Cultural Materialism

Transcript

Cultural Materialism: Food for Thought or Bum Steer? [and Comments and Replies] Drew Westen; Michael Chibnik; Paul Diener; Jeffrey Ehrenreich; Madhav Gadgil; B.G. Halbar; Marvin Harris; Thomas W. Hill; Allen Johnson; Roger Joseph; Paul J. Magnarella; Michael Painter; Andrew P. Vayda Current Anthropology, Vol. 25, No. 5. (Dec., 1984), pp. 639-653. Stable URL: http://links.jstor.org/sici?sici=0011-3204%28198412%2925%3A5%3C639%3ACMFFTO%3E2.0.CO%3B2-9 Current Anthropology is currently published by The University of Chicago Press. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/ucpress.html. 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. The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected]. http://www.jstor.org Thu Mar 20 11:01:31 2008 CURRENT ANTHROPOLOGY V01. 25, NO. 5, December 1984 O 1984 by The Wenner-Gren Foundation for Anthropological Research, ail rlghts reserved 0011-320418412505-0004 $ 1 85 Cultural Materialism: Food for Thought or Bum Steer? by Drew Westen THE QUESTION OF THE RELATION between "ideal" and "real" has long vexed students of the human condition. Ever since Plato's contention that reality is merely an imperfect reflection of a world of ideal forms, the issue of real and ideal has preoccupied philosophers. This question was central in Marx's critique of Hegel. Hegel had argued that history is the progressive realization of Reason or Spirit. Marx "stood Hegel on his head" by arguing instead that the real dialectics lie in the economic infrastructure and that human consciousness and culture are products of material life, not vice versa. Being "stood on one's head" for a prolonged period of time tends to affect one's ideas, as nonmaterial blood rushes to an immaterial brain, and the "idealist camp" has, in anthropology, come a long way since Hegel's Absolute unfolding itself through time. Cognitive, structural, and symbolic anthropology all share a belief in the irreducibility of meaning to matter (though Lkvi-Strauss would ultimately make such a reduction to brain physiology), and writers such as Rappaport (1979) who are not themselves associated with "idealism" have argued that a hermeneutic approach can-and in fact must-coexist with a causalmechanical model of culture and social reality. Materialism, too, has developed in the past century and a half. With the possible exception of Godelier's (1974, 1977a) attempt to apply Marxist analysis to anthropological data, the most comprehensive model of culture and society from a materialist perspective is Marvin Harris's cultural materialism. In many ways, Harris's advocacy of his brand of anthropology in Cultural Materialism (1979) is a model of exposition: he unambiguously states his epistemological position, spells out in detail his method and theoretical assumptions, presents specific theories utilizing the method, and systematically critiques competing paradigms. In so doing he makes conscious his scientific unconscious and thus both helps others to understand his ideas and opens them to rigorous challenge. is a doctoral candidate in clinical psychology at the DREWWESTEN University of Michigan and a psychotherapist at the University of Michigan Counseling Center. Born in 1959, he received his A.B. from Harvard University in 1980, an M.A. in social and political thought from the University of Sussex in 1981, and an M.A. in clinical psychology from the University of Michigan in 1983. His research interests are culture and personality, cultural conceptions of self, and the role of emotion in motivating individual and collective behavior. He is the author of Self and Society: Narcissism, Collectivism, and the Development of Morals (Cambridge: Cambridge University Press, in press). The present paper was submitted in final form 11 I 84. Vol. 25 . No. 5 . December I984 My aim in this article is to present a comprehensive critique of the method and theoretical assumptions of Harris's materialism. I do not intend to argue with his analysis of specific events, except in the course of making a more general point. In criticizing his overall model I am not suggesting that we throw out a number of vigorously kicking theoretical babies with what I hope to show is a rather stagnant bath. While many specific cultural materialist explanations can be integrated into other models which pay attention to "demo-techno-econoecological" variables, my intention here is to demonstrate that cultural materialism as a paradigm (or preparadigmatic "school") for explaining culture is not only flawed, but irremediably so. I shall first briefly review Harris's model, then provide a critique of cultural materialism as a method and model, and finally reanalyze Harris's discussion of cattle killing in Kerala to suggest a more fruitful approach. CULTURAL MATERIALISM: A THEORETICAL OVERVIEW Harris's system emphasizes the distinctions between emic and etic and between behavioral and mental events. In dividing sociocultural reality into infrastructure, structure, and superstructure and making the methodological assumption of infrastructural determinism, Harris gives priority to the etic and behavioral over the emic and mental. Infrastructure, for Harris, includes the mode of production and what he calls the "mode of reproduction." Rejecting Marx's categorization of relations of production as part of the infrastructure, Harris includes subsistence technology, "techno-environmental relationships," ecosystems, and work patterns as components of the mode of production. He contends, further, that Marx failed to consider the importance of the "mode of reproduction" in determining culture and social structure, and he includes in this category such factors as demography, mating patterns, infant care (evidently child care is somehow derivative, whereas infant care is not), and population control mechanisms. Cultural materialism rests upon the assumption that most of sociocultural reality is determined by and thus predictable from a knowledge of the material infrastructure. What Harris calls the "structure" includes domestic and political economies (including household patterns, roles, political organization, social hierarchies, war, etc.). The "superstructure" includes the arts, ritual, advertising, recreational activities, and science, which constitute the behavioral superstructure, and knowledge, thoughts, and ideology, which comprise the emic and mental superstructure (197952-54). Infrastructure is central, according to Harris, because it is the main interface between nature and culture (p. 55). The basic postulate of cultural materialism is the "principle of infrastructural determinism": "The etic behavioral modes of production and reproduction probabilistically determine the etic behavioral domestic and political economy, which in turn probabilistically determine the behavioral and mental emic superstructures" (pp. 55-56). What Harris means by this, in terms of research method, is that one should assume that any aspect of culture is determined by infrastructural variables and construct explanations accordingly. Any residual "error variance" should then be explained in terms of structure or, if need be, superstructure. The model is primarily unidirectional, with causality running from infrastructure to structure to superstructure. Though he does postulate a "feedback" process from emic mental superstructure to the rest of the system, he has never elaborated this process in his theoretical work or explicitly resorted to such explanation in his analyses of specific phenomena. Having sketched Harris's model in broad strokes, I will now try to show why it does not work. I will focus upon his Cultural Materialism, since it is there that he presents his view systematically. A CRITIQUE O F CULTURAL MATERIALISM Distinguishing between elements of infrastructure and superstructure is by no means as easy as it appears (Wallace 1980, Adams 1981). First, Harris classifies science as part of the behavioral superstructure. Yet it is difficult to see how science can be superstructural while technology is infrastructural. How does one classify a computer? Does its categorization as infrastructural or superstructural depend on its use (e.g., when used by an academic on university time it is superstructural, but when used by the same academic while consulting for a business it is infrastructural)? Further, it is difficult to see how the thought processes of a technician trying to find the problem with a malfunctioning machine can be emic and superstructural while the machine itself, the tools he is using, and his behavior in trying to fix the machine are infrastructural. Are his thought processes (e.g., problem solving by trying out operations in his head before actually performing them on the machine) determined by his behavior, or is his problem-solving behavior a reflection or outcome of his thought? Secondly, Harris considers both subsistence-maintaining labor and "work patterns" infrastructural. This leaves open the question of whether all labor is infrastructural. If so, the labor of the artist is infrastructural, which makes the relegation of art to superstructure untenable in that the products of other types of labor (e.g., for subsistence) are considered infrastructural (such as food). In a less extreme case, the labor of a mill worker manufacturing women's blouses is not subsistencedirected but certainly appears to be an aspect of the mode of production. The number of blouses produced, however, is a function of supply and demand, and demand is in large measure dependent upon fashion trends and culture-specific "desires" for clothing beyond the bare necessities. These factors ("fashion" and "desires") are superstructural and ernic, yet they determine production a t the infrastructural, economic level. If, on the other hand, mode of production is defined as including only production for subsistence, difficulties multiply. One can hardly deduce American culture and political structure from knowledge about the technoeconomic structure of American agriculture. One cannot even deduce the rest of the American economy from such knowledge. Defining infrastructure in this limited way would also raise the question why farmers, as the ones with a hold over subsistence, do not control every human culture. Thirdly, as Friedman (1974) and Godelier (1977) have forcefully argued, one cannot meaningfully separate the forces of production (technology) from the relations of production. It is difficult to imagine how one could study "work patterns" without looking at class and relations of ownership. Industrial technology appears to be compatible with both capitalism and state ownership; thus one could not argue that modern technology produces one political structure or another in any direct way (Friedman 1974). If the principle of infrastructural determinism cannot, for example, largely account for the difference between the political systems of the United States and the Soviet Union, industrial powers locked in the same struggle for political and economic security or hegemony (depending on whether one takes an emic or an etic view), then it does not appear very useful. Fourthly, as Magnarella (1982) argues, if one takes Harris at his word about "probabilistic" determination, then cultural materialism can explain only a very circumscribed part of cultural "variance." If one attaches a high estimate (.8) to the causal probability of each link in the chain from infrastructure to structure to etic superstructure to emic superstructure, then the actual variance accounted for by the infrastructure is [.814 or .41. (I have altered Magnarella's argument somewhat here.) This figure itself is bloated, since, as Magnarella points out, "infrastructure" is not a sinele variable but a combination of " many variables with numerous causal links. At best, one could hope for a "probability" around . 2 . In social science, this is a very high figure, but an approach that on principle excludes, or at best provides no insight into, the processes that account for 80% of the variance cannot be the general science of culture Harris intends. If one thinks of pure infrastructural determinism as one end of a continuum of which no infrastructural determinism is the other, then it is incumbent upon the anthropologist to develop a more general theory to subsume Harris's that elucidates the conditions which make for relatively more or less infrastructural determinism. Finally, how does one decide when one has exhausted the possible infrastructural explanations and therefore must look for causal factors in the structure or superstructure? In his concrete analysis, Harris has never found a need to go beyond the infrastructure and often, despite his lip service to "feedback," claims to explain the entirety of a phenomenon infrastructurally. For example, in critiquing Lewis's "culture of poverty" (1966), Harris writes: "None of the victims of the culture of poverty are victims of a self-replicating design for living because every aspect of that design for living is determined by the etic infrastructure and structure of U.S. capitalism" (1979:302, emphasis added). I agree with Harris's liberal sentiments for moral (and not scientific) reasons, but I would challenge him to explain the 14-year-old with two children and a free family-planning clinic down the street. Cultural materialism lacks explanatory mechanisms for crucial links in the general theory. First, Harris does not elucidate the process by which infrastructural etics are transformed into superstructural emics. Is emic understanding a direct reflection of reality? If not, how is it produced? Secondly, cultural materialism does not provide a theory explaining how etic infrastructural functional necessities are translated into conscious and unconscious intentions. How, for example, does the greater availability of beef (Harris and Ross 1978) become a psychological preference for beef? Thirdly, a mechanism explaining how infrastructural adaptiveness creates superstructural ideology is also missing. Harris's (1966) seminal essay on India's sacred cattle provides a good illustration. Harris seems to vacillate between saying that CURRENT ANTHROPOLOGY the prohibition on killing bovines is adaptive and therefore explainable and saying that the peasants do not really follow it anyway. The former suggests infrastructural determinism, whereas the latter runs counter to a materialist explanation in implying a conflict between infrastructural necessity and culture. Assuming that the prohibition makes ecological sense, one