Chronicle 192 mentioned Emile Durkheim’s refutation of Max Müller’s “naturist” theory of religion, which boils down to the denial that the radical otherness of the sacred can be derived from human reaction to nature, even the unusual in nature. As we observed in passing, Durkheim’s far more sophisticated theory of the sacred as an ideal projection of society nevertheless fails to explain how this projection is to be generated. In gaining his understanding of the place of religion in the social order, Durkheim abandons Müller’s most valuable intuitions: that the sacred is the product of a memorable event, and that the most concrete and durable trace of this event is to be found not in ritual but in language.

This having been said, Durkheim comes qualitatively closer to a generative conception of religion and culture than anyone before Girard. His insight into the sacred might be more appreciated today had he not made the unforeseeable marketing error of describing primitive religion in terms of “totemism.” (The neglect of Freud’s Totem and Taboo, another proto-generative work, springs no doubt in part from the same source.) The idea that totemism is a constant feature of “elementary” religion has been generally discredited. Claude Lévi-Strauss’s Le totémisme aujourd’hui (1962) reduces totemism, to the extent that it may be said to exist at all, to a classificatory system disassociated from sacrificial ritual; the totem is, as Lévi-Strauss puts it, “good to think” rather than good to eat.

The rejection in principle of Durkheim’s search for a universal model of religion is one more case of the empirical trees overgrowing the theoretical forest. It is important to study the facts in their diversity, but it is equally important to formulate models that parsimoniously explain these facts. Such models have not been forthcoming in this area for a long time. The fact that the elements of totemic systems are better to think than to eat does not discredit the hypothesis that the totem was originally a sacrificial animal any more than the arbitrariness of a given signifier discredits the hypothesis that this signifier was originally motivated. To reduce the sacred to a product of our classifying intellect is to ignore its role in the genesis of the human–more specifically, in bridging the gap between animal and human communication systems. Contemporary research into the origin of language has revived interest in this gap, but the neuroscientists and paleoanatomists who conduct this research are not professionally inclined to take an interest in the sacred.


The strength of Durkheim’s understanding of religion lies in his uncompromising insistence on the separateness of the sacred from the profane. This makes Durkheim a “dualist” in a sense far more anthropologically significant than that defined by the incommensurability of mind and brain (see Chronicle 193). Although he never discusses the origin of language, it is a small extrapolation from statements like the following to the more fundamental hypothesis that sacred difference (if not precisely the sacred-profane distinction) is the origin of meaning itself.

Religious forces . . . are moral powers because they are made up entirely of the impressions this moral being, the group, arouses in those other moral beings, its individual members . . . Their authority is only one form of the moral ascendancy of society over its members. But, on the other hand, since they are conceived of under material forms, they could not fail to be regarded as closely related to material things. Therefore they dominate the two worlds. . . . It is this double nature which has enabled religion to be like the womb from which come all the leading germs of human civilization. Since it has been made to embrace all of reality . . . the forces that move bodies as well as those that move minds have been conceived in a religious form. That is how the most diverse methods and practices, both those that make possible the continuation of the moral life (law morals, beaux-arts) and those serving the material life (the natural, technical and practical sciences), are either directly or indirectly derived from religion.
    The Elementary Forms of the Religious Life (tr. J. W. Swain), Book 2, Ch. 7, sec. 3, p 254-55)

All that is lacking here is consideration of the relation of religion to language. Religion is the “womb” or matrix of culture because it “dominate[s] the two worlds.” The notion of “religious forces” as “conceived of under material forms” is a reference to totemism, but what is at stake is the more general point that what is sacred and therefore the basis for significance manifests itself at the same time in the material world of appetite. The meaning, Idea, or signified can only be grasped by being attached to a material referent whose mortality demonstrates the formal separation of words and things, the ideal and the real.

