The original idea behind Generative Anthropology before it was called GA was to provide a minimal, or in other terms, a maximally parsimonious theory of the origin of human language, and thereby of the human itself. But in the beginning I emphasized this minimalism above all as a feature of the theory, and given that all theories strive to be parsimonious, the application of minimality to language itself was not as salient as it might have been. Nonetheless, the most important feature of this minimal theory was that it began with a minimal construct of language.

Minimality has not been a feature of the latest trend in “explanations” of language. Since Terrence Deacon’s 1997 The Symbolic Species, it has been generally recognized that human language is not a development of animal signals but something entierly different. But the current studies of the connection between words and motor neurons, etc., make no real attempt to understand what specific feature distinguishes human language from animal signals. In all this “empirical” study, no one seems to consider the obvious idea that the first language must have been minimally simple, one step removed from animal signals, and could not have depended on a “language module” for its implementation. But if this is true, then studying language in the 21st century to find out the essence of human language is about as far from parsimony as one can possibly get.

Alas, this point has the vast disadvantage of being purely logical, “analytic,” inherent in the problem rather than in its “empirical” solution. Instead of doing experiments to find out what parts of the brain light up when different ideas are expressed, a research mode that Raymond Tallis never tires of mocking, it would pay those who study the question of the origin and nature of human language to close up the lab and sit down at their desks and THINK for a moment about what specifically it is that distinguishes human language from animal communication—even before, as in GA, concocting a hypothetical scenario to explain how this development might have come about.

Let us reflect on the kind of animal interactions that I assume led to the originary scene. Ethologists may correct me, but it seems pretty clear that an animal has an essentially linear relationship to objects in the world, on a gradient of highly attractive to highly avoidable.Attractive does not mean desirable; it means inducing me to attempt to appropriate it. Thus if I’m not hungry, I won’t recognize the desirability of a piece of food. This does not exclude behaviors for storing up food for later, as do squirrels, but it would be inappropriate to attribute such behaviors to an intuition of desirability (“if I hadn’t just had dinner I’d really like to eat that acorn”) rather than to conditioning either inherited or, occasionally, individual.

The need to avoid conflict promotes, as we have seen, pecking-order hierarchies of food (and sex) distribution. But the presence of an inhibiting rival is part of the overall scene in which the object is located. Until, in our “Girardian” schema of hominization, the contravening factor of mimetic attraction overcomes the inhibitory power of a stronger rival, leading to the breakdown of the Alpha-Beta hierarchy, the object’s attractiveness is necessarily diminished by the rival’s presence, as it would also be diminished by the presence of a dangerous member of a rival species. An animal needs no “theory of mind” to experience a fellow as an obstacle.

What kind of “language” is necessary to such encounters? Perhaps the higher-ranked animal will make a gesture such as beating its chest that signals its higher status to its inferior, who may respond by a gesture of submission. Here the interaction between the two individuals does not involve the object that occasioned their potential rivalry. The point is to assert dominance, and whether it be with an end to food or sexual favors or a better sleeping place is for the moment immaterial. There is no true ternary relationship, no “joint attention” to the object. I don’t deliberately show my rival that I’m interested in the object; indeed, to do this (as opposed to signaling the presence of an object, desirable or dangerous, as do the famous vervet monkeys with their varied vocabulary of signals) would require language. This is a point easy to miss, and when Tallis talks about pointing as a step beyond animal behavior, he is mistaken only to the extent that, I think out of a kind of sacred modesty, he presupposes an unnecessary and indeed concretely inconceivable tertium quid between the animal and the human, a “pointing-but-prelinguistic” stage. Détrompez-vous: language doesn’t need “words”: a gesture of pointing to initiate joint attention is already (and not “always already”) linguistic, even if etymology suggests that linguistic “utterances” must be verbal.

I think that once one takes this in, the originary hypothesis is just common sense, even if in “empirical” reality it is understood by something like 0.0000002% of the world population. (Which should, I think, make those of you who really take GA seriously feel pretty good about yourselves.) The only way we can assume that pointing will be initiated rather than grasping or withdrawing (which is not the same as deferring) is out of an inhibitory fear. But in the absence of any Alpha to defer to, and in the absence of conflictual movement toward the object, the configuration is meta-stable, and pointing-to the object would become a sign in the service of reinforcing that stability by showing, in effect, that the configuration has a “value” in itself, not simply as a stage toward the object’s appropriation. The sign, in other words, makes itself the matter at hand.

