Have Wittgensteinians Found A Way Of Diagnosing Nonsense?

Ian Dearden 21/10/2014 11:52 AM Have Wittgensteinians Found A Way Of Diagnosing Nonsense? At Tractatus 6.53 Wittgenstein writes: The correct method in philosophy would really be the following: to say nothing except what can be said, i.e. propositions of natural science – i.e. something that has nothing to do with philosophy – and then, whenever someone else wanted to say something metaphysical, to demonstrate to him that he had failed to give a meaning to certain signs in his propositions. Although it would not be satisfying to the other person – he would not have the feeling that we were teaching him philosophy – this method would be the only strictly correct one. There have been some pretty radical interpretations of the Tractatus but, so far as I am aware, no one has claimed that Wittgenstein actually used the ‘strictly correct’ method in that book. The question arises whether he did so in his later work. As does the question whether he could have done so. And whether anyone could have done so. Is there such a thing as demonstrating to someone that he has ‘failed to give a meaning to certain signs in his propositions’? Let us note at the outset that Wittgenstein’s formulation allows for the fact that people can use words deviantly and introduce neologisms. The question is whether the would-be metaphysician means anything by his terms, not whether the language as it now is assigns them a meaning or, if it does, whether he is using them correctly. Can one show that someone means nothing by his words? 1 Ian Dearden 21/10/2014 11:52 AM SECTION I Criticising An Utterance Without Assigning It A Meaning: The Obvious Problem At first sight it seems that in order to criticise an utterance, one must attribute some meaning to it. One could express this as a dilemma or fork. Either one ascribes some meaning to what is said, in which case one can’t then go on to say it is meaningless, or one doesn’t (perhaps it is ungrammatical or contains neologisms), in which case the most one can say in criticism is ‘I don’t understand’ and not a great deal could be concluded from a mere failure to understand. Dilemmas like this furnish a great test of philosophers’ ingenuity: there is nothing they like better than wrestling with such problems. But I want to suggest that this particular dilemma is as bad as it seems and that no amount of ingenuity will enable one to evade it. I would concede1 that a failure to understand can be expressed as a criticism – ‘That’s simply gobbledygook’ – but one has not proved anything by saying this, certainly not that the utterer means nothing, still less that he means nothing but thinks he means something. ‘The Sense That Is Senseless’ Consider this well-known and frequently quoted remark in the Investigations: When a sentence is called senseless, it is not as it were the sense that is senseless. But a combination of words is being excluded from the language, withdrawn from circulation. (PI500) What might be the point of this observation? Someone might, I suppose, think that one had to understand a ‘sentence’ in order to recognise it as senseless. This would indeed be an error. (Contrast: in order to recognise it as false). Perhaps Wittgenstein is making this point.2 But I feel sure that there is more to the remark than this. Another thing one might find in it is what one might call the ‘Problem of Specification’, a problem that goes back to the Tractatus. Suppose I say of a locution that it is nonsense. Assuming I am right, have I done more than point out a trivial linguistic fact? After all, there is nothing about any sequence of sounds or marks on paper that prevents it from being given a meaning. I wanted to say something non-trivial about a particular combination of sounds or marks. Now it is no good my saying that the locution is nonsense when it has the meaning it has in, say, English. For if it is nonsense, it has no meaning in English. I seem to be trying to specify a 1 I make this concession once and for all here and shall not repeat it. See P.M.S. Hacker, Analytical Commentary on ‘The Philosophical Investigations’, Volume IV, Blackwell, 1996, p283. 2 2 Ian Dearden 21/10/2014 11:52 AM particular something and point out that it is nonsense, to rule something out. But what? No one bothers to point out that ‘The the at of’’ is nonsense.3 A passage in Philosophical Grammar (p. 130) sheds some light on this. We say of a locution that: these words are senseless. But it isn’t as it were their sense that is senseless; they are excluded from our language like some arbitrary noise, and the reason for their explicit exclusion can only be that we are tempted to confuse them with a sentence of our language. The point then is that in philosophy we often have to say that particular combinations of words do not constitute meaningful sentences in our language, and the combinations in question are ones that philosophers have confused or are likely to confuse with genuinely meaningful ones. Somehow this must be done without making it look as though it is the meaning of these combinations that is meaningless. There is a third point that Wittgenstein might have been making, though I am not sure of this. The Problem of Diagnosis If one cannot say of a senseless string of words that it is its sense that is senseless, then surely this has an obvious corollary, namely that if someone does not mean anything by what he says, one cannot say that what he means by it is meaningless. Perhaps Wittgenstein had this personalised corollary in mind. I do not know that he anywhere formulated it explicitly. And of course a clear recognition of it leads straight to the dilemma propounded above. Let us call it the ‘Problem of Diagnosis’. How do you show that there is no meaning behind someone’s words? What do you get to work on? Wittgenstein, in his later philosophy, certainly paid more attention to those who utter alleged nonsense than he did in the Tractatus. In the early philosophy (apart from the occasional remark such as 6.53) one gets the impression, as one does with the Logical Positivists, that the problematic locutions just occur. It is as though one could simply assess sounds or marks on paper for their meaningfulness without reference to the fact that they are produced by human beings. The later philosophy puts human beings, the philosophers who supposedly talk nonsense, at the centre of the stage.4 There is much about how such things as mental pictures and false grammatical analogies mislead the philosopher into thinking he is producing meaningful utterances when he is not. So one might expect to find a clear formulation of the Problem of Diagnosis. Yet I do not know of one. Still less do I know of a solution to it – in Wittgenstein or anyone else. 3 Here I follow Kenny’s exposition of Wittgenstein’s criticism of Russell’s Theory of Types in Wittgenstein, Pelican, 1975, pp 43 - 4 4 He does not often mention other philosophers by name, it is true, but he does keep in mind the fact that philosophical utterances are the utterances of philosophers. 3 Ian Dearden 21/10/2014 11:52 AM I can imagine followers of Wittgenstein, and perhaps others, thinking that there must be some solution latent in his work: even if only by implication he must have shown how it is possible to get to grips with someone’s utterance and demonstrate that it is devoid of meaning. What was he doing in the later philosophy, it might be asked, if not addressing this problem? Yet the sheer difficulty of imagining what a solution – any solution, perhaps an unWittgensteinian one – would look like has to be weighed against this. I think that the piecemeal approach of the later Wittgenstein may have helped conceal the difficulty. If one starts with a general theory of what can be said, as in the Tractatus (on the usual interpretation – see below), or with a general criterion of meaningfulness, as in Logical Positivism, it soon becomes evident that there is going to be a difficulty in applying the theory or the criterion to show that a certain string of sounds or marks lacks meaning. How and to what does one apply it? How is one to bring it to bear upon the utterance? But if one takes a case-by-case approach, it might seem that there could be a way of showing that a particular philosopher on a particular occasion had unwittingly deprived his own words of meaning. The later Wittgenstein’s attention to the fact that alleged philosophical nonsense is produced by human beings together with his piecemeal approach might suggest that it is at least possible that he had found a way of getting behind someone’s utterance to show that there is no meaning there. And yet these aspects of his later work have no magical qualities. The problem does not disappear. Suppose that one is in the presence of a philosopher who has just said ‘something metaphysical’. How does one show that he has ‘failed to give a meaning to certain signs in his propositions’? One must somehow engage with his utterance, but not by attributing a meaning to it. Deliberately Talking Nonsense It is of course usually assumed in philosophical discussions of nonsense that a person accused of talking nonsense is not deliberately doing so. It is assumed that he thinks he means something. But it is easy to forget that our dilemma, the Problem of Diagnosis, arises even if one suspects that the other is only pretending to talk meaningfully. Since Hegel, accusations of outright charlatanry have not been infrequent in philosophy and one can think of a number of recent Continental writers whom it is tempting to dismiss in this way. But a moment’s thought shows that any attempt to prove that Hegel, say, did not mean anything by a particular passage in his works would meet the same difficulty whether or not one assumed he thought he meant something.5 The problem however is nicely illustrated by something one might feel is nearer home: a dispute about the interpretation of the early Wittgenstein. Until recently it had been assumed that in the Tractatus Wittgenstein attempted to convey insights about the relationship between language and the 5 One problem with claiming that Hegel, say, was deliberately talking nonsense would be that of giving an account of those who claim to understand Hegel, especially perhaps those, such as Kierkegaard and Marx, who claim to understand him yet disagree with him. 4 Ian Dearden 21/10/2014 11:52 AM world which strictly cannot be expressed in language at all. Consider the famous self-condemnation at 6.54: My propositions serve as elucidations in the following way: anyone who understands me eventually recognises them as nonsensical, when he has used them – as steps – to climb up beyond them. (He must, so to speak, throw away the ladder after he has climbed up it.) He must transcend these propositions and then he will see the world aright. This is usually taken to imply that by means of the nonsensical ‘propositions’ Wittgenstein does try to convey something but not in the way that one conveys something by ordinary meaningful propositions, that is by depiction. Certainly discomfort has been felt with this idea, but it has been assumed that if we are to do justice to the Tractatus, we must see it in this way. Wittgenstein once said to Anscombe that it was not all wrong, not like a bag of junk professing to be a clock but like a clock that did not tell you the right time.6 Recently, however, some commentators have suggested that the Tractatus – or the bulk of it - really is just nonsense and intended as such. Look again at the above passage. Wittgenstein says: My propositions serve as elucidations in the following way: anyone who understands me eventually recognises them as nonsensical… (emphases added). He does not say ‘anyone who understands them’. So perhaps one cannot assume that it is possible to understand them in any sense. Perhaps they are sheer nonsense, so that anyone who thinks he finds a meaning in them has gone wrong somewhere. He has misunderstood, but it is Wittgenstein he has misunderstood, not the ‘propositions’, for there is nothing there to be understood.7 This is not a straightforward dispute between people who think something makes sense and people who think it does not. Both sides agree that the ‘propositions’ of the Tractatus are in some sense nonsense. But one side sees them as illuminating nonsense, which has to be credited with something like meaning (though just what it is and what it should be called is not entirely clear). And the other side denies this: almost all the ‘propositions’ are nonsense pure and simple. How might this dispute be resolved? Those who argue for the sheernonsense view seem to rely on certain framing remarks, remarks which tell us how the book is to be read, such as 6.54. Rather awkwardly, not all these occur in the preface or at the end of the book. Some, such as 5.4733, occur 6 G.E.M. Anscombe, An Introduction To Wittgenstein’s Tractatus, Hutchinson, 1967, p78 Those who take this line include Cora Diamond, James Conant and a number of the contributors to The New Wittgenstein, ed. Alice Crary and Rupert Read, Routledge, 2000. 