Sparagmos! is a Vancouver-based reading and discussion group that meets regularly to discuss the mimetic theory of René Girard and the generative anthropology of Eric Gans.

Sparagmos! reads all kinds of books together, including novels; Dostoevsky’s The Devilswas a recent project. Interested locals are invited to contact the group through the GA List about joining the Vancouver meetings. Andrew Bartlett, with great hospitality and style, has been hosting the meetings at his home for some years now.

The dialogue below is compiled and edited from a recent exchange concerning the originary scene. In it, various Sparagmos! members are grappling with the details and merits of both Girard’s and Gans’s hypotheses concerning hominization.

The dialogue begins with Pablo who, from the perspective of the Girardian hypothesis, starts to ask the local GA expert, Richard, questions about the Gansian hominization hypothesis.

Pablo: Both Girard and Gans agree that the central object/victim must be arbitrary. How can a truly arbitrary object be an object of appetitive desire?

Richard: How can the words “roast beef” both be arbitrary (you need to learn their meaning from someone else) and refer to real roast beef? What enables humans to distinguish arbitrary signs from non-arbitrary objects?

Suppose I put a juicy piece of roast beef in front of my dog, then point at it and say, “roast beef.” My words do not “mediate” the object for the dog. If he pays any attention to them at all, it will only be as another index that contributes to the central focus of his attention, which is that delicious smelling piece of beef.

The point of the sign’s (or victim’s) arbitrariness is not that there exists no previous motivation for attending to objects. It is that the new mode of attention inaugurated by the originary symbolic sign transcends the previously existing appetitive relation. “Transcends”: not in the sense that it means we have become angels instead of animals–far from it. But that we can no longer attend to the object independently of the mimetic awareness that the other model/mediator/rival is attending to it as well. Psychologists call this “theory of mind,” the ability to see things from someone else’s perspective. (It appears to develop in children around age 2–in other words, when they begin to acquire language.) I see it as the basis of symbolic thought, grounded in the originary scene of collective representation.

Why, if animals are also mimetic, do only humans superimpose on the mimetic relation the mimesis of signs, the scene of representation? But the question implies a teleology from mimesis to language which does not exist, which is why only a punctual originary hypothesis can explain this difference. Mimesis leads us to attend to objects by watching–“imitating”–others. Throw a piece of bread off the wharf and immediately seagulls swarm the morsel. Mimesis works. What makes human mimesis different? The difference is difference itself: i.e., the arbitrariness of the relation between imitated gesture and the object it “points” to (cf. my dog and the roast beef). In the originary scene, this gesture has to be motivated appetitively, just as in the case of all other animals. If it isn’t, then we are attributing to the scene a transcendental motivation that does not yet exist. But the end result of the originary scene–if it is truly originary–is not just another swarming around an appetitive object, but the designation of it as something other–the appetitive mimetic gesture becomes a gesture of designation, separating sign from object, periphery from center, human community from sacrificial other.

By the way, I included the victim in my account of Gans as a pedagogical concession. The violence of the originary scene–the sparagmos–is in the first place an act of designating the object as central, which is to say, as different from the human periphery. This is the originary transcendental separation between (peripheral) human and (central) god. It is an unverifiable empirical matter whether this object is a conspecific or (as seems more likely) a nonconspecific game animal. What is important is the separation between center and periphery. The ensuing sparagmos is the “honor” paid to the god for having “created” this originary separation between sacred center and profane periphery.

Pablo: You say, “In the originary scene, this gesture has to be motivated appetitively, just as in the case of all other animals. If it isn’t, then we are attributing to the scene a transcendental motivation that does not yet exist.” Isn’t mimetic violence just as natural to animals as appetite? I’ve been careful not to use the term “mimetic desire” in the context of this pre-human mimetic crisis, precisely to avoid this false injection of humanity into a pre-human scene. But mimetic violence can (and usually does) ultimately lead to an all-against-one gesture of the group against the victim–or in Gansian terms, the periphery against the center.

Chris: Do animals really scapegoat this way, i.e., with a scapegoat as the inevitable end result of “mimetic violence”? Doesn’t designation have to happen before there is a scapegoat (pace Girard, who reverses this order)? In other words, isn’t scapegoating a human phenomenon?


Mimetic animals use pecking-order hierarchies to preserve the species; hyper-mimetic proto-humans, however, unable to be constrained by the animal hierarchy straightjacket, would have died off from bashing each other’s skulls in: but they suddenly evolved language instead, when their hyper-mimesis allowed them to stumble into suddenly representing an appetitive object that their mimetic crisis had originally been fighting over by means of escalating imitation.

What makes imitation become representation? Isn’t it the shift of attention from the mimetic rival back to the appetitive object (now being designated thanks to the logical outcome of the hyper-mimetic crisis)? In other words, isn’t Gans on to something with that “counterintuitive result” of his from Signs of Paradox, Chapter Two?

