of the words of which it seems to be composed. The nonlinguistic props and surroundings of a linguistic expression—this person speaking in the presence of that other, at this time and in that place, and so on—are what really allow language users to do things with words.
Many actions can be carried out with words without using any of the verbs that allegedly “perform” the action. I can promise to marry someone by saying “Sure I will” in response to a plea, and that’s just as binding as saying “I promise.” I can warn somebody with an imperative—“Stay away from the cliff!”—just as I can threaten someone by asking them to step outside in a particular tone of voice. The force of an utterance is not related solely to the meanings of the words used in the utterance. In many instances, it is hard to show on linguistic evidence alone that they are related at all.
Intentional alteration of one or more of the basic contextual features of an utterance usually turns a meaningful expression into some kind of nonsense. But the reverse can also be achieved: nonsense can be made to make sense by supposing some alternative context for it. At the start of his revolutionary work Syntactic Structures (1957), Noam Chomsky cooked up a nonsense sentence in order to explain what he saw as the fundamental difference between a meaningful sentence and a grammatical one. “Colorless green ideas sleep furiously” was proposed as a fully grammatical sentence that had no possible meaning at all. Within a few months, witty students devised ways of proving Chomsky wrong, and at Stanford they were soon running competitions for texts in which “Colorless green ideas sleep furiously” would be not just a grammatical sentence but a meaningful expression as well.
Here’s one of the prizewinning entries:
It can only be the thought of verdure to come, which prompts us in the autumn to buy these dormant white lumps of vegetable matter covered by a brown papery skin, and lovingly to plant them and care for them. It is a marvel to me that under this cover they are labouring unseen at such a rate within to give us the sudden awesome beauty of spring flowering bulbs. While winter reigns the earth reposes but these colourless green ideas sleep furiously.1
Nowadays the expression “colorless green ideas” could perhaps refer to the topics of negotiation at the Copenhagen Climate Summit of December 2009; to say that they “slept furiously” may be no more than to name the paltry outcome of the conference. The point of this is not just to say that people play with language and often make mincemeat of authoritative generalizations about it. It is this: no grammatical sentence in any language can be constructed such that it can never have a context of utterance in which it is meaningful. That also means that everything that can be said or written—even nonsense—can (at some time or another) be translated. Verdi idee senza colore dormono furiosamente.
To translate utterances that perform a conventional action by the fact of being uttered—greeting, ordering, commanding, and so on—requires the target language to possess parallel conventions about things you can do with words. But there are significant differences between cultures and languages in how people do things with words. A promise may be a promise the world over, but the conditions of felicity, as well as the forms of language that are appropriate to the making of a promise, may vary greatly between, for example, Japan and the United States. It’s not the linguistic meaning of “I promise, cross my heart and hope to die” that needs to be translated if the aim is to make a similar commitment in the target language. Once again, the expression uttered (in speech or writing) is not the sole or even the primary object of translation when the force of an utterance is what matters, as it always does.
These considerations don’t affect just the set of verbs that Austin called performatives. The range of things you can do with words goes far beyond the promising, warning, knighting, naming, and so on that attracted the philosopher’s attention, and it would be better to see those not-so-special verbs of English as only one way of grasping a more general aspect of language use. When I say “How are you?” to an acquaintance I run across, I am performing the social convention of greeting with an utterance that is conventionally attached to it. Whether I use a performative verb