STOR
®
Forget Culture: Replacement, Transcendence, Relexification Author(s): Robert Brightman
Source: Cultural Anthropology, Vol. 10, No. 4 (Nov., 1995), pp. 509-546
Published by: Blackwell Publishing on behalf of the American Anthropological Association Stable URL: http://www.jstor.org/stable/656256 Accessed: 09/09/2008 12:27
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=black.
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.
http://www.jstor.org
Forget Culture: Replacement, Transcendence, Relexification
Robert Brightman
Department of Anthropology Reed College
I could stay with the transformationalists pretty well, until they attacked my darling, the phoneme.
—Archibald Hill
Introduction
In his article "How Many Revolutions Can a Linguist Live Through?" Hill (1980:74) thus reflected on one by-product of the generativist revolution in linguistics, the critique of the taxonomic phoneme. Hill's lament exhibits a certain topicality for anthropology during a period in which culture, the discipline's longstanding darling, is increasingly embattled. The utility, not to mention the integrity, of the construct of culture—as expounded by Tylor, relativized by Boas, and thereafter refracted through diverse functionalist, ecological, cognitive, transactionalist, structuralist, Marxian, and hermeneutic perspectives—is increasingly being challenged. These recent objections to culture receive both absolutist and historically relativist phrasings, the former holding that the culture concept has been flawed from its inception and the latter that culture—viable enough as a device in earlier historical moments—can no longer engage a world in which social identities, practices, and ideologies are increasingly in-congruent and volatile. What I propose to do here, in brief compass, is to examine the defects of the culture construct as currently represented in anthropological writing, to discuss in somewhat more detail the characteristics of three critiques of the concept (by James Clifford, Lila Abu-Lughod, and Pierre Bourdieu), and finally to reflect on the essentialist ideology at play in the current disciplinary self-consciousness of paradigmatic transition or emancipation. The objective is neither to defend the received culture concept from its critics (indeed, most of the criticisms are well founded) nor to articulate a version of the fatigued message that no new critical perspectives exist in the profession today, that "it's all been said" earlier and better. Rather, my purpose is to indicate how
Cultural Anthropology 10(4):509-546. Copyright © 1995, American Anthropological Association.
509
510  CULTURAL ANTHROPOLOGY
certain contemporary critiques of culture derive their cogency and persuasiveness from a strategic and selective retrospective construction of the meaning of the concept in earlier conditions of anthropology. Reconstituted precisely as the antithesis of theoretical agendas currently in place, culture is presented in this criticism as an antiquity from the past to be transcended or replaced, a kind of conceptual Paschal Lamb whose death is at once the atonement for the elisions and distortions in earlier anthropological practices and the precondition for disciplinary renewal. The current consciousness that the anthropological profession has gotten or should get "beyond" culture can thus be read, in some measure, as the effect of rhetorical strategies that (re)construct an essentialized culture concept in the antipodes of contemporary theoretical orientations.
The manifest signs of anthropological uneasiness with culture are nomen-clatural. First, the diminishing legitimacy of the construct is overtly signaled these days by lexical avoidance behavior. While the adjective "cultural" continues as an acceptable predicate—as, for example, in the title of this journal, or in the designation "cultural studies"—such phrases as "culture" or "Kwakiutl culture" or "the culture of the Nuer" are of increasingly infrequent occurrence. Second, when the word "culture" does occur, it frequently bears the stigmata of quotation marks (see Abu-Lughod 1991; Bourdieu 1977; Clifford 1986; Rosaldo 1989), indexing the writer's ambivalence, self-consciousness, or censure. At the same time, terminological substitutions are presently creating a consciousness of conceptual transition. The symbolism of these innovations is classically Sausurrean in the proportionality presupposed between difference in form and difference in meaning. Such terminological items as "habitus," "hegemony," and "discourse" are increasingly opposed to "culture" as new concept to old, as useful to defective. The performative deployment of these novel phonological shapes seems to be decisive in emergent disciplinary beliefs that both the analytical concepts in play and the fields of social experience that they construct or refer to are qualitatively distinct from those that have gone before.
Reconceptualization or Transcendence
Neither calls for the radical reconfiguration of the culture construct nor events of substantial reconceptualization are anything new in anthropology. Rosaldo's argument for "the remaking of social analysis," specifically "with a view toward redefining the concept of culture" (1989:208), and Appadurai's assertion that "our very models of cultural shape will have to alter" (1990:20) are recent instances. Somewhat more novel in the disciplinary moment are assertions that the culture construct is so hopelessly flawed as to require not rehabilitation but exile, replacement by another analytic construct substantively distinct in definition, characterization, and reference. From this point of view, the concept embodies fundamental misconceptions with respect to the spheres of human experience it represents. So substantial are the disparities between the received construct and these spheres as currently theorized that culture itself is judged expendable, evanescent, or already "dead."
FORGET CULTURE 511
Some 20 years ago, for example, after arguing that culture had outlived its ideological functions, John Moore wrote that "the [culture] concept died in American anthropology, or at least is now in the process of dying" (1974:546). More recently Edward Said asked, "Is the notion of a distinct culture (or race, or religion, or civilization) a useful one, or does it always get involved either in self-congratulation (when one discusses one's own) or hostility and aggression (when one discusses the 'other')?" (1978:325).' By 1988 Paul Rabinow (1988:358) described anthropologists as witnesses to "the concept of culture's partial triumph and contemporary decline." For Joel Kahn (1989:16-17), postmodern criticisms "take us so far away from the classical concept of culture, that it would be far better for the latter to be quietly laid to rest." As with race, says Kahn, "we must similarly abandon the notion of culture" (1989:20). James Clifford, while allowing that "culture is a deeply compromised idea I cannot yet do without" (1988:10), appears nevertheless convinced that we will shortly be obliged to do so.
It may be true that the culture concept has served its time. Perhaps, following Foucault, it should be replaced by a vision of powerful discursive formations globally and strategically deployed. Such entities would at least no longer be closely tied to notions of organic unity, traditional continuity, and the enduring grounds of language and locale. But however the culture concept is finally transcended, it should, I think, be replaced by some set of relations that preserves the concept's differential and relativist functions. [Clifford 1988:274]
In contrast, for Lila Abu-Lughod (1991) the differential function reproduces hierarchy between the metropolitan West and the peripheral rest, reason enough for anthropologists to "write against culture" and to introduce Bourdieu's "practice" and Foucaultian "discourse" as analytical replacements.2
The question of how conceptual reconfiguration is to be distinguished from conceptual replacement will be addressed further on. Here, it is sufficient to note that in this current regime of nomenclatural experimentation, lexical selection can take on new significance, indexing a diversity of perspectives on the resilience of the culture construct. While Rosaldo, for example, talks of redefinition, Rabinow speaks of decline, Kahn of putting to rest, and Abu-Lughod of writing against.
The Defects of Culture
What, then, is wrong with culture in the 1990s? A turn through the literature discloses the following objections, anachronistically arranged below as a sequence of oppositive pairs. The recent critics of culture in no respect comprise an internally homogeneous block, and the objections currently in play represent a complex skein of partially discrete, partially convergent influences from political economy, modernist and postmodernist anthropologies, varieties of feminist writing, cultural studies, and diverse other sources. While it is not possible (in most instances, in any case) to delineate doctrinal "schools" of cultural criticism, certain themes are nevertheless recurrently identifiable and occur as mem-
512   CULTURAL ANTHROPOLOGY
bers of a relatively stable set: holism, localism, totalization, coherence, homogeneity, primordialism, idealism, ahistoricism, objectivism, foundationalism, discreteness, and divisive effects.3 It is of some interest that many of the same objections are invoked both by those who speak of reconceptualization and by those who advocate replacement. Perusal indicates that many of these criticisms are interrelated and also that most possess a complex history both in anthropology and in Western social thought more inclusively. Many objections address not explicit definitions of culture but rather diverse implications and connotations held to be entailed by it. The discussion below is in no way comprehensive and doubtless excludes many criticisms currently being raised in an increasingly interdisciplinary discourse.
Culture Is a Reified Abstraction (versus Practice, Action, and Interaction)
Recently and most influentially raised by Bourdieu (1977:26-27), this objection has a long history in both American and British anthropologies. "Where is this 'culture' which you talk about as doing this and that?" asked the empiricist skeptical of invisible entities in Kluckhohn and Kelly's (1945:81) simulated dialogue. Radcliffe-Brown (1940:10) of course rejected culture, a "fantastic rei-fication of abstractions," in favor of "actually occurring social relations."4 More recently, from a utilitarian posture, Murdock (1972) wrote that "culture, social system and all comparable supra-individual concepts such as collective representation, group mind, and social organism, are illusory conceptual abstractions inferred from the very real phenomena of individuals interacting with one another and with their natural environments." As it has entered into recent debate on culture, largely via Bourdieu's writing, the objection to reification concerns less the ontological status of culture as an abstraction than the attribution to it of an autonomous and regulatory position relative to human agents and their conduct.
Culture Is Ideation or Meaning (versus Behavior, Practice, Action, and Interaction)
Abu-Lughod recommends the replacement of culture with Foucaultian "discourse" which, she says, "is meant to refuse the distinction between ideas and practices or text and world that the culture concept too readily encourages" (1991:147). As earlier adumbrated by Boas and his students, culture inclusively referred to people's ideation, actions, and manufactures. In 1958, Kroeber and Parsons attempted a segmentation of the social field, and in the division of labor thereafter, as Wolf (1980) cogently put it, "sociology was permitted to claim all the social action and anthropology retained the residual values." Thus was culture more narrowly delimited, excluding what people do and make and referencing ideas, symbols, knowledge, beliefs, meanings, values, dispositions, and classifications. Interactive social conduct, thus deprived of its status as an object of analysis, became the material of observation in which culture was objectified and from which it could then be analytically abstracted and formally described.
FORGET CULTURE 513
Ortner's (1984) elucidation of an emergent "practice" orientation in anthropology can be read, in part, as the chronicle of widespread dissatisfaction with this marginalization of conduct as disciplinary subject matter. If we mean by culture a system of symbols and meanings abstracted from institutions and practices, decontextualized, formalized, and valorized as a thing-in-itself, then anthropology's center of gravity has indeed shifted from culture to practice. It is in these terms that Yengoyan (1986) eulogized culture as a casualty of the practice orientation. The framing of recent disciplinary history in these oppositive terms— the substitution of practice for culture, either as ongoing process or as fait accompli—is sustainable only to the degree that the Parsonsian segmentation was taken as authoritative. Murphy (1971), for example, is one among many for whom interactive conduct remained a de jure object of analysis all along, regardless of its exclusion from Parsons-derived definitions of culture. Whether culture is seen to be objectified both in ideational-dispositional domains and in conduct or as an ideational-dispositional field that articulates with conduct, it remains unclear how anthropological indifference to action and agency follows as its necessary entailment.
Culture Is Legalism (versus Agency, Strategy, and Improvisation)
Bourdieu (1977:24-25) has argued that culture (and the Durkheim-derived concept of social structure) "implies the construction of a notion of conduct as execution." That culture commits anthropology to a legalist perspective on conduct as rote enactment of cultural rules has been argued more recently by Abu-Lughod, for whom Bourdieu's practice orientation "is built around problems of contradiction, misunderstanding, and misrecognition, and favors strategies, interests, and improvisations over the more static and homogenizing cultural tropes of rules, models, and texts" (1991:147). By the mid-1980s, Ortner (1984:150) was able to remark shrewdly that this particular facet of practice theory had been rather overdone, with decisions and strategy almost entirely displacing unreflecting reproduction of custom as the disciplinary stock-in-trade.5 As concerns the resilience of the culture construct, the issue would appear to be whether, in any of its received senses, culture can articulate with a theory of conduct that takes account of improvisation and interested strategy.
Culture Is Objectivism or Super organicism (versus Constructivism)
Congenial to, although not identical with, a legalistic perspective on conduct is the conception of culture as a field entirely sui generis, the superorganic of Kroeber or the collective consciousness of Dürkheim. Says Rosaldo, for example, "In this [earlier anthropological] tradition, culture and society determined individual personalities and consciousness; they enjoyed the objective status of systems. Not unlike a grammar, they stood on their own, independent from the individuals who followed their rules" (1989:32). Essential to the practice orientation described by Ortner was a shift of the individual actor (or actors) from the status of empirical exemplar of culture to the status of subject matter.
514  CULTURAL ANTHROPOLOGY
Within this orientation, individuals have become central as analytic constructs, specifically as sites of agency both in relation to conduct and to the reproduction and transformation of systems or structures (in their diverse senses). On the one hand, the legalistic representation of action as behavioral execution of the system is rejected. On the other, the image of culture as an autonomous system is itself discredited. Instead, anthropological writing has increasingly focused upon culture as a system constructed, reproduced, and transformed in and through the ideation and practices of agents, either by deliberate design or as contingent by-product. The distinctive character of this constructivist theme in practice theories, relative to earlier actor-centered approaches in British and American anthropologies (see, for example, Barth 1966; Leach 1954), is the rejection not of economism but of methodological individualism, coupled with an insistence that the culture or system comprises the actors' moves and strategies, in addition to existing as their context or constraint.
In recent criticism, the culture concept is characteristically represented as incompatible with constructivist perspectives. Approaches in ethnomethodol-ogy (see, for example, Button 1991) focus on the microprocessual coordination (or lack thereof) of individuals' situated understandings and practices as these engender social forms (variously construed as norms, institutions, classifications, "structures") in their "objective" facticity. Another perspective links agency and the inception of new cultural forms to the locals' encounters with states and transnational systems. Says Wolf, for example,
Once we locate the reality of society in historically changing, imperfectly bounded, multiple and branching social alignments, however, the concept of a fixed, unitary, and bounded culture must give way to a sense of the fluidity and permeability of cultural sets. In the rough-and-tumble of social interaction, groups are known to exploit the ambiguities of inherited forms, to impart new evaluations or valences to them, to borrow forms more expressive of their interests, or to create wholly new forms to answer to changed circumstances. [Wolf 1982:387]
Recent approaches to local-national-global interactions have foregrounded the locals' constructive effects—deliberate or unintended—on the cultural transformation precipitated out of such contexts, replacing earlier images of their passive subjection to determining exogenous forces (world capitalism, McDonaldization, etc.). Increasingly, attention has shifted from the organization of the local by the global to the reverse. Thus, for example, attention is directed to the "indigenization" of exogenous elements—the discrepant reactions of Israeli Arabs, kibbutzim, and Russian immigrants to the television series Dallas, for example (Hannerz 1989:72-73)—or to the impossibility of borrowing cultural material without reinventing it, as with Filipino musicians who replicate Motown but whose lives are "not in complete synchrony with the referential world which first gave birth to these songs" (Appadurai 1990:3).
FORGET CULTURE 515
Culture Is Generalization (versus Individuals and Events)
"The method," wrote Radin in 1933, "of describing a culture without any reference to the individual except insofar as he is an expression of rigidly defined cultural forms, manifestly produces a distorted picture" (Radin 1933:42). More recently, Abu-Lughod has characterized the culture construct as referring only to typifications and abstractions, at the expense of persons, events, and the qualities of lived experience: "By focusing on particular individuals and their changing relationships, one would necessarily subvert the most problematic connotations of culture: homogeneity, coherence, and timelessness" (1991:154). As used here, timelessness indicates the culture construct's failure to engage the contingent temporal character of particular events in progress and people's subjective experience of them.
Culture Is Holistic (versus Fragmentary)
Of the currently identified defects of culture, "holism" is the most commonly invoked, and the term is used to refer to several distinct but interrelated ideas. In recent cultural criticism, the attribution of holism means variously that the culture construct ignores intracultural diversity and variation (see Culture Is Homogeneity, below), elides contradictory or conflictual elements and represents the constituent forms of culture as globally interarticulated (see Culture Is Coherence and Totalization, below), or postulates that cultures are discrete entities (see Cultures Are Discrete, below).
At a more general level, the imputed wholeness of cultures or societies— the interrelatedness of differentiated constituent forms which then comprise a bounded entity—has increasingly been examined as the invented artifact of anthropology's theoretical projects and literary practices, with the implication that no such properties characterize social fields "out there." Wholeness is, in a now-conventional idiom, constructed rather than discovered, and this objection to culture cross-references others that draw upon rejections of foundationalism (see Culture Is Foundationalism, below). Curiously, these arguments often begin with the crypto-positivist observation (see Friedrich 1992) that anthropologists cannot experience wholeness as sense data in the ethnographic field situation, that wholeness or the interarticulation of collocated cultural forms is not the material of empirical observation. Thus Tyler (1986:132) speaks of "these [ethnographic] invocations of holism, of functionally integrated systems" as "literary tropes, the vehicles that carry imagination from the part to the whole, the concrete to the abstract."
Life in the field is itself fragmentary, not at all organized around familiar ethnological categories such as kinship, economy, and religion, and except for unusual informants like the Dogon sage Ogotommeli, the natives seem to lack communicable visions of a shared integrated whole; nor do particular experiences present themselves, even to the most hardened sociologist, as conveniently labeled synecdoches, microcosms, or allegories of wholes, cultural or theoretical. [Tyler 1986:131]
516  CULTURAL ANTHROPOLOGY
Tyler here both questions the integrity of the traditional monographic categories and refers us to the locals' perspective: Where is holism if the folk themselves have no experience of it? Similarly, Thornton characterizes cultural or social wholeness as the product of objectified tropes: sociological holism involves the mistaken transference of mereological (spatiotemporal part-whole) relationships from such tropes of society as organic bodies, machines, trees, and buildings to society itself (1988:293). Thornton sees wholeness as additionally created through "mistaken analogy with the [ethnographic] text whose parts— namely chapters, titles, subheadings, paragraphs, and so on—are truly constitutive of the textual whole" (1988:291). Herbert, who follows Raymond Williams (1983) in pursuing the genealogy of the culture construct into 19th-century British literary traditions, likewise levels the criticism that holistic interrelations among cultural forms are nonobservable, as with his assertion that the Tylorian culture concept "posits a metaphysical, immaterial substance, complex wholeness, that is not commensurate with observed data and can only be perceived (if at all) by a kind of extrasensory perception" (1991:14; emphasis added). For Herbert, this "nonvisibility" of cultural wholeness engendered an epistemologi-cal anxiety of which anthropological positivism was the compensatory expression. For Clifford (1992) and others (see Culture Is Localism, below), the "localizing strategies" by means of which anthropologists invent boundaries around cultures are linked to the exclusion of intercultural interaction from ethnographic writing. He suggests, for example, that bounded locales—village field sites, for example—become points of reference for representing a "whole culture" whose external connections are then elided (1992:98).
Culture Is Homogeneity (versus Intracultural Variability)
The culture construct, recent critics observe, fails to represent or theorize adequately the heterogeneous character of disposition and conduct within cultures or societies. Thus, one reads Rosaldo's (1989:207) assertion that "human cultures are neither necessarily coherent nor always homogeneous" and Abu-Lughod's (1991:154) inclusion of homogeneity as among the "most problematic connotations of culture." Rosaldo's (1989) elucidation of "cultural borderlands"—such intercultural and intracultural spheres as interaction across ethnic boundaries, movement between discrepant statuses and relationships in daily life, and disparities attendant upon difference in gender, age, status, and life experience—addresses questions of both homogeneity and coherence. In question is what Romney et al. (1986) call the "division of labor in who knows what," the nonuniform distribution of knowledge and conduct among individuals and subsets of individuals occupying different positions in the social field. Ethnicity, occupation, age, class or status group, and gender are the typically foregrounded sites of intracultural diversity. An exceptionally interesting perspective on such diversity is offered by Drummond's (1980) analysis of "plural" cultures in Guyana. As against recouping cultural homogeneity by segmenting heterogeneous cultural participants into homogeneous subsets, Drummond's "creole" culture concept explains, for example, not only that men do X and women do Y but that
FORGET CULTURE 517
women also know how to do X and sometimes do so, that individuals are the loci of diverse repertoires.
Culture Is Coherence and Totalization (versus Disorder, Contradiction, and Contestation)
"Coherence," among the most ubiquitous epithets in recent critiques, seems to reference images of institutional or logical consistency and order. Rosaldo, for example, writes that "[culture] emphasizes shared patterns at the expense of processes of change and internal inconsistencies, conflicts and contradictions" (1989:28; see also Abu-Lughod 1991:154; Clifford 1988:232). The recent message is clearly that the culture construct falsely ascribes coherence to fields of social experience which are incoherent or, at least, less coherent than they have been imagined to be. The claim is not only that cultures are internally diverse (versus homogeneous) but that they are disordered, contradictory, and sometimes disputed.
Contestation, entropy, and chaos have long since displaced coherence and integration as the privileged disciplinary themes, and a variety of distinguishable modes of disorder figure in critical writing on culture. At one level, coherence concerns the imputed inability of the culture construct to refer to what anthropologists (if not demonstrably their local interlocutors) deem to be coexisting but contradictory modes of talk and action—judicial torture by the Apollonian Zunis, for example. At another, the term totalization can refer to one aspect of coherence, the posited global interarticulation of all the concurrent forms in a culture. Ortner, writes, for example, that the concepts of system and structure carry an "implication of singularity and of totalization: a 'society' or a 'culture' appears as a single 'system' or as ordered by a single 'structure,' which embraces (or pervades) virtually every aspect of that social and cultural universe" (1990:43). Totalization may refer to multiple interconnections among cultural forms or their regimentation by a single dominant component, principle, or design pervasive throughout the system. Anthropological totalizations have taken diverse forms: the configurationalism of Benedict, the integration of Malinowski and of structural functionalism, the "order of orders" of Levi-Strauss, Marxian modes of production, and, most recently, that whole of wholes in which all others are subsumed, the encompassing world system. In terms of functional integration or of thematic configuration, cultures are noncoherent to the degree that some constituent elements lack systemic integration and simply coexist in collocation. Perspectives skeptical of totalization need not entail rejection of all interconnections between coexisting cultural forms. Rather it is a question of more or less interrelatedness, of allowing scope for such less orderly arrangements as contradictory or unaligned elements, more or less loosely articulated subsystems, and redundancy.
A different index of noncoherence concerns cultural forms that coexist in oppositive or contradictory relation: "Culture as multiple discourses, occasionally coming together in large systemic configuration, but more often coexisting within dynamic fields of interaction and conflict" (Dirks et al. 1994:4). On the
518   CULTURAL ANTHROPOLOGY
one hand, there is the Two Crows phenomenon, the question of disagreement as to what does or should exist in society and the cosmos.6 Such disagreement becomes especially relevant as it pertains to valued stakes and becomes, therefore, the basis of tacit or overt contestation and debate. Contestation engages, in turn, with the Gramscian question of whose culture shall be dominant and with whose well-being the officially valorized forms most congenially articulate.
The socially universal allocation of persons to distinct groups with unequal or differentiated access to material or symbolic capital has multiple relevance to questions of cultural homogeneity and coherence. That cultures or social structures are socially, cognitively, and ecologically good for all the people who participate in them has been a fundamental assumption, explicit or tacit, of much anthropological writing. The anthropological culture construct, it could be claimed, thus guarantees a certain obliviousness to contending interests, to inequalities predicated upon age, gender, class, status group, or position in junctures of intersocietal interaction. The consequences are then, at once, a failure to register the diversity of practices and discourses exhibited by agents occupying different sites in the system, and a blindness to politico-economic criteria—both within and beyond the relevant social boundaries—as exceptionally privileged loci of cultural organization. In specific relation to the culture concept, a significant theme in anthropological writing on gender has been to relate gender-linked intracultural differences to the varying positions men and women occupy in local schemes of authority and prestige (see, for example, Ardener 1975).
