Wittgenstein on das Innere

The following are notes that I wrote soon after coming out of hospital early in 2002, and in my book on Wittgenstein, which I am now completing early in 2004, I promise that they will be the basis of the penultimate section of the last chapter, ie coming between Colour and Certainty. That order, incidentally, is one of convenience, because the dates of their respective contributions are very higgledy-piggledy, and not always known. The word "basis", however, has to be taken with a pinch of salt, because my notes are far too detailed for the book, and I shall ensure that they remain extant on my website after my book is published. There is another thing that may need to be taken with a pinch of salt: in writing the final chapter I have been tending to the conclusion that I had been over-impressed by Elizabeth Anscombe's stories about the effect on Wittgenstein's thought of his anti-cancer drugs. He was only free of these in the last months of his life, but this is not to say that what he wrote in 1950 and at the end of 1949, when cancer was diagnosed after his return from the States, was in any way feeble. I am not expunging criticisms in these notes that I make based on my drug-assumption, but I do warn readers to beware of them.

Wittgenstein's late work on what he called das Innere is a development of his pre-war notebook work on privacy and his immediately post-war working up of that in Investigations Part I (§§ 243-315). Daunted, however, by trying to explain how 'privacy' and 'das Innere' differ, I would rather concentrate on the latter, which, for all its difficulty, does at least allow its aim to be put in a nutshell --- while this very aim makes clear why Wittgenstein's attempts to achieve it cannot go into a nutshell. The crucial passage comes on page 84 of the volume called Last Writings on the Philosophy of Psychology Volume II.

Das 'Innere' ist eine Täuschung. D.h.: Der ganze Ideenkomplex, auf den mit diesem Wort angespielt wird, ist wie ein gemalter Vorhang vor die Szene der eigentlichen Wortwerwendung gezogen.

What is important to understand about this is that it does not mean that there is no 'inner-and-outer', but that the way we describe it results in illusion. Wittgenstein's attempts to show up our misleading tricks of description occupy the pages of this volume, though some relevant passages are omitted and others included which belong rather with Colour and Certainty (and there is one particular group of paragraphs which make Colour and das Innere overlap in a manner which would drive any conscientious editor to despair).

There is another passage which will help make my point. On page 55, at the beginning of MS 171, Wittgenstein writes

Inneres, in dem es entweder so oder so auschaut; wir [ie, we outsiders] sehen es nicht. In meinem Innern ist es entweder rot oder blau. Ich weiss es, der Andre weiss es nicht.

and in the fifth paragraph on the same printed page he modifies this as follows:

Nicht das ist der wichtige Aspekt, dass die Evidenz das Erlebnis des Andern 'nur wahrscheinlich macht', sondern dass wir gerade diese Erscheinung als Evidenz für etwas Wichtiges betrachten.

For on the one hand it is indisputably true that if I am sitting still and visualising, alternately, red and blue objects or patches, nobody (outside a laboratory, perhaps) will be able to detect which is which, but at the same time nobody (outside a laboratory) will have any interest in which is which. Yet on the other hand, the subtle facial betrayals of what is going on 'inside us' can in other circumstances be of extreme importance (Wittgenstein is thinking of anxieties about whether a friend is telling us the whole truth). Naturally, between these two poles, of our indifference to and our anxiety about what another person is thinking or feeling, there is a huge psychological area for Wittgenstein to try to map with his technique of offering examples of the different things we might say. (Incidentally, I suspect that MS 171, from which these page 55 quotations come, was written while Wittgenstein's clarity of mind was affected by cancer drugs, and I shall later have a number of criticisms to make of the way he expressed himself in it, and sometimes of its content, but I do not think this affects the use I have made of its first and fifth paragraphs.)

While recovering in hospital a few months ago from an operation, I had the pleasure of teaching a nurse the medical distinction between signs and symptoms. If she pressed me somewhere sensitive and saw my face twitch with pain, she would have observed a sign, of pain and of whatever was being diagnosed. If she had to ask me whether or not it hurt, I should be reporting a symptom to her. I have expressed elsewhere my regret that Wittgenstein never took advantage of his war-time hospital experience to make any philosophical use of the sign-symptom distinction, which would have been grist to his mill, but the point I want to make here is that, contrary to the impression given by "Das Innere ist eine Täuschung", there is nothing in his notes on the subject to suggest seriously that if he had overheard us he would have disagreed in any way whatever with what I taught my nurse.

The opening pages of Last Writings II are (in their notebook, MS 169) mainly crossed out vertically, Wittgenstein's code for indicating that a passage has been put to use elsewhere. These pages probably belong to revisions of the undictated passages of MSS 137-8, printed as Last Writings I, in preparation for either MS 144 or its typescript successor, the long lost typescript from which Investigations Part II (henceforth Pt II for short) was edited. However, almost at the top of printed page 26 there are two paragraphs which are not marked as belonging to this revision process. Incidentally, in everything that follows I have assumed that readers will have the printed book to hand, so that I do not constantly need to say which bits are by me and which by Wittgenstein.

Zeig, wie das ist, wenn man Schmerzen hat. --- Zeig, wie das ist, wenn man heuchelt, dass man Schmerzen hat.

In einem Theaterstück kann man beides dargestellt sehen. Aber der Unterschied!

After three more vertically crossed paragraphs there is a longer one, similarly crossed out, which returns to the inner-outer theme, and begins "Und doch bin ich über diesen Ausdruck nicht glücklich", the expression in question presumably expressing the concepts of simulation, dissimulation, pretence and hypocrisy. Then there are two Pt II related lines that are not crossed out, bringing us to the top of page 27, and a paragraph crossed out diagonally, indicating that it was not to be used at all. Finally, a paragraph beginning "Ein Stamm, in dem sich niemand je verstellt" can be taken as the opening of the real das Innere. It ends with the somewhat inscrutable observation "Und das ganze Leben sieht nun ganz anders aus, aber darum im Ganzen nicht notwendigerweise schöner".

Having set the notebook scene I shall begin by quoting a few isolated paragraphs in this one which help make Wittgenstein's view clear, though textual detail will be called for near the end. One such paragraph comes on page 29.

Auch wenn ich jetzt alles hörte, was er zu sich sagt, wüsste ich sowenig wie wenn ich einen Satz aus der Mitte einer Geschichte läse. Auch wenn ich alles wüsste, was in ihm jetzt vorgeht, so wüsste ich doch nicht, auf wen sich z.B. die Namen und Bilder in seinen Gedanken beziehen.

This expresses a point made a number of times elsewhere, that if God observed our entire consciousness in detail, the details would not reveal to him what we meant or intended by them. It also makes clear that a paragraph in inverted commas at the top of page 31 is a deliberate non sequitur --- a didactic non sequitur, one might say.

"Aber für ihn gibt es doch keinen Zweifel darüber, ob er sich verstellt. Wenn ich also in ihn hineinschauen könnte, gäbe es für mich darüber Keinen."

At the bottom of page 33 a paragraph begins consisting of four subparagraphs, for which, respectively, Wittgenstein's terms were "Absatz" and "neue Zeile". This paragraph is one of the few marked at the opening with a slanting line, indicating that he had approved it for inclusion in whatever text he was preparing, so I reproduce it in full.

