What does "signify" signify? more

PHILOSOPHICAL PSYCHOLOGY , VOL. 14, NO . 4, 2001 What does “signify” signify?: a response to Gillett RUPERT READ ABSTRACT Gillett argues that there are unexpected con uences between the tradition of Frege and Wittgenstein and that of Freud and Lacan. I counter that that the substance of the exegeses of Frege and Wittgenstein in Gillett’s paper are awed, and that these mistakes in turn tellingly point to unclarities in the Lacanian picture of language, unclarities left unresolved by Gillett. Lacan on language is simply a kind of enlarged/distorted mirror image of the Anglo-American psychosemanticists: where they emphasize information and representation, he emphasizes evocation and connotation. Neither contrasting emphasis is remotely adequate to linguistic action-in-the-world. Is “the unconscious”, as Lacan claims, a “network of signi ers”? Arguably, yes; but most ordinary/ actual language does not involve such “signi cation”. Words primarily “signify” concepts or things only in exceptional circumstances; normally, words are transparent, and nothing at all is meant by them except in an actual situation of use of a sentence. Second, is “the unconscious” structured like a language? Again, yes—if we understand by “language” what Lacan asks us to. “The unconscious” arguably is structured like a language—as Lacan (inadequately) understands language. Grant Gillett’s paper is a bold piece of work on a potentially important topic. He explicitly claims to be able to show serious and neglected “conceptual links” between the work of Jacques Lacan and the work of “analytic philosophers” (Frege and Wittgenstein are the two by far the most prominently featured in his paper). This would be quite an achievement, one worthy of being noted and acted upon by some theoretically minded psychoanalysts; and by philosophers in the “Continental” and “Analytic” traditions of philosophy alike. By the former, because it should then lead them to note the relevance to their work of English-speaking philosophical authors who often they do not acknowledge; by the latter, because it should lead them to read Lacan and hopefully to take his potential contribution seriously. Gillett is someone who over the past 20 or so years has made a major contribution to philosophy of mind and to the philosophy of psychoanalysis. This serious contribution, of an avowedly Wittgensteinian nature, makes his paper especially worth grappling with. My main argument shall be that the substance of the invocations of Frege and Wittgenstein in Gillett’s paper are, regrettably, at times Dr Rupert Read, Lecturer in Philosophy, SOC, University of East Anglia, Norwich NR4 7TJ, UK, email: R.Read@uea.ac.uk ISSN 0951-5089/print/ISSN 1465-394X/online/01/040499–16 Ó DOI: 10.1080/09515080120088139 2001 Taylor & Francis Ltd 500 R . READ seriously philosophically awry, and that these mistakes tellingly point to unclarities in the Lacanian picture of language, unclarities which are left unresolved by Gillett at the ultimate cost of his paper’s central thrust. Frege The key problem with Gillett’s use of Frege is that no attention whatsoever is paid in the body of the paper to Frege’s “Context Principle” [1]. One has the sense from reading Gillett’s paper that its author thinks that Frege (and even the early Wittgenstein) was a Russellian logical atomist, and that he (Frege) was primarily concerned with producing a compositionalist theory of meaning. Recent Frege scholarship—especially the powerful work of Joan Weiner (1990, 1999, 2001) and Cora Diamond (1991)—has put this interpretation of Frege in serious doubt, to say the least [2]. The key signi cance of this omission for the paper to which I am responding is that it is only because of our author’s ignoring the Context Principle (CP) that he can claim the relatively close ties of Frege with Lacan on questions of ordinary linguistic meaning/performance that he wants to suggest. When one takes the CP properly into account, a gulf opens up between the Frege/Wittgenstein view of meaning on the one hand and the Saussure/Lacan (and to a considerable extent similarly Levi-Strauss and Derrida [3] (and also, actually, pretty similarly Fodor and Millikan, etc.)) views on the other [4]. Because the signi er/signi ed basis for theorizing meaning is broadly compatible with representationalism and with arguments based around compositionality, but is incompatible with arguments and “reminders” which take sentential and contextual wholism more seriously—i.e. with Frege and Wittgenstein, especially. This incompatibility is generally concealed in those in uenced by the “Continental” tradition or by Literary Theory beneath the very—almost “congenitally”—vague word, “signify” and its cognates/derivatives. This word conceals a vast difference in “depth grammar” between: · · · · · · ordinary linguistic meaning (use); “implicatures”; “special” literary (e.g. metaphorical) signi cances (which of course themselves come in many varieties); purely personal psychological associations; portents; etc., etc. Now, on p. 493 Gillett does, for the rst and only time in his essay, make some effort explicitly to distinguish between different meanings of “meaning”, different signi cations of “signify”, different aspects of use; in short, between different kinds of “thing” that language effectuates and creates. But the metaphor he uses to express the distinctions, rather than highlighting the importance of context and the various kinds of differences between everyday language and its others, only muddies the waters: for the metaphor employed by Gillett is “layers” of meaning. Gillett writes that “[E]very signi er carries layers of meaning”, and this makes it sound as though RESPONSE TO GILLETT 501 these are all layers of the same kind of “stuff”. Whereas my suggestion has been that one has to see, minimally, that there is a deep difference between the use of the word “salt” in “Pass the salt”, on the one hand … and the way the word “salt” could work on one if (say) it featured in a poem at some particular point, or if one had traumatic childhood memories associated with salt, on the other. The term “layers of meaning” preserves the systematic ambiguity (concerning the meaning of meaning/signi cation) that I have suggested that one nds in Gillett (just as in Lacan and Derrida and so on). Gillett notes the similarity of the “signi cation” view of language to Lockean pictures of language and mind—these latter have been subject to vast criticism in the last 250 years. Is the signi er/signi ed distinction, etc., any better off? Why is there no (mention of) criticisms of the signi er/signi ed picture in the entirety of Gillett’s paper (not even of the “internal” criticisms of the “signi cation” view offered by, for example, Derrida) [5]? In sum: we get about half of Frege featured in Gillett’s paper as Frege, simpliciter—and