ABSTRACT
When one studies Wittgenstein’s post-Tractatus work, twin enthusiasms develop. The first is to wonder at the impressive progress of his ideas, from February 1929 to April 1951. The second follows later, as one gets deeper into his Nachlass-the extraordinary way in which he edited and re-edited his expressions of those ideas. He only gave this passion up at the very end, when he no longer had any time for it. Even Remarks on Colour is much edited. The second volume of Last Writings on the Philosophy of Psychology known as “Das Inner und das Äußere” is too scattered in its notebooks for one to know what he might have hoped to do with it. Only On Certainty, and in particular from §300 to the end, flows so well as to convince one that it is as it was meant to be. The most extreme example of successive editing is what was printed as
Part I of Philosophische Grammatik, edited for Blackwell by Rush Rhees.1 Its publication date is 1969, but I received unbound page proofs from Blackwell in January 1970 and it was only late in that year that it actually saw print. In 1976 Anthony Kenny published a criticism of Rhees’s editing.2 I cannot claim to have read it but I must at least have heard about it on the grapevine, and it reinforced suspicions of my own that began early in 1976 when a sabbatical six months enabled me to devour the Cornell microfilms, on which, in spite of their imperfections, nearly all the relevant manuscripts and typescripts had been recorded. My further researches during that half-year made me even more suspi-
cious, for I had also obtained (from Rhees, in Elizabeth Anscombe’s absence) permission to read originals in the Wren Library of Trinity College, Cambridge, starting with Band X. This, rather puzzlingly, consisted of two parts, the first being the end of a long series of dated first or second drafts, its last date 4.6.32 on page 41 of that volume, the second the opening of an undated revision. In between, TSS 212 and 213 (the Big Typescript, derived from 212) had been dictated, and the revision in question was of the first 404 pages of 213 minus a segment that Wittgenstein apparently did not think needed revising so much, in print as the opening two sections of Rhees’s Part II of Philosophische Grammatik, called “Logischer Schluss” and “Allgemeinheit.” All the remainder of 213 similarly escaped minute revision.
In other words, the Big Typescript falls into two parts, a much revised part consisting of its first 404 pages, minus the sections I have mentioned and a few smaller ones, and the little-revised remainder. Rhees ought to have made clear that his Part I was the upshot of repeated
Wittgenstein revisions, and given a clear account of the form those revisions had taken, not calling his volume anything but Philosophische Grammatik simply. Then he should have printed the totality of the Big Typescript in a separate volume, entitled as such and not confusing us by referring to Philosophische Grammatik at all. (The Anhang to his Part I was cobbled together from the smaller sections mentioned above and from various other sources. As it is, in the middle of Rhees’s volume, it is a serious distraction to the reader.) In any case, from his so-called Part II Rhees had dropped three major Big Typescript sections, called “Philosophie,” “Phänomenologie” and “Idealismus etc.,” and they are extremely interesting and important, but they are now easily accessible, not only in the Bergen Electronic Edition but in two editions of the complete Big Typescript: Michael Nedo’s in the form of Volume 11 of the Wiener Ausgabe and a recent Blackwell edition, edited and translated on opposing pages by Luckhardt and Aue. What primarily concerns me here, however, is what I mean by calling
Philosophische Grammatik’s Part I the upshot of repeated revisions, because misunderstandings about this go back to the Kenny article in Acta Philosophica Fennica.3 What I had heard about this seemed to me entirely reasonable in 1976, until, at the Wren Library, I moved from the second part of Band X to the first part of Band XI. There I next found a further revision in the form of a large loose-leaf manuscript called Grosses Format, MS 140, and finally a manuscript written on both sides of about 100 half-sized pages called Kleines Format. This had, and still has, no assigned G. H. von Wright number, a fact that no one has ever explained to me. It was having all four manuscripts at one large Wren table that enabled me
to find the missing clue. This was that in all of them there were markings between paragraphs directing the reader, or Wittgenstein himself in making a dictation, from one manuscript to another. I also had my printed volume with me and was able to compare it with these directions and found that they tallied almost exactly. Unfortunately I did not publish this discovery and was content to mark details in the margins of my printed copy. I still have the most important of these, one directing from page 38 of Grosses Format to page 180 of Kleines Format (taking one to the opening of §41) and another, after three paragraphs of §42, taking one back to page 38 of Grosses Format. Anscombe arrived at the Wren Library as I was examining these manu-
scripts and was much reassured to be shown that Kenny’s criticisms were unfounded. That seemed to me to be all I needed to do with my discovery. I did, however, report it to Isaiah Berlin, who was as pleased as Anscombe to be told that Kenny was wrong. Some years later I was told by the Wren that Kleines Format was missing,
in spite of my having returned it in the normal manner. I had done so under
the Wren’s original coding, but now they used only von Wright numbering and so had no way of tracing it. The only alternative to its being somewhere in the Wren but inaccessible is that Anscombe might have been so intrigued as to take it home with her, which as a trustee she had every right to do, in which case one can only hope that it will turn up among her papers when they are catalogued. In 1994, when he was launching the first volume of his Wiener Ausgabe
edition, Nedo told me that a typescript vindicating Rhees’s editing had been found in Moritz Schlick’s Nachlass. With the help of Alois Pichler of the Wittgenstein Archives at the University of Bergen and Brigitte Parakenings of the Philosophical Archive at the University of Konstanz I have been able to obtain a copy. It is catalogued at Konstanz as D.5, but it turns out to be only a copy of the original, held by the Vienna Circle Archive at Haarlem in the Netherlands and catalogued there as Mulder V. It stops short where §42 ends, just after the Kleines Format insertion, followed by three short paragraphs from elsewhere. Nevertheless, it is a vindication as far as it goes, for it coincides with the printed version with very few inclusions and omissions. It is perhaps significant that the next manuscript written on the same large
paper as Grosses Format is the first sketch for the Brown Book, written in German (MS 141). That, and its subsequent English version, forms an opening sketch for Philosophical Investigations. So as well as corroborating Rhees’s Philosophische Grammatik editing, it is also possible that the Schlick Mulder V dictation betrays Wittgenstein’s loss of interest in that particular version. The fact still remains, however, that it was the last version before Wittgenstein turned his attention to writing the Philosophical Investigations. Now I had always assumed that Kenny’s criticisms of Rhees were a con-
sequence of his not knowing the editorial signs that indicate this version (a virtual version, as Nedo calls it), but his article in Wittgenstein: The Philosopher and His Works,4 volume 17 of the Working Papers of the University of Bergen, puts the matter in a different light. First, he reveals that he had discussed Rhees’s editing with him, as a result of being commissioned to translate the book, in which case it is inconceivable that Rhees would not have shown him these editorial marks, or at least some of them. But then, second, he uses a phrase that reveals his attitude:
