Far Afield: French Anthropology between Science and Literature

By Vincent Debaene

Anthropology has long had a vexed relationship with literature, and nowhere has this been more acutely felt than in France, where most ethnographers, upon returning from the field, write not one book, but two: a scientific monograph and a literary account. In Far Afield—brought to English-language readers here for the first time—Vincent Debaene puzzles out this phenomenon, tracing the contours of anthropology and literature’s mutual fascination and the ground upon which they meet in the works of thinkers from Marcel Mauss and Georges Bataille to Claude Lévi-Strauss and Roland Barthes.
           
The relationship between anthropology and literature in France is one of careful curiosity. Literary writers are wary about anthropologists’ scientific austerity but intrigued by the objects they collect and the issues they raise, while anthropologists claim to be scientists but at the same time are deeply concerned with writing and representational practices. Debaene elucidates the richness that this curiosity fosters and the diverse range of writings it has produced, from Proustian memoirs to proto-surrealist diaries. In the end he offers a fascinating intellectual history, one that is itself located precisely where science and literature meet.

• Language: English
• Publisher: University of Chicago Press
• Release date: Apr 4, 2014
• ISBN: 9780226107233

Book preview

VINCENT DEBAENE is associate professor of French at Columbia University. He is the critical editor of the Pléiade edition of the works of Claude Lévi-Strauss and coauthor of Claude Lévi-Strass: L’Homme au regard éloigné.

JUSTIN IZZO is assistant professor of French Studies at Brown University.

Cet ouvrage publié dans le cadre du programme d’aide à la publication bénéficie du soutien du Ministère des Affaires Etrangères et du Service Culturel de l’Ambassade de France représenté aux Etats-Unis. This work received support from the French Ministry of Foreign Affairs and the Cultural Services of the French Embassy in the United States through their publishing assistance program.

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2014 by The University of Chicago

All rights reserved. Published 2014.

Printed in the United States of America

Originally published as Vincent Debaene, L’Adieu au voyage: L’ethnologie française entre science et littérature (Paris: Éditions Gallimard, 2010)

© Éditions Gallimard, Paris, 2010.

23 22 21 20 19 18 17 16 15 14       1 2 3 4 5

ISBN-13: 978-0-226-10690-8 (cloth)

ISBN-13: 978-0-226-10706-6 (paper)

ISBN-13: 978-0-226-10723-3 (e-book)

DOI: 10.7208/chicago/9780226107233.001.0001

Library of Congress Cataloging-in-Publication Data

Debaene, Vincent, author.

[Adieu au voyage. English]

Far afield : French anthropology between science and literature / Vincent Debaene ; translated by Justin Izzo.

pages cm

"Originally published as Vincent Debaene, L’adieu au voyage : L’ethnologie française entre science et littérature (Paris: Éditions Gallimard, 2010)

© Éditions Gallimard, Paris, 2010." — Title page verso.

Includes bibliographical references and index.

ISBN 978-0-226-10690-8 (cloth : alkaline paper) — ISBN 978-0-226-10706-6 (paperback : alkaline paper) — ISBN 978-0-226-10723-3 (e-book)

1. Ethnology—France—History—20th century. 2. Ethnology—Authorship. 3. Literature and anthropology—France—History—20th century. I. Izzo, Justin, translator. II. Title.

GN308.3.F8D4313 2014

306.0944'0904—dc23

2013028984

This paper meets the requirements of ANSI/NISO Z39.48–1992 (Permanence of Paper).

