Wittgenstein and Anscombe’s Intention

Rachael Wiseman has argued that we cannot make sense of G.E.M. Anscombe’s Intention unless we recognise that it is an “exemplification of [Wittgenstein’s] grammatical investigation”. While Wiseman is alive to the Wittgensteinian nature of Anscombe’s method , and to her deep Wittgensteinian sympathies, she is not preoccupied with the question of influence. This is the question I am concerned with in the current paper. I argue that in focusing on the concept of intention, Anscombe was homing in on a pivotal co ncept in Wittgenstein’s later philosophy, and that most of the basic elements of her account were being worked out by Wittgenstein during the period when she was his pupil. However, as Anscombe worked through Wittgenstein’s idea’s afresh, in her own more s ystematic and analytic mode of philosophical investigation, she often cast ideas that were sometimes merely nascent in Wittgenstein’s work in a new light. Moreover, some of her most seminal ideas had their origin in concerns which she did not share with Wittgenstein and were entirely original to her.


1.
Anscombe's Intention is generally acknowledged to have been deeply influenced by Wittgenstein.Although Anscombe refers to Wittgenstein twelve times in the text, she herself does little to make the nature of her debt to Wittgenstein explicit.John Haldane sees this as typical of Anscombe's writings in general.He writes that they "usually give the sense of starting afresh each time in response to some newly discovered, or freshly returned to puzzle, not writing to a plan but setting down the movement of thought and only occasionally referring to contemporary philosophers, even more rarely quoting particular texts."(Haldane,  2016:378)  This is certainly true of Intention.And in at least some of its movements, Anscombe's thought is driven by interests that are remote from those she shared with Wittgenstein, namely her Catholicism and her interest in making philosophy of psychology the basis for a study of ethics which focuses on "considering the concept of 'virtue'" (Anscombe,  1958:15).Nonetheless, Rachael Wiseman has argued persuasively that we cannot begin to make sense of "Anscombe's seemingly meandering reflections" (Wiseman 2016:55) in Intention unless we recognise the fundamentally Wittgensteinian conception Anscombe has of her task.In her Routledge Guidebook, Wiseman claims that Intention "is an exemplification of the method of grammatical investigation" (Wiseman,  2016:16).
Wiseman sees this grammatical investigation as working against a number of temptations.For example, the temptation to think that the concept of intention refers to an inner mental state, which expressions of intentions give expression to; or to think that the difference between an intentional and an unintentional action is some extra property of intentional actions, which philosophers must identify.It works against these temptations and clarifies the real nature of the concept of intention by "examin [ing] and describ[ing] a human life shaped by the concept of intention" (Wiseman, 2016:4).It thereby uncovers "an order that belongs to the concept of intention" (Wiseman, 2016:4), an order that characterizes our mode of description whenever we employ intentiondependent concepts in accounts of what oneself or another is doing.
Central to Wiseman's Wittgensteinian reading of Intention is Anscombe's eschewal of any explanatory ambitions.According to Wiseman, Anscombe's aim is purely one of conceptual clarification: to expose the order that is there in the enormously complex tacit conventions we master in coming to understand this region of our language.Anscombe sets out to uncover the order that is there to be seen in our ordinary practice of saying what one is doing oneself or what another is doing, of expressing and attributing intentions and motives, of giving grounds and offering excuses, and so on.It is important to the kind of investigation it is that we are being invited to investigate our own practice, as it were, from within -a practice of which we ourselves are the practitioners.Anscombe, like Wittgenstein, "is marshalling recollections for a particular purpose" (PI §127). 1 The concepts and distinctions we're concerned with are ones we have already fully grasped -even a child has mastered them -but their nature and their grounding are obscure to us.Anscombe aims to bring order to the chaos and show through a process of self-recognition that the ground of the distinctions that puzzle us lie open to view in the complicated details of our everyday life with language.It is, therefore, important for the nature of Anscombe's enquiry that it starts from the point at which she can "note the fact" (I: 8) that we can simply look at a man and say what he is doing. 2 And what we say will generally coincide with what he could say he was doing, usually without reflection, and "certainly without adverting to observation" (I:8).However, about the character of the concept of intention -which we unreflectively employ -she believes we are "in fact pretty much in the dark" (I:1).
Anscombe's investigation of the concept of intention is necessarily complex, as she sets out to do justice to the nuances and indeterminacies that characterise our ordinary language-game.Her enquiry does not, however, range freely over this chaos.It is given shape by her initial proposed account of intentional actions -"'Intentional actions are ones to which a certain sense of the question 'Why?' has application'" (I:11)a sense which is gradually clarified and made more perspicuous as she explores the terrain of our language-game, by examining a range of cases in which we would or would not describe someone as having done something intentionally, or as having done something with some specific motive or intention.
In the second half of the book, Anscombe approaches the order she has uncovered from a new perspective: that of Aristotle's account of practical reason.However, she is clear that the order which she sees as definitive of the concept of intention is one that she has already fully revealed by means of her special sense of the question "Why?".She writes: The interest of [Aristotle's] account is that it describes an order which is there whenever actions are done with intentions, the same order as I arrived at in discussing what 'the intentional action' was, when the man was pumping water.3I did not realise the identity until I had reached my results; for the starting points for my enquiry were different from Aristotle's, as is natural for someone writing in a different time.(I:80)   The crucial point about the order Anscombe has uncovered, which Wiseman underlines in describing the work as a grammatical investigation, is that it belongs to "a form of description of events" (I:84).However, while Wiseman is alive to the Wittgensteinian nature of Anscombe's method, and to the deep Wittgensteinian sympathies that inform her account of intention, she is not preoccupied with the question of influence.It is this question I'm concerned with in this paper.I don't mean thereby to underestimate the extent to which Anscombe did, as Haldane suggests, start afresh and think the puzzles through anew.Even where ideas have a clear origin in Wittgenstein's work, Anscombe casts them in a new light, as she works them into her own distinctive, more systematic and more analytic mode of philosophical investigation.Her reading of Wittgenstein is, moreover, influenced by her knowledge of the work of Aristotle and Aquinas.There is no doubt that in investigating the concept of intention, Anscombe focused on a key concept in Wittgenstein's later philosophy and that she derived fundamental insights from his work.But she also made significant progress in thinking the puzzles through anew, which holds out the hope her work might illuminate Wittgenstein's, as well as vice versa.
