Bilingual language behaviour is driven by numerous factors that are usually studied in isolation, even though individual factors never operate alone. Bringing together key insights from psycholinguistics and sociolinguistics, Luna Filipovic presents a new model of bilingual language processing that captures bilingualism within and across minds. The model enables readers to explain traditional puzzles in the field, and accounts for some apparently contradictory reports in different studies. It shows how theory can be applied in practice and how practice feeds back into theory, with mutual benefits. Bilinguals are studied in action, when they interact with other bilinguals or monolinguals, when they recall witnessed events in real life and in the lab and when they translate and interpret for the benefit of monolinguals. This interdisciplinary take on bilingualism in action will lead to new research on bilingualism itself, and to applications in forensic linguistics and translation studies.

This volume introduces a new concept, 'criterial features', for the learning, teaching and testing of English as a second language. The work is based on research conducted within the English Profile Programme at Cambridge University, using the Cambridge Learner Corpus. The authors address the extent to which learners know the grammar, lexicon and usage conventions of English at each level of the Common European Framework of Reference (CEFR). These levels are currently illustrated in functional terms with 'Can Do' statements. Greater specificity and precision can be achieved by using the tagged and parsed corpus, which enables researchers to identify criterial features of the CEFR levels, i.e. properties that are characteristic and indicative of L2 proficiency at each level. In practical terms, once criterial features have been identified, the grammatical and lexical properties of English can be presented to learners more efficiently and in ways that are appropriate to their levels.