What accounts for boosts in downstep? – Syntax-prosody mapping revisited Midori Hayashi University of Toronto
Japanese downstep is a phonologically conditioned phenomenon in which a lexical
accent triggers the lowering of a subsequent H tone. It is generally assumed that the domain of downstep is the Major Phrase (MaP). Within this domain, however, tone is not lowered consistently, but is ‘boosted’ at certain points. The question of where these boosts occur remains open.
It has been assumed that there is a strict mapping between syntax and prosody (Poser,
1984; Pierrehumbert and Beckman, 1988; Kubozono, 1989; Selkirk and Tateishi, 1991). Kubozono (1989) proposes that syntactic binary branching structure maps into the prosodic structure; therefore, he assumes the prosodic structure (1b) for the sentence (1a) (*indicates an accented mora). The boost occurs at the left edge of branching constituent, as indicated by an arrow. Selkirk and Tateishi (1992) claim that boost occurs at the left edge of XP, which corresponds to the left boundary of MaP; hence (1c) is the prosodic structure that they would predict for (1a). Although these theories differ in certain respects, they agree that the prosodic structure is fully determined by syntax. To account for those boosts which do not conform to syntax, a restructuring rule is postulated, such as Kubozono’s Rhythmic Principle (2), which states that an utterance in a uniformly left branching structure is restructured into an equally-branched structure. I will call theories which make the same assumption “strong syntax-prosody” mapping theories.
I will argue against strong syntax-prosody mapping theories (see Ghini (1993) for
similar arguments). I propose that downstep domains are organized by means of a combination of syntactic constituency and a weight-balancing strategy, which groups Minor Phrases (MiP) into equally-weighted units. This is different from Kubozono’s (1989) Rhythmic Principle because the weight-balancing strategy applies regardless of the syntactic branching direction. An experiment was conducted using five sentences, each of which has four MiPs and a major syntactic boundary between the first and second MiPs (MiP1 and MiP2) as in (3) (the boundaries tested are a topic boundary, a relative clause boundary, a reported speech clause boundary, a conjoined clause boundary and a sentence boundary). Strong syntax-prosody mapping theories would assume MiP2 will be boosted.
The major findings follow: The result for the sentence with a topic boundary (3) is
summarized in Figure1 as an example. First, one out of three speakers boosted MiP2 for all the sentences except the sentence with a conjoined clause boundary, where no boost was observed. Second, the other two speakers showed a strong tendency to boost MiP3; they resorted to the weight-balancing strategy. Third, MiP2 was boosted for all the speakers only at a sentence boundary. Based on this, I will propose two derivational rules and their constraints; First, speakers may choose to wrap MiPs in a certain syntactic relationship (e.g., a modifier-modifiee relationship). Second, the weight-balancing strategy applies, but it cannot interrupt the wrapped MiPs. My presentation will conclude with further constraints and how the observed boosting patterns are derived from these rules.
(3) [hana*ya-wa [huro*ya-ni [kone*ko-o florist-TOP
‘The florist left the kitten to the bath-house owner.’
Figure 1: Means of peaks for (2) for each speaker.
References Ghini, M. (1993). Phonological Phrase Formation in Ligurian Italian. ms. University of
Kubozono, H. (1989). Syntactic and rhythmic effects on downstep in Japanese. Phonology 6.
Pierrehumbert J. and M. Beckman (1988). Japanese Tone Structure. Cambridge, Mass.:MIT
Poser, W. J. (1984). The Phonetics and Phonology of Tone and Intonation in Japanese. Ph.D.
Selkirk, E. and K. Tateishi. (1991). Syntax and Downstep in Japanese. In Georgopoulos, C.
and R. Ishihara. (Eds.). Interdisciplinary Approaches to Language: Essays in Honor of S. –Y. Kuroda. Dordrecht, Netherlands: Kluwer Academic Publishers.
Malaysia’s pioneer in alkaline anti-oxidant water and distributor for world’s leading brand of water ionizer – Toyo Mavello Alkalife Sdn Bhd (603) 78774846; [email protected] ; www.alkalifewater.com ___________________________________________________________________________ Why Don’t Doctors Operate on Patients on Blood Thinners? The human blood has many God-given life-s
Daystar Church Fast Overview Fasting is a spiritual discipline designed to better connect us with God. At the beginning of 2010, we'll take 21 days, to deepen our relationship with God and center ourselves on what God is going to do in our church and in our lives this next year. We are praying that hundreds of people will come to faith in Christ this next year and that each family in