The Origin of Language Read online

  The Origin of Language

  A New Edition

  Eric Gans

  Introduction by Adam Katz

  Spuyten Duyvil

  Editor’s Introduction:

  The Origin of Language and the Anthropological Imagination

  Foreword

  Chapter 1: Introduction

  Chapter 2: The Linguistic Dialectic

  Chapter 3: A Derridean Parenthesis

  Chapter 4: Formal and Institutional Representation

  Chapter 5: The Ostensive

  Chapter 6: Linguistics of the Ostensive

  Chapter 7: The Imperative I

  Chapter 8: The Imperative II

  Chapter 9: Elementary Forms and Grammatical Structures

  Chapter 10: The Fundamental Asymmetry of the Speech Situation

  Chapter 11: Imperative Dialogue

  Chapter 12: Dialectic of the Imperative (II)

  Chapter 13: Negation as Predication: The Origin of the Declarative

  Chapter 14: The Declarative Model

  Chapter 15: The Esthetics of Linguistic Forms

  Conclusion

  GA Bibliography

  About the Author

  Copyright Information

  Introduction:

  The Origin of Language

  and the Anthropological Imagination

  In this new edition of The Origin of Language, originally published in 1981, Eric Gans provides a hypothesis for the origin of language, and then extends that hypothesis to account for the most fully developed linguistic form, the declarative sentence. In doing so, he shows how the earliest human group would have gone from using a single, “ostensive” sign, to a variety of ostensive signs, to the use of imperatives and then, finally, the declarative sentence.

  It is astonishing that no one had ever considered the necessity of doing something like this, much less attempted it, because how else could one have imagined that humans would have gone from no language at all to such complex linguistic structures as the subject-predicate relation comprising the declarative sentence? Gans shows how the sequence he follows here is the only plausible way of imagining the evolution of the earliest linguistic forms, and he does so by analyzing the sequences in intricate detail, addressing everything that is necessary to account for linguistic evolution, and nothing that isn’t. The Origin of Language remains as completely original and unprecedented (and intellectually demanding and satisfying) today as when it was originally published, so much so as to constitute a kind of intellectual scandal.

  In all of the decades in which immense intellectual energy has been put into “dismantling” and “deconstructing” metaphysics, hardly a single one of our leading thinkers has found it necessary to address the startlingly simple definition provided by Gans: metaphysics is the assumption that the declarative sentence is the primary linguistic form. And if you assume that the declarative sentence is the primary linguistic form, you will never think to ask, or to think one can ask, whence it derived—even if a moment’s reflection must convince us that it must have derived from some previous linguistic form. If you don’t ask these elementary questions, you remain within metaphysics, regardless of how you have “implicated” it in power relations, or Eurocentrism, or technology, or whatever. Gans, in this book, stays focused on the dialectic of the linguistic forms, without many side comments on metaphysical obfuscation, because how could he point out that every moment in that dialectic is obscured by metaphysical assumptions and still get on with the only work that could really clear out those assumptions?

  I hope that the effect of this publication of a condensed edition of Eric Gans’s The Origin of Language, originally published by the University of California Press in 1981, will be to make Gans’s work visible. I don’t just mean “visible” in the sense of more widely available, or more effectively publicized, although those things would be nice as well. I mean “visible” in the sense of no longer occluded, conceptually, by the reigning “language games” and “problematics” in the human and social sciences. We have known for a long time that inquiry takes place within a Gestalt, or frame, that makes some things visible, other things obvious and other things invisible or unthinkable. That’s why science advances through “epistemological breaks” that don’t just correct previous mistaken conclusions, but reorganize the entire field on conceptual terms. For there to be any possibility that this will happen, though, discoveries must be made that, as Michael Polanyi argued, “must be not only true, but also interesting, and more particularly, interesting to science” (66). What is “interesting” is generally what continues along the paths already laid out by scientific authority and tradition, reinforced institutionally in many ways.

  Polanyi provides an example of how scientific theories get sorted out into “interesting” and “not interesting” from the reception of “a paper by Lord Rayleigh, published in The Proceedings of the Royal Society in 1947”:

  It described some fairly simple experiments which proved, in the author’s opinion, that a hydrogen atom impinging on a metal wire could transmit to it energies ranging up to a hundred electron volts. Such an observation, if correct, would be far more revolutionary than the discovery of atomic fission by Otto Hahn in 1939. Yet when this paper appeared, and I asked various physicists’ opinions about it, they only shrugged their shoulders. They could not find fault with the experiment, yet they not only did not believe its results, but did not even consider it worthwhile to consider what was wrong with it, let alone check up on it. They just ignored it. About ten years later some experiments were brought to my notice which accidentally offered an explanation of Lord Rayleigh’s findings. His results were apparently due to some factors of no great interest, but which he could have hardly have identified at the time. He should have ignored his observation, for he ought to have known that there must be something wrong with it. The rejection of implausible claims has often proved mistaken, but safety against this danger could be assured only at the cost of permitting journals to be swamped by nonsense. (65)

