A corpus is a collection of specimens of a language as used in real life, in writing and/or speech. Corpus linguistics is research, carried out in university linguistics departments and computing departments (and nowadays in industrial research labs too), which uses corpora as crucial sources of evidence on the structure and properties of languages. Because corpus linguistics has grown fast from small beginnings, newcomers to the field often find it hard to get their bearings. Important papers can be difficult to track down. This volume reprints 42 corpus linguistics articles which first appeared at dates ranging from 1952 to 2002, and which between them illustrate all the main directions in which the subject is developing. It includes articles that are already recognized as classics, and others which deserve to become so, supplemented with editorial introductions relating the individual contributions to the field as a whole.