Facts, Promising and Obligation

John Searle attempts to show through a consideration of promising that at least some ‘ought’ statements can be derived from ‘is’ statements ['How to Derive “Ought” from “Is”’, Philosophical Review, 73, 1964, pp. 48-58]. He thinks (1) that you can determine on purely factual grounds that a person has made a promise, and (2) that it follows logically from the statement that a person has made a promise that he has at least a prima facie obligation to do the thing he promised to do. I agree with (2) but not with (1) [I follow Antony Flew in this, but my strategy is in other respects different (A. G. N. Flew, ‘On Not Deriving “Ought” from “Is”’, Analysis, 25, 1964)].

The statement that a person has made a promise can be analyzed as containing two separable parts, one of which gives the facts of the matter— and I mean all of them—while the other makes an evaluative judgment. Searle would disagree. He would argue that you cannot give an account of all the facts without making an evaluation. And he would perhaps add that philosophers have only thought that they could because they failed to recognize that ‘promise’ is a performative, and that promising is making a performance that can only be described in full if one grants implicitly that the promisor has acquired a prima facie obligation. Searle would suppose that a proper account of a particular promise must (1) acknowledge the existence of an institution which enables people to acquire obligations by saying certain words such as ‘I promise . . .’ in the appropriate circumstances (e.g. the speaker must not be play-acting, he must understand the language, he must be conscious, etc.) and (2) state that someone has said the required words in appropriate circumstances—that is, that he has followed the directions. I would suppose rather that to cover the facts, all you need to acknowledge is that there is an institution designed to enable people to acquire obligations by saying certain words in certain circumstances and that someone has indeed followed these directions. To give the facts, you do not need to say that the institution actually enables people to acquire obligations. The facts only require you to say that it is designed to enable them to do so. If you go on and add that it succeeds—that it does enable them to acquire obligations, you are adding an evaluation. You could, of course, say that the institution succeeds in another respect—that it enables people to acquire the belief that they have obligations and enables them to lead others to hold the same belief, but none of this implies that they actually have an obligation.

The claim that I have made should not be confused with another, namely, that a person is free to decide that the institution of promising is not a good thing. This is true but, as Searle sees, perfectly compatible with his thesis. Searle can admit that we are free to decide that the institution of promising is bad but hold that, even so, we are forced to admit that the institution does enable people to acquire obligations.

My criticism of Searle is similar in certain respects to Flew’s, but whereas Flew cannot cope with Searle’s rejoinder (Speech Acts, pp. 188-198), I believe that I can. Flew argues that given the fact that a person has said the prescribed words in the prescribed circumstances, we are only forced to admit that he has made a promise if we construe the statement that he has done so as oratio obliqua, but that the statement that he has made a promise only implies that he has an obligation if we construe it as oratio recta, and therefore that Searle’s argument has failed. Searle replies that even if we are speaking oratio recta, we can be forced by the facts to acknowledge that a person has made a promise. Fie claims (Speech Acts, p. 187) that although the statement that a person has made a promise is evaluative, it is also ‘purely descriptive’ [W. D. Hudson (‘The “is-ought” Controversy’, Analysis, 25, 1965) provides an alternative reply. He argues that the statement that a person has made a promise (oratio recta) is not evaluative because we can be forced by the facts to accept it, but that it does imply an evaluative statement, namely, that the person has a prima facie obligation to keep the promise. His response, like Searle’s, would fail if it can be shown that we cannot be forced by the facts to acknowledge that a promise has been made].

Is Searle right? Can the facts really force us to acknowledge that a person has made a promise, and therefore has a prima facie obligation? On first glance, it would seem that they can. If a person has said that he promises to do a certain thing and the circumstances are appropriate, you are forced at any rate not to deny that he has made a promise, and this is true even if you suppose that the institution of promising does not enable us to acquire obligations. But as I shall show, this does not block the case against Searle. If I am right, the statement that a person has made a promise can be broken down into two parts, a presupposition and an assertion:
  1. It is presupposed that an institution exists which enables people to acquire obligations by following certain directions.
  2. It is asserted that the person in question has followed the directions.
If I am right, then, to say that a person has made a promise is both to state something and to presuppose something rather than to state the conjunction of two things; therefore, to say that a person has not made a promise is not simply to state that at least one of a pair of conjuncts is false. Instead, it grants the presupposition with its implicit evaluation and denies that the person has followed the directions. But the question of whether he followed the directions is purely factual, so facts alone may make it incorrect to deny that a person has made a promise.

