November-December 2001
I. Wittgenstein, Contemporary Philosophy, and Neuroscience
Edward Kanterian: Peter, the year 2001 marked the anniversary of 50years since Ludwig Wittgenstein's death. Your philosophical career,spent in its entirety in Oxford, was mostly dedicated to theinterpretation of Witt-genstein. Your numerous books and articles onhim, partly written together with your colleague Gordon Baker, belong to some of the most authorita-tive in this field. How did you come across Wittgenstein in Oxford? Peter Hacker: The first 'wave' of Wittgenstein's influence upon philosophy at Oxford was from the late 1940s until the end of the 1950s. When I was a student in the early sixties, the posthumous publications of J.L. Austin dominated Oxford. He had died in 1960, and with the publication of Sense and Sensibilia and How to do Things with Words in 1962, interest in his work, for a while, eclipsed Wittgenstein's influence. Of course, like any other serious student of philosophy at the time, I had a copy of the Investi-gations and dipped into it. But I could make little of it. My tutor at Queen's, Jonathan Cohen, had no interest in Wittgenstein and steered me down quite different paths, and although my supervisor, Herbert Hart, with whom I wrote my doctorate in philosophy of law, directed me to read what Wittgenstein had to say about family resemblance and vagueness, I really had little acquaintance with Wittgenstein's work in my student days. In-deed, I do not think that anyone lectured on Wittgenstein in those years; certainly I attended no lectures on his work. But when I came to St John's as a young Fellow, I found that John Mabbott, who was then President of the College, expected me to give tutorials on Wittgenstein. He thought, quite rightly, that any undergraduate in philosophy should know at least something about the private language argument. So, for the first time, I had to sit down and read the Investigations carefully. In many ways it was lucky that I had to find my own way through this masterwork - for I did not know what I was, so to speak, supposed to think about it all. But I was fortunate to have a guide. I had been a Junior Research Fellow at Balliol before coming to St John's, and while I was there Anthony Kenny be-friended me. So when I started working on Wittgenstein, he gave me guid-ance and answered my many questions.
2. But you acquired a real passion for Wittgenstein. How did that come about?
Well, I was greatly impressed by the power of his thought, fascinated by the Bemerkungen mode of exposition that so often merely intimates a tra-jectory of thought, leaving it to the reader to follow through, and I was be-guiled by his wonderful similes and metaphors. That, I suppose, was the initial attraction. But what was more important, Wittgenstein answered questions about the nature of philosophy that had plagued me for years. I had become interested in philosophy as a teen-ager, and I came to Oxford primarily to read philosophy. I was fascinated with the subject, but also had grave qualms. I could not really understand what the subject dealt with, and was deeply unsure whether it was right to spend one's life worrying about such strange and seemingly insoluble questions. Wittgenstein resolved these doubts. His conception of philosophy as a conceptual investigation, with constructive and destructive aspects to it, seemed to me to be not only right, but also immensely liberating.
3. Why did this seem liberating?
It seemed, and still seems, to me to be liberating in so far as it made it clear why philosophy is altogether unlike other subjects, why there is not a ac-cumulated store of solid philosophical knowledge, why there seems no progress in philosophy - and yet it is nevertheless not a futile endeavour. If one thinks of philosophy, as many philosophers today do, as a cognitive discipline, if one thinks that it aims to add to the sum of human knowledge, then the spectacle of its achievements must be very depressing indeed. Where are the libraries incorporating the sound knowledge achieved by 25 centuries of philosophical investigation? There are thousands of books con-taining secure knowledge of physics, chemistry, biology, and so forth - but none in philosophy. But that is neither because philosophers are so stu-pid, nor because their questions are so difficult. It is because philosophical questions are conceptual, not factual questions. They call out for clarifica-tion, not for information. The aim is to clarify our conceptual scheme, the way we conceive of things - not to add to our knowledge about the world. The objective is a certain, distinctive, form of understanding.
4. And what are the constructive and destructive aspects of this endeavour that captured your imagination?
The constructive aspect is just the clarification, by description, of the ar-ticulations, in any domain of thought, of a fragment of our conceptual scheme. The goal is to display, in an illuminating manner, the connections, compatibilities, incompatibilities, and implications of an array of concepts - to draw a map, as it were, of a conceptual field. We want to be clear how this concept is related to that one, how it differs from a related one, and how very different it is from one to which it seems to be similar. So, for example, we want to know how perception is related to sensation (does it always, sometimes or never involve sensation?), how it is related to knowledge or belief (are the senses cognitive faculties?), what are the con-ceptual differences between veridical and mistaken perception (are veridi-cal perceptions caused by the objects perceived?), how perception differs from imagination (can one mistake a mental image for a perception?) and so on. The destructive aspect is no less important. We are, in philosophical reflection, beset by conceptual illusion. This, to be sure, is true not only of philosophers, but also of others who venture into these quagmires of thought - anyone who raises, and thinks about, conceptual questions, questions about the nature of things or the essential nature of things. We are inclined to think of the mind, for example, as a mental or spiritual en-tity that is in principle separable from the body and may survive the death of the body - or, on rebound from such Cartesianism, we may be inclined to think that the mind is just the brain. It is a destructive task of philosophy to show that both these conceptions are incoherent. Similarly, we are in-clined to think that mathematics is the study of the objective, mind-independent properties of abstract entities - numbers. Again, it is a task of philosophy to expose this form of Platonism as an illusion. Philosophy, as a purely critical discipline, is the logic of conceptual illusion. These two aspects of philosophy are complementary.
