A story of two schwas: a production study from Tashlhiyt

Abstract

What is the status of schwa in Tashlhiyt? How does it pattern at the segmental and suprasegmental levels? This paper studies these questions by examining the frequencies with which schwa is inserted in different segmental contexts, and by exploring the nature of the interaction between schwa insertion and prosodic prominence. The results show the coexistence of two types of schwa in the language: a transitional schwa and a prosodically triggered schwa. The former is ignored by the phonological system of the language; its occurrence depends solely on the phonetic characteristics of the adjacent consonants. The latter surfaces whenever a prosodically prominent feature does not attach to a vowel or sonorant already present within the word. Both schwas are argued to be manifestations of a more general principle governing the coordination pattern of consonant sequences in Tashlhiyt.

Expanding the gestural model of lexical tone: evidence from two dialects of Serbian

“…This relationship need not be unidirectional; many previous studies have shown that pitch contours can influence segmental timing, e.g. the duration of rising vs. falling tones (Yu 2010) or even the insertion of segmental material in Italian (Grice, Savino, & Roettger 2018) or Berber (Ridouane & Cooper-Leavitt 2019). It is not possible to express this conditioned variance with a range of stiffness values specified for a gesture, analogous to a window of F0 values for a tone target, or a window of constriction degree or location for an oral gesture: these ranges of target values are typically meant to express unconditioned variability in the production of a gesture.…”

Micro‐prosody

“…2 This timing pattern has been termed an 'open transition' because there is a period of open vocal tract, that is, no constriction, in the transition between the first consonant and the second (see, e.g., Catford, 1977, p. 213). Notably, open transition is not the same as having a vowel between the consonants (Figure 1 conditions for voicing may be met, and it is possible to observe a short period of voicing in the transition between consonants, as has been documented in, for example, Tashlhiyt Berber (Ridouane, 2008;Ridouane & Cooper-Leavitt, 2019) and varieties of Moroccan Arabic (Gafos et al, 2010;Heath, 1987). This micro-prosody is schematised in the bottom right of Figure 1.…”

The secret morphology of Tashlhiyt Berber

“…One argument for this is that schwa is never a segment in Tashlhiyt (Dell & Elmedlaoui 1985, 2002; cf. also Ridouane 2008, Ridouane & Cooper-Leavitt 2019. Hence, the schwa should not be a version of the underlying segment /a/.…”