7 5 Ian Dearden 21/10/2014 11:52 AM in the body of the book – which rather spoils the ‘framing’ analogy. Behind the sheer-nonsense view there is no doubt a desire to maximise Wittgenstein’s consistency: a thinker of his calibre, it is felt, would not have equivocated over whether something is nonsense in the way that the traditional interpretation of the Tractatus seems to require. Supporters of the traditional ineffable–insights view have perhaps more that they can appeal to.8 There is the difficulty about framing. There are remarks Wittgenstein made about the Tractatus to others, such as the one to Anscombe mentioned above.9 There is the fact that some things in it seem genuinely illuminating and have found a place in modern logic. And there is the negative point that Wittgenstein never seems to have said or written anything apart from the framing remarks that lends support to the sheernonsense view. Certainly the criticisms of the Tractatus in the Investigations read like criticisms of a work that was at least intended to convey something. But what the dispute surely illustrates is that even deliberate nonsense can be hard to unmask. We have here a conflict between an ‘interpretation’ of a work that sees it as almost entirely deliberate nonsense and one that sees it as an attempt to convey something, even though this is by means of what is strictly nonsense. If the dispute is ever definitively resolved – not something I would predict with any confidence10 - it will almost certainly be in large part by attention to what Wittgenstein says about the Tractatus (framing remarks, remarks to others etc.) Imagine that a diary were found in which Wittgenstein confessed to producing a work that was sheer nonsense, perhaps in order to mock the less astute members of the philosophical community. Alternatively suppose that in it he expressed the worry that philosophers might take it to be sheer nonsense and so miss the insights he was trying to convey. Perhaps nothing less than discoveries such as these would settle the question decisively. The Scope Of My Discussion In order to keep this article to a manageable length I shall restrict its scope in certain ways and make a number of assumptions. a) Of those who make use of the notion of philosophical nonsense (henceforth to be called ‘nonsensicalists’) I shall largely confine myself to Wittgenstein and his followers, though something will be said of the Logical 8 See P.M. Hacker, ‘Was he trying to whistle it?’ in The New Wittgenstein, Crary and Read (ed), op. cit. Note also that Wittgenstein once said that he regarded the Tractatus as the only alternative to his later work. (Norman Malcolm, Ludwig Wittgenstein – A Memoir, O.U.P., 1958, p.69) 10 I express no opinion as to who is right, but I will make one comment. The difficulties that confront Wittgenstein’s interpreters here might be a reflexion of Wittgenstein’s own difficulties. It is the burden of this article that it is not clear that there is any way of condemning the utterances of others as nonsense that is not radically self-defeating. It would not be entirely surprising if condemning one’s own utterances as nonsense resulted in confusion and paradox. 9 6 Ian Dearden 21/10/2014 11:52 AM Positivists. As it happens, most accusations of talking nonsense that one hears in philosophy today seem to come from followers of the later Wittgenstein.11 b) I shall mainly confine myself to alleged philosophical nonsense. The possibility that the error of thinking there is something one means when there is not might occur outside philosophy, in dreams for example or insanity, will only be touched upon. c) I shall assume that illusions of meaning (IOMs) are possible, that is that it is possible for someone to believe that there is something he means by what he says when there is not. I would emphasise however that I make this assumption only for the sake of argument. d) I shall assume that one cannot mean a nonsense, that what one means cannot be meaningless. e) I shall assume for the reasons already outlined that no criterion of meaningfulness will work. If there is a way of diagnosing philosophical nonsense, it will have to be more subtle that the mere application of a criterion.12 One further point should be borne in mind. Even if the attempt to show that someone is wrong to think he means anything by an utterance fails, this does not mean that all is well with that utterance. There might be something of substance in the reasons given for calling it nonsense. It may be, for example, that in certain circumstances unverifiability is a defect in a claim, even if one cannot argue from unverifiability to meaninglessness.13 Reactions To The Dilemma I have the impression that many philosophers do not take the problem of Diagnosis very seriously or appreciate its generality.14 Take the case of verificationism. It was pointed out against the verificationism of the Logical Positivists that one has to understand a claim, ascribe some meaning to it, in order to assess it for verifiability. The conclusion that it is unverifiable can 11 My impression is that most contemporary philosophers in the Anglo-American tradition believe that other philosophers sometimes talk nonsense but that only Wittgensteinians (and perhaps not all of them) regard philosophy as being mainly nonsense. It would be interesting if someone were to do a survey on this. I would also like to know what proportion of philosophers can recall being accused of talking nonsense themselves. 12 For a discussion of the topics excluded and the assumptions made see my Do Philosophers Talk Nonsense? - An Inquiry into the Possibility of Illusions of Meaning, Teller Press, 2005. 13 I discuss another example illustrating this possibility in Dearden, op. cit., pp. 119-20. See also what the Wittgensteinian Lars Hertzberg has to say on pp. 14-15 of ‘The sense is where you find it’ at web.abo.fi/fak/hf/filosofi/…/The_Sense_Is_ Where_You_Find_It.pdf. The admonition ‘That wouldn’t make sense’, according to him, ‘is a mark of philosophical impatience’. 14 I do not really know how widespread the recognition of the Problem of Diagnosis is, and some wavering on this point may perhaps be detectable in this article. When I come to discuss the ‘austere’ view of nonsense and what I call ‘therapeutic nonsensicalism’ I treat them as contributions to a discussion of the problem. 7 Ian Dearden 21/10/2014 11:52 AM only be attained by attention to what it means.15 One would have thought that this objection would have conclusively disposed of verificationism. Yet is has received less attention that other objections which are inconclusive or worse. The difficulty of formulating the Verification Principle so as to exclude metaphysics without also excluding scientific laws is a real one, but by its very nature is an inconclusive objection – has anyone ever proved that a sufficiently complicated version of the Principle could not do the trick?16 The objection that the Verification Principle is self-defeating because it is not itself verifiable is based on a misunderstanding. According to the Principle a claim is meaningful if it is verifiable or analytic. Clearly the Logical Positivists could have claimed that the Principle itself was analytic (wrongly, I believe, but that does not restore relevance to the objection that it is self-defeating).17 In point of fact, they preferred to reinterpret it as a methodological recommendation. In my view therefore the most devastating objection to verificationism has been neglected in favour of objections that are either significantly weaker or off-target. This is especially regrettable because the objection can be generalised so as to confront other kinds of nonsensicalism: one gets our dilemma, the Problem of Diagnosis.18 Why should there be this neglect? I can only suggest that the thinking might be along the following lines: Obviously, one must somehow grasp what someone is saying if one is to assess it in any way. One can do nothing, for example, if it is in a language of which one is entirely ignorant. But this grasping is not a full-blooded attribution of meaning but something more rudimentary. One seems to discern something that may turn out to be a meaning but may not. The trouble comes when one tries to be more precise about what it is that one must attribute to the utterance in order to assess it. A suggestion I have heard more than once is that all that is necessary is that one can see what the philosopher is ‘trying to say’. But is not what someone is ‘trying to say’ normally just a way of referring to what he means? (It is also an extremely condescending way – who has not been irritated by being told, ‘I can see what you are trying to say’?) Once we have set this aside we are back where we 15 See, for example, Morris Lazerowitz, The Structure of Metaphysics, Routledge and Kegan Paul, 1955, pp 54 – 6. 16 I realise that many philosophers – particularly in America – believe that Quine revealed the underlying source of the difficulty. But I still don’t see how it could be proved conclusively that no version of the Verification Principle could get round it. 17 The earliest version of verificationism equated the meaning of a proposition with the method of its verification. This would surely make the Verification Principle analytic enough for anybody (or at least anybody who does not reject the very concept of analyticity). For a discussion of what would be wrong with treating the Principle as analytic see Charles Pigden, ‘Coercive theories of meaning, or why language shouldn’t (so much) to philosophy’, Logique et Analyse, 2010, pp. 174-79. 18 One place where this objection to verificationism is clearly seen to be a special case of a general difficulty for nonsensicalism is in the entry under ‘nonsense’ (by Bede Rundle) in the Oxford Companion to Philosophy, ed. Ted Honderich, O.U.P., 1995. 8 Ian Dearden 21/10/2014 11:52 AM started. It is just not clear what it is that one sees when one sees what someone is trying to say, if this is not what he means. It looks as though one will have to attribute to the utterance: a) a proto- or potential meaning: something that may yet turn out to be a meaning19 or b) a quasi-meaning, something that is not a meaning but is sufficiently similar to a meaning for it to be somehow grasped, if not in the full sense understood.20 or c) (if this is different from (b)) a pseudo-meaning: something that is deceptively like a genuine meaning, even to the extent of being something one can grasp by a kind of pseudo-understanding. These all raise the question: why not just admit that in order to criticise an utterance, you have to understand it, assign a meaning to it? Why are these proto-, quasi- and pseudo- meanings not just meanings? After all, one will have somehow to understand, grasp, them to make one’s criticisms. Why is one not understanding what the utterance means? I refer the reader to Witherspoon’s criticisms of the notion of quasi-meanings, which can easily be converted into criticisms of those of proto- and pseudo-meanings. Are there any other reasons why the Problem of Diagnosis has received so little attention? Perhaps some people suspect a trick, a sophistry. But if the dilemma is based on a trick, it ought to be fairly easy to point out what it is and to say in the light of this how philosophical nonsense might be diagnosed. Perhaps some people, when confronted with the more blatant cases of assigning a meaning to an utterance in order to prove it has no meaning think, ‘This is just carelessness. It surely cannot be typical.’ I do not deny that some cases are more blatant than others - verificationism tends towards blatancy21 but this still leaves the question whether there is a legitimate way of diagnosing nonsense. Finally, there may be a tendency for philosophers to give all their attention to the Problem of Specification at the expense of that of Diagnosis. Perhaps they think that one can judge someone to be talking 19 David Pears argues that for Wittgenstein there are no ‘candidates for possibility’. (The False Prison, Clarendon Press, Vol.II, 1988, pp 215-16). A possibility is simply what makes sense. This, I think, amounts to the claim that there are no ‘proto-meanings.’ Another way of putting it might be to say that although there are candidates for meaningfulness – sounds or marks on paper – there are no candidatemeanings, entities which may or may not turn out to have the status of real meanings. 20 Edward Witherspoon in ‘Conceptions of nonsense in Carnap and Wittgenstein’ in The New Wittgenstein’ op. cit. argues that Carnap and others in effect attribute quasi-meanings to the utterances they claim are really nonsense. 21 Some of the most extreme examples are to be found among the verificationist arguments employed in Norman Malcolm’s ‘Dreaming, Routledge and Kegan Paul, 1959, which I discuss in Do Philosophers Talk Nonsense? In fairness to Malcolm I should have said there that one of the things that made for blatancy is his refusal to avail himself of a characteristic hedge of the Logical Positivists. He never says that the claims he is criticising lack ‘factual’, ‘cognitive’ or ‘empirical’ meaning. He says forthrightly that they lack meaning, that they are nonsensical. 9 Ian Dearden 21/10/2014 11:52 AM nonsense on a more or less intuitive basis22 and then start to get rigorous about specifying what that nonsense is. This might be what happened with Wittgenstein’s criticism of Russell’s Theory of Types. Wittgenstein did not deny that Russell had discovered a certain type of nonsense; it was his attempt to formulate rules for identifying and excluding such nonsense that he objected to.23 But speculations aside, the Problem of Diagnosis needs to be taken seriously. The Austere View Of Nonsense Recently a number of philosophers have realised the need for greater rigour in the employment of the concept of nonsense. This ‘austere’ view of nonsense has been well summarised by Cora Diamond, who was one of the first to take it: Anything that is nonsense is so merely because some determination of meaning has not been made; it is not nonsense as a logical result of determinations that have been made.24 There is no ‘positive nonsense’, no such thing as nonsense that is nonsense on account of what it would have to mean, given the meanings already fixed for the terms it contains.25 If one takes PI 500 seriously, one cannot allow for the possibility of ‘positive nonsense’, for that would be to allow that it had a ‘sense that is senseless’. Wittgenstein puts it thus: Most of us think that there is nonsense which makes sense and nonsense that does not – that it is nonsense in a different way to say, ‘This is green and yellow at the same time’ from saying ‘Ab sur ah’. But these are nonsense in the same sense, the only difference being in the jingle of the words.26 The following example should help to illustrate what is at issue. A certain locution in Heidegger has been rendered into English as ‘The Nothing noughts’, ‘The Nothing nothings’ and ‘The Nothing noths’.27 Suppose it is said 22 ‘What is required is an acquaintance with the language, an eye and an ear for sense. This is a requirement and may be compared to the musical ear in the case of music … there does seem to be something comparable in the case of reading philosophy. To some people everything makes sense’, O.K. Bouwsma, Philosophical Essays, University of Nebraska Press, 1965, p.196. 23 See Kenny, Wittgenstein, op. cit., pp 43 – 4. 24 The Realistic Spirit: Wittgenstein, Philosophy and the Mind, MIT Press, 1991, p 106 25 Ibid., p 107 26 Wittgenstein’s Lectures, Cambridge, 1932 – 35, from the notes of Alice Ambrose and Margaret Macdonald, ed. Alice Ambrose, Blackwell, 1979, p.63. 27 The translation ‘Nothing noths’ is, I think, semi-facetious. It is from Peter Geach, Mental Acts, Routledge and Kegan Paul, 1957, p.10 10 Ian Dearden 21/10/2014 11:52 AM (rightly or wrongly) that these are nonsense. We are likely to think that they are nonsense in different ways. The last is nonsense because there is no word ‘noth’ in English, so the ‘sentence’ fails to have meaning because, recalling the words of Tractatus 6.53, no meaning has been given to a certain sign. But the first two, one might think, are nonsense in a quite different way. Take ‘The Nothing noughts’. The English vocabulary does contain the words ‘nothing’ and ‘nought’ and these have perfectly good meanings. But when one uses them to construct the locution ‘The Nothing noughts’ the meanings clash, jam, conflict, collide – and one gets nonsense, positive nonsense. But this, on the austere view, is wrong. We ought to say that in English no meaning has been given to ‘nought’ as a verb.28 ‘The Nothing noughts’ and ‘The Nothing noths’ are nonsense for essentially the same reason, though this is obscured by the fact that ‘nought’ does have a meaning in English – it can be used to construct meaningful sentences – whereas ‘noth’ does not.29 I assume that proponents of the austere view would wish to extend it to the utterer of the alleged nonsense. It is not that he means something by the individual sounds or marks on paper and these inner mental meanings clash to produce overall meaningless. Rather he does not mean anything by any of them.30 This is how things look to the nonsensicalist who wishes to avoid the lack of rigour that has characterised so many discussions of alleged philosophical nonsense. But our own discussion so far might suggest that it would be better to adopt an ‘austere view of sense’ rather than an ‘austere view of nonsense’. If one is compelled to assign some sense to an utterance in order to criticise it, should not this requirement be clearly recognised? If one cannot get away from meaning, should one not drop the pretence that one can? Perhaps it will be thought that I am urging the replacement of one bias by another, the opposite one. I would argue that the second is a far more natural and reasonable bias than the first. After all the utterances that are alleged to be specimens of philosophical nonsense seem to mean something (in most cases at least – the example from Heidegger would need to be considered in context). It is this appearance that the nonsensicalist has to show is deceptive and it is far from obvious that there is any way of doing this. Still, until one is absolutely sure that the Problem of Diagnosis is insuperable it is no doubt better to adopt a neutral terminology. One might speak of ‘the austere view of the distinction between sense and nonsense’. If, as Diamond 28 I have ignored the status of the locution ‘the Nothing’. It is clear what a proponent of the austere view would say about it: simply that no meaning has been given to it or to ‘the’ and ‘Nothing’ in this combination of words. The proponent of the naïve view that the austere theorist is criticising might however say that there is a clash of meanings when ‘the’ is followed by ‘Nothing’ and the resultant is nonsense. (I am not sure whether any importance should be attached to the capital letter: it may just record the fact that we are attempting a translation from German, where all nouns are capitalised.) 29 I should say, I suppose, that we are discussing current, standard English. The Oxford English Dictionary gives ‘noth’ as an obsolete Scottish variant of ‘nought’; and ‘nought’ used as a verb as obsolete and rare, meaning, when transitive, ‘to set at nought’ and, when reflexive, ‘to efface onself’. 30 Any of them? Or just some of them? On this see Witherspoon, ‘Conceptions of nonsense’, op.cit, p.347, fn.18. 11 Ian Dearden 21/10/2014 11:52 AM et al claim, senses do not combine to produce nonsense, there is going to be a difficulty in specifying the nonsense one is talking about and demonstrating that it is indeed nonsense. To my knowledge all the well-known advocates of the austere view of nonsense are Wittgensteinians31 (There are to a large extent the same philosophers as those who take the sheer-nonsense view of the Tractatus). I find this surprising since an appreciation of the need for the nonsensicalist to avoid giving the impression that he is postulating a ‘sense that is senseless’ seems to lead pretty directly to an awareness of the difficulties with the idea of philosophical nonsense and one might have expected a certain unease about Wittgenstein’s general nonsensicalism to have arisen.32 My own development was in the opposite direction. I began by becoming suspicious of nonsensicalism and then gradually became more rigorous in my view of the distinction between sense and nonsense as I wondered whether there was any way in which nonsensicalism might be defended. What suggestions have been made by the austere theorists and others about such a defence? 31 Opposed to the ‘austere’ conception of nonsense advocated by these writers is the ‘substantial’ (Conant’s term) or ‘Carnapian’ (Witherspoon’s term) conception. In Diamond’s terminology substantial or Carnapian conceptions allow for ‘positive’ as well as ‘negative’ nonsense. 32 But then of course there is the question of the position of Wittgenstein himself. Did he see how the ‘senseless sense’ point of P1 500 leads straight to a fundamental difficulty with nonsensicalism, and, if he did, how did he think the difficulty could be dealt with? 12 Ian Dearden 21/10/2014 11:52 AM SECTION II Making Disguised Nonsense Patent One way of sharpening our sense of the difficulties which the Problem of Diagnosis creates for nonsensicalism is to consider the suggestion made by Wittgenstein in various places that philosophical nonsense is disguised, not obvious, nonsense and that our task is to penetrate the disguise.33 How might this penetration be effected? Obviously, one cannot transform the disguised nonsense into something that means the same thing but is obvious nonsense, since it is not supposed to mean anything. That would be to jump into the ‘sense that is senseless trap’ with lead-lined boots. It is no less absurd to suppose that one could show that something was disguised nonsense by showing that it had nonsensical implications. For if it is nonsense, it has no implications; nothing follows from it; it does not stand in logical relations with anything else.34 So what remains? The only possibility that occurs to me is that one might seek to show that something is disguised nonsense by comparing it to something that is obvious nonsense.35 There are several difficulties with this. Here is one of them. Wittgensteinians and other nonsensicalists are always warning us against being taken in by the surface similarities between locutions. Meaningful sentences that have the same surface grammar might have quite different depth-grammars, different logics. I have no quarrel with this. But they may also tell us that we can be misled into thinking a locution makes sense by its surface resemblance to something that really does make sense. Perhaps, but this cuts both ways. If one can’t be sure that ‘p’ makes sense just because it resembles ‘q’, which makes sense, how can one be sure that ‘r’ does not make sense on the basis of its resemblance to ‘s’, which (let us assume) is admitted to be nonsense? I have always been wary of the expression ‘non-starter’, as used in philosophy. I often get the feeling that an attempt is being made, perhaps unconsciously, to divert attention from an idea that deserves serious consideration. But if the use of the expression is ever justified, it is in connection with the suggestions just considered.36 Understanding The Language The above was really concerned with ‘pre-austere’ modes of thinking. Let us turn to approaches to diagnosing philosophical nonsense that try to take account of the 33 See for example P1 464. In Wittgenstein’s Lectures, op. cit., p.64, he speaks cryptically of ‘operations’ as being needed to recognise nonsense. 