/ Meine Gedanken sind ihm nicht verborgen, wenn ich sie unwillkürlich ausspreche und er hört es. Doch, denn auch dann weiss er nicht, ob ich wirklich meine, was ich sage, und ich weiss es. Ist das richtig? [nZ] Aber worin besteht nun das, dass ich weiss, ob ich's meine? Vor allem: kann er's nicht auch wissen? [nZ] Wie wäre es denn, wenn mein ehrliches Geständnis unzuverlässiger wäre als das Urteil des anderen? [nZ] Oder auch: Was ist das für eine Tatsache: dass es nicht so ist?

This paragraph is important, because it not only has Wittgenstein's imprimatur but elaborates a point that is at the heart of his cagey ambivalence towards das Innere. I had such ideas myself in the summer of 1951, concerning a friend who, I believed, did not understand his own feelings or know his own intentions, and Wittgenstein's personal life had been giving him frequent occasion for such suspicions.

On page 34, Ein Relativitätsproblem. This is a Wittgenstein code-term. It can be found on page 315 of Volume 2 of the Wiener Ausgabe, from MS 108, late July 1930: Es is ein Schritt nötig der dem der Relativitätstheorie ähnlich ist. There, the necessary step was to justify his dissatisfaction with the idea expressed in his previous paragraph, that our only way of knowing that something that turns up is precisely what we were expecting to happen is to examine our feelings of satisfaction about the result. The idea in that paragraph had been Russell's, and Wittgenstein had been criticising it for some months. His correction for it is expressed two manuscript pages and one printed page later:

Die Sprache als Ausdruck der Erwartung ist das Vorbereitete.

Die Sprache kann nur sagen: Ich habe früher zur Vorbereitung den Satz "p" verwendet und verwende zur Beschreibung wieder den Satz "p".

In other words, what settles whether something that happens is what we were expecting is not our feelings about the event but the language (assuming we can remember it) in which we originally expressed our expectation. The strange thing is that Wittgenstein had already coined a catchphrase for this solution a month earlier, on 28.6.[30], namely:

In der Sprache wird alles ausgetragen.

This insight into Wittgenstein's earlier habits of notebook writing will, I hope, explain the difficulty of untangling his notebooks of 1949-1951. As to what the relativity-type problem of MS 169 was, I can only suggest that it was the difficulty of including in the same account, on the one hand, that

Ich kann im allgemeinen ein klareres zusammenhängenderes Bild von meinem Leben entwerfen als der Andre

and, on the other, that I can sometimes be wrong in making such a sketch, with, as it were on the third hand, that I can indubitably know some (possibly trivial) things about myself that others can't --- and finally finding a way of justifying Wittgenstein's extreme unwillingness to admit that this latter irreducible privacy brings any illumination to the problem.

Three paragraphs below relativity there is another phrase that, as code, harks back to earlier notes, this time to late 1929: "eine allgemeine Tatsache", this one being our ability to predict [some of] the movements of our own body. In the opening pages of MS 108 (see page 133 of Volume 2 of the Wiener Ausgabe), on 13.12.[29], he had written

Es gibt allerdings sehr interessante ganz allgemeine Sätze von grosser Wichtigkeit, wirkliche Sätze die also auch eine wirkliche Erfahrung beschreiben, die also auch hätte anders sein können aber nun einmal so ist. Z.B. dass ich nur einen körper habe . . . [and other examples]. Das sind merkwürdige und interessante Tatsachen.

In other words they are facts that we could easily be tempted to consider a priori propositions (leading us to invent a category of the synthetic a priori, if we were so inclined). And similarly, replacing one person not being able to have two bodies with two people not being able to have one, he had written on 8.[11.29] in MS 107 (see page 109 of the same Wiener Ausgabe volume)

Hat es einen Sinn zu sagen dass zwei Menschen denselben Körper haben? Das ist eine ungemein wichtige und interessante Frage. Having strongly suggested that it is indeed meaningless (and the impossibility therefore a priori) he invents an extremely improbable but intriguing thought experiment to argue that it is meaningful (and the impossibility therefore empirical).

At the foot of our 1949 page (34) there is a declaration that our thoughts' being open to ourselves in no way contradicts their being open to others:

Meine Gedanken sind ihm nicht verborgen, sondern nur auf andre Weise offenbar, als sie's mir sind.

Half way down the following page he elaborates this, in a paragraph where he marks approval at the beginning and then changes his mind and marks approval after the first colon:

Man kann sagen "Er versteckt seine Gefühle". Das heisst aber [because it refers to what he sometimes does], dass sie nicht a priori immer versteckt sind. Oder auch: / Es gibt zwei Aussagen, die einander wiedersprechen: Die eine ist, dass die Gefühle wesentlich versteckt sind; die andre, dass jemand seine Gefühle vor mir versteckt.

Kann ich nie wissen, was er fühlt, dann kann er sich nicht verstellen. This, on the same page, goes further than that and is a puzzling non sequitur. He seems to mean, if his interior is hidden from me absolutely there is nothing left for him to disguise by pretence, but even if the separation is absolute there will be no contradiction if this chap decides that his impenetrable exterior will henceforth not merely hide in the normal manner what is actually going on inside him but give an impression of hiding something different, which is not going on inside him at all.

This rather tortuous possibility expresses, I believe, what Wittgenstein means in a paragraph on page 36 where he speaks of a language game that suggests privacy or hiddenness, as distinct from actually hiding our feelings in a familiar manner. Es gibt offenbar ein Zug des Sprachspiels . . . but there is also something that one can [sanely] call hiding the inner.

In the paragraph above this there occur the words äussert and Aeusserung, which sometimes seem to have for Wittgenstein a special meaning that has puzzled me ever since my attempt in 1952 to correct the translation of Philosophische Untersuchungen. Und dem Besitzer ist es nicht verborgen in dem Sinne dass er es äussert --- that he gives vent to it, I should like to say colloquially, and I leave people who do not like colloquialisms to invent something better. Of course, there are many occasions when these words do not carry this immediacy of meaning, for example on page 35, where "die lügenhafte Aeusserung" is simply "the mendacious utterance".

On the same printed page there is a paragraph that is marked by the slanting line of approval and also by a marginal vertical line which might mean "n.b.":

| / "Er schreit, wenn er Schmerzen hat, nicht ich." Ist das ein Erfahrungssatz?

He seems to imply that it is not, or perhaps to indicate a temptation to think that it is not. For a mother can scream when her child is in such pain as to be unable to scream, and that is certainly empirical. In the next paragraph we have a Witttgenstein "Aeusserung":

"I am feigning pain" doesn't stand on the same level as "I am in pain". After all, it is not a giving vent to feigning.

"When does one say that someone is in pain?" That is a sensible question, and has a clear kind of answer. --- "When does one say that someone is feigning pain?" After all, that must be a sensible question too [but it has a quite different kind of answer].

Can one imagine the signs and the occasions of pain being something utterly other than what they actually are? Say their being signs, etc., of joy? --- So the signs of pain and pain-behaviour determine the concept 'pain'. And they also determine the concept 'feigning pain'.

This important paragraph (ending on printed page 37) draws our attention to the fact that our concept of pain will not allow us to imagine worlds in which pain is expressed by signs that we consider to be signs of joy --- and if I appear to be putting Wittgenstein's horse, as I believe, properly in front of his cart, I must agree that it works both ways: our experience of pain and our knowledge of its signs do not allow us to form any other concept.

The next paragraph seems to suggest that we cannot imagine a world without pretence or hypocrisy. Ahead, in MS 171, Wittgenstein considers such worlds again, but still without offering any explanation of them, though a little later he has second thoughts and suggests that it could be a result of aesthetic revulsion.

For a last detailed comment on the present notebook I must mention the next paragraph, which has its own horse and cart problem. It refers to an idea in the psychology of William James, which James makes too much of but is still plausible.

Wenn man 'traurig ist weil man weint', warum hat man nicht auch Schmerzen weil man schreit?

Common sense is needed here. One can sometimes reinforce sorrow by crying, and perhaps sometimes actually induce it, while doing the same for pain seems excessively unlikely. I believe James would infer from this that pain is not an emotion but sorrow is. What of the emotion of aggression in tennis? One has the impression that aggression induces grunts and groans and that these quite definitely reinforce aggression. I am sure there are passages in earlier post-war notebooks where Wittgenstein takes James's idea more seriously than he does here.

After five more paragraphs Wittgenstein draws a short horizontal line to show that he is changing the subject, to the if-feeling. This sounds rather like sensing the colours of vowels, but for Wittgenstein it had serious significance. His aim was to argue that feelings accompanying words were not what constituted words' meanings (see Pt II, vi). He never gave up the Blue Book claim that the meaning of a word was its use. Yet, when the if-feeling comes back on page 38 after a five paragraph interpolation marked by short horizontal lines and he says "Die Atmosphäre des Wortes ist seine Verwendung", how else can one translate that than, as Luckhardt and Aue do, "The atmosphere of a word is its use"? Fortunately for my thesis, the next sentence shows that this is not meant to contradict the Blue Book. "Or: We imagine its use as an atmosphere."

In the ten paragraphs of this split passage on the if-feeling (ending on page 39 with "('Sabre'-feeling.)", whatever that might mean), Wittgenstein tries to do justice to the reality of our feelings for words (equivalently their atmosphere) while defending his conviction that the nub of a word's meaning is its use. For example, in the first four of these paragraphs are a pair (of which the first, at the foot of page 37, is marked with approval), namely:

/ We interpret the word "if", spoken with this expression, as the expression of a feeling. [My italics to go with the emphasis of the German word-order.]

The use (of the word) seems to fit the word.

and in the second six paragraphs:

We look at [regard, perhaps] a word in a certain environment, spoken with a certain intonation, as an expression of feeling.

The interpolation, on page 38, is on our attitude to the soul as being an expression of our attitude to human beings. We might be brainwashed (at school, perhaps) into saying that people are automata, but this will not influence our attitude towards people [as being people]. For [if so brainwashed] one could say it about oneself [and still know one's human nature]. While applauding his basic thesis I am afraid Wittgenstein was optimistic here.

We return to das Innere on page 39 (where an omitted private passage marks a change of subject), where I need to paraphrase.

Only I can give vent to my thoughts, feelings, etc.

The words in which I [appear to] do so can be insincere. In particular they can be feigned. That is a different language game from the primitive game, of genuinely 'giving vent'.

A change of view comes about between page 39 and page 41.

The child doesn't yet know enough to pretend. [Wittgenstein's own italics, or rather underlining.] [nZ] The question is: What things must it be able to do [was muss es alles können] for us to say that? [nZ] Only when there is a relatively complicated pattern of life do we speak of pretending.

But on page 41, not actually retracting that but changing its emphasis, he says

Can an idiot be too primitive to pretend? He could pretend in the way an animal does. This shows that from there upwards there are levels of pretence.

There are very simple forms of pretence. [nZ] It is therefore perhaps untrue to say that a child has to learn a lot before it can pretend. To be sure, to do this it must grow into it, develop.

Just where he puts the balance I cannot say, but on page 42 he he expresses himself in his earlier manner and marks his paragraph with approval:

/ A child must have developed far before it can pretend, must have learned a lot before it can simulate.

The well-known Pt II concept (see Pt II, i) of a pattern of life or a pattern in our tapestry of life returns on this page. Two other sub-topics need to be mentioned in the four and a half pages between "('Sabel'-Gefühl)" and the next change of topic (at the top of page 44). One is the importance of evidence (facial expressions, groans etc) and the other is the concept of 'concepts'. The former is dealt with further in later notebooks printed in this volume and the latter in the next of those, MS 170. Wittgenstein's central view of concepts is expressed in the penultimate paragraph of this MS 169 passage, on page 43.

As children we simultaneously learn our concepts and what we do with them. [nZ] Sometimes it happens that we later introduce a new concept that is more practical for us. But that will only happen in very definite [Wittgenstein's underlining] and small areas, and it presupposes that most concepts remain unaltered.

Four paragraphs above this there is one about the difficulty of distinguishing between patterns and (should Wittgenstein life-patterns ever get into the curriculum) teaching them. This paragraph is important and needs elucidation (but certainly not retranslation --- I hope my occasional tinkerings with the English do not show lack of admiration for the translators' heroic efforts with a difficult text). I leave a similar elucidation of the next paragraph as an exercise for the reader.

The main difficulty arises from our imagining the experience (the pain, for instance) as a thing, for which of course we have a name and whose concept is therefore quite easy to grasp.

The idea that we interpret a sensation as an object that can be given a name and thereby a category which has only to be formulated to be understood, goes back to Wittgenstein's many pre-war notebooks on privacy. A particularly helpful one is MS 166, notes written in English for a lecture that he may have delivered during his 1939 philosophy of mathematics lectures when Turing was absent.

The subject that takes over at the top of page 44 is knowledge-and-certainty. I cannot guess whether this passage of two and a half printed pages was written before or after the loose sheets called MS 172 [but before them I now suspect, in 2004], on knowledge and colour and printed in their respective volumes. Wittgenstein left these sheets in Elizabeth Anscombe's Oxford house when he travelled with her to Cambridge to die, but she did not find them, if I understood her aright, until the early months of 1967. That was when she asked me to translate the knowledge portion but did not wait for me to do so. The passage beginning on page 44 would have made a useful introduction to the rather abrupt opening of On Certainty which the MS 172 passage offers.

Only nine paragraphs, starting in the middle of page 46, form the passage on concepts, nearly all related to das Innere, while the notebook devoted to concepts, MS 170, making up a mere two and a half printed pages, is free of that restriction. In MS 169 the next subject is colour, one and a half printed pages and unfortunately not included in the colour volume. Then, at the top of page 49, a paragraph on knowledge of another's thoughts is crossed out with double diagonals, followed by two on knowledge and uncertainty crossed out vertically. Finally, and started at the very end of the notebook but completed at the front, there are two paragraphs on knowledge in relation to das Innere, not crossed out.

This notebook was incidentally not one of those which Elizabeth Anscombe gave me access to in 1952 and 1964, and I only noticed its concluding textual peculiarity at the Bodleian in 1992, from photocopies of the Cornell microfilms. I was very relieved, when I bought the volume shortly afterwards, to see that the editors had got the notebook's ending right.

The second section of the printed volume reproduces MS 170, which is something of an oddity. It is a notebook of the same format as MS 169 (grey aand narrow in shape) but the entries end after a few pages. They concern concepts, and although these are concepts related to das Innere they are not strictly part of it. On the other hand, if they had not been printed in this volume there would have been nowhere to print them -- yet it would have been absurd to treat them as a separate unit.

MS 171, of which I have already quoted the first paragraph and the fifth, has a second paragraph whose logic depends on an unstated clause. Remember that when Wittgenstein was shifting his viewpoint about when in infancy pretence could begin, he did not consider new-born children as possibly pretending, even in a 'tierisch' manner. So this paragraph, supplemented, could run:

If pretending were not a complicated pattern, it would be imaginable [which it is certainly not] that a new-born child pretends. [Therefore, pretending is a complicated pattern.]

But the sixth paragraph begins by asking us to imagine just that. Arguing from the impossible is not of the same order of philosophical delinquency as arguing from the meaningless, but it makes one uncomfortable. This is how it begins.

But let's assume that from the very first moment a child was born it could pretend, indeed in such a way that its first utterance of pain is pretence. --- We could imagine . . .

Incidentally, the word "Wehweh" comes here, which I had always taken to be a way of spelling a cry of pain, like "wau wau", but I believe it can be used as a noun, which is how Wittgenstein spells it, as in "Das war ein Wehweh!", making it properly translated as "a hurt".

The concept one has to teach a child in the next paragraph must be that of pretended pain. A philosopher who has been brought up on Frege will find it natural to say that a child who can use the word "pain" must have the concept of pain, but there is no need for children to be taught this concept. And in teaching pretence one must warn against other people's pretence, and thus teach children how to look out for signs of pretence or honesty, making the whole concept of which this search for evidence is part a very remarkable one indeed.

On the next page, 56, there is what I take to be evidence of a lack of clarity in Wittgenstein's thought, caused, I presume, by anti-cancer drugs that he was taking, the giving up of which led to a last burst of admirable clarity. The question here is whether we can imagine people who neither know pretence and lying nor can understand us if we try to explain them. It is not clear what psychological mechanism inhibits these people, convention and upbringing perhaps. A little way ahead Wittgenstein digs deeper. For all he says here it could be mere magic, like the Irish legend of the island where only truth could be spoken, belonging to a great young good god from the south and east, or, less poetically, like the film Liar Liar. The bone I want to pick with him, however, is independent of the mechanism, for he declares that no such people could create or understand fiction, or take part in a play. I maintain that, whatever mechanism inhibits them, such people could find story telling a perfectly proper activity, and similarly for drama. Telling a story is not telling a lie, and the only requirement would be for these people's narrators to distinguish between their histories and their fictions. Nor is drama a lie, and an actor who is not in pain but says he is because his part requires it is not telling a lie either. What these people would not be able to perform or understand would be parts depicting hypocrites. Elizabeth Anscombe would have put him right on this point if he'd asked her.

In the next paragraph it is true that we can imitate pain behaviour without pretending to be in pain. A first aid instructor could do this. And so it is also true that pretence resides in our intention. But:

The capacity to pretend therefore resides in the ability to imitate [and, surely not or] in the ability to have this intention [to deceive, not merely to imitate].

This confusion continues into the next sub-paragraph, but then there is an entirely reasonable suggestion that introduces a possible mechanism: these people might consider lying an aesthetic perversion, not a moral one. This reminds me of Erewhon.

Again, on the next page is a paragraph that is entirely reasonable if one expands it sufficiently to see what Wittgenstein is driving at, namely the importance of what we regard as evidence.

The remark [that such and such is pretence] is not important to us [ie, to Wittgenstein and his followers], but rather the remark that this convoluted set up is evidence for us.

A paragraph about doubt about pain in an anaesthetised patient, and in some cases knowing with and in other cases knowing without specifiable criteria, leads to the point (I do not think necessarily true always, but entirely reasonable in an operating theatre) that there must sometimes be specifiable criteria.

Suddenly, we have a paragraph of pure knowledge-and-certainty (sycamore trees being objects), another on pain and pretence, nine paragraphs that would also have done very well to introduce MS 172 at the beginning of On Certainty, a reworking of the paragraph on evidence, a paragraph on das Innere, one on colour and, ending the notebook, one on whether believing and knowing can be experiences that one recognises as such in the immediacy of having them --- put into inverted commas to show that they aren't.

Of the three sets of notes that follow, the last is only an extract from its notebook, MS 176, which is the penultimate of the three responsible for § 300 onwards of On Certainty. This short set, marked fore and aft with horizontal lines as being on another subject, was written between §§ 523 and 524 of On Certainty. It is an attempt finally to wind up Wittgenstein's thoughts on das Innere. (Another, extremely brief, interpolation that should have gone into this last section separately comes in MS 176's earlier pages, on colour, and is printed as § 57 of Part I of Colour, on page 9). The first of the three sets, MS 173, also opens in much the same way, with well constructed final considerations, but it deteriorates around page 70, and on page 74 it turns to colour and colour blindness, while the last three paragraphs (on page 79) are part of Wittgenstein's complex of ideas about knowledge and certainty. These are the passages that I said would drive an editor to despair, and to reinforce this, when I first saw them in Colour I thought they ought to have been reserved for an eventual volume on das Innere. In fact it does no harm to have them in both. In Colour they start at § 318 of III, and it is what precedes them that ought to have been reserved for das Innere. At § 297 of Colour I noted late in 1990 "My das Innere typescript starts here". That typescript, made in 1965, is now lost. Thanks to Dr Rothhaupt I now know that §§ 296 - 316 of III of Colour are the passages on das Innere without colour complications that are properly in the present das Innere volume, and thanks to a footnote on page 71 of the das Innere volume I know that § 317 was left out of that, having been put in Vermischte Bemerkungen by von Wright because it is about God. The trustees' bound photographs from which I dictated my typescript included both MS 173 and MS 174. Joseph Rothhaupt's Farbthemen in Wittgensteins Gesamtnachlass (Betz) has a table on page 368 which has been very helpful in piecing together my memories of the manuscripts and bound photographs. Quite a few paragraphs in MSS 173 and 174 as now printed were familiar to me in 1965 but one is nowhere to be found in the present volume. It compares trying to solve philosophical problems to needing to scratch an itch. In 2004 (which means after nearly forty years) I have found it and quote it ahead. One paragraph about itching and another about God may not seem much to make a fuss about, but they do typify the insensitivity of the original trustees' editing.

MS 173, which opens with a colour passage not printed here, is where, on printed page 61, I must now take up meaning and leave textuality behind. The opening points out that we can know that someone is happy by his facial expression without necessarily being able to describe what makes us certain. Or we can be wrong, or we can be unsure, or again we might feel that neither genuineness nor pretence hits the mark. Wittgenstein suggests that people we feel fall into neither category might relate to normal people like colour-blind people to those of normal sight. I find this a most unconvincing analogy, but it may explain why colour-blindness keeps complicating das Innere --- and in any case he changes his mind at the end of MS 174. Summarising in my own words two paragraphs on this page:(printed 61 -- we are still in MS 173):

Similarly, I can make reliable predictions as to someone's behaviour without being able to specify my grounds.

There could even be countries where ungrounded assertions about people's character made by other people who knew them well were accepted in a court of law.

Wittgenstein then declares that his aim is to to describe the rules of evidence for empirical sentences, but asks if this is a good way of characterising what we mean by mental attributes.

He goes on to try to do better. The idea that we must guess from their outside what is going on in people's inside, and that it is only our own interior that we know, has gone up in smoke (he seems to take for granted that he has achieved this successfully), and what remains is to express the truth of the matter. He attacks this task by a kind of via negativa. For example (on page 62), it is not true that the inner has become something external, nor that we have direct inner and indirect external evidence for the mental. Those are meaningless. What is meaningful is 'evidence for what is interior' and 'evidence for what is external'.

Similarly, it is not true that all we ever perceive is the external, indeed it is not meaningful, for if it were we should have concepts of the external and the interior. We just perceive the other chap's dubiety. And it often happens that we can describe what is within him and not what is on his outside. The connection between the two is a matter of these [reasonable] concepts. What we have is inner concepts and external concepts. I do feel that he has gone round the houses rather, but he hasn't contradicted himself, and he insists that all this to-do is not to magic the inner away, in agreement with my basic thesis. He sums his position up by saying (at the foot of this page) that inner and outer differ in their logic, and that logic really does make our metaphor of inner and outer comprehensible.

Two paragraphs at the top of page 63 continue this attempt to sum up. We do not need the concept 'mental' to justify the fact that some of our conclusions are indefinite. Rather, this indefiniteness illuminates our use of the word "mental". And when someone says that all we really see is the external, Wittgenstein, having put this in inverted commas to make it seem an over-simplified point of view, points out that there is truth in it. Talking about when people say this or that is to refer to external circumstances, and gives an impression that our aim is to explain the inner by means of the external, yet that emphatically isn't the case.

The next paragraph seems to me a red herring: the false impression above is caused by the fact that a language game is something external. So, strictly, language games are, but this fact never influenced anybody in grasping at an over simple picture of inner-and-outer.

The next paragraph contains the mysterious expression "die psychologische Aeusserung". This simply expresses what I have termed "giving vent". Naturally no evidence teaches us to give vent to our feelings.

The rest of this page is important for its summaries, but I think readers can examine them without my guidance. Page 64 has two paragraphs that appear to be asides about colour, but they will be found to be illustrations of how language games and concepts are constantly at work in what we say.

An important idea begins at the bottom of page 65, predictability of behaviour, and at the foot of page 66 this leads to: in what ways might a human body behave so as to give us no temptation to speak of inner-and-outer in respect of it? Wittgenstein's answer is "mechanically", and he goes on for many paragraphs to tease out the manifestations of the mechanical and the non-mechanical, but I can safely leave these to the reader and concentrate on a particular detail at the top of page 66. If behaviour were predictable down to fine shades that matter to us, might we give up speaking of inner as opposed to outer? The answer (in the form of further questions) seems to imply that in abandoning these metaphors we should have misunderstood psychological predictability. For (to give my own illustration) we might know a friend's behaviour so well that we can predict it confidently, but it will still be meaningful to ask him for a decision (Wittgenstein's words but not his italics). This hint is the closest I know to Wittgenstein's confronting the apparent paradox between predictability and free will. He maintained the freedom of the will resolutely, as Isaiah Berlin did, but I do not remember any explicit arguments for it. Berlin's line was that our attributing moral value to behaviour entailed that it must at least sometimes be unpredictable, which of course it very frequently is, but the hint here is that even in those cases where it is successfully predictable there remains a residual meaning to asking for a decision, which in my view entails that, however predictably one's friend decided, he was free to decide otherwise. As I say, this is a mere hint and what Wittgenstein is arguing for here is a residual meaning for the metaphor of inner-and-outer.

An important attempt at summary comes at the top of page 68. It is not that the relationship between inner and external explains the uncertainty of evidence, but the other way round: the relationship is a metaphorical depiction of this uncertainty.

In a longer paragraph on this page I feel that Wittgenstein is being too clever. Knowing what is going on behind someone's brow can be a matter of indifference to us, but in his post-war notebooks he has given many examples of his finding whatever this phrase means extremely important. Whether the question is trivial (like visualising blue or red) or matters to us (like the tribulations of friendship) is irrelevant to the phrase's meaning, but it is precisely in cases that matter to us that we are likely to use the phrase. And our uncertainty does indeed refer to the whole human being who responds to our friendship, but if trivial things like red and blue can be hidden from us it is reasonable that feelings about friendship can be hidden as well, and while it is true that the general mental contribution to friendship finds much expression in the bodily, there can still be hidden components, a proper description for which has still to be found.

In a sub-paragraph he says reasonably, about such contributions, that to an uncertainty about the inner there corresponds an uncertainty about the external; but this does not mean that there is no hidden component left for us to wonder about, and nor is any light thrown on this by the final sub-paragraph, which appeals to the completely irrelevant Fregean distinction between number and numeral in the bottom line of a bill.

And after all that, the next paragraph expresses just what I have maintained, that his arguments do not mean that, in general, uncertainty about the mental can be expressed as uncertainty about the external, and he gives an illustration of this: grief in its very essence expresses itself in our facial features, but I may have no readier way of describing such features than by the word "grief-stricken". In other words, we need an 'interior' word to express something external (distortions of the face). To me, there still remains a natural wish to describe subtle distinctions between different feelings of grief.

On this page my first edition has a mis-printing of "zugelassen" which I trust has been corrected, and on the next page, 69, there is a mistranslation: "inkonsequent" means "inconsistently". In between there is the sensible remark that a witness's views on what the accused thought might or might not be accepted in a court depending on the degree of acquaintance, followed by the absurd surmise that in some jurisdictions no utterance [being external] could be admitted if its only relevance was to mental events [being internal]. In MS 174, ahead, Wittgenstein admits this absurdity.

After examples of how we express ourselves in respect of other people's thoughts and feelings, he declares that our various language games about this reinforce each other in encouraging an inner-outer picture, with further reinforcement from the basic fact that we are frequently in doubt about the matter. Then, just as he previously imagined a world in which no-one lies, he now imagines one in which facial features are never disguised, and people are reduced to disguising their feelings by covering up their faces. The relevance of this is not clear, except that in the next paragraph (on page 70) it gives him an excuse for a metaphor which he dismisses here but makes much of in the next notebook, an inner face which is perfectly attuned to the truth, the soul's face as it were. The serious point of this paragraph, however, is to deny what he sometimes seems to have suggested, that uncertainty about someone's annoyance is simply uncertainty about his future behaviour. There is just uncertainty of criteria --- and lest we still think there is uncertainty about some interior entity, he brings in the inner face metaphor, only to deny it. And he denies it again in the next paragraph, where even if the other chap says truthfully that he is annoyed, this doesn't mean seeing an angry face inside him: we only have words from him whose meaning is still not clear to us. This might seem to mean that the meaning of the image of the inner face is still not clear, but in the next paragraph it is just the words "I was a bit annoyed" that aren't, because even if I say this of myself I don't know their application precisely. This seems absurd, but in the next paragraph he tries to make it reasonable.

For at the top of page 71 he protests that he never meant when he said that these words' application was indeterminate that one wouldn't know when to utter them, but that "One simply mustn't forget which connections are made when we learn to use expressions such as 'I am annoyed'." Unfortunately, in the paragraphs that follow he gets no closer to explaining what he means by connections than asking what the bare statement that certain people play chess without a king could mean. Knowing the normal rules of chess (like knowing when to say one is annoyed) leaves that 'abnormal chess' possibility in the air, and all kinds of different connections could be called in to offer an alternative --- so what connections could be called up if someone says he is annoyed but not in the normal manner? I find this airy fairy. No child who had seriously explored the ways games' rules can be changed would be taken in by it as a comparison. This chess paragraph, following a sub-paragraph about people whose language games differ from ours in crucial respects, is precisely where Wittgenstein has come back to das Innere after an interlude on colour, and perhaps he needed a rest from his problem of 'connections'. If so, in coming back to it I do not think he made great progress, and I recommend readers to read these paragraphs to give themselves a feeling for Wittgenstein's ideas on how "The rule-governed nature of our language permeates our life" without expecting very precise argument. These paragraphs run to two thirds down page 74, where the paragraph on God has been left out, and I believe the paragraph beginning "Ich beobachte diesen Fleck" is really part of the 'connections' context, while the one beginning " 'Die Psychologie beschreibt . . . ' " is where we came in with colour blindness and psychology intertwining with das Innere.

The next notebook extract, from MS 174, is written in Wittgenstein's best concise style, and is a serious attempt to summarise his ideas, including some passages where he has thought better of them. My memory of reading it in 1965 (and dictating some in translation) is what induced me, in my draft of my book on Wittgenstein, to put 'das Innere' with the concise notes on colour (now Part I of Remarks on Colour) and the best of On Certainty (I had in mind § 193 onwards, and in particular § 300 onwards, but detailed inspection in 2004 convinces me that everything from § 66 onwards is a unified whole) as 'last quartets'. Since there seems to be much resistance to giving Wittgenstein's late colour ideas this honour, I will repeat what I have said elsewhere on this site, that his telling analysis of the concepts of colour would have been a great deal better as concept analysis if he had kept up his early interest in colour science.

The first four paragraphs all make points made before, and with such concision that the reasoning behind them may not be obvious. Expressing pain straightforwardly does not relate to the pain as expressing it with pretence relates to the pretence, because one is in no sense 'giving vent' to one's pretence. Pretence is not as simple a concept as being in pain, because the former has to be learnt or at least in some way acquired. And even if acquired through the development of natural dishonesty, formulating it requires the child to be taught, and as a practical necessity this will entail teaching it to look out for evidence of dishonesty in others. The fact that the evidence available to us only gives probability to what another person is experiencing is less important to us than the fact that this pattern that is so difficult to describe matters to us for what it is evidence of.

"He smiled" can be infinitely important to us. But does a small twitch of the face have to be important to us? And if so, because of probable practical consequences?

Complications between das Innere and the concept of knowledge arise. "He cannot know what's going on within me." But he can guess. So all he can't do is know. Therefore we are only making a distinction in the use of the word "know".

There follows a remark taken from MS 169 (see printed page 49, where the second and third paragraphss are crossed out vertically in the manuscript to show that it has been used elsewhere, presumably here): does an astronomer predicting an eclipse say that we cannot of course know the future? We say this when we are in doubt about the future, like a landlubber about the weather. A cabinet-maker doesn't say that one cannot know whether his chairs will collapse.

So much of what follows relates das Innere to knowing, knowledge and certainty that one might think it has been printed in the wrong volume, but there is no doubt that the prime emphasis is on das Innere. One does often say that one knows someone was glad to see one, but what consequences does that have? Confidence in the assertion, and that other people will understand it (he does not say believe it), but this seems feeble in comparison with our interest in the truth, and what exactly is that? In an example of his trick of ascribing his own philosophical temptations to the rest of us, he says that we wish to project everything into the other person's interior. This is to evade the difficulty of describing the remark's field, a metaphor from physics and mathematics that is quite helpful, but one from chemistry is less so, but if you want to understand the metaphor you must read a little way ahead, because what he says immediately is disappointing. "But why do I say that I 'project' everything into the inner? Doesn't it reside in the inner? No. It doesn't reside in the inner, it is the inner."

I find this a vivid way of expressing what I have taken to be Wittgenstein's aim, to show how the expressions we use in our instinctive inner-outer language act as a veil across the real truth of the matter, while our interest in using those expressions, what makes them matter to us, is the core of the real truth of the matter. Yet here, at the foot of page 82, he says "And that is only a superficial logical classification and not the description we need."

On the next page he re-introduces a dismissed idea that I warned he was going to make much of, the face within. Imagine that the soul is a face, and when someone is glad this inner face smiles. Take this seriously --- but we still want to know what importance this smile (or any other facial expression) has. This could actually be our normal way of expressing ourselves: "His inner face smiled when he saw me" etc.

Now it is clear that the fact that we can imagine a society in which that was accepted idiom does not mean that it is any more than idiom, but Wittgenstein procedes to apply his best philosophical analysis to it, the upshot being that whatever makes the inner smile important to us makes the outer one important too. Then he confesses "It is not easy to realize that my manipulations are justified", and one can only suspect that they are not. Finally, if "I know that he was glad" certainly does not mean "I know that he smiled", then what I know and is important is something else --- for even for people who took the inner smile seriously the question of its meaning (ie, its significance for them) would remain.

At the foot of page 83 and the top of 84 we return to what might be accepted in a court of law, namely "I'm certain that he was glad to see me", provided I knew the chap well. Here (but Wittgenstein does not mean here in the courtroom but in the case of this claim) the practical consequences are clear --- and not my friend being acquitted but the behaviour I expect of him, and similarly but contrarily if I say I think he is pretending. But are those consequences what matters to me? What matters is that I feel wellbeing when this particular person (who has done this and this before) behaves in this way. And the 'in this way' is admittedly a very complicated pattern. If the reader finds my elaborations gratuitous, they are to emphasize that Wittgenstein is harking back to earlier notes that were less concise.

If one doesn't want to solve (double underlining, not capitals) philosophical problems, why doesn't one give them up? Solving them means changing one's point of view, one's old way of thinking. And if you don't want to do that you should consider them insoluble.

That is on page 6r of MS 174, and at the bottom of page 10r he thinks of an image to encourage him to keep trying, in the paragraph that took me nearly forty years to rediscover.

Philosophy hasn't made any progress? -- If someone scratches because he has an itch, must there be some progress to be seen? Is it otherwise not a genuine scratching or not a genuine itch? And can't this reaction to the irritation go for a long time before a cure for itching is found?

That should have been printed at the top of page 87, but going back to page 84, where the other paragraph was printed, there comes immediately beneath it:

One always presupposes that the person who smiles is a person, not just that a human body is smiling. Much else is presupposed. But with all those presuppositions behind me, someone's smiling pleases me. If I ask someone the way I prefer a friendly answer to an unfriendly one. And even more welcome, I feel I must add personally, is a smile from someone whom one is merely passing and will probably never see again. He presupposes the inner, he says, in so far as he presupposes a person, but I feel he misses something about human society by drawing his examples from people with whom he has contact, even the minimal contact of asking the way.

This context, where he has given a serious meaning to the inner, is where he says, putting the word in single inverted commas in my opening quotation, that it is an illusion, meaning, to quote him afresh in English, that what is illusory is "the whole complex of ideas that this word alludes to, like a painted curtain drawn across the scene of our actual use of [other] words".

Here, from a few paragraphs above, the involvement of das Innere in the problems of knowing and certainty returns, until the last two paragraphs of all. A change of mind indicated in those, an important paragraph on page 86, one on page 87 that mentions a personal attitude that turns out not to be a belief, and another change of mind at the foot of page 84, are the only things I need to comment on. Everything else is straightforward and should be familiar.

In MS 173, printed at the foot of page 68 and the top of 69, after saying that a statement about someone's thoughts might be accepted in a court of law, depending on length of acquaintance, Wittgenstein had gone on to take seriously the possibility that in some courts not even a [reported] utterance of the accused, being external, would be accepted if it were offered only to describe the accused's mental events. Now, in MS 174, on pages 84 and 85, he says it would be ridiculous for counsel to argue thus, in this case in respect of anger.

The paragraph on page 86 points out that there is a 'why' to which answers permit no predictions. He has not in mind, though he could well have had, Marxist predictions that are explained away when they do not hit the mark, but, in the first place, what he calls animistic explanations. Then, Freud's explanations, and Goethe's in his theory of colours. These give us an analogy, put a phenomenon in a context, and thereby reassure us. I think it is important to point out that the syndrome he is referring to here does in fact merge with an area of pretended predictions, and that these play a part in making the mere reassurance, which would be quite rational if it claimed to be no more than reassurance, spuriously scientific.

In the paragraph at the foot of page 87, Wittgenstein examines a possibility that he might appear already to have upheld: namely that "I know that he is glad" means no more than that I am certain of his joy, I react to him in such and such a way, and in particular without uncertainty [and don't expect anyone else to]. This would make the remark similar to "I know that everything is for the best", which Wittgenstein explicitly says expresses his own attitude to whatever might happen. The translators fudge this, I assume not being able to believe that their hero could say anything so apparently Panglossian. It is a point of view that he had expressed before in notes going back, I believe, to the thirties, and it is not Panglossian. As he says here, and I feel with understatement, there would be grounds for saying that this wasn't really knowledge, and the statement would certainly not convince anybody in a court of law that everything was for the best. The most accessible example of a similar attitude comes in the 1929 Lecture on Ethics (in the January 1965 Philosophical Review, pages 8 and 9), where the attitude "I am safe, nothing can injure me whatever happens" is cited. He declares it to be nonsense and a misuse of language, but (on page 12) it exemplifies ethics, which "does not add to our knowledge in any sense" but "is a document of a tendency in the human mind which I personally cannot help respecting deeply . . . ". In this printing there are also two quotations from Vienna conversations of December 1929 and December 1930, translated by Max Black. These depict Wittgenstein's views on ethics and religion vividly. They should be compared with the late 1937 Norway notebooks, MSS 117-120. The first quotation has been topped and tailed (see Wittgenstein und der Wiener Kreis), for it opened with an uncharacteristically kind remark about Heidegger and ended with a characteristically appreciative one about Augustine. The second has been spoiled by Black's mistranslation of "da ist" as "is there" instead of "exists". (Colloquially, of course, "Wer ist da?" means "Who is there?", but Wittgenstein has been talking about his first example in his lecture, a wonder at the fact that the world exists, or that there is a world at all, and what this wonder could mean.)

Finally, the last two paragraphs, on page 90. They hark back to the opening paragraph on das Innere of MS 173, on page 61. Appearing to feel neither genuine nor pretence joy but something in neither category would be, in relation to normal people, like being colour blind compared to people with normal sight. Now, quite differently, "It seems to me as little a fact that there can only be genuine or feigned expressions of feeling as that there can only be major or minor keys". This observation has stayed in my memory for fifty years, and with the help of Dr Rothhaupt's table on his page 368 I can explain how. I did know this notebook in 1952, but not all of it, and was familiar with the end of its das Innere passages, which came just before the beginning of its On Certainty passages, ie § 66 - § 192; and even though I digested the end and the beginning of those respectively, I only came to know them in full when I was lent the trustees' bound photographs in 1964 (and retained them well into 1966).

The final section, from MS 176, was one I knew much better in 1952, since it came in the knowledge and certainty passages that I was concentrating on then. I made a scrupulous copy of those and also began translating them when Elizabeth Anscombe left me looking after her family and her accumulated Wittgenstein manuscripts and typescripts (far from complete at that time, of course), in order to go to Vienna to check and make dictations from items that had been left there. She had encouraged me, as soon as the translation of Investigations was finished, to make a translation of my own of something. She was taken aback when I chose the knowledge notes, which I took rather uncritically as starting in their most telling form with § 300, because she said it would be a long time before they were published. I stuck to my belief that they were the most important part of the Nachlass. My impression is that she and von Wright had already formed a plan to publish knowledge, colour and das Innere last of all, not only because they were the last written but because they were the least important.

This intention was thwarted by an accident over which I had no control. Elizabeth told me off roundly when she found that I had shown my first draft of my knowledge translation to Iris Murdoch, whom I had innocently assumed to be 'part of the family'. She made me swear to show nothing to anybody without her express permission. I had the greatest difficulty in getting her to agree to my showing anything to Delia Macbeth for her criticism of my English style. A few years later a cousin of Delia's typed my translation and I hoped to show it to Isaiah Berlin, but he would only let me if I did so without Elizabeth's permission, which my oath made impossible. I did however still feel, in spite of my reprimand over Iris Murdoch, that it would be honourable to give a carbon copy to Yorick Smythies, who surely really was part of the family, and moreover could be trusted to know the Wittgenstein rules. This was when, in return, Yorick gave me the mid-war Investigations typescript that many years later, after a romantic history, was delivered to the Wren Library at Trinity by Eliane Flach. Unfortunately, Yorick got into debt quite soon after I gave him my carbon copy and settled it by giving that to a friend, who took it to America and circulated it on the black market. This was what made the trustees, against their will, publish what they brought out as On Certainty, commissioning me to complete the translation but not waiting for me to do so (I was shortly to take up an extremely demanding job lecturing in mathematics at Nottingham College of Education).

Returning, however, to 1952. Among the Wittgenstein notebooks, while Elizabeth was in Vienna, I found a little grey one, now listed as MS 166, with English notes on privacy. Not trusting Elizabeth to look after it I made a copy. I took advantage of my permission to show this to Delia Macbeth, who made some very cogent comments on how Wittgenstein's faulty English had trapped him in what she took to be philosophical errors. The notes were so captivating that I even read them into a tape recorder, on the Jesuitical grounds that I had only promised not to show anybody anything, though I never, in the end, took advantage of this.

