Roehampton. university, United Kingdom
A syntactic variationist study of a German/English mixed code: the blueprint is not a Red Herring
A reply to Garnder-Chloros & Edwards (2004)
This paper presents the grammar of a German/English mixed code. It demonstrates that this code can be adequately and appropriately described with a text-friendly grammatical model that allows a completely surface analysis, i.e. Word Grammar. It shows that the notion of a matrix language as suggested by some researchers (Joshi 1985 & Myers-Scotton 1993) is not tenable in the light of the data this paper is based on. Based on the grammatical analysis it demonstrates that the principles guiding code-switching are probabilistic (rather than universal) but nonetheless interesting and valuable for structural linguistics for several reasons. First, consistent properties of mixed corpora provide us with a rich source of information for identifying grammatical regularities across languages. Second, they give us insights into bilingual processing and production, and third, they throw light on interfaces, how various linguistic (and non-linguistic) factors interact. The paper concludes that the blueprint is not a red herring, i.e. a detailed grammatical analysis of a mixed code does not divert attention form the real question. Mixed corpora, like all corpora, may initially worsen the theoretical problem of limiting hypotheses (because they provide large infinites of things to calculate), but they do contain the information for identifying grammatical regularities.
Session: Paper session
Contact / Code-Switching
Friday, April 4, 2008, 10:30-12:00