What We Can Do

Citation preview

What We Can Do Arthur Danto; Sidney Morgenbesser The Journal of Philosophy, Vol. 60, No. 15, Symposium: Human Action. (Jul. 18, 1963), pp. 435-445. Stable URL: http://links.jstor.org/sici?sici=0022-362X%2819630718%2960%3A15%3C435%3AWWCD%3E2.0.CO%3B2-7 The Journal of Philosophy is currently published by Journal of Philosophy, Inc..

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/jphil.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 Fri May 18 08:28:01 2007

outlined them "causal explanations"? There is the extreme view, endorsed by many philosophers, which categorically denies that they are causal explanations-and indeed that any alleged explanation of action on the basis of wants could succeed in being a genuine causal explanation. However, what these philosophers mean by 'causal explanation' is often unclear, and, when it is clear, it is not a t all certain that all the familiar explanations found in the natural sciences are "causal" in any single required sense. Sometimes these philosophers seem to have in mind something analogous to the mechanical model of classical physics; sometimes, deductive explanations with general laws and antecedent conditions; sometimes any explanation in which reference is made to antecedent events or states; sometimes an explanation in which the explanans includes facts that are occurrents in a sense in which wanting, at least as we have explained it, is not; and so on. TITefind it hard to attach much importance to this issue, in the absence of a generally accepted and methodologically interesting conception of "causal explanation" that is appropriate to the issue a t hand.j What we think clear and important in this dispute is the question of whether or not sensible explanations of human actions exhibit the appropriate inferential and nomological pattern of explanations found in physical and biological sciences-in other words, whether explanations of action form a unique type of explanation with special logical and methodological requirements distinct from those of explanations in natural science. Obviously, if our analysis of wanting and of explanations of action in terms of wants is plausible, a t least in basic outline, it is one good reason for thinking that explanations of action do not differ from explanations in natural science, in inferential nomological pattern. RICHARDBRANDT JAEQWON KIM SWARTHXORE COLLEGE

WHAT W E CAN DO this paper I shall presuppose the following notions: * (1) B is a hasic action of a if and only if ( i ) B is an action and (ii) whenever a performs B, there is no other action A performed by a such that B is caused by A.

IN

5 We also agree, on all essential points, with W. P. Alston's recent able criticisms of some philosophers' arguments for the thesis that explanations of actions are not causal explanations, in his paper '(Wants, Beliefs, and Action," read to a conference a t Wayne State University in April, 1963, of which he kindly furnished us a copy. These notions are defended in a companion paper, "Basic Actions," to be published subsequently.

436

THE JOUBA'AL O F PHILOSOPHY

( 2 ) B is a nonbasic action of a if there is some action A, performed by a, such that B is caused by A. ( 3 ) Whatever further may be said in the analysis of ' m causes n', m causes n only if m and n are distinct events. There may be other senses of '. . . causes -' but this is the only sense I am interested in here. ( 4 ) There is a basic repertoire R of basic actions such that a is a normal person only if a possesses R. a is positively abnormal if ( i ) a performs B , (ii) B is a basic action, and (iii) B is not included in R. a is negatively abnormal if ( i ) a cannot perform B , (ii) B is a basic action, and (iii) B is included in R. Any paralytic is negatively abnormal; anyone who can dilate his pupils "at will," i.e., without performing some action which causes them to dilate, is positively abnormal; raising an arm without performing some action that causes it to rise is included in R. ( 5 ) If there are any actions at all, there are basic actions. ( 6 ) There are basic actions. (7) Not every action is a basic action. (8) If B is an action performed by a, then either (i) B is a basic action or (ii) B is the terminal effect of a causal series the originating member of which is a basic action.

