Towards a useful theory of language

Richard Hudson

last changed 19 March 2009

Bibliographical information

This short paper was written as a pre-view of a plenary talk to be given in the School of Oriental and African Studies, London, on Dec 7th 2007 as part of a celebration at and of the Department of Linguistics. Unfortunately I had to withdraw at the last moment, but in March 2009 it was put on the SOAS website, where it can be downloaded.

Abstract

I survey some tensions between truth and utility in linguistic theory, and reach the following conclusion: In conclusion, then, I have suggested that the truth about language is that it is much more complicated than mainstream linguistic theory suggests; and although good conventional dictionaries and descriptive grammars give some idea of the scale of the problem, they don’t even start to model the mental structures that are actually needed for language. This is a serious problem for anyone who wants to use linguistic theory in describing a language or in solving a practical problem, because it seems to force a choice between a theory which is usable but wrong and one which is right but useless. At present I think this is indeed how things are, but I can imagine a somewhat distant future in which it will be possible to use computers to reconcile truth and utility.