Multilingualism acts as a motor of linguistic developments, and accordingly, multilingual communities can afford us a privileged view onto ongoing tendencies of language variation and change. However, in order to make full use of this opportunity, it is not enough to measure multilingual language use against the yardstick of a putative monolingual standard language. Rather, we need to look at the gamut of speakers’ repertoires, which means taking into account both formal and informal registers and, crucially, doing so for multilinguals and monolinguals alike.

In my talk, I address the methodological challenges for such an enterprise, and in particular the importance to elicit data that is representative of speakers’ natural behaviour in different communicative situations, and discuss set-ups to overcome this challenge. I show that looking at matching repertoire data from multilinguals and monolinguals can reveal interesting similarities of noncanonical patterns, often with quantitative advantages for multilingual communities that make them particularly valuable as a domain of study.

By way of example, I discuss evidence from the domain of noun phrases, drawing on findings from current projects on heritage speakers’ linguistic repertoires, urban contact dialects (especially Kiezdeutsch), Namibian German, and language use at a multilingual Berlin street market.