Pablo: Chris, I remember getting stuck on this point, too–talking about “mimesis” in a pre-human context. But the difficult thing is precisely the fact that we do indeed observe mimetic escalations of violence in the animal kingdom, especially among the “higher” animals like primates. As you go up the evolutionary ladder you generally find an increasing capacity for imitation. One can imagine that the hominids we’ve been talking about, situated on the highest pre-human rung of the ladder, must have been extremely mimetic. But, as you imply, this cannot be “mimetic desire” in the human sense yet. When we see mimetic violence growing in a group of chimpanzees it cannot be that they are caught in any existential predicament like Girard’s double-bind or Gans’s esthetic paradox. I think what these higher animals are actually imitating in this situation is each other’s emotions (that is, the basic animal emotions of fear, anger, etc.). This is mimesis in its most basic form–direct imitation of emotion. Fear inspires fear, anger inspires anger, and then violence inspires violence. It is only among humans that desire enters the scene, and basic mimesis becomes “mimetic desire.”

So, in short, animals do not “scapegoat” in the human sense. That is, they do not think of their victim as a scapegoat. This would require not merely language or designation, but the quality of transcendence that makes recognition of the Other, and therefore language, possible. But they do kill just the same, collectively, as the result of a basic mimetic crisis. My point to Richard was that this violence is just as natural to higher animals as any appetite.

I wanted to pick up on a question that Richard asked, but for some reason got lost in the discussion. He asked what it was in Girard’s “originary scene,” involving the victim rather than an appetitive object, that made the members of the group remember that moment? Or in other words, what was it that allowed the originary moment to have a lasting (i.e. permanent) effect?

Gans addresses this on page 34 of Science and Faith:

Without the peaceful division and distribution of the central appetitive object, the peace bestowed by the aborted gesture would have endured but little, and its memory even less; the survival of the system of representation inaugurated in the originary scene required that it lead to an appetitive satisfaction greater, or at any rate more secure, than that which had previously been available.

This seems to imply that after the originary event–after the pre-human animal crosses the threshold into humanity–it is actually possible to revert back to animality if the originary event is forgotten. In fact, it may be that humanity was born more than once in the history of the world before it finally “stuck” due to the particular events that immediately followed it. Do I have this right?

If this is correct then, of course, I have a problem with it. As we were discussing before, the crossing over into humanity involves the acquisition of the quality of transcendence. In Gans’s words, it involves the recognition of a “transcendental ontology.” It seems to me that this threshold is irreversible; once you acquire the quality of transcendent thought you can’t just give it up or forget it. From that point on you see things differently, whether you remember the originary event or not. The notion of becoming human and then spontaneously losing your humanity again is disturbing.

I also have a related question, referring to the second part of that Science and Faith Gans quote. The crucial moment in the originary scene, the aborted gesture, is the result of escalating mimetic interaction, culminating in extreme resentment (the first actual sensation of resentment). Can we really say that the subsequent peaceful satisfaction of the usual appetite is more memorable than the originary event itself? It is certainly more “stable,” but I think Gans’s point is that this action is more memorable, or makes the originary event more memorable. I’m not sure why this is necessarily so.

Moreover, distributing the food to each member of the group certainly results in more appetitive satisfaction than nobody getting any of it. But it results in less satisfaction than getting all of it, which is really what each animal was trying to accomplish before the originary event. So, again, from each animal’s point of view, why should this distribution of food, which results in less appetitive satisfaction than each animal was hoping for, make the originary event more memorable?

Chris: No, I don’t see the Science and Faith passage opening up the possibility of “reverting back”. It’s simply stating the necessary connection of the peace of the sparagmos to the invention of the first sign.

What interests me is whether the originary scene only happened once, or whether or not it had to occur multiple times in prehistory among different communities. This question applies to both Girard’s and Gans’s hypothesis.

Pablo: The problem is that, according to Gans, the peace of the sparagmos comes after the first sign… after the animal becomes human. If this peace is in fact necessary to remain human, than this implies that without this peace the humanizing effect of the first sign would not last–i.e., the human would revert back to animality.


Richard: Yeah, I agree with Chris that what makes the originary event “originary” is whether it is remembered or not. If it isn’t, then it isn’t originary. The aborted gesture wasn’t a sign and the “sparagmos” was just the usual business of animals competing for a piece of the appetitive object.

It is of course possible–indeed likely–that there were countless times when this envelope was pushed, when mimetic interaction tended toward “something like” the originary event. But the whole point of the originary hypothesis is that this “something like” cannot be substituted as an explanation for the origin of symbolic culture. The definition of the latter is that it is conscious, that is, memorable to its participants in a way radically different from the instinctual mechanisms of animals. This is why Gans objects to the “gradualism” of evolutionary theory when it comes to explaining culture. Evolutionism does not explain culture so much as explain it away by reducing it to noncultural precursors, which is to say, to events that are non-originary or non-symbolic.

All this is apart from the question of monogenesis or polygenesis. It is logically possible that different hominid groups in different geographical regions invent/discovered language. But, as Gans observes, this is not very parsimonious. It is far more likely that the originary event–which is itself a highly unlikely and anomalous occurrence–occurred just once among one group of (proto-) humans. One can then assume that language/culture spread by diffusion. Once culture originates it can be learned by imitation. This is presumably the point of religious ritual, which produces a type of membership that transcends the kinship relations of animal social groups.

