The most widely held view on the syntax-phonology interface is built on the assumption that phonological processes do not refer to syntactic constituency directly, but rather to the derived prosodic structure. Revising this standard prosodic hierarchy theory, Selkirk (2009) puts forward a new universal Match theory, focusing on Japanese wh-questions. In particular, Selkirk maintains that the syntax-phonology interface universally calls for a clause to match up with the intonational phrase. Jung (2010, 2012, 2013), on the other hand, advocates a direct feature reference analysis, based on Busan Korean wh-questions, according to which phonology interprets the syntactically-licensed wh-agreement features directly with its own constraints on the formation of wh-prosody. The present paper tries to show that the prosodic domains of wh-questions cannot be predicted by Selkirk's Match theory either in Japanese or in Korean due to the complexity and variability of wh-prosodic patterns, and that the precise nature of wh-prosody can be best explained by an interplay between syntax and phonology under the direct feature reference analysis.
Abstract
1. Introduction
2. Prosodic Patterns of Wh-Questions in Busan Korean and Fukuoka Japanese
3. Theories on the Prosodic Boundaries of Wh-Questions
4. Can the Match Clause Constraint Be a Universal Prosodic Condition on Wh-Questions?
5. Revisiting the Issue Under the Direct Feature Reference Analysis
6. Conclusion
References
(0)
(0)