Corpus-based methodology and critical discourse studies : context, content, computation

The presentation will discuss practices employed in current and recent relevant research projects in order to address two interrelated criticisms of corpus approaches to critical discourse studies, namely that a) they do not (sufficiently) take account of the relevant context; b) they only/mainly examine lists of words (keywords, collocates), or, at best, concordances of a few words around the node. The examples come from the completed project Discourses of refugees and asylum seekers in the UK press, 1996-2006, and the ongoing project The representation of Islam and Muslims in the UK press, 1998-2008, which used corpora comprising all British national newspapers (both projects were funded by the ESRC; principal investigator: Paul Baker). The central aim is to demonstrate that these criticisms stem from restricted conceptions of corpus research. That is, irrespective of whether they may be justified in individual cases, these criticisms do not concern the potential of corpus-based approaches. More specifically, the following points will be illustrated: • Corpus researchers have no less access to sources of relevant contextual information than (critical) discourse analysts. • A preliminary quantitative analysis of a corpus reveals patterns which not only pinpoint areas that can be usefully examined in detail, but can also uncover helpful contextual elements. • Corpus analysis need not stop at the examination of short concordance lines, but can involve the close analysis of longer stretches of text, or whole texts if necessary – while at the same time retaining the ability to quantify the results.