Presented as part of the ANU Linguistics Seminar Series

We begin by describing a sub-class of non-factive sentences containing factive verbs. In particular, we focus on sentences such as ‘Rob has not realized that Obama is in jail’ which we call ‘quasi-factives’. To explain the jargon, the verb ‘realize’ is “factive” (in our sense of that term), in that, given the right linguistic context, the verb’s complement is analytically entailed. If Rob has genuinely realized that p, then p must be true. (Similarly for ‘know’, ‘regret’, ‘inform’, ‘recognize’, etc.) But, as some reflection will show, the above sentence is “non-factive” (in our sense of that term), in that the linguistic context – here, negation via the word ‘not’ – eliminates the entailment. What’s peculiar about the sub-class of non-factive sentences which interests us is that they are, as it were, heard as factive – and hence lend themselves to “verbal sleight of hand”. For instance, ‘Rob has not realized that Obama is in jail’ would typically be heard as entailing, as a matter of meaning, that Obama is in jail, and only careful logico-semantic scrutiny shows otherwise.

We then offer a positive account of how this surprising effect is achieved. Logically speaking, the sentences at issue analytically entail a pair of negative disjuncts – one about the relation (e.g., Rob failing-to-realize) and one about the propositional relatum (e.g., that Obama isn’t in jail). Psychologically, the latter disjunct would violate a requirement of cognitive efficiency, and hence the former is seized upon. But the former entails the truth of the embedded proposition: if Rob fails-to-realize p, then p must be true. In the right contextual circumstances, then, the complement may be heard as “said” and even as “main point” – though as a matter of linguistic convention it isn’t in fact entailed, nor is it even “weakly logico-semantically supported”.

We end by considering an important objection, namely that our approach is overly narrow. Specifically, there seem to be closely related phenomena which are not best explained in this way. For instance, an out-of-the blue use of ‘Axel hopes to return to space’ might seem, to the incautious, to entail that Axel has been to space; and the sentence could be used with “verbal sleight of hand” to suggest just this. However, ‘Axel hopes to return to space’ is not a quasi-factive (in our sense of that term) because ‘hope’ is by no means a factive verb.