ProjectSocially-relevant pragmatic inference

Basic data

Socially-relevant pragmatic inference
01/04/2024 to 31/03/2027
Abstract / short description:
In order to understand whether the sentence The results of the election were interesting expresses a positive or a negative evaluation of the election, and exactly what in the outcome of the elections the speaker found interesting, one would need to know the relevant background beliefs of the speaker and possibly her political affiliations. The speaker might mean that the results were not transparent or that the candidate who lost the election did not accept the defeat gracefully, to name a few options. Understanding what the speaker meant also involves reasoning about what else she possibly could have said. If she was dissatisfied with the counting procedure or the behavior of one of the candidates, she might have used more direct statements, such as The counting procedure was flawed or The candidate's behavior was not acceptable. We could then reason why the speaker did not choose a more straightforward expression, possibly contemplating two hypotheses: the speaker either did not think that a more direct statement holds, or she stays indirect to avoid a potential conflict. It is the interaction of rational pragmatic reasoning and background beliefs of conversation partners that is the focus of this project. The goals of the project are fourfold. First, it seeks to understand how speakers pursue multiple communication goals at the same time. Second, the project questions how background beliefs about what is predictable affect the way speakers choose to encode their utterances. Third, the project considers an additional step in communication: what the utterance interpretation reveals about the background beliefs of the listener. The inferences concerning background beliefs carry social relevance: they allow conversation partners to learn about each other in situations when direct questioning is not appropriate or possible. Finally, the project investigates the cognitive effects of belief accommodation. One main result of the project will be the development of a computational probabilistic model that captures the integration of background beliefs in the process of choosing and interpreting utterances. Such a model is critical for building computational theories of human communication that span beyond simple information transfer.

Involved staff


Research training groups
University of Tübingen

Local organizational units

Institute of Linguistics (SfS)
Department of Modern Languages
Faculty of Humanities


Bonn, Nordrhein-Westfalen, Germany

will be deleted permanently. This cannot be undone.