Evidentiality in Time and Space

When making a simple statement, speakers of some languages spoken in, e.g., the Himalayas, the Americas, and Papua New Guinea must choose one or another grammatical marker depending on how they obtained the information they want to convey. These languages encode what linguists came to widely refer as ‘evidentiality’ in the 1980s. Interestingly enough, some of the semantic and distributional properties of Tibetic evidentials, which figured prominently in the development of the concepts involved, appear to challenge widely accepted theories of evidentiality. At the same time, scholars working on Tibetic evidentials have not worked out the details of how these phenomena have arisen. 

In this project, we address these two issues by providing a diachronic-functional account of all evidentials observed in Tibetic and neighboring languages. We have already identified two opposite ways in which evidentiality developed in these languages. These two diachronic types differ in terms of their original grounding, the linguistic constructions they involve, the different contrasts that emerge, and the semantics that individual markers have. The project collects data from the two other main hotbeds of evidentiality of the Tibetic type, one in the New Guinea Highlands and the other in northwestern South America. The thorough synchronic grammatical description of individual languages is essential as a basis for diachronic-functional accounts, particularly of evidential systems, and one important goal of the project is to complement the extant synchronic accounts for these two areas. Our preliminary diachronic typology of evidentiality – based on languages of the Himalayas – will support us in accomplishing this goal, and the new data will in turn inform our typology, allowing us to extend and refine it, which is the second main goal of the project. 

Project leader: Fernando Zúñiga
Post-doctoral researcher: Marius Zemp
Former collaborators: Benjamin Brosig (postdoctoral researcher) and Sara Schindler (research assistant)