Wittgenstein Research Revisited
Conference at the HIT Centre - University of Bergen
12th - 15th of December 2001

Kristóf Nyíri:

Wittgenstein's Philosophy of Pictures

1. Introduction
2. What the printed corpus offers
    Philosophical Investigations
    Philosophical Remarks
    Philosophical Grammar
    The Blue and Brown Books
3. Using the Nachlaß: towards a re-interpretation
4. Summing up


1. Introduction

Wittgenstein's philosophy of pictures is commonly regarded as comprising two contrasting positions. The Tractatus is taken to argue for a picture theory of meaning, summed up by Wittgenstein's dictum: "The proposition is a picture of reality."(1) The later Wittgenstein is interpreted as holding a use theory of pictures, according to which pictures by themselves do not carry any meaning; they acquire meaning by being put to specific uses and by being applied in specific contexts. Those uses and contexts are defined by language; pictures are subservient to words, and indeed not even mental images mean by virtue of their resemblance to some external reality.

Now of course neither the early nor the later views of Wittgenstein on picturing are as straightforward as common opinion suggests. Recall the Tractarian notion of the abbildende Beziehung or "pictorial relationship"(2), a relationship consisting of "the correlation of the picture's elements with things"(3). This "pictorial relationship" has exactly the same function as the later concept of a "method of projection": the idea of convention is there in the Tractatus, too. Nor is the idea of resemblance missing from the Investigations.

The standard opinion did not go uncontested. In 1973 already Kenny emphasized that "the picture theory needs supplementing rather than [being] false ... the theory of meaning as use is a complement rather than a rival to the picture theory".(4) The discontinuity view, however, remains predominant. In his Picture Theory W.J.T Mitchell writes in reference to Wittgenstein of "a philosophical career that began with a 'picture theory' of meaning and ended with the appearance of a kind of iconoclasm, a critique of imagery that led him to renounce his earlier pictorialism...".(5)

Wittgenstein's later philosophy of pictures has been taken note of in the so-called imagery debate. Fodor in his 1975 The Language of Thought paraphrases insertion (b) at §139 of the Philosophical Investigations when he writes: "A picture which corresponds to a man walking up a hill forward corresponds equally, and in the same way, to a man sliding down the hill backward."(6) By omitting the second half of the passage - "Perhaps a Martian would describe the picture so. I do not need to explain why we do not describe it so" - Fodor fosters the one-sided image of a unequivocally propositionalist Wittgenstein. Fodor's interpretation is taken up by Stephen Kosslyn, for many years the main protagonist on the "images exist" side of the imagery debate, in his 1994 book Image and Brain.(7) For him, too, Wittgenstein stands for the view that pictures without a verbal interpretation cannot carry meaning.

Beyond the boundaries of the imagery debate Wittgenstein's later philosophy of pictures has not received much attention. Thus from the very extended discussions surrounding Goodman's Languages of Art(8), Wittgenstein's name is practically absent,(9) even though one of the first reviewers of the work, Richard Wollheim, did to some extent rely on Wittgenstein.(10) My suggestion is that the relative lack of interest in Wittgenstein's later philosophy of pictures is not independent of the fact that his full Nachlaß was, until the publication of the Bergen electronic edition, not actually available. The printed corpus only partially conveys the richness, complexities, continuities of, and changes in, Wittgenstein's ideas on pictorial representation.(11) And it fails to convey the significance of the later Wittgenstein's method of explaining philosophical points with the help of diagrams - his Nachlaß contains some 1300 of them. This method would have made no sense if he had really adhered to the position that images do not have a meaning unless interpreted verbally.(12)

2. What the printed corpus offers

What picture of Wittgenstein's philosophy of pictures does there emerge from his printed work? I will deal with some of the more important volumes one by one. No attempt at completeness is made, and I am taking only the most significant passages into account.

Philosophical Investigations

The picture theme makes its appearance at the very beginning of the volume - in the 1945 preface, immediately after the passage explaining that the author has at long last been forced to give up the idea of writing a proper book, the idea of composing extended texts which would progress in a linear order; and that this has to do with the nature of the task itself. Instead, he has produced an album, made up of ever new pictures of the same sites.(13) The word "picture" is a metaphor here; but the metaphor - entirely absent in the 1938 version of the preface - is quite elaborate, the author likening himself to a poor draughtsman, with references made to picture cuts and to observers of landscapes. And the difference between pictures and unidirectional texts is real enough. Wittgenstein in fact seems to suggest that the written text is an inappropriate medium in which to conduct his specific investigations. Towards the end of my talk I will formulate a hypothesis as to why this might actually be the case.

In the so-called Part I of Philosophical Investigations the insertion at §22 gives the example of a picture ("a picture representing a boxer..."(14)) that can be used to convey various meanings. Pictures, or some pictures at least, are not self-explanatory. §23 introduces the concept of language-games, relates this concept to that of a form of life, and lists a number of language-games. One of them is: "Constructing an object from a description (a drawing)." Wittgenstein does not explicitly say so here, but he clearly implies that one needs training, that one has to become acquainted with the workings of an institution, in order to be able to construct an object from a drawing. §139 introduces the notion of a method of projection. "The picture of the cube", Wittgenstein here writes, "did indeed suggest a certain use to us, but it was possible for me to use it differently." There are two insertions at §139. The first is a reminder that "pictures are often used instead of words, or to illustrate words". The second ("I see a picture; it represents an old man walking up a steep path..."(15)), referred to above, makes the point that, although pictures can indeed be misinterpreted, a great many of them are unambiguous because they are integrated in a specific way into our form of life. In §140 Wittgenstein repeats that "there are other processes, besides the one we originally thought of, which we should sometimes be prepared to call 'applying the picture of a cube'". And towards the end of §141 we read: "Can there be a collision between picture and application? There can, inasmuch as the picture makes us expect a different use, because people in general apply this picture like this." What §§139-141jointly suggest is that it is not the single picture, but rather the institution of how we use pictures that is decisive.

§291 contrasts a blueprint with those pictures "which seem simply to portray how a thing looks". Such pictures, Wittgenstein says, "are as it were idle". Pictures, just like words, can serve as instruments. §396 asserts that in order for words to be understood, no supplementary pictures - no mental images or physical drawings(16) - are needed. §§432-434 introduces the idea of gestures as pictures, and suggests that only use gives life to pictures. In §449 there occurs an intriguing passage: "We do not realize that we calculate, operate, with words, and in the course of time translate(17) them sometimes into one picture, sometimes into another." §450 relates calling up the image of someone to mimicking the person's expression. Since to mimic is to evoke a resemblance, and since the ability to mimic is in important respects more fundamental than the ability to speak, Wittgenstein here implies that some kinds of visual representation can convey meaning without relying on verbal appendage. §454 makes the point that the arrow symbol does not, by itself, point: "The arrow points only in the application that a living being makes of it."

§518 applies to the painting of pictures the Platonic riddle of the unthinkable nonexistent,(18) and concludes with the far-reaching question: "Well, tell me what the object of painting is: the painting of a man (e.g.), or the man that the picture portrays?" The issue reverberates in §520, with Wittgenstein's imaginary interlocutor saying that "a painting or relief or film ... can at any rate not set forth what is not the case". §522 introduces the distinction between a "portrait" ("a historical representation") and a "genre-picture". "When I look at a genre-picture", Wittgenstein writes, "it 'tells' me something, even though I don't believe (imagine) for a moment that the people I see in it really exist, or that there have really been people in that situation. But suppose I ask: 'What does it tell me, then?'" The answer is given in §523: "I should like to say 'What a picture tells me is itself.' That is, its telling me something consists in its own structure, in its own lines and colours." The idea that pictures need to be backed by words in order to be unequivocal is entirely missing here; it reappears in §§663 and 683.

