The British National Corpora

Abstract

The British National Corpora are two large corpora of British English representing the 1990s and 2010s. This chapter evaluates and demonstrates the potential of these corpora for accessing regional linguistic variation. The chapter argues that the spoken components of these corpora, totalling over 20 million tokens of transcribed speech, may be suitable for supra-regional analysis of variation in England through the use of speaker metadata. This is demonstrated by a brief case study which investigates was/were alternation in subject-verb agreement with singular and plural pronominal subjects. The case study shows that the non-standard agreement combinations appear to have declined between the 1990s and 2010s and, to an extent, lost their regional distinctiveness, with the exception of singular + were, which remains distinctively Northern.