Some of the most striking and seminal ideas in Intention, 4 even if they were made possible by her Wittgensteinian approach, were driven by concerns that Anscombe doesn't share with Wittgenstein, and are entirely original to her.And it may be, as we shall see, that these concerns also lead her to put forward views that are prima facie in tension with the naturalistic outlook that characterises Wittgenstein's conception of the task of description.One cannot but be struck, for example, that the question with which Intention begins -the question of whether the concept of intention is univocal across its use in expressions of intention for the future, descriptions of intentional action, and statements of the intention with which an action is done -is not typically Wittgensteinian.Even if Wiseman is correct in arguing that in responding to the question Anscombe employs a Wittgensteinian method of grammatical investigation, the question itself leads her to draw conclusions about the essential nature of the concept of intention, which are unlike anything one finds in Wittgenstein's philosophy and are, I shall argue, closely tied to Anscombe's conception of the human soul. 5

2.
Anscombe's Intention is generally credited with reviving philosophy's interest in the philosophy of action after centuries of neglect.It has not always been noted that she had implicitly identified a concept that is not only central to Wittgenstein's later philosophy, namely the concept of the deed, but one that is also an important focus of his philosophical investigations.Its centrality is clearly expressed in his quoting with obvious approval the lines from Goethe's Faust: "…and write with confidence 'In the beginning was the deed.'"'(On Certainty §402).More importantly, it is manifest in the form his opposition takes to the Augustinian picture of meaning in the opening paragraph of PI.The scene Wittgenstein describes shows a speaker engaged in an activity with a specified end and operating with words within that wider context.The crucial point, Wittgenstein emphasises, is that the speaker "acts" as he describes.Likewise, in the language of the builders in PI §2, we see speakers engaged in a recognisable activity and giving orders and responding to them within that context.They are seen to be acting in ways which we recognise straight-off as intentional in much the way Anscombe notes as a fact. 5Wiseman contrasts her grammatical reading of Anscombe, and thus her approach to answering the question she raises in §1, with how Intentions has traditionally been understood in the literature on philosophy of action.She calls the traditional or standard understanding of the question of how to give a unified account of the three aspects of intention Anscombe identifies "Anscombe's Question".To answer Anscombe's Question would be to describe the connectionscausal, explanatory and rationalthat exist between the phenomenathe mental statesthat are picked out by the use of the concept of intention in each of its three guises.Wiseman argues that "Anscombe herself neither poses nor answers this question" (Wiseman,  2016:49); those who seek for answers to the question in the pages of Intention will render it "at worst incoherent, at best suggestive but incomplete" (Wiseman, 2016:52-3).Wiseman sums up the contrast between the standard and the grammatical reading as follows: For a philosopher engaged in answering Anscombe's Question, the question Anscombe is sometimes supposed to be addressing, the answer is: we should look more closely at the phenomena to which 'intention' refers, the alleged states of mind and bodily movements.For a philosopher like Wittgensteinand Anscombewe should look not at the phenomena but at the linguistic practices and activities in which it comes to be that 'intention' refers to those phenomena.(Wiseman, 2016:60)  McGinn 6 When Wittgenstein lists the variety of language-games in PI §23, it is significant that he does not confine himself to what J.L. Austin called illocutionary acts (giving orders, describing, reporting, requesting,  thanking, cursing, greeting, praying).He goes beyond the illocutionary acts speakers perform to include some of the different things, as Austin would put it, we use language for: making jokes, guessing riddles, making up stories, and singing.He also includes activities in which the speaker's use of language is an element in a wider undertaking: constructing an object from a description (or drawing), forming and testing a hypothesis, acting in a play.Austin believed Wittgenstein both blurred the distinction between illocutionary and perlocutionary force and mixed in other, irrelevant aspects of language-use. 6But this is to misunderstand the nature of both Wittgenstein's philosophical aims and his philosophical method.Wittgenstein is not concerned with the sort of systematic classification of speech acts that interested Austin, but in a task of conceptual clarification, in particular of the concepts of meaning and understanding, which he undertakes by looking at language in use.PI §23 should be read in that light.
If we take the language-games of PI § §1 and 2 as models for the way Wittgenstein, with the aim of conceptual clarification, intends us to undertake the investigation of our own immensely complicated languagegames, it is clear he means us to look at language where it is being used by speakers, on particular occasions, in the context of everyday activities.It is important, therefore, that the list of language-games in PI §23 points to a speaker's use of language as it meshes with life in a variety of ways.Wittgenstein thereby indicates the nature of the terrain at which his investigations are directed: our life with language; the practice of employing the expressions of our language within our everyday lives.Clearly, this makes the topic of the actions of speakers the focus of enquiry from the outset in PI.His investigation of the topics of meaning, understanding and rule-following should thus be seen as making a vital contribution to the picture of intentional action that emerges in Wittgenstein's later work.It is not only that these remarks can be seen as part of an anti-Cartesian project to which Anscombe's Intention is also a contribution.Wittgenstein's reflections on the concepts of understanding and rule-following lead to a clarification of these concepts which, given his approach to the speaker as an agent who operates with words, is by the same stroke a clarification of the concept of intention. 7et's start with his investigation of the use of the words "Now I know!", "Now I can do it!",and "Now I understand!", which begins at PI §151.These exclamations are akin to expressions of intentions for the future insofar as they share a predictive aspect: the speaker is asserting that, if the occasion arises, he will go on correctly. 8The aim is to clarify the way these expressions function by becoming clearer about their antecedents -what leads up to them -about the circumstances in which their use is justified, those which would show it is mistaken, and so on.Wittgenstein imagines an example (PI §151) in which A writes down a number series -1, 5, 11, 19, 29 -and B has to try to find the rule of the series.He imagines that when A gets to 29, B says, "Now I can go on."The first question is: What actually happened here?
