AbstractThe modern field of corpus linguistics – based around the computer-aided analysis of extremely large databases of text – is largely a phenomenon of the late 1950s onwards. Its early history was marked by opposition from, in particular, Noam Chomsky, who favored a rationalist view over the empiricism associated with corpus-based approaches. However, corpora have been shown to be highly useful in a range of areas of linguistics (but perhaps most notably lexicography and grammatical description). Modern corpus linguistics was formed in the context of work on English, though it is now applied to many different languages; it was in this context that techniques such as corpus annotation, and important concepts such as collocation, emerged. Alongside this history of corpus linguistics considered as a methodology stands the history of an alternative approach, sometimes called neo-Firthian, within which the study of words, phraseology and collocation in corpora are the keystone of linguistic theory.