When one thinks of Wittgenstein’s Philosophical Investigations, and in particular of §§243–315, the word ‘übersichtlich’: that is, ‘easy to survey’, ‘synoptic’,1 does not spring to one’s lips. One is much rather reminded of the way Wittgenstein characterizes the book in the Preface: as ‘travel(s) over a wide field of thought, criss-cross in every direction’:
The same or almost the same points were always being approached afresh from different directions, and new sketches made. … Thus this book is really only an album.
(PI, Preface)
If this is an apt description (and I think it is), what are we to make of Wittgenstein’s claim in §122 that ‘the concept of a synoptic representation (übersichtliche Darstellung) is of fundamental significance’ for him? Here is §122 in full:
A main source of our failure to understand is that we lack a synoptic view [nicht übersehen] of the use of our words.–Our grammar is not easy to survey [fehlt es an Übersichtlichkeit].–A synoptic representation [übersichtliche Darstellung] produces just that understanding which consists in ‘seeing connexions’. Hence the importance of finding and inventing connecting links.
The concept of a synoptic representation [übersichtliche Darstellung] is of fundamental significance for us. It characterizes our form of representation, the way we look at things. (Is this a ‘Weltanschauung’?)
(PI §122)
Where in the Philosophical Investigations do we find such synoptic representations?–I believe what Wittgenstein had in mind here are simple, fictitious language-games, like that of buying apples (§1), the builders (§§2, 8, 15, 21), the description of coloured squares (§§48, 64), ordering someone to fetch composite objects (§§60, 62), or the reading of a table according to different schemas (§86).
There is strong evidence for this interpretation in §5 of the Investigations. Having mentioned the distorting influence of a wrong philosophical notion of meaning, Wittgenstein continues:
It disperses the fog to study the phenomena of language in primitive kinds of application in which one can command a clear overview [übersehen] of the aim and functioning of the words.
(PI §5)
Here, simple language-games are characterized, almost verbatim, as übersichtliche Darstellungen of the use of words. The same point is made in the Blue Book:
If we want to study the problems of truth and falsehood, of the agreement and disagreement of propositions with reality, of the nature of assertion, assumption and question, we shall with great advantage look at primitive forms of language in which these forms of thinking appear without the confusing background of highly complicated processes of thought. When we look at such simple forms of language the mental mist which seems to enshroud our ordinary use of language disappears. We see activities, reactions, which are clear-cut and transparent. On the other hand we recognize in these simpler processes forms of language not separated by a break from our more complicated ones. We see that we can build up the complicated forms from the primitive ones by gradually adding new forms.
(BB 17)
In this passage, for ‘synoptic’ (übersichtlich) we have another visual metaphor: ‘transparent’.
In the Brown Book, too, Wittgenstein defends the idea that a complicated and confusing part of grammar may be represented by judiciously invented simple language-games, each exhibiting one aspect of the grammar of the word under discussion: ‘some examples showing these features in exaggeration, others showing transitions, certain series of examples showing the trailing off of such features’ (BB 125). Thus, the reader should be given a whole series of such simplified examples; as indeed Part I of the Brown Book consists largely of a succession of simple language-games with some comments interspersed. Although in the Investigations Wittgenstein used this device much less frequently, still there are instances where a simple language-game is made more complicated and thus brought closer to our actual use of language (esp. §§2, 8, 15). This fits well what Wittgenstein says in §122 about the importance of finding or inventing ‘connecting links’. The word ‘inventing’, incidentally, makes it clear that fiction is a legitimate device for a synoptic representation.
Further evidence can be found in Friedrich Waismann’s Logik, Sprache, Philosophie, a book that was written in the 1930s in close co-operation with Wittgenstein as an attempt to give a systematic account of the latter’s new philosophy. Chapter IV.2 is entitled ‘Language Games’ and explains the method of giving ‘synoptic representations’:
We could invent new language games and imagine that, for instance, some primitive tribe knows only one of these language games or only a certain combination of them; and thus we illuminate the boundless [unübersehbare, literally: unsurveyable] ever-changing complex of language by contrasting or comparing it with clearly defined constructions which nevertheless we cannot help calling ‘language’.
(LSP 125; PLP 79)
Then this procedure is compared to Goethe’s conception of an ‘original plant’ (Urpflanze):
His conception of the original plant implies no hypothesis about the temporal development of the vegetable kingdom such as that of Darwin. What then is the problem solved by this idea? It is the problem of synoptic representation. ‘All the organs of plants are leaves transformed’ offers us a schema in which we may group the organs of plants according to their similarities as if around some natural centre. … That is precisely what we are doing here. We are collating one form of language with its environment, or transforming it in imagination so as to gain a view of the whole of the space in which the structure of our language has its being.
(LSP 127f., cf. PLP 81)2
In a recently published variant of this passage the final sentence translates:
we situate a linguistic form in its surroundings, we see the grammar of our language against a background of similar and related games, and that banishes disquiet.
(VW 310f.)
A simple, possibly invented, language-game, or even a series of them, is an illuminating object of comparison (PI §130). Like the schema of the ‘original plant’, it provides a synoptic representation of some potentially puzzling features of our language.
Apart from such textual support, the interpretation that simple language-games are instances of synoptic representations seems also to win by default. No other candidates are plausible. In particular, there are no systematic lists of grammatical rules in the Investigations.3 There are only sporadic grammatical propositions or reminders that some philosophical claim is in conflict with common usage (e.g. that the bearer of a name is not what one calls its meaning (§40), or that it is often correct to say ‘I know he’s in pain’ (§246)). Such sporadic reminders certainly do not merit the title ‘synoptic representation’. So if Wittgenstein’s simple language-games cannot be called ‘synoptic representations’, we would have to dismiss §122 as an entirely unfulfilled programme. Furthermore, it is certainly true that such simple language-games are what synoptic representations are supposed to be: namely, a characteristic feature of Wittgenstein’s mature writings (cf. §122b).