These notions taken together constitute a fragment of what I shall call a theory of action. I t is not difficult to find a familiar analogue to this theory of action in a certain theory of knowledge, e.g., by substituting 'basic sentence ' for 'basic action ', ' is inferred from' for 'is caused by', 'sentence' for 'action', and by specifying a notion of "epistemic normality" for (4) (e.g., for 'paralytic' substitute 'blind man'; for 'can dilate his pupils . . .' substitute 'has a sixth sense'; and let "seeing" be the paradigm case of epistemic normality.) One could retain 'distinct events' in (3) if one wished to keep the notion of performatives; otherwise one might give another interpretation for 'distinct events' or just drop (3) altogether. I am only indicating the analogue in a rough way, but there is, I think, a suf6ciently plain structural parity for us to suppose that this theory of action and this theory of knowledge are, as it were, "models" for some uninterpreted and, I may say, heretofore unwritten calculus. If someone were to work the calculus out, he would have the logical armature for a whole class of philosophical theories, including the two I have mentioned. I am not to be concerned here, however, with issues in metaphilosophy. Instead, I want to give a highly informal gloss on (7).

WHAT W E CAN DO

43 7

I t is radically unclear from the bare description "a's right arm rises at t" whether a performs here a basic action or, indeed, whether a has performed an action at all. We can determine this only by contextual specification. I t will have been a basic action only if a raised his arm without causing it to rise, and a nonbasic action only if a caused it to rise by performing a basic action distinct from it, e.g., by pushing it up with his other arm. Otherwise it is not an action of his at all; e.g., if someone or something other than a caused the arm to rise, or if a suffers a special nervous disorder and has undergone a spasm, etc. On the other hand, it is commonly taken as clear from the bare description "a's hat rises at t" that whether or not a has performed an action here, he has at least not performed a basic action. Any event that is sometimes a basic action admits of a description which leaves it unclear whether a basic action has really been performed. But some actions admit of comparable descriptions which, while they leave it unclear whether an action has really been performed, at least make it clear that a basic action has not been performed. So not every action is a basic action. Nevertheless, some individuals are positively abnormal. So "a's pupils dilate at t", while it normally leaves it clear that a basic action has not been performed, still leaves room for the possibility that a has performed a basic action if he is positively abnormal in the required way. Whatever we can do as a basic action, we can also cause to happen. But some things that normal persons can only cause to happen, some individuals can also perform as basic actions. They have, to put it platitudinously, two ways of doing what we can do in only one way. I am to be concerned with the limits of positive abnormality, with actions that are never basic actions. This is a purely philosophical concern, I think, for so long as we can do a certain thing, it little matters practically whether we can in fact do it in only one way (and all ways of causing something to happen constitute, from my point of view, one way of doing it). Comparably, there may be many sentences that we can only know via inference, but this does not entail any ultimate ignorance-which is cognitive impotency-unless we have, unreasonably, restricted knowledge to direct, or nodinferential knowledge. 1 I take the expression 'a causes B', when a stands for an individual, to be a shorthand way of saying that, if B is an event, then some other event in a's biography stands to B as cause to effect. The relationship between this event and a is not easily analyzed, but for purposes of this discussion I shall leave i t unanalyzed.

438

T H E JOURh7AL OF PHILOSOPHY

Now raising a hat seems to be a natural candidate for an action which lies beyond these limits and which no individual, however otherwise positively abnormal, can claim as included in his repertoire of basic actions. The question is only whether this limit (and this ' can') is logical or factual. And this question may have some implications for philosophical theology. Does the alleged omnipotency of God require that whatever God can do as an action He also can do as a basic action? It is sometimes held that not only is God omniscient, but whatever He knows, He knows directly and non-infer en ti all^.^ So I am considering the natural analogue to this in the theory of action. True, we would doubtless find it unintelligible were someone to claim that God can raise a hat without causing it to rise. But then we also find it unintelligible, because we cannot do i t ourselves, that someone, e.g., Felix Krull, can dilate his eyes without causing them to dilate. And so, too, does a paralytic, for similar reasons, find it unintelligible when we claim to be able just to raise our arm. We cannot answer him when he asks us how we do it if, for instance, he expects a recipe. For there is nothing we do first in order to cause our arm to rise. Or rather, raising our arm is what we do first. And this is what he cannot understand.

I1

I t is useful here to consider telekinesis. This is a possible view of how things might be moved, which arises, I think, only because people are dominated by a causal picture of action. Regarding an arm as, so to speak, an alien entity which, in its owner's case at least, is caused to move through the issuance of some inner summons, e.g., a volition, it becomes genuinely puzzling that we cannot, and genuinely possible that we might, through issuing an exactly similar summons, cause a similarly alien entity-a hat perhapsto move. As though there were a kind of psychic energy which, since it undeniably (sic) causes some crass objects (arms) to move, might be harnessed and redirected to cause the movement of other crass objects (hats). And indeed this is a wholly plausible extension of an unfortunately wholly implausible theory of how men move their limbs. Indeed, if those who believe i n telekinesis were so much as able to find one man who moved his arm in the way in which they hope to be able to move hats, they would already have established their claims. The only thing is that they 2 ( ( T h e manner in which God knows the infinity of propositions is utterly more excellent than is the manner by means of which we know the few that we do. We proceed by argumentation, and advance from conclusion to conclusion, while God [apprehends] through a simple, sudden intuition." Galileo, D i d o g o Sopra i due Massimi Sistemi del Mondo.

TVHAT TVh' C A X DO would have found a rare specimen, for I think few people, if any, move their arms by causing them to move through discharging psychic energy across whatever strange distance i t is that supposedly separates them from their arms. Such rare specimens might move hats in much the same way. But the question I am concerned with is whether it is possible that we might move hats the way we do move an arm, not by causing them to move (and discharging psychic energy would only be a special kind of cause), but by just moving them-the way we move our arms: as basic actions. A natural suggestion for answering "no" to this question is that hats are not parts of us, though arms are, and a man's repertoire of basic actions must be performed with parts of himself. But of course, there are parts of every normal person with which he cannot perform basic acts: fingernails, hair, teeth, not to mention the interior organs. So at best "P being part of a" is only a necessary condition for "a can perform basic acts with P." But it immediately follows that the latter is a sufficient condition for the former; so if I can perform basic actions with P, P is part of myself. If I then could move a hat as a basic action, that hat would be part of me. And so others could only cause it to move, for others have only one way of moving parts of me. This creates what I should call the Problem of Other Bodies. I should in fact like to adopt as a criterion for something being a part of a man's self that he should be able to perform basic actions with it. This would suggest a form of metaphysical dualism not greatly in favor since, I believe, the vague days of German Idealism, a dualism f a r more interesting and rather more plausible than the customary dualism it intersects with, namely, the dualism of mind and body. Roughly, it would be a dualism with the self on one side and everything else on the other, and the self would consist in everything with which a man performs basic actions. The self would admit of increase and diminution, in the form of positive and negative abnormalities. When it diminishes to zero (death) a man is no longer able to act, for he is no longer able to perform basic actions: nothing is any longer part of him, because he is nothing. But this is in fact a digression, helping not a t all with our problem. Or a t all events, it merely restates it. F o r the limits of positive abnormality and the ultimate limits of the self are one and the same thing.

Obviously, the normal man is not even in a position to try. Fur I understand by 'trying to do B' not "making a special effort

