Philosophy as a Kind of Writing

Philosophy as a Kind of Writing

Volume 39, Number 1, Winter 2008

E-ISSN: 1080-661X Print ISSN: 0028-6087

DOI: 10.1353/nlh.0.0014

Philosophy as a Kind of Writing: An Essay on Derrida

Richard Rorty Princeton University

I

Here is one way to look at physics: there are some invisible things which are parts of everything else and whose behavior determines the way everything else works. Physics is the search for an accurate description of those invisible things, and it proceeds by finding better and better explanations of the visible. Eventually, by way of microbiological accounts of the mental, and through causal accounts of the mechanisms of language, we shall be able to see the physicists’ accumulation of truths about the world as itself a transaction between these invisible things.

Here is another way of looking at physics: the physicists are men looking for new interpretations of the Book of Nature. After each pedestrian period of normal science, they dream up a new model, a new picture, a new vocabulary, and then they announce that the true meaning of the Book has been discovered. But, of course, it never is, any more than the true meaning of Coriolanus or the Dunciad or the Phenomenology of Spirit, or the Philosophical Investigations. What makes them physicists is that their writings are commentaries on the writings of earlier interpreters of Nature, not that they all are somehow “talking about the same thing,” the invisibilia Dei sive naturae toward which their inquiries steadily converge.

Here is a way of thinking about right and wrong: the common moral consciousness contains certain intuitions concerning equality, fairness, human dignity, and the like, which need to be made explicit through the formulation of principles—principles of the sort which can be used to write legislation. By thinking about puzzle-cases, and by abstracting from differences between our (European) culture and others, we can formulate better and better principles, principles corresponding ever more closely to the moral law itself. [End Page 101]

Here is another way of thinking about right and wrong: the longer men or cultures live, the more phronēsis (wisdom) they may, with luck, acquire—the more sensitivity to others, the more delicate a typology for describing their fellows and themselves. Mingling with others helps; Socratic discussion helps; but since the Romantics, we have been helped most of all by the poets, the novelists, and the ideologues. Since the Phenomenology of Spirit taught us to see not only the history of philosophy, but that of Europe, as portions of a Bildungsroman, we have not striven for moral knowledge as a kind of epistēmē (knowledge). Rather, we have seen Europe’s self-descriptions, and our own self-descriptions, not as ordered to subject matter, but as designs in a tapestry which they will still be weaving after we, and Europe, die.

Here is a way of looking at philosophy: from the beginning, philosophy has worried about the relation between thought and its object, representation and represented. The old problem about reference to the inexistent, for example, has been handled in various unsatisfactory ways because of a failure to distinguish properly philosophical questions about meaning and reference from extraneous questions motivated by scientific, ethical, and religious concerns. Once these questions are properly isolated, however, we can see philosophy as a field which has its center in a series of questions about the relations between words and the world. The recent purifying move from talk of ideas to talk of meanings had dissipated the epistemological skepticism which motivated much of past philosophy. This has left philosophy a more limited, but more self-conscious, rigorous, and coherent area of inquiry.

Here is another way of looking at philosophy: philosophy started off as a confused combination of the love of wisdom and the love of argument. It began with Plato’s notion that the rigor of mathematical argumentation exposed, and could be used to correct, the pretensions of the politicians and the poets. As philosophical thought changed and grew, inseminated by this ambivalent erōs, it produced shoots which took root on their own. Both wisdom and argumentation became far more various than Plato dreamed. Given such nineteenth-century complications as the Bildungsroman, non-Euclidean geometries, ideological historiography, the literary dandy, and the political anarchist, there is no way in which one can isolate philosophy as occupying a distinctive place in culture or concerned with a distinctive subject or proceeding by some distinctive method. One cannot even seek an essence for philosophy as an academic Fach (because one would first have to choose the country in whose universities’ catalogs one was to look). The philosophers’ own scholastic little definitions of “philosophy” are merely polemical devices—intended to exclude from the field of honor those whose pedigrees are unfamiliar. We can pick out “the philosophers” in the contemporary intellectual world only by [End Page 102] noting who is commenting on a certain sequence of historical figures. All that “philosophy” as a name for a sector of culture means is “talk about Plato, Augustine, Descartes, Kant, Hegel, Frege, Russell . . . and that lot.” Philosophy is best seen as a kind of writing. It is delimited, as is any literary genre, not by form or matter, but by tradition—a family romance involving, e.g., Father Parmenides, honest old Uncle Kant, and bad brother Derrida.

