The role of contrast in limiting vowel-to-vowel coarticulation in different languages

Languages differ in their inventories of distinctive sounds and in their systems of contrast. Here, it is proposed that this observation may have predictive value with respect to how extensively various phones are coarticulated in particular languages. This hypothesis is based on three assumptions: (1) There are "output constraints" on just how a given phone can be articulated; (2) output constraints are, at least in part, affected by language-particular systems of phonetic contrast; and (3) coarticulation is limited in a way that respects those output constraints. Together, these assumptions lead to the expectation that, in general, languages will tend to tolerate less coarticulation just where extensive coarticulation would lead to confusion of contrastive phones. This prediction was tested by comparing acoustic measures of anticipatory vowel-to-vowel coarticulation in languages that differ in how they divide up the vowel space into contrastive units. The acoustic measures were the first and second formant frequencies, measured in the middle and at the end of the target vowels /a/ and /e/, followed by /pV/, where /V/ was /i,e,a,o,u/. Two languages (Ndebele and Shona) with the phonemic vowels /i,e,a,o,u/ were found to have greater anticipatory coarticulation for the target vowel /a/ than does a language (Sotho) that has a more crowded mid- and low-vowel space, with the phonemic vowels /i,e,e,a,c,o,u/. The data were based on recordings from three speakers of each of the languages.