THE JOURNAL O F PHILOSOPHY

in the hope of doing B, hoping that the special effort will cause B to happen," but instead "doing B with a special effort," e.g., in adverse circumstances-like moving one's arms against ropes that bind one. One feels the effort in doing B thus, and the temptation is to isolate the feeling, dignify it as "an act of will," then transfer it to the normal case, where circumstances are benign, and finally suppose that there are episodes of trying that cause the actions we perform. To try to move a hat as a basic action would then be to move it with effort, e.g., when it is nailed to the table, and this is already to have it in one's repertoire. So the normal man cannot try to move a hat as a basic action. But neither can he try to dilate his pupils or to retract his finger nails. Or to move his ears. There are those who are positively abnormal in being able to move their ears as a basic act. They have that gift. And let us imagine that we, too, one day find ourselves with this gift, our self having become augmented in some dark way. How should a man know, on the very first occasion, that he is doing it, and that it is not just happening, that he has not been suddenly afflicted with a nervous spasm? Finding ourselves newly in control of a part of our body heretofore alien is not an experience most of us have this side of infancy, when, in time, it became clear that some things we could do directly, and some things we could only cause to happen. Perhaps the concept of cause only becomes clear when the limits of the self become clear. If so, then Hume was looking in just the wrong place when he sought to find the origin of our idea of causal power in our ability to move our limbs, for it is just there that causes are not to be found. If we did everything in the way in which we move our limbs (normally), we should have no notion of cause. So small wonder that Hume was di~appointed.~Anyway, I am speaking of a man who has these concepts, and finds that he can do as a basic act what he before could only cause to happen. And I am asking how, the first time, he knows that he has done it. I should think that the power and the knowledge of the power come a t the same time. How does one, for example, know that one is raising one's arm when one does this as a basic action? This seems to be one of the things a man knows directly and not, as i t were, on the basis of evidence. Nor is i t something I know only because I notice it happening. That is, had I not noticed it (or had it brought to my attention because someone else noticed it and told me), I might not otherwise have known. If one day I 3 David Hume, An Inquiry Concerning Human Understanding, Sec. VII, Pt. 1.

W H A T TVE CAN DO

44 1

should notice that my arm was rising and lowering, and then realize that, if I had not noticed, I would not have known it was doing this, this would be for me a terrifying experience, a sign that I had lost contact with part of myself, that my arm had become an alien entity. Now a man who looked in the mirror and found his ears moving in this way, that is, noticed that they were, would have hardly a less terrifying experience. I t would be like finding that his face was covered with blotches he had had no suspicion of. So I am suggesting that even on the first occasion, when there has been the sudden acquisition of a new gift, this is not something a man would know about by only noticing that, say, his ear was moving. I can imagine that one to whom this happened might rush to a mirror in order to check up. I should only find i t hard to imagine him not finding what he expects. Largely because the gift consists in being able to start and stop the motion of one's ears without having to cause these things to happen, and checking up would consist in doing these things in front of the mirror; and if one does not know what to do, how can one check u p ? Notice that I might feel my ears moving and go to the mirror to check whether they were moving, but this will not tell me whether I was moving them. The mirror shows the same thing whether I am moving them or they are only moving. So I would have to know already if I were to know at all. Do I after all know that I am seeing by noticing that I a m ? So that, had I not noticed, I would not have known? But in what should such noticing consist? Surely not in having some auditory experience.

Of course, one can sometimes notice that one's arm is moving, when one would not otherwise have known. Someone has anesthetized my arm, and sent a small electrical current through to some part of my brain. My mind is taken up with other things, and the experimenter cries out in great excitement: "Look! the arm is moving!" And I look and notice that it is moving, though I would not otherwise have known. The experiment is a success. I am in no special panic, for I know what is happening. I t is not terrifying as it would have been had I not known what was happening, and awoke to see my alienated arm traversing space. I n a case such as I have described, though, I have no better, and perhaps have a lesser authority for pronouncing on the motion of my arm than a favorably situated second person. But is it possible that it should be brought to my attention, or

