TS 239, the Smythies Typescript
Yorick Smythies gave me this typescript in 1957, in return for a copy of my Certainty translation, which at that time consisted only of § 300 onwards. He imposed on me such strong conditions as to what I could do with it that I was not at all sure whether it counted as my property, though I have since been given legal assurance that it did. In 1966 I offered it to Eliane Flach, explaining these conditions and making it clear that it could only be a loan. Unfortunately, she forgot this restriction and let it fall into the hands of her boyfriend, whose interest in it was that he might be able to sell it for enough money to enable him to leave her and set up home with somebody else. I discovered this deplorable fact in the autumn of 1977, and asked her to do what she could do to recover it so that it could find a permanent home in the library of Trinity College, Cambridge. Her negotiations with this entreprenurial type were made easier by the fact that he had already tried to sell it to Cornell University, where the librarian had refused to recognise his claim to own it, and sent him packing. Eliane did recover it, and in the spring of 1978 brought it to Cambridge, where, in my presence, she handed it to Philip Gaskell, the librarian. He asked me who had the title to it, and I answered "Yorick Smythies". I have never been able to discover whether he received the College’s letter of thanks before he died.
I asked Trinity to give me, for my efforts, a photocopy of the typescript and to send another to Norman Malcolm, who I thought was the person who had refused to buy the original. Malcolm wrote and thanked me, and told me that he was arranging for the photocopy to be added to the Cornell series of Wittgenstein microfilms. He did not mention the part the Cornell librarian had played, and I only discovered this when I examined this last Cornell microfilm later. That, incidentally, makes no reference to the fact that it is made from a photocopy.
The typescript begins with a copy of the Preface to Philosophical Investigations, dated, like the printed Preface, 1945, but not identical to it For example, the last sentence of the third paragraph of the Preface as printed is missing from the typescript, and can be found written on its own on page 22 of MS 130. Similarly, the opening phrase of the Smythies Preface was "In dem folgenden teile ich Gedanken mit", but the change to "In dem Folgenden veröffentliche ich" is noted on page 55 of MS 130. These early pages of that manuscript were almost certainly written in1945, but presumably after the Preface to the Smythies typescript, which is dated, like the printed version, ‘im Januar 1945’. Another significant difference is that whereas in print the reacquaintance with the Tractatus (from reading it with Bachtin) is described as taking place ‘vor vier Jahren’, in the Smythies typescript it is ‘vor zwei’. The only reasonable explanation I can give for this (though it presupposes an untypical weariness with detail on Wittgenstein’s part) is that he completed his full typescript (one copy of which can be seen in photocopy at the Bodleian --- the other was apparently lost by the printers) in 1947, made this adjustment as a dating clue, but left other clues as for 1945, including the date. There is evidence in some book of reminiscences which I no longer have to hand that the reading with Bachtin took place in 1943, not 1941.
1943 is my guess for when the main Smythies typescript was prepared --- not as a fresh typescript but one using a carbon copy of the pre-war typescript, TS 220, with many manuscript changes added in ink, with two extra manuscript paragraphs inserted on additional pages (§§ 117 and 142 in print) and a number of pages in the middle rearranged, crossed out, retyped or omitted. Among the latter is an important clue to dating. TS 220 included, somewhere in this much amended region, the phrase "die sprachwidrige Verwendung des Wortes ‘Gegenstand’ und ‘Komplex’." I no longer have easy access to that typescript, and the notes I made in 1976 when I read it on microfilm do not include the phrase’s position, but it can be found in manuscript in MS 127, mostly written in 1943. There the phrase comes with Tractatus references, I assume connected with the Bachtin readings. The phrase also goes with Wittgenstein’s dissatisfaction with his picture theory of language. I hope I am not going too far in saying that dropping it from the Smythies typescript went with a tendency to cover up his tracks, and copying it into MS 127 with a guilty conscience about doing so.
