We necessarily think in propositions and for this reason cannot escape the metaphysical, which is merely a synonym for the transcendental; a declarative sentence creates a “fictional” reality “next to” (meta) or independent of the physical world. But we need not follow the metaphysical tradition in taking propositional or declarative language for granted. The corollary assumption that language is essentially independent of its specific origin among the beings who use it allows us to describe the natural world, and the human world, too, to the extent that it likewise takes for granted the existence of propositional language. What metaphysical thinking is incapable of describing is the one moment at which language cannot be taken for granted, the originary moment that inaugurates the transcendental relationship between representation and reality that uniquely characterizes the human. This moment has been from the beginning the province of religion; more recently, it has become that of generative anthropology.

The interface between the real and transcendent worlds is paradoxical in a more fundamental sense than logical paradox; it is literally outside or “beside” (para) the realm of linguistic representation because it is where representation came into being. What is “absurd” in religious discourse from the standpoint of propositional reason can always be understood as the worldly incarnation of the otherworldly characteristics of language, or more generally of representation; religious narratives describe the miracle of becoming-language. The specific irrationalities of these stories are not only a source of solidarity and exclusiveness to their communities of believers; each story formulates an originary hypothesis of the human, albeit one not characterized by parsimoniousness. The common characteristic of the subjects of religious narratives is that they share the non-mortality that belongs only to the world of the sign. Signs do not die, and the gods who share this quality explain by their presence on earth the presence of these signs.

For René Girard, the immortality of the gods is a mythical deformation of the role of the emissary victim in bringing peace to the community. The victim is killed, but he remains present through the expurgated memory of the murder, which brings peace to the community by concentrating its aggression on the single “scapegoat.” Thus the myth tells an anthropological truth at the expense of a lie; its dead protagonist is presented as eternally alive, disculpating the murderers of their crime. But what Girard describes as an example of méconnaissance is the basis for all our religious ideas, including his own; whether or not there is a “true” immortality and a “true” God, the only anthropological source of these categories is the transfer to the victim of the categories of the sign by means of which we commemorate him. To call the primitive forms of transcendence misapprehensions is to go beyond the anthropological lesson of Christianity itself, which is that there is no ontological separation between the human and the transcendent. Jesus’ role is not to end this gap but to declare it fictitious.

In his little book Saint Paul (PUF, 1997), Alain Badiou makes the point that Paul’s only proposition about Jesus is that he was resurrected, an “absurd” affirmation that cannot be understood as a simple empirical statement.

Paul . . . reduces Christianity to a single proposition: Jesus is resurrected. Now this is the [my emphasis] fabulous (fabuleux) element, since all the rest–birth, preaching, death–can after all be defended [as plausible]. What is “fable” is the element in a story that we conceive as having no contact with reality, save through the invisible and indirectly accessible residue [of the fabulous] that clings to every overt figure of the imagination (qui colle à tout imaginaire patent). (p. 5; my translation)

Badiou’s reference to the “residue” of the fabulous that clings to the reality-based imagination expands on my point above concerning the originary “absurdity” of the interface between the worldly and the transcendent. The originary hypothesis explains the pertinence of this particular “absurdity.” The resurrection is, once more, a transcendence mediated by, or modeled on, the act of representation; like Girard’s mythical victims who appear to avoid death, Jesus accedes from the mortal state of the inhabitants of the real world to the immortality of the sign, the difference being that he is explicitly described prior to his miraculous resurrection as both human and innocent. Jesus’ transcendence of death through the sign is revealed to Paul on the road to Damascus as the product of his persecution of Jesus’ original followers, a point I developed in Science and Faith. Paul’s reduction of Christianity to the “fabulous” affirmation of the resurrection is, in religious rather than scientific terms, a minimal statement of the hypothetical communal discovery/invention of transcendence through the reciprocal exchange of the sign as the “aborted gesture of appropriation.”

