The electrons have not even dried on the PDF of my dissertation — I submitted it and handed in the final lump of paperwork a few hours ago — but you can download it here. Here’s the abstract:
This dissertation examines the strategic deployment of evidential resources in communicative interactions among Nantis, an Arawak people of Peruvian Amazonia. In particular, this work focuses on Nantis’ uses of evidentials to modulate representations of responsibility, and shows that two distinct types of responsibility must be distinguished in order to account for the socially instrumental properties of evidential resources: event responsibility and utterance responsibility. Event responsibility concerns praiseworthiness or blameworthiness for happenings in which the relevant individual is causally implicated; while utterance responsibility concerns the socially salient attributes of an utterance (e.g. truthfulness), and not the utterance’s consequences. Evidential resources are shown to mitigate event responsibility in Nanti interactions by serving as a pragmatic metaphor, whereby the sensory directness or indirectness encoded by evidentials yields inferences regarding individuals’ participation in, and responsibility for, events. The use of evidential resources, principally quotative resources, to modulate utterance responsibility operates on quite different principles. Specifically, quotative resources serve to individuate utterances by attributing them to a particular source, thereby rendering explicit that individual’s commitment to the stances expressed by the quoted utterance. In doing so, the use of the quotative resource emphasizes that individual’s responsibility for the expressed stance. Quotative resources are also employed to decrease a first party’s responsibility for a stance, by attributing it to a third party. In this case, inferences based on the Maxim of Quantity lead interactants to infer reduced commitment on the part of the first party on the basis of the attribution of strong commitment to a third party. Both epistemic stance and a variety of moral and evaluative stances are relevant to utterance responsibility. Significantly, utterance responsibility is one of the few areas in which a pragmatic tie exists between evidentiality and epistemic modality, indicating the relative marginality of epistemic modality to evidentiality in Nanti, even at the level of pragmatics. An ethnographic and historical sketch of the Nanti people is provided, and a grammatical description of the Nanti language is also included.