Pablo: What exactly is the “originary event?” The aborted gesture as first sign or the sparagmos that comes afterward? From everything we’ve discussed so far, I understood that it is the aborted gesture that is the critical moment when the threshold from animality to humanity is crossed. The sparagmos comes afterward, and helps to explain the characteristic violence and distribution of food we see in myth and ritual. Now it sounds like we’re saying that if the sparagmos doesn’t happen then the originary sign wasn’t actually a sign in the first place, which doesn’t make sense. Either it was the first sign or it wasn’t. What happens afterward can’t change that.

Maybe what you guys are trying to get through my thick skull is that the originary event is remembered by future generations only if the sparagmos occurs. In other words, the original group of proto-humans that experienced the event remain human afterward, and because it is remembered via the sparagmos this experience is taught to other proto-humans. Is that it?

If so, this brings up another related question (sorry)… Is it really possible for proto-humans (who, by definition, are still only animals) to “learn” to be human? Can you really teach an animal to be human? Is it the fact that these proto-humans we’re talking about are “hyper-mimetic” that makes this learning possible? If so–if these proto-humans are on the edge of humanity, so to speak–then I’m not sure we need something as drastic as an originary mimetic crisis to make them cross the threshold.

Richard: The aborted gesture is the difference. If the gesture is recognized as a symbol (= Gans’s “mimetic paradox”), then the originary event occurs. If it isn’t so recognized, then no originary event. Hence, no humanity.

The sparagmos is the aftermath of the sign: the release of the tension of the originary mimetic/symbolic paradox on the central object. When I referred to the “sparagmos” of animals competing over an appetitive object, I was speaking rather loosely (mea culpa). There is no sparagmos–i.e., no sacrificial crisis–for animals. Just the usual wolf-pack scramble for a piece of the appetitive object.

As to how the originary event is remembered, this is the nub of the issue. In order for the scene to be remembered as a scene with a periphery and center, the fundamental structure of the sign must have been understood, which is to say, consciously performed and remembered (which is why Gans insists that it precedes the sparagmos–something I probably didn’t emphasize sufficiently during our last meeting). Henceforth the object is not merely a piece of meat, which I attend to because my body has evolved biologically to attend to such things. I now attend to it as an object of desire, which is to say, because it is situated at the center of the scene of representation. The object also partakes of the signifying structure of the sign; it is the imaginary “referent” of the sign, and as such it transcends its appetitive reality. I now do not merely want the object to fulfill an appetitive need. I desire it for its imagined significance at the center.

Gans’s point is that scene cannot be remembered without the “placeholder” of the sign, which, in any case, is all that is left of the scene by which to remember it. The sign is “minimal,” in that all it requires is the utterance of the word. But in the elementary case of ostensive culture, the sign is not merely a word independent of its worldly object (that comes only with declarative culture): it is a word together with its object, the ritual reproduction of the originary scene. The original participants of the originary scene discovered this relationship between sign and object by a “revelation”–at least that is how it would have appeared to them. Newcomers to the postoriginary world would have to learn this relationship, not by revelation, but by participating in the ritual reproduction of that original revelation. That’s why Gans objects to polygenesis–it would be like asking God for another thunderbolt. “Aw shucks, come on. Just give me one more proof, please.” No way.

Latecomers to the scene are drawn to it by mimesis: they are instructed in how to defer resentment by the “rites of passage” of postoriginary repetitions, which is to say, by the culture they did not invent but inherited from their forefathers, beginning with those participants in the originary event, who made the most revelatory discovery of all: the virtual membership (the “city of God”) that is humanity.


Pablo: I think Richard and Chris are finally getting through to me. The idea that the “peaceful distribution of the appetitive object” follows logically from the originary sign does indeed make sense. It effectively confirms the significance of the originary event in the minds of the new humans, and therefore prolongs the memory of it. In this sense it is analogous to the peace that follows the killing of the victim in Girard’s originary scene.

I was basically confusing the “peaceful distribution” that Gans talks about in Science and Faith with the sparagmos that he talks about in Signs of Paradox. Now, I should probably just go back and re-read Signs of Paradox, but maybe one of you can remind me of how these two things fit together. Gans says that the aborted gesture is necessarily followed by the peaceful distribution of the object, without which it would not last in the memory of the group. The sparagmos, by contrast, is a violent release of mimetic tension that also follows the aborted gesture, and results in the violent tearing apart of the object. How do both of these coexist in the originary scene?

To follow up on what Richard says… The notion of pre-human animals “learning” culture from those hominids that actually experienced the originary event seems troubling to me. I can certainly understand why one would say that these animals must learn this culture mimetically, since that is the only mode of psychological interaction available to an animal. But the fact is that no matter how many times you repeat a gesture in front of a chimpanzee, it will never be understood as a representative sign by the chimp. An animal cannot “learn” to be human. So it’s not obvious to me how the animals that were not part of the originary scene managed to become human as well.

As I said previously, it may be that the “hyper-mimetic” nature of these hominids allows them to cross over into humanity by mere imitation more easily than other animals. But if it was so easy for them to cross this threshold in the first place, why was something as dramatic and special as the originary event even necessary?

