Innateness, internalism and input: Chomskyan rationalism and its problems

This paper is about the way linguists do linguistics and the constraints on linguistic theories. It looks at the problem of handling data, what counts as data for linguistics, and discusses two macro paradigms in linguistic metatheory: the inductivist and the deductivist approaches exemplified here by Bloomfieldian and Chomskyan linguistics. It concludes that neither approach can justifiably exclude the other.