An acoustic study of palatalization of fricatives in American English

The alveolar fricative consonants (/s,z/) are sometimes palatalized when they appear adjacent to a palatal consonant (/š,ž,y/) as, for instance, in “this ship” or “us yet.” Two questions about this process were studied for six speakers recorded during both spontaneous conversation and sentence reading: (1) When do these palatalizable sequences merge toward a single phonetic unit? and (2) How similar are the resulting merged sounds to the corresponding English palatal segments (/š,ž/)? Both duration and spectral measurements were made. Preliminary results indicate that an alveolar fricative palatalizes more frequently and more completely when it is followed by a palatal fricative than when it is preceded by one. In addition, some palatalizable sequences merge toward the form of a palatal segment, but others merge into acoustic forms which do not correspond to any English segment. Detailed analyses of temporal and spectral characteristics can be used to evaluate alternative accounts of palatalization in terms of, e.g., deletion, gemmination, and/or anticipatory coarticulation.