From the ALWS archives: A selection of papers from the International Wittgenstein Symposia in Kirchberg am Wechsel

I

On 22 October 1915 Wittgenstein wrote to Russell, teling him that was doing a satisfactory job of synthesis in treatise form, which would be finally fit for publication. Publication could not take place before Russell looked at the work, and thus this would only be possible after the war. But he could not survive: therefore he invited Russell, in this case, to ask for all his manuscripts, among which he would base the final summary written in pencil on loose sheets.¹ Taking in consideration this letter in his "Historical Introduction" to the first edition of Prototractatus, von Wright does not think that speaking about a “last" version implies that there must have been at least another one, and that at the time there existed two "Zusammenfassungen"; it can be only derived from this - he supposes - that the Prototractatus note-book, preliminary drawing up of the Tractatus Logico-Philosophicus, in turn had at least one predecessor (Wittgenstein 1971, 6). Von Wright thought that the Prototractatus was composed immediately before the final version, in summer 1918; with much caution, and it was only 30 years later that McGuinness (McGuinness 2002) came to the conclusion that both of von Wright theses are wrong. On October 1915 already two prototypal versions of the Abhandlung existed, and one of these was the Prototractatus (more exactly, the first pages of Prototractatus note-book).²

In the dating of literary works, one usually agrees to consider the moment of completion, imagining a sequential process on whose end the entire job is dated. But there are cases that break off the outline and completely puzzle the classic interpreter. The Prototractatus note-book is one of these, because it accompanies the entire composition of the Abhandlung, from 1915 to 1918, and in every moment, by virtue of its structure and the ingenious expedient of its numbers, it turns out to be up-to-date. In fact the modifications that followed over the course of time consist in adding new propositions and do not consider the already written text. Since material organization is based upon the decimal numeration, the note-book does not require any rewriting and it is indeed in every moment the exact documentation of itself. It happens therefore that it turns out to be prior to as well as later than October 1915. Not in the banal sense of a work in progress, and therefore of indeterminate and insignificant dating, but rather in the sense that in every moment the Prototractatus is a last copy. It is enough to read it watching to the numeration, and not to the physical order of the phrases. Apart from this, the composition order is totally different from the sequence in which we are accustomed to reading the Tractatus: to say, the cardinal propositions 1 – 6 are all on the first page of the notebook; the "comments" that Wittgenstein gradually adds can be connected to whichever previously inserted proposition. Such a top-down technique of drawing up gives to the work a second, valuable characteristic: it is also, in every moment, complete. It can have more or less details, a comment in one or in another point, but it appears always virtually finished, totally readable. This particularity is doubly loved by Wittgenstein, who risks life in any moment and who, in any case, always thought he was destined to die a premature and unexpected death. Nevertheless, in October 1915 Wittgenstein claims to have constructed a still better version than the Prototractatus notebook, a representation method on melted sheets which he advises Russell to take mainly into consideration as "last summary".

II

We can reasonably assume that on October 1915 the notebook reached the separation line on page 28 (on 121 in total): the "original" part of the treatise, which was not resumed from previous manuscripts, was nearly completely carried out and comprised all the entire structure and various detail coppers. What did the letzte Zusammenfassung suggest to Russell to possess more? Here McGuinness puts forward an important hypothesis (McGuinness 2002), that nevertheless neglects Wittgenstein’s overall sense. He imagines that the version on loose sheets contained the same propositions of the notebook, but in numeration order, exactly as the final typescript. McGuinness thinks that later on Wittgenstein wrote up other versions ordered by decimals, some on typescript, some on intermediate note-book, some on melted sheets. But in so doing he disregarded the fundamental structural difference between the notebook and the typescript, on one side, and melted sheets, on the other. A drawing up in tightened decimal order is sequential by definition, and demands a sequential support. Loose sheets are of opposite nature, have the faculty and the aim to be put in different orders and they certainly do not introduce any advantage; they demand indeed sagacity and attention in order not to be melted and for the sequence not to disappear. Why then does Wittgenstein choose just a melted sheets support, as he specifies to Russell? If we place such question, McGuinness hypothesis isn’t tenable. Can we think that at the moment Wittgenstein did not have a normal notebook and predisposed just for Russell a package of sheets that goodness knows in which order and after how many passages would have reached him?

Imagining the route of the sheets from Wittgenstein to Russell, we can approach the kernel of the objection. In this type of interpretation one is inclined to discuss only one side of the semantic of "the last": that is, the reference to a temporal order, to a question of “before” and “after”. It is neglected instead the observer point of view: when something turns out "last", and how longer eventually it can remain such. Since that was a letter in which Wittgenstein speaks about his recent work, it was considered obvious that “die letzte Zusammenfassung” was “the last” only at that time when Wittgenstein writes the letter. Obviously - thinks who has in mind a normal sequential copy - very soon die letzte Zusammenfassung is no more the last one, and is forgotten, rejected, integrated in other versions (see McGuinness 2002). But if we examine the step more thorough, we see how Wittgenstein nearly represents the moment in which Russell will read his papers: if I don’t survive, get my people to send you my manuscripts, and among them you’ll find the last summary made so and so; it will cost you some trouble to understand it etc. All makes think that the main interest of the letzte Zusammenfassung is to be still the last version at the moment in which Russell will have it in his hands. Die letzte Zusammenfassung is such not only because was composed in der letzten Zeit, i.e. little before 22 October 1915, but because it will be the final synthesis, that is the most up-to-date, also when Russell will receive his manuscripts.

If Wittgenstein’s letter was that one of a condemned in the imminence of his execution, the two interpretations would be equivalents: the state of manuscripts during the letter writing down would be the same one at the moment of the testamentary bequest reception. But Wittgenstein cannot know if and when he will dead: who knows whether I shall survive the war, he writes; and he expects a long war. Therefore he’s imagining that in case of his sudden death, perhaps after some month, perhaps after some year of further work, among his manuscripts there will still be the Zusammenfassung written in pencil on loose sheets, and it will be indeed classifiable, for Russell, as die letzte. How could he prefigure something of this sort, if the drawing up on loose sheets was a simple sequential rewriting, immediately obsolete and quickly to eliminate? Nevertheless Wittgenstein feels himself truly, in an obscure way, condemned to death; his letter is at all effects a testamentary disposition. Make Russell read die letzte Zusammenfassung is not a simple fantasy, but a precise will, that formally obligates Russell as Wittgenstein himself. Therefore the letter, while engages Russell as publishing executor, equally engages Wittgenstein to make possible what he promised. His punctilios rigor would not tolerate the idea that Russell would not found any letzte Zusammenfassung written in pencil on loose sheets, any future time Wittgenstein would die.

A testamentary disposition remains valid until its revoke, or until it is exceeded by ulterior actions: in this case, until when, on 1919, Wittgenstein writes to Russell again and he succeeds to send him the definitive version. Until that moment, or at least until August 1918 dictate, we must presume that, in force of his engagement, between Wittgenstein’s papers remained die letzte Zusammenfassung written in pencil on loose sheets. The hypothesis seems fantasious because a loose sheets support is rather unusual, and above all because, in general, it is difficult to understand how an on-paper drawing up can remain so along the last version, in spite of its radical restructures. On the contrary, if Wittgenstein had had a PC and had written to Russell: "you’ll find in the 'Dokumente' directory the file: letze_Zusammenfassung.doc", we would not doubted that this would have been the last synthesis for every following month, and not only in the moment in which the letter was sent. Analogous, if we admit that the flying sheets support had a structure finalized to favor a continuous upgrading of the propositional device, it becomes obvious that Wittgenstein’s words testify not only the existence of a copy written "on loose sheets", but the persistence of the support and the relative method of representation for the entire period of treatise composition - in symbiotic parallel with Prototractatus note-book.

III

This allows a more precise hypothesis on the real structure of the text on loose sheets, that cannot be that one assumed by McGuinness. Sure the melted sheets must respect the propositional numeration, but without falling back in the absolute sequence of final Tractatus editing. In order to use the heuristic hypothesis that the Tractatus is an hypertext organized by hierarchical levels, one has to imagine that the melted sheets structure was very liked as that one of electronic virtual pages. The method with which a today computer science engineer plans a Web site can help us to understand the sense of Wittgenstein job. If the engineer was forced to work with paper and pencil, he would predispose an increasing set of single sheets, on everyone of which to develop different sights. In case he decided to represent the connections by means of absolute numerical indices, a good solution would be just the Wittgenstein decimal notation, that specifies pages virtual positioning and links sequences inside every sight. Structural modifications are done simply changing the numerical index of the single object and they correspond to different arrangements of paper sheets, or of objects inside the same sheet. On paper, this is the best way in order to visualize the dependencies net, which becomes more and more articulate and impossible to reconstruct in abstract on the basis of only pointers.

The package of loose sheets remains an optimal work support for all the planning phase and does not demand any overall rewriting. On the contrary of whichever sequential drawing up in decimal order, the version on flying sheets is always up-to-date and available for several representations - by hierarchical lines, sequences of equal level, local or general structures. Additions are possible in every part of the configuration, in particular by putting new objects at the bottom of every sheet, or by inserting ulterior detail sheets – witch is exactly the proceeding way that Prototractatus note-book documents (see Geschkowski 2001, pp.64-66). Probably, every new proposition is put by attempt into different places, reconstructed thanks to various sheets dispositions. Once accepted the new object, it is recorded inside the destination sheet and, with the consequent decimal number, on Prototractatus note-book.

If Wittgenstein therefore had not clearly insisted on the "last synthesis written on melted sheets", all this would not be that one pale conjecture. But if we put that indication beside the other clues we have, the picture becomes finally clear:

    a) The choice of a loose sheets support, with all the risks of conservation, ordering etc., is comprehensible only if it’s strongly finalized to a modelling work

    b) Such version in 1915 is imagined as "die letzte Zusammenfassung" for all the incoming period; that is, it’s always upgradeable and always concluded, ready for editing

    c) The Prototractatus note-book, at least after page 28 separation line, has evidently a parallel alter-ego from which to obtain the number of every new proposition and on which to reorganize the propositional cascade; otherwise, the continuous numeration modifications remain inexplicable

    d) The reference to "melted sheets", that some times seems to return and that puzzled critics, is not else than the echo of only one gradually increased package of sheets. Therefore, the manuscript “written in pencil on loose sheets with numbered propositions" that Heinrich Groag remembers to have had in winter 1917-18 is not properly another new version, like von Wright said (Wittgenstein 1971, p.6 fn.2). Neither it’s necessary, for economy of versions, to anticipate the episode to the previous year and to imagine a memory fault only because in a letter datable in spring 1917 Wittgenstein already asked Groag to return his manuscript sheets (McGuinness 1988, p.265). The loose sheets accompanied the entire Tractatus composition

    e) The 1918 "Korrectur", cited on Prototractatus page 103, that involves a nearly complete propositions renumbering, necessarily happened on another support, used at least until August 1918 final dictate

    f) After the "Korrectur", the Prototractatus operativity remains unchanged and the job proceeds in the same exact way of the previous pages, also in presence of a new numeration and therefore with necessary reference to a parallel document. This appears so incomprehensible to the Prototractatus editors that they decide that the note-book properly finishes at page 103 and they don’t publish the successive 18 pages, which they consider simple copy of Tractatus parts and not a preparatory work to it. Viceversa, these pages continue to collect new additions, obviously with the definitive numeration that the loose sheets now have. Some proposition continues to be composed and reviewed on the note-book, with all the corrections of a first drawing up

    g) The melted sheets instrument is perfected during its use and allows the tuning of the decimal indexes, which is testified – but not explained – by the note-book. The enigma of the numeration, origin of infinite arguments, is all in being the product of a long formal elaboration that is carried out in part on Prototractatus (moreover very rarely taken in consideration from critics) and in part elsewhere, quickly growing rich of multiple uses and means. From a point on, the ulterior evolution remains invisible; the "Korrectur" pointed out on page 103 exalts an architectonic and aesthetic function that was able to be fine-tuned only through the formal work on melted sheets.

IV

The representation on loose sheets is not the only possible one; Wittgenstein, so as our hypothetical computer science engineer, considers it a work support. It isn’t evidently proposable for a sequential edition, because it would lose its main value: the possibility to freely reorganize the pages at different deepening levels. While the engineer can decant the sheets in their electronic analogous, Wittgenstein is forced to decide for one definitive paper sequence. This can only become in tight numbers order, assigning to every proposition its absolute logical place. The iconic-formal composition remains at reader responsibility, who must have the necessary ability to logical abstraction. But if an exhaustive representation turns out nearly impossible for a normal reader, it would remain to ask how could therefore Wittgenstein create a so pregnant architecture, in converging of logical, aesthetic and (for their needing power) ethic requests. The elaboration by loose sheets, to which Wittgenstein points out, answers well to the question, because it allows complex structural alternatives and it was able to represent much sophisticated formal aspects. This hypothesis naturally emerges in experimenting the effectiveness of the hypertextual transcription of the Tractatus, which exalts similitudes, symmetries and shapes that Wittgenstein fine tessellated and assembled in one only harmonic organism.