  What makes a claim “implausible” is its misfit with the prevailing research program, and Polanyi’s example is of a case where that misfit turned out to be a reliable indicator of the observation’s falsity. But if such rejection proves mistaken, what remedies are available? In the physical sciences, we might be able to expect self-correction and what Polanyi calls “mutual control” to maintain steady progress towards a more truthful and comprehensive understanding of the natural world; at any rate, it serves no purpose of mine to contend otherwise here. In the human sciences, though, “interesting” is a far more loaded term, as is “implausible,” since these sciences contain far more disagreement regarding the basic assumptions of what constitutes a legitimate research program, and are far less able to control for (and perhaps should not exactly “control for”) the moral, ethical, religious and political convictions of the inquirer. The only remedies here lie in pointing out the sources of the refusal to grant sufficient “interest” and “plausibility” to an observation to make it worth one’s while to even “check up on it.” These sources would be “obstacles” to scientific thought, of the kind Gaston Bachelard “psychoanalyzes” in his The Formation of the Scientific Mind. In this case, the identification of such obstacles would not only aid in making the more genuinely scientific theory more visible, but would serve as a demonstration of the power of the theory. I will right away give a name to the primary obstacle, which includes the others I will mention below: the anthropomorphism of the human, that is, the explanation of human activity by qualities that rely upon the prior constitution of the human, which is in turn simply taken as given. In other words, anything we could say about human beings always already presupposes beings capable of l
anguage—this capability is therefore retrojected back to the earliest models of humanity we can imagine, making entertaining an originary hypothesis that would have the human and language emerging together extraordinarily difficult.

  Eric Gans’s “originary hypothesis” of the origin of language runs up at every point against the anthropomorphism of the human. I think the most obvious source of opposition is the prevalence of what Gans has examined very thoroughly over the past several decades, and has recently described (in his online Chronicle of Love and Resentment 563, “Victimary Humanism”) as “extreme humanism”: what he has termed “victimary thinking.” Victimary thinking, as a product of the twin iconic markers of World War II, Auschwitz and Hiroshima, frames all inequalities as instances of “oppression,” to which the Nazi-Jew model can be applied, in a social and technological environment in which the potential for violence is virtually unlimited. For victimary thinking, although this may be expressed with greater or less explicitness, any “ascriptive difference,” that is, any “inherent” difference between groups, portends unacceptable levels of violence. And we have more recently discovered (as Gans shows in the aforementioned Chronicle) that any difference can ultimately be framed as an “ascriptive” one, that is, attributed to racism, sexism, homophobia, transphobia, Islamophobia, and so on. Gans’s hypothesis, meanwhile, locates the origin of language and hence of the human in the unanimous acceptance of the sign as the deferral of violence on a scene of extreme mimetic danger. There is no room for “ascriptive” differences there, no way of conceiving of the scene as an “exclusionary” one, even if it is possible to conceive of it as an “uneven” one, in which the sign is repeated in different ways at different paces by the members on the scene. So, the originary hypothesis provides no foothold for victimary thinking—indeed, it seems unlikely that contemporary thinkers of the victimary could even find much use in evoking the egalitarian hunter-gatherer communities that once inspired the left, insofar as the various “exclusionary” elements of those communities are all too evident. If one were to devise a theory directly aimed at occluding the originary hypothesis, one could not do better than victimary thinking, which is perhaps why Gans has found it “interesting” and, perhaps, in its own way, “plausible”: victimary thinking isolates and thereby brings into heightened focus a single moment of the originary scene, the equality or symmetry of all before the sacred center, while obscuring all other pertinent elements of the scene, such as the existence of the center itself.

  The anthropomorphism of the human multiplies differences within the human, while considering the difference constitutive of the human beyond the realm of plausible inquiries. It’s as if as soon as the question, why do humans engage in rituals? is raised, the question is shut down by pointing to yet another ritual, different from all the others, that has been discovered. A good example of such thinking, which in its mixture of blindness and insight will also help me to clarify the conditions under which the originary hypothesis might be found both highly plausible and of the greatest possible interest, is the anthropologist C.R. Hallpike’s “Rene Girard’s World of Fantasy” (from Hallpike’s website: hallpike.com). As the title suggests, the essay is intended as a “debunking” of Girard’s theory of mimetic violence, a theory which lies at the origin of Gans’s own originary hypothesis. Hallpike begins by restating Girard’s theory, which “begins with the premise that all human behavior is learned, and is therefore based on imitations,” and goes on to assert that mimesis leads to violence because individuals imitating each other converge on the same object of desire, and concludes by arguing that the resolution of this mimetic violence is the arbitrary selection of a scapegoat to be sacrificed as the “cause” of the violence (1-2).

  Hallpike begins his critical examination by questioning this last part of the sequence, noting with surprise that Girard seems to assume that the original act of communal violence

  took place far back in prehistory before humans acquired language. The first scape-goating ritual, being pre-linguistic, was simply based on instinct, and since scape-goating is the substitution of one thing for another it is also the origin of language, since words themselves are also substitutes for things. Sacrifice and the prohibitions associated with it would have created communal peace for early hominid groups and a safe space for mothers and their babies in particular.

