the cooked (1970), the second Jêvariation on M1, the Bororo key myth which starts off the whole cycle.Lévi-Strauss also later drew further for Mythologiques on Métraux’scollections of Toba-Pilaga, Mataco and Chiriguano myths, but there areother authors who pay far less attention to his material. John Renshaw,in his The Indians of the Paraguayan Chaco (2003), describes his work as‘the first attempt to offer an overview of the Chaco since AlfredMetraux’s “Ethnography of the Chaco”’, published in the HSAI in 1946,but he makes no reference to any of Métraux’s numerous fieldwork-based publications. Métraux’s work on voodoo has suffered a similarfate. If one looks at more recent studies of the phenomenon, there is

Alfred Métraux 161

little reference to his work. For example, Wade Davis, in his The serpentand the rainbow (1986), makes two very brief references to him. Indeedtoday few people cite his fieldwork results.

There may, however, be a clue as to why Métraux confined himselfto description and eschewed interpretation. There are two publicationsin which Métraux expresses his views on fieldwork, ethnography andanthropology, one published at the beginning and the other at the endof his career. In 1925, while still a student at l’École des Languesorientales vivantes, but after his first field trip to the Calchaquí and afterhaving declared to Bataille his intention of becoming an ethnographer,he published as his second academic article, ‘De la méthode dans lesrecherches ethnographiques’. There are two points which arise fromthis article which deserve attention here. The first is that Métraux seesa clear division of labour within the anthropological endeavour whichharks back to the nineteenth century and Frazer’s ‘men in the field’,whose duty it was to collect the facts that the anthropologists, in theirarmchairs, could use in their theorising. In Métraux’s case it is betweenthe ethnographers and those he variously refers to as scholars, scientistsor sociologists. The nature of Métraux’s fieldwork publications whichwe have just reviewed indicates that he continued to subscribe to thisview throughout his working life; he saw his job as providing the factswhich would speak for themselves, as Wagley put it (1964: 606), or,perhaps, on which other people could theorise. This concern with factsrather than theories did not escape his contemporaries and friends.Michel Leiris, in his preface to the French edition of Le Vaudou haïtien(not included in the English translation), wrote:

In his work, Métraux seems like someone who cared above all for concreteknowledge and for whom the study of societies was, not a path opening outinto theoretical insights, but a way of knowing men and of approachingthem as nearly as possible, in all the diversity of their usages and customs.(1958: 7)19

The second point that can be taken from the 1925 article is that he appearsto have a remarkably pessimistic view of the ethnographic endeavour.Basically he argues that the science of society must rest on a body of secureand precise facts, but it is questionable whether these can be collected. Thereason for this is that the civilised mind is unable to comprehend theworkings of the primitive mentality. Indeed, the only author he refers to isLévy-Bruhl.20 To the question, ‘Do ethnographic facts have the high degreeof precision and exactness indispensable to science?’, he answers:

If, as has been believed and continues to be believed, there is an identitybetween primitive mentality and civilized mentality, the response to this

162 Peter Rivière

question cannot be other than affirmative. This universality of humanreason has implicitly been admitted by the British sociological school, whichis responsible for the illusion shared by many scholars regarding the quasi-absolute value of the result of these investigations. The critique presented byLévy-Brühl in these two works, les Fonctions mentales dans les sociétésinférieures and la Mentalité primitive, has cast a legitimate suspicion over thispostulate. The analysis of the representations of primitives has revealed theirfundamental differences from those of the civilized. ‘Wild’ humanityappears like a closed world, impermeable to our experience, which remainsincomprehensible so long as one attributes those who create it with mentalprocesses similar to ours. This misunderstanding, demonstrated withevidence only in the past fourteen years, is the reason for the imperfectionof evidences obtained from the natives and will affect the results ofethnographic enquiries to come, as well as those that have already beenprovided to us. (1925: 276–77)21

The article continues with a litany of the difficulties and obstacles thatimpede the collection of accurate and reliable information. One mightbe led to think that this is rather a paradoxical position, given theauthor’s declared commitment to ethnography. However, he goes on toargue that, with sufficient ability, patience and the ingenuity to adapt tothe circumstances, the ethnographer can overcome many difficultiesand achieve an increasing degree of accuracy. He ends:

The task of the ethnographer must be to succeed in assembling a collectionof information and data whose value and precision place them beyond allcriticism. The facts must dictate the hypotheses and not submit to them. Itis on greater rigour in observation that the progress of sociology will depend,and it will only become a science of societies under this sole condition.(1925: 289–90)22

In other words and in the end, Métraux is claiming that the role of thecollector of information is a highly skilled and not a purely mechanicaloccupation that anyone can undertake. It is a bid for the properrecognition of the ethnographer’s status, one to which he already aspired.

The other article is entitled ‘Entretiens avec Alfred Métraux’, basedon three interviews with Fernande Bing in 1961 and publishedposthumously in L’Homme (Bing 1964). The second and thirdinterviews, dealing respectively with his work on Easter Island and onvoodoo, are of no immediate interest, but it is worth looking at the firstinterview in some detail, for it is entitled, ‘Comment et pourquoidevient-on ethnologue?’

In answer to this question, Métraux states that ethnographersbecome ethnographers because they are ill at ease in their own society,because they do not belong – something which certainly seems true of

Alfred Métraux 163

Métraux.23 He then looks back at the mid-1920s when he decided tobecome an ethnographer. It was period of ferment and of surrealismthat turned to exotic peoples to fulfil its aspirations, first in the fields ofprimitive art, but then, for some, to become channelled into scientificends. It was at this time that Métraux undertook his first field researchamong Amerindians, and he found that among these people, he felt,unlike in his own culture, entirely at ease. The rhythm of life seemedslower and less problematic. He then embarks on a piece of romanticismwhich is best left in his own words:24

I also believe that this taking up contact with primitive civilisations has mademe feel that, at root, the protest that has pushed me precisely towardscivilisations that are so much removed from our own finds its motivation ina sort of nostalgia, a nostalgia that we, men of the West, have, I believe, feltat all times and which I call, using a term that may be humorous – or atleast I intended it to be so – a nostalgia for the Neolithic. It seems to me,without wishing to fall into a facile Rousseauism, that humanity may havebeen wrong in going beyond the Neolithic. (Bing 1964: 21–22)25

He goes on to argue that the people among whom he spent time inSouth America do not differ in their style of life greatly from that of theNeolithic, and paints an extraordinarily rosy picture of such a life. Hewould, he states, be very happy living in the Neolithic age, if only therehad been dentists. He admits, however, that there is no returning to theNeolithic, and the last people who are at that stage are rapidlydisappearing. What has inspired his ethnographic career is the need torecord, in as much detail as possible, this passing way of life, for hebelieves that they have been able to resolve, better than we, certainproblems that confront humanity.26

In practice, however, this Neolithic utopia seemed to have escapedhim, and often he does not seem to have enjoyed his fieldwork. Forexample, he wrote to Georges-Henri Rivière from Easter Island on 4December 1934, complaining that life on the island is sad andmonotonous. His stay on the island, which he describes as a ‘devil ofan island’, has undermined his resistance, and he has rheumatism anda serious stomach complaint. He continues:

All this for the modest glory of having compiled a dictionary, made agrammar and created a corpus of the traditions of these mongrelizedPolynesians. I have a horror of the inhabitants of this island: it is difficult toimagine a population more vilely degenerated. South American contact hasbeen enough to introduce the foulest vulgarity. (Laroche 1992: 64)27

164 Peter Rivière

In February 1954, he wrote to Pierre Verger:

Through an act of masochism and pure stupidity, I have spent a monthwandering around the Aymara country, in the rain, in the fog and withweather that was all the colder given that I was equipped for the tropics. Ihave a horror of the landscapes of the high plateaux, I can’t bear altitude,and the Aymara Indians inspire physically in me a violent aversion. In fact,they give me nausea. (Verger 1992: 182)28

Although he was looking forward to working among Tropical Forestpeople, his experience with the Kayapó in Central Brazil, a little laterthat year, was no better. He was, like most others, a victim to the endlessdelays that accompanied travel in the interior of Brazil at that time. Theaeroplane that he expects on 4 April to take him from the Kayapó villageof Kubenkankrey does not arrive and when by 6 May, desperate to leave,there is still no sign of it, he writes:

I would feel much better … if I did not have at the bottom of my heart the fearof not being able to leave this place next Sunday. The prospect of anotherfortnight waiting here terrifies me. I won’t be able to stand it.29

The aeroplane finally arrives on 24 May. By that time, presumably, in astate of frustration and anxiety, he has even given up writing hisjournal. Later, however, no mention is made of his problems andworries, and the single difficulty to which he refers is his lack of aninterpreter (d’Ans 1992: 11–15).

Finally, right at the end of his career, in December 1959, in the littleAndean village of Carhuauz, he confides these words to his journal:‘The relentless silence of the main square under the sun, and the vagueterror of being condemned to live here, even as an ethnographer’ (d’Ans1992: 24).30

There is something enigmatic about Alfred Métraux. He was a manopenly dedicated to fieldwork, which, in practice, he often founddisagreeable. He devoted himself tirelessly to the accumulation of facts,whether from the field or from the writings of others, and to theproduction of descriptive accounts, but, at the same time, he seems tohave had a principled unwillingness to submit the material to furtheranalysis.

A survey of South American ethnographies from the past threedecades indicates that rarely do his anthropological colleagues refer tohis work. On the other hand, Rhoda Métraux, Alfred’s second wife and ananthropologist in her own right, noted in her article on him inInternational Dictionary of Anthropologists (1991) that, while his work hadbeen neglected for some years, ‘now it is recognized as an invaluable base

Alfred Métraux 165

on which to build new knowledge and new theory’. This may be so, andthere are signs of a recent revival of interest in Métraux. In Geneva hasemerged the Société d’Études Alfred Métraux, and in 2005, at asymposium in Paris to commemorate 60 years of UNESCO, a paper (seePrins 2005) celebrated his many years work as an applied anthropologist.

Notes

1. This was the Revista del Instituto de Etnología de la Universidad Nacional de Tucumán(hereafter Revista).

2. ‘Mon Institut et ma Revue sont les seuls organismes vivants de cette Université et monactivité est considérée comme un bon élément de propagande.’

3. ‘Ma situation est loin d’être aussi brillante que l’année dernière en raison de l’effroyabledébâcle économique de l’Argentine.’

4. ‘J’ai quitté l’Argentine où mon activité à la suite de l’idiosyncrasie du pays risquait d’êtrecomplètement paralysée. J’ai échappé à la stagnation la plus lamentable.’

5. For an account of Métraux’s work with UNESCO, see Prins (2005).6. Nordenskiöld also rejected the idea of recent Asian or Oceanic influence in

Amerindian culture to which Rivet subscribed. Interestingly enough, when Métrauxpublished an article on tapirage (the treatment of birds, such as parrots, in order tochange the colour of their plumage) in South America (1928c), he not only used aNordenskiöld-type distribution map of the practice, but explicitly stated that he couldnot relate it to the Malayo-Polynesian culture area.

7. In April 1963, the month of his self-inflicted death and a few months after he hadturned sixty, he published an article entitled ‘Does life end at sixty?’ Curiously, in noneof the obituaries or commentaries on Métraux’s life that I have read is there anyreference to what proved to be the rather ominous title of this article. The articlemainly consists of examples of how well old people are treated and respected in simplesocieties, but ends on the slightly sour note that in modern societies we haveexchanged these for increased longevity. Although the reason for Métraux’s suicideis outside the scope of this paper, Lévi-Strauss made a remark that suggests that hewas suffering from a loss of what today we call self-esteem: ‘And what increases ourdesolation even more is the thought that he might not have overrated death if he hadnot so unfairly underrated his work, and that he left us under this doublemisunderstanding’ (‘Et ce qui aggrave encore notre désolation, c’est de penser qu’il n’auraitpeut-être pas surestimé la mort s’il n’avait injustement sous-estimé son oeuvre, et qu’il nousa quittés sur ce double malentendu’; 1964: 8).

8. There is a comprehensive, but not exhaustive, list of Alfred Métraux’s publications inL’Homme, 4 (see Tardits 1964).

9. Métraux himself wanted to do more. He was disappointed that he was not to writethe articles on the Altiplano, and he made a strong claim to write those on myth andreligion, especially the former (Murra 1992: 78).

10. ‘la voie contemporaine des recherches sur les basses terres.’11. Louis Faron’s Hawks of the sun (1964), which is a fieldwork-based study of Mapuche

shamanism, refers only twice to Métraux’s work on Araucanian shamanism, bothtimes critically. Whereas Métraux refers to the work of Audrey Butt Colson in hisupdating of his 1944 article (1944b) on tropical South American shamanism in hisposthumous collection, the latter makes reference only to his HSAI contributions inany of her works on the topic.

166 Peter Rivière

12. For example, ‘Contribution à l’ethnographie et à la linguistique des Indiens Urod’Ancoaquil (Bolivie)’ (1935a), of 35 pages, is based on just one week’s stay in the Urovillage of Ancoaquil, on the Bolivian Altiplano.

13. Given its date and place of publication, it is not too surprising that this edition isdifficult to find. I am grateful to Jean-Pierre Chaumeil, who has a copy in his extensivecollection of South Americana and provided me with a detailed account of how itdiffers from the 1951 edition.

14. A decade later, in the course of an interview, Métraux was much more scathing aboutHeyerdahl’s theory: ‘This is a perfectly unsustainable theory, which no man of science hasaccepted, I would even say so absurd that no scholar has dreamt of examining it seriously’(‘C’est une théorie parfaitement insoutenable qu’aucun homme de science n’a acceptée, je diraimême si absurde qu’aucun savant n’a pu songer à l’examiner sérieusement’; see Bing 1964: 26).

15. An English version, Haiti: Black peasants and their religion, appeared in 1960.16. In 1961 he stated that it was only in 1948, when he became involved with the

UNESCO project in the Marbial Valley, that he began his study of voodoo (Bing 1964:28). It is not clear why he should have claimed this, since he declares elsewhere thathe became interested in the subject during his visits in 1941 and 1944, and hepublished his first article on voodoo in 1946 (Métraux 1959: 17).

17. This work still commands a sizeable readership and apparently remains popular inHaiti. I am grateful to Harald Prins for pointing this out to me, as well as for makinga number of other invaluable comments on an earlier draft.

18. This against a background when there is likely to have been much discussion inParisian anthropological circles about mythology, as Lévi-Strauss was then workingon the first volume of Mythologiques, which appeared in 1964.

19. ‘Métraux apparaît dans son oeuvre comme quelqu’un qui se souciait avant tout deconnaissance concrète et pour qui l’étude des sociétés était, plutôt qu’une voie débouchant surles aperçus théoriques, un moyen de connaître les hommes et de les approcher du plus près,dans toute la diversité de leurs us et coutumes.’

20. To whom, at least in those of his publications which are listed in the bibliography andcomprise all his main ones, he never makes another reference. Perhaps it might beadded that the Revue d’ethnographie, in which this piece appeared, was edited by,amongst others, Lévy-Bruhl.

21. ‘S’il y avait, comme on l’a cru et comme on continue à le croire, identité entre la mentalitéprimitive et la mentalité civilisée, la réponse à cette question n’aurait pu être que l’affirmative.Cette universalité de la raison humaine a été implicitement admise par l’école sociologiqueanglaise qui est responsable d’illusion partagée par beaucoup de savants sur la valeur quasiabsolue du résultat de ces investigations. La critique présenté par Lévy-Brühl dans ses deuxouvrages, les Fonctions mentales dans les sociétés inférieures et la Mentalité primitive, a jetésur ce postulat une suspicion légitime. L’analyse des représentations des primitifs a révélé lesdifférences foncières qu’elles présentent avec celles des civilisés. L’humanité ‘sauvage’ apparaîtcomme un monde clos, imperméable à notre expérience, qui reste incompréhensible si l’onattribue à ceux qui le composent des processus mentaux sembables aux nôtres. Cemalentendu, signalé avec évidence depuis quatorze ans seulement, est cause de l’imperfectiondes témoignages obtenus des indigènes et affecte le résultat des enquêtes ethnographiques àvenir aussi bien que de celles qui nous sont déjà données.’

22. ‘La tâche de l’ethnographe doit être de parvenir à rassembler un ensemble de renseignementset de données dont la valeur et l’exactitude seront au-dessus de toute critique. Les faits doiventimposer les hypothèses et ne pas s’y plier. C’est d’une plus grande rigueur dans l’observationque dépendront les progrès de la sociologie, qui ne deviendra la science des sociétés qu’à cetteseule condition.’

23. See Lévi-Strauss (1976 [1955]) and Leach (1984) for similar arguments.

Alfred Métraux 167

24. It should be noted that such romanticism was not that uncommon. One of the mostfamous examples of it is to be found in Lévi-Strauss’s Tristes tropiques (1976 [1955]).

25. ‘Je crois aussi que cette prise de contact avec les civilisations primitives m’a fait sentir qu’au fond,la protestation qui m’avait précisément poussé vers des civilisations tellement éloignées de lanôtre, trouvait son motif dans une sorte de nostalgie, une nostalgie que nous, hommesd’Occident, avons, je crois, ressentie de tout temps et que j’appelle d’un terme peut-être comique,enfin que je veux tel, la nostalgie du néolithique. Il me semble, et cela sans vouloir tomber dansun rousseauisme facile, que l’humanité a peut-être eu tort d’aller au-delà du néolithique.’

26. It comes across more or less explicitly in many of Métraux’s field studies that he seeshimself engaged in salvage ethnography, that what he is observing is the debris of apast civilisation and that he is recording what will soon have gone forever.

27. ‘Tout ceci pour la modeste gloire d’avoir réuni un dictionnaire, fait une grammaire et constituéen corpus les traditions de ces Polynésiens abatardis. J’ai les habitants de cette île en horreur:on peut difficilement imaginer population plus vilement dégénérée. Il a suffi du contactsudaméricain pour y introduire la plus crapuleuse vulgarité.’

28. ‘Par un acte de masochisme et de pure imbécillité, j’ai parcouru le pays aymara pendant unmois, sous la pluie, dans la brume et par des froids d’autant plus acerbes que j’étais équipétropicalement. J’ai horreur des paysages de haut-plateau, je supporte mal l’altitude et lesIndiens aymara m’inspirent une aversion physique violente. En fait, ils me donnent lanausée.’

29. ‘Je serais beaucoup mieux…si je ne gardais au fond du coeur la crainte de ne pouvoir quittercet endroit dimanche prochain. La perspective de quinze nouveaux jours d’attentem’épouvante. Je ne pourrai pas les supporter.’

30. ‘implacable silence de la grande plaza sous le soleil, et vague terreur d’être condamné à yvivre, même en tant qu’ethnographe.’

References

Auroi, C. and A. Monnier. 1996. Du pays de Vaud au pays du Vaudou: ethnologies d’ AlfredMétraux, Geneva: Musée d’ Ethnographie.

Bing, F. 1964. Entretiens avec Alfred Métraux, L’Homme, 4: 20–31.Cooper, J. 1942. Areal and temporal aspects of aboriginal South American culture,

Primitive Man, 15: 1–38.d’Ans, A.-M. 1992. Le contenu d’itinéraires 2, 1953–1961, Présence de Alfred Métraux,

Cahiers Georges Bataille (Special Issue): 5–28.Davis, W. 1986. The serpent and the rainbow, London: Collins.Faron, L. 1964. Hawks of the sun: Mapuche morality and its ritual attributes, Pittsburgh:

University of Pittsburgh Press.Fernandes, F. 1963. Organização social dos Tupinamba, São Paulo: Difusão Européia do Livro.Grenand, P. 1982. Ainsi parlaient nos ancêtres: essai d’ éthnohistoire ‘Wa api’, Paris:

ORSTOM.Laroche, M.-C. 1992. Alfred Métraux à l’île de Pâques, Présence de Alfred Métraux, Cahiers

Georges Bataille (Special Issue): 47–65.Le Bouler, J.-P. 1992. Alfred Métraux et Georges Bataille en 1922: de l’École des Chartes

à l’Amérique du Sud, Présence de Alfred Métraux, Cahiers Georges Bataille (SpecialIssue): 129–39.

Leach, E. 1984. Glimpses of the unmentionable in the history of British socialanthropology, Annual Review of Anthropology, 13: 1–23.

Leiris, M. 1958. Preface, in A. Métraux, Le Vaudou haïtien, Paris: Gallimard.Lévi-Strauss, C. 1964. Hommage à Alfred Métraux, L’Homme, 4: 5–8.

168 Peter Rivière

——— 1976 [1955]. Tristes tropiques, London: Penguin. Métraux, A. 1925. De la méthode dans les recherches ethnographiques, Revue d’

ethnographie et des traditions populaires, 6: 266–90.——— 1927a. Le bâton de rythme: contributions à l’étude de la distribution

géographique des éléments de culture d’origine mélanésienne en Amérique du Sud,Journal de la Société des Américanistes de Paris, 19 (n.s.): 117–22.

——— 1927b. Migrations historiques des Tupí-Guaraní, Journal de la Société desAméricanistes de Paris, 19 (n.s.): 1–45.

——— 1928a. La Civilisation matérielle des tribus Tupi-Guarani, Paris: Librairie Geuthner.——— 1928b. La Religion des Tupinamba et ses rapports avec celle des autres tribus tupi-

guarani, Paris: Librairie Ernest Leroux.——— 1928c. Une découverte biologique des Indiens de l’Amérique du Sud: la

décoloration artificielle des plumes sur les oiseaux vivants, Journal de la Société desAméricanistes de Paris, 20 (n.s.): 181–92.

——— 1929a. Contribution à l’ethnographie et à l’archéologie de la province deMendoza (R.A.), Revista del Instituto de Etnología de la Universidad Nacional de Tucumán,1: 5–73.

——— 1929b. Deux anciens textes peu connus concernant l’institution du potlatch enFloride et dans le Gran Chaco, Journal de la Société des Américanistes de Paris, 21 (n.s.): 417.

——— 1930. Études sur la civilisation des Indiens Chiriguano, Revista del Instituto deEtnología de la Universidad Nacional de Tucumán, 1: 295–493.

——— 1931. Les hommes-dieux chez les Chiriguano et dans l’Amérique du Sud, Revistadel Instituto de Etnología de la Universidad Nacional de Tucumán, 2: 61–91.

——— 1932. Mitos y cuentos de los Indios Chiriguano, Revista del Museo de La Plata, 33:119–84.

——— 1934. Contribution au folk-lore andin, Journal de la Société des Américanistes deParis, 26 (n.s): 67–102.

——— 1935a. Civilización material de los Indios Uro-Chipaya de Carangas (Bolivia),Revista del Instituto de Etnología de la Universidad Nacional de Tucumán, 3: 85–129.

——— 1935b. Contribution à l’ethnographie et à la linguistique des Indiens Urod’Ancoaquil (Bolivie), Journal de la Société des Américanistes de Paris, 27 (n.s.): 75–110.

——— 1935c. La religión secreta y la mitologia de los Indios Uro-Chipaya de Carangas(Bolivia), Revista del Instituto de Etnología de la Universidad Nacional de Tucumán, 3: 7–84.

——— 1935d. Les Indiens Uro- ipaya de Carangas, Journal de la Société des Américanistesde Paris, 27 (n.s.): 111–28.

——— 1936a. La structure sociale, Encyclopédie française 7: Section A, Chapter 2, Paris:Comité de l’Encyclopedie Française Éditeurs.

——— 1936b. Numerals of Easter Island, Man, 36: 190–91.——— 1937a. Easter Island sanctuaries (analytic and comparative study), Ethnological

Studies, 5: 104–53.——— 1937b. Études d’ethnographie Toba-Pilaga (Gran Chaco), Anthropos, 32: 171–

94, 378–401.——— 1938. The Proto-Indian script and the Easter Island tablets (a critical study),

Anthropos, 33: 219–39.——— 1939. Myths and tales of the Matako Indians (the Gran Chaco, Argentina),

Ethnological Studies, 9: 1–127.——— 1940. Ethnology of Easter Island, Honolulu: Bernice P. Bishop Museum.——— 1941. L’île de Pâques, Paris: Gallimard.——— 1942. Le shamanisme araucan, Revista del Instituto de Antropología de la

Universidade Nacional de Tucumán, 2: 309–62.——— 1944a. La causa y al tratamiento mágico de las enfermedades entre los Indios de

la región Tropical Sud-Americana, America Indigena, 4: 157–64.

Alfred Métraux 169

——— 1944b. Le shamanisme chez les Indiens de Amérique du sud tropicale, ActaAmericana, 2: 197–219, 320–41.

——— 1945. Le shamanisme chez les Indiens du Gran Chaco, Sociologia, 7: 157–68.——— 1946a. Myths of the Toba and Pilaga Indians of the Gran Chaco, Philadelphia:

American Folklore Society.——— 1946b. La civilisation Guyano-Amazonienne et ses provinces culturelles, Acta

Americana, 4: 130–53.——— 1946c. The concept of soul in Haitian vodou, Southwestern Journal of

Anthropology, 2: 84–92.——— 1946d. El dios supremo, los creadores y héroes culturales en la mitología

Sudamericana, America Indigena, 6: 9–25.——— 1947. Mourning rites and burial forms of the South American Indians, America

Indigena, 7: 7–44.——— 1948a. Ensayos de mitología comparada Sudamericana, America Indigena, 8: 9–30.——— 1948b. Étude sur l’agriculture paysanne dans une vallée haïtienne, Acta

Americana, 6: 173–91.——— 1950. The contribution of the Rev. Father Cooper to South American

ethnography, Primitive Man, 50: 39–48.——— 1951a. Droit et coutume en matière successorale dans la paysannerie haïtienne,

Zaïre, 5: 339–49.——— 1951b. L’île de Pâques (Édition revue et augmentée), Paris: Gallimard.——— 1952. Les croyances animistes dans le vodou haïtien, Mémoires de l’Institut

Français d’Afrique Noire, 27: 239–44.——— 1956. Le mariage mystique dans le vodou, Cahiers du Sud, 43: 410–19.——— 1957a. Easter Island: A stone-age civilization of the Pacific, London: André Deutsch.——— 1957b. Haïti: la terre, les hommes et les dieux, Neuchâtel: À la Baconnière.——— 1958. Le Vaudou haïtien, Paris: Gallimard.——— 1959. Voodo in Haiti, London: André Deutsch.——— 1960a. Haiti: Black peasants and their religion, London: George G. Harrap.——— 1960b. Mythes et contes des Indiens Cayapo (Groupe Kuben-kran-kegn), Revista

do Museu Paulista, 12 (n.s.): 7–35.——— 1962. Les Incas, Paris: Éditions du Seuil.——— 1963. Does life end at sixty? The UNESCO Courier (April): 20–23.——— 1967. Religions et magies indiennes d’Amérique du Sud, Paris: Gallimard.Métraux, R. 1991. Métraux, Alfred, International Dictionary of Anthropologists, New York

and London: Garland Publishing.Murra, J. 1992. Correspondance entre Julian H. Steward et A.M. à propos du Handbook

of South American Indians, Présence de Alfred Métraux, Cahiers Georges Bataille (SpecialIssue): 75–78.

Nordenskiöld, E. 1919–1938. Comparative ethnological studies (vols 1–10), Göteborg:Elanders Boktryekeri Aktiebolay.

Prins, H.E.L. (with E. Krebs) 2005. Vers un monde sans mal: Alfred Métraux, unanthropologue à l’UNESCO, 60 Ans d’Histoire de l’Unesco, Paris: UNESCO.

Renshaw, J. 2003. The Indians of the Paraguayan Chaco: identity and economy, Lincoln andLondon: University of Nebraska Press.

Rivière, P. 1967. Chaco and Mapuche, The Times Literary Supplement, 13 July 1967.Steward, J. (ed.). 1946. Handbook of South American Indians, Smithsonian Institution,

Bureau of American Ethnology, Bulletin 143, vol. 1.Tardits, C. 1964. Bibliographie, L’Homme, 4(2): 49–62.Verger, P. 1992. Trente ans d’amitié avec Alfred Métraux: mon presque jumeau, Présence

de Alfred Métraux, Cahiers Georges Bataille (Special Issue): 173–91.Wagley, C. 1964. Alfred Métraux 1902–1963, American Anthropologist, 66: 603–13.

Chapter 7

ROGER BASTIDE OR THE‘DARKNESSES OF ALTERITY’

Stefania Capone

Introduction

Within the French anthropology of the 1950s, Roger Bastide played apreeminent role in the foundation of what was an entirely new domainof studies at the time, namely ‘Afro-Americanism’, defined in France byits notion of ‘Black Americas’.1 If his heritage continues in France,thanks especially to the journal Bastidiana, it was in Brazil that Bastideacquired his reputation, profoundly marked by the development ofstudies on the religions and cultures of African origin. His slightlymarginal position within the French academic world2 seems to havederived from his critical position with regard to the theories that weredominant at this period. As we shall see, his critical distance from thesociology of Durkheim and Weber, as well as from Lévi-Strauss’sstructuralism,3 was motivated by a profound questioning of the role ofthe researcher and his involvement with the field.

If fieldwork constitutes the privileged moment of the anthropologicalenterprise, in which alterity is fully at work, the reflexivity of theBastidian intellectual project was not limited to encounters with theOther but became a constant aspect of his entire work. In Bastide’s case,theory cannot be the simple product of deductive reasoning or Cartesianrationality, as was the case for many of his colleagues, but involves adifferent rationality, one that seeks its chosen field in art and mysticalstates. Bastide never ceased to reflect on his sociological work, which wasprofoundly marked by the discipline of anthropology.

Figure 7.1. Roger Bastide, seated left on bench, Ifanhin, Bénin, 1958. Fundação PierreVerger, Salvador de Bahia, Brazil.

172 Stefania Capone

However, the complexity of Bastide’s intellectual trajectory cannot beunderstood without taking into account another figure who plays therole of a go-between linking the worlds of the candomblé and academia,namely Pierre Verger. Bastide’s friend and ‘accomplice’ for more thanthirty years, Verger incarnated that ‘ethnographic essentialism’ that hasbeen so much decried by the critics of a certain type of Frenchanthropology. For Verger, theory is only a secondary aspect, of littleimportance and even carefully avoided in a personal project which wasvery remote from the scientific standards of the time. Bastide needed his‘alter ego’ to reflect on his own practice and to theorise a ‘sociology indepth’. Bastide’s interest in mysticism, which marked his intellectualtrajectory from the start, led him to think that the only way to understandthe world of candomblé was for the anthropologist himself to go throughan initiation.4 The first period of his intellectual biography, in which artand mysticism combine, thus proved crucial for the future developmentof his method, in which poetry would play a fundamental role.

Roger Bastide or the ‘darknesses of alterity’ 173

The relationship between Bastide and Verger can be thought of as aperfect example of the anthropological division of labour betweenethnographic description and theoretical analysis. The trajectory involvingBastide’s break with the dominant conceptualisations and his position as anoutsider in the intellectual milieu of his period made him an exemplarycase of another way of creating a sociology of religion, taking fully intoaccount the ‘encounter with the other’, its chimeras and its dangers.

Bastide and French sociology in the inter-war period

Roger Bastide was born at Nîmes in the south of France on 1 April 1898.The son of primary school teachers, he was educated in the Protestantreligion and spent his infancy in Anduze in the Cévennes, which hadbeen ‘a refuge of Protestant rebels’ at the start of the eighteenth century.As Maria Isaura Pereira de Queiróz, a former student of Bastide’s, hasstated (1983: 7): ‘Right from his infancy, he was plunged into socio-cultural contradictions which, throughout his life, constituted thebackground to his researches, the basic food for his observations andreflections .... Although feeling himself to be profoundly French, he wasa member of a minority religious group which had known and kept themarks of its oppression and persecution.’

This specificity of Bastide, his ‘marginality’ with respect to the worldthat surrounded him, would be one of the marks of his scientific andbiographical trajectories. The links he maintained throughout his lifewith other figures in the French academic world who were alsoProtestant by confession, such as Raoul Allier, Gaston Richard, PaulArbousse-Bastide and Maurice Halbwachs, bears witness to thepregnant nature of a religious identity that was never completelyabandoned and that marks his trajectory in different ways.

Trained as a philosopher, it was under the direction of GastonRichard (1860–1945)5 that Bastide discovered sociology by followingRichard’s courses at the University of Bordeaux, where he obtained alicence (first-degree) bursary at the end of the First World War. Richard’sinfluence (1923), and in particular his anti-Durkheimianism, wouldplay a continual role in his work overall. Trained in an academicuniverse that was strongly marked by the Durkheimian school, Bastidenever ceased to rethink the relationship to the sacred and the religious,questioning Durkheim’s theories while taking into account certain ofthe master’s formulations.

Nonetheless, in Les problèmes de la vie mystique, published in 1931(Bastide 1931b), Bastide adopted a similar procedure to Durkheim’s inhis study of religion, starting out from elementary forms of mystical

174 Stefania Capone

life, that is of primitive mysticism, in order to construct a ‘mysticalchain’ to more elevated forms. As Fernanda A. Peixoto recalls (2000:28), Bastide had already addressed some severe criticisms ofDurkheim’s Les formes élémentaires de la vie religieuse (1912) in an articleof 1928 entitled ‘Mysticisme et sociologie’. In it, he questionsDurkheim’s ‘ethnocentrism’, especially for the importance he gives tothe exaltation of religious ceremonies in his work, which Bastide findsexcessive. For Bastide, this was the outcome of a ‘western gaze’.

Durkheim (1895) defined the ‘social fact’ as an entity sui generis, atotality which contains within itself its own explanation and whichcannot be reduced to the sum of its parts. To explain it, it was necessaryto use the tools of the nascent sociology and to reject contributions fromother disciplines, like psychology. In dissociating the individual from thecollective, Durkheim saw in social norms and institutions the meanswhereby society exercises a constraining action on individuals.

In his earliest writings, Bastide developed a whole series of criticismsof Durkheimian sociology, addressing himself to what he called the‘collectivist emphasis’ and the effacing of the ‘individual fact’. But itwas above all Durkheim’s attitude to belief which led Bastide to distancehimself from Durkheim’s teaching and join the ranks of his critics.Thus, quoting Richard in his introduction to Eléments de sociologiereligieuse (1935: 8), Bastide emphasises the importance of experience insociological reflection: ‘the first condition to be satisfied by anyonewishing to understand the believer and the society of believers is to haveparticipated oneself in a belief at some moment in one’s life, at leastthrough emotion or sentiment’. In taking up Richard’s ideas, whichwould play a very important part in his preparations for his agrégationin philosophy, Bastide saw in the refusal of French positivism to takethe individual into account one of the main limitations of theDurkheimian approach. In reducing the religious to the social,Durkheim had therefore lost the very essence of the phenomenon.

By contrast, Bastide underlines the importance of individual factorsin understanding religious phenomena while still retaining fromDurkheimian theories the idea that the collective deeply penetrates thereligious and the distinction between sacred and profane. Theimportance accorded, from his very first writings, to the individualdimension of religious practices evokes the theories of Durkheim’s greatrival at this period, Gabriel Tarde (1843–1904), the founder of Frenchsocial psychology. The latter had developed a sociological theory basedon criteria that were essentially individual, showing that the social isnothing other than the expression of individual forces, especiallypsychological ones. For Tarde, the domain of sociology boils down tointer-individual or ‘inter-mental’ communication. He thus opposes to


Durkheim’s ‘collective self ’ a multiplicity of individuals, irreducible tothe collective consciousness.6

As Peixoto rightly stresses (2000: 35), where Durkheim affirms thelegitimacy of science in opposition to literary style, history andpsychology, Tarde ‘combines literary leanings and the image of the poetwith scientific work’. This intellectual posture was much closer to thetemperaments of Richard and Bastide, who cultivated a taste forliterature and the arts, philosophy and psychology. For them, sociologyis not enough on its own, as Durkheim affirmed, but has to be completedby other disciplines.

For a rapprochement between sociology and social psychology, one hasto wait for Marcel Mauss (1872–1950), Durkheim’s nephew and pupil:

For our part, we observe the complete and complex reactions of numericallydefined masses of men, complete, complex beings. We, too, observe whatconstitutes their organism or their psyche. At the same time we describe thebehaviour of this mass and its corresponding psychoses: sentiments, ideas,and the volitions of the crowd, or of organized societies and their subgroups.(Mauss 1990 [1922]: 103)

The notion of the ‘social fact’, introduced by Mauss, offers a holisticapproach which encompasses the totality of human manifestations.Mauss’s interest in ‘sentiment’ found echoes in Bastide’s writings,notably in Eléments de sociologie religieuse of 1935. But the attempt todraw closer to Mauss, whom Maurice Halbwachs had advised him tomeet, was not a very happy one. Thus, in a letter dated 3 November1936, Mauss, acknowledging Bastide having sent him this work, writes:

Your introduction strikes me as far too philosophical. I add that one of thoseyou particularly appreciate, Max Weber, is among those with whomDurkheim, Hubert and I communicated the least. […] As for the rest, I alsobelieve that you have exaggerated appreciably the metaphysical basis ofDurkheim’s thought. He was not only one of the founders of science, butalmost exclusively a founder who used philosophy less and less. And toregard this philosophy as a metaphysics is equally wrong in my view.7

In the same work, Bastide also moved nearer to the interpretations ofLévy-Bruhl (1857–1939), for whom mysticism did not havepathological causes but constituted a collective template of perceptionexpressing a different logic, a different type of relationship to the world,than our own. Being inspired by the notion of the collectiverepresentation of Durkheim, whose collaborator he had been, Lévy-Bruhl affirmed that, if representations are the product of the social, itbecomes possible to explain mystical representations with reference to

176 Stefania Capone

the principles that are proper to them and which are linked to a well-defined social milieu. However, to understand a ‘pre-logical’ mentality,it was also necessary to include supra-sensitive elements like the spiritsof the dead, which form part of ‘primitive reality’.

Bastide adopted some of Lévy-Bruhl’s formulations (1922),especially in the elaboration of his key concept, the ‘principle ofcompartmentalization’.8 The influence exerted by reading the firstworks of Maurice Leenhardt (1878–1954), who had practisedparticipant observation well before Malinowski, thus inaugurating theanthropological fieldwork advocated by Mauss, helped Bastide torethink Lévy-Bruhl’s theories and to espouse an ‘obscure and confusedthought’ which seemed to him better able to express mystical states.9

This calling into question of the dominant theories of his period putBastide at odds with the French intellectual milieu, in which theory didnot necessarily arise from field data. Durkheim had advocated theexteriority of the researcher in relation to the object of his research,this being the sole guarantee of objectivity. Mauss had reintroducedsubjectivity in underlining the unconscious dimension of socialphenomena, while recognising, like his uncle, the subordination ofpsychological to social phenomena.

This attitude regarding the role of the researcher and his immersionin the field, as well as the epistemological status of mysticism and of thereligious life in general, led Bastide to position himself critically withregard to Durkheimian sociology.10 Thus, in the introduction to his workof 1960 (see 1978: 4), Bastide writes:

Durkheim always seems to hesitate between religion as ‘product’ andreligion as ‘expression’. The two themes are always closely intertwined in hiswork, and it is difficult to separate them. […] Durkheim’s conclusion goesbeyond the multitude of examples he compiles in support of his thesis,inasmuch as they all merely show that religion is always incarnate in thesocial structure, but not that social structure creates religion.11

Bastide, who in this period was working as a teacher agrégé de philosophiein the lycées and had been appointed to Valence in 1928, was alreadyattracted by doing fieldwork. In 1930 he became involved in his firstsociological enquiry on groups of immigrants in France, which wouldlead to ‘Les Armeniens en Valence’, a study that appeared in the RevueInternationale de Sociologie in 1931 (Bastide 1931a). It was in Valence,where he lived from 1928 to 1937, that Bastide wrote his first texts onreligious sociology and mysticism, a theme that accompanied him rightup until his death in 1974. Although a critic of Durkheimian theories,he made contact with Maurice Halbwachs (1877–1945), whom he metat Strasbourg and with whom he discussed his works on mysticism and


religion (Peixoto 2000: 39). His influence would become apparent inthe Bastidian analysis of ‘collective memory’ (cf. Capone 2007).

The years in Brazil, or the discovery of the Other

In 1937 Bastide was appointed to Versailles, but he only stayed therefor a year, since Georges Dumas offered him a post as professor at theUniversity of São Paulo. The author of Le traité de psychologie (Dumas1923–24), ‘at once a doctor and a philosopher’ (Carelli 1993: 193),had originally been sent to Rio de Janeiro in 1908 as a spokesman forthe Groupement des Universités et Grandes Écoles de France in order toset up scientific cooperation with Brazil. His role was to be central in thefounding of the French lycée in Rio in 1916 and the sending ofuniversity missions to São Paulo between 1935 and 1939 (ibid.).Supported by a fund for French university and scientific expansionoverseas, created in 1912, and a service promoting French worksoverseas, Dumas also founded Instituts franco-brésiliens de HauteCulture in Rio in 1922 and in São Paulo in 1925, which helpedestablish French higher education in Brazil (Lefebvre 1990).

Paul Arbousse-Bastide, a former student of Dumas and long-termfriend of Bastide, who was installed in Brazil from 1934, wrote toBastide as follows in 1937: ‘I am reading you from afar, and our subjectsof study are very close. I would like to have you for a colleague. Youknow that this is not impossible. A chair of sociology may becomevacant from one day to the next, and sociologists are rare in theUniversity … Think about it in principle. It would be for March 1938’(quoted in Morin 1994: 36). Thanks to Dumas, the University of SãoPaulo (USP), which had only just been founded in 1934, had alreadybrought together an important number of French intellectuals, amongwhom were Fernand Braudel (1935–38) in history, Pierre Monbeig(1935–46) in geography and Claude Lévi-Strauss (1935–38), whobenefited from the splitting of Arbousse-Bastide’s chair of sociology.12

In 1938, Dumas invited Bastide to come to Brazil to take up the postof professor of sociology at the University of São Paulo, which had beenvacated by Lévi-Strauss. The contract Bastide signed with the Brazilianuniversity envisaged him teaching Durkheimian sociology and – inreaction to Lévi-Strauss’s resignation, presented in 1937 so that hecould conduct research among the Nambikwara and Bororo – it alsoobliged him to restrict his research to the university vacations. Thisarrangement with his employers is the origin of his brief fieldexperience. In Le candomblé de Bahia (1958: 14), Bastide admits tohaving passed no more than nine months in the field, including five at

178 Stefania Capone

Salvador de Bahia, spread out over seven consecutive years (1944 to1951).13 His work therefore proceeded more by following contacts with‘local experts’ than through prolonged observation of the rituals he wasstudying: ‘as a result, I might get to know a candomblé during threemonths of vacation. The ceremonies that took place in months withoutvacations I could not get to know’ (quoted in Cardoso 1994: 72).

At the start of his Brazilian sojourn, Bastide therefore devotedhimself to his teaching and to intensive activities as a critic in severalBrazilian journals. He attended the lectures of Brazilian sociologists andfrequented Paulist intellectual circles. But it was especially the affinitiesbetween Bastide and the modernist group in São Paulo, notably withMario de Andrade, that marked his first steps in Brazil.

The Modern Art Week, put on in São Paulo in 1922 upon thecentenary of the country’s independence, offered a group of youngintellectuals an opportunity to assert themselves on the national scene.Bastide quickly inserted himself into this group, to which belonged thepoet Mario de Andrade, the painter Lazar Segall, the archaeologistPaulo Duarte, the writer Oswald de Andrade and the art critic SergioMillet. It was in his debates with the modernists that the Frenchsociologist refined his view of the Other in his search for the ‘Braziliansoul’. This intense dialogue becomes clear in Bastide’s writings onBrazilian art, especially his reflections on the Baroque and the work ofAleijadinho (Peixoto 2000: 64). In his first writings can already befound his questioning of the authenticity and originality of Brazilianculture, with a particular interest in syncretism, without which it isimpossible to understand the reality of Brazil.

In these earliest Brazilian writings can be found themes dear toBastide: the links between art, mysticism and religion in the article‘Pintura e mística’ (‘Painting and mysticism’) of 1938, and thepsychoanalytical analysis applied to a sociological topic in ‘Psicanálisedo cafuné’ (‘A psychoanalysis of the cafuné’) of 1940. The link betweenart and mysticism is one of the constant themes in Bastide’s writings.For Bastide, mysticism and painting are intimately interwoven with oneanother, since mysticism is ‘the search for unity, the will to integration’,while painting is ‘the search for and conquest of unity’. Art and religionare thus for Bastide ‘two forms of ecstasy, two forms of contact with thesacred, as manifested in styles of life and artistic expressions’ (ibid.: 70).

But Bastide did not restrict himself to maintaining his numerouscontacts with Brazilian sociologists and intellectuals.14 He kept up hiscontacts with the French academic world, trying to help his colleagueswho were in danger from the imminent war. Thus in 1940 he receiveda letter from Georges Gurvitch, who was to play a central role in hisfuture return to France. Gurvitch wrote to him from Clermont-Ferrand,


where the University of Strasbourg had withdrawn, to ask Bastide toconsider an application from Gurvitch to teach in Brazil, ‘since the lawof 18 July 1940 on the dismissal of naturalized officials had led to hissuspension, before being permanently dismissed’ (Morin 1994: 38).Despite his efforts Bastide could do nothing, and in 1942 Gurvitchdeparted for New York. Having returned to France at the end of the war,he joined Bastide at the University of São Paulo, where he taught for ayear. In 1948 he wrote to Bastide again, regretting the fact that thelatter, ‘the best candidate and [his] preferred candidate’, had not agreedto replace him at the University of Strasbourg, ‘which would haveallowed him to be appointed Maître de Conférences and, when he hadfinished his thesis, titular professor’ (ibid.).

During his first years in Brazil, Bastide devoted himself to histeaching and to studying the Brazilian literature on his chosen topics.But it was also in Brazil that he discovered American sociology,especially studies of acculturation and culture contact, through the‘Memorandum for the Study of Acculturation’ by Robert Redfield,Ralph Linton and Melville Herskovits (Redfield et al. 1936).15 The latterhad defined the process of acculturation as ‘those phenomena whichresult when groups of individuals having different cultures come intocontinuous first-hand contact, with subsequent changes in the originalculture patterns of either or both groups’ (Herskovits 1938: 10). Thisprocess of social and cultural change also involved the disorganisationand reorganisation of individual personalities. Robert Park (1928) usedthe expression ‘marginal man’ to designate a member of a culturalgroup that has lost its characteristics in contact with another group,though without having been integrated into the dominant group. Hetherefore becomes ‘marginal’ since, as Arthur Ramos writes (1979[1937]: 244), he is on the margins of two cultures: his own, which hehas lost, and the other one, which he has not yet assimilated.

When Bastide arrived in Brazil, Arthur Ramos was already recognisedas the great specialist in Afro-Brazilian cultures and religions. Bastide methim in 1938, when he was occupying the chair of anthropology at thenew National Philosophy Faculty (FNFi), founded that year in Rio deJaneiro, and he rapidly became his friend (Teixeira and al. 1952). Ramosundoubtedly had a great influence on the evolution of Bastide’s thought,in particular on his interest in social psychology.16

Bastide had to wait until the start of 1944 to undertake his first Afro-Brazilian fieldwork, travelling to the Nordeste between 19 January and28 February. On this journey, which led him to Recife, João Pessoa andSalvador de Bahia, he met Jorge Amado, who was to become his guidewithin the world of the Bahian candomblés. His contacts with thoseinitiated into the candomblé were thus made through the intermediation

180 Stefania Capone

of a great Bahian writer, as well as through rather precarious linguisticexchanges, mixing, according to Amado, ‘French and Nagô’, since,despite having already spent six years in Brazil, Bastide had not yetmastered Portuguese (Lühning 2002: 10).17 In fact, Bastide was neverto master it. In the course of time, he fashioned a language of his own,‘a bizarre mixture of French, Latin and Provençal, in which, here andthere, Brazilian words surfaced’ (Queiróz 1994: 218). This does notseem to have prevented Bastide from grasping deeply the point of viewof the Other, sometimes appropriating it in order to reason, as his life-long friend Pierre Verger stresses, with his interlocutors’ arguments byseeing things ‘through their own eyes’ (Verger 1978: 52).

Poetry as a scientific method

This attempt to plunge as deeply as possible into the ‘Brazilian soul’ ledBastide to revise his approach to the field and his methodology. Duringthis initial journey, despite the very short period he spent in the field,with Amado’s help Bastide had the opportunity to attend severalceremonies in the Bahian candomblé. Thus he attended the‘confirmation of an ogan’, a ritual title bestowed on men who do not gointo trance, in the terreiro (cult house) of Mãe Cotinha. He was alsopresent at the ‘exit of an iaô’, the ceremony which marks the end of theperiod of initiation by publicly presenting a newly initiated person, in aterreiro of the Angola nation, that of the celebrated Joãozinho daGoméia. It was with this pai-de-santo (cult leader) that he was to havehis first experience of the ‘sacred’:

The inspired [sic] to be cannot be visited, not even by the ogans or the obaj [obá],in the traditional terreiros. João da Gavea [Goméia] nonetheless permitted meto violate this secret taboo, and I saw, spread out along the boards, eachwrapped from head to foot in a large white sheet, five or six bodies. They lookedlike larvae, or some white cockchafer grubs, and indeed, during this periodthey are truly human larvae. A new ‘self ’ is being born, a metamorphosis ofthe personality going on in the belly of the sanctuary, from which will veryshortly emerge the sons of the gods [the initiated], who will then openthemselves out in the shade of the hut, their wings still fragile, the wings ofbeings who henceforward will be saints [orixás]. (Bastide 1995 [1945]: 63)18

In this first work on the candomblé, Bastide seems to hesitate between ascientific approach and poetic writing, which he feels more suitable indescribing this new universe that is opening before his eyes. Thus, inhis introduction, he writes:


This is just a bundle of images. It is not purely a scientific book, let alone atype of lyrical chant. […] Perhaps the main fault with this work actuallyresides in its hesitation between science and poetry. However, this hesitationexactly translates the spiritual state in which I found myself at that moment,torn between a very great degree of fervour and the desire to conductobjective research. (ibid.: 11–12)19

However, Bastide’s scientific care and the reflections he had alreadymade from the writings of his Brazilian predecessors also led him toquestion the interpretations of these religious phenomena, as well astheir generalisations:

Up until now, studies have been restricted to superficial descriptions of thefetish cult of Bahia. The moment has come to devote ourselves to researchesthat require greater patience. It is necessary to construct monographs on allthe terreiros, at least the more traditional ones, together with all theiractivities. When such monographs have been completed, studies carried outon the same nation can be compared first, and then the various nationscompared among themselves.20 I believe this method will be particularlyfertile in separating false generalizations from true ones, for the researchertends to take everything he states about a terreiro as a general characteristic.In doing this, a false idea is provided of a cult which should only be spokenof with the greatest respect. For my part, every time I made any observationwhatever, I sought to localize it in order then to pose the same question toother informers. Now, I have often stated that such-and-such gesture ormyth did not go beyond a restricted domain, that of a nation, and often eventhat of a unique terreiro. (ibid.: 73–74)

This recommendation was not actually followed, since, in the workBastide was to write in 1958 on the candomblé of Bahia, all referencesto particular individuals or places have been removed. The descriptionhe presents therefore forms the basis for an analysis of the ‘subtlemetaphysics’ of the candomblé in which all the internal differences, soevident in his first work of 1945, have been finally erased. In the 1958work, one discerns the influence of Marcel Griaule and of a certain typeof Africanism, which, since the 1930s, had sought to cast light onAfrican ‘philosophies’ and ‘metaphysics’.

Motta (1994: 105) has raised again the question of Griaule’s possibleinfluence over Bastide, affirming that a certain number of ideas developedin Le Candomblé de Bahia (1958) had already been anticipated in Imagesdu Nordeste mystique en noir et blanc (1945; see 1995), in which Bastidedeclared that the ‘philosophy of the candomblé’ is not a ‘barbarousphilosophy’ but a ‘subtle form of thought’ that had to be deciphered. Now,this same Bastide was aware of the closeness of his 1958 volume toGriaule’s work. Thus, in the introduction, written in 1972, to his Estudos

182 Stefania Capone

afro-brasileiros (‘Afro-Brazilian studies’; Bastide 1973b), a collection of aseries of essays originally published between 1946 and 1953, Bastidewrites as follows about this first stage of his Brazilian sojourn:

I would like to stress that, at this period, I had not yet had an opportunity toread Griaule’s first books (the circumstances of the Second World War hadprevented me from receiving these books in Brazil). At around the sametime, Griaule arrived at some conclusions that were close to mine on thesubject of Dogon religion, which concealed a philosophy as rich and asvaluable as those of a Plato or an Aristotle. (Bastide 1973b: xii)

Other authors have underlined the affinities between Bastide andcertain parts of French intellectual production, especially with thefounding group of the Collège de Sociologie, Michel Leiris, GeorgesBataille and Roger Caillois. In his 1945 volume, the only one of thisgroup whom Bastide quotes is Leiris, an emblematic figure in thinkingabout the intimate articulation between art and anthropology. Leirishad discovered anthropology thanks to surrealism, which ‘represents arebellion against western rationalism, being conveyed, among otherthings, through a curiosity about primitive peoples and about primitivementality’ (Peixoto 2005: 136). The initiative in creating the Collègede Sociologie must therefore be understood in this context of culturalcriticism, fed as much by surrealism as by anthropology. Bastide’sthematic affinities with this group are obvious: common themes includethe dream, the unconscious, mysticism and the festival. The notion ofthe sacred was the intellectual focus of the group which had for itsultimate aim, according to Caillois (1950 [1939]), to ‘restore to societyan active, epidemiological and contagious sacred’, a project echoed inthe pages of Sacré sauvage et autre essais, published by Bastide in 1975.

It was with Leiris that Bastide maintained relations the most. Thus,‘in a warm letter’ dated 15 May 1958 replying to Leiris, who had senthim a copy of his La possession et ses aspects théâtraux chez les Ethiopiensde Gondar (1958), published that year, Bastide writes of havingsubscribed in one of his Brazilian books to the thesis of the ‘wardrobe’of personalities, which to him ‘appears as necessary to apply to Americaas to Africa’ (Leiris 1996: 909).21 The affinities with Leiris are not justa matter of the topics dealt with but also, and above all, concernmethod. If Leiris was a writer and a poet concerned to explore all theresources of a language, he also sees in poetry a way of being able toadvance knowledge about human beings. Bastide stressed several timesin his writings the importance of poetry as a ‘form of knowledge aboutthe world’. Thus, in Brésil terre de contrastes (1957: 16), he writes:


The sociologist who studies Brazil does not know which system of conceptsto use. The ideas he has learned from European or North Americancountries are no longer any good. The old is mixed with the new. Historicalperiods are all tangled up with one another. The same words, such as ‘socialclass’ or ‘historical dialectic’, do not have the same meanings and do notapply to the same concrete realities. Instead of rigid concepts, it is necessaryto discover notions that are to some extent fluid, capable of describing thephenomena of fusion, turmoil and interpenetration, and moulded on aliving reality in perpetual transformation. The sociologist who wants tounderstand Brazil must often transform himself into a poet.

This link between scientific research and ‘poetic knowledge’ had alreadybeen formulated in a double article published in 1946 in the Diário deSão Paulo, under the title ‘A propósito da poesia como método sociológico’(‘On poetry as a sociological method’). In order to grasp a moving socialreality, the sociologist must situate himself within the social experiencehe is studying; he must adhere to the ‘soul’ of the fact being studied.Understanding only becomes possible through this ‘transfusion of souls’,which forces the researcher to abandon his position as an externalobserver. Any type of judgement concerning the social reality beingstudied that proceeds from external categories must be rejected:

For the sociologist, it is a matter of not situating himself externally inrelation to social experience, but of living it […] we have to transformourselves into what we are studying, whether crowd, mass, class or caste …As in an act of love, we must transcend our own personalities in order toadhere to the soul that is attached to the phenomenon being studied.(Bastide 1946, quoted in Queiróz 1983: 17)

But, to do this, it was necessary to modify profoundly the researcher’sown logical categories if one were to expect finally ‘an anti-ethnocentricmentality’ (Bastide 1973b: xi). The ‘crisis of conscience’ in what Bastidecalled his ‘spiritual itinerary’ would lead him to ‘convert’ himself to adifferent mentality. It was necessary to abandon a mentality producedby ‘three centuries of Cartesianism’ by fully accepting a ‘conversion’presented as a crucial stage in the work of research: ‘Scientific researchrequires from me the initial passage through a rite of initiation’ (ibid.).

Initiation, or the ‘anthropology of chiasmus’

After his trip to the Nordeste in 1944, Bastide only returned to Salvadorin January 1949 and August 1951.22 In 1946 Bastide met PierreVerger, who had arrived in Brazil after a long journey through LatinAmerica. Bastide advised him to go to Bahia ‘to find Africa again’, which

184 Stefania Capone

Verger already knew, having worked there as a photographer. Vergerarrived in Salvador on 5 August 1946. Having fallen for the charms ofthe city and its religious traditions, he decided to establish himself there,where he was to remain until his death in 1996.23 The spiritual son ofSenhora d’Oxum, the mãe-de-santo of Axé Opô Afonjá, Verger was a sortof alter ego to Bastide, opening doors for him to the Bahian candomblé,like Amado in earlier years.

When Bastide undertook his second trip to Salvador in 1949, Vergerwas in Africa and could not accompany him on his visits to the cult housesof the candomblé (Verger 1978). It was on this trip that Bastide achieved hisfirst divination séance with Vidal, the pai-de-santo or cult leader of thequarter of Brotas, who revealed to him his mystical connection with thegod Xangô. On his third trip in 1951, in the celebrated terreiro of Axé OpôAfonjá, to which Verger was already affiliated, Bastide completed theceremony of the consecration of the necklaces, called the lavagem dascontas (‘washing of the necklaces’), through which is established theminimum of involvement between an individual and his protectingdivinity. By this time, Verger had already become his main interlocutor asort of local representative of the world of candomblé and principal‘translator’ of the religious universe for his compatriot.

This event, the consecration of the necklaces in the colours of theprotecting divinities, is the source of a sort of ‘mysticism of initiation’ inthe writings of certain authors who use Bastide as their authority,though this seems to mistake the internal organisation of the candomblé.Thus, Claude Ravalet (2005: 124) writes that Bastide was ‘initiatedunder Xangô’s aegis in the night of 3 to 4 August 1951’. However,Bastide, in an article devoted to this ceremony, states that this was not aninitiation: ‘It was a private ceremony, of little importance, quite ordinaryand, doubtless for this reason, neglected by researchers’ (Bastide 1973a:363–64). The washing of the necklace is thus represented as ‘a form ofincorporation into the candomblé in which one does not pass throughinitiation or the rite of “feeding the head”. […] The washing of thenecklace constitutes the first moment in this incorporation; “feeding thehead” is the second moment, initiation the third’ (ibid.: 364).24 Thus, ifthe washing of the necklace does not strictly speaking constitute aninitiation, it nonetheless marks the individual’s link with the cult group,placing him under the authority of the spiritual chief, the mãe or pai-de-santo who has conducted the ritual. The washing of the necklace –‘which, as I have just said, is merely the first degree of incorporation ofthe candomblé, the lowest level, the most ordinary of all’ (ibid.: 373) –marks the acceptance, on the part of the individual, of a range ofprohibitions and financial duties in respect of the cult house to whichhe belongs. In no way is he acknowledged by the initiated group as one


of them. He is, on the other hand, part of the group of abian, the futurenovices, candidates for initiation proper.

This becomes evident if one reads attentively the passage fromBastide’s thesis of 1960, where he declares: ‘Africanus sum, inasmuchas I have been accepted by one of those religious sects, which regardsme as a brother of the faith, having the same obligations and the sameprivileges as the other members of the same degree’ (Bastide 1978: 28, myemphasis). Bastide knew very well that his incorporation into the cultgroup was very superficial and did not permit him to attend all the ritualsof the cult house. This reluctance to cross the threshold towards a ‘trueconversion’ has been made the subject of analyses by two of Bastide’sclosest collaborators, Maria Isaura de Queiróz and Françoise Morin. ForBastide, genuine participant observation, the ‘transfusion of souls’, hadto be transformed into ‘controlled observation’ through the ‘self-criticism’ of the researcher (Queiróz 1983: 17). If it is true that Bastidesearched throughout his life for what Morin (1975) calls ‘ananthropology of the abyss’, his fascination with ‘the possibility ofvertigo’ persisted up until his last years. Thus, in some notes edited in1968 and entitled ‘Réflexion sur une agitation’, he wrote, of May 1968:‘I find in it one of my fundamental desires (I have been defined previouslyas someone who circles around the abyss in order to feel the seductionof its vertigo, but who is firmly attached to the mast of the ship; one musthear the Sirens but not drown)’ (quoted in Morin 1994: 24).

As Morin emphasises, for Bastide the seduction of the abyss was foundabove all in the study of the trance, the dream and ‘polytheistic mysticism’.Thus, ‘doubtless through a fear of drowning’, Bastide resisted the singingof the Sirens, clung to the mast of reason and did not pursue his initiationfurther (ibid.: 25). In 1965, in an interview he gave to the journal Combat,he confided to Jacques Delpeyrou: ‘In my passionate quest for mysticalexperiences, I have always had a fear of going mad’ (ibid.).

Another route had been taken by Bastide’s friend and ‘accomplice’ PierreVerger. Far from wanting to devote himself to scientific studies, Verger washardly interested in the work of the anthropologist. In his journeys to Africa,he took notes in order to ‘fulfil his role as a messenger’, making clear thefaithfulness of the Blacks of Bahia to African traditions by comparing Africaand Bahia. His wish was to be able to ‘recount Africa’ to his Bahian friends(Métraux and Verger 1994: 62).25 Verger spent many years going betweenBrazil and Africa, where in 1953 he had been initiated into the Ifá cult andbecome a babalawo (diviner) with the ritual name of Fatumbi: ‘Ifá broughtme back into the world’. Verger had already conducted a borí in the Bahianterreiro of Axé Opô Afonjá in 1948. Mãe Senhora, who, ‘in a clever move’,had brought Verger under her spiritual protection (cf. Nobrega andEcheverria 2002: 178–9), consecrated his head to Xangô.26 When Verger

186 Stefania Capone

returned from Africa, Senhora bestowed on him the title of Oju Oba (‘theeyes of the king’, a ritual title given to one of Xangô’s sons), thusincorporating him finally into the hierarchy of her terreiro.

Now, initiation carried out in Africa seemed to obey a very differentlogic from that put forward by Bastide:

I went through my initiation not to ‘see’ (predict the future), but because itgave me access to the knowledge of the babalawo, those who transmit orally allthe knowledge of the Yoruba people. […] It was this side that interested me,since I had not only the right to learn, but the duty to do so. This is completelydifferent position from that of the anthropologist who enters like an analyst,with more or less idiotic questions which make no sense to the people … Iconducted all my researches without posing any questions, merely collectingwhat the people judged important and what was linked to the corpus ofknowledge of the babalawo. (cited in Nobrega and Echeverria 2002: 202)

In reality, what stimulated Verger was his desire to rid himself of hisformer bourgeois identity in order to identify fully with the Nagô, thatis the inhabitants of the eastern region of Dahomey (present-dayBenin). As he wrote in his correspondence with Métraux, whom hedescribed as ‘almost his twin’, since 1952 his ‘sympathies’ for thispeople had been confirmed even more (Métraux and Verger 1994: 157).Verger never ceased to reaffirm his desire to ‘go native’, to ‘becomeNagô’. Thus, on 12 May 1956 he wrote as follows:

And yes! I have become very conformist, seeking Yoruba dignities andhonours, while I have only sarcasm for those I could acquire in my ownsocial milieu. This is very much what Lévi-Strauss describes [in Tristestropiques]: ‘Willingly subversive among his own kind, and rebellious withregard to traditional usages, the ethnographer appears respectful to theextent of being conservative as soon as the society being envisaged is foundto be different from his own’. If I do not wear a ribbon, on the contrary Idisplay a bracelet of yellow and green pearls when in the milieu of theYoruba Nagô, the insignia of my dignity as a babalawo. I refuse to kiss thehands of the dowagers, but I willingly and ostentatiously make extravagantbows in another style and utter interminable salutations in front of strangeold crabs. (Métraux and Verger 1994: 229)

Verger’s ‘conversion’, unlike Bastide’s, does not seem to have beendictated by any inclination towards belief. Verger never ceased to declarehis incredulity. Thus, in an interview he gave to Gilberto Gil a few daysbefore his death in February 1996, he stated that he had neverexperienced the state of trance, since he was ‘an idiot as a Frenchman,a rationalist … I am not an idiot for believing in these things’ (quoted inNobrega and Echeverria 2002: 388). In reality, Verger was never


inspired by profound mystical or metaphysical questioning as wasBastide: his transformation into Nagô was not the result of any spiritualquest, but the keystone to his quest for an identity. Upon his death,Verger was finally incorporated into the Nagô cult, attaining the statusof an ancestor venerated in the cult of the eguns or revenants.Paradoxically for someone who remained an unbeliever right up to theend, Verger’s Baba Egun ‘manifested himself ’ for the first time inJanuary 2005 in the terreiro of Balbino and received the name offunlade, ‘the ancestral spirit made of efun [chalk, a sacred substancelinked to Oxalá]’ (Souty 2007: 366).27

Verger’s African initiation and the privileged links he was able toweave with the Bahian cult houses made him an ideal collaborator forBastide, who, as we have seen, was greatly restricted in his researchwork. The extensive correspondence between Verger and Bastide,consisting of 227 letters over 27 years, reveals a genuine collaboration‘between Verger, more nomadic, travelling unceasingly between Africa,Brazil and the Caribbean, and Bastide, more sedentary because he wasteaching in São Paulo and Paris’ (Morin 1994: 40). Thus, in a lettersent to Verger on 30 June 1947 while leaving for Belém de Pará, Bastidewrites: ‘Believe me, I envy you, being imprisoned by my students, farfrom the most cherished objects of my research. If you can, in thecourse of your travels, collect some texts of hymns, at Belém orMaranhão. I would be interested in having them to compare them withthose of Recife or Bahia’ (quoted in Morin 1994: 40).

At the end of the 1950s, Verger’s works inaugurated a new phase inAfro-Brazilian studies by suggesting links providing continuities withAfrican cultures. His 1957 work had ‘the impact of a bomb’ in anenvironment in which all the specialists were insisting on acculturation,syncretism and change (cf. Bastide 1996: 18). The study of Afro-Brazilian religions therefore became a hunting ground for the‘Africanists’, a new generation of Brazilian anthropologists who,inspired by Verger’s work, went to Africa to search for proof of thefaithfulness of certain Afro-Brazilian religious practices to the Africantradition. For Bastide, Verger embodied the ‘transfusion of souls’ whichhe had pursued throughout his own life:

No one could make these Afro-Brazilian cults better known and loved thanPierre Verger. Not only have his numerous voyages to the two coasts allowedhim to compare the Brazilian ceremonies with those of Africa and todiscover their perfect unity, but he is not an ‘outsider’, a stranger lookingwith curiosity and capturing hieratic gestures or faces in trance on thesensitive plate. He belongs to the world of the candomblés and has beenaccepted by the Blacks of Bahia as one of themselves, like a true brother, awhite brother. North American sociologists have invented a term to

188 Stefania Capone

designate a research method which consists precisely of identifying oneselfwith the milieu one is studying, namely ‘participant observation’. However,Pierre Verger is more than a participant observer, for the word ‘observer’still marks out a certain barrier, splitting the ethnographer in two in a ratherunpleasant fashion as a ‘man of the outside’ and a ‘man of the inside’. ForPierre Verger, knowledge is the fruit of love and communion. (Bastide,Preface to Verger 1995 [1954]: 11)

Bastide tried several times to make his collaboration with Verger official,regretting that their two names did not figure ‘fraternally, one alongsidethe other’ (Morin 1994: 41). Verger almost always opposed the ‘law ofsecrecy’, to which he should have submitted as one who had beeninitiated, at the insistent requests of his friend.28 The works of these twoauthors demonstrate two models of ethnographic texts: one, Bastide’s,a highly abstract model of the candomblé, which, as da Silva recalls(2000: 127), owes its form to specific strategies of the textualdescription and interpretation of ethnographic data; and anothermodel of ethnographic narration, Verger’s, that of ethnography as the‘photography’ of a reality which becomes apparent in the choice ofscenes to be shown in his work as a photographer, always presented asundeniable testimonies of the continuity of African traditions withBrazil and of the nearness of the candomblé rituals to those of the WestAfrican coast (ibid.: 130). For Verger, photography had to suffice byitself, without any other commentary or explanation. It is only at asecond period, motivated by his desire to demonstrate the Africanfaithfulness of Brazilian cults, that Verger was to use the image ‘in amore and more didactic fashion’ (Souty 2007: 133).

The ambiguous relationship that Verger maintained with religious beliefmade him a sort of alter ego of Bastide’s, who remained, right up until hisdeath, much more fascinated by mysticism than Verger. His distance, at theexistential level, thus seems to be inversely proportional to this fascination.

Return to France, or ‘those excluded from the horde’

After 1951, and up to 1954, Bastide started to prepare his return toFrance, dividing his time between São Paulo and Paris. Lucien Febvrehad offered him a post as Directeur d’études at the École Pratique desHautes Études (EPHE). He finally returned to his own country in 1954.Bastide multiplied his teaching by giving a course on the sociology ofBrazil and another on Brazilian literature to the Institut d’AmériqueLatine. Art and sociology continued to be closely linked in Bastide’strajectory. And it was in Paris that he resumed his association withGeorges Gurvitch, who encouraged him to prepare his thesis.29


Upon Gurvitch’s death in 1965, Bastide became the director of theLaboratoire de Sociologie de la Connaissance, which he had created.30

Gurvitch’s position in the French academic world was unique: very muchinfluenced by Marxism and German phenomenology, and concerned tocombine certain Durkheimian themes with phenomenology, he never felthimself to be a true representative of French sociology, defining himselfas being ‘excluded from the horde’ (Gurvitch 1966).31

Gurvitch was persuaded that the discipline was in crisis because ithad still not succeeded in joining together theory and empiricalresearch. Confronted with Durkheim’s theories, Gurvitch adopted theidea of social life being divided into different levels of reality, whilecriticising Durkheim’s ‘dogmatism’ and his view of God as a projectionof the collective consciousness (Gurvitch 1959: 4). Such would be theconsequences of a ‘conceptual deficiency, the cause of which is theDurkheimian definition of the social fact, from which unfolds thecollective consciousness, to which is attributed the status of thesupreme being’ (Marcel 2001: 13).

Gurvitch was very critical of German sociology, and especially ofMax Weber. The ‘young Marx’, by contrast, was one of his main pointsof reference because of the attention Marx gave to ‘the reciprocalimmanence between society and the individual’ (Gurvitch 1948: 22).Mauss, with his notion of the ‘total social fact’, helped him work out hismethod, in which sociology and psychology were mixed together inorder to achieve a genuine ‘reciprocity of perspectives’. The synthesis ofthe teachings of Mauss and Marx would lead Gurvitch to develop atypology of ‘depth levels’, which constituted an attempt to resolve thequestion of the separation between individual and society, thanks tolevels which go from the morphological surface of society to the mentaland psychic states of the individual.

However, in order to arrive at a unified paradigm for the discipline,North American sociology also had to be confronted:

The enormous descriptive work provided by American sociology has shownthe way here, although to bear all its fruits, and even to become usable, itneeds to be rooted in conceptual schemes that are clearer, more refined andmore flexible, such as those which are the strength of French sociologicalthought. (Gurvitch 1950: 4)

In this respect, Bastide signed up fully to the Gurvitch school, since forhim North American anthropology was merely the point of departurefor the analysis of the interpenetrations of civilisations. He found theintellectual tools for the development of this analysis in Frenchsociology and anthropology: Durkheim’s and Mauss’s notion of the

190 Stefania Capone

collective representation, Halbwach’s definition of collective memoryand Lévy-Bruhl’s law of participations.

Between 1945 and 1960 in French sociological circles, Americanauthors were commented on in terms of a conception that referred toa ‘methodological holism’ inherited from Durkheimian and Marxistthought (Marcel 2004). It was a matter of appropriating methods ofenquiry created across the Atlantic while rebuilding around them aconceptual inventory compatible with the need to explain the parts ofthe whole in all circumstances. American authors were therefore nomore than ‘purveyors of methods permitting the collection of micro-sociological data’ (ibid.). Bastide adopted this critique by affirming that‘the great task of sociology was to establish valid generalizations frommonographic descriptions’ (Bastide 1960: 324).

In the introduction to his major thesis, defended in 1957 at EPHEunder Gurvitch’s supervision, Bastide examines the contributions ofMarxism, Durkheimian sociology and French anthropology in theanalysis of processes of the interpenetration of civilisations. Positivistsociology must be replaced by a ‘sociology of understanding’, whichmust be freed from Weberian subjectivism: ‘For the understanding heseeks is attained by reference to the observer, i.e. the sociologist whointerprets the correlations. This is to ignore the fact that the sociologisthimself is part of society, that he has been shaped by a given culture,and consequently that his own psychology has also been conditionedby social factors’ (Bastide 1978: 5).

For Bastide, however, the determining factor of the new sociologicalconceptions must be sought in the very evolution of anthropology atthe start of the twentieth century, and especially in the contributions ofGranet and Leenhardt (1930, 1947): ‘A civilization does not reveal itstrue meaning unless it is grasped through its mythical vision of theworld, which is more than its expression or justification, being indeed itsvery mainstay’ (Bastide 1978: 9). In American sociology, ‘what had soclosely unified French sociology … was split up into three differentdisciplines: sociology proper, cultural anthropology and socialpsychology’. With Gurvitch and his ‘depth sociology’, ‘the stratificationof the depth levels of social reality made a much richer dialectic possible… it made possible the transition from statics to change, from situationto cause; in brief it allowed explanation to be more effectively modelledon concrete fact in its perpetual transformation’ (ibid.: 10–11).

Now, the sole means open to Bastide to achieve genuineunderstanding was ‘to become the Other’. But how could one integrateoneself into an ‘African’ religion? This was possible since, in Brazil,‘there is a dissociation between culture and race’:


In the candomblé we find Spanish ‘daughters of the gods’ as well as Frenchand Swiss members of the priestly hierarchy with various titles. […] All thatis required is to accept African law wholeheartedly. From that moment on,however white one’s skin, one is caught up in the shared mystic rites, intaboos, and in susceptibility to magical vengeance. In fact it is possible inBrazil to be a Negro without being African and, contrariwise, to be bothwhite and African. (ibid.: 28)

If, therefore, Bastide declares ‘Africanus sum’, he knows that his positionin relation to the candomblé prevents him from achieving this deepunderstanding:

it was evident that, even though I entered the candomblé as a ‘member’ andnot at all as a simple observer, the law of the maturing of the secret, whichdominates any religion of initiation, forced me to remain still too much ofan outsider for me to be able to provide anything other than an introductionto a certain Negro vision of the world. Only a priest of the cult who occupiesan elevated position in the hierarchy could provide us with the work I waspraying for. (Bastide 1996: 18–19)

This ‘sociology of the encounter’ (Balandier 1995 [1960]), which demandsthe fruit of both scientific knowledge and shared belief, was thus condemnedto failure. Only those who had been initiated could achieve genuineunderstanding, but they could not reveal it, since, as in Verger’s case, theywere bound to the ‘law of secrecy’. Like the mystic, Bastide tried to positionhimself on another plane of existence, gazing towards somewhere elsewhich was in reality at the root of himself. As he noted in his article of 1965on ‘obscure and confused thought’, to lose consciousness to the point oflosing oneself is equivalent to being plunged into darkness, to forgettingoneself in the silence of the self, until the ultimately revealing encounterwith the Other. Right up to the end, Bastide was bound to the mast of reason,a new Ulysses trying to resist the Sirens’ songs.

Notes

1. For a critical analysis of this domain, see Capone (2005).2. In his preface to a work of Bastide (1960), Georges Balandier writes (1995: v): ‘The

separation, the delayed access to Parisian academic positions, the gap with respect tothe ideological confrontations of the 1950s were for him a handicap. Besides thesecircumstances, equally important was the distance he was able to maintain betweenwhat was important to him and the concessions that had to be made to maintain hisreputation.’

3. On the ‘missed encounter’ between Bastide and Lévi-Strauss, see Mary (2000a,2005).

4. For an analysis of the limits of this approach to the field, see Capone (1999: 41–48).

192 Stefania Capone

5. Richard taught at the University of Bordeaux, where he succeeded Emile Durkheim,who occupied the Chair of Pedagogy and Social Sciences from 1887 to 1902. He waslinked to the Année Sociologique group, with whom he collaborated until 1907. ForBastide’s biography and bibliography, see Ravalet (1993a, 1993b).

6. Bastide stresses Tarde’s contribution (1890, 1898) to the study of the‘interpenetrations of civilizations’ in one of his last works, Le prochain et le lointain(1970), where he defines him as ‘the true founder of cultural anthropology’.

7. This letter from Mauss to Bastide has been published in Bastidiana no. 49–50 (2005).See also Morin (1994: 38).

8. I have devoted another work to Bastide’s ‘principle of compartmentalization’ andtheory of syncretism. See Capone (2007).

9. On this, see Bastide (1965), Beylier (1977: 234–37).10. Brazil seems to have been a fertile field for the rediscovery of Durkheimian sociology,

given Durkheim’s influence in the Paulist academic milieu of the period, especiallyamong the members of the French mission in São Paulo (cf. Peixoto 2000). In hisdiscussion of the relationships between magic and religion, Bastide partly adoptsDurkheim’s formulations (1970) in opposing a ‘disaggregative’ magic to an‘integrative’ religion.

11. This opposition to Durkheimian theories was not so clear in previous writings. Thusin 1945 Bastide wrote as follows on the subject of the trance or ‘mystical crisis’: ‘it isinscribed in a cultural ensemble, it follows a certain number of collectiverepresentations, and one can say of the candomblé what I have said elsewhere withregard to Durkheim, that a mysticism that commences at a fixed moment in order toend at a given moment while always following certain rules, far from explaining thesocial, can only be explained by the priority of the social over the mystical. … theexplanation for the trance must be sought in sociology, in the constraint of the milieuover the individual’ (1995: 100). This approach was modified fundamentally someyears later. Thus, in an article on the ‘washing of the necklace’, originally publishedin 1953, he writes: ‘the structure of the social is determined by religious conceptionsand by the African philosophy of the universe. If we want to understand themorphological organisation of groups, we are obliged to pass through religioussociology, since it alone possesses the explanatory key’ (1973a: 370).

12. Jean-Paul Lefebvre (1990) lists thirty-eight French professors who took part inuniversity missions to Brazil from 1934 to 1944. They occupied especially the chairsin the humanities and social sciences, while German and Italian professors sharedthe chairs in the exact and natural sciences. See also Peixoto (1989), Carelli (1993).

13. The sites and periods of his fieldwork were as follows: Bahia and Recife (December1943 to February 1944), Porto Alegre (July 1945), Bahia (December 1948 toFebruary 1949), and Bahia and São Luiz do Maranhão (July–August 1951). Theseperiods correspond to university vacations in Brazil.

14. On the fertile ‘dialogues’ between Bastide and Brazilian intellectuals, as well as hisdebt to them, see Peixoto (2000), Capone (2007).

15. I have dealt with the relations between Herskovits and Bastide elsewhere (Capone2007). Herskovits spent six months in Brazil at the start of the 1940s and supervisedanthropological theses there, such as those by Octavio da Costa and René Ribeiro,while maintaining close links with Brazil, especially with Arthur Ramos. Cf.Guimarães (2004).

16. Ramos had already published, in 1936, an introduction to social psychology beforeBastide arrived. For his interpretation of the process of acculturation, see Ramos(1979 [1937]).

17. From this journey Bastide would draw a book, published in Brazil in 1945 (see Bastide1995). In this text, there are numerous errors concerning the terminology of the


candomblé and great confusion in transcriptions of the names of the orixás ordivinities. The text was only published in France after Bastide’s death.

18. An obá is a particular type of ogan (cf. Capone 1999: 260–70). In this text, Bastidecalls the mãe-de-santo (chief priestess) of the Casa de Oxumarê ‘Dona Cotilha’ insteadof ‘Dona Cotinha’. Similarly, the celebrated pai-de-santo Joãozinho da Goméia, hisprincipal informant at this time, becomes ‘João da Gávea’ (cf. Bastide 1995: 85, 63).

19. Here is an example of the Bastidian style, in which poetry seems to be the principalroute to understanding reality: ‘The twilight chased me, forcing me to take the tram,which was full of black beggars, dreaming workers, amorous soldiers, ironic mulattos,brigands decked out in the most incredible shirts, sky blue with golden braids. Ireturned to my hotel, waiting for the tom-toms of the festival to call me back to thoseplaces, this time dominated by the night, the music and the divine madness’ (ibid.: 92).

20. The candomblé is divided into several ‘nations’: the nagô (ketu, ijexá, efon), the bantu(angola and congo), the jeje and the candomblé de caboclo, in which the indigenous spiritsare venerated. Today this division into nations reflects a difference in the unfolding ofthe ritual rather than actual ‘ethnic’ origins.

