Theories of Culture Revisited*

Roger M. Keesing

Note from your teachers: This is a wonderful essay in many respects, but we are asking you to read it mainly because of what it has to say about the essentialist vs.constructionist debate about the nature of culture. This debate overlaps with others such as the modernity vs. post-modernist debate, which Keesing also discusses. For our purposes, the key parts of this reading are pp. 301-303 and 308-310 (the page numbers are in the text, in square brackets; so are the footnotes).


If radical alterity did not exist, it would be anthropology's project to invent it.[1] Radical alterity — a culturally constructed Other radically different from Us — fills a need in European social thought: what Trouillot (1991) calls "the savage slot." I believe that in important senses, the radical otherness anthropologists have sought has not existed, at least for very many millennia. The tribal world in which we have situated that alterity — the world of Levi-Strauss's "cold societies" of unchanging tradition — was our anthropological invention. We continue to invoke it; and some of us journey ever deeper into darkest New Guinea to find it, existing still.

The invention and evocation of this radical otherness which has been anthropology's project required a conceptual universe, a mode of discourse. Especially as the idea of "a culture" was developed in the Boasian tradition as a bounded universe of shared ideas and customs and as the idea of "a society" was developed in functionalist social anthropology as a bounded universe of self-reproducing structures, these concepts provided a framework for our creation and evocation of radical diversity. "A culture" had a history, but it was the kind of history coral reefs have: the cumulated accretion of minute deposits, essentially unknowable, and irrelevant to the shapes they form.

The world of timeless, endlessly self reproducing structures, social and ideational, each representing a unique experiment in cultural possibility, has (we now know) been fashioned in terms of European philosophical quests and assumptions, superimposed on the peoples encountered and subjugated along colonial frontiers. The diversity and the uniqueness are, of course, partial "truths": the Tupinamba, the Aranda, the Baganda, the Vedda, the Dayak challenged comprehension and still do. But I believe we continue to overstate Difference, in search for the exotic and for the radical Otherness Western philosophy, and Western cravings for alternatives, demand.

I will touch again on this question of radical alterity, as it has been interpreted and created in anthropological discourse. My main concern here is to reexamine the concept of "culture," particularly our ways of talking and writing about "a culture." Hence I return to issues I addressed in a paper on "Theories of Culture" over fifteen years ago (Keesing 1974). I will begin by setting out a series of ironies and contradictions.

The Emphasis on Difference

A first irony is that the presently fashionable — in some quarters, at least, ascendant — symbolist/interpretive modes of anthropology require radical otherness more than ever, in a world in which such boundaries as there ever were are dissolving by the day. To show that conceptions of personhood, of emotions, of agency, of gender, of the body are culturally constructed demands that Difference be [302]demonstrated and celebrated, that. "cultures" be put in separate compartments and characterized in essentialist terms. Yet the formerly tribal or peasant peoples whose lives we engage are caught up in a world system through which ideas flow freely. I have just come back from a Solomon Islands where dreadlocks in the style of Bob Marley and Kung Fu videos are the stuff of contemporary "culture." More than ever, the boundedness and the essentialism that motivate it must depart from observed "realities"; the gulf between what we see in the field and the ways we represent it widens by the minute.

In its current postmodernist mutations, cultural anthropology can effectively engage the complexities of the present as collages, juxtapositions of the old and the new, the endogenous and the exogenous; and can transcend our old preoccupations with "authenticity" and closed boundaries. In their general pronouncements, postmodernist anthropologists have often highlighted such complexities. Thus, Marcus and Fischer tell us (1986:78) that:

most local cultures worldwide are products of a history of appropriations, resistances, and accommodations.The [present] task . . . is . . . to revise ethnographic description away from [a] self-contained, homogeneous, and largely ahistorical framing of the cultural unit toward a view of cultural situations as always in flux, in a perpetual historically sensitive state of resistance and accommodation to broader processes of influence that are as much inside as outside the local context.
Amen. Yet in practice, American postmodernist anthropologists, with their roots in the interpretive/cultural constructionist tradition, often rhetorically invoke radical alterity in the same ways. Marcus and Fischer talk about "the most intimate experiences of personhood . . . distinctive of particular cultures" (1986:62) and of "Moroccan masculinity" as "superficially similar to masculinity in other cultures" (1986:62). "What if persons in certain other cultures act from different conceptions of the individual?" (1986:45).

