Phonological aspects of Kipwanji

This study attempts an account of the phonology of Kipwanji from a primarily descriptive view point. Specifically, the study sought to examine the role assumed by the phonological processes that affect vowels and consonants in preserving the phonological configuration of the phonologically possible word or morpheme. Accordingly, this study is approached from a non-linear representation (auto segmental theory as postulated by Goldsmith, 1976) perspective. This theory treats phonological representations as consisting of parallel tiers of phonological segments, both tonal and segmental. With regards to methodological approaches, the study employed a descriptive research design. This enabled the researcher to collect the data and describe the state of affairs as existed in the Kipwanji speaking community. Primary data were collected using focus group discussion, participant observations, and interviews. On the other hand, secondary data were extracted from Bible story books and tale books as described in Chapter Three. The findings reveal that Kipwanji language attests for twenty three (23) consonant sounds, glides and prenasalised consonants inclusive as shown in Table 2. The distributional test indicated that the language only allows consonants in the word initial and medial position not in the word final. Moreover, Kipwanji language attests for seven vowel phonemes (both long and short) which are divided into two categories; +ATR and —ATR, depending on whether the tongue root is moved forward enlarging the throat cavity or is in a neutral position or moved backwards reducing the throat cavity (see table 4). Furthermore, it is revealed that the canonical structure of Kipwanji syllable structure is CV but not limited to $V$, $C$, $CV$ and $CGV$. The syllables are conditioned by constraints that govern the sequence of segments (phonological sequential constraints (PSCs) in the language. However, once words are formed out of syllables, they are subject to rule-sensitive mechanisms (phonological processes) through which the native speakers are able to recorganise and correctly pronounce the acceptable words of the language by posing the phonological rules which are in conspiracy. It is also revealed that, the rules in auto segmental representations have the target, the change, and the environment all written as they would appear in the auto segmental tree format. Moreover, the rules are ordered with respect to the satisfaction of the structural descriptions that allow more than one rule to apply.