Far Afield

French Anthropology between Science and Literature

VINCENT DEBAENE

TRANSLATED BY JUSTIN IZZO

The University of Chicago Press

CHICAGO AND LONDON

Contents

Preface to the English Edition

Introduction

The Ethnographer’s Two Books

Science and Literature: A Genealogy

I. ETHNOGRAPHY IN THE EYES OF LITERATURE

1. The Birth of a Discipline

Breaks and Discontinuities

Fieldwork

Ethnography’s Prestige

2. The French Exception

The Speculative Origins of French Ethnography

Everything involving the exercise of the mind

Malinowski: A Counterexample

3. Rhetoric, the Document, and Atmosphere

From the Science of Customs to Total Social Facts

Evocative Documents

The Supplement to the Ethnographer’s Expedition

The Impossible Return to Belles Lettres

The Human Document and the Living Museum

4. A literature that is not meaningless like our own

Some of the innocent flavor of the original text

L’Île de Pâques: 1941, 1951

Mauss, Fieldwork, and Ethnographic Documents

5. The Lost Unity of Heart and Mind

The Philosophical Voyage as Paradise Lost

From the Enlightenment to the Renaissance

A New Humanism

II. L’ADIEU AU VOYAGE

6. Ceci n’est pas un voyage

Travel: Polemics, Prestige, and Legitimacy

The Ethnographer, the Adventurer, and the Tourist

Spatializing Cultural Difference

L’Afrique fantôme and Tristes Tropiques: Impossible Intimacy

This is not travel writing

7. Les Flambeurs d’hommes: The Ethiopian Chronicles of Marcel Griaule

The Ethnographer and the Littérateur

The Inadequacies of the Ethnographic Document

The Impossible Evocative Document

Excursus: Sociology and Cruelty

Ethnography and Cultural Knowledge

8. L’Afrique fantôme: Leiris and the Living Document

The Impossible Foreword

Reading L’Afrique fantôme

From Communion to Representation

Theatricality and the Family

Living Document, Phantom Africa

9. Tristes Tropiques: The Search for Correspondence and the Logic of the Sensible

The boat entered the harbor at 5:30 in the morning

From Conrad to Proust

From the Deserts of Memory to the Science of the Concrete

History, Entropy, Entropology

Doorways that reveal other worlds and other times

III. LITERATURE IN THE EYES OF ETHNOGRAPHY

10. Literature, Letters, and the Social Sciences

Lanson, 1895: The Dispossession of the Artist by the Scientist

The Man of Letters and the Social Division of Labor

Humanities, Sciences, and Counterrevolutionary Thought

Lanson, 1904: From Literature to Science

11. Disputes over Territory

Ramon Fernandez, 1935: A Conversation between the Scientist and the Essayist

Breton, 1948–1966: You will never really know the Mayas

Bataille, Barthes, Blanchot, 1956: The Reception of Tristes Tropiques

12. 1955–1970: A New Deal

The End of the Documentary Paradigm

Ethnography and Literature in the Real World

(Post)colonial Literature and the Ethnographic

The Terre humaine Series: Literature from Within and Without

Barthes and Structures

Barthes, 1967: From Science to Literature

Conclusion

Literature

Ethnography

Acknowledgments

Notes

Bibliography

Index

Preface to the English Edition

This book was published in French under the title L’Adieu au voyage. This phrase is an allusion to the last page of Tristes Tropiques, in which Claude Lévi-Strauss invites us to seize the essence of humankind not through geographical or anthropological explorations of the planet (fond farewell to savages and explorations!), but through the ephemeral contemplation of the works of nature: a crystal, a perfume, or, famously, the eye of a cat. In my mind, this phrase did not refer to such a project, and even less to some historical moment: the farewell to journeying does not designate some historical realization through which, after explorations and empires, the West would observe with bitterness the end of exoticism or the vanishing of differences (these topoi date back at least to the eighteenth century). It designates rather a moment within ethnography, through which the anthropologist relinquishes any idealized conception of difference. It is thus not only a farewell to some idealized Other, but also a farewell to oneself, in other words the redefinition of the relationship between subject and object. Like in Lévi-Strauss’s original phrasing, the farewell to the journey does not point to any conclusion, or disenchantment, but to the reconfiguration of a relationship, a twofold process of objectivation and subjectivation.

In the French anthropological tradition, this reconfiguration appears through a striking phenomenon: upon their return from fieldwork, most twentieth-century ethnographers produced, in addition to the expected scholarly monograph, a second book, a book that was often more literary, or at least freer in its form and intended for a wider audience than specialized publications. L’Île de Pâques by Alfred Métraux, L’Afrique fantôme by Michel Leiris, Les Flambeurs d’hommes by Marcel Griaule, Tristes Tropiques by Claude Lévi-Strauss: all of these books supplement anthropological works on the inhabitants of Easter Island, on the Dogon of Mali, on the Amhara of Ethiopia, or on the Nambikwara Indians of Brazil. Many other examples would be possible.

This second book is sometimes merely a symptom. In that case, it just betrays the insufficiencies of the epistemological model of the time; it testifies to the inner contradictions of the conception of the anthropological object but without solving them. This is what happens with the second books studied in chapters 3, 4, and 5. Written in the 1930s, they try to compensate for the shortcomings of the documentary anthropology that was then being developed around the Musée de l’Homme. By their mere existence, these second books reveal the inadequacy of the positivist and museum-based paradigm on which French anthropology thought it could be founded—but they don’t modify this paradigm.

But sometimes, there is more. It may happen that the writing itself of the second book contributes to this redefinition of the anthropological relationship, to this twofold process of subjectivation and objectivation. This is what happens with the second books studied in chapters 7, 8, and 9. In those cases, the second book does more than just betray a contradiction or an imbalance; it allows for a complete reorientation of the anthropological project. This is particularly obvious in the case of Lévi-Strauss, since one can contend that the composition of Tristes Tropiques played a crucial role in the development of the second structural anthropology (after the first and more sociological moment of The Elementary Structures of Kinship). Indeed, it is in this 1955 text that Lévi-Strauss undertakes for the first time what he will continue in The Savage Mind and The Mythologiques: the search for correspondences, analogies, and differences between various ensembles whose intelligibility and organizing principles are not extracted by the anthropologist’s rationality, but revealed through variations and comparisons. The only difference is that in Tristes Tropiques, Lévi-Strauss is not going back and forth between classificatory systems or variants of a myth, but between various layers of his own past. In this case, then, the literary work—or at least its writing—is less a supplement to the scholarly work than a condition of it, making possible the anthropological work to come. Literature here has nothing to do with representation anymore; it is just the site for experiences made with writing and through writing—in that sense, it is a true continuation of fieldwork (and whether or not the ethnographer’s own society then recognizes the book as a piece of literature is in itself of secondary importance). Although in more modest proportions one could observe the same type of dynamic with other ethnographers, for whom the experience of writing provides an occasion not only for reorganizing ethnographic data but for deeply changing the relationship to that data, allowing for a reconstruction both of oneself and of the object. This work of writing is not necessarily ethnographic in itself. Think of Michel Leiris: it is not the exploitation or the amplification of his field notes, but the passage through the difficulties of autobiography, which allowed him, twenty-five years after his fieldwork in northern Ethiopia, to formalize his experience in his book on possession cults in Gondar.

Finally, it may also happen that the second book has no direct relationship to the scholarly work: it is neither a refutation, nor a rectification or a supplement, but a mere testimony. In that case, it simply fulfills the elementary need to recount the fieldwork experience or to make use of the elements that the scholarly work left aside. It is just that the (very badly named) ethnographic encounter is tremendously richer, more complex, and more multilayered than a scientific event, and it is much more than the reference it will become in the conversations among experts in which ethnographer engages with his peers. The second book also plays the role of memoir of an asymmetrical and almost always nonreciprocal encounter, of which the ethnographer was the agent, and whose trace, because of guilt or psychological necessity, he wants to inscribe in his own culture and society.

