“In this work more than in any other it is worth looking at apparently solved questions again and again from new sides as unsolved,“ Ludwig Wittgenstein jotted in his philosophical notebook in November of 1914. “Don’t get stuck with what you once wrote. Think always of a fresh beginning, as if nothing had as yet happened.” (p. 30)  The First World War had been raging for months; Wittgenstein was serving as an outlook on an Austrian gunboat; but he remained determined to continue the philosophical work he had been doing before the war with Bertrand Russell at Cambridge. “Logic must take care of itself,” had been the opening entry in his new notebook on August 22. He called it “a singularly profound and significant insight.” (p. 2) The sentence was intended to say, first of all, that logic is self-contained, that it does not rest on anything outside it. But by putting it at the head of his notebook Wittgenstein may also have been expressing the hope that his work in logic would not be affected by the vagaries of the war. “Will I be able to work now?” he had asked himself anxiously on the first page of the private diary he attached to his philosophical notebook. It turned out that he could do so even under heavy bombardment. “Canons shook the boat as they fired near us at night. Worked much and with success,” he wrote on December 6. But progress was often slow and he feared that “the redeeming word has not been spoken.” As long as that was the case, he could only go over the same ground again and again. His most vexing problem at the time was that of “the logical form of the proposition,” a topic he had been exploring with Russell in the preceding years. But his view on the topic was still far from settled. “Does the subject-predicate form exist,” he now asked himself. “Does the relational form exist? Do any of the forms exist at all that Russell and I were always talking about?” (pp. 2-3) And so it went with questions but no definitive answers.
Some fifteen years later, Friedrich Waismann was trying to pin down Wittgenstein’s thinking for an expository book he was hoping to write. Wittgenstein had by then achieved some fame with his Tractatus Logico-Philosophicus, composed as the war was ending and published in 1921 with Russell’s effusive introduction. Back in Austria, the scholars in the Vienna Circle treated the work as a revelation. But their attempts to get its author to explain the book to them met with little success; Waismann’s discussions with Wittgenstein proved equally frustrating. “He has the wonderful gift of always seeing things as if for the first time,” Waismann noted. “He always follows the inspiration of the moment and tears down what he has previously sketched out.” After ten years of almost complete philosophical silence, Wittgenstein’s thinking had just entered a newly volatile phase. In 1930 he returned to Cambridge to sort out what he now thought to be missing in his earlier work. He began to lecture on the themes of the Tractatus, but the book struck him now increasingly as a piece of unbearable dogmatism. That conclusion sparked new investigations which led him on an exhausting journey, ”criss-cross over a wide field of thought,” in which “the same points were always being approached afresh from different directions and new sketches made,” as he wrote later in the preface to his Philosophical Investigations. Many of these sketches, he added, “were badly drawn or lacking in character, marked by all the defects of a weak draughtsman. And when they were rejected, a number of half-way decent ones were left, which then had to be arranged.” This is how the new book was composed. It was a collection of remarks, no more than “an album.”
Wittgenstein never quite finished his Philosophical Investigations. Its initial parts were firm enough in his mind, but he was unsure about how to complete the book. In 1948 he began to look at the issues once again in new ways. At the end, he confided to yet another notebook: “I do philosophy now like an old woman who is always mislaying something and having to look for it again: now her spectacles, now her keys.”
There emerges from all this the picture of a restless thinker set on constantly revisiting and reworking what he has previously thought. Waismann who had initially heard of Wittgenstein as the author of the Tractatus had expected a very different person. The book consisted of apodictic statements put forth with a minimal amount of argument and formulated in an often hieratic style. Its aura of absolute certainty was reenforced, moreover, by the elaborate numbering of its propositions that Wittgenstein had adopted from Principia Mathematica, Russell and Whitehead’s majestic logical treatise. The Tractatus presented itself thus to the unwary reader as a work of a stern, logical order and the author as someone endowed with unconditional truths.
That was, however, a piece of fiction. The book was, in fact, largely composed by extracting diverse propositions from earlier writings and each one of them had originally been surrounded by doubts and questions. Some of those propositions came from notes Wittgenstein had compiled for Russell in 1913; many others were taken from the philosophical notebooks he kept during the First World War. The Tractatus was, in this respect, just like Wittgenstein’s Philosophical Investigations, more of “an album” of sketches made on an extended intellectual journey than a tightly constructed treatise. We can see this clearly when we turn from the book to the source of its propositions in the war-time notebooks. Though not all of them have survived, the three notebooks that have permit us to reconstruct much of Wittgenstein’s course of thinking from the beginning of the First World War to January 1917. What we find in them is the record of an intricate back and forth in his thinking on a wide range of philosophical topics. They document continuous small shifts as well as some major turns in Wittgenstein’s thinking and in the end the emergence of a radically new way of conceiving philosophy. The Tractatus was a record of this entire development. Taking note of the long, treacherous journey that led to its composition is, in fact, indispensable for understanding its highly condensed formulations and for recognizing their varied and unstable subsoil.
The journey recorded in Wittgenstein’s war-time notebooks begins with questions provoked by Russell’s logical atomism. Russell had come to that doctrine in the late 1890s as a result of his break with F. H. Bradley’s monistic idealism. His friend G. E. Moore had led the way in an 1899 essay on “The Nature of Judgment.” With the help of a somewhat rudimentary logic Moore had sought to establish that reality was not a single thing – Bradley’s “One” – but consisted of a multitude of concepts. This so-called turn to realism was thus, in effect, one to a metaphysical pluralism. Russell readily adopted Moore’s pluralistic view and found support for it in the philosophy of Leibniz. His view was, as he later put it, “that you can get down in theory, if not in practice, to ultimate simples out of which the world is built, and that those simples have a kind of reality not belonging to anything else.” But he remained quite uncertain about the nature of those simples. Were they concepts and judgments, as Moore had maintained? Were they particulars or universals or, perhaps, both? Could they be spatial points or even sense data? When Wittgenstein arrived in Cambridge in 1911, he found Russell at work on a theory of knowledge that was to settle the issue. Russell’s tool for achieving that end was the logic he had devised in the previous years and, in particular, his theory of descriptions which seemed to allow one to distinguish between merely apparent objects that could be analyzed away and objects resistant to further analysis.
