Bozena Tieszen Charles Read
Dept. of Linguistics, Univ. of Wisconsin, 1220 Linden Dr., Madison, WI 53706
In Polish, as in some other Slavic and Germanic languages, word-final obstruents are devoiced. Three monolingual speakers in Krakow recorded a sample of ine minimal pairs containing final /p,t,k,b,d,g/ preceded by each of the vowels /i,a,u/ in a meaningful carrier sentences. Each word was followed by a voiceless obstruent, in one case, and by a vowel in another. The speakers had also recorded some of the same words in the same environments in a narrative passage. Three durations were measured: Of the preceding vowel, of the stop closure, and of glottal pulsing into the closure. For all three speakers and all environments, the durations of the preceding vowel and of the stop closure did not differ significantly between underlyingly voiced and underlyingly voiceless stops. However, for all three speakers and all three places of articulation, glottal pulsing into the closure was significantly longer for underlyingly voiced stops than for their underlyingly voiceless counterparts. This outcome held for the words in the narrative passage as well as those in the carrier sentences. These results contrast with those obtained by Slowiaczek and Dinnsen [J. Phon. 13, 325--341 (1985)].