We can see then that, in the French tradition, the pattern of two books is striking and strikingly recurrent (it still continues today), but this pattern is in itself an empty form. What matters is not the doubling, but the relationship between the two books and the historical evolution of this relationship.

The English title, Far Afield, seeks to preserve something of this imbalance, of the slight discrepancy created by the second book. Rather than remoteness or geographical distance (which are not necessary conditions for ethnography anymore, and have not been for a long time), it expresses a detour, a straying: not a refutation or an invalidation of the anthropological project per se, but rather a drift away from the expectations and assumptions that originally presided over ethnographic investigation. It is not the second book that has gone astray, afield, or off track, but rather the opposite: it reveals that the original expectations were flawed. And the second book is not an invitation to get back on track but, quite the contrary, it is an invitation to reorient the entire space of the anthropological relationship.

The second chapter of this book is devoted to the specificity of the French anthropological tradition. It shows that this divide between the two books finds its source in the ambivalent (and probably contradictory, in the eyes of an English-speaking reader) relationship of French scholars to literature. This paradoxical relationship is a consequence of the specific historical construction of French discursive space, that is to say a consequence of both the elaboration of the Parisian literary field and the gradual organization of the space of knowledge into academic disciplines. It combines both a rejection of and a desire for literature, a refusal and an aspiration; as a consequence literature is seen by the sociologist and the anthropologist as a temptation that one should avoid and, simultaneously, an accomplishment that one dreams of reaching. I would like now to add three elements about this French specificity that did not appear to me as clearly when I was writing this book.

I first would like to acknowledge that what I call the French anthropological tradition is more complex and heterogeneous that what I sometimes imply, especially after World War II. Even if I mention many ethnographers, some of them not well-known or forgotten, my general argument remains centered around the lineage of Durkheim–Mauss–Lévi-Strauss–Bourdieu, a lineage characterized by a distinct philosophical tendency and a strong connection between sociology and psychology, between the study of social facts and the study of mental facts. The theoretical description of French anthropology at the end of the twentieth century, and even more in the early twenty-first century, would be more complex and more diverse, even if several recent publications show that the relationship to literature remains a central concern for French anthropologists. But of all the tendencies of French anthropology, past or present, this lineage is doubtlessly the least compatible or the most difficult to translate into the terms of the American culturalist tradition (as shown, for instance, by Alfred Kroeber’s and Robert Lowie’s misgivings toward the Durkheimian group of L’Année sociologique in the 1930s or by Clifford Geertz’s reluctance toward Lévi-Strauss’s cerebral savage). This accounts for its ambiguous status in the American anthropological corpus. Mauss and Lévi-Strauss, Durkheim and Bourdieu, are doubtlessly part and parcel of the disciplinary canon and they remain required readings for any aspiring anthropologist, but they are read more as social theorists than as anthropologists per se; they are not integrated into the accepted narratives of the discipline.¹ Implicitly, French anthropologists are blamed for their nonexistent, or very poor, field experience and for their excessive theoretical ambition, the latter being generally seen as a consequence of the former. In this respect, it seems revealing to me that the lack of scholarly work on the French tradition by American historians of anthropology knows but two exceptions: the excellent studies by James Clifford on Maurice Leenhardt and Marcel Griaule. These seminal works (which were translated into French, as opposed to Writing Culture) helped to enrich and complicate our vision of the history of French anthropology, but they are devoted precisely to two scholars who were outsiders, in terms both of theory and of institutions, and who were both inscribing themselves in an exegetic approach toward native cultural texts, world-views, or wisdom.

Another specificity of French anthropology, which did not initially appear to me as clearly (and which is in direct connection with the previous point), is the striking absence of the term culture. This notion—and the discussions it unavoidably elicits about the relationship between culture and civilization—was at the core of the exchanges between anthropologists, writers, and literary critics in Great Britain and the United States during the twentieth century. It is on the grounds of his readings in anthropology that, in his Notes Towards a Definition of Culture, T. S. Eliot calls into question Matthew Arnold’s analyses in Culture and Anarchy, before being himself criticized by the anthropologists Alfred Kroeber and Clyde Klukhohn in their famous inventory Culture: A Critical Review of Concepts and Definitions. It is also the notion of culture that is at the heart of Edward Sapir’s (himself an ethnographer, linguist, and poet) twofold reflection on cultural genuineness and individual artistic creativity. It is again the notion of culture that allows Ruth Benedict to compare the work of the anthropologist with that of the literary or art critic.²

But this notion—as well as the controversy on cultural relativism, on culture with or without a capital C—is strikingly absent from the French social sciences. Contrary to Franz Boas’s students, ethnographers trained by Mauss were not trying to grasp an ethos, they wanted to breathe an atmosphere; they were not deciphering patterns, they were seeking a radical mental transformation. Although it was rarely explicit, the bodily experience of the ethnographer was seen as central since the social is by essence incarnated in body and in mind (as eighteenth-century theories of climate would have it). This bodily experience became a precondition for the comprehension of the social fact as opposed to the understanding of culture, a concept that Mauss described as even worse than that of civilization, itself already pretty bad ["assez mauvais"].³ The foreignness of French anthropology from an American perspective comes in part from this very simple fact: it is an intellectual tradition that never made culture a key concept and that, from Durkheim to Lévi-Strauss to Bourdieu (one could add Foucault), always refused to acknowledge what Bourdieu called the old Diltheyan distinction between explaining and understanding.⁴ I insist in the conclusion of this book on the fact that Writing Culture was never translated into French, and this is first and foremost because the title of this infamous collection does not quite make sense for a French anthropologist. French anthropology’s relationship to literature, when it is present, does not imply any hermeneutic stance and rarely involves a cultural representation; it takes place either earlier in the construction of anthropological discourse—in the bodily experience of the ethnographer and its translation into words—or later—in the reflections on ways to use, display, and present the results of the work performed.

