My topic is the conception of clarity in Wittgenstein’s later thought–or, more precisely, his understanding of clarification, taken to be the task of philosophy. I am picking up a suggestion of Warren Goldfarb’s and Steven Gerrard’s, that we see the change in Wittgenstein’s view of clarity as central in the transformation that his thought underwent (Gerrard 2002: 69).
It is sometimes argued that a reading of the Tractatus that takes seriously its description of its own propositions as nonsensical, as genuinely without content, cannot allow for the depth and significance of the later change in his thought. But such an idea rests, I think, on failure to see how far-reaching the change is in Wittgenstein’s understanding of clarification, and what it demanded of him, and of us as readers. These issues seem to me to shape Part I of Philosophical Investigations. I can discuss only some of this. My argument is meant to provide a commentary on a number of sections in the Investigations, but also on the claim in the Preface that Wittgenstein’s new thoughts could be seen in the right light only by contrast with, and against the background of his old way of thinking. I believe that that claim holds very strongly indeed of his new thoughts about what is involved in philosophical clarity in relation to his old way of thinking about it.
I begin by making things worse. That is, one might take a central contrast between Wittgenstein’s early approach to philosophy and his later approach to be that Wittgenstein didn’t, in his later thought, attempt to get all of philosophy done at once, as it were. ‘Problems are solved’, he says, ‘not a single problem’ (PI §133). And we take that to mark a deep difference from the Tractatus. Well, it may be. But I want to make that contrast less clear-cut, or indeed more puzzling, by looking at clarification as we are supposed to go in for it from the point of view of the Tractatus. The correct method in philosophy, Wittgenstein said, would be to say nothing except propositions of natural science, and then, when someone said something metaphysical, you could show him that he had failed to give a meaning to certain signs in his propositions (TLP 6.53). So you are essentially waiting for something metaphysical to turn up in this other person’s statements, and dealing with it when it does turn up. Wittgenstein says that the method wouldn’t satisfy the other person, but surely, if one did go in for that approach, it would appear to be every bit as piecemeal as the approach of the Investigations. Indeed, when we have, at the beginning of the Investigations itself, an approach to philosophical problems about language that appears to be quite piecemeal, we have an interlocutor speaking up at §65, plainly feeling dissatisfied. So it at least appears that we have a situation like this: the Tractatus specifies a piecemeal method of approach which can leave the person on whom it is tried feeling dissatisfied, feeling (that is) as if the philosophical issues have been evaded, and the Investigations demonstrates a method in its first 64 sections which can leave the person on whom it is tried feeling dissatisfied and as if the philosophical issues have been evaded.
I believe that the idea that we can make out a big difference between Tractatus and Investigations by taking the former to be committed to an allat-once demonstration that philosophical propositions are nonsensical, where the latter works on a problem-by-problem basis, is confused. Not that there is no difference, no important difference, in the vicinity of this idea, but it’s not that easy to see what it is. What gets in the way is precisely the idea that the Tractatus provides a wholesale approach to the demonstration of the meaninglessness of philosophical propositions. (This isn’t the only thing that gets in the way.)
Let me do this wholesale–retail thing in more detail: the wrong idea and the right one. I’m going to argue that there is no wholesale demonstration of the nonsensicality of philosophical propositions that we are supposed to have available to us at the end of the Tractatus, and that therefore there can be no attempt to contrast the Tractatus with the Investigations on any supposed basis of wholesale dismantling of philosophy by early Wittgenstein versus patient disentangling by later Wittgenstein. If there is something wholesale in the approach of the Tractatus, it will need some patient disentangling to see what it is.
There are various ways in which one might take the Tractatus to provide a wholesale method for criticizing philosophical propositions. It might be said that we can infer from the Tractatus that such propositions violate logical syntax, and are therefore nonsensical. Or it might be thought that we can use the Tractatus doctrines as the basis of an inference from the failure of philosophical propositions to be bipolar, together with their not being tautologies or contradictions, to the conclusion that they are nonsensical. In the case of a large group of such propositions, including (supposedly) the propositions of the Tractatus itself, it might be thought that the Tractatus allows us to infer, from the fact that they are attempts to say what can only be shown, that they are nonsensical. But there is a devastating problem with any such attempt to see the Tractatus as providing a wholesale method for dealing with philosophical propositions. Any propositional sign can be used in various ways; there is no reason to doubt, of anything that looks like a propositional sign, that it can be used to express a thought, or to name a cat, or in other ways. So, if there is a ‘wholesale’ approach to demonstrating of any philosophical or metaphysical proposition that it is nonsense, there must first be some way of making clear how the proposition is to be taken, since it can be used to say something perfectly intelligible. It isn’t to be taken in any of the ways in which it wouldn’t be nonsense. How, then, is the intended nonsensical use to be made clear? For only if that can be done could the wholesale approach catch hold of the proposition in question. So some kind of clarification, or attempt at clarification, is going to be involved if the wholesale approach is even to have a chance to connect with some purportedly nonsensical proposition. The devastating problem for a reading of that general type is this: to attempt to specify which way of taking the propositional sign makes it nonsensical, you have to make clear what use of the sign you have in mind. Any such clarification deals with the detail of the individual sentence; it is an essentially retail proceeding. But, in the case of a nonsensical proposition, the attempt at clarification will reveal that it is nonsense by making plain that there is no particular use of the propositional sign that is clearly in focus; there is no way in which the sign is being meant. The ‘wholesale’ approach requires that there be some way of taking the propositional sign, such that the sign, taken that way, can be recognized to be an attempt to express something which propositions allegedly can’t be used to express, or in some other way to violate some or other rule. But then that use must be specifiable, and distinguishable from other uses. But this attempt to specify a use proceeds by attempting philosophical clarification. In the course of that attempt the proposition’s character will be revealed, without any appeal to supposed general Tractatus doctrines.1
I shall put the argument in two further forms.