We now come to "Part II" of Philosophical Investigations. Here Wittgenstein in section xi begins by pointing out that texts supply interpretations of illustrations, but adds that with each different interpretation we really see the illustration differently.(19) The idea of a "picture-object" is introduced. Wittgenstein gives the example of a "picture-face", and remarks: "In some respects I stand towards it as I do towards a human face. I can study its expression, can react to it as to the expression of the human face. A child can talk to picture-men or picture-animals, can treat them as it treats dolls."(20) Some pages later follows the tantalizing passage: "The only thing that is natural to us is to represent what we see three-dimensionally; special practice and training are needed for two-dimensional representation whether in drawing or in words. (The queerness of children's drawings.)"(21)

More than once Wittgenstein returns to the question of how it is possible "to see an object according to an interpretation", and more and more he tends to reject, rather than to answer, this question. There is just no "queer fact" that would stand in need of explanation.(22) There are pictures we do not interpret at all, but react to, as Wittgenstein puts it, in an immediate way. Whether we do so react, might be a question of "custom and upbringing".(23) And then there are instances in which it is more appropriate to speak of merely knowing what a picture represents rather than to speak of directly seeing. Such is the case when someone "treats the picture as a working drawing, reads it like a blueprint". Wittgenstein here remarks: "You need to think of the role which pictures such as paintings (as opposed to working drawings) have in our lives. This role is by no means a uniform one."(24)

Grappling with the problem of seeing as, Wittgenstein discusses the figure of the "double cross". This can be seen either as a white cross on a black ground or a black cross on a white ground. The two aspects of the figure could be called attention to, he writes, "simply by pointing alternately to an isolated white and an isolated black cross". And Wittgenstein adds: "One could quite well imagine this as a primitive reaction in a child even before it could talk." Here we have, then, a case where understanding a picture is entirely independent of language use.

From the Philosophical Investigations there does not emerge a unified philosophy of pictures. The remarks in "Part I" embody viewpoints markedly different from those in"Part II". This, of course, is what one would expect. Von Wright's observation, according to which "Wittgenstein's writings from 1946 onwards represent in certain ways departures in new directions",(25) applies to the remarks pertaining to pictorial meaning, too. Those in Part II tend to accept the possibility of autonomous pictorial representation; those in Part I tend to reject it. But, as must have become clear from the foregoing, neither Part II, nor, especially, Part I, do even by themselves convey a really coherent view.

Philosophical Remarks

The bulk of this book(26-27) is a more or less faithful edition of TS 209, which in turn is based on manuscripts written by Wittgenstein in 1929 and in the first half of 1930. The book contains quite a number of remarks mentioning pictures. However, the remarks are occasional; random. Wittgenstein at this time was quite clearly not concerned with working towards anything like a theory of pictorial meaning. On pp.53f. there is a passage containing two interesting remarks. First: when a child thinks, "it forms for itself pictures", pictures that are arbitrary "in so far as other pictures could have played the same role". Secondly: "Of course, the thought processes of an ordinary man consist of a medley of symbols, of which the strictly linguistic perhaps form only a small part." On p.57 Wittgenstein says that the pictorial nature of propositions becomes even clearer if one thinks of the latter as instructions for preparing models. Wittgenstein here also allows for the element of negation in the course of making models.(28)

On the next page the idea that "you 'imagine' the meaning of a word when you hear or read it" is called "a naïve conception of the meaning of the word", with Wittgenstein however adding: "Yet the naïve theory of forming-an-image can't be utterly wrong." On p.61 we read: "The agreement of a proposition with reality only resembles the agreement of a picture with what it depicts to the same extent as the agreement of a memory image with the present object."(29)

On p.63 two consecutive remarks are: "If you exclude the element of intention from language, its whole function then collapses. - What is essential to intention is the picture: the picture of what is intended."(30) On p.65 Wittgenstein writes: "How is a picture meant? The intention never resides in the picture itself, since, no matter how the picture is formed, it can always be meant in different ways." On p.73 we encounter the interesting passage: "Our ordinary language has no means for describing a particular shade of colour, such as the brown of my table. Thus it is incapable of producing a picture of this colour." On p.81 Wittgenstein introduces the film metaphor he often applies in the Philosophical Remarks. He writes of "a confusion of the time of the film strip with the time of the picture it projects".(31)

On p.115 general propositions are likened to incomplete pictures, and on the next page the possibility of expressing negation by means of incomplete pictures is explored. These are momentous ideas, but Wittgenstein does not further pursue them. A passage on pp.118f. deals with ways the interpretation of images depends on methods of projection. On p.133 we read: "Of what 3 strokes are a picture, of that they can be used as a picture." Insofar as the topic of pictures is concerned, there now follows a long silence in the Philosophical Remarks. On p.272 we encounter the interesting passage: "In films, when a memory or dream is to be represented, the pictures are given a bluish tint. But memory images have no bluish tint, and so the bluish projections are not visually accurate pictures of the dream [nicht korrekte anschauliche Bilder der Träume], but pictures in a sense which is not immediately visual." On p.284 Wittgenstein says: "You could obviously explain an hypothesis by means of pictures. I mean, you could, e.g., explain the hypothesis, 'There is a book lying here', with pictures showing the book in plan, elevation and various cross-sections." Finally, on p.293 we encounter the remark: "A Galtonian photograph is the picture of a probability."

Philosophical Grammar

In contrast to the Philosophical Remarks, the volume entitled Philosophical Grammar contains extended and focussed passages pertaining to problems of pictorial meaning. As a starting point here let me however select a remark which, ostensibly, is not about pictures at all. It is printed on p.42: "How curious: we should like to explain the understanding of a gesture as a translation into words, and the understanding of words as a translation into gestures. - And indeed we really do explain words by a gesture, and a gesture by words."(32) The language of gestures - a pre-verbal, visual language - appears to possess a certain autonomy.

On p.102 we encounter this crucial set of remarks:

"That's him" (this picture represents him) - that contains the whole problem of representation.

What is the criterion, how is it to be verified, that this picture is the portrait of that object, i.e. that it is meant to represent it? It is not similarity that makes the picture a portrait (it might be a striking resemblance of one person, and yet be a portrait of someone else it resembles less).

How can I know that someone means the picture as a portrait of N? - Well, perhaps because he says so, or writes it underneath.

What is the connection between the portrait of N and N himself? Perhaps, that the name written underneath is the name used to address him.

When I remember my friend and see him "in my mind's eye", what is the connection between the memory image and its subject? The likeness between them?

Well, the image, qua picture, can't do more than resemble him.

The image of him is an unpainted portrait.

In the case of the image too, I have to write his name under the picture to make it the image of him.

Pictures, or at least an important class of pictures, depend on words to designate unequivocally.

On p.145 Wittgenstein writes: "Think of a sign language, an 'abstract' one, I mean one that is strange to us, in which we do not feel at home, in which, as we should say, we do not think ... , and let us imagine this language interpreted by a translation into - as we should like to say - an unambiguous picture-language, a language consisting of pictures painted in perspective." Now while in the case of the first - written - language, Wittgenstein goes on to say on p.146, it is easy to think of various interpretations, the picture language seems to be unambiguous. And this, Wittgenstein adds, "is connected with the fact that what we call a 'picture by similarity'(33) is not a picture in accordance with some established method of projection. In this case the 'likeness' between two objects means something like the possibility of mistaking one for the other." But on p.147 doubts again arise. An image, Wittgenstein here repeats, could not qualify as a portrait unless it bore the name of its subject. This does not mean that one has to imagine the subject and the name at the same time; but in some way the name does play a role. "I may go on from the picture to the name", Wittgenstein writes, "or perhaps say that I imagined N, even though at the time of the imagining there wasn't anything, except a kind of similarity, to characterize the image as N's."(34) As a remark on p.148 suggests: An isolated picture is as it were dead; it has no meaning by itself; or rather, it can be variously interpreted. It is the system of language, Wittgenstein says on p.149, in which propositions come alive; the implication seems to be that the meaningful use of pictures, too, depends on that very system.

The next sequence of passages pertaining to our topic begins on p.163 with two striking remarks: "Anything can be a picture of anything, if we extend the concept of picture sufficiently." And: "Thinking is quite comparable to the drawing of pictures." On p.164 we read: "If we compare a proposition with a picture, we must think whether we are comparing it to a portrait (a historical representation) or to a genre-picture. And both comparisons have point." On the same page there is this important remark: "for the picture to tell me something it isn't essential that words should occur to me while I look at it; because the picture should be the more direct language."(35) There then follows on p.165: "what the picture tells me is itself. - Its telling me something will consist in my recognizing in it objects in some sort of characteristic arrangement." The phrase "the picture tells me itself" repeatedly occurs on the subsequent pages.