could explain the peculiar convergence of religious belief and functional necessity by arguing for a kind of cultural selection by which groups with the prohibition would supplant those without it. This is untenable for three reasons. First, we have no evidence that that type of cultural selection of small groups took place over the last three millennia in India. Secondly, evolution (or more precisely adaptation) of this sort requires random variation (or at least significant variation among local groups) and selective retention of adaptive traits through cultural selection. Since the appearance of world religions (as opposed to "randomly" local beliefs), however, the ideological variation has been too small to account for differential selection. Thirdly, a prohibition on the slaughter of cattle rationalized by a doctrine of the sanctity of life could not have developed and been selected randomly; rather, it would have required a rational assessment of the situation and an intentional (i.e., emic mental) act on the part of a shrewd religious leader and a population anxious to believe. Clearly lacking is a theory of how this intention is transformed into religious dogma and why it is accepted by a population that may be starving in the short run and have every reason not to heed it. Assuming instead that peasants routinely evade the proscription of "bovicide" in various ways (e.g., by denying that they are actually starving the cattle to death [see Harris 1966]), one has left the province of cultural materialism, because now one is arguing for a conflict between ecological adaptiveness and cultural prohibitions that impinges upon individual actors. This view calls for a psychocultural explanation: the individual faces a conflict between an impulse (eating) and an internalized norm (preserving the life of cattle) and resolves the conflict by consciously maintaining a belief while simultaneously subverting it. Fourthly, all of this raises a central question about rationality, ideology, and "mystification": why are cultural prescriptions and proscriptions presented as sacred obligations instead of as utilitarian rules of thumb? Harris's answer is straightforward: "Total interdiction by appeal to sacred sanctions is a predictable outcome in situations where the immediate temptations are great, but the ultimate costs are high, and where the calculation of cost-benefits by individuals may lead to ambiguous conclusions" (1979:193). Harris must presuppose a Platonic or Rousseauian lawgiver who, unlike everyone else around him, correctly sees the long-term good of the group and has the guile to convince everyone that his costibenefit analysis is holy writ. Harris runs into further difficulty in trying to deal with this problem. He argues that the investment of incest taboos with "so much guilt, anxiety, and symbolism reflects deep confusion and ambivalence concerning the cost/benefits of incest"; invoking the sacred "cuts through" this ambivalence and prevents "each new generation from repeating the trials and errors of past generations" (p. 81). Yet if the costibenefit analysis is so ambiguous, the relative advantage or disadvantage of a practice such as incest should fluctuate with ecological conditions, so that a rigid rule going one way or the other would be clearly dysfunctional. If values and religion are important in determining survival, then ideological false moves can have massive implications for an entire culture, including its infrastructure, and the premise of infrastructural determinism is invalid. If, on the other hand, Vol. 25 . No. 5 . December I984 westen: CULTURAL MATERIALISM ideas and ideology do not play such a significant role, it is impossible to account for the persistence over time of rules such as the prohibition of incest. Functionally inert ideas should show minimal resistance to change over time, whereas in reality taboos of various sorts are universal and often longlived. Indeed, the role of values in the cultural materialist system is problematic. For example, recent studies of revolutions (e.g., Skocpol 1979) have revealed a considerable degree of autonomy of the political from the economic sphere, and there is no reason to assume that this was not also true before the great revolutions of the late 18th century. Harris himself admits the existence of "system-transforming values" which can be conceived as "feedback between infrastructure and superstructure" (p. 303). The problem with this is that in times of social change and disintegration of older structures of meaning and authority, one cannot speak of systemic feedback because a functioning system n o longer exists. The use of the word "feedback" here masks a significant problem with the principle of infrastructural determinism. The analysis of "systemtransforming values" brings one inevitably to the Weberian notion of ideas as "switchmen" that connect a society to one path of development or another in times of social change. This notion was intended by Weber as an antidote to Marxian determinism, and it is fundamentally incompatible with cultural materialism. If ideas can be the switchmen of modern events, they no doubt could have played a similarly influential role in the revitalization movements that have appeared throughout human history (Wallace 1980). The idealistic blueprints of a group of German-influenced Russian intellectuals in 1917 have surely altered the infrastructures as well as the ideologies of hundreds of millions of people. Dismissing this influence with the label of "feedback" does not bolster anthropological or historical understanding. When pressed (see Magnarella 1982, Harris 1982), Harris distinguishes emic innovations in art, religion, or ritual with no utilitarian results and no impact on anything of substance from "charismatic ideologies and religious prophets," as in contemporary Iran (1982:142). He concludes that emic superstructural movements of the latter sort will be short-lived and rather inconsequential in the long run (1982:142; 1979:160), but it is difficult in practice to see what this means. Iran's future course would surely have been altered if Marxist ideology had prevailed over Islamic. Infrastructural conditions, along with structural political events such as a long history of American interference, were instrumental in creating an opening for revitalizing ideology, but the ideology that took root did not merely reflect infrastructural variables. Similar infrastructure produced Marxist Libya, which is likely to move in a very different direction because of its political and ideological, as well as economic, ties to the Soviet Union. Once again, ideas appear to have been the switchmen of history, and a theory that relegates ideas to the bottom of the causal heap loses explanatory value. Harris chastises Mao for believing that "superstructure was as important as infrastructure for the birth of a communist society," and he notes, with justification, that the "Chinese people have paid dearly for Mao's eclecticism" (1979:313). Yet what is paradoxical is that Mao proved precisely what he believed, that one could drastically alter the lives of millions of people through radical ideology. Mao's emic mental superstructural beliefs about the efficacy of will and the necessity for continuing revolution deeply affected the infrastructure of Chinese society during the Great Leap Forward and the Cultural Revolution and resulted in mass starvation. Since both economics and demographics are part of Harris's infrastructure, this is a clear case in which ideology played far more than a vague "feedback" role. Mao's ideology was certainly not adap- tive, but it was nonetheless causally dominant. Ideology "going against the grain" of infrastructure is not always economically maladaptive. Taiwan's infrastructure was not so different from that of the mainland, but Taiwanese anticommunist ideology facilitated an alliance with the Americans, who pumped a fortune into Taiwanese infrastructural development. Harris's epistemology conflicts with his theoretical assumptions. Throughout Cultural Materialism he argues for the superiority of a modified empiricist conception of science over obscurantism, mysticism, animism, and other modes of knowing. Not only does he make an impassioned plea for science, but he advocates the acceptance of a particular scientific view. Unfortunately, the eloquently articulated philosophy of science he espouses is negated by his assumption of infrastructural determinism. If science is superstructural, then a particular science is to be explained in terms of the infrastructure that gave birth to it, in this case American capitalism. T o be consistent, then, Harris must be willing to accept Sahlins's (1978) criticism that his explanation of sociocultural phenomena in terms of rational costibenefit analysis is a reflection of an ethnocentric bourgeois view. Harris is thus faced with a dilemma: either he is right about determinism, in which case his view of culture should not be taken seriously because it is culture-bound, or he is wrong about determinism, in which case his view of culture should also not be taken seriously. What is worse, his methodological assumption of infrastructural determinism mandates precisely the cognitive relativism he wishes to combat: if cognition (whether scientific or otherwise) reflects infrastructural forces, then all cultures' cognitive constructs are equally culturally and historically specific and the problem of emics and etics dissolves because there can be no etics. As Beidelman (1982:1245) has pointed out, Harris "fails to provide a proper epistemological apparatus by which one's own emic thought is related to interpretation of the etic of others." Again, by his theoretical principles, Harris would have difficulty explaining why he has spent so much of his lifetime dealing in ideas, trying to change the way people think. If he can change the way a reader thinks without changing the reader's or society's material conditions, then the reader who has come to believe him has been duped. Further, he has the same difficulty as Marx in explaining why people such as intellectuals frequently act in ways not in accordance with their material interests. Harris uses different principles to understand the superstructure of his own society as opposed to that of other societies and historical periods. His analyses of contemporary affairs, especially intellectual affairs, run counter to his theoretical system. For example, "The success of structuralism remains incomprehensible until one grasps that it is the most important surviving European representative of the cultural idealist tradition" (p. 165). Geertz would glow at this explanation of ideology in terms of endurance of symbolic structures. Similarly, according to Harris, Jack Goody failed to see the importance of infrastructural variables because "he is committed to an eclectic strategy" (p. 307). Phenomenology "appeals to many anthropologists who are dissatisfied with the status quo, and who identify with the aspirations of oppressed minorities, the young, and the third world" (p. 325). If one can explain anthropological ideology in terms of symbolic structures and psychological processes, it is difficult to see why one cannot apply similar causal explanations to the belief systems of other peoples. UNIT OF ANALYSIS An equally significant problem for the cultural materialist is determining the appropriate unit of analysis. In arguing that rational costlbenefit analysis essentially determines cultural traits, one must be careful to specify which costs and which benefits accrue to which individuals or groups. Not only has Harris not been careful in this regard, but one cannot be so within the context of cultural materialist principles. A good illustration of the problem is Harris's previously cited comments about appeal to the sacred when decisions by individuals may be shortsighted. For whose benefit are these appeals to the sacred? At times, as in this example, he seems to see the proper unit of analysis as the entire population. Yet the problem of delimiting the relevant group has been with us at least since Leach's Political Systems of Highland Burma (1954), and it is one of the difficulties that brought functionalism to its knees (and threatens any approach based on systems theory). Further, one cannot analyze intrasocietal conflict if one assumes that ideology conforms to the long-term needs of the population or culture as a whole (see Sahlins 1979:42). This is all the more evident when one turns one's attention to stratified societies, in which various individuals and subgroups have conflicting material interests. As soon as elite classes arise, regardless of the standard of living of the masses, subsistence should no longer be so determinative of culture. This is because, assuming that Harris is right about class domination, if the struggle for subsistence is not central for elites it should not determine the social structure or ideology of the culture. According to Harris (1966:54), much of the flesh of the 25 million cattle and buffalo that die every year in India is probably consumed by human beings-Moslems, Christians, and particularly untouchables: "Indeed, could it be that without the orthodox Hindu beef-eating taboo, many marginal and depressed castes would be deprived of an occasional, but nutritionally critical, source of animal protein?" This is an odd brand of functionalism that sees the upper castes as benevolently restraining themselves from eating meat so that their poor brethren can occasionally do so. I t also runs counter to Harris's more general rule of seeing domination by selfinterested elites. I t is also difficult to understand how men and upper classes manage to pursue their interests to the detriment of everyone else. In many hunter-gatherer societies women provide the greater proportion of subsistence and should thus, if infrastructural determinism is valid, tend to dominate. Similarly, the exploitation of peasants by landed elites exists despite the numerical superiority of peasants. That traditional peasantries often failed to throw off these seeming parasites has as much to do with emic superstructural beliefs and values as with the military muscle of elites who were by definition dependent for subsistence on the productive powers of the masses that outnumbered them. This raises the issue of how ideology comes to reflect the interests of dominant classes, which is central to Harris's claims but in direct opposition to his theoretical assumptions. For