DescriptionAccording to a familiar and appealingly simple picture of assertion, the propositional content asserted by a speaker is both that which she communicates to members of her audience and that to which she undertakes a distinctive sort of commitment. In what follows, I develop a criticism of this familiar picture and propose an alternative account according to which assertoric content is more intimately associated with commitment than communication.
I begin by motivating a common claim about rational communication: that it cannot proceed when interlocutors are uncertain which contents utterances contribute to discourse. It emerges during my discussion of this claim that, given certain natural assumptions about the conditions under which it is rational for interlocutors to assert, it is possible to construct an argument from the premise that the propositional content asserted by a speaker is the information she communicates to members of her audience to the conclusion that speakers always assert diagonal propositions of their utterances — that is, propositions which characterize the information interlocutors can learn from their utterances by assuming that the propositions they semantically determine are true.
I proceed to argue that this latter claim, to which I refer as Diagonalism, systematically conflicts with our intuitive judgments about the conditions under which the contents of speakers' assertions would be true or false. It follows that the failure of Diagonalism requires us to abandon the claim that the propositional content asserted by a speaker is the information she communicates to members of her audience.
I then argue that, even if assertoric content does not play the role in communication it has often been thought to play, the familiar picture of assertion is at least half right: the propositional content asserted by a speaker is indeed that to which she undertakes a distinctive sort of commitment. I motivate this claim in two ways. First, I propose a diagnostic for identifying the propositional content to which a speaker becomes primarily committed in asserting and show that this diagnostic singles out assertoric content even in cases where what is asserted is not what is communicated. In the course of this argument, it emerges that the content of an assertion is usually its horizontal proposition — that is, the proposition it semantically determines in the context in which it is in fact uttered. Second, then, I argue that a linguistic community which adopts a practice of regarding its speakers as assertorically committed to the horizontal propositions of their utterances enjoys benefits not enjoyed by communities which adopt alternative committal practices, such as the practice of regarding speakers as assertorically committed to the diagonal propositions of their utterances.