On What Sort of Speech Act Wittgenstein's Investigations Is and Why It Matters

(The Philosophical Forum, XXVIII, no. 3, 1997. )

I

I ought to be no more than a mirror, in which my reader can see his own thinking with all its deformities so that, helped in this way, he can put it right. (Wittgenstein 1980, p. 18)

Philosophers concerned with speech acts, or Wittgenstein's uses of language, mostly fix their attention on actions done by issuing just a phrase or short sentence (in the appropriate circumstances with the proper qualifications, feeling, intent, uptake, etc.). "Five red apples" is Wittgenstein's paradigm example in his Philosophical Investigations. "There's a bittern at the bottom of your garden" plays a similar role in J. L. Austin's most central and ambitious essay, "Other Minds." Indeed, as Wittgenstein points out, a single word or gesture may do the job perfectly well, just as an illiterate man can make a valid contract by marking an "X" on a piece of paper. And, of course, in all relevant philosophical respects, a speech act may be written rather than spoken: in his leitmotiv example, Wittgenstein writes "five red apples" on a slip of paper and hands it to his shopper.

He takes the slip to the shopkeeper, who opens the drawer marked "apples"; then he looks up the word "red" in a table and finds a colour sample opposite it; then he says the series of cardinal numbers -- I assume that he knows them by heart -- up to the word "five" and for each number he takes an apple of the same colour as the sample out of the drawer. (Philosophical Investigations, par. 1; hereafter I just give the numbered paragraphs for Part I, and page numbers for Part II )

However, a speech act obviously may use several sentences, or indeed several thousand. Dostoevsky stages Raskolnikov's confession with relentlessly laconic plainness: "It was I who killed the official's old widow and her sister Lizaveta with an axe and robbed them." (less than twenty words to end a third of a million, words so tersely definitive that Dostoevsky can only think to add a string of dots to indicate that Raskolnikov repeats himself word for word, tone for tone, without extenuation, variation, explanation, rationale, or excuse). But many actual criminal confessions take several hundred sentences elicited over several hours if not several thousand over days. Indeed, Augustine's Confessions and Tolstoy's A Confession are book length. Wittgenstein was much struck by both authors, although ultimately in rather different ways. Investigations begins with a lengthy quote from Augustine's Confessions and Wittgenstein makes three further quotes from Augustine and several other references as Investigations unfolds (aside from a couple of lines from Plato's Theatetus, and a phrase from William James, Augustine is the only philosopher whom Wittgenstein quotes, though he occasionally jousts with Bertrand Russell and, much more frequently, his own Tractatus views, and he wrestles with particular lines of Meditations, though he does not mention Descartes by name ). No allusion to Tolstoy appears, although Wittgenstein had been overwhelmed by Tolstoy's Gospels in Brief while in the tumult of war and he continued to vigorously recommend Tolstoy's shorter fictions -- Hadji Murad in particular -- to people for the rest of his life, while issuing more circumspect comments about Tolstoy's confessional and religious ramblings (Engelmann 1967, pp. 66, 79-80, 91, 109; McGuinness 1988, pp. 220-221, Bartley 1985, pp. 72-73, 84-85, 110; Redpath 1990, pp. 50, 61; Monk 1990, pp. 115-117, 132, 136; Malcolm 1962, p. 38; Wright 1982, p. 23; Hallett p. 774; McGuiness 1988, especially, p. 110).

Raskolnikov's confession, however brief and heartfelt, is not of course a real confession by a real person who has committed a real crime, but rather, like Raskolnikov himself, a part of a brilliant artist's fictional creation (although we bring our everyday understanding to the narrative, its characters, and their speech acts). But it is important to recognize that Augustine's and Tolstoy's books are put forward as real confessions of real offenses by their authors. These books call out, demand, to be taken as such, and not, say, as creative fictions to be enjoyed for their coherence, characterization, drama, or moral example. Indeed, both confessions call out to be assessed by the criteria laid down in Christian rite for a valid confession: the acts confessed did take place and are truthfully recounted and recognized as offenses, worse sins have not been left out, and there is genuine contrition with manifest intent to struggle to do better henceforth. Since both books treat of several decades, confessing lives of sin and self-deception and, eventually, throes of repentance, conversion, and redemption, one expects a positive as well as negative example, a convincing and coherent account of sinful life but also of a sea change to a better and truer one, thus offering instruction to the reader. One can of course to some degree assess the truthfulness of the confession and redemption on internal textual grounds, without as much as one might like of external evidence -- particularly as the sins confessed are intellectual, attitudinal, and affective. Indeed, just as John Donne devoutly harnessed the enormous linguistic, expressive, and logical skills of his poetry and prose to holy purpose in his sermons and meditations with overwhelming truthfulness, to use Wittgenstein's word for confessional virtue, so the truthfulness of Augustine's Confessions or Tolstoy's A Confession derives most importantly from the text itself -- yes, but if inescapable historical evidence showed up that the man who wrote Confessions, became Bishop of Hippo, and composed most of the other works attributed to St. Augustine, was in fact a scholarly satirist who expressively invented a convincing account of Christian self-exposure, sin to expiation, and then found himself trapped by an eager community of the faithful, and the career that now opened up, we would read the book differently (for supportable hypotheses see also Alfaric 1918 and Augustine 1992, p. xx). On the other hand, when E. A. Poe issues the "Tell-Tale Heart," in which a madman, plaintively insisting that he is not mad, confesses to killing his employer because he found his heart beat too noisy and still finds it deafening after the murder, we have a luridly self-deluding character who unwittingly reveals himself in a caricature of a valid confession that makes, for those with a taste for Poe, a neatly chilling horror story.

The claim that there is no principled limit to the number of sentences (or subordinate and bracketed speech acts) in a speech act reminds one of the paradigmatic claim of contemporary linguistics that there is no longest sentence. Perhaps the simplest way linguists seek to prove this is to point out that English has a recursive rule to the effect that two well-formed sentences can be made into a third sentence by conjoining them with the word and. But this is not a helpful, nor a sufficiently fine-grained, way to make the point: you could, following the crude rule, simply run on indefinitely with "It is flat and it is flat and it is flat and . . . " or join by and a random list of sentences with no coherent anaphoric, implicative, logical, or illocutionary interrelationship (reminiscent of the atomic propositions of Wittgenstein's Tractatus). Of such products the later Wittgenstein might have said You have noticed that you can replace the periods in a piece of writing with ands and it won't seem to make much difference, so you conclude that you have found out that the period dot is really an and, but you might as well have said that an and is really a period dot -- you can say it is many sentences or one, only note what makes you want to say it is one.

A second paradigmatic claim of contemporary linguistics is that what we natively and consciously hear or say as a sentence is generated characteristically as a hierarchical merger, with deletion, of subsidiary sentences; equally, in coherent discourse, anaphoric and implicative relationships, etc., show us that several sentences are indeed one sentence. So the point is that much the same grammatical restraints that determine the logical/semantical/syntactical properties of sentences are also those that govern collections of sentences. The imperative sentence, "Call me Ishmael," requires that the pronoun and noun be co-referential; equally it initiates a book length first person narrative within which the first person singular pronouns and "Ishmael" will continue, for the same syntactical reasons, to be co-referential (similarly with the richly- structured narrative web of nominal and pronominal references to other personae with their attendant characterizations and actions) with all the demands and possibilities of consistency, coherence, and implication, of elegant detail, striking portraiture, and expressive power, that this vast first person narrative act afforded Herman Melville. The possibility of structures such as these gives full, fine-grained richness to the claim that there is no longest sentence. In Requiem for a Nun William Faulkner does indeed begin with a sentence that only gets a period at the book's end. But he does not add logician's ands between sentences that otherwise have no grammatical relationships, rather he makes use of the rich panoply of syntactical devices (the whichs, whos, buts, durings, sinces, sos, etc.) through which we indicate grammatical relationships within complex sentences (or between sentences ). That Faulkner mostly manages to pull off his "one sentence" narrative, that Requiem for a Nun is not difficult to read and make sense of as a story, that it has so much the sinewy cadences of the oral story telling that we soon forget the lack of dots, strongly suggests that narratives at least provide the richest and most intriguing actual exemplifications of the "no longest sentence" thesis.

Investigations is certainly not a conventional biographical narrative of the Austro-Hungarian nobleman genius, son to one of the richest families in central Europe, who worked with Bertrand Russell on the logical foundations of mathematics, who rubbed himself raw so to speak in individual guilt, had a harrowing experience of world war which led him to read Tolstoy's A Confession and related works, who like Count Tolstoy was to give up his inheritance, repeatedly contemplate suicide, and take up school teaching, but who returned to Cambridge eventually, lecturing to select small groups of advanced students and professors. But clearly nonetheless Investigations is straightforwardly first person narrative: The I is co-referential with Ludwig Wittgenstein of the title page, and then some, the narrative anchor for the piece, like the I of Descartes Meditations, although Wittgenstein's I easily becomes we when a general human understanding is examined. But Investigations is also second person: you are asked questions, your answers are suggested or implied and then explained, criticized, or expanded; indeed there is even second person narration in which you are described as going through various exercises or routines. There is no book I know that is more conversational, interactive, and narrational: you almost hear your responses (and when he provides them -- "You will say . . . " -- you read them as your own) and then find yourself honestly caught and turned about by his reply. You want to say, how can I be having an intense conversation with a man who died many decades ago?

Taking a path that Wittgenstein explicitly and emphatically avoided, J. L. Austin offered a general account or theory of the "felicities" and "infelicities" of speech acts, distinguishing locutionary, illocutionary, and perlocutionary aspects of speech acts, etc. (Austin 1962). He explained what he was about, in one of his more rotund and pontifical metaphors, by saying that the "seminal and tumultuous" sun (philosophy) periodically casts forth material which will harden into one more comparatively settled science -- in this case, a linguistics that will go beyond purely formal phonological and syntactical theory to take on a fuller semantical, pragmatic, and evaluative account of our speech actions (Austin 1979, p. 232). To a degree he succeeded in that some linguists have adopted some of his terminology and distinctions, along with H. P. Grice's conversation theory, etc. (speech act, however, certainly has not replaced sentence as the focus of linguistics). For many philosophers his most striking discrimination was of "performatory utterances" or (in his later terminology) "performatives." These were delivered through sentences in the first person present tense whose main verb specified the speech act done by saying these words in the appropriate conditions, etc.: for instance, "I christen this ship the Mary Ann" or "I promise to water the azaleas." While Austin never gave up the view that performatives are paradigmatic doings, not true or false descriptions, he came to insist that all speech acts have such an illocutionary aspect (along with the locutionary aspects of reference and predication). Particularly since I mentioned Austin's overweening "seminal and tumultuous" posture that his heavenly profession will deign to rain down a proper conceptual foundation on the grateful natives, so they may become proper linguists, I must add that J. R. Firth, the most influential and popularly known British linguist of Austin's day, had since the early 1930s championed a use theory of meaning as the keystone in what has been called the London School of Linguistics, emphasizing,

situations in which words, often fixed by law or custom, serve to bind people to a line of action or free them from certain duties . . . [and that] many other words and phrases are used with a similar binding effect in everyday life . . . [in, e.g.] the language of agreement, endorsement... wishing, blessing, cursing, etc. " (Firth 1935/1957, pp. 30-31).

