The Logical Character of Action-Explanations

Citation preview

The Logical Character of Action-Explanations Paul M. Churchland The Philosophical Review, Vol. 79, No. 2. (Apr., 1970), pp. 214-236. Stable URL: http://links.jstor.org/sici?sici=0031-8108%28197004%2979%3A2%3C214%3ATLCOA%3E2.0.CO%3B2-J The Philosophical Review is currently published by Cornell University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/sageschool.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sat May 12 00:24:33 2007

THE LOGICAL CHARACTER O F

ACTION-EXPLANATIONS1

I

S I T THE CASE that garden-variety explanations of human actions-explanations in terms of wants, beliefs, and so onare sketches or enthymemes of deductive-nomological arguments ? A considerable number of contemporary writers maintain that they are not. Some claim that there simply are no empirical laws governing human action as a function of configurations of mental states and/or episodes, or at least that if there are such laws, the ordinary man is quite ignorant of them. Indeed, some insist that, for purely logical reasons, there could not possibly be any empirical laws connecting wants and the like with actions. Accordingly, we are invited to embrace various alternative theories as to the logical character of action-explanations. By some it is argued that action-explanations are an instance of a general type of explanation distinct from the D-N (or at least from the causal) type (for example, Charles Taylor2 and Alan Donagans) ; others claim that action-explanations are of a distinct and unique type in themselves (for example, A. I. Melden4 and William Dray5). I cannot hope to examine all of them, but one of the more interesting of these alternative theories will receive critical attention in the latter part of this essay. My immediate aim is to make a prima-facie case in favor of the view that action-explanations are indeed of the familiar D-N mold. This sort of task has been made less arduous than it might have been by Donald Davidson's most enlightened attack on many of the standard arguments 1 A major acknowledgment is owed to Miss Patricia Smith. Most of the major details in what follows were forged in conversations with her during the summer of 1968. 2 The Explanation o f Behaviour (London, I 964). 8 "Explanation in History," in Theories o f History, ed. by Patrick Gardiner (Glencoe, Ill., 1959). Free Action (London, I 96I ) 6 See esp. "The Historical Explanation of Action Reconsidered," in Philosophy and History, ed. by Sidney Hook (New York, I 963).

.

ACTION-EXPLANATIONS

against the causal theory of action- explanation^;^ unfortunately, Davidson makes only a minimum positive contribution with respect to outlining the structural details of our explanatory capacity in this area, apparently regarding it as consisting of only the most inchoate of insights into possible and probable nomic connections. I shall try to make out that, on the contrary, there are some fairly sophisticated nomic principles or "laws" specijically presupposed by our ordinary action-explanations, and that a variety of interesting features of action-explanations can be clearly understood when this background structure is brought to light. That our ordinary explanatory practices with respect to human actions are presumptive of what is prima facie a general law can fairly easily be brought to light by a systematic examination of those practices. Any adequate theory of the logical character of action-explanations must be able to account for the undoubted propriety of the various types of everyday objections to which they can be subject, and this fact provides us with a strategy for winnowing out the underlying law, if it happens there is one. We need only examine and classify the types of objection which can legitimately be raised against an ordinary explanatory statement of the form "X A-ed because he wanted ~ 3 ' in ' ~order to bring out the entire set of necessary conditions (short of the law itself, it being, presumably, immune from casual denial) for the correctness of that explanatory statement. If action-explanations are sketches of D-N arguments, then there must be some putative law which sanctions an inference from the conjunction of these explanatory conditions to the desired explanandum, "X A-ed." Presumably, it will simply be a universal conditional containing the conjunction of these explanatory conditions as its antecedent, and "X A-s" as its consequent. The strategy is to winnow out, from examples of objections, these implied but usually unstated explanatory conditions, construct -- -

"Actions, Reasons, and Causes," Journal of Philosophy, LX ( I 963). One can substitute for 0 here anything that fits. I n what follows, however, I shall understand it as running surrogate for the completely general "that P (be the case)." In one indirect way or another, wants, like beliefs, are all identified by reference to some specific proposition. 7

