ELISA – a pedagogically enriched corpus for language learning purposes

The aim of this paper is to introduce a methodological solution for the design and exploitation of a corpus which is dedicated to pedagogical goals. In particular, I will argue for a pedagogically appropriate corpus annotation and query, and for the enrichment of such a corpus with additional materials (including corpusbased tasks and exercises). The solution will be illustrated with the help of ELISA, a small spoken corpus of English containing video interviews with native speakers. However, the methodology is transferable to the creation of pedagogically relevant corpora with other contents and for other languages. 1 Corpora in language learning: where, what corpora and how While corpora have made their way into many areas of language pedagogy, for instance to inform teaching materials or learners' dictionaries, their direct exploitation by learners and teachers has to date largely been confined to academic settings. In other learning environments, especially in the school context, the application of corpora has so far remained an exception (cf. Davies & Russell-Pinson 2004, Kaltenböck & MehlmauerLarcher 2005) and even the mere awareness of corpora among teachers is low (cf. Mukherjee 2004). One could argue that this is not too surprising because the process of 'percolation' from corpus-based research to corpus use in teacher training and eventually in teaching practice simply takes some time (cf. Chambers 2005, McEnery & Wilson 1997). However, according to Granger (2004) and Tribble (2000) the uptake of corpora in language pedagogy remains generally lower than might be expected in view of the huge amount of corpus-based research carried out over the past decades (see also Mukherjee, this volume), and even in universities corpus-based language courses are mainly run "by researchers and enthusiasts" (Chambers 2005: 1). This suggests that there are more fundamental reasons which prevent a broader uptake of corpus consultation 'in the classroom'. In search of possible explanations it will be helpful to take a look at current practice, especially at the types of corpora used and at the methods of their exploitation. The corpora used 'in the classroom' range from existing reference corpora (e.g. the British National Corpus, Bernardini 2000) and subsections thereof (e.g. BNC sampler, Gavioli & Aston 2001) to a number of smaller, 'self-made' corpora, especially newspaper