While a performative sentence (used appropriately) often conveniently indicates, in its main verb, what speech action it does, a host of other factors also may or must serve in such a determination. Notice, for example, that the force of Raskolnikov's confession would if anything be weakened if Dostoevsky had added "I confess that . . . " to its beginning. And I must confess that in everyday talk I expect the words "I confess that . . . " or "I must confess that . . . " to introduce the admission of a minor shortcoming or to soften a claim or critical comment (e.g., "I must confess that I just haven't got around to reading your paper" or "I must confess that I don't see why that follows"). A customary beginning for the Christian rite, "Father, I have sinned in thought, word, and deed . . . ," clearly marks the sentences that follow as a confession; similarly, "This is my last will and testament . . . ," "You are hereby requested and required . . . ," or "I swear to tell the truth, the whole truth, and nothing but the truth . . . ," the first marking what follows as a will, the second, a military order, and the third (a proper performative), legal testimony. Such prefatorial markers function rather like the titles of books, journals, and papers (with an author's name serving like the signature on the will, etc.); subtitles and prefaces also serve this purpose, crucial markers of what the author will do in the many sentences that follow. When Wittgenstein prepared his manuscript for publication, he intended it to be titled Philosophical Investigations; although he clarifies this in the first line of his 1946 preface by writing that what follows is the carefully-winnowed "precipitate" of a number of years of philosophical investigations (strictly, Part I is the precipitate of 1936-1939 and Part II of 1946-1949 (Wright, pp. 110-136)). But there are clear indications in that preface, in the crucial opening leitmotifs, throughout the fabric of the work, and in its concluding passages, that indicate that the work is something else, or something more specific, as a work or action as well; to call something a "precipitate" is to avoid saying how it is to be taken, aside from the suggestion that what follows is a carefully crafted and definitive formulation of a number of years of philosophical inquiry. Of the man who adamantly insisted, against his earlier self and generations of philosophers, that there are countless uses of language, we have a right to ask: yes, and what are you doing with language here? .

II

That I measure time I know. But I measure not the future for it is not yet; nor do I measure the present because it is extended by no space; nor do I measure the past, because it no longer is. What, therefore, do I measure? Is it times passing, not past?. . . But how is that future diminished or consumed which as yet is not? Or how does the past, which is no longer, increase, unless in the mind which enacts this there are three things done? . . . Future time, which is not, is not therefore long; but a "long future" is "a long expectation of  the future." Nor it time past, which is no longer, long; but a long past is "a long memory of the past." (Augustine 1948, p. 199-201.)

In his Tractatus Logico-Philosophicus Wittgenstein offered to the world the culmination of his early philosophical thought and, for a time, left philosophy. This latter abjuration is prefigured by the claim about "the problems of philosophy" with which he prefaced Tractatus, after ruefully admitting that his "expressive craftsmanship," on which we know he labored long and hard, has fallen short.

On the other hand the truth of the thoughts communicated here seems to me unassailable and definitive. I am, therefore, of the opinion that the problems have in essentials been finally solved. (Wittgenstein 1922, p. 29)

To this proud, I dare say arrogant, announcement the youthful Wittgenstein adds, dismissing centuries of the modern effort begun in Descartes as "a misunderstanding of the logic of our language."

And if I am not mistaken in this, then the value of this work secondly consists in the fact that it shows how little has been done when these problems have been solved. (Wittgenstein 1922, p.27)

Today Wittgenstein's later philosophy is represented by a number of books compiled from the oft differing lecture notes of Wittgenstein's students and from Wittgenstein's own often disordered notes. If we take books like Tractatus or Descartes' Meditations as large scale speech acts, delivered to the world by their designers with clear indications of their character as actions, these compilations may well be regarded not as full-fledged speech acts authored by Ludwig Wittgenstein but as notes, drafts, preparations, etc., that did not mature and join into something that he would willingly issue into the public arena (notes a person might make for himself with the intention of drafting an appropriately signed and witnessed will do not constitute a will; a promise planned is not a promise made, an insult contemplated but unvoiced is not an insult, etc.). Philosophical Investigations is the only MS of Wittgenstein's later philosophy that he prepared for publication and intended, when he wrote its preface, to publish. This gives us a particular license to take it as a speech act, a finished and issued work, one whose title, initial lines, consistencies, order, self-labeling, etc., constitute a work, an evaluable action, rather than compilations of students' lecture notes, or their notes with Wittgenstein's additions and corrections, latterly shuffled together by others without a credible claim of unity. It also supports the claim that the force, the achievement, of Investigations substantially determines what we should make of the many oddities in the work, of the extraordinary way in which it can be seen as simply an early presentation of the claims of "ordinary language philosophy" (of Gilbert Ryle, Austin, etc.), only one notable for its unscholarly, disordered, egocentric, overheated but often striking and memorable prose style, while it can also be seen as a profoundly original and shockingly strange work that operates orthogonally to the whole enterprise of philosophy as a particular academic discipline -- even, perhaps, a work, indeed an act, that sets us free both to revere and live honestly within the world of everyday experience, and, if one is inclined to as Wittgenstein of course was not, to embark on real (what Wittgenstein would call "natural") cognitive science.

In the preface to Investigations, Wittgenstein indicates his recognition of this dilemma in that he says he has decided to publish because his

results, variously misunderstood, more or less mangled or watered down, were in circulation. This stung my vanity. (Wittgenstein 1953, p. v-vi)

Latterly, he adds,

If my remarks do not bear a stamp which marks them as mine, --I do not wish to lay any further claim to them as my property. (Wittgenstein 1953, p. vi)

Note the shift. In the Tractatus preface he submits that "the truth of the thoughts communicated here is definitive and unassailable [and all the problems of philosophy are solved]," while his "expressive craftsmanship" leaves much to be desired: hence, the view that there is a translative relationship between what we say in the "apparent form" of an ordinary human language and its deep logical structure and the view that this relationship can only be pointed to or more or less shown, not clearly said or stated -- it is too profound for the ordinary words of everyday life. With Investigations, however, there are no definitive and unassailable true thoughts to be more or less successfully expressed, there are only "sketches of landscapes" made in "a battle against the bewitchment of our intelligence by means of language" (par. 109). Investigation's Preface also displays a humility foreign to Tractatus's Preface. Wittgenstein submits he was made to see his mistakes by the "always certain and forcible" criticisms afforded him by Frank Ramsey during "the last two years of his life," he similarly credits Piero Sraffa as the "stimulus for the most consequential ideas of this book," and he concludes, modestly and somberly,

I should have liked to produce a good book. This has not come about, but the time is past in which I could improve it. (Wittgenstein 1953, p. vi)

Augustine, whose mistake-breeding "tractatus-esque" account of language is quoted at the beginning of Investigations as foil to the leitmotiv counterpoint of "five red apples," is again made introducer when, in the text of Investigations, Wittgenstein first gives a somewhat general account of what he is doing. After quoting Augustine's famous remark about time, Wittgenstein generalizes it,

This could not be said about a question in natural science. Something that we know when no one asks us, but no longer know when we are supposed to give an account of it, is something we need to remind ourselves of . . . We feel as if we had to penetrate phenomena: our investigation, however, is directed towards the 'possibilities' of phenomena. We remind ourselves of the kind of statement that we make about phenomena. Thus Augustine recalls to mind the different statements that are made about the duration, past present future, of events. Our investigation is therefore a grammatical one . . . Our considerations could not be scientific ones. It was not of any possible interest to us to find out empirically 'that, contrary to our preconceived ideas, it is possible to think such and such' . . . And we may not advance any kind of theory. There must not be anything hypothetical in our considerations. We must do away with all explanation, and description alone must take its place. And this description gets its light, that is to say its purpose, from the philosophical problems. These are not empirical problems; they are solved, rather, by looking into the workings of our language, and that in such a way as to make us recognize those workings: in despite of an urge to misunderstand them. The problems are solved, not by giving new information, but by arranging what we have always known. (par.s 89, 90, 109)

Now what strikes one overwhelmingly about this whole procedure, and the justification for puritanical restrictions (no theory, no technical language, etc.), is that it sounds much like the way in which linguists get the native-speaker sentences (usually serially numbered for ease of reference) from which they extract evidence for various grammatical rules or restrictions, all part of the attempt to characterize the language, or better, the grammatical competence instantiated in the native speaker's mind/brain. Linguists, surely rightly, take the view that native speaker's intuitions about grammaticality ("Is this English?, Would you say this?") are reliable, robust, and uniform to the degree that the linguist can simply ask himself/herself Would I say this, would I take this as an English sentence, etc.? Sometimes the linguist may supply a helpful context, or some starred (ungrammatical) numbered "sentences" to show what the grammar rules out, etc. For example,

1a) John shaved him. 1b) Alex expected John to shave him.

2a) John shaved himself. 2b) Alex expected John to shave himself.

With such samples, the linguist shows us that for the competent native speaker (as in 1a and 1b) him cannot be co-referential with the nearest nominal, and must be so with the more distant one (i.e., him cannot be John and, in 1b, must be Alex), while (as in 2a and 2b), himself must be co-referential with John and cannot be so with the more distant Alex (the anaphoric relationships that are ruled out here could be called starred sentences, like the nonsense, or useless, speech that Wittgenstein considers). Now the linguist might try go on to extend the analysis by considering a number of other sentences. Thus:

3a) I knew the man who Alex expected to shave him.

3b) I knew the man who Alex expected to shave himself.