Wittgenstein suggests various things might have happened in the lead up to B's saying these words.Various formulae might occur to him, when A reaches 19, B thinks of the formula a n =n 2 +n-1 and 29 confirms it.Or perhaps B doesn't think of formulae; he just watches A with certain feelings of tension while all sorts of vague thoughts go through his head; then he asks himself, "What is the series of differences?"-4, 6, 8, 10and he says, "Now I can go on!"Or perhaps B watches A and says, "Yes, 7 These remarks are usually read in the context of the question: what settles what counts as the correct application of a rule in the future (as the fulfilment of the intention a speaker has expressed, e.g., to use a word with a particular meaning).In the course of his investigation of this question, Wittgenstein repeatedly pivots away from thinking of meaning and understanding as states of mind towards presenting the speaker as an agent who has mastered the techniques of an established linguistic practice.It is the picture of agency that thereby emerges in this investigation of our ordinary language-game that I want to focus on in this paper. 8Wittgenstein remarks on the predictive aspect of expressions of intention, for example, in PI § §629-32.Anscombe takes up Wittgenstein's suggestion that expressions of intention are a certain kind of prediction: "I first, in considering expressions of intention, said that they were predictions justified, if at all, by a reason for acting, as opposed to a reason for thinking them true" (I:15).Treating expressions of intention as a form of prediction is in part intended to counter the idea that they are reports of occurrent, internal mental states.Wittgenstein's remarks on "Now I can do it" are also intended to counter the idea that these words describe a mental state which makes its appearance "in an instant" (PI §139); drawing attention to its predictive aspect fits with his aims.
I know that series" -and he goes on with it, as we would have done with the series 1, 3, 5. Or perhaps B says nothing at all, but simply has the feeling, "That's easy", and just goes on with the series.
Wittgenstein describes all these possible scenarios and asks: "are the processes which I have described here understanding?" (PI §152).Even in the case where the formula occurs to B, can we say that is the understanding?"For it is perfectly conceivable that the formula should occur to him and that he should nevertheless not understand."(PI §152) There seems to be something missing.We feel, Wittgenstein suggests, "'He understands' must have more to it than: the formula occurs to him" (PI §152).Similarly, for all the other scenarios: they merely describe the "more of less characteristic concomitant processes or manifestations of understanding," (PI § 152) but they are not the understanding itself.
It is at this point that Wittgenstein makes the crucial move away from looking for something behind or beside the processes which are characteristic of understanding and attends to the "particular circumstances, which warrant my saying that I can go on -if the formula occurs to me" (PI §154).It is not enough that the formula comes to mind: it could be an accident and not be a mark of understanding.There must be a connection between the formula's coming to mind and my going on to continue the series.But isn't that a connection that is established through experience?And wouldn't that mean I would have had to observe the connection empirically to justify my saying, "Now I can go on", when the formula occurs to me?But this is to mistake the kind of prediction "Now I can go on" is.B is not making an estimation of what will happen on the basis of empirical evidence.We are describing a different language-game herean entirely different pattern of use -one that is manifest when we are describing speakers who are giving expression to what they can do.This language-game is grounded in the abilities that the speaker has acquired, which put B in a position to say, straight-off, not on the basis of observation, as soon as the formula occurs to him, "Now I can go on".The circumstances which connect the formula's coming to mind and B's going on correctly do not figure as evidence for his prediction, but he "had a right (my emphasis) to say the words…" "…under certain circumstances.For example, if he had learnt algebra, had used such formulae before" (PI §179).That is, if he had acquired the background abilities on which a correct application of the formula depends; abilities which ground, if the occasion arises, his going on correctly, in the way he predicts, and which make the formula's coming to mind a mark of understanding.
As we saw, not every case of coming to understand the series involved B's thinking of a formula.It might be that nothing at all went through B's mind, he just had a feeling of sureness that he could go on and then did go on developing the series correctly."And in this case too we should say -in certain circumstances -that he did know how to go on."(PI §179) These circumstances include B's having been trained in mathematics, or his having developed a capacity for recognising mathematical patterns, and so on.A speaker's background training and know-how set the scene for his use of the expression, "Now I know how to go on."The speaker's confidence in his prediction of what he will do if the occasion arises is not based on evidence, but in his self-conscious awareness of his own capacities and ability to act as he intends.We judge whether his expression is correct on the basis of a complex set of tacit conventions to do with our everyday practices and institutions: training, the past history of the speaker, and so on.We do not generally wait on the confirmation of a speaker's actually going on correctly.And, as Wittgenstein notes, if we find that when a speaker tries to go on, he hesitates and can't do it, we would not necessarily say he was wrong to say he could go on, when he said it: "Clearly, we shall say different things in different cases" (PI §181)."The criteria which we accept for… 'being able to,' 'understanding,' are much more complicated than might appear at first sight" (PI §182).
It is through such steps as these that Wittgenstein moves to clarify the idea of the agency of speakers.He has uncovered a pattern in our use of words that is there when we speak of coming to understand -in one's own case and in the case of another -that depends upon immersion in our practices and the building up of a repertoire of linguistic techniques and abilities.It has introduced a number of significant ideas.First of all, the idea that there is a fundamental 1 st -person/3 rd -person asymmetry that is key to the pattern of use definitive of agency.The speaker does not say, "Now I can do it", "Now I understand!", on the basis of observation; it is something he says straight-off, a form of expression of the dawn of the capacity he has acquired.Wittgenstein says it would be better to "call them a 'signal'" (PI §180), "an instinctive sound, a glad start" (PI §323).Our 3 rdperson criteria on the basis of which we judge whether the words are correctly used are complex and involved, but what forms the background to their use is an existing linguistic practice and a speaker's manifest possession of abilities to participate in it, which both he, and we, confidently expect to continue into the future.

3.
In his remarks on rule-following, Wittgenstein further underlines two ideas which are central for the picture of agency that emerges in the later philosophy.First of all, there is the idea that what intentional action an agent is described as performing on a particular occasion depends upon the existence of a context of an established practice.For example, an agent is described as going by a signpost, playing a game of chess, or making a report only if there exists a practice of going by signposts, of playing chess, of speaking a language.He sums up the point at PI §199: To follow a rule, to make a report, to give and order, to play a game of chess, are customs (usages, institutions).