34 There is no such thing as ‘the logic of nonsense’ though there is a book with that title by Sören Halldén (Uppsala, 1949) 35 For example, Baker and Hacker (Anal. Comm., Vol. II, 1985, p208) cite a manuscript of Wittgenstein’s in which ‘A thing is identical with itself’ and ‘A thing is very similar to itself’ are given as examples of disguised and patent nonsense respectively. 36 It is interesting to note some of the metaphors and analogies one meets in connection with this supposed unmasking of disguised nonsense (itself a metaphor): one pushes the disguised nonsense a little further until it falls over the edge; one follows further the trajectory of thought that has led to the disguised nonsense and one finds patent nonsense; the disguised nonsense is said to crumble under close inspection or fall apart when suitably treated. Similarly metaphors and analogies are often used as if they could explain how it is possible to talk nonsense unwittingly. These include: making an illegal move in chess; confusing the layout of an unfamiliar house with that of one’s own; wrongly fitting together the pieces of a jigsaw puzzle; and failing to see immediately what is amiss in an engraving by M. C. Escher. Such metaphorical language is not inherently objectionable. Perhaps the nonsensicalist could if necessary say what he had in mind without the use of metaphor or perhaps some things can only be said metaphorically. But I suspect a double standard. Suppose someone accused of talking nonsense were found to be constantly resorting to metaphor in his attempts to explain himself. How much mercy could he expect from the nonsensicalist? 13 Ian Dearden 21/10/2014 11:52 AM ‘sense that is senseless’ danger. We shall find, I think, that the practice is always less austere than the theory. Perhaps all that is necessary to diagnose nonsense is to understand the language in which the nonsense has been wrongly supposed to make sense. The idea has been given expression by the Wittgensteinian, Guy Robinson. He wants to argue that the locution ‘machine intelligence’ is nonsensical. He realises that there is a danger that he will seem to be saying that it is the sense of the locution that is senseless that it is, in Cora Diamond’s terminology, senseless in virtue of determinations of meaning that have been made. He writes that certain: ways of demonstrating a failure of sense depend on ‘determinations that have been made’. And how else could it be? But, he continues: [T]here is an ambiguity in that phrase. The individual or subgroup may not have made a determination of sense with the expression ‘machine intelligence’, but the reason for that failure lies precisely in the determinations that have been made within the language. It is certainly true that those who think that there might be artificial or mechanical intelligence have not determined for themselves any meaning for ‘machine’ or ‘artifact’ in that attempted use, but if we are to demonstrate that failure to them we are going to have to appeal to the sense that is established in the language. The word ‘machine’ is not like ‘runcible’, something into which anything can be put: it has a place and a role in our language, and it is precisely that which we can appeal to in showing that it cannot be used in this way, and that the resulting phrase is without sense.37 I have heard this idea neatly summed up as follows: ‘It is because words have the meanings that they do that particular combinations fail.’ Unfortunately, it is a neat formulation of an error. To get it right, one would have to say something like, ‘It is because words don’t have the meanings that they don’t have that particular combinations fail’, or, less awkwardly, ‘It is because words only have the meanings that they do that particular combinations fail.’ Once this correction has been made, it is evident that we are no nearer a solution to the Problem of Diagnosis. How, to recall Tractatus 6.53, does one show that someone has given no meaning to certain signs in his proportions? How indeed does one show that the language does not already give meanings to the signs such that his propositions make perfect sense? The reader will have to take my word for it that I did once meet the formulation in question. But it is surely clear that Robinson’s suggestion amounts to the same thing. How is he to get from his knowledge of what ‘machine’ and ‘intelligence’ mean in English (which we can surely credit him with) to the conclusion that the combination ‘machine intelligence’ has no meaning in English. (Let us not forget that many people think it has a meaning: that is why we are discussing the matter.) 37 Guy Robinson, Philosophy and Mystification – A Reflection on Nonsense and Clarity, Fordham U.P., 2003, p. 36 14 Ian Dearden 21/10/2014 11:52 AM Suppose someone thought he could he could prove that the expression ‘machine intelligence’ was self-contradictory. I have no idea how this could be done but let us assume that did offer a proof. It is sometimes possible – in mathematics, for example – to show that a certain expression is (perhaps surprisingly) selfcontradictory. But this is because one is dealing with terms whose meaning is precisely defined. To show that something is self-contradictory one must operate with ‘determinations of meaning that have been made’ and one must be sure that these terms are being used univocally. So, if anyone were misguided enough to call contradictions nonsense38, they would be, in Diamond’s terminology, ‘positive nonsense’ and have a clear-cut ‘senseless sense’. What one could not do would be to assign no meaning whatsoever to one’s signs and then prove that their combination was a contradiction.39 It would indeed be an achievement to prove that in view of the ordinary meanings of ‘machine’ and ‘intelligence’, the phrase ‘machine intelligence’ was a contradiction in terms. But that is irrelevant to the enterprise of showing that those who utter the sounds ‘machine intelligence’ think they mean something by them when they do not. Robinson’s view that the expression ‘machine intelligence’ is nonsense needs to be distinguished not only from the view that it is self-contradictory but from another: one that might find some support in the writings of the later Wittgenstein and which moreover might well be correct. Some people are struck by certain analogies between what certain machines can do and human thinking, and to them this is sufficient to ‘give sense’ to the idea of machine intelligence. Others are struck by certain dissimilarities between the doings of machines and those of human beings, and to them this means that ‘machine intelligence’ has no clear sense. Must one side be right and the other wrong? Suppose that words like ‘intelligence’, ‘think’ and ‘reasoning’ are too vague for there to be any definite answer here. A few centuries ago no one needed to decide whether such words could be applied to machines: the question had not been raised. Now that it has, perhaps all we can do is stipulate – a decision is called for. Robinson seems to have some awareness of this40, but makes it sound as though it is those who opt to accept the phrase ‘machine intelligence’ who are stipulating. This is an easy mistake to make. Suppose we have a word ‘W’, which turns out to have a certain vagueness, there being cases where people disagree about whether it applies. It can easily look as though those who want to apply it are being innovatory and those who refuse to do so are being conservative. In fact, it is those who want a definite decision - to apply it or not – who are being innovatory and those who are content to let it remain vague who are being conservative. Cross-Examining The Accused Edward Witherspoon believes that the later Wittgenstein’s method can be characterised as follows: 38 Additional reasons against regarding contradictions as meaningless: a) They have a truth value. The fact that they and their truth-value are in a certain sense degenerate surely does not warrant our saying that the truth-value is not really a truth-value. b) They can occur in inferences. c) ‘It is raining here now and it is not raining here now’ is a different contradiction from ‘It is snowing here now and it is not snowing here now’. Surely they mean different things. They would have different translations into another language. 39 One can of course assign no specific meaning to the symbol ‘p’ in such a formula as (p & not-p). It is a variable and one can leave it as such. 40 In his discussion of Margaret Boden’s view, pp 36-37. Rereading the passage I wonder whether I am being too charitable. 15 Ian Dearden 21/10/2014 11:52 AM [W]hen Wittgenstein is confronted with an utterance that has no clearly discernible place in a language-game, he does not assume he can parse the utterance; rather, he invites the speaker to explain how she is using her words, to connect them with other elements of the language-game in a way that displays their meaningfulness. Only if she is unable to do this in a coherent way does Wittgenstein conclude that the utterance is nonsense; ideally the speaker will reach the same conclusion in the same way and retract or modify her words accordingly.41 This is surely best seen as a suggestion about how the nonsensicalist ought to proceed when engaged in debate – as a kind of late–Wittgensteinian equivalent of 6.53’s ‘correct method in philosophy’. Does it avoid the ‘sense that is senseless’ trap? Does it furnish a way of diagnosing nonsense without attributing a meaning to what is said and then concluding that that meaning is meaningless? First, let us look at the utterer of the alleged nonsense and assume, perhaps somewhat naively, that she is simply unable to explain what she means. What are we to conclude from this? Socrates seems to have thought, and has certainly been taken to have thought, that if someone cannot give you a definition of a word, then he does not know the meaning of that word. Few philosophers would accept that today. Yet, a very similar assumption is often made when it is felt that there is a real possibility that someone, probably a philosopher, is not using words in the normal way. But the assumption is surely unjustified. Even if we allow that IOMs are possible, it does not follow that if she is unable to explain how she is using her words, then she must be the victim of one. Still, it might be felt that if someone is using words in a novel way, the onus is on her to explain the new use. If she is unable to do so, we are justified in discounting what she says and moving on to more profitable matters. We must not forget that the terminology of any field, however technical, has to be explained and must ultimately be related to ordinary language.42 Someone who has mastered the terminology of relativistic quantum electrodynamics did so gradually by having new terminology explained by means of terminology she already knew, leading back ultimately to her first physics lesson. So let us turn our attention to Witherspoon’s Wittgensteinian interrogator. It is a well known debater’s trick to keep on demanding that the other define her terms. If one refuses to accept any definition offered and insists that the words are used in the definition must themselves be defined, one can prevent the other from stating her case. It is also