two birds
Photo: Pixabay

A dialogue-based theory of evidentiality

Research project
Active research
Project period
2021 - 2024
Project owner
Department of Philosophy, Linguistics and Theory of Science

Swedish Research Council (Vetenskapsrådet)

Short description

How do we claim knowledge in everyday talk? This project investigates the linguistic category of evidentiality and how it is used in spoken language. The project focuses on data from indigenous languages of the Americas.

Evidentiality signals the speaker's claim of knowledge and conveys their source for such knowledge. This project aims to formulate a theory of evidentiality based on how the category is used in spoken interaction. The project focuses on endangered minority languages from South America, in which evidentiality is expressed by grammatical markers. The data used in the project comes from corpora of relevant languages, deposited in online archives dedicated to safeguarding endangered language materials. The project is innovative in terms of its research goals (defining a linguistic category based entirely on spoken language) and methodology (research on existing minority language corpora).

Project description

This project advocates a dialogue-based approach to defining evidentiality, which signals how speakers claim knowledge of events (through seeing/hearing/reasoning/report). In European languages, such as English and Spanish, evidentiality is not part of grammar and as a result, evidentiality has long been marginalized as an ‘exotic’ category. However, as new evidential systems are being documented, a definition of this category must rest on theoretical generalizations that reflects the richness of attested evidential systems and the use of individual forms.

The goal of the project is to propose a coherent, interaction-based theory of evidentiality: one that makes adequate predictions about the systems we encounter, rather than confining us to make neatly-structured, ad hoc taxonomies. To this end, we will use spoken-language corpora (collected by ourselves and ones deposited in language archives) to capture generalizable characteristics of evidential markers and provide analytical support for hypotheses regarding a cross-linguistically viable definition of evidentiality.