Jean-Paul RIOPELLE, L'hommage à Rosa Luxemburg, 1992 (détail)[ * ]

From Plato to Socrates: Wittgenstein's
Journey on Collingwood's Map [1]

Evan Cameron


How can I learn and help others to learn to mean more precisely by saying, doing and making things? By attending, I suggest, to how Ludwig Wittgenstein and Robin Collingwood answered that question during the first half of the twentieth century. In this paper I show how the last of three answers given by Wittgenstein, and the journey he undertook to arrive at it, exemplify the kind of answer that Collingwood had been advocating and exemplifying. I conclude, however, by suggesting as well that a fourth answer upon which they converged unwittingly points even further along the road to philosophical understanding than either of them was able to go, namely that if we are to answer such questions exactly, we must approach them not only historically, but biographically

Consider for example the proceedings that we call "games". I mean board-games, card-games, ball-games, Olympic games, and so on. What is common to them all? - Don’t say: "There must be something common, or they would not be called 'games'" - but look and see whether there is anything common to all.  For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that. To repeat: don't think, but look! … And this is just how one might explain to someone what a game is. One gives examples and intends them to be taken in a particular way. - I do not, however, mean by this that he is supposed to see in those examples that common thing which I - for some reason - was unable to express; but that he is now to employ those examples in a particular way. Here giving examples is not an indirect means of explaining - in default of something better. For any general definition can be misunderstood. The point is that this is how we play the game. (I mean the language-game with the word "game".)  Ludwig Wittgenstein (1953) [2]

"The whole value of an example is lost unless it is historical."
Robin Collingwood (1916) [3]

How can I learn and help others to do so? How, in particular, can I learn and help others to learn to say, do and make things that will serve better as means toward ends of a kind that I and they would like to realise?

How, that is, can I learn and help others to learn to mean more precisely by saying, doing and making things? [4]

Long before, and ever after, Plato placed such questions at the centre of philosophical enquiry, philosophers have been provoked by the answers that they and others have given to them, and understandably so; for the constraints upon how we learn determine, when institutionalized, the very horizons of the cultures within which we live, move and have our being. Only the wisest among them, however, have been sufficiently perplexed to subject the answers that they were themselves exemplifying to sustained and vigilant reconsideration.

In this essay I shall tell a story of how two philosophers, Ludwig Wittgenstein and Robin Collingwood, answered my questions, partly through precept but largely by example, during the first half of the twentieth century. By doing so, I intend to show how the last of three answers given by Wittgenstein, commonly overlooked, and the arduous journey that he had to undertake to arrive at it, seldom traced, exemplify the kind of answer that Collingwood had concurrently been advocating, refining and exemplifying. I shall conclude by suggesting, however, that a fourth answer upon which they converged unwittingly points toward a way-station even further along the road to philosophical understanding than either of them was able to go.

More exactly, I shall devote most of this essay to describing how Wittgenstein came, by his own account, to reconfigure the problems and solutions of his first book, the Tractatus Logico-Philosophicus, into those of his last book, the Philosophical Investigations, showing how he came to replace the first of the answers that he had given to our questions with a second. I shall tell this story, however, only to confirm a third answer that Wittgenstein was to exemplify in a sequel to it, namely that one can come to understand the meaning of a text only by coming to understand it historically – the answer that Collingwood would have applauded. [5] I shall then suggest, however, that both men, as they tried late in life to solve the problem of how best to ensure that their life's work would be understood as they intended (as a means, that is, to the kinds of ends they wished), showed us – without realising it – how, and why, the answer to our questions, if philosophically exact, must be not only historical but biographical.

To open a window onto the relevance to artistical learning of the issues of this essay, however, I shall begin by describing two events that occurred a third of a century apart within my own 'autobiography', showing how I came at last to understand better how to design and construct screenplays that mean more exactly, having learned what I could from the lives and works of Collingwood and Wittgenstein about how to think better philosophically. [6]

September 1967

In September of 1967 I entered a university classroom for the first time as a 'professor'. I was only a student, having enrolled within a doctoral programme in philosophy after completing the requirements for a master's degree in filmmaking yet to be received; but I had been asked by the film faculty to teach a course for graduate students on any subject of interest to me, 'provided it had something to do with film history', and I, needing the money, had agreed, 'provided' that the course could concentrate upon the history of Russian and Soviet film. For a good text with that title had become available, a selection of representative films could be obtained from a dedicated distributor, the filmmakers had been uncommonly articulate about how they had been thinking when working upon them, and I, although unaccustomed to teaching, had surmised that I should learn better about them if compelled to introduce them coherently to others.

Within weeks, however, I remarked to my wife, colleagues and friends that I was going to write a book on 'the foundations of filmmaking', for the texts I was pondering by the filmmakers upon the work that they had done seemed to me to be, however intriguing, barely intelligible; and the commentaries by critics upon the films and writings of those filmmakers, and much else, were even worse.

My pronouncement must have amused some of my peers, for I had as yet made few films, and assuredly none informed by any mastery of the craft (my degree notwithstanding). But I knew what I intended to do, and how to do it, and would have dismissed the cautions of others; for, as a promising young logician, I had been well­ trained in the techniques of analysis pervading the dominating tradition of philosophical enquiry within North America. I had also, of course, though unwittingly, been indoctrinated into the pretences pervading the practice.

With an ignorance and arrogance that now astonishes me, I set out to do for filmmaking what others had done (or better, had claimed to have done, or to be doing or to be about to do) for much of mathematics, physics, biology, psychology, sociology and even ethics and aesthetics: I would harvest the remarks made by respected filmmakers, winnow the wheat from the chaff, convert it into flour and then, after adding the yeast of my training and acumen, knead the dough into a logically ordered loaf that, when baked, would render the precepts of filmmaking forever after palatable.

Looking back, I was merely intending to do as I had been trained to do, having been misled into believing that I could analyze and clarify how films could more meaningfully be constructed by applying the logical tools I possessed to a body of texts about filmmaking that I had read and pondered without having bothered to learn, through long and hard practice at making films informed by how others had made them before me, how to distinguish the problems that those filmmakers had faced and tried to solve by constructing their films as distinctively as they did.

At the back of my mind, however, serving as the beacon of promise of the tradition within which I was working, was a text compiled from notes near the end of the first world war by the young Wittgenstein, the Tractatus Logico-Philosophicus. This little book, consisting of short paragraphs encompassing often only a single assertion or two of remarkable brevity and provocation, arranged and numbered in supposed logical order and bereft of argument, examples, articulation of questions, discussion or historical context, was as alluring to me as it had been upon publication to the members of the Vienna Circle, and thereafter to most workers within the tradition they engendered. [7]

I hardly wished to mimic the design of the book, for, as its author had anticipated even before publication, it had been misused by careless readers while proving unintelligible to careful ones, largely, or so it seemed to me, because of the oddities of its exposition. The fascination of the book lay rather in the transcendent promise of its pretensions, namely that one could instruct human beings on how to think better by articulating in logical order the few foundational propositions upon which a discipline of knowledge rested.

As Wittgenstein would later admit, he had exaggerated his achievement when he had claimed in his prefatory remarks to the text to have "found, on all essential points, the final solution of the problems" addressed within it, [8] for its "grave mistakes" were hardly confined to a muddied exposition. [9] But if the promise of his text was legitimate, then a young man like me might hope to articulate within a work of comparable scope and nature, constructed solely with the logical tools that Frege had taught us to sharpen, the primary precepts of some narrower part of the task, 'the foundations of filmmaking'.

I was presuming, of course, as I had been trained to do, that how filmmakers were speaking, thinking and working ought to conform to how I, in company with other philosophers (and especially logicians), had been trained to speak, think and work; for I had been led to believe, as the young Wittgenstein had believed, that I knew of a shortcut to learning how better to mean that would enable me to evade the clutter of historical enquiry, skirting around the tough task of emulating, and then cautiously extending, the hard-earned practices of others.

The book I promised to write in 1967 was (thankfully!) never completed. Why not? Because I began to realise, little by little, that I could learn how better to make films only by learning how better to think of the films made by others as they had thought of them. Only by emulating imaginatively how filmmakers had tried to solve the problems they faced at the time and place of their making could I learn better how to make films. I began to learn, in other words, how better to make things as writers, composers, painters and filmmakers have always had to learn it, by coming to understand how artefacts made by others had come to mean as they did, having been made as they were.

By doing so, however, I was also weaning myself, though unwittingly, from the pretensions of the Tractatus, retraining myself to learn how texts mean, including philosophical texts. And the lesson I was learning, though it would take me three decades to comprehend it, was almost exactly that which Wittgenstein had himself had to learn when transfiguring the habits of thinking exemplified by the Tractatus into those exemplified by the Philosophical Investigations – Collingwood's lesson, or almost so. But that brings me to the second story from within my 'autobiography'.

April 2003

A third of a century later, in April, 2003, I exited in the early afternoon from a university classroom, having met for the last time with a few advanced students of screenwriting who had persevered with me to the conclusion of a year-long double-weighted course in the history of screenwriting that had met for eight hours weekly – a course that proved to be, in their judgment and mine, remarkably successful.  

Each of those students had witnessed with me over eighty feature films written over the past seventy-five years by noteworthy screenwriters, had compiled a notebook of weekly meditations on problems of screenwriting exemplified by the films screened, and had submitted two major master-scene treatments for screenplays of feature-length of their own, at least one of them a rewriting of a major novel, play or screenplay.

During our final discussion, I asked the students which, if any, of the popular 'theories' of screenwriting they still thought to be useful, having pondered the range of screenplays, and films made from them, that we had been privileged to encounter. After a moment of silence, the most accomplished of them replied that he believed no longer in any of the precepts of screenwriting that he had been trained to respect, for he had now encountered counterexamples to every one of them; and yet, he added, he had by doing so learned far more than ever before about how better to ponder, write and rewrite screenplays.

Readers unacquainted with the ways of emulation through which neophyte artists have always learned their crafts, but familiar with the common misappropriation by many of the later writings of Wittgenstein, may have been tempted by now to consign my remembrances of things past into the waste-bin of 'just so' stories told so often by moribund philosophers of how they emerged from darkness into light by attending to whatever they took Wittgenstein to have meant by whatever he seemed to have said (that one ought, for example, to seek meaning in use, to avoid searching for 'essences', to discriminate among particulars rather than lumping them together 'theoretically', and, most enticing of all to those who never learned how to speak elegantly, to show rather than say).

