Você está na página 1de 393

The Predicam.

ent of Culture
Copyright © 1 988 by the President and Fellows of Harvard Col l ege
All rights rese rved
Pri nted in the United States of America
10 9 8 7 6 5 4 3 2 1
Desig n ed by Joyce C. Weston

This book is printed on acid-free paper, and its b i n d i n g materials have been chosen
for strength and d u ra b i l i ty.

Library of Congress Cataloging in Publication Data

Clifford, j a mes, 1 945-

The predicament of c u l t u re twentieth-ce n t u ry eth nography,
l i t e ratu re, and art I James Clifford .
p. em.
B i b l iography: p .
I ncl udes index.
ISBN 0 -674-69842-8 (al k. paper) ISBN 0 -674-69843-6 ( p b k . : alk. pape r)
1 . Ethnology-H istory-20th century. 2. Ethnology-Philosop hy.
I . Title.
GN308.C55 1 988 87-241 73
306'.09-dc1 9 CIP

The chapters gathered in this book were written between 1979 and
1986. During these years I have enjoyed the encouragement of
friends and colleagues in many fields, most of whom I have publicly
thanked in earlier versions of some of these chapters. To mention
their names again here would result in a long, ultimately impersonal
list. I trust that those who have helped me know of my continuing
This book emerges from a period of unusual theoretical and
political questioning in several disciplines and writing traditions.
The provocation, criticism, and guidance I have received from many
others working along similar lines are only imperfectly acknowl­
edged in the book's citations.
For help in thinking through the chapters composed specifi­
cally for this volume I would like to thank James Boon, Stephen
Foster, George Marcus, Mary Pratt, Paul Rabinow, Jed Rasula, Renata
Rosaldo, W illiam Sturtevant, and Richard Wasserstrom.
I am grateful for fellowship support during the past seven years
viii A C KN O W L E D G M E N T S

from the American Council of Learned Societies, the American Phil­

osophical Association, and the National Endowment for the Human­
ities Summer Stipend program.
This book was written during my academic affiliation with the
History of Consciousness Program at the University of California,
Santa Cruz. I t reflects something of the ethos and energy of that
extraordinary group of scholars and graduate students. I would like
to single out for special thanks my colleagues Donna Haraway, Hay­
den W hite, and Norman 0. Brown.
Lindsay Waters of Harvard University Press offered acute edito­
rial suggestions. Special thanks to jacob, who shared the word pro­
cessor, and to my wife, Judith Aissen, for love and for not being
infinitely patient.

Santa Cruz, California j.C.

Permission to quote the following is gratefully acknowledged:

"To Elsie." William Carlos Williams, Collected Poems, Volume 1:
1909-1939. New York: Copyright 1938 by New Directions Publishing
Corporation. For British rights: The Collected Poems of William Car­
los Williams, Volume 1: 1909-1939. Edited by A. Walton Litz and
Christopher MacGowan. Carcanet Press, London, 1987. James Fen­
ton, "The Pitt-Rivers Museum, Oxford." Quoted from Children in
Exile: Poems 1968-1984. Copyright 1984 by Random House I nc. For
British rights: The Salamander Press, Edinburgh, 1983. Excerpt from
j. H. M. C. Boelaars, Headhunters About Themselves: An Ethno­
graphic Report from Irian }aya, Indonesia, Dordrecht: Martinus Ni­
joff, 1981. Copyright Koninklijk lnstituut voor Taal-, Land- en Volk­
enkunde, Leiden. Priere d'inserer for L'Afrique Fantome. From
Michel Leiris, Brisees, Paris: Mercure de France, 1966. For English
language rights, North Point Press. Lines from "Notebook of a Re­
turn to the Native Land," "Reply to Depestre," and "A Freedom in
Passage." From Aime Cesaire: The Collected Poetry, translated by
Clayton Eshleman and Annette Smith, Berkeley: University of Cali­
fornia Press, 1983.

Introduction: The Pure Products Go Crazy 1

Part One: Discourses 19

1. On Ethnographic Authority 21
2. Power and Dialogue in Ethnography: Marcel Griaule's
Initiation 55
3. On Ethnographic Self-Fashioning: Conrad and
Malinowski 92

Part 1\vo: Displacements 1 15

4. On Ethnographic Surrealism 1 17
5. A Poetics of Displacement: Victor Segalen 1 52
6. Tell about Your Trip: Michelleiris 1 65
7. A Politics of Neologism: Aime Cesaire 1 75
8. The Jardin des PI antes: Postcards 1 82

Part Three: Collections 1 87

9. Histories of the Tribal and the Modern 1 89
10. On Collecting Art and Culture 215

Part Four: Histories 253

11. On Orienta/ism 255
12. Identity in Mashpee 277

References 349
Sources 37 1
Index 373

18 "White man," Onyeocha. Cole and Aniakor 1984:150. Photograph by

Herbert M. Cole. Courtesy UClA Museum of Cultural History.

66 Marcel Griaule developing photographic plates. Courtesy Mission

Dakar-Djibouti Collection, Musee de I'Homme, Paris.

69 Marcel Griaule photographing. Courtesy Mission Dakar-Djibouti

Collection, Musee de I'Homme, Paris.

81 Marcel Griaule and Michel leiris prepare to sacrifice chickens.

Courtesy Mission Dakar-Djibouti Collection, Musee de I'Homme,

114 Store, avenue des Gobelins. Photograph by Eugene Atget. Courtesy

The Museum of Modern Art, New York; Abbott-levy Collection;
partial gift of Shirley C. Burden.

164 lgorot man, Philippines, exhibited at the 1904 St. louis World's Fair.
Neg. No. 324375. Courtesy Department library Services, American
Museum of Natural History, New York.

186 Indian woman spinning y arn and rocking cradle. Neg. No. 11604.
Courtesy Department library Services, Boas Collection, American
Museum of Natural History, New York.

194 The Making of an Affinity. (a) Pablo Picasso, Girl before a Mirror.
1932. Oil on canvas, 64 x 51%". Courtesy The Museum of Modern
Art, New York; gift of Mrs. Simon Guggenheim. (b) Kwakiutl Mask,
painted wood. Photograph by Gisela Oestreich. Courtesy Museum
fUr Volkerkunde, Berlin. (c) Girl be fore a Mirror (detail).

199 Affinities Not Included in the MOMA "Primitivism" Show. 1. Bodies.

(a) Josephine Baker. Courtesy the Granger Collection, New York.
(b) Wooden figure (Chokwe, Angola) published in Carl Einstein,
Negerplastik, 1915. (c) Fernand Leger, costume design for The Crea­
tion of the Worl d , 1922-23. Courtesy the Kay Hillman Collection,
New York.

204, Affinities Not Included in the MOMA " Primitivism" Show. 2. Collec-
205 tions. (a) Interior of Chief Shake's House, Wrangel, Alaska, 1909.
Neg. No. 46123. Photograph by H. I. Smith. Courtesy Department
Library Services, American Museum of Natural History, New York.
(b) View of the Margaret Mead Hall of Pacific Peoples. Courtesy De­
partment Library Services, American Museum of Natural History.

208 The Earth Deity, Ala, with her "children." Cole and Aniakor 1984:9.
Photograph by Herbert M. Cole. Courtesy UCLA Museum of Cul­
tural History.

2 1 0, Affinities Not Included in the MOMA "Primitivism" Show. 3. Appro-

211 priations. (a) Mrs. Pierre Loeb in her family apartment, Paris, 1929.
Courtesy Albert Loeb Gallery, Paris. (b) New Guinea girl with pho­
tographer's flash bulbs (included in the "Culture Contact" display at
the Hall of Pacific Peoples). Neg. No. 336443. Photograph by E. T.
Gilliard. Courtesy Department Library Services, American Museum
of Natural History, New York.

252 Inside a Hopi kiva. Neg. No. 28345. Photograph by Robert Lowie.
Courtesy Department Library Services, American Museum of Natu­
ral History, New York.
The Predicam.ent of Culture
We were once the masters of the earth, but since the gringos
arrived we have become veritable pariahs . . . We hope that
the day will come when they realize that we are their roots
and that we must grow together like a giant tree with its
branches and flowers.

Introduction: The Pure

Products Go Crazy

SOMETIME AROUND 1 92 0 i n a New Jersey s u b u rb of New York City, a

you ng doctor wrote a poem about a girl he ca l led E l s i e . He saw her
worki ng i n his kitchen or l a u n d ry room, h e l p i n g his wife with the house
c l ea n i ng or the kids. Someth i ng about her brought him u p short. She
seemed to sum up where everyth i ng was goi n g- h i s fam i l y, his fledgl i ng
practice, h i s art, the modern world that su rrounded and caught them a l l
i n its caree n i n g movement.
The poem Wi l l i am Carlos Wi l l iams wrote was a rush of associations,
beg i n n i ng with a famous assertion :

The p u re products of America

go c razy-

and conti n u i n g a l most without stopping for breath . . .

mounta i n fol k from Kentucky

or the ri bbed north end of

with its i so l ate l a kes and

va l l eys, its deaf-mutes, th ieves

old names
and prom iscuity between

dev i l-may-care men who have taken

to rai l road i ng
out of sheer l ust for adventu re-

and young s l atterns, bathed

in fi lth
from Monday to Saturday

to be tricked out that n ight

with ga uds
from i magi nations which have no

peasant trad itions to give them

but fl utter and fl a u nt

sheer rags-succu m b i n g without

save n u m bed terror

u nder some hedge of choke-cherry

or v i burnum-
whi ch they can not express-

U n less it be that marriage

with a dash of I nd i a n blood

wi l l th row up a gi rl so desol ate

so hemmed rou nd
with d i sease or m u rder

that she' l l be rescued by an

reared by the state and

sent out at fifteen to work in

some hard pressed
house in the suburbs-

some doctor's fam i l y, some E l s i e­

vol u ptuous water
ex pressing with broken

bra i n the truth about us-

her great
u n ga i n l y h i ps and floppi n g breasts

add ressed to cheap

jewel ry
and r i c h you n g men with fine eyes

when sudden l y the a ngry descri ption veers :

as i f the earth u nder o u r feet

an excrement of some sky

and we degraded prisoners

desti ned
to h u nger u nt i l we eat fi lth

wh i le the i magi n ation stra i n s

after deer
goi n g by fields of go ldenrod i n

the stifl i n g heat o f Septem ber

it seems to destroy u s

It i s o n l y i n i so l ate flecks that

someth i ng
i s given off

N o one
to witness
and adj u st, no one to d rive the car

These l i nes emerged en route i n Wi l l i ams' dada treati se on the i mag­

i n ation, Spring & All ( 1 9 2 3 ) . I hope they can serve as a pretext for th i s
book, a way o f starting i n w i t h a pred i cament. Cal l t h e pred i cament eth­
nograph i c modern ity: eth nograph ic beca use Wi l l i ams finds h i mself off
center among scattered traditions; modern ity s i nce the cond ition of root­
lessness and mob i l ity he confronts i s an i nc reas i ngly common fate . "EI-

sie" stands s i m u ltaneously for a l oca l c u ltural brea kdown and a col lec­
tive future . To Wi l l i a ms her story is i nescapably h i s, everyone's . Looki ng
at the "great/unga i n ly h i ps and flopp i ng breasts" he feel s th i ngs fa l l ing
apart, everywhere. Al l the beautifu l , pri m itive pl aces are ru i ned . A k i nd
of cu ltura l i n cest, a sense of runaway h i story pervades, drives the rush of
assoc iations.
Th i s fee l i ng of lost authenticity, of "modern ity" ru i n i ng some essence
or sou rce, is not a new one. In The Country and the City ( 1 97 3 ) Raymond
Wi l l iams finds it to be a repetitive, pastora l "structu re of feel i ng." Aga i n
and aga i n over the m i l le n n i a change i s configu red a s d isorder, pure prod­
ucts go c razy. B ut the i mage of E lsie suggests a new tu rn . By the 1 920s a
tru ly globa l space of c u ltural con nections and d i ssol utions has become
imaginable: local authenti cities meet and merge in trans ient u rban and
suburban setti ngs-setti ngs that wi l l i nc l ude the i m m i grant neighbor­
hoods of New J ersey, m u lticu ltura l sprawls l i ke Buenos A i res, the town­
sh i ps of Joha n nesb u rg. Wh i le Wi l l iam Carlos Wi l l iams invokes the pure
products of America, the "we" caree n i n g i n h i s driverless car is clearly
someth i n g more . The eth nograp h i c modern i st searches for the u n iversa l
i n the loca l , the whole i n the part . Wi l l ia ms' famous choice of a n Amer­
ican (rather than Engl ish) speec h , h i s regiona l ly based poeti c and med i ­
cal practice m ust not c u t h i m o ff from t h e most genera l h u m a n processes.
H is cosmopo l ita n i sm req u i res a perpetual veering between local attach­
ments and genera l poss i b i l ities .
E l s i e d i srupts t h e project, for her very exi stence ra i ses historical un­
certai nties u nderm i n i ng the modern i st doctor-poet's secu re position . 1 H i s
response t o t h e d i sorder she represents is comp lex a n d ambiva l ent. I f
authentic trad itions, t h e p u re products, are everywhere yielding to prom­
i scu ity and a i m l essness, the option of nosta l g i a holds no charm . There i s
n o go i ng back, no esse nce t o redee m . Here, a n d th roughout h i s writi ng,
Wi l l i ams avo ids pastora l , fol kl oristic appea l s of the sort common among
other l i bera l s in the twenties-exhorti ng, preserv i ng, col lecti n g a true
rural cu ltu re in endangered p laces l ike Appa lach i a . Such authentic ities
wou l d be at best a rtific i a l aestheti c purifications (Wh i snant 1 983). Nor
does Wi l l iams settle for two other common ways of confronti ng the rush

1. "Elsie" a lso d isplaces a l i terary trad ition . I n Western writing servants have
always performed the chore of representing "the people" -lower cl asses and
d i fferent races . Domesticated outsiders of the bou rgeois imagi nation, they regu­
larly provide fictional epi phan ies, recogn ition scenes, happy endings, utopic and
d i stopi c transcendences. A bri l l iant survey is provided by Bruce Robbins 1986.

of h i story. H e does not evoke E l s i e and the i d i ocy of rural l i fe to ce lebrate

a progress i ve, tech nological futu re . He shares her fate, for there rea l l y i s
"no o n e t o d r i ve t h e car"-a fri ghten i ng cond ition . N o r does Wi l l iams
res ign h i mself sad l y to the l oss of l oc a l trad itions i n an entropic moder­
n ity-a v i s i o n com mon among prophets of c u l tural homogen ization, l a­
mente rs of the ru i ned tropics. I n stead, he c l a i m s that "someth i ng" i s sti l l
bei n g "given off" - i f o n l y i n " i so l ate flecks ."
It is worth dwe l l i ng on the d iscrepancy between th i s emergent, d i s­
persed "someth i ng" a nd the car i n w h i c h "we" a l l ride. Is it poss i b l e to
res i st the poem 's momentu m , i ts rushed i nevitab i l ity? To do so i s not so
m u c h to offe r a n adeq uate read i n g (of a poeti c seq uence a bstracted from
Spring & A//) as it i s to reflect on severa l read i ngs, on severa l h i storical
" E i s i es ." Let th i s p rob lematic figu re with her "dash of I nd i a n blood," her
u n ga i n l y fem a l e form, her i na rti c u l ateness stand for grou ps margi n a l i zed
or s i lenced i n the bou rgeoi s West: "natives," wome n , the poor. There i s
violence, c u riosity, p ity, a nd desi re i n t h e poet's gaze. E l s i e provokes very
m i xed emotions. Once aga i n a fem a l e, poss i b l y colored body serves as
a s i te of attracti o n , repu l s i o n , symbo l i c appropriati o n . E l s i e l ives o n l y for
the eyes of pri v i l eged men . An i n a rti c u l ate m udd l e of lost origins, she i s
goi n g nowhere. Wi l l iams evokes this w i t h h is angry, b l e a k sym pathy­
and then turns i t a l l i n to modern h istory. Two-th i rds of the way th rough
the poem, E l s i e's perso n a l story sh ifts toward the genera l ; her own path
through the s u b u rban kitchen va n i shes. She, Wi l l iams, a l l of us a re
caught i n modern ity's i nescapable momentu m .
Someth i n g s i m i l a r occu rs whenever marg i n a l peoples come i nto a
h istoric al or ethnogra p h i c s pace that has been defi ned by the Western
i magi n ation . " Ente r i n g the modern worl d ," the i r d i sti nct h istories q u ickly
van i s h . Swept u p in a destiny dom i nated by the capita l i st West and by
various tec h n o l ogica l l y advanced soc i a l isms, these sudden l y " back­
ward" peoples no longer i nvent local futu res . What i s d i fferent about
them rem a i ns tied to trad itiona l pasts, i n herited structu res that e i ther re­
s i st or yield to the new but can not prod uce it.
Th i s book proposes a d i fferent h i storical vision . It does not see the
world as pop u l ated by endange red a uthentic i ties-p u re prod ucts a l ways
goi ng crazy. Rather, it m akes space for specific paths through modern i ty,
a recogn ition a ntici pated by Wi l l iams' d i screpant question: what i s
"given off" b y i nd iv i d u a l h i stories l i ke E l s i e's ? Are t h e " i solate flecks"
dyi ng sparks? N ew begi n n i ngs ? Or . . . ? "Com pose . (No ideas/but i n
th i ngs) I nvent!" T h i s was Wi l l iams' s l ogan ( 1 967 : 7) . I n Spring & All the

h u m a n futu re i s someth i ng to be creatively i m agi ned, not s i m p l y en­

d u red : "new form dea lt with as rea l ity itsel f . . . To enter a new world,
and have there freedom of movement and newness" ( 1 923 : 70, 7 1 ) . But
geopo l itical q u estions m ust now be asked of every i nventive poetics of
rea l ity, i nc l ud i ng that u rged by th i s book: Whose rea l ity? Whose new
world ? Where exactly does anyone stand to write "as if the ea rth u nder
our feet/were a n excrement of some s ky/and we . . . destined . . ." ?
Peopl e and th i ngs are i ncreas i ngly out of pl ace. A doctor-poet­
fieldworker, Wi l l iams watches and l i stens to N ew Jersey's i m m igrants,
workers, women giving b i rth, p i m p l y-faced teenagers, mental cases. I n
the i r l ives and words, encountered through a privi l eged part i c i pant ob­
servation both poetic and scientific, he finds material for h i s writi ng. Wi l­
l i ams moves freel y out i nto the homes of h i s patients, keeping a med ica l­
aesthetic d i stance (though someti mes with great d ifficu lty, as i n the
"beautifu l th i ng" seq uences of Paterson, book 3). The meeting with E l s i e
i s somehow d i fferent: a troubl i n g outsider t u r n s up inside bourgeoi s do­
mestic space . She can not be held at a d i stance.
Th i s i nvas ion by a n ambiguous person of q uestionable origin antic­
i pates deve lopments that wou ld become widely apparent only after the
Second World War. Co lon i a l relations wou ld be pervasively contested .
After 1 9 5 0 peoples long spoken for by Western eth nographers, adm i n i s­
trators, and m i ssiona ries began to speak and act more powerfu l l y for
themsel ves on a global stage . It was i ncreasi ngly d i fficu lt to keep them
in thei r (trad itional) pl aces . D i sti nct ways of l i fe once desti ned to merge
i nto "the modern world" reasserted the i r d i fference, i n novel ways. We
perceive E l s i e d i fferently i n l ight of these developments.
Read i ng aga i n st the poem's momentu m , from new positions, we are
able to wonder: What becomes of th i s girl after her sti nt i n Wi l l ia m Car­
los Wi l l iams' kitchen ? Must she sym bo l ize a dead end ? What does E l s i e
prefigure ? A s woma n : h e r unga i n l y body is either a symbol o f fa i l u re i n
a world dom i nated by t h e male gaze or t h e i mage o f a powerfu l , "dis­
orderly" fem a l e form, a n a l ternative to sexi st defi n itions of bea uty. As
i m p u re prod uct: th i s m i x of backgrounds i s either an u prooted lost sou l
o r a new hybrid person , less domestic than the subu rba n fam i ly home
she passes t h rough . As American I nd i a n : E l s i e i s e ither the l ast a l l-but­
assi m i l ated remnant of the Tuscaroras who, accord ing to trad ition, settled
i n the Ramapough h i l l s of Northern N ew Jersey, or she represents a Na­
tive American past that i s be i ng tu rned i nto an u nexpected futu re. (Dur­
i ng the l ast decade a gro u p of E l s i e's k i n ca l l i ng themselves the Rama­
pough Tri be have active ly asserted an I ndian identity. )2 Wi l l iams'
T H E P U R E P R O D U C T S G O C R AZ Y 7

assi m i lation of h i s sym bo l i c servant to a shared destiny seems l ess defi n­

itive n ow.
" E l s i e," read i n the l ate twentieth centu ry, is both more spec ific and
l ess determ i ned . Her poss i b l e futu res reflect an u n resol ved set of c ha l­
lenges t o Western v i s i ons o f modern ity-ch a l lenges that reson ate
throughout t h i s book. E l s i e is sti l l l arge l y si l ent here, but her d i sturb i ng
presences-a p l ural ity of emergent subj ects-can be felt.3 The time i s
past w h e n priv i l eged a uthorities cou ld routi nely "give voice" (or h istory)
to others without fea r of contrad iction . "Croce's great d i ctum that a l l h is­
tory is contem porary h i story does not mean that a l l h i story is our contem­
porary h i story . . . " (Jameson 1 98 1 : 1 8) When the preva i l i n g narratives of
Western identity are contested , the pol itica l i ssue of h i story as emergence
becomes i n escapable. j u l i et Mitchel l writes in Women: The Longest Rev­
olution ( 1 984) : "I do not th i n k that we can l i ve as h uman subj ects without
in some sense tak i n g on a h i story; for us, it i s m a i n ly the h i story of bei n g
men or women u nder bou rgeoi s capita l i s m . I n decon structi ng that h i s­
tory, we can o n l y construct other h istories. What are we i n the process
of beco m i ng?" (p. 294). We a re not a l l together in Wi l l i ams' car.

O n l y one of E l s ie's emergent possibi l ities, the one con nected with her
"dash of Indian b l ood ," i s explored i n th i s book. During the fal l of 1 9 7 7
i n B oston Federa l Court t h e descendants o f Wam panoag I nd i ans l iv i ng i n
Mash pee, "Cape Cod's I n d i a n Town," were req u i red t o prove thei r iden­
tity. To esta b l i s h a lega l right to sue for lost l ands these citi zens of modern
Massachusetts were asked to demonstrate conti nuous tri bal ex istence
s i nce the seventeenth centu ry. Life in Mashpee had cha nged d ramati-

2 . The Native American ancestry of the isol ated and i n bred Ramapough
mounta i n people ("old names" . . . from "the ri bbed north end of/Jersey") i s de­
batable. Some, l i ke the fol klorist David Cohen (1974), deny it al together, de­
bunking the story of a Tuscarora offshoot. Others bel i eve that this m i xed popu­
lation (formerly cal led jackson's Wh ites, and drawing on bl ack, Dutch, and
Engl ish roots) probably owes more to Del aware than to Tuscarora Indian blood .
Whatever i ts real h i storica l roots, the tribe as presently constituted is a l ivi ng
i m p u re prod uct.
3. " Natives," women, the poor : this book d iscusses the eth nographic con­
struction of only the first group. In the dom inant ideological systems of the bour­
geo is West they a re i n terrel ated , and a more systematic treatment than mine
wou l d bri ng th is out. For some begi n n i ngs see Duvignaud 1973; Alloula 1981;
Tri n h 1987; and Spivak 198 7 .

cal l y, however, si nce the fi rst contacts between Engl ish Pi lgrims at Plym­
outh and the Massachusett-speaking peoples of the region . Were the
p l a i ntiffs of 1 9 7 7 the "same" I nd i ans? Were they someth i ng more than a
col lection of ind ivid u a l s with varying degrees of Native American ances­
try ? If they were d i fferent from the i r neigh bors, how was the i r "tribal"
difference man ifested ? During a long, wel l -publ icized trial scores of I n­
d ians and whites testified about l i fe i n Mashpee. Professional h i storians,
anth ropologists, and soc iologists took the stand as expert witnesses. The
bitter story of N ew England I nd ians was told i n m i n ute deta i l and vehe­
mently debated . In the confl ict of interpretations, concepts such as
"tri be," "cu l tu re," "identity," "ass i m i l ation ," "ethn icity," "po l itics," and
"co m m u n i ty" were themse l ves on trial . I sat through most of the forty
days of argu ment, l i stening and tak i ng notes.
It seemed to me that the tri a l - beyond its i m med iate pol itical
stakes-was a cruc i a l experiment in cross-cu ltural translation . Modern
Indians, who spoke in N ew England-accented Engl ish about the G reat
Spi rit, had to convi nce a wh ite Boston j u ry of their authenticity. The
translation process was fra ught with ambigu i ties, for a l l the c u l tu ra l
bou ndaries at issue seemed to b e b l u rred and shifting. T h e t r i a l raised
far-reach i n g question s about modes of cu ltu ra l i nterpretation, i m p l icit
models of wholeness, styles of d i stanci ng, stories of h i storica l deve lop­
I began to see such q u estions as symptoms of a pervas ive postcolo­
n i a l crisis of eth nographic authority. Wh i l e the crisis has been fe lt most
strongly by former l y hegemonic Western d i scou rses, the q uestions it
ra ises a re of global sign ificance . Who has the authority to speak for a
group's identity or authenticity ? What are the essential e lements and
bou ndaries of a cu lture ? How do self and other c l ash and converse in the
encounte rs of eth nography, travel , modern intereth n i c re lations? What
narrati ves of deve lopment, l oss, and i n novation can account for the pres­
ent range of local oppos itional movements ? During the tri a l these q ues­
tions assumed a more than theoretical u rgency.
My perspective i n the cou rtroom was an obl ique one. I had j ust
fi n i shed a Ph . D. thesis i n h istory with a strong i nterest i n the h i story of
the h u man sciences, particu larly c u ltural anthropo logy. At the time of the
trial I was rewriti ng my d i ssertation for publ ication. The thesis was a
biography of Mau rice Leenhardt, a m i ssionary and eth nographer i n
French N ew Caledonia a n d an eth nologist i n Paris (C l ifford 1 982a) . What
cou l d be farther from New England I nd ians ? The connections turned out
to be c lose and provocative .

I n Melanesia Leen hardt was deeply i nvo l ved with tribal groups who
h ad experienced a co l o n i a l assa u l t as extreme as that i nfl icted i n Mas­
sac h usetts. He was preoccupied with practical and theoretical prob lems
of c u ltural change, syncretism, conversion, and s u rviva l . L i ke many
American I nd ians the m i l itari l y defeated Kanaks of New Caledon ia had
"triba l " i n stitutions forced on them as a restrictive reservation syste m .
Both gro u ps wou l d m a ke strategic accom modations with these external
forms of government. N ative Americans and Melanesians wou ld s u rvive
periods of acute demogra p h i c and c u ltura l crisis, as wel l as periods of
cha nge and revi va l . Over the l ast h u nd red yea rs New Caledo n ia's Kanaks
h ave managed to find powerfu l , d isti nctive ways to l ive as Melanesians
in a n i n vasive world . It seemed to me that the Mash pee were strugg l i n g
toward a s i m i l a r goa l , reviving a n d i nventing ways t o l ive as I ndia ns i n
the twentieth centu ry.
Undoubted l y what I heard i n the New England cou rtroom i nflu­
enced my sense of Melanesian identity, someth i n g I came to u nderstand
not as an arc h a i c s u rvival but as an ongoing process, po l itica l l y con­
tested and h i storica l l y u nfi n i shed . I n my stud ies of Eu ropean eth no­
graphi c i nstitutions I h ave c u l tivated a s i m i l a r attitude.

This book i s concerned with Western visions and practices . They are
show n , however, respond i n g to forces that cha l l enge the authority and
even the futu re identity of "the West." Modern ethnography appears i n
several forms, trad itional a n d i n novative . A s a n academ i c practice it can­
not be separated from anth ropo logy. Seen more genera l ly, it i s s i m p l y
d i verse ways o f th i n ki n g a n d writi ng about c u lture from a standpoint of
part i c i pant observation . I n t h i s expanded sense a poet l i ke Wi l l iams is an
eth nogra pher. So are many of the people soc i a l scienti sts h ave ca l led
"native i nformants." U lti mate ly my topic is a pervasi ve cond ition of off­
centered ness i n a world of d i sti nct mea n i n g systems, a state of being i n
c u l tu re w h i l e looking a t c u ltu re, a form o f personal a n d col lective self­
fas h i o n i n g . T h i s p red icament- not l i m ited to schol a rs, writers, artists, or
i nte l l ectuals - responds to the twentieth centu ry's u n precedented overlay
of trad itions. A modern "eth nography" of conj u nctu res, constantly mov­
i ng between c u ltures, does not, l i ke its Western alter ego "anthropo l ogy,"
asp i re to s u rvey the fu l l range of h u m a n divers ity or development. It is
perpetu a l l y d ispl aced, both regiona l ly focu sed and broad ly comparative,
a form both of dwe l l ing and of trave l in a world where the two experi­
ences a re less and l ess d i sti nct.
10 I NTR O D U C T I O N

Th is book migrates between local and global perspectives, con­

stantly recontextua l i z i n g its topic. Part One focuses on strategies of writ­
i ng and representation , strategies that change h i storica l ly i n response to
the genera l s h ift from h igh colon i a l ism around 1 900 to postcolon ial ism
and neocolon i a l i sm after the 1 950s. In these chapters I try to show that
eth nograph ic texts a re orchestrations of mu ltivocal exchanges occu rri ng
in pol itica l ly charged situations. The subjectivities prod uced in these
often u neq ual exc hanges-whether of "natives" or of visiting partici pant­
observers-are constructed domains of truth , serious fictions. Once this
is recogn ized, d i verse i nventive possibi l ities for postco lonial eth no­
graphic representation emerge, some of which are su rveyed in th is book.
Part Two portrays eth nography i n a l l iance with avant-garde art and c u l ­
tura l critic ism, activ ities with wh ich it shares modern i st proced u res of
co l l age, juxtaposition, and estrangement. The "exotic" is now nearby. I n
th is section I a l so probe the l i m its of Western eth nography through sev­
era l self-reflexive forms of travel writi ng, exploring the poss i b i l ities of a
twentieth-centu ry " poetics of d i splacement." Part Three turns to the h i s­
tory of co l lecting, particularly the c l assification and display of "pri m i ­
tive" art a n d exotic "cultures." M y genera l a i m i s to d i splace any
transcendent regime of a uthenti city, to argue that a l l authoritative col lec­
tions, whether made in the name of art or sc ience, are h i storica l ly con­
tingent and subject to local reappropriation . In the book's fi nal section I
explore how non-Western h i storica l experiences-those of "orientals"
and "triba l " N ative Americans-are hem med i n by concepts of conti nu­
ous trad ition and the u n i fied se lf. I argue that identity, considered eth­
nograph ica l l y, m ust always be m ixed, re l ation a l , and i nventive .
Se lf-identity emerges as a complex c u ltural problem in my treatment
of two polyglot refugees, joseph Con rad and B ron i s l aw Mal i nowsk i ,
Poles s h i pwrecked i n England a n d Engl i s h . Both m e n prod uced sem inal
med itations on the local fictions of col l ective l i fe, and, with different
degrees of i rony, both constructed identities based on the acceptance of
l i m ited rea l ities and forms of expression. Embracing the serious fiction of
"culture," they wrote at a moment when the ethnograph ic (relativist and
p l u ral) idea began to attai n its modern cu rrency. Here and e l sewhere i n
the book I try to h i storic ize a n d see beyond t h i s cu rrency, stra i n i ng for a
concept that can preserve cu lture's differenti ati ng functions w h i l e con­
ceiving of col l ective identity as a hybrid, often d i sconti nuous inventive
process. Cu lture is a deeply com prom ised idea I cannot yet do without.
Some of the pol itical dangers of cu ltura l i st red uctions and essences
T H E P U R E P R O D U C T S G O C R AZY ll

are explored i n my analysis of Edward Sa id's po lemica l work Orienta/ism

( 1 978a). What emerges is the i nherently d i sc repant sta nce of a post­
colon i a l "oppositional" critic, for the construction of simp l ify i n g es­
sences and d i sta nci ng d ic hoto m i es is clearly not a monopoly of Western
Orienta l ist experts . Said h i mself writes in ways that s i m u ltaneously assert
and su bvert h i s own authority. My analysis suggests that there can be no
fi nal smooth i n g over of the d i sc repanc ies in h is d i scou rse, s i nce it i s i n­
c reas i n g l y d ifficu lt to ma i nta i n a c u ltural and pol itical position "outside"
the Occident from w h i c h , in security, to attack it. Critiq ues l i ke Said's are
caught i n the doub l e eth nograph i c movement I have been evoking. Lo­
cal ly based and pol itica l ly engaged , they m ust resonate globa l ly; wh i l e
they engage pervasive postcolo n i a l processes, they do s o without over­
view, from a b latantly partial perspective.
I n terve n i n g in an i ntercon n ected world, one is a lways, to vary i ng
degrees, " i n authentic " : caught between cu ltu res, i m p l icated i n others .
Because d isco u rse i n global power systems is elabo rated vi s-a-vis, a
sense of d i fference or d isti nctness can never be located so lely i n the con­
t i n u ity of a c u lture or trad itio n . Identity is conj u nctu ra l , not essentia l .
Sa i d add resses these i ssues most affecti ngly i n After the Last Sky, a recent
evocatio n of " Palest i n ian Lives" and of his own pos ition among them
( 1 986a : 1 5 0) : "A part of somet h i n g i s for the foreseeable futu re goi ng to
be better than a l l of it. F ragments over wholes. Restless nomad i c activ ity
over the settleme nts of held territory. Criticism over res i gnation . The Pa l­
est i n i a n as self-consciousness in a barre n plain of i nvestments and con­
sumer a ppetites . The hero i s m of anger over the begg i ng bow l , l i m ited
i ndependence over the status of c l ients . Attention, alertness, focu s . To
do as others do, but somehow to stand a part. To te l l you r story i n pieces,
as it is ." This work appeared as I was fi n i s h i n g my own book . Th u s my
d iscussion of Orienta/ism merely antici pates Said's ongo i n g searc h for
nonessentia l ist forms of cu ltura l pol itics. After the Last Sky active l y i n ­
habits t h e d i screpancy between a spec ific condition o f Pa l esti n i a n ex i l e
and a more general twentieth-centu ry range of options. It i s (and i s not
on ly) as a Pa l esti n i a n that Said movi ngly accepts "ou r wanderings,"
pl ead i ng for "the open sec u l a r element, and not the symmetry of re­
demption" ( p . 1 50) .

I share th i s suspicion of "the symmetry of redemption ." Questionable acts

of p u rification a re i nvolved in any atta i n ment of a prom i sed land, retu rn
12 I N T R O D UC T I O N

to "origi n a l " sou rces, o r gathering u p of a true trad ition . Such c l a i m s to

pu rity a re i n any event a l ways su bverted by the need to stage authenticity
in opposition to externa l , often domi nating alternatives . Thus the "Th i rd
World" p l ays itsel f against the " F i rst World," and vice versa. At a local
leve l , Trobriand I s l anders i nvent their cu lture with i n and aga i n st the con­
texts of recent co lonial h i story and the new nation of Papua-New
G u i nea. If authenticity is relationa l , there can be no essence except as a
pol i tical , c u ltura l i nvention , a loca l tactic.
I n th is book I question some of the loca l tactics of Western eth nog­
raphy, focus i n g on redemptive modes of textual ization and particularly
of col lecting. Severa l chapters analyze i n some deta i l the systems of au­
thenticity that have been i m posed on creative works of non-Western art
and cu lture . They l ook at col lecting and authenticating practices in con­
temporary setti ngs : for exa m p l e the controversy su rrou nding an exh ibi­
tion at the Museum of Modern Art i n N ew York City over the relations
between "tri ba l " and "modern" a rt. How have exoti c obj ects been given
va l u e as "art" and "cu lture" i n Western col lecting systems? I do not argue,
as some c ritics have, that non-Western objects are properl y understood
o n l y with reference to the i r origi nal m i l ieux. Eth nograph ic contextua l i­
zations are as p roblematic as aesthetic ones, as susceptible to purified,
a h i storical treatment.
I trace the modern h i story of both aesthetic and eth nographic c l as­
sifications in an earl ier setting: avant-garde Paris of the 1 920s and 1 93 0s,
a rad ical context I ca l l eth nographic su rreal ism. Two i nfl uential mu­
seums, the Musee d ' Eth nographie d u Trocadero and its scientific succes­
sor, the M u see de I ' Hom me, symbol ize d i sti nct modes of "art and cultu re
col lecti ng." Thei r j u xtaposition forces the q u estion : How are eth no­
graphic worlds and thei r meani ngfu l a rtifacts cut up, sal vaged, and va l ­
ued ? Here c u lture appears not as a trad ition t o b e saved b u t a s assembled
codes and artifacts always suscepti ble to critical and creative recombi­
natio n . Eth nography is an expl icit form of cu ltura l critique shari ng rad ical
perspectives with dada and su rrea l ism. I n stead of acq u iesci ng in the
separation of ava nt-garde experi ment from d i sc i p l i nary science, I reopen
the frontier, suggesting that the modern d ivision of art and eth nography
i nto d i sti nct institutions has restricted the former's analytic power and the
latter's su bversive vocation .
S i nce 1 900 i nc l usive co l l ections of "Manki nd" have become i nsti­
tutional ized in academ ic d i sc i p l i nes l i ke anth ropo logy and in m useums
of a rt o r eth nology. A restrictive "art-cu ltu re system" has come to contro l
T H E P U R E P R O D U C T S G O C R AZY l3

the authenticity, va l ue, and c i rc u l ation of artifacts and data . Analyz i ng

this system, I propose that any co l l ection i m p l ies a tem pora l vision gen­
e rat i n g rarity and worth , a meta h istory. This h i story defi nes which groups
or th i ngs w i l l be redeemed from a d i s i ntegrating h u man past and which
w i l l be defined as the dynamic, or tragic, agents of a common desti ny.
My a n a l ysis works to bring out the loca l , pol itica l conti ngency of such
h i stories and of the modern col lections they j ustify. Space is c l eared ,
perhaps, for a l ternatives .

T h i s book is a spl iced eth nographic object, an i ncomplete col l ecti o n . It

consists of explorations written and rewritten over a seven-yea r period .
Its own h i storical moment has been m arked by ra pid changes i n the
terms-sc iefltific, aesthetic, and textua l - gover n i n g cross-c u ltu ra l rep­
resentation . Wri tten from with i n a "West" whose authority to represent
u n ified h u man h i story is now widely c h a l lenged and whose very spatial
identity is i n c reasi ngly p roblematic, the explorati ons gathered here can­
not- s h o u l d not-add up to a seam l ess vision . The i r partial ity i s appar­
ent. The chapters vary i n form and style, reflecting d i verse conj u nctu res
and spec ific occasions of composition. I h ave not tried to rewrite those
a l ready p u b l ished to prod uce a consistent veneer. Moreover, I have in­
c l uded texts that active l y break u p the book's preva i l i ng tone, hoping in
th i s way to manifest the rhetoric of my accou nts . I p refer sharply focused
pictu res, composed i n ways that show the frame or lens.
Eth nography, a hybrid activity, thus appears as writing, as col lecti ng,
as modern ist col l age, as i mperi a l power, as su bversive criti q u e . Viewed
most broad l y, perhaps, my topic is a mode of trave l , a way of understand­
ing and getting around in a d i verse world that, si nce the si xteenth cen­
tury, has become cartograph i ca l ly u n i fied . One of the principal fu nction s
o f eth nography is "orientatio n " (a term l eft over from a t i m e when E u rope
traveled and i n vented itself with res pect to a fantastica l l y u n i fied " East") .
B u t i n the twentieth centu ry eth nography reflects new "spati a l practices"
(De Certea u 1 984), new forms of dwe l l i ng and c i rc u l ating.
T h i s centu ry has seen a d rastic expansion of mobi l ity, i nc l ud i ng tou r­
ism, m i grant labor, i m m i gration, u rban sprawl . More and more people
"dwe l l " with the h e l p of mass trans it, automobi l es, a i rp l a nes. I n c ities on
six conti nents foreign pop u l ations have come to stay- m i x i n g in but
often in parti a l , spec ific fash ions. The "exotic" i s u ncan n i l y c l ose . Con­
verse l y, there seem no d i stant places l eft on the pl anet where the pres-
14 I N T R O D UC T I O N

ence of " modern" prod ucts, med i a, and power can not be felt. An older
topography and experience of travel is exploded . One no longer leaves
home confident of find ing something rad i ca l l y new, another time or
space. D ifference is encountered in the adjoi n i ng neighborhood , the fa­
m i l iar turns up at the ends of the ea rth . Th is d i s-"orientation" is reflected
throughout the book. For example twentieth-centu ry academic ethnog­
raphy does not a ppear as a practice of i nterpreting d i stinct, whole ways
of l ife but i nstead as a series of s pecific d i a l ogues, i m positions, and i n­
venti ons. "Cu ltura l " d i fference is no longer a stable, exotic otherness;
self-other relations are matters of power and rhetoric rather than of es­
sence . A whole structure of expectations about authenticity in cu lture
and i n art is th rown i n doubt.
The new rel ations of eth nographic d i splacement were registered
with precoc ious c l a rity i n the writi ngs of Victor Sega len and Michel
Lei ris. Both wou ld h ave to u n learn the forms that once organized the
experience of travel i n a time when " home" and "abroad," "self" and
"other," "savage" and "c i v i l i zed" seemed more clearly opposed . The i r
writings betray a n u nease w i t h narratives o f escape a n d retu rn, o f i n itia­
tion and conqu est. They do not c l a i m to know a distanced "exotic,'' to
bring back its secrets, to objective ly describe its landscapes, customs,
languages. Everywhere they go they register complex encou nters. In Se­
ga len's words the new traveler expresses "not simply h i s vision, but
through an i n stanta neous, constant transfer, the echo of his presence."
C h i na becomes an a l legorical m i rror. Lei ris' fie ldwork in a "phantom
Africa" throws h i m back on a relentless se lf-eth nography-not auto­
biography but an act of writi ng h i s existence in a present of memories,
d reams, pol itics, d a i l y l ife.
Twentieth-centu ry identities no longer presuppose conti nuous c u l ­
tu res or trad itions. Everywhere i nd ividuals a n d groups i mprov ise local
performa nces from (re)col lected pasts, d rawing on foreign media, sym­
bo ls, and la nguages . Th is existence among fragments has often been por­
trayed as a process of ru i n and c u ltura l decay, perhaps most eloquently
by C l aude Lev i-Strauss in Tristes tropiques ( 1 9 5 5 ) . In Levi -Strauss's global
visi on-one widely shared today-authentic h u man differences are d is­
i ntegrating, d isa ppearing in an expansive commod ity cu lture to become,
at best, col lectible "art" o r "fo l klore." The great narrative of entropy and
l oss in Tristes tropiques expresses an i nesca pable, sad truth . B ut it i s too
neat, and it assumes a q uestionable E u rocentric position at the "end " of
a u n ified h u man h i story, gathering up, memoria l iz i ng the world's loca l
T H E P U R E P R O D U C T S G O C R AZY 15

h istoricities . Alongside th is narrative of progressive monoc u lture a more

ambiguous "Cari bbean" experience may be g l i m psed . I n my account
Aime Cesa i re, a p ractitioner of "neo logistic" c u ltura l pol itics, represents
such a poss i b i l ity-orga n i c c u lture reconceived as inventive process or
c reo l i zed " i nterc u lture" (Wagner 1 980; Drum mond 1 98 1 ) . 4 The roots of
trad ition a re cut and retied, col lective sym bols appropriated from exter­
nal i nfluences. For Cesai re cu lture and identity are i nventive and mobi le.
They need not take root i n ancestra l p l ots; they l ive by pol l i nation, by
( h i stori ca l ) transplanti ng.
The "fi lth" that a n expansive West, accord i ng to the d i s i l l usioned
traveler of Tristes tropiques (p. 38), has thrown i n the face of the world's
soc ieties appears as raw materia l , com post for new orders of d i fference.
It i s a l so filth . Modern c u l tu ra l contacts need not be romanticized, eras­
ing the v i olence of e m p i re and conti n u i ng forms of neocolon i a l dom i­
nation . The Cari bbean h i story from which Cesai re derives an i nventive
and tactica l "negritude" is a h i story of degradation , m i m i c ry, violence,
and b l ocked poss i b i l ities. It i s a l so rebe l l ious, syncretic, and creative .
This k i nd of a m b i gu ity keeps the p l anet's loca l futu res u n certa i n and
open . There is no master narrative that can reconc i l e the tragic and
com ic plots of global cu ltural h i story.
It i s easier to register the loss of trad itional orders of difference than
to percei ve the emergence of new ones . Perhaps this book goes too far
in i ts concern for eth nograp h i c presents-becom ing-futu res . Its utopian,
pers i stent hope for the rei nvention of d i fference risks down p l aying the
destructive, homoge n i z i n g effects of global econo m i c and c u ltural cen­
tra l i zation . Moreover, its Western assu m ption that assertions of "trad i ­
t i o n " a r e a l ways responses t o the n e w (that there i s no rea l recu rrence i n
h i story) may exc l ude loca l narratives o f c u ltura l conti n u ity a n d recovery.
I d o not tel l a l l the possi b l e stories . As an l gbo sayi ng has it, "You do not
stand in one pl ace to watch a masq uerade."
My pri mary goa l is to open space for c u ltura l futu res, for the recog-

4. For recent work on the h i stori ca l -pol itical i nvention of cultu res and tra­
d i tions see, among others, Comaroff 1 985; G u ss 1 986; Handler 1 98 5 ; Hand ler
and L i nnekin 1 984; H obsbawm and Ranger 1 98 3 ; Taussig 1 980, 1 987; Wh is­
nant 1 98 3 ; and Cantwe l l 1 984 . Fam i l iar approaches to "cultu re-contact," "syn­
cretism," and "accu lturation" are pressed farther by the concepts of "interfer­
ence" and " i n terreference" (fi sc her 1 98 6 : 2 1 9, 232; Baumga rten 1 982 : 1 54),
"transc u lturation" (Rama 1 98 2 ; Pratt 1 987), and " i nterc u l tural i n tertexts" (Ted­
lock and Ted lock 1 98 5 ) .
16 I N T R O D UC T I O N

n ition of emergence. Th i s req u i res a critique of deep-seated Western hab­

its of mind and systems of va l u e . I am espec i a l l y skeptical of an a l most
automatic reflex-in the service of a u n ified vision of h istory-to rele­
gate exotic peoples and objects to the col lective past (Fabian 1 983). The
i n c l usive orders of modern ism and anthropology (the "we" rid ing i n Wi l­
l iams' car, the Manki nd of Western soci a l science) are always deployed
at the end po i nt or advancing edge of H i story. Exotic trad itions appear as
archaic, purer (and more ra re) than the d i l uted i n ventions of a syncretic
present. In th i s tem pora l setup a great many twentieth-century creations
can only appear as i m itations of more "deve loped" models. The E l s ies of
the planet a re sti l l trave l i ng nowhere thei r own .
Throughout the world i nd igenous popu l ations have had to reckon
with the forces of " p rogress" and "national" u n i ficati on . The resu l ts have
been both destructive and i nventive . Many trad itions, l anguages, cos­
mologies, and val ues a re lost, some l itera l ly m u rdered; but much has
s i m u ltaneously been i nvented and rev ived in com plex, oppositiona l con­
texts. If the victi ms of progress and empire a re weak, they are seldom
pass ive. It used to be assumed, for example, that conversion to Ch ristian­
ity in Africa, Me l a nesia, Lati n America, or even colon i a l Massach usetts
wou ld l ead to the exti nction of i nd igenous cu ltures rather than to thei r
transformatio n . Someth i n g more ambiguous and h i storica l l y com plex
has occu rred , req u i ri n g that we perceive both the end of certai n orders
of d i versity and the c reation or transl ation of others ( Fernandez 1 978) .
More than a few "exti nct" peoples have retu rned to haunt the Western
h i storical i m agi natio n . 5 It is d ifficu l t, in any event, to equate the future of
"Catho l ic i sm " in New G u i nea with its current prospects in Italy; and
Protestant Christian ity in N ew Ca l edon i a is very d i fferent from its d i verse
N igerian forms . The futu re is not (on ly) monoc u l tu re.6

5 . T h e conti n ued tribal l ife o f Cal iforn ia I nd i ans is a case i n point. Even ,
most notorious of a l l , the genocidal "exti nction" of the Tasmanians now seems a
much less definitive "event." After systematic dec i mations, with the 1 876 death
of Trugani na, the l ast "pure" spec imen (playing a myth ic role s i m i lar to that of
lshi i n Cal iforn ia), the race was scientifica l l y dec lared dead. But Tasmanians did
survive and interma rried with aboriginals, wh ites, and Maori . In 1 978 a commit­
tee of i n q u i ry reported between fou r and five thousand persons el igible to make
land claims i n Tasmania (Stocking 1 987 :283).
6. Research speci fica l l y on th i s issue is being conducted by U lf Han nerz
and his col leagues at the U n i versity of Stockho l m on "the world system of cul­
ture." I n an early statement Han nerz confronts the wi despread assumption that
"cu ltu ra l diversity is wan i ng, and the same s i ngle mass culture w i l l soon be
T H E P U R E P R O D U C T S G O C R AZY 17

To reject a s i ng l e progressive or entropi c metana rrative i s not t o deny

the existence of pervasive global processes uneven l y at work. The world
is i ncreas i ngly connected , though not u n i fied , econom i ca l ly and cu ltur­
a l ly. Loc a l part i c u larism offers no escape from these i nvol vements . I n ­
deed, modern eth nogra p h i c h i stories are perhaps condem ned t o osc i l l ate
between two metana rratives : one of homoge n i zation, the other of emer­
gence; one of loss, the other of i n vention . In most spec ific conj u n ctu res
both narrati ves are relevant, each u nderm i n i ng the other's c l a i m to tel l
"the whole story," each deny i n g to the other a privi leged , Hege l ian vi­
s i o n . Everywhere i n the world d i sti nctions are bei ng destroyed and c re­
ated ; but the new identities and o rders of d ifference are more rem i n i scent
of Wi l l iams' E l s i e tha n of Edwa rd Curtis' ideal ized "va n ish i ng" American
I n d i a n s . The h i stories of emergent d i fferences req u i re other ways of tel l­
i n g : Cesai re's i m p u re cu ltu ra l poeti cs, Sa id's d i spersed " Pa l esti n i a n
Li ves," Mash pee's rei nvented trad ition-there i s no s i ngle model . Th i s
book s urveys several hybrid and s ubve rs i ve forms o f c u ltural representa­
tion , forms that prefig u re an i nventive futu re . In the l ast decades of the
twentieth centu ry, ethnography begi ns from the i nescapable fact that
Westerners a re not the o n l y ones goi ng p l aces in the modern world .
But have not travel ers a l ways encountered world ly "natives" ?
Strange antici pation : the Engl i s h P i l grims arrive at Plymouth Rock in The
New World o n l y to find Squanto, a Patuxet, j ust back from E u rope.

everywhere ." He i s skeptica l : "I do not th i n k it is only my bias as a n anthropol­

ogist with a vested interest i n c u ltural variation which makes i t d iffic u l t for me to
recogn i ze that the s i tuation for example i n N i geria cou ld be anything l i ke this.
The people i n my favorite N i gerian town drink Coca Cola, but they drink buru­
kutu too ; and they can watch Charlie's Angels as wel l as Hausa drum mers on the
television sets which spread rapidly as soon as electricity has arrived . My sense
is that the world system, rather than c reati ng massive cultura l homogeneity on a
globa l sca le, is replaci ng one d i versity with another; and the new diversity i s
based relatively more on i n terrel ations and less on autonomy" (Hannerz n . d . : 6 ) .
"White Man, " Onyeocha, a performer at Igbo
masquerades. Amagu Izzi, southeast Nigeria, 1982.
Part One � Discourses
Clifford takes as his natives, as well as his informants, . . .
anthropologists . . . We are being observed and inscribed.

1. On Ethnographic Authority

THE 1 724 frontispiece of Father Lafitau 's Moeurs des sauvages ameri­
qua ins portrays the ethnographer as a you n g woman sitti ng at a writi ng
ta ble amid a rtifacts from the New World and from c l assical G reece and
Egypt. The author is accompan ied by two cherubs who assist i n the task
of com parison and by the bearded figure of Ti me, who poi nts toward a
ta bleau representi n g the u lti mate sou rce of the truths issu i n g from the
writer's pen . The i mage toward which the you n g woman l ifts her gaze is
a ban k of c louds where Ada m , Eve, and the serpent appea r. Above them
sta nd the redeemed man and woman of the Apocal ypse, on either side
of a rad iant tri angle bearing the Hebrew script for Yahweh .
The fronti s p i ece for Ma l i nowski's Argonauts of the Western Pacific is
a photograph with the ca ption "A Ceremon i a l Act of the K u l a ." A she l l
neckl ace i s bei n g offered to a Trobriand c h i ef, who sta nds a t the door of
h i s dwel l i ng. Beh i n d the man presenting the nec klace is a row of six
bowing youths, one of them sou nd ing a con c h . A l l the figu res stand in
profi le, thei r attention apparently concentrated on the rite of excha nge,
a rea l event of Melanesian l i fe . B ut on c loser i n s pection one of the bow­
i n g Trobrianders may be seen to be look i n g at the camera .

22 D I S CO U R S E S

Lafitau's a l l egory i s the less fam i l iar: h i s author transcri bes rather
than origi nates . U n l i ke Mal i nowsk i 's photo, the engrav i ng makes no ref­
erence to eth nographi c experience-despite Lafitau's five years of re­
searc h among the Mohawks, research that has earned h i m a respected
place among the fieldworkers of any generation . H i s account is pre­
sented not as the p rod uct of fi rsthand observation but of writi ng, in a
crowded workshop. The frontispiece from Argonauts, l i ke a l l photo­
graphs, asserts presence-that of the scene before the lens; it a l so sug­
gests another presence-that of the eth nographer active l y com posing
th is fragment of Trobriand rea l ity. K u l a exchange, the su bject of Ma l i­
nowski's book, has been made perfectly visible, centered i n the percep­
tua l frame, wh i le a partic i pant's glance red i rects our attention to the ob­
servational standpoi nt we share, as readers, with the eth nographer and
h is camera . The predomi nant mode of modern fieldwork authority is sig­
naled: "You a re there . . . because I was there."
Th i s chapter traces the formation and breakup of eth nograph i c au­
thority i n twentieth-centu ry soc i a l anthropology. It is not a com plete ac­
cou nt, nor is it based on a fu l l y rea l i zed theory of eth nograph i c i nterpre­
tation and textu a l ity. 1 Such a theory's contou rs are problematic, s i nce the
activity of cross-c u l tural representation is now more than usua l l y in q ues­
tion . The present pred icament is l i n ked to the breakup and red i stribution
of colon i a l power in the decades after 1 950 and to the echoes of that
process in the rad ical cu ltura l theories of the 1 9 60s and 1 9 70s. After the
negritude movement's reversal of the E u ropean gaze, after anth ropo l ogy's
crise de conscience with respect to its l i beral status with i n the i m peria l
order, a n d now that the West can no longer present itse lf a s the u n iq ue
pu rveyor of anthropological knowledge about others, it has become nec­
essa ry to i m agine a world of genera l ized ethnography. With expanded
com m u n i cation and i n tercu ltural i nfl uence, people i nterpret others, and
themselves, in a bewi ldering d i versity of idioms-a global condition of

1. Only E ngl i s h , American, and French examples are discussed. If it i s l i kely

that the modes of authority analyzed here are able to be genera l ized widely, no
attempt has been made to extend them to other national traditions. It is assu med
a l so, in the anti pos i tivist tradition of Wi lhelm D i l they, that eth nography is a pro­
cess of interpretat ion, not of explanation . Modes of authority based on natu ral­
scientific epistemologies are not discussed . I n its focus on participant observation
as an i ntersubjective process at the heart of twentieth-centu ry eth nography, this
discussion scants a n u m ber of contributing sou rces of authority: for exam ple the
wei ght of acc u m u l ated "archival" knowledge about particular groups, of a cross­
cu ltural com parative perspective, and of statistical survey work.

what M i kh a i l B a khti n ( 1 9 5 3 ) cal l ed "heteroglossi a ." 2 T h i s ambiguous,

m u ltivocal world m a kes it i n c reas i ngly h a rd to conce ive of h u m a n d i ver­
sity as i nscri bed i n bou nded, i ndependent cultures. D i fference is an ef­
fect of i nventive syncretism . I n recent years works such as Edward Said's
Orienta/ism ( 1 9 78) and Pau l i n Hou ntondj i 's Sur Ia philos ophie afri­
" "

caine ( 1 9 7 7) h ave cast rad ical doubt on the proced u res by w h i c h a l ien
h u man groups can be represented without proposi ng systematic, sharply
new methods or epistemologies . These stud i es suggest that wh i le eth no­
gra p h i c writing can not enti re l y escape the red uctio n i st use of d i choto­
m i es and essences, it can at least struggle self-consc iously to avoid por­
traying a bstract, a h i storical "others ." It is more than ever c ruc i a l for
d i fferent peoples to form com p l ex concrete i mages of one a nother, as
wel l as of the rel ations h i ps of knowledge and power that con nect them;
but n o sovereign scientific method o r eth ical stance can guarantee the
truth of such i mages. They a re constituted-the critique of colo n i a l
modes o f representation h a s shown a t l east th i s much- i n spec i fic h i s­
torical re lations of d o m i nance and d i a logue.
The experiments in eth nogra p h i c writ i n g su rveyed in th i s chapter do
not fal l i nto a c lear reform i st d i rection or evo l ution . They a re ad hoc
i nventions and can not be see n in terms of a systematic an alys i s of post­
co lon i a l representation . They a re perhaps best u nderstood as com po­
nents of that "too l kit" of engaged theory recentl y recommended by G i l les
Deleuze and M i c he l Foucau lt: "The notion of theory as a tool kit means
(i) The theory to be constructed i s not a system but an i n stru ment, a log ic
of the spec ific ity of power rel ations and the struggles a round the m ; ( i i )
That th i s i nvestigation c a n o n l y b e carried o u t step b y step on the bas i s
o f reflection (wh ich w i l l necessa ri l y b e h i storical i n some o f i t s aspects)
on give n s i tuations" ( Foucau lt 1 980 : 1 45 ; see a l so 1 9 77 : 208) . We may
contri bute to a practical reflection on c ross-c u ltural representation by u n­
derta k i n g a n i nventory of the better, though i m perfect, approaches c u r­
rently at hand . Of these, ethnographi c fieldwork rem a i ns a n u n usua l l y

2 . " Heteroglossia" assu mes that " l a nguages d o not exclude each other, but
rather i nte rsect with each other in many d i fferent ways (the U krai n i a n language,
the language of the epic poem, of early Symbolism, of the student, of a particular
generation of ch i l d ren, of the run-of-the-m i l l i nte l l ectu a l , of the N ietzschean,
and so on). It m i ght even seem that the very word ' l an guage' loses all mea n i ng
in th i s process-for a pparently there i s no single plane on wh ich a l l these 'lan­
guages' might be j u xtaposed to one another" (29 1 ) . What is said of languages
appl ies equ a l l y to "cultures" and "subcu ltures." See a l so Vo losi nov (Bakhti n ?)
1 95 3 : 29 1 , esp. chaps. 1 -3 ; and Todorov 1 98 1 :88-93 .
24 D I S C O UR S E S

sensitive method . Part i ci pant observation obl iges its practiti oners to ex­
perience, at a bod i l y as wel l as an i nte l l ectu a l l eve l , the vic issitudes of
translation . It req u i res ard uous language learn i ng, some degree of d i rect
i nvolvement and conversation, and often a derangement of personal and
cu ltura l expectat ions. There is, of cou rse, a myth of fie ldwork. The actua l
experience, hedged around with conti ngenc ies, rare l y l ives u p to the
idea l ; but as a means for prod u c i n g knowledge from an i ntense, i ntersu b­
j ective engagement, the practice of ethnography reta i n s a certa i n exem­
plary status. Moreover, if fieldwork has for a time been identified with a
u n iquely Western d i sc i pl i ne and a tota l i z i n g science of "anthropology,"
these assoc iations are not necessari l y permanent. Cu rrent styles of c u l ­
tu ra l description are h i storica l l y l i m i ted a n d are u ndergo i ng i m porta nt
metamorphoses .
The deve lopment of eth nographic science can not u lti mate l y be
understood i n isolation from more genera l pol itica l -epistemologica l de­
bates about writi ng and the representation of otherness. In th i s d i scus­
sion, however, I have mai nta i ned a focus on professiona l anthropol ogy,
and spec ifica l ly on ethnography s i nce 1 95 0 . 3 The cu rrent crisi s-or bet­
ter, d ispers ion-of eth nographic authority makes it possible to mark off
a rough period, bou nded by the yea rs 1 900 and 1 960, d u r i ng which a
new conception of field research esta b l i shed itse lf as the norm for Eu ro­
pea n and .American anthropology. I ntens i ve fiel dwork, pu rsued by
u n i versity-tra i ned spec i a l i sts, emerged as a priv i l eged , sanctioned sou rce
of data about exotic peoples. It is not a q uestion here of the dom i nance
of a single research method . " I ntensive" eth nography has been variously
defi ned . (Co m pa re Griaule 1 95 7 with Mal i nowski 1 92 2 :chap. 1 ) . More­
over, the hegemony of fieldwork was establ i shed earl ier and more thor­
ough l y i n the U n ited States and i n England than i n France. The earl y
examples of Franz Boas and the Torres Stra its exped ition were matched
o n l y be l ated l y by the fou nd i ng of the l n stitut d ' Ethnologie i n 1 92 5 and

3. I have not attempted to survey new styles of eth nograph ic writing that
may be origi nati ng outside the West. As Edward Said, Pau l i n Hountondj i , and
others have shown, a considerable work of ideological "cleari ng," oppositional
critical work, remains; and it is to this that non-Western i ntel lectua ls have been
devoting a great part of thei r energies. My discussion rema i ns i nside, but at the
experimental bounda ries of, a rea l ist cultural sc ience elaborated i n the Occ ident.
Moreover, it does not consider as areas of i n novation the "para-eth nographic"
gen res of ora l history, the nonfiction novel, the "new journ a l ism," travel l i tera­
ture, and the documentary fi l m .
O N E T H N O G R A P H I C A UT H O R I T Y 25

the m uch-pu b l icized Mission Dakar- Dj ibouti of 1 932 (Karady 1 982 ;

J a m i n 1 982a; Stock i n g 1 98 3 ) . N everthe l ess, by the m id- 1 930s one can
fa i r l y speak of a deve l o p i n g i ntern ational consensus : va l id anthropolog­
ica l abstractions were to be based , wherever possible, on i ntensive c u l ­
tura l desc ri ptions by q u a l ified sc holars. By this point t h e n e w styl e h ad
been made popu l a r, i nst itutional ized , and em bod ied i n specific textual
It has recently become possi b l e to identify and take a certa i n d i s­
tance from these convention s . 4 If eth nography prod uces c u ltural i nter­
pretations through intense research experiences, how is u n r u l y experi­
ence transformed i nto a n authoritative written account? How, precisely,
is a garru lous, overdeterm i ned c ross-c u ltura l encou nter shot through
with power re lations and perso n a l c ross-purposes c i rcumscribed as an
adeq uate version of a more or less d i sc rete "other world" com posed by
an i nd iv i d u a l author?
I n a n a l y z i n g this com p l ex transformati on one m u st bear i n m i nd the
fact that ethnography is, from begi n n i ng to end, en meshed i n writi ng.
This writi ng i n c l udes, m i n i ma l l y, a tra n slation of experience i nto textual
form . The process is com p l i cated by the action of m u lti ple subjectivities
and pol itical constra i nts beyond the contro l of the writer. I n response to
these forces eth nogra p h i c writing enacts a speci fic strategy of authority.
Th i s strategy has classical l y i n volved a n u n q uestioned c l a i m to appear as
the pu rveyor of truth in the text. A com p l ex cu ltural experience is enun­
c i ated by an i nd i vidua l : We the Tikopia by Raymond F i rth ; No u s avons
mange Ia foret by Georges Condom i nas; Coming of Age in Samoa by
Margaret Mead ; The Nuer bv E. E. Evans-Pritchard .
The d iscussion that fol lows fi rst locates th i s authority h i storical l y i n
the development of a twentieth-century sc ience of parti c i pant observa­
tion . It then p roceeds to a critique of u nderl y i n g ass u m ptions and a re­
view of emergi ng textual practices. A l ternate strategies of eth nographic
authority may be seen i n recent experiments by eth nographers who self­
consc i o u s l y reject scenes of cu ltural representatipn i n the style of Mali­
nowski's frontispiece. D i fferent sec u l a r versions of Lafiteau 's crowded
scriptoria l workshop a re emergi ng. I n the new parad igms of authority the

4 . In the present crisis of authority, eth nography has emerged as a subject

of h i storical scruti ny. For new critical approaches see Hartog 1 9 7 1 ; Asad 1 9 7 3 ;
Bu rridge 1 9 73 :chap. 1 ; Duchet 1 9 7 1 ; Boon 1 98 2 ; D e Certeau 1 980; S a i d 1 9 78;
Stock i ng 1 98 3 ; and Ri.Jpp-Eisenreich 1 984 .

writer is no longer fasc i nated by transcendent figures-a Hebrew­

Christian deity or its twentieth-centu ry repl acements, Man and Culture.
N oth i ng rema i n s of the heaven l y tableau except the anth ropo logist's
scu m bled i mage in a m i rror. The s i l ence of the ethnographic workshop
has been broken-by i nsistent, heteroglot voices, by the scratc h i ng of
other pens. 5

At the c l ose of the n i neteenth centu ry noth i ng guaranteed, a prior i , the

eth nogra pher's status as the best i nterpreter of native l ife-as opposed to
the traveler, and espec i a l l y the m i ssionary and ad m i n i strator, some of
whom had been i n the field far longer and had better research contacts
and l i ngu istic ski l l s. The development of the fie ldworker's i mage i n
America, from Fra n k H a m i l ton Cushing (an oddba l l ) to Ma rgaret Mead
(a nationa l figu re) is sign i ficant. During this period a particular form of
authority was created-an authority both scientifical ly val idated and
based on a u n ique personal experience. During the 1 920s Mal i nowski
pl ayed a centra l role in esta b l i s h i n g cred it for the fie ldworker, and we
shou ld reca l l i n th i s l ight h i s attacks on the com petence of com petitors
in the fie l d . For example the co lonial magistrate Alex Rentou l , who had
the temerity to contrad i ct sc ience's fi ndi ngs concerning Trobriand con­
cepti ons of patern i ty, was excom m u n i cated i n the pages of Man for h i s
unprofess ional " pol ice cou rt perspective" (see Rento u l 1 93 1 a , b ; Ma l i­
nowski 1 93 2 ) . The attack on amateu rism i n the field was pressed even
fu rt her by A. R. Radcl iffe-B rown , who, as lan Langham has shown , came
to epitomize the sc ientific profess iona l , d iscovering rigorous soc i a l laws
( Langham 1 98 1 :chap. 7). What emerged d u r i ng the fi rst half of the twen­
tieth century with the success of profess ional fiel dwork was a new fusion
of general theory and empi rical researc h , of cu ltura l analysis with eth­
nographic descri pti on .
The fie ldworker-theorist repl aced an older partition between the
"man on the spot" ( i n J a mes Frazer's words) and the soc iologist or anth ro­
po logist i n the metropol e . This d ivision of l abor varied in d ifferent na­
tional trad itions. In the U n ited States for example Morgan had personal
knowledge of at least some of the c u l tures that were raw material for h is

5 . On the su ppression of d i a logue i n Lafitau's frontispiece and the consti­

tution of a textua l i zed , a h i storica l , and visua l ly oriented "anthropology" see
Michel de Certeau 's deta i led analysis ( 1 980) .
O N E T H N O G R A PH I C A U T H O R I T Y 27

sociological syntheses ; and Boas rather e a r l y on made i ntensive fie ld­

work the sine qua non of serious anth ropological d i scou rse. In general ,
however, before Ma l i nows k i , Radc l iffe-Brown, and Mead had success­
fu l ly establ i shed the norm of the u n i versity-tra i ned scholar testing and
deriving theory from fi rsthand researc h , a rather d ifferent economy of
eth nographic knowledge preva i led . For exam p l e The Melanesians ( 1 89 1 )
by R. H . Cod r i n gton is a deta i led comp i l ation of fol kl ore and custom ,
d rawn from h i s re l ative l y long term o f research a s a n evangel ist and
based o n i ntensive co l l a boration with i n d i genous transl ators and i n for­
mants . The book is not organ i zed around a fie l dwork "experience," nor
does it advance a u n ified i nterpretive hypothes is, fu nction a l , h i storica l ,
o r otherwise. It i s content with low- level genera l i zations a n d the amass­
i n g of an ec lectic ran ge of i nformation . Codrington is acute l y aware of
the i ncom p l eten ess of h i s know l edge, bel ievi ng that rea l understa nd i ng
of native l ife beg i n s o n l y after a decade or so of experience and study
(pp. vi-v i i ) . Th i s u nderstand i n g of the d ifficu lty of grasp i n g the world of
a l ien peoples-the many years of learn i ng and u n learn i ng needed, the
problems of acq u i ri n g thorough l i ngu i stic com petence-tended to dom­
i nate the work of Codringto n 's generation . Such assu mptions wou ld soon
be chal lenged by the more confident c u l t u ra l relativism of the Ma l i now­
ski an mode l . The new fieldworkers sharply d i stingu ished themselves
from the earlier "men on the s pot" -the m i ssionary, the ad m i n i strator,
the trader, and the traveler-whose knowledge of i n d i genous peoples,
they argued , was n ot i n formed by the best sc ientific hypotheses or a suf­
fic ient neutral ity.
Before the emergence of profess ional ethnography, writers such as
j. F. Mclen n a n , J o h n Lu bboc k, and E . B . Tylor had attem pted to control
the q u a l ity of the reports on w h i c h the i r anthropological syntheses were
pased . They d i d th is by means of the g u i de l i nes of Notes and Queries
and, i n Tylor's case, by c u ltivati ng long-term work i n g re l ations with so­
ph isticated resea rc hers in the fie l d such as the m i ssionary Lo rimer F i son .
After 1 88 3 , as n ew l y appoi nted reader i n anth ropology at Oxford , Tylor
worked to encou rage the systematic gathering of eth nograph i c data by
q u a l ified profess ion a l s . The U n ited States B u reau of Eth nol ogy, a l ready
com m i tted to the u ndertaki ng, provided a mode l . Ty lor was active i n
fou nd i ng a com m i ttee on the Northwestern Tri bes of Ca nada . The com­
m i ttee's fi rst agent in the field was the n i neteen-year-veteran m i ssionary
among the Oj i bwa, E . F. Wi l son . He was replaced before long by Boas,
a phys i c i st in the process of turn i ng to profess ional eth nography. George
28 D I S CO U R S E S

Stocking has persuasive l y argued that the repl acement of Wi l son by Boas
"marks the beg i n n i ng of an i m porta nt phase in the deve l opment of B ritish
ethnographic method : the col l ection of data by academ ica l l y tra i ned nat­
ural sc ientists defi n i ng themse l ves as anth ropo logi sts, and i nvo lved a l so
i n the form u lation and eva l uation of anthropological theory" ( 1 983 : 74) .
With Boas' ea rly su rvey work and the emergence i n the 1 890s of other
natu ra l -sc ientist fie l dworkers such as A. C. Haddon and Ba ldwi n Spen­
cer, the move toward professional eth nography was u nder way. The
Torres Straits exped ition of 1 899 may be seen as a c u l m i nation of the
work of th i s " i ntermed iate ge neration," as Stoc k i n g ca l l s them . The new
style of resea rc h was c learly d i fferent from that of m issionaries and other
amateu rs in the field, and part of a genera l trend si nce Tylor "to draw
more c l osel y together the empi rical and theoretical components of an­
th ropological i n q u i ry " ( 1 983 : 72 ) .
T h e establ ish ment o f i ntensive partic i pant observation as a profes­
sional norm, however, wou l d have to await the Ma l i nowskian cohort.
The " i ntermed iate generation" of eth nographers did not typica l l y l ive i n
a s i ngle loca le for a year o r more, mastering the vernac u l a r and u nder­
goi ng a personal learn i n g experience comparable to an i n itiation . They
d id not speak as cu ltura l i ns iders but reta i ned the natural scienti st's doc­
umentary, observationa l stance. The principal exception before the th i rd
decade of the centu ry, Fra n k Hami l ton C u s h i ng, remai ned an isolated
i n sta nce . As Curtis H i n s ley has suggested, Cush i ng's long fi rsthand study
of the Zu n i s , h i s q u asi-absorption i nto their way of l ife, "raised problems
of veri ficati on and accou nta b i l ity . . . A comm u n ity of scientific anthro­
pol ogy on the model of other sciences req u i red a common l anguage of
d i scou rse, channels of regu lar com m u n ication, and at least m i n i ma l con­
sensus on j udging method" ( 1 983 : 66) . Cush i ng's i ntuitive, excessively
personal u nderstand i ng of the Zu n i cou ld not confer scientific authority.
Schematica l ly put, before the l ate n i neteenth centu ry the ethnogra­
pher and the anthropologist, the describer-tra nsl ator of custom and the
b u i l der of genera l theories about humanity, were d i sti nct. (A clear sense
of the tens ion between eth nography and anthropol ogy is i mporta nt i n
correctly perce iving the recent, and perhaps temporary, conflation of the
two projects . ) Ma l i nowski gives us the i mage of the new "anth ropolo­
gi st" -squatt i ng by the campfi re; look i n g, l iste n i ng, and questioni ng; re­
cord i ng and i nterpreting Trobriand l ife. The l i terary charter of th i s new
authority i s the fi rst chapter of Argonauts, with its prom i nently d isplayed
photographs of the eth nographer's tent pitc hed among Ki riwi n i a n dwe l l ­
i ngs . T h e sha rpest methodological j u stification for the new mode is to be

fou nd i n Radel iffe- B rown 's Andaman Islanders ( 1 922) . The two books
were p u b l i shed with i n a year of each other. And a lthough the i r authors
developed q u ite d i fferent fieldwork sty les a nd visions of cu ltural science,
both early texts p rovide expl icit arguments for the spec i a l authority of the
eth nographer-a nthropo l ogist.
Ma l i nows k i , as his notes for the cruc i a l i n troduction to Argonauts
show, was greatly concerned w ith the rhetorical problem of convi n c i n g
h i s readers that t h e facts he w a s putti ng before t h e m were objective l y
acq u i red, n o t s u bjective c reations (Stock i n g 1 983 : 1 0 5 ) . Moreover, he
was fu l l y aware that "in Ethnography, the d i stance i s often enormous be­
tween the brute materi a l of i nformation-as it is presented to the student
in h i s own observations, in native statement, in the kaleidoscope of tribal
l i fe-and the fi n a l a uthoritative presentation of the res u l ts" (Ma l i nowski
1 92 2 : 3 -4 ) . Stoc k i ng has n ice l y anal yzed the various l iterary a rtifices of
Argonauts ( its engag i n g narrati ve constructs, u se of the active vo i ce i n
the "eth nogra p h i c p resent," i l l us i ve d ramatizations of the author's partic­
i pation i n scenes of Trobriand l ife) , tec h n i ques Ma l i nowsk i used so that
" h i s own experience of the natives' experience [ m ight] become the read­
er's experience as wel l " (Stock i n g 1 983 : 1 06; see a l so Payne 1 98 1 , and
Chapter 3). The problems of verification and accou ntabi l ity that had rel ­
egated C u s h i n g t o t h e professional marg i n were very m u c h on Ma l i ­
nowsk i 's m i nd . T h i s a n x i ety i s reflected i n the mass o f data conta i ned i n
Argonauts, i ts s i xty-s i x photogra p h i c p l ates, the now rather c u rious
"Chronological L i st of Kula Events Witnessed by the Writer," the constant
a l ternation between i m personal descr i ption of typical behavior and state­
ments on the o rder of " I wi tnessed . . ." a nd "Our party, sai l i ng from the
North . . . "
Argona uts i s a com plex narrati ve s i m u ltaneously of Trobriand l ife
and ethnograph i c fiel dwork. It is archetypical of the generation of eth­
nogra p h i es that successfu l ly esta b l i shed the sc ientifi c va l id ity of partici­
pant observati o n . The story of research bu i l t i nto Argonauts, i nto Mead's
popu l a r work on Samoa, and i nto We the Tikopia became an i m p l i c i t
narrati ve u nderl y i ng a l l professional reports on exotic worlds. If su bse­
q uent eth nogra p h ies d i d not need to i n c l ude developed fieldwork ac­
cou nts, it was because s u c h accou nts were assumed , once a statement
was made on the order of, for example, Godfrey Lienhardt's s i ngle sen­
te nce at the beg i n n i ng of Divinity and Experience ( 1 9 6 1 :vi i ) : "Th i s book
is based u po n two yea rs' work among the D i n ka, spread over the period
of 1 94 7-1 9 5 0 ."
I n the 1 920s the new fieldworker-theorist brought to comp l etion a
30 D I S CO U R S E S

powerfu l new sc ientific and l iterary genre, the ethnography, a synthetic

cu ltura l descr i ption based on parti ci pant observation (Thornton 1 983).
The new style of representation depended on i nstitutional and method­
ologica l i nnovations c i rcu mventi ng the obstacles to rapid knowiedge of
other cu ltures that had preoccu pied the best representatives of Cod ri ng­
ton's generation . These may be briefly summarized .
F i rst, the persona of the fieldworker was va l idated , both publ icly
and professiona l ly. I n the popu l a r doma i n , visi ble figu res such as Ma l i ­
nowski, Mead , and Marcel G ri a u l e com m u n i cated a v i s ion of eth nogra­
phy as both sc ientifica l ly demand ing and heroic. The professional eth­
nographer was tra i ned i n the latest analytic tec h n iques and modes of
scientific expl anation . Th i s conferred an adva ntage over ·amateurs i n the
fiel d : the profess ional cou ld c l a i m to get to the heart of a c u l tu re more
q u ickly, grasp i n g its essenti a l i nstitutions and structu res. A prescri bed at­
titude of c u ltural relativism d i sti ngu ished the fieldworker from m ission­
aries, ad m i n istrators, and others whose view of natives was, presu mably,
less d i s passionate, who were preocc upied with the problems of govern­
ment or conversion . In add ition to scientific soph i stication and re lativist
sympathy, a va ri ety of normative sta ndards for the new form of research
emerged : the fieldworker was to l ive in the native vi l l age, use the ver­
nac u l a r, stay a suffic ient (but seldom speci fied) length of ti me, investigate
certa i n c lass ic su bjects, and so o n .
Second, it w a s tac itly agreed that t h e new-sty le eth nographer, whose
soj o u rn in the field seldom exceeded two years, and more freq uently was
much shorter, cou ld efficiently " use" native languages without "master­
i ng" them . In a s i gn ificant a rti cle of 1 939 Margaret Mead argued that the
ethnographer fol lowing the Ma l i nowskian prescri ption to avo id i nter­
preters and to conduct resea rc h in the verna c u l a r d i d not, i n fact, need
to atta i n "vi rtuosity" i n native tongues, but cou l d "use" the vernac u l a r to
ask q u esti ons, m a i nta i n rapport, and genera l ly get a l ong in the culture
w h i l e obta i n i ng good researc h resu lts i n particu l a r areas of concentra­
tion . Th i s in effect j u stified her own practice, wh ich featu red re lative l y
short stays a n d a focu s on specific doma i ns s u c h as c h i ldhood or "per­
sona l i ty," foc i that wou ld fu nction as "types" for a cu ltural synthesis. Her
attitude towa rd language "use" was broad l y characteristic of an eth no­
graphi c generation that cou ld, for example, cred it as authoritative a
study ca l l ed The Nuer that was based on on l y eleven months of very
d i fficu lt research . Mead's article provoked a sharp response from Robert
Lowie ( 1 940) , writi ng from the older Boas ian trad ition, more ph i lologica l
i n its orientation . But h i s was a rear-guard action ; the poi nt had been

genera l l y esta b l i s hed that va l i d research cou ld, i n practice, be accom­

p l ished on the basis of one or two yea rs' fam i l iarity with a foreign ver­
n ac u l a r (even thou gh, as Lowie suggested , no one wou ld cred it a trans­
lation of Prou st that was based on an eq u iva lent knowledge of French).
Th i rd , the new eth nography was marked by an i ncreased emphasis
on the power of observation . Cu lture was construed as an ensem ble of
characteristic behaviors, ceremon ies, and gestu res suscepti ble to record­
i ng and expl anation by a tra i ned o n l ooker. Mead pressed this poi nt fu r­
thest ( i ndeed , her own powers of visual a n a l ys i s were extraord i nary) . As
a genera l trend the parti c i pant-observer emerged as a research norm . Of
cou rse s uccessfu l fieldwork mobi l ized the fu l lest poss i b l e ra nge of i nter­
actions, but a d i sti nct pri macy was accorded to the visual : interpretation
was tied to desc ri pti o n . After Ma l i nowski a genera l suspicion of "privi­
l eged i nformants" reflected this systematic preference for the (method i ­
c a l ) observations o f the eth nogra pher over t h e ( i nterested) i nterpretations
of i n d i genous a uthorities.
Fou rt h , certa i n powe rfu l theoretical abstractions pro m ised to h e l p
academ i c eth nographers "get t o t h e heart" o f a c u lture more rapi d l y than
someone u ndertaki ng, for exa m p l e, a thorough i nventory of customs and
bel iefs . Without spend i n g years getting to know natives, their complex
l a nguages and habits, i n i nti mate deta i l , the researc her cou ld go after
se l ected data that wou ld yield a centra l a rmatu re or structure of the c u l ­
tura l w h o l e . R i vers' "genea l ogica l method," fol lowed by Radc l iffe­
B rown's model of "soc i a l structu re," provided th i s sort of shortcut. One
cou ld , it seemed, e l icit kin terms without a deep u nderstanding of local
vernac u l a r, and the range of necessary contextual knowledge was con­
ven iently l i m ited .
F i fth, s i nce c u l tu re, seen as a comp l ex whole, was a l ways too much
to master i n a short research span, the new ethnographer i ntended to
foc u s thematica l l y on pa rticu l a r i n stitutions. The a i m was not to contrib­
ute to a com p l ete i n ventory or description of custom but rather to get at
th e whole th rough one or more of its parts . I h ave noted the priv i l ege
given for a time to soc i a l structu re . An i nd ividual l ife cyc l e , a ritual com­
plex l i ke the K u l a ring or the N aven ceremony, cou l d a l so serve, as cou ld
categories of behavior l i ke econom ics, po l itics, and so o n . I n the pre­
dom i nantly synecdoc h i c rhetorical stance of the new eth nography, parts
were assu med to be m ic rocosms or ana logies of wholes. Th i s setting of
i n stitutio n a l foregrounds aga i nst c u ltural backgrou nds i n the portraya l of
a coherent world lent itsel f to rea l i st l ite rary conventions .
S i xth , t h e wholes th u s represented tended t o b e synchron ic, prod-
32 D I S CO U R S E S

ucts of short-term researc h activity. The i ntensive fieldworker cou l d plau­

s i b l y sketch the conto u rs of a n "eth nographic present" -the cyc le of a
year, a ritu a l series, patterns of typi cal behavior. To i ntroduce long-term
h i storical i nqu iry wou ld have i m poss i b l y compl icated the task of the
new-style fieldwork. Thus, when Ma l i nowski and Radc l iffe-Brown estab­
l is hed thei r critique of the "conjectural h istory" of the d iffusionists, it was
a l l too easy to exc l ude d i ach ronic processes as objects of fiel dwork, with
conseq uences that have by now been sufficiently denou nced .

These i n novations served to va l idate an efficient eth nography based on

scientific part i c i pant observatio n . Their com bined effect can be seen i n
what m a y we l l be t h e tou r d e force o f the new eth nography, Eva ns­
Pritchard's study The Nuer, p u b l i s hed in 1 940. Based on eleven months
of researc h cond ucted-as the book's remarkable i ntroduction tel l s us­
i n a l most i m poss ible cond itions, Evans-Pritchard nonetheless was able
to com pose a c l as s i c . He arrived i n N uerland on the hee l s of a punitive
m i l itary exped ition and at the u rgent req uest of the government of the
Anglo-Egyptian Suda n . He was the object of consta nt and intense suspi­
cion . Only i n the fi n a l few months cou ld he converse at a l l effectively
with i nformants, who, he te l l s us, were ski l led at evad ing h is questions .
I n t h e c i rc u m stances h i s monograph is a kind o f m i racle.
Wh i l e adva n c i ng l i m ited claims and making no secret of the con­
stra i nts on h i s resea rch, Evans-Pritchard manages to present h i s study as
a demonstration of the effectiveness of theory. He focuses on N uer po l it­
ical and soc i a l "structu re," a n a l yzed as an abstract set of re l ations
between territo r i a l segments, l i neages, age sets, and other more fl u id
groups. T h i s ana l ytica l l y derived ensemble is portrayed aga i nst a n "eco­
logica l " backd rop composed of m igratory patterns, re l ation s h i ps with
cattle, notions of time and s pace. Eva ns-Pritchard sharply d i stingui shes
his method from what he ca l ls "haphazard" (Ma l i nowskian) docu men­
tatio n . The Nuer i s not an extensive compend i u m of observations and
vernac u l a r texts in the styl e of Ma l i nowski's Argonauts and Cora/ Gar­
dens. Evan s-Pritchard a rgues rigorously that "facts can o n l y be selected
and a rranged in the l ight of theory." The frank abstraction of a pol itical ­
soc i a l structu re offers t h e necessary framework. If I am accused o f de­
scri bing facts as exem p l i fications of my theory, he then goes on to note,
I have been u nderstood ( 1 969 : 2 6 1 ) .
I n The Nuer Evans-Pritchard makes strong c l a i m s for the power of
O N E T H N O G R A P H I C A UT H O R I T Y 33

sc i entific a bstraction t o focus research and arrange complex data. The

book often presents i tsel f as an argument rather than a description, but
not consi stent l y : its theoretica l a rgument i s su rrou nded by ski l lfu l ly ob­
served and n a rrated evocations and i nterpretations of N uer l ife . These
passages function rhetori ca l l y as more than s i m p l e "exempl ificati on," for
they effective l y i m p l i cate readers i n the com p l ex subjectivity of parti ci­
pant observation . This may be seen in a characteristic paragraph, whic h
progresses t h rough a series o f d i sconti n uous d iscu rsive positions :

It is d ifficu l t to fi n d an English word that adequately descri bes the so­

c i a l pos ition of die/ in a tri be. We have ca l led them aristocrats, but do
not wish to i m ply that N uer regard them as of superior rank, for, as we
have emphatica l ly dec l a red , the idea of a man lording it over others is
repugnant to the m . On the whole-we wi l l q u a l i fy the statement
later-the die/ have prestige rather than ran k and i n fl uence rather than
power. If you a re a dil of the tribe i n which you l i ve you are more than
a s i m p l e tri besm a n . You are one of the owners of the coun try, its v i l ­
l age sites, i t s pastu res, i t s fish i n g pools and wel ls. Other people l ive
there by v i rtue of marriage i n to you r c l a n , adoption i nto you r l i neage,
or of some other soc i a l tie. You are a leader of the tribe and the spear­
name of your c l a n is i n voked when the tri be goes to war. Whenever
there is a dil in the v i l lage, the vi l l age c l u sters arou nd h i m as a herd of
cattle c l usters around i ts bu l l . ( 1 969: 2 1 5 )

T h e fi rst t h ree sentences a r e p resented as an a rgument about translation,

but in passi ng they attribute to " N uer" a stable set of attitudes . ( I wi l l
have more to say l ater about th i s styl e of attri bution . ) N ext, i n the fou r
sentences begi n n i ng " I f you a re a dil . . . , " the second-person construc­
tion br in gs together reader and native in a textual parti c i pation. The fi n a l
sentence, offered as a d i rect desc r i ption o f a typical event (wh ich the
reader now ass i m i l ates from the standpo i n t of a partic i pant-observer),
evokes the scene by means of N uer catt l e meta phors . I n the paragraph's
eight sentences an argu ment about translation passes th rough a fiction of
partic i pation to a metaphorical fusion of external and i nd igenous cu ltura l
descri ptions. T h e s u bjective j o i n i ng o f abstract analysis a n d concrete ex­
perience is accomp l i shed .
Evans-Pritchard wou l d l ater move away from the theoretical pos ition
of The Nuer, reject i n g its advocacy of "soc i a l structu re" as a priv i l eged
framework. I ndeed each of the fieldwork "shortcuts" I enumerated earl ier
was and remai ns contested . Yet by the i r deployment in d i fferent com bi-

nations, the authority of the academic fie ldworker-theori st was estab­

l i shed in the years between 1 920 and 1 950. This pec u l iar amalgam of
i ntense personal experience and scientific analysis (u nderstood in this
period as both " rite of passage" and " l a boratory") emerged as a method :
participant observatio n . Though variously u nderstood, and now d isputed
in many quarters, th i s method remains the c h ief d i sti ngu i s h i ng featu re of
professional anthropology. Its complex subjecti vity is routi nely repro­
d u ced in the writi ng and read i ng of ethnographies.

" Partic i pant observation" serves as shorthand for a conti nuous tacking
between the " i n side" and "outside" of events : on the one hand graspi ng
the sen se of spec ific occu rrences and gestu res empathetica l l y, on the
other steppi ng back to situate these mea n i ngs in wider contexts . Partic­
u l ar events thus acq u i re deeper or more ge neral sign ificance, structu ra l
ru les, and so forth . U nderstood l itera l l y, partic i pant observation is a par­
adoxica l , m i slead i n g fo rm u la, but it may be taken seriou s l y if reformu­
l ated i n hermeneutic terms as a d i a l ectic of experience and interpreta­
tion . T h i s is how the method's most persuasive recent defenders have
restated it, i n the trad ition that l eads from Wi l he l m D i l they, via Max We­
ber, to "symbo l s and mea n i ngs" anthropologists l i ke C l i fford Geertz . Ex­
per ience and i nterpretation have, however, been accorded d i fferent em­
phases when presented as c l a i ms to authority. In recent years there has
been a marked shift of emphasis from the former to the l atter. Th i s section
and the one that fol l ows wi l l ex p lore the rather d ifferent c l a i m s of expe­
rience and i n terpretation as we l l as the i r evolving i nterre l ation .
The growing presti ge of the fieldworker-theorist down played (with­
out e l i m i n ating) a n u mber of processes and med iators that had figu red
more prom i nently in previous methods. We have seen how l a nguage
mastery was defi ned as a l evel of use adeq uate for amassing a d iscrete
body of data in a l i m ited period of time. The tasks of textual transcri ption
and translation, along with the cru c i a l d i a logical role of i nterpreters and
"privi leged i nformants," were relegated to a secondary, sometimes even
despised status. Fieldwork was centered in the experience of the parti ci­
pant-observing scholar. A sharp i mage, or narrative, made its appear­
ance-that of an outsider entering a c u l ture, undergoing a k i nd of i n i­
tiation lead ing to "ra pport" ( m i n i ma l l y acceptance and empathy, but
usua l l y i mp l y i n g someth i n g akin to friendsh i p) . Out of th is experi ence

emerged , i n unspecified ways, a representational text written b y the

partic i pant-observer. As we shal l see, th i s version of textual prod uction
obscu res as much as it revea l s . But it is worth tak i n g seriously its pri nci­
pal ass u m ption : that the experience of the researc her can serve as a u n i ­
fyi n g sou rce o f authority i n t h e f i e l d .
Experiential authority i s based on a "feel " for t h e foreign context, a
k i nd of accu m u l ated savvy and a sense of the style of a people or place.
Such a n appea l i s freq uently exp l icit in the texts of the early professional
partici pant-observers . Margaret Mead's claim to grasp the underly i ng
princ i p l e or ethos of a c u l tu re through a heightened sen sitivity to form ,
tone, gest u re, and beh aviora l styles, and M a l i nowski's stress on h i s l i fe
in the v i l l age and the comprehension derived from the " i mponderabi l ia"
of d a i l y existence, a re pro m i nent cases i n po i nt. Many ethnographies­
Col i n Tu rnbu l l 's Forest People ( 1 962), for example-are sti l l cast in the
experienti a l mode, assert i n g prior to any specific research hypothes i s or
method the "I was there" of the ethnographer as i nsider and partici pant.
Of cou rse it i s d iffic u lt to say very m uch about experience. L i ke
" i ntu ition," it is someth i ng that one does or does not have, and its i n vo­
cation often smacks of mystification . N evertheless, one shou ld resist the
temptation to transl ate a l l mean i ngfu l experience i nto i nterpretatio n . If
the two a re reci proca l l y re lated , they a re not i dentica l . It makes sense to
hold them apart, if o n l y because appea l s to experience often act as val i ­
dations for eth nogra p h i c a uthority.
The most serious argu ment for the role of experience i n the h i storical
and c u ltu ral sc iences i s conta i ned i n the genera l notion of Verstehen .6 In
the i nfl uenti a l view of D i lthey ( 1 9 1 4) u nderstand i ng others arises i n itial l y
from the sheer fact of coex istence i n a shared wor l d ; but th i s experienti al
world, a n i ntersu bjective ground for objective forms of knowledge, i s
prec ise l y what i s m i ssi ng or problematic for a n eth nographer enteri ng an
a l ien c u l tu re . Thus, d u ri n g the early months in the field (and i ndeed
throughout the resea rc h), what i s go i ng on is language learn i ng in the
broadest sense . D i l they's "common sphere" m ust be esta b l ished and re­
esta b l i s hed, bu i l d i ng up a shared experiential world in rel ation to w h i c h
a l l "facts," "texts," "events," and the i r i nterpretations wi l l b e constructed .

6. The concept i s someti mes too read i l y associ ated with i ntuition or em pa­
thy, but as a description of eth nograph i c knowledge Verstehen properly involves
a critique of e m pathetic experience. The exact mea n i ng of the term is a matter of
debate among D i l they scho lars (Makreel 1 9 7 5 : 6-7) .

Th i s process of l iv i n g one's way i nto a n a l i e n expressi ve u n i verse i s al­

ways subjective i n natu re, but it q u i c k l y becomes dependent on what
D i l they cal ls "permanently fixed expressions," stable forms to which
u ndersta nd i ng can retu rn . The exegesis of these fixed forms provides the
content of a l l systematic h i storica l-cu ltural knowledge. Th us experience
is c lose l y l i n ked to i nterpretation . ( D i lthey is among the fi rst modern
theorists to com pare the u nderstand i ng of cultural forms to the read i ng
of "texts .") But this sort of read i ng or exegesis cannot occur without a n
i ntense personal part i c i pation, a n active at-homeness i n a common u n i ­
verse .
Fo l low i ng D i lthey, eth nogra p h i c "experi ence" can be seen as the
bu i ld i ng up of a com mon, mea n i ngfu l world, d rawing on i ntu itive styles
of feel i ng, perception, and guesswork. Th i s activity makes use of cl ues,
traces, gestu res, and scraps of sense prior to the development of stable
interpretations. Such piecemea l forms of experi ence may be c l assified as
aesthetic a nd/or d iv i n atory. There is space here for only a few words
about such styles of comprehens ion as they rel ate to ethnography. An
evocation of an aesthetic mode is conven iently provided by A. L. Kroe­
ber's 1 93 1 revi ew of Mead's Crowing Up in New Guinea .