With respect to such concerns, Gramsci's (1971) concept of hegemony has been influential in relating intracultural differences to the politico-economic position of agents. As Raymond Williams put it, hegemony "goes beyond 'culture' in its insistence on relating the 'whole social process' to specific distributions of power and influence" (1977:108-109). Not the least of the concept's advantages is its attention to difference and contradiction not only between but within the cultural repertoires of groups or individuals occupying like positions, as when Gramsci writes of "two theoretical consciousnesses (or one contradictory consciousness)" in which the "masses" participate (1971:326). Ortner (1990:44-45; see Ortner 1984) subsequently used the concept not as a substitute for but as a characterization of culture: in any society, certain cultural forms may be dominant or hegemonic, others counterhegemonic, and still others simply present in nonconflictive relation to the others.
Cultures Are Discrete (versus Overlapping and Unbounded)
Perhaps the major problem of holism concerns the delimitation of boundaries between cultures. Early on, Lowie, exhibiting much the same lack of enthusiasm for cultural boundaries that Sapir and Goldenweiser expressed for the su-perorganic, wrote, "In defiance of the dogma that any one culture forms a closed system, we must insist that such a culture is invariably an artificial unit segregated for purposes of expediency" (1937:235). Recent culture criticism represents the culture concept as postulating objective boundaries and as eliding interaction and resemblance between cultures. Asks Thornton, for example,
FORGET CULTURE 519
Is social life radically continuous, only broken by the periodicity of text? ... Apparent boundaries, then, such as those that define nations, ethnic groups, age-grades, or classes, are seen to be relative to time, the observer, or to each other, and thus not "really" there at all. In any case, we cannot always assign them unambiguous ontological status. [1988:299]
With respect to cultural boundaries, two conventional anthropological strategies of delimitation can be identified. First, it may be argued that people exhibit qualitatively distinguishable constellations of cultural forms, identifiable zones of sameness and difference. Coextensive distributions of traits or elements thus specify cultures and the boundaries between them. Fortes (1949:2), for example, wrote, "The Tallensi have more in common among themselves .. . than the component segments of Tale society have with other like units outside of what we have called Taleland." In each case, intracultural sameness and intercultural difference are assumed to exceed intracultural difference and inter-cultural sameness, and this becomes the justification for discriminating discrete cultures. The second and more common strategy deals with the problem of boundaries between cultures by deriving them ready-made from boundaries between social collectivities. Thus the boundaries of a culture have been guaranteed by the cultural criteria—variously ethnic, political, linguistic, or especially territorial—defining the boundary of a social collectivity in which the culture is contextualized, the society whose culture it is. Kwakiutl culture is both a culture and Kwakiutl culture neither because its content is unique (many elements are present elsewhere) nor because of its unique configurational gestalt but because it is contextualized in a social collectivity delimited by territory, identity, linguistic criteria, and the classifications anthropologists have made of these. The transposition of the problem of demarcation from the fields of cultural practice and discourse themselves (which could ideally delimit bounded social collectivities) to bounded social collectivities (which then return the favor and delimit discrete cultures) is not an advance in precision. Although Hastrup has asserted confidently that "unlike a society which is an empirical entity, culture is an analytical implication" (1990:45-61), society, as an anthropological construct, has exhibited lability and instability in like measure with culture and exhibited similar problems of delimitation.
An exceptionally privileged and authoritative diacritic of cultural boundaries has been local conceptions of social likeness and distinctiveness. To the degree that the local reckonings of we-ness and other-ness ("We are Karoks, they are Hupas") are reproduced in anthropological representations of boundaries ("Karok society/culture is distinct from Hupa society/culture"), the work of delimiting cultural units is displaced onto the people themselves. And, while such reckonings indeed construct and reproduce tangible zones of difference, the native point of view may be inadequate for many purposes. Recent criticism has foregrounded the degree to which criteria of delimitation are multiple, redundant, incongruent, and overlapping. From one point of view, it is not that there exist no boundaries that could delimit cultures or collectivities but that there is a superabundance of them, no two of which segment the social or cultural field
520   CULTURAL ANTHROPOLOGY
in precisely the same way. Some years ago, Southall wrote that "the close identity of language, culture, and society (if it ever existed) is now blurred and has become a series of alternatives" (1970:29-30; see Drummond 1980:354). For Southall, "The representation of adjacent stateless societies as a neatly discrete series of named units is to misunderstand and misrepresent them" (1970:40-41). The culture or "social structure" of the 1990s is one in which the distributional fields of social self-identification, social interaction, cultural resemblance, linguistic codes, polity, and territory are characteristically noncongruent.
The particular facet of this noncongruence foregrounded in recent critical writing on culture concerns what was earlier called diffusion and has now been relexified as cultural transfer and flow: movement of cultural forms across social and cultural boundaries. For Joel Kahn (1989), the culture concept has scientific utility only if it classifies human beings into discrete groups. His argument is predicated upon an analogy between the relationship of physical traits to nominally discrete races and the relationship of cultural traits to nominally discrete cultures. Describing the arbitrary character of racial classifications based on trait distributions, Kahn argues that the same is true of culture.
I would argue that the notion of a culture is formally identical [to race]; that those markers used to assign people to one or the other of the world's cultures are equally ambiguous, and are far from enabling us to demarcate discrete, to say nothing of unchanging, cultural units except by reference to some boundary which is purely spatial and, hence, largely arbitrary, especially in the modern world. [Kahn 1989:18-19]
Kahn argues, for example, that the Minangkabau of Sumatra do not possess a discrete culture because there exists no set of traits that uniquely characterize all Minangkabau and no other group. Cultures overlap by sharing traits; by virtue of this overlap they are not discrete; and because they are not discrete they do not exist. Thus there are no cultures, only overlapping distributions of traits in space. Kahn astutely identifies the problem of whether cultural boundaries and thus distinct cultures can be identified without reference to the social collectivities in which cultures are conventionally contextualized by anthropologists. Given the incongruent and overlapping distributions of cultural forms themselves, some specification is required of the criteria by which particular cultural forms are assigned significance as the discrimina of different cultures.7
From a different perspective, the absolute incompatibility of cultural boundaries with intercultural resemblance and borrowing remains undermoti-vated in Kahn's argument. The claim that cultural borrowing precludes identification of distinguishable cultures ignores the configurational process of indi-genization. The Minangkabau, to be sure, may watch Dallas, but it is questionable that they experience it in the same terms as its (multiple) American audiences. Cultural transfer, from this perspective, compounds boundaries as much as it erodes them; movements of "exotic" materials reinforce as well as erode local reckonings of "we" and "they."8
FORGET CULTURE 521
Appadurai (1988, 1990) has formulated somewhat similar criticisms of cultural discreteness, suggesting that the culture concept entails inattention to regional and global interactions. Appadurai's elucidation of "ethnoscapes" concerns incongruent boundaries, the increasingly discrepant distributions of social collectivities, ethnic and social identities, manufactures, and ideologies: "Our very models of cultural shape will have to alter, as configurations of people, place and heritage lose all semblance of isomorphism. Recent work in anthropology has done much to free us from the shackles of highly localized, boundary-oriented, holistic, primordialist images of cultural form and substance" (1990:20). The dominant image in Appadurai's exposition of global culture is motion: the dispersion of groups through transnational space, the transfer of cultural materials between multiple spatial loci, the movement of people and of culture across boundaries. From this perspective, cultures lack boundaries not only because their forms and contents are increasingly mobile but because emergent cultural materials are precipitated out of the interaction between collectivities— localized or dispersed—within the global ecumene. Thus such elements of culture as desires, aspirations, and social identities are no longer reproduced trans-generationally via enculturation but created and continually reconfigured in changing contexts of cultural transmission.
The obvious question is again how (or whether) one delimits cultures in such a continuum. Appadurai suggests
that we begin to think of the configuration of cultural forms in today's world as fundamentally fractal, that is, as possessing no Euclidean boundaries, structures, or regularities. Second I would suggest that these cultural forms, which we should strive to represent as fully fractal, are also overlapping, in ways that have been discussed only in pure mathematics (in set theory for example) and in biology (in the language of polythetic classifications). Thus we need to combine a fractal metaphor for the shape of cultures (in the plural) with a polythetic account of their overlaps and resemblances. [1990:20]
Appadurai goes on to pose the relevance of chaos theory for anthropology, an appeal to the authoritative practices of the hard sciences, but this time one in which process, flow, and uncertainty rather than stability and structure are the privileged foci. Here, cultures remain integral enough to be pluralized and to exhibit overlaps and resemblances. The image of a fractal configuration of cultural forms suggests, however, that no segmentations of human populations— whether by territory or by polities or by self-defined social collectivities—qualify as criteria for the delimitation of cultural boundaries. The collapse of these conventional reference points produces a salutary sense of vertigo. Since people everywhere persist in classifying themselves and others, often in multiple ways, as members of distinguishable if not discrete social collectivities with distinguishable if not discrete practices, the question remains as to whether this particular facet of culture(s) can continue to serve, as it typically has in the past, to contextualize representations of plural cultures.
522   CULTURAL ANTHROPOLOGY
Culture Is Localism (versus External Interaction)
As an attribute of the culture concept, localism conveys at once images of cultures as discrete territorial units and as isolated from external interactions, problems closely aligned with those to discreteness (see above) and primordial-ism (see below). By fabricating images of exotic provincialism and ignoring external interactions, culture guarantees both inattention to borrowed forms (Boas did not study Kwakiutl Christianity) and obliviousness to processes of invention and transformation arising from exogenous cultural transfers and the politico-economic contexts in which these are embedded. Anthropology's interchange with political economy (see Marcus and Fischer 1986:77-110; Ortner 1984:141-144) has directed attention to the creation of cultural forms in the context of interactions with local, regional, and global systems and to the increasing frequency of cultural transfer within the global ecumene.
In their strongest phrasing, arguments from political economy hold that the very existence of cultures—and of the "tribes" or "societies" in which they are contextualized—can be the emergent effect of such interactions. Thus, Southall (1970:35) wrote that, much as postcolonial states were fabricated at conference tables, many African "tribes" came into existence "through a combination of reasonable cultural similarity with colonial administrative convenience." More recently, Wolf (1982) has argued that both the existence and the content of what we call cultures are the product of the organizing forces of world capitalism. For Wolf (1982:388), there are no "self-contained societies and cultures"; those that may once have existed have long since become the appendages of the world system.
[Furthermore], if we think of such interaction ["historically changing, imperfectly bounded, multiple and branching social alignments"] not as causative in its own terms but as responsive to larger economic and political forces, the explanation of cultural forms must take account of that larger context, that wider field of force. "A culture" is thus better seen as a series of processes that construct, reconstruct, and dismantle cultural materials, in response to identifiable determinants. [Wolf 1982:387]
In the wake of Wolf's contributions, assertions that cultures are not pristine and isolated have proliferated, hand in hand with claims that the anthropological culture construct represents them as being so. Marcus, for example, asks, "What is holism once the line between the local worlds of subjects and the global world of systems becomes radically blurred?" (1986:171). For Appadurai, the dominant metaphor is confinement: while there never existed groups "confined to and by the places to which they belong, groups unsullied by contact with a larger world," anthropology's localizing strategies have effected a representational "'incarceration' of non-Western peoples in time and place" (Appadurai 1988:38-39; see Abu-Lughod 1991:146).9 Rosaldo (1989:44), putting a spin on Cora DuBois's characterization of anthropology's topical disarray as a "garage sale," characterizes the latter instead as a."precise image for the postcolonial situation where cultural artifacts flow between unlikely places, and nothing is
FORGET CULTURE 523
sacred, permanent or sealed off." For Rosaldo, "the guiding fiction of cultural compartments has crumbled. So-called natives do not 'inhabit' a world fully separate from the one ethnographers 'live in.' Few people simply remain in their place these days" (1989:45). Likewise, for Clifford (1992:97), "The people studied by anthropologists have seldom been homebodies."
A second theme links localism to stereotypy. Said (1978:332), for example, characterizes as "highly debatable" the idea that "there are geographical spaces with indigenous, radically 'different' inhabitants who can be defined on the basis of some religion, culture, or racial essence proper to that geographical space." Appadurai uses "totalization" to refer to "making specific features of a society's thought or practice not only its essence but also its totality" (1988:41), a synecdoche he traces from German romanticism, Hegelian holism, Marxism, and Mauss's concept of total social phenomena. Anthropology selectively identifies cultural forms that stereoptypically epitomize groups—or places inhabited by groups—and differentiate them from others: "Hierarchy is what is most true of India and it is truer of India than of any other place" (1988:40).