But my students, early and late, had been constrained to encounter the works of others only in the historical order of their creation, and my later students had been required as well to discriminate among them by performing imaginative acts of biographical emulation, pondering what they could have said to the screenwriters in their time and place, had they been working alongside them trying to solve the problems that they were facing, that would have prompted them to have done something better. By the end of the later course, my students had indeed learned how to say, do and make things better by attending to 'examples' of how screenwriters had 'spoken ordinarily' while solving the commonplace problems of their work, and by doing so had freed themselves from the bewitching 'theoretical' chatter that had previously misled them. Unsurprisingly, therefore, the therapeutic game they had played bore a 'family resemblance' to that which Wittgenstein had recommended we play to overcome the seductive misuses of 'language'.

But the difference is crucial. Rather than "examining what we should say when" (John Austin's apt précis of the kind of ahistorical game that Wittgenstein and he had thought adequate to avoiding linguistic clutter, replete with royal 'we', subjunctive 'should' and ahistorical 'when'), [10] my later students had been required to imagine how they might have responded to what a screenwriter, or writers, had said when designing a film within a studio in Hollywood in 1928, and then to another writer, or writers, working upon a film on the streets of Rome in 1946, and then to yet others working … . Rather than presupposing that any of the writers whose screenplays, novels, plays, letters, notebooks, etc., they had encountered had been speaking within some kind of ahistorical 'ordinary language', using words constrained ahistorically by some kind of 'grammar' (or 'rule') of which we could become aware by discriminating ahistorically among examples, and which, once recognised, would 'tell us what kind of objects' we were encountering, [11] my students had been required to presuppose that there could be no such things.

Despite the superficial 'family resemblance' to the ahistorical game of Wittgenstein, my later students had instead been playing the kind of historical game that Collingwood had recommended.

Later in this essay I shall describe how Wittgenstein came 'suddenly' in 1941 to confirm that Collingwood's game, rather than his own, was to be preferred. I must first, however, introduce those precepts to you as Collingwood summarised them in 1938 when obliged to try one last time, or so he thought, to ensure that readers would be able to understand how he intended them to be used (to mean); for not only had those precepts immunized him from the 'analytical' tradition of enquiry, derived from Wittgenstein's Tractatus, that had dominated the English-speaking philosophical world during his lifetime, isolating him within it and delaying his influence upon it until recently, but how he chose to summarise them exemplifies exactly the lesson that Wittgenstein was to learn three years later – though neither of them was ever to know what the other had done.

Collingwood's 'Autobiographical' Precepts

Collingwood, while studying classics and philosophy as a student at Oxford, had trained as well to become an archaeologist of Roman Britain, continuing the practical work that he had begun under the tutelage of his father but now concentrated under the supervision of the most noted authority on the subject, Francis Haverfield – the "least philosophical of historians". [12] As one of the very few of Haverfield's students to survive the first world war, Collingwood felt obliged after it to continue the work of his mentor, and was to spend almost all of his time, when not teaching philosophy, as a working archaeologist. By the end of his life, he had become England's ranking authority on Roman Britain.

In February of 1938, at age forty-nine, Collingwood, who had been in increasingly poor health since the early 1930s, suffered a major stroke. Fearing that he would die without having completed any of the major projects upon which he was working, he decided "to put on record some brief account of the work I have not yet been able to publish, in case I am not able to publish it in full". (C38:118) Remarkably, he chose to write it autobiographically. The result is one of the modest miracles of philosophy, for never has a major philosopher summarised, with less clutter or greater style, the conclusions of a considered life while exemplifying them by doing so. [13]

Near the end of his Autobiography, Collingwood turned to an example of the kind of problem that he had been obliged to confront every day when working as an archaeologist, and whose consequences for philosophical thinking he had been pondering and refining ever since his first publication in 1916, for it defied accommodation within the pseudo-perceptual models of 'knowing' everywhere advocated within the dominant tradition of 'analytical' positivism. Suppose, Collingwood remarked, that an archaeologist at work upon a site between Tyne and Solway were to uncover yet another elongated section of shaped rock, aligned with others, that might seem to have been part of a wall. [14] What must the archaeologist do to come to understand what has been uncovered?

(a) The archaeologist must acknowledge that the object is an artefact that was constructed by human beings in the past to serve as a means toward ends they had wished to accomplish. To understand why the artefact was shaped and positioned as it is, the archaeologist must determine the meaning (use) that its makers had intended it to have.

(b) The question 'What was the artefact intended to mean?' can only be answered, however, through an empathetic act of imagination. The archaeologist must imagine the problem that its makers had faced, rethinking the object, as they had thought of it when making it, as a solution to that problem.

(c) The problem solved by the making of that artefact at that time and place, however, differs from the problems solved by other artefacts made by other makers at other times and places.

(d) To learn how an artefact was intended by its makers to mean (to be used), therefore, an archaeologist must engage unexceptionally in the evidentiary and open-ended task of coming to imagine better how its makers had tried to solve the historically specific problem they had faced by making it as they did. As Collingwood summed up the practice in his Autobiography,

For the archaeologist this means that all objects must be interpreted in terms of purposes. Whenever you find any object you must ask, 'What was it for?' and, arising out of that question, 'Was it good or bad for it?' i.e. was the purpose embodied in it successfully embodied in it, or unsuccessfully?' These questions, being historical questions, must be answered not by guesswork but on historical evidence; any one who answers them must be able to show that his answer is the answer which the evidence demands. (C38:128)

Unless the archaeologist is content merely to describe what he or some one else has found, which it is almost impossible to do without using some interpretative terms implying purpose, like 'wall', 'pottery', 'implement', 'hearth', he is practising Baconian history all the time: asking about everything he handles, 'What was this for?' and trying to see how it fitted into the context of a peculiar kind of life. (C28:133 and 134)

Since archaeologists must presume that the meaning of an object is the use to which its makers had intended it to be put, identifiable only by imagining the problem that they had intended it to solve, saying so was simply to repeat, as Collingwood observed, "the tritest of commonplaces", precepts "obvious" to every archaeologist. (C38:128 and 133) Were he to have suggested applying them only to neighbouring disciplines, he might have garnered a modest if prosaic reputation as an observer of how archaeologists do, and other historians might better, behave.

Instead, Collingwood insisted that his precept applied with equal warrant to the craft of philosophy, for texts, whether philosophical or otherwise, are artefacts akin to Roman walls, paintings, letters, screenplays or any other objects left behind by human beings, having been created by them to solve problems they had faced at that time and place. To uncover the meaning of a philosophical text, one must engage in the kind of imaginative, painstaking and evidentiary enquiry required to construe it as the intended solution to the historically unique problem that it had been constructed by its author to address. We can come to understand how Aristotle intended his Metaphysics to be used, for example, or Descartes his Meditations, Kant his critiques or Wittgenstein his Tractatus, only by approaching the text as historically as we should have had to approach any other artefact to come to understand how it was intended by its maker(s) to be used.

But then a remarkable consequence follows with respect to philosophical authorship. If readers of philosophical texts must construe them historically to understand what they mean, then authors wishing to ensure understanding of their texts must make them historically comprehensibly. Were Wittgenstein, for example, ever to have wanted the meaning of his second book to prove accessible to readers, he would have had to insist that it be published in such a way that it could be encountered as a historical consequence of his first book.

But that brings me to the turning-point of this essay.

The Convergence

The precepts and examples found within Collingwood's Autobiography, like those found within Wittgenstein's Philosophical Investigations, are among the most provocative philosophical legacies of the past century, contrasting starkly with the moribund pretensions of positivism. The pathway of historical learning along which Collingwood was traveling, however, seems far removed from the ahistorical trail along which Wittgenstein was concurrently journeying.

Must we, then, choose between Collingwood and Wittgenstein? Remarkably, no, for the ahistorical precepts and examples that Wittgenstein articulated within the text of his second book are only part of a broader lesson that he had learned by then, and would have had us learn by its publication, had it been published as he wished.

In 1945, when Wittgenstein wrote the preface to the text that he thought would soon be published as his second book, he confirmed that he had "up to a short time ago … really given up the idea of publishing my [later] work in my lifetime" (W45:ixe), for so many readers had for so long misunderstood what he had been trying to say. The aims and purposes of his first book, the Tractatus, had been overlooked or disregarded even by those members of the Vienna Circle who, like Carnap and Neurath, had once claimed most loudly to have been informed by it; and while later pursuing his 'investigations', he had been angered recurringly upon learning "that my results (which I had communicated in lectures, typescripts and discussions), variously misunderstood, more or less mangled or watered down, were in circulation. This stung my vanity and I had difficulty in quieting it." (W45:ixe and xe)

What then changed Wittgenstein's mind? In 1941, as he was working despondently upon the draft of what would appear as Part 1 of the Philosophical Investigations, he found himself rereading the Tractatus while trying to explain to an acquaintance the problems that he had faced, and tried to address, when writing it. As he was doing so, he realised "suddenly" how to solve the conclusive practical problem that he was facing, namely how to ensure that the meaning of his second book would be less easily misunderstood than his first book had been: his 'new thoughts' ought only to be published, he insisted, as the second half of a book whose first half would consist of a reprint of his 'old thoughts', the Tractatus.

Four years ago I had occasion to re-read my first book (the Tractatus Logico-Philosophicus) and to explain its ideas to someone. It suddenly seemed to me that I should publish those old thoughts and the new ones together: that the latter could be seen in the right light only by contrast with and against the background of my old way of thinking. (W45: xe; italics added) [15]

Wittgenstein's wish was never to be granted. His 'new thoughts' were never to appear after his 'old thoughts' within a single publication, and the result was as he had feared: misunderstanding by readers of the meaning of the culminating work of his life. For his literary executors, and thereafter the publisher, failed to register the importance of what he had 'suddenly' recognised in 1941, namely that he had not only been provoked into thinking his 'new thoughts' by having been compelled over many years to ponder criticisms of his 'old thoughts' by colleagues, [16] but that the precepts and examples of his 'new thoughts' could only be understood by readers willing to undertake imaginatively something like the same kind of historical journey that he had had to take to transfigure the one into the other. Wittgenstein, having been compelled by happenstance in 1941 to try to enhance someone's understanding of his 'new thoughts' by explaining his 'old thoughts', had come 'suddenly' to realise what Collingwood would have predicted, namely that the meaning of the precepts and examples within his new book could only be understood if registered historically as a consequence – as a part of the (unintended) meaning – of his first book.

