Temporal (in)stability in English monosyllabic and disyllabic words: Insights on the effect of voicing on vowel duration
English is one in the wide range of languages in which the duration of vowels is modulated by the voicing of the following consonant: Vowels are shorter when followed by voiceless stops, and longer when followed by voiced stops. The so-called voicing effect has been attributed to a variety of mechanisms. Temporal compensation between the duration of the vowel and the following stop closure is one of these mechanisms. Based on acoustic data from Italian and Polish disyllabic words, the compensatory mechanism has been proposed to be a consequence of the temporal stability of the interval between the consonant releases flanking the vowel. The timing of the VC boundary within this interval determines the respective durations of the vowel and the stop closure. In this paper, it is shown that the duration of the release-to-release interval is not affected by the voicing of the second consonant in English disyllabic words, but that it is in monosyllabic words. It is argued that the stability of the interval can be derived from the isochronous phasing of the vocalic gestures in the VCV sequence of disyllabic words. The absence of the temporal anchor of a second vowel in monosyllabic words, on the other hand, allows the vocalic and the consonant gesture durations to be modified independently. Other aspects of production and perception behind the voicing effect can coexist with a temporal compensation mechanism and cannot be excluded.