Sentence Prosody in Japanese
I proposed a theory of focus and intonation, according to which the focus structure of a sentence determines the distribution of prosodic prominence. It is suggested by this theory that there are strings which can, or are likely to, become a focus (defined by Rooth (1992)), and those which cannot.Any string can be a focus except when either of the following holds: (i) it consists of a subject NP and a predicate in a sentence which does not express the existence, the appearance, or the arrival of the denotation of the subject (Yatabe(1997)); or (ii) it indudes two dependents. On the other hand, a string is likely to constitute a single focus when it meets at least one of the following conditions: (i) What it expresses it not expected by the speaker; (ii) it is semantically heavy (Ladd(1996)); or (iii) it is an adverb expressing a certain kind of meaning. Then, by the rule proposed, the leftmost word in each of the focused strings becomes prosodically prominent. The theory presented is supported by the F0 contours for sentences studied in a series of experiments conducted for the purpose of writing this paper.