Poststructuralist thought, in its concern with texts as pervasive fictions (e.g. Derrida, Foucault, écriture feminine), has been caught in similar contradictions. Critically examining the takens-for-granted of Western thought, poststructuralism has undermined the old dualisms — civilized vs. primitive, rational vs. irrational, Occident vs. Orient — on which anthropology's exoticizations have implicitly rested. Yet at the same time, poststructuralist thought, too, urgently needs radical alterity, to show that our takens-for-granted represent European cultural constructions. To argue that our "logocentrism," our focus on reason, is a legacy of Greek philosophy, for example, requires a nonlogocentric alterity — somewhere — uncontaminated by Greeks. The various modes of feminism have always been caught in a dialectical tension between evoking tribal societies at least partly free from patriarchy and seeing the subordination of women as universal, but nonetheless historically contingent — hence preserving the theoretical and political possibility of a world not dominated by men or pervasively gendered by male-oriented, phallocentric thought. Either way, anthropology is drawn upon to provide the alterities to which envisioned ones are counterposed.

Reification and Essentialism

I will suggest that our conception of culture almost irresistibly leads us into reification and essentialism. How often, still, do I hear my colleagues and students talk as if "a culture" was an agent that could do things; or as if "a culture" was a collectivity of people. Of course, we profess that we don't really mean that "Balinese culture" does or believes anything, or that it lives on the island of Bali (it is all a kind of "shorthand"); but I fear that our common ways of talk channel our thought in these directions. Moreover, attributing to "Balinese culture" a systematic coherence, a [303] pervasive sharedness, and an enduring quality — so that Bali remains Bali through the centuries, and from south to north, west to east (even nowadays, despite the tourists) — commits us to essentialism of an extreme kind. Balinese culture is the essence of Bali, the essence of Balineseness.

The essentialism of our discourse is not only inherent in our conceptualizations of "culture," but it reflects as well our vested disciplinary interests in characterizing exotic otherness. If we arrive in a New Guinea or Amazonian community and find people listening to transistor radios or watching videos, planting cash crops or working for wages, going to church and attending schools instead of conducting rituals in men's houses — and if what we came to study was their conception of Personhood or their cultural constructions of time and space — then we have to believe that their essential culturalness lives on despite the outward changes in their lives.

Everyday ways of contemporary talk have been heavily influenced by our anthropological concept of culture. In pervading popular thought, anthropology's concept of culture has been applied to complex, contemporary ways of life — "Greek culture," "French culture," "Chinese culture" — as well as to the exotic "primitive" ones in TV documentaries. Ironically, with our all-inclusive conception of "culture," as it has passed into popular discourse, have gone our habits of talk that reify, personify, and essentialize. I recently heard a radio announcer in Australia talk about "the different cultures living in our area."

Our essentialist, reified conception of "culture," having passed into everyday Western discourse, has been adopted by Third World elites in their cultural, nationalist rhetoric. If "a culture" is thinglike, if cultural essences endure, then "it" provides an ideal rhetorical instrument for claims to identity, phrased in opposition to modernity, Westernization, or neocolonialism. A crowning irony is that through this borrowing, our own conceptual diseases may strike us down from unexpected directions. "Culture," so essentialized and reified, can serve as an ideal symbol to deploy against foreign researchers, who can be pilloried for having stolen "it," having sold "it" for profit in the academic marketplace, or simply (as outsiders seeking to interpret someone else's mystical essence) having misunderstood and misrepresented "it."

The "Cultural" of Cultural Studies

At the same time that our anthropological concept of culture has been increasingly pervading popular thought and talk, social theorists of various persuasions have started to take "culture" much more seriously. "Cultural studies" has become a burgeoning field. Is this "culture" of cultural studies "culture" as we anthropologists have conceptualized it? In general, the answer is No. The "culture" of "cultural studies" (whether post-Marxist, postmodernist, or post-whatever) has been developed through a broadening and critical sharpening of the conception to which ours has for decades been counterposed: "culture" as the highest artistic and aesthetic refinements and achievements of a complex society. It is not that "cultural studies" are preoccupied only with paintings, statues and symphonies. An increasing engagement with language and semiotic theory[2] has led to a considerable broadening of the old concept of "culture" as high art — a broadening in an anthropological direction.

