A reexamination of the compensation effect in the mora hypothesis

Abstract

The objective of this paper is to reexamine the compensation effect of the mora hypothesis in mora timing. An early investigation of the mora hypothesis presented the compensation effect in Japanese as evidence in support of the mora hypothesis [Port et al., Phonetica 37, 235–252 (1980)] and another presented contrary evidence against it [M. Beckman, Phonetica 39, 113–135 (1982)]. However, since Port et al.'s evidence was obtained in comparison with different test words produced by Japanese and Arabic speakers, their results need to be reexamined. In this paper, an experiment was conducted as to whether the compensation effect observed in Port et al.apos;s experiment will be relevant only to a mora-timed language by using the identical test words produced by Japanese and Arabic speakers. Contrary to Port et al.'s results, the results of this experiment indicated that the compensation effect was observed not only in Japanese but also in Arabic. These data indicate that the compensation effect may not be evidence for the mora hypothesis.