And returning not only to 1952 but to my muttons, the notebook in which I made my privacy copy had some spare pages at the end, and it seemed appropriate to use these to translate the brief interpolation on das Innere which now completes the printed book, under the heading of MS 176. Wittgenstein had made his intention that this was separate from the notes on knowledge so clear that, although I included them in my transcription complete with their separating horizontal lines, I was never tempted to include them in my main translation. There is a similar interpolation, an extremely brief one mentioned above, in MS 176's early colour pages, printed as a numbered paragraph on page 9 of Colour, between square brackets -- in the manuscript it is between vertical lines. I also find that some personal remarks in this April 1951 region are included in On Certainty, but one is not, and so I will add it at the end, together with one written in that March. Prior to that, to relieve myself of the duty of commenting on or tacitly changing the Luckhardt and Aue translation, I shall give my 1952 translation of the interpolation, made in my spare pages later that summer (for I certainly would not have wasted the precious weeks of Elizabeth's absence with a translation that that had no hope of being published). The main theme of this, starting in its sixth paragraph, is examining what difference it would make to our concepts if we could use something along the lines of a clinical thermometer to detect other people's pain.

The interpolation begins after § 523 of On Certainty and comes, like that, under the date 14.4.[51].

/ "Can one know what is going on in another person in the way he knows it himself? --- How does he know it himself? He can (for example) express what he feels. A doubt for him whether he is really having this experience - analogous to a doubt whether he really has such and such a disease - does not enter the game; and it is therefore wrong to say he knows what he is experiencing. But the other person can very well doubt whether the one [ie the first chap] is having this experience. Doubt, in other words, does enter the game, but it is just for that reason that it is also possible for entire certainty to exist.

Must I be less certain that someone is suffering pain than that 12 x 12 = 144?

And yet one does sometimes say, that is something one can't know. Well, to start with, it is something one can't prove. That is to say, there is no such thing in this case as a proof based on generally recognised principles.

/ But how can I see something that is inside him? Between his experience and me there is always self-expression! [nZ] This is the picture: he sees it directly, I indirectly. But it isn't like that. He doesn't see something and describe it to us.

If 'something is going on in him' of course I don't see it, but who knows whether he sees it himself. --- --- ---

Don't I, in fact, often see what is going on in him? "Yes, but not in the way he sees it himself. I see that he is in pain, but I don't feel any pain myself. And if I did feel a pain it wouldn't be his." That means nothing. --- On the other hand it is conceivable a connection could be set up between us so that I felt the same pain (ie the same kind of pain) as he, and in the same place. But that's being the case would have to be established by our respective expressions of pain.

And if this means of getting to know another person's pain had proved itself, then conceivably one would make use of it in despite of someone's expression of pain, in other words one would mistrust what someone said if it contradicted this test. [nZ] And then one can surely also imagine that there are people who have followed this method from the beginning, and call that pain which it gives as its results. In that case their concept will be related to our own but different from it. (But of course it isn't a question of whether they call this concept by the same word as we do our related one, but only of whether it corresponds in their lives to our concept of pain.)

To this analogon of our our concept the uncertainty of evidence would be lacking. In this respect our concepts would not be similar.

(If we call this analogous concept "pain", then these people will be able to believe that they have a pain and also to doubt whether they have. But if someone says: "But then there is just no essential similarity between the concepts" --- we can rejoin to this: there are enormous differences here, but great similarities as well.)

One could imagine a kind of clinical thermometer being used to settle whether someone was 'in pain'. If someone cries out or groans they put the thermometer in him and only begin to pity the poor fellow and to treat him as we treat someone who 'is obviously in pain' if it shows such and such a reading.

Does the logic of our concept of pain hang together with the actual non-occurrence of certain possibilities of thought and feeling [literally, of reading thought and feeling]? If that question is one of causality, how can I answer it?

The question could really be put like this: how does what is important to us depend on what is physically possible?

Where measurement is unimportant to us we do not measure, even if we can.

15.4.[51]

"Is the impossibility of knowing what is going on in someone else a physical or a logical one? And if both, --- how do the two connect?" [nZ] First and foremost: possibilities of discovering about other people can be imagined which do not exist in reality. So there is a physical impossibility. [nZ] The logical impossibility lies in the lack of precise evidence. (Hence our sometimes expressing ourselves in this fashion: "we can always make a mistake; we can never be certain; what we observe can always be simulation." In spite of the fact that [other people's] simulation is only one of many possible causes of a false judgement.) --- We can imagine an arithmetic in which problems involving small numbers can be solved with certainty, but the results become more uncertain as the numbers become greater. So that people who have this manner of calculating can never be quite certain of the product of two big numbers; and no boundary can be given between the small numbers and the big ones. [nZ] But naturally it isn't true that we are never certain of the mental processes in someone else. We are in countless cases. [nZ] And now the question remains, whether we would give up our language-game which rests on uncheckable evidence if we had the chance of changing it for a more exact one that in certain cases had similar consequences. We could, for example, work with a mechanical lie detector, and then define a lie as what is given by a reading of the apparatus. [nZ] The question is therefore: should we change our form of life if such and such were put at our disposal? And how could I give an answer?

This is followed by § 524 of On Certainty, under the same date. Two private asides come soon after, the one about doing philosophy like an old woman losing her spectacles, printed correctly under § 532, and the one about pretensions, which should come after and not before the date 18.4. If any library in America has a copy of my black market typescript, it will reveal an embarrassing mistranslation. Convinced that my German was too good to need a dictionary, and impressed by Elizabeth's story of Wittgenstein's being held back by drugs, I jumped to the conclusion that a Hypothek (a mortgage) was something one bought in an Apothek, and translated it "a drug". Between these two asides, however, and following § 534, the following paragraph, marked off by Wittgenstein with vertical lines, has been left out.

| One cannot judge oneself if one is not at home in the categories. (Frege's style is sometimes great; Freud writes excellently, and it is a pleasure to read him, but he is never great in his writing.) |

Some psychologically revealing notes had been written in MS 175, after its earlier notes on knowledge, and under the date 10.3.51, when the last knowledge notes began, in which Wittgenstein polished again and again his views on philosophers whom he regarded as his plagiarisers. These were Ayer, Wisdom and a third whose name escapes me. In one set of these notes he used an analogy of someone picking up a bunch of keys and having no idea of which doors to unlock with them. Obediently to Elizabeth, I did not copy these but simply noted "| Interpolation |", except for copying the final version of a similar set which mentions neither names nor key-rings. I must excuse myself from trying to give its paranoid flavour in translation.

Das entschuldigt die Unredlichkeit derer nicht, die ihren Veröffentlichungen durch meine von mir nicht veröffentlichten Einfälle (Beispiele, Methoden), ein Ansehen verschaffen. Denn wenn auch, was sie davon tragen können, nicht wertvoll ist, so halten sie selbst es doch für wertvoll, und es ist auch besser, als was sie selbst erdenken können.

Comments will be welcome at: admin@wittgenstein.co.uk