http://www.jstor.org
Presence and Representation: The Other and Anthropological Writing
Author(s): Johannes Fabian
Source: Critical Inquiry, Vol. 16, No. 4, (Summer, 1990), pp. 753-772
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/1343766
Accessed: 09/07/2008 23:14
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=ucpress.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.
Presence and Representation: The Other
and Anthropological Writing
Johannes Fabian
In discussions of the Orient, the Orient is all absence, whereas one
feels the Orientalist and what he says as presence; yet we must not
forget that the Orientalist's presence is enabled by the Orient's
effective absence.
-EDWARD W. SAID, Orientalism
Representationsand Presence
It is not without significance that in anthropological and sociological
usage representation most often appears in the plural. The singular
would put the emphasis on representation as an activity or process.
Instead, by privileging the plural, we invoke entities, products of knowledge
or culture. That this is not merely a matter of practicality-of
devising terms that best fit the analytical tasks to which we put themwill,
I hope, become clear from the reflections that follow.
Taken as a philosophical issue, the idea of representation implies
the prior assumption of a differencebetween reality and its "doubles."
This essay was developed from a contribution to a panel on "Othering: Representations
and Realities," organized by Smadar Lavie, Kirin Narayan, and Renato Rosaldo for
the 85th Annual Meeting of the American Anthropological Association in Philadelphia,
December 1986.
In an earlier version, I wrote at this point: "I want to thank my friend and colleague
Bob Scholte for his untiring readiness to be a partner in discussions and to share his
knowledge of the literature." I now dedicate this essay to his memory.
CriticalInquiry 16 (Summer 1990)
? 1990 by The University of Chicago. 0093-1896/90/1604-0006$01.00. All rights reserved.
753
Presenceand Representation
Things are paired with images, concepts, or symbols, acts with rules
and norms, events with structures. Traditionally, the problem with
representations has been their "accuracy," the degree of fit between
reality and its reproductions in the mind. When philosophers lost the
hope of ever determining accuracy (and thus attaining Truth), they
found consolation in the test of usefulness: a good representation is one
that works. The proof of its working is that it enables us to act on the
world together.' In such a frame, science, including anthropology, is
conceived as the pursuit of privileged representations, privileged in
that, by their nature or by their combination, they establish knowledge
of a special kind. In the case of anthropology, "culture" has served as a
sort of umbrella concept for representations. The structuralists have
been most explicit about the need to think of representation in the
plural, but their position is shared, in varying degrees, by all those who
conceive of (cultural) knowledge as the selection and combination of
signs in systems, patterns, or structures, in short, as some kind of
conceptual order ruling perceptual chaos.
The postulate of a difference between reality and its doubles
generates another assumption of difference, or rather, distance: that
between the knower and the known. This comes naturally, as it were,
with conceiving of the (scientific-philosophical) knower as a viewer and
observer. It is reinforced by the idea that to know scientifically is to
interpose a system of concepts (a method or a logic) between reality and
1. Rememberthe connectionbetween the Kantianquest for syntheticformsand
EmileDurkheim'sideaof collectiverepresentationssustainedby the moralauthorityof a
society.Durkheimcertainlywasone to look for the "ethic"in the "ethnic"-primitive,and
it makesme wonderwhetherStephenA. Tyler'scharacterizationof postmodernethnographyas
a returnto "anearlierandmore powerfulnotion of the ethicalcharacterof all
discourse,as capturedin the ancientsignificanceof the familyof terms'ethos,' 'ethnos,'
'ethics"' might not signal a return to the Durkheimian fold (Tyler, "Post-Modern
Ethnography:From Documentof the Occult to Occult Document,"in WritingCulture:
ThePoeticsand Politicsof Ethnography,ed. James Clifford and George E. Marcus [Berkeley
andLosAngeles, 1986],p. 126).
Johannes Fabian is professor of cultural anthropology at the
University of Amsterdam. His publications includeJamaa: A Chdrismatic
Movementin Katanga (1971), Timeand theOther:HowAnthropologyMakes
Its Object(1983), and Language and Colonial Power:The Appropriationof
Swahili in the FormerBelgian Congo, 1880-1938 (1986). Two books will
appear in 1990: Historyfrom Below: The Vocabulary of Elisabethville by
Andre Yav, a commented edition/translation of a colonial history written
in Swahili by the colonized for the colonized, and Power and
Performance,a study of conceptions of power through popular wisdom
and theater in Shaba/Zaire.
754 Johannes Fabian
Summer1990 755
the mind. If I had to specify anthropology's special contribution to
debates about the nature of knowledge, I would say that reflecting on
the nature of ethnography has led us to abandon naive faith in distance.
One form this realization may take is to reject the givenness of the
Other as the object of our discipline. Awkward and faddish as it may
sound, othering expresses the insight that the Other is never simply
given, never just found or encountered, but made. For me, investigations
into othering are investigations into the production of
anthropology's object. In Timeand theOther:How AnthropologyMakesIts
Object(1983), I tried to show connections between representation as an
idea central to various kinds of semiotic and symbolic approaches and
the creation of distance. I have developed the argument against
"representationism" (not to be confused with a theory of representation)
more fully in my later essay, "Culture, Time, and the Other"
(1985).2 Of course, our malaise regarding distancing as both a prerequisite
and a possibly autodestructive device in anthropological writing
needs to be explored and argued further. In this paper I attack representation,
but I urge readers also to see Marilyn Strathern's thoughtful
defense of representation in a review of Vincent Crapanzano's Waiting:
The Whitesof SouthAfrica.s Strathern argues a point that is also formulated
succinctly byJean-Paul Dumont:
The others are not any more present in the text the anthropologist
offers to her readers than their voices are present in (or even
behind) the phonetic transcription of their utterances. In fact, it is
because of this, because of this phonic and existential absence that
these others can in the end be represented by the anthropologist,
for, if they were here, there would be no point in representing
them, that is, to stand for them and to speak for them.4
Perhaps it is possible to continue the debate (without simply
excluding convinced "representationists") if one locates the problem
with representations not in a difference between reality and its images
but in a tension between re-presentation and presence.First of all, this
would help to revindicate the primacy of experience as something that
requires presence (as sharing of time and place). Second, it would stress
the processual and productive nature of representation, not in the weak
sense of the term in which processcan signify any activity or sequence
2. See Johannes Fabian, Timeand the Other:How AnthropologyMakesIts Object(New
York, 1983) and "Culture,Time, and the Objectof Anthropology,"BerkshireReview20
(1985):7-23.
