Abstract
In this paper we focus on the logicality of language, i.e. the idea that the language system contains a deductive device to exclude analytic constructions. Puzzling evidence for the logicality of language comes from acceptable contradictions and tautologies. The standard response in the literature involves assuming that the language system only accesses analyticities that are due to skeletons as opposed to standard logical forms. In this paper we submit evidence in support of alternative accounts of logicality, which reject the stipulation of a natural logic and assume instead the meaning modulation of nonlogical terms.
Highlights
The ‘logicality of language’ hypothesis refers to the idea that the language system contains a deductive device to exclude analyticities – that is, contradictions and tautologies (Gajewski 2002, 2009; Chierchia 2013, 2021; Del Pinal 2019, 2021; PistoiaReda and Sauerland 2021)
On the one hand, we have that the exceptive should make the quantified formula “some students smoke” false on student but true with respect to a relevant subset X of student, e.g. student {j }; yet, on the other hand, we know that the existential quantifier is an upward monotone quantifier, which implies that if the quantified formula is true over X, it should be true over all supersets of X, including student
As discussed by Del Pinal (2019, pp. 2124), there are versions of the skeleton proposal (e.g. Chierchia’s influential account of negative polarity constructions) that appeal to classical principles such as the law of non-contradiction to account for the triviality of ungrammatical structures such as (3), which makes one wonder whether the idea of a linguistic logic is tenable for those cases
Summary
The ‘logicality of language’ hypothesis refers to the idea that the language system contains a deductive device to exclude analyticities – that is, contradictions and tautologies (Gajewski 2002, 2009; Chierchia 2013, 2021; Del Pinal 2019, 2021; PistoiaReda and Sauerland 2021). Structures like (1), (2), and (3) are said to be excluded from the language on account of their obligatory association with logically trivial contents. The contradiction emerges from a conflict between the exceptive (“but”) and the existential quantifier (“some”). Let us denote by student and smoke, respectively, the collection of all students and all smokers, and by the constant j the individual John. According to the analysis we have a contradiction, and (4b) provides the logical reason why the language system excludes (1), i.e. why (1) is not an acceptable sentence. In the context of universal quantification, exceptive constructions are associated with clearly informative contents, as is illustrated in (5b) and in (6b).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