the part of Frege which is omitted raises serious questions about whether our author’s project can be pulled off. This response is not however a hatchet-job. I believe that there are interesting things to be said about possible relations of Lacan to Frege/Wittgenstein, and that Gillett comes close to some of them. To explain, let me turn to discussion of Wittgenstein. Wittgenstein One of the attractive ideas of Gillett’s paper is obviously his effort to bring Wittgenstein and Lacan into conversation. This is rare [6]. What is not so rare, however, is the bringing into some kind of alignment of Wittgenstein and psychoanalysis in general. I am thinking for example of the work of Jonathan Lear, Louis Sass, and Richard Moran, writers who, like Gillett, see a potentially positive relation between Wittgenstein and (say) Freud. What is somewhat strikingly absent from Gillett is any attention to the work of three writers who have perhaps made the most signi cant contributions to this relation: Stanley Cavell (1999), David Finkelstein (1976, 1979), and James Guetti [7]. For the contributions of these latter three in each case raise real concerns about the substance of Gillett’s contribution. I will take them in turn. The most crucial relevance of Cavell is not directly to his various interesting re ections on matters Freudian, etc., but to moments where Gillett’s presentation of psychoanalytic thought appears to court extremely un-Wittgensteinian “private language” type ideas, assuredly without intending to [8]. Let me mention one such moment in particular. Gillett writes that the thoughts of other human beings … are incomprehensible in that one can never know exactly what others think of a given situation nor can one be in touch with the whole of language (as it is spoken by all) … Speech (Parole) is that part of impersonal exchange derived from language and its “combinatory power” and is the means by which others convey to me their worlds and what is signi cant about their worlds. These communications are not transparent to me (in 502 R . READ that I do not see clearly through them the encounters of others with the real) but they are the only basis I have for knowing what the world is like. [9] Does our author have in mind here Cavell on other minds? It is very opaque otherwise how talk of “non-transparency” could be helpful here—it appears to presume that human beings are isolated monads or some such—a presumption sometimes seemingly present in Lacan as in most of “his” tradition (and indeed in most of the Anglo-American tradition prior to Wittgenstein). This is not to say that what Gillett is trying to speak of here is not a real and important phenomenon—only a behaviourist, I think, would say that. But what is troubling is the strong impression—given by the nature of Gilett’s prose—that one is always trying to grasp what is “hidden” in others as if it were the same kind of thing as what is plainly open to view. Is it true that “one can never know exactly what others think of a situation”? Perhaps so—if by this remark one means to index the kind of thing Cavell speaks of when he speaks of the deep dif culties in acknowledging other persons, in acknowledging their reality, especially if they are initially culturally, etc., “distant” from us. One might choose to express this Cavellian (similarly Kuhnian, Winchian) sense of “distance” and dif culty in understanding the sensibility of others in an expression like: “My relation to them is not one of knowing how they think”. Otherwise, Gillett’s Lacanian claim is either false or nonsensical; for I can be told exactly what another thinks of some situation. The words, “I understand your situation exactly” can be used, and used appositely and even perhaps ordinarily. Note furthermore Gillett’s troubling use of words such as “incomprehensible”, “convey”, “only” and “knowing” in the passage quoted above. The only way I could see to recuperate these remarks of his, again, would be to attempt to read them in relation to Cavell’s discussions. Otherwise, Gillett’s employment of the concept of “signi cation” risks riding roughshod over important distinctions (e.g. between the way I “feel” the word “America”, and how the word “America” actually functions in sentences in contexts) [10]; while his employment of the concept of “knowledge” risks, on the other hand, absolutizing and reifying an important but not utterly schismatic distinction (e.g. between my knowledge of American geography and my knowledge of American people). Finkelstein’s paper, “On the distinction between conscious and unconscious states of mind”, is in my view the most signi cant effort by a Wittgensteinian philosopher of mind—perhaps by any recent philosopher—to understand what discourse about “the unconscious”—what psychoanalysis—can actually be. Gillett presents Lacan’s version of post-Freudian psychoanalytic theory as a series of theses/claims; Finkelstein’s paper has the virtue of presenting talk of “the unconscious” rather as something which people actually engage in, as a practice, rather than as a set of theoretical gambits. In short, Finkelstein offers a sophisticated understanding of how one can understand “the unconscious” [11] without violating Wittgensteinian precepts, an undertaking seemingly presupposed by but not perspicuously represented in Gillett. RESPONSE TO GILLETT 503 A number of pieces by Guetti are relevant to Gillett’s topic, but especially his Wittgenstein and the Grammar of Literary Language (1993). It is worth quoting extensively from pp. 19–20 of that work: [T]he logical problem of Freud’s argument concerning slips of the tongue and the like—in which “unconscious” processes are unsatisfactorily distinguished from “conscious” ones—goes very deep. For it bespeaks his own actual unrecognition of the unconscious itself. There is nothing free or wild or irrational about the unconscious—attributes that Freud will claim for it elsewhere—in its function in Freud’s “literary” cases. It traf cs in the same sort of meanings and purposes as do the conscious operations it sometimes interrupts and replaces. It is not revelatory of associative possibilities, but rather of “meaningful” chains more determined and foregone—and, of course, more boring—than anything the conscious mind has to offer. It should be noted … that I am not proposing that the “unconscious” does not exist, but only that the modelling of it upon processes and motives that are if anything hypermeaningful is nonsense. More important, I am certainly not objecting to the notion that an “unconscious” might be composed of some sort of language. Rather, I would suggest … that “accidents” in speech are the results of our assimilation of and perpetual access to the grammatical possibilities of our language. This is a recognition that some of Freud’s revisionists—most notably Jacques Lacan—sometimes seem to approach. Lacan