Language universals and linguistic argumentation: a reply to Coopmans

Abstract

In the scientific community, reviews have a sacrosanct status, and it is therefore with some trepidation that I requested permission to reply to Coopmans' review article of Comrie (I98I; hereafter LULT). I have tried to ensure that in my reply I do not simply defend myself against the general and specific criticisms of LULTmade by Coopmans, but rather that I address the general issues raised by Coopmans in the spirit of an ongoing debate about the relation between data from individual languages, analyses of individual languages, formal theories of language, and language universals. Just as Coopmans (I 983) is a review article rather than just a review, this is a ' reply article' rather than just a reply. In the space at my disposal, I am not able to reply to all the criticisms made by Coopmans, and I have therefore selected, first, those general points that seem most indicative of the differences between my approach and Coopmans' (as a representative of current generative grammar) and, secondly, those analyses discussed by Coopmans that seem particularly illustrative of differences in methodology. The fact that I do not reply to a specific criticism in Coopmans (I983) should not be taken as an indication that I accept the validity of that criticism. The manuscript of LULT was delivered to the publisher at the beginning of I98 I, and therefore did not take into account work in generative grammar published after that date, in particular Chomsky (I98I), a major work which clearly establishes a new period in the development of generative grammar. In LULT I claimed that one of the major differences between my approach to language universals and that of generative syntax is that I believe it necessary to consider data from a wide range of languages (5-12), whereas generative grammar typically considers only data from English or English plus a narrow range of other languages. Chomsky (i98 i) makes clear that generative grammarians have come to realize that an adequate study of syntax within universal grammar requires the study of languages of different types (in the current terminology: with different parameter values); I am delighted that this difference between the generative approach and my own approach