Incidentally, this question applies to Girard as well as Gans.

Richard: I’m glad I haven’t made things more confusing Pablo.

As to your more recent query, I think you may be trying to ask too much of the originary event, the function of which is epistemological, not empirical. That is, you seem to be saying that if one can’t teach chimpanzees to go through something similar (for example, in the primatologist’s laboratory), then the originary hypothesis is “falsified.” Two points:

  1. The hypothesis is neither definitively confirmed nor definitively falsified by such experiments. Even if you could teach a chimpanzee to understand symbolic signs, this will not confirm the truth of the hypothesis. Incidentally, things aren’t as clear-cut as you seem to assume. Under very artificial laboratory conditions, it appears that chimpanzees are able to acquire a rudimentary understanding of symbolic signs. (I would cite Sue Rumbaugh’s experiments with the bonobo or pygmy chimpanzee Kanzi, the results of which are very perceptively analyzed by Terrence Deacon in his book The Symbolic Species [1997].) What these “successful” experiments show is that there is no innate genetic reason preventing chimpanzees from acquiring language. In a sense, this provides corroboration (but not proof) of the originary hypothesis, because it proves that the deciding factor is not biological. The nonbiological origin of language is something assumed by the originary hypothesis; the latter is a memorable event and cannot be explained either as a genetic mutation (Chomsky) or as a gradually evolved extension of preexisting animal systems of communication (again, Deacon is instructive here).
  2. Relatedly, you have misgivings about my suggestion that “latecomers” to the originary scene of culture must learn to participate in it by imitating their (cultural) precursors. But is this so objectionable? Is it not more implausible to expect that each newcomer must–in isolation from the original participants–go through the originary event all over again? What ritual repetition does is to reproduce the structure of the originary event in each “new” individual. What is “originary” about ritual is that it refers to the originary event itself.

The language-training experiments with chimpanzees are instructive in this regard. As I suggested, under special conditions it is in fact possible to impart a rudimentary symbolic “culture” to chimpanzees. But in their natural habitat, no chimpanzee ever comes close. Why? Clearly, the infrastructure provided by the experimenters, which requires extremely tedious “ritual” repetitions of indexical sign/object relations, reproduces something like an “originary event” for the chimpanzees. Deacon even suggests that the moment of “recognition” for the chimpanzee occurs as an “insight” rather than as a repetition of previous sign/object relations (which are indexically learned). In other words, the realization that the relationship is symbolic rather than indexical is something that must be “discovered” by each chimp. And this discovery occurs only after laborious ritualized training. I would suggest that the function of what Gans calls “elementary” culture is analogous with the chimp experiments: a long period of co-evolution between elementary culture and human brain structure, lasting perhaps millions of years, in which the human brain evolved to become predisposed to acquire symbolic culture. The reason why children acquire language so effortlessly is the consequence of this long period of co-evolution.

It is also extremely significant that so-called “wolf” children–i.e., those who receive no exposure to language before their brain matures–never acquire a normal capacity for language. In other words, despite the fact that we are–thanks to millennia of co-evolution–biologically predisposed to acquire symbolic culture, that acquisition still depends–in the final analysis–upon a level of interaction that is not biological but anthropological, which is to say, dependent upon our mimetic participation in (a descendant of) the originary scene.


I guess what I’m saying is that your objections are too focused on the empiricism of the event. If the prehumans are so predisposed to crossing the “threshold” of symbolic culture, why bother with the originary event at all? Why not just say that gradually more and more prehumans crossed the threshold as they learned to focus their mimetic aggressions on a single victim.

I think that the above “gradualization” of the event of human origin is in fact implied by Girard’s idea that scapegoating is a “mechanism” rather than a conscious, memorable, representable event. Girard indeed sometimes talks as though scapegoating is “innate” in human nature, and here he joins ranks with the sociobiologists. Gans, on the other hand, does no such thing. For him, human violence is radically distinct from animal violence because it assumes the structure of the originary event, which is conceivable only in terms of scene of representation, with a periphery and center and all the other “fundamental anthropological categories,” including linguistic, aesthetic, sacred, and economic exchange.

Pablo: You’re actually responding to a slightly different problem than what I had in mind. I’m not really concerned with “falsifying” either Gans’s or Girard’s theory (my questions apply to both). Both Gans and Girard are of the same mind with regard to the question of whether or not their theories are “theories” in the sense of Karl Popper, and the value of a theory that cannot in principle be falsified. I completely agree with what they (and you) say about this. The fact that we have trouble reproducing an originary event in the lab does not mean that such an event did not occur.