Although the carbon pages of the pre-war typescript were used up to and including page 77, radical alterations start at the foot of page 71, and it is here that I recommend a detailed study to begin. Running into page 72, two paragraphs are crossed out, beginning "Betrachte auch diesen Satz" and "Das Bekenntnis zu einer Ausdrucksform" --- while the next paragraph, § 100 in print and part of § 91 in the pre-war TS 220, is left, except for a tantalising parenthesis crossed out at the top of typescript page 73. This deserves to be quoted in full. It includes an alternative written in ink, presumably while the parenthesis was still being considered for inclusion.
(Es is ähnlich, als wenn Du sagtest: //als sagte Einer:// "Der Umfang dieses Rades ist wirklich D.pi"; so genau ist es gearbeitet.)
When I worked on the 1929/30 manuscripts this observation haunted me: it was just what I needed as an example of unattainable practical precision, but I could not remember where I had read it. Clearly, in my 1976 reading of TS 220 --- but when I did find it, it was no use to me, because as the printed § 100 shows, the context there was an ideal of language, not an ideal of practical workmanship or measurement. The latter problem may be the cause of Wittgenstein’s dissatisfaction with the discarded paragraph mentioned above, beginning "Betrachte": for it goes on to compare the leeway a language game leaves its users to the freedom of someone imprisoned between elastic walls. These must have a precise degree of elasticity, argues the arguer. Certainly --- in a practical sense, but not with the precision of the ideal that Wittgenstein is tilting against, and one can understand his not wanting to hold up his discussion with this kind of detail.
On typescript page 73 another paragraph is crossed out, the crossing out queried, and then the English "No" wtitten and underlined. It was:
Wie kann ich den Satz jetzt verstehen, wenn die Analyse soll zeigen können, was ich eigentlich verstehe? --- Hier spielt die Idee des Verstehens als eines sonderbaren geistigen Vorgangs hinein. ---
The first two sentences of what follows this are in print as § 102, but then the typescript continues without being printed for some three paragraphs, until § 105 (much cut) begins at the botttom of ts page 74, followed by §§ 106, 107 and the first sub-paragraph of 108, followed by §§ 109, 110 and 111. With philosiphical depth, more serious textual problems begin, and before trying to disentangle them I must mention some other details.
§§ 101 and 103 both appear on typescript page 73, above § 102, but § 104 is missing in the typescript, and of the paragraphs of ts page 74 not printed, one is of particular interest., because it appears to declare that Wittgenstein has, at last, a theory, but he makes it clear that it is a dream by adding the manuscript footnote "Freud spricht von einer ‘dynamischen’ Theorie des Traums".
Wir haben nun [ms, besitzen] eine Theorie; eine ‘dynamische’ Theorie des Satzes, [ms, der Sprache,] aber sie erscheint [ms, uns] nicht als Theorie. // Es ist ja das Charakteristische einer solchen Theorie, dass sie einen besonderen, klar anschaulichen, Fall ansieht, und sagt: "Das zeigt, wie es sich überhaupt verhält; dieser Fall is das Urbild aller Fälle." --- "Natürlich! so muss es sein", sagen wir und sind zufrieden. // Wir sind auf eine Form der Darstellung gekommen, die uns einleuchtet. Aber es ist, als haben wir nun etwas gesehen, was unter der oberfläche liegt. [Not all textual details noted. One is that after "sagen wir" there is a typist’s comma that Wittgenstein has crossed out. The double lines presumably mark a passage that he was thinking of changing. There is a barely legible manuscript addition in the margin that does not refer to this or the following paragraph but to the typescript original of § 105.]
The margin and top of typescript page 75 has further manuscript changes that contribute towards the final printed version of § 105. Underneath that comes, not changed at all, printed § 106, but underneath that is a parenthetical paragraph about which Wittgenstein had second thoughts. His first was to move it up to come between §§ 105 and 106, and his second to cross it out altogether. It contains an echo of the 1929 struggle with the phenomenological language ("die letzten Feinheiten" of § 106). This paragraph was
(Auch in diesen Ueberlegungen rührt das problematische nicht daher, dass wir noch nicht auf den Grund der Erscheinungen gekommen wären; sondern daher, dass wir uns in der Grammatik unserer Ausdrücksweise, die Zeichen, die physikalischen Gegenstände betreffend nicht auskennen.)
After that we have, from the foot of typescript page 75 to the foot of page 77, comparative textual peace. The final parenthetical sentence of § 108’s first sub-paragraph is crossed out in the Smythies typescript but re-included in print, while the second, third and fourth sub-paragraphs are not in the typescript at all. It is between §§ 111 and 112 that serious problems return. These are heralded by a short paragraph cut from elsewhere and set about an inch to the left of the rest of the page. Its opening phrases, up to the first dash, are added in manuscript, but give an impression of being copied from a cut up typescript rather than being newly composed.
Eine ähnliche Gedankenbewegung: Wie kann man die Zeit schätzen --- da das Leben doch fern von einer Uhr ist? --- Dass uns die Zeiten übereinstimmend mit der Uhr einfallen; dass wir die Zeit schätzen können; ist ein Grund, warum, was die Uhr misst, die Zeit, so wichtig ist.
What follows at the top of the next page (with no numeral) is part of § 98 of TS 220 but does not appear in print at all, and begins (with the "tortois" of that typescript corrected by hand):
Worin liegt z.B.die Tiefe des Witzes: "We called him tortoise because he taught us"? Wir werden plötzlich aufmerksam darauf, dass eine solche Ableitung des Substantivs unmöglich ist.
This has an extraordinary echo. In February 1929 Wittgenstein began writing in a large manuscript book which he called "Band I" (MS 105), using at first only right hand pages for his philosophy and filling in the left hand pages subsequently, to the confusion of unwary scholars. These pages can now be read quite easily, and in the order in which they were written, in the first volume of the Nedo-edited Wiener Ausgabe (Springer Verlag). On page 19 (a right hand page) of that manuscript there is a passage that was not dictated into TS 208 and so does not appear in Philosophische Bemerkungen, based on cut and pasted paragraphs from that typescript. It begins (in my English) "Against my will I am apparently being forced back to arithmetic", and discusses the relationship between cardinal numbers and the objects they count, represented by nouns. On the next right hand page (21) this enquiry leads to the problem of the derivation of nouns by some kind of definition. Wittgenstein attempts to define the noun "Liebespaar". Translating out of his predicate calculus, one could try to go from "Jack is going for a stroll, Jill is going for a stroll, Jack loves Jill and Jill loves Jack" to "this couple is going for a stroll" via an attempted "Jack-and-Jill-love-each-other is going for a stroll", which he gives up as impossible. No juggling can force this attempt into a definiendum defined by a definiens because, while "couple" is a noun, "Jack-and-Jill-love-each-other" is not, and writing "complex" underneath it cannot turn it into one.
In other words, the tortoise who taught us is simply Wittgenstein’s code for his 1929 failure to derive the noun "Liebespaar".
Things get even curiouser than that, however. The unnumbered typescript page we have reached continues, after "unmöglich ist", with a shift to the right. It continues to use material from TS 220. It suggests that if the impossible were accomplished the philosophical depth of Carroll’s joke would be lost, and goes on to mention a joke of Lichtenberg’s.. This is from his "Briefen von Mägden über Literatur", in which one maid, in a letter to another, writes of a hundred as "001". The joke itself is not quoted and we are merely told that if we just say that there is nothing to stop a hundred being written like that we miss its depth. Then Lichtenberg’s title is crossed out. Then we are told that the depth will be apparent to anyone who can draw the mathematical consequences of the maid’s error. These sentences are crossed out with a vertical line, indicating that they may have been transferred elsewhere, while further sentences of explanation (which can be identified on microfilm by the word "Grundstellung") are crossed out altogether and replaced in the margin by "Wir empfanden ihn als tief, weil er ein grelles licht auf das System unsrer Sprache zu werfen schien". Finally, written more carefully in the margin, is
In Lichtenbergs "Briefen von Mägden über Literatur" schreibt eine Magd der andern:
with a colon suggesting that the joke is going to be quoted after all. And it was, in TS 227 of 1947, the typescript from which Investigations was printed. It was a beautiful joke, going far beyond "001", and I remember it very well. I was very disappointed when it did not appear in print in 1953. I also remember very well that Elizabeth Anscombe translated the whole of the Lichtenberg passage. The explanation only came to me in 1993 when I examined the Bodleian’s photocopy of the surviving copy of TS 227. In that, Wittgenstein had crossed everything out --- the tortoise, the maid’s dialect German, 001 and all, including, if my 1976 notes are correct, the passage about a verb’s ‘Grundstellung’ that Wittgenstein had crossed out in the Smythies typescript decisively and doubly.
While it is clear from this that the lost copy (I cannot say which was top and which carbon) was the one Elizabeth Anscombe translated from, it does not follow that she was responsible for its loss. Much the most likely explanation is that she recalled the more corrected copy from the publishers for checking purposes (including excising my lamented Lichtenberg maid), and replaced it by the less corrected copy, which I guess was passed on to the printers for proof reading and lost.
At the foot of the Carroll and Lichtenberg page in the Smythies typescript comes a paragraph which is not crossed out but does not appear in print. Its typing is continuous with TS 220 but I did not note it in 1976 as being in that. I quote it here.
Die Tiefe der Absurdität liegt hier inVerhältnissen, die eine längere Erklärung zulassen; weil sie den eigentümlichen Bau unserer Sprache betreffen. --- Wenn wir auf das System unserer Sprache sehen, dann haben wir das Gefühl der Tiefe. Es is, als sähen wir durch ihr Netz hindurch die Ganze Welt.
What follows in print as § 112 is marked by me as part of § 98 in TS 220 (§ 111 was § 97 in TS 220), and begins "Ein Gleichniss", but on the Smythies page (also unnumbered) it has written in front of it in pencil "Das phil Problem". The next printed paragraph, § 113, does not come until the bottom of that typed page, and I have noted it as being part of § 110 of TS 220. What comes in between is largely crossed out, but includes the printed § 114, its opening Tractatus quotation being put in by hand but giving the impression of being copied from TS 220 (and also noted by me as part of § 110 of TS 220, if anyone wishes to check).
The next Smythies item is the manuscript of printed § 117, but while printed §§ 115 and 116 are noted by me as coming from §§ 109, 110 and 111of TS 220, they are not to be found in Smythies. That, instead, jumps to its numbered page 90, with "Woher nimmt die Betrachtung ihre Wichtigkeit", § 118 in print, paired with § 119 as part of § 111 in TS 220. Our next textual anchor is § 134 in print, on page 94 of Smythies, which began in TS 220 (crossed out in Smythies) "Lass uns zu dem Satz zurück kehren: ‘Jeder Satz sagt:’ ", replaced in print by "Betrachten wir den Satz:" and in Smythies by nothing. Between these two safe textual markers, there come important revisions in print which are prepared by revisions in Smythies to which they are far from identical.
Indeed, the first of these two textual anchors is best taken as including § 112 of MS 220, which extends into printed § 120 with two extras. The first of these is another of those where I find echoes of the 1929 phenomenological language quest but can never convince other people of this. Is everyday language perhaps too coarse and material for what we want to say? Und wie wird denn eine andere gebildet? Wittgenstein had tried to fashion two other languages, the formal one of the Tractatus and the immediate-experience-only one of 1929, and I read his italics as an expression of the unworkability of both, however much he still admired the Tractatus as an attempt worth preserving. The second extra is the last sub-paragraph of printed § 120, a word’s meaning not being an entity of the same kind as a word. I see no revision of ideas in this but only a polishing of a point often made.
Printed § 121 is then inserted at the foot of Smythies page 90, cut from slightly earlier in the original typescript, and "Wenn die Philo" copied by hand. In Smythies "Wenn die Philosophie . . ." is the paragraph’s opening, but in § 220 that was prefaced by "Hier gehört auch:" and in print by "Man könnte meinen:". Where on the last line of print there is "eine solche", the original typescript had "eine R" and in Smythies this is changed by hand to "Rechtchreibelehre" (my Fregean inverted commas, of course).
From there until we reach the second anchor (in other words printed paragraphs 122 to 133) the revision is achieved by a cutting up and reordering of old typescript pages (of which 82 and 85 have their numerals visible) and what may be a later typescript with smaller type, of which the numeral of page 83 has survived. Paragraphs that are completely new in print are § 125, on which I have commented in my "45 Minutes", and the last sub-paragraph of § 133. Paragraphs 122 to 131 were noted by me in 1976 as coming from §§ 100-107 (with omissions) of TS 220, while printed §§ 132-133 are from the old typescript §§ 113-116, and thereafter, from printed § 134 = old typescript § 117 to the typescripts’ end, the printed book, TS 220 and Smythies march reasonably well together, with minor changes of which I will note the more important when they are reached.
The important qustion, of course, is whether one can conclude, from the changes in the much revised region sketched above, anything about the progress of Wittgenstein’s ideas during and immediately after the war.
Textually, I note that the next Smythies page, with no number, has small type, but visible against its opening paragraph, § 122 in print but with the slightly different opening "Eine Hauptquelle unseres Unverständnisses ist,", there is the old typescript numeral "100", which I noted in 1976 from the Cornell microfilms of TS 220. From this it is reasonable to infer that these small-type pages were part of the 1938 dictation and not freshly dictated in 1943. I should be very grateful if scholars who have access to the original TS 220 or microfilms of it would let me know. At present I cannot even reach the nearest Cornell microfilms. Similarly (but this would require access to the Smythies original in the Wren) there is the question of whether these rearranged pages have their paragraphs not only cut but pasted. That is certainly the impresion given by my photocopy, and of the small-type pages as much as the large.
After what is now § 122, § 123 is added to the Smythies typescript by hand. Then comes § 124, with Ramsey mentioned as the author of its quotation, but the whole of this mathematical sub-paragraph is crossed out, and so is a revealing paragraph that follows.
Ein Gleichnis gehört zu unserem Gebäude; aber wir können auch aus ihm keine Folgen ziehen; es führt uns nicht über sich selbst hinaus, sondern muss als Gleichnis stehen bleiben. --- Wir können keine Folgerungen daraus ziehen. [The rest ia crossed out more decisively.] So, wenn wir den Satz mit einem Bild vergleichen (wobei ja, was wir unter "Bild" versstehen, schon früher in uns festliegen muss), oder die Anwendung der Sätze, das Operieren mit Sätzen, mit der Anwendung eines Kalküls, z.B. des Multizipierens.
The cutting out of language as a kind of calculus goes with the post-war admission of language-free thoughts, as I hope to show in a later study, but what is revealing in the less firmly cut out sentences is not so much a change of mind as a wish not to give a false impression. Not being able to draw consequences is a serious matter. The viewpoint Wittgenstein wants to put over is expressed at the bottom of this typescript page by what is now § 126, separated in print by § 125, the 1945 interpolation on mathematics. "Die Philosophie stellt eben alles blos hin, und erklärt und folgert nichts" and "was etwa verborgen ist, interessiert uns nicht" are not remarks in which I find anything to apologise for. Much of philosophy’s task is to state what only a convoluted mind would think of denying --- but that philosophy has nothing whatever to do with drawing conclusions . . . that has a kind of New Age ring that Wittgenstein could no more have welcomed than Russell or Popper would.
The next Smythies page is of full-sized type, is numbered 82 and has no hint of cutting or pasting, but has two paragraphs decisively crossed out. That leaves only the second sub-paragraph of printed § 126, with § 127 and § 128 (the remark about "Thesen", in italics in print but spaced in typescript, the title of Waismann’s aborted book, always making me want to say "Poor Waismann!"). The first discarded paragraph awakes memories which make me think it was in the lost copy of TS 227, but I cannot be certain. I may be remembering TS 220, in which it was not crossed out.
Wenn Einer die Lösung des ‘Problems des Lebens’ gefunden zu haben glaubt und sich sagen sollte, jetzt sei alles ganz leicht, so brauchte er sich zu seiner Widerlegung nur erinnern, dass es eine Zeit gegeben hat, wo die Lösung nicht gefunden war; aber auch zu der Zeit musste man leben können, und im Hinblick auf sie erscheint die gefundene Lösung als ein Zufall. Und so geht es in der Logik. Wenn es eine ‘Lösung’ --- wie eines mathemathisches Problems --- der logischen, d.i. philosophischen Probleme gäbe, so müssten wir uns nur vorhalten, dass sie einmal nicht gelöst waren (und auch da musste man leben und denken können).
Between §§ 127 and 128 are two sub-paragraphs, a parenthetical one not printed and a normal one crossed out.
(Die Anlage der Philosophie beruht auf der Fähigkeit, von einer Tatsache der Grammatik einen starken und nachhaltigen Eindruck zu empfangen.) And:
Das Lernen der Philosophie ist wirklich ein Rückerinnern. Wir erinnern uns, dass wir die Worte wirklich auf diese Weise gebraucht haben.
The next page is numbered 83 and is in large type. It opens with a paragraph which is marked by hand "[gehört nicht hierher]" but does come in print (§ 129, with "Die philosophisch wightigsten Aspekte") followed by three which are not crossed out and are missing from print.
The first of these includes the phrases "das erlösende Wort" and "ein Haar auf der Zunge", so familiar that I feel I must have known them in 1952, and it is faintly crossed out. The second is about hitting a false thought so neatly on the head by an accurate phrasing of it that its perpetrator immediately admits that it is exactly what he thought and must therefore be wrong. This procedure is compared to psychoanalysis. The third paragraph is set about a quarter of an inch to the right, and so may be cut from another page.
This paragraph has an echo of what comes a little ahead in TS 220, the old opening of printed § 132. This opening is neither in print nor in Smythies, but read "Da unser Ziel ist, den Bann zu brechen, in welchem uns gewisse Sprachformen halten, so wollen wir...". The unused Smythies paragraph begins "Der Bann in dem uns eine Analogie hält, kann gebrochen werden, wenn man ihr eine andere an die Seite stellt, die wir als gleichberechtigt anerkennen."
The next Smythies page returns to large type. Its page number is obscured by four lines cut from a similar page. These, with three more from the current page, form the first of three paragraphs of which the middle one is the most interesting. It half-echoes a point made in the unprinted ‘theory’ paragraph on typescript page 74. In that a complaint was made about taking one case as typical of all. Now, we are encouraged to take a particular application of a logical calculus as what it is --- an example --- provided we do not take a set of examples and say of them that they are not the ideal that this calculus is really meant for. "Kann ich den Kalkul überhaupt verwenden, dann ist das auch die ideale Verwendung, und die Verwendung, um die es geht."
This paragraph and the next are crossed through very lightly, and the second of these is an allegory on the same point. There is a collection of bundles of rods, and someone declares that the rods cannot be the ‘real’ bundles, because they can break off and fall out of them. How can such impure things be counted with our pure, clear numbers 1, 2, 3, . . . ?
The next page has its number, 85, and is in big type like all the remainder. It opens with the end of the paragraph just summarised, and there is a skewing that suggests that the second of its two full paragraphs is from some other page. It is printed § 130, and its displacement led me to note it in 1976 as not in TS 220 at all. The paragraph above it is printed as § 131, and the "nämlich" originally referred to the bundle analogy, was then crossed out in Smythies, and reinstated to refer to the "Vergleichsobjekte" paragraph when that was placed above it (in TS 227).
The page that follows has no number, and it opens with a paragraph already mentioned, printed § 132, in which a long opening clause has been replaced in Smythies and in print by "Wir wollen". Printed paragraphs 132 and 133 correspond almost exactly to three numbered paragraphs on this page, except that the last sub-paragraph of § 133 (‘therapies’) appears neither in TS 220 nor in Smythies.
Following that, typescript page 94 introduces printed § 134. The TS 220 opening, "Lass uns zu dem Satz zurückkehren: ‘Jeder Satz sagt:’", has been crossed out illegibly in Smythies, with nothing replacing it, while print has "Betrachten wir den Satz:". Moreover, the idea that "Es verhält sich so und so" is simply one sentence among others is elaborated, with "Der Himmel ist blau", and with the chap who needs an advance turning up a second time, reported as saying "p --- und ich brauche daher einen Vorschuss". (Elizabeth Anscombe had a joke about Wittgenstein saying innocently in a class or lecture "Would you know what I meant if I said ‘Miss Anscombe, p’?")
This paragraph ends on typed page 95, including an unprinted sub-paragraph:
Es wäre mir, z.B., nicht eingefallen, statt jenes Satzschemas die Form "es so" zu setzen, und doch könnte in einer Sprache, die ([crossed out] wie z.B. die russische) keine Kopula verwendet, dies sehr wohl als Satzvariable gebraucht werden.
With printed § 135 on the same page (of TS 220 and Smythies equally) we reach textual plane sailing and I shall mention only the more significant points of difference that remain. One recurrent pattern, which I leave people to investigate for themselves, is that short passages from TS 220 that fail to appear in print are crossed out in Smithies. Of course, my assumption that the Smythies text, aside from the preface, was rearranged from a copy of TS 220 in 1943, does not imply that these deletions were made then. Indeed, it is very unlikely that they were made before 1944-5. A more trivial recurrent detail is that where the first person singular pronoun is used, it is "Du" in typescript (presumably at Wittgenstein’s request, because he wrote it thus in manuscript), but is changed to "du" by hand on the Smythies pages. In a romantic mood I once dreamed that he wrote "Du" in his notebooks because he felt that he was addressing his philosophical antagonist in a letter --- that is how one writes to one’s friends but not how one reports speech in print. I hope native German speakers will tell me if this could be defended seriously.
The first large textual addition comes between typed pages 100 and 101, where printed § 142 is inserted on a full sheet of blank paper with the number 158, which dictates its place between § 157 (printed § 141) and § 159 (printed § 143), the numbers on page 101 having been altered to prepare for this insertion. A manuscript draft for the new paragraph can be seen (faintly) in the margin and top of page 101.
Between printed §§ 148 and 149 a whole sub-paragraph is dropped from page 106 of the typescript, part crossed out in Smythies and part simply dropped in print (but after many manuscript changes).
What is more interesting and requires to be detailed is what has happened to the printed paragraphs 149 and 150. "Einen solchen Zustand nennt man eine Disposition" was originally "Einen seelischen Zustand in diesem Sinne will ich eine Disposition nennen." The following printed sentence is completely new, and the final parenthesis of § 149 had originally "irreleitender" for "verwirrender". The comma that follows is Wittgenstein’s, correcting his typist, and the original "hier" is crossed out in Smythies and replaced in print. § 150, on the grammar of "wissen", had, in TS 220 but crossed out in Smythies, "Denn ich verstehe --- schon seit Jahren --- wie eine Dampfmaschine funktioniert, wie ich seit jahren das ABC weiss, und Schach spielen kann", with "Einmaleins" and "ein Spiel" put in by hand in Smythies with, finally, the "(Eine Technik ‘beherrschen’.)" of print put in by hand.
In printed § 134, the fourth sub-paragraph is added in the margin of typed page 109, while the parenthetical fifth sub-paragraph is not present at all. Before printed § 135 appears on typed page 110, however, there is a long crossed out paragraph which I hope is not important enough for me to quote. The changes from printed § 135 on until the end of printed § 176 are certainly minor, except that the final parenthetical sub-paragraph of printed § 166 is a later addition.
What separates §§ 176 and 177 (and two experiences of the because) are a crossed out sub-paragraph and a parenthetical one that is merely queried. They are (on typed page 130):
Vergleiche damit diesen Fall: Jemand soll sagen, was er fühlt, wenn ihm ein Gewicht auf der flachen Hand ruht. --- Ich kann mir nun vorstellen, dass hier ein Zwiespalt entsteht: Einerseits sagt er sich, was er fühle sei eine Pressung der Handfläche und eine Spannung in den Muskeln seines Arms; anderseits will er sagen: "aber das ist doch nicht alles; ich empfinde doch ein Streben des Gewichts nach unten!" --- Empfindet er denn ein solches ‘Streben’? Ja: wenn er nämlich an das ‘Streben’ denkt. Mit dem Wort "Streben" geht hier ein bestimmtes Bild, eine Geste, ein Tonfall; und in diesen siehst du das Erlebnis des Strebens.
