The way sentences are pronounced signals listeners how they should relate information in the sentence to the preceding discourse, a phenomenon known as ‘focus’ or ‘information structure’. As in many languages, expression of focus in English is determined by how constituents are grouped into domains and how pitch accents are realized within domains. For instance, if She teaches in China is an answer to the question ‘What does she do?’, pitch accents give prominence to both teaches and China, but if the same question is answered She teaches linguistics, only linguistics is accented, though teaches is also focused as part of the answer. The difference here is due to verbs and objects forming different phrasal constructions from verbs and prepositions.
Languages differ in the way phrasing and prominence interact to signal focus. Chinese dialects present a challenging and revealing research area for this interaction. These dialects use pitch changes (tone) to indicate word meanings in isolation, but tonal realization of words in phrases is determined by domain formation (i.e. phrasing effect) and pitch changes (i.e. prominence effect) within the domain, together known as tone sandhi. Crucially, focus affects tone sandhi. Thus far, studies on tone sandhi, however, rarely consider the effect of different focus conditions. Work on focus realization has also been limited to simple morpho-syntactic constituents in few dialects. I propose a systematic experimental approach to examining the interaction of these two hitherto independent lines of research (tone sandhi and focus realization) in six dialects of Chinese, which lie on a continuum between dialects with dense tonal distributions and sparser distributions. In so doing, I hope to contribute to the understanding of the phonetics and phonology of tone sandhi, and to a cross-linguistic theory of focus realization, which has largely been based on European and African languages.