| This dissertation concerns the best approach for examining usage of what Japanese descriptive grammars call ‘feminine’, ‘neutral’, and ‘masculine’ sentence-final forms. My dissertation shows, based on Holmes' (1984) and Schegloff's (1993) discussions, why a detailed discourse analysis is a prerequisite to understanding the precise conditions for actual usage of the sentence-final forms. Holmes (1984) is showing that linguistic choices need to be understood within discourse contexts.;Previous studies claimed that a speaker's sex (Ide 1979), age (Okamoto 1995), and/or occupation affect(s) usage of the sentence-final forms. However, little is known about discourse conditions for occurrence of the forms because previous studies were mainly quantitative and did not microanalytically study discourse contexts.;I recorded a total of ten hours of Japanese university orchestra meetings involving eighteen women and twelve men. Through a sequential analysis of their uses of 139 so-called ‘neutral’ nonparticle questions, I refute the common assumptions of previous quantitative approaches that overlook close discourse contexts: (1) all speakers have equal opportunity to use the form in question, and hence, (2) if the frequency of the form differs between women and men, the speakers' sex is responsible.;The dissertation demonstrates that when women and men use nonparticle questions at different frequencies, the sex of speakers is not directly responsible. Rather, individual speakers' engagement in activities specific to particular discourses affects the frequency of usage. One speaker who takes notes after the questions is more likely to use nonparticle questions. Another speaker, the sole announcer of practice schedules immediately before the questions, is less likely to ask them. What often matters to the frequencies of usage are the discourse activities before, after, or when the question occurs. Individual participants do not have equal chances to use nonparticle questions (Schegloff 1993) because of what they are doing when talking.;To consider such connections between linguistic choices and discourse activities (Freed 1996), we must closely examine the context of specific occurrences of the choice. This dissertation demonstrates that a close discourse analysis provides data that reorient our thinking about the influence of sociolinguistic variables on speakers' linguistic choices. |