My question is not so much about the specific details of how prehumans actually “learned” culture from their ancestors, but whether or not such a thing is possible at all, from an epistemological perspective. Gans objects to any theory that attributes human-like qualities to what would have been pre-human animals. I’m basically raising the same sort of objection. Precisely because language can only be “learned” by experiencing the originary event (as opposed to genetically or via evolution), how do prehumans that did not experience this event learn language? You say that they learned this by imitating their ancestors’ ritualistic repetition of the event. This makes sense for children or “wolf children,” who are already human, but not so much for prehuman animals. As you say, it may be possible for an animal to spontaneously have an “insight” after many, many years of repetition (I have my doubts about this, but put them aside for now). But the original members of the group that are capable of ritualistic repetition died after only a few years. I realize this is dipping into the empirical side of things, but it’s a basic fact that must be accounted for. How is it possible for a prehuman animal to “learn” culture from a member of the group in only a few years?

Whatever the exact details of the origin of humanity (or language, for you Gansians), I think we can safely say two things about it: (1) It was a single, sudden event, and (2) it involved a relatively small group of proto-human animals (even 50 or 100 animals is a small number on a planetary scale). This small group could only have survived for some relatively small number of years. These basic facts, empirical though they may be, cannot be ignored. Whether the actual originary gesture was a pointing finger or a swinging fist is secondary. Whether the ritualistic repetition of the event involved making noises or some kind of charades is secondary. Whether or not prehuman animals have the psychological capacity to learn transcendent thought, and whether or not their teachers were around long enough to do the job, are critical questions.

Richard: Actually, I did take your originary query in the sense you intended, namely, how did all those prehumans that didn’t actually participate in the originary event become human?

My answer: the originary scene spreads by diffusion, that is, in the same way all symbolic culture spreads: by new individuals participating in the ritual reproduction that “refers” back to the originary event.

You objected: “Precisely because language can only be “learned” by experiencing the originary event (as opposed to genetically or via evolution), how do prehumans that did not experience this event learn language?”

In the post-originary world, one experiences the originary event by reproducing it. That’s what ritual (and even its “minimal” form, language) does. Ritual is a form of representation of the originary event. That’s why I object to Girardians who insist that the scene of victimage doesn’t imply representation. Ritual is always representational: it represents the originary scene.

You also objected: “You say that they learned this by imitating their ancestors’ ritualistic repetition of the event. This makes sense for children or “wolf children,” who are already human, but not so much for prehuman animals.”

Well, actually it “makes sense” for anybody who can make sense of the ritual repetition, including wolf children, chimpanzees, and– god forbid–robots. That is the “anthropological” definition of the human: you are human if you are able (like Data from Star Trek, or R2D2 from Star Wars, or Kanzi in Sue Rumbaugh’s laboratory, or Prehuman X from 3 minutes after the originary event) to participate in the originary event. The human/prehuman divide falls there. If prehuman animals couldn’t “get it,” then they’re not human.

So, permitting ourselves some indulgent empirical speculation, let us suppose that a small group of prehumans (let’s call them group A) are roaming the African savanna 2.5 million years ago. They stumble on a group of hyenas who have made a fresh kill of antelope. They scare off the hyenas with rocks and sticks and screaming and such. Surrounding the kill they approach the bleeding gobs of fur and meat. But then, each sees the others’ appropriative gesture, and each hesitates for a moment. Each individual perceives the other’s gesture in a paradoxical oscillation between the central object and the peripheral aborted gestures of the others. Lo! the aborted gesture of appropriation is transformed into a sign: the deferral of mimetic conflict through representation. After this hesitation, the “division” of the kill follows in the ensuing sparagmos. But this sparagmos is forever haunted by the memory of the sign, which demonstrated to each individual that his relationship to the central appetitive object is mediated by the other’s desire. Henceforth each individual will be unable to appropriate the central object without realizing that he is participating in a social act that is mediated by the other’s desire.


Okay. So now suppose that, in a neighboring territory, another group of prehumans (group B) have surrounded and killed a hapless vervet monkey. A fight for the meat ensues, and, as usual, the alpha male (call him Conan) grabs the whole chunk of meat and settles down to his dinner, while the rest of his group stand on the sidelines begging for scraps from the mighty Conan. Conan gives a few females some meat, but he tells the other junior males to “beat it” by sinking his teeth into their begging hands. They bite back, and painfully too, but Conan will not give up his prize. Eventually, tired of the contest, they retreat to lick their wounds, and watch sulkily as the triumphant Conan eats his meal.

Now suppose that group A rape and kidnap a young female from group B. She now becomes part of their group. On their next foraging expedition, they disturb the same group of hyenas around another fresh kill (perhaps it’s time the hyenas moved on?). They scare the cats off, and move in for the meat. But this time the moment of hesitation is prolonged: the central object brings to mind the paradoxical oscillation of the previous night (the originary scene). Each individual lingers a little longer on the aborted gesture of appropriation; the memory of the previous event remains vivid in each individual’s mind. They recall the moment of delay before the sparagmos in which the meat was more equitably divided among themselves, so they delay again, perhaps bowing their heads before the meat, which now seems irresistibly to resist their desires. There is one exception: the new female recalls no such event. But, not being one used to “dominating,” she bows her head too in imitation of her fellows, and she is rewarded by the same prolonged desire to possess the meat that bleeds so invitingly before her alerted senses. Nor does this desire stop there. Once the bowing has ended, the sparagmos ensues. But, to her surprise, the female is treated as an “equal” in this sparagmos. She too gets to rip a piece from the quivering flesh. This is her reward for obeying the god of the meat. Side by side with Maximus (Conan’s equivalent in group A) she partakes of the sacred flesh. And it is precisely because Maximus remembers the miraculous success of the originary event, in which he achieved a portion of the kill without having to fight for it with his rivals, that he is willing, once again, to defer his appetite and bow his head to the meat that quivers before him.