On p.171 there is a two-paragraph passage which I quote here in full:

Let us imagine a picture story in schematic(36) pictures, and thus more like the narrative in a language than a series of realistic pictures. Using such a picture-language we might in particular e.g. keep our hold on the course of battles. (Language-game.) And a sentence of our word-language approximates to a picture in this picture language much more closely than we think.

Let us remember too that we don't have to translate such pictures
into realistic ones in order to 'understand' them, any more than we ever translate photographs or film pictures into coloured pictures, although black-and-white men or plants in reality would strike us as unspeakably strange and frightful. - Suppose we were to say at this point: "Something is a picture only in a picture-language"?
The concluding question appears to be a fair summary of the passage.

On p.176 we read about "a diagram representing the inside of a radio receiver". For someone with no knowledge of diagrams and radios this will be "a jumble of meaningless lines". For someone possessing the necessary knowledge, the drawing will be "a significant picture". Under certain conditions, then, even single pictures can serve as instruments.

On p.179 there is the remark: "Think of the multifariousness of what we call 'language'. Word-language, picture-language, gesture-language, sound-language." On p.182f. there follows a crucial passage: "If one takes it as obvious that a man takes pleasure in his own fantasies, let it be remembered that fantasy does not correspond to a painted picture, to a sculpture or a film, but to a complicated formation out of heterogeneous components - words, pictures, etc. Then one will not contrast operating with written and spoken signs with operating with 'imagination-pictures' of events. - (The ugliness of a human being can repel in a picture, in a painting, as in reality, but so it can too in a description, in words.)"(37) And finally we find this remark on p.213: "We may say: a blueprint serves as a picture of the object which the workman is to make from it. - And here we might call the way in which the workman turns such a drawing into an artefact 'the method of projection'. ... what we may call 'picture' is the blueprint plus the method of its application."

So what view of Wittgenstein's philosophy of pictures does the Philosophical Grammar offer? The idea of an institutional embeddedness of picture use is not yet prominent, but otherwise all the familiar elements of Wittgenstein's thinking on the topic are more or less present. However, these elements do not make up a coherent whole; the contradictions are not resolved. The Philosophical Grammar does certainly not encourage any systematic interpretations. And how could it? By 1976 the latest, the year the von Wright Festschrift containing Kenny's "From the Big Typescript to the Philosophical Grammar" was published,(38) it must have become clear that this volume is a mis-edited aggregate of various separate, unfinished texts; not anything Wittgenstein in this form would have, or indeed could have, put together.

The Blue and Brown Books

In the Blue Book Wittgenstein focusses on problems of meaning and intention; the few remarks on pictures arise mainly in connection with the leitmotiv "We could perfectly well, for our purposes, replace every process of imagining by a process of looking at an object or by painting, drawing or modelling".(39) On p.32 Wittgenstein puts the question "'What makes a portrait a portrait of Mr. N?'". The answer, he writes, which might first suggest itself is: "'The similarity between the portrait and Mr. N'". It is quite clear however, Wittgenstein points out, that "similarity does not constitute our idea of a portrait; for it is in the essence of this idea that it should make sense to talk of a good or a bad portrait. ... An obvious, and correct, answer to the question 'What makes a portrait the portrait of so-and-so?' is that it is the intention." To which Wittgenstein adds: "To intend a picture to be a portrait of so-and-so (on the part of the painter, e.g.) is neither a particular state of mind nor a particular mental process. But there are a great many combinations of actions and states of mind which we should call 'intending...'" For instance, the painter might have been told to paint a portrait of N. In this case it is words together with the picture that makes the portrait a portrait of N.(40)

On p.36 Wittgenstein calles attention to the possibility of "a picture which we don't interpret in order to understand it, but which we understand without interpreting it." There are, he writes, "pictures of which we should say that we interpret them, that is, translate them into a different kind of picture, in order to understand them; and pictures of which we should say that we understand them immediately, without any further interpretation." Later in this rather exceptional passage Wittgenstein acknowledges that there occur mental images making up as it were a pictorial language,(41) and continues by introducing the notion of a "picture by similarity". As he writes on p.37: "One might use for this kind of picture the word 'copy'. Roughly speaking, copies are good pictures when they can easily be mistaken for what they represent."

The focus of the Brown Book is on language games. References to pictures here enter as illustrations of specific aspects of language use. Wittgenstein mentions cases where "the pointing gesture" is "part of the practice of communication itself" (p.80) and tables "in which written signs are placed opposite to pictures of objects" (p.82), and remarks that such gestures or pictures are "elements or instruments of language" (p.84). On p.105 a pictorial language is introduced, involving two sequences of images running in parallel to each other in such a way as to provide for "a primitive kind of narration of past events". It consists of a "sun series", representing the passage of time during the day, and of "life pictures", showing the activities of a child. The two rows of pictures, when properly correlated, "tell the story of the child's day". Wittgenstein certainly implies here that this would be a feasible child language; that a child could learn to think in such a language.

Another type of autonomous pictorial communication is described on p.125: the "facial characteristics of a certain family" could be shown by "a proper arrangement" of a set of family portraits. On pp.144f. Wittgenstein demonstrates his view that mental states like believing are not invariably accompanied by characteristic feelings or experiences by asking us to consider "an analogous case drawn from facial expressions". There is, he says, "a family of friendly facial expressions". It is not the case however "that there are certain traitswhich one might call friendly traits, each of which makes the face look friendly to a certain degree, and which when present in a large number constitute the friendly expression". On the contrary, "in the wide family of friendly faces there is what one might call a main branch characterized by a certain kind of eyes, another by a certain kind of mouth, etc.; although in the large family of unfriendly faces we meet these same eyes when they don't mitigate the unfriendliness of the expression."

We have now come to the point in the Brown Book at which - as Rush Rhees registers, in his 1958 "Preface", with a for him unusual editorial precision - Wittgenstein in 1936 gave up making a German version of the text.(42)

At this point, after making a passing reference to the possibility of describing the position of an object "by words or pictures" (p.154), Wittgenstein on p.162 again touches on the issue of facial expressions. He asks us to "contemplate the expression of a face primitively drawn in this way":

We should let this face, Wittgenstein writes, "produce an impression" on us. We will then say: "Surely I don't see mere dashes. I see a face with a particular expression." And the point Wittgenstein here makes is that we cannot actually explain what this particular expression consists in. As he puts it: "'Words can't exactly describe it', one sometimes says. And yet one feels that what one calls the expression of the face is something that can be detached from the drawing of the face.(43) It is as though we could say: 'This face has a particular expression: namely this' (pointing to something). But if I had to point to anything in this place it would have to be the drawing I am looking at." One has an experience here, Wittgenstein implies, which cannot be conveyed by words; although it can be conveyed by pointing to a drawing. It appears our system of communication is incomplete, unless pictures play a part in it. This implication will be explicitly spelled out on p.174.

On p.163 Wittgenstein draws attention to two other cases where we would insist that we do not see "mere strokes" or "mere dashes". First, when we say "This is a face, and not mere strokes", distinguishing, for instance,


Secondly, the case of picture puzzles, when for instance "what at first sight appears as 'mere dashes' later appears as a face. We say in such cases: 'Now I see it as a face'." Wittgenstein stresses that this "'seeing it as a face'" does not indicate any delusions; rather, it "must be compared with seeing this drawing

either as a cube or as a plane figure consisting of a square and two rhombuses". On p.164 Wittgenstein adds: "'seeing dashes as a face' does not involve a comparison between a group of dashes and a real human face; and, on the other hand, this form of expression most strongly suggests that we are alluding to a comparison." On p.169 Wittgenstein remarks that seeing the drawing as a cube does not consist "in seeing it as a plane figure plus having an experience of depth". We are "puzzled by the three-dimensional appearance of the drawing", but this puzzlement is caused by the form of the question "'What does seeing it three-dimensionally consist in?'", for "this question really asks 'What is it that is added to simply seeing the drawing when we see it three-dimensionally?'". On p.170 Wittgenstein insists that instead of saying "I see this as a face" we should really say "'I don't see this as a face, I see it like this'". We should refrain from circumscribing verbally what we can simply point to.