Now we seem to have found a contradiction, for in 3a him is co-referential with Alex, not the more distant man or I, while in 3b himself cannot be co-referential with Alex and must be so with the more distant man, thus seeming to directly contradict our tentative generalization. The linguist may now propose (I am giving a glimpse of a large body of research covering a process that seems universal) that man leaves a psychological "trace" between expected and to that does not appear phonologically but is real enough for the competent native speaker (similarly, phonologists point out that the "l" sound of "split," which distinguishes it from "spit," does not actually exist sonically, though it is "heard" psychologically by competent speakers).The linguist now seems on the way to establishing some generalizations about anaphora in English that presumably apply very generally to pronouns and reflexives. (Note of course that while the linguist's sample sentences are restricted, as Wittgenstein's are in general, to the everyday English of the native speaker (to that speaker's most basic intuitions about "Is this English? Would I, or my shopkeeper, say this?"), while the surrounding text bristles with technical terms, theoretical generalizations, complicated arguments, all delivered in the impersonal, nonconversational prose of academic journals -- nothing is supposed to be decided by "ordinary competent speaker's intuitions" at the surrounding text level.)

It is of the utmost importance that linguists do not in fact disagree about the sample sentences or about the minimal intuitions of competent speakers (i.e., that "John shaved himself" is English and that himself must mean John, etc.); if they did, the notion of competence is in disarray, and they would have to do what developmental linguists do, namely, record children's spontaneous speech, placing no trust whatsoever in their own intuitions about what children "ought to" say. The same goes for Wittgenstein's enterprise, where all the sentences (and related sign or perceptual phenomena) are to be like the linguist's samples: this is exemplified in his comment that "If someone tried to advance theses in philosophy, it would never be possible to debate them, because everyone would agree to them" (para. 128). For Wittgenstein, the linguist's text (apart from the samples) advances the theories, explanations, generalizations, "the search for what is hidden," etc.: it is, in his terminology, nonsense or natural science. Something like "trace" for example, while perhaps "tacit knowledge" in the competent speaker, is not available in the ordinary speaker's consciousness, not something "in plain view," so it is wholly excluded from Wittgenstein's enterprise (except in one way, which we will eventually consider).

But while Wittgenstein restricts himself to the language of the linguist's samples, and does so in many respects on much the same basis as they do, he does not pick his examples, such as "five red apples" and its surrounding narrative, willy-nilly or from some general, still less systematic, interest in grammar or speech acts, nor does the "discussion" that follows these examples display such an interest. Such a "description gets its light, that is to say its purpose, from the philosophical problems" which are to be solved when our description and discussion make us understand the in-plain-view "workings of our language . . . in despite of an urge to misunderstand them." What is left us then is in a sense nothing.

Where does our investigation get its importance from, since it seems only to destroy everything interesting, that is, all that is great and important? (As it were all the buildings, leaving nothing behind only bits of stone and rubble.) What we are destroying is nothing but houses of cards and we are clearing up the ground of language on which they stand. [par. 118]

Well, one wants to ask, what is the harm in that? Why not let people build cardhouses like you and Russell did -- it is not many people and it seems harmless enough -- you yourself make the point that such philosophical monocular visions don't have any practical or testable significance. Even one caught by such an ultimately impoverished but utterly seductive illusion, when he leaves his study, easily re-enters our common everyday world.(And if someone did get emotionally affected -- you imagined a case where you try to regard the children playing across the street as mere automata, and you get a strange feeling -- why not a drug or some good hard physical exercise?)

Well, one wants to ask it but I think it would be impertinent. Philetus of Kos reportedly grew thin and died, circa 270 BC, of brooding about the liar paradox. Accounts abound, especially from the period when Wittgenstein first was captivated by philosophy until he was well into writing the Tractatus, of the passionate agony, strain, exalting highs, and devastating lows, that consumed him as he wrestled with philosophy. Undoubtedly, there were peculiarities in his body chemistry, family matrix, sexuality, etc., and he was fascinated as well as repulsed by Tolstoy's example (of confessing his bodily and individual sins, of exposing the hypocrisy and empty artificiality of intellectual life, of living with the lowly, making boots, etc.). But it is beyond question that the philosophical problems consumed and tormented him, that they drove him to work and work and work, to write and revise endlessly, until the vast "unassailable and definitive" tractatus card house vision was chiseled into the Tractatus and he hoped (falsely) to be at rest. No one in our era would have seemed more likely to self-destruct from philosophical perplexity than this man (McGuinness 1988, p. 100).

Though Wittgenstein begins Investigations with the Augustine quote to represent the beginnings of the temptation that would flower into the tractatus bewitchment, he at one time planned to publish both the "old thoughts" of Tractatus and the "new ones" of Investigations together, and this project survives to a considerable degree in the actual text,

113. "But this is how it is -----------" I say to myself over and over again. I feel as though, if only I could only fix my gaze absolutely sharply on this fact, get it in focus, I must grasp the essence of the matter.

114. (Tractatus Logico-Philosophicus, 4.5): "The general form of propositions is: This is how things are."-------That is the kind of proposition that one repeats to oneself countless times. One thinks that one is tracing the outline of the thing's nature over and over again, and one is merely tracing round the frame through which we look at it.

This certainly sounds like a description of bewitchment (the word at root means taking someone over through incantations and sorcery; instrumentally, it is the very sequence of words through which the spell is cast: similarly, confession is a speech transaction between the confessant and the listening confessor, but also the very sequence of words used; confessor also means someone who publicly confesses the articles of faith in face of persecution). The bewitchment is personal but purely cognitive or philosophical: the I is the I of Meditations or Tractatus, and the one is that individual tersely and instructively recollected in philosophical sin by an abased and determined later self, confessant and confessor. There is just the slightest hint as to what this bewitchment might mean to a particular embodied human individual who instantiated this captivation.

But, we want to say, you have really quite toned down what a threatening and compulsive experience bewitchment was for you, not for the I of Investigations or even the arrogant Cartesian I of Tractatus, but for you, plain old Ludwig Wittgenstein (the personal, bodily Wittgenstein, the one who angrily accused Russell, when he published Problems of Philosophy, of issuing a "shilling shocker," sullying, cheapening, vulgarizing, the lonely but absolutely required enterprise, begun in Frege and Russell himself , of finally seeing the starkly minimalist logically necessary structure of the world and of language through determining what (profoundly unspeakably) had to hold of each that one could picture the other). Russell tells us that Wittgenstein had all the passionate mad drive, exultantly creative and savagely self-critical in his thinking, "of genius as traditionally conceived" (Russell 1968, p. 136). He also tells us that on several occasions Wittgenstein would come to Russell's Trinity College rooms of a midnight and might sit silently for hours, Russell reluctant to expel him since, as Russell explains, Wittgenstein would announce upon arrival that he would commit suicide after he left. Russell adds that on one occasion, he called out to Wittgenstein to ask whether he were brooding about logic or his sins, to which Wittgenstein replied, "Both." (Russell 1968, p. 99, etc.). (Russell's after dinner caricature, written and published many decades later, is narrated with dotty light-heartedness and races full tilt toward its punch line, for otherwise we will fill in what must have a wrenching and perplexing experience for Russell and an unimaginably tortured, hysterical, and driven maelstrom for Wittgenstein.)

We laugh at the punch line precisely because Wittgenstein's answer appears emotionally absurd, of a piece with "She went home in a flood of tears and a sedan chair," for brood seems to undergo a category switch like in. Specifically, brooding about logic sounds like an impersonal esoteric activity, something, to use Investigations' most commanding global metaphor, that speaks to the layout, or ideally supposed design, of some straight-lined scientific suburb of our language's outskirts, while brooding about your sins belongs to the "maze of little streets and squares" of the old city that is central to us all as persons, as competent speakers of a human language, as "masters of a technique" (sin here, you want to say, is real sin -- and agency agency, and confession confession, as well). Worse still, we understand brooding as an compulsive spell of monomaniacal obsession; paragraphs 113-114 describe brooding, particularly the "countless times" and the yearning and self laceration of "If only I could fix my gaze absolutely sharply." "Brooding" means not being able to stop (in Investigations Wittgenstein writes that he finally has learned to do philosophy so that he can stop doing philosophy, the very motto for a born philosophical brooder striving for redemption). So we might imagine that Wittgenstein is obsessively entangled in tracing the general form of the proposition, or that he is lacerating himself, like a character of out of Kafka, about his relationship with his father. But he can't be doing both at the same time, or hopping from one to the other every few seconds,

Grief describes a pattern which recurs, with different variations, in the weave of our life. If a man's bodily expression of sorrow and of joy alternated, say with the ticking of a clock, here we should not have the characteristic formation of the pattern of sorrow or of the pattern of joy. "For a second he felt violent pain." ---Why does it sound queer to say: 'For a second he felt deep grief'? Only because it so seldom happens?

But don't you feel grief now? (But aren't you playing chess now? The answer may be affirmative, but that does not make the concept of grief any more like the concept of a sensation. . . .). ( p. 174.)

Let us then exchange the cheap joke for a more charitable and instructive interpretation. Wittgenstein is again and again striving to fix his gaze absolutely sharply, tracing and retracing how the world and language ultimately have to be, but there are false moves and blurs, a lack of sufficient intensity and purity of vision, and already a frightening history of sloppy and superficial mistakes -- in a word, sins. So when Wittgenstein says "Both," he takes logic to be the profound "unassailable and definitive truths" he desperately seeks, and his sins are his all too many hateful errors, lack of vision and dedication, and the ease with which he lets himself be seduced by superficial features of language (raised in a Catholic community, Wittgenstein did not need to read Augustine's Confessions to be thoroughly familiar with this use for the word sin, and the related old city neighborhood of useful tools to which it belongs). There is also a further irony: The world and language as they must be, the "truths" he had strained so proudly to see, is card house, the shadow cast by some homely features of language, of perception and procedure, which come to seem the real world; while the superficial, paradoxical, and highly-localized features of language and thought, "the maze of little streets and squares" common to us all, is the foundation and the very face of our human existence, worthy of reverence and a demonstrable faith that its countless uses have for us no useful summary or regimentation, that the manifest contradictions or incoherences between our various language games are bedrock for us as human persons, and that for that very reason there is no hope that this tangle can be "translated" into a consistent, unified, deep language, or a level of neurological description for that matter, that can tell us what is really going on, that can give us the true account, a more useful picture for practical purposes, than the one already available to us. In a narrative a remark can have an ironic resonance because of what is to come.

Well, someone may say, it may make a better story -- Wittgenstein will look more coherent and self insightful, even more confessional. But there must have been a literal fact of the matter: what the real flesh and blood Ludwig happened to be thinking of when he replied "both." Indeed, we might find a confessional diary in which Ludwig tries to recall what he meant or even records that Russell had asked him what he meant by "both" and he had explained that . . . Stories, however plausible, can crash against facts.

I agree. To give an actual and excruciating case, Tolstoy's biographer, A. N. Wilson, makes this point about the self-deluding gap in A Confession between sketch and landscape.