example, Harris explains the relationship between unilineal descent groups, stratification, and the emergence of the state by showing how the "interests of increasingly powerful chiefs" came to conflict with, and predominate over, the interests of their lineages (1979:99). Elsewhere, he writes of the "deepening misery of the lower castes and classes" brought on by "politico-economic exploitation and Malthusian penalties" (p. 108). His "demystification" of the world religions shows that Confucianism, Taoism, Buddhism, Hinduism, Christianity, and Islam were invented or coopted by self-interested elites: "By spiritualizing the plight of the poor, these world religions unburdened the ruling class of the obligation of pro- CURRENT ANTHROPOLOGY viding material remedies for poverty" (p. 110). In sum, peasants tend to act on their rational interests, "provided, of course, that their perceptions are not subject to continued manipulation by classes or factions that benefit from the status quo" (p. 300). I t should be apparent that this is an emic superstructural explanation: the structure of society depends upon the "perceptions" of the peasants. Moreover, it is one that places the problem of unit of analysis in stark relief. What would happen if the peasants stopped believing these selfserving mystifications of the ruling classes? Would this destroy "adaptiveness" to the ecosystem? In all probability it would alter the balance of power in the society. Such a move would be "adaptive" for one class and "maladaptive" for another. Even if one accepts that class domination may originally be maintained by military force, one cannot explain why peasants frequently buy into the ideological superstructure that reinforces domination. A far more "adaptive" strategy, from the peasants' point of view, would be to reject that belief system and overthrow the upper classes the moment the political climate was right. The difficulty with the cultural materialist position is that as soon as intracultural conflict becomes as significant as convergent interests, one can no longer make simple arguments about adaptiveness producing cultural traits; instead one must show why one group or class sacrifices its "adaptiveness" for the "adaptiveness" of another group or class. Harris claims that "it is always cheaper to produce obedience through mystification than through police-military coercion" (p. 102). This, however, negates the whole point he has so forcefully argued throughout his work: behind apparent mystification (whether in pollution rites, food prohibitions, or incest taboos) lies good common sense from an adaptive point of view. In analyzing stratified societies he abandons this assumption and starts talking about the force of irrational beliefs. He would thus have us believe that until 10,000 B.C. people were not mystified and tended to operate as rational actors, but with the rise of agriculture, literacy, increased knowledge, and the like they suddenly lost their rationality. T o put it differently, Harris assumes a rational psychology of selfinterest that applies to everyone except peasants and proletarians. With the rise of stratification comes the development of two kinds of people with radically different psychologies: those who think well (and can thus trick their fellows) and those who do not. Cultural materialism rests upon an untenable psychology, and the principle of infrastructural determinism precludes the possibility of a more adequate one. When he is at his most consistent, Harris locates costlbenefit analysis at the level of the individual and even speaks of it as "infrastructural" (p. 193). The problem is that such analysis is emic and mental, not infrastructural and behavioral, and therefore the ultimate causes of behavior and culture must be superstructural. This is all the more true if the categories that mediate between reality and individual costlbenefit computations are at least in part culturally constructed. In fact they are: economists and economic anthropologists have spoken of the "mediating role" of ' Perhaps it would make more sense to say that human beings have always been mystified about social reality and have expressed their understanding in, and been directed by, systems of symbols that are not always reflective of subsistence needs. As I have argued elsewhere (Westen n.d.), such meaning systems, which serve as both blueprints for and explanations of social life, reflect compromises between the various individual and collective needs operative in a culture; with the rise of radical inequality, the compromises encoded in those structures of meaning are bound to change and to reflect the differential power in obtaining need satisfaction of different individuals and groups. Vol. 25 . No. 5 . December I984 Westen: CULTURAL MATERIALISM "preference schedules" (Cook 1973:840) that underlie demand curves and that are shaped by culture. The example presented earlier of the influence of fashion on blouse production shows that if the infrastructure is dependent upon costibenefit analyses that are psychological, and psychological processes and desires are culturally conditioned, then the infrastructure is as much the product as the producer of culture. This can hardly be understood as intrasystemic "feedback." Harris never specifies where psychological needs fit into his system. Are they emic or etic? What is fatal for his system is that they are actually determinative of infrastructure while themselves being structurally and superstructurally conditioned. The psychologist Bandura (1977) has convincingly shown the extent to which preferences as well as behaviors are produced simply through copying or "modeling" the behaviors and motives of others. If one looks carefully at Harris's infrastructure, one finds that it exists in response to psychological needs. At the simplest level, the mode of production is important because people need to eat; if that were not so, production of subsistence would determine very little. Harris claims to rely on only four psychological assumptions: people need to eat, try to expend minimal energy, are highly sexed, and need love and affection (p. 63). But what about the care of infants, an infrastructural variable included under "mode of reproduction"? None of these four principles explains it. What does explain it is the bio-psychological motive to take care of one's young that humans share with other animals. Similarly, demography is drastically affected by the tendency of humans not to kill off every child at birth. This infrastructural phenomenon can only be explained bio-psychologically. If psychology thus underlies infrastructure, and psychological needs are influenced by culture, infrastructural determinism cannot be sustained. Not only are Harris's four psychological principles inadequate, but his approach precludes the possibility of an acceptable psychology. In particular, his relegation of the emic and the mental to the superstructural and largely epiphenomena1 renders a comprehensive understanding of behavior impossible. His epistemology tends toward that espoused by Skinner in psychology, with its denigration of the mental, the unobservable, and the nonenvironmental. According to Harris, "the objective of cultural materialism is to predict both ideas and behavior from a knowledge of behavior" (p. 271). This is a far more ambitious goal than even that of Skinner, who was content to ignore ideas and predict behavior from behavior. I t is ironic that Harris published his book in 1979, at the end of a decade in which psychologists had come to abandon the behavioral, nonpurposive psychology of Skinner because it had proven incapable of dealing with the complexities of cognitively mediated behavior. One by one psychologists came to look to the mind, to people's understandings of themselves, their situations, and other people, for an explanation of behavior. Cultural materialism cannot incorporate such a psychology because it separates behavior and thought and considers the former primary. Harris justifies this move on the operationalist ground that one can use different operations to make scientifically acceptable statements about each realm. The problem is that psychologists have found that this is precisely what one cannot do-that behavior cannot be explained without examining the cognitions that mediate between stimulus and response. If we accept for a moment Harris's notion of "individual men and women who respond opportunistically to cost-benefit options" (p. 61), the question is to what ends they direct this "opportunism." Are they Hobbesian creatures who seek pleasure and avoid pain? Are they Freudian creatures who seek to discharge their sexual and aggressive impulses? If they are selfinterested, why do they rear offspring, especially in times of 643 scarcity when they have their own mouths to feed? Is it not perhaps the sociobiologists' "inclusive fitness," not individual happiness, that is the goal of their optimizing? The fact that the human being is an incomplete animal who completes himself through culture (Geertz 1973:49) suggests that culture is instrumental in mediating conflicting claims of self and significant others that instinct resolves in other creatures; surely the Hobbesian image of the war of all against all is not an accurate portrayal of human social life. That Harris's implicit psychological premise is a Hobbesian one is clear in his specific analyses: the true aims of human endeavor are "wealth and power" (p. 107). In discussing the origins of female infanticide, he contends, against the sociobiologists, "The genesis of this system lies in the struggle to maintain and enhance power and wealth, not in the struggle to achieve reproductive success" (p. 138). Clearly a materialist argument is appropriate in explaining Harris's choice of psychological principles: as Macpherson (1962) has argued, Hobbes's psychology of "possessive individualism" is congruent with capitalism. Harris's psychology is thus indeed ethnocentric. Worse still for his theoretical system, the needs for power and wealth have been empirically shown to vary across cultures (McClelland 1961), which makes the psychological infrastructure dependent upon the cultural superstructure (such as the myths and legends that reflect and transmit cultural values) and structural child-rearing practices. Finally, if one accepts that bio-psychological needs are at the heart of infrastructure, then one must consider explanations of cultural phenomena in terms other than the rationalistic utilitarian mode of cultural materialism. For example, ethologists such as Bowlby (1969) have shown conclusively that human beings have an innate need for attachment, which I have argued elsewhere (Westen n.d.) gradually differentiates into a wide range of social needs, including those listed by Harris in his four psychological principles. If people become attached to other people, then much of culture may be explicable in terms of these "expressive," noninstrumental needs, which have little to do directly with Harris's infrastructure. Similarly, the need for meaning and identity emphasized by psychologists such as Frank1 (1959) and Erikson (1968) appears so central to human life that prolonged identity confusion and absence of a sense of meaning are diagnostic of severe forms of psychopathology. If, as is likely, these needs in part determine such phenomena as revitalization movements, religion, and ritual, then the materialist bent of Harris's science of culture necessarily leads one astray. CATTLE-KILLING I N KERALA: AN ALTERNATIVE VIEW I have now shown several ways in which cultural materialism is deficient as a method and a theory. I would now briefly like to suggest an alternative approach, which has been spelled out in much more detail in a book (Westen n.d.), and to illustrate its advantages by reexamining data provided by Harris about the covert killing of cattle in Kerala. The issue of the relationship between individual and group is not unique to cultural materialism; it has emerged repeatedly throughout the history of the social sciences. In the early part of this century, Durkheim (1938) demanded that social facts be explained only in terms of other social facts, while Freud (1950 [1913]) attempted to explain culture as a vast sublimation, or taming, of psychological drives. Most subsequent approaches have fallen somewhere between sociocultural and psychological reductionism. Over the last three decades it has become apparent that cultures are not the monolithic, homogeneous structures once imagined and that diversity exists even within the least technologically advanced societies (Pelto and Pelto 1975). This has led many (e.g., Britan and Denich 1976, Barth 1967) to focus on individual decision-making, even in domains typically believed to be regulated by cultural norms.2 One must be careful to distinguish between the analysis of individual decision-making and the assumption of individualistic decision-makers. Given that societies consist of individuals whose thoughts and actions concretely influence, if not constitute, the reality that an observer will perceive, the examination of processes at the individual level is indispensable. Yet one must be cautious, when moving to this level of analysis, lest one import naive and ethnocentric psychological principles into anthropological thinking to explain cultural phenomena. It is important to keep in mind that, while concepts similar to "rational self-interest" were suggested by some of the Sophists, this view of human motivation is rare among preindustrial peoples and did not begin to take hold in the West until enunciated by the social-contract philosophers, notably Hobbes. Aristotle, for example, began with the state and deduced the individual, not the other way around. Individuals do, indeed, make decisions, and their decisions are guided by the affective consequences of their actions (i.e., choices are "hedonically selected," to use Boehm's [I9821 term). Yet individual actions are not always aimed at maximizing individualistic aims, just as their affects are not always responses to changes in material welfare. For example, to the extent that human beings internalize norms, they will experience guilt, shame, and other socially related emotions upon transgression or anticipated transgression of these norms, and their "decisions" will be influenced by such feelings. We have every reason to believe that people in all cultures will have affective reactions to perceived threats to their lives or livelihoods and that they will