Anscombe's "Intention" and Practical Knowledge

7 Anscombe’s Intention and Practical Knowledge M IC H A E L T HOM P S ON One of the impediments to our comprehension of G. E. M. Anscombe’s Intention is that it is packed full of bits of jargon and peculiar obsessive theoretic tics that were characteristic of Anscombe’s teacher, Ludwig Wittgenstein. For example, the famous dogma about the relation of knowledge to narrowly psychical phenomena like pain or belief and intention itself—the insistence that I shouldn’t be said to know that I am in pain, or to know that I intend this or that, or that I believe one thing or another. Whether my foot hurts or whether I intend to leave Uppsala some day, or whether I think that some day I’ll leave Uppsala—these are supposed to be things I can say. I cannot myself be said to know them “except perhaps as a joke.” (Shooting pains through the leg, you say? Well, you should know . . . ) Insofar as I can make something of this once famous article of the faith, it is a protest against understanding the case in which you come to know that I am in pain, or intend something, from me telling you—it is a protest against understanding this coming to know on your part by a simple reduction the standard case of the spread or movement of knowledge by hearsay, as when I tell you what the weather has been like in Pittsburgh. In the standard case, you then have secondhand knowledge of the weather in Pittsburgh, and can transmit this important knowledge to your friends—who A N S C O M B E ’S I N T E N T I O N A N D P R A C T I C A L K N O W L E D G E  are evidently moved, like you, by the Love of Truth and the Desire to Know. he sensible formulation of Anscombe and Wittgenstein’s point, if they have one, is thus this. In the case where I tell you that I’m in pain or that I think that something is wrong with my liver, you don’t have it secondhand by simple knowledge transfer; nor, on the other hand, is your knowledge of my pain a mere matter of observation exactly. When it comes to the narrowly psychical features of the speaking animals, the humans—well, there is, among them, a way of coming to know certain propositions about one another from one another that is sui generis, or anyway, in need of another account: it is neither testamentary in any ordinary sense, nor observational, nor is it derived by inference from these things, and so on. hat this is what Anscombe at least has in mind is her repeated opposition of the things I can be said to know, and the things I can say. But the idea that the one from whom and about whom we know in this special way sui generis—namely, the pained intending believer—the one who talks and can talk to us about it—the idea that he does not himself know is quite unnecessary to preserve the distinctiveness of our coming to know from him by that USE . . . which expresses the psychical, as Anscombe’s translation of her master’s Investigations puts it. his particular bit of egregious Wittgensteinism is a buzzing noise that distracts the reader, and may dispirit him or her quite quickly. In my own view, though, it is a sort of obnoxious but superficial terminology that one can replace with one’s own, though it takes a little labor. And many features of the book are like this. But it may also be that when we clear away this detritus, we will find that the essential core of Anscombe’s teaching is in some sense dependent on that of the so-called later Wittgenstein. his may or may not be benign, or mostly benign; I cannot say, since I cannot say I understand this Wittgenstein. Of course, I can’t claim to understand Anscombe either. But this much seems plain: If Anscombe’s Intention is a genuine illustration of the mysterious “method” advanced by Wittgenstein in the 1930s and 1940s, still, it is a far better illustration of it than the old man himself ever managed to produce. I have mentioned the business about knowledge of the inner or psychical, because it will bear on my principal topic, namely, Anscombe’s doctrine of practical knowledge, or of my knowledge of what I am doing intentionally, that is my knowledge of what I’m up to.  E S S A Y S O N A N S C O M B E ’S I N T E N T I O N We today sometimes speak of self-knowledge—as opposed, if you like, to knowledge of something that is other, or other-knowledge. And on a certain going using use of the term, it does not simply mean knowledge that would be put forth with the first person. My knowledge that I have a coffee stain on my shirt, or that I have a little scar on my left knee—these knowledges pertain to me. We should call them cases not of self-knowledge, but rather of knowledge of oneself, or rather as knowledge of oneself as other, to put it in the original Aristotelian, a language to which I will return. he possible contents of my proper self-knowledge, on the contemporary conception of it, are things like my pain, my intention, my belief, and so on; not my condition in respect to trouser health and shirt origin. It may be that I am hearing too much in the contemporary use of the expression “self-knowledge.” But if Anscombe could have broken the spell of her master on the points mentioned earlier, she might have employed it to express herself more lucidly. he overarching thesis of Intention was that selfknowledge