Towards a comparative typology of emphatics: across Semitic and into Arabic dialect phonology

This thesis investigates the role of emphatics within the Semitic sound system as the basis for a typology of Semitic emphatics. In seeking to defme the term 'emphatic', since emphatics are realised in some Semitic languages as ejectives, and in others as 'pharyngealised', or 'backed', the phonetic aspects of both are investigated. I present acoustic analyses of Tigrinya and Arabic (Peninsula Arabian and Iraqi) emphatics, paying particular attention to perceptual salience. Firstly, the notions of 'noise-lag' and 'stop-lag' are discussed and exemplified in relation to ejectives; secondly, I present and evaluate analyses of VOT in Arabic, showing that there is dialectal variation in the voicing series (i.e. two-way vs three-way). Further to this, I discuss the phonological composition of the various emphatics and gutturals, proposing structural representations broadly within an element -theoretic framework. I then take a diachronic angle, looking at Proto-Semitic and the development of the sound systems of the Semitic languages, in particular the Semitic triads, and the development of 'backed' emphatics as a product of changing sound systems. I argue that Proto-Semitic laterals were not part of the 'triad' system and that the voiced lateral fricative was 'backed'. The emphatic trajectory hypothesis is evaluated and theoretically contextualised, and I show that dialectal variation in the voicing series of Arabic is relevant to the variant phonological systems of the dialect types discussed. A preliminary comparative investigation into Arabic dialect sound systems is then presented. I discuss dialect classification and detail a set of key variables for each dialect group. The thesis then discusses the issue of 'emphasis spread', analysing data from four different dialect types. The data is discussed in terms of sound systems, and the traditional analysis of 'emphasis spread' is disputed. I show how the various sound systems of Arabic are characterised by resonance patterns, which are a crucial part of what is normally taken to be 'emphasis spread', and that there is an active process of 'fronting' (imiila) which is crucial to an analysis of 'emphasis' (tafx'im). The thesis concludes with an evaluation of the research, stressing the need for systematic and consistent cross-dialectal analyses of both the phonetics and the phonology of Semitic emphatics. I outline how this can be used in future work to develop a comprehensive comparative typology, towards which this thesis is a preliminary contribution.