  The victimising process was therefore the missing link between the animal and human worlds that explains the humanisation of primates, and hunting and the domestication of animals were also motivated by the need for a stock of sacrificial victims. Scape-goating and sacrifice are the basis of all ritual and archaic religion generally, and archaic religion is the basis of all political and cultural institutions. Girard claims that the victimisation process is the rational principle that explains the infinite diversity of culture, and compares it to the principle of natural selection, which cannot be proved experimentally but convinces us by its great explanatory power. (2-3)

  So far, Hallpike might be arguing that Girard’s hypothesis is insufficient to account for the emergence of language and hence to initiate the humanization process. In that case, there would be some overlap between Hallpike’s critique of Girard and Gans’s. But the “substitution of one thing for another” cannot be the origin of language, because words are not “substitutes for things”—for one thing, how would anyone know that the first word or sign was a substitute for a thing? They would already have to have language: joint attention mediated by a sign. Hallpike also makes a point of mentioning Girard’s claim that “experimental proof” cannot be expected with regard to a hypothesis of origin, but we must rely on its “great explanatory power” (evidence of which it already seems Hallpike does not believe can be forthcoming). Hallpike continues:

  Girard’s belief that scape-goating could have been the source of language because it involves the substitution of the arbitrarily chosen victim faces two major problems, the first of which is a simple matter of evidence, or rather the lack of it. We simply know nothing about the thought processes of early hominids such as Homo erectus. Nor can we imagine what the social relations of pre-linguistic Homo sapiens might have been either, and attempts to do so are pure speculation. Indeed, we actually have no direct evidence for when grammatical language emerged. By ‘grammatical’ I mean, for example, predication – the ability to say that something or someone has certain qualities; distinguishing between acting on and being acted upon; questions; negation, and referring to past and future. This raises the second problem, which is that it is hard to see how any symbolic culture would be possible at all without language. This is because the relation between a symbol and what it stands for, while drawn from nature, is not a representation of it. For example, among the Konso of Ethiopia, white is an inauspicious colour, but without language how could a group of people decide that white rather than black or some other colour should be regarded as inauspicious? (Indeed, how could the very idea of ‘inauspicious’ come to be understood by a group of people without language?) In fact, the Konso regard white as inauspicious because it is the colour of bone, of death, therefore, and also the colour of cotton, which ripens during the hottest and driest part of the year. Black, on the other hand, is the colour of the life-giving rain-clouds and is therefore auspicious. But these are simply one set of symbolic values and other cultures have chosen different ones. In short, Girard does not explain how symbolic culture could have existed in a pre-linguistic society. (3)

  Hallpike immediately focuses on the question of evidence, but again there are a couple of possibilities here. We don’t know what the thought processes or social relations of pre-linguistic hominids might have been. Hallpike seems to have in mind the kind of pre-linguistic hominid that has already gone through Girard’s scapegoating event, but has not yet completed the process of acquiring language, or what we could recognize as language, as a result of that event. Fair enough, but what, exactly, do we need to do here to assess, first of all the plausibility, and eve
n probability, of Girard’s account?

  Gans’s originary hypothesis is predicated upon a gap in Girard’s account very similar to the one Hallpike identifies here, but Gans’s critique addresses the nature of language and its constitution of the human: the originary event could not be significant without being memorable, and it couldn’t be memorable, which is to say it couldn’t be repeated, without a sign. There is no sign in Girard’s account. What does Hallpike think language is? Well, he knows that language is necessary to attribute qualities to objects, to describe relations between objects and actions, to publicly distinguish between the more and less preferable. And it is certainly true that without being able to do these things, the ritual culture Girard imagines could not have existed. But Hallpike seems uninterested in the question he has implicitly raised here—we might say he cannot even see the question: how, exactly, did it become possible for humans to do all these things? What interests Hallpike, and what we can say interests the modern anthropological imagination, is all the different ways humans have found of doing it. We could say this is a primary obstacle to seeing the originary hypothesis: the principled, constitutive fascination with “diversity” (not, of course, in the sense this takes for a contemporary HR department) precludes an interest in the originary.

  I would suggest that it is that fascination which prevents Hallpike from considering that we do know quite a bit about the “social relations” of pre-linguistic, that is, pre-human hominids, and how they differ from even the most primitive human societies of which we are aware. At the very least, hominids have a pecking order determining access to food and mates; no such pecking order exists in humans. The coordination of a group against an individual is possible for human, but not for animals; hierarchies among humans are organized institutionally, not established through one-on-one confrontations. Are we forever barred, due to “lack of evidence,” from constructing plausible accounts of what replaced the pecking order, and what role language might have played in the transition? Aren’t some “speculations” along these lines going to be of greater explanatory and heuristic power than others, and couldn’t we make those yet more powerful? You can only answer such questions in the negative if you are determined to bar the way to any inquiry into the transformation of the “pre-human” into the “human,” and therefore to accept that we can never understand the constitution of the human, but only observe all the different things beings we know to be human do.