Fronted /s/ in General White South African English1

1. The author would like to thank Ms Sally Hunt from the English Language and Linguistics Department, Rhodes University for first drawing his attention to the feature which is the focus of this article. Thanks are also owed to anonymous reviewers whose comments and suggestions made this paper a far better one.

Abstract

General White South African English (WSAE) is distinguishable from other accents of WSAE by a number of Lanham and MacDonald's (1979, 41–51) so-called ‘Natal variables’ (e.g. a fronted and glide-weakened /aI/) and other prestige values such as lowered and retracted /æ/ (Bekker and Eley 2007). Advanced values for these variables have been found to be particularly prevalent among young females from the Johannesburg area (Lanham and Macdonald 1978, 62; Bekker and Eley 2007). A variable that has not, however, been mentioned in the literature is what appears on an articulatory level to be a fronted denti-alveolar or dentalised /s/ i.e. In order to verify this impression, data drawn from recordings of subjects belonging to two groups were subjected to an acoustic analysis. The first group consisted of young white female speakers from the northern suburbs of Johannesburg who had attended private schools in the area. The second group consisted of similar subjects from elsewhere in South Africa. Four parameters (or spectral moments) are the focus of this analysis: the mean of each spectrum derived from each fricative, as well as each spectrum's variance, skewness and kurtosis (see Jongman et al. 2000, 1253–1254). Basic statistical methods were then employed to determine whether the two groups differ significantly along any of these parameters, and thus to confirm or disconfirm the initial impressions mentioned above. Results indicate that, at least with regard to the variance parameter, there is a substantial difference between the two groups, thus providing suggestive evidence for the existence of /s/-fronting in at least one subvariety of WSAE.

Notes

1. The author would like to thank Ms Sally Hunt from the English Language and Linguistics Department, Rhodes University for first drawing his attention to the feature which is the focus of this article. Thanks are also owed to anonymous reviewers whose comments and suggestions made this paper a far better one.