An acoustic analysis of the monophthongs of mesolectal-to-acrolectal Nigerian English

Abstract

This dissertation provides an acoustic analysis of the monophthongs of Nigerian English using data harvested from selected television newscasters, who are expected to be mesolectal-to-acrolectal speakers of this variety of English. The "special speech training" which, as Gut (2004: 817) notes, newscasters receive, arguably makes their speech a reliable source of the standard variety. The study not only confirms the existence of such sociolects, which have been impressionistically identified, for example, by Banjo (1971, 1993, 1996), but also includes a gender dimension. Thus, in the process of unveiling which gender gravitates further away from the influence of the mother tongue and towards the reference accents, that is, RP or GenAm, the relevant acoustic properties of the monophthongs of the prestige Nigerian English norm were identified. To accomplish this also required secondary data for comparative purposes, which included acoustic data for the reference accents and impressionistic data for the relevant Nigerian (mother-tongue) indigenous languages. A balanced representation of both genders (seven male newscasters and seven female newscasters) was ensured in terms of the research subjects from whom the primary data was sourced. Approximately one-hour-long news broadcasts anchored by male and female newscasters were subjected to PRAAT instrumental analysis to determine first and second formants (corresponding to the vowel height and frontness/blackness) as well as the length of each monophthong. The findings revealed that while the subjects as a whole were able to reliably contrast the length difference between the traditionally short and long monophthongs, as established by the statistical instrument of t-testing, the female subgroup showed longer length contrasts than their male counterparts. In terms of quality, the subjects were generally able to show quality contrasts between relevant minimal pairs, say KIT-FLEECE, in relation to the available RP acoustic data. The patterns of contrast, however, show that mesolectal-to-acrolectal Nigerian English is modeling RP rather than GenAm. Notably, the female speakers were found to be making considerably better quality contrasts than the male speakers. All of this is in alignment with an already established finding in sociolinguistics, as enunciated by Labov (1990) and Trudgill (1972), that the female gender tends to use prestige variants more than their male counterparts. Among the L1 subgroups, the L1-Hausa and L1-Igbo speakers modeled RP more closely than did the L1-Yoruba speakers, a development that was traced to both linguistic and historical factors.