Enthusiasm for reanalyzing all non-Western cultures and cultural materials as the by-product of European expansion has, of course, recently begun to subside in the face of more nuanced assessments. Says Clifford, for example, "Anthropological 'culture' is not what it used to be. And once the representational challenge is seen to be the portrayal and understanding of local/global historical encounters, co-productions, dominations, and resistances, then one needs to focus on hybrid, cosmopolitan experiences as much as on rooted, native ones" (1992:101).
Culture Means Ahistoricity (versus History and Change)
The question of the imputed ahistoricism of the culture construct is logically distinguishable from localism, but the two are typically conjoined. "Culture," for Clifford (1988:235), "is enduring, traditional, structural (rather than contingent, syncretic, historical)." For Rosaldo (1989:28), "[culture] emphasizes shared patterns, at the expense of processes of change." He goes on to describe his and Michelle Rosaldo's resistance, during their second period of field research in 1974, to studying external influences on the Ilongots: "The broad rule of thumb under classic norms to which Michelle Rosaldo and I still ambivalently subscribed seems to have been that if it's moving it isn't cultural" (1989:208). The "then-fading classic concept of culture" could not, for them, encompass "flux, improvisation, and heterogeneity" (1989:208).
The claim that the culture concept is ahistorical exhibits distinct modalities. First, it can mean that the concept represents other people's cultures as lacking an internal historical dynamic: the Asiatic mode of production of Marx, for example, or the "cold societies" of Levi-Strauss. Second, it can mean that the concept represents other people's cultures as lacking notions commensurable with those of Western history (the "past," irreversible event sequences, humans as constructive agents, etc.). Third, it can mean that the culture concept is defini-tionally synchronic, necessarily excluding study of past conditions of particular
524  CULTURAL ANTHROPOLOGY
cultures and therefore precluding attention to external contacts and cultural change. As Rosaldo suggests, in certain disciplinary contexts the phrase "Ilon-got culture" could more readily refer to head-hunting than whatever Ilongots were making of a recently introduced Christianity; the latter was "acculturation." Rosaldo (1989:28) specifically qualifies his generalization, noting both the obvious historicism of the Boasians and the reemergence of history in the 1980s as a major theoretical orientation in anthropology. These exceptions cast doubts on the necessary ahistoricism of the culture construct. One can add that there has existed a certain continuity in American attention to history and to so-ciocultural change, manifested variously in acculturation studies, in neoevolu-tionary writing, and in the congeries of approaches labeled ethnohistory.
Culture Means Primordialism (versus Syncretism and Invention)
The rubric "primordialism" encompasses several of the criticisms enumerated above, specifically as they address a set of related ideas that commit the culture construct to spurious notions of authenticity. Specifically, the authentic "native" or the authentic culture is contextualized in a social collectivity that is local (versus mobile or dispersed) and isolated (versus externally interactive). Additionally, the forms comprising cultures are authentic insofar as the forms are indigenous (versus being borrowed) and are continuously and transgenera-tionally reproduced in unchanged form (versus being improvised or invented) (see Appadurai 1988:37; Clifford 1988). It is presumably to such images that Appadurai (1990:20) refers when he speaks of "holistic, primordialist images of cultural form and substance." The antithesis to the primordialist culture construct is one whose forms are selectively and deliberately appropriated from a heterogeneous assortment of repertoires. In a world of accelerating cultural transfer, "culture becomes less what Bourdieu would have called a habitus (a tacit realm of reproducible practices and dispositions) and more an arena of conscious choice, justification, and representation, the latter often to multiple and spatially dislocated audiences" (Appadurai 1990:18).
Culture Is Representationalism and Foundationalism
Culture or cultures, recent criticism states, are anthropologically imagined as though they are objects comparable to those theorized by the physical sciences. Antipositivist counterassertions that cultures are not such objects have recently proliferated. Tyler, for example, derides "the absurdity of 'describing' nonentitites such as 'culture' and 'society' as if they were fully observable, though somewhat ungainly bugs" and asserts that "in ethnography, there are no 'things' there to be the object of a description... there is rather a discourse, and that too, no thing" (1986:130-131). Similarly, in Rosaldo's parodic exemplification of anthropology's "classic norms," "The product of the Lone Ethnographer's labors, the ethnography, appeared to be a transparent medium. It portrayed a 'culture' sufficiently frozen to be an object of 'scientific' knowledge" (1989:31).
FORGET CULTURE 525
A more fundamental criticism, one linked to postmodern misgivings about the representational project in modernist science, is that anthropology misrecognizes culture and cultures, its constructed inventions, as discoveries about and representations of external reality. As seen from within the postmodern sensibility (see, for example, Fabian 1990; Lyotard 1984; Tyler 1986), hermeneuticians as much as ecologists appear committed to a naive representational realism insofar as they conceptualize the culture concept as mirroring independently existing social fields "out there" and thus perpetuate a spurious dualism of knowing anthropologist-subject and knowable other-object. In anthropology since the "crisis," the recurring aria has been that what we have taken to be our representations of cultures, more or less valid by truth value assessments, are constructions, precipitated out of field encounters, disciplinary genre conventions, theoretical perspectives, and Western literary traditions, especially as these are conditioned by politico-economic criteria. Culture shares this status equally with ethnography, with other constructs in social theory, and with modernist science in general. Fou-cault's preoccupations with the socially and historically situated process of knowledge production—how truth is constituted, what can count as a fact— have been central to the perspective. At its most playfully solipsistic, such criticism asserts an absolute discontinuity between cultures and what we can know and say of them, as with Tyler's claim that "no object of any kind precedes and constrains the ethnography. It creates its own objects in unfolding and the reader supplies the rest" (1986:138). Kahn writes that "the view that what we as anthropologists call culture is something that we produce, in definitive social and historical contexts, seems to me to take us so far away from the classical concept of culture, that it would be far better for the latter to be quietly laid to rest" (1989:16-17). Similarly, Herbert has asked, "Where did so problematic, so self-defeating a concept [culture], one vulnerable to so many 'fairly obvious' objections, one which leads in practice to such dubious scientific results—where did such a concept originate, and in obedience to what influences?" (1991:21).
The implications of the various criticisms of representation for the viability of the culture construct remain rather less clear-cut than Tyler or Kahn indicate. In at least certain of its expositions, skepticism about representation does not entail an absolute relativization of judgment, an inability to assess competing truth claims (see Rabinow 1986:236-238; Watson 1991:83), and the absolutist dichotomy premised in much postmodern writing between the construction and the representation of culture remains, of course, philosophically undermotivated for many nonconverts. Many anthropologists would probably affirm Ortner's (1984:143) argument that the representational project is, at the very least, worth attempting. An interesting desideratum here is greater attention to non-Western theories of representation and to local constructs that seem to reference the spheres of social experience we call cultures.
526  CULTURAL ANTHROPOLOGY
Culture Means Difference and Hierarchy (versus Resemblance and Humanism)
While writing from the postmodern turn has foregrounded literary and tro-pological influences on the invention of the culture construct, criticism from political economy has emphasized a more specifically ideological genesis. Culture, from this point of view, was forged in the crucibles of nationalism and colonialism. The particular development of the construct in the present century, for example, has been linked by Wolf (1982:385) to the sovereign aspirations of emergent European polities and to struggles for domination between them. More commonly, the culture construct and its modes of ethnographic reference have been represented as conditioned by the hegemonic politico-economic articulation of the West with its diverse peripheries and as inventing and reproducing images of difference between Western self and non-Western other congenial to its perpetuation. The ahistoricism, discreteness, homogeneity, totalization, and localism anthropologically ascribed to others' cultures have all been interpreted in these terms. Culture, it is argued, guarantees a certain obliviousness to the anthropologist's own position in the relevant fields of power.
While certain of these issues were raised earlier (see Hymes 1974), recent criticism derives most proximally from Said's (1978) reflections on Orientalism, from his speculation that the spatially homogeneous and temporally static Orient is a constructed image, one comprising both an antithesis to Western progress and a terrain for imperialist projects. The lesson for anthropology is that what is true of Orientalism is true of culture and cultures. For Rosaldo, stasis, homogeneity, and harmony invite colonial supervention or Western development:
The Lone Ethnographer depicted the colonized as members of a harmonious, internally homogeneous, unchanging culture. When so described, the culture appeared to "need" progress or economic and moral uplifting. In addition, the "timeless traditional culture" served as a self-congratulatory reference point against which Western civilization could measure its own progressive historical evolution. [Rosaldo 1989:31]
Similarly, for Dirks (1992:3), "Even as much of what we now recognize as culture was produced by the colonial encounter, the concept was in part invented because of it." Appadurai (1988:43), in the tradition of Said, identifies an "exoticizing" movement in anthropology, the valorization of difference as the sole basis for comparison between Western metropolitan selves and non-Western peripheral others (see also Abu-Lughod 1991; Kahn 1989:20-21).
The Construction of Defective Culture
Certain of the criticisms sketched above take assertions about the culture concept's past definitions and connotations as evidence for claims that it should be, is being, or has already been displaced by more useful substitutes. Such criticism presupposes some stability and uniformity in past definitions. It can plausibly be argued that a set of core meanings has been commonly (if not univer-
FORGET CULTURE 527
sally) present in anthropological definitions and characterizations of culture. Stocking (1982:230), for example, identifies the essential elements of the Boasian construct as "historicity, plurality, behavioral determinism, integration and relativism." Likewise, assertions that culture is nonbiogenetic, contextualized in some ethnic or social collectivity, and global—that it comprises a "whole way of life"—have been and continue to be common meanings of the construct. Culture, nonetheless, has undergone a career of multilinear development, and if we talk of a single construct it is one exhibiting exceptional synchronic and dia-chronic lability. Thus, more iconoclastically, culture can be represented as "a class of phenomena conceptualized by anthropologists in order to deal with questions they are trying to answer" (Kaplan and Manners 1972:3) or as representing "a shared terminology rather than substantial conceptual agreement" in the discipline (Kahn 1989:6). There exist, of course, both resemblances and differences in how anthropologists have defined the concept and characterized what it refers to. Recent arguments that the culture construct is evanescent and dispensable foreground conceptual stability rather at the expense of lability, presupposing that there existed in the past and into the present a culture construct with a determinate definition, now discredited. Such stability in definition is not readily apparent. When we encounter arguments today that the culture construct should be abandoned, we must naturally wonder which of its formulations from among all the possible ones we should be rid of. Such criticism affords an exceptional opportunity to investigate what it can mean to say of a construct that it is dispensable when there has existed so little disciplinary consensus as to its definition, characterization, and reference.
The critics' assertions regarding what culture means and references in anthropology are themselves questionable as principled representations of the concept's complex and heterogeneous intellectual history. Put another way, these images of culture are themselves inventions rather than representations. Recent critics, through selective forays into disciplinary history, have retrospectively synthesized images of the culture concept, devising essentialist representations of what culture has signified or connoted in its anthropological usages. This culture concept—as thus reconstructed—exhibits, to be sure, from the perspective of social theory in the 1990s, defects and biases of sufficient gravity to warrant dispensing with it, if indeed they are all that those who theorized culture in the past ever had to offer us. The rhetorical strategies utilized in this recent cultural criticism identify as essential to the culture construct just those assertions about cultural experience which disciplinary practice today construes as logically or empirically misconceived. Such characterizations identify as constitutive of the culture construct certain earlier meanings (coherence, ahistoricism, homogeneity) uncongenial to contemporary disciplinary beliefs, while at the same time selectively excluding certain other earlier meanings (constructivism, disorder, diversity) that happen, ironically, to be continuous with such beliefs. The culture concept is therefore retrospectively positioned in the antipodes of the theoretical perspectives currently in place.
528   CULTURAL ANTHROPOLOGY
Given the centrality of the culture concept in anthropology, considerable symbolic capital accrues to authoritative pronouncements that it is changing or that it is evanescent. The question then arises as to whether an expendable "straw culture" is thus being retrospectively devised. It is not that the defects foregrounded in recent cultural criticism are fabrications or that they have not been integral to many influential definitions and deployments of culture. But the recent literature exhibits, to say the least, a strategic inattention to the many (equally influential) exceptions and alternatives to the essentialized concept thus reconstructed.