While Wittgenstein’s war-time notebooks show him deeply immersed, from the first page, in the problems of logical atomism it wasn’t, however, this doctrine that had initially brought him to work with Russell. When Wittgenstein got to Cambridge in 1911, he was mostly familiar with Russell’s 1903 book The Principles of Mathematics. As a result he had visited Gottlob Frege in Jena, whose logical theories Russell had discussed in a long appendix, and Frege had advised him to go to Cambridge to study with Russell; he had also become fascinated with Russell’s discussion of the logical paradoxes in a second appendix to his book and had written short piece on that topic. He was thus initially primed to focus on Russell’s logic and convinced that he could make a contribution to its further development. “Logic is still in the melting pot,” he wrote brazenly to his mentor in 1912 within a year of having embarked on that project. He was sure that it “must turn out to be a totally different kind than any other science.” Russell, who had spent years of excruciating work on his logic, seems to have taken the remark in good spirits. But, as far as Wittgenstein was concerned, that logic still lacked sufficient unity and simplicity. In logic, he wrote later on in the Tractatus, simplicity was a sign of correctness. In June 1913, he informed Russell accordingly of some ideas he had in this respect: “One of the consequences of my ideas will – I think – be that the whole of logic follows from one proposition only.” His efforts to improve on Russell’s logic took form in two early writings: a set of dense notes written for Russell in 1913 and another set of equally dense remarks dictated to G. E. Moore a year later. Fundamental in them was the distinction between propositions and proper names. “Frege said ‘propositions are names’, Russell said ‘ proposiitons correspond to complexes’. Both are false; and especially false is the statement ‘propositions are names of complexes’,” he wrote in the first set of notes. (p.97) The distinctive characteristic of propositions was that they had two poles of in that they could be either true or false. This bi-polarity of the proposition he expected to be one of the unifying principles of his logic. It justified truth-functional logic, it suggested a diagrammatic representation of their relations, and eventually their depiction in a system of truth-tables. But that still left the logic of generalized propositions unexplained and Wittgenstein found himself wrestling with that topc. In addition there was also still the need to solve the logical paradoxes with helps of a theory of types whose justification was, however, again a problem. Wittgenstein’s notes for Russell and Moore were suggestive, but they did not make a conclusive case for a newly unified and simplified science of logic..
Wittgenstein’s preoccupation with these technical problems, however, made Russell anxious that his student would remain a narrow specialist, lacking an understanding of the broader and deeper philosophical issues with which Russell saw himself concerned. That was, however, a misjudgment. At the end of his notes for Russell, Wittgenstein had sought to to soothe Russell’s anxieties by announcing that “philosophy consists of logic and metaphysics. Logic is its base.” (p. 106) So, metaphysics was after all the goal of Wittgenstein’s project. But he added at this point three other propositions that must have given Russell some pause. “In philosophy there are no deductions; its is purely descriptive. Philosophy gives no picture of reality. Philosophy can neither confirm nor confute scientific investigation.” (p. 106) These posed, in effect, a direct challenge to Russell whose logical atomism was meant precisely to give a deductive and scientific picture of reality. It is far from clear how Wittgenstein himself thought of ho those claims were to be reconciled with his belief that philosophy consisted of logic and metaphysics. He held consistently that philosophy was not a science and in this respect he was certainly at odds with Russell and that proposition was not necessarily incompatible with saying philosophy was made up of logic and metaphysics. But the two other propositions were more of a challenge. How could philosophy be purely descriptive and how could it not give a picture of reality while being at the same time a metaphysics? As it turned out, Wittgenstein was not to make anything more of these pronouncements till later. It was only in the 1930’s when the Tractatus system had broken down that he came back to the idea of philosophy as a purely descriptive undertaking. And that philosopher might not be able to provide a picture of reality became clear to him only in the course of his war-time reflections. It entered from there into the Tractatus and became one of Wittgenstein’s leading ideas in the 1930s.
Two things are, in any case, clear from Wittgenstein’s notes for Russell. The first is that already in 1913 Wittgenstein subscribed to a broader philosophical agenda than Russell realized and the second is that this agenda remained undeveloped until he turned to Russell’s logical atomism. That his wartime notebooks begin with than examination of key issues of logical atomism signals, thus, one of those transitions in Wittgenstein’s thinking that were to become definitive of his philosophical life. But it is not difficult to understand why he embarked at this point on the exploration of logical atomism. Atomism has proved an attractive position in the history of philosophy. Democritus and Leucippus were inspired by it and they invented the word “atom for the assumed simple elements. Plato sketched a version of it in the Theaetetus. Leibniz became its most prominent exponent in modern philosophy and even today it lives a ghostly life in what we call model theory. Wittgenstein himself had, however, grown up with other, more holistic forms of thinking and so Russell’s atomism must have come to him as an appealing and liberating alternative.
His war-time notebooks indicate that he initially identified with much of the program of Russell’s atomism though by no means to all of its details. He never bought into his mentor’s entire philosophy and certainly not into his social and political opinions. His philosophical discussions with Russell were often stormy. In the notes he wrote for his mentor in 1913, he highlighted various disagreements and laid out how he meant to go beyond Russell’s work. The war-time notebooks continue that independent line of thinking. From the start, Wittgenstein objects to Russell’s appeal to self-evidence in order to settle questions about logical structure and the nature of the simple objects. “Russell would say ‘Yes! That’s self-evident,’” he wries, but Wittgenstein considers this ridiculous. (p. 3) Russell had, in fact, devoted a whole chapter of his Theory of Knowledge to defend his reliance on self-evidence, but for Wittgenstein the idea of self-evidence “is and always was whole deceptive.” (p. 4) Such disagreements did not mean that Wittgenstein rejected logical atomism altogether. But the doctrine needed sifting and clarification. In contrast to Russell – and, in fact, in contrast to all the earlier atomists – Wittgenstein recognized the profound challenges the doctrine raises. The most difficult one was, perhaps, how the doctrine could be justified. It could, obviously, not be deduced from observation since that would never deliver “ultimate constituents.” Some kind of analysis was called for and that analysis had to use some kind of logic. That raised, however, immediately two set of questions. The first concerned the logic to be used in the process of analysis, and the other the outcome of the analysis, the determination of the sought-after simples. Wittgenstein’s notebooks show him to be working tenaciously and for months on these interconnected issues.
As for the logic, Wittgenstein asked himself at the start of his notebooks what logical structures the process of analysis could discover. Hence, his preoccupation with the question of the logical form of propositions. How could one be sure that one had given the right analysis? And how could one determine that the analysis was complete? Wittgenstein struggled hard over these questions. There were seemingly insurmountable difficulties with negative propositions and likewise with the generalized ones. The notebooks contain, in consequence, repeated reminders such as: “In all these considerations I am somewhere making some sort of FUNDAMENTAL MISTAKE.” (p. 10) Russell had naively assumed that one could read the structure of reality off from the logical form of our propositions. But he had no satisfactory account of how this was to be done. In his Theory of Knowledge he had said only that the proposition “A is similar to B” is true “when there is a complex composed of A and B and similarity.” Wittgenstein had at first agreed with this, writing in his 1913 notes: “The form of a proposition has meaning in the following way… I say that if an x stands in the relation R to a y the sign ‘xRy’ is to be called true to the fact and otherwise false. This is a definition of sense.” (p. 91) But that was certainly no definition. Both Russell and Wittgenstein were considering only particular relational propositions. They soon moved on, however, to a more considered view, if we can trust the lectures on the philosophy of logical atomism Russell delivered in 1918. According to Russell, those lectures reflected the state of their joint thinking at the moment the war broke out and Wittgenstein was forced to return to Austria. Their view was, in Russell’s words, that “in a logically correct symbolism there will always be a certain fundamental identity of structure between a fact and the symbol for it …and the complexity of the symbol corresponds very closely with the complexity of the facts symbolized by it .” But this was not yet enough in Wittgenstein’s eyes. Early on in his war-time notebook he still expresses uncertainty over “the logical identity of sign and signified.” (p. 3) We must, perhaps, think of a proposition as a model in which “a world is as it were put together experimentally.” Or, better still, we must think of it as a picture. (p. 7) But was that satisfactory? “On the one hand my theory of logical depiction seems to be the only possible one, on the other there seems to be an insoluble contradiction in it.” (p. 17) It is not obvious what contradiction Wittgenstein has in mind. It seems to have involved generalized propositions. His dissatisfaction with the “theory of logical depiction” is, any case, evident. One of its problems is that there are different ways or methods of representation. And: “The method of depiction must be completely determinate before we can at all compare reality with the proposition to see whether it is true or false.” (p. 25) That answer created, however, problems of its own. Would one not have to be able to identify both the structure of the proposition and that of the fact in order to see whether they were one and the same? The logical atomists assumed, however, that one could read the structure of reality off from that of the fully analyzed proposition. That was the whole strategy of their position. But what would guarantee that the structure of the proposition was a clue to that of the facts? .
Considerations of this kind led Wittgenstein, in turn, to ones concerning the supposed elements of reality. Like Russell, he kept going back and forth over the possibilities. Were they sense-date? Could they be universals? Were they spatial points?And so on. In the end he came to ask himself whether logic could even tell us in principle what they were. Perhaps it could assure us only that there were such but without identifying the atoms themselves. He was certainly ready to assume that “the world has a fixed structure.” (p. 62) This was equivalent, he thought, to saying that our words must have definite sense since “the demand for simple things is the demand for definiteness of sense.” (p. 60) One could then conclude that there were ultimate constituents of reality, “simple objects,” without having to specify their exact nature. “It seems that the idea of the simple is already to be found contained in that of the complex and in the idea of analysis, and in such a way that we come to the idea quite apart from any examples of simple objects, or of propositions which mention them, and we realize the existence of the simple object – a priori – as a logical necessity.” (p.62) But was it obvious that the analysis would lead to simple, unanalyzable elements? Was that already contained in the concept of analysis? Or could the analysis go on ad infinitum? Perhaps, one could even speak of simple constituents. Wittgenstein was willing to put particular weight on the transcendental argument that Leibniz had used in his Monadology to argue for the existence of simples. There had to be simples because there were complexes. Had Wittgenstein been inspired by Russell’s book to adopt this Leibnizian line of reasoning? Or had he read Leibniz on his own? There are certainly surprising similarities between Wittgenstein’s Tractatus and the Monadology even though the simples they speak of are quite different. What unites the two writings is their logical starting-point, their contraposition of will and representation, the denial of causal relations, and most of all the course of their reasoning from metaphysical foundations to the questions of ethics, not to speak of the numbering of their propositions. It is useful to remind ourselves here that Wittgenstein drew not only on Frege and Russell for philosophical inspiration. From the private war-time diaries we learn that he was also reading Tolstoy, Emerson, and Nietzsche at the same time as he was laboring over Frege’s and Russell’s ideas. Traces of all these influences are, in fact, noticeable at some point in his notebooks.
His attachment to Russell was to weaken as the war went on. Nine months into the conflict, Wittgenstein’s notebook records a first rupture with Russell’s way of thinking and an indication that a major transition in Wittgenstein’s thinking is on the way. The passage in question occurs in the middle of extensive reflections on the simple constituents of reality. On May 23, 1915, Wittgenstein writes in a tone quite alien to anything he had written before in his notebook: “The limits of my language mean the limits of my world. There really is only one world soul, which I for preference call my soul and as which alone I conceive what I call the souls of others.” This, he adds, provides the key for deciding “in how far solipsism is a truth.” (p. 49) In the same tone he continues two days later: “The urge towards the mystical comes from the non-satisfaction of our wishes by science. We feel that even if all possible scientific questions are answered our problem is still not touched at all.” (p. 51) None of those words had been prepared by anything earlier in his notebooks. The limits of language and the world, the stress on “my language,” the conception of a world soul, solipsism, the limits of science, the non-scientific character of “our problems,” the mystical — none of those themes had been addressed in the earlier pages. All of them were, moreover, entirely foreign to Russell and his mode of thinking. In his Introduction to the Tractatus, Russell would later dismiss such speculative thoughts with the words: “Mr. Wittgenstein manages to say a good deal about what cannot be said.” His caustic remark provoked Wittgenstein, in turn, to accuse Russell of not having understood his book. May 23, 1915 marks thus a first major break-with Russell and his way of doing philosophy even though its full implications may not have been obvious to Wittgenstein at the time since he turns immediately back from his speculative remarks to further reflections on the nature of the simple objects – almost, as if nothing had happened. But those few sentences mark, nonetheless, a first step away from Russell and his program and, as it turns out, also from Russell’s understanding of philosophy itself.
From where, then, did these strange new thoughts come? What motivated their intrusion into Wittgenstein’s apparently straightforward attempt to sort out the problems of logical atomism? We don’t know exactly what occasioned them. Was there an external motivation for them? Wittgenstein’s first sentence on May 23 may, in fact, be read as continuing his preceding discussion of simple objects but In a new register suggesting that we have no language-independent means of establishing what is simple. Some such thought had, indeed, already occurred to him two weeks earlier: “The simple thing for us is: the simplest thing that we know – the simplest thing to which our analysis can advance – it need appear only as a prototype, as a variable in our propositions.” (p. 47) Even so, the new emphasis on language being “my language” introduces an element of the subjective into Wittgenstein’s thinking that had not been visibly there before and that has no equivalent in Russell’s philosophy. Wittgenstein’s turn against Russell becomes even more evident in the next two sentences with their talk of solipsism and the world soul, themes onc again alien to Russell. And, if this is not enough, his subsequent appeal to the mystical and his insistence that there are “personal” problems that cannot be resolved scientifically are clearly meant to draw a line between his own view and Russell’s scientistic outlook.
Though none of these reservations had made an appearance before in his notebook, Wittgenstein was also not making them up on the spot. He was not just relying here on momentary inspiration. His turn against Russell signaled, rather, a return to another, earlier set of ideas that had become submerged when he went to Cambridge and had aligned himself there with Russell’s program. Those earlier ideas had come to Wittgenstein largely from Arthur Schopenhauer’s book The World as Will and Representation. Georg Henrik von Wright recalled later that Wittgenstein had told him once that “his first philosophy was a Schopenhauerian epistemological idealism.” This is not implausible since Schopenhauer’s book had been immensely a popular in Fin de Siècle Vienna. Its pronounced pessimism and implied skepticism fitted the mood of the Austro-Hungarian empire in decline. From Ludwig Boltzmann to Sigmund Freud and from Fritz Mauthner to Otto Weininger, the Viennese authors with whom Wittgenstein was familiar, had all been affected by Schopenhauer’s philosophy. But we have to assume that by the time he arrived in Cambridge Wittgenstein had abandoned (or was about to abandon) this early excursion into philosophy. He may have been drawn to Russell’s atomism, in fact, precisely because it provided an antidote to the overwhelming cultural influence of Schopenhauer’s view of the world as unindividuated metaphysical will. But it would be mistaken to think that the association with Russell completely extinguished Wittgenstein’s interest in Schopenhauer. The later parts of his war-time notebooks and the Tractatus itself show how much Schopenhauer was still very much on his mind. He appears to have remained particularly attracted Schopenhauer’s heterodox views on ethics which rejected moral rules and, in particular Kant’s categorical imperative, and described ethics as a way of seeing the world rather than of acting in it. Wittgenstein’s notebooks and the Tractatus also draw on Schopenhauer’s reflections on the will and, not least, on his philosophy of art. Most striking is perhaps that the beginning and end of the Tractatus mimic the first and last sentence of Schopenhauer’s book.