F i rst o f a l l , it i s c lear that s h e possesses t o an outstand i ng degree the

facu lties of sw iftly apperceivi ng the pri ncipal cu rrents of a culture as
they i m p i nge on i ndividuals, and of del i neating these with com pact
pen-pictures of aston ishing sharpness. The result is a representation of
q u ite extraord i nary vivid ness and semblance to l i fe. Obviously, a gift
of i ntel lectu a l i zed but strong sensational ism underl ies this capacity;
a l so, obviously, a h i gh order of i ntuitiveness, i n the sense of the a b i l ity
to com plete a convincing picture from cl ues, for cl ues is a l l that some
of her data ca n be, with only six months to learn a language and enter
the i nwards of a whole culture, besides special izing on c h i ld behavior.
At any rate, the pictu re, so far as it goes, is whol ly convincing to the
reviewer, who unreserved ly adm i res the sureness of insight and effi­
ciency of stroke of the depiction . (p. 248)

A d i fferent form u l ation is prov ided by Maurice Leenhardt in Do Kama:

Person and Myth in the Melanesian World ( 1 9 3 7), a book that in its some­
ti mes crypti c mode of expos ition req u i res of its readers j ust the sort of
aesthetic, gesta lti st perception at which both Mead and Leenhardt ex­
cel led . Leenhardt's endorsement of th i s approach is significant s i nce,
given h i s extreme l y long fiel d experience and profound cultivation of a
O N E T H N O G R A P H I C A UT H O R I T Y 37

Melanesian l a nguage, h i s "method" cannot b e seen as a rationa l i zation

for short-term eth nogra p h y : " I n rea l ity, our contact with another is not
accomp l ished through analysis. Rather, we apprehend h i m in his en­
ti rety. F rom the outset, we can sketc h our view of h i m using an out l i ne
or symbo l ic deta i l w h i c h conta i n s a whol e i n itself and evokes the true
form of h i s be i ng. T h i s l atter i s what escapes u s if we approach our fel l ow
creatu re u s i n g o n l y the categories of o u r i ntel l ect" (p. 2 ) .
Another way o f tak i n g experience seriously as a sou rce o f eth no­
graphic knowl edge i s provided by Carlo G i nzburg's i nvestigations ( 1 980)
i nto the com p l ex trad ition of d i vi n ation. H is resea rc h ranges from early
h u nters' i nterpretations of a n i ma l tracks, to Mesopotamian forms of pre­
d iction, to the dec i pheri ng of sym ptoms i n H i ppocratic med i c i ne, to the
focu s on deta i l s i n detecti n g art forgeries, to Freud, Sherlock Hol mes,
and Proust. These sty l es of nonecstatic d i v i nation apprehend spec ific c i r­
c u mstantial rel ations of mea n i ng and are based on guesses, on the read­
i ng of apparently d i spa rate c l ues and "chance" occ u rrences. G i nzburg
proposes h i s model of "conjectural know l edge" as a d isc i p l i ned , non­
genera l iz i ng, abductive mode of comprehension that i s of centra l ,
though u n recogn i zed, i mporta nce for t h e c u l tu ral sc iences. I t m a y be
added to a rather meager stock of resou rces for u nderstand i ng rigorously
how one feel s one's way i nto an u nfam i l iar eth nographic situatio n .
Prec ise l y because i t is hard to p i n down, "experience" has served as
an effective guara ntee of eth nographic a uthority. There is, of course, a
tel l i n g a m bigu ity i n the term . Experience evokes a partici patory pres­
ence, a sensitive contact with the world to be u nderstood , a rapport with
its people, a concreteness of perception . It a l so suggests a c u m u l ative,
deepen i n g knowledge ("her ten years' experience of New G u i nea" ) . The
senses work together to authorize an eth nographer's rea l but i neffable
feel or fla i r for " h is" or " her" people. It i s worth noti ng, however, that t h i s
"world," when conceived as an experiential c reation, i s subjective, not
d i a logical o r i nters u bjective. The eth nographer acc u m u l ates person a l
knowl edge o f the fie l d (the possessive form my people h a s u nti l recently
been fa m i l i a r l y u sed i n anth ropological c i rc l es, but the ph rase i n effect
s i g n i fies " m y experience" ) .

I t i s u n derstandab l e, given the i r vagueness, that experiential criteria of

authority- u nexam i ned bel iefs i n the "method " of part i c i pant observa­
tion, in the power of rapport, empathy, and so on-have come u nder
38 D I S C O UR S E S

criticism by hermeneutica l l y soph i sticated anthropologists. The second

moment in the d i a lecti c of experience and i nterpretation has recei ved
i n c reas i n g attention and el aboration (see, for example, Geertz 1 9 7 3 ,
1 976; Rabi now and Su l l ivan 1 979; Winner 1 9 76; Sperber 1 98 1 ) . Inter­
pretation, based on a phi lological model of textual "read i ng,'' has
emerged as a soph i sticated alternative to the now apparently naive
c l a i m s for experiential authority. I nterpretive anth ropology demystifies
much of what had previously passed u nexa m i ned in the construction of
ethnogra phic n a rratives, types, observations, and descriptions. It contrib­
utes to an i ncreas i ng v i s i b i l ity of the creative (and in a broad sense po­
etic) processes by w h i c h "cu ltura l " objects are i nvented and treated as
mean i ngfu l .
What i s i nvolved i n looki ng at c u l tu re a s a n assem b lage of texts to
be i nterpreted ? A cl assic account has been provided by Pa u l Ricoeu r, i n
h i s essay "The Model o f Text: Mea n i ngfu l Action Considered a s a Text"
( 1 9 7 1 ) . C l ifford Geertz in a n u mber of sti m u l ating and subtle d i scussions
has adapted Ricoeu r's theory to anthropological fieldwork ( 1 973 : chap.
1 ) . "Textua l ization" i s u nderstood as a prereq u i site to i nterpretation, the
constitution of D i lthey's "fixed expression s ." It is the process th rough
which unwritten behavi or, speec h , be l iefs, ora l trad ition, and ritual
come to be marked as a corpus, a potentia l l y mea n i ngfu l ensemble sepa­
rated out from an i mmed i ate d iscu rsive or performative situati o n . In the
moment of textu a l ization th is mea n i ngfu l corpus assu mes a more or less
stable re lation to a context; and we are fam i l i ar with the end res u l t of th i s
process i n m u c h o f what counts as eth nograph i c t h i c k desc ription . For
example, we say that a certa i n i nstitution or segment of behavior is typ­
ical of, o r a com m u n i cative element with i n , a surrou ndi ng cu lture, as
when Geertz's famous cockfight ( 1 973 : chap. 1 5 ) becomes an i ntense l y
sign i ficant locus o f Ba l i nese cu ltu re . F i e l d s o f synecdoches are c reated
in whi ch parts are rel ated to wholes, and by which the whole-what we
often cal l c u lture- i s constituted .
Ricoe u r does not actu a l l y privi l ege part-whole rel ations and the spe­
c i fic sorts of a nal ogies that constitute fu nctional i st or rea l ist representa­
tions. He mere l y pos its a necessary rel ation between text and "world." A
world can not be apprehended d i rectl y; it i s always i nferred on the basi s
o f i t s parts, and t h e parts m u st b e conceptual l y and perceptua l l y c u t out
of the flux of experience. Thus, textua l ization generates sense th rough a
c i rc u l a r movement that i solates and then contextua l i zes a fact or event
in its englob i n g rea l ity. A fam i l iar mode of authority is generated that
O N E T H N O G R A P H I. C A U T H O R I T Y 39

claims to represen t d i screte, mean i n gfu l worl ds. Eth nography is the i n ter-
pretation of c u l tu res.
A second key step i n R icoeu r's analysis is h i s account of the process
by w h i c h "d iscou rse" becomes text. D i scourse, in E m i l e Benve n i ste's
c l assic d i scussion ( 1 9 7 1 : 2 1 7-230), is a mode of comm u n ication i n
which the presence o f the spea k i n g subject and o f the i m med i ate s itua­
tion of com m u n ication are i ntri nsic. D i scou rse i s m arked by pronouns
(pronou nced or i m p l ied) I and you, and by deictic i nd i cators-this, that,
now, and so o n -that s i g n a l the present i n stance of d i scou rse rather than
someth i ng beyond it. D i scou rse does not transcend the specific occas ion
i n which a s u bject appropriates the resou rces of l a n guage in order to
com m u n i cate d i a logica l l y. Ricoe u r argues that d i scourse can not be i nter­
preted in the open-ended , potenti a l l y pu b l ic way in which a text i s
" read ." To u nderstand d i scou rse "you h a d t o have been there," i n the
presence of the d i scou rs i n g su bj ect. For d i scou rse to become text it m ust
become "autonomous," in Ricoeu r's terms, separated from a spec ific ut­
terance and author i a l i n tenti o n . I nterpretation i s not i nterlocuti o n . I t does
not depend o n be i n g in the presence of a speaker.
The rel evance of t h i s d i sti nction for eth nography is perhaps too ob­
v i o u s . The eth nographer a l ways u lti mate l y departs, tak i n g away texts for
l ater i n terpretation (and among those "texts" taken away we can i nc l ude
memories-events patterned , s i m p l i fied, stri pped of i m med i ate context
in order to be i n terpreted in l ater reconstruction and portraya l ) . The text,
u n l i ke 9 i�ourse, can trave l . If much eth n ographic writi ng is prod uced i n
the fiel d , actu a l composition o f an eth nography i s done el sewhere . Data
consti tuted i n d i sc u rs i ve, d i a l ogica cond itions are appropri ated o n l y i n
textua l i zed forms. Researc h events a n d encounters become fie l d notes .
Experiences become narratives, mea n i n gfu l occ urrences, o r exam ples.
Th i s tra n s l ation of the research experience i nto a textual corpus
sepa rate from its d i sc u rs i ve occas ions of prod uction has i m portant con­
seq uences for eth n ogra p h i c authority. The data thus reform u l ated need
no longer be u nderstood as the com m u n ication of spec ific persons. An
i nformant's exp l anati o n or desc r i ption of custom need not be cast in a
form that i n c l udes the message "so and so said th i s ." A textual ized ritu a l
o r event i s no lon ger c lose l y l i n ked t o t h e prod uction o f t h a t event by
spec ific actors. I n stead these texts become evidences of an englob i n g
context, a "cu l tu ra l " rea l ity. Moreover, as specific authors a n d actors a re
severed from thei r prod uctions, a genera l ized "author" must be i n vented
to accou nt for the world or context with i n w h i c h the texts are fictiona l l y
40 D I S C O UR S E S

re l ocated . Th i s genera l i zed author goes u nder a variety of names : the

native po i n t of v i ew, "the Trobria nders," "the N uer," "the Dogan," as
these and s i m i l a r ph rases appea r in eth nographies. "The B a l i nese" fu nc­
tion as author of Geertz's textu al ized coc kfight.
The eth nographer thus enjoys a s pec i a l re l ationsh i p with a cu ltural
origi n or "a bso l ute su bject" (Michel-Jones 1 9 7 8 : 1 4) . It i s tempting to
compare the eth nographer with the l iterary i n terpreter (and this compar­
i son is i nc reas i ngly com mon p l ace)-but more spec i fica l l y with the tra­
d i tional critic, who sees the task at hand as l ocating the u n r u l y mea n i ngs
of a text in a s i ngle coherent i ntenti o n . By rep resenting the N uer, the
Trobria nders, or the B a l i nese as whole subjects, sou rces of a mea n i ngfu l
i ntention, the eth nographer tran sforms the research situation's ambigu i ­
ties a n d d i versities of mea n i ng i nto an i ntegrated portra it. It i s i m porta nt,
though, to notice what has d ropped out of sight. The research process i s
separated from the texts it generates a n d from t h e fictive world they are
made to ca l l u p . The actua l i ty of d i sc u rsive situations and i nd iv i d u a l
i nterl ocutors i s fi l tered out. B u t i nformants-along with field notes-are
cruc i a l i n termed i a ries, typ ica l l y exc l uded from authoritative ethnogra­
phies. The d i a logica l , situational as pects of ethnograph ic i nterpretation
tend to be ban i shed from the final representati ve text. Not enti re l y ban­
i shed , of cou rse; there ex i st approved topoi for the portraya l of the re­
sea rc h process .
We a re i ncreas i ng l y fa m i l i a r with the sepa rate fie ldwork account (a
su bgen re that sti l l tends to be c l assi fied as subjective, "soft," or u nscien­
tific), but even with i n c l assic ethnograph ies, more-or-less stereotypic
"fables of ra pport" narrate the atta i n ment of fu l l partic i pant-observer sta­
tu s . These fa bles may be to ld elaborate ly or in pass i ng, naive l y or i ron i­
ca l l y. They norma l l y portray the ethnographer's early i gnorance, m i s­
u ndersta nd i ng, lack of contact-freq uent l y a sort of ch i ld l i ke status
with i n the c u lture . In the Bildungsgesch ichte of the eth nogra phy these
states of i n nocence o r confusion are rep l aced by ad u l t, confident, d i s­
abu sed knowledge. We may cite aga i n Geertz's cockfight, where an early
a l ienation from the B a l i nese, a confused "nonperson" status, i s tra ns­
formed by the appea l i n g fable of the po l i ce ra id w ith its show of com­
p l i city ( 1 973 : 4 1 2-4 1 7). The anecdote establ i shes a pres u m ption of con­
nected ness, which perm its the writer to fu ncti on in h i s su bseq uent
analyses as an o m n i present, knowl edgeab l e exegete and spokesma n .
Th i s i nterpreter si tuates the ritual sport as a text i n a contextu a l world and
bri l l iantly " read s" its cu ltura l mea n i ngs. Geertz's abru pt d i sappearance

i n to h i s rapport-the quasi-i nvisi b i l ity of partic i pant observation-is pa r­

ad i gmatic. Here he makes use of an establ ished convention for staging
the atta i n ment of ethnogra p h i c authority. As a res u lt, we a re seldom
made awa re of the fact that an essenti a l part of the cockfi ght's construc­
tion as a text is d i a logica l -the a uthor's ta l k i ng face to face with partic­
ular B a l i nese rather than read i n g cu l t u re "over the [ i r] shou lders"
( 1 9 7 3 :45 2 ) .

I nterpreti ve anthropology, b y view i n g c u l t u res a s assembl ages of texts,

loosely and someti mes contrad ictora l l y u n ited , and by h i gh l ight i n g the
i nventive poes i s at work in a l l co l lective representations, has contri buted
sign ificantly to the defa m i l i a rization of eth nogra p h i c authority. I n i ts
m a i n stream rea l i st strands, however, it does not escape the general stric­
tu res of those critics of "colon i a l " rep resentation who, s i nce 1 950, have
rejected d i sco u rses that portray the c u l tu ra l rea l ities of other peo p l es
without p l a c i n g the i r own rea l i ty i n jeopardy. I n M i chel Lei ris' early cri­
tiq ues, by way of J acques Maq uet, Ta l a l Asad, and many others, the un­
rec i procal q u a l ity of eth nogra p h i c i nterpretation has been cal led to ac­
cou nt ( l e i r i s 1 95 0 ; Maquet 1 964; Asad 1 9 7 3 ) . Henceforth neither the
experience nor the i nterpretive activity of the scientific researc her can be
con s idered i n nocent. I t becomes necessary to conceive of eth nography
not as the experience and i nterpretation of a c i rcu mscri bed "other" rea l ­
i ty, but rather as a constru ctive negotiation i nvo l v i n g a t l east two, and
u su a l l y more, consc ious, po l itica l l y s i g n i ficant subjects . Parad igms of ex­
perience and i nterpretation a re y i e l d i n g to d i sc u rs i ve parad igms of d i a­
logue and po lyphony. The rema i n i n g sections of th i s cha pter w i l l su rvey
these emergent modes of authority.
A d i sc u rs i ve model of eth nogra p h i c pract i ce bri ngs i nto pro m i nence
the i n ters u bjectivity of a l l speech , along with its i m med i ate performative
context. Benven i ste's work on the constitutive ro l e of personal pronouns
and deixis h i gh l i ghts j u st these d i mensions. Every use of I pres u pposes a
you, and every i n stance of d i scou rse is i m med i ate l y l i n ked to a spec ific,
shared situation : no d i sc u rs i ve mea n i ng, then , w i thout i n terlocution and
context. The relevance of th i s emphasis for eth nography i s evident. F i eld­
work i s sign i ficantly com posed of language events; but language, in
Bakht i n 's words, " l ies on the borderl i n e betwee n onese l f and the other.
The word i n l a n gu age is h a l f someone e l se's." The Russian critic u rges a
reth i n k i n g of l a nguage i n terms of s pec ific d i sc u rsive situations : "There
42 D I S C O UR S E S

are," he writes, "no ' neutra l ' words and forms-words and forms that
can belong to 'no one' ; l anguage has been complete l y taken over, shot
through with i ntentions and accents ." The words of ethnographic writing,
then, can not be construed as monologica l , as the authoritative statement
about, or i nterpretation of, a n abstracted , textua l i zed rea l ity. The lan­
guage of eth nography i s shot through with other subjectivities and spe­
c i fic contextual ove rtones, for a l l language, in Bakht i n 's v iew, is "a con­
c rete heterog l ot conception of the world" ( 1 953 : 2 9 3 ) .
Forms o f eth nographic writ i n g that present themselves i n a "d i sc u r­
sive" mode tend to be concerned with the representation of resea rc h
contexts and s i tuations of i nterlocut i o n . Thus a book l i ke Pa u l Rab i n ow's
Reflections on Fieldwork in Morocco ( 1 977) is concerned with the rep­
resentation of a spec ific resea rch s ituation (a series of constra i n i ng ti mes
and pl aces) and ( i n somewhat fictional ized form) a seq uence of i nd ivid­
ual i nterloc utors. I ndeed an enti re new su bgenre of "fie ldwork accounts"
(of w h i c h Ra b i n ow's is one of the most trenchant) may be s ituated with i n
the d iscu rsive pa rad i gm o f eth nograph ic writing. Jea n ne Favret-Saada's
Les mots, Ia mort, les sorts ( 1 977) is an i n s i stent, self-conscious experi­
ment with eth nography i n a d i sc u rsive mode . 7 She argues that the event
of i nterloc ution a l ways assigns to the eth nographer a spec ific pos ition i n
a web o f i ntersu bject i ve re l ations. There i s n o neutra l standpo i nt i n the
power- laden fie l d of d i sc u rsive position i ngs, in a sh ifting matrix of rela­
tionsh i ps, of / 's and you 's .
A n u m ber o f recent works h ave chosen t o present the d i scurs i ve
processes of eth nography i n the form of a d i a l ogue between two i nd i vid­
uals. Ca m i l l e Lacoste- Dujard i n 's Dialogue des femmes en ethnologie
( 1 977), jean-Pa u l D u mont's The Headman and I ( 1 978), and Marjorie
Shostak's Nisa : The Life and Words of a !Kung Woman ( 1 98 1 ) are note­
worthy examples. The d i a l ogical mode is advocated with considerable
soph i stication in two other texts . The fi rst, Kev i n Dwyer's theoretica l re­
flections on the "d i a logic of eth nology" spri ngs from a series of interviews
w i th a key i n formant and j ustifies Dwyer's dec i sion to structu re h i s eth­
nography i ri the form of a rather I itera l record of these exchanges ( 1 97 7 ,
1 9 79, 1 98 2 ) . The second work i s Vi ncent Crapanzano's more com plex
Tuhami: Portrait of a Moroccan, another account of a series of i nterv iews

7. Favret-Saada's book is translated as Deadly Words (198 1); see esp. chap.
2 . Her experience has been rewritten at a nother fictional level i n Favret-Saada
and Contreras 198 1 .
O N E T HN O G R A P H I C A U T H O R I T Y 43

that rejects any sharp sepa rat i o n of an i nterpret i n g self from a textua l i zed
other ( 1 980; see a l so 1 9 7 7 ) . Both Dwyer and Crapanzano l ocate eth nog­
raphy in a process of d i a logue where i nterlocutors active l y negoti ate a
shared v i s i o n of rea l i ty. Crapanzano a rgues that th i s mutua l construction
m ust be at work in any ethnograph i c encou nter, but that partici pants tend
to assume that they have s i m pl y acq u iesced to the rea l ity of their coun­
terpa rt . Thus, for example, the ethnographer of the Trobriand I s l anders
does not openly co ncoct a version of rea l ity in col l aboration with h i s
i n forma nts b u t rather i nterprets t h e "Trobriand po i n t o f v i ew." Crapan­
zano and Dwyer offer soph i sti cated attem pts to break with th i s l iterary­
hermeneutical convention . In the process the eth nographer's authority as
narrator and i n terpreter i s a l tered . Dwyer proposes a hermeneutics of
"vu l nera b i l i ty," stres s i n g the ruptu res of fiel dwork, the d ivi ded pos ition
and i m pe rfect control of the eth nographer. Both Cra pa nzano and Dwyer
seek to represent the resea rch experience in ways that tear open the tex­
tua l i zed fabric of the other, and th u s a l so of the i nterpret i n g self.8 (Here
etymo l ogies a re evocative : the word text. is re l ated , as is we l l known, to
weavi ng, vulnerability to ren d i n g or wou n d i ng, in t h i s i n stance the open­
i n g up of a c l osed authority. )
The model of d i a l ogue bri ngs to pro m i nence prec i se l y those d i sc u r­
s i ve- c i rcu mstant i a l and i n tersu bject i ve--:;elements that R i coeur had to
exc l ude from h i s mode l of the text. B ut if i n terpretive a uthority is based
on the exc l usion of d i a l ogue, the reverse is a l so tru e : a p u rely d i a l ogical
a uthority wou l d repress the i nesca pable fact of textu a l i zation . Wh i l e eth­
nograph ies cast as encou nters between two i n d iv i d u a l s may successfu l l y
d ramatize the i ntersu bj ective give-and-take of fiel dwork and i n trod uce a
counterpo i nt of authori a l vo ices, they rema i n representations of d i a­
logue. As texts they may not be d i a l ogical i n structu re, for as Steven Tyler
( 1 98 1 ) poi nts out, a lthough Soc rates appears as a decentered partic i pant
in his encounters, Pl ato reta i n s fu l l contro l of the d i a l ogue. Th i s d i s p l ace­
ment but not e l i m i n ation of monological a uthority is characteristic of any

8 . It wou ld be wrong to gloss over the d ifferences between Dwyer's and

Crapanzano's theoretical positions. Dwyer, fol lowi ng Georg Lukacs, translates
dialogic i nto Marxian-Hegel ian d i a lectic, thus hol d i n g out the possi b i l i ty of a
restoration of the human su bject, a kind of completion i n and through the other.
Crapanzano refuses any anchor i n an englobing theory, h i s only authority bei n g
that o f t h e d i a l ogue's wri ter, an authority u nderm i ned b y a n i nconclusive narra­
tive of encou nter, ru ptu re, and confusi o n . (It is worth noti ng that d ialogic, as used
by Bakhti n, is not reducible to d ia lecti c . ) For an early advocacy of dialogical
anthropology see a l so Ted lock 1 9 79.
44 D I S CO U R S E S

approach that portrays the eth nogra pher as a d i screte cha racter i n the
fiel dwork narrative. Mo reover, there i s a freq uent tendency in fictions of
d i a logue for the eth nogra pher's counterpart to appear as a representative
of h i s or her c u l tu re-a type, i n the l a n guage of trad itional rea l ism­
through wh ich genera l soc i a l processes are revea led . 9 Such a portraya l
re i nstates the synecdoc h i c i nterpretive a uthority by which the ethnogra­
pher reads text in re lation to context, thereby constituti ng a mea n i ngfu l
"other" wor l d . If it is d i ffi c u l t for d i a l ogical portraya ls to escape typify i n g
proced u res, they can, t o a s i gn i ficant degree, res i st t h e pu l l towa rd a u ­
thoritative representation o f t h e other. T h i s depends on the i r a b i l i ty fic­
tion a l l y to m a i nta i n the stra ngeness of the other voice and to hold in view
the spec ific conti n gencies of the exchange.

To say that a n eth nography i s com posed of d i scou rses and that its d i ffer­
ent com ponents a re d i a logica l l y re l ated is not to say that its textua l form
shou ld be that of a l i teral d i a logue. I ndeed as Crapanzano recogn i zes i n
Tuhami, a t h i rd parti c i pa n t, rea l o r i magi ned, m u st fu nction a s med i ator
i n any encou nter between two i nd i v i d u a l s ( 1 980 : 1 4 7-1 5 1 ) . The fictiona l
d i a logue is i n fact a condensation, a s i m p l ified representation of complex
m u l tivoc a l processes. An a l tern ative way of representing th i s d i scurs ive
complexity is to u nderstand the overa l l cou rse of the research as an on­
goi ng negoti ation . The case of Marcel G r i a u le and the Dogon i s wel l
known a nd part i c u l arly clear-c ut. G r i a u l e's account of h i s i nstruction
in Dogon cosmological w i sdom, Die u d'eau ( 1 948a), was an early exer­
c i se in d i a l ogical eth nograph i c na rration . Beyond th i s spec ific i n ter­
locutory occas ion, however, a more com plex process was at work, for
it is apparent that the content and t i m i n g of the G r i a u le team 's long­
te rm researc h , spa n n i ng decades, was c lose l y monitored and sign ifi­
cantly shaped by Dogon tri bal authorities (see my d i scussion in Chap­
ter 2 ) . T h i s is no longer news . Many eth nogra phers have commented
on the ways, both subtle and b l atant, i n which the i r research was
d i rected or c i rcu mscri bed by the i r i nformants. In h i s provocat ive d i scus-

9. On rea l ist "types" see Lukacs 1 964, pass i m . The tendency to transform
an i ndividual i nto a c u l tu ra l enunciator may be observed in Marcel G riau le's
Dieu d'eau ( 1 948a). It occ u rs a mbivalently i n Shostak's Nisa ( 1 98 1 ) . For a d is­
cussion of th i s ambivalence and of the book's resu lting discursive complexity see
C l i fford 1 986b : 1 03-1 09 .

sion of this issue loan Lewis ( 1 9 7 3 ) even ca l l s anth ropology a form of

"plagiarism ."
The give-a nd-take of eth n ography is c l ea rly portrayed i n a 1 980
study noteworthy for its presentation with i n a si ngle work of both an
i nterpreted other rea l ity and the research process itse lf: Ren ata Rosaldo's
1/ongot Headhunting. Rosa ldo arrives i n the Ph i l i pp i n e h i g h l ands i ntent
on writ i n g a sync h ro n i c study of soc i a l structu re ; but aga i n and aga i n ,
over h i s o bj ections, he i s forced to l i sten to end less l l ongot narrati ves of
l oca l h istory. Dutifu l l y, d u m b l y, i n a k i nd of bored trance he tra nscri bes
these stories, fi l l i n g notebook after notebook with what he considers d i s­
posa b l e texts . O n l y after leav i n g the fie l d , and after a long p rocess of
re i n terpretation (a process made m a n i fest i n the ethnography), does he
rea l i ze that these obsc u re ta les h ave in fact provided him with his fi n a l
topic, the c u l t u ra l l y d i sti nctive l l ongot sense o f narrative a n d h i story. Ro­
sal do's experience of what m ight be cal led "d i rected writing" sharply
poses a fu ndamenta l question : Who i s actua l ly the author of fie l d notes ?
The issue is a subtle one and deserves systematic study. B ut enough
has been sa id to make the general poi nt that i n d i genous control over
knowl edge gai ned in the fie l d can be considerable, and even determ i n ­
i ng . C u rrent eth nogra p h i c writi ng i s see k i n g new ways t o represent ade­
q u atel y the a uthority of i nformants. There a re few mode l s to look to, but
it is worth reco n s ide r i n g the older textual com p i l ations of Boas, Ma l i­
nowsk i , Lee n h a rdt, and others . I n these works the eth nogra p h i c gen re
has not coal esced aro u n d the modern i n terpretational monograph
c l ose l y id entified with a person a l fieldwork experience . We can contem­
pl ate a n eth nogra p h i c mode that i s not yet authoritative i n those spec ific
ways that are now po l itica l l y and epistemologica l l y i n q uestio n . These
older assembl ages i n c l ude much that is actua l l y or a l l but written by
i nformants . One th i n ks of the ro le of George H u nt i n Boas' eth nography,
or of the fifteen "transcri pte u rs" l i sted i n Leenha rdt's Documents neo­
ca ledoniens ( 1 9 3 2 ) . 1 0
Ma l i nows k i is a com p l ex transitional case. H i s eth nogra ph ies reflect

1 0 . For a study of this mode of textual prod uction see C l i fford 1 980a . See
a l so in t h i s context Fontana 1 9 75, the i ntrod uction to Fra n k Russe l l , The Pima
Indians, on the book's h idden coauthor, the Papago Indian jose Lewis; Lei ri s
1 948 d i sc usses col l aboration as coauthors h i p, as does Lewis 1 9 7 3 . F o r a
forwa rd-looking defense of Boas' emphasis on vernacu lar texts and h i s co l labo­
ration with H u nt see Goldman 1 980.
46 D I S C O UR S E S

the i ncomplete coa lescence of the modern monograph . If he was cen­

tra l l y respo n s i b l e fo r the we l d i n g of theory and description i nto the au­
thority of the professional fieldworker, Ma l i nowski nonetheless i ncl uded
material that d id not d i rectly su pport h i s own a l l -too-c lear i nterpretive
s l ant. I n the many d ictated myths and spe l l s that fi l l h i s books, he pub­
l i shed much d ata that he ad m i tted l y d i d not u nderstand . The res u l t was
an open text subject to m u l ti p le rei nterpretations. It is worth compari ng
such o l der com pend i u ms with the recent model eth nography, which
c i tes evidence to support a focused i nterpretation but l i ttle e l se . 1 1 In the
modern , authoritative monograph there are, in effect, no strong vo ices
present except that of the writer; but in Argonauts ( 1 922) and Coral Car­
dens ( 1 9 3 5 ) we read page after page of magi ca l spel ls, none i n any es­
senti a l sense i n the eth nographer's word s . These d i ctated texts in a l l but
the i r physical i nscription a re written by specific u n named Trobrianders .
I ndeed a n y cont i n uous eth nographic expos ition routi nely fol d s i n to itself
a d i vers i ty of descri ptions, transc r i ptions, and i n terpretations by a variety
of i n d igenous "authors." H ow shou ld these authori a l presences be made
man ifest?

A u sefu l - if extreme-standpoi nt i s provided by Bakhti n 's analysi s of the

" polyphon ic" nove l . A fund amental cond ition of the gen re, he argues, is
that it re presents spea k i n g s u bjects in a fie ld of m u ltiple d i scou rses. The
nove l grapples with, and enacts, heterogloss i a . For Bakht i n , preoccu pied
with the representation of nonhomogeneous wholes, there a re no i ryte­
grated cu l tu ra l worlds or l a nguages . A l l attempts to posit such a bstract
u n i ties a re constructs of monologica l power. A "cu l ture" is, concrete l y,
an open-ended , creat ive d i a l ogue of subcu ltures, of i n s iders and outsid­
ers, of d i verse factions. A " l anguage" i s the i nterp l ay and struggle of re­
gional d i a l ects, professional jargons, generic commonpl aces, the speech
of d i fferent age gro u ps, i nd i v i d u a l s , and so fort h . For Bakhti n the po l y­
pho n i c novel is not a tou r de force of cu ltura l or h i storical tota l i zation (as
rea l ist critics such as Georg Lu kacs and Erich Auerbac h have argued) but
rather a carn ivalesque a rena of d i versity. Bakhti n d i scovers a utopian tex-

1 1 . )a mes Fernandez' elaborate Bwiti ( 1 985) is a self-conscious transgres­

sion of the tight, monograph ic form, returning to Ma l i nowskian sca le and reviv­
ing eth nography's "archival" fu nctions.

tual space where d i sc u rsive com plexity, the d i a logical i nterp l ay of

voices, can be accommodated . I n the nove l s of Dostoyevsky or D i ckens
he va l ues prec i se l y the i r res i stance to tota l ity, and his idea l nove l i st i s a
ventri l oq u i st- i n n i netee nth-centu ry parla nce a "polyphon i st." " H e do
the po l i ce i n d i fferent voices," a l i stener exc l a i m s ad m i ri n g l y of the boy
S l o ppy, who reads pu bl i c l y from the newspaper i n Our Mutual Friend.
But D i ckens the actor, oral performer, and po l yphon i st mu st be set
aga i nst F l a u bert, the master of a uthori a l contro l , mov i n g god l i ke among
the thoughts and fee l i ngs of his c h a racters . Eth nography, l i ke the nove l ,
wrestles w i t h these a l ternatives . Does t h e ethnographic writer portray
what natives th i n k by means of F l a u bert i a n "free i nd i rect sty l e," a sty l e
t h a t su ppresses d i rect quotation i n favor o f a control l i ng d i scou rse a lways
more or l ess that of the author? (Dan Sperber 1 98 1 , ta k i n g Eva ns­
Pritchard as h i s exa m p le, has conv i n c i ng l y shown that style indirect i s
i ndeed t h e preferred m ode o f ethnogra p h i c i n terpretati o n . ) Or does the
portraya l of other subjectivities req u i re a vers i o n that i s sty l i stical l y l ess
homogeneous, f i l led with D i c kens' "d ifferent voices" ?
Some use of i n d i rect sty l e is i nevitable, u n l ess the novel or eth nog­
ra phy is com posed ent i re l y of q u otations, somet h i n g that is theoret i ca l l y
poss i b l e b u t seldom attem pted . 1 2 I n pract i ce, however, t h e ethnogra phy
and the novel have recou rse to i nd i rect sty le at d i ffe rent l eve l s of a bstrac­
tion . We need not ask how F l a u bert knows what Emma Bovary is th i n k­
i ng, but the a b i l ity of the fieldworker to i n habit i n d igenous m i nds is
a l ways i n doubt . I ndeed t h i s i s a perm anent, u n resol ved problem of eth­
n ograph ic method . Eth nographers have genera l l y refra i ned from ascri b­
i n g bel iefs , fee l i ngs, and thoughts to i nd iv i d u a l s . They have not, how­
ever, hesitated to asc r i be subjective states to c u ltures. Sperber's analysis
reveals how ph rases such as "the N uer think . . ." o r "the N uer sense of
ti me" a re fu ndamenta l l y d i fferent from q u otations o r translations of i nd ig­
enous d i scou rse. Such statements are "without any specified speaker"
and a re l itera l l y eq u ivoca l , combi n i ng in a sea m l ess way the eth nogra­
pher's affi rmations with that of an i nformant or i nformants ( 1 9 8 1 : 7 8) .
Eth nograph ies a bou nd i n u nattri buted sentences such as "The spi rits re-

1 2 . Such a project is announced by Evans-Pritchard i n h i s i ntrod uction to

Man and Woman among the Azande ( 1 9 74), a late work that may be seen as a
reaction aga i nst the closed , analytical natu re of h i s own ea rlier eth nograph ies.
His acknowledged i n s p i ration is Mal i nows k i . (The notion of a book entirely com­
posed of quotations i s a modern ist dream associ ated with Wa lter Benj a m i n . )
48 D I S CO UR S E S

turn to the v i l l age at n i ght," descri ptions of bel iefs i n wh i c h the writer
assumes i n effect the voice of c u l tu re .
A t t h i s "c u l tu ra l " l evel eth nographers aspi re t o a F l a u bertian o m n i ­
science t h a t moves free ly th roughout a world of i n d i genous subjects. Be­
neath the su rface, though, the i r texts a re more u n r u l y and d i scordant.
Victor Tu rner's work prov ides a te l l i ng case i n poi nt, worth i nvesti gat i n g
more c lose l y as an exa m p l e o f t h e i nterp l ay o f monophon ic and po l y­
phonic exposi ti o n . Tu rner's eth nograph ies offer s u perbly com p l ex por­
traya l s of Ndembu ritu a l symbo l s a n d be l iefs ; and he has provided too
an u n usua l l y expl i c i t g l i m pse beh i n d the scenes . I n the m idst of the es­
says col lected i n The Forest of Symbols, h i s th i rd book on the Ndembu,
Tu rner offers a portra it of his best i nformant, "Muchona the Hornet, I n­
terpreter of Re l i gion" ( 1 9 6 7 : 1 3 1-1 50) . Muchona, a ritua l heal er, and Tu r­
ner a re d rawn together by thei r shared i nterest i n trad itional symbols,
etymologies, and esoteric mea n i ngs . They a re both " i nte l l ectu a l s," pas­
sion ate i nterpreters of the n u a nces and depth s of custom ; both are u p­
rooted scho l a rs s h a r i n g "the quench less th i rst for objective knowledge."
Tu rner compares Muchona to a u n i vers i ty don; h i s account of the i r co l­
laboration i n c l udes more than pass i n g h i nts of a strong psycho logic a l
doubl i n g .
There is, however, a th i rd p resent i n the i r d i a l ogue, Wi ndson Kash­
i na kaj i , a Ndembu sen ior teacher at the l ocal m ission schoo l . He brought
Muchona and Tu rner together and shares the i r pass i o n for the i n terpre­
tation of customary re l i g i o n . T h rough h i s b i b l ical ed ucation he "acq u i red
a fl a i r for e l ucidating knotty q u estions." N ew l y skeptical of Ch ristian
dogma a n d m i ss ionary priv i l eges, he i s l oo k i n g sympathetica l l y at pagan
re l igion . Kash i nakaj i , Tu rner te l l s us, "spanned the c u l tural d i stance be­
tween Muchona and myse lf, tra nsform i n g the l i ttle doctor's tec h n ical j a r­
gon and salty vi l l age argot i nto a prose I cou ld better grasp ." The three
i nte l l ectua l s soon "sett led down i n to a sort of d a i l y sem i n ar on rel i gion ."
Tu rner's acco u n ts of t h i s sem i n a r a re sty l ized : "eight months of exh i l a rat­
i n g q u i c kfi re ta l k among the th ree of us, m a i n l y about Ndembu ritual ."
They revea l an extraord i n a ry ethnogra p h i c "co l l oq uy" ; but s i g n i ficantly
Tu rner does not make h is th ree-way co l l aboration the crux of his essay.
Rather he focuses on Muchona, thus tra n sform i n g trialogue i nto d i a logue
and flatten i ng a com plex prod uctive rel ation i n to the " portrait" of an " i n­
formant." (Th i s red uction was i n some degree req u i red by the format of
the book i n w h i c h the essay fi rst a ppea red , J oseph Casagrande's i m por-

ta nt 1 960 co l l ection of "Twenty Portra its of Anthro po l ogica l l n formants,"

In the Company o f Ma n . ) 1 3
Tu rner's pu b l i s hed works vary cons iderably i n thei r d i sc u rs i ve struc­
ture. Some a re com posed l a rgel y of d i rect q u otations; in at least one
essay M uchona i s ident ified as the pri n c i pa l sou rce of the overa l l i nter­
pretation ; e l sewhere he is i n voked anonymously, for exa m p l e as "a m a l e
ritu a l spec i a l i st" ( 1 9 7 5 :40 - 4 2 , 87, 1 54 -1 5 6 , 244 ) . Wi ndson Kas h i na kaj i
is identified as a n ass i sta nt and tra n s l ator rather than as a sou rce of i nter­
pretat ions. Overa l l , Tu rner's eth nographies are u n usu a l l y po l ypho n i c,
open l y bu i l t u p from q u otations ( "Accord i ng to an adept . . . ," or "One
i n formant guesses . . . "). He does not, however, do the Ndembu in d i f­
ferent voi ces, and we hear l ittl e "sal ty v i l l age a rgot ." A l l the voi ces of the
field h ave been smoothed i nto the expos itory prose of more-or-l ess i nter­
changea b l e " i nformants." The stagi n g of i nd i genous speec h in an eth nog­
ra phy, the degree of tra n s l ation and fam i l i arization necessary, are com­
p l icated pract ical and rhetorical problems . 1 4 But Tu rner's works, by
givi ng v i s i b l e p l ace to i n d igenous i nterpretations of custo m , expose con­
cretel y these i ssues of textu a l d i a logism and po lyphony.
The i n c l u s i o n of Tu rner's portra it of Muchona i n The Forest of Sym­
bols may be seen as a sign of the ti mes . The Casagrande col lection i n
w h i c h i t origi na l l y appea red had the effect o f segregat i n g the cruc i a l i ssue
of re lations between eth nogra phers and the i r i nd igenous co l l aborators .
D i scussion of these i ssues sti l l had no p l ace with i n scientific eth nogra­
ph ies, but Casagrande's col lection shoo k the post-Ma l i nowski profes­
s i o n a l taboo on "priv i l eged i n forma nts." Raymond F i rth on Pa Fen u atara,
Robert Lowie o n J i m Carpenter-a long l i st ofd i sti ngu i shed anthropolo­
gi sts h ave descri bed the i n d igenous "ethnographers" with whom they
shared , to some degree, a d i sta nced , ana lytic, even i ron ic v i ew of c u s­
tom . These i nd iv i d u a l s became va l ued i nfo rmants because they u nder­
stood , often with rea l s u btl ety, what an ethnographic att itude towa rd c u l ­
ture entai led . I n Lowie's q u otation o f h i s C row i nterpreter (and fel l ow
" ph i l o l og i st") J i m Carpenter, one senses a shared outlook : "When you

1 3 . For a "group dynam ics" approach to eth nography see Yan nopoulos and
Martin 1 9 78 . For an ethnography expl icitly based on native "sem i n a rs" see Jones
and Kan ner 1 9 76.
1 4 . Favret-Saada's use of d i a l ect and ita l i c type i n Les mots, I a mort, les sorts
( 1 9 7 7) is one solution among many to a problem that has long preoccupied re­
al i st nove l i sts.
50 D I S C O UR S E S

l i sten to the o l d men te l l i ng about thei r v i s ions, you've j u st got to be l i eve

them" (Casagra nde 1 9 60 : 4 2 8 ) . And there is considerably more than a
w i n k and a nod i n the story recounted by F i rth about h i s best Ti kop ian
friend and i n formant:

On another occasion ta l k tu rned to the nets set for salmon trout i n the
lake. The nets were becom i ng black, possibly with some organic
growth , and tended to rot eas i l y. Pa Fen uatara then to ld a story to the
crowd assembled i n the house about how, out on the lake with his nets
one time, he felt a spirit goi n g among the net and making it soft . When
he held the net up he found it s l i my. The spirit had been at work. I
as ked h i m then if this was a trad itional piece of knowledge that spi rits
were responsible for the deterioration of the nets . He answered , "No,
my own thought." Then he added with a laugh, "My own piece of
trad itional knowledge." (Casagrande 1 960: 1 7-18)

The fu l l methodological i m pact of Casagrande's co l lection remains

l atent, espec i a l l y the s i g n i fica nce of its accou nts for the d i a l ogical pro­
d uct ion of eth nogra p h i c texts and interpretations. T h i s sign ificance is ob­
scu red by a tendency to cast the book as a u n i versa l i z i ng, human i st doc­
u ment reveal i ng "a ha l l of m i rrors . . . in fu l l variety the end less reflected
i mage of man" (Casagra nde 1 9 60 : x i i ) . In l i ght of the present crisis i n
et h nogra p h i c authority, however, these revea l i ng portraits spi l l i nto the
oeuvres of thei r authors, a l ter i n g the way they can be read . If eth nogra­
phy is part of what Roy Wagner ( 1 980) ca l l s "the i nvention of c u l t u re,"
its act ivity is p l u ra l and beyond the control of any i nd i v i d ua l .

One i n c reas i n g l y common way to man ifest the co l l aborative prod uction
of ethnograph i c knowledge i s to quote reg u l arly and at length from i n ­
form ants. ( A stri k i n g example i s We Eat the Mines, the Mines Eat Us
( 1 979) by J u ne Nash . ) B u t such a tactic o n l y beg i n s to break up mono­
phon i c authority. Quotations are always staged by the quater and tend
to serve mere l y as exam p l es o r confi rm i ng testi monies. Looking beyond
q u otation, one m ight i magine a more rad ical po l yphony that wou l d "do
the natives and the eth nographer in d i fferent vo ices " ; but t h i s too wou l d
o n l y d i splace eth n ographic authority, sti l l confi rm i ng the fi nal vi rtuoso
orchestrat ion by a s i ngle author of a l l the d i scou rses in h i s o r her text . I n
t h i s sense Bakht i n 's po l yphony, too narrowly identified with the nove l , i s
a domesti cated heterogloss i a . Ethnograph i c d i scou rses are not, i n any
ON E T HN O G R A P H I C A U T H O R I T Y 51

event, the speec hes of i n vented c haracters . I nfo rma nts a re spec ific i n d i ­
v i d u a l s with rea l proper names-n a mes that c a n b e cited, i n a l tered form
when tact req u i res . I nforma nts' i nte ntions a re overdeterm i ned, the i r
words po l i tica l l y and metapho rica l l y com p l ex . I f accorded an autono­
mous textu a l space, transcribed at suffic ient length , i n d i genous state­
ments make sen se in terms d i fferent from those of the arrangi ng eth nog­
rapher. Ethnography is i n vaded by heterogloss i a .
T h i s poss i b i l i ty suggests a n a l tern ate textu a l strategy, a utopia o f p l u ­
r a l authors h i p that accords t o co l l aborators n o t merely t h e status o f i n­
dependent e n u n c i ators but that of writers . As a fo rm of authority it m u st
sti l l be considered utopian for two reaso n s . F i rst, the few recent experi­
ments with m u l t i p l e-author works appear to req u i re, as a n i n stigat i n g
force, the research i n terest o f an eth nographer w h o i n t h e end assumes
an executive, ed itori a l pos ition . The authoritative sta nce of "gi ving
voi ce" to the other i s not fu l l y transcended . Secon d , the very idea of
p l u ra l authors h i p c h a l lenges a deep Western identifi cation of any text's
order w i th the i ntention of a s i ng l e author. If t h i s identification was l ess
strong when Lafita u wrote h i s Moeurs des sauvages ameriquains, and if
recent criticism has t h rown it i nto q uestion, i t i s sti l l a potent constra i nt
on eth n ogra p h i c writ i n g . Nonethel ess, there a re signs of movement i n
th i s domai n . Anth ropologi sts w i l l i n c reas i ngly have to share the i r texts,
and someti mes thei r title pages, with those i n d i genous col l aborators fo r
whom the term informants is no longer adeq uate, if it ever was .
Ra l p h B u l mer and l a n Maj nep's Birds of My Kalam Country ( 1 9 7 7)
is an i m po rtant prototype . (Sepa rate typefaces d i sti ngu ish the j u xtaposed
contri butions of eth nographer and New G u i nean, co l l abo rators for more
than a decade . ) Even more s i g n i ficant is the co l l ective l y prod u ced 1 9 74
study Piman Shamanism and Staying Sickness (Ka :cim Mumkidag),
w h i c h l i sts on i ts title page, without d i sti nction (though not, it may be
noted , i n a l phabetical o rder) : Don a l d M. B a h r, anthropo l ogist; j u a n G re­
gorio, shaman ; Dav i d I . Lopez, i n terpreter; and Al bert Alva rez, ed i tor.
Th ree of the fou r a re Pa pago I nd ians, and the book is consc ious l y de­
signed "to tra n sfe r to a shaman as many as possi b l e of the functions nor­
m a l l y assoc i ated with authors h i p . These i n c l ude the se lection of an ex­
pos ito ry style, the d u ty to make i n terpretations and ex planations, and the
right to j udge w h i c h th i ngs a re i m portant and w h i c h a re not" (p. 7 ) . B a h r,
the i n itiator and o rga n i zer of the p roject, opts to share authority as much
as poss i b l e . G regorio, the shaman, appears as the pri nc i pa l sou rce of the
"theory of d i sease" that i s transcri bed and transl ated , at two sepa rate
52 D I S CO U R S E S

l evels, by Lopez and A l varez. G regorio's vernacu lar texts i n c l ude com­
pressed , often gnom ic expl anations, which are themselves i nterpreted
and contextua l i zed by Bah r's separate com mentary. The book is u n us u a l
i n i t s textu a l enactment o f t h e i nterpretation o f i nterpretations.
I n Piman Sha manism the transition from i nd ividual enu nciations to
cu ltura l genera l izations i s always v i s i b le i n the separation of G regori o's
and Bah r's vo ices . The authority of Lopez , less visi ble, is a k i n to that of
Windson Kas h i nakaj i in Tu rner's work . H i s b i l i ngual fluency guides Bahr
through the s u btleties of G regorio's l a nguage, th us perm itti ng the shaman
"to speak at length on theoretical topics." Neither Lopez nor Alvarez a p­
pears as a spec ific voice i n the text, and thei r contribution to the ethnog­
raphy rem a i n s l arge l y i nv i s i b l e to a l l but q u a l ified Pa pagos, able to ga uge
the accu racy of the transl ated texts and the vernac u l a r nuance of Bah r's
i nterpretations. A l va rez' a uthority i n heres i n the fact that Piman Shaman­
ism is a book d i rected at separate aud iences . For most readers foc u s i n g
on t h e tra n s l ations and expla nations t h e texts pri nted i n P i m a n wi l l b e of
l i ttle or no i nterest. The l i ngu ist A l varez , however, corrected the tran­
scriptions and trans lations with an eye to the i r use in l a nguage teac h i ng,
u s i n g an orthography he had developed for that pu rpose . Thus the book
contributes to the Papagos' l i terary i nvent ion of the i r cu lture . Th i s d iffer­
ent read i ng, bu i l t i nto Piman Shamanism, is of more than l oca l signifi­
cance .
It i s i ntri n s i c to the brea ku p o f monological authority that eth nogra­
ph ies no longer add ress a s i ngle ge nera l type of reader. The m u l t i p l ica­
tion of poss i b l e read i n gs reflects the fact that "ethnographic" consc ious­
ness can no l o nger be seen as the monopo l y of certa i n Western cu ltures
and soc i a l c l asses. Even i n eth n ograph ies lacking vernac u l a r texts, i n d i g­
enous readers w i l l decode d ifferently the textual ized i nterpretations and
lore . Po lyphon ic works are part i c u l a r l y open to read i ngs not spec i fica l l y
i ntended . Trobriand readers may find Ma l i nowski's i n terpretations ti re­
some but h i s examp les and extended transcri ptions sti l l evocative.
N dembu wi l l not g l oss as q u ic k l y as E u ropean readers over the d ifferent
voi ces embedded in Tu rner's works.
Recent I iterary theory suggests that the abi I ity of a text to make sense
in a coherent way depends less on the wi l led i nten tions of an origi nating
author than on the c reative activ ity of a reader. To quote Ro l and Barthes,
if a text i s "a tissue of q u otations d rawn from i n n u merable centers of
c u l tu re," then "a text's u n ity l i es not in its origi n but in its desti nati on"
( 1 9 7 7 : 1 4 6 , 1 48) . The writ i n g of eth nography, an u n ru l y, mu ltisubjective
O N E T H N O G R A P H I C A UT H O R I T Y 53

activity, i s g i ven coherence i n particu l a r acts of read i ng . B u t there i s a l ­

ways a va riety o f poss i b le read i ngs (beyond mere l y i nd i v i d u a l appropria­
tions), read i n gs beyond the contro l of any s i ngle authority. One may ap­
proach a c l ass ic eth nography see k i ng s i m p l y to grasp the mea n i ngs that
the resea rcher de r i ves from represented c u l t u ra l facts. Or, as I have sug­
gested , one may a l so read aga i nst the gra i n of the text's domi nant voice,
see k i n g out other ha lf- h idden authorities, re i nterpret i n g the descriptions,
texts, and q uotati o n s gathered together by the writer. With the recent
q uestion i ng of colon i a l sty l es of representation, with the expansion of
l iteracy and ethnogra p h i c con sc ious ness, new poss i bi l ities for read i n g
( a n d th u s for writi ng) c u l tu ra l desc r i ptions are emerg i n g . 1 5
The textua l em bod i ment of authority i s a recurring problem for con­
tem pora ry experi ments in eth n ography. 1 6 An older, rea l i st mode-fig­
u red i n the frontispiece to Argonauts of the Western Pacific and based on
the construction of a c u l t u ra l ta b l eau v i vant desi gned to be seen from a
si ngle vantage poi nt, that of the writer and reader-can now be identi­
fied as only one poss i b l e pa rad igm for authority. Po l itical and epistemo­
logical ass u m ptions are bu i lt i nto th i s and other styl es, ass u m ptions the
eth nogra p h i c writer can no l onger afford to ignore. The modes of author­
ity revi ewed here-experientia l , i nterpreti ve, d i a l ogica l , po lyphon i c­
a re ava i lable to a l l writers of ethnograph i c texts, Western and non-

1 5 . An extremely suggest ive model of polyphon ic expos ition i s offered by

the projected four-vol u me edition of the eth nograph i c texts written, provoked ,
and transcri bed between 1 896 and 1 9 1 4 by James Wal ker on the Pine Ridge
Sioux Reservation . Three titles h ave appeared so far, ed ited by Raymond J .
DeMa i l le and E l a i ne Jah ner: Lakota Belief and Ritual ( 1 982a), Lakota Society
( 1 982b), and Lakota Myth ( 1 983). These engross i n g vo l u mes in effect reopen the
textual homogeneity of Wal ker's classic monograph of 1 9 1 7, The Sun Dance, a
sum mation of the i n d ividual statements pu b l i shed here i n translation. These
statements by more than t h i rty named "authorities" complement and transcend
Wa l ker's synthesis. A long section of Lakota Belief and Ritua l was written by
Thomas lyon , Wal ker's i nterpreter. The co l l ection's fourth vol ume wi l l be a trans­
lation of the writi n gs of George Sword, an Oglala warrior and j udge encou raged
by Wa l ker to record and i nterpret the trad itional way of l i fe. The first two vol u mes
present the u n publ ished texts of knowledgeable Lakota and Wa l ker's own de­
scriptions in identical formats . Ethnography appears as a process of col lective
prod uctio n . I t i s essential to note that the Colorado H i storica l Society's decision
to publish these texts was sti m u l ated by increasing req uests from the Oglala com­
m u n i ty at Pine R idge for copies of Wa l ker's materials to use in Oglala h i story
cl asses. (On Wa l ker see a l so C l i fford 1 986a : 1 5-1 7 . )
1 6 . F o r a very usefu l and complete su rvey o f recent experi mental eth nogra­
ph ies see Marcus and Cushman 1 982; see also Webster 1 98 2 ; Fa h i m 1 982; and
C l i fford and Marcus 1 98 6 .
54 D I S C O UR S E S

Western . None i s obsol ete, none pure : there i s room for i nvention with i n
each pa rad igm . We have seen how new approac hes tend t o red i scover
d i scarded practi ces . Po lyphon ic authority l ooks with renewed sympathy
to compend i u m s of vernac u l a r texts-expository forms d i sti nct from the
focused monograph tied to partic i pant observation . N ow that naive
c l a i ms to the authority of experience have been su bjected to hermeneu­
tic suspi c i o n , we may antici pate a renewed attention to the subtle i n ter­
play of person a l and d i sc i p l i nary components in eth nographic researc h .
Experi entia l , i n terpret ive, d i a logica l , a n d po lyphon ic processes are
at work, d i scordantly, i n any eth nography, but coherent presentation pre­
supposes a contro l l i ng mode of authority. I have argued that th i s i m po­
sition of coherence on an u n r u l y textu a l process is now i nescapably a
matter of strateg ic choice. I h ave tried to d isti ngu ish i m portant styles of
authority as they have become v i s i b l e i n recent decades. If eth nographic
wri ting i s a l ive, as I bel ieve it is, it i s strugg l i ng with i n and aga i nst these
poss i b i l ities.
In fact the sociologist and his "object " form a couple where
each one is to be interpreted through the other, and where the
relationship must itself be deciphered as a historical

2 . Power and Dialogue in Ethnog­

raphy: Marcel Griaule's Initiation

M A R C E L G R I A U L E cut a figu re-se lf-confident and theatrical . He be­

gan h i s career as an aviator i n the yea rs j ust after the Fi rst World War.
( later, i n 1 946, as holder of the fi rst chai r i n eth nology at the Sorbon ne,
he wou l d lect u re in his air force officer's u n iform . ) An energeti c promotor
of fieldwork, he portrayed it as the conti n u ation-by scientific means­
of a great trad ition of adventure and exploration ( 1 948c : 1 1 9) . In 1 92 8 ,
encou raged by Marcel M a u s s and t h e l i ngu i st Marcel Cohen, G r i a u l e
s pent a yea r i n Eth i o p i a . He retu rned a v i d for n e w exploration, a n d h i s
p l a n s bore fru it two years l ater i n t h e m uch-pu b l ic i zed M i ssion Dakar­
Dj i bouti , w h i c h for twenty-one months traversed Africa from the Atlantic
to the Red Sea along the lower r i m of the Sahara . Largely a m useu m ­
col lect i ng enterprise, the m i ss i o n a l so undertook extended eth nographic
soj o u rn s i n the French Sudan (now Mal i), where Griaule fi rst made con­
tact with the Dogo n of Sanga, and i n Eth iopia (the reg ion of Gondar) ,
where t h e exped ition spent five month s . A m o n g t h e m i ssion's n i ne m e m ­
bers (some com i n g and goi n g at various poi nts) were Andre Schaeffner,
Deborah Lifch itz, and M ichel Le i ris, each of whom wou ld make sign ifi­
cant eth nograph i c contri butions.

56 D I S C O UR S E S

Thanks large l y to the p u b l i c ity sense of Georges-Henri Riviere-a

wel l -con nected jazz amateu r engaged by Pau l Rivet to reorga n i ze the
Trocadero Eth nogra p h i c Museum-the M i ssion Dakar- Dj i bouti was pa­
tro n i zed by Pa risian h igh soc iety. The Chamber of Deputies voted a spe­
c i a l l aw of authorizatio n , and G r i a u l e and Rivi ere ski l lfu l l y explo ited the
postwar vogue for th i ngs Afri can i n sol i c i ti ng fu nds and personnel . The
underta k i n g partook a l so of a certa i n techno logical bravado rem i n i scent
of the period's famous exped itions, fi nanced by C i troen , La Croisiere
jau ne, and La Croisiere N o i re, each a tou r de force of mob i l ity crossi n g
whole continents b y automobi l e . G riau le, an early a i rpl ane enthusiast,
wou ld be fasc i nated th roughout h i s career by technological aids to eth­
nography: conventional and aer i a l photography, sou nd-record i ng de­
vices, and even the project of a researc h boat-cum-l aboratory for use
on the N iger.
The m i ssion's " booty," i n R i vet and Riviere's term ( 1 93 3 : 5 ), i n c l uded
among i ts many photos, record i ngs, and docu ments 3 , 500 obj ects des­
ti ned for the Trocadero m u seu m, soon to become the Musee de
I ' Homme. The idea was o n l y j u st wi n n i ng accepta nce in England and
America, with Rockefe l ler fu nd i ng of the I ntern ati onal African I nsti tute,
that i nten s i ve fie l d stud ies were in themselves enough to j ustify major
subventio n s . Thus co l l ecti ng wa s a fi nanc i a l necessity, and the m ission
brou ght back wh atever authentic objects it could decently-and occa­
siona l l y su rreptitiously-acq u i re . The postwar pass ion for /'art negre fos­
tered a c u l t of the exotic artifact, and the carved figures and masks of
West and Equatorial Africa sati sfied perfectly a Eu ropean fetishism nour­
ished on c u b i st and su rrea l i st aesthetics (see Cha pter 4 ; a l so j am i n
1 982a) .
From 1 93 5 to 1 9 3 9 G r i a u l e orga n i zed group exped itions to the
French Sudan, Camero u n , and Chad, in wh ich mu seu m co l lecting
p l ayed a lesser ro l e . I n annual or biannual visits to West Africa focusing
i ncreas i n g l y on the Dagon, he worked out a d i st i n ctive ethnographic
" method ." For Griau le the co l l ection of artifacts was part of the i ntensive
docu mentat ion of a u n i fied cu lture area, a region center ing on the bend
of the N i ger, and particu l arly the home of the Bambara and Dago n , with
whom he spent a bout th ree years over ten ex ped itions ( Lettens
1 9 7 1 : 504) . G riau le's desc riptions were ca rtographic and archaeo log i ca l
as wel l a s eth nogra p h i c ; he was concerned w i t h variations i n c u l tura l
tra its, t h e h i story of m i grations, a n d t h e overlay o f civi l izations i n West
Afri ca. I ncreasi ngly, however, h i s i n terests focu sed on synchron ic c u i -

tura l pattern s . Over t i m e he esta b l i shed , t o h i s own sati sfaction, t h e ex­

istence of a ram ified but coherent cu lture a rea he l ater portrayed as one
of th ree major d ivisions of s u b-Saharan Africa : the Western Sudan, Bantu
Africa, and a n i ntermed iate zone in Camero u n and C h ad . Each region
was c h a racterized by a trad itional sophie o r science-a mode of knowl­
edge i n scribed i n language, habitat, oral trad ition, myth, technol ogy, and
aesthetics. G ri a u l e d i scerned common pri n c i p les u nderl y i n g the three
African epi stem o l ogical fie lds, and t h i s perm i tted h i m to u se the Dogan
and thei r neighbors as priv i l eged exa m p l es of l'homme nair- m i cro­
cosms of "African" thought, civi l ization, ph i losophy, and re l i g i o n . A
c haracteristic movement from parts to wholes to more i nc l u s i ve wholes
was G r i a u l e's bas i c mode of eth nogra p h i c representation . It m i rrored,
and fou n d confirmation i n , Dagon sty les of thought, with i ts encom pass­
i n g symbo l i c correspondences of m i crocosm and macrocosm , of body
and cosmos, of everyday deta i l s and patterns of myth .
A n u mber of d i fferent approaches are su bsu med u nder the general
label of the Gri a u l e schoo l . 1 The tota l project spans five decades, fa l l i n g
rough l y i nto two phases : before and after Ogotem mel i . I n 1 94 7 , i n a now
l egendary series of i nterv i ews, the Dogan sage Ogotemme l i , apparently
acti n g on i n structions from tri bal e l ders, i ndoctr i n ated G riau l e in the
deep w i sdom of his peo p l e ( G ri a u l e 1 948a) . The fi rst decade of research
at Sanga had been exh austive l y documentary i n character; now, with
access to the knowledge revea led by Ogotemme l i and other q u a l i fied
i nformants, the task became exegetical. Ogotemme l i 's el aborate know l ­
edge- re i nfo rced a n d extended b y other sou rces-appea red t o provide
a potent " key" to Dogan c u l t u re ( G r i a u l e 1 95 2 c : 548). Seen as a ki nd of
l ived myth o l ogy, it provided a framework for graspi ng the Dogan world
as a n i ntegrated whole. This i m manent structu re-a "metaphysic," as
Griaule l i ked to ca l l i t-offered a p u re l y i nd igenous orga n ization of the
com p lex tota l soc i a l facts of Dogan l i fe.

1 . There are many personal varia nts, and one shou ld d i stinguish the fol l ow­
i n g standpoints: the "core" of the ongoing resea rch on the Dogon and Bam bara
is that of Marcel Griau le, Germa i ne Dieterlen, and Solange de Ganay. Genevieve
Ca lame-G ria u l e and Dom i n ique Zahen contributed d i rectly to the project, but
from d i stinct methodological standpoi nts. jean-Pa u l Lebeuf, an earl y co-worker,
shared Griau le's general v iewpoi nt, but h i s work was concentrated in Chad. jean
Rouch, Luc de Heusc h , and various later students rema i n ambivalently loyal to
the "trad ition." Den ise Pau l me, Michel Lei ris, and Andre Schaeffner, early con­
tributors to the Dogon project, have always mai nta i ned a skeptical d i sta nce from
the u ndertaking and shou ld not be i n c l uded i n the "school ."
58 D I S CO U R S E S

Fu l l com p i l ations of th i s sagesse, an enormously deta i l ed system of

symbo l ic and na rrative correspondences, appeared only after Griau le's
death i n 1 95 6 . The masterpieces of the G r i a u l e schoo l 's second peri od
are Le renard pale, written with h i s c losest col l aborator Germa i ne D ie­
terlen ( 1 965), and Ethno/ogie et langage: La parole chez les Dogan by
h is daughter, the d i sti ngu ished ethnol i ngu i st Genevi eve Ca lame-G r i a u l e
( 1 9 6 5 ) . I n these works o n e hears, as i t were , two fu l l chords o f a Dogan
sym phony : a myth ic ex p l a nation of the cosmos and a native theory of
language and express ivity. More than j ust native exp l a nations or theories,
these s u perb compend i a present themselves as coherent arts of l ife, so­
c iomyth i c l a ndscapes of physiology and persona l i ty, symbo l i c networks
i n carnate in a n i nfi n ity of d a i l y deta i l s .
The work o f G ri a u l e a n d h i s fo l l owers i s o n e o f the c l assic ach ieve­
ments of twentieth-centu ry eth nography. With i n certa i n areas of empha­
s i s its depth of comprehension and com pleteness of deta i l a re u n pa ra l ­
leled . B u t given i t s rather u n usual focus, t h e extreme natu re o f some of
its c l a ims, and the cruc i a l , problematic ro le of the Dogon themselves as
active agents in the long ethnograph ic process , G r i a u le's work has been
subj ected to sharp criticism from a variety of standpoi nts. Some have
noted its idea l i stic bias and its lack of h i storical dynamism (Baland ier
1 960; Sarevskaj a 1 9 6 3 ) . B ritish soc i a l anth ropo logi sts have ra i sed skep­
tica l q uestions about G ri a u le's fieldwork, notably his l ife long re l i a nce on
transl ators and on a few privi leged i nforma nts attu ned to his i nterests
(whose i n itiatory knowledge m i ght not be read i l y genera l i zable to the
rest of soc i ety) . Fol l owers of Ma l i nowski or Evans-Pritchard have m i ssed
i n G r i a u le's work any susta i ned attention to da i l y exi stence or po l i tics as
actu a l l y l ived , and in general they are wary of a too perfectly ordered
vision of Dogon rea l i ty ( R ic hard s 1 9 6 7 ; Douglas 1 9 6 7 ; Goody 1 9 6 7 ) .
Reread i n g t h e Dogon corpus c l ose l y, other cri tics have begu n , on
the basis of i nte rnal contrad ictions, to u n rave l the eq u i l i b r i u m of Dogon
mytho l ogy and to questi on the processes by wh ich an "abso l ute su bject"
(here a u n i fied construct cal led "the Dogan") is constituted in ethno­
gra p h i c interpretation ( Lettens 1 9 7 1 ; M i c hel-J o nes 1 9 78) . In the wake of
co lon i a l i s m , G r i a u l e has been taken to task for his con sistent preference
for an Afr i can past over a modern i z i n g present. Africans have criticized
h i m for essenti a l i z i n g trad itional c u l tu ra l patterns and repressi ng the ro le
of i nd iv i d u a l i nvention i n the elaboration of Dogon myth (Hou ntondj i
1 9 7 7 ) . After 1 95 0 G r i a u l e's work reso nated strongly with the negritude
movement, particu l a rly with Leopo ld Senghor's evocati on of an African
P O W E R A N D D I A L O G U E I N E TH N O G R A P H Y 59

essence. But as Senghor's brand of negritude has yielded to Aime Ce­

sai re's more syncretic, i m p u re, i n ventive conception of c u ltural identi ty,
G ri a u le's African metaphys i c h as beg u n to seem an a h i storica l , idea l ized
a l ter ego to a tota l iz i ng occ identa l h u m a n i s m .
It is i m poss i b l e here t o eva l uate m a n y o f t h e spec ific criticisms lev­
e l ed at G r i a u le, espec i a l l y in the a bsence of deta i l ed restudy of the
Dago n . A few methodological warn i ngs are necessary, however, for
approac h i ng s u c h a contested oeuvre. The h i storian of fiel dwork is ham­
pered by l i m i ted and foreshortened ev idence; i t i s a l ways d ifficult, if not
i m poss i b le, to know what ha ppened i n an eth nograp h i c encou nter. (Th i s
u nce rta i nty i s at l east part l y respo n s i b l e for t h e fact t h a t t h e h i story of
anth ropol ogy has tended to be a h i story of theory, even though the
modern d i sc i p l i ne has defi ned itsel f by reference to i ts d i sti nct "method .")
U s ua l l y, as in G ri a u le's case, one m ust re l y heav i l y on the eth nographer's
own ex post facto narrations, accou nts that serve to confi rm h i s authority.
One can a l so d raw o n h i s methodological prescr i ptions and those of h i s
co l l aborators ; b u t these too tend t o b e overly systematic rational izations
com posed after the fact . A scattering of re l evant journals and memoi rs
can he l p somewhat ( Le i ri s 1 9 34; Rouch 1 9 78b; Pa u l me 1 9 7 7) , as can a
critica l read i ng of p u b l i sh ed eth nograph ies and field notes-where ava i l ­
a b l e (and comprehen s i b l e) . 2 D i rect evidence o f the i n terpersonal dynam­
ics and po l i tics of researc h , however, i s l a rge l y absent. Moreover, there
is an enormous gap i n a l l h i stories of fie ldwork: the i n d i genous "side" of
the sto ry. How was the research process u nderstood and i nfl uenced by
i n formants, by tribal authorities, by those who d i d not cooperate (d.
Lew i s 1 9 7 3 ) ? G ri a u le's story has the merit of making th i s part of the
encou nte r i nescapa b l e . Yet our knowledge of Dogan i nfluences on the
eth nogra p h i c process rem a i n s fragmenta ry.
It i s s i m p l i stic to tax G r i a u l e w i th project i n g onto the Dogan a s u b­
ject i ve v i s i o n , with deve l o p i n g a research method for e l i c i t i n g essenti a l l y
what h e was l oo k i n g for ( Lettens 1 9 7 1 : 3 9 7 , pass i m ) . Even the more cred-

2. Anyone who has tried to rei nterpret field notes wi l l know it is a problem­
atic enterprise. They may be gnom ic, heteroglot shorthand notes to oneself, or
the sorts of "field notes" often q uoted in publ ished eth nographies-form u l ated
summaries of events, observations, and conversations recom posed after the fact.
It i s wel l-n igh i m possible to d i sentangle the i nterpretive processes at work as field
notes m ove from one level of textua l i zation to the next. Griau le's 1 73 rich ly
deta i led "fiches de terra i n " for the cruc ial i nterviews with Ogotem me l i (Griaule
1 946) are clearly the prod uct of at least one rewriti ng, e l i m i nating specific l i n ­
gu i stic problems, the presence o f the translator Kogem, a n d s o o n .
60 D I S C O UR S E S

i b l e c l a i m that G r i a u l e overstressed certa i n parts of Dogan rea l ity at the

expense of others assumes the existence of a natura l entity ca l l ed Dogan
c u l t u re apart from its ethnogra p h i c i nventions. Even if it is true that key
i nformants became " G r i a u l ized ," that G r i a u l e h i mse l f was " Dogo n i zed ,"
that Ogotemmel i 's w i sdom was that of an i n d i vi d u a l "theo l ogian," and
that the "sec ret," i n itiatory nature of the revea led knowledge was system­
atica l l y exagge rated , even if other priorities and methods wou ld certa i n l y
have produced a d iffe rent eth nography, it does not fo l low that Griau le's
version of the Dogan is fa lse. H i s writi ngs, and those of h i s assoc i ates,
express a Dogan truth-a complex, negot iated , h i stori ca l l y conti ngent
truth spec ific to ce rta i n re lations of textual prod uction . The h i storian asks
what kind of truth G ri a u l e and the Dogan he worked with prod uced , i n
what d i a logical cond itions, w i th i n what po l itical l i m i ts, i n what h i stori­
cal c l i mate.
Masterpieces l i ke Le renard pale and Ethnologie et langage are elab­
orate i n ventions a uthored by a va riety of su bjects, E u ropean and Africa n .
These compend i u m s do not represent the way "the Dogan" th i n k : both
the i r enormous com plexity and the absence of fem a l e i nformants cast
doubt on any such tota l i z i n g c l a i m . Nor is the i r "deep" knowledge an
i nterpretive key to Dogan rea l i ty for anyone beyond the eth nographer
and a sma l l n u m ber of native " i nte l lectua l s ." To say that these Dogan
truth s are s pec ific i nventions ( rather than parts or d i stortions of " Dagon
c u l tu re") , however, is to take them seriously as textual constructions,
avo i d i ng both celebration and pole m i c .
T h e G r i a u l e trad ition offers o n e o f t h e few fu l ly e l aborated a l terna­
tives to the Anglo-American model of i ntensive pa rt i c i pant observation .
For t h i s reason a lone it is i m portant for the h i story of twentieth-century
eth nography-parti c u l a r l y with the recent d i scovery i n America of
"long-term fie l d researc h " ( Foster et a l . 1 9 79). G ri a u le's writings are a l so
i m po rtant (and here we must separate the man from h i s "schoo l " ) for the i r
u n usual d i rectness i n portray i ng resea rc h as i n herently agon i stic, theat­
rica l , and fra ught w i th power. H i s work be longs manifestly to the co l o­
n i a l pe riod . Thanks to G r i a u le's d ramatic fla i r and fond ness for overstate­
ment, we can perceive clearly certa i n key ass u m ptions, ro les, and
systems of meta phors that em powered ethnography d u ri n g the th i rties
and forties.

One can not speak of a French "trad ition" of fiel dwork, as one refers (per­
haps too eas i l y) to B ritish and American schoo l s . N onetheless, if o n l y by

contrast, G ri a u l e's eth nography does appear to be pecu l i arly French . We

can suggest t h i s rather e l u sive q u a l i ty by evoking briefly two i nfluent i a l
precu rsors . I n Pa ris the most i m portant advocates o f fieldwork d u r i n g the
1 920s were M a rcel Mauss and Mau rice Del afosse, who col la borated
with Lucien Levy- B ru h l and Rivet to found the l n stitut d ' Eth nologie. Here
after 1 92 5 a generation of "Africa n i st" eth nographers was trai ned .
I n the fi rst th ree decades o f the centu ry B l ac k Africa was com i n g
i nto focus, separated from t h e "ori enta l " Magh reb . By 1 9 3 1 , when the
Journal de Ia Societe des Africanistes was fou nded , it had become pos­
s i b l e to s pea k of a fie l d cal led "Africa n ism" ( modeled on the older syn­
thetic d i sc i p l i n e of orienta l i s m ) . The fash ionable vogue for /'art negre and
black music contributed to the formation of a cu ltural object, a civilisa­
tion about w h ich synthetic statements cou ld be made. Maurice Dela­
fosse's Noirs de / 'Afrique and L 'ame nair contributed to th i s deve lop­
ment, a long with the transl ated writi ngs of F roben i u s . G r i a u le's work
u n fo lded with i n the Afri ca n ist parad igm, mov i n g assoc i atively from spe­
c i fic stud ies of parti c u l a r popu l ations to genera l izations a bout l'homme
noire, African c i v i l izat i o n , and metaphysics (G riau l e 1 9 5 1 , 1 95 3 ) .
A t t h e l n stitut d ' Et h n o l og i e a reg u l a r stream o f colon i a l officers stud­
ied eth nogra p h i c method as part of their tra i n i n g at the Eco le Colon i a l e ,
w here Del afosse w a s a pop u l a r teacher before h i s death i n 1 9 2 6 . As a
veteran of exte nded d u ty i n West Africa, Del afosse knew African lan­
gu ages and cu ltures i nti mate l y. When his health was u ndermi ned by the
rigors of consta nt travel and resea rc h , he reti red to France, becomi ng the
fi rst p rofessor of B l ack African l a n gu ages at the Eco l e des Langues Ori­
enta les. A sch o l a r of great erud ition, he made contri butions to African
h i story, ethnography, geography, and l i ng u i stics. At the Eco l e Co l o n i a l e ,
where Africans had long been cons idered c h i l d l i ke i nferiors, he taught
the fu ndamenta l eq u a l ity (though not the s i m i l arity) of races. D i fferent
m i l i eux prod uce d i fferent civi l i zations. If the Africans are tech n i ca l l y and
mate r ia l l y backward , th i s i s a h i storical accident; the i r art, the i r mora l
l ife, the i r re l i gions a re nonetheless fu l l y developed and worthy of es­
teem . He u rged h i s students toward eth n ography and the mastery of i n­
d i genous l a n guages . H i s a uthority was concrete experience, h i s persona
that of the broussard-the man of the bac k country, tough-mi nded , icon­
oc l astic , h u mane, i m pat ient with h i erarchy and the a rtifices of po l i te so­
c i ety ( D e lafosse 1 90 9 ; cf. Desch a m ps 1 9 7 5 : 9 7 ) . For a generati on of
you ng, l i bera l l y i n c l i ned colon i a l officers he represented an authentic,
concrete way to " know" Africa and to com m u n icate its fasc i nati on .
After Del afosse's death the pri n c i p le i nfluence on the fi rst generation

of professional fieldworkers i n France was exerted by another charis­

matic teac her, Marcel Mauss. Though he never undertood fiel dwork,
Mauss cons i stently depl o red France's backward ness in th is doma i n
(Mauss 1 9 1 3 ) . A t the l nstitut d' Eth nologie h e ta ught a yearly cou rse (eth­
nograph ie descriptive) spec i fica l ly geared to fiel dwork methods . Mauss
was a nyth i ng but an abstract, bookish sc hol ar; anyone who looks at h i s
"Tech n iq ues d u corps" ( 1 934) c a n see an acute power o f observation, a n
i nterest i n t h e concrete a n d the experimental (d. Condom i nas 1 9 72a).
Mauss u rged all h i s students toward eth nography; between 1 92 5 and
1 940 the l n stitut sponsored more than a h u n d red field tri ps (Karady
1 98 1 : 1 76) . U n l i ke W H . R. Rivers, Ma l i nowski, and later Griaule,
whose teac h i ng reflected the i r own experiences i n the fie l d , he did not
propou nd a d i sti nct research "method " ; but if he lacked i nti mate expe­
rience, he d i d not feel compe l l ed to rational ize or j usti fy h i s own prac­
tice. D rawn from the fieldwork trad itions of vari ous nations, h i s cou rse
was an i nvento ry, c l assification, and critique of poss i b l e method s . Mauss
provided a sense of the com p l ex i ty of "tota l soc i a l facts" (Mauss
1 92 4 : 2 74) and the d i fferent means by which descriptions, record i ngs,
textua l accou nts, and col lections of artifacts cou l d be constituted . H i s
wide-rangi n g Manuel d'ethnographie ( 1 947), a com p i l ation of cou rse
notes brought together by Den ise Pau l me s ho rt l y before h i s death , makes
it c l ear that the idea of a priv i l eged approach was q u ite foreign to h i m .
Mauss strongly suppo rted the general trend o f modern acade m i c
fiel dwork, u rgi ng "the professional eth nographer" to adopt "the i ntensive
method" ( 1 947 : 1 3 ) . Serious com parative work depended on the com ple­
tion of fu l l local desc r i ptions. A l though the Manuel's recom mendations
reflect a c l ose know l edge of American and B ritish tec h n i q ues, there is no
emphasis on i nd ividual parti c i pant-observation . Mauss endorses team re­
searc h ; overa l l h i s approac h is documentary rather than experiential and
hermeneuti c .
Th i s documenta ry concern wou l d b e reflected i n t h e i n trod uction to
G ri a u l e's fi rst major field monograph : "Th i s work presents documents
re lative to the Masks of the Dogon, co l lected d u r i n g research tri ps among
the c l iffs of Band i agara" ( G ri a u l e 1 93 8 : v i i ) . 3 I t is hard to imagine an ac­
cou nt in the Mal i nowskian trad ition beg i n n i ng in this way. Although
G r i a u l e does considera b l y more in Masques Dogons than s i m p l y d i splay

3. Here and throughout th is book translations of foreign works are my own

u n l ess I c i te a publ i shed Engl ish translation .

co l lected docu ments, the metaphor revea l s a particu l a r e m p i rical styl e

(cf. Lee n hardt 1 9 3 2 ; C l i ffo rd 1 98 2 a : 1 3 8 -1 4 1 ) . F o r Mauss, w h o accepted
an o l der d iv i s i o n of l abor between the man i n the field and the theorist
at home, descri ption shou l d never be governed by explanatory concerns
(Mauss 1 94 7 : 3 8 9 ) . To provide the k i nd of i n formation u sefu l to a com­
parative soc i o l ogy, the eth nographer shou ld avo id bu i l d i ng too much i m­
p l i c i t explanation i nto the ethnographic data i n the process of i ts const i ­
tution . M a u s s gave no spec i a l status t o t h e i d e a that a synthetic portra it
of a c u l tu re (someth i ng for h i m massive l y overdetermi ned) cou l d be pro­
d u ced t h rough the research experience of an i n d i v i d u a l s u bject or bu i lt
a ro u n d the a n a l ys i s of a typ i c a l or central i n stitution . H i s l i m it i n g notion
of "tota l soc i a l facts" l ed h i m rather to recom mend the deployment of
m u l t i p l e documentary methods by a variety of s pec i a l i zed observers.
Worki n g at a h i gher l evel of a bstraction, the soc i o l ogist cou ld perhaps
"gl i m pse, meas u re, and hold in eq u i l i b r i u m " ( 1 9 2 4 : 2 79) the d i fferent
strata of "tota l " facts-tec h n o l ogica l , aesthetic, geographica l , demo­
gra p h i c , econom ic, j u r i d i ca l , l i ngu i stic, rel i gious, h i storica l , and i nter­
cu l t u ra l . B u t the task of the ethnographer, whether alone or in a research
tea m , was to amass as com p l ete a corpus as poss i b l e : texts, a rtifacts,
maps, photographs, and so forth- "doc u ments" p recisely local ized and
coveri n g a b road range of c u ltu ra l phenomen a . F i e ldworkers should con­
struct "series and not panop l i es" (p. 2 1 ) . Mauss used old terms p rec i se l y :
a panoply i s a fu l l com p l ement o f arms, a s u i t o f armor w i th a l l i ts ac­
cou terments. The term suggests a fu nctional i n tegration of parts depl oyed
and d i s p l ayed arou nd a coherent, effecti ve body. Mauss d i d not see so­
c i ety or cu l t u re t h i s way. One shou l d be wary of red u c i n g h i s concept of
total soc i a l facts (re m i n i scent of Freud's "overdeterm i nation") to a fu nc­
tiona l i st notion of the i n terre l ation of parts .
Mau ss's e l u s i ve concept neverthel ess articu l ated a fu ndamental pre­
d icament for twentieth-centu ry ethnographers. If every "fact" is suscep­
ti b l e to m u l t i p l e e ncod i ng, m a k i n g sense in d i verse contexts and i m­
p l i cati n g i n its com prehension the "tota l " ensemble of re l ations that
constitutes the soc iety u nder study, then th i s assumption can serve as
encou ragement to grasp the ensem b l e by foc u s i n g on one of its pa rts .
I ndeed t h i s i s what fiel dworkers h ave a l ways done, bu i l d i ng u p soc i a l
who les ( "c u lture" i n t h e American trad ition) through a concentration o n
sign i ficant e l ements . Many d i ffe rent approaches have emerged : t h e focus
on key " i n stitutions" (Mal i nowski 's Trobriand Kula, Eva ns-Pritchard 's
Aza nde witchcraft) ; the bri ngi n g to the foreground of "tota l i z i n g c u l tu ra l

performances" ( B a l d w i n Spencer and F. G i l len's Arunta i n i tiation , G reg­

ory Bateson 's latm u l N ave n , Geertz's B a l i nese cockfight) ; the identifica­
tion of privi l eged armatu res to which the whole of culture cou l d be
rel ated ( R i vers' "genea logical method" and Radel iffe-Brown 's "soc i a l
structu re") ; or even G riau le's l ate concepti on o f i n itiatory knowledge as
the key to a u n ified representation of West Afri can c u l tu res. In d i fferent
ways the new generati on ot academic fieldworkers were a l l looking for
what G ri a u l e wou l d recommend, defend i ng h i s practice of teamwork i n
the fie ld-a " ra p i d , s u re method " able t o grasp synthetica l l y a n over­
determ i ned c u l tu ra l rea l ity ( 1 9 3 3 : 8) . Thus Mau ss's be l i ef that the tota l ity
of soc i ety is i m p l icit in i ts parts or organ i z i n g structu res may appear as a
kind of ena b l i ng charter for a broad range of fieldwork tactics (ap­
proac hes to soc i a l representation in the rhetorical mode of synecdoc he),
w ithout w h i c h re l atively short-term professional fiel dwork wou l d be
q uestionable-pa rt icu l a r l y researc h a i m i ng at portraya l s of whole c u l ­
tures . S i nce o n e can not study everyth i ng a t once, o n e must b e a b l e to
h i gh l i ght pa rts or attack spec ific prob lems i n the confidence that they
evoke a wider context.
There is another side to tota l soc i a l facts : the i dea i s ambiguous and
fi n a l l y trou b l i ng . If i t . legiti mates partial cu l tu ra l descriptions, it offers no
guidance as to w h i c h code, key, or l u m i nous example is to be preferred .
L i ke N ietzsche's vision of i n fi n i te i n terpretations, Mau ss's idea sees soc i a l
rea l i ty and t h e mora l world as constructed i n m a n y poss i b l e ways, none
of which may be privi l eged . Modern eth nography took shape i n a shat­
tered world hau nted by n i h i l i s m , and Mauss in h i s portraya l s of the con­
stitution of col lective order was acutely aware of the poss i b i l ity of d i s­
order (see Chapter 4 ) . The Gift is an a l legory of reconc i l iation and
rec i proc ity in the wake of the F i rst World War. As i s we l l known, the war
had a d evastati ng i m pact on Mauss; its seq uel i n 1 940 wou l d deprive
h i m of the wi l l to work and t h i n k . With the breakdown of evo l ution ist
master na rratives, the re l ativist science of c u l t u re worked to reth i n k the
world as a d i s persed whole, com posed of d i sti nct, fu nction i ng, and i nter­
re lated c u l tu res. It reconstituted soc i a l and mora l who l eness p l u ra l l y. If
synecdoc h i c ethnography a rgued, in effect, that "cu l tu res" hold together,
it d i d so i n response to a pervasive modern fee l i ng, l i n k i n g the I rishman
Yeats to the N i gerian Achebe, that "th i ngs fa l l apart ."
For a com m i tted soc i a l i st l i ke Mauss, the study of soc iety was a
refusal of n i h i l i s m ; its constructions of soc i a l wholeness served mora l
a n d po l itical as wel l as scientific e n d s . But he was too c l ea r-sighted a n d

know l ed geable t o espouse any sovereign method for t h e constitution of

tota l ities. He contented h i mself w ith a kind of gay science; he was gen­
erous, rather than, l i ke N i etzsche, sardo n i c . He presen ted a generation
of eth nographers with a n aston i s h i n g reperto i re of obj ects for study and
ways to put the world together: eth nography was a d i pp i n g of d i fferent
nets in the teem i n g ocean , each catc h i ng its own sort of fis h . Schoo l ed
i n C us h i n g's work, he knew that the task of representi ng a culture was
potenti a l l y end less. "You say you have s pent two and a half years with
one tri be," he rema rked to Meyer Fortes. " Poor man, it w i l l ta ke you
twe nty years to write it u p " (Fortes 1 9 73 : 2 84) .
Mau ss's Manuel was not a methode but an enormous checkl ist; t h u s
one can not speak of a "Mauss i a n " as one can o f a "Ma l i nows kian" or a
" Boas i a n " eth nography. (Th is fact may exp l a i n , i n part, why F rench field­
work has never assumed a d i st i nct identity and has, in effect, been i nv i s­
i b l e to anthropologi sts of other trad itions . ) H i s students d i ve rged ma rk­
ed l y. A l fred Metrau x pu rsued a d i st i ngu is hed career of Ameri can-style
pa rtici pant observation . Michel Le i ris, wh i l e making original contr i bu ­
tions t o Doga n and Eth iopian ethnography, never stopped questi o n i n g
t h e s u bj ective confl i cts and po l itical constra i nts of c ross-cu ltural study as
suc h . Maurice Lee n h a rdt, whose l ate entry i nto the U n ivers i ty of Paris
was m u c h encou raged by Mauss, represented a n older style of research
whose authority was rooted in years of m ission ary work rather than i n
academ i c trai n i n g . C harles LeCoeu r, who attended Ma l i nows k i 's sem i n ar
at the London School of Econom ics, l ived among the Teda, l earned their
langu age, and forma l l y, at least, conducted fie l dwork a I'anglais. Of
Mau ss's other students-vi rtua l l y every m ajor French ethnographer be­
fore 1 95 0 --o n l y G ri a u l e developed a systematic method and a d i st i nct
trad ition of resea rc h .

Two loose m etaphoric structu res govern G riau le's conception of field­
work: a documen tary system (governed by i mages of col lection, obser­
vat i o n , and i nterrogation) and an initiatory com p l ex ( i n wh i c h d i a logi cal
processes of ed u cati o n and exegesis come to the fore) . G r i a u l e h i m se l f
presented the two approaches as com p l ementary, each req u i ri n g a n d
bu i l d i n g on t h e other. O n e ca n , however, d i scern a s h i ft from t h e docu­
mentary to the i n itiatory as his career progressed and as his person a l
i nvolvement w i t h Dogan modes o f thought a n d be l ief deepened . F o r the
sake of a n a l ytic c l a rity I s ha l l consider the a pproac hes separate l y. It

Marcel Griaule developing photographic plates. Sanga,

October-November 1 93 1 .

shou l d be u nderstood, however, that both are attem pts to account for a
com p l i cated , evo l v i ng ethnographic experience-an experience tra­
versed by i n fluences, h i storica l and i n tersubjective, beyond the contro l
of G ri a u l e's metaphors .
T h e notion that eth nography was a process o f col lection dom i nated
the M i ssion Dakar- Dj i bouti, with its m u seographical emphasis. The eth­
nogra p h i c object-be it a tool , statue, or mask-was understood to be a
pec u l iarly re l i able "witness" to the truth of an a l ien soc iety. The Maussian
ratio n a l e i s evident i n a set of " I n structions for Co l lectors" d i stributed by
the m i ssion .

