Phrasing, prominence, and morphotonology: How utterances are divided into tone groups in Yongning Na

Yongning Na is a Sino-Tibetan language spoken in an area straddling the border between Yunnan and Sichuan. The Yongning Na tone system is based on three levels: L, M, and H. It comprises a host of rules that are specific to certain morphosyntactic contexts. These rules represent the bulk of what language learners must acquire to master the tone system. Different rules apply in the association of a verb with a subject or an object, the association of two nouns into a compound, that of a numeral and classifier, and that of a word and its affixes, for instance. The domain of tonal computation is referred to here as the tone group; tonal processes never apply across tone-group junctures. The present study investigates how utterances are divided into tone groups in Yongning Na, building on examples from narratives and elicited combinations. There is no hard-and-fast correspondence between syntactic structure and tone group divisions: several options are generally open for the division of an utterance into tone groups. The choice among these options depends on considerations of information structure. This study is intended as a stepping-stone towards the long-term goal of modelling the Na tonal system (its morpho-phonology and its phonetics), and placing the findings in a typological perspective.