Date: Mar 27, 2014
What: Intergestural timing and aerodynamic parameters in Spanish syllables: three cases.
Where: BCBL auditorium
Who: Mark Gibson. Universidad de Navarra.
When: 12 noon
This presentation centers discussion round the constraints which govern systematic patterns of intergestural timing in Spanish syllables as well as their aerodynamic consequences in a global framework based on vocal tract gestures. Three different cases will be highlighted and experimental data from each case will be addressed.
The first cases address the spatio-temporal patterning of articulatory gestures in Spanish complex onsets and the repair processes which result from non-native timing configurations. Drawing on acoustic and articulatory data it is proposed that Spanish onset formation falls out from a tendency to couple pre-vocalic consonant gestures in an in-phase relationship (0º ϕ) with the following vocalic target:
Initial results indicate that Spanish complex onsets generally exhibit a stable c-center effect (consistently lower standard deviation) for the center of the consonant cluster and right vocalic anchor. There is a systematic reduction in the timing of linguo-palatal contact for the second consonant (/l/) in all /Cl/ clusters. When /s/ is added to the cluster, however, it does not form part of this temporal configuration, corroborating predictions made in Gibson (2011, 2013) and substantiating a theoretical correlation between intergestural timing and /e/ prosthesis in word-initial /sC/ clusters.
The second case addresses a series of claims vis-à-vis the syllabic association of intervocalic [h], and the correlation between aspiration and resyllabification, in Western Andalusian Spanish. In this dialect, the oral gesture for /s/ is suppressed as glottal widening allows airflow to escape through the oral cavity producing a gradient continuum of aspiration between [h] and [ɦ]. Prototypically aspiration affects phonological codas, though a number of studies have reported cases of onset aspiration. Ordinarily aspiration emerges in onsets via a resyllabification rule which reorganizes a word-final coda as the onset of the following initial vowel: (Vs#V→V.hV). To date, however, it has not been tested experimentally whether resyllabification obliges a reorganization of glottal timing such that a novel temporal relationship with the following vocalic target is formed. It is in this context that the present study asks whether the aspirated segment is really an onset.
To determine the syllabic constituency of intervocalic [h/ɦ] and the correlation between aspiration and resyllabification, simultaneous audio-signal, oropharyngeal pressure, airflow and electroglottographic signals were obtained for three native speakers of Western Andalusian Spanish (one male, two females). The subjects read paradigms of identical phonetic strings which differed with regard to the syllabic affiliation of /s/:
(a)…haz la masa ya, dijo. (/s/=phonological onset) ‘Make the dough now, he said’
(b)…hay que ir más allá, dijo. (/s/=phonological coda) ‘It has to go beyond, he said’.
As predicted, the results indicate that duration measures (for acoustic, aerodynamic and glottal abduction) for the intervocalic segment were generally higher than for syllable-final aspirated segments, matching roughly the duration values for onset [s]. Corroborating the second prediction, aerodynamic patterns show that peak flow was achieved faster and with higher amplitudes for the intervocalic segment than for aspirated codas at all prosodic margins. Coordination patterns between the glottal target and V2 were by and large sporadic and provide little suggestion of a global arrangement for the glottal gesture in relation to the vocalic target (if it were indeed possible). Timing instability between the glottis and V2 suggest that suppression of the oral gesture for [s] does not necessarily induce a timing reconfiguration such that the glottal gesture enters into an independent relationship with the vowel (i.e. glottal timing seems to remain synchronous to the informational temporal specification of the oral gesture for /s/). Thus, although the data insinuate that the aspirated segment exhibits certain acoustic and aerodynamic characteristics common to onset fricatives, it does not seem to constitute an onset in terms of coordination dynamics.
Finally, the third case addresses the variation of aerodynamic and articulatory parameters in voiced apical trill production in Spanish codas. The aim of the study is to characterize the behavior and robustness of trill production in different phonetic environments within a general framework based on speech gestures. Simultaneous acoustic, oropharyngeal pressure and airflow signals were obtained for two native Spanish-speaking subjects in a rate-controlled repetition task using /Vr.CV/ sequences. The results indicate that although the phonetic specifications of the flanking segments do affect local articulatory and aerodynamic parameters of trill production, the spatio-temporal characteristics of the trill’s closure phase remain relatively stable across categories. The results of this study may bear directly on the tap/trill variation attested in Spanish codas as well as dialectal differences in trill production.