Wittgenstein's journey from his first to his second book is therefore a 'case study' of especial relevance by which to measure the usefulness of the admonition of Collingwood that he had appropriated, that we must approach texts historically if we are to learn better how to use them. But where should we begin if we wish to reconstruct a more insightful narrative of that journey? By pondering, I suggest, some conversations that Wittgenstein had in the midst of it with Moritz Schlick, founder of the Vienna Circle, and his young assistant, Friedrich Waismann, during the crucial period just before and after 1930 when he was beginning to sense that the mistakes that had disfigured the Tractatus had arisen from a single pervasive presumption whose rejection would have radical consequences.

First, however, we must attend to an analogy that Schlick was to draw in 1931, shortly after those conversations had taken place, between the approach to philosophy that he had learned from Wittgenstein and the behaviour of Socrates; for, by conjoining Schlick's analogy with a second one that Wittgenstein was soon after to draw between how he had mistakenly been thinking when writing the Tractatus and how Plato had thought, we shall be able to plot the trajectory of how Wittgenstein was to transfigure his first book into his second.

A Pair of Analogies, 1931 & 1933

In 1922 Schlick was appointed to the chair of Philosophy and History of the Inductive Sciences at the University of Vienna, convening soon thereafter the first meeting of the select group of bright and industrious students of science and mathematics that was to become known as the 'Vienna Circle'. Wittgenstein's Tractatus had been published a year before, and Schlick, after reading it, convinced the members of the seminar to subject the text to a line-by-line reading and discussion, the first of two such studies, taking most of an academic year, that the Circle was to devote to the book.

By 1924 Schlick had become so intrigued with Wittgenstein's thinking that he wrote to him, trying to secure a meeting between them. Wittgenstein, though flattered by Schlick's interest, resisted the overture and several sequels. At last, however, he agreed, and in February 1927 the two men met at one of the residences of the Wittgenstein family. So taken were they with one another, and with their discussions, that further conversations ensued, attended almost always by Waismann and frequently by Carnap, Feigl and other members of the Circle.

By 1929 it had become obvious, however, that Wittgenstein was no longer willing to talk with anyone other than Schlick and Waismann, for only they had seemed interested in how his thinking was now evolving beyond what he had said in the Tractatus. Their curiosity was genuine, for Schlick, Waismann's mentor, alone among the senior members of the Circle, had seriously studied philosophy, knew something other than hearsay about its history and respected it. Unlike Carnap, for example, who knew little and cared less about philosophy but was convinced that it had to be destroyed, [17] Schlick saw himself, and the movement that he headed, as trying to accomplish what philosophers had before tried to do, but inefficiently, namely to reform the discipline. To the dismay of Carnap, Feigl and the other senior members of the Circle, Schlick had by 1929 become more rather than less interested in what Wittgenstein was coming to say, for he alone had been able to sense that it might prove to be significant.

When Wittgenstein and Schlick, accompanied by Waismann, resumed their conversations during the Christmas vacation of December 1929, Waismann was delegated to transcribe them. We have thus a provocative record, subject to the infelicities of transcription, of much of what Wittgenstein said to the two of them through January 1931 (and thereafter, though less complete, to Waismann alone through July 1932), responding often to their queries, during the decisive period when he, having now 'returned to philosophy', was for the first time lecturing at Cambridge and trying to articulate the redirection of his thinking.

Later in 1931 Schlick addressed an audience in Stockton, California on the topic "The Future of Philosophy". The aim of the talk, given in English, was to acquaint his American listeners with the nature of the philosophy of Wittgenstein that had so influenced the work of the members of the Vienna Circle, a subject upon which only he among its senior members was now qualified to speak, for only he had persevered in tracking its evolution. Schlick concluded his address by summarising the conception of philosophy that he had been expounding.

The view which I am advocating has at the present time been most clearly expressed by Ludwig Wittgenstein; he states his point in these sentences: "The object of philosophy is the logical clarification of thoughts. Philosophy is not a theory but an activity. The result of philosophy is not a number of 'philosophical propositions', but to make propositions clear." This is exactly the view which I have been trying to explain here. [18]

Schlick had tried to explain it, however, by drawing the attention of his listeners to a figure whose conduct almost two and a half millennia before had exemplified the kind of philosophical behaviour that he, after Wittgenstein, was advocating.

All of you have probably read some of Plato's Dialogues, wherein he pictures Socrates as giving and receiving questions and answers. If you observe what was really done – or what Socrates tried to do – you discover that he usually did not arrive at certain definite truths which would appear at the end of the dialogue but the whole investigation was carried on for the primary purpose of making clear what was meant when certain questions were asked or when certain words were used. … Socrates asks 'What is Justice?'; he receives various answers to his question, and in turn he asks what was meant by these answers, why a particular word was used in this way or that way, and it usually turns out that his disciple or opponent is not at all clear about his own opinion. In short, Socrates' philosophy consists of what we may call 'The Pursuit of Meaning'. He tried to clarify thought by analyzing the meaning of our expressions and the real sense of our propositions.

Here then we find a definitive contrast between this philosophical method, which has for its object the discovery of meaning, and the method of the sciences, which have for their object the discovery of truth. In fact, before I go farther, let me state shortly and clearly that I believe Science should be defined as the 'pursuit of truth' and Philosophy as the 'pursuit of meaning'. Socrates has set the example of the true philosophical method for all time. (S31:216 and 217)

Four years later, in 1936, Schlick was to be murdered on a staircase leading to the philosophy rooms of the University of Vienna by a deranged student, depriving Wittgenstein of the understanding and support of the only living philosopher he had ever respected as a peer, and affecting him profoundly. Schlick was never to know of the severe constraints upon philosophical behaviour that Wittgenstein would soon after recommend after having travelled further down the path that Schlick's analogy had predicted; and, had he known, he might never have accepted them. Nevertheless, he was the first of only a few philosophers ever to sense and respect the continuity of the evolution of Wittgenstein's thinking during and after the writing of his first book, having recognised that Wittgenstein's 'newer thoughts' were simply the latest in a sequence of attempts, reaching back through the 'old thoughts' of the Tractatus, to address an evolving series of profound problems distinct from the logical niceties that had comprised its surface – and that Wittgenstein had become increasingly like Socrates in his approach to them.

Wittgenstein was never to compare himself to Socrates. Strikingly, however, he was soon after to identify the fundamental mistake that he had made when writing the Tractatus with the kind of mistake that Plato had made. Remarking around 1933 upon the 'objects' that he had once insisted must 'make up the substance of the world' for propositions to have sense (see Table, 2.02-2.0212, and the next section of this essay), Wittgenstein acknowledged that he had been misled by the kind of ‘picture of reality’ that Plato had drawn:

'Nothing is as dead as death; nothing as beautiful as beauty itself!' The picture according to which reality is thought of here is that beauty, [or] death, etc., is the pure (concentrated) substance, whereas in a beautiful object it is contained as an admixture — And don't I recognize here my own observations about 'object' and 'complex'? (Plato.) [19]

Schlick had observed in 1931 that Wittgenstein's thinking had been directed toward ends increasingly Socratic. Wittgenstein, in turn, had by 1933 come to realise that the root mistakes of the Tractatus had been Platonic. We need only to conjoin those analogies to grasp that the transfiguration of Wittgenstein's first book into the second was to be an extended journey backward from Plato to Socrates.

Wittgenstein's Tractatus must then have been erected upon a foundation of Platonic presumptions, centred around the notions of 'elementary proposition' and the 'objects' named within them, whose Socratic rejection by Wittgenstein as useless (that is, meaningless) would culminate in the precepts and examples of the Philosophical Investigations. How then did the notions of 'elementary proposition' and 'objects' mislead the young Wittgenstein as he wrote the Tractatus? How were they Platonic? How were they to be Socratically overcome?

The 'Elementary Propositions' and 'Objects' of the Tractatus

The aim of the Tractatus, as evident from the ordering and numbering of its remarks, was to enable readers to distinguish what could be said from what could only be shown, and both from nonsense, for only after having learned how to do so could readers observe the seventh and concluding commandment of the book:

7.0 Whereof one cannot speak, thereof one must be silent. [20]

Only then would they be able to appreciate how the young author of the book had "found, on all essential points, the final solution of the problems" that he had addressed, and "how little is achieved when these problems are solved" (W18:5), for only then would they have come to recognise that, since only propositions can say anything about the world, sentences purporting to articulate the 'problems of philosophy' or their solutions, having no sense, are useless (ethical and aesthetic pronouncements among them).

The central task of the book was therefore to disclose 'the general form of a proposition', the shape that a sentence must have to be capable of representing a fact about the world – capable, that is, of being used to say something about it.

Wittgenstein solved the problem, or seemingly so, by stipulating that propositions were of two kinds, distinguished by how their truth (or falsity) could be established. Some propositions, the elementary ones, could be determined to be true (or false) only by attending to how the world happened to be, for having no logical structure, they required no other proposition to be true or false. The truth (or 'falsity') of all other propositions, however, compounded of elementary ones through repeated applications of the fundamental logical operation of 'joint [or alternative] denial' (indicated within the text by the sign '|'), could be determined by noting whether the elementary propositions of which they were compounded were true (or false). [21]

If Wittgenstein, when drafting the Tractatus as an answer to the problems he had found within Russell and Whitehead's Principia Mathematica, had been able to anticipate the "recursive mode of thought" that the young Thoralf Skolem would soon introduce into mathematics in response to the same challenge, he might well have summarised in three sentences the 'general form of a proposition'. [22]

(1) An elementary proposition is a proposition.
(2)  If A and B are propositions, then A|B is a proposition.
(3) There are no other propositions.

Had he done so, readers might have recognised more easily that his account of the world, and how we may speak of it, rested upon his conviction that there must be elementary propositions. Wittgenstein had come to believe so, of course, not by having uncovered any such things (he gave within his book no examples of them), but rather by having inferred that they must exist if we are to account for our ability to speak sensibly, as we sometimes do, about the world. But why had he concluded that we should be prohibited from speaking sensibly of the world were every proposition to be compounded logically of others?

Wittgenstein believed at the time that a proposition was true if and only if it 'pictured' one or another of the positive (or negative) facts of which the world (reality) consisted, and that the sense of a proposition consisted in how it could be verified. (See Table, 2.04-2.063, and the following sentences.)

