What AVT can make of corpora: some findings from the Pavia Corpus of Film Dialogue

This contribution reports on a research project aimed at identifying frequent phraseology in both original and translated filmic speech. Findings based on the Pavia Corpus of Film Dialogue, a parallel corpus of original British and American film dialogues and their dubbed Italian versions, are presented and discussed with a view to showing how corpus methods can be of use to audiovisual translation (AVT). It is argued that, on the one hand, corpus investigations provide a deeper understanding of film dialogue and the ways in which such dialogue functions. On the other, quantitative information gleaned from observation of translators' recurrent behaviour can help pinpoint linguistic and translational areas that are specific to the genre and, consequently, develop translators' awareness, improve translation quality and thus foster alignment with acceptability standards.