5. Wittgenstein said in his Lectures 1930-3, his later philosophy was meant to be a 'kink' in the history of the subject. Looking at current philosophical debates, it is doubtful whether this 'kink' has taken place. Instead, 'tradi-tional' philosophers such as Kant and Hegel experience revivals every other decade. More recent thinkers such as Frege, Russell and Carnap continue to be very influential in the philosophy of language and logic. Cognitive science has a leading position within the philosophy of mind, and there is a new rise of scientifically inspired metaphysics. Is Wittgen-stein not rather about to become one of those ingenious, but ephemeral figures in the history of philosophy?
No, I don't think so. I don't have a crystal ball into which to gaze to see the future course of philosophy. But it seems to me that Wittgenstein's place in the pantheon of the philosophers is perfectly secure. The Tractatus and the Philosophical Investigations remain two of the greatest works of twentieth-century philosophy, and I doubt whether that judgement is likely to change. The real danger is that Wittgenstein's books will become 'classics', i.e. books everyone wishes they had read. There is a danger that the interest in Wittgenstein's philosophy will degenerate into a form of scholasticism, i.e. a subject for interminable conflicting interpretations, which is no longer used in the struggle with fresh philosophical problems. That could happen. I think it would be very regrettable. And it would be a terrible mistake - for profound insights would be lost. But it could happen, it could happen...
6. It seems to me that the metaphysics and semantics of the Tractatus have come to be - via the Russell-Vienna Circle-Quine route - more influential than the Investigations. How is it to be explained that the Tractatus, a work Wittgenstein himself came to consider as full of errors, could exert such a fascination on 20th century philosophy?
I think that you exaggerate, although, to be sure, there is something right about what you say. I don't think that the Tractatus is more influential than the Investigations; that is, I don't think that many philosophers read the Tractatus and try to execute its programme, or that many philosophers would recognize themselves as followers of the early Wittgenstein. But what is true is that the spirit of the Tractatus has triumphed over the spirit of the Investigations. This is especially manifest in the philosophy of lan-guage, and also in the revival of metaphysics. It seems to me that it can be explained by four factors, which should be seen in the context of the growth of scientism and the widespread accep-tance of a continuity between philosophy and science.
7. What do you mean by 'the growth of scientism'?
I mean the illegitimate extension of scientific modes of explanation to phe-nomena for which they are inappropriate, or the belief that any serious question is to be answered by science or at least by scientific methods - and if it cannot be so answered, then it is not a real question. And, of course, if philosophers think this, they will be prone to think of philosophy as an extension of science, embarked upon the same general enterprise as the sciences. They will be inclined to think that philosophy, like science, is concerned with the attainment of knowledge about the world, and that like sciences, it is concerned with the construction of theories about the nature of things.
8. And you consider this to be misguided?
Indeed I do. It seems to me to be an intellectual disaster. As I have already tried to explain, I think the methods of philosophy are altogether unlike the methods of the sciences. I do not believe that philosophy is concerned with theory construction, but with concept clarification and with the critique of conceptual illusion. And I do not think that the problems of philosophy are amenable to any form of scientific investigation, let alone solution.
9. I see. Now, what are the four factors that you mentioned?
Ah yes - the four factors that perhaps explain the triumph of the spirit of the Tractatus. What I had in mind was first, the emergence of new kinds of philosophy of language - the endeavour to construct what is fondly thought of as a 'theory of meaning for a natural language' - in which the vision of the Tractatus was clearly operative. Both Davidson and Dum-mett, the two main advocates of this enterprise, were fascinated by the thought - clearly articulated in the Tractatus - that we are able to under-stand an infinity of sentences on the basis of learning a finite vocabulary and formation rules. The second factor, operative at the same period, was the emergence of a new linguistics at the hands of Chomsky. This seemed to complement the aspirations of philosophers of language. The linguists could justify their conception, and paper over the cracks in the coherence of their programme, by appealing to the philosophers. At the same time, the philosophers could appeal to empirical linguistics to show how scientific their endeavours were - a happy exchange of dirty washing with no laundry. The third factor I had in mind was the computer revolution and its apparent implications. It inclined philosophers, scientists and psychologists to construe the operations of the mind and the workings of the brain on the model of a computer. And it was no coincidence that Chomskian linguis-tics is eminently computerizable. It is amusing to recollect that when I was a boy, the mind was a telephonist and the brain the central telephone ex-change; now the mind is computer software and the brain is the hardware. Finally, the emergence of cognitive psychology, and subsequently of cognitive science - an imaginary subject which is neither cognitive nor scientific. In rather complicated ways, these different factors supported a marked move away from the modes of thought characteristic of the Inves-tigations towards a much more theory-oriented mode of thought that has much in common with the Tractatus. So depth grammar, of the kind fos-tered by the Tractatus, became popular again, despite its abandonment by the author of the Investigations. Theory construction displaced the patient description of grammatical facts. 'Isms' proliferated and philosophers be-came lost in the pursuit of one chimerical theory and its advocacy in oppo-sition to another equally misconceived will-o'-the-wisp of a theory.