Against the idea that simple language-games might be intended as synoptic representations it has been objected that ‘we cannot represent the rules of chess through describing draughts’ (Glock 1996: 281).–Of course one cannot represent all the rules of one game by describing a different game, but in philosophy we never need to have a comprehensive set of grammatical rules (philosophy is not linguistics); we are only concerned with certain puzzling features of grammar (cf. Baker and Hacker 1980: 95).4 To develop the games analogy: suppose someone is puzzled to hear that in chess a pawn can be promoted to a queen. Now this sort of promotion: that in a board game one piece is turned into another, can indeed be demonstrated and explained by using a less complicated game, like draughts. And in such a case one might well say that the teacher makes the promotion of a draughtsman stand for or represent the promotion of a chess pawn. Similarly, Wittgenstein uses the simple language-game of the builders (PI §2) to make, or illustrate, some fundamental points about language in general: that the use of language is essentially interwoven with non-linguistic activities; that even where Augustine is correct in so far as each word is associated with an object, such a correlation is not enough to account for the meaning of a word (PI §10); rather, the meaning of a word is its use (PI §43). And in the subsequent extensions of the builders’ language-game (PI §§8, 15, 21) Wittgenstein illustrates some basic features of colour words, numbers, proper names, and the difference between reporting and ordering.
Simplification is what makes a representation synoptic (cf. TS 302, 21). In a lecture Wittgenstein used the analogy of country and map: for language and an account of its grammar (AWL 43). Maps are so much easier to survey than the country they represent because they simplify matters; they leave out millions of details, which makes it possible to print them on a small sheet of paper. (A map that depicted every pebble would be pretty useless.) Similarly, a written grammar that conscientiously recorded each and every detail of our use would probably be not much less confusing than the real thing. The wealth of details would prevent us from seeing the crucial points and connections–the main contours of the grammatical landscape–that are philosophically relevant.–Note also that in a good map features of particular interest may be exaggerated. Street maps, for example, depict streets much wider than they actually are. That is not a mistake, but a way of making the map easier to read.–A map is a synoptic representation; but a long and detailed account of one’s travels criss-cross through the country is not, even if it serves the same purpose of giving someone a good idea of how the land lies. Therefore, neither the Philosophical Investigations as a whole, nor what could be regarded as its chapters (like the Private Language Discussion), can be called synoptic representations–whatever their enlightening effects on a reader.
If this interpretation is correct, there are some synoptic representations in the Investigations, but not that many. There was a time when Wittgenstein intended to make more use of this device than eventually he did. This earlier stage is embodied in the Brown Book. Why did he soon give up the method he employed there: of simply sketching one fictional language-game after another? Perhaps he found that such synoptic representations are not quite the panacea he had once hoped they were. They may be sufficient to dissolve some philosophical problems, but others are more recalcitrant.
One can distinguish between two stages of a philosophical problem. At first a misleading analogy (or form of expression) produces a disquietude (Beunruhigung), as Wittgenstein often calls it: something appears paradoxical, impossible or inexplicable. Such a disquietude typically manifests itself in a question of the form ‘How is it possible … ?’ (e.g. ‘How is it possible to follow a rule?’(PI §217)), or–in all those cases where a noun produces the wrong impression of there being an object–by a ‘What is … ?’-question (e.g. ‘What is time?’(PI §89); ‘What is language?’(PI §92)). A disquietude is the early stage of a philosophical problem. Given a certain temperament it may lead to a more advanced stage. We may be led to answer those questions: to develop metaphysical views, a philosophical theory, in conflict with common usage and without any clear use. Thus, at a second stage, philosophical confusion becomes dogmatic.
It’s easy to see how a synoptic representation can provide an appropriate therapy for a disquietude. Seeing clearly how a language-game works ‘disperses the fog’ (PI §5), exposes the ‘false appearance’ (PI §112) as mere appearance, and those questions can be dispatched lightly, by trivial answers that make fun of the questions (TS 220, §111).5 Things are different with substantial philosophical positions. Consider, for instance, §273:
the word ‘red’ means something known to everyone; and in addition, for each person, it means something known only to him
(PI §273)
Here, one could of course present a simple language-game of colour words, thus giving a synoptic representation of the essential features of our colour language and, of course, a private meaning of the word ‘red’ would play no role in such an account. But that would hardly convince those who believe in private meanings of colour words. Once we are not just baffled or disquieted by a philosophical picture, but firmly in its grip, a demonstration that it is in conflict with common usage is unlikely to have much effect. Indeed, some philosophical pictures are presented quite expressly against common usage, and as a correction of it (e.g. idealism, solipsism (PI §402)). Obviously, in such a case much more is required than an account of the workings of our language. A considerable part of the therapy will be concerned with the philosophers’ anomalous use of language, trying to expose it as pointless or inconsistent.
I shall now turn to the Private Language Discussion–and its lack of a synoptic view. I shall outline two cases where the reader is likely to be puzzled by the way Wittgenstein presents his text.
(A) It is natural to take §243b as providing the title question, as it were, to this chapter of the book: ‘Could one have a language that referred to one’s private experiences?’ But note the exchange that leads to the final wording of the question. The sensible reminder that it is perfectly possible in our ordinary language to record one’s feelings and moods is not contradicted, but dismissed as irrelevant (‘that is not what I mean’), followed by a more precise description of the subject matter of the envisaged language: not any old sensation, but one’s immediate private sensations. The implication seems to be that not all our sensations etc. are strictly private. What the interlocutor appears to have in mind are those diffuse feelings and finer nuances of experience that are so notoriously difficult to put into words. However, no such thing is discussed in the following sections. After the interlude of §§244–5, instead of answering the question from §243 Wittgenstein appears to change the subject. He turns to the somewhat different issue whether others can know for certain whether I am really in pain (§246). This is not an answer to the question of §243 as, for one thing, of all sensations pain must be the last thing the interlocutor had in mind, because pain is obviously an experience for which we have a word in our public language. Moreover, §246 is concerned only with the sceptical doubt whether others are ‘really’ in pain, presupposing a general understanding of the word ‘pain’, whereas the initial question is about the far more radical idea that others cannot even understand the words that denote my private sensations.
