A sociophonetic analysis of Jamaican vowels.

This thesis provides an acoustic examination of the vowel systems of eleven Jamaican Creole-dominant (St. Thomas parish) and nine Jamaican English-dominant (Kingston) speakers and seeks to link results with sociolinguistic factors. This research was motivated by an interest in two enduring questions in creole linguistics: the possibility of contrastive vowel duration in Jamaican language varieties, and the magnitude of vowel quality differences between Jamaican English and Jamaican Creole. Data were collected in rural and urban Jamaica using word list, casual conversation, and picture description tasks. Although fourteen orthographic vowel categories were investigated, of particular interest were the relative roles of spectral (quality) and temporal (quantity) characteristics in contrasts between vowels traditionally described as belonging to long:short oppositions (high front /i:, i/, mid front /e:, &egr;/, low /alpha:, a/, and high back /u:, u /). Spectral overlap was assessed using a formula based on a representation of vowel distributions as ellipses in acoustic space. Temporal overlap was assessed using duration ratios of long:short vowel pairs. Variation in vowel quality and quantity was correlated with both extralinguistic and phonetic factors. St. Thomas and Kingston speakers displayed spectral differences between vowel quality oppositions, but for some contrasts temporal differences were of greater magnitude for St. Thomas speakers than for Kingston speakers. Additionally, females in both groups showed larger temporal differences within long: short pairs than males. Manner of articulation and voicing of a following consonant primarily affected vowel duration. A social network strength index, based on degree of orientation into a close-knit rural community, was adapted for use in Jamaica. However, with one exception, network strength was not a reliable predictor of acoustic features: males in Kingston with low network scores tended to show a high degree of r-lessness. In an informal study of metalinguistic awareness, a listening task was used to elicit speakers' comments concerning what constitutes and distinguishes Jamaican Creole and English speech. Speakers related age, social class, and region of residence with differences in Jamaican speech and described a small number of stylistic, morphosyntactic, and phonological features which they ascribed to Jamaican English, Jamaican Creole, or American English.PhDLanguage, Literature and LinguisticsLinguisticsUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/131814/2/9929976.pd