442

THE JOURNAL O P PHILOSOPHY

that I myself should discover through noticing the fact, that my arm was n o t moving when I should have said, with my usual authority, that I was moving my arm? I am told to raise my arm, and then berated for not obeying, and I insist that I have obeyed, and then am told to look. I look. And there is my arm which I thought was aloft hanging instead at my side. Can there be such errors? Epistemologists often enough tell us of the strange cases of phantom limbs, where a man feels pains in limbs no longer his, where he would point to a space left empty by a sectioned arm if asked where the pain was. Had he not known the arm was gone, he would have said the hurt was there, so there might have been a point when he would have said the pain was there and then noticed his arm was gone. Might there not then be amputees who feel themselves to be raising an arm when they have no arm to raise? And then might there not have been that awful first moment when they noticed that they had no arm though they would have said they were raising i t ? And then might this not be our situation at any minute? One discerns here the first crack which can widen into a skeptical abyss, but I am less troubled by this than I am by the fact that my own position is threatened if not destroyed. For I have said that when one moves one's arm, one knows this, knows the thing itself, and not on the basis of some kind of evidence. But if illusions are possible, there must be space for them to enter, and if there is the required space, where, and between what, can it, on my account, be located? But if there is no room for a gap on my account, how can I be right? I t might then be thought that the correct account cannot be very different from one I shall now sketch. If a man discovers, but would otherwise never have known, that his arm was not moving when he would have said that he was moving his arm, the question must be raised as to what his assertion would have been based upon. Whatever it would have been based upon, i t could not have been based upon something different from what it would have been based upon in the normal case where a man would say that he was moving his arm and his arm in fact was moving, so that he would be saying something true. The two cases cannot differ except for the moving of the arm in the one case and its not moving in the other. For if what his assertion were based upon could be different, then the man would be able to tell, on the basis of this difference, whether his arm was moving or not, and the assumption is that he i n fact finds this out only by noticing that the arm is moving or that it is not. So something must be invariant in the two cases, it must

W H A T TVE C A N DO

443

be this upon which his assertion is based, and plainly the behavior of the arm is ?tot invariant in the two cases. This invariant factor, let us suppose, is a kinesthetic sensation, a feeling, perhaps, that one is moving one's arm. Call this K. Then the account-which I shall term the Inductivist Account-runs this way. A man associates K, over time, with the moving of his arm. A t some time this noticed correlation becomes for him a habit of expectation. So whenever he has K he expects that his arm is moving, and commonly he is right, since his arm moves. B u t we all know how i t is with inductions. It is possible a t any point that K will fail to precede the arm motion it is taken to herald, it being only our man's great luck that it never actualizes. B u t we are thinking now of a case where i t does become actual. B y contrast, the ilormal case is this: ( a ) the man has K ; (b) the men expects his arm to move; (c) the man's arm moves. I am unable to accept the Inductivist Account of how we know that we move our arms. F o r I can imagine a case where ( a ) , ( b ) , and (c) all are true though the man does not move his arm. The two events "a has K" and "a's arm rises" may be situated in parallel causal chains. Rather as though some metaphysical jokester were to sever the connections between pilot room and ship and reattach them elsewhere, so that the pilot, though he makes all the right moves, is not steering the ship, even though the ship behaves just as i t would if he were in fact driving it. Knowing that he is making a turn and noticing that the ship in fact turns in the direction he wanted it to turn in, the pilot rests complacent in a sense of efficiency when credit ought properly to go to a pre-established harmony between pilot-moves and ship-moves. This difficulty cannot be forestalled by the Inductivist Account unless it adds a new truth condition: ( d ) the man moves his arm. B u t how can this be added when in fact i t is supposed to be exiiaustively analyzed as (a)-(c) ? This cannot be a n analysis, however, if (a)-(c) are true and ( d ) is false. I should prefer to say that there is no connection between me and my body to be cut. There is that much point to the negative inetaphor that I am not in my body the way a pilot is in a ship. There is no empty space between me and parts of me for a Cartesian spirit to haunt. Which does not mean that the self, as I have characterized it, cannot admit of temporary or permanent diminutions, but only that i t should know when this happens (barring the final diminution to zero). There cannot then be errors of the kind I have been considering. J l y arm can move without my knowing it, but I cannot be moving i t without knowing it. My arm can fail to move when I mould

T H E JOURNAL OF PHILOSOPHY

have thought it was moving, but I cannot fail to move my arm without knowing that I have failed; and so I cannot be in a position to say that I am moving my arm and then find out, through noticing that my arm is not moving, that I was wrong. If, as a basic action, I could raise a hat, I would know.