Finally, an English-speaking reader might be surprised by the recurring use of the term science to designate French anthropology. But here again, this is a local historical reality. At the time of their foundation, French sociology and French anthropology were conceived as sciences (sciences of man was originally the most usual designation)—what else could they have been? The phrase human sciences appears much later in French (for the first time in 1942, to translate Dilthey’s Geisteswissenschaften) and the notion of humanities (which, in France, used to designate education based on the study of Greek and Latin literatures) became rarer and rarer in the twentieth century. In France, the scholarly disciplines that in German- or English-speaking countries are part of the humanities (or would hesitate between humanities and social sciences) strove throughout the twentieth century to be recognized as nothing more than plain sciences. As for literature per se, its destiny was played out elsewhere, above or beyond disciplines, but in any case in a space other than that of knowledge.

The mistake would be to believe, as Clifford Geertz did, that the American culturalist tradition constitutes the standard of true anthropology (the anthropology of professionals, as he says in The Cerebral Savage) and that this use of the term science is nothing but a cultural archaism, a sort of folkloric vestige. French anthropologists would fall victim to their inextirpable universalism and their narrow rationalism, an inheritance from the Enlightenment that blinds them to the complexity of symbols and meanings. But one should be a little more Foucaultian, and not assume the naturalness of one’s own theoretical framework: there is no reason to believe that anthropology cannot be a science; it all depends on what is meant by this word, on the local divides and oppositions that define science. To justify the proximity between anthropology and literature (or literary studies), one often recalls that philology is, after all, that nineteenth-century ancestor of both anthropology and literary study.⁵ That might very well be true in the United States, but it is not the case everywhere. Change the foundational operation—for example the way anthropology defined itself in contradistinction to literature or to folklore or to the natural sciences—and the subsequent relationships of anthropologists to literature (considered as both a corpus and a writing practice) and literary studies (considered as both a type of scholarship and an institutionalized discipline) will be greatly modified.

This book thus considers itself a history of French anthropology, but also a contribution to a wider comparison between intellectual traditions. The issue of the relationship between literature and anthropology cannot be abstracted from the local paradigms in which it was and continues to be framed. And this framing can be very different, depending on language, power relationships, space, and time. Given that anthropological traditions are local and that literary fields emerged on national bases, it is not surprising that the divide between disciplines or between discursive formations—as Foucault would have it—differs when considered from a French, British, German, Japanese, or American perspective. This shows that what need to be compared are not only the disciplines or writing practices, but the various divisions between disciplines and writing practices. This may be a lesson we can take from structural anthropology: what matters is not the differences, but the differences between the differences.

Introduction

Between 1925, the founding date of the Institut d’ethnologie in Paris, and the 1970s, there were countless exchanges between literature and anthropology in France. In many respects these decades comprise the ethnographic moment of French culture: First, ethnographers wrote works of anthropology, of course, but also stylized books that were more difficult to classify such as Mexique, terre indienne, L’Île de Pâques, Tristes Tropiques, Afrique ambiguë, and Chronicle of the Guayaki Indians. Second, literary writers, poets, and intellectuals read ethnographers: the Surrealists, for example, enthusiastically received Lucien Lévy-Bruhl’s Primitive Mentality before Lévy-Bruhl moved from the Do read to the Don’t read section on the back of the Surrealist publications catalogue. In the fall of 1937, Georges Bataille and others founded the College of Sociology, a moral community intended to promote a sacred sociology and to extend to modern societies the analysis of so-called primitive societies.¹

Twelve years later, Bataille and Simone de Beauvoir reported on Claude Lévi-Strauss’s Elementary Structures of Kinship in Critique and Les Temps modernes, respectively, before Roland Barthes and Gilles Deleuze commented on The Savage Mind and Introduction to the Work of Marcel Mauss, also by Lévi-Strauss. Some of these writers prided themselves on their expertise in anthropology: in 1938, Roger Caillois, who took Mauss’s courses at the Institut d’ethnologie, aspired to broad anthropological reflection in Le Mythe et l’homme, and in 1947 Bataille sought to generalize Mauss’s The Gift in The Accursed Share. Others played with conventional forms of ethnographic discourse or drew inspiration from it: Henri Michaux offers an imaginary ethnography in his Voyage en Grande Garabagne, and Georges Perec parodies Tristes Tropiques in Life: A User’s Manual. Sometimes the reaction was hostile, as with André Breton, who condemned the all too often icy gaze of the ethnographer,² or with Aimé Césaire, who, in his Discourse on Colonialism, defended Michel Leiris and Lévi-Strauss against attacks made by Caillois but mocked ethnographers who go in for metaphysics.³

Artistic forms circulated, as well: after Blaise Cendrars tried his hand at Negro poems beginning in the 1910s, around 1945 Breton wrote xenophilic poems in which he took up myths of the Maoris or the Easter Islanders. In the interwar period, La Nouvelle Revue française published Malagasy proverbs and Indian texts from Argentina at the same time as promotional excerpts from Lévy-Bruhl’s La Mythologie primitive. Literary series were created that tried to straddle both literature and ethnography, such as Gallimard’s L’Espèce humaine series, edited first by Alfred Métraux and then by Michel Leiris, or Plon’s Terre humaine series, edited by Jean Malaurie. Certain journals were also points of convergence, such as Documents (1929–30), edited by Bataille, in whose table of contents could be found the names of scholars from the Musée de l’Homme and those of dissident Surrealists who had just broken with Breton. In Martinique, Aimé and Suzanne Césaire’s journal Tropiques (1941–45) placed ethnography squarely within the range of the journal’s concerns. Other periodicals that situated themselves at the crossroads of intellectual trends opened their columns to ethnographers: Arnold van Gennep had an Ethnology, Folklore column in the Mercure de France from 1905 until his death in 1949, and in 1934 La Nouvelle Revue française created a Sociology column that actually dealt with ethnology. Finally, people themselves circulated, and the names of Leiris, Bataille, Métraux, Lévi-Strauss, and Barthes embody a permeability of modes of thought and types of discourse that seem to fall sometimes under the auspices of science and at other times under those of literature. In a certain sense, then, they appear to be continuing a specifically French tradition that, from Montaigne to Rousseau by way of Montesquieu, has always blended together philosophical reflection, curiosity about the exotic, introspection, and meditations on human nature.

