No abstract

a b st r a c tThis paper studies two types of cognitive factors which have been assumed to underpin people's interpretation of conditional promises and threats: logic and socio-cognitive assumptions about what conditional promisors and threateners are obliged and permitted to do. We consider whether the logic of conditionals is compatible with the socio-cognitive assumptions underlying their interpretation or whether the two come apart. From the classical logical accounts of conditionals, almost all modern theories have inherited a constraint which specifies that a conditional cannot be true if its antecedent is true and consequent false. This logical constraint is widely assumed to constitute, at least partially, a conditional's semantics, or 'core meaning'. A replication of Beller et al. 's (2005) study, reported in this paper, calls for revisiting this longstanding, cross-theoretically assumed constraint. As predicted, we have found that, in English, conditional promises are generally consistent with this logical constraint, but threats are not. Our findings provide evidence for the existence of a new usage-based category of conditional threats, and support the claim that the observed logical asymmetry in the interpretation of conditional promises versus threats is just an[*] We are extremely grateful to Sieghard Beller for providing us with the German versions of the 2005 questionnaires. We would like to thank Linda Walz for translation and Stephanie Foxton, Arianna Nargis Hussain, and Tilly Flint, our student research assistants, for their help with the data collection. Our thanks also go to Andrew Merrison for discussions about the questionnaires and to York St John University for sponsoring various stages of this project with funding from the Student Researchers Scheme and the Ad Hoc Fund. Finally, we are grateful to Language and Cognition's reviewers and editors for extremely useful comments, which have helped improve this paper.terms of use, available at https://www.cambridge.org/core/terms. https://doi.

There is a growing body of research which undermines the traditional dualprocessing model of utterance interpretation, whereby pragmatic inference is preceded by the context-independent process of linguistic decoding. This body of research suggests that utterance interpretation is a wholly pragmatic inferential process. In this paper, I seek to defend a wholly pragmatic approach by investigating the role of the purported process of context-independent decoding and the process of pragmatic inference in determining when a conditional is false. I show that material conditionality, like all kinds of conditionality, lies in pragmatically derived holistic thought, i.e. not in any putative linguistic semantics.

Conditionals and modals work in tandem in some instances of practical reasoning, or decision making. Consider the following example (from Kratzer 2012): a. I want to become a mayor. b. (q) I will become a mayor only if (p) I go to the pub. c. Therefore, I should go to the pub. Given what the cogniser wants (a) and the relevant circumstances (b), the conclusion that the cogniser goes to the pub comes out as necessary. Hence, the presence of the necessity modal should in (c). Indeed, given the context of (a), the necessity modal in (c) is simply a reflection of the necessity of p for q, which is overtly represented by the use of the 'only if p, q' construction. This chapter looks into whether indirect reports of conditionals-in particular, indirect reports which involve the use of a modal verb-are sensitive to the necessity of p for q in cases where necessity is not overtly represented in a conditional, as in 'if p, q' formulations. We report on two online experiments into the relation between (i) perceived necessity or sufficiency of the truth of a conditional antecedent for the truth of the consequent, and (ii) the formulation of an indirect report of a conditional with necessity or possibility modals (have to, should, could). In Experiment 1, the 'necessity/sufficiency of p for q' variable was manipulated by contextually altering the number of alternative antecedents (e.g. Cummins et al. 1991; Thompson 1994; Politzer 2003). It was found that modals used in indirect reports of 'if p, q' conditionals co-vary with the number of alternative antecedents in predictable ways. This suggests that modals used in indirect reports of 'if p, q' conditionals may be a diagnostic for biconditional versus material interpretations of conditionals. The aim of Experiment 2 was to find out whether the results of Experiment 1 could be replicated in contexts which lower/eliminate the believability of the conditionals. It was found that manipulating the believability variable has no reliable effect on the results.

This paper investigates the relation between conditionals and modals in Polish. In particular, it looks into whether modalized indirect reports of conditionals – specifically, reports which involve the use of root modality verbs – reflect the number of alternative antecedents foregrounded in the context of the reported conditional utterance. It finds that the root possibility modal mogę is preferred when many alternative antecedents are foregrounded (i.e. when p is interpreted as a sufficient but not necessary condition for q), whereas the root necessity modal muszę is preferred in contexts where there are no alternative antecedents (i.e. when p is interpreted as a necessary condition for q). The results are comparable to those obtained for English (see Sztencel and Duffy 2018), thus providing initial support for their cross-linguistic generalizability. Furthermore, a parallel is drawn between the basic structure of instrumental practical reasoning (e.g. Walton 2007) and modalized indirect reports of conditionals. This parallel offers a rationale for the method of investigating instrumental practical reasoning by the proxy of indirect speech reports.