in this familiar sense extends beyond the inner recesses of the mind, beyond the narrowly psychical, and into the things that I am doing. One form of self-knowledge, she thinks, is practical knowledge. Or, in a different jargon: one form of self-consciousness is self-consciousness in respect of one’s actions. Or again, now in the formula of Sebastian Rödl: one form of knowledge from spontaneity is a form of knowledge of a material process. Her most famous formula, which appears very early in the book, is this: My knowledge of what I’m doing intentionally, or of what I’m up to, is knowledge “without observation.” his is a purely negative specification, though, and one that, as we will see, must be expelled from the final theory; a theory tells us what things are, not what they are not. On the face of it, there are lots of kinds of knowledge that come to us “without observation,” for example, my knowledge of certain familiar facts about the number 17, e.g., that it is prime, that the associated polygon is constructible, that it is 2 + 1. Her point, as I said, is rather that my knowledge of what I’m doing is self-knowledge, alike in this respect to my knowledge of what I think. In alluding to observation, we characterize the intended form of self-knowledge more closely. It is not founded on the knowing and known substances’ entering into a relation of intuition of the other by the one; nor even on a special case of this in one individual substance entering into the relation on both sides, which in the Aristotelian jargon would be intuitive knowledge of oneself as other. A N S C O M B E ’S I N T E N T I O N A N D P R A C T I C A L K N O W L E D G E  In Aristotelian craft, in techné, we have he Craftsman, and that to which he is opposed, really opposed: he Material, the raw materials. In medicine we ourselves give these opposed items the titles Doctor and Patient, still speaking a half-Aristotelian. When the doctor heals himself, the substance comes into both positions and wears both hats. And so the Philosopher says that the physician heals himself, but only as other. So it is with empirical knowledge of oneself; self-knowledge is not like that, no more than poésis is like praxis. But there is more to the contrast with the observation nexus: Anscombe’s practical knowledge is precisely self-knowledge in respect of what might in principle be observed. Or rather, it is knowledge of something that is of such a kind as is regularly observed, and, presumably, necessarily typically observable—namely, what someone is doing, what he’s up to. his is something that is already encoded in her emphasis on the question, “Why are you doing A?” as I will suggest in a moment. hat is, for the agent herself, the intentional action, what she’s up to, is indeed not known “as something alien, from without, through the medium of the senses,” as Frege would put it. But on the other hand, her action is not like Frege’s numbers, since that is precisely how it really is for others: her action comes to them as something “alien,” as pertaining to an alius, another, and “from without,” and “through the medium of the senses.” Frege’s numbers are not objects of experience for anybody; we have no Anschauung of them, as Frege says, they can’t even make a dent; but in the case of someone’s action, some of us might have such an Anschauung. Nor, on the other hand, is the agent’s intentional action quite like the narrowly psychical determinations mentioned earlier. hese too are not given to subject as something alien from without. But they are not properly matters of experience and intuition for outsiders either; they are not contents of simple perceptual knowledge, but enter the outsider’s cognition in other ways. By contrast, Anscombe insists everywhere that an agent’s intentional action is just another sort of process in the world, something perceptible and watchable by others. Indeed her first attempt to isolate the category of intentional action, in a section that might easily be overlooked, depends on this. She says that if I ask you to enter a room with some people in it and to report what’s up with them, the process-descriptions you return with will mostly be descriptions of intentional actions. When it comes to each other, the processdescriptions to which our perceptual power is especially attuned are just these, the intentional actions.  E S S A Y S O N A N S C O M B E ’S I N T E N T I O N I am sitting in a chair writing, and anyone grown to the age of reason in the same world would know this as soon as he saw me, and in general this would be his first description of what I was doing. (emphasis added) Too bad she didn’t say: I am sitting in a chair writing, and anyone grown to the age of reason in the same world would get his mirror neurons triggered. Maybe she would have won the Nobel Prize. he kind of “world” she’s talking about, if it is a human world, is something that essentially and at one stroke decks its bearers out with a pile of possible objects of volition, or descriptions of intentional action, and possible contents of perception. To paraphrase Parmenides, then, we can say that they are the same that are for doing and for perceiving people doing. Let us leave that thought aside. he connection with observation is why