In this final chapter I shall try to answer briefly the question: What difference have corpora made to the working lives of applied linguists?

Corpora have made language investigations possible that were not possible before

Examples of new possibilities include all kinds of quantitative linguistic work, such as obtaining accurate frequency counts for words, lemmas and grammatical features, or comparing one speaker/ writer's usage against a norm. This can be used to identify register differences, or to characterise the style of an individual writer, with applications in literary stylistics, forensic linguistics and literary translation.

Corpora have changed the way we look at language and, for teachers at least, the way we see our own roles

In terms of language, new concepts such as the ‘unit of meaning’ or ‘semantic prosody’ are dependent on the availability of large quantities of language which can be manipulated electronically. These concepts in turn bring into question traditional language units such as the phrase and the clause.

For the language teacher, data-driven learning and reciprocal learning necessitate a re-evaluation of the teacher's role. More generally, whatever role a corpus plays in the learning process, the very activity of designing and compiling corpora forces attention upon the assumptions in place. Without an explicit corpus, for example, the question of what varieties of English a learner should have as a model may remain implicit, or may be different in theory and in practice.