1 Wittgenstein believed that, if a propositional sign expressed a thought, the proposition could be clarified. The activity of clarification, as he understood it, involved rewriting propositions in a way that would enable us to see clearly what, in a sense, the proposition had shown all along; we would be able to see the use of the proposition more clearly. Since, on the Tractatus view, any propositional sign can, in some use, express a thought, any criticism of something that looks like a proposition as being a mere nonsense must involve distinguishing the supposed use-as-nonsense from other possible ways of using the sign, which might be legitimate. We need to try to specify the use we have in mind. But the attempt to specify the use we have in mind is carried on by attempting philosophical clarification. That is, philosophical clarification is an activity which we can and, indeed, must attempt to carry through if we want to criticize a thing that looks like a proposition, and claim that it is nonsense. It is, essentially, in the failure of the attempt at clarification of the particular proposition with which we are concerned that we are able to come to recognize that there was nothing there to clarify. There is no philosophical critique of propositions available on the basis of the Tractatus, separate from the Tractatus conception of clarification of genuine propositions.
2 In this version of the argument, I consider an example of a wholesale approach. Suppose someone claimed that, from the recognition that a particular proposition lacked bipolarity, and was neither a tautology nor a contradiction, that it therefore had to be nonsensical, from the Tractatus point of view. The question then is: what supposedly lacks bipolarity? No mere sign has or lacks bipolarity. And, again, if we call something a tautology, we are taking it that the names in it have a particular use: if two occurrences of the same letter, say, are not names for the same thing, the sense cannot ‘cancel out’ as it does in a tautology. Take a typical Tractatus proposition, of the sort that appears to lack bipolarity, ‘Propositions are truth-functions of elementary propositions’. What use do we want to give the first word of that proposition, as it occurs there? It is hardly meant to refer to all things that look like propositions. Nor do we intend to use the word “propositions” to mean truth-functions of elementary propositions; we don’t want to use the quoted propositional sign to say that truth-functions of elementary propositions are truth-functions of elementary propositions. In order to make this Tractatus proposition out as lacking bipolarity, and as not being a mere tautology, we should have to specify some other sort of use for that first word. If there is no specifiable use that we will accept as what we want there, we can recognize a kind of failure, but it is a failure to give any meaning to one of the words we are using. The idea we may have, that ‘absence of bipolarity’ might be directly available to us as we consider a proposition, comes from failure to take seriously that a sign does not itself determine a use. We have to make clear that use of the propositional sign such that we want to say: the sign used that way expresses something that is not bipolar. Before there is any attempt to apply some general doctrine about non-bipolar propositions, we’ve got to have such a proposition. But what will happen if we attempt to spell out the use we mean is that the attempt at clarification will show us that there is nothing we will accept as what we mean. The attempt at clarification has to precede the supposed application of doctrine; and, if indeed the proposition-like thing in question is philosophically problematic, what will happen is that the attempt will bring out a kind of failure to mean anything clear at all. We shall never get as far as the supposed application of doctrine. What does the work is the attention to the particular problematic sentence itself, the attempt to clarify it, and the failure of that attempt. Again: the idea we may have that such-and-such proposition is not bipolar is, itself, a symptom of not having attempted clarification. It involves an irremediably cloudy impression-of-having-a-thought, attached to a sign the use of which we are not attending to. In general: there is no description available to us, of any proposition, which will enable us to connect it with some supposed Tractatus doctrine about nonsensicality and, thereby, to deduce the non-sensicality of the proposition. We have first to attempt to clarify the use of our signs. If that attempt falters, we shall have come to see that the prop-ositional sign in question is nonsensical. If it does not falter, then nothing else is going to show that the proposition is nonsensical. There is here no special Tractatus sense of ‘nonsensical’, only the ordinary idea of not meaning anything at all.2
I have given that argument in general terms, but it is meant to cover a variety of readings of the Tractatus, in which the ‘wholesale’ criticism of philosophical propositions is made to depend on this or that different supposed principle for demonstrating nonsensicality. There are, however, special points to be made about two of these ‘wholesale’ readings.
I have argued that the Tractatus does not provide a general principle which can be used to demonstrate that philosophical propositions are nonsensical, and that therefore we cannot differentiate the philosophical method of the Tractatus from that of Wittgenstein’s later philosophy by reference to the place of such a general principle in the Tractatus. The Tractatus approach to philosophical confusion is, in an important sense, piecemeal; it depends essentially on enabling a person to see that the attempt to clarify the use of his or her words falters. In the attempt at clarification, one comes to be able to recognize that one has failed to say anything, that one has not given meaning to one or more of the signs one has used.
Let me turn here to Philosophical Investigations §65. The person who objects to Wittgenstein’s approach in the previous 64 sections says: ‘You take the easy way out! You talk about all sorts of language-games, but have nowhere said what the essence of a language-game, and hence of language, is: what is common to all these activities, and what makes them into language or parts of language. So you let yourself off the very part of the investigation that once gave you yourself most headache, the part about the general form of propositions and of language.’
The propositions of the Tractatus which appear to be statements about the nature of language are supposed to be recognizable by us as nonsensical, and the very questions we ask in philosophy are supposed to be recognizable as not genuinely questions at all. We are, however, supposed (when we get to the end of the book) to be able to engage in philosophical clarification; this activity includes not only clarification that turns out to be successful but also attempts which fail, where the failure itself can help us to see that we had meant nothing by our words. Wittgenstein’s later criticism of his earlier thought has at its center his own reflections on how the questions which we supposedly renounce in the Tractatus, and supposedly recognize not to be questions, nevertheless shadow the kind of clarification which the Tractatus recommends. The book leaves us with a method that is in the shadow of the big questions he had been asking. The search for the essence of language is, in theory, überwunden, overcome. But it is really still with us, in an ultimately unsatisfactory, unsatisfying, conception of what it is to clarify what we say.
We can see Wittgenstein, in §65 of the Investigations, to be critically engaged with a conception of the problems of philosophy, with the idea that there are certain fundamental essential problems. He took himself to have been in the grip of this idea. Although he had (at the time he finished the Tractatus) believed himself to have resolved these fundamental problems, and indeed to have shown that they were, in a sense, not questions at all, he had had a conception of clarification, of the method of philosophy, which, although he was unaware of it, was distorted by the idea of big fundamental problems. So, to understand the difference between Wittgenstein’s earlier and later thought one has to see the difference in his conception of clarification; and to understand that, one has to see how his earlier thought was in the grip of the idea of big fundamental philosophical problems, and how the later thought was meant to free us from the grip of that idea. The idea of philosophical clarification in Wittgenstein’s later thought is tied closely to his idea of how our thinking can be distorted by the conception of big essential philosophical problems, a conception which it is enormously difficult, in practice, to renounce.
(Here, I ‘d like to clarify something about what is involved in reading the Tractatus. In reading any work of philosophy, we may operate with some distinction between what the author thought had been achieved by it and what he or she did achieve. In the case of the Tractatus, though, we need particular care in distinguishing Wittgenstein’s own conception of the Überwindung of his own propositions and what we may want to say about the achievement of that Überwindung. It may be that, as Wittgenstein understood his achievement at the time of completing the Tractatus, he took the metaphysical-seeming propositions of the work, including propositions about the essence of language, to have been completely überwunden, and to be recognizable as nonsense, plain and simple. It is entirely consistent with such a reading of what Wittgenstein took his achievement to have been at that time to hold that a metaphysical view of language had not been completely überwunden and was, indeed, reflected in the work in some way or other, for example, as I shall argue, in the conception of philosophical method with which the book leaves us. To argue that there is, in some sense, metaphysics remaining in the Tractatus is not to hold that Wittgenstein took any of his metaphysical-seeming propositions to be meant to convey some sort of metaphysical insight. I have emphasized this point because it may appear possible to infer from ‘There is something metaphysical in the Tractatus’ to ‘Not all the metaphysical-seeming propositions in the Tractatus are meant to be recognizable as plain and simple nonsense’. I have elsewhere argued for an interpretation which says that Wittgenstein did mean the metaphysical-seeming propositions, including those about the nature of language, to be recognizable as plain and simple nonsense; I am not here arguing for such a reading but only noting that they can all have been so intended, and it nevertheless be the case that the book is unwittingly metaphysical in some respect or other.)4
I have argued that the Tractatus leaves us with a conception of philosophical method that in some respects resembles the ‘piecemeal’ method of the Investigations. We wait for someone to say something metaphysical; we show the person who said it that attempts at clarifying his or her use of signs reveals that some sign is not being used to mean anything. But what Wittgenstein had in mind as clarification had in it a conception of the general logical character of all thought and speaking and inferring. The propositions of the Tractatus, for example about the truth-functional character of logic, may be überwunden, but the picture of logic as having a certain general character is then present in the recommended method of the Tractatus in the way inferential relations are supposed to be treated in the clarifying of propositions. If, in the Tractatus, ‘a picture held us captive’, that captivity can be seen in the way the clarification of propositions proceeds in accordance with a model taken to have totally general applicability. The Tractatus treatment of the Big Question of the nature of language leaves behind (once the Question is supposedly shown to be not a question at all) a philosophical method that pays no attention to differences, to the complex reality of our propositions and our modes of inferring, or to the reality of our particular philosophical difficulties. The failure of the direct attack on the Big Question of the nature of language, the attack that shares with the Question a conception of the generality of the issue, can be seen (supposedly) in the persistence, in the philosophical method of clarification, of a false understanding of philosophy itself. The Big Question does not disappear; the Tractatus had only seemed to provide a route to genuine clarity.
In §133 of the Investigations, Wittgenstein says that the real discovery is the one that makes it possible for philosophy no longer to be tormented by questions which bring philosophy itself into question. He says that problems are solved (difficulties eliminated), not a single problem.–How, then, is it possible for us to see philosophical problems as problems, and not in terms of a single problem? What the real discovery is, then, is what will enable us to see philosophy in that way.
This leads to a question. In what way does the Investigations enable us to leave behind a captivity to the idea of the Big Question? I take this question to be vital in our attempt to see how Wittgenstein’s understanding of clarification shifts between the Tractatus and the Investigations and, hence, also vital in the reading of the Investigations itself.
The points about philosophy that I have been discussing are already present, most of them anyway, in the notes on philosophy in Wittgenstein’s Big Typescript of 1933. The version of PI §133 in the Big Typescript lacks the Investigations remark that there is not a philosophical method, but rather methods, like different therapies (PO, p. 195). And, indeed, the chapter on philosophy in the Big Typescript itself seems to work with the idea of a single method. I don’t think that Wittgenstein became clear until somewhat later what might be a way of responding to our tendency to think in terms of Big Questions, what it might be to write a book that could show what it is to think in terms of philosophical problems, rather than of a single essential problem. He did come to believe that one should demonstrate with particular examples the usefulness of different methods, for example, of inventing language-games unlike ours in important respects. But he did not, I think, believe that there was anything that could, as it were, insulate his approach from a kind of misreading that would see the treatment of particular examples as derived from or embodying some supposed response to the Big Question of the nature of language, or the conditions of sense. Such misreadings may have various sources, but one of them is the feeling that his philosophy would lack depth or would be unsatisfying if it did not deal with some such Big Question. To see him as an important philosopher, so it may seem, we must see him as moving from one approach to the Big Question in his earlier thought to another later on, not as serious in what he says about Big Questions. So, here is a brief summary of one kind of understanding of Wittgenstein’s later thought about philosophical clarification, a summary of what I take to be a misunderstanding: Wittgenstein gives a new answer to the Big Question of the nature of language, which dictates an understanding of clarification as making plain the rules of grammar of the language-games in which we engage. We are thus (supposedly) enabled to see that we have got into our philosophical puzzlements by using words in ways that have no place in those language-games. Here, a general kind of approach to philosophical problems is established first, and in advance of its application to particular problems. It is no part of clarification (thus understood) that we should be enabled to see our problems themselves as standing on their own, to be dealt with on their own, or that indeed seeing them in that way constitutes a great part of the difficulty of responding to the problems. I am arguing for a different understanding of Wittgenstein’s later ideas about philosophical clarification, taking seriously the remark from the Big Typescript (also in Zettel), that what makes the greatest difficulty in philosophy is the kind of reordering of our understanding that enables us to see philosophy as cross-strips, each of them a whole definite piece.5
Let me try here to avoid two possible misconceptions. (1) I spoke just now of seeing our problems as, in a sense, standing on their own. I don’t mean that Wittgenstein thought of philosophical problems as having no light to shed on each other. He very definitely thought they did shed light on each other; but the question is how they do so. It is very natural for us to organize our thought about the connection of problems to each other through the idea of something both deeper and more general than the individual problems themselves. This conception is present in the Tractatus and survives the supposed Überwindung of the questions. Some or other version of it characterizes many readings of the Investigations, for example, that of Kripke, who sees the treatment of particular problems as falling out of a general treatment of Big Questions about Meaning. (2) In saying that Wittgenstein was not, in the Investigations, concerned to provide a new answer to the Big Question of the nature of language, I don’t mean that he was not concerned with the nature of language. I mean that there is a difference between seeing such a question as a Big Question and seeing it simply as a problem or rather a group of problems, philosophical problems that can be approached through the methods he had developed.
The presence of some or other Big Answer in our understanding of the Investigations is responsible for a kind of incoherence in many readings. That is, we may say that Wittgenstein thought that in philosophy we need to remind the person who is confused of how we use words. And we may say that, because we are simply issuing such reminders, philosophical disputes should not arise; if they do, we can simply provide other reminders instead. But if, in our understanding of Wittgenstein, we take it that the point of the reminders is tied to an idea that we speak nonsense when we depart from the ways of using language that are in accordance with the rules of our language-games, we are making use of a general conception of language, of sense and nonsense, of the conditions of sense, which would seem to be highly disputable. So far as we ascribe to Wittgenstein a conception of clarification which is tied to some such general account of language, the fact that the bits of clarificatory discourse are not themselves interestingly disputable cannot begin to show that philosophical practice, as Wittgenstein understood it, doesn’t give rise to serious disputes. If we want at least to try to see Wittgenstein’s understanding of philosophical method as coherent, we have to see his idea that it does not give rise to serious dispute as not combinable with the intention of providing this or that Answer to a Big Question about language. The real difficulty is in not thinking Big Questions; the real discovery is how not to do it. When Wittgenstein said ‘Don’t think, look’, the hardness of looking is that of seeing the case with which we are puzzled as treatable genuinely on its own, the hardness of letting what can be said about it help, letting it satisfy us. This is at the heart of his later conception of clarification.
The authors in this volume have been asked to focus on the Investigations and related works. But we can, I think, be helped to see what Wittgenstein is doing in the examples of philosophical method which he provides in the Investigations if we consider a simple example from his actual practice with his students. Such an example may help us to see the transformation of our problems at which Wittgenstein aims: we may be able to see a problem losing its tie to Big Questions, the tie which keeps us, in our struggles with the problem, from going out the unlocked door.
The example I shall consider comes from Elizabeth Anscombe’s account of how she had felt trapped by phenomenalism, and how Wittgenstein’s ‘medicine’ enabled her to see her way out of it.6 She had felt trapped by phenomenalism because she had responded strongly against a Lockean representative realism which insisted that colors as she saw them were not genuinely part of the external world. But, finding herself insisting that blue (this blue), or yellow (this), were there, out there, she was on a path that led, or seemed to, in a direction in which she did not want to go, to a reading of the world as itself made of these items that she was thus aware of, a world constructed out of the ‘this’es: out of the yellow of which she was aware as she stared at the cigarette packet in front of her, and of more things like it. We need then to imagine her, sitting in Wittgenstein’s classes, hearing his discussion of the ostensive definitions we may think we give ourselves. What he says seems to allow no place for the thises of which she is aware. If he says that words for the colors things are are public words, not words we define by concentrating on a this, then it seems that what is there, given his understanding of the world, cannot be this. But it is this, blue, or this, yellow, that is what she still wants to say is there. Take away the ostensive definition she gives herself and the thises that make, or seem to make, those definitions possible, and you take away the character of the world as she is aware of it. You take away what she wants to say is there.
