A-plus Writer

I

111

72 '* MUST WE MEAN WHAT WE its own risks of dishonesty and hostility. But I do not see that th faults of explicit discipleship are which come from subjection to whose origins are unseen or create a different blindness inaccessible in other ways to cure. Be­ tween control by the living and control by the dead there is nothing to choose.

Because the breaking of such control is a constant purpose of the later Wittgenstein, his writing is deeply practical and negative, the way Freud's is. And like Freud's therapy, it wishes to prevent under­ standing which is unaccompanied by inner change. Both of them are intent upon unmasking the defeat of our real need in the face of self-impositions which we have not assessed (§108), or fantasies ("pic­ tures") which we cannot escape (§l1S). In both, such misfortune i. betrayed in the incongruence between what is said and what is meant or expressed; for both, the self is concealed in assertion and action and revealed in temptation and wish. Both thought of their negative soundings as revolutionary extensions of our knowledge, and both were obsessed by the idea, or fact, that they would be misunderstood -partly, doubtless, because they knew the taste of self-knowledge, that it is bitter. It will be time to blame them for taking misunder­ standing by their disci pIes as personal betrayal when we know that the ignorance of oneself is a refusal to know.

SAY?

more dangerous than the faul u modes of thought and sensibility

unremembered and which therefore III

Aesthetic Problems

of Modern Philosophy

rhe Spirit of the Age is not easy to place, ontologicaHy or empirically; Mid it is idle to suggest that creative effort must express its age, either hecause that cannot fail to happen, or because a new effort can create

new age. Still, one knows what it means when an art historian says, I hinking of the succession of plastic styles, "not everything is possible

every period." 1 And that is equally true for every person and ('very philosophy. But then one is never sure what is possible until it happens; and when it happens it may produce a sense of revolution, of the past escaped and our problems solved-even when we also know that one man's solution is another man's problem.

Wittgenstein expressed his sense both of the revolutionary break his later methods descry in philosophy, and of their relation to meth­ ods in aesthetics and ethics.2 I have tried, in what follows, to suggest ways in which such feelings or claims can be understood, believing lhem to be essential in understanding Wittgenstein's later philosophy as a whole. The opening section outlines two problems in aesthetics 'ach of which seems to yield to the possibilities of Wittgensteinian procedures, and in turn to illuminate them. The concluding section

1 Heinrich WolfHin, Principles of Art History. foreword to the 7th German edition. Quoted by E. H. Gombrich, Art and Illusion (New York: The Bollingen Series. Pan· theon Press. Ig60). p . 4·

• Reported by G. E. Moore, "Wittgenstein's Lectures in Ig3()-1I3," reprinted in Moore's Philosophical Papers (London: George Allen and Unwin. 1959). p. 315.

73

WE SAY?

suggests resemblances between one kind of judgment recognizable a aesthetic and the characteristic claim of Wittgenstein-and of ordi·

we should

suppose the way I have written, grows from a sense that philosophy is in one of its periodic crises o(

am sure is not mine alone, that method dictates to content; that, for example, an intellectual com· mitment to analytical philosophy trains concern away from the wider, traditional problems of human culture which may have brought one to philosophy in the first place. Yet one can find oneself unable to

A free eclecticism of method is one obvious solution to such a problem. Another solution may be to discover further freedoms or possibilities within the method one finds closest to oneself. I lean here towards the latter of these alternatives, hoping to make philos· ophy yet another kind of problem for itself; in particular, to make the medium of philosophy-that is, of Wittgensteinian and, more generally, of ordinary language philosophy-a significant problem

TWO PROBLEMS OF AESTHETICS

sheer matter of words-the controversy about whether a poem, or more modesdy, a metaphor, can be pata­ phrased. Cleanth Brooks, in his Well Wrought Urn,s provided a convenient title for it in the expression "The Heresy of Paraphrase," the heresy, namely, of supposing that a "poem constitutes a 'state­ ment' of some sort" (p. 179); a heresy in which "most of our difficul­

The truth of the matter is that all such formulations (of what a poem says) lead away from the center of the poem-not toward it: that the "prose sense" of the poem is not a rack on which the stuff of the poem is hung; that it does not represent the "inner" structure or the "essential" structure or the "real" structure of the poem (p.

(New York: Harcourt. Brace Be Co., 1947). AlI page references to Brooks are to this edition. "The Heresy of Paraphrase" is the title of the

74 '* MUST WE MEAN WHAT

nary language philosophers generally-to voice "what ordinarily say."

What I have written, and I

method, heightened by a worry I

relinquish either the method or the alien concern.

for aesthetics.

Let us begin with a

ties in criticism are rooted" (p. 184).

• The Well Wrought Urn

concluding chapter.

AESTHETIC l'ROBLEMS OF MODERN PHILOSOPHY '* 75 182). \Ve can very properly use paraphrases as pointers and as short­ hand references provided that we know what we are doing. But it is highly important that we know what we are doing and that we see plainly that the paraphrase is not the real core of meaning which constitutes the essence of the poem (p. 180).

We.: may have some trouble in seeing plainly that the paraphrase is .1/ the real core, or essence, or essential structure or inner or real

.fl ilcture of a poem; the same trouble we should have in understand­ "f( what is any or all of these things, since it takes so much philosophy lI'l to state them. It is hard to imagine that someone has just flatly Iven it out that the essence, core, structure, and the rest, of a poem

iLS paraphrase. Probably somebody has been saying that poetry uses 'fil aments of style, or requires special poetic words; or has been say­

IIlg what a poem means, or what it ought to mean-doing something lin t makes someone else, in a fit of philosophy, say that this is distort­

fUg a poem's essence. Now the person who is accused in Brooks' writ II probably going to deny guilt, feel that words are being put into his II louth, and answer that he knows perfectly well that a "paraphrase,

course, is not the equivalent of a poem; a poem is more than its I' .• raphrasable content." Those are the words of Yvor Winters, whose ,...ork Professor Brooks uses as "[furnishing] perhaps the most respect­ hIe example of the paraphrastic heresy" (p. 183).4 And so the argu­

Il1 cnt goes, and goes. It has the gait of a false issue-by which I do not Hlean that it will be easy to straighten out.

One clear symptom of this is Brooks' recurrent concessions that, If course, a paraphrase is all right-if you know what you're doing. Which is about like saying that of course criticism is all right, in its !,Iace; which is true enough. But how, in particular, are we to assess .1 critic's reading the opening stanza of Wordsworth's "Intimations"

de and writing: ". . . the poet begins by saying that he has lost omething" (Brooks, p. 116)? We can ransack that stanza and never

li nd the expression "lost something" in it. Then the critic will be ,,!Eended-rightly-and he may reply: Well, it does not actually say •his, but it means it, it implies it; do you suggest that it does not mean .hat? And of course we do not. But then the critic has a theory about

(For Winters' posltlon. I have relied solely on his central essay, "The Exped. lIIenral School in American Poetry." from PrimitirJ;-sm and Decadence, itself republished. logcther with earlier of his critical works. HOller the title III Defense of Reason (Denver:

Ian Swallow, 1917).

76 '* MUST WE MEAN WHAT WE SAY? what he is doing when he says what a poem means, and so he will have to add some appendices to his readings of the poetry explaining that when he says what a poem means he does not say exactly quite just what the poem means; that is, he only points to its meaning, or rather "points to the area in which the meaning lies." But even this last does not seem to him humility enough, and he may be moved to a footnote in which he says that his own analyses are "at best crude approximations of the poem" (p. 189). By this time someone is likely to burst out with: But of course a paraphrase says what the poem says, and an approximate paraphrase is merely a bad paraphrase; with greater effort or sensibility you could have got it exactly right. To which one response would be: "Oh, I can tell you exactly what the Ode means," and then read the Ode aloud.

Is there no real way out of this air of self-defeat, no way to get satisfying answers? Can we discover what, in such an exchange, is causing that uneasy sense that the speakers are talking past one another? Surely each knows exactly what the other means; neither is pointing to the smallest fact that the other fails to see.

For one suggestion, look again at Brooks' temptation to say that his readings approximate to (the meaning of) the poem. He is not there confessing his personal ineptitude; he means that any para· phrase, the best, will be only an approximation. So he is not saying, what he was accused of saying. that his own paraphrase was, in some more or less definite way, inexact or faulty: he denies the ordinary contrast between "approximate" and "exact." And can he not do that if he wants to? Well, if I am right, he did do it. Although it is not clear that he wanted to. Perhaps he was led to it; and did he realize that. and would his realizing it make any difference? It may help to say: In speaking of the paraphrase as approximating to the poem (the meaning of the poem?) he himself furthers the suggestion that para­ phrase and poem operate. as it were, at the same level, are the same kind of thing. (One shade of color approximates to another shade. it does not approximate. nor does it fail to approximate, to the object of which it is the color. An arrow pointing approximately north is exactly pointing somewhere. One paraphrase may be approximately the same, have approximately the same meaning, as another .para­ phrase.) And then he has to do everything at his philosophical disposal to keep paraphrase and poem from coinciding; in particular, speak of cores and essences and structures of the poem that are not reached by

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY *' 77 fhe paraphrase. It is as if someone got it into his head that really "linting to an object woulci require actually touching it, and then.

.lIizing that this would make life very inconvenient. reconciled him­ \( to common sense by saying: Of course we can point to objects, but

he must realize what we are doing, and that most of the time this only approximately pointing to them.

This is the sort of thing that happens with astonishing £re­ 'l "eDcy in philosophy. We impose a demand for absoluteness (typi- I .lI ly of some simple physical kind) upon a concept, and then, finding. ( , I.at our ordinary use of this concept does not meet our demand, we ( rcommodate this discrepancy as nearly as possible. Take these famil·

" Ir patterns: we do not really see material objects, but only see them IIldirectly; we cannot be certain of any empirical proposition, but Illy practically certain; we cannot really know what another person

., feeling, but only infer it. One of Wittgenstein's greatest services, to lilYmind, is to show how constant a feature of philosophy this pattern j,: this is something that his diagnoses are meant to explain ("We have a certain picture of how something must be"; "Language is !tiling; not doing work; being used apart from its ordinary language .Imes"). Whether his diagnoses are themselves satisfying is another

question. It is not very likely, because if the phenomenon is as com­ Illon as he seems to have shown, its explanation will evidently have to lie very much clearer and more complete than his sketches provide.

This much, however, is true: If you put such phrases as "giving I he meaning," "giving a paraphrase," "saying exactly what something means (or what somebody said)," and so on, into the ordinary con­ texts (the "language games") in which they are used, you will not find that you are worried that you have not really done these things. \Ve could say: That is what doing them really is. Only that serenity will last just so long as someone does not start philosophizing about it. Not that I want to stop him; only I want to know what it is he is then doing, and why he follows just those particular tracks.

We owe it to Winters to make it clear that he does not say any ,f the philosophical things Brooks attributes to him. His thesis. hav­ ng expressed his total acquiescence in the fact that paraphrases are

not poems, is that some poems cannot be paraphrased-in particular, poems of the chief poetic talent of the United States during the sec­ ond and third decades of the twentieth century; that poems which are unparaphrasable are, in that specific way, defective; and that there­

78 .. MUST WE MEAN WHAT WE SAY?

fore this poetic talent was led in regrettable directions. The merit this argument for us, whether we agree with its animus or not, an trying to keep special theories about poetic discourse at arm's length, is its recognition that paraphrasability is one definite characteristic uses of language, a characteristic that some expressions have and som do not have. It suggests itself that uses of language can be distin guished according to whether or not they possess this characteristi and further distinguished by the kind of paraphrase they demand. us pursue this suggestion with a few examples, following Wittgen stein's idea that we can find out what kind of object anything (gra matically) is (for example, a meaning) by investigating expression which show the kind of thing said about it (for example, "explainin the meaning").

It is worth saying that the clearest case of a use of languag having no paraphrase is its literal use. If I tell you, "Juliet [the girl next door] is not yet fourteen years old," and you ask me what I mean, I might do many things-ask you what you mean, or perhaps try t teach you the meaning of some expression you cannot yet use (which, as Wittgenstein goes to extraordinary lengths to show, is not the sam thing as telling you what it means). Or again, if I say, "Sufficient unto the day is the evil thereof," which I take to be the literal truth, then if I need to explain my meaning to you I shall need to do other things:

. I shall perhaps not be surprised that you do not get my meaning and so I shall hardly ask you, in my former spirit, what you mean in asking me for it; nor shall I, unless my disappointment pricks me into offense, offer to teach you the meaning of an English expression. What I might do is to try to put my thought another way, and perhapt refer you, depending upon who you are, to a range of similar or identical thoughts expressed by others. What I cannot (logically) do in either the first or the second case is to paraphrase what I said.

Now suppose I am asked what someone means who says, "Juliet is the sun." Again my options are different, and specific. Again I am not, not in the same way, surprised that you ask; but I shall not try to put the thought another way-which seems to be the whole truth in the view that metaphors are unparaphrasable, that their meaning il bound up in the very words they employ. (The addition adds nothing: Where else is it imagined, in that context, that meanings are bound, or found?) I may say something like: Romeo means that Juliet is the warmth of his world; that his day begins with her; that only in her

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY "* 79 II ishment can he grow. And his declaration suggests that the moon, , II other lovers use as emblems of their love, is merely her reflected It , and dead in comparison; and so on. In a word, I paraphrase it. ,,('over, if I could not provide an explanation of this form, then I is a very good reason, a perfect reason, for supposing that I do

I know what it means. ~able. (And if that ~ I lie, it is tautologous.) When Croce denied the possibility of para- I.,~e, he at least had the grace to assert that there were no meta­ ,rs.

Two points now emerge: (1) The "and so on" which ends my wple of paraphrase is significant. It registers what William Emp·

II l;a11s the "pregnancy" of metaphors, the burgeoning of meaning I hem. Call it what you like; in this feature metaphors differ from t1C, but perhaps not all, literal discourse. And differ from the simi­

I device of simile: the inclusion of "like" in an expression changes \. rhetoric. If you say "Juliet is like the sun," two alterations at kast I' m obvious: the drive of it leads me to expect you to continue by

'Y"lg in what definite respects they are like (similes are just a little t pregnant); and, in complement, I wait for you to tell me what 1Il mean, to deliver your meaning, so to speak. It is not up to me to lid as much as I can in your words. The over-reading of metaphors I often complained of, no doubt justly, is a hazard they must run

'I'll their high interest.· (2) To give the paraphrase, to understand the m'laphor, I must understand the ordinary or dictionary meaning of I;hc words it contains, and understand that they are not there being '1~C'd in their ordinary way, that the meanings they invite are not to

found opposite them in a dictionary. In this respect the words in lIIe taphors function as they do in idioms. But idioms are, again, specif­ 1l.,lly different. "I fell flat on my face" seems an appropriate case. To ' plain its meaning is simply to tell it-one might say you don't :.: plain it at all; either you know what it means or you don't; there is

110 richer and poorer among its explanations; you need imagine noth­

• [Added 1968. I should have made it more explicit that throughout this essay Illn using "paraphrase" to name solely that specific fonn of account which suits

n~laphors (marked, for example, by its concluding sense of "and so on''). So when I say Il lII t stretches of literal prose "cannot be paraphrased," I mean to imply the specilka­ ,Ion "•.. in that way." Certainly an exercise useful in the teaching of reading can I.e given as "Paraphrase the following passage," where what is wanted is a resume "I the passage which shows a grasp of the difficult words and constructions in it and I1f Its over· all sense. Dut in that context, paraphrase IS explicitly not a candidate for anything likely to be taken as a competitor of the passage In question.]

80 '* MUST WE MEAN WHAT WE SAY? ing special in the mind of the person using it. And you will find it a dictionary, though in special locations; which suggests that, unli metaphors, the number of idioms in a language is finite. In so though not all, of these respects the procedure of "giving the me ing" of an idiom is like that in translating: one might think of it translating from a given language into itself. Then how is it differ from defining, or giving a synonym?

One final remark about the difference between idioms and me phors. Any theory concerned to account for peculiarities of metaph of the sort I have listed will wonder over the literal meaning words, in that combination, have. This is a response, I take it, to t fact that a metaphorical expression (in the "A is B" form at lea sounds like an ordinary assertion, though perhaps not made by ordinary mind. Theory aside, I want to look at the suggestion, oft made, that what metaphors literally say is false. (This is a response the well-marked characteristic of "psychic tension" set up in me phors. The mark is used by Empson; I do not know the patent.) Bu to say that Juliet is the sun is not to say something false; it is, at be wildly false, and that is not being just false. This is part of the fa that if we are to suggest that what the metaphor says is true, we shall have to say it is wildly true-mythically or magically or primitivel true. (Romeo just may be young enough, or crazed or heretic enough to have meant his words literally.) About some idioms, however, it i fair to say that their words literally say something that is quite false: something, that is, which could easily, though maybe comically, imagined to be true. Someone might actually fall fiat on his face, hay a thorn in his side, a bee in his bonnet, a bug in his ear, or a fiy in hi ointment-even all at once. Then what are we to say about the literal meaning of a metaphor? That it has none? And that what it literally says is .not false, and not true? And that it is not an assertion? But it sounds like one; and people do think it is true and people do think ie is false. I am suggesting that it is such facts that will need investigatinS if we are to satisfy ourselves about metaphors; that we are going to keep getting philosophical theories about metaphor until such fact are investigated; and that this is not an occasion for adjudication, for the only thing we could offer now in that line would be: all the theories are right in what they say~ And that seems to imply that all are wrong as well.

At this point we might be able to give more content to the idea

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY *' 81 I some modes of figurative language are such that in them what " pression means cannot be said at all, at least not in any of the

I!' or less familiar, conventionalized ways so far noticed. Not be­ these modes are flatly literal-there is, as it were, room for an

"I..nation, but we cannot enter it. About such an expression it may light to say: I know what it means but I can't say what it means.

111\ this would no longer suggest, as it would if said about a meta­ Ir, that you really do not know what it means-or: it might sug­ t it, but you couldn't be sure.

Examples of such uses of language would, I think, characteristi­ Ily occur in specific kinds of poetry, for example Symbolist, Surreal- or Imagist. Such a use seems to me present in a line like Hart

IlIIe's "The mind is brushed by sparrow wings" (cited, among hers, in the Winters essay), and in Wallace Stevens' "as a calm I kens among water-lights," from "Sunday Morning." Paraphrasing

Ie" lines, or explaining their meaning, or telling it, or putting the ,,'light another way-all these are out of the question. One may be hIe to say nothing except that a feeling has been voiced by a kindred I'lr it and that if someone does not get it he is not in one's world, or IIII of one's flesh. The lines may, that is, be left as touchstones of " imacy. Or one might try describing more or less elaborately a par~

If 1Iiar day or evening. a certain place and mood and gesture, in hose presence the line in question comes to seem a natural expres­

'lPn, the only expression. This seems to be what Winters, who profitably distinguishes

u'veral varieties of such uses of language. distrusts and dislikes in his 1d ense of reason. as he also seems prepared for the reply that this is Illlt a failing of language but a feature of a specific approach of lan­ \lage. At least I think it is a reply of this sort. which I believe to be

t1ght, that he wishes to repudiate by appealing to "the fallacy of ex­ pressive (or imitative) form." instanced by him at one point as "Whit­ man trying to express a loose America by writing loose poetry." or 'Mr. Joyce [endeavoring] to express disintegration by breaking down his form." It is useful to have a name for this fallacy. which no doubt

me people commit. But his remarks seem a bit quick in their nota­ tion of what Whitman and Joyce were trying to express, and in their e"xplanation of why they had to express theinselves as they did; too ure that a break with the past of the order represented in modern art was not itself necessary in ' order to defend reason; too sure that con­

'* 82 MUST WE MEAN WHAT WE SAY? vention can still be attacked in conventional ways. And they suggest scorn for the position that a high task of art has become, in our bombardment of sound, to create silence. (Being silent for that pur· pose might be a good example of the fallacy of imitative form. But that would depend on the context.) The fact is that I feel I would have to forgo too much of modem art were I to take his view of it.

Before we leave him, we owe it to Brooks to acknowledge feature of Winters' position which may be causing his antipathy to it. Having wished to save Winters from a misconstruction of paraphrase. we gave back to that notion a specificity which, it now emerges, open him to further objection. For his claim that poems that cannot be paraphrased--or, as he also puts it, do not "rest on a formulabl logic"-are therefore defective now means or implies that all poem not made essentially of metaphorical language (and/or similes, idioms, literal statements) are defective. It is certainly to be hoped that all criticism be rational, to be demanded that it form coheren t propositions about its art. But to suppose that this requires all poetry to be "formulable," in the sense that it must, whatever its form and pressure, yield to paraphrase, the way single metaphors specifically do. is not only unreasonable past defense but incurs what we might call the fallacy of expressive criticism.

In summary: Brooks is wrong to say that poems cannot in prin. ciple be fully paraphrased, but right to be worried about the relation between paraphrase and poem; Winters is right in his perception thal some poetry is "formulable" and some not, but wrong in the assur· ance he draws from that fact; both respond to, but fail to follow, the relation between criticism and its object. And now, I think, we can be brought more unprotectedly to face the whole question that motivates such a conflict, namely what it is we are doing when we describe or explain a work of art; what function criticism serves; whether different arts, or forms of art, require different forms of criticism; what we may expect to learn from criticism, both about a particular piece of art and about the nature of art generally.

The second problem in aesthetics must be sketched even more swiftly and crudely.

Is such music as is called "atonal" (not distinguishing that, for our purposes now, from the term "twelve-tone") really without tonality? (The little I will say could be paralleled, I think, in discussing the nature of the painting or sculpture called abstract or non-objective.)

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY '* 83 I he arguments are bitter and, to my knowledge, without issue; and I"\oy musicians have felt within themselves both an affirmative and lIegative answer.~ Against the idea that this music lacks tonality are I.) the theory that we are so trained to our perception of musical rganization that we cannot help hearing it in a tonal frame of refer­

, flce; and (2) the fact that one can, often, say what key a so·called Itonal" piece is in. In favor of the idea that it lacks tonality are (1) a

Iheory of composition which says that it does, and whose point was Just to escape that limitation, while yet maintaining coherence; and

) the fact that it simply sounds so different. Without our now even boeing at the theories, let us look at the fact we recorded as "being ule to say, often, what key a piece is in." Does that have the weight

II seems to have? An instance which once convinced me of its deci­ lveness was this: in listening to a song of Schoenberg'S, I had a clear :nse that I could, at three points, hear it cadence (I almost said, try

III resolve) in F# minor. Then surely it is in F# minor? Well, the ' :hopin Barcarolle is in the key of F# major. How do I know that? Because I can hear it try to cadence in F# major? Three or more li mes? And after that I am convinced it is, feel slightly relieved and r'ven triumphant that I have been able to hear some F# major? But tha t is absurd. I know the key; everyone knows it; everyone knows it Ir om the opening measure-well, at least before the bass figure that begins on the pitch of F#: it does not take a brick wall to fall on us. r would not even know how to go about doubting its key or trying to hear it in its key. And I know it because I know that now it has moved If) the subdominant of the key, and now the dominant of the key is heing extended, and now it is modulating, and now it is modulating I n a more distant key. And to know all this is to know the grammar IJf the expression "musical key." Sometimes, to be sure, a solidly tonal 10m poser ,,,ill, especially in "development sections," obliterate the

nse of placement in a key; but this is here a special effect, and de­ pends upon an undoubted establishment of key. So if I insist upon aying that atonal music is really tonal (and to be said it has to be

insisted upon) I have, so far as my ear goes, to forgo the grammar of Ihe expression "tonality" or "musical key"-or almost all of it: I can tctain "almost cadences in" and "sounds like the dominant of," but lIot "related key," "distant key," "modulation" etc. And then I am in

• I am told, by Professor David Lewin . that this was true of Anton Webern, who was in doubt about his own music in this regard.

84 *' MUST WE MEAN WHAT WE SAY? danger of not knowing what I am saying. Wittgenstein says th " . .. the speaking of language is part of an activity, or of a form life" (Investigations, §23), and also "To imagine a language means imagine a form of life" (ibid., §19). The language of tonality is p" of a particular form of life, one containing the music we are m familiar with; associated with, or consisting of, particular ways being trained to perform it and to listen to it; involving particul ways of being corrected, particular ways of responding to mistak to nuance, above all to recurrence and to variation and modification No wonder we want to preserve the idea of tonality: to give all th up seems like giving up the idea of music altogether. I think it is· like it.

I shall not try to say why it is not fully that. I shall only mention that it cannot be enough to point to the obvious fact that musical instruments, with their familiar or unfamiliar powers, are employ -because that fact does not prevent us from asking: But is it music Nor enough to appeal to the fact that we can point to pitches, inte vals, lines and rhythm-because we probably do not for the most pa know what we are pointing to with these terms. I mean we do n know which lines are significant (try to play the "melody" or "bass" of a piece of Webern's) and which intervals to hear as organizin More important, I think, is the fact that we may see an undoubted musician speak about such things and behave toward them in wa similar (not, I think, more than similar) to the ways he behaves t ward, say, Beethoven, and then we may sense that, though similar, it is a new world and that to understand a new world it is imperative tl concentrate upon its inhabitants. (Of course there wiII be the usual consequences of mimicry and pretension.) Moreover, but still perha even more rarely, we may find ourselves within the experience

I \ such compositions, following them; and then the question whether ~ this is music and the problem of its tonal sense, will be-not answered

or solved, but rather they will disappear, seem irrelevant. That is, of course, Wittgenstein's sense of the way philosophical

problems end. It is true that for him, in the Investigations at an rate, this happens when we have gone through a process of bringin ourselves back into our natural forms of life, putting our souls bac into our bodies; whereas I had to describe the accommodation of th new music as one of naturalizing ourselves to a new form of life, new world. That a resolution of this sort is described as the solution

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY *' 85 philosophical problem, and as the goal of its particular mode of

111 icism, represents for me the most original contribution Wittgen­ IIl"in offers philosophy. I can think of no closer title for it,.in an

t;tblished philosophical vocabulary, than Hegel's use of the term II /hebung. We cannot translate the term: "cancelling," "negating," Il lfilling" etc. are all partial, and "sublate" transfers the problem. It r f'ns to me to capture. that sense of satisfaction in our representation rival positions which I was asking for when I rehearsed the prob­

illS of Brooks and Winters. Of course we are no longer very apt to IIl ppose, with Hegel, that History will make us a present of it: we are

aware of its brilliant ironies and its aborted revolutions for that. "lit as an ideal of (one kind of) philosophical criticism-a criticism in

hich it is pointless for one side to refute the other, because its cause nd topic is the self getting in its own way-it seems about right.

In the Tractatus Wittgenstein says: "The solution of the proh­ I, In of life is seen in the vanishing of the problem" (6.52 I); and in he Investigations he says: "... the clarity that we are aiming at is IIdeed complete clarity. But this simply means that the philosophical

I'loblems should completely disappear" (§133). Yet he calls these IIoblems solved (Investigations, ibid.); and he says that "... when 10 questions remain ... just that is the answer" (Tractatlls; 6.52, lIy emphasis). In the central concept of his later work, this would em to mean that the problems of life and the problems of philos­ pity have related grammars, because solutions to them both have

Ihe same form: their problems are solved only when they disappear, lid answers are arrived at only when there are no longer questions­

\~ h en, as it were, our accounts have cancelled them. But in the Investigations this turns out to be more of an answer

IlI:'m, left this way, it seems to be; for it more explicitly dictates and It'plays the ways philosophy is to proceed in investigating problems,

WH yS leading to what he calls "perspicuous representation" (iiber­ Ir:htliche Darstellllng). It is my impression that many philosophers If) not like Wittgenstein's comparing what he calls his "methods" to

cl l(;rapies (§ 133); but for me part of what he means by this comparison " brought out in thinking of the progress of psychoanalytic therapy. I'he more one learns, so to speak, the hang of oneself, and mounts lit e'S problems, the less one is able to say what one has learned; not

)Jecause you have forgotten what it was, but because nothing you said would seem like an answer or a solution: there is no longer any

86 *' MUST WE MEAN WHAT WE SAY? question or problem which your words would match. You hay reached conviction, but not about a proposition; and consistency, bu not in a theory. You are different, what you recognize as problems

~ \ different, your world is different. ("The world of the happy man i. different one from that of the unhappy man" (Tractatus; 6.43).) AncS this is the sense, the only sense, in which what a work of art mea: cannot be said. Believing it is seeing it.

When Wittgenstein says that "the concept of a perspicuous rep sentation .•. earmarks the form of account we give" (§12Sl), I tak him to be making a grammatical remark about what he calls a "gra matical investigation," which is what his Investigations consist i (§go): no other form of resolution will count as philosophical. H says of his "form of account" that it is "the way we look at thinga" and he then asks, parenthetically, "Is this a 'Weltanschauung' (§U2). The answer to that question is, I take it, not No. Not, pe haps, Yes; because it is not a special, or competing, way of looking things. But not No; because its mark of success is that the world se -be-different. As usual, the claim to severe philosophical advan entails a reconception of the subject, a specific sense of revolution.

AESTHETIC JUDGMENT AND A PHILOSOPHICAL CLAIM

Another good cause for stumbling over the procedures ordinary language philosophy lies in its characteristic appeal to wha "we" say and mean, or cannot or must say or mean. A good cau since it is a very particular, not to say peculiar appeal, and one woul expect philosophers dependent upon it themselves to be concern for its investigation. I will suggest that the aesthetic judgmen models the sort of claim entered by these philosophers, and that familiar lack of conclusiveness in aesthetic argument, rather tha. showing up an irrationality, shows the kind of rationality it has, an needs.

Hume is always a respectable place to begin. Near the middle his essay "Of the Standard of Taste," he has recourse to a story fro' Don Quixote which is to illustrate that "delicacy" of taste said to essential to those critics who are to form our standard of it.

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY '* 87 It is with good reason, says Sancho to the squire with the great nose, Ihat I pretend to have a judgment in wine: This is a quality heredi­ tary in ollr family. Two of my kinsmen were once caBed in to give their opinion of a hogshead, which was supposed to be excellent, being old and of a good vintage. One of them tastes it; considers it; and after mature reflection pronounces the wine to he good, were it not for a smali taste of leather, which he perceived in it. The other, after using the same precautions, gives also his verdict in favour of the wine; hut with the reserve of a taste of iron, which he could easily distinguish. You cannot imagine how much they were hath ridiculed for their judgment. But who laughed in the end? On emptying the hogshead, there was found at the bottom, an old key with a leathern thong tied to it.

ItU of all, the fine drama of this gesture is greater than its factual i ~iveness-a bit quixotic, so to say: for the taste may have been ~ent and the object not, or the object present and the taste not.

l ond, and more important, the gesture misrepresents the efforts of "ritic and the sort of vindication to which he aspires. It dissociates

It" exercise of taste from the discipline of accounting for it: but all I LL makes the critic's expression of taste worth more than another 1 111'S is his ability to produce for himself the thong and key of his J

ponse; and his vindication comes not from his pointing out that it r- Ill' was, in the barrel, but in getting us to taste it there. Sancho's

lu 'cstors, he tells us, in each case after the precautions of reflection, II h pronounced in favor of the wine; but he does not tell us what ,"se reflections were, nor whether they were vindicated in their vorable verdict. Hume's essay, I take it, undertakes to explore just

'I(' h questions, but in his understandable difficulty in directing us to It· genuine critic and distinguishing him from the pretender, he says

ut him just what he, or anyone, says about art itself: that he is Illable, that we may disagree about his merits in a particular case,

III that some, in the long run, "will be acknowledged by universal Ill iment to have a preference above others." But this seems to put \(. critic's worth at the mercy of the history of taste; whereas his lue to us is that he is able to make that history a part of his data,

Howing that in itself, as it stands, it proves nothing-except pop­ I,Irity. His value to art and culture is not that he agrees with its taste which would make him useful for guiding one's investments in the

,L market-but that he sets the terms in which our tastes, whatever Iry happen to be, may be protected or overcome. Sancho's descend­

c

88* MUST WE MEAN WHAT WE SAY?

ants would, by the eighteenth century, have risen to gentlemen, exercising distinction in a world which knew what was right, and not needing to make their tastes their own. But it is Quixote who is the

r patron saint of the critic, desperate to preserve the best of his culture against itself, and surviving any failure but that of his honesty and his expression of it.

The idea of the agreement or "reconciliation" of taste controll Hume's argument; it is agreement that the standard of taste is to provide, so far as that is attainable. Hume's descendants, catching the assumption that agreement provides the vindication of judgment, but no longer able to hope for either, have found that aesthetic (and moral and political) judgments lack something: the arguments that support them are not conclusive the way arguments in logic are, nor rational the way arguments in science are. Indeed they are not, and if they were there would be no such subject as art (or morality) and no such art as criticism. It does not follow, however, that such judgmenu are not conclusive and rational.

Let us turn to Kant on the subject, who is, here as elsewhere, deeper and obscurer. Universal agreement, or as he also calls it, the "harmony of sentiment" or "a common sense of mankind," makes it appearance in the Critique of Judgment not as an empirical problem -which is scarcely surprising about Kant's procedure-but as an a priori requirement setting the (transcendental) conditions under which such judgments as we call aesthetic could be made ilberhaupt. Kant begins by saying that aesthetic judgment is not "theoretical," not "logical," not "objective," but one "whose determining ground can be no other than subjective." 6 Today, or anyway the day hefor, yesterday, and largely under his influence, we would have said it j not cognitive; which says so little that it might have been harmless enough. Kant goes on immediately to distinguish two kinds of "ae thetical judgments," or, as he also calls them, judgments of taste; and here, unfortunately, his influence trickled out. The first kind he call. the taste of sense, the second the taste of reflection; the former con· cerns merely what we find pleasant, the latter must-logically must, some of us would say-concern and claim more than that. And it i only the second whose topic is the beautiful, whose role, that is, would be aesthetic in its more familiar sense. The something more these

• All quotations from Kant are from sections 7 and 8 of the Critique of Judgmen ,

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY * 89 Judgments must do is to "demand" or "impute" or "claim" general v;"didity, universal agreement with them; and when we make such j Ildgments we go on claiming this agreement even though we know from experience that they will not receive it. (Are we, then, just willful or stupid in going on making them?) Kant also describes our Ircling or belief when we make such judgments-judgments in which we demand "the assent of everyone," although we cannot "pbstulate" this assent as we could in making an ordinary empirical Jlldgment-as one of "[speaking] with a universal voice." That is •he sort of thing that we are likely nowadays to call a piece of psy­ . hology, which is no doubt right enough. But we would take that to mean that it marks an accidental accompaniment of such judg­ lII,ents; whereas Kant says about this claim to universal validity, this \'oice, that it "so essentially belongs to a judgment by which we de­

ribe anything as beautiful that, if this were not thought in it, it would never come into our thoughts to use the expression at all, hut everything which pleases without a concept would be counted 15 pleasant." 1 The possibility of stupidity here is not one of con­ ti nuing to demand agreement in the face of the fact that we won't luain it; but the stupidity of going on making aesthetic judgments l all (or moral or political ones) in the face of what they cost us,

the difficulties of finding them for ourselves and the risk of ex­ plicit isolation.

Kant seems to be saying that apart from a certain spirit in which ",Ie make judgments we could have no concepts of the sort we think of

aesthetic.s What can the basis for such a claim be? Let us look at he examples he gives of his two kinds of aesthetic judgments.

T One might compare with this Wittgenstein'S question: "What gives us so much the idea that living beings, things, can feel?" (Investigations, §28!j).

• Another way of describing this assumption or demand, this thing of speaking ith a universal voice, of judging "not merely for himself. but for all men," Kant

100 describe.s as "[speaking] of beauty as if it were a property of things." Only "as if" I\'­ lK'Cl)use it cannot be an ordinary property of things: its presence or absence cannot ,... established in the way ordinary properties are; that is, they cannot be established Ill1bl!cly, and we don't know (there aren't any) causal conditions, or usable rules, for

.oducing. or altering, or erasing. or increasing this "property." Then why not 11,( say it isn't a property of an object? I suppose there would be no reason not

h ; l ay this, if we could find another way of recording our conviction that it is one, "yway that what we are pointing to is there, in the object; and our knowledge that

"len make objects that create this response in us, and make them exactly with the Ira that they will create it; and the fact that. while we know not everyone will agree hh us when we say it is present, we think they are missing something if they don't.

90 '* MUST WE MEAN WHAT WE SAY? ... [someone] is quite contented that if he says, "Canary wine i pleasant," another man may correct his expression and remind him that he ought to say, "It is pleasant to me." And this is the case not only as regards the taste of the tongue, the palate, and the throat, but for whatever is pleasant to anyone's eyes and ears .... To striv here with the design of reproving as incorrect another man's judg ment which is different from our own, as if the judgments were logically opposed, would be folly ...•

The case is quite different with the beautiful. It would (on th contrary) be laughable if a man who imagined anything to his own taste thought to justify himself by saying: "This object (the house we see, the coat that person wears, the concert we hear, the poem submitted to our judgment) is beautiful for me." For he must not call it beautiful if it merely pleases him ....

What are these examples supposed to show? That using a form of expression in one context is all right, and using it in another is not all right. But what I wish to focus upon is the kind of rightness and wrongness invoked: it is not a matter of factual rectitude, nor of formal indiscretion but of saying something laughable, or which

. would be folly. It is such consequences that are taken to display a difference in the kind of judgment in question, in the nature of the concepts employed, and even in the nature of the reality the conceptt capture. One hardly knows whether to call this a metaphysical or a logical difference. Kant called it a transcendental difference; Wittgen­ stein would call it a grammatical difference. And how can psycholog­ ical differences like finding something laughable or foolish (which perhaps not every person would) be thought to betray such potent, or anyway different, differences?

Here we hit upon what is, to my mind, the most sensitive index of misunderstanding and bitterness between the positivist and the post-positivist components of analytical philosophy: the positivist grits his teeth when he hears an analysis given out as a logical one which is so painfully remote from formality, so obviously a question of how you happen to feel at the moment, so psychological; the phi­ losopher who proceeds from everyday language stares back helplessly, asking, "Don't you feel the difference? Listen: you must see it." Surely, both know what the other knows, and each thinks the other is perverse, or irrelevant, or worse. (Here I must appeal to the experi· ence of anyone who has been engaged in such encounters.) Any explanation of this is going to be hard to acquire. I offer the follow-

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY * 91 IIg guess, not because it can command much attention in itself, but

a way of suggesting the level I would expect a satisfying explana­ ,m to reach, a 'way of indicating why we lack as yet the concepts, n :1l the facts, which must form a serious accommodation.

We know of the efforts of such philosophers as Frege and Hus­ I I to undo the "psychologizing" of logic (like Kant's undoing IlI me's psychologizing of knowledge): now, the shortest way I might

I( scribe such a book as the Philosophical Investigations is to say that II :Ittempts to undo the psychologizing of psychology, to show the Il'cessity controlling our application of psychological and behavioral ' Icgories; even, one could say, show the necessities in human action

11 11 passion themselves.9 And at the same time it seems to turn all of .I tilosophy into psychology-matters of what we call things, how we

Ill';lt them, what their role is in our lives. For one last glance, let us adapt Kant's examples to a form which

I~ more fashionable, and think of the sort of reasons we offer for such Ju dgments:

1. A: Canary wine is pleasant. B: How can you say that? It tastes like canary droppings. A: Well, I like it.

2. A: He plays beautifully doesn't he? B1: Yes; too beautifully. Beethoven is not Chopin.

r he may answer:

B2: How can you say that? There was no line, no structure, no idea what the music was about. He's simply an impres­ sive colorist.

Now, how will A reply? Can he now say: "Well, I liked it"? Of course he can; but don't we feel that here that would be a feeble rejoinder, ., ,·etreat to personal taste? Because B's reasons are obviously relevant

o Consider, for exampJc. the question: "Could someone have a feeling of ardent Illve or hope for the space of one second-no matter what preceded or followed this

:ond?" (Investigations, §583). We shall not wish to say that this is logically impossible, I,r that it can in no way be imagined. But we might say : given our world this cannot I I~ rpen; it is not, in ollr language, what "love" or "hope" mean; necessary in our world . hnt this is not.what love and hop::! are. I take it that our most common philosophical. ullderstanding of such notions as necessity, contingency, synthetic and analytic state­ ments, will not know what to make of our saying such things.

92 * MUST WE MEAN WHAT WE SAY? to the evaluation of performance, and because they are arguable, in ways that anyone who knows about such things will know how to pursue. A doesn't have to pursue them; but if he doesn't, there is a price he will have to pay in our estimate of him. Is that enough to show it is a different kind of judgment? We are still in the realm or the psychological. But I wish to say that the price is necessary, and specific to the sorts of judgments we call aesthetic.

Go back to my saying "he doesn't have to pursue" the discussion, and compare the following case:

A: There is a goldfinch in the garden. B: How do you know? A: From the color of its head. B: But goldcrests also have heads that color. A: Well, I think it's a goldfinch (it's a goldfinch to me).

This is no longer a feeble rejoinder, a retreat to personal opinion. and the price that would be paid here is not, as it would be in th former case, that he is not very articulate, or not discriminating, has perverse tastes: the price here is that he is either mad, or doesn'. know what the word "know" means, or is in some other way unintel ligible to us. That is, we rule him out as a competent interlocutor i matters of knowledge (about birds?): whatever is going on, he doesn" know there is a goldfinch in the garden, whatever (else) he thinks h "knows." But we do not, at least not with the same flatness and g' conscience, and not with the same consequences, rule out the peno who liked the performance of the Beethoven: he still has a claim u us, however attenuated; he may even have reasons for his judgmen or counters to your objections, which for some reason he can't gi (perhaps because you've brow-beaten him into amnesia).

Leaving these descriptions so cruelly incomplete, I think on can now imagine the familiar response: "But you admit that ar ments in .the aesthetic case may go on, may perhaps never end, 31 that they needn't go on, perhaps can't go on in some cases, and t: they may have different 'prices' (whatever that may mean), presu ably depending on where they stop. How do you get logic out of th What you cannot claim is that either party to the dispute, whether I the case Kant calls the taste of sense or the case he calls the taste I

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY * 93 I eflection, can prove his judgment. And would he want to, even if he I ould? Isn't that, indeed, what all your talk about criticism was about: I J'he person accounts for his own feelings, and then, at best 'proves' ( .hem to another, shows them to whomever he wants to know them, (' I tlte best way he can, the most effective way. That's scarcely logic; and I /low can you deny that it is psychology?" V

It may help to reply to this: You call it psychology just because II so obviously is not logic, and it must be one or the other. (1 do think Ihnt is the entire content of "psychology" in such objections. Such a t>t' l'son knows what he means by logic: how to do it, how to recognize I when he sees it done, what he can expect from it, etc. But who IlOWS any of this about the "psychology" in question?) Contrariwise,

r ~ hould admit that I call it "logic" mostly because it so obviously is HI t "psychology" in the way I think you mean it. I do not really think I is either of those activities, in the senses we attach to them now; III I cannot describe to anyone's satisfaction what it is. 'Vittgenstein li ed it "grammar"; others might call it "phenomenology."

Those of us who keep finding ourselves wanting to call such IIfcrences "logical" are, 1 think, responding to a sense of necessity

feel in them, together with a sense that necessity is, partly, a matter I rhe ways a judgment is supported, the ways in which conviction I it is produced: it is only by virtue of these recurrent patterns of Ipport that a remark will count as-will be-aesthetic, or a mere I.mer of taste, or moral, propagandistic, religious, magical, scien­ 11f , philosophical. ... It is essential to making an aesthetic judg-

III that at some point we be prepared to say in its support: don't .\1 ,ee, don't you hear, don't you dig? The best critic will know the

I points. Because if you do not see something, without explanation, 1\ there is nothing further to discuss. Which does not mean that

ritic has no recourse: he can start training and instructing you I preaching at you-a direction in which criticism invariably will I to veer. (A critic like Ruskin can be a bit eager in seizing this

lion, but it is a measure of his honesty, and his devotion to art, (' I' to shrink from it; as it is part of the permanence of his writing fxcmplify that moral passion which is a natural extension of the 111';\1 task.) At some point, the critic will have to say: This is what

Reasons-at definite points, for definite reasons, in different cir­ ,"lances-come to an end. (C£. Investigations, §217·)

94 '* MUST WE MEAN WHAT WE SAY? Those who refuse the tenn "logic" are responding to a sense of

arbitrariness in these differences, together with a sense that "logic" is a matter of arriving at conviction in such a way that anyone who can follow the argument must, unless he finds something definitely wrong with it, accept the conclusion, agree with it. I do not know what the gains or disadvantages would be of unfastening the term "logic" from that constant pattern of support or justification whose peculiarity is that it leads those competent at it to this kind of agree­ ment, and extending it to patterns of justification having other purposes and peculiarities. All I am arguing for is that pattern and agreement are distinct features of the notion of logic.

If we say that the hope of agreement motivates our engaging in these various patterns of support, then we must also say, what I take Kant to have seen, that even were agreement in fact to emerge, our judgments, so far as aesthetic, would remain as essentially subjective, in his sense, as they ever were. Otherwise, art and the criticism of art would not have their special importance nor elicit their own fonns of distrust and of gratitude. The problem of the critic, as of the artist, is not to discount his subjectivity, but to include it; not to overcome it in agreement, but to master it in exemplary ways. Then his work outlasts the fashions and arguments of a particular age. That is the beauty of it.

Kant's "universal voice" is, with perhaps a slight shift of accent, what we hear recorded in the philosopher's claims about "what we say": such claims are at least as close to what Kant calls aesthetical judgments as they are to ordinary empirical hypotheses. Though the philosopher seems to claim, or depend upon, severer agreement than is carried by the aesthetic analogue, I wish to suggest that it is a claim or dependence of the same kind.

We should immediately notice an obvious failure in the analogy between aesthetic judgments and the philosophical claim to voice what we say. The philosophical claim seems clearly open to refutation by an empirical collection of data about what people in fact say, whereas it makes no obvious sense to confirm or disconfirm such a judgment as "The Hammerklavier Sonata is a perverse work" by collecting data to find out whether the Sonata is in fact perverse. It is out of the question to enter into this difficult range of problems now. But I cannot forbear mentioning several points which I have tried

AESTHETIC PROBLEMS OF MODERN PHILOSOPHY '* 95 Isewhere to suggest, with, to judge from results, evident unsuccess.10

1. I take it to be a phenomenological fact about philosophizing 'rom everyday language that one feels empirical evidence about one's I.mguage to be irrelevant to one's claims. If such philosophizing is to he understood, then that fact about it must be understood. I am not .l ying that evidence about how (other) people speak can never make i,ll ordinary language philosopher withdraw his typical claims; but I 'Ind it important that the most characteristic pressure against him is 'i pplied by producing or deepening an example whiCh shows him ,I .at he would not say what he says "we" say.

2. The appeal to "what we should say if ..." requires that we tn)agine an example or story, sometimes one more or less similar to ,'vents which may happen any day, sometimes one unlike anything we have known. Whatever the difficulties will be in trying to charac­ terize this procedure fully and clearly, this much can be said at once: ,f we find we disagree about what we should say, it would make no OI bvious sense to attem'pt to confirm or disconfirm one or other of our I esponses by collecting data to show which of us is in fact right. What we should do is either (a) try to determine why we disagree (per­ k Ips ,...e are imagining the story differently)-just as, if we agree in t esponse we will, when we start philosophizing about this fact, want It) know why we agree, what it shows about our concepts; or (b) we will, if the disagreement cannot be explained, either find some ex­ planation for that, or else discard the example. Disagreement is not t1 1sconfirming: it is as much a datum for philosophizing as agreement ,t At this stage philosophizing has, hopefully, not yet begun.

3. Such facts perhaps only amount to saying that the philosophy uf ordinary language is not about language, anyway not in any sense rn which it is not also about the world. Ordinary language philosophy ,. about whatever ordinary language is about.

The philosopher appealing to everyday language turns to the •eader not to convince him without proof but to get him to prove \t)mething, test something, against himself. He is saying: Look and

10See J. Fodor ann J. Katz, "The Availability of What We Say," In the Philosoph- QI Review, Vol. LXXn (1963), an attack, primarily, on my paper "Must We Mean

What We Say?" which appears as the first essay in this book. [Added 1968. A paper "l' Professor Richard Henson ("What We Say," American Philosophical Quarterly, Vol. t/No. 1, January 1965, pp. 52-62) includes specific rejoinders to a number of the points r~ i sed by Fodor and Katz.]

96 '* MUST WE MEAN WHAT WE SAY? ...fu!4-uu_L:wll1:th.er you can see what I see, ~ say. Of course he often seems to ali'SWer or beg his own questi by posing it in plural form: "We say ... ; We want to say .. . We can imagine . . . ; We feel as if we had to penetrate pheno en a, repair a spider's web; We are under the illusion ..• ; are dazzled ... ; The idea now absorbs us ... ; Weare dissat tied ...." But this plural is still first person: it does not, to u Kant's word, "postulate" that "we," you and I and he, say and w and imagine and feel and suffer together. If we do not, then t philosopher's remarks are irrelevant to us. Of course he doesn't th in they are irrelevant, but the implication is that philosophy, like a is, and should be, powerless to prove its relevance; and that says so thing about the kind of relevance it wishes to have; All the phil pher, this kind of philosopher, can do is to express, as fully as he ea his world, and attract our undivided attention to our own.

Kant's attention to the "universal voice" expressed in aesthtl judgment seems to me, finally, to afford some explanation of that of dogmatism which claims about what "we" say seem to carry fl critics of ordinary language procedures, and which they find repu nant and intolerant. I think that air of dogmatism is indeed pres in such claims; but if that is intolerant, that is because toleran could only mean, as in liberals it often does, that the kind of clai in question is not taken seriously. It is, after all, a claim about 0 lives; it is differences, or oppositions, of these that tolerance, if it to be achieved, must be directed toward. About what we should when, we do not expect to have to tolerate much difference, believi that if we could articulate it fully we would have spoken for al1 me found the necessities common to us all. Philosophy has always ho for that; so, perhaps, has science. But philosophy concerns those nee sities we cannot, being human, fail to know. Except that nothing more human than to deny them.

IV

Austin at Criticism

lept for the notable translation of Frege's Foundations of Arith· tic and whatever reviews there are, Philosophical Papers collects

I Ihe work Austin published during his lifetime.1 In addition, this ~s t volume includes two papers which will have been heard

t" 'lI t, but not heard, outside Oxford and Cambridge. The first is one Ihe two pieces written before the war ("Meaning," 1940) and

IIJ WS more clearly than the. one published a year earlier ("Are There , Priori Concepts?," 1939) that the characteristic philosophical turns

r which Austin became famous were deep in preparation.2 The nnd previously unpublished paper ("Unfair to Facts," 1954) is

',,,Lin's rejoinder to P. F. Strawson's part in their symposium on II lh, a debate which, I believe, Austin is widely thought to have I initially, and to lose finally with this rejoinder. Austin clearly I lIot concur in this opinion, repeating the brunt of his counter· rge at the end of the course of lectures he gave at Berkeley in

,1$- 1959.8 The remaining five papers have all become part of the

I J. L. Austin, Philosophical Papers, ed. by J. O. Unnson and G. J. Warnock .!I 'rd: The Clarendon Press, 1961).

• Curiously, the 1940 paper is the most ·Wittgensteinian of Austin's writings, in ntlng an explicit theory of what tauses philosophical disability and in the partie.

0' theory it offers (sc., "We arc using a working-model which fails to fit the facts tha t Hally wish to talk about").

• These lectures, which he gave for many years at Oxford, were published post­ IfllJsly under their Oxford title, Sense and Sensibilia, edited by G. J. Warnock (Ox­ , The Clarendon Press, 196:1). Austin's original paper on "Truth" (1950) is, of

97