samedi 14 août 2010

Pharyngealization and pharyngeals in Tashlhiyt Berber

Pharyngealization and pharyngeals in Tashlhiyt Berber:
A descriptive and comparative study
Rachid Ridouane
Laboratoire de Phonétique et Phonologie (UMR 7018) CNRS/Sorbonne Nouvelle
rachid.ridouane@univ-paris3.fr


Berber is an Afro-Asiatic language spoken in large parts of North Africa, mainly in
Morocco and Algeria. Tashlhiyt, the variety investigated here, is spoken in the Southern
part of Morocco. The phonemic inventory of this language is argued to include the
following set of dorsopharyngealized phonemic consonants, all of which are coronals:
/t, d, s, z, n, l, r, , /. Each of these consonants has a geminate counterpart,
yielding thus a system of eighteen pharyngealized consonants. These phonemes spread
emphasis to all the other segments present in a word or even a sequence of words
(Elmedlaoui 1985, Boukous 1987). In addition, Tashlhiyt also has the so-called
pharyngeal fricatives, which may be either singletons /, / or geminates /, /.
In this presentation, I will first provide a brief survey of the distribution of these
consonants, and show that while some pharyngealized consonants are very common (i.e.
/t, d, s, z/), others are quite rare and found mainly in some Arabic loanwords (e.g. /n,
l, r/). The productivity of their geminate counterparts will also be discussed. The core
of this presentation is concerned with the phonetic characteristics of these segments. The
phonetic study is divided in two parts.
The first part is concerned with the phonetic description of the following pharyngealized
consonants: /t, d, s, z/ and their geminate counterparts /tt, dd, ss, zz/. The aim is
to determine how these segments are phonetically implemented and how they differ
from their non pharyngealized counterparts /t, d, s, z/ as well as from the uvular
consonants /q, , /. I will present some acoustic, fiberscopic and ultrasound data to
tackle this issue. The acoustic data, based on the productions of five subjects, concern
both the temporal characteristics of these consonants and their qualitative effects on the
surrounding vowels /i, u, a/. Fiberscopic data, based on the production of two speakers,
show how voiceless pharyngealized consonants /t, s/ are produced at the laryngeal
level. The aim is to show how the timing and the amplitude of glottal opening may
account for the differences observed at the acoustic level between these segments and
their non-pharyngealized counterparts /t, s/ and the uvulars /q, / (namely with respect
to the feature [spread glottis] and the shorter VOT displayed by /t/ and /q/, compared
to /t/). The ultrasound data, based on the productions of three speakers, provide lingual
measures of dorsopharyngealized consonants compared to their plain counterparts as
well as to uvulars /q, , /. The results will be presented concerning tongue root activity,
timing of articulatory events, overall tongue shape, and interactions between vowels and
lingual consonants.
The second part of the phonetic study provides fibroscopic data during the production
of /, / and their geminate counterparts. Based on the laryngeal model developed by
Esling and colleagues (e.g. Esling 1996, Edmondson & Esling 2005), the aim is to determine
whether these fricatives are real pharyngeals or whether they are best described as epiglottals
or ayepiglottals. To conclude, a brief comparative study will be provided where data from
Tashlhiyt are compared to data from Moroccan Arabic, a national language with which
Berber has been in contact for centuries and which resembles it in many respects. Data
from Moroccan Arabic are based mainly on findings by C. Zeroual (2002) on the variety
of Moroccan Arabic spoken in Oujda area.

source: http://www.ncl.ac.uk/linguistics/assets/documents/RachidRidouane.pdf

Aucun commentaire: