Performative uses and the temporal interpretation of modals

Abstract

Expressions with a modal semantics vary with respect to whether they are suited to ‘performative uses’. Modals like must and have to can easily be used to give a command and hence create an obligation, but the same does not seem to be true for predicates like be obligated to and be under an obligation to. This fact poses a challenge for an otherwise attractive class of analyses that take the ‘performative effect’ (i.e., the creation of an obligation) to arise pragmatically from a claim made with the usual, descriptive modal semantics. Seeing as the different modals are typically assigned the same truth-conditional content, a pragmatic account predicts that there should not be a difference in the availability of performative uses. In this paper, I will explore possible avenues for meeting this challenge while preserving the attractive features of a pragmatic account. My starting point will be the observation that certain commonly-made assumptions about temporal interpretation in fact block a pragmatic derivation of the performative effect. Then I will consider how this conclusion can be avoided for modals that have performative uses. We either have to assume that these modals have a more liberal temporal interpretation or that the performative effect arises in a different manner than assumed by existing accounts. Quite independently from the issue of anti-performativity, this paper demonstrates that two seemingly independent phenomena—temporal interpretation and performative uses of modals—are in fact intertwined, hence we can shed light on one by studying the other.