According to a recent article by Eric Gans, the question of the origin of language, after a century-long ban by the disciplines of history and linguistics, and a decades-long interdict as a non-question by the deconstructionists, is once again receiving attention from the academy. "Today," states Gans, "there are dozens of books and thousands of articles. The Language Origins Society or LOS will hold its fifteenth annual meeting this year. Language origin study has become an interdisciplinary subfield of the human sciences." In this renewed pursuit of the question, the humanities, however, continue to be relegated to the sidelines by sciences like biology and neurology, which view human language as the product of instinct and evolution and favor a gradualist view of its emergence. Against these, Gans, citing the historical record, particularly that of this century, argues that mimetic theory provides the best source of insight into the origin of language because it puts in relief the primary feature of human culture: its propensity to degenerate into internecine violence. Before language could have served as a means of communication, he reasons, its first function must been to mitigate violence within the group. Otherwise there simply would not have been a human culture.

In this paper I will not attempt to assess the scientific plausibility of mimetic theory’s contribution the language-origin question, but will compare its two main versions as formulated by René Girard and Eric Gans. My purpose is not to champion one against the other, but rather to articulate some of the issues at stake between them. To this end I have found that Kenneth Burke’s independent speculations on this question offer a useful third viewpoint on the question, in that they offer a means to formulate and negotiate the differences between Girard and Burke. At this point in the development of mimetic theory, it has perhaps reached the point where rather than an orthodoxy it can be seen as a hypothesis that can be fruitfully unfolded in a variety of modes, for as René Girard has frequently declared, the mimetic factor in human behavior is by no means his exclusive discovery. Rather, again as he has suggested, it is to be found revealed in foundational religious texts and in major works of world literature. Girard’s particular exploit was to develop his insight into the mimetic aspect of individual behavior into a general theory of human culture. By taking the quantum step from individual (or "interdividual") to group psychology, he fashioned a flexible hermeneutic susceptible to an open-ended process of refinement and development. Thus in this paper, I will begin with Girard's account of the origin of language, and then juxtapose it to Gans’s account, and finally look at both of these from the vantage of Burke’s speculation on the origins of language.

Girard and the Origin of Language

In the section entitled Fundamental Anthropology in Things Hidden Since the Foundation of the World, René Girard distinguishes several steps in proto-humanity’s accession to language. The first step occurred, he writes, when the increase of mimetic rivalry in the group had exceeded the point where it could be contained by dominance patterns:

Why, one might wonder, after he has plausibly pursued the process which leads from the situation of hyper-mimetic violence to the signifying function of the body of the victim, does Girard take care to distinguish this signifier from the true transcendental signifier? As the factor in the scene that embodies difference, from which all articulated meaning in essence derives, does it not thereby "transcend" all the other elements in the scene? Let us leave the question unanswered for the moment and turn to Eric Gans’s narrative of a similar scene of origin, similar insofar as it is faithful to Girard's notion of mimetic conflict, but divergent in its view of the effect of this conflict’s consequences.

Eric Gans and the Origin of Language

Of the many accounts Gans has given of the origin of the human sign, I am quoting the one from Originary Thinking: Elements of Generative Anthropology (1993) because of its succinctness. Gans invites us to imagine that:

In sum, from the identical cause, i.e., mimetic rivalry, two distinct scenarios unfold. On the one hand mimetic rivalry leads directly to violent conflict which generates the sign, on the other it leads to a moment of stasis which generates the sign. What are the stakes involved in these competing scenes? First and foremost there is the question of verisimilitude, of likelihood. But to try to decide on the basis of our present sensibilities which of these scenarios was more likely to have occurred at the dawn of culture is a fruitless task. Whatever their scientific aspirations, both Gans’s and Girard’s accounts are primarily useful as heuristic narratives not unlike Freud's in Totem and Taboo. Of the two, Gans seems less interested in fictional plausibility and more in the rigor of a logical argument, for he places only those elements in the scene that he feels are strictly necessary for later extrapolation as culture unfolds in history. Girard, on the other hand, seems more interested in verisimilitude. If violence is the keynote of our present existence, both in the form of end results and efficient cause, then perhaps it served the same functions at the start.

Against this, however, Gans argues that Girard's scenario commits the logical error of positing a signifier prior to signification, as if there might be such a thing as a word before there was such a thing as language. Gans asks how Girard’s reconciling victim could perform its cultural function prior to the advent of human culture. Summarizing Girard's scenario, Gans concludes:

Only when the sign has been brought into being through general "ostensive" attention can the ensuing violence and its cessation start to take on meaning for humans.

Such notional sharing, no matter how brief, allows for the object to be "possessed" by all equally. Thus, in an important alternative to Girard's notion of the scene, Gans allows that the object in dispute could have been equitably shared. Its apportionment could have been peaceably negotiated. Although the potential for a frenzied "sparagmos" must always have been great, the primal scene might conceivably have unfolded in perfect peace. For the essence of the human signifier, as Gans never tires of stating, is to defer violence through representation. Thus for Gans the human signifier is truly "transcendent" in a way that it never is for Girard. In rises "vertically" above the "horizontal" object by replacing it and making itself equally available to all.