2.201 A picture depicts reality by representing a possibility of existence and non-existence of states of affairs.

2.203 A picture contains the possibility of the situation that it represents.

2.21 A picture agrees with reality or fails to agree; it is correct or incorrect, true or false.

2.22 What a picture represents it represents independently of its truth or falsity, by means of its pictorial form.

2.221 What a picture represents is its sense.

2.222 The agreement or disagreement of its sense with reality constitutes its truth or falsity.

2.223 In order to tell whether a picture is true or false we must compare it with reality. 

He believed, that is, in what would become known as the 'correspondence theory of truth' and the 'verification theory of meaning'. [23] Were there to be no elementary propositions, deriving their sense by representing independently of others one of the elementary states of affairs (Sachverhalte) of which the world consisted, [24] then, or so he concluded, every proposition, compounded of others, would lack sense; for no matter how finely one distinguished its compounds, one would never arrive at any part of it representing the world directly.

A dozen years later, in 1930, Waismann was to capture within his Theses the gist of the argument that had compelled Wittgenstein to conclude that there must be elementary propositions.

A verification-path cannot lead to infinity. (An 'infinite verification' would no longer be a verification.)

To be sure, a proposition can lead back to other propositions, and those back to yet other ones, etc., but ultimately we must reach propositions that do not indicate further propositions, but point to reality. …

If it were otherwise, no proposition could be verified. There would then be no connection between language and the world.

The propositions that deal with reality immediately are called elementary propositions.

It is not an hypothesis that there are elementary propositions. The requirement that elementary propositions should exist is the requirement that our statements have sense. The fact that we understand the propositions of our ordinary language already guarantees that there are elementary propositions.

The elementary propositions are what give all other propositions sense. (FW29-32:247 and 248)

Wittgenstein, however, had only hinted at the argument within the Tractatus, suggesting that the need and nature of 'elementary propositions' were as "obvious" as the need and nature of the 'objects' named within them, leaving it for readers to reconstruct the argument on their own.

2.02 Objects are simple.

2.021 Objects make up the substance of the world. That is why they cannot be composite.

2.0211 If the world had no substance, then whether a proposition had sense would depend on whether another proposition was true.

2.0212 In that case we could not sketch out any picture of the world (true or false).

3.202  The simple signs employed in propositions are called

3.203 A name means an object. The object is its meaning. …

3.21 The configuration of objects in a situation corresponds to the configuration of simple signs in the propositional sign.

3.22 In a proposition a name is the representative of an object.

4.21 The simplest kind of proposition, an elementary proposition, asserts the existence of a state of affairs.

4.22 An elementary proposition consists of names. It is a nexus, a concatenation, of names.

4.221 It is obvious that the analysis of propositions must bring us to elementary propositions which consist of names in immediate combination. [25]

4.2211 Even if the world is infinitely complex, so that every fact consists of infinitely many states of affairs, and every state of affairs is composed of infinitely many objects, there would still have to be objects and states of affairs.

What, then, were these 'objects'? Surely, or so it seemed to Carnap, Neurath and most other senior members of the Vienna Circle, as philosophically illiterate as they were scientifically enthused, but enraptured by the promise of the logical techniques encompassed within Wittgenstein's little book, the 'objects' named by the names comprising the most 'elementary' of the propositions true of the world must be the most 'elementary' things of which the world consisted. As 'obviously' as the need for elementary propositions naming 'objects' had seemed to Wittgenstein, it now seemed to them that 'objects' had to be the simplest parts of the world – the smallest irreducible bits of which the world was built, akin in scope and nature to the elementary particles being sought by physicists as the fundamental building blocks of the world when construed materially, or to the 'coloured patches' being postulated by sense-data theorists as the building blocks of the same world when construed phenomenologically.

Before presuming so, however, they would have been well advised to have pondered a short-list of the remarks that Wittgenstein had made within his book about its 'objects' (see Table), for, had they done so, they could hardly have failed to recognise that Wittgenstein's 'objects' could not be parts of the world, whether materially or phenomenologically construed. To the author of the Tractatus,

(a) The world, being the "totality of facts, not things", encompasses "all that is the case" (1 and 1.1);

(b) The "substance" of the world, however, "subsists … independently of what is the case" (independently, that is, of the world), determining only its "form, and not any material properties" of it (2.024 and 2.0231; italics added); and

(c) "Objects make up the substance of the world". "Objects are just what constitute [its] unalterable form". "Objects, the unalterable and the subsistent are one and the same" (2.021, 2.023 and 2.027; italics added).

Wittgenstein's 'objects', in other words, were hardly irreducible parts of the world, elementary entities within it whose natures, material or phenomenological, we might hope someday to confirm by closer examination of it. Rather, they 'subsisted independently' of the world as a priori, 'formal' pre-conditions of it. [26] The existing 'states of affairs' that comprised the world consisted of the 'changing and unstable configurations' of these 'objects', rather than of the 'objects' themselves. [27]

One may sympathise with the difficulties that the members of the Vienna Circle faced when trying to comprehend what the young Wittgenstein had meant by saying what he did about 'objects', for he had been playing as fast and loose with the supposed difference between 'existence' and 'subsistence' as any theologian of the Middle Ages had done. As Wittgenstein was to confirm in 1937 (or 1938), he had been 'misusing' the word 'object', 'misleading' himself and others.

I at one time said (in the Tractatus Logico-Philosophicus), that the 'elementary proposition' is a concatenation of names. For, to the names correspond objects, and to the proposition corresponds a complex of them. The proposition 'The bottle is to the right of the glass', if it is true, corresponds to the complex consisting of the bottle, the glass, and the relation to-the-right-of. ... What a misuse of the words 'object' and 'complex'!! ... To say a red circle 'consists of' redness and circularity, is a complex of these constituents, is a misuse of these words, and is misleading. [28]

He had done so, however, because, as he had noted in 1933, he had been misled as Plato had been misled: the 'objects' that he had intended to be understood as the 'unalterable and subsistent' formal preconditions of the world were comparable to the forms, or ideas, of Plato (see W33:199, quoted above at the conclusion of the previous section of this essay). Just as a pious act, for example, had been for Plato a 'concatenation' of a multitude of forms, one of them distinguishing it essentially from all non-pious acts, and all of them distinguishing it collectively from other pious acts, so, for Wittgenstein, the 'configuration of objects', designated by the 'concatenation of names' constituting an elementary sentence, 'produced' one of the 'states of affairs' of which, if existent, the world consisted.

Wittgenstein's Tractatus remains, in my judgment, the most profoundly Platonic text to have been written after Plato's own, though decorated beyond easy recognition. Its young author had fallen into the habit of misusing language as Plato had done, and was to grope his way out of the "fly-bottle" that he had himself constructed only by learning, step by slow step, how to speak without using the words 'elementary proposition' and 'object' – only, that is, by coming to realise that they, having no use, had become meaningless. [29]

The Way Out of the Fly-Bottle

To serve as the logical building blocks of the world, elementary propositions, each describing an 'elementary' state of affairs, had to be logically independent of one another, their truth (or falsity) being determined solely by how the world is rather than being inferable from the truth (or falsity) of some other propositions.

4.211  It is a sign of a proposition's being elementary that there
can be no elementary propositions contradicting it.

5.1314 One elementary proposition cannot be deduced from  another.

Even careful readers of the Tractatus were perplexed, therefore, to discover that Wittgenstein believed as well that we could be assured that nothing in the world could appear, for example, to be both red and green, for propositions such as 'A is red' and 'A is green' appear to be as logically pristine (that is, truth-functionally uncompounded) as any propositions could be. Yet Wittgenstein had insisted that they were not only incompatible, but logically so.

6.375 Just as the only necessity that exists is logical necessity, so too the only impossibility that exists is logical impossibility.

6.3751 For example, the simultaneous presence of two colours at the same place in the visual field is impossible, in fact logically impossible, since it is ruled out by the logical structure of colour. … The statement that a point in the visual field has two different colours at the same time is a contradiction.

We could be assured a priori, that is, without examining any red or green things, that were a proposition 'A is red' to be true, we could infer that 'A is green' is false, even though the inference appears to be non-logical, controverting what would seem to be the primary principle of the remainder of the book. [30]

The problem had seemed insignificant to Wittgenstein when writing the Tractatus only because he had earlier concluded that the simple appearance of propositions such as 'A is red' and 'A is green' was deceiving. As he had remarked in his notebooks,

If the logical product of two propositions is a contradiction, and the propositions appear to be elementary propositions, we can see that in this case the appearance is deceptive. (E.g. A is red and A is green.) (W14-17: 91e, 1 August 1917)

By 1929, however, when conversing with Schlick and Waismann, Wittgenstein had come to recognise that he had made a mistake with respect both to propositions about colour and to the bounds of inference, and had begun to comprehend what would have to be done to correct it. He had been able to do so, however, only by having been compelled to attend to a far more encompassing problem of mathematical representation, namely that elementary propositions, if they were to serve their intended purpose, would have to encompass real numbers. As he would soon discover, however, elementary propositions could encompass real numbers only by becoming non-elementary – only, that is, by destroying their sense, rendering them, and thus the foundation of his first book, useless.

How did Wittgenstein come to acknowledge the problem and try to solve it? By realising, in effect, that he had failed for over fifteen years to recognise the fecundity of a metaphor by which he had tried as early as 1914 to grasp the nature of the relation between elementary propositions and the world.

Proposition and situation are related to one another like the yardstick and the length to be measured. (W14-17:32e, 24 November 1914)

The proposition is a measure of the world. (W14-17:41e, 3 April 1915)

By the time of the Tractatus, Wittgenstein had begun to refine the metaphor. An elementary proposition, deriving its sense from its capacity to 'picture' one of the fundamental states of affairs of which the world consists, did so by taking the 'measure' of that state of affairs as if one were applying only the end-points of a segment of a ruler to it.

2.1511 That is how a picture is attached to reality; it reaches right out to it.

2.1512 It is laid against reality like a measure.

2.15121 Only the end-points of the graduating lines actually touch the object that is to be measured.

A decade later, however, Wittgenstein had come to realise that he had seriously underestimated how accurately the metaphor of an elementary proposition measuring the world like a ruler pictured the act of 'picturing'. Were, for example, some A to be located within the world 3.6237… metres to the left of some B, the elementary proposition picturing that elementary state of affairs would to represent the fact that 'A is 3.6237… metres to the left of B'; for otherwise one of the elementary states of affairs of the world would have escaped representation as it is (that is, as it is exactly) by any elementary proposition.