considers, for instance, that phrases like Freud’s “unconscious thought” are not contradictory precisely because “the unconscious participates in the functions of ideation, even of thought”, and it does so just because it is itself composed of language—as if the unconscious might be gured as the entire learned totality of the possibilities of language—which I have called the “grammar”—unrestricted by any immediate and particular applications or intentions. Here one may be so grati ed to discover that the importance of such grammatical possibility is acknowledged by a psychologist that Lacan’s formulation may become too eagerly accepted. But we should proceed with caution here, for this agreeable assumption of “the omnipresence of human discourse”—in which even the unconscious mind should be considered as somehow “structured like a language”—often reveals itself as a much more restrictive conception where language is not present as a system of possibilities but working, even unconsciously, as a secret but quite meaningful code. A “subject” is introduced by the analyst into a “primary Language in which, beyond what he tells us of himself, he is already talking to us unbeknownst to him.” This conception of “unconscious language” as a source of meanings and knowledge to which only psychoanalysts are privy may be indistinguishable from Freud’s own. The powerful in uences and effects of the body of 504 R . READ language, of the grammar, within our speaking behaviour that I am describing [here, in this book] are neither so occult as this nor, when discovered, so revelatory, exactly because they are merely a matter of language’s complex and variegated texture, of its surface of possibility as we learn it. And if they seem magical—or “signi cant” beyond anything we might say in a meaningful way—this is not because they are varieties of some sort of hypermeaning, but because they are not meanings at all; rather, they are signs of our very capability of, and of our language’s potential for, meaningful expression, exhibitions of verbal power and freedom that in fact can never be realized since they are meaningful only in restrictive applications, when we know what we must say … [12] We can see that Guetti’s account of meaning, etc., differs notably from Gillett’s (whose account Guetti would, I think, regard as an account of a fantasized “hypermeaning”). Encoded in this long quote from Guetti is a case in some ways really rather similar to Gillett’s—plus objections to some of what is central to Gillett’s case. Guetti provides us the resources not only to worry about but to get beyond Gillett’s phrase “layers of meaning”, for example—for he points out that “grammatical possibilities”, the kinds of in ections and associations found in much modernist poetry and in some parapraxes, are not best seen as meanings (uses) at all. Guetti distinguishes, notably (though not “absolutely” nor rei catorily), between meanings (uses) and “grammatical effects” [13]. “Grammatical effects” and associations are more or less systematic effects of language which words and their interconnections have on us, effects of great “signi cance” to us, but not well understood as “signifying” strictly in the sense of being used to name something, or to describe something [14], still less to accomplish some action. As implied above in my discussion of Frege, such a distinction (between meaning considered as use on the one hand and effects of language/grammar on the other) is needed and lacking at certain points in Gillett’s paper; for instance, where “[discursive] signi cations” and “semantic content” are run together in quasi-post-modernist fashion, in a fashion which Saussure unwittingly laid the groundwork for long ago, in a fashion which Guetti can facilitate one’s resisting. Guetti thinks then that the idea that “ ‘the unconscious’ is structured like a language” can indeed be fruitful for one thinking, in broadly Wittgensteinian vein, about philosophy, language, and varieties of “meaning”—but in a way more concrete and less exegetically and philosophically problematic than our author’s. One can think about “the unconscious” as generating or even being the kinds of effects that “grammar” has (more or less systematically) upon us—without misleadingly assimilating these effects to ordinary applications of language (or indeed to some other things that can be meant by the word “signify”). A clear place in Wittgenstein wherein one can see clearly the signi cance and power of the kinds of points that Guetti makes is as early as section 6 of Philosophical Investigations (PI). Wittgenstein remarks that the “… ostensive teaching of words can be said to establish an association between the word and the thing”. So far, no-one, certainly not Locke nor Gillett nor myself, is very likely to disagree. Wittgenstein RESPONSE TO GILLETT 505 however presses for clari cation: “But what does this mean? Well, it can mean various things; but one very likely thinks rst of all that a picture of the object comes before the child’s mind when it hears the word”. That is classic psychologism. Does Wittgenstein repudiate it? Not straightforwardly; he does something subtler: “But now, if this does happen, can it be the purpose of the word?—Yes, it can be the purpose.—I can imagine such a use of words (of series of sounds)”. The parenthetic clari cation is pretty crucial. One might express it this way: using a word as a sound, or so as to create a certain picture in the mind, is not really using the word as a word at all. To use a word as a word is to use it to say something, to carry out some linguistic act, to make a move in a language-game. When what one has is what Wittgenstein immediately goes on to term, rather, an utterance of “a word [so as to strike] a note on the keyboard of the imagination”, one doesn’t yet have a move in a language-game [15]. Now, as Wittgenstein says, it can be the purpose of words to work that way. When? Well, for example, in much literature. Also, quite possibly, in attempting to train or control someone (e.g. a child), one might well work to evoke certain images, so as to encourage or discourage a behaviour associated with them. But according to the dominant pictures of philosophy of mind and language in recent Anglo-American philosophy (as for so long before), it is in the rst instance precisely the aim of uttering words, all the time, to strike notes on a mental keyboard. This is what Cognitivism and Mentalism say [16]. And Lacan? It appears to be what he says too: for the subject is to be “de ned as the effect of the signi er”. And this effect, presumably, is primarily one of construction through “evocation” (of images, of atmospheres), which for Lacan is what language is essentially about (see my “Conclusions”, below). We see here how Lacan is still rather beholden to psychologism—he has not, and this Gillett crucially fails to see, made the turning, the Wittgensteinian turning, away from static evocation and toward actual use. In “the unconscious”, grammatical effects of words—not their uses—may indeed be king. But the running together of the workings of “the unconscious” (and of aspects of language which play upon it) with the dynamics of actual everyday use of most language is a recipe for unclarity. In sum, our author’s use of Wittgenstein fails to connect with—fails to be compatible with—the best “pro-psychoanalysis” work already accomplished by thinkers who, like our author, would like to see a more positive relation between Freud/Lacan and Wittgenstein than has reigned until recently; and our author’s use of Wittgenstein is in any case rather vague and inaccurate. But, it might be asked, what of our author’s employment of the early Wittgenstein? Surely the early Wittgenstein at least is susceptible of the kind of “theoreticist” reading practised by Gillett [17]? The reader may already have guessed the general character of my answer to these questions, given my favourable reference to revisionist Frege scholarship; for Diamond and Conant (especially) have been at the forefront of radically revising the standard interpretation of early Wittgenstein, too [18]. The logical atomist reading of Tractatus Logico-Philosophicus (TLP), and the linked idea that TLP envisages any gap whatsoever between language and reality, have been thoroughgoingly questioned [19]. Compare TLP 2.1511ff.—pictures 506 R . READ reach right out to reality. This is far closer to recent Putnam/McDowell/Diamond than it is to (say) Kripke, Derrida or Fodor [20]. For the early Wittgenstein, it is crucial to understand that, if we are to speak of “showing”, we must not enter into the illusion that what can be shown can also be said (say “indirectly”, or by means of “gestures”, or by means of extra “layers of meaning”); it is crucial to understand that the “limits of language” are only “Pickwickian” (i.e. there is no such thing as an outside to the “limits” of language) [21]; and it is crucial to understand that one is in any case not being presented by Wittgenstein with a theory of language but rather with a set of sentences which have to be overcome, “thrown away”. Regrettably, Gillett’s essay shows little sign of understanding these things. Weiner, Diamond, Conant et al. provide one with the resources to see deep continuities between (as well as progress through) Frege, TLP and PI. These continuities provide resources to undermine thoroughly the pictures of mind, language and meaning present in philosophers and theorists as “diverse” as generative linguists, semantic theorists, cognitive scientists, structuralists and post-structuralists—and to undermine the very pictures of Frege and Wittgenstein’s own thought found in standard interpretations of that thought. How exactly does all this relate to discussions of “the unconscious”? Well, “the unconscious” may indeed be like the way that a certain word unavoidably strikes us and reminds us of certain other connected words/structures (cf. Guetti); but Wittgenstein does not identify ordinary language use as similarly “non-negotiable” [22]. On the contrary, ordinary language use is almost in nitely negotiable [23]. For example, I can even say “bububu” and mean “If it rains, I will go for a walk”, if only a language-game is (co-)constructed to this end, only if a sentence which can be made sense of occurs in the course of an activity [24]. We look for a sense in what others say: only as a last resort do we nd their talk nonsensical, and even then not because it “breaks grammatical rules”. Wittgenstein simply had no role in his philosophy for “logical and grammatical truths”. (As for the early Wittgenstein—this was in large part his view too: check out for instance the astonishing and too-often-ignored rst sentence of TLP 5.5563.) Wittgensteinian philosophy leaves the use of ordinary language as it is: the practitioners of ordinary language can do all the negotiation and clarifying of it that is required. This de ationary approach is perhaps what is especially uncomfortable for Chomsky and Dummett and Derrida (and Gillett?) alike. Thus, if we are thinking of everyday language, then Lacanian images of language as a total system through which we are structured and in which we are “imprisoned” surely do signi cantly more harm than good. Conclusions Gillett tells us that Lacan can tell us “important things about psychosemantics” [25]. I have suggested, rather, that Lacan is likely only to confuse us if we try to read and use him, as I fully admit it is reasonably natural to do, as a “psychosemanticist”, i.e. as someone trying, absurdly, to discover things about meaning by looking at the mind. I have also suggested that Lacan can hint to us important things about what RESPONSE TO GILLETT 507 we might call “psycholinguistic effects”—about certain phenomena, not happily identi ed with meaning, which nevertheless do have a lot to do with language in its non-transparent, “non-used” aspects. The difference between where Lacan is helpful and where he is not becomes stark at the close of Gillett’s paper. That is, on p. 495 the difference between Lacan and (say) Wittgenstein on fundamental issues to do with ordinary meaning/use is plain: “the function of language”, according to Lacan, “is not to inform but to evoke”. This ts perfectly with the idea that “signifying”—producing “signi eds”— is what language is all about, but contradicts completely a conception of (roughly) meaning as use. For sure, Wittgenstein thinks “the” function of language is not primarily to inform—but the function only of literary etc. language is (primarily) to evoke. That Gillett does not recognize this is disturbing. Thus one is driven strongly to suspect that Lacan (perhaps after Freud, and certainly after Saussure and along with Derrida) is systematically unclear about what the word “signify” means (“signi es”…), and that Gillett does very little to rectify this situation. I am not arguing that Lacan should be ignored; but I am arguing that, if one expects intellectually minded psychoanalysts and open-minded philosophers in the English-speaking world to take Lacan seriously, and work on bringing his ideas into creative dialogue with Wittgenstein et al. (who must rst be adequately interpreted), then one must provide an account of Lacanian insights which does not founder on a basic unclarity about language—in particular, on fairly basic distinctions between different uses of the word “mean”. The central thrust of my critical remarks might be summed up thus: evoking is one thing, informing another, doing (which, roughly, subsumes informing) still another … Our author assimilates Lacan to Wittgenstein, etc., because Lacan opposes an informing-centred picture of language; but this is to ignore Wittgenstein’s doing-centred picture (especially as we nd it in