Performatives are conceptually special in that they involve illocutionary acts that can be performed simply “by uttering a sentence containing an expression that names the type of speech act” (Searle 1989: 536), as in I (hereby) quit. A grammatical reflex of this in English is that performatives, unlike canonical present-time event reports, prefer the simple present to the present progressive (cf. *I talk right now versus I promise to quit). Assuming a perfective aspectual value for the English simple present, this is indeed remarkable: in English as well as cross-linguistically, present-time events cannot normally be reported by means of perfective constructions (De Wit forthcoming). However, these observations about English performatives cannot readily be extended to other languages: Slavic languages, for one, hardly ever use perfective aspect in performative contexts (Dickey 2000).

In this study we chart the aspectual characteristics of performative utterances in a typologically diverse sample of sixteen languages on the basis of native-speaker elicitations. We conclude that there is not one single aspectual type (e.g., perfectives) that is systematically reserved for performative contexts. Instead, the aspectual form of performative utterances in a given language is epistemically motivated, in the sense that the language will turn to that aspectual construction which it generally selects to refer to situations that are fully identifiable as an instance of a given situation type at the time of speaking. We use Croft & Poole's (2008) method of multidimensional scaling to demonstrate this: whatever the exact value of a given aspectual marker, if it is used to mark performatives, then it also commonly features in the expression of states and habits, which have the subinterval property (they can be fully verified based on a random segment), live sports broadcasts, and other special contexts featuring more or less predictable and therefore instantly identifiable events.