He does little to unpack the notions of "customs (usages, institutions)", but they clearly imply agreed, regular, stable and established ways of acting, not only in the sense of the way speakers use words, but in the way their use of words meshes with their lives.Wittgenstein also makes clear that it is the existence of this context of an established practice -of customs and institutions -that makes the connection between a speaker's expressed intentions and the thing intended.It is, for example, the existence of the everyday practice of playing chess, and my own participation in it, that makes it chess that I intend to play when I say, "Let's play chess": Where is the connection effected between the sense of the words "Let's play a game of chess" and all the rules of the game?-Well, in the list of the rules of the game, in the teaching of it, in the everyday practice of playing.(PI  §  197)   An intention is embedded in a setting, in human customs and institutions.If the technique of the game of chess did not exist, I could not intend to play a game of chess.To the extent that I do intend the construction of an English sentence in advance, that is made possible by the fact that I can speak English.(PI §337) This raises one of the central questions of the remarks on rulefollowing: how are the rules of a game, or the rules for the use of words, present to the mind of someone who intends to play chess or who understands a language.Wittgenstein raises the question at PI §205: McGinn 11 [I]sn't chess defined by its rules?And how are these rules present in the mind of someone who intends to play chess?
The whole thrust of the remarks on rule-following has been to get us to see that it is not a question of what is "present in the mind".As he says in PI §199: To understand a sentence means to understand a language.To understand a language means to have mastered a technique.
We come back to the idea of a certain pattern in the life of the speaker.What is in question here is that the words -"Let's play chess", "I meant he should say 1002 after 1000" -are spoken by a speaker who has mastered certain techniques, in the context of a practice which he points to if he wants to specify which game he intends to play, which rule he meant by "Add 2".The practice depends upon agents' acquiring the ability to act confidently and independently and autonomously -without any further guidance -as members of a practice of playing games, adding, measuring, inferring, describing, acting, forming and testing hypothesesall the things Wittgenstein lists in PI §23.
What has also become clear is that it is not anything that accompanies an act of following a rule that constitutes it as a case of rule-following, and makes it an event we can, for example, describe as playing chess, adding 57 and 68, or developing the series +2.It just is a fact about us that, after a certain training, we do, for the most part, go on independently in a way that sustains our practices.We may, in certain circumstances, give justifications for how we apply a particular rule.But in the end, as Wittgenstein observed in PI §1, "[e]xplanations come to an end."And then we come back to the action of an autonomous agent who applies the techniques he has been trained in, by acting, without guidance, in the way that counts, in our practice, as "following the rule".If he has mastered the rules of chess, or the relevant mathematical techniques, then he has a form of knowing how which, when it is exercised on particular occasions, is an expression of what Anscombe calls practical knowledge. 9he participant in our practice, who has mastered the relevant techniques, is in a position to say when he is engaged in doing these things, not on the basis of observation, that what he is doing is playing chess, adding 57 and 68, or developing the series +2.It his background abilities that put him in a position to say these things, straight off, not on the basis of observation.It is the practice that provides the context in which those abilities have been developed, that shows what counts as their correct exercise; and it is the practice that the agent acknowledges, in a certain sense, as sovereign and understands himself to be participating in.What is important for our purposes is to see that what has emerged in the course of these remarks is a picture of agency which emphasises 1 st -person/3 rdperson asymmetry, and the speaker's self-conscious awareness of, and ability to say, what he is doing, not on the basis of observation, grounded in background of abilities that constitute it as a form of what Anscombe calls practical knowledge.

4.
Given the place that the concept of agency occupies in the later philosophy, the grammar that is definitive of intention has, as we've just seen, already begun to emerge before Wittgenstein turns his attention to the topic of voluntary action.There are remarks on the concept of intention scattered throughout PI, but there is a more focused discussion of it, following remarks on willing and voluntary action, in the 600s.Intention and voluntary action continue to be a topic of remarks following the completion of Part 1 of PI.Although these concepts never receive the kind of systematic or analytic treatment that Anscombe gives them in Intention, nevertheless it is possible to discern some of the key elements of her account present in Wittgenstein's remarks, at least in nascent form.
Wittgenstein takes up the investigation of the concepts of intention and voluntary action at PI §627: Consider the following description of a voluntary action: "I form the decision to pull the bell at 5 o'clock; and when it strikes, my arm makes this movement."-Is that the correct description, and not this one: "…and when it strikes 5, I raise my arm"? 10 10 Although Wittgenstein's examples of action often involve bodily movements, there is no reason to think he believed that actions were restricted to movements of the body.He does also include descriptions that go beyond the body in some of his examples, e.g., "lifting a thing" (PI §615), "When I see the milkman coming, I fetch my jug and go to meet him" (RPP, vol.1, §185).He also notes that not moving may be voluntary: "Can't rest be just as voluntary as motion?Can't abstention from movement be voluntary?"(RPP, vol.1, §845).
There is a clear suggestion that the two descriptions -the passive "my arm makes this movement" and the active "I raise my arm" -to some extent at least, exclude one another: if it is correct to say "I raise my arm", then it is not correct, or at least misleading, to say "my arm makes this movement".Wittgenstein spells the point out more fully as follows: One would like to supplement the first description: "And lo and behold!My arm goes up when it strikes 5." And this "lo and behold!" is precisely what doesn't belong here.I do not say "Look, my arm is going up!" when I raise it.
