The Problem of Ethnographic Representation

Gold rule

In his review of Edward Said's Orientalism, James Clifford asks,

Should criticism work to counter sets of culturally produced images like Orientalism with more "authentic" or more "human" representations? Or, if criticism must struggle against the procedures of representation itself, how is it to begin? [1]

Said's answer to Clifford's first question is unequivocal: "It is not the thesis of this book to suggest that there is such a thing as a real or true Orient."[2] Said's statement seems paradoxical since he devotes over 300 pages to an assault on inauthentic portrayals of the Orient. However, Said points out that "the Orient" is itself a constituted entity."[3]Since "the Orient" does not exist except as a fabricated concept, the question of authenticity or inauthenticity is irrelevant.

Said believes his function as a critic is to struggle against the procedures of representation itself:

[T]he real issue is whether indeed there can be a true representation of anything, or whether any and all representations, because they are representations, are embedded first in the language and then in the culture, institutions, and political ambiance of the representer. If the latter alternative is the correct one (and I believe it is), then we must be prepared to accept the fact that a representation is eo ipso implicated, embedded, interwoven with a great many other things besides the "truth," which is itself a representation.[4] (Emphasis in original).

Said's study of Orientalism is negative in two respects. First, he claims that representations cannot be avoided, even when truths are conveyed. Second, he illustrates the misuse of various representational strategies, but he never suggests a correct approach to the problem of representation. Said provides readers with critical tools for detecting suspect representations, but he provides no guidance for authors who may wish to confront the problem of representation. Clifford asks where criticism is to begin with the struggle against the procedures of representation, but the more pertinent question should be where does the author begin in the struggle against the procedures of representation.

Authors write texts in order to accomplish certain purposes. A purpose of the poet or novelist is creative self-expression. For the creative writer, representation is the vehicle for expression; the creative writer consciously chooses representations as representations. The writer of nonfiction, however, typically focuses on the substance of what she wishes to communicate, and often fails to realize that she uses representations when communicating her ideas. Some writers of nonfiction attempt to eliminate all vestiges of expressiveness or rhetoric from their texts, but this attempt must fail:

Anything which makes functional use of words will always be involved in all the technical problems of words, including rhetorical problems. The only road from grammar to logic, then, runs through the intermediate territory of rhetoric.[5]

"[R]hetoric ... is the characteristic manner by which a text's language and organization convinces its readers of the truth, or at least of the credibility of its claims."[6] The nonfiction writer focuses on the need to convince or persuade her reader, and is concerned with rhetorical strategies only insofar as they meet this need. The creative writer and nonfiction writer both use rhetoric, but, generally, the creative writer uses rhetoric as both a means and an end in itself, while the nonfiction writer tends to view rhetoric solely as a means to an end.

Writers of nonfiction may begin to approach the problem of representation by becoming aware of their use of rhetoric in representation. Some of Said's detractors have declared that Said's point concerning the inevitability of representation is trivial. However, Said's observation is far from trivial if writers incorporate that observation into their texts. All writers must become aware of the constructed nature of their texts; in this sense all texts are fictions. The word "fiction" comes from the Latin word fingere which means "to form".

Ethnography, the anthropological description of technologically "primitive" societies (although contemporary ethnographies are no longer limited to the study of primitive societies), is a paradigm of representation of "the other". During the past several decades anthropologists have been attempting to deal with the problem of representation of other cultures, and the remainder of this essay will examine some of the solutions proposed by theorists and practitioners of ethnography.

Gregory Bateson's Naven is an example of the self-conscious use of rhetoric. He begins by presenting the naven ritual as a puzzle. Bateson then describes other components of Iatmul culture as he recreates (represents) the development of his interpretation. Bateson draws the reader's attention to language, style and rhetoric throughout. In an epilogue, Bateson admits that his purpose was not so much to describe Iatmul culture, "but to suggest methods of thinking about anthropological problems."[7] Nevertheless, Bateson represents another culture and simultaneously  makes explicit his methods of representation.