If §246 does not really address the question raised in §243, but only a related one, §253, where the discussion of privacy is taken up again, seems completely irrelevant. ‘Nobody else can have my pains’ is of course another possible interpretation of the expression ‘privacy of sensations’–but clearly not the one that is at issue in §243. The discussion in §253 would apply just as well to ‘Nobody else can have my smile’ or ‘Nobody else can have my cough’, though evidently neither a smile nor a cough is a candidate for something that ‘can only be known’ to its owner.
So, here Wittgenstein begins a discussion with a question, but what at first glance appears to be an attempt to answer it is, if taken as such, rather unfocused or entirely off the mark.
(B) Consider the famous centre-piece of the Private Language Discussion, the private diary of §258. Many readers, even among Wittgenstein’s admirers, have felt somewhat uncomfortable about it. True, the no-criterion objection seems correct and unanswerable; but still, there is a lingering feeling that this sort of diary is possible after all; and would that not mean that a private language is possible after all?6 Has the ‘perfect clarity’ Wittgenstein aimed at (PI §133) really been achieved in this case? Is there not some disquietude left? (Incidentally, Wittgenstein himself occasionally seems to have felt a little unsure about the private diary. In 1946, two years after the writing of those sections, he returned to it in a way that showed that some of the disquietude about a private sensation language had not been entirely cured (RPP I §§393–9)7.) The problem about §258 is this. It is unclear whether the scenario is supposed to be inconsistent, or whether it just falls short of what Wittgenstein’s interlocutor meant to achieve. The stronger reading is suggested by the final conclusion of §258:
And that only means that here we can’t talk about ‘right’.
(PI §258)
And by §260, which implies that the diarist failed to make a note of anything:
‘Well, I believe that this is the sensation S again,’–Perhaps you believe that you believe it!
Then did the man who made the entry in the calendar note down nothing whatever?–Don’t consider it a matter of course that a person notes something down when he makes a mark–say in a calendar.
(PI §260)
But should it really be impossible to associate a certain sensation–for which we have no natural expressive behaviour–with a sign and keep a record of the re-occurrence of this type of sensation? That seems to be what Wittgenstein is saying here; but it is contradicted in §270, where he describes circumstances under which the diary becomes useful (nützlich):
I discover that whenever I have a particular sensation a manometer shews that my blood-pressure rises.
(PI §270)
Clearly, this discovery and the ensuing usefulness of the diary presuppose that it already was a sensation diary before (cf. Hanfling 2002: 31–3). For it to be possible to discover that a certain sensation is always followed by rising blood-pressure, it must also be possible to find that it is not always so followed. Hence, it must be meaningful to speak and keep a diary of that sensation independently of any such useful correlation.
What then are we to make of the sensation diary of §258?–I shall argue that it is best understood as a synoptic representation, as promised in §122; though somewhat misleadingly presented. Still, the Private Language Discussion is not easy to survey. And to become clear not only about the apparent leaps at the beginning of the Private Language Discussion, but also about the role of §258 it is important first to have a clear idea of the overall aims and strategy of these passages and their arrangement. That I shall turn to now before returning to §258.
One of the things that make it so difficult to understand what is going on in these sections is that Wittgenstein does not always bother to preface his discussions with an explicit and comprehensive account of the philosophical picture he is engaged to combat. Of course he often quotes with disapproval a philosophers’ statement, but such statements are only the corollaries of an underlying picture. And a clear understanding of that picture is required to see how those philosophers’ statements hang together, as symptoms of the same ‘disease’. Otherwise crucial sections can easily appear in the wrong light (as I began to illustrate about §258), and it will seem as if he jumped capriciously from one question to another (as demonstrated above with §§243, 246 and 253). Well, to some extent he does jump from one question to another; that cannot be denied. But there is some method in his crisscrossing in so far as all his remarks are united by a common target: a particular philosophical picture with its ramifications.
So to make the Private Language Discussion easier to survey I shall now spell out the underlying picture, draw out its consequences in a synoptic diagram, and then try to correlate that diagram with some of Wittgenstein’s remarks.
As P.M.S. Hacker observed, Wittgenstein’s main concern in the Private Language Discussion is an offshoot of the Augustinian picture of language (Hacker 1990: 16), which has it that:
(AP) Meaning is due to reference: words stand for objects.8
From which it is tempting to derive the negative claim that:
(AP.n) The use of a word is irrelevant to its meaning (semantics ≠ pragmatics).
The Augustinian picture may be roughly appropriate for some words,9 but it is not acceptable in its intended generality (PI §§2f).10 Some counterexamples are fairly easily recognized as such. But sensation words are a more difficult case.
Applying the Augustinian picture to sensation words (and other psychological terms) means that sensations must be objects. Evidently they are not physical objects; they are objects perceived by one’s inner sense inside one’s mind. And the mind is a private container to which only its owner has access. This inner-object conception of sensations is explicitly mentioned in §293 and §304 (cf. §374), but unfortunately not at the very beginning of the Private Language Discussion.
The inner-object conception of sensations has three corollaries:
These have further consequences. With its roots and branches the inner-object conception can be represented by a diagram (see Table 7.1).
