Abstract

There is cultural as well as biological evolution; but the "evolution of language" is frequently discussed in terms that suggest that the complex forms of present-day languages are the outcome of a biological evolutionary process. This point of view has been powerfully advocated by Steven Pinker’s popular and very influential book The language instinct, together with several other relatively popular expositions of linguistic nativism during the past decade. These works in turn rely heavily on arguments originally put forward by Noam Chomsky in the 1960s and 1970s in less popular but again very influential writings (though Pinker is more sophisticated than Chomsky about biological evolution, and quotes some additional categories of evidence for the claim that language is an instinct).

My Educating Eve (revised paperback edition, Cassell, 1999) urges that human language is in all important respects a purely cultural rather than biological phenomenon: there is no good reason to believe in a "language instinct" in Pinker’s sense. I look at each strand of argumentation which appears to me to have been influential in winning converts to linguistic nativism since the 1960s, and show that in every case the argument either rests on factually false premises, or is logically fallacious (or, in some cases, both). Thus, the various defences of nativism put forward in Chomsky’s writings can be classified as variants of seven distinct empirical arguments (together with a number of apparently harmless rhetorical moves which on closer examination involve question-begging). These writings inspired independent contemporary research by e.g. Philip Liberman, and B. Berlin & P. Kay, which appeared to reinforce Chomsky’s case. But none of Chomsky’s or his contemporaries’ empirical arguments succeed. Writers of the 1990s have cited further empirical evidence, for instance Pinker discusses the family of alleged "language mutants" discovered by Myrna Gopnik, but these new categories of data also fail to support the consequences drawn from them by nativist writers.

My Paris paper will outline the nature of the case for linguistic nativism, and will illustrate its emptiness by refuting as large a sample of its various strands as can be covered in the time (alluding briefly to the fact that strands which have to be omitted from the oral presentation are also covered in the book). The multi-strandedness of the nativists’ case is one of the features which has helped it acquire more converts than it deserves: someone who looks at any one of the arguments and notices something problematic about it is usually aware that this is only one of many arguments which have been put forward, so that the isolated problem may not seem a fatal flaw. But in reality every one of the arguments is fatally flawed.

 Conference site: http://www.infres.enst.fr/confs/evolang/