21. See also the letters sent by Leiris to Bastide, published in no. 49–50 of Bastidiana(2005).

22. Two other brief trips were made after his return to France, in September 1962 andAugust 1973.

23. On Verger’s itinerary between Brazil and Africa, see Métraux and Verger (1994). Agreat number of works were published in Brazil at the centenary of Verger’s birth in2002. Among others, see Le Bouler (2002), Lühning (2002, 2004), Moura (2002),and Nobrega and Echeverria (2002). On Verger, see also Capone (1999), Souty(2007).

24. The words ‘feeding the head’ designate the ceremony of the bori (bo + ori = ‘tovenerate the head’), in which the head, the receptacle of the gods, is ‘fed’ by the bloodof the sacrifices and the offerings of sacred food. Initiation proper is called feitura dosanto, ‘the making of the saint [orixá]’.

25. When he started, Verger had no vocation for anthropological work, as he declaredhimself: ‘I did this research for myself and my friends in Bahia. The idea of publishingthe results for a wider public never entered my mind. It was Monod who made mepublish’ (Verger 1982: 257). Théodore Monod, Director of the Institut Françaisd’Afrique Noire or IFAN in the 1950s, had invited Verger to publish a study of cultsin Brazil and Africa, which appeared in 1957. Verger entered the Centre National dela Recherche Scientifique in 1962, finally becoming a directeur de recherche in 1972,at the age of 70.

26. See Souty (2007), Le Bouler (2002), on the tension – expressed in the attribution oftwo different protecting divinities (Xangô in Axé Opô Afonjá and Oxalá in the CasaBranca) – that brought the mães-de-santo of two of the most prestigious terreiros inSalvador into mutual opposition, since both wished to incorporate Verger into theircult group.

27. On the ‘counter-acculturative’ effect of Verger’s work and its consequences for theritual practice of the candomblé, see Capone (1999: 253–59), Souty (2007: 256–62).The search for a ‘Nagô purity’ in the creation of a new terreiro, that of his ‘protégé’Balbino Daniel de Paula, in which Verger participated actively, is an excellentillustration of the role he played in legitimating a ritual model (ketu) in the Bahiancandomblé. Cf. Capone (2003).

28. They only published two articles together, a study of divination in Salvador in 1953,and an article on the Nagô markets of Dahomey in 1959.

29. For information on Gurvitch’s biography, see Gurvitch (1966), Duvignaud (1969),Balandier (1972).

194 Stefania Capone

30. This raises a question about Marcel’s statement (2001: 6) that, after 1965, ‘therewas no longer a Gurvitch school’. Gurvitch had left his mark on Bastide’s work,notably on his theory of syncretism and of the ‘Negro-African collective memory’(cf. Capone 2007). See also Mary (2000a, 2000b).

31. One could say the same about Bastide, cf. Balandier (1995).

References

Balandier, G. 1972. Georges Gurvitch, sa vie, son oeuvre, Paris: Presses UniversitairesFrançaises.

——— 1995 [1960]. Une sociologie de la rencontre, Préface to Roger Bastide, Les religionsafricaines au Brésil: contribution à une sociologie des interpénétrations de civilisation,Presses Universitaires Françaises.

Bastide, R. 1928. Mysticisme et sociologie, Revue Internationale de Sociologie, 36(5/6):297–306.

——— 1931a. Les Arméniens de Valence, Revue Internationale de Sociologie, 39(1/2), 17–42.

——— 1931b. Les problèmes de la vie mystique, Paris: Armand Colin.——— 1935. Eléments de sociologie religieuse, Paris: Armand Colin.——— 1938. Pintura e Mística, Revista do Arquivo Municipal (São Paulo), 50: 47–60.——— 1940. Psicanálise do Cafuné, Revista do Arquivo Municipal (São Paulo), 6(70):

118–30.——— 1946. A propósito da poesia como método sociológico, 2 parts, Diário de São Paulo,

8 and 22 February.——— 1957. Brésil, terre de contrastes, Paris: Hachette.——— 1958. Le candomblé de Bahia (rite nagô), Paris: Mouton.——— 1960. Problèmes de l’entrecroisement des civilisations et de leurs œuvres, in G.

Gurvitch (ed.), Traité de sociologie, Paris: Presses Universitaires Françaises, vol. 2.——— 1965. La pensée obscure et confuse, Le Monde Non-Chrétien, 75/76: 137–56. ——— 1970. Le prochain et le lointain, Paris: Edition Cujas.——— 1973a. Algumas considerações em torno de uma ‘lavagem de contas’, in R.

Bastide, Estudos Afro-brasileiros, São Paulo: Perspectiva.——— 1973b. Estudos Afro-brasileiros, São Paulo: Perspectiva.——— 1975. Le sacré sauvage et autres essais, Paris: Payot.——— 1978 [1960]. The African religions of Brazil: toward a sociology of the

interpenetration of civilizations [Les religions africaines au Brésil: contribution à unesociologie des interpénétrations de civilisation], Baltimore: Johns Hopkins UniversityPress.

——— 1995. Images du Nordeste Mystique en Noir et Blanc, Paris: Actes Sud/Babel (orig.Brazilian edn. 1945).

——— 1996. Etat actuel des recherches afro-américaines en Amérique Latine,Bastidiana, 13/14: 11–28.

Bastide, R. and P. Verger. 1953. Contribuição ao Estudo da Adivinhação em Salvador(Bahia), Revista do Museu Paulista, 7: 357–80.

——— 1959. Contribution à l’étude sociologique des marchés nagô au Bas-Dahomey,Cahiers de l’Institut de science économique appliquée, 95: 33–65.

Beylier, C. 1977. L’œuvre brésilienne de Roger Bastide, Thèse de doctorat, Paris, 2 vols.Caillois, R. 1950 [1939]. L’homme et le sacré, Paris: Gallimard.Capone, S. 1999. La quête de l’Afrique dans le candomblé: pouvoir et tradition au Brésil, Paris:

Karthala.


——— 2003. Regards contemporains sur les premiers candomblés à Salvador de Bahia,in A. Kouvouama and D. Cochart (eds), Modernités transversales: citoyenneté, politiqueet religion, Paris: Éditions Paari.

——— 2005. Repenser les ‘Amériques noires’: nouvelles perspectives dans la rechercheafro-américaniste, Journal de la Société des Américanistes, 91(1): 83–91.

——— 2007. Transatlantic dialogue: Roger Bastide and the African American religions,Journal of Religion in Africa, 37(3): 1–35.

Cardoso, I. 1994. Entretien avec R. Bastide, Bastidiana, numéro spécial, ‘Roger Bastide:Claude Lévi-Strauss. Du principe de coupure aux courts-circuits de la pensée’, 7/8:69–73.

Carelli, M. 1993. Cultures croisées: histoire des échanges culturels entre la France et le Brésilde la découverte aux temps modernes, Paris: Nathan.

da Silva, V.G. 2000. O antropólogo e sua magia, São Paulo: EDUSP.Dumas, G. 1923–24. Traité de psychologie, Paris: Alcan.Durkheim, E. 1895. Les règles de la méthode sociologique, Paris: Alcan.——— 1912. Les formes élémentaires de la vie religieuse, Paris: Alcan.——— 1970. L’avenir de la religion, in E. Durkheim, La science sociale et l’action,

Introduction et présentation de J.-C. Filloux, Paris: Presses Universitaires Françaises.Duvignaud, J. 1969. Georges Gurvitch, symbolisme social et sociologie dynamique, Paris:

Seghers.Guimarães, A.S.A. 2004. Comentários à correspondência entre Melville Herskovits e

Arthur Ramos (1935–1941), in F.A.Peixoto, H. Pontes and L.M. Schwarcz (eds),Antropologia, história, experiências, Belo Horizonte: Editora UFMG.

Gurvitch, G. 1948. La sociologie du jeune Marx, Cahiers Internationaux de Sociologie, 4,3−47.

——— 1950. La vocation actuelle de la sociologie, Paris: Presses Universitaires Françaises.——— 1959. Les cadres sociaux de la connaissance sociologique, Cahiers Internationaux

de Sociologie, 26: 165–72.——— 1966. Mon itinéraire intellectuel ou l’exclu de la horde, L’Homme et la société, 1:

3–12.Herskovits, M. 1938. Acculturation: the study of cultural contact, New York: Augustin.Le Bouler, J.-P. 2002. Pierre Fatumbi Verger: um homen livre, Salvador: Fundação Pierre

Verger.Leenhardt, M. 1930. Notes d’ethnologie néo-calédonienne, Paris: Institut d’Ethnologie.——— 1947. Do Kamo: la personne et le mythe dans le monde mélanésien, Paris: Gallimard.Lefebvre, J.-P. 1990. Les professeurs français des missions universitaires au Brésil (1934–

1944), Cahiers du Brésil Contemporain, 12: 89–100.Leiris, M. 1958. La possession et ses aspects théâtraux chez les Ethiopiens de Gondar, Paris: Plon.——— 1996. Miroir d’Afrique, textes édités par Jean Jamin, Paris: Quarto Gallimard.Lévy-Bruhl, L. 1922. La mentalité primitive, Paris: Alcan.Lühning, A. (ed.) 2002. Verger-Bastide: dimensões de uma amizade, Rio de Janeiro: Bertrand

Brasil.——— 2004. Pierre Verger, repórter fotográfico, Rio de Janeiro: Bertrand Brasil.Marcel, J.-C. 2001. Georges Gurvitch: les raisons d’un succès, Cahiers Internationaux de

Sociologie, 110: 97–119.——— 2004. Une réception de la sociologie américaine en France (1945–1960), Revue

d’histoire des sciences humaines, 11: 45–68.Mary, A. 2000a. Le bricolage africain des héros chrétiens, Paris: Editions du Cerf.——— 2000b. Le défi du syncrétisme: le travail symbolique de la religion d’Eboga (Gabon),

Paris: Editions de l’EHESS.——— 2005. Métissage and bricolage in the making of African Christian identities, Social

Compass, 52(3): 281–94.

196 Stefania Capone

Mauss, M. 1990 [1922]. The gift, London and New York: Routledge.Métraux, A. and P. Verger. 1994. Le pied à l’étrier: correspondance 1946–1963, Paris: Jean-

Michel Place (présenté et annoté par Jean-Pierre Le Bouler).Morin, F. 1975. Roger Bastide ou l’anthropologie des gouffres, Archives de Sciences Sociales

des Religions, 40: 99–106.——— 1994. Les inédits et la correspondance de Roger Bastide, in P. Laburthe-Tolra (ed.),

Roger Bastide ou le réjouissement de l’abîme, Paris: L’Harmattan.Motta, R. 1994. L’apport brésilien dans l’œuvre de Bastide sur le candomblé de Bahia, in

Philippe Laburthe-Tolra (ed.), Roger Bastide ou le réjouissement de l’abîme, Paris:L’Harmattan.

Moura, C.E.M. de. (ed.). 2002. Uma saída de iaô: Pierre Verger, São Paulo: Axis Mundi Editora.Nobrega, C. and R. Echeverria. 2002. Verger: um retrato em preto e branco, Salvador:

Corrupio.Park, R.E. 1928. Human migration and the marginal man, The American Journal of

Sociology, 33: 881–93.Peixoto, F.A. 1989. Franceses e Norte-Americanos nas ciências sociais brasileiras 1930–

1960, in S. Miceli (ed.), História das ciências sociais no Brasil, vol. 1, São Paulo: Vértice.——— 2000. Diálogos brasileiros: uma análise da obra de Roger Bastide, São Paulo:

EDUSP/FAPESP.——— 2005. Roger Bastide: Nordeste mystique, itinéraires africains et villes brésiliennes,

Bastidiana, 49–50: 127–40.Queiróz, M.I.P. de. 1983. Nostalgia do Outro e do Alhures: a obra sociológica de Roger

Bastide, in M.I.P. Queiróz (ed.), Roger Bastide, São Paulo: Ática.——— 1994. Roger Bastide, professor da Universidade de São Paulo, Estudos avançados,

8(22): 215–20.Ramos, A. 1936. Introdução à psicologia social, Rio de Janeiro: José Olympio.——— 1979 [1937]. As culturas negras no Novo Mundo, São Paulo: Editora Nacional.Ravalet, C. 1993a. Bio-bibliographie de Roger Bastide, Bastidiana, 1: 39–48.——— 1993b. Bibliographie de Roger Bastide, Bastidiana, 3: 7–110.——— 2005. Roger Bastide et le Brésil, Bastidiana, 49–50: 117–26.Redfield, R., R. Linton and M. Herskovits. 1936. Memorandum for the study of

acculturation, American Anthropologist, 38: 149–52.Richard, G. 1923. L’athéisme dogmatique en sociologie religieuse, Revue d’histoire et de

philosophie religieuse, 3: 125–37, 229–61.Souty, J. 2007. Pierre Fatumbi Verger: du regard détaché à la connaissance initiatique, Paris:

Maisonneuve et Larose.Tarde, G. 1993 [1890]. Les lois de l’imitation, Paris: Kimé.——— 1999 [1898]. Les lois sociales: esquisse d’une sociologie, Paris: Synthélabo.Teixeira, A. et al. 1952. Arthur Ramos, Rio de Janeiro: Ministério de Educação e Saúde,

Serviço de Documentação.Verger, P. 1957. Notes sur le culte des Orisa et Vodun: à Bahia, la Baie de tous les saints, au

Brésil et à l’ancienne côte des Esclaves en Afrique, Dakar, Mémoires de l’Institut Françaisd’Afrique Noire (IFAN), no. 51.

——— 1978. Roger Bastide, Revista do Instituto de Estudos Brasileiros, 20: 52–53.——— 1982. 50 anos de fotografia, Salvador: Corrupio.——— 1995 [1954]. Dieux d’Afrique: culte des Orishas et Vodouns à l’ancienne Côte des

Esclaves en Afrique et à Bahia de Tous les Saints au Brésil, Paris: Editions Revue Noire.

Chapter 8

THE ART AND CRAFT OFETHNOGRAPHY:

LUCIEN BERNOT 1919–1993

Gérard Toffin

In the landscape of the French anthropology of the second half of thetwentieth century, Lucien Bernot (1919–1993) appears as a quiteoriginal figure.1 Like certain other anthropologists we are dealing within the present volume, he did not occupy the front stage and was notwell-known abroad, except perhaps by Western colleagues working onhis particular geographical areas of research, namely the ChittagongHills, Burma and Southeast Asia. He has nevertheless been veryinfluential in France, through his books and his teaching at the ÉcolePratique des Hautes Études (sixth section), which in 1975 became theÉcole des Hautes Études en Sciences Sociales (Paris). In fact, Bernottrained a whole generation of young anthropologists between 1960until 1980, mainly on Asiatic studies but also a few on Europeanethnography. I myself was a doctoral student of Bernot’s for both myPh.D. dissertations.2 From 1972 till 1986 I was in very close contactwith him and learned a great deal about anthropology from him. Histeaching was for me a kind of revelation, an immersion into the realityof fieldwork and the craft of ethnography, as well as a move fromconcept to action. Throughout his career, he defended a vision ofanthropology based on intensive fieldwork and knowledge of thevernacular languages. He campaigned in favour of an ethnographyoriented towards facts, realities and detailed observations. As Clifford remarks (1988: 139), this ethnographic vision, which isoften associated with the Musée de l’Homme, formerly located in thePalais du Trocadéro (Paris), is rooted in a long French tradition. It dates

198 Gérard Toffin

back to the very origins of French anthropology, through figures likePaul Rivet, Marcel Griaule, Michel Leiris and, as far as Asia isconcerned, ethnographers belonging to the École Française d’Extrême-Orient (Charles Archaimbault, Guy Moréchand and Jean Boulbet, toname but a few). After the mid-1950s, this approach became in manyways linked with the name of André Leroi-Gourhan (1911–1986),prominent archaeologist and ethnographer specialising in technology,who was a professor at the Collège de France from 1969 until 1980.For a number of French anthropologists of the second half of thetwentieth century, Leroi-Gourhan embodied an alternative to thestructuralism of Lévi-Strauss, which is often considered too distant fromethnographic realities. Yet although he was undoubtedly an eminentscholar and great spirit, Leroi-Gourhan cannot be reduced to just oneschool. In the course he taught in the 1960s at the Sorbonne, Iremember that he often claimed to be closer to the culturalanthropology of the United States than to the French or British schoolsof anthropology, a view of himself corroborated by a carefulexamination of his works.

Although this ethnographic and empirical perspective has neverbeen totally rejected by anthropologists who are more oriented towardstheory, and although its practitioners generally maintained closerelations with the latter, it retains its vitality as a stream ofanthropology that is very different in its style and intellectual concernsfrom the Lévi-Straussian structural school (Toffin 1995).

Figure 8.1. LucienBernot, on the occasionof his being honouredwith a Festschrift at theÉcole des Hautes Étudesen Sciences Sociales,Paris, 1987. To the rightin the background,Claude Lévi-Strauss (left)is talking to Marc Augé.Taken by J.C. Vaysse.Collection BernardKoechlin.

The art and craft of ethnography 199

Life and works

Lucien Bernot was born in Gien (Loiret department), central France, in1919.3 Unlike most French anthropologists of the twentieth century, hecame from a very modest family with strong rural roots. His father, whotaught himself to read and write, was a worker in the PTT state telephonecompany. Throughout his life he cultivated a small garden and somevines, as was the custom in the region. Lucien became a typographer andfollowed this job from 1935 (when he was fifteen years old) until 1943 inseveral workshops all over France, including the Imprimerie Nationalein Paris. But he soon became interested in other cultures, starting withAsiatic writing systems, which were different from the graphic signs hewas using in his job. From 1944 until 1947, he learned Chinese, thenTibetan, at the École Nationale des Langues Orientales Vivantes in Paris.4

This degree gave him an equivalent to the baccalauréat and enabled himto enter university. In 1946 in the Musée de l’Homme he met André Leroi-Gourhan, who found him a post in the Asiatic department with a smallsalary. Bernot was in charge of filling out card indexes for the varioustools and items that the museum acquired.

In 1949, Claude Lévi-Strauss raised some funds from UNESCO forthis young student in ethnology and oriental languages. Bernot wassent by the historian Lucien Febvre to a village located on the borderbetween Normandy and Picardy, northwest of Paris, with RenéBlancard, a psycho-sociologist (see Lévi-Strauss 1995, Zonabend1995). This UNESCO project was very much under the influence of the‘Culture and Personality’ American school. Its emphasis was on theinteractions between individuals and their society, as well as on thecultural models affecting children at an early age. The project was alsointended to highlight contacts between cultures. A border villagebetween two geo-cultural regions was an excellent choice in thatconnection. The two young men spent eight months there and in 1953published a book, entitled Nouville, an invented word derived from nous,‘we’, and ville, ‘town’. This monographic study, which was republishedin 1995, was the first to focus on a French village from a genuinelyanthropological perspective. The earlier studies on French communitiesby Robert Hertz (1913) on ‘Saint Besse’ in the Aosta valley, and by LouisDumont (1951) on La Tarasque in Provence were mostly concerned withlocal cults. As for the study by Laurence Wylie on Roussillon (Village inthe Vaucluse), this only dates from 1957.

Between 1951 and 1952, Bernot undertook fieldwork with his wife,Denise, in the Chittagong Hills, in what was then East Pakistan (todayBangladesh). He was advised to do research in this area by Lévi-Strauss,who thus for a second time directed the young ethnographer’s career in

200 Gérard Toffin

a decisive manner. Lévi-Strauss had actually just returned himself fromthe Chittagong Hills, as he narrates in Tristes tropiques (1958). Bernotjoined CNRS in 1954 and returned a second time to East Pakistan,thanks to a scholarship from the Ministry of Foreign Affairs (October1959–February 1960). He worked mainly on a Tibeto-Burman-speaking population, the Marma, who originally came from Burma. Hesubmitted his doctorat d’état on this Buddhist group in 1967 with a full-length study under the supervision of André Leroi-Gourhan, the thesisbeing published the same year in a two-volume monograph, Les paysansarakanais du Pakistan oriental (1967a). In 1967 another book dealingwith another Tibeto-Burman-speaking ethnic group, Les Cak:Contribution à l’étude ethnographique d’une population de langue loi (1967b),whom Bernot had studied in 1960, also appeared. The Cak are very closelinguistically to the Kadu, who live on the upper Mu River, Burma.

By then Bernot had already joined the École Pratique des HautesEtudes (in 1964) as a Directeur d’Études and had started his importantteaching to students in ethnology (or anthropologie sociale, theexpression made more and more familiar in France under the influenceof Louis Dumont and Claude Lévi-Strauss). His courses focused ontechnology, one of his favourite themes, but also on other topics, suchas kinship and ethnobotany. This was before the anthropologydepartment at Nanterre University (Paris X) had really got going: it onlystarted in 1967 and became really effective in the early 1970s. Thesixth section of the École Pratique des Hautes Études was therefore themain place for training advanced students in anthropology.

Bernot participated with Georges Condominas and André-GeorgesHaudricourt in the foundation of the Centre de Documentation et deRecherches sur l’Asie du Sud-Est et le Monde Indonésien (Cedrasemi).He played an active role in this Centre, stimulating a programme on alinguistic atlas of Tibeto-Burman languages to which I shall return. Hewent back several times to Asia, especially during the summer vacations,which in Southeast Asia corresponded with the rainy season. He workedmainly in Burma with his wife, who became Professor of Burmese at theInstitut National des Langues et des Civilisations Orientales. He studieda number of Burmese craftsmen all around Mandalay and becameinterested in another ethnic group, the Intha, who lived on the edges ofthe Inle Lake and practised agriculture on floating islands.

Although he was in some ways marginal to the dominant Frenchacademic establishment, Bernot was appointed professor in theprestigious Collège de France in 1979, where he mainly taught Burmeseand Assamese ethnography. He entitled his chair the ‘Sociographie del’Asie du Sud-Est’, resuscitating a term current at the beginning of thetwentieth century (e.g. the chair of Alfred Le Chatelier, then of Louis


Massignon, in the Collège de France was in the ‘Sociologie etSociographie du monde musulman’). Bernot gives the followingdefinition of this term: ‘To observe the facts and isolate those thatcharacterize the society being studied without losing sight of theircontext, to search the disciplines that relate to this context for the meansto clarify these facts and, through this, to enrich and deepenobservations and descriptions: these are the steps that, for me too, theterm “sociography” covers’ (Bernot 2000 [1978]: 28).5 He wanted tobring ethnography and history together in the Boasian manner, and tostudy literate as well as preliterate societies (Leçon inaugurale au Collègede France 2000 [1979]: 500). Yet, a few years later, he was no longersure that sociography was the right word (personal communication).

He died in 1997, at the age of 72, just after mowing his lawn in thegarden of his house in Brantes, in southern France. Thus we lost an eruditescholar, a great connoisseur of rural life, a bon vivant, a very warm personwith an open nature, expansive, but also modest and quite untypical inmany other respects in the intellectual milieu of French anthropology.

Monographic studies and attention to detail

Of all the professional ethnographers we are dealing with in thisvolume, perhaps nobody placed more emphasis on the importance ofmonographic study than Bernot. Even though he occasionally defendedhimself from the charge of relying exclusively on this method (‘Avant-Propos’ to Les Paysans arakanais), most of his books have a monographicstructure focused either on a village or a minority group. Nouville, forexample, is the study of a specific locality inhabited by 594 individualsworking in either agricultural activities or two local glassmanufacturers. Les Paysans arakanais deals specifically with a particularethnic group (about 100,000 in number in 1960), the Marma. Thesame can be said for his books on the Cak (about 3,000 individuals) andhis article devoted to the Intha published in the Journal d’agronomietropicale et de botanique appliquée (2000 [Bernot and Bruneau 1972]).

Bernot was fully aware of the limitations of the monographicapproach, especially with regard to a village. As he explained in one ofhis best articles, ‘Les Plein-Vent’,6 published in Ethnos (1975a), a villagedoes not live in total isolation from the external world, but has links withsurrounding villages and overlapping economic or social interests; italso has a relationship with cities as well as various markets, on whichit may depend. Bernot also acknowledges that a particular village isfrequently visited by people from different localities and that marriagerules often unite different communities through strong links. Quoting

202 Gérard Toffin

the French Marxist historian Charles Parain, he stressed that a regionalapproach is sometimes a useful complement to village study. Likewise,he mentioned Louis Dumont, his colleague at the École Pratique desHautes Études, who argued that, in the Indian context, the supposedlyself-sufficient multi-caste village is partly irrelevant from thesociological point of view (Dumont 1966).

Nonetheless, Bernot defended the monographic approach, which heconsidered a good point of departure. Admittedly, the village frameworkis more appropriate for studying localities inhabited by a single ethnicgroup, such as the ones he worked in, than those consisting of multi-caste communities. But more fundamentally, he was of the opinion thatpeasant life and rural social units are best considered at the local level,because farmers are attached in a very intimate manner to a specificterritory. Bernot even suggests that the size of the ideal village to beinvestigated intensively by an anthropologist staying alonecontinuously for a year is 200 to 300 persons, including children (2000[1973]: 325).7 It is only at this micro-level that the professionalfieldworker can be on close terms with the local population and get toknow everybody. It is also at this level that s/he can realise the strongcontrast which opposes ‘us’ to ‘others’ in all rural worlds, a theme whichoccurs repeatedly in his works.

Bernot devoted a whole article on how to conduct a census in a smallterritorial unit without official census data or a civil register. In thiswork, the author of Nouville is stimulated by a concern to collect datawith as much rigour as possible (‘Le recensement d’un village’, 2000[1973]), but unfortunately he ignores sampling problems and therepresentativeness of the materials collected. In his view, the census ofa village of forty houses normally takes about forty hours. He suggeststhat the investigator should collect genealogies for the study of thesocial organisation at the same time. He also gives some advice on thepossible use of photography to provide better identification of thepopulation during an enquiry and to give each of them the pleasure ofreceiving their own photograph:

The work of taking a census can be facilitated by means of photographs.One can ask a certain number of individuals to pose (it does not matterwhom so long as they live in the village). From these photographs one hasenlargements made, as many as the people in the photographs, plus two.When the photographs return from town, two are kept and [the others]distributed. On one of the retained photographs is indicated, using tracingpaper, the name of each individual and his or her number. The otherphotograph is cut up and the corresponding face glued on to the individualpage [for each person]. (ibid.: 327–28)


This article may be considered as a late but nonetheless usefuldevelopment of the innovative works of W.H.R. Rivers (1900) on theuse of the genealogical method to study kinship systems and socialorganisation. By and large, indeed, Bernot’s fieldwork method was moreor less in accordance with the methods laid down by Rivers in the fourthedition of Notes and Queries (Royal Anthropological Institute 1912):8

the necessity of learning the local language, the search for spontaneousinformants, the concern to create as great an empathy as possiblebetween the enquirer and the population being studied (Stocking1995). This method of intensive enquiry, with both direct andparticipant observation, which Malinowski developed so wellsubsequently, contrasts strongly with the practice of the first tribalethnographers in India and Burma, British for the most part, whogathered their information mainly while ‘touring’ or ‘surveying’ wholedistricts, spending a few nights camping in successive villages, and interviewing informants in isolation rather than in any detailed context.9

Ultimately, Bernot validates the monographic approach in these terms:

Monograph, rural sociology within the monograph, structural study – noone of these forms of study can exclude the two others. But the excellentmonograph can always, sooner or later, lend itself to structural analysis,while a structural study, however excellent, can never permit themonograph to be reconstructed. (1973: 245)

In other words, monographs are credited with a more permanent lifethan theoretical studies, which often fall under the influence oftransitory moods, and whose life expectancy is much more limited. Agood monograph will always be useful to future generations, being aninventory, a document made at a certain point in time and space, towhich it can always be referred.

It is instructive to examine closely the synopses and divisions of hisworks into chapters. For instance, Nouville falls roughly into three parts.The first four chapters are devoted to the geographical, historical,demographic and economic settings. The second part includes fourchapters dealing with the life-cycle (infancy, adolescence, marriage andfamily life, and old age). The third and last part focuses on theinhabitants of the village and their relationships with outsiders(communal life, values, information, space and time, attitudes towardoutsiders). The contents correspond to the expectations of those whoconceived the UNESCO project. Among the various appendixes, onepresents the result of the psychological tests that Blancard performed ona selected sample of the population. Included is an interesting chapter

204 Gérard Toffin

(Ch. 5), written in fact by Blancard, on childhood and education, asubject too often neglected by anthropologists even today.

Let us now consider Les paysans arakanais, Bernot’s masterpiece,which contains an impressive body of ethnographic materials. Part 1looks at the narratives of a Hinduised group through historical sources,mainly European (Portuguese, British, Dutch). Part 2, ‘Le monde végétal’,includes chapters on the elementary principles of Marma technology(Ch. 3), slash and burn agriculture (Ch. 4), occasional agriculture(ploughing and gardening) (Ch. 5), food (Ch. 6), bamboo and cotton (Ch.7) and the house (Ch. 8). The third part is devoted to society – men andwomen – with chapters focusing on the economy (Ch. 9), the life-cycle(Ch. 10), the village and its environment (Ch. 11) and kinship (Ch. 12).The conclusion is devoted to the social division of labour.

As a whole, such lists of contents are very classical, cover mostaspects of society and culture, and adopt a holistic viewpoint, exceptfor the two important domains of religion and politics, which are nevertreated as such, but only in relation to the facts of social organisationand material life. In the conclusion to Nouville, the two authors explainthat their aim has been to throw light on the links between all aspectsof the life of a social group: psychological, sociological, historical andeconomic. These divisions into chapters can be considered typical ofthe ethnographic monographs prepared in the Musée de l’Homme (andthe École Française d’Extrême-Orient too) at that time.

But what marks Bernot out as original and distinctive is his scientificattention to detail. Undoubtedly his powers of observation were acute,his interest in the concrete passionate. He himself gives someillustrations of his ethnographic method:

For example, let us cite the different ways of lighting a fire according towhether it is done by a man or a woman, the places in the house classifiedaccording to whether they are occupied by the father, the mother or thechildren, the differences between the birth of a boy and that of a girl,between the cremation of a deceased male or female, the smell of a mother’sclothing in infancy, what happens to the hair of a young monk, the yamsthat wind themselves round in the opposite direction, etc. (1967a: 15)

What he thinks more important is the noting of ‘little details’ or ‘micro-details’, ‘hidden or not in the discourse of the informants’, inaccordance with Mauss’s instructions in his Manuel d’ethnographie(2007 [1947]: 7). These details are in fact ‘crucial, primordial’, inBernot’s words. He presents his work as an attempt, ‘starting with alittle, to understand the whole, that is, to make the whole understood:that is, to make the Marma ethnic group understood’ (1967a: 14). Theaim is to leave as little as possible behind the scenes and to list all the


details, so that the underlying determinants of the social and culturalrules, the patterns of a group, can be highlighted.

The result is a somewhat puzzling way of writing anthropology, onethat recalls Franz Boas’s works, as well as the volumes of theEncyclopédie agricole of the Éditions J.B. Baillière10 or the agronomicalworks of the Maison rustique publishing house.11 These two types ofwork, which are a mine of inexhaustible information on the history ofagricultural techniques in France, were very well known in the Frenchprovinces in the nineteenth and twentieth centuries, though more bythe rural elites in the strict sense than the minor peasants. This type ofwriting accumulates a tremendous quantity of detail on the socialgroup studied, providing a sort of ‘naturalistic’ ethnography devotedto the daily routines, the chores of ordinary activities, repeatedly citingthe terms of one’s informants and interlocutors. Thus when Bernot isstudying agricultural techniques, he gives a lengthy description of howHari Phru ploughs his two-acre plot of land before sowing wheat from8.30 in the morning until 6.15 in the evening. Similarly, when he dealswith marriage, he gives a description of a particular wedding, at acertain time and place. The farmers and their families are presentedpersonally in the context, without any literary effect, and are placed atthe heart of the ethnographic description. This particularistic approachalso characterises chapters on rituals. Interestingly, the ethnographicdata are often presented in a rough form, not very different, in allprobability, from his field notes. Besides, the sentences in his texts,intricate and often quite long, are not written to be elegant but to reflectthe complexity of the realities observed, the very opposite of anyjournalistic or literary type of writing. This direct style conveys a senseof reality that is expressed in the uniqueness of individuals and events.

The ethnography of techniques and of the botanical world

In his teaching as well as in his writings, Bernot stressed the importanceof the study of techniques and of material culture for a thoroughunderstanding of pre-modern societies and cultures. After beingappointed Directeur d’Études in the École Pratique des Hautes Études,he focused his seminar on ‘Les techniques de consommation (habitat,alimentation, vêtement) en Asie du Sud-Est’. His aim was to study thelinks between these techniques and the society using them and to bringout the articulations between the two. These aspects became more andmore important as he advanced in his career. He identified amethodological interest in starting with these studies: at the beginningof his work, the ethnographer often does not understand enough of the

language of the unfamiliar culture in which he has chosen to settle tocarry out research on more abstract subjects.

In advocating this method, he was inspired by his doctoralsupervisor, André Leroi-Gourhan, whose main achievements in thefield of anthropology were to reflect on technology and on theclassification of pre-industrial techniques from all over the world,taking as his main parameter the various technical operations on rawmaterials (hammering, cutting, scraping, shearing, moulding, etc.).Bernot was also influenced by André-Georges Haudricourt, a colleaguewith whom he was in close contact throughout his professional life andwith whom he shared a common vision of anthropology, or perhapsmore broadly, a conception of research in the social sciences andhumanities. Bernot was also a steady reader of the seven volumes ofJoseph Needham’s Science and civilisation in China, a series published byCambridge University Press from 1954 to 1985. This great work isconceived as a history of science, technology and medicine in China,seen in its fullest social and intellectual context. Bernot used to takeexamples from these books to illustrate his teaching.

His concern for material culture was first of all motivated by his feelingthat too much attention had been paid to religion and symbolism by mostanthropologists, especially French ones, and that the ordinary means ofsubsistence had been neglected in the majority of studies. In fact, one ofBernot’s main lessons was that rural communities in either the hills orthe plains were made up of peasant agriculturists, whose mainpreoccupation was to produce food for their own consumption, provideshelter, exchange and carry goods, wear clothes, produce tools, and in amore general way adapt themselves to a somewhat hostile environment.

In a lecture delivered in the Maison franco-japonaise (Kyoto) in1974, Bernot argued that a technique has to be studied from threedifferent aspects: historical, geographical (including the influence ofsoil, fauna and flora) and how it is used. In this last respect, he proposeda useful functional classification of techniques into three parts:production (how tools and objects are acquired or made), distribution(how they are given away, exchanged or stored), and consumption(how they are used). Significantly, in each of these techniques, he placedthe emphasis on gestures (the ‘techniques of the body’ of Marcel Mauss,1936), and on the movements of the hand, the feet and of the wholebody in daily work activities. Studying fieldwork, he recommendedasking the following questions for each tool and object: ‘What? Where?When? How? How many? By whom? Why?’ (2000: 270).

However, compared to Haudricourt and Leroi-Gourhan, Bernot’sstudies of material culture pay less attention to classification proper. Inaccordance with the museological emphasis of the Musée de l’Homme,

206 Gérard Toffin

he took seriously every ethnographic object as a witness of the societybeing studied. Yet, on the whole he was more interested in the context,the relationship between various techniques and their relationshipswith social facts. Religious ideas interested him mostly when they couldbe connected with material life. Although he did not theorise the topic,he argued that technology is socially and culturally constructed. Thisappears clearly in his research on the house among the Marma andCak. For instance, among many other topics, he considered suchmatters as the positions allocated to the different family members in thedomestic space, the social and religious uses of different rooms(whether there was partition), the division into female and males areas,the values attributed to each part of the dwelling place, the economicaspects of the different divisions of the house, the inventory of all theobjects kept in it, how they are kept, what happens in case of their lossand the psychological aspect of the house as a refuge. Very often, allthese indicators vary from one ethnic group to another. More broadly,he considered the house ‘as the most useful, if not the most preciousobject of all ethnographic study’ (2000 [1982]: 142). In other words,for Bernot techniques were a key to entering into the social world, todeciphering cultural categories, modes of thought and local values.With food or habitat, he was confronted not only with technology, butalso with economy, religion, family life, law and so on. My own study ofthe anthropology of space among the Newar and Tamang of Nepal isgreatly indebted to these ideas (e.g. Toffin 1994).

The similarity with Mauss’s Manuel d’ethnographie (2007 [1947]) isparticularly evident here. In his courses delivered at the Institutd’Ethnologie in the University of Paris from 1920 to 1939, under thetitle ‘Instructions d’ethnographie descriptive’, Mauss actually devoteda great deal of space to technology. For the great master ofanthropological studies of this period, the study of clothing, the house,pottery and fishing provided a way of addressing problems of economicsor social morphology, of religious or legal anthropology. His teachingprovided a vivid introduction to the study of societies populating theFrench colonies between the wars and ‘societies at the same stage’ (ibid.:5). In this work, from considering the sandal as footwear Mauss passesto African secret societies, from agricultural works in Oceania toPolynesian chiefs, from boat technology to the chants of rowers.‘Learning to observe’ is specified as the goal from the first sentence ofthe Manuel (ibid.). Although Bernot only came to know Mauss towardsthe end of the latter’s life, one finds the same inspiration in his owncourse. Indeed, his teaching seems to have come straight out of theManuel. It is this global vision of anthropology that the title of thevolume published in his honour in 1987 – De la voûte céleste au terroir,


du terrain au foyer: mosaïque sociographique – with contributions fromsixty-five anthropologists, tries to capture (see Koechlin et al. 1987).

Closely linked to the study of material culture for Bernot was theexamination of plants and the botanical world. In this respect too hewas probably influenced by his long-term friend Haudricourt, anagronomist himself and a specialist in cultivated plants. Together theyedited the Journal d’agronomie tropicale et de botanique appliquée, a well-known scholarly journal of ethno-botany published by the Muséumd’Histoire Naturelle in Paris since 1923. Bernot collected from theMarma a herbarium of about 160 vegetables, giving the names anduses of ninety-two edible plants, either wild or domesticated. Suchbotanical variety is nothing exceptional in Southeast Asia, where forestpeople often use more than four hundred species of plants, all named,as food. The Hanunoo of Mindoro described by Harold C. Conklin(1975), an important reference in Bernot’s works, distinguish morethan 1,600 different plant types, reaching the astounding number of430 cultivars. The samples collected in the field by Bernot were given tothe Muséum d’Histoire Naturelle and identified by Haudricourt and J.E.Vidal. For most of these plants, the author of Les paysans arakanais alsodocumented how they were cooked and what they were eaten with.

Bernot developed a strong interest in the agriculture of these self-contained societies, which was organised mainly to fulfil the subsistenceneeds of its people rather than to produce a marketable surplus. On theone hand, he dealt with swidden agriculture, which was practised by theMarma on the peripheral land of their territory, as in a great number ofother Asian rural groups. He even gave a yearly course on that topic inthe Collège de France. On the other hand, he spoke and wrote a greatdeal about rice cultivation, either wet or dry, which is particularlyimportant in Southeast and South Asia, being the staple cereal of anumber of these countries. His rich article on rice cultivators, includedin a manual published by Armand Colin in 1975 (1975b), is a first-ratestudy, embracing the history of the plant Oryza, its origins and varieties,wild and cultivated, the various ways of cultivating it, the vocabularyof its different organic parts, the associated techniques of irrigation, theharvest tools (knives and sickles), the processing of the grain, how tomeasure and store it and so on.

All these researches played a great role in the development ofethnobotany in France, stimulating many of Bernot’s students to takean interest in the complex relationships between humans andvegetables, as well as to collect data on the names of plants from thefour corners of the world. It is from these seminars on ethnobotany andethnozoology, in which Haudricourt, Jacques Barrau and ClaudineFriedberg also participated – held in the delightfully antiquated rooms

208 Gérard Toffin

of the Muséum d’Histoire Naturelle, among specimens of snakespreserved in jars of formaldehyde and various stuffed animals – that awhole series of researches on indigenous systems of classification andlocal taxonomies was born.

The taste for words

One of the outstanding characteristics of Bernot’s ethnography is itsinterest for linguistics, more specifically for ethno-linguistics. Theseconsiderations appeared very early, at the time of the Nouville study. Ina lecture delivered in 1949 entitled ‘Les confins picards de la Normandied’après quelques considérations d’histoire économique et sociale’(2000: 337–8), he raised the question of the significance of localtoponyms and village names ending with ville, mesnil court, bec, budh orboeuf between the departments of Seine-Maritime (Normandie) andSomme. He linked these suffixes with the remote history of the regionand the influence of the various groups that had migrated throughNormandy or settled there (mainly Frankish, Roman, Germanic andScandinavian). Thus ville denotes Roman influence, boeuf and be pointto Scandinavian origins and so on. Each of these groups has thus leftbehind certain words that are still alive today.

The link between ethnography and linguistics is also prevalent in hisresearch on Southeast Asia. With his wife, who specialised in linguistics,in 1958, he published a short book (Bernot and Bernot 1958) on theTibeto-Burman language of the Khyang of the Chittagong Hills, a Chingroup. The book contains a general presentation of the Khyang anddata on their clan organisation, kinship terminology and technology,but at its core is an account of Khyang phonology and vocabulary.Some of the articles he wrote during the same period concern mainlythe classification of the Tibeto-Burman language groups in the area,taking the work of Robert Shafer (1967) as a guide. Similarly, the paperhe published in the volume offered to G.H. Luce, a leading epigraphistand historian of ancient Burma, is entitled ‘Éléments de vocabulaireCak recueilli dans le Pakistan Oriental’ (2000 [1966]).

In addition, the team he chaired within his laboratory (Cedrasemi;see above), with the aim of drawing up an ethno-linguistic atlas ofSoutheast Asia, was a joint research project on the mapping of differentword series. The words were either collected in the field or taken fromthe published documentation of the time. The main linguistic families ofthe area – Tibeto-Burman (on that group alone, 230 languages wereselected), Austro-Asiatic, Mon-Khmer, Thai and so on – were all takeninto account. Such tremendous work supposed first of all the precise


localisation of ethnic groups and languages on a map. The western partof Burma even included, at least partly, the Nepalese Himalayas, whichwas filled mainly using data collected by Brian Hodgson, a former BritishResident in Kathmandu during the nineteenth century. Through thismethod, Bernot and his colleagues sought to throw light on the historyof the local populations and on migrations. The team started by mappingthree words: ‘dog’, ‘teeth’ and ‘salt’, a list afterward extended to 21, thento 80 words, though it is unclear to me how they were chosen. The maps(size 55 x 70 cm, then reduced to 42 x 59 cm) were printed off by thecartographic laboratory of the École Pratique des Hautes Études in smallnumbers for preliminary research use only. Bernot also thought ofmapping information on rice cultivation – wet and dry rice, modes ofpreparing ricefields, use of nurseries, techniques of harvesting and soon – but this project remains largely unrealised. The models for theseundertakings were the various linguistic atlases published by linguistson French regions,12 such as that on the Massif Central published in fourvolumes in 1959–63 by Pierre Nauton, one of Bernot’s favouritesbecause of its richness and its numerous ethnographic aspects. The finalwork of the Southeast Asia ethno-linguistic atlas was published partly inthe journal ASEMI (1972, III, 4), without much analysis or comment.On the whole, the atlas remains mainly a research project.

What chiefly interested Bernot in the words taken from differentlanguages was their own identity – which was jealously guarded byeach ethnic group – their links with the social and material worlds, theirmigrations and the testimony they reveal of a disappeared past. He likedto trace a word’s history, how it travelled over various geographicalareas in the course of time, within a diffusionist framework. In a paperpresented to a symposium in the Sorbonne in 1986 on the Routes d’Asie(2000 [1988]), he mentioned the different plants which arrived inSoutheast Asia from elsewhere: maize (from America), sorghum (fromAfrica), nuts, yams, varieties of beans. Some of these plants had beenintroduced to China through Southeast Asia. He gave the Chinesenames borne by these plants today and described how the words hadbeen formed. He noted in passing that the introduction of new plants inAsia has not acted to the detriment of local ones, as has happened inEurope. In a similar vein, he noted that certain spices came to SoutheastAsia from India together with their original names, such as pepper, Pipernigrum L., whose Sanskrit name, marica, has been adopted by somelanguages of Asia for either pepper or pimento. Then taking theexample of the plough, Bernot stressed that, even though the name forit in Southeast Asia may have come from Indian languages, itsconstruction has become highly differentiated today, from northern tosouthern regions: North Burma, North Laos and North Vietnam have

210 Gérard Toffin

adopted the Chinese plough, whereas the southern parts of thosecountries still used the Indian type. Through all these notations, theresearcher can point out certain ancient links and speculate about thepast influence of one group over another.

Bernot’s ethnographic descriptions are filled with the indigenousexpressions that are used to speak about different objects. He was awareof the danger of using preconceived European categories and wanted tounderstand each culture in its own terms, from the inside. Similarly, itis worth mentioning his interest in folk classifications. One of his mainguides in this respect was again Conklin, whose work on the Hanunoo(1975) is still a key reference today. It is unfortunate that the neworientations in linguistics, with their shift towards cognition (amongother things), have made such approaches if not marginal then at leastless valued than formerly.

Conclusions: achievements and limitations

Besides the attention to detail and the descriptive aspect of his works,Bernot emphasised what he considered a ‘global’ conception ofethnography. As we have already noted, this global conception isreflected in the synopses of his main books, the most notable exceptionsbeing firstly, religion – which is considered only in passing, never as awhole – and secondly, politics and power, whose role is rarely taken intoaccount at all. These lacunae are, of course, important and represent asignificant difference between Bernot’s ethnography and that of theBritish school of anthropology. But by and large, Nouville and Lespaysans arakanais are characterised by their holistic viewpoints, aholism, however, that is not sociological, properly speaking, butprincipally methodological. These works aim to draw a picture of awhole society or a whole culture. Bernot himself expresses his concernwith this aspect in the preface to his two volumes on Les paysansarakanais, when he explains that one of his crucial aims was ‘to attemptto acquire a view of the whole of Marma culture’ (1967a: 13), the othergoal being ‘the search for detail’ (ibid.: 15). The two approaches arepresented as complementary, each requiring and building on the other.

Bernot mentioned explicitly the influence of Marcel Mauss (‘uneapproche maussienne’) in this respect (ibid.: 10), in that he applied theadjective total, invented by Mauss to characterise pre-modern society,to the individual in preliterate and traditional societies: ‘Having beensomething of a sociologist in studying a French village, and anethnographer in studying several villages in East Bengal, I believe I canaffirm that, in both areas, I have encountered “total” men’ (2000


[1975a]: 235). In fact, it is rarely remembered today that Mauss usedthe expression ‘total’ not only to designate social institutions but alsohumans themselves. In a lecture delivered in 1924 to the SociétéFrançaise de Psychologie, he wrote:

The average man of our days – and this is true above all of women – andalmost all the men of archaic or backward societies, is a ‘total man’: he isaffected in all his being by the least of his perceptions or by the least mentalshock (Mauss 1979 [1950]: 28)

[And further:] The study of the complete man is among the most urgent ofthese studies we would ask you to make. […] It is this man, this indivisible butmeasurable but not dissectible being that we met in our moral, economicand demographic statistics. It is this man that we find in the history ofmasses and people, and of their practices, in the same way that historymeets him in the history of individuals. (ibid.: 26, emphasis in the original)

Bernot uses the expression ‘total men’ in a rather different sense,namely that agriculturists, men and women in the forest and otherpeople in traditional societies maintain close contact with the soil, trees,plants, cattle and fields. This is the meaning of that elegant expression‘les Plein-vent’ (see also above), which Lucien Febvre (1942) uses,outside its original context of arboriculture, to denote ‘peasant’.

Bernot’s conception of ethnography is also distinguished by itsinterdisciplinarity. In his own words:

I am convinced that the human sciences demand of the researcher that hedevelops his faculties of observation … that he knows perfectly thebibliography of his subject, that he avoids becoming a prisoner of theory orof any single discipline, but uses one and the other as a means of guiding histhought, a programme which is not original, since it was already expressedin l’Année sociologique at the moment it was founded. (2000 [1978]: 28)

The parallel here with Mauss’s Manuel is also striking: ‘Ideally, anexpedition should not set out without its geologist, botanist andethnographers. […] So, set out as a group’ (2007 [1947]: 13). In fact,Bernot made a great many references in his works to the French schoolof geography of Albert Demangeon, Max Sorre, Roger Dion and so on, aswell as to the historical École des Annales, quoting frequently Marc Blochand Lucien Febvre. With all these scholars, he shares the same scientificvalues, the same trust in the progress of science and pure scholarship.

The diachronic dimension is also prevalent in all his works, fromethnography to linguistics, contrary again to the British school, whichwas sceptical of using historical facts during most of its classical

212 Gérard Toffin

functionalist period (ca. 1922–1970). When Bernot spoke of ArthurHocart, he mostly mentioned The Progress of Man (1933), anevolutionary work forgotten today, not the better known books, Kingsand Councillors (1936) or Les Castes (1938). In his studies, he alwaysused written sources if they were available, using them either todemonstrate the influence of the past on the present, as in his work onthe Marma (1967a), or else to derive the etymology of a word orindicate parallel usages in other languages. However, the informationdrawn from books was never used as a way of devaluing observation.Similarly, the concerns with space, with the influence of geography onhuman settlements and processes of diffusion from one country toanother, are recurrent themes in all his works. The author of Nouvillealso summoned up other disciplines: botany and linguistics, as alreadymentioned, but also architecture, which he found very useful in respectof the ethnographic study of vernacular houses and settlements. Myown ethno-architectural studies in Nepal (Toffin 1991, 1994) entirelyconfirms this recommendation.

Ultimately, Bernot regarded himself as a ruralist, that is, a specialistin peasant societies. His family background gave him an immediateunderstanding of agriculturists’ concerns and attitudes. His students,brought up for the most part in an urban culture, learned a great dealfrom him about these rural topics. Clearly Bernot was first of all anethnographer; but he was also eager to collaborate with otherdisciplines that were closely linked to his field of research and topromote a global approach to rural men and societies that was valid forFrance as much as for Asia. He wrote:

I would not hesitate even to write that, among these three examples ofpeasants [those of Nouville, East Bengal and Haiti, where Bernot spent abrief time13], I have felt less difference than that which exists between mostcitizens of Paris and the peasants of the Bresle valley from whom they areseparated by some 150 kilometres. I have no pretensions to a desire to definerural sociology, but wish to try and research the common denominatorsbetween types of countryside established in a landscape ‘which belongs tothe fields, to the countryside’ – the definition of ‘rural’ according to Littré.(2000 [1975a]: 235)

By writing this sentence, Bernot was obviously recalling his countryorigins. However, it is a contestable thesis he presents, given thedemonstration by a scholar like Haudricourt of the profound differencesthat oppose rice-cultivators to cereal-cultivators, or peasants in Westerncountries to tuber-planters in Oceania. Whatever the case may be, thisnotion explains the frequent comparisons he made between westernrural communities and Asian farmers in the east.


Some words should also be devoted to Bernot’s relationship with thechief authority on structuralism in France. Interestingly, throughouthis career there seems to have been a strange complicity, at least at firstsight, between Bernot and Claude Lévi-Strauss. Of course, the latterembodies a totally different trend in anthropology, oriented mainlytowards theory, universal rules of marriage, structures ofcommunication and the mind. He developed a highly intellectualschool, in which Bernot, like Haudricourt and Condominas, was not atease. Moreover, some of the structuralist followers of Lévi-Strausslooked down rather disparagingly on the research of someone likeBernot, which was so attached to realities, and they considered suchethnographic studies at best peripheral and subordinate. In the dualgeography of French anthropology of the second part of the twentiethcentury, the first dominated by Leroi-Gourhan, the second by Lévi-Strauss, Bernot evidently situated himself much closer to the former.Nevertheless, Lévi-Strauss supported Bernot at different key momentsin his career, even writing a preface to the re-publication of the Nouvillemonograph in 1995. This cannot be explained only in terms of tacticalor political reasons: from the very beginning, Lévi-Strauss wasimpressed by Bernot’s unconventionality and deep knowledge,immediately recognising his astonishing erudition in so many differentfields. But more broadly, unlike some of his imitators, the leader ofstructural anthropology has always defended ethnographic studies, thecollection of materials through fieldwork, without which, as he wrote,a more interpretative anthropology cannot exist. Everybody can recallthe considerable use Lévi-Strauss made of ethno-botanical data andfolk classifications in La pensée sauvage (1964).

What about Lucien Bernot today? Re-reading his monographsclosely, some limitations naturally emerge. The sociological dimensionsof the groups studied – the conflicts, the relations of power withincommunities, for instance – remain largely unexamined. Attempts to gobeyond particular descriptions and to propose cultural generalisationsare extremely rare. The style also sometimes sounds old-fashioned todayin some ways. In brief, Bernot’s main achievements are based on acombination of erudition and fieldwork: they are in no sense guided bythe anthropological theory of his time. Nevertheless, his monographsare models of their kind and exert a continuing influence,unfortunately far too limited and too much restricted to Frenchresearchers and students. His book with Blancard on Nouville (Bernotand Blancard 1953) is a landmark in the study of the anthropology ofFrance, breaking with the former folklorist approach, and for the firsttime applying anthropological methods and addressing concerns infavour in countries distant from the French field. Similarly, his works

214 Gérard Toffin

on Tibeto-Burman-speaking ethnic groups remain essentialintroductions to the material culture of Southeast Asia and are stillunrivalled, being among the best books to be prescribed for theeducation of professional fieldworkers. English translations of themwould undoubtedly be useful tools and would help extend theirreadership. Interestingly, Bernot never challenged his naturalistapproach to ethnography, nor did he ever raise epistemologicalquestions about the validity of the fieldwork experience or ofethnographic knowledge: he just explained straightforwardly how thematerial was collected, in which village, and on the basis of whichobservations. All signs of subjectivity are banned from his books.

However, Bernot’s achievement was undoubtedly to associate data-collecting with a good sense of humour, an immediate and friendlycontact with his interlocutors and informants. All this cannot be learntfrom books and is contrary to what happens in most other socialsciences and humanities, most of which deal exclusively with writtendocuments. The anthropologist, as is well known, works with livingpeople and has to rely on his or her own human qualities to establishpersonal contacts with informants and win their co-operation. In thisconnection, Bernot raised ethnographic fieldwork not only to arespected craft, a necessary discipline, but also to something of an art.

Notes

1. I would like to thank Robert Parkin, Anne de Sales and Geneviève Bédoucha for theircomments on an earlier version of this article.

2. The first, on Pyangaon village, was submitted in 1974 (EPHE, 6 ème section), thesecond (my doctorat d’état), entitled ‘Société et religion chez les Néwar du Népal’, in1982, Paris, École des Hautes Études en Sciences Sociales.

3. On Bernot’s life, see especially Thomas et al. (1987), Toffin (1995), D. Bernot (2004).4. The École Nationale des Langues Orientales Vivantes was transformed into the

Institut des Langues et Civilisations Orientales (INALCO) in 1971, following thestudent movement of 1968. It had originally been created during the FrenchRevolution, in 1795, on Lakanal’s initiative. At this period, it was called the ÉcoleSpeciale des Langues Orientales. Since the beginning of the twentieth century, thisÉcole has been known to successive generations of students as ‘Langues O’.

5. The quotations taken from Bernot’s articles and similar items are given hereaccording to the pagination in Bernot (2000), a collection of all his works publishedunder this form. The references to his books are given according to the pagination intheir original editions.

6. He took this word, which refers to arboriculture, especially of fruit trees in unshelteredlocations that are exposed on all sides to the wind, from Lucien Febvre’s book onRabelais (1942).

7. Interestingly, the population of Pyangaon, the first Newar village in which I carriedout fieldwork in the Kathmandu Valley, was 484 in 1971 (close to the 594


inhabitants of Nouville). I chose this village mainly because of its low population,more or less in accordance with Bernot’s recommendation.

8. One cannot help being struck by the considerable gap between the dates of theappearance of the two fieldwork manuals, Notes and Queries (first edition 1874) andMarcel Mauss’s Manuel d’ethnographie (2007 [1947]), which provided a broadlysimilar framework for investigation and type of questionnaire.

9. However, Bernot himself never actually spoke Marma or Cak fluently, nor evenBurmese, as he himself confessed in his introduction to Les Paysans arakanais (1967a:11); instead, he relied mainly on English-speaking informants and his wife. Nonethelesshe was always passionately interested in words, vocabularies and dictionaries.

10. The publishing house of J.B. Baillière, founded in 1818 in Rue Hautefeuille in Paris,has been one of the great scientific and medical publishers in France since thenineteenth century. It has reissued extracts from the Encyclopédie of Diderot andAlambert several times in different forms. Its collection Encyclopédie agricole covers awhole series of subjects concerning agricultural techniques, the care of animals,plants and forests, the use of manure, etc.

11. La Maison rustique, founded in 1836 in Paris, is a bookshop and publisher specialisingin agriculture, stock-raising, horticulture, hunting and fishing, and is still activetoday.

12. This linguistic atlas of France is different from the geographical atlas of Francepublished under the auspices of the National Committee of Geography from thebeginning of the twentieth century. Cf. Febvre (1962).

13. Before researching in East Pakistan, Bernot started a study in Haiti, thanks to a grantfrom UNESCO. He did not publish anything on this.

References

Bernot, D. 2004. Les collines de Chittagong, in M. Izard (ed.), Lévi-Strauss, Paris: Éditionsde l’Herne (‘Cahiers de l’Herne’).

Bernot, L. 1966. Éléments de vocabulaire Cak receuilli dans le Pakistan Oriental, in B.Shin, J. Boisselier and A.B. Griswold (eds), Essays offered to G.H. Luce, Artibus Asiae, vol.1: 67–91 [republished in L. Bernot, 2000, 389–414].

——— 1967a. Les Cak: contribution à l’étude ethnographique d’une population de langue loi,Paris: Éditions du CNRS.

——— 1967b. Les Paysans arakanais du Pakistan Oriental: l’histoire, le monde végétal etl’organisation sociale des réfugiés Marma (Mog), Paris: Mouton (2 vols).

——— 1972. Essai pour la présentation de la carte des langues, ASEMI, 3(4): 1–6[republished in L. Bernot 2000, 482–94].

——— 1973. Le recensement d’un village, in L’homme, hier et aujourd’hui: receuil d’étudesen homage à André Leroi-Gourhan, Paris: Cujas, 17–24 [republished in L. Bernot 2000,325–32].

——— 1975a. Les Plein-Vent, Ethnos 1975: 1–4, 73–90 [republished in L. Bernot 2000,235–47].

——— 1975b. Riziculteurs, in R. Creswell (ed.), Éléments d’ethnologie,1: huit terrains,Paris: Armand Colin (‘Collection U’) [republished in L. Bernot, 2000, 101–42].

——— 1978. Titres et travaux, Paris: Imprimerie Commerciale d’Anthony. [republishedin L. Bernot, 2000, 27–41].

——— 1979. Leçon inaugurale au Collège de France, Paris: Collège de France [republishedin L. Bernot 2000, 497–510].

216 Gérard Toffin

——— 1982. The house of swidden farmers as a special object for ethnological study, inK.G. Izikowitz and P. Sorensen (eds), The house in East and Southeast Asia:anthropological and architectural aspects, Richmond: Curzon Press, 35–40 [republishedin L. Bernot 2000, 143–48].

——— 1988. Transmissions des techniques et des produits entre la Chine et l’Asie duSud-Est, in Routes d’Asie: marchands et voyageurs, XVe–XVIIe siècle (Varia Turcica XII),Paris and Istanbul: Isis, 87–101 [republished in L. Bernot, 2000, 157–69].

——— 2000. Voyage dans les sciences humaines: qui sont les autres? Paris: PressesUniversitaires de la Sorbonne.

Bernot, L. and D. Bernot. 1958. Les Khyang des collines de Chittagong: matériaux pour l’étudelinguistique des Chin, Paris: Plon (Cahiers de l’Homme).

Bernot, L. and R. Blancard. 1953. Nouville: un village français, Paris: Institut d’Ethnologie.Bernot, L. and M. Bruneau. 1972. Une population lacustre: les Intha du lac Inlé (États

Shan du Sud, Birmanie), Journal d’Agriculture Tropicale et de Botanique Appliquée, 19:10–11, 410–41 [republished in L. Bernot, 2000, 63–100].

Clifford, J. 1988. The predicament of culture: twentieth-century ethnography, literature, andart, Cambridge, MA: Harvard University Press.

Conklin, H. 1975. Hanunoo agriculture: a report on an integral system of shifting cultivationin the Philippines, Yale: FAO Forestry Development Paper No. 12.

Dumont, L. 1951. La Tarasque, Paris: Gallimard.——— 1966. Homo hierarchicus, Paris: Gallimard.Febvre, L. 1942. Le problème de l’incroyance au XVIème siècle: la religion de Rabelais, Paris:

Albin Michel.——— 1962. Pour une histoire à part entier, Paris: Éditions de l’EHESS.Hertz, R. 1913. ‘Saint Besse’, Revue de l’Histoire des Religions, vol. 67.Hocart, A.M. 1933. The progress of man: a short survey of his evolution, his customs and his

works, London: Methuen.——— 1936. Kings and councillors: an essay in the comparative anatomy of human society,

Cairo: Paul Barbey.——— 1938. Les castes, Paris: Guethner (Annales du Musée Guimet, vol. 54).Koechlin, B., F. Sigaut, J. Thomas and G. Toffin (eds). 1987. De la voûte céleste au terroir, du

jardin au foyer: mosaïque sociographique, Paris: Éditions de l’EHESS.Lévi-Strauss, C. 1958, Tristes tropiques, Paris: Plon.——— 1964. La pensée sauvage, Paris: Plon.——— 1995. Pour la réédition de Nouville, in L. Bernot, Nouville, Paris: Archives

contemporaines.Mauss, M. 1936. Les techniques du corps, Journal de Psychologie, 33 (3–4).——— 1979 [1950]. Sociology and Psychology (tr. B. Brewster), London: Routledge &

Kegan Paul.——— 2007 [1947]. Manual of Ethnography (tr. D. Lussier), New York and Oxford:

Berghahn.Nauton, P. 1959–1963. Atlas linguistique et ethnographique du Massif Central, Paris:

Éditions du CNRS (4 vols).Needham, J. 1954–1985. Science and civilization in China, Cambridge: Cambridge

University Press.Rivers, W.H.R. 1900. A genealogical method of collecting social and vital statistics,

Journal of the Royal Anthropological Institute, 30: 74–82.Royal Anthropological Institute. 1912. Notes and queries on anthropology, 4th edition,

London: Routledge & Kegan Paul.Shafer, R. 1967. Introduction to Sino-Tibetan, Part 2, Wiesbaden: Otto Harrassowitz.Stocking, G.W. 1995. After Tylor: British social anthropology 1888–1951, Madison: The

University of Wisconsin Press.


Thomas, J. 1987. À Lucien Bernot, in B. Koechlin, F. Sigaut, J. Thomas and G. Toffin (eds),De la voûte céleste au terroir, du jardin au foyer: mosaïque sociographique, Paris: Éditionsde l’EHESS.

Toffin, G. 1994. Ecology and anthropology of traditional dwellings, Traditional dwellingsand settlements review, 5(2): 9–20.

——— 1995. Lucien Bernot (1919–1993), L’Homme, 133: 5–8.Toffin, G. (ed.). 1991. Man and his house in the Himalayas, Delhi: Sterling Publishers. Wylie, L. 1957. Village in the Vaucluse, Harvard: Harvard University Press.Zonabend, F. 1995. Nouville après Nouville, in L. Bernot, Nouville, Paris: Archives

contemporaines.

218 Gérard Toffin

Chapter 9

ANDRÉ-GEORGES HAUDRICOURT: A THOROUGH MATERIALIST

Alban Bensa1

Accessing the real

In an interview given to the Union Rationaliste in 1984, André-GeorgesHaudricourt summed up the initial context in which he learned to doresearch, and set the tone for the whole encounter: ‘I’m my parents’son, and my parents were city people who’d gone back to the countrybecause they didn’t like living in the city, having to say “Good morning,good evening,” etc. I obviously inherited some of that.’ Born in 1911and raised on his father’s farm in Picardy, where a good number of farmworkers were employed, André-Georges was apparently too sickly to goto school. His mother taught him to read, and he took correspondencecourses until the age of fifteen. He was then admitted to the Lycée SaintLouis in Paris, and three years later, in 1929, he passed the nationwidecompetitive examination for admission to ‘L’Agro’, the Institut Nationald’Agronomie. During his youth in Picardy, he acquired anextraordinary culture as an autodidact, observing the rural world andagrarian techniques in actual use, listening to the surrounding patois,initiating himself into a knowledge of foreign alphabets (Greek,Russian) by means of a stamp collection, and showing a passionateinterest in botany, history and biology, among other subjects. He thusacquired a free-ranging, acute perspective on human knowledge thatwas far removed from the conventions and disciplinary distinctions thatcharacterise ordinary schooling. His ability to rid observation of all

Figure 9.1. André-Georges Haudricourt: ‘le maître à la recherche de la petite bête ou lemaître dans l’exercice de ses fonctions’, June 1972. Fonds André-Georges Haudricourt/Archives IMEC.

preconceptions led him to the understanding that the only genuine‘fact’ was what could be known through the senses. Cleaving to the realin all its strangeness, his words and texts do have a certain abruptnessto them, somewhat like a whack of the baton a Zen master uses to wakehis disciples from their dreams.

Haudricourt’s anthropological thinking was thus always grounded intangible material traces of human activity, what is presented to oursensory experience: sight, hearing, smell, touch, taste. Palpating leaves,crushing them between his fingers to bring out or ‘express’ their odour;cupping his hand around his ear to apprehend better the sound of a voice,the ‘interesting phonemes’ it produced; scrutinising landscapes, buildingsand tools so as to identify spaces and the particular plants in them, theirmaterials, forms and functions – this extraordinary observerconcentrated all his attention on what can be grasped of the world as itappears to us. Being anchored in this way in the materiality of the real

220 Alban Bensa

presupposes an ability to clear out the mind. To remain in the empiricalworld and perceive effectively, a would-be observer needs to know how toexpel social, psychological and intellectual considerations that mightcloud or obstruct the full receptiveness required, namely the conventionsof human relations (politenesses, phatic or narrative speech, compulsorybanalities, etc.) and the ordinary flow of psychic life, in which ideas,images, and feelings screen out the world of the visible, the audible, thesensory. It is only by fully exposing ourselves to that world, returning towhat is there, right in front of us, that we can grasp and analyse it.

Though Haudricourt’s close-up view may be hyperrealist, it is notcontemplative, but rather active and critical. He worked to discern theconnections, schemata and logical relations that are common toactions and representations operative in objects, words, and physicalgestures and practices. His fascination for the material world in thebroad sense constantly pushed him to wonder about the origin andhistory of practices. Being anchored in the materiality of things andliving beings caused him to develop a general attitude of caustic doubtabout the supposed autonomy of representations and ideas. ForHaudricourt, it is clear that the ideas which stir and agitate humanbeings have their source in the experiences that link us to the world.The origin of what seem the most cerebral ideologies can therefore, inhis view, be related to habits acquired in hunting or domesticatinganimals and plants; that is, through humans’ interaction with theecological and historical constraints (climatic, geological, botanical,zoological; migrations, confrontations, ways of doing) that haveaffected diverse peoples, groups and populations.

For Haudricourt, focusing on the natural world and techniques workslike a kind of stripping agent, making it possible to see and correct ourmisguided wandering in the supposedly pure sphere of ideas and feelings.Consistent with this view, he never explains texts by other texts or entersthe history of ideas by way of ideas, but always refers ideas and affects tothe social, linguistic and above all sensory and technical experience that,in his understanding, enabled them to emerge. The domain of abstractideas (‘abstracted from what?’, he might ask) is only comprehensible if itis not dissociated from the hold of the physical and social worlds of oursenses, themselves the very tools of our first investigatory experience.

It would be wrong, however, to consider Haudricourt’s way ofproceeding as mere ‘sensualism’. As he understood it, sensoryexperience proposes different values which can be connected withsystems of ideas, such as that by which pork is valued in China andreviled in the Judeo-Christian world, or by which the plant world hasserved as a referent for the East, whereas the West has tended toconceive of itself in terms of the animal-breeding model. Looking at

André-Georges Haudricourt 221

human beings’ relations to the natural world brings to light not directdeterminations but rather choices based on their observations of andattempts to imitate animals: swimming like frogs, running like horses,singing like birds, digging in the earth like pigs, counting in a waysuggested by leaf lobes and so on.

Conversely, animals imitate us when they live in our company: onlydomesticated dogs bark as if they wanted to speak to us – wild dogs livein packs and howl. With comments of this kind, often made jokingly,Haudricourt sought to draw attention to both the reciprocal ‘learning’that might be said to take place between the natural and humanmilieus, and the biological processes common to all living species. ‘Idon’t separate the natural and the human sciences’, he liked to say. Thisapproach constitutes a refusal to imagine any kind of autonomy for thesymbolic order or any dissociation between the biological and the socialthat might be arbitrarily imposed by culture. Close in this to GregoryBateson, Haudricourt conceived of the natural milieu and humanbeings as having a kind of mutual hold on each other, and he sought toconceive of that hold by suggesting the possibility of organisationalschemata common to both the natural and cultural orders.

This approach meant that he rejected all spiritualism, as well asliberal individualism and its psychological presuppositions, workinginstead to link conceptions of the world and nature to the history of themost material practices. He thus identified a few fundamentalfunctional wholes within which necessary ties might be understood tohave developed between the environment, techniques and representations.As he saw it, these kinds of contiguities between natural milieus,societies and schemata for interpreting the world were determiningtotalities in which all were inescapably caught up. By refusing torecognise or posit any break in the continuity between physical andmental activities, body and soul, Haudricourt developed an anti-dualist,immanentist, deterministic conception of the human sciences thatintimately links human life, gestures and practices, as well as moral andreligious values, to plant and animal life.

As repeated action, technique is aimed at transforming a materialworld that is resistant to humans, their needs and specific interests. ForHaudricourt, technical action is effective both physically and sociallybecause handling or treating plants and animals forms a continuumwith humans’ treatment of each other, a continuum both cultural andgeographical, a continuum that might in turn be called a civilisation.The gradual development or constitution of that continuum and therelational principles that prevail within it were inspired by technicalinitiatives that involved a savoir-faire (physical know-how) and savoir-penser (intellectual understanding) that are closely entwined. The

222 Alban Bensa

contours of these techno-cultural worlds could only be discerned byattending to ‘typical, ancient’ facts. Techniques had been invented bytrial and error and transmitted by teaching and learning; they werelater adapted and modified in the course of a history in whichestablished ways of doing interacted with local innovations,conformism with ingenuity, in which local innovations came tointerfere with established ways of doing, and ingenuity withconformism. Haudricourt’s method is thus genealogical in the sensethat it encompasses both inheritances and transformations, more orless unconscious old ways (pesanteurs) and conscious practices, withoutestablishing any simplistic correspondence between a series ofmovements or gestures and a particular ‘mentality’. As a historicalconception of societies, it is aimed less at identifying the causes ofindividual or collective behaviour than in shedding light on the specificecological constraints that may have oriented social relations in this orthat direction and enforced ways of acting and interpreting, each withits particular logic, ways that varied according to whether one wasraised in China or the Fertile Crescent, for example.

Haudricourt’s model of determinant interactions

It was during his stay in Vietnam, lasting over a year (1948–49), thatHaudricourt developed his model of an opposition in the practices ofagriculture and animal breeding between Asia and the West, a model henever ceased to enrich. Before this decisive experience and the sense itbrought of being ‘entirely elsewhere’ (Haudricourt and Dibié 1987:94), there is no mention of any ideas on ‘the pastoral peoples and thegardening peoples’. When discussing his reading of the Bible – ‘one ofthe funniest books there is’ – with Charles Parain in 1937, for example,he was more directly interested in farming vehicles and threshing, citingIsaiah 28:27–28 in this connection:

27. For dill is not threshed with a threshing sledge, nor is a cart wheel rolledover cummin; but dill is beaten out with a stick, and cummin with a rod.28. Does one crush bread grain? No, he does not thresh it for ever; when hedrives his cart wheel over it with his horses, he does not crush it.

It was his experience of the Far East as ‘the world turned upside down’that led him to wonder about the source of behavioural differencesbetween persons who had originated in distinct civilisations and todevelop explanatory hypotheses for these differences. He wrote his first,as yet unpublished text, ‘Recherche des bases d’une étude comparative


des mentalités extrême-orientales et occidentales’ in Hanoi in 1948 onthe basis of his unpublished correspondence of the time. Also aroundthis time, he spoke of his ideas to Jacques Gernet, who found inspirationin them for a talk entitled ‘Le comportement en Chine archaïque’,published in 1952 in the Annales. As Haudricourt later explained: ‘Formy part, I waited ten years longer than Gernet to publish an article Icalled ‘Domestication des animaux, cultures des plantes et traitementd’autrui’ (Haudricourt 1962). Though L’homme et les plantes cultivées,written with Hédin and published in 1943, and L’Homme et la charrueà travers le monde, written with M. Delamarre and published in 1955,may be considered a point of departure, the fact is that his own writingson the subject amount to no more than fifty pages, namely a series ofshort articles published from 1962 to 1986.

In the general schema that Haudricourt constructed, an oppositionis developed between sheep, goat and cattle-breeding societies, whichin the Neolithic Near East also cultivated grain plants (wheat, barley),and societies in ancient Asia and Oceania, where pig-breeding andtuber plant gardens (i.e., rice paddies) predominated. These twotechnical worlds are understood to have inspired two different sets ofmoral, religious and philosophical attitudes, characteristic of two‘mentalities’ that cannot be reduced to one another: the western andthe eastern. The separating of wheat from chaff and the shepherd’sdirect authority over his herd may be seen as opposite to the gesturesand practices of a Chinese or Oceanian horticulturist or plant grower orfarmer, who ‘accompanies’ the growth of various tubers withoutintervening directly in nature.

Herbivores, cereal grains and transcendence

The first western herdsmen-breeders were governed by specificconstraints in their approach to animals of the bovine, ovine and equinefamilies. These animals lived together in herds in wide-open spaces; theywere not at all in competition with humans to find food, and they did notspontaneously come near them. To get close to them, hunt or gatherthem, and thereby subject them to human authority – the law of the stickor whip – camp or village dwellers first had to follow them and channelthem into tight mountain valleys. In Haudricourt’s view, this violentcoercion was of crucial importance – historically structural, in that itendowed what would become the monotheistic world of Jews, Christiansand Muslims with a religious, philosophical and political model, themodel of transcendence, dualism and absolute hierarchical authority.

The distance separating wandering herds of herbivores from theirmasters-to-be, and shepherds from their beasts after the latter had comeunder human control, appears as the first technical experience of

224 Alban Bensa

transcendence. The gulf separating these animals from thecommunities that domesticated them is cast in the image of the distanceseparating God from humans, a god whose authority is expressed by thearm of Jehovah-the-father raised to threaten humans, homologouswith the stick the shepherd uses to beat the animals in his herd. Thisdistance is attested by the arbitrary separation of body from soul, ofideas from their shadow in Plato’s cave, and it is also to be found in theAristotelian political model, where the leader is understood to be afather imposing his law on a childlike people. Subaltern internalisationof transcendent authority was nothing more than this feeling ofobligation that dominated the western moral conscience up throughthe writings of Kant. For Haudricourt, the categorical imperative had itsorigin in what the shepherd does when he leads or directs his herd withhis crook. It is diametrically opposed to eastern pragmatism, whereinsituations are evaluated as a function of immediate circumstances,without any projection of abstract moral demands from without. Thisis particularly clear in the history of pig domestication.

Pigs, tubers and immanence

Haudricourt located the beginning of the chain of constraints hediscerned with regard to pigs in the domain of plants of the variety thatstock a great quantity of nourishment: ‘vegetal growth is indirect in away because it is always preceded by the stocking of energy reservesover a variable length of time, food that will be used later’ (Haudricourtand Hédin 1943: 70). Plants that stock great quantities ofunderground reserves in the form of carbohydrates or fats and lipidsgrow in ‘variable climates where there is one fairly long season that isunfavourable to vegetation in that it is either extremely dry or cold’(ibid.: 72). This describes the case of rhizomic plants and tubers intropical or monsoon regions, where the rain, which brings to an end along dry period, enables yams, taro and other edible roots to growrapidly and accumulate new reserves for the following season.

Remarkably, this distribution of these plants coincided with thedistribution of zones that were formerly inhabited by wild pigs. Theseanimals ate tubers and small animals living in fairly loose soil; beforehorticulture and breeding, they lived only in the tropical forest zones ofAsia (Sus scrofa, which later moved into Europe), Africa (Potamochoerusporcus) and North and South America (Dicotyles ajacu). Wild pigs usedtheir snouts, teeth and tusks to dig up nourishing tubers. As Haudricourtunderstood it, as soon as the humans who had settled on the edges oftropical forests in Africa, and later Asia and Oceania, learned to makethese plants their staple nourishment, they found themselves incompetition with the wild pigs. And he hypothesised that, during the


proto-agricultural period, they learned from the pigs how to find and digup wild yams and taro roots, replacing the pig’s snout with a digging stick.

To eat the tubers found in the forest, people needed to learn how tomake a fire and how to wash them before cooking them, because certaintaro and manioc roots become toxic on exposure to the air. Whilehumans learned to light a fire 1.8 million years after the first hominidsappeared in East Africa in 2.5 million BC, hunting and gathering onlygradually and partially gave way to horticulture, plant growing andfarming much later. Yam and taro cultivation is only attested around12,000 BC in Southeast Asia; significantly, it is contemporaneous withthe domestication of Suidae or porcine animals.

With the invention of plant cultivation by means of cuttings andtransplantation into prepared soil, humans entered into directcompetition with wild pigs, which ruined human planting work bydigging up the soil in search of yams and taro roots. To prevent thisdamage and neutralise the pigs’ destructive power, hunting had to beintensified, planted gardens enclosed and pig movements monitored,while pigs themselves also became a source of meat for the domestic unit.

In any case, Haudricourt points out, these animals could only havebeen domesticated after agriculture began, not before, because raisingand breeding them presupposed their being fed by people, first indirectly,when the pigs pillaged the humans’ gardens and stayed in the generalvicinity of their settlements to eat human excrement, and later directly,when tubers came to be grown specifically as animal fodder. Obviously,he notes in passing, the domestication of wild pigs, like that ofherbivores, could not be thought of as deliberate. It should rather beseen as the effect of the gradual invention of tuber plant cultivation,which required intensified hunting to protect gardens, which in turnmade it possible to produce an abundance of food, the remains of whichcould be used to feed the pigs and thereby domesticate them.

A major advance in the process was made when men begancapturing piglets in the forest and giving them to the women of thevillage to suckle or nourish with porridge made from tubers, as was stilldone until very recently in certain regions of Melanesia. These animals,which were now attached to the house and family, were not rigorouslyseparated from their as yet wild relatives. Adult sows were mounted bywild boars in a kind of semi-breeding, where herd reproduction was notmonitored or controlled. All human efforts were concentrated on‘mothering’ the infant piglets found in distant or relatively close forestareas. Like children, the piglets benefited from their ‘mother’s’ care; shefed them, attended to their bodies, even gave them a name. Cohabitationof this sort, notes Haudricourt, no doubt constituted an importantenrichment of humans’ relations with natural species, and thus of

226 Alban Bensa

inter-human, i.e., interpersonal relations. The extension of affectivity toinclude nature was an expression of the new closeness to animalsinduced by domestication. In return, it increased mutual socialrelations among people with personalised affective ties to pigs, dogs andeven plants, which they now thought of as members of their families.Hunters in the Palaeolithic period and in the area of the Lascaux cavescould only act with regard to wild animals by means of signs. But withthe domestic breeding of animals, namely pigs and dogs, humans begancommunicating more directly and indeed physically with individualisedanimals on a daily basis, speaking to them, stroking them, caring forthem and feeding them (sometimes even mouth to mouth).

In China, Southeast Asia and Oceania, pigs live under humandwellings. In contrast to the western attitude, their scatophagicbehaviour is not rejected as dirty or disgusting. A pig’s stomach is longenough to allow it to digest without ruminating, and eating excrement(i.e. food pre-digested by other animals) has the effect of a kind of Alka-Seltzer, enabling them to digest yet other foods. Pigs, then, suckled at thehuman breast almost like children, became human beings’ permanentcompanions and waste disposers.

Pigs reproduced in the forest, as did yams that had been forgotten inspaces that were once gardens. The yams reverted back to the wild andproduced new clones (Haudricourt 1964). Forests and former gardens,places where nature manifested its singular power to provide humanswith unexpected resources, were endowed with ambivalent meaning.Far from village dwellings, they were places where one might get lostand even perish, but where one might just as easily find plants andanimals which could be brought back to the family to enrich the usualmenu. Such discoveries were made possible by the ancestors, humanbeings who had returned to nature after their death. The ancestors’power was ‘taboo’; that is, both negative and positive, because in theplaces where they operated, far from habitations, one could meet witheither death or life; there was the risk of losing one’s way, but potentiallyalso the opportunity of finding something that would improve one’sdaily life. In Oceania, then, the taboo has nothing to do with notions ofpurity and impurity but is what characterises human relations withspaces that have been abandoned to spontaneous vegetation.

