Philosophy & Rhetoric

"How to do things with words?" How can you really do things with nothing but words? It seems to me that sophistics is in a way the paradigm of discourse that does things with words. Doubtless it is not a "performative" in Austin's sense of the word, although Austin's sense varies considerably in extension and intension. But it is for real a discourse that operates, that transforms or creates the world, and has what I call a "world effect."

Making the connection with performativity is all the more tempting, as epideixis , the word that serves in Plato to designate sophist discourse, cannot be rendered better than by "performance," on condition that "performance" is also understood in the sense of contemporary aesthetics as a "happening," an "event," an improvisation that requires engagement (Gorgias is the inventor of discourse ex tempore, according to Philostratus)—something like an "exploit."¹

"Performative" is Austin's own invention, acclimated to French by Austin himself at a colloquium held at Royaumont (Austin 1962); thereafter it was immediately adopted and popularized by Émile Benveniste (1966). "Performance" is a much older term, which, after ceaseless borrowings to and fro between English and French, has seen its meaning shifted and extended accordingly. Klein's Comprehensive Etymological Dictionary of English Language (Klein 1971) maintains that in English "performer" [End Page 349] was coined from Old French parfournir (from medieval Latin perfurnire) and/or parformer; in addition French borrowed the term at least three times, if the Dictionnaire culturel de la langue francaise (Rey 2005) is to be believed: in 1869, by analogy with the vocabulary of horse races to mean the "manner of developing a subject, of executing a work in public"; in 1953, to denote "individual result in the accomplishment of a task"; and in 1963, in the wake of Chomsky, to mean the opposite of "competence." In sum, the word is a fluid, bilingual term that bridges sport (performance in the sense of a record), technique (performance in the sense of the output of a machine), psychology (performance of a test), linguistics (performance/competence), and modern art (performance in the sense of happening).

Let us start with the relationship between performance and performative. It is a way to interrogate the status of rhetorics, for which Austin, without naming it, reserves a somewhat unstable place between the "locutionary" on the one hand and the "illocutionary" or performative on the other, calling it the "perlocutionary" ("per" precisely as in "performative").²

But it is not of Austin that I will speak. Austin is simply the contemporary frame of reference that informs us today: he "invented" the performative as such for us, by trying to isolate it. And he never hides the difficulty, the permeability, of his taxonomy. Just one citation is enough to show the difficulty. In the seventh lecture of the twelve that make up How to Do Things with Words (quite late then) he writes: "It is time [. . .] to make a fresh start on the problem. We want to reconsider more generally the senses in which to say something may be to do something, or in saying something we do something (and also perhaps to consider the different case in which by saying something we do something). Perhaps some clarification and definition may help us out of our tangle. For after all, 'doing something' is a very vague expression. When we issue any utterance whatsoever, are we not 'doing something'?" (1975, 91–92).

In the framework of the generalized theory of speech acts, the difference between the locutionary, the illocutionary, and the perlocutionary has for a long time been in a "tangle". It is not so easy to differentiate between the three. All three are, precisely, "acts" of language, and without doubt the categories are at once abstract, slippery, and overlapping. The "locutionary" or "constative," a "normal" statement, is an utterance that "says something"; it is an "act of saying something" (1975, 100): "The cat is on the mat" has a meaning (both a "sense" and a "reference") and is susceptible of being either true or false. For...