Title:	Pragmatic use & interpretation of conditionals
Authors:	Grusdt, Britta
ORCID of the author:	https://orcid.org/0000-0001-7938-3734
Thesis advisor:	Prof. Dr. Michael Franke
Thesis referee:	Prof. Dr. Mingya Liu Prof. Dr. Nicole Gotzner
Abstract:	‘If’-sentences can be used to express many different beliefs and intentions, including, among others, warnings, promises, causal relations, or the speaker’s uncertainty with respect to the propositions mentioned in the conditional, an expression of the form ‘if . . . , (then) . . . ’. Thus, speakers use conditionals in a wide range of situations and listeners easily interpret them accordingly. However, it remains an unsolved question how such diverse interpretations of the word ‘if’ arise. Across a number of different disciplines, people, therefore, aim to find answers to the supposedly simple question of what ‘if’ means. In this thesis, I investigate the pragmatic meaning of conditionals. In the first part of the thesis I present a probabilistic Bayesian model, more precisely a Rational-Speech-Act model, building on recent developments in computational pragmatics. We show that pragmatic processes formalized by the model, in combination with a representation of rich structural prior beliefs of the interlocutors, can explain common observations within the communication with conditionals. These include, among others, the usually inferred relation between the propositions mentioned in the conditional (‘If A, C’ suggests that A and C are related) as well as the infelicity of so-called missing-link conditionals which, as suggested by their name, lack this relation entirely. These are two observations that have been the focus of a current debate about whether, and if so, how, it is possible to explain them within pragmatics instead of ascribing them to the semantics of conditionals. In the second part of the thesis, I first present experimental data, collected in a behavioral experiment that we designed, on speakers’ use of conditionals. Then, I compare this empirical production data with the quantitative predictions of our model, which shows that the model is able to account for major aspects observed in the data. The third part of the thesis addresses a particular phenomenon called conditional perfection which refers to the interpretation of a conditional ‘if A, (then) C’ as biconditional ‘if and only if A, (then) C’. In this part, I focus on the empirical side again, presenting two behavioral experiments with which we aim to test an account proposed by von Fintel (2001) about the influence of the focus of a conversation, a so-called Question under discussion, on the listener’s interpretation of a conditional, in particular as biconditional.
URL:	https://doi.org/10.48693/419 https://osnadocs.ub.uni-osnabrueck.de/handle/ds-202310249948
Subject Keywords:	computational pragmatics; conditionals; Rational-Speech-Act model; experimental; causal reasoning
Issue Date:	24-Oct-2023
License name:	Attribution-NonCommercial-NoDerivs 3.0 Germany
License url:	http://creativecommons.org/licenses/by-nc-nd/3.0/de/
Type of publication:	Dissertation oder Habilitation [doctoralThesis]
Appears in Collections:	FB08 - E-Dissertationen

This item is licensed under a Creative Commons License