Countless times group A repeats this structure. And, surprisingly perhaps, instead of getting weaker, they get stronger. They now have the advantage in territorial fights with group B. More and more of the members of group B are either destroyed, like the stubborn Conan who would not bow to the new god of group A, or absorbed into their religion, like the first female who gratefully accepted the “equal” piece of meat in her first “rite of passage” into this strange new community that insists on delaying before eating their meat. Group A expands its territory irrepressibly. It is now bound by a network, not of biological kinship relations, but by cultural relations: all those who participate in their “rite of passage” of delayed gratification before the meat are a part of their community.

After approximately 2 millions years of this kind of thing, a full blown symbolic culture has evolved, culminating in the “cultural explosion” of the upper paleolithic: stone tools, cave art, and burial sites are the material signs left behind of this explosion. To paraphrase Neil Armstrong: “One small aborted gesture of appropriation for group A, one giant cultural explosion for humankind.”

Pablo: Excellent story! Basically, you’re saying that, in the originary event, prehumans learn language from the other prehumans on the periphery of the scene. After that, prehumans learn language in the same way from the now humans on the reproduced periphery of the scene. Every time the originary event is reproduced and repeated (or maybe almost every time) new prehumans effectively experience the originary event by experiencing the reenactment of that event.

Now, let me step back from the empirical details again… There is, I think, a basic assumption that, whatever the precise details of the originary event itself, it was a big deal. That is, it was something special and unlikely, otherwise it would have probably happened more than once in more than one place. The aborted gesture wasn’t just any old gesture–it was more than just two animals butting heads–it was the result or culmination of some highly specialized set of conditions that do not normally exist in nature. Remember, we are not just talking about a few animals learning a new trick, even an impressive trick like a new gesture. We’re talking about animals acquiring the capacity for transcendent thought, without which language is impossible.

It therefore seems to me a little strange that, while a successful originary event is so difficult to achieve, the mere reproduction of that event, which is not very special, is equally successful. To put it another way, if you strip away all the elements and parameters of the originary event and reduce it to a single gesture, as important as that gesture is, do you still have the originary event, with all its power? Even if we accept the notion that this type of “learning” can happen over the course of many years, it’s hard to believe that it can happen during the few years that the original humans are alive. Moreover, it must have happened with extreme efficiency (literally almost every time the originary event is reproduced) to have maintained an increasing number of humans from generation to generation.

I don’t mean to belabor the point. Quite honestly, it is this sort of thing that makes a purely naturalistic explanation of the origin of humanity unconvincing to me. Ultimately, the creation of transcendent thought from immanent mechanisms just doesn’t seem logical. It is the same problem as trying to explain the origin of life itself from purely naturalistic means. You can create a plausible “scene” involving a puddle of primordial goo, some lightning, and–voila!–life is born. But this doesn’t really explain anything–it’s just a possible setting for the “scene” of the origin of life. To put it another way, imagine assembling a human body from healthy organs and bones. Each individual body part is intact and used to belong to some living body, and you put it all together so that each piece down to the last capillary is connected with perfect precision. Do you end up with a living person? No; you end up with a pile of well-connected body parts and nothing more. It takes something beyond the natural mechanisms available to us to create life from scratch. It takes something above life to create life. In the same way, I think it takes something transcendent to create transcendent thought–something above humanity to create humanity.

Andrew: This implication of the supernatural makes me nervous.


Pablo: I’m afraid it’s unavoidable. Anyway, the existence of the supernatural doesn’t eliminate or invalidate the natural.

I ask the question, “If you strip away all the elements and parameters of the originary event and reduce it to a single gesture, as important as that gesture is, do you still have the originary event, with all its power?” You may very well answer “Yes, Pablo. The critical event in the originary scene is the aborted gesture. It was difficult for this gesture to occur accidentally in nature, but once it did then the deliberate reproduction of it was straightforward.”

What bothers me about this is the notion that a single gesture, so easily reproduced and independent of any special context or specific scenario, could be responsible for the creation of transcendent thought. Basically, if it’s that easy to turn an animal into a human (once you know the trick), you would think we’d be able to do it in a laboratory.

Richard: Essentially what you are arguing is that, given the tremendous difference between symbolic culture and all other presymbolic systems of social organization, it’s impossible to believe that this difference can be reduced to one miserly little “aborted gesture of appropriation.” If the originary event really is the origin of “transcendence” then let’s not hold back, let’s talk about transcendence in terms that are suitably transcendent.

I sympathize with your skeptical transcendentalism, but the minimalism of the originary scene is the prerequisite of any fundamental anthropology that wishes to avoid the designation “myth of origin.” If you start loading the scene up with all kinds of figures (e.g., Girard’s human victim), you no longer have a minimal anthropology. What you have is a ritual/mythical reproduction that runs quite counter to the minimalism of the originary thinking.