While the ethnographer should approach her text with an awareness of its fictional nature, and should self-consciously adopt some of the rhetorical and literary techniques ordinarily associated with creative writing, e.g., genre, narrative, and metaphor, she must avoid carrying the parallels between the two forms of writing too far. For example, an omniscient or totalizing perspective may be appropriate for the creative writer communicating her personal vision, but not for the writer describing another culture.

In order to establish anthropology on a scientific foundation, early ethnographers adopted a realistic style derived from natural science writing. Ethnographic realism is defined as "a mode of writing that seeks to represent the reality of a whole world or form of life."[8] (Emphasis added). According to Marcus and Cushman, "what gives the ethnographer authority and the text a pervasive sense of concrete reality is the writer's claim to represent a world as only one who has known it first-hand can."[9] As James Clifford puts it, the goal of ethnographic realism is to give the reader a sense of "you are there, because I was there."[10] Note the tenses used in Clifford's statement: the ethnographer organizes his past experiences in order to give the reader the illusion of an experience in the present.

Marcus and Cushman identify nine characteristics of ethnographic realism: (i) a totalizing description of another culture; (ii) an omniscient, unintrusive narrator; (iii) substitution of composite creations for individuals; (iv) references to fieldwork only to the extent necessary to establish the actual presence of the ethnographer; (v) focus on everyday life situations; (vi) dogmatic claim that the native point of view is being represented; (vii) generalizations are favored over detailing of particular facts; (viii) use of jargon; (ix) conceptual abstractions which bypass attention to the context of native language.[11]

The result of the foregoing characteristics is a radical separation between the fieldwork experience and the ethnography which is the product of the fieldwork. The attempt to maintain realism actually results in an absolute representation. Ethnographic realism produces a mimesis and synthesis of the fieldworker's experience; the text is an imitation of an experience.

Recent experiments in ethnography have attempted to deal directly with the elements of ethnographic realism which paradoxically falsify and distort other cultures. In one experiment the fieldworker and his experience becomes the focus of the text. Jean-Paul Dumont's The Headman and I: Ambiguity and Ambivalence in the Fieldworking Experience is an example of the self-reflexive approach.

Dumont sees fieldwork as a dialectical experience: native culture (thesis) and anthropologist (antithesis) interact and both are changed in the process (synthesis). This is not Marxist dialectics strictly speaking, because the native culture and anthropologist, although changed, remain separate. Instead, Dumont's observations suggest that fieldwork is an illustration of the Heisenberg Uncertainty Principle: an investigator, through the very process of conducting an experiment, alters the conditions which he examines and is himself changed. Dumont believes intersubjectivity resolves the dilemma of objectivism versus subjectivism:

The problem is not to eliminate the distortions of subjectivity and objectivity, but mainly to reinstall experience in its place; in other words, to let it all happen, to accept the radical character of the fieldworking experience. Once subjectivism and objectivism are rejected, what is left to turn to? The answer was given to me indirectly in the field and amounts to the experience of intersubjectivity. ... Intersubjectivity depends exclusively upon the possibility of establishing a dialogue, that is, upon the reversal of perspective whereby not only are the natives anthropologized -- they are also, in turn, anthropologizing.[12]

The fieldworker must not only be aware of intersubjectivity, he must also make it an explicit component of the ethnography.

Dumont is alert to the psychological aspects of fieldwork. He is informally adopted as the "brother" of the headman of the tribe he studies. As time passes, Dumont recognizes the phenomenon of transference in which he reacts to the headman just as he would with an actual sibling. Dumont realizes transference will interfere with his interpretation of Panare culture, but rather than attempt to identify and exclude the impact of the transference, which is impossible, he lets the reader know of his tainted perspective. It then becomes the reader's responsibility to evaluate the effects of the transference.

Dumont is fully aware that his task as an ethnographer is analogous to translation. While the translator of a text from one language to another seeks to remain transparent, if the cultural translator is invisible, the reader of his text cannot detect the crucial factors which effect the translation, i.e., the cultural, institutional and political biases of the translator. Dumont compares the process of cultural translation to a Rorschach test. Dumont also recognizes the impossibility of some translations. For example, when he attempts to describe the significance of a victory in the World Series, he can only say "They have killed a lot of game and now they are going to drink a lot of beer."[13] Though Dumont does not explicitly say so, he implies that translation in the reverse direction is equally problematic. In any event, Dumont makes the problems of translation and his role in translation explicit.