Beca use of the need that has always driven men to i mprint the
traces of their activity on matter, nearly all phenomena of col lective

l ife are capable of expression i n given objects. A col lection of objects

systematically acq u i red is thus a rich gathering of admissible evidence
[p ieces a conviction] . The i r col lection creates archives more revea l i n g
a n d sure t h a n written a rc h i ves, since these are authentic, autonomous
objects that can not have been fabricated for the needs of the case [/es
besoins de Ia cause] and that thus characterize types of civi l izations
better than anyth i ng e l se. (Mauss 1 93 1 : 6-7)

" Dead ," decontextua l i zed objects, the broch u re goes on to argue, can
be restored to " l i fe" by s u rro u n d i n g "docu mentation" (descri ptions,
d raw i n gs, photos). The l i n ks ty i n g any object or i nsti tution to the "en­
sem b l e of soc iety" can th u s be reconstituted and the truth of the whole
e l ic i ted sc ientifica l l y from any one of its pa rts .
The rec u r r i n g j u ridical metaphors ( pieces a conviction, besoins de Ia
ca use) are revea l i ng ; if a l l the pa rts of a c u lture can i n pri nc i p l e be made
to y i e l d the whole, what j ustifies a n eth nographer's parti c u l a r selection
of revea l i ng "evidence" ? Some "witnesses" m ust be more re l i ab l e than
others . A coro l l ary of the va l ue p l aced o n objects as "authentic and au­
tonomous," not "fabri cated for the needs of the case," i s the ass u m ption
that other forms of evidence, the "arc h i ves" com posed on the bas i s of
personal observati o n , desc r i ption, and i n terpretatio n , a re l ess p u re, more
i nfected with the conti n gent eth nogra p h i c encounter, its c l ash of i nter­
ests, and part i a l truth s . For G r i a u l e fie ldwork was a perpetual strugg l e for
con trol ( i n the pol itical and scientifi c senses) of th i s enco u n ter.
G r i a u l e assu med that the oppos i n g i nte rests of ethnographer and na­
tive cou l d never be ent i re l y harmon i zed . Re l ations someti mes romanti­
c i zed by the term rapport were rea l l y negoti ated settlements, outcomes
of a cont i n uous push and p u l l determ i n i ng what cou l d and cou l d not be
known of the soc iety u nder study. The outsider was a l ways in danger of
l os i n g the i n i tiative, of acq u iesc i n g in a su perfi c i a l mod u s v i vend i . One
cou l d not learn what was systematica l l y h idden in a cu l tu re s i m p l y by
becom i ng a tem porary mem ber of a common mora l com m u n i ty. I t cou l d
be reveal ed o n l y by a k i n d of v i o lence : t h e ethnographer m ust keep u p
t h e p ressu re ( G r i a u l e 1 9 5 7 : 1 4) . G ri a u l e m a y h ave h a d no choice : i n S u ­
danese soc ieties, with the i r l o n g processes of i n itiation, one had either
to force the reve l ation of occ u l t trad itions o r to be on the scene for dec­
ades .
Of a l l the poss i b l e ave n u es to h idden truths the l east rel i a b l e was
speec h-what i nformants actu a l l y said in response to q uestions. T h i s

was due not mere l y to consc ious l y i ng and res i stance to i n q u i ry ; i t fol ­
l owed from d ramatistic ass u m ptions that were a le itmotif o f h i s work. For
G riau l e every i nformant's self-presentation (a long with that of the eth nog­
ra pher) was a d ramatization, a putt i n g forward of certai n truths and a
hold i ng back of others . I n penetrati n g these consc i ous or u nconscious
d i sgu ises the fieldworker h ad to exploit whatever adva ntages, whatever
sou rces of power, whatever knowledge not based on i nterl ocution he or
she cou l d acq u i re ( 1 9 5 7 : 9 2 ) .
G riau l e l ooked i n itia l l y t o v i s u a l o bservation as a sou rce o f i n for­
mation that cou ld be obta i ned without depend i n g on u ncerta i n ora l co l ­
laboration a n d cou ld provide t h e edge needed t o provoke, contro l , and
verify confessional d i scou rses . Accustomed to actu a l l y l ook i n g down on
thi ngs (his fi rst job in the air force had been that of an aeri a l spotter and
navigator), G r i a u l e was particu l arly consc ious of the advantages of over­
view, of the prec ise mappi n g of habitats and thei r su rrou nd i n g terra i n .
T h i s visual preocc u pation, apparent i n a l l h i s methodologica l works,
emerges with d i sconcert i n g c l arity i n Les Sao Jegendaires, h i s popu lar
account of eth nograph i c and archaeo logical work i n Chad ( 1 943 ) :

Perhaps it's a q u i rk acq u i red i n m i l itary ai rcraft, but I always re­

sent hav i ng to explore an u n known terrai n on foot. Seen from high i n
the a i r, a d i strict holds few secrets . Property i s del i neated a s if i n India
ink; paths converge i n critical poi nts; i nterior cou rtyards yield them­
selves up; the i n habited j u mble comes clear. With an aerial photo­
graph the com ponents of i nstitutions fal l i nto place as a series of th i ngs
d i sassembled, and yielding. Man is s i l ly: he suspects h i s neigh bor,
never the sky; i n s ide the fou r wal l s, pa l i sades, fences, or hedges of an
enclosed space he th i n ks all is permitted . But all his great and sma l l
i ntentions, h i s sanctuaries, h i s garbage, h i s careless repa i rs, his ambi­
tions for growth appea r on an aerial photograph . I n a v i l l age I know i n
t h e French S u d a n , I recal l havi ng d iscovered fou r i m portant sanctu­
aries at the cost of much hard land travel , along with platitudes, flat­
tery, payoffs, and u n redeemable promises. Seventeen sanctuaries ap­
peared o n an aerial photo thanks to the m i l let pulp spread out on their
domes. A l l at once the open ness of my informants increased to an
unbe l ievable degree. With an a i rplane, one fixed the u nderlying struc­
tu re both of topography and of m i nds. (pp. 6 1 -62)

I t i s not c lear whether th i s passage sho u l d be read as enth usiastic publ ic­
ity for a new scientific method (on aeri a l photography see G riau l e 1 9 3 7)
P O W E R A N D D I A L O G U E I N ETH N O G R A P H Y 69

Marcel Griaule photographing from cliff top near Sanga,

October-November 1 93 1 . Andre Schaelfner holds him by
the ankles.

o r as a somewhat d istu rb i n g fantasy of observati o n a l power. G ri a u l e sel­

dom had a n a i rp l a n e at h i s d i sposa l in the fie l d , but he adopted its pan­
optic viewpo i n t as a habit and a tact i c .
The s i m p l e fact o f d raw i n g u p a map cou l d g i ve an overvi ew and
i n i tial mastery of the c u l tu re i nscribed on the l a nd . Recounting the ex­
cavation of a n c i ent fu neral rem a i n s aga i nst the wi shes of local i n habi­
ta nts who considered the graves to be ancestra l , Griaule prov ides a n
extraord i na ry phenomeno l ogy o f t h e wh ite outsider's strugg l e to ma i nta i n
a n edge i n d ea l i ngs w i t h t h e native counc i l o f e l ders. B ecause thei r ora l
trad ition i s a key sou rce of i n formation for where exactl y to d ig, they
m u st be i nd uced to ta l k ( 1 94 3 : 5 8) . G r i a u l e is a l ive to a l l man ner of signs
in behavior and espec i a l l y in the terra i n that may eventu a l l y serve as
70 D I S C O UR S E S

entrees i n to the h idden world of custo m . H i s q uestions a i m to provoke

and confuse, to el i c i t u nguarded responses . H av i n g a rduously mapped
the l andhold i n g and habitations of the region, he is able to pose u n ­
expected l y acute queries about i n congruous sites that are i n fact sa­
c red- a l ta rs, a stra nge door in a wa l l , a cu rious topograph i c featu re­
traces of sec rets written on the s u rface of the habitat. The map-making
outs ider holds a d i sconcert i ng authority : he seems a l ready to know
where everyth i ng is. Reve l ations fo l l ow. New sites are excavated .
For G ri a u l e a map is not o n l y a p l a n of work but "a base for com bat"
where "every i nscribed pos ition is a conq uered pos ition" ( 1 943 : 66).
Throughout his account he i s consc ious of the aggressive, d i sruptive
power of the gaze . I nvestigation, l ook i n g i n to someth i ng, is never neu­
tra l . The researchers fee l themse lves u nder s u rve i l l a nce : " H u nd reds of
eyes fo l low u s . We' re i n fu l l view of the v i l l age ; i n every crack i n the
wa l l , be h i n d every granary, a n eye i s attentive" (p. 64) . In opposition
stands the i r sc i entific observati on : "To d i g a hole i s to com m it an i n d i s­
creti o n , to open an eye onto the past" (p. 68) . Every i n q uest is "a s i ege
to be o rga n i zed" ( p . 60) . T h i s pa rticu l a r war of gazes ends with a nom i n a l
truce, a compro m i se perm itti ng t h e col lection o f certa i n a rtifacts wh i l e a
few espec i a l l y sacred ones are spared (p. 76) . The theatri cal tug of war
actu a l l y ends with a n a rran gement enti re l y to the advantage of the out­
siders, who a re able to com plete the i r excavation, remove n u merous
rel ics, and esta b l i s h ground ru les for l ater intensive eth nography.
For G r i a u l e the exhaustive doc u mentation of a cu lture was a pre­
cond ition for p l u m b i n g i ts "secrets" th rough long-term, contro l led i n ter­
rogation of i nformants . He did not, of cou rse, be l ieve that com plete de­
scri ption was poss i b l e ; but often-espec i a l l y when defend i n g h i s
practice o f tea mwork aga i nst the Anglo-American model o f i nd ividual
part i c ipant observation-he wou l d betray panopti cal aspi rations. H i s fa­
vorite example was the problem of descr i b i ng a Dogan fu neral cere­
mony, a spectacle i n vo lvi ng h u nd reds of part i c i pants. An ind ivid u a l
part i c i pa nt-observer wou l d be l ost i n t h e melee, jott i n g down more or
less arbi tra ry i m press ions, and with l ittle grasp of the whole.
G r i a u le a rgues that the o n l y way to document such an event ade­
q uate l y is to deploy a tea m of observers. He offers, characteristica l ly, a
map of the performance s i te and a set of tactics for i ts coverage, pro­
ceed i ng rather in the manner of a modern te levision crew report i ng on
a n American po l i tical convention ( 1 9 3 3 : 1 1 ; 1 95 7 : 4 7-5 2 ) . Observer
n u m ber one is stationed atop a c l iff not far from the vi l l age square with

the job of photogra p h i n g and noti n g the l a rge-scale movements of the

rite; n u m ber two i s among the menstruati n g women to one side; th ree
m i xes with a band of you n g torc h bea rers ; four observes the gro u p of
m u s i c i a n s ; five is on the roof tops "charged with survei l lance i n the w i ngs
with the i r thousand i n d i sc retions, and go i n g freq uent l y, along with n u m ­
b e r s i x , t o t h e dead man's house i n searc h o f t h e l atest news" ( 1 9 5 7 : 49 ) .
N u m be r seven observes the reactions o f t h e women and c h i l d ren t o the
masked dances and ritual com bats tak i n g p l ace at center stage. A l l o b­
servers note the exact ti mes of their o bservations so that a synthetic por­
tra i t of the ritual can be constructed .
T h i s o n l y begi ns the task of adeq u ate docu mentation . The synoptic
outl i ne thus constructed w i l l l ater be augmented and corrected by pro­
cesses of "verification" and "commenta ry." Witnesses m u st be asked for
the i r exp l a n ations of obsc u re gestures. " H o l es" i n the fabric w i l l be fi l l ed
i n , i nc l u d i ng those that a re d u e to cont i ngenc ies of a spec ific perform­
a nce-the absence or presence of part i c u l a r groups or i n d i v i d u a l s, the
forgetfu l ness of the acto rs, or any d ivergences from the rite's " idea l har­
mony" ( 1 9 5 7 : 5 0 ) . S l ow l y, over a n u m ber of years, b u i l d i ng on repeat
performa nces if poss i b le, an idea l type of the rite wi l l be l aboriously
constructed . But th i s enormous "dossier" spi l l s out i n many d i rections,
and "each pa rt of the observation becomes the core of a n enq u i ry that
sooner or l ater w i l l fu rn i s h a vast network of i nformation" (p. 5 1 ) .
Griau le's Methode de /'ethnograph ie, from which th i s account i s
d rawn, p rovides a rat iona l i zed vers ion of h i s o w n research practice. It is
often u n c l ea r whether the methods propou nded are those G r i a u l e actu­
a l l y u sed or ideal recom mendations based on a rather more messy ex­
perience. B u t the Methode gives a good sense of the overa l l ass u m ptions
and pa rameters of his fie l dwork. I n Sanga the M i ssion Dakar- Dj i bouti
had in fact encounte red a Dogan fu nera l , a d ramatic, confu s i n g rite fea­
t u r i n g s pectac u l ar performances by masked dancers . G r i a u l e set about
its doc u mentatio n : h i s su bseq uent work wou l d center on the sec ret so­
c i ety of masks, and various of h i s co-workers contributed rel ated stud ies
(Lei ris 1 94 8 ; De Ganay 1 94 1 ; D ieterlen 1 94 1 ) By d i nt of repeated visits

and i nten s i ve col labo rati ve work a n orga n i zed corpus of "docu ments"
was bu i l t u p .
G r i a u le's foc us o n t h e i nstitution o f masks d id n o t i nvo lve a synec­
doc h i c representati o n of c u lture as a whole in the functi onal ist trad ition
(using the mask soc i ety as either a n idea l -typ ical " i nstitution" or its ritua l s
a s "tota l i z i n g c u l tu ra l performa nces " ) . Rather, work i ng o u t from th i s

dense c l u ster of tota l soc i a l facts, he and h i s associ ates constructed a

"vasi network of informati on" as a context and contro l for what natives
themse l ves said about the i r c u lture . I n i tia l l y, in h i s "documentary" phase,
Griaule used the ex p l i cations of i nforma nts as commentaries on observed
beh avior and co l l ected artifacts; but th is attitude wou l d change, espe­
c i a l l y after Ogotemmel i : once properly tested and qual ified , i nformants
cou l d be trusted w ith research tasks. With proper contro l they co u l d be­
come regu lar auxi l iaries and, in effect, members of the team . The net­
work of observation and doc u mentation cou l d th us be d ramatica l l y ex­
tended ( G r i a u l e 1 9 5 7 : 6 1 - 64). Teamwork was an efficient way to deal
with tota l soc i a l facts, to prod uce a fu l l documentation on a m u lt i p l icity
of s u bjects treated i n d i verse man ners .
A s conceived b y G r i a u l e the team was much more than a makes h i ft
co l l aboration of i n d ividua l s . It embod ied the principle u nderl y i ng a l l
modern i nq u i ry : s pec i a l i zation a n d the d ivision of l abor. Because soc i a l
rea l ity is too com p l ex for t h e s i ng l e researcher, he must " rely on other
spec i a l ists and try to form with them a th i n k ing group, an element of
com bat, a tactica l u n it of resea rch in wh ich each person, wh i le hold i n g
t o h i s o w n person a l q u a l it ies, knows he i s an i ntel l i gent cog o f a mac h i ne
i n which he is i n d i spensable but without which he is noth i n g" ( 1 9 5 7 : 2 6) .
Some o f G riau le's early co-workers, l i ke Leiris, Schaeffner, a n d Pau l me,
d i d not fi nd end u r i n g p l aces with i n this prod uctive mechan i sm . Le i r i s's
scandalous L Afriq ue fan tome ( 1 934) was a c l ear breach of d i sci p l i ne .

But othe rs ( De Ganay, D i eterlen, Lebeuf, a n d Ca l ame-G riau le), if not

prec isely " i n te l l igent cogs," worked free l y with i n the developing para­
d i gm . G ri a u l e spoke of h i s ideal team in terms of o rga n i c sol ida rity and
a quasi-m i l itary esprit d e corps, and the works of the schoo l do suggest
an effic ient co l l aborative enterpr i se ; but as a prod uctive mechanism the
"team" cou l d never be tightly control led . When one i n c l udes as active
agents the Dogo n i nformants, transl ato rs, and tribal authorities, whose
i nfl uence on the content and t i m i ng of the knowledge ga i ned was cru­
c i a l , it becomes appa rent that the co l l aborative documentary experience
i n itiated by G r i a u l e in 1 93 2 had by the 1 9 5 0s undergone a metamor­
phos i s .
H o w, before Ogotem mel i , d i d G riau l e "choose," " identify," " i nter­
rogate," and "uti l ize" i nformants ( 1 9 5 2 c : 542-5 4 7 ; 1 9 5 7 : 54 - 6 1 )? H i s
method o l ogica l stri ctu res are particu l a r l y revea l i ng s i nce, a s h i s respect
for African ora l trad ition grew, he came i ncreasingly to center h i s re­
search in c l ose work with a l i m ited n u m ber of col/aborateurs indigenes.

T h e i nformant m u st fi rst be ca refu l l y identified and l ocated i n a spec ific

gro u p or set of gro u ps with i n the soc i a l fabric. In th i s way one can a l l ow
for exaggerations and for om issions rel ated to gro u p l oya lty, taboos, and
so on. He o r she- i n fact G ri a u l e's i nformants, as he regretfu l l y noted ,
were a l most enti re l y men ( 1 95 7 : 1 5)- has to be q u a l i fied to p ronou nce
on particu l a r s u bjects, whether tec h n o l ogica l , h i storica l , lega l , or rel i­
gious. H i s "mora l q u a l it ies" a re to be assessed : s i ncerity, good fa ith,
memory. A l though many of his i nformants were s i gn i ficantly i nfl uenced
by "outs ide" perspectives ( Lettens 1 9 7 1 : 5 2 0 -5 3 5 ) , G r i a u l e we i ghed
heavi l y the attac h ment to trad ition, m i strust i n g Ch ristians, M u s l i ms, and
i nd i v i d u a l s with too much prior contact with wh ites ( 1 95 7 : 5 7 ) .
Every i nformant, G ri a u l e assu mes, e n u nc iates a d i fferent k i nd of
truth, and the eth nographer m u st be constantly a l i ve to its l i m i tations,
strengths, and wea knesses . In his Methode he d i scusses various types of
" l i a rs ." I n deed throughout his work he i s preoccu p i ed with l ies-al­
though n ot as simple u ntruths . Each i nformant, even the most s i ncere,
experiences a n " i n st i nctive need to d i ss i m u l ate part i c u l arly del icate
poi nts . He w i l l g l ad l y ta ke advantage of the s l i ghtest cha nce to escape
the s u bject a n d dwe l l on a nother" ( 1 95 7 : 5 8) . Native co l l aborators " l ie"
in j est o r t h rough ven a l ity, the des i re to pl ease, or the fear of neighbo rs
and the god s ( p . 5 6 ) . Forgetfu l i nformants and Europea n i zed i n formants
are part ic u l a r l y dangerous types of " l i a rs ." In a n eth nograph ic "strategic
operatio n " ( p . 5 9 ) the i n vesti gator m u st brea k through i n it i a l defenses
and d i ssi m u l at i o n s . Often a n i nd i v i d u a l i nformant m ust be isol ated for
i nten s ive q u estion i n g so as to remove i n h i bi t i n g soc i a l pressu re ( p . 60) .
When the i r testimony is confronted with d iffering vers ions ga i ned from
other i n terviews, h a rd-pressed i nformants e n u n c i ate truths they had not
i ntended to revea l . On one occasion G r i a u l e perm i ts h i m se l f to d ream of
a n " idea l " s ituation : "an i nfin i ty of separated i n fo rmants" ( 1 943 : 6 2 ) . Yet
it may sometimes be p rofita b l e to p u rsue i nq u i ri es i n p u b l i c , espec i a l l y
ove r del icate p roblems s u c h a s l a n d ten u re, where t h e researc her can
provoke reveal i n g d i s putes w i th thei r i nevita b l e i n d i scretions ( 1 943 : 66-
68; 1 9 5 7 : 60 ) .
G ri a u le's tactics a re varied, but they h ave i n common an active,
aggress i ve postu re not u n l i ke the j u d i c i a l process of i nterrogat ion
( 1 9 5 2 : 5 4 2 , 547) : "The ro l e of the person sn iffing out soc i a l facts i s ofte n
com parable to that of a d etecti ve or exam i n i ng magi strate . The fact is the
cri me, the i n terlocutor the gu i lty party; a l l the soci ety's mem bers a re ac­
com p l i ces" ( 1 9 5 7 : 5 9 ) . He is fasc i nated by the tactics of ora l i n q u i ry, the

p l ay of truth and fa l sehood that can lead i nto " l abyri nths" that are "or­
gan ized ." L i ke a psyc hoanal yst, he begins to see patterns of res istance,
forgetfu l ness, and o m i ss i o n not as mere obstacles but as signs of a deeper
structu ring of the truth :

The informant, on fi rst contact, seldom offers much resistance. He lets

h i mself be backed i nto positions he has been able to organize in the
course of feel i ng out the situation, observing the qui rks, ski l ls, and
awkward nesses of h i s i nterlocutor. The va lue of these posi tions de­
pends on what he can make of them ; he resists as best he can . And if
they are taken by force? After other similar res istances he wi l l retreat
to a final position that depends neither on h i mself nor on h i s "adver­
sary" but on the system of proh i bitions of custom . ( 1 952c: S9-60)

For G ri a u l e the deep structu re of resi stances is not spec ific to an i ntersub­
jective encounter but derives from a genera l sou rce, the ru l es of "cus­
tom ." T h i s hypostatized entity is the l ast bastion to be stormed . As we
sha l l see, it can not be conquered by frontal assa u l t, by the tactical pro­
cesses of observation, docu mentation, or i n terrogation . A d i fferent " i n i ­
ti atory" process must come i nto p l ay.
Des igned for begi n n i ng fie l dworkers, G r i a u le's treati ses on eth no­
graphic tec h n ique rem a i n l a rgel y with i n the "docu mentary" parad i gm .
Moreover, G r i a u l e probably d i d not have time to d i gest fu l l y the meth­
odo logical con seq uences of Ogotemme l i 's reve l ations or of the gather i n g
c r i t i q u e o f colon i a l know l edge i n t h e decade before Methode w a s pub­
l i shed . It is proba b l y best to read t h i s rather mechan i stic com pend i u m of
tec h n i q ues as a l ess-than-successfu l attempt to contro l an u n r u l y re­
search process, i n Geo rges Devereux's terms ( 1 967), a passage from anx­
iety to method . G r i a u l e's u lti mate complex rec i procal i n vo l vement with
the Dogan i s hard l y captu red in section titles such as "The Detect ion and
Observation of H u ma n Facts" or in the portrayal of eth nographers and
i nd i genous co l l aborators as b u i lders of i nformation networks, co l l ectors
of "documents," com p i l ers of "doss iers." Eth nography, in G r i a u le's j u rid­
ical l a nguage, i s sti l l aki n to the process of instruction- i n F rench law,
the pre l i m i nary estab l i shment of the facts of a case before the jugement
proper ( 1 95 7 : 5 1 ) . Worki n g among i n terested parties the ethnographer
uses the far-reac h i ng powers of the juge d'instruction (one of G r i a u l e's
favorite metaphors) to smoke out the truth (cf. Ehrmann 1 9 76). Genera l l y
respecti ng the division o f l abor l a i d down by Mauss, and suspicious of
abstractions and systematic cross-cu ltural comparison, G r i a u l e leaves

matters o f theory a n d exp l a n ation t o others outs ide t h e fray. T h e juge

d'instruction, having co l lected enough rel i able docu ments and havi n g
cross-checked h i s witnesses' versions o f t h e facts, has i n h i s possession
everyth i n g h e needs to d eterm i ne the truth .
By 1 95 0 these attitudes toward observation and i n terrogati o n were
beco m i n g genera l l y suspect, and G r i a u le's earl y documentary metaphor
was no longer adeq uate to a research process that was tak i n g o n a l i fe of
its own . G rad u a l l y G ri a u le's u nderstand i n g of the Dogon was becom i n g
i n d i sti n g u i s h a b l e from the i r i n c reas i ngly e l a borate exp l i cations. The
ori g i n a l ity of the ethnograph i c activ ity he set in motion was that i t u n ­
covered-and t o a n u nd eterm i ned extent provoked -a sop h i sti cated i n ­
terpretation o f thei r c u lture b y a gro u p o f i nfl uent i a l Dogo n .

Before we consider the second p h ase o f G ri a u l e's work, i t i s worth step­

p i n g back for a moment from h i s researc h sty les and tactics to s uggest
the i r re lation to the colon i a l situatio n . Griau le provides us with a k i n d of
d ramatu rgy of eth nograph i c experience before the fifties. I n an extraor­
d i nary passage- i nc l uded in both h i s early and h i s l ate d i scussions of
methodo l ogy- h e evokes the gamut of power-l aden ro l es adopted by an
eth n ographer e l i c i t i n g i nformation from a n i nformant. Ethnographie ac­
t ive, h e writes, i s "the a rt of bei n g a m idw i fe and an exam i n i ng magis­
trate" :

By tu rns an affable comrade of the person put to cross-exam i nation , a

d i stant friend, a severe stranger, compassionate father, a concerned
patro n ; a trader paying for revelations one by one, a l i stener affecti n g
d i straction before t h e open gates o f t h e most dangerous mysteries, an
obliging friend showi ng l ivel y i n terest for the most insi pid fam i l y sto­
ries-the eth nographer parades across h i s face as pretty a col l ection
of masks as that possessed by any m useu m . ( 1 93 3 : 1 0; 1 95 2 c : 5 4 7 ;
1 95 7 : 59)

The passage evokes a theme i nfu s i n g a l l of G r i a u l e's work-that eth­

nography i s a theatrical u n derta k i n g . His d ramatu rgy does not, however,
i n c l ude a ro l e popu l a r among fie l dworkers i n the Anglo-American trad i­
tio n : the person a of the earnest l earner, often cast as a ch i l d i n the pro­
cess of acq u i ri ng, of be i n g taught a d u l t knowledge . Perhaps t h i s persona
did not occ u r to G ri a u l e beca u se, seconded by i nterpreters and E u ropean
co-workers, he never actua l l y experienced the pos ition of bei n g a stam-

merer, h e l p l ess i n an a l ien c u l tu re . I t was o n l y after 1 9 50, l ate i n h i s

ca reer, that he began t o adopt t h e sta ndo i nt o f a student with respect to
Dogan cu lture; but th i s ro l e was a lways m i xed with the less vu l nerable
authority of i n itiate, spokesma n , and exegete. At least i n h i s writi ngs,
G riau le never abandoned a basi c confidence, a sense of u lti mate control
over the research and its products . But m a i nta i n i ng control was always a
battle, at best a j o k i n g re latio n . G riau l e never presented fieldwork as an
i n nocent atta i n ment of rapport a n a l ogous to fr iends h i p . Nor d i d he neu­
tra l i ze the process as a n experience of ed ucation or growth (ch i l d or
ado l escent becom i n g ad u l t) o r as acceptance i nto an extended fam i l y (a
k i n sh i p ro l e given to the ethnographer) . Rather, his accounts assumed a
rec u rring confl ict of i n terests, an ago n i stic d rama resu l t i n g i n mutual re­
spect, com p l i c i ty i n a p rod uctive bal ance of power.
G r i a u le's writ i n gs are u n usual i n the i r sharp awareness of a structura l
power d i fferenti a l a n d a s u bstratum of vio lence underlying a l l re lations
between w h i tes and b l acks i n a co l o n i a l s i tuation. For example i n Les
flambeurs d'hommes, an adventu re story Griau l e ca l led "an objective
desc r i ption of certa i n epi sodes from my fi rst trip to Abyss i n ia" ( 1 934a :vi),
he coo l l y notes a "given" of colon i a l l ife : the mem bers of his caravan
hav i n g shown themse l ves re l uctant to attempt a tricky ford i n g of the N i l e,
"there fo l l owed b l ows, given by the White Man and not retu rned ; for a
White is a l ways a man of the govern ment, and if you touch h i m com p l i ­
cations ensue" ( p p . 7- 8 ) . A revea l i ng sty l i stic device i s deployed here,
as e l sewhere in G r i a u l e's accou nts of fieldwork ( 1 948a) : a use of the
pass ive voice and of generic terms for h i m se lf- "the White Man," "the
E u ropea n ," "the Trave ler," "the Nazarite," "the Foreigner." The story of
the beati n gs suggests an automatic series of events to which a l l parties
acq u iesce . A E u ropean i n Africa cannot, shou l d not, avoid the pasts re­
served for h i m . G riau l e does not th i n k of e l ud i n g the priv i l eges and con­
stra i nts of h i s ascri bed status-a d ream that obsesses, and to a degree
para l yzes, M i c h e l Lei ris, h i s co l league of the M i ssion Dakar-Dj i bouti .
Lei ri s' fie l d journal ( 1 934) and h i s l ater writi ngs, both ethno logical and
l i terary, portray a s l ow reconcil i ation with a theatrical conception of the
self; but his acceptance i s a l ways ambivalent, i n creative confl ict with a
des i re for i m med i ate contact and part i c i pation (see Chapter 6 ) . Griau l e,
by contrast, harbors no q u a l m s about h i s own theatri ca l i ty. Once this i s
p l a i n , puzz l i n g aspects o f h i s practice become c learer-for example h i s
idea l "coverage" o f t h e Dogan fu nera l .
G r i a u l e's e l a borate panoptic p l a n w i l l ra i se the hackles of any eth-

nogra pher schoo l ed in part i c i pant observation . The crew he envi sages
m u st necessari l y d i st u rb and perhaps orient the cou rse of the ceremony,
but th i s does not seem to concern G r i a u l e . Does he n a i vely i magi ne that
seven observers w i l l n ot exert a cons iderable i nfluence? The question i s
beside t h e poi nt, for G r i a u le never thought o f bei ng a n unobtrusive par­
tici pant. H i s researc h was m a n i fest l y an i ntrusion; he made no pretense
that it was otherwise. Thus, to an i m portant degree the truth he recorded
was a truth p rovoked by ethnography. One is tem pted to speak of an
ethnographie verite a n a l ogou s to the cinema verite p i oneered by
G ri a u l e's l ater assoc i ate Jean Rouch-not a rea l ity objective l y recorded
by the camera but one p rovoked by its active presence (Rouch 1 9 78a) .
One suspects that G ri a u l e saw cu ltu re itse lf, l i ke perso n a l i ty, as a
performance or a spectac l e . I n the years fol lowi ng the Dakar- Dj i bouti
m i ssion G r i a u l e and his teams tu rned up every yea r o r so at Sanga . The
arrival of these i nc reasi n g l y fam i l i a r outsiders was a d ramatic event. Ti me
was of the essence; i nformants we re mobi l i zed , ritua l s were acted for the
cameras, and as much Dogan l ife as poss i b l e was recorded . I n fact
G ri a u le's early research tended to concentrate on aspects of cu ltura l l i fe
su scepti b l e to demonstration and performance: masks, publ ic ritu a l s ,
and games . It i s s i gn ificant i n th i s regard that Sanga, the Dagon com­
m u n ity m ost accustomed to eth nography, i s today the region's pri n c i pa l
tou r i st center, routi n e l y perform i ng its dances for outsiders ( I m perato
1 9 7 8 : 7-3 2 ) .
G r i a u le's penchant for t h e d ra m atic i n fuses h i s work; for the h i stor­
i a n th i s poses p roblems of i nte rpretation . For exam p l e a heightened but
c h a racteristic passage in Les Sao legendaires exu lts in a b reakth rough .
H av i n g maneuvered n ative i nter l ocutors i nto givi ng u p i nformation they
h ad not i ntended to d i vu l ge, G ri a u l e conte m p l ates the prom i se of futu re
work i n the area :

We wou l d be able to make asses of the old hesitators, to confound the

traitors, abom i nate the s i l ent. We were goi n g to see mysteries leap l i ke
repti les from the mouths of the neatly caught l iars. We wou l d play with
the vict i m ; we wou l d rub his nose i n his word s . We'd make him smi le,
spit u p the truth , and we'd turn out of his pockets the last secret pol ­
i shed b y t h e centu ries, a secret t o make h i m w h o h a s spoken it blanch
with fear. (Griaule 1 943 : 74)

How i s one to read such a passage ? G r i a u l e a l ways l i ked to provoke : a

passage written to shock i n 1 943 i s sti l l shoc k i n g and puzz l i ng . I n the

narrative to w h i c h it i s a kind of c l i max, one watches with d i scomfort

and w i th grow i n g anger as the ethnographer bu l l ies, caj o les, and man i p­
u l ates those whose resista nce i nterferes with h i s i n q u i ry, natives who do
not wish to see the i r a ncestra l rem a i n s co l lected in the i n terests of a
foreign sc ience. But G ri a u l e wi l l not perm it us to d is m i ss h i m out of
hand . If we now perceive such attitudes and acts as an embarrassment,
it i s thanks to G r i a u l e that we see them so c l early. He rubs our nose i n
them .
Beca use G r i a u l e p l ayed co l o n i a l ro les with gusto and with a certai n
i rony, the words j u st quoted can not be p l aced neatl y i n their h i storical
context and d i s m i ssed as attitudes u nfortunate l y poss i b l e i n the colon i a l
period . I t was more typical o f t h e period t o h ide s u c h vio l ence than to
bri n g it to the fore. Yet if the vio lence is, in some sense, G riau le's po i nt,
nowhere does he su ggest a criticism of forced confess ions in eth nogra­
phy. On the contra ry, h is methodologica l writi ngs give i n structi ons on
how to provoke them . Griau l e does not ex press serious second thoughts
about estab l i sh i n g dom i na nce, find i n g and exploiting the weakness, d i s­
u n ity, and confusion of h i s native hosts . Thus a h i storical read i n g of such
awkward passages can not u nderstand G r i a u l e as either a typical partici­
pant o r a self-conscious critic with i n the co l o n i a l s ituation . His pos ition
i s more com plex.
One i s tem pted to asc ri be such passages to G riau le's "sty le" - h i s
penchant for banter, for charged metaphors, for provocation; b u t th i s
mere l y ra ises t h e question of how a sty le fu nctions a s part o f a research
activity and how it p l ays aga i nst a n ideo l ogical m i l ieu. G riau l e's styl e is
not merely, as some have assumed, a faiblesse, a d i stract i n g and unfo r­
tunate dev i ation from the sc ientific busi ness at hand ( lettens 1 9 7 1 : 1 2 ,
49 1 ). It is rather a mea n i n gfu l response to a pred i cament, a set of ro les
and d i sc u rs i ve poss i b i l ities that may be cal led ethnographic liberalism.
A com plex, contentious debate on anthropo logy and em p i re has l a rgel y
esta b l i shed that eth nographers before t h e 1 950s acq u i esced i n colon i a l
regi mes (Lei ris 1 9 5 0 ; Asad 1 9 7 3 ; Copans 1 9 74) . Wh ite rule or c u l tu ra l
dom i nance was a given context for the i r work, a n d they adopted a range
of l i beral pos itions with i n it. Seldom "co lon i a l i sts" i n any d i rect, i nstru­
menta l sense, eth nographers nonethe l ess accepted particu lar constra i nts
w h i l e q uesti o n i n g them to vary i ng degrees . Th i s ambivalent pred i cament
i m posed certa i n ro les.
Griau le's sty le of ethnographic l i bera l ism may be u nderstood as both
a d ramatic performance a nd a mode of i rony. The most ac ute observers
of the colon i a l s i tuati on, Orwe l l and Conrad for example, have portrayed

i t as a power- l aden, a m b i guous world of d iscontin uous, c l ash ing rea l i­

ties . L i ke Orwe l l 's you ng d i strict officer who u nwi l l i ng l y shoots an ele­
phant to avo id bei ng l a u ghed at by a c rowd of B u rmese, and l i ke a l l the
cha racters in Heart of Darkness, d is p l aced E u ropeans m ust l abor to m a i n ­
ta i n the i r cu ltural identities, however artifi c i a l these m a y appear. Both
co l o n i a l and ethnogra p h i c s i tuations p rovoke the u n nerv i n g feel i ng of
bei n g o n stage, observed and out of p l ace. Part i c i pa nts in such m i l ieux
are caught in ro les they can not choose . We have seen G r i a u le's hei ght­
ened awareness of the masks worn as part of fie ldwork's c l ash of wi l l s ,
wits, b l u ffs, and strategies. H e i s n o t u n ique i n stressing the i m portance
of theatri ca l ity and i m press ion management in eth nography, the sense
that researc h rel ationsh i ps deve l o p "beh i nd many masks" ( Berreman
1 9 7 2 ) . Most eth nographers h ave, l i ke h i m , rej ected the pretense of goi n g
native, o f bei ng a b l e to s hed a fu ndamental Eu ropea n ness; but o n l y a
few have portrayed so clearly the tactical d i ss i m u l ations and i rred u c i b l e
violence o f eth nogra p h i c work (Rabi now 1 9 7 7 : 1 2 9-1 30).
U n l i ke Conrad , O rwe l l , or Lei ri s , Griaule seem s not to be oppressed
by h i s rol e p l ay i n g . B u t a l though he is not critica l , he i s i ro n i c . If he
com pares eth nography to a theater of war o r a j u d i c i a l proceed i ng, one
need not ass u m e that in the fie ld he acted cons i stently as a company
commander o r a n exam i n i n g magistrate . To take G ri a u le's metaphors at
face va l ue i s to m i ss thei r i m p l i c i t analytical fu nctio n . I t i s a l so to push
aside his other personae: his charm, his tem per, his p l ayfu l banter, h i s
grow i n g sym pathy, even love, for the Dago n .
Eth nogra p h i c l i bera l s , o f w h i c h there a re m a n y sorts, have tended
to be i ro n i c part i c i pants . They h ave sought ways to stand out or apart
from the i m perial ro les reserved for them as wh ites. There have been
freq uent variations on Del afosse's broussard. Many h ave in one way or
a nother p u b l i c l y identified themse l ves with exotic modes of l ife and
thought o r c u l tivated a n i m age of m a rg i n a l ity. G riau l e's exaggeration i s
another response. Eth nograph i c l i bera l ism i s a n a rray o f i ron i c positions,
ro les both w ith i n and at a certai n remove from the colon ial s i tu ation . Its
complete d ramatu rgy rem a i n s to be writte n .
T h e pol itica l a n d eth ical tensions v i s i b l e i n G r i a u l e's writi ngs have
o n l y recently become ex p l icit su bjects of a n a l ys i s . A penetrating para­
graph written in 1 968 by C l i ffo rd Geertz reflects the begi n n i n g of the end
of i n nocence in fieldwork:

Usua l l y the sense of being members, however temporari ly, i nsecu rely,
and i n completel y, of a s i ngle moral com m u n i ty, can be mai nta i ned

even i n the face of the wider social rea l i ties which press i n at al most
every moment to deny it. It is fiction-fiction, not falsehood-that l ies
at the very heart of successfu l anthropological field researc h ; and, be­
cause it is never completely convincing for any of the partici pants, it
renders such research , considered as a form of cond uct, contin uously
i ron ic. (p. 1 54)

By the l ate s i xties the romantic mytho l ogy of fieldwork rapport had begu n
to d i ssolve publ ic ly. S i nce then a growing reflexivity in eth nograph i c
thought and practice has deepened t h e recognition o f i t s i ro n i c structu re,
its rel i ance on i m prov i sed , h i storica l ly contingent fictions. Th i s new
awareness makes poss i b l e a read i n g of G r i a u l e that sees a theatrica l ,
i ro n i c stance a s centra l t o h i s eth nographic work.

Although G riau l e's sense of the moral tension and violence i n herent in
fieldwork was u n usual l y acute, he deve loped nonetheless an enab l i ng
fiction of rec i proca l encounter w ith the Dogo n . Th i s fiction, not fa l se­
hood , is most c learly em bod ied in the work after Ogotem me l i . I n
Griau l e's ongo i ng research (c l osely l i n ked with that o f Dieterlen) one
sees the overlay of an eth nograph ic fiction ( Dogon i n itiatory knowledge)
by a fiction of eth nography (fieldwork as i n itiation ) . To account for th i s
dou b l i ng we may return to Geertz's i ron i c fiction o f moral com m u n i ty,
w h i c h he sees as d iss i pating, temporari l y at least, the eth ical tensions
i n herent in fieldwork. Geertz u nderm i nes the myth of ethnograph ic rap­
port before re i n stating it in an i ronic mode. L i ke G r i a u l e he seems
to accept that a l l parties to the encou nter recogn i ze i ts elements of i n­
si ncerity, hypocri sy, and se lf-decepti on . He sees t h i s recogn ition as a pre­
condition for a l i ved fiction (a d rama in G r i a u l e's terms) that is in some
very guarded but real se nse gen u i ne . Just how this prod uctive compl icity
is actua l l y enacted is a l ways d ifficu l t to know; but if, as Geertz suggests,
such l i ved fictions are centra l to successfu l eth nographic resea rc h , then
we may ex pect to fi nd them reflected in the texts that orga n i ze, narrate,
and genera l l y account for the truths learned in fieldwork. In fact many
ethnograph ies i n c l ude some part i a l account of fieldwork as part of the i r
representation of a c u l tural rea l ity. But whether or n o t an ex pl icit or im­
p l icit fie ldwork narrative appea rs i n the eth nography, its very shape-the
defi n i tion of its top ic, the horizon of what it can represesnt- is a textua l
expression of t h e performed fiction o f com m u n ity that h a s made t h e re-

Marcel Griaule and Michel Leiris prepare to sacrifice

chickens before the Kono altar at Kemeni, September 6, 1 93 1 ,
as a condition of entering the sanctuary.

searc h poss i b l e . Thus, and w i th vary i n g degrees of exp l icitness, eth n og­
raph ies a re fictions both of another c u ltu ra l rea l ity and of the i r own mode
of prod uctio n . Th i s i s u n usua l ly c lear in the l ate work of Gria u l e and
D ieterlen , where i n itiation provides the common o rga n i z i n g metaphor.
To say that eth nogra phy is like i n itiation is not to recommend that
the researc her actu a l l y u ndergo the processes by which a native atta i n s
t h e w i sdom o f t h e gro u p . G ri a u l e has l ittle u s e for such a "comed ie"
( 1 9 5 2 c : S49). The metaphor of i n i t i ation evokes, rather, the deepen i n g of
u nderstand i n g that accrues to long-term field research with repeated v i s­
i ts throughout the anth ropologi st's career. It evokes too a q u a l itat ive
cha n ge i n eth nogra p h i c re l ati o n sh i ps occ u rr i n g as a c u l m i n ation of the

long, pers istant documentary process . I n i tiation fi na l l y gives access to a

privi leged stratu m of native understandi ng, someth i n g Griaule c l a i med
was "a demonstration, summary but com plete, of the fu nction i n g of a
soc iety." The eth nograher, rather than tryi ng to blend i nto the soc iety
u nder study, " p l ays h i s stranger's ro le." A frien d l y but determ i ned out­
sider, press i n g constantly aga i nst customary i nterdictions, the ethnogra­
pher comes to be seen as someone who, prec i sely because of h i s or her
exteriority with respect to native i n stitutions, i s u n l i kely to fa lsify the m .
" I f he i s t o receive i n structions a n d reve l ations t h a t are t h e eq u ivalent of,
and even s u perior to, those enjoyed by i n itiates, the researcher must re­
m a i n h i mself. He w i l l be ca refu l not to try to ga i n time by telescoping
the i nformati o n ; rather he wi l l fo l l ow steps para l lel to those of i n itiation
as it i s practiced by the men of the soc iety" (p. 548) .
The narrative of "para l l e l " (or spec i fica l l y ethnographic) i n itiation
appears pro m i nently i n Le renard pa le and Conversations with Ogotem­
meli. The fi rst decade of doc u mentary work at Sanga u nfo lded at the
lowest of fou r stages of Dogon i n itiatory knowledge. A l l the early ques­
tions of the G r i a u l e team were answered at a l evel of i nstruction offered
by e lders to beg i n ners-the parole de face. But the ethnographers re­
tu rned repeated l y. They p roved the i r good fa ith : G riau le, for example,
used his aeri a l photography to advise the Dogon on crucial questions of
water man agement. G rad u a l l y the pers i stent researc hers approached
deeper, secret leve l s of cu ltura l knowledge. Then, "the Dogon made a
decision" ( G r i a u l e and D i eterlen 1 965 : 5 4 ) . The l ocal patriarchs met and
dec i ded to i nstruct Griau le in /a parole cla ire - the h i ghest, most com­
p lete stage of i n i tiatory knowledge . Ogotemme l i wou ld beg i n the tas k.
Others cont i n u ed when he d ied shortly after h i s famous conversat ions
with G r i a u l e .
Taken a s a w h o l e th i s na rrat ive is certa i n l y too neat a n d patently sel f­
j u sti ficatory.4 But whether or not the "dec ision" by "the Dogon" was mo­
tivated i n j ust th i s way, and whatever the exact status of Ogotemmel i 's
d i scou rse ( i nd i vidual spec u l ation or c u l tu ra l knowledge), the overa l l i n i­
ti atory parad igm does ra i se i m porta nt q uestions about short- and l ong-

4. We need not go as far as Lettens ( 1 9 7 1 :509), who suggests that the entire
i n itiatory logic of progressively revealed secrets was an i nvention of Griau le's to
cover u p the fa i l u res of h i s fi rst phase of research i n the l ight of Ogotemmel i 's
revelations . Letten's extreme skepticism is largely u nsu bstantiated and uncon­
vincing, given widespread evidence for Sudanese i n itiatory systems, and given
h i s rather rigid and l itera l i st conception of i n itiatory processes.

term eth nography. There c a n b e no doubt t h a t G r i a u l e's repeated v i s i ts

res u l ted i n a progressive, q u a l i tati ve deepe n i n g of h i s understand i ng.
Open-ended long-term study may wel l yield resu l ts that d i ffe r i m por­
ta ntly from those of i nten s i ve soj o u rn s of a year or two, fo l l owed perhaps
by a l ater return v i s i t to measu re "c hange" ( Foster et a l . 1 9 79). The agi n g
o f both fieldwo rkers and i nformants and t h e acc u m u l ated experience o f
cooperative w o r k over decades prod uce at least t h e effect of a deepen i ng
know l ed ge . To conceive of th i s experience as an i n itiation has the merit
of i n c l u d i n g i nd i genous "teachers" as centra l subj ects i n the process. Do­
gan i n struction of G ri a u l e in /a parole claire is a l so an i m p l icit criticism
of the earl ier "docu mentary" research ; i ndeed one wonders if most eth­
nogra p h ies generated over a rel ative l y narrow time span may not be pa ­
roles de face. The n arrative of i n itiation sharply q uestions approac hes
that d o not strive for a certai n l evel of com plexity in representi n g "the
native po i n t of v iew." Ogotemmel i 's i n itiative need not be portrayed as a
com p l etion ( i n G ri a u le's word s a "couronnement") of the earl ier re­
searc h . It can a l so be seen as a comment on it and a s h i fti ng of its epis­
temological bas i s . Here the Dagon "si de" of the story rema i n s prob lem­
atic : d i rect evidence i s lacki ng, and the i n it iatory narrative with its
assu med teleology-a progress towa rd the most complete possible
know ledge-ceases to be h e l pfu l .
I t i s clear that Ogote m me l i 's i n tervention was a cruc i a l tu rn i n the
research p rocess. I t revealed the extent of Dagon contro l over the k i nd
of i nformation access i b l e to the eth nographers. It annou nced a new styl e
of research i n w h ich the authority of i n formants w a s more expl icitly rec­
ogn i zed . N o l o n ger u ntrustworthy witnesses s u bjected to cross-exam i n a­
tion, the Dogan "doctors ," Ogotemmel i and h i s successors, were now
lea rned i n terlocutors . D u r i n g the "doc u mentary" ph ase of the researc h
the eth nographer had been an aggress ive co l l ector of observations, arti­
facts, and texts . N ow he or she was a tra nscri ber of form u l ated lore, a
tra n s l ator, exegete, and commentator. I n G r i a u le's account of the i r meet­
i ngs Ogote m me l i is not i nterrogated i n the man ner outl i ned in Methode
de l'ethnographie. " Le b l a nc," "the N azarite," as G r i a u l e now someti mes
ca l l s h i m se l f, has become a student; the secret is com m u n icated free l y,
not confessed .
The doc u m entary and i n itiatory pa rad igms, however, are l i n ked by
i m porta n t u nder l y i n g assu m ptions. To see ethnography as e i ther extract­
i n g confessions or u ndergo i ng i n itiatio n , one m u st assume the exi stence
and i m po rtance of secrets . Cu l t u ra l truth i s structu red in both cases as

someth i n g to be revea l ed (Griau le's freq uent work i s decele: d i sc l osed ,

d i v u l ged, detected , u ncovered ) . Moreover, the new pa rad igm i n co rpo­
rates the theatrica l conception of fieldwork. In a "para l l e l " i n itiation the
eth nographer p l ays the pa rt of an i n itiate, the i nformant, an i nstructor. A
d ramatic re l ationsh i p, recogn i zed as such by both parties, becomes the
enabl i n g fiction of encounter. I ndeed if a l l performances a re contro l l ed
revel ations presu ppos i n g a " back region" h idden from view where the
performance is prepa red and to w h i c h access i s l i m i ted (Goffman 1 95 9 :
2 3 8 ; Berreman 1 9 7 2 :xxx i i), then a theatrical model o f re lationsh i ps nec­
essari l y presu p poses secrets . Thus an underl ying l ogic of the secret u n i tes
the two phases of Griau le's career. 5 Whether the eth nographer is a re­
lentless "judge" or a hel p i n g "m idwife," the truth must always emerge,
be brought to l ight. As an i n itiate, the researc her rece ives and i n terprets
Th is view of the emergence of truth may be contrasted with a con­
ception of eth nography as a d i a l ogical enterprise in w h i c h both research­
ers and natives a re active creators or, to stretc h a term, authors of c u l tu ra l
representations. I n fact Griau l e's experience with the Dogan may b e bet­
ter accounted for in th is second perspective; but to say this presu pposes
a critique of i n itiatory authority. D i a l ogica l , constructivist pa rad igms tend
to d i sperse or share out eth nographic authority, wh i le narratives of i n itia­
tion confirm the resea rc her's spec i a l com petence. I n itiation assumes an
experience of progressi ve, connected reve l ations, of gett i n g beh i nd h a l f­
truths and ta boos, of bei n g i n structed by authentica l ly q u a l i fied members
of a com m u n i ty. Th i s experience of a deepe n i n g "education" em powers
the eth nographer to speak as an i nsider on behalf of the com m u n i ty's
truth or rea l i ty. Though a l l c u l tu ra l l ea rn i n g i n c l udes an i n itiatory d i men­
sion, G r i a u l e presses th is logic to the l i m i t: "proceed i n g by means of
successi ve i nvest i gations among more and more knowl edgeable strata
of the soc iety, it is possi bl e to consi derably red uce a pop u l ation's area of

5. jam i n ( 1 982a : 88-89) disc u sses this aspect of Griau le's work. For a stim­
u l ati ng treatment of the soc i a l functions of secrets see his Les Lois du silence
( 1 977). Sec rets are part of the mise en scene s oc ia le generators of group identi­

ties and of c u l tu ra l mea n ings which, not goa l s to be fi nal ly attai ned, are "end­
lessly deferred and d i ss i m u l ated" (p. 1 04) . My d i scussion of the exegetical fu nc­
tion of Ia parole cla ire draws on this genera l perspective, as wel l as on Kermode
( 1 980) . For a trenchant critique of the "cryptologica l " assumptions u nderl ying
G riau le's practi ce and that of many "symbolic anthropologists" see Sperber
( 1 9 7 5 : 1 7-50) . Perhaps the most subtle critique of the logic of secrecy is con­
tai ned in Victor Segalen's Rene Leys ( 1 922); see Chapter 5 .
P O W E R A N D D I A L O G U E I N E TH N O G R A P H Y 85

esoteric know l edge, the o n l y o n e , to te l l the truth , that i s i m porta nt, s i nce
it constitutes the nat i ve key to the system of thought and action"
( 1 9 5 2 c : 54 5 ) .
T h i s " n ative key" began t o emerge for G ri a u l e and h i s co-workers i n
t h e l ate forties and early fifties. T h e l a ndmark books a n no u n c i n g i t s d i s­
covery were Dieu d'ea u (Conversations with Ogotemmeli) ( 1 948a) and
D i eterlen's Essai sur Ia religion Bambara ( 1 9 5 1 ). The two works revea led
a "deep thought among the b l acks," "an i ntricate network of representa­
tions" ( D ieterlen 1 95 1 : 2 2 7) . The " i n n u merable correspondences" of the
B a m ba ra and Doga n emerged as a "coherent tableau ," a "metaphysic"
(Gri a u l e 1 9 5 1 : i x) . Once Ogotemme l i had, i n t h i rty-th ree days of mean­
deri n g ta l k, e n u n c i ated the basic outl i nes of Dagon cosmogo n i c myth ,
an enormous work of e l u c idatio n rem a i ned . As recorded i n G r i a u le's
d ay-by-day accou nt, h i s d i sco u rse was ridd l ed with gaps and contrad i c­
tions. The c u ltural m aster sc r i pt he had sketc hed wou l d req u i re el aborate
exeges i s , c ross-c hec k i n g aga i nst other versions of myths, and attention
to the scri pt's enactment i n v i rtu a l l y every doma i n of col l ective l ife.
T h i s work was to occu py G ri a u l e and h i s co-workers for decades . It
wou l d a l so occ u py thei r smal l gro u p of key i nforma nts, d rawn from the
esti mated 5 percent of "com p l ete l y i n structed" Dogan in the Sanga re­
g i o n , as we l l as from the 1 5 percent of the popu l ation who possessed a
fa i r portion of the sec ret know l edge (Griau l e 1 95 2 a : 3 2 ) . There is d i s­
agreement about the p rec i se nat u re of the Dagon " revel ations" prod u ced
i n t h i s col l a borat i o n . Some h ave seen them as theo l ogical spec u l ations
by i nd i v i d u a l Dagon o r as mythopoe ic i n ventions (Goody 1 9 6 7 : 24 1 ;
Lew i s 1 9 7 3 : 1 6 ; Copans 1 9 7 3 : 1 5 6 ) . G r i a u l e and D i eterlen, however,
stro n g l y reject the noti o n that the know l ed ge they report is i n any sign if­
icant sense the origi n a l c reation of spec ific Dago n . I n the i r v iew the u n i­
form ity of c ustom and the w idespread behav ioral a rticu l ation of the eso­
teric knowled ge makes it u n l i ke l y that any i nd ividual cou ld have done
more than s l i ghtly i nflect the end u r i n g myth i c structu res . But to pose the
issue as a debate between personal origi n a l ity and c u ltu ra l typical ity
( H o u ntondj i 1 9 7 7 : 79-1 0 1 ) i s probab l y fru itless, given o u r ignora nce
about key i n formants. T h i s view is based a l so o n a fa l se d i c h otom y : a l l
authors, whether African o r E u ropea n , a re o ri g i n a l o n l y with i n l i m ited
resou rces and i n restricted relations of textu a l prod uction .
It is tem pti n g to portray the l ate works of the G r i a u l e schoo l , i n the
words of Pierre Alexandre, as "second l evel ethnography-the eth nog­
raphy of Dagon eth n ography" ( 1 9 7 3 :4). Th i s notion of " leve l s " does not

do j u stice, however, to the way i n w h i c h Griaule's vers ion of custom and

the versions e n u n c i ated by Dogan i nforma nts are d ia l ogica l l y i m p l i cated
i n one another, for it is d ifficu lt, if not i m poss i b l e, to sepa rate c l early
Dogan ethnography from G riau le's eth nography. They form a common
proj ect : the textu a l ization and exegesi s of a trad itiona l system of knowl­
edge . The cu ltura l "text" does not ex i st prior to i ts i n terpretati on; it i s not
d i ctated by fu l l y i n structed i nforma nts and then exp l i cated and contex­
tual ized at a second " l eve l " by European ethnographers. G r i a u l e and
D ieterlen g i ve evidence that there can i n fact be no complete version of
the Dogan "metaphys i c ." If, i n G riau le's te l l i ng meta phor, it is "written"
throughout the cu lture- i n the habitat, i n gestu res, i n the system of
gra p h i c signs-these traces a re of the o rder of a mnemon icon rather than
of a com plete i nscri ptio n . In fact a "fu l ly i n structed" Dogan wi l l spend a
l ifetime mastering /a parole cla ire. To grasp the fu l l range of its sym bo l i c
correspondences, signs, myths, rites, a n d everyday gestu res req u i res a
conti nuous process of concrete poes i s . The myth i c "word" is end l essly
mate ri a l i zed, exchanged, i n terpreted . Beca u se stable order is relen tl ess ly
d i sru pted by the forces of d i sorder, i ncarnate i n the myth i c renard pa te,
cosmos and soc iety are constantly re i nscri bed .
The ethnographic encou nter is one of the occasions of th i s rei n scri p­
tion, but with a s i gn ificant d i fference. Now the Dogan d i a l ectic of order
and d i sorder takes p l ace on a world stage, lead i n g to the i nscri ption of a
new k i nd of tota l ity, a Dogan essence or cu lture . I n Le renard pate we
see an attem pt to esta b l i s h a cultura l base l i ne, to separate, for example,
"commentaries" by i n formants from the recorded myths and va riants. It
is u n c l ea r, thoug h , how rigorously such a separation can be made, for as
Dieterlen says, these gl osses demonstrate the Dogan propensity to "spec­
u l ate on the h istory of creati on," an ongoing "native deve l opment of
thought on the bas i s of myth i c facts" (Griau le and Dieterlen 1 965 : 5 6) .
The deve l opment o f myth i c thought, a s o f any thought, i s both structu red
and open-ended, but the activity of exeges is depends on the pos iting of
a restricted set of symbo l s by the hermeneutica l i magi nation . There m u st
i n pri nc i p l e be a stable corpus for i nterpretation . Griau le's "fu l l " i n itiatory
know l edge-which can never be ex pressed i n its enti rety-fu nctions i n
th i s canon ical man ner. It provi des a stoppi ng poi nt for t h e process of
cu ltura l representation . On the bas i s of this original master script a po­
tenti a l l y end l ess exegetical d i scou rse can be generated . La parole cla ire,
l i ke any primal text or ground of authori ty, acts to structu re and em power
i nterpretat i o n .

G ri a u le's parad igm of i n iti ation fu nctioned to tran sform the eth nog­
rapher's ro l e from o bserver and doc u menter of Dagon cu lture to exegete
and i n terpreter. It p reserved and refo rm u l ated , h owever, the domi nant
themes of h i s ear l ier p ract ice: the l ogic of the secret, a n aspi ration to
exhaustive knowled ge, a vision of fieldwork as ro l e playi ng. I t expressed
a l so the sense one h as throughout G r i a u l e's career of h i s Dogan counter­
parts as powerfu l agents i n the eth n ographic process, i n iti a l l y clever tac­
ticians and w i l l fu l res isters, l ater teachers and co l leagues . By atta i n i n g /a
parole cla ire and worki ng l i ke any i n iti ate to grasp the "word 's" i ncar­
nation i n the experienti a l wor l d , G ri a u l e becomes (always i n h i s para l l e l ,
"eth nogra p h i c " pos ition) o n e o f a restri cted gro u p o f "docto rs" o r "meta­
phys i c i a n s" who control and i nterpret Dogan knowledge. G r i a u l e is an
i ns i de r, but with a d i fference. It is as though the Dagon had recogn i zed
the need for a k i nd of cu ltura l am bassador, a q u a l i fied representative
who wou l d dramatize and defend the i r c u l tu re i n the co l o n i a l world and
beyon d . G ri a u l e in any case acted as if this were his ro l e .
T h e sta nce of the eth nographer w h o speaks as an i nsider on beh a l f
o f h i s or her people i s a fa m i l i a r one; i t i s a stoc k ro l e o f t h e ethnographic
l i bera l . Griaule adopted th i s stand po i n t i n the ear l y fifties with confi­
dence and authority. An active advocate and med iato r in the colon i a l
pol itics of t h e Sanga reg i o n , h e effected a reconc i l iation between trad i ­
tional Dogan authorities and the n e w c h i efs i n sta l led b y the government
(Ogono d ' Arou 1 9 5 6 : 9 ) . In a variety of forums, from the pages of Pre­
sence africaine to U N ESCO i n ternational gatheri ngs to the Assem b l y of
the U n ion Fran�a i se (where he served as president of the Com m i ssion on
Cu ltura l Affa i rs), h e u rged respect for the trad itions of Afri ca . Fortified by
Ogote m m e l i 's reve lations, he portrayed in e l a bo rate deta i l a mode of
knowl edge to rival or su rpass the occ identa l l egacy of the G ree ks . Speak­
i n g perso n a l l y, in the voice of an i n iti ate, he cou ld report about the Do­
gan that "with them , everyth i n g seems truer, more noble, that is to say
more classica l . T h i s m ay not be the i m press ion you have from the out­
side, but as for me, each day I seem to be d i scoveri n g someth i n g more
beautifu l , more shaped , more sol id " ( 1 9 5 2 b : 1 66 ) .
One senses i n the work o f G r i a u l e a n d a m o n g h i s co-workers-es­
pec i a l l y Germa i ne D ieterlen-a profo u n d , sometimes mystical engage­
ment with the Dogan sophie ( Rouch 1 9 78b : 1 1 -1 7 ) . B u t whereas D ieter­
len has ten ded to efface her own a uthority beh i nd that of the Doga n ,
G r i a u le, who l ived t o see o n l y t h e begi n n i ngs o f "deco l o n i zation," spoke
in fra n k l y paterna l i st acce nts as an advocate for African trad itional c u i -
88 D I S C O UR S E S

tures. H i s l ate genera l i zations a re governed by a fa m i l i a r c h a i n of syn­

ecdoches. Ogotemme l i and Sanga stand for the Dogan, the Dogan for
the trad itiona l Suda n , the Sudan for B l ack Africa, Africa for l'homme noir.
G ri a u l e moves free l y from level to l eve l , constructing an elementa l civi­
l i zation stri k i ngly d ifferent from that of E u rope; but d i fference i s estab­
l i shed o n l y to be d i sso l ved in a tota l iz i n g human ism ( 1 9 5 2 b : 24). Once
trad itional African essence is characteri zed and sym pathetica l l y de­
fended , it i s then portrayed , in the l ast i n stance, as a response to "the
same great pri n c i p l e , to the same great h u man u ncerta i n ties" that West­
ern science and ph i l osophy h ave engaged ( 1 9 5 1 : 1 66) . The eth nographer
speaks as a part i c i pant in two civi l izations that by means of h i s i n iti atory
experience and spec i a l knowl edge can be brought together at a "human"
leve l .
I n the ea rly fifties G r i a u l e presents h i mself a s someone who knows
Africa and who knows too what is good for Africa . Eth nographic u nder­
sta nd i ng is critical i n a changing colon i a l context: it perm its one to "se­
lect those moral va l ues which are of merit and sho u l d be preserved ," to
"decide what i nstitutions and what systems of thought shou l d be pre­
served and propagated in B l ack Afri ca" ( 1 95 3 : 3 72 ) . Trad ition must be
wel l u nderstood so that change can be properly gu ided . " I t is a question
of tak i n g what's the i rs that i s rich and tra nspos i ng it i nto our own situa­
tion, or i n to the s i tuation we w i sh to make for them" ( 1 95 1 : 1 63 ) .
G r i a u l e's "we" be l ongs to 1 95 1 a n d the colon i a l U n ion Franc;aise, of
which he was a cou nc i l o r.
The c u ltura l ric hes that wi l l somehow be preserved or transposed
are a l ways l ocated in the doma i n of trad ition or "authentic" custom-an
area more or l ess free of E u ropea n of I s l a m i c i nfl uences . The eth no­
gra p h i c l i beral who represents the essence of a cu lture aga i nst impure
"outs ide" forces encounters sooner or l ater a contrad iction bu i l t i nto a l l
such d i scourses that resist or try to stand outside h i stori cal i nvention . The
most persi stent critics of G r i a u l e's defense of Africa were ed ucated Afri­
cans, evo/ues, who rejected any re ification of the i r c u l tu ra l past, how­
ever sym pathetic . G r i a u l e tended to exp l a i n away these resistances as
u n fortunate con seq uences of an u n ba l a n ced education : "You can't be
s i m u l taneously at schoo l a n d i n the sac red grove" ( 1 95 1 : 1 64 ; see a l so
Ma l raux 1 95 7 : 1 5 ) . The b l ack i nte l lectu a l s who objected to h i s eloq uent
portraya l s of the i r trad itions were no longer authentica l l y African but
were vict i m s of "that k i nd of ' l ead i n g astray of m i nors' wh ich a l l colon i a l
powers h ave i n d u l ged i n " ( 1 95 3 : 3 7 6 ) .

Such statements no longer ca rry the authority G r i a u l e was able to

i m part to them in the ea rly fifties; in fact they were cha l l enged even on
the occasion of their e n u n c i ation ( G r i a u l e 1 95 2 b : 1 4 7-1 66). More con­
gen i a l today are the views expressed at the same ti me by G r i a u l e's earl y
co l l eague M i c h e l Le i r i s . A brief fi n a l contrast w i l l evoke the changing
ideo logical s i tuation i n the years before G r i a u le's death , a s i tuation in
w h i c h ethnography i s sti l l en meshed .
Le i r i s was perhaps the fi rst eth nographer to confront sq ua re l y the
po l i tical and epistemo logical constra i nts of colon i a l ism on fieldwork
( Le i r i s 1 95 0 ) . H e v i ewed the eth nogra pher as a natu ra l advocate for ex­
p l o i ted peo p l es , and he wa rned agai nst defi n itions of authentic ity that
exc l uded evolues and the i m puriti es of c u ltura l syncreti s m . Both Le i ri s
a n d G r i a u l e contri buted essays i n 1 95 3 t o a U N ESCO col lection entitled
Interrelations of Cultures. The d i fferences i n thei r approaches a re sti l l
i nstructive today. G ri a u l e's essay, "The Problem of N egro C u l tu re," argues
that "trad itional re l i gions, as we l l as the soc i a l and l ega l structure and
tec h n ical c rafts of the b l ac k races emanate from a s i ngle, rigid system of
thought-a system that prov ides an i nterpretation of the u n i verse, as we l l
as a p h i losophy enabl i n g the tribe to carry o n and the i nd ividual to l ead
a balanced l i fe" ( 1 9 5 3 : 3 6 1 ) Dogon and B a m ba ra exa m p l es are el i c i ted

to i l l ustrate t h i s " m etaphysical su bstratu m ," w h i c h G r i a u l e presents

th roughout as c h a racteristic of "the Negro" or of "negro c u l tu re" (p. 3 6 2 ) .
Le i r i s , i n approac h i ng h i s top i c , "The African Negroes a n d t h e Arts of
Carv i ng and Scu l ptu re," evokes a h i stori ca l l y s pec ific problem of i n ter­
c u ltural tra n s l ati o n . He begi ns by trac i n g the d i scovery of art negre
among the ava nt-garde early i n the century- E u ropea ns i nventing an Af­
rican aesthetic for the i r own a rti stic pu rposes. He then th rows doubt o n
h i s o w n u nderta k i n g b y po i nt i n g out the absurd i ty of an African attem pt­
i n g i n a short essay to deal with the whole of " E u ropean scu l ptu re ." He
p roceed s to base h i s genera l izations about "African" art not on any pre­
s u m ption of a common essence but on a conti n gent perspective . H e
writes as a Westerner perce i v i n g s i m i l arities a m o n g t h e d i verse sc u l ptu res
of Africa and even p resenting them as expressions of a "civi l i zation"
w h i l e u ndersta n d i n g these ensemb les to be i n a sense optica l i l l usions.
The a pparent u n ity of black art forms i n h eres o n l y i n a perception of the
common ways i n wh i c h they d i ffer from those to w h i c h a Eu ropea n is
accu stomed . (See Cha pter 9 , n . 3 . ) T h i s refu sal to represent an exotic
essense-an i m portant i ssue of epistemological tact- i s based (in pa rt at
l east) on the ways in which Le i ris' eth nographic career d i verged from

that of h i s co-worker on the M i ssion Dakar- Dj i bouti . Lei r i s never under­

went any " i n itiation" i nto an exotic form of l i fe or bel ief. I ndeed h i s work
(espec i a l l y L'Afrique fantome) i s a re lentless critique of the parad igm of
i n itiation . H i s l iterary work, largely devoted to a heterodox, end less
autobiography, rei nforces the ethnographic po i nt. (See Chapter 6 . ) H ow
cou l d Le iris presume to represe nt another c u l tu re when he had trou ble
enough representing h i mself? Such an attitude made susta i ned fie ldwork
i m poss i b l e .
Griau le's energetic confidence i n cu ltura l representation cou ld not
be farther from Le i r i s' tort u red, l ucid u ncerta i n ty. The two pos itions mark
off the pred icament of a postcol o n i a l ethnography. Some authoriz i n g fic­
tion of "authentic encounter," in Geertz's ph rase, seems a prereq u i site for
i ntensive research ; but i n iti atory c l a i m s to spea k as a knowled geable i n ­
sider reveal i n g essenti a l c u ltu ra l truths are no longer cred i b l e . Fie ldwork
can not a ppea r pri mari l y as a c u m u l ative process of gathering "ex peri­
ence" or of c u l tu ra l " l ea rn i ng" by an automonous subject. I t m ust rather
be seen as a h i storica l l y conti ngent, u n r u l y d ia l ogical encou nter i nvo lv­
i n g to some degree both confl ict and co l l aboration i n the prod uction of
texts. Eth nographers seem to be condemned to strive for strive for true
encou nter wh i l e s i m u l taneously recog n i z i ng the po l itica l , eth ica l , and
person a l cross-purposes that u nderm i ne any transmiss i o n of i n terc u l tu ral
knowl edge. Poi sed between G ri a u le's enactment and Le i r i s's refusal of
th i s i ro n i c pred i cament, and worki ng at the now b l u rred bou ndaries of
eth nogra p h i c l i beral i s m , fiel dworkers struggle to i m provi se new modes
of a uthority.
They may perhaps fi nd some retrospective encou ragement i n the
Griau le trad ition of eth nogra p h i c cu ltura l i nventi on, for the story con­
ta i n s elements that poi nt beyond i n itiatory authority and the neoco l on i a l
context. To d ate the most i l l u m i nating account o f how research pro­
ceeded in the wake of Ogotemme l i i s Genevieve Ca l ame-Griau le's pref­
ace to Ethno/ogie et /angage: La parole chez /es Dogan ( 1 965). She tel l s
how "the extremely prec ise v iews" she gathered from h e r i nterlocuto rs
led to the e l a boration of "a veritable Dagon 'theory' of speech " ( p . 1 1 ) .

She i ntrod uces her fou r key co l l aborators, giving h i nts of the i r personal
styles and preoccu patio n s . We learn that one of them , Manda, was the
Dagon eq u ivalent of a "theologian" and that he gu ided the eth nographer
toward the re l ations of speech and the person that became the book's
orga n i z i ng pri n c i p l e . Even the book's descriptions and i n terpretations of
everyday behavior were the work of both eth nographer and i nforma nts,

many of the l atter possess i ng extraord i na ry "fi nesse i n o bservation" (p.

1 4) . Wh i le Cal ame-G riau l e sti l l m akes a guarded claim to represen t an
overa l l Dogan "cu ltural orientation," her preface goes a long way toward
cast i n g the ethnogra p h i c process i n spec ific d i a l ogical term s . The theory
of speech t h at Calame-G riau l e has bri l l i antly com p i l ed i s i nescapably a
co l laborative work, cont i n u i n g her father's prod uctive encou nter with
the i n habitants of Sanga . And it is an authentic c reation of " Dogan
thought's need in expressi ng i tse l f for d i a l ectic, for an exchange of ques­
tions a n d an swers that i nterpenetrate and weave themse l ves together"
(p. 1 7) .
. . . the age in which we are camped, like bewildered
travelers in a garish, unrestful hotel.

My whole ethics is based on the fundamental instinct of

unified personality.

3 . On Ethnographic Self-Fashioning :
Conrad and Malinowski

To S A Y T H A T the i n d ividual is c u ltura l l y constituted has become a

tru i sm . We a re acc ustomed to hearing that the person i n Bal i or among
the Hopi or in med ieval soc iety i s d i fferent-with d ifferent experiences
of ti me, space, k i n s h i p, bod i l y identity-from the i nd ividual in bou r­
geo i s E u rope or i n modern America . We assume, a l most without q ues­
tion, that a self be longs to a spec ific c u l tu ra l world much as it spea ks a
native l a nguage : one se lf, one cu lture, one language. I do not wish to
d i spute the considerable truth contai ned i n even so bald a form u l a ; the
idea that i n d i v i d u a l ity i s a rti cu l ated with i n worlds of s i g n i fication that are
col lective and l i m ited i s not i n q uesti o n . I want, however, to h i storic ize
the statement that the se l f i s cu ltura l l y consti tuted by exam i n i n g a mo­
ment a round 1 900 when this idea bega n to make the sense it does today.
I n the m id - n i neteenth century to say that the i nd ividual was bound
u p i n cu lture meant someth i ng q u ite d i fferent from what it does now.
"Cu lture" referred to a s i ngle evo l utionary process. The European bou r­
geoi s ideal of autonomous i nd ividual ity was wide ly be l ieved to be the
natura l outcome of a long deve lopment, a process that, a lthough th reat-

O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 93

ened by various d i sruptions, was assumed to be the basic, progress i ve

movement of h u m a n ity. By the turn of the centu ry, however, evo l ution i st
confidence began to fa lter, and a new eth nographic conception of c u l ­
ture became poss i b l e . T h e word began t o b e u sed i n the p l u ra l , suggest­
i n g a world of sepa rate, d i st i nctive, and eq u a l l y mean i n gfu l ways of l i fe .
T h e idea l o f a n a u tonomous, c u ltivated s u bject cou l d appear as a local
project, not a telos for a l l h u m a n k i n d . 1
The u nderly i ng causes of these ideo logical devel opments are be­
yond my scope here . 2 I wa n t o n l y to ca l l attention to the deve l o pment i n
t h e early twentiet h cent u ry o f a new "eth nograp h i c s u bjectivity." Modern
anthropology-a S c ie nce of Man l i n ked c l osel y to c u ltura l desc r i ption­
pres u p posed the i ro n i c stance of parti c i pant observati o n . By profession­
a l i z i n g fiel dwork anthropology transformed a widespread pred i cament
i nto a scientific method . Ethnograph i c know l edge cou l d not be the prop­
e rty of any one d i scou rse or d isc i p l i ne : the cond ition of off-centered :
ness i n a wor l d of d i sti nct mean i n g systems, a state of be i n g i n c u l tu re
wh i le looki n g at cu lture, permeates twentieth-century a rt and writing.
N ietzsch e had c learly a n nou nced the new stance i n h i s famous fragment
"On Truth and Lie in a n Extra-Mora l Sense," aski n g : "What, then i s truth ?
A mob i le army of meta phors, metonyms, and anth ropomorph isms- i n
short, a s u m o f h u m a n re l ations, w h i c h have been enhanced , trans­
posed , and e m be l l i s hed poetica l ly and rhetorica l ly, and wh i c h after long
use seem fi rm, canon i ca l , and ob l i gatory to a people" (Kaufman
1 954 :46) . N ietzsc he, perhaps more than Ty l o r, was the main i nvento r of
the rel ativist i dea of cu lture : th is c h a pter cou ld we l l have bee n ca l led
"On Truth and L i e i n a Cultural Sense."
I have i n stead taken my title from Stephen G reen b l att's Rena issance
Self-Fash ion ing ( 1 980) , a work that traces a n emergi ng, bou rgeo is, mo­
b i le, cosmopo l itan sen se of the se l f. The ethnographic s u bjectivity I a m
concerned w i t h m a y b e seen as i ts l ate variant. T h e s i xteenth-century

1 . On the development of the concept of c u lture see Wi l l iams 1 966, Stock­

i n g 1 968, and Chapter 1 0 . The novelty and frag i l ity of the Western notion of the
individual was noted in Mauss 1 938, perhaps the fi rst eth nograph i c overview of
the subject.
2. A fu l l analysis of changes i n the "cu l ture" response wou ld presu ppose
those forces taken by Raymond Wi l l iams ( 1 966) as determi nants: i ndustria l ism,
social confl ict, the rise of mass culture. To these wou l d be added the needs of
h i gh colon i a l soc ieties to understand the increas i ngly accessi ble d i versities of the
planet as a d i spersed tota l ity. The mappi ng of the world's h u man arra ngements
as d i sti nct c u ltures asserts that thi ngs hold together-separately.

figures of More, Spenser, Marlowe, Tyndale, Wyatt, and Shakespeare ex­

empl ify for G reen b l att "an i n c reased self-consciousness about the fash­
ion i n g of human identity as a m a n i p u lable, artfu l process" (p. 2). I can not
do j ustice to the book's subtle and persuasive i n d i vidual analyses, but I
want to note G reenbl att's own eth nograph i c standpo i nt, the com p l ex at­
titude he m a i nta i ns towa rd fashioned se l ves, i nc l ud i ng h i s own . He rec­
ogn i zes the extent to wh ich recent questions about freedom, identity, and
l a nguage have shaped the version he constructs of si xteenth-centu ry c u l ­
ture . He i m ports a modern critica l a pproach t o h i s materi a l . Yet he writes
too as someone caught up with and loyal to a trad ition . He expresses i n
a movi n g epi l ogue h i s stu bborn com m itment to the poss i b i l ity of shap i ng
one's own identity, even if th i s means o n l y to "selfhood conceived as a
fiction" ( p . 2 5 7) . H e is l ed to what Con rad approvingly cal led a "del i b­
erate be l ief."
G reen b l att is a part i c i pant-analyst, construct i n g and engag i n g a c u l ­
tural formation that i s both d istanced i n t h e si xteenth centu ry a n d d i a lec­
tical l y conti nuous with the present. H i s " l ate," reflex i ve vers ion of Ren­
a i ssance self-fash io n i n g rel i es on a sharply articu l ated eth nograph ic
v i ew po i nt. The fash i oned , fictional self i s a l ways l ocated with reference
to its cu lture and coded modes of expression, its language. G reenbl att's
study conc l udes that Renai ssance self-fas h i o n i ng was anyth i ng but the
u nconstra i ned emergence of a new i nd ividua l i st autonomy. The subjec­
ti vity he fi nds is " not an epi phany of i dentity free l y chosen but a c u ltural
artifact" ( p . 2 5 6) , for the se lf maneuvers with i n constrai nts and poss i b i l ­
ities given b y an i nstitutional ized set o f co l lective practices a n d codes.
G ree n b l att i nvokes symbo l ic-i nterpretive anthropology, particularly the
work of Geertz (a l so Boo n , Douglas, Duvignaud, Rabi now, and Tu rner) ;
and he knows, moreover, that cu ltura l symbo l s and performances take
shape in situ ations of power and domi nance. One hears echoes of Fou­
cau lt in G ree n b l att's warn i n g : "The power to i m pose a shape u pon one­
se lf i s a n aspect of the more general power to control identity-that of
others at l east as m u c h as one's own" (p. 1 ) . It fo l l ows that eth nograph ic
d i scou rse, i n c l u d i ng G reenbl att's l iterary varia nt, works in this double
man ner. Though it portrays other se lves as c u l t u ra l ly constituted , it a l so
fas h i o n s an identity authorized to represent, to i nterpret, even to be­
l ieve-but a l ways with some i rony-the truths of d i screpant worl ds.
Ethnogra p h i c subjectivity i s com posed of part i c i pant observation in
a world of "cu ltura l art i facts" l i n ked (and th i s i s the origi nal ity of
N ietzsche's form u l ation) to a new conception of l a nguage-or better,
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 95

l a nguages-seen as d i sc rete systems o f signs. Along with N ietz sche, the

th i n kers who stake out my a rea of exploration are Boas, D u rkhe i m , and
Mal i n owski ( i n ventors and pop u l arizers of the ethnographic c u l tu re
idea) , and Saussure . They i naugurate an i ntercon n ected set of ass u m p­
tions that a re n ow i n the last q u a rter of the twentieth centu ry j ust becom­
ing v i s i ble. A n i nte l lectual h i storian of the year 2 0 1 0, i f such a person is
i magi nable, may even look bac k on the fi rst two-th i rds of our centu ry
and observe that th i s was a ti me when Western i nte l lectu a l s were pre­
occ u p ied with grounds of mea n i ng and identity they cal l ed "culture" and
" l a n guage" (much the way we now look at the n i neteenth centu ry
and perce ive there a probl ematic concern with evo l utionary " h i story"
and "progress" ) . I t h i n k we a re see i ng signs that the pri v i l ege given to
natu ra l l a n guages and, as it were, natural c u l tu res, is d i sso l v i n g . These
objects and epi stemological grounds are now appeari ng as constructs,
ach ieved fictions, conta i n i ng and domesti cat i n g hete rogl oss i a . In a world
with too many voices spea k i n g all at once, a world where syncreti sm
and pa rod i c i nvention are beco m i ng the rule, not the exception, an u r­
ban , m u lti nati onal world of i nstitutional tra n s ience-where American
c l ot h es made in Korea a re worn by you ng people i n Russia, where every­
one's " roots" a re in some degree cut- i n such a world it becomes i n ­
creas i n g l y d iffic u l t t o attach h u man identity a n d mean i ng t o a coherent
"cu lture" o r " l angu age."
I evoke t h i s syncretic, " postcu ltura l " s ituation o n l y to gestu re towa rd
the standpo i nt (though it can not be so eas i l y spati a l i zed) , the cond ition
of u n certa i nty from which I a m writi ng. B u t my concern i s not with the
poss i b l e d i sso l ution of a s u bjectivity anchored i n cu lture and la nguage .
Rather, I w a n t to exp lore two powerfu l art i c u l ations o f th i s subjectivity i n
t h e works o f Conrad a n d Ma l i nowski , two d i s p l aced person s both of
whom stru ggl ed in the ea rly twentieth cent u ry with cosmopo l itan ism and
com posed thei r ow n vers i o n s of "On Truth and Lie in a Cu ltura l Sense."
Con rad may have seen more dee p l y i nto the matter, for he b u i l t i nto h i s
work a v i s i o n o f t h e constructed nature o f cu lture a n d language, a seri ous
fictiona l i ty he de l i berate l y, a l most absurd l y, embraced . But a comparable
grapp l i ng with c u ltu re and l a n gu age may be seen i n Ma l i nows k i 's work,
parti c u l a r l y in the d iffic u l t experience and l iterary represe ntation of h i s
famous Trobriand fieldwork. (Th i s fie ldwork has served a s a k i n d of
fou n d i n g charter for the twentieth-centu ry d i sci p l i ne of anthropology. )
Con rad acco m p l i shed t h e a l most i m poss i b l e feat o f beco m i ng a great
writer ( h i s mode l was F l a u bert) in Engl i s h , a th i rd l a n guage he bega n to

acq u i re at twenty yea rs of age . I t is not su rpri s i ng to fi nd throughout h i s

work a sense o f t h e s i m u l ta neou s artifice a n d necess ity of c u l tu ra l , l i n­
gu i stic conventions . H is l i fe of writing, of constantly beco m i ng an En­
gl i s h writer, offers a parad igm for ethnographic subjectiv i ty ; it en acts a
structu re of feel i ng conti nuously invo l ved i n translation among lan­
gu ages, a consc iousness d eeply aware of the arbitra ri ness of conven­
tions, a new sec u l a r rel ati v i s m .
Ma l i nows ki remarked , " [W. H . R . ) Rivers i s t h e R i d e r Haggard of
Anth ropol ogy : I s h a l l be the Con rad ! " (to B . Z. Sel igman, q uoted i n F i rth
1 95 7 : 6) . He probably h ad i n m i nd the d i fference between Rivers' m u lti­
cu ltu ra l su rvey methodo logy (co l lecting tra i ts and genea logies) and h i s
own i ntensive study o f a s i ngle grou p . For Ma l i nowski t h e name Con rad
was a symbol of depth , complexity, and s u btlety. (He i n vo kes h i m in t h i s
sen se i n t h e fie l d d i a ry. ) B u t Ma l i nowski w a s n o t t h e Con rad o f anth ro­
po logy. H i s most d i rect l ite rary model was certa i n l y James F razer; and i n
much o f h i s o w n writi n g he was rem i n i scent o f Zo la-a natura l i st pre­
senti ng facts p l u s hei ghtened "atmosphere," h is sc ientific-cu ltura l de­
sc ri ptions yield i ng mora l ly cha rged h u ma n i st a l l egories. Anthropo l ogy is
sti l l wa i t i n g for its Conrad .
My comparison of Mal i nowski and Con rad foc uses on the i r d iffic u l t
access ion to i n novative professional expression . Heart o f Darkness
( 1 899) is Con rad 's most p rofound med itation on the d ifficu l t process of
giving h i mself to England and Engl ish . 3 I t was written in 1 898-99, j ust
as he d ec i s ive l y adopted the land l oc ked l i fe of writing; and it looked
back to the begi n n i ng of the process, h i s l ast, most audacious voyage out
to his "fa rthest poi n t of navigation ." O n the jou rney u p the Congo a dec­
ade earl i er, Kon rad Korzen iowsk i had carried with h i m the i n itial chap­
ters of h i s fi rst nove l , A/mayer's Folly, written i n an awkward but powerfu l
Engl ish . My read i n g of Heart of Darkness embraces a com plex decade of
choice, the 1 890s, begi n n i ng with the African voyage and end i ng with
its na rration . The choice i nvolved career, language, and cu l tu ra l attach-

3. The i n terpretation I suggest owes a good dea l to previous expl icators of

Con rad, most notably Edwa rd Said and lan Watt. In its b iographical d i mensions
it d raws on the standard works : Bai nes 1 960; Watt 1 9 79; Karl 1 9 79; and Najder
1 983 . My focus on Heart of Darkness as an al legory of writing and of grappling
with language and c u lture i n the i r emergent twentieth-century defi n itions is, I
bel ieve, a new one, but it d raws on many poi nts wel l establ ished i n Con rad
stud ies. I have not cited specific sources for biograph ical facts, si nce those I bu i ld
with are not, to my knowledge, disputed i n the l iterature.
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 97

ment. Ma l i nows k i 's para l l e l experience i s marked off by two works,

w h i c h may be treated as a s i ngle expanded text: A Diary in the Strict
Sense of the Term ( 1 967), h i s i nti mate Trobri a nd journal of 1 9 1 4 -1 9 1 8,
and the c l assic eth n ogra phy that emerged from the fieldwork, Argonauts
of the Western Pacific ( 1 9 2 2 ) .
A word o f method o l ogical caution i s needed a t t h e outset. To treat
the Diary and Argonauts together n eed not i m ply that the former is a true
revelation of M a l i nows ki's fieldwork. (Th i s i s how the Diary was widely
u nderstood on i ts pu b l i cation i n 1 96 7 . ) The Trobriand field experience is
not exhausted by Argonauts o r the Diary or by the i r comb i n ation . The
two texts a re pa rt i a l refractions, spec ific experi ments with writi ng. Re­
corded l a rgel y in Po l is h and c learly n ot i ntended for publ ication, the
Diary caused a m i no r scandal in the pu b l i c i mage of anthropol ogy­
a l though fiel dworkers recogn i zed m uc h that was fam i l iar. One of the
d i sc i p l i ne's fou nders was see n to have fe l t considerable anger toward h i s
n ative i n formants. A field experience that h ad set the standard for scien­
tific c u ltural desc r i ption was fra ught with am bivalence. The authoritat ive
anthropo l og i st i n h i s i nti mate journal appeared a self-absorbed hypo­
chondriac, freq uently depressed , prey to constant fantas ies about E u ro­
pean and Trobriand women, trapped in a n end l ess struggle to mainta i n
h i s mora l e , t o pu l l h i mse l f together. H e was merc u r i a l , try i n g o u t d iffer­
ent vo i ces, personae. The a n g u i s h , confusion, e l at i o n , and anger of the
Diary seemed to leave l i tt l e room for the stable, comprehend i ng posture
of rel ativist eth nography. Moreover, i n its rawness and v u l nerab i l i ty, its
u nq uestionable s i ncerity and i nconc l u s i veness, the Diary seemed to de­
l i ver a n u nvarn i shed rea l i ty. But it is o n l y one i m po rtant version of a
com p l ex, i nters u bj ective s i tuation (wh i c h a l so p rod uced Argonauts and
other eth nogra p h i c and popu l a r accou nts) . The Diary i s an i nventive,
po lyphonic text . I t i s a c ru c i a l docu ment for the h i story of anthropol ogy,
not because it revea l s the rea l i ty of eth nographic experience but because
it forces us to grapple with the com plexities of such encou nters and to
treat a l l textu a l accou n ts based o n fieldwork as part i a l constructions.4

Mal i nowski and Con rad knew eac h other, and there i s evidence from
Ma l i nows k i 's comments on the older, a l ready wel l known writer that he

4. I j u xtapose Argonauts and the Diary to highl ight a critical disc repancy
between the two best-known accou nts of Ma l i nowski's research process. At ti mes

sensed a deep affin ity i n their pred icaments . With reason : both were
Poles condemned by h i storical contingency to a cosmopo l itan E u ropean
identity ; both purs ued ambitious writing careers in England . Drawing on
Zdz islaw Najder's excel lent stud ies of Conrad, one can spec u late that
the two ex i les shared a pecu l iarly Pol ish c u l tu ra l d i sta nce, h aving been
born i nto a nation that had si nce the eighteenth centu ry existed o n l y as
a fiction-but an i ntensely bel ieved , serious fiction -of col lective iden­
tity. Moreover, Poland's pec u l iar social structu re, with its broad ly based
smal l nobi l i ty, made aristocratic va l ues u n usua l ly evident at a l l levels of
soc iety. Poland's c u l tivated ex i les were not l i kely to be charmed by Eu­
rope's reign i ng bou rgeois va l ues; they would keep a certa i n remove . This
viewpo i nt outside bou rgeois soc iety (but mai ntai ned with a degree of
artifice-rather l i ke Balzac's stand point in the France of the 1 830s) is
perhaps a pec u l iarly adva ntageous "eth nograph ic" position . Be that as it
may, there i s no doubt about Ma l i nowski's strong affi n ity for Con rad . (J ust
before the war he prese nted the older man with a copy of his fi rst book,
The Fa mily among the Austra lian Aborigines, with a Po l i sh i nscri ption ;
what Con rad made of Aru nta notions of patern ity remains, perhaps for­
tunately, u n known . ) Although the i r acq uai ntance was brief, Ma l i nowski
often represented his l ife i n Con rad ian terms, and i n his diary he seemed
at ti mes to be rewriting themes from Heart of Darkness.
Nearly every commentator on the Diary has plausibly compared it
with Con rad 's African ta l e (see for exam ple Stocking 1 9 7 4). Both Heart
of Darkness and the Diary a ppear to portray the crisis of an identity-a
struggle at the l i m its of Western c i v i l ization against the th reat of moral
d isso l ution . I ndeed this struggle, and the need for personal restra i nt, is a
com monplace of col o n i a l l iteratu re . Thus the para l lel is not particu l arly
revea l i ng, beyond showing l i fe (the Diary) i m i tati ng " l iteratu re" (Heart of
Darkness) . I n addition to Ku rtz's mora l d isi ntegration, however, Con rad

I oversimpl ify the cou rse of Mal i nowski's research and writing; the Diary actual l y
covers work done i n both t h e Trobriands and Mai l u . By concentrating on two
texts I ignore other com p l i cati ng ones, most notably certai n unpubl ished and
currently u navai lable diaries, along with Ma l i nowski's "Natives of Mai l u " ( 1 9 1 5)
and "Baloma: The Spi rits of the Dead i n the Trobriand Isl ands" ( 1 9 1 6). I n these
last two works he can be seen working out the personal and scientific eth no­
graph ic style that achieves fu l l expression in Argona uts . A biograph ical account,
or a thorough portrayal of Ma l i nowski's fieldwork, or a depiction of Melanesian
culture and hi story each wou l d select a d ifferent corpus. Moreover, by stoppi ng
at 1 922 I neglect Ma l i nowski's ongoing rewriting of the dia logue with the Trobri­
ands. I n im portant ways his last major monograph, Coral Gardens and Their
Magic ( 1 935), experi menta l l y and self-critically questions the rhetorical stance
constructed in Argonauts .
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 99

i ntroduces a more p rofound, s u bvers ive theme : the famous " l ie" -actu­
a l l y a series of l ies that in Heart of Darkness both u nderm i ne and some­
how e mpower the complex truth of Marlow's narration . The most prom­
i nent of these l ies is, of cou rse, Marl ow's refusal to tel l Ku rtz's I ntended
h i s l ast words, "The Horro r," su bstituting i n stead words she can accept.
T h i s l ie i s then j uxtaposed with the truth -also h ighly c i rc umstanti a l ­
to l d t o a restri cted gro u p o f Engl i s h men on t h e deck o f t h e cru i s i n g yawl
Nellie. Mal i nows k i 's u n settled Diary does seem to enact the theme of
d i s i ntegration . B ut what of the l ie? The a l l -too-bel ievable accou nt? Mal­
i nowsk i 's sav i n g fictio n , I wi l l argue, i s the c l assic eth nography Argo­
nauts of the Western Pacific.
Heart of Darkness is notoriously i nterpretable; but one of its ines­
capab l e themes is the problem of truth-speaki ng, the interplay of truth
and l ie in Marlow's d i scourse. The l ie to Ku rtz's I ntended has been ex­
haustive l y debated . Very schematica l ly, my own position is that the l ie is
a saving l ie . In sparing the I ntended Ku rtz's l ast words, Marlow recog­
n izes a n d constitutes d i fferent dom a i ns of truth-male and female as
wel l as the truths of the m etropole and the frontier. These truths reflect
elementary structu res in the constitution of o rdered mea n i ngs-knowl­
edge d ivided by gender and by c u ltura l center and periphery. The l ie to
the I ntended i s j u xtaposed with a d i fferent truth (and it too is l i m i ted ,
contextua l , and problematic) tol d on the deck of the Nellie to Engl i sh m en
identified on l y as soc i a l types-the Lawyer, the Accou ntant, the D i rector
of Com pan ies. If Marlow succeeds i n com m u n icat i n g, it is with i n this
l i m ited doma i n . As readers , however, we identify with the u n identified
person who watches Marlow's dark truths and wh ite l ies enacted on the
stage of the yaw l 's dec k . This second narrator's story is not itself u nder­
m i ned or l i m ited . It represents, I propose, the eth nographic standpoi nt,
a s u bjective position and a h i storical site of narrative authority that truth­
fu l l y j u xtaposes d i fferent truth s . Wh i l e Marlow i n itial l y "abhors a l ie," he
learns to l ie-that i s, to com m u n i cate with i n the col l ecti ve, partial fic­
tions of c u ltu ra l l ife. He te l l s l i m i ted stories . The second narrator sal­
vages, com pares, and ( i ronical l y) be l ieves these staged truths. This i s the
ach i eved perspective of the serious i nterpreter of c u ltu res, of local , pa r­
tial knowledge. The voice of Con rad 's "outermost" narrator is a stabi l iz­
i n g voi ce whose words are not meant to be m i strusted . 5

5 . For a read i ng c lose to my own but with a d i fferent overal l emphasis see
). H i l l is M i l ler 1 96 5 . Here we fi nd strong argu ments for seeing Heart of Darkness
not as a positive choice for the " l ie of cu lture" but as someth ing that u nderm i nes

Heart of Darkness offers, then, a parad igm of ethnographic subjec­

tivity. In what fol l ows I wi l l be exploring spec ific echoes and analogies
l i n k i ng Conrad's situation of c u ltura l l i m i n a l ity i n the Congo with Mal i ­
nowski's i n the Trobriands. T h e correspondence i s not exact, however.
Perhaps the most i m porta nt textual d i fference is that Con rad takes an
i ron ic position with respect to representational truth , a stance only i m­
pl icit i n Mal i nowski's writing. The a uthor of Argonauts devotes h i mself
to constructing rea l istic c u ltura l fictions, whereas Conrad, though s i m i ­
l a r l y comm itted , represents the activity as a contextua l ly l i m ited practice
of storyte l l i n g . 6

a l l truth, a more tragic, dark, u lti mate l y n i h i l istic text. U ndoubted ly i n both form
and content the tale grapples with n i h i l ism . Nonetheless, it does d ramatize the
successfu l construction of a fiction, a contingent, u nderm i ned , but fina l l y potent
story, a mean i ngfu l economy of truths and l ies. B iograph ical evidence rei nforces
my suggestion that Heart of Darkness is a story of qual ified but d i sti nct success
in truth-tel l i ng. I have al ready noted that the tale was written j u st when Conrad
fi n a l l y decided to stake everything on his career of writing in Engl ish. In the
autumn of 1 898 he left Essex and the Thames estuary (the place between land
and sea) for Kent to res ide near other writers- H . G . Wel l s, Stephen Crane, Ford
Maddox Ford , Henry James. The move, immed iately fol l owed by h i s last re­
corded search for a maritime post, i naugurated h i s most prod uctive years of l it­
erary work. A serious writing block was broken ; Heart of Darkness emerged i n
an u ncharacteristic rush . F rom th is standpoint o f decision t h e tale reaches back
a decade to the begi n n i n g of Korzen iowski's turn to writing, when, in the Congo,
h i s luggage contained the first chapters of A/mayer's Folly. I n the read ing I am
sketc h i ng out Heart of Darkness is centra l l y about writing, about tel l i ng the truth
i n its most a l ienated , nondia logical form . Conrad does succeed in becom ing an
English writer, a l i m ited truth-te l ler. It is not surprisi ng, then, that in the b l u rred
cacophony of the j u ngle Marlow yearns for Engl ish words. Ku rtz was partly ed­
ucated in Brita i n , and h i s mother, we reca l l , was half-Engl ish . From the begin­
n i ng Marlow searches for Ku rtz's i ntimate and elemental voice; and i n the end
"th is i n itiated wra ith from the back of Nowhere honoured me with its amazing
confidence before i t van ished altogether. This was because it could speak English
to me" (p. 50) . I can not here d iscuss the many com plexities i n the staging and
va l u i ng of d i fferent languages in Heart of Darkness.
6. I n Reading for the Plot ( 1 984 : 25 9-260) Peter Brooks nicely observes that
Heart of Darkness presents its truth as a "narrative transaction" rather than a
"summing up" (as i n Ku rtz's last words). Mean ing in the narrative is not a re­
vealed kernel ; it exists outside, dia logica l l y, in specific transmissions; it is "lo­
cated in the i n terstices of story and frame, born of the relationship between tel lers
and l i steners." In stressing the ta le's " i nterm inable analysis" B rooks m i n i m izes the
fi rst narrator's stab i l izing fu nction as a spec ial l i stener (reader), not named or
given a l i m ited c u l tu ra l fu nction l i ke the others on the deck. This l istener's invis­
i b i l ity guarantees a certai n i ronic authority, the possibi l ity of see i ng and not bei ng
seen, of speaking without contrad iction about relative truths, or deciding thei r
u ndecidabi l ity.
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 101

I n com paring the experiences of Mal i nowski and Con rad , one is
struc k by the i r l i ngu i stic overdeterm i natio n . I n each case th ree languages
are at work, p rod u c i ng constant trans l ation and interference. Conrad's
pred icament is extreme l y complex. J u st before leav i ng for Africa he h ad
u naccou ntably begu n writi ng what wou ld become A/mayer's Fo lly. After
com pos ing the ope n i ng chapters, he ran i nto obstacles . Around th is time
he came to know a cou s i n by ma rriage, Marguerite Poradowska, with
whom he became i n some significant way amorously i nvol ved . She was
married and a wel l-known French author; it was l a rge ly a l iterary en­
tanglement. Con rad wrote her rather pass ionate and self-revelatory let­
ters- i n F rench . Poradowska, who l ived i n B russels, was i n stru menta l i n
a rranging h e r ki nsman's Congo employment. Then, i n the months j ust
before he left for Africa, Con rad retu rned to Po land for the first time si nce
he had run away to sea fifteen years before . T h i s renewed h i s Pol is h ,
w h i c h h a d rem a i ned good, and revived its assoc iation w i t h c h i ldhood
p l aces and a m bivalent fee l i ngs . F rom Pol and (actu a l ly the Russian
U krai ne) he rushed a l most d i rectl y to take u p h i s post i n the Congo .
There h e spoke Frenc h , h i s most fl uent acq u i red tongue, b u t kept a d i ary
i n Engl ish and may have worked on the chapters of A/ma yer. ( H e c l a i m s
a s m u c h i n h i s " B i ographical N ote" of 1 900 . ) I n Africa Con rad estab­
l ished a friends h i p with the I rishman Roger Casement and genera l l y
m a i nta i ned a pose o f a n Engl ish nautical gentleman . H is i ntense letters
to Poradowska conti n ued , as a l ways, i n F ren c h . H is mother tongue h ad
j u st been revived . The Congo experience was a ti me of m ax i mal l i ngu is­
tic complexity. In what language was Con rad consistently t h i n k i ng? It is
not surprising that words and t h i ngs often seem d isjointed in Heart of
Darkness as Ma rlow searches in the dark for mea n i ng and i nterlocutio n .
As for Mal i nows k i , i n t h e field he kept h i s private d iary i n Pol ish and
corresponded i n that langu age with h i s mother, who was beh i nd enemy
l i nes i n Austr i a . H e wrote i n Engl i s h on anthropological topics to h i s
p rofessor, C . G . Sel igman, i n London . To h i s fiancee, " E . R . M ." ( E l sie R .
Masson), i n Austra l i a he wrote frequently, a l so i n Engl ish . There were,
however, at l east two other women, old flames, on h is m i nd , at least one
of them associ ated with Pol a n d . H is most i nti mate Pol is h friend, Sta n i s l as
Witkiew i cz ("Stas" i n the Diary), soon to become a major ava nt-garde
a rtist and writer, a l so hau nted h i s consc iou sness. The two h ad traveled
together to the Pac i fic and had fa l l en out j u st before Ma l i nowski's Trobri­
and sojourns. H e yearned to set thi ngs right, but his friend was now in
Russia . These powerfu l Engl i s h and Pol i sh associations were i n te rru pted

by a th i rd l i ngu i stica l l y coded world, the Trobriand u n i verse, i n which

he had to l ive and work prod uctively. Ma l i nowski's d a i l y transactions
with Trobria nders were cond ucted i n Kiriwi n ian, and i n time his field­
notes were recorded l a rgely i n the vernacu l ar. 7
We can suggest a tentati ve structu re for the th ree active languages
of Con rad's and Mal i nowski's exoti c experiences . Between Pol ish, the
mother tongue, and Engl i s h , the language of future career and marriage,
a th i rd i ntervenes, assoc i ated with eroticism and vio lence . Con rad's
F rench i s l i n ked with Poradowska, a problematic love object (she was
both too i ntim idating and too i nti mate) ; French is a l so l i n ked with Con­
rad's rec kless youth in Marse i l le and with the I m perial Congo, wh ich
Con rad abhorred for its viol ence and rapaci ty. Ma l i nowski's i nterfering
l angu age was Kiriw i n i a n , assoc iated with a certai n exuberance and l ud i c
excess (wh ich M a l i nowsk i enjoyed a n d portrayed sympathetica l l y i n h i s
accounts o f K u l a rituals a n d sexual c ustoms) a n d a l so with the erotic
tem ptations of Trobriand women . The Diary struggles repeated l y with
th i s K i ri w i n i a n rea l m of des i re .
S o i t i s poss i ble to d i stinguish i n each case a mother tongue, a lan­
guage of excess, and a l anguage of restra int (of marri age and authorsh i p) .
This i s surely too neat. The languages wou ld have interpenetrated and
i nterfered in h igh l y conti ngent ways ; but enough has been said, perhaps,
to make the m a i n poi nt. Both Con rad in the Congo and Mal i nowski in
the Trobriands were e nmeshed in complex, contrad ictory subj ective sit­
uations articu l ated at the leve ls of language, des i re, and c u l tu ra l affi l ia­
tion .

