GwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwD T h e G R E E N Y w o r l d D o m i n a t i o n T a s k F o r c e , I n c o r p o r a t e d Presents: __ __ 999999999 22222222 _____ ____ _| |__| |_ 9999 9999 22 222 // | \ |_ __ _| 999 999 222 || ____ | || | | | | | 9999 9999 222 || || \ / | || | _| |__| |_ 9999999999 222 \\___// \/\/ |____/ |_ __ _| 999 222 |__| |__| 999 2222 999 22222222222 "Language Acquisition: Philosophical Variations on a Theme by Steven Pinker" by Bob the Master of the World ----- GwD: The American Dream with a Twist -- of Lime ***** Issue #92 ----- ----- release date: 01-03-01 ***** ISSN 1523-1585 ----- I. Introduction Language acquisition, though seemingly a purely scientific phenomenon, profoundly affects the way in which philosophers understand what sort of thing language is and through what intellectual paradigm it might best be comprehended. Does language, in both its acquisition and subsequent usage, manifest some basic human instinct, such that human beings obtain it and employ it before or without considering its pragmatic utility in reacting to their environment? Or is language learned in response to some overwhelming communicative need imposed on humanity by the circumstances in which they find themselves, and, while perhaps augmenting the fundamental human constitution, is not an "essential" ability, in the nebulous, ontological sense of the word, and does not arise in the absence of external stimuli? Answers to these questions could shed light on how linguistic rules arise, whether these rules constitute language or should be thought of merely as ad hoc descriptions of an essential unknowable natural process, and whether possessing language is an incidental or essential feature of human beings, to name but a few queries of philosophical import. Steven Pinker, a cognitive scientist at the Massachusetts Institute of Technology, has recently written an article entitled "Language Acquisition," soon to be published as a chapter in An Invitation to Cognitive Science, Second Edition, Volume One (1) and presenting a well-detailed overview of, and commentary on, the latest research that either directly contributes to the issue of language acquisition or addresses concerns posed at the periphery of language acquisition. Pinker attempts two separate tasks in this article: first, he seeks to summarize the available scientific knowledge pertaining to language acquisition; secondly, and perhaps more importantly, he then draws conclusions from the established empirical facts illuminated by the first task. Furthermore, in setting out the empirical data gleaned by cognitive science, he also presents the paradigm through which he interprets the data, which consequently has a bearing on the content and form of his subsequent inferences. I do not seek to challenge the actual data presented by Pinker nor even his commentary on that data; instead, I wish to elucidate the existence of possible alternative interpretive paradigms as well as examine the validity of the implications that Pinker constructs. In effect, as the title suggests, I desire to simply bring out concerns and problems of philosophical interest that lie dormant in Pinker's analysis, veiled under the well-meaning shroud of scientific rhetoric. I will not attempt herein to investigate the entirety of "Language Acquisition," given the antecedently constrained scope of this assignment and the relative girth of that chapter, nor shall I dare to introduce or recommend a rival account of the body of knowledge that constitutes language acquisition, since I find Pinker's work, despite its need for philosophical emendation, rather sound, well-considered, and reasonably well-supported by the empirical observations of cognitive science. Instead, I will examine what I consider to be the sections of greatest philosophical interest and those that incidentally come closest to the paper topic on language acquisition that had been given to the class, namely specific subsections of the sections entitled "The Biology of Language Acquisition" and "Explaining Language Acquisition." II. The Biology of Language Acquisition: Evolution of Language (2) Pinker begins his discussion of language acquisition by examining the adaptation of the human vocal tract to the "demands of speech" as a function of evolutionary progress. Pinker relies on the evolutionary paradigm to initiate his argument for an account of language that underscores its innate position in the constitution of a human being. Yet, evolution can merely demonstrate a correlation between a feature of humanity and its contribution to human survival; it fails, at a purely scientific level, to manifest a causal connection between some human need and the acquisition of a trait to satisfy that need. The aforementioned trait might arise because of some other, less obvious, evolutionary interest, or it might simply have been acquired for the purposes of some future need rather than an existing one. Furthermore, one may note that the structure of the human vocal tract and its consequent ability to produce linguistic utterances may be entirely the product of evolutionary coin- cidence; perhaps the peculiar shape of human vocal tracts has some other natural function, of which language is merely important but subsidiary and unforeseen consequence. Pinker appears to assume that the design of the human vocal tract is optimal for the use of spoken language, and this assumption is critical to his establishment of the evolutionary link between the two. However, an alternative design of the human vocal tract might facilitate speech to an even greater degree; merely because the scientific community has not witnessed a more efficient tract design does not mean that a more efficient one does not exist, one that might even make a human being's relatively advanced tract construction seem primitive and inefficient. Pinker concludes by observing that "the evolutionary selective advantages for language must have been very large to outweigh such a disadvantage [the "sacrifice of efficiency for breathing, swallowing, and chewing"]." (3) In this claim, Pinker assumes that the ability to produce audible sounds necessitated a "sacrifice of efficiency," when in fact the efficient functioning of both abilities might be perfectly compatible; the current human vocal tract design may be deficient for this very reason, insofar as it hampers those other essential human functions. Furthermore, Pinker suggests that language caused what he perceives to be the human vocal tract's manifest inefficiencies. Pinker seems to imply that the need for language altered the vocal tract design through evolution instead of the design itself antecedently allowing the production of aural utterances; in essence, Pinker chooses one possible line of causation, from language to biology, instead of an equally plausible and possible one, from biology to language. Pinker does not provide evidence to support his interpretation over the rival account nor does there appear to be a justifiable way to arbitrate between the conflicting views. Since human beings already possess language and an anatomy that provides for the production of language, how can we come to know which one caused the other, if any causal connection between the two actually exists, an empirical fact that currently lacks conclusive confirmation? Pinker continues by discussing recent efforts to locate language in our closest biological relatives, the primates. He maintains that, in this effort, one must keep in mind that "the scientific question is whether the chimps' abilities are homologous to human language." (4) Why, one might ask, should this question be the relevant one? Pinker denies that an analogous system to our linguistic system, meaning one that possesses a similar function, should be permitted to be subsumed under the category of language. Why should analogy be rejected as a criterion for categorization? Airplanes are radically different from automobiles, in the same sort of way that the wings of bats are different from the "wings" of gliding rodents (Pinker denies that the latter are truly wings), since, in both examples, the entities in question reflect "a different evolutionary history." (5) However, do we not take both airplanes and automobiles to satisfy the basic definition of a vehicle, despite their overt dissimilarities? The definition provided for "vehicle" by The Random House Dictionary of the English Language yields the following: "any means in or by which someone travels or something is carried or conveyed." By this definition, even the legs of animals would count as vehicles, though they evince no apparent homology whatsoever with automobiles. Analogy in function in the case of vehicles seems to provide for equal satisfaction of the definition; why might the case not be the same for language, however different the methods of communication of chimps and human beings might appear to be? Also, it seems relevant to remark that the speech of chimps and human beings appears to differ only as a matter of degree and not of kind, the speech of chimps being noticeably less complex than the speech of human beings but retaining the atomic form of short sounds. Furthermore, Pinker appears to arbitrarily appoint one thing as the paradigmatic satisfier of a definition to the exclusion of another and by which that other should be compared against for satisfaction; he takes the communication of humans to be this ultimate satisfier and consequently rejects that which is not homologous to human communication. Why might not the rudimentary communication of animals be taken as the "default" empirical satisfaction of the term "language"? By what right or through what means does human communication acquire this exalted status? Pinker provides no justification for judging what is or is not a language against the standard of human communication. Thus, not only does he promote homology as the criterion for categorization to the exclusion of the more common-sense criterion of analogy, he further takes it as given that, for anything to count as a language, it must be in agreement, not merely analogously but also and necessarily homologously, with the communication peculiar to human beings. Pinker observes a common scientific belief that human beings developed language out of evolutionary need yet assumes that chimps are incapable of producing a linguistic system as complex as that of human beings. Perhaps chimps and other primates never needed, as a matter of evolution, to develop anything beyond their rudimentary system of communication. The current simplicity of the speech of chimps does not by itself preclude their future ability to advance their system beyond this primitive state. Given an evolutionary impetus, chimps and other primates might develop more sophisticated speech but perhaps not otherwise. This scenario would account for the documented failure of scientists' efforts to recreate human speech in chimps. Furthermore, why should we assume that our language represents the only kind of language? Perhaps chimps cannot learn human language because that language was designed precisely for human beings and not chimps; scientists have attempted to make chimps speak a human language rather than a chimp language and consequently should not be surprised by their failure. This attempt seems analogous in its futility to the project of getting a baby duck to swim by attaching fins to it. Chimps may not be able to learn our language, but such inability does not imply that they might not be able to learn another language of some sort, a language particularly suited to their unique constitution. III. Explaining Language Acquisition: Learnability Theory In this section, Pinker proffers a method for understanding language acquisition through an application of the learnability theory, a branch of theoretical computer science that provides for the circumstances in which and by which learning may be said to have successfully occurred. Pinker realizes that learning a language possesses a difficulty common to all "'induction problems', [in that] there are an infinite number of hypotheses consistent with any finite sample of environmental information." (6) Pinker believes that it is "the role of negative evidence" (7) that reduces the "infinite number" of possible grammatical sentences into a finite set that becomes a communal language. Pinker takes negative evidence to occur when "human children...