On p.171 we read: "We should here ask ourselves in what sense we can call mental images pictures, for in some ways they are comparable to drawn or painted pictures, and in others not. It is, e.g., one of the essential points about the use of a 'material' picture that we say that it remains the same not only on the ground that it seems to us to be the same, that we remember that it looked before as it looks now. In fact we shall say under certain circumstances that the picture hasn't changed although it seems to have changed; and we say it hasn't changed because it has been kept in a certain way, certain influences have been kept out." The institution of referring to mental images, Wittgenstein here reminds us, is different from the institution of referring to, and dealing with, pictures.

I now come to the remark on p.174. Wittgenstein writes: "When I say 'I don't see mere dashes (a mere scribble) but a face (or word) with this particular physiognomy', I don't wish to assert any general characteristic of what I see, but to assert that I see that particular physiognomy which I do see. And it is obvious that here my expression is moving in a circle. But this is so because really the particular physiognomy which I saw ought to have entered my proposition." What ought to have entered the proposition, Wittgenstein implies, is a non-verbal, pictorial, sign. This is the conclusion towards which the train of thought in the Brown Book in fact leads. And we are now in a position to see that what Wittgenstein in the so-called Part II of the Philosophical Investigations did was to take up, again, this train of thought. When studied together with the Brown Book, Part II of the Philosophical Investigations goes a long way towards giving a picture of what Wittgenstein's philosophy of pictures might amount to. By two editorial acts however an appearance was created which made it unlikely that the reader would attempt such a study. First, people were discouraged from alloting much attention to the Brown Book, and especially to the later pages of it, by what Rhees had said in his "Preface" about Wittgenstein giving up the idea of a revision at a certain stage and calling the whole attempt worthless. Secondly, people were encouraged not to notice the tensions between TS 227 and TS 234 by the publication of them together, as a single volume - the so-called Philosophical Investigations.


The volume(44) contains quite a number of remarks on pictorial representation.(45) These remarks however, when compared with the material covered above, do not cosiderably enrich our understanding of Wittgenstein's philosophy of pictures. I'll pass them over.


Among Wittgenstein's printed volumes there are four others in which the topic of pictures plays a more or less significant role. These are the Remarks on the Philosophy of Psychology I-II,(46) and the Last Writings on the Philosophy of Psychology I-II.(47) In contrast, however, to the books discussed in the present section, these volumes are printed editions of typescripts and manuscripts which Wittgenstein could not have regarded as anything but a preliminary stock of remarks - some quite raw, some taken from older typescripts - to serve as a background for his further work. In other words, the volumes in question should be regarded as the first items of a complete Nachlaß-edition rather than as the concluding pieces of Wittgenstein's collected works. Also, the material they contain on the topic of pictures does not add anything substantial(48) to what will already be familiar to the readers of Part II of the Philosophical Investigations. Thus I will not consider them here.

3. Using the Nachlaß: towards a re-interpretation

As I tried to show in the foregoing, Wittgenstein's printed writings offer a wealth of important ideas on the social function of pictures, on pictorial meaning, and on pictorial communication. These ideas, however, do not add up to a unified philosophy of pictures. And in fact the later Wittgenstein at no stage of his thinking possessed such a unified philosophy. He had significant insights, but no clear views as to what his problems actually were, or what he was striving to achieve. Hence he often abandoned ideas his contemporary interpreters might find promising; and many ideas have never made it to the printed editions of his writings. No attempt at constructing out of his insights a genuine philosophy of pictures can, then, succeed without taking account of the entire Nachlaß.

Using the Bergen edition I will, in what follows, provide examples of what working with the Nachlaß from this perspective might amount to. I will exhibit five samples. I have labeled them PHANTASIE, ALLES KANN, PHILEBOS, SCHLINGE, and KINEMAT.


On p.30 of MS 109, on Aug. 22, 1930 Wittgenstein wrote:

Daß die Sprache ein Bild hervorbringt zeigt sich schon darin, daß
Bilder -- im gewöhnlichen Sinn des Wortes -- sich ihr natürlich

Die Illustration in einem Buch ist dem Buch nichts fremdes,
sondern gesellt sich ihm zu wie ein verwandter Behelf einem anderen,
-- wie «etwa» eine Reibahle dem Bohrer.
Wenn einen die Häßlichkeit eines Menschen abstößt so kann sie
einen im Bild (im gemalten) gleichfalls «ebenso» abstoßen, aber auch
in der Beschreibung, durch Worte «in den Worten».

On p.199 of MS 110, we find the entry, dated June 22, 1931:
Wenn man es für selbstverständlich hält daß sich der Mensch an
seiner Phantasie vergnügt so bedenke man daß diese Phantasie nicht
wie ein gemaltes Bild oder ein plastisches Modell ist sondern ein
kompliziertes Gebilde aus heterogenen Bestandteilen: Wörtern &
Bilder. Man wird dann das Operieren mit Schrift- & Lautzeichen
nicht mehr in Gegensatz stellen zu dem Operieren mit
„Vorstellungsbildern" der Ereignisse.
These two passages reappear, in reverse order, on pp.320 and 337 of TS 211 (probably 1932). It was from TS 211 Rhees edited the selection "Bemerkungen über Frazers The Golden Bough", published in the journal Synthese in 1967. The passage "Wenn man es ..." appears there on p.240.

In TS 213 (the "Big Typescript", probably 1933) the two passages, finally, come together. On p.86 we read:

Wenn man es für selbstverständlich hält, dass sich der Mensch an
seiner Phantasie vergnügt, so bedenke man, dass diese Phantasie
nicht wie ein gemaltes Bild oder ein plastisches Modell ist,
sondern ein kompliziertes Gebilde aus heterogenen Bestandteilen:
Wörtern und Bildern. Man wird dann das Operieren mit Schrift- und
Lautzeichen nicht mehr in Gegensatz stellen zu dem Operieren mit
"Vorstellungsbildern" der Ereignisse.

Die Illustration in einem Buch ist dem Buch nichts fremdes,
sondern gesellt sich hinzu wie ein verwandter Behelf einem andern,
-- wie etwa eine Reibahle dem Bohrer.

(Wenn einen die Hässlichkeit eines Menschen abstösst, so kann
sie im Bild, im gemalten, gleichfalls abstossen, aber auch in der
Beschreibung, in den Worten.)

This is the text that, one would suppose, is reproduced on pp.182f. of the Philosophical Grammar. However, the second of the three paragraphs here is missing in the printed version; and the reference to the film there is missing in the Big Typescript.

With minor modifications, the Big Typescript version reappears again, in handwriting, on pp.155f. of MS 114. It has "Wörtern, Bildern, u.a.", and the reference to the reamer is absent.

On p.66 of MS 116 we encounter - again in handwriting - a shorter variant. The text displays numerous insertions and deletions:

Wenn man es für selbstverständlich hält, daß der
Mensch sich an seiner Phantasie vergnügt, so bedenke man «möge man
bedenken», daß // wie // diese «die» Phantasie nicht wie einem «gleich
einem» gemalten«m» Bild oder ein plastisches«m» Modell ist, //
«daß diese» nicht von der Art eines gemalten
Bildes, oder plastischen Modells «gemalter Bilder» ist, // sondern
ein komplexes Gebilde «Wesen» // ein Konglomerat //
«ist» aus heterogenen Bestandteilen --
allerlei Zeichen & Bildern. Man wird dann das Operieren
«Erzählen Beschreiben» mit Schrift- & Lautzeichen nicht
«(mehr)» in «schroffem» Gegensatz stellen
«sehen», zum «zu dem» Operieren mit 'Vorstellungsbildern' (der

Die Häßlichkeit eines Menschen «‹Gesichts›» kann uns
im gemalten Bild abstoßen, aber auch in den Worten der «einer»
Beschreibung. ‹abstoßen.›

The paragraph on phantasy reappears on p.14 of "Bemerkungen I" (1945/46), and the same paragraph is printed as Zettel §652. Here, along "painted picture" and "sculpture", "film" is, finally, also listed. Of the paragraph in this form no handwritten antecedents are known.

This is a set of remarks, then, Wittgenstein again and again returned to all through the years from 1930 to 1948. The fullest version of the set is the one in the Big Typescript. That version has never been printed, although its message is momentous: mental operations involve both words and visual images; pictures are instruments of communication in the same way written texts are; and both in the pictorial and the verbal medium it is possible to preserve real-world visual information.


On pp.153-156 of MS 114, written probably around 1933-34, we can single out three mutually incompatible passages:


Alles kann ein Bild von allem sein: wenn wir den Begriff des
Bildes entsprechend ausdehnen. Und sonst müssen wir eben sagen
«erklären», was wir ein Bild von etwas nennen, & damit auch, was wir
noch die Übereinstimmung der Bildhaftigkeit, die Übereinstimmung der
Formen nennen wollen.
Das Denken ist ganz dem Zeichnen von Bildern zu vergleichen.
Wenn man es für selbstverständlich hält, daß sich der Mensch ‹an›
seiner Phantasie vergnügt, so bedenke man «man», daß diese Phantasie
nicht wie ein gemaltes Bild oder plastisches Modell ist; sondern
«einem gemalten Bild oder … ähnlich ist; sondern …» ein
kompliziertes Gebilde aus heterogenen Bestandteilen: Wörtern,
Bildern, u.a. Man wird dann das Operieren mit Schrift -- &
Lautzeichen nicht mehr in Gegensatz stellen zu dem Operieren mit
"Vorstellungsbildern" der Ereignisse.

Die Illustration in einem Buch gesellt sich zum Wort, wie ein
verwandter Behelf zu‹m›
einem andern. (Die Häßlichkeit eines Menschen kann im
Bild«im gemalten,» abstoßen, wie in der
Wirklichkeit, aber auch in der Beschreibung, in den Worten.)

Passage [C] we are already acquainted with from the sample PHANTASIE. It clearly contradicts [B]: if the mental comprises both words and and pictures, thinking cannot be quite comparable to the drawing of pictures. [C] contradicts [A]: if pictures preserve real-world visual information, than it is not the case that anything can be a picture of anything. And [B] of course contradicts [A] unless we are prepared to say that thoughts can be the thoughts of anything.

[A] expresses a verbalist or propositionalist bias; [B] represents an imagistic or pictorialist extremism; [C] formulates a comprehensive view. One might be interested to find out what the ensuing fate of these passages in Wittgenstein's notes was: "Alles kann" reappears, or appears, in TS 213 and in the first part of MS 116 (1936). "Das Denken ist ganz dem Zeichnen von Bildern zu vergleichen" does not occur again. By contrast, "Wörter, Bilder, u.a.", as we saw, is still there in the Bemerkungen I / Zettel stage (1945 or 1946 to 1948).


On July 14, 1931 Wittgenstein copied into his notebook (MS 111, p.14) a passage from Plato: "Sokrates zu Theaitetos: 'Und wer vorstellt, sollte nicht etwas vorstellen?' - Th.: 'Notwendig' - Soc.: 'Und wer etwas vorstellt, nichts Wirkliches?' - Th.: 'So scheint es.'" Some lines later he added (pp.14f.): "Man vergleiche das Vorstellen mit dem Malen eines Bildes. Er malt also ein Bild des Menschen wie dieser in Wirklichkeit nicht ist. Sehr einfach. Aber warum nennen wir es das Bild dieses Menschen? Denn, wenn es das nicht ist, ist es (ja) nicht falsch. -- Wir nennen es so, weil er selbst es drübergeschrieben hat. Also hat er nichts weiter getan, als jenes Bild zu malen & jenen Namen drüber zu schreiben. Und das tat er wohl auch in der Vorstellung." There follows a brief reference to Augustine, and then the sentence (the very first entry from July 15): "Plato nennt die Hoffnung eine Rede. (Philebos)" This is what Socrates in the Philebus says: "In jedem von uns also sind solche Reden, welche wir Hoffnungen nennen", continuing: "Und doch auch die gemalten Bilder."(49) Our feelings of hope are embodied both in inner speech and visual imagery. Plato introduces this idea some passages earlier by comparing the soul to a book, adding however that besides the "scribe" who writes "within us" there is also "another artist, who is busy at the same time in the chambers of the soul": "The painter, who, after the scribe has done his work, draws images in the soul of the things which he has described."(50)

Now while in Plato's dialogues the traces this "scribe" leaves in our minds - namely abstract notions - are amply discussed, we find there no comparable analyses of the work of the "painter", i.e. no analyses of mental pictures or visual images. Wittgenstein was certainly conscious of the fact that his later philosophy represented a markedly anti-Platonic approach;(51) what these entries in MS 111 make us realize is that rectifying Plato's one-sided handling of the topic of pictures was part of that approach.


On pp.4r-5r of MS 159 (1938) we encounter the following entry:

Die Erinnerung ist ein Bild & Worte. Es ist klar daß diese
nur in einer ganz bestimmten Umgebung bedeutungsvoll sein können.
Die Bilder können bedeutungslos, die Worte ein leerer Schall sein.

Das Symbol des gesprochenen Wortes Schriftzeichen in einer Schlinge
die aus dem Mund des Sprechers kommt.

Dies Bild erscheint uns ganz natürlich, obwohl wir doch dergleichen
nie gesehen haben.

This reference to the speech bubble, which nowhere reoccurs in the Nachlaß, alerts us to problems connected with the emergence of pictorial conventions. The speech bubble functions like a natural sign, although it is clearly conventional.(52)


On p.65r of MS 118 (1937) Wittgenstein suggests a way to prove the equation 3 + 2 = 5:

        ... dies als Beweis von 3 + 2 = 5

(Oder kinematographisch vorgeführt.)

That is, the proof would consist in drawing a series of pictures, or in the "cinematographic" presentation of the same series - an animation.

There is a related remark on p.45r of MS 122 (1939/1940):

Wenn ich sage: "der Beweis ist ein Bild" -- so kann man sich ihn
auch als kinematographisches Bild denken.
This remark was published in Part II, §23, of the Remarks on the Foundations of Mathematics (1964), with the word "auch" left out: "When I say 'a proof is a picture' -- it can be thought of as a cinematographic picture."(53)

The idea of a very different use of animations is experimented with in MS 129 (1944), p.130. As Wittgenstein here writes:

Denke Dir statt Momentaufnahmen«photographien» unserer Bekannten
benützten wir eine Art kinematographischer Bilder, die eine ganz
kleine Bewegung wiedergäben. Und das nennten wir
«bloß» ein lebendes Bildnis, im Gegensatz zu
einem toten, & faßten es nicht als Bild einer Bewegung,
«einer Lageveränderung» auf.
Practically the same passage reappears in TS 228 ("Bemerkungen I"), pp.81f. On p.9, TS 230 ("Bemerkungen II"), a sentence is added in parentheses:
Denke, statt Momentphotographieen unserer Bekannten benützten
wir eine Art kinematographischer Bilder, die eine ganz kleine
Bewegung wiedergäben. Und das nennten wir ein 'lebendes' Bildnis,
im Gegensatz zu einem 'toten', und faßten es nicht als Bild einer
Bewegung, einer Lageänderung, auf. (Das vibrierende Leben der
This is an inspired idea. As contemporary cognitive science makes us realize, mental imagery appears to be a matter of dynamic, rather than static, pictorial representations; still images are, psychologically speaking, but limiting cases of dynamic ones. With the development of twentieth-century visual culture, this seems to have become the case with regard to physical pictures, too. In my talk given at the 2000 Kirchberg symposium I found it difficult to explain that "Wittgenstein, who was a movie addict, and who regularly employed the film metaphor especially in his middle phase, did not make use of the idea of animation when discussing pictorial representation".(54) With the availability of the Bergen edition one now sees that Wittgenstein was not unaware of that idea. Indeed at one place at least he made momentous use of it. On pp.70f. of MS 145 (1933) he asks us to consider a certain proposal as to what a wish consists in, and says that the situation described is not satisfactory, since it is not embedded in a proper context; that situation, or its elements, somehow stand isolated.
Wir würden sagen: weil das in dieser
Isolierung kein Wunsch //noch nicht der Wunsch// ist. Wir müssen nun
uns jenes «das beschriebene» Bild als Wunsch zu denken es uns
«im Zusammenhang» mit
Bildern zusammen «im Zusammenhang» denken. Wir möchten sagen: Ja,
das ist das Bild eines Wunsches.
Wie ist es wenn wir uns statt dieses ruhenden Bildes eine in der
Zeit ablaufende Handlung, ein Durchlaufen von Situationen «eines
Wechsels der Situation», denken? Fühlen wir uns dann von diesem
kinematographischen Bild noch immer unbefriedigt? Ich meine würde
es uns seltsam vorkommen wenn man sagte das sei der Wunsch? Würde
man noch immer sagen: "so kann man wünschen"? Man möchte sagen der
Vorgang zeigt klarer als die ruhende Situation was der Wunsch
This is a stunning passage. What Wittgenstein here in fact suggests is that writing a caption under a picture is not the only way to disambiguate it; turning a static picture into an animated one might solve the problem, too. However, Wittgenstein never returned to this suggestion,(55) and the idea does certainly not surface in his printed writings. It is an idea his embittered adversary H.H. Price brilliantly elaborated in his1953 book Thinking and Experience.(56)

