Oral-laryngeal timing patterns in voiceless stops

Abstract

Voice-onset time (VOT) in voiceless stops is typically said to be shortest for labial stops, intermediate for alveolars, and longest for velars. One proposed explanation states that oral release is actively coordinated relative to the onset of glottal adduction (OGA) to produce place-related VOT differences−specifically, the earlier the release, the longer the VOT. The present study uses transillumination and acoustic data to test this claim directly. The results confirm that oral release tends to be timed earliest for /p/ relative to the OGA, intermediately for /t/, and latest for /k/, thus reflecting the rank order of VOT differences. Interestingly, another pattern of oral-laryngeal timing emerged from the data. Despite place-related differences in the magnitude and duration of the devoicing gesture, the OGA occurred at a fixed interval after oral closure across all stop categories for one speaker and at a fixed interval after closure across /p,t/ for the other. Thus while oral release may be coordinated relative to the OGA, the OGA itself appears be fixed relative to oral closure. [Work supported by NIH Grant DC-00121.]