Intonation in Ekegusii, Its Interaction With Information Structure and How It is Perceived by Native Speakers

Abstract

The work presented in this thesis is a description of the structure of intonation and its interaction with information structure units, perception and interpretation in Ekegusii. This description was aimed at meeting the following objectives: to identify and describe the phonetic structure of intonation phrases in Ekegusii; to analyse how intonation encodes focus in Ekegusii utterances and to investigate Ekegusii native speakers‟ level of precision in the perception of intonation phrases in the language. Consequently, 3744 utterances were analysed following the basic principles of the Autosegmental-Metrical Theory. Data were analysed at the production and perception levels. At the production level, 24 purposively selected participants read aloud utterances presented to them. The utterances of each one of them were recorded and fed into a speech analysis software, PRAAT. The recorded data were used in the description of the structure of intonation phrases in Ekegusii. This structure, the research has established, consists of boundary intonemes and F0 fluctuations. The major finding in this part is that all Ekegusii utterances, irrespective of the lexical tones, have final L%, L-L% and H-L % boundary intonemes but could be distinguished by the differences in the gradient of declination and the F0 fluctuations associated with them. Findings also indicate that polar interrogative sentences were pronounced at the highest F0, 211 Hz. These were followed by the constituent interrogatives, 202 Hz; the imperative sentences, 201 Hz and the echo interrogatives, 194 Hz. Declarative paratones were articulated at the lowest F0, 185 Hz. In addition, results have shown that participants had different F0 registers with findings indicating that children and the advanced-aged participants had higher F0 registers (242 Hz and 205 Hz, respectively) than the youth (185 Hz) and middle-aged participants (163 Hz). An overall decrease in F0 with advancement in age upto the middle-aged period before an increase at the advanced-aged period for both male and female participants was observed. Equally, female speakers produced utterances at higher F0 registers than their male counterparts did across all the age categories. The female children‟s pitch register was 249 Hz while the male children‟s was 233 Hz. The female youth‟s register was 223 Hz while their male conterparts‟ was 147 Hz. The middle-aged females spoke at 211 Hz while their male counterparts at 119 Hz. The advanced-aged female participants spoke at 231 Hz while the advanced-aged males spoke at 183 Hz. The female participants‟ F0 bottom line was 184 Hz while their top line was 244 Hz. The male participants‟ F0 bottom line was 157 Hz and the top line 179 Hz. The average F0 register for the female participants was 218 Hz while the male participants‟ was 168 Hz. The study equally found out that two intonation strategies were used to signal focus in Ekegusii. These were F0 fluctuations and rephrasing. The argument, predicate, contrastive and sentence focus structures were articulated at different F0s (199 Hz, 198 Hz, 196 Hz and 195 Hz, respectively). Focus was also signalled through the insertion of an intermediate intonation phrase to the left or right of the focused constituent. This intermediate intonation phrase was marked with an H- or L-tone. In terms of perception and interpretation of intonation types, the study has established that the echo interrogative utterances were the easiest to identify while the polar interrogative ones posed the greatest challenge to interpret. The contrastive focus structure also presented the greatest challenge to interpret while the predicate focus posed the least challenge to the participants. Again, female participants had a higher identification rate than their male counterparts. The male participants seemed to concentrate more in correcting the utterances given than classifying them. They also gave arbitrary responses. This study is invaluable in acoustic phonetics research especially in tone languages.