It is not always easy to find one’s way in the abundance of these perspectives. Should we speak of exchanges, permeability, or circulation? But just what is exchanged? What exactly is circulating and where, between what spaces? Should we conceive of anthropology and literature as worlds (as in the world of La Nouvelle Revue française and the world of the Musée de l’Homme) that, like geometric shapes, might share common boundaries or areas of intersection? Or should we think of them as discourses? However, literature is not exactly a discourse, in the sense of a corpus of texts belonging to a specific period whose principles we could set out or whose internal rules we might discern. Should we instead think of them as types of writing, opposing the anthropologist’s scientific and restrained writing to the free and sovereign writing of the literary author who gives himself over entirely to language? Should we consider them as forms of scholarship? The term scholarship undoubtedly does not apply to literature, but is literature not still the site of a certain mode of knowledge? Since it also speaks to us about humankind, must we not acknowledge that literature has anthropological merit?

THE ETHNOGRAPHER’S TWO BOOKS

This book aims to respond to these types of questions but begins from the premise that it is impossible to do so by preserving the terms that they accept. These questions presuppose clearly defined concepts whose overlapping we might observe; and yet these presuppositions also block our line of investigation. This is to say not that anthropology and literature do not exist but rather that thinking in terms of intersections between readymade categories assumes that these latter are already defined and qualified and that we are simply responding to the questions, What is anthropology? and What is literature? These questions are legitimate—essential, even—and we should in no way dismiss them in favor of the richness of individual works, but they cannot constitute a point of departure; we will therefore address them in this book’s conclusion. In effect, this is the lesson of the ethnographer: rather than taking the risk of importing categories and artificially grafting them onto an evasive reality, before all else we must look at what people do. Thus our inclination should be to tend in the opposite direction, not of positing a preliminary definition of anthropology or an initial characterization of literature, but instead in the direction of examining texts and what ethnographers have written.

Indeed, this book takes as its point of departure the writings of ethnographers and not those of literary authors, and for several reasons: First, the former are more stable, more easily identifiable, and they do not constantly refer us back to the inevitable question of value (Is this really important? Is this really literature?). We will come back to this. Second, if we begin from the texts of literary writers, setting out to demonstrate the ways in which they borrow from ethnography, we lose sight of the singularity of our object since the same mode of inquiry would be possible for history, geography, or even physics. Yet we can already sense that the relationship between literature and anthropology is more profound, more fundamental, than a simple adoption or appropriation of scholarly knowledge by poets or essayists.

What we must do, therefore, is study what ethnographers have written and the ways in which those writings were conceived. We will rely in particular on a curious fact concerning the first generation of French ethnographers trained by Mauss in the 1920s and 1930s: nearly all the French ethnographers who left for the field before 1939 wrote, upon their return, not only a scholarly study of the people they lived with but often a second book as well, a more literary work that did not adhere to the canonical forms of the scholarly monograph. The rules of the ethnographic monograph in France were set by Mauss, who mixed the Anglo-American ethnographic account with the French tradition of positivist history. The organization of a text was more or less as follows: climate and geography, social organization (clans, moieties, lineages, etc.), technology (housing, tools, hunting and fishing techniques, etc.), social and family life (games, life cycles, etc.), economics, law, and religion. Although they were not very numerous, every ethnographer who followed Mauss’s teaching at the Institut d’ethnologie left for the field with this model in mind and when they returned published a book (usually their doctoral thesis) that generally respected the conventional framework. However, in addition to this book nearly all wrote an account of their experience that, if not literary, was at any rate not scholarly and unlike their scientific work was published by a generalist press such as Grasset, Gallimard, or Plon. The most famous instances are those of Leiris, who wrote La Possession et ses aspects théâtraux chez les Éthiopiens de Gondar (1958) and L’Afrique fantôme (1934), and Lévi-Strauss, who published La Vie familiale et sociale des Indiens nambikwara (1948) and Tristes Tropiques (1955). These are not the only cases, though, and this curious form of splitting and duplication can be observed in a number of ethnographers from the same era.

After Silhouettes et graffiti abyssins and Jeux et divertissements abyssins, two collections that emerged from the first expedition he led in Ethiopia in 1929, Marcel Griaule won the Prix Gringoire for Les Flambeurs d’hommes, published by Calmann-Lévy in 1934. This text was a strange hapax legomenon that consisted of an ethnographic narrative written entirely in the third person. At the same time that Jacques Soustelle defended his doctoral theses (La Famille Otomi-Pame du Mexique central was the principal thesis, La Culture matérielle des Indiens lacandons the secondary one),⁴ he also published Mexique, terre indienne with Éditions Grasset, a book that, according to commentary from the time, combines human emotions with scholarly observation⁵ and whose learned insight matches the heartfelt generosity.⁶ Between 1932 and 1939, Maurice Leenhardt published three lengthy volumes in the Travaux et mémoires de l’Institut d’ethnologie series that were devoted to the ethnography of New Caledonia and the Kanak language, but it was for Gens de la Grande Terre, a much more accessible work published by Gallimard in 1937, that he was awarded the Prix La Pérouse the following year.