Anscombe’s later formulation of her difficulty in comprehending practical knowledge is a bit closer to the core of the matter: Now if there are two ways of knowing here, one of which I call knowledge of one’s intentional action and the other of which I call knowledge by observation of what takes place, then must there not be two objects of knowledge? (51) Here the contrast is best framed with respect to the inner psychical—pain and belief and so on. We will all or mostly all say that, on the part of their bearer, inner psychical things are “known from within,” and not as something alien and from without; they are contents of her self-knowledge or selfconsciousness in standard cases; but, for others, on the other hand, as I just suggested, they are not exactly directly intuitable, potentially observable, features, like hair color or posture. How can something that is captured in an observational concept, as we might put it— something that is such as to be observed, or known from without through the impact of the thing known—in this case something that is happening or going on, a material process—how can such a thing be, at the same time and intrinsically, such as to be “known from within”? It is difficult to combine the two models of knowledge; the idea of self-knowledge seems to draw the thing known into the agent and away from others. And the picture of things knowable from an intuition of an external substance seems to put them at a distance from the cognition even of that substance, if it is a knower. He must look down to see the gravy on his jacket or the coffee stain on his pants. A N S C O M B E ’S I N T E N T I O N A N D P R A C T I C A L K N O W L E D G E  Now it is, I think, increasingly common for philosophers to attempt to find a place for self-knowledge in respect not just of thoughts and intentions and pains and so on, but in respect of action. But we should see that this was her program all along. Indeed, in the end we may say that her definition of intentional action is in terms of a form of self-knowledge, or self-consciousness: what’s up with me is an intentional action precisely where it is a content of specifically practical knowledge, otherwise not. his is the real definition. It is a nominal definition of intentional action that is propounded earlier on in the complicated account in terms of what is known, and non-observationally known, and so forth: a mere fi xing of the extension of the term, to put it crudely; something that is to hold the topic in place as we seek to divine this real definition. A principal difficulty in comprehending Anscombe’s teaching on this point arises, I think, from a distorted emphasis in the choice of examples in the literature and also on unnoticed presuppositions of the use of nominalizations like “action” and “event,” which of course Anscombe herself uses freely. Terms like these make the construction of grand philosophical generalizations easy and smooth. But they bring along a danger that we will lose contact with the more fundamental judgments in respect of which we are generalizing, the ones that would exemplify our claims. hus we will fail to follow the fundamental principle of Fregeanism, of which Anscombism is a subordinate sect, in always keeping before our minds a complete thought. But let us consider some of the complete thoughts that might be at issue. One striking difference between the approaches Anscombe and Davidson respectively take to this material shows up in an apparently subtle feature of their illustrations. he typical Davidsonian example is in the past tense, the typical Anscombian example is in the present. here are exceptions to this rule, in either author, but consider a few familiars from Davidson’s great essays on action: I flipped the switch because I wanted to turn on the light. Cass walked to the store. Smith coughed. Alice broke the mirror. Brutus kissed Caesar. I flew my spaceship to the morning star. Amundson flew to the North Pole.  E S S A Y S O N A N S C O M B E ’S I N T E N T I O N And everyone’s favorite: Strange goings on! Jones did it slowly, deliberately, in the bathroom, with a knife, at midnight. What he did was butter a piece of toast. Flipped, Walked, Broke, Kissed, Flew, Buttered, did. It is not just that these propositions all concern the past. hey all exhibit what linguists call perfective aspect. What is represented is represented as a completed whole that falls under a certain heading, whether as switch-flipping, kissing, or buttering toast. he metaphor of completion and wholeness is what is intended by the word perfect or perfective. If, for each verb phrase, we introduce a common noun by nominalization, and speak of switch-flippings, kissings, and so on, then the complete thoughts Davidson puts before us, the “he did it” propositions, will all entail the existence of something falling under such a noun. hat is, they will all entail the existence of an event of a certain type. And this is indeed Davidson’s theory of what makes the propositions true. But the class of complete thoughts to consider contains some quite different forms. his is something that E. J. Lemmon pointed out in his comment on the original essay on Action Sentences, and which Davidson pooh-poohed as false hairsplitting in his response. If