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

1. Warning

On presenting the so-called Prototractatus notebook to the press, von Wright pointed out the enigmatic nature of the remark at the manuscript’s beginning, which so recites: "Zwischen diese Sätze werden alle guten Sätze meiner anderen Manuskripte gefügt. Die Nummern zeigen die Reihenfolge und die Wichtigkeit der Sätze an. So folgt 5.04101 auf 5.041 und auf jenen 5.0411 welcher Satz wichtiger ist als 5.04101" (Wittgenstein 1971, p.39). "A student of the manuscript - von Wright observed - is struck by the use of the accusative in `Zwischen diese Sätze' and by the phrase `werden alle guten Sätze... gefügt'. The question arises which are the sentences, of which is said that in between them all the good sentences of Wittgenstein’s other notes will be inserted. My conjecture is that they are the sentences listed on the first text-page, or possibly first few text-pages. Perhaps the note at the very beginning was written after Wittgenstein had on the first text-page written down the remarks which so to speak constitute the backbone of the entire work" (Wittgenstein 1971, p.3). The attempt to face the question posed by von Wright leads us to try a more complete inquiry: what, when, where, who and above all why.

2. What

The first phrase of the note distinguishes between "these sentences" and "all the good sentences of my [of Wittgenstein] other manuscripts". Von Wright tries to identify the first ones; the answer becomes easier if one asks, instead, which are "the good sentences of the other manuscripts" that are inserted between those at the moment in the notebook. If we examine the current content of the notebook, it is easy to see that no citation from other manuscripts does appear until the horizontal line traced at page 28, after very 282 sentences that, so long as we know, are formulated ex novo.¹ Beginning from such line, 29 consecutive sentences follow, drawn from the 1913 Notes on logic; and so on. In total, the literal or nearly literal citations from known manuscripts are beyond 200, all successive (with one exception) to the line of page 28. We can therefore deduce that "these sentences" of the present manuscript, to which the note referred, are all those that precede the page 28 line.

If this is true, then the remark is not describing a state of fact: it rather prescribes a series of operations to do in order to complete the text according to the author’s intent. In particular, the accusative of motion into place indicates that such job is in some way traced through numerals, but it is still in action, in an indeterminate stage at the moment in which the manuscript was taken in consideration. The part of von Wright’s analysis tending to establish when the note has been affixed becomes therefore crucial.

3. When

The content of the warning leads us to exclude that it can have been formulated immediately after the first text-page. It is more logical to think that it was written only once Wittgenstein arrived at page 28, since until that point there is no trace of the idea to systematically use material from any previous manuscripts. But when in effect the note is written up? That is: when Wittgenstein finishes the first 28 pages of the notebook?

According to (McGuinness 2002), the first 70 pages of the Prototractatus were composed between April 1915 and August 1916, in twofold shape: on notebook and on loose sheets. If we connect the structural change (implied from the passage from the notebook to the scattered sheets) with the methodological change (from the former drawing up ex novo to the systematic retrieval of previous material), it becomes natural to distinguish within these seventy pages the first layer, with all evidence directly composed on the notebook. Once Wittgenstein reached page 28, he would have written down the note and would have continued the job, now consisting in finding good sentences from other manuscripts, using the support offered by loose sheets, which better favour complex structural compositions. Therefore, in October 1915, when we have notice of a loose sheets support, the draft of the notebook is suspended at page 28; the numeration method now also concurs to virtually connect material of assorted origin, written in effects in physically different but logically coherent sequences.

4. Where

Thanks to the virtual structure determined by the numerical notation, the effective physical site in which the sentences are recorded becomes relatively indifferent. The diaries proceed in date order. Prototractatus notebook collects sentences in order of draft and/or of finding. The version on not-consecutive scattered sheets presumably reproduces essentially the sentences of that one, but by hypertextual pages (that is, for levels and subsequences). Such a version is moreover collecting the usable sentences drafted from previous manuscripts, by placing them on respective page and so assigning the right number to each. It is reasonable to think that Prototractatus notebook already carried its title ”Logisch-Philosophische Abhandlung”. The back page was still white, since Wittgenstein can have inserted the dedication to David Pinsent only after Mrs Fanny Pinsent’s letter that on 6th July 1918 announced his death. The back of the sheet with Kürnberger’s maxim, in front of the first text-page, is free too. So, where did Wittgenstein write his warning? The editors of the bilinguist 1971 edition reproduce it on the first text-page, over sentence number 1, but Wittgenstein really hasn’t put it near the text. He wrote it on notebook’s very first page, before the title, the maxim and the other free pages. Perhaps this note, unlike that one in the Tractatus, does not properly address the final reader.

5. Who

Who is therefore the addressee of the warning? Evidently, the addressee is the eventual publisher, an editor of Wittgenstein’s papers who, opening the notebook, must be made aware of its possible use. So the remark acquaints him that the notebook contains only the skeleton of the work, which is still in progress: it has to be integrated with sentences of previous manuscripts, separately documented. Wittgenstein had often indicated Russell as his posthumous editor, even when the forecast of his imminent death sounded less outlandish; now, the hypothesis is anything but remote, and the papers are predisposed to such evenness. The purpose is reaffirmed in the letter of 22nd October 1915: "If I don’t survive - he writes to Russell – get my people to send you all my manuscripts: among them you’ll find the final summary written in pencil on loose sheets. It will probably cost you some trouble to understand it all, but don’t be put off by that". The difficulty "to understand it all" does not refer as much to the content of the sentences, as instead to the unusual structure of the text, that has to be entirely recomposed.

