Vis sammendrag
This thesis presents a systematic experimental study of putative negative evidence in pragmatics. Because pragmatic meaning depends on the context, pragmatics is traditionally considered independent of the grammar. This post-semantic view of pragmatics entails that, unlike ungrammaticalities and infelicities that constitute negative evidence in syntax and semantics, pragmatic inferences are not expected to result in contradictions since they arise only when the context is such that they are meaningful. However, different cases of such putative negative evidence in pragmatics, i.e., inferences that lead to infelicities, have been discussed in the literature.
The existence of negative evidence in pragmatics would question a post-semantic view on pragmatics. A framework within which such negative evidence could be predicted would need to bridge semantics and pragmatics and allow for pragmatic meaning to be derived alongside grammar. The grammatical account of implicatures offers such a framework, also known as the exhaustivity-based framework.
While the exhaustivity-based theories successfully account for putative negative evidence in pragmatics, they have been questioned on several grounds. On the one hand, some such theories appear to be ad hoc. On the other, the automatization of the derivation of inferences means that the exhaustivity-based theories may lack flexibility in accounting for positive evidence, i.e., they may over-generate predictions of negative evidence.
The three articles of this thesis each concerns a different instance of putative negative evidence for which an exhaustivity-based theories have been proposed. These are, first, odd scalar sentences with the exhaustivity-based analyses proposed by Magri (2009), Del Pinal (2021) and Bassi, Del Pinal, Sauerland (2021), second, the limited distribution of distributive inferences with the exhaustivity-based analysis proposed by Bar-Lev & Fox (2023), and third, the obligatory ignorance of one of several disjunctive forms in Russian with the exhaustivity-based proposal by Ivlieva (2024). The aim of the thesis is to experimentally test the predictions of these theories, evaluate them against experimentally collected data, and provide alternative accounts of the negative evidence in question, when appropriate.
The main conclusion of this thesis is that, despite the power of the exhaustivity-based framework to predict negative evidence and to bridge semantics and pragmatics, its predictions do not always hold. At times, there are other pragmatic factors that need to be considered. At other times, the putative negative evidence in pragmatics is better accounted for as an instance of negative evidence in semantics. The thesis highlights the importance of testing hypotheses of putative negative evidence in pragmatics and exemplifies how such tests can be conducted using experimental methods.