10. Nowadays Wittgenstein is, for the most part, regarded to be the author of several interesting philosophical arguments, most notably the private language argument, family resemblance and rule-following. They are be-ing discussed and used in various contexts, no less than Kant's transcen-dental deduction or Hume's views on causality still are. But how would you explain the fact that neither his conception of philosophy nor his philo-sophical method have been widely accepted by the philosophical commu-nity?
It is certainly not the case that no significant figures employ Wittgen-steinian methods. But it is seems that his is indeed not the predominant ap-proach in current philosophy. If you are right, the price of neglect will be high. For the errors or kinds of errors that he tried to teach us to avoid will be committed again, either in venerable form or in new guise.
11. That seems to be a rather pessimistic view of the present scene.
Yes, it is. I am inclined to think that what is going on today is essentially regressive. Philosophical error is analogous to a disease, not a disease of the body, but of the intellect. And the same old diseases keep cropping up, sometimes having undergone some degree of mutation. So in the middle of the twentieth century Wittgenstein and others tried to cure us of the disease of sense-datum theory in perception - itself a mutant of the disease of ideas and impressions in the seventeenth and eighteenth centuries. That was suc-cessful for a generation. But now the virus is recurring, after due mutation, in the form of 'internal representations'. Similarly, Wittgenstein taught us that metaphysics is an illusion - that so-called metaphysical propositions are at best grammatical propositions, i.e. rules for the use of words dressed up in the guise of descriptions of reality. But today metaphysics has been revived, and the lessons Wittgenstein tried to teach us have been forgotten. It is not that his criticisms of metaphysics have been answered or rebutted. Far from it. They are not understood.
12. And why do you think this is happening?
Well, let me elaborate a little more on what I said before. The linguistic turn that Wittgenstein initiated early in the 20th century was one of the most insightful shifts of perspective in the history of philosophy. Today there has been a move away from this approach. Why has this happened? I sup-pose that there are many different factors at work.
There is a foolish tendency to belittle linguistic concerns, to contrast something that 'is merely linguistic' or 'just about words' with 'substantial matters'. This has undoubtedly played a role in eclipsing Wittgenstein's philosophy and philosophical methods. There are three points worth ob-serving: First, no one criticizes linguistics on the grounds that it is 'just about words' - why should it be beneath the dignity of philosophy to concern itself with words and their use? After all, the particular forms of its concern with language are very different from those of the linguist or grammarian. Secondly, why 'mere words' or 'just about language'? We are what we are because we are language users. The last thing we should belittle is language and its importance to us. Without language we really would be no more than naked apes. Thirdly, the linguistic turn in philosophy did not advocate the view that all philosophical problems are really problems about language. Rather it claimed that philosophical problems often arise out of linguistic unclarity of one kind or another, that a primary method of solving or dissolving philosophical problems is by means of investigations into the uses of words and clarification of their meaning.
13. But few people continue to practice these methods today.
Yes, that's right - and it is an aspect of what I mean in saying that current philosophy is regressive. Wittgenstein said that with the change in his phi-losophy in 1929, with the shift, as he put it, from the method of truth to the method of meaning, it was possible for the first time in the history of phi-losophy to have skilful philosophers. I think that this is true. But the skills are not easy to acquire. And they are rapidly vanishing today.
14. What were the other factors you had in mind?
One is clearly the predominance of American philosophers on the scene. American culture is mesmerized by science. It seems to me that all too many American philosophers think that all real problems can be resolved and answered by scientific methods and that philosophy is either continu-ous with science or at any rate ought to emulate the methods of the sci-ences. I have the impression that such a view is widespread in America. It is certainly a view that was encouraged by Quine. And most American phi-losophers seem to think that Quine showed that the analytic/synthetic, a priori/a posteriori, conceptual/empirical distinctions are obsolete, invalid and to be rejected. It seems to me that he showed no such thing. In particu-lar, Wittgenstein's distinction between what is grammatical and what is empirical is not in the slightest tainted by Quine's (very questionable) criticism of Carnap's conception of the analytic/synthetic distinction. Quine's view that philosophy is continuous with science has certainly con-tributed to the growth of a certain form of scientism in American philoso-phy. And this has been influential in Britain too. I suppose that a related factor is the difficulty in accepting the view that philosophy is not a cognitive discipline, that it does not add to the sum of human knowledge. To those infected by scientism, this seems to dero-gate from the grandeur of philosophy. Many philosophers in America and in Britain find themselves attracted to the older view that philosophy is part of the overall activity of the pursuit of knowledge, perhaps the most august branch. What is surprising is that they are not more depressed by the pov-erty of its achievements over the past two thousand years if the quest is in-deed for new knowledge.
15. Don't you think that this departure from Wittgenstein was also self-induced in Britain, for example by Dummett's promotion of a Fregean ver-sion of the linguistic turn, with its attraction to formal logic and mathemat-ics as a guide to the understanding of the structure of language?
Yes. Michael Dummett's passionate advocacy of Frege's philosophy as he understood it made a considerable contribution to the demise of the kind of analytic philosophy that was practised in Oxford from 1945 to 1975. It seems to me that Dummett pursued neo-Fregean projects virtually oblivi-ous of Wittgenstein's manifold criticisms of Frege and Fregean ideas. Dummett cleaved to the idea that the predicate calculus with identity con-stitutes the fundamental deep-structure of any natural language. Wittgen-stein showed this conception to be misguided in the 1930s. So I agree with you that there were significant local sources of philosophical waywardness. But I doubt whether Dummett's views would have had the impact they did in Oxford but for Davidson's powerful influence. Although Dummett and Davidson disagree over some very important issues, nevertheless their overall projects are similar.
16. If one could maintain that current philosophy is rooted in culture, which in our age is predominantly American culture, could one say that the lessons to be drawn from Wittgenstein's philosophy will be only fully ac-knowledged once our culture will be transformed? This might sound a bit pessimistic.
That philosophical ideas develop in a given cultural context is indisputable. One should surely strive to understand many aspects of philosophical movements against a more general cultural context. It is not surprising that the philosophies of Bacon and Descartes, Locke and Leibniz developed in the context of the scientific revolution of late 16th and 17th century science. Whether there is anything pessimistic about this depends on the fur-ther question of whether one thinks that the influence of the cultural con-text can be broken or shaken. I think it can. I don't think that there is any form of cultural determinism at work here. While current philosophy of mind is greatly influenced by the cultural context of the rise of computer sciences, of developments in computationalist cognitive psychology, of ad-vances in artificial intelligence, and of fantasies about a grand synthesis in 'cognitive science', I am strongly inclined to think that this misguided wave of enthusiasm will spend itself in due course. If, as I believe, the con-ception that informs cognitive science is nonsense, then it is not unreason-able to suppose that after a generation or two people will wake up to the fact that this bogus science does not actually do anything. You can perhaps do things with faulty hypotheses, or with hypotheses that are merely prob-able. You may, in the fullness of time, even arrive at hypotheses that are true. But you cannot do anything at all with nonsense - except breed more of it. So it seems to me that cognitive science will run its course and will be abandoned. The moot question is whether, when it is abandoned as futile and empty, people will turn back to the kind of philosophy of mind that Wittgenstein and his pupils cultivated. I don't know whether that will hap-pen. I hope it will. I hope that people who have done good work in Witt-gensteinian philosophy of mind, such as Anthony Kenny, Bede Rundle, Alan White, Georg Henrik von Wright, and, in due course, their pupils too, will succeed in keeping this form of philosophy of mind alive. I hope that when the wave of cognitive science has broken, sufficient continuity with Wittgenstein's philosophy of mind will have been maintained to enable the next generation to pick up the threads and develop the subject further.
17. In your recent work you seem to be more optimistic about the possibil-ity of philosophy countering scientistic preconceptions. Currently you are working on a substantial book on neuroscience, a rather unusual undertak-ing for a Wittgensteinian. What is this book about?
The book is called The Philosophical Foundations of Neuroscience. I have written it together with the distinguished Australian neuroscientist Profes-sor Max Bennett of Sydney University. He approached me almost two years ago, having read some of my papers on conceptual difficulties in some neuroscientific and psychological theories about human cognitive functions. After some correspondence, we resolved to write this book, as well as a couple of review articles, as a joint venture. It is written specifi-cally for neuroscientists, but I hope that it will be of some interest to psy-chologists and philosophers too. What we have tried to do is to select theories, hypotheses and con-jectures from the works of leading neuroscientists, and to identify and, we hope, disentangle conceptual confusions. I found it interesting, as well as encouraging, that Professor Bennett, who until we started working to-gether, was committed to many of these theories, became persuaded that one need not, and indeed should not, think thus - that considerably more clarity can be obtained by thinking along the very different Wittgensteinian lines that I advocate. Philosophy is not in competition with cognitive neuroscience. In-deed, I believe that it can work together with neuroscience in fruitful coop-eration. The task of philosophy, here as elsewhere, is to disentangle con-ceptual confusions and to clarify the relevant conceptual structures. Unless we are much mistaken, many of the theories of contemporary neuroscien-tists, such as Crick, Edelman, Damasio and Kandel are vitiated by failure to pay adequate attention to conceptual questions. Here is an area in which analytic philosophy of mind of the kind Wittgenstein cultivated and taught us can make a very significant contribution.
18. What would be examples of the kind of conceptual confusions neurosci-ence is entangled in?
I shall give you a couple of examples. One is from current neuroscientific theories of perception. Most neuroscientists think that when a human being perceives objects in his en-vironment, the brain creates images, and what the person actually perceives are these brain-created images. This view, which is in effect a variant of 17th century representationalism, is supported by the widespread acceptance among neuroscientists of the 17th century distinction between primary and secondary qualities. Now, there are numerous reasons for being sceptical about 17th century metaphysics. There are even stronger reasons for being sceptical about 17th century metaphysics when it is presented in the guise of the latest discoveries of 20th century science. There are very powerful ar-guments against any form of representationalism, and equally powerful ar-guments against the conception of the ideality of secondary qualities as conceived by Galileo, Descartes, Boyle and Locke. Philosophy can con-tribute to the disentangling of contemporary neuroscientific theories of perception from the residues of 17th century metaphysics. There really is no reason under the sun why contemporary neuroscience should nail its col-ours to such an old and rotten mast. Another example concerns current neuroscientific work on memory. It is widely believed that memory is stored in the form of increased strength of synapses. But there are problems with that thought. They turn on the question of what exactly is meant by 'a memory'. Is a memory what one remembers, or is it a symbolic (typically linguistic) representation of what one remembers? Either way, the thought that one can store a memory in the brain is problematic. I have the impression that many neuroscien-tists, such as Kandel, or indeed my co-author Max Bennett, committed themselves to research work on memory, indeed on the character of mem-ory storage, in advance of clarifying to themselves exactly what they mean by 'storage of memory'. For while we can store a written representation of a given fact in a filing cabinet, or in a computer, or store a photographic representation of a picturable state of affairs in an album or on tape or diskette, it is far from clear what can be meant by 'storing a memory in the brain'. For one obviously cannot store what is remembered (e.g. a past event or a general truth antecedently learnt) in one's brain, and it is not ob-vious what could be meant by the claim that one stores a representation (a sentence or picture) of what is remembered in the brain. Another point is this: a great deal of the work by Eric Kandel, a very distinguished neuro-scientist, was done on such lowly animals as the Drosophila or the humble sea-slug. In effect what was being investigated were the neural effects con-sequent upon a conditioned reflex. It may well be of great interest that after the conditioned reflex is established, the synaptic connections associated with gill withdrawal are strengthened. But memory is the retention of knowledge antecedently acquired, and there is no reason whatsoever to at-tribute any knowledge to creatures as primitive as a sea slug. All one has there is a conditioned reflex, not knowledge of anything at all - a pattern of behaviour in recurrent circumstances, not cognition. So valuable as Kandel's work may be, its relevance to memory has yet to be demon-strated.