Although a doctor by training, Jean-Albert Vellard (who would accompany Lévi-Strauss on his second Brazilian voyage) was assigned by Paul Rivet to carry out ethnographic observation during his July 1931–January 1933 expedition to Paraguay. Vellard returned with a long study titled Les Indiens guayakis, which was published in two parts in the Journal de la Société des américanistes in 1934 and 1935. This text was followed in 1939 by Une Civilisation du miel, which appeared in Gallimard’s Géographie humaine series and which, for Rivet, markedly distinguished itself from the hurried and superficial literature that the taste for exoticism and the ease of communication have so regrettably made fashionable.⁷ If Paul-Émile Victor’s voluminous field notes taken during his lonely stay with the Ammassalik Eskimos had not been misplaced at the Musée de l’Homme at the end of the 1930s, he would surely have written more than the two articles he published on string figures and cup and ball games in the Journal de la Société des américanistes in 1937 and 1938.⁸ At roughly the same time, however, he published his diary with Éditions Grasset in two revised volumes, titled Boréal and Banquise, that were accompanied by an appendix of over sixty pages (containing maps and a glossary, notes on equipment, and methods for attaching dogs to a sled, etc.) and that have since been reissued a number of times. One year before Victor left for Greenland, Alfred Métraux spent five months on Easter Island and supplemented his time in the field with two years of research at the Bishop Museum in Honolulu, after which he published the substantial Ethnology of Easter Island in 1940. A year later, his L’Île de Pâques⁹ appeared in Gallimard’s L’Espèce humaine series, a book Georges Bataille considered to be one of the masterpieces of French literature today . . . [that] leaves far behind the mass of novels received by the public as ‘literature.’¹⁰

If we allow for some expansion, our model of the ethnographer’s two books could also include the writings of a number of travelers from the 1930s who had an assignment from the Institut d’ethnologie or the Musée de l’Homme and who often returned to France with a haul of previously unpublished data (as the saying went at the time) and a wondrous story of their exotic experience. This was the case with Théodore Monod, for example, who published a predominantly archeological study titled L’Adrar Ahnet in 1932 in the Travaux et mémoires de l’Institut d’ethnologie series and, five years later, wrote a grandiose account of his travels across the Sahara called Méharées.¹¹

The recurrence of this second book prompts us to make two observations. First, this phenomenon is all the more striking since the interwar period saw the institutionalization of anthropology and saw ethnographers ceaselessly reiterate that they were not writers dabbling in literature and that the study of humankind had in fact broken with literature to enter the scientific age. The new discipline had cut its ties with the travel narrative and facile exoticism in order to ground itself in documentary evidence and established facts, eschewing local color and the desire to please a readerly public. Second, whether we think of Georges Balandier’s Afrique ambiguë, Georges Condominas’s We Have Eaten the Forest, Robert Jaulin’s La Mort sara, or Chronicle of the Guayaki Indians by Pierre Clastres (to name only books written before 1975), we can observe that, curiously, the pattern of the ethnographer’s two books has continued in other forms into the present, as if to solidify a specifically French tradition. Although all these books share a certain formal heterogeneity (since nearly all include maps, photographs, citations from indigenous literature, citations from travel journals, etc.), they also belong to widely varying genres ranging, for instance, from a protosurrealist journal secretly haunted by the model of Breton’s Nadja (L’Afrique fantôme), to an intellectual autobiography inspired by Proust (Tristes Tropiques), to a legendary account of the ancient Ethiopians (Les Flambeurs d’hommes), to a more conventional travel narrative (Mexique, terre indienne). Nonetheless, all of these supplements to the ethnographer’s expedition evince—and often explicitly so—an ambiguous relationship to the scholarly work they claim to complement. How can we account for this unusual distribution of texts? Why write two books at all, and why have French ethnographers sought to write literary works in addition to their scholarly texts? These are some of the major questions that pervade this book.

Since the 1980s, there has been considerable scholarly interest (principally in the United States but also in France and Switzerland) in the writing of ethnography and the construction of anthropological discourse. In the wake of Writing Culture, James Clifford and George Marcus’s foundational collection, ethnography from the so-called heroic period has been amply deconstructed.¹² Scholars have focused specifically on piecing together the genealogies of anthropological texts by tracing the process that leads from the field note to the finished work, examining all the operations that such a process entails: how does one move from firsthand, diverse, often turbulent and polyglot anthropological experiences to a monograph or ethnographic account intended for scholarly publication? Despite the obvious relevance of these questions and the excellent work to which they have given rise, this book proceeds differently, not in a genetic or vertical manner but in a horizontal or cross-sectional direction, one that considers all the written work published upon return from a given anthropological expedition. Thus I focus less on the construction of ethnographic knowledge than on its distribution among places and publics and on the ways in which ethnographic texts have been conceived and characterized. In order to study the relationship between anthropology and literature it does not suffice simply to uncover the rhetorical strategies or the tropes that ethnographers deploy. We must also situate their writings in a wider discursive space and consider how this space has been organized at a given historical moment, asking how these writings, and especially these second books, have been received, described, consecrated, or, on the contrary, neglected. Thus, this book does not take as its object the poetics of ethnographers or the question of writing in the social sciences, but, rather, investigates the relationship between anthropology and literature by considering how each developed in parallel to the other.

