Durational correlates for differentiating consonant sequences in Russian

In Russian, the same consonant sequences are permitted in various phonotactic environments. The presence of a word boundary or reduced vowel can be phonologically contrastive, and both learners and experienced listeners may rely on fine acoustic cues to discriminate between phonotactic possibilities. In this study, durational characteristics of consonant sequences are examined to establish whether speakers use duration to distinguish (a) word-initial clusters (#C1C2), (b) consonant–schwa–consonant sequences (#C1əC2), and (c) sequences divided by a word boundary (C1#C2). Both monolingual native Russian speakers and bilingual Russian–English speakers produced several types of target sequences: stop+consonant, fricative+consonant, and nasal+consonant. Results show that C2 is significantly longer in C1#C2 than in other sequences. For #C1C2, when C1 is a stop, there is no significant difference in duration when compared with other sequence types, though C1s of other manners are significantly shorter. Differences in C1 burst duration for stops are consonant-specific, but a longer interconsonantal duration is a reliable cue to schwa presence in #C1əC2. These results are discussed with respect to their implications for gestural coordination, segmentation, and language learning.