The hardest thing to accept in GA’s new way of thinking is that “ordinary” human language is a form, indeed, the fundamental form of transcendence, and that the atheistic critiques that proclaim the superfluity of God and miracles are blind to the fact that language is as much of a miracle as the biblical act of creation. This does not mean that representation cannot be explained without supernatural forces, merely that the ontology of language has more in common with that of God than with that of animal communication or the genetic code. The “absurd” affirmations of religious discourse are mediations between the worldly and the transcendental, not alternatives to the propositions of science, which do not so mediate. Similarly, the originary hypothesis is not an alternative to religious discourse but a minimal model of the first event that is in principle compatible with all such discourses, if not with the absence of any such discourse. For to claim that it is unnecessary to narrate an inaugural event of our mutual communication through signs is less to deny the transcendental status of representation than to leave it unthought.

The new anthropic principle

The term anthropic principle is commonly used as a means of dismissing claims of divine creation: if the latest cosmological model makes the conditions for the existence of (human) life highly improbable, it suffices to note that if we didn’t exist, we wouldn’t be observing these conditions; hence if the probability of a universe’s meeting these conditions is 1/n, it suffices to remove this anomaly to assume that ours is the one “anthropic” case out of k universes, where ((n-1)/n)^k <.5.

The sheer arrogance this kind of “reasoning” is comparable only to its naiveté, as though we were certain enough about the origin of the universe to assess the probability of the conditions of life and then hypothesize the existence of zillions of other universes of which we can never have the slightest evidence. Instead, we should redefine the anthropic principle inanthropological rather than cosmological terms: the universe, but particularly, life on Earth, must be such as to permit the origin of representation. Since the emission of a sign is a voluntary, conscious, and memorable act that commemorates the objects and acts of which it is the representation, this new anthropic principle entails that at a given moment in the evolution of our ancestors, a first act of representation occurred, commemorating the first event. A world without events is a world without representation. The Darwinian model of gradual, genetic changes is not inconsistent with the anthropic principle, but does not entail or imply it.

Creation stories

The “absurd” religious narratives that mediate between the world of appetite and the world of transcendence are all more or less explicitly creation stories that tell of the irruption of the sacred into the appetitive world that creates us as humans through the gift of representation. The creation of the natural universe by this same force may then be understood as an affirmation of the new anthropic principle: the creator of a universe within which humans can be created must be as powerful as, if not superior to, the creator of humanity.

I need not dwell here on the importance of the creation story for the Judeo-Christian and Islamic traditions. Yet some religions place little emphasis on creation stories; either the universe is considered as eternal and/or cyclical, or, in the words of the Buddha, knowledge of the origin of [human] life “is not fundamental to the holy life.” We are invited to turn away from concern with an originary event in a doubling of the originary renunciation of the central object. Not seeking to know the moment of origin sets an example for renouncing knowledge of any event, that is, any memorable violence done by human desire. This constitutive renunciation makes Buddhism both more and less than a religion in the Western sense. A major factor in its appeal to many Westerners is that, in contrast to the Abrahamic religions, it requires no belief in a sacred history. GA was born in the context of the Western tension between man creating God and God creating man; the Buddhist alternative detaches the originary act of renunciation from its locus in an event in such a way as to obviate the need for either mode of creation. But this detachment is an act of askesis that demonstrates the “necessity” of a creation story as much as the stories themselves.


The “absurd” narratives of the various religions are translations of the same basic insight that the human is defined by the transcendental; each draws different anthropological consequences from this fundamental dependency. Indeed, the same may be said of the fictions of art, which also depend on our faith in the artist’s ability to give human plausibility to the intentions of his subjects. The originary hypothesis provides a minimal principle of interpretation for these narratives. Whether or not the hypothesis is acceptable to the faithful of any given religion, I can at least claim that the hypothesis in its minimality is designed to have a minimally perturbing effect on believers in general.

Were a proponent of GA to participate in a discussion among partisans of various belief systems, the point would not be to argue about the specifics of the originary hypothesis, but to begin from the agreement that the hypothesis of an originary event is necessary. Religious believers would agree, because they have already implicitly accepted this assertion; the holdouts would be the atheists. But we would hope to persuade someone with no use for the supernatural of the usefulness of a hypothetical natural event on the basis of which he could both begin to understand religious narratives and enter into dialogue with those who believe in them.