Zwara (Zuara, Zawara, Zouara) Berber is spoken in Zwara, a coastal city with over 40,000 inhabitants in Libya near the border with Tunesia. It belongs to the eastern Zenati group of Northern Berber. A close variant is spoken in Ghadamès, an oasis near the Algerian-Tunisian border 400 km south of Zwara (Louali-Raynal 2000). Research on the language is restricted to phonological alternation in the verbal and nominal morphology (Mitchell 1953, 1957), and a response to Mitchell (1953) on personal affixes by Hamp (1959). Preliminary fieldwork has been carried out by Carlos Gussenhoven but is limited to segmental issues.
Zuara appears to have a fixed location for the syllable with which some pitch obtrusion is aligned, which is provisionally taken to be a stressed syllable. In most words, it is penultimate, as in /1;J.fud/ ‘knee’, /l.1qah.wut/ ‘coffee’, less commonly, this is the word-final syllable, as in /am.1naj/ ‘horseman’.
Because word-final /rt/ is realized as [rət], stress may appear to be antepenultimate in words like /l.1k;Jn.d-;Jrt/ ‘shoe’, /1t-;J-n-.y-;J-r-t-/ ‘pan’, pronounced [l.1k;Jn.d-;Jr-;Jt-], [1t-;J-n-.y-;J-r-;J-t-]. The auditory impression of penultimate stress can be quite different from equivalent structures in European languages. This is particularly true for words in which the stress would appear to fall on a syllable in which the V-position is filled by the first half of a geminate, as in /a.1tr.ras/ ‘people’, /a.1wf.far/ ‘grass’, /a.1df.fu/ ‘apple’ – where the stress falls on [r], [f], and [f], respectively. Strikingly, the transition from the onset consonant to the geminate does not give the impression of a vocoid, however brief. In non-geminate cases, the syllable nucleus more clearly has a brief [ə], which may have to be interpreted as an intrusive vocoid, as in /da.1mz.war/ [da1məzwar] ‘he is first’.
In other cases, such as when a geminate precedes or follows an identical consonant, a brief [ə] is inserted, as in /1tt.t;J15/ [1t;Jt.t;J15] ‘I eat’. At the same time, there does not appear to be a phonological shift in the position of the pitch obtrusion, but the way in which transitional vocoids affect the phonetic alignment of these pitch obtrusions is currently unknown, as is the shape of these obtrusions.
For the intonation analysis of Zwara, the analysis of the segmental and word prosodic system in Zwara will be completed, in particular the manifestation of word stress and the independent status of schwa.