II. The New Wittgenstein, Nonsense, and Beyond
19. In recent years a new interpretation of Wittgenstein has stirred contro-versy. Its proponents, mostly American philosophers such as Cora Dia-mond and James Conant, have criticized interpretations such as Elizabeth Anscombe's, Anthony Kenny's and yours as dogmatic. According to the New Wittgensteinians, throughout his lifetime Wittgenstein was not so much interested in putting forward philosophical arguments against other doctrines, but in demonstrating the futility and nonsensicality of any philo-sophical reasoning. These interpreters maintain that this alleged continuity between the early and later Wittgenstein is more essential than any of the differences between the two periods. How do you relate to this interpreta-tion?
I think that these arguments are profoundly misguided and ill-informed. A number of points come to mind here. There is an extraordinary lack of at-tention on behalf of the proponents of the 'New Wittgenstein' to the actual texts. As far as I can see, they pay no attention whatsoever to what Witt-genstein himself said about his early work, both in the course of the com-position of the Tractatus and in the immediate aftermath of its completion, or to what he said and wrote about the Tractatus in later years. Virtually everything that he every said or wrote about the book contravenes what the New American Wittgensteinians claim. This is the first point I wish to make. The second point is this: there have been many topics from Wittgen-stein's work that have attracted attention over the past fifty years. Family resemblance concepts, criteria, private languages, rule following succes-sively caught the imagination of philosophers. All of these great topics had significant implications for the correct methods of philosophising, whether one was a follower of Wittgenstein or not. By contrast, the particular issues that so preoccupy the new American Wittgensteinians seem to me to be fu-tile. Suppose they are right - so what? What follows from this about the correct way to tackle philosophical problems in any domain of thought? Surely, nothing whatsoever. And philosophers who are not well disposed towards Wittgensteinian philosophy will look at these local controversies and say that all the followers of Wittgenstein can do is quarrel over whether the Master talked utter nonsense or only partial nonsense in his first book. - All of this is singularly futile. I think that the new American Wittgensteinians are doing little other than muddying the waters.
20. The New Wittgensteinians claim that the real issues Wittgenstein was always concerned with were ethics and therapy, the non-argumentative struggle against the will to engage in philosophy. They accuse the tradi-tional interpreters of ignoring this engagement which underlies his texts and philosophy. Was sin, rather than logic Wittgenstein's primordial con-cern?
No, I think that idea is deeply misleading. Let's look first at the Tractatus. One of the tasks of the Tractatus was to address the conception of logic that was to be found in the works of Wittgenstein's predecessors, in par-ticular in Frege and Russell. Frege imagined that logic is a science that dis-covers fundamental truths about relationships between thoughts (that is - propositions), truths that hold for thoughts on any topic whatsoever. The result of a logical investigation would be a general law of thoughts. Russell imagined that logic is concerned with the description of the most general features of the universe - the most general facts of the world, a kind of 'super-physics'. Wittgenstein thought them both to be profoundly mis-taken. A very considerable part of the Tractatus is concerned with deep criticism of these conceptions of logic. The Tractatus set the nature of logic in a completely new light. This was a great achievement. It was not mere therapy; it provided a much needed clarification of the nature of logic and logical truth - something that had puzzled philosophers since the days of Plato. It was not a preoccupation with sin and redemption. And it was argumentative through and through - the Fregean and Russellian concep-tions of logic were refuted with powerful arguments. In addition, the Trac-tatus attempted to give a wholly original account of the relation between thought, language and reality, and, associated with this, a particular solu-tion to fundamental problems about intentionality. This account, although mistaken, as Wittgenstein later came to realize, is deep. To dress this up, as the new American interpreters do, as merely a piece of Kierkegaardian irony seems to me to be a grotesque distortion of great philosophy. Many years after he had written the Tractatus, Wittgenstein said of it not that it was a joke or Kierkegaardian irony or Zen therapeutics but that it was like a watch that didn't work. I can see no value whatsoever in the interpretation that the new American Wittgensteinians are offering. Nothing, in my opinion, speaks for it, and all the evidence speaks against it.