In response to her expression of that idea, Wittgenstein asked her to suppose that we had the word ‘painy’ as a word for the property of some surfaces.7 This ‘medicine’, Anscombe says, was effective. She did not (before or after the ‘medicine’) think of ‘blue’ as the name of this sensation that she was having; and Wittgenstein’s suggestion of a word working as a secondary-quality word for surfaces with a property through which they caused pain did not lead her to the idea that, so far as she was inclined to say ‘Blue is there’ she might equally be inclined to say ‘Painy is there’. Quite the reverse. She had no inclination to say ‘Painy is there’; and she could see the contrast clearly between a word like ‘painy’ and a color word, like ‘blue’. Before the medicine, it had seemed that, if one were dissatisfied with Lockean realism, and did not take blue-as-one-was-aware-of-it to be something internal in contrast to the ‘colorless’ external world, one would have to ask whether blue-as-one-was-aware-of-it was a part of the surface of things, or one of the items out of which the external world was composed, or something else again. One would focus on what one was aware of, and ask about it. The clarity produced by Wittgenstein’s suggestion lay in the capacity of the example to make the Lockean question disappear, the question where blue, this, really is. The question arises out of a kind of unclarity. ‘Blue’ is not like ‘pain’/’painy’, but Lockean realism gets its convincingness from that contrast being out of sight. ‘Painy’, as a secondary quality word, works just fine; but it works as such a word precisely because ‘pain’ is not a word like ‘blue’ but a word for what we feel. If ‘painy’ (for surfaces) together with ‘pain’ (for what we feel when we come in contact with a painy surface) is our model of how secondary quality words work, ‘blue’ is not a secondary quality word. It appears to us, though, as we move into the Lockean quagmire, that, if there can be a word for those features of blue things that make them look the way they do to us, then what else there is to blue must be purely a sort of given. When we are gripped by this idea, there appears to be a question where blue as this-we-are-aware-of really is. Anscombe rejected the idea of it as purely internal, but the only alternative (before the medicine) appeared to be that it was somehow ‘out there’. A recognition (as in Anscombe’s case) that there is no need to say ‘Painy is there’ may help reveal the contrast between ‘pain’ and ‘blue’, and the way a not-fully-thought-through analogy between the two falsifies our thought.
1 It is clear in Anscombe’s account of the ‘medicine’ that it did not have to work. She ascribes to Wittgenstein the idea that it was the right medicine for her; it would not have been the appropriate medicine for someone who had taken ‘blue’ to be a word for a sensation, or who moved to that view in response to Wittgenstein’s suggestion. If I speak of the capacity of the example to make the Lockean problem lose its apparent inevitability, I don’t mean to suggest that it would do so for everyone.
2 Wittgenstein provides various metaphors to describe the kind of transformation of thought that can be effected by such ‘medicine’. One is the metaphor of the condensation of a cloud of philosophy into a drop of grammar. Another, already referred to, is the metaphor of seeing philosophy in cross-wise strips. I want to say more about how that metaphor connects with the Anscombe case.
Before the ‘medicine’, Anscombe’s problem is one of philosophy’s Big Questions. It is a form of the question how our thought is able to connect with reality. She is aware of, has in her mind, this, the blue; is it or is it not there, in the world? There are various Big Answers one can give to this Question. Wittgenstein invites her to think about ‘painy’; one might ask whether it is there or not, but she has no inclination to ask or answer that question. The question she was asking, about where the blue is, loses its capacity to puzzle her, as she comes to see clearly the contrast between ‘blue’ and ‘pain’/’painy’. Her problem about color is resolved; it is at the same time seen to be one particular problem. The problem had been irresolvable when it was seen as Big Question: for that pointed her attention away from the details of the use of words. The medicine makes possible attention to details that are capable of helping her reach a satisfactory resolution of the problem; but also makes possible the re-conception of the problem as a particular problem.
Wittgenstein says in §92 of Philosophical Investigations that in his investigations he is concerned with the essence of language. But he says that it is possible to be concerned with that question in different ways. There is the Big Question way, that asks ‘What is language?’, ‘What is a proposition?’; and there is an investigation which transforms such questions into cross-wise strips, which comes at them over and again from different directions, and reaches satisfaction not ‘ein für allemal’ but for the case of this and that particular problem. In reading Wittgenstein, the attempt to put together an overall account of the nature of language goes against precisely that idea of philosophical method. So I am suggesting this way of thinking of clarification, as understood in Wittgenstein’s later thought: Wittgenstein clarified part of the grammar of ‘blue’, as a way of dealing with one particular problem, where the clarification itself included allowing the problem to be seen as a particular problem. So, too, parts of the grammar of ‘proposition’ (say) or ‘language’ may be clarified, in response to particular problems, where the clarification will include allowing us to see our problem as a particular problem, not as the problem, not as an infinitely long lengthwise strip.
In this section I want to try to clarify the conception of clarification which I have sketched by considering an example of what it is like to do something different, something which will fail to clarify by providing or attempting to provide a lengthwise strip of philosophy.
Here, then, is my example. It’s meant as an example of reading Wittgenstein without taking in the significance of the contrast between lengthwise and crosswise philosophy. Wittgenstein, so it may be suggested, gives us an account of sense as constituted by grammar. Grammar is different in different language-games; so (reading Wittgenstein as holding that sense is constituted by grammar), it follows that what is said in one language-game cannot mean the same as what is said in a language-game in which the rules of grammar are somewhat different. What we might take to be a case of people within one language-game contradicting people in another cannot literally be contradiction.8
Someone who reads Wittgenstein that way thinks that we should look within particular language-games to see the sense of what is said in them, but does not think that that applies to words like ‘sense’ or ‘meaning’ or ‘contradiction’. That is, if in our ordinary talk about who means the same as someone, or about who contradicts someone, and so on, we actually accept that people whose modes of speech have different rules may mean the same, or may contradict each other, that doesn’t count. There is (supposedly) a deeper sense of meaning the same, or of contradiction, or a more literal sense. That is, if grammar constitutes sense, we can gather from that insight a more fundamental understanding of sameness and difference of sense than might be got by what would be a merely slavish adherence to the vagaries of ordinary language. Wittgenstein himself may sometimes seem to recommend such an attitude, but, in giving an account of his overall view of language, we should not follow him.
But the issue here is distorted if seen in terms of whether we can be too slavish in adhering to ordinary language. Consider the ways in which people dispute, or seem to be disputing, whether Jesus is the Messiah. On the face of it, it looks as if Jews and Christians contradict each other about this. ‘Jesus is the fulfilment of these texts’, ‘Jesus is not the promised fulfilment’. Jews and Christians may point to the same texts to explain what they mean by being the Messiah, but they do not use the same criteria for settling whether someone is or isn’t the Messiah. So, if someone holds that criteria for settling an issue belong to grammar, and if grammar is supposed to constitute sense, Jews and Christians cannot mean the same by ‘Messiah’ and are not literally contradicting each other about who is Messiah. Here, the concepts of meaning and not meaning the same, and of contradicting someone, are given a comparatively simple grammar, so that whether or not two people are contradicting each other can be read off from the rules for the use of a word, here ‘Messiah’. If the rules in the respective language-games are different, the two people don’t mean the same, and aren’t contradicting each other. But Wittgenstein thought that, in philosophy, we are frequently inclined to just that sort of simplification; we don’t see how complicated the use of our words is. The way in which we learn to use expressions like ‘mean the same’ and ‘contradict’ is extremely complicated, is extremely difficult to describe; and in philosophy we may let ourselves off from seeing that complication. The case of the dispute about the Messiah is one part of the complicated story here, and should be looked at on its own. What is interesting about it is precisely that it is like some cases that we may have in the forefront of our minds when we think about whether people mean the same by their words, but unlike other cases. The complicated grammar of ‘contradict’, of the ways we use that term, includes this sort of case. If we see this case, with its resemblances and differences from other cases in which we say that such-and-such people deny what others assert, that one group contradicts the other, there is not here some further question whether Jews and Christians ‘literally’ contradict each other. If we look at the cross-strip, the uses of terms like ‘contradict’ in connection with a variety of circumstances including this particular case, the philosophical question may disappear; there would be no remaining puzzlement; we see this case in its relation to others. This isn’t to say that there is a foolproof way of achieving such a reconception of the problem, to make it become a question about the grammar of ‘contradict’ and its complications in such cases. We may go on wanting to insist (say) that there is no literal contradiction in such cases; but that insistence reflects the idea of contradiction as not something open to view. Contradiction is seen as a kind of logical doing, fundamentally as one thing which is either there or not. Whether one person is asserting that which the other is denying depends on sameness of sense, which depends upon what determines sense, about which there is a general account, to which the ways we actually use words like ‘contradict’ are irrelevant, and needn’t be attended to. The infinite lengthwise strip is, in this case, the question how sense is determined. Here the focus on a Big Question is tied to a falsification of the use of our words, of what complicated things we do with them.9
The example illustrates how Wittgenstein is concerned with ordinary language–not, that is, as a slavish devotee of leaving it as it is. Rather, the refusal to attend to what we say frequently shows a kind of misconception of our problems themselves, a misconception which keeps us turned away from the unlocked door, and which makes it impossible for us to achieve clarity. The point of attention to ordinary language is not that it is in any way sacred; rather, attention to ordinary language is capable of transforming our problems and our sense of where we are with them.
I have argued that we cannot succeed in reaching genuine philosophical clarity, as Wittgenstein understands it, so long as we take philosophy, or try to, in lengthwise strips. If we are not able to transform our understanding of our problems, we are locked into unsettlable disputes. This comes out, for example, in the interminable dispute about whether a person isolated from birth could possibly have a language, and what Wittgenstein’s view of the matter is. One side holds that, so long as the speaker is conceived as using words the meanings of which are public, there is nothing in Wittgenstein that disallows the possibility of such a case, even if the speaker were isolated from birth; the other side holds that, apart from a speaker’s relations with other speakers there can be no such thing as his or her meaning anything determinate by sounds or written marks. The focus of the dispute is on a lengthwise strip of philosophy, the possibility of meaning anything by anything uttered or written. There are supposed to be general answers to that Big Question to be found in Wittgenstein; but obviously whichever Answer one chose to ascribe to him would be highly disputable. How could the question be turned into one which draws us back to crosswise strips of philosophy? What would that be like? What is the work of clarification really like here? I’ve suggested that the real work of clarification is whatever makes possible the move to the crosswise strip, the transformation of the question itself. (In this particular case, the transformation may be effected by considering in detail the character of the kind of case we are asked to imagine; for it is no good saying that we are to imagine someone whose words have ‘public’ meanings. What does the person do, what exactly (in the absence of other people) shows the shareableness of what she is doing? My word ‘tree’ has a public meaning that can be seen in how I learned it, and how I use it with other people. If we are to imagine someone who has a word supposedly meaning ‘tree’, but who has not learned words from others and who does not use the words with others, where exactly, in what she does, am I supposed to see what the shareableness of what she says comes to? To describe this in detail would be to describe a case which is, in many ways, unlike our own in what we see as showing the shareableness of what is said; so whatever we may adduce by way of resemblances, the case will be one which is connected with ours as much and as deeply by contrast as by resemblance. Such attention to the description of the case may transform the question, so it is no longer about the Conditions for Meaning Something By Our Words.)10
I gave as an example in Section 8 the case of someone who ascribes to Wittgenstein a general account of the relation between sense and grammar, and who takes it to be a consequence of the views ascribed to Wittgenstein that, if the rules of two language-games are different, the people playing one game cannot literally contradict the people playing the other game. The picture there, of contradiction as a sort of logical reality that may or may not be present, is tied in two ways to the mode of thought of the Tractatus. In the first place, the picture is present in the Tractatus, and is indeed part of what remains behind, influencing Wittgenstein’s conception of the activity of philosophical clarification, even after the problems of the book are supposedly überwunden. More significantly, though, it reflects an approach to philosophy itself that Wittgenstein came to see made impossible genuine philosophical satisfaction. At 3.3421, Wittgenstein says that particular cases have no significance for philosophy, except so far as the possibility of such-and-such individual case can open up for us something of the essence of the world. Grasping the essence of the world is, here, a matter of getting hold of a lengthwise strip. And individual cases can matter, on this understanding of philosophy, only so far as they enable us to get hold of the lengthwise strip.