What distinguishes "their" culture from "our" culture most strikingly is the stress in cultural studies on the articulation of symbolic systems with class and power — the production and reproduction of cultural forms. I will suggest that instead of our traditional anthropological conception of culture being extended to complex contemporary "societies," we urgently need to draw on the conceptual refinements in contemporary social theory, including the "cultural" of "cultural studies," to interpret the production and reproduction of symbols among the peoples we have [304] encountered along the colonial margins and in peasant communities.

We confront a deep irony, then, that our anthropological conception of culture has pervaded popular thought and has been applied willy-nilly to contemporary life; while, at the same time, some of us are beginning to question its utility and to look at alternative ways of thinking about collective symbols and meanings being explored outside our own discipline.


Let me unpack some of these arguments. First, let me say something more about the invention of radical alterity, radical Otherness. Just how different the thought and experience of non-Western peoples are from our own is a moot point about which we could all argue ad nauseam. I recently spent some weeks immersed in conversation with a brilliant young Kwaio (Solomon Islands) man who still practices his ancestral religion and lives in a world where magic, ritual, and conversations with the dead are the stuff of everyday life. Maenaa'adi's cultural alterity is perhaps as radical as any in the world of the early 1990s (although he too lives in the collages of our time, riding buses and checking the time on his watch when he comes to town). He takes for granted that if his shadow were cast on a fissure where a leprosy victim's body had been thrown, he would die of leprosy. He takes for granted that every night, his shade encounters the shades of his ancestors, who give him messages of impending events. He recites magical spells a dozen times a day, with complete faith that they should work. Obviously, I am not claiming that Maenaa'adi's world of experience and mine are minor variants of one another: There is more to it than that. Yet I see no reason, in all the texts, to infer that the pragmatic way in which he finds his way through his world is qualitatively different from the way in which I find my way through mine; or that his culturally constructed senses of individuation and agency (or personhood or causality or whatever) are strikingly different than mine.

We could argue endlessly about how radically diverse culturally constructed concepts of personhood and agency or experiences of emotion are. I believe that anthropologists have disciplinary vested interests in construing cultural diversity in more extreme terms than our ethnographic evidence justifies (Keesing 1989a). Moreover, we have ignored the implications of steadily mounting evidence that casts doubt on our extreme relativisms. Some of this evidence comes from the neurosciences and cognitive sciences, where the constraints of what can be learned and remembered by members of our species loom large, and logics of thought begin to look not at all exotic. Some comes from research on the neurobiology of mammalian emotional systems, including those of humans and other higher primates. Some comes from studies of language, both the formal structures of syntax and the logics of semantics and conceptualization. The burgeoning field of cognitive linguistics increasingly unites grammatical theory with other realms of cognition.[3]

Metaphor and Embodiedness

Languages, seen through this lens, look not at all exotic or radically diverse (even though particular languages obviously explore different logical and organizational possibilities in different ways). I don't want to go off on a tangent about this, but I want to note two lines of development I find particularly important with regard to anthropology's exaggerated relativisms. The first is the pervasive importance of metaphor in language: both (1) the way languages as conceptual systems are pervaded by conventional metaphor, as explicated (albeit imperfectly) by Lakoff and Johnson (1980), and (2) the way change in a language [305] over time whereby lexical forms acquire grammatical functions — the process that has come to be called "grammaticalization" — follows metaphoric pathways (and in ways that reveal striking similarities from language family to language family as Bernd Heine and others have shown — see Heine and Reh 1984; Heine, Claudi, and Hünnemeyer 1992; Heine and Traugott 1992). I have suggested elsewhere that with our predilection to take the most exotic possible readings of cultural texts, in fieldwork languages we learn to quite limited degrees, ethnographers run the danger of elevating conventional metaphoric schemes into cosmological structures and religious philosophies (Keesing 1985, 1987, 1989a).

The second point, which has emerged partly through studies of language and conventional metaphor but has much wider implications, is the degree to which the embodiedness of cultural experience is turning out to be critical. Studies of conventional metaphor and the linguistic encoding of relations of time, space, perception, and causality increasingly point to the centrality of embodied experience. Mark Johnson's The Body in the Mind (1987) powerfully argues the costs of Western philosophy's mind/body dualism and its focus on disembodied reason. Anthropology is now engaging embodiedness and the cultural construction of the body largely through "medical anthropology" (see e.g. Nancy Scheper-Hughes in this volume). But as Johnson's philosophical argument and feminist reflections on embodiedness (see e.g. Gatens 1983) illustrate, progress is being made on other fronts as well.

