Research in Corpora in Language Teaching and Learning

A corpus (plural corpora) is simply a large collection of texts, typically nowadays stored on a computer. The idea of understanding individual texts or any part of a text by assembling large quantities of texts that have something in common long predates modern computerized corpora (for a brief history, see McCarthy and O’Keeffe 2010: 3-5). The use of the word corpus to refer to a body of texts (e.g. all the works of one author) goes back to at least the early 18 century, and for many centuries before that, biblical scholars treated the Bible as a corpus, seeking out all instances of particular words and bringing them together in manually-produced concordances. Equally, scholars of literature have long sought to understand an author’s mindset or ‘fingerprint’ by examining their work as a body. And that is what we still do when we use corpora in language teaching: we seek evidence as to how the language is used and how it is learnt by looking at lots of texts assembled together. This way of approaching language is considered by corpus linguists to be a more reliable yardstick than looking at any single text or expression in a text, or else putting aside the textual evidence and looking inwardly, the process of introspection, which some linguists believe to be equally valid.