great detail in his later work), and moreover to miss the interesting and genuine connection (between Lacan on the one hand and Frege/Austin/Wittgenstein on the other) around the “subsidiary” issue of evocation by means of language. To put this in a cute and capsule-like form: If “the unconscious” is structured like a language, then that language is at the least rather more like Keats’s or Faulkner’s or Wallace Stevens’s or L.A.N.G.U.A.G.E . poetry or Artaud’s writing than it is like Clift’s Notes or a lawyer’s description of a traf c accident or a car-repair manual or the dictionary or “The cat sat on the mat”. In conclusion: Gillett deserves praise for having taken on the tricky task of attempting to take Lacan seriously in an “analytic” idiom. But there are serious dif culties with Lacan’s conception of language-in-general , dif culties which emerge into clear view only I think if one uses the likes of Frege and Wittgenstein seriously, rather than merely translating them more or less reductively into “analytic” or quasi-scienti c “verities”. Gillett wants to call on Frege and Wittgenstein as support for his “analytic” project—but Frege and (especially) Wittgenstein problematize precisely the kind of “picture” of language (of words as akin to names, of sentences as getting their meaning through combining the meanings (references) of the words of which they are composed [26]—in Wittgenstein scholarship, this is known usually 508 R . READ as “the Augustinian picture of language”) that in practice is shared by Fodor, Millikan, Devitt, Dummett and Lacan alike. Lacan is simply a kind of enlarged/ distorted mirror image of the Anglo-American psychosemanticists … Where their bottom-line is a single-minded focus on information [27], his focus is singlemindedly on information’s “other”, on evocation; what they all have in common is an unhealthy preoccupation with bells being rung in the mind, with “names” which—through the “medium” of the mind—are held to produce images, or words, or feelings, or otherwise to yield a “picture” of the world and of oneself in (or apart from) it. The concomitant failure of all these thinkers is a failure to take seriously the actual use of words (sentences) to accomplish real actions: to be, with others, in the world. To return then to the two questions around which Gillett structures his paper. First, is “the unconscious” a network of signi ers? We can now answer: roughly, yes, insofar as one can understand and make something of the question; but don’t be under the illusion that what actual language most of the time is like is: like this. Words primarily “signify” concepts or things only in exceptional circumstances; normally, words are transparent, and nothing at all is meant by them except in an actual situation of use of a sentence. Second, is “the unconscious” structured like a language? Again, yes—if we understand by “language” what Lacan asks us to. Lacan can help us to understand “the unconscious”—because it does, arguably, operate in something like the manner of the “Augustinian picture”, i.e. it does, arguably, operate through the weighty resonance of names, and through the hiding and exposing of suspended grammatical possibilities; through, in short, the kinds of processes long known to Freud’s great predecessors, namely, Shakespeare and those poets who, like him, revel in the form and materiality of language, and in the ways words work on our minds. Especially then if one emphasizes a psychologistic reading of the “Augustinian picture”, one can say that “the unconscious” is structured like a language. It is structured like a language—as Lacan (inadequately) understands language. Unlike ordinary, everyday language, wherein use and context—not names and their combinations and the effects of both—are paramount. Lacan’s misreadings of the use of everyday language may not in the end be of much moment, insofar as he is read above all as a psychoanalyst, who is naturally therefore predominantly interested precisely in the pathological. But then it will be dangerous to use and trumpet Lacan much beyond the important but relatively narrow domain wherein he deserves his fame. It is Wittgenstein, above all, who offers us a proper orientation towards (everyday) language, and who enables us thereby to understand the different aspects of language which words such as “signify” tend to tun together. I look forward to how Gillett’s writing will look different, if and when he recognizes this [28]. Acknowledgements My thanks for helpful comments on this paper to Nadine Cipa and Luke Mulhall. RESPONSE TO GILLETT 509 Notes [1] The importance of the principle is acknowledged in notes 5 and 8 of Gillett’s paper—but this acknowledgement plays no role in the paper’s actual argument. It is striking, in fact, that the kind of holism that Gillett does discuss—more or less the kinds of “ eld” holism that one nds most prominently in Quine, Davidson, Saussure and Derrida—is presented without any textual acknowledgement of its deep difference from the (ignored) sentential and contextual holism of Frege and Wittgenstein. To say, as Davidson and Derrida, etc., do, that words take on their meanings only in relation to and in their difference from all other words, is still to risk leaving room for a picture of meaning in which words have meanings “attached” to them, in which meanings are “signi ed” by (individual) words (in their differences from other words). In contrast, Frege and Wittgenstein emphasize that there is just no point looking for word meanings outside of sentences used in contexts, if what one is hoping to do is to understand and not be bemused by the workings of language. (For detail, see Read, 2000c.) Gillett falls into thinking of meanings still as some kind of thing, rather than placing use centrally, even in the very parts of his text where he tries to paint himself as a Wittgensteinian. Take his remark that “meanings [are tied] to the uses we make of the words we use (PI no. 43)”. But if we actually look at PI 43, we nd that Wittgenstein actually says that “For a large class of cases—though not for all—in which we employ the word ‘meaning’ it can be de ned thus: the meaning of a word is its use in the language”. The noting of exceptions, the “though not for all”, not only ags Wittgenstein’s antipathy to the production of semantic theory but also indexes exactly the cases for which Guetti’s post-Lacanian account of non-transparent language works—see my account of Guetti, Gillett and Lacan, below. But more importantly still, note that Wittgenstein does not in PI 43 say that meanings are things tied to uses; he says that where philosophers have used the word “meaning”, they would do best to think of it as or substitute for it the phrase “use in the language[-game]”. There is no role in Gillett’s paper for what is paramount in Frege and Wittgenstein (especially later Wittgenstein) and Austin: words being used to do things. (For a rather more detailed account of the absolute importance of “contextualism”, see Read, 2000c; Gustafson, 1998.) [2] It hardly seems to me that note 9 of Gillett’s paper is suf cient acknowledgement of the highly controversial status of the interpretation of Frege that he presumes throughout his paper. There has been a wealth of revisionist scholarship on Frege in the last 20 years, thoroughgoingly questioning the kind of picture of Frege that Gillett offers up. Some of the best of the revisionist work, besides Weiner’s seminal books, is that of Cora Diamond (1991). (See, also the work of Tom Ricketts, e.g. 1985, and of James Conant, e.g. 2000) and Ray Monk’s (1996) useful account of the deep differences between the atomism of Russell and Moore on the one hand, and the “contextualism” of Frege and Wittgenstein on the other. [3] The case of Derrida is, of course, complex, given that he aims to critique “the metaphysics of presence”, and thus ostensibly opposes psychologism and hopes (sometimes) to deconstruct metaphysics in general. However, he leaves in place the kind of picture—or categorization—of language (“signifer” and “signi ed”) that causes so many of the problems in Saussure and Husserl and their successors. (For a central instance of where “signi cation” talk gets Derrida into trouble, see p. 12, 1988.) Derrida’s innovation is to bracket or erase the “signi ed”. But that is only a reactive response within the categories allowed for by metaphysics (i.e. it is structurally parallel to certain forms of scepticism), and is not yet to move to a genuinely post-metaphysical and “contextualist” approach of the kind one nds in (e.g.) Wittgenstein. This is one reason why Derrida remains fatally in thrall to metaphysics, and does not see the possibility of a mode of philosophizing that would actually take the ordinary seriously. (For argument, see Martin Stone, 2000; Read, 2000b, especially the Appendix on Derrida.) A similar aw, I shall suggest below, af icts Lacan. Lacan, like Derrida, makes it sound as though our not “having” (access to) the “signi ed” is a loss. Whereas Wittgenstein makes plain that we lack nothing, insofar as our ordinary grasp (“mastery”) of the language, of meaning as use, is concerned. [4] A similar gulf opens up between the “generativist” assumptions made by Gillett when he writes 510 R . READ [5] [6] [7] [8] [9] [10] [11] that “a nite number of signs or concepts can be used to generate an in nite number of sentences of thoughts” on the one hand and Frege/Wittgenstein on the other. A number of writers in uenced by Wittgenstein (e.g. Martin Gustafson, Ian Niles, H.A. Nielsen) have fundamentally questioned Chomskian and Davidsonian “paradigms” hereabouts. Such questioning is already implicit in Frege’s CP; and yet Gillett claims, in note 5, that there is a “generative strand of semantics” in Frege and Wittgenstein. He gives no attribution or citation for this highly controversial claim. I see no reason to think that there is a generative semantics—if by that we mean any claim about the “generativity” of language, any claim remotely resembling what one nds in (say) Chomsky, Fodor and Davidson—in either thinker. See note 3, above, on Derrida. Gillett speaks of some things being “captured” by signi cation, and other things eluding it: But what is it to be “captured” by signi cation? Since we don’t really know what exactly “signi cation” is, since “signify” seems to signify nothing (in particular, or else anything and everything), since the term is systematically ambiguous, we have very little idea what we are being told cannot be done, when we are told by Gillett, after Lacan, that “many parallel messages which have made their mark on us … are not [and cannot be] wholly captured in an iconic or formal language”. This suggestion that “formal” (???) language fails to “capture” all that we would capture with it tends to produce a sense of language as limited, or as a “trap” or a “prison”. This deeply problematic un-Wittgensteinian idea is grist to the mill of deconstruction and of post-modernism—and is, interestingly, shared by Fodor and Chomsky (under the guise of “cognitive closure”). It is also a common way of (mis-)interpreting Wittgenstein’s early work on “the limits of language”, a way seemingly favoured by Gillett (see my brief discussions of TLP, below). Though not altogether unprecedented—see the quotation from Guetti, below. Gillett merely refers the interested reader to his own 1999; but, on my understanding, that book in its aims and execution is simply consistent with the paper to which I am responding: Gillett wants both to present a Wittgensteinian approach to mind and language and to put forward a set of theoretical claims about “the unconscious”. These two aims cannot be combined—and, in part by problematizing the Rorty/Davidson theory of “the unconscious”, Finkelstein (1999) makes this perspicuous. On a linked point, I am not altogether reassured by Gillett’s invocation of a Wittgenstein who apparently “… argues that rule-following is the basis of meaning and that the conditions to which I adapt my rule following should be public”. This sounds rather too much like a substantive thesis is being attributed to Wittgenstein: namely, that “there can be no language which is not the actual following of public rules”. But a less theory-like and more “therapeutic” version of Wittgenstein on rules has recently become perspicuously available: see, for instance, Phillips (1993, 2000), Pleasants (1999), Read (2000a), and the preface to Winch (1990). Nested quote from Lacan (1977). The same kind of worry arises again, for example, when Gillett says that “In any linguistic exchange, a subject is attentive to the words used and what they seem to mean but may also be uncertain about the exact meanings or nuances of meaning intended by the speaker”. This claim is surely too general (there are plenty of such exchanges in which there is no real room for a doubt as to what is meant); but it might be right as a claim about the meanings of words if by “meaning” we mean roughly grammatical or psychological effects. Unfortunately, here as elsewhere, Gillett does not distinguish for us between different meanings of the word “meaning”, and so, as discussed above in relation to Cavellian considerations, he gives the impression of being vulnerable to anti-private-language objections, where clearly that is not his desire. (See also my presentation of Guetti, below, for more on “experiencing meanings”, etc.) I shall normally scare-quote this phrase, on the grounds (not remarked by Gillett) that powerful philosophical voices have often been raised to the effect that to speak of “the unconscious” using a noun-phrase at all is to risk an unhealthy rei cation. I cannot explore this further here: but some strong more or less Rylean reasons for