McGuinness thinks that, as can be assumed for the following year’s typescripts, the loose sheets listed the sentences in strict numeration order, according to the method of the final printed version. However numbered typescripts and loose sheets don’t have the same structure at all. A drawing up in numeration order is closely sequential, it doesn’t favour inserts and it doesn’t take advantage of a single sheet support. On the contrary, loose sheets – if exactly every sheet is dedicated to a specific level of numeration – allow alternated orderings and different reading modalities; on this reticulum of sheets it is easier to make additions and transfers. It is the technique today used for hypertextual documents. It doesn’t astonish that Wittgenstein did not think it was immediate not even for his elective addressee, Bertrand Russell, who even had a good familiarity with the decimal structure used also by Principia Mathematica.

6. Why

A last issue is: why to conclude the remark with such a specific example, and why just with the three propositions of the 5.401 branch? The reading of the passages themselves offers one answer that is in some way amazing.

5.041

In particular, the truth of a proposition p follows from [folgt aus] the truth of an other one, q, if all the truth-grounds of the former are truth-grounds of the latter.

5.04101

We also say that the truth-grounds of the one are contained in those of the other; p follows from q.

5.0411

That one proposition follows from an other, we can see it from the structure of the proposition.

As we can see, the sentences chosen to exemplify the way in which a sentence of the Abhandlung follows [folgt auf] an other, by consequence of theirs numerals, deal in effect with the manner how a sentence logically follows from [folgt aus] an other. Wittgenstein evidently suggests that his propositions follow from the departure sentence in a some analogous sense in which in logic p follows from q. To enunciate comments to the sentence q is equivalent to explicate the content of q. Wittgenstein’s example suggests that his way to enunciate tends to find the intimate nexus between thoughts. So his propositions can be cited outside context, but the general sense of the work cannot be set aside from the organic relation between them, on vertical and on horizontal line.

A dramatic turn of events happens finally: in the Tractatus, the sentence 5.041 has been visibly corrected, because it contained a logical error. The truth-grounds of a complex sentence q are all those combinations of truth-values of the elementary propositions that render the complex sentence true. Given a list of states of things represented from a truth-table with 4 rows that arranges two elementary sentences a and b, let q their logical product (a . b) and p their logical sum (a v b). We have that p logically follows from q; however, q has only one truth-ground (only one line of the four) while p has three truth-grounds. Therefore it is not true that all the truth-grounds of the consequence p are comprised between the truth-grounds of the premise q, but it exactly the opposite is true. The consequence has more truth-grounds in general than the premise from which it follows: its grounds are not ”contained in those of the other", like instead also 5.04101 says. The error is discovered in 1922 by Ramsey during the job of translation; Ramsey suggests that there was an oversight, an exchange between "ersten" and "zweiten" in sentence 5.041, now numbered as 5.12. Wittgenstein answers in a very dry way, accepting the relief:

Your correction, the transposition of ”first" and "second", is quite right. Is "first" and "second" in the wrong order in the German too? If so, please set it right. Otherwise leave the German as it is (Wittgenstein 1973, p.31).

The correction, although necessary, really distorts the whole passage. In effect it seemed natural that the premise sentence had more content than the single consequence: "If p follows from q, the sense of p is contained in the sense of q" (5.04102). However, the content of a sentence, how much it says, does not correspond to the truth content as number of "truth-grounds". If Wittgenstein had become aware in advance of the oversight, he could have made a step ahead, rather than to limit himself to a formal correction that leaves one inelegant situation. The content of a sentence is not in function of its possible verifications: it is not the number of "V”s in the truth-table to measure the empiricist content of a sentence, as rather the number of "F”s! In order to maintain the intuitive sense of "comprehensiveness" and "content", it would have been unavoidable to enter the road taken by (Popper 1934), recognising that the content of a sentence is proportional to its falsifiability, to the number of states of things that correspond to the "F"s of the table. In this way, the parallelism suggested by the sentences cited in the note to the Prototractatus would be totally strengthened and it could induce to an interpretation of the structure by decimals, so as the note suggested, still more meaningful and essential.

7. Epilogue

After writing the remark at the beginning of the Prototractatus, a little before October 1915, the systematic rereading of previous manuscripts begins, to find by successive screenings the best usable material. The new sentences are brought back on a loose sheets support – sheets organised as hypertext’s pages. With the final 1918’s restructure, the architectonic complexity and the formal design become so expressive that they can totally be appreciated perhaps only with an analogous representation by hypertextual pages.²

The final skim focuses on the second diary (and than the following one, today lost); the "good sentences" ultimately recovered have been copied (perhaps on little sheets singularly numbered) and marked at the same time on the notebooks. Of these first 42 special marked phrases, only four³ are comprised in the final typescript (but no more in the Prototractatus). A similar origin reasonably had the "supplement 72", stuck in extremis to the typescript sent in 1919 to Russell (and through Dorothy Wrinch to Ostwald for the publication in the Annalen der Naturphilosophie). The adding was not printed by Ostwald, who in its place leaves the reference: "4.0141 (Siehe Ergänzung Nr. 72)". The dance of the flying sheets, with the sentences to insert in between those of the book, continues until the end, and perhaps beyond. Until when, after an incautious solicitation of Ogden, publisher of the English edition, about the eventuality of other possible supplements, Wittgenstein loses his temper: "The supplements are exactly what must not be printed – he answers on May 1922 - To let you print the Ergänzungen would be no remedy. It would be just as if you had gone to a joiner and ordered a table and he had made the table too short and now would sell you the shavings and sawdust and other rubbish along with the table to make up for its shortness. (Rather than print the Ergänzungen to make the book fatter leave a dozen white sheets for the reader to swear into when he has purchased the book and cannot understand it)" (Wittgenstein 1973, p. 46).

The last "sentences drawn from other manuscripts", evidently not equally ”good", are so omitted and the process of insertion, announced in 1915 note, finishes therefore in definitive way.