PAUL M. CHURCHLAND

the universal conditional as indicated, and see whetherwerecognize it as a nomological, as a true, nontrivial, unrestricted universal conditional capable of supporting subjunctive conditionals. One look may not suffice to settle the issue, but if it does so appear, the conflict will be usefully joined. Such a strategy is simple and potentially decisive, but for a variety of reasons it is difficult to carry through cleanly. First, the ways in which one can object to an explanatory statement of the form "X A-ed because he wanted 0"are legion, a consequence of the fact that one can object by denying anything which, in context, is implied by any one or any conjunction of the basic explanatory conditions (the singular premises in the relevant form of D-N argument). The filtering out of these basic explanatory conditions is therefore something less than a mechanical process. Prospects for immediate and clear-cut success are further clouded by the possibility that the putative law we seek to unearth is, in fact, very much a law sketch,8 and that one or more of the corresponding explanatory conditions may therefore be vague and hence somewhat elusive. And lastly, it appears that there are, in addition to the normal case, two or three slightly aberrant cases of "intentional action," and that there are distinct though only slightly different sets of correctness conditions on explanatory statements appropriate to each of these kinds of cases. To make anything approaching an irrefragable case for the specific conditions to which I have been led would therefore require a dismal number of examples, and the following minimum set must be viewed as being merely illustrative of (what I take to be) the basic explanatory conditions in the normal case. The reader can check them out and perhaps improve upon them by stumping through further examples of his own contrivance. I n the normal case, if one says, "X A-ed because he wanted 0 ," one implies that ( I ) X wanted %, and (2) X believed (judged, saw) that A-ing was, under the circumstances, a means for him to achieve %, an action which would achieve % or contribute to his achievement of % . One's explanatory statement would normally be defeated if either of these were shown to be false. Is anything This notion will be made clearer presently.

ACTION-EXPLANA T I O N S

further implied by such an explanatory statement? Apparently so. Suppose the following conversation to take place at a noisy party.

P: "Why did Peter crook his finger back and forth like that?" Q: "Well, he wanted Mary to come across the room to him, and, strangely enough, he thought that would bring her." P: "Yes, I know he wants very badly for Mary to come over and be sociable to him-and don't you kid yourself, because Mary would come at the crook of his finger, and he knows it-but Peter has a lot more style than that: he'd have called her over with a fetching grin or some such, not a crooking finger. He must have been calling Bill over; he's standing over there with Mary." In this case, neither of the explanatory propositions-(I) and (2)-put forth by Q i s denied by P. What is denied is that they explain or are relevant to the explanation of Peter's crooking his finger. The point of the objection is this: if, at the time at which Peter was in fact crooking his finger he had been doing something as a means to fulfilling his desire for Mary's company, then that something would not have been his crooking his finger, but something else-his smiling fetchingly at her or some such. And the general point behind this claim seems to be that if he had been doing something as a means to the desire cited, then that something would have been the means he thought or took to be the most preferable to him, all things considered, of the means he thought open to him. P's objection is simply that, in virtue of being presumptuously blatant, crooking his finger was unlikely to have been the means thought of or judged by (stylish) Peter as the means most preferable to him of the means he thought open to him, given that he was in fact aware of other alternatives, and hence most unlikely that e s explanation is correct. What this example illustrates is the following apparently general truth. If one says, "X A-ed because he wanted 0 ," one has made a true statement only if (3) there was no other action believed by X to be a means for him to bring about 0 , under the circumstances, which X judged to be as preferable to him as, or more preferable to him than, A-ing. Further implications can be brought to light in the same

PAUL M. CHURCHLAND

pre-theoretical manner. Suppose that P had objected to c s explanation not as above, but as follows.

