In American English, phonologically voiced stops are often phonetically voiceless in utterance-initial position. Other than Westbury (1983), there is little articulatory evidence regarding utterance-initial voicing in American English. The current study uses ultrasound imaging and acoustic measures to examine how tongue position correlates with phonation in American English, comparing phonated voiced stops, unphonated voiced stops, and voiceless stops in utterance-initial position. Eight speakers of American English recorded voiced/voiceless stops at three places of articulation (labial, alveolar, and velar). One adjustment for initiating or maintaining phonation during the closure is enlarging the supraglottal cavity volume primarily via tongue root advancement. In utterance-initial position, there was a clear distinction between voiced stops and voiceless stops in the tongue root for the alveolar and velar places of articulation. Even without acoustic phonation during closure, the tongue root is advanced for voiced stops in comparison to voiceless stops. These results suggest that speakers have the same target for both phonated and unphonated stops in utterance-initial position, but other articulatory adjustments are responsible for the presence or absence of phonation. The difference in tongue root position reflects the phonological laryngeal contrasts in American English.

This content is only available via PDF.