It is crucial, indeed, not to take the opposition between literature and the social sciences for granted. As we will see, the pairs formed by juxtaposing the ethnographer’s two books do not all express the same relationship: the relationship between L’Afrique fantôme and Leiris’s work on spirit possession among the Ethiopians of Gondar is not identical to the relationship obtaining between Métraux’s L’Île de Pâques and his Ethnology of Easter Island. These relationships are themselves quite different from the contrast between Bronislaw Malinowski’s Argonauts of the Western Pacific and the diary published by Malinowski’s widow twenty-five years after his death, to take a well-known example that allows us to highlight the originality of the French tradition.¹³

A major weakness of many studies that have taken up this question of the links between the social sciences and literature lies in the fact that these links are immediately set against analytic categories whose opposition is taken for granted: these studies oppose the objective to the subjective, explanation to interpretation, dispassionate reasoning to refined taste, the restrained writing of the scientist to the free and sovereign writing of the literary author, or the rigor and progress of science to the impressionism and traditionalism of letters. For example, the writing of Malinowski’s diary (which is indeed quite literary and clearly influenced by the novels of Conrad) is often explained by the opposition between objectivity and subjectivity since the scientific demands of ethnography are seen to suppress subjective expression which, by a sort of psychological mechanism, calls out for another type of book. From this perspective, because it is less distorted by the objectifying imperatives of scholarly science, A Diary in the Strict Sense of the Term provides us with a more authentic rendering of Malinowski’s experience in the Trobriand Islands and, at the same time, reveals the flaws and blind spots of a discipline that makes scant mention of the subjective conditions in which information is gathered.¹⁴

A similar reading is also common for Tristes Tropiques: after the figures and diagrams of The Elementary Structures of Kinship, the richness of Lévi-Strauss’s actual experience needed an outlet for its expression, and the man behind the mask of the ethnographic scholar needed at last to reveal himself with all his beliefs and emotions. These types of readings leave much to be desired. They are based on psychological assumptions and a limited conceptualization of science as the mere objectification of data. In short, they ignore the fact that anthropology, no less than the other sciences, constructs its objects. Only this work of construction (rather than the psychology of an author who feels torn between objective and subjective modes of self-expression) can explain the existence of an ethnographic residue that, by default, appears to be literary.

The opposition between scientific progress and the resistance of letters is the guiding principle of analyses inspired by Pierre Bourdieu’s sociology, analyses that conceive of science as a field whose development proceeds by way of a progressive emancipation from social constraints and demands, eventually allowing for the production of rational discourse. The reactions of men of letters to the rise of Durkheimian sociology have been examined in these terms, notably the virulent critiques made by Charles Péguy or the reactionary right led by Charles Maurras and the long-forgotten sociological projects of Paul Bourget and Henry Bordeaux.¹⁵ However, by opposingscientific culture to literary culture as rationality to blind faith, this type of analysis is too quick to assign right and wrong in the name of historical progress and a historical present taken for the norm. Once Durkheimian sociology is identified with the forward march of reason, every reaction it generates outside of itself (whether these be explicit attacks or attempts at reappropriation or expansion) is automatically disqualified since it is seen to sin against reason, democracy, and the progress of history. This approach thus ties together in the same obscurantist reaction such diverse critiques as those made by Péguy, Pierre Lasserre, Maurice Barrès, Bourget, Agathon, and, later, Breton and Bataille’s reproach of ethnography in the university. In addition to its Manichaeism, the major flaw of this type of explanation lies in the fact that it is always threatened with tautology: the reaction of literary writers to the social sciences is thus explained by the importation of the scientific model that transforms the organization of the university and by a consequent restriction of the prerogatives of the humanities. In short, there was a battle and a defeat, and the defeat is explained by the arrival of the victors.

One of the major stakes of this book is thus to avoid historical finalism and to reject the teleological narratives proposed by both scholars of science and literary histories. Likewise, this book is concerned not with advocating for the reinstatement in our cultural present of more or less forgotten ethnographic texts, but, rather, with extricating ourselves from facile narratives of historical progress, whether we are dealing with the history of science, of literature, or even of our own political and moral principles. After the end of the heroic period of anthropology and the collapse of colonial empires, postcolonial remorse, the critical history of science, and the new epistemology that claimed to reveal the relationship between knowledge and power all oriented numerous studies of ethnography toward a suspicious historicism that considers ethnography to be a misguided invention, an expression of latent ethnocentrism, or even a form of epistemic violence. This sense of historicism is bent on reestablishing certain continuities, which it accomplishes in several broad analytic moves: first, by highlighting what ethnography owes to nineteenth-century natural sciences and by evoking the close relationship between ethnography and early physical anthropology (but without elaborating on this point, since the observation alone is seen as damning enough); second, by retracing the links that connect the exhibition of savages at the Jardin d’acclimatation at the end of the nineteenth century, the colonial expositions, the Musée de l’Homme, and the more recent Musée du Quai Branly; finally, by isolating propositions made by sociologists from the interwar period and finding in them echoes of theories from past centuries of whose outdated or dangerous logic we are now well aware (that is, theories from whose barbarity we are fortunately far removed today).

As a result, this patient historical work on continuities claims to unveil the true professional ideology of this human science whose postulates finally come to light and are based on objectifying observation, that is, on the paradigm of the natural sciences, through which, in the reified categories of questionnaires . . . the ‘ethnographified savage,’ reduced to identity stereotypes, is trapped in the gaze of even the most open-minded white man.¹⁶ Thus the modern scholar expresses sincere remorse and, buoyed by the new lucidity afforded by this historical perspective, wholeheartedly rejects this blighted heritage: human sciences cannot, of course, be based on the model of the natural sciences, since humans are subjects and not objects. To conflate the two is an error that even common sense can discern (such an obvious fact should, however, inspire our mistrust) and, above all, a moral failing, since who among us would not staunchly resist being confined to a box on a sociological questionnaire?