I n both Heart of Darkness and the Diary we see the crisis of a self at some
"fu rthest po i nt of n avigation ." Both works render an experience of lone­
l i ness, but one that is fi l led with other people and with other accents and
that does not perm it a feel i n g of centered ness, coherent d ia logue, or au­
thentic com m u n io n . In Con rad's Congo h is fe l low wh ites are d u p l i c itous
and u ncontro l led . The j u ngle i s cacophonous, fi l led with too many

7 . The " Pol ish" d i a ry is extraord i nari ly heteroglot. Mario Bick ( 1 967: 299),
whose task was to com p i l e a glossary and genera l l y to "sort out the l i ngu istic
melange," spec ifies that Mal i nowski wrote "in Po l i sh with freq uent u se of En­
glish, words and ph rases in German, French, Greek, Spanish, and Latin, and of
cou rse terms from the native languages" (there were fou r : Motu, Ma i l u , Kiriwin­
ian, and Pidgi n ) .
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 1 03

voices-therefore m u te, i n co herent. Mal i nowski was not, of cou rse, iso­
lated in the Trobriands, either from natives or from local whites . But the
Diary is an u nsta b l e confusion of other voices and worl d s : mother, l ov­
ers, fiancee, best friend, Trobrianders, loca l m i ss ionaries, traders, as wel l
as the escapist u n i ve rses, the nove l s h e can never resi st . Most fieldwork­
ers w i l l recogn ize th is m u l tivoca l pred i cament. But Ma l i nowski experi­
ences (or at l east h i s Diary portrays) somet h i n g l i ke a real spi ritu a l and
emotional c r i s i s : each of the vo ices represents a tem ptation ; he is p u l led
too many ways . Thus, l i ke Marlow in Heart of Darkness, Ma l i nowski
c l i ngs to his work routi nes, his exerc ises, and his d iary-where con­
fused l y, bare l y, h e brings h i s d i vergent worlds and desi res together.
A passage from the Diary wi l l i l l ustrate h i s pred icament:

7 . 1 8 . 1 8 . . . On the theory of religion . My eth ical position i n rel ation

to Mother, Stas, E . R. M . Twi n ges of conscience resu lt from lack of
integrated fee l i ngs and truth in rel ation to ind ividua l s . My whole eth ics
is based on the fu ndamental insti nct of u n ified personality. From this
fol lows the need to be the same i n d i fferent situations (truth i n rel ation
to oneself) and the need, i n d i spensa b i l ity, of si ncerity: the whole val u e
o f friendsh i p i s based on t h e poss i b i l ity o f expressi n g oneself, o f being
oneself with complete fra nkness. Alternative between a l ie and spo i l ­
i n g a relationsh i p . ( M y attitude t o Mother, Stas, and a l l my friends was
stra i ned . ) Love does not flow from eth ics, but eth ics from love. There
is no way of ded u c i n g Ch ristian eth ics from my theory. But that eth ics
has never expressed the actual truth-love you r neighbor-to the de­
gree actu a l l y possible. The real problem is: why must you a lways be­
have as if God were watc h i n g you ? (pp. 2 96-297)

The passage i s confu sed ; but we can extract perhaps the central i ssue on
which it turns : the i m poss i b i l ity of bei n g si ncere and thus of havi ng an
eth ical center. Ma l i nowski feel s the req u i rement of personal coherence.
A p u n itive God i s watc h i n g his every (i nconsi stent) move. He is thus not
free to adopt d i fferent personae in d i fferent situations. He suffers from the
fact that th i s r u l e of si ncerity, an eth ics of u n i fied persona l i ty, mea ns that
he wi l l have to be u n p l easantly truthfu l to va rious friends and lovers. And
th i s wi l l mean-has a l ready meant- los i n g friend s : "Alternative be­
tween a l i e and spo i l i ng a re l ations h i p ."
There i s no way out. There m u st be a way out. Too much truth­
tel l i ng u nderm i nes the com prom i ses of co l lective l i fe . Ma l i nowski's so­
l ution consists of construct i n g two rel ated fictions-of a self and of a

c u lture . Although my task here is ne ither psycho logica l nor biographica l ,

l et m e s i m p l y suggest that the personal style-extravagant, operatic­
that both charmed and i rritated Mal i nowski's contem poraries was a re­
sponse to th is d i lemma. He indu l ged in "Slavic" extremism; h i s revela­
tions about h i mself and his work were exaggerated and ambiguously pa­
rod ic. He wou ld stri ke poses (he c l a i med to have s i ngle-handed l y
invented "The F u nctional Method"), chal lenging the l iteral-mi nded to
see that these persona l truths were i n some degree fictions. H i s character
was staged but a l so truthfu l , a pose but nonetheless authenti c . One of
the ways Ma l i nowski p u l led h i mself together was by writing eth nogra­
phy. Here the fash ioned wholes of a sel f and of a cu ltu re seem to be
mutua l l y rei n forc i ng a l l egories of identity. An essay by Harry Payne,
"Ma l i nowski's Style" ( 1 9 8 1 ), suggestive l y traces the complex combi na­
tion of authority and fictiona l ity that the na rrative form of Argonauts en­
acts : "With i n th e i mmense latitude of [ its] structu re Ma l i nowski can de­
term i ne s h i fts in focus, tone, and objects; the cyc l ical th read wi l l always
provide a place of return. F u nctional therapy � cts only heu ristica l l y.
Si nce everyth i ng adheres to everyth i ng, one can wander without ever
getting fu l l y d i sconnected" (p. 438) . 8 The l iterary problem of a uthorial
poi nt of view, the Jamesian req u i rement that every nove l reflect a "con­
tro l l i ng i nte l l igence," was a painfu l personal problem for the Trobriand
d i arist. The ample, m u l t i perspectiva l , meandering structu re of Argonauts
reso l ves th is crisis of s i ncerity. I n effect, as the scientific, persuasive au­
thor of th is fiction, Mal i nowski can be l i ke Flau bert's God, omni present
in the text, arranging enthus iastic descriptions, scientific exp lanations,
enactments of events from d i fferent standpoi nts, personal confess ions,
and so fort h .
C u l tu ra l descri ptions i n Mal i nowski's style o f functiona l ism strove
for a ki nd of u n i fied persona l ity, but a convincing tota l ization a l ways
escaped them . Mal i nows ki never d i d pu l l together Trobriand c u ltu re; he
prod uced no synthetic portra it, only densely contextua l i zed monographs
on i m po rtant i n stitutions. Moreover, his obsessive i n c l u s ion of data, "im­
ponderabi l ia," and vernac u l ar texts may be seen as a des i re to u n make
as wel l as to make a whole; such add itive, metonym ic empiricism un­
derm i nes the construction of functiona l , synecdoc h i c representations.
Ma l i nowski's ethnograph ies- u n l i ke Rad c l i ffe-B rown's spare, analytic,
fu nctional portra its-were m u ltifarious, loose, but rhetorica l ly success-

8. There is an interesting s l i p between th is passage and its footnote: func­

tional ist "therapy" becomes fu nctional ist "theory."
O N E T H N O G R A P H I C S E L F - FA S H I ON I N G 105

fu l narrative form s ( Payne 1 98 1 :420-42 1 ) . F i ctional express ions o f a c u l ­

ture a n d o f a s u bjectivity, they provided a way out o f t h e b i n d o f s i ncerity
and wholeness, the Con rad i an problematic of the l ie at i ssue i n the Diary.
There are more specific echoes of Heart of Darkness in Mal i nows k i 's
i nt i m ate Pol is h text. Spea k i n g of h i s Trobriand i nformants who wi l l not
cooperate with h i s researc h , he damns them i n Ku rtz's term s : "At mo­
ments I was fu rious at them , partic u l arly because after I gave them their
portions of tobacco they all went away. On the whole my feel i ngs toward
the natives are decided ly te n d i n g to 'Exterminate the brutes, ' " ( p . 69).
Mal i n owski fl i rted with various colon i a l wh ite ro les- Kurtz-l i ke excess
i n c l uded . H e re the i ron ic i nvocation provides h i m with a fiction a l grasp
of the stresses of fie l dwork and the violence of h i s feel i ngs. I n the Diary,
l i ke Marlow i n h i s ambivalent dou b l i n g with Kurtz , Ma l i nowsk i often
faces the i nsepara b i l ity of d i scou rse and power. He m ust struggle for con­
tro l in the eth nogra p h i c encou nter.
Another non i ro n i c echo of Heart of Darkness is heard i n M a l i now­
ski's grief-stricken response to the news of h i s mother's death, which shat­
ters the Diary's l a st pages : "The terrible mystery that su rrounds the death
of someone dear, close to you . The u nspoken l ast word-someth i ng that
was to cast l ight is b u ried, the rest of l ife l ies half h idden i n da rkness" (p.
2 9 3 ) . Ma l i nowsk i fee ls he has been denied Marlow's rescued ta l i sman,
a n ambiguously i l l u m i nati ng, potent l ast word breathed i n the moment
of death .
Beyond the more-or- l ess d i rect citations i n the Diary one notes a l so
more genera l thematic and structu ra l para l lels with Heart of Darkness .
Both books are reco rds of wh ite men at the frontier, at poi nts of danger
and d i s i n tegration . In both sexua l ity i s at issue : both portray an other that
is conventiona l ly fem i n ized, at once a danger and a tem ptation . Fem i­
n i ne figu res in the two texts fa l l i nto either spi ritual (soft) or sensual (hard)
categories . There is a common thematization of the pu l l of desi re or ex­
cess ba re l y chec ked by some crucial restrai nt . For Ma l i nowski the re­
stra i nt is em bod ied by· h i s fiancee, l i n ked i n h i s m i nd to an Engl ish aca­
dem ic career, to an e levated love, and to marri age. "Thought of E. R. M .
. . ." is the Diary's censor for l ascivious thoughts about women, native
or wh ite : " I m u st not betray E. R . M . menta l l y, i . e . , recal l my previous
relations with women, o r th i n k about future ones . . . Preserve the essen­
tial i nner personal ity t h rough a l l d ifficu lties and vicissitudes : I m u st never
sacrifice moral pri n c i p l es o r essential work to ' posi n g' to convivial Stim ­
mung, etc . My m a i n task m u st be work. E rgo : work ! " (p. 268).
L i ke Con rad 's protago n i st, the eth nographer strugg les constantly to
1 06 D I S C O UR S E S

mainta i n an essenti a l i n ner self-rel iance- h i s "own true stuff," as Marlow

puts it. The pu l l of dangerous others, the d i s i ntegrati n g frontier, is resisted
by method ic a l , d i sc i p l i ned work. For Marlow obsessive attentions to h i s
steamboat a n d i t s navigation provide t h e "su rface wisdom" needed to
hold h i s personal ity i n p l ace. As invoked in the Diary, Ma l i nowsk i 's
scientific l a bors serve a s i m i lar pu rpose. Restrai ned , eth ical personal ity
is relentlessly ach i eved th rough work. Th i s structu re of fee l ing can be
located with some prec i sion in the h istorical pred icament of l ate Vi cto­
rian h igh colon i a l society, and it is close ly re l ated to the emergence of
eth nographic c u lture .
Victorian soc i a l critics d i scerned a pervasive crisis for which Ma­
thew Arnold's title Cu lt u re and Anarchy provided the bas ic d i agnosi s :
aga i n st t h e fragmentation o f modern l ife stood t h e order a n d wholeness
of c u lture. Raymond Wi l l iams ( 1 966) has offered a subtle account of
these h u m a n i st responses to the u n precedented tech nological and ideo­
logical tra nsformations at work in the m id-ni neteenth centu ry. George
E l iot's stra nge affi rmation is characteristi c : of the three words "God ," "Im­
morta l ity," and " Duty," she pronounced , "with terri ble earnestness, how
i nconceivable was the first, how unbel ievable was the second, and yet
how perempto ry and abso l ute the third!" (quoted in Houghton 1 95 7 : 4 3 ) .
D uty had become a del i berate bel i ef, a w i l led fide l i ty to aspects o f con­
vention and to work (Carlyl e's so l ution) . lan Watt has persuasively tied
Con rad to t h i s response ( 1 9 7 9 : 1 48 -1 5 1 ) . Marlow, in the m idd le of Af­
rica, c l i ngs for dear l ife to h i s steam boat, to the routi ne duties of its mai n­
tenance and navigation . And the structu re pers ists in Ma l i nowski's Diary,
with its constant se lf-exhortati ons to avo id loose d i stractions and get
down to work. I n the c u lture and anarchy problematic (wh ich persi sts i n
p l u ra l , anth ropologi cal cu ltu re concepts that privi lege order and system
ove r d i sorder and confl ict), personal and col lective essences must con­
tinuous l y be maintained. The ethnograph i c standpoint that concerns us
here stands h a l f-outside these processes, observing their loca l , arbitrary,
but i n d ispensable worki ngs .
Cu ltu re, a col lect ive fiction, is the ground for i nd ividual identity and
freedom . The self, Marlow's "own true stuff," is a product of work, an
ideo logica l construction that is nonetheless essenti a l , the fou ndation of
eth ics. But once c u lture becomes visible as an object and ground, a sys­
tem of mean i n g among others, the ethnograph ic self can no longer take
root i n u n med iated identity. Edward Sa id has said of Conrad that h i s prin­
c i pal struggle, reflected i n his writing, was "the ach ievement of charac­
ter" ( 1 9 6 6 : 1 3 ) . I ndeed he reconstructed h i mself q u i te carefu l l y in the
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 107

persona o f a n "Engl ish" author, t h e character w h o speaks i n the "Author's

Note" he wou l d l ater add to each of h i s works. T h i s construction of a se lf
was both artific i a l and dead l y serious. (We can see the process parod i ed
by the accou ntant i n Heart of Darkness, who seems l itera l ly to be held
together by h i s rid i c u lously forma l but somehow adm i rable get- u p . ) Al l
of th i s gives spec i a l poignancy to the sentence that ends the publ ished
Diary: "Tru l y I lack real character."

Ma l i nowsk i d id , howeve r, rescue a self from the d i s i ntegration and

depress ion . That self was to be tied , l i ke Con rad's, to the process of writ­
ing. I n th i s context it is worth explori ng another region of s i m i l arity be­
tween the Diary and Heart of Darkness : The ro le of i n congruous written
texts. The fragmented subj ectivity mani fested i n both works is that of a
writer, and the pu l l of d i fferent des i res and languages is man i fested i n a
n u m ber of d i screpa nt i nsc r i ptions. The most famous exam ple i n Heart of
Darkness is K u rtz's strident essay on the s u ppression of sa"l{age customs,
abruptly canceled by his own scrawled comment, " Exterm i n ate a l l the
brutes." B ut another eq u a l l y s i g n i ficant text loose in Con rad's j ungle i s a
strange book Marlow d iscovers on one of o n l y two peri lous departu res
from the deck of h i s steam boat (on the other he wrestles Kurtz back from
the w i lderness) . I n a shack by the riverba n k he fa l l s i nto an al most mys­
tical reve r i e :

There remai ned a r u d e table-a plank on two posts; a heap o f rubbish

reposed i n a dark corner, and by the door I picked u p a book. It had
lost its covers, and the pages had been th u m bed i n to a state of ex­
treme l y d i rty softness; but the back h ad been lovingly stitched afresh
with white cotton thread, w h i c h looked c lean yet. It was an extraor­
d i nary fi nd . Its title was, An Inquiry Into Some Points of Seamanship,
by a man Towser, Towson-some such name-Master i n H i s Majesty's
Navy. The m atter looked d reary read i n g enough , with i l l u strative d i a­
grams and repu l sive tables of figures, and the copy was si xty years o l d .
I hand led this a m a z i n g antiqu ity w i t h t h e greatest possible tenderness,
lest it shou ld d i ssolve in my hands. With i n , Towson or Towser was
i nq u i ring earnestly i nto the breaking stra i n of s h i ps' chains and tackle,
and other such matters. Not a very enth ra l l i n g book; but at the fi rst
glance you cou l d see there a singleness of intention, an honest con­
cern for the h u mble pages, thought out so many years ago, l u m inous
with another than a professional l ight. The simple old sa i l or, with his

ta l k of chains and purchases, made me forget the j u ngle and the pil­
grims in a del icious sensation of having come u pon someth ing u n m i s­
takably rea l . Such a book being there was wonderfu l enough ; but sti l l
more astounding were the notes penci l led in the margi n , and plai n l y
referri ng t o t h e text. I cou ldn't bel ieve my eyes ! They were i n c i pher!
Yes, it looked l i ke c i pher. Fancy a man l uggi ng with him a book of that
description into th is nowhere and studying it-in ci pher at that ! I t was
an extraagant mystery. (pp. 3 8 -39)9

The passage has re l i gious overtones-a m i racu lous rel ic, an abrupt
movement i n i m agery from d i rt and decay to transcendence and l i ght
and thence i nto myste ry, the naive witness i n g of a moment of faith . We
m ust be carefu l not to i nterpret the Inquiry's appea l to Marlow s i m p l y as
nosta lgia for the sea , though that is part of its charm . The Russian "har­
leq u i n " who turns out to be the book's owner seems to read the treatise
pri mari ly in t h i s way; for he takes carefu l notes, presumably on the
book's content, as i f he were studying seamansh i p . For Marlow, however,
the i nspiration of the book proceeds i n some way d i rectly out of the
writ i ng itse lf, wh ich, tra nscend i ng the c h a i n s and s h i ps and tackle, is
" l u m i nous with another than a professiona l l ight." Marlow heeds not the
content but the language . H e is i nterested i n the old sai lor's pai nstaking
craft; h i s way of making the book and his "ta l k" seem conc rete-even to
the abstract n u merical tab l es.
What charms Marlow is not primari l y the poss i b i l ity of s i ncere au­
thors h i p . The old sa lt, "Towser or Towson-some such name-Master i n
H i s Maj esty's N avy," i s persona l l y e l usive ; i t is not h i s bei ng that cou nts
but h i s language. The man seems to d i ssolve i nto vague typical ity; what
matters is h i s p l a i n Engl i s h . Significantly, though, the text fa i l s to u n i te its
two equ a l ly devout readers; for when they fi nal ly meet, the Russian is
overjoyed to greet a fel low seaman, whereas Marlow is disappoi nted not
to fi nd an Engl i s h m a n . Readers h i p is in q uestion . The same physical
book provokes d i fferent, eq ua l l y reverent reactions. I can not expl ore
here the biograph ical significance of the disj u nctu re : Con rad had j ust
shed h i s offic i a l Russian citizensh i p for British national ity, and arguably
the harleq u i n is connected with the young wanderer, Korzen iowski, who
was beco m i n g Con rad . It is enough to notice the rad ical relativity : the
d i stance between two read i ngs. The "cipher" makes the point graph i ­
ca l ly, a n d if t h e margi na l i a t u r n o u t later t o b e i n a European language,

9. Page references here and el sewhere are from the 1 9 7 1 Norton ed ition .
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 1 09

that i n no way d i m i n ishes the graph i c i m age of a separateness. (It is rem­

i n i scent of the sense of unease one experiences on fi nd i ng strange ma rk­
i ngs in a book and then recog n i z i n g hav i n g made them oneself-another
person - i n a previous read i n g . )
What pers ists is t h e text itsel f-bare ly. Worn b y th u m b i n g a n d cut
loose from its covers-wh i c h may symbo l ize the context of its origi nal
p u b l i cation-the written text m ust resi st decay as it travel s through space
and ti me. After s i xty years-a h u man l ifeti me-the moment of d i s i nte­
gration has come. The author's creation faces ob l ivion, but a reader
stitches the pages lov i ngly back together. Then the book is abandoned to
its death somewhere on a strange conti nent, its nautical content run
aground in the a bsence of context-and once more a reader to the res­
cue. Resc ue is one of Con rad's key i mages for h i s work; the act of writi ng
a l ways reac hes toward rescue in an i magi ned act of read i ng. S i gn ifi­
cantly the text that mea ns most in Heart of Darkness i s the one with the
least reference to the situation at hand .
Mal i nowsk i 's fieldwork experience is fi l l ed w i th d iscrepant i n scrip­
tions: h is deta i led field notes, written i n Engl i s h and K i ri w i n i a n ; vernac­
u lar texts, often recorded on the back of l etters from abroad ; h i s Pol ish
(actu a l l y heteroglot) d i ary; the m u lti l i ngual correspondence; and fi na l l y
a corpus worth l i nge r i n g on for a moment, the nove l s he can not res ist.
These l ast conta i n whole narrated worlds that seem at ti mes more rea l
( i n any case more des i rable) than the day-to-day busi ness of research ,
with its m a n y i ncom p l ete, contrad ictory notes, i m p ressions, data that
m u st be made to cohere . Ma l i nowsk i catches h i mself "escaping from"
Trobriand actu a l ity "to the company of Thackeray's London s nobs, fo l­
low i n g them eagerly around the streets of the big city." (The escapist
read ing of eth nographers i n the fie l d may req u i re an essay of its own . )
Ma l i nowsk i 's nove l s suggest a revea l i ng though i m perfect para l lel
with Towser's Inquiry-another wonderfu l l y compel l i ng fiction i n the
m idst of confu s i n g experience. Towser's book shows the poss i b i l ity of
persona l l y and a uthenti ca l ly s peaki ng the truth ; and it poi nts toward writ­
i ng (a m i racu lous presence i n absence) as salvation . B u t Towser is a l so a
tem ptation, l i ke Ma l i nowski's novels, d rawing Marlow away from h i s
work, h i s steam boat, into a k i nd o f vertiginous reverie. S u c h read i ngs are
des i red comm u n ions, pl aces where a coherent subjectivity can be recov­
ered in fictional identification with a whole voice or wor l d . Towser and
the novels do suggest a viable path beyond fragmentation, not for the
ch armed reader but for the hard-worki ng, constructive writer. For Mal i ­
nowski rescue l ies i n c reating rea l ist cu ltura l fictions, o f which Argonauts

i s h i s fi rst fu l l y rea l ized success . I n both nove l s and eth nograph ies the
se lf as a uthor stages the d i verse d i scou rses and scenes of a bel i evable
worl d .

The loose texts i n Heart o f Darkness a n d t h e Diary are scraps o f worlds;

l i ke fie l d notes they are i ncongruent. They must be made i nto a probable
portra it. To u n ify a messy scene of writi ng it i s necessary to se l ect, com­
bine, rewrite (and th u s efface) these texts. The resu lting true fictions for
Ma l i nowski are Argonauts and the whole series of Trobriand eth nogra­
ph ies, for Con rad A/mayer's Folly and the long process of learn i ng to
write Engl ish books, cu l m i nati ng in h i s fi rst masterpiece, Heart of Dark­
ness. Obviously these are d ifferent writing experiences : ethnographies
are both l i ke and u n l i ke nove l s . But i n an i m portant genera l way the two
experiences enact the process of fictional se lf-fash ioning in rel ative sys­
tems of c u lture and l a nguage that I ca l l eth nograph i c . Heart of Darkness
enacts and i ron i ca l l y ca l l s attention to the process. Argonauts is less re­
flexive, but it does both prod uce a cu ltura l fiction and an nounce the
emergence of an authoritative persona: B ron i s l aw Mal i nowski, new-sty le
anth ropologist. Th is persona, endowed with what Ma l inowski cal l ed
"the ethnographer's magic," a new kind of i nsight and experience, was
not, properly speaking, constructed in the fiel d . The persona does not
represent but rationa l i zes a resea rch experience . The Diary shows th i s
clearly, for t h e fie ldwork, l i ke most s i m i lar research, was ambiva lent and
u n r u l y. The confused s u bj ectivity it records i s sharply d i fferent from that
staged and recou nted in Argonauts . When the Diary was fi rst publ i shed
i n 1 967, the d iscrepancy was shocking, for the authoritative partici pant­
observer, a locus of sympathetic u nderstanding of the other, is s i m p l y not
visible in the Diary. Conversely, what is visible, a pronounced ambiva­
lence toward the Trobrianders, em pathy m i xed with desi re and aversion,
i s nowhere i n Argona uts, where com prehension, scru pulousness, and
generosity reign .
One is tem pted to propose that ethnograph ic comprehension (a co­
herent position of sym pathy and hermeneutic engagement) is better seen
as a creation of eth nograph ic writing than a consistent qual ity of eth no­
graphic experience. In any event what Ma l i nowsk i ach i eved in writi ng
was s i m u ltaneously ( 1 ) the fictional i n vention of the Trobri anders from a
mass of fie ld notes, documents, memories, and so forth , and (2) the con­
struction of a new publ i c figure, the anth ropologist as fieldworker, a per­
sona that wou l d be further elaborated by Margaret Mead and others . It is
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G I ll

worth noti ng that the persona of the pa rtici pant-observer anth ropologist
was not the professional i m age about which Ma l i nowski fantasized i n
the Diary (wh i c h i n volved kni ghthoods, " Royal Soc ieties," " New H u­
manisms," and the l i ke) . Rather it was an arti fact of the version he con­
structed retrospective l y in Argonauts. In fus i n g anthropol ogy with fie ld­
work M a l i nowski made the most, the best story, of what c i rcu mstance
had obl iged h i m to attempt.
Such considerations lead us to a problem in d iscuss i ng M a l i now­
ski's-and i ndeed nearly a l l -ethnographic prod uctio n . Than ks to a
growi ng n u m be r of confessional and analytic accounts, we know more
and more about fieldwork experiences and the i r constra i nts . B ut the ac­
tual writi ng of ethnograph ies rem a i n s obsc u re and u nanalyzed . We know
someth i n g about Ma l i nowski's Trobriand research of 1 9 1 4 and 1 9 1 8 but
v i rtu a l l y noth i n g of what he was doing in the Canary I s l ands d u r i n g 1 92 0
a n d 1 92 1 . ( H e w a s writi ng Argonauts of the Western Pacific. )
T h e Diary l eaves u s hangi ng. There is a sudden gap i n t h e writi ng
that, as we learn from sma l l revel ations when the text struggles to re­
sume, signal s the arriva l of word that h i s mother has d ied . Then the des­
perate envoi : "Tru l y I l ac k rea l character." S i lence. Three years l ater Mal­
i nowski reappears as the author of Argona uts, the charter of the new
fieldworker-anthropologist. What has i ntervened ? L i ke Con rad i n the pe­
riod between the rout of h i s African adventu re and the success of Heart
of Darkness he has accepted th ree major comm itments : ( 1 ) to writi ng, (2)
to marri age, and (3) to a l i m ited audience, language, and cu lture .
T h e Canary Islands a re a n i ntrigu i ng scene for Ma l i nowski's writi ng
c u re . H e goes there for his health, but the choice is overdeterm i ned . One
is tem pted to see th is place as a l i m i na l site at the outer edge of E u rope,
propitious for a d i s p l aced Po le writing Pac ific eth nography. More i m por­
tant, however, is the fact that he had earl ier vacati oned in the Canaries
with his mother. Now he is there aga i n with his new wife, completing
his fi rst major work. He is fu l l y i n the rea l m of su bstitution, of a series of
com pro m i ses and rep l acements . For Ma l i nowski , as for Con rad , th ree
such su bstitutions are cruc i a l : ( 1 ) fam i l y, with mother repl aced by wife;
(2) la ngu age, with the mother tongue aba ndoned for Engl ish ; and (3)
writing, with i nsc r i ptions and texts s u bstituted for i mmed iate ora l expe­
rience . The arbitrary code of one language, Engl ish, is fi n a l l y given pre­
cedence. The mother tongue recedes, and (here the persona l and the
po l i tical coi ncide) English dom i n ates-represents and i nterprets- Kiri­
w i n i a n . C u ltural attach ment is enacted as marri age. The yearn i n g for
s i n cere i nter l ocutory speech gives way to a p l ay on written substitutes .

Some of these trans itions and repl acements were s u rely at stake i n the
successfu l writing on the Canary I s lands . Mal i nowsk i 's Diary ends with
the death of a mother; Argonauts i s a rescue, the i nscription of a c u l tu re. 1 0

A few fi nal reflections on the cu rrent status of the ethnograph i c author:
When Ma l i nowski's Diary was fi rst pub l i shed , it seemed scandalous. The
q u i ntessenti a l a nth ropo l ogist of Argonauts did not, in fact, a l ways main­
ta i n an u nderstand i ng, benevo lent attitude toward h i s informants; his
state of m i nd i n the field was anyth ing but coo l l y objective; the story of
ethnogra p h i c research i n c l u ded in the fi n i s hed monograph was styl ized
and selective . These facts, once entered i nto the publ i c record of anthro­
po logical science, shook the fiction of cu ltura l rel ativism as a stable sub­
jectivity, a standpo i nt for a self that understands and represents a cu ltural
other. I n the wake of the Diary cross-c u l tu ra l com prehension appeared a
rhetorical con struct, i ts balanced com prehens ion traversed by ambiva­
lence and power.
We recal l the fate of Ku rtz's violent scrawl i n Hea rt of Darkness,
" Exterm i nate a l l the brutes." Marlow tears off the dam n i ng, truthfu l s u p­
plement when he gives K u rtz 's d isq u i s ition on savage customs to the Bel­
gian press. It i s a tel l i ng gesture, and it suggests a troubl ing q uestion
about Ma l i nowski and anth ropo logy : What i s a l ways torn off, as i t were,
to construct a pub l i c , bel i evab l e d i scourse? In Argonauts the Diary was
excl uded , written over, in the process of giving wholeness to a culture
(Trobriand) and a self (the scientific eth nographer) . Thus the d isc i p l ine of
fie ldwork-based anth ropo logy, in constituting its authority, constructs
and reconstructs coherent c u l tu ra l others and i nterpreting selves . If th i s
eth nographic self-fash io n i ng pres u pposes l ies o f o m i ss ion a n d o f rheto­
ric, it a l so m a kes possi b l e the te l l i ng of powerfu l truths. But l i ke Marlow's
accou nt aboard Nellie, the truths of c u ltura l descri ptions are mean i ngfu l
to spec i fi c interpretive com m u n ities i n l im iting h i storical c i rcumstances .
Thus the "tearing off," N ietzsche rem i nds us , is s i m u l taneously an act of
censors h i p and of mea n i ng creation, a suppression of i ncoherence and
contrad ict ion. The best eth nographic fictions are, l i ke Ma l i nowski's , i n -

1 0. It wou l d b e interesting to analyze systematica l ly how, o u t o f the heter­

oglot encou nters of fieldwork, ethnographers construct texts whose preva i l i ng
language comes to override, represent, or translate other languages. Here Tal a l
Asad's conception o f a persi stent, structu red i nequal ity of la nguages gives pol iti­
cal and h i storical content to the apparently neutra l process of cultu ra l translation
(Asad 1 986).
O N E T H N O G R A P H I C S E L F - FA S H I O N I N G 1 13

tricate l y truthfu l ; but the i r facts, l i ke a l l facts i n the huma n sc iences, are
c l assified , contextual ized , na rrated , and i nten s ified .
I n recent years new forms of ethnographic rea l ism have emerged ,
more d i a logical and open-ended i n narrative sty l e . Self and other, c u ltu re
and its interpreters, appear less confident entities . Among those who
have revised eth nogra phic authority and rhetoric from with i n the d isci­
p l i ne I s ha l l mention j ust th ree (whom C l ifford Geertz has marked off for
critique i n a series of p rovocative lectu res on the writi ng of eth nography) :
Pau l Rab i now, Kev i n Dwyer, and Vi ncent Crapanzano . 1 1 ( For the i r s i n s
of self-d i s p l ay Geertz ca l l s t h e m "Ma l i nowski 's C h i l d ren .") These th ree
can stand for m a n y others cu rrently engaged i n a com plex field of textual
experi ments at the l i m i ts of academ ic eth nography. 1 2 I have said that
anth ropo l ogy sti l l awa its i ts Con rad . I n va rious ways the recent experi­
menta l ists are fi l l i ng that role. They teeter prod uctive l y, as Con rad did­
and as, more a m b iva lently, Geertz h i mself ba l ances-between rea l ism
and modern i s m . The experi menta l ists revea l i n the i r writi ngs an acute
sense of the fash ioned, conti ngent status of a l l c u ltu ra l descri ptions (and
of a l l cu l tural descri bers) .
These self-reflexive writers occu py i ron i c pos itions with i n the gen­
era l project of eth nograph ic s u bjectivity and cu ltu ra l descri ptio n . They
stand, as we a l l do, on an u n certa i n h i storical ground, a p l ace from
which we can beg i n to a n a l yze the ideological matrix that prod uced eth­
nography, the p l u ra l defi n ition of c u ltu re, and a self pos itioned to me­
d i ate between d i screpant worlds of mean i ng . (To say that this h istorical
ground is, for example, postco l o n i a l or postmodern i s not to say much­
except to name what one hopes no longer to h ave to be . ) I n fact most of
the self-conscious hermeneutic ethnographers writing today get about as
far as Con rad did i n Heart of Darkness, at least i n thei r presentations of
narrative authority. They now gestu re toward the problematic other nar­
rator on the deck of the Nellie as they say, with Marlow : "Of cou rse i n
th i s you fel l ows see more than I d i d the n . You see me, whom you know."

1 1 . G ee rtz's lectures ( 1 983), "Works and Lives: The Anth ropologist as Au­
thor," were not yet publ i shed as of this writing. I n the section of the oral presen ­
tation I a m d i scussing he refers pri mari l y to Rabinow 1 9 7 7 ; Crapanzano 1 980 ;
and Dwyer 1 98 2 .
1 2 . T h e d iscu rsive fie l d cannot, o f cou rse, b e l i m ited t o t h e d isci p l i ne of
anthropology or its frontiers; nor is i t adeq u ate l y captu red in terms l i ke reflexi\·e
or dialogica l. For provisional su rveys see Marcus and Cushman 1 98 2 ; Cl ifford
1 986a; and Chapter 1 .
Store, avenue des Gobelins. 1 920s. Photograph by Eugene Atget.
Part 1\vo � Displacements
The coupling of two realities, irreconcilable in appearance,
upon a plane which apparently does not suit them . . .

4. On Ethnographic Surrealism

A N D R � B R E T O N often i n s i sted that su rrea l i sm was not a body of doc­

tri nes or a defi nable idea but a n activity. T h i s chapter is an exploration
of ethnograph i c activity set, as it m ust always be, i n s pec ific c u ltura l and
h i storical c i rc u m stances. I wi l l be concentrating on ethnography and s u r­
rea l i sm i n F rance between the two world wars . To d i sc u ss these activities
together-at ti mes, i ndeed, to perm i t them to merge-is to question a
n u m be r of common d i sti nctions and u n ities. I am concerned l ess with
charti ng i ntel lectua l or a rtistic trad itions than with fol lowi ng some of the
byways of what I take to be a crucial modern orientation toward c u l tu ra l
order. I f I someti mes u se fa m i l iar terms against t h e gra i n , m y a i m is to
cut across retrospective l y establ i shed defi n itions and to recaptu re, if pos­
s i ble, a situation in w h i c h ethnography is aga i n someth i n g u n fa m i l iar and
su rrea l i s m not yet a bounded provi nce of modern art and l iteratu re .
The orientation toward c u ltura l order that I evoke can not be neatly
defined . It is m ore p roperly c a l led moder n i st than modern , ta king as its
problem-and opportu n ity-the fragmentation and j uxtaposition of c u l ­
tu ra l va l ues. From th i s d i sencha nted viewpoi nt stable orders o f col l ective


mean i n g appear to be constructed , artific i a l , and i ndeed often ideologi­

ca l or repress ive. The sort of norma l i ty or common sense that can amass
empires in fits of absent- m i nded ness or wander routinely i nto world wars
is seen as a contested rea l ity to be subverted , parod ied, and transgressed .
I w i l l suggest reason s for l i n king eth nographic activity to this set of criti­
cal attitudes, d ispos itions usua l l y assoc i ated with the artistic avant-garde.
I n F rance partic u l a r l y the modern h u man sciences h ave not lost contact
with the world of l iteratu re and art, and in the hothouse m i l ieu of Parisian
cu ltural l ife no field of soc i a l or artistic research can long remain i nd if­
ferent to i nfl uences or provocations from beyond its d i sc i p l i nary bou nd­
aries . In the twenties and th i rties, as we shal l see, ethnography and sur­
real ism developed i n c lose prox i m ity.
I a m u s i n g the term surrealism i n an obvious l y expanded sense to
c i rc u mscribe an aesthetic that va l ues fragments, cu rious co l l ections, un­
expected j u xtapos itions-that works to provoke the manifestation of ex­
traord inary rea l ities d rawn from the doma ins of the erotic, the exotic,
and the u nconsc ious. This set of attitudes cannot, of cou rse, be l i m ited
to B reton's group; and the su rrea l i st movement narrowly defi ned-with
its manifestoes, sch isms, and excomm u n ications- is not the concern
here . I ndeed the figu res I wi l l be d i scuss i ng were at best fel l ow travelers
or d issidents who broke with B reton. They partook nonethe less of the
genera l attitude I ca l l su rrea l i st, 1 a tangled d isposition foreshortened here
i n an attem pt to d isengage its eth nograph ic d i mensi o n. Ethnography and
surrea l ism are not stable u n ities; my subject i s not, therefore, an overlap­
ping of two c learly d isti ngu i shable trad itions . 2 Moreover, I have tried not
to th i n k of my topic as a conj u nctu re restricted to French cu lture of the
twenties and th i rties. The bou ndaries of art and sc ience (espec i a l l y the
h u man sciences) are ideo logical and sh ifti ng, and i ntel lectual h i story is
itself en meshed in these shifts. Its gen res do not rema i n fi rm ly anchored .
Changing defi n itions of art or science must provoke new retrospective

1 . My broad use of the term rough l y coincides with Susan Sontag's ( 1 9 77)
view of su rrea l ism as a pervasive- perhaps dom inant-modern sensi b i l ity. For a
treatment that d i sti ngu i shes the specific trad ition I am d iscussing from the sur­
rea l ism of Breton's movement see Jam i n 1 980. A "corrective" to this cha pter,
reasserting strict defi nitions of both su rreal ism and ethnography, can be found i n
jam i n 1 986.
2 . Research on the common ground of twentieth-centu ry social science and
the avant-garde is sti l l u ndeveloped . Thus my d i scussion is very pre l i m i nary. On
the French context see Boon 1 9 72; Duvignaud 1 9 79; Hol l ier 1 9 79; jamin 1 979,
1 980; Lourau 1 9 74; and Tiryakian 1 9 79.
O N E T H N O G R A PH I C S U R R E A L I S M l l9

u n ities, new ideal types for h i stori cal descr i ption. In th is sense eth no­
gra p h i c su rreal is m i s a utopian construct, a statement at once about past
and futu re poss i b i l ities for c u ltura l analysis.

The Ethnographic Surreal

In "The Storytel ler" Wa l ter Benj a m i n descri bes the transition from a
trad itiona l m ode of com m u n i cation based on conti n uous ora l narrative
and shared experience to a c u ltura l style characteri zed by bu rsts of " i n ­
formation" -the photograp h , t h e newspaper c l i p, t h e perceptual s hocks
of a modern city. Benj a m i n beg i n s h is essay with the Fi rst World War :

A generation that h a d gone to schoo l on a horse-drawn streetcar

now stood under the open sky i n a cou n tryside i n which noth ing re­
mai ned unchanged but the clouds, and beneath the clouds, in a field
of force of destructive torrents and explosions, was the tiny, fragile
h u man body. ( 1 969 : 84)

Real ity is no l onger a given, a n atu ra l , fam i l iar envi ron ment. The
self, cut loose from its attachments, m u st d iscover mean ing where it
may-a pred icament, evoked at its most n i h i l istic, that u nderl ies both
su rrea l is m and modern eth n ography. Earl ier l i terary and arti stic refrac­
tions of Benj a m i n's modern world are we l l known : the experience of
Baudelai re's u rban flaneur, R i m baud's systematic sensual dera ngements,
the ana lytic decom position of rea l ity begu n by Cezanne and com pleted
by the c u b i sts, and espec i a l l y Lautreamont's famous defi n ition of beauty,
"the chance encou nter on a d i ssect i n g tab l e of a sewing m ac h i ne and an
u m bre l l a ." To see c u ltu re and its norms-beauty, truth, rea l i ty-as a rti­
fici a l arrangements susceptible to detached analysis and com parison
with other poss i b l e d ispos itions is c ru c i a l to a n eth nographic attitude.
In his c l ass ic History of Surrealism ( 1 965) Mau rice N adeau stressed
the formative i m pact of warti me experiences on the fou nders of the s u r­
rea l i st movement- B reton, E l uard , Aragon, Peret, Sou pa u l t. After Eu­
rope's col lapse into barbarism and the manifest ban kru ptcy of the ideol­
ogy of progress, after a deep fiss u re h ad opened between the experience
of the trenches and the offi c i a l language of heroism and victory, after the
romantic rhetorica l conventions of the n i neteenth century had proved
themsel ves i ncapab l e of representi ng the rea l ity of the war, the world was
permanently su rrea l i st. F resh from the trenches, G u i l laume Apo l l i n a i re
coi ned the term i n a letter of 1 9 1 7 . H i s Calligrammes ( 1 9 1 8 : 34 1 ) with
1 20 D I S PL A C E M E N T S

the i r fractu red form and heightened attention to the perceived world,
annou nced the postwar aesthetic :

The Victory above a l l wi l l be

To see c learly at a d i stance
To see everything
Near at hand
And may all th i ngs bear a new name.

Wh i l e for Fernand Leger:

The war had th rust me, as a soldier, i nto the heart of a mechanical
atmosphere . Here I discovered the beauty of the fragment. I sensed a
new real ity i n the deta i l of a machi ne, in the common object. I tried
to fi nd the p l astic value of these fragments of our modern l i fe . 3

Before the war Apol l i n a i re had decorated h i s study with African "fet­
ishes," and in h i s long poem "Zone" these objects wou ld be i nvoked as
"des Ch rist d ' u ne autre forme et d ' u ne autre croyance." For the Paris
avant-garde, Africa (and to a l esser degree Oceania and Ameri ca) pro­
vided a reservo i r of other forms and other bel iefs . This suggests a second
element of the eth nograph ic su rreal i st attitude, a bel i ef that the other
(whether access i b l e in d reams, fetishes, or Levy-B ru h l 's mentalite primi­
tive) was a cruc i a l obj ect of modern researc h . U n l i ke the exoticism of
the n i neteenth centu ry, which depa rted from a more-or- less confident
cu ltura l order in search of a temporary frisson, a c i rcu mscribed experi­
ence of the bizarre, modern su rrea l ism and eth nography began with a
rea l ity deep l y i n questio n . Others appeared now as serious human alter­
natives ; modern cu ltu ra l rel ativism became possible. As artists and writ­
ers set about after the war putti ng the pieces of cu lture together i n new
ways, the i r field of selection expanded dramati cal ly. The "prim itive" so­
cieties of the planet were increasingly avai lable as aesthetic, cosmolog­
ical, and scientific resou rces . These possibi l ities d rew on someth ing
more than an o l der Orienta l ism; they req u i red modern ethnography. The
postwar context was structured by a bas ica l l y i ron i c experience of c u l ­
tu re . F o r every l ocal custom or truth there was a lways an exotic alterna­
tive, a possible j uxtaposition or i ncongru ity. Be low (psychologica l ly) and

3. Quoted i n Sontag 1 97 7 : 204. Pa u l Fusse l l 's incisive study The Great War
and Modern Memory ( 1 975) a l so stresses the F i rst World War's i n itiation of a
generation into a fragmented , "modern ist" world .

beyond (geograph i ca l l y) ord i na ry real ity there existed another real ity.
Su rrea l ism shared t h i s i ro n i c situation with rel ativist eth nography.
The term ethnograph y as I am u s i n g it here is evidentl y d i fferent from
the empi rical researc h tec h n i q ue of a h u man sc ience that in France was
cal l ed eth nology, in England soc i a l anthropo logy, and in America c u l ­
tura l anthropology. I am referring t o a more general c u l t u ral p red isposi ­
t i o n that c uts th rough modern anthropology a n d that t h i s science shares
with twentieth-centu ry a rt and writi ng. The eth nographic l abe l suggests
a characteristic attitude of part i c i pant observation among the artifacts of
a defam i l iarized c u ltural rea l ity. The su rrea l i sts were i ntensely i nterested
in exotic worlds, among w h i c h they i n c l uded a certai n Paris. Thei r atti­
tude, wh i le comparabl e to that of the fieldworker who strives to render
the u nfam i l iar com prehensible, tended to work in the reverse sense,
making the fam i l iar strange . The contrast i s in fact generated by a contin­
uous p l ay of the fam i l ia r and the strange, of wh ich eth nography and s u r­
rea l ism are two e l ements . Th is p l ay is constitutive of the modern c u ltura l
situation I a m ass u m i ng as the ground for my account.
The world of the c ity for Lou is Aragon's Payson de Paris or for B reton
in Nadja was a sou rce of the u nexpected and the sign ificant-significant
i n ways that suggested beneath the d u l l veneer of the rea l the poss i b i l ity
of another more m i raculous world based on rad i ca l ly d i fferent pri n c i ples
of c l ass i fication and o rder. The su rreal ists freq uented the Marche aux
Puces, the vast flea m arket of Paris, where one cou ld red iscover the ar­
tifacts of c u l tu re, sc ram bled and rearranged . With l uc k one cou ld bring
home some b izarre or u nexpected object, a work of art with nowhere to
go- " ready-mades" such as Marce l Duchamp's bottle rack, and objets
sauvages, African or Ocea n ian scu l pt u res . These objects-stripped of
their fu nctiona l context-were necessary fu rn i s h i ngs for the avant-garde
stud io.
It is best to suspend d i sbe l ief i n considering the practi ces-and the
excesses-of s u rrea l i st "ethnographers." And it is i m po rtant to u nder­
stand the i r way of ta king c u l tu re seriously, as a contested rea l ity-a way
that i nc l uded the r i d ic u l i ng and reshuffl i n g of its orders. Th is m uch i s
necessary if one is t o penetrate t h e m i l ieu that spawned a n d oriented the
emerging French scho l a r l y trad ition . More genera l l y, it i s advisable not
to d i sm iss su rrea l i s m too q u ickly as frivo l ous, in contrast with the serieux
of eth nographic science. The connections between anth ropological re­
searc h and researc h i n l iteratu re and the arts, a l ways strong in this cen­
tu ry, needed to be more fu l ly explored . Su rrea l ism i s eth nography's secret

sharer-for better or worse- i n the description, ana l ysis, and extension

of the grounds of twentieth-century expression and mean i ng.

Mauss, Bataille, Metraux

Paris 1 92 5 : the Revue negre enjoys a smash season at the Theatre

des Champs- E iysees, fol low ing on the heels of W. H . Wel l man's Southern
Syncopated Orc hestra . Spi ritu a l s and /e jazz sweep the avant-garde bou r­
geo isie, w h i c h hau nts Negro bars, sways to new rhyth ms i n search of
someth ing prim itive, sauvage . .and completely modern . Styl ish Paris

i s transported by the pulsing stru m of banjos and by the sensuous Jose­

phine Baker "abando n i ng herself to the rhythm of the Charleston" ( Leiris
1 96 8 : 3 3 ) .
Paris 1 92 5 : a n ucleus of U n i versity scho lars-Pa u l Rivet, Lucien
Levy-Bru h l , and Marcel Mauss-esta b l i shes the l nstitut d ' Ethnologie . For
the fi rst time in F rance there exists an organization whose primary con­
cern is the tra i n i n g of profess ional fie ldworkers and the publ ication of
eth nographic sc holarsh i p .
Paris 1 92 5 : i n the wake o f t h e Fi rst Surreal ist Manifesto t h e move­
ment begins to make itself notorious. F rance is engaged in a m i nor war
with anticolonial rebe l s in Morocco ; B reton and company sym path ize
with the i n s urgents . At a banq uet i n honor of the symbo l i st poet Sai nt­
Pol-Raux, a me lee eru pts between the su rrea l ists and con servative patri­
ots . Epithets fl y; "Vi ve I 'AI I emagne!" ri ngs out; Ph i l i ppe Sou pa u l t swi ngs
from a chandel ier, kicking over bottles and gl asses . Michel Leiris is soon
at an open wi ndow, denouncing France to the growi ng crowd . A riot
ensues; Lei ris, nearly lync hed , is arrested and manhand led by the pol ice
( N adeau 1 965 : 1 1 2-1 1 4) .
T h e th ree events were connected b y more than a coi nci dence of
date . For example when Lei ris, whose evocation of Joseph ine Baker I
have j ust quoted , defected from the su rrea l i st movement i n the late twen­
ties seeki ng a more concrete app l i cation for his subversive l iterary tal ­
ents, it seemed natu ra l for h i m to study with Mauss a t the l nstitut
d' Eth nologie and to become an ethnographer of Africa-a partic i pant i n
France's fi rst major fie l dwork expedi tion, the Mission Dakar-Dj i bouti of
1 93 1 -1 93 3 . Scientific, or at least academ ic ethnography had not yet
come of age. Its development i n the early th i rties, through successes l i ke
that m uch-pu b l i c ized Da kar-Dj i bouti expedition, was continuous with
the su rreal ism of the twenties . The organizational energies of Rivet and
the teac h i ng of Mauss were domi nant factors . I sh a l l d i scuss Rivet's insti-

tutional accomp l ish ments l ater i n th i s chapter, notably h i s creation o f the

M u see de I ' Homme. Mauss's pervas ive i nfluence is harder to pin down
si nce it took the form of ora l i ns p i ration in his teac h i n g at the Eco le Pra­
tique des H autes Etudes and the l nstitut d ' Eth nologie.
Nearly every major F rench eth nographer before the m i d-fifties­
with the notable exception of Levi-Strauss-was the benefici ary of
Mauss's d i rect sti m u l atio n . From the perspective of today's i nte l l ectu a l
regi me, where p u b l i cation i s a t a prem i u m a n d where a n y idea o f va l u e
tends t o b e guarded for t h e next a rticle or monograph, it i s astonish i ng,
indeed movi ng, to note the tremendous energies that Mauss pou red i nto
h i s teac h i ng at Hautes Etudes . A g lance through the schoo l 's Annuaire,
where cou rse s u m maries are recorded , revea l s the extraord i nary wea lth
of learn i ng and analysis made ava i l able to a few students, year in and
year out, w ithout repetition, much of which never saw pri nt. Mauss gave
courses on topics from S i berian shaman i s m to Au stra l ian oral poetry to
Polynesian and West Coast I nd i a n ritua l , bringing to bear h i s profo u nd
knowl edge of orienta l re l i gions and c l assical antiqu ity. Readers of
Mau ss's essays-the pages half-devou red by footnotes-wi l l recogn i ze
the breadth of references; they wi l l m i ss, however, the wit and verve, the
give-and-take of h i s oral performances.
Mauss was a research scho l a r. He taught a select gro u p . In the th i r­
ties a band of devotees, some of them amateu rs of the fash ionable exotic,
others eth nographers preparing to l eave for the field (some of the former
in the process of beco m i n g the l atter), wou l d fol low Mauss from hal l to
hal l . At Hautes Etudes, the l nstitut d ' Eth nologie, and l ater the Col lege de
France they reve led i n his erudite, loquacious, and a l ways provocati ve
tou rs th rough the world's c u ltural d i versities. Mauss's lectu res were not
theoretical demonstrations. They stressed , in their d i vagating way, con­
crete eth nogra p h i c fact; h e had a sharp eye for the sign ificant deta i l .
Though h e never d id fiel dwork h i mself, Mauss was effective i n pressi ng
h i s students toward fi rsthand research (see Condomi nas 1 9 72a, b; Mauss
1 947).
H is essay "Tech n i q u es of the Body" ( 1 934) gives a h i nt of Mauss's
ora l style. Here a re a few l i nes from what i s essentia l l y a long l i st of the
th i ngs people in d i fferent parts of the world do with thei r bod ies :

It's normal for c h i l d ren to squat. We no longer know how to squat. I

consider this to be an absurd i ty and i nferiority of our races, civi l iza­
tions, societies.

The notion that sleeping is someth i n g natural is completely inexact.


Noth ing i s more d izzying than to see a Kaby l ie come downsta i rs with
babouches on . How can he stand without los i ng his s l i ppers? I 've tried
to watch, to do it. I don't see how. And I don't understand either how
women can wa l k on their h igh heels.

Hygiene of natural body functions. Here I could l i st numberless facts.

Final ly, it m u st be u nderstood that danci n g while embracing is a prod­

uct of modern E u ropean civ i l i zation. Th is shou ld show you that thi ngs
q u ite natura l for us are h i storica l ; they may horrify everyone else in the
world except us. (pp. 3 74, 3 78, 3 8 1 , 383)

The pre h i storian Andre Leroi-Gourhan remembers his teacher as a man

of " i nspired confu s i o n ." In an interview he is asked what he reca l l s of h i s
teacher's speech :

H i s s i l ences, i f I may put i t thus. I can't provide a n i m itation ; so many

years have passed, and I have an ideal ized image of Mauss; but he
constructed his sentences i n a way that suggested thi ngs without de­
claring them i nflexibly. H is d i scou rse was all a rticulations and elastic­
ity. Most of his sentences came u p empty, but it was an empti ness that
invited you to b u i l d . That's why I said the most characteristic thi ngs
were h is si lences.
He was especial ly amazing when he did textua l explications on
authors who had worked in S i beria on the G i l iaks or Goldies . I remem­
ber sessions at H autes Etudes-there were never more than ten of us,
and yet ! We gathered around a table l i ke this one, not q u i te so long;
Mauss translated from German to French with commentaries that drew
comparisons from every corner of the globe. H is erudition was fantas­
tic, and we took it in without rea l ly being able to say afterwards how
he had managed to be so engrossing. ( 1 982 : 3 2 )

Mauss d i d not write books . H i s Oeuvres ( 1 968 -69) is composed of

essays, scholarly a rtic l es, interventions at meeti ngs, countless book re­
views . Compressed c l ass ics such as The Gift ( 1 923) and A General
Theo ry of Magic ( 1 902) were pub l ished i n the Annee sociologique. H i s
magn u m opus, a d issertation on prayer, remai ned a col l ection o f d rafts,
essays, scraps, and notes. So d id other synthetic works on money and
the nation . Perhaps because so much was connected i n his m i nd, Mauss
cou l d eas i l y be sidetracked ; and he was profl igate with com mitments
and loya lties. He lectu red constantly and spent years bri nging work by
O N ETH N O G R A P H I C S U R R E A L I S M 125

deceased col leagues ( D u rkheim, Robert Hertz, Henri H u bert) to com ple­
tion . A D reyfu sard and soc i a l ist in the trad ition of j a u res, he wrote for
L'h umanite and took part i n str i kes, elections, and the pop u l a r u n i versity
movement. U n l i ke D u rkhe i m , h i s rather austere uncl e, Mauss was gre­
garious, bohe m i a n , and someth i ng of a bon vivant.
Some reca l l Mauss as a loyal D u rkhe i m i a n . Others see a foreru nner
of stru ct u ra l i s m . Some see pri mari ly an anthropologi st, others a h istori an .
Sti l l others, c iting h i s rabb i n ical roots, h i s tra i n i ng i n Sanskrit, and h i s
l ife long i nterest i n ritu a l , a l l y h i m with students o f re l igion s u c h as h i s
friends Marcel G ranet, H u bert, a n d Leenhardt. Some stress Mauss's icon­
oclasm, others h i s coherent soc i a l i st-h u ma n i st vis ion . Some see a bri l­
l i ant armc h a i r theorist. Others remember a sharp empirical observer. The
d i ffe rent versions of Mauss a re not irreconci lable, but they do not q u ite
add u p . People read ing and remembering h i m a l ways seem to find some­
th i ng of themselves (from Leroi-Gou rhan 1 982 : 3 2-3 3 ) :

For a period o f two years when I was attend i n g nearly a l l h i s cou rses i t
was agreed that a comrade and 1 - a Russian Jew, Deborah Lifchitz
who d i ed i n the Nazi deportation-would take notes i n turn and i n a
way that wou ld l et us compare them to determ i ne the real content of
Mauss's teac h i ng. We never managed to construct anyth ing coherent
because it was too rich and always ended u p at the horizon. Later a
record of h i s cou rse was publ i shed by a group of former students.
Wel l , there was a total d i vergence between what they noted and what
Deborah and I took down ! T h i s i s the secret, I bel ieve, of the rea l spe l l
he cast on h i s fo l l owers .

An exa m p l e of how Mauss's pec u l iar brand of i ntel lectual sti m u l a­

tion got a ro u nd is provided by the great fieldworker Alfred Metraux, who
was h i s student d u r i n g the m id-twenties ( B i ng 1 964 : 2 0 -2 5 ) . Being of a
carefu l , empi rica l tem perament, Metraux soon d i strusted the fast-and­
loose way that eth n ogra p h i c fact was bei n g u sed by the early su rreal i sts .
H e devoted h i s l i fe to firsthand research , becomi ng, i n the words of Sid­
ney M i ntz ( 1 9 7 2 : 2 ) , the "fieldworker's fieldworker." But he remai ned in
tou c h with the avant-garde . Wh i le a student at the Ecole des Chartes,
Metraux had establ is h ed a lasting friend s h i p with Georges Bata i l le, the
id iosyncratic scholar, essayi st, and pornographer, whose i n fl uence has
been so pervasive on the p resent generation of rad ical c ritics and write rs
i n Paris. The work of the two friends cou ld not be more d i fferent: the one
restra i ned, a l most p u rita n ical in tone, though with a flai r for i so l ating the

tel l ing deta i l ; the other provocati ve, far-fl u ng, N ietzschean . Yet i n a cu­
rious, compel l i n g way the two are complementary : while Bata i l le was
steadied by Metraux's erudition, Metraux found h i s passion for eth nog­
raphy confi rmed by h i s friend's wi l l i ngness to express what, accord i ng to
Le iris, they h ad in com mon-"a violent ardor for l ife combi ned with a
piti less awareness of its absu rd i ty" ( Leiri s 1 966a : 2 5 2 ; see a l so Bata i l l e
1 95 7 : 1 4 ; a n d Metraux 1 963 : 6 7 7-684) . The l ifelong association between
Bata i l le and Metra ux can be seen as emblematic of that enduring conti­
gu ity, if not a l ways s i m i larity, that has kept French eth nography on speak­
i ng terms with the avant-garde .
Batai l le's most i nfl uenti a l book was h i s l ate treatise L 'erotisme
( 1 9 5 7) . Its orientation, and that of Bata i l le's work genera l ly, can be traced
to Mauss by way of Metraux's report of a lectu re aro u nd 1 92 5 . I n L'ero­
tisme Bata i l l e i ntrod uces the book's key chapter, on transgression, with
the p h rase "Transgression does not negate an i nterd i ction, it transcends
and completes it." Metraux spec ifies that h i s characteristic form ula is
o n l y a pa raph rase of "one of those profound aphorisms, often obscu re,
that Marcel Mauss wou ld th row out without worrying about the confu­
sion of h i s students ." Metraux h ad heard Mauss say in a lectu re, "Taboos
are made to be violated ." Th is theme, which Bata i l l e wou ld often repeat,
became a key to h i s th i n ki ng. Cu ltu re is ambivalent in structu re. One
may refra i n from m u rder, or one may go to war; both acts are, for Ba­
ta i l le, generated by the i nterd iction on ki l l i ng . Cu ltu ra l order incl udes
both the r u l e and the transgress ion. T h i s logic appl ies to a l l man ner of
ru les and freedoms-for example to sex ual norma l ity and its partner the
perversions. In Metraux's words, "Mauss's proposition, in the apparent
absurd ity of its form, manifests the i nev itable con nection of confl icti ng
emotions: [quoti ng Bata i l le) ' U nder the i m pact of negative emotion, we
must obey the i nterd icti o n . We violate it if the emotion is positive' " (Me­
traux 1 963 : 682-683 ; Batai l le 1 95 7 : 72-7 3 ) .
Bata i l le's l ifelong project was to demystify a n d valorize th is "positive
emotion" of transgress ion in a l l its various forms, and i n this he was true
to h i s su rreal ist beg i n n i ngs. ( I n the twenties Bata i l le was fi rst an assoc iate
then a critic of the B reton grou p . ) One of h i s fi rst publ ished texts was part
of a col lection on pre-Co l u m bian a rt, i n which he col laborated with Me­
tra ux and Rivet. H is appreciation of h uman sacrifice ( " For the Aztecs
death was not h i ng") j u xtaposes in surrea l i st fashion the beautifu l and the
ugly, the normal and the repugnant. Thus Tenochtitlan is s i m u l taneously
a " h u man slaughterhouse" and a gorgeous "Ven ice" of canals and flow-
O N ETH N O G R A P H I C S U R R E A L I S M 127

ers . The sacrifi c i al victi ms dance i n perfumed garlands; the swarms of

fl ies that gather on the r u n n i n g b lood are beautifu l (Bata i l le 1 930 : 1 3 ) .
"A l l writi ng is garbage," said Anto n i n Artaud, another renegade su rreal­
ist, who wou ld flee France to h i s own d ream of Mex ico-co u rting mad­
ness among the Tarah u mara I ndians (Artaud 1 976). The exotic was a
primary court of appea l aga i nst the rational , the beautifu l , the normal of
the West. B u t Bata i l le's i nterest in the world's c u ltu ral systems, however,
fina l l y went we l l beyond mere delectation or escapism. U n l i ke most sur­
rea l i sts he worked toward a more rigorous theory of co l lective order
based on the double logic of i nterd iction . A l ways au cou rant with eth­
nographi c sc holars h i p, he conti nued to d raw heav i l y on Mau ss-La part
maudite ( 1 949) is an elaborate extrapol ation of The Gift a nd l ater on

Levi-Strauss. The logic developed by Bata i l le, which I can not p u rsue
here, has provided a n i mportant conti nu ity in the ongo i n g rel ation be­
tween c u ltural ana l ys i s and early s u rrea l i s m i n France. It l i n ks the twen­
ties' context to a later generation of rad ical critics, i n c l u d i n g Michel Fou­
cau lt, Rol a nd B a rthes, j acq ues Derrida, and the Te l Quel group.4
I t i s worth noti n g that the col lection of essays i n which Metraux,
R i vet, and Bata i l le co l la borated was part of the fi rst popu lar exh ibition of
pre-Co l u mbian a rt in France. The exh ibit h ad been orga n ized by
Georges-Henri Riviere, a m u s i c student and amateur of jazz who wou l d
become France's most energetic eth nographic m u seo logist. Riviere was
wel l con nected soc i a l l y, Rivet pol itica l l y. The latter u nderstood perfectly
that the c reation of anthropo l ogical researc h i n stitutions req u i red a fash­
ionable enth u s i as m for th i ngs exotic. Such a vogue cou ld be exploited
fi nanc i a l l y and chan neled i n the interests of science and p u b l i c i n struc­
tion . Ri vet, i mpressed by Riviere's su ccessfu l pre-Co l u m bian show, h i red
h i m on the spot to reorgan ize the Trocadero m u seu m , whose col lections
were in a state of d i sorga n ization and d i srepa i r. Th i s was the begi n n i ng
of a prod uctive co l l aboration between the two ch ief animateurs of
French ethnograph ic i n stituti ons, one that wou l d resu lt in the Musee de

4. The trad ition is visible in "Hom mage a Georges Bata i l le," publ ished i n
1 963 b y Critique, which i n c l udes essays b y Alfred Metraux, Michel Leiris, Ray­
mond Queneau, Andre Masson, and Jean Wah l of the prewar generation, and by
Michel Fouca u l t, Roland Barthes, and Phi l i ppe Sol lers of the emerging critical
trad ition . (Another o utgrowth of eth nographic su rreal i s m that can not be pu rsued
here is its con nection with Th i rd World modern ism and nascent anticolon ial d is­
cou rse. It is enough to mention a few promi nent names : Aime Cesai re (a long­
time friend of Leiris), Octavi a Paz, and Alejo Carpentier, who was a col laborator
on the journal Documents . )

I ' Homme and i n Riviere's Musee des Arts et Trad itions Popu l a i res (see
Riviere 1 9 68, 1 979) .
Before the fu l l deployment of these i n stitutions, i n the early years of
the l n stitut d' Eth nologie, Mauss's cou rses remai ned the cruci a l forum for
an e mergi ng eth nography. This teac h i n g was a c u rious scholarly i nstru­
ment, not fu ndamenta l ly at odds with surreal ism and capable of sti mu­
lating the l i kes of both Metraux and Bata i l le . I t is reveal i ng to consider i n
th i s l ight a we l l -known evocation o f Mauss :

In h i s work, and sti l l more in h i s teach i ng, unthought-of comparisons

flou rish. Wh i le he i s often obscure by the constant use of antitheses,
shortcuts, and apparent paradoxes which, later on, prove to be the
result of a deeper insight, he gratifies his l i stener, sudden ly, with fu l­
gu rating i ntuitions, providing the su bstance for months of fru itfu l thi nk­
ing. I n such cases, one feel s that one has reached the bottom of the
soc ial phenomenon and has, as he says somewhere, "hit the bedrock."
This constant stri ving toward the fu ndamenta l , this w i l l i ngness to sift,
over and over aga i n , a h uge mass of data until the purest material on l y
remains, expl a i n s Mauss's preference for t h e essay over t h e book, and
the l i m i ted size of his published work. ( Levi-Strauss 1 94 5 : 5 2 7)

Th i s account from the pen of Levi -Strauss suffers perhaps from a tendency
in its fi nal sentences to portray Mauss as a protostructu ra l ist. 5 The drive
to reach bed rock, to grasp on l y the pu rest u nderl ying material is an as­
p i ration more characteristic of Levi-Strauss than of Mauss, who publ ished
relative l y l ittle not because he had d i st i l led elemental truths but because
he was preoccu pied with teac h i ng, ed iti ng, and pol itics, and because,
know i n g so much, he found that the truth had become too complex. As
Lou i s Du mont reca l l s, " H e had too many i deas to be able to give com­
plete expression to any of them" ( 1 9 7 2 : 1 2) . Levi-Strauss's descri ption of
the great teacher's provocative u se of antithesis and paradox in the pre­
sentation of ethnograph i c knowledge ri ngs true, however, in the context
I have been d i scussing. Eth nograph i c truth for Mauss was restless ly sub­
versi ve of su rface real ities . Its pri ncipal task was to d iscover, i n h i s fa­
mous phrase, the many " l u nes mortes," pale moons in the "fi rmament of

5 . Levi -Strauss's most elaborate attempt in th i s vei n is h i s bri l l iant "I ntroduc­
tion a !'oeuvre de Marcel Mauss" ( 1 950). For a good corrective see Maurice
Leenhardt 1 95 0 .

reason" ( 1 924 : 3 09 ) . There is no better s u mmary o f the task o f ethno­

gra p h i c su rrea l ism, for the " reason" referred to is not a paroch i a l Western
rational ity but the fu l l h u man potential for c u ltura l expression .


In an avan t-garde period ical's " Homage to Picasso" one is not s u r­

prised to find a statement from Mauss ( 1 930) . The journal in q uestion,
Documents, was a glossy review ed ited by Georges Batai l le . I t offers a
revea l i n g case of eth nographic su rrea l ist col laboration . Bata i l le h ad l eft
B reton's su rrea l ist movement a l ong w ith Robert Desnos, Lei ris, Artaud,
Raymond Quenea u , and various others d u r i ng the sch i sms of 1 929, and
his journal fu nctioned as a forum for d i ssident views . It h ad , moreover, a
d i sti nctly eth nographi c bent, w h i c h wou l d attract the col l aboration of
future fie ldworkers s u c h as G r i a u l e , Andre Schaeffner, and Lei ris, as wel l
a s Riviere and Rivet. G ri a u le, Schaeffner, and Lei ri s wou l d depart for
Africa on the M i ssion Dakar- Dj i bouti soon after the demise of Docu­
ments in 1 93 0 . If Documents appears tod ay as a rather strange context
for p u rvey ing ethn ogra p h i c know l edge, in the l ate twenties it was a per­
fectly appropri ate-that is, outre-foru m .
I ndeed , it req u i res an effort o f i magination t o recaptu re the sense, o r
senses, o f t h e word ethnography a s it was u sed i n t h e su rrea l i st twenties.
A defi ned soc i a l science with a d iscern i b l e method , a set of c l assic texts,
and u n i vers ity cha i rs was not yet fu l ly formed . Exam i n i ng the word's u ses
in a p u b l i cation l i ke Documen ts, we see how eth n ographic evidence and
an ethnographic attitude cou ld fu nction in the service of a sub·.rersive
c u ltu ra l critic i s m . In the subtitle of Documents- "Archeologie, Beaux
Arts, Eth nograph ie, Va rietes" -the wild card was " Eth nographie." It de­
noted a rad ical q uestio n i n g of norms and an a ppea l to the exotic, the
paradox ica l , the inso/ite. I t i m p l i ed too a leve l i ng and a rec lassification
of fam i l i a r categories . "Art," spe l l ed with a capital A, h ad a l ready suc­
c u m bed to dada's h eavy arti l le ry. "Cu ltu re," hav i n g bare l y survived th i s
postwar barrage, w a s n o w reso l ute l y lower-case, a pri n c i p l e o f re l ative
order in w h i c h the subl i me and the vu l ga r were treated as symbols of
eq u a l s i g n i ficance. S i nce c u l tu re was perceived by the col laborators of
Documents as a system of mora l and aesthetic h iera rc h ies, the rad ical
critic's task was one of semiotic decod i ng, with the aim of deauthenti­
cating and then expand i ng or d isplacing the common categories. The

cubists' break with the canons of real ism had set the pace for a genera l
assa u lt on t h e norma l . Ethnography, w h i c h shares w i t h su rreal ism a n
abandonment o f t h e d isti nction between h igh a n d l o w cu ltu re, provided
both a fu nd of non-Western a l ternatives and a preva i l i ng attitude of i ro n i c
partic i pant observation a m o n g t h e h ierarc h ies a n d mea n i ngs of col lec­
tive l i fe .
It i s i n structi ve t o attem pt an i nventory o f eth nograph ic perspectives
as revealed by thei r use in Documen ts. Before one has caught the drift,
one is su rprised, for example, to come u pon an artic l e by Car l E i n stei n ­
author o f Negerp/astik ( 1 9 1 5 ) , a pioneering account o f African scu l ptu re
viewed in the l i ght of c u b i sm-entitled "Andre Masson, etude ethnolo­
gique." What d id it mean i n 1 92 9 to study an avant-garde pai nter "eth­
nologica l ly " ? From the outset E i n ste i n sounds the cubist-su rrea l i st battle
c ry :

One thing i s i m portant: to shake what i s ca l l ed rea l i ty by means of

nonadapted hal l u c i nations so as to a l ter the value h i erarchies of the
rea l . H a l l uc i natory forces create a breach i n the order of mecha n i stic
processes; they introd uce blocs of "a-causa l ity" i n th i s rea l ity which
had been absurd l y given as such . The u n i nterru pted fabric of th i s re­
al ity is torn , and one inhabits the tension of dual isms. ( 1 929:95)

The " h a l l uc i n atory forces" of Masson's pa i nting represent, accord ing to

E i nste i n , "the return of mythological creation, the retu rn of a psyc holog­
ical archaism as opposed to a p u re l y i m itative archaism of forms" (p.
1 00) . E i nste i n descri bes this myth i c psychology as "totem i c ." To grasp
the s i g n i ficance of Masson's metamorphoses and u nexpected a n i mal­
h u man com b i n ations, " i t i s e nough to reca l l the prim itive mask-costumes
that i nc i te identi fications with a n i ma l s, ancestors, etc ." (p. 1 02) . E i n­
ste i n 's casual a l l usion en passant to masks (African ? Ocean i a n ? Al askan ?
H is aud ience wi l l know what he means) suggests a context i n which
exotic or arc h a i c poss i b i l ities are never far from the su rface of consc ious­
ness, are ever ready to offer confi rmation for any and a l l breaks opened
i n the Western order of th i ngs. I n E i nste i n 's essay two key e lements of
eth nograph i c su rreal ism are noticeable: fi rst, the corrosive analysis of a
rea l i ty now identified as loca l and artifi c i a l ; and second, the supplying
of exotic a l ternatives .
There is a th i rd aspect of t h i s attitude that spri ngs to one's attention
as one l eafs through the pages of Documents. Marcel Griaule provides
a c lear statement in an essay ridicu l i ng the aesthetic assumptions of
O N ETH N O G R A P H I C S U R R E A L I S M 131

pri m i tive-art amateu rs who doubt the purity of a Bao u l e drum because
the figure carved d n it is hold i n g a rifle. The eth nographic su rrea l i st,
u n l i ke either the typical art critic or anth ropologi st of the period, d e l i ghts
in c u ltura l i m pu rities and d i sturb i n g syncretisms. G r i a u l e eq uates the Eu­
ropean 's del ectation of African art with the African's taste for texti les, gas
cans, a lcoho l , and firearms. If Africans do not choose to i m itate o u r h igh­
c u ltural prod ucts, tan t pis ! H e conc l udes:

Eth nography- i t i s q u i te ti resome to have to keep repeating th is- i s

interested i n the beautiful and t h e ugly, i n t h e European sense o f these
absurd words. It has, however, a tendency to be suspicious of the
beautifu l , which is rather often a rare-that is monstrous-occu rrence
in a c i v i l ization . Eth nography is suspicious too of itsel f-for it is a
wh ite science, i . e. , sta i ned with prej udices-and it w i l l not refuse
aesthetic va lue to an object because it is u p-to-date or mass-pro­
duced . ( 1 930 :46)

Andre Schaeffner u rges a s i m i l a r poi nt in a scholarly su rvey of " Les in­

struments de m us i q u e dans u n m usee d'eth nographie." H i s strictu res a re
n ow an anthropo l ogical com monplace . Read, however, i n the su rreal i st
context of Documents, they recover the i r fu l l su bversive effect.

Whoever says eth nography ad m i ts necessari l y that no object designed

to produ ce sound or music, however ."pri mitive" or formless it may
seem, no musical i nstrument-whether i ts existence is accidental or
essenti al-sha l l be excl uded from a m ethod ical classification . For this
p u rpose any percussive procedu re, on a wooden box or on the earth
itself, i s of eq ual i m portance with the melod i c or polyphon i c means
ava i lable to a viol i n or a gu itar. ( 1 929: 248)

Schaeffner, a n early authority on Stravi n sky, wou ld come, by way of j azz,

to study the m u s i c of the Dagon and l ater to fou nd the eth nomusicology
section of the Musee de I' Homme.
The "ethnogra p h i c " attitude provided a styl e of scientifical ly va l i­
dated c u ltu ra l leve l i ng, the red i stri bution of val ue-charged categories
such as " m u s ic ," "art," " beauty," "sophisticatio n ," "clean l i ness," and so
fort h . The extreme re l ativism, even n i h i l ism, l atent in the eth nographic
approach d id not go u nexploited by the more extreme col laborators of
Documents . The i r view of c u lture d i d not featu re conceptions of orga n i c
structure, fu nctional i ntegration, wholeness, or h i storical conti n u i ty.

Thei r conception of cu ltu re can be cal led, without u nd ue anach ronism,

sem iotic . Cu ltu ra l rea l ity was composed of artificial codes, ideological
identities, and objects suscepti ble to i nventive recom bi nation and j uxta­
position : Lautreamont's u mbre l l a and sewing machi ne, a viol i n and a
pai r of hands s l apping the African d i rt.
The conception, h i gh l i ghted i n Schaeffner's title, of an "eth no­
graph ic m u seu m " is of more than pass ing i m portance here . The fragmen­
tation of modern c u l tu re perceived by Benjam i n , the dissoc i ation of
cu ltu ral knowledge i nto j u xtaposed "citations," is presupposed by Doc­
uments . The journ a l 's title, of cou rse, is i nd i cative. Cu ltu re becomes
someth ing to be col lected , and Documen ts itself is a k i nd of eth no­
graph ic d i splay of i m ages, texts, objects, l abels, a playfu l museum that
s i m u ltaneous l y col lects and rec l ass ifies its speci mens.
The j o u rna l 's basic method is j u xtaposition-fortu itous or i ron ic col­
l age . The proper arrangement of cu ltura l symbol s and artifacts is con­
sta ntly p l aced in doubt. H igh a rt is combi ned with h ideously e n l a rged
photographs of big toes; fol k crafts; Fant6mas (a popu lar mystery series)
covers ; H o l l ywood sets ; African, Melanesian, pre-Co l umbian, and
French carn iva l masks; accou nts of music hall performances; descrip­
tions of the Paris s l aughterhouses. Documents poses, for the cu ltu re of
the modern city, the problem fac i ng any organ izer of an eth nograph ic
m useum : What belongs with what? Should masterpieces of scu l pture be
isolated as such or d isplayed i n proxim ity with cooki ng pots and ax
blades ? (see Leiris 1 966b.) The eth nographic attitude must conti n ua l l y
pose these sorts o f q u estions, compos i n g and decom pos i ng cu lture's
"natu ra l " h ierarc h ies and re l ationsh i ps . Once everyth i n g in a c u l tu re i s
deemed worthy i n pri nciple o f col l ection a n d d isplay, fu ndamenta l issues
of c l assification and va l ue are raised .
I n Documents we observe the use of eth nographic juxtaposition for
the purpose of pertu rbing commonplace symbols. A regu lar section of
the journal is a so-cal led d i ctionary of u nexpected defin ition s . The entry
for the word homme is characteristi c . It rec ites a researcher's brea kdown
of the chemical composition of the average human body: enough i ron to
make a nai l , enough sugar for one c u p of coffee, magnes i u m sufficient
to take a photograph, and so on-market val ue twenty-five francs. The
body, a privi leged i mage of order, is a favorite target. Together with a
va riety of other "natu ra l " entities, it is recoded, and in the process it is
th rown i n doubt. Robert Desnos contributes a d i sconcerting i nventory of
rhetorical forms concern ing the eye, and h i s entry for the mobi le symbol
O N ETH N O G R A P H I C S U R R E A L I S M 133

" n ightingale" begi n s , " Except i n spec i a l cases, this does not have to do
with a b i rd " ( Desnos 1 92 9 : 1 1 7) .
Crachat, "spittle," i s redefined b y G riau l e using b l ack African and
I s l a m i c evidence with the res u l t that spit becomes assoc iated with the
sou l , and with both good and evi l spi rits. In E u rope, natu ra l ly, to spit i n
someone's face i s an a bsol ute d i shonor; i n West Africa i t can be a mode
of bless i n g . "Spittle acts l i ke the sou l : bal m or garbage" (Gri a u l e 1 929).
The eth nographer, l i ke the su rrea l i st, i s l icensed to shock. Lei ris takes u p
Griau l e's defi n ition a n d goes fu rther: spittle i s t h e permanent sperm l i ke
s u l l y i ng of the noble mouth , an organ associated i n the West with i ntel ­
l i gence and language. S p i t thus resym bo l i zed denotes a cond ition o f i n ­
escapable sacri l ege ( L e i r i s 1 92 9 ) . I n t h i s newly recom posed defi n ition to
tal k or to th i n k is a l so to ejac u l ate.
An a pproach to representation by means of j uxtaposition or col l age
was a fam i l iar su rreal i st tac k (Matthews 1 9 7 7 ) . Its i ntent was to break
down the convention a l " bod ies"-objects, identities-that comb i ne to
produce what Barthes wou l d l ater ca l l "the effect of the rea l " ( 1 968) . I n
Documents the j uxtapos ition o f t h e contributions, a n d espec i a l l y o f their
photographi c i l l ustrations, was designed to provoke th i s defam i l i ariza­
tion . The fi rst issue of 1 92 9 begins, for example, with an art i c l e by Lei ris,
" Picasso's Recent Canvases," profuse l y i l l ustrated with photographs.
(These were years when Picasso seemed to be break in g and ben d i ng,
a l most savagel y, the norm a l shape of the h u ma n frame . ) These deformed
i m ages are fol lowed by "The Outcasts of N ature" by Bata i l le, a charac­
teristic apprec iation of frea ks, i l l ustrated by fu l l -page eighteenth-century
engravi n gs of S i amese twi n s . N ext an i l l u strated review of an exh i b ition
of African scu l ptu re provides fu rther visual d i s location of the " natu ra l "
body a s rea l i stica l ly conceived i n t h e West. T h e body, l i ke a c u l tu re sem i ­
otica l l y i magi ned , i s n o t a conti nuous w h o l e b u t an assembl age o f con­
ventional sym bo l s and codes .
Documen ts, part i c u l a r l y i n its u se of photographs, creates the order
of a n u nfi n i s hed col l age rather than that of a u n ified o rganism . Its i mages,
in the i r eq u a l iz i n g gloss and d i stan c i ng effect, present in the same plane
a Chatelet show advertisement, a Hol l ywood mov i e c l i p, a P icasso, a
G i acomett i , a doc u mentary photo from co l o n i a l N ew Cal edo n i a, a
newspaper c l i p , an Eskimo mask, an Old Master, a m usical i n strument­
the world's i conography and c u ltural forms presented as evidence or
data. Evidence of what? Evidence, one can only say, of su rpri s i ng, de­
c l assified c u l t u ra l orders and of an expa nded range of h u man a rtistic

i nvention . Th is odd m u seum merely docu ments, j u xtaposes , rel ativ­

izes-a perverse col lectio n .
T h e m u seum o f eth nograph ic su rreal ism was t o be improved and
channel ed i nto more stable, contin uous i n stitutions . In 1 93 0 Documents,
which h ad become l ess and less recognizably a review of art, \vas aban­
do ned by its ch ief fi nancial backer. Three years l ater a reconstituted cat­
egory, eas i l y identifiable today as mod�rn art, wou l d be i ncarnated in the
legendary l'vfinotaure. A thing of beauty, Minotaure i n terspersed no pho­
tographs of s l aughterhouses, Movietone Fol l ies, or big toes among its
l avish l y rep rod u ced Picassos, Dal is, or Massons. After turning over its
second issue to the Dakar-Dj i bouti team for a handsomel y i l l ustrated re­
port on their African research (Gria u l e 1 934b), M inota u re did not s u b­
seq uently reserve any sign ificant p l ace for eth nograph i c evidence. The
artifacts of otherness were rep l aced , genera l ly, by B reton's category of
the su rrea l - l ocated in the myth ic or psychoanalytic u nconscious and
al l too eas i l y co-opted by romantic notions of artistic gen i u s or inspira­
tion . The concrete c u l tural artifact was no longer cal l ed upon to p l ay a
disru ptive, i l l u m i natory rol e . Modern a rt and eth nography had emerged
as fu l ly d i sti nct positions, i n com m u n i cation to be sure, but from a d is­
tance .
I have dwe lt on Documents because it exem p l ifies with unusual
clari ty the chief areas of convergence ben.veen eth nography and sur­
rea l ism d u ri ng the t\venties and because a n u mber of its contri butors
went on to become i nfluenti a l fieldworkers and m useum organizers .
Documen ts reveal s too i n i ts su bversive, nearly anarc h i c docu mentary
attitude an epistemo l ogica l horizon for twentieth-centu ry c u l tural stud­
ies. If Documents was, as Leiris reca l l s, " i m poss i b l e," it wou ld be hasty
to d ismiss it as an aberration, a person a l c reation of the " i m poss i b l e"
Georges Batai l le (Lei ris 1 9 63). It attracted the partici pation of too many
serious schol a rs and artists to be written off as mere l y self-indulgent or
n i h i l istic . It exem p l i fied , rather, an extreme sens itivity (more character­
istic of the French eth nographic trad ition than is often recogn ized) to the
overdeterm i ned character of what Mauss h ad cal led "total soc i a l facts"
( 1 924 : 76-7 7 ) . Rea l i ty, after the s u rrea l ist t\venties, cou ld never aga i n be
seen as s i m p l e or conti nuous, descr i bable empi rica l l y or th rough i nduc­
tion . It was Mauss who best exem p l i fied the u nderlying attitude when he
rem arked , as he l i ked to : " Ethnology is l i ke the ocean . A l l you need is a
net, any k i nd of net; and then if you step i n to the sea and swi n g you r net
about, you' re sure to catch some k i nd of fish" ( Fortes 1 9 7 3 : 2 84).

I n the Museum o f Man

The h i sto ry of French eth nography between the world wars can be
told as a ta l e of two m useu m s . The old Trocadero m useu m and the new
M u see de I ' Homme exerted sign i ficant i nfl uences, both practical and
ideolog i c a l , on the cou rse of research and the comprehension of i ts re­
s u l ts . If the "Troca" of the tvventies, with i ts m i sl abeled, m i sc l assified
objets d ' a rt, corresponded with the aesthetics of ethnograph i c s u rrea l­
i s m , the com p l ete l y modern Pa l a i s de Cha i l l ot i ncarnated the emerg i ng
scholarl y parad igm of eth nograp h i c h u m a n i s m . The scientific gai n s rep­
resented by the M u see de I ' Homme were cons iderable. It provided both
needed tec h n ical fac i l ities and the eq u a l ly necessary del i neation of a
field for study-the " h u man," i n a l l its physica l , archaeologica l , and eth­
nogra p h i c m a n i festations. The coa l escence of a researc h parad igm c re­
ates the poss i b i l i ty of an acc u m u lation of knowled ge and thus the phe­
nomenon of scholarly progress. What is less often recogn ized, for the
h um a n sciences at l east, i s that any conso l idation of a parad igm depends
on the exc l us i o n or re l egation to the status of "art" of those elements of
the changi n g d i sc i p l i ne that cal l the c redent i a l s of the d i s c i p l i ne itself
i nto q u estion, those research practi ces that, l i ke Documents, work at the
edges of d i sorder.
Before 1 93 0 the Trocadero was a j u mb l e of exotica. Its arrangements
em phasized " l ocal color" or the evocation of foreign sett i ngs : costumed
m a n n eq u i ns, panopl ies, dioramas, m assed specimens. A journa l i st cou ld
write that a visit was l i ke " u n voyage en p l e i ne barbarie" ( D i az
1 9 85 : 3 78). S i nce the co l l ection lacked an u p-to-date scientific, pedagog­
ical v i s i o n , i ts d i sorder made the m useu m a pl ace where one cou ld go to
encou nter c u riosities, fet i s h i zed obj ects . It was there that P i casso,
a round 1 908, began to make a serious study of / 'art negre.

When I went for the first time, at Derain's u rg i ng, to the Trocadero
M u seu m , the smel l of dam pness and rot there stuck i n my throat. It
depressed me so m uch I wanted to get out fast, but I stayed and stud­
ied. ( G i lot 1 9 64 : 2 6 6)

" Le Troca" was a c u rious Byzanto-Moorish structure, u n heated , u n l it. Its

l ac k of coherent scientific contextua l i zation encou raged the appreci ation
of i ts objects as detached works of a rt rather than as cu ltural artifacts .
After the F i rst World War, as the enth u s i asm for t h i n gs pri m i ti ve b los­
somed , the scandalous m u se u m beca me tem porari l y a museu m of "art."

