Chie Sakuta, University of Tokyo

Intonation units in Japanese and English discourse

Previous studies have argued that Intonation Units (IUs) tend to be more fragmented (i.e. clauses are more frequently broken down into more than one IU) in Japanese than in English. This trend has often been related to the interactional characteristics of Japanese spoken discourse, such as the presence of interactional final particles. In this
presentation, however, I will show that there is a large difference between discourse genres in the structure of IUs, and in the cross-linguistic difference in the tendency for IU fragmentation. Furthermore, the genre variation suggests that the interactional motivation for this difference, if any, is not as significant as other motivations, such as the way that information structures are marked in Japanese and English discourse.

Huei-ju Huang, National Taiwan University

Clause structure in Tsou-discourse-based observations

Tsou is a mildly endangered Austronesian language spoken in the southwest mountain area in Taiwan. As in many other western Austronesian languages, one of the typologically unusual linguistic features of Tsou is its voice
construction. In this talk, I will introduce the four voice constructions in Tsou, an actor voice and three undergoer voices. The actor voice is intransitive and undergoer voices are transitive. Tsou makes no syntactic distinction between argument and adjunct-what is grammatically an adjunct in English tends to be coded as an argument of the predicate in a voice construction. Voice selection is determined both by the semantic roles of the nominative NPs and by the transitivity of the predicates. These four voice constructions constitute the basic clause structure in Tsou.