Morphological vs. phonological contrastive topic marking

In recent studies, several authors have claimed that a contrastive topic (CT; also known as topic-focus, contrastive focus, etc.) is morphologically marked in languages like Japanese (Hara in press) and Korean (Lee 1999a, b), while it is phonologically (tonally) marked in other languages. In English, for example, it is said that a contrastive topic is marked with the so-called B-accent (fall-rise tone; (L+)H^*L-H%), while a focus is marked with the A-accent (fall tone; H^*L-L%) (Jackendoff 1972; Büring 2003; Kadmon 2001). In this paper, I argue that information structurebased analyses of CT-contours (e.g. English B-accent) along the lines of Büring (2003) and Roberts (1996) cannot be applied to CT-morphemes, and propose an alternative semantic analysis of CT-morphemes. I briefly review the analysis developed by Roberts (1996) and Büring (2003) and discuss that Roberts-Büring's analysis cannot be applied to CT-morphemes, demonstrating that certain crucial assumptions that it makes about CT-contours do not hold for CT-morphemes. I propose an alternative analysis of CT-morphemes; I argue that the semantic contribution of a CT-morpheme is antonymous to that of the additive particle ‘also’; and I examine whether the proposed analysis can be extended to CT-contours, so that CT-morphemes and CT-contours can be given a uniform analysis.