The point is a grammatical one.If we express the decision to pull the bell at 5 o'clock in words -"I will pull the bell at 5 o'clock" -then it is clear that it has the same predictive aspect as an expression of intention.Wittgenstein's aim is to make clear the distinction between the fulfilment of a decision and what Anscombe calls a true estimate for the future.In the case of a decision which I make and later fulfil, I do not observe that something which, on the basis of evidence, I predicted would happen happens.The problem with "Lo and behold" and "Look, my arm is going up", is precisely that they imply my relation to the later action is one of observer, that is, that I know that the action is taking place on the basis of observing it.In the following remark, Wittgenstein captures this essential 1 st -person/3 rd -person asymmetry by suggesting that we "might say: voluntary movement is marked by the absence of surprise" (PI §628).But one could see the absence of surprise as another way of expressing the fact that my own voluntary behaviour is not something I know on the basis of observation, but is something I can report straight-off, without adverting to observation.And in remarks he made after he'd completed Part 1 of PI, Wittgenstein makes the point explicit: One might…put the matter, I think, like this: my relation to my actions is not one of observation.(RPP I, §712) [W]hen is something an expression of intention?Well, when the act follows it, when it is a prediction.I make the prediction, the same one as someone else makes from observation of my behaviour, without this observation.(RPP I, §788) My own behaviour is sometimes, but rarely the object of my own observation.And this hangs together with the fact that I intend my behaviour.(RPP I, §838) 11 The final element in Wittgenstein's account that I want to draw attention to focuses on the concept of intentional action more directly.At PI §631, Wittgenstein introduces the following example: '"I'm going to take two powders now, and in half an hour I shall be sick."'He says that "[i]t explains nothing to say that in the first case I am the agent, in the second merely the observer" (PI §631).There is a grammatical distinction between the two kinds of statement that needs to made clear and it is not made clear by a statement which, even if it is true, does nothing to illuminate the distinction we trying to grasp.Once again, we need to look at the antecedents.In the case of the second statement, the antecedents are knowledge of how the powders work that is based on observational evidence.By contrast: It wasn't on the basis of observation of my behaviour that I said I was going to take two powders.The antecedents of this statement were different.I mean the thoughts, actions and so on which led up to it.(PI §631) In this case, the antecedents are reasoning that has led to my taking a certain sort of action, reasoning, for example, about how to relieve an attack of indigestion, as well as the exercise of know-how in performing the actions needed to obtain the powders and dissolve them in water, and so on.Such a train of thought presupposes know-how, knowledge of means-ends relations, and practical capacities.
At PI §490, Wittgenstein describes such a train of thought, when it has resulted in an action in fulfilment of an intention, as akin to a calculation: How do I know that this train of thought has led me to this action?-Well, it is a particular picture: for example, of a calculation leading to a further 11 Wittgenstein also gives a negative characterisation of voluntary movement: a piece of behaviour will count as involuntary if I was not aware I was doing it: "A movement of my body, of which I do not know that it is taking place or has taken place, will be called involuntary" (RPP, vol.1, §844).'"That glance was not intended" sometimes means "I didn't know that I gave such a look" or "I didn't mean anything by it'" (RPP I, §846).Wittgenstein also notes that a movement will be involuntary, rather than voluntary, if "one can't prevent" it (RPP, vol.1, §761).And he connects this with the fact that involuntary movements "don't happen in obedience to orders, like voluntary actions" (RPP, vol.1, §840).
experiment in an experimental investigation.It looks like this -and now I could describe an example. 12ttgenstein had expressed the same idea in the Blue Book (BB), where he links it with a distinct sense of the question "Why?": The double use of the word "why", asking for the cause and asking for the motive, together the idea that we can know, and not only conjecture, our motives, gives rise to the confusion that a motive is a cause of which we are immediately aware, a cause 'seen from the inside', or a cause experienced.-Giving a reason is like giving a calculation by which you have arrived at a certain result.(BB, p.15)   Again, these ideas are not developed, but the linking of the concept of intentional action with the idea of reasons for action, and the linking this in turn with the idea of the action's being associated with a calculative order and the applicability of a certain sense of the question "Why?", all seem ripe for further elucidation.
Wittgenstein is clear that we should not think of the expression of intention "I am going to take two powders" as the efficient cause of its fulfilment.However, he notes that it is important that we can often predict a man's actions from his expression of an intention.And this is connected with the fact that the first-person expression of intention is not a report of a mental state, but an exercise of an agent's ability to say, on the basis of the capacities he possesses, what he is doing or about to do.There is, in general, a connection between voluntary actions and intentions.He writes: Voluntariness hangs together with intentionalness.And therefore with decision as well.(RPP I, §805) Voluntary movements are certain movements with their normal surroundings of intention, learning, trying, acting.(RPP I, §776) And these normal surroundings mean "[o]ne draws quite different conclusions from an involuntary movement and from a voluntary: this characterises voluntary movement" (Z §599).And among the conclusions that one draws are that the movement -the action -is one that the agent himself is aware of and not on the basis of observation, that it is done intentionally and perhaps with certain further intentions that are not immediately obvious and can only be discovered by asking the agent why he acts as he does.These further intentions may be among the thoughts that are the characteristic antecedents of intentional action, although it is important that having the thought should not be understood to require the agent's having explicitly said something to himself.The "normal surroundings" of voluntary movement makes the question "Why?" applicable in a sense which will be answered, not by speculation about causes, but by the agent's giving the calculation which arrives at the action he is performing, or has performed, as a result. 13lthough these ideas remain undeveloped, it seems clear that the connection between intentional action and a certain sort of calculative order, and thus with the applicability of a certain sense of the question "Why?", is nascent in Wittgenstein's thought.In PI, he is particularly concerned with the way that our first-person account of past, unfulfilled intentions draws on this order.He spends some time considering examples in which "I was going to say…", "For a moment I was going to say…", and "For a moment I was going to deceive him".We need, he suggests, to ask ourselves: "how did human beings ever come to make the kind of linguistic utterance which we call 'reporting a past wish' or 'a past intention'?"(PI §656).It is not, he tries to show, a matter of recalling having said something to oneself, or reporting the scanty details of what passed through one's mind at the time.It would, he suggests, be less misleading to say "[t]hat the purpose of such a report might be to acquaint someone with my reactions" (PI §657).We might see these statements as the expression of my capacity to speak for myself in describing my attitude towards the events of the past, and what I say is given its significance, not by some momentary process, but by the whole background to the incident.
