|dc.description.abstract||Although research in bilingual studies has widely investigated language alternation in various settings, very few studies probe into bilingual conversation in specific institutional contexts, in which the power relations of speakers’ institutional roles are relatively discernible. In view of this,
this dissertation undertakes the research targeting police-public interactions in police check-up encounters in the An-Nan District of southern Taiwan, in which most speakers are Mandarin-Minnanyu bilinguals. While my previous project (Lee, 2011) indicates a tendency for officers to
accommodate people’s language preferences, the present research begins with the hypothesis of “power as medium negotiation,” attempting to analyse the new data of eleven recordings conducted in the same research site. In order to verify the hypothesis, analysed instances are categorised into
three parts: (1) speakers’ negotiation toward a monolingual medium (2) speakers’ negotiation toward a bilingual medium (3) transitional places of medium switching.
By extensively applying Gafaranga’s (1999, 2000, 2007a, 2007b) notion of medium and
making appropriate use of Auer’s (1984, 1988, 1995) sequential approach to look at the local environments of the speech turns for how medium negotiation sequences unfold, it has been found that there are no consistent results to support the hypothesis. By contrast, it is observed that both the officers and the public negotiate for their preferred codes throughout conversations. Furthermore, officers and the public propose their language preferences with different considerations, such as
one’s “competence-related preference (Gafaranga, 2001:1916),” “the participant-related factor (Gafaranga & Torras, 2002a:7),” and “the medium switching proposal as a contextualisation cue for
a topic shift” (Torras, 1998; Gumperz, 1982). For this study investigating the inter-relevance between institutional power and medium negotiation, the current results suggest that the institutional parameter of officers’ power does not permeate to the level of bilingual speaker’s medium negotiation; additionally, it is hoped that this study will indicate a researchable facet for further investigation studying bilingual conversations in other institutions.||en