Culture, Authentic and Inauthentic
Current objections to culture associated with political economy and postmodern writing intersect in the writing of James Clifford. Clifford neither writes the concept off as hopeless nor issues performative announcements of its demise. Rather culture emerges in his criticism as a construct valuable for its pluralism and relativism but seriously flawed in its primordialist assumptions. Clifford's reflections on culture foreground both the textually-theoretically constructed (versus discovered) character of the concept and the historically constructed (versus primordial) character of its referents.
Cultures are not scientific "objects" (assuming such things exist, even in the natural sciences). Culture, and our views of "it," are produced historically and are actively contested. [1986:18]
If culture is not an object to be described, neither is it a unified corpus of symbols and meanings that can be definitively interpreted. Culture is contested, temporal, and emergent. Representation and explanation—both by insiders and outsiders— is implicated in this emergence [1986:19]
Cultures cannot be scientific objects if they are historical products, and neither can representations of culture be scientific because they are also historical products. Clifford, nonetheless, implies that there are cultures "out there" and that our constructs can exist in some (perhaps improvable) referential relationship to them.
Clifford's critique of primordialism (he does not privilege the word) in the culture concept focuses on the contextualization of cultures in discrete populations and territories, on the elision of intercultural interaction and intracultural contradiction, and on objectivist assumptions of unbroken transgenerational continuity. Cultures that are spatially dispersed, invented, or externally interactive are, Clifford argues, proportionately inauthentic by conventional anthropological criteria. He cites, for example, Margaret Mead's sobering reflections during her 1932 sojourn with the Mountain Arapesh. Remarking, "We are just completing a culture of a mountain group here in the lower Torres Chelles," Mead goes on to complain that the locals are excessively receptive to external influences: "A picture of a local native reading the index to the Golden Bough
FORGET CULTURE 529
just to see if they had missed anything, would be appropriate" (quoted in Clifford 1988:232). For Clifford, this exemplifies how
this culture with a small c [the pluralized culture concept] orders phenomena in ways that privilege the coherent, balanced, and "authentic" aspects of shared life. Since the mid-nineteenth century, ideas of culture have gathered up those elements which seem to give continuity and depth to collective existence, seeing it whole rather than disputed, torn, intertextual, or syncretic. Mead's almost postmodern image of "a local native reading the index to The Golden Bough just to see if they had missed anything" is not a vision of authenticity. [1988:232]
During the course of a suit filed by the Mashpee [Wampanoag] Indians of Cape Cod in 1977, Clifford examined the fortunes of the culture concept in a legal context where cultural authenticity figured as a decisive factor in the determination of the plaintiffs' "tribal" status. Clifford's point is that primordialist definitions of culture confounded realistic assessment of the Wampanoags identity and thus of their claim. Specifically, cultural forms deriving from Euro-American sources were seen as diluting or contradicting the authenticity of Indian identity.
[Culture] was too closely tied to assumptions of organic form and development. In the eighteenth century culture meant simply "a tending to natural growth." By the end of the nineteenth century the word could be applied not only to gardens and well-developed individuals but to whole societies. Whether it was the elitist singular version of a Matthew Arnold or the plural, lower-case concept of an emerging ethnography, the term [culture] retained its bias toward wholeness, continuity, and growth. Indian culture in Mashpee might be made up of unexpected everyday elements, but it had in the last analysis to cohere, its elements fitting together like parts of a body. The culture concept accommodates internal diversity and an "organic" division of roles but not sharp contradictions, mutations, or emergences. It has difficulty with a medicine man who at one time feels a deep respect for Mother Earth and at another plans a radical real estate subdivision. It sees tribal "traditionalists" and "moderns" as representing aspects of a linear development, one looking back, the other forward. It cannot see them as contending or alternating futures. [Clifford 1988:338; emphasis added]
Clifford adds that "The idea of culture carries with it an expectation of roots, of a stable, territorialized existence" and that "the culture idea, tied as it is to assumptions about natural growth and life, does not tolerate radical breaks in historical continuity" (1988:338). Or, as Landsman and Ciborski (1992:428) noted, "This notion of what constitutes authentic culture [elements that "can be traced to the past and confirmed in documentary sources"] implies that contemporary cultural expressions are genuine or spurious depending on their degree of correspondence to the documented original culture patterns." To this, one can add that exogenous cultural material is maximally inauthentic when it emanates from the West; cultural transmission among the locals themselves was a primary focus of Boasian historicism.
Clifford provides a summation of how culture, as articulated by anthropologists testifying as expert witnesses, fared under legal examination.
530  CULTURAL ANTHROPOLOGY
This cornerstone of the anthropological discipline proved to be vulnerable under cross-examination. Culture appeared to have no essential features. Neither language, religion, land, economics, nor any other key institution or custom was its sine qua non. It seemed to be a contingent mix of elements. At times the concept was purely differential: cultural integrity involved recognized boundaries; it required merely an acceptance by the group and its neighbors of a meaningful difference, a we-they distinction. But what if the difference were accepted at certain times and denied at others? And what if every element in the cultural melange were combined with or borrowed from external sources? At times the experts seemed to suggest that culture was always acculturating. But then how much historical mix and match would be permissible before a certain organic unity were lost? Was the criterion a quantitative one? Or was there a reliable qualitative method forjudging a culture's identity? [1988:323]
Notwithstanding Clifford's (1988:337) claim that "by 1978 the modern [anthropological] notion of culture was part of the trial's common sense," it can be suggested, after reflecting on these passages, that the difficulty concerned rather the tension between the primordialist connotations of an exoteric culture concept as legally construed and the nonprimordialist (and, to be sure, proportionately ambiguous) characterizations of culture ("a contingent mix of elements," "culture was always acculturating") expressed by anthropologists brought in as expert witnesses. As paraphrased by Clifford, certain of these witnesses represented culture as referencing a mutable, syncretic, and dispersible field of ideation and practice, lacking any essential core or content and detachable from territory, language, and heredity. Such testimony, clearly refractory to folk-American notions of genuine Indian culture, parallels Clifford's own reflections on groups, communities, and tribes:
Groups negotiating their identity in contexts of domination and exchange persist, patch themselves together in ways different from a living organism. A community, unlike a body, can lose a central "organ" and not die. All the critical elements of identity are in specific conditions replaceable: language, land, blood, leadership, religion. Recognized, viable tribes exist in which any one or even most of these elements are missing, replaced, or largely transformed. [1988:338]
And so, also, for cultures. Clifford, however, in these contexts, seemingly ascribes to the culture concept the same properties of organic unity, coherence, and continuity that he claims the culture concept itself attributes to the fields of social experience it represents. It is as though these defects are both essential to the concept and exhaustively characteristic of all its uses. To be sure, Mead's privileging of the "authentically indigenous" over the "inauthentically borrowed" is refractory to current disciplinary concerns with margins, borderlands, and intersystems. But objections to such notions of discreteness and authenticity are hardly unprecedented.
As there is no Simon-pure race nowadays, so there is no Simon-pure culture. Quite apart from the spread of Caucasian civilization, Congolese Pigmies have been influenced by Bantu neighbors, one Australian group makes distant trips to others, Papuan sailors carry their earthenware hundreds of miles from the place of
FORGET CULTURE 531
manufacture.... In short, White influence, however devastating in its ultimate effect, is not a thing sui generis; aboriginal peoples have borrowed from one another for thousands of years, and the attempt to isolate one culture that shall be wholly indigenous in origin is decidedly simple-minded. [Lowie 1935:xvii-xviii]
Writing in 1935, Lowie here expounded an inherently syncretic rather than primordialist culture construct, at the same time characterizing Euro-American influences on aboriginal people as exceptionally volatile instances of a ubiquitous syncretic process. Given such longstanding differences in its characterization, it would appear preferable, as Clifford also suggests, to view culture as a more labile construct which changes to engage new perspectives and issues: "There is no need to discard theoretically all conceptions of 'cultural' difference, especially once this is seen as not simply received from tradition, language, or environment but also as made in new political-cultural conditions of global relationality" (Clifford 1988:274).
Culture and the Politics of Difference
Lila Abu-Lughod's article "Writing against Culture" (1991) constitutes a synoptical compendium of the defects of culture currently in play:
Culture as ideation: "[A discourse-centered approach] refuses the distinction between ideas and practices or text and world that the culture concept too readily encourages" (1991:147).
Culture as legalism: "[Practice theory] favors strategies, interests, and improvisations over the more static and homogenizing cultural tropes of rules, models, and texts" (1991:147)
Culture as localism: "All these projects [examining trans regional, transnational, and global connections] expose the inadequacies of the concept of culture and the elusiveness of the entities designated by the term cultures" (1991:149).
Culture as localism and ahistoricism: "Denied the same capacity for movement, travel and geographic interaction that Westerners take for granted, the cultures studied by anthropologists have tended to be denied history as well" (1991:146).
Culture as holism, coherence, and discreteness: "Organic metaphors of wholeness and the methodology of holism that characterizes anthropology both favor coherence, which in turn contributes to the perception of communities as bounded and discrete" (1991:146).
Culture as coherence, timelessness, and discreteness: "If 'culture,' shadowed by coherence, timelessness, and discreteness, is the prime anthropological tool for making 'other,'... then perhaps anthropologists should consider strategies for writing against culture" (1991:147).
Culture as homogeneity, coherence, and timelessness: "By focusing closely on particular individuals and their changing relationships, one would necessarily subvert the most problematic connotations of culture: homogeneity, coherence, and timelessness" (1991:154).
532   CULTURAL ANTHROPOLOGY
Culture as hierarchy: "I will argue that 'culture' operates in anthropological discourse to enforce separations that inevitably carry a sense of hierarchy" (1991:137-138).
For Abu-Lughod, anthropology is multiply implicated in hierarchy, both as the intellectual project of the hegemonic West and as an arrangement in which Western selves control representations of non-Western others. While acknowledging the egalitarian nuances of relativism, she asserts that, given the hierarchical positionality of anthropology and its objects, writing about difference reinforces hierarchy. The perspective is clearly Saidian, and Abu-Lughod, after citing Said on Orientalism's implication in political and economic domination, asks, "Should anthropologists treat with similar suspicion 'culture' and 'cultures' as the key terms in a discourse in which otherness and difference have come to have, as Said points out, 'talismanic qualities'?" (1991:147).
Clearly the culture concept has been influenced by and influential upon regional and international projects of politico-economic domination, just as its differentiating functions may reiterate those of hereditarian racialism. Abu-Lughod's conclusions seem particularly persuasive when considering, for example, Kluckhohn and Kelly's (1945) harrowing reflections on the superiority of culturalism over racialism. For postwar America, culture
does carry an overtone of legitimate hope to troubled men. If the Germans and the Japanese are as they have been mainly because of their genes, the outlook is an almost hopeless one, but if their propensities for cruelty and aggrandizement are primarily the result of situational factors ("economic" pressures and so on) and their cultures, then something can be done about it. [1945:99]
Whether any concept predicating differences in fields of social experience necessarily inscribes itself in hierarchy is a complex question. Baudrillard's (1993:124-139) reflections on modalities of difference and otherness suggest interesting approaches to the problem. I focus here more narrowly on Abu-Lughod's characterization of culture as a flawed concept which incorrectly characterizes its object, on the concept's imputed ahistoricity, coherence, discreteness, and the rest.