I may be permitted to superimpose phases of human history, consider that it might have been like this: at the fateful moment when Brutus kissed Caesar, or when the kissing occurred, Cass was walking to the store, moseying along with thoughts of the tobacco he was proposing to purchase. But then the sight of this same-sex Roman kissing so alarmed his tender modern sensibilities that his heart failed for good there and then. Or, to superimpose again, it might be that at the moment when I flipped the switch and there was a switch-flipping, Jones was buttering the toast. he wiring I triggered was faulty, and so I thereby electrocuted the poor devil, who was leaning against a copper water pipe. Toast and butter were only a halfcentimeter apart when the knife and toast fell to the ground, and with them Jones, quite dead. Let the stories be placed in time definitely past. In that case, we will have to say that Cass did not walk to the store in fact; Jones precisely did not butter the toast. Moreover, there was in neither case an event of toast-buttering or to-the-store-walking. Event is eventus; it is what came out, what turned out, what happened; X eventus est, or X est eventus V, means: it happened. But a A N S C O M B E ’S I N T E N T I O N A N D P R A C T I C A L K N O W L E D G E  walk to the store and a toast-buttering are not among the things that happened. he kiss and the flip happened; there were such events. It is true that in Cass’s case, a walk to the scene of the kiss happened; and in Jones’s case, a beginning of buttering occurred. If the question is, “What happened on that fateful day?; render unto me a syllabus of Events, or Eventa,” we will get the answer that Brutus kissed Caesar, and that Cass walked to the scene of it, but of course not that Cass walked to the store. If the question is, “What happened on that fateful night,” we will get the answer, I flipped the switch, and Jones started to butter the toast, but not Jones buttered the toast. Nevertheless, the verb phrases “walk to the store” and “butter a piece of toast” have a certain foothold in reality in the two cases. We express this by using imperfective aspect, i.e., as English speakers, with the past progressive. here is thus something that Davidson’s doctrine of events or of things that happened is missing, namely, not to put too fine a point on it, the things that didn’t happen. hat is, he forgets about the things that didn’t happen, but were happening. For this reason, I will suggest, he is not capable of properly grasping the nature of what really did happen, that is, of events, especially where these are completed intentional actions. he whole system is formally structured to repel this content. By contrast, Anscombe’s illustrations are unrelentingly present, and for this reason always imperfective in character. Since she is speaking English she expresses the propositions or states of affair by means of the present progressive. here are of course exceptions, but we typically come upon Anscombe’s agents in medias res, so to speak, or in flagrante delicto: here’s a couple of examples from page 35: Why are you setting up a camera on that pavement?—Marilyn Monroe is going to pass by. Why are you crossing the road?— I am going to look in that shop window . . . (or more obscurely: Eclipse in July.) One agent is setting up a camera, the other is crossing the road—we say, speaking progressively. If philosophers perchance arrange that the H-bomb goes off just now, these propositions will not be falsified. But then it will never be true to say that the one did set up the camera, or the other did cross the road. No such events will have occurred. I will be laboring the imperfective element in the complete thoughts that we Fregeans must keep ever before us, and also the contrast with Davidson.  E S S A Y S O N A N S C O M B E ’S I N T E N T I O N But we should bring out immediately that Anscombe’s standard examples are not only progressive and imperfective and in medias res, but typically first person in character, or rather, as we might say, first and second person. For example, she begins by describing the man replenishing the house water supply in the third person, “He is moving his hand up and down . . . He is earning his wages. . . . ,” etc. But soon enough she goes Anscombe on us, and we are asking him, “Why are you moving your hand up and down?” and he is replying, “I’m pumping.” Whereupon we ask him, “Why are you pumping?” and he says, “I’m pumping the water supply for the house.” Formally speaking, this interaction is in some sense the fundamental scene Anscombe is working with throughout the book. Although it is of course perfectly contingent whether someone’s intentional action ever gets brought up in such an exchange, all of the crucial elements of her thought emerge in it. One human being comes upon another and perceives her doing something, or observes her at it—he is setting up a camera, she is crossing a road, he is moving a pump handle. he enquirer knows by perception, by an intuition or Anschauung of the other as other, by observation of her, that the agent is setting up a camera or is crossing a road. Notice again that this is exactly what is given to the senses of the suitably cultivated observer, even though, of course, a prudent philosopher will arrange that the hydrogen bomb is just now going off; that does not make the perception anything but veridical. In the first instance this empirical knowledge is demonstrative and third person in character. But the difference between the observer’s observational thought in the case at hand, and his thought in other cases of observed things happening, like trees falling down, comes next. he observer moves into what we might call a cognitive relation with the agent herself and asks her why she’s doing it. He does not do this with falling trees. he mark of the cognitive relation is the use of the second person, “Why are you doing A?” hat he addresses himself to the observed individual substance is already a clue that something is different. We should notice that the answer the agent gives will itself frequently supply a first-person–present-progressive formula. he agent might say, “Well, I’m doing B.” hat is, he might come out with the formula of the type that was contained in the question, but something that is by chance not so readily observed, at least not by the enquirer; it is as it happens not given in her intuition of the agent. Of course not all answers are in the first person of the present progressive: “I’m doing B.” Some are like: “He killed my brother,” or “I like it, it’s fun,” or “It befits a Nazi,” or “No particular reason.” But even some that A N S C O M B E ’S I N T E N T I O N A N D P R A C T I C A L K N O W L E D G E  aren’t could be put this way, for example some that take the form, “To do B.” Of the one who is doing A, and is asked “Why?” she says (pp. 38–39—I substitute my preferred letters, for her X and Y): Let the answer contain a further description B, then sometimes it is correct to say not merely: the man is doing A but also: ‘the man is doing B’—if that is, nothing falsifying the statement ‘He is doing B’ is observed. E.g., Why are you pumping?—‘To replenish the water supply’. If this was the answer then we can say: He is replenishing the water supply; unless indeed he is not. [Much later it emerges that maybe the pipe is broken and the water is running out.] h is will appear a tautologous pronouncement, but there is more to it. If after his saying ‘To replenish the water supply’ we can say ’He is replenishing the water supply,’ then this would, in ordinary circumstances, be enough to characterize that as an intentional action. . . . Now that is to say, as we have already determined, that the same question ’Why?’ will have application to this action in its turn. Of course, in that case, if we ask him why he’s pumping he might have just said, “I’m replenishing.” And again this use of the progressive would be marked as the sort to which one can annex the question “Why?” What I observe, on the one hand, and what I am told on the other, are each marked out in the same way, as inviting the question “Why?” in her special employment of it. his nexus of descriptions, of doing A with doing B and so forth, or rather the nexus that is here given as founding it, the one that might, if you like, be expressed with the infinitival formula He’s doing A in order to do B, or He’s doing A to do B, but in many others as well—this is something that Anscombe has just argued is essential to the constitution of her material. his is in the all-important sections 19 and 20. Section 20 begins with the question Would intentional actions still have the characteristic ‘intentional’, if there were no such thing as expression of intention for the future, or as further intention in acting? he answer is supposed to be “No.”  E S S A Y S O N A N S C O M B E ’S I N T E N T I O N Of course, not everything I do intentionally is done in order to do something else; sometimes, as we saw, it’s for no particular reason, or because I like it, or because he killed my brother, and so on. Nevertheless, apart from the possibility of this sort of connection— of doing A with doing B, and of doing B with doing C and so on—there could be no intentional doing of anything, no Will and no articulate discourse that latches onto it, as the elements in the fundamental scene do. Consider those rational trees I read about in an essay by Colin McGinn a million years ago, rational, sighted trees who observe sheep moving across their visual fields, and, as it happens, round the hill. hey can’t go around to the other side, they can’t even move their eyes, they peer firmly ahead. Isn’t it obvious that they could come to know that the sheep go around to the other side of the hill? (He was giving a counter-example to Michael Dummett’s philosophy, securing his position at the head of a then very happening lynch mob.) Anyway, let’s expand on the example, and suppose that once in a while, the rational trees just spit, and just for the hell of it. hat’s all they do and all they can do. hat’s impossible, she argues: “Some chains must begin.” Piecing these remarks together with the one about the man who keeps answering to do B and then to do C and so on, we can see that Anscombe’s proposition, “Some chains must begin,” entails the proposition, “Some practical progressives must stack.”