Realis and irrealis: Chini verb morphology, clause chaining, and discourse

At this point in time, realis/irrealis distinctions have been described for many languages, particularly of North America and New Guinea. There remains considerable debate, however, about the language-specific functions of these categories and to what extent they are comparable cross-linguistically. From a descriptive point of view, they are a challenge in part due to the ways in which they do not quite fit better understood areas of typology, in particular tense and its semantic basis in temporal reference. Another part of the analytical challenge lies in the contextual uses and interpretations of realis and irrealis constructions, which crosscut temporal reference, aspect, modality, and polarity, epistemic concepts involving presupposability and expectability, illocutionary and perlocutionary force. In this dissertation, I seek to contribute to what we know about this little-understood and often misunderstood area of grammar by discussing the workings of realis/irrealis distinctions in a language where they are unusually highly elaborated. Chini, the traditional language of the Awakŋi people of Andamang village and the Yavɨnaŋri of Akrukay in Madang Province, Papua New Guinea, is well-positioned in that respect, because distinctions recognizable in terms of realis and irrealis meaning are marked in all of three unrelated sets of forms: twice in the inflectional morphology of the Chini verb and then again in the forms of the linking enclitics that attach to dependent ('medial') clauses in clause chains. One set of inflectional markers occur in multiple parts of the verb morphology, where the distinction has taken on certain specialized functions as a result of grammaticalization and other historical processes. In order to understand the workings of realis and irrealis constructions in Chini grammar and discourse, I draw on nine months of documentary fieldwork and a corpus of 10.5 hours of annotated connected speech and additional fieldnotes. This multiplicity of realis/irrealis distinctions encoded in the language allows us to see that the functions of the markers are at once language-specific but also dependent on the area of the grammar where the marking occurs. In Chini, this can be seen in the divergent functions of realis and irrealis marking in the verb morphology versus in the chain linkage devices. The inflectional distinction has a semantic basis I describe as 'within (realis) versus beyond (irrealis) experience'. The distinction in clause chaining is pragmatically based. Roughly put, the linkers signal events in medial clauses as being 'within (realis) or beyond (irrealis) the expected or expectable course of reality'. Irrespective of where in the grammar the distinction occurs, the underlying concepts involve a general notion of reality, one that does not refer to a universal conceptual space, philosophical notion, or set of notional semantic features. It refers to the real, culturally-specific world of lived experience. In order to contextualize the discussion on realis/irrealis distinctions, also embedded in the structure of the dissertation are descriptions of the ethnographic and historical setting, as well as descriptions of the structures involved in Chini verbal morphology and clause chaining. The appendix includes three texts: a narrative and two excerpts from Chini conversation