(Denke auch daran: Manche Leute sagen, von dem und dem ‘gehe ein Fluidum aus’. --Daher fiel uns auch das Wort "Einfluss" ein.)
Printed § 182 is missing from TS 220 and the Smythies typescript alike.
The Smythies typescript ends, as TS 220 does, on its page 137, but in Smythies the page is torn and the number is missing with more than half the page. It is fortunate that the lower half is what has survived, because this is a clue as to when TS 220 was dictated. Von Wright has expressed the view that a previous typescript was dictated in 1937 and discarded, TS 220 itself coming in 1938. TS 220 and Smythies end with the first sub-paragraph of printed § 189. This was not written until the 11th of September 1937, when it was written twice --- in MS 118 (Band XIV), and then fair copied into MS 117 (Band XIII) on the same day, which appears to have been a febrile one, in Norway. The journey back to Vienna might theoretically have left him a few days for dictation at the end of that year, but practically one must assume that the dictating was done in 1938 (in Ireland or back in Cambridge).
A memory involving Yorick Smythies has come to me from 1945, which I should like to end by mentioning. In October of that year, when the Michaelmas term had begun, I visited Erich Heller before joining a section of the Friends Ambulance Unit which was stationed in the Russian Zone of Germany, north of Berlin. Talk turned to Wittgenstein. Erich declared that his philosophy was sheer nonsense. He knew this with absolute confidence because he had attended (I think "gatecrashed" would be a proper word) one of his lectures and had been able to understand none of it. Afterwards he had had coffee with a group of students, his description of one of them sounding exactly like Yorick. He (assuming my identification is correct) told Erich that it was impossible to summarise Wittgenstein’s philosophy --- one had to persevere with studying it if one wished to understand it. This assurance did not convince Heller at all. Fortunately, a fortnight or so later, in Lüneburg on my way to my remote posting, an FAU colleague showed me a copy of the Tractatus and told me forcibly that I was wrong to be prejudiced. What I should like, having told that story, is to ask a question which it poses. What records are there of what Wittgenstein lectured on in the academic year 1945-6? Records of the following year are copious and well-known Have any notes taken in 1945-6 survived? My postal address is given at the end of my "45 Minutes", and with e-mail I can be reached at admin @ wittgenstein.co.uk.
In August 2004 I must add a further note. With the help of the Oxford/Bergen electronic edition I have been able to make a more precise analysis of the Smythies typescript, to be found in Chapter 8 of my book, and if anybody should find discrepancies between that and the above, the book is to be preferred. I also came to the conclusion that the typescript was not so much 'mid-war' as 'late-war', having been made after Wittgenstein left his second hospital, in Newcastle, in February 1944 (see the Moore-Malcolm letter of Oct 13 / 44). And a point has occurred to me a propos of this typescript being given me in return for a copy of my early translation of what is now § 300 onwards of On Certainty. To pay a debt, Yorick gave it to someone who flogged it in America, where it was circulated. Discovering this was what made the trustees decide to bring the full notes out earlier than they had intended and to ask me to translate them. In Additional Note 2 in the second edition of his Memoir, Malcolm says that only after its first publication in 1958 had he known of Wittgenstein's notes on the subject, knowledge, that he had discussed with him in America. They were not published until 1969, and it now seems to me much more likely to me that Malcolm first met them in their contraband circulation, and that he had been the person who alerted Elizabeth that my translation was on the market. (That, incidentally, contained a howler confessed in my book.) However, in whichever form he first met them, it seemed to me typical Elizabeth insensitivity that she had never told him about them, just as she never showed the notes to Moore. My disapproval of both failings is the last item in my book.