3. Marilyn Strathern, "Intervening," review of Waiting:TheWhitesofSouthAfrica,by
VincentCrapanzano,CulturalAnthropology2 (May1987):255-67.
4. Jean-PaulDumont,"Prologueto Ethnographyor Prolegomenato Anthropography,"Ethos14(Dec.
1986):359.
CriticalInquiry
756 Johannes Fabian Presenceand Representation
occurring in time and production anything that precedes what one
chooses to take as a result or outcome, but in the strong sense of transforming,
fashioning, and creating.5 In other words, we anthropologists
should perhaps not think of representation in the first place as some
enabling capacity of the human mind (although investigations of such a
capacity remain legitimate for psychologists, brain scientists, and
philosophers) but, more modestly, as something that we actually do, as
our praxis. This would help us to realize that our ways of making the
Other are ways of making ourselves. The need to go there (to exotic
places, be they far away or around the corner) is really our desire to be
here (to find or defend our position in the world). The urge to write
ethnography is about making the then into a now. In this move from
then to now the making of knowledge out of experience occurs. Both
movements, from here to there and from then to now, converge in
what I called presence. This is the way I would define the process of
othering.
I may be accused of spreading confusion when, in one place, I
oppose representation to praxis, and then talk about representation as
praxis, or the praxis of representation, and so forth. I am afraid I know
of no less ambiguous way of talking about the issue that, in my mind, is
"representationism"-roughly the philosophical stance, together with
its hegemonical claims, which also is the object of Richard Rorty's
critique (and not only his; I only cite him as a convenient example).6
The aim is to assign to representing a more modest, less imperial, place
in a theory of knowledge, not to exorcise by decree what is obviously an
important human capacity.7 Incidentally, if representation has to do
above all with power, then it may not only be thought of as praxis but it
is praxis. All the attempts to make it unassailable by declaring it essentially
theoretical would then be so many practical moves designed to
preserve its hegemony.
If representation is thought of above all as praxis, this has two
consequences: (1) The foremost problem with it will not primarily be
accurate reproduction of realities but-how shall we call it?-repetition,
reenactment. (2) Representations (in the plural) will then be
5. Dean MacCannell and Juliet Flower MacCannell's Time of the Sign: A Semiotic
InterpretationofModernCulture(Bloomington, Ind., 1982) is an intelligent and thoughtful
recent defense of representation (as semiosis). Their proposal to redefine semiotics as
"the study of the meansoftheproductionof meaning" moves them closer to a conception of
representation as praxis (MacCannell and MacCannell, The Timeof theSign, p. 9). But not
close enough, since they are not ready to abandon the idea that semiotics (legitimately)
occupies a central, privileged position in their theory of knowledge.
6. See Richard Rorty, Philosophyand theMirrorofNature (Princeton, N.J., 1979).
7. And an important aspect of all scientific practice. See Human Studies:AJournal
for Philosophyand theSocialSciences11 (Apr./July 1988), a special issue, "Representation in
Scientific Practice," ed. Michael Lynch and Steve Woolgar.
CriticalInquiry Summer1990
considered as acts or sequences of acts, in short, as performances.
Performances need actors and audiences, writers and readers. Therefore,
representations ought to draw their convincingness primarily
from communication, rhetoric, and persuasion, and only secondarily
from systemic fits or logical proofs. The logic of symbols, to name a
conceptualization of knowledge much in favor among anthropologists,
is much more a matter of Socratic persuasion in conversation than of
Platonic appeal to transcendent truths.
Representationand theStrangeStoryof "Ethnography"
Before I return to othering I shall permit myself a slight digression.
I may have to stand corrected, but it is my impression that recent
awareness of textuality, authorship, style, and so forth has not gone
very far in critically examining our seemingly mindless, but actually
quite revealing, ways with the word ethnography.
Ignorance of Greek may have helped, but it is hardly the only
explanation for general insensibility to terminological misfits such as
"ethnographic writing." At best it is an awkward pleonasm (because
writing is already contained in the Greek graphein); at worst it signals a
shift to a merely indexical use of the qualifier "ethnographic." It
doesn't mean a thing; itjust indicates the place of a kind of writing in a
taxonomy of writing. But that is not the whole story. Without feeling
much discomfort we also use expressions such as "ethnographic collections/exhibits/museums"
(which are institutions that neither write nor
usually exhibit writings). And what about "ethnographic observation,"
"ethnographic data," and "ethnographic subjects"? Easy, one may
respond, these are observations, data, and subjects written about in the
manner of ethnography. While they are less than elegant, these expressions
ought not to be more offensive than such accepted compounds as
an "ethnographic publication" or an "ethnographic authority."
At any rate, it might be argued further, we can avoid the semantic
tangle if we simply accept as a fact that in current technical usage
ethnographyand ethnographictake their meaning in contrast to theoryand
theoretical,roughly in the manner our French colleagues have come to
oppose ethnographyto ethnology.So we seem to have returned to a purely
indexical or classificatory function of the term. Perhaps, but the layman
who picks up a book or visits an exhibition designated as ethnographic
may be ignorant of disciplinary classifications; his attention is attracted
by the lure of connotations-ethnographic evokes the exotic. In more
prudish times it promised (and delivered) bare bosoms and frightening
fetishes; today it satisfies a new aesthetic of the fantastic and imaginary
among some, and the need for "inside knowledge" of the third world
among others. As Roland Barthes has observed, the signified of conno-
757
758 Johannes Fabian Presenceand Representation
tation is "atonce general, global and diffuse; it is, if you like, a fragment
of ideology."8 Whenever in the use of "technical" terms the balances of
signification tip heavily on the side of connotation, ideology, obfuscation,
and plain intellectual fraud loom large, whereby the blame is first
of all ours, not the layman's.
