906 SPOKEN ACADEMIC CORPORA APPLIED TO LANGUAGE TEACHING AND RESEARCH: TOWARDS A MULTIMODAL APPROACH Begoña Bellés-Fortuño Universitat Jaume I 1. From corpus linguistics to applied corpus linguistics? Over the last 25 years there have been developments in corpus linguistics both on the part of European as well as North American researchers. Although not initially with a pedagogical goal in mind but with a research end, most corpus linguistics projects undertaken lately have recognized the necessity of bringing in a pedagogical aim towards the teaching and learning of a language, resulting in what I will call here applied corpus linguistics. European researchers in the study of corpus linguistics have been particularly active specially in the United Kingdom and Scandinavia as the prolific publications and creation of several corpora show. Take for instance the COBUILD project (earlier Bank of English (BoE project) ) which started in the year 1991 and where the collaboration of John Sinclair towards a pedagogical lexical computing orientation was seminal (cf. Sinclair 1987 and 1991), it covers over 450 million words of British English (spoken and written), it is basically a specialized corpus devoted to single genres. Another important European corpus is the BNC (British National Corpus), a 100 million word collection of samples of mainly written but also spoken English language from a wide range of sources which was also started in the year 1991. Also relevant among corpora is the ICE (International Corpus of English) project, which began in 1990 with the main aim of collecting material for comparative studies of English worldwide. The British component of the corpus was published in 2001 with 1 million words. At the moment 13 research teams are participating in this project following Greenbaum’s (1991) initial idea. Also important is the CIC (Cambridge International Corpus); started in 1992, it holds 600 million words but this time including British and American English discourse. Within the ICI we can find corpora such as the British CANCODE (Cambridge and Nottingham Corpus of Oral Discourse in English) or the CAMSNAE (Cambridge Corpus of Spoken North-American English). The ICI tries to help writers in their task of writing books for learners of English and it is only available to authors and writers of books working for Cambridge University Press publishing house. More recently compiled is the BASE corpus project (The British Academic Spoken English Corpus), developed at the University of Warwick under the leadership of Hilary Nesi and Paul Thompson; it consists of 160 lectures and 40 seminars recorded in a variety of university departments, this corpus aims to be a counterpart of the North-American MICASE (Michigan Corpus of Spoken Academic English). As to North-American scholars, there have also been remarkable developments in corpus linguistics, we could name Douglas Biber et al. at Northern Arizona University with the Longman Grammar of Spoken and Written English (1999) (the majority of co-authors being European, though), Michael Barrow at Rice University and other corporist groups at The University of California or the University of Pennsylvania. Standing out we find the ANC (American National Corpus), started in 1990, it includes a massive electronic collection of American English including texts of all genres and transcripts from spoken data that is still in progress. When completed, the ANC will contain a core corpus of at least 100 million words, comparable across genres to the BNC described above. Undoubtedly significant in North-