2 - Learnability and the acquisition of syntax
Published online by Cambridge University Press: 15 October 2009
Summary
Introduction
… there is no a priori reason to expect that the languages permitted by UG be learnable – that is, attainable under normal circumstances. All that we can expect is that some of them may be; the others will not be found in human societies. If proposals within the P[rinciples] & P[arameters] approach are close to the mark, then it will follow that languages are in fact learnable, but that is an empirical discovery, and a rather surprising one.
(Chomsky and Lasnik, 1995: 18)The above passage appears in a discussion in which Chomsky and Lasnik argue against the proposition that an adequate grammar of a language must provide an efficient basis for parsing the language's sentences. Well-known examples of well-formed sentences which are not efficiently parsable provide the crucial cases for the argument, which, of course, is ultimately based on the importance of the competence–performance distinction. Consistency then requires that skepticism be extended to another aspect of linguistic performance, that of acquiring a natural language. Just as there is no reason to believe that a grammar is well-designed to parse over an infinite range, so it is not necessary for universal grammar (UG) to be designed so as to make available only learnable languages. The optimism which appears in Chomsky and Lasnik's final sentence is closely linked to the theme of this book, since it is based on an accumulation of evidence indicating that linguistic variation, at least that obtaining within core grammar, can be accommodated by a finite number of finitely valued parameters.
- Type
- Chapter
- Information
- Language Acquisition and Learnability , pp. 15 - 80Publisher: Cambridge University PressPrint publication year: 2001
- 2
- Cited by