Volume 55, Issue 1-2

An emerging concept for the characterization of the form of babbling and early speech is ‘frame dominance’: most of the variance arises from a frame provided by open-close mandibular oscillation. In contrast, the tongue – the most versatile articulator in adults – plays only a minor role in intersegmental and even intersyllabic changes. The contribution of another articulator – the soft palate – to time-domain changes in babbling was evaluated in an acoustic analysis of 433 consonant-vowel-consonant sequences produced by 3 infants. Strong nasal effects on vowels in symmetrical consonantal environment were observed in the form of a lower frequency first formant region in low vowels and a lower frequency second formant region in front vowels. These results, the first of which also occurs in adults, were complemented by perceptual tendencies for transcribers to transcribe more mid vowels relative to low vowels and more central vowels relative to front vowels in nasal environments. Thus the soft palate is like the tongue in making only minor contributions to time-domain changes in babbling, and this is considered to be additional evidence for the frame dominance conception.

Voiceless velar stops may become palatoalveolar affricates before front vowels. This sound change is not only one of the most common types of palatalization, but is a very common sound change in the world’s languages. Nevertheless, we do not have an adequate understanding of how this sound change takes place. Three experiments reported here test the hypothesis that velar palatalization is the result of listeners’ on-line perceptual reanalysis of fast rate speech. It is shown that velars before front vowels are both acoustically and perceptually similar to palatoalveolars. This supports the proposal that velar palatalization is perceptually conditioned.

Electropalatographic and F2 frequency data in /VlV/ sequences reveal more prominent C-to-V effects for Catalan dark /l/ than for German clear /l/, more so in the /i/ context than in the /a/ context, which is in agreement with the existence of high lingual requirements on the formation of two constriction places for dark /l/. German clear /l/ exerts a similar amount of F2 displacement on both vowels which may be indicative of the tongue dorsum being directed towards a target position; this is also suggested by dorsopalatal contact and formant frequency data showing less vowel-dependent variability than clear /l/ in other languages though more so than Catalan dark /l/. Salient anticipatory requirements for the implementation of /l/ in the two languages block V1-dependent carryover effects to a large extent which results in more prominent vocalic anticipation than vocalic carryover. This directionality trend in vocalic coarticulation is more obvious for Catalan dark /l/ than for German clear /l/ (in agreement with the former consonantal variety requiring more anticipation than the latter) and opposes German /l/ to clear /l/ in other languages (i.e., the less constrained /l/ variety of Spanish may favor vocalic carryover over vocalic anticipation in VCV sequences).

Degema is an Edoid language of Nigeria whose ten vowels are organized phonologically into two sets of five. The two sets are thought to be differentiated by the degree of tongue root advancing. This paper examines the acoustic nature of these vowels as represented in field recordings of six speakers. The most consistent acoustic correlate of the tongue root contrast was found to be the first formant frequency which consistently distinguishes four of the five vowel pairs, the exception being the two low vowels. Three of the five pairs could also be distinguished by F₂ , though the direction of the difference was not consistent. Additionally, a comparison of corresponding advanced and retracted vowels using a normalized measure of relative formant intensity demonstrated that this correlate could also distinguish them in general, but only operated reliably in two of the five vowel pairs. The pair of low vowels could not be distinguished from each other by any of these measures. Finally, a perceptual study was conducted which demonstrates that Degema speakers do not classify their vowels very well using formant frequencies as the sole acoustic variable; only the two pairs of mid vowels were reliably singled out by native listeners from an array of synthesized vowels.