Attested versus unattested contrafactive belief verbs
Main Article Content
Abstract
While factive belief reports (x knows p) are said to presuppose p, Holton (2017) suggests that no language shows the opposite: there appear to be no CONTRAFACTIVE belief verbs presupposing not-p. Recent work seeks to explain why not. In this squib, I suggest that a contrafactive could be defined in two different ways: (i) it could require not-p (i.e., that all worlds in the Context Set are not-p worlds); or (ii) it could require that the Common Ground is compatible with not-p (i.e., that some worlds in the Context Set are not-p worlds). So far no verb of type (i) has been demonstrated, but those of type (ii) arguably do exist -- for example, Mandarin yıwei, analyzed by Glass (2023). Refining the typology of (un)attested belief verbs, the question of “Why are there no contrafactives?” becomes: “Why are there belief verbs requiring that p is or is not Common Ground, but none requiring that not-p is Common Ground?”
Article Details
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
Articles appearing in Semantics and Pragmatics are published under an author agreement with the Linguistic Society of America and are made available to readers under a Creative Commons Attribution License.