The same goes in speaking of my further intentions or motives in acting: Why do I want to tell him about an intention too, over and above telling him what I did?-Not because the intention too was something going on at the 13 It is important that Wittgenstein recognises that not all intentional action is done with a further intention, but there are limits: "Not all that I do, do I do with some intention.(I whistle as I go long, etc.)But if I were now to stand up and go out of the house and then come back inside, and to the question 'Why did you do that?'I answered 'For no particular reason' or 'I just did' this would be found queer, and someone who often did this with nothing particular in mind would deviate very much from the norm" (RPP, vol.1, §224).-Not, however, on grounds of self-observation, but by way of reaction (it might also be called intuition).(PI §659) The fact is we are supremely interested in these narratives.In telling someone what I did I am telling him what took place.There is a distinction between true and truthful, and there is the possibility of confirmation by others.But when I tell someone of my further intentions in doing what I did, there is no distinction between true and truthful, and the concept of truthfulness relates to ideas of character rather than cognitive capacities.Whether someone accepts what I say will depend on their willingness to take me at my word, and on the way my narrative fits with the surrounding evidence.Wittgenstein's suggestion that we might call these statements "intuition" draws attention to the fact that they are made straight off, not on the basis of observation, in a narrative in which I speak revealingly and for myself.

5.
It is clear that in focusing on intention Anscombe was homing in on a pivotal concept in Wittgenstein's later philosophy.Wiseman is surely correct in claiming that Wittgenstein's method of grammatical investigation at last provided Anscombe with the means by which she believed she could make real philosophical progress.It also seems clear that the basic elements of the account of intention that she gives in the first half of Intention were being worked out by Wittgenstein during the period when she was his pupil and one of his principal interlocutors.In no particular order, this includes 1 st -person/3 rd -person asymmetry and the idea that one knows one's own intentional actions not on the basis of observation, the idea that expressions of intentions are a form of prediction, the importance of context for the formation of intentions, the connection between the concept of intentional action and a calculative order, the applicability of a special sense of the question "Why?" to intentional actions, the distinction between what someone does intentionally and the further intentions with which he acts, the role of the concept of truthfulness in 1 st -person accounts of motives and further intentions, and the recognition of a degree of indeterminacy that affects both our 1 st -person account and our 3 rd -person criteria for the ascription of motives and further intentions.However, it was clearly Anscombe's willingness to start from scratch and work out these ideas afresh and for herself that enabled her to develop them in the way that she did, systematically with a new kind of analytic penetration, and with questions in mind that were distinctively her own.
Anscombe's positive account of intentional action begins with the proposal "'Intentional actions are ones to which a certain sense of the question 'Why?' has application" (I:11).Her primary aim is to make this sense perspicuous.One of the first observations she makes is that the question is refused application if the agent's answer is "I was not aware I was doing that."She then notes that any single action can have many different descriptions and that an agent may know what he is doing under one description, and not under another.This idea is one of the main elements in Anscombe's analytical working out afresh of the idea of the calculative order that Wittgenstein suggested is definitive of intentional action, but before she gets to that point, she spends time clarifying a number of distinctions: the distinction between mental causes and reasons, between motives of various kinds and intentions, and between voluntariness and involuntariness.With these distinctions in place, she is in a position to give her summary account of intentional action: Intentional actions are a sub-class of events in a man's history which are known to him not just because he observes them.[…] Intentional actions […]  are the ones to which the question 'Why?' is given application, in a special sense which is so far explained as follows: the question has not that sense if the answer is evidence or states a cause, including a mental cause; positively, the answer may (a) simply mention past history, (b) give an interpretation of the action or (c) mention something future.In the cases of (b) and (c) the answer is already characterised as a reason for acting, i.e. as an answer to the question 'Why?' in the requisite sense; and in the case (a) it is an answer to that question if the ideas of good or harm are involved in its meaning as an answer 14 ; or again if further enquiry elicits that it is connected with 'interpretative' motive, or intention with which.(I:24-5) Anscombe holds that any true description of what an agent, A, does intentionally will be a description of an action of his, but only those descriptions under which the action is one to which the question "Why?", 14 This is taken to characterise backward looking motives.Nordic Wittgenstein Review 12 | 10.15845/nwr.v12.3683|[prepublication for open review]   McGinn 19 in this special sense, has application, will count as an intentional action of A's."A is X-ing" is a description of A's intentional action if (a) it is true and (b) A answers the question, "Why are you X-ing?", with an answer from within the permissible range of reasons for action (a), (b), or (c). 15Let's suppose the answer to the question is the mention of something future, (c), which is also a wider description of the action that makes the agent's intention in acting clear: "I'm Y-ing."If there is something that can be observed that falsifies the description "A is Y-ing," then you cannot say "A is Y-ing."But if this is not the case, the question "Why?" applies again.We can, Anscombe suggests, continue this line of questioning until we reach a point at which the expression of the objective can be given only by means of the expression 'in order to.' Below this point, all the descriptions are of intentional actions which A is currently in the process of enacting; above it, we can say "A is X-ing in order to Y," and there is an implicit recognition that success is uncertain: "the failure to execute intentions is necessarily the rare exception…the failure to achieve what one would finally like to achieve is common…What is necessarily the rare exception is for a man's performance in its more immediate description not to be what he supposes."(I:87) 16 Anscombe's account can be seen as combining Wittgenstein's special sense of the question "Why?" and the idea of a calculative order that characterises intentional action and thinking it through afresh to reveal what is essential to "the form of description of intentional actions" (I:84).By repeated application of the question "Why?", an agent's reasons 15 Anscombe follows Wittgenstein in observing that there are cases in which the answer the answer to the question "Why?", in this special sense, may be "For no particular reason", "I just thought I would", and so on.And such answers "are often quite intelligible" (I: 26), but "their interest is slight, and it must not be supposed that because they can occur that answer would everywhere be intelligible, or that it could be the only answer ever given" (I:34).She also notes that "I don't know why I did it" may be given in answer to the question "Why?", in a sense which does not mean there is a causal explanation which one does not know; she suggests the answer represents an intermediary case in which the question both "has and has not an application" (I:26). 16An agent can, of course, be doing something which he nevertheless fails to do, if it is something which takes time to complete and he is interrupted, or changes his mind.In this case, it will be true to say "A was Y-ing", although it may be that this could only be known to others on the basis of A's own testimony, as not enough had been done for it to be clear to others what it was that A was doing.Anscombe concedes that the break between cases in which we can say "A is Y-ing" and "A is X-ing in order to Y" is not a sharp one, "[b]ut the less normal it would be to take the achievement of the objective as a matter of course, the more the objective gets expressed only by 'in order to'" (I:40).