As Riviere's i m provements of the early th i rties progressed , the m u ­

seum bega n t o featu re a n u m ber o f exh i bitions o f African, Ocea n i a n , and
Eskimo art. The d i splay of objects col l ected by the Dakar-Dj i bouti ex­
ped ition wou ld, to a large degree, fa l l i nto th i s category. A devoted group
of vo l u nteers- prospecti ve eth nographers l i ke Den i se Pau l me and fash­
ionable sixteenth-arrond issement l ad ies , amateurs of the exotic- hel ped
with the renovations. The m useu m was becom i n g c h i c . At the open i n g
o f a n e w Ocea n ian exh i bition ha l l model s from t h e great Parisian fashion
houses went on parade exotical ly and a l l uri ngly attired . The M i ssion
Dakar-Dj i bouti drew its fu nds, beyond government and Rockefe l ler
Fou ndation subventions, from private patrons of the arts (among them
the wea lthy protosurrea l ist author of Impressions d'Afrique, Raymond
Rousse l ) . Before the departure of G r i a u l e's team for its twenty-month re­
con n a i ssance, a ga l a fu nd-raiser was orga n i zed by Riviere at the C i rque
d ' H i ver, a box i n g event featu ring the "African" featherweight champ AI
Brown and attended by le tout Paris in even i n g attire. Accord ing to leg­
end the champion shadowboxed with Marce l Mauss, a l egend not en­
tire l y i mprobable (the great scholar was a good ath lete and a practitioner
of savate) . 6
These anecdotes give a sense of t h e Trocadero's extrasc ientific am­
biance around 1 93 0 . The museum was rid i n g the crest of the wave of
enthusiasm for /'art m ?gre . ? D u r i n g the twenties the term negre cou ld
embrace modern American jazz, African tribal masks, voodoo ritua l ,
Oceani a n scu l pture, a n d even pre-Co l u mbian artifacts . I t had attai ned
the proportions of what Edwa rd Sa id has cal led an "oriental ism"-a
knitted-together col l ective representation figu ring a geograph ica l ly and
h i storica l l y vague but symbo l ical ly sharp exotic world ( 1 978a) . 8 If the
notion of the African "fetish" had any mea n i ng in the twenties, it de­
scri bed not a mode of African bel i ef but rather the way in which exotic
artifacts were consu med by E u ropea n aficionados . A mask or statue or
any shred of black c u lture cou ld effectively summon a com p lete world

6. My accou nt is based largel y on personal comm u n i cations from Georges­

Henri Riviere and on h i s two memoi rs ( 1 968, 1 9 79). See also Pau l me 1 9 77 and
jam i n 1 982a.
7 . On this negrophifie see Laude 1 968 : 5 28-5 3 9 ; also Lei ri s 1 968 and
B l achere 1 98 1 . For a pa rticularly reveal i ng example see Le negre by Ph i l i ppe
Sou pa u l t ( 1 927). Sou pa u l t's negre is a kind of destructive-regenerative force,
more N i etzschean than Afro-American .
8. Said's account u nderplays the positive va l uations of the exotic freq uently
associ ated with such projections. See Chapter 1 1 .
O N ETH N O G RA P H I C S U R R E A L I S M 137

of d ream s and poss i b i l ities-passionate, rhyth m ic, concrete, mystica l ,

u nc ha i n ed : an "Africa."
By the time of the M i ssion Dakar- Dj i bouti th i s i nterest i n Africa had
become a fu l ly developed exotisme. The p u b l i c and the m u seu m s were
eager for more of a n aestheticized com mod i ty, and it was i n this c l i mate
that the F rench legis l ature was preva i l ed u pon to enact a spec i a l l aw
u n derwriting a n exped ition whose c h i ef offi c i a l tas k was to enrich the
nation's col l ecti o n s . The M ission Dakar-Dj i bouti satisfied the demand; it
brought back data that cou ld be cou nted and d isplayed (J am i n 1 982a) . 9
The eth nographers depa rted i n 1 93 1 with a structu red aesthetic i n
m i nd , a vision o f Africa, a n d a certa i n (essential ly fetish i st) conception
of how " i t" shou ld be col lected and represented . They d id not, in the
m a n ner of E n g l i s h and American fie ldworkers of the ti me, set out to ex­
perience and i nterpret d i sc rete c u ltura l wholes. Fieldwork rapport i n
Lei ri s' account ( 1 934) emerges a s l ittle more than a romantic fa ntasy; and
in G r i a u l e's report ( 1 9 3 3 ) eth nography is portrayed as a process fraught
with ro le playi ng and man i pu lation, in which power is centra l l y at sta ke
(see Chapters 2 and 6) . Even i n the l ater work of Griau le and h is col lab­
orators, w h i c h l ooks far beyond the m useu m col lecting that dom i nated
the early m iss i o n , there is l ittle attem pt to present a u n ified version of a n
African rea l ity (Gri a u l e strongly em phasized m u lti perspectiva l gro u p re­
search) free of the gaps and d i sconti n u ities of a doc u me ntary, exegetical
presentation .
The research process that bega n with the M i ssion Dakar-Dj i bouti
has prod u ced o n e of the most com p l ete descriptions of a tri bal gro u p (the
Dogan and the i r neighbors) on record anywhere . Yet, as Mary Douglas
( 1 967) has comp l a i ned, the p ictu re is curiously skewed . We can never
grasp, for i n stance, j ust how d a i l y l ife is conducted, how c i rc u mstantial
po l itical dec i s i o n s a re actua l l y made . 1 0 There is an overemphasis on

9 . Accord i ng to Rivet and Riviere's proud cal c u l ations i n Minota ure no. 2
( 1 933), 3 , 500 "ethnographic objects" were co l l ected , along w i th six thousand
photographs, a l arge col l ection of Abyss i n ian pai nti ngs, th ree h u n d red manu­
scri pts and a m u l ets, notations of thirty languages and d i a lects, and h u n d reds of
record i ngs, "ethnographic observations," botan ical spec i mens, and so on. Th is,
the m i ssion's "booty," i n Rivet and Riviere's words, was the public measure of a
successfu l m i ssion. Barthes ( 1 95 7 : 1 40) d issects the word mission : an i m perial
"mana term ," h e cal ls it, which can be appl ied to any and all colon ial u ndertak­
i ngs, giving them as req u i red a heroic, redemptive a u ra .
1 0 . This account shou ld serve as a corrective t o Douglas' tendency to por­
tray Griau l e and the French trad ition genera l l y as formal istic and enamored of

e l aborate ly c ross-referenced native theories of the way th i ngs are or

shou l d be-a myth i c conception of cosm ic order that aspires to embrace
every gestu re and deta i l of the profa ne world . The extraordi nary beauty
and conceptu a l power of Dogon wisdom, known in its fu l l ness to only a
sma l l group of elders, never sati sfies the nagging q uestion : What are the
Dogon rea l l y l i ke ? The G riau l e trad ition gives us a scru pu lously exp l i ­
cated ensemble o f documents, t h e most i mportant o f w h i c h , t h e cosmo­
gon i c myth, i s man ifestly composed by Dogon . L i ttle effort i s expended
on a natu ra l i stic account in the manner of, say, Mal i nowski 's Argonauts;
i ndeed, in the wake of su rrea l ist fragmentation what wou ld be the poi nt ?
If t h e M i s s i o n Dakar-Dj i bouti brought back considerable q uantities
of "art" for d i sp l ay at the Trocadero, its objects fou nd thei r permanent
home in a rather d i fferent m u seu m . N o sooner had Riviere completed h i s
restorations i n 1 9 34 than Rivet annou nced the approval o f a grandiose
new p l a n . The old Byzant i n e structure was to be razed to make way for
a d ream bu i ld i n g that wou ld s u b l i mate the anarc h i c cosmopol itan ism
of the twenties i nto a monumenta l u n ity: " h umanity." The Musee de
I ' Homme, a name that has only recently become m u lti ply i ronic, was i n
the m i d-th i rties an adm i rable idea l , a t once scientific and po l itical i n
sign ificance. T h e n e w institution combi ned under a s i n g l e roof t h e tech­
n ical laboratories from the Musee d ' H i stoi re Nature l l e and the l nstitut
d' Eth nologie, formerly housed at the Sorbonne. The museum brought
together a l i bera l , synthetic i m age of "man," a v ision conce ived by Rivet,
wh i c h wove together in a powerfu l symbo l i c ensemble a n u m ber of the
ideological th reads I have been trac i ng. Rivet had gathered together a
talented group of eth nologists i n c l ud i ng Metra ux, Leroi-Gourhan, Leen­
hardt, G riau le, Leiris, Schaeffner, Dieterlen, Pau l me, Lou is Dumont, and
jacq ues Souste l l e . He provided the i nstitutional su pport that, along w ith
Mauss's teach i ng, formed a center for an emergi ng fieldwork trad ition .
For most of these scholars the con nection between art and ethnography
was cru c ia l .
Mauss and Rivet's brand of h u ma n i sm envisaged a n expansion and
an ope n i n g-out of local conceptions of h u man natu re . N o one time or

abstract systems. It shou ld a lso reinforce her suggestive rapprochement between

Dogon culture and su rrea l ism. On th is correspondence see a l so Guy Davenport's
( 1 9 79) imaginative pl acement of the Dogon, along with Charles Fou rier, in 1 920s

c u lture cou ld c l a i m t o i n carnate the mankind o n display a t the Musee de

! ' Homme. The species in its total ity wou l d be represented there, begin­
n i ng with biologica l evo l ution, mov i ng through the archaeological re­
ma i n s of early c i v i l izations, and end i n g with a fu l l array of actu al cu ltural
a l ternatives . The d i fferent races and c u l tu res of the planet were to be
success ively d i splayed, a rra nged in ga l leries organ ized synthetical ly on
one s ide, ana lytica l l y on the other. Mauss's homme tota l wou l d be
brought together for the fi rst time for the ed ification of the publ i c . Also
for the i nstruction of the scientist the Musee de I ' Homme contai ned ex­
tensive research laboratories and sc holarly col l ecti ons. Less than 1 0 per­
cent of its tota l col l ection was on d i s p l ay at any given moment (see Ri­
viere 1 968, 1 9 7 9 ; Rivet 1 948 : 1 1 0-1 1 8) .
T h e wedd i ng o f sc ience and p u b l i c ed u cation with i n a progress ivist
h u m a n i s m s u ited Rivet's world view perfectly. He was a soc i a l ist with a
vision-and w ith the pol itical and soc ial con nections necessary to rea l ­
i z e i t . T h e M u see de ! ' Homme was conceived a s part o f t h e I n ternational
Exh i bition of 1 93 7 , a symbol of Popu l a r Front ideals. Rivet, whose spe­
c i alty was American archaeo l ogy and pre h istory, tended to see mankind
i n a n evo l utionary, d iffusio n i st frame, stressi n g lon g-term biocu l tu ra l de­
velopment and the reconstruction of h i storical seq uences through the
exte nsi ve co l l ection and comparison of traits . In an early arti c l e on
method, which a ppeared in Documents, he announced the underl y i n g
themes o f h is d rea m m useu m ( R i vet 1 92 9) . I n the study o f man, he
writes, the bou ndaries between eth nography, a rchaeology, and pre­
h istory are "abso l ute l y artific i a l ." ( I n a l ater version he wou ld add physi­
cal anth ropo logy to the m ixtu re . ) Eq ual ly artificial are c lassifications of
h u ma n rea l ities accord ing to the d i visions of pol itica l geography. " H u­
man ity i s an ind ivisible whole, i n space and ti me." "The science of man"
no longe r need be subd ivided a rbitrari ly. "It [is] h i gh ti me to break down
the barriers . And that is what the Musee de ! ' Homme has tried to do"
(Rivet 1 948 : 1 1 3) . The pol itical message for 1 93 7 was c lear.
The M u see de I ' H o m me provided a l i bera l , prod uctive environ ment
for the growth of French eth nographic science. Its gu i d i ng va l ues were
cosmopo l ita n , progressive, and democrati c; one of the fi rst cel ls of the
Resi stance formed with i n its wa l l s in 1 940 ( B i u menson 1 97 7 ) . The mu­
seu m encou raged international u nderstanding and globa l val ues, an ori­
entation that wou ld conti n u e after the Second World War with the
i n vo l vement of Riviere, Rivet, G r i a u l e , Le i ris, Metraux, and other eth nol-

ogi sts i n U N ESC0 . 1 1 The i rs was a cosmopol itan trad ition that had re­
mai ned congruent, i n i m portant ways, with the eth nograph ic su rrea l ism
of the twenties. It shou ld be remem bered that su rreal ism has been a gen­
u i nely i nternational phenomenon with branches on every conti nent. It
has sought the articu lation less of c u l tu ra l differences than of h u man dif­
ferences. The same can be said overa l l of French eth nographyY But
though it shared the scope of su rrea l ism, the eth nograph ic humanism of
the Musee de I ' Homme d id not adopt a n earl ier su rrea l ism's corrosive,
defam i l iariz i n g attitude toward c u l tu ral rea l ity. The aim of sc ience was
rather to col lect eth nographic artifacts and data and to d isplay them i n
reconstituted , eas i l y i nterpretable contexts . This enta i l ed losses a s we l l
as gai ns . I n deed i t i s possible to imagine a n eth nograph ic su rrea l i st c ri­
tiq ue of the Museum of Man poi nting tentatively at the shape-or rather
at the activity-of a more supple and less authoritative h u manism .
The Musee de I ' Homme's African scu l ptu res were displayed region­
a l l y along with re l ated objects, their sign i ficance function a l l y i nter­
preted . They d i d not find a pl ace beside the Picassos of the Musee d'Art
Moderne, l ocated a few streets away. As we have seen , the emergi ng
doma i n s of modern art and eth nology were more d i sti nct in 1 93 7 than a
decade before. 1 3 It i s not merely whi msical to q uestion these apparently
natura l c l assifications. At issue is the loss of a d i sruptive and creat ive p l ay
of hum an categories and d ifferences, an activity that does not s i m p l y
d isplay a n d com prehend t h e d i versity o f cu ltura l orders b u t open ly ex­
pects, a l lows, i ndeed desi res its own d i sori entation .
Such an activity i s lost i n the consol idati on and d i splay of a stable
eth nograph ic knowledge. In the twenties the knowledge brand ished by
a you nger eth nography a l l ied with surrea l ism was more eccentric, u n-

1 1 . Two characteristic U N ESCO publ ications are Interrelations of Cultures

( 1 953), with contributions by Griaule and Leiris, and Race and History by Claude
Levi -Stra us ( 1 952).
1 2 . An i m p l icitly su rrea l ist ("anthropological") conception of m i nd as a cre­
ative sou rce capable of generating the entire range of h u man express ions-both
exist i ng and potenti a l , both myth i c and rational-finds perhaps its most pro­
grammatic expression in Lev i-Stra uss's structu ra l i st esprit humain . See Chapter
1 0, section 3 .
1 3 . The d i sti nction was not ach ieved without conscious effort. Accord ing
to Mi chel Lei r i s (personal communication) i n the Musee de I ' Homme Rivet issued
a formal i njunction aga i nst treati ng artifacts aesthetical ly. The new i nstitution had
to pu rge the legacy of the Trocadero and the twenties, a period when the contexts
of science and art b led into each other. Rivet's taboo remai ned in effect unti l the
1 960s.

formed , a n d w i l l i ng to d islocate t h e orders o f its own c u lture-the cu l­

tu re that b u i l t great m useu m s of eth nographic sc ience and modern art.
The Musee de I ' Homme opened its doors to the public i n J u ne 1 93 8 .
D u r i n g t h e previous s u m mer a c u rious alternative h a d been c reated by
Bata i l le , Leiris, Roger Cai l l ois, and a loose col lection of avant-garde i n ­
tel l ectu a l s (some of t h e m students o f Mauss) who cal led themse l ves the
Col l ege de Soc io l ogie. Wh i l e the name s uggests the trad ition of Du rk­
hei m , the group's renewed i nterest i n the Annee sociologique i nvo lved a
considerab l e d egree of rei nvention . The i r tu rn toward sociology ( l ess
sharply d i sti ngu i sh ed from eth nology than i n England or the U n i ted
States) signaled a rejection of what they saw as su rrea l i sm's overidentifi­
cation with l iterat u re and art, its excessive s u bj ectivism and concern with
automatic writi ng, i nd iv i d u a l dream ex perience, and depth psychol ogy.
The Col lege de Soc i ologie-w h i c h met for two years i n the d i n i ng room
of a Lati n Quarter cafe then folded because of i nternal d i ssension and
the outbreak of war-was an attempt to rei ntegrate scientific rigor w i th
person a l experience i n the study of c u ltural processes . L i ke the a uthor of
Elementary Forms of the Re ligio us Life, the fou nders of the Col lege were
preoccu p i ed with t hose ritu a l moments when experi ences outside the
norma l flow of ex istence can fi nd col lective express ion, moments when
c u ltural order i s both transgressed and rej uvenated . They adopted the
D u rkhe i m ian concept of the sacred to c i rc u msc r i be th i s recreative do­
mai n .
I f D urkhei m d i scovered the roots of soc i a l sol idarity i n d i spl aced
ethnograph i c exam ples such as the "co l l ective effervescence" of aborig­
i na l rites, Bata i l le envi saged col lective expressions of transgress ion and
excess i n contemporary Par i s . He was obsessed with the power of sacri­
fice and with the p l ace de I a Concorde, which he hoped to rec l a i m as a
site for ritual acts orga n i zed by the Co l lege. Ca i l lois, more tem perate,
was engaged in the research that wou ld res u l t in L 'homme et /e sacre
( 1 9 3 9 ) . He wou l d lect u re to the Col lege on " I a fete," a tou r of the world's
c u ltures, d raw i n g o n h i s teachers Mauss, Georges Dumez i l , and Marcel
G ranet, as wel l as on the eth nographers A. P. E l k i n , Dary l l Forde, and
Mau rice Lee n h a rdt. Ca i l lois' d iverse sacn§ i n c l uded ritu a l express ions of
primord ia l c h aos, excess, cosmogony, fert i l ity, debauchery, incest, sac­
ri l ege, and parod ies of a l l sorts . Wh i le they shared D u rkheim's i n terest
in the constitution of col lective order, the mem bers of the Col lege de
Soc i o l ogie tended to focus on the regenerative processes of d i sorder and
the necessa ry i rr u ptions of the sacred i n everyday l ife . F rom th is stand-

point the su bvers ive critical activities of the ava nt-garde cou l d be seen
as essenti al for the l ife of soc iety; the c i rcumscribed position of "art" i n
modern c u l t u re cou l d b e transcended , a t least programmatical ly.
It is hard to genera l ize about the Col lege, a body so short- l ived and
idiosyncratic in its members h i p . Lei ris, for example, was preoccu pied
not with col lective rites but rather with those autobiograph ical moments
i n whi ch the a rticu l ation of self and soci ety can be brought to conscious­
ness . To this end he cu ltivated a k i nd of method ical c l u ms i ness, a per­
manent i nabi l ity to fit. H i s own ch ief contri bution to the Co l lege (before
resign i ng because of q u a l ms about loose sta ndards of evidence and the
danger of fou n d i n g a coterie) was an essay entitled "The Sacred in Every­
day Life " ( 1 93 8b) . In t h i s text, a bridge between eth nography and self­
portra iture, Leiris s ketched many of the topics he later developed i n La
regie du jeu ( 1 948 -1 976) . Obj ects of u n usual attraction ( h i s father's re­
vo lver), dangerous zones (the racetrack), taboo sites (the parental bed­
room), secret spots (the W. C . ) , words and ph rases with a spec ial magical
resonance-these sorts of data wou l d evoke "that ambiguous attitude
tied to the approach of someth i ng both attracti ve and da ngerous, presti­
gious and rejected , that m ixtu re of respect, des i re, and terror that can be
taken as the psychological mark of the sacred" ( lei ris 1 93 8 b : 60) .
I n L'Afrique fantome ( 1 934) Le iris sharply q uestioned certa i n scien­
tific d i sti nctions between "subj ective" and "objective" practices. Why,
he wondered , are my own reactions (my dreams, bod i l y responses, and
so on) not i m portant parts of the "data" prod uced by fie l dwork? I n the
Co l lege de Soc iologie he gl i m psed the possibi l ity of a ki nd of eth nogra­
phy, analytica l ly rigorous and poetic, focused not on the other but on the
se lf, its pecu l i ar system of symbols, rituals, and soc i a l topograph ies. The
exception wou ld be made to i l l u m inate the ru le without confirm ing it.
B u i l d i ng on the work of Robert Hertz, Lei ris and his col l eagues cu ltivated
a gauche, or l eft-handed , sense of the sacred . I n Leiris' case th is attitude
generated a l ifework of self-portraiture, an awkward and forever­
i mperfect process of socia l ization whose title, La regie du jeu, wou l d
express t h e ambiguous two-sidedness o f order t h e Col lege was con­
cerned to i n vestigate . From the late th i rties on, however, Leiris held h is
l i terary and eth nograph i c work rigorously apart. H is provocative field
journa l , L 'Afrique fantome, rema i n s an iso lated example of su rreal ist eth­
nography. (see Chapter 6) . 1 4

1 4 . A n essay that h ighl ights the "ethnographic" d i mensions of Lei ris' career
is Cl ifford 1 986c, from which parts of this d iscussion are adapted . Cha ney and
O N ETH N O G R A P H I C S U R R E A L I S M 1 43

The Co l lege de Soc iologie was freq uented by a d i verse p u b l i c that

i n c l uded Jean Wah l , Pierre K lossows k i , A l exand re Kojeve, Jean Pau l han,
J u les Mon nerot, and Wa lter Benjam i n . Long a s u bj ect of legend and m is­
i nformation, the Co l lege can now be d i sc ussed with some degree of con­
fidence tha n ks to the l a bors of Den i s Hol l ier ( 1 979), who has brought
together v i rt u a l ly every surviv i n g documentary trace of i ts exi stence . 1 5
The pictu re i s com p l ex and i n many ways sti l l mysterious; i t i s enough
here to e n u merate those concerns of the Col lege that resonate with what
I have been ca l l i n g eth nographic su rrea l ism-concerns that sti l l occu py
the marg i n s of the h u ma n sciences.
The mem bers of the Co l lege struggled in an exem p l ary way aga i n st
the opposition of i n d iv i d u a l and soc i a l knowledge ( D uvignaud 1 9 79 : 9 1 ) .
A l though they never successfu l l y reso lved a tension between sc ientifi c
rigor and the c l a i m s of activism, they nevertheless resi sted any easy com­
pro m i se with one side or the other. The Col lege envisaged a critical "eth­
nology of the q uoti d i a n ," as Jean J a m i n puts it, which cou ld react s i m u l­
taneously o n soc iety and on a gro u p of activist researchers constituted as
a kind of vanguard or i n iti atory body. In Jam i n 's s u mmary:

The n otions of d i stantiation, exoticism, representation of the other, and

d i fference a re i nflected , reworked , readjusted as a fu nction of criteria
no longer geograph i ca l or c u l tu ra l but methodological and even epis­
temological i n n ature : to make foreign what appears fam i l i ar; to study
the ritua l s and sacred sites of contem porary i n stitutions with the m i -

Pickeri n g ( 1 986a , b) offer a rich account o f a nother poss ible example o f "sur­
rea l i st eth nography" : Mass Observation, the B ritish soc i a l documentary project
of 1 93 7-1 943 . I nstigated by Charles Madge, a journa l i st and su rreal i st writer,
Tom H arrisson, an ethnographer and orn ithologist, and H u mphrey Jen n i ngs, a
documentary fi l m maker and su rrea l i st painter, Mass Observation envisaged a
comprehen sive ethnography of B ritish popu lar c u l ture conceived as a defa m i l ­
iarized, exotic worl d . Its goa l was t o mobi l ize eth nographers from a l l cl asses i n
a democ rati c expansion o f social consc iousness a n d a constant interchange of
observations. As Madge and jen n i ngs put i t, these observations, "though subjec­
tive, became object ive because the subjectivity of the observer is one of the facts
u nder observation" (quoted in Chaney and Pickeri n g 1 986a :47). The project an­
tici pated l ater conceptions of reflexi ve ethnography and anth ropology as cultura l
criti c i s m . (See Chapter 1 ; also Marcus a n d F i scher 1 986; jackson 1 98 7 . ) The
speci fic m i xtu res of soc i a l , aesthetic, and scientific aims in the i n terwar "docu­
mentary" movements of France, England, and the U n i ted States deserve system­
atic comparison . (See a l so Stott 1 9 73 . )
1 5 . T h e col lection i n c l udes texts b y Bata i l le, Cai l lois, G uastal la, Klos­
sowski, Kojeve, Lei ris, Lewitsky, Mayer, Pau l han, and Wah l , with extensive com­
mentaries by the ed itor. O n the College see a l so Lourau 1 9 74 and an exce l lent
account i n J a m i n 1 980.
1 44 D I S PLAC E M E NT S

nute attention of a n "exotic" eth nographer, and using his methods; to

become observers observing those others who are m• rselves-and
at the l i m it, this other who is oneself . . . The i rru ption of the sociol­
ogist in the field of h i s research, the i nterest devoted to h i s experi­
ence, probably constitutes the most original aspect of the Col­
lege . ( 1 980 : 1 6)

The Col lege de Sociologie, i n its conception of an ava nt-garde, ac­

tivist science, in its ded icati on to brea k i ng through the veneer of the
profane, in i ts gaucherie, and in its someti mes grand iose ambitions, was
a l ate ema nation of the su rrea l ist twenties. It offers a part i c u l arly stri king
exa m p l e of that d i mension of su rreal ism that struggled against the gra i n
o f both modern art a n d science to deploy a fu l ly eth nograph ic cu ltu ral
criti c i sm .
I f the Co l lege was u n stable, ad hoc, and amateurish, the Musee de
I ' Homme bore a l l the marks of an offi c i a l l y sanctioned , scientific, mon­
u me nta l learn i ng. In an ambivalent report on the open i ng of the i nstitu­
tion where he wou l d be e mpl oyed for the next three decades, Lei r i s
dwelt o n t h e paradox o f a m u s e u m devoted t o t h e arts o f l ife. T h e danger,
he wrote, was that " i n the service of those two abstractions ca l led Art
and Science, everyt h i n g that is l iv i n g fermentation" wou l d be "system­
atica l l y exc l uded ." Wh i le pra i s i n g the h u ma n i st, progressive a i m s of the
n ew eth nographic m useology, Leiris a l lowed h i mself a regretfu l glance
backwa rd to the old Trocadero m useu m , with its d i stinctive ambi ance
and a "certa i n fam i l i a r air (lacking d idactic rigid ity)" ( 1 938a : 344) .
On the h igh parapet of the Musee de I ' Homme, in go ld letters, are
engraved words by Pau l Va lery (wh i l e below stands the statue of a mus­
cular man subd u i ng a buffalo) :

Every man creates without knowing it, a s he breathes. But the artist is
aware of h i mself creating. His act engages his entire being. He is for­
tified by h i s wel l -loved pa i n .

A rt , now a u n iversa l essence, is d isplayed and approved b y an ideal­

i stic, confident good sense. A part i c u l a r version of h u man authenticity,
featuring personal i nteriority and romantic agony, is projected onto the
rest of the planet. A l l people c reate, love, work, wors h i p . A stable, com­
plete " h u man ity" is confirmed . 1 6 Such a whole presu pposes an omission,

1 6 . For a stingi ng critique of these assumptions see "La grande fam i l le des
hommes," in Barthes 1 95 7 .
O N ETH N O G R A P H I C S U R R E A L I S M 1 45

the excl uded sou rce of the projection . What was not d isplayed i n the
Musee de I ' Homme was the modern West, its art, i n stitutions, and tech­
n i q ues . Thus the orders of the West were everywhere present i n the Mu­
see de I ' Homme, except on d ispl ay. An i m portant i mpact was lost in the
wel l -c l assified h c. ' l s , for the m u se u m encou raged the contemplation of
m a n k i n d as a whole, see n , as it were, from a d i stance, coo l l y, tolerantly.
The identity of the West and its " h u ma n i s m " was never exh i b ited or ana­
l yzed , never open ly at issue.
To speak of "man" and the " h u ma n " is to run the risk of red u c i n g
conti ngent d ifferences t o a system o f u n i versal essences . Moreover, the
authority arrogated by the h u m a n i st too often goes unquestioned . As
Maurice Merlea u - Ponty wou ld po i nt out: "In its own eyes, Western hu­
m a n i s m is the l ove of h u m a n i ty, but to others it i s merely the custom and
i nstitution of a gro u p of men, the i r password , and somet i mes the i r battle
cry" ( 1 9 4 7 : 1 8 2 ) . The problems assoc iated with a h u m a n i st (or anthro­
pological) vision have lately become a l l too apparent. T h i rd-world voi ces
now ca l l i nto q uestion the right of any local i nte l l ectu a l trad ition to con­
struct a museum of m a n k i nd (see, for example, Adotevi 1 972-7 3 ) ; and
in France rad ical c u l tu ra l critics have announced with eq uan i m ity the
death of m a n . I can not dwe l l here on the ambigu ities of such analyses of
the h u m a n ist West and its globa l d i scou rses (see Chapter 1 1 ). One sho u l d
be wary, i n any event, o f abandon i n g too q u ickly t h e v i s i o n o f a Mauss
or a R ivet-a h u m a n ism that sti l l offers grounds for resistance to oppres­
sion and a n ecessary counsel of tolerance, com prehension, and mercy.

C ulture/Collage

To stress, as I have, the paradoxical natu re of eth nographic knowl­

edge is not necessa r i l y to abandon the ass u m ption of h u man connect­
ed ness, although it does mean q uestion i ng any stable or essential
grou nds of h u ma n s i m i l a ri ty. Anth ropo logical h u ma n i sm and eth no­
graphic su rreal ism n eed not be seen as m utua l l y excl us i ve; they are per­
haps best u nderstood as anti nom i es set with i n a transient h i storica l and
c u ltural pred icament. To state the contrast schematical ly, anthropologi­
cal h u m a n i sm beg i n s with the different and renders it-through nam i ng,
c l assifyi ng, desc r i b i ng, i n terpreti ng-comprehen s i b l e . It fam i l iarizes . An
eth nograph ic su rreal i st practice, by contrast, attac ks the fam i l iar, provok­
i ng the i rruption of othe rness-the u nexpected . The two attitudes pre­
s u ppose each other; both a re e lements w i th i n a com plex process that

generates c u l tu ral mea n i ngs, defi n i tions of se lf and other. Th i s process­

a permanent i ronic p l ay of s i m i larity and d i fference, the fam i l iar and the
strange, the here and the e l sewhere-is, as I h ave argued , c haracteristic
of globa l modern i ty.
I n explori ng th is pred icament I h ave dwelt on the practice of eth­
nographic su rrea l i sm, paying l ess attention to its converse, su rreal i st eth­
nography. Let me offer a few hypotheses concern i n g the l atter. There are
no p u re examp les, except perhaps Lei r i s' L'Afrique fantome; but I wou ld
l i ke to suggest that su rrea l ist proced u res are always present in ethno­
graphic works, though seldom expl icitly acknowledged . (For exam ple
see the addend u m to th is chapter. ) I have noted some of them i n Griaule's
documentary approach . More genera l l y the mechan i sm of col l age can
serve as a hel pfu l parad igm . I n every i ntroductory anthropology cou rse,
and i n most ethnograph ies, moments are prod uced in whic h disti nct c u l ­
tural real ities are cut from the i r contexts a n d forced i nto jarring prox i m i ty.
For exam ple i n Mal i nowski's Trobriand Islands behavior we label eco­
nomics or trade is identified with ca noe magic and myth . Ritual ex­
change va l uab les, vaygu 'a (shel l necklaces), a re j uxtaposed with the En­
gl ish c rown jewe l s . Even to bri ng an a l ien kinsh i p system i nto the
conceptu al dom a i n of Western marriage is to provoke a defa m i l iarizing
effect; but it is essentia l to d i stinguish th i s moment of metonym i c j u xta­
position from its norma l sequel , a movement of metaphorical compari­
son in wh i c h consistent grounds for s i m i l a rity and d ifference are elabo­
rated .
The s u rrea l i st moment i n eth nography is that moment in w h i c h the
possibi l ity of comparison exists in u n med iated tension with sheer i ncon­
gru i ty. This moment i s repeated l y prod uced and smoothed over in the
process of ethnograph ic comprehen sion . But to see th i s activity in terms
of co l l age is to hold the s u rrea l i st moment in view-the start l i n g co­
presence on Lautreamont's d i ssecting tabl e . Col l age bri ngs to the work
(here the eth nographic text) e l ements that conti nual l y proc l a i m thei r for­
eign ness to the context of presentation . These elements- l i ke a news­
paper c l i pp i ng or a feather-are marked as real , as col lected rather than
invented by the art ist-writer. The proced u res of (a) cutting out and (b)
assembl age are of cou rse basi c to any sem iotic message ; here they are
the message . The cuts and sutu res of the research process are left visible;
there i s no smooth i ng over or blend i ng of the work's raw data i nto a
homogeneous representation . To write ethnograph ies on the model of
col l age wou l d be to avo id the portraya l of cultures as organic wholes or
as u n ified , rea l i stic worlds subj ect to a conti nuous explanatory d i scou rse .

(G regory Bateson's Na ven is a n early and, i n the gen re, unclassifiable

exa m p l e of what I am suggesting here . On Naven as an experi ment in
eth nograph i c writi n g see Marcus 1 980 : 5 09 and 1 98 5 . ) The ethnography
as col lage wou ld leave manifest the constructivist procedu res of ethno­
gra p h i c knowledge; it wou ld be an assembl age conta i n i ng voi ces other
than the ethnographer's, as wel l as exa m p les of "found" evidence, data
not fu l ly i ntegrated with i n the work's govern i ng i nterpretation . F i n a l l y it
wou l d not exp l a i n away those elements i n the forei gn c u lture that render
the i nvesti gator's own c u ltu re new l y i n comprehensible.
The su rrea l ist elements of modern ethnography tend to go un­
acknowl edged by a science that sees itse lf engaged i n the red uction of
i ncongru ities rather than, s i m u ltaneously, i n the i r prod uction . B ut is not
every ethnographer someth i ng of a su rrea l ist, a re i nventor and resh uffler
of rea l ities? Eth nography, the science of cu ltura l jeopa rdy, presupposes a
constant wi l l i ngness to be surprised, to u nmake i nterpretive syntheses,
and to va l ue-when it comes-the u n c l assified, u n sought other.
Ethnograph i c su rreal ism and su rrea l ist eth nography are utopian con­
structs; they mock and rem i x i nstitutional defi n itions of a rt and science.
To th i n k of su rrea l ism as eth nography is to q uestion the centra l rol e of
the c reative "artist," the shaman-ge n i u s d iscovering deeper rea l ities i n
t h e psyc h i c rea l m o f d reams, myths, hal l u c i nations, a utomatic writi ng.
This role is rather d i fferent from that of the c u ltural anal yst, i nterested i n
t h e m a k i n g and u n making o f com mon codes a n d conventions. Su rrea l ­
i s m coup l ed with ethnography recovers i t s earl y vocation a s critical c u l ­
tura l pol itics, a vocation lost i n later deve lopments (Max Ernst devoting
h i s energies to design i ng a n oneiric double bed for Nelson and Happy
Rockefe l l er, the genera l prod uction of "art" for the "art world").
Eth nography com b i ned w ith s u rreal ism can no longer be seen as the
empirica l , descri ptive d i mension of anthropology, a genera l science of
the h u m a n . Nor is it the interpretation of c u ltures, for the planet can not
be seen as d i vided i nto d i sti nct, textual ized ways of l ife. Eth nography cut
with su rrea l ism emerges as the theory and practice of j u xtapositio n . It
stud ies, and i s part of, the i nvention and i nterruption of mea n i ngfu l
wholes i n works of c u ltura l i m port-export.
Two fi nal exa m ples (parables) of j u xtaposition and i nvention in the
modern world system : both ca l l for a n ethnographic su rrea l ist attitude.
The fi rst i s perhaps too fam i l ia r. Around 1 905 Picasso acq u i res a West
African mask. It is beautifu l , a l l p l a nes and cyl i nders . He d iscovers cub­
ism . (Other versions of the story locate the epiphany i n the old "Troca .")
Much i n k has been spi l led i n tryi ng to account for the rol e of African

scu l ptu re i n the emergence of cubism . D i d P icasso recognize primari l y

a forma l affi n ity ? Was /'art negre essentia l ly "ra i son nable," as he once put
it? Or was he moved-as he i ntimated much l ater-by a quasi-re l igious
"magic" felt i n African a rt? The debate cont i n ues (see Rub i n 1 984b :268-
3 3 6 ; Foster 1 98 5 : 1 8 1-208) . Whatever i n spirations and affi n ities may be
retrospectively constructed by and for Picasso it seems c l ea r that the ex­
oti c objects he co l lected were too l s su ited for doing speci fic jobs : the
projecti ng cy l i ndrical eyes of a Grebo mask, for exam ple, suggesting the
sou nd hole of a meta l guitar constru ction . A cubist sol ution to various
problems of com position wou l d doubtless have emerged without the
masks; but the fact that Picasso, Dera i n , and others noticed and appre­
c iated African a rtifacts at this h i storical moment is sign ificant. Someth i ng
new was occ u rri ng i n the presence of somet h i n g exotic. It is a common
process; for example Monet's house at G i verny overflowed with J apanese
pri nts . Around · 1 920 when /'art negre was i n vogue, an i nquest wou l d be
spon sored on the subject. Picasso repl ied i n a famous sal l y : " L'art negre ?
Con n a i s pas ! " I ndeed he had l ittle i nterest i n Africa per se . There had
bee n noth i n g essential l y negre about the masks he found powerfu l and
i n structive fifteen years ea rlier. They had come in handy for making a
d i fference.
My second example comes from the Trobri a nd I s l ands. It occu rs in
the classic eth nograph i c fi l m made by Jerry Leach and Gary Ki ldea in
col l aboration with a local Trobriand pol itical movement: Trobriand
Cricket: An Ingen ious Response to Co lo nia lis m The gentleman's game,

brought by B ritish m i ssionaries about the time Mal i nowski was on the
scene, has been taken over and m ade new. Now it i s l ud i c warfare, ex­
travaga nt sexual d i spl ay, po l itical com petition and a l l iance, parody.
Someth i n g amaz i ng has been concocted from elements of trad ition,
b u i l d i ng on the m i ssiona ries' ga me which has been "rubbished ," worki ng
in sym bo l s d rawn from the m i l itary occu pation of the i s lands during the
Second World Wa r. The fi l m takes us i nto a staged swi rl of brightly
pai nted , feathered bod ies, ba l ls, and bats . In the m idst of all th i s on a
cha i r sits the u m p i re, ca l m l y i nfluenc i ng the game with magical spe l l s .
He i s chew i n g bete l n ut, w h i c h he shares out from a stash h e l d on h i s
l a p . I t i s a bright b l u e pl astic Ad idas bag. It is beautifu l .
Perhaps a n acq uai ntance with ethnograph ic su rrea l ism can help us
see the b l ue p l astic Adidas bag as part of the same kind of i nventive
cultura l process as the African-looking masks that i n 1 907 sudden l y ap­
peared attached to the p i n k bod ies of the Demoiselles d'Avignon .

Dada Data-An Addendum

One was free to go and stand on another man 's shadow.

Excerpt from J. H . M . C. Boe l aa rs, Headh unters about Themselves : An

Ethnographic Report from Irian Ja ya . Indonesia. The H ague: Marti n u s N i ­
joff, 1 98 1 , pp. 6 7- 6 9 .

Here fo l lows a l ist o f notes about t h e various parts o f t h e h u man body.

1 . The h a i r, muku-rumb, was paid attention to when somebody was

i l l and d u ri n g the age ceremony for the c h i l d ren . In both cases the h a i r
was shaved o ff b u t i n t h e latter case repl aced b y ornaments. T h e hair of
a captured head was u sed to decorate spea rs and to make wa ist-ba nds,
qowa, for the great head h u nters .
2 . The face m ight be pa i nted for celebrations. They made use of a
butterfly pattern, rur-doka k. By pai nting the parts around the eyes i n
bright colours the eyes l ooked l i ke the black dots o n the w i n gs of a but­
3 . The eyes, kind, may represent a person as is stressed i n the ad­
vice : "The wedd i ng sago shou l d be given u nder the eyes of the s u n ,
tapaq-kind-kan . "
4 . T h e ear is assoc i ated with using one's bra i n s . A stu pid person is a
person without ears , mono-a in-mbek. The expression mono-koame,
there i s an ear, means, "we too are able to th i n k."
5 . The nose, tamangk, was spec i a l l y adorned by pieces of a she l l
a n d c l aws o f b i rds. Tamangk qana, hard nose, i s a sombre a n d deter­
m i ned face.
6 . The mouth , mem, is always associated with eating. Mem reng­
gembak, b i g mouth , does not refer to an i m pudent person but to a gl ut­
ton . The mouth has a spec i a l fu nction in the custom of taki n g a mouthfu l
of water and spri n k l i ng this over the face of a person who has lost con­
sc iousness. The gestu res of putti ng out one's tongue or of spitting on the
gro u nd in front of somebody's feet a re i n su lts, which lead to fights . The
tongue is a l so said to be a del icacy (j ust as is the bal l of the th u mb) for
ca n n i ba l s .
7 . Shruggi ng one's shou lders was not a n express ion o f ignorance but
of fear. In the p resence of men, women may stand together with h u nc hed
sho u l ders as a token of decency, but the men know better. They say, "If

they were a lone with a man, they wou ld a l l too much l i ke to have sexual
i ntercou rse."
8 . C h i n rubbing or nose rubbi ng, as among the Asmat, is not a Jaqaj
custom of greeti ng. They u sed to lay thei r right hand i n the l eft hand of
the other, who then c lasped the first man's fingers . Men kiss each other
on the cheeks but do not kiss any women, not even the i r own wives.
Women usual l y do not k i ss one another.
9 . B reath i s not associ ated with the notion of spi rit. B reath ing proves
that somebody is sti l l al ive .
1 0 . Women adorn the i r upper arms and the place between the
breasts with scarifications. The g i r l s glad ly suffered any pa i n in order to
have these marks. The men are greatl y interested in the breasts, abur,
and i n the size of the female gen i tals, jo. The women i n tu rn gossi ped
about the be l l ies, kandom, and the anuses, mo, of the men . These pa rts
of the body always tu rned up i n my l i st of words of abuse.
1 1 . C h i l d ren were not a l lowed to touch the inside of their mother's
thighs. The " i nside of the th ighs of h i s wife" was the pl ace where the
a ncestor Kapaqait took the seeds for p lanti ng vegetables. The pubic hair
of women and the fi bres of the i r peri nea l bands were smoked i n the
peace pipe. They sa id that the hymen should rema i n u ntouched u nti l
after the first menstruation . Sperm and u r i ne cou ld be used as med i c i ne.
The myth of Ujoqot rel ates how he c reated a h u man be ing by smearing
his sperm on a coconut.
1 2 . The anus, mo, had a spec ial cover, a tai l , ek, of fibres. Whenever
a man was i l l , he wou l d always ask whether he was lying decently. To
touch a man's anus was e ither an appea l to h i s strength or a very serious
i nsu lt. B rea k i n g wind proved that one had eaten too m uch . If it happened
in the presence of men o n l y it did not matter, but in the presence of
women, espec i a l l y i n the presence of one's own wife, it cou l d be per i l ­
o u s for t h e m or her. Women i ncu rred t h e r i sk o f be i ng ki l led if they
looked at the excrements of the i r husbands. Husbands feared the i r end­
less reproaches about eati ng too much .
1 3 . The pen i s , paqadi, or the pubic h a i r of a man drew l ess atten­
tion . They d id not wea r any shame cover. The term paqadi, penis, was
often heard as an ex pletive. The most stu pid th ing a person cou ld do,
they sa id, was to h u rt his own anus or pen is.
1 4 . The fl u i d ooz i ng out of a decaying corpse was not used for any
spec i a l pu rpose . The on l y thi ng that happened after a corpse had de­
cayed was that c h i l d ren were told to tram ple the ground where the burial
O N E TH N O G R A P H I C S U R R E A L I S M 151

platform had bee n . T h i s was done so that they m i ght become the worthy
successors of the deceased .
1 5 . Body odou r, espec i a l ly that of the armpits, was bel ieved to have
a spec i a l defens ive power aga i n st spi rits . A person 's shadow did not get
any attention . One was free to go and stand on another man's shadow.


1 . I am gratefu l to Renata Rosaldo for bringing Boelaars' l i st to my attention .

"Borges," he said, "could not have i m proved on this one."
2 . Ma l i nows ki was i n terested by what he cal led the "coefficient of wei rd­
ness" in cross-c u ltu ral descriptions. It had always to be ba lanced, however, by
the "coefficient of rea l ity." Other ways of l i fe shou ld be made rea l and com pre­
hensible wh i le at the same t i me preserving a sense of the i r strangeness and d if­
ference. One way of preserv i n g this strangeness was to i n c l ude data not fu l ly
contextua l i zed- random, odd facts, " i m ponderab i l i a ." Rea l i st ethnograph ies,
Mal i nowski thought, shou ld mai nta i n a prod uctive ba lance between the coeffi­
c ients of wei rd ness and rea l ity, leaving readers to c i rcle hermeneutical ly (hap­
p i l y) between plaus i b i l ity and surprise, coherence and smal l bits of data . But the
interpretive balance someti mes s l i ps; and when i t does, the i mage of the other
disintegrates i n to partial col lections of facts and j uxtaposed statements from het­
erogeneous sou rces. The l i st i ng, select i ng, sorting processes by which certa i n
kinds o f i n formation emerge a s significant become suddenly visi ble .
3 . Eth nographic l i sts tend t o indu ce reverie, t h e way Joseph Cornel l 's pa rti­
tioned " hote ls," "habitats," and " m useu ms" do: b i rds and clocks, star charts, ba l l
beari ngs, p i pes , body parts . . . An unexpected beauty can b e found i n class i fi­
cations or i n sentences l i ke "The fl u i d ooz i ng out of a decayi n g corpse was not
used for any s pec ial pu rpose." "To touch a man's anus was either an appea l to
h i s strength or a very serious insu lt."
4 . Boe laars was a m issionary-ethnographer and l i nguist (Father of the Sa­
cred Heart) for nearly ten years in I ndones ian New G u i nea.
5. Eth nograph ies generate m u ltiple read i n gs. For exa mple Trobri anders are
free to read Ma l i nowsk i 's accou nts of the i r cu lture as parodies. By selecting out
individual sentences from any c u l tu ra l description, one can eas i l y prod uce series
l i ke Boelaars' .
6. When the "coefficient of we i rd ness" floats free from the "coefficient of
rea l i ty," the res u l t is a new sort of exoticism. The strangeness that's prod uced
does not i n here i n the c u lture or world of the peoples represented . This exoticism
is d i fferent from earl ier varieties-romantic, Orienta l i st, and poetic-for what
has become i rred u c i b l y c u rious is no longer the other but cultural description
itse l f.
7 . "At 1 0 I went to Tegava, where I took pictures of a house, a grou p of
girls, and the wasi, and stud ied construction of a new house" (Ma l i nowski's Tro­
briand d iary) .
8. What's c a l l ed for, then, is an eth no (G RAPHIC) poetics . . .
. . . dire, non pas tout crument sa vision, mais par un
transfert instantane, constant, / 'echo de sa presence.

5 . A Poetics of Displacement :
Victor Segalen

THE F 1 D 1 LED account of Pau l Gaugu i n's last weeks was sent to
Paris from the Marq uesas by a you ng naval doctor who had arrived j ust
too late to meet the great rec l u se. It was a sign ificant m i ssed rendezvous,
for Victor Sega len was to become an i m portant contributor to what may
be cal l ed a postsymbol ist poetics of d isplacement. Th is poetics, d rama­
tized by Gaugu i n 's fl ight from Europe, had a l ready sent Arth u r Rimbaud
to Abyss i n i a . It wou ld prope l B l a i se Cend rars arou nd the globe, Lei ris to
Africa, Artaud to the Tara h u maras . The new poetics rejected establ ished
exoticism s-those for example of a Pierre Loti-and it d iffered from Pau l
Claudel 's q u est for a profound, " i n s ide" Connaissance de /'est. The new
poetics reckoned with more troub l i ng, less stable encou nters with the
exotic .
Born i n Brittan y i n 1 8 78, Segalen voyaged widely i n Po lynesia from
1 902 to 1 905 and i n C h i na, where he spent nearly five years before h i s
death i n 1 9 1 9 . A poet, nove l i st, archaeologist, a n d trave l writer, Sega len
partici pated in the Paris l i terary m i l ieu of late symbo l ism-but from a
d i stance. H i s work i s hard to define. An expanded gen re of travel l itera-

1 52

tu re comes c losest but can not fi n a l l y accommodate the fu l l range . Sega­

len wou ld have been most content to be cal led a writer of exotic i s m ; but
the word wou ld fi rst need to be c l eansed of its myriad con nections with
swayi ng pa lms, beaches, teem i ng markets, Ti betan monasteries, danger­
ous (Africa n , Malays i a n , Amazon ian) j u ngles, "the wisdom of the East,"
the pleasures and i ron ies of travel by rai l or on s h i pboard , and so fort h .
Sega len redefi nes exoti cism a s an "aesthetic o f t h e d i verse." Th i s i s the
subtitle of h i s long essay on the subj ect, begu n many ti mes but never
fi n i s hed . I n it he attacks the pred ictab l e na rratives and decor of most
trave l writi ng ( Loti is h i s c h ief target) . H i s own writi ng substitutes trou­
b l i ng encou nters with the u n ex pected , the strangely fam i l ia r, the un­
formed . Exot i c i s m emerges as an extension of his friend J u les de G a u l ­
tier's " Law of Bovaryism " : i n Sega len's paraph rase, " Every bei ng i n
conceivi ng o f itsel f conceives itse l f as necessari ly other t h a n it is"
( 1 9 7 8 : 2 3 -24) . Making the most of modern anomie, Segalen's exoti cist
extends and red iscovers an identity by means of a perpetual series of
deto u rs, of encou nters w i th " l e D i vers."
H i s own l ife of travel was an i ncomplete quest for a se l f among the
others- i n Po lynesi a and, most tro u b l i ngly, in C h i na . By the time he d ied
at the age of forty-one, wasted by an u n d i agnosed i l l ness, he had pro­
d uced the elements of a major "exot i c i st" oeuvre, most of it u n p u b l ished .
Over the years Sega len h a s enjoyed a secret reputati on, nou ris:1ed b y the
publ ication of new inedits. Recently, with a grow i n g sensitivity to the
epistemological problems of writi ng across c u l t u ra l boundaries, Sega­
len's m ed itations on h i s exoti c encounters have acq u i red fresh resonance
for a wide range of modern cu ltu ra l studies. (For d i scussions extend i ng
h i s s i g n i ficance, see J a m i n 1 979 and G i lsenan 1 98 6 . ) More than twenty
titles a re now in print, making it poss i b l e fi n a l l y to perceive the variety
and development of h i s writi ng, its grow i ng reflex ivity, and the troubled
q uesti o n i n g of the exot i c ist proj ect.

A trave ler's phenomenological " body" can sometimes be rather prec i se l y

l ocated . Certa i n write rs a re ha ppiest w i t h t h e view a n d conversation or­
gan ized by a compartment on a mov i n g tra i n . Sau l Ste i n berg l i ked Amer­
ica as seen from a G reyhound bus (the old mode l , without ti nted gl ass) ;
it gave h i m , he said, a "cava l ier's" perspective. T h i s was c l ose to Sega­
len's v i ewpo i nt, parti c u l a r l y in C h i n a-a d i stance both aesthetic and po-

l itica l from which to engage the other. Sega len d id not warm to the Chi­
nese. He fe lt none of the i nsti nctive sym pathy and erotic attraction he
had for Tah itians. For aristocratic C h i nese perhaps; but here he lacked
soc i a l access. Sega len preferred C h i na's monuments and what he cou l d
i magi native l y resc ue from an i m perial trad ition that seemed menaced
after 1 9 1 0 by mass violence, rebel l ion, and head long modern ization (Se­
ga len and Manceron ( 1 985 : 9 2 , 1 20, 1 3 7).
Trave l i ng through C h i n a (as portrayed i n h is texts), he seldom looks
i nto people's i ntimate l ives, l i ke a candid photographer, or rubs e l bows
with the crowd ; he is se ldom face to face with i nd ividua l s . Sega len often
seem s to be on horseback-wa l ki ng, in physical contact with the u neven
grou nd, but at a certa i n height. The mounted trave ler sees out over th i ngs
wh i le avoiding the map maker's com mand i n g overview. In a world of
gates, porta l s, and courts, the horseman rides through Chi nese places
but without presu m i n g to be " i n side Ch i na." He rejects Claudel ian par­
ticipation, knowledge of the East as co-bi rth ("co-naissance" ) . Segalen
does not experience and revea l the deep, h idden truths of C h i n a .
Mou nted a n d mobi le, he moves over a n d around i t s s u rfaces. T h e sur­
faces are com p l ex, loop ing.

I n h i s fi rst book, Les immemoriaux ( 1 907a), Sega len tried rea l i stica l ly to
evoke i nd i ge nous experience, what ethnographers of the time were be­
gi n n i ng to cal l the native poi nt of v iew. Th is is probably h i s best-known
and least characteristic work. It speaks e loquently on behalf of trad itiona l
Po lynes ia-and it fa l l s rather too easi ly i nto an elegiac lament for the
van i s h i ng prim itive. The novel 's standpo i nt is that of Teri i , a recitant, or
ora l performer of genea l ogies and myths. It begi ns with a cri s i s : Teri i
forgets, fa lters i n the m idst of an i m portant recitatio n . Th i s ruptu re of ora l
trad ition i s tied to t h e a rriva l of E u ropean s h i ps i n Tahiti, the presence of
a new, confusing power. The novel fo l lows Teri i's d i sgrace and fl ight, h i s
travels, h i s encou nters with m i ssionaries; and it ends with a tragi c prog­
nos i s : the death of "Maori Civi l ization ." Les immemoriaux is a rather suc­
cessfu l ethnographic nove l . Sega len was a sensitive observer of the c u l ­
tura l situation i n F rench Po l ynesia aro u nd t h e turn o f t h e century, a n d h i s
d iscri ptions o f trad ition a l ritu a l a re based on t h e best scholarsh i p ava i l ­
a b l e a t t h e t i m e . I n add ition there is a happy correspondence between
Sega len's own symbo l ist fasci nation for the orph ic power of ora l expres-
A P O E T I C S O F D I S PL A C E M E NT : V I C T O R S E G A L E N 155

sian, the m u s i c a l Word, and a Polynesian emphasis o n cosmogonic

speec h .

And it's the busi ness o f the n i ght-wa l kers, the haere-po with long mem­
ories, to pass on from a ltar to alta r, from priest to d i sciple the pri mal
stories and exploits that m u st never die. And so as soon as n i ght fal l s
t h e haere-po h u rry about thei r task; from every o n e o f the d iv i ne ter­
races, from every marae bu i l t on the c i rcle of beaches, a monotonous
m u r m u r rises in the darkness, mixing with the howl ing voice of the s u rf
and su rroun d i ng the isl and with a gird l e of prayers. ( 1 907a : l l )

B u t if Les immemoriaux touched a rom antic chord among its ( E u ro­

pean ) readers, certai n of its deepest associ ations may have been prob­
lematic for its author. The doomed trad ition of Polynes i a is rendered as a
sonorous world, a n environ ment of s poke n and heard i ntimacies. Such
presences were strongly but dangero u s l y attractive to Segalen . One of
h i s short fictions, " Dans u n monde sonore" ( 1 907b) , imagi nes a man
who chooses to l ive i n a darke ned room fi l l ed with subtle sounds, who
touches and apprehends s pace acousti ca l l y. Sight i n this world seems
crude and i ntrus ive. The tale's protagonist is beautifu l ly i nsane . Mo re­
over the sonorous world of Les immemoriaux is assoc i ated with an i nevi ­
tab l e c u l tu ra l death . I n C h i na Segalen moved away from th is style o f
c u ltural evocation, b u t h e fel t a constant nostalgia for the sen suous ab­
sorption assoc i ated with sou n d . H i s l ast, fragmentary poems were a se­
ries of long- l i ned odes- Thibet (he c l u ng to the aspi rated spe l l i ng)­
songs of the most exotic p lace, a p u re, transcendent echochamber
( 1 9 7 9 ) . Sega l e n never reac hed Ti bet, the u ltimate, deferred goal of a l l h i s
exped itions.

The move from Tah iti to Ch i na was a shift from the sonorous and ora l to
the v i s u a l and writte n . Sega len, who p layed the piano, com posed a bit,
and even col l aborated with Debussy, fou nd C h i n a an acoustic desert . Its
m u s i c and song repel led h i m ( 1 985 : 1 4 3 -1 44) . C h i nese speech i s bare l y
evoked i n h i s writings, but i n scriptions-ch a racters, gestu res, arc h i tec­
ture, pa i ntings-abound . It is no l onger a q u estion of evocation, of Se­
gal e n merging h i s voice with that of the other. As he put it in a l etter to
Debussy, " I n the e n d , I came here looking for neither Europe nor C h i n a
b u t for a v i s ion of C h i na" ( Bou i l ler 1 9 6 1 : 1 00) . Sega len's other is a con-

struct of des i re and a manifest fiction- l i ke its recent analogue, the "J a­
pan" of Roland Barthes in Empire of Signs ( 1 970).
Though he was a scholar and con noisseu r of thi ngs C h i n ese, Sega­
len often portrayed an u ncerta i n real ity-multiform, sh ifti ng, giving way.
H i s co l lection of travel observations written i n 1 9 1 0, Briques et tuiles
( 1 975), i s a series of d i screte encou nters, notes, and prose poems that
enact the movement of a traveler th rough a cou ntry that is, to adapt B re­
ton's phrase, an "erosion fixe ." H is fasc i nation with ru i ns is a pos itive
aestheti c of movement and process . C hina appears as l i ght su rfaces and
cru m b l i ng forms, wal ls and doors with noth ing beh i n d . Sega len wa l ks­
rides-t h rough th i s cou ntry, entranced by its wooden structures, the de­
cay accepted and bu i l t i n . (Co u l d not a French traveler see the same
today among Cal iforn ia's ru i n s ?) He mocks E u rope, where stone cathe­
dra l s are bu i lt as if for the ages: " D u ration does not come from so l id ity;
i m m utabi l ity l ives not i n you r dwe l l i ngs but i n you , slow men, ongoing
men ! " ( 1 9 7 5 : 4 7) .
H e wrote to h i s friend Henry Manceron : " I th i n k I've h it u pon a
rather sati sfactory formu l a for the art of C h i nese monuments by simply
replac i ng the stasi s the Egypti ans and G reeks have taught us to put there
with a k i nd of dynamism that m ust not be stripped of its perpetu a l l y no­
mad ic character. Houses and tem ples are sti l l tents and p latforms, j ust
waiting for the process ion to depart" ( 1 985 : 9 1 ) .
T h i s fee l i ng for the dynam i s m o f C h i nese monuments provoked a
correspond ing movement on the part of the trave ler, abandon i ng any
fixed p lace of "observation ." From Briques et tuiles (Segalen's e l l i pses
regi ster the bumpy motion) :

Pal aces, i m mobi le by accident and agai nst you r nature; l ight construc­
tions . . . can't return you to the swaying of the pl atform bearers . . .
It's I that w i l l move toward you ; and the undulation of my wa lki ng,
with each of you r cou rtya rds a station, wi l l retu rn to you the shoulders'
rhythms and the osci l l ations by which you once were ani mated . I w i l l
wa l k to you . ( p . 3 2 )

Sega len's C h i nese landscape is barely stopped motion . Mounta i ns are

"frozen waves ." He rides with fasc i nation over the "ye l l ow country" of
the north ( " I mage de Ia Ch i ne?" he wonders), a fu rrowed , cut land, yel ­
low dust i n t h e w i n d and water constantly mov i ng; everyth i n g erodes.
H is road maneuvers across the l and, tacki ng to skirt each new cave-i n or

alteration of a stream's cou rse. Segalen writes the modern experience of

d isplacement: self and othe r a seq uence of encou nters, detou rs, with the
sta b l e identity of each at issue.

Sega len's C h i na i s a m u ltiform a l l egory, a sou rce of i n c reasingly personal

(if carefu l ly eq u ivocal ) mea n i ngs . Steles ( 1 9 1 2), poems written i n the
manner of fu nerary i nscriptions, do not so much translate a C h i nese c u l ­
tura l content as provide the i r author with a n i m perso n a l , official vo ice, a
d i sgu ise a l low i n g h i m a degree of expressive freedom. Segalen is not
given to person a l statements of emotio n ; but his Peintures ( 1 9 1 6), poems
descri b i n g a series of C h i nese pa i nti ngs, are facets of an i nti mate i magi­
n ation . The "pai nti ngs" are i nsc ribed on s i l k, porce l a i n , wool, water,
even i n the a i r by a mov i n g fan . Some u n ro l l as long scro l l s . I n the
opaq u e gaze of a woman, the fee l of a tapestry, the cold su rface of a
vase, Sega len explores a ga l lery of personal fasc i nations and fears .

The pa inting that comes next isn't one that hangs h igh u p but i s
t o b e opened with a touch o f t h u m b and i n dex, l i ke t h e half-moon fan
carried in Autu m n and Spring . . . and actua l l y it's called :


Don't give it any rest : don't try to inspect it laid out flat, or cou nt
the ivory i n l ays ; but give it m ovement, a lways; stroke the air and se­
cretly, out of the corner of an eye, look into each gentle breath it sends,
bit by bit guess at the fu rtive scenes : the background, black and sh in­
ing. Sudden ly a battlement opens: wi ngs beat : eyes rol l : a sku l l caves
i n : out comes a pagoda, with a s i ngle spurt spread i n g to the open
sky . . .
D i d you see it? Fan , keep fan n i n g .
A figu re composes itself: a naked m o n k , ecstatic . j ust two eyes
are left of h i s enti re body, but they' re very m uch a l ive . (The rest is d ry
or rotted . ) He l ets u s know that what's seen, alone, is good . Fan, keep
fan n i n g . . .
Now a wide-open face stares out at you ; so magical l y and deeply
that it wi l l fix itself to you r features and may become your face if you
don't, sti l l fan n i ng, change it to something else less question ing: the
curved stroke of a Pai nter's horizon; the vast u nd u lation of the sea ;
slow wi ng-beat of the great rose goose i n the sky; the gathered ,
stri pped , spare caress of every des i re . . . Fan , keep fan n i ng . . .
1 58 D I S PLAC E M E NT S

But the pai n ted face evokes itself again, with insolence, clearer
at every turn . It gazes from too c lose. What can it mea n ? Are you
provoking it? Met anywhere else: what an i ntolerable experience ! L i ke
the appearance of an overly i nsistent friend, l i ke a too-faithfu l regret,
l i ke a m ute wa nting to ask a q uestion.
But we don't i n habit the true world . We can reject what offends
or troubles u s, effaced more eas i l y than a regret, with a quick fl ick of
the hand .
So c lose up you r fi ngers : at once, the face is gone . . . (pp. 34-

The seq uence of Ch i nese pa i nti ngs is control led by a con sciousness
that moves through a n exotic but i nti mate imaginaire. As Sega len wrote
i n a letter, "The tra nsfer from the E m p i re of C hina to the Empire of the
se lf is constant" (Bou i l ler 1 9 6 1 : 1 0) . Read i ng h i s l ater works, one begins
to suspect why he never fin ished the long-pl a n ned essay on exoticism,
"an aesthetic of the d i verse." China h ad confou nded the exoticist's q uest
for d i versity. Sega l en's C h i n a was more d i stant and mysterious than the
sensuous, acoustica l ly present world of Tah iti . But d i sta nce and mystery
wou ld not be paths to " l e D i vers ." They wou ld provoke the end less con­
struction of doubles and a l legories of the self.

Sega len never wrote a coherent theory of his exotic encou nters . I n stead
he engaged i n a series of writi ng experi ments, se lf-conscious fictional
exc ursions that probed and q uestioned the search for d i versity. Theory,
as the word's etymo l ogy m i ght i m p l y, was i n separable from displace­
ment, transfer, and trave l .
Sega len's nove l Rene Leys ( 1 922) is perhaps h i s most sustai ned se lf­
reflexive work. Th i s bri l l iant mystery story about the i m peria l Forbidden
City u nderm i nes the classic exoticist topography of barriers and thresh­
olds surroundi ng a "secret ." Rene Leys is a subtle med itation on depth­
truth, its disclosure, and the end l ess wi l l to know. The narrator, named
Sega len (the novel is loosely based on rea l events he witnessed i n Peking
d u r i ng 1 9 1 2 , the last days of the Ch' i ng dynasty), is obsessed with the
Forbidden City and with the h idden center of Ch ina, the emperor. He
m ust know everyth i n g possi ble about "The With i n ." (Sega len in fact
d reamed of writi ng a large book on C h i n a as seen by the emperor, Le fils
du cie/.) A you ng Belgian named Rene Leys who has grown up in Peking

and is a master of l a n guages (and of role play i ng) serves as i ntermed iary.
The youth passes i n and out of the wal led com pound, revea l i ng to the
avid n arrator more and more amaz i n g stories of sed uction and i ntrigue.
In a n echo of Lot i 's famous seduction of Az iyade, v i o l ating the Tu rkish
harem, Rene Leys becomes the secret lover of the empress. I n a forbid­
den place the u l t i mate (female) other is possessed .
Soon, however, . t h i s fam i l ia r Orienta l ist d rama goes awry. Segalen's
knowledge is vicarious. He and Rene Leys a re doubles, sec ret sharers. A
too-i nti mate u ndersta n d i n g u n ites them, makes them com p l i c it-we are
led to s u spect- i n the very i nvention and revel ation of the exotic
"With i n ." The narrative begins to u n rave l . Dou bts emerge; stories be­
com e contrad ictory. We begi n to q u estion the exi stence of any secret or
centra l truth i nside the pal ace, a world that emerges as m u ltifarious and
labyri nth i ne, where no one, espec i a l l y not the em peror, can know a l l
that goes o n . A t the same t i m e w e are unable to d is m i ss what w e hear as
l ies and fantasy. The story has too m u c h h i storical spec ific ity, fo l l ow i n g
very c l osel y as it does t h e overth row o f the e m p i re . Rene Leys is final ly
ki l l ed for his " i ns ide" activities : there m u st have been someth ing; it can­
not all h ave been made u p .
Rene Leys mai nta i n s a subtle u ncertai nty a s t o what if anyth i ng goes
on with i n the pa lace. We are fi nal l y brought to see the sed uctive, even
leth a l force of the narrator's d es i re for know l edge, penetration, and d is­
closu re . I ndeed the pa rable resonates widely: cou ntless stories of con­
cea l ment, reve l ation, and i n itiation are structu red by a s i m i l a r d es i re that
posits secrets i n order to revea l them, i magi nes an other with a true
"with i n ." By the end of Rene Leys there are no more u lti mate depths: the
sea rc h for revelations i s shown to be end l ess. What rem a i n s are su rfaces,
m i rrors, doubles-an eth nography of signs without essential content.

Segalen's i magi nary detou rs t h rough C h i n a wou ld become more and

more obviously persona l . B y the end of h i s l ife the search for d ivers ity
retu rned h i m relentlessl y to h i mself, to h i s fam i l iar obsessions . Severa l
late texts d ra matize th i s short c i rc u it of " l e D ivers." The death of Sega l e n 's
you n g confidant i n Rene Leys s i gn ifies, among other t h i n gs, the end of
that part of h i s bei ng that cou l d " pass" with i n an exotic Forbidden C ity,
that cou l d bel ieve i n the poss i b i l ity of sharing other l i ves, of erotica l ly
possessi ng the other, of shedd i ng a given identity. L ittl e rema ins of the
exoticist project. The nove l 's end i n g is i nfu sed with l ucid sadness, a po i-

gnant sense of l oss . ( Perhaps a ppropriately Segalen's own l i fe wou ld end

in u ncerta i nty. H is strange fi n a l i l l ness, l in ked to a s p i ritual crisis, remains
obscure. C l audel wou l d u rge him toward a recon c i l iation with Catho l i ­
c i s m , without success . Ru mors o f s u i c ide pers ist. )
Segalen's last fi n i shed work was Equipee ( 1 929); t h e title means
someth i n g bet-.veen "trek" and "escapade ." It records his longest archae­
ological exped ition, j ust before the F i rst World War, a journey that ap­
proached but d id not cross i nto Ti bet. Subtitled "Voyage au pays d u ree l ,"
Equipee i s Sega len's most d i rectl y personal work of travel writi ng. The
fi rst-person s i ngu lar preva i l s . But th i s je is fa r from s i m p l e : it moves
th rough the C h i nese l andscape in two d i stinct registers . Equipee records
a permanent alternation bet\.veen " imagi nary" and " real"-"betv·:een the
s u m m i t conq uered by a metaphor and heights ard uously gai ned by the
l egs" (p. 1 2 ) , bet\.veen what one seeks and what i s grasped . Th i s is not
s i mply a matter of i l l usion versu s rea l ity or of "menta l " aga i nst "phys i ca l "
events . Rather i t i s a process o f desi re, a forever-un satisfied quest for
d ivers ity constituting the body and s u bj ectivity of the traveler.
I n the m o u nta i nous l andscapes of the south , so distant from the
world around Pek i ng, Segalen u n learns m u c h of what he knew of C h i n a .
B ut h e seldom desc r i bes t h e p laces a 11 d people he meets, a s if they cou ld
be held at a d i sta nce, p i ctu red i n deta i l . Equipee provides i nstead a s u b­
jective rhyth m-the perceptions and fee l ings of a body mov i ng through
a s pace that i s both real and visionary. If at ti mes it i s un c lear whether
what i s evoked is an extern a l perception or a d ream, the n arrative sti l l
preserves a n i rred u c i b l e concreteness. Th i s q u a l i ty i n heres i n i ts beauti­
fu l ly articulated steps-the va riable stages of the journey, each a negoti­
ation with the rea l .
For Sega len true d ivers ity is not what has been preceded as exotic
or "Ch i nese" but rather the sensations and desi res that surprise h i m and
seek h i m out. In the mou nta i n s near Ti bet the exhausted narrator of Equi­
pee fi n a l ly encounters the Autre, spel led now with a capital A-the end
of h i s long " i n itiation au ree l " (p. 1 2 1 ) This Other sends h i m back on h is

tracks. Met on the path a strangely fam i l iar man, blond, fifteen years
you nger, wa ndering, " ready for anyth i ng, ready for other pl aces, ready
to l ive other poss i b i l ities . . . ." Victor Segalen starti ng out for Tah iti .
Is Segalen's quest for d iversity fi nal l y trapped by a field of s u bjective
desi res ? Yes and no. H is writing departs i n search of the Diverse, only to

confront the Same i n new gu i ses . Each time, h owever, there i s a sma l l
difference. Segalen encounters doubles a n d reflections, but the m i rrors
a re never perfect. A d ispl acement occ u rs . By the end of h i s career the
self, not the other, has become exotic . It is th i s ope n i ng of a fissu re in the
s u bj ect, however s l ight-a passage in time, a su rpri s i n g angle of vi­
sion-that constitutes " l e D i vers ."
Sega len someti m es writes i n stereotypic terms a bout the tropics. I n
a l ette r from C h i n a he d reams o f a retu rn to Po lynes i a , remem bering its
sensuous ease : "The who le i s land came to me l i ke a woman. And i ndeed
woman out the re gave me gifts that whole cou n tries can't give anymore
. . . I knew caresses and rendez-vous, l i berties th at req u i red noth­
i n g more than a vo ice, eyes, a mouth , and lovely ch i l d ish words"
( 1 985 : 1 06) . Th is vision of the other as fem i n i zed and c h i l d is h is an ob­
vious projection . The exotic is domesticated to male yearn i ng .
Even fam i l i a r Orienta l ist vis ions refract strange l y i n Segalen . H e
presses h i s des i re t o an i m possible l i m i t and t h u s t o its possi b l e su bver­
sion . Sexua l possession and easy eroticism are n ot h i s u lti mate goa l s . H i s
"veritable amoureuse" i s a you ng gi rl , a v i rgi n : "My Essai sur l exo tisme '

wi l l say it: the you n g girl is farthest from us, thu s i ncom parably precious
to all devotees of the D i verse" (pp. 1 06-1 07). The most d istant, i n acces­
s i b l e , tabooed obj ect provokes h i s strongest des i re, a des i re that does not
a i m at penetration or possession. I n 1 9 1 2 he wrote from Tientsi n that h i s
wife wou l d soon b e givi ng bi rth to " a c h i l d I hope wi l l b e o f the fem i n i ne
sex, for pure reasons of exoticism" (p. 1 1 9) . Segalen's exotic "amou­
reu se" was not s i m p l y the c h i ldish woman of color, of the harem , Rene
Leys's em press, the yiel d i n g female, a forbidden place to be entered . Th i s
object was more com plex.

What fol lows i s another


of a girl, p l a i n ly;-th i s h a i r style, th i s carriage ! Even these eyes looking

straight at you and me . . . or perhaps over our shou lders i nto the
space beh i n d ? ( Don't turn arou n d . )
T h i s face reveal s no emotio n . T h e del icate brow i s smooth ; the
eyebrows sedatel y a rched ; the lids pu rsed or open . . . look aga i n :
t h i s chaste curve o f t h e shou lders, a n d t h e hands c lasped arou nd the
stomach from decency and a good education, as if for a bow she's
abou t to make, or to h ide an encumbering pregnancy. A l l i n a l l , a great

Sti l l , you 'd l i ke to know what vision o r turn of m i nd gives her

whole young body th i s d iscreet demeanor . . .
Very wel l ! Look her stra ight i n the eyes, as she seems to be doing
to u s . If the Pa inter is equal to the masters (to the one who enclosed i n
t h e pupil o f a cowherd t h e perfect i mage o f a cow, with spots, h ide,
and halter), if the Painter has been ski l led and c lever, the R E F L ECTION
I N T H E EYES should conta i n everything they see or d ream. So gaze i n to
them, from very close . . .

Oh-th i s m i n ute m i rage, marvelous and magical l y enclosed in

the shining l i ttle shield ! Accord i n g to the commentary, we d iscern
"two g i rls, naked from breast to foot, one on the knees of the other
who crad les and caresses her." (We can even d i stinguish the fi nger­
tips ! ) What integrity in the Pai n ter's craft! So this is the scene the pu re
face reflects, decently contemplates .
But the eyes are sti l l riveted to our own . So what's the sou rce of the
reflection ?
Our own eyes? The space beh i nd u s ? ( 1 9 1 6:4 1-43)

The l ast l i ne of " Reflection in the Eyes" suggests a cruc i a l u ncer­

ta inty. Does Segalen see i n the maskl i ke face someth i ng of h i s own
psyche projected , its "deep" wishes, a perverse u nconscious revea l ed ?
Or does h e d iscern someth i n g h idden b y the gesture of turning toward
that part i c u l a r face, the act of position i ng a se lf vi s-a-vis an other? Is the
self com posed of i n ner depths or of s pecific encounters with alterity that
prod uce areas of bl indness and potentia l i nsight? U nderstood psycholog­
ica l l y, the i magi ned pai nti ng reflects repressed fee l i ngs about sexual ity.
The on l ooker sees o n l y what is a l ready i n h i s own eyes, and the other
becomes a screen for projected des i res. Segalen's vision of two naked
gi r ls is a c l iche of pornograph i c voyeurism.
We are l eft, however, with a question rather than a revelation. The
scene's poss i b l e d i splacement suggests that the pai nting regi sters not a
psyc h i c projection but someth i ng h i dden "beh i nd" the se lf i n the spec ific
act of i magi n i ng a n en igmatic woman . I n his turn toward a fem i n i ne
other the heterosexual male turns away from l esbian eroti c i s m . What i s
occ u l ted i s not a pha ntom or h idden des i re b u t t h e rea l possibi l ity o f a
female sexual ity i ndependent of the male. G l i mpsed as a reflection i n the
other's eyes, th is i magined rea lity confu ses the dom i nant h i storica l cate­
gory : woman as oppos ite sex, myste ry, and object of man's des i re. The
heterosexual exotic wavers .

Sega l e n 's program of exoticism is a fa i l u re . There is no escape; nei­

ther is there a stable home. The fai l u re enacted i n Segalen's poetics of
d is placement is both an epitome and a critique of the wh ite man's re­
lentless q uest for h i mself.


Severa l years ago, w h i l e doing arc h ival researc h on the h i story of

ethnographi c photographs, I found in a file a face that stuc k l i ke "an
ove rly i n s i stent friend, l i ke a too-fa ithfu l regret, l i ke a m ute wanti ng to
ask a q u estion ." No amount of fl ipping through other fi les-cou ntless
i mages of I n d ians, Africans, Melanesians, Eski mos-could fan this face
away. Nor cou ld I penetrate its fixed, eloquent si lence.
The arch ive's caption records an " lgorot Man" (brought from the
P h i l i pp i ne H ighlands to be ex h i bited at the 1 904 World's Fa i r in St.
Lou is) . If we look i nti mately i nto th is face, what d i stu rbances appear
beh i n d ? (Don't turn arou n d . )
Igorot man, Philippines, exhibited at the 1 904 St. Louis
World's Fair.
Guinee, de ton cri, de ta main, de ta patience
il nous reste toujours des terres arbitraires.

6. Tell about Your Trip :

Michel Leiris

L' A F R I Q U E F A N TOM E is a monster: 533 dense pages of eth nography,

trave l d i a ry, self-exp loration, "onei rography." Take the book's prif�re d'in­
s ere r a p u b l i c ity flyer s l i pped between the pages of the fi n i s hed work.

Throughout his career M i c h e l Le iris has c u ltivated this m i c roscopic

gen re : the cool essay that descri bes a book to which it is both i nti mate l y
and bare l y attached, l ead ing or m i slead i n g its readers, perm itting the
writer to cover his tracks . Of late the p rie re d'inserer has come to be
pri nted o n the back, or jacket flap, of p u b l ished books-an i m mobi l i­
zation Lei r i s regrets . That of the fi rst ed ition of L'Afrique fantome ( 1 934)
was a l oose sheet :

Sick of h i s l ife i n Paris, viewing travel as poeti c adventure, a

method of concrete knowledge, an ordea l , a sym bo l i c way to stop
growi ng o l d , to deny time by crossi ng space, the a uthor, i nterested i n
eth nography for t h e va l u e he gives that science i n t h e clarification of
h u ma n rel ations, joins a scientific exped ition c rossing Africa.
What does he fi nd ?

1 65

Few adventu res, research that i n itially excites h i m but soon re­
veal s itself too i n h u man to be satisfyi ng, an i ncreased erotic obsession,
an emotional void of growing proportions. Despite h i s d i staste for c iv­
i l ized people and for the l i fe of metropol itan cities, by the end of h i s
jou rney he yearns for t h e return .
H i s attempted escape has been a complete fai l u re, and anyway
he no longer bel ieves in the value of escape. Even with capital ism's
increasi ng tendency to render a l l true human contacts i mpossi ble, isn't
it with i n his own c i v i l ization that a Westerner can find opportu n ities
for self-rea l ization at the emotional level ? In any case he wi l l learn
aga i n that here as everywhere else man can not escape h i s i solation :
the result bei ng that he wi l l sta rt out aga i n , one day or another, caught
up in new phantoms-but this time without i l l usions. Such is the
schema of the work the author would perhaps have written if, con­
cerned above a l l to offer as objective and s i ncere a document as pos­
sible, he had not stuck to h i s travel notebook, publish i ng it as is.
This schema i s perceptible, at least i n latent form, th roughout a
journal i n which are noted, pel l-me l l , events, observations, fee l i n gs,
dreams, ideas.
It's u p to the reader to d i scover the germs of a com ing to con­
sc iousness attai ned only wel l after the return, wh i l e at the same time
fol lowi n g the author among peoples, sites, vicissitudes from the Atlan­
tic to the Red Sea . ( Leiris 1 966a: 54-55)

The priere d'inserer i s u n bound, neither preface nor conc lusion,

written for readers without the time to read-journal ed itors, booksel lers,
d istri butors, rev iewers. (The customs agents of genre : where to pl ace this
awkward Afrique ?) And for the cu rious tu rners of pages a sma l l sheet
fl utteri ng out i nto wastebaskets . The author desc ri bes pages destined for
anonymous readers : a c hance to start them on the ri ght track, to tel l what
(whom) the book i s about, to give the pages, fi n a l l y, a subject. A l ast
cha nce to say what was bei n g said, to evoke a sc hema, the story he'd
intended to write . ( B ut this author descri bes the story he had not i ntended
and that he refused to write . ) A cha nce to begin writing aga i n . . .
F i fty years l ater, w i th the h e l p of a new i ntrod uctory explanation and
yet another " Preambu le," it i s sti l l hard to know what to make of the 638
entries of th i s book that isn't one: " It's up to the reader to d i scover the
germs of a com i n g to consc iousness attai ned only we l l after the retu rn,
wh i l e at the same time fol lowing the author among peoples, sites, vicis-

situdes . . ." An i m poss i ble double read i n g : for if we keep i n m i nd the

n arrative shape offered (a lways) by h i ndsi ght, we can not fo l low the jour­
n a l 's m u ltid i rectional peri p l u s ; and if we do give oursel ves to these ad
hoc wanderi ngs, then the creation of any story to account for them be­
comes problemati c . The author refuses to narrate the scraps of experi­
ence, pu b l i s h i n g them te l que/, in chrono logica l series-as if th i s cou ld
solve the u lti mate d i lemma of giving public form to person a l experiences
without betrayi n g thei r pec u l iar l ived authenticity. Leiris to reader:
"Warn i ng-th is book i s u n readable."
" . . . as objective and s i n cere a docu ment as poss ible." L 'Afrique
fantome wi l l not amass its objects as if they were artifacts desti ned for
waiti ng m useu m cases . I ts eth nographic co l lecti ng is without c lea r
gu idel i nes, aesthetic or scientific. Nor can its pages reflect an authorita­
tive viewpo i nt or adopt a d i spassionate tone : they must contrad ict one
another. And they wi l l be stra nge l y metic u l o us: "My boots are m uddy,
my h a i r long, my n a i l s d i rty. B ut I enjoy th is fi lth, where everyth i ng I love
becomes so p u re and d ista nt" (p. 2 8 7 ) . By excess of s u bj ectivity, a kind
of objectivity is guaranteed-that (paradoxica l l y) of a personal eth nog­
raphy. The rea l ist i magi nation , fabricator of the vra isemblab/e, is refused
in favor of an i m poss i b l y s i ncere record of the rea l : perceptions, moods,
In Africa Leiris beg i n s to keep field notes on h i mse l f, or more pre­
c i se l y on an u ncerta i n existence. These notes, on carefu l ly col l ated
cards, wi l l form the data for L 'age d'homme (Manhood, translated by
Richard Howard) and fou r vo l u mes of La re gie du jeu : not autobiogra­
p h i es but col lections of "facts and i mages which I refused to exploit by
l etti ng my i magi nation work on them ; in other words, the negation of a
nove l . To reject a l l fables . . . noth i n g but these facts and a l l these facts"
( 1 946 : 1 5 6) .
" Rien q u e ces fa its ." " B ut a voyage m u st be told . It cannot be a heap
of observations, notes, souve n i rs-the pieces are d isplayed in se­
quences . A jou rney makes sense as a "com i n g to consciousness" ; its
story hardens a rou nd an identity. (Te l l us about you r tri p ! ) B ut what if one
refuses to tel l ? ( L i ke every ch i ld Lei ris has learned to tel l a proper story.
What d i d you do i n schoo l ? No, i t's not i m portant to say j ust what hap­
pened , that you wen t to the c l assroom, that it was hot and bor i n g, there
were fl ies, you sha rpened you r penc i l , went to the b l ac kboard . And you
don't have to recal l a l l the l i ttle thi ngs that were love l y or that set you on
edge : a bi rd's w i n g through the wi ndowpa ne, an ugly tu rd in the lava-
1 68 D I S P L A CE M E N T S

tory. ) " F rom the start, writing th i s jou rna l , I 've struggled agai nst a poison :
the idea of publ ication" ( 1 9 3 4 : 2 1 5 ) .
Wou ld it b e enough to retu rn from Africa, l i ke Con rad's Marlow, with
only a s i ngle potent word ? What sorts of erasu res, l i es, are necessary to
make an acceptable story? Or cou ld one outmaneuver narrative and
somehow tel l a l l , transc ri b i n g with eq ual rigor the boring, the pass ionate,
the i n teresting, the u nexpected , the bana l ? Another way of tel l ing: as if a
thousand snaps hots cou ld testify to the rea l i n their own way : this was .
c;a a ete. Et r;a, r;a, r;a . "To be i n facts l i ke a c h i l d . Th at's where I ' d l i ke to
get" ( 1 934 : 2 3 4 ) . Desi re for a regress ion to exi stence before the need to
col lect oneself, to account for th i ngs and one's l ife.
But L 'Afrique fa n to m e portrays the su rrea l i st-ethnographer en­
meshed i n writi ng- h i mself through the others . Toward the end of an
i ntense period of resea rch on za r possess ion in Eth iopia, a sacrifice is
made spec i fica l ly for Lei ris. His jou rna l records that he tasted the blood
of the a n i mal but d i d not perform the gourri, the dance of the possessed .
We see h i m seated a mong the zar adepts, the room th ick with i ncense,
sweat, and perfu me. H i s head is smeared with butter, and-as requ i red
by ritua l-the dead a n i m a l 's entra i l s are co i led around h i s brow. He does
not, however, i nterrupt h i s note taki ng.

Le iris holds the superb title of "secretary-archivist" for the Dakar- Dj ibouti
m i ssion . As such he is expected to produce a h i story of the exped ition
and its h i storic crossi ng of the Dark Conti nent; but this story is, i n effect,
a l ready i nscribed before he has taken down a single note or written out
his fi rst identification card for one of the 3 , 600 objects the m ission wi l l
acq u i re . A narrati ve i s i m pl i c i t i n the very name of the u ndertaking: Mis­
sion Dakar-Dj i bouti . Mission fu nctions as an al l-pu rpose term for any
redem ptive colon i a l errand, whether m i l itary, evangel i ca l , ed ucational,
med ica l , or eth nographic (see Barthes 1 979) . It suggests hundreds of
other voyages, a l l of them heroic, confident gestures of a stable subject
who conquers, i n structs, converts, decri bes, adm i res, represents . . .
other people and the i r wor lds.
"Ne v i sitez pas ! ' Exposition Colon i a l e" (surrea l i st slogan of 1 93 1 ).
J ust as the Dakar- Dj i bouti team is prepa ring for its departu re, an
enormous panoply of exotic worlds is being laid out i n the boi s de Vi n­
cennes. Pav i l ions from a l l the colon ies, costumes, statues, masks, curi-

osities of every sort, "savage dances" regal e the trave ler i n a l and of wel l ­
ordered enchantment. Offi c i a l m arked paths l ead the vis itor without
confusion from one outpost of progress to the next- I ndoc h i na, French
West Africa, Madagascar, N ew Ca ledon i a, G u i nea, Marti n i q ue, Re­
u n io n . A h i story of the M i ssion Dakar-Dj i bouti, the one Leiris is expected
to write, of an exped ition passing through th i rteen African cou ntries of
wh ich ten are under Frenc h dom i nation risks appearing to be this k i nd of
Then Eth iopia, never colon i zed, i nterrupts the exped ition's smooth
progress and provokes the longest, most troubled pages from the pen of
its secretaire-arch iviste. Here the m i ssion encou nters the fi rst serious ob­
stac les to its a uthority; it m ust a lter its cou rse, make the best of a tense
po l itical s i tuation . At Gondar Leiris grapples with the s h i fting ro les, de­
cepti ons; and u ndomesti cated eroticism of h i s work with the zar adepts ;
and he loses for good whatever sh reds rem a i n of the confidence needed
to shape an authoritati ve story about Africa. The narrative i m p l ied in the
m i ssion's name u n rave l s i n the day-to-day ephemera of h i s journ a l .
To be replaced by what? Le i r i s has for some time been strugg l i n g
aga i n st certa i n na rrati ve pos itions, standpoi nts fi rm ly ass igned t o whites
in the colon ies, whatever the i r personal po l itical or aesthetic proc l ivities.
Early in the trip at a performance of d r u m m i n g and danci n g : " I rem a i n
for a moment, lost i n t h e crowd , then , seeing that a seat is reserved for
me bes ide the adm i n i strator, I dec ide, with many hes itations, to take it"
( 1 934 : 3 2 ) .
I f t h e co l o n i a l standpoint c a n b e recogn ized a n d , to a degree, held
at a d i stance, others are less perceptible. It is not u nti l l ate in the voyage
that Le i ris breaks with the a l ternative, l i bera l stance offered by scienti fic
eth nography, a d iscu rsive position that " u nderstands" Africa, its peop l es,
and its cu ltures, i n the i r own terms if poss i b l e . Eth nography studies its
objects sym pathetica l ly, systematical ly. " I ntense work, to which I give
myself with a certa i n ass i d uousness, but without an ounce of passion . I'd
rather be possessed than study possessed people, have carn a l knowledge
of a 'zari ne,' rather than sc ientifica l l y know a l l about her. For me, ab­
stract know l edge wi l l never be anyth i ng but a second best" ( 1 934 : 3 24 ) .
Sti l l another position from w h i c h , confidently, t o tel l a story is of­
fered by the voyager who goes native and returns to evoke i n itiation, loss
of self, terror, e n l ightenment. Before leaving for Africa Lei ris had been
i m p ressed by Wi l l iam Seabrook's adventu re story of H a i tian voodoo, The

Magic Island (translated i n to French i n 1 92 9 ) . Seabrook appears i n a pho­

tograph beside a voodoo a ltar with a cross in blood on h i s brow, the sign
of h i s i n itiation . Le i r i s rereads the traveler's fantastic African tal e Les se­
crets de Ia jungle ( 1 93 1 ) d u r i n g an i n term i nable delay at the border of
the Anglo-Egyptian Sudan . Aga i n he is seduced by this " rather bri l l i ant
fantasy" ( 1 934 : 2 02). But a certa i n pudeur always seems to restrai n Leiris,
who, i n any event, a ppears to derive as much i nspi ration from Notes and
Queries on An thropology (read i ng W. H. R. Rivers on Freud's and j u ng's
theories of d reams, he i s gu ided i n h i s ongo i ng se lf-eth nography) and
from Pickwick Papers, fou n d by cha nce in a guest house.
Stuck at the Eth iopian borde r, read i n g whatever turns u p and scrib­
b l i ng to fi l l the ti me, Le iris becomes preoccu pied with the kind of narra­
tive he is col l ecti ng. Which of a l l the poss i b l e enunc iative positions
shou l d the rel uctant h istorian adopt and which avoid? How not write the
travelogue, the adventure story, the grand reportage, the utopia, the p i l ­
gri mage, t h e ecstatic (or i ron ic) access t o wisdom, t h e eth nograph i c fable
of rapport, the h u man ist rite of passage, the scientific myth of d i scovery,
the q uest (for woma n , for the bizarre, for sufferi ng, for art, for renewa l ,
for an authentic voice) ? We come across l i sts of " i m agerie afri ca i ne" (to
be forgotten)-Prester j o hn, death of Livi ngston, Fachoda, Ri mbaud,
Kitc hener, Raymond Roussel 's Impressions d'Afrique, " l es amazones de
Behanz i n " . . . (p. 2 94) .
Lei r i s passes the s l ac k days drafti ng prefaces (two of which appear
in the m idst of L 'Afrique fantome) . In add ition to issues of gen re and
narrative form he worries about pri nci p les of i nc l usion and exc l usion.
He defends a rigorous s u bjectivity, the right (the d uty) to record the
course of a d ream o r a bowe l movement-along with observations of the
locale, events of the m i ssion, and scientific i n q u i ries . He wi l l l eave h i s
text open t o objective chance, record i ng whatever ideas, problems, or
fantasies i m pose themselves.

Leiris conti n ues to searc h , however, for a satisfactory way of tel l i ng-of
col lecti ng and d i splayi ng-an exi stence. The l ast pages of L'Afrique fan­
tome conta i ns a s ketc h for a nove l centering on a patent alter ego, a
character named after Axel Heyst from Con rad's Victory. Heyst enacts
Lei ris' various sexua l obsessions and fears-his worries about the i m mi­
nent return to E u rope, reu n ification with his wife, the eternal problem of
measuring u p to an obscu re, pun itive standard of man hood . The convo-

l utions of the p l ot are i ntrigu i ng, i f i ncon c l u sive (pp. 499-5 04) . More
i m portant is the i m p l icit n arrative model for the work, wh ich prefigures
lei ris' later l iterary prod uctions.
The nove l 's projected form owes less to Con rad's Victory than to
Hea rt of Darkness, a tale leiris m u c h ad m i red (p. 1 96) . l i ke Con rad he
portrays the death of a mysterious colon ial figure ( HeysUKu rtz) as seen
by a second character ( " l e docte u r" /Marlow) who pieces together h i s
story from fragments - l etters, doc u ments, hearsay, a n d a n el usive per­
sonal contact. Once a p l a u s i b l e account of the protagon ist's death is es­
tab l ished, the second figu re fabricates a fa l se vers ion for use in a partic­
u l a r context where it wi l l be bel i evable. The enacted process of
col l ecti n g and tel l i ng a personal story becomes itself the focu s of narra­
tion . lei ris' n ovel out l i ne i nc l udes the l aborious documentation of a l ife
story, the lie of any s i ngle vers ion of it, and the i nterp lay of character,
writer, and aud ience i n its m ise en scene.
A theatrical conception of the subj ect appears l ater i n Lei ris' sc hol­
arly recko n i n g with h i s zar research , its ambiguous, d i stu rbi ng p l ay of
roles: La possession et ses aspects theatra ux chez les Ethiopiens de Con­
dar ( 1 958). I ndeed his l iterary works a l ways m a n i fest their "aspects thea­
traux," giv i n g freq uent gl i m pses beh i nd the scene of writing. Leiris' prac­
tice resem bles that of a d i sci p l i ned actor, comb i n i ng s i m u ltaneously
d i ss i m u lation and s i n cerity i n a q uest for presence that never q u ite
comes off.
Th is d isci p l i ne is v i s i b l e in the seq uel to L 'Afrique fantome. Man­
hood ( 1 946) adopts a narrative form that successfu l l y d raws on both the
i nti m ate jou rna l and the nove l i stic fiction w h i l e fa l l i ng i nto neither gen re .
I n t h e book's fi rst priere d'inserer (i nserted i n a l ater prefatory essay, "On
literatu re Considered as a B u l lfight") the author sti l l seeks a way to
"speak of h i mse l f with a maxi m u m of l uc i d i ty and si ncerity." He does this
paradox i ca l l y, though, by avoid i n g forms that present themselves as
expressions of a self-reve l atory s u bject. Le i ri s turns our attention away
from an authentic voice to " I ' objet fabri q u e," a blatant self-c reation that
he offers, dead pan, to the publ i c . Manhood, a novel of education, ends
with the emergence not of a n identity but of a personage . It stops short,
u nfi n i s hed, with words q uoted from a d ream : "I expl a i n to my m istress
how necessary it is to construct a wa l l arou nd oneself by means of
clothi ng."
The "si ncerity" Lei ris seeks has as l ittle to do with the romantic no­
tion of confession (an u n med i ated true speech) as the "objectivity" he

cu ltivates has to d o with scientific detachment. I n each case the author

seems to accept a ru le of p u b l i c compo rtment but then, by pressi n g it
rigorously, e laborately to its l i m it, exposes the proceed ing as yet another
ruse of a subjectivity in process, forever te l l ing and rete l l i ng itself.
("Ruse" is not q u i te right, for there is a lways another turn by which Lei ris
convi nces us somehow of the s i m p l icity of the u ndertaking.) "The undis­
s i m u l ated use of rhetoric," Leiris' ph rase appl ied to Raymond Queneau
(preface to Contes et propos, 1 98 1 ), desc ri bes eq ual l y his own narrative
constructions of and aro u nd h i msel f-clothes that make the man .
L'Afrique fan tome, stubborn ly naive, holds off acceptable forms of
narrative w h i l e h i nting at the i r necessity ( i n its priere d'inserer) . Ma nhood
goes beyond the journa l 's anti narrative, the mere l y chronologica l col l ec­
tion of c itations and snapshots . It constructs its story, Le i ris te l l s us, on
the model of photomontage ( 1 946 : 1 5 ) . This arranged anthology of the
se lf sti l l cu ltivates a photographic viewpoi nt-a documentary, q uasi­
sc ientific, but a l so su rrea l ton e . There i s no attempt-as in the anti­
rhetoric of romanticism-to speak without a rtifice or from the heart . Lei r­
is's "objective," "si n cere" sta nce obsessively reveal s itself as an effect of
sty le, l argely through a systematica l l y c l u msy and com p l i cated staging of
the text for which the various elaborate explanations, su pplementary
notes, h idden prefaces, and prieres d'inserer are props.
What is most i nexpl icable about L 'Afrique fantome, however, is not
its awkward ness, its dada ideas of data, its refusals, even its boredom (a
form of disponibilite) . Nor is it the persistent d i sa ppoi ntment that the
journal enacts . (If someth i ng l u m i nous occu rs, it tends q u ickly to appear
as a shabby spectac le, a commerc i a l transaction, a fu rther occasion for
amb ivalence, depression, and so o n . ) After Con rad we are accustomed
to the tristes tropiques with the i r fables of d i senchantment. What rema i n s
most i nexpl i cable is t h e strange c h i ld l i ke i n nocence emergi ng somehow,
each time, after experience. It is i n c red i b l e that Lei ris keeps on writi ng,
and that we keep on read i ng, d i pping in and out of these pages. Yet every
day the jou rna l 's scru pulous entries appear- long, short, elaborate,
terse-each prom i s i ng that something wi l l somehow happen and that
soon we w i l l see what the re lent less ser ies is lead ing to. We never do.
No moment of truth : Afrique fant6me is only a pen starti ng u p each day.
We recal l afterward the i ntensities, the confrontations, the i n c idents
of self-doubt, the d i atri bes agai nst colon ial ism and eth nography, as if
they marked a thread , the progress of the ta le. We forget a l l the ti ny
beg i n n i ngs, entries : "Co u p de theatre s u r coup de theatre" . . . "Slept

bad l y" . . . " I ntense work, to w hich I give myself with a certa i n assidu-
ousness" . . . "We're bored , all of us" . . . "The masks' mothers used to
be offered h u m a n sacrifices; th i s is Tabyon's story" . . . " Departu re from
Bordeaux at 5 : 5 0 P . M . " . . . "Another n ight at Mal kam Aya hou's" . . .
"We're approach i n g Malaka l . G reen grasses. Yel low grasses."
Le i ri s' l ife of writi ng combi nes an acute sense of the fut i l ity of ex is­
tence with a tenac ious des i re to sa l vage its mea n i ngfu l deta i l s-q uota­
tion, perception, memory. He retu rns to h i s field notes. H i s 1 98 1 work,
Le ruban au cou d'Oiympia, adopts once aga i n a fragmentary form­
co l l ected textu a l evidences of a n existence. I ts priere d'inserer records a
double goa l : "for a moment, to give the protagon i st of this sort of, some­
ti mes open someti mes d i sgu i sed , p u b l i c confession the i ntox icated feel ­
i n g of l iv i ng a second l ife; t o make t h e rece i ver perce ive what, speaking
of a n actor and h i s p l ay [son jeu) he'd ca l l 'presence .' "
Priere d'inserer- l oose somewhere between the written book, the
des i red reader. Starti ng u p . The next Leiris . . .

L 'Afrique fantome beg i n s a writi ng process th at wi l l end lessly pose and

recom pose a n identity. Its poetics is one of i ncom p letion and process,
with space for the extraneous. I nterru pti ng the smooth eth nographic story
of an access to Africa, it u nderm i nes the assu m ption that se lf and other
can be gathered i n a stable narrative coherence. Le i ris' strange, open­
ended " book" m ay be s ituated with i n a new, heterogeneous h i storical
s ituati o n . Le iris wou ld become a friend of A i me Cesai re i n the cruc ia l
decade after the Second World War, w he n surreal ism, as cu ltural­
po l itical criti c i s m , retu rned to its native l and, Pa ris, but now s peaking
the accents of negritude. (Leiris was perhaps the fi rst professional eth nog­
rapher to name and analyze co lon i a l is m , in 1 950, as an inescapable
ideo logical grou nd . ) It is becom i n g common to d i stinguish two negri­
tudes. Senghor's looks back to trad ition and eloquently gathers u p a co l ­
lective "African" essence . Cesai re's is more syncreti c, modern ist, a n d
parod ic-Caribbean i n its acceptance o f fragments and i n i t s apprec ia­
tion of the mec h a n i s m of col l age in cu ltura l l ife.
We are a l l Cari bbean s now i n o u r u rban arc h i pel agos. " G u i nea" (old
Africa, writes Cesa i re) "from you r c ry from your hand from your patience/
we sti l l have some arbitrary l a nds" ( 1 983 :207). Perhaps there's no return
for anyone to a native l and-o n l y field notes for i ts rei n ventio n . The
G u ya nese novel ist and critic Wi l son Harris recommends a "pri n c i p l e of
1 74 D I S PL A C E M E NTS

j u xtaposition" as a way to account for "the making of trad ition . . . the

heterogeneous grou ndwork of authentic com m u n i ty." He i s i nterested in
somet h i n g he ca l ls "the j igsaw of natu re, and the d i a l ogue of rea l ity"
( 1 973 : 7, 9, 8 1 ) . We can recogn i ze in th i s vision the jagged setting for
modern eth nography and eth nopoetics. Begi n n i ng with Cesa i re's unset­
tl i ng i rony ( 1 983 : 5 1 ) :

And you know the rest

That 2 and 2 are 5

that the forest meows
that the tree p l u c ks the maroons from the fire
that the sky strokes its beard
etc etc . . .

Who and what a re we ?

A most worthy question !
1\vo cultures seem to intermingle in a fascinating, ambiguous
embrace only so that each can inflict on the other a more
visible denial.

7 . A Politics of Neologism :
Aime C esaire

"V E ERITION" ? The l ast word o f A i me Cesai re's " Notebook o f a Re-
turn to the N ative Land" bri ngs the whole i n c red i ble poem to an i m pos­
s i b l e term-or tu rn . The " N otebook" is a tropological landscape i n
w h i c h syntactic, semantic, a n d ideological tran sformations occ u r. Ce­
sa i re's poems m a ke demands. To engage t h i s writi ng (the best Engl i s h
tra n s l ation t o date is by Clayton Esh leman and A n nette S m i th) is an active
work of reth i n k i n g . 1 How does one grasp, translate a l anguage that i s
b l atantly m a k i n g itse l f u p ? Esh l eman and Sm ith have gone to great
lengths of accu racy and daring; but Cesai re sti l l sends readers to d i ction­
aries in several tongues, to encyc loped ias, to bota n i ca l reference works,
h i stories, and atlases. H e is attached to the obsc u re, acc u rate term and
to the new word . He makes readers confront the l i m its of thei r l anguage,
or of a n y s i ng l e language. He forces them to construct read i ngs from a
debris of h i storical and futu re poss i b i l ities. H i s world is Cari bbean-hy­
brid a n d heterog lot.

1 . All poetry by Cesa i re i s quoted from the 1 983 Eshleman and Smith trans­
lation (Cesai re 1 9 7 3 ) .

1 75

Cesai re's poems veer. This req u i res a special page ; and the page
itself is q uestioned by h i s verse . N owhere are the size and format of "the
book" so standard ized as in France. On the na rrow pages of earl ier ed i­
tions Cesa i re's exorbitant l i nes were stu bbed . Lengthy conti nuations
separated i nto d iscrete u n its. Where these had hardened i nto printing
errors, Esh leman and Sm ith, with Cesai re's help, have corrected the pros­
ody. Their ed ition provides an u n usual ly large page, giving the poetry the
space it needs to swerve extravagantly between vertical and horizontal
momentu ms. For example the famous end ing of the " Notebook" :

then , stra ng l i ng me with you r l asso of stars

I fo l low you who a re i m pri nted on my ancestra l white cornea
rise sky I ic ker
and the great b l ac k hole where a moon ago I wanted to d rown it is
there I wi l l now fish the malevo lent tongue of the n ight i n its
motio n l ess veerition !

No page can rea l ly accommodate the fi nal horizonta l rush of words from
"and the great" to "veerition ." Esh l eman and Smith print it as a cont i n u­
ous u n it, ru n n i ng out of page only once (before "ma levolent" ) . By con­
trast the French of the "defi n i tive" Pres en c e africaine ed ition brea ks th is
long seq uence into two syntactical ly and spati a l l y d i stinct l i nes. E m i l e
Snyder's we l l -known translation opts for th ree separate l i nes, wh i l e john
Berger's Pengu i n version carves out fou r, moving even farther i n the
wrong d i rection of i magistic compression . After the p l u mmeti ng vertical
seq uence of "rise"s, Eshleman and Smith stay with Cesai re's fi nal ecstat ic
run-on sentence. On a page accommodating one h u ndred horizonta l
characters (Presence africaine, Snyder forty-five characters, Berger fifty­
five) the " l i ne" zooms across and off-a long expu lsion.
The poem "stops" on a coi n age, itse lf a new tu rn . Cesa i re's great
lyric about fi n d i n g a voice, about retu rning to native ground, strands us,
fina l l y, with a made-up, Lat i n ate, abstract-sound i ng q uestion mark of a
word . So much for expectations of d i rect, i m med iate l i ngu i stic "authen­
ticity." With Cesa i re we are i nvo lved in a poetics of cu ltu ral invention .
Eshleman and Sm ith have done we l l -as we l l as possib le-with the
poem's various neologisms (rhizulate, effarade, desencastration . ) ; but . .

as they write i n the i r i n trod u ction : "Only Cesa i re h i mse l f was i n a posi­
tion to reveal (in a private com m u n i cation) that 'verrition ' which preced­
i n g transl ators and sc holars had i nterpreted as 'fl i c k' and 'sw i r l ' had been
co i ned on a Lati n verb, 'verri ,' mean i n g 'to sweep,' 'to sc rape a surface,'
and u l ti mate l y 'to scan .' O u r rend ition ('veerition') attem pts to preserve
the tu rn i n g motion (set aga i nst i ts oxymoro n i c mod ifier) as we l l as the
Lati n sou n d of the ori g i n a l -thus restitut i n g the long-lost mea n i ng of an
i mportant passage" (p. 26). The transl ators may be forgive n their c l a i m
to h ave restored a " l ong-lost mea n i ng." I n fact rad ical i ndeterm i nacy i s
t h e essence o f neolog i s m . N o d i ctionary or etymology c a n n a i l down the
s i g n i ficance, nor can an i nventor's (remem bered) i ntention . The rea l
strength of Cesa i re's l ast word i s that it forces open aga i n the semantic
u n i verse of the " Notebook" -j u st as it i s about to c l ose. Cesa i re does not
restore the "mea n i ngs" of language, c u lture, and identity ; he gives them
a turn .