thinking this are to be found scattered through the work of the Wittgensteinian ethnomethodologist, Jeff Coulter. For parallel worries about the very idea of “the conscious” as a quasi-thing, worries which a fortiori there is no space for here, see Read (forthcoming a). RESPONSE TO GILLETT 511 [12] Nested quotes from Lacan (1981). Guetti makes clear here the issues at stake; whereas at points in Gillett’s presentation, Lacan’s notoriously dif cult writing remains elusive and the point of his “theses” opaque. It is perhaps surprising that Gillett did not, to counter this dif culty, draw on pre-existing and relatively accessible discussions by others of Lacan and his picture of language: for example, those of his Feminist inheritors and critics, such as Mitchell, Rose, Gallop (1982, 1985), and some of the French Feminists; and more recently the intriguing and helpful writings of Slavoj Zizek. [13] See Guetti (1993); Read and Guetti (1999). [14] My point here is that, whereas ordinary naming is preparatory to making a move in a languagegame, the function of “names” and “signi cations” in poems or in the psychopathology of everyday life is not in fact normally that. But as I explain later, a “name-theory” of language is in fact arguably the lowest common denominator between post-modernism, Lacan, and AngloAmerican psychosemantics (cf. also note 15, below). [15] Gillett alludes vaguely to the Builders’ “language-game”. If we look however at what Wittgenstein says of the Builders in PI 6, we nd the following clear remarks: “Uttering a word [simply as part of a series of sounds] is like striking a note on the keyboard of the imagination. But in the language of PI 2 [the Builders’ language-game] it is not the purpose of the words to evoke images. (It may, of course, be discovered that that helps to attain the actual purpose.)” How does Gillett square these criticisms of the “evocatory” view with Lacan’s general claim that “the function of language … is to evoke” (discussed further below); or with his own remark that “the point of … structural distinctions is to mark the world in ways apt for structuring our interaction with it”? For the latter makes it sound as if the world is above all to be named (“marked”)—but just that is the very fantasy which motivates Mentalism. [16] Gillett writes, “The need for philosophical semantics to account for both sense and reference poses the need for it to engage with psychosemantics”. Here, Gillett betrays clearly a psychologism which runs completely counter to everything which Frege and Wittgenstein hoped in their work to establish. Don’t look for a psychosemantics, look at use. And don’t turn “use” into the centre-piece of a theory, either. Later Wittgenstein no more held a use-theory of language than early Wittgenstein held a picture theory. For powerful argumentation in support of this exegetical point, see Crary’s essay in Crary and Read (2000). [17] What do I mean by a “theoreticist” reading? Well, take for example this declaration of intent, on p. 5 of Gillett (1992): “The present view aims to use an account derived from Wittgenstein’s later philosophy to link a number of disparate topics in philosophy of mind, epistemology, philosophy of language, and philosophical psychology so as to present a workable theory of mental content” (italics mine). This seems to me to be in deep tension with (for example) PI paras 108–134. I am trying to urge in this “Response to Gillett” that Wittgenstein’s philosophy is of necessity missed if one misses his methodology. [18] See also Crary and Read (2000), Read (2000d), Read (forthcoming b), and Michael Kremer’s recent work. [19] For a useful source, see Crary and Read (2000); see also the latest (special) issue of the journal Philosophical Investigations, especially Tony Palmer’s piece. [20] Also, the Pears/McGuiness translation of TLP 7 is in any case now widely regarded as highly suspect—just try nding the sense of “pass over” (suggesting that there is some thing to pass over, intimating an ineffable mystery) in the actual German. So when Gillett writes that there is a “… sense of [a] gap between language (or thought) and reality which many have discerned in [TLP 7]”; then we can reply: why yes, indeed; and they are wrong. [21] Thus it is a complete mistake to speak of TLP 5.6 (and TLP 7), as Gillett does (1992, p. 184), as “nihilistic”. [22] Nor I think does pragmatism, to which our author expresses allegiance. A small, related point: Wittgenstein was worried that pragmatism did not altogether escape philosophical theorizing/“thesisizing” hereabouts—see for example his Lectures on the Foundations of Mathematics. Gillett tends to assimilate pragmatism and Wittgenstein—a tempting and often useful but potentially misleading endeavour (for a partly contrasting vision of Wittgenstein’s 512 R . READ [23] [24] [25] [26] [27] [28] relation to pragmatism, see e.g. Read and Guetti, 1999). Our author wants to take “use” and perhaps “pragmatics” seriously—but it is not clear that this is well-described simply as “pragmatism” (or as Wittgensteinian). Taking use seriously involves questioning the technical distinctions between “syntax” and “semantics” and “pragmatics”, while of course remaining open to important differences between different aspects of (uses of) language. When in ected explicitly by Lacan, Gillett’s picture sometimes becomes still less clear: e.g. is “creative subjectivity” really a kind of synonym for “use”? (See my note 24, below.) Similarly, Gillett writes that provision of a referential semantics is “a pivotal point linking the Lacanian view with mainstream analytic philosophy of language”. This may be technically true; but that would only be because Lacan here is arguably well behind Wittgenstein and Guetti, who, unlike the “mainstream”, take seriously the central place of use, and context, and “pragmatics” (one might say: there is no semantics without pragmtics), in their “account” of language. For explication, and for implicit guidance on how this point is not to be confused with post-modernist talk of “free play”, etc., consult especially Conant (2000). Now, it may seem as though Gillett recognizes this kind of possibility in moments such as the following: “[One] has an important (and creative) role in the discursive project of speaking life into language because it is in use (as Wittgenstein notes at PI no.432), including my use, that words take on life or meaning”. So far so good. But Gillett goes on by quoting Lacan to “explain” what this means: “[C]reative subjectivity has not ceased in its struggle to renew the never-exhausted power of symbols that brings them to the light of day”. The trouble with this is that it now sounds rather too free; it sounds like there is no language-game discipline there. For Lacan appeals to “creative subjectivity”, which sounds like an artistic process; where Wittgenstein, if what we