That all humans everywhere experience the world "out there" in and through their bodies, and that this embodiedness becomes the model for cultural conceptualization of spatial orientation, agency, perception, emotion, and thought, is not to say that the cultural elaborations of embodied experience all go in the same direction. Let me come back for a moment to my friend Maenaa'adi. In one sense, Maenaa'adi's experiences when he dreams and mine must be quite similar; yet he attributes a reality status to the nocturnal wandering of his soul and its encounters with others, while I assume that it is all being fabricated in my dreaming imagination. To see one's shadow as a component of the body that may be damaged or polluted, to fear the possibility of soul loss, to attribute forms of illness to magically injected foreign objects, entails having conceptions of the boundaries and dynamics of one's body very different from mine.

Culturally constructed bodies and bodily images and experiences unquestionably vary in different times and places. Yet anthropology's theoretical predilections and disciplinary interests run in this direction and make us prone both to overstate the case and to miss or underestimate a reverse phenomenon. We know a good deal, through the writings of Foucault (e.g., 1973a, 1977) and recent anthropological explorations, about how the power of the body social is inscribed on the body physical. Yet we have not yet seen clearly enough, I think, the power of the body physical — as subjectively experienced by its "occupant"to inscribe itself on cultural traditions and thus to constrain cultural diversity.

My point is not to argue endlessly what, I admit, is a partisan position with regard to radical cultural diversity. But I do want to register deep skepticism about much of what I read nowadays about the cultural construction of personhood, agency, and emotion. I believe that in a decade or two, when some of the biological/cognitive constraints on human thought, emotion, and learning have come more clearly into view, the extreme relativisms and cultural constructionist positions of our time will seem quaint in the extreme. That is what I mean by a radical alterity that does not exist.


The second point I want to expand concerns the hidden agendas in conventional anthropological [306] conceptions of "culture." I have suggested that our quest for radical alterity shapes and is shaped by our conceptualizations of "cultures" as discrete, self-contained, self-reproducing universes of shared customary practices and beliefs. For "a culture" to be a separate experiment in human possibility, it must not only be separate and internally coherent and homogeneous; the "experiment" must, as it were, be "natural" (in the sense that intentional agency and interest do not contribute to its cumulative coral reef-like form). There is a whole set of hidden agendas here. Indeed, anthropological theories of culture can be subjected to the same order of political critique as functionalist sociology in, say, its Parsonian variants — as stressing order, integration, and stasis, and hiding conflict, contradiction, and the ideological and hegemonic force of "shared" symbols and institutions. Marxists, feminists, and other critics of societies as-they-are have aptly pointed to all that is hidden by representations of "culture" and "society" as consensual, collective, coherent, integrated, and self-reproducing. The voices of subalternity (Guha 1983a, 1984a), contradictions and conflicts, the hegemonic force of dominant ideologies, cleavages of class and gender, are glossed over with a wave of the analytical brush.

Anthropological conceptualizations of "culture" have been — shall we say — innocent (in the sense of naivete, not culpability) in terms of the battle lines of social theory. Our ways of conceptualizing what used to be called the "primitive" world still embody a set of assumptions deriving from the nineteenth century about the collectiveness and sharedness of "custom." Counterposing Them to Us, nineteenth-century ethnology characterized "primitive" peoples subjugated along colonial frontiers in terms of the deep conservative force of tradition, the lack of individuation in the way they lived their lives, the unseen force of social convention. As anthropology refined "custom" into "culture," it preserved largely unexamined the assumptions about collectivity and uniformity of culturally,, defined beliefs, norms, and experiences (although there were a few dissenting queries, notably those posed by Radin).[4] The coral reef conception of how cultures cumulate remained intact. Cleavages of gender and social inequality were unreflectively hidden. The production, ideological force, and hegemonic power of cultural meanings went substantially unexamined, in a discipline that grew up (especially in North America) in curious isolation from continental social theory.

What, then, of the present? Obviously things have changed, most notably through a serious input from Marxism in the 1960s and 1970s, the emergence of an anthropology of gender and the serious study of women's lives, and a much greater openness to issues of social theory. But have they changed enough in the needed directions?