This metaphysical picture, with its implications, is the landscape which Wittgenstein traverses criss-cross in §§243–315. Table 7.2 is a correlation to show which point is to the fore in which sections.
Ad (A). Following the order of his remarks, one can see that he begins at the very end: the idea of a private sensation language is an extreme consequence of the Inner Object Picture. Then, after a short sketch of how we learn sensation terms (§244)–indicating a better construal of this part of grammar–in §246 Wittgenstein moves back to an earlier, less radical implication of the picture: direct and indirect perception providing certainty or mere surmise respectively. Section 253 is not strictly pertinent to the private language discussion, but–as the diagram shows–it is concerned with another branch sprouting from the Inner Object Picture: if you have a sensation as one has a beetle in a private box, then of course what is in your box cannot be in mine. The question of a private sensation language from §243 is resumed only in §256, after various remarks showing pain to be a sensation far from private in virtue of its straightforward connection with natural expressive behaviour.
I do not think too much should be read into the order and arrangement of the remarks. It would be nice if one could argue that there was some kind of hidden necessity in this particular arrangement, but there is not. Wittgenstein himself cautioned us that there was not–‘really only an album’–and his method of composition makes it a priori unlikely that there should be. Remember that these remarks were not written to fit each other, like the pieces of a jig-saw puzzle; rather, they were taken out of quite different contexts and joined together as well as their diversity would allow. And it is remarkable that Wittgenstein never did what most people would have done: add some new sentences in between those remarks to weld them together; to make smoother transitions from one to the next. On the contrary, fairly tight lines of argument in an earlier version of the book were broken up by the insertion of other remarks, related but also distracting and making it more difficult to see some crucial links. Thus, the Intermediate Version has §256 immediately followed by §258, before Wittgenstein decided to insert what is now §257, a remark that was written eleven years earlier (in 1933); which effects a retardation in the progress of the action. Section 257 gives another version of the setting for the private diary and a very succinct anticipation of some comments on it that are made in §§260, 261 and 270. Obviously that kind of insertion does not add to the text’s Übersichtlichkeit. Again, in the Intermediate Version, §261 was immediately followed by §270. That means, the whole presentation and discussion of the diary issue was given in one block of five sections. Now, however, there is a long dramatic pause before the final act: eight new sections separate the first part of the discussion from its resolution in §270.
Spelling out systematically what Wittgenstein argued against helps one to find one’s bearings; shows that the remarks are all connected through their opposition to one ramifying metaphysical conception; but it does not make the book any less like an album, in which one moves backwards and forwards, and sometimes leaps from one end to the other. Having said that, the jump from §243 to §246–from (2?3 e) back to (2a) and (3a)–is not incomprehensible. To begin with, Wittgenstein gives an idea of the far-reaching consequences of the Inner Object Picture: a private sensation language. However, the treatment starts in more familiar regions, with the common-or-garden example of pain and the common place of the comparative uncertainty of our knowledge of others’ pain. This is in accordance with Wittgenstein’s maxim:
In philosophy treat the clear cases first, not the unclear ones. The latter will be dissolved when the former are.
(TS 211, p. 510)11
How does Wittgenstein attack the Inner Object Conception? First, there are various pertinent reminders (cf. PI §127): how in fact we learn to speak of pain (PI §244); that often we do say of others that they know that I’m in pain (PI §246). Apparent linguistic evidence for the Inner Object Picture is shown to be specious by a reminder of its actual context and meaning (PI §247); and we are reminded of cases where the claim of epistemic privacy begins to look ridiculous (PI §249; cf. §303b). Second (though not in order of appearance, but much later in the text), some of those metaphysical statements are denounced as pointless: as nothing but idle expressions of a picture without any use outside philosophy (PI §§295f, 298, 303). The most powerful presentation of this kind of criticism is the simile of the beetle in the box (PI §293): the Inner Object Picture is exposed as entirely irrelevant to our actual talk about sensations. Third, in one short (but wonderfully crafted) paragraph (PI §302) Wittgenstein draws attention to an inconsistency in the Inner Object Picture (2?3 c): that I know sensations only from inner experience is incompatible with (3) and (3c): that I attribute sensations, though precariously, to others.12
However, the two most prominent criticisms of the Inner Object Picture emerge out of the discussion of one of the few instances of a synoptic representation in the Investigations: the private sensation diary in §§258 & 270, to which now I return.
Ad (B). Wittgenstein remarks that the private sensation diarist has no criterion of the correctness of his entries, and he concludes:
One would like to say: whatever is going to seem right to me is right. And that only means that here we can’t talk about ‘right’.
(PI §258)
The verdict is a little too harsh. For it does not mean that. ‘Whatever is going to seem right to me is right’ is not the same as, ‘Anything is right’. It would still be wrong for me to write down ‘S’ on a day when none of my sensations seemed to be of the same kind as the one I initially called ‘S’. Hence, one can very well speak of a right or correct application of ‘S’. As Wittgenstein remarked in a different place, truth can coincide with truthfulness (PI, p. 222). Seeming right is the same as being right in the domain of sensations and feelings. The conclusion that is warranted by the lack of a criterion in §258 is rather: And that only means that here we can’t talk about an error.’ Which is exactly what Wittgenstein says about reporting one’s pain:
But I can’t be in error here; it means nothing to doubt whether I am in pain!
(PI §288)13
This is a powerful argument directed at the very core of the Inner Object Picture. If a sensation were an inner object perceived and identified through introspection, it would be conceivable that one should misperceive and misidentify it. It would be possible to be mistaken in one’s belief that one was in pain. That sounds absurd, and §258 and §290 explain why: we do not identify our sensations by criteria. And where there is no criterion, there is no possibility of error. The crucial move is a re-interpretation of modal words: from a metaphysical explanation why something seems impossible to a logical one. Wittgenstein makes parallel moves in response to all three branches of the Inner Object Picture, as shown (in italics) at the bottom line of Table 7.3.