Unlike Clifford, Abu-Lughod explicitly advocates replacing culture: the proposed substitutes are Bourdieu's "practice" and theories of "discourse" deriving from Foucault and from sociolinguistics. Abu-Lughod enumerates what she takes to be the advantages of practice and discourse over culture: practice stresses strategies over rules, Foucaultian discourse "refuse[s] the distinction between ideas and practices," linguistic discourse perspectives capture "multiple, shifting, competing statements," and all three work against boundedness and idealism (1991:147-148). In the exigencies of the encounter between theory and field data, Abu-Lughod argues that the substitutes exhibit superior interpretive and analytical power. A footnote, for example, provides concrete exemplification of the superiority of discourse as an antidote to the "timelessness, coherence, and homogeneity" entailed in the culture construct:
FORGET CULTURE 533
In my own work on an Egyptian Bedouin community I began to think in terms of discourses rather than culture simply because I had to find ways to make sense of the fact that there seemed to be two contradictory discourses on interpersonal relations—the discourse of honor and modesty and the poetic discourse on vulnerability and detachment—which informed and were used by the same individuals in differing contexts [Abu-Lughod 1991:162]
She notes further that two distinct discourses on death characterized men and women and legitimized power differences between them. The import is clear. "Discourse" permits us to theorize and represent contradiction and heterogeneity, whereas "culture," by definition, cannot. In another context, remarking on the apparent contradiction between lewd humor and Islamic religiosity in an old woman's speech, Abu-Lughod writes, "How does this sense of humor, this appreciation of the bawdy, go with devotion to prayer and protocols of honor? .. . What can 'culture' mean, given this old woman's complex responses?" (1991:155).
Obviously nothing, if the culture construct is characterized in such a way that it cannot refer to (nominally) contradictory or heterogeneous discourses. An equally obvious response is that culture can continue to mean what it has meant in the diversity of its past usages because it is not in these but only in Abu-Lughod's representation of them that contradiction and heterogeneity are excluded from consideration. In question is whether "discourse" is better able than "culture" to elucidate, for example, Malinowski's assertion in 1926 that "human cultural reality is not a consistent logical scheme, but rather a seething mixture of conflicting principles" (1926:121) or Sapir's conclusion in 1938 that anthropology is concerned
not with a society nor with a specimen of primitive man nor with a cross-section of the history of primitive culture, but with a finite, though indefinite, number of human beings, who gave themselves the privilege of differing from each other not only in matters generally considered as "one's own business" but even on questions which clearly transcended the private individual's concern and were, by the anthropologist's definition, implied in the conception of a definitely delimited society with a definitely discoverable culture. [Sapir 1949:569-570]
Distinct concepts are usefully signaled by distinct labels. Conversely, distinct labels engender consciousness of conceptual distinctiveness. The relation of Foucault's "discursive practices" or Bourdieu's "practice theory" to "culture" or "culture theory" is genealogically complex. Either resemblance and difference can be foregrounded, depending especially on whose characterizations of culture are given cognizance. Plausible arguments could be advanced that these concepts address certain facets of social experience neglected or ignored by the culture construct. The difficulty is that the facets upon which Abu-Lughod focuses—interested strategic conduct, diverse and contradictory propositions, and the like—figure both in practice-discourse theories and in earlier anthropological discussions of culture. The claim that the meanings of practice or of discourse engage specific facets of social experience that the meanings
534  CULTURAL ANTHROPOLOGY
of culture definitionally exclude is questionable. In this respect, the phonological shapes /praektis/ and /diskowrs/ are no better or worse than /kAlcar/ as ways of talking about contradiction and agency in Bedouin social experience. Again, it could be argued that these concepts go "beyond" what is possible with the culture concept—delineating novel or superior perspectives on social diversity or contradiction, for example—but the argument does not take this direction either. Rather, in Abu-Lughod's discussion, nomenclatural practices become detachable from the question of conceptual likeness or unlikeness. Specifically, through an implicit ideology of lexical hygiene, discourse and practice seem to derive their imputed conceptual superiority from their graphic or phonological difference from culture. What results is reference to familiar signifieds—contradictory principles, heterogeneous actions—with new and more impressive signifiers, an eventuality Abu-Lughod herself prophetically envisions when she writes that "there is always the danger that these terms will come to be used simply as synonyms for culture" (1991:147).
Another facet of Abu-Lughod's critical treatment of culture concerns generalization. "Anthropologists," she writes, "commonly generalize about communities by saying that they are characterized by certain institutions, rules, or ways of doing things" (1991:153). In her argument, generalization both creates exaggerated notions of cultural difference (by homogenizing the others) and impoverishes ethnography's object by eliding the experiences of particular individuals in relation to particular events. Dissatisfaction with the position that culture or social structure concerns only typifications abstracted from the plurality of unique instances is, of course, hardly novel. Consider Radin's objections to what he conceived to be the ahistoricism of Boasian anthropology.
We are dealing with specific, not generalized, men and women, and with specific, not generalized, events. But the recognition of specific men and women should bring with it the realization that there are all types of individuals and that it is not, for instance, a Crow Indian who has made such and such a statement, uttered such and such a prayer, but a particular Crow Indian. It is this particularity that is the essence of all history and it is precisely this that ethnology has hitherto balked at doing. [Radin 1933:184-185]
More recently, concern with individual persons and events emerged as a focus of "performance" and "experience" orientations in the 1980s (see Turner and Bruner 1986). The value of this perspective as an antidote to an exaggerated preoccupation with scripts, rules, and decontextualized structures is eloquently attested in Abu-Lughod's (1986) writing on the Awlad Ali Bedouins.
It is less clear, however, that an anthropology attentive to "lived experience" can dispense with generalization. At what level of generality does generalization become objectionable? What of the locals' own generalizations, both implicit and articulated, in whose terms they necessarily interpret and act upon their experiences? Are we to focus only on the particulars and neglect the regularities? How then can the two be differentiated? Generalization is unavoidable even when it engages heterogeneity, as with Abu-Lughod's observation that
FORGET CULTURE 535
"the two primary discourses [on death]—ritual funerary laments and the Islamic discourse on God's will—were attached to different social groups, men and women" (1990:162). Abu-Lughod again observes that "the pattern of the [Bedouin wedding night] defloration, as I have written elsewhere, is standard" but adds that "every defloration involves a specific set of people and takes place in a particular way" (1990:157). This example places special emphasis on the fact that a cultural reading cannot predict the outcome of defloration, whether or not there will be blood to index the bride's virginity. "Events," Abu-Lughod informs us, "take different courses. That is the nature of 'life as lived,' everywhere. Generalizations, by producing effects of timelessness and coherence to support the essentialized notion of 'cultures' different from ours and peoples separate from us, make us forget this" (1990:158). Leaving aside the issue of whether the culture concept makes people forget that events have multiple outcomes, it is questionable whether disciplinary attention to the particular can proceed independently of generalizations about cultural form. Abu-Lughod's examples affirm that the Bedouins are indeed "characterized by certain institutions, rules, or ways of doing things," not least with respect to the protocol of defloration displays which retain a recognizable character from one instance to the next, even as the contextual variables shift to produce diverse outcomes.
Culture, Legalist and Objectivist
Of contemporary critics of culture, Pierre Bourdieu is arguably the principal architect of recent nomenclatural instability. Bourdieu, in a sense, does away with culture by failing to talk about it and his terminological practices deserve particular attention. The concept of the habitus, first encountered by many readers in Bourdieu's Outline of a Theory of Practice (1977 [ 1972]), has been exceptionally influential in anthropological theory in the 1980s and 1990s. The term is introduced in the phrase "the agents' habitus" (1977:9), and Bourdieu thereafter allocates considerable space to expounding the concept (1977:72-95; see also 1984:466-484, 1990a:52-65, 1990b: 12-13, 107-109). For habitus, the Oxford Latin Dictionary (Glare 1982) lists such glosses as "state of being, condition, expression, demeanour, manner, bearing, physical attitude, posture." Bourdieu defines the concept, in the context of his theoretical project, as "a system of acquired dispositions, functioning on the practical level as categories of perception and assessment or as classificatory principles as well as being the organizing principles of action" (1990b: 13).
In Outline, the only information given as to the lexical choice is located in a footnote. There Bourdieu observes, "One of the reasons for the use of the term habitus is the wish to set aside the common conception of habit as a mechanical assembly or performed programme." (1977:218). In his subsequent reflection on the derivation of the concept, Bourdieu (1985) makes no reference to culture, even as he positions his use of "habitus" within a continuous theoretical tradition. Alluding briefly to Mauss (1973[1935]), who used the term to refer to socially learned bodily dispositions, Bourdieu notes that the most proximate influence on his lexical choice was the art historian Erwin Panofsky.
536  CULTURAL ANTHROPOLOGY
Still, I believe that the choice of this old word which, despite some occasional uses, has for so long been relegated to oblivion is not alien to the subsequent accomplishments of the concept. Those who, intending to reduce or destroy, wish to take the word back to its point of origin will surely discover if they carry on their research with intelligence, that its theoretical advantage resided precisely in the direction of research that it designated and which is at the very source of the overstepping which it made possible. It seems to me in fact that in all cases those who used the word habitus were inspired by a theoretical intention not far removed from mine which was: to get out from under the philosophy of consciousness without doing away with the agent. [1985:14; emphasis added]
This appropriation of an existing word whose earlier uses prefigure or are consistent with a developing conceptual project is, Bourdieu adds, "diametrically opposed to the strategy which consists of an attempt to associate its (i.e., the project's) name with a neologism ... and thereby to make its rating rise in the Citation Index" (1985:14). While the geneaology is persuasively motivated, it is nevertheless the case that few sociologists or anthropologists encountered the old word as anything other than a neologism, a fact that invites reflection on the significance of the lexical innovation. As Ortner (1984:148) wrote, "Bourdieu's habitus ... behave[s] in many ways like the American concept of culture, combining elements of ethos, affect and value with more strictly cognitive schemes of classification." The point is not that Bourdieu should have used the term culture to refer to agents' ideational and dispositional repertoires; the term, of course, has different resonances in French sociology than in American anthropology. No one would question Bourdieu's right to explicate a concept and to call it whatever he likes. Rather I focus here on his nomenclatural practices and on his representation of culture as a construct with determinate meanings, meanings inherently different from those of the habitus.
Clearly, in the passage cited above, Bourdieu accords considerable agency to the linguistic sign "habitus," taken as a phonological or graphic shape conjoined with a meaning that has accrued from earlier scholarly uses a valuable conceptual orientation. Specifically, these earlier uses entail obligatory reference to agents, a denial of objectivism congenial to Bourdieu's project. In contrast, by implication, such imaginable alternatives as conscience collective or culture would be inappropriate because they have accrued from their past uses meanings antithetical to this project. That culture possesses such meanings for Bourdieu is clear from his passing remarks about the concept. First, he understands culture as referring to an abstraction to which deluded anthropologists ascribe agency:
In short, failing to construct practice other than negatively, objectivism is condemned either to ignore the whole question of the principle underlying the production of the regularities which it then contents itself with recording; or to reify abstractions, by the fallacy of treating the objects constructed by science, whether "culture," "structures," or "modes of production," as realities endowed with a social efficacy, capable of acting on agents responsible for historical actions or as a power capable of constraining practices. [1977:26-27]
FORGET CULTURE 537
The second difficulty with culture is its misrepresentation of conduct as the behavioral enactment of rules and its exclusion of strategy and improvisation. After rejecting Saussure's conception of parole as the implementation of langue, Bourdieu writes,
It would not be difficult to show that the construction of the concept of culture (in the cultural anthropology sense) or "social structure" (in Radcliffe-Brown's sense or that of British social anthropology) similarly implies the construction of a notion of conduct as execution which coexists with the primary notion of conduct as simple behavior taken at face value. The extreme confusion of debates on the relationship between "culture" (or "social structures") and conduct generally arises from the fact that the constructed meaning of conduct and the theory of practice it implies lead a sort of underground existence in the discourse of both the defenders and the opponents of cultural anthropology. [1977:24-25]
Thus is culture implicated, like structuralism, in portraying an "agent reduced to the role of bearer—Trager—of the structure" (Bourdieu 1985:13).
While culture and habitus both refer to people's socially acquired ideational and dispositional systems, these passages suffice to indicate that, for Bourdieu, culture is precisely what habitus is not. Culture entails legalism and objectivism, meanings incompatible with Bourdieu's practice orientation. Habitus, insofar as it excludes these, is represented as a qualitatively distinct concept. The lexical ideology prefigures and parallels that of Abu-Lugod for whom culture cannot, by definition, refer to contradictory or heterogeneous discourses. But here, the effect of the terminological innovation is all the more striking in the absence of explicit discursive comparison of habitus and culture. Bourdieu doesn't call for the replacement of the culture construct: he simply replaces it.