Cesa i re's most fa mous neo l og i sm, negritude, has by now l ost its new­
ness. It is too fam i l ia r as a l iterary movement and as a set of " pos itions"
in a n ongo i n g debate about black identity, essenti a l i s m , and oppos ition a l
consciou sness . Negritude, i n many o f i ts senses, h a s become w h a t Ce­
sa i re never wanted it to be, an abstraction and an ideology. When the
word fi rst appeared in the " Notebook," i t was sheer pol itica l , poetical
i nventio n . Any neo l og i s m , perceived as such, annou nces itsel f as made.
N egritude i s less an e n d u r i n g fact or cond ition to be d i scovered and
named than i t i s a h istorica l creation, a language process . In an interview
with Ren e Depestre ( 1 980) Cesai re dec l i nes to defi ne h i s coinage i n any
way except h istorica l l y and conti ngently : "There's been a lot of theoriz­
ing about negritude. I 've kept out of it, from personal d iscretion . B ut if
you ask me how I concei ve of negritude I ' l l say that in my opi n ion negri­
tude i s primari l y a concrete, not abstract, com i ng to consc iousness ." He
goes o n to reca l l a generation's response to the dom i nant "atmosphere of
ass i m i lation" in the th i rties and forties. Spea k i n g with L i l ya n Kesteloot,
Ces a i re is even more ca refu l i n h i s hand l i ng of the term :

It's an obvious fact: negritude h as brought dangers . It has tended to

become a schoo l , to become a churc h , to become a theory, an ideol­
ogy. I'm in favor of negritude seen as a l iterary phenomenon, and as a
personal eth ic, but I ' m aga i n st b u i l d i ng an ideology on negritude . . .

I f negritude means a k i nd of prophecy, wel l then no, because I strongly

bel ieve there's a c l ass struggle, for example, and there are other ele­
ments, ph i losoph ical elements, that certa i n l y determ i ne us. I abso­
lutely refuse any sort of confused, idyl l i c pan-Africanism . . . As a re­
sult, although I don't reject negritude, I look on it with an extremely
critical eye. Critica l , that's b asica l ly what I mean : l ucid ity and d i scern­
ment, not confused l y mixing everything. In add i tion, my conception
of negritude is not biologica l , it's cultural and h istorica l . I th i n k there's
always a certai n danger i n basing someth i ng on the black blood in our
vei ns, the three drops of black blood . (Depestre 1 980: 1 44-1 45)

Two of those who partici pated i n the early creation of negritude as a

movement (and who wou ld both sharply criti c i ze it from a standpo i nt of
Marxist h u m a n i sm) have now pu b l i shed i m portant books that look back
reflectively on the phenomenon . The title of Rene Depestre's col lection
of essays, Bonjour et adieu a Ia negritude, i s i n d icative of a c l ear d i stance
take n . S i m i larly Rene Men i l 's Tracees ( 1 9 8 1 ) reflects a des i re to place
negritude in a h i storical context and to see it as part of a general N ew
World pred icament.
Men i l , who was Cesa i re's main contact with su rrea l ism in the th i rties
and who partici pated i n both of the formative gro u p endeavors, Legitime
defense and Tropiques, has repu b l i shed a selection of essays from h i s
l o n g career, i n c l ud i ng a subtle reread i ng o f Cesai re's " N otebook ." Here
the negritude of Leopold Senghor and that of Cesa i re are clearly d isti n­
guished . The former elaborates a "backward- looking idea l i sm ," a fa l se l y
natura l ized , consi stent African menta l ity that tends to rei nscri be the cat­
egories of a romantic, someti mes rac i a l i st Eu ropean eth nography. Ce­
sai re's Cari bbean negritude, by contrast, rejects a l l essenti a l i st evoca­
tions. I nstead, in the Notebook, accord i n g to Men i l , "anti ph rasis and
e l l i ps i s are constantly brought to the service of poetic condensatio n . The
poet consistently holds h i mself at a d i stance from what he is and from
what he says, i n order to prod uce the literary effect of derision" (p. 80,
emphas i s in the origi n a l ) . I rony is i n herent i n the West I nd ian pred i ca­

It can be said that our West Ind ian consciousness is necessari ly par­
od ic, si nce it's caught i n a game of doubling and redoub l i ng, m i rrori ng
and separation, i n the face of a French colon ial consciousness embod­
ied in ru l i ng institutions and the mass media. For this kind of d ivided ,
worried consciousness, naivete in art is forbidden . This is the sou rce

of those d i ssonances i n our a rt that, as Baudelaire said i n the n i ne­

teenth centu ry, are agreeable to modern ears. (pp. 2 23-2 24)

Depestre too i n h i s "ad i e u " to negritude merges it with a broader

modern ism, with "the essenti a l creolite of the Cari bbea n and of Lat i n
America" ( p . 1 5 1 ) . To b e "America n " is t o b e hybrid, metis; a n d i n De­
pestre's vision the true h e i rs of negritude are writers l i ke Carpentier, G u i l­
len, Amado, Val lejo, Cortazar, Marquez. Aga i n negritude is transmuted ;
it is no longer about roots but about present process i n a polyphonous
rea l ity.
With t h i s i n m i nd it is good to ret u rn to the word's origi nal coi nage
in the " N otebook" (p. 67), where we see not the elaboration of a broad
black identity but rather very specific affi rmations and negations. "My
negritude i s not . . . "

my negritude is not a stone, its deafness h u rled aga i nst the c l amor
of the day
my negritude is not a l e u koma of dead l iq u id over the earth's dead
my negritude is neither tower nor cathed ra l
it takes root in the red flesh of the soi l
i t takes root i n t h e ardent flesh o f the sky
it brea ks through the opaq ue prostration with its u pright patience

Cesa i re writes " i n " Fre n c h , but . . . "Marronnerons-nous Depestre mar­

ron nerons-nous?" The l i ne appears i n a poem of 1 95 5 entitled "The Verb
'Marro nner'/for Rene Depestre, H a itian Poet." Here neo logism defeats
the best of transl ators, for the o n l y possi b l e eq u ivalent of the coi ned mar­
ronner is "to maroon," w h i c h , though derived from the same root, is
dom i n ated by i mages of s h i pwreck and abandonment. The noun marro n ­
age has been adopted by anglophone scholars of maroon societies i n the
G u i anas, B raz i l , and the Cari bbea n ; but the verb, Cesai re's i nvention, is
sti l l without trans l ation . The sou rce is o l d Span ish : cima, or "mounta i n­
top" (th u s a p l ace of escape), lead i ng to the l ater cimarron, "wi ld," "run­
away" (th u s the maroon, or fugitive slave) . Cesa i re's marronner i nvokes
escape and somet h i n g more.
Recent stu d i es of maroon societies h ave made apparent the very
com plex m i x of i ngred ients that combi ned in origi n a l ways to form res i l ­
ient, flex i b l e c u l tu res. T h e ties with Africa were rea l , b u t notions o f a

"col lective memory" o r o f cu ltural "survivals" cou l d not fu l l y account for

the spec ific Afro-American forms constructed from diverse tribal trad i­
tions, from the new "slave c u ltures," from various creolizing processes,
and from local , properly h i storica l experiences . (See Price 1 9 73, 1 98 3 . )
Cesai re's new verb marronner i tself changes sign i ficantly i n succes­
sive revisions of the poem for Depestre. In i ts fi rst vers ion, dated 1 95 5 ,
the work was cal led " Reponse a Depestre, poete haitien : Elements d ' u n
art poetique." Depestre, then ex i l ed i n Braz i l , h a d recently ra l l ied, in the
pages of Presence africaine, to the F rench Com m u n i st party's emerging
conservative l i ne on poeti c experimentation . I n the wake of su rrea l ism,
Lou i s Aragon was press ing for a return to more trad itional prosody, to
s i m p ler forms and messages, l i nking these with the interests of revo l u­
tionary workers . Free ve rse and rad ical i n novation of many ki nds were
now proscribed as "form a l i nd ividua l is m ." Cesai re's reply to Depestre,
a l so pu b l ished in Pres e n ce africa ine, rejected the conservative trend and
prepa red his own break w ith French com m u n ism a year l ater in his Letter
to Maurice Thorez. " I s it true th i s season that they' re pol ish i ng up son­
nets ?" he asks Depestre, and i mmed i ately ties the new constra i nts to H a i ­
t i 's colon i a l sugar m i l l s ( 1 983 : 369) :

when s low s k i n ny oxen make the i r rou nds to the whine

of mosq u i tos

Bah ! Depestre the poem is not a m i l l for

gri n d i ng sugar cane absol utely not
and if the rhymes a re fl ies on ponds
w ithout rhymes
for a whole season
away from ponds
u nder my persuasion
let's l augh drink and escape l i ke slaves
[rions buvons et marronnons]

The fi rst and second pri nti ngs of the " Repl y to Depestre" conta i n a spe­
cific reference to Aragon : "To hel l with it Depestre to he l l with it let
Aragon ta l k ." The reference wou l d be d ropped from Cesai re's Oeuvres
completes ( 1 9 76), the sou rce for Esh l eman and Sm ith . This movement
away from a speci fic controversy is reflected i n a more sign ificant textual
change . I n 1 95 5 Cesa i re had exhorted Depestre : "Marron nons-les De­
pestre ma rron nons- les I comme j ad i s nous marron n ions nos maitres a

fou et" ( let's escape them Depestre let's escape them I a s i n the past we
escaped our wh i p-wie l d i n g m asters) . I n the l ater version the exhortation,
once cast in a transitive form, " let's escape them" (the sl ave drivers, the
party l i ne), has been altered to an i nterrogative future form i n which
the play of sou nd becomes a s l i ghtly glosso l a l i c ri pple of n 's and r's.
"Marronnerons-nous Depestre marron nerons-nous ?" The reference to
"wh i p-wield i n g masters" is gone, and marronage is now a less l i m ited
and ongo i n g act of escape. I n the poem it is enacted as the m i x i n g up of
sound and sense, the ru n n i ng away with l anguage. Thus marronage is no
longer a bout s i m p l y esca p i n g (them) . It is a l so about reflexive poss i bi l ity
and poes i s . Cesa i re makes rebe l l ion and the remaki ng of cu lture-the
h i storical maroon experience- i nto a verb. A necessary new verb names
the New World poetics of conti n uous tra nsgression and cooperative c u l ­
tura l activity ( "Marronnerons-nous Depestre") . Fugitive s laves w h o c re­
ated c u ltures in the swam ps of the G u ianas represented d isti n ct African
trad itions. Livi ng together they took over, used , and a l tered one another's
c u stoms, words, and pasts. So Cesai re, born i n Mart i n ique, i n vokes i n ­
cidents of H a itian h i story i n h i s letter t o Depestre, w h i l e pressi n g a poetic
rad ica l is m derived from R i m baud and su rrea l i sm . The fi n a l l i nes of "The
Verb 'Marro n ner' " a re scattered with words and p lace names from West
Afr i ca , France, H i spa n i c America, B raz i l , H a iti . Cesa i re veers among the
trad itions that h i story has offered to and i mposed on a Cari bbean identity.
H is bei ng and h i s poetics are elements of "A F reedom i n Passage" (the
l ast poem of the co l l ection) :

h e l ped so m u c h by b i rd s
whose m i ssion i s b y means o f po l le n

We sti l l need a verb marronner.

vegetate [from L. vegetare, to enliven, quicken . . .]

8 . The Jardin des Plantes : Postcards

Paris, 912184
Dear A,
Around the Jard i n des Pl ante s : bou l es p l ayed in the old Roman space
. . . remember? Arenes de Lutece, h idden beh i nd bu i ld i ngs of the rue
Monge; Mouftard and its ma rket (gentrified); or the mosq ue, where you
can sti l l take a steam bath and drink m i nt tea from gold trays . Th i s year
the gardens are l ush-bl u r of blossoms from everywhere, goi n g to seed .
Peopl e scattered on green c h a i rs observe the plants. And statues : Bernar­
d i n de Sai nt-Pierre s m i les down i n b l u i sh bronze at the myt h i c kids Pau l
et Vi rgi n i e . B uffon , back tu rned to everyone : a pigeon twitches o n h i s
meta l head . Over near the zoo and the Sei ne Lamarck i n a n attitude of
thought-above a risi ng s u n (sc ience? nature?). See you at the end of the
month . . .

Dear P,
NAT U R E/C U LT U RE NAT U R E/C U LT U R E NAT U R E/C U LT U R E : it's sti l l the most
beautifu l , i ntense, fu n ny, etc . etc . exh ibit in Pa r i s : " Les Plus Beaux I n -

T H E J A R D I N D E S P L A NT E S : P O S T C A R D S 183

sectes d u Monde" ( I n stitute o f Entomology, r u e B uffon , afternoons) . One

bright room . . . p i n ned fi reworks, faces, masks, eyes, bones, skin . . .
then sudde n l y fou r exactl y torn leaves (with antennae), a band of tri­
colores, the d u sty s u btlety of moth s ( B raque, K l ee), touched up, sprayed,
day-glo, lacquer, cera m i c , m i c roc h i p . . . laugh of c l assifications. Bou­
ga invi l le, Borneo, S u m atra, J ava; with wi ngs, without wings; "Hercu l es,"
go l d meta l , b l ackblue, neongreen . . . Didn't Levi-Strauss write some­
where that modern art shou ld be i n s p i red by butterfl ies, not Picasso?

Dear S,
Tal l tree, SO P H O RA JAPO N I CA, planted by B . J ussieu, 1 74 7 , i n the J a rd i n d u
Roy, seed sent from C h i n a b y R . P. d ' l ncarv i l le, . . . o r another, brought
from "The Levant." Strange th i ngs, a live and historica l (not at a l l l i ke
those seq uoia ri ngs with dates, 1 49 2 , 1 776, 1 9 1 4 ; or "the bed Napoleon
slept i n" ) . They' re l iv i n g in planetary- h u m a n time and space . . . the Age
of D i scovery transplanted . Outl iving u s . By the way, there's a new book­
store with a p retty good poetry section, Mouftard and Pot de Fer:
"L'Arbre Voyageu r." Wi l l you be pass i ng through Ca l iforn i a this wi nter?

Dear T,
I n h u man Robinsonade-
RO B I N I E R de RO B I N ( Rob i n a Pseudo Accaci a L i n ne)
The fi rst s u bject i ntroduced i nto E u rope from seeds origi nati ng i n
N O RT H AM E R I CA b y Jean ROB I N i n h i s garden o f the pl ace Dauph ine
i n 1 60 1 . Transplanted to th i s spot i n the Roya l Garde n i n 1 63 6 by
Ves pas ier ROB I N , son of the above.

Dear B ,
You won't h ave forgotten t h e fantastic, l o ng, dappled a l l eyways of Ti l ­
leu ! . B u t maybe y o u d idn't s e e a l ittl e rock garden where they c u ltivate
shrubs, flowers, cacti, and herbs from C h i na, the Caucasus, Corsica,
N ew Zea land, Morocco, the H i m a l ayas, Pyrenees, B a l kans, Arctic, J a­
pan . . . Cont i nents beside eac h other i n h u n d reds of beds. On a tru n k
supported b y a n i ron post :
PI STAC H E R ( P i stac i a Vera l . )
P l anted i n the seed l i n g garden aro u nd 1 700

(the present A l p i ne Garden) .

Permitted Sebastian Va i l lant to d iscover the
sexual ity of plants in 1 7 1 6.
So what's become o f M . Vai l l ant? O r the sexua l ity of pl ants ? The pista­
chio tree l ives . With love . . .

Dea r N ,
"Pri m itive" pa inting b y H a itians i s a recent avocation . ( B u t you know a l l
about th i s . ) And they took to it s o " natu ra l l y." A friend tel l s m e h e once
saw a H a itian artist pa i n t i ng the complex forests of " G u i nee" (place of
origin) with H e n ri Rousseau reprod uctions at hand. There are no African
j u ngles in H a iti . And the Douanier hadn't seen them either but copied
his in Paris from tropical spec i mens at the Jard i n . Right now I'm looking
i n at the entrance to one of the old d ream l i ke green houses. A tyger?
Beh i nd tal l panes . . . fabu lous, sha rp, sagging leaves . Fable of our
"Cari bbean" selves ?

Dear L,
About the view from you r hote l , rue L i n ne . . . I can i magine the ivy­
d renched gate of the Jard i n , and with i n , dark wa l ls of the " Labyri nth ." (Its
rising c i rcu l a r paths, assignations, strangers . . . ) And isn't th i s near the
great Cedar of Lebanon where around 1 860 ( i n an old pri nt) people with
top hats and long d resses stro l l ed to ad m i re the su perb, spread i ng form
and to marve l at the gathered i m perial u n i verse ? They m ust have heard­
as one sti l l does-noises from the zoo, a n i m a l s that wou ld be devoured
some years l ater by bes ieged citizens of the Comm u ne. Pl ease stay i n
tou c h .

Dear C,
Thanks for sen d i ng me the new book by A l icia D ujovne O rtiz, Buenos
Aires . One of those m i racles of travel , the vertige horizontal. Jews from
Moldavia marrying Argenti n ians ( i m m igrants : Span ish, Ital ian, Albanian
. . . ), then a daughter: portena, i n Paris, writing French, remembering. I
l i ke her ambiva lent g l i m pses of Borges and the tango. A l so, espec i a l ly,
her love for the city's Jard i n Bota n i co-giant pl ants visited by cats and
old lad i es with bags of l i ver. The zoo's " h i nd u pl ace i n habited by a dusty

elephant . . . " Transpl anted c i v i l ization . " B u t if I have no roots, why

have m y roots h u rt me so?"

9/ 1 0184
Dear J,
Paris of the ren tree : streets fi l l ing aga i n , tem po picking u p . The low s u n
h a s a n a rtific i a l g l aze. I n t h e J a rd i n des Pla ntes, they' re bu i l d i ng a new
"Zootheq ue," u ndergro und. Pru n i ng has begu n . Contemplating winter.
I ' m i nfatuated yet aga i n with the pa l m s of the Luxembo u rg Gardens ("Au­
tou r d ' u n e meme p l ace I I ' ample pal me ne se l asse . . . "), sym metrical,
perfect, i n boxes w i th i ro n feet. Vegetable extraterrestrials . . . six i nches
of a i r between the path and thei r . . . earth .
Indian woman spinning yam and rocking cradle with a
cord tied to her foot. In the background Franz Boas and
George Hunt help compose the picture.
Part Three � Collections
You do not stand in one place to watch a masquerade.

9 . Histories of the Tribal

and the Modem

DURING THE WINTER of 1 984 - 8 5 one cou ld encou nter tribal objects
in an u n u s u a l n u m ber of locations around New York C ity. T h i s chapter
su rveys a half-dozen, foc u s i n g on the most controversi a l : the major ex­
h i bition held at the Museum of Modern Art (MOMA), " ' Pri m itivism' i n
2 0th Centu ry A rt : Affi n ity of t h e Tri ba l a n d the Modern ." The chapter's
"eth nogra p h i c present" is l ate December 1 984 .

The "tri ba l " objects gathered on West F i fty-th i rd Street have been a rou n d .
They are travelers-some a rriv i n g from fo l klore a n d ethnograph i c m u ­
s e u m s i n E u rope, others from a rt ga l l eries a n d private co l l ections. They
have trave l ed fi rst c l ass to the Museum of Modern Art, e l aboratel y c rated
and i ns u red for i m portant s u m s . Previous accommodations have been
l ess l uxurious: some were stolen, others " p u rc hased" for a song by co­
lon i a l ad m i n i strators, trave lers, anthropologists, m i ssionaries, sa i l ors i n
African ports . These non -Western objects have been by tu rns c u riosities,
eth nograph i c spec i mens, major art creations. After 1 900 they began to
turn u p i n E u ropean flea m arkets, thereafter mov i ng between avant-garde

1 89

stud ios and co l l ectors' apartments . Some came to rest i n the u n h eated
basements or " l a boratories" of a nth ropology museums, su rrounded by
objects m ade in the same region of the worl d . Others encountered odd
fel low trave lers, l ighted and l abel ed i n strange d isplay cases. Now on
West Fifty-th i rd Street they interm i ngle with works by E u ropean mas­
ters- Picasso, G iacomett i , B ranc usi, and others . A th ree-d i mensional
Eskimo mask with twel ve arms and a n u m ber of holes hangs beside a
canvas on w h i c h joan M i r6 has pai nted colored shapes. The people i n
N ew York look a t the two objects and see that they are a l i ke.
Travelers tel l d i fferent stories i n d i fferent places, and on West F i fty­
th i rd Street an origi n story of moder n i sm is featu red . Aro u nd 1 9 1 0 Pi­
casso and his cohort sudden l y, i ntu itive l y recogn ize that "pri m i tive" ob­
jects a re in fact powerfu l "art." They col l ect, i m itate, and are affected by
these objects . Thei r own work, even when not d i rectly i nfl uenced , seems
odd l y rem i n i scent of non-Western forms. The modern and the pri m itive
converse across the centu ries and conti nents. At the M useum of Modern
Art a n exact h i story i s to ld featu ring i nd ividual artists and objects, the i r
encou nters i n spec ific stud ios a t prec i se moments. Photographs docu­
ment the cruci a l i nfl uences of non-Western artifacts on the pioneer mod­
ern ists . T h i s focused story is surrou nded and infused with another-a
loose a l l egory of re l ations h i p centeri ng on the word affinity. The word is
a k i nsh i p term , suggesti n g a deeper or more natu ral rel ationsh i p than
mere resemblance or j uxtaposition . It con notes a com mon q u a l ity or es­
sence joi n i n g the tri bal to the modern . A Fam i ly of A rt is brought to­
gether, globa l , d i verse, rich l y i nventive, and m i racu lously u n i fied , for
every object d i s p l ayed on West Fifty-th i rd Street looks modern .
The exh i b ition at MOMA is h i storica l and d idactic. It is com ple­
mented by a comprehensive, scholarly cata l ogue, which i n c l udes d i ver­
gent views of its topic and i n which the show's organ izers, Wi l l iam Rubin
and K i rk Varnedoe, argue at length its u nderl ying premises (Rubin 1 984) .
One of the v i rtues of an exh i b i tion that bl atantly makes a case or tel ls a
story is that it encou rages debate and makes possible the suggestion of
other stories. Th u s i n what fol l ows d i fferent h i stories of the tribal and the
modern wi l l be proposed in response to the sharply foc used h i story on
d isplay at the Museum of Modern Art. But before that h i story can be seen
for what it is, however-a spec ific story that exc l udes other stories-the
u n iversal i z i n g al legory of affi n i ty must be cleared away.
Th is a l legory, the story of the Modern i st Fam i l y of Art, is not rigo r­
ously argued at MOMA . (That wou ld req u i re some expl icit form of either
H I S TO R I E S O F TH E T R I B A L A N D TH E M O D E R N 191

an archetypal or structu ral analys i s . ) The a l legory i s , rather, bu i lt i nto the

exh i bition's form , featu red suggestive l y i n its p u b l i c i ty, left u ncontra­
d i cted, repetitio us ly asserted - "Affi n i ty of the Tri ba l and the Modern ."
The a l l egory has a hero , whose v i rtuoso work, an exh i bi t caption tel l s
u s , conta i n s more affi n ities with the tri bal than that o f any other pioneer
modern ist. These affi n ities "meas u re the depth of Picasso's grasp of the
i nform i ng pri n c i p l es of tribal scu l pt u re, and reflect his profo u nd identity
of s p i rit with the tribal peoples." Modern ism is thus presented as a search
for " i n form i n g pri n c i p l es" that transcend cu ltu re, pol itics, and h i story.
Beneath t h i s generous u m b re l l a the tribal is modern and the modern
more r i c h l y, m ore d iversel y h u m a n .

T h e power o f t h e affi n ity idea i s s u c h ( i t becomes a l most self-evident i n

the MOMA j u xtapos itions) that i t i s worth reviewing the major objections
to it. Anthropologists, long fam i l iar with the issue of c u ltura l d i ffusion
versus i ndependent i nvention, a re not l i kely to find a nyth i n g spec i a l i n
the s i m i l arities between selected triba l and modern obj ects. A n estab­
l i shed pri n c i p l e of anthropologica l com parative method asserts that the
greater the range of c u l t u res, the more l i kely one is to find s i m i lar tra its .
MOMA's samp l e is very l arge, embrac i ng African, Ocea n i a n , North
America n , and Arctic "tri ba l " gro u ps . 1 A second pri n c i p l e, that of the
" l i m itatio n of poss i b i l ities," recogn izes that i nvention, w h i l e h i gh l y d i ­
verse, is not i nfi n ite . The h uman body, for exam ple, w i t h its two eyes,
fou r l i m bs, b i l atera l a rrangement of features, front and back, and so on,
wi l l be represented and styl i zed i n a l i m ited n u m ber of ways . 2 There i s
th u s a priori no reason t o c l a i m evidence for affi n i ty (rather than mere

1 . The term tribal i s used here with considerable rel uctance. It denotes a
kind of soc iety (and a rt) that cannot be coherently spec ified. A catchal l , the con­
cept of tribe has its source in Western projection and adm i n i strative necessi ty
rather than i n any essential qual ity or group of traits. The term is now common l y
used i nstead of primitive i n ph rases s u c h as tribal art. T h e category t h u s denoted ,
as th i s essay a rgues, is a product of h i stori ca l l y l i m i ted Western taxonom ies.
Wh i l e the term was orig i n a l l y a n i m position, however, certai n non-Western
groups have embraced it. Tri ba l status is in many cases a crucial strategic ground
for identity. I n thi� essay my use of tribe and tribal reflects common usage wh i l e
suggesting ways i n w h i c h t h e concept is systematica l l y d i storting. See Fried 1 9 75
and Stu rtevant 1 98 3 .
2 . These poi nts were made b y Wi l l iam Stu rtevant a t t h e sympos i u m o f an­
thropologi sts and art h i storians held at the Museum of Modern Art i n New York
on November 3 , 1 984.

resembl ance o r coi nc idence) because an exh i b ition of tribal works that
seem i m pressi vely " modern" in sty l e can be gathered . An eq u a l l y stri king
col lection cou ld be made demonstrating sharp d i ssi m i l arities between
tribal and modern objects.
The q u a l ities most often said to l i n k these objects are the i r "concep­
tua l ism" and "abstraction" (but a very long and u lti mate l y incoherent l ist
of s hared tra i ts, i n c l u d i n g "magic," "ritua l ism," "envi ron menta l ism," use
of "natura l " materials, and so on, can be derived from the show and
espec i a l l y from its cata l ogue) . Actual ly the tri bal and modern a rtifacts
are si m i lar o n l y in that they do not featu re the pictorial i l l usion ism or
scu l ptu ra l natura l ism that came to dom i nate Western European art after
the Renaissance. Abstraction and conceptual i sm are, of cou rse, perva­
sive in the arts of the non-Western World . To say that they share with
modern i s m a rejection of certai n natura l ist projects is not to show a ny­
th i n g l i ke an affin ity. 3 I ndeed the "triba l i sm " selected in the exh i bition to
resemble modern ism i s itself a construction designed to accom pl ish the
task of resembl ance. lfe and Ben i n scu l ptures, h ighly natura l i stic in style,
are exc l uded from the "tri ba l " and pl aced i n a somewhat arbitrary cate­
gory of "court" soc i ety (wh i c h does not, however, i n c l ude large c h i eftan­
sh i ps) . Moreover, pre-Co l umbian works, though they have a pl ace in the
cata logue, are large l y omitted from the exh i bition . One can q uestion
other selections and exc l usions that result i n a col lection of only " mod­
ern" - looki ng tribal objects . Why, for example, are there relative ly few
" i m p u re" obj ects constructed from the debri s of colo n i a l cu lture con­
tacts? And i s there not an overa l l bias toward c lean, abstract forms as
aga i n st rough or crude work?
The "Affin ities" room of the exhi bition i s an i ntrigu ing but enti rely
problematic exerc i se i n formal m i x-and-matc h . The short i ntroductory

3. A more rigorous formu l ation than that of affi nity is suggested in Le iris
1 95 3 . H ow, Leiris asks, can we speak of African scu l ptu re as a single category ?
He warns of "a danger that we may u nderestimate the variety of African scu l p­
ture; as we are less able to appreciate the respects in which cultures or th i ngs
unfa m i l iar to u s differ from one another than the respects in which they differ
from those to which we are used , we tend to see a certa i n resemblance between
them, which l ies, in point of fact, merely in the i r common differentness" (p. 3 5 ) .
Thus, t o speak o f African scu l pture o n e i nevitably shuts one's eyes "to t h e rich
diversity actua l l y to be found i n th is scu l ptu re in order to concentrate on the
respects i n which it is not what our own scu lptu re genera l l y i s ." The affin ity of
the tribal and the modern is, in this logic, an i m portant optical i l lusion-the
measure of a common differentness from artistic modes that dom inated in the
West from the Renaissance to the late n i neteenth centu ry.

text beg i n s wel l : " A F F I N ITI ES presents a group of tribal obj ects notable for
the i r a ppea l to modern taste." I ndeed th i s i s a l l that can rigorously be said
of the objects in this room. The text conti n u es, however, "Se lected pai r­
i ngs of modern and triba l obj ects demonstrate common denominators of
these a rts that a re i ndependent of d i rect i nfluence." The ph rase common
denominators i m p l i es someth i n g more systematic than i ntrigu i ng resem­
blance. What can it poss i b l y mea n ? Th i s i ntrod uctory text, cited i n its
enti rety, i s e m b l ematic of the MOMA u ndertaki ng as a whole. Statements
carefu l l y l i m iting its p u rv iew (spec ify i n g a concern o n l y with modern i st
pri m itivism and not with tribal l ife) coexi st with freq uent i m p l i cations of
someth i n g more. The affi n ity idea itself i s wide-ranging a nd pro m i sc u ­
ous, as a re a l l u s i o n s t o u n i versa l h u m a n capac ities retrieved i n the en­
cou nter between modern and tri ba l o r invocations of the expans i ve h u ­
man m i nd-the hea l thy capac ity of modern ist consciousness t o q u estion
its l i m its and engage otherness.4
N owhere, h owever, does the exh i b ition or cata logue u nderl ine a
more d isqu ieti n g q u a l ity of modern ism : its taste for appropri ati ng or re­
deem i ng otherness, for constituti ng non-Western a rts in its own i mage,
for d i scoveri n g u n iversa l , a h i storical " h uman" capacities. The search for
s i m i l a rity itse lf req u i res j ustification, for even if one accepts the l i m ited
task of exploring " modernist primitivism," why cou ld one not learn as
m u c h about Picasso's or Ernst's creative processes by analyzing the dif­
ferences separat i n g the i r a rt from tri bal models or by trac i ng the ways
the i r a rt moved away from, gave new twists to, non-Western forms ? 5 T h i s
s i d e of t h e process is u nexpl ored i n t h e exh i b ition . T h e preva i l i n g view­
point is made a l l too clear in one of the "affi n ities" featu red on the cata­
logue's cove r, a j u xtapos ition of Picasso's Girl before a Mirror ( 1 932) with
a Kwakiutl half-mask, a type q u ite rare among Northwest Coast c rea­
tions. Its task here is s i m p l y to prod uce an effect of resemblance (an effect
actu a l l y c reated by the camera angle) . I n th i s exh i bition a u n i versal mes­
sage, "Affi n ity of the Tri ba l and the Modern ," is prod uced by carefu l se­
l ection and the m a i ntenance of a speci fic angle of vision .
The notion of affi n i ty, an a l legory of kinsh i p, has an expans ive, eel-

4 . See, for exa mple, Rubi n's d i scussion of the myth ic u n i versals shared by
a Picasso pa inting and a Northwest Coast half-mask (Rubin 1 984 : 3 28-330). See
a l so K i rk Varnedoe's assoc iation of modern i st pri mitivism with rationa l , scientific
exp loration (Rubin 1 984 : 2 0 1 -203, 652-65 3 ) .
5 . This point was made by C l i fford Geertz a t t h e November 3 , 1 984, sym­
pos i u m at the Museum of Modern Art (see n . 2 ) .

(a) (b)

The Making of an Affinity

(a) Pablo Picasso, G i rl before a
Mirror, 1 932 (detail)
(b) Kwakiutl mask
(c) Picasso, Girl before a Mirror
The detail from the Picasso paint­
ing and the Kwakiutl mask were
juxtaposed on the cover of the ex­
hibition catalog "Primitivism"
in 20th Cen tury Art: Affinity
of the Tribal and the Modern,
volume I.


ebratory task to perform . T h e affi n ities shown a t MOMA are a l l on mod­

ern i st terms. The great modern ist "pioneers" (and the i r m useum) are
shown promoti n g formerly despised triba l "fetishes" or mere eth no­
gra p h i c "spec i mens" to the status of h igh art and in the process d iscov­
ering new d i mensions of the i r ("our") creative potential . The capacity of
art to transcend its c u ltura l and h i storical context is asserted repeated ly
( R u b i n 1 984 : x , 73). In the cata logue R u b i n tends to be more i nterested
in a recovery of e lementa l expressi ve modes, whereas Varnedoe stresses
the rationa l , forward-loo k i n g i nte l lect (wh ich he opposes to an u n healthy
pri m itivism, i rratio n a l and escap ist) . Both celebrate the generous spirit of
modern ism, p i tched now at a globa l sca l e but exc l ud i ng-as we shal l
see-Th i rd World modern isms.
At West F i fty-th i rd Street modern ist prim itivism is a goi ng Western
concern . It is, Varnedoe tel ls us, s u m m i n g u p in the l ast se ntence of the
cata logue's second vo l u me , "a process of revo l ution that begins and ends
in modern c u lture, and because of that- not i n sp ite of it-can conti n­
u a l l y expand and deepen ou r contact with that which i s remote and d if­
ferent from us, and conti n u a l l y th reate n , c h a l l e nge, and reform o u r sense
of self" ( R u b i n 1 984 : 682 ) . A s kepti c may doubt the abi l ity of the mod­
ern ist pri m itivism exh i b i ted at MOMA to th reaten or chal l enge what i s
b y n o w a thorough l y i nstitutiona l i zed system o f aesthetic (and market)
val u e ; but it is appropriate, and i n a sense rigorous, that this massive
co l l ection span n i ng the globe sho u l d end with the word self.
I ndeed an u n i ntended effect of the exh i b ition's comprehensive cat­
a l ogue is to show once and for a l l the i n coherence of the modern Ror­
schach of "the pri m itive ." From Robert Goldwater's formal ism to the
tra n sform i n g "magic" of Picasso (accord i ng to R u b i n ) ; from Levy-Bruh l 's
mystical m en tal ite primitive ( i nfl ue n c i ng a generation of modern a rti sts
and writers) to Levi-Strau ss's pensee sauvage (resonating w i th "systems
a rt" and the cyberneti c b i n arism of the m i n imal i sts) ; from D u buffet's fas­
c i n ation with i nsan ity and the c h i ld i s h to the en l i ghtened ration a l sense
of a Gaugu i n , the p l ayfu l experi menta l ism of a Picasso or the new "scien­
tific" s p i r i t of a J a mes Tu rre l l (the l ast th ree approved by Varnedoe but
cha l len ged by Rosa l i nd Krauss, who i s more attached to Bata i l le's decap­
itation, bassesse, and bod i l y deformations6) ; from fetish to icon and back

6. The c l ash between Krauss's and Varnedoe's dark and l ight versions of
pri m itivism is the most stri king i ncongru ity with i n the cata logue . For Krauss the
cru c i a l task is to shatter predom i nant European forms of power and subjectivity;
for Varnedoe the task i s to expand the i r purview, to q uestion, and to innovate.
1 96 C O LLE CTI O N S

aga i n ; from aborigi nal bark pa i nti ngs (Kiee) to massive pre-Co l umbian
monu ments (Henry Moore) ; from weightless Eskimo masks to Stone­
henge-the cata logue succeeds in demonstrating not any essenti a l affin­
ity between tri bal and modern or even a coherent modern ist attitude to­
ward the pri miti ve but rather the restl ess des i re and power of the modern
West to col l ect the world .

Setting aside the a l l egory of affi n ity, we are left with a "factua l ," narrowl y
foc u sed h i story-that o f t h e "d iscovery" o f pri mitive art b y Picasso a n d
h i s generation . It is tem pti ng t o s a y that t h e " H i story" section o f t h e ex­
h i bition is, after a l l , the rigorous part and the rest mere l y suggesti ve as­
soc i ation . U nden i a b l y a great deal of sc holarly research in the best
Kunstgesch ich te trad ition has been brought to bear on this specific h i s­
tory. N u merous myths a re usefu l l y q uestioned ; i m portant facts are spec­
ified (what mask was in whose studio when); and the pervasiveness of
tri ba l i nfl uences on early modern ist a rt- E u ropean, Engl ish, and Ameri­
can - i s shown more a m p l y than ever before . The cata logue has the merit
of i n c l ud i ng a n u m ber of articles that dampen the celebratory mood of
the exhi bition : notably the essay by Krauss and usefu l contri butions by
Christian Feest, P h i l i ppe Peltier, and Jean-Lo u i s Pa ud rat deta i l i ng the a r­
rival of non-Western artifacts i n Europe . These h i storical artic les i l l u mi­
n ate the less ed ify i ng i m perial ist contexts that su rrou nded the "di scov­
ery" of tri ba l obj ects by modern i st artists at the moment of h igh
colon i a l ism .
If we ignore the "Affi n ities" room at MOMA, however, and focus on
the "serious" h i storica l part of the exh i bition, new critical questions
emerge. What is exc l uded by the spec ific foc us of the h i story? Isn't th is
factual narration sti l l i nfused with the affinity a l legory, si nce it is cast as
a story of c reative gen i u s recogn izing the greatness of triba l works, dis­
coveri ng common art i stic " i nform i ng pri nciples" ? Cou ld the story of th i s
i ntercu ltural encou nter b e told differently? I t is worth m a k i n g the effort to
extract another story from the materials i n the exh i bition-a h i story not
of redemption or of d i scovery but of rec lassification . This other h i story
assumes that "art" is not u n i versa l but is a changing Western cu ltural
category. The fact that rather abru ptly, in the space of a few decades, a
l a rge c l ass of non-Western a rtifacts came to be redefi ned as a rt is a tax­
onomic sh ift that req u i res critica l h i storica l d i scussion, not ce lebration .
That this construction of a generous category of art pitched at a globa l
H I S T O R I E S O F T H E T R I B A L A N D TH E M O D E R N 1 97

sca l e occu rred j ust as the planet's tri ba l peoples came massively u nder
E u ropean po l itica l , economic, and evange l ical domi nion can not be i r­
relevant. But there i s no room for such complexities at the MOMA show.
Obviously the modern ist a ppropriation of triba l prod uctions as art i s not
s i m p l y i m perial ist. The project i nvol ves too many strong critiq ues of co­
l o n i a l ist, evo l uti o n i st assu m ptions. As we s h a l l see, though , the scope
and u nderlyi n g logic of the "d iscovery" of tri bal a rt reprod uces hege­
m o n i c Western ass u m ptions rooted in the colon ial and neoco l o n i a l
epoch .
Picasso, Leger, Apo l l i n a i re, and many others came to recogn ize the
e l ementa l , "magica l " power of African scu l ptu res i n a period of growing
negrophilie, a context that wou l d see the i rru ption onto the E u ropean
scene of other evocative black figu res : the j azzman, the boxer (AI
B rown), the sauvage joseph i ne Bake r. To tel l the h i story of modern ism's
recognition of Afri can "art" i n th is broader context wou ld ra ise ambigu­
ous and d i sturb i n g q u estions about aesthetic appropriation of non­
Western others, issues of race, gender, and power. Th i s other story is
large ly i nv i s i bl e at MOMA, give n the exh i bition's narrow focu s . It can be
g l i m psed o n l y in the sma l l section devoted to " La c reation du monde,"
the Afr i can cosmogony staged i n 1 92 3 by Leger, Cend rars, and M i l haud,
and i n the broad ly pitc h ed if sti l l largely u ncritical catal ogue a rticle by
Lau ra Rosenstock devoted to it. Overa l l one wou ld be hard p ressed to
ded uce from the exh i bition that a l l the enth usiasm for t h i ngs negre, for
the "magic" of African art, had anyth ing to do with race. Art in th i s
focused h i sto ry h a s n o essential l i n k w i t h coded perceptions o f black
bod ies-the i r vita l ism, rhythm, magic, eroti c power, etc . -as seen by
w h i tes. The modern i s m represented here is concerned o n l y with artistic
i nvention, a positive category separable from a negative pri m itivism of
the i rrationa l , the savage, the base, the fl ight from civi l i zation .
A d i ffe rent h istorical focu s m ight bring a photograph of j osep h i n e
Baker i n to the v i c i n ity o f t h e African statues that were exciting t h e Pari­
sian avant-garde i n the 1 9 1 Os and 1 920s; but such a j u xtaposition wou l d
b e unth i n kable i n t h e MOMA h istory, for it evokes different affi n i ties from
those contributing to the category of great art. The black body i n Paris of
the twenties was an ideo l ogical a rtifact. Archaic Africa (wh ich came to
Paris by way of the futu re-that is, America) was sexed , gendered , and
invested with "magic" in specific ways . Standard poses adopted by " La
Bakai re," l i ke Leger's designs and costumes, evoked a recogn izable "Af­
rican ity"-the naked form emphas i z i ng pelvis and buttocks, a seg-

mented sty l i zation suggesting a strangely mechan i ca l vital ity. The inclu­
sion of so ideologica l ly loaded a form as the body of Josephine Baker
among the figu res c l assified as a rt on West Fifty-th i rd Street wou ld sug­
gest a different account of modernist primitivism, a d i fferent analysis of
the category negre in /'art negre, and an exploration of the "taste" that
was someth i ng more than j ust a backd rop for the d iscovery of tribal art
i n the open i ng decades of th i s centu ry. 7
Such a focus wou ld treat art as a category defined and redefi ned i n
specific h i storical contexts and re lations of power. Seen from this angle
and read somewhat agai nst the gra i n , the MOMA ex h i bition docu ments
a taxonom ic moment: the status of non-Western objects and "h igh" art
are i m porta ntly redefined , but there is noth i ng permanent or transcen­
dent about the categories at stake. The appreciation and i nterpretation of
tri ba l objects takes place with i n a modern "system of objects" which
confers va l u e o n certa i n th i ngs and withholds it from others (Baudri l lard
1 968). Modern ist pri mitivism, with its claims to deeper humanist sym­
pathies and a w ider aesthetic sense, goes hand-i n-hand with a developed
ma rket i n tribal art and with defi n itions of arti stic and cu ltural authentic­
ity that are now widely contested .
S i nce 1 900 non-Western objects have genera l l y been c lass ified as
either pri m i tive art or eth n ographic spec i mens. Before the modern i st rev­
o l ution assoc iated with Picasso and the s i m u ltaneous rise of c u ltural an­
thropo logy assoc i ated with Boas and Ma l i nowski, these objects were d if-

7. On negrophilie see Laude 1 968; for paral lel trends i n l i teratu re see B la­
chere 1 98 1 and Lev i n 1 984 . The d iscovery of thi ngs " negre" by the E u ropean
avant-garde was med i ated by an i magi nary America, a land of noble savages
s i m u ltaneously standing for the past and future of human ity-a perfect affi n ity of
prim itive and modern . For example, jazz was associated with primal sources
(wi ld, erotic passions) and with technology (the mechan ical rhyth m of brushed
drums, the g leam i n g saxophone). Le Corbusier's reaction was characteristic : " I n
a stupid variety show, joseph i ne Baker sang ' Baby' with such an intense and
d ramatic sensibi l ity that I was moved to tears . There is in this Ameri can Negro
music a lyrical 'contem porary' mass so invincible that I cou ld see the fou ndation
of a new sentiment of music capable of being the expression of the new epoch
and also capable of c l assifying its E u ropean origins as stone age-just as has
happened with the new architecture" (quoted in Jencks 1 973 : 1 02). As a sou rce
of modern ist i nspiration for Le Corbusier, the figu re of Josephine Baker was
matched only by monumenta l , almost Egyptian, concrete gra i n elevators, rising
from the American plains and built by nameless "pri m i tive" engi neers (Banham
1 986: 1 6) . The h i stori cal n arrative impl icit here has been a feature of twentieth­
century l i terary and artistic i n novation, as a redemptive modernism pers istently
"di scovers" the pri m itive that can justify its own sense of emergence.

Affinities Not Included in the MOMA "Primitivism " Show.

1 . Bodies
(a) Josephine Baker in a famous pose, Paris, ca. 1 929
(b) Wooden figure (Chokwe, Angola)
(c) Fernand Leger, costume design for The Creation of
the World, 1 922-23

ferently sorted-as anti q u ities, exotic curiosities, orienta l ia, the remains
of early man, and so o n . With the emergence of twentieth-century mod­
ern ism and anth ropology figu res formerly ca l l ed "fetishes" (to ta ke j ust
one c lass of object) became works either of "sc u l ptu re" or of "materia l
cu ltu re ." The d i sti nction between the aesthetic and the anthropological
was soon i n stitutiona l l y rei nforced . I n art ga l leries non-Western objects
were d isplayed for the i r formal and aesthetic q u a l ities; in eth nograph i c
m u s e u m s they were represented i n a "cu ltura l " context . I n t h e l atter an
African statue was a ritual object belonging to a d i sti nct group; it was
d i splayed in ways that e l ucidated its use, sym bol ism, and fu nct ion . The
i nstitutiona l i zed d i sti nction between aesthetic and anthropo logical d i s­
cou rses took form d u r i ng the years documented at MOMA, years that
saw the com pl eme ntary d iscovery of pri m i tive "art" and of an anth ropo-

logica l concept of "culture" (Wi l l iams 1 966) . 8 Though there was from
the start (and cont i n ues to be) a regu l a r traffic between the two domains,
th i s d i sti nction is u n c h a l l enged in the ex h i bition . At MOMA treating
tribal objects as art means exc l ud i ng the origi nal cu ltura l context. Con­
s ideration of context, we are fi rm ly told at the exh i b ition's entrance, i s
the busi ness o f anth ropologists. Cu ltural background i s not essential to
correct aesthetic apprec iation and analysis: good a rt, the masterpiece, i s
u n i versa l l y recogn izable.9 The pioneer modern ists themsel ves knew l i ttle
or noth i n g of these objects' eth nographic meani ng. What was good
enough for Picasso is good enough for MOMA. I ndeed an ignorance of
c u l t u ra l context seems a l most a precondition for artistic apprec iatio n . I n
th i s object system a tri ba l piece i s detached from one m i l ieu i n order to
c i rc u l ate freely i n a nother, a world of art-of museums, markets, and
connoi sseurs h i p .
S i nce t h e early years o f modern ism a n d cultural anth ropo logy non­
Western objects have found a " home" either with i n the d i scourses and
i nstitutions of a rt or with i n those of anthropology. The two dom a i ns have
exc l uded and confi rmed eac h other, inventive l y d isputing the right to
contextual ize, to represent these objects. As we shal l see, the aesthetic­
anth ropo l ogical opposition is systematic, presuppos i ng an u nderl y i ng set
of attitudes towa rd the "tri bal ." Both d i scou rses assume a prim itive world
in need of preservation, redem ption, and representation . The concrete,
i n ventive existence of tribal cu ltures and artists is suppressed in the pro­
cess of e ither constituting authentic, "trad itiona l " worlds or appreciati ng
the i r prod ucts i n the t i meless category of "art."

Noth i ng on West F i fty-th i rd Street suggests that good triba l art is being
produced i n the 1 980s . The non-Western artifacts on d i splay are l ocated

8. The twentieth-century developments traced here redeploy these ideas i n

a n i ntercu ltural domai n wh i l e preserving their older eth ical a n d po l itical charge.
See Chapter 1 0, section 2 .
9 . O n the recogn ition o f masterpieces see Rubin's confident claims
( 1 984 : 20-2 1 ) . H e i s given to statements such as the fol lowi n g on tribal and mod­
ern art: "The sol utions of gen i u s in the plastic arts are a l l essential ly i n sti nctual"
(p. 78, n . 80). A stubborn rejection of the supposed views of anthropologists (who
bel ieve in the col lective prod uction of works of tri bal art) characterizes Rubin's
attempts to clear out an autonomous space for aesthetic judgment. Suggest ions
that he may be projecting Western aesthetic categories onto trad itions with d i f­
ferent defi n itions of art are made to seem simpl istic (for example p. 28).

either i n a vague past (remi n i scent of the l abel " n i neteenth-twentieth

centu ry" that accom pan ies African and Ocea n ian pieces in the Metro­
pol itan M u seum's Rockefe l ler Wi ng) or in a pure l y conceptua l space de­
fi n ed by "prim itive" q u a l ities : magic, ritu a l i sm, c l oseness to natu re,
myth ic or cosmological a i m s (see R u b i n 1 984 : 1 0, 6 6 1 -689) . In this re l ­
egat ion of t h e tri bal or prim itive t o either a van i s h i n g past or an a h i stori­
cal, conceptua l present, modern i st apprec iation reprod uces common
eth nogra p h i c categories.
The same structu re can be seen i n the H a l l of Pacific Peoples, ded­
icated to Margaret Mead, at the American Museum of N atura l H istory.
This new permanent ha l l i s a superbly refu rbished anthro po l ogical stop­
ping p l ace for non-Western obj ects . In Rotunda ( December 1 984), the
m u seum's p u b l i cati o n , a n arti c l e anno uncing the i nsta l l ation conta i n s
t h e fol lowi n g paragraph :

Margaret Mead once referred to the cu ltures of Paci fic peoples as "a
world that once was and now is no more." Prior to her death in 1 9 78
she approved the basic plans for the new Hall of Pacific Peoples. (p. 1 )

We are offered treasu res saved from a destructive h i story, rel i cs of a

van i s h i ng world . Visitors to the i nsta l l ation (and espec i a l l y members of
present Pac ific c u ltures) may fi nd a "world that is no more" more appro­
priately evoked in two ch arm i n g d i sp l ay cases j u st outside the hal l . It i s
t h e world o f a dated anth ropology. Here one finds a neatly typed page of
notes from Mead's m uch-di sputed Samoan research , a picture of the
fieldworker i n teracting "closely" with Mel anesians (she is carrying a
c h i ld on her back), a box of brightly col ored d i scs and triangles once
u sed for psyc hological testing, a copy of Mead 's col u m n in Redbook. I n
the H a l l o f Pac ific Peo p l es artifacts suggest i n g change and syncretism a re
set apart i n a smal l d i s p l ay entitled "Cu lt u re Contact." It i s noted that
Western i nfl uence and i n d i genous response have been active in the Pa­
c i fic s i nce the e ighteenth centu ry. Yet few signs of th is i n vo l vement a p­
pear anywhere e l se i n the l arge h a l l , despite the fact that many of the
objects were made in the past 1 50 years in s ituations of contact, and
despite the fact that the museum's ethnographic explanations reflect
q u i te recent researc h on the cu ltures of the Pac ific. The h i storical con­
tacts and i m p u rities that are part of ethnographic work-and that may
signal the l i fe, n ot the death , of soc ieties-are systematica l l y excl uded .
The tenses of the h a l l 's explanatory captions are revea l i ng. A recent
color photograph of a Samoan kava ceremony is accompa n i ed by the

word s : " STATUS and RAN K were [sic] i m po rtant featu res of Samoan soci­
ety," a statement that wi l l seem strange to anyone who knows how i m­
portant they remai n i n Samoa today. E lsewhere in the hal l a blac k-and­
wh ite photograph of an Austra l ian Aru nta woman and c h i ld, taken
a round 1 900 by the pioneer eth nographers Spencer and G i l len, is cap­
tioned in the present tense. Aboriginals apparently must a l ways i n habit
a myth ic time. Many other examp les of tem pora l i ncoherence cou ld be
c i ted-old Sepi k objects descri bed in the present, recent Trobriand pho­
tos labeled in the past, and so forth .
The poi nt is not s i mply that the i m age of Samoan kava d r i n k i ng and
status soc iety presented here is a d i stortion or that in most of the Ha l l of
Pac ific Peoples h i story has been a i rbrushed out. (No Samoan men at the
kava ceremony are wearing wri stwatches; Trobriand face pa i nting is
shown without noti ng that it i s worn at cricket matches . ) Beyond such
q uestions of accu racy i s an i ssue of systematic ideological cod i ng. To
locate "triba l " peoples i n a non h i storical t i me and ourselves i n a d i ffer­
ent, h i storical t i me is c learly tendentious and no longer c red i ble (Fabian
1 98 3 ) . This recogn ition throws doubt on the perception of a van i s h i ng
tribal world, resc ued , made val uable and mean ingfu l , either as eth no­
graph ic "c u l tu re" or as prim itive/modern "art." For in th i s tem poral or­
dering the rea l or gen u i ne l ife of tri bal works always precedes the i r col­
lection, a n act of salvage that repeats an a l l-too-fam i l iar story of death
and redem ption . In this pervasive a l l egory the non-Western world is a l ­
ways va n i sh i ng and modern i z i ng-as i n Wa lter Benjam i n 's a l legory of
modernity, the tribal world is conceived as a ru i n . (Benjam i n 1 9 77) . At
the H a l l of Pac ific Peoples or the Rockefe l ler Wi ng the actual ongo i ng
l ife and " i mpure" i nventions of tri ba l peoples are erased in the name of
c u l tu ra l or a rtistic "authentic ity." S i m i larly at MOMA the prod uction of
tribal "art" is enti rely i n the past. Tu rn i ng up in the flea markets and
museums of l ate n i neteenth-centu ry E u rope, these objects are desti ned
to be aesthetica l ly redeemed , given new va l ue in the object system of a
generous modern ism .

The story reto ld at MOMA, the struggle to ga i n recogn ition for tribal art,
for its capac ity " l i ke a l l great art . . . to show i mages of man that tran­
scend the parti c u l a r l ives and ti mes of the i r creators" (Rubi n 1 984 : 73), is
taken for granted at another stoppi ng place for triba l travelers i n Manhat­
tan , the Center for African A rt on East Sixty-eighth Street. Susan Vogel ,

the executive d i rector, procl a i m s i n her i ntrod uction to the catal ogue of
its i naugural exh i b ition, "African Masterpieces from the Musee de
! ' H o m me," that the "aestheti c-anthropological debate" has been re­
so lved . It is now widely accepted that "ethnographic speci mens" can be
d i stinguished from "works of art" and that with i n the l atter category a
l i m ited n u mber of " masterpieces" are to be fou n d . Vogel correctl y notes
that the aesthetic recogn ition of tribal objects depends on changes i n
Western taste . For example i t took the work o f Francis Bacon, Lucas Sa­
maras, and others to m a ke it poss i b l e to exhibit as art " rough and horri­
fyi n g [African) works as we l l as refined and lyrical ones" (Vogel
1 985 : 1 1 ) . Once recogn i zed, though, art is apparentl y a rt. Thus the se­
lection at the Cente r i s made on aesthetic criteria alone. A pro m i nent
p l acard affi rms that the abi l ity of these objects "to transcend the l i m ita­
tions of t i me and p l ace, to s peak to us across time and c u lture . . . p laces
them among the h i ghest poi nts of h u m an ac h ievement. It is as works of
a rt that we regard them here and as a testament to the greatness of thei r
There could be no c learer statement of one side of the aesthetic
anthropo logical "debate" (or better, system) . On the other (anth ropo l og­
i c a l ) side, ac ross town, the H a l l of Pac ific Peoples presents co l lective
rather than i nd i v i d u a l prod uctions-the work of "c u lt u res." But with i n
a n i nstitutiona l i zed pol arity interpenetration o f d i scou rses becomes pos­
s i b l e . Sc i ence can be aestheticized, art made anth ropo logi ca l . At the
American M useu m of N atura l H istory ethnographic exh i bits have come
i n c reas i n g l y to rese m b l e a rt shows. I ndeed the H a l l of Pac ific Peoples
rep resents the l atest i n aestheticized scienti sm . Objects are d ispl ayed in
ways that h i g h l i ght the i r formal properties. They are suspended i n l ight,
held in s pace by the i ngen ious use of Plexiglas. (One is sudden l y aston­
i shed by the sheer wei rd ness of a smal l Ocea n i c figu rine perched atop a
th ree-foot-ta l l transparent rod . ) Wh i l e these artisti ca l l y d i splayed artifacts
are sc ientifical l y exp l a i ned , an older, functional ist attempt to present an
i ntegrated picture of spec i fic soc i eties or c u lture areas is no l onge r seri­
ously p u rsued . There i s a n a l most d adaist q u a l ity to the labe l s on eight
cases devoted to Austra l i a n aborigi nal soc iety (I c ite the com plete series
in order) : " C E R EMONY, S P I R I T F I G U R E , MAG I C I A N S AND SORC E R E RS, SAC R E D
E l sewhere the h a l l 's pieces of c u lture have been recontextual ized with i n
a new cybernetic, anthropo l ogical d i scourse. For i nstance fl utes and
str i n ged i n struments are captioned : " M U S I C is a system of organized

Affinities Not Included in the MOMA "Primitivism " Show.

2. Collections
(a) Interior of Chief Shake 's House, Wrangel. Alaska, 1 909

sound i n man's [sic) a u ra l environ ment" or nearby : " COMM U N I CAT ION is
an i m portant fu nction of organized sou nd ."
I n the anthropological H a l l of Pacific Peoples non-Western objects
sti l l have primari ly scientific va l ue . They a re in add ition beautifu l . 1 ° Con­
verse ly, at the Center for African Art artifacts are essentia l ly defi ned as
"masterpieces," the i r makers as great artists . The d i scourse of connois­
seu rsh i p reigns. Yet once the story of art told at MOMA becomes dogma,

1 0 . At the November 3 , 1 984, sympos i u m (see n . 2) Christian Feest pointed

out that the tendency to reclass ify objects in eth nographic col lections as "art" i s
i n part a response t o t h e much greater amount of fu nding ava i l able for art (rather
than anthropological) exh i b itions.

(b) View of the Margaret Mead Hall of Pacific Peoples

it is poss i b l e to rei ntroduce and co-opt the d i scou rse of eth nography. At
the Center tribal contexts and fu nctions are descri bed a l ong with i nd ivid­
u a l h i stories of the objects on d i sp l ay. N ow firm ly c l assified as m aster­
pieces, African objects escape the vague, a h i storical location of the
"tri ba l " or the "prim itive ." The cata logue, a sort of c atalogu e raisonne,
d i scusses each work i ntensive l y. The category of the masterpiece i nd ivid­
uates : the pieces on d i splay are not typica l ; some are one of a kind . The
famous Fan god of wa r o r the Abomey shark-man lend themse l ves to
precise h i stories of i nd ividual creation and appropriation in v i s i b l e co­
l o n i a l situations. Captions spec ify which Griau l e exped ition to West Af­
rica i n the 1 9 3 0s acq u i red eac h Dogan statue (see Le i ri s 1 934 and Chap­
ter 2 ) . We learn in the cata logue that a su perb Bam i l eke mother and c h i l d
was carved b y a n artist named Kwayep, that t h e statue was bought by
the colon i a l ad m i n istrator and anthropo l ogist Henri Labou ret from K i ng

N 'J i ke . Wh i le tribal names predom i nate at MOMA, the Rockefel ler Wi ng,
and the American Museum of Natu ra l H i story, here personal names make
thei r appearance.
I n the "African Masterpieces" cata logue we learn of an eth nogra­
pher's exc i tement on fi nding a Dogan hermaphrod ite figu re that wou ld
l ater become famous. The letter record i ng th i s exc itement, written by
Den i se Pau l me in 1 93 5 , serves as evidence of the aesthetic concerns of
many early eth nographic col l ectors (Vogel and N'diaye 1 985 : 1 2 2 ) .
These i nd i v i d u a l s, w e a re to ld, cou ld i ntu itively d i stinguish masterpieces
from mere art or eth nographic spec i mens. (Actua l l y many of the i nd ivid­
ual eth nographers beh i nd the Musee de I ' Homme col l ection, such as
Pau l me, M ichel Le i ris, Marcel Griau le, and Andre Schaeffner, were
friends and col laborators of the same "pioneer modernist" arti sts who, i n
the story told a t MOMA, constructed the category o f prim itive art. Thus
the i ntu itive aesthetic sense i n q uestion i s the product of a h istorica l l y
spec ific m i l ie u . See Chapter 4 . ) T h e "African Masterpieces" catal ogue
i n s i sts that the fou nders of the Musee de I ' Homme were art con noisseu rs,
that th is great anthropological m u se u m never treated a l l its contents as
"ethnograph ic spec i mens." The Musee de I ' Homme was and is secretly
an a rt m u se u m (Voge l 1 985 : 1 1 ) . The taxonom ic spl it between art and
artifact i s thus hea l ed, at least for se lf-evident "masterpieces," enti rely i n
terms o f the aesthetic code. Art is art i n any m useum .
I n th i s exh i bition, a s opposed to the others i n New York, i nformation
can be provided about each i nd i vidual masterpiece's h istory. We learn
that a Kiwara n i a ntel ope mask studded with m i rrors was acq u i red at a
dance given for the colon i a l ad m i n istration i n Ma l i on Basti l le Day 1 9 3 1 .
A rabbit mask was pu rchased from Dogan dancers at a ga l a so i ree i n
Par i s d u ri ng the Colo n i a l Exh i bition of the same year. These are no longer
the datel ess "authentic" triba l forms seen at MOMA. At the Center for
African Art a d i fferent h istory documents both the artwork's u n i q ueness
and the ach i evement of the d i scern i ng col lector. By featuring rar ity, ge­
n i us, and con noisseu rs h i p the Center confirms the exi stence of autono­
mous a rtworks able to c i rc u l ate, to be bought and sold, in the same way
as works by Picasso or G iacomett i . The Center traces i ts l i neage, appro­
priately, to the former Rockefe l ler Museum of Pri m itive Art, with i ts c lose
ties to col lectors and the art market.
I n its i naugura l exh i b ition the Center confi rms the predomi nant
aesthetic-eth nographic view of tribal art as someth ing located i n the past,
good for bei ng col l ected and given aesthetic val ue. Its second show
(March 1 2 J u ne 1 6 , 1 985) is devoted to " l gbo Arts : Com m u n ity and Cos-

mos ." It te l l s another story, l ocating art forms, ritua l l i fe, and cosmo logy
in a s pec ific, changing African soc iety-a past and present heritage.
Photographs show "trad ition a l " masks worn i n danced masquerades
a round 1 98 3 . (These i n c l ude sati ri c figu res of w h i te colon i sts. ) A deta i l ed
h istory of c u ltura l c h ange, struggle, and reviva l i s provided . In the cata­
logue C h i ke C. A n i a kor, an l gbo scholar, writes along with co-ed i tor H e r­
bert M . Cole of "the conti n u a l ly evo l v i ng lgbo aesthetic" : " I t is i l l usory
to th i n k that w h i c h we comfortably l abel 'trad itiona l ' art was i n an earl ier
time i m m u ne to changes i n sty le and form; it i s th us u n p roductive to
lament changes that reflect c u rrent rea l ities. Conti n u ity with earl ier forms
w i l l a l ways be fou n d ; the p resent-day pers i stence of fam i ly and com­
m u n ity va l ues e n s u res that the arts w i l l thri ve . And as a l ways, the l gbo
w i l l c reate new art forms out of the i r i nventive spi rit, reflecting the i r dy­
nam i c i nteractions with the environment and the i r neighbors and ex­
pressing c u ltu ra l ideals" (Cole and A n i akor 1 984 : 1 4) .
Cole and A n i akor p rovide a q u ite d i fferent h i story o f "the tri ba l " and
"the modern " from that tol d at the Museum of Modern Art-a story of
i nvention, not of rede m ption . In his foreword to the cata logue C h i n u a
Achebe offers a v i s i o n o f c u lture a n d o f objects that sharply c h a l lenges
the ideology of the art co l lection and the masterpiece. l gbo, he tel l s us,
do not l i ke col l ecti ons.

The pu rposeful neglect of the pai nstakingly and devoutly accom­

pl i shed mbari houses with a l l the art objects in them as soon as the
primary mandate of their c reation has been served, provides a signifi­
cant i n s i ght i nto the l gbo aesthetic va lue as process rather than prod­
uct. Process is motion wh i l e product is rest . When the prod u ct is
preserved or venerated, the i m pu l se to repeat the process is com pro­
m i sed . Therefore the l gbo choose to e l i m i nate the prod uct and reta i n
t h e process s o that every occas ion a n d every generation wi l l receive its
own i m p u l se and experience of creation . Interestingly th is aesthetic
d ispos ition receives powerfu l endorsement from the tropical c l i mate
which provides an abu ndance of materials for making art, such as
wood , as wel l as form idable agencies of d issol ution, such as h u m i d ity
and the term ite . Vi sitors to lgboland are shocked to see that arti­
facts a re ra rely accorded any particular value on the basis of age
alone. (Ac hebe 1 984 : ix)

Achebe's i mage of a "ru i n " suggests not the modern i st a l l egory of re­
dem ption (a yearn i ng to make thi ngs whole, to th i n k archaeo logica l ly)

The Earth Deity, Ala, with her "children " in her mbari
house. Obube Ulakwo, southeast Nigeria, 1 966.

but an acceptance of e n d l ess seri a l i ty, a des i re to keep th i n gs apart, dy­

namic, and h i storica l .

The aesthetic-anth ropo l ogica l object systems of the West a re c u rrently

u nder chal lenge, and the pol itics of col l ecting and exh ibiting occasion­
ally become v i s i b l e . Even at MOMA evidence of l iv i ng tri ba l peoples has
not been enti re l y exc l uded . One sma l l text breaks the spel l . A s pecial
labe l exp l a i n s the absence of a Z u n i war god figu re c urrently housed i n
the Ber l i n M u s e u m fu r Vol kerunde. We learn that l ate i n its preparati ons
for the show MOMA "was i nformed by know l edgeable authorities that
Zu n i people consider any p u b l i c exhi bition of the i r wa r gods to be sac­
ri legiou s ." T h u s, the l abel conti n ues, although such figu res a re routinely
d i spl ayed e lsewhere, the m u seu m dec ided not to bri ng the war god (an
i nfl u ence on Pau l K l ee) from Berl i n . The terse note ra i ses more q uestions
than it answers, but it does at l east establ i s h that the objects on d i splay
may i n fact "belong" somewhere other than i n an art or an eth nograph ic
m u seu m . Livi n g trad itions have claims on them, contesting (with a d i s­
tant but i nc reasingly pa l pable power) the i r present home i n the i n stitu­
tiona l systems of the modern West. 1 1
E l sewhere i n New York th is power has been made eve n more vis­
i b l e . "Te Maori ," a show visiting the Metropo l ita n , c learly establ ishes that
the "art" on d isplay i s sti l l sacred , o n loan not mere l y from certa i n New
Zea l and m useu m s but a lso from the Maori people. I n deed tri ba l art is
pol i tical th rough and th rough . The Maori have a l lowed their trad ition to
be exploited as "art" by major Western c u ltura l i nstitutions and their cor­
porate sponsors i n order to en hance thei r own i nternational prestige and

1 1 . The sh ifting balance of power i s evident i n the case of the Zun i war
gods, or Ahauuta . Zuni vehemently object to the d isplay of these figures (terrify­
ing and of great sacred force) as "art." They are the only trad itional objects
si ngled out for th is objection . After passage of the N ative American Freedom of
Rel i gion Act of 1 9 78 Z u n i i n itiated th ree forma l legal actions c l a i m i n g return of
the Ahauuta (wh ich as comm u n a l property are, in Zuni eyes, by defin ition stolen
goods). A sale at Sotheby Parke-Bernet i n 1 9 78 was interru pted, and the figure
was eventua l ly retu rned to the Zu n i . The Denver Art Museum was forced to
repatriate its Ahau utas in 1 98 1 . A c l a i m agai nst the Sm ithsonian remains u n re­
solved as of this writing. Other pressu res have been appl ied elsewhere in a n
ongoing campaign . I n these new conditions Zun i Ahauuta c a n no longer b e rou ­
t i n e l y displayed . I ndeed t h e figure Pau l Klee saw i n Berl i n wou l d have run the
risk of being seized as contraband h ad it been s h ipped to New York for the
MOMA show. For general background see Tal bot 1 98 5 .

Affinities Not Included in the MOMA "Primitivism " Show.

3. Appropriations
(a) Mrs. Pierre Loeb in her family apartment with modern
and tribal works, rue Desbordes-Valmore, Paris, 1 929

th us contri bute to the i r c u rrent resu rgence i n New Zealand soc iety (Mead
1 984) . 1 2 Tri bal authorities gave perm ission for the exh i bition to travel ,
and they partici pated i n its open i ng ceremonies i n a visible, d isti nctive
man ner. So d id Asante l eaders at the exh ibi tion of the i r art and cu ltu re at
the Muse u m of Natural H i story (October 1 6, 1 984 -March 1 7, 1 98 5 ) .
Although t h e Asa nte display centers on eighteenth- a n d n i neteenth­
centu ry artifacts, evidence of the twentieth-century colonial suppression
and recent renewa l of Asante cu lture is incl uded , a long with color photos

1 2 . An article on corporate fund ing of the arts in the New York Times, Feb.
5, 1 985, p. 27, reported that Mob i l O i l sponsored the Maori show in large part
to please the New Zea l and government, with which it was col laborating on the
construction of a natural gas convers ion plant.

(b) New Guinea girl with photographer's flash bulbs

of modern ceremon i es and newly made "trad itiona l " objects brought to
New York as gifts for the m useu m . I n th i s exh i b ition the location of the
art o n d i splay-the sense of where, to whom, and in what ti me(s) it be­
longs- i s q u ite d ifferent from the location of the African objects at
MOMA or i n the Rockefe l ler Wi ng. The tri bal is fu l ly h i storica l .
Sti l l a nother representation of tribal l ife and art can be encountered
at the Northwest Coast co l lection at the I BM Gal lery (October 1 0 -De­
cem ber 2 9 , 1 984), whose obj ects have traveled downtown from the Mu­
seu m of the American I n d i a n . They are d i splayed i n poo ls of i ntense l ight
(the beautifyi ng "boutique" decor that seems to be modern ism's gift to
m useum d i spl ays, both eth nographi c and artisti c) . But th i s exh i bition of
trad itiona l masterpieces ends with works by l iv i n g Northwest Coast art­
i sts . Outside the ga l lery i n the I BM atri u m two large totem poles have
been i nsta l l ed . One i s a weathered spec i men from the Museum of the
American I n d i a n , and the other has been carved for the show by the

Kwaki utl Ca l v i n H u nt. The artist put the fi n is h i ng touches on h i s creation

where it sta nds i n the atr i u m ; fresh wood c h i ps are l eft scattered around
the base. Noth i n g l i ke th i s i s possi b l e or even th i n kable at West Fifty-th i rd
The orga n i zers of the MOMA exh i bition have been c l ear about its
l i m itations, and they have repeated l y spec ified what they do not c l a i m to
show. It i s thus i n a sense unfa i r to ask why they did not construct a
d i fferently focused h istory of re lations between "the tribal " and "the mod­
ern ." Yet the exc l usions bu i l t i nto any col l ection or na rration are l egiti­
mate objects of critiq ue, and the i nsistent, d idactic tone of the MOMA
show o n l y makes its focus more debatable. If the non-Western objects on
West Fifty-th i rd Street never rea l l y question but conti n u a l l y confirm es­
tabl i shed aesthetic va l ues, this raises q uestions about "modern i st primi­
tivism's" pu rported l y revo l utionary potenti a l . The absence of any ex­
amples of Th i rd World modern ism or of recent tri ba l work reflects a
pervasive "se l f-evident" al legory of redemption .
The final room of the MOMA exh i bition, "Contemporary Explora­
tions," w h i c h m ight h ave been used to refocus the h istorical story of mod­
ern ism and the triba l , i nstead strai ns to find contemporary Western arti sts
whose work has a "prim itive feel ." 1 3 Diverse criteria a re asserted : a use
of rough or "natu ra l " materials, a ritua l i stic attitude, ecological concern,
archaeological i n s p i rati on, certa i n tec h n iq ues of assembl age, a concep­
tion of the artist as shaman, or some fam i l i arity with "the m i nd of pri mi­
tive man i n h i s [sic] science and mytho l ogy" (deri ved perhaps from read­
i ng Lev i-Strauss) . Such c riteria, added to a l l the other "pri m itivist"
q u a l ities i nvoked i n the ex h i bition and its cata logue, u n ravel for good the
category of the pri m i tive, exposi ng it as an incoherent c l u ster of q u a l ities
that at d i fferent ti mes have been used to construct a sou rce, origin, or
alter ego confi rm i ng some new "d iscovery" with i n the territory of the
Western self . The exh i b ition i s at best a h i storical account of a certa i n
moment i n th i s relent less process. B y the end the fee l i ng created is one
of claustrophobi a .

1 3 . I n places the search becomes sel f-parod ic, a s i n the caption for works
by jackie Winsor: "Wi nsor's work has a prim itivist feel , not only in the raw phys­
ical presence of her materi a l s, but a l so in the way she fabri cates. Her labor­
driving nails, b i n d i ng twi ne-moves beyond simple systematic repet ition to take
on the expressive character of ritua l i zed action."

The non-Western objects that exc ited Picasso, Dera i n , and Leger
broke i nto the rea l m of offi c i a l Western art from outside. They were
q u ickly integrated , recogn i zed as masterpieces, given homes with i n an
anthropological -aesthetic object system . By now th i s process has been
suffi c iently ce lebrated . We need ex h i bitions that question the boundaries
of art and of the art world, an i nfl ux of tru l y i n d i gesti ble "outside" arti­
facts. The rel ations of power whereby one portion of h u ma n ity can se­
lect, va l ue, and col l ect the pure prod ucts of others need to be criticized
and transformed . T h i s i s no sma l l task . I n the meanti me one can at least
i magi ne shows that featu re the i m pu re, " i nauthentic" productions of past
and prese nt tribal l ife; exh i bitions rad i ca l ly heterogeneo us in the i r global
mix of styles; exh i bitions that locate themselves i n spec ific m u lticu ltura l
j u nctu res ; exh i bitions i n w h i c h nature rema i n s " u n natu ra l " ; exh i b itions
whose pri n c i p l es of i ncorporation are open ly q uestionable. The fol low­
i n g wou ld be my contri bution to a d i fferent show on "affi n ities of the
tribal and the postmodern ." I offer j ust the fi rst paragraph from Ba rbara
Ted l ock's su perb descr i ption of the Z u n i Shal ako ceremony, a fest ival that
is o n l y part of a com plex, l iv i ng trad ition ( 1 984 : 246) :

Imagine a sma l l western New Mex ican v i l l age, its snow- l i t streets
l i ned with white Mercedes, q uarter-ton pickups and Dodge vans. Vi l­
l agers wra pped i n black blankets and flowered shaw l s are stand ing
next to visitors i n blue velveteen blouses with rows of d i me buttons
and volu m i nous sat i n skirts. The i r men are in black Stetson s i l ver­
banded h ats, pressed jeans, Tony Lama boots and m u lticolored Pen­
dleton blan kets . Strangers d ressed in dayglo orange, p i n k and green
ski jackets, stocking caps, h i k i n g boots and m i ttens. Al l crowded to­
gether they are looking i nto newl y constructed houses i l l u m i nated by
ba re l i ght b u l bs dangl i n g from raw rafters edged with Woolworth's red
fabric and flowered b l ue print cal ico. Ci nderblock and pl asterboard
white wa l l s are layered with str i ped serapes, C h i mayo blankets, Nav­
ajo rugs, flowered fri n ged em broidered shawls, black si l k from Mex i co
and purple, red and b l u e rayon from Czechoslovakia. Rows 'of Hopi
cotta � dance k i l ts and ra i n sashes; I s leta woven red and green belts;
Navajo and Zuni s i lver concha belts and black mantas covered with
si lver brooches set with ca rved lapidary, ra i n bow mosa ic, channel in­
lay, turquoise need lepoi nt, pink agate, alabaster, bl ack cannel coal
and bakel ite from old ' 78s, cora l , abalone shel l , mother-of-pearl and
horned oyster hang from poles suspended from the cei l i ng. Mule and

white-ta i led deer trophy-heads wearing squash-blossom , cora l and

chunk-tu rquoise neckl aces are hammered up around the room over
rearing buckskins above Arabian tapestries of Mart i n Luther King and
the Kennedy brothers, The Last Su pper, a herd of sheep with a haloed
herder, horses, peacocks.
There is a Third World in every First World, and vice-versa.

1 0 . On C ollecting Art and Culture

T H I S C H A P T E R is composed of fou r loosely connected parts, each con­

cerned with the fate of tribal artifacts and c u ltura l practices once they are
re located in Western m u se u ms , exchange systems, d isc i p l i nary a rc h i ves,
and d iscu rs ive trad itions. The fi rst pa rt proposes a criti ca l , h istorical ap­
p roach to col lecti ng, foc u s i ng on s u bj ective, taxonom ic, and pol itical
processes . It sketches the "art-cu lture system" through which i n the last
centu ry exotic objects h ave been contextual ized and given va l ue in the
West. This ideo l ogical and i nstitutiona l system i s fu rther explored in the
second part, where c u l tu ra l descr i ption is presented as a form of col l ect­
i ng . The "authentic i ty" accorded to both h u man groups and the i r artistic
work i s shown to proceed from s pec ific ass u m ptions about tem pora l ity,
wholeness, and conti n u ity. The t h i rd part focuses on a revea l i ng moment
i n the modern appropriation of non-Western works of "art" and "c u ltu re,"
a moment portrayed in severa l memo i rs by C l a ude Levi-Strauss of h i s
wartime years i n N e w York. A critical read ing makes expl icit the
redemptive meta h i storical narrati ve these memo i rs presu ppose . The
genera l a rt-c u lture system su pported by such a narrati ve is contested


th roughout the chapter and particularly i n the fou rth part, where alter­
native "triba l " h i stories and contexts are suggested .

Col lecting Ou rselves

You wi l l find yo urs e l f in a c l i m ate of nut castanets,
A musical whip
From the Torres Straits, from Mirzapur a sistrum
Called j umka, "used b y Aboriginal
Tri bes to attract small game
On dark nights," coo l i e ci gar e tt e s
And mask of Saagga, the Devi l Doctor,
The eyelids wo rked by strings.

james Fento n 's poem "The Pitt Rivers Museum , Oxford" ( 1 984 : 8 1 - 84),
from w h i c h th is stanza is take n , red iscovers a pl ace of fasci nation in the
eth nographic col lection . For th is visitor even the museum's desc ri ptive
labe l s seem to i n c rease the wonder (" . . . attract smal l game I on dark
n i g hts" ) and the fear. Fenton i s an adu lt-ch i ld exploring territories of dan­
ger and desi re, for to be a c h i l d i n this co l lection ("Please s i r, where's the
withered I Hand?") is to ignore the serious ad monitions about h u man
evol ution and cu ltura l d ivers ity posted i n the entrance hal l . It is to be
interested i nstead by the c law of a condor, the j aw of a dolph i n , the h a i r
o f a witc h , o r "a jay's feather worn a s a charm I i n Bucki nghamsh i re ."
Fenton's eth nograph ic m u seum is a world of i ntimate encou nters with
i nexpl icably fasc i nati n g objects : personal fetishes. Here col lecting is
i nescapably tied to obsess ion, to recol l ection . Vi sitors "fi nd the l and­
scape of the i r c h i l dhood marked out I Here in the chaotic pi les of sou­
ven i rs . . . boxroom of the forgotten or hard l y poss i b l e ."

As a h i storian of ideas or a sex-offender,
For the prim itive art,
As a dusty sem iologist, eq u i pped to u n ravel
The seven components of that witch's c u rse
Or the syntax of the m uti l ated teeth . Go
In groups to giggle at c u rious fi nds.
But do not step i nto the ki ngdom of your promi ses
O N C O L L E C T I N G A R T A N D C U LT U R E 217

To you rse lf, l i ke a c h i ld entering the forbidden

Woods of h is lonely p layti me.

Do n ot step i n th is tabooed zone " l aid with the snares of pri vacy and
fiction I And the dangerous th i rd w i s h ." Do not encou nter these objects
except as curiosities to giggle at, art to be ad m i red, or evidence to be
u nderstood scientifica l l y. The tabooed way, fol low�d by Fenton, is a path
of too- intimate fantasy, reca l l i ng the d reams of the sol ita ry c h i l d "who
wrestl ed with eagl es for the i r feathers" or the fearfu l vision of a you n g
gi r l , h e r turbu lent lover seen as a hou nd with "strange preterca n i ne eyes."
Th is path th rough the Pitt Rivers Museum ends with what seems to be a
scrap of autobiography, the vision of a personal "forbidden woods"­
exoti c, desired , savage, and governed by the (paterna l ) l aw :

H e h a d known w h a t tortures t h e savages h a d prepared

For h i m there, as he cal m l y pushed open the gate
And entered the wood near the placard : " TAK E NOT I C E M E N
For h i s father had protected h i s good estate.

Fento n 's jou rney i nto otherness l eads to a forbidden area of the self. H is
i nti m ate way of engag i n g the exotic co l lection finds an area of des i re,
marked off and po l iced . The law is preoccu pi ed with property.
C. B . Macpherson's c l assic analysis of Western " possessive i nd i vid­
u a l i sm " ( 1 962) traces the seventeenth-centu ry emergence of an idea l self
as owner: the i nd iv i d u a l s u rro u nded by accu m u l ated property and
goods . The same idea l can hold true for co l l ectivities making and rem ak­
i ng thei r c u ltura l "sel ves ." For exa m p l e R i chard Hand ler ( 1 985) analyzes
the m a k i n g of a Quebecoi s c u ltural "patrimoi ne," drawing on Macpher­
son to u n ravel the assu m ptions and paradoxes i nvol ved in "havi ng a cu l ­
tu re," selecti ng and cherish i ng an authentic col l ective "property." H i s
ana lysis s uggests that t h i s identity, whether c u ltura l or personal , presu p­
poses acts of col lection, gatheri ng u p possess ions i n arbitrary systems of
va l u e and m ea n i ng . Such systems, always powerfu l and rule governed ,
change h i storica l l y. One can not escape the m . At best, Fenton suggests ,
one can transgress ( " poach" i n their tabooed zones) or make the i r se l f­
ev ident o rde rs seem stra nge . I n Handler's s u btly perverse analysis a sys­
tem of retrospection-revealed by a H istoric Mon u ments Comm ission's
selection of ten sorts of "cu l t u ra l property" -appears as a taxonomy war-

thy of Borges' "Ch i nese encyc loped ia" : " ( 1 ) commemorative monu­
ments; (2) c h u rc hes and chape l s ; (3) forts of the French Regime; (4) wind­
m i l l s ; (5) roadside c rosses ; (6) commemorative i nscriptions and p l aques;
(7) devotiona l mon u ments; (8) old hou ses and manors; (9) old fu rn itu re ;
( 1 0) ' l es c hases disparues"' ( 1 985 : 1 99 ) . I n H and ler's d iscussion the col ­
lection a n d preservation o f an authentic doma i n o f identity can not be
natu ra l or i nnocent. It is tied up with national ist pol itics, with restrictive
l aw, and with contested encod i n gs of past and futu re .

Some sort of "gatheri ng" arou nd the self and the group-the assemblage
of a mate ri a l "world," the marki ng-off of a subjective doma i n that is not
"other" - i s probably u n i versal . Al l such col lections embody h ierarch ies
of val ue, exc l usions, ru le-governed territories of the self. But the notion
that this gathering i nvo lves the acc u m u lation of possess ions, the idea that
identity is a kind of wea lth (of objects, knowledge, memories, experi­
ence) , is sure l y not u n iversa l . The i nd i vid u a l i stic acc u m u l ation of Mel a­
nesian "big men" is not possessive in Macpherson's sense, for in Mela­
nesia one acc u m u l ates not to hold objects as private goods but to give
them away, to red istri bute . In the West, however, col lecting has long
been a strategy for the deployment of a possess ive self, c u lture, and au­
thentic ity.
C h i ldren's co l l ections a re revea l i ng i n th is l ight : a boy's accumu la­
tion of m i n iatu re cars, a g i r l 's dol ls, a s u m mer-vacation "natu re m useum "
(with labeled stones a n d shel ls, a h u m m i ngbi rd i n a bott l e), a treasu red
bow l fi l led with the bright shavi ngs of crayons. In these sma l l ritu a l s we
observe the channe l i ngs of obsess ion, an exerc ise in how to make the
world one's own , to gather thi ngs around oneself tastefu l ly, appropriately.
The i n c l u sions i n a l l col lections reflect wider cu ltural ru les-of rational
taxonomy, of gender, of aesthetics. An excessive, someti mes even rapa­
cious need to have is transformed i nto ru l e-governed , meaningfu l des i re.
Thus the self that m u st possess but can not have it a l l learns to se l ect,
order, c lassify in h ierarc h ies-to make "good " col l ections. 1

1 . On col lecting as a strategy of des i re see the h ighly suggestive catalogue

(Hainard and Kaehr 1 982) of an exh i b ition entitled "Co l l ections passion" at the
Musee d' Eth nograph ie, Neuchatel, J u ne to December 1 98 1 . This analytic col lec­
tion of col l ections was a tou r de force of reflexive museology. On col lecti ng and
des i re see also Don na Haraway's bri l l iant analysis ( 1 985) of the American Mu­
seu m of Natural H istory, American manhood, and the th reat of decadence be-
O N C O L L E C T I N G A R T A N D C U LT U R E 2 19

Whether a c h i ld co l l ects model d i nosa u rs or dol ls, sooner or later

she or he wi l l be encou raged to keep the possessions on a shelf or i n a
speci a l box or to set u p a dol l house. Personal treasures wi l l be made
pub l i c . If the passion i s for Egyptian figu ri nes, the col l ector w i l l be ex­
pected to l a be l them, to know the i r dynasty (it i s not enough that they
s i m p l y exude power or m ystery), to te l l " i nteresti ng" th i ngs about them,
to d isti n g u i s h copies from origi n a l s . The good col lector (as opposed to
the obsess ive, the m i ser) is tastefu l and reflective . 2 Acc u m u l ation u nfo lds
in a pedagogica l , ed ifying man ner. The co l l ection itself- i ts taxonom ic,
aesthetic structu re - i s va l ued, and any private fixation on s i ngle objects
is negativel y m arked as fetish i s m . I ndeed a "proper" rel ation with objects
(ru l e-governed possess ion) presupposes a "savage" or deviant relation
(idolatry or erotic fixation) . 3 In S usan Stewa rt's gloss, "The bou ndary be­
tween col lection and fetishism is med i ated by classification and d i splay
i n ten sion with acc u m u l ation and secrecy" ( 1 984 : 1 6 3 ) .
Stewart's wide-ranging study O n Longing traces a "structu re o f de­
si re" whose tas k is the repetitious and i m possi ble one of closing the gap
that sepa rates language from the experience it encodes. She exp lores
certai n recu rre nt strategies pu rsued by Westerners si nce the sixteenth
centu ry. In her analysis the m i n i ature, whether a portrait or dol l 's hou se,
enacts a bou rgeois longing for " i n ner" experience. She a l so explores the

tween 1 908 and 1 93 6 . Her work suggests that the passion to col lect, preserve,
and d i splay is artic u l ated in gendered ways that are h i storica l ly specific. Beau­
cage, Gom i l ia , and Va l l ee ( 1 976) offer critical med itations on the ethnographer's
com plex experience of objects .
2 . Walter Benj a m i n's essay " U n packing My Li brary" ( 1 969 :59-68) provides
the view of a reflecti ve devotee. Col lecting appears as a n art of l iving i ntimately
a l l ied with memory, with obsession, with the salvaging of order from d i sorder.
Benjam i n sees (and takes a certa i n p l easure in) the precariousness of the subjec­
tive s pace atta i ned by the col lection. "Every passion borders on the c haotic, but
the col lector's passion borders on the c haos of memories. More than that: the
chance, the fate, that suffuse the past before my eyes are conspicuously present
in the accustomed confusion of these books. For what else is th i s co l l ection but
a d i sorder to wh ich habit has accommodated itself to such an extent that it can
appear as order? You have a l l heard of people whom the loss of their books has
tu rned i nto inva l i ds, of those who i n order to acq u i re them became criminals.
These are the very areas i n which any order i s a ba lancing act of extreme precar­
iousness." (p. 60)
3. My u nderstanding of the role of the fetish as a m ark of otherness i n West­
ern i n te l lectual h istory-from DeBrosses to Marx, Freud, and Deleuze-owes a
great dea l to the largely u n publ ished work of Wi l l iam Pietz; see "The Problem of
the Feti sh, I" ( 1 985 ) .

strategy of gigantism (from Rabe l a i s and G u l l iver to earthworks and the

b i l l board), the souve n i r, and the col l ection . She shows how col l ections,
most notably m u se u ms-create the i l l usion of adeq uate representation
of a world by fi rst c utti ng objects out of spec ific contexts (whether c u l ­
tu ra l , h i storical , or i ntersubjective) a n d making them "stand for" abstract
wholes-a " Bambara mask," for example, becom i ng an ethnograph ic
metonym for Bambara cu lture. Next a scheme of c l ass ification is elabo­
rated for stori ng or d isplaying the object so that the rea l ity of the col lec­
tion itself, its coherent order, overrides specific h i stories of the object's
prod uction a nd appropriation (pp. 1 62-1 65). Para l lel i ng Marx's account
of the fantastic objectification of commod ities, Stewart argues that i n the
modern Western museum "an i l l usion of a rel ation between th i ngs takes
the p lace of a soc i a l re l ation" (p. 1 65 ) . The co l lector d iscovers, acq u i res,
salvages objects . The objective world i s given, not prod uced, and thus
h i stori ca l rel ations of power i n the work of acq u i sition are occu lted . The
making of mean i n g in m useum c l assification and d i splay is mystified as
adeq uate representation. The time and order of the col l ection erase the
concrete soc i a l labor of its making.
Stewart's work, along with that of Ph i l l i p F isher ( 1 975), Krzysztof
Pom ian ( 1 978), j ames B u n n ( 1 980), Daniel Defert ( 1 982), johannes Fa­
bian ( 1 983), and Remy Saisse l i n ( 1 984), among others, bri ngs col lecting
and d isplay sharply i nto view as crucial processes of Western identity
formation . Gathered artifacts-whether they fi nd the i r way i nto cu rio
cabi nets, private l iving rooms, m useu ms of eth nography, fo l klore, or fine
art-fu nction with i n a developing capita l ist "system of objects" (Baudri l ­
l a rd 1 968). By v i rtue o f t h i s system a world o f value is created a n d a
mea n i ngfu l deployment and c i rc u l ation of artifacts mainta i ned . For
Baudri l la rd co l lected objects create a structu red env i ronment that substi­
tutes its own tem pora l ity for the " real ti me" of h istorica l and productive
processes : "The env i ronment of private objects and their possession-of
wh ich col lections are an extreme manifestation-is a d i mension of our
l ife that i s both essential and i magi nary. As essential as d reams"
( 1 968 : 1 3 5 ) .

A h i story of anth ropology and modern art needs to see i n col l ecting both
a form of Western subjectivity and a changing set of powerfu l i nstitu­
tional practices. The h i story of col lections (not l i m ited to museums) is
centra l to an u nderstand ing of how those soc i a l groups that i nvented
O N C O L L E C T I N G A R T A N D C U LT U R E 22 1

anthropology and modern art have appropriated exoti c th i ngs, facts, and
mea n i ngs. (Appropriate: "to make one's own," from the Latin proprius,
" proper," " property.") It is i m portant to analyze how powerfu l d i sc r i m i ­
nations m a d e a t partic u l a r moments consti tute the genera l system of
objects with i n w h i c h va l ued artifacts c i rc u l ate and make sense. Far­
reach i ng q uestions are thereby rai sed .
What c riteria va l idate an authentic c u ltura l or artistic prod uct? What
are the d ifferenti a l val ues p laced on old and new creations? What mora l
and pol itical criteria j ustify "good ," respons i b l e, systematic col l ecti n g
practices? Why, for example, do Leo Froben i us' wholesale acq u i sitions
of African objects aro u nd the turn of the centu ry now seem excess i ve ?
(See a l so Co l e 1 985 a n d Pye 1 98 7 . ) How i s a "complete" co l lection de­
fined ? What i s the proper bal a nce between scientific analysis and publ ic
d i splay? ( I n Santa Fe a su perb col lection of Nati ve American art i s housed
at the School of American Research i n a bu i l d i ng constructed , l itera l l y,
a s a vau lt, with access carefu l l y restricted . The Musee d e ! ' Homme ex­
h i bits less than a tenth of its co l lections; the rest is stored in steel cabi nets
or heaped in corners of the vast basement . ) Why has it seemed obvious
u nti l recently that non-Western objects shou ld be preserved in E u ropean
museums, even when this means that no fine spec i mens are v i s i b l e i n
thei r cou ntry of origi n ? How are "anti q u ities," "curiosities," "art," "sou­
ven i rs," "mon u ments," and "eth nographic artifacts" d i sti ngu i shed-at
d i fferent h i storical moments and in spec ific m arket cond itions? Why
h ave many anthropological m useu m s in recent years begu n to d isplay
certa i n of the i r obj ects as " masterpieces" ? Why has tou r i st art only re­
cently come to the serious attention of anth ropologists ? (See G raburn
1 976, j u les-Rosette 1 984 . ) What has been the changing i nterpl ay be­
tween natu ral - h i story col l ecti ng and the selection of anthropological ar­
tifacts for d i splay and analysis? The l ist cou ld be extended .
The c ritica l h i story of col lecti ng i s concerned with what from the
materi a l world s pec ific gro u ps and i n d i v i d u a l s choose to preserve, va l ue,
and exchange. A l though th i s complex h istory, from at least the Age of
D i scovery, rem a i n s to be written , Baudri l l ard prov ides an i n itial frame­
work for the deployment of obj ects in the recent capita l i st West. In h i s
accou nt it i s axiomatic that a l l categories o f meani ngfu l objects- i n c l ud­
ing those m arked off as sc i entific ev idence and as great a rt-fu nction
w ith i n a ram i fied system of symbo l s and val ues.
To take j ust one exa m p l e : the New York Times of December 8, 1 984,
reported the w idespread i l lega l l ooti ng of Anasaz i archaeological sites in

the American Southwest. Pai nted pots and u rns thus excavated i n good
cond ition cou ld bring as much as $30, 000 on the market. Another arti c l e
i n t h e same issue conta i ned a photograph o f B ronze Age pots a n d j ugs
sa lvaged by a rchaeo logists from a Phoenician sh i pwreck off the coast of
Turkey. One accou nt featu red c l a ndest i ne col lecting for profit, the other
scientific col lecti ng for know l edge. The mora l eva l u ations of the two acts
of salvage were sharply opposed, bul_ the pots recovered were a l l mean­
i ngfu l , beautifu l , and o l d . Commerc i a l , aestheti c, and sc ienti fic worth in
both cases presupposed a given system of va l u e . Th i s system finds i ntrin­
sic interest and beauty i n objects from a past t i me, and it assu mes that
co l lecti ng everyday objects from ancient (preferably vanished) civi l iza­
tions w i l l be more rewarding than col l ecti ng, for exam ple, decorated
thermoses from modern C h i n a or custom i zed T-sh i rts from Ocea n i a . Old
objects a re endowed with a sense of "depth" by the i r h i storica l ly m i nded
col lectors . Tem pora l ity is reified and sa lvaged as origi n , beauty, and
know l edge .
T h i s a rc h a i z i n g system h a s not a l ways dom inated Western col lect­
i ng. The c u riosities of the New World gathered and appreciated in the
si xteenth century were not necessari ly va l ued as antiqu ities, the prod ucts
of pri m itive or "past" civi l izations . They frequently occu pied a category
of the marvelous, of a present "Golden Age" (Honour 1 9 75 ; M u l l aney
1 98 3 ; Rabasa 1 98 5 ) . More recently the retrospecti ve bias of Western ap­
propriations of the world's c u ltu res has come u nder scruti ny (Fabian
1 98 3 ; C l ifford 1 986b) . C u ltura l or a rtistic "authenticity" has as m uch to
do with an i nventive present as w ith a past, its objectification, preserva­
tion, o r reviva l .

Si nce the turn of the centu ry obj ects col lected from non -Western sou rces
have been c l ass i fied in two major categories : as (scientific) cu ltura l arti­
facts or as (aesthetic) works of art. 4 Other col lecti b les-mass-prod uced
com mod ities, "tou rist art," cu rios, and so on-have been less systemati-

4 . For "hard" articulations of eth nographic cu ltural ism and aesthetic for­
malism see S ieber 1 9 7 1 , Price and Price 1 980, Vogel 1 985, and Rubi n 1 984 .
The first two works argue that art can be u nderstood (as opposed to merely ap­
prec iated) only in i ts original context. Vogel and Rubin assert that aestheti c qual­
ities transcend the i r original local articulation, that "masterpieces" appea l to u n i ­
versal or a t least transcu ltural human sensibil ities. F o r a glimpse o f how the often
incompati ble categories of "aesthetic excel lence," "use," "rari ty," "age," and so
O N C O L L E C T I N G A R T A N D C U LT U R E 223

ca l l y va l ued; at best they f i n d a p l ace i n exh i bits o f "tech nology" or "fo l k­

lore ." These and other locations with i n what may be ca l led the "modern
a rt-cu lture system " can be v i s u a l ized with the h e l p of a (somewhat pro­
crustian) d iagra m .
A . J . G re i m as' "sem iotic sq uare" (Gre i mas a n d Rastier 1 968) shows
u s "that any i n i t i a l b i n a ry oppos ition can, by the operation of negations
and the a ppropriate syntheses, generate a much larger fiel d of terms
w h i c h , however, a l l necessa ri l y rem a i n loc ked in the c l osure of the i n itial
system " (Jameson 1 98 1 :62). Adapting G re i m as for the pu rposes of c u l ­
tural criti c i s m , Fred ric Jameson uses t h e sem i otic sq uare to revea l "the
l i m its of a specific ideo l ogical consc iousness, [marki ng] the conceptua l
poi nts beyond w h i c h that consc iousness can not go, a n d between w h i c h
it i s condemned to osc i l l ate" ( 1 9 8 1 :47). Fol lowing h i s example, I offer
the fol lowi ng map (see d i agram) of a h i storica l l y spec ific, contestible
fie l d of mea n i ngs and i nstitutions .
Begi n n i ng with an i n itial oppos ition, b y a process o f negation fou r
terms a re generated . T h i s establ ishes horizonta l a n d vertica l axes and
between them fou r semantic zones : ( 1 ) the zone of authentic master­
pieces, (2) the zone of a uthentic artifacts, (3) the zone of i nauthentic
masterpieces, (4) the zone of i n a uthentic artifacts . Most objects-old
and new, rare and common, fam i l ia r and exotic-can be located i n one
of these zones or am biguously, i n traffic, between two zones .
The system c lass i fies objects and ass igns them rel ative va l ue . It es­
tab l i shes the "contexts" in w h i c h they prope rly belong and between
w h i c h they c i rc u late . Regu lar movements toward positive va l ue proceed
from bottom to top and from right to left. These movements select arti­
facts of end u ri n g worth or ra rity, the i r va l u e norma l ly guaranteed by a
"va n i s h i ng" cu ltu ra l status or by the selection and pricing mechan isms of
the art ma rket. The va l u e of Shaker crafts reflects the fact that Shaker
soc iety no longer exists : the stock is l i m ited . I n the art world work i s
recogn i zed as " i m portant" b y con n o i sseu rs a n d co l l ectors accord i ng to
criteria that a re more than s i m p l y aesthetic (see Becker 1 98 2 ) . I ndeed ,
preva i l i ng defi n itions of what is "beautifu l " or " i nteresti ng" someti mes
cha nge q u ite rapid l y.
An area of freq uent traffic i n the system i s that l i nk i n g zones 1 and

on a re debated in the exerc ise of assign ing a uthentic va l u e to tribal works, see
the rich ly i nconclusive sym pos i u m on "Authenticity in African Art" orga n i zed by
the journal African Arts (Wi l l ett et al . 1 9 76).


A Machine for Making Authenticity


connOisseurship history and folklore
t h e art museum the ethnographiC museum
the art m arket material culture, craft
art 41----1•
... � culture
orig i n a l , singular traditional, collect1ve

(masterpiece) (artifact)

not-culture not-a rt
new, uncommon
3 4
fakes, inventions tourist art, commodities
the museum of technology the curio colleCtiOn
ready-mades and ant1-art utilities


2 . Objects move in two d i rections a long th i s path . Thi ngs of cu ltural or

h i storical va l ue may be promoted to the status of fine art. Examples of
movement i n this d i rection, from eth nographi c "culture" to fine "art," are
plentifu l . Tri ba l objects located i n art ga l leries (the Rockefe l ler Wi ng at
the Metropo l itan Museum in N ew York) or d i splayed anywhere accord­
i n g to "forma l i st" rather than "contextua l i st" protocol s (Ames 1 98 6 : 3 9-
42) move i n th i s way. Crafts (Shaker work co l lected at the Wh itney Mu­
seum i n 1 986), "fo l k art," certa i n antiq ues, "naive" art a l l are subject to
period i c promotions. Movement i n the inverse d i rection occ urs when­
ever art masterworks are cu ltura l ly and h istorica l ly "contextua l ized,"
someth i n g that has been occurring more and more expl icitly. Perhaps the
most dramatic case has been the relocation of France's great i mpress ion­
ist col lection, formerly at the Jeu de Pau me, to the new Muse u m of the
N i neteenth Centu ry at the Gare d'Orsay. Here art masterpieces take thei r
O N C O L L E C T I N G A R T A N D C U LT U R E 225

p l ace i n t h e panorama of a h i storica l -cu l tu ra l "period ." T h e panorama

i n c l udes a n emerg i ng i n d u stri a l u rban ism and its tri u m phant technology,
"bad" as we l l as "good" a rt. A l ess d ramatic movement from zone 1 to
zone 2 can be seen i n the routine process with i n art gal leries whereby
objects become "dated," of i n terest l ess as i m med iatel y powerfu l works
of gen i u s than as fi ne exa m p les of a period style.
Movement a lso occu rs between the l ower and u pper h a l ves of the
syste m , u s u a l l y in an u pward d i rection . Commod ities in zone 4 regu larly
enter zone 2 , beco m i n g rare period pieces and th us col lectibles (ol d
green g l ass Coke bottles) . Much c u rrent non-Western work m igrates be­
tween the status of "tou rist a rt" and creative c u ltura l -artistic strategy.
Some c u rrent prod uctions of Th i rd World peoples have ent i re l y shed the
stigma of modern commerc i a l i nauthenticity. For examp le H aitian "prim­
itive" painti ng-commerc i a l and of rel ative l y recent, i m pu re origi n ­
has moved fu l l y i nto t h e art-cu lt u re c i rc u it. S i gn i ficantly th i s work e n ­
tered t h e a rt market by assoc i ation w i t h z o n e 2 , becom i n g va l ued as t h e
work not s i m p l y of i n d i v i d u a l a rtists b u t o f Haitians. H a itian pai nting i s
su rrounded by spec i a l associations w i t h t h e land o f voodoo, m a g i c a n d
negritude. T h o u g h speci fic a rti sts have come t o b e known and prized ,
t h e a u ra o f "cu ltu ra l " p rod uction attaches t o them m u c h more than, say,
to Pi casso, who is not i n any essenti a l way va l ued as a "Spanish artist."
The same i s true, as we sha l l see, of many recent works of tri bal a rt,
whether from the Sep i k or the American Northwest Coast. Such works
h ave l argely freed themselves from the tou rist or com mod ity category to
w h i c h , beca use of their m odernity, purists had often relegated the m ; but
they can not move d i rectly i nto zone 1 , the a rt market, without tra i l i n g
c l ouds o f a uthentic (trad itional) c u lture . There can b e no d i rect move­
ment from zone 4 to zone 1 .
Occasi o n a l trave l occ u rs between zones 4 and 3 , for examp le when
a commod ity or tec h n o logical artifact i s percei ved to be a case of spec i a l
i n ventive c reation . T h e object i s selected o u t o f commerc i a l or m ass c u l ­
ture, perh aps t o b e featu red i n a m u seum o f technology. Someti mes s u c h
objects fu l l y enter t h e rea l m o f a rt : "tec h nologica l " i n n ovations or com­
mod ities may be contextua l i zed as modern "design," thus passi ng
through zone 3 i nto zone 1 (for example the fu rn iture, household ma­
c h i nes, cars, and so on d i sp l ayed at the Museum of Modern Art in New
York) .
There is a l so regu l a r traffic between zones 1 and 3 . Exposed a rt for-

geries are demoted (wh i le nonetheless preserving someth i ng of the i r orig­

i na l a u ra). Converse ly various forms of "anti-art" and a rt parad ing its un­
ori g i n a l ity or " i nauthenticity" are col lected and val ued (Warho l 's soup
can, S herrie Levi ne's photo of a photo by Wal ke r Evans, Duchamp's
u ri na l , bottle rack, or shove l ) . Objects i n zone 3 are a l l potential l y col­
lecti ble with i n the ge nera l dom a i n of art: they are u ncommon, sharply
d i sti nct from or blatantly cut out of cu lture . Once appropriated by the art
world, l i ke D u cham p's ready-mades, they c i rcu l ate with i n zone 1 .
The art-cu lture system I have d iagramed exc l udes and margi nal izes
various res idual and emergent contexts . To mention on ly one: the cate­
gories of art and cu lture, technology and com mod ity are strongly sec u l a r.
" Re l i g ious" objects can be val ued as great art (an a ltarpiece by Giotto),
as fo l k a rt (the decorations on a Lat i n American popular sai nt's shri ne),
or as cu ltural a rtifact (an I nd ian rattle) . Such objects h ave no i n d i v i d u a l
"power" or mystery-qual ities once possessed b y "fetishes" before they
were rec l assi fied in the modern system as pri m i tive art or cu ltura l artifact.
What "va l ue," however, i s stripped from an altarpiece when it i s moved
out of a fu nction i ng c h u rch (or when its church begins to fu nction as a
muse u m ) ? Its spec ific power or sacred ness is relocated to a genera l aes­
thetic rea l m . (See Chapter 9, n . 1 1 , on a recent chal lenge by Zuni triba l
authorities to such secu lar contextual i zations . )

It is i m po rta nt t o stress t h e h i storic ity o f th i s art-cu lture syste m . It h a s not

reached its fi n a l form : the pos itions and va l ues assi gned to col lecti ble
artifacts have changed and wi l l continue to do so. Moreover a syn­
chronic d i agram can not represent zones of contest and transgress ion ex­
cept as movements or ambigu ities among fixed po les. As we sha l l see at
the end of th i s chapter, m u c h c u rrent "tri bal art" partici pates in the reg­
u lar art-cu lture traffic and in trad itional spiritu a l contexts not accou nted
for by the system (Coe 1 986). Whatever its contested doma i n s , though,
genera l l y spea k i ng the system sti l l confronts any co l lected exotic object
with a stark alternative between a second home in an ethnographic or an
aesthetic m i l ie u . The modern ethnographic museum and the a rt museum
or private art col l ection have developed separate, comp lementary modes
of classification . In the former a work of "sc u l ptu re" is d i sp l ayed a long
with other objects of s i m i lar function or i n proxi m ity to objects from the
same cu ltural group, i n c l u d i n g uti l itarian a rtifacts such as spoons, bowls,
or spears . A mask or statue may be grou ped with form a l l y d i ssimi lar ob-
O N C O L L E C T I N G A R T A N D C U LT U R E 227

jects a n d exp l a i ned as part o f a ritual or i nstitution a l complex. T h e names

of i ndividual scul ptors a re u n known or s u ppressed . In art museums a
scu l pture i s identified as the c reation of an i ndividual : Rod i n , G i aco­
metti, Barbara Hepworth . I ts pl ace in everyday c u ltural practices ( i n c l ud­
ing the market) is i rre levant to its essenti a l mea n i ng. Whereas i n the eth­
nogra p h i c museum the object is c u ltura l ly or h u m a n l y " i n te resti ng," in
the a rt m useum it is primari ly " beautifu l " or "orig i na l ." It was not a l ways
thu s .
E l izabeth Wi l l i a m s ( 1 985) has traced a revea l i ng chapter i n t h e sh ift­
i n g h istory of these d i sc ri m i nations. I n n i neteenth-centu ry Pa ris it was
d ifficu lt to conceive of pre-Co l u m bian artifacts as fu l ly "beautifu l ." A pre­
va i l i n g n at u ra l i st aesthetic saw ars Americana as grotesque or crude. At
best pre-Col u m bian work cou ld be ass i m i l ated into the category of the
antiqu i ty and a pprec iated through the fi lter of Vio l l et- le-Duc's med i eva l­
ism . Wi l l iams shows how Mayan and l ncan a rtifacts, their status uncer­
ta i n , m ig rated between the Louvre, the B i b l iotheq ue N ationale, the Mu­
see G u i met, and (after 1 8 78) the Trocadero, where they seemed at l ast to
find an ethnograph ic home i n an i n stitution that treated them as scientific
evidence. The Trocadero's fi rst d i rectors, Ernest-Theodore Hamy and
Remy Vernea u , showed scant i nterest i n the i r aesthetic q u a l ities.
The "beauty" of much non-Western "art" is a recent d i scovery. Be­
fore the twentieth centu ry many of the same objects were col lected and
valued, but for d ifferent reasons. In the early modern period the i r rarity
and strangeness were pri zed . The "cabi net of c u riosities" j u mbled every­
th i n g together, with each i n d i v i d u a l object stand ing metony m i ca l l y for a
whole region or popu l ation . The co l lection was a m ic rocosm, a "su m­
mary of the u n iverse" ( Pom i a n 1 978). The eighteenth centu ry introd uced
a more serious concern for taxonomy and for the e l aboration of com p l ete
series. Col lect i n g was i n c reas ing l y the concern of sc ientific natu ra l i sts
(Feest 1 984 : 90), and objects were val ued because they exe m p l i fied an
array of syste m atic categories : food , c loth i ng, bu i ld i ng materi als, agri­
c u l t u ra l tool s , weapon s (of war, of the h u nt) , and so forth . E . F. jomard's
eth nogra p h i c c l ass i fications and A. H . L. F. Pitt Rivers' typo l ogical d i s­
pl ays were m id - n i n eteenth-century cu l m i nations of this taxonom ic vision
(Chapman 1 985 : 24-2 5 ) . Pitt Rivers' typo logies featu red deve lopmenta l
seq uences . By the end of the centu ry evo l utionism had come to domi­
nate arrangements of exotic a rtifacts . Whether objects were presented as
antiqu ities, arranged geograph i ca l l y or by society, spread in panopl ies,
or arranged i n rea l i stic " l ife gro u ps" and d ioramas, a story of human

deve lopment was tol d . The object h ad ceased to be primari l y an exotic

"curiosity" and was now a sou rce of i nformation entirely integrated in
the u n iverse of Western Man (Dias 1 985 : 3 78 -3 79). The val ue of exotic
objects was the i r abi l ity to testify to the concrete rea l ity of an earl i er stage
of h u man Cu lture, a common past confi rm i ng E u rope's tri u m phant
With Franz Boas a nd the emergence of re l ativist anth ropology an
emphasis on p l ac i ng objects in spec ific l ived contexts was consol idated .
The "cu l tures" th us represented cou ld either be arranged i n a modified
evol utionary series or d ispersed in synchronous "eth nograph ic presents."
The l atter were ti mes neither of antiqu ity nor of the twentieth centu ry but
rather representing the "authentic" context of the col lected objects, often
j u st prior to thei r co l lection or d i spl ay. Both co l lector and salvage eth­
nographer cou ld c l a i m to be the l ast to rescue "the rea l thi ng." Authen­
ticity, as we sha l l see, is produced by remov ing objects and cu stoms from
the i r cu rrent h i storical situation-a present-becom i n g-futu re.
With the con so l idation of twentieth-centu ry anth ropology, a rtifacts
contextua l i zed ethnograph i ca l l y were val u ed because they served as ob­
jective "wi tnesses" to the total m u ltid i mensional l ife of a cu lture (Ja m i n
1 982a: 89-9 5 ; 1 98 5 ) . S i m u ltaneously with new developments i n art and
l iteratu re, as Pi casso and others began to visit the "Troca" and to accord
its tribal objects a nonethnographic ad m i ration, the proper pl ace of non­
Western objects was aga i n thrown in q uestio n . In the eyes of a tri um­
phant modernism some of these artifacts at l east cou ld be seen as uni­
versa l masterpieces . The category of "pri m i ti ve art" emerged .
Th i s development i ntroduced new ambigu ities and possibi l ities i n a
changing taxonom i c syste m . I n the mid-ni neteenth centu ry pre­
Col u m bi a n or tribal objects were grotesq ues or antiq u ities. By 1 920 they
were cu ltura l witnesses and aesthetic masterpieces . Si nce then a con­
trol led m i gration has occu rred between these two i nstitution a l i zed do­
mains. The bou ndaries of art and sc ience, the aesthetic and the anth ro­
pological, are not permanent ly fixed . I ndeed anthropology and fine a rts
museums have recently shown signs of i nterpenetration . For exam ple the
Hal l of Asian Peoples at the New York Museum of Natura l H i story re­
flects the " boutiq ue" style of d i spl ay, whose objects cou ld never seem
out of pl ace as "art" on the wal l s or coffee tables of m iddle-class l iv i ng
rooms. I n a complementary development downtown the Museum of
Modern Art has expanded its permanent exh i bit of c u l tu ra l artifacts : fu r­
n iture, a utomobi les, home app l iances, and utens i ls-even hangi ng from
O N C O L L E C T I N G A R T A N D C U LT U R E 229

the cei l i ng, l i ke a Northwest Coast w a r canoe, a m uch-ad m i red bright

green h e l i copter.

Wh i l e the object systems of a rt and anth ropology are i nstitutional ized

and powerfu l , they a re not i m m utabl e . The categories of the beautifu l ,
the c u ltura l , a n d the a uthentic have changed and are changing. Thus it
i s i mporta nt to resi st the tendency of co l lections to be self-sufficient, to
suppress the i r own h i storica l , economic, and pol itical processes of pro­
d uction (see Haacke 1 9 7 5 ; H i l ler 1 9 79) . Idea l ly the h i story of its own
co l lecti o n and d isplay shou ld be a v i s i b l e aspect of any exh i bition . It had
been rumored that the Boas Room of Northwest Coast artifacts i n the
American Museum of N atural H istory was to be refu rbished , its style of
d i s p lay modern i zed . Apparently (or so one hopes) the plan has been
abandoned , for th i s atmospheric, dated ha l l exh i bits not merely a su perb
col lection but a moment in the h i story of col lecti ng. The widely publ i ­
ci zed M u se u m o f Modern Art show o f 1 984, " ' Pri m itivism' i n Twentieth­
Centu ry Art" (see Chapter 9), made a pparent (as it celebrated) the prec ise
c i rc u m sta nce in w h ich certa i n eth nograph i c objects suddenly became
works of u n i versal art. More h i storica l self-consc iousness in the d i s p l ay
and v iewi ng of non-Western obj ects can at least jostle and set i n motion
the ways in which anthropolog i sts, a rti sts, and the i r pub l i cs col lect them­
se lves and the worl d .
A t a more i nt i m ate l eve l , rather than graspi ng objects on l y a s c u l ­
tura l s i g n s and artistic icons (G u i d ieri a n d Pe l l izzi 1 98 1 ), we c a n retu rn
to them , as J ames Fenton does, thei r lost status as fetishes- not spec i­
mens of a devi ant or exotic "fet i s h i s m " but our own fetishes. 5 Th i s tactic,
necessari l y pe rso n a l , wou ld accord to th i ngs i n co l l ections the power to
fixate rather than s i m ply the ca pac ity to ed i fy or i nform . African and
Ocea n i a n artifacts cou ld once aga i n be objets sauvages, sources of fas­
c i n ation with the power to d isconcert. Seen in thei r resistance to c l assi­
fication they cou ld rem i nd u s of our lack of self-possess ion, of the arti­
fices we e m ploy to gather a world around us.

5. For a post-Freudian positive sense of the fetish see Lei ris 1 929a, 1 946;
for fetish theory's rad ical possibi l ities see Pietz 1 985, which d raws on Deleuze;
and for a repentant sem iologist's perverse sense of the fetish (the "punctum") as
a p l ace of strictl y personal mean ing unformed by c u l tu ra l codes (the "stu d i u m " )
see Barthes 1 980. Gom i l a ( 1 9 76) reth i nks eth nographic material cultu re from
some of these surrea l ist-psychoanalytic perspectives .

C ulture Collecting

Fou nd i n American Anthropologist, n . s. 34 ( 1 932) : 740:

A l iatoa, Wiwiak D istrict, New G u i nea
April 2 1 , 1 932
We a re just com pleting a culture of a mounta i n group here in the
lower Torres Chel les. They have no name and we haven't decided
what to cal l them yet. They are a very revea l i ng people in spots, pro­
viding a final basic concept from which a l l the mother's brothers'
cu rses and father's sisters' c u rses, etc . derive, and having articulate the
attitude toward incest which Reo [ Fortune] outli ned as fundamental in
h i s Encyc loped ia a rticle. They h ave taken the therapeutic measu res
which we recom mended for Dobu and Manus-having a dev i l in ad­
d ition to the neighbor sorcerer, and havi ng got the i r dead out of the
vi l lage and local ized . But in other ways they are annoying: they have
bits and snatches of a l l the rag tag and bob tai l of magical and ghostly
bel ief from the Pacific, and they are somewhat l i ke the Plains i n their
receptivity to strange ideas. A picture of a local native reading the
i ndex to the Golden Bough j ust to see if they had missed anyth ing,
wou l d be appropriate. They are very d ifficu lt to work, l iving a l l over
the place with half a dozen garden houses, and never staying put for a
week at a time. Of course this offered a new chal lenge i n method
which was interesting. The difficu lties i ncident upon being two days
over i m possible mountains have been consum i ng and we are goi ng to
do a coasta l people next.
Sincerely yours,

"Cu ltures" are eth nographic col lections. S i nce Tylor's fou nd i n g def­
i n ition of 1 8 7 1 the term has designated a rather vague "complex whole"
i n c l ud i ng everyth i n g that is l earned group behavior, from body tec h­
n iq ues to sym bo l i c orders . There have been recurring attem pts to defi ne
c u l tu re more prec ise l y (see Kroeber and Kl uc khoh n 1 952) or, for ex­
ample, to d isti ngu ish it from "soc i a l structu re ." But the i n c l usive use per­
sists. For there are ti mes when we sti l l need to be able to speak hol i stic­
a l l y of Japanese or Trobriand or Morocca n c u l tu re in the confidence that
we are designating someth i ng rea l and d ifferenti a l ly coherent. It is in-
O N C O L L E C T I N G A R T A N D C U LT U R E 23 1

creas i ngly clear, however, that the conc rete activity of representi ng a
cu lture, subcu lture, or i ndeed any coherent doma i n of col l ective activity
is always strategic and selective . The worl d 's soci eties are too systemati­
cal l y i nterconnected to perm i t any easy isolation of separate or i ndepen­
dently fu ncti o n i n g systems (Marcus 1 986) . The i ncreased pace of h i stor­
ical c h ange, the com mon rec u rrence of stress in the systems u nder study,
forces a new self-consciousness about the way c u ltura l wholes and
bou ndaries are constructed and transl ated . The pioneering elan of Mar­
garet Mead "com plet i n g a c u l tu re" in h i gh l and New G u i nea, col l ecti ng
a d ispersed popu lation, d iscovering its key c u stoms, nam i n g the resu l t­
i n th is case "the Mounta i n Arapes h " - i s no longer poss i bl e .
To s e e eth n ography as a form o f c u l tu re co l lecti ng (not, o f cou rse,
the only way to see i t) h i g h l ights the ways that d i verse experiences and
facts a re selected , gathered, detached from the i r original tem pora l oc­
casions, and give n enduring va l u e in a new arrangement. Col lecti n g­
at l east i n the West, where t i m e is genera l ly thought to be l i near and
i rrevers i ble- i m p l ies a rescue of phenomena from i nevitable h i storical
decay or loss. The co l lection conta i ns what "deserves" to be kept, re­
membered, and treasu red . Artifacts and customs a re saved out of t i me . 6
Anthropologica l c u lture co l l ectors h ave typica l ly gathered what seems
"trad ition a l "-what by defi n ition i s opposed to modern ity. From a com­
plex h i storical rea l i ty (wh i c h i nc l udes cu rrent eth nograph ic encounters)
they sel ect what gives form , structu re, and cont i n u ity to a worl d . What
i s hybrid or " h istorica l " in an emergent sense has been less common l y
co l lected and presented as a system o f authenticity. For example i n New
G u i nea Margaret Mead and Reo Fortune chose not to study gro u ps that
were, as Mead wrote in a letter, " bad l y m ission i zed" ( 1 9 7 7 : 1 2 3 ) ; and it
h ad been self-evident to Ma l i nowsk i in the Trobriands that what most

6 . An exh i bition, "Temps perd u , tem ps retrouve," held during 1 985 at the
Musee d ' Eth nograph ie of Neuchatel systemati cal ly interrogated the tem poral pre­
d i cament of the Western eth nographic m useu m . Its argu ment was condensed i n
the fol lowi n g text, each proposition o f wh ich was i l l ustrated museograph ical ly:
" Prestigious pl aces for locking thi ngs up, museums give val ue to thi ngs that a re
outside of l i fe : i n this way they resemble cemeteries. Acq u i red by d i nt of dol l ars,
the memory-objects partici pate in the grou p's changing identity, serve the powers
that be, and acc u m u l ate i n to treasu res, w h i l e personal memory fades. Faced with
the aggressions of everyday l i fe and the passing of phenomena, memory needs
objects-always man i p u l ated through aesthetics, selective emphasis, or the m ix­
ing of gen res. F rom the perspective of the future, what from the present should
be saved ?" ( H a i nard and Kaehr 1 98 6 : 3 3 ; also H a i nard and Kaehr 1 985 . )

deserved scientific attention was the c i rcu mscri bed "cultu re" th reatened
by a host of modern "outside" i nfl uences . The experience of Melanesians
becom i n g Ch ristians for thei r own reasons- l earn i ng to pl ay, and p l ay
with, the outsiders' games-did not seem worth sa l vagi ng.
Every appropriation of c u l tu re, whether by i n siders or outsiders, im­
p l i es a spec ific tempora l pos ition and form of h i storical narratio n . Gath­
ering, own i ng, c l assifyi ng, and val u i ng are certa i n l y not restricted to the
West; but e l sewhere these acti vities need not be assoc i ated with accu­
m u lation (rather than red i stribution) or with preservation (rather th an nat­
ural or h istorical decay) . The Western practice of cultu re col lecti ng has
its own local genea l ogy, enmeshed in d isti n ct E u ropean notions of tem­
poral ity and order. It is worth dwel l i ng for a moment on this genealogy,
for i t o rga n i zes the ass u m ptions bei ng arduously u n lea rned by new theo­
ries of practi ce, process, and h i storicity (Bou rd i eu 1 9 77, G iddens 1 979,
Ortner 1 984, Sah l i n s 1 98 5 ) .
A cruc i a l aspect o f t h e recent h i story o f t h e cu l tu re concept h a s been
its a l l iance (and d ivision of labor) with "art." C u l tu re, even without a
capital c, stra i n s toward aesthetic form and autonomy. I have a l ready
suggested that modern c u l tu re ideas and art ideas fu nction together in an
"art-cu l ture system ." The i nc l usive twentieth-centu ry cu lture category­
one that does not privi lege "h igh" or " l ow" c u l tu re- i s plausible only
with i n th is syste m , for wh i le i n pri n c i p l e ad mitti ng a l l learned human
behavior, th i s c u lture with a smal l c orders phenomena i n ways that priv­
i l ege the coherent, ba lanced , and "authentic" aspects of shared l ife.
S i nce the m id-n i neteenth centu ry, ideas of c u ltu re have gathered u p
those elements that seem to give conti n u ity a n d depth to col lective ex­
i stence, see i n g it whole rather than d i sputed , torn , i ntertextua l , or syn­
cretic . Mead 's a l most postmodern i m age of "a local native read i ng the
i ndex to The Colden Bough j ust to see if they had m issed anyth i ng" is
not a vision of authenticity.
Mead found Arapesh receptivity to outs ide influences "an noyi ng."
Their culture col lecti ng com p l i cated hers . H istorica l deve lopments
wou ld later force her to provide a rev ised picture of these d ifficu lt Mela­
nesians. In a new preface to the 1 9 7 1 reprint of her th ree-vo l u me eth­
nography The Mountain Arapesh Mead devotes severa l pages to letters
from Bernard Narokobi , an Arapesh then studyi n g l aw i n Syd ney, Austra­
l i a . The anthropologist read i ly admits her astonishment at heari ng from
h i m : "How was it that one of the Arapesh-a people who had had such
a l ight hold on any form of co l lective style-shou ld have come fu rther
O N C O L L E C T I N G A R T A N D C U LT U R E 233

th a n any i nd iv i d u a l a m o n g t h e Man us, w h o h ad moved as a group i nto

the modern world i n the years between our fi rst study of them, i n 1 928,
and the begi n n i ng of o u r restudy, i n 1 95 3 ?" (Mead 1 97 1 : ix) . She goes on
to exp l a i n that N a rako b i , along with other A rapesh men study i ng i n Aus­
tra l ia, had "moved from one period in h u man c u l tu re to another" as " i n ­
d iv i d u a l s ." T h e Arapesh were " less tightly bound with i n a coherent c u l ­
ture" t h a n Manus (pp. i x-x) . Narakobi writes, however, as a member of
h i s "tri be," spea k i n g with pride of the va l ues and accom p l i shments of h i s
"c l a n sfo l k." ( H e uses t h e n a m e Arapes h spari ngly. ) H e arti c u lates t h e pos­
s i b i l ity of a new m u ltiterritorial "cu ltura l " identity : "I feel now that I can
fee l proud of my tribe and at the same time feel I belong not only to
Papua-New G u i nea, a nation to be, but to the world com m u n i ty at
large" (p. x i i i ) . Is not this modern way of bei ng "Arapesh " a l ready prefig­
u red in Mead's earl ier i m age of a resou rcefu l native pagi ng through The
Colden Bough ? Why m u st such behavior be margi n a l i zed or c l assed as
" i nd i vid u a l " by the anth ropological cu lture col l ector?
Expectations of wholeness, conti n u ity, and essence have long been
b u i lt i n to the l i n ked Western ideas of c u l tu re and a rt. A few word s of
recent background m ust s u ffice, s i nce to map the h i story of these con­
cepts wou l d lead u s on a c hase for origi ns back at l east to the G reeks.
Raymond Wi l l iams provides a start i n g poi n t i n the early n i neteenth cen­
tu ry-a moment of u n precedented h istorica l and soc i a l d isruption . I n
Culture and Society ( 1 966), Keywords ( 1 976), and e lsewhere Wi l l iams
has traced a para l l e l deve lopment in u sage for the words art and culture .
The changes reflect com p lex responses to i n d ustri a l i s m , to t h e specter of
"mass soc iety," to acce lerated soc i a l confl ict and change . 7
Accord i n g t o Wi l l iams i n t h e e ighteenth century the word art meant
pred o m i n a ntly "ski l l ." Cab i n etmakers, cri m i na l s, and pa i nters were each
in thei r way artfu l . Culture design ated a tendency to natural growth, its
u ses predo m i nantly agri c u ltura l and personal : both pla nts and h uman
i nd iv i d u a l s cou l d be "cu ltured ." Other mea n i ngs a l so present in the e i gh­
teenth centu ry did not predom i nate u nti l the n i n eteenth . By the 1 82 0s
art i nc reasi ngly designated a s pec ial doma i n of creativity, spontaneity,
and purity, a real m of refi n ed sensi b i l ity and expressive "ge n i us ." The

7 . Although Wi l l iams' analysis i s l i m ited to England, the general pattern

appl ies e lsewhere i n E u rope, where the t i m i ng of modern ization d i ffered or
where other terms were u sed . In France, for example, the words civilisation or,
for D u rkhei m , societe stand i n for culture. What i s at issue are genera l q u a l i tative
assessments of col lective l ife.

"artist" was set apart from, often aga i n st, soc iety-whether "mass" or
"bou rgeo i s ." The term culture fol lowed a para l lel cou rse, com i ng to
mean what was most e l evated , sen sitive, essenti a l , and prec ious-most
u ncommon- i n soc iety. L i ke art, cu lture became a genera l category;
Wi l l iams ca l l s it a "fi n a l cou rt of appea l " aga i nst th reats of vu l garity and
leve l i ng . It ex i sted in essenti a l oppos ition to perceived "anarchy."
Art and c u l tu re emerged after 1 800 as m utua l l y rei nforc i ng domains
of h u man va lue, strategies for gatheri ng, marki ng off, protecting the best
and most interesting creations of "Man ." 8 I n the twentieth century the
categories u nderwent a series of fu rther deve lopments . The p l u ra l , an­
th ropologica l defi n ition of cu lture ( l ower-case c with the possibi l ity of a
fi nal s) emerged as a l i bera l a lternative to rac i st c l assifications of human
d i versity. It was a sensitive means for understand i ng different and d is­
persed "whole ways of l ife" in a h igh colon i a l context of unprecedented
global i ntercon nectio n . Culture i n its fu l l evol utionary rich ness and au­
thenticity, formerly reserved for the best creations of modern E u rope,
cou ld now be extended to a l l the world's pop u l ations. In the anthropo­
logica l vision of Boas' generation "cu l tu res" were of eq ual val ue. In the i r
new p l u ra l ity, however, t h e n i neteenth-centu ry defi n itions were not en­
t i rely transformed . If they became l ess el itist (d i sti nctions between "h igh"
and " l ow" cu lture were erased) and less E u rocentric (every human soc i ­
ety w a s fu l l y "cu ltu ra l " ) , nevertheless a certai n body o f assu m ptions were
carried over from the older defi n itions . George Stocking ( 1 968 : 69-90)
shows the com plex i nterre l ations of n i neteenth-century h u man i st and
emerg i n g anthropo logical defi n itions of cu lture . He suggests that anth ro-

8. As Vi rgi n i a Dom i nguez has argued, the emergence of this new subject
i m pl ies a spec ific h i storicity close ly tied to ethnology. Drawing on Foucault's
Order of Things ( 1 966) and writing of the scramble for eth nographic artifacts
during the "Museum Age" of the late n i n eteenth centu ry, she cites Douglas Cole's
summation of the preva i l i ng rationale: "It is necessary to use the time to col lect
before it i s too late" (Cole 1 985 : 50). "Too late for what?" Dom i nguez asks.
"There is a h i storical consciousness here of a special sort. We hea r an u rgency i n
the voices o f the col lectors, a fear that w e wi l l no longer b e able t o get o u r hands
on these objects, and that this wou ld amount to an i rretrievable loss of the means
of preserving our own h i storicity. There is a twofold d isplacement here. Objects
are col lected no longer because of their i ntri nsic value but as metonyms for the
people who prod uced the m . And the people who produced them are the objects
of exa m i nation not because of their i ntri nsic va lue but because of their perceived
contribution to our understand i n g of our own h i storica l trajectory. It is a certa in
view of 'man' and a certa i n view of ' h i story' that make this double displacement
possi ble" ( Dom i nguez 1 98 6 : 548) .
O N C O L L E C T I N G A R T A N D C U LT U R E 235

pol ogy owes as m uch to Matthew Arnold a s to i ts official fou n d i n g father,

E . B . Tylor. I ndeed much of the vis ion embod ied i n Culture and Anarchy
has been transferred d i rectly i nto re l ativist anthropol ogy. A powerfu l
structu re of fee l i ng cont i n ues to see c u l t u re, wherever it is fou nd , as a
coherent body that l ives and d ies. Cu lture is enduri ng, trad itiona l , struc­
tu ra l ( rather than conti ngent, syncretic, h i storical ) . Cu lture is a process
of ordering, not of d i sru ption . It changes and develops l i ke a l iv i ng or­
gan is m . It does not norma l l y "s u rvive" abrupt a lterations.
I n the early twentieth centu ry, as culture was bei n g extended to a l l
the world's fu nction i ng soc ieties, an i n c reasi ng n u m ber of exotic, pri m i ­
tive, or archaic obj ects came to b e seen a s "art." They were eq u a l i n
aesthetic a n d moral va l u e with the greatest Western m asterpieces . B y
m idcentu ry the new attitude toward "prim itive art" had been accepted
by l a rge n u m bers of educated E u ropeans and Americans. I ndeed from
the standpo i nt of the late twenti eth century it becomes c lear that the
para l l e l concepts of a rt and c u lture did successfu l ly, a l beit temporari ly,
com prehend and i ncorporate a p l ethora of non-Western artifacts and c us­
toms . This was accompl i shed through two strateg ies. F i rst, obj ects re­
c l assi fied as "pri m itive art" were adm itted to the i magi nary m u seum of
h u m a n c reativity and, though more slowly, to the actual fi ne arts mu­
seu ms of the West. Second, the d i scou rse and i nstituti ons of modern
anthropology constructed comparative and synthetic i m ages of Man
d raw i ng even handed l y from among the world's authentic ways of l ife,
however strange in appearance or obscu re in origi n . Art and c u lture,
categories for the best c reations of Western h u m a n i s m , were i n pri nciple
extended to all the world's peoples.
It i s perhaps worth stress i n g that noth i ng said here about the h i stor­
ic ity of these c u ltural or artistic categories sho u l d be construed as c la i m ­
i n g t h a t t h e y are fa lse or denyi ng that many o f thei r val ues are worthy of
su pport . L i ke any successfu l d i sc u rs i ve arrangement the art-c u l tu re au­
thentic i ty system a rt i c u l ates co nsiderable dom a i n s of truth and scientific
progress as we l l as a reas of b l i n d ness and controversy. By emphas i z i n g
t h e tra n s ience o f t h e system I do s o o u t o f a convi ction (it i s more a
fee l i ng of the h i storical ground mov i n g u nderfoot) th at the c lassifications
and generous appropriations of Western a rt and c u lture categories are
now m u c h l ess stable than before . Th i s i n stabi l ity a ppears to be l i n ked to
the grow i n g i ntercon nection of the world's popu l ations and to the con­
testation s i nce the 1 950s of co lon i a l ism and Eurocentrism. Art co l lecti n g
a n d c u ltu re co l lecti ng now take p l ace with i n a c h a n g i n g field o f coun-

terd iscou rses, syncretisms, and reappropriations origi nating both outside
and i n s ide "the West." I cannot d i scuss the geopol itica l causes of these
deve lopments . I can o n l y h i nt at the i r tra nsform i ng conseq uences and
stress that the modern genea logy of cu lture and a rt that I have been
sketc h i n g i ncreasi ngly appears to be a l oca l story. "Cu lture" and "art"
can no longer be s i m p l y extended to non-Western peop l es and thi ngs.
They can at worst be imposed, at best translated-both h i storica l ly and
po l itica l ly conti ngent operations.
Before I su rvey some of the cu rrent cha l l enges to Western modes of
col lection a nd authentication, it may be worth portraying the sti l l­
dom i nant form of a rt and c u lture col lecting i n a more l i m ited , concrete
setting. The system's underl y i n g h i storical asu m ptions w i l l then become
i nescapable. For if col l ecting i n the West sa l vages thi ngs out of non­
repeatable ti me, what i s the assumed d i rection of th i s time? How does it
confer rarity and a uthenticity on the varied prod uctions of human ski l l ?
Col lecti n g presupposes a story; a story occu rs i n a "chronotope ."

A Chronotope for Collecting

Dans son effort pour comprendre le monde, l 'homme dispose

done toujours d 'un surplus de signification.

THE TERM chronotope, as u sed by Bakht i n , denotes a configu ration of

spatia l and tem poral i n d i cators i n a fictional setting where (and when)
ce rta i n activities and stories ta ke p/ace.9 One can not rea l i stica l l y situate
h i storical deta i l - putt i n g someth i n g " i n its ti me"-without appea l i ng to
exp l i c i t or i m p l icit c h ronotopes . C laude Levi-Strauss's poi nted , nosta lgic
reco l lections of N ew York during the Second World War can serve as a
chronotope for modern art and c u lture col l ecti ng. The setting is e labo­
rated in an essay whose French title, "New York post-et prefigu ratif"

9 . Chronotope: l itera l l y "ti me-space" with no priority to either d i mension

(Bakhti n 1 937). The chronotope is a fictional setting where h i storica l l y spec ific
relations of power become visi ble and certai n stories can "take place" (the bou r­
geois salon in n i neteenth-century social novels, the merchant ship in Con rad's
ta les of adventu re and empire). As Bakhtin puts it: " I n the l i terary artistic chron­
otope, spatial and tempora l indicators are fused into one carefu l l y thought-out,
concrete whole. Time, as it were, th ickens, takes on flesh, becomes artistica l ly
visible; l i kewise space becomes charged and responsive to the movements of
time, plot and h istory" (p. 84) .
O N C O L L E C T I N G A R T A N D C U LT U R E 237

( 1 983), suggests its u nderlying spatia-tempora l pred icament more

strongly than the publ is hed Engl ish translation, " N ew York in 1 94 1 "
( 1 985 ) . The essay fa l l s with i n a m ic rogenre of Levi-Strauss's writi ng, one
he deve loped with v i rtuosity in Tristes tropiques. Specific pl aces- Rio,
F i re Island, new B raz i l i an c ities, I nd i a n sacred s i tes-appear as moments
of inte l l i g i b l e h u man order and transformation su rrou nded by the de­
structive, entropic c u rrents of global h istory.
I n what fol lows I have su pplemented the essay on New York with
passages from other texts written by Levi-Strauss either d u ring the war
years or in reco l lection of them . In read ing them as a u n ified ch ronotope,
one ought to bear in m i nd th at these are not h i storical records but com­
p lex l iterary commemorations. The time-space in question has been ret­
rospective l y composed by Levi-Strauss and recom posed , for other p u r­
poses, by myself.

A refugee in New York d u ri n g the Second World War, the anthropo logist
is bewi ldered and del ighted by a landscape of u nexpected j u xtapos i­
tions. H is recol lections of those sem i n a l years, d u ri ng which he i nvented
structural anthropology, a re bathed in a magical l ight. N ew York is fu l l of
del ightfu l i ncongru ities. Who cou ld resist

the performances that we watched for hours at the C h i nese opera

u nder the fi rst a rch of the Brooklyn B ridge, where a company that h ad
come long ago from C h i na had a large fol l owing. Every day, from m id­
afternoon u nti l past m id n ight, it wou ld perpetuate the trad itions of
c lassical C h i nese opera . I felt myself goi n g back in time no less when
I went to work every morn ing in the American room of the New York
Publ i c Li brary. There, u nder its neo-classical arcades and between
wa l l s paneled with old oak, I sat near an Indian in a feather headd ress
and a beaded buckski n jacket�who was taking notes with a Parker
pen . ( 1 985 : 266)

As Levi-Stra uss te l l s it, the New York of 1 94 1 is an anth ropologi st's

d rea m , a vast selection of h u m a n cu lture and h i story. 1 0 A brief wa l k or
subway ride w i l l take h i m from a G reenwich Vi l l age rem i n iscent of Bal-

1 0 . It sti l l is. Retu rn i ng to the neighborhood where I grew u p on the U pper

West S ide and walking between 1 1 6th and 86th Streets, I i nvariably encou nter
several races, cultures, languages, a range of exotic smel ls, "Cuban-C h i nese"

zac's Pa ris to the towering skysc rapers of Wa l l Street. Tu rn ing a corner i n

th i s j u mb l e of i m m i grants and ethn ic groups, the strol ler sudde n l y enters
a d i fferent world with its own language, customs, c u i s i ne . Everyth i n g i s
avai lable for consu m ption . I n N ew York o n e c a n obtai n a l most a n y trea­
s u re . The anth ropologist and h i s artistic fri ends Andre B reton, Max Ernst,
Andre Masso n , Georges Duthu it, Yves Tanguy, and Matta find master­
pieces of pre-Co l u mbian, I nd ian, Ocean ic, or Japanese art stuffed i n
dea lers' c losets o r apartments. Everything somehow finds i t way here .
For Lev i-Strauss New York i n the 1 940s i s a wonderl and of sudden open­
i ngs to other ti mes and p l aces, of c u ltura l matter out of p l ace :

New York (and this i s the source of its charm and its pecul iar fasci na­
tion) was then a c i ty where anyth i n g seemed poss ible. Like the urban
fabric, the social and c u l tu ral fabric was ridd led with holes. All you
had to do was pick one and s l i p through it if, l i ke Al ice, you wanted
to get to the other s ide of the looki ng gl ass and find worlds so enchant­
ing that they seemed u n real . (p. 2 6 1 )

The anth ropological flaneur is del ighted, amazed , but also troubled by
the chaos of s i m u ltaneous poss i b i l ities . Th i s New York has someth i ng i n
common with the early-centu ry dada-surrea l i st flea market- but with a
d ifference . Its objets trouves are not j ust occasions for reverie. T h i s they
s u re l y are, but they are a l so signs of van i s h i ng worlds. Some are trea­
su res, works of great art.
Levi-Strauss and the refugee su rrea l i sts were passionate col l ectors.
The Th i rd Ave n ue a rt dealer they frequented and advised, J u l i u s earle­
bac h , always had several Northwest Coast, Melanesian, or Eskimo
pieces on hand. Accord i n g to Edmund Carpenter, the su rrea l i sts felt an
i mmed iate affi n i ty with these obj ects' pred i l ection for "visual pu ns" ; thei r
selections were nearly a l ways of a very h igh quality. I n add ition to the
a rt dea lers anothe r sou rce for th i s band of pri m itive-art connoi sseu rs was
the Museum of the American I n d i a n . As Carpenter tel l s it: "The Su rrea l­
i sts began to visit the B ronx warehouse of that Museu m , select i ng for
themse l ves, concentrating on a co l lection of magn ificent Eskimo masks.
These h uge visual puns, made by the Kuskokw i m Eskimo a century or
more ago, constituted the greatest col lection of its kind i n the world . But

restau rants, and so o n . I t i s enough to seriously smudge at least the spatial dis­
ti nction between F i rst and Th i rd Worlds, center and peri phery i n the modern
world syste m .
O N C O L L E C T I N G A R T A N D C U LT U R E 239

t h e M u s e u m D i rector, George Heye, cal led them 'jokes' a n d so ld h a l f for

$ 3 8 and $ 5 4 eac h . The S u rrea l i sts bought the best. Then they moved
happ i l y through Heye's Northwest Coast col lection , stri pping it of one
masterwork after another" (Carpenter 1 9 75 : 1 0) . In 1 946 Max E rnst, Bar­
nett N ewman , and severa l others mou nted an exh i bit of Northwest Coast
I nd ian pa i nt i n g at the Betty Pa rsons G a l lery. They brought together pieces
from the i r private col l ections and a rtifacts from the American Muse u m of
N atu ra l H i story. By moving the m useu m pieces across town, "the S u r­
rea l i sts dec lassified them as scientific spec i mens and rec l assified them as
art" (Carpenter 1 9 7 5 : 1 1 ) .
The category of p r i m itive a rt was emergi ng, with its market, its con­
noisseu rsh i p , and its c l ose ties to modern i st aesthetics. What had begun
w ith the vogue for /'art n egre i n the twenties wou ld become i n stitution­
a l ized by the fifties and s i xties; but i n warti me N ew York the battle to
ga i n widespread recogn ition for tri ba l objects was not yet won . Levi­
Strauss reca l l s that as c u ltura l attache to the French Embassy i n 1 946 he
tried in va i n to a rrange a trad e : for a massive col l ection of American
I nd i a n art a few Matisses and Picassos . But "the French authorities tu rned
a deaf ea r to m y entreaties, and the I nd i a n col l ecti ons wou nd up i n
American m u seums" ( 1 985 : 2 62 ) . T h e co l lecti ng o f Levi -Strauss and the
su rrea l ists d u ri n g the forties was pa rt of a struggle to ga i n aesthetic status
for these i n c reas i ng l y rare masterworks.

N ew York seemed to have someth i n g u n usua l , va l uable, and beautifu l

for everyone. F ranz Boas l i ked to tel l h i s E u ropea n visitors about a Kwa­
k i utl i n formant who had come to work with h i m in the c ity. As Roman
Jakobson reca l ls :

Boas l oved to depict the i n d i ffe rence of this man from Vancouver I s­
land toward Manhattan skyscrapers ( "we bu i l t houses next to one an­
other, and you stack them on top of each other"), toward the Aq uari u m
("we th row s u c h fish back i n t h e lake") or toward t h e motion pictures
which seemed ted ious and senseless. On the other hand, the stranger
stood for hours spe l l bound i n the Ti mes Square freak shows with the i r
giants and dwarfs, bearded ladies and fox-ta i l ed gi rls, or i n t h e Auto­
m ats where d ri n ks and sandwiches appear m i raculously and where he
felt transferred i nto the u n iverse of Kwakiutl fa i ry-ta les. (Jakobson
1 95 9 : 1 42)

I n Levi -Strauss's memory brass ba l l s on sta i rcase ban n i sters a l so figu re i n

the col lection o f fasc i nating phenomena ( 1 960 : 2 7) .

F o r a E u ropean N e w York's sheer space i s vertiginous:

I strode up and down m i les of Man hattan aven ues, those deep chasms
over which loomed skyscrapers' fantastic c l i ffs. I wandered random ly
i nto c ross streets, whose physiognomy changed drastica l ly from one
block to the next : someti mes poverty-stricken , someti mes m iddle-class
or provincial, and most often chaotic. New York was decided l y not the
u ltra-m odern metropol i s I had expected , but an i m mense, horizontal
and vertical d isorder attri butable to some spontaneous upheava l of the
urban crust rather than to the del iberate plans of b u i l ders. (levi­
Strauss 1 98 5 : 258)

Levi -Strauss's New York i s a j uxtapos ition of ancient and recent "strata,"
chaotic rem n ants of former " upheava l s ." As in Tristes tropiques meta­
phors from geo l ogy serve to tran sform e m p i rical s urface i ncongru ities or
fau lts i nto leg i b l e h i story. For Lev i-Strauss the j umble of Man hattan be­
comes i ntel l igible as an over l ay of past and futu re, legible as a story of
cu ltu ra l deve lopment. Old and new are side by side. The E u ropean ref­
ugee encou nters scraps of h i s past as wel l as a trou b l i n g prefigu ration of
common destiny.
N ew York i s a site of trave l and reverie u n l i ke the onei ric c ity of
B reton 's Na dja or Aragon's Paysan de Paris . For Parisian emigres fi n d i n g
the i r feet on its streets and aven ues it is never a known pl ace, someth i n g
to b e made strange b y a certa i n su rrea l i st a n d eth nograph i c attention .
I nstead they are am bushed by the fam i l iar-an o l der Paris i n Greenwich
Vi l lage, gl i m pses of the E u ropean world i n i m m i grant neighborhoods,
med ieva l bu i l d i ngs reassembled at the Cloisters . B ut these rem i nders are
masks, surviva ls, mere col lectibles. I n New York one is permanently
away from home, depayse, both i n space and i n time. Post- and pre­
figu rative N ew York is fa ntastica l l y suspended between a j umble of pasts
and a u n i form future.

Whoever wanted to go h u nting needed only a l i ttle culture, and fla i r,

for doorways to open i n the wal l of ind ustrial civi l ization and revea l
other worlds and other ti mes . Doubtless nowhere more than i n New
York at that time were there such faci l ities for escape. Those poss i bi l i-
O N C O L L E C T I N G A R T A N D C U LT U R E 24 1

ties seem a l most mythical today when we n o longer dare to dream of

doors : at best we may wonder about n i ches to cower i n . But even
these have become the stake in a fierce competition among those who
a re not w i l l i n g to l i ve i n a world without friend l y shadows or secret
shortcuts known only to a few i n i ti ates . Los ing its old d i mensions one
after a nother, this world has pushed us back i nto the one rem a i n i n g
d i mension : one wi l l probe it i n va i n for secret loopholes. ( 1 985 :262)

The resi gned "entropologist" of Tristes tropiques remembers New York as

the fi nal glow and p rophetic d i s i ntegration of a l l rea l c u l t u ral d iffe rences .
Soon even the loopholes w i l l be gone. M i l lennia of h u man d i versity and
i nvention seem to have been s h i pwrecked here, remnants and broken
shards, good to evoke in escap ist reveries, good to col l ect as a rt (or an­
tiques), and "good to th i n k with" in salvaging the c u ltural structu res of a
tra n s h i stori cal esprit humain . The c h ronotope of New York prefigures an­
th ropo logy.
Structu ra l i st anthropology at least was conceived and written there .
It is hard to i magi n e a better setting. Among New York's j u mble of c u l ­
tures, arts, a n d trad itions, a s a professor a t t h e Ecol e Libre des Ha utes
Etudes, Levi -Strauss attended Roman Jakobson's celebrated l ectu res on
sound and mean i ng. On many occasions he h as testified to the i r revol u­
tionary i m pact. J a kobson's demonstration that the bew i lderi ng d iversity
of m ea n i ngfu l h u ma n sou nds cou ld be red u ced to d i screte d i fferentia l
system s through the appl ication o f phonem ic analys i s offered an i m me­
diate model for studyi ng the plethora of h u m a n kinsh i p systems. More
genera l ly J a ko bson's a pproach suggested a research program-that of
d iscoveri ng elementa ry cogn itive structu res beh i nd the many "language­
l i ke" productions of h u man c u lture. Amid the c u ltura l - h i storical j u mble
of wartime New York-too much i n the same pl ace at the same ti me­
Levi-Strauss gl i m psed an u nderlyi ng order.
The Elementary Structures of Kinship was researched i n the New
York Publ i c L i brary read i ng roo m , where, beside what seemed to be a
parody of a feathered I nd i a n with a Parker pen , Levi-Strauss pored over
accou nts of tribal marriage ru les. The fou n d i ng text of structu ral anth ro­
pology was d rafted in a sma l l , d i lapidated stud io in G reenwich Vi l l age,
down the street from Yves Tanguy and a few yards (through the wa l l s)
from C l aude S h a n n o n , who, u n known to h i s neighbor, "was creating
cybernetics" ( 1 985 : 2 60 ) .

U ptown at the American Museum of Natural H i story Levi-Strauss cou ld

wander and wonder among the i ntimate, hyperrea l d ioramas of African
an i ma l spec ies. Or he cou ld marvel in the H a l l of Northwest Coast I nd i ­
a n s , where Kwa k i utl a n d Tl i ngit masks i n thei r glass cases whispered to
h i m of Baudel a i rean correspondances (levi-Strauss 1 943 : 1 80) . I ndeed
by the 1 940s a deep correspondence between prim itive and modern art
was widely assumed i n avant-garde m i l ieux. The anthropologist friend of
the su rrea l ists saw these magica l , archaic objects as l u m i nous exampl es
of human c reative gen i us . He wrote i n 1 943 for the Gazette des beaux
arts :

These objects-bei n gs transformed into thi ngs, human an i mals, l i ving

boxes-seem as remote as possible from our own conception of art
si nce the time of the G reeks. Yet even here one wou ld err to su ppose
that a si ngle poss i b i l ity of the aesthetic l ife had escaped the prophets
and vi rtuosos of the Northwest Coast . Several of those masks and stat­
ues are thoughtfu l portraits which prove a concern to attai n not only
physical resemblance but the most subtle spiritual essence of the sou l .
The scu l ptor of Al aska and British Col u mbia i s not only the sorcerer
who confers u pon the su pernatural a visible form but a l so the creator,
the i nterpreter who translates i nto eternal chefs d'oeuvre the fugitive
emotions of man . ( 1 943 : 1 8 1 )

H uman artistic c reation transcends location and time. To com mu­

n icate the i ncred i b l e i n ventiveness he sees i n the Northwest Coast Ha l l ,
Lev i-Stra uss fi nds a revea l i ng com parison : "Th is i ncessant renovation,
th i s s u reness which i n no matter what d i rection guarantees defi n i te and
overwhel m i ng success, th i s scorn of the beaten path, th i s ceasel ess driv­
i ng toward new feats which i nfa l l i b l y ends i n dazz l i ng res u l ts-to know
th i s our c iv i l ization had to await the exceptiona l destiny of a Picasso . It
i s not futi l e to emphasize that the daring ventu res of a s i ngle man which
have left us breath l ess for th i rty years, were known and practiced d u r­
i n g one h u n d red and fifty years by an entirely i nd igenous culture"
( 1 943 : 1 75) . The passage i s u ndoubted ly adapted to its occasion : the
need to promote tri ba l works for an art-world p u b l i c . ( E l sewhere Levi­
Strauss wou l d stress the systems l i m iti ng and maki ng possi b l e i nventions
by any loca l group or i nd i vidual creator. ) Here he i n s i sts only that tribal
works are as i n ventive as that modern paragon of creat ivity, Picasso . Im­
p l icit i n the conceit was a vision of h u man cu ltures as comparable to
O N C O L L E C T I N G A R T A N D C U LT U R E 243

c reative a rtists . As I have a l ready argued, the twentieth-century cate­

gories of a rt and c u lture presupposed each other.
The categories were, however, i nstitutiona l l y separated . If the sur­
rea l i sts cou ld rec l assify tribal objects by mov i ng them across town from
an anth ropo logy m useu m to an art ga l l ery, the end poi nts of the traffic
were not thereby u nderm i ned . The d i scou rses of anth ropol ogy and art
were deve l o p i n g on separate but complementary path s . Thei r evo l v i n g
relationsh i p may be seen i n a legendary su rrea l ist j o u r n a l o f 1 942-43
ed ited by David H a re and dom i n ated by i ts "ed i tori a l advisers" And re
B reton, Max Ernst, and Marcel Duchamp. VV\1, accord ing to i ts subtitle,
aspi red to cover the fields of " poetry, p l asti c arts, anth ropo logy, sociol­
ogy, psych o logy." I n fact it d i d j u stice to the fi rst two, with a spri n k l i ng of
the th i rd . (On ly fou r issues of VVV appeared i n two years . ) N u mber 1
contai ned two short artic l es by Levi-Strauss, one on Kad uveo I n d i a n face
pa i nting, the other an obituary for Ma l i nowski . The fol lowi ng n u m ber
conta i n ed a note by Alfred Metraux on two ancestra l figu ri nes from Easter
Is land . And in the fi n a l i ssue Robert A l l erton Pa rker fancifu l l y i nterpreted
com p l ex l i ne d rawi ngs from the New Hebrides (extracted from A. B .
Deacon 's eth nography) u nder the title "Ca n n i ba l Desi gns." I n general
materi a l from non-Western c u ltures was i n c l uded as exoticism or naive
art . There were occas ional photos of an A l askan mask or a kach i na .
I n VVV anth ropo l ogy was part o f t h e decor o f avant-garde art and
writi ng. Serious cu ltura l a n a l ysis made no rea l i n roads i n to what were by
now canon ical su rrea l i st notions of gen i us, i ns p i ration, the i rration a l , the
magi cal, the exotic, the pri m itive . Few of those around B reton (with the
possi ble exception of Max Ernst) had any systematic i n terest in eth no­
logical sc ience. Levi-Strauss's contri butions to VVV seem out of p l ace.
Essenti a l l y a jou rn a l of art and l i terature, VVV was preoccu p i ed with
d ream s , arc h etypes, gen i us, and apoca l ypti c revol ution . It engaged in
l i ttle of the u nsettl i ng, reflex ive eth nography practi ced by the d i ssidents
of the earl ier jou rna l Documents (see Chapter 4) . "Ma i n strea m" su rreal­
ism did not typ i ca l l y bri ng c u ltural analysis to bear on its own categories.
Su rrea l ist a rt and structu ral anthropology were both concerned with
the h u man s p i rit's "deep" shared spri ngs of creativity. The common a i m
was to tra nscend-not, a s i n Documents, t o describe critica l l y or sub­
vert-the loca l o rders of c u lture and h istory. S u rrea l i s m 's subject was an
i nternational and e lementa l h u man ity "anth ropologica l " i n scope. I ts ob­
ject was Man, someth i ng it shared with an emergi ng structura l i s m . But a
conventi o n a l d i vision of l abor was so l id ifyi ng. With i n the project of

prob i ng and extend i ng humanity's creative esprit, the two methods d i ­

verged , o n e p l a y i n g art t o t h e other's science.

Modern practices of art and cu lture col lecti ng, scientific and avant­
ga rde, have situated themselves at the end of a global h istory. They have
occ u p i ed a pl ace-apocalyptic, progressive, revol utiona ry, or tragic­
from which to gather the val ued i n heritances of Man . Concretizing th i s
tem pora l setup, Levi-Strauss's "post- and prefigu rative" N e w York antici­
pates h u m a n ity's entropi c future and gathers up its d i verse pasts i n de­
contextu a l ized , col l ecti b l e form s . The eth n i c neighborhoods, the prov i n­
c i a l rem i nders, the C h i nese Opera Company, the feathered I ndian i n the
l i brary, the works of art from other conti nents and eras that turn up i n
dea lers' c l osets : a l l are s u rvivals, rem nants o f th reatened o r va nis hed tra­
d itions. The world's c u l tu res appear in the c h ronotope as sh reds of hu­
ma n i ty, degraded commod i ties, or e levated great art but a l ways fu nction­
i ng as va n i s h i ng " l oopho l es" or "esca pes" from a one-d i mensional fate .
I n New York a j u mb l e of h u man ity has washed up i n one vertigi nous
place and t i me, to be grasped s i m u ltaneously in al l its prec ious d i versity
and emergi ng u n i form ity. I n th i s c h ronotope the pure products of h u man­
ity's pasts are rescued by modern aestheti cs only as subl i mated art. They
are sa lvaged by modern anthropo logy as consu ltable archives for th i n k­
i n g about the range of h u man i nvention . I n Levi -Strauss's setting the
prod ucts of the present-becom i ng-future are shal l ow, i m p u re, escapi st,
and " retro" rather than tru ly d ifferent- "antiq ues" rather than genu i ne
antiqu ities . C u l t u ral i nvention is subsumed by a com mod ified "mass c u l ­
tu re" ( 1 985 : 2 64 -2 6 7) .
The c hronotope o f New York su pports a gl obal a l legory of fragmen­
tati on and ru i n . The modern anthropologist, lamenti ng the pass i ng of
h u man d ivers ity, col l ects and val ues its surviva l s, its enduring works of
art . Levi -Strauss's most prized acq u i sition from a marvelous New York
where everyth i ng seemed ava i l able was a nearly complete set of vol u mes
1 through 48 of the Annual Reports of the B u reau of American Ethnology.
These were, he tel l s us i n another evocation of the war yea rs, "sacrosanct
vol u mes, representing most of our knowledge about the American In di­
ans . . . It was a s though the American Ind ian c u ltu res h a d sudden l y
come a l ive a n d become a l m ost tangi ble through t h e physical contact that
these books, written and publ ished before these cultures' defin ite exti nc­
tion, establ i shed between the i r ti mes and me" ( Levi-Stra uss 1 9 76 : 50) .
O N C O L L E C T I N G A R T A N D C U LT U R E 245

These prec ious records of h u man d iversity h ad been recorded by an eth­

nology sti l l in what he cal l s its " p u re" rather than "d i l uted" state ( Levi­
Strauss 1 960 : 2 6 ) . They wou ld form the authentic eth nograp h i c materi a l
from w h i c h structu ra l i sm's metacu ltural orders were constructed .
Anthropo logical col l ections and taxonom ies, however, are con­
stantly menaced by tempora l conti n gencies. Levi -Strauss knows th i s . It is
a d i sorder h e a l ways holds at bay. For example in Tristes tropiques he is
acute l y awa re that foc u s i n g on a tribal past necessari l y b l i nds him to an
emergent present. Wandering through the modern landscape of New
York, far from encou ntering less and l ess to know, the anth ropologist i s
confronted with more and mare-a heady m i x-and-match o f possi b l e
h uman com bi nations. H e struggles t o mainta i n a u n ified perspective; he
looks for order in d eep "geo logica l " structu res . But in Levi -Strauss's work
genera l l y, the englob i n g "entropo logica l " narrative barely conta i n s a c u r­
rent h i story of loss, tran sformation, i nvention, and emergence.
Toward the end of his bri l l iant i naugu ra l lectu re at the Col l ege de
France, "The Scope of Anthropology," Lev i -Strauss evokes what he ca l l s
"ant h ropo l ogical dou bt," the i n evitab l e resu lt of eth nographic risk­
tak i ng, the "buffeti n gs and den i a l s d i rected at one's most cherished ideas
and habits by othe r i deas and habits best able to rebut them" ( 1 960 : 2 6 ) .
H e po ignantly reca l ls Boas's Kwaki utl v i s i tor, transfixed b y t h e freaks a n d
automats o f Ti mes Square, and h e wonders whether anthropology may
not be condem ned to eq ua l ly bizarre perceptions of the d i stant soc ieties
and h i stories it se�ks to grasp. New York was perhaps Levi-Strauss's on l y
t r u e "fieldwork" : for once he stayed l o n g enough a n d m astered t h e l ocal
language . Aspects of the p l ace, such as Boas's Kwakiutl, h ave cont i nu ed
to charm and haunt h i s anth ropo l ogical c u lture co l l ecting.
But one N ew York native s its with spec i a l d i scomfort i n the c h rono­
tope of 1 94 1 . Th i s i s the feathered I n d i a n with the Parker pen work i n g i n
the Pu b l i c L i b ra ry. For Levi -Strauss the I n d i a n i s primari ly assoc i ated with
the past, the "extinct" soc ieties recorded i n the prec ious B u reau of Amer­
ican Eth nology Annual Reports . The anthropologist feel s h i mse l f "go i n g
back i n ti me" ( 1 985 : 2 66). I n modern N ew York an I nd i a n c a n appear
on l y as a s u rvival or a k i nd of i ncongruous parody.
Another h i storica l vision m i ght h ave pos itioned the two schol a rs i n
the l i brary d i ffe rently. The decade j u st preced i n g Levi-Strauss's arriva l i n
N ew York had seen a d ramatic turnaround i n federal pol icy. U nder John
Co l l ie r's leadersh i p at the B u reau of I n d ian Affa i rs a " N ew Indian Po l i cy"
actively encou raged tribal reorgan i zation a l l over the cou ntry. Wh i le

Levi-Strauss stud ied and col lected their pasts, many "exti nct" Native
American groups were i n the pmcess of reconstituting themselves c u ltur­
a l l y and pol itical ly. Seen i n th i s context, d i d the I nd ian with the Parker
pen represent a "go i ng back in ti me" or a g l i m pse of another futu re ? That
i s a different story. (See Chapter 1 2 . )

Other Appropriations

To te l l these other stories, loca l h i stories of c u ltural s u rvival and

emergence, we need to resist deep-seated habits of m i nd and systems of
authentic ity. We need to be suspicious of an a l most-a utomatic tendency
to re legate non-Western peoples and objects to the pasts of an increas­
ingly homogeneous h uman ity. A few examples of cu rrent i nvention and
contestation may suggest d i fferent ch ronotopes for art and cu ltu re col­
Anne Vita rt-Fardou l is, a cu rator at the Musee de I ' Homme, has pub­
l i shed a sensitive account of the aesthetic, h i storica l , and cu ltura l d i s­
cou rses routinely used to exp l icate i nd ividual museum objects. She d i s­
cusses a fa mous i ntricate l y pa i nted a n i m a l skin ( its present name : M . H .
34 . 3 3 . 5 ), probably origi nati ng among the Fox Indians of North America.
The ski n tu rned u p i n Western co l lecting systems some time ago in a
"cabi net of curiosities " ; it was used to educate aristocratic c h i ldren and
was much adm i red for its aesthetic q u a l ities. Vitart-Fardou l is te l l s us that
now the s k i n can be decoded ethnographica l l y in terms of its com bi ned
"mascu l i ne" and "fe m i n i ne" graphic sty les and u nderstood in the context
of a probable rol e in spec ific ceremon ies. But the mea n i ngfu l contexts
are not exhausted . The story takes a new tu rn :
The grandson of one of the Indians who came to Paris with Buffalo B i l l
was searc h i n g for the [pai nted ski n] tu n i c h i s grandfather had been
forced to sel l to pay h i s way back to the U n ited States when the circus
co l lapsed . I showed him a l l the tunics i n our col l ection, and he paused
before one of them . Contro l l i ng h i s emotion, he spoke. He told the
mean i n g of th i s lock of hair, of that design, why this color had been
used , the mean i n g of that feather . . . Th is garment, formerly beautifu l
and i nteresting but passive and i nd i fferent, l ittle by l ittle became mean­
i ngfu l , active testimony to a l iving moment through the mediation of
someone who d i d not observe and analyze but who l ived the object
and for whom the object l ived . It scarcely matters whether the tunic is
rea l l y h i s grandfather's . (Vitart-Fardou l i s 1 986: 1 2)
O N C O L L E C T I N G A R T A N D C U LT U R E 247

Wh atever is happe n i ng in t h i s encounter, two th i n gs are c l early not

happen i ng. The grandson is not rep l ac i ng the obj ect in its origi n a l or
"authentic" c u ltural context. That i s long past. H i s encou nter with the
pa i nted s k i n is part of a modern reco l l ection . And the pa inted tu n i c is
not be i n g appreci ated as a rt, as an aesthetic object. The encou nter is too
spec ific, too enmeshed i n fam i ly h i story and eth n i c memory. 1 1 Some as­
pects of "cu ltura l " and "aestheti c" appropriation are certa i n l y at work,
but they occ u r with i n a current tribal h istory, a different tem pora l ity from
that govern i n g the domi nant systems I d i agrammed earl ier. I n the context
of a present-becom i ng-future the old pai nted tu n ic becomes newly, tra­
d ition a l l y mea n i ngfu l .
The c u rrency of "tr i ba l " artifacts i s beco m i n g m ore visible to non­
I nd ians. Many new tribal recogn ition c l a i m s are pend i n g at the Depart­
ment of the I nterior. And whether or not they are form a l l y successfu l
m atters less than what they make manifest : the h i storical and pol itical
rea l ity of I n d i a n s u rvival and resu rgence, a force that i m p i nges o n West­
ern art and c u ltu re col lections. The " proper" place of many objects i n
m useum s is now su bject to contest. The Zu n i who prevented the loan of
the i r war god to the Museum of Modern Art (see Chapter 9) were c h a l ­
lenging the dom i nant a rt-c u ltu re system, for i n trad itional Z u n i bel ief war
god figu res a re sacred and dangerous . They a re not ethnographic a rti­
facts, and they a re certa i n l y not "art." Zuni c l a i m s on these obj ects s pe­
cifica l l y reject the i r "promotion" ( i n a l l senses of the term) to the status
of aesthetic or scientific treasu res.
I wou ld not c l a i m that the only true home for the objects i n question

1 1 . I n his wide-ranging study "Eth n i city and the Post-Modern Arts of Mem­
ory" ( 1 986) Michael F i scher ident ifies general processes of cultural rei nvention,
personal search, and futu re-oriented appropriations of trad itio n . The specificity
of some Native American relations with col lected "triba l" objects is revea led i n
a grant proposa l to the National E ndowment for the Human ities by the Oregon
Art I nstitute (Monroe 1 986) . In preparation for a rei n stal l ation of the Rasmussen
Col lection of Northwest Coast works at the Portland Art Museum a series of con­
su ltations i s envisioned with the pa rticipation of Haida and Tl i ngit e lders from
Alaska. The proposal makes c lear that great care m ust be given "to matching
spec i fic groups of objects i n the col lection to the clan members h i p and knowl­
edge base of spec ific elders. Northwest Coast Natives belong to specific clans
who have extensive oral trad itions and h i stories over which they have owner­
s h i p . Elders are responsible for representing their clans as wel l as thei r group"
(Mon roe 1 98 6 : 8 ) . The rei nsta l l ation "wi l l present both the academic interpreta­
tion of an object or objects and the interpretation of the same material as viewed
and understood by Native elders and artists" (p. 5 ; original emphasis) .

i s i n "the tri be"-a locati on that, i n many cases, i s fa r from obvious. My

poi nt is j u st that the dom i nant, i nterloc k i ng contexts of art and anthro­
pology a re no longer self-evident and u ncontested . There are other con­
texts, h i stories, and futures in which non-Western objects and cu ltura l
records m a y " belong." T h e rare Maori artifacts that i n 1 984 -85 tou red
m u seum s in the U n ited States norm a l l y res ide in New Zeal and museums.
But they are contro l led by the trad itional Maori authorities, whose per­
m i ss ion was req u i red for them to l eave the country. Here and e lsewhere
the c i rc u lation of m useu m col lections i s significantly i nfl uenced by re­
su rgent i nd i genous comm u n ities.
What i s at stake i s someth i ng more than conventional museum pro­
grams of com m u n ity ed u cation and "outreach " (Alexander 1 9 79 : 2 1 5 ) .
C urrent deve lopments question the very status of m useu ms a s h i storica l ­
cu ltura l theaters o f memory. Whose memory? F o r what pu rposes ? The
Provi n c i a l Museum of British Col umbia has for some time enco u raged
Kwaki utl ca rvers to work from models in its col lecti o n . It has lent out old
pieces and donated new ones for use i n modern potlatches. Surveying
these developments, M i c hael Ames, who d i rects the U n i versity of B ritish
Col u mbia Museu m , observes that " I nd ians, trad itiona l l y treated by mu­
seums only as objects and c l ients, add now the rol e of patrons." He con­
tinues : "The n ext step has a l so occu rred . I ndian com m u n ities establ ish
their own m useu ms , seek the i r own N ationa l Museum grants, i nsta l l the i r
own cu rators, h i re the i r o w n anthropol ogists on contract, a n d ca l l for
repatri ation of the i r own co l l ections" (Ames 1 986 : 5 7 ) . The Quad ra I s­
land Kwaki utl Muse u m l ocated i n Quath raski Cove, British Co l u mbia,
d isplays triba l work returned from the national col l ections in Ottowa.
The obj ects are exh i bited i n gl ass cases, but arranged accord i ng to the i r
ori g i n a l fam i l y owners h i p . I n Alert Bay, British Co l u m bia, t h e U ' m i sta
C u l tu ral Centre d i sp l ays repatriated artifacts i n a trad itiona l Kwaki utl "big
house" arranged i n the seq uence of the i r appeara nce at the potlatch cere­
mony. The new i n stitutions fu nction both as publ ic exh i bits and as c u l ­
tural centers l i n ked t o ongo i n g tribal trad itions. Two Ha ida m u seums
have also been establ i shed i n the Queen Charlotte I s l ands, and the
movement i s grow i n g e lsewhere in Canada and the U n ited States .
Resou rcefu l N ative American groups may yet appropri ate the West­
ern m u seu m-as they h ave made thei r own another E u ropean i nstitu­
tion, the "tribe ." Old objects may aga i n partici pate in a triba l present­
beco m i n g-futu re . Moreover, it is worth briefly noting that the same th i ng
is poss ible for written artifacts co l lected by salvage eth nography. Some
O N C O L L E C T I N G A R T A N D C U LT U R E 249

of these old texts (myths, l i ngu i stic sam p l es, lore of a l l ki nds) are now
bei n g recyc l ed as l oca l h i story and triba l " l iteratu re." 1 2 The objects of
both a rt and c u lture col lecti ng are susceptible to other appropriations.
This d i stu rbance of Western object systems is reflected i n a recent
book by Ra l p h Coe, Lost and Found Traditions : Native American Art:
1 965-1 985 ( 1 986) . (On i n ventive tribal work see a l so Macnai r, Hoover,
and Neary 1 984; Ste i n bright 1 98 6 ; Babcock, Montha n , and Monthan
1 986). Coe's work i s a col lector's tou r de force. Once aga i n a w h i te au­
thority "d i scovers" true tribal a rt- but with sign ificant d ifferences. H u n­
d reds of photographs document very recent works, some m ade for loca l
use, some for sale to I ndians or wh ite outs iders. Beautifu l obj ects-many
formerly c l assified as c u rios, fo l k a rt, or tou rist art-a re located in on­
goi ng, i nventive trad itions. Coe effecti ve l y q uestions the widespread as­
s u m ption that fi ne tri bal work is d i sappearing, and he th rows doubt on
common c riteria for j udging pu rity and authenticity. In his col lection
among recogn izably trad itional kac h i nas, totem pol es, b l a n kets, and
p l a i ted baskets we find ski l l fu l ly beaded ten n i s shoes and baseba l l caps,
articles deve l o ped for the c u rio trade, q u i lts, and decorated leather cases
(peyote kits mod e l ed on old-fash ioned too l boxes).
Si nce the N ative American C h u rc h , i n whose ceremonies the peyote
kits are used, d i d not exist in the n i n eteenth centu ry, the i r c l a i m to tra­
d ition a l status can not be based on age . A stronger h i storical c l a i m can i n
fact b e made for many prod uctions o f t h e c u rio trade, such a s the beaded
"fancies" (hanging b i rds, m i rror frames) made by Mati lda H i l l , a Tusca­
rora who sel l s at N iaga ra Fal ls :

"j u st try tel l i ng Mati lda H i l l that her 'fancies' (cat. no. 46) are tourist
cu rios," said Mohawk Rick H i l l , author of an unpubl ished paper on
the subject. "The Tuscarora have been able to trade pieces l i ke that
b i rd or beaded frame (cat. no. 47) at N iagara s i nce the end of the War

1 2 . The arch ives of james Wa l ker, produced before 1 9 1 0, h ave become

relevant to the teach i n g of local h i story by S ioux on the Pine R idge Reservation
(see Chapter 1 , n. 1 5 , and C l i fford 1 986a : 1 5 -1 7). Also a corpus of transl ated and
u ntrans lated Tolowa tales and l i ngu istic texts collected by A . L. Kroeber and P. E .
Goddard are i m portan t evidence i n a plan ned petition for tribal recogni tion . The
texts were gathered as "salvage eth nography" to record the shreds of a pu rport­
ed ly van i s h i n g c u lture . But in the context of To Iowa persi stence, retranslated and
interpreted by Tolowa e l ders and their Native American l awyer, the texts yield
evidence of tribal h i story, territorial l i m its, group d i sti nctness, and oral trad ition .
They are Tolowa " l i teratu re" (Slagle 1 986) .

of 1 8 1 2 , when they were granted excl usive rights, and she would n't
take kindly to anyone sl ighti ng her cultu re ! "

"Sure l y," Coe adds, "a trade privi lege establ i shed at N iagara Fal l s i n 1 8 1 6
sho u l d be acceptable as trad ition by now" ( 1 986 : 1 7) . He drives the gen­
era l poi nt home1 3 : "Another m i sconception derives from our fai l u re to
recogn ize that Ind ians h ave a l ways traded both with i n and outside the i r
c u l tu re ; it is second natu re t o t h e way they operate i n a l l thi ngs . Many
objects are, and a lways have bee n , created in the Indian world without
a spec ific desti nation in m i n d . The h istory of I n d i a n trad i n g predates any
white i nfl uence, and trad ing conti n ues today u nabated . It is a fasc i nating
instrument of soc i a l cont i n u ity, and i n these modern ti mes its scope has
been greatl y en l a rged " (p. 1 6) .
Coe does not hesitate to com miss ion new "trad itiona l " works, and
he spends considerabl e time el iciti ng the spec ific mea n i ng of obj ects
both as i nd ividual possess ions and as tri ba l art. We see and hear particu­
l a r a rtists ; the coexi stence of spi ritu a l , aesthetic, and commerc i a l forces
is always v i s i b l e . Overa l l Coe's col lecting project represents and advo­
cates ongo i n g art forms that are both re l ated to and separate from domi­
nant systems of aesthetic-ethnographic va l u e . I n Lost and Found Tradi­
tions a uthenticity is someth i n g prod uced , not salvaged . Coe's co l lecti on,
for all its love of the past, gathers futures.
A long c h apter on "trad ition" resi sts sum mary, for the d iverse state­
ments quoted from practice a rtists, old and young, do not reprod uce
preva i l ing Western defi n itions. "Wh ites th i n k of our experience as the
past," says one of a group of students d i scussing the topic. "We know it
is right here with us" (p. 49) .

1 3 . The common presumption that tribal art is essentially noncommercial

("sacred ," "spiritu a l ," "envi ronmenta l ," and so on) is of q uestionable va l ue every­
where. A revea l i ng case is the New Gu inea Sepik region, where customary ob­
jects and lore have long been traded, bought, and sol d . To a sign i ficant degree
the i nvolvement of loca l grou ps in the a rt markets of a wider world can be "tra­
d itional ." I ndigenous com mod i ty systems interact with outside capita l ist forces;
they do not s i m p l y give way to the m . The world system is thus dynamically and
loca l l y organ ized . A persistent tendency to see non-Western societies as lacking
h i storical agency i s corrected by a growing number of academ i c stud ies; for ex­
ample Rosaldo 1 980, R. Price 1 983, and Sah l i ns 1 985. These works u nderm i ne
the bi nary ("Orienta l i st") division of h u man groups i nto h i storical and myth ica l ,
"hot" a n d "cold," d iachronic a n d synchronic, modern a n d archaic. Sa l l y Price
( 1 986) d raws attention to the d iverse historica l visions of non-Western, "triba l "
peoples a n d t o t h e role o f art i n articulating these visions.
O N C O L L E C T I N G A R T A N D C U LT U R E 251

"We always begi n our sum mer dances with a song that repeats only
four words, over and over. They don't mean much of anyth i ng i n En­
glish, 'young c h i efs sta nd up.' To us those words demonstrate our pride
in our l i neage and our happiness i n always rememberi ng it. It i s a
happy song. Trad ition is not someth i n g you gab about . . . It's i n the
doi n g ." (p. 46)

"Yo u r trad ition i s 'there' always. You're flexible enough to make

of it what you want. I t's always with you . I pray to the old pots at the
ru i n s and d ream about making pottery. I tel l them I want to learn it.
We l ive for today, but never forget the past." (p. 47)

"Our job as artists i s to go beyond, w h i ch i m p l ies a love of

change, [a lways accom pl ished with] trad itions i n m i nd, by tal king to
the elders of the tribe and by being with you r grandparents. The stories
they tel l are j ust amaz i ng. When you become ex posed to them, every­
th i n g becomes a reflection of those events. There's a great deal of sat­
isfaction bei n g an a rtist of trad itions." (p. 47)

"We've always had charms : everyth i n g that's new is old with

us." (p. 79)
Inside a Hopi kiva.
Part Four � Histories
There are many different kinds of Palestinian experience,
which cannot all be assembled into one. One would therefore
have to write parallel histories of the communities in
Lebanon, the occupied territories, and so on. That is the
central problem. It is almost impossible to imagine a single
narrative: it would have to be the kind of crazy history that
comes out in Midnight's Children, with all those little strands
coming in and out.

1 1 . On Orienta/ism

IN 1 93 9 Aime Cesa i re pu b l i shed h i s seari ng long poem "Cah ier d ' u n

retou r au pays n ata l ." I n it he wrote o f h i s native Marti n iq ue, o f co l o n i a l
oppression ; o f red iscovered African sou rces; he coi ned t h e te rm negri­
tude. H is poem was written i n the langu age of Lautreamont and R i m ­
b a u d , b u t it was a F rench spattered w i t h neologisms, pu nctu ated b y new
rhyth ms. For Cesa i re a " n ative land" was something com plex and hybrid,
salvaged from a lost origi n , con structed out of a sq ual id present, arti cu­
l ated with i n and aga i n st a co l o n i a l tongue.
By the ea rly 1 95 0s the negritude movement was in fu l l swi n g, thrust­
i n g an a l ternative h u ma n ism back at E u rope; and i n this new context it
became possi b le to q uestion E u ropea n ideologica l practi ces i n rad i ca l
ways. M ichel Lei ri s , who was a friend and col l aborator of Cesai re's,
com posed the fi rst exten ded analys i s of the rel ationsh i p between an­
thropo l ogical knowledge and colon i a l ism ( Le i ris 1 950). H is d i scou rse
opened a debate that has conti n ued , with varying degrees of i ntensity,
d u ri ng the su bseq uent decades . How h as E u ropean knowledge about the
rest of the p l anet been shaped by a Western w i l l to power? How have
Western writers, both i magi n ative and scientific, been en meshed in co-


lonial and neocolon i a l s ituations? How, concrete ly, have they ignored ,
res isted , and acq u iesced i n these end u r i ng cond itions of inequal ity?
Lei ris po i nted to a basic i m ba l ance. Westerners had for centu ries studied
and spoken for the rest of the wor l d ; the reverse had not been the case.
He annou nced a new situation, one in which the "objects" of observation
wou ld begin to write back . The Western gaze wou l d be met and scat­
tered . S i nce 1 950 Asians, Africans, Arab orientals, Pacific islanders, and
Native Americans have in a variety of ways asserted the i r independence
from Western cu ltura l and po l itical hegemony and establ ished a new
m u l tivocal fie ld of i nterc u l tural d i scou rse . What w i l l be the long-term
conseq uences of such a s ituation- i f it end u res ? How has it al ready al­
tered what one can know about others, the ways such knowledge may
be form u l ated ? It i s sti l l early to j udge the depth and extent of the epi ste­
mologica l changes that may be u nder way. (The l iterature on anth ropol­
ogy and colon ial ism is q u ite large . A few i m portant works are Maquet
1 964 ; Hymes 1 969; Asad 1 9 7 3 ; F i rth 1 9 7 7 ; Copans 1 9 74, 1 975 ; Lec lerc
1 97 2 ; and N ash 1 9 75 . I n the field of Orienta l and I s l a m i c stud ies see
Ti bawi 1 96 3 ; Abdei-Malek 1 96 3 ; Houran i 1 96 7 ; and Khatibi 1 9 7 6 . )

Edward Sa id's Orienta /ism ( 1 978a), a critical study o f Western knowledge

about the exotic, occu pies th is indeterm i nate h i storical context. If it pre­
sents itse lf as part of the genera l "writi ng back" agai nst the West that
Lei r i s a n nou nced , Orien ta/ism's pred icament is an ambiguous one that
shou l d be seen not in terms of a simple anti-i m peri a l ism but rather as a
symptom of the unce rta i nties generated by the new global situation . It is
i m porta nt to situate Sa id's book with i n this wide perspective, for it wou l d
b e a l l too easy t o d ismiss Orienta/ism as a narrow polemic dom i n ated by
i m med i ate ideo l ogical goa l s in the Midd le East struggle. It cou l d be seen
too as merely the personal protest of a Palesti nian deprived of h i s home­
land by a " u n iquely pu n i s h i ng desti ny," sufferi ng from his extern a l l y i m­
posed , abstract identity as "an Orienta l ," oppressed by "an a l most u nan­
i mous consensus that pol itica l l y he does not exist" (pp. 26-2 7) . I ndeed
Sa id writes forthrightly and e loq uently of this, h i s own pred icament; and
he writes a l so from a conviction that "pure" scholars h i p does not ex ist.
-r Knowledge in his view is i nextricably tied to power. When it becomes
i institutiona l ized, c u l t u ra l l y accu m u l ated , overly restrictive in its defi n i -
1 ' tions, it m ust b e actively opposed b y a cou nterknowl edge. Orienta/ism
is pole m i ca l , its analysis corrosive; but Said's book operates in a number
ON O R I E N TA L I S M 257

o f regi sters, a n d i t wou l d b e wrong t o restrict its significance u n d u ly.

Orienta/ism is at once a serious exerc i se i n textual criti cism and, most
fu ndamenta l l y, a series of i m portant if tentative epistemologica l reflec­
tions on genera l styles and proced u res of c u ltural d i scou rse .
Sa id's top i c is u s ua l l y thought of as a rather old-fash ioned scholarly
d i sc i p l i ne a l l i ed with n i neteenth-centu ry p h i l o logy and concerned with
the col lection and analysis of texts i n Eastern l a nguages . Raymond
Schwab's encycloped i c R e n a issan ce orienta le ( 1 9 5 0) i s of cou rse the
c l assic h istory of t h i s ensemble, w h i c h i n c l uded S i nologists, l s l a m i c i sts,

l ndo- E u ropea n i sts, l i te rat i , travelers, and an ec lectic host of afi c i onados.
Said does not atte m pt to revi se or extend Schwab's work, for h i s ap-
proach is not h i stor i c i st or e m p i rical but ded uctive and constructivist. H is
study u ndertakes a s i m u l ta neous expan sion and forma l ization of the
fie l d , transfo rm i ng Orienta l i s m i nto a synecdoch e for a much more com­
plex and ram i fi ed tota l ity. Said ca l l s t h i s total ity a "discou rse," fol lowi n g
Fouca u lt. I sha l l d iscuss Said's adoption o f a Fouca u l d ian methodol ogy
and its hazard s . For the moment, though, it is enough to say that the
Orienta l i st "discou rse" i s c h a racterized by an oppress ive systematicity, a
"sheer knitted-together strength " (p. 6) that Said sets out to reveal through
a read i ng of representati ve texts and ex periences.
Although Sa id d iscovers "Orienta l ism" i n Homer, Aeschylus, the
Chanson de Roland, and Dante, he situ ates its modern origi ns i n Barthe­
lemy d' Herblot's Bibliotheque orienta le. T h i s com pend i u m of orienta l
knowledge is criticized by Said for i ts cosmological scope and for its
construction as a "systematic" and " rationa l " oriental panorama. It is sig­
n i ficant that Said's read i n g of H e rb l ot's seventeenth-century work makes
n o attem pt to a n a l yze it as Fouca u lt wou ld in Les mots et les chases­
that is, "arc h aeo l ogica l ly" - i n relation to a syn chro n i c epi stemo logica l
fie l d . The approach of Orienta/ism i s thus c l early i n d icated as genea log­
i ca l . Its centra l task is to descri be retrospective l y and cont i n uo u s l y the
structu res of an Ori enta l i s m that ach ieved its c l ass ical form in the n i ne­
teenth and early twentieth centu ries. Sa i d 's two criticisms of Herblot are
constitutive of h is object : Oriental i s m i s a l ways too broad l y and ab­
stractly pitched, and it i s a l ways overly systematic.
Sa id p roceeds to apply these reproaches, with varying degrees of
p l a u s i b i l ity, to a d i ve rse ra nge of a uthors, i nstitutions, and typical expe­
riences. There a re a n a l yses of Sylvestre de Sacy, E rnest Ren an and the
N a poleo n i c exped ition to Egypt's scho larly prod uct, the mass ive De­
scription de I'Egypte. The s peeches of po l iticians such as Balfo u r and

Cromer (j uxtaposed with Henry Kissi nger); the I nd ian journa l ism of
Marx; the orienta l voyages of Chateaubriand, Lamarti ne, N erval , and
F l a u bert; the adventu res of B u rton and Lawrence; the scholars h i p of
H. A. R . G i bb and Lou is Mass ignon are a l l woven i nto an i ntertextual
u n ity. This ensembl e-though it leaves some room for h i storica l muta­
tion, d i fferent national trad itions, personal id iosyncrasies, and the gen i u s
o f "great" write rs- is desi gned to emphasize the systematic and i nvariant
nature of the Orienta l i st d i scourse. There is no way to summarize the
complex i nterweavi n gs of Said's critical method-assoc iative, some­
ti mes bri l l i a nt, someti mes forced , and in the end n u m bingly repetitive . It
succeeds at l east in isolati ng and d i scred iting an array of "orienta l "
stereotypes : t h e eternal a nd u nchanging East, the sex ua l l y i n sati able
Arab, the "fem i n i ne" exotic, the teem i ng marketplace, corrupt despo­
tism, mystical rel igios ity. Said is particu l arly effective in h i s critica l anal­
ys is of Orienta l i st "authority" -the paterna l i st privi leges u n hesitati ngly
assumed by Western writers who "speak for" a m ute Orient or reconsti ­
tute its decayed or d i smembered "truth," who lament the passing of its
authenticity, and who know more than its mere natives ever can . Th is
method ical suspicion of the reconstitutive proced u res of writing about
others cou ld be usefu l l y extended beyond Orienta l ism to anthropo logica l
practice genera l ly.
If Oriental ism, as Said descri bes it, has a structu re, th i s resides in its
tendency to dichotom ize the h u man conti n u u m i nto we-they contrasts
and to essentialize the resu ltant "other" -to speak of the oriental m i nd ,
for exa m ple, or even t o genera l ize about " I s l am" or "the Arabs." A l l of
these Orienta l i st "visions" and "textua l i zations," as Said terms them,
fu nction to s u ppress a n authentic " h uman" rea l ity. This rea l i ty, he im­
p l ies, is rooted i n ora l encou nter and rec i procal speech , as opposed to
the processes of writing or of the visual i m agi nation . Said's l i m ited po­
lemical goa l is we l l served by such an anal ysis. "Authentic" human en­
cou nter can be portrayed as subj u gated to the dead book. (Fiaubert does
not, for example, rea l l y experience Egypt as much as he recopies a pas­
sage from earl ier "voyages to the East.") The theoretical issues rai sed by
Orienta/ism as a case study of a cu ltural d i scou rse can not be d isposed
of, however, by means of any simple contrast between experience and
textua l i ty.
Said is not a s i m ple po lemicist. H i s critical approach is restless and
mordant, repeated ly push i n g its analyses to epistemological l i m its . Be­
h i nd the i mmed i ate i nfl uence of Fouca u l t l ies an ambivalent adm i ration
ON O R I E N TA L I S M 259

for N i etzsche. At various moments i n h i s book Said i s led to argue that

a l l c u ltura l defi n itions m u st be restrictive, that a l l knowledge i s both
powerfu l and fiction a l , that a l l language d istorts . He suggests that "au­
thentic i ty," "experience," " rea l ity," "presence" are mere rhetorica l con­
ventions. The genera l i nfl uence of the French theory that Said has done
so m uc h to i nterpret for American readers i s here most apparent (see
particularly h i s "Abcdari u m Cu lturae" in Said 1 9 75 : 2 7 7-344) . Wh i le he
c i tes Levi -Stra uss and Barthes as wel l as Foucau lt, at the same ti me Said
makes freq uent a ppea l s to an old-fash ioned ex i stenti a l rea l ism . I n the
m u ltivoca l world situation I h ave outl i ned th i s sort of u ncerta i nty i s cru­
cial . Shou l d criticism work to cou nter sets of c u ltura l l y prod uced i mages
such as those of Orienta l ism with more "authentic" or more " h u man"
representations? Or i f criticism m ust strugg le aga i n st the proced u res of
representation itse lf, how i s it to begi n ? How, for example, i s an opposi ­
t i o n a l c r i t i q u e of Orienta l ism t o avoid fa l l ing i nto "Occidenta l is m " ?
These a re fu ndamental issues- i n separably pol itical and epistemologi­
ca l - ra i sed by Sa i d 's work.