P: "No, no, that can't be right. Peter knows that Jack is watching him very closely, and he also knows that the jealous ruffian would cause a fearful row if he tried in any way to gain Mary's company. I'm sure Peter does want Mary's company, and I agree that crooking his finger is just what the graceless fool would do, but I'm sure the whole thing isn't worth an ugly scene to him. He must have been calling Bill; he's standing over there with Mary." As before, P does not deny ( I ) or (2), and in this case P even agrees that crooking his finger is precisely the means Peter would have (and perhaps has) judged most preferable to him of the means (to bring Mary over) currently open to him. This time the objection is to the effect that Peter had a further desire which, under the pregnant circumstances, overrode his desire to bring Mary over. In brief, Peter would rather bide his time than precipitate a scene, and he judges the former to be the price of avoiding the latter. The general lesson here is roughly this: if one says, "X A-ed because he wanted 0," one has made a true statement only if (4) X had no other want (or set of them) which, under the circumstances, overrode his want 0 . Now suppose the following objection raised against c s explanation. R: "I'm afraid that's quite wrong. You see, Peter is subject to an unfortunate variety of nervous paralysis which comes and goes quite without warning. One can tell when an attack hits him because his eyes droop glassily and his finger twitches spasmodically. He may have been about to crook his finger, doubtless for the very reasons you mention, just as the attack came, but with it came a complete inability to control any part of his body. His finger did crook, perhaps right on time, but Peter didn't crook it: I saw his eyes droop glassily the instant before his finger began to move." Unlike the previous objections, this is not just a challenge to the offered explanation in particular; it is a challenge to the propriety of an action-explanation in the case at issue. It denies that Peter

ACTION-EXPLANATIONS

crooked his finger-in the full-blooded sense that would merit an explanation of the sort offered by Q-on the grounds that he was unable to crook his finger. The lesson here is simply this: to give an action-explanation is to imply that the agent had the ability to perform the alleged action being explained. That is, if one says, "X A-ed because he wanted 0 ," one has made a true statement only if (6) X was able to A. Finally, consider the following conversation. P: "1 just saw John dial Bill's number on the telephone. Why do you suppose he did that ?" Q: "What's the problem? He wanted to talk with Bill. He thought dialing Bill's number the best way to achieve this under the circumstances-Bill went back upstairs to his office, you know-and since no other desire(s) overrode that inclination, he, being able, went ahead and dialed Bill's number !" R: "That's not quite right. John believes that Bill's number is extension 3, whereas of course it is 2. You are right about his wants, beliefs, preferences, and abilities, and Bill's number did get dialed by John, but the pencil must have slipped in his hand as he went to dial 3, Bill's real number, 2, being dialed only accidentally. I t rather looked like it, you'll notice, for John hung up immediately and dialed again." Because John was mistaken as to how (or what it was) to dial Bill's number, he cannot properly be said to have dialed Bill's number in the full-blooded sense to which JL's explanation would have been appropriate, even though, as it chanced, Bill's number did get dialed by John. In this respect the case is like the preceding e ~ a m p l efor , ~ the relevant condition here is perhaps best described as a condition on the propriety of an action-explanation (for the action under the description at issue). The general I might mention that I have switched examples here (from the fingercrooking case) for reasons of simplicity only. I t would have been prima facie desirable to hold the example constant for all six conditions, but since one can imagine being able to crook one's finger, but not knowing how and, further, doing it accidentally anyway, only in the most unusual and complicated cases, time and space would seem to justify illustrating the point with a more manageable example.

PAUL M. CHURCHLAND

lesson here is roughly this: if one says, "X A-ed because he wanted 0 ," one has made a true statement only if it is the case that (5) X knew how to A (or, at least, that X's belief or judgment, concerning how, in those circumstances, to A, was correct).1° An interesting, but mistaken, objection is possible here. One might suppose that it would be correct to say, "X A-ed because he wanted % ," if X A-ed, in complete ignorance of that fact, in the belief that he was B-ing, and where in fact he B-ed because he wanted % . If this were correct, it would undercut more than just condition ( 5 ) , but one need only consider an example to see that there is something wrong with this suggestion. Suppose that X brings the water to the boil because he wants to make a cup of coffee. Now in bringing the water to the boil, he therein raised its vapor pressure to equal atmospheric pressure (Pa) ;but suppose he is quite ignorant of this fact, not knowing even what vapor pressure is. That is, he does not know how (or what it is) to raise the water's vapor pressure to Pa. Clearly then, there is something wrong with saying that X raised the water's vapor pressure to Pa because he wanted to make a cup of coffee, because he did not