Intonation variation of declarative questions by Kenyan speakers of english

The focus of this paper is to highlight the various manifestations of intonation patterns to declarative questions by Kenyan speakers of English who use English as a Second Language (ESL). Specific references are made to two ethnic groups (Bukusu and Nandi), who produced maximally distinct accents during the pilot study. The paper describes the range of intonation patterns that exist among Bukusu and Nandi ESL speakers and further establishes whether there are features of intonation that distinguish the two groups. The objectives are achieved with respect to auditory and acoustic analysis of both read and spontaneous speech. Acoustic analysis by use of Prosogram script for Praat software and Tone and break indices (ToBI) for prosodic annotation is adopted to depict nuclear accent placement, accent types and their associated boundary tones. Additionally, the intonation patterns are coded and plotted in Ms Excel for variability tests. The British School is combined with the Autosegmental-Metrical (AM) theory of intonation for discussion. The findings are explored departing from an assumption that these ethnic groups demonstrate dissimilar intonation patterns based on their underlying first languages (L1) substrate. Instrumental analysis reveals that the Bukusu adopt a high nuclear that ends in a low boundary tone (H* L%). While the Nandi ESL speakers predominantly yield a rising nuclear followed by a high boundary tone (L*H H%).