But we can do more than cast suspicion. Without pretending to do
what should be the work of a careful conceptual history (in several
languages), we can state that the term ethnographyacquired its present
free-floating character as a result of certain disjunctions that still pose
problems and are the reasons why we worry about the meaning of the
term in the first place. Summarily, these disjunctions can be stated as
follows:
1) Assuming that the term ethnographywas coined following models
such as cosmographyor geography,its "innocent" meaning would simply
be "description of peoples."9 Except that by the time the word was
coined, ethnoi had long since lost its innocence as a neutral term and
taken on an evaluative notion (somewhat like the Latin gentes,which, to
put it mildly, meant "non-Christians" at the height of the Middle Ages).
More important, the element of de-scription, writing about, had from
the beginning a nominal slant, suggestive less of the activity of writing
than of its products: descriptions, tableaux, in short, representations.
2) When, at a much later time, description of peoples became a
professional activity, when field research became an accepted and even
required practice, the disjunction of ethnography from writing was so
advanced that one could, without flinching, designate observation on
the spot as "doing ethnography." Bronislaw Malinowski, the man on the
spot, was considered an ethnographer; Sir James George Frazer, the
writer in his armchair, not really. Yet Malinowski, as James A. Boon
8. Roland Barthes, Writing Degree Zero and Elements of Semiology,trans. Annette
LaversandColinSmith(Boston,1970),p. 91.
9. A quick check of a few dictionaries confirms this. TheShorterOxfordEnglish
Dictionary,3d ed., 2 vols.(Oxford, 1975)datesthe introductionof the termethnographyin
1834anddefinesitas "thescientificdescriptionof nationsor racesof men,theircustoms,
habits,anddifferences"(s.v."ethnography,"1:685).Ethnologyappearsin 1842 andis "the
sciencewhichtreatsof racesandpeoples,their relations,theirdistinctivecharacteristics,
etc."(s.v. "ethnology,"1:685-86). The compiler'sdifficulty in distinguishingbetween
-graphyand -logy is obvious.The use of etc.suggeststhat he abandonedthe quest for a
precise statement. In a modern French dictionary, Lexis-Larousse:Dictionnairede la langue
frangaise(Paris,1975),ethnographieis datedin 1823 anddefinedas "etudedescriptivedes
ethnies"(s.v."ethnographie,"p. 655). Notice that whilethis formuladesignatesan activity,the
referenceto writingisfoundin anadjectivemodifyingstudy.Ethnologieappeareda
decadelater(1834) and is describedas "sciencequi a pour objet l'etudedes characteres
ethniques,en vue de degagerdes loisgeneralesdes societeshumaines"(s.v."ethnologie,"
ibid.).
CriticalInquiry Summer1990
reminds us, emulated Frazer's writing in his early ethnographies
(before he became scientific and boring).'0
3) The disjunction between ethnography and writing has, of
course, also been at the bottom of making a distinction between
research and write-up that was masked as a mere sequence but was in
fact a ritual dramatization of spatial distance between the sites of observation
and the places of writing.
4) The process of disjunction had its apotheosis when professionalization
of anthropology reached its peak (probably in the 1950s). Two
things happened:
a) In common usage, ethnography had become synonymous with
empirical research and data collection. As such, it was opposed to
theory. This was expressed most openly in course requirements and
reading assignments; it was, to my knowledge, not often discussed in
the literature except in reviewers' complaints about too much or too
little ethnography in a given publication. It also gave rise to a curious
sort of double-talk: praise was heaped on "theory"; dissertation proposals
and Ph.D. theses had no chance of passing if they were considered
short of theory, yet the same judges could be caught nostalgically
invoking the great ethnographies of the past.
b) However, accuracy demands that we note the appearance of an
avant-garde of young American anthropologists in the fifties who registered
discontent with this mindless devaluation of ethnography. They
became the inventors of "new ethnography." The old opposition
between theory and ethnography was abolished and ethnography itself
was declared a theoretical enterprise. Great efforts went into attempts
to provide ethnography with rules (such as in the "ethnography of
speaking") and more formal and sophisticated methods (in the various
kinds of ethnoscience)." The former branch of the movement-Dell
Hymes and a few others-was eventually led to consider anthropological
writing itself important, inevitably so, given the attention paid to
texts, rhetoric, style, genres, and so forth. The ethnoscientists had put
their bets on schemata, tables, and graphs, and algebraic representation.
By the time they discovered that there is no way to get around
10. See James A. Boon, Other Tribes, Other Scribes: SymbolicAnthropology in the
ComparativeStudyofCultures,Histories,Religions,and Texts(Cambridge, 1982), pp. 9-21.
11. An example of this new attitude is Harold C. Conklin's entry for "Ethnography"
in TheInternationalEncyclopediaof theSocialSciences,18 vols. (New York, 1968), 5:172-78,
an important text given its task to formulate a succinct contemporary definition of the
term. That the definition was a matter of a group defining what it does or thinks ought to
be done can be seen if Conklin's statement is compared to the article that follows it,
Harold E. Driver's "Ethnology" (in ibid., 5:178-86). The difference is much less systematic
than historical. Driver represents the view of a different generation.
759
760 Johannes Fabian Presenceand Representation
texts, the discussion on ethnographic writing was already in full bloom,
and "new ethnography" looked old. Nevertheless, I find in current
writings about ethnographic writing a curious lack of acknowledgment
of that "turn to language," brought about by both ethnographers of
communication and the ethnoscientists. It is to them, above all, that we
owe our present critical position. Or, to put it somewhat differently,
had we not been prepared by the "turn to language," the reception of
hermeneutics (Paul Ricoeur, Clifford Geertz), critical theory (Jiirgen
Habermas, for instance), and literary and historical "deconstructionism"
would probably not have had the impact they have had.
5) From an outside point of view, which anthropology acquired
under the influence of literary theory and criticism, the disjunction
between ethnography and theory was not just a matter of assigning
different hierarchical status to different sorts of activities (which could
be taken as a sign of anthropology reaching maturity as a science). It
was expressive of a differentiation of genres of writing or, as some
observers have stated, of a generic crisis.