Few would dispute (at least with respect to the social sciences) Bourdieu's characterization of "a fair attitude toward theoretical tradition" as "one which consists of affirming as inseparable both continuity and rupture, conservation and going beyond it" (1985:15). But if continuity and rupture are truly inseparable, they fail to appear so in Bourdieu's characterization of culture and habitus. If Kroeber, for example, is taken as the authoritative source on the concept, culture becomes strategy-proof, and the differences from habitus are clear for all to see. But others have deployed culture in ways more consistent with a practice orientation. Sapir, for example, professed an explicitly constructivist concept of culture, one that anticipated, in many crucial respects, contemporary concerns with social construction, reproduction, and microprocess. Remarked Sapir, in 1931:
While we often speak of society as though it were a static structure defined by tradition, it is, in the more intimate sense, nothing of the kind, but a highly intricate network of partial or complete understandings between the members of organizational units of every degree of size and complexity.... It is only apparently a static sum of social institutions; actually it is being reanimated or creatively reaffirmed from day to day by particular acts of a communicative nature which obtain among individuals participating in it. [Sapir 1949:104]
538  CULTURAL ANTHROPOLOGY
An abiding theme of much of Sapir's writing, especially into the 1930s (see Sapir 1994), is the methodological and theoretical necessity of rejecting the objectivism that Bourdieu construes as essential to the culture construct. Indeed, it could be argued that Sapir's writing negotiates the higher ground between objectivism and methodological individualism more successfully, in certain respects, than does Bourdieu's practice theory. Sapir's culture—historical, sui generis, and subject to the transformative effects of individual improvisation on social consensus—accords considerable agency to its participants (1949:571; see also Goldenweiser 1917; Sapir 1917). In this, it differs from the habitus, which "makes possible the free production of all the thoughts, perceptions, and actions inherent in the particular conditions of its production—and only those" (Bourdieu 1990a:55).
Bourdieu's assertion that the culture construct inevitably construes conduct as the behavioral execution of rule is likewise subject to question. As is well known, Malinowski was overtly critical of legalism, repeatedly addressing his ethnographic virtuosity to its demolition. And if his exposition draws too heavily on aHobbesian antipathy between the social rule and the refractory egoism of "human nature," it is, nonetheless, the case that his Trobrianders are im-provisers, manipulators, and strategists. "Take the real savage," wrote Malinowski, "keen on evading his duties, swaggering and boastful when he has fulfilled them, and compare him with the anthropologist's dummy who slavishly follows custom and automatically obeys every regulation" (1926:30). For Malinowski, legalism exists only rhetorically in the locals' exegesis. Their disposition and conduct are products of an unstable dialectic of strategy, compliance, and improvisation: "[The savage's] sentiments, his propensities, his bias, his self-indulgences as well as tolerance of others' lapses, he reserves for his behavior in real life ... the natural, impulsive code of conduct, the evasions, the compromises and non-legal usages are revealed [through observation] only to the field-worker" (1926:120). Neither was it the case that Malinowski's skepticism toward legalism was limited to the observation that people are ambivalent toward rules and sometimes violate them. For Malinowski, both officialized law and legalized usage emerge from strategic practices; practices in turn transpire amid contradictory laws.
We have to abandon now definitely the idea of an inert, solid "crust" or "cake" of custom rigidly pressing from outside upon the whole surface of tribal life. Law and order arise out of the very processes they govern. But they are not rigid nor due to any inertia or permanent mould. They obtain on the contrary as the result of a constant struggle not merely of human passions against the law, but of legal principles with one another. [1926:122-123]
We have here, then, a representation of Trobriand "legal principles" situated in the discourses of interested agents. The point is not that Malinowski (or anyone else to date) explicated a satisfactory theory of the articulation of system with agency but simply that he proposed something rather more complex than Bourdieu's "conduct as execution." If culture is internally contradictory (the
FORGET CULTURE 539
matrilineal clan versus the family, official conduct versus individual interest), no objectivist determination of conduct is possible.
Conclusion
Of Eric Wolf's (1980) observation that "the old culture concept is moribund," Ward Goodenough (1989:93) remarked easily that "the same thing could have been said thirty years ago. The term culture has a long history of meaning different things to different people." Unstable in meaning and reference both synchronically and over time, the culture construct has exhibited exceptional lability. Culture's versatility in sense and reference—its capacity to mean something simultaneously to Edward Sapir and Alfred Kroeber or to Ward Good-enough and Clifford Geertz—has variously been cited as grounds for dispensing with it or represented as its great strength. To the question of culture's contemporary relevance, Freilich affirms that the concept is undergoing "progressive simplification and clarification" and that "over time, much superfluous information is weeded out, so that the concept's essence is able to shine through" (1989:2). Regardless of whether one shares his optimism, the question of how the essence is to be specified is clearly relevant to assertions that culture has been, is being, or should be replaced.
By asserting that the old culture concept is dying, Wolf implies that a new culture concept—or some entirely distinct concept—is currently in play. Assertions of this kind pose, of course, complex questions of conceptual periodization and transition. If culture entails legalism, coherence, homogeneity, discreteness, and timelessness, during which (overlapping? discrete?) phases of disciplinary history has this been the case? If the anthropologically derived culture construct has been, is being, or will be displaced, how will we know that the event has occurred? What criteria exist that allow us to assert of a construct like culture that, at some point in time, it has been reconfigured or replaced? How would we distinguish reconfiguration of the existing construct(s) from its replacement by something altogether more novel? If the culture concept is analyz-able into a subset of constitutive elements, what are these elements? Which and how many of them must change before a new culture concept supplants the old, or a different concept entirely emerges? How do current reformulations of culture compare in scope and magnitude with changes that the concept has undergone in the past? Much recent criticism seems to concur with Luhrmann's (1993:1058) assertion that culture is presently "more unsettled than it has been for forty years." This is a defensible conclusion but hardly self-evident. One could argue, on the contrary, that culture is not measurably more (or less) unsettled than in the past, but that debate about it is increasingly phrased in Kuhnian idioms of transition rather than in idioms of reconceptualization. Such judgments are, of course, perspectival, and will certainly be easier to assess 50 years from now. As Stocking writes,
It is an artifact of historical periodization that the last period always ends in the present moment; but whether that moment, or any other recent moment, marks a
540  CULTURAL ANTHROPOLOGY
significant historical transition is another matter—about which my backward-looking historicist temperament makes me disinclined to speculate. [1992:345-346]
Seemingly implicit in Stocking's remarks is the admonition that consciousness of paradigmatic transition may be gravely exaggerated. Thus a possible reaction to the critical writing on culture, and a common reaction to varieties of postmodern writing more generally, is that anthropologists have always known what the critics are trying too earnestly to tell us, that the nominally new perspectives and debates are so many reinventions of the wheel. And, to be sure, certain of the recent criticisms of culture gamble rather poignantly for their topicality on an increasingly pervasive disciplinary amnesia, a lack of familiarity with what has gone before. Neither, however, can it be argued persuasively that no new criticisms are being articulated, that no directional (rather than cyclical or oscillatory) changes are underway. Concern with the distributions of power, for example, constitutes a substantially new perspective on what we call culture and cultures. Further, while many of the culture construct's difficulties have been addressed and debated in the past, they have never adequately been resolved. Older debates acquire new and emergent meanings in the changing disciplinary and historical contexts of the 1980s and 1990s, assuring, for example, that recent concerns with individuals, events, and experience are not and cannot be the same as those expressed by Radin 60 years before.
Rather than attending both to continuity and transition in the history of anthropological ideas, much of the recent cultural criticism rests upon an essential-ist understanding of what culture as an analytical concept means, such essential-ism then guaranteeing the argument for conceptual transition. For Bourdieu, Clifford, and Abu-Lughod, every occurrence of the lexeme "culture" seems indifferently to evoke the specters of Benedict's configurationalism, Kroeber's superorganicism, and Malinowski's ahistoricism. But consider what is not evoked by culture: the constructivism of Sapir, the historicism of Radin, the an-tilegalism of Malinowski. And this is only to cite the ancestors. Does anyone really want to say in the mid-1990s that deployment of the anthropological culture concept necessarily commits us to coherence, homogeneity, ahistoricity, and a legalistic theory of conduct? History, chaos, contestation, and strategy have been anthropological growth stocks since at least 1980, and disciplinary writing reflects this state of affairs, both in theoretical exposition and in the interpretation of ethnographic materials. I suspect that it would be exceptionally difficult to find any theoretically oriented writing in the last ten years that represents fields of social meaning and practice—however labeled—as interactively pristine, configurationally coherent, and lacking history. To take four quite distinct examples, Sahlins (1985), Rappaport (1984), Wagner (1981), and Rosaldo (1980) do not represent culture or cultures in these terms.
Few would dispute that identifiable shifts in thematic emphasis—configurationalism to disorder, consensus to contestation, synchrony to diachrony, and so forth—have recently occurred in the profession. It is less clear that these shifts index a qualitative transformation of the culture construct or a replace-
FORGET CULTURE 541
ment of it. Many of the contemporary concerns—constructivism, historicity, skepticism about boundaries—figure in earlier theoretical projects, variously as subdominant themes or articulated policy positions. They may appear in surprising places, sometimes conjoined uneasily with their opposites, as when Mead (1932) wrote of The Changing Culture of an American Indian Tribe. Implicit in the appeal to practice, discourse, and other substitutes is the assertion that analytic constructs such as culture do not and cannot change as they engage new insights, emphases and topics. The new cultural criticism exhibits a perspective on such constructs—both synchronically and in long and short durees—as discrete, discontinuous, and rigidly periodized rather than continuous, dynamic, and overlapping. The culture construct in these representations is immutable, rather like the primordialist culture criticized by Clifford. It cannot survive modifications of its (imputed) content, and when one or more of its attributes alters, it dies.
The tempered reservations expressed here toward the imminent demise of culture are subject to an important qualification. Increasingly, critical judgments of culture exhibit the illocutionary force (if not the surface form) of such performative speech acts as promising, christening, or sentencing: "I [anthropologist of record] hereby pronounce you [culture] dead or dying in the 1990s" [sound of breaking glass]. The diminishing frequency of the lexical item "culture" in the journals is a tangible sign that the retrospective images of culture synthesized by its critics are eminently performative and imbued with historical efficacy. It is not only that new labels are creating an exaggerated consciousness of paradigmatic transition, that anthropologists who self-consciously reject "culture" in favor of "discourse," "hegemony," or "habitus" will traffic partly in old signifieds with new signifiers. At issue is the question of securing and exercising the authority to define and characterize culture. Neither in earlier disciplinary history nor as deployed in recent anthropological writing does the culture concept consistently exhibit the attributes of ahistoricism, totalization, holism, legalism, and coherence with which its critics selectively reconstitute it. These are invented images of culture, both arbitrary and partial with respect to a much more diverse and versatile field of definition and use. Such images, nonetheless, are rapidly acquiring more authoritative perlocutionary effects. As more and more anthropologists decide that the culture concept does entail these determinate and conceptually defective characteristics, its meaning(s) as a disciplinary construct, presently heterogeneous, will progressively converge with the meanings constructed in the critical literature. When this happens, the phonological shape will exit the active anthropological vocabulary and those features of culture's meaning consistent with current theoretical orientations will be relexi-fied.
Notes
Acknowledgments. I would like to thank David Dinwoodie, Raymond Fogelson, David French, Kathrine French, Herbert Lewis, and Aidan Southall for valuable suggestions and input; Maria Lepowsky and Robert Moore gave the manuscript especially
542   CULTURAL ANTHROPOLOGY
close and constructive readings. My thanks also to two anonymous referees. All the usual disclaimers apply.
1. Self-congratulation and hostility hardly exhaust the undesirable possibilities: the others' cultures have frequently enough gotten involved as well in Western projects of self-flagellation and self-improvement. A lamentable lacuna in postcolonial criticism is attention to the primitivistic representation of the "native" as exemplary.
2. Abu-Lughod's proposal for "writing against culture" invites comparison with Fabian's (1990) strategy of eschewing ethnographic writing altogether; it possesses, of course, greater congeniality with conventional measures of professional activity.