respond, to the best of their ability and largely within the confines of culturally accepted or delineated behavior, as adaptively as they can. We also have reason to believe, however, that they will react with distress to threats to significant others, to their ways of understanding the world (which are significantly influenced by culture), and to the values upon which they base a sense of meaning. As I have argued elsewhere, the emotional reaction of distress will produce a response that may be either behavioral (an attempt at adaptation) or intrapsychic (what psychodynamic psychologists call "defense mechanisms" and some academic psychologists-e.g., Lazarus 1981-refer to as "coping mechanisms"). At the cultural level, ecological understanding and an examination of individual actors can be accommodated without abandoning functionalist insights into the role of various actions or institutions in the pursuit of collective ends (e.g., solidarity). Rather, both may be incorporated into a single model by recognizing that in each case social action is being activated by a "need," whether it be a social need such as intragroup solidarity or an individual need such as the biological need for protein. From a systems point of view, in both cases the discrepancy between an ideal state or "set-goal" (e.g., solidarity or nutritional satiety) and cognized reality activates a control mechanism (e.g., ritual or migration) to minimize the discrepancy. The feedback mechanism that evokes the response may be affective (e.g., the feeling that things are not going well or hunger) or encoded in the structure of a ritual or other institution itself. A few examples should make this more clear. Rappaport's (1968) classic analysis of Maring ritual cycles suggests that * A problem that looms as large today as it did at the turn of the century is how to incorporate a recognition of the role of the individual without jettisoning an understanding of structure and function in collectivities. I t is paradoxical that what is perhaps most threatening to the development of anthropological theory is an exclusive emphasis on individual behavior to the detriment of what historically has been one of anthropology's key contributions, the examination of irreducibly collective phenomena. CURRENT ANTHROPOLOGY when cognized reality indicated by various culturally encoded signs deviates from "reference values," control mechanisms such as ritual are set in motion to achieve a number of goals, including regulation of the pig population and solidifying alliances through distribution of pork. At the cultural level, a mechanism that activates the ritual process when the pig population grows too large is the hardship imposed upon the women who must care for the pigs. At the individual level, the affect (e.g., anger, annoyance) elicited by tending a burdensome pig population motivates the inaiviaual s acrion. airnllarly, LJlCKeman (1975) has shown that the Tikopia limit sexual intercourse when the population threatens to become too large. Similar control mechanisms emerge in relation to social needs unrelated to material welfare. For example, Turner (1969) discovered that periods of heightened social conflict among the Ndembu correlate with an increase in the frequency of rituals, which suggests that rituals may function as control mechanisms activated by intragroup conflict to restore social order. Cultural ideals come to function as set-goals, so that a discrepancy between the ideal and cognized reality will often produce a response aimed at reducing the discrepancy. The utility of this approach is illustrated by reanalysis of Harris's discussion of covert cattle-killing in Kerala. Harris (1979) observed that the mortality rate was twice as high among male as female calves, yet the farmers "ardently affirmed the legitimacy of the standard Hindu prohibition against the slaughter of domestic bovines" (pp. 32-33). The farmers denied any deliberate shortening of the lives of domestic cattle, though they were aware of the differential mortality rates of males and females. Harris elicited from the farmers their explanations of this phenomenon (p. 33): When I asked farmers to explain why male calves got sick more often, several suggested that the males ate less than the females. One or two suggested that the male calves ate less because they were not permitted to stay at the mother's teats for more than a few seconds. But no one would say that since there is little demand for traction animals in Kerala, males are culled and females reared. Harris notes here the discrepancy between emic and etic reality, arguing that emically the farmers were caring for their cattle in accordance with Hindu law, while etically male cattle were being systematically killed. He examines the four possible combinations of etic, ernic, behavioral, and mental in this situation in terms of four propositions (p. 38): I I1 I11 IV EmiciBehavioral: "No calves are starved to death." EticlBehavioral: "Male calves are starved to death." EmiciMental: "All calves have the right to life." EticlMental: "Let the male calves starve to death when feed is scarce. " It is difficult to see, however, how any one of these statements is more mental or behavioral than any other. The emicl behavioral proposition, for example, is a cognition, not a behavior, and it is no more behavioral than Proposition IV. Propositions I and IV are both mental: I is a (mistaken) construal of reality and IV an (unconscious) rule for behavior. Further, what Harris cannot do is to explain the relationship between these four propositions. How, for example, does one account for the discrepancy between the emic and the etic understanding of the situation? From a different perspective, this discrepancy is readily understood. The central question is how and why the actual (etic) situation (Proposition 11, male calves are starved to death) is transformed into an inaccurate cognized (ernic) model (Proposition I , no calves are starved to death). The key is the conflict between two set-goals. On the one hand, when feed is scarce, preservation of useless male cattle threatens the material welfare of the individual farmer. On the other hand, a religious ideal grants all calves the right to life (Proposition 111) and makes killing them morally wrong. Material interest was apparently more potent in determining behavior here, since the Vol. 25 . No. 5 . December 1984 Westen: CULTURAL MATERIALISM behavioral norm emerged to let male calves starve to death when feed is scarce (Proposition IV). This behavioral outcome is, however, only part of the story and is a response to only one of these conflicting needs or setgoals. Failure to satisfy the cultural ideal of not murdering cattle would, if recognized, produce fear and guilt. The discrepancy between the reality of male "bovicide" (Proposition 11) and the ideal of preservation of the lives of all cattle (Proposition 111) produced an altered cognition: no calves are being starved to death (Proposition I). The culturally approved mechanism of denial accompanied by the systematic starvation of male calves is similar to what Freud called, at the individual level, a "compromise-formation," i.e., a compromise between two conflicting motives. The function of a compromiseformation is to maximize pleasurable and minimize unpleasurable emotion, and in the Keralan case the combination of a tabooed behavior and a cognitive distortion was hedonically selected because it did precisely that. This suggests that the farmers had, in fact, two emic understandings that are contradictory: male calves are starved, and male calves are not starved. The latter is a transformation of the former, distorted by a hedonically selected control mechanism: denial. The belief that calves are being starved is what one may call an unconscious emic understanding, replaced defensivelv bv a conscious emic one. One could, of course, analyze this case strictly at the level of the individual farmer in terms of a conflict between what I have elsewhere called "self-needs" and internalized values, producing an affect which motivates a behavior (starving males) and a defense (denial). This would be valid, but it would miss the fact that the ideal of not killing cattle is a cultural one and thus has a wider significance. Perhaps more importantly, it would ignore the extent to which this defense is culturally constituted. Not only do individual farmers deny the reality of what they are doing, but they turn a blind eye to the similar misdeeds of their compatriots, who are also breaking culturally shared rules. Without cultural acceptance, this defensive distortion at the individual level could not be maintained. I should again make clear that in providing a critique of Harris's method I am not arguing that cultural materialist explanations should be abandoned. Harris has made numerous brilliant contributions to the understanding of culture, and much of his mode of explanation is in accord with current ecological thinking and can be integrated into a model that allows greater causal efficacy for ideas and ideals. Rather, I am arguing that it is time we put to rest the misbegotten hope that we will ever reduce meaning to mechanism and began to work toward an understanding of that peculiar species that thrives on symbols as well as proteins. < < Comments by MICHAELCHIBNIK Department of Anthropology, University of Iowa, Iowa City, Iowa 52242, U . S . A . 3 v 84 Harris's most important contribution to anthropology may be his identification and lucid discussion of fundamental problems associated with the description and explanation of sociocultural phenomena. He argues that anthropologists have frequently been inexplicit about the nature of cultural causality The reanalysis here of Harris's Keralan example is an abridged version of a discussion of conscious and unconscious emics from my forthcoming book. 645 and inconsistent in their use of "inside" and "outside" categories in their ethnographic descriptions. He clearly states the theoretical and methodological assumptions about causality and description used in cultural materialism and strongly criticizes those of alternative research strategies. In part because of Harris's polemical writing style and his intolerance of opposing theoretical positions, cultural materialism has been the target of numerous criticisms. Harris and other materialists have been accused of epistemological confusion (Geertz 1973:3-30; Paul and Rabinow 1976:123-24), incomplete (Chibnik 1981:257-58) and inadequate (Fisher and Werner 1978) explanation, insufficient attention to the causative power of structural and superstructural factors (Service 1969, Sahlins 1976), and failure to recognize the infrastructural determinants of their own views (Sahlins 1978). Harris has responded to these attacks, sometimes effectively, in his many publications. Westen provides a useful service in pulling together in one place many of the more frequent criticisms of cultural materialism. Perhaps because he is not an anthropologist, he does not seem fully aware that many of his points have been made elsewhere. Aside from his incisive remarks comparing Harris's ideas to those of B. F . Skinner, I find little new in Westen's "comprehensive critique." Although Harris's views have not been generally accepted by sociocultural anthropologists, he has defined the terms in which many arguments are phrased. This can be seen in Westen's discussions of causality. Westen begins his critique of cultural materialism by presenting several instructive examples of the difficulties involved in distinguishing elements of the infrastructure, structure, and superstructure. Nonetheless, Westen later argues that "psychological needs are actually determinative of the infrastructure while themselves being structurally and superstructurally conditioned." Westen thus manages in the space of his article both to question and to use Harris's terminology. by PAULDIENER 2407 W . Hermosa S t . , Sun Antonio, Tex. 78207, U.S.A. 5 VI 84 Westen criticizes Harris's materialism from the perspective of idealism. The critique is cogent, although these points have been made before. More interesting is the parallel between Westen's "alternative approach" and the theory he criticizes. Westen lauds Rappaport's stability model of ritual; so has Harris. Westen describes human choices as "hedonistically selected"; Harris also sees people as selfish. Westen feels that values maintain psychological and social "order" in the face of "disruption"; Harris shares this conservative viewpoint. Both theories lack mechanisms of historical process and a concern for origins. Can stability theories (Freudian or Skinnerian) explain the origins of revolution and revitalization in modern China and Iran, for example? Westen realizes that his theory and Harris's differ little. Thus, he credits Harris with "brilliant" contributions and claims that minor compromises will yield a synthesis. Rose and Rose (1976:7) have also noted this affinity between crude materialism and idealism: Knowledge of the social separated from the natural becomes idealism, knowledge of the natural separated from the social becomes mechanical materialism. . . . Within capitalism each becomes elevated into a world view which can serve only to prevent the realisation of true knowledge; idealism by giving primacy to ideas, to human consciousness separated from the material conditions which create it; mechanical materialism by giving primacy to nature separated from humanity's actions upon it. These world views constitute the dominant ideas within capitalist society; they are held and promulgated by and on behalf of the ruling class. Both Westen and Harris misrepresent Marxist epistemology. Although Westen claims that Marx was a mechanical materialist, for example, in fact Marx "never concealed his hostility to what both he, and Engels, repudiated as 'bourgeois materialism'-a dry, gloomy, and melancholy doctrine" (Lewis 1972:80). For Marx, "man makes the world, and what he knows is the material world made over by his work. Marx thus maintains the world as unitary reality, not divided into 'appearances which are not real, and a reality which does not appear,' which is the consequence of every dualistic theory of knowledge. " Westen claims that the "mind-matter" problem is eternal, but common people solve it in practice. While near the war zone in Nicaragua recently, I was cut off in a rural area for several days and passed the time chatting with a young studentlsoldier. When I described my own arguments with Harris, he