In each case there is something impossible: to have another’s sensation; to perceive another’s sensation directly; to misperceive one’s own sensation.
According to the Inner Object Picture such impossibility is due to the metaphysical constitution of things: to the fact that everybody’s beetle is in a box to which no one else has access (cf. PI §293). But in that case the opposite state of affairs should at least be conceivable. And as this is not so, the picture must be wrong. What looks like a metaphysical impossibility is, in fact, a logical one, like the impossibility of scoring a goal in tennis:
Do not say ‘one cannot’, but say instead: ‘it doesn’t exist in this game’. Not: ‘one can’t castle in draughts’ but–‘there is no castling in draughts’; and instead of ‘I can’t exhibit my sensation’–‘in the use of the word “sensation” there is no such thing as exhibiting what one has got’.
(Z §134; cf. PI §114)
That is the rejection of (3x); in the Investigations the same point is made in §374. (1x) is dispatched in §253; while (2x) is the target of the no-criterion objection launched in §258 (and pursued further in §§259; 265–7; 270f; 288; 290; 292). The point to emphasize is this: The ‘no criterion’–objection of §258 is not an objection to the private sensation diary described in this section, but an objection to a misconstrual of this or indeed any other sensation language: namely the Inner Object Conception and its immediate consequence that one perceives one’s own sensations. It is directed at (2) (in my diagram), not directly at (2?3 e), and not at all at the idea of keeping a record of sensations without a natural expression. The point concerns ordinary sensations, like pain, just as much as the private (i.e. fictitious) scenario without the distraction of natural expressions or a familiar use of the sensation word.
The scenario itself–and not just a misconstrual of it–is criticized in §260:
a note has a function, and this ‘S’ so far has none.
(PI §260)
But the criticism is not particularly convincing: many respectable applications of language have no function; are just idle talk. Perhaps the word ‘use’ would have been better than ‘function’ (Hanfling 2002: 31). Anyway, as the words ‘so far’ indicate, this remark is not meant to point out an irremediable flaw in the set-up of the diary, it is merely an admonition that the language-game is not yet complete.
Next, Wittgenstein complains that we have no reason ‘for calling “S” the sign for a sensation’ (PI §261). Is this supposed to mean that the diarist has not really got a sensation when he writes ‘S’? No. After all the question was whether someone who had sensations without any natural expression for them could keep a diary about them. So the worry is not that those entries may not be correlated with occurrences of sensations–it is presupposed that they are; the worry is that such a correlation does not suffice to make ‘S’ a sensation word.14 Such a correlation is not enough to establish that the diarist has mastered the concept of a sensation (cf. PI §384). (As an animal could be trained to make a certain noise at the sight of a bicycle, without having the concept of a bicycle.) What then would be a reason for calling ‘S’ the sign for a sensation? Well, suppose we asked the diarist, ‘Where is that S?’ and he answered, ‘In my left knee’, then we had some reason for regarding ‘S’ as a sensation word. If he added further that usually S lasted for about half an hour and was slightly unpleasant, especially when he was running–he would display an ability to talk about a sensation and thus show some grasp of the concept of a sensation; even if those entries did not fulfil any function. However, finding the right sort of function, or usefulness for ‘S’ may make it particularly clear that it has the use of a sensation word. That is what Wittgenstein demonstrates in §270. When recording S the diarist applies a manometer, and perhaps a few days later he is able to say, correctly, that his blood-pressure is rising without using any apparatus. We ask, ‘How do you know?’, and get the reply, ‘There’s S again’. An awareness of S is understood to be informative about one’s bodily state;–so the use of ‘S’ displays an important and typical feature of bodily sensation language-games. This should not be read as implying that in order to be meaningful the sign for a sensation must be useful (although, unfortunately, this is what §260 seems to suggest). In fact, it is not even clear that it is the sign ‘S’ or the diary entries that are particularly useful; primarily it is the regular occurrence of a certain sensation that comes in handy, whether it is given a name or not. But even if we change the description so that the diary becomes instrumental in the discovery of the link between S and rising blood-pressure, such usefulness provides only a particularly clear example of a sign being used in a respectable language-game, suitable for a stylized account of grammar in a synoptic representation. This is comparable to the obvious usefulness of the builders’ language-game of §2. A’s utterances help them in constructing a building; and that underscores Wittgenstein’s central point that language is an activity insolubly intertwined with other activities in our lives. But the linguistic meaningfulness of A’s utterances would not be undermined if he ordered B to give him those building-stones only for fun. Similarly in §270, Wittgenstein might just as well have shown ‘S’ to be a sensation word by connecting those entries with other behaviour in a way that does not make either of them appear useful. Suppose, for instance, that after every ‘S’ entry the diarist went to bed for a couple of hours, and somehow it became clear that the occurrence of S was his reason for lying down,–then, again, it would be reasonable to regard ‘S’ as a sensation word. Yet, it does not matter whether taking a rest after S is, in fact, sensible from a medical point of view; the important thing is that taking a rest is a typical reaction to bodily feeling. Hence, when it occurs as a response to something the diarist calls ‘S’ we have a reason to regard this sign as reporting some sort of bodily feeling.
Taken together, §§258 and 270 make as good an instance of a synoptic representation as one can find in the Investigations. A simplified, hence easily surveyable representation of a segment of sensation language. In two steps we are given a scenario that fulfils the minimum requirements of a language-game of naming and referring to a comparatively private sensation:
(i) Every now and again, in response to certain sensations that have no other expression, someone enters the sign ‘S’ into a diary. (ii) After a while he is able to say correctly when his blood-pressure is rising, citing S as evidence.