Wittgenstein speaks of someone locating his pain in a table. We might speak of locating a K-like kinesthetic sensation in a hat. So imagine that a feels K in his hat, and then notices that the hat moves. And over time a habit builds up, etc. Or let's not worry about whether K is located one place rather than another, but say only that one has some kind of feeling and then notices a hat move, and that, whenever one has the feeling, the hat moves, and this goes on. Would one ever say that one was moving the hat? If not, wherein does this case differ from the case of the arm? Are not ( a ) , ( b ) , (c) all satisfied ? By opening a gap between ourselves and our arms, the Inductivist closes the gap between arms and hats. True, he might say we never do have the appropriate K. But suppose we did; would he not then have to say that we were moving that hat in just the same sense as he says we move our arm? The Inductivist can give only an inductive answer to the question, What are the limits of positive abnormality? I t is just a matter of fact that we cannot move our hats in the way in which we move our arms. It is only that I have never had the right things to correlate with hat-moves. I have challenged the Inductivist Account, but am I any better able than the Inductivist to show that the limits here are logical and not empirical? Someone may be able to show this, but I am not clever enough for that. True, at a certain level of description, hats are differently constituted than arms are, but this gives us only necessary conditions, for we have the case of negatively abnormal persons whose arms, like ours, are flesh and blood, but who cannot move their arms, and the case of positively abnormal persons whose eyes, like ours, are cells and nerves, but who can dilate their pupils. Nor would it help if there were hats made of flesh and blood and connected through nerves to someone's brain, special hats, so to speak, which were more like arms than hats. I t is essential to our problem that we leave our criteria as they are. Or does this admit a trivial solution to the problem? For it might be our criterion, or part of our criterion, for a hat that a hat can move only when something causes it to move. The limits of positive normality then become logical by courtesy of the appropriate paradigm-case arguments.

WHAT W E CAN DO

Wrongly, perhaps, I am not satisfied with this. A man who could just dilate his eyes, without them being caused to dilate (by him or someone or something) would, by perhaps a similar appeal to paradigm usage, perhaps not be regarded as human. So whether or not a human can dilate his eyes that way admits of a final, logical answer. Our repertoire of basic actions R would specify the essence of being human. But by a comparable criterion, paralytics would not be human; and this is only a bitter piece of irony. If we relax these criteria, however, our problem once more becomes empirical. I don't say I would not be dubious if someone were to claim the hat-moving ability, even if his hat rose and fell: I should suspect fine wires or air jets. But I don't believe the thing is even physically impossible. Physics is concerned only with the way things move when they are caused to move (forgetting just now the First Law). But this, being quite another way of moving an object, would leave physics intact.

Besides, something may be a perfectly ordinary arm-and for all that we should be unable to move it as a basic action. Specifically, this is so when it is not our arm. This lies in the area of what I briefly indicated as the Problem of Other Bodies, which I cannot treat of here. But the fact that I can no more raise your arm than I can feel your feelings, so that privacy extends out into the world of crass objects, should serve to disabuse those who feel that the Problem of Other Minds is due to some logical feature peculiar to minds. The feature, which indeed marks the limits of what we can do, is peculiar rather to selves, and the distinction between a self and everything else cuts, I have suggested, across the distinction of mind and body. If someone could raise a hat as one of his basic actions, the hat would be his in a philosophical, rather than a legal sense: it would be part of him. So not just any hat would do. The theological implications of all of this are rather hazy. Logical impossibility is alleged not to affect an ascription of omnipotence to God, but God could not even do some of the things that are done in the world; e.g., none of the basic actions that are performed by us are His actions. Unless He causes us to do them, in which case not all His actions are basic actions. A believer might nevertheless take some comfort from the fact that "God made the world" and "The world was not caused" are quite compatible if the former describes a basic action. ARTHUR D ANTO COLGMBIAUNIVERSITY