4. Summing up

In a series of papers since 1989 I have undertaken to show that Wittgenstein's later work can be usefully interpreted as a philosophy of post-literacy, and that his frequent references to Plato - the first and foremost philosopher of literacy - should be explained as attempts to arrive back at the juncture where Plato took the wrong turn.(57) Throughout its history Western philosophy reflected the influence of linear written language;(58) Wittgenstein was trying to liberate himself from that influence precisely at a time when post-literary modes of communication began to transform the civilization of the West. Written language as a source of philosophical confusion was Wittgenstein's real foe. He was not clearly aware of this, perhaps since his insights were made possible, to some extent at least, by an impairment: dyslexia.(59) He was striving to overcome the pitfalls of written language by elaborating a philosophy of spoken - oral - language. And he attempted to overcome the barriers of verbal language by working towards a philosophy of pictures. It is this latter dimension in Wittgenstein's thinking I hope to have directed attention to in the present paper.(60)


1. Tractatus Logico-Philosophicus, 4.01, Ogden transl. As Wittgenstein then goes on to explain: "In order to understand the essence of the proposition, consider hieroglyphic writing, which pictures the facts it describes. - And alphabetic script developed out of it without losing what was essential to depiction. - This we see from the fact that we understand the sense of the propositional sign, without having had it explained to us." (TLP, 4.016, 4.02, the sentence "And alphabetic script..." rendered in the Pears-McGuinness transl.)

2. Pears-McGuinness translation. Ogden has "representing relation".

3. TLP, 2.1514.

4. Anthony Kenny, Wittgenstein, Penguin Books, 1973, p.226. - Recent contributions arguing against the standard view are Anat Biletzki and David Berlin, "The Logic of Making Pictures", in: R. Casati - G. White, eds., Philosophy and the Cognitive Sciences, Kirchberg am Wechsel: ÖLWG, 1993, pp.47-50; Judith Genova, "Wittgenstein on Thinking: Words or Pictures?", in: R. Casati - G. White, eds., Philosophy and the Cognitive Sciences, pp.163-167; Judith Genova, Wittgenstein: A Way of Seeing, London: Routledge, 1995.

5. Chicago: The University of Chicago Press, 1994, p.12.

6. Jerry A. Fodor, "Imagistic Representation", in Ned Block, ed., Imagery, Cambridge, Mass.: The MIT Press, 1981, p.68. This text in Block, ed., is taken from Fodor's The Language of Thought (1975). The remark appears again, quite disfigured by then, in Zenon W. Pylyshyn's Computation and Cognition: Towards a Foundation for Cognitive Science: "As Wittgenstein points out, the image of a man walking up a hill may look exactly like the image of a man walking backward down a hill; yet, if they were my images, there would be no question of their being indeterminate - I would know what they represented" (Cambridge, Mass.: The MIT Press, 1984, p.41).

7. Stephen M. Kosslyn, Image and Brain: The Resolution of the Imagery Debate, Cambridge, Mass.: The MIT Press, 1994.

8. Nelson Goodman, Languages of Art, Indianapolis: Bobbs-Merrill, 1968.

9. Compare my "The Picture Theory of Reason", in: Berit Brogaard - Barry Smith, eds., Rationality and Irrationality, Wien: öbv-hpt, 2001, pp.242-266.

10. The review, published in The Journal of Philosophy in 1970, utilized arguments previously formulated in the reviewer's book Art and Its Objects (New York: Harper & Row, 1968), the book itself heavily relying on both Part I and Part II of the Philosophical Investigations. Compare also Wollheim's lecture On Drawing an Object (London: H.K. Lewis, 1965).

11. As Hintikka has put it:"discussions of whether Wittgenstein 'gave up the picture theory ' in his later philosophy offer an instructive example of the confusion one inevitably runs into if one does not distinguish the different components of the syndrome that usually goes by the name 'Wittgenstein's picture theory'." (Jaakko Hintikka, "An Anatomy of Wittgenstein's Picture Theory", 1994, here quoted from Hintikka, Ludwig Wittgenstein: Half-Truths and One-and-a-Half-Truths, Dordrecht: Kluwer, 1996, p.21.)

12. This is the point Andreas Roser makes in his important paper "Gibt es autonome Bilder? Bemerkungen zum grafischen Werk Otto Neuraths und Ludwig Wittgensteins", Grazer Philosophische Studien 1996/97. An earlier version of Roser's paper was read at the conference Wittgenstein y el Circulo de Viena, organized by the Universidad de Castilla-La Mancha with the collaboration of the Forschungsstelle und Dokumentationszentrum für Österreichische Philosophie, at Toledo, November 3-5, 1995. Roser's main argument, very briefly, is that one could not speak of different applications of the same picture if one did not distinguish between the picture and its application.

13. The word "Bild" occurs twice in the paragraph examined, but in the English translation only the second occurrence is translated as "picture". The first occurrence, "immer neue Bilder entworfen", is rendered as "new sketches made".

14. The remark first occurs on p.29 of MS 113. It was written in February 1932.

15. The text of this insertion first occurs on p.175 of MS 129. It was written late 1944 or early 1945.

16. The English translation here has "sketch" for "Zeichnung".

17. The German word is "überführt".

18. "Socrates toTheaetetus: 'And if someone thinks mustn't he think something? - Th.: 'Yes, he must.' - Soc.: 'And if he thinks something, mustn't it be something real?' - Th.: 'Apparently.'" To which Wittgenstein adds: "And mustn't someone who is painting be painting something - and someone who is painting something be painting something real?" (Anscombe's translation ends with an exclamation mark here, but Wittgenstein's text in the original has the question mark.)

19. Philosophical Investigations, Oxford: Basil Blackwell, 1963, p.193.

20. Ibid., p.194.

21. Ibid., p.198. - In section xii, p.230, Wittgenstein asks: "is even our style of painting arbitrary? Can we choose one at pleasure? (The Egyptian for instance.)"

22. Cf. e.g. ibid., p.200.

23. Ibid., p.201.

24. Ibid., pp.204f. Wittgenstein then adds: "we regard the photograph, the picture on our wall, as the object itself (the man, landscape, and so on) depicted there. This need not have been so. We could easily imagine people who did not have this relation to such pictures. Who, for example, would be repelled by photographs, because a face without colour and even perhaps a face reduced in scale struck them as inhuman." - On p.213 we read: "If you look at a photograph of people, houses and trees, you do not feel the lack of the third dimension in it. We should not find it easy to describe a photograph as a collection of colour-patches on a flat surface; but what we see in a stereoscope looks three-dimensional in a different way again."

25. G.H. von Wright, "The Origin and Composition of the Philosophical Investigations" (1979), here quoted from the revised version in von Wright, Wittgenstein, Oxford: Basil Blackwell, 1982, p.136. - I would like to avail myself of this opportunity to express my heartfelt gratitude to von Wright for the support and friendship he has extended to me since the early 1970s.

26. Ludwig Wittgenstein, Philosophische Bemerkungen, as vol.2 of Schriften, Frankfurt am Main: Suhrkamp.

27. Chicago: The University of Chicago Press. Translated by Raymond Hargreaves and Roger White. The English and the German page numberings are identical.

28. For example, "someone might show his understanding of the proposition 'The book is not red' by throwing away the red when preparing a model".