21. The new interpreters hold on to a peculiar notion of nonsense: non-sense is plain, austere nonsense, as opposed to illuminating one. There is no difference between nonsense such as 'Good is and' and 'The world is everything that is the case', and Wittgenstein was not trying, especially not in the Tractatus, to demonstrate that the latter instance of nonsense had some special metaphysical status which, although unsayable, nevertheless could be shown. How do you view this account of nonsense?
I think it is completely misguided. The new Wittgensteinians, in particular Professor James Conant, want to contrast what they call 'an austere' con-ception of nonsense with what they call 'a substantial' conception. They accuse Carnap, and also the received interpretations of Wittgenstein, in-cluding mine, of ascribing to the Tractatus a substantial conception of non-sense. On this conception, a nonsensical sequence of words is nonsense be-cause the constituent words are meaningful but their meanings will not fit together to constitute a meaningful sentence. Substantial nonsense, Conant argues, has a sense that is defective, a sense that makes no sense. Contrary to what they allege, Carnap never held any such view. And indeed, neither Conant nor any of his colleagues among the new Wittgen-steinians, has provided any evidence whatsoever to support this allegation. I have certainly never embraced such a view. Indeed, I cannot think of anyone among 'the old Wittgensteinians' who has. Let me clarify things. Wittgenstein thought that nonsense results from combining words contrary to the rules of logical syntax. At the end of the Tractatus, he pointed out that all the sentences of the book are actually nonsensical, in as much as they transgress the rules of logical syntax. Why do they do so? Because they employ formal concepts, such as 'fact', 'proposition', 'number', 'object', as if they were material or genuine con-cepts. Formal concepts, Wittgenstein held, are in effect variables, and can only occur within the scope of a quantifier. So they cannot occur in an elementary proposition, and so too, they disappear on analysis. The issue is a complex and more or less technical one. But it is evi-dent, both from the Tractatus and from many later remarks of Wittgen-stein, that he thought that rules for the use of words, which he earlier called 'logical syntax' and later called 'rules of grammar', lay down the combina-torial possibilities of words in significant sentences. If you fail to comply with these rules, then the result of combining the words contrary to logical syntax or grammar is that nothing whatsoever is said.
22. So what is the difference between 'Good is and' and 'The world is eve-rything that is the case'?
Well, according to the Tractatus both are nonsense. But the second is non-sense that deliberately misuses expressions the legitimate and correct use of which is as formal or categorial concepts. Moreover, it does so in the endeavour to get the reader to apprehend something that is shown by ordi-nary propositions with a sense. The first is not more nonsensical than the second - there are not different degrees of nonsense. To be sure, unlike the second, it contravenes ordinary syntax - which, as Wittgenstein notes, does not suffice to exclude various forms of nonsense. The suggestion of the New American Wittgensteinians that the Tractatus is not trying to get its readers to recognize certain fundamental, structural features of the world, features that are shown by the forms of language, seems to me to be wholly mistaken. Indeed, it conflicts with eve-rything Wittgenstein said about the book.
23. So are the propositions of the Tractatus really nonsense? All of them?
Well, by the standards of the book itself, they are nonsense. That is, they are not pictures of possible states of affairs that could be otherwise. They do not have a sense that consists in the agreement or disagreement of the propositions with possibilities of the obtaining and non-obtaining of states of affairs. You must remember that the concept of sense in the Tractatus is a highly technical one - so, for example, mathematical equations such as '25x25=625' are also nonsense, for they surely do not describe possible states of affairs that might have been otherwise. However, if you mean: are they really nonsense, and did Wittgen-stein later think they are nonsense - the answer is surely No. The brilliant criticisms of Frege and Russell are anything but nonsense. They are power-ful criticisms of a mistaken conception of logic. The profound insights into the nature of logic and logical truth are anything but nonsense - they are, as Wittgenstein would later put it, grammatical propositions that clarify concepts. There is nothing nonsensical about the assertion that the proposi-tions of logic are tautologies, or that the logical connectives are not names of functions, but operators. On the other hand, many metaphysical propositions of the book are awry. It was surely a mistake, for example, to say 'The world is everything that is the case'. But one might say 'A description of the world consists of statements of what is the case'. And the conceptions of objects, properties and relations, of facts and states of affairs were all confused, as Wittgen-stein later realized.
24. So you think that Wittgenstein was propounding a metaphysical system in the Tractatus?
Yes, I do. When he wrote the book he really did think that there were things that could not be said in any possible language, but were shown by the forms of a language. These were precisely truths about the essential structure of the world. As I just explained, the sentences of the book, by its own standards, are indeed nonsense, but they are not like gibberish, either in their constitu-tion or in the intent with which they are uttered. The pseudo-proposition that there are no logical objects, or that the logical connectives are not rep-resentatives, or that the propositions of logic are senseless tautologies are, according to the pictorial conception of the proposition, nonsense. But they are not just like 'Higgledy-piggledy' - even though they no more de-scribe possible states of affairs than does 'Higgledy-piggledy'. Remember, again, what I said before: mathematical equations such as '25x25=625', indeed all mathematical propositions are also said to be pseudo-propositions, but this does not mean that they are as useless as 'Higgledy-piggledy'. You might compare the sentences of the Tractatus with the para-doxical etchings of Moritz Escher. Escher draws, as it were, logically im-possible pictures. They do not depict possible states of affairs - but they are not mere scribbles. They are pictures that transgress the rules of per-spectival representation. Now, imagine that the point and purpose of these pictures is to draw your attention to the correct forms of representing three-dimensional objects on a two-dimensional surface. Of course the pictures are nonsense pictures. But if you examine them carefully, you can see why they are nonsense, and you can also see what the correct method of repre-sentation is. This is analogous (but only analogous) to what is going on in the Tractatus. Wittgenstein thought that by employing formal concepts in ways that are actually illegitimate, he can bring his readers to apprehend something profoundly important about the metaphysical nature of the world, of language, and of the relation between language and the world. So, the New American Wittgensteinians' view that all this is an elaborate form of gibberish propounded in a Zen-like manner seems to me to be misguided.