There, then, are two ways of thinking about various things. I have drawn them up as reminders of the differences between a philosophical tradition which began, more or less, with Kant, and one which began, more or less, with Hegel’s Phenomenology. The first tradition thinks of truth as a vertical relationship between representations and what is represented. The second tradition thinks of truth horizontally—as the culminating reinterpretation of our predecessors’ reinterpretation of their predecessors’ reinterpretation. . . . This tradition does not ask how representations are related to nonrepresentations, but how representations can be seen as hanging together. The difference is not one between “correspondence” and “coherence” theories of truth—though these so-called theories are partial expressions of this contrast. Rather, it is the difference between regarding truth, goodness, and beauty as eternal objects which we try to locate and reveal, and regarding them as artifacts whose fundamental design we often have to alter. The first tradition takes scientific truth as the center of philosophical concern (and scorns the notion of incommensurable scientific worldpictures). It asks how well other fields of inquiry conform to the model of science. The second tradition takes science as one (not especially privileged nor interesting) sector of culture, a sector which, like all the other sectors, only makes sense when viewed historically. The first likes to present itself as a straightforward, down- to-earth, scientific attempt to get things right. The second needs to present itself obliquely, with the help of as many foreign words and as much allusiveness and name-dropping as possible. Neo-Kantian philosophers like Putnam, Strawson, and Rawls have arguments and theses which are connected to Kant’s by a fairly straightforward series of “purifying” transformations, transformations which are thought to give clearer and clearer views of the persistent problems. For the non-Kantian philosophers, there are no persistent problems—save perhaps the existence of the Kantians. Non-Kantian philosophers like Heidegger and Derrida are emblematic figures who not only do not solve problems, they do not have arguments or theses. They are connected with their predecessors not by common subjects or methods but in the “family resemblance” way in which latecomers in a sequence of commentators on commentators are connected with older members of the same sequence. [End Page 103]

To understand Derrida, one must see his work as the latest development in this non-Kantian, dialectical tradition—the latest attempt of the dialecticians to shatter the Kantians’ ingenuous image of themselves as accurately representing how things really are. Derrida talks a lot about language, and it is tempting to view him as a “philosopher of language” whose work one might usefully compare with other inquiries concerning the relations between words and the world. But it would be less misleading to say that his writing about language is an attempt to show why there should be no philosophy of language. (1) On his view, language is the last refuge of the Kantian tradition, of the notion that there is something eternally present to man’s gaze (the structure of the universe, the moral law, the nature of language) which philosophy can let us see more clearly. The reason why the notion of “philosophy of language” is an illusion is the same reason why philosophy—Kantian philosophy, philosophy as more than a kind of writing—is an illusion. The twenti-eth-century attempt to purify Kant’s general theory about the relation between representations and their objects by turning it into philosophy of language is, for Derrida, to be countered by making philosophy even more impure—more unprofessional, funnier, more allusive, sexier, and above all, more “written.” Thus, insofar as he has an attitude towards, for example, the mini-tradition which stretches from Frege to Davidson, it is the same as his attitude towards Husserl’s discussion of language. The attitude, roughly, is that most twentieth-century concern with language is Kantian philosophy in extremis, a last desperate attempt to do on a pathetically small scale what Kant (and before him Plato) attempted to do on a large scale—show how the atemporally true can be contained in a spatio-temporal vehicle, regularize the relation between man and what man seeks by exhibiting its “structure,” freezing the historical process of successive reinterpretations by exhibiting the structure of all possible interpretation.

Derrida, then, has little to tell us about language, but a great deal to tell us about philosophy. To get a handle on his work, one might take him as answering the question, “Given that philosophy is a kind of writing, why does this suggestion meet with such resistance?” This becomes, in his work, the slightly more particular question, “What must philosophers who object to this characterization think writing is, that they should find the notion that that is what they are doing so offensive?” Whereas Heidegger, Derrida’s great father-figure, was the first to “place” (or if you prefer, “transcend” or “castrate”) Hegel by giving a historical characterization of Hegel’s historicism, Derrida wishes to “place” (or whatever) Heidegger by explaining Heidegger’s distrust of writing. Heidegger, it is true, wrote a lot, but always (after the “turn”) in the interests of urging us to be still and listen to the single line of verse, the individual Greek [End Page 104] word. Derrida is suspicious of Heidegger’s preference for the simplicity and splendor of the word spoken on the hill, and also of his contempt for the footnote scribbled in the ergastulum, down in the valley. The preference, he thinks, betrays a fatal taint of Kantianism, of the Platonic “metaphysics of presence.” For it is characteristic of the Kantian tradition that, no matter how much writing it does, it does not think that philosophy should be “written,” any more than science should be. Writing is an unfortunate necessity; what is really wanted is to show, to demonstrate, to point out, to exhibit, to make one’s interlocutor stand at gaze before the world. The copy theory of ideas, the spectator theory of knowledge, the notion that “understanding representation” is the heart of philosophy, are expressions of this need to substitute an epiphany for a text, to “see through” representation. In a mature science, the words in which the investigator “writes up” his results should be as few and as transparent as possible. Heidegger, though struggling manfully against this cluster of notions, and especially against the notion of the “research project” as model for philosophical thinking, in the end succumbed to the same nostalgia for the innocence and brevity of the spoken word. His substitution of auditory for visual metaphors—of listening to the voice of Being for being a spectator of time and eternity—was, Derrida thinks, only a dodge. The Kantian urge to bring philosophy to an end by solving all its problems, having everything fall into place, and the Heideggerian urge towards Gelassenheit and Unverborgenheit, are the same urge. Philosophical writing, for Heidegger as for the Kantians, is really aimed at putting an end to writing. For Derrida, writing always leads to more writing, and more, and still more—just as history does not lead to Absolute Knowledge or the Final Struggle, but to more history, and more, and still more. The Phenomenology’s vision of truth as what you get by reinterpreting all the previous reinterpretations of reinterpretations still embodies the Platonic ideal of the Last Reinterpretation, the right interpretation at last. Derrida wants to keep the horizontal character of Hegel’s notion of philosophy without its teleology, its sense of direction, its seriousness.

II

So far in this paper I have merely been trying to locate Derrida in philosophical space. Now I want to focus on a few of his remarks about writing, with an eye to seeing more clearly how he answers the question, “What must philosophers think writing to be that they resent so much the suggestion that this is what they do?” His answer is, roughly, that they think that writing is a means of representing facts, and that the more [End Page 105] “written” writing is—the less transparent to what it represents and the more concerned with its relation to others’ writings—the worse it must be. The way he spells out this answer, I think, can help us see why he thinks writing about writing will help to “deconstruct” the Kantian way of looking at things. Consider, to begin with, the following passage:

There is therefore good and bad writing: the good and natural is the divine inscription in the heart and the soul; the perverse and artful is technique, exiled in the exteriority of the body. A modification well within the Platonic diagram: writing of the soul and of the body, writing of the interior and of the exterior, writing of conscience and of the passions, as there is a voice of the soul and a voice of the body. . . .

The good writing has therefore always been comprehended. Comprehended as that which had to be comprehended: within a nature or a natural law, created or not, but first thought within an eternal presence. Comprehended, therefore, within a totality, and enveloped in a volume or a book. The idea of the book is the idea of totality, finite or infinite, of the signifier; this totality of the signifier cannot be a totality, unless a totality constituted by the signified preexists it, supervises its inscriptions and its signs, and is independent of it in its ideality. The idea of the book, which always refers to a natural totality, is profoundly alien to the sense of writing. . . . If I distinguish the text from the book, I shall say that the destruction of the book as it is now under way in all domains, denudes the surface of the text (2)

Consider Derrida as trying, in such passages as this, to create a new thing for writing to be about—not the world, but texts. Books tell the truth about things. Texts comment on other texts, and we should stop trying to test texts for accuracy of representation: “reading . . . cannot legitimately transgress the text toward something other than it, toward the referent (a reality that is metaphysical, historical, psychobiographical, etc.) or toward a signifier outside the text whose content could take place, could have taken place outside of language, that is to say, in the sense that we give here to that word, outside of writing in general. . . . There is nothing outside of the text. (3) Derrida regards the need to overcome “the book”—the notion of a piece of writing as aimed at accurate treatment of a subject, conveying a message which (in more fortunate circumstances) might have been conveyed by ostensive definition or by injecting knowledge straight into the brain—as justifying his use of any text to interpret any other text. The most shocking thing about his work—even more shocking than, though not so funny as, his sexual interpretations of the history of philosophy—is his use of multilingual puns, joke etymologies, allusions from anywhere to anywhere, and phonic and typographical gimmicks. It is as if he really thought that the fact that, for example, the French pronunciation of “Hegel” sounds like the [End Page 106] French word for “eagle” was supposed to be relevant for comprehending Hegel. But Derrida does not want to comprehend Hegel’s books; he wants to play with Hegel. He doesn’t want to write a book about the nature of language; he wants to play with the texts which other people have thought they were writing about language.

At this point one can imagine serious-minded philosophers on both sides of the Channel murmuring about “idealism.” There is a deep terror among Kantian philosophers of a certain job-related health hazard: the philosopher, after overstrenuous inquiry into our relation to the world, may lose his nerve, his reason, and the world simultaneously. He does this by withdrawing into a dream world of ideas, of representations—even, God help us, of texts. To guard against this temptation, Kantian philosophers tell us, we must remember that only the transcendental idealist can be an empirical realist. Only the man who comprehends the relation between representation and represented, in that arduous but rigorously scientific way characteristic of the epistemologist in the last century and the philosopher of language in this, can be transcendental in the required sense. For only he can represent representing itself accurately. Only such an accurate transcendental account of the relationship of representation will keep the Knowing Subject in touch with the Object, word with world, scientist with particle, moral philosopher with the Law, philosophy itself with reality itself. So whenever dialecticians start developing their coherentist and historicist views, Kantians explain that it is another sad case of Berkeley’s disease, and that there is no cure save a still better, more luminously convincing, more transparent philosophical account of representation.

When dialectical philosophers are accused of idealism, they usually reply as Berkeley replied to his critics—by explaining that they are only protesting against the errors of a certain philosophical school and that they are really not saying anything at which the plain man would demur. As Austin said in this connection, “There’s the bit where you say it and the bit where you take it back.” The nice thing about Derrida is that he doesn’t take it back. He has no interest in bringing “his philosophy” into accord with common sense. He is not writing a philosophy. He is not giving an account of anything; he is not offering a comprehensive view of anything. He is not protesting against the errors of a philosophical school. He is, however, protesting against the notion that the philosophy of language, pursued “realistically” as the study of how language hooks on to the world, is something more than one more quaint little genre, that it is first philosophy. But the protest is not because he has a different candidate for the position of “first philosophy,” it is against the notion of “first philosophy.” He could, if he liked, say that he, too, can pass judgments within this genre—that he recognizes better and worse “realistic” [End Page 107] philosophy of language, that he agrees with all up-to-date philosophers of language that Strawson and Searle were terribly wrong about the referents of proper names, and so on. But what he really wants to do is to say, “You used to think that it was terribly important to get meaning and reference, and all that, right. But it isn’t. You only thought that because. . . .” He might be compared with the secularist who says not “There is no God” but rather “All this talk about our relation to God is getting in our way.” James, when he said that “the true is what is good in the way of belief” was simply trying to debunk epistemology; he was not offering a “theory of truth.” So Derrida, when he says “il n’y a pas de hors-texte,” is not putting forward an ontological view; he is trying to debunk Kantian philosophy generally.

Well then, one might reply, he does take it back. For he admits that all this stuff about there not being any such thing as accuracy of representation is metaphorical, just a way of speaking. But why doesn’t he say what he means? Why doesn’t he come right out and tell his views about language and about reality? To this one can only reiterate that Derrida is in the same situation in regard to language that many of us secularists are in regard to God. It isn’t that we believe in God, or don’t believe in God, or have suspended judgment about God, or consider that the God of theism is an inadequate symbol of our ultimate concern; it is just that we wish we didn’t have to have a view about God. It isn’t that we know that “God” is a cognitively meaningless expression, or that it has its role in a language-game other than the fact-stating, or whatever. We just regret the fact that the word is used so much. So it is for Derrida with the vocabulary of Kantian philosophy. His attitude towards centuries of worry about the relation between subject and object, representations and the real, is like the Enlightenment attitude toward centuries of worry about the relation between God and man, faith and reason. Indeed, for Derrida as for Heidegger, these worries are all the same worry: the worry that we may lose touch with certain exigencies, conformity with which is the whole duty of man. For Derrida as for Freud, these are all forms of the worry about what our fathers require of us. For Derrida as for Sartre, these are all forms of the attempt to know oneself by transforming oneself into a knowable object—an être-en-soi which obeys the laws of its kind.

So, to sum up the gloss I want to put upon the texts I have quoted from Derrida, Derrida is trying to do for our highbrow culture what secularist intellectuals in the nineteenth century tried to do for theirs. He is suggesting how things might look if we did not have Kantian philosophy built into the fabric of our intellectual life, as his predecessors suggested how things might look if we did not have religion built into the fabric of our moral life. The secularists I speak of were continually assailed