for action are made perspicuous in the characteristic A-B-C-D means/ends structure that is present whenever intention-dependent concepts are used to describe an agent's intentional actions.A's reason for doing A is to do B, his reason for doing B is to do C, and so on.Each A, B, C, and D gives a description under which A's action is intentional.The final description in the series, the one before the break ushered in by the "in order to" locution, swallows up all the descriptions below it and gives the intention with which all the actions described earlier in the series are done.Anscombe recognises that B, C and D are all redescriptions of a single action, A, and that each redescription in the series depends on the one below and is independent of the one above; only more circumstances are required for A to be B, B to be C, and so on.What she is concerned with, however, is our practice of giving and asking for reasons, and there is no assumption that the first member of the series is a description of a bodily movement; it may be a description of what A is doing which we would give straight off, which A would give straight off not on the basis of observation, and to which the question "Why?" applies.
There is no assumption that there is a determinate number of steps between the first member of the series and the last, or that there is a single series that applies in any particular case: there may be a number of intentions with which A does what he's doing, which do not swallow up one another, but which swallow up A, B, and C.There are no absolute answers.Nor is there any assumption that the agent thought through the calculative order that repeated applications of the question "Why?" reveals prior to acting.The whole practice is grounded in an agent's capacity to say, not on the basis of observation, what he is doing in doing something else, and thus to make his intentions in acting clear.Anscombe's account of intention makes perspicuous the connection between our employment of intention-dependent concepts in accounts of what oneself or another is doing and an agent's capacity to answer, without observation, the question "Why?", by giving A-B-C-D type descriptions that make clear one's intentions in acting.
But can one, Anscombe asks, know without observation, say, that one is painting a wall yellow?The question is one which her more systematic approach leads her to confront more directly than Wittgenstein did and to make the vital distinction between practical and speculative knowledge.Anscombe claims that if one knows that Z happens if one does A, B, C, then one can have the intention to do Z and know without observation that one is doing Z. Anscombe calls the knowledge without observation that one is doing Z practical knowledge. 17The very same thing -that Z is happening -can be known by others on the basis of observation, as an object of speculative knowledge.When the description of what happens is the very thing which I should say, without observation, is the thing I am doing, then there is no distinction between my doing and the thing that happens: in Anscombe's formula, "I do what happens" (I:52).My intentions, and thus my capacity to know without observation, can reach as far as my practical knowledge.This is an observation about how we use the word "intend".The concept of practical knowledge is fundamental for Anscombe's account and it represents an important contribution to how to think about Wittgenstein's grammatical observation of the 1 st -person/3 rd -person asymmetry that characterises the use of intention-dependent concepts. 18

6.
There are at least two ways in which Anscombe's account develops that cannot be put down as attempts to think anew ideas that are there, even in nascent form, in Wittgenstein's later philosophy.There is no reason to think that Wittgenstein made any clear distinction between voluntary and intentional, or that he thought the distinction carried any ethical significance: "Voluntariness hangs together with intentionalness" (RPP I, §805).The important distinction, for him, is between voluntary and involuntary, and that is understood in terms of movements that one knows nothing about, is unconscious of, or knows only on the basis of observation.However, while Anscombe acknowledges that "[e]very intentional action is also voluntary" (I:90), she is, given her development 17 Anscombe notes that in most of the things we do we rely on the use of our senses to guide us, and this is to make use of speculative rather than practical knowledge: practical knowledge is restricted to "the account that one could give of what one was doing, without adverting to observation; and the account of exactly what is happening at a given moment (say) to the material one is working on" (I:89).Although the term is most often used in connection with specialised skills, Anscombe recommends that the term should be used more widely insofar as intentional action generally presupposes "what might be called 'knowing one's way about' […]  and this knowledge is exercised in the action and is practical knowledge" (I:89). 18Anscombe thinks that the concept of practical knowledge can only be understood "if we first understand 'practical reasoning'," (I:57) which suggests that, despite her claim that her question "Why?" describes the same order as the one revealed by Aristotle (I:80), she does believe that Aristotle has a vital contribution to make when it comes to understanding the nature of practical knowledge.Nordic Wittgenstein Review 12 | 10.15845/nwr.v12.3683|[prepublication for open review]   McGinn 22 of the idea that actions are intentional under a description, in a position to make an important distinction between the voluntary and the intentional that was not available to Wittgenstein: Something is voluntary though not intentional if it is the antecedently known concomitant result of one's intentional action, so that one could have prevented it if one would have given up the action; but it is not intentional: one rejects the question 'Why?' in its connexion.(I:89) The importance of this distinction is that it also opens up the possibility that something can be an involuntary antecendently known concomitant of one's intentional action, but one may still have reasonsin particular, ethical reasons -that compel one to perform the intentional action: From point of view, however, such things can be called involuntary, if one regrets them very much, but feels 'compelled' to persist in the intentional actions in spite of that.(I:89) There is not space her to examine Anscombe's moral philosophy in any detail, but with the above distinctions in place she had developed a philosophy of psychology that provided the foundation for an ethics which focused on the concept of virtue.Anscombe believed that the concept of virtue was best explored through examples.For example, the concept of justice could be investigated through the distinction between what is intrinsically unjust -punishment for something one manifestly hasn't done -and what is merely unjust in certain circumstances -for example, non-payment of a debt -in cases where we have to exercise judgement and decide according to what is reasonable.The concept of virtue emerges as the idea of an abiding character trait that is ultimately to be understood in terms of a capacity for practical reason and what it is for a human being to live well: "a good man is a just man; and a just man is a man who habitually refuses to commit or participate in any unjust actions for fear of any consequences, or to obtain any advantage, for himself or anyone else" (Anscombe 1958: 16).