Said never defi nes Orienta l i sm but rather q u a l i fies and designates it from
a variety of d i sti nct and not a l ways com pati ble standpoi nts. The book
beg i n s by postu l ating th ree loose "mean i ngs" of Orienta l i sm , " h istorical
genera l i zations" that com prise the " backbone" of his su bseq uent analy­
ses . F i rst, Orienta l i s m is what Orienta l i sts do and have done. An Orien­
ta l i st i s "anyone who teac hes, writes about, o r researches the Orient . . .
either i n its spec i fic or its genera l aspects." I n c l uded i n th i s gro u p are
academ ics and government experts : ph i lo l ogists, soc iologists, h i stori ans,
and anth ropo l ogists. Second, Orienta l ism i s a "sty le of thought based
u pon an onto logical and epistemologica l d i sti nction made between 'the
Orient' and (most of the ti me) 'the Occ ident"' (p. 2 ) . Any writi ng, Said
goes on to suggest, at any period i n the h i story of the Occident that ac­
cepts as its sta rting poi nt a basi c d i chotomy between East and West and
that makes essentia l i st statements about "the Orient, its people, customs,
' m i nd,' destiny, and so on" i s Orienta l i st. F i n a l ly, Orienta l ism i s a "cor­
porate i nstitution for dea l ing with the Orient," wh ich, d u ring the colon i a l
period fo l low i n g roug hly t h e l ate eighteenth century w i e l d s t h e power of
"dom i n ati ng, restructu ri ng, and havi ng authority over the Orient" (p. 3 ) .
This th i rd designation, u n l i ke t h e other two, is pitched a t a rigorously
trans i nd iv i d u a l , c u l tu ra l level and suggests "an enormously systematic"

mechan ism capable of orga ll i z i ng and large l y determ i n i ng whatever may

be said or written about the Orient.
One notices i m med iate ly that i n the fi rst and th i rd of Said's "mean­
i n gs" Orienta l ism is concerned with someth ing cal led the Orient, w h i l e
i n t h e second t h e Orient exists merely a s t h e construct o f a questionable
menta l operation . This ambival ence, which someti mes becomes a con­
fusion, i n forms much of Said's argument. Frequently he suggests that a
text or trad ition d i storts, domi nates, or ignores some rea l or authentic
featu re of the Orient. El sewhere, however, he den ies the existence of any
"rea l Orient," and in th is he i s more rigorously faithfu l to Foucau lt and
the other rad ical c ritics of representation whom he c ites. Indeed the ab­
sence of anyth i n g more than a brief a l l usion to the "brute rea l ity" of the
"cultures and nations whose location is in the East . . . the i r l ives, h i sto­
ries and customs" represents a significant methodological choice on h i s
part. Orienta l i st i n authentic ity is not an swered b y a n y authenticity. Yet
Said's concept of a "di scou rse" sti l l vaci l lates between, on the one hand,
the status of an ideo l ogical d i stortion of l ives and cu ltu res that a re never
concretized and, on the other, the cond ition of a persi stent structu re of
sign i fiers that, l i ke some extreme exam ple of experimenta l writing, refers
solely and end l essly to itself. Said is thus forced to rely on nearly tauto­
logical statements, such as h i s freq uent comment that Oriental i st d i s­
cou rse "oriental izes the Ori ent," or on rather u n h e l pfu l spec ifications
such as : "Orienta l ism can th us be regarded as a manner of regu larized
(or Orienta l i zed) writi ng, vision, and study, dom i nated by i m peratives,
perspectives, and ideological bi ases ostensibly suited to the Orient"
(p. 202).
If red undancy hau nts Said's accou nt, th is is not, I t h i n k, merely the
resu lt of a hermeneutical short c i rc u it i n which the critic d iscovers i n h i s
topic what he h a s al ready p u t there. N o r i s it s i m p l y an effect o f h i s
i nsistence on t h e sheer knitted -togetherness o f a textual u n ity that is con­
stantly in danger of decomposi ng i nto its d i scontin uous functions, au­
thors, i nstitutions, h i stories, and epistemo logi ca l l y d i stinct epochs. Be­
yond these problems (faced by any i nterpreter of constructed , complex
c u l tu ra l ensembles) l ies a substantial and d isq u ieting set of q uestions
about the ways in w h ich d isti nct groups of h uman ity (however defi ned)
i magi ne, describe, and com prehend eac h other. Are such discou rses u l ­
timately condem ned t o red u ndancy, t h e prisoners o f thei r own authori­
tative im ages and l i ngu istic protocol s ? Orienta l i sm-"enormously sys­
tematic," cosmo logical in scope, i ncestuously se lf-referential -emerges
ON O R I E N TA L I S M 261

a s much more than a mere i nte l l ectu al o r even ideologica l trad ition . Said
at one po i nt cal ls it "a considerable d imension of modern pol itical­
inte l l ectu al c u l tu re ." As such it " h as less to do with the Orient than it
does with 'our' worl d " (p. 1 2) .
The q u otation ma rks placed b y Said around our may b e understood
to h ave generated h i s enti re study. The reasons for this are not s i m p l y
personal b u t l ead us t o what Sa id rightly identifies a s "the m a i n i nte l l ec­
tual issue ra i sed by Orienta l ism. Can one d i vide h uman rea l ity, as i ndeed
h u man rea l i ty seems to be gen u i ne l y d ivided , i nto c l early d ifferent c u l ­
tu res, h i stories, trad itions, soc ieties, even races, a n d survive t h e conse­
q uences h u ma n l y ? " (p. 4 5 ) . The res u l t of such d i sti nctions, he argues, i s
t o c reate invid ious and i m peri a l ly usefu l oppositions that serve t o " l i m i t
t h e h u man encou nte r between d i fferent c u ltures, trad itions, a n d soc i ­
eties" (p. 46) . (It i s worth not i n g i n passi n g that we-they d i sti nctions of
the kind Sa id condem ns are a l so u sefu l to anti - i m peri a l ism and nation a l
l i beration movements . ) The key theoretical issue ra i sed b y Orienta/ism
concerns the status of all forms of thought and representation for dea l i ng
with the al i e n . Can one u lt i m ately escape proced u res of d i c hotom i z i ng,
restructu ri ng, and textu a l i z i n g in the making of interpreti ve statements
about foreign c u l tu res and trad itions? If so, how? Said frankly ad m its that
a l ternatives to orienta l ism a re not h is subj ect. H e merely attacks the d i s­
cou rse from a variety of positions, and as a res u l t h i s own standpoi nt i s
not sharply defi ned or logica l l y gro u nded . Someti mes h i s analysis fl i rts
with a c ritique of representation as such; but the most constant position
from which it attacks Orienta l ism is a fam i l iar set of val u es associ ated
with the Western anth ropological h u man sciences-exi stential sta ndards
of " h u m a n encounter" and vague recommendations of " persona l , au­
thentic, sympathetic, h u man istic know ledge" (p. 1 9 7 ) .
I n Said's d i scussion o f t h e Orienta l ist as h u man ist these ass u m ptions
are thrown i nto s h a rp re l ief. There has, of cou rse, been a sym pathetic,
nonred uctive Oriental ist trad ition, a strand that Said down p lays . He
does, however, on one occasion grapple w ith this "good" Orienta l ism in
the person of its most representative figure, Lou i s Mass i gno n . Massignon
m u st stand for those Orienta l i sts-one th i n ks of scholars such as Sylva i n
Lev i , Marcel Mauss, H e n ry Corb i n-whose involvement with t h e foreign
trad itions they stud i ed evolved i nto a deep personal and d i a logica l q u est
for comprehension . Such writers h ave characteristica l l y presented them­
se l ves as s pokesmen for orienta l or pri m itive "wisdom " and a l so as dem­
ocrati c reformers and h u m a n i st critics of i mperia l i s m .

Sa id's d i scussion of Massignon, the most i nteresting i n h i s book, i s a

cruc i a l test case for the theory of Oriental ism as a pervasive and coerc ive
c u ltu ra l d i scou rse . Here Said can no longer genera l ize sweepingly and
categorica l ly about "the Orienta l i st" and "Orienta l i sm ." ( I ndeed his crit­
ical manner someti mes a ppea rs to m i m ic the essential izing d i scou rse it
attacks . ) Said gives fu l l and generous recogn ition to Massignon's pro­
fou nd empathy with I s l a m i c mystic i sm , to h i s subtlety and range of
expression, and to h i s po l itical comm itment on behalf of exploited ori­
enta l s ; but he a rgues that the great sc holar's work i s sti l l fi nal l y defined
with i n a restri cted "d i scu rsive consistency." He deploys h i s most N ietz­
schean arguments to the effect that any re p resentation m ust be " i m p l i ­
cated , i ntertw i ned, embedded , i nterwoven with a great many other
th i ngs bes ides the 'truth,' w h i ch i s itself a representation" (p. 2 72 ) .
Sa id shows rather effective l y t h e l i m its o f Massignon's i nte l l ectual
worl d . The most i m portant of these i s the scholar's tendency to perceive
present M idd le Eastern rea l ities with reference to trad itiona l ly defi ned
c u ltural or s p i ritual va l ues . Massi gnon saw the earthbound experiences
of colon i a l i sm , econom ic oppression, love, death, and so on th rough the
"deh u ma n i zed lens" of a q uasi-metaphysical conception of Sem itic es­
sence. He perceived the Pa lest i n ian confl ict, for exam ple, in terms of the
q u a rre l between I saac and Ishmae l . Here as e l sewhere Said makes short
work of appea l s beyond a corrupt present to an authentic trad ition. Such
appea ls, howeve r · sym pathetic, are always suspect i n thei r d i sparage­
ment of cu rrent processes of c u l tu ra l and po l itical i nvention . U lti mate l y
Massignon cou ld not avo id partici pation i n a "wi l l t o knowledge over the
Orient and on its beha lf" (p. 2 7 2 ) .
If even a "ge n i us" s u c h a s Massignon c a n be s o restricted, i t be­
comes d i fficu lt to escape the bleak though rigorous conc l usion that a l l
h u man expression i s u lti mately determi ned b y cu ltura l "archives," and
that global truth m u st be the resu lt of a battle of "d iscursive formations"
i n w h i c h the strongest prevai l s . Said i s uneasy with so Foucau ldian a
conc l usion . He goes on to reassert a transcendent h u man i st standard,
rescu i ng Mass ignon, who i s after all "a very human bei ng" from an i nsti­
tutional determ i nation now q u a l i fied as only a "d i mension" of his "pro­
d uctive capac ity." Massignon does in the end rise above h i s cu lture i nto
a "broader h i story and anthropology." Massignon's statement "nous
sommes tau s des Sem ites" shows, accord i ng to Sa id, "the extent to which
h i s ideas about the Orient cou ld transcend the local anecdotal c i rcum­
stances of a Frenchman and of French society" (p. 2 74) . A very h u man
ON O R I E N TA L I S M 263

bei ng becomes a h u m a n i st. But the privi l ege of stand i ng above c u l tu ra l

particu l arism, o f asp i r i n g t o t h e u n iversal ist power that speaks for hu­
m a n ity, for the u n i versa l experiences of love, work, death , and so on, i s
a privi lege i nvented b y a tota l izi ng Western l i bera l ism. Th i s benevolent
com prehension of the visions p rod uced by mere " l ocal anecdota l c i r­
c u m stances" is an authority that escapes Said's criticism .
Said sometimes presents h i s critical postu re as "oppos itiona l " (p.
3 2 6), a stance of open attac k on i m perial power and knowledge (see Said
1 9 76, 1 9 79) . More frequently, though, he q u a l i fies h i m self positively as
a h u m a n i st. This stance seems to presu ppose a pa rticula rist, even i n d i ­
v i d u a l ist attitude com bi ned with cosmopol ita n i s m a n d a general va lori­
zation of c reative process. For example T. E . Lawrence is taken to task
for writi ng ( i n a rather adm i rably self-conscious passage) of "Arabs"
rather than of " i nd ividual Arabs with narratable l i fe h i stories" (p. 2 2 9 ) .
S u c h genera l statements, S a i d a rgues, "necessari ly subord i n ate" an Ar­
ab's s pec ific fee l i ng of j oy, of sadness, of i nj u stice i n the face of tyranny,
and so o n . Sa id castigates Orienta l ism for its construction of static images
rather than h i storical or personal "na rrati ves." The " h u man experience,"
whether that of the i nd i v i d u a l Orienta l ist or of h i s or her objects of study,
is flattened i n to an asserted authority on one side and a genera l ization
on the other. Said characterizes the h u ma n rea l ities th us e l ided with q uo­
tations from Yeats-'"the u ncontro l lable mystery on the besti al floor,' i n
which a l l h u mans l ive,'' a n d "the fou l rag and bone shop o f the heart"
(pp. 2 3 0 , 1 1 0) .
It is sti l l an open q uestion, of cou rse, whether an African pastora l i st
shares the same exi stenti a l "besti a l floor" with an I rish poet and h i s read-
ers. And it i s a general featu re of h u ma n i st common denom inators that
they a re m ea n i ng l ess, s i nce they bypass the loca l c u ltural codes that
make persona l experience a rti c u l ate. Sa i d 's resort to such notions u nder-
l i nes the abse n ce i n h i s book of any developed theory of c u l t u re as a
d i fferentiati n g and expressive ensemble rather than as s i m ply hegemonic . \ ' " ·

and d i sc i p l i na ry. H i s basic va l ues a re cosmopol i ta n . He approves as an

alte rn ative to Orienta l i s m the c u ltura l hermeneutics of Erich Auerbach,
Ernst Robert C u rti us, and C l i fford Geertz . H e appears to endorse the an­
th ropologic a l commonplace that "the more one i s able to leave one's
c u l tu ra l home, the more eas i l y i s one able to j udge it, and the whole
world as wel l , with the spi ritu a l detac hment and generos ity necessary for
true vision" ( p . 2 5 9 ) . The anthropologist as outsider and partici pant­
observer (ex i stenti a l shorthand for the hermeneutical c i rc l e) is a fam i l iar

modern topos . Its wisdom-and authority- is expressed with a d i stu rb­

i n g beauty by H ugh of St. Victor (q uoted by Said from Auerbac h ) : "The
man who finds h i s homel and sweet is sti l l a tender beginner; he to whom
every so i l i s as h i s native one i s a l ready strong; but he i s perfect to whom
the entire world is as a fore ign land" (p. 2 5 9 ) .

Sa id's h uman ist perspectives do not harmon ize with h i s u s e o f methods

derived from Foucau lt, who is of cou rse a rad ical critic of h u man i s m . But
however wa ry and i nconsistent its appea ls, Orienta/ism is a pioneeri ng
attempt to use Fouca u l t systematica l l y in an extended cultura l analysis.
Its d ifficu lties and successes shou ld thus be of interest to h i storians, crit­
ics, and anth ropo logists .
We have a l ready encountered the centra l notion of d i scourse. For
Said a d i scou rse is the c u ltural-po l itical configu ration of "the textua l at­
titude" (pp. 9 2-94) . The most extreme exam ple of th i s attitude is Don
Quixote; its condensed modern form u l ation is Flau bert's Dictionnaire
des idees rer;ues . People prefer order to d i sorder; they grasp at formu las
rather than actu al ity; they prefer the guidebook to the confusion before
them . " I t seems a common h u man fa i l i ng," Said writes, using the word
human with significant ambivalence, "to prefer the schematic authority
of a text to the d i sorientations of d i rect encounters with the h u man" (p.
93). In certai n cond itions th i s textua l attitude hardens i nto a body of rigid
cu ltura l defi n itions that determ i ne what any individual can express about
a certa i n acuta l ity. T h i s " rea l ity" coa lesces as a field of representations
prod uced by the d i scou rse. The cond itions for d i sc u rsive harden i ng are
not clearly defi ned by Said, but they appear to be rel ated to an ongoing
i m bal ance of power that perm its-perhaps obl iges-a pol itica l l y and
tech no logi ca l ly stronger cu lture o r group to defi ne wea ker groups. Th us
in Said's analys i s occidental cu lture through the d i scou rse of Oriental ism
"suffused" the activity of orienta l s with " mean i ng, i ntel l igibi l ity, and re­
a l ity." The Orienta l i st d i scou rse, which, accord ing to Sa id, did not sig­
n i ficantly change after the l ate e i ghteenth centu ry, generated a d u m b
show o f orienta l i m ages . "Actua l h u man i nterchange between Oriental
and Westerner" (p. 95) was systematica l ly repressed . Orientals had no
vo ice o n the "Orienta l i st Stage ."
Said's general attempt to extend Foucault's conception of a d i scourse
i nto the area of c u ltura l constructions of the exotic is a prom i s i ng one.
Fouca u l t's overa l l u ndertaking has of cou rse been scru pulously ethno-
ON O R I E N TA L I S M 265

centr i c . I n atte m pt i ng t o isol ate t h e epi stemologica l strata o f E u ropean

thought he has avoided a l l com parative appea l s to other worlds of mean­
i ng. There a re no evocati ons of pensee sauvage, of Hopi l i ngu i stic
categories, and the l i ke . Fouca u l t probably bel ieves such appea l s to be
methodologica l l y dubious, and he contrasts Western c i v i l ization o n l y
wh i msica l l y t o Borges' "Chi nese encyc loped ia" a t t h e outset o f Les mots
et les chases . Foucau l t is i n te rested i n the ways i n w h i c h a given c u l tu ra l
order constitutes itsel f b y means o f d i s c u rsive defi n itions: sane-mad,
hea lthy-sick, lega l -cri m i na l , normal-perverse. The i l l ic i t categories for
Fouca u l t exist not as a reas of an outlaw freedom but as c u ltura l l y pro­
d uced, a rra nged experiences .
Said extends Foucau lt's a n a l ysis to i n c l ude ways i n whic h a cu ltu ra l
order is defi ned externa l l y, w i t h respect t o exoti c "others ." I n an i m peri­
a l i st context defi n itions, representations, and textu a l izations of subj ect
peo p l es and p laces play the same constitutive role as " i nterna l " represen­
tations (for exa m p l e of the cri m i nal c l asses i n n i neteenth-century E u rope)
and h ave the same conseq uences-d isc i p l i ne and confinement, both
physical and ideo l ogica l . Therefore "the Orient," in Said's ana l ysis, ex i sts
u n i q u e l y for the Occident. H is task in Orienta /ism is to d ismantle the
d i scou rse, to expose its oppressive syste m , to "clear the arc h i ve" of its
received ideas and static i mages.
Fou ca u l t i s not eas i l y i m itated . H is writi ng has been a series of ex­
periments and tactica l interventions rather than a method ical progra m .
Sa id's a ppropriation o f Fouca u lt stri kes a comm i tted , mora l note. Con­
trasti n g (and preferri ng) Foucau lt to Derrida, Said notes that the latter's
"end l ess worry i n g of representation" from "with i n " canon ical Western
texts does not perm it critical attention to move beyond the written (how­
ever " i ndec idable") to the soc i a l and pol itica l , to the institutions u nder­
l y i n g an i m perial and hegemon i c "Western thought." Fouca u l t's brand of
critici sm , u n l i ke Derrida's, "reads" a prison or a hospita l , a lega l system,
or-as Sa id does in Orien ta/ism-a geopo l itical a rtifact such as De Les­
seps' ca nal (seen as an Orienta l i st i nscri ption). "By v i rtue of Fouca u l t's
criticism we a re able to understand c u lture as a body of d i sci p l i nes h av­
i n g the effective force of knowledge l i n ked systematica l l y, but by no
means i m med i ate l y, to power." Cu lture as Said conceives it i s l i ttle more
than "a massive body of se lf-congratulating ideas" and of "disc i p l i nes"
that the critic m ust u n mask and oppose without c l a i m i ng-by v i rtue of a
system or sovereign method-to stand outs i de of " h istory, subjectivity, or
c i rc u m stance." 'The critical consc iousness . . . havi ng i n iti a l l y detached

itse lf from the dom i nant c u ltu re" thereafter adopts " a situated and re­
s ponsible adversary pos ition" (Said 1 978b: 709, 690, 7 1 3 ) .
It i s rather d ifficu lt, however, to qual ify Foucau lt's restless guerri l l a
activity on beh a lf o f t h e exc l uded , agai nst a// tota l i z i ng, defi n i ng, essen­
tia l iz i n g a l l i ances of knowledge and power as "situated and respons i ble."
Said h i mself deploys a rather l oose co l l ection of "adversary theoretical
models" derived from Foucau lt, G ramsci, Lu kacs, Fanon, and others
( 1 9 79 : 1 6) . A key pol itical term for Said is oppositional, and it is fairly
clear what th i s means in the l i m ited context of a book such as Oriental­
ism, w h i c h "writes back" at an i m perial d iscou rse from the position of an
oriental whose actua l ity has been d i storted and den ied . More genera l ly,
however, it is apparent that a wide ra nge of Western humanist assu mp­
tions escape Said's oppositiona l analysis, as do the d iscu rsive a l l iances
of knowledge and power prod uced by anticolonial and partic u l arly na­
tiona l ist movements .
Beyond h i s overa l l stance as "oppos itiona l " c u l tu ra l critic Said makes use
of other Fouca u l d i an approaches that should be d iscussed briefly. Most
sign ificant is h i s adoption of the postu re of crit.i�aJ retrosQec.UQ.n that
N ietzsc he cal led genealogy. In th i s Said is true to Foucau lt's l ater evolu­
tion away from the methodology of l ayered "archaeologica l " d i sconti n­
u i ty exempl ified in The Order of Things and The Archaeology of Knowl­
edge and towards a presentation of the l i neages of the present, as
exempl ified in Discipline and Punish and espec i a l ly The History of Sex­
uality, vol u me 1 .
The field of Orienta l ism is genealogica l l y d istri buted i n two ways :
sync h ron ica l ly (constituting in a u n ified system a l l Western textual ver­
sions of the Orient) and d i ac h ro n i ca l l y (plotting a si ngle l i neage of state­
ments about the East, r u n n i n g from Aeschylus to Renan to modern pol it­
ical soc iology and ;'area stud ies") . L i ke a l l genealogies Said's grows more
specific as it approaches the present it has been constructed to explai n
and affect. Thus the bu l k of h i s account describes the heyday of Orien­
ta l ism in the n i neteenth and early twentieth centu ries. Th is is fol lowed
by an attem pt to generate mea n i ngs i n the c urrent Middle East situation
with reference to this c l assical trad itio n . The aim here is not, of cou rse,
the one most usual in genea logies-a new legiti mation of the present­
but rather, as in Foucau lt's History of Sexuality and Madness and Civili­
zation, rad ical de-legiti mation . A certa i n degree of anac h ron ism is
ON O R I E N TA L I S M 267

open l y embraced . 1 Genea logy, l i ke a l l h i storical descri ption and analy­

sis, is constructive. It makes sense i n the present by making sense sel ec­
tive l y out of the past. I ts i n c l usions and exc l usions, its narrative conti­
n u ities, i ts j udgments of core and peri p hery are fina l l y legiti m ated either
by convention or by the a uthority granted to or arrogated by the genea l­
ogi st. Genealogy is perhaps the most pol itical of h i storical modes ; but to
be effective it can not appear too ope n l y tendentious, and Said's genea l­
ogy suffers on th i s score . To his c red it he makes no sec ret of the restrictive
choices i nvol ved .
F i rst, Said l i m its h i s attention a l most exc l usively to statements about
the Arab M i d d l e East-o m itti ng, regretfu l ly but fi rm ly, the Far East, I n d i a,
the Pac ific, and North Africa . The om i ssion of the Magh reb is c ruc i a l , for
it ensures that Said w i l l not have to d iscuss modern French Orienta l i st
c urrents . I n a French context the ki nds of critical q uestions posed by Said
have been fam i l ia r s i nce the A l gerian war and may be found strongly
expressed wel l before 1 95 0 . It wou l d s i m p l y not be poss i b l e to castigate
recent French "Orienta l ism" in the way that he does the d i scourse of the
modern American Middle East "experts," w h i c h is sti l l shaped by Cold
War patterns and by the po larized Arab-Israe l i confl i ct.
Sa i d 's second genea logical l i m i tati on restri cts the national trad itions
u nder consideration to the B ritish and French strands, with the add ition
of a recent American offspri ng. H e i s obl i ged to ru le out Ita l ian, Span i s h ,
Russian, and espec i a l l y German Orienta l isms. T h e h igh l y developed
n i neteenth-centu ry German trad ition is cast as peri phera l to French and
Engl ish p i oneers but, more i m portant, as not constituted l i ke these two
i n a c l ose re l ationsh i p with co l o n i a l occu pation and dom i nation of the
Orient (pp. 1 6-1 9 ) . In effect, German Orienta l ism i s too d i s i nterested
and thus atypical of a genea logy that defines the d i scou rse as essentia l ly
colo n i a l i st. If Said's pri mary aim were to write an i ntel l ectual h i story of
Orienta l i sm or a h i story of Western ideas of the Orient, h i s narrow i n g
a n d rather obviously tendentious s h a p i n g o f t h e f i e l d cou ld be taken as a

1 . In Discipline and Punish ( 1 975 : 3 5 ) Foucault writes of h i s i ntention to

produce a h i story of the prison : " Par un pur anachronisme? Non, si on entend
par Ia fai re l ' h istoire d u passe dans les termes du present. O u i , si on entend par
Ia fa ire l ' h istorie d u present" (p. 3 5 ) . His fu l lest statement on genealogy is
" N ietzsche, Genea logy, H istory" ( 1 977). This chapter d iscusses only those works
by Fouca u l t that were avai lable at the ti me of publ i cation of Orienta/ism. I do
not consider h i s refinements and transgressions of h i sto rical method fol lowi n g
t h e fi rst vol ume of History of Sexuality.

fata l flaw. But h i s u ndertaking i s conceived otherwise and is ope n l y an

oppositional genea logy. If Said's genea logy someti mes appears c l u m s i l y
rigged (the fi nal a l l-too-pred ictable zeroing i n on t h e Midd le East and
abrupt j u m p from Conti nenta l to American "Oriental ism" is the least con­
vincing of its "conti n u ities") , one need not reject the enti re critical para­
d igm.
Said i s perfectly co rrett to identify retrospectively a "d iscou rse" that
d i chotomizes and essential izes in its portraya l of others and that func­
tions i n a comp l ex but systematic way as an e lement of colon ial dom i­
nation . It is i m portant that this d i scou rse be recognized wherever it ex­
ists; but the d i scou rse shou ld not be closely identified with the spec ific
trad ition of Orienta l i s m . Its field of appl ication has been far more gen­
era l . The problem with the book, at least from a theoretical standpoi nt,
is its title. In attempti ng to derive a "d i scourse" d i rectly from a "trad ition,"
Said abandons the level of c u ltural criticism proposed by Foucau lt and
rel apses i n to trad itional i nte l l ectua l h i story. Moreover, i n portray ing the
d i scou rse as based on essenti a l ly n i neteenth-century modes of thought,
Said gives h i mself too easy a target. He does not q uestion anthropologi­
cal orthodoxies based on a mythology of fieldwork encounter and a her­
meneutica l l y m i nded cu ltura l theory-orthodox ies he often appears to
share .
It is apparent that "discou rse" analysis cannot safe ly b e fou nded o n
redefined "trad itions." N o r c a n it b e derived from a study of "authors ."
The general tendency i n modern textual stud ies has been to red uce the
occasion of a text's c reati on by an i nd ividual subject to merely one of its
generative or potenti a l l y mean i n gfu l contexts. Wh i le recogn i z i ng the im­
portance of th i s separation of the text from the work (Barthes : "The work
is held in the hand, the text in language"), Said has resi sted rad ical struc­
tural ist attacks on phenomenology and on the essential (begi n n i ng and
conti n u i ng) function of an authorial i ntention . Beginn ings ( 1 975), which
preceded Orienta/ism, i s a deta i l ed and perspicuous med itation on this
set of issues. It i s concerned precisely with the prob lem, experienced by
a wide range of modern ist writers, of bei ng an "a uthor." Steering a com­
plex cou rse between i ndividual ist conceptions of creativity on the one
hand and on the other red uctions of "the mov i ng force of l ife and beh av­
iour, the forma informans, intention" (p. 3 1 9) to an external system,
whether c u ltural or critica l , Said suggests an i ntermed i ate analytical to­
pas that he cal l s a "career." The modern author's i ntention is not so m uch
to produce works as it is to beg i n (and to continue beg i n n i ng) to write. A
O N O R I E N TA L I S M 2 69

career is the ensem ble of these complex h istorica l l y and c u l tu ra l l y s itu­

ated i ntentions. It i s a l ways i n process, a lways being begun in speci fic
s ituations, and never possessi ng either a stable essence or a shaped bio­
graphical fi n a l ity. The author i s reconceived , and i n the face of structu r­
a l ist d i ssol ution rescued .
It is not s u rprisi ng, then, that Said, i n d i scussing Orienta l ism as a
d i scou rse and a trad ition, adopts what he ca l ls a "hybrid perspective."
"Fouca u l t bel ieves that i n genera l the i nd ividual text or author counts for
very l ittle; empi rical ly, i n the case of Orienta l ism (and perhaps nowhere
e l se) I find th i s not to be so" ( 1 978a : 2 3 ) . T h i s dogged ly empi rical and
curiously q u a l i fied assertion separates Said sharply from Fouca u l t . What
is i mportant theoretica l ly is not that Foucau lt's author cou nts for very
l i ttle but rather that a "d isc u rs i ve formation" -as opposed to i deas, c ita­
tions, i nfl uences, references, conventions, and the l i ke-is not prod uced
by authori a l subjects or even by a group of authors arra nged as a "trad i­
tion ." Th i s methodolog i ca l (not empi rica l ) po i nt i s i m portant for anyone
i nvolved in the k i n d of tas k Said is attempti ng. One can not com bine
with i n the same analytic tota l ity both personal statements and discursive
statements, even though they may be lexica l ly identica l . Said's experi­
ment seems to show that when the analysis of authors and trad itions is
i nterm i xed with the analysis of d iscurs ive formations, the effect is a mu­
tua l weake n i ng.
None of the authors d iscussed in Orienta/ism i s accorded a "career"
in the comp l ex se nse pos ited by Beginnings, but a l l a re portrayed as i n ­
stances o f Orienta l i st d i scou rse. U n l i ke Fouca u lt, however, for whom
authori a l n ames fu nction as mere labels for discurs ive statements, Said's
authors m ay be accorded psychoh istorical typical ity and are often made
through thei r texts to have representative Orienta l i st experiences. One
exa m p l e among many, chosen for the fam i l i arity of its subject, i s Said's
read i n g of a passage from Marx-the end of h i s articl e "The B ritish R u l e
i n I nd ia" ( S a i d 1 9 78a : 1 5 3 -1 5 7 ) .
Marx denounces an affront t o " h u ma n fee l i ng" -the spectac le of
I n d i a n soci a l l ife bruta l l y disrupted , "th rown i nto a sea of woes" by im­
peri a l i s m ; but he q u ickly rem i nds his readers that "these idyl l i c v i l l age
com m u n iti es" have always been the fou ndation of "Orienta l despotism ."
They have " restra i n ed the h u m a n m i nd with i n the sma l l est poss i b l e com­
pass, making it the u n resisting too l of su perstition, enslaving it beneath
the trad ition a l r u l es, deprivi ng it of a l l its gra ndeu r and h i storical e ner­
gies ." England, Marx goes on to say, is h i story's agent; its task is to " lay

the material fou ndations of Western Society i n Asi a ." Said scents Orien­
ta l i sm in the reference to despotism and in a l ater citation of Goethe's
Westostlicher Diwan. He identifies a "romantic redemptive project,"
which assumes the genera l Western privi l ege of putting the Orient-stag­
nant, d ismembered , corru pt-back together. Marx is a l so convicted of
subsu m i ng " i nd ividuals" and "ex i stenti a l h u man identities" u nder "arti­
ficial ent ities" such as "Orienta l ," "As iatic," "Sem itic," or with i n col lec­
tives such as " race," " menta l i ty," and "nation ."
Here an effective read i ng begins to get out of hand . It is unclear why
Said does not a l so conv i ct Marx of subsu m i n g individuals u nder the "ar­
tifici a l entit ies" "class" and " h i story." Furthermore, if Marx's participation
i n Orienta l ism derives from his i n attention to ex i stenti a l , i nd ivid ual
cases, one wonders how soc i a l or c u ltura l theory is ever to be "human l y"
bu i lt. I n add ition, it is wel l known that Marx heaped "Orienta l ist" scorn
and condescension u pon the "id iocy of rural l ife" wherever he found it,
bel ievi ng that such stagnant, repressive s i tuations had to be violently
transformed before they cou l d i m prove. Here Said skirts " unfa i rness" to
Marx . Wh i l e legitimatel y iso l ating Orienta l ist aspects of the text, he too
q u ickly skims over its rhetorical i ntentions. Moreover, Said soon aban­
dons any d i scussion of Orienta l ist statements and goes on to uncover i n
the text a typical Orienta l i st experience. Marx, we are told, at first ex­
pressed "a n atura l h u man repugna nce" toward the sufferi ng of orientals;
he felt a "h uman sympathy," a "fel l ow fee l i ng." This "persona l human
experience" was then "censored" by a process of Ori enta l i st label i ng and
abstraction, "a wash of sentiment" was repressed by " u nshakable defi­
n itions." (Sai d writes in the past tense, as i f this i s what rea l l y happened
i n Marx's m i nd . ) "The vocabu lary of emotion d i ssi pated as it submitted
to the l ex icograph ical pol ice action of Orienta l i st science and even Ori­
enta l ist art. An experience was d i slodged by a d i ctionary defi n ition" (p.
1 5 5 ) . By now Said cou ld not be farther from Fouca u lt's austere pages,
where al l psycholog i z i n g is forbidden and where authors escape at l east
having to go through such i n structive "experiences ." Said's descriptions
of Orienta l i st d iscourse are frequently sidetracked by h u man ist fables of
suppressed authenticity.
Discourse analysis is a lways i n a sense unfa i r to authors . It is not
i nterested in what they have to say or feel as subj ects but is concerned
mere l y with statements as related to other statements in a field . 2 Escaping

2 . On the i n itial defin ition of this field, which he ca l l s a "discu rsive forma­
tion," see Foucault's strictu res in The Archaeology of Knowledge ( 1 969 : chap.
ON O R I E N TA L I S M 271

an i mpress ion o f u n fa i rness and red uctionism i n t h i s k i nd o f analysis is a

matter of methodological r i gor and styl istic tact. Foucau lt, at least, does
not appear u n fa i r to authors becau se he seldom appea l s to any i ndividual
i ntentional ity or subj ectivity. " H ybrid perspectives" such as Said's h ave
considerabl y more d i fficu lty escapi ng red uction ism . 3
I ndeed Said's methodo l ogica l catho l i c ity repeated l y b l u rs h i s a n a l ­
ysi s . If he i s adva n c i n g a nthropological arg u ments, Orienta l ism appears
as the cu ltura l q u est for order. When he adopts the stance of a l iterary
critic, it emerges as the processes of writing, textual izi ng, and i nterpret­
i ng. As an i nte l l ectua l h i storian Said portrays Oriental ism as a specific
series of i nfl uences and schools of thought. For the psychohistorian Ori­
enta l i st d i scou rse becomes a representative series of persona l - h i storical
experiences . For the Marx i st critic of ideology and cu lture i t is the expres­
sion of defi n ite po l itical and economic power i nterests . Orienta l i sm i s
a l so a t ti mes conflated w i t h Western positivism, with genera l defi n itions
of the pri m itive , with evo l ution i s m , with rac i s m . One cou ld conti nue the
l i st . Said's d i scou rse analysis does not itse lf escape the a l l - i n c l usive "Oc­
c i denta l ism" he speci fica l l y rejects as a n a l ternative to Orienta l ism
(p. 3 2 8 ) .

Though Said's work freq uently rel apses i nto t h e essent i a l izing modes it
attacks and is a m bivalently en meshed i n the tota l izin g habits of Western
h u m a n i s m , it sti l l succeeds i n q uestion i ng a n u m ber of i m portant anthro­
pological categories, most i mportant, perhaps, the concept of c u lture . I n
th i s fi nal section I shal l s ketch out some of these i ssues, the most far­
reac h i n g q u estions raised by Orienta/ism.
The effect of Said's gene ra l argument i s not so much to u nderm ine

2 ) . Fouca u l t's method ignores " i nfl uences" and "trad itions," demotes "authors,"
and holds in suspense any criteria of d i sc u rsive u n ity based on the persistence or
common a l i ty of "objects," "styles," "concepts," or "themes ." It may be noted that
Said makes u se of all these fam i l i a r elements from the h i story of ideas.
3. Said's critical approach can in fact be q u i te d i sturb i n g, espec i a l l y when
he i s u n covering Orienta l i s m i n lesser-known figures than Marx, among whom
the disjuncture between d i sc u rsive statements and persona l expressions is less
i mmed iately apparent. A particu larly blatant example may be seen in his use of
the great Sanskrit scholar and h u ma n i st Sylva i n Levi in order to show the con­
nection of Oriental ism with i m perial pol itics (Said 1 9 7 8 : 249-250). The m i s l ead­
ing i m age of someone i ntense l y concerned with European " i nterests" in the Ori­
ent (the word interest i s i nserted i n to Levi 's d i scourse) i s nowhere qual ified . For
an affi rmation that modern Orienta l i sts have been far less red uctive than Said
portrays them to be see Houra n i 1 9 79.

the notion of a su bstantial Ori ent as it i s to make probl ematic "the Oc­
cident." It is less com mon today than it once was to speak of "the East,"
but we sti l l make casual reference to "the West," "Western cu l tu re," and
so on. Even theorists of d i sconti n u ity and deconstruction such as Fou­
ca u lt and Derrida continue to set their analyses with i n and agai nst a
Western tota l ity. Said shares the i r ass u m ptions i nasmuch as he portrays
the Western c u l tu re of which Orienta l ism is an exemplar as a d iscrete
entity capable of generating know ledge and i nstitutional power over the
rest of the p l a net. Western order, seen t h i s way, is i m peria l , u n reciproca l ,
aggressive, and potentia l ly hegemon i c . At times, though, Said perm its u s
to see the fu nction ing o f a more complex d i a l ectic b y mea ns o f which a
modern c u ltu re cont i n uously constitutes itself through its ideo logica l
constructs of the exotic. Seen i n this way "the West" itself becomes a
play of projections, dou b l i ngs, idea l izations, and rejections of a com­
plex, s h ifting otherness . "The Orient" always plays the ro le of origi n or
alter ego . For exa m p l e Renan worki ng i n h i s "ph i l o logical laboratory"
does not s i m p l y concoct the scholarly topos of the Semitic Orient but i n
t h e same process produces a conception o f what i t means to b e E u ropean
and modern (pp. 1 32 , 1 46) .
Here Said's argument rei n forces Stan ley Diamond's ( 1 9 74) conten­
tions that Western cu lture can conce ive of itself c ritica l ly only with ref­
erence to fictions of the pri m i tive . To this d ia lectical view we may use­
fu l ly add the overa l l perspective of Marsha l l Hodgson's h i storical work,
which portrays " E u rope" as, u nti l the late eighteenth centu ry, mere l y "a
fri nge area of the Afro- E u roasian zone of agrarianate citied l ife" (see par­
tic u l arly Hodgson 1 974, 1 963, and Bu rke 1 979, an excel lent su rvey of
Hodgson's com p l ex work). If we adopt along with these perspectives a
genera l ly structu ra l i st suspicion of a l l q uests for origins (the origins of the
West in G reece or in C hristian ity), we are left with a tota l ity in process,
com posed and recomposed in changing external rel ations.
When we speak today of the West, we are usua l l y referring to a
force-tech nologica l , econom ic, pol itica l-no longer rad iating i n any
simple way from a d iscrete geographical or cu ltura l center. This force, if
it may be spoken of in the s i ngular, i s d i ssem i nated i n a divers ity of forms
from m u ltiple centers-now i n c l ud i ng Japan, Austral ia, the Soviet
U n ion, and C h i na-and is artic u l ated in a variety of "m icro­
soc iological" contexts (see Duvignaud 1 9 73). It is too early to say
whether these processes of change wi l l resu lt in globa l c u l tu ra l homoge­
n ization or in a new order of d i vers ity. The new may always look mono-
ON O R I E N TA L I S M 273

l ith i c to the o l d . For the moment, i n any event, all d ichotom i z i n g

concepts should probably b e held i n suspicion, whether they b e the
West-rest ( "Th i rd World") spl it or devel oped-u nderdeveloped , modern­
premodern, and so on. It i s at this level that Said's critique of the d i s­
cou rse he ca l l s Orienta l ism becomes most sign ificant. Moreover, if a l l
essenti a l i z i n g modes o f thought m ust a l so b e held i n suspense, then we
sho u l d attem pt to th i n k of c u ltu res not as organ ica l l y u n ified or trad ition­
a l l y cont i n uous but rather as negoti ated, present processes . F rom this
standpo i nt Said's refu sal to appea l to any authentic and espec i a l l y trad i­
tional orienta l rea l ities aga i n st the fa l se stereotypes of Orienta l ism i s ex­
emplary. H is m a i n concern is not with what was or even what is but with
what is beco m i ng. A lthough of th i s process he tel l s us very l ittle, the
fu ndamenta l q u estion i s posed : on what basis may h u man groups accu­
ratel y (and we must a l so add mora l ly) be d i sti ngu ished ?
The concept of c u lture used by anth ropo l ogists was, of cou rse, i n ­
vented by E u ropea n theorists t o account for t h e col lective art i c u l ations
of h u m a n d i versity. Rejecting both evol utionism and the overly broad
entities of race and civi l i zati on, the idea of c u lture posited the exi stence
of loca l , fu nctio n a l l y i ntegrated u n its. For a l l its supposed rel ati vism,
though, the concept's model of tota l ity, basica l l y organic i n structu re,
was not d ifferent from the n i neteenth-centu ry concepts it rep laced . O n l y
its p l u ra l ity w a s new (see Chapter 1 0, section 2 ) . Despite m a n y s ubse­
quent redefi n itions the notion's orga n i c ist ass u m ptions have persi sted .
Cu ltural system s hold together; and they change more or l ess cont i n u ­
ously, anchored pri mari ly b y language a n d p l ace . Recent sem iotic o r
symbol ic models t h a t conceive o f c u l ture as com m u n i cation are a l so
fu nctiona l i st i n th i s sense (see Leach 1 9 76 : 1 , Geertz 1 9 7 3 , Sch neider
1 968).4
A submerged but cru c i a l emphasis of Said's study is h i s restless sus­
picion of tota l ity. H is critique of Orienta l i st proced ures for enclosing and
characterizing "the Orient" may be appl ied to the presumably more pre­
c i se and even " n at u ra l " entity of cu l tu re . I have a l ready noted with the
example of Massignon Said's d istaste for the most sym patheti c appeal s to

4 . Geertz offers a stri king and problematical image of cultu ral organization
not as a spider or a p i l e of sand but as an octopus "whose tentac les are i n a la rge
part separate l y integrated, neura l l y q u ite poorly con nected with one another and
with what i n the octopus passes for a bra i n , and yet who nonetheless manages
to get around and to p reserve h i m se lf, for a wh i le anyway, as a viable, if some­
what u n ga i n l y entity" ( 1 9 7 3 :407-408) . Cultu re rema ins, bare l y, a n organism.

trad ition . Having stressed so thorough ly that the Orient i s a constituted

entity, he goes on to suggest "that the notion that there are geographica l
s paces with i nd igenous, rad ica l ly 'different' i n habitants who can be de­
fi ned on the bas i s of some rel igion, c u l tu re or rac i a l essence proper to
that geograph ica l s pace is eq ual ly a h igh l y debatable idea" ( 1 978a : 3 3 2 ) .
I n h i s final pages he asks the most i m porta nt theoretica l q uestions o f h i s
study. " How does one represent other c u ltu res? I s t h e notion o f a d i sti nct
c u lture (or race, or rel igion, or civi l ization) a usefu l one?" (p. 3 2 5 ) .
S u c h q uestions need t o b e posed a n d need to b e a l lowed t o stand i n
sharp rel i ef. Having asked them, one does wel l to avo id q u i c k recou rses
to a l ternate tota l ities . (As we have seen, Said h i mself has recou rse to
h u ma n ist cosmopo l itanism and conceptions of personal integrity as wel l
as to a notion of authentic development alternate l y glossed as "narrati ve"
or as a vaguely Marxist " h istory.") It is h igh time that c u l tu ra l and soci a l
tota l ities are su bjected t o t h e kind o f rad ical question i ng that textua l en­
sembles have u ndergone in recent critica l practice (for example Derrida
1 970; Barthes 1 9 7 7 ; Said 1 9 78b and 1 9 7 5 ) . Said's attack on essences
and oppositiona l d i sti nctions is here very m u c h to the poi nt; but col l ec­
tive l y constituted difference is not necessari ly static or positional l y d i ­
chotomous i n t h e m a n n e r o f Orienta l ism a s S a i d describes it. There i s n o
need to d i scard theoreti ca l l y a l l conceptions o f "cu ltu ra l " difference, es­
pec i a l l y once this is seen as not s i m p l y received from trad ition, language,
or envi ronment but a l so as made in new pol itica l-cultu ra l cond itions of
globa l relationa l ity.
H ow are these new conditions to be conceived now that the "si­
lence" of the Orient is broken; now that eth nography, as Lei ris suggested ,
can be m u ltid i rectiona l ; now that authenticity, both personal and c u l ­
tural , i s seen a s someth ing constructed vi s-a-v i s others ? I n these c i rcum­
stances shou ld our ideas of re lationa l ity be drawn from the metaphors of
conversation, hospita l ity, and exchange, as h u man ists such as Massi­
gnon, Sylva i n Lev i , and Mauss have u rged ? Or m u.st we prefer the figu res
of m i l itary maneuver someti mes i nvoked by Foucau lt. It may be true that
the cu ltu re concept has served its time. Perhaps, fol lowi ng Foucau lt, it
should be rep l aced by a vision of powerfu l d i sc u rsive formations globa l l y
and strategica l l y deployed . Such entities wou ld a t l east no longer be
close ly tied to notions of orga n i c u n ity, trad itional cont i n u i ty, and the
enduri ng grounds of la nguage and loca le. But however the c u l tu re con­
cept is fina l l y transcended , it shou ld, I th i n k, be repl aced by some set of
rel ations that preserves the concept's d ifferential and re lativist fu nctions
ON O R I E N TA L I S M 275

and that avo ids the positing o f cosmopol itan essences a n d human com­
mon denom i n ators .
It shou l d be poi n ted out that these prescr i ptions are i n the nature of
what Con rad u rged i n Heart of Darkness-a "de l i berate be l ief." The
planet's c u ltura l future may indeed reside i n the entropy Levi-Stra uss la­
ments i n Tristes tropiques or i n the ideological hegemony Said portrays
in h i s blea ker passages ( 1 978a : 3 2 3 -3 2 5 ) . L i ke Sa id's comm itment to the
h u ma n , any res i d u a l faith in c u l tu re-that is, in the cont i n u i ng abi l ity of
groups to make a rea l d i fference- i s essenti a l l y an idea l i stic choice, a
po l itical response to the present age i n w h i c h , as Con rad wrote, "we are
camped l i ke bew i l dered travel lers i n a garish, u n restfu l hote l " ( 1 9 : 1 1 : 1 ) .
It is the v i rtue of Orienta /ism th at it obl iges its readers to confront such
issues at once person a l l y, theoretica l l y, and pol itica l l y. For its author, as
for Con rad , there can be no natural sol utions. Palestine i s perhaps the
twentieth centu ry's Po land, a d i smembered nation to be rei nvented .
Sa id, l i ke the Po l ish-Engl ish writer whom he adm i res and freq uently
q u otes, recogn izes that persona l and c u l tu ra l identities are never given
but m ust be negoti ated . Th i s is a n i m portant em phasis of Said's fi rst book,
a penetrat i n g study of Con rad ( 1 966) . I t wou l d be wrong to d ismiss t h i s
kind o f situation as a berrant, as the cond ition o f exi l es . T h e u n restfu l
pred icament of Orienta/ism, its methodo logical ambivalences, are char­
acteristic of an i ncreasi ngly genera l global experience.
Its a uthor's com p l ex critical postu re may i n th i s sense be taken
as representative . A Pal est i n ian nationa l ist ed ucated in Egypt and the
U n i ted States, a sc holar deeply i m bued with the E u ropean h u man ities
and now professor of Engl i s h and com parative l i terature at Col u mbia,
Said writes as an "oriental ," but only to d i sso l ve the cateogry. H e writes
as a Pa lesti n i a n but takes no support from a spec ifica l l y Pa lesti nian cu l­
ture or identity, tu r n i n g to European poets for h i s expression of essenti a l
va l ues a n d t o French p h i l osophy for h i s a n a l ytical too l s . A rad ica l critic
of a major com ponent of the Western c u l tu ra l trad ition, Said derives most
of h i s sta ndards from that trad ition . The poi nt i n saying this is to suggest
someth i n g of the s i tuation with i n w h i c h books such as Orienta/ism m ust
i nevitably be writte n . It is a context that Sa id has e l sewhere ( i n d iscuss i ng
George E l i ot and the roots of Zion i sm) cal led "a genera l ized cond ition of
homelessness" ( 1 9 7 9 : 1 8) . Such a situation generates d iffic u lt q uestions.
What does it mean, at the end of the twentieth centu ry, to speak l i ke
Aime Cesa i re of a " native land" ? What processes rather than essences are
i nvolved i n present experiences of c u ltura l identity ? What does it mean

to write as a Pa lest i n i a n ? As an American ? As a Papua-New G u i nean ? As

a E u ropea n ? From what d i sc rete sets of cu ltura l resou rces does any mod­
ern writer construct h i s or her d i scourse ? To what world aud ience (and i n
what l a nguage) are these d i scou rses most genera l l y add ressed ? Must the
i nte l lectua l at least, in a l iterate global situation, construct a native land
by writi ng l i ke Cesa i re the notebook of a retu rn ?
Lo, a people dwelling alone, and not reckoning itself among
the nations! -NUMB E R S 2 3 : 9

1 2 . Identity in Mashpee

IN A U G U S T 1 9 76 the Mas h pee Wam panoag Tri ba l Cou nc i l , Inc., sued

i n federal court for possess ion of about 1 6, 000 ac res of land constituti n g
th ree-q uarters o f Mash pee, "Cape Cod's I nd ian Town ." (The townsh i p of
Mash pee extends i n land from the Cape's southern shore, fac ing Marth a's
Vi neyard , between Fal mouth and Barnsta b l e . ) An u n precedented tri a l
ensued whose pu rpose w a s n o t t o settle t h e q uestion o f l a n d owners h i p
b u t rather t o dete rm i ne whether t h e gro u p cal l i ng itself t h e Mash pee Tri be
was i n fact a n I nd i a n tribe, and the same tribe that i n the m id-n i neteenth
centu ry had lost its l ands through a series of contested l egi s l ative acts .
The Mas h pee s u it was one of a gro u p of l and-c l a i m actions fi l ed i n
the late 1 9 60s and 1 9 70s, a rel ative l y favorable period for red ress of
Native American grievances in the courts . Other c l a i m s were bei n g i n i­
tiated by the Gay Head Wam panoag Tri be on Martha's Vi neya rd ; the N a r­
ragan sets of Charlestown , Rhode I s l a n d ; Western Peq uots, Schaghti­
cokes, and Mohegans in Con necti cut; and Onei das, St. Regis Mohawks,
and Cayugas in New York. The Mash pee action was s i m i l a r in concep­
tion to a much-pu b l i c ized suit by the Passamaq uoddy and Penobscot


tri bes laying c l a i m to a l a rge portion of the state of Maine. The i r suit,
after i n itial successes i n Federal D i strict Cou rt, d i rect i ntervention from
Pres ident j i mmy Carter, and five years of hard negoti ation, resu l ted i n a
favora ble out-of-co u rt settlement. The tri bes received $ 8 1 . 5 m i l l ion and
the authority to acq u i re 300, 000 acres with Indian Cou ntry status.
The lega l bas is of the Penobscot-Passamaquoddy su it, as conce ived
by the i r attorney, Thomas Tu reen , was the Non-I ntercourse Act of 1 790.
Th i s patern a l i st legislation, designed to protect tri ba l groups from spo l ia­
tion by u nscru pulous whites, decl ared that a l i enation of I n d i a n lands
cou ld be lega l l y accom p l ished o n l y with perm ission of Congress. The act
h ad never been resci nded , a lthough throughout the n i n eteenth centu ry it
was often honored i n the breac h . When i n the 1 9 70s I nd ian groups ap­
pea led to the Non-I ntercou rse Act, they were attempti ng, i n effect, to
reverse more than a centu ry of attacks on Ind ian lands. The a l ienations
h ad been particularly severe for eastern groups, whose c l a i m to col lec­
tive l a nd was often unclea r. When court dec i s ions confi rmed that the
Non-I ntercou rse Act appl ied to non reservation I n d ians, the way was
opened for su its, l i ke those of the Mai ne tri bes, c l a i m i ng that nearly two
centu ries of I nd i a n land transfers, even ord i nary purchases, were inva l i d
s i nce they had been made without perm ission of Congress.
Although the Mash pee c l a i m was s i m i lar to the Maine I nd ians',
there were cruc i a l d i fferences . The Passamaq uoddy and Penobscot were
genera l ly recogn i zed I n d i a n tri bes with d i stinct comm u n ities and c lear
aboriginal roots i n the a rea . The Mash pee plai ntiffs represented most of
the nonwh ite i n h abitants of what, for over three centu ries, had been
known as a n " I n d i a n town" on Cape Cod ; but their i n stitutions of tribal
governance had long been e l us ive, espec i a l l y d u ring the century and a
half preceed i ng the s u it. Moreover, s i nce a bout 1 800 the Massachusett
language had ceased to be common ly spoken in Mash pee . The town was
at fi rst l a rgely Presbyterian then Baptist in its publ ic rel igion . Over the
centuries i n habitants had i ntermarried with other I nd ian groups, whites,
b l acks, Hessian deserters from the British Army d u r i ng the Revol utionary
War, Cape Verde i s landers . The i n habitants of Mashpee were acti ve i n
the economy a n d soc iety o f modern Massach usetts . They were busi ness­
men, schoolteac hers, fishermen, domestic workers, smal l contractors .
Cou ld these people of I nd ian ancestry fi le su it as the Mashpee Tri be that
had, they c l a i med , been despo i l ed of col lectively held l ands d u r i ng the
m id-n i neteenth century ? Th i s was the q u estion a federa l j udge posed to
a Boston j u ry. O n l y if they answered yes cou l d the matter proceed to a
land-c l a i m trial .

The forty-one days of testi mony that u nfolded i n Federa l D i strict

Cou rt d u ri n g the l ate fa l l of 1 9 7 7 bore the name Mashpee Tribe v. New
Seabury et a/., s horthand for a comp l ex, m u lti partied d i spute . Mash pee
Tri be referred to the p l a i ntiffs, the Mash pee Wam panoag Tribal Cou n c i l ,
I nc . , descri bed b y its members a s a n arm of the Mash pee Tri be . A team
of l awyers from the Native American R ights F u nd , a nonprofit advocacy
group, prepared thei r su it. Its c h i ef arch i tects were Thomas Tu reen and
Ba rry Margo l i n . In cou rt the p l a i ntiffs' case was argued by the tri a l l awyer
Lawrence S h ubow, with ass i stance from Tu reen, Margo l i n , Ann G i l more,
and Moshe Genauer. New Seabury et a l . referred to the New Seabury
Corporation (a l arge development com pany), the Town of Mash pee (rep­
resenting over a h u n d red i nd i v i d u a l landowners), and various other
c l asses of defendant ( i ns u rance compan ies, busi nesses, property own­
ers) . The case for the defense was argued by j ames St. C l a i r (Ric hard
N ixon's Watergate attorney) of the l arge Boston fi rm Hale and Dorr, and
A l l an Van G estel of Goodwi n , Proctor, and Hoar. They were assi sted by
a team of eight other lawyers .
The presence of the Town of Mashpee among the defendants re­
q u i res explanation . It was not u nti l 1 869 that the com m u n ity l iv i ng i n
Mashpee was accorded formal towns h i p statu s . F rom 1 869 u nti l 1 964
the town government was overwhe l m i ngly i n the hands of I n d i an s . D u r­
i n g th i s period every selectman but one was an I n d i a n or married to an
I nd i a n . Genea logica l evidence presented at the tria l showed that the
fam i l ies of town officers were c l osely i nterre lated . No one contested the
fact that before the 1 9 60s Mashpee was governed by I n d i ans. The d i s­
agreement was over whether they governed as an " I nd i an tri be."
T h i s basi c demogra p h i c and pol itical situation, wh ich had not a l ­
tered drastical ly for over th ree centu ries, was revo l ution ized d u r i n g the
early 1 960s. Before then census figu res showed a popu l ation in Mash pee
fluctuat i n g in the neighborhood of 3 5 0 I ndians and "negroes," "col­
oreds," o r " m u l attoes" (the official categories sh ifted), and 1 00 or fewer
wh ites . A rel iable count of 1 859, wh ich served as a benchmark in the
tri a l , l i sted o n l y one white resident. After 1 960 for the fi rst ti me wh ites
were recorded i n the majority, and by 1 9 70 whites outnu m bered I nd i ans
and other people of color by 982 to 306. By 1 968 two of the town's
selectmen were w h i tes, the t h i rd I n d i a n . Th i s proportion was i n effect at
the ti me of the l awsu it. Mashpee's wh ite selectmen voted th at the town
should lega l l y rep resent the non- I nd i a n majority of property holders who
were th reatened by the l a nd c l a i m .
"Cape Cod's I nd ian Town" had fi na l l y been d iscovered . For centu-

ries a backwater and a curiosity, i n the 1 950s and 1 960s Mash pee be­
came des i rable as a site for reti rement, vacation homes, condom i n i ums,
and l uxury deve lopments . Fast roads now made it accessi ble as a bed­
room and weekend suburb of Bosto n . The new i nfl ux of money and jobs
was first we lcomed by many of Mash pee's I nd ian residents, i nc l ud i ng
some of the l eaders of the land-c l a i m su it. They took advantage of the
new s ituatio n . The town government, sti l l ru n by Indians, enjoyed a
su rge i n tax reve n ues. But when loca l government passed out of I nd i a n
contro l , perhaps for good, a n d a s t h e sca le o f deve lopment i n c reased ,
many I nd i a ns bega n to fee l q u a l m s . What they had taken for granted­
that t h i s was the i r town-no longer held true. Large tracts of u ndevel­
oped land formerly open for h u nting and fish i n g were sudden l y ringed
with " N o Trespassi ng" signs. The New Seabury development, on a
choice stretch of coastl i ne, with its two go lf cou rses and expansionist
p l ans, seemed particu larly egregious. Tensions between trad itional resi ­
dents and newcomers i ncreased, fi nal l y lead i ng t o the su it, fi led w i t h the
su pport of most, but not a l l , of the Ind ians in Mash pee . The land c l a i m ,
wh i le foc u s i ng on a loss o f property i n t h e n i neteenth century, was actu­
a l l y an attempt to rega i n control of a town that had s l i pped from I n d i a n
h a n d s very recently.

Earl Mills

Earl Mills has taught high school in the Falmouth Public School sys­
tem for over twenty-five years . 1 Between 1 952 and 1 967 he lived in Fal­
mouth, ten miles from Mashpee. Mills has ta ught physical education,
hea lth, and social studies . He advises the student council and directs
various other extracurricular activities.
In Mashpee he shares ownership of the town s best restaurant with
his ex-wife, Shirley. He is its principa l cook.
Since the mid-fifties Mills has held the title of Chief Flying Eagle of
the Mashpee Wampanoag Tribe.
On the witness stand he is earnest, engaging, very much the coach
or Boy Scout leader. Forty-eight years old, trim, athletic-looking, he
wears a striped necktie, blue blazer, loafers .
Mills recalls his youth in Mashpee during the thirties and forties. He

1 . The descriptions of persons and places date from the fa l l of 1 9 77. Readers
shou ld bea r in m i nd that i n d ividuals' l i ves have changed si nce then, as have
aspects of the situation i n Mash pee .

was never a s good a h unter a s h is brother, Elwood, s o h e often skipped

the frequent hunting trips . Early on he asked questions and read books .
He questioned his grandmother, "the strong arm" behind his uncle (who
had held the formal title of chief and was "a drifter'), and also h is mother,
the treasurer and tax collector of the town, "the strong arm behind me:'
In the th irties, Mills recalls, some townspeople wore regalia occa­
sionally, and a few spoke a little Indian dia lect. He remembers the festive
atmosphere of a close community-selling corn at town meetings, the
yearly beach outings, the annual herring run .
As a child h e was shown the location o f "Indian taverns:' These were
not drinking places, according to Mills, but just places where paths
crossed. You would pick up a stick, spit on it, and throw it on a heap to
appease spirits in the area .
Mills sa ys he can s till identify two "Indian ta verns;' but most have
long since been cleared a way because the sticks, piled h igh, were a fire
This is the exten t of the Indian rituals Mills reports . Raised a Baptist,
he does not now consider h imself a Christian; but he believes in a crea­
tor, "something greater than me."
Mills says tha t when he inquired after Indian artifacts, especially the
traditional Mashpee pla ited baskets, he was told by his father that "those
fellows up around Cambridge m ust've taken them" (a reference probably
to the Harvard Anthropology Department). His father showed him how
to plait bark baskets, a skill he had acquired as a young man from Eben
Queppish, a master basket maker in Mashpee.
Mills recalls tha t as a boy he made fun of the old-timers, including
the medicine man of the period, William ]ames .
In Fa lmouth High School Mills excelled in athletics . ("You had to be
a scrapper to make it:') Sports were a road to con fidence in a threatening
environment. Outside of school, like h is father and other Mashpee Indi­
ans, he served as a guide for hunting and fishing parties in the region .
Q. : "How was your youth different from that of any sma ll-town
youth ?"
A . : "We were different. We knew we were different. We were told
we were different:'
On ly in the late forties did Mills learn Indian dancing-in the army.
On a lonely even ing during basic tra ining at Fort Dix two comrades, a
Montana Chippewa and a New York Iroquois, performed their dances.
Mills was chagrined to admit that he knew none h imself.
Earl Mills tells about h is five children-four by his first wife, who is

part Navajo, and one by his second wife, who is Caucasian. The eldest,
Roxanne, is married to a Choctaw. Earl Jr. (called "Chiefy'') lives in Fal­
mouth and in recent years has become a champion drummer at various
Indian gatherings and powwows . Shelly, also a fine drummer, attends
Native American festivals all over the Northeast. Robert lives on Com­
monwealth Aven ue in Boston . "He 's into quill work, leather work, skins:'
Nancy, the child of Mills 's second marriage, is now six years old. She
does Indian dances. Her parents agree that she is a Wampanoag.
Mills explains his duties as tribal chief. He teaches beadwork, leath­
erwork, and basketry in Mashpee. Overa ll his job is to be a mediator, to
keep h is people "on balance:'
Under question ing he cannot or will not give any specific examples
of his mediations . Mills tells how in the late fifties and early sixties he
and three whites formed a committee to restore the Old Indian Meeting
House in Mashpee. The meeting house, which had fallen into disrepa ir,
had for many years been the most visible symbol of Indian life in the
town .
During the fifties there had been a tribal constitution of some sort
(the document is introduced into evidence), but Mills testifies that the
tribe did not follow the constitution as written . Tribal meetings were held
irregularly, with notice passed by word of mouth. (Where, St. Clair asks
on cross-examina tion, are the minutes for these purported tribal meet­
ings ?)
In the early seven ties, Mills sa ys, he attended a grant-writing sem inar
at Dartmouth College, along with Amelia Bingham, a state employee
(s ister of john Peters, the tribal medicine man, and Russell Peters, cha ir­
man of the Tribal Council, Inc.). Mills says he had little originally to do
with the land suit. As ch ief he simply approved the action of the incor­
porated body on beha lf of the tribe. It was discussed in his restaurant
Earl Mills testifies that he respects john Peters. The two of them rep­
resent the Mashpee traditionalist wing. The modernists, he says, people
like Russell Peters, are the legal arm of the tribe and represent its interests
in dealings with the government, the courts, and foundations .
St. Clair's questions portra y Chief Flying Eagle as an opportunist fol­
lowing rather than leading h is people. They reveal that Mills's traditional
authority was recently challenged by Russell Peters and others who
wanted to sell beer at the annual Mashpee powwow, a festival attended
by a considerable number of tourists and other outsiders. Over the chief's
I D E NT I T Y I N M A S H PEE 283

objections beer was sold. St. Cla ir harps on this evidence of lack of lead­
ership. Rebutta ls follow, concerning different tribal responsibilities and
roles . There are references to President Carter's inability to control the
(beer-related) behavior of h is brother, Billy.
On the stand Chief Flying Eagle often sounds like a social studies
teacher; his speech is larded with pat anecdotes and homilies .
Only once, toward the end of his testimony, does he do something
unexpected. Asked whether he often wears Indian regalia, Mills answers
no, only at powwows . Then he sudden ly tugs at his necktie, pulling two
thin strings of beads from under h is shirt. One, he says, is turquoise, from
the Southwest. The other small strand was a gift from his father.
Man y people in the courtroom are surprised by this apparently spon­
taneous revelation -surprised and, as Mills stuffs the beads back into h is
sh irt and fumbles to readjust h is tie, a little embarrassed.


At the end of the tri a l Federa l J udge Wa lter j. S k i n ner posed a n u m ­

ber o f specific q u estions t o the j u rors concern ing tri ba l statu s a t certai n
dates i n Mashpee h i story; b u t throughout the proceed i ngs broader q ues­
tions of I n d i a n identity and power permeated the cou rtroo m . Although
the l and c l a i m was formal l y not at issue, the l awyers for N ew Seabury et
a l . someti mes seemed to be playing on a new n i ghtmare . At the door of
you r suburban house a stranger i n a busi ness suit appears . He says he is
a N ative America n . You r land has been i l lega l l y acq u i red generations
ago, and you m ust rel i nq u ish you r home. The stra nger refers you to h i s
l awyer.
Such fears, the th reat of a "giveaway" of private lands, were much
exploited by pol iticians and the press i n the Penobscot-Passamaq uoddy
negoti ations. Actua l l y sma l l h o l d i ngs by private citizens were never i n
danger; o n l y l arge tracts o f u ndeve loped land h e l d by timber compan ies
and the state were in q uestion . In Mashpee the plai ntiffs red uced thei r
c l a i m to e l even thousand acres, form a l l y exc l u d i n g a l l private homes and
lots up to an acre i n size. Large-sca le devel opment, not smal l ownersh i p,
was man ifestly the target; but the i r opponents refused pretrial compro­
m i ses and the ki nds of negotiation that h ad l ed to settlement of the Ma i ne
d i spute .
Accord i n g t o Thomas Tu reen t h e sorts o f land c l a i m s pu rsued i n
Mai ne, Mashpee, Gay Head, a n d Charl estown were always d rastical l y

c i rc u m scri bed . At that h istorica l moment th � courts were re l atively open

to N ative American c l a i ms, a situation u n l i ke l y to l ast. In a decision of
1 985 perm itting Oneida, Mohawk, and Cayuga Non-I ntercou rse Act
su its the Supreme Cou rt made it abundantly clear, in Tu reen 's words,
"that I n d ians a re dea l i n g with the magnan i m ity of a rich and powerfu l
nation, one that is not about to d i vest itself or its non-I nd ian citizens of
large acreage in the name of its own laws . In short, the U n ited States wi l l
perm it I nd i ans a measure o f recompense through the l aw-i ndeed , i t has
done so to a n extent far greater than any other nation i n a comparable
situation-but it u l ti mate l y makes the ru les and a rbitrates the game (Tu r­
een 1 985 : 1 4 7 ; a l so Barsh and Henderson 1 98 0 : 2 89-2 9 3 ) .
Seen i n t h i s l ight, t h e Mashpee t r i a l was s i m p l y a clarification o f the
ru les in an ongo i n g struggle between parties of greatly u neq u a l power.
But beneath the exp l icit fea r of white c itizens losing thei r homes because
of an obsc u re past i nj u stice, a trou b l ing u ncertai nty was find i ng its way
i nto the domi nant i m age of I nd ians in America. The plai ntiffs in the Non­
Intercou rse Act su its had power. In Ma i ne pol iticians lost office over the
issue, and the Mashpee case made national head l i nes for severa l months.
Scandalously, it now paid to be I n d i a n . Acti ng aggressively, tribal groups
were d oin g soph isticated , "nontrad itional" th i ngs . A l l over the cou ntry
they were beco m i n g i n volved in a variety of businesses, some c l ai m i ng
exemption from state reg u l ation. To many whites it was comprehensible
for Northwest Coast tri bes to demand trad itional sal mon-fish i ng privi­
leges; but for tri bes to run h igh-stakes bingo games i n violation of state
l aws was not.
Ind i ans had long fi l led a pathetic i magi native space for the dom inant
c u l t u re ; they were always survivors, noble or wretched . Thei r c u ltures
had been stead i l y erod i ng, at best hanging on in m useu m l i ke reserva­
tions. Native American soc ieties cou l d not by defn ition be dynam ic, in­
ventive, or expansive. I n d ians were lovi ngly remembered i n Edward Cur­
tis' sepia photographs as proud, beautifu l , and "vanish i ng." But Curtis,
we now know, ca rried props, costumes, and wigs, freq uently d ressing up
h i s mode l s . The i m age he recorded was carefu l l y staged (lyman 1 982).
I n Boston Federa l Cou rt a j u ry of white c itizens wou ld be confronted by
a co l l ection of h ighly ambiguous i m ages . Cou ld a group of fou r women
and eight men (no m i norities) be made to be l ieve i n the persistent " I n ­
d ian" exi stence o f t h e Mashpee p l a i ntiffs without costumes a n d props ?
This q uestion su rrounded and i n fu sed the tri a l 's tec h n ical foc us on
whether a partic u l ar form of po l itical-cultural orga n i zation ca l l ed a tribe
had exi sted conti nuously in Mashpee s i n ce the si xteenth centu ry.
I D E NT I T Y I N M A S H P E E 285

T h e i mage o f Mash pee I nd ians, l i ke that o f severa l other eastern

gro u ps such as the L u m bee and the Ramapough, was com p l i cated by
issues of race ( B i u 1 980; Cohen 1 9 74) . Sign ificant i nterma rriage with
blacks had occu rred si nce the m id-eighteenth centu ry, and the Mash pee
were, at ti mes, widely identified as "colored ." I n cou rt the defense oc­
casiona l l y suggested that they were rea l ly blacks rather than Native
America n s . L i ke the L u m bee (and, less successfu l ly, the Ramapough) the
Mashpee p l a i ntiffs had struggled to d isti ngu ish themselves from other m i ­
norities and eth n i c gro u ps, asserting triba l statu s based on a d i stinctive
pol itica l -c u ltura l h i story. I n court they were not hel ped by the fact that
few of them looked strongly " I nd ia n ." Some cou l d pass for black, others
for wh ite.

Hazel Oakley, Hannah Averett

Mrs . Oakley is membership cha irman of the tribe; Mrs. Averett, who
is an active member of the Mashpee Baptist Church, works with Indian
ch ildren in the public schools .
These women wear no Indian jewelry. They speak simply with New
England accents about their childhood experiences, their values, their
parents and grandparents .
They look like what they are: ordinary pillars of the community,
church women .
They describe their activities on behalf of the tribe. Mrs . Oakley has
recently been establishing a membership list. It includes people living
out of town as well as Indians who oppose the suit and who will testify
for the defense in court.
Mrs . Averett looks to be in her fifties . She says that her earliest mem­
ories of community life in Mashpee were the powwows. She also recalls
regular Sunday school picnics at a place ca lled Daniel's Island, attended
by Mashpee Wampanoags and their children . They played games and
sang hymns . Her mother, grandfather, and relatives in the town told her
Indian legends and stories-about Granny Squanett and Mausop and
one about "some Indian maiden that swam in the lake with the trout:'
English was spoken in the family, but Mrs . Averett recalls that some
of her older relatives knew "Wampanoag language:' The only time she
heard her grandfather speak the old tongue was once when his mother
was sick and he held a long con versation with her in her room. Mrs .
Averett's mother sa id to h im, "Dad, why didn 't you tell me you could
speak "indian ?" When he made no reply, Hannah asked, "Grandpa, why

didn 't you tell us you could speak Indian ? Wh y didn 't you teach us ?" He
said, "I just want my children to learn the English language and learn it
as well as they can :'
Mrs. Averett reca lls her mother's herbal remedies-teas and cough
medicines, skunk grease rubbed on for a chest cold-some of which she
still uses .
Mrs . Averett has done housework for a living since she was a girl.
During the war years, however, she went to New Bedford to do defense
work in a Goodyear plant, then to the Boston Naval Yard, where she was
a rope maker. From there she went to the Hood Rubber company, did a
few years in Boston shoe factories, and then went back to housework. In
1 952 she married William Averett. He died in 1 958, and she returned to
Mashpee. "I had two sons to bring up. I felt I could do it better down
there. I felt that if anyth ing should happen to me, my people were there.
If I needed help, my people would be there to help me:'
Three years ago Mrs. Averett joined the Mashpee Wampano?g Tribal
Council, Inc ., and became active in federally funded Indian education
programs. She testifies that the immediate motiva tion for her in volvement
came from her youngest son . He used to take walks in the woods after
school-she didn 't know where. "One day he came in and he asked me,
he said, 'Wh y don 't you do something about this ?'" He explained that he
often went to a favorite spot where deer grazed. "He said, 'It was the
most beautiful sight you could ever see . Now they're putting up a golf
link, and I 'll never see it again ."'
Since 1 9 74 Mrs . Averett has been cha irperson of the Indian Educa­
tion Parent Comm ittee, a federa lly funded Indian education program to
help Mashpee Wampanoag children in the schools. The committee or­
gan izes tutoring, arts and crafts programs, loca l and genera l Indian his­
tory classes, sess ions with Chief Mills and Medicine Man Peters, visits
from other native groups, field trips to the United Nations in New York,
to the Museum of the American Indian, to historic sites in Cay Head.
"Th is is to expand the culture of our people, to see how other tribes, other
people live:'
Mrs. Averett is also chairperson of the board of trustees of the Mash­
pee Baptist Church .
On cross-examination she is asked about possible inconsistencies in
her claim to Indian identity: You don 't eat much Indian food, do you ?
Only sometimes. You use regular doctors, don 't you ? Yes, and herbs as

How do you know your ancestors ? My mother, grandparents, word

of mouth . Have you traced your ancestry? Did you use the 1 859 census ?
(introduced as evidence at the tria l). What about being a devout Baptist
and an Indian ?
Mrs. Averett testifies tha t she respects the medicine man, John Peters
(calling him a "counselor''). She respects Indian beliefs : the Great Spirit,
the land, "grandmother moon and the earth and all those things . . . They
are very dear to me, and I respect them . But I also respect God through
my Christian belief. And to me God and the Great Spirit are the same:'
She has recently received an Indian name, Bright Star, from the med­
icine man. She isn 't sure she is allowed to go into details about her nam­
ing ceremony: there is a pra yer, a circle formed. "Everything is round, as
our lives :'
She likes to see drumm ing and dancing being taught to today's chil­
dren. They do it the way other kids skip rope.

Outside the courtroom during a recess some of the Mashpee women

talk fondly about their children 's Indian activities . "But he's got to cut that
hair!" "If on ly they took care of it, but, you know . : :' "He looks like a
wild man!"

The Sea

Ramona Peters is a co/lege-educa ted woman in her twen ties. After

participa ting in a training program at the Children 's Museum in Boston,
she has recently returned to Mashpee, where she teaches classes in In­
dian language and lore. During her testimony she relates a myth about a
giant who swims over to the Wampanoag tribe at Gay Head on Martha 's
Vineyard. On h is return the giant turns in to Moby Dick, the great wh ite

(I suddenly rea lize that the Indians in th is courtroom are descendants

of Tashtego, the Gay Head harpooner of Melville 's Pequod . This connec­
tion somehow gives everything a strange rea lity and depth grafted now
onto my own literary mythology.)

A good deal of testimony at the trial concerns Cape Cod Indians '
closeness to the sea- long traditions of shellfish ing and work on whaling
vessels in the n ineteenth century. Vernon Pocknett, an activist and

nephew of Mabel Avant, the town 's leading traditionalist and h istorian
during the forties and fifties, tells of a federal CETA program Title 3 grant
to encourage modern aqua-farming by Mashpee Indians.
Would the jury see aqua-farming as a "traditional" activity ?


Mashpee I n d i a n s suffered the fate of many sma l l N ative American

groups who rem a i n ed in the origi n a l th i rteen states . They were not ac­
corded the reservations and sovereign status (stead i l y eroded) of tri bes
west of the M i ssissippi . Certa i n of the eastern comm u n ities, such as the
Seneca and the Semi noles, occupied general ly recogn ized tribal lands.
Others-the L u m bee, for example- possessed no col lective lands but
c l u stered i n d iscrete regions, m a i nta i n i ng ki nsh i p ties, trad itions, and
sporad ic tri bal i nstitutions. In all cases the bou ndaries of the com m u n ity
were permeable. There was i ntermarriage and routi ne m igration i n and
out of the tri ba l center-sometimes seasona l , someti mes longer term .
Aborigi nal languages were much d i m i n ished , often entire l y lost. Rel i ­
gious l ife was d i verse-someti mes Christian (with a d i stinctive twist),
someti mes a transfo rmed trad ition such as the I roq uois Longhouse Rel i­
gion . Mora l and spi ritual va l ues were often N ative American amalgams
com pou nded from both loca l trad itions and pan- I n d i a n sources. For ex­
ample the ritua l and rega l i a at N ew England powwows now reflect Sioux
and other western triba l i nfluences; i n the 1 920s the feathered "war bon­
net" made its appearance among Wam panoag leaders. Eastern Indians
genera l l y l ived i n c l oser prox i m ity to wh ite (or b l ac k) society and in
sma l ler gro u ps than thei r western reservation cou nterpa rts. I n the face of
i ntense pressu re some eastern com m u n ities have managed to acq u i re
offi ci al federa l recognition as tri bes, others not. During the past two dec­
ades the rate of appl ications has risen d ramati cal ly.
With i n th is d i versity of local h i stories and i nstitutional arrangements
the long-term residents of Mashpee occ u pied a gray area, at least in the
eyes of the su rrou nd i ng soc iety and the l aw. The I n d i a n identity of the
Penobscot and Passamaquoddy was never seriously chal lenged, even
though they had not been federal l y recogn ized and had lost or adapted
many of their trad itions. The Mashpee were more problematic. Partisans
of the i r land c l a i m , such as Pau l B rodeur ( 1 985), tend to accept without
q u estion the right of the tri ba l cou nc i l , i ncorporated i n 1 9 74, to sue on
behalf of a group that had l ost its lands i n the m id-n ineteenth centu ry.

They see the q uestion of tribal status as a l ega l red herri ng, or worse, a
calcu l ated p l oy to deny the tribe its b i rthright. However proc rustian and
colon i a l in origin the l ega l defi n ition of tribe, there was nonethe less a
rea l issue at stake i n the trial . Although triba l status and I n d i an identity
have long been vague and pol itica l l y constituted , not j ust anyone with
some native blood or c l a i m to adoption or shared trad ition can be an
Indian; and not j ust any Native American group can decide to be a tribe
and sue for l ost col l ective lands.
I nd ians i � Mashpee owned no tribal lands (other than fifty-five acres
acq u i red j u st before the tri a l ) . They had no s u rvivi ng language, no c l early
d i sti nct re l igion, no bl atant po l itical structu re. The i r kinsh i p was much
d i l uted . Yet they d i d have a p l ace and a reputation . For centu ries Mash­
pee had been recog n i zed as an I nd i an tow n . Its bou ndaries had not
cha nged si nce 1 66 5 , when the land was formal ly deeded to a group
ca l l ed the South Sea I nd i ans by the neighbori ng leaders Tookonchasun
and Weepq u is h . The Mashpee p l a i ntiffs of 1 9 77 cou ld offer as evidence
s u rv i v i n g p i eces of N ative American trad ition and po l itical structures that
seemed to h ave come and gone. They cou ld also poi nt to a sporad ic
h i story of Indian revi va l s cont i n u ing i nto the present.
The Mash pee were a border l i n e case. In the cou rse of thei r pec u l iar
l itigation certa i n u nderly i ng structu res govern i n g the recogn ition of iden­
tity and d i fference became v i s i b l e . Looked at one way, they were I n d i a n ;
seen another way, t h e y were not. Powerfu l ways of looking thus became
i nescapably p roblematic . The tri a l was less a search for the facts of Mash­
pee I n d i a n c u lture and h istory than it was an experi ment i n translation,
part of a long h i storical confl ict and negotiation of " I nd ian" and "Ameri­
can" identities .

(Th i s is how I came to see the Mash pee case, and the account I give of it
reflects my way of see i ng. As a h i storian and critic of anth ropology I tend
to focu s on the ways in w h i c h h i storical stories are to l d , on the alternate
c u ltu ra l models that h ave been appl i ed to h u m a n gro u ps. Who speaks
for c u ltural authenticity? How i s col lective identity and d i fference repre­
sented ? How do peop l e defi ne themselves with, over, and in spite of
others ? What a re the changing loca l and world h i storical cond itions de­
term i n i ng these processes ?
At the Mash pee tri a l these were the k i nds of questions that i nterested
me and that now o rgan ize my accou nt. I am not fiction a l i z i n g or i nvent-

i n g anythi ng, nor am I presenting the whole picture . The rea l ity pre­
sented here is the rea l ity of a s pecific i nterest and field of vision .
I attended most of the tri a l , and I ' ve u sed my courtroom notes as a
gu iding th read . I 've read what has been published about the h istory of
Mash pee and the l itigation , notably F rancis H utc h i ns' Mashpee : The
Story of Cape Cod's Indian Town ( 1 979), Pau l B rodeu r's Restitution : The
Land Claims of the Mashpee, Passamaquoddy, and Penobscot Indians of
New England ( 1 985), and Wi l l iam S i m mons' Spirit of the New England
Tribes ( 1 986). I 've had access to Rona Sue Mazu r's Ph . D . thesis in an­
thropol ogy at Col u mbia U n iversity, "Town and Tri be i n Confl ict: A Study
of loca l -level Pol itics in Mashpee, Massac h usetts" ( 1 980) . And I have
con s u l ted the tri a l record . But I haven't systematica l l y i nterviewed par­
tici pants or done firsthand researc h in the arch i ves or in Mash pee.
It shou l d be c lear from what fol lows that I am portraying primari ly
the tri a l , not the complex l i ves of Indians and other eth n i c groups i n
Mashpee. Sti l l , i n t h e process I make strong gestu res toward truths m issed
by the dom i nant categories and stories in the cou rtroom . Th u s I i nvoke
as an absence the rea l ity of Mashpee and particularly of its Indian l ives .
I do th i s to mai nta i n the h i storical and eth nograph ic seri ousness of the
accou nt, a seriousness I wish both to assert and to l i mit.
I accept the fact that my vers ion of the tri a l , its wi tnesses, and its
stories may offend people on several sides of the i ssue. Many i nd ividual
positions are more complex than I have been able to show. My account
may be objectionable to Native Americans for whom cu lture and trad i­
tion are cont i n u ities, not i nventions, who fee l stronger, less compro­
m i sed ties to aboriginal sou rces than my analysis a l l ows. For them th is
version of " identity i n Mashpee" may be about rootl ess people l i ke me,
not them .
It is, and is not o n l y, that.
When I report on witnesses at the tri a l , the i m pressions are m i ne .
Others I spoke with saw thi ngs d ifferently. T h e trial record-wh ich sten­
ograph i ca l l y preserves, by a prec ise but not i nfa l l i ble tec h n iq ue, the
mean i ngfu l , s po ken sou nds of the tria l-provides a check on my im pres­
sions. It does not, of course, provide much information on the effect of
witnesses or events in the cou rtroo m . It omits gestu res, hesitations, c l oth­
ing, tone of voice, laughte r, i rony . . . the someti mes devastati ng si­
I offer vignettes of persons and events i n the cou rtroom that are ob­
viously composed and condensed . Testi mony evoked in a page or two

may ru n to h u n d red s of pages i n the tra nscri pt. Some witnesses were on
the stand for several days. Moreover, rea l testi mony al most never ends
the way my vignettes do; it tra i l s off in the q u i bb l es and corrections of
red i rect and rec ross-exam i nation . Wh i l e I have i nc l uded for comparison
a verbati m excerpt from the transcript, I have genera l l y fol lowed my
cou rtroom notes, checked aga i nst the record, and h ave not hesitated to
rearrange, select, and h i g h l ight. Where q u otation m arks a ppea r, the
statement i s a fa i r l y exact q u otatio n ; the rest i s paraph rase .
Overa l l , if the witn esses seem flat and somewhat e l u sive, the effect
is i ntentiona l . U s i n g the usual rhetorical tec h n i q ues, I cou ld have given
a more i nti mate sense of peoples' personal ities or of what they were
rea l l y try i n g to express; but I h ave preferred to keep my d i sta nce. A c o u rt­
room i s more l i ke a theater than a confess ional .
M i strustfu l of tra nspa rent accou nts, I want m i ne to man ifest some of
its frames and angles, i ts wavelength s . )

John Peters

Peters is about fifty years old. He wears a sport jacket with a turtle­
neck swea ter. He is graying, dignified, and appears somewhat taciturn .
He speaks with broad New England vowels .
Peters is medicine man of the Mashpee Wampanoag Tribe.
He testifies that whereas he and Earl Mills, Ch ief Flying Eagle, are
leaders of the tribe, they are not alone. All elders are leaders . Women are
leaders : Mary Lopez, Hannah Averett ("in the education field"), Mrs .
Mills (Earl Mills 's mother), Hazel Oakley.
He describes a growing interest in Native American religion . Asked
about traditiona l ritua ls in Mashpee, he reca lls participating in the peace
pipe ceremony, offerings to the Great Spirit, "for as long as I can remem­
Newpaper notices from 1 93 6 are introduced as evidence. They re­
port tribal meetings and "Indian Da y" services at the Old Indian Meeting
House. Peters remembers these services. He and his father were in the
choir. Reverend Redfield (the Baptist m inister) and William james (the
medicine man) both participated. Christian hymns were sung. The peace
pipe was smoked.
Peters tells of his youth and training as a medicine man . There were
no specific ceremonies or rites of passage. An old man, Russell Mingo,
talked at Saturday dinners about Indian matters . "We kids didn 't under-

stand:' In his late teens Peters approached William james with questions,
and he learned "things about Mother Earth :' Neither james nor his par­
ents ever spelled much out. "A medicine man doesn 't force th ings
on you:'
Peters is a strong, understated presence. Also a savvy witness. Under
cross-examination he pauses for a long time after hostile or baited ques­
tions. Often he slowly repeats the question before answering, turning it
over in public view.
Peters reca lls his school days . The Mashpee were called "thieving
Indians" or "Womps" (from Wampanoag). The latter was not always pe­
jora tive, he adds . The school basketball team was called the Womps, the
cheerleaders the Wompettes.
On cross-examination Peters is reminded that the traditional cere­
mon y "Indian Day" was proclaimed by the governor of Massachusetts,
who was not an Indian . Reverend Redfield was not an Indian either. And
smoking the peace pipe was not limited to Indians. The Mashpee parish
and the Old Mashpee Meeting House Corporation all depended on non­
Indian participants. How can he cla im, as he does, that they are "arms of
the Mashpee Tribe" ?
Q. : What does it take to be a member of the Mashpee Tribe ? A . :
Tracing ancestry back to your great-grandfather or great-grandmother.
How do you know? We know each other. Who was your great­
grandmother? (Peters cannot recall her first name.) Was she an Indian ?
Yes. Have you traced this specifically? No. Who was your great­
grandfather? Charles Peters . Wasn 't he a preaching Indian from Martha 's
Vineyard? Yes . Not a Mashpee Wampanoag Indian ? No .
Peters is asked about the years he has lived away from Mashpee
while still serving as medicine man. In 1 9 64 he worked for more than a
year in Hawaii as a private detective. Between 1 973 and 1 976 he was
based on Nantucket. He testifies that from the latter residence he was
able to stay in touch . He employed Mashpee people in his various busi­
ness ventures .
With a brother and nephew he is curren tly part owner of Peters Fuel
Oil. The concern owns two trucks and does a moderate business. Q. :
This is a priva te, not a comm unal tribal business ? A . : Yes .
Peters testifies about a /and-development plan proposed some years
back by a company he founded with various family members, the Ashers
Path Development Corporation . He is confronted with blueprin ts for the
development that was subm itted to the town but never carried through .
The blueprints show an extensive subdivision into thirty-seven Jots .

Peters is reminded of h is years in business as a general contractor

and of making bids on projects for the New Seabury Corporation (now a
symbol to many Indians of excessive development). He is unclear about
whether he has since left contracting and rea l estate.
When pressed on the con flict between business activities and his
role as medicine man, he comments that the art of making money is
probably inconsisten t with being an Indian; but a ll Mashpee Indians
make money. It depends on how you do it.
He is asked, Was what you were doing as a developer a destruction
of the earth ? Exactly what part of your business was consisten t with In­
dian values ? Peters offers no clear answers . But, he adds, "if I had devel­
oped that plan, I'd have been violating my principles:'
Peters testfies that until recently he didn 't spend enough time as
medicine man counseling people. Early in his career he was much less
conscientious than he is now. During the last five years he has changed
Peters is asked about current Indian religious practices in Mashpee.
He can specify no customary rites at birth or at puberty. He has himself
been married and divorced twice without any special Indian ceremony.
He was married in a Baptist church by a wh ite minister. Asked whether
it was a typica l Baptist ceremony, Peters replies that he can 't say, since
all Ch ristian marriages seem the same to h im .
How many "followers o f traditional religion" are there now in Mash­
pee ? Peters won 't guess and questions the word fol lowers. It's not a ques­
tion of authority, he says, of leaders and followers . A person can be very
religious without worshipping in a building or participating in a formal
Who are the "traditionalists" in Mashpee ? It's hard to be sure. Ma ybe
there are a hundred or so. His brother Russell (chairman of the Tribal
Council, Inc.) is not a traditionalist. As for h is other brothers and sisters,
he hasn 't formed an opinion .
Peters testifies that the Supreme Sachem, Elsworth Oakley, a Mash­
pee, has recently designated him Supreme Medicine Man of the Wam­
panoag Nation. He has passed his local duties to a young man, Skip
Black. "He's been in my mind for some time:'
Peters sa ys tha t he and Oakley are currently helping groups of Wam­
panoags in New Bedford and Brockton to form tribal structures .
Q . : "Do the new tribes you 're c reating there have any plans to bring
lawsuits ?"
A . : "You mean, taking over New Bedford and Brockton ?"

History I

The case aga i nst the plai ntiffs was stra ightforward : there never had
been an I ndian tri be in Mas h pee. The com m u n ity was a creation of the
colon ial encounter, a co l lection of d i s parate I ndians and other m i norities
who sought over the years to become fu l l c itizens of the Com monweal th
of Massac h u setts and of the Repu b l i c . Dec i m ated by d i sease, converted
to Christian ity, des i rous of freedom from patern a l i stic state tutelage, the
people of m i xed I nd i a n descent in Mashpee were progressively ass i m i ­
lated into American society. The i r I n d i a n identity had been lost, over and
over, si nce the mid -seventeenth centu ry.2

The plague . When the English P i l grims arrived at Plymouth i n

1 620, they fou nd a region devastated by a d i sease brought b y white sea­
men . The settlers wa l ked i nto empty Indian v i l l ages and planted in a l ­
ready c l eared fields. T h e region was seriously u nderpop u l ated . I n the
years that fol lowed Pu ritan leaders l i ke Myles Standish pressed stead i l y
t o l i m it I nd ian territories and t o estab l i sh c l ear "properties" for t h e grow­
i ng n u mber of newcomers . Misu nderstandi ngs i nev itably ensued : for ex­
ample wh ites c l a i med to own u noccupied l and that had been ceded to
them for tem pora ry use.
Richard Bourne of Sandwich, a farmer near what is now Mashpee
Pond and a tenant . on I n d i a n lands, studied the l anguage of h i s land lords
and soon became an effective med iator between the soc ieties. He was
friend l y to the a rea's i n habitants, remnants of earl ier grou ps, who came
to be cal l ed South Sea Indians by the settl ers to the north . He be l ieved
that they needed protection; beco m i ng thei r advocate, he negotiated for­
mal title to a large tract adj oi n i ng h i s farm (wh ich in the meanti me he
had managed to p u rc hase) . H i s a l l y in these transactions was Pa upmun­
nuck, a leader o f the nearby Cotac hesset.

2. The two "h i stories" that fol l ow represent the best brief i nterpretive ac­
counts I cou ld construct of the contending versions of Mashpee's past . They draw
selectively on the expert testi mony presented at the trial-testimony much too
long, complex, and contested to summarize adeq uately. The overa l l shape of the
two accounts reflects the sum mation provided at the end of the testimony by each
side's princ i pal attorney. " H i story I" owes a good deal to Francis H utch i ns' book
Mashpee: The Story of Cape Cod's Indian Town ( 1 979). This book takes a some­
what more moderate position than the courtroom testimony on wh ich it is based .
" H i story I I " owes someth i ng to the genera l approach of James Axte l l 's book The
European and the Indian ( 1 98 1 ) . Axte l l was witness for the plaintiffs.

Bourne's "South Sea I nd i an P l antation" was to become a refuge for

Ch ristian converts, for as white power i n c reased , it became i ncreasi ngly
dangerous for I nd i a n s to l ive around Cape Cod u n l ess they came together
as a com m u n ity of "pray i n g I n d i an s ." U nder Bourne's tutelage the Mash­
pee p l antation was a center for the fi rst I ndian c h u rch on the Cape, or­
gan i zed i n 1 666.
Th u s Mashpee was origi n a l l y an artific i a l comm u n i ty, never a tribe.
It was created from I nd i a n su rvivors in a n area between the traditional
sachemdoms of Manomet and N a uset-the former centered on the p res­
ent town of Bou rne at the Cape's western edge, the l atter near its tip.

Conversion to Christianity. Bad l y d i sorga n i zed after the p l ague

and confronted by a grow i n g n u m be r of determi ned sett lers, the Cape
Cod I nd i ans made accom modations. L i ve and l et l i ve was not the Purita n
way, espec i a l l y once thei r power h a d been conso l idated . Tensions a n d
confl icts grew, l ead i n g t o w a r i n 1 675 with t h e forces of t h e Wam panoag
S u preme Sachem Metacomet ( " K i n g Ph i l i p" ) . After Metacomet's defeat
I nd i an s who sympath i zed with h i m were expe l led from thei r lands.
Many, i n c l u d i ng some who had remai ned neutra l , were sold i n to s l avery.
The price for l iv i n g on ancestra l lands i n easte rn N ew England was
cooperation with w h i te soc iety. The Mash pee, u nder Bourne's tutel age,
became model Ch rist i a n s . By 1 6 74 n i nety Mash pee i n h abitants were
cou nted as baptized , and twenty-seven were ad m i tted to fu l l comm u ­
n io n . T h e " pray i n g I nd ians" were enteri ng a n e w l ife. They stopped con­
s u l t i n g " powwows" (med i c i ne men, in seventeenth-century usage); they
respected the Sabbath and other holy days, severed ties with " pagans,"
a l te red c h i l d- rea r i n g p ractices, d ressed i n new ways, washed d i fferently.
The changes were grad u a l but te l l i ng. They reflected not o n l y a tactical
accom modation but a l so a new be l i ef, born of defeat, that the powerfu l
w h ite ways m ust be superior. When Bourne d ied i n 1 68 2 , h i s successor
as Protestant m i n i ster was an I nd i a n , Si mon Popmonet, son of Bourne's
o l d a l l y Pau pm u n n uck. Th is was a fu rther s i gn that the I ndians were wi l l ­
ingly givi ng u p their o l d ways for the new fa ith .

"Plantation" status. O nce the South Sea I nd i a n Plantation had

been establ i shed, its i n habitants' c l a i m to the i r land rested on a written
deed and on E n g l i s h law rather than on any aboriginal sovereignty. L i ke
other " p l a ntations" i n N ew England, the com m u n ity at Mash pee was a

joi nt-owners h i p arrangement by a group of "proprietors." U nder Engl i sh

law proprietors were l i censed to develop a vacant portion of land, re­
serving part for com mons, part for the c h u rch, and part for i ndividual
hol di ngs. All transfers of l and were to be approved col lectively. Th i s
p l antation-proprietary form, as appl ied to early Cape Cod settl ements
such as Sandwi c h and Barnstable, was i ntended to evol ve q u ickly i nto a
townsh i p where freemen held ind ividual private property and were rep­
resented in the General Cou rt of the col ony. The white pl antations
around Mashpee did evo l ve d i rectly i nto towns. F rom the late seven­
teenth century on the i r common lands were converted i nto private i n d i­
v i d u a l holdi ngs i n fee s i m p l e . Mash pee fo l l owed the same cou rse, but
more s l ow l y. As l ate as 1 830 its lands were the j o i nt property of proprie­
tors .
For com p l ex h i stori cal reasons Mash pee's progress toward fu l l citi­
zensh i p l agged a l m ost two centuries beh i nd that of its neighbors . An
end u r i n g prej udice aga i nst Indians and the i r supposed l ack of "civi l ity"
certa i n ly p layed a part, for d u ri ng the early and m i d-ei ghteenth centu ry
the I nd ian p l antation was governed i n h u m i l iating ways by white "guard­
ians." Nonetheless, development toward a utonomy, wh i le delayed , d i d
occur. I n 1 76 3 , after a d i rect appeal to K i n g George I l l , Mash pee won
the right to i n corporation as a d i stri ct, a step on the road to town s h i p
status and a l i beration from oppressive medd l i ng b y wh i te outsiders .
Then, begi n n i ng i n 1 834 and c u l m i n ating i n 1 8 70, a series of acts o f the
Massachu setts legi s l ature changed the Mash pee plantation i nto an i n cor­
porated tow n . Its i n habitants had overcome the prej udice and paternal­
ism that had so long hem med them in. They were now fu l l -fledged ci ti­
zens of Massach u setts .

Taking the colonists' side . From early on the I ndian i n habitants

of Mash pee gave s i gns of active identification with the new white society.
During King Ph i l i p's War a ce rta i n Capta i n Amos, probably a Nauset from
near Sandwich, led a gro u p of I nd ians aga i nst Metacomet. Amos became
a pro m i nent i n habitant of Mashpee after the confl ict ended . A century
later the d i strict of Mash pee sent a conti ngent to fight in the Revo l ution­
ary War aga i nst the Britis h , a com m itment of troops even greater than
that of the surround i n g wh ite towns. Rel iable accou nts estimate that
about half the ad u lt male popu lation d ied i n the war. A Mash pee I nd ian,
Joshua Pocknet, served at Va l l ey Forge with George Wash i ngto n . At these
critica l moments, therefore, the descendants of the South Sea I n d i ans

showed someth i n g more than simple acq u i escence u nder co l o n i a l rule.

The i r enthu s iastic patriotism strongly s uggests that they had identified
with white society, rel i n q u i s h i n g any sense of a separate tribal pol itical

Intermarri age. Mash pee's popu l ation showed two significant pe­
riods of expansion . D u r i n g the 1 660s and 1 6 70s there had been an i nfl ux
of I ndi an s