As Wittgenstein said to Schlick and Waismann on 22 December 1929,

Now I think that there is one principle governing the whole domain of elementary propositions, and this principle states that one cannot foresee the form of elementary propositions. It is just ridiculous to think that we could make do with the ordinary structure of our everyday language, with subject-predicate, with dual relations, and so forth. Real numbers or something similar to real numbers can appear in elementary propositions, and this fact alone proves how completely different elementary propositions can be from all other propositions. And what else may appear in them we cannot possibly foresee today. Only when we analyze phenomena logically shall we know what form elementary propositions have. Here is an area where there is no hypothesis. The logical structure of elementary propositions need not have the slightest similarity with the logical structure of propositions.

Just think of the equations of physics – how tremendously complex their structure is. Elementary propositions, too, will have this degree of complexity. (FW29-32:42)

Wittgenstein had by then realised that he might have avoided the gravest of the mistakes of his first book had he attended more exactly in 1918 to the implications of the metaphor that he had envisaged as early as 1914.

Once I wrote, 'A proposition is laid against reality like a ruler. Only the end-points of the graduating lines actually touch the object that is to be measured.' I now prefer to say that a system of propositions is laid against reality like a ruler. What I mean by this is the following. If I lay a ruler against a spatial object, I lay all the graduating lines against it at the same time.

It is not the individual graduating lines that are laid against it, but the entire scale. If I know that the object extends to graduating line 10, I also know immediately that it does not extend to graduating lines 11, 12, and so forth. The statements describing for me the length of an object form a system, a system of propositions. Now it is such an entire system of propositions that is compared with reality, not a single proposition. (FW29-32:64)

By rethinking the metaphor, Wittgenstein had solved as well, or so he thought, the problem of incompatible colour predicates. [31]

If I say, for example, that this or that point in the visual field is blue, then I know not merely that, but also that this point is not green, nor red, nor yellow, etc. I have laid the entire colour-scale against it at one go. This is also the reason why a point cannot have different colours at the same time. For when I lay a system of propositions against reality, this means that in each case there is only one state of affairs that can exist, not several – just as in the spatial case. (FW29-32:64)

By doing so, however, Wittgenstein was unwittingly driving a wedge into what was to become an ever-widening crack in the foundations of the account he had given within the Tractatus of the form of the world, and how we represent it logically.

All this I did not yet know when I was writing my work: at that time I thought that all inference was based on tautological form. At that time I had not yet seen that an inference can also have the form: This man is 2m tall, therefore he is not 3m tall. This is connected with the fact that I believed that elementary propositions must be independent of one another, that you could not infer the non-existence of one state of affairs from the existence of another. But if my present conception of a system of propositions is correct, it will actually be the rule that from the existence of one state of affairs the non-existence of all the other states of affairs described by this system of propositions can be inferred. (FW29-32:64)

Looking back, we can begin to understand how far Wittgenstein had yet to travel before he could complete the transfiguration of the Tractatus into the Philosophical Investigations, for, even while affirming the above, he still believed that he could somehow salvage the notions of 'elementary proposition' and 'objects' by further complicating his understanding of them. Three days later he would continue the conversation by saying, 

I want to explain my views on elementary propositions and first I want to say what I used to believe and what part of that seems right to me now.

I used to have two conceptions of an elementary proposition, one of which seems correct to me, while I was completely wrong in holding the other. My first assumption was this: that in analysing propositions we must eventually reach propositions that are immediate connections of objects without any help from logical constants, for 'not', 'and', 'or', and 'if' do not connect objects. And I still adhere to that. Secondly I had the idea that elementary propositions must be independent of one another. A complete description of the world would be a product of elementary propositions, as it were, these being partly positive and partly negative. In holding this I was wrong … . (FW29-32:73 and 74; italics added)

Every proposition is part of a system of propositions that is laid against reality like a yardstick. (Logical space.)

What I first paid no attention to was that the syntax of logical constants forms only a part of a more comprehensive syntax. Thus I can, for example, construct the logical product p.q only if p and q do not determine the same co-ordinate twice.

But in cases where propositions are independent everything remains valid – the whole theory of inference and so forth. (FW29-32:76; italics added)

Wittgenstein had a long way to go, therefore, before he would be able to free himself from the habit of speaking of 'elementary propositions' and 'objects', and from the compulsions of the 'verification theory of meaning' within which he had for so long been entrapped. But he was by now moving in the right direction.

I have here been retracing the first steps that Wittgenstein took, when 'returning to philosophy' in 1929, upon the path that would lead from the Tractatus to the Philosophical Investigations; and enough has been said, I think, to show that coming to understand how Wittgenstein came to say what he said, when he said it, is crucial to coming to understand what he meant by it, as he was himself to conclude a decade later.

I shall only summarise, then, the remainder of his journey, commending it to you for further thought. Wittgenstein had written the Tractatus believing that elementary propositions, and the objects named within them, were required if any proposition was to have a sense (was to be capable, that is, of representing the world); for propositions compounded of others could only have a sense if their 'method of verification' rested ultimately upon the truth (or falsity) of elementary propositions whose sense derived from their ability to represent directly, and without logical mediation, one of the fundamental states of affairs comprising the world.

By the time of his conversations with Schlick and Waismann, Wittgenstein had come to recognise that no propositions of independent sense were possible, for only a proposition selected from within a family of alternative propositions could be used to say anything about the world. No proposition, therefore, however 'elementary', could derive its sense solely from how it described the world. By linking the sense of every proposition to the senses of the others in its family, Wittgenstein had begun to concentrate upon the use to which we put families of propositions rather than individual ones, destroying the raison d'être of the 'verification theory of meaning'.

Soon thereafter he was to recognise that we use many kinds of sentences other than propositional ones, each consisting of words selected from within families of alternatives, as means toward ends. We question, command, exclaim, promise, condemn, etc., using words to mean as we intend. To use words is to mean by them, and conversely; hence, almost always, "the meaning of a word is its use in the language". (W45:43) But then, as Wittgenstein recognised, the 'verification theory of meaning' is redundant, and thus otiose, with respect to propositional sentences, for they, too, are meaningful only when used as means to ends. The 'verification theory of meaning', lacking use, had at last become meaningless.

Philosophy was to remain for Wittgenstein, as it had always been, a way of behaving rather than a doctrine, aimed at helping us to avoid misusing words. Were a successful move to be made within the language game of philosophy, someone would indeed have learned to think more clearly about the world, but without having learned anything more about it. "Philosophy … leaves everything as it is." (W45:124)

Now, however, the norms of clarity constraining the use of a word (its 'grammar', or 'rule'), unlike those envisaged by the young author of the Tractatus, were exemplified by how we ordinarily use it. Only by attending to how we use a word customarily could one relearn how to avoid misusing it. The vocation of philosophers is to distinguish examples of common use that, when pondered 'in a particular way', will immunize us from the lure of the mistaken analogies that tempt us recurringly into misusing the very words that we already know better how to use. The 'methods' of philosophy, properly and humbly pursued, are therapeutic. [32]

Conclusion: The Unfinished Business of 'Essentialism'

In 1933, Wittgenstein, looking back upon the struggle that he had already had to wage with himself on the road to overcoming the mistakes of the Tractatus, remarked that

... philosophy does not lead me to any renunciation, since I do not abstain from saying something, but rather abandon a certain combination of words as senseless. In another sense, however, philosophy requires a resignation, but one of feeling and not of intellect. And maybe that is what makes it so difficult for many. It can be difficult not to use an expression, just as it is difficult to hold back tears, or an outburst of anger." (W33:161)

Wittgenstein was speaking of himself as well as others, for few philosophers have had to struggle harder or longer to disentangle themselves from the habit of pretending to use such useless expressions as 'elementary proposition', 'object' or 'verification', and he was almost immediately to accentuate the autobiographical resonance of his remark.

Work on philosophy is – as work in architecture frequently is – actually more of a work on oneself. On one's own conception. On the way one sees things. (And what one demands of them.) (W33:163)

Five years later, in 1938, Collingwood was to write what he thought might be the final text of his life, insisting within it that a philosophical work is an artefact whose meaning can only be understood by coming to comprehend it as the solution to a problem faced by its author at the time and of its creation – a problem that might resemble others faced at other times by other philosophers, but whose identity was determined by its historical difference from them; and he showed what he meant by saying it autobiographically.

Both men were confirming, each in his own way, that 'work on philosophy' is a transfiguration by a human being of his or her way of thinking. Neither of them, however, was to sense that they were thereby pointing further along the road to philosophical understanding than either of them would be able to go, namely toward acknowledging that the intended meaning (use) of a philosophical text can be ascertained only by coming to understand it biographically. For if 'work on philosophy' is 'more of a work on oneself', then we cannot acquire philosophical understanding simply by registering how an author came at some time and place to construct a text as a solution to a problem bequeathed to him by others. We must rather rewrite that narrative to centre within it the story of how its author came, at that time and place, to transfigure himself by writing it. Philosophical understanding accrues exactly only to biographical enquiry.

Why, then, were Wittgenstein and Collingwood unable to acknowledge it, each of them having come so close to doing so? Because they had yet to overcome the bewitchments of 'essentialism' that each in his own way, had resisted so strongly.

Wittgenstein was to admonish us in his second book to avoid searching for the 'meaning' of a word, an essence supposedly common to, and hence transcending, each of the things to which we may use the word to refer (see the opening quotation of this essay). As often, however, he was to insist, and even more often to imply, that when we use a word (correctly), we are exemplifying the 'grammar' (or 'rule') that constrains its use. When misled into misusing a word, we may, by attending to examples of its ordinary use, be reminded of how we had once been able to use it (correctly), thereby relearning how to do so, [33] but only by attending to it 'in the particular way' that will enable us to become aware again of the 'grammar' (or 'rule') constraining its use.

By speaking nominatively as he did, Wittgenstein was suggesting not only that we are constrained when using a word, but that there must be a constraint that constrains us, the 'grammar' (or 'rule') circumscribing in common, and hence transcending, every instance of its use; for were nothing constraining us when using a word, we should be free to utter it at any time in any situation, but would be unable to mean anything by having done so. But then the 'grammar' (or 'rule') of the use of a word transcends the instances of its use exactly as its purported 'meaning', or 'essence', was once supposed to have done. Rather than disposing of the 'essential', Wittgenstein had pushed it across the line from what can be said into what can only be shown.

I sympathise with commentators who have tried to pretend that Wittgenstein, when speaking of the 'grammar' (or 'rule') of the use of a word, was misusing those words (though few have put it that way), misleading us by using a nominative figure of speech that could be eliminated by some kind of adverbial sleight of hand. Wittgenstein knew better, however, having meant exactly what he said by speaking as he did, as he confirmed later in the book when making two remarks impervious to adverbial tampering.