— or in her jargon, “An A–D order must be possible.” his thought is most easily understood if we take her to be characterizing the Will as a capacity or faculty or power, rather than hunting for something to characterize each case of intentional action. he Will is the power to be subject of intentional actions, or, if you like, apt for process ascriptions that can be queried with her question “Why?” Not every exercise of the Will is an intentional action; for example, an intention to do something and a certain sort of wanting to do it are exercises of the same power, and might not accompany any action. Similarly, not every exercise that does take the shape of an intentional action, an action of doing A, say, need be accompanied with other simultaneous exercises that take the shapes of doing B, doing C and doing D, where these are fitted together in what she calls the A–D order. But if this is not possible, there is no Will, no action, no agency. I propose to piece this point together with the doctrine of practical knowledge. Now, when Anscombe speaks of practical knowledge, of knowledge without observation of my intentional action, there is a danger of misunderstand- A N S C O M B E ’S I N T E N T I O N A N D P R A C T I C A L K N O W L E D G E  ing. I said that the word “event” taken as applying to individuals, properly applies to what happened, and not what was merely happening; so the word “action” or “deed,” as applied to individual tokens, properly applies to what was actually done, not to what the agent was merely doing; the etymology is similar: ago, agere, egi, actum. Our new-modeled Cass did not perform any such ACTION as walking to the store; our new-modeled Jones did not perform any ACTION of toast-buttering. Nihil actum est. Nevertheless, each had the knowledge Anscombe is talking about. he late Cass knew that he was walking to the store; the late Jones knew that he was buttering toast. he content of Anscombe’s practical knowledge is progressive, imperfective, in medias res. Its character as knowledge is not affected when the hydrogen bomb goes off and most of what the agent is doing never gets done. In this respect it is to be compared with the more familiar forms of self-knowledge. I do not have self-knowledge in respect of another person’s pain or belief; and I do not have self-knowledge of my own pains and beliefs of another day or week. An affirmation of self-knowledge in the sense in question will be, as is said in a famous passage of Wittgenstein, the first person of the present tense. It is here that we find the so-called expression of the psychical. So it is with specifically practical knowledge, Anscombe’s practical selfknowledge. Only here, as with all event-description or process-description, the present tense entails not-yet-completeness and imperfective aspect, and thus (in English) a present-progressive formulation. My thoughts and pains are matters for self-consciousness, only as long as I have them, as long as they are present. It is the same with practical knowledge, and thus since the present in this case must be imperfective, there is practical knowledge only when the thing is precisely NOT done, not PAST; there is more to come, something is missing, and the H-bomb may hit before it does. My so-called knowledge of my intentional action in truth exists only and precisely when there is no action, but only something I am doing. Now, this might be enough to bring out the defect in a celebrated argument of Davidson’s, which unfortunately killed off the sublime topic of practical knowledge for a few decades, until Velleman, in his way, revived it. Davidson considers someone making several carbon copies of a document as he signs it. He needs ten, one for the bank, two for his lawyers, two for the other guy’s lawyers, one for the county, and so on. He starts to write his name on the top. He doesn’t know, as he writes that first bit, whether he is making an impression on the last copy. But suppose he is and keeps it up through all  E S S A Y S O N A N S C O M B E ’S I N T E N T I O N the letters of his name. In the end he has signed all ten documents, and did it intentionally. But he didn’t know it. Now there are some real estate closings that are like that. But the more ordinary case is like this: you write on the top sheet, trying to make a good impression to get through all the carbon, then look to see if your impression made it through to all of them. If it did, you stop. If it didn’t, you remove the last properly impressed sheet and begin again. If necessary, you repeat. Even the man who has to go through five stages is all along, from the first feeble impression, making ten copies of the document, and he knows it, all along. he one who doesn’t know it, Davidson’s man, must be under some strange mafia threat: he gets one chance, no checking, and he’s dead if he doesn’t manage it. Well, for him, the making of the inscription is like the buying of a lottery ticket. You can say he made ten copies intentionally if you like, but it will not be an illustration of the topic of Anscombe’s book, any more than lottery-winning is when you bought the ticket with that aim. he failure to perceive this comes from failing to meditate on the progressive, on the fact that the content of practical knowledge is something present, and thus something of which more is to come, perhaps including several attempts at it or some part of it.