29. On p.82 there is a remark: "It's clear of course that speaking of memory as a picture is only a metaphor; just as the way of speaking of images as 'pictures of objects in our minds' (or some such phrase) is a metaphor. We know what a picture is, but images are surely no kind of picture at all. For, in the first case I can see the picture and the object of which it is a picture. But in the other, things are obviously quite different." (Cf. p.81: "If we take memory as a picture, then it's a picture of a physical event. The picture fades, and I notice how it has faded when I compare it with other evidence of what happened.") - On pp.77f. Wittgenstein writes: "You cannot compare a picture with reality, unless you can set it against it as a yardstick. - You must be able to fit the proposition on to reality. - The reality that is perceived [angeschaute Wirklichkeit] takes the place of the picture."

30. Wittgenstein here speaks of his "picture conception" ("Bild-Auffassung"), contrasting it with "the conception of Russell, Ogden and Richards". The picture conception "regards recognition as seeing an internal relation [das Wiedererkennen als das Erkennen einer internen Relation sieht], whereas in their view this is an external relation. - That is to say, for me", Wittgenstein explains, "there are only two things involved in the fact that a thought is true, i.e. the thought and the fact; whereas for Russell, there are three, i.e. thought, fact and a third event which, if it occurs, is just recognition."

31. On pp.83 and 86 references are made to "the picture on the screen". They add nothing to our understanding of pictorial meaning. In a different context however Wittgenstein's remarks on the film are very important. I have made use of them in my "Wittgenstein as a Philosopher of Secondary Orality",Grazer Philosophische Studien 52 (1996/97), pp.45-57. See also my references further below to Wittgenstein's remarks on "kinematographische Bilder".

32. The German edition (Philosophische Grammatik, ed. by Rush Rhees, published as vol.4 of Ludwig Wittgenstein, Schriften, Frankfurt am Main: Suhrkamp, 1969) and the English translation by Anthony Kenny (Berkeley: University of California Press, 1974) have identical page numberings.

33. Actually: a portrait by similarity, "ein 'ähnliches Porträt'".

34. "Or again", the passage continues, "there might be something preceding the image that made the connection with N. And so the interpretation isn't something that accompanies the image; what gives the image its interpretation is the path on which it lies. - That all becomes clearer if one imagines images replaced by drawings, if one imagines people who go in for drawing instead of imagining."

35. "Denn das Bild sollte doch die direktere Sprache sein." Kenny has: "the picture was supposed to be the more direct language".

36. The German edition has "in systematischen Bildern". It should be "in schematischen Bildern".

37. This is how the German text, in the Philosophische Grammatik, runs: "Wenn man es für selbstverständlich hält, daß sich der Mensch an seiner Phantasie vergnügt, so bedenke man, daß diese Phantasie nicht einem gemalten Bild oder plastischen Modell ähnlich ist; sondern ein kompliziertes Gebilde aus heterogenen Bestandteilen: Wörtern, Bildern, u.a. Man wird dann das Operieren mit Schrift- und Lautzeichen nicht mehr in Gegensatz stellen zu dem Operieren mit 'Vorstellungsbildern' der Ereignisse. - (Die Häßlichkeit eines Menschen kann im Bild, im gemalten, abstoßen, wie in der Wirklichkeit, aber auch in der Beschreibung, in Worten.)" The reference to the film is missing in the German version. As I will show later, this is not just an editorial lapse.

38. Essays on Wittgenstein in Honour of G. H. von Wright. Acta Philosophica Fennica, 1976/1-3, ed. by Jaakko Hintikka, Amsterdam: North-Holland Publishing Company.

39. Preliminary Studies for the "Philosophical Investigations". Generally Known as the Blue and Brown Books. By Ludwig Wittgenstein. [Preface by Rush Rhees.] Oxford: Basil Blackwell, 1958, repr. 1964, p.4. The issue of pictures also occurs in the context of Wittgenstein's criticism of the idea that "the meaning of a word is an image, or a thing correlated to the word". As Wittgenstein writes (p.18): "we are inclined to think that the general idea of a leaf is something like visual image, but one which only contains what is common to all leaves. (Galtonian composite photograph.)"

40. As Wittgenstein writes on p.39: "'There's no doubt I imagine King's College and no other building'. But can't saying this be making the very connection we want? For saying it is like writing the words 'Portrait of Mr. So-and-so' under a picture. ... The fault which in all our reasoning about these matters we are inclined to make is to think that images and experiences of all sorts, which are in some sense closely connected with each other, must be present in our mind at the same time."

41. As he puts it: "in some cases saying, hearing, or reading a sentence brings images before our mind's eye, images which more or less strictly correspond to the sentence, and which are therefore, in a sense, translations of this sentence into a pictorial language".

42. As Rhees writes, Wittgenstein thought of the Brown Book "as a draft of something he might publish. He started more than once to make revisions of a German version of it. The last was in August, 1936. He brought this, with some minor changes and insertions, to the beginning of the discussion of voluntary action - about page 154 in our text. Then he wrote, in heavy strokes, 'Dieser ganze "Versuch einer Umarbeitung" vom (Anfang) bis hierher ist nichts wert'. ('This whole attempt at a revision, from the start right up to this point, is worthless.') That was when he began what we now have (with minor revisions) as the first part of the Philosophical Investigations."

43. For some related remarks, see also pp.179f.

44. Ludwig Wittgenstein, Zettel, Oxford: Basil Blackwell, 1967. English translation by G.E.M. Anscombe. German text also in vol.5 of Ludwig Wittgenstein, Schriften, Frankfurt am Main: Suhrkamp, 1970.

45. Let us briefly list them. §§218f.: "I interpret words; yes - but do I also interpret looks? Do I interpret a facial expression as threatening or kind? - That may happen. ... - We don't understand Chinese gestures any more than Chinese sentences." -- §§231-233: "Think of a sign language..." (practically identical with PG pp.145f.) -- §239: "'At that moment the thought was before my mind.' - And how? 'I had this picture.' -So was the picture the thought? No; for if I had just told someone the picture, he would not have got the thought." -- §§241f.: "Let us imagine a picture story in schematic pictures..." (practically identical with PG pp.171, but with the drawing missing) -- §243: "Certainly I read the story..." (identical with PG p.171) -- §245: "I understand the picture exactly, I could model it in clay. - I understand this description exactly, I could make a drawing from it. - In many cases we might set it up as a criterion of understanding, that one had to be able to represent the sense of a sentece in a drawing (I am thinking of an officially instituted test of understanding). How is one examined in map-reading, for example?" -- §246: "And the significant picture is what can not merely be drawn, but also represented plastically. And saying this would make sense." -- §621: " ... Images are not pictures. I do not tell what object I am imagining by the resemblance between it and the image. - Asked 'What image have you' one can answer with a picture." -- §652: "If one takes it as obvious that a man takes pleasure in his own fantasies, let it be remembered that fantasy does not correspond to a painted picture, to a sculpture or a film, but to a complex formation out of heterogeneous components - signs and pictures."

46. Oxford: Basil Blackwell, 1980.

47. Oxford: Basil Blackwell, vol. I: 1982, vol. II: 1992.

48. I am not implying they contain nothing of interest. For instance the remarks §§1017-1019 of RPhPs, vol.I, first formulated in MS 135 (1947), have nowhere been published before. Wittgenstein here suggests, among other things, that the double cross "can be seen, not just in two but in very many different ways"; or that it would possible "that we had first to learn with some pains to understand a method of depiction, in order to be able later on to use it as a natural picture". Or the passage on gestures as "assimilated" though "not innate", written in 1949, variations of which are published as §712 of LWPhPs, vol.I, and §XXX of LWPhPs, vol.II {p.30 of the Suhrkamp ed.}.

49. 40a, Schleiermacher transl. - The passage fully, in Jowett's translation: "SOCRATES: And all men, as we were saying just now, are always filled with hopes? - PROTARCHUS: Certainly. - SOCRATES: And these hopes, as they are termed, are propositions which exist in the minds of each of us? - PROTARCHUS: Yes. - SOCRATES: And the fancies of hope are also pictured in us..."

50. 39a-b, Jowett transl.