It’s not news that Wittgenstein took particular cases seriously in his later thought in a way he had not earlier. I don’t think it is easy to see, though, what the change actually comes to, what its significance is, and how it is connected with his changed conception of philosophy itself. It is easy to read the emphasis on attention to particulars as itself reflective of an Answer to a Big Question. We may be directed to look at the language-games we play with some word, with a view to showing us that our philosophical uses violate the rules for that word, or project the rules of one language-game into another, and so on. And that then is supposed to imply, given some general conception of meaning that we ascribe to Wittgenstein, that we had been talking nonsense. Here, attention to the particularities of this or that language-game is supposed to be the method Wittgenstein teaches us, but what is meant here by ‘attention to particulars’ has no tie to the idea that our problems can be transformed if we are able to take them as problems on their own, without connection to a Big Question. The method of attention to language-games, understood as dictated by an Answer to a Big Question, e.g. about the conditions of sense, I shall call the dogmatic method. Dogmatism enters it in two ways: it is dogmatic in its reliance on an Answer to a Big Question; it is dogmatic in its way of telling people that what they are inclined to say can be shown to be nonsensical because of the supposed failure to play the language-game right. So, I want to argue here that there are at least two piecemeal approaches that are not the later Wittgenstein’s: one is the piecemeal approach of the Tractatus, in the shadow of a picture of language arrived at by working through a Big Question, but not having got genuinely past it. The other is the piecemeal approach of the dogmatic method of language-games, in the shadow of a different Answer to the same Big Question.
At Investigations §122, Wittgenstein emphasizes the importance of finding and inventing intermediate cases. The German is Zwischenglieder and has (much more than the English) the idea of links. We want cases that link up with ours, that connect with ours but are different, and that take away from the case we are considering its capacity to send us in search of something deeper and hidden from view. The importance of language-games partly like ours, partly unlike ours, the importance of seeing our language-games as neighbored, as having nearby neighbors and neighbors very much further off, is that such neighbors let us see what we do. Wittgenstein says that what he aims at is a perspicuous presentation of the use of our words. Perspicuity here means: we can look at it, we can take it in philosophically, can find ourselves again in that use of words, can see the problem we had had as one problem, a cross-wise strip.11 Here there is the idea of a piecemeal approach which is genuinely not in the shadow of Big Questions.
In the Preface to the Investigations, Wittgenstein writes about his struggle to force his thoughts into a ‘natural order’, without breaks. He uses the first person singular in describing the struggle and his abandonment of it. He then switches to the plural when he notes that it belongs to ‘the very nature of the investigation’ that it compels us to travel over a wide field of thought criss-cross in every direction. (It wasn’t just that he hadn’t been able to do anything else.) His own struggle gives an indication of the difficulty here. He had written of the importance of philosophy in ‘crosswise strips’ as early as 1933, but the attempts he went on making to impose an order on his thoughts show there to have been a continuing tension within his understanding of his own philosophical activity.12 He had had to learn from the criss-cross character of the investigations as he was actually engaged in them and the criss-cross path that the investigations took, from the fact that crisscross was the only way that he could operate without crippling his thought, that criss-cross was not a failure or fault, not the best he could do where someone else might do better, or where there was a better that could be conceived of but that was unfortunately out of reach. He was able to break off the struggle when he saw that the idea of something better than his crisscross sketches reflected a false conception of the nature of the investigation, of what it came to to attend to particular problems, this one and that one, and what it came to to write a book that showed the progress of such an investigation. What matters–for the path of the Investigations as much as for the path of our investigations–is the kind of clarity that can come from going criss-cross. I close with a quotation from Simon Glendinning’s ‘rewriting’ of part of §133, which helps brings that out.
We always say: ‘The real help here would be a map that would take us from A to B in a direct line–the one that misses out the mountains and forests.’–Instead, I will take you another way. I have no such map and we will stick to natural paths. But we should be able to take some breaks along the way.–And one of the things we will do, each time we get going along, is (in various ways) try to rid ourselves of the idea that what we really need is a map that takes us from A to B.
(Glendinning 2004: 162; cf. Glendinning 2002: 76, n. 3)13
1. I wasn’t aware of the importance of this argument when I wrote Diamond (1988) or (1995).
2. Some critics of Wittgenstein suggest that he made too much of the criticical term ‘nonsensical’. But this criticism of Wittgenstein rests on the idea that he calls certain sentences nonsensical because they fail to measure up to some special standard or other. The criticism depends upon ascribing to Wittgenstein some special notion of nonsensicality, over and above the ordinary idea of having, perhaps unwittingly, failed to say anything. We can fail to think through what we take ourselves to be saying; if we had, or had tried to, we might have recognized an incoherence in our intentions. We might, e.g. not have decided whether to mean one thing or another by some word; we might have left it unsettled which meaning we wanted, although a clear look at both possibilities would have shown us we did not really want to come down for either, or for anything else. This sort of phenomenon isn’t being labelled as ‘saying nothing’ on account of some Tractatus doctrine about nonsensicality. The Tractatus point is rather that the activity of attempted philosophical clarification can reveal to us that we have got into that sort of position.