Just as significant as the presence of Schopenhauer in the notebook passage from May 23 is that of Fritz Mauthner, the author of three expansive volumes of Contributions to a Critique of Language. The work argued for “a critique of language which would be a meta-critique of reason” in a phrase Mauthner had borrowed from Fritz Jacobi, the friend of Immanuel Kant, who had taken the idea in turn from the dark musings of Johann Georg Hamann. Wittgenstein had certainly read the first volume of Mauthner’ book by the beginning of the war and would later familiarize himself with Hamann’s writings. He was interested, thus, in ways of thinking about language that were distinct from what he had learned from Frege and Russell. It was Mauthner who impressed on him the idea that what we abstractly call “language” is in reality always “my language.” It was Mauthner also who had discussed solipsism and the world soul in his magnum opus.. Wittgenstein’s notebook entry of May 23 marks thus a tur and return not only to Schopenhauer but also to Mauthner. And this return to Mauthner was potentially most devastating for his attachment to Russell’s atomism For it would lead Wittgenstein in the later parts of his notebooks to adopt Mauthner’s skepticism about all philosophical theorizing. It was this “pyrrhonian” Mauthner who supplied him with his concluding metaphor in the Tractatus of the ladder one must throw away after one has climbed up on it. Even so, Wittgenstein’s adoption of Mauthner’s views had been selective. In the Tractatus he was still sufficiently in the thrall of Russell’s logic to conclude that critique of language could not be conducted “in Mauthner’s sense” by remaining within the confines of ordinary language. At that point Wittgenstein had still faith in Russell’s distinction between the apparent and the real logical form of the proposition. It was only after he had abandoned the Tractatus that he discovered the whole force of Mauthner’s meta-critique of reason by means of a critique of language.
Perplexing as Wittgenstein turn and return to Schopenhauer and Mauthner on May 23 proves to be, just as perplexing is the abrupt way the excursion ends. Within two days, we find him back at the pains-taking task of trying to determine the identity of the simples of atomist lore. It was as if nothing had happened. But how solid was his apparent recommitment to the atomist program? Where was he moving to at this point? We have unfortunately no record of the course of Wittgenstein’s thinking between June 2015 and April 2016 since the notes covering that period are now lost. But it is clear from the following notebook that a year later he has not yet given up on logical atomism. That notebook begins with further reflections on simple objects and the form of the proposition. But his words echo now what he had said the previous year about the subjective character of our thinking about them. On April 15 he writes: “We can foresee only what we ourselves construct. But then where is the concept of the simple object still to be found?” (p. 71) Since simple objects are not immediately accessible to us, our claim that there are such comes to being a prediction that logical analysis will ultimately deliver them. But if we can foresee only what we ourselves construct then our account of the simple objects will depend on our construction. He goes on to speak similarly of the construction of simple functions and since such functions determine the form of the propositions in which they occur, it now appears that the forms of propositions we can identify must also be relative to what we are able to construct. The realistic picture of logical atomism seems to have given way here to a constructivist one. But the issue is not yet completely settled. A year earlier he had already entertained the possibility that one might need to distinguish between what is simple for us and what is simple absolutely speaking. Could it not even be the case that when we speak of an ordinary object, such as the watch on the table, we treat it, in effect, as logically simple? Even so, there may still be absolute simples in addition. What if our construction is somehow determined by what is out there? In that case, it may still be possible to give a definite and “objective” answer to the question of the simple objects and the logical forms of the proposition. Towards the end of the notebook, on Nov. 11, Wittgenstein gets back to this point once more when he writes that it must be possible “to set up the general form of the proposition, because the possible forms of propositions must be a priori.” (p. 89)
All the same, his notes indicate also a growing distance to Russell’s version of logical atomism. That becomes dramatically apparent in the spring of 1916. The episode begins on May 6 with Wittgenstein addressing once more the theme of the limits of science. “The whole world view of the moderns is grounded in this illusion that the so-called laws of nature are the explanations of the natural phenomena. We thus stop with the “laws of nature” as something inviolable just as the ancients did with God and fate. And both are right and wrong. The old ones are, however, clearer in that they acknowledge a clear endpoint, whereas it appears in the new system, as if everything was grounded.” (p. 72) This renewed critique of scientism expresses at once reservations about Russell’s world view and fresh attention to Schopenhauer and Mauthner who both question the explanatory power of science. There follows a short passage on logical operations followed by a hiatus of three weeks in which Wittgenstein remains silent. When he resumes his notebook entries on June 11 we are faced with the unexpected question: “What do I know about God and the purpose of life?” (p. 72) The two topics prove to be intimately linked for him. “To believe in God means that life has a sense,” he writes shortly later. (p. 74) We can, in fact, call the meaning of life by the name of God. We can also identify God with fate and even with the world. Prayer he says is “thinking of the meaning of life.” The question about God and the purpose of life initiates a whole series of further thoughts on the world and our place in it, on good and evil, and the nature of the will. As he pursues these themes still other topics make their appearance: the human self, idealism and solipsism, ethics and art, happiness, death and suicide. The tightly argued examination of logical atomism in the earlier pages of his notebooks has exploded into a wild array of bewildering new thoughts. “Yes,” he writes, “my work has expanded from the foundations of logic to the essence of the world.” (p. 79)
In contrast to the previous moment in May 1915 when he had turned away from the problems of logical atomism only to return to them shortly afterwards, these new reflections continue for several months interspersed with only occasional glances back at his previous concerns with the atomist program. Finally, on November 21, the current of speculative thought appears to be running dry and we find Wittgenstein fretting once more over the question of the general form of the proposition. While we can’t say what had occasioned the disruption of his train of thought in May 1915, we do know what triggered this one. On May 6, five days before Wittgenstein asks his disturbing question about God and the purpose of life, his private diary records that he feels in imminent danger of losing his life in the war. How can one find inner peace in this situation? “Only by living in a way that pleases God. Only in this way is it possible to bear life.” He adds on May 10: “I am doing well now due to the grace of God. … He will not abandon me in this danger.” On May 16: “I am sleeping today under infantry fire and will likely perish. God be with me. I surrender my soul to the Lord.”  Worse is still to come. In June, the Russians launch a major attack on the Austrian forces in the so-called “Brusilov offensive.” Thousands of Austrian soldiers lose their lives, many others become Russian prisoners of war. Wittgenstein’s own unit is directly engaged in this deadly battle which lasts till the middle of August. He is certain that he will not survive. On July 29 he writes: “I was shot at yesterday. Was scared. I was afraid of death. I have such a wish now to go on living. And it is difficult to renounce life…”
His thoughts in this period are often feverish and obscure – as he realizes. “Here I am still making crude mistakes. No doubt about that.” he notes on July 29. (p. 78) “I am conscious of the complete unclarity of all these sentences.” (p. 79) They are driven by existential anxieties rather than logical ratiocination. Only some of them will make it into the Tractatus. His reflections on God do not, but those on the meaning of life will become essential for the coda of his book. The world” will also remain a topic of concern in it and that in two ways. The world is all that is the case, as the first sentence of the book will say, and as such a matter of logic.. But ethics also demands that we see the world in the right way, as the end of the book will declare. “Ethics does not treat of the world,” he writes in his notebook. It is, rather, “the condition of the world, just like logic.” (p. 77) We don’t have to look far to find the inspiration for this dual picture. We can find it in Schopenhauer World as Will and Representation.