saw the point quickly. Pointing first to his books, then to a machete and rifle lying nearby, he said, "We use these to better use those." The young Nicaraguan realized that neither disembodied ideas nor ahuman demo-techno-econo-environmental spirits (and cultural materialism is a form of animistic superstition) shape the world. He said, and I agree, "The people make the world-and they can remake it." If Harris's and Westen's views are flawed, how can we explain their wide acceptance? Here we must consider the social structure of anthropology. Pelto and Pelto (1973:258) remark: "from the anthropologist's point of view there is nothing altruistic about [anthropological research]. The information he receives is directly translatable into economic, professional, and social advantage-when he gets home." Harris tells us that his materialism seeks to strengthen "the scientific credentials of cultural anthropology within the prestigious and wellfunded natural sciences" (Harris 1968:508). This funding comes from the American government and the large corporate interests (Ford, Rockefeller, etc.). The institution with which Harris is now affiliated, for example, is very heavily involved in Latin American "development" work, including work in Central America. The idealists appeal to a different market segment. Although much research funding reflects the monopoly capitalist agenda, many anthropology departments are also dependent upon the still-large American petite bourgeoisie (through local and state governments, endowments, tuition, or influence on federal expenditures). Poor and working people are almost totally absent from American anthropology. The petite bourgeoisie experiences an "alternative realityn-that is, its interests differ from those of monopoly capital. Entrepreneurial values still matter here, as Ronald Reagan likes to remind us. This is the land of the New Right, commitments to Jesus and the Reverend Moon, communes and alternative lifestyles, Carlos Castaneda and psychic phenomena. By appealing to both the "central-state materialists" and the "petitbourgeois idealists," American anthropology maximizes its academic market share. This process is not innocent. In Nicaragua, I witnessed the removal of murdered children from a raid site an Americanconstituted army had attacked. A few miles away, on the other side of the border, I met American anthropologists and other scientists working in counterinsurgency (sometimes called "development") programs. One anthropologist working in Honduras told me that his U.S. government job was the best, and best-paid, he had ever had. Human culture is not "hedonistically selected," but clearly many American anthropologists are. To counteract the mystification in the Westen-Harris debate, I suggest that all readers of this article immediately read the Anthropology Newsletter, official publication of the American Anthropological Association. There they will find, among the several hundred funding sources and grant announcements listed in every issue, "$253 million at the National Science CURRENT ANTHROPOLOGY Foundation, $30.8 million at the National Institute of Education, $190 million a t the National Institute of Mental Health, $110,000 for the study of peasant economics, $68,000 for the analysis of warfare, $70,000 for the study of disease in the Mayan area, $50,000 for the study of Honduran settlement patterns." No one questions the propriety of this scientific espionage. It is universally assumed that there is "nothing altruistic" about anthropology. Indeed, the hundreds of millions are not enough, and we are urged to lobby for new money, because "the interest of the social and behavioral science research community coincides with the national interest" (Miller 1984:s). I find all of this tragic. by JEFFREYEHRENREICH Department of Sociology and Anthropology, University o f Northern Iowa, Cedar Falls, Iowa 50614, U.S.A. 15 vI 84 Thus, the superior man consolidates his fate By making his position correct. Ting (the Caldron), from the I Ching If the number of vocal opponents generated is any measure of social science paradigms, then cultural materialism is a theoretical perspective to be reckoned with in anthropology. No living anthropologist, with the possible exception of LCviStrauss, has had more theoretical poison-tipped darts shot his way than Harris, cultural materialism's most ardent and prolific theorist, and none has so thoroughly responded to the challenges of his critics. Westen, a psychologist, is the latest to aspire to deliver the final blow to Harris's position and influence in cultural theory. His intention is to "present a comprehensive critique of the method and theoretical assumptions of Harris's materialism" and to "demonstrate that cultural materialism as a paradigm . . . is not only flawed, but irremediably so." T o the extent that these goals become the standard of judgment, Westen has failed; cultural materialism will easily sustain this latest assault. Just as critics of Marx have consistently tried to render the thrust of his analysis false by attacking selected parts carefully isolated from the complex whole, so Westen attempts to paint Harris into a mechanistic, reductionistic, and contradictory corner. It is of course easy to dispose of cultural materialism if, as Westen implies, Harris's stress on infrastructural causality "reduces meaning to matter." (Westen confuses metaphors in this regard. Marx did not stand "Hegel on his headn-Hegel and idealism were already there. Rather, Marx claims to have set things "right side up again" by resort to a materialist perspective which by his account facilitates rational understanding and explanation of the way the world works [see Marx 1972 (1873):198;Harris 1968:233-351.) But, does the picture painted by Westen really represent the cultural materialist position? I think not. Westen's presentation accurately summarizes some of Harris's central contentions but then seriously distorts their meanings. For example, he lays out the "principle of infrastructural determinism" and then characterizes it with the suggestion that cultural materialists proceed under the assumption "that any aspect of culture is determined by infrastructural variables" and with the intent to "construct explanations accordingly" (emphasis added). But Harris's position (19795-58) defies this analysis. In brief, Harris says (p. 56): cultural materialism asserts the strategic priority of etic and behavioral conditions and processes over emic and mental conditions and processes, and of infrastructural over structural and superstructural conditions and processes; but it does not deny the possibility that emic, mental, superstructural, and structural components may achieve a degree of autonomy from the etic behavioral infrastructure. Rather, it merely postpones and delays that possibility in order to guarantee the fullest exploration of the determining influences exerted by the etic behavioral infrastructure. Val. 25 . No. 5 . December 1984 Westen: CULTURAL MATERIALISM I t is "priority" in research strategy, probabilistic determinism, and the causes of sociocultural variance, all considered in broad perspective, that are stressed in Harris's theoretical position, not the mechanical reduction o f meaning (cognition) to matter (infrastructural conditions). Giving infrastructure priority and designating it as a key to understanding sociocultural causality from the cultural materialist perspective does not imply that it is the sole basis from which all else must flow culturally or cognitively. Westen also suggests that there are difficulties for cultural materialists in "distinguishing between elements of infrastructure and superstructure." But what if there are? The conclusion to be drawn from this is not that cultural materialism is somehow damaged goods to be thrown away. Is it not more sensible to conclude that guidelines need to be established to help make classificatory distinctions easier to draw, more accurate, and more consistent (see Ehrenreich 1981)? Granted that all classification systems, by definition, draw some artificial and arbitrary lines, the specific example of the computer used by Westen to make his point is especially unconvincing. In claiming that "it is difficult to see how science can be superstructural while technology is infrastructural," Westen reveals a lack of understanding of what "science" is and is not. Science is neither the bodies of knowledge produced using its techniques and methods (e.g., physics, chemistry, astronomy, etc.) nor the products or technology which may result from its enterprise. Science is a technique for perceiving, deciphering, and understanding reality-any reality. In the eloquent phrase of White (1949), "Science is sciencing." In this sense, science is unambiguously superstructural. It follows that a computer is technology, not science, and is therefore best interpreted as infrastructural. Along similar lines, Westen makes his most intriguing suggestion in charging Harris, and by implication all materialists (including Marxists and structural Marxists), with not being able to explain himself by reference to his own epistemology and theoreticalimethodological assumptions. His argument is that it makes no sense to be in the business of trying to change the minds and thought processes of others in order to change society if one subscribes to infrastructural as opposed to ideological determinism. Harris's life work is thus trivialized, reduced to a logical contradiction. But this is absurd. The act of persuading others of the significance of infrastructural priority is neither folly nor idle, self-indulgent intellectualism-it is serious enterprise with serious consequences for real people. The issue is not merely getting people to believe but getting them to plan and to act with these principles and perceptions in mind. Knowing the powerful causal role of the infrastructural permits a society and its leaders (politicians, academics, corporate heads, et al.) to understand the past and more intelligently anticipate the future. Ideally, it also pushes people against the ideological tide of mysticism, obscurantism, fatalism, and antiintellectualism that, as Harris has so forcefully argued, stands in the way of humane and intelligent policy and contributes to political indifference, passivity, and oppression. It is notable that Marx, who believed that aspects of history were "inevitable," nevertheless advocated revolutionary action to make things happen. Harris and many other cultural materialists share with Marx the conviction that "men make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly found, given and transmitted from the past" (Marx 1972 [1852]:437). Harris's theories permit and encourage activism, even though he has personally shied away from formulating specific social and political programs or policies. In stressing infrastructural priority there is no logical requirement to cease engaging in intellectual activities on the grounds that ideological shifts are inherently beside the point. The Marxist axiom that "the philosophers have only interpreted the world, in various ways; the point, however, is to change it" (Marx 1972 [1845]:109) is perfectly compatible with Harris's cultural materialism. The record indicates that Harris firmly believes that adhering to the cultural materialist perspective in no way curtails or compromises the capacity of individuals to act or think socially and politically. Furthermore, Harris has never denied the efficacy of ideas in human history. He has argued that the transformation and implementation of major ideological trends are often based on, and traceable to, material conditions and realities. These conditions can be, have been, and will be defied ideologically, but not for long and not without paying a heavy price in the functional and organizational capacity of a society. A major problem with Westen's analysis is that he would have us believe that cultural materialism is a view of social or individual psychology rather than an approach to and perspective on the origins and causes of macro-patterns of cultural development. Likewise, he confuses questions of cultural origins with those of function and continuance (see Ehrenreich 1981), while equating individual choice and decision making (psychology) with broadly conceived ideological belief systems (cultural context). The strategy of cultural materialism predicts that certain kinds of infrastructural conditions, e.g., persistent scarcity of critical resources, help to account for the structural and superstructural elements of a culture. By use of the cultural materialist strategy, broad patterns of ideology and beliefs become more discernible and predictable. But at no time does the strategy of cultural materialism attempt to explain the nuances of all individual behavior or cognition. It is, in contrast, concerned with the widest cultural context in which behavior and cognition take place. What all this seems to suggest is that Harris and his critics are often grappling with questions that are substantively different in nature and purpose. As long as critics pose fundamentally different questions from those Harris seeks to address, they will inevitably find his answers wanting. Meanwhile, Harris continues to expand his program and strategies and to clarify his meanings, while many of his opponents choose to ignore the refinements and adjustments he offers. To push anthropological theory farther down the road, it is obligatory for critics such as Westen to consider fully the real positions of the cultural materialist strategy and not some distorted caricature conjured up for quick slaughter and disposal. To date, they have failed to do so. by MADHAVGADGIL Centre for Theoretical Studies, Indian Institute of Science, Bangalore 560012, India. 9 v 84 Westen's critique of Harris's cultural materialism is a welcome contribution to the debate. I would like to make three points: 1. Harris's position implies that human social behaviour is largely determined by the way a society interacts with its environment. In fact, the debate has gone on to argue about the proportion of the variance in an attribute that can be explained on these grounds. That Hanuman langur troops in the desert generally contain a single adult male while troops in the South Indian forest have several was unexpected and has not been convincingly explained. While ecological factors must play a significant role in determining animal social organization, this quite typical example suggests that we are still far from explaining any large proportion of its variation in ecological terms. It would obviously be far more difficult to do so for human societies. 