THE BRITISH NATIONAL CORPUS AS A LANGUAGE LEARNER RESOURCE

Guy Aston
Scuola Superiore di Lingue Moderne, University of Bologna

This is a lightly revised version of a paper presented at the second conference on Teaching and Language Corpora, (held at Lancaster University 9-12 August 1996) in a session jointly organized with Lou Burnard. Lou's paper Introducing SARA provides complementary detailed information about the software described here. A brief report on the whole conference is also available.

Abstract

While large corpora are now extensively used in the production of language learning materials (dictionaries, grammars, concordances, course books), hands-on use by teachers and learners has generally been limited to relatively small collections of text (Johns & King 1991; Murison-Bowie 1993, 1996; Aston 1995; Wichmann et al 1996). Over the last few months, the Bologna University School for Interpreters and Translators in Forli' has had access to the 100-million-word British National Corpus in order to test the specially-designed interrogation software, SARA, with the aim of preparing a handbook for naive users of the BNC (Aston and Burnard, in press). The project has involved experimental use of the BNC as a self- access resource by a small group of advanced learners of English. This experience suggests that access to very large corpora can be highly motivating for such learners, since the quantity and quality of evidence available concerning language use potentially places them in a position to critique the pedagogically-filtered material provided by teachers and textbooks, as well as by corpus-based reference works.