Approaching A Confession 'blind', the reader will indeed be arrested by its overpowering emotional force, and might even mistake its apparently ratiocinative thrust, its burning intellectual sincerity, for a piece of argument. But for those who have followed Tolstoy's life and work in a chronological order, its ninety or so pages give off disconcerting impressions. It is not the book which its author intends us to read. Doubtless, while he was writing it, A Confession felt as noble and courageous as some modern readers have found it. But it is not, as Tolstoy so heart-rendingly believes, the record of a mind clearing, of a troubled soul coming at last to peace . . . [Tolstoy tells us he was ] a thoughtless sensualist, who had put all thoughts of God, the meaning of life, soul or goodness aside . . . who had pursued the sins of the flesh, the cruel pleasures of war, wrote only for fame and money, and enjoyed the didactic role thrust upon the Russian writer, even though he had nothing to teach. (Wilson 1988, p. 310-311)

However, after marriage, many novels and decades, he eventually realizes the pointlessness of the lives of the aristocrats and intelligentsia, "which led only to despair," considers suicide, and finding that the peasants know life's purpose, he sees his sinful life for what it was, totally turns himself round, gives himself to Christ's beatitudes which will give his soul peace, etc. In fact, Tolstoy had always displayed unease with aristocratic and intellectual life, a respectful and helping regard for country virtues and his peasants, whom he had decades ago freed, putting together a school for them and writing ABCs -- that and so many other efforts, while all the while writing War and Peace and Anna Karinina. Perhaps the most telling little detail Wilson shows us is that the account Tolstoy gives of the moral stance of the St. Petersburg writers (i.e. Turgenev and Fet) he joined in the late 1850s is not in the least true of them but fits quite well the Dostoyevsky who latterly had developed into Tolstoy's chief literary and religious rival -- a rival Tolstoy never mentions by name though he refers to dozens of lesser Russian authors.

Ludwig Wittgenstein was an intensely private man, one who published nothing after his first book, though he winnowed, prefaced, and shaped the two precipitates, The Philosophical Investigations to a degree tantamount to publication, and though he also left a much later body of notes, or let his students (often professors or soon to be that or more: Turing, Moore, Anscombe, Geach, Toulmin, Malcolm, etc. ) collate their notes, all of which now could sprawl across a book shelf, not to mention the library you might form from the several hundred thousand journal articles and citations, the horde of biographical and hagiographical material dispersed through countless columns, articles, letters, biographies, novels, movies, paintings, etc., etc., along with a choral incanting Tractatus' seven cardinal sentences to music and Palozzi's painting Wittgenstein Meets Marilyn Monroe.

While Ludwig Wittgenstein knew Tolstoy's A Confession, and manifestly showed in private personal conversation, and occasionally in personal jottings and letters, some Toystoyesque broodings, petulant screeds, mawkish apologies, and hysterically wrought confessional outcries, he clearly shrank from any public expression of this whatsoever. It is barely possible that some lurid scribblings lurk in some Cambridge attic, some Lamentations of Ludwig, to which A. N. Wilson's comments about A Confession might apply have hypothetical application, as they might to some Nachlass items. But this would be, so to speak, empirical biographical information about Ludwig, information with no relevance to our understanding of Investigations or its Is (and similarly for the oral broods we already know about). Indeed, I would like to think that as the Wittgenstein who wrote Tractatus, and resonated with Tolstoy, grew into the man who was to write Investigations, he became more and more painfully sensitive to the self-delusive and unintentionally self-revealing aspects of Tolstoy's biographical and spiritual writings and that this (and Augustine, of course, whose Confessions he'd read first in boyhood) helped him see the powers and all too evident pitfalls of the confessional mode, its peculiar felicities and infelicities, its distinctive standards and instructional uses, particularly if it were adapted to non-empirical narrative, to the Is of Meditations and Tractatus. It is just philosophically unrewarding to think that Tractatus and Investigations are clues to something called "what Wittgenstein really thought," or that anything like that could matter at all as much as what, so to speak, Investigations thinks, what the I there, and the attendant personal voices, achieve as a speech act. Wittgenstein certainly taught us this, though the technicalese term "speech act" was out for him as was anything like Austinian systematic speech act linguistics.

(It is thus orthogonal to the thesis of this paper that, in his private life, Ludwig was on occasion passionately given to personal, nonreligious, acts or near acts of confession. The self-critical near confessional mode appears in a number of letters and in others' accounts of conversations with him; more explicitly, for example, we have what Ludwig refers to as "the confession," which he had sent Engelmann, but which he had also delivered one-on-one to his closest and most respected associates around the New Year of 1937, including close family members, G. E. Moore, Maurice Drury, Fania Pascal, and Rowland Hutt, with a dreadful time on all accounts had by the participants, often illustrating the difficulties about the form that we have mentioned. Several such as Moore felt it excruciatingly to be forced into the painful confessor role, including Fania Pascal, who recalls that Wittgenstein confessed that while he had let people believe his ancestry was three-quarters Aryan he was in fact three-quarters Jew (Engelmann 1967, p. 58; Pascal 1979, pp. 45-50; Monk 1990, pp. 367-72). Pascal, a Russian Jew who had grown up amidst the threat of pograms, was latterly startled to learn that two of Wittgenstein's "Jewish" grandparents were baptized as children and the other was baptized on marriage Pascal writes "'Some Jew,' my grandmother would have said" (Monk, over-simplifying, attributes the "some Jew" to Fania herself). There was general agreement that Wittgenstein took his worst offense to be a beating he gave to one of his girl pupil which he, subsequently, denied to authorities having administered. Subsequently, Wittgenstein revisited the tiny mountain village where he had taught over a decade before and apologized to at least four of his former pupils. I called Wittgenstein's confession "non-religious" because, throughout his life, Wittgenstein did not indicate or express any religious commitments or beliefs, and he explicitly denied such on a number of occasions, though he shifted, during World War I, from vocal contempt for believers to a grudging respect in some cases for what he could not join; thus, in "Remarks on Frazer's Golden Bough," which Rush Rhees compiled from various notes written in the 1930s, Wittgenstein writes,

If I, a person who does not believe that there are human-superhuman beings somewhere which one can call gods -- if I say: "I fear the wrath of the gods," that shows that I can mean something by this, or give expression to a feeling which is not necessarily connected with that belief. (Luckhardt 1979, p. 68).

Here Wittgenstein thus also denies Frazer's claim that the ritual of the priest-king of Nemi rests on
superstitious false beliefs or factual errors.

No opinion serves as the foundation for a religious symbol. And only an opinion can involve an error. . . . The religious actions, or the religious life, of the priest-king are no different in kind from any genuinely religious action of today, for example, a confession of sins. (p. 64).

While I think Wittgenstein is right to think that religious forms can well have valid uses apart from the truth or falsity of various beliefs, and even, hence, outside the official canons of the various sects, I do believe that Investigations was a rather more successful application of this insight than that of "the confession" of 1937, even granting that Wittgenstein certainly expected confidentiality from his confessors, which he had except for Pascal and Hutt.)

III

I lighted upon a certain book of Cicero. . . . In Greek the love of wisdom is called philosophy, with which that book inflamed me. . . . There also is disclosed that most salutary admonition of Thy Spirit, by Thy good and devout servant: Beware lest any man spoil you through philosophy and vain deceit . . . I resolved, therefore, to direct my mind to the holy Scriptures, that I might see what they were. But behold, I perceive something not comprehended by the proud, nor disclosed to children . . . and I was not such as could enter into it, or bend my neck to follow its steps. For not as I now speak did I feel when I turned to those Scriptures; but they appeared to me to be unworthy to be compared to the dignity of Tully: for my swelling pride shrunk from their lowliness, nor could my sharp wit pierce their inner meaning. Yet were they such as would develop in little ones; but I scorned to be a little one; and, swollen with pride, looked upon myself as a great one. (Augustine 1948, p. 37.)

Philosophical Investigations is a confession: perhaps the most masterful and instructive use of the form ever issued. There is an economy and simplicity of wording, a sharpness of vision, a deft and utterly convincing sense of what the various voices, including the reader, must say; the voice's words chant in against each other liturgically as if they were carefully honed spiritual exercises (only not for the glory of God but for the cleansing, instructing, and reverencing of our old city, our everyday life and thought.) It is also, to again appropriate vocabulary, a valid confession: it is truthful. Wittgenstein is exquisitely aware of this requirement and of how the criterion should be understood to apply in this case, particularly, as we will see, in Part II, Section XI.

In the preface, Wittgenstein says that his book contains "sketches of landscapes" made in criss-crossing "a wide field of thought" which would "be crippled if I tried to force them on in any direction against their natural inclination,"

The same or almost the same points were always being approached afresh from different directions, and new sketches made. Very many of these were badly drawn or uncharacteristic, marked by all the defects of a weak draughtsman. And when they were rejected a number of tolerable ones were left, which now had to be arranged and sometimes cut down, so that if you looked at them you could get a picture of the landscape. Thus this book is really only an album. (p. v)

[One however notes the appropriateness of a crisscrossed album to Wittgenstein's most basic theme: that a regular systematic map is a seductive but unavailable illusion -- what better than the confessional mode to teach such a lesson?]

To begin this book of "sketches" with the long quote from Confessions, counterpointed with "five red apples," is both to keynote and contrast the two general views of language (reference, meaning, thought, etc.) that weave through the book and to establish the confessant/confessor (etc.) narrational dynamic that animates it (since the preface announces that it is not his "philosophical investigations" but a "precipitate" we are free to consider the hint here offered). But Wittgenstein is much more explicit -- toward the end of Section XI -- a position that also highlights his use-indicating comments. The first part of Investigations consists of arabic numbered paragraphs, mostly a few sentences long, a few of page length. The shorter second part of Investigations is comprised of fourteen roman numbered sections; the first ten take up fifteen full pages, the last three are each half page in length and obviously are postscripts (though important), but section XI, however, is thirty-six pages, more than twice as long as the rest. Given the painstaking care that Wittgenstein devoted to shaping his album for publication, one is led to expect a master sketch, a recapitulatory play-within-a-play, and of course a denouement, conclusion, or signature. (For Part I, which comprises three/fourths of the book, the executors had to make a few choices, as there were several, minorly-differing versions; Part II is MS 144, which Wittgenstein extracted and fair-copied from his 1946-1949 notes. (Wright 1982))

I want to call XI the Ecce homo! of Investigations. It begins with the humblest perceptions of "seeing as," where we learn that cognition and experience seamlessly, and generally unnoticeably, fuse into what we see, and we learn -- under the I's gentle but persistent instruction -- to resist the seductive thought that "the interpretation" is a separable "hidden" (inside) mind picture that must be available if we concentrate sufficiently (there may of course be an explanatory and causal neurological process but that is "natural science" ). Now we weave up to the "seeing as" aspects of words, sentences, and discourses, again with instructive examples, that show us that intentionality fuses with speech and facial expression, with words said that are deeds done, thoughts thought, within the whole of language and related sign use and our way of life (emotions can be in the face and in the phrase). Finally, we fire up the full-fledged I (Wittgenstein minus empirical peculiarities, like the I of the cogito), the thinker who might be tempted to say "I know I exist." In two stunning paragraphs, to my mind rather like the two sequences of four notes that open Beethoven's Fifth Symphony, Wittgenstein dispels (exorcizes?) Moore's claim that he knows he has two hands and Russell's claim to know that the world has existed for the last few minutes (though perhaps not before then). One now hears differently when the siren I, in Meditations, enchants "But I now know with certainty that I am and also that all these images--and generally, everything belonging to the nature of the body--could turn out to be nothing but dreams.(Meditations p. 28)" The Confessor I of Investigations inverts and disenchants the cogito with a simple plainsong lesson (and a parenthetical comment),

I can know what someone else is thinking, not what I am thinking.
It is correct to say "I know what you are thinking", and wrong to say "I know what I am thinking." (A whole cloud of philosophy condensed into a drop of grammar.)

