In response to:

A Special Supplement: The Development of Wittgenstein's Philosophy from the January 16, 1969 issue

To the Editors:

Professor Pears’s account of the development of Wittgenstein’s philosophy is the same, old, slightly less than enlightening, version of the mythical Wittgenstein, which was conceived by Russell, nurtured by the Vienna Circle, and sustained by a segment of Anglo-Saxon philosophers for the last forty-five-odd years. What I find appalling about this is that it should occur in an article which has as part of its raison d’être the reviewing of Paul Engelmann’s Letters From Wittgenstein with a Memoir. Englemann’s work has quite clearly and unequivocally brought out the nature of Wittgenstein’s well-known protestations against such a bowdlerization of his thoughts as that of Russell in his introductory essay to the Tractatus. Engelmann’s Wittgenstein is quite different from the Wittgenstein that Pears is reporting upon; he is a genuine mystic whose mysticism, far from being the inexplicable curiosity of Russell’s introduction, is rooted in the silence to which language is consigned when it confronts the transcendent sphere of values. To a philosopher whose thinking has been formed in the Anglo-Saxon tradition the notion of language-mysticism appears as a hideous monstrosity. However, to one familiar with German letters in the period around the turn of the century parallels are not unknown….

Now, the force of Wittgenstein’s letters to Engelmann and Engelmann’s memoir, is that if one is to view what Wittgenstein was up to in the Tractatus as he himself did, he must not be taken as an Anglo-Saxon philosopher but as one who has emerged from a German, or to be more precise, an Austrian intellectual milieu. That Wittgenstein would so identify himself while writing the Tractatus is attested to in McGuinness’s appendix to the Englemann volume in a letter to Ludwig von Ficker, editor of the Catholic, existentialist periodical Der Brenner in which Wittgenstein asserts,

The book’s point is an ethical one. I once meant to include in the preface a sentence which is not in fact there now but which I will write out for you here because it will perhaps be a key to the work for you. What I meant to write, then, was this: My work consists of two parts: the one presented here plus all that I have not written. And it is precisely this second part that is the important one. My book draws limits to the sphere of the ethical from the inside as it were, and I am convinced that this is the ONLY rigorous way of drawing those limits. In short, I believe that where many others today are just gassing, I have managed in my book to put everything firmly into place by being silent about it. And for that reason, unless I am very much mistaken, the book will say a great deal that you yourself want to say. Only perhaps you won’t see that it is said in the book. For now, I would recommend you to read the preface and the conclusion, because they contain the most direct expression of the point of the book.

Wittgenstein thus saw himself as accomplishing the same task as the language-mystics who maintained that language was incapable of expressing any sort of meaningful discourse concerning values. In this respect values were “transcendent” or “higher.” What distinguished Wittgenstein from the other language-mystics was the manner in which he showed that this was the case, i.e., utilizing the tools which he had obtained from his study of physics and logic….

Finally, Pears regards the task that Wittgenstein set for himself in the Tractatus to be an examination of the foundations of logic. Such a view is hardly tenable if the Engelmann volume is to be taken seriously. Pears evidently maintains that this is the case because Wittgenstein speaks of the foundations of logic chronologically earlier and the limits of language only later in the Notebooks 1914-1916. Had Pears glanced at the editor’s Preface to that work, I do not think that he would have advanced such an argument, for it is quite clearly stated there that these are not the only notebooks for the Tractatus but only three of several, the rest of which were destroyed in 1950 in compliance with Wittgenstein’s wishes. Since we do not know what the other notebooks contained, it is fallacious to base any argument about the development of the Tractatus upon the chronology of those that we possess….

Allan Janik

Milton, Massachusetts

D. F Pears replies:

It may be easy to describe Wittgenstein’s demarcation of the frontier between philosophy and the sciences in a general way, but, as Mr. Gildin demonstrates in his letter, the difficulties lie in the details. Wittgenstein’s general idea was that philosophy is an investigation of the conceptual apparatus of all other disciplines and modes of thought. So between philosophy and the sciences there would be a clear line to be drawn. Scientists divide the world among themselves, and investigate its phenomena in a piecemeal way, theorizing and experimenting. But the philosopher makes no experiments and constructs no theories, because he has no such assignment. He is concerned with their thoughts and their language about the phenomena which have been assigned to them. They browse and ruminate in their appropriate fields: he is a gadfly.

But it is one thing to say that there is a clear line to be drawn between philosophy and the sciences, and another thing to draw it. Mr. Gildin asks how anyone who takes this view of philosophy could possibly philosophize against behaviorism. Perhaps this question is not so difficult to answer as his next question, which arises out of it. For behaviorism evidently sprawls across the line between philosophy and science, and there is some justification for cutting it in two. If this is done, there will be on the scientific side not a theory, but a methodological rule: concentrate on stimulus and response, which are the only firm and measurable data in psychology, and so the only data which might be discovered to be connected by functional laws, and ignore the data of introspection, which cannot be extracted from animals, and which come from human beings in a form which is neither firm nor measurable—verbal responses which have to be checked against non-verbal responses, and even then cannot be squared off against a proper scale of measurement.

On the other side of the dividing line there will be the theory that all mental phenomena are reducible to patterns of response. This reductive theory is, of course, related to the methodological rule: it too is an exasperated protest against the elusiveness not only of other minds but also of one’s own mind. Nevertheless, the form which the protest takes on this side of the line is quite different. It really is expressed in a theory, and, what is more, in a theory which is stunningly paradoxical if the word “response” bears its normal meaning. So the next episode will be some sort of apology: the word “response” was being used in a way which does not require responses to be overt. But I need not pursue these developments. The point that I want to make is that, when Wittgenstein offered an alternative to the theory that all mental phenomena are reducible to patterns of response, he was offering an alternative not to a scientific theory, but to a philosophical theory.

Mr. Gildin’s second question is harder to answer. He asks whether anything in Wittgenstein’s philosophy of mind implies the thesis that the acquisition of language precedes cognition and makes cognition possible. The difficulty is that this thesis seems to belong to psychology, and the question, whether it is true or false, seems to be a factual question to be settled empirically. So how can a philosophy which keeps within the bounds set by Wittgenstein presume to settle it?

If philosophy is conceptual analysis, more trouble is to be expected on the frontier between philosophy and psychology than on the frontier between philosophy and any other science except linguistics. The point is that psychology, like linguistics, but unlike physics or mechanics, is concerned with mental phenomena and with their expression. So encroachments on these two frontiers will be particularly difficult to detect and rectify. I cannot unravel the skein of issues which lie behind Mr. Gildin’s second question, but perhaps I can make a start.

First something needs to be said about Wittgenstein’s general strategy in the philosophy of mind. What he offered was an alternative to philosophical behaviorism. But it was also an alternative to psycho-physical dualism. For when he argued that there could not be a private language—i.e. a language which is necessarily untouchable—he was not tilting against windmills. For psycho-physical dualism treats mental phenomena in a way that is very closely analogous to the way in which it treats physical phenomena—the two coordinate divisions of reality—and Wittgenstein believed it to be a consequence of this treatment that the language of mental phenomena would be necessarily unteachable. For on this theory how could anyone know that his use of a word belonging to this part of language was the same as anyone else’s use of it? Wittgenstein saw two distinct absurdities in this consequence of dualism. It conflicts with the evident fact that we really do communicate with one another about mental phenomena, and it conflicts with his thesis that, quite apart from communication with one another about mental phenomena, and it conflicts with his thesis that, quite apart from communication with others, nobody could set up such a language even for his own use.

I am, of course, not giving Wittgenstein’s arguments, but only a general sketch of his strategy. On the one hand, mental phenomena ought not to be treated in a way that is so closely analogous to the way in which physical phenomena are treated. But on the other hand, the paradox of philosophical behaviorism had to be avoided. For though mental phenomena stand in need of behavioral criteria (this is the kernel of truth in philosophical behaviorism) they cannot be reduced without remainder to patterns of response.

So much for the negative aspect of Wittgenstein’s strategy. His positive contribution to the philosophy of mind was a careful description of the language of mental phenomena, and its essential dependence on the language of physical phenomena. This description is designed to reveal the limit of the analogy between, say, sensations and material objects, and to reveal this limit against the background of the natural temptation to overstep it. The temptation is great, and the limit of the analogy is extremely difficult to determine. Everyone can see that a philosopher like Berkeley, who describes an isolated mind packed with ideas, must have overstepped it. But just how far does the analogy extend? Wittgenstein tried to answer this question by carefully describing our linguistic system as it is. But the effectiveness of this description as a piece of philosophy depends not only on its truth, but also on its background. Its background is the picture which leads us to misconstrue our own language, like savages listening to civilized men—the powerful picture which one gets when, for example, one reads Berkeley: mental processes and states are just like physical processes and states except that they unfold in a different medium. Wittgenstein neutralizes the attraction of this picture by heightening it: whenever it seduces us into undetected nonsense, he lays on the colors, and makes the transition to obvious nonsense. It is against this background that he places his description of our linguistic system as it is. Without the background the description would simply be flat.

I cannot answer Mr. Gildin’s second question, but I can now put it in its place, and formulate it in more detail. Anyone who philosophizes in Wittgenstein’s way will make statements about human beings which must at least be true. These truths owe their philosophical effectiveness to their setting: they point towards our linguistic system and the a priori checkerboard of our concepts. Nevertheless, they are empirical truths and this is one source of trouble on the frontier between philosophy and the human sciences.

But disputes of a far more important kind would break out if, as Mr. Gildin implies, Wittgenstein offered philosophical arguments for what are, in fact, empirical truths, if they are truths. It is one thing to take admitted empirical truths and use them in philosophy, but another thing to settle controversial empirical questions by philosophical arguments. Does Wittgenstein ride in the psychologist’s saddle, as so many philosophers have done in the past?