25. I mentioned earlier that the new interpreters stress the continuity be-tween the early and the later Wittgenstein in terms of a predominantly ethi-cal concern. It seems to me that your own interpretation, your work on Wittgenstein stresses rather the discontinuity.
I think that there is a profound break that begins when he returned to phi-losophy in 1929. Of course, there are continuities as well as discontinuities, and I hope I have stressed each to the right degree. The Tractatus collapsed as a result of the colour-exclusion problem. The conception of logic had to be substantially changed once he realized that not all logical relations are determined by truth-functional combination of elementary propositions, and that elementary propositions need not be independent. Similarly, the conception of word-meaning had to be abandoned, since it was wrong to suppose that the meaning of a word is an object for which it stands. It was misguided to think that the relation between a word and its meaning is a word-world connection. In that sense, there is no con-nection between language and reality. So too, the solution to the problem of intentionality undergoes radi-cal change, while preserving a core of truth. There is indeed an internal re-lation between a proposition and the fact that makes it true, between an ex-pectation and its satisfaction, and between desire and its fulfilment. In the Tractatus this internal relation is articulated in terms of an isomorphism between the logical structures of world and language. After 1931 he real-izes that this relation is in fact an intra-linguistic one, not a word-world re-lation at all. The relation between the thought that p and the fact that p con-sists in this, that 'the thought that p' and 'the thought that is made true by the fact that p' are simply two different ways of referring to the same thought. So there is both continuity and very dramatic change. To get the balance right is not easy, but it is important to do so.
26. So do you think that his ethical struggle is of essential importance for our understanding of Wittgenstein as a philosopher?
No, I don't. I think that Wittgenstein, as a human being, was intensely, pas-sionately, concerned with moral questions, with how one should live, and with how he himself ought to live. But moral philosophy was a minor con-cern that preoccupied him only during one short phase of his philosophical career. Remember that he worked for seven long years on the Tractatus. Only for a few months out of those seven years was he concerned with ethical questions. During that phase, in 1916, the ethical involvement is in-deed deep and sincere. But to pretend that this is the main theme of the book seems to me to be wrong. It is true that when he returned from the war he told von Ficker that the book consists of two parts, the one he wrote and the one that he didn't write - namely on absolute value - and that the second is the more important. I do not doubt his sincerity, but I am in-clined to question his judgement about his own achievement. The main achievement of the Tractatus consists in its insights into the nature of logic and its criticisms of Frege and Russell - not in its remarks on ethics. After the Tractatus he wrote only one short piece on ethics, the 1929 lecture - which is not very good. The idea that moral philosophy was at the centre of his philosophical concerns seems to me nonsense. His later work was fo-cussed on four great themes: philosophy of logic and language, philosophy of mathematics, philosophy of psychology, and metaphilosophical con-cerns about the nature of philosophy itself. Moral philosophy was of no in-terest to him and he wrote nothing on the subject.
27. It is a common view, whether held by the new Wittgensteinians or by other interpreters that the lessons to be drawn from the later Wittgenstein are purely negative. Is this so?
No, I think that is mistaken. It is true that much of his writing is concerned with dismantling philosophical houses of cards - that he was fascinating by the logic of illusion. But it is also true that a great deal of his later work sheds unprecedented light on conceptual structures that have fascinated philosophers since the inception of their subject. Let me give you a couple of examples. Between 1929 and 1944 almost half of Wittgenstein's endeavour was dedicated to philosophy of mathematics. He produced a conceptual overview of mathematics unprecedented in the history of philosophy and of mathematics alike. He suggested that we view mathematics not as a de-scription of relations between abstract entities, nor as a description of a formal game with signs, but rather as a system of norms of representation. He thought that mathematical propositions are rules, and that mathematical proof is a method of concept formation. Whether this is correct or not is a subject of deep debate. I find it profoundly illuminating. But at any rate, it is anything but negative and destructive. If true, it is a great constructive contribution to our understanding of mathematics. Another example is provided by his philosophy of psychology. It sheds much light upon a wide range of psychological concepts - on sensa-tion and perception, on thinking and imagining, on belief, on aspect perception, on volition and intention. Here the work, although also concerned with disabusing us of various illusions, provides us with the main contour lines of the structure of the relevant concepts and the pattern of their relationships. This is constructive, not destructive, work.
28. To take up something you mentioned earlier: In his Lectures 1930-3 Wittgenstein says that now at last there will be such a thing as skilful phi-losophers. Where there is a skill, there is a method, a teachable procedure to come to terms with, in our case, philosophical problems. Do you think there is such a method, and if so, what does this method consist in accord-ing to Wittgenstein?