Contrary to (AP.n), (i) alone does not suffice to endow the sign with meaning. Correlation of a sign with some phenomenon is not enough for the sign to become a name of that phenomenon.15 The sign must have an understandable use (PI §260),16 and its grammar must show some typical features of the use of sensation words (PI §261).
Furthermore, the description of (i) affords Wittgenstein a good opportunity to show that sensations are not inner objects perceived through introspection (the ‘no criterion’ objection levelled at (0) and (2)).
Wittgenstein’s moves against the Inner Object Picture
Finally, I would like to attempt an explanation why both the overall structure of the Private Language Discussion and the presentation of its crucial arguments leave so much to be desired; that is to say, leave so much exegetical work to be done by Wittgenstein scholars.
First, Wittgenstein found it impossible to write a clearly structured treatise. He thought he was able to teach philosophy, but not to write a philosophical book (MS 118, p. 172). The best he could write were ‘philosophical remarks’; his ‘thoughts soon went lame’ if he ‘tried to force them on in any single direction against their natural inclination’ (PI, p. vii). This was partly due to his intellectual temperament, but it was also ‘connected with the very nature of the investigation’ (PI, p. vii).
Wittgenstein likened his philosophy to therapy: the treatment of diseases of the understanding (PI §255). That is to say, it offers no positive doctrine, but merely the dissolution of conceptual confusions. It is essentially a reaction to something else. It needs to follow someone’s conceptual confusions and someone’s inclinations to respond to objections or defend a philosophical picture through further refinements. Thus, the dialogue form is essential to it. But to the extent to which one’s interlocutor’s worries and responses move, sometimes abruptly, from one issue to another, one needs to follow these movements–even though they may seem confusing to others–if one is to convince one’s interlocutor. And perhaps this is what individual worries and responses are usually like: they frequently jump from one point to another, in a way that seems natural to oneself but less so to others.
Even so, Wittgenstein felt that his book was not as well-presented as it could have been:
I should have liked to produce a good book. This has not come about.
(PI, p. viii; cf. Monk 1990: 503; MS 152, p. 13)
Its shortcomings were partly due to his own temperament. His natural inclination was to ‘jump about all round a topic’ (CV 33); to keep changing his position: ‘not to stand too long on one leg, so as not to go stiff’ (CV 32). Obviously, this natural inclination to jump around is still manifest in the Investigations, in spite of the author’s attempts to control himself. This jumpiness was connected with some uncertainty as to what he did and what he did not need to discuss. As he noted in 1947 (when the book was largely finished):
I still keep getting entangled in details without knowing whether I ought to be talking about such things at all.
(CV 74)
He accused himself of asking ‘countless irrelevant questions’ (CV 77) and of making repetitions that others might find boring (CV 3); and in a draft of a preface to the Investigations, from 1948, he described the movements of his remarks thus:
Only every so often does one of the sentences I am writing here make a step forward; the rest are like the snipping of the barber’s scissors, which he has to keep in motion so as to be able to make a cut with them at the right moment.
(CV 76)
In a lecture in 1939, having compared philosophical problems to the difficulty of finding one’s way in an unfamiliar town (cf. PI §123), and likening his own task to that of a guide, he adds:
In order to be a good guide, one should show people the main streets first. But I am an extremely bad guide, and am apt to be led astray by little places of interest, and to dash down side streets before I have shown you the main streets.
(LFM 44)
Second, Wittgenstein was not particularly concerned to make himself generally understood. For one thing, he loathed the idea of giving results to people who had not done any honest work to deserve them. He despised popular science: what he thought of as attempts to give people the impression of understanding something that they did not really understand (RW 117). And it is easy to imagine the disgust which he would have suffered at seeing publications like Wittgenstein in 90 Minutes. In the Preface to the Investigations he announced:
I should not like my writing to spare other people the trouble of thinking.
(PI, p. viii)17
And in a remark from 1948 he enjoins himself: ‘Anything the reader can do for himself, leave it to the reader’ (CV 88). No doubt Wittgenstein did a good job of putting his readers on their mettle. Apart from the fact that no synoptic accounts are given of the targets of his remarks, almost every section leaves something to be pursued further by the reader.18
More importantly, Wittgenstein’s main concern was to solve his own philosophical problems; to achieve clarity for himself. The dialogues in the Investigations are quite authentic in that the interlocutor, far from being an invented straw man, gives expression to what Wittgenstein himself, at some point, felt inclined to say. As he remarked in 1948:
Almost the whole time I am writing conversations with myself. Things I say to myself tête-à-tête.
(CV 88)
In many instances it is views held by Wittgenstein when he wrote the Tractatus that are presented and responded to in the Investigations (e.g. PI §§23, 39ff., 65, 108, 114). He was fully aware that those who did not happen to have wrestled with very similar problems would find it extremely difficult to understand his remarks (cf. TLP, Preface). About the Blue Book, which is far less aphoristic and hence much easier to read than the Investigations, Wittgenstein warned Russell (not exactly a tiro in philosophy) that he thought it very difficult to understand ‘as so many points are just hinted at’ (CL 296). Clearly, he must have been aware that that was true in even greater measure of his mature work. Indeed, in the Preface he reckons that:
It is not impossible that it should fall to the lot of this work … to bring light into one brain or another–but, of course, it is not likely.
(PI, p. viii)
He certainly hoped to be read and appreciated, but only by a small number of kindred spirits, ‘a few friends spread out over the world’ (CV 9; cf. 12f.); those whose style of thinking was sufficiently akin to his own to enable them to understand what must remain obscure to others:
For if a book has been written for only a few readers that will be clear just from the fact that only a few understand it. The book must automatically separate those who understand it & those who do not.
