Aside from experiments with Mark Davies' COCA interface and WordSmith Tools, this book was my first exposure to corpus analysis.[As an aside: corpus analysis is a research methodology originating in linguistics. Today the methodology is associated primarily with large databases and specialized software packages that facilitate the examination of large bodies of text in order to uncover language patterns above the level of the sentence. An example of one such corpus is the Corpus of Contemporary American English (COCA), which is continuously updated and at present contains about 425 million words. Corpus analysis uses empirical methods (as opposed to armchair introspection) to develop theories of how natural language is used in specific contexts.] I think this book works quite well for those who have little experience with the field. As the title suggests, it's also geared more towards the researcher who wants to add some quantitative methods to the usual qualitative tools deployed by a discourse analyst (and probably mostly a critical discourse analyst: Fairclough is cited with fair regularity).The text's failings are perhaps only by design, since it is meant to be only an introduction and a how-to guide. However, for my tastes I would have preferred more theoretical framing of these methods (particularly in terms of epistemology, semiotics, or social theory). This would have helped me to wrap my mind around the kind of machine-aided pattern-matching typical of corpus analysis. (In general it seems as if the cultural/philosophical/historical significance of corpus as a method isn't touched in this book). At any rate, a brief survey of the literature on topics such as these would have been useful. ( )