Closed-syllable vowel laxing: A contrast enhancement strategy

Abstract

Closed-syllable vowel laxing describes the cross-linguistic tendency for high and mid vowels to have higher F1 values and more central F2 values in closed than in open syllables. This pattern is often analyzed as resulting from vowel shortening in closed syllables. However, vowel undershoot does not generally result in an increase of F1 for mid vowels. This paper tests an alternative hypothesis according to which laxing is a strategy to enhance coda-consonant place contrasts, with lower and more central vowels providing more informative closure transitions than higher and more peripheral vowels. Two native French speakers were recorded uttering C1VC2 nonce words with C1, C2={p,t,k} and V={i,y,u,e,ø,o,ɛ,œ,ɔ,a}. 85 English and French hearers were presented with the stimuli without word-final bursts and were asked to identify the place of the word-final consonant. In accordance with the enhancement hypothesis, lowering accompanied by centralizing was found to improve [p]-[k] and [t]-[k] contrasts for front unrounded vowels and [p]-[k], [t]-[k], and [p]-[t] contrasts for back vowels. These contrasts were not systematically more distinct after front rounded vowels than after back and unrounded front vowels with similar F1 values (e.g., [y] vs. [i]/[u]), suggesting that centralizing alone is not sufficient to enhance place contrasts.