And that, very simply, is what language is and does: it makes itself the point of behavior that previously had no “point” since it did not exist as a theme for anyone to point to. Tallis’ thatter is what-is-the-point, and although reference to pointing here is a pun, it is justified by the catachretic trace of the parallel between (I imagine) the point of a sword and of a finger; if I stick you with either one, I get your attention. All of a sudden, the members of what was previously just a crowd with a hierarchy are all simultaneously motivated to maintain a communalconfiguration in which absence of conflict around a desirable object (which for the first time we can actually call desirable) has become a value in itself, a “pleasure” in the absence of pain, but also, we should imagine, a sense of numinousness in the non-instinctual contemplation of the object, a state that the intentional sign embodies in what we may call a memory-trace.


I could go on, as I have elsewhere, but I think this is pretty close to explaining the difference between language and pre-linguistic states, between the prehuman and the human. The space between the participants and the object is an empty space, a néant, that is traversed only by language but not in reality. Thinking the object, thematizing it, thatting it, charges that space with an interdictive energy that we call the sacred. It is this space that by deferring the appetitive presence of the object allows us to intend it in a human, signifying way.

Thus the famous deconstructive critique of presence is both productively and disastrously ambiguous. Presence in the human sense is precisely not the kind of presence that obtains in an animal’s field of sensual awareness; it exists only as deferral, whereas presence in the sense of “immediacy,” whether controlled as in ritual or occasioned by loss of control as in Girard’s scenes of lynch-mob violence, is always accomplished by filling in and effectively abolishing this space of representation—a space that elsewhere in the animal kingdom simply does not exist. The human scene is a scene of representation, of which language is the most universal, least energy-consuming form. And thus we can define violence—in its exclusively human form, not the mere aggression of “nature raw in tooth and claw”—as the breakdown of the scene of representation, the overwhelming with action of the néant in which the sole expense of human energy should be that of the sign.


Perhaps this is a good moment to formulate a proper response, after a two-year delay, to Jean-Pierre Dupuy’s showing of a provocative mob scene from Fritz Lang’s Fury in a session at the joint COV&R-GASC meeting in Tokyo in 2012. The clip began with the sheriff facing down the mob gathered with the intention of breaking into the jail to lynch a murder suspect. At first he cows them with his legitimate authority, but then a rock is thrown from beyond the group gathered immediately around the sheriff, and this gives the signal for the mob to attack.

At the time, I reacted as though this was being proposed as a kind of Girardian originary scene, which wasn’t really the point. On reflection, I was struck by the fact of the stone’s thrower being outside the “stage” of the event, making it essentially different from the adulteress-stoning scene in John 8 where Jesus prevents the throwing of the “first stone.” In the Lang scene, the “first stoner” is anonymous and not accountable, the scene is open, not closed, and we have a “mimetic crisis” that is the inverse of the orderly, ritualized execution that Jesus’ gesture paralyzes.

The more general point I would make now is that the breakdown of order comes as the result of a violation of the néant that is figured here by the empty zone between the sheriff and the immediate crowd. This physical violation is precisely not that of a human but of a projectile. Although apes occasionally throw things, they do not use projectiles as weapons in a systematic way. This use (and Tallis would no doubt agree) is itself a product of human intentionality; it is not an instinctive movement toward the object nor an example of animal “tool” use. In order to take aim at something, we must first stand back from our appetitive movement toward it; we intend it. The possibility of peacefully intending the sacred object via the sign is also the possibility of violently intending it through a missile. My difference from Girard on this point is “merely” to insist that the latter intention is dependent on the former.

Jean-Pierre’s scene thus illustrates the dark underside of human intentionality and the deferral that occasions it. The scene of representation is the sacred space in which the new dimension of the sign originates and is shared and internalized as a cultural memory within a universally maintained history recallable by each individual as the memory of the sign. But this same space is by the same token a potential space of violence. Deferral is not demise; to defer violence is also to accumulate it, just as in the originary hypothesis the final sparagmos releases the deferred appetitive energy. More specifically, the space of deferral is vulnerable to its deliberate violation from “outside,” that is, from the profane world that inevitably surrounds the sacred inner space—and that is ritually excluded in the stoning scene in John. Just as the erotic exists only as a transgression of the néant, so does mob violence. These uniquely human categories all exemplify the power of the paradigm manifested in the originary hypothesis.


If as a result of this discussion, another 0.0000001% or so can be brought to this realization, it will have accomplished its object. Let no one say we lack either faith or patience.