Working paper Open Access
Tilsen, Sam; Zec, Draga; Bjorndahl, Christina; Butler, Becky; L'Experance, Marie-Josee; Fisher, Alison; Heimisdottir, Linda; Renwick, Margaret; Sanker, Chelsea
This paper is a preliminary report on a series of experimental investigations of articulatory coordination in word-initial consonants in a variety of languages. These experiments were designed, conducted, and analyzed in the context of a Linguistics graduate seminar on syllable structure taught at Cornell University in the spring of 2012.
We aimed to replicate and extend previous studies of complex onset coordination in Italian, French, and English, and to investigate coordinative patterns in Serbian and Hebrew, languages in which complex onsets have not been previously studied. The cross-linguistic coverage offered by previous studies is currently quite limited, hence our replications of prior studies and investigations of unstudied languages are a valuable contribution to the understanding of phonetic and phonological aspects of syllable structure.
The motivation for studying word-initial consonant sequences is derived from two sources. First, investigations of the syllable as a phonological domain have in recent years turned to articulatory differences between onset and coda consonants as a source of evidence for this domain (Browman & Goldstein, 1988; Krakow, 1989; Sproat & Fujimura, 1993; Byrd, 1995).
Because the phonotactics of word-initial consonants are often similar to the phonotactics of syllable onsets, our study is relevant to understanding how articulatory patterns are associated with syllable structure. Second, available evidence indicates that two distinct coordinative patterns are available for word-initial consonant sequences, and in some cases (discussed below) this distinction correlates with phonological patterns involving syllabification. Hence cross-linguistic differences in articulatory timing of word-initial consonant sequences may be useful for understanding crosslinguistic differences in syllabification.
Our investigation and analysis of articulatory timing is guided by theoretical approaches to gestural coordination in the frameworks of Articulatory Phonology (Browman & Goldstein, 1988, 1990, 2000) and task-dynamics (Saltzman & Munhall, 1989).
Articulatory phonology holds that in some languages, articulatory gestures associated with consonants in the onset of a syllable exhibit an in-phase timing relation to the following vowel. This C-V coordinative pattern contrasts with a theorized anti-phase C-C coordinative pattern. In a complex CC onset, both consonantal gestures are hypothesized to exhibit C-V coordination, but this competes with the anti-phase C-C timing relation.
The result of these competing coordinative patterns is a compromise known as the c-center effect: relative to a singleton C, the timing of the prevocalic C gesture in a CC cluster is shifted closer to the following vowel, and the timing of the initial C gesture is shifter earlier relative to the vowel. Crucially, the midpoint (c-center) of these gestures maintains a stable relation to the vocalic gesture. This type of timing pattern in a word-initial cluster is henceforth referred to as complex organization.
In contrast, in languages where word-initial consonants preceding the prevocalic C are not syllabified with the following V, these consonants are hypothesized not to exhibit C-V coordination. In that case, the timing of the rightmost C gestures should remain stable when preceded by one or more consonants. This type of pattern is referred to as simplex coordination.
Articulatory studies have tested these predictions by comparing a variety of timing intervals in #CVC, #CCVC, and #CCCVC forms. Below we review these studies.