But what is the right way to look at the world? The question leads Wittgenstein to pose two others. What is the place of the subject, the I, the self in relation to the world? And where do we find good and evil? As to the first question, he says that what we call “the world” is really always only “my world.” Or as he puts it provocatively: “The world and life are one.” (p. 77) Death is therefore not an event or fact in the world. In death the world ceases to be. But what makes the world my world? What is the subject? Russell had talked of knowledge as based on a fundamental relation of acquaintance and of acquaintance as “a dual relation between a subject and an object.” But this subject, he had insisted, was not to be identified with an I or self. There was no need to assume a persistent self. Wittgenstein had never felt much attraction to Russell’s doctrine of acquaintance. His work was not meant to be epistemological in character. Epistemology is the philosophy of psychology,” he had written to Russell in 1913, and as such of minor interest. (p. 106) That was also something he also meant to convey in his programmatic statement that logic must take care of itself. In his notebook he rejects Russell’s conception of acquaintance out of hand. “The representing subject is surely mere illusion,” he writes. But: there is still the I or self to be considered as distinct from the Russellian subject. “The I, the I is what is deeply mysterious.” (p. 80) The I is mysterious because we never confront it as an object. I am in the world, he writes, in the way my eye is in my visual field. The perceiving eye is, in fact, not part of the visual field. It is what makes the visual field possible. In the same way the I is not part of the world but makes my world possible. It makes its appearance “through the world’s being my world.” (p. 80) This I is, however, not the individual self, but the “one world soul” of which he had said a year earlier that it manifests itself as that “which I for preference call my soul.” (p. 49) This subject or self it to be conceived will as he writes, once again in line with Schopenhauer. While the representing subject is a mere illusion, “the willing subject exists.” (p. 80) But the will must not be conceived as a causal power by which the individual can affect events in the world. For “the world is independent of my will.” (p. 73) Standing apart from the world, the will can only define an attitude to the world as a whole. It can lead us to reject or accept the world for what it is.
With this we have entered the domain of value or ethics and aesthetics. To accept the world for what it is means to be happy and happiness is what we should strive to attain. The happy life, is one lived in accord with the world. It is in this state that we grasp the meaning of life. “Is this not the reason why human beings to whom after many doubts the meaning of life has become clear, cannot then say in what that meaning consists.” (p. 74) The problems of life find their solution in their disappearance. Ethics is thus not concerned with advancing a theory; it does not assert propositions or promulgate rules. It is, in fact, clear that ”ethics cannot be expressed in words.” (p. 78) . Ethics, he says, is thus “transcendent” – or, rather, “transcendental,” as he will phrase it in the Tractatus. It is a way of seeing the world from outside as a whole sub specie aeternitatis. And this is, at the same time, the aesthetic way of seeing things. “The work of art is the object seen sub specie aeternitatis, and the good life is the world seen sub specie aeternitatis. This is the connection between art and ethics.” (p.83) Ethics is the aesthetics of the world.
The claim that ethics cannot put into words puts Wittgenstein on the path to an even more radical conclusion. He had previously talked of logical limits to what can be said. What do I know when I understand a proposition but do not know whether it is true or false, he had asked himself in 1914. And he had replied to that question: “At this point I am again trying to express something that cannot be expressed. (p. 31) That had been, in fact, yet another thought embedded in the first, introductory sentence of his war-time notebook. That logic must take care of itself meant among other things that “all we have to do [and, in fact, all we can do] is to look and see how it does it.” (p. 11) But Wittgenstein’s conclusion that ethics, too, cannot be put into words, adds a new dimension to those earlier considerations. Since logic and ethics form the arc on which Wittgenstein’s philosophical thinking in this period moves he is made to conclude: “The correct method would be to say nothing, except what can be said, i.e. what belongs to natural science, i.e. something that has nothing to do with philosophy.” (p. 94) And with this conclusion, the break with Russell has become complete. The entire project of logical atomism has been consigned now to what cannot be said and is therefore without sense. Wittgenstein’s thinking has broken through all the earlier layers in which it had been confined and confronts philosophy now in a new, profoundly different way. It will take him the rest of his life, though, to discover the consequences of this revelation.
