Theoretical linguists tend to start from the assumption that the native speaker’s competence can be seen as a homogenous entity with a well-defined grammar as its underlying basis. Speaker variation is variation between speakers, i.e. variation between the individual grammars of the speakers.

However, it has long been observed in bilingual acquisition studies, contact-language studies and sociolinguistic studies that the homogenous view of the linguistic competence is problematic because individual speakers often adjust their particular language use to a number of discourse variables, leading to style shifting or code-mixing sometimes leading to the emergence of a new language. This conference brings together experts from different domains: formal linguistics, bilingual acquisition, psycholinguistics, sociolinguistics and cultural studies to reflect on how this style shifting or code-mixing should be captured in terms of one speaker’s linguistic knowledge and what it says about the concept of ‘grammar’ as the underlying system for the performance.