It is therefore not at all surprising that both pigs and yams werethought of as children found in the wild and thus as assimilable to theancestors, who prolonged human society into both the future and thepast. A regularly developed theme in Kanak tales is the discovery, inspaces far removed from human settlement, of beings who, whenbrought back into the domestic space, enrich it with forces that are sostrong and in some cases so new that they become difficult to control


(Bensa and Rivierre 1983). Making a detour through these wild spacesfavours the renewal and diversification of nature and society. Theproducts found there are integrated into the household like providentialfamily extensions. This would seem to explain why plant gathering andcultivating are described by Oceanians as maternal, motheringpractices. The yam laid in the soil like a newborn in its crib andtransported and swathed like a child at harvest time is likened to a giftfrom the ancestors that humans figured out how to reproduce, makingit yield others of its kind. Likewise the pig brought back from the wild,where the bodies and spirits of the founders of human lineages lie, isbrought up as if it were a child. In this view, there is no boundarybetween the natural and supernatural separating socialised beings froma distant ‘beyond’ where inaccessible entities live. The ancestors arethere in the visible world, just behind all that is manifest in it. They carefor their descendants by providing the living with food; in return, or forpropitiatory purposes, the living maintain them through rituals thatare in fact gifts of food. In the Kanak world, boiled yams are offered (inthe past it was also the heart and liver of sacrificed persons).

These products of the wilderness – such as forest yams and pigs,which were simultaneously ancestors to be maintained and appeased bygifts of food, and animal ‘babies’ whose growth was likened to that ofhuman babies – were dependent on the persons who found, captured,suckled and fed them. The substitution and substitutability obtainingbetween the human, the vegetal and the animal accords well with thenotion of the immanence of ancestral powers and the sacred. Theactions of horticulturalists or plant growers and of the breeders of semi-wild pigs only facilitated the manifestation of these forces. Theyaccompanied them rather than sought to impose their will on them,since they were ‘thought of’ or ‘understood’ as integral parts of a wholein which nature, society and the ancestors were associated rather thandissociated. The point, then, was to favour the growth of plants bypulling up weeds and digging a cavity beneath a planted tuber for it togrow into. In Vanuatu, for example, people feed pigs with boiled tubersso they do not wear down their teeth by digging in the forest groundand so that their incisors grow into each other in such a way as to formbracelets, a precious exchange commodity.

A similar gesture of reception and welcome, of care andmaintenance, can be found in Oceanian political rhetoric, in which thetaboo power of the chief (his mana) is linked to that of a plant or animaldiscovered in the bush or on distant shores, a being then transformedinto a child-ancestor that needs to be raised. In New Caledonia, peoplespeak of ‘child chieftains’ who bear within them the powers of the bushand will gradually be domesticated by the old men of the land, the

228 Alban Bensa

masters of the soil. Further north, in the Melanesian archipelago, pigaccumulation is the focus of rituals accompanying the ‘Big Men’: theirpower is proportionate to the number of bush-discovered piglets theyare able to raise, that is the number of relations these warrior-huntersmaintain with ancestors who have become their children through thedomesticating labour performed by their wives.

Such was Haudricourt’s way of thinking about the way the societiesof Asia and Oceania made pigs and yams into ancestors and childrenproduced by their care, in opposition to the societies of the FertileCrescent, where humans were made to depend on the protection theywere granted by larger herbivorous animals. In the Indo-Pacific region,his thinking went, pigs were thought of as child-ancestors, whereas inthe ancient Mediterranean world, domesticated bovine and equineanimals were understood to have direct kinship with humans. Thecontrast is indeed striking between the image of women in Papua NewGuinea suckling piglets and the image of the Christian crib, in which ababy in a stable is warmed by the breath of a cow and a donkey.

The relation of continuity between human society and animal andvegetal species obtained in the East was hardly conducive to theemergence of anything resembling the dualist opposition between thepure and impure characteristic of civilisations of grain-growingherbivore breeders who ‘separated the wheat from the chaff’ and rejected‘black sheep’ (in French, brebis galeuse, literally ‘gall-infected sheep’) fromthe herd. As Haudricourt liked to point out, the sequence in the scriptureswhere Jesus exorcises a man by transforming the demons in him into pigswas simply unimaginable in the Asian-Pacific cultural universe.

According to this scheme, large herbivorous animals and theirmasters were the parents terribles of Middle Eastern humans, whereaspigs were the pampered, liberally raised children of the Chinese andOceanians. The figure of the Biblical or Koranic herder-farmer is thusfor Haudricourt the symmetrical opposite of the figure of the Chinesegardener with his pigs living beneath the house.

Materialism and determinism

In their famous 1903 article, ‘De quelques formes primitives declassification: Contribution à l’étude des représentations collectives’,Durkheim and Mauss claimed that sociology was in a position to ‘shedlight on the genesis and thereby the functioning of logical operations’.Similarly, the conclusion of Les formes élémentaires de la vie religieuse(Durkheim 1912) sums up the point of view of the Ecole Française deSociologie on the matter: categories are ‘social things’ that ‘were not


made to apply exclusively to the social kingdom’, but ‘extend to realityas a whole’. Although Haudricourt was a diligent student of Mauss, hedid somewhat modify this argument by making technical gestures andpractices the source of representations, that is, by working to anchorthe social in material practices. In this understanding, categories didnot arise from projecting the social on to the real but rather throughthe extending of technical experience and the effectiveness of thatexperience to representations of social relations. The social and logicalcategories that humans would later think of as obvious, as the givens oftheir consciousness of the world, might in fact be thought of asmodelled on the vital relations that humans had with plants andanimals. In the model presented above, two wholes are identified,corresponding to the different forms taken by agriculture and animalbreeding in the west and in Asia and Oceania, wholes whosecomponents – contexts or milieus (geography), practices and know-how, ideas and values – were tightly intertwined. Two types ofexplanation were needed in Haudricourt’s view to account for the originof these networks of correspondences, one biological, focused on thebiological rules that governed interaction among living species, theother emphasising the history of human populations.

Natural sciences and social sciences

Haudricourt’s particular way of looking at social facts was profoundlyshaped by his training as an agronomist and his experience as abotanist. He liked jokingly to differentiate himself from both Leroi-Gourhan, who he said had come to human sciences via zoology, andLévi-Strauss, whose structuralism seemed marked by his geology walks.

As a knowledgeable biologist investigating the capacity of our senses toinform us about the world, Haudricourt regularly looked to such fields ofinquiry as brain chemistry, the symmetries and asymmetries of the body,the loss of a developed sense of smell that followed on from anthropoids’adoption of the vertical position (distancing the nose from the ground),and so on. Because we share with nature the same chemical make-up andthe same physical and biological organisation, it was important never tolose sight of the continuity between the natural and human sciences.

Plant and animal properties and ways of life were like a force fieldthat humans were compelled to deal with. Conversely, life altogetherwas organised by rules that applied to or ran through all species.Haudricourt perpetually situated himself at the junction betweenbiological inheritances that either accumulated or were lost throughevolution, and inheritances that were transmitted, abandoned orrecurred through teaching, learning and imitation, as well as habitsthat had developed within human societies in the course of their

230 Alban Bensa

histories. It was as a resolute opponent of any form of innatism,including Chomsky’s in the case of linguistics, that he explored boththe biological and social origins of sensory and cognitive schemata.What was left to humans of the sense of smell that their primate cousinsstill retained? How was it that seeing was so difficult for people in theMiddle Ages that no one seemed able to draw a plant or animal at allrealistically? How was it that, if everything came down to differences,‘classifications’ could be established that grouped together certainfeatures and excluded others? Constantly seeking explanations,Haudricourt refused to imagine a cause that was not itself determinedby another cause; he therefore moved ineluctably from the social to thebiological via techniques. The body governed the mind because it wasboth a living and an interactive arrangement. The constant externality ofcausality was no doubt one of the major traits of his thought, and he usedit to combat not only innatism, mentalism and spiritualism, but also theidea of arbitrary cultural power or imposition. He was strongly attachedto the determinist dimension of the materialist sciences, and thereforestrongly denied that there was any value in attempts to demonstrate anysort of discontinuity between the biological, the social and the symbolic.It is on this essential point that his theoretical orientations seemincompatible with Lévi-Strauss’ structural anthropology.

Human sciences, historical sciences

In Haudricourt’s view, the link between the life sciences and the sciencesof man and society had to do with their shared inscription in afundamentally historical order. The relations he established among thevarious levels of reality were first and foremost historical, whether thismeant the biological history of the species or the history of populationsand their environments. We cannot doubt that for Haudricourt therespective models of grain-growing herdsman and pig-raising tubergardener obviously correspond and apply to members of these twomajor sets of civilisations, and that this obviously preconditions theways they look at the world. However, it was just as clear to him thatthe people of the Fertile Crescent and those of the Middle Kingdom orthe Oceanian continent were in control of their breeding and farmingtechniques – that is, they adjusted their know-how to the situationsthey had to deal with in such a way as to ensure at least a minimum ofeffectiveness. They were thus inscribed in a historical dynamic ofconstant adaptation that accounted for local variations in the model,which in their turn were to be studied on a case-by-case basis.

In fundamental agreement with Bachelard’s critique of generality andcorresponding defence of the inescapable importance of detail,Haudricourt perpetually vaunted particularisms. Differences constituted


genuine challenges for the idea of the absolute, for the systematic spiritthat had been generated by the concern particular to all the descendantsof Abraham for transcendence, purity and order. And differences werealways to be explained in historical terms. Haudricourt used hisextraordinary erudition in matters of event history, battles and themovements of peoples and kings – knowledge which, of course, implied aprecision regarding dates and locations – to reflect on thoroughly concretesingular attitudes. Universality was reserved for the domain of biology,and Haudricourt took great pleasure in relating his colleagues’ theoreticalorientations (like the valuing of this or that animal and a specific way ofpruning grapevines) to their particular individual itineraries and thehistory of the social groups they belonged to. ‘My Marxism is limited torelating what people think to the situations they find themselves in eitherby choice or in spite of themselves’, he remarked to me one day.

In his view, recurrent attitudes or ones that could be transferred fromone level to another (for example, a way of handling plants to a way oftreating other humans; links between sensory experience and the mostabstract conception, etc.) bore the mark of technical gestures andpractices and the representations governed by them. Those attitudesalso bore traces of the history that had been lived through by the peoplewho manifested them. Thus rejecting, as Mauss had, any reading ofindividual behaviour in strictly psychological terms, he refused, forexample, to imagine that Heidegger’s interest in being could beunrelated to the German quest for communal unity, that Levi-Strauss’concern for universalism was not an indirect effect of a desire for socialintegration, or that the fashion for developing theories about métissage,that is, racial or ethnic mixing and interbreeding, was not also an echoof the existential problematic at the heart of the experience of so manyexcluded communities. At the risk of assuming or seeming to haveassumed a reductionist position as a provocateur, Haudricourtcontinually sought out traces of the most concrete experiences inmoral, philosophical and political attitudes.

When his comparative approach encountered linguistic, sociologicalor technical similarities in practices, Haudricourt always gave a priorityto the hypothesis of direct contact, either long past or recent, betweenthe peoples using them. ‘How can we really know’, he threw out at theend of an interview, ‘whether Roman tiles do or do not come fromcountries that used split bamboo?’ This remark evokes his stronginterest in the work of Graebner and diffusionism, a theory that clearlyposed the question of exogenous versus endogenous change. Thisproblematic is also at the core of Haudricourt’s linguistic work, inwhich he sought to understand the emergence of phonological systemsin a perspective that embraces structural necessities and the effects of

232 Alban Bensa

historical contact. He was just as willing to underscore the importanceof the sudden and the new (the appearance of a high tone, theappropriate adoption of a hitherto unknown technical movement, afashion, etc.), as of what has been maintained and resists change.History, as he saw it, was not merely a series of transformations, butalso a storing up of acquisitions in which teaching and learningreplaced the genetic – a power, that is, that extended back into the pastand established something similar to an identity.

This oscillation between accumulated, consolidated histories andhistorical movement was based on an idea, close to Bourdieu’s notion ofhabitus, of the unconscious recurrence of physical and mental attitudes– Haudricourt compared them to survivals in biology. This enabled himto engage in a kind of archaeology of societies whose primary materialswere their languages and objects. In this work he was mistrustful ofabstract conceptualisation, and easily raged against the Marxist notionsof ‘modes of production’, ‘productive forces’ and ‘state ideologicalapparatus’. Notional Marxism, which talked of production but not plantsor animals, power and control but not know-how, relations ofdomination but not the domestication of pigs and so on, prevented peoplefrom observing the extraordinary material singularity of the ordinary,the enormous impact of seemingly utterly unremarkable actions. On avisit to Leningrad in 1934, he left Soviet researchers thoroughlyperplexed when he explained to them that the fall of the Roman Empirehad less to do with the collapse of a ‘proslavery ideology’, programmed bysome ‘sense of history’ or other, than with the fact that the barbarianshad better military equipment (lances, bridles, etc.).

His functional historicism, with its foundation of concrete erudition,his defence of facts against linguistic effects and the related suggestionthat reality was the same as mere grammatical games or vague termsthat made no explicit reference to the visible, material world meant that,throughout his life, Haudricourt stood in an antagonistic relation to thedominant intellectual environment.

A singular way of seeing

In asserting that he was ‘a real ethnographer’, Haudricourt wasclaiming the right to look at people and things directly, without mentalor linguistic flourishes, in a way that set reflection into motion on thebasis of what is accessible here and now, a way that dismisses everythingthat is not first grounded in observation. He was always surprised to seethat most people were absorbed in their interpersonal relations, and that,like Victor Hugo on the heath, they kept ‘their eyes fixed on theirthoughts’ rather than on the world around them. Concerned to shatterall self-generated and self-fuelled mentalism, Haudricourt continually


reminded us of the presence of the natural world and of human beingsas integral parts of and partakers in this world. His highly concentratedattention to the most visible details and their infinite complexity was asource of profound jubilation, as if, for him, sensuality and affectioncould only be reached through the exercise of observation. In return, henever failed to fulminate against purveyors of ideas without materialballast, ‘philosophers’ of the sort who, like Sartre, did not know thedifference between a lime tree and a chestnut tree.

In direct contrast to many people, specifically intellectuals,Haudricourt did not have to make any particular effort to bring about thecatharsis that frees seeing from parasitical ideas. Spontaneously graspingthe world in its material nudity, and in a perpetual state of amazementas he did so, he was, as it were, quite naturally disconnected from the kindof ordinary social communication that is often in itself a problem forsocial science researchers wishing to accede to objective seeing. In apermanent state of scientific wakefulness and attentiveness, Haudricourthad instead to make his way back to the necessary illusion of social life soas not to seem completely out of phase with others and society. Let it besaid that he often only managed to do this by means of the salutary half-distance from one’s fellow human beings that humour affords us.

Notes

1. Translated from the French by Amy Jacobs.

References

Bensa, A. and J.-C. Rivierre. 1983. Histoires canaques, Paris: Conseil international de lalangue française, Edicef.

Durkheim, É. 1968 [1912]. Les formes élémentaires de la vie religieuse, Paris: PUF.Durkheim, É. and M. Mauss. 1903. De quelques formes primitives de classification:

Contribution à l’étude de représentations collectives, L’Année sociologique, 6 (1901–2): 1–72.

Gernet, J. 1952. Le comportement en Chine archaïque, Les Annales en Sciences Sociales,Paris, Éditions Armand Colin, March 1952.

Haudricourt, A.-G. 1962. Domestication des animaux, culture des plantes et traitementd’autrui, L’Homme, 2(1): 40–50.

——— 1964. Nature et culture dans la civilisation de l’igname: l’origine des clones etdes clans, L’Homme, 4: 93–104.

Haudricourt, A.-G. and J.-B. Delamarre. 1986 [1955]. L’homme et la charrue à travers lemonde, Paris: La Manufacture.

Haudricourt, A.-G. and P. Dibié. 1987. Les pieds sur terre, Paris: Éditions Métailié.Haudricourt, A.-G. and L. Hédin. 1943. L’homme et les plantes cultivées, Paris: Éditions

Gallimard.

234 Alban Bensa

Chapter 10

LOUIS DUMONT: FROM MUSEOLOGY TO STRUCTURALISM VIA INDIA

Robert Parkin

Life and career

Louis Dumont was born in Salonika, Greece, in 1911, where his father,an engineer, was manager of a company building a railway from thereto Constantinople.1 Louis’ grandfather, Victor Emile Dumont, was acommercial artist who created designs for wallpaper in France and forcashmere produced in India in the nineteenth century. Louis’ first wife,Jennie, died in 1977 after forty years of marriage to him. He latermarried Suzanne Tardieu, an expert in Norman furniture at the Muséedes Arts et Traditions Populaires. He died without issue in Paris on 19November 1998, at the advanced age of eighty-seven.

As a youth, Dumont went to the Lycée Saint-Louis to prepare forentry into the École polytechnique in Paris, but dropped out at eighteenbecause of disgust with the bourgeois lifestyle into which this wasleading him. His mother, who had made considerable sacrifices for thesake of his education, threw him out, and he turned to a series of jobs,in insurance, as a proof-reader, and so on. During this time he becamepolitically engaged as a communist fellow-traveller in support of thePopular Front government. However, he eventually returned toacademic interests, frequenting the Collège de Sociologie of GeorgesBataille, Roger Caillois and Michel Leiris in the late 1930s andsimultaneously gaining indirect exposure to India for the first timethrough a group calling itself Le Grand Jeu. In 1936, thanks to GeorgesHenri Rivière, perhaps the most important French museologist of the

Figure 10.1. Louis Dumont, taken by himself, among the Kallar, Tamil Nadu (India),with his chief informant, Muttusami Tevar, 1949. Courtesy Mme Dumont.

inter-war period, he obtained a clerical job in the Musée des Arts etTraditions Populaires, which had just been separated from the Musée del’Homme. Among his tasks was to type up Lévi-Strauss’s notes on theBororo. He also discovered, and followed, Mauss’s courses inanthropology, which he was to describe later as a sort of ‘conversion’.Among other things, this inspired his interest in India, and he passed acertificate in ethnology in 1938. The following year he enrolled in theÉcole du Louvre with a view to preparing a thesis in the history of arton Celtic survivals in modern French tools. However, the Second WorldWar intervened and put an end to this project.

236 Robert Parkin

Dumont was taken prisoner early in the war and sent to Germany. Hewas set to work as a field hand, then as a worker in a factory in aHamburg suburb. During his spare time in captivity he not only learnedGerman but also translated three German books on French folklore.Feeling at some point that he had done enough of this, he asked his wifeto send him materials with which he could learn Sanskrit. Even moreextraordinary, with the connivance of a guard he was not only able tomeet Walther Schubring, an expert on the Jains, but to take weeklylessons from him in Sanskrit too.

After the War, in 1945, he resumed his activities at the Musée desArts et Traditions Populaires, in which capacity he undertook a studyof the southern French festival of La Tarasque, the subject of his firstmajor written work (Dumont 1951). Simultaneously he studied Hindiand Tamil in preparation for fieldwork in India, which he was able toundertake from 1948 through a scholarship obtained for him by theeminent French Sanskritist, Louis Renou. He himself describes thisperiod as one of unremitting hard work.

Dumont therefore started his substantive career in anthropologyrelatively late, at the age of 38. His first trip to India lasted two yearsaltogether, including eight months with the Pramalai Kallar, a Shudracaste of former warriors and bandits in Tamil Nadu. It resulted in hisonly fieldwork monograph, Une sous-caste de l’Inde du Sud (1957c). Aftera further brief sojourn with the Musée des Arts et Traditions Populaires,at the instance of Fürer-Haimendorf he went to Oxford in 1951 toreplace M. Srinivas, a former student of Radcliffe-Brown’s, as Lecturerin Indian Sociology in the then Institute of Social Anthropology. Thiswas during Evans-Pritchard’s tenure of the chair, and Dumont referredto this period as a kind of ‘second training’ (in Galey 1982b: 18). In1955 he returned to Paris, took his doctorate, and was appointed to thechair of the Sociology of India, later changed to a chair in ComparativeSociology, at the 6th section of the École Pratique des Hautes Études(later the Maison des Sciences de l’Homme), where he remained for therest of his career. He apparently owed this appointment in greatmeasure to Lévi-Strauss and Lucien Febvre. Immediately after hisappointment he set up the Centre d’Études Indiennes en SciencesSociales, which became the Centre d’Études de l’Inde et de l’Asie du Sudin 1967, though he left this to pursue other interests in 1970. In 1976he founded ERASME (Équipe de Recherche en Anthropologie Sociale:Morphologie, Échanges), a research team set up by CNRS (the CentreNational de la Recherche Scientifique) with the aim of comparing wholecultures on the basis of their key values (the latter being a basic conceptin Dumont’s mature thought). These years also saw the launch in 1977of a book series jointly published by the Maison des Sciences de

Louis Dumont 237

l’Homme and Cambridge University Press. In 1982 ERASME was takenover by Daniel de Coppet, but following his death in 2002 it broke up.

From 1955, Dumont spent fifteen months intermittently in a villagein Uttar Pradesh, but was neither as inspired by nor as successful in thissecond period of fieldwork, of which little was published, despite plansfor a monograph on mourning. It was here, however, that he wasconfronted with renunciation, from which he later developed theimportant notion of the out-worldly individual, a fulcrum in his latercomparison between India and the West. Instead of writing up this stintof fieldwork, he turned to global accounts of Indian civilisation, first inthe semi-popular La civilisation indienne et nous (1964) and then in hismajor work, Homo hierarchicus: essai sur le système de castes (1966a).Many of the articles in Contributions to Indian Sociology, the journal hefounded in 1957 with his former student David Pocock, were apreparation for this task (Pocock ceased to be editor in 1964, Dumontin 1967). Subsequently Dumont turned to the study of Europeanideology, which he saw as fundamentally reversing Indian ideology instressing both equality and individualism. This led to the two volumesof Homo aequalis, Genèse et épanouissement de l’idéologie économique(1977) and L’idéologie allemande (1991), the first charting theemergence of economic thought as a separate domain from politics inEurope, the second demonstrating variations in individualism in Europe(specifically Germany in relation to France). This was supplemented bya collection of papers entitled Essais sur l’individualisme: une perspectiveanthropologique sur l’idéologie moderne (1983b). All of these works wereeventually translated into English apart from La civilisation indienne etnous. To complete the picture there is his work on kinship, some of themajor works being collected together in the volume Affinity as a value(1983a), including the comparative paper Hierarchy and marriagealliance in south India (originally 1957b), but also consisting of a courseof lectures given in Paris on descent theory and alliance theory,Introduction à deux théories d’anthropologie sociale (1971).

Ideas

Dumont is known today principally as a structuralist, indeed the leadingstructuralist of his generation in French anthropology after Lévi-Strauss himself. Intellectually, however, his thought developed, evenchanged radically during the early part of his career. His involvementwith the Musée des Arts et Traditions Populaires seems to have givenhim not only an interest in material culture, which was still to be foundin Sous-caste, but also a diffusionist perspective on the past, which

238 Robert Parkin

stimulated his interest in south India specifically. This was because hewas influenced to begin with by sub-stratum theories that postulated, inthis part of India, a Brahmanical or ‘Aryan’ veneer to local society andideology that had been diffused to the area from north India but thatnow existed over an earlier and indigenous Dravidian base. Althoughthis view has since been superseded in mainstream anthropology, notleast thanks to Dumont’s own efforts, it has more recently enteredpolitical discourses advocating a specifically Dravidian national identityand the rejection of ‘Aryan’ influences from the north. But it wasinitially because of a desire to get at the Dravidian sub-stratum thatDumont chose to study a middle-ranking caste remote fromBrahmanical influences, in a village without Brahmans.

However, Dumont had also gone out to India having read in proofthe relevant chapters of Lévi-Strauss’s Les structures élémentaires de laparenté (1949), with which the author had himself provided him. This,plus the fact that the Tamils themselves, ‘born sociologists’ according toDumont (in Galey 1982b: 21), thought like structuralists in terms ofbinary oppositions between kin and affines, quickly led Dumont to seestructuralism, not diffusionism, as the best approach towardsunderstanding south Indian social organisation. This was the secondprofound change in attitude he experienced, the first having been thediscovery of Mauss, which initiated the shift in his thought fromcultural to sociological approaches that was completed by hisexperiences of south India and of Oxford (the latter still had somethingof its Radcliffe-Brownian tradition of social rather than culturalanthropology that itself drew on Durkheimian precedents). Sub-stratum theories and survivals were therefore progressively abandonedby Dumont in favour of a combination of Maussian sociological holismand what became an original form of structuralism. I shall return tothe significance of Mauss’s teaching for Dumont’s comparisonsbetween India and the West later.

These sociological and structuralist influences remained withDumont henceforward, though he was never a slavish imitator of anyof them, his structuralism in particular developing in markedly differentdirections from Lévi-Strauss’s. As Toffin (1999: 12) remarks, althoughmeaning is still present in Dumont’s structuralism, values replacesignifiers and signifieds, and hierarchy replaces structuralism of whatmight be called here the ‘simple’ sort of Lévi-Strauss: in short,hierarchical oppositions involving encompassment, and reversalbetween differently constructed levels, replace simple binaryoppositions that may or may not be asymmetric and whose reversal isa matter of different contexts only. However, as Toffin also points out,these differences from Lévi-Strauss are more apparent in Dumont’s

Louis Dumont 239

work on India and modernity than in his studies of kinship, whereindigenous values are of lesser importance than general principles.Another way of putting this is that Dumont’s work on kinship freeditself less from Lévi-Straussian structuralism than his studies of casteand European modernity. And Galey makes the further point that,unlike Lévi-Strauss’s structuralism, Dumont’s ‘does not aim at definingthe universal through the similarities of a human nature postulated apriori’ (1982a: 8). The focus on values permits Dumont to attemptinstead ‘to recapture the entirety of mankind through the recognitionof intrinsic dissimilarities’ (ibid.: 9).

As Moffat hints, however, Sous-caste too is closer to Lévi-Straussianstructuralism than Dumont’s later work on India: ‘In the years since Unesous-caste, the divergences have become increasingly apparent.Dumont’s structuralism is more concretely grounded in particular inter-societal comparisons; it has a stronger interest in social action; and it ismore relative and reflexive’ (1986: xix). Sous-caste recognises fully theimportance of the dichotomy between pure and impure, and therefore ofstatus, among the Pramalai Kallar, but the absence of Brahmans fromthe village Dumont worked in means that the relation of status to poweris left aside here. And although the importance of hierarchy isrecognised, it is not problematised as it was to be in Homo hierarchicusand related work. This is significant in light of the frequent charge thatDumont’s overall account of caste is excessively Brahmanical: to theextent that this charge can be made to stick, it does not apply to Sous-caste. The early numbers of Contributions also represent a decisive andexplicit break with the past. This was quite deliberate from the outset.Thus Dumont and Pocock downgraded earlier tribal studies, which haddominated the anthropology of India hitherto – despite tribes being adefinite minority of the population – in favour of an advocacy of thestudy not just of caste, but of the caste system.

Thus although history, including Maussian world history, remainedimportant in Dumont’s later work, with world-historical perspectiveseven becoming central later on, the flirtation with diffusionist survivalsand sub-strata had disappeared from his writings by the late 1950s. Atthat point in his career, his attention became focused rather on the needto study a society like the caste system synchronically and holistically,as a coherently functioning and structured phenomenon, not as a seriesof historical layers and accidental accretions. Caste was also to betreated as comprehensible in terms of its own values, which werefundamentally religious, not as a pathological or degenerate system ofnaked power and oppression. As Madan (1999: 479) points out,Dumont controversially saw caste as resolving conflict, unliketotalitarianism, which was the elevation of power as a value in its own

240 Robert Parkin

right. Dumont also doubted whether caste could be reformed: it couldonly cease to exist. From this point of view, modern competitionbetween castes represented a change not in fundamental values but inbehaviour, a change produced, furthermore, by external influencesintroduced under the cover of modernity.

His work on kinship is perhaps even more striking in respect of theadoption of synchronic, holistic, sociological perspectives over purelyhistorical and cultural ones, as in the famous article of 1953 on ‘TheDravidian kinship terminology as an expression of marriage’. Althoughthis part of his corpus is relatively small, it is in part concerned to see inwhat he called ‘positive marriage rules’ – elsewhere ‘prescriptivealliance’ or ‘cross-cousin marriage’ – a system found in all parts of theworld, even where historical links are unknown, in the manner of Lévi-Strauss. This stands in marked contrast to his arguments that caste isunique to India. It is exemplified by his re-examination of Australian aswell as Indian material on kinship (the parallels between Australia andsouth India in their both having systems of prescriptive alliance, thoughwith differences in detail, are well known), as well as his lectures onkinship (Dumont 1971). But even within India, there were two otherexamples in which he sought to understand kinship in comparative,universal terms. One was his largely failed attempt to argue away theacknowledged differences between north and south Indian kinship,which he himself recognised was problematic (Dumont 1957a). Theother, much more convincing, was his demonstration that the highlyunusual system of affinity (if such it was) of the matrilineal Nayar inKerala could be understood in terms of wider, pan-Indian values andpractices (Dumont 1983a).

As already indicated, a third major impact on Dumont, apart fromMauss and structuralism, was his four-year sojourn in Oxford withEvans-Pritchard. Although Dumont evidently doubted whether thegreat man entirely understood what he (Dumont) was trying to do, hesaw in The Nuer, with its demonstration of the relativity of groupsthrough unending processes of fission and fusion, the work of astructuralist manqué (Dumont 1968). He may also have been influencedby Evans-Pritchard’s notion of anthropology as essentially a process oftranslation (cf. Madan 1999: 476–77). More than anything, though,the experience seems to have drawn Dumont away from what I wouldclaim is the common division of labour in France between ethnographyand theory, and towards a more Anglo-Saxon situation where it is moreusual for anthropologists to contribute to both. Certainly, according toGaley (2000: 325), ‘he admired British ethnography’, and Sous-caste,his only ethnographic monograph, is noticeably influenced by it. This,plus his interest in India, led him to produce the great majority of his

Louis Dumont 241

work in English, either originally or through eventual translation,which he always controlled very directly. He is therefore probably thebest-known French anthropologist of his generation in the Anglo-Saxon world, for many even ahead of Lévi-Strauss himself. Indeed, heis perhaps better known in Britain, America and of course India than inhis native France, despite his being made a Chevalier de la Legiond’Honneur in 1987 (with Georges Dumézil making the presentation atDumont’s home in Paris). Together with his reluctance to becomeinvolved in wider political issues and his readiness to occupy a singlepost in his career after 1955, this helped cushion him from some of thecompetition and rivalries of the Parisian academic hothouse.

Other intellectual positions that Dumont adopted can be traced rightback to his study of the popular festival in southern France known as LaTarasque, which he set not only in its regional context in southern France,but also in the wider context of Mediterranean Christianity. Thus thestudy combined anthropological fieldwork with a consideration of thewider context of his study, using insights drawn from history. Thisrecognition of different contexts is also found in his later work on India –though stretched out over all his subsequent major writings rather thancondensed into just one – and with a similarly varied methodology.Fieldwork in south India described the specifics of a particular caste in aparticular region, Tamil Nadu. This led to a regional south Indiancomparison of kinship in an extended paper, Hierarchy and marriagealliance in south India (Dumont 1957b), then to wider comparisonsfocusing essentially on the different forms of relationship between kinshipand caste in north and south India (see especially Dumont 1966b). Asfar as India was concerned, this process culminated in the overall accountpresented in Homo hierarchicus (Dumont 1966a), which drew, as alreadynoted, on localised ethnographies (mainly by other anthropologists) aswell as the more global insights of history and Indology. In this regard, itis a pity that Dumont’s fieldwork in north India was so much lesssuccessful that his research in Tamil Nadu. Not only was the areaphysically less pleasant, dry and dusty, and the people not really ‘bornsociologists’ like the Tamils, but the absence of his wife on this tripevidently upset him somewhat, as did persistent sickness. Morespecifically, though, the village he chose had thirty-six castes living in it,unlike the Tamil village, where it was a simpler matter to concentrate onjust one caste. The idea behind his later trip to Uttar Pradesh was to extendregional comparison within India. In the event, in Homo hierarchicusDumont had to rely instead on the often outmoded work of earlieranthropologists to give him a solid ethnographic basis for north India.

But this was not the end of the process of continually expandingcomparative horizons, for Dumont’s work on European ideology, taken

242 Robert Parkin

together with that on India, could be seen as forming the starting pointfor a global comparison of what he called non-modern and modernsocieties. By the time of the work on Europe, ethnography had been leftbehind and general ideas had replaced observed facts. The sourcematerial is rather history, especially the history of ideas, to whichDumont’s own account is often seen as having contributed. The chiefinspiration here again appears to have been Mauss. Although Mauss’sinfluence on Dumont is usually seen in terms of his holism andsociology, Dumont’s overall approach to historical change, even afterhis conversion to structuralism, was also influenced by the distinctiveevolutionism of the Année sociologique school that is perhaps mostclearly represented by Mauss. Indeed, Dumont’s overall comparison ofIndia and Europe is cast in the world-historical terms of a contrastbetween non-modern and modern ideologies, in a manner very similarto that routinely adopted by this school. And like much of its work,Dumont’s typological sequences do not entirely match the historicalones: in particular, while the India Dumont discusses as the paradigmof non-modern societies is contemporary, the Europe of modernideology is mostly historical. Similarly, the separation of economic frompolitical ideas charted in Homo aequalis I resembles a disassembling inmodernity of aspects of a phenomenon that were fused togetherprimordially, which one finds regularly in the writings of the Annéesociologique school and forms a significant aspect of their specificversion of evolutionism (cf. Parkin 2001: Ch. 13).

A further influence of the Année Sociologique on Dumont asrepresented by Mauss concerns the virtues of cooperative work inacademic activities. But this was not the only, nor even the first examplehe had encountered that had this impact on him: there was also his earlywork in the Musée des Arts et Traditions Populaires, where Dumont andhis colleagues saw their work in preserving France’s folklore heritage as‘a duty as much as a profession’, in Toffin’s words (1999: 8). Similarattitudes informed his later work. Early articles in Contributions wereunsigned, to indicate that they were the joint work of the two editors(Dumont and Pocock), a policy soon abandoned, however, as it came tobe feared that it was discouraging other scholars from taking part inthese debates. Dumont conceived of the study of India as a joint projectnot only between Indologists and anthropologists, but also betweenanthropologists undertaking fieldwork in different parts of India, whoprovided the local factual underpinnings to his synthetic view of thewhole. This is represented not only in the use made of various materialsin Homo hierarchicus, in terms of both geographical regions and differentdisciplines, but also in his engagement with other specialists from thesedisciplines after his foundation of the Centre d’Études Indiennes en

Louis Dumont 243

Sciences Sociales (Toffin 1999: 9). ERASME too was intended to gobeyond the usual levels of cooperation, even producing one text jointlyauthored by four of its members (Barraud et al. 1984). In generalDumont appears to have regarded joint work as more akin to scientificresearch, though he also felt that this was difficult to sustain in the socialsciences, where the individual researcher is the norm, resulting in whathe called ‘a chronic instability in the major interest or interests of theprofession’ (in Galey 1982b: 20).

This feeling can also be connected with Dumont’s overall view of thescholar’s task. An immensely hard-working and precise scholar, withan eye for detail as well as the wider picture, he explicitly saw himself asan artisan or craftsman as much as an intellectual, as comes outespecially strongly in his interviews with Jean-Claude Galey (1982b)and Christian Delacampagne (1981: 4), to whom he described himselfas ‘a jobbing social anthropologist’. He felt he had a duty to otherresearchers coming later who might want to use his work in being ascomprehensive as possible. Thus Sous-caste, being intended as acomprehensive account of a particular caste, contains data on manymatters not of pressing concern to Dumont himself but provided in casethey might be of value to scholars coming afterwards. Indeed, as Moffatpoints out (1986: xviii), while the earlier chapters in that book are basedon observation, the later ones reflect more directly the people’s owncollective representations; it is easy to see that it is the latter that mostinterested Dumont, especially in respect of his later work. Anotheraspect of his craftsmanship was that, although allegedly sensitive, evenhostile to criticism (cf. Madan 1999: 490), he was also prepared torevise his own work, as shown in his successive studies of north Indiankinship (e.g. Dumont 1962, 1975) and his occasional replies to hiscritics. Moffat called him ‘a good experimentalist’ (1986: xvi), whileGaley remarked that he ‘was neither a man of systems nor a figurehead’(2000: 326), but one scholar among many cooperating scholars,though undoubtedly at least primus inter pares to his followers.

Rarely engaging as a scholar in the wider world of affairs, Dumontnonetheless clearly had a scepticism of egalitarianism, recognising thatit had its limits, beyond which ordinary moderns were no longerprepared to recognise it (as with race in the West); thus his attitude herewas, in a sense, ethnographic, not ideological. He was similarly sceptical,mainly in conversations reported by others, of the notion of humanrights in contemporary international discourses, seeing it as a form ofuniversalism based ultimately upon the atomising and egalitarian valuesof Western modernity, and therefore quite possibly of doubtful relevanceto other traditions (cf. de Coppet 1990: 123–24, Galey 2000: 327).

244 Robert Parkin

Applications

Although he is often compared with de Tocqueville, Dumont’s careerthus reversed the earlier Durkheimian project of Célestin Bouglé, whobegan studying Western notions of equality before turning to India.Bouglé, who never visited India, certainly understood it less well thanDumont and blamed all its alleged problems on the Brahmans. Fewanthropologists have capitalised more literally than Dumont on theprinciple that studying another society teaches us a lot about our own.As for structuralism, as already noted, in Dumont’s case this was alwaysmore ethnographically specific, less universalistic, than Lévi-Strauss’s.But it is Dumont’s development of Lévi-Strauss’s structuralist device ofbinary opposition into what Dumont called hierarchical opposition thatI want to focus on here. Not only is it the key to Dumont’sunderstanding of India, it also raises a number of interesting issuesregarding how the West too views itself, as I shall argue below. However,it has also been widely misunderstood; at the same time, it provides amethod of relating ideology and practice in a way that was not open, Iwould argue, to Lévi-Strauss’s simpler form of structuralism.

Dumont initially applied this revised form of opposition to therelationship between the Brahman and the Kshatriya in Indian society.Varnas rather than castes in the strict sense, both Brahmans andKshatriyas were associated with different forms of authority. In theBrahman’s case, this meant spiritual authority in a broad sense. Thecanonical depiction of the Brahman as a priest reflects reality in Indiaonly partly. There are priests who are not Brahmans, especially thosewho serve lower status castes and tribes. There are also Brahmans whoare not priests but landholders, having their land worked by oftenuntouchable labour, but seeing themselves as restricted or even non-transactors whose lack of dependence on the gifts of clients and the sinsembodied in those gifts allows them to claim superiority over Brahmanpriests. The role that these landowning Brahmans claim for themselvesis to study the ancient texts, the Vedas, and to perform rituals, includingexact repetitions of these texts, of profound cosmological significance.

In the traditional system the Kshatriyas, by contrast, have authorityin the secular sphere and are associated with secular rule, power andwarfare. Everyone, including the Brahman, is subject to them, but onlyin that sphere. This indicates the inferior status of that sphere comparedto that of the Brahman, who is responsible for cosmic goalstranscending the narrow domain of the practical affairs of the man inthe world, the domain of the Kshatriya. In short, Dumont says, this isnot an ordinary binary opposition of the type exploited by Lévi-Strauss,whether the poles are seen as equivalent in status (or in ‘value’, to use

Louis Dumont 245

Dumont’s term) or as asymmetric. The nature of the Brahman–Kshatriya relationship is that it is a hierarchical opposition in which thevalues represented by the Kshatriya are ‘encompassed’ by those of theBrahman. This is because, while even the Brahman landowner issubject to the Kshatriya in the secular sphere, that sphere is not onlyinferior to, it is also encompassed by, the sphere of the Brahman, whichby virtue of its transcendence is superior overall. This view is reinforcedby the fact that the Kshatriya supports the Brahman in the latter’s task(by giving him land in the first case, and in pre-Hindu, Vedic times, byproviding the sacrifice) and protects him physically by providing socialorder. In other words, the secular sphere has no purpose other than tosupport the transcendental activities of the sphere of the Brahman.

As already noted, this notion of hierarchical opposition is certainlyamong the most misunderstood in the whole of post-war anthropology.A more familiar, though much abused example may make clearer justwhat is involved (cf. Dumont 1980: 239–40).

