Volume 75, Issue 2

This study examines formant and spectral moment data for the apical and retroflex lateral sounds /l ɭ/ of the Dravidian languages Kannada and Malayalam, together with the rhotic /ɻ/ of Malayalam. Data are presented for 10 male speakers of each language. We find that the first spectral moment is lower for retroflex laterals than for alveolar laterals, and lower for the rhotic /ɻ/ of Malayalam than for the retroflex lateral in the same language. Differences emerge when the retroflex lateral of Kannada is compared with the same sound in Malayalam. For both languages, F1 is higher and F3 and F4 are lower for the retroflex /ɭ/ than for the alveolar /l/. However, F2 is higher for the retroflex than for the alveolar sound in Kannada, but lower in Malayalam. This difference is also reflected in differences in the second spectral moment between the languages. It is suggested that since proximity of F2 and F3 is known to be a defining feature of the rhotic /ɻ/ in Malayalam, principles of phonetic dispersion apply to keep F2 from becoming too close to F3 for the retroflex lateral /ɭ/ of Malayalam, but not for the same sound in Kannada.

Although numerous studies have investigated supraglottal strategies for signalling voicing in fricatives, there is still no agreement about the precise characteristics of tongue-to-palate contact timing during voiced as opposed to voiceless fricatives. In this study we use electropalatography (EPG) to investigate articulatory and coarticulatory characteristics of tongue-to-palate contact timing during /s/ and /z/ in English. Five typically speaking participants, speakers of Southern British English, produced 500 trochaic words containing the intervocalic alveolar fricatives /s/ or /z/. The time between the start of the frication and the maximum contact at the place of articulation was expressed as a percentage of each fricative's total duration (time to target, TT).This measure was used to analyse articulatory and coarticulatory timing during /s/ and /z/. Data for absolute timing were also presented. The results showed that the time between the start of the frication and the maximum contact point was longer for /s/ than for /z/. This difference was consistent across speakers but was not significant for all of them. The results of the coarticulatory effects showed that the influence of vowel context on TT values for /s/ and /z/ did not differ significantly, but there was a tendency for /z/ to be more resistant to coarticulation effects than /s/.

Two of the major problems with rhotics are: (1) rhotics, unlike most other classes, are highly resistant to secondary palatalization, and (2) acoustic cues for rhotics as a class have been elusive. This study examines the acoustics of Upper and Lower Sorbian rhotics. Dynamic measures of the F1-F3 and F2-F1 were recorded and compared using SSANOVAs. The results indicate there is a striking delay in achievement of F2 for both the palatalized rhotics, while F2, F1, and F2-F1 are similar for all the rhotics tested here. The results suggest an inherent articulatory conflict between rhotics and secondary palatalization. The delay in the F2 increase indicates a delay in the palatalization gesture. This is likely due to conflicting constraints on the tongue dorsum. There was also an overlap in the F2 and F2-F1 for both the uvular and alveolar rhotics. This suggests a strong acoustic cue to rhotic classhood is found in the F2 signal. The overall formant similarities in frequency and trajectory also suggest a strong similarity in the vocal tract shapes between uvular and alveolar rhotics.

Consonant reduction is often treated as an allophonic process at the phonological planning level, with one production target (allophone) being substituted for another. We propose that, alternatively, reduction can be the result of an online process driven by prosodically conditioned durational variability and an invariant production target. We show that this approach can account for patterns of coronal stop (/t/, /d/, and /n/) production in both American English and Spanish. Contrary to effort-driven theories of reduction, we show that reduction does notdepend on changes to gestural stiffness. Moreover, we demonstrate how differences between and within a language in the particular articulatory postures used to produce different coronal stops automatically lead to reduction to what have normally been considered distinct allophones - coronal approximants ([ð̞]) and flaps ([ɾ]). In this way, our approach allows us to understand different outcomes of coronal stop reduction as the dynamic interaction of a single process (durationally driven undershoot) and variable spatial targets. We show that these patterns are reflected across a wide variety of languages, and show how alternative outcomes of reduction may fit within the same general framework.