2. Westen brings out the flaws in Harris's hedonistic model of human behaviour and suggests that the sociobiological postulate of maximization of inclusive fitness would provide a superior explanation. While agreeing with Westen about Har- ris's model, I doubt the adequacy of this alternative. Intergroup conflict appears to have played a critical role in human evolution, creating conditions favouring the operation of group selection with the result that human societies exhibit pronounced group loyalties and enmities. Thus warfare rather than subsistence may have moulded much of human culture. Harris, in his preoccupation with the role of cannibalism in the supply of protein, seems largely to have missed this point. Westen also fails to elaborate on it. 3. Westen rightly criticizes Harris's preoccupation with the primacy of subsistence strategies. In western India there are three major endogamous castes, the Nandivallas, the Vaidus, and the Phaseparadhis, that derive a major proportion of their subsistence from hunting. Each of these castes has its own hunting technique and specializes in particular prey animals. Thus the Nandivallas employ dogs and spears to hunt wild pigs and porcupines, the Vaidus trap smaller carnivores, and the Phaseparadhis snare antelopes and birds. Although none of these hunting techniques is so elaborate as to preclude its acquisition by the other groups, each caste shuns the techniques of the others. Faithful pursuit of the subsistence strategy of one's caste is a major tenet of the Hindu religion. This specialization and diversification of subsistence strategies has played an overriding role in moulding Indian society (Gadgil and Malhotra 1983). As an ecologist, I am struck by the profound influence culture has had on the pattern of resource use on the Indian subcontinent, something that could not have been predicted on the basis of Harris's infrastructure alone. by B. G . HALBAR Department of Anthropology, Karnatak University, Dharwad 580003, India. 4 VI 84 Westen has done a real service in moderating the extreme positions of cultural materialism and psychocultural explanation (cultural idealism). Seeking to present a comprehensive critique of the method and theoretical assumptions of Harris's cultural materialism, he does not reject the cultural materialist approach outright. He takes into account the many instances of brilliant explanation of culture in terms of Harris's model as well as pointing to the pitfalls of materialist reductionism. Even Harris should feel gratified by this exercise. By reanalysing and reinterpreting Harris's case material from Kerala, Westen has shown a more reasonable alternative approach that explains the phenomena better. Though many other anthropologists have undertaken the reinterpretation of the work of earlier writers, much more such scrutiny is needed. However, Westen's criticism of Harris's categories of infrastructure, structure, and superstructure as ambiguous, overlapping, and not mutually exclusive is, I think, unfair. Any classification or typology is intrinsically arbitrary, artificial, heuristic, and methodologically determined rather than unambiguous, real at the phenomenal level, or exclusive and exhaustive. The categories depend upon one's purpose, one's viewpoint, and the model in terms of which the observed phenomena are arranged and rearranged. This does not mean that any sort of categorisation is to be accepted as scientific. Rather, scientific categorisation should reflect reality in terms of one's methodological assumptions and conform to accepted rules of logic and thought. There is no perfect way of dissecting social reality into definite, exclusive, and exhaustive categories. by MARVINHARRIS Department of Anthropology, University of Florida, Gainesville, Flu. 32611, U . S . A . 20 vr 84 I regret that other commitments prevent me from responding to Westen's criticisms at length. Almost all of his objections were anticipated in the Rise of Anthropological Theory or in CURRENT ANTHROPOLOGY Cultural Materialism. Suffice it to say at this point that Westen does not know how to think nomothetically about sociocultural phenomena. Almost any other diachronic and synchronic nomothetic sociocultural paradigm besides cultural materialism could have been a target for most of his objections. These objections were long ago disposed of by Alfred Kroeber and Leslie White. What remains is a compilation of transparent sophistries. Example: Westen warns, "If [Harris] can change the way a reader thinks without changing the reader's or the society's material conditions, then the reader . . . has been duped." As a nomothetic research strategy, cultural materialism does not predict the behavior of individuals; it predicts (or retrodicts) the behavior of aggregates of individuals. And in fact, although I have had some degree of success in changing people's minds, the infrastructural and political-economic conditions in the U.S.A. are so overwhelmingly supportive of idealist, eclectic, and obscurantist social science as to leave them in the ascendancy despite their demonstrable inability to develop a coherent corpus of testable theories. Obviously, cultural materialists want as many people on our side as possible, but the material conditions guarantee that for every cultural materialist thrown up by our educational system, twenty Westens will be thrown up against that one. I have always said that the day cultural materialism becomes the dominant paradigm, without infrastructural changes, is the day I will cease to be a cultural materialist. And the day critics like Westen abandon their purely destructive sophistries and offer a coherent and testable corpus of theories for explaining the evolution of cultural differences and similarities is the day I will take their epistemological and theoretical pronouncements seriously by THOMASW. HILL Department of Sociology and Anthropology, University of Northern lowa, Cedar Falls, l o w a 50614, U . S . A . 9 VI 84 Westen's article does a good job of identifying conceptual ambiguities, theoretical gaps, and internal contradictions within Harris's research strategy of cultural materialism. Although the article raises a number of points that deserve further discussion, I will limit my comments to the manner in which Harris handles the relationship between biopsychological states/processes and overt behavior. As Westen notes, although Harris grants "a degree of autonomy" (1979:65) to biopsychological states and processes and talks in passing about feedback between personality configurations and infrastructure (p. 260), the role that biopsychological factors play in producing overt behavior is severely limited (pp. 59-60): Conscious thoughts in the form of plans and itineraries certainly help individuals and groups to find a path through the daily complexities of social life. But these plans and itineraries merely chart the selection of preexisting behavioral "mazeways." . . . [Pllanned actions . . . are never conjured up out of thin air but are drawn from the inventory of recurrent scenes characteristic of that particular culture. Harris appears to be arguing that while biopsychological stateslprocesses, such as plans, do help guide overt behavior, the plans of any one individual are merely taken from the total stock of plans found in the society. The exact manner in which an individual acquires his particular set of plans from the total stock is never made clear; nor, as Westen points out, are the processes by which infrastructural factors determine the content of these plans adequately specified. A further indication of the way Harris treats these issues becomes apparent when he criticizes idealists for maintaining a research interest in rules. Harris opposes this interest because it "runs counter to cultural materialist assumptions about the dependent nature of the entire mental and emic sector" (p. 269). Because humans can formulate, if the need arises, a "rule" to explain an action, Harris mistakenly concludes that V o l . 25 . N o . 5 . December 1984 Westen: CULTURAL MATERIALISM "people have a rule for everything they do" (p. 275). As a result, "every act will be deemed appropriate by at least one actor and possibly many more" (p. 276). The conclusion to be drawn from these observations "is not that people behave in order to conform to rules, but they select or create rules appropriate for their behavior" (p. 275). Nonetheless, Harris concedes that people will sometimes, at least, follow rules if the rules are endorsed by authorities, i.e., people who possess power by virtue of the infrastructure (p. 277). Now this is a curious position for Harris to adopt. To admit that "rules facilitate, motivate, and organize our behavior" (p. 275) is to accept, in spite of Harris's assertions to the contrary, that biopsychological stateslprocesses can become, from one point of view, elements in a causal network that produces behavior. Although it is true that from another point of view a researcher can ask how particular individuals acquired or adopted specific rules, this does not, as Harris seems to believe, invalidate the first viewpoint. The same event may play different causal roles in relation to different sets of background conditions (see Mandelbaum 1977). In discussing these issues, Harris sets up a straw man to attack, presenting idealists as demanding an exclusive focus on cognition. In fact, the research strategies of many "idealists" consider both biopsychological and situational factors as potential causes of overt behavior (see, for example, Goodenough 1963, Wallace 1970). In place of these more balanced views, Harris would have us all adopt his extremely restricted approach. But as Westen correctly concludes, Harris's delineation of cultural materialism actually "renders a comprehensive understanding of behavior impossible." by ALLENJOHNSON Department of Anthropology, University of California, Los Angeles, Calif. 90024, U . S . A . 15 v 84 The difficulty with critiques such as Westen's is that they are too scattershot to offer coherent alternatives to the theories they criticize. Materialists take many shapes, and Harris does not always speak for all of them-penetrating criticism by someone knowledgeable in the field is always welcome. But Westen writes as one who "knows the words but not the music," and he even has some of the words wrong, as in his evident confusion over Harris's use of the phrase "emicl behavioral." Once again we are given the example of the artist (Burling [I9621 had the same difficulty with the opera singer): is he part of the infrastructure or of the superstructure? Why not both? The artist works for a living, and the art gallery and opera house are centers of elite social activity that certainly have materialist consequences. On the other hand, the artist is manipulating symbols and reflecting or transforming images in the superstructure. Of course the distinction between base and superstructure is artificial and open to criticism, just as any typological division of a cultural whole must be. Dividing a cultural unit into economic, political, social, and ideological realms, as anthropologists do every day, is no less fuzzy and arbitrary. Does Westen have inside information on an analytic scheme that does not share these weaknesses? As for Westen's claim that Harris cannot explain why peasants do not overthrow their numerically inferior overlords, had he read Harris's (1959) "The Economy Has No Surplus?" he would have learned that Harris early recognized that landlords and other elites in state society do not only skim off surplus parasitically but simultaneously provide essential managerial services symbiotically. Hence, for many peasants, an ideology that emphasizes loyalty to a landlord makes sense, even when the anthropologist is inclined to interpret the class relation as exploitative. Here as elsewhere, ideology and material reality are related through the continuous feedbacks that Harris describes, and it is misleading to imply that such an ideology arises arbitrarily as "pure culture," without material antecedents. The most promising line of argument in Westen's paper is never developed: Harris's theory is characterized by a lack of development in its psychological aspect. I infer from several statements that Westen may embrace a variety of humanistic psychology, "ego psychology," or "self psychology" from psychoanalysis. Each of these theories has an underlying theory of needs that is structurally similar to Harris's more . of them, including Harbiological one (Johnson 1 9 8 2 ~ )None ris's, assumes that "social needs" exist apart from individuals, an error Westen commits in this paper. For all of them, human behavior and culture can in large measure be accounted for as attempts to meet basic needs. Broadening the theory of needs to include such psychological concepts as identity would be helpful, and Westen might have made better use of his (and our) time had he shown how this could be done. This leads to the heart of the matter. By showing that materialism is limited in its explanatory power, Westen hopes to discredit it altogether. He does not confront the reality that any explanation is reductionistic (Johnson 19826). He seems a t times to prefer a psychological determinism and at other times a cultural determinism to Harris's materialist determinism. But each of these, taken alone, will be found to provide only a partial explanation of any given aspect of human behavior or belief. Thus it seems Westen has to do one of two things that he has not done in this paper: either accept Harris's (1979:75-76) challenge to come up with a deterministic theory that explains more than his does or else propose a nondeterministic theory that is of any use at all. I suspect he would find the former quite difficult and the latter next to impossible, for the hard truth is that explanation is reductionistic, and the opposite of reductionism in most cases is confusion, not enlightenment. by ROGERJOSEPH Department of Anthropology, California State University, Fullerton, Calif. 