Whereas the lives of those we now study are situated in the 1980-1990s world of global mass culture, consumerism, and capitalist labor relations and dependencies, their cultures still can be subjected to that analytical sleight of hand, that denial of "coevalness," so aptly characterized by Johannes Fabian in his powerful book Time and the Other: How Anthropology Makes Its Object (1983). Their cultures are hermetically sealed, beyond the reaches of time and the world system. I have recently written (Keesing and Jolly 1992) on the way anthropological characterizations of Melanesia, where I mainly work, persistently edit out Christianity, trade stores, labor migration, contemporary politics, and cash economy, exoticizing and essentializing "traditional culture" as it ostensibly survives in hinterland villages. This is not to say that contemporary anthropology is still unwaveringly committed to the portrayal of exotic cultural alterity, as a cursory reading of any recent American Anthropological Association annual meeting program will attest. Yet this pursuit of the exotic Other is still a persistent theme, and "culture" is a powerful device for its perpetuation (as a score of recent ethnographies of [307 New Guinea and eastern Indonesia would attest).

"Culture" in Third World Discourse

Let me return briefly to the way in which anthropological talk about "culture," with its irresistible temptations to reify, personify, and essentialize, has passed into the cultural nationalist discourse of Third World elites. For the latter, Westernized though they may be, to claim that "it" is "our culture" is to make claims of identity, authenticity, resistance, and resilience. Culture, so reified and essentialized, can be subjected to metonymic transformation, so that the cultural heritage of a people or a postcolonial nation can be represented by its fetishized material forms and performances: "traditional dress," dances, artifacts. So transformed, "it" — the cultural heritage, semiotically condensed (in the sense specified in footnote 2) — can be deployed in rituals of state, art festivals, tourist performances, and political appearances to reaffirm that "it" survives despite Westernization (and hence to deny the erosion, capitalist reorganization, and pauperization of rural life). It is worth observing that such a semiotic of cultural identity has its origin in nineteenth-century cultural nationalism in Europe, expressed in an intense search for ethnic roots and folk origins, for primordiality and cultural tradition. The museum and folkloric traditions so strong in eastern and northern Europe grew out of this romantic quest for origins and local folk tradition. The Third World has inherited these European semiotic systems, and institutions, in the colonial process, and they have been deployed in a parallel affirmation of cultural identity.

I have noted (1989a) the ironies that emerge when a conception of culture, indirectly borrowed from anthropology, is used to denounce foreign researchers, with anthropologists as the quintessential villains. They, as outsiders (it is argued), can never penetrate Our essence, never really understand Us. Once "a culture" has been reified and hypostatized as a symbol, the outside researcher can also be accused of appropriating "it." "It" can be commoditized as well, depicted as having been alienated by an anthropologist and sold for profit in the academic marketplace. I take it as the crowning irony that our own conceptual diseases should be deployed against us.


Let me come back to the extreme cultural-constructionist and relativist positions, the postmodernist elevations of ethnography as the core of the discipline, and the indulgent subjectivism and narcissism of "experimental" ethnographic accounts, that have been so prominent in the practice and politics of recent American anthropology (see e.g. Clifford and Marcus 1986). I do not find it coincidental that these approaches have flourished during the conservative era of Thatcherism and Reaganism.[5]

There is no reason why the new modes of representation need be conservative, in ignoring contemporary state terror and taking for granted the political economy of global capitalism that continues to drain wealth out of Third World countries and hold them bonded in debt, pauperizes hinterland communities, and generates ecological devastation.[6] What remains unsaid in much of current anthropological writing speaks deafeningly, if we only step back to listen — as Edward Said has pointed out (Said 1989).

The not-so-hidden agendas in anthropological conceptions of culture take me to a further and final expansion of my argument. I have suggested that conceptions of culture narrower than those conventional in anthropology are being developed in the field of "cultural studies," as a burgeoning concern in social and critical theory. I have further suggested that the theoretical developments in this area could usefully illuminate anthropological understanding [308] of the communities in which we work, "tribal," peasant, or urban.

