A Corpus-Based Study of Compounding in English

It is long established that corpus-based studies force the linguist-analyst to come faceto-face with a number of phenomena that might easily be overlooked in an armchair-type study. In this article, we demonstrate the validity of this truism once again in a study of English compounding patterns. We report here on a study of word-formation patterns in words from a large corpus of British newspaper English. In this article, we consider only new compound formations, and we show that considering real data can cause problems for the theoretician of word-formation and for the descriptive grammarian alike. Not only do we report on patterns that are not described in the major handbooks (Adams 1973; Bauer 1983; Marchand 1969), but we show that some of the patterns that are being used productively in the English of the early 1990s break principles that are laid down as absolute in some of the theoretical works.