Throughout his ethnography, Dumont takers great pains to emphasize the contingent, incomplete, emerging nature of the knowledge he acquires. He explicitly acknowledges his limitations and fallibility. Interpretation and speculation are clearly signaled. Dumont does not resort to composite characters, and he successfully portrays members of the Panare village understudy as individuals. Furthermore, Dumont uses particularized statements such as "I saw certain Panare do X" rather than more generalized statements such as "the Panare do X."

The problem of representation cannot be avoided altogether, but Dumont suggests that the explicit handling of representation gives the reader the opportunity to detect the truth which may be present in his representation. Although Dumont violates each principle of classical ethnographic realism identified by Marcus and Cushman, his work is more realistic since it rejoins the fieldwork experience (encounter with "the other") and the text created as a result of that experience (representation of "the other").

The Headman and I was published two years after an earlier book by Dumont based on the same fieldwork experience. The earlier ethnography is more straightforward and less self-reflective than The Headman and I, and perhaps a combination of the two books could achieve a further improvement in the handling of representation. At times The Headman and I seemed overly confessional, and would probably be more meaningful if the reader was given more concrete information about Panare culture.

Texts utilizing the conventions of ethnographic realism attempt to establish authorial authority. Monologue is the dominant form of such ethnographies. Marjorie Shostak's Nisa: The Life and Words of a !Kung Woman[14],however, is presented in the form of a dialogue. The conditions of Shostak's fieldwork are described in an introduction and an epilogue. The main portion of the book consists of thematically organized chapters which begin with Shostak's brief description and interpretation of !Kung culture, followed by Nisa's commentary on the chapter theme. Although the ratio of material provided by Nisa to that provided by Shostak is 2:1, Shostak is listed as the sole author; Shostak, of course, selected, edited and organized Nisa's statements. Even though the form of the text is a dialogue, Shostak's ultimate control over the text makes it a monologue. The monological aspect is repeated within the text itself: there is no true discourse between Shostak's and Nisa's portions of the text, only alternating monologues.

However, Shostak's text does move away from the central position of the ethnographer (implicit in ethnographic realism and explicit in Dumont), and brings the importance of native informants to the foreground. "The other" is given the opportunity, albeit limited, to represent herself in Shostak's text. Shostak's text is also significant because it attempts to incorporate dialogue as a structural feature. As noted above, this attempt is not entirely successful; however, Shostak demonstrates the potential usefulness of multiple voices to disperse authorship.

The strategies for dealing with the problem of representation discussed thus far involve making those strategies explicit. Stephen A. Tyler, however, proposes an entirely different definition of ethnography which causes the problem of representation to vanish altogether:

A post-modern ethnography is a cooperatively evolved text consisting of fragments of discourse intended to evoke in the minds of both reader and writer an emergent fantasy of a possible world of commonsense reality, and thus to provoke an aesthetic integration that will have a therapeutic effect.[15]

Tyler's definition emphasizes the dialogical nature of ethnography, but here the discourse is between reader and writer rather than between the writer and the culture he studies. Tyler maintains that the experience which matters is not the fieldwork but the writing of the ethnography; the ethnographer does not attempt to represent another culture to the reader, but rather to evoke in the reader a recollection of his own culture. Ethnography is a way to make the familiar unfamiliar and then familiar again.

Tyler's essay, however fascinating, is somewhat inconsistent: although he claims that all ethnographies are post-modern in effect, he also states that the post-modern ethnography has not yet been written and may not even be possible! However, he adds,

The point anyway is not how to create a post-modern ethnography or what form it ought to take. The point is that it might take any form but never be completely realized. Every attempt will always be incomplete, insufficient, lacking in some way, but this is not a defect since it is the means that enables transcendence. Transcendence comes from imperfection not from perfection.[16]

Furthermore, Tyler acknowledges that an author's intention and a reader's interpretation often diverge. While a reader sympathetic to post-modern thinking might see an evocation, a reader with Said's sensitivities might see a representation. According to Tyler's theory, neither reader would be wrong.