Baring his soul (so the speak), the confessor/confessant I of Investigations now must turn the lesson, conclusively and concludingly, on himself, on the criterion for understanding his thoughts (which also has to mean, at this point, Wittgenstein's as Investigation's author).

We hear first the confessant's quote-marked voice, the proud voice that leads us to solipsism.

"A man's thinking goes on within his consciousness in a seclusion in comparison with which any physical seclusion is an exhibition to public view."

Then the unmarked confessor I makes two disenchanting moves designed to remind us how to prevent a picture from "holding us captive." The first is to have us imagine a physical change (i.e., something that could happen and so should not affect a logical claim) which would undermine the picture by "outing" the mental (a technique emblemized in the drawer labeled "apple," the color chart, and rote counting, etc., of the "five red apples" leitmotiv)). The second is to remind us that in fact there are such "outed" cases in our everyday life in which it makes perfectly good sense to say that someone's thoughts really are hidden, though such a case provides no support for the confessant's spell-binding picture. Indeed it positively dispels it in that it shows us a healthy everyday use of "his thoughts are hidden," one which by the same token suggests a straightforward use for its denial, for his thoughts being known, rather than leaving us to be seduced by a useless cardhouse-breeding notion of "hidden thoughts."

If there were people who always read the silent internal discourse of others--say by observing the larynx--would they too be inclined to use the picture of complete seclusion? If I were to talk to myself out loud in a language not understood by those present my thoughts would be hidden from them.

And now, at last, behold the man:

Let us assume there was a man who always guessed right what I was saying to myself in my thoughts. (It does not matter how he manages it.) But what is the criterion for his guessing right? Well, I am a truthful person and I confess that he has guessed right. -- But might I not be mistaken, can my memory not deceive me? And might it not always do so when--without lying--I express what I have thought within myself?--- But now it does appear that 'what went on within me' is not the point at all. (Here I am drawing a construction line.)

The criteria for the truth of the confession that I thought such-and-such are not the criteria for a true description of a process. And the importance of the true confession does not reside in its being a correct and certain report of a process. It resides rather in the special consequences which can be drawn from a confession whose truth is guaranteed by the special criteria of truthfulness.

This is the signature: for what on earth has Wittgenstein been telling us but that he "thought such-and-such"? The first paragraph says: I am my words here. The second: I have here sketched how I might or might then have thought (in Tractatus, in Meditations, and in a medley of simpler confusions); to do this well is to confess under the criterion of truthfulness. Wittgenstein is insisting that the truth (or confessional validity) of Investigations is not evaluable as a description of the process through which he (or Descartes, etc.) did in fact manage to bewitch himself. Description of such a process would be an empirical and causal account within some future cognitive science (if such proved possible), or a biographical account that provides data for such, as one postscript, summarizing a litany, suggests

But our interest does not fall back upon these possible causes of the formation of concepts; we are not doing natural science; nor yet natural history--since we can also invent fictitious natural history for our purposes. (Part II xii)

Wittgenstein adds a parenthetical comment to the signature paragraph to clarify the distinction between confessional truthfulness and a true description of a process.

(Assuming that dreams can yield information about the dreamer, what yielded the information would be truthful accounts of dreams. The question whether the dreamer's memory deceives him when he reports the dream after waking cannot arise, unless indeed we introduce a completely new criterion for the report's 'agreeing' with the dream, a criterion which gives us a concept of 'truth' as distinct from 'truthfulness' here.)

I don't think there is any coherent way to read this except:
"information about the dreamer" = "that I thought such-and-such";
"truthful accounts of dreams" = "truthful confessions of thinking such-and-such."

Wittgenstein reminds us that Meditations speaks this way as well: Insofar as its I cannot tell whether or not it dreams and yet makes brashly authoritative claims about its thinking, its reasoning comes under some criterion of "truthfulness." More generally, we look for meditative coherence and sincerity in the I's speaking to itself, not criteria-specified correspondence. It would not matter in the least philosophically were someone to establish that in fact Descartes wrote Meditations lying in his bed rather than at his writing desk; it does perhaps matter, if we take his "letter to the sacred faculty of theology at Paris" as a preface, that its pledge to observe the proper aims and limits of Christian meditation is clearly violated by the text. Wittgenstein, of course, does not rail at this but rather at a possible lack of clear vision and truthfulness -- for example when he wonders what he (or you) could see or mean in saying sotto voce "Those children over there are mere automata" (par. 420). Empirical dream research has made Wittgenstein's analysis particularly perspicacious, for indeed the rudiments of a new criterion have begun to appear: one can imagine a research subject, awakened by a researcher, saying "You interrupted a fabulous dream: I was . . .," to which the researcher replies, looking at the electroencephalogram tracings, "Not true, I didn't interrupt you: your R.E.M sleep stopped fifteen minutes before I woke you." Similarly, Wittgenstein says he would only uselessly decide whether to record S in his diary when he suspected his private sensation , which he'd dubbed S, had recurred, but he also pointed out that if S turned out to be correlated with a rise in his blood pressure, then it would be useful and the criterion would be tied more to rises in blood pressure than to an indefinite shade of a subjective feeling.

The confessional I of "that I thought such-and-such" cannot hence be anything more or less than the "I" comprised by the words of Investigations. Of course, Investigations exemplifies fumbles and falsities in thinking (confessant avowals) and the confessor's corrections, and the exercises -- "You will think that . . . But . . .," etc. -- that follow like choruses, all honing, cleaning, clarifying, and bearing witness to the old city's unavoidable and defining (our lot and opportunities) informal instruments, the make do, strange, tempting, and incommensurable structures of our personal linguistic, communicative, cognitive, emotive, and cognitive way of life. Obviously the confessional form fits hand and glove the aim of reviewing the cognitive temptations that lead to "philosophy" (confessant sense) and holding them up to a cleansing examination and demystification. But what then can confessional truthfulness really amount to? To be sure the sentences must (like the linguist's articulate very long sentence) form a coherent anaphoric, implicative, and interactional narrative. Narrative verisimilitude to be sure -- convincing stories, "draughtsman"ship -- but what more and what particularly?

Confession calls for the plainest homely language, no word wasted, spoken in wholly serious simple truthfulness alone, with reverence for the commonplace, owning and conveying the commonality of temptation and illusion, accompanied by the unassuming correction provided by the confessor, who claims no special status, who only reminds us of (bears witness to) simple, instructive truths perspicaciously surveyed, and all this delivered without a touch of ill feeling, contempt, arrogance, bias, frivolity, or any further (perhaps covert) theoretical agenda. Ill-feeling, contempt, arrogance, carelessness, bias, frivolity, floridity, insufficiently intense concentration, jargon, word intoxication, all may distort the descriptions people give of intentional or cognitive doings in their biographical lives; we can often hear one or several of these quite various faults in the description itself, but in important cases -- criminal or civil legal proceedings for example -- we are disposed to defer to a large degree to the doings themselves as determined empirically. How much more strict and demanding our judgement of truthfulness when we cannot seek help from external biographical sources or natural science. (The confessor of the Christian rite is to some degree in a similar epistemic boat; while he may introduce commonplace facts, he is not supposed to engage in an independent factual investigation of the confessant's assertions.)

As an example of what these strictures might exclude, consider the opening three and last two sentences of the first paragraph of J. L. Austin's "Truth."

'What is truth?' said jesting Pilate, and would not stay for an answer. Pilate was in advance of his time. For 'truth' itself is an abstract noun, a camel, that is, of a logical construction, which cannot get past the eye even of a grammarian. . . . What needs discussing rather is the use, or certain uses, of the word 'true'. In vino, possibly, 'veritas', but in a sober symposium'verum'. (Austin 1979, p. 117)

The "camel," "past," and "eye" presumably stem from the Biblical assurance that a wealthy man's chances of heaven are comparable to those of a camel attempting to pass through the eye of a needle. But then what sense can we attach to "which cannot get past the eye even of a grammarian"? "Even of" suggests that grammarians are criminally unwilling to recognize that "truth" is an abstract noun while for everyone else the answer is obvious, it is an "abstract noun" "that is, of a logical construction," as everyone recognizes immediately except grammarians who need more time. But the sketched metaphor is absurd: the root metaphor is one of "passing through" an "opening," supposedly sternly narrow, while Austin's application has the eye not as an opening to pass through but as a scanner one passes by (pass means success in the root case and failure in Austin's maladaptation) -- and who but a grammarian would feel called upon to identify "truth" as an abstract noun? "Even a grammarian" is surely "no one but." Austin's title is "Truth," but by his opening paragraph's end the topic of the paper's title -- truth -- is ridiculed offstage as too hopelessly profound, and we go on with "true": Austin's misinformative title is there to trigger some horsing around.

And then there is the cute bit about Pilate being "in advance of his time." That phrase and "even a grammarian" suggest that Pilate recognizes that "truth" is an abstract noun and that he somehow has the advanced wisdom to recognize that discussion of it is ill-advised hot air (as the Anchor Bible commentary drily counters, "Even John is not likely to have painted a venal politician as a philosopher" [Brown1970, p. 869]).Moreover, the clear suggestion that Pilate would have stayed for answer if Christ had used the adjectival true is nonsense, for he is off to try to dispel a vengeful mob by insisting, "I find no fault with this just man"). Since Pilate was responding to Christ's "Everyone that is of the truth heareth my voice," it is perfectly obvious that in context St. John means by "the truth" what the Christian has faith in, what he sees and reverences. Since all of this is ground floor biblical information (which Austin of course knew), are we then to conclude that Austin meant to insinuate that St. John's platonistic Christianity is a particular case of the reifying spell cast by "truth"? No. Does he mean that St. John and his brethren are drunk? No. Austin doesn't mean anything, nor is there anything philosophically instructive in his wrenched metaphorical use of "eye." Austin is just horsing around.