Critical Theory

"Cultural studies" is very much contested ground, and it is being entered from a number of directions. One incursion comes from critical theory. Critical theory in its various forms has as its central concerns literature and "the arts." I would include here much of postmodernism, and much of continental poststructuralism, especially its American manifestations ("deconstruction" in literary criticism, etc.). The key intellectual sources for contemporary critical theory are extremely diverse in terms of "traditional" disciplinary compartments, and include not only literary critics, but also philosophers, psychoanalysts, anthropologists, linguists, and semioticians. "Cultural studies" in this mode takes as the "core" of the "cultural" the literary and artistic productions of complex contemporary "societies," [and] hence builds from a conception of culture against which anthropology's encyclopedic conception has been opposed for decades. However, the concept of the cultural in "cultural studies" as envisioned by critical theorists has been broadened considerably, on the one hand by a pervasive concern with language in all its manifestations (the influence of semiotics and structuralism), and on the other, by postmodernism's refusal to persist with the "great divide" between "high" and "popular" culture (see Huyssen 1986).

"Post-Marxist" and Feminist Approaches

Other incursions into the realm of "cultural studies" have come from various forms of social theory. I will illustrate this with what I will (in the spirit of the time) call "post-Marxist" approaches. Here we can take as illustrative the work of Willis, Hall, Hebdige and the colleagues in England, the work of Ranajit Guha and his Subaltern Studies colleagues,[7] and the writings collected in the recent volume on Marxism and the Interpretation of Culture (Nelson and Grossberg 1988). The inspirations here in some cases come more directly from Gramsci than from Marx. But the ultimate origins lie in Marx's own insights regarding symbols and ideology.

For most post-Marxist theorists, too, the core of the "cultural" lies in publicly developed symbolic productions. However, there is a second and more anthropological reading of culture as concerned with the symbolic aspects of "everyday life" (notably in the work of Bourdieu, Guha and his colleagues, and Willis). As with the critical theorists, concerns with language and semiotics have led to a considerable broadening in conceptions of "the cultural." Grossberg and Nelson (1988:5-6) write that:

Cultural theory has now expanded the category of culture well beyond "the best that has been thought and said," beyond the general forms of art, language, and entertainment, beyond the leisure (that is, non-labor) activities of the general population.... A foregrounding of issues of language [has] promoted a broad concern with culture, symbolic forms, communication, and meaning.... Marxism began to [recognize] that cultural analysis needed to be concerned with all the structural and meaning-producing activities by which human life is created and maintained.

In this broadening, Gramsci's insights regarding hegemony and Foucault's insights regarding power and knowledge and critiques of ideology loom large.[8] A more subtle conception of the ideological process and the hegemonic power of meanings is emerging in recent post-Marxist writing. Ideologies, in this view, create not illusions per se, but idealized subject positions in terms of which (some elements of) the social and cultural worlds are portrayed: they define perspectives and hence stances and takens-for-granted, rather than simply veiling realities with deceptions. I quote from Stuart Hall (1988:44) to exemplify:

The circle of dominant ideas does accumulate the symbolic power to map or classy the world for others; its classifications do acquire not only the constraining power of dominance over other modes of thought but also the inertial authority of habit and instinct. It becomes the horizon of the taken-for-granted: what the world is and how it works ... setting the limit to what will appear as rational, reasonable, credible, indeed sayable or thinkable, within the ... vocabularies of motive and action available to us.
Feminist theory in its various forms has been another area where important theoretical clarity has been achieved. Feminists share with scholars in the Marxist tradition (and of course there is considerable overlap here) a theoretical starting point that questions and challenges any symbolically constructed status quo. No coral reefs here: Who produces the dominant symbolic systems of a society, what interests they serve and hide, what ideological force they carry, and how they operate hegemonically to shape the consciousness of those they subordinate are all rendered problematic from the outset.

Toward a Different View of the Cultural

If social theories in their various forms — poststructuralist, post-Marxist, feminist — have learned usefully from anthropology's broader conception of the cultural (albeit perhaps mainly indirectly via semiotics), I would argue that these alternative approaches to cultural theory being developed in "cultural studies" have much to teach anthropology. They are squarely concerned with precisely what anthropology's "culture" as coral reef hides: the historical situatedness, production, and hegemonic force of cultural meanings in terms of the internal structures and cleavages of "society."

Even in those tribal communities where we anthropologists have sought radical alterity, without cleavages of class, the production of cultural forms and their hegemonic force demand interpretation. As we have sat in Amazonian or New Guinea men's houses recording cosmologies, rituals, pollution taboos, our coral reef conception of culture has deflected us away from such questions. Even when cleavages and inequalities beyond those of gender have been expressed all around us in daily life and talk, as in Bali or India, we have been prone to immerse ourselves in wondrous cultural richness, and not analyze the political economy of symbolic forms.