(CV 10)
By no means did he address himself to the community of academic philosophers, the ‘philosophical journalists’ (CV 75), those who wrote for, and enjoyed reading, as ‘impotent and bankrupt’ a periodical as Mind.19
Anyway, Wittgenstein did not think that the study of philosophy should be encouraged. ‘I don’t recommend it’, he said to O.K. Bouwsma, ‘It’s for people who cannot leave it alone’ (WC 68). Wittgenstein’s own answer to Ryle’s question ‘What has a fly lost who never got into a fly-bottle?’ (cf. PI §309) was a resounding ‘Nothing’. Far from trying to persuade anyone to study philosophy, he would strongly advise his best students against pursuing it any further.20
Finally (and perhaps most importantly), Wittgenstein believed that a philosopher should be a poet (MS 120, p. 145r).
I believe I summed up my attitude towards philosophy by saying: really philosophy should only be written as poetry.
(CV 28)21
This is not to be dismissed as merely a somewhat purple way of saying that philosophy should be presented in a neat and elegant form. That Wittgenstein was serious about the idea of philosophy as a kind of poetry is indicated by the solemn way the remark is prefaced. What exactly does it mean?–To begin with, Wittgenstein was obviously concerned about the right flow of language, its quasi musical aspects. The way he constantly revised his remarks shows that he came as close as probably no other philosopher to Nietzsche’s maxim: that one should work on a page of prose as scrupulously and painstakingly as a sculptor works on a statue (Human, All Too Human, vol. II, 2. §95).22
One poetic ideal that appealed strongly to Wittgenstein was naturalness, the avoidance of all mannerisms, and false pathos.23 On that account he admired the unpretentious prose of Gottfried Keller, Hebel and Tolstoy; but disliked the preciousness of Rilke (WC 72). A mark of great poetry (of the cultural tradition Wittgenstein appreciated) is the appearance of perfect ease and naturalness in spite of all constraints imposed by form (rhythm and rhyme). Something analogous, it seems to me, is aimed at and often achieved in the Philosophical Investigations: what is difficult and highly complicated is to appear light and effortless.24 The constraints that Wittgenstein had to struggle with, while trying to give the impression of a natural succession of seemingly simple observations, were of course not rhythm and rhyme, but his task of dissolving recalcitrant philosophical problems and tracing their interconnections. But just as the choice of words in a poem should not appear to be dictated by rhythm and rhyme (although largely it is), it seems that Wittgenstein preferred to give the impression that one remark casually led on to the next, while their real purpose: the way they relate to a network of philosophical confusions was left largely unexplained. ‘If this book is written as it should be written’, he once remarked, ‘then everything I say must be easily understandable, indeed, trivial, but it should be difficult to understand why I say it’ (MS 117, pp. 140f.).25 Wittgenstein preferred not to push the point of his remarks under the reader’s nose. The light touch he aimed at in his writing precluded him from giving pedantic circumstantial explanations of all his respective targets and their interconnections.
Not only is the point of Wittgenstein’s remarks often left unexplained, many of the remarks themselves have the terseness and density that are typical of poetry, and leave so much to say and write to readers and scholars. We already came across that brilliant §302, which gives the outlines of a thought weighty enough to be spelled out at the length of a substantial journal article.
Earlier I mentioned how peculiar it was that although Wittgenstein constantly polished and reshuffled his remarks, he never tried to fuse them together by inserting smooth transitions from one to the next. He insisted on emphatic pauses between his remarks, which of course is a formal device closely related to the division of a long poetic text into stanzas.
Finally, Wittgenstein had a poet’s taste for striking figures of speech. ‘I always take joy in my own good similes’ (D 144), he once wrote in his diary.
Sometimes the same point is made by a whole series of similes: looking up a table in the imagination (PI §265); buying several copies of today’s paper (PI §265); different ways of looking at a clock to determine the time (PI §266); testing the stability of a bridge in the imagination (PI §267). Are all of them necessary to get the point across, or is it not also artistic enjoyment that accounts for such variations on the same theme?
The peculiarities of Wittgenstein’s style that can be annoying from an academic philosopher’s point of view, are largely those that make his book aesthetically so attractive and that add to Wittgenstein scholarship some of the enjoyment of the interpretation of poetry.
1. The German word ‘übersichtlich’ and its cognates are notoriously difficult to translate consistently into English. In lectures Wittgenstein used the word ‘synoptic’ (AWL 43), and for the most part I shall follow his example.
2. Note that PI §122, the remark about the importance of a synoptic representation, originates in Wittgenstein’s discussion of Frazer’s Golden Bough in MS 110, p. 257 (PO, p. 132), where it is preceded and followed by remarks on Goethe’s morphological method of setting things in juxtaposition rather than hypothesizing about their development.
3. And the lists that are sometimes quoted from RPP II §§63 and 148 (= Z§§472, 483, 621, 488–92) are mere plans for the treatment of that area, not intended to be offered to the reader in this form.
4. MS 121, p. 18: ‘we describe [the rôle of a word] only so far as it is necessary for the clarification of philosophical problems (wir beschreiben [die Rolle eines Wortes] nur so weit, als es nötig ist, philosophische Probleme zu klären)’. Note, incidentally, that Wittgenstein does not normally speak of rules as the object of a synoptic representation: in PI §122 it is grammar, in other writings it is the use of words, concepts or linguistic facts that are to be represented in a synoptic manner.
5. ‘“How do sentences manage to represent?” … Don’t you know? You can see it when you use them … For nothing is concealed’ (PI §435).
6. ‘You approach from one side and know your way about; you approach the same place from another side and no longer know your way about’ (PI §203).
7. On 8 September 1946 (MS 131, pp. 216f.), he expressed exactly the worry I just mentioned: But isn’t it possible after all?–:
But for feelings of bodily movements and the like isn’t there a kind of private ostensive definition after all? I bend a finger, for example, and remember the sensation. … Now could I not, for my private use, call this sensation ‘S’, use my memory as criterion of identity, and now say ‘Yes, that is S again’etc. … Of course being the same here means seeming to be the same.