Whatever else this literary turn accomplished, it restored writing
to its central position in debates about the nature of ethnography, and
this occurred in more than one sense. It not only raised questions of
authorship and literary form, which, after all, are concerns that do not
transcend the confines of our own culture,12 but it also revived critical
thought about the role of literacy in relation to orality. The latter
makes ethnography political. Othering, in my view, is cut short when
awareness of the political dimension of writing remains limited to
insights about the political character of aesthetic standards and rhetorical
devices. In such critical discourse, anthropology's Other is said to be
dominated by ethnography. But to be dominated, it takes more than to
be written about.To become a victim the Other must be written at (as in
"shot at") with literacy serving as a weapon of subjugation and discipline.
Conversely, to stop writing about the Other will not bring
liberation. But more about that later.13
Experiments with Genre: Symptomor Therapy?
How do the changes that came about as a result of the so-called
generic crisis-comprising, roughly, the demise of the monograph as
12. Limitation to such concerns is what can be held against Clifford Geertz's latest,
and again brilliant, work, Worksand Lives: TheAnthropologistas Author (Stanford, Calif.,
1988). It is a critique of anthropological writing in a voice that talks from, to, about, and
perhaps around ethnography as a self-contained practice. For a similar critique (with
many references to the recent debate about Geertz's approach), see Bob Scholte, "The
Charmed Circle of Geertz's Hermeneutics: A Neo-Marxist Critique," CritiqueofAnthropology6
(Spring 1986): 5-15.
13. See also Fabian, "Dilemmas of Critical Anthropology" (forthcoming).
Summer1990 761
the model form, the softening of strictures regarding an "objective"
and "realistic" prose style, the refusal to maintain a clear separation of
systematic description and narrative account-affect representation? Is
the generic crisis a phase withinthe representationist stance, a period of
emerging new genres that will eventually replace the old ones as privileged
representations of the Other? Should we therefore fear that what
looks like a crisis isjust a lot of noise made by anthropologists regrouping
in their attempts to save their representer's privileges? Or does the
generic crisis go far enough to become a crisis of representation, leading
perhaps to a rejection of representation? In which case
nonrepresentation, including its most radical form, not-writing,
graphic silence, would have to be an inescapable consequence. I shall
say something about not-writing in the next section, but first let me
offer a few thoughts on the debate about genres.
One of the confusing aspects of the debate is a lack of precision in
using the term genre. Perhaps this is salutary at this early stage; I find it
obfuscating. In the course of its history the term has sometimes taken a
singular, or adjectival, meaning (in such expressions as "genre painting,"
or "this is not his genre") but in the present critical discussion, I
believe its primary use should be to signal differentiation within a
domain of activity. If, for instance, ethnography qualifies as a genre,
this can only mean that it is, without residue, subsumed under literature
(and its other genres). This is a way to preempt whatever useful
insights might be gained from looking at ethnography as literature.
What else would be left over for the critic but to recommend good over
bad writing? (Come to think of it, Geertz probably deserves credit for
initiating the new literary awareness in anthropology not so much
because he fraternized with literary critics but because he dared to
write well and got away with it.)
The broadest claim made about the generic crisis has been epistemological.
Experimenting with new kinds of writing took many
different forms, but it has been said that their common background is a
crisis of realism.14 Inevitably, this has caused critics to express fears that
to abandon realism may be tantamount to giving up on objectivity,
which in turn would mean the end of anthropology as a science.'5 In my
view, it is too early to sound a general alarm. Instead I would suggest
the following line of argument: What was wrong with ethnographic
realism (as a literary convention) was not its realism (as an epistemological
stance) but the surreptitious substitution of the former for the
latter. As a literary form, realism has been a mode of representation,
complete with its various privileged representational devices. As such,
14. See Marcus and Dick Cushman, "Ethnographies as Texts," Annual Review of
Ethnography11 (1982): 25-69.
15. See, for example, Jonathan Friedman, "Prolegomena to The Adventures of
Phallus in Blunderland: An anti-anti-discourse," Cultureand History1 (1987): 31-49.
CriticalInquiry
762 Johannes Fabian Presenceand Representation
realism has been a style of anthropological writing corresponding to a
mirror-of-nature theory of knowledge. A kind of knowledge that is
really created by conventions of writing claims to be the reflection of
that which is real. Realism's often unspoken, and unwarranted, pretention
has been that a reality-a practice of writing "realistic"
ethnographies-be representative of the reality. The turn to genres
could be regarded as a turn not to less but to more realism, if one avoids
confusion between literary and philosophical realism.
That the issue is even more complicated may be taken from Steven
Webster's recent critique of the ways in which ethnographic realism has
been perhaps too easily linked to nineteenth-century literary realism.16
This is not the place to comment on his remarks on George Marcus and
Dick Cushman, and James Clifford. But I agree with the general direction
of his argument, which is, if I understand him correctly, that
ethnographic writing has been naively naturalist rather than realist.
And I share his fears that a reification of genre, however critical and
sophisticated it may be, may result in the degeneration of critical epistemological
diagnosis into literary "therapy" of ethnography. Even more
depressing is the prospect that the discovery of generic constraints in
ethnography may have an effect analogous to that of, say, dependency
theory in "explaining" relations between the first and third worlds.
Having demonstrated the inescapable logic of oppression and exploitation,
such a theory threatens to eternalize them (because it is more
easily taken as a proof of superiority than as a program for change).
Yes, generic constraints are at work when we write, but to discover
them does not absolve us. Indeed, as Webster says, "the genre be
damned." 17
In discussions about genres anthropology questions itself, not just
incidentally but in principle; at issue is not just the style but the nature
of what we are doing when we write. As I suggested before, this regards
praxis, not just modes of representation. Anything that leads to considering
the practical aspects of writing about the Other is in my view not
less but more realistic. Furthermore, with this perspective we will also
be led to ask how a praxis of writing relates to the praxis written about.
That relation cannot be exhausted by mirroring because praxis is doing.
