Felicity and Performative Efficacy — Epoche C1
If a bystander at a launching ceremony seizes the bottle, smashes it against the bows and shouts "I name this ship the Mr Stalin", the ship is not named — and the reason is not that he got the words wrong. He said exactly the right words. The example is J. L. Austin's own, from the lectures he delivered at Harvard in 1955 and published as How to Do Things with Words (1962), and it isolates the condition this essay is about. Austin called it appropriateness: the requirement that the particular persons and circumstances be right for the procedure being invoked. The claim to be defended is that this condition does not behave like the others on Austin's list, that it is claimed by an utterance rather than merely presupposed by it, and that this is visible with unusual clarity in a language whose grammar obliges every speaker to encode his standing relative to his hearer in the shape of the verb. Austin's list, and the two quite different ways it can fail This section sets out the conditions and one distinction the original version of this essay elided, because the argument later depends on it. A performative utterance, in Austin's early terminology, is one whose saying is the doing of something — naming, betting, bequeathing, marrying, promising — as opposed to a report of something. Austin's felicity conditions fall into three groups: A.1 There must exist an accepted conventional procedure, having a certain conventional effect, which includes the uttering of certain words by certain persons in certain circumstances; and A.2 the particular persons and circumstances in the case at hand must be appropriate for invoking that procedure. B.1 The procedure must be executed by all participants correctly, and B.2 completely. Γ.1 Where the procedure is designed for persons having certain thoughts, feelings or intentions, the participants must actually have them; and Γ.2 must conduct themselves accordingly afterwards. The two groups fail in different ways, and Austin gave them different names. Breaches of A or B produce misfires : the purported act is void, nothing has been done, no ship is named. Breaches of the Γ conditions produce abuses : the act comes off, but hollowly. This is where the earlier version of this essay went wrong in saying that if the conditions are not met "the speech act simply misfires, failing to achieve its intended effect or normative bindingness". A father who promises his child something he has no intention of delivering has violated Γ.1, and he has nonetheless promised. That is precisely why he is blameworthy: the obligation was incurred. If insincerity voided promises, the insincere promiser would be off the hook, which inverts the moral facts. Sincerity is a condition on promising well, not on promising. Appropriateness sits in the other group, and it really does void. But notice already that it is an odd member of that group. B.1 and B.2 concern what happens in the performance — whether the words were completed, whether the other party said "I do" in turn. A.2 concerns something outside the performance altogether: who the speaker is. Searle's refinement, and where the oddity shows John Searle's Speech Acts (1969) recasts Austin's conditions for the particular case of promising, and doing so makes the oddity of appropriateness sharper by contrast with conditions that are demonstrably about the content of what is said. Searle's analysis has four parts. The propositional content condition: in uttering the sentence the speaker predicates a future act of himself — which is why "I promise that it rained yesterday" is not a defective promise but not a promise at all. The preparatory conditions: the hearer would prefer the act to be done, and it is not obvious to both parties that the speaker would do it anyway in the normal course of events — which is why "I promise to keep breathing" is peculiar. The sincerity condition: the speaker intends to do it. And the essential condition: the utterance counts as the undertaking of an obligation. That last formulation points to the general form Searle later gives for institutional facts in The Construction of Social Reality (1995): a constitutive rule of the shape "$X$ counts as $Y$ in context $C$", where the $Y$ term assigns a status that the $X$ term does not carry by its physical nature. Marks on paper count as money; a man in a particular chair counts as a judge. Searle's crucial addition is that a status function of this kind exists only for as long as it is collectively accepted. The judge's authority is not a property of the man, and it is not a brute fact about the room; it is a standing assignment that a community keeps in force. Once that is stated, the question the essay's title poses becomes precise rather than merely suggestive. Is the appropriateness condition a fact about the world that an utterance must match, or is it itself something that utterances participate in maintaining? The precondition reading, and the political work it does The standard answer is the first, and it is not a naive position: it does real analytical work, which is why the alternative has to earn its place. Rae Langton's 'Speech Acts and Unspeakable Acts' (1993) is the clearest demonstration. Her argument, stated accurately, is that certain speech can silence in a specifically Austinian way. A speaker may be free to utter any words she likes and still be unable to perform the act those words would ordinarily perform, because a felicity condition fails for her in particular — most importantly the condition of uptake, that the hearer recognise the act for what it is. Langton calls this illocutionary disablement : a woman's refusal fails to count as refusal, not because she is prevented from speaking, but because the conventions of recognition have been arranged so that her refusal is not taken up. The point of framing this in Austin's terms is that it identifies a harm to liberty rather than merely to feelings, and it does so