Language variation and corpus linguistics

Abstract

ABSTRACT: Corpus linguistics deserves serious attention from linguists and applied linguists, since it is of direct relevance to linguistic description, language variation, lexicography, and language education. Linguists tend to be indifferent to corpora, however, as the predominant paradigm in linguistics is based on introspective data, i.e. native speaker intuition. Research has shown that intuitions are not 100 per cent reliable; the notion of ‘core’ grammar needs to be modified to accommodate the systematic differences across registers at all linguistic levels. Moreover, what linguists perceive as significant principles of linguistic organization may not coincide with their distribution in patterns of use. One goal of applied linguistics is to see what correspondences can be established between the two sets, i.e. the set of underlying principles of linguistic organization and the patterns of use of these principles revealed by analyses of corpora. Regrettably, applied linguists have not embraced corpus linguistics any more enthusiastically than formal linguists. Corpus linguistic analyses have their problems, too. This paper examines a set of data from the lexicon and grammar of world Englishes to suggest that a complete reliance on patterns of use may not solve all the problems of language description, study of variation, language instruction, translation, and lexicography. Furthermore, whereas analyses of corpora are effective in revealing dialect variation, they are not of great use in accounting for diatypic variation, i.e. the permanent characteristics of users of language and recurrent features of their language use, which are crucial for understanding human linguistic behavior.