Doing does not mirror anything and that alone is reason enough to
discard naive realism.18 Doing is acting on, making, transforming (giv-
16. See Steven Webster, "Realismand Reification in the EthnographicGenre,"
CritiqueofAnthropology6 (Spring1986):39-62.
17. Ibid.,p. 59.
18. I fullyagreewithTyler whenhe statesthatthe problemwithrealism"isnot...
the complexity of the so-calledobject of observation, nor failure to applysufficiently
rigorousandreplicablemethods,nor even lessthe seemingintractabilityof the language
of description. It is instead a failure of the whole visualist ideology of referential
discourse,withits rhetoricof 'describing,''comparing,''classifying,'and 'generalizing,'
and its presumptionof representationalsignification"(Tyler, "Post-ModernEthnography,"p.
130).
CriticalInquiry Summer1990
ing form to), not regrettably so or incidentally (as complaints about
subjective distortions in writing would have it), but inevitably. Praxis as
transformation is the condition of relations with the Other. Othering in
the sense of "making the Other" is the way in which that particular
form of intercultural relation called ethnography occurs, in real space
and real time, not (to use Michel Foucault's expression) in the "tabular
space" of prefabricated systems of representation.
If this perspective is adopted, it follows that genres deserve our
attention not so much because they are helpful in inventorying and
classifying kinds of writing (or even in locating anthropological writing
in Western literature) but as a conceptual element in a theory of text
production that is unified in the sense that it encompasses the
"ethnographic text" and the texts of ethnographic writing. A gap in the
generic debate, indicating in my view a direction to be explored with
profit, is a failure to recognize that some anthropologists began to think
about genres in an attempt to understand text production "on the
ground." At issue was not the shape of the final, published product but
the form of (mostly) oral accounts on which such products must be built
if they are to be ethnography: texts as a record of communicative
events. 19
I now want to take up briefly two issues that have played a prominent
part in recent genre debates: dialogic and poetic modes of writing.
Dialogue, perceived vaguely as an alternative to isolating or domineering
monologue, has been en vogue more than once during this
century. It was part of what one might call soft existentialism (Martin
Buber comes to mind) and part of hard critical theory (Habermas); it
really became fashionable as various "encounter" doctrines swept the
globe, and it never fails to serve as a device of obfuscation in the
mouths of politicians. Anthropologists began to think seriously about
dialogue at a time when, in general usage, the term had reached a low
in signification. It had acquired a nonspecific ethical bonhomie, oozing
good will, apparently lacking any cutting edge that would be required
for critical work. Who could be against dialogue? This has tended to
obscure the fact (for some outside critics, perhaps even for some of the
proponents some of the time) that dialogue was first introduced not to
signal an ethical attitude, much less a literary form, but to serve as a
19. See my own attempts in "Genres in an Emerging Tradition: An Approach to
Religious Communication," in Changing Perspectivesin the ScientificStudy of Religion, ed.
Allan W. Eister (New York, 1974), pp. 249-72, with acknowledgments to Dell Hymes,
Dan Ben Amos, and others, and a more generalized application in a study of colonial
language description, Language and Colonial Power: The Appropriation of Swahili in the
Former Belgian Congo, 1880-1938 (Cambridge, 1986). Incidentally, anthropologists
concerned with this issue will profit much from watching recent developments in African
history, or historiography: see Jan Vansina, Oral Tradition as History (Madison, Wis.,
1985), and AfricanHistoriographies:WhatHistoryfor WhichAfrica?ed. Bogumil Jewsiewicki
and David Newbury, Sage Series on African Modernization and Development, vol. 12
(Beverly Hills, Calif., 1986).
763
Presenceand Representation
reflection about the nature of anthropological fieldwork. So once
again, as in the case of realism, epistemological and literary or generic
meanings should be distinguished before their relation is considered.
Having established the communicative, and in this sense dialogical,
nature of fieldwork, having realized that it is an activity that rests on
exchange and conversation at least as much as it does on observation
and recording, reflection moved on where it perhaps should have
lingered to ponder epistemological implications.
Kevin Dwyer and Dennis Tedlock have been two outspoken advocates
of dialogue in anthropology. In their well-known essays and later
in their books, I have understood them to ask (by opposing dialogical to
monological, or to analogical) whether dialogue, being central in the
early phase of knowledge production, should not also generate a
specific discourse in later phases, up to the final presentation of ethnography
in a published text.20
Critics of Dwyer and Tedlock inevitably put their teeth into the
most palpable part of what is really a complex epistemological argument,
namely the dialogical form of ethnographic writing. Having
found what is indeed a weak spot (the question whether actual dialogue
as it occurs in fieldwork should or could ever be represented by written
dialogue), they have been quick to dismiss dialogical ethnography as
well intentioned but utopian. I should like to plead here for continued
exploration of the dialogical nature of ethnographic research.21 Meanwhile
we can continue to clarify relations between dialogue as event and
dialogue as literary form. One thing seems to be certain: dialogue as a
genre of ethnographic writing will not automatically preserve the dialogical
nature of the knowledge process. I can think of two lines of
argument leading to that conclusion.
First, the most compelling reason for acknowledging dialogue is
not that the ethnographic process most often involves actual conversation.
More significant is that our mind or thinking can produce
ethnographic knowledge only because it is capable of being intersubjective.
Thought does not function only by imposing logic on data but also
by addressing statements and questions, by listening and responding, in
short, by communicating with a real Other. Inasmuch as invoking
20. See Kevin Dwyer, "On the Dialogic of Field Work," Dialectical Anthropology2
(May 1977): 143-51; "The Dialogic of Ethnology," Dialectical Anthropology4 (Oct. 1979):
205-24; and MoroccanDialogues: Anthropologyin Question(Baltimore, 1982). See Dennis
Tedlock, "The Analogical Tradition and the Emergence of Dialogical Anthropology,"
Journal of AnthropologicalResearch35 (Winter 1979): 387-400, and The SpokenWordand
the Workof Interpretation(Philadelphia, 1983). See also TheInterpretationof Dialogue, ed.
Tullio Maranhao (Chicago, forthcoming).