Similarly, more briefly, and more flatfootedly, Austin remarks, in "Performative-Constative," to the Royaumont Conference, "But there are more ways of killing a cat than drowning it in butter;*" (his footnote reads "*English proverb. I am told that this rather refined way of disposing of cats is not found in France." (Austin 1963, p. 27). Does Austin really mean to say that on occasion an upper class Englishman will dispose of Puss by, so to speak, buttering it? No. Does Scotland Yard find itself at times confronted with a cat tail sticking up like a candle's wick from a silver family heirloom murderously full of congealed butter fat? No. Indeed, surely he was responding to Frenchmen who told him not that they lacked such a practice, but rather that they did not have such a proverb. And surely they were right to carp, for English doesn't of course have any such absurdly-inflated proverb either. As one would expect, in the eldest, sixteenth century proverb "milk" is the instrument, upgraded to "cream" in the eighteenth century; the first and only substitution of "butter" occurs in a joke book published a few years before Austin spoke.

Given the rich, endlessly inventive and supercharged rhetorical prose and pose that Austin typically adopted when philosophizing, these instances appear as slightly over-done, off-key passages in Austin's baroque repertory. By the confessional criteria of truthfulness, however, there is every sort of offense I mentioned four paragraphs back. Even when Austin lands his best, most characteristic, punches there is kind of excess, a kind of gleeful malice, that makes our chuckle turn awry; we learn perhaps but are not so much instructed as scared, as some lines about pretending suggest,

On a festive occasion you are ordered, for a forfeit, to pretend to be a hyena: going down on all fours, you make a few essays at hideous laughter and finally bite my calf, taking, with a touch of realism possibly exceeding your hopes, a fair-sized piece right out of it. Beyond question you have gone too far. Try to plead that you were only pretending, and I shall advert forcibly to the state of my calf -- not much pretense about that, is there? There are limits, old sport. This sort of thing in these circumstances will not pass as '(only) pretending to be a hyena.' True -- but then neither will it pass as really being a hyena. (Austin 1979, p. 256).

While there is much fun to had here (of a slightly malicious sort) there is not quite as much philosophy. The luxuriant details make no philosophical point at all, but convey much about social class minutiae: the slim philosophical point floats in Falstaffian fat, and again Austin has to maneuver (this time more fluently) simply to get "pass" set up so the punch line arrives in parallel. The papers of Philosophical Papers make it forthrightly evident that their author is a Fellow of a traditional Oxford College, a Lit. Humaniores Don and proud of it, used to playing forfeits, dining at High Table, and having friends whose gardens are sufficiently vast that someone might informatively report, at the house end, that there is a bittern "at the bottom of your garden," someone indeed given to the would-be-aristocratic (or High Tory) intuition that, on the whole and sadly, the only decent, straight-talking folk, aside from themselves, are the honest English yeomen (undefiled by jargon ridden bogus technicaleses of lower middle class union types, social workers, scientists, businessmen, and bureaucrats -- i.e., by subliterate people with non-U, non-Oxbridge, credentials).

Except where absolutely unavoidable, the austere, plain speaking I of Investigations, while creating the most intimate familiarity between us about our cognitive/linguistic foibles, has no individuating personal qualities whatsoever: no class, no nationality, no occupation, no political, religious, ethnic, cultural identity except that of confessorial philosophizing and having the basic cognitive/perceptual competencies common to us all, an I who in his confessant voice owns to thoughts such as those of "the author of Tractatus Logico-Philosophicus," but whose most self-individuating description is perhaps "there is a hammer, pliers, a saw, a screw-driver, a rule, a glue-pot, glue, nails and screws.----- The functions of words are as diverse as the functions of these objects." It is of course "unavoidable" that the I speak some human language (and indicate a familiarity with a few philosophical texts, and bare smatterings of scientific and mathematical information) but even here, the parallel presentation of the German and English texts, so easily and so often intertranslatable word for word, makes the language seem shorn of nationality, region, temporality, ethnicity or class. Wittgenstein's confessional mode abets this in sticking to the simplest words (often cognates) and avoiding any of the poetic devices, intranslatably bound to a language's sound system, or class accent, in which Austin's prose abounds. In this respect Wittgenstein can remind one of the "daughtsmanship" of some of Jorge Borges' little philosophical pieces, such as "Borges and I" or "Everything and Nothing," which translate so well into English (as if they, like the paragraphs of Investigations, are really written in the language of thought, or slightly more modestly, some common Indo-European creole).

Consider once more and by way of comparison "five red apples"; this time I add the confessor/confessant chorus:

Now think of the following use of language: I send someone shopping. I give him a slip marked "five red apples." He takes the slip to the shopkeeper, who opens the drawer marked "apples"; then he looks up the word "red" in a table and finds a color sample opposite it; then he says the series of cardinal numbers -- I assume he knows them by heart -- up to the word "five" and for each number he takes an apple of the same color as the sample out of the drawer. -----It is in these and similar ways that one operates with words. -----"But how does he know where and how he is to look up the word 'red' and what he is to do with the word 'five'?" -----Well, I assume that he acts as I have described. Explanations come to an end somewhere. -----But what is the meaning of the word 'five'? ---No such thing was in question here, only how the word "five" is used.(PI 1)

Here we are led, tersely, through the instance and exercise. The minimal sketch shows us that
this use of language needs no hidden "mental events"; the boy simply presents the slip and the shopkeeper opens the apple drawer, locates the color sample, and plucks apples, saying "one," "two," etc. until "five" in parallel, and then hands the result to W.'s shopper. (All in the starkest contrast with Augustine's account in which meaning is "learned" through associating sounds and things inwardly so that the "essential" link is eventually (and uselessly) made between the mental word and mental thing.) In despite of this, the confessant interlocutor tries to get the old regressive picture going again, "But how does he know . . .," but he is firmly squelched. To give even this still very incomplete exegesis is to realize that an essay would be required and then of course fail abysmally to match the original: it is like being asked to explain all that is packed into "Like diamonds, we are cut by our own dust" (Webster, The Duchess of Malfi).

A meditation (one kind of monologue) has but one speaker, who may be regarded as addressing himself; a dialogue has at least two interlocutors; a confession, in its literal ritual sense, is a three person speech act. In the rite of confession, the confessant confesses to God, with the confessor taking the role of editor/instructor, who listens, clarifies, questions, reformulates, and so on, ideally until the goal of truthfulness is fully reached, so that a true penance and absolution may be effected. The selection of an appropriate confessor is thus one of considerable importance. The rite of course is a most private one, whose absolute confidentially on the confessor's part, has long been recognized in both church and secular law. While the confessant is not bound to confidentially in quite the same way, to give an account to others of what one said, or was said to one, in the confessional is a violation, while it is of course proper to admit to the same acts to relevant secular authorities. All these considerations suggest that to publish a confession (as opposed to a meditation, dialogue, etc.) is an act fraught with problems. First off, while Descartes can suggest that he is simply offering to the world a record of a meditation he had had with himself, and Plato have Socrates, in Republic for example, simply say that he is repeating from memory a conversation he had had yesterday at Piraeus, without it mattering to our judgement of their work whether there is much literal truth to these claims, it would be another matter if a confessant or a confessor published a purported record of an actual confession.

Neither Augustine nor Tolstoy purport to be doing exactly this but rather offer to the world an instructive account of the errors of their earlier days, their travail, and eventual correction and redemption, thus producing something analogous the confessant/confessor interlocution, with the world as well as God listening. Still, flaws are amply possible here. In the rite, the confessant and the proper confessor interact to produce a confession which is instructive to the confessant: someone secretly and improperly listening is a voyeur not another confessant or confessor. Indeed, you can develop a uneasy or false feeling in reading or hearing public confessions. What right have I to your ears? What legitimating motive have I? Why do you need to know, to witness, this lurid breast-beating? I recall what my greatgrandmother, Flora Holcomb, wrote for her children about what she did, age 11 in 1862, inspired by some Pennsylvania revivalists

I remember one girl coming up from the water shouting & spluttering "Glory! Glory!" as she ran up the bank. If the person baptized caught cold it was a sign he did not have the holy spirit, but was an imposter. At church they had the "power" and became rigid. Once surrounded by four or five grown-ups on their knees I rose and said I knew I was a great sinner and asked the prayers of the congregation.

That night at Uncle Dolph's as they had devotions before going to bed he prayed the Lord to bless "thy handmaiden if such she really be" and I felt like a little hypocrite. (Leiber, et al., 1983, p. 291)

Confessional literature is fraught with related difficulties. If it is to follow the logic of the rite, it might be presumed to be addressed to those who may be tempted to the confessant's sin. Then, at least, it might be said that where would be no idle voyeurism, and the retrospective confessant now confessor may offer such a reader a realization of his own similar temptations and offenses, and possibilities of correction and redemption. By playing all the personal pronominal voices flawlessly in an austerely cognitive plainsong narrative that powerfully draws in those tempted and those alone, Wittgenstein refines the potential weaknesses of the form into strengths.

Tolstoy's confessional writing was Wittgenstein's great negative example, one that showed him how not to confess, one that was such an effective negative example because there was so much in Tolstoy's life and character to attract him. Wittgenstein bore witness to the way in which the dross of Tolstoy's false, hysterical, biographical broodings fell off from the true silver of his fictions. Augustine suggested some of the structural, instructional, and narrative possibilities of the confessional form, and perhaps as well made it clear (or clearer, since Wittgenstein had long seen the virtue of passing by in silence) that at least in our time and for him biographical and religious themes must give way to exclusively cognitive/perceptual ones. That Wittgenstein succeeded makes it possible to say that philosophy, which for many of us began in dialogue and entered its modern moment in meditation, has come to one sort of culmination in confession. Pace Tolstoy, Investigations indeed is the book which its author intends us to read.

IV

It is humiliating to have to appear like an empty tube which is simply inflated by a mind. (Wittgenstein 1980, p. 7)

To see clearly, to rehearse and instruct, is the confessional aim: this means not to "sublimern," except as confessant example apt for demolition, and not to theorize or seek causal explanations. Wittgenstein announces the same corrective aim, and exclusion, in his stress on "countless uses" and "description alone":

we do not command a clear view of the use of our words. -----Our grammar is lacking in this sort of perspicuity. A perspicacious representation produces just that understanding . . . It earmarks the form of account we give, the way we look at things. (par. 122)

When he says our grammar lacks perspicuity, this is not a criticism and still less is there any suggestion that a better grammar can be produced (it is exactly that hope for a systematic "subliming" that produced Tractatus and similar illusionary work). "Perspicacious representations" are precisely what the "reminders" of Investigations are intended to be: cognitive vignettes that knock down what are, for those with philosophical inclinations, sublimely attractive cardhouses. They are not intended, and will not serve, as some clumsy and informal presentation of an Austinian new science of speech act linguistics, or of any other sort of systematic science or philosophy that would -- to use current terminology -- systematically reduce or explicate our folk psychological or our intentional idiomatical understanding of ourselves as whole everyday persons to some explanatory natural language semantics, language of thought, neurosemantics, or cognitive science. Our reverent appreciation of the tangled old city demands we resist the temptation to blink at its tangled, self-contradictory peculiarities, to bewitch ourselves into thinking that it is really only our mis-seeing, or superficial misunderstanding, of what is merely the inner city extension of the systematic suburbs. We can exit the old city when we adopt the neatly designed "languages,"the artificially and stipulatively formalized constructs of our sciences, sometimes indeed in order to map the modularized substructures of our subpersonal cognition, but this whole round of understanding is orthogonal to anything that will make better sense of our intentional whole-person situation within the homey, tangled, incommensurable, and contradictory structural concepts of the old city.

For there is a reverse stamp to this confessional coin. For in his relentless effort at pure descriptions that will squelch our temptation to "sublimern" the logic of our ordinary conscious cognition and language, Wittgenstein at the same time inevitably highlights intellectual traps that can as well snare cognitive scientists, and may also suggest important avenues for subpersonal, or modular, psychological research. What is, at the level of the person, a temptation to "sublimern," can be grist to the mill of modular subpersonal cognitive science, what Wittgenstein calls "natural science" in the penultimate section of Investigations (to avoid confusion, let me add that what Wittgenstein calls "psychology," and holds to be barren because of "conceptual confusion," might be called a sort of person level, "little man in the head" psychology).

If the formation of concepts can be explained by facts of nature, should we not be interested, not in grammar, but in that in nature which is the basis of grammar? . . . But our interest certainly does not fall back upon these possible causes of the formation of concepts; we are not doing natural science. (p. 230)

Why should the investigation of "what is in clear view" (and "grammatical" (or apriori)) in our everyday conscious cognitive/perceptual life) -- "in despite of" our tendency to mistake or "sublimern" it -- be grist to the mill of modular subpersonal cognitive science? The short answer is that what can lead a philosopher of mind/language to build a cardhouse can equally well lead a psychologist, linguist, or neurologist to do so; and to identify robust data of just this sort is to identify anomalies that permanently resist smooth explanation at the personal conscious level (what we try to blink away), and hence may suggest possible explanation at another level (natural science). But let me, briefly, give three examples of this.

1) The first, from Part II Section IX, is the duck/rabbit discussion (and related ones about "seeing an aspect). Without hoping to recapitulate the instructive complexity of Wittgenstein's discussion, whose final aim, of course, is to undermine dualism, let me simply point to his insistence that when someone, seeing the figure for the first time and "seeing it as a rabbit," reports "It's a rabbit," his perceptual experience is indistinguishable from someone who has seen it both as duck and rabbit and now sees it as a rabbit, and that his perceptual experience is subjectively indistinguishable from all figural perceptions, indeed all visual perception (that is why Wittgenstein insists that someone who succumbs to the temptation to say, "Now it's a rabbit" is not making a perceptual report and should, to be honest to the phenomena, rather simply say, "It's a rabbit.). Our everyday intelligence, as Wittgenstein amply shows us, strongly resists the suggestion that we SEE according to an interpretation. Now the obvious point is that the drawing itself (the object independent of our perceptual activity) is neutral between the rabbit/duck interpretation (and that definitely doesn't mean that it flickers between rabbit and duck while no one looks at it). So one point of the example is that peculiarities of our perceptual apparatus inaccessible to consciousness (i.e. subpersonal, modular aspects) determine the "shape of what we see"; but another point is that we cannot tell through concentrated introspection whether such paradoxical seeings are available -- there is absolutely nothing apriorily "open to clear view." As Wittgenstein engagingly puts it,

We find certain things about seeing puzzling, because we do not find the whole business of seeing puzzling enough. (p. 212) [Wittgenstein also says, speaking of such paradoxes, with a wave to cognitive science]

When it looks as if there were no room for such a form between other ones you have to look for it in another dimension. If there is no room here, there is room in another dimension. (p. 200)

And, of course, he is right, and he makes a medley of related points about script sentences or drawings of faces looking wholly different when right side up or upside down. Wittgenstein always returns to the point that we often see according to an interpretation but that this is in no way evident in the preceptual experience and that there is no way to "peel off" the interpretation in actual experience (though of course we can in retrospect recognize ("in despite") that we indeed are seeing according to an interpretation -- all without changing our actual visual experience). The resilient and specific character of such visual paradoxes has been a major ingredient in the considerable enterprise that the study of visual perception has become. As the linguists have noticed, there are related syntactical "paradoxes," such as Flying planes can be dangerous sentences, which most native English speakers "hear" in one of two syntactical ways, without noticing the possibility of the other reading. Again, there is nothing in the conscious hearing experience of the native speaker that indicates to her that she is hearing "according to an interpretation"; nor is there any way to "peel off" the "interpretation" to get down, ultimately, to the "real" continuous and unintelligible sound stream that the monolingual Hindi speaker can hear. Such robust syntactical and phonological phenomena (the numbered sentences and native speakers' everyday structural intuitions) form the rich data base by which the linguist derives and tests her theories and explanations, in the technical and formally-focused discourse of the linguist's text, of her profession's language game.

2) In the decades following Investigations' publication, nothing seemed more central than Wittgenstein's anti-essentializing" use of the example of games, particularly in criticism of his own earlier quest for the form of the proposition, the meaning of words, the nature of objects, etc. Thus we have of course a useful warning for someone who might otherwise be disposed to write or think something like the Tractatus; but it is also -- particularly in the richness and sharpness with which cognitive temptations are pictured and dispelled -- splendid instruction for any thinker who investigates phenomena through the lens of an everyday lexical item: always keep in mind the possibility that there is no unitary phenomenon or concept, no set of sufficient and necessary conditions, but perhaps just several typical examples, a "family of resemblances" with no common thread running through all! But, as with duck rabbit, we have not only a useful warning for theorists but also we find displayed appealing venues for empirical research into everyday cognition. Through work as various as that of Eleanor Rosch, et al., and that of Amos Tversky, et al., we find striking empirical accounts of the role of stereotypes and exemplars in our thinking: an account of the ways we may feel pulled down an illusory path serves as well to tellingly illustrate the peculiar contours of our actual folk psychological cognitive apparatus. Seeing how you can misunderstand everyday notions is also, most importantly, seeing how they actually operate.

I am inclined to give a similar account of a related theme: Wittgenstein's curious skepticism about naive, conventionalist, or formalist accounts of "rule-following," as exemplified in his discussion of the pupil who is going on with the sequence 2, 4, 8, . . . and does so "according to rule" until 100, when he continues 104 + 108 + 112 . . . In part, Wittgenstein means to dispel the vision that I must have thought out the series, including 102, 104, 106, 108, 110, etc. when I formulated the rule or told the pupil to go on, that somewhere inside of me accessible to consciousness there is the absoluteness of sense that I meant. More generally, Wittgenstein compellingly insists that "a rule," whether lodged in a sign post, voiced phonological sequence, or memory, must be embedded within the common and pragmatically viable fabric of practices of a community of signer users, fortuitously made possible by various background features of human nature and the physical world (such as that there are plenty of discrete, movable, and markable physical objects around with relatively enduring properties accessible to common human senses, such as that humans have some commonalities in cognition and perception that might explain why we go on in certain ways, why we fortunately avoid some of the many breakdowns in learning and communication that a Wittgenstein can imagine).

3) Beginning with "five red apples," Wittgenstein again and again shows us how tempting (and useless) it is to postulate private mental meanings as a necessary and apriori explanation for our public agency and sign usages, all this that had once been the high road to the atomic propositions, to the logical structure of the world dreamed in Tractatus, as well as more common varieties of solipsistic dualism. While Gilbert Ryle dubbed this mythology the "ghost in the machine," present day psychologists will more likely recognize it as "the little man in the head," stressing the many temptations to "explain" by placing a thinker with all the powers of an embodied human person inside the head. The Wittgenstein who wrote Tractatus was apoplectic that Russell's Introduction suggested that Tractatus described an ideal language, for Wittgenstein believed that his atomic propositions were a necessary "deep structure" of actual human language to the degree that it is actually meaningful, which would seem also to have to meant to him that there must be some regular logical/transformational relationship between the sentences of human languages and the atomic propositions (Tractatus insists that there must simples because there are complexes, that there must be definiteness of sense). But young Wittgenstein was hardly the only person so seduced, nor the last. The "generative semantics" movement in linguistics of the early 1970s collectively found compelling an even more expansive view: not only was the totality of the meaning of a sentence available (apriori so to speak) in its fully-articulated semantic deep structure but also its full status as a speech act within all possible social and material contexts. Thus J.R. Ross proposed that the deep structure of sentences contains an "illocutionary marker" that indicates their force as speech acts (promise, order, threat, question, naming, etc.) but which marker generally is "deleted" by something supposedly like the sort of narrowly syntactical process that allows us to understand that He saw the antelope and then the lion abbreviates He saw the antelope and then he saw the lion (Ross 1970). Similarly, generative semanticists proposed that the deep structure of the English word killed as it appears in sentences such as Jill killed Jack is decomposable into a concatenation of universal semantic atoms CAUSE + TO BE + NOT (ALIVE) (where "CAUSE," etc., means some concept employed by all human languages). But now everything in the act and context of utterance became part of syntax; this burden made the movement incoherent and it collapsed. As critics of the movement put it

[G]enerative semantics has distorted grammar by including within its goals a complete theory of acceptability. This assimilation of the phenomenon of performance into the domain of grammaticality has come about as consequence of an empiricist criterion for determining what counts as grammatical. In almost every paper Lakoff makes explicit his assumption that the explanatory goal of a grammar is to state all the factors that influence the distributions of morphemes in speech. On this view, any phenomenon systematically related to cooccurrence is ipso facto something to be explained in the grammar. Since in actual speech almost anything can influence cooccurence relations, it is no wonder that Lakoff repeatedly discovers more and more new kinds of "grammatical phenomena." (Katz and Bever 1976, p. 58)

As an historian of linguistics, Frederick Newmayer, latterly commented

The dynamic that led generative semantics to abandon explanation flowed irrevocably from its practice of regarding any speaker judgement and any fact about morpheme distribution as a defacto matter for grammatical analysis. In retrospect, it is easy to appreciate the a priori nature of the practice of reducing all linguistic facts to grammatical facts. It is no more logically necessary that, say, the proper use of please in English and the surface order of clitic pronouns in Spanish be treated within the same general frame work than it is that any two physical phenomena be necessarily derivable from the same set of equations. Other sciences take a modular approach to their subject matter: They take it for granted that a complex phenomenon is best explained by regarding it as a product of the interaction of a number of autonomously functioning systems, each governed by its own general principles. Generative semantics, on the other hand, attempted to force all phenomena into one and only system.

Attributing the same theoretical weight to each and every fact about language had disastrous consequences. Since the number of facts is, of course, absolutely overwhelming, simply

describing the incredible complexities of language became an all-consuming task, with formal explanation postponed [indefinitely]. (Newmayer 1986, p. 133)

As usual, Wittgenstein is much plainer and pithier when he makes the same point about his much earlier attempts at the same comprehensiveness:

In the course of our conversations Russell would often exclaim: "Logic's hell!" -- And this perfectly expresses the feeling we had when we were thinking about the problems of logic; that is to say, their immense difficulty, their hard and slippery texture. I believe that our main reason for feeling like this was the following fact: that every time some new linguistic phenomenon occurred to us, it could retrospectively show that our previous explanation was unworkable. . . Again and again a use of the word emerges that seems not to be compatible with the concept that other uses have led us to form. We say: but that isn't how it is! -- it is like that though! And all we can do is keep repeating these antitheses. (Wittgenstein 1937/1980 p. 30)

Wittgenstein's uses "grammar" (or "language game" feature) even more broadly than Lakoff because he is concerned as well with a vast melange of communicative sign use, facial expression, drawing, etc. that has little or nothing to do with natural language (writing systems, signals, artificial and formal languages, pictures, practices and procedures of all sorts). "Five red apples" is not only written (thus at one remove in artificiality from natural language), it is also not a well-formed linguistic structure; and Wittgenstein carefully blocks any easy avenue for saying it is linguistically elliptical (Wittgenstein just says "I send someone shopping. I give him a slip marked 'five red apples'": there are countless (and therefore no) sentential candidates, such as, Buy five red apples, Get me five red apples, I want five red apples, It's your job to get me five red apples, Get Schultz to bag you five red apples, I order you to bring me five red apples, and so on. The lack of regular linguistic structure in "five red apples" is, of course, no flaw for Wittgenstein, indeed it allows him to stress that there are no necessary formal properties for a use of language, for an effective act of communication: the written phrase does its work in this particular context, as a wink or a nod or simply an expressionless face may do as well in other contexts. Perhaps, as Wittgenstein suggests, there is no profound deep logic and system underlying all these countless uses of signs. On the other hand, a modest, and modestly modular, linguistics may feel constrained to focus attention on the systematic and narrowly syntactical phenomena of full-fledged natural language.

Wittgenstein was always fascinated by the idea of picturing something indirectly, by saying what it is not: in describing what can positively be said one gives the negative -- what cannot be said, what might be thus shown -- only one does not say it. The Tractatus explicates the empirical world -- all that is contingent, all that can be described meaningfully -- through a skeleton of sentences of which none is supposed to be an empirical assertion. You could consider Investigations as a transcendental deduction of the existence and some features of the organic neurological/cognitve/perceptual structures that subserve our everyday folk psychological experience while not open to clear view. After all, to confess that Cartesian dualism is false is to find one's body and brain, and the world, inescapably present. Today, the midwife metaphor, the Delphian claim to know only that one knows nothing, are for us more Wittgensteinian than Socratic.

And finally someone may say, perhaps in certain respects it does make sense to call Investigations a confession, but so what? -- as Aristotle pointed out you can do a physics treatise in verse, so you can do philosophy in a confessional style but this no more affects the validity of the philosophical views expressed than putting them in iambic pentameter. This paper is a lengthy attempt to reply to this question. But a more summary one is this. Investigations is a richly personal act, the sharpness and economy of instruction and insight marshaled into intense and wholly captivating truthfulness. Meditations, the confident assertion of the dualist, metaphysical I, had to be written in the first person, self-establishing, meditative mode: we now can see that in almost all of those possible worlds most akin to ours in which Descartes dies before writing Meditations a similar work, in first person meditative prose, will appear that plays a similar role in intellectual history to that of the one Descartes penned in ours (a rather more traditional way to express this, cribbed from Voltaire, might be: if there were no Meditations we would have to invent it). For us I think it may eventually become strikingly clear that Wittgenstein's confessional Investigations has had a similar career: we believe him, we validate his confession, we recognize ourself and chastise ourselves in his many persons. Yes, Wittgenstein's course is winding, for that is the real nature of the territory. Indeed, the same points are addressed from different directions, and some series of paragraphs might be changed in position, though not as many as one might think. But there is a continuous necessity of narrative form. We must have the confessant and confessional first person and the we and you of instruction: that is the profound and inevitable unity and force of his speech act. That dualism and solipsism today appear forlorn, fragile, unhuman, self-delusional, conceited, unheroic, and above all dishonest, is surely in large measure Investigations' achievement or at least its aim. If there are possible worlds in which Investigations is watered out and down in journal articles and self-help logical work books, they will be clumsier, dumber, emptier, and less truthful worlds. Investigations not only is the book Wittgenstein intends us to read: it does what he intended it to do -- or rather, it has the uses he wanted it to have. To quote the penultimate paragraph of Investigation's preface,

I should not like my writing to spare other people the trouble of thinking. But, if possible, to stimulate someone to thoughts of his own.

We must agree he has managed that.

Justin Leiber

Philosophy Department

Florida State University

Tallahassee, FL 32306

References

Alfaric, P. 1918. L'Evolution intellectuelle de Saint Augustin, I. Du Manichesisme au Neoplatonisme. Paris: Emile Nourry.

Austin, J. L. 1962. How to Do Things with Words. Oxford: Oxford University Press.

----- 1963. In Caton, C. E., ed., Philosophy and Ordinary Language. Urbana-Champaigne, Illinois.

----- 1979. Philosophical Papers. Oxford: Clarendon Press.

Bartley, W. W. 1985. LaSalle, Illinois: Open Court.

Brown, R. E. 1970. The Anchor Bible, vol. 29 (New York: Doubleday).

Chomsky, Noam 1995. "Language and Nature." Mind, vol. 104 No. 413 (Jan. 1995).

Engelmann, Paul, 1967. Letters from Ludwig Wittgenstein with a Memoir. New York: Horizon Press.

Firth, J. R. 1930. Speech. (reprinted in Firth 1964, p. 139-211).

----- 1957. Papers in Linguistics, 1934-1951. London: Oxford University Press.

----- 1964. The Tongues of Men and Speech. London: Oxford University Press.

Frongia, Guido, and Brian McGuinness 1990. Wittgenstein: A Bibliographical Guide. Oxford: Basil Blackwell.

Katz, Jerrold, and Bever, Thomas 1976. "The Fall and Rise of Empiricism." In An Integrated of Linguistic Ability, eds. Thomas Bever, Jerrold Katz, and D. Terence Langendoen, pp. 11-64. New York: Crowell.

Langendoen, D. Terence 1968. The London School of Linguistics: A Study of the Linguistic Theories of B. Malinowski and J. R. Firth. Cambridge, Massachusetts: MIT Press.

Leiber, Justin, James Pickering, Flora Bronson White 1983. "Mother By the Tens." Journal of the

Illinois State Historical Society, LXXVI no. 4, Winter.

Luckhardt, C. G., ed., 1979. Wittgenstein: Sources and Perspectives. Ithaca, New York: Cornell University Press.

McGuinness, Brian 1988. Wittgenstein: A Life: Young Ludwig 1889-1921. London: Duckworth.

Monk, Ray 1990. Ludwig Wittgenstein: The Duty of Genius. New York: Macmillan.

Newmayer, Frederick J. Linguistic Theory in America. Orlando, Florida: Academic Press, 1986.

Pascal, Fania 1979. "Wittgenstein: A Personal Memoir." In Lockhardt, pp. 23-60.

Pitcher, George, ed. 1966. Wittgenstein: The Philosophical Investigations: A Collection of Critical Essays. New York: Doubleday.

Redpath, T. 1990. Ludwig Wittgenstein: A Student's Memoir. London: Duckworth.

Rhees, Rush 1984. Recollections of Wittgenstein. Oxford: Oxford University Press.

Rosch, E. 1978. "Principles of Categorization." In E. Rosch and B. B. Llyod, eds., Cognition and Categorization. Hillsdale, New Jersey: Erlbaum.

Rosch, E., and C. B. Mervis 1975. "Family resemblances: Studies in the internal structure of categories. Cognitive Psychology Vol. 3, pp. 382-439.

Ross, J. R. 1970. "On declarative sentences." In Jacobs, Roderick A. And Rosenbaum, Peter S., eds., Readings in English transformational grammar. Washington: Georgetown University Press.

Routley, Erik 1962. Creeds & Confessions: The Reformation and its Modern Ecumenical Implications. London: Duckworth.

Russell, Bertrand 1968. Autobiography, II. London: George Allen and Unwin.

St. Augustine 1948. Whitney J. Oates, ed. The Basic Writings of St Augustine. New York: Random House, 1949.

St. Augustine 1992. James J. O'Donnell, ed. Augustine Confessions I: Introduction and Text. Oxford:

Clarendon Press.

Shields, Philip R. 1993. Logic and Sin in the Writings of Ludwig Wittgenstein. Chicago: University of Chicago Press.

Tolstoy, Leo 1879/1987. A Confession and Other Religious Writings. New York: Viking Penguin.

Turing, A. M. 1950. "Computing Machinery and Human Intelligence," Mind, 59, 236, pp. 433-460.

Tversky, A. and Kahneman, D. 1973. "On the psychology of prediction." Psychological Review vol. 80, pp. 237-251.

Wilson, A. N. 1988. Tolstoy. New York: Norton.

Wittgenstein, Ludwig 1922. Tractatus Logico-Philosophicus. London: Routledge & Kegan Paul.

----- 1937/1980. Culture and Value, ed. G. H. von Wright, trans. Rush Rhees. Chicago: University of Chicago Press.

----- 1953. Philosophical Investigations. New York: Macmillan.

Wright, Georg Henrik von 1982. Wittgenstein. Oxford: Basil Blackwell.