3. For a reading of the Tractatus which takes an opposed view of logical syntax, see Hacker (2003). I discuss the issues raised by Hacker in Diamond (in preparation).
4. I did not make this point sufficiently clear in Introduction II to Diamond (1991: 13–39). I did not there anticipate possible misunderstandings, and what I wrote has been taken to mean that a ‘resolute’ interpretation of the Tractatus should not be extended to (some or all of) Wittgenstein’s propositions about the nature of language. The issue of a ‘resolute’ interpretation of the book should be separated sharply from the question whether Wittgenstein had genuinely freed himself from metaphysics in his thought about language. In fact, I think that propositions like ‘Propositions are truth-functions of elementary propositions’ are particularly good examples of propositions we are meant to be able to recognize to be plainly nonsensical, to have no content, speakable or unspeakable. It is nevertheless arguable that they are also excellent examples of how Wittgenstein had not freed himself from a metaphysical conception of language.
5. For a very interesting treatment of the issues in this paragraph, see Schulte (2002).
6. See the ‘Introduction’ to Anscombe (1981: vii–x).
7. Compare PI §312. Here, too, Wittgenstein invites us to consider the possibility of our distinguishing certain surfaces which have patches that produce pain when we touch them.
8. The case, as I discuss it, is not meant to correspond to any specific reading, but my description was guided in some respects by Michael Forster’s reading in Forster (unpublished).
9. See also Wolfgang Freitag’s discussion of the issues here, in Freitag (B.Phil thesis, Oxford University). He takes the reflexive character of philosophy as Wittgenstein understood it to be central, where reflexivity would (in the case I was discussing) involve treating ‘same sense’, ‘contradict’ and so on, not as terms whose application was determined by a kind of meta-account but as terms of our language, like any others, the grammar of which we may need to attend to. I should also mention that the role I have given above to complicatedness is meant to pick up on ideas which are especially prominent in Wittgenstein’s Last Writings on the Philosophy of Psychology, vol. II. I have discussed some of the issues in this paragraph in Diamond (1999) and Diamond (forthcoming).
10. For a further discussion of the issues here, see Diamond (1989).
11. This reading of Wittgenstein on perspicuity is, in some respects, close to that of G.P. Baker, criticized by P.M.S. Hacker in Hacker (2001: 346). Hacker dismisses all such readings of Wittgenstein on perspicuity by appeal to Wittgenstein’s writings and lectures of the early 1930s. For a discussion of this view of the relevance of Wittgenstein’s thinking about method in the early 1930s to his understanding of philosophy in the Investigations, see Schulte (2002).
12. Again, see Schulte (2002) on the various strands in Wittgenstein’s conception of his ‘method’.
13. This paper was read at the conference on Wittgenstein’s later philosophical methods, in Venice in September 2002. I am very grateful for the helpful comments and discussion on that occasion. I am also grateful for comments and suggestions from James Conant.
Anscombe, G.E.M. (1981) ‘Metaphysics and the Philosophy of Mind’, Collected Papers, vol. II, Oxford: Blackwell.
Diamond, C. (1988) ‘Throwing Away the Ladder’, Philosophy 63 (1988), 5–27.
Diamond, C. (1989) ‘Rules: Looking in the Right Place’, in D.Z. Phillips and P. Winch (eds) Wittgenstein: Attention to Particulars, Basingstoke: Macmillan.
Diamond, C. (1991) The Realistic Spirit, Cambridge, MA: MIT Press.
Diamond, C. (1995) ‘Wittgenstein, Ludwig Joseph Johann’, in J. Kim and E. Sosa (eds) A Companion to Metaphysics, Oxford: Blackwell.
Diamond, C. (1999) ‘How Old are the Bones?’, Proceedings of the Aristotelian Society, supp. vol. 73, 99–134.
Diamond, C. (in preparation) ‘Logical Syntax in Wittgenstein’s Tractatus’.
Diamond, C. (forthcoming) ‘Wittgenstein on Religious Belief’, in D.Z. Phillips (ed.) Religion and Wittgenstein’s Legacy, Aldershot: Ashgate.
Forster, M. (unpublished) ‘Wittgenstein’s Later Philosophy: The Question of the “Arbitrariness” of Grammar’.
Freitag, W. (unpublished) ‘Truisms and Altruisms: Some Reflections on Wittgenstein and the Private Language Argument’ (B.Phil. thesis, Oxford University).
Gerrard, S. (2002) ‘One Wittgenstein’, in Erich Reck (ed.) From Frege to Wittgenstein: Perspectives on Early Analytic Philosophy, New York: Oxford University Press.
Glenndinning, S. (2002) ‘Wittgenstein’s Apocalyptic Librarian’, in R. Haller and K. Puhl (eds) Wittgenstein and the Future of Philosophy. A Reassessment after 50 years/Wittgenstein und die Zukunft der Philosophie. Ein Neubewertung nach 50 Jahren, Vienna: öbv & hpt.
Glenndinning, S. (2004) ‘Wittgenstein’s Nomadism’, in H. Carel and D. Gomez (eds): What Philosophy Is, London: Continuum.
Hacker, P.M.S. (2001) ‘Philosophy’, in H.J. Glock (ed.) Wittgenstein: A Critical Reader, Oxford: Blackwell.
Hacker, P.M.S. (2003) ‘Wittgenstein, Carnap and the New American Wittgensteinians’, Philosophical Quarterly 53, 1–23.
Schulte, J. (2002) ‘Wittgenstein’s Method’, in R. Haller and K. Puhl (eds) Wittgenstein and the Future of Philosophy. A Reassessment after 50 years/Wittgenstein und die Zukunft der Philosophie. Ein Neubewertung nach 50 Jahren, Vienna: bv & hpt.