But I don’t think that the aborted gesture as a symbolic sign can be reduced to the notion of a simple trick that any bright hominid could come up with. The point of course is that the symbolic sign is not a “trick” (like a chimp reaching for a banana with a long stick), but a condition of highly unstable mimetic scene: the minimal mimetic triangle or originary scene of representation. I therefore disagree with you when you say that you can “strip” this scene down to the aborted gesture without losing anything. The aborted gesture exists as a sign only within the context of the scene as a whole. That’s why the indexical signs produced by nonhuman primates don’t count as language.

Of course the reason why chimps don’t produce symbolic culture is that mimetic conflict never presented the kind of absolute problem that it presented for our hominid ancestors. As Gans says, the human may be most easily defined as that species for which internal conflict, rather than external conflict with the environment, presents the biggest problem. You only have to look at the dismal situation of chimps today to see the truth of this. Chimps are endangered because of us, not because of themselves.

The other point is that you seem to be confusing the originary scene with the reproduction of that scene. You seem to be assuming that if the probability of the originary event is low, it follows that the probability of its reproduction is equally low. But that doesn’t follow.

An analogy: Think of one of those mind-bending puzzles that asks you to see the hidden figure in a picture.

Chris: “Magic eye” puzzles! Oh, I hate those!

Richard: Why?

Chris: I can never see the hidden figure!

Richard: Okay, you just can’t see it. Then suddenly (be patient, Chris) you see it. After that, it’s impossible to miss. Now you “reproduce” the original “insight” time and again without difficulty. Perhaps you even show someone who can’t see it how to see it. Indeed, perhaps that’s how you learnt to see it in the first place, i.e., by someone else showing you how to find it.

Now extrapolate back to the originary event. Before the event, the object was just another piece of meat. During the event, the meat becomes the paradoxical object of the originary sign, the first figure is an infinite series of such figures. Voila! the generative scene of origin. The first “originary” moment provides the crucial difference.

Andrew: This is great. I feel like this dialogue is helping us see the hidden figure on the originary scene…

Chris: Pablo, the thing to remember concerning your objection–“if it’s that easy to turn an animal into a human (once you know the trick), you would think we’d be able to do it in a laboratory”–is that the origin of language is a “little bang”. That is how Gans describes it (see

Pablo: The “little bang” analogy makes sense to me. My point is that the humanizing event, originary or reproduced, must be a “bang” for each and every hominid that experiences it. It sounds like, for every subsequent repetition of the originary event, Gans keeps the “little” but takes out the “bang.”

Richard: I see what you’re saying Pablo. From the point of view of those participating in the postoriginary ritual reproductions of the event, these reproductions cannot be considered as mere supplements. On the contrary, they are indispensable to the act of sacralization. From an internal viewpoint, there is no difference between the originary event and its repetition. Each repetition is originary, at least to those doing the reproduction (which hence would not be considered a “reproduction” at all).


But Gans is speaking from the theoretical or epistemological viewpoint of generative anthropology. From the point of view of the latter, the minimal difference takes place at the very beginning, in the aborted mimetic gesture of the originary scene.

Remember also that–anthropologically speaking–the key to the success of the originary event is not individual but collective. Once the collective scene (which requires at least two individuals) is established in the crucial originary event, introducing subsequent “newcomers” into this now-existing collective scene is an event of a quite different order.

The analogy would be to the “rites of passage” that mark an individual’s life: birth, adulthood, marriage, death. These are “little bangs” for the individual, but they assume the existence of the collective originary scene. Individuals don’t invent culture. It preexists their existence. The exception, of course, is the originary event, in which at least two individuals invented, or more accurately, discovered the “deferral of violence through representation.” Once this “virtual” scene has been discovered, it becomes a question of “teaching” it to others.

But I agree that in the immediate postoriginary era, subsequent ritual reenactments of the originary event would be rather spectacular. The best way to persuade others as to the efficacy of the originary event is to “overengineer” it. Ritual always takes place as an overengineering of the originary.

Chris: So does this mean that the children of the first humans do not become human themselves until they are taught how to be human?

Richard: Yes, insofar as participating in the ritual scene is a form of “teaching.”

Chris: In other words, are you saying that ontogeny doesn’t recapitulate phylogeny until the kids learn how to be human from their parents’ rituals?

Richard: Depends which children you are looking at. Children of modern Homo sapiens“recapitulate” phylogeny, in the sense that their ontogenetic development assumes the existence of a huge amount of coevolutionary or “phylogenetic” prehistory (roughly, from Homo erectus to Homo sapiens, or about 2 million years.) But the children of the first humans don’t have any human evolution to recapitulate. The only thing they can “recapitulate” is the originary event. So that’s what they do.

Chris: This doesn’t seem right to me. Don’t human kids have the capacity to form abstract ideas even before they learn about some communal sparagmos feast?