51. As he said to Schlick in 1931: "I cannot characterize my standpoint better than by saying that it is opposed to that which Socrates represents in the Platonic dialogues." (MS 302:14)

52. The speech bubble, or speech balloon, is of course a familiar symbol from the world of comic strips. It appears to have emerged towards the beginning of the twentieth century. In the American comic strip Katzenjammer Kids it was employed as early as 1901. By contrast, speech bands, with texts emanating from the mouth of the speaker, are age-old. See especially Meyer Schapiro, Words, Script, and Pictures: Semiotics of Visual Language (New York: George Braziller, 1996, pp.117-119). Medieval books, as Schapiro writes on p.118, "sometimes depict speech naively as a string of letters issuing from the mouth of a person represented in the miniature... In some medieval book illustrations and also in sculptures, mosaics, and wall paintings, the figures hold scrolls on which their recorded speech is transcribed. The scroll itself becomes a sign of speech." On p.173 Schapiro points out that "visual rendering of speech in the Middle Ages may be matched on Greek vases", and refers to "an often cited one, made shortly before 500 B.C., [on which] are painted three young man who see a swallow; one cries: 'How lovely!'; another: 'Spring is here". A swallow flies above them; their speech, issuing from very near the mouth, is directed upward in the air." On the emergence and the varieties of the speech bubble as a comics and cartoon convention, cf. e.g. Carl G. Liungman, Dictionary of Symbols (New York: Norton & Co., 1991, pp.358f., original Swedish edition1974), William Horton, The Icon Book: Visual Symbols for Computer Systems and Documentation (New York: John Wiley & Sons., Inc., 1994, p.69), and Robert E. Horn, Visual Language: Global Communication for the 21st Century (Bainbridge Island, WA: MacroVU, 1998, pp.141f.). Horton in particular indicates that convention and intuition both play a role in the family of speech balloon symbols. "Consider", he writes, "the meaning conveyed by the shape of speech balloons in cartoons. What kind of message would you expect each of these speech balloons to deliver?"

A related point is made by Ian Hacking, in his essay "Dreams in Place", Journal of Aesthetics and Art Criticism, Summer 2001, p.251. I am indebted to Carolyn Korsmeyer for bringing this essay to my attention.

53. The word is missing from the German edition, too (cf. Bemerkungen über die Grundlagen der Mathematik, as vol.6 of Ludwig Wittgenstein, Schriften, Frankfurt am Main: Suhrkamp, 1974, p.159).

54. J.C. Nyíri, "The Picture Theory of Reason", in: Berit Brogaard - Barry Smith, eds., Rationality and Irrationality, Wien: öbv-hpt, 2001, p.253.

55. There is an early remark on pp.49f. of MS 153 (1931) which might seem promising:

Wenn man sagt: Ich stelle mir die Sonne vor wie sie rasch über
den Himmel zieht; so
ist doch nicht die Vorstellung damit beschrieben daß „die Sonne rasch
über den Himmel zieht"!

Nun könnte ich einerseits sagen «fragen»: ist nicht, was Du vor Dir
siehst «etwa» eine gelbe Scheibe in Bewegung
aber doch nicht gerade die Sonne?-- andrerseits, wenn ich sage „ich
stelle mir die Sonne so & so vor" so ist das nicht dasselbe als wenn
ich -- etwa kinematographisch -- ein solches Bild zu sehen be-

Ja es hätte Sinn von diesem Bild zu fragen: „stellt das die
Sonne vor?"

The remark reappears, with slight variations, on p.290 in MS 110, and on p.305 of TS 211. It also occurs in the Big Typescript (TS 213, pp.290f.); it is there immediately followed by an important remark a variant of which I have quoted earlier from the Philosophical Grammar (where, on p.102, the "Ich stelle mir die Sonne vor..." paragraph is however absent):
Wenn man sagt: Ich stelle mir die Sonne vor, wie sie über den
Himmel zieht; so ist doch nicht die Vorstellung damit beschrieben,
dass "die Sonne über den Himmel zieht"! Nun könnte ich einerseits fragen: ist
nicht, was Du vor Dir siehst, eine gelbe Scheibe in Bewegung? aber
doch nicht gerade die Sonne. -- Andrerseits, wenn ich sage "ich
stelle mir die Sonne in dieser Bewegung vor", so ist das nicht
dasselbe, wie wenn ich (etwa kinematographisch) ein solches Bild zu
sehen bekäme.

Ja, es hätte Sinn, von diesem Bild zu fragen: "stellt das die
Sonne vor?"

Das Porträt ist nur ein dem N ähnliches Bild (oder auch das
nicht), es hat aber nichts in sich (wenn auch noch so ähnlich), was
es zum Bildnis d i e s e s Menschen, d.h. zum beabsichtigten
Bildnis machen würde. (Ja, das Bild, was dem Einen täuschend
ähnlich ist, kann in Wirklichkeit das schlechte Porträt eines
Anderen sein.)

The idea of animated pictures and the idea that pictures do not become unambiguous merely by resembling an object occur together in this passage. Still, it is far from clear what Wittgenstein was actually trying to say here.

56. London: Hutchinson's Universal Library. I have provided a summary of Price's argument in my "The Picture Theory of Reason".

57. "Wittgenstein and the Problem of Machine Consciousness", Grazer Philosophische Studien 33/34 (1989), pp.375-394; "Schriftlichkeit und das Privatsprachenargument", Deutsche Zeitschrift für Philosophie 40/3 (1992), pp.225-236; "Heidegger and Wittgenstein", in: Nyíri, Tradition and Individuality: Essays, Dordrecht: Kluwer, 1992;" "Wittgenstein as a Philosopher of Post-Literacy", in: K.S. Johannessen and T. Nordenstam, eds., Culture and Value: Philosophy and the Cultural Sciences, Papers of the 18th International Wittgenstein Symposium, Kirchberg am Wechsel: 1995, pp.82-88; "Wittgenstein as a Philosopher of Secondary Orality". Grazer Philosophische Studien 52 (1996/97), pp.45-57.

58. I elaborate this point in my "The Picture Theory of Reason", loc. cit.

59. The thesis of Wittgenstein the dyslexic was formulated by Jaakko Hintikka and Anna-Maija Hintikka. Both gave talks at the 2001 Kirchberg symposium. In her lucid and thorough talk "Dialogues with Inner Pictures: Ludwig Wittgenstein as Dyslexic" Anna-Maija Hintikka, MPh, a speech therapist, marshalled facts of family history, biographical data, and autobiographical testimony to prove beyond any possible doubt that Wittgenstein has indeed suffered from dyslexia. Of the wealth of observations and details she offered let me here mention just one that I find, in the present context, very pertinent. Drawing, she said, was for Wittgenstein "a means of communication. Von Wright provides an example of this in telling that in his Charlottenburg days Ludwig Wittgenstein had a friend with whom he 'conversed' by means of drawing pictures."Jaakko Hintikka's paper "Ludwig Wittgenstein: The Bewitched Writer" constituted a brilliant survey of Wittgenstein's philosophy, demonstrating that many of its central ideas were indeed attempts to cope with the dyslexia condition. There exists an earlier, unpublished version of this paper, entitled "Ludwig Wittgenstein - A Case Study in Dyslexia". In what follows, I am quoting from this version. Dyslexia, that is, as Hintikka puts it, the "slow, impaired recovery of the phonetics and semantics of written text from visual clues", is a "cognitive challenge" which "forces a dyslexic person to look upon language and linguistic skills in a way we usually do not do". Hintikka points out that: "In the same way ... a dyslexic has difficulties in keeping in mind the meaning of a sentence because of the need of concentrating on particular words, [so also] a dyslexic finds it hard or even personally impossible to keep track of an argument or other similar line of thought or at least articulate it verbally." Certainly Wittgenstein was unable to maintain, and, as Hintikka stresses, indeed programmatically denied the possibility of, "a linear or progressive mode of organization of his ideas". Much of Wittgenstein's "actual philosophical thought can be viewed", Hintikka writes, "as a series of attempts to understand his own handicaps and to overcome them or as attempts to articulate and to generalize philosophically his experiences as a dyslexic".

60. At the Bergen conference the discussion of the present paper war introduced by Ralph Jewell. In his congenial comments he made valuable additions to my argument. I look forward to see them published in the envisaged volume containing the materials of the conference.