Abstract:
This thesis aims to describe how direct reported speech (DRS) is constructed in Turkish and investigate how narrators make use of different reporting strategies in order to position themselves, their audience and the speakers they report in order to construct relational identities. Our analysis is based on 6 hours of video and audio recordings composed of both life story interviews and naturally occurring conversations. The most frequent DRS strategies in our data are DRS with the quotative de-, DRS with zero quotative and DRS with the converbial diye. DRS with the quotative de- is categorized into DRS with single quotative and DRS with multiple quotatives. In the latter category, the irregular behavior of de- in terms of sentential distribution, argument structure and prosodic features challenges its status as a verb. As for the narrative functions of DRS, we argue that there are three main categories: event description, theatrical representation and character assertion. Event description is carried out by DRS with single quotative, embedded in the narrative as part of complicating action or evaluation. Theatrical representation function is fulfilled by DRS with zero quotative, which enables the reporter to play the part of the reported speaker. Character assertion function is achieved when the narrative itself is the reported conversation and the multiple quotatives work towards reinforcing the reported speaker's position in the narrative. We further show that narrators are well aware of the representation and positioning effects of different strategies and make conscious choices regarding the DRS strategy to be used, as the organization of experiences via narratives requires the organization of characters.