Wittgenstein certainly had a specific range of methods in mind. The reason he thought that his work in 1929 and later constituted a break with the an-tecedent tradition was this: he conceived of previous philosophies, for ex-ample Descartes', or Kant's, or his own work in the Tractatus, as commit-ted to solving a single fundamental problem from which the solution to all philosophical problems will follow. When he wrote the Tractatus he thought that identifying the essence of the proposition, the general proposi-tional form, will disclose the essential nature of logic, of language and of the world. And, of course, if this core notion was wrong, then everything else would be wrong too. In much the same way, Kant's critical philosophy turns on the correctness of his notion of synthetic a priori truths as descrip-tions of the conditions of the possibility of experience - and if that notion is incorrect, everything else collapses with it. And Descartes' philosophy depends upon the cogency of his conception of clear and distinct ideas as marks of truth. Wittgenstein saw his new philosophy in 1929/30 as revolu-tionary - not a culmination of an antecedent tradition, but a break with the past. Philosophy, as he now practised it, was not tackling one fundamental problem. Problems are tackled piecemeal. Each problem can be solved or resolved, and put in the archives.
29. So what kind of method or methods did he have in mind?
Surely, above all, a careful and methodical description of the uses of ex-pressions in order to lay bare their logical characteristics - their combina-torial possibilities, their compatibilities and incompatibilities with related expressions, their implications and the contextual presuppositions of their use. Of course, this also involves the subtle discrimination between logi-cally dissimilar types of expression and uses of expression. Here Wittgen-stein's genius is patent - where his predecessors had seen uniformity, he discerned diversity, where they had assimilated logically different concepts and concept types, he differentiated them. And in the course of so doing, he taught us to ask new questions - and showed us the importance of their answers to the resolution of philosophical problems.
30. For example?
Well, he taught us to avoid the question of what an expression stands for, and focus instead on the question of what it is used for or what its distinc-tive role is or what need it answers in discourse. Again, he taught us not to search for the phenomena that are described by certain sentences - say mathematical sentences, or ethical sentences, or first-person psychological sentences, or so called metaphysical sentences - but rather to ask whether their role is descriptive at all. Or again, when faced with century-old philo-sophical conflicts, such as between idealism and realism, or Platonism and constructivism, he taught us not to try to decide between the conflicting arguments in favour of one or other of the different positions, but rather to identify the shared views, the common ground between the conflicting schools, and to challenge that - to question what both parties to the dis-pute agreed on.
31. Philosophers such as Bernard Williams have maintained the thesis that Wittgenstein was a linguistic idealist. They base this thesis on the latter's discussion of statements such as 'Nothing can be red and green all over'. This seems to be a statement about the structure of the world, of colours, but if we are to follow Wittgenstein, it is actually a grammatical rule. Simi-larly with other basic concepts of our conceptual scheme. Does this not imply that language alone determines the way we perceive the world? Was Wittgenstein a linguistic idealist?
No. That is quite wrong. The term 'linguistic idealism' is unclear. But if it is supposed to mean that Wittgenstein thought that sentences of our lan-guage do not refer to items in the world, or do not involve objective refer-ence, then that is obviously wrong. On the other hand, if it means that Wittgenstein objected to the notion of de re necessity, and held instead that all necessity is de dicto, then that is correct. When it comes to propositions such as 'Red is more like orange than it is like yellow', Wittgenstein would concede that this looks like a state-ment about the objective, language-independent, nature or essence of col-ours. But he argued that appearances here are deceptive. In fact, the propo-sition is a rule of grammar in the guise of a statement about the world, about colours. It is a rule of grammar in so far as its role, its sole role, is that of a rule that allows us to infer from the fact that A is red, B is orange and C is yellow, that A is colourwise more like B than like C. And we can draw that inference without looking to see. The proposition is part of the 'geometry of colour'. It is no more a proposition about colour than geomet-rical propositions are about space - rather, it is a norm of representation for statements about coloured things, just as geometrical propositions are norms of representation for spatial objects and their spatial relations. I don't think that this makes Wittgenstein a 'linguistic idealist'. What it does make him is an opponent of metaphysical realism and de re necessity.
32. What about On Certainty, which is considered by some interpreters as representing a third or even fourth period in Wittgenstein's development. Do you think one can speak of such a period?
I think that 'third' or 'fourth' period is perhaps exaggerated. But, to be sure, there are important new movements of thought in the last notes on certainty. Much work has been done on them, although how good it is, is perhaps debatable. Unlike the material on philosophy of mathematics, On Certainty is of modest proportions, written over a relatively short period, and unpolished. The work is important, but it is a fragment of a larger whole that was never completed. I once compared it to Michelangelo's Rondanini Pietá - a similar unfinished fragment of genius. I don't think that we shall ever know in what direction these thoughts might have devel-oped. All we can do is make the best we can of incomplete and not wholly consistent notes.
33. Do you see any topics in Wittgenstein which, despite the vast amount of work done so far, have been neglected and are still in need of explanation and interpretation?
The part of his philosophy that has been the least explored is his philoso-phy of mathematics. There is a great deal there that is at best poorly under-stood. I think that there are insights of genius hidden in this disorderly mass of unfinished material. It would be well worth investing a great deal of effort in trying to understand his thoughts in this domain. Of course, it requires someone who has a deep understanding of Wittgenstein, a good grasp of the history of mathematics and of the philosophy of mathematics, and a reasonable flair for mathematics as well. That is a pretty tall order. So far no one has answered to this description. But I am sure that the work will one day be done.