A SUBVERSIVE ACCOUNT OF LINGUISTIC META-THEORY

This work criticizes the articles of faith of Chomskyean linguistic meta-theory and in so doing seeks to vindicate Chomsky's teacher, Zellig Harris. I argue that Chomsky's revolution in linguistics, launched by the publication of Syntactic Structures, was misguided from its inception. Chomsky misconstrued the old paradigm in linguistics by unjustly burdening Harris with a meta-theory that Harris did not hold. Chomsky charged Harris with seeking procedures for the discovery of grammars; I argue that what Chomsky called "discovery procedures" was a straw man, revealing from the outset Chomsky's mentalistic outlook.;Recalling an earlier argument of Charles Hockett, and agreeing with Harris, I argue that not only is language not well-defined, contra Chomsky, but that it is essential that language be ill-defined. I tie Chomsky's view to his version of innateness, and use a feature of Harris' grammatical machinery to point to Chomsky's error.;I summarize the history of Chomsky's nativism and attack what I take to be Chomsky's primary argument in support of his views. I argue that learned syntactic correlates of semantic properties of lexical items explain how a child may come to know what Chomsky claims we must suppose to be innate.;Finally, I discuss the recent grammatical theory of Zellig Harris, which explains linguistic structure as the resultant of external demands placed upon speakers of a language, and hence upon language. If such an account is successful, it obviates the need for an explanation of linguistic structure in terms of a correlative mental structure.;I argue that Chomsky's idealization to a homogeneous speech-community of ideal speaker-listeners conceals strong presuppositions which are not warranted. By invoking this idealization, Chomsky was able to gain a foothold for mentalism, from which he would soon leap to innateness.