Department of French
Los Angeles CA 90095-1550

Although he is not primarily known as a religious scholar, few thinkers have influenced our conception of the sacred as much as Emile Durkheim. In particular, Durkheim should be considered the principal theoretical ancestor of René Girard’s notion of the sacred and subsequently of that embodied in the originary thinking of Generative Anthropology. Whence my interest in defining–in appropriating, if you like–Durkheim’s anthropological legacy.

With the passing of the millennium, and of the twentieth century’s political millennialism, the sacred, even in the strict, traditional sense of the term, seems much healthier than nearly anyone at the turn of the previous century expected it to be. It is also increasingly on the intellectual agenda on both ends of campus–at UCLA, the North and the South, the respective domains of the Humanities and the Sciences. With respect to the Humanities, this colloquium (June 2000) on “Transforming the Sacred” and its parent Consortium Seminar on “Sacred and Profane” provides one example; a conference scheduled for Fall 2000 on “The Pious and the Profane” at the University of Washington is another. One effect of the globalization of the study of culture has been to make humanists habituated to the Enlightenment dichotomy between reason and obscurantism aware of the salience of the sacred in other cultures and perhaps more tolerant, or at least more intellectually curious, of its importance in our own.

Yet I would venture to say that the general level at which the sacred is theorized among humanist intellectuals today is far lower than that obtaining in Durkheim’s day, which was also that of Max Müller, James Frazer, Edward Tylor, and W. Robertson Smith, not to speak of Durkheim’s own students. The attention Durkheim’s work on the sacred has received in recent years–in a colloquium on Durkheim and religion entitled “Vive Durkheim!” on the UCLA campus in 1997, in Vincent Pecora’s talk on “Modernity, Religion, and the Social Imagination” in October 1999 to kick off the Mellon lecture series on “Sacred and Profane,” and in a steady stream of scholarly works including a volume of essays on the Elementary Forms published by Routledge in 1998, not to speak of a number of active web sites–suggests an awareness of this lacuna and an increasing desire to fill it. As the notion of the sacred moves from the outlands to the periphery and thence toward the center of humanistic studies, the search for theoretical models leads in a privileged fashion to Durkheim as the last major scientific theorist of the sacred–a qualification that separates him from the provocative but dilettantish writers of the Collège de Sociologie such as Georges Bataille and Roger Caillois. As for Claude Lévi-Strauss, the successor to Durkheim’s mantle in French anthropology, what excludes him from consideration is not that he dismisses Durkheim’s association of the sacred with the social, but that he evacuates the difference between them by subsuming them under the neutral category of “structure.” Lévi-Strauss qualifies the same structure found in the relations among gods, the patterns of tattoos, the village topography, or the distribution of marriage exchanges as “good to think”–and as resolving tensions within the group that we are tempted to qualify as mimetic. What he fails to consider is the specific dynamic of the sacred in the discovery/invention of the “good to think” as a means for achieving social cohesion.


On the South campus, the realm of science, the return to the sacred is real but more hesitant, and the need to redefine Durkheim’s legacy therefore all the more urgent. What I believe to be of greatest significance in this area is the still controversial movement of evolutionary psychology. The ambition of this movement, as canonically expressed in Leda Cosmides and John Tooby’s introduction to their 1992 collection The Adapted Mind, is to recapture the study of the human–and animal–mind from the hegemony of what the authors call the Standard Social Science Model or SSSM. In this model, the mind is conceived as a tabula rasa, an all-purpose information-processing organ whose content is entirely supplied by the environment (nihil in intellectu quod non prius fuerit in sensu). Rejecting this model as both improbable in itself and as conducive to impressionistic descriptions of cultural phenomena rather than the formulation of testable hypotheses, the evolutionary psychologists propose in its place a mind composed of special-purpose modules pre-calibrated by evolutionary adaptation, in the Pleistocene era when homo sapiens emerged, for sensitivity to significant aspects of the human environment, including the social environment.

Durkheim and the sacred enter into this polemic in two places.

First, the human universality of the sacred and its practices give it a respectability from the evolutionary perspective that it did not have in latter-day versions of the SSSM. We have heard a great deal, following a notorious 1979 paper by Stephen Gould and Richard Lewontin, about “spandrels” or “exaptations,” the contingent consequences of evolutionary adaptations, themselves unselected by evolution. Although the obvious importance of language to humanity gives this position an air of deliberate perversity, many Chomskian linguists, and Chomsky himself at certain moments, seem to have decided that the communicative aspect of human language, as opposed to its cognitive function, is such an exaptation. If human language, and the culture that depends on it, however adaptive they later proved to be, can be shown to be mere accidents of evolution, then, the implicit argument goes, we have no reason to feel superior to (or “fitter” than) other species; as Gould likes to say, “there is no progress in evolution.”

Unfortunately, good science is not made with good–or, rather, falsely modest–sentiments. Evolutionary psychology, I think quite logically, considers universal, significant human traits as specific adaptations absent proof to the contrary. It suspects, again I think quite logically, that the enthusiastic adoption of Gould’s para-evolutionary exaptations by “soft” scientists, not to speak of humanists, is a stalking-horse for the fashionable relativism reflected in the dominant position of the SSSM. Thus, in its attempt to understand the human mind in its emergent cultural context, evolutionary psychology considers the sacred along with every other universal feature of culture as a primary candidate for the status of adaptation. A typical example is Pascal Boyer’s 1994 article “Cognitive constraints on cultural representations: Natural ontologies and religious ideas.” Boyer’s thesis is that when “counter-intuitive” religious beliefs are maintained in conjunction with “intuitive” or rational beliefs about natural phenomena they provide a “cognitive optimum” that attracts and maintains our attention. Whatever the limitations of Boyer’s cognitivism, I read here a faint but real echo of Durkheim’s central idea that sacred beliefs are always rational in reference, not to the natural, but to the social world.


By the logic of evolutionary psychology, religion, as a universal human trait, must be treated as an adaptation on an equal footing with language. Yet there is little likelihood that we will discover the existence in the brain of a “religion module” or “religion acquisition device” to parallel the language-related structures situated in the Broca and Wernicke areas of the brain, not to speak of a “religion organ” comparable to the highly specialized human vocal apparatus. Precisely because the relation between religion and language–a subject virtually abandoned since the days of Max Müller in the second half of the 19th century–is not directly accessible to the hypotheses of biological evolution, it is a crucial subject for evolutionary thinking in a broader sense. Thus it is no coincidence that the relation between religion and language is both the focal point of Generative Anthropology and the point on which its debt to Durkheim is most apparent. I will return to this point below.

Durkheim’s second and more visible role in the evolutionary polemic demonstrates even more directly than the first the necessity not merely of preserving the Durkheimian sacred but of fleshing out its anthropological abstraction with a concrete and, I would specify, evenemential or “scenic” mediation between the individual and society. For, ironically enough, from the perspective of evolutionary psychology, Durkheim is the arch-villain, the principal initiator of the SSSM. In the introductory chapter of The Adapted Mind, the epigraph to the section devoted to the SSSM is a paragraph from The Rules of the Sociological Method, described as “perhaps [this model’s] most famous early expression.” The authors italicize within this paragraph the following passage:

. . . but [the] individual natures [of the members of society] are merely the indeterminate material that the social factor molds and transforms. Their contribution consists exclusively in very general attitudes, in vague and consequently plastic predispositions . . . (24-25)

Durkheim’s notion of “society” as determining the cognitive operations of its members makes the individual mind wholly dependent on its social environment. However far Durkheim’s unitary and functional notion of the social is from that, descriptive and pluralistic, of Clifford Geertz (archly noted by Cosmides and Tooby for his “literary ability to express the humanly familiar and intelligible as the exotic”), Durkheim and Geertz hold in common the view of the individual human mind as a blank slate dependent for all its categories of thought on its social environment, a view alleged as the foundational principle of the SSSM.

But if evolutionary psychology is justified in denouncing as unscientific anthropologies such as Geertz’s that disdain the attempt to formulate general hypotheses concerning the human, this is hardly a reproach one could make of Durkheim. However excessive Durkheim’s reliance on the “social factor” may be, this “factor” is never, as in Geertz, a pretext for culture-dependent “thick description,” but always an element of a hypothetical model that seeks to derive the empirical diversity of the present from originary, fundamental characteristics of human communal organization. To quote from the first lines of Durkheim’s essay on Incest:

In order to understand a practice or an institution, . . . it is necessary to trace it as nearly as possible to its origin; for between the form it now takes and what it has been, there is a rigorous relationship. (13)

But beyond claiming that the evolutionary-psychological critique of Durkheim throws out, as the Marxists used to say, the baby with the bathwater, I believe that the weaknesses this critique points out are remediable through a reformulation of Durkheimian thought precisely, but more radically, in the sense of the passage just quoted: by tracing the ensemble of social practices and institutions to their origin. If what is lacking in Durkheim is a concrete articulation between the individual mind and the community that supplies that mind with its language and its particular understanding of the sacred, the central aim of Generative Anthropology is to supply such an articulation.


If Durkheim is increasingly remembered as the author of the Elementary Forms of the Religious Life rather than the Division of Labor or even Suicide, it is because his work on religion grounds the otherwise diffuse notion of the “social” in a set of collective representations that embody a given society’s self-consciousness. Durkheim’s insistence that even the most apparently irrational religious ideas correspond to real needs of the social order makes use of an argument not unlike that of the evolutionists: if all societies invest a great deal of mental energy in religious ideas that convey no useful information about the natural world, these ideas must be of value, not in some nebulous “spiritual” sense, but for the very survival of the social order.

But this reduction of religion to “ideas” is the sign of a problem. For Durkheim, religious representations are a projection of a society’s ideal vision of itself; to share these representations is to internalize the beliefs of this society, beliefs that could in principle be formulated as a set of propositions. Durkheim is aware that the most salient aspect of tribal religion is ritual ceremony, and Book 3 of Elementary Forms includes descriptions–at second-hand–of such ceremonies. In Durkheim’s theoretical model, religious practice is a means to reinforce within the participants’ minds the values of society as a whole over the selfish concerns of everyday life:

On feast days . . . their thoughts are centered upon their common beliefs, their common traditions, the memory of their great ancestors, the collective ideal of which they are the incarnation; in a word, upon social things. . . . So it is society that is in the foreground of every consciousness; it dominates and directs all conduct; this is equivalent to saying that it is more living and active, and consequently more real, than in profane times. . . . The individual soul is regenerated too, by being dipped again in the source from which its life comes; consequently it feels itself stronger, more fully master of itself, less dependent upon physical necessities. (390-91)

Even when his description goes beyond “common beliefs” to “common traditions” and the “collective ideal,” the sacred remains for Durkheim a set of representations that resist what he calls “the antagonistic tendencies aroused and supported by the necessities of the daily struggle.” Durkheim conceives these “antagonistic tendencies” as entropic forces that threaten to sap the vigor of sacred representations, rather than, as I believe they must be conceived, as posing the very critical danger to human survival that the sacred and its representations came into being to defer.

So long as we understand sacred representations as the equivalent of logical propositions, we risk falling into a cultural relativism that misconstrues the social as permitting any belief at all. Because someone who can learn one proposition can just as well learn another, the basis for the choice can only lie in the social group that inculcates it rather than in the individual mind itself. Durkheim certainly understands religious beliefs as selected for by social evolution if not by human evolution stricto sensu. But even if all societies shared a single set of beliefs, the abstract relationship of the proposition of religious belief to the individual mind still leaves the latter an undefined tabula rasa.

For Durkheim the educator, religion is analogous to a school in which collective representations are the subject-matter and religious rites, pedagogical techniques to insure their retention. The limit of this analogy is situated precisely at the juncture of the individual and society or “culture” that evolutionary psychology seeks to investigate. Although ritual may plausibly be explained as a technique for regenerating the beliefs of the clan, to seek the common origin of the ritual and the beliefs leads one to formulate a hypothesis concerning the genesis of human society and culture, if not of “the human” itself. For if ritual came first, what beliefs would it transmit? If beliefs came first, how were they shared and affirmed? Once we begin to follow this line of reasoning, we can no longer accept as fundamental, or, in a terminology that I prefer, as minimal, Durkheim’s conception of the sacred as a projection of the social ideal.