Essence is expressed by grammar.

Grammar tells us what kind of object anything is. (Theology as grammar.) (W45:371 and 373; italics original)

Wittgenstein was here reaffirming a sentiment that had long suffused his thinking, namely that when we speak of a thing (in contrast to naming it), we do so by conceiving of it as one among other things 'of its kind', and do so essentially. As early as 1916, Wittgenstein, believing in 'elementary propositions', had insisted that "there must be something common to them; otherwise I could not speak of them collectively as the 'elementary propositions' at all". (W14-16:90e; 23 November 1916) Fifteen years later, during the period when he was conversing with Schlick and Waismann, Wittgenstein was to refine his conviction, having begun to realise that only by means of the unspoken 'rules' of language could we express 'what kind of thing an object is'.

For what belongs to the essence of the world simply cannot be said. And philosophy, if it were to say anything, would have to describe the essence of the world. But the essence of language is a picture of the essence of the world; and philosophy as custodian of grammar can in fact grasp the essence of the world, only not in the propositions of language, but in rules for this language which exclude nonsensical combinations of signs. [34]

By 1945, however, thirty years after having first affirmed the need for the 'essential', Wittgenstein had come to recognise, as Peter Hacker would say, that "grammar is autonomous … what we conceive to be natures or essences are …the shadows cast by grammar." (H90:438) Were I to say to you, for example, 'May I please have that cookie?', I should, by speaking as I do, have expressed 'what kind of thing' the cookie is, distinguishing it not only from the plate upon which it rests, my teacup and other things, but from the distinctive 'kinds of things' they are. I should have done so, however, by having used the word 'cookie' in conformity with the 'grammar' (or 'rule') governing in common the ordinary uses of it within my linguistic community, rather than by having referred to something 'within' the cookie common to all cookies.

Hacker, unfortunately, augmented the phrase 'the shadows cast by grammar' by inserting the word 'merely' before it, [35] showing that he had misunderstood how essentially Wittgenstein believed us to be constrained by the 'grammar' (or 'rule') of a word when using it, contrary to the plain sense of W45:371 and 373. To Wittgenstein, the 'shadow cast by the grammar' of the word 'cookie' constrains in common every user of the word within my linguistic community, transcending every instance of its use. We are thus bound as essentially by the 'grammar' (or 'rule') governing the use of the word 'cookie' as we should have been had there been something common to all cookies; for, were we constrained less essentially, we should be unable, when using the word 'cookie' to refer to one of them on some occasion, to show 'what kind of object' the cookie is. Wittgenstein, contra Hacker, had redirected rather than abandoned the 'essential', switching its locus from the cookie to the cookie cutter – the 'grammar' (or 'rule') determining how we must use the word 'cookie' to attend to one or another of the cookies of the world.

Thirty years before, Wittgenstein had insisted that we could name, but could not express, the 'objects' that he would soon thereafter suggest must comprise the 'substance' that, by transcending the world, determines how it is 'formed'. [36] Now, a third of a century later, he had disposed of those 'objects', but only by having come to believe that every instance of the use of a word had to conform to a transcending 'grammar' (or 'rule') that we could express, but of which we could not speak. A heavenly host of 'essences' of which we could only 'speak', determining supposedly how the world could be, had been replaced by others that we could only 'express', determining supposedly how we could speak usefully within it.

No wonder the writings of Wittgenstein, Austin and others, elucidating supposedly by example 'what we should say when', describe so seldom how anyone actually spoke on a specific occasion when trying to solve a problem unique to that time and place. The so-called 'examples' are hardly descriptions of 'use', but rather 'schema', or 'models' to which genuine instances of 'use' are presumed to conform. As anyone would know, however, who has ever tried to track exactly how different people have spoken when speaking 'ordinarily', the differences defy abstraction. No wonder biographical studies have seemed as irrelevant to these 'investigations' as Collingwood's work seemed to philosophers of his day, for to search for the 'essential', one must divert one's eyes from the world.

I can understand, therefore, how Wittgenstein, unaccustomed to registering texts and their makers historically, and searching throughout his life for a "perspicuous representation" (übersichtliche Darstellung) of how we are constrained when speaking usefully, some kind of 'form' of the 'given' transcending sub specie aeternitatis the untidy details that have distinguished human beings and their artefacts from one another, should have failed to register the philosophical significance of approaching biographically what he and others had said and done. [37]

I remain astonished, however, that Collingwood, among the most historically aware and articulate philosophers ever to work in English, should have failed to do so. Collingwood had long insisted that a philosophical text could only be understood historically as the intended solution to a problem peculiar to its time and place. There are no 'permanent philosophical problems' transcending philosophers or their world.

If there were a permanent problem P, we could ask 'what did Kant, or Leibniz, or Berkeley, think about P?' and if that question could be answered, we could then go on to ask 'was Kant, or Leibniz, or Berkeley, right in what he thought about P?' But what is thought to be a permanent problem P is really a number of transitory problems p1p2p3 … whose individual peculiarities are blurred by the historical myopia of the person who lumps them together under the one name P. It follows that we cannot fish the problem P out of the hyperuranian lucky-bag, hold it up, and say 'what did So-and-so think about this?' We have to begin, as poor devils of historians begin, from the other end. We have to study documents and interpret them. (C38:69)

To understand the problem that a philosophical text was intended to solve, we must place it within a historical sequence of problems, each having arisen as a transfiguration of its predecessors but bearing only a 'family resemblance' to them.

But then, as Collingwood might well have recognised, having chosen to write his Autobiography when he did, and for the reasons he wrote it, if we are to come to understand how a text written by a philosopher was intended to be used, measuring what it was intended to mean as a transfiguration of the problems addressed in texts that preceded it, we must begin by placing it within the sequence of problems that the philosopher had previously confronted in his own works. The problems that Kant addressed by writing the second edition of his first critique, for example, published in 1787, differ importantly, and consequently, from those that he had tried to solve by writing its first edition of 1781, and these in turn from those that he had been trying to solve when writing other texts in 1768 and before; and we can come to understand any one of them only by registering biographically how Kant came to write it as a consequence of what he had written before.

What, then, prevented Collingwood from broadening the precepts and examples of his Autobiography to encompass his own writing of it? As with Wittgenstein, a pervasive essentialism.

Collingwood had claimed unexceptionally that we could come to understand a text by Aristotle, or by Kant, Leibniz or Berkeley, only by coming to emulate how the author had thought when creating the text in response to the problem he faced. By saying so, however, he meant something other than that we must try to think as the author had thought. Rather, he insisted,

… the historian of a certain thought must think for himself the very same thought, not another like it. If some one, hereinafter called the mathematician, has written that twice two is four, and if some one else, hereinafter called the historian, wants to know what he was thinking when he made those marks on paper, the historian will never be able to answer this question unless he is mathematician enough to think exactly what the mathematician thought, and expressed by writing that twice two are four. (C38:111; italics added)

For Collingwood, the world encompassed not only thinking authors and their texts, but the 'thoughts' that they had been thinking when writing them; and only by thinking those 'very same thoughts' could we come to know of the problem they faced and thus understand the texts they had created to solve them. "Historical knowledge", he insisted, "is the re-enactment in the historian's mind of the thought whose history he is studying" (C38:112; italics added), for were there no 'thoughts' common to the present and the past, we should be unable to encounter and 'know' anything within the present that anyone had encountered and 'known' in the past.

Collingwood always resisted the charge, laid recurringly against him, that he was an 'idealist' trotting after Hegel, and his concerns were importantly different. Nevertheless, the compulsion to believe that there had to be 'thoughts' transcending the time and place of their expression by thinking authors was to prompt Collingwood to suggest, as well, that philosophers like Aristotle, Descartes and Kant had by articulating their 'thoughts' been able to express the 'absolute presuppositions' of their age. Not only were there 'thoughts' as well as thinkers within history, but some thinkers have been able to articulate the 'thoughts' presupposed by all thinkers at that time and place. [38]

No wonder Collingwood, whenever commenting upon the work of Kant, for example, failed to attend to the complexities of how Kant had managed over years of struggle to transfigure the problems and solutions of his earlier works into the text of his first critique. Unable to forsake the essentialist notion that there must be a 'thought' common to Kant's thinking and our own, were we to come to understand him, Collingwood had taken a shortcut, reconstructing supposedly Kant's 'thought' but only by bypassing the tough task of coming to imagine how Kant's 'thinking' had evolved step-by-step, decisive for understanding it.

Collingwood, although committed ostensibly to the historically particular, had remained in the grip of a Platonic compulsion comparable to that from which Wittgenstein was never able to free himself. Both men had fought hard against the prevailing essentialisms of their day, each with weapons of uncommon power, only to fall prey to others that they had failed to recognise. I respect them for having done what they did, when they did it, and owe each of them more than I can say or show. I can only wonder, given their example, what pretensions lurk unseen in the unexplored shadows of what I have written here.

Nevertheless, I reaffirm the moral of the story that I have been telling of them, namely that there can be no shortcuts to understanding, for understanding is historical, and at root biographical, and nothing transcends the historical. When Whitehead observed that "the safest general characterization of the European philosophical tradition is that it consists of a series of footnotes to Plato", [39] he was reminding us that philosophers have invariably presupposed, though each in their own way, that we can come to understand the things we encounter, and learn better how to say, do and make newer ones, only by presupposing that some other things transcend them. But nothing transcends them, neither Plato's 'forms', nor the 'elementary sentences' and 'objects' of Wittgenstein's first book, nor the 'grammars' or 'rules' of his second book, nor even the 'thoughts' of Collingwood's Autobiography; for however things may be, they can be no more to us than whatever we shall make of them, and whatever we shall make of them, having been made at some time and place, will be historical, and having been made by us, will be at root biographical.

Learning to think better philosophically is therefore as arduous a task of bringing artefacts of the past to bear historically upon the present, and biographically so, as learning to think better of how to play the piano, compose, paint, lawn bowl, write screenplays, design experiments, reconstruct hypotheses or prove theorems – a hard lesson worth learning, and especially so for philosophers. To learn it we must humble ourselves, lower our gaze and look at the trees rather than trying, godlike, to acquire a "perspicuous representation" of the wood. For as Collingwood reminded us,

… if anybody had objected that in what I call 'open' history one couldn't see the wood for the trees, I should have answered, who wants to? A tree is a thing to look at; but a wood is not a thing to look at, it is a thing to live in. (C38:76)


Wittgenstein on the 'Objects' of the Tractatus

1.  The world is all that is the case.

1.1  The world is the totality of facts, not of things.

2.0123 If I know an object, I also know all its possible occurrences in states of affairs. (Every one of these possibilities must be part of the nature of the object.) A new possibility cannot be discovered later.

