- 著者
-
河野 守夫
- 出版者
- 日本音声学会
- 雑誌
- 音声研究 (ISSN:13428675)
- 巻号頁・発行日
- vol.2, no.1, pp.16-24, 1998-04-30
Human beings innately have two neuropsychologically different rhythm processing systems: holistic and analytic. The former copes with rapid tempos of rhythm with less than 330ms inter-beat intervals (IBIs), and the latter with slow ones with more than 420ms IBIs. These two systems constitute a hierarchical structure: the holistic system comprises the base, and the analytic system, the superstructure component. These systems and their relationship were detected in Kohno's accumulated experiments using a split-brain patient, a patient with pure anarthria, and normal persons as subjects (Kohno, 1992, 1993, etc.). On the basis of these experiments, the underlying reasons for correction of truisms long-held by phoneticians will be discussed and the following claims will be made: 1) The traditional dichotomy of speech timing such as 'stress-timed' and 'syllable-timed' should be reclassified into 'stress-timed' and 'mora-timed'. 2) 'Mora-timed' is not a subcategory of 'syllable-timed'. 3) Equal interval accent is not a language universal, but a language specific. The timing regulation mechanism in a mora-timed language and the possibility of foot structure in Japanese (cf. Poser, 1990) will also be discussed.