familiar to anyone who has ever been in teaching (at any level) that, confronted with a sufficiently obtuse or refractory pupil, one will, after making a number of unsuccessful attempts to explain something, simply run out of ideas and 41 Edward Witherspoon, ‘Conceptions of nonsense in Carnap and Wittgenstein’ in The New Wittgenstein, ed. Crary and Read, op. cit. p.345. I take the reference to parsing to be a way of saying that Wittgenstein does not assume he can break the ‘sentence’ up into meaningful parts combined in a determinate way and then ask whether the combination makes sense. That would be to treat the combination as ‘positive nonsense’ and to fall into the ‘sense that is senseless’ trap. 42 See T. P. Uschanov, ‘The Strange Death of Ordinary Language Philosophy’, http://www.helsinki.fi/~tuschano/writings /strange/ 16 Ian Dearden 21/10/2014 11:52 AM have to give up.43 How therefore do we know that the fault does not lie with the interrogator? It would be unjustified to assume this, of course, just as it would be unjustified to assume that because someone cannot explain what she means by what she says, she therefore means nothing even if she thinks she does. Obviously one needs to establish some rules for fair discussion but it is not immediately obvious what they are going to be, given that we are allowing for the possibility of the radical error of thinking there is something one means when there is not. Witherspoon suggests that ideally the speaker will also come to the conclusion that she has been talking nonsense. But how often does this happen? Frequently? Rarely? Never? These questions will be taken up again when I consider the possibility of cooperation between the accuser and the accused. For the present, I want to comment on a particular word that Witherspoon uses in the passage quoted. He speaks of explaining how words are being used ‘in a coherent way’. Now ‘coherent’ may be being used to mean ‘meaningful’ here, in which case the same question is being raised about the explanation as about what is being explained. We have just discussed this case. But perhaps ‘coherent’ means ‘consistent’. If she can be convicted of inconsistency, then no doubt something has been achieved, but, as I have said before and shall be saying again, this possibility is not well described by using words like ‘nonsense’ and ‘meaningless’.44 Inconsistent Rules Hans-Johann Glock suggests that Wittgenstein’s later method involves confronting the traditional philosopher with a trilemma: It may be granted that mere deviance from ordinary linguistic practice does not constitute a philosophical mistake. But the idea is to demonstrate a certain kind of inconsistency in the philosophical positions or questions attacked, an inconsistency concerning the use of words. The point is that it is constitutive of metaphysical theories and questions that their employment of terms is at odds with their explanations of these terms and that these theories use deviant rules along with the ordinary ones. As a result, traditional philosophers cannot coherently explain the meanings of their questions and theories. They are confronted with a trilemma: either their new, technical uses of terms remain unexplained (unintelligibility), or it is revealed that they cross language-games by using incompatible rules (inconsistency), or their consistent employment of new concepts simply passes by the original problem, which is based on our ordinary use.45 43 Consider this passage from On Certainty: A pupil and a teacher. The pupil will not let anything be explained to him, for he continually interrupts with doubts, for instance as to the existence of things, the meaning of words, etc. The teacher says, ‘Stop interrupting me and do as I tell you. So far your doubts don’t make sense at all.’(On Certainty, 310) 44 I wish philosophers would settle on some precise technical meaning for the terms ‘coherent’ and ‘incoherent’ or stop using them. (On this see also fn. 45.) 45 ‘Philosophical Investigations section 128: ‘theses in philosophy’ and undogmatic procedure’ in Wittgenstein’s Philosophical Investigations – Text and Context, ed. R.L.Arrington and H–J. Glock, 17 Ian Dearden 21/10/2014 11:52 AM Let us set aside the third horn of the trilemma, since it does not seem to have much to do with nonsense. Consider the first horn. Is ‘unintelligibility’ meaninglessness? Well, one might not understand what the traditional philosopher is saying and one might not understand his attempts to explain what he is saying. But suppose one tried to show that the traditional philosopher could not explain his use of words. Would not this run straight into the ‘sense that is senseless difficulty? As already pointed out, there is nothing about any particular sequence of sounds as marks on paper that prevent their being given a meaning. What might well happen is that the attempt would turn into an attempt to convict him of the second type of mistake, inconsistency. But surely trying to convict a traditional philosopher of inconsistency is just – traditional philosophy. Socrates would have been quite at home with it!46 It seems nothing remains of nonsensicalism. Glock’s trilemma (which certainly merits further investigation) seems to amount to this: either the traditional philosopher fails to explain himself or he is inconsistent or he is irrelevant. There is nothing here about nonsense. Am I overlooking something? Does the use of ‘deviant rules along with the ordinary ones’ somehow generate nonsense? On reflexion, there seem to be two forms which this inconsistency might take. It might be that a philosopher uses a word according to one rule (the correct rule for that word in the language in question) in one sentence and according to a different, deviant rule in another. But this is surely equivocation.47 It will not generate nonsense. There is no loss of meaning: the philosopher means something different by the word in one sentence from what he means by it in the other. It may be an unusual kind of equivocation. Perhaps it is only philosophers who equivocate by using words in accordance with the ordinary rules one minute and in accordance with deviant rules the next. But this is not a matter of talking nonsense. The other form the inconsistency might take is the simultaneous application of inconsistent rules (or what is little different, the application of a rule that is internally inconsistent). But, so far as I can see, this will just generate an ordinary contradiction. If, for example, something is an X only if it is an A and only if it is a not-A, then the assertion that it is an X will be a contradiction. More generally, assume the inconsistent rule(s) satisfied and you get a contradiction. Once again, nonsensicality – the failure to mean anything – does not enter into the matter. Routledge,, 1991, p. 84. In this passage the claim that ‘traditional philosophers cannot coherently explain the meanings of their questions and theses’ (emphasis added) appear to sum up the trilemma they face. But in a footnote to the paragraph from which I have taken the quotation ‘incoherence’ is used to refer to the second horn of the trilemma (inconsistency). 46 I do not say that it is easy to convict philosophers of inconsistency. There may indeed be very few knock-down cases in the history of philosophy. It can seem as though, if one is convict another philosopher of self-contradiction, that philosopher has almost to co-operate in his own refutation, a situation which has an interesting parallel in the idea to be discussed in Section III. 47 ‘Logic is about consistency – but not about all types of consistency. For example, if a man supports Arsenal one day and Spurs the next, then he is fickle but not necessarily illogical’ (Logic, Wilfrid Hodges, Pelican, 1977, p.13). The kind of inconsistency I am considering here – equivocation – is closer to Hodges’ example than to logical inconsistency (self-contradiction). 18 Ian Dearden 21/10/2014 11:52 AM Direct Pointing Let us look again at Tractatus 6.53 The correct method in philosophy would really be the following: to say nothing except what can be said, i.e. propositions of natural science – i.e. something that has nothing to do with philosophy – and then, whenever someone else wanted to say something metaphysical, to demonstrate to him that he had failed to give a meaning to certain signs in his propositions. Could there be some direct way of drawing someone’s attention to the fact that he had given no meaning to certain signs? After all, it is nonsense we are supposed to be dealing with, sounds or marks that do not mean anything. It is an absence we are trying to direct attention to. The constant temptation is to postulate some kind of quasi-meaning and direct attention to that, whilst claiming that it is not a real meaning. But if we do this, we have already gone wrong. The accused can justifiably retort that their quasi-meaning is in fact a real meaning, whatever else might be said against it. So we must somehow gesture towards an absence of meaning in the other’s utterance. How is this to be done? Here we seem to draw a blank. We can cross-examine him about what he means, but we have seen how inconclusive that is likely to be. But let us try supposing that, somehow or other, we get him to admit that he has given no meaning to certain signs. What reliance should be placed upon his admission? How can we be sure that it is not like a confession make to the police under duress, which the accused may later retract? Of course, some confessions are genuine: they are truthful and obtained without the application of undue pressure. What is the equivalent to this in the present case? The best I can think of is that the metaphysical utterance should cease even to seem meaningful to the utterer. I would not say dogmatically that this never happens or never could happen. As Wittgenstein points out, a word can seem to lose its meaning when repeated many times.48 Perhaps the same can happen with a ‘proposition’, if attention is focused on it in the right way. Let us assume that the interrogator knows how to achieve this. Why should we treat this as a revelation rather than as a curious mental phenomenon that can be induced by suitable psychological techniques? The question will be especially urgent if the ‘proposition’ later comes to seem meaningful again. But, of course, their experience of loss of meaning, even if it occurs, is rare. And here we come to what might be called the ‘staying power’ of supposed philosophical nonsense. Consider the case of scepticism. Many philosophers 48 PI, II, xi, p 214. I can never get this to work properly. Certainly the word can quickly come to seem unfamiliar. It might, for example, come to have doubts about the spelling. But it never seems to me to lose its meaning completely. Perhaps I have just not continued long enough. If I were to treat it like a mantra, I might be more successful. 19 Ian Dearden 21/10/2014 11:52 AM today consider sceptical doubts to be nonsensical. (I have sometimes wondered whether it might be the majority view.) But I venture to say that sentences expressing sceptical doubt still seem meaningful to all of them. It is not that ‘How do I know that I am not now dreaming?’ and similar questions have come to seem gibberish. No professional philosopher is likely to have much difficulty in giving an acceptable summary of the philosophy of Descartes. This would be remarkable if ‘How do I know that I know that I am not now dreaming?’ had come to seem no more meaningful than ‘Ab sur ah’ followed by a question mark or pronounced with an interrogative intonation. Whatever the reason for the contemporary disdain for scepticism - and I suspect that intellectual fashion is not entirely uninvolved – it is not that sceptical doubts have lost all appearance of meaningfulness. I do not wish to overstate the case. There is an explanation of the staying power of philosophical nonsense to which the nonsensicalist can appeal. He might remind us that something similar occurs with perceptual illusions: lines continue to look curved even when one knows they are straight and two lines can continue to look to be of different lengths even when one knows them to have the same length. How things seem to the senses is to a certain extent independent of how one knows them to be. 49 Could not IOMs be like this? Our susceptibility to certain IOMs, it might be suggested, is hard-wired into us and is not affected by the purely intellectual recognition that an appearance of meaningfulness is illusory. I admit the possibility. There is no obvious equivalent to the contrast between sensation and cognition in the meaning case, but this may not be decisive. It cannot be ruled out that Satzklang has a certain independence of intellectual conviction.50 The trouble is that there has got to be some reason to override appearances, some reason for insisting that, in spite of how things seem, there is just no meaning there. And this is just what we have yet to find. What Kind Of Understanding Does The Accuser Claim To Have Of The Accused? Obviously, the nonsensicalist who accuses another of producing philosophical nonsense is claiming to have some kind of understanding of the accused. Indeed there is a sense in which he is claiming to understand the accused better than the accused understands himself; for the accused is supposed to be in error about whether he himself means anything. Yet the accuser cannot be claiming to understand what the accused says, his utterance. So what is he claiming to understand? We need something more precise than the truism that he is claiming to understand not his utterance but him.51 There are passages in the writings of the later Wittgenstein where he claims to understand the moods, states of mind, attitudes, temptations and 49 See Colin McGinn, Problems in Philosophy – The Limits of Inquiry, Blackwell, 1993, p 132 An examination of nonsense verse might shed some light on this matter. 51 This is not quite the same contrast as that between ‘He who understands me …’ and ‘He who understands them … ‘ insisted upon by those who favour the sheer-nonsense view of the Tractatus, because the nonsense of the accused is not supposed to be deliberate. 50 20 Ian Dearden 21/10/2014 11:52 AM prejudices that are conducive to the production of philosophical nonsense.52 But I do not believe they help much with the task of explaining how IOMs might be possible or – what is under discussion here – how they might be diagnosed. Rather, they assume that IOMs are possible and diagnosable. This judgment will no doubt seem dogmatic, so I suggest that it be taken as a challenge to anyone who disagrees to show how such passages can be used shed light on our problem. Remember that we are asking how someone can become aware that, contrary to what he thought, he has given no meaning ‘to certain signs in his propositions’. This is a more fundamental problem than that of identifying the states of mind that encourage the error. If one did have a solution of diagnosing IOMs, then one could go on to investigate the factors that facilitate them. One thing that is clear from the passages is which Wittgenstein claims to have some kind of understanding of the nonsense-talker is that he does so because he has made the same mistakes himself or at least has been tempted to make them. This suggests that there must be some common ground between the accuser and the accused. We have noted the obvious fact that one is powerless to criticise utterances in a language one does not understand. But perhaps the common ground must amount to more than a common language. Many readers might feel – as I do – that the Heideggerian utterances discussed earlier are simply alien. They purport to be translations into English of a German sentence but they might as well be translations into Martian. Faced with them all one can do, perhaps, is say, ‘I don’t understand’. For criticism to be possible one must be able to go some way towards putting oneself in the place of the other and seeing things from his point of view. But whether the kinds of criticism that then become possible include showing that the other is wrong to think he means anything is a further question. Anyhow, the idea of common ground between accuser and accused suggests that we consider the possibility that there may be more co-operative approaches to detecting IOMs than those considered so far. 52 For example BB, pp 65-66, and PI, I, 398. There are cases where he comes dangerously close to admitting that that confused philosopher means something by his words. Consider Section 87 excerpted from the Big Typescript in Philosophical Occasions 1992 – 1951, edition James C Klagge and Alfred Nordmann, Hackett, 1993, pp 162 -65. Is there not something curiously un-Wittgensteinian about this? – One of the most important tasks is to express all false thought processes so characteristically that the reader says, ‘Yes, that’s exactly the way I meant it’ (emphasis added). 21 Ian Dearden 21/10/2014 11:52 AM SECTION III Conflict and Co-operation In Philosophy We have been considering accusations of talking nonsense as aggressive moves on the part of the accuser. Whereas traditional philosophers accused each other of such things as contradicting themselves, asserting necessary falsehoods and (sometimes) asserting empirical falsehoods, recent philosophers have been just as likely to accuse their opponents of talking nonsense. For some philosophers nonsense now plays the same sort of role as necessary falsehood once did. Indeed it plays a somewhat bigger role since it is often used to eliminate not just theses but questions. (Traditional philosophy would have tried to answer the questions, in part by eliminating wrong answers.53) One does however hear of a different way of employing the concept of philosophical nonsense, a therapeutic way. The idea is that the traditional philosopher is someone who is perplexed. He is perplexed because he has misunderstood the logic of his language and this has led him to ask nonsensical questions and propound nonsensical answers. The therapeutic nonsensicalist, as I shall call him,54 tries to guide the perplexed philosopher out of his confusion. Tractatus 6.53 could be read in this light: demonstrating to someone ‘that he failed to give a meaning to certain signs in his propositions’ could be seen as a way of being helpful rather than hostile. Might one perhaps have a better chance of dealing with the Problem of Diagnosis if this unmasking of nonsense were seen as a co-operative venture? I said that one hears of the therapeutic use of the notion of philosophical nonsense, for one seldom meets it. The reader will probably not need reminding that Twentieth Century nonsensicalism was nothing if not aggressive. Think of Ayer’s Language, Truth and Logic. Or of a book with a similar title, Baker and Hacker’s Language, Sense and Nonsense. Though the idiom is new, these are as polemical as anything in Plato, say, or Berkeley or Nietzsche. Language, Sense and Nonsense, for example seems to dismiss the whole of modern linguistics as a pseudo-science.55 The voice of therapeutic nonsensicalism by contrast has been muted. Nevertheless I suggest that we take it seriously. 53 As Russell put it much traditional philosophy treated logic as ‘constructive through negation’: Where a number of alternatives seem, at first sight, to be equally possible, logic is made to condemn all of them except one, and that one is then pronounced to be realised in the actual world … (Our Knowledge of the External World, Open Court, 1914, p 8.) 54 B.A.Farrell used the term ‘therapeutic positivism’ to characterise Wittgenstein’s later philosophy (or what was then known of it outside the circle of his pupils). See ‘An appraisal of Therapeutic Positivism’, Mind, 1946. 55 G.P. Baker and P.M.S. Hacker, Language, Sense and Nonsense, Blackwell, 1984, pp.314-315, for example, or the blurb. 22 Ian Dearden 21/10/2014 11:52 AM Let me dwell a little longer on the polemical character that philosophy has never lost. Schopenhauer gives the following advice (of which Wittgenstein apparently approved): when you find yourself stumped in an attempt to convince someone of something, tell yourself that it is his will you are up against, not his intellect.56 This prompts the retort: can’t the other fellow tell himself the same about you? But there is something in the advice, all the same. Intellectual battles do tend to become battles of wills. Now Wittgenstein in several places suggests that philosophy, if it to be done properly, requires work on oneself: one has to combat a tendency to see what one wants to see.57 Thus the battle of wills becomes internal: one wants to be cured of one’s own confusions, let us assume, and this desire comes into conflict with the tendency to see what one wants to see. So the idea that there could be a kind of philosophy that is therapeutic is not perhaps as distant from the concept of philosophy as polemic as might at first be thought. In both cases the will is involved. Not surprisingly, therapeutic nonsensicalism is often likened to psychoanalysis.58 Therapeutic Nonsensicalism And Psychoanalysis Before asking to what extent the methodology of therapeutic nonsensicalism might be analogous to that of psychoanalysis, I would like to mention certain similarities between the issues raised by the idea of IOMs and those raised by the idea of the Unconscious.59 There may be some significance in the fact that, simplifying somewhat, psychoanalysis and nonsensicalism originated or at least rose to prominence at about the same time. Perhaps both stem from a certain self-questioning attitude that is specifically modern. I do not know; I 56 G.P. Baker and P.M.S. Hacker, ‘Analytical Commentary on Philosophical Investigations’ Vol. I Blackwell, 1980, p.487fn. 57 See Philosophical Occasions, ed. Klagge and Nordmann, op cit., pp160 – 163. I cannot develop the point here but it seems to me that Wittgenstein is postulating a very peculiar kind of self-deception: clinging to nonsense. Standard examples of self-deception involve someone’s clinging to a belief in the face of overwhelming contrary evidence, but the belief at least makes sense. There will be no problem about saying what is believed. But if someone clings to some metaphysical ‘claim’ because he wants it to be true, and the ‘claim’ is said to be nonsense, the problems explaining what is going on, difficult enough in the case of ordinary self-deception, will be staggering. 58 In one respect it is surprising. P.M.S. Hacker writes: [Wittgenstein] thought that Freud was an important thinker, that he had created not a new science of the mind but a new mythology of the mind. His pseudo-explanations are, Wittgenstein held, brilliant and dangerous. Wittgenstein’s critical attitude does not affect the analogy in question nor do the criticisms of psychoanalysis apply to Wittgenstein’s own philosophical techniques. (Insight and Illusion, revised edition, Clarendon Press, 1986, p154 Fn.) No doubt there is no outright inconsistency between being highly critical of a thinker and yet comparing one’s own method to – not just with – his. But unless one clearly states the points of agreement and those of disagreement, others are likely to be baffled. My discussion of the relationship between psychoanalysis and therapeutic nonsensicalism may shed some light on this puzzle but I make no great claims to this effect. 59 I shall focus on classical Freudian psychoanalysis. I do not think that bringing in other psychodynamic or depth-psychological schools would essentially alter the picture. I do not know that they are ever mentioned by Wittgenstein. 