21. This is being done in an as-yet-unpublished paper by Paul Friedrich and John
Attinasi, "Dialogic Breakthrough: Catalysis and Synthesis in Life-Changing Dialogue."
764 Johannes Fabian
CriticalInquiry Summer1990
dialogue serves to stress the intersubjective nature of ethnographic
investigations, it signals an intent to go beyond positivism and scientism.
To assure that this (epistemological) position informs all stages of
the ethnographic knowledge process, it is not enough to cast the
"results"into a dialogic form.
Admittedly, to be attentive to literary form (rather than passing off
ethnography as neutral reporting according to unreflected canons)
expresses recognition of the dialogical nature of ethnography inasmuch
as it reflects recognition of the multiple audiences a writer addresses
and of "constituencies" for whom he presumes to speak. But to insist
that only one form, written dialogue, can do this would in my view run
against the declared epistemological intentions of a dialogic approach.
There is another way to arrive at the same conclusion. It consists of
applying what was said earlier about representation and mirror theories
of knowledge. If dialogue as a literary genre were the only adequate (or
accurate) representation of dialogic experience made during field
research, it would thereby become "analogical" (in Tedlock's understanding
of the term). If representationism is to be abandoned and
replaced with a notion of praxis as described above, then the appropriate
ways of conveying knowledge based on dialogic experience with the
Other would be those that transform that experience in a struggle with
"means of production" of discourse that include autobiography, political
economy, relations of power, scientific canons, and literary form,
but do not privilege dialogue as a genre. Conversely, ethnographic
representations that are (or pretend to be) isomorphic with that which
is being represented should be met with suspicion; more likely than not,
they lack what distinguishes knowledge from mimicry.22
With that I have rehearsed my argument regarding poetry in
anthropology. I do not pretend to have a full grasp of the issues
involved in that particular debate. The remarks I am about to make are
addressed to a recent publication, Reflections:The AnthropologicalMuse,
especially to the editor's essay, J. Iain Prattis's "Dialectics and Experience
in Fieldwork: The Poetic Dimension." Prattis diagnoses a "gap"in
field reporting and suggests that this gap may be filled with a new,
different "language of experience," which he and others have found in
poetry as a form of ethnography. Throughout he defines the problem
as one of overcoming the -emic/-etic distinction-between an inside
and an outside view of other cultures-and the stifling effect it has had
on theory and method in anthropology. At the same time he seems to
accept that the constituting acts of field research consist of "observa-
22. Which is not to deny that mimesis is probably an aspect of all forms of knowing
that qualify as ethnographic. On the issue of mimesis and recognition of an Other, see the
interesting study by Fritz Kramer, Der roteFes:UberBesessenheitund Kunstin Afrika(Frankfurt
am Main, 1987).
765
766 Johannes Fabian Presenceand Representation
tion."23 If I understand him correctly, he advocates poetry as a new,
possibly more adequate and hence privileged representation. With that,
poetry, a literary genre, is set to compete with scientific prose, another
literary genre. I do not deny that such competition may be healthy and
entertaining, and that writing poems has helped individual anthropologists
to overcome alienation from experience. At the same time I fear
that it may do little to further the project of othering. To seek the solution
for a problem regarding the production of knowledge in different
or better representations of knowledge is to reaffirm, not to overcome,
the representationist stance.
Matters are different with an approach exemplified by Paul Friedrich.
His accomplishments as an ethnographer and as a poet put him
above suspicion of using poetry as an escape from ethnography. More
important, in his theoretical reflections Friedrich does not begin by
advocating poetry as a literary genre but by arguing for poesis as an
alternative to representationist conceptions of language and culture. I
consider his eloquent critique of linguistic and anthropological obsession
with order (and indeed of equating knowledge with the imposition
of order on some presumed chaos) one of the most important theoretical
advances in recent anthropology.24
In sum, dialogical and poetic conceptions of ethnographic knowledge
touch the heart of questions about othering. But they have a
chance to change the shape of ethnography only if they lead to literary
processes that are hermeneutic-dialectical, or "practical," rather than
representational. To preserve the dialogue with our interlocutors, to
assure the Other's presence against the distancing devices of anthropological
discourse, is to continue conversing with the Other on all levels
of writing, not just to reproduce dialogues. In fact, I have gone as far as
saying precisely that f fieldwork is conducted dialogically, problems of
writing will not be resolved by adopting the dialogical form. Similarly,
that we remain attentive to what I called the transformative, creative
aspects of ethnographic knowledge, for which I gladly adopt the term
poetic,will not be guaranteed by poetry as a form of writing.
I am willing to entertain the idea that adopting a dialogical and
poetic master trope might change anthropologists from natural histori-
23. SeeJ. Iain Prattis, "Dialectics and Experience in Fieldwork: The Poetic Dimension,"
in Reflections:TheAnthropologicalMuse, ed. Prattis (Washington, D.C., 1985), pp.
266-83.
24. See Friedrich, "Linguistic Relativity and the Order-to-Chaos Continuum," in
On LinguisticAnthropology:Essaysin Honor ofHarryHoijer, 1979, ed. Jacques Maquet (Malibu,
Calif., 1980), pp. 89-139; reprinted in Friedrich, The Language Parallax: Linguistic
Relativismand PoeticIndeterminacy(Austin, Tex., 1986), pp. 117-52. An appreciation of
Friedrich's poetry as well as of his theoretical contribution was given by Tyler in "The
Poetic Turn in Postmodern Anthropology: The Poetry of Paul Friedrich," American
Anthropologist86 (une 1984): 328-36.
CriticalInquiry Summer1990
ans into itinerant bards, clowns, or preachers, and that this may
ultimately be the aim of "being with others." But I still think that such
purposeless conviviality must be earned by the critical labors of inter-
pretation.
