Regional Variation in New Zealand English: the Taranaki Sing-Song Accent
Although lay people confidently assert the existence of regional varieties of New Zealand English, linguists have produced very little evidence to support such claims. There are vocabulary items special to, or favoured by, the people of Southland and the West Coast of the South Island; there are traces of non-prevocalic /r/in Southland and Otago; and there are regional differences in the playground language of New Zealand school children. Attempts to identify further differences between regions have generally not been successful. In most cases linguistic evidence has pointed to either social class or ethnic variation, but not to regional variation. Nevertheless, many New Zealanders assert that a Taranaki variety of New Zealand English exists. This study was designed to test the validity of the claim by comparing samples of New Zealand English from Taranaki with samples from Wellington. The Taranaki sample included speakers from New Plymouth (population 50,000) and the South Taranaki dairy farming community. The Wellington sample was drawn from the Greater Wellington region extending from Porirua in the north to suburbs on the southern coast of the city. Interviewees were located by the social network approach, otherwise known as the 'friend of a friend' approach advocated by Lesley Milroy (1980, 1987a). An index of rural orientation was devised to indicate the degree to which a speaker was oriented towards town or country. This proved helpful in distinguishing between genuinely regional differences, and rural versus urban differences. Factors of gender and age were also considered. It has been claimed that Taranaki English has a 'sing-song' quality, suggesting that an investigation of the intonation of Taranaki speakers would be worthwhile. Comparing features of the intonation of a Taranaki sample with a Wellington sample, this thesis attempts to isolate and measure what contributes to the 'sing-song' perception of Taranaki English. 'Singsong' in this context was taken to mean that the speaker had dynamic pitch; in other words their speech was characterised by a lot of movement up and down in pitch. Auditory analysis of speech samples was undertaken, and intonation features were derived from that analysis. Averaging the number of times a speaker changed pitch direction in each intonation group and then in each accent unit provided global measures of changes in pitch direction. Analysis of nuclear accents gave an indication of whether speakers favoured tunes which were characterised by pitch movement. And analysis of the manner in which accents were approached, whether with a boosted step up in pitch, or with a more standard onset, provided a narrower focus on the amount of pitch movement present. Results indicated that, in general, most Taranaki speakers in the sample showed more pitch dynamism than the Wellingtonians; for some features the males showed more pitch dynamism than the females; and, overall, the elderly speakers showed more pitch dynamism than the younger speakers. There were, however, important exceptions to these generalisations. Factors of Location, Gander and Age interacted significantly for all but one of the features examined and there were clear indications that intonational patterns are undergoing change in both regions studied. Explanations for the exceptional cases are explored in the thesis, and sociolinguistic, social network and geolinguistic theories provide possible clues as to the sources of the differences. Evidence of differences in the degree of pitch dynamism present in the intonation of the Taranaki and Wellington speakers supports claims about regional variation in New Zealand English intonation, but it does not in itself prove the existence of a uniquely Taranaki or a uniquely Wellington way of speaking English.