23 Ian Dearden 21/10/2014 11:52 AM leave the question to the historian of ideas. There seem to me to be three other striking similarities that are worth mentioning. a) The idea of IOMs and the idea of unconscious motivation60 both leave one wondering whether there are any limits to their application. Once one has accepted that someone might have unconscious reasons for what he does, one is bound to wonder how common this is. Perhaps most of what we do is not done for the reasons we think but for unconscious ones. Similarly, once one has accepted the possibility of IOMs, one is bound to wonder how much of our talk is nonsense. Most of our philosophical talk perhaps? And what of our non-philosophical talk? Those who objected to Logical Positivism on the grounds that it would be surprising if large numbers of people were going round producing meaningless utterances failed to see that if – and it is a big if – the kind of error the Positivists claimed to detect is a possible one, there is no obvious reason why it should not be common. b) One cannot straightforwardly defend oneself against the accusation of being motivated by an unconscious desire. One can deny it, of course, but how does one back up one’s denial? If one accepts the possibility of unconscious motivation, one seems at an immediate disadvantage: in the nature of the case one will not be aware of the alleged desire. Similarly, if one is accused of being mistaken in thinking that one means anything by what one says, one can insist that one does and one can give further explanations, but if these are rejected, then, as long as one accepts the possibility of IOMs, one is at a disadvantage. c) Both the psychoanalyst and the nonsensicalist are claiming to understand those in whom they detect unconscious motives or IOMs better than they understand themselves. Of course, there are differences. One could perhaps sum them up by saying that in one case the focus is on ignorance and in the other on error. If I have an unconscious motive for something I do, I am ignorant of the existence of that motive, yet I do not have to be mistaken about anything (though I might be, if one of the conscious reasons I give myself for the action is a rationalisation and nothing more). By contrast, if I am a victim of an IOM, I am mistaken in thinking that there is anything I mean by what I say, yet I do not have to be ignorant of anything, except in the trivial sense that anyone who falls into error is at the time ignorant of the fact that he is making an error. Another contrast that might be suggested is that in one case the focus is on a presence (of a motive behind the scenes) and in the other of an absence (of meaning in what is said). I suspect that, when thought through, this contrast will turn out to be amount to much the same as the first. 60 I shall confine myself to discussing unconscious desires and ignore other mental items such as beliefs, memories and fantasies which might also be said to be unconscious. I shall also ignore the possibility that a desire that is conscious might be unconsciously efficacious, i.e. be the motive for an action without one’s realising it. 24 Ian Dearden 21/10/2014 11:52 AM The Methodologies Of Psychoanalysis And Therapeutic Nonsensicalism Psychoanalysis has to face its own Problem of Diagnosis. To see this imagine someone who fully accepted the logical possibility of unconscious motivation but doubted whether Freud or anyone else had developed a reliable method of uncovering it. There would be no obvious inconsistency in his position.61 Psychoanalysts speak of making that which is unconscious conscious, but there is at least a prima facie difficulty in knowing that something that appears in consciousness was there in the mind all along, but unconscious. It is also not entirely clear – to me at any rate – why following a person’s ‘free associations’ can be expected to lead one back to unconscious material of any sort. Yet I do not think the analogy between the diagnostic problems is very close. The difficulty for nonsensicalism is that it is not clear how one can criticise an utterance without assigning a meaning to it. And this is not a matter of uncovering something hidden. One can of course say, as I have, that one is trying to get behind the utterance to show that there is no meaning there. But the analogy with trying to get behind behaviour (including utterances) to show that there is an unconscious motive there is still not close. With the latter there seems no parallel to the question: Is there anything about an utterance other than its meaning that you can get to work on in order to criticise it? In general psychoanalysis does not have a problem about the meaningfulness of utterances. The possibility of genuine communication is not in doubt. An apparent exception might be thought to arise with the interpretation of dreams.62 There are dreams which at least seem to be nonsensical, in a fairly literal sense of the word ‘nonsensical’. A character in Proust dreams he is a string quartet. Let us assume that this is a possible dream and that it is a nonsensical one. This would not block the application of the psychoanalytic method. The man could simply be asked to free-associate to certain elements in his dream, for example: string quartet. An interpretation of the utterance ‘I am a string quartet’ would not be required. Free-associating in the presence of the analyst may be an unusual form of communication, but it clearly is communication. What analogies do those who advocate therapeutic nonsensicalism see between it and psychoanalysis? The main one seems to be that in both cases a certain agreement has to be secured. First, just as the client has to agree to undergo psychoanalysis, the - let us call him ‘the client’ as well – has to agree to undergo nonsensicalist therapy. This is the difference between therapeutic nonsensicalism and aggressive, polemical, refutational nonsensicalism. But if progress is to be made, further agreement will be needed. The therapies will both proceed by stages, at each of which 61 The verificationist might have scruples, of course, depending on just how undetectable the unconscious motivation was supposed to be. 62 Another exception might be the apparently nonsensical utterances of some schizophrenics. But according to classical psychoanalysis, schizophrenics cannot be analysed. The reasons for this are unconnected with the question of whether their utterances make sense. 25 Ian Dearden 21/10/2014 11:52 AM agreement will be crucial. If the psychoanalyst claims to have uncovered an unconscious desire, the client must acknowledge that he does have this desire and this acknowledgement must not be a matter of merely intellectual assent but something more: he must become conscious of the desire. Merely believing he has the desire is not enough. The equivalent of this with therapeutic nonsensicalism is that the client must come to acknowledge that he has been using, or rather misusing, words in a certain way.63 Let us leave for the present the question of whether there is any analogue to the distinction between mere intellectual acceptance of the existence of a desire and consciousness of it and ask: What exactly is the client admitting? And here all our previous problems seem to reappear in the therapeutic setting. The client might admit that he has been misusing words, so the question will arise whether he can say what he wanted to say without misusing words. Suppose he can. Then attention will switch to his reformulation. There may be something wrong with it, but it will not be nonsense. (And if it turns out that he has been using words inconsistently, this will be equivocation not meaninglessness). But suppose he cannot say what he means. Then the question will arise whether his inability to do so is, so to speak, terminal. Perhaps it is just a matter of a lack of verbal skills and shows nothing about the meaninglessness of his earlier utterance. Or perhaps it is just that the therapeutic nonsensicalist refuses to accept any explanation of the utterance or demands an explanation of the explanation, and an explanation of that, and so on ad infinitum. Could the therapeutic nonsensicalist show that no one could possibly explain the utterance? Of course not: any sequence of sounds could have a meaning. What might well happen is that he will give it a meaning and then try to prove that, so interpreted, it is self-contradictory. Here I am simply recapitulating what I said about the suggestions of Robinson, Witherspoon and Glock. The therapeutic model seems to have introduced nothing new other than a less polemical atmosphere and a willingness on the part of the client to be converted. The therapeutic nonsensicalist seems unable to prove anything. Some nonsensicalists might admit this. Hacker writes: Wittgenstein thought that the notion of proof in philosophy was obsolete … Scepticism is not to be answered by proving that we do know what the sceptic doubts, but rather by showing that the sceptical doubts make no sense . . .64 The trouble here is that it is not clear what this ‘showing’ is going to be. It should be clear by now that one cannot prove that something is nonsense, and Hacker or Hacker’s Wittgenstein appears to be admitting this. But the ‘showing’ will have to be something like proving if it is to have any claim on our attention and respect. Must it not have something analogous to rigour, to 63 See P.M.S.Hacker, Wittgenstein’s Place in Twentieth-Century Analytic Philosophy, Blackwell, 1996, p.236 64 Insight and Illusion, op. cit., p 208 26 Ian Dearden 21/10/2014 11:52 AM cogency? One ought not to have the feeling that the conclusion (if that is the word) of the argument (if that is the word) has been reached arbitrarily. Not only might one wonder whether there is any equivalent to rigour here, one might wonder whether there is any equivalent to objectivity. So far we have been proceeding on the assumption that there is a fact of the matter: the accused/client either does or does not mean something by what he says. Earlier, when we were discussing the tendency of philosophical polemic to become a clash of wills, the worry might have surfaced that there may not always be anything objective that is at stake. It certainly surfaces here. What reason is there to think that the therapeutic procedure, whatever it is, is leading the client to a true belief or something like a true belief, something that can at the very least be assessed?65 The Client’s Motivation I mentioned above that Freud insisted that a mere intellectual acceptance by the client of the analyst’s interpretation is not enough: he must somehow become conscious of the previously unconscious desire. Is there anything equivalent to this with therapeutic nonsensicalism? The philosopher who willingly undergoes nonsensicalist therapy presumably wants to be relieved of his perplexities. For example, he does not see how to avoid scepticism (and is not too embarrassed to admit to being worried by anything so passé).66 He is therefore motivated to accept what the therapeutic nonsensicalist tells him: in particular, that he has been taken in by what is in fact nonsense. Clearly a blunt assertion by the therapeutic nonsensicalist that scepticism is nonsense is unlikely to have much effect. But it is hoped that the therapeutic procedure, whatever it is, will gradually lead – no doubt by revealing the nonsensicality of many different locutions – to this conclusion. Is there not a danger that he will assent to some of the claims about nonsensicality simply because he wants to be liberated from his perplexities? After all, it is not as though scepticism is ever going to cease even to seem meaningful. No one surely expects that ‘How do I know I am not dreaming?’ is going to end up seeming like ‘Ab sur ah?’67 So the best that can be hoped 65 James C. Edwards in Ethics Without Philosophy – Wittgenstein and the Moral Life, University Press of Florida, 1982, (see p 109 in particular) takes the view that Wittgenstein did not believe in objective standards of sense. To reject an utterance as nonsense is like saying, ‘I can’t go along with you there.’ One is drawing one’s own boundary setting one’s own limits, for a particular purpose. This interpretation has not, so far as I am aware, been particularly influential, but the fact that it has been seriously proposed should encourage us to ask whether the therapeutic procedure is likely to uncover anything objective. Another problem that is worth mentioning in this connection is that of vagueness. Suppose the concept of meaning is a vague one. Suppose there are borderline cases, ones where there is no definite answer to the question whether someone means anything by what he says. Whether this would be a serious problem for nonsensicalism would depend on how common such cases were and whether they occurred in philosophically sensitive contexts. 66 What of those philosophers who propound theses, who believe they have solved – not dissolved – philosophical problems? They will not come to the therapeutic nonsensicalist asking to be cured. There are obvious limits to the applicability of therapeutic nonsensicalism, even assuming that it is sometimes effective. 67 Am I being dogmatic here, or exaggerating perhaps? Well, I would like to hear from anyone who does expect this or has experienced it himself. 27 Ian Dearden 21/10/2014 11:52 AM for is that the locutions making up classical scepticism will, for reasons explored in the therapy, be rejected as meaningless, in spite of their continuing to seem meaningful. Like the other analogies between psychoanalytic and nonsensical therapy that I have discussed, this one has fairly obvious limitations. Freud is concerned that a client might be too intellectual in his approach to an interpretation, too ready to settle for a mere belief that he has a certain desire when what he needs is to become conscious of that desire; whereas I am worried that the client might allow his intellectual judgement to be swayed by a desire, the desire to have his perplexity relieved. But there is this resemblance: in both cases there is a need for something to clinch the matter and that something must be something of which the client himself becomes aware. What shows that the therapy is making progress? Freud has an answer: the client must become conscious of the previous unconscious desire. But – here again a disanalogy – it is hard to see what the clincher could be in the case of therapeutic nonsensicalism. It is our old problem: what would show that one has given no meaning to certain signs? Not the vanishing of all appearance of meaningfulness. So what else is there? I have twice spoken of ‘the therapeutic procedure, whatever it is’, and the reader will surely have noticed that I am being extremely vague about it. There is an obvious reason for this: the therapeutic setting does not in itself constitute a new method. Let us imagine the most co-operative atmosphere possible. The client is genuinely perplexed about some philosophical problem, wants to be relieved of his perplexity and accepts the possibility that he may have got into this state at least in part by taking nonsense for sense – he does not share my doubts about nonsensicalism. The therapeutic nonsensicalist, for his part, genuinely wants to help and is willing to forgo all aggressive, polemical techniques. There is nothing here that might suggest that ‘a method has been found’.68 I can see no reason why the approaches considered in Part II should fare any better in the relaxed atmosphere of the therapeutic session than they did in the charged atmosphere of philosophical debate. Psychoanalysis does have a method. This is not the place to attempt an assessment of it, but it is distinctive and well-known. There does not as yet seem to be anything that therapeutic nonsensicalism can set beside it. Self-Diagnosis One difference between psychoanalysis and therapeutic nonsensicalism is that with the former the therapy came first and the theory was based on what was learned from it, whereas the latter has yet to be given a serious trial. I am simplifying somewhat. The development of Freud’s thought involved a continuous interplay between observation and theory and his many critics 68 See Moore’s report of Wittgenstein’s 1930 – 33 lectures in Philosophical Occasions, ed. Klagge and Nordmann, op. cit., p 113. The full sentence reads: As regards his own work, he said that it did not matter whether his results were true or not: what mattered was that ‘a method had been found’. Does the reader share my unease about Wittgenstein’s confidence in a method that might not yet have produced any correct results? But the more pressing question is that of what the method is. 28 Ian Dearden 21/10/2014 11:52 AM would argue that his clinical observations give far weaker support to his theories than he thought. Nevertheless, psychoanalysis was a therapeutic procedure from the beginning, whereas the therapeutic version of nonsensicalism seems more like a last-ditch attempt to save it from seeming insuperable objections.69 Psychoanalysts, in so far as they confine themselves to analysing patients, practise what they preach in a way that nonsensicalists, however therapeutic they may aspire to be, rarely if ever do. I say ‘in so far as they confine themselves to analysing patients’, because psychoanalysts have offered ‘interpretations’ of major figures in politics, religion and the arts who have never voluntarily entered into analysis or acknowledged the correctness of the ‘interpretations’ and who may even be long dead. It is far from clear that this is justifiable in terms of psychoanalytic theory itself, though Freud himself was not above it. If one wants an analogy between the usual practice of nonsensicalists and psychoanalysis, one must look to this sort of thing. Imagine that Freud had only produced works like the study of Leonardo and had never dealt with actual patients. If psychoanalysis seems generally to be in a stronger position than therapeutic nonsensicalism, it does have one weakness from which therapeutic nonsensicalism is free. Psychoanalysis is essentially an interaction between two persons and psychoanalytic theory provides reasons why this must be so. It is also important that the analyst should himself have been analysed. But the whole tradition could never have got started unless at least one analyst had either never been analysed at all or had analysed himself. In fact Freud analysed himself. This is presented by psychoanalysts as a heroic venture but in terms of the theory it is an embarrassment. Therapeutic nonsensicalism does not appear to be faced with parallel difficulty. There is no special difficulty with the idea of self-diagnosis, no extra difficulty. If it is possible to show someone that he has given no meaning to certain signs in his ‘propositions’, it should also be possible for him to discover that for himself. Indeed Wittgenstein’s later writings often read like a dialogue with himself and he stressed that it was important that is followers learn to fend for themselves.70 But of course self-diagnosis will face all the difficulties that (two-person) therapeutic nonsensicalism faces. I leave it to the reader to satisfy himself that this is so. Therapeutic Nonsensicalism Should Be Given A Chance I would like to end this discussion with a plea that someone should make a real attempt to apply the therapeutic approach. I do not believe that the attempt will be successful (whether or not it is influenced by psychoanalysis) and it may seem strange that a sceptic about a method should advocate its adoption. But we are probably faced with one of those cases where the only way to be sure that something does not work is to do one’s level best to make it work and to fail. What is needed is for philosophers who seriously believe 69 This assumes that those philosophers who gesture in the direction of a therapeutic approach are aware of the seriousness of the Problem of Diagnosis. As I remarked in footnote 14, I am not clear how widespread this realisation is. 70 See PI, II, xi, p.206 (second edition); p.175 (third edition). 29 Ian Dearden 21/10/2014 11:52 AM there is something in therapeutic nonsensicalism to take up the challenge. Thus my suggestion is neither entirely serious nor entirely tongue-in-cheek. Let me make a few suggestions about how the experiment should be conducted if anything is to be learnt from it. a) The therapeutic nonsensicalist should always make it clear whether he is attributing meaning to the other’s utterance or not. In some cases he may only be doing so tentatively – ‘Perhaps you mean …’. This too needs to be made clear. b) The therapeutic nonsensicalist should always make it clear what sort of understanding (if any) of the other he is claiming to have. If he is not claiming to understand the other’s utterance but only the state of mind that leads him to come out with that utterance, the nature of this understanding must be clearly specified. c) The status of contradictions should be agreed upon. I do not believe that anyone who considers the matter seriously can hold that they are simply nonsensical, meaningless. But if anyone persists in claiming this, I suggest that the discussion be agreed to be about nonsenseother-than-contradictions. d) There is no obvious way of banning debaters’ tricks such as that of continually demanding that the other define his terms. Obviously no one is going to admit that is what he is doing; he will just claim that he genuinely does not understand. The best I can suggest is that one should keep in mind the existence of such tricks. e) It should be clearly understood that if a person cannot explain what he means to the other’s satisfaction it by no means follows that he does not mean anything. f) It will be noticed that the suggestions I am making seem mainly to be restrictions on the therapeutic nonsensicalist rather than on his client. In fact I cannot think of any special conditions that apply to the latter, other than the ones that apply in any fair and fruitful debate: he should be willing to explain what he means when asked to do so and he should always be ready at least to consider the possibility that the other might be right. Whatever the final outcome of the investigation, the details should prove interesting and illuminating. ____________ ((In the above I would like to have been able to say something about Gordon Baker’s ‘Wittgenstein’s Method and Psychoanalysis’ in Wittgenstein’s Method – Neglected Aspects, Gordon Baker, ed. Katherine J. Morris, Blackwell, 2006. But he seems to be focusing on a different kind of case of alleged nonsense 30 Ian Dearden 21/10/2014 11:52 AM from the ones I have in mind. He discusses Heidegger’s ‘Das Nichts nichtet’ and describes it as ‘patently nonsensical’, adding that ‘Heidegger could scarcely have failed to notice this “defect”’.71 Certainly Heidegger could scarcely have failed to notice that it is not an acceptable sentence in standard German. But I would have said – at least before reading Baker and ignoring the possibility of charlatanry – that Heidegger would most certainly have credited himself with meaning something by it. Baker however writes of this sort of case that a philosopher thinks that to convey an important insight he is compelled to say something which seems, even to himself, empty, self-contradictory or meaningless.72 This makes such a philosopher sound rather like the Wittgenstein of the Tractatus (on the standard ineffable-insight interpretation). I do not know whether Baker is right to see Heidegger in this light.73 But Heidegger, when so interpreted, is a very different figure from the alleged nonsense-talkers discussed above. Even the philosopher who is worried about scepticism and who comes to the therapeutic nonsensicalist to be cured of his perplexities is only open to the possibility that he may have been confused by taking nonsense for sense. I have emphasised that such locutions as ‘How do I know that I am not now dreaming?’ do not, and are not likely ever to, seem to him to be nonsensical. The case of the philosopher who is supposed to be aware or half-aware of talking nonsense yet nevertheless feels compelled to do so requires separate discussion elsewhere.)) 71 Wittgenstein’s Method, p208 Ibid, p208 73 As Baker points out (p. 220 n.13), Heidegger uses ‘Das Nichts nichtet’ to attack the sovereignty of logic. I suspect that any interpretation of Heidegger will have difficulty accommodating this line of thought comfortably. 72 31