A text that allows us a particularly sharp look at both the strengths and the weaknesses of Durkheim’s sociological epistemology is Primitive Classification (De quelques formes primitives de classification), written in collaboration with his nephew Marcel Mauss in 1901-02, a decade before Elementary Forms, and published in the Année sociologique. This essay, in which Australian examples are followed by Amerindian and Chinese, deserves to be called the cornerstone of French structural anthropology. What distinguishes it from the work of Claude Lévi-Strauss two generations later is its conception of social causality. Where Lévi-Strauss agnostically describes a “structure,” Durkheim and Mauss insist that classificatory systems have their origin in social organization. Although they admit in their conclusion that “we still do not know what the forces were which induced men to divide things as they did between the classes,” they nevertheless remain steadfast in asserting that the logical connections between the spaces and objects they allude to are “represented in the form of familial connections, or as relations of economic or political subordination, so that the same sentiments which are the basis of domestic, social, and other kinds of organization have been effective in this logical division also” (85).

We gain insight into the difficulties this doctrine poses to empirical anthropology by the Introduction to the English translation of Primitive Classification by the British structuralist Rodney Needham, published in 1963. It is striking that although Needham, no less than Durkheim and Geertz, falls under Cosmides and Tooby’s condemnation of the SSSM, his fundamental criticism of Durkheim is nearly identical to theirs. After first reminding Durkheim à la Lévi-Strauss that mere correspondence between social and symbolic classification tells us nothing about causality, Needham continues:

The second point of general criticism is the more serious, since it shows Durkheim and Mauss’s entire venture to have been misconceived. They aptly call their essay a “contribution to the study of collective representations,” but their real concern throughout is to study a faculty of the human mind. They make no explicit distinction between the two topics, and indeed they argue as though there were none to be made, so that conclusions derived from a study of collective representations are taken to apply directly to cognitive operations. . . . Durkheim and Mauss are led by this ambiguous conception of mind to assert that the individual mind is incapable of classification, and their venture as they conceive it derives much of its justification from this assumption. Now no one would pretend that the individual could ever construct, without education in the categories of his society, a complex classification of collective representations such as the society has inherited from a long history. But this in no way implies that the individual mind lacks the innate faculty of classification; and it would be difficult to conceive, in any case, how an individual might even apprehend a classification unless the mind were inherently capable of the essential operations by which classes are constituted. (xxvi-xxviii)

That is, whatever the contribution of “society,” the individual mind must possess specific faculties prior to and independent of any given social organization. The human mind must have the capacity to process information concerning its spatial environment and to distinguish significant objects in that environment, as well as, in Needham’s words, “an innate capacity to learn to classify.” No doubt Needham’s conception of these faculties is broader and less modular than Cosmides and Tooby’s, but the conceptual point is the same.


Thus evolutionary psychology’s critique of structural anthropology as over-determined by the “cultural” is presaged by the critique advanced by structural anthropology itself against Durkheim’s radical attempt to derive all human thought from the single independent variable of the “social.” Because Durkheim’s hypothesis fails to articulate the relationship between the social and the individual, it is incompatible with the empirical study of the behavior of human individuals not merely in the context of evolutionary psychology but even in that of the SSSM.

But if the critiques are similar, the proposed solutions are not. Where evolutionary psychology demotes the social to the status of a cultural environment for the adaptations of the individual mind, structural anthropology, denying the possibility of making either the social or the psychological the independent variable, contents itself with elucidating the “structures” that lie behind both.

I believe an alternative solution is possible that permits us to protect Durkheim and Mauss’s essential anthropological intuition from the empiricist critique to which, in its original form, it is all too vulnerable. By rethinking Durkheim from the perspective of originary anthropology, we can show why, despite the abstract treatment of the social individual that no doubt made inevitable the long dominance of social science by the relativistic SSSM, the fundamental thrust of Durkheim’s work is not vulnerable to the reductive critique of evolutionary psychology.

The rethinking I am suggesting begins by violating Durkheim’s taboo concerning the origin of the sacred representations by means of which the social is generated as a conscious human reality. Durkheim claims that “[t]here was no given moment when religion began to exist . . . Like every human institution, religion did not commence anywhere. Therefore, all speculations of this sort can be discarded” (EF 20). He wants to “get as near as possible to the origins” through the study of empirical data, but not to speculate on the origin itself. Yet as Durkheim recognized in other contexts, it is only by considering human society as a phenomenon emerging in time that we can begin to understand the fundamental benefits that its specifically human aspects–language, religion, and other forms of representation–provide to its members.

It is inconceivable that our physical and mental adaptations to human language–the lowered larynx studied by Philip Lieberman, Broca’s and Wernicke’s areas in the brain–evolved before the use of language itself. This use must have become adaptive before any adaptations designed to facilitate it. Durkheim is constantly concerned with collective “representations,” but shows little interest in language per se, as though there were no need to explain the evolutionary genesis of the most important medium for such representations. Yet the crucial importance of these representations to the social order suggests that they were themselves the originary material of language, and that the social is dependent on the existence of language as the carrier of these representations. If this is so then we are obliged to found the empirical study of human culture, even “elementary” human culture, on a hypothesis concerning the coeval origin of language and the sacred representations constitutive of social self-consciousness.

No attempt to tell the story of the origin of language and culture can ever wholly disembarrass itself of the chicken-and-egg paradox in which the existence of the thing that originates is presupposed by the explanation of its origin; but–and this is my key contention–the minimal kernel of such narratives is by no means invalidated by this fact. Culture–language and the sacred–is in the first place deferral; the originary event of culture is a non-event of the natural world of appropriation that opens a space for a new social or cultural form of evolution that comes in turn to drive human biological evolution. By assigning to Durkheim’s social sacred the hypothetical originary function not simply of maintaining human society against “antagonistic tendencies” but of generating this society in order to preserve the members of our species from self-destruction through these all-too-familiar “tendencies,” we not only become able to defend Durkheim’s social sacred from the critique of evolutionary psychology, but we make possible a genuine dialogue between a humanistic anthropology and evolutionary science.


When in 1972 René Girard formulated his generative model of the sacred in Violence and the Sacred, he proved himself the true successor of Durkheim in French anthropology. For Girard, the sacred is essential to the functioning of human society because it alone can protect us from self-destruction through mimetic violence. What is missing from Girard, but what becomes possible to envisage on the basis of his work, is a hypothesis concerning the origin of language and other forms of representation. If language originated as a means to defer mimetic violence, then that very freedom for general-purpose thinking that is the central intuitive basis of the SSSM itself derives from the cultural, if not in the first place genetic, adaptiveness of the origin of representation. To reason in this manner is to reject the insipid relativism of the tabula rasa without denying or ignoring the freedom and variability of individual human beings and their cultures. This strikes me as the best way to appropriate, and to honor, Durkheim’s legacy.

Works Cited

Barkow, Jerome H., Leda Cosmides, and John Tooby. The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York & Oxford: Oxford UP, 1992.

Boyer, Pascal. “Cognitive constraints on cultural representations: Natural ontologies and religious ideas.” In Mapping the mind, ed. Lawrence Hirschfeld and Susan Gelman, Cambridge UP, 1994.

Durkheim, Emile. The Elementary Forms of the Religious Life. Tr. Joseph Swain. New York: Free Press, 1965 [1915].

———-. Incest: The Nature and Origin of the Taboo. Tr. Edward Sagarin. New York: Lyle Stuart, 1963 [1897].

Durkheim, Emile, and Marcel Mauss. Primitive Classification. Translated and edited with an Introduction by Rodney Needham. U of Chicago P, 1963 [1903].

Girard, René. La violence et le sacré. Paris: Grasset, 1972.

Gould, Stephen J. & R. C. Lewontin. “The Spandrels of San Marco and the Panglossian program: A critique of the adaptionist programme.” Proceedings of the Royal Society of London 250 (1979): 281-88.