2.0124 If all objects are given, then at the same time all possible states of affairs are also given.

2.014  Objects contain the possibility of all situations.

2.02 Objects are simple.

2.021 Objects make up the substance of the world. That is why they cannot be composite.

2.0211 If the world had no substance, then whether a proposition had sense would depend on whether another proposition was true.

2.0212 In that case we could not sketch out any picture of the world (true or false).

2.022 It is obvious that an imagined world, however different it may be from the real one, must have something – a form – in common with it.

2.023 Objects are just what constitute this unalterable form.

2.0231 The substance of the world can only determine a form, and not any material properties. For it is only by means of propositions that material properties are represented – only by the configuration of objects that they are produced.

2.024 Substance is what subsists independently of what is the case.

2.025 It is form and content.

2.026 There must be objects, if the world is to have an unalterable form.

2.027 Objects, the unalterable, and the subsistent are one and the same.

2.0271 Objects are what is unalterable and subsistent; their configuration is what is changing and unstable.

2.0272 The configuration of objects produces states of affairs.

2.04 The totality of existing states of affairs is the world.

2.06 The existence and non-existence of states of affairs is reality. (We call the existence of states of affairs a positive fact, and their non-existence a negative fact.)

2.063 The sum-total of reality is the world.

Ludwig Wittgenstein, Tractatus Logico-Philosophicus, vis-à-vis translation by D. F. Pears and B. F. McGuinness (London: Routledge & Kegan Paul, 1961 [1921]).

[1] The gist of this essay was presented on 30 May 2003 to the Annual Meeting of the Canadian Society for Aesthetics in Halifax, Nova Scotia, as part of a panel organized by Béla Szabados on 'Wittgenstein and the Arts'. I thank him, the other participants, Gary Hagberg and Claude Savary, and the members of the audience for responding generously to it.

[2] Ludwig Wittgenstein, Philosophical Investigations, vis-à-vis translation by G. E. M. Anscombe, edited by Anscombe and Rush Rhees (Oxford: Basil Blackwell, 1963 [1953]), 66 and 71. Since Part I, the only section of the book that Wittgenstein intended to be published as it appears, was completed by 1945 (see note 14 below), I shall refer to the book hereafter as W45. I shall also use Wittgenstein's numbering of the items I cite, rather than the page numbers of the text.

[3] Robin G. Collingwood, Religion and Philosophy (Bristol: Thoemmes Press, 1997 [1916]), 53.

[4] I shall be concerned in this essay, as in life itself, with the problem of how better to say, do and make things for 'readers' of good will who are willing to learn from whatever their 'authors' intended them to mean. My disdain for those who would wilfully misconstrue (read: deconstruct) the intended meaning of a work created by another human being, misusing it and its author to contrary ends of their own, is almost boundless.

[5] Ludwig Wittgenstein, Tractatus Logico-Philosophicus, vis-à-vis translation by David F. Pears and Brian F. McGuinness (London: Routledge & Kegan Paul, 1961 [1921]). Since the work was completed in August 1918, but published only in 1921 in the final volume of Wilhelm Ostwald's Annalen der Naturphilosophie, I shall hereafter refer to it as W18, and shall use Wittgenstein's numbering of the items I cite, rather than the page numbers of the text.

[6] I shall use single quotation marks throughout this essay to signal that the word or phrase enclosed is worthy of special attention, sometimes because I am referring to the word or phrase itself, but more often because it resonates with other uses of it, or of other words or phrases, elsewhere in the essay, is a near-but-not-exact quotation or is simply being used atypically.

[7] I overstate in both directions. Some 'arguments', or pretenders to them, may be found within the text (see Table, 2.021–2.0211), and, as Brian McGuinness has remarked, "no one has yet completely elucidated" the supposed logical arrangement of the paragraphs of the book patterned after the numbering system of Russell and Whitehead's Principia Mathematica (Wittgenstein: A Life. Young Ludwig 1889–1921 [London: Gerald Duckworth & Co., Ltd., 1988], 300). (Hereafter BM88) The notebook that we now call the Prototractatus contains an early draft of much of Wittgenstein's text replete with paragraphs comparably numbered, so the book may well have originated, as McGuinness suggests, as an attempt by Wittgenstein to comply early on with a request by Russell that he rewrite the opening chapters of Principia Mathematica (see BM88:264 and 265).

[8] W18:5 (Preface, dated "Vienna, 1918"). As McGuinness notes (BM88:302), Wittgenstein was here echoing Gottlob Frege's remark in his Introduction to The Foundations of Arithmetic, translated by John L. Austin, revised edition (New York: Harper & Brothers), xviii: "I hope to settle the question finally, at least in its essentials". The arrogance of Wittgenstein's claim was mollified somewhat by the sentence that followed it, "And if I am not mistaken in this belief, then the second thing in which the value of this work consists is that it shows how little is achieved when the problems are solved", but only, as McGuinness says, by heightening the oddity of a book "half of whose value consists in showing its own unimportance". (BM88:302)

[9] "Grave mistakes" is the phrase Wittgenstein used a quarter-century later when assessing his first book within the Preface, dated "January 1945", to his second book, the Philosophical Investigations. See note 15 below for the full quotation.

[10] John L. Austin, "A Plea for Excuses", reprinted from the Proceedings of the Aristotelian Society, 1956-57, as Chapter 6 of Austin's Philosophical Papers (Oxford: Clarendon Press, 1961), 123-52, at 129.

[11] Wittgenstein: "Grammar tells us what kind of object anything is". (W45:373) See the Conclusion of this essay for commentary upon this sentence.

[12] Robin G. Collingwood, An Autobiography (Oxford: Oxford University Press, 1939), 83. Since the work was completed in 1938, I shall refer to it hereafter as C38.

[13] What scholars would give, for example, for a comparable volume by Leibniz, or by the elderly Kant.

[14] I condense here a discussion extending before, over and after pages 128 to 130 of Collingwood's account.

[15] A footnote to this paragraph, appended by the editors of the volume, reads: "It was hoped to carry out this plan in a purely German edition of the present work". The hope was never fulfilled. Worse, the text that the editors published as Wittgenstein's Philosophical Investigations encompasses not only Part I, which he had completed as it appears by 1945 but of which the editors remark that, had he "published his work himself", he would have "suppressed a good deal of what is in the last thirty pages or so", but they included within a Part II material written by Wittgenstein between 1947 and 1949 that he never intended to be published as it appeared, if published at all, suggesting that he would have "worked what is in Part II, with further material, into [the] place" of the material that he would have 'suppressed'. However one may sympathise with the difficulties faced by the editors, the fact remains that Part I is the only material in the book that we can trust to have been put forward for publication by Wittgenstein as it appears. With the exception of the final quotation in note 36, therefore, I shall restrict my citations to it.

[16] "For since beginning to occupy myself with philosophy again, sixteen years ago, I have been forced to recognise grave mistakes in what I wrote in that first book. I was helped to realise these mistakes – to a degree which I myself am hardly able to estimate – by the criticism that my ideas encountered from Frank Ramsey, with whom I discussed them in innumerable conversations during the last two years of his life. Even more than to this – always certain and forcible – criticism, I am indebted to that which a teacher of this university, Mr. Sraffa, for many years unceasingly practised on my thoughts. I am indebted to this stimulus for the most consequential ideas of this book." (W45:xe)

[17] "… we give no answer to philosophical questions and instead reject all philosophical questions, whether of Metaphysics, Ethics or Epistemology". Rudolf Carnap, The Unity of Science (London: Kegan Paul, Trench, Trubner & Co., 1934), 21, a translation, with introduction, by Max Black of Carnap's "Die physikalische Sprache als Universalsprache der Wissenschaft", Erkenntnis 2, nos. 5/6 (1932): 423-65. Schlick and Wittgenstein, on the other hand, were to converse spontaneously and respectfully about aspects of the work of Heidegger, Husserl and Kierkegaard. See the transcriptions by Friedrich Waismann of the conversations between Wittgenstein, Schlick and himself, later edited by Brian F. McGuinness and published in translation by Joachim Schulte and McGuinness, with appendices of material by Waismann, under the title Wittgenstein and the Vienna Circle: Conversations Recorded by Friedrich Waismann (Oxford: Basil Blackwell, 1979), 67f. and 78f., 22 and 29 December 1929, and 2 January 1930. (Hereafter FW29-32)

[18] Moritz Schlick, "The Future of Philosophy", College of the Pacific Publications in Philosophy I (Stockton, California, 1932), reprinted in Schlick's Philosophical Papers: Vol. II (1925-1936), edited by Henk L. Mulder and Barbara F. B. Van de Velde-Schlick (Dordrecht, Holland: D. Reidel Publishing Company, 1979), 210-24, at 222. (Hereafter S31) Schlick is paraphrasing a part of W18:4.112.

[19] "Philosophy: Sections 86-93 of the so-called 'Big Typescript' (Catalogue Number TS 213)", trans­lated and incorporated within the anthol­ogy of Wittgen­stein's writings entitled Ludwig Wittgen­stein: Philosophical Occa­sions, edited by James C. Klagge and Alfred Nordmann (Indianapolis: Hackett, 1993), 199. Georg Henrik von Wright, on page 490 of the appendices to this volume (hereafter WW:69, a revised and expanded reprint of his "Special Supplement: The Wittgenstein Papers", Philosophical Review 78, no. 4 [October 1969]: 483-503), dates the work as "probably 1933", so I shall refer to it hereafter as W33.

[20] I use here the translation as given in the original English edition of 1921, attributed to Charles K. Ogden but probably by Frank P. Ramsey, whom Braithwaite  remembered as "sailing into Miss Pate's typing office in Cambridge and dictating the translation with Ogden at his side". This description of Braithwaite's recollection is McGuinness's, who likewise cites Ramsey's translation, rather than his own, "to give the effect" of the "mystical adjuration to silence in the face of the ineffable, as it were a form of negative theology" with which the book ends. (BM88:300)

[21] Henry Sheffer had shown in 1913 that any proposition, compounded logically of others and representable by means of the now-customary signs for logical connection ('and', 'or', 'not', 'only if', etc.), could be represented equivalently by a proposition using only a single connective '|', now commonly referred to as the 'Sheffer Stroke function', standing for 'Neither … nor …' (or alternatively 'Not both … and …'). (Henry Maurice Sheffer, "A Set of Five Independent Postulates for Boolean Algebras, with Applications to Logical Constants", Transactions of the American Mathematical Society 14 [1913], 481-88.) As Wittgenstein notes in 5.1311, the "inner connection" between the propositions of an inference "becomes obvious" only when using Sheffer's function (though he never mentions him by name in the text).

