Abstract
This study is concerned with the voice onset time (VOT) of the voiced plosives /b, d, g/ in word initial position, especially with the co-occurrence relationships between the half voicing (VOT taking plus value) of voiced plosives in word initial position and nasalization of voiced plosives or voicing of voiceless plosives in the word medial position. The data were collected from 20 residents of Kagoshima prefecture, and results were compared to those previously published for the Tohoku, Kitakanto, northeastern Nigata, Kanto and Shikoku regions. Additionally, the Kagoshima VOT type in word initial voiced plosives was studied. Results disprove the supposed co-occurrence between half voicing and nasalization/voicing. Also, they indicate that Kagoshima's VOT type is comparable to the "unified type for full voicing" found in Kanto and Shikoku, as well as to the "intermediate type" between the "unified type for full voicing" and the "unified type for slight voicing" found in Kitakanto and northeastern Nigata.
