Coarticulation, assimilation and blending in Catalan consonant clusters

Abstract

Electropalatographic data on C-to-C coarticulatory effects were analyzed for consonant clusters composed of an extensive set of Catalan consonants, i.e., dentals (t), alveolars (n, dark l, s, trilledr ), alveolopalatals (∫, ⋌ n), palatals (j) and velars (k). Regarding tongue dorsum coarticulation, results show that consonantal effects in CC clusters are more prominent than vocalic effects in VCV sequences which is attributed to differences in articulatory control between consonants and vowels. Moreover, tongue dorsum lowering for the alveolar fricative and for the alveolar trill appears to be more coarticulation resistant than tongue dorsum raising and fronting for alveolopalatals. Data at the place of articulation show some interesting trends: on the one hand, sequences made of dentals (t), and fronter alveolars and alveolopalatals (i.e., n, l, ⋌, n) yield articulatory blending; on the other hand, any of these consonants may assimilate to those alveolar and alveolopalatal consonants which exhibit a more retracted place of articulation (s, r, ∫), but not vice versa. These findings are in agreement with the “degree of articulatory constraint” (DAC) model which relates coarticulatory and assimilatory effects to the degree of articulatory constraint involved in consonantal production, and predicts that fricatives and trills should be highly constrained both at the tongue front and at the tongue dorsum. Data on the relative strength of the anticipatory and carry-over effects reported in this paper are also to a large extent in agreement with predictions of the DAC model.