   Charles S. Pierce had discovered the same function thirty years before, but, as usual, no one paid attention until long after Sheffer rediscovered it. See his paper of 1880, entitled by the editors "A Boolean Algebra With One Constant", reproduced as sections 12-20 of the fourth volume of The Collected Papers of C. S. Peirce, edited by Charles Hartshorne, Paul Weiss and Arthur Burks (Cambridge, Mass.: Harvard University Press, 1931-53).

[22] Thoralf Skolem, "Begründung der elementaren Arithmetik durch die rekurrierende Denkweise ohne Anwendung scheinbarer Veränderlichen mit unendlichem Ausdehnungsbereich", Bidenskapsselskapets skrifter, I. Matematisk-naturvidenskabelig klasse, no. 6 (1923) (written Autumn, 1919), reprinted in translation by Stefan Bauer-Mengelberg as "The Foundations of Elementary Arithmetic Established by Means of the Recursive Mode of Thought, Without the Use of Apparent Variables Ranging over Infinite Domains", in From Frege to Gödel: A Source Book in Mathematical Logic, 1879-1931, edited by Jean van Heijenoort (Cambridge, Mass.: Harvard University Press, 1967), 302-33.

[23] Wittgenstein was to continue to believe so, though with increasing difficulty, into the early 1930s, as his conversations with Schlick and Waismann confirm (see FW29-32:47f., 79, 97f. and 126). I shall conclude this essay by condemning Wittgenstein's later use of the words 'grammar' and 'rule', but will avoid the issue of whether his use of the word 'criterion' trades upon the notion of 'verification'. Schlick, I think, would assuredly have thought so.

[24] "What is the difference between Tatsache and Sachverhalt? Sachverhalt is, what corresponds to an Elementarsatz if it is true. Tatsache is what corresponds to the logical product of elementary [propositions] when this product is true. The reason why I introduce Tatsache before introducing Sachverhalt would want a long explanation." (W1914-17:129, from a letter to Russell of 19 August 1919)

[25] Whenever Wittgenstein remarks that 'It is obvious … ', one can quite safely conclude the opposite. The phrase should be understood to mean rather that 'I have come to believe what I am about to say, and have reasons for doing so that I shall not bother to tell you here, since it would be wasting my time', echoing the sentiment Wittgenstein so often expressed in his letters to Russell, who found unobvious, and often rightly so, much of what Wittgenstein had said. For example, "An account of general indefinables? Oh Lord! It is too boring!!! Some other time! – Honestly – I will write to you about it sometime, if by that time you have not found out all about it (because it is all quite clear in the manuscript I think)"; or again, "Please think the matter over for yourself, I find it awful to repeat a written explanation, which I gave the first time with the greatest reluctance." (W1914-17:124 and 128, excerpts from two letters of late 1913)

[26] See note 29 for Schlick's concern about the seemingly synthetic a priori nature of Wittgenstein's claims here and elsewhere in the book.

[27] Wittgenstein had struggled long and hard to arrive at this conclusion, as his notebooks attest, for he, too, had had to overcome the assumption that the 'simple things' to which the simplest propositions referred could in any way be 'material' or, like 'the points of the visual field', phenomenological. (W14-17:16ef. and 43ef.) His circuitous journey to the conclusion of the Tractatus that the 'objects' named within 'elementary propositions' could only be 'logical', passing by way of the puzzles of negation through the interim insights that a "proposition is a logical picture of a situation" (25e), that a "name" might be a "logical concept" (52e), that "relations and properties, etc., are objects, too" (61e), that "all objects" must "in a certain sense be simple objects" (61e) and that "the demand for simple things is the demand for definiteness of sense" (63e), is among the most philosophically illuminating stories of which I am aware. See, as well, his later comment of 29 December 1929 to Schlick and Waismann, contrasting how Frege and Russell had "spoke[n] of objects". (FW29-32:41 and 42)

[28] Ludwig Wittgenstein, TS 220, section 108, an unpublished typescript on deposit in the Bodlian Library, dated "1937 or 1938" by von Wright (VW69:490 and 499). I am indebted to Dale Lindskog for drawing my attention to this remark in his dissertation "Diagnosis and Dissolution: From Augustine's Picture to Wittgenstein's Picture Theory" (York University, Department of Philosophy, 1999), 201, from which I have quoted it.

   Note that Wittgenstein's remark answers a question that has provoked and misled even careful commentators, namely whether Wittgenstein intended 'relations' to be included among the 'objects' of the Tractatus. Wittgenstein had earlier said so in an entry in his notebooks of 16 June 1915, remarking that "relations and properties, etc. are objects too" (W14-17:61e), but had failed to repeat the claim in the Tractatus. Hence even David Keyt, one of the few commentators before McGuinness to sense and respect the Platonic ambience of the Tractatus, refused to believe that the author of the book still thought so, and gave an account of its 'objects' as excluding 'relations' that, however thoughtful, can now be discounted. See Keyt's "Wittgenstein's Notion of an Object", The Philosophical Quarterly 13, no. 50 (January, 1963): 3-13, and his "Wittgenstein's Picture Theory of Language", The Philosophical Review 73 (1964): 493-511, reprinted respectively as pages 231-48 and 377-92 of Essays on Wittgenstein's 'Tractatus', edited by Irving M. Copi and Robert W. Beard (Bristol: Thoemmes Press, 1993 [1966]), and especially the section of the latter called "Proposed Solutions", 381f.

[29] "What is your aim in philosophy? – To shew the fly the way out of the fly-bottle." (W45:309)

[30] No wonder Schlick, when questioning Wittgenstein, invited him recurringly to affirm that the remarks of the Tractatus were intended to be free of any taint of the synthetic a priori, so dreaded by the members of the Circle; and no wonder Wittgenstein evaded those invitations, for, as Schlick suspected, the seemingly synthetic claims of the book could be rendered meaningful only if acknowledged to be a priori (see FW29-32:65f. and 76f.).

[31] Peter Hacker, among others, less attuned than Wittgenstein to things mathematical, missed the conundrum about real numbers that precipitated Wittgenstein's attack upon the integrity of 'elementary propositions', claiming instead that "the colour-exclusion problem led to the idea of a Satzsystem, and the independence of the elementary propositions was accordingly relinquished." (Peter M. S. Hacker, Wittgenstein: Meaning and Mind, vol. 3 of An Analytical Commentary on the "Philosophical Investigations" [Oxford: Basil Blackwell, 1990], 548). (Hereafter H90) As we have seen, it was rather Wittgenstein's realisation that 'elementary propositions' could be no more independent than the 'real numbers' contained within them that led to the notion of a Satzsystem, and therewith to the 'solution' of the 'colour-exclusion problem'.

[32] See the opening quotation of this essay, and "... we now demonstrate a method, by examples ... There is not a philosophical method, though there are indeed methods, like different therapies." (W45:153)

[33] Wittgenstein had insisted as early as 1933 that "the work of the philosopher consists in assembling reminders for a particular purpose"; for "learning philosophy is really recollecting. We remember that we really used words in this way". (W33:173 and 179)

[34] Ludwig Wittgenstein, Philosophical Remarks, edited from his posthumous writings by Rush Rhees and translated into English by Raymond Hargreaves and Roger White (London: Basil Blackwell, 1975), 85. The editor confirms on page 347 that "All the passages in it were written in manuscript volumes between February 2nd, 1929, and the last week of April, 1930".

[35] "… what we conceive to be natures or essences are merely the shadows cast by grammar." (H90:438)

[36] "Objects I can only name. … I can only speak of them, I cannot express them." (W14-17:51e; 26, 27 May 1915) He was to reaffirm this conviction within the Tractatus at 3.221.

[37] 'Sub specie aeternitatis', Spinoza's phrase, was assimilated by Wittgenstein into his notebooks on 7 October 1916. (W14-17:83e) Readers of Wittgenstein's first and second books alone may easily be diverted from the consistency of his pursuit by his shifts of phrase. Wittgenstein had said in the Tractatus that "… the limits of language (of that language which alone I understand) mean the limits of my world”. (W18:5.62, italics original) Fifteen years later he was to shift the locus of understanding of how language delimits the world from the private 'I' to a public 'we' comprising a community of language users: "The concept of a perspicuous representation is of fundamental significance for us. It earmarks the form of account we give, the way we look at things. (A kind of 'world-view', as is apparently typical of our time. Spengler.)” (W33:175) A dozen years later, however, having recognised that no 'perspicuous representation' of how words were used within any of the language games played by a community of human beings could match the pretensions of a Spenglerian 'world-view', he was to repeat the first two sentences, but to replace the global affirmation within the parenthesis by a question, "Is this a 'Weltanschauung'?" (W45:122); for he had now learned, as he put it, that "… to imagine a language game means to imagine a form of life" (W45:19; italics added), a way of living within the world common to the members of a linguistic community, rather than an 'overview' of it. To say so, however, was but another way to reaffirm essentialism, for to 'imagine a language game', or 'form of life', was to register how the words used within the game exemplify the 'grammar' (or 'rule') of their use, transcending every instance of their use – however culturally or historically delimited the game itself may prove to be. A 'form of life', expressed however briefly by the instances of linguistic behaviour of the members of a cultural community, and transcending them however transiently, remains as essential as Plato's 'form of piety' (or, as we shall see, Collingwood's 'thoughts'). Wittgenstein was never to forsake the search for a 'perspicuous representation' of the 'accepted', the 'given' (later, the 'certain') – the 'forms' of life that constrain us linguistically; for "forms of life", as he would insist even after 1945, are "what has to be accepted, the given …" (W45, page 226e)

[38] See C38:65f. Collingwood, able to live a few years longer than he had expected, was to amplify this suggestion in Part 1, Chapters IV through VI, of An Essay on Metaphysics (Oxford: Clarendon Press, 1940), 21-57.

[39] Alfred North Whitehead, Process and Reality, Corrected Edition, edited by David Ray Griffin and Donald W. Sherburne (New York: The Free Press, 1978 [1929]), 39.

Previous article /

Article précédent

Top / Début

Table of content / Table des matières
Next article /

Article suivant