Against Gans’s logical claim that the signifier cannot precede language, Girard had suggested, as we saw, that there might indeed have been something like a pre-linguistic signifier. We saw in Girard's view of the hominization process that the body of the victim did indeed become significant, in the sense that it was both differentiated and a source of difference, before it emerged as a full blown sign, i.e., before it was abstracted to the level of a repeatable immaterial signifier. Girard suggests that ritual, with its roots in prehuman animal societies, must have functioned as the bridge between collectively meaningful routines and genuine human signifiers. As protohumans repeatedly extracted themselves from the chaos of mimetically induced aggression through the expedient of the killing of a victim, the functional significance accruing to the process must have "stuck," so to speak, in the habitual repertory of a group’s resources before making its way into the clear awareness of anyone in particular. Perhaps under threat of renewed violence a group might have proceeded to a "ritual" reenactment of murder, either in fact or by proxy, without being driven to it "as if for the first time". In short, although for both Girard and Gans the gap between nature and culture is one of radical difference, Girard nevertheless appears to be open to the notion of a temporal period of preparation when the victim could have carried out its meaningfully pacifying role before their was a fully conscious medium capable of accomplishing this mediation. Time is an important factor here. As Girard points out, the process took place over millions of years:

Rather than simply being a question historical verisimilitude, i.e., of which version seems more likely to have actually occurred, I would suggest that point at issue here concerns divergent understandings of the mechanics of violence. Both Gans and Girard agree that violence springs from mimetic desire, but part company concerning the role and function of the desired object. For Girard, once the conflict has been engaged, the object that served as its source tends to lose importance and fade from the consciousness of the disputants:

For Gans, on the other hand, it is not the model/obstacle but the object that remains primary. All the energy in his model is located at the center of the scene, where it remains, even in the absence of the object. Thus for Gans the generative energy of the mimetic situation is not overt violence but covert resentment. In his scenario, the crucial moment is when the object has assembled the group around it and fixated its attention.

Rather than try to adjudicate the difference between Girard and Gans, let us now turn to the independent viewpoint of Kenneth Burke, which will serve to suggest that perhaps the clue to their differences has to do with their frames of reference. As we stated, René Girard's starting point was individual (or interdividual) psychology (ontogeny), which then shifted to anthropology. Gans has proceeded in the inverse direction, beginning with anthropology (phylogeny) and using its framework to understand the individual.

Kenneth Burke and the Origins of Language

One of the reasons why the writings of Kenneth Burke offer an interesting vantage on the differences dividing Girard and Gans is because he is less consistent than either. He clearly definitely a "fox" of theory to their "groundhogs," to make use of Isaiah Berlin’s famous distinction between the two basic types of theorists. Burke coined the terms "logology" and "dramatism" for the kind of rhetorical/cultural criticism he engaged in, which we might understand as a sort of meta-rhetoric and ritual analysis respectively. Burke locates the focus for his investigations neither in the psychology of interpersonal human relations (like Girard) nor in the anthropology of the group (like Gans), but rather in both, which in turn are viewed from the perspective of language as such. That is to say that Burke takes language as a given, not in the sense of a transcendent given, but as a historical cultural entity animated by its own suprapersonal dynamic. Looking at the difference between Girard’s and Gans’s view of the origin of language from the vantage of Burke, one can sense the importance played by their primary frames of reference. Burke’s starting point is not so radical as Girard's or Gans’s but it is akin to theirs. He too is concerned with the violence that characterizes human culture and attempts to find a heuristic model with which to understand it. Rather than contrive one, he finds his model in drama, an aesthetic form which offers a partial glimpse of its scene of origin in ritual. Thus in his own way he accomplished a parallel move to Gans and Girard, except that for him his model could serve equally well as an "originary" or "culminating" hypothesis.

Let me briefly review Burke’s demonstration from his essay "A Dramatistic View of the Origins of Language and Postscripts on the Negative," in his volume Language as Symbolic Action.

Taking his cue from Bergson (and Hegel) that the negative is nowhere to be found in nature, each thing or situation being positively what it is, Burke concludes that the negative is purely linguistic. In fact, since the negative it is only to be found in language, it can in turn be taken to language’s distinctive feature, as well as a clue to its origin. As Burke conducts his argument he moves back and forth between historical and "cumulative" perspectives. "Cumulatively" speaking, says Burke, the negative reveals itself most fully in the Decalogue, the "thou shalt nots" regulating the moral and ethical behavior of humans. Hunting for clues to the negative’s historical origin, Burke finds them in the "negative" behavior of animals, which, of course, is not really negative, but a kind of forerunner. Grunts of rejection or disgust, acts of aversion, acts of flight, these, although again they are simple positives taken in and of themselves, give evidence of acts of discrimination that can later "flower" as negative judgments in human language. The human body of course is the link to the animal sphere. And it too, says Burke, with its visceral rejections and aversions stands as source of negation that precedes the negative of language.

In a move surprisingly akin to Gans, Burke places the true human origin of the negative, and hence of language, in a kind of negative ostensive. For Burke, it would have to be a verb, a sound, and not, as with Gans, a gesture, but it is a sound that both "points" and is full of implication, just as does Gans’s mute gesture:

My purpose in this essay has been to bring to light some of the seeming contradictions in mimetic theory by considering Girard's and Gans’s heuristic descriptions of the scene or process that gave rise to human language, and to offer a fresh perspective on them by bringing to bear Kenneth Burke’s reflections on the same topic. I think Burke helps us to understand that it is their frames of reference rather than a fundamental disagreement that separate Gans and Girard. In turn, Burke’s kindred but independent thinking might perhaps be fruitfully incorporated into mimetic theory. In addition to his views concerning the evolutionary process of language development and its hypothetical scene of origin, Burke never forgets the presence of the non-human animal and the human physical body as conditioning factors that continue to operate in all aspects of human culture. In short, he offers a perspective or a method which bridges the gradualist and punctual view of the scene of origin of language. And finally, as I hope to have indicated, his flexible method of analysis, moving from "logological," (proto-Gansian) to "dramatistic" (proto-Girardian) perspectives as the argument requires, offers a model for deploying mimetic theory’s basic concepts in a creative and productive manner.