In pre-politically correct times, the English word ‘man’ had a doublemeaning. On one ‘level’, to use Dumont’s term, ‘man’ was simply opposedto ‘woman’ as its opposite. On the other level, it stood for the whole ofhumanity, including ‘woman’ (as in ‘mankind’). On this latter level, inother words, it ‘encompassed’ its contrary, ‘woman’. Clearly this wentalong with a whole set of circumstances in which things male were seenas ideologically more important, of higher value and so on, than thingsfemale. On the level involving encompassment, moreover, women aresimply invisible, thanks precisely to their encompassment. It is only on thesecondary level, that of distinction, that the category ‘woman’ appears atall. The two ‘levels’ are thus differently structured. They are alsoideologically unified into a single structure: they do not simply representdifferent ‘contexts’ in which first one pole of a binary opposition, then theother, is prominent. The contexts produced by reversing one of Lévi-Strauss’s merely asymmetric binary oppositions are equivalent, in thatmoving between them simply involves reversing the polarity of theopposition. In moving between Dumont’s levels, on the other hand, one ismoving between a superordinate situation of the encompassment (i.e.non-visibility) of one pole by the other, and a subordinate situation inwhich both are present by being distinguished. Thus, to return to India, theBrahman either stands for (encompasses) the whole of society in itsrelations with the cosmos, in rituals in which only he is evident; or else heappears alongside the Kshatriya as subject to the latter’s authority in asubordinate (secular, non-transcendent) situation or level.

So much for encompassment – what about hierarchy? First, giventhat levels are unequally valued, there must be hierarchy. It is fairly easyto relate this to a society like India’s, which is still hierarchical today to

246 Robert Parkin

a high degree (e.g. the persistence of caste, but also aspects of modernlife, like office politics, as well as kinship). The problem for Dumont’scritics has been his insistence that this model is pertinent to the West too,despite the West’s claims to egalitarianism. One result of this has beenthat Dumont has been accused of mistaking or ignoring the nature ofequality, and even of actually preferring hierarchy to it as a mode of life.Even within India, his account has repeatedly been criticised for givingthe Brahman’s point of view and ignoring those of other sectors ofsociety. This can be seen partly with respect to the values of therenouncer, who turns his (sometimes her) back on society in order topursue personal salvation as an individual. Yet the landowningBrahman may be considered closest to these ideals of anyone still insociety, given his status as a minimal transactor – like the renouncer, thelandowning Brahman tries to minimise his dependence on thehouseholder and also avoids exchange transactions, since they carrywith them some of the sin-laden and otherwise inauspicious substanceof their inferior givers. As already noted, it is this that distinguishes thelandowning from the priestly Brahmans, who are more or less entirelydependent on such transactions (cf. Dumont 1966, 1980).

Dumont’s critics have made some significant points, but they stilltend to misconstrue both his own position and the nature, let aloneexistence, of hierarchy in the West. Again, this often reflects sheermisunderstanding. Hierarchy is not simply the basis of the model ofhierarchical opposition seen objectively – as a subjective cultural value,it may itself partake in this very model by actively being one of the polesof a hierarchical opposition. It is perhaps a failure to recognise this thathas most misled Dumont’s critics. Hierarchy in Dumont’s terminologyis not just social stratification: it is the operation of according differentvalues to different things. Here it is useful, I think, to invoke the notionof ‘preference’. Briefly, we may say that while India prefers the values ofhierarchy to those of equality, so that the former encompass the latter,the West does the reverse. In other words, in the West the value‘equality’ itself encompasses the value ‘hierarchy’ in what is clearlyanother hierarchical opposition. That is, equality is an ideal, oneassociated with other ideals like individuality and freedom. As such,ordinarily it is stressed to the exclusion of hierarchy. Yet Western societyis still hierarchical in many respects, which mostly relate to practical(i.e. non-ideological) matters. The world of work in particular ishierarchical, since – however much this may be mystified by modernindustrial relations and personnel practices – orders are still given andobeyed, and firms managed through processes of hiring and firingsubject others. Similarly, the law, government and the military aredomains that are rarely endowed with more than the status of

Louis Dumont 247

necessary evils, in that they too limit the practical exercise of theWestern ideals of liberty and equality, not least because they arehierarchical, generally quite literally so. Yet significantly they alsoappeal to their own support or protection of the ideals of liberty andequality as their ultimate justification, that is, they explicitlysubordinate themselves to them. In short, there is a hierarchicalopposition in the West that places equality in a superordinate andtherefore encompassing position in relation to its opposite, hierarchy.The latter only emerges in domains proper to it, and then as a practicalmatter necessitated by, but also supporting, the level of fundamentallyegalitarian and individualistic ideals. Thus the relationship betweenequality and hierarchy in the West is itself a hierarchical oppositioninvolving levels and encompassment.

This formulation may seem strange, but that is simply because thereis a fundamental contradiction in the Western way of life that ahierarchical society like India is not faced with. For Dumont, hierarchyis unavoidable, anywhere. In India, the parallel to the hierarchicalopposition described above for the West is the reverse situation, in whichhierarchy encompasses egalitarianism, just as society encompasses theindividual and duty encompasses both material interest and freedom(sometimes represented by pleasure). This can be expressed inindigenous terms, in respect of the triple but still hierarchical distinctionbetween the ends of life, dharma, artha and kama, or duty, work andpleasure: all have their place, but in a descending order of value, andtherefore encompassment. Certainly, as has often been remarked, thevalues of the renouncer, which are ultimately concerned with personalsalvation, appear to stress both individuality and the basic equality ofall transcendent approaches to that end. This is far from being anegligible point, since this is an important form of transcendence,though one only pursued by a minority of Indian society (since,moreover, the aim is moksha, that is, liberation from the cycle ofrebirths, it can also be seen as encompassing the above three values). Yetideologically the renouncer is not in society, and indeed often marks hisor her removal from it by undergoing a symbolic death ritual, quitepossibly complete with shrouds and immersion into the Ganges or ariver assimilated to it. Conversely the Brahman’s role is a social one,since he keeps the cosmos in being for the good of society and ultimatelyof humanity; he can therefore claim to lack the self-centredness of thepath of the renouncer. This is one area where, as Richard Burghartshows us (1978), the values of the renouncer and the Brahmanconflict, both politically and ideologically.

However that may be, Dumont argues that, in expressing apreference for hierarchy, the model of hierarchical opposition therefore

248 Robert Parkin

accords with the superordinate value in Indian society; in the West itdoes not. In other words, hierarchical societies comfortably exist inconformity with their ideals, which stress that very hierarchy; egalitarianones never can, because for practical reasons they can never entirelyevade the hierarchy they are opposed to – all they can do is subordinateit as a value to the higher emphasis placed on egalitarianism.

This has at least three further consequences. One relates to racism.Dumont does not oppose this to egalitarianism so much as see it as one ofthe latter’s pathologies. Although the West has influenced some Indiandiscourses in the direction of racism, and differences between castes haveprobably always been seen in part as differences of substance, modernracism never appears to have existed in traditional India. In the first place,the caste system is a system of inclusion rather than exclusion. Althoughone’s practices (cousin marriage, consuming beef and alcohol, polygyny)may consign you to a low status within the system, the system will stillfind you a place. In other words, hierarchies are flexible, because they areessentially relational: you are not absolutely different from me, just moreor less pure, and your rank with respect to me reflects this. Even if youare impure, and although I might shun all contact with you, ideologicallyI do not dismiss you entirely but rank you accordingly.

Egalitarianism, on the other hand, can only produce definitions insubstantial terms, since adopting a relational approach to definitionalong Indian lines would involve introducing the very hierarchy thategalitarianism rejects. However, no society that sees itself as ideologicallyegalitarian seems to be able to free itself entirely from some urge todiscriminate in practice: this is one of the ironic contradictions thatDumont locates in modernity. And if a social group is to discriminatewhile still maintaining equality within its own boundaries, it can only doso through a process of exclusion, that is, by defining the object of itsdiscrimination as wholly different in substantial terms, for exampleracially or ethnically. We should not forget that the white populationsthat dominated certain multi-racial societies in the fairly recent past, asin the southern United States or South Africa, saw themselves asinternally equal, at least racially and therefore in terms of substance, ifnot always socially (e.g. class). This was, of course, contrasted with thedraconian and often vicious discrimination, also in substantial terms,meted out to non-whites in the same society. The practical outcomes ofthis discrimination were the colour bar in the US and apartheid in SouthAfrica (cf. Dumont 1980, Appendix).

The second consequence of applying hierarchical opposition to theegalitarian West relates to questions of discrimination in other ways.Dumont repeatedly insists on the relationship between makingdistinctions and differentially valuing what is distinguished: indeed, for

Louis Dumont 249

him we only distinguish in order to state a preference, that is, to allocatedifferent values. This is another reason why oppositions cannot be otherthan hierarchical. A corollary of this is that if we wish to cease todiscriminate we should not distinguish, that is, should not drawattention to difference. Up to a point, this reflects practice in the West asthis has evolved over the past few decades. There is a contrast here withmany other parts of the world, where it may be much more common forindividuals to be identified casually with others in terms of race orethnicity. In the West such usages are increasingly felt to be wrong, orat least impolite, and as making an unnecessary point – which is not tosay that they have disappeared entirely, of course. Only if ethnic orracial discrimination is being discussed substantively, for example as anissue that still needs addressing, does reference to such differences seemjustified. Similar practices have extended to other domains of potentialor actual discrimination, such as sexual orientation, disability or foodpreferences (vegetarianism, for example). One might also add gender,though here the situation is complicated by a continuing compulsion tomake often oblique reference to difference in circumstances of, forexample, flirting and seeking partners. But even here – in workenvironments, for example – discussing work in relation to genderdifferences is no longer seen as acceptable except in the context ofovercoming any remaining discrimination (again, I am not saying thatit never happens, only that it is no longer considered politically correct).Gender also becomes interesting when it is combined with aconsideration of sexual orientation, or at any rate gay politics. InBritain, at least, the term ‘gay’ tends to have a double reference of thesort discussed earlier for ‘man’ and ‘woman’, sometimes covering bothgenders, sometimes only men, as in the frequent identifier ‘gay andlesbian’; conversely ‘lesbian’ is categorically female, never male. Thissurely demonstrates the continuing power of hierarchical opposition,even in social circles that would appear to have the greatest interest inrejecting not only discrimination, but also the sense of differential valueand hierarchy that goes along with it.

The third consequence of applying hierarchical opposition to theegalitarian West is the relation of ideology to practice. As Allen notes(1998: 3), Dumont was well aware of Weberian and Parsoniansociology, and unlike Lévi-Strauss took them into account by giving thepractical activities they stressed their due place, while characteristicallysubordinating them to the level of ideals. It is in the nature of pragmaticactivities that they may conflict with ideals while at the same timesupporting them (e.g. the offerings the wealthy make to the church ortemple from their ill-gotten gains; see Parry and Bloch 1989). As Parryreminds us (1994), so long as humans have values distinct from the

250 Robert Parkin

world of the practical, their ideals will be unattainable and will thusalways have a separate existence from the actual and the practical. Ihave no qualms in suggesting that, even in a period that stressespractice and agency, ideals remain equally valid objects of enquiry forthe anthropologist. One of the reasons I believe Dumont’s work to be ofvalue is that, through the notion of hierarchical opposition, he hasoffered us a way of relating ideology and practice that is rooted in theDurkheim tradition, yet also goes beyond it. As noted above, the level ofideals and values always encompasses that of practice, since althoughthe former may be reliant on the latter for its fulfilment, the latter isideologically subordinate, sometimes even ideologically unrecognised.It is only when the pragmatics of providing worship or the morallycompromised nature of the world of practice become focal points fordiscussion that they are at all evident, and then only at the subordinatelevel of distinction, not the superordinate level of encompassment.

I therefore suggest that Dumont’s name should be added to thosewho have attempted to combine practice and agency with ideology,including in the most recent period Giddens and Bourdieu, in the middledistance Parsons, and originally Weber himself. However, Dumontdiffers from all of these in according ideology a clearly superordinatevalue with respect to practice, thus keeping him closer to Durkheim,while articulating this difference through the uniquely Dumontialhierarchical opposition.

To recap, therefore, Dumont’s intellectual trajectory can be seen asinvolving a series of shifts. The first was from early diffusionist,culturological approaches drawn from his museum experience toMaussian sociology, holism and world-historical perspectives, supportedby a growing appreciation of Lévi-Straussian structuralism focused onsimple binary oppositions. A second shift was from an early familiaritywith this latter form of structuralism, stimulated also by his earlyfieldwork in Tamil Nadu, to the revision of structuralism in thedirection of hierarchical opposition that was stimulated by his widercomparisons within India, as well as sustained in his yet widercomparison between India and the West. A third shift was fromobservation to ideas, from fieldwork to writings, in forming theevidential basis of his work, though in both producing ethnography andusing it theoretically he was also adopting a distinctly Anglo-Saxonrather than French anthropological methodology. Yet, in combinationwith a lesser but still real familiarity with non-Durkheimian writerssuch as Parsons and Weber, who emphasise practice as much as ideasand values, hierarchical opposition also gave Dumont a way of relatingand reconciling ideology and practice that was simply not open to Lévi-Strauss’s simpler form of structuralism. This is in addition to the

Louis Dumont 251

manner in which he used hierarchical opposition to render less opaqueaspects of Western ideology and practice as well as Indian. It is theselatter aspects, I suggest, that give Dumont his greatest claim to ourcontinuing attention.

Note

1. I never met Dumont personally, though I corresponded with him on one occasion andsaw him speak on another. In this appraisal I am therefore relying on publishedsources, especially for Dumont’s life (Part 1) and his more private views, namely theobituaries by Allen (1998), Galey (1999, 2000), Madan (1999) and Toffin (1999),two interviews given by Dumont (Delacampagne 1981, Galey 1982b), and appraisalsby Galey (1982a) and Moffat (1986). Many of the views I give voice to in Part 2 are alsoanticipated in these writings. Part 3 has a little more claim to originality. A slightlydifferent version of this chapter has already appeared in Spanish (Parkin 2006).

References

Allen, N.J. 1998. Obituary: Louis Dumont 1911–1998, Journal of the AnthropologicalSociety of Oxford, 29(1): 1–4.

Barraud, C., D. de Coppet, A. Iteanu and R. Jamous. 1984. Des relations et des morts:quatre sociétés sous l’angle des échanges, in J.-C. Galey (ed.), Différences, valeurs,hiérarchie: texts offerts à Louis Dumont, Paris: Éditions de l’École des Hautes Études enSciences Sociales.

Burghart, R. 1978. Hierarchical models of the Hindu social system, Man (n.s.), 13(4):519–36.

de Coppet, D. 1990. The society as an ultimate value and the socio-cosmic configuration,Ethnos, 1990(3–4): 140–50.

Delacampagne, C. 1981. Louis Dumont and the Indian mirror, Royal AnthropologicalInstitute News, 43: 4–7.

Dumont, L. 1951. La Tarasque: essai de description d’un fait local d’un point de vueethnographique, Paris: Gallimard.

——— 1953. The Dravidian kinship terminology as an expression of marriage, Man, 53:34–39.

——— 1957a. For a sociology of India, Contributions to Indian Sociology, 1: 7–22.——— 1957b. Hierarchy and marriage alliance in south India, London: Royal

Anthropological Institute.——— 1957c. Une sous-caste de l’Inde du Sud: organisation sociale et religion des Pramalai

Kallar, Paris: Mouton.——— 1962. Le vocabulaire de parenté dans l’Inde du Nord, L’Homme, 2(2): 5–48.——— 1964. La civilisation indienne et nous: esquisse de sociologie comparée, Paris: Armand

Colin.——— 1966a. Homo hierarchicus: essai sur le système des castes, Paris: Gallimard.——— 1966b. Marriage in India, the present state of the question III: north India in

relation to south India, Contributions to Indian Sociology, 9: 90–114.——— 1968. Preface, in E.E. Evans-Pritchard, Les Nuers, Paris: Gallimard.——— 1971. Introduction à deux theories d’anthropologie sociale, Paris and The Hague:

Mouton.

252 Robert Parkin

——— 1975. Terminology and prestations revisited, Contributions to Indian Sociology(n.s.), 9(2): 197–215.

——— 1977. Homo aequalis: genèse et épanouissement de l’idéologie économique, Paris:Gallimard.

——— 1980. Homo hierarchichus: the caste system and its implications (2nd ed.), Chicago:The University of Chicago Press.

——— 1983a. Affinity as a value: marriage alliance and south India, Chicago: The Universityof Chicago Press.

——— 1983b. Essais sur l’individualisme: une perspective anthropologique sur l’idéologiemoderne, Paris: Le Seuil.

——— 1991. Homo aequalis II: l’idéologie allemande, France-Allemagne et retour, Paris:Gallimard.

Galey, J.-C. 1982a. The spirit of apprenticeship in a master craftsman, in T.N. Madan (ed.),Way of life: king, householder, renouncer. Essays in honour of Louis Dumont, Delhi: Vikas,and Paris: Éditions de la Maison des Sciences de l’Homme.

——— 1982b. A conversation with Louis Dumont, Paris, 12 December 1979, in T.N.Madan (ed.), Way of life: king, householder, renouncer. Essays in honour of Louis Dumont,Delhi: Vikas, and Paris: Éditions de la Maison des Sciences de l’Homme.

——— 1999. Obituary. Louis Dumont (1911–1998): an enduring consistency, EASANewsletter, 25: 13–17.

——— 2000. Louis Dumont (1911–1998): a committed distancing, AmericanAnthropologist, 102(2): 324–29.

Lévi-Strauss, C. 1949. Les structures élémentaires de la parenté, Paris: Presses Universitairesde France.

Madan, T.M. 1999. Louis Dumont (1911–1998): a memoir, Contributions to IndianSociology (n.s.), 33(3): 473–501.

Moffat, M. 1986. Preface to Louis Dumont, A south Indian subcaste: social organization andreligion of the Pramalai Kallar, Delhi: Oxford University Press.

Parkin, R. 2001. Durkheimian evolution in the work of Marcel Mauss, in R. Parkin,Perilous transactions: papers in general and Indian anthropology, Bhubaneswar:Sikshasandhan.

——— 2006. Louis Dumont: estructuralismo, jerarquía e individualism, Revista deOccidente, 299: 9–34.

Parry, J. 1994. Death in Banares, Cambridge: Cambridge University Press.Parry, J. and M. Bloch (eds). 1989. Money and the morality of exchange, Cambridge:

Cambridge University Press.Toffin, G. 1999. Louis Dumont, 1911–1998, L’Homme, 150, 7–13.

Louis Dumont 253

Chapter 11

WILL THE REAL MAURICELEENHARDT PLEASE STAND UP?FOUR ANTHROPOLOGISTS IN SEARCH OF AN

ANCESTOR

Jeremy MacClancy

Students of anthropology have it tough: the courses they are taught onthe history of the subject are usually boring, blinkered and Whiggish toboot.1 All too often the subject is presented as a deadening chronicle ofdisciplinary self-improvement, with each generation identifying, thenmoving beyond, the sins of their forefathers (and mothers). Evolutionism,this story tells us, was racist, functionalism dovetailed with colonialism,structural-functionalism ignored history, high structuralism was formystics, postmodernism was an apolitical dead-end, while diffusionismwas just plain wrong-headed. Only the present holds out much promise.‘Onward, ever upward’ is the underlying agenda to this all-too commontale. At times I am surprised our students stay with us.

Of course, this party line of constant self-advancement only appearscoherent because it wilfully excludes so much. At an Oxford lecture Iattended several years ago, Stephen Jay Gould argued that we humans arenot at the apical growing tip of some evolutionary tree but out on a limb, andwe ignore all the other branches we happened not to go down at our peril.It was a salutary reminder of just how random our development can be.

Anthropology is little different. The conventional format of ourhistory remains oddly silent about a whole host of different approacheswhich did not make it for the wrong reasons. And there are even more,still worthy of our consideration today, which contain valuable insightsand suggestive agendas. In other words, what our students need is not

Figure 11.1. Maurice Leenhardt (back row, centre), with Melanesian pastors during aconference, Nouvelle Calédonie 1916. Archives de la Nouvelle-Calédonie, FondsMaurice et Raymond Leenhardt, 12 J.

just a history of anthropology, but an anthropology of that history aswell. They need to learn why some ideas are yet propagated while therest are left to lie fallow. For if anthropology is about putting ideas andcustoms into their contexts, surely it behoves us to do the same withour own practices. We are not special.

There are, of course, exceptions. Not all historians of anthropologyare Whigs with a wilfully exclusionary style. Some have tried to rewritethe past by vigorously promoting the forgotten, the neglected, themarginalised. But this is not a high-minded, Lazarus-like revival of theotherwise dead, merely a variant of the tired self-interested strategy ofthose striving for hegemony. I could cite several examples. Instead I willstick with the one I know best. In the 1960s and 1970s, RodneyNeedham persuaded the University of Chicago Press to establish a series,with him as editor, dedicated to the re-printing of nineteenth-centuryworks, each with a lengthy introduction by himself or one of his brighterstudents. Needham’s aim was manifold. First, these books gentlysubverted established historiographies by exposing their structuringconventions. Second, they acted as cautionary tales for students,reminding them that ideas then being touted as new were not in factquite that novel. Third, they implicitly criticised high structuralistabstraction and love of apparent paradox by demonstrating thatanthropological ideas could be discussed in a pellucid, seemingly

256 Jeremy MacClancy

unpretentious nineteenth-century prose (which happened to chime withNeedham’s own literary style). Fourth, and perhaps above all, thesebooks reflected, and so further established, the singularity and historicaldepth of his own vision of our discipline: who else had had the insight torealise the contemporary value of these forsaken classics? These bookscould thus act as supplementary means of advancing the project moreexplicitly stated in his major essays and books. They were, if you like, thesoft sell complementing the hard sell of his key articles and tomes.

Though a former student of Needham’s, I am still rather surprised tofind myself treading a somewhat similar path, albeit in a far less illustriousmode. In a variety of papers, I have tried to re-illuminate once brightcorners of anthropology, almost forcibly bringing them to others’attention (e.g. MacClancy 1986, 1995, 1996, 2000). In this chapter,however, I wish to do something different. Instead of attempting toresuscitate a long-dead figure, I wish to examine how others have tried todo so, and to what effect. In the process we might learn something aboutthe way histories of anthropology are negotiated for present-day purposes.

My object of attention is Maurice Leenhardt. If the aim of thisvolume is to demonstrate that French anthropology has not been, asthe stereotype has it, all grand theorising from afar and that, to thecontrary, it has in fact a long tradition of empirical fieldwork with itsown grounded theory, then Leenhardt fits the bill extremely well. For inthe 1930s and 1940s, he was hailed by his peers as one of the greatestfieldworkers of his day, and certainly the most long-term one. Moreover,as we shall see, he had his own particular theoretical approach, born inhis case out of his missionary concerns.

In this chapter, I first sketch his life; then examine the interests of hisvaried would-be resuscitators, including those who wish to revive the man,only in order to put the knife back in; I end with some general comments.

A life

We can be brief. Our subject’s biography has already been recountedmany times, most memorably by Clifford (1982).

The Leenhardts were a pious family of bourgeois Protestants inclinedtowards the liberal professions and the pastorate. Franz Leenhardt(1846–1922) was an eminent geologist who desired to fuse theology withpositive science. His fourth child, Maurice (born 1878), was a mediocre,occasionally troublesome student, who went deaf in one ear and failed hisbaccalaureate three times. Inclined towards the missions from an earlyage, he married in 1902 and was ordained three months later; four daysafter that, the couple left to establish a mission in New Caledonia.

Will the real Maurice Leenhardt please stand up? 257

Life on the former penal colony was not easy. The locals wereincreasingly hemmed in by expansionist colonists: farmers, herdsmen,miners, Catholic missionaries, administrators. Between 1855 and 1900the local population dropped by almost half, their death hastened byimported alcohol, diseases and firearms, while the rebellion of 1878–79led to the killing of two hundred Europeans and several times thatnumber of indigenes. Leenhardt had to contend with demoralised butstill proud locals (known as Canaques), territorial priests and colonialskeen to keep their potential labour-force subdued, not educated. The high-minded young missionary was forced to learn, by painful mistakes, howto be diplomatic yet firm if he wished to assist those he defended. To thecolonial government of the day, he became a long-term irritant.

He established his mission station, Do Neva, on the eastern coastand worked intensively with his natas, pastor-evangelists. They were thecentral plank of his conversion strategy: he saw his own role asprotecting and encouraging an autonomous Melanesian church. Thuseducating the natas became his first priority, essentially throughpractical exercises in the comparative analysis of religious languages:the biblical and the indigenous forms. Leenhardt would make tours tovisit his natas stationed in villages along the coast and in the bush, andhe and his wife also ran a school for local children.

Though his church steadily grew, Leenhardt remained concerned aboutthe thoroughness of locals’ conversions. In order to understand this processbetter, he realised he would have to comprehend indigenous ways as deeplyas possible. So, during his first leave home in 1908–9, he broached the workof Durkheim’s Année Sociologique group and of Lévy-Bruhl, while his father,‘as usual, urged him toward more precise observation and stressed theimportance of collecting genealogies’ (Clifford 1982: 75). On his return toDo Neva, he commenced in earnest an ethnographic study of hisneighbours. He wished, through repeated discussions and encounters, toprobe the lived reality of customary life, and so derive the most appositeterms and forms for a New Caledonian Bible. Working with his natas, heaimed to translate the words of God into expressions meaningful to a livingMelanesian language. To Leenhardt, this laborious co-operative translationwas the key to any worthwhile conversion.

In 1920 he and his family returned to France. There he began toengage seriously with university anthropology. At his father-in-law’shouse, he met Lucien Lévy-Bruhl. This proved to be the start of a longand productive friendship, the philosopher eager to learn from thefieldworker about Melanesian conceptions and Leenhardt keen todiscuss his companion’s ideas about the ‘prelogical’ and ‘modes ofparticipation’. He gave papers at academic meetings, met with MarcelMauss, and at his request published a long article on a key New


Caledonian festival. In 1922 he embarked on a seventeen-monthinvestigative tour of all the French Protestant missions in sub-SaharanAfrica. During these visits he was able to check and prove the worth ofthe guiding ideas he had elaborated in New Caledonia.

In late 1923 he went back to Do Neva, this time without his family.It was to be a difficult stay, with Leenhardt wishing to see the principlesof his evangelising approach well-rooted, so that it would survive hisdeparture. He believed in a strong indigenous church, with greatautonomy, little hierarchy, and reliant on local pastors. To his greatdisappointment, his ideas would be rejected by his metropolitansuperiors. They continued to regard their independent, criticalcolleague with suspicion; his recommendations were quietly ignored.

Back, this time definitively, in Paris in 1926, Leenhardt had to look forwork, as his Mission Society would not give him a position of responsibility.He gained employment as an ‘urban missionary’, attending to the needyof the city. That took up half his time; the rest of each week he dedicatedto writing and teaching, for both anthropological and missionaryaudiences. He established and edited a bimonthly journal aboutmissionary matters, Propos missionaires, and besides producing numerousarticles of ethnography, published a trilogy of works: Notes d’ethnologienéo-calédonienne (1930), Documents néo-calédoniens (1932), and Vocabulaireet grammaire de la langue houaïlou (1935). Thanks to his growing friendshipwith Mauss, he began in 1933 to teach at the École Pratique des HautesÉtudes. Within a few years he was doing half of Mauss’s teaching.

In 1937 he accepted an invitation to publish a popular account ofindigenous life, grounded in fieldwork. His Gens de la Grande Terrepresents a sympathetic account of New Caledonians as bearers of arich, rounded tradition, worthy of our interest and increasinglythreatened by colonialist inroads. Shortly after its publication in 1938Leenhardt and his wife returned for more than a year’s stay to carryout a survey of the languages and dialects of New Caledonia andnearby islands. The result, Langues et dialects de l’Austro-Melanesie, is asomewhat forbidding compendium of grammatical sketches andvocabularies. Of course Leenhardt, while conducting his survey, did notact as an aloof academic, but attempted to intercede in important issuesof the day. Many of his supporters now venerated their white-bearded‘Missi’, while his white opponents continued to regard him, with unease,as an influential, interfering négrophile.

When, in 1940, the anti-Semitic laws of the Vichy governmentforced Mauss to resign from the École Pratique des Hautes Études, hehad Leenhardt replace him. Leenhardt turned the lectures he gave thereinto his most famous, most challenging book, Do Kamo: la personne et lemythe dans le monde mélanésien (1947). The next year he returned, for


the last time, to New Caledonia as the founding director of agovernment research station in the capital, the Institut Françaisd’Océanie. As before, colonialist differences made his stay initially verytense. While waiting patiently for these criticisms to peter out, he gentlyencouraged the work of his junior colleagues and so cemented thereputation of the fledgling institute.

In Paris, Leenhardt, who was now well into his seventies, kept up anexemplary list of activities: organising, teaching, writing. In 1952 heaccepted the presidency of the Alliance Evangélique Universelle. Thenext year he planned a final trip to la Grande Terre. But that summer hewas diagnosed with cancer, and died in January 1954.

In search of an ancestor

In my sketch of his life, I have coasted over Leenhardt’s central ideas,because exactly what they were is the key issue of this chapter. Insteadof advancing my own exegesis of his words, I wish to examine howothers have interpreted his thoughts.

Leenhardt the post-structuralist avant la lettre

In the mid-1970s Jim Clifford was an unknown, young academic witha first degree in literature from Harvard, where he had becomeacquainted with the first signs of what is now termed a post-structuralist approach. He went to Paris to do doctoral research on thehistory of French anthropology during the interwar period. Afterreading Do Kamo and then Leenhardt’s unpublished letters andjournals, he chose to focus on this by then neglected figure.

Throughout his book, Clifford takes pains to stress the open-ended,dynamic nature of Leenhardt’s thoughts and approaches. He portraysan assiduous, sensitive priest who engaged with anthropology in orderto further his missionary project. According to Clifford, Leenhardtwanted to comprehend the amplitude and profundity of NewCaledonian thought in order to ascertain the most effective way towardsmeaningful conversion. He did not arrive at a final, definitive positionbut ‘thought and rethought a difficult and inspiring involvement withthe Melanesian world’ (Clifford 1982: 1). Moreover, this rethinking hadreflexive effect, making Leenhardt reconsider the very nature ofChristianity and its teachings. In the process, theological abstractioncame to yield first place to the power of a concrete immediacy.

Clifford lays stress on Leenhardt’s radical notion of selves withoutunifying centres: ‘There is no experience of a defining “body”. TheMelanesian feels no physical envelope that separates a personal “inside”


from an objective “outside”’ (ibid.: 185). Rather, the self is thatamalgamation of dualistic relationships which each New Caledonianaccumulates and participates in. Instead of a centralising ‘ego’, anindigene is made up of dualities, between himself and his parent, uncle,spouse, cross-cousin, clan, ancestor, totem, and so on. Leenhardt didnot speak of traditional locals as individuals but as each having a‘personage’, a relational ensemble of such dualities.

Myths are not to be seen as stories but as geographically rooted providersof images which, through juxtaposition, enable locals to experiencecomplex emotional states. Thus myths are not so much narrated as lived.Especially at heightened, ritualised moments, a New Caledonian mayparticipate, via these dualities, in a socio-mythic space, where time anddistance, in conventional Western terms, are collapsed, transcended.

Leenhardt wished to forge a Western vocabulary to render thedistinctiveness of New Caledonian ways. He wished, for instance, toemphasise the lack of distinction for indigenes between thought, on theone hand, and expression and concrete action on the other. In modernparlance, he wanted to emphasise the illocutionary dimension ofutterance. Here, saying is doing, and utterance a speech act. For him‘parole’, which his English commentators have translated as ‘the word’ or‘words’, dissolved the conventional gap between speech and language.But he wished to give ‘parole’ a much broader compass than thoselinguistic dimensions. Not tied to elocution, it was positioned moreconcretely in gestures. While it could not be separated from thinking,‘thought’ was here understood as solidified emotion rather than intellect.In sum, this remarkably wide version of ‘parole’ was whatever manifeststhe person, was more likely to be exemplified by things than words, andwas to be understood as expressivity rather than structure.

Clifford’s analysis is subtle and gracefully couched. Indeed, he doesthe job so well that almost all subsequent Anglophone re-interpreters ofLeenhardt are to a certain extent influenced by his reading of the man’swork. However, what is of particular interest here is his stress on thecontemporary relevance of Leenhardt, whom he portrays as a bypassedtrailblazer: ‘What is important ... is that his mistrust of systematicclosure, his emphasis on reciprocal interpretation and culturalexpressivity, placed him on the boundary of a science that, since Tylor,had concerned itself with the study of whole, integrated ways of life inmore or less continuous development’ (ibid.: 173). Clifford’s Leenhardtis an institutionally marginal figure, whose interests and approachesdovetail remarkably well with ideas only then beginning to percolatethrough into the anthropology of the early 1980s. On a perhapsflippant, more likely designedly polemical note, Clifford even termsLeenhardt a post-structuralist, albeit one avant la lettre (ibid.: 173).


The various concerns which Clifford underlines in his biography –reciprocity, reflexivity, open-endedness, persistent provisionality, thedecentred self, and the problems of translation – are all, of course, nowclassically post-modernist matters. And the first raisers of that bannerin social anthropology were Clifford himself and his colleague GeorgeMarcus in the book they edited, Writing culture: the poetics and politics ofethnography (1986), today as notorious as it is famous.

In a recently translated book, the Parisian historian of ideas FrançoisCusset (2008) argues that in the 1980s American academicstrumpeted the ideas of French theoreticians (e.g. Derrida, Foucault,Baudrillard), which they branded ‘poststructuralism’, just as thoseapproaches were rapidly losing favour on their home turf. The very term‘poststructuralism’ is itself an American coinage; it does not exist inFrance (Wolin 2008). The postwar saw was ‘When a philosophy dies, itcomes to Oxford’. For the 1980s, replace ‘Oxford’ with ‘America’. Cussetdoes not discuss anthropology in his account, but Clifford’s importationof Leenhardt into US academe slips all too easily into this interpretationof the trans-Atlantic trade in ideas.

For these reasons, it is not overly cynical to see Clifford as in effectexploiting the figure of Leenhardt as a means to bring post-structuralism into anthropology. At the same time, of course, it is alsoan exploitation of Leenhardt in order to advance his own academiccareer, to be recognised as a standard-bearer of the thenanthropological avant-garde. I contend that this interpretation is nottoo cynical because Sangren, in one of the first incisive critiques of post-modernism, exposed the hegemonic pretensions of Clifford andMarcus’s movement (Sangren 1988). For, by claiming the equality anddiversity of different theoretical approaches, they quietly failed to statethe superiority of one theory: their own. When Sangren links this bidfor hegemony to their own desire for institutional self-advancement,Clifford, an otherwise very astute respondent, complains of ‘innuendoabout career strategies’ (Clifford 1988: 425).2

Leenhardt the phenomenologist Romantic

The American anthropologist Thomas Maschio did fieldwork in themid-1980s among the Rauto of southwestern New Britain. In hismagisterial ethnography of them, he openly acknowledges theinfluence of ‘Leenhardt’s seminal work on the character of NewCaledonian religious experience. It is an insight that to my mind hasneither been applied, nor even recognised, save tangentially, by theanthropology of religion’ (Maschio 1994: 28).

What Maschio finds so useful in Leenhardt is his conception of therole of mythic consciousness in a New Caledonian’s progress towards


authentic personhood. Dissatisfied with well-established anthropologicalapproaches which stress the discursive interpretation of symbolicrepresentation, Maschio is stimulated by Leenhardt’s phenomenologicalfocus. Following his exemplar, Maschio does not wish to comprehendlocal ways according to the structuring constraints of a Westernsemiology, but to explore the ways Rauto ritual and poetic performancesenable the enactment, expression and invention of the self. Thus,openly inspired by Leenhardt, he sees Rauto religious performance andexpression as ways to transform nostalgia, anger and other emotionsinto a style of cultural memory, one which juxtaposes patternedcultural meaning with an emotional feeling of plenitude.

While he acknowledges the use of Clifford’s reading of Leenhardt,Maschio goes further, for he wishes to associate Leenhardt with alongstanding strand within Western ideas. First he points out howLeenhardt’s portrayal of the link between individuation and mythicconsciousness resonates with Jung’s characterisation of theindividuation process. The ideas of both intellectuals about the mythicimage being ‘somehow basic to human existence’ dovetail with Barthes’conception of it as retaining the obtuse and ‘sometimes obscuremeaning of primary intuitive experience ... as a way of knowing thateschews clear conceptual thought and language’ (ibid.: 31). WhileLeenhardt and Jung saw mythic thought as central, they both regardedit as only the foundation for individuation. Both considered that mythicthinking needed to be coupled with rationality ‘so that a person couldbring about a psychologically balanced form of individuation’ (ibid.:220). Openly opposing them to modern symbolic anthropologists whoemploy metaphors of reading, writing and editing, Maschio groupsLeenhardt and Jung into a long line of Romantic thinkers, especiallyVico, who were concerned with the relations between image, memoryand experience (ibid.: 33).3

In other words, in order to understand Leenhardt within hisplenitude, Maschio places him within a strand of Western thoughtwhich goes back centuries while retaining much relevance foranthropologists of religion today. According to Maschio, Leenhardt canyet be a guide for our times.

Leenhardt the existentialist

Deborah van Heekeren sees Leenhardt as a crypto-existentialist whoembedded a Heideggerian perspective into his ethnography. SinceLeenhardt did not explicitly acknowledge any debts to the philosopher,she grounds her argument on several struts: Clifford’s note aboutLeenhardt’s regular conversations with a translator of Heidegger (Clifford1982: 250, n. 39); the ethnographer’s concern to focus on experience


and thus, if need be, to transcend western categories of analysis; and theparallels between his concept of mythic participation and Heidegger’snotion of Dasein, ‘an ontological term which he employed to designateman in respect of his being’ (van Heekeren 2004: 433).

Heidegger argued that the essential nature of existence could berevealed in certain ‘limit situations’, such as struggle and death. Onlyin these situations might the presence of being become evident. To vanHeekeren, these situations, which she designates ‘existential events’,resonate with Leenhardt’s stress on clusters of participations in whichmythic thought is lived. Just as the philosopher thought authentic beingwas discovered in a situation such as death, so the ethnographerconsidered that New Caledonian authenticity was discovered in socio-mythic events: ‘a moment experienced as passion/flight/transformation,or perhaps fear or despair, that is universally recognised yet deeply andindividually experienced’ (ibid.: 438). Similarly, she sees Leenhardt’sidea of the collapse in socio-mythic space of distance between peopleand things as strikingly similar to Sartre’s comment on the annihilationof distance between subject and object (ibid.: 438). She concludes thatboth Leenhardt and Heidegger recognised a mode of being thatparticipates with the world. This being-with-the-world is at the same timea being-with-others. However, to be with others authentically, one has toexperience the mode of relation to the other which promotes existencein the full sense (ibid.: 446). If Heidegger was concerned about thesurvival of authenticity in the modern world, Leenhardt was similarlytroubled about the continuation of ‘plenitude’ in colonialist times.

The value of van Heekeren’s approach is heuristic, or pragmatic: it isto be judged in terms of its results. A fieldworker of Papua New Guinea,she claims, ‘I have been particularly impressed by the way Leenhardt’swriting resonates with the work of indigenous authors in so far as eachseems to capture a fundamental sense of being that other models elide’(ibid.: 432). Analysing her own field-data, she wishes to demonstratethat mythic dimensions to ontology can also be uncovered in other partsof Melanesia. To her, this is the greatest legacy of Leenhardt: hisinterpretation of myth as more than story or charter, leading to an‘outstanding philosophy of Melanesian existence’ (ibid.: 433).

The problem with approaches such as van Heekeren’s is that the gameis rarely worth the candle. It is all too easy to speculate on possibleprecursors whom Leenhardt might have read and who might haveinfluenced him. Heidegger is one candidate. Bergson and Mach are others.But how to choose between them, unless we have substantiated evidence?4


Leenhardt the decentred soul

If there is a key volume in Marilyn Strathern’s oeuvre, it is her Gender ofthe gift (Strathern 1988). And if there is a key, much commented-uponidea in that book, it is her portrayal of Melanesians as ‘dividuals’, anidea first formulated by Marriott in his South Asian ethnography (ibid.:348–49, n. 7). Recognising dividuality means regarding persons inprimarily relational terms, constructed as the plural and composite siteof the relationships that produced them (ibid.: 13). While praising the‘brilliance’ of Leenhardt for charting the extent to which personsappear through their relationships, Strathern bluntly states that hemade the mistake of thinking there is a centre, albeit an empty one, toNew Caledonian personhood (ibid.: 268–69). Instead she wishes tospeak almost exclusively of relationships.