Writingand Not-Writing
Different from, but not unrelated to, the kinds of critique of
anthropological writing I have commented on so far is the debate about
writing framed in the opposition of literacy and orality. Participants
in this discussion take varied points of departure: some begin with
writing as a technology, or with the special constraints imposed by our
system of alphabetic writing; others ponder the changes that take place
when sound-events are represented in writing, an issue that is related to
that of contrasts between the verbal-aural and the graphic-visual; still
others start with the societywide or global impact of literacy as a historical
phenomenon; and so forth.25 It is my impression that this debate,
perhaps due to the influence of thinkers such as Walter Ong, Jacques
Derrida, and Foucault, questions anthropological writing as a praxis of
representation in a context of power more radically than the critique of
genres.
That the problem with anthropological writing is a problem with
relating to an Other is much more in the foreground in the literacy
than in the literary discussion. The former also tends to give a more
compelling expression to the anthropologist's dilemma: if writing is
part of a system of intellectual and political oppression of the Other,
how can we avoid contributing to that oppression if we go on writing?
There are those who respond to a seemingly radical question with a
radical answer; they give up writing about the Other and drop out, if
not out of anthropology, then out of ethnography. But if the premise is
correct, namely that given the power relations in this world writing as
such is an act of oppression, then writing need not have the Other as its
subject matter in order to oppress the Other. To stop writing altogether
would seem to be a logical consequence (for those who care), but
25. SeeJack Goody, TheDomesticationof theSavageMind (Cambridge, 1977), and his
earlier work, Literacyin Traditional Societies(Cambridge, 1968). See also Ton Lemaire,
"Antropologie en Schrift," in Antropologieen Ideologie,ed. Lemaire (Groningen, 1984), pp.
103-24; Spokenand WrittenLanguage: Exploring Oralityand Literacy,ed. Deborah Tannen
(Norwood, N.J., 1982); Tedlock, The SpokenWordand the Workof Interpretation;Tyler,
"The Vision Quest in the West, or What the Mind's Eye Sees," Journal of Anthropological
Research 40 (Spring 1984): 23-40; and Literacy and Society, ed. Karen Schousboe and
Mogens Trolle Larsen (Copenhagen, 1989), especially the essay in that collection by
Michael Harbsmeier, "Writing and the Other: Travellers' Literacy, or Towards an
Archeology of Orality," pp. 197-228.
767
768 Johannes Fabian Presenceand Representation
it is hard to see how it could ever become more than an isolated, and
probably temporary, act of demonstration. No one expects critical
anthropologists to regroup as a community of silent meditators or
vociferous oral raconteurs.26
Perhaps we dare not say what we dare not hope: power relations
must change.27 What can experimenting with genres or the critique of
writing accomplish toward that end? Well, to begin with, they can help,
have already helped, to undermine the kind of naive security that went
together with belief in scientific objectivity and the neutral nature of
scientific prose. But is there a guarantee that oppressors will be less
oppressive just because they become self-conscious? Consciousness-raising
can only be preparatory to a critique that might have a chance of
being truly subversive. I am not sure that I can make my point at this
time, and within the limits of this essay, but I'll give it a try.
What needs to be subverted is a stubborn residue of positivity in
conceptions of anthropological writing. Philosophical representationism
engendered sociological positivism: human conduct, of which
writing is one sort, is understood and explained as enactment of collective
representations, as belief in beliefs, as valuation of values, as
obedience to rules, as exercise of grammars. Such positivity makes it
impossible to think negatively of conduct (as nonconduct) except as bad
conduct, deviance. True, much has been accomplished in the last few
decades in criticizing sociological positivism, but I wonder how far this
went in the critique of ethnographic writing. A dialectical conception
of writing (which I do not pretend to invent here, but neither do I want
to bother at this moment with tracing its ancestry) holds that every act,
every production, must contain its negation to become realized. Notwriting
is a "moment" of writing. This sounds abstract and must be so
in order to be thought. It is nevertheless eminently practical, as I would
like to illustrate now with the help of some autobiographical reflec-
tions.
26. The problem is tricky enough as it stands, but it should at least be noted that to
locate literacy on our side, and orality on the Other's, as some of the contributors to this
debate seem to do, is a simplification that rapidly becomes too gross to be permitted even
for the sake of the argument. As some of us who started as emissaries of literate societies
facing oral cultures began to discover, literacy is catching up with us from the other side.
In my own case, this has been one of the most striking findings in a project of "revisiting"
the Jamaa movement in Shaba/Zaire twenty years after I had first written about it. See
also Fabian, Historyfrom Below: The Vocabularyof Elisabethville byAndre Yav (Amsterdam
and Philadelphia, 1990).
27. This is, of course, a point made by Edward W. Said in a recent essay published
in this journal, "Representing the Colonized: Anthropology's Interlocutors," Critical
Inquiry 15 (Winter 1989): 205-25. I was invited to comment on that essay but could not
do so for reasons of time. Although it was written earlier, my paper may now be read as a
response to Said after all.
CriticalInquiry Summer1990
?Of the three projects of fieldwork I have undertaken so far (all of
them in urban-industrial southern Zaire), the first one was dissertation
research. It took the conventional eighteen months and produced a
monograph and numerous articles. The latest one consisted of summer
visits in 1985 and 1986; a paper and two books are about to be
published. But there was also (now fifteen years ago) a period of almost
three years of research, much better planned and perhaps more carefully
carried out than the other two, which so far has yielded only one
major article and no monograph. I have spent much time coming to
grips with my unwillingness or inability to "write up" the rich data I
collected in those years. In fact, part of my trouble has been to decide
whether it was unwillingness or inability. Or was it just the absence of
pressure because I had gotten professional security in the meantime?
Perhaps my failure to turn this project into ethnography was a sign
of liberation from scientism and from a conception of writing as a sort
of production line, running from "raw data" through theoretical
processing to final monographic assembly. Between the first and the
second fieldwork, I had spent much of my theoretical energy criticizing
this view as positivist and scientistic. Was I now paying the price for
rocking the scientific boat? I also found out that if one has abandoned
empiricist-positivist notions of data as collected pieces of information
waiting to be fed into one or the other analytical machinery; if one has
come to realize that ethnographic data are not "given" but "made"
through communication; if, in a sense, ethnographic "material" has
become subjective, autobiographical, then the problem of unused data
or unwritten monographs becomes more, not less, pressing. The expectation
that it should be less urgent might arise from an easing of
pressures to execute research projects according to rigid canons. But it
does not seem to work that way. No tension is more acute than the
intrapersonal experience of unfinished work, all the more so because,
on this personal level, we experience the presence of the Other. It is
this experience of presence-an experience, by the way, that may grow
with time and, at any rate, needs time to grow-that lies at the origin of
my struggling with the idea of representation as praxis.