The philosophy of psychology Anscombe developed in Intention allowed her, in combination with the idea of a modern version of Aristotelian ethical naturalism, to focus on an agent's intentional action and raise the question whether it was good or bad, permissible or impermissible.What matters in the moral assessment of man's actions is what he does and the practical reasoning that plausibly emerges, given the circumstances, 19 as the reason he gives for acting in response to repeated applications of the question "Why?".She could then distinguish between a man's intentional actions, which includes the action described by his ultimate intention in acting and all the intentional actions in the series which specify the means by which he achieved it, and the merely foreseen but unintended consequences of what he intentionally did.If someone performed a bad action, then he should get no credit for its good consequences; conversely, if someone performs a good action, he should not be held responsible -should not be blamed -for its bad consequences. 20This element in Anscombe's account is made possible by the development of the idea of actions being intentional under a description, which emerges in the context of her use of Wittgenstein's method of grammatical investigation, but reflects her own distinctively systematic and analytical approach to the task of conceptual clarification.
At §20 in Intention, Anscombe takes up the question she raised in §1: is the concept of intention univocal when it is used in expressions of intentions for the future, when we speak of intentional actions, and the intention with which an action is done.She now expresses the question as follows: 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?I.e. is 'intentional' a characteristic of the actions which 19 The circumstances provide the context in which the answers an agent gives to the question "Why?" are assessed for sincerity, honesty, credibility, degree of self-deception, and so on.It is here that Anscombe notes the complexity in our ordinary criteria for judging whether or not a man did or did not have a certain intention in action, which shows what "kind of truth there is in the statement 'Only you can know if you had such-and-such an intention or not'.There is a point at which only what the man himself says is a sign; and here there is room for much dispute and fine diagnosis of his genuineness" (I:44).However, she also notes that "while we can find cases where 'only the man himself can say whether he had a certain intention or not'; they are further limited by this: he cannot profess not to have had the intention of doing the thing that was a means to an end of his" (I:44). 20This distinction is crucial to the Catholic doctrine of double-effect, which explains the permissibility of an action that causes harm, if it is the antecedently foreseen side-effect of a permissible action which is done to promote some good end.It was also crucial in Anscombe's criticism of Truman's order to drop the atomic bombs on Hiroshima and Nagasaki.According to Anscombe, the deaths of between 130,000 and 225,000 innocent Japanese civilians was not a foreseen but unintended consequence of a permissible intentional action.It was the means by which Truman set out to achieve his ultimate intentionthe surrender of the Japanese governmentand was therefore something he intended, for which he should be held responsible.

McGinn 24
have it, which is formally independent of those other occurrences of those other occurrences of the concept of intention?(I:30) At this point, Anscombe's philosophical aims cease to be purely descriptive and she sets out to show that something is "essential to the existence of the concept of an intention or voluntary action" (I:33).
Anscombe's argument here is fairly tortuous.It takes the form of a reductio, in which we assume that (a) the concept of intention only occurs in "intentional actions", and (b) the only answer to the question "Why are you X-ing?", where it is not refused application, is "I just am".Anscombe considers what might still be permitted by such restrictions -the appeal to backward-looking and interpretative motives as reasons for actionbut essentially her claim is that if the only occurrence of intention were as the intention of doing whatever one is doing, the notion of intentional action itself would be a very thin one; it is not clear why it should be marked off as a special class among all those of a man's actions and movements which are known to him without observation.(I:32) Moreover, if the only answer to the question "Why?" is "I just am," then there would in effect be no special sense of the question "Why?" "and no distinct concept of intentional action at all" (I:32).Anscombe concludes that the occurrence of other answers to the question "Why?" besides ones like "I just did", is essential to the existence of the concept of an intention or voluntary action.And that means that the concept of intention must also occur in expressions of intention for the future and in speaking of the further intention with which an action is done.
Anscombe's argument clearly goes beyond the weak claim that an answer to the question "Why are you X-ing?" which sometimes makes sense in our language-game -"I just am" -would not be intelligible if we think of it applying universally.It is rather that it is essential to our language that it include the resources to ask the question "Why?" in the special sense and that the concept of intention should also occur in expressions of intentions for the future and in speaking of the further intentions with which an action is done.The question is why does Anscombe wish to show that our concept of intention necessarily takes the form it does?Why wasn't she content to describe how our concept actually functions and leave it at that?It seems likely that the answer lies in Anscombe's belief that her account of intention captures something essential about the human soul, namely that it possesses a rational nature.Human beings, insofar as they are agents who have the capacity to describe their own intentional actions -to say what they are doing -essentially have the capacity to answer the question "Why?" in the special sense, in a way that makes their further intentions clear, and are thus essentially autonomous, social, self-aware, practical reasoners who have intentions for the future, who are capable of self-reflection and ultimately of moral self-criticism.Anscombe accepts that intention-dependent concepts apply to animals which have no language: "we certainly ascribe intention to animals" (I:86).This means that we recognise that the cat, say, sees the bird, aims to catch it in order to kill it, so we can say what the cat is further doing in crouching, slinking along with its eye fixed on the bird, in just the way that is characteristic of description of intention in acting.However, the cat -non-linguistic creatures in general -cannot give expression to any knowledge of its own action, only creatures with linguistic capacities and the concept of intentional action can do that.Anscombe uses her grammatical enquiry into the nature of this concept to show us what it means to be a human being, and she wants to show that the features she uncovers belong to the concept of intention, and thus to human beings, as a matter of necessity.
The description of a human intentional action is the description of a self-consciously executed intention which is an expression of practical knowledge.She wants to show that the existence of the question "Why?" and answers to it besides "I just did" are essential to the existence of the concept of intentional action because she wants to show that the human soul is essentially characterised by a rational nature, which is the basis of the autonomous, moral self.This is not a project that Wittgenstein shares.Indeed, he seems to warn against it: [I]f anyone believes that certain concepts are absolutely the correct ones, and that having different ones would mean not realizing something that we realize -then let him imagine certain very general facts of nature to be different from what we are used to, and the formation of concepts different from the usual one will become intelligible to him.(PI, PPF, xii, §366) However, Wittgenstein also raises the question whether our concepts "are determined by our interest, and therefore by our way of living?" (LWPP II, p.43).He asks whether a legislator could abolish the concept of pain and responds as follows: time.But because I want to tell him something about myself, which goes beyond what happened at the time.I reveal to him something of myself when I tell him what I was going to do.