I doubt it. But only a prolonged examination of his later work would put anyone in a position to answer this question with certainty. Mr. Gildin rests his case on a very problematical instance. Certainly Wittgenstein argues that the language of mental phenomena is necessarily dependent on the language of physical phenomena, and his argument (which I have not given) is based on the concept of language. But this necessary dependence does not imply that cognition cannot precede the acquisition of language. Cognition presumably requires some discriminating response, either an overt response, or perhaps a covert one. But the response need not be either vocal or sub-vocal language. In fact at one point in Philosophical Investigations, when Wittgenstein is discussing the connection between the word “pain” and the sensation, he says “Here is one possibility: words are connected with the primitive, the natural, expressions of the sensation and used in their place. A child has hurt himself and he cries; and then adults talk to him and teach him exclamations and, later, sentences. They teach the child new pain-behavior” (§ 244).

I can deal with Mr. Janik’s remarks more briefly. It is difficult to see how anyone could persuade himself that the task which Wittgenstein undertook in the Tractatus Logico-Philosophicus was not an investigation of the foundations of logic. Mr. Janik did it by “taking the Engelmann volume seriously” (Paul Engelmann: Letters from Ludwig Wittgenstein with a Memoir, edited by B. F. McGuinness; Basil Blackwell 1967). The point is that, according to Engelmann, the reason why Wittgenstein tried to fix the limit of factual discourse was that he was interested in what lies beyond that limit. There is, of course much truth in Engelmann’s view. Certainly Wittgenstein believed that it was important to demonstrate that religion and morality (among other things, and let us not forget that philosophy itself is one of them) lie beyond the limit, because that was, he thought, the only way to put them once and for all out of reach of the sort of pseudo-scientific presentation which so discredits them. This enterprise is, of course, a Kantian one. Wittgenstein’s view of its importance may be gathered from the last sentence of his Preface, in which he observes how little has been achieved by his treatment of philosophical problems in the Tractatus. This estimate is confirmed by the interesting letter from Wittgenstein to Ficker which Mr. Janik quotes (Editor’s Appendix to Engelmann’s Memoir, p. 143), in which he says that the point of the book is ethical, and that the important things are the other things (not philosophy) about which he has said that they lie beyond the limit of factual discourse.

But from this there is no need to draw the conclusion that the Tractatus is not an investigation of the foundations of logic, or that, though it is such an investigation, the only point of it is that it puts religion and morality in their appropriate place. It would be more plausible to infer that Wittgenstein stresses the importance of what he did not write about partly because what he did write speaks for itself (after all, he did express his own philosophy in non-factual propositions), and partly because religion and morality have more than theoretical importance. Why assume that a complex work like the Tractatus must have a single purpose? Would anyone accept the parallel supposition that, when Kant wrote the Critique of Pure Reason he cannot have been interested both in what lies beyond the bounds of possible experience and in what lies within them?

My statement, that the Tractatus is an investigation of the foundations of logic, was meant to be based not on Wittgenstein’s Notebooks 1914-1916, but on the Tractatus. But I did observe that in the Notebooks Wittgenstein begins by setting himself this task. Mr. Janik suggests that I overlooked the fact that not all the notebooks are extant. I did not overlook it. The manuscript of the Notebooks, as we have them, starts in code as a soldier’s diary, and when the philosophy begins, it begins with logic. This in itself hardly proves any conclusion about the development of Wittgenstein’s ideas in those years, and was not meant to do so. But it is something that the argument, such as it is, should have a true premise. If confirmation of my statement were needed, it is provided by Wittgenstein’s Notes on Logic, written in 1913, and by his Notes Dictated to G. E. Moore in April 1914 (both included as appendices in Notebooks 1914-1916). But it is quite absurd to have to prove to someone that the Tractatus is an investigation of the foundations of logic. If he doubts it, let him meditate on the full title, or read the book again.

On the other hand, it would be equally absurd to deny the influence of Kantian philosophy on the Tractatus. As a matter of fact, I was afraid that I had exaggerated that influence in my article, because I put that aspect of the work in the forefront. But I hope that I need say no more about the naive dilemma, that either Wittgenstein was influenced by the great logicians Frege and Russell (as he himself says that he was, in his Preface to the Tractatus), or else he was influenced by Kant and Schopenhauer. Of course, both influences were at work, and their interaction with one another can be described in detail. I would like to end this unconscionably long letter with a point which is not too remote from this issue, but is really quite interesting. It cannot be a coincidence that Schopenhauer’s treatment of proofs of geometrical theorems (The World as Will and Idea: Vol. I, p. 98) is so like Wittgenstein’s treatment of proofs of theorems in all branches of mathematics and in logic (Tractatus 6.iff and 6.2ff).

This Issue

July 10, 1969