Freed from the strictures to view personal experiences and their
embodiment in recordings, notes, pictures, and so forth, as ethnographic
data that become only what they ought to be through
representation, the ethnographer has several alternatives. One (see
above) is to experiment with writing in an effort to communicate
"ethnography" as directly and vividly as possible. Another is to dissociate
these data from any scheme or purpose and to treat them as
necessary but gratuitous, like the air we breathe (if this is still an appropriate
image), as elements that nourish anthropology without having
first been processed into ethnographic monographs. This is one aspect
of what I meant by not-writing as part of writing. It is unlikely that
769
770 Johannes Fabian Presenceand Representation
funding agencies will accept this view and start sponsoring gratuitous
field experience (that they have often done this in the past cannot have
escaped their attention); still, to leave much or most of what was experienced
unwritten about is how anthropological writing works and always
did work, even though this was not always recognized.28
Because the problem is not so much writing in general but ethnographic
writing, there is yet another form of not-writing to be
considered. To show this, I return to my own case. I may have given the
impression that I felt that my first fieldwork was adequately "written
up" with ethnographic publications, that the third project was well on
its way to turning into ethnography, while I had hardly written
anything about the project in between. So strong are our internalized
conceptions of ethnography as representation that it takes a special
effort of reflection to realize that all of this is inaccurate. Throughout
the years I produced "theoretical" and "historical" studies that seemed
to be not-writing from the point of view of ethnography because they
were not descriptive of what I had found out in field research. Yet none
of them would have been conceived and carried out without those
experiences. In this sense, these writings are the results of processes
that originated "in the field" without being representations of them.
They are attempts to formulate insights in response to concrete, practical
demands-such as teaching, lecturing, publishing, and making a
living in academic institutions-while trying to stay in the presence of
the Other.29
If recognizing the necessity of what I called not-writing helps to
further a conception of ethnography as praxis, all the better. I realize
28. I remember Paul Bohannan declaring many years ago that should he ever
discover a Ph.D. candidate having put everything he knew about the subject into his
thesis, he would refuse to certify the candidate as an anthropologist. The realization that
ethnographic notes, far from only providing a finite quantity of information, are only
contingent extractions from an inexhaustible reservoir of matters for thought shows how
misguided (maybe not ethically, but epistemologically) demands are to consign one's
"data"to public depositories. This is, of course, not to deny that certain kinds of material
that have a documentary character of their own can be shared by more than one ethnographer
and may even be of importance to the people studied.
29. Time and the Other(1983), for instance, is in no obvious way about the Jamaa
movement, which I began to study in 1965. But I know-and an outside reader might
notice this even more than I do-that some of the theoretical issues I address in this book
were first experienced as practical problems; and there are some theoretical insights I
owe to the teachings of the movement. Similarly, Language and ColonialPower (1986) is
not an ethnography of sociolinguistic research on language and labor conducted in 1972-
74. It is a historical study with some theoretical pretensions, but to me it has been the
only practical way to make sense of what I had naively taken to be a "given" linguistic
situation.
CriticalInquiry Summer1990
that it is not as such a remedy for what is wrong in our relations with
the Other. It helps to create conditions for othering-recognition of
the Other that is not limited to representations of the Other.
All this makes little more than a start on a long road toward undoing
ethnographic representationism. Several of the issues I touched on
need to be recast in more subtle terms. Take not-writing, for instance.
In its first form discussed above, I described it as a rather trivial case of
inevitable selectivity: some data are chosen, others omitted. Positivists
write that way, what is so dialectical about it? I hope to have given part
of an answer, but the idea needs to be worked out more fully. Furthermore,
are cases of not-writing that seem to be a personal affliction and
are known to have destroyed careers something altogether different
from the ethnographer's temporary silence or the preservation of
unwritten-about experience as an ethnographic branch to sit on? How
much confusion is spread in discussions of anthropological writing by a
failure to distinguish not only between literature and literacy but also
between writing and publishing? An important part of myself is now
defined as a writer. Given my lack of talents en belleslettres,this would
not have happened had I not as an anthropologist sought to converse
with an Other and then been pressured to communicate my insights in
writing. Through writing I work as best I can on making my experiences
critically useful both to the community for which I write and the
community about which I write (and there are signs that the two are
merging). But does that mean that the people I studied would profit
much from reading those of my writings that conventions of anthropological
publishing have qualified as acceptable ethnography? In Timeand
the OtherI made an oblique remark to the effect that given the dominant
rhetoric of anthropological discourse, the Other's ethnographic
presence goes together with his theoretical absence. In ethnography as
we know it, the Other is displayed, and therefore contained, as an
object of representation; the Other's voice, demands, teachings are
usually absent from our theorizing.
Should this current search attain its highest aims-to transform
ethnography into a praxis capable of making the Other present (rather
than making representations predicated on the Other's absence)-then
the work of interpretation can begin. In the end all questioning of the
How must be in the service of understanding the What, what it is all
about. And the answer to that question carries no promise of illumina-
tion:
There are those who say that a native will not speak to a white man.
Error. No man will speak to his master; but to a wanderer and a
friend, to him who does not come to teach or to rule, to him who
asks for nothing and accepts all things, words are spoken by the
camp-fires, in the shared solitude of the sea, in riverside villages, in
771
Presenceand Representation
resting-placessurroundedby forests-words are spoken that take
no account of race or colour. One heart speaks-another one
listens;and the earth, the sea, the sky, the passing wind and the
stirringleaf, hearalsothe futile taleof the burdenof life.30
30. Joseph Conrad, "Karain:A Memory,"SelectedTalesfromConrad,ed. Nigel
Stewart(London,1977),pp.65-66.
772 Johannes Fabian