3. While there exists a shared vocabulary that has been used to characterize the defects of the culture construct, there is less consistency in how terms are used. For some critics, a single item is polysemous, referring to logically interrelated images or connotations of the culture construct. In many cases, the terms are only indirectly characterized.
4. The irony of this substitution was not lost on Bourdieu (1977:201) who contrasted Radcliffe-Brown's "naive realism" with culture's "realism of the intelligible."
5. In a similar vein, Rosaldo .prefaced his critique of legalism by allowing, "in certain respects, after all, cultural practices do conform to codes and norms. People make plans and sometimes their plans work out. Not all expectations remain unmet. Conventional wisdom does not always fail" (1989:102).
6. Two Crows, an Omaha consultant of James Dorsey (1882) in the 1870s, provoked earlier criticisms of cultural homogeneity (Sapir 1949:569-570) by systematically disagreeing with ethnographic characterizations provided by other Omahas.
7. The obvious disciplinary precedent is the culture area concept whose boundaries were generally acknowledged to be heuristic and arbitrary, different in purpose and scale from what the culture construct conventionally refers to.
8. Drummond (1980) has also questioned the integrity of discrete cultures, but in the somewhat different terms raised by Bickerton's (1975) critique of the integrity of discrete languages. For Bickerton the concept of a language is problematic both because of the continuousness of elements across the nominal boundaries (as in Creoles) and because any set of speech habits "has the potential for merging into any other in a principled way" (1975:180). Given continuousness, there are no languages but only language, the faculte de langage of which particular natural languages are "arbitrary interpretations." Says Drummond, "To be consistent with the creole metaphor, one would have to assert that there are no cultures, only Culture. Any cultural system would contain analytical specifications—or invariant properties in the structural metaphor— already contained in some other, supposedly distinct, cultural system" (1980:372).
9. Avoiding the historical amnesia characteristic of much critical writing on the culture construct, Appadurai (1988:38) also registers earlier anthropological counterpoints to localism, citing specifically Boas, Mauss, Benveniste, Dumezil, Lord Raglan, and Hocart.
References
Abu-Lughod, Lila
1986 Veiled Sentiments: Honor and Poetry in a Bedouin Society. Berkeley: University of California Press.
1991 Writing against Culture. In Recapturing Anthropology: Working in the Present. R. Fox, ed. Pp. 137-162. Santa Fe: School of American Research Press.
FORGET CULTURE 543
Appadurai, Arjun
1988  Putting Hierarchy in Its Place. Cultural Anthropology 3(1):36-47.
1990 Disjuncture and Difference in the Global Cultural Economy. Public Culture 2(2): 1-24.
Ardener, Shirley
1975  Introduction. In Perceiving Women. Shirley Ardener, ed. London: Malaby. Barth, Fredrik
1966  Models of Social Organization. Royal Anthropological Institute Occasional Paper 23. London: Royal Anthropological Institute. Baudrillard, Jean
1993 The Transparency of Evil. London: Verso. Bickerton, Derek
1975   Dynamics of a Creole System. Cambridge: Cambridge University Press. Bourdieu, Pierre
1977  Outline of a Theory of Practice. Cambridge: Cambridge University Press.
1984 Distinction: A Social Critique of the Judgment of Taste. Cambridge, MA: Harvard University Press.
1985 The Genesis of the Concepts of "Habitus" and "Field." Sociocriticism 2(2): 11-24.
1990a  The Logic of Practice. Stanford: Stanford University Press. 1990b  In Other Words: Essays Towards a Reflexive Sociology. Stanford: Stanford University Press. Button, Graham
1991 Ethnomethodology and the Human Sciences. Cambridge: Cambridge University Press.
Clifford, James
1986 Introduction: Partial Tuths. In Writing Culture: The Poetics and Politics of Ethnography. James Clifford and George Marcus, eds. Berkeley: University of California Press.
1988  The Predicament of Culture. Cambridge, MA: Harvard University Press.
1992 Traveling Cultures. In Cultural Studies. Lawrence Grossberg, Cary Nelson, and Paula Treichler, eds. Pp. 96-116. New York: Routledge.
Dirks, Nicholas
1992  Colonialism and Culture. Ann Arbor: University of Michigan Press. Dirks, Nicholas, Geoff Eley, and Sherry Ortner
1994 Introduction. In Culture/Power/History: A Reader in Contemporary Social Theory. Nicholas Dirks, Geoff Eley, and Sherry Ortner, eds. Pp. 3-45. Princeton: Princeton University Press.
Dorsey, James
1882  Omaha Sociology. Third Annual Report of the Bureau of American Ethnology. Pp. 205-371. Washington, DC: Government Printing Office. Drummond, Lee
1980  The Cultural Continuum: A Theory of Intersystems. Man (n.s.) 15(2):352-374. Fabian,Johannes
1990 Presence and Representation: The Other and Anthropological Writing. Critical Inquiry 16(3):753-772.
544  CULTURAL ANTHROPOLOGY
Freilich, Morris
1989  Introduction: Is Culture Still Relevant? In The Relevance of Culture. Morris Freilich, ed. Pp. 2-19. New York: Bergin and Garvey. Friedrich, Paul
1992  Interpretation and Vision: A Critique of Cryptopositivism. Cultural Anthropology 7(2): 211-231. Fortes, Meyer
1949  The Web of Kinship among the Tallensi. Oxford: Oxford University Press. Glare, P. G. W., ed.
1982  Oxford Latin Dictionary. Oxford: Clarendon Press. Goldenweiser, Alexander
1917  The Autonomy of the Social. American Anthropologist 19(3):447-449. Goodenough, Ward
1989  Culture: Concept and Phenomenon. In The Relevance of Culture. Morris Freilich, ed. Pp. 93-97. New York: Bergin and Garvey. Gramsci, Antonio
1971 Selections from the Prison Notebooks of Antonio Gramsci. Quentin Hoare and Geoffrey Nowell Smith, eds. New York: International Publishers.
Hannerz, Ulf
1989 Notes on the Global Ecumene. Public Culture l(2):66-75. Hastrup, Kirsten
1990 The Ethnographic Present: A Reinvention. Cultural Anthropology 5(1):45-61.
Herbert, Christopher
1991 Culture and Anomie: Ethnographic Imagination in the Nineteenth Century. Chicago: University of Chicago Press.
Hill, Archibald
1980  How Many Revolutions Can a Linguist Live Through? In First Person Singular. Boyd Davis and Raymond O'Cain, eds. Pp. 69-78. Amsterdam: John Benjamin. Hymes, Dell, ed.
1974  Reinventing Anthropology. New York: Vintage Books. Kahn, Joel
1989  Culture, Demise or Resurrection? Critique of Anthropology 9(2):5-25. Kaplan, David, and Robert Manners
1972 Cultural Theory. Englewood Cliff: Prentice Hall. Kluckhohn, Clyde, and William Kelly
1945  The Concept of Culture. In The Science of Man in the World Crisis. Ralph Linton, ed. Pp. 78-105. New York: Columbia University Press. Kroeber, Alfred, and Talcott Parsons
1958  The Concepts of Culture and of Social System. American Sociological Review 23(3):582-583. Landsman, Gail, and Sara Ciborski
1992 Representation and Politics: Contesting Histories of the Iroquois. Cultural Anthropology 7(4):425-448.
Leach, Edmund
1954  Political Systems of Highland Burma. Boston: Beacon Press. Lowie, Robert
1935  The Crow Indians. New York: Irvington.
1937  The History of Ethnological Theory. New York: Holt.
FORGET CULTURE 545
Luhrmann, T. M.
1993  Review of Hermes' Dilemma and Hamlet's Desire: On the Epistemology of Interpretation. American Anthropologist 95(4): 1058-1059. Lyotard, Jean Francois
1984  The Postmodern Condition: A Report on Knowledge. Minneapolis: University of Minnesota Press. Malinowski, Bronislaw
1926  Crime and Custom in Savage Society. New York: Harcourt Brace and Co. Marcus, George
1986  Contemporary Problems of Ethnography in the Modern World System. In Writing Culture: The Poetics and Politics of Ethnography. James Clifford and George Marcus, eds. Pp. 165-193. Berkeley: University of California Press. Marcus, George, and Michael Fischer
1986  Anthropology as Cultural Critique. Chicago: University of Chicago Press. Mauss, Marcel
1973[1935]  Techniques of the Body, Economy and Society. New York: Oxford University Press. Mead, Margaret
1932 The Changing Culture of an Indian Tribe. Columbia University Contributions to Anthropology 15:1-313.
Moore, John T.
1974 The Culture Concept as Ideology. American Ethnologist 1:537-549. Murdock, George Peter
1972  Anthropology's Mythology. Proceedings of the Royal Anthropological Institute of Great Britain and Ireland for 1971. Pp. 17-24. London: Royal Anthropological Institute. Murphy, Robert
1971   The Dialectics of Social Life. New York: Columbia University Press. Ortner, Sherry
1984 Theory in Anthropology since the Sixties. Comparative Studies in Society and
History 26:126-166. 1990  Gender Hegemonies. Cultural Critique 14 (winter):35-80. Rabinow, Paul
1986 Representations Are Social Facts: Modernity and Post-Modernity in Anthropology. In Writing Culture: The Poetics and Politics of Ethnography. James Clifford and George Marcus, eds. Pp. 234-261. Berkeley: University of California Press.
1988  Beyond Ethnography: Anthropology as Nominalism. Cultural Anthropology 3(4):355-364. Radcliffe-Brown, Arthur Reginald
1940 On Social Structure. Journal of the Royal Anthropological Institute 70:1-12. Radin, Paul
1933 The Method and Theory of Ethnology. New York: McGraw Hill Rappaport, Roy
1984  Pigs for the Ancestors. New Haven: Yale University Press. Romney, A. K., Susan Weller, and William Batchelder
1986 Culture as Consensus: A Theory of Culture and Informant Accuracy. American Anthropologist 88(2):313-338.
546   CULTURAL ANTHROPOLOGY
Rosaldo, Renato
1980 Ilongot Headhunting, 1883-1974. Stanford: Stanford University Press. 1989  Culture and Truth: The Remaking of Social Analysis. Boston: Beacon Press.
Sahlins, Marshall
1985 Islands of History. Chicago: University of Chicago Press. Said, Edward
1978  Orientalism. New York: Pantheon Books. Sapir, Edward
1917  Do We Need a Superorganic? American Anthropologist 19(3):441-444. 1949  Selected Writings of Edward Sapir. D. Mandelbaum, ed. Berkeley: University of California Press.
1994  The Psychology of Culture: A Course of Lectures. J. Irvine, ed. New York: Mouton de Gruyter. Southall, Aidan
1970  The Illusion of Tribe. Journal of Asian and African Studies 5( l-2):28-50. Stocking, George
1982 Race, Culture and Evolution: Essays in the History of Anthropology. Chicago: University of Chicago Press.
1992  The Ethnographer's Magic. Madison: University of Wisconsin Press. Thornton, Robert
1988 The Rhetoric of Ethnographic Holism. Cultural Anthropology 3(3):285-303. Turner, Victor, and Ed Bruner, eds.
1986 The Anthropology of Experience. Urbana: University of Illinois Press. Tyler, Stephen
1986  Post-Modern Ethnography: From Document of the Occult to Occult Document. In Writing Culture: The Poetics and Politics of Ethnography. James Clifford and George Marcus, eds. Berkeley: University of California Press. Wagner, Roy
1981 The Invention of Culture. Chicago: University of Chicago Press. Watson, Graham
1991   Rewriting Culture. In Recapturing Anthropology: Working in the Present. Richard Fox, ed. Pp. 73-92. Santa Fe: School of American Research Press. Williams, Raymond
1977  Marxism and Literature. New York: Oxford University Press.
1983 Culture and Society, 1780-1950. New York: Columbia University Press. Wolf, Eric
1980  They Divide and Subdivide. New York Times. November 30.
1982 Europe and the People without History. Berkeley: University of California Press.
Yengoyan, Aram
1986  Theory in Anthropology: On the Demise of the Concept of Culture. Comparative Studies in Society and History 24(2):368-374.