抄録
The purpose of this study is to elucidate the nature of disfluencies produced by stutterers and normal speakers of Japanese. The first part of the paper reports the result of a detailed analysis of over 2000 stuttering samples (repetition) produced by 36 stutterers and 286 normal speakers. The analysis has revealed that most of the segmentation patterns in the samples are mora-based segmentations. The second part is to seek the difference between stutterers' and normal speakers' repetitions. The statistic analysis from the viewpoint of a phonetic transition defect has revealed that the trigger of nonstutterers' disfluency is not so relevant to the phonetic transition as that of stutterers'. This implies that the two kinds of disfluencies have different backgrounds. On the other hand, the triggers of stuttering, which are stops, can be interpreted as stutterers' flaws in their phonetic plans. This finding makes up for a weak point of The Covert Repair Hypothesis.