92634, U.S.A. 14 v 84 The degree of seriousness one accords Westen's thesis depends on how impressed one is with Harris's argument. Harris made important contributions to anthropological theory, none of which are cited here. His more recent work fails to differentiate between the construction of a model and the system being modeled, and it is not made intellectually richer by the simplistic commentary of advocates such as Magnarella or detractors such as Westen. Cultural materialism lacks the vigor to reconstitute itself as a moral, economic, or ideological statement. Elevating "bovicide" to the status of theoretical cause merely suggests the bourgeois gap between theoretical anthropologists and disenfranchised peasants. That Westen believes the Iranian revolution is Islamic and not economic must raise a few eyebrows among area specialists, as must his calling Libya a Marxist regime. Westen believes that a political economy must "begin" someplace. The political economy of any society is both a model for its historical development and that historical development itself. The historical part of this business is a miasma of counterinfluences; the modeling portion includes production, distribution, and consumption. One can conceive the model as circular, a process in which each aspect implicates the rest. The process can be modeled in any one of a variety of waysbut each intellectual act on the model (as opposed to the historical events) simply causes a relationship to appear. The appearance of a relationship in the model signifies not historical interconnection but an interaction between the model and our thinking about the model. The model reflects neither nature nor historical development but rather our reflections and interpretations of the incidents we assign the role of active intellectual agencies. The model, always exogenous to history and to the activities of eating, laboring, or whatever, is a flow diagram of our cognitive processes in apprehending this history and those activities. Most of this is very difficult to think about, and Marx often got it wrong. Harris inevitably muddles it further. There are a number of anthropologists who understand it better: Wolf, Sahlins, Geertz, Bourrillard, and Bourdieu. Having failed to do a bit of preliminary reading, Westen might have examined Godelier more closely. by PAULJ. MAGNARELLA Department of Anthropology, University of Florida, Gainesville, Flu. 32611, U.S.A. 30 v 84 Westen's provocative article deals in significant part with the problem of determining the roles played by ideas in sociocultural systems. According to Harris's principle of infrastructural determinism, "the etic behavioral modes of production and reproduction probabilistically determine the etic behavioral domestic and political economy, which in turn probabilistically determine the behavioral and mental emic superstructures" (Harris 1979:55-56). But "determine," Harris explains, does not mean "cause the origin of." He writes (p. 59): In asserting the primacy of the behavioral infrastructure over the mental and emic superstructure, cultural materialism is not addressing the question of how technological inventions and other kinds of creative innovations originate in individuals but rather how such innovations come to assume a material social existence and how they come to exert an influence on social reproduction and social production. Thoughts in the minds of geniuses like Hero of Alexandria, who invented the steam turbine in the third century, or Leonardo d a Vinci, who invented the helicopter in the sixteenth century, cannot assume a material social existence unless appropriate material conditions for their social acceptance and use are also present. According to this statement, the nature of the ideas created is not dependent on the infrastructure and structure in place, but the two do determine (probabilistically) which ideas will take hold by being acted upon. Farther along in the same paragraph, however, Harris seems to revert to a deterministic position concerning idea creation. He argues that "the conclusion seems inescapable that when the infrastructural conditions are ripe, the appropriate thoughts will occur, not once but again and again." Hence, an ambiguity exists: what exactly is the theoretical status of mental phenomena in the cultural materialist research strategy? Ideas are of different kinds and weights. Some arise repeatedly under diverse sets of infrastructural-structural conditions; many diverse and even contradictory ideas arise within the same infrastructural-structural context. There are certain kinds of ideas (e.g., religious, aesthetic, recreational) that people can act on in ways that do not alter the existing infrastructural-structural arrangement. There are other ideas which appear to be closely related to the reigning infrastructure (e.g., new invention ideas that build upon previous inventions), and once people act upon them they alter the infrastructure. We could say that these ideas are both determined and deterministic, although as Sahlins (1976) convincingly argues, the relationships among ideas, behavior, and the resulting material products are much more complex. The above examples suffice to support the position that different kinds of ideas should be assigned different theoretical statuses in the cultural materialist research paradigm. This position is not foreign to Harris. He alludes to it in Cultural Materialism. After describing the etic behavioral infrastructure, structure, and superstructure, he states that each of these levels has associated with it a set of mental and emic components. But instead of distinguishing them according to the strength of their relationship to specific etic behavioral C U R R E N T ANTHROPOLOGY components, he "lump[s] them together and designatels] them in their entirety as the mental and emic superstructure" (p. 54). Creating this fourth level distorts the structure of the cultural materialist paradigm and the relationships of its constituent parts. The result is a large residual category-mental and emic superstructure-whose contents are misplaced. A rectification of this situation should clarify some of the ambiguities in this research strategy. by MICHAELPAINTER Institute for Development Anthropology, 99 Collier S t . , Suite 302, P.O. B o x 818, Binghamton, N . Y . 13902, U . S . A . 25 V 84 Westen voices criticisms of cultural materialism that have been made in other contexts by a number of authors. The contribution of his article is that it attempts to draw together arguments that have been presented in diverse quarters into a comprehensive critique. The major weakness is that his review of the relevant discussion is not complete and his treatment of the criticisms upon which he bases his own argument is sometimes shallow. As a result, while his argument that cultural materialism is hopelessly flawed as a paradigm for anthropological inquiry is convincing, he does not say as much as he might about the elements of a more adequate materialist paradigm. I shall mention two points that I was disappointed not to see discussed in greater detail and that I feel must be central elements of any paradigm that seeks to offer an alternative to cultural materialism. Westen properly discusses the conceptual difficulties associated with Harris's idiosyncratic use of the notions of infrastructure and superstructure. However, he does not explain that a major source of these difficulties is Harris's inadequate appreciation of history as a force shaping the organization and functioning of human societies. Although cultural materialism may draw upon history to explain individual cases, as a general paradigm it is ahistorical. As a result, Harris must attempt to tell us for all times and places what constitutes infrastructure and superstructure and then perform conceptual gymnastics to explain why exceptions or errors in his typology really are not exceptions or errors at all. As Godelier (1977) points out, the institutional location of the infrastructure of a society is the result of historical process and must be determined empirically. Harris's ahistoricism also leads him to differentiate infrastructure and superstructure as if they pertained to different institutions rather than treating them as distinct roles within a single institution. Because Harris does not assign sufficient importance to history, he overlooks the fact that his own society is one of the very few in which the functional distinctions between infrastructure and superstructure correspond to institutional distinctions and that this coincidence of role and institution arises from the historical development of capitalism (Godelier 1978:764-65). Secondly, Westen points out that if cognitive constructs, such as the concept of science, are the result of infrastructural factors as Harris contends, then all cognitive constructs are culturally and historically specific, and there can be no truly etic explanations. This epistemological inconsistency arises from Harris's view that a true science of culture should be based upon deductive reasoning. He thus treats the consummately emic categories of Western science as if they were etic concepts from which cross-culturally generalizable interpretations of emic components of behavioral systems might be deduced. He ignores the fact that in those few areas of inquiry where well-developed bodies of etic concepts exist-such as phonology or genealogical kinship-these have been derived inductively, through painstaking comparison of numerous emic analyses (Goodenough 1970:104-30; Harris 1976). Because they are derived inductively, etic concepts are subject to Vol. 25 . N o . 5 . December 1984 Westen: CULTURAL MATERIALISM constant review and refinement as new emic data are introduced, even as they are being utilized to interpret the new emic data. Thus, a new paradigm for materialist explanations of culture must differ from cultural materialism in at least two respects. First, it must accord a more central role to history as a determiner of what constitutes infrastructure and superstructure in human societies. Secondly, it must be built upon an epistemological foundation that includes an explicit recognition of the interrelationship between deductive and inductive analyses in fruitfully studying human behavior. Cultural materialism is flawed because it apes the folly of neoclassical economics, treating the historical and cultural particularities of Western societies as if they were natural laws. by ANDREWP. VAYDA Graduate Program i n Ecological Anthropology, University of Indonesia, Rawamangun, Jakarta, Indonesia. 16 v 84 While having only minor disagreements with his specific criticisms of Harris's cultural materialism, I find Westen's statement of his own "more fruitful approach" rather unclear, even regarding what it is an approach to. Although he refers to paradigms for explaining culture, his final section gives examples of the explanation of such specific actions as cattle killing and the initiation of pig festivals. If his approach involves something very different from what I have elsewhere described as attention to the "ways in which responses to changing circumstances are made by particular human beings, acting either together or separately and making use of whatever technological, organizational, and cultural means are available to them" (Vayda 1983: 270), he needs to indicate better what that something is and how it can be used not, as in his examples, simply to show cultural influences on behavior but rather to explain culture. by DREWWESTEN A n n Arbor, iMich., U . S . A . 27 VII 84 Were I an idealist, I would certainly be appreciative of the comments by Harris and Diener, because both nicely confirm the theory that our values and our constructions of reality in large measure determine our responses to it. In more psychological terms, people process information through schemas, and the responses evoked by situations are influenced by both their motivations and the categories they use to understand reality. Harris, for example, accuses me of "purely destructive sophistries." In contrast, Halbar suggests that Harris should be pleased by this attempt to moderate extremes, and I referred in the article to Harris's brilliant contributions, arguing not that his method should be abandoned but that it should be subsumed into a more inclusive approach. I will not speculate on motivations, but it is clear that Harris's response rests upon his cognitive construal of my arguments and intentions and that others who construe them differently respond otherwise. Perhaps this cognitive influence on behavior is not so unusual. Similarly, Diener views the world (or at least the intellectual world) as consisting of those with white hats (those whose views agree with his) and those with black hats (everyone else). This permits him to categorize as "idealist" an article which explicitly states that human beings are motivated by both values and the needs underlying production and reproduction and to lump together a critique and its target. If Diener's behavior can reflect his cognitive and evaluative construction of a situation, if this construction can differ from that of others reading the same article, and if his response is correspondingly different from theirs, does this not demonstrate the role of ideas and ideals in influencing human action? In response to my argument about changing minds without changing "infrastructures," Harris contends that his view is meant to explain the behavior of aggregates of individuals, not individuals. The problem, as I argued in the article, is that Harris cannot focus on aggregates without settling on a unit of analysis and that this cannot be done from a cultural materialist perspective. One cannot simply view culture as a reflection of material interests because culture defines interests. Cultural norms or values dictate which unit is given primacy in decisions, i.e., which group's "interests" are to be optimized. In some cultures it is the lineage, in others the nuclear family, and in the contemporary West many people make decisions strictly on the basis of self-interest. One cannot understand these different emphases solely on the basis of demo-technoeconomic factors, though certainly these have an important causal impact. Nor can one try to divorce a theory of the behavior of aggregates of individuals from a theory of individual behavior. If individual actors pursue ideals or identity as well as material self-interest, then an aggregate of individuals will not somehow be seen to pursue only the latter. Cultural materialism lacks a theory of the relation between individual action and infrastructural determinants that can account for the convergence of aggregate behavior and material demands. I agree with Harris that many of the criticisms of cultural materialism could equally be applied to other approaches, though I do not see how that exonerates his own system. Diener's comments exemplify the illegitimate intrusion of polemics into anthropological discourse. While he and I essentially agree in our political views (I, too, would like to see the Nicaraguans free of American imperialism), I think there is a difference between moral-political ideology and scientific doctrine. His argument, defending Marx, contains several of the errors Marx himself tried to expose. First, Marx criticized the "bourgeois" economists for dressing up values as scientific laws, as in their assumptions about free trade and the classical economic theory of value. Yet what is the belief that people should not be exploited, or that the current world distribution of economic resources should be changed, or that supply and demand should not determine distribution of goods, if not a moral-political philosophy? Secondly, we owe to a great extent to Marx the recognition that all thought is socially conditioned. I have not discovered the Olympus from which one determines that everyone else's beliefs are ideology while his own are Truth. That strikes me as a peculiarly elitist doctrine. Diener and Harris have emerged from the same economic system in the same historical period, and it is difficult to see how, for example, Diener can claim as etic and objective the view that Harris's materialism is animism in disguise. Thirdly, Diener's criticisms of Harris and myself invoke the same bourgeois psychology that Marx both criticized and unwittingly accepted, the view that people act solely on the basis of their economic self-interest. Diener explains the views of those who disagree with Marxist analysis as based on their "market segment"; we are all lackeys either of monopoly capital or of the petite bourgeoisie. Surely this is a shallow, "bourgeois" view of human motivation and one that certainly cannot explain Diener's own ideology, given that his class interests are probably quite similar to those of many other anthropologists. Freud recognized years ago that neurosis is not something that everyone else has, and perhaps it would be worthwhile for Diener and those who share his views to reflect on the possibility that ideology is not something endemic to everyone but oneself. Indeed, Diener's ideology has blinded him to much of what I am arguing. For example, "Westen describes human choices as 'hedonistically selected'; Harris also sees people as selfish." I spoke of hedonic selection, by which I meant selection on the basis of emotional consequences. I explicitly argued, against Harris and many rational decision models, that people respond emotionally to threats to the welfare of many people other than themselves. This has been a stumbling block for Marxist theory, which accepts the rationalist psychology inherited from liberal philosophy and ignores the extent to which people work and die for collectivities that may exploit them economically, in the view of some outside observer, but with which they identify or to which they are emotionally attached. As for Diener's assumption that all anthropologists (with the exception of himself, of course) are simply capitalist agents scouting for booty, there are some of us who are not yet convinced that we know everything there is to know about cultural evolution or history and therefore consider studying cultures in different parts of the world and in the process of change a useful enterprise. I am not denying that anthropology can be and has been used for political-economic purposes; rather, I am simply suggesting that deductive reasoning from emotion-laden premises cannot substitute for observation. I agree with Johnson that psychological, cultural, and material determinism are all inadequate. That is why I advocate the abandonment of single-cause theories like Harris's and a move toward a conception of humans as creatures motivated by symbolically mediated desires and needs such as attachment and identity as well as by the need for nutrients. In this limited space, I could not systematically suggest directions a broader model might take; that is why I chose to focus on a specific case. I also agree with Johnson that all explanation is reductionistic. The question is not whether we reduce one thing to another but what our rules for that reduction are. My argument is simply that Harris's rules are inadequate. I appreciate Ehrenreich's attempt to answer arguments rather than sling epithets, though I should mention that my intent was less to shoot a poison-tipped dart at Harris, for whom I have a great deal of respect, than to contain a materialist bull's-eye which has tried to take over the whole dartboard. Ehrenreich points out that Harris only gives research priority to infrastructural explanations and does not exclude causality running in the other direction. My argument, however, is that his having proposed no theory to explain reciprocal links renders his approach a partial one. I , too, believe that material forces are often the most important causal factor, but I disagree that that represents the comprehensive statement about culture Harris claims to provide. Ehrenreich accuses me of failing to understand science, arguing that "science is sciencing" and therefore unambiguously superstructural. I doubt that many philosophers of science would agree that science includes neither bodies of knowledge nor evolving tools or technologies. Popper did not view the process of conjecture and refutation as contentless, nor does Kuhn see paradigms as independent of bodies of knowledge. If science is really sciencing, then why could one not similarly say that production is really "producing," i.e., a method in the mind of producers, or that reproduction is really "reproducing," i.e., a feeling in the minds of sexual partners? In that case, it is difficult to see what, other than physical artifacts, would not be superstructural. A number of commentators suggest that all categorization is flawed and therefore my criticism of cultural materialism for failing to provide firm boundaries between infrastructure, structure, and superstructure is misplaced. I agree that every conceptual system is in some sense arbitrary and certainly imperfect. The question is whether it is useful. I argue that in Harris's case it is not, as in the example of the blouse or the place of psychological needs. I agree with Ehrenreich that the act of persuading others of one's ideas is not folly and that, if one has a political program in mind, the "issue is not merely getting people t o believe but CURRENT ANTHROPOLOGY getting t h e m to plan and act w i t h these principles a n d perceptions i n m i n d . " B u t I believe this because I believe i n t h e causal efficacy of ideas a n d ideals a s well a s material forces. H o w c a n a materialist hold this v i e w regarding his o w n actions a n d his o w n society w i t h o u t extending i t t o other cultures? Finally, Ehrenreich suggests t h a t Harris's critics a r e "often grappling w i t h questions t h a t a r e substantially different i n nature a n d purpose" f r o m Harris's. I d o u b t it. H a r r i s h a s proposed a theory of t h e structure, d y n a m i c s , a n d evolution of culture a n d i n s o d o i n g h a s discussed s u c h issues a s t h e relation between material forces a n d beliefs, between ideals a n d t h e culture o n t h e g r o u n d , a n d between intrasocietal conflict a n d culture. Surely these m u s t b e s o m e of t h e s a m e questions w i t h which any theory of culture m u s t grapple. References Cited ADAMS,R. N . 1981. Natural selection, energetics, and "cultural materialism." CURRENT ANTHROPOLOGY 22:603-24. BANDURA, A. 1977. Social learning theory. Englewood Cliffs: Prentice Hall. BARTH, F . 1967. On the study of social change. American Anthropologist 69:661-69. BEIDELMAN, T . 1982. Review of: Cultural materialism, by M. Harris (New York: Random House, 1979). American Journal of Sociology 87:1245-47. BOEHM,C. 1982. A fresh outlook on cultural selection. American Anthropologist 82:105-25 BRITAN,G . , and B. DENICH.1976. Environment and choice in rapid social change American Ethnologist 3:55-72. BOWLBY,J . Attachment and loss. Vol. 1. Attachment. New York: Basic Books. BURLING,ROBBINS.1962. Maximization theories and the study of economic anthropology. American Anthropologist 64:802-21. [AJ] CHIBNIK,MICHAEL.1981. The evolution of cultural rules. Journal of Anthropological Research 37:256-68. [MC] COOK,S. 1973. "Economic anthropology: Problems in theory, method, and analysis," in Handbook of social and cultural anthropology. Edited by J . J . Honigmann, pp. 795-860. Chicago: Aldine. DICKEMAN,M . 1975. Demographic consequences of infanticide in man. Annual Review of Ecology and Systematics 6:107-37. DURKHEIM,E . 1938. The rules of sociological method. New York: Free Press. EHRENREICH, JEFFREY.1981. Comment on: Natural selection, energetics, and "cultural materialism," by R. N. Adams. CURRENT ANTHROPOLOGY 22:614-16. UE] ERIKSON,E . 1968. Identity: Youth and crisis. New York: Norton. FISHER,LAWRENCEE . , and OSWALDWERNER.1978. Explaining explanation: Tension in American anthropology. Journal of Anthropological Research 34:194-218. [MC] FRANKL,V. 1959. Man's search for meaning. New York: Pocket Books. FREUD,S. 1950 (1913). Totem and taboo. New York: Basic Books. FRIEDMAN, J . 1974. Marxism, structuralism, and vulgar materialism. Man 9:444-69. GADGIL,M . , and K. C. MALHOTRA.1983. Adaptive significance of the Indian caste system: An ecological perspective. Annals of Human Biology 10:465-78. [MG, KCM] GEERTZ,C. 1973. The interpretation of cultures. New York: Basic Books. GODELIER,M . 1974. Infrastructures, societies, and history. CURRENT ANTHROPOLOGY 19:763-7 1. . 1977a. Perspectives i n Marxist anthropology. Cambridge: Cambridge University Press. -. 19776. "Anthropology and economics," in Perspectives i n Marxist anthropology, by M . Godelier, pp. 15-62. Cambridge: Cambridge University Press. [MP] -. 1978. Infrastructures, societies, and history. CURRENT ANTHROPOLOGY 19:763-71. [MP] WARD.1963. Cooperation i n change. New York: RusGOODENOUGH, sell Sage Foundation. [TWH] , 1979. Description and comparison i n cultural anthropology. Chicago: Aldine. [MP] HARRIS,MARVIN.1959. The economy has no surplus? American Anthropologist 61:185-99. [AJ] -. 1966. The cultural ecology of India's sacred cattle. CURRENT ANTHROPOLOGY 7:s 1-59. -. 1968. The rise of anthropological theory. New York: Crowell. [PD, JEI Vol. 25 . No. 5 . December 1984 Westen: CULTURAL MATERIALISM -. 1976. History and significance of the emicletic distinction. Annual Reviews i n Anthropology 5:329-50. [MP] -. 1979. Cultural materialism: The struggle for a science of culture. New York: Random House. -. 1982. Reply to P. J. Magnarella. American Anthropologist 84:142-45. HARRIS,M . , and E . Ross. 1978. The origins of the U.S. preference for beef. Psychology Today, October, pp. 88-94. JOHNSON,ALLEN. 1982a. "Nonreductionistic cultural ecology," in Culture and ecology: Eclectic perspectives. Edited by John G. Kennedy and Robert B. Edgerton. American Anthopological Association Special Publication 15. Washington, D.C. [AJ] -. 1982b. Reductionism in cultural ecology: The Amazon case. CURRENT ANTHROPOLOGY 23:413-28. [AJ] LAZARUS, R. 1981. "The stress and coping paradigm," in Models for clinical psychopathology. Edited by C. Eisdorfer, D. Cohen, and A. Kleinman. New York: Spectrum. LEACH,E . R. 1954. Political systems of Highland Burma: A study of Kachin social structure. London: Humanities Press. LEWIS, JOHN. 1972. The Marxism of Marx. London: Lawrence and Wishart. [PD] LEWIS, 0. 1966. The culture of poverty. Scientific Anterican 215(4):19-25. MCCLELLAND, D. 1961. The achieving society. Princeton: Van Nostrand. C. B. 1962. The political theory of possessive individMACPHERSON, ualism: Hobbes to Locke. Oxford: Oxford University Press. MAGNARELLA, P. J . 1982. Cultural materialism and the problem of probabilities. American Anthropologist 84: 138-45. MAURICE.1977. The anatomy of historical knowledge. MANDELBAUM, Baltimore: Johns Hopkins University Press. [TWH] MARX, KARL. 1972 (1845). "Theses on Feuerbach," in The MarxEngels reader. Edited by Robert C. Tucker, pp. 107-9. New York: Norton. UE] -. 1972 (1852). "The Eighteenth Brumaire of Louis Bonaparte," in The Marx-Engels reader. Edited by Robert C. Tucker, pp. 436525. New York: Norton. UE] -. 1972 (1873). "From the afterword to the second German edition of Capital," in The Marx-Engels reader. Edited by Robert C. Tucker, pp. 195-98. New York: Norton. WE] MILLER,ROBERTA BALSTAD.1984. A look at the F Y 1985 budget. . . and a proposal for the future. Anthropology Newsletter 25(4):8. [PD] PAUL,ROBERT,and PAULRABINOW. 1976. Bourgeois rationalism revived. Dialectical Anthropology 1:12 1-43. [MC] PELTO,PERTTIJ., and GRETELH . PELTO. 1973. "Ethnography: The fieldwork enterprise," in Handbook of social and cultural anthropology. Edited by John J . Honigmann, pp. 241-88. Chicago: Rand McNally. [PD] -. 1975. Intra-cultural diversity: Some theoretical issues. Anterican Ethnologist 2:l-18. RAPPAPORT, R. 1968. Pigs for the ancestors: Ritual i n the ecology of a New Guinea people. New Haven: Yale University Press. . 1979. Ecology, meaning, and religion. Richmond, Calif.: North Atlantic Books. ROSE, HILARY,and STEVENROSE. 1976. "The problematic inheritance: Marx and Engels on the natural sciences," in The political economy of science: Ideology ofiin the natural sciences, pp. 1-26. London: Macmillan [PD] SAHLINS,MARSHALL.1976. Culture and practical reason. Chicago: University of Chicago Press. [MC, PJM] -. 1978. Culture as protein and profit. New York Review of Books 25(17):45-53. SERVICE,ELMAN. 1969. The prime mover of cultural evolution. Southwestern Journal of Anthropology 24:396-409. [MC] SKOCPOL, T . 1979. States and social revolutions: A comparative analysis of France, Russia, and China. Cambridge: Cambridge University Press. TURNER,V. 1969. The ritual process. Chicago: Aldine. VAYDA,ANDREWP. 1983. Progressive contextualization: Methods for research in human ecology. Human Ecology 11:265-81. [APV] WALLACE, ANTHONY F . C. 1970. Culture and personality. New York: Random House. [TWH] -. 1980. Review of: Cultural ntaterialisnt, by Marvin Harris (New York: Random House, 1979). American Anthropolofist 82:423-26. WESTEN. D . n.d. Self and society: Narcissism, collectivism, and the development of morals. Cambridge: Cambridge University Press. In press. WHITE, LESLIEA. 1949. The science of culture: A study of man and civilization. New York: Grove Press. UE]