[are] corrected every time they speak ungrammatically" (8); the problem arises because of the fact that children are not typically corrected on every single occasion of ungrammatical utterance. Pinker asserts, correctly, it seems, that this situation never obtains, and, consequently, some other explanation is necessary to account for the development of the speech of children into the standardized speech of adult members of a linguistic community, one that focuses on some sort of internal grammatical mechanism. Yet, it seems that negative evidence might exist in a different form and consequently obviate the scientific need to look inward for an explanation of proper language development in children. Children learn language not only through direct communication but also, and primarily, through observing the linguistic interactions between adults. Given their possible and even frequent possession of a superset of the language used in their community, children might perceive that the superfluous elements of their language are not used, and hence not endorsed, by the adults whose speech patterns they perceive. In such a way, after repeated exposure to the conversations of adults and the absence of the child's linguistic excesses in those conversations, the child might comprehend that those extra utterances of his contrivance are not accepted in the mainstream language and gradually eliminate them from his speech, wishing to mimic his linguistic exemplars as much as possible in the aim of correct speech. This account, which seems plausible on the face of it, would only require the standard mechanism by which children learn that a dog is not a cat, by repeatedly perceiving that only certain entities are called "dogs" while certain dissimilar entities are called "cats" and consequently learning the correct application of those words. Pinker appears to be too eager to discount the model of language acquisition as simply a subset of general learning in favor of postulating a separate internal mechanism specifically designed for language acquisition. Why does he choose this route, which appears at first glance to be the more difficult path, given the enormous difficulties associated in defending the existence of innate mental constructs? Pinker's acceptance of an internal linguistic mechanism might have the unfortunate consequence of allowing him to appeal to innate structures as first options for solutions to any kind of mental phenomenon, such answers being easier to construct than through the often difficult task of locating the source of whatever phenomenon is in question in the external world; this approach becomes problematic insofar as explanations that center in the working of inner processes seem more difficult to confirm or disprove than empirical explanations and therefore are of lesser scientific value, despite the relative ease with which they might be produced. Such a perspective may yield a certain kind of scientific complacency in which scientists immediately turn to innate mechanisms as soon as any kind of difficulty is encountered in finding an external, empirical solution. I do not wish to assert herein that Pinker is wrong; in fact, he continues throughout the rest of "Language Acquisition" to demonstrate the plausibility and probability of his account. I do wish to contend, however, that this perspective on language acquisition may be a slippery slope in which scientific investigation yields to scientific speculation and consequently loses a great deal of its potential for and utility in the advancement of knowledge. IV. Conclusion I may be criticized for not covering, in the commentary above, the breadth of Pinker's article, nor even its essential contentions. However, I do not believe such criticism is just, for I consider myself justified in examining those issues that I feel require philosophical elaboration and on which I had something meaningful to say. A large portion of "Language Acquisition" is concerned merely with presenting the accumulated empirical data for a given hypothesis, and with such I could do little in the way of philosophical commentary. I have sought herein only the modest goal of illuminating some of Pinker's assumptions as well as challenging some of these assumptions for their lack of adequate justification in the hopes of encouraging further rumination on his work. ----- -=[Footnotes]=- 1. Edited by L. R. Gleitman, M. Liberman, and D. N. Osherson and soon to be published by the MIT Press in Cambridge, Massachusetts. I discovered Pinker's article at the following URL: http://www.cogsci.soton.ac.uk/~harnad/Papers/Py104/pinker.langacq.html 2. For convenience, the section headings of this essay will correspond to the sections of "Language Acquisition" on which they comment. 3. Pinker, 3. 4. Ibid. 5. Ibid. 6. Ibid, 9. 7. Ibid. 8. Ibid. ----------------------------------------------------------- GwDweb: http://www.GREENY.org/ GwD Publications: http://gwd.mit.edu/ ftp://ftp.GREENY.org/gwd/ GwD BBSes: C.H.A.O.S. - http://chaos.GREENY.org/ Snake's Den - http://www.snakeden.org/ E-Mail: gwd@GREENY.org * GwD, Inc. - P.O. Box 16038 - Lubbock, Texas 79490 * -------------------------------------------------------------------------------- "Semantics." -------------------------------------------------------------------------------- -+- F Y M -+- GR33NY LIK3S mash3d p0tat03s MORE THAN FIVE YEARS of ABSOLUTE CRAP! /---------------\ copyright (c) MM Bob the Master of the World/GwD Publications :SHIT YOUR PANTS: copyright (c) MM GwD, Inc. : GwD : All rights reserved \---------------/ GwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwDGwD92