CONDITIONS FOR AN ACT OF EXPLAINING

Explaining is what Austin calls an illocutionary act.¹ Like warning and promising, it is typically performed by uttering words in certain contexts with appropriate intentions.

The illocutionary character of explaining can be exposed by formu­lating a set of conditions for performing such an act. To do so I shall consider sentences of the form “S explains q by uttering u,” in which S denotes some person, q expresses an indirect question, and u is a sentence. (I will assume that any sentence of this form in which q is not an indirect question is transformable into one that is.)^{[54] [55]}

The first condition expresses what I take to be a fundamental rela­tionship between explaining and understanding. It is that S explains q by uttering u only if

(1) S utters u with the intention that his utterance of u render q understandable. This expresses the central point of S's act. It is the most important fea­ture which distinguishes explaining from other illocutionary acts, even ones that can have indirect questions as objects. If by uttering u I am asking you, or agreeing with you about, why the tides occur, by contrast to explaining it, I will not be doing so with the intention that my utter­ance render why the tides occur understandable. (I shall return to the concept of understanding in section 2 after formulating the remaining conditions.)

To explain q is not to utter just anything with the intention that the utterance render q understandable. Suppose I believe that the words “truth is beauty” are so causally efficacious with you that the mere utter­ing of them will cause you to understand anything, including why the tides occur.

(2) S believes that u expresses a proposition that is a correct answer to Q. (Q is the direct form of the question whose indirect form is q.)

Often people will present hints, clues, or instructions which do not them­selves answer the question but enable an answer to be found by others. To the question “Why do the tides occur?” I might respond: “Look it up in chapter 10 of your physics text,” or “Newton's Principia has the answer,” or “Think of gravity.” Some hints, no doubt, border on being answers to the question. But in those cases where they do not, it is not completely appropriate to speak of explaining. By uttering “Look it up in chapter 10 of your physics text” I am not explaining why the tides occur, though I am uttering something which, I believe, will put you in a position to explain this.

These conditions are not yet sufficient. Suppose that S intends that his utterance of u render q understandable not by producing the knowl­edge that u expresses a correct answer to Q but by causing people to come to think of some nonequivalent sentence u' which, like u, S believes expresses a proposition that is a correct answer to Q. In such a case, although S utters something which he believes will cause others to be able to explain q, S does not himself explain q by uttering u. For example, to an audience that I believe already knows that the tides occur because of gravitational attraction, I say

u: The tides occur because of gravitational attraction of the sort described by Newton.

Although I believe that u does express a correct answer to Q (Why do the tides occur?), suppose that I utter u with the following intention: that this utterance will render q understandable not by producing the knowledge of the proposition expressed by u that it is a correct answer to Q, but by causing my audience to look up the more detailed and pre­cise answer actually supplied by Newton, which I don't present. This is like the situation in which I give the audience a hint that in this case is a correct answer, but is not the answer in virtue of which I intend q to be understandable to that audience.

To preclude such cases we can say that S explains q by uttering u only if

(3) S utters u with the intention that his utterance of u render q understand­able by producing the knowledge, of the proposition expressed by u, that it is a correct answer to Q.

In the case of the tides, I do not intend that my utterance of u render q understandable by producing the knowledge of the proposition expressed by u that it is a correct answer to Q, but by producing such knowledge with respect to another proposition. So, according to condition (3), in such a case by uttering u I am not explaining why the tides occur.

Suppose, by contrast, I know that my audience is familiar with the answer supplied by Newton, but its members have no idea whether this answer is correct. Since the audience knows what sort of gravitational at­traction Newton describes, I might explain why the tides occur, simply by uttering u. In this case I intend to render q understandable by producing the knowledge, of the proposition expressed by u, that it is a correct an­swer to Q. It is possible for me to have this intention with respect to u since I know that the audience is aware of the sort of gravitational attrac­tion described by Newton.

Let us change the example once more. Suppose I believe that the au­dience does not know that the tides are due to gravitational attraction. I now proceed to utter u above with the intention that my utterance of u will render q understandable by the following combination of means (which I regard as jointly but not separately sufficient for rendering q un­derstandable): (i) producing the knowledge, of the proposition expressed by u, that it is a correct answer to Q; and (ii) causing others to look up some different, more detailed, proposition (supplied by Newton) which is also a correct answer to Q. By uttering u am I explaining why the tides occur?

One might be inclined to say that I am both explaining q by uttering u and giving a clue about where to find another answer to Q.

In section 5, some further conditions (involving restrictions on q and u) will be suggested whose formulation requires concepts to be intro­duced later. For the present I shall treat these three conditions as not only necessary but jointly sufficient. If so, then the same honor can be accorded to (3) by itself, since (3) entails both (1) and (2).

Although “explain” may be used in describing an act governed by these conditions, it can also be employed in a more restricted way to cover only correct explainings. We can say that Galileo explained why the tides occur, even though he did so incorrectly, or that he failed to explain this, even though he tried. When one has correctly explained q by uttering u one has performed the illocutionary act of explaining q and in doing so one has provided a correct answer to Q. In what follows, however, when reference is made to acts of explaining I shall mean acts for which this is not a requirement. (See chapter 1, section 8, for an account of “correct explanation.”)

2.