Tyler's essay is important for its emphasis on discourse analysis, the ethical character of ethnography, and the relationship between writer, text and reader. But the problem of representation cannot simply be defined away. Said is not interested in a "real or true Orient" because he denies that "the Orient" exists in the first place. Nevertheless, anthropologists and other social scientists want to produce representations of "the other" which are as authentic as possible because "the other," i.e., not-self, does exist unless one is a radical solipsist.

According to Aristotle, "All men by nature desire to know." The applicability of Aristotle's observation to all humans is questionable, but he does identify the distinguishing characteristic of scholars. Francis Bacon observed that "Knowledge is power." The thesis Said defends in Orientalism, the conjunction of Aristotle's and Bacon's propositions, seems to be "All men by nature desire to know in order to acquire power." I do not concur with this thesis: although the results of scholarship certainly can be used to facilitate domination, I do not believe domination of others is the objective of scholarship. If indeed power and domination have a necessary relationship to knowledge, then I suggest it is simply a matter of the knower having power over what is known.

Obviously, power/knowledge can exert external effects, but the use of power/knowledge then raises issues of ethics, and the potential consequences of its use must be considered. Although a scholar has limited ability to control the use by others of the knowledge she produces, she must never forget her responsibilities.

The scholar's first responsibility, then, is to be aware that knowledge, like action, has consequences. When other humans are the subject of knowledge, she must pay attention to the power relations involved. The scholar must always struggle against the procedures of representation, even while recognizing that the problem of representation can never be overcome completely.

This essay has considered some strategies which can be brought to bear on this struggle. First, the role of rhetoric in all forms of writing must be acknowledged and handled in an explicit manner. Second, the tactics of ethnographic realism must be avoided, and the role of the scholar in the procedures of the investigation must be emphasized. Third, polyphony or dispersed authorship should be utilized so that the objects of study may speak for themselves to the greatest extent possible. Plato's dialogues may be a useful model here: through Platonic dialectic, various voices can explore and challenge the representational strategies of other voices.

ts Finally, the reader, too, must become involved in the struggle against the procedures of representation. The author is only a mediator between the reader and the object of knowledge, and responsibilities arising from the use of knowledge thus pass to the reader. The model of Platonic dialectic suggested above emphasizes the role of the reader since the reader must actively engage the text and construct knowledge out of the information provided by the various voices. Reading can no longer be a passive activity.

Gold rule


[1]    Clifford, James, review of Orientalism, History and Theory, 19:204-223 (1980), p. 208.

[2]    Said, Edward I., Orientalism. New York: Vintage (1979), p. 322.

[3]    Ibid., p. 322.

[4]    Ibid., p. 272.

[5]    Frye, Northrop, Anatomy of Criticism: Four Essays. Princeton: Princeton University Press (1957), p. 331.

[6]    Marcus, George E., Rhetoric and the Ethnographic Genre in Anthropological Research, Current Anthropology 21: 507-10 (1980), p. 508.

[7]    Bateson, Gregory, Naven, 2d ed. Stanford: Stanford University Press (1958), p. 260.

[8]    Marcus, George E. and Dick Cushman, Ethnographies as Texts, Annual Review of Anthropology 11:25-69 (1982), p. 29.

[9]    Ibid., p. 29.

[1]0    Clifford, James, On Ethnographic Authority, Representations1(2):118-146 (1983), p. 118.

[11]    Marcus and Cushman, pp.31-36.

[12]    Dumont, Jean-Paul, The Headman and I: Ambiguity and Ambivalence in the Fieldworking Experience. Austin: University of Texas Press (1978), pp. 60-61.

[13]    Ibid., p. 109.

[14]    Shostak, Marjorie, Nisa: The Life and Words of a !Kung Woman. Cambridge, Mass.: Harvard University Press (1981).

[15]    Tyler, Stephen A., Post-Modern Ethnography: From Document of the Occult to Occult Document, p. 125. In Writing Culture: The Poetics and Politics of Ethnography, James Clifford and George E. Marcus (eds.). Berkeley: University of California Press (1986).

[16]    Ibid., p. 136.

Gold rule


Gold rule

Please use the links below to reach other areas of this site:

Last revised: June 5, 2015.

Contact Me   About Me