Bill Maurer, who wishes to promote a ‘lateral anthropology’, stronglyqueries Strathern’s desire to call Leenhardt mistaken: ‘Refusing thestructure of error, I would simply add that this language lies alongsideothers, where mistakes can be made and where the very idea of amistake can be obviated by multiple and polyvalent emergences’(Maurer 2005: 19). Edward LiPuma, a fellow Melanesianist, criticisesher on precisely this point. He contends that she has exploited themuch-used tactic of criticising predecessors on the grounds that theyhave been compromised by ethnocentric presuppositions, in this caseLeenhardt. He considers that the power of her argument rests on ausually unexamined ‘theory of anthropological “progress” based onincreasing epistemological awareness of the uniqueness of other’scultures’ (LiPuma 1998: 55). In other words, he regards her as a Whig.

Eric Hirsch, a Melanesianist colleague of Strathern’s, counter-arguesthat LiPuma has misrepresented her. She did not ignore the individualaspect of personhood. As Hirsch emphasises, Strathern states early onin her book: ‘Far from being regarded as unique entities, Melanesianpersons are as dividually as they are individually conceived’ (ibid.: 13,Hirsch 2001: 140).5 Even though we take this point, it does not puncturethe power of LiPuma’s criticism of Strathern’s discriminating dismissalof Leenhardt. I might add that if, as Hirsch claims, ‘her interest is not todeny the relevance of the individual to Melanesian social life’ (ibid.: 140),how are we meant to conceive of individuals without centres? It wouldseem difficult for Hirsch to answer that convincingly without having tomodulate Strathern’s criticism of Leenhardt.


Towards an assessment?

Leenhardt as post-structualist (Clifford), romantic (Maschio),existentialist (van Heekeren), decentred soul (Strathern): one man, fouranthropologists, four rather different interpretations of the same work:why the diversity? We have already mentioned the felt need of someacademics to cut their own road, and to be seen to be doing so. In otherwords, it pays the ambitious to come up with an individual interpretationof Leenhardt which advances their own interests. But there is a furtherreason why he is such a suitable candidate for multiple interpretation.

Do Kamo is not easy to read. It is at times a confusing jumble ofinconsistent language and unannounced shifts in style. Its Englishtranslator stresses its labile vocabulary and eccentric organisation(Gulati 1979; see also Clifford 1982: 172). Bensa claims that Leenhardtproceeds ‘less by progressive analyses than by inspired affirmations’(Bensa 2000: 95). Crapanzano notes that in Do Kamo, Leenhardt canconfuse role and person, and fails to separate the concept of the personfrom the experience of being a person. He also highlights ‘the suddenintrusion of the concrete in the abstract and the abstract in theconcrete, indeed the idiosyncrasy of its language’ (Crapanzano 1979:xvi, xxiv). Jamin, while sympathetic to Leenhardt’s desire to producean effective translation, queries the limits of his endeavour:

By turning translation upside down, that is to say by trying to adapt andbend his own language to that of others, he certainly reinstated the originalgrammar, but immediately risked a loss of meaning: because he so wanted to learn, say and transcribe difference, he risked making itincomprehensible. (Jamin 1978: 56)

In other words, by attempting to render into French the almostunnameable, Leenhardt produces the almost unreadable.

Several criticise his inconsistent use of ‘myth’. Crapanzano contendsthat he confuses ‘a mode of knowledge with a cultural reality of somenever quite clear status’. According to him, Leenhardt elided aconstruct, analytically derived by himself, from observed behaviourwith an experientially felt reality portrayed by him and then attributedto New Caledonians (Crapanzano 1979: xviii–xix; see also Young 1983:15). To put that another way, how much of Do Kamo’s ‘mythicconsciousness’ represents indigenous patterns of thought, and howmuch is Leenhardt’s own mystifying creation? One reason this questionis so difficult to answer is that, unlike Malinowski, Leenhardt did notprovide the social contexts associated with individual myths (Young1983: 15–16). Thus several of his analyses can appear more literarythan social, and some even anecdotal (Crapanzano 1979: xx).


Furthermore, Leenhardt’s central notion of participation can beeasily questioned. Thus Naepels points out that Leenhardt, whenspeaking of a related pair, such as an uncle and his nephew, claims thata local does not ‘perceive two persons but the unity of that pair, withouttruly being conscious of the individuals composing it’. As he argues,though indigenes might speak of their ‘parents’, that does not meanthat they are not clearly aware of the individual existence of theirmothers and fathers (Naepels 2007: 77). Naepels, himself a veteranfieldworker of New Caledonia, here speaks with some authority.

Some have argued that Leenhardt lacked a modern appreciation ofmetaphor, in effect denying New Caledonians’ ability for metaphoricalplay and beauty. Instead he tended to confound local descriptions ofexperience with evidence for description (Young 1983: 15; see alsoNaepels 2007: 79). This criticism could stand as a particular instanceof the more general fear that Leenhardt appears to give a determiningweight to linguistic structures as reflecting ways of thought (Naepels2007: 71). It is therefore of great concern to learn that Leenhardt, eventhough he had lived among the locals for so long, could still get it badlywrong. Clifford starts the key chapter of his book, Chapter Eleven‘Structures of the person’, with the following:

Leenhardt never tired of recounting a conversation with Boesoou Erijisi inwhich he proposed to his oldest convert: ‘In short, what we’ve brought intoyour thinking is the notion of spirit.’ To which came the correction: ‘Spirit?Bah! We’ve always known about the spirit. What you brought was the body.’

The largest part of Leenhardt’s ethnological theorizing was direct orindirect exegesis of this retort. (Clifford 1982: 172)

Clifford makes such a grand claim because, he states, the response madeLeenhardt revaluate the applicability of the Western notion of ‘body’ toNew Caledonians, and so enabled him to conceive of indigenouspersonages as transcending European ideas of corporal limits. It is allthe more worrying, therefore, that Leenhardt can here be accused ofmistranslation. Naepels, who is able to read Leenhardt’s texts in theirindigenous languages, contends that he incorrectly translated the termused by Boesoou Erijisi, ‘karo’, as ‘body’ (corps). The nata made hisremark in the course of a conversation about Paul’s Epistle to theRomans, in which he argued that what the missionaries had broughtwas the flesh, in the Pauline sense of the carnal, the fleshy, the sinful.Naepels is emphatic that Leenhardt’s transcription of his friend’s wordsdoes not justify his claim that New Caledonians had been unaware oftheir bodies as bodies, as the support or prop of individual existence:


When Leenhardt, unconscious of the conceptual difference between fleshand body, quoted the decontextualized translation of this dialogue, heturned his very subtle indigenous pupil into the involuntary witness of theprimitive absence of the consciousness of self. (Naepels 2007: 84)

If Naepels is correct, a central feature of Do Kamo is exposed as theoutcome of a single miscomprehension.

It is criticisms such as these which enable some of Leenhardt’scommentators to characterise him in less than eulogistic terms. ThusL’Estoile points out that his work in places suffers from an evolutionist,truly primitivist vision of New Caledonians (L’Estoile 2007: 29–32),while even Clifford, otherwise a panegyrist of Leenhardt, feels forced atone point to admit that ‘his rather mystical Canaque is an exaggeration’(Clifford 1982: 137). Not surprisingly, then, at least one Melanesianisthas accused Leenhardt of explicit racism (Guidieri 1984: 75, n. 1).Alban Bensa, a French anthropologist of New Caledonia, has questionedmuch of Leenhardt’s ethnographic analysis.6 But he goes further.According to him, Parisian intellectuals cited Do Kamo so often becauseits portrait of Canaques dovetailed neatly with the long-groundedstereotype of the ‘primitive’, still inhabiting an Eden long denied to us.In other words Leenhardt’s language seduced an anguishedintelligentsia. Bensa concludes, somewhat rhetorically, by asking, howLeenhardt, who had done such lengthy, painstaking fieldwork, could‘come up with a vision of the Canaque world so strange and so distantfrom those of his predecessors and of his successors?’ (Bensa 2000: 97).

Jean Guiart, Leenhardt’s most distinguished pupil, stung by thecriticisms of his Parisian colleagues, tried to defend his revered teacherby writing a short biography of him. Unintentionally, in his hagiographyhe provides his enemies with further arms. For he reveals that, inDocuments néo-calédoniens (1932), Leenhardt quoted long extracts,without any attribution whatsoever, from the notebooks of one especiallytalented nata (Guiart 1998). As Naepels notes, this is surely plagiarismand verges dangerously close to colonial exploitation (Naepels 2000).

While Leenhardt’s work is now much utilised and praised by Canaquesin their struggle to regain a sense of cultural dignity, Bensa worries that theideas in Do Kamo are also being used to explain, ‘against all evidence’, thescholastic failure and economic difficulties of native New Caledonians.Far from assisting politicised Canaques, Do Kamo may be exploited to showthe indigenes up as living in an archaic world (Bensa 2000: 94).7

Leenhardt’s eulogisers might wish us to view him as an exemplary,activist anthropologist, prepared to fight against the worst excesses ofcolonialism, but, according to his critics, he was still very much a manof his time. However hard he was prepared to rethink his position and


understanding of New Caledonian ways, he was, unsurprisingly, notable to free himself of all the prejudices of his varied milieux. Indeed, isit unfair to think that he could?

Towards a lateral history of anthropology

Every complex thinker, over the course of their life, stakes a variety ofpositions. Moreover, the different facets of his or her work can be pickedup by commentators in a plurality of ways. We have already seenLeenhardt in four different guises. More, of course, are possible: such asLeenhardt as a ‘pioneer of ethnolinguistics’ (Calame-Griaule 1978: 43,Laroche 1978: 46) or as ‘the first French anthropologist to carry outserious ethnography’ (Cavignac 2001: 8); he is praised byphenomenologists and mythologists for his contribution to their fields,and by clinical psychologists as ‘one of the very few anthropologists whoinfluenced the psychological and psychiatric theories of his time’(Dardel 1954, Garelli 1995, Mouchenik 2005).8 Others are surelypossible.9 This interpretative process is further enriched if the thinkerdoes not write in a pellucid prose. Given the spasmodic abstruseness ofDo Kamo, it is not surprising that many of his commentators prefacetheir remarks with ‘If I understand him correctly’. Of course, this lackof clarity only fuels the exegetic challenge, facilitating multiple andoften competing interpretations.

The obvious question most commentators here raise is why, if theirman is so worthy of critical attention, was his work marginalised for solong? Jamin, for instance, says Leenhardt was ‘unappreciated, forgotten,neglected’ (Jamin 1978: 55). No commentator suggests that there wasany discernible, deliberate intention by succeeding generations of Frenchanthropologists to sideline him and his oeuvre. Rather, structuralismcame very strongly to the fore, and his approach simply went out offashion. Lévi-Strauss, who replaced Leenhardt at the École Pratique desHautes Études, practised a very different kind of anthropology, muchmore rationalist in style and with almost no concern forphenomenological issues. He did not need to criticise Leenhardt – he justasked an alternative, maybe complementary set of questions. Thisexplanation is all the more likely because it was precisely when the voguefor structuralism had passed and its replacements were first being toutedthat the re-evaluation of Leenhardt’s work commenced.10

If (and that is a big ‘if ’) the influence of Leenhardt’s writings can beeasily summarised for today’s anthropologists, it is as an exemplar ofexperience-rich ethnography which strives to grapple with translatingthe non-discursive. What the resurrection of his work strongly suggests


is that Whiggish approaches to the history of our subject are deeplymisleading, if not downright distorting. Imagine the difference if,instead of teaching it in a hierarchical format, grounded on an illusionof progress, we presented its past practitioners and their ideas in alateral, multi-layered manner (e.g. Maurer 2005: 19). How muchricher, more suggestive, more open-ended, more receptive to alternativestyles our subject could be!

Notes

1. My thanks to Peter Parkes for comments, and to Anne de Sales for assistance inobtaining French references. All translations from the French are by myself.

2. Marcus, in a joint reply with his fellow postmodernist promoters Michael Fischer andStephen Tyler, complains of an ‘unsavory, ad hominem charge of bad faith, a totallyunsupported charge of scheming careerists who wish merely to advance themselves’(Fischer, Marcus and Tyler 1988: 426).

3. Stephen, who did fieldwork in southeast Papua New Guinea, judges Leenhardt’sunderstanding of New Caledonians’ mythic participation to have been ‘romantic’(Stephen 1995: 141).

4. Given that several commentators on Leenhardt note parallels between his work andMalinowski’s, especially regarding the former’s concern with myth and the latter’swith magic, it might be entertaining to play with the idea of Mach (on whomMalinowski wrote his doctorate) influencing both, but in divergent ways. But wouldthe enterprise rise above the level of entertainment?

5. Hemer, who quotes the same phrase, states that it is only on close reading of your bookthat one can spot that ‘Strathern’s analysis does allow space for non-relational aspectsof Melanesian personhood’. In a review of recent work on personhood in the region,she notes the divergence of subsequent Melanesianists from Strathern’s approach andstresses the need to make a distinction between individuality (recognised and perhapsvalued) and individualism (recognised and not valued) (Hemer 2008).

6. See, for example, Bensa’s critique of Leenhardt’s approach to totemism (Bensa 1990)and of his conception of the relation between grammatical categories and forms ofthought (Bensa 1995); also Bensa and Leblic 2000. For further criticisms ofLeenhardt’s notions of totemism, see Naepels (1998), Salomon (2000).

7. For examples of Leenhardt’s relevance to contemporary New Caledonia, seehttp://www.adck.nc/html_en/programme/mwavee.pho?num=38 (accessed 20 April2005). Mouchenik, a clinical psychologist, is concerned that Leenhardt’s ideasdirectly influenced psychiatric ideas in New Caledonia until recently, allowing mostlocal psychiatrists to neglect ‘the more multiple evolutions of contemporarypsychoanalysis: familial, group, and transcultural’ (Mouchenik 2006: 664).

8. Dardel, one of the earliest exponents of social geography in France, was alsoLeenhardt’s brother-in-law.

9. There is, for instance, the literary Leenhardt, in a short story, ‘Boys smell like oranges’,by Guy Davenport, a university friend of Needham’s (Davenport 1996). Needhamhimself praised Leenhardt for Do Kamo, which he regarded as an exemplaryethnography of a particular concept (Needham 1972: 152–53). When I asked him,in 2006, where he had first heard of Leenhardt, he said he did not know. Leenhardtwas someone he seemed to have learnt of from very early on (Needham, personalcommunication).


10. My colleague Peter Parkes comments: ‘Lévi-Strauss’s refusal to speak of Leenhardt isstill curious: I don’t think he was anti-phenomenology, at least with respect toMerleau-Ponty, more against what he called “shop-girl philosophy”, maybe regardinghis EHESS predecessor as a sentimental precursor to Sartre’ (Parkes, personalcommunication.). In his reply dated 23 September 2008 to a letter of my own abouthis silence, Professor Lévi-Strauss stated, ‘I am unfortunately too old…to tryanswering your query’. The old fox!

References

Bensa, A. 1990. Des ancêtres et des hommes: introduction aux théories kanak de lanature, de l’action et de l’histoire (Nouvelle-Calédonie), in R. Boulay (ed.), De jade etde nacre: patrimoine artistique kanak, Paris: Réunion des Musées nationaux.

——— 1995. Chroniques kanak: l’ethnologie en marche, Paris: Ethnies-Document.——— 2000. Les ‘réalités mythiques’ de Maurice Leenhardt, Gradhiva, 27: 93–97.Bensa, A. and I. Leblic (eds). 2000. En pays Kanak: ethnologie, linguistique, archéologie,

histoire de la Nouvelle Calédonie, Paris: Éditions de la Maison des Sciences de l’Homme.Calame-Griaule, G. 1978. Maurice Leenhardt, pionnier de l’ethnolinguistique, Journal de

la Société des Océanistes, 34: 43–44.Cavignac, J.A. 2001. Maurice Leenhardt e o início de pesquisa de campo na antropologia

francesa, cchla.ufrn.br/tapera/equipe/julie/maurice_leenhardt.pdf (accessed 4 July2008).

Clifford, J. 1982. Person and myth: Maurice Leenhardt in the Melanesian world, Berkeley:University of California Press.

——— 1988. Comment (on Sangren 1988), Current Anthropology, 29: 424–25.Clifford, J. and G. Marcus (eds). 1986. Writing culture: the poetics and politics of ethnography,

Berkeley: University of California Press.Crapanzano, V. 1979. Preface to English translation of Do Kamo by Maurice Leenhardt,

Chicago: The University of Chicago Press.Cusset, F. 2008. French theory: how Foucault, Derrida, Deleuze, & Co. transformed the

intellectual life of the United States, Minneapolis: University of Minnesota Press.Dardel, E. 1954. The mythic: according to the ethnological work of Maurice Leenhardt,

Diogenes, 7: 33–51.Davenport, G. 1996. The Cardiff team: ten stories. New York: New Directions.Fischer, M. and G. Marcus with S. Tyler. 1988. Comment (on Sangren 1988), Current

Anthropology, 29: 426–27.Garelli, J. 1995. La phénoménologie du jugement et la dimension ‘cosmomorphique’ du

corps chez les Canaques, selon Maurice Leenhardt, Droits et cultures, 29: 255–74.Guiart, J. 1998. Maurice Leenhardt: le lien d’un homme avec un peuple qui ne voulait pas

mourir, Nouméa: Le Rocher-à-la-Voile.Guidieri, R. 1984. L’abondance des pauvres: six aperçus critiques sur l’anthropologie, Paris:

Seuil.Gulati, B.M. 1979. Translator’s note, to English translation of Do Kamo by Maurice

Leenhardt, Chicago: The University of Chicago Press.Hemer, S.R. 2008. Poit, personhood, place and mobility in Lihir, Papua New Guinea,

Oceania, 78(1): 109–25.Hirsch, E. 2001. When was modernity in Melanesia?, Social Anthropology, 9(2): 131–46.Jamin, J. 1978. De l’identité à la difference: la personne colonisée, Journal de la Société des

Océanistes, 34: 51–56.Laroche, M.-C. 1978. L’enseignement de Maurice Leenhardt, Journal de la Société des

Océanistes, 34: 45–48.


Leenhardt, M. 1930. Notes d’ethnologie néo-calédonienne, Paris: Institut d’Ethnologie(Travaux et Mémoires, 8).

——— 1932. Documents néo-calédoniens, Paris: Institut d’Ethnologie (Travaux et Mémoires, 9).——— 1935. Vocabulaire et grammaire de la langue houaïlou, Paris: Institut d’Ethnologie

(Travaux et Mémoires, 10).——— 1937. Gens de la Grande Terre: Nouvelle Calédonie, Paris: Gallimard.——— 1946. Langues et dialects de l’Austro-Melanésie, Paris: Institut d’Ethnologie

(Travaux et Mémoires, 46).——— 1947. Do Kamo: la personne et le mythe dans le monde mélanésien, Paris: Gallimard.L’Estoile, B. de. 2007. Une politique de l’âme: ethnologies et humanisme colonial, in M.

Naepels and C. Salomon (eds), Terrains et destins de Maurice Leenhardt, Paris: EHESS.LiPuma, E. 1998. Modernity and forms of personhood in Melanesia, in M. Lambek and

A. Strathern (eds), Bodies and persons: comparative perspectives from Africa andMelanesia, Cambridge: Cambridge University Press.

MacClancy, J. 1986. Unconventional character and disciplinary convention: John Layard,Jungian and anthropologist, in G. Stocking (ed.), Malinowski, Rivers, Benedict andothers: essays on culture and personality (History of Anthropology 4), Madison:University of Wisconsin Press.

——— 1995. Brief encounter: the meeting, in Mass Observation, of British surrealism andpopular anthropology, Journal of the Royal Anthropological Institute, 1(3), 495–512.

——— 1996. Popularizing anthropology, in J. MacClancy and C. McDonaugh (eds),Popularizing anthropology, London: Routledge.

——— 2000. The decline of Carlism (The Basque Series), Reno: University of Nevada Press.Maschio, T. 1994. To remember the faces of the dead: the plenitude of memory in southwestern

New Britain, Madison: University of Wisconsin Press.Maurer, B. 2005. Mutual life, Limited: Islamic banking, alternative currencies, lateral reason,

Princeton: Princeton University Press.Mouchenik, Y. 2006. Maurice Leenhardt et l’invention d’une personnalité indigène en

Nouvelle-Calédonie, Annales médico-psychologiques, 164(8): 659–67.Naepels, M. 1998. Histoires de terres kanaks: conflits fonciers et rapports sociaux dans le region

de Houaïlou (Nouvelle-Calédonie), Paris: Belin.——— 2000. Review of Guiart 1998, Oceania, 70(4): 370–71.——— 2007. Notion de personne et dynamique missionaire, in M. Naepels and C.

Salomon (eds), Terrains et destins de Maurice Leenhardt, Paris: EHESS.Needham, R. 1972. Belief, language and experience, Oxford: Basil Blackwell.Salomon, C. 2000. Savoirs et pouvoirs thérapeutiques kanaks, Paris: PUF-INSERM.Sangren, S. 1988. Rhetoric and the Authority of Ethnography: ‘Post-Modernism’ and

the Social Reproduction of Texts, Current Anthropology, 29: 405–35.Stephen, M. 1995. A’aisa’s gifts: a study of magic and the self (Studies in Melanesian

Anthropology 13), Berkeley: University of California Press.Strathern, M. 1988. The gender of the gift: problems with women and problems with society

in Melanesia, Berkeley: University of California Press.van Heekeren, D. 2004. ‘Don’t tell the crocodile’: an existentialist view of Melanesian

myth, Critique of Anthropology, 24(4): 430–54.Wolin, R. 2008. The state of literary theory: America’s tolerance for French radicalism

(review of Cusset 2008), The Chronicle of Higher Education, The Chronicle Review, 13 June,http://chronicle.com/temp/reprint.php?id=ww474bgn53kn7fv2g17ytbnzwr2rxhpt(accessed 26 June 2008).

Young, M. 1983. Magicians of Manumanua: living myth in Kaluana, Berkeley: University ofCalifornia Press.


NOTES ON CONTRIBUTORS

Alban Bensa, anthropologist and Directeur d’Études at l’École desHautes Études en Sciences Sociales (IRIS), Paris, is a specialist in theKanak societies of New Caledonia, where André-Georges Haudricourtsent him on a mission starting in 1973. He has conducted ethno-linguistic and social anthropological research focused on politics andnarrative, as well as on the relationship between anthropology andhistory. Among his recent publications are La fin de l’exotisme:l’anthropologie autrement, Toulouse: Anacharsis 2006; and with DidierFassin (eds.), Les politiques de l’enquête: épreuves ethnographiques, Paris: LaDécouverte, 2008.

Margaret Buckner completed graduate studies in anthropology atthe University of Paris X-Nanterre under the direction of Professor Ericde Dampierre, whom she met as a Peace Corps volunteer in Bangassou,Central African Republic. Still a member of the Laboratoire d’ethnologieet de sociologie comparative, she now teaches cultural and linguisticanthropology at Missouri State University. She has published severalarticles on the Zande, but since 1991 has also been carrying outresearch among the Manjako of Guinea-Bissau.

Stefania Capone took her PhD from Paris X-Nanterre in 1997 andher habilitation in 2005. She is currently Directrice de recherche(Tenured Senior Researcher) at Centre National de la RechercheScientifique. She is currently a Visiting Scholar at the New YorkUniversity, and a researcher at CIRHUS, the Center for InternationalResearch in the Humanities and Social Sciences, CNRS/NYU (NewYork). She is the author of La quête de l’Afrique dans le candomblé (Paris,1999; Brazilian edition, 2004; American edition, 2010, DukeUniversity Press); and Les Yoruba du Nouveau Monde: religion, ethnicité etnationalisme noir aux États-Unis (Paris, 2005; Brazilian edition, 2009).

Girodana Charuty is a Directrice d’Études at the École Pratique desHautes Études in Paris. A Europeanist, she has worked on themedicalization of madness, Christian custom in Mediterranean Europe

and social practices of writing. Her current research is on the history ofItalian anthropology and on ethnographic knowledge developed in thecourse of the nineteenth century in the context of missionary activity.Her publications include: Le Couvent des fous: l’internement et ses usagesen Languedoc aux XIXe et XXe siècles, Paris, Flammarion, 1985; Nel paesedel tempo: antropologia dell’Europa cristiana, Naples, Liguori, 1995; Folie,mariage et mort: pratiques chrétiennes de la folie en Europe occidentale, Paris,Le Seuil, 1997; and De Martino: les vies antérieures d’un anthropologue,Aix-en-Provence and Marseille, Parenthèses/MMSH, 2009.

Anne de Sales is a Researcher in anthropology at the Centre Nationalde la Recherche Scientifique, a member of the Laboratoire d’Ethnologieet de Sociologie Comparative in Nanterre (CNRS-LESC), and a ResearchAssociate at the Institute of Social and Cultural Anthropology (Oxford).She has carried out extended periods of fieldwork in Nepal, where herstudies of shamanic practices and oral literature include the monographJe suis né de vos jeux de tambours: La religion chamanique des Magar du nord(Nanterre: Société d’ethnologie, 1991). More recently she has publisheda series of articles towards a comprehensive ethnography of the Maoistinsurrection that overthrew the royal regime in Nepal in 2006.

Paul Henley is Professor of Visual Anthropology at the University ofManchester, where he has been Director of the Granada Centre forVisual Anthropology since its foundation in 1987. Having begun hiscareer as an Amazonist specialist, he moved into visual anthropologyafter training as a documentarist at the National Film and TelevisionSchool through a scheme managed by the Royal AnthropologicalInstitute and funded by the Leverhulme Trust. He has recentlypublished The Adventure of the Real, a major study of the film methodsof Jean Rouch, with the University of Chicago Press.

Jeremy MacClancy, Professor of Social Anthropology, Oxford BrookesUniversity, has done major fieldwork in Vanuatu and the BasqueCountry. He has published extensively on the histories of anthropologyand of its interchange with literature. His latest book is Expressingidentities in the Basque arena (Oxford: James Currey, 2008).

Peter Parkes is Reader in Historical Anthropology at the Universityof Kent. His fieldwork has concentrated on the Hindu Kush region ofeastern Afghanistan and northern Pakistan, particularly among thenon-Islamic Kalasha (Kalash Kafirs) of Chitral District, NWFP,Pakistan. His research interests include mountain subsistence anddevelopment, verbal arts, visual anthropology, and historical

274 Notes on Contributors

anthropology. He has also published extensively on fosterage andadoptive kinship in Eurasia.

Robert Parkin is a Departmental Lecturer in Social Anthropology at theUniversity of Oxford. His interests include kinship, the anthropologies ofSouth Asia and eastern Europe, and the history of French anthropology.His main works regarding the latter are The dark side of humanity: the workof Robert Hertz and its legacy (Harwood, 1995), Louis Dumont andhierarchical opposition (Berghahn, 2003) and ‘The French-speakingcountries’, in Fredrik Barth et al., One discipline, four ways (The Universityof Chicago Press, 2005). He has also translated work by Henri Hubert,Robert Hertz and Louis Dumont into English, including Dumont’sIntroduction to two theories of social anthropology (Berghahn, 2006).

Laura Rival is University Lecturer in Ecological Anthropology andDevelopment, and a Fellow of Linacre College. Her doctoral research wasamong the Huaorani Indians, on whom she has published numerousarticles and two books (Hijos del Sol, Padres del Jaguar: Los Huaorani deAyer y Hoy, Abya Yala 1996, and Trekking Through History: The Huaoraniof Amazonian Ecuador, Columbia University Press 2002). Her researchinterests include the impact of development policies on indigenouspeoples; Amerindian conceptualisations of nature and society; andnationalism, citizenship and state education in Latin America.

Peter Rivière is Professor Emeritus of Social Anthropology at theUniversity of Oxford and Emeritus Fellow of Linacre College, Oxford.He has had a life-long interest in the Native Peoples of Lowland SouthAmerica. His main works include Marriage among the Trio (1969),Individual and Society in Guiana (1984), Absent-minded Imperialism(1995), and a two-volume work, The Guiana Travels of RobertSchomburgk 1835–1844 (2006). He has recently edited A History ofOxford Anthropology (2007), published by Berghahn.

Gérard Toffin is an anthropologist, Directeur de recherche at CNRSand a specialist on the Himalayas. He has worked on the materialculture, social structures and religions of several ethnic groups in Nepal,in particular the Newar of the Kathmandu valley. His main publicationsare Man and his house in the Himalayas (1991), Le palais et letemple (1993), L’ethnologie: la quête de l’autre (2005) and Newar society:city, village and periphery (2007). He is currently preparing a book on amajor Nepali royal festival and is interested in the anthropology of space.

Notes on contributors 275

Aabian (novices), 185Académie des Sciences de Paris, 127acculturation, 16, 179, 187action, 197, 261action, social, 19, 240activism, political, 20‘ada (custom), 64adatrecht, 62, 68advocacy, 57, 59affinity, 241Africa, 14, 16, 21, 77, 80, 103, 105,

106, 109, 119, 153, 171, 182–8,190–2, 207, 225East, 226French West, 62South, 19, 22, 249West, 78, 79, 81–3, 93, 97

Afro-Americanism, 171agency, 251agriculture, 204–6, 208, 212, 213

swidden (slash and burn), 208 Aguaruna, 137Ait ‘Amar, 49, 59Ait Iraten (Kabyles), 48, 49, 52, 57, 61Albania, 67Algeria, 11, 12, 27, 45, 47, 60, 65, 66Algiers, 47, 48, 49, 50, 51, 58, 60, 68‘alim (ulema) (religious scholar[s), 49alliance, prescriptive, 241Alliance Evangélique Universelle, 260alliance theory, 238alterity, 9, 29, 156, 171Altiplano, 152, 157, 165, 166Amazon, 7, 23, 143America, 22, 182, 225, 242

Black, 171Latin, 143South, 142, 154, 163, 225

American Man, 138–40Amerindians, 21, 148, 163amin (president), 54, 61

‘anaya (protection), 54, 57, 63Ancoaquil, 166Andes, 125, 136, 139, 140, 142, 143, 164Anduze, 173Angola, 180, 193animism, 159Année du Patrimoine, 25Année sociologique, 6, 8, 22, 26, 31, 160,

192, 243, 258Anthropologie, 28anthropology

Anglo-Saxon, 76, 81, 83, 86, 87, 95–100, 143, 241, 242, 251

armchair, 43, 46biological, 144British, 2, 6, 8, 26, 109, 162, 198,

211, 212cognitive, 9cultural, 8dialogical, 86, 95four-field, 105French, passim lateral, 265, 269–70physical, 3, 29, 105, 128, 130, 136,

139, 141, 144, 147shared, 86, 96, 97structural, 144visual, 75–100

anthropometry, 128, 133, 138, 139, 142Antioquia, 131antisemitism, 41anti-structuralism, 17, 20Aosta, 199apartheid, 249apegado, 134–5Apulia, 39Arabic, 48Arabs, 134archaeology, 8, 129, 138, 139, 143,

198, 233archives, 27Argentina, 15, 151, 152, 157

SUBJECT INDEX

Argonauts of the Western Pacific, 99Armenians, 16, 176art, 171, 172, 175, 178art, history of, 236Art Nègre, 77artefacts, 145artha, 248Arts populaires et loisirs ouvriers, 34Aryans, 239Ashluslay, 154Asia, 143, 224, 225, 229, 230

South, 208Southeast, 197, 200, 208, 209, 210,

215, 226, 227Asie du Sud-Est et le Monde Indonésien

(ASEMI), 210Assam, 200Astrophel and Stella, 1asymmetry, 118atlases, linguistic, 210Auraucanians, 156, 165Aurès, 10Australians, 143, 148, 153, 241Austroasiatic, 209Aymara, 153, 164Azande history and political institutions, 114

Bbabalawo (diviner), 185, 186Bahia, 180, 181, 184, 185, 187, 192, 193Bahnar, 4ballet, 91Bandia, 113–16, 120, 122Bandiagara Cliffs, 77, 82Bangassou, 103, 104, 107–9, 113, 114, 120Bangladesh, 199Bangui, 109bantu, 193barbarians, 233Baroque, 178Bastidiana, 171Belém de Pará, 187belief, 174Bella, 84Bendo, 120Bengal, East, 211, 213Bénin, 172, 186Berbers, 12, 45, 46, 48, 60, 62, 66–8Bering Straits, 143Bernice Bishop Museum, 152Beyliks, 59bia (soul etc.), 93

Bible, 223, 258Big Men, 229binaries (binary oppositions), 239, 245,

246, 251biology, 30, 133, 142, 219, 232, 233bitti (drum), 89, 91Blénod-les-Toul, 126blood groups, 142Bocage, 39body, 23, 41, 42, 206, 230, 231, 260, 267Bogatá, 131Bolivia, 151–3, 157, 166Bonneville, 30borí, 185, 193Bororo, 160, 177, 236botany, 208, 213, 230Bourg la Reine, 34Bourgogne, 39Brahmans, 239, 240, 245–8Brantes, 201Brazil, 15, 153, 164, 171, 177–83, 187,

188, 190–3Bresle Valley, 213Briançon, 36brideprice, 67bridewealth, 114Brotas, 184Bureau of American Ethnology, 153Bureau Politique des Affarires Arabes

(bureaux arabes), 47, 50, 53, 58, 66, 67Burma, 197, 200, 203, 209–11Burmese, 17, 216bush school, 30

CCak, 200, 201, 206, 209, 216Calchaquí, 151, 157, 158, 161Canaques, 23, 258, 268; see also Kanakscandomblé, 16, 172, 177–81, 184, 187,

188, 191–3candomblé de caboclo, 193cannibalism, 156, 158canons, 46Carhuauz, 164Caribbean, 140, 187Cartesianism, 171, 183cartography, 34Casa Branca, 193Casa de Oxumarê, 193caste, 19, 21, 160, 183, 202, 240, 241,

245, 247, 249catalogues, 35

278 Subject index

categories, 230Catholicism, 134Cauca, 131Caucasus, 67causality, 231Central African Republic, 14, 103, 104, 107Centre d’Anthropologie de Toulouse, 40Centre d’Etudes de l’Inde et de l’Asie du

Sud, 237Centre d’Etudes Indiennes en Sciences

Sociales, 237, 243–4Centre d’Etudes Sociologiques, 106Centre de Documentation et de

Recherches sur l’Asie du Sud-Est et leMonde Indonésien (Cedrasemi), 200

Centre National de la RechercheScientifique (CNRS), 8, 78, 89, 93,106, 108, 193, 200, 237

Cévennes, 173charivari, 41, 43chiasmus, 183Chicago, 107Chile, 151Chin, 209China, 210, 220, 223, 224, 227, 229Chinese, 199, 211Chipaya, 152, 157Chiriguano, 152, 157, 158, 160Chittagong Hills, 197, 199–200, 209Christ, 125Christianity, 39, 41, 43, 59, 67, 91, 140,

224, 242, 260Christmas, 41Chronicle of a summer, 89, 92Chroniques, 11, 28–32, 35ciné-eye, 93ciné-poems, 78ciné-trance, 13, 16, 21, 75, 89–92, 94,

95–6cinema, ethnographic, 75–100cinema-vérité, 92, 93, 94circumcision, female, 21class, 249Clermont-Ferrand, 178Clisthenes, 120Code Civil, 55, 56, 57Code Pénal, 57coffee, 116cognition, 211Collège de France, 6, 28, 198, 200–1, 208Collège de Sociologie, 182, 235Colombia, 15, 131, 138, 140, 141, 143, 147

Colombian Society of Ethnology, 131Colonial Exhibition (French), 7, 9colonialism, 9, 12, 14, 21, 46, 50, 60–5,

80, 84, 97, 117, 129, 134, 135, 207,255, 258–60, 264, 268

Comité de Vigilance des IntellectuelsAntifascistes, 131

Commune, Paris, 60comparison, 11, 38compartmentalization, principle of, 176,

192concierto, 134–5conflict, 240congo, 193Congress of Popular Art, 28consciousness, collective, 175, 189Constantople, 235Contributions to Indian Sociology, 238,

240, 243cosmology, 81, 83, 246, 248coutumiers, 5, 61Creoles, 143, 144cults, 21, 81, 135, 181, 185, 187, 191, 193cultural relativism, 8culture

contact, 179heroes, 156material, 133, 136, 138, 140–2, 147,

158, 159, 205, 206, 208and nature, 18and personality school, 199

DDahomey, 186Dakar, 79Dakar-Djibouti Expedition, 7, 32, 77dance, 22, 38, 81, 85Dasein, 264Dauphinois, 36deduction, 2, 171Dellys, 50democracy, 50, 58depth levels, 189depth sociology, 17, 190descent theory, 238dharma, 248Diaguita, 151dichotomies, 63Dieu d’eau, 98diffusionism, 7, 17, 20, 140, 142, 143,

210, 213, 232, 238, 239, 240, 251,255

Subject index 279

diglossia, 40Dionysian, 91, 96Direct Cinema Group, 87–8, 92disability, 250discourse, 14discrimination, 249, 250distinction, 246, 251dividuals, 265Do Kamo, 259, 260, 266, 268, 269, 270Do Neva, 258, 259Dogon, 16, 77–85, 98, 99, 160, 182domestication, 227Dra el-Mizan, 48, 52, 67Dravidian, 239drawing, 33dreams, 182dualism, 18, 224, 261Dutch East Indies, 62

EEaster Island, 152, 153, 154, 155, 157,

158, 159, 162, 163Ecole des Annales, 212Ecole des Chartes, 151Ecole des Hautes Etudes en Sciences

Sociales (EHESS), 197, 198, 271Ecole du Louvre, 28, 236Ecole Française d’Extrême-Orient, 5, 198,

204Ecole Française de Sociologie, 229–30Ecole Nationale des Langues Orientales

Vivantes, 25, 151, 161, 215Ecole Polytechnique, 47, 56, 58, 235Ecole Pratique des Hautes Etudes (EPHE),

25, 27, 108, 130, 151, 153, 188,190, 197, 200, 202, 205, 210, 237,259, 269

Ecole Speciale des Langues Orientales, 215Ecuador, 15, 125, 127, 129, 132, 138–

47education, 204efon, 193efun (chalk), 187egalitarianism, 19, 244–50egun (revanent), 187Empire, French, 4, 7Empire, Second, 60, 64empiricism, British, 2, 95encompassment (of contrary etc.), 239,

246, 248, 251Encyclopédie, 25epilepsy, 41

Epiphany, 41Equipe de Recherche en Anthropologie

Sociale: Morphologie, Echanges(ERASME), 237–8, 244

Escuela Nacional de Antropología, 153Essai sur la religion bambara, 81essentialism, ethnographic, 2, 3, 8, 9, 11,

17, 20, 22, 172Ethiopian church movement, 19ethnicity, 250ethnobotany, 17, 200, 208, 214ethnocentrism, 174ethnofiction, 78Ethnographie, 28ethnography

ancient, 146canonical, 12, 13, 66dialogical, 12, 13French, passim

ethnolinguistics, 209, 269ethnology, 6, 27, 28, 37, 42, 77, 103,

105, 109, 129, 130, 132, 141, 144,146, 147, 199, 200, 236

ethnopoetics, 66ethnopsychiatry, 41ethnozoology, 208Europe, 37, 242–3evolutionism, 4, 20, 213, 243, 255, 268exchange, 247existential events, 264existentialism, 8, 19, 263–4, 266experience, 16, 18, 21, 22, 29, 33, 41,

174, 221, 263, 266Exploration scientifique d’Algérie, 66

Ffacts, 2, 6, 7, 11, 15, 17, 18, 20, 21, 22,

137, 155, 161, 164, 220, 223, 233,243

facts, social, 174, 175family, 4, 134, 136, 137, 160, 203, 207Far East, 18fascism, 21, 146feeding the head, 193feitura do santo, 193Fertile Crescent, 223, 229, 231fertility, 39, 40fieldwork, passimfiqh, 49, 56First World War, 6, 43, 129, 146, 173fission, 241Florida, 153

280 Subject index

folklore, 4, 5, 11, 20, 25, 27, 29, 32, 36,37, 39, 139, 214, 243

Folklore du Dauphiné, 34food, 133, 136Fort-Napoléon, 12, 47, 48, 51, 52, 53,

57, 58, 59, 60, 61, 67, 68Fort-National, 61, 68Franco-Prussian War, 126Freudianism, 87Fulani, 84, 90functionalism, 213, 255funlade (spirit), 187fusion, 241