But his thoughts in the summer and fall of 1916 were not yet the end of the road he had been traveling since the beginning of the war. The year before he had written to Russell: “I have recently done much work and, I believe, with good result. I am now busy bringing the whole thing together and writing it down in the form of a treatise.” That treatise was still to be written a year later. Wittgenstein did not get around to that concluding task till the middle of 1918. The war-time notebooks in our possession end unfortunately in January of the preceding year. It is thus impossible for us to track the course of Wittgenstein’s thinking to the moment when he set to work on the text of the Tractatus. The assumption is that he completed at least one more notebook and possibly two before that he began work on his book. It is plausible to think that whatever he wrote in that period was more or less like the earlier notebooks, that he continued to record individual thoughts on variety topics and put them down as they struck him. Those lost notes may have covered much of the same ground as the notebooks we have and displayed the same kind of back and forth in his thinking. There was, no doubt, also additional material in them. Some passages in the Tractatus have no precedents in the existing notebooks. One striking example is the discussion of mathematics in the Tractatus. Was that part of the book based on notes he made between the end of 1916 and the middle of 1918? We can be fairly certain, however, that he did noy sit down in this period two write a deductive treatise. All he had, when he finally began to write the Tractatus was a set of scattered notes from which he quickly excerpted his text. That should, in fact, be evident from the look of the finished product.
When he started the composition of his book in the summer of 1918, Wittgenstein was on a temporary leave from the war front. Time was short and he was determined to bring the work to an end. There was no other option for him than to draw on what he had written earlier. He relied for this purpose on all the notes he had made in the previous five years and selected from them what he found suitable. That was no easy task, since his thinking had changed so much in those years and in so many different directions. His ambition was, surely, to present the material as a single coherent whole. In putting it together and selecting from what he had, he must have sought to clarify and unify what he wanted to say. But how far did he succeed in this? How far could he have succeeded, given that his thinking had moved across such a wide arc from the foundations of logic, through Russell’s atomism, to Schopenhauer’s ethics and Mauthner’s critique of language.
In putting his material together, Wittgenstein sought to impose on it a semblance of order by arranging his propositions according to the numbering system he had borrowed from Principia Mathematica. There were the major proposition numbered 1 through 7. Others were given decimal figures to indicate their place in the text and their relations to each other. Proposition 1.1 was meant to be subsidiary to proposition 1 and proposition 1.11 to 1.1, and so on. Some propositions ended up with lengthy decimals to indicate their minor place in the whole, as, e. g., proposition 5.47321. The scheme was ingenious and the scholars have been busy trying to decipher its significance. But how seriously can one take it? Not all of the seven major propositions appear to be of equal weight. Proposition 6 provides a notation for truth-functions which appears to be a greatly more specific subject-matter than the ones he addresses in the other major propositions. The principle that “logic must take care of itself” has been relegated to number 5.473 even though Wittgenstein had declared it to be a singularly profound insight in his initial notebook entry and the proposition retained a fundamental significance. Russell’s distinction between the apparent form of a proposition and its real logical form on which so much in Wittgenstein’s own account of atomism depends is mentioned only in 4.0031. The claim that logical constant do not represent anything, which Wittgenstein himself characterizes a Grundgedanke, is stated only in 4.0312. Caution is thus in place. One must not be seduced by Wittgenstein’s numbers.
While the formal presentation, the sustained voice of certainty, and the rhetorical pull of the propositions suggest a unitary system of ideas, it is far from clear that Wittgenstein has succeeded in producing any such thing. Looked at from the perspective of his war-time notebooks, the text suggests rather a course of thinking that leads from an apparently self-evident logical atomism to the concluding call to overcome its propositions in order to see the world in the right way. We should the perhaps read the Tractatus despite its name not as a treatise but as the record of a philosophical journey that Wittgenstein has undertaken in the previous years. Interpreters tend to look at philosophical texts in another way. They operate on a principle of charity which assumes that a philosophical text must be advancing a a coherent theory. In the face of obvious difficulties in applying that principle they often struggle in their attempts to find the expected theoretical lesson. The readers of Wittgenstein’s Tractatus have often done so by focusing only on part or another of the text. Some have taken it to be a straightforward contribution to the philosophy of logical atomism; others have considered it to be of piece of logic and a theory of meaning. For yet another group of interpreters the Tractatus is a lesson in philosophical skepticism. There are finally those for whom it is an essay on ethics. Each one of those interpretations has to ignore, reject, or downplay parts of the text. The Tractatus is, in fact, all and none of what the interpreters have made of it. It is an exercise in philosophical thinking that takes one over uncharted territories and produces a map that can be confusing. It exemplifies what philosophy is like when once one has crossed such territories and tried to decipher their maps. Wittgenstein’s hope in the Tractatus was that after this exercise one might be able to leave philosophy alone. Philosophy is “not a doctrine but a practice. A philosophical work consists essentially of elucidations.” At the time, Wittgenstein thought that one would have to give such elucidations only once and could thus be done with the work of philosophy. That was an error as he came ultimately to see.
When he spoke to members of the Vienna Circle in 1929 Wittgenstein appeared once again to be drawn to the question how to understand atomism. He still wanted to think of propositions as pictures of states of affairs, but he allowed now that they might turn out to be “incomplete pictures” that represented only some features of what they depicted. An identity of sign and signified was thus not required. The form of the elementary proposition could, moreover, not be determined once and for all by logic. It could be determined only in an empirical fashion. “It is simply ridiculous to believe that we can do here with the usual forms of ordinary language, with subject-predicate, with two-place relations, and so on.” Objects were simply “equivalent elements of the representation;” they might be represented, for instance, by the equations of physics.
But after his return to Cambridge, there was no more talk of simple objects. G. E. Moore attended his lectures in this period and kept copious notes of them but in none of them did Wittgenstein contemplate such simples. Instead, he said now that both he and Russell had been confused about what logical analysis could deliver, “thinking that further work at logic would show us the elements…. Russell had no right to say that the result of analysis would be 2-term, 3-term relations, etc. … I was wrong in supposing that it had any sense to talk of [the] result of [a] final analysis.” They had both been misled by relying on an idea of analysis derived from natural science. “To analyze water is to find out something new about water… In philosophy we know all we need to know at the start, we don’t need to know any new facts.”
