Atelier de Phonologie - Brandão, Faust & Lampitelli

Joaquim Brandão de Carvalho (Université Paris 8, CNRS SFL), Noam Faust (idem), Nicola Lampitelli (Université Paris 10, CNRS MoDyCo)

Bendjaballah & Ségéral (2014) propose a new binary feature [±idle glottis]: voiced and ejective consonants are [+idle glottis] and voiceless ones are [-idle glottis]. This distinction, they convincingly argue, plays a central role in several phenomena in the South Semitic language Mehri, spoken in Oman. In this talk, we propose an answer to the question of how to express this binary feature in Element Theory (e.g. Backley 2011). We propose that the answer is to be found in the meeting between Georgian (Kartvelian) and Tush (Nakh), both spoken in modern day Georgia, and more specifically in the distribution and realization of the epiglottal trill /ʜ/ in Tush. In Tush, a voiced allophone of /ʜ/ emerges in initial clusters after voiceless ejectives or voiced consonants, yielding [Tʜ, Tˀʢ, Dʢ], where T/D are any plosive.

We begin our analysis by showing that Tush took from Georgian the possibility of forming “harmonic clusters”. These are clusters in which C1 is a non-dorsal obstruent and C2 is any dorsal one - but the two share a laryngeal feature. We hold that the structure of a consonant employs at least three tiers: place, manner and larynx. Unmarked /T/ involves no element on the laryngeal tier; voiced D involves the element |L|, which also heads the expression; and ejective [Tˀ] involves a non-head |L|. Given that the laryngeal feature must be shared, this leads to the attested [Tʜ, Tˀʢ, Dʢ]. The binary feature [±idle glottis] is expressible as the presence vs absence of (head or non-head) |L| on the laryngeal tier.

Backley, Phillip. 2011. An Introduction to Element Theory. Edinburgh: Edinburgh University Press.

Bendjaballah, Sabrina & Philippe Ségéral. 2014. The Phonology of ‘Idle Glottis’ Consonants in the Mehri of Oman (Modern South Arabian), Journal of Semitic Studies 59 (1): 161-204.