But are we forced therefore to accept the presupposition with its implicit claim that, because of the existence of the institution of promising, one can acquire an obligation by following certain instructions? Surely not. We are free to take the evaluative position of denying that the institution in question enables us to acquire obligations. This would require us to take the somewhat awkward stand of refusing to admit that people have made promises even when they have followed the instructions. But on my view you can justify this on the ground that to admit that the promise has been made would commit you to accepting the presupposition that the institution of promising does enable people to acquire obligations. Just as one can recognize that it would be wrong to say that the present King of France is not bald, but balk if asked to admit that therefore he is bald, similarly one can acknowledge that it would be wrong to say that a person has not promised, yet legitimately refuse to say that he has promised.

Searle insists quite rightly that you cannot cover all the facts implicit in the statement that a person has made a promise without talking about obligations. But this does not mean that to cover the facts you must make an evaluative judgment. The statement that an institution is designed to enable people to acquire obligations is about obligations, but it is not an evaluation. The statement that an institution enables people to acquire obligations looks very similar. The difference between the two statements is that the first asserts that an attempt has been made; the second, that the attempt has succeeded. If you are talking about the digging of a ditch, facts may require you to admit the occurrence of a success just as readily as they may require you to admit the occurrence of an attempt. If the statement that a person tried to dig a ditch is purely factual, so is the statement that he succeeded in digging a ditch. But the same thing does not hold for the creation of obligation. The statement that a person tried to create an obligation is purely factual, but the statement that he succeeded is not. Attempt statements about the creation of obligation (henceforth ‘attempt statements’) may be implied by an account of the facts; success statements about the same subject (henceforth ‘success statements’) may not.

But why, after all, cannot we be forced by the facts to accept success statements? I could ask to be shown just what fact would make me go from admitting an attempt to admitting a success, and I can think of none. However, there is another reason, namely, that success statements are evaluative but attempt statements are not. But has not Searle already said that we can be forced by the facts to accept a statement that is evaluative? Indeed he has, but he did not expect us to accede to that claim unless he could force us to do so by a clear cut example. He seemed to have given us one in the case of the statement that a person made a promise. But we are not forced by the facts to accept the statement that a person made a promise unless we are forced by the facts to accept the success statement presupposition. The statement that a person made a promise is not a clear- cut example of a judgment that we are forced to make by the facts unless the success statement is. But the success statement is not, so Searle has failed to provide his clear-cut example, so we are entitled to stick to the view that if a judgment is evaluative we cannot be forced by the facts to accept it.

If my analysis of the statement that a person made a promise is correct, the truth of my claim that we cannot be forced by the facts to accept success statements about the creation of obligations is not only sufficient to provide a refutation of Searle’s attempt to derive a prima facie  ‘ought’ from an ‘is’ but also necessary for such a refutation. For if we could be forced by the facts to accept such a success statement, we could be forced to admit that a promise has been made and that, other things being equal, there was an obligation to keep it.

I have tried to show that Searle has failed to derive an ‘ought’ from an ‘is’. I shall now denigrate the task I set myself. To put it baldly, if you could get an ‘ought’ from an ‘is’, it would not be worth the trouble of doing it. If ought-statements involve an attitudinal commitment, you cannot derive an ‘ought’ from an ‘is’. If they do not involve such a commitment, you must still decide what you are basically in favor of doing after you have decided what you ought to do [For a more extensive defense of this point see Peter Singer, ‘The Triviality of the Debate over “Is-Ought” and the Definition of “Moral”’, American Philosophical Quarterly, January 1973]. Whether ought-statements involve a basic attitudinal commitment or not, you still have to make your basic decisions, and as Hume pointed out it is not ‘contrary to reason to prefer the destruction of the whole world to the scratching of my finger’. This does not mean that anyone who has the basic desires of a typical human being is at all likely, when he has his wits about him, to prefer the destruction of the world. But it does mean that if he did not already have a basic desire to save the world, or to do something that required the world to be saved, facts alone could not force him to choose to save it. I should add that although Hume is credited with the view that you cannot derive an ‘ought’ from an ‘is’, I think that he would agree that there is a sense in which it does not really matter.


Facts, Promising and Obligation
Author(s): R. I. Sikora
Source: Philosophy, Vol. 50, No. 193 (Jul., 1975), pp. 352-355
Published by: Cambridge University Press on behalf of Royal Institute of Philosophy
Stable URL: http://www.jstor.org/stable/3749860