are talking about is (ordinary) use, would surely appeal instead to something like the perfectly ordinary engendering of linguistic etc. novelty. I am, as should be clear, very unclear on how Lacan is supposed to tie in with Frege (see e.g. Gillett’s peculiar claim that “Lacan’s view introduces subtleties … partly arising from the interactions in which speech is used to shape the signifying structure of the subject. The result [of this?] is to make sense or cognitive signi cance centrally important in meaning and understanding (as Frege saw)”). But I am not entirely clear either on how Gillett hopes to make Lacan relevant to Fodor, and this may be partly due to an odd feature of his interpretation of Fodor. He says that Fodor has “recently conceded” that sense is centrally important in meaning and understanding. The implication appears to be that Fodor has only recently grudgingly moved away from a radical externalism toward the “concession” that sense is important too. But, as I read Fodor, this gets the story of his development backwards. Fodor has gradually and reluctantly (half-heartedly?) moved away from a pure and roughly sense-based “signi cation” view to one allowing some externalist elements, not the other way around. Fodor was at heart a good old-fashioned “Lockean”, until forced over the last 10–15 years to admit that that left him in some very uncomfortable philosophical positions. His psychosemantics seems then still more vulnerable than Lacan’s to pressure from “externalist” and Wittgensteinian philosophical arguments alike; though, regrettably, it is only the former that Fodor has bothered actively responding to, in the last two decades. Compare the following passage of Gillett’s paper: “signi ers are the currency of speech and therefore they are the raw material from which an individual assembles thoughts about the world … The distinctions picked out by signi ers are the joints at which my psyche carves the world”. Gillett’s pragmatist–Wittgensteinian strand sits most ill with the name-ist and quasiatomist emphasis of moments such as these. For a detailed treatment of this concept, see Read (forthcoming a). It is of course possible that I have been too hard on Gillett. And I am writing very much, I hope, as a constructive critic. Clearly, he wants to avoid some of the conclusions which I have suggested he is unfortunately committed to. I applaud his desire, for example, to nd a middle way between analytic philosophy and post-modernism; I applaud his self-avowed “pragmatism” (though see note 22, above); I applaud his anti-Correspondencism; I applaud his questionings of idea-ism/idealism. However, my paper has, among other things, raised serious concerns about whether as RESPONSE TO GILLETT 513 things stand he is actually entitled to think of himself as a “pragmatist–Wittgensteinian” at all. I would be pleased to be wrong. References CAVELL, S. (1976). Must We Mean What We Say? Oxford: Oxford University Press. CAVELL, S. (1979). The Claim of Reason. Oxford: Oxford University Press. CONANT, J. (2000). Frege and early Wittgenstein on elucidation and nonsense. In A. CRARY & R. READ (Eds) The New Wittgenstein. London: Routledge. CRARY , A. & READ, R. (2000). The New Wittgenstein. London: Routledge. DERRIDA , J. (1988). Limited Inc. Evanston: North Western. DIAMOND, C. (1991). The Realistic Spirit. Cambridge: MIT Press. FINKELSTEIN , D. (1999). On the distinction between conscious and unconscious states of mind. American Philosophical Quarterly, 36, 79–100. GALLOP , J. (1982). The Daughter’s Seduction: Feminism and Psychoanalysis. Ithaca: Cornell. GALLOP , J. (1985). Reading Lacan. Ithaca: Cornell. GILLETT , G. (1992). Representation, Meaning and Thought. Oxford: Clarendon. GILLETT , G. (1999). The Mind and its Discontents. Oxford: Oxford University Press. GILLETT , G. (2001). Signi cation and the unconscious. Philosophical Psychology, 14, 477–498. GUETTI J. (1985). Freudian slippage. Raritan, 8, 1. GUETTI , J. (1993). Wittgenstein and the Grammar of Literary Experience. Athens, GA: University of Georgia Press. GUSTAFSON, M. (1998). Systematic meaning and linguistic diversity: the place of meaning—theories in Davidson’s later philosophy. Inquiry, 41, 435–453. LACAN, J. (1981). Speech and Language in Psychoanalysis. Baltimore: Johns Hopkins. MONK, R. (1996). Bertrand Russell 1872–1921: The Spirit of Solitude. London: Jonathan Cape. PHILLIPS , D.Z. (1993). Ten questions for psycho-analysis. Philosophy, 68, 183–192. PHILLIPS , D.Z. (2000). Beyond rules. History of the Human Sciences, 13, 17–36. PHILLIPS , D.Z. & WINCH, P. (1989). Wittgenstein: Attention to Particulars. New York: St. Martin’s. PLEASANTS , N. (1999). Wittgenstein and the Idea of a Critical Social Theory. London: Routledge. READ, R. (2000a). Getting rule-following right: the anticipation in philosophical investigations of para.s 201–2 by para.s 197–9. UEA Papers in Philosophy, New Series No. 11. READ, R. (2000b). Wittgenstein and Marx on ordinary and philosophical language. Essays in Philosophy, 1, 2; available online at: , http://www.humboldt.edu/ , essays/ . . READ, R. (2000c). What “There can be no such thing as meaning anything by any word” could possibly mean. In A. CRARY & R. READ (Eds) The New Wittgenstein. London: Routledge. READ, R. (2000d). The New Hume’s “new” antagonists: on the relevance of Wittgenstein and Goodman to the New Hume Debate. In READ & RICHMAN (Eds) The New Hume Debate. London: Routledge. READ, R. (forthcoming a). The “hard” problem of consciousness is continually reproduced and made harder by all attempts to solve it. Theory, Culture and Society. READ, R. (forthcoming b). On approaching schizophrenia through Wittgenstein. Philosophical Psychology, 14, 449–475. READ, R. & GUETTI , J. (1999). Meaningful consequences. Philosophical Forum, XXX, 289–317. RICKETTS, T. (1985). Frege, the Tractatus, and the logocentric predicament. Nous, 19, 3–15. STONE, M. (2000). Wittgenstein on deconstruction. In A. CRARY & R. READ (Eds) The New Wittgenstein. London: Routledge. WEINER , J. (1990). Frege in Perspective. Ithaca: Cornell. WEINER , J. (1999). Frege. Oxford: Oxford University Press. WEINER , J. (2001). Theory and elucidation: the end of the age of innocence. In FLOYD & SHIEH (Eds) Future Pasts: Perspectives on the Place of the Analytic Tradition in Twentieth Century Philosophy. Cambridge: Harvard. 514 R . READ WINCH, P. (1990). The Idea of a Social Science and its Relation to Philosophy. London: Routledge. WITTGENSTEIN , L. (1922). Tractatus Logico-Philosophicus. New York: Routledge. WITTGENSTEIN , L. (1958). Philosophical Investigations. London: Macmillan. WITTGENSTEIN , L. (1970). Lectures and Conversations on Aesthetics, Psychology and Religious Belief. Berkeley: University of California Press. WITTGENSTEIN , L. (1976). Wittgenstein’s Lectures on the Foundations of Mathematics. Hassocks, Sussex: Harvester.