Let me try to be quite clear about what I am saying, and not saying. I am not arguing that we should adopt a concept of culture that takes paintings to be more cultural than cookbooks or umbrellas or pollution taboos, and try to force our comparative data into such compartments. I am saying that what anthropologists and other social theorists need is a concept of the cultural that adequately characterizes both complex modern ways of life and those of small-scale communities, past and present:

1. Such a view of the cultural (I avoid "culture" deliberately here, to avoid reification as best I can) would take the production and reproduction of cultural forms as problematic; that is, it would examine the way symbolic production is linked to power and interest (in terms of class, hierarchy, gender, etc.) and would hence probe what I have elsewhere called the "political economy of knowledge" (Keesing 1987).

2. Such a conception would assume that (many elements of) cultural traditions carry ideological force — again, not in a crude Marxist sense of distorting reality or creating false consciousness, but in Stuart Hall's (1988) reformulation of a Gramscian conception: that ideologies define the world in terms of idealized subject positions: "a brave warrior," "a virtuous woman," "a loyal subject," "a dutiful son."

3. Further, a critical conception of the cultural would begin with an assumption that in [310] any "community"/"society," there will be multiple subdominant and partially submerged cultural traditions (again, in relation to power, rank, class, gender, age, etc.), as well as a hegemonic force of the dominant tradition. In these respects, feminist theory and post-Marxist theory in particular have opened to view and critically examined precisely what anthropological theory has been at pains to hide or deny.

4. Finally, a more critical cultural theory would make no assumptions about closed boundaries within which cultural meanings hold sway: "a culture" as bounded unit would give way to more complex conceptions of interpenetration, superimposition, and pastiche.

Revisiting theories of culture, then, I think in this realm we now have more to learn than to teach.


1. The initial version of this paper was presented in a symposium on "Assessing Developments in Anthropology" organized by Rob Borofsky at the American Anthropological Association meetings in Washington, D.C., in November 1989. I am grateful to George Stocking for helpful comments in his role as discussant. An expanded version was presented as a lecture at the University of Manitoba in February, 1990; for comments and questions that led to further revisions, I am indebted to Rod Burchard, Jean-Luc Chodkiewicz, Yngve Georg Lithman, James Urry, and Raymond Wiest.

2. Editor's note: "The central idea in semiotics is a particular conception of the structure of the sign which is defined as a bond between a signifier and the signified: for example, the bond that exists between a series of sounds (signifier) and their meaning (signified) in a given language, or the social convention that the colour red stands for danger. Semiotic research involves...all the mechanisms which serve both to produce and obscure meanings, and to change meanings in sign systems" (Kuper and Kuper 1985:743).

3. Here I commend George Lakoff's book Women, Fire and Dangerous Things (1987); the new journal Cognitive Linguistics is also a valuable source.

4. Yngve George Lithman has usefully reminded me that the anthropological conception of culture was itself strongly influenced by wider intellectual developments (such as German romanticism, as it became expressed both in intellectual debates and in cultural nationalism); and George Stocking has commented in a similar vein on the transformations in social theoretic vision that separate the Boasian tradition and its invocations of "culture" from the Tylorian conception and project.

5. Cultural interpretations illuminate some questions, but they hide a host of others. If classical Orientalism, including its anthropological forms, was part and parcel of the imperialist process, the neo-Orientalism in which we are caught up is part and parcel of the global economy and political climate of our time.

6. See e.g. the recent writings of Michael Taussig on the culture of terror (Taussig 1987).

7. See Guha (19836, 19846, 1985, 1986, 1987).

8. Foucault demolished the classical Marxist theory of ideology as a class-interested disguise of underlying truth. To see ideologies as masking the truth assumes some privileged access to truth which is no longer philosophically admissible. Marx himself glimpsed in places much more subtle views of ideologies and their force, but it has remained for Gramsci and his contemporary successors to develop them coherently.

*This essay is from Assessing Cultural Anthropology, ed. R. Borofsky. New York: McGraw-Hill, Inc., 1994, pp. 301-310. The bibliographical references in this essay are found in the consolidated reference list at the end of this volume.