GGanges, 248gay, 250gender, 250Gender of the gift, 265genealogical method, 203genealogies, 202, 223, 258Geneva, 165geography, 35, 139, 177, 212, 213, 243

human, 34, 38social, 270

Germans, 11, 189, 232Germany, 21, 22, 134, 146, 237Gestapo, 131gestures, 261Gien, 199gifts, 245Godeste, 108Gold Coast, 76Gothenburg, 151, 154government, 247Gran Chaco, 152, 153, 154, 155, 156,

157, 159Grand Jeu, 235Greece, 235Greeks, 14, 119Groupement des Universités et Grand

Ecoles de France, 177Guianas, 156

Hhabitus, 233Haiti, 155, 159, 166, 213, 216Handbook of American Indian Languages, 147Handbook of South American Indians, 153Hanoi, 5, 224Hanunoo, 208, 211harp, 112, 117, 118

Harvard, 260Hautes-Alps, 35Hawaii, 152hegemony, 262hierarchy, 3, 18, 191, 224, 239, 240,

245, 246, 247, 248, 249, 250, 251,252, 259, 270

Hierarchy and marriage alliance in southIndia, 242

Himalayas, 210Hindi, 237Hispanics, 140Historical Society of Algeria, 50, 66Histories of the Berbers, 48historiography, 256history, 18, 42, 81, 105, 110, 139,

144–6, 154, 156, 159, 175, 177,201, 219, 231–3, 240, 242, 243, 255cultural, 39of ideas, 243life, 111oral, 5, 110of religions, 25, 42social, 38world, 240, 251

holism, 20, 175, 204, 239–1, 243holism, methodological, 190homicide, 57L’Homme, 162Homo Aequalis, 238Homo Hierarchicus, 238, 240, 242Honolulu, 152horticulture, 226householder, 247houses, 41, 204, 207, 213Human Relations Area Files (HRAF), 152human rights, 244humanism, 15, 144–5hurma (moral reputation), 54, 57hybridity, 140, 143Hylean Amazon Project, 153hysteria, 41

Iiaô, 180Ibarra, 147ideals, 251ideas, 221identity, 15, 19ideology, 19, 245idiophones, 112Ifá cult, 185

Subject index 281

Ifanhin, 172ijexá, 193ikhwan (religious confraternities), 59illocutionary, 261‘ilm i-qalam (Islamic jurisprudence), 60immigrants, 176Imprimerie Nationale, 199impurity, 227, 229, 240, 249; see also

pure-impureimrabden (marabouts), 59Incas, 134, 139, 140India, 18, 21, 203, 211, 235–52Indians (American), 133–6, 139, 140, 156indirect rule, 66individualism, 19, 238, 247, 248, 265, 270individuation, 263Indo-China, 62, 68Indo-European, 140, 142Indology, 242, 243Indus Valley, 154Informants

indirect, 138intermediary, 137secondary, 144

inheritance, 52, 53initiation, 172, 184, 187, 188, 191, 193Inle Lake, 200Institut d’Amérique Latine, 188Institut d’Ethnologie, 6, 14, 129, 146, 207Institut d’Etudes Politiques, 103Institut des Langues et Civilisations

Orientales, 215Institut Français d’Afrique Noire (IFAN),

79, 193Institut Français d’Anthropologie, 128–9Institut Français d’Océanie, 260Institut International de Coopération

Intellectuelle, 28, 34Institut National d’Agronomie, 219Institut National des Langues et des

Civilisations Orientales, 200Institut pour la Recherche et de

Développement (IRD), 22Institute of Social Anthropology

(Oxford), 237Instituto de Etnología, 152Instituts franco-brésiliens de Haute

Culture, 177Instruments, musical, 111intellectualists, 5, 88interdisciplinarity, 212International Labour Office, 153

internationalism, 146interpreters, 20Intha, 200, 201Inuit, 85Iroquois, 46Isère, 34Islam, 59, 60, 62, 68Italy, 6, 37, 42izerf (Berber customary law), 68

JJains, 237jama’a (public assemblies), 46, 53, 54,

57, 61, 64jazz, 77J.B. Baillière, 216Jê, 160jeje, 193Jerusalem, 123Jews, 224Jibaros, 132, 136–8jihad (holy war), 59, 60Journal d’Agronomie Tropicale et de

Botanique Appliqué, 208Journal de la Société des Américanistes, 146,

147Judeo-Christians, 221jurisprudence, 63, 64Jurjura massif, 45, 51, 61, 66

KKabylia, Kabyle, 12–13, 21, 27, 45–68Kadu, 200Kallar, Pramalai, 236, 237, 240kama, 248Kanaks (see also Canaques), 227–8karo (body), 267Kathmandu, 210, 215Kayapó, 160, 164Kengu (river), 116ketu, 193khoja (interpreter/secretary), 12, 48–9Khyang, 209kinship, 58, 112, 158, 160, 200, 203,

204, 209, 238, 240, 241, 242, 244,247

kinship, spiritual, 41kisoro (board game), 112Kontiki expedition, 159Kshatriyas, 245–6Kubenkankrey, 164Kyoto, 206

282 Subject index

LL’Anthropologie, 29L’ethnographie, 28La Kabyle et les coutumes kabyles, 45–68La Maison Rustique, 216La notion de personne en Afrique noire, 89La pensé sauvage, 27La religion et la magie songhay, 81La Tarasque, 4, 199, 237, 242Laboratoire d’Ethnologie et de Sociologie

Comparative, 108, 122Laboratoire de Sociologie de la

Connaissance, 189labour, division of 204language(s), 111, 118, 138, 140, 142,

182, 197, 199, 203, 233, 258Indo-European, 142Melanesian, 19North American, 141Romance, 35South American, 142

Langues O, 215Laos, 210–11Lascaux caves, 227Latin, 180Lausanne, 151lavagem das contas (washing of the

necklaces), 184, 192law(s), 45, 46, 48, 51–6, 58, 62, 68,

160, 191, 207, 247Le rêve plus fort que la mort, 75League of the Iroquois, 46learned societies, 3, 33Leiden, 62, 68Leningrad, 233Les demi-savants, 27Les forms élémentaires de la vie religieuse,

26, 174, 229Les langues du monde, 129Les rites de passage, 5, 26, 27Les structures élémentaires de la parenté, 239Les tabours d’avant: tourou et bitti, 89lesbian, 250levels, 239, 246, 247, 250L’exotique est quotidian, 23liberty, 248life-cycle, 204liminality, 12, 37limit situations, 264linguistics, 25, 40, 129, 130, 135, 138,

140, 142–5, 146, 147, 209–13,231–3, 267structural, 8

literature, 15, 35, 42, 126–7, 175, 179,180, 188oral, 112

Loiret, 199Lorraine, 126Lucania, 37Lycée St Louis, 219, 235

MMadabazouma, 104mãe-de-santo (chief priestess), 184, 193Maghrib, 60, 62magic, 37–9, 155, 156, 191, 192, 270magico-religious, 42Maison des Sciences de l’Homme, 237Maison Franco-Japonaise, 206Malayo-Polynesian, 143, 148, 153, 165Mali, 77Maliki, 49, 52, 56, 59, 63, 64Malvire-sur-Desle, 105ma’mara college, 67Man with a movie camera, 92–3mana, 228Mandalay, 200mandrel, 110Manuel d’ethnographie, 6, 207, 212, 216Manuel de folklore française contemporaine,

4, 25, 28, 30, 32, 36, 37, 39–43maps, 210Mapuche, 165marabouts, 59Maranhão, 187, 192Marbial Valley, 166marginal man, 179marginality, 173Marma, 200, 201, 204, 207, 208, 211,

213, 216marriage, 40, 43, 115, 134, 136, 158,

160, 201, 203, 205, 214, 249cross-cousin, 241rules, positive, 241voodoo, 159

Marxism, 2, 8, 17, 38, 87, 189, 190,202, 232, 233

Masai, 29Masques dogon, 98Massif Central, 210Mataco, 152, 154, 157, 160materiality, 219, 221, 222, 230, 233, 234matriarchy, 67Max Planck Institute of Social

Anthropology (Halle), 22Mboki, 108, 111

Subject index 283

Mbomu (river), 116meaning, 42medicine, 41, 43, 127Mediterranean, 37, 242mediums, 13Melanesia, 18, 19, 226, 229, 256, 258,

260, 264, 265, 268, 270Mélanges de sociologie religieuse et de

folklore, 31memory, 39, 177, 190, 263Mendoza, 151mentality, 183mentality, pre-logical, 176, 258Mercure de France, 11, 26, 28, 29, 32, 33,

36, 43mestizos, 132, 136, 143, 144metaphor, 13, 117–18, 263metaphysics, 41, 175, 181, 187Méthode de l’ethnographie, 81métissage, 21, 140–3, 232Mexico City, 153Middle East, 18Middle Kingdom, 231migration, 143, 156, 221mind, 231Mindoro, 208Minot, 39Minotaure, 77miscenegation, 140Mission Society, 259Mission Sociologique du Haut-Oubangui

(MSHO), 108, 122, 123missionaries, 4, 5, 19, 23, 37, 135, 156,

257, 258, 260Modern Art Week, 178modernism, modernity, 14, 87, 178,

240–4, 249moksha, 248Mon-Khmer, 209monographic approach, 201–3montage, 93Montauban, 19Moqrani, 68Morocco, 62Mu river, 200Mukhtasar, 56multiculturalism, 9Musée d’Ethnographie, 4Musée d’Histoire Naturelle, 3Musée de l’Homme, 9, 14, 22, 108, 125,

127, 130–2, 146, 147, 198, 199,204, 206, 236

Musée des Arts et Traditions Populaires,4, 25, 28, 29, 235–8, 243

Musée du Quai Branly, 9, 22, 131, 146museology, 235Muséum National d’Histoire Naturelle

(MNHN), 28, 127–30, 146, 147, 208,209

museums, 3, 9–10, 27, 141, 251music, 14, 111, 113, 117, 193Muslim (s), 51, 52, 55, 59, 62, 64, 224mysticism, 15, 16, 172–6, 178, 182–8,

191, 192myth(s), 19, 137, 138, 155–7, 160, 165,

181, 261–4, 266, 270Mythes et legends d’Australie, 26mythico-ritual, 63mythologists, 269

NNagô, 180, 186, 193Nambikwara, 177Nancy, 126Nandi, 29Nanook of the North, 85Nanterre, 14, 103, 108, 122, 123, 200nata (pastor-evangelist), 258, 267, 268National Committee of Geography, 216National Film Board of Canada, 87National Institute of Anthropology, 147National Institute of Ethnology, 131National Philosophy Faculty, 179National University of Tucumán, 152nationalism, 146nature, 18Nayar, 241Near East, 224Negev, 123Neolithic, 16, 21, 163, 224Nepal, 207, 210, 213, 215Neuchâtel, 27, 33New Britain, 262New Caledonia, 4, 19, 21, 228, 256–62,

265–9, 270New Guinea, 229, 264, 270New School of Social Research, 8New World, 4New York, 8, 179Newar, 207, 215Ngbandi, 114nguinza (money), 105Niamey, 79, 89Nice, 30

284 Subject index

Niger river, 78, 80territory, 79, 84, 89

Nîmes, 173Nivacle, 154non-modern, 243Normandy, 39, 199, 209Notes and Queries, 203, 216Nouvelle Calédonie, see New CaledoniaNouville, 17, 199, 201–4, 209, 211,

213, 214, 216Nzakara, 14, 21, 103, 107–22

Oobá, obaj, 180, 193observation, 19

participant, 9, 83, 185, 188, 203Oceania, 224, 225, 227–231ogan, 180, 193Oju Oba, 186opposition

binary, 19, 239, 245, 246, 251hierarchical, 18, 239, 245–7,

249–52Organisation pour la Recherche

Scientifique et Technicque de l’Outre-Mer (ORSTOM), 8, 107

orixá (saint), 180, 193Oryza, 208Oxford, 18, 237, 241, 255

Ppai-de-santo (chief priest), 184, 193Pakistan, East, 199–200, 209, 216Papua New Guinea, 229, 264, 270Paris, 7, 19, 25, 77, 78, 107, 109, 128,

130, 143, 187, 188, 199, 207, 237,242, 258, 259, 260

parody, 40parole, 261participation, mythic, 264, 267, 270participations, law of, modes of, 190, 258paysannat system, 116peasants, 213personage, 261, 267personhood, 19, 38, 41, 42, 261, 263–7,

270perspectivism, 156Peru, 151, 153phatic speech, 221phenomenology, 8, 12, 19, 189, 269, 271philology, 57philosophy, 15, 42, 105, 126–7, 130,

146, 173, 175, 258photography, 33, 98, 184, 188, 202Picardy, 199, 219pigs, 228–9Pilaga, 157, 160plants, 18poetry, poets, 48, 77, 109, 111–12, 126,

172, 175, 181, 182, 193, 263Poland, 27politeness, 221politics, 14, 20, 204, 211polygyny, 249Polynesia, 155, 158, 159, 163, 207Ponts et Chaussées (Grand Ecole), 79Popular Front, 131Portuguese, 180, 204positivism, 56, 174, 190possession, spirit, 13, 16, 78, 79, 84, 89,

90, 93–4postmodernism, 8, 13, 76, 86–7, 255,

262, 271post-structuralism, 19, 23, 260–2, 266potlatch, 153pottery, 27power, 42, 211, 240, 245practice(s), 8, 13, 21, 62, 221, 222, 230,

245, 250–1, 252practice, theory of, 63Prague, 28Pramali Kallar, 237, 240Pre-Colombian, 139, 147predation, 156prehistory, 29, 105, 147pre-logical, 176, 258primitivism, 268Protestant(s), 19, 173, 257, 259Provence, 36, 199psychiatry, 270psychoanalysis, 8, 32, 178, 270psychology, 18, 105, 174–6, 189, 203,

204, 269social, 105, 174, 179, 192

pure/impure, 18, 227, 229, 240, 249purity, 232Pyangaon, 215

Qqadi (judge), 61qanun (qawanin) (canon), 12, 46–68Quechua, 132–6, 140, 142, 147, 153questionnaires, 137Quito, 147

Subject index 285

Rrace, 15, 128, 134, 136, 139–41, 153,

190, 244, 250racism, 61, 135, 141, 144, 146, 249,

255, 268Rafai, 114Rahmaniyya order, 59, 60, 67raqba (feud), 57rationality, 2, 135, 182, 263, 269

Cartesian, 171Rauto, 262, 263realism, 93reason, 162, 191reasoning, deductive, 2, 171rebirth, 248Recherches en sciences humaines, 106Rechtsgeschichte, 55Recife, 179, 187, 192reflexivity, 171regionalism, 33–5relativism, cultural, 8relativity of groups, 241religion(s), 6, 15, 18, 39–42, 81, 135–8,

155, 158, 159, 165, 171, 173, 174,176–9, 182, 187, 204, 206, 207,211, 215, 240, 262, 263African, 16, 190Afro-Brazilian, 16history of, 25, 29Muslim, 59

remainders, 118Renaissance, 143renunciation, 238, 247, 248representations, 18, 221, 230

collective, 175, 244reproduction, cultural, 7republicanism, 58resistance, 11reversal, 239Revue Africaine, 50Revue de l’Histoire des Religions, 26rice, 208, 210rights, human, 244Rio de Janeiro, 177, 179Riobamba, 132–6ritual(s), 7, 12, 30, 33, 35–43, 77, 85,

133, 135, 158, 178, 184–5, 205,228, 245, 246, 248

rituals of atonement, 42role, 266Romans, 55, 233, 267rôtie, 40

Roussillon, 199Rules of sociological method, 22

Ssacred, 178, 182sacrifice, 246saints, 43, 180Salonica, 129, 235Salvador de Bahia, 178, 179, 183, 184, 193Salvation, 247Sanga, 82Sango, 122Sanskrit, 237sanza (thumb piano), 112São Liuz do Maranhão, 192São Paulo, 16, 177, 178, 187, 188, 192Sar Luk, 23Savoy, 30, 36Sayyid, 12, 49science, 181science fiction, 78Sciences Po, 103Second Empire, 60, 64Second World War, 7, 8, 11, 27, 39, 153,

236Sedan, 60Seine-Maritime, 209self, 260–1, 263self-development, 19semiology, 263Senegal, 4shamanism, 156, 165shari’a, 51, 55, 59, 61Shudras, 237Sigui, 85, 98singularity (seeing, thinking in), 14,

118–21, 233–4Sirens, 185, 191Smithsonian Institution, 131, 147, 153socialism, 134Socialist Party, 131Sociedad Colombiana de Etnología, 131Société d’Anthropologie de Paris, 128Société d’Etudes Alfred Métraux, 165Société de Folklore Français, 28Société des Américanistes de Paris, 128,

129, 131, 147Société des Nations, 28Société des Observateurs de l’Homme, 3, 10Société des Sultanats, 117Société Française de Psychologie, 212sociography, 200–1

286 Subject index

sociology, 4–8, 18, 20, 27, 42, 105, 109,130, 172–9, 183, 187–92, 204, 237,239, 243depth, 17, 190rural, 213

sociometry, 104–5Somme, 209Songhay, 13, 78, 80, 81, 84, 85, 93, 94, 99Sorbonne, 28, 79, 130, 198, 210sorcery, 39Sorko, 84Soudan, 77soul(s), 23, 159, 183, 185, 187Soviet Union, 233space, 207, 213, 220Spain, 67Spanish, 134, 139, 142, 191Sre, 22St Besse (cult), 6, 31, 199St Paul, 267state, 19status, 240Stimmung, 91, 94, 96Strasbourg, 179stratification, social, 247structural-functionalism, 8, 255structuralism, 2, 3, 7, 8, 17, 18, 20, 40,

87, 88, 125, 144, 145, 148, 171,198, 202, 213, 230, 231, 235, 238–45, 251, 255, 256

structure, 19substance, 249sub-stratum theories, 239, 240Sud e magia, 38suelto, 134–5suicide, 137, 153superstition, 15, 39surrealism, surrealists, 13, 76–7, 88, 91,

96, 99, 163, 182survivals, 32, 239Sweden, 15, 151Switzerland, 15, 27, 35, 151, 191symbolism, 19, 37–42, 206, 263symmetry, 118syncretism, 16, 178, 187, 192, 194

Ttaboo, 227, 228Tabou et totemisme à Madagascar, 28taddart (village), 54Tahoua, 75tajmaat (assembly), 53

Tamang, 207Tamazirt, 51, 61, 67Tamil, 237Tamil Nadu, 236, 237, 242, 251Tamils, 18, 239, 242tapirage, 165tarentism, 39techniques, technology, 17, 200, 205–9,

222–3, 232Terre Humaine, 10terreiro (cult house), 180–1, 184–7, 193Thai, 209theology, 257theory, 2, 10, 15– 22, 160, 172, 189,

198, 213, 241, 257of practice, 13

Third Republic, 21, 46Tibetan, 199Tibeto-Burman, 200, 209, 215time, 38–9, 40, 118, 203Toba, 152, 157, 160Torres Straits expedition, 7total, totalities, 211–12, 222totalitarianism, 240totemism, 6, 30, 136, 160, 270Toulouse, 40tourou (drum), 89, 90, 91, 94tradition, invention of, 12trance, 13, 16, 21, 91, 180, 186, 192transactionalism, 12transactions, 63transcendence, 224–5, 232, 246, 248transition, 12translation, 241, 262travellers, 9, 28tribes, 240Trickster, 119Tristes tropiques, 16, 22, 167, 186, 200Trobriands, 125Trocadéro, 4, 9, 10, 14, 22, 28, 130,

146, 147, 198Tsachilas, 136Tuareg, 48, 65Tucumán, 152, 153Tupí-Guaraní, 15, 151, 156Tupinamba, 151, 156, 157Turks, 59twins, 118

UUbangui river, 113, 120, 122Uele river, 113

Subject index 287

‘ulama, 46, see also ‘alimUlysses, 191Une esthétique perdue, 14UNESCO, 21, 123, 153, 159, 165, 166,

199, 203, 216Union Rationaliste, 219United States, 249universalism, 232universities, 3, 8Untouchables, 245‘urf (customary usages), 64Uro, 157, 166Uro(u)-Chipaya, 152, 157, 158Uttar Pradesh, 238, 242

VValence, 16, 176value(s), 3, 18–19, 207, 237, 239–41,

245–50Vanuatu, 228varnas, 245Vedas, 245, 246vegetarianism, 250Versailles, 177Vichy, 79, 84, 131, 259Vienna school, 140, 143Vietnam, 4, 17, 22, 210–11, 223Vincennes, 7Virgin, 41voodoo, 159, 160, 162, 166Vungara dynasty, 113, 114

Wwar, 158, 245washing of the necklace, 184, 192Wayapi, 156Wasigny, 126weapons, 136West, 19, 239, 244, 245, 247–52wheat, 18Whigs, 255, 256, 265, 270Whites, 134, 191witch doctors, 133witchcraft, 136, 137work, 250Writing culture, 262

YYale, 152Yoruba, 186youth, 41Yuma, 148

ZZande, 14, 107, 108, 111–23Zen, 220Zerma, 84, 89, 93zima (cult priests), 90zoology, 230

288 Subject index

AAgeron, C.R., 58, 61, 66, 67, 68, 69Alain, 131Aleijadinho, 178Allen, N.J., 22, 23, 24, 70, 130, 148,

250, 252Allier, R., 173Althusser, L., 1, 2, 8Amadigné Dolo, 85Amado, J., 179, 180, 184Andrade, M. de, 178Andrade, O. de, 178Ans, A.-M. d’, 164, 167Anthony, R., 128Aráuz, J., 146, 147, 148, 149Arbousse-Bastide, P., 173, 177Archaimbault, C., 198Aristotle, 182Armstrong, L., 91Aron, R., 106, 117, 123Aucapitaine, H., 66, 67, 69Aurégan, P., 23Auroi, C., 152, 154, 167

BBachelard, G., 231Baillière, J.B., 205, 216Baker, J., 77Balandier, G., 9, 191, 193, 194Bangbanzi, R., 117Barrau, J., 208, 244, 252Barraud, C., 244, 252Barthes, R., 1, 263Bastide, R. 5, 14, 16, 17, 19, 20, 21, 70,

171–196Bataille, G., 151, 161, 167, 169, 182, 235Bate, D., 98, 100 Bateson, G., 222Baudrillard, J., 1, 262Beaud, S., 22, 23Bédoucha, G., 215Behar, R., 67, 69Bellwood, P., 143, 148Belmont, N., 25, 28 Bensa, A., 17, 18, 266, 268, 270, 271,

273

Berbrugger, L.A., 50, 66Berger, L., 22, 24Bergson, H., 264Bernand, C., 147, 148, 149Bernard, A., 47, 50, 51, 52, 56, 60, 62,

66, 67, 68, 69Bernot, D, 199, 216Bernot, L., 2, 5, 9, 10, 14, 16, 17, 18, 20,

21, 197–218Berque, J., 45, 66, 67, 68Berrouet, E., 159Bertholet, D., 148Betts, R.F., 61, 69Beylier, C., 192, 194Bing, F., 162, 163, 166, 167Blancard, R., 199, 203, 204, 214, 217Blanckaert, C., 9, 23, 63, 69, 149Bloch, Marc, 212Bloch, Maurice, 145, 149, 250, 253Blue, J., 88, 100Boas, F., 8, 140, 141, 142, 144, 145,

146, 147, 148, 151Bouglé, C., 160, 245Boulbet, J., 198Bourdieu, P., 1, 8, 13, 62, 63, 68, 69, 71,

233, 251Bousquet, G.H., 55, 62, 67, 68, 69, 71 Braudel, F., 177, 180Breton, A., 91Broca, H., 28 Bromberger, C., 11, 24, 43Bruneau, M., 201, 217Buckner, M., 14, 273Burghart, R., 248, 252Burke, E., 68, 69Burns, P., 68, 70

CCaillois, R., 182, 194, 235Campbell, L., 142, 149Capone, S., 273Cardoso, I., 178, 195Carelli, M., 177, 192, 195Carette, E., 48, 66, 70Carlos, A., 150Castelain, J.P., 136, 149

NAME INDEX

Céfaï, D., 9, 23Certeau, M. de, 1Cervoni, A., 99, 100Chaix, B., 35Chanock, M., 68, 70Charuty, G., 4, 11, 12, 274Chevasse, P., 127, 148, 149Chirac, A., 105, 123Chiva, D., 25, 28, 43Chomsky, N., 231Christelow, A., 68, 70Clancy-Smith, J.A., 67, 70Clastres, P., 16, 19Clifford, J., 19, 22, 77, 80, 83, 98, 100,

198, 217, 257, 258, 260, 261, 262,263, 266, 267, 268, 271

Cohen, M., 28, 150Colleyn, J.P., 97Colonna, F., 68, 70Comte, A., 56Condominas, G., 9, 17, 23, 200, 214Conklin, H.C., 208, 211, 217Cooper, J., 154, 155, 167, 169Copans, J., 10, 22, 23Coppet, D. de, 3, 238, 244, 252Crapanzano, V., 43, 57, 266, 271Cuisenier, J., 25, 43Cusset, F., 10, 22, 23, 262, 271, 272

DDa Silva, V.G., 188, 195Damouré Zika, 84Dampierre, E. de, 9, 103–124Daumas, E., 48, 57, 66, 67Davis, N.Z., 24, 43Davis, W., 161De Martino, E., 37, 38, 39, 43, 44, 274Delacampagne, C., 3, 23, 244, 252Delafosse, M., 6, 7, 9Delamarre, M., 224, 234Demangeon, A., 212Derrida, J., 23, 262Descola, P., 8Dias, N., 129, 138, 141, 145, 146, 149Dieterlen, G., 7, 11, 79, 80, 81, 82, 85,

88, 98, 100Dion, R., 212Djabir, 115,Dostal, W., 48, 57, 66, 67 Douglas, M., 98, 100Dournes, J., 22Dresch, P., 68, 70

Drew, R., 87Duarte, P., 131, 178Dumas, G., 177, 195Dumézil, G., 242Dumont, J., 235Dumont, L., 1, 2, 3, 4, 5, 14, 15, 16, 18,

19, 20, 21, 23, 199, 200, 202, 217,235–254

Dumont, V.E., 235Dundes, A., 27, 43Durham, M.E., 67, 70Durkheim, E., 1, 5, 6, 7, 11, 18, 22, 23,

26, 28, 31, 37, 42, 43, 68, 72, 128Dussán de Reichel, 126, 132, 140, 142,

148, 149Duvigneaud, J., 193, 195

EEaton, M., 98, 100Echeverria, R., 185, 186, 193, 196Ellington, D., 91Emerit, M., 67, 70Evan-Pritchard, E.E., 14, 18, 23, 112, 113,

114, 122, 123, 124, 237, 241, 252

FFabar, P., 48, 57, 70Fabre, D., 25, 28, 40, 43Fabre-Vassas, C., 41, 44Faron, L., 165, 167Favret-Saada, J., 39, 44, 45Febvre, L., 177, 188, 192, 195, 199,

212, 215, 216, 217, 237Féraud, L.C., 66, 70Fernandes, F., 156Ferry, J., 117, 124Fieschi, J.-A., 88, 100Fine, A., 41Flaherty, R., 85Forbes, J., 98, 100Foucault, M., 1, 8, 262Fox, R.G., 11, 24Francqui, Commandant, 115Frazer, J.G., 37, 148, 150, 158, 161Frémaux, J., 66, 70Friedberg, C., 208Friedemann, N.S. de, 131, 149Fulchignoni, E., 91, 94, 99, 100

GGaillard, G., 3, 4, 24Galey, J.-C., 240, 241, 244

290 Name index

Gellner, E., 68, 69, 70Georgakas, D., 98, 100Gernet, J., 224, 234Giddens, A., 251Gil, G., 186Gilissen, J., 68, 70Gingrich, A., 11, 24Giorgio de Chirico, 77, 99Gjeçov, S., 67, 70Godelier, M., 2, 8, 24Gónzalez Suárez, F., 136, 139, 147, 149Goodman, J.E., 66, 70Gould, S.J., 255Gourmont, R. de, 28, 36 Graebner, F., 232Granet, M., 190Grenand, P., 156, 167Griaule, M., 6–8, 11, 13, 16, 22–4, 32,

77, 79–85, 88, 98, 99, 100, 160,181, 182, 198, 269, 271

Gruzinski, S. 136, 149Gueydon, (Admiral de), 61, 70 Guiart, J., 268, 271, 272Guimaraes, A.S.A., 192, 195Gurvitch, G., 8, 17, 98, 178, 179, 188,

189, 190, 193, 194, 195

HHacoun-Campredon, P., 67, 70Haekel, J., 140, 149Haffner, P., 97, 100Hall, G., 147, 149Hallbwachs, M., 173, 175, 176Hamy, E.-T., 4, 8, 17, 98, 128Hannemann, T., 45, 52, 56, 60, 63, 66,

67, 70Hanoteau, A., 12, 13, 20, 21, 45–73Harcourt, R. d’, 126, 149Harris, O., 147Hasluck, M., 67, 71Haudricourt, A.-G., 9, 10; 17, 18, 20,

200, 206, 208, 213, 214, 219–234Heekeren, D. van, 263, 234, 266, 272Heidegger, 19, 232, 263, 264Henley, P., 7, 13, 16, 75–102, 274Hennebelle, G., 97, 101Héritier, F., 8Herskovits, M., 179, 192, 195, 196Hertz, R., 6, 24, 31, 43, 44, 128Heusch, L. de, 1, 8, 100Hevezy, G. de, 154Heyerdahl,T., 159

Hirsch, E., 265, 271Hobbes, T., 63Hocart, A., 213, 217Hodgson, B., 210Holleman, J.F., 68, 71Hornborg, A., 143, 149Horton, R., 106Hubert, H., 175, 275Hugo, V., 233Hurgronje, R., 8Hurgronje, S., 62

IIllo Gaoudel, 84Izard, M., 147, 148

JJamin, J., 10, 23, 43, 44, 77, 98, 101,

130, 148, 148, 195, 266, 269, 271Jijón caamano, J., 147Johnson, C., 141, 144, 146, 148, 149Jolas, T., 39Julien, C.-A., 66, 68, 71Jung, 19, 272, 263Just, R., 71Jutra, C., 99, 101

KKalt, J.P., 147, 149Kant, E., 225Karsten, 147 Kemlin, J., 4Kemper, M., 67, 68, 70, 71Kerns, V., 131, 149Khalil ibn Ishaq, 56, 60, 71Koch-Grünberg, 147Kovalevsky, M.M., 68, 71Krickeberg, 147Kropotkin, P., 68, 71

LL’Estoile, B. de, 7, 9, 10, 22, 24, 268, 272Lacan, J., 1, 8Lam Ibrahim Dia, 84Landaburu, J., 142, 147, 149Langevin, P., 131Laplantine, F., 22, 24Laroche, M.C., 163, 167, 269, 271Latour, B., 1Laurière, C., 127, 128, 129, 130, 136,

137, 138, 141, 146, 147, 148, 149Laveleye, E., 68, 71

Name index 291

Lazreg, M., 67, 71Le Bouler, J.-P., 151, 167, 193, 195, 196Le Chatelier, A., 200Le Cour Grandmaison, O., 68, 71Le Play, F., 4, 5, 28Leach, E., 138, 149, 166, 167Leacock, R., 87Léautaud, P., 36Leenhardt, F., 257Leenhardt, M., 5, 19, 21, 23, 160, 176,

190, 195, 255–272Lefèvre, L., 67, 71Lehmann, 147Leiris, M., 7, 22, 24, 43, 44, 100, 101,

107, 108, 161, 167, 182, 193, 195,198, 235,

León, L., 25, 127, 139, 146, 147, 149Leontovitch, F.I., 71Leroi-Gourhan, A., 4, 6, 17, 198, 199,

200, 206, 214, 216, 230Letourneux, A., 12, 13, 45–74Levallois, M., 67, 71Lévi-Strauss, C., 1, 2, 7, 8, 14, 16, 17,

18, 19, 22, 24, 27, 37, 39, 42, 44,125, 141, 143, 144, 145, 146, 148,149, 160, 165, 166, 167, 171, 186,191, 195, 198, 199, 200, 214, 216,217, 230, 231, 232, 236, 237, 238,240, 241, 242, 245, 246, 250, 251,253, 269, 271

Lévy-Bruhl, L., 6, 14, 28, 37, 128, 129,130, 131, 146, 161, 162, 166, 175,176, 190, 195

Linant de Bellefonds, Y., 66, 71Linton, R., 179, 196LiPuma, E., 265, 272Lorcin, P.M.E., 61, 66, 67Lottman, H., 131, 149Lowie, R., 131, 154, 160Luc, B., 51, 57, 71Luce, G.H., 209, 216Lühning, A., 180, 193, 195Lyautey, M., 7, 68, 69

MMacCarthy, O., 47, 71MacClancy, J., 5, 6, 19, 24, 255–272Mach, 264, 270MacMahon (Marshal), 57, 71Madan, T.M., 240, 241, 244, 252, 253Mahé, A., 45, 53, 57, 61, 63, 66, 67, 68Maine, H., 160

Malaurie, J., 22Malinowski, B., 7, 15, 19, 42, 53, 62, 63,

81, 82, 98, 99, 101, 125, 149, 150,160, 176

Mamber, S., 87, 98, 101Mamdani, M., 68, 71Marcel, J.-C., 189, 190, 194, 195Marcorelles, L., 87, 101Marcus, G., 19, 271Marcy, G., 67, 71Marillier, L., 25Marin, L., 28Mary, A., 191, 194, 195Maschio, T., 262, 263, 266, 272Masqueray, E., 45, 67, 72, 68, 70Massignon, L., 201Maunier, R., 63, 72Maurer, B., 265, 270, 272Mauss, M., 1, 5, 6, 7, 8, 12, 13, 14, 15,

17, 18, 19, 20, 22, 23, 24, 25, 26,27, 28, 37, 44, 63, 80, 81, 86, 88,128, 129, 130, 146, 148, 151, 152,153, 175, 176, 189, 192, 196, 204,206, 207, 211, 212, 216, 217, 229,230, 232, 234, 236, 239, 240, 241,243, 251, 253, 259

Meillassoux, C., 2Merleau-Ponty, M., 1, 8, 271Métraux, A., 15, 16, 19, 20, 21, 105,

131, 151–169, 185, 186, 193, 196 Mézières, B. de, 116Michelet, J., 67, 72Millet, S., 178Milliot, L., 47, 50, 51, 52, 56, 60, 62, 66,

67, 68, 69, 72Mistral, F., 36Moffat, M., 240, 244, 252, 253Monbeig, P., 177Monnier, A., 152, 154, 167Montagne, R., 45, 68, 72Montesquieu, 55Moore, S.F., 68, 72Morand, M., 67, 72Moréchand, G., 198Moreno, Dr., 104, 124Morgan, L. H., 46, 73Morin, E., 89, 92Motta, R., 181, 196Moura, C.E.M., 193, 196Moussa Hamidou, 84Murra, J., 193, 196Muttusami, Tevar, 236

292 Name index

NNaepels, 267, 268, 270, 272Nauton, P., 210, 217Needham, J., 206, 217Needham, R., 27, 44, 256, 257, 270, 272Nef, J., 105Neumann-Spallart, F.X., 105Neveu, E. de, 67, 72Nietzsche, F. 91, 99Nobrega, C., 185, 186, 193, 196Nordenskiöld, N.E., 15, 143, 147, 151,

154, 155, 158, 165, 169

OOuld-Braham, O., 67, 72

PParain, C., 202, 223Park, R.E., 179, 196Parker, E., 46Parkes, P., 12, 13, 270, 271, 274Parry, J., 250, 253Parsons, T., 251Patorni, F., 53, 72Patrinos, H.A., 147, 149Paulme, D., 7, 11, 107, 160Peixoto, F.A., 174, 175, 177, 178, 182,

192, 195, 196Pergaud, L., 36Perkins, K.J., 66, 67, 72Pessoa, J., 179Piault, M.-H., 98, 101Pina-Cabral, J., 149Pineda Camacho, R., 126, 130, 138,

141, 142, 143, 147, 148Pingaud, M.C., 39Plato, 119, 182, 225Pocock, D., 238, 240, 243Poussereau, L.M., 66, 67, 72Prédal, R., 75, 100, 101Preuss, 147Price, S., 67, 98, 101Prins, H., 152, 165, 166, 169Privat, J.M., 29, 44Pythagorus, 119

QQuatrefages, A. de, 4, 9Queiróz, M.I.P. de, 173, 180, 183, 185, 196

RRabih, 110Radcliffe-Brown, A.R., 160, 237, 239Ramos, A., 179, 192, 195, 196Randon, Marshal, 48, 49, 58, 61, 64, 66Ravalet, C., 184, 192, 196Redfield, R., 105, 179, 196Reichel-Dolmatoff, G., 131Renan, E., 36, 51, 72 Renou, L., 237Renshaw, J., 160, 169Rey, P.P., 2, 8Richard, G., 173, 174, 175, 192, 196Richardson, M., 98, 101Ricoeur, P., 1Rinn, L., 68, 72Rival, L., 1, 8, 14, 15, 23, 275Rivers, W.H.R., 203, 217Rivet, P., 6, 10, 14, 15, 17, 18, 20, 21, 22,

28, 125–150, 152, 153, 154, 165, 198Rivière, G.-H., 4, 22, 28, 235Rivière, P., 15, 16, 21, 275Roberts, H., 68, 72Robin, J.N., 68, 72Rochereau, H.J., 138, 150Rogers, S., 5, 24Rohmer, E., 87, 101Rosfelder, R., 82Rouch, J., 7, 9, 11, 13, 16, 20, 21,

75–102, 274Rouget, G., 113, 124Rousseau, 163, 167Royer, L., 36

SSabatier, L., 5, 68Sadoul, G., 99, 101Saint-Simon, 50, 58, 67, 71Saintyves, P., 31Salemink, M.B., 68, 72Salemink, O., 68, 72Sand, G., 35Sangren, P., 262, 271, 272Sartre, J.P., 8, 234, 264, 271Saussure, F. de, 8, 9Savigny, 55Schaeffner, A., 107Scheele, J., 63, 66, 67, 68, 72Schubring, W., 237Schweinfurth, G., 110, 124Segalen, M., 5 Segall, L., 178

Name index 293

Senhora, 184, 185, 186Shafer, R., 209, 217Shinar, P., 68, 72Si Lunis, 49, 51, 60, 61, 67Si Mula, 49, 51, 52, 53, 56, 57, 59, 60,

61, 64, 66, 67, 68Sibeud, E., 22, 24, 68, 72Slane, W. de, 48Snyder, F., 68, 72Sorre, M., 212Souty, J., 187, 193, 196Sperber, D., 9Srinivas, M., 237Stendhal, 35, 36Steward, J., 131Stewart, F., 68, 72Stocking, G., 127, 132, 138, 140, 142,

150, 203, 217, 225Stoller, P., 98, 99, 100, 101Strathern, M., 9, 265, 266, 270, 272, 275Strauss, L., 105, 106Surdon, G., 68, 73

TTallou Mouzourane, 84Tarde, G., 174, 175, 192, 196Tardieu, S., 235Tardits, C., 165, 169Taylor, L., 78, 98, 99, 102, 191Teixeira, A., 179, 196Ter Haar, B., 68, 73Terray, E., 2, 8 Thomas, J., 215, 217, 218Thompson, C., 77, 98, 100, 101, 102Tillion, G., 11, 24Tocqueville, A., 245Todorov, T., 11, 24Toffin, G., 14, 17, 239, 243, 244, 252,

253, 275Tooker, E., 46, 73Trautmann, T.R., 46, 73Tupper, C.L., 68, 73Tylor, E., 150, 160, 217, 261

UUrbain, I., 50, 58, 59, 66, 67, 71Uribe, T. 139, 150

VVallès, J., 36van Gennep, A., 4, 5, 6, 11, 12, 20, 21,

23, 24, 25–44, 160Vangele, 110Vaux, P. de, 113, 124Velay-Valentin, C., 25, 44Verdier, Y., 39, 40, 44Verger, P., 16, 164, 169, 172, 173, 180,

183, 184, 185, 186, 187, 188, 191,193, 194, 195, 196

Verneau, R., 128, 129Vertov, D., 92, 93, 94, 95, 96Vico, 263Vidal, J.E., 208Viveiros de Castro, E., 156Vollenhoven, C. van, 62, 71

WWagley, C., 154, 157, 160, 161, 169Weber, F., 22, 23Weber, M., 106, 113, 124, 171, 175,

189, 190, 250, 251Wilder, G., 148, 150Winston, B., 98, 102Wylie, L., 199, 218

YYoung, C., 98, 102, 266, 267

ZZerilli, F., 126, 137, 140, 141, 147, 150Zin, D.L., 43Zonabend, F., 5, 39, 199, 218

294 Name index