The glottalic consonants of Hausa.

This thesis is concerned with the examination of the glottalic consonants of Hausa, a Chadic language spoken in northern Nigeria, Niger, Cameroon, Togo and Ghana. The glottalic consonants constitute a set of phonemes in the language whose historical and phonetic properties merit further investigation. The study is laid out in three parts. Part one (Chapters 1, 2 and 3) is the historical section. Here, a general overview of the Hausa language is given. Also discussed are several specific points made by pioneers of the genetic classification of Chadic within Afroasiatic and the reconstruction of glottalic consonants in both Chadic and Afroasiatic. The discussion here is not new but presents a summary of the literature. Hausa native words that have glottalic consonants are compared with possible cognates from other related Chadic languages from West, Central and East branches of Chadic and also from other Afroasiatic languages. Part two (Chapters 4 and 5) of the study concerns the investigation of phonation types in general. Chapter 4 gives a short account of the larynx, the mechanism of the vocal fold vibration and classification of phonation types. Chapter 5 is devoted to a review of instrumental techniques used in voice measurement. Part three (Chapter 6, the instrumental section) presents and discusses the results of a detailed electro- laryngographic analysis of the activity of the behaviour of the vocal folds in the production of the glottalic consonants and their non-glottalic counterparts as observed in the speech of educated native speakers of the language. The chapter begins with a review of the literature reporting early instrumental and non-instrumental studies of the segmental phonemes of the language. This is followed by a description of the techniques used to record, display and annotate both speech pressure and laryngographic waveforms. Both qualitative and quantitative analysis of the waveforms are presented. The most important parameter in the quantitative analysis is estimated open quotient (OQ) derived from the Lx waveforms measurements of Fundamental Frequency, duration and Voice Onset Time are also given. The chapter concludes by presenting the results of the experiment: 1 OQ increases in anticipation of plain voiceless consonants, and is relatively high at the consonantal release; 2 OQ decreases in anticipation of laryngealized consonants including the glottal stop (for most speakers) and less sharply for the ejectives; 3 OQ remains approximately at the speaker's modal value for the plain voiced consonants; 4 The laryngealized segments tend to lower pitch at the left of the consonant; and 5 They also tend to be longer than their plain counterparts.