ABSTRACT
This chapter explores the ways in which corpus linguistics can contribute to the study of discourse. Combining the two approaches makes it possible to investigate language as a social phenomenon through representative corpora that would be too large for purely manual analysis. The chapter explains and illustrates how standard corpus linguistic tools (frequency analysis, keywords, collocations and concordances) can reveal the genre-specific representation of themes, social actors and perspectives. Although researcher bias cannot be completely eliminated, it can be reduced if qualitative discourse analysis is enriched with quantitative, corpus-assisted methods. Concluding with epistemological reflections, the chapter also discusses what corpora cannot tell us about discourse.
