AbstractThis paper aims to explore the sound change in the production of velar nasals /ŋ/ in Taiwan Mandarin from a sociolinguistic perspective. Some studies have suggested that velar nasal /ŋ/ is undergoing a process of ‘alveolarization’; that is, the occurrences of non-prestigious [n] variant are on the increase. In addition, the phonological variation of non-prestigious [n] and prestigious [ŋ] is correlated with external social factors. However, there are no agreements on what kinds of social factors or on to what extent the social factors correlated with the variation.
In this study, one hundred and twenty-one informants, selected from four districts and social classes in Kaohsiung City, participated in the research. Four contextual styles were used to gather the linguistic date: oral interview, a short passage reading, sentence reading, and a wordlist reading. All interviews were recorded and the statistical analysis was tested to examine the interaction between velar nasal (ŋ) and five social variables—gender, age, social class, ethnicity and context.
The results indicate that (1) gender was not a determinant factor in the velar nasal /ŋ/ production. It was suggested that male speakers favor non-prestigious [n] variant. However, senior female subjects and lower middle-class female subjects also produced a lot of [n]. Two main possible explanations are given: some senior women received less education and are in a situation that prestigious forms [ŋ], are less needed; the gender of the researcher is also a key element. It was found that being interviewed by a female interviewer, lower middle-class women feel more relaxed and more non-prestigious [n] were produced. (2) Age was found to correlate with the informants’ linguistic performance. Compare to the younger people (16~30) and older people (above 51), those who aged between 31 to 50 produced the most prestigious forms [ŋ]. (3) A marked correlation was found between social class and linguistic behavior. The lower the class, the more non-prestigious forms [n] are produced. In addition, evidence of hypercorrection was found in the performance of upper working class and lower middle class subjects. The position of these two classes on the borderline between the working class and middle class causes them to be highly sensitive to the social pressure from higher social classes and thus hyper-correct in their use of standard speech forms. (4) People adjust their speeches in response to the formality of social situations. Informants, regardless of gender, age and social classes, produced the prestige forms of the variable most frequently in the most formal speech context—wordlist reading. (5) Finally, in the use of the (ng) variable, it was found that ethnicity did not play an important role. There is no statistical difference in the velar nasal production speeches of the Hakka, Minnan Ren, and mainlanders.
|Date of Award||2006|
|Supervisor||Mei-Hui Tsai (Supervisor)|