Richard: Depends what you mean by “abstract ideas.” All “higher” animals are capable of cognitive categorization. Monkeys have an “idea” of kinds of predator (e.g., leopard, snake, or eagle). And they can even emit calls that associate the call (= indexical sign) with the “idea” or category of a particular predator. But some ideas are not a question of “abstraction” from “natural kinds”: e.g., “anthropological” categories like the self, desire, morality, etc. These are learned only by participating in a human culture, which finds its historical origin in the originary scene.

Chris: How can someone be taught about a “virtual” scene if they do not already have the capacity to abstract ideas on such a scene?

Richard: The difference is between the idea as a form of cognitive abstraction or categorization and the idea as a signified or symbolic category. Anthropological ideas (the self, the moral, the good, the beautiful) exist only as “abstractions” from the originary scene. For example, the idea of the beautiful is not a category that can be derived solely from theories of sexual selection: e.g., male cognition is “selected for” admiring young females, etc. I don’t doubt that there are deep-rooted genetic factors that play a role in anthropological categories like (female) beauty: a philosopher might say these are necessary but not sufficient conditions for the anthropological idea of beauty. The sufficient condition is symbolic abstraction, which requires the minimal mimetic configuration of the originary scene. As Deacon’s data and research confirms, symbolic abstraction is not something that can be genetically assimilated. Modern human ontogeny is evolutionarily selected for a predisposition toward symbolic abstraction, but without the “virtual” infrastructure of culture, the kernel of which is the originary scene, the child will not acquire symbolic “behavior.”

A thought experiment: Imagine the entire human species is suddenly wiped out by a virus. One human baby survives. She is rescued by some chimpanzees who raise her as one of their own. (Yes, this is “Planet of the Apes” for philosophers.) Though she is genetically predisposed for symbolic cognition, her new situation never encourages her to develop this capacity. As her brain matures through childhood and adolescence, all those brain cells that would normally be recruited for associating a signifier with a signified during language acquisition would be recruited for chimpanzee behaviors, which is to say for non-anthropologically specific categories (e.g., indexical alarm calls, grooming behaviors, assessing the landscape, tool use etc.). No doubt she would prove exceptional at certain tasks (e.g., tool making), and rather poor at others (e.g., smelling predators). But the point is that she would never acquire symbolic culture because there is no existing infrastructure for her to “learn” it.

Pablo: Chris, it sounds like you’re thinking along similar lines as I am with respect to the originary scene (although you guys have indeed corrected my thinking on a few things). In response to your question, I think Richard may point to the fact (so to speak) that human kids are biologically conditioned to “learn” culture after millions of years of repetition of the originary scene, but ultimately the roots of that conditioning go back to that scene (sorry to put words in your mouth, Richard). But I think your last question, Chris, really gets to the heart of the matter: “How can someone be taught about a ‘virtual’ scene if they do not already have the capacity to abstract ideas on such a scene?” This is a better rephrasing of my original simple question, “Is it really possible to ‘teach’ an animal to be human?”


The sticking point is that this “learning” must be done in a conscious way (this is an important requirement for Gans). I, too, have trouble with the notion of a pre-human animal recognizing a deliberate gesture as an abstract representation. In the case of the originary event itself it may be possible due to the particularly dramatic and unlikely accumulation of parameters that make it such a special event in the first place. But to obtain the same effect in subsequent reproductions of the event seems strange.

This ultimately goes back to the concept of “transcendence” as the defining human characteristic, from which language comes. This may be the main divide between Girard and Gans–Gans situates the origin of the human in the origin of language, which must be a consciously recognized event. Girard situates the origin of the human in the origin of transcendent thought, which is the first recognition of the Other as Other, and this most likely occurred as the result of an unconscious mechanism.

I was hoping to get your thoughts on a couple questions that are still on the table…

Gans says that the aborted gesture is necessarily followed by the peaceful distribution of the object, without which it would not last in the memory of the group. The sparagmos, by contrast, is a violent release of mimetic tension that also follows the aborted gesture, and results in the violent tearing apart of the object. How do both of these coexist in the originary scene?

The second question is more technical. Gans says that in order to make the originary event last in the memory of the group, it had to have directly led to something that resulted in more appetitive satisfaction than “that which had previously been available.” Presumably, the peaceful distribution of the appetitive object fulfilled this function by providing every member of the group with some of the object, rather than no one getting any of it. But isn’t it true that each member of the group was originally hoping and trying to get all of it, rather than just some of it? The thought of getting none of it probably never entered anyone’s brain. Each animal was going after the object–the whole object, with no intention of sharing. So, by comparison, getting only a piece of it would be less satisfying, not more. Doesn’t this contradict Gans’s requirement?

Andrew: Well, time’s up. Maybe all your questions weren’t answered, but we certainly addressed more than none. If you find this communal ritual strangely satisfying–despite the fact that you got less rather than more–let us meet again soon for our next Sparagmos!

This Sparagmos! dialogue was compiled and edited by Christopher Morrissey.

Sparagmos! is a Vancouver-based reading and discussion group that meets regularly to discuss the mimetic theory of René Girard and the generative anthropology of Eric Gans.

Interested readers should contact Andrew Bartlett ( about participating. Andrew has been organizing and hosting the Sparagmos! meetings in Vancouver, BC.