The same process holds when it comes to textual analysis. In the United States in particular, scholars have demonstrated the persistence of tropes from travel or exploration narratives in ethnographic texts: Malinowski’s descriptions, for example, are based on the same literary devices as those of the explorer Mungo Park, and the arrival on the island of Tikopia in Raymond Firth’s We, the Tikopia is of a piece with the type of genre writing whose rules were notably established by James Cook and Louis-Antoine de Bougainville.¹⁷ Clifford Geertz claims to find in Tristes Tropiques a Richard Burton . . . sort of tone, returning from the source of the Nile, and calls for "a systematic attempt to connect Tristes Tropiques with the French travel literature [Lévi-Strauss] was supposedly reacting against, though actually reincarnating. For Geertz, such a study would highlight how Gide’s Voyages au Congo, the intensely read romantic travelogues of Pierre Loti, or even such a classic mandarin figure as André Malraux . . . seem the prototypes of the attitude, and the style, Lévi-Strauss is adopting here."¹⁸ An interpreter as skilled as Geertz, however, should have known that any act of interpretative reading runs the risk of finding in texts exactly what it has already put into them, but catching the author in his own trap and showing that a denunciation does not escape from what it denounces is a particular pleasure that the meticulous critic has trouble avoiding. Nonetheless, unveiling a contradiction in a text cannot constitute the final word on the matter, since it is also important to outline just what the contradiction produces. An author who does not do what he claims to do does not only contradict himself: he says things, he does others, and he contradicts himself. We must grasp all of this together if we want to understand more clearly what an author actually does in a given text.

Our task in this book will thus be to avoid the suspicious analytic position, and the chapters that follow will not raise these types of questions. This book does not seek to establish continuities among texts or practices, nor does it seek to reveal the artificial nature of the differences between ethnography and literature or to oppose the delusions of a field blinded by disciplinary logic and the need to justify itself. It is a matter not of exonerating ethnographers or blaming them, but rather of creating a position that does not hold up the present as a norm of any kind, whether scientific (We know today that Griaule’s theoretical explanations of Dogon cosmology are nothing but fiction, or, We have renounced the scientific pretensions of structuralism), moral (We are more respectful of differences and will no longer reify individuals by categorizing them), or aesthetic (We know today that the literature of the 1930s was more on the side of Georges Bataille than of Maurice Leenhardt). From the history of science, and from anthropology itself, we must retain the need for immanent understanding, and we must therefore refuse to distinguish between serious anthropological developments and other texts that the current state of the discipline allows us to dismiss as pseudoethnographic. Granted, we still read Mauss’s The Gift, while we hardly ever read Une Civilisation du miel by Jehan Vellard, who, after taming a young boy and capturing several adults, noted regretfully that the Guayaki Indians tolerate captivity poorly.¹⁹ Similarly, it hardly seems reasonable to consider the writings of Lévi-Strauss alongside those of Paul-Émile Victor or Monod who, we are given to understand, are great scientific thinkers but who, we suspect, are not actually great writers. However, this is precisely the type of legacy from which we must distance ourselves at the outset.

This is why this book takes as its point of departure a phenomenon as specific and identifiable as the double books of French ethnographers. By taking up these textual objects, which are historically situated and speak to the epistemological configuration of their time, we avoid the potential pitfall of speculating about the relationship between literature and anthropology in general as well as the dangerous axiological questions that would condemn us to retrospective illusion: Is this really anthropology? Is this really literature? These pairs of publications will thus serve as a focus for reflection, both because they seem objectively to embody the divisions between science and literature (without for the moment assigning too much content to these notions), and because they allow us to think through the meaning of such a distribution of knowledge with the help of concrete, historically situated documents.

Before turning to the texts themselves, however, we should perhaps linger for a moment on this relationship between science and literature that is taken almost naturally to be one of opposition. Indeed, it is usually taken for granted that these two categories constitute heterogeneous orders of discourse and thought, since the split between them is so often felt to be obvious in our everyday apprehension of cultural realities: it not only determines how schools and universities are organized but also appears to influence our understanding of psychology, since a person can be said to have a scientific or literary personality.

SCIENCE AND LITERATURE: A GENEALOGY

Since the beginning of the nineteenth century and from the perspective of scientific scholars, science’s relationship to literature was, first, one of emancipation. The disciplines constituted themselves through the delimitation of a field of knowledge that was carved out from the generalized knowledge of men of letters, and in this sense by becoming sciences they freed themselves from literature. Let us take the example of Georges Cuvier, who became director of the Muséum national d’Histoire naturelle which was founded in 1793. In the introduction to his Le Règne animal distribué selon son organisation, he begins by defining the proposed object of this science [natural history] and by establishing the rigorous limits between it and the contiguous sciences. He enumerates the different conceptions of nature, the general study of which is called physics, and goes on to distinguish general physics from particular physics as well as, within each of these, the various sciences according to their objects of study. Thus, general physics is subdivided into dynamic physics (itself subdivided into mechanics, statics, hydrostatics, etc.) and chemistry, particular physics overlaps with natural history with the exception of astronomy and meteorology, and natural history considers only inanimate bodies, called minerals, and the various kinds of living beings. Life being the most important of all the properties of beings, and the highest of all characters, minerals will be de facto excluded from Cuvier’s work, which makes him the founder of biology in the strict sense of the term.²⁰

Leaving aside the suitability or historicity of these divisions, it is clear that a science is first and foremost characterized by the delimitation of a territory that is proper to it from within the field of knowledge in general. Since we can see how this territory is marked off on one of its sides, namely in its distinction from the other sciences, we must also examine its definition on the other side in relation to those discourses that deal with various kinds of living beings but that do not fall under the auspices of natural science. In Cuvier’s day, this domain went by the name of letters or by what was then called literature (a meaning no longer in use), in reference to a broad form of knowledge, the res