(RPP I §§393, 395)
(Note that this is not just another draft of §258. Rather, the same philosophical question resurfaced in a different context.) Then, once more, he managed to reassure himself that the lack of a criterion was indeed an insuperable objection: ‘No, this difficulty is not an affectation’(RPP I §396). On the following day, again, the rejection of the scenario was qualified by considerations similar to those of §270 of the Investigations:
Such utterances could be important; if for example we observe certain physiological correlates of the sensations.
(RPP I §399)
8. There is good evidence that Wittgenstein himself regarded the Augustinian picture as the ultimate source of the private language confusions: i The reader is reminded of it in §264 of the Investigations; and its application to sensation language is rejected in §293 and §304. ii The text of §257, right at the core of the Private Language Discussion, was originally (in MS 115, p. 91) a continuation of the first paragraph of §27, where Wittgenstein lists some one-word exclamations and then asks rhetorically: ‘Are you inclined still to call these words “names of objects”?’ iii Manuscript 166, entitled ‘Notes for the Philosophical Lecture’, contains three drafts of a lecture on the private language problems; each one begins with the Augustinian picture:
Meaning consisting of the word referring to an object. … As introduction:/Word referring to an object. … Common idea: a word has meaning by referring to something….
(PO 447, 451, 454)
9. Provided it is not taken to imply (AP.n).
10. Cf. BT 209v: ‘Language consists in naming objects, namely: people, species, colours, pains, moods, numbers, etc. [Sprache besteht darin, daß man Gegenstände benennt & zwar: Menschen, Gattungen, Farben, Schmerzen, Stimmungen, Zahlen, etc.]’.
11. ‘Behandle die deutlichen Fälle in der Philosophie, nicht die undeutlichen. Diese werden sich lösen, wenn jene gelöst sind.’
12. A more detailed, and probably more influential, account and explanation of this argument was given by Peter Strawson (1959: ch. 3).
13. Also in the case of pain Wittgenstein once used that slightly exaggerated expression (‘no “right”’ instead of ‘no error’): ‘One cannot “be right (or wrong)” in saying: “I am in pain” [Es kann Einer nicht “Recht (oder Unrecht) haben”, wenn er sagt: “ich habe Schmerzen”]’ (MS 121, p. 5). Cf. MS 159, p. 11v: ‘“You are immediately aware” makes one think you are right about something, that you can be shown to be wrong about; whereas the point is that there is no right (or wrong) about it. (And of course no one would say: I’m sure I’m right that I have pain.)’
14. Here, again, it is important ‘not to represent the matter as if there were something one could not do’ (cf. PI §374). Of course, it is possible to have a sensation without a natural expression for it and then write down the sign ‘S’. The question is whether we should accept that as a piece of sensation language.
15. Cf. MS 119, p. 275: ‘But what is the point of language? Is it to accompany experiences with noise? (Wozu dient denn die Sprache? Dazu die Erlebnisse mit Lärm zu begleiten?)’
16. Cf. Schroeder 1998, §§42–7.
17. Cf. MS 119, p. 64.
18. Consider, for example: ‘
This body has extension.’ To this we might reply: ‘Nonsense!’–but are inclined to reply ‘Of course!’–Why is this?
(PI §252)
That is not just a rhetorical question. Or:
Think of the recognition of facial expressions. … Think, too, how one can imitate a man’s face without seeing one’s own in a mirror.
(PI §285)
It is less than self-evident what exactly one is to think of here and how these matters are supposed to help the discussion in hand.
19. Malcolm 1984: 100; cf. also 107.
20. Cf. also Malcolm 1984: 28.
21. Cf. MS 115, p. 30; MS 146, pp. 32 and 50; MS 133, p. 13r.
22. In MS 120, p. 145r Wittgenstein compares his aesthetic attitude towards philosophy to Nietzsche’s.
23. CV 10: ‘Everything ritualistic (everything that, as it were, smacks of the high priest) is strictly to be avoided because it straightaway turns rotten.’
24. An unintentional tribute to Wittgenstein’s success in his attempt to make his writing appear light and effortless is Russell’s notorious comment that in the Investigations Wittgenstein seemed to have grown ‘tired of serious thinking’ (Russell 1959: 161).
25. ‘Wenn dieses Buch geschrieben ist, wie es geschrieben sein sollte, so muss, was ich sage, alles leicht verständlich, ja trivial sein, schwer verständlich aber, warum ich es sage.’
Baker, G.P. and Hacker, P.M.S. (1980) Wittgenstein: Understanding and Meaning. An Analytical Commentary on the ‘Philosophical Investigations’, vol. 1, Oxford: Blackwell.
Glock, Hans-Johann (1996) A Wittgenstein Dictionary, Oxford: Blackwell.
Hacker, P.M.S. (1990) Wittgenstein: Meaning and Mind, vol. 3 of An Analytical Commentary on the ‘Philosophical Investigations’, Oxford: Blackwell.
Hanfling, Oswald (2002) Wittgenstein and the Human Form of Life, London: Routledge.
Malcolm, Norman (1984) Ludwig Wittgenstein: A Memoir, 2nd edn, Oxford: Oxford University Press. Monk, Ray (1990) Wittgenstein: The Duty of Genius, London: Cape.
Russell, Bertrand (1959) My Philosophical Development, London: Unwin.
Schroeder, Severin (1998) Das Privatsprachen-Argument. Wittgenstein über Empfindung und Ausdruck, Paderborn: Schöningh.
Strawson, P.F. (1959) Individuals. An Essay in Descriptive Metaphysics, London: Methuen.