That doesn’t mean that he had given up altogether on the pluralism that had been integral to the atomistic view-point. But his pluralistic outlook took now a different form. How many kinds of sentences are there, he asked in his Philosophical Investigations. “There are countless kinds, countless different kinds of use of all the things we call ‘sign’, ‘words’, sentences’.”  There were numerous kinds of language-games. There were multiple world views, he said in his late notes, each with its own internal logic. And there were even multiple kinds of mathematics. Russell’s atomism, it appears, had cured him once and for all of a hankering for an ultimate unity of things.
But his own reflections on logical atomism had also cured him of the wish to look for a single, unifying philosophical theory. Looking back at the Tractatus in 1932, he told his students: “Philosophical trouble is essentially this: You seem to see a system, yet [the] facts don’t seem to fit it; & you don’t know which to give up. Near it looks like a stump., then like a man. Then again…” His experience with the atomist program made him give up on the traditional understanding of philosophy as aiming at the construction of a theory. He now thought that this kind of theorizing was due to philosophers being blinded by the method of science which had tempted them “to ask and answer questions in the way science does. This tendency is the real source of metaphysics, and leads the philosopher into complete darkness.” That sort of philosophizing he now considered to be dead and what was left of it was dispersed in different places. His own work concerned only some of what remained. ”One might say that the subject we are dealing with is one of the heirs of the subject which used to be called ‘philosophy’.” In a daring metaphor he suggested that the only kind of philosophy that was now possible was like arranging books on a bookshelf. The goal was to achieve a transparent order but there was no ideal, absolute, final arrangement. He tried several formulas for characterizing the task of this new philosophizing. Philosophy, he said, was purely descriptive. Its goal was to describe the different ways we speak and think. It was phenomenological in trying to present a perspicuous overview of our grammar. It was therapeutic somewhat in the way Freud’s psychoanalysis was, relieving us of our perplexities. It was meant to liberate us from ways of thinking that entrapped our minds like flies in a fly-bottle. Philosophy was, in fact, a continuous battle against the bewitchment of our mind by language We were bound to find ourselves again and again not knowing our way about. There could be no end to these entanglements. We were bound to remain in transition.
 Page references are to Ludwig Wittgenstein, Notebooks 1914-1916, edited by G.H. von Wright and G.E.M Anscombe; 2nd edition, University of Chicago Press, Chicago, 1979. Translations modified (Text also referred to hereafter as NB)
 Ludwig Wittgenstein, Geheime Tagebücher 1914-1916, edited by Wilhelm Baum, Turia & Kant, Vienna 1991, p. 13. (Hereafter referred to as GT)
 GT, p. 49.
 GT, p. 45.
 Brian McGuinness, “Vorwort,” in Friedrich Waismann: Wittgenstein und der Wiener Kreis, edited by B. F. McGuinness, Basic Blackwell, Oxford 1967, p. 26.
 Ludwig Wittgenstein, Philosophical Investigations, translated by G.E.M. Anscombe, P.M. S. Hacker, and Joachim Schulte, Wiley-Blackwell, Chichester 2009, p. 3.
 Ludwig Wittgenstein, On Certainty, edited by G.E.M. Anscombe and G.H. von Wright, Harper & Row, New York 1972, p. 70.
 G. E. Moore, “The Nature of Judgment,” Mind, vol. 8, 1899, p. 182: “It seems necessary then to regard the world as formed of concepts.”
 Bertrand Russell, A Critical Exposition of the Philosophy of Leibniz, Cambridge U. P., Cambridge 1900
 Bertrand Russell, “The Philosophy of Logical Atomism,” in Russell, Logic and Knowledge. Essays 1901-1950, edited by Robert Charles Marsh, George Allen & Unwin, London 1956, p. 270.
 Letter to Russell of June 12, 1912, NB, p. 120.
 Letter to Russell of October 30 1913, NB, p. 123.
 Published as appendix I and II in NB.
 Bertrand Russell, Theory of Knowledge. The 1913 Manuscript, edited by Elizabeth Ramsden Eames, Routledge, New York and London 1992, p. 149.
 Russell, “The Philosophy of Logical Atomismn,” p. 275.
 Leibniz, Monadology 1-3.
 Bertrand Russell, „Introduction,“ in Ludwig Wittgenstein, Tractatus Logico-Philosophicus, Routledge & Kegan Paul, London 1922, p. 22
 Norman Malcolm, Ludwig Wittgenstein. A Memoir with a biographical sketch by Georg Henrik von Wright, Oxford U. P., London 1962, p. 5.
 Alexander Stern, The Fall of Language: Benjamin and Wittgenstein on Meaning, Harvard U. P. Cambridge Mass. 2019.
 Ludwig Wittgenstein, Tractatus Logico-Philosophicus, 40031.
 GT, p. 70.
 GT, p. 71.
 GT, p. 74.
 Russell, Theory of Knowledge, p. 5.
 Georg Henrik von Wright, “The Origin of the Tractatus,” in v. Wright, Wittgenstein, Blackwell, Oxford 1982, p. 70
 Tractatus, 4.112.
 Wittgenstein und der Wiener Kreis, p. 42.
 Ludwig Wittgenstein, Lectures, Cambridge 1930-1933. From the Notes of G. E. Moore, edited by David G. Stern, Brian Rogers, and Gabriel Citron, Cambridge U. P., Cambridge 2016, p. 253
 Lectures, Cambridge 1930-1933, p. 88.
 Philosophical Investigations, 23
 Ludwig Wittgenstein, The Blue and Brown Books, Harper & Row, New York 1960, p. 18.
 The Blue and Brown Books, p. 28.