Durkheim rejects “naturism” because, as he puts it, “The impressions produced in us by the physical world can, by definition, contain nothing that surpasses this world” (2.7.4, p. 256). A worldly reality can be elevated into a totem only through the mediation of the group or clan. The sole element lacking in Durkheim’s model is an explanation of how the clan comes to be “represented” by the totem. The originary hypothesis generates this representational relation, which is precisely that of language, from the concentration of the group’s mimetic desire on the object. In the light of this hypothesis, Durkheim’s tentative insights into the origin of the totemic representation of the clan may be applied, more radically, to that of language itself.

Durkheim’s explanation of the clan’s attachment to its totem as a symbol of itself is a petitio principi:

We have shown how the clan, by the manner in which it acts upon its members, awakens within them the idea of external forces which dominate them and exalt them; but we must still demand how it happens that these forces are thought of under the form of totems, that is to say, in the shape of an animal or plant.

It is because this animal or plant has given its name to the clan and serves it as emblem. In fact, it is a well-known law that the sentiments aroused in us by something spontaneously attach themselves to the symbol which represents them. . . . This transference of sentiments comes simply from the fact that the idea of a thing and the idea of its symbol are closely united in our minds; the result is that the emotions provoked by the one extend contagiously to the other.

At this point, we still have not learned why the specific totem is chosen as this “emblem.” The word “contagiously” refers ostensibly to a relation of signs in the mind, but its more fundamental referent is the group itself.

Durkheim continues:

For we are unable to consider an abstract entity, which we can represent only laboriously and confusedly, [as] the source of the strong sentiments which we feel. We cannot explain them to ourselves except by connecting them to some concrete object of whose reality we are vividly aware. [my emphasis] Then if the [abstract entity] itself does not fulfil this condition, it cannot serve as the accepted basis of the sentiments felt, even though it may be what really aroused them. Then some sign takes it[s] place; it is to this that we connect the emotions it excites. . . . The soldier who dies for his flag, dies for his country . . .

Now the totem is the flag of the clan. . . . It is therefore natural that the impressions aroused by the clan in individual minds–impressions of dependence and of increased vitality–should fix themselves to the idea of the totem rather than that of the clan; for the clan is too complex a reality to be represented clearly in all its complex unity by such rudimentary intelligences. More than that, the primitive does not even see that these impressions come to him from the group. He does not know that the coming together of a number of men associated in the same life results in disengaging new energies, which transform each of them. . . . However, he must connect these sensations to some external object as their cause. Now what does he see about him? On every side those things which appeal to his senses and strike his imagination are the numerous images of the totem. . . . Placed thus in the centre of the scene, it becomes representative. The sentiments experienced fix themselves upon it, for it is the only concrete object upon which they can fix themselves. It continues to bring them to mind and to evoke them even after the assembly has dissolved, for it survives the assembly, being carved upon the instruments of the cult . . . It is still more natural to attribute them to it for, since they are common to the group, they can be associated only with something that is equally common to all. Now the totemic emblem is the only thing satisfying this condition. By definition, it is common to all. During the ceremony, it is the centre of all regards. (2.7.3, p.251-52)

I have quoted this passage at length in order to demonstrate both how close Durkheim comes to the originary hypothesis and how effectively this hypothesis resolves the residual obscurities of his text. As we see, the specific reason why this object is the totem of this clan is never given. On the level of individual psychology, the association is explained in the italicized phrase by the individual’s vivid awareness of a concrete reality. In what follows, the totem is presented as such a reality in the context of a preexisting totemic ceremony. Yet although the origin of the ceremony itself is not given, the centrality of the object and its “vivid” presence to each individual participant are precisely the conditions for the arousal of mimetic desire in the scene postulated by the originary hypothesis.


Durkheim is truly the founder of modern religious anthropology. We can only regret that, despite all the attention his work continues to receive, his most fundamental contribution to human science has been put aside by his empirically minded successors. It would be comforting to attribute this neglect to Durkheim’s failure to grasp the connection between mimetic desire, potential violence, and representation, but my own experience suggests that had he made this connection, the neglect would have been still more radical.

Durkheim’s thought on religion is far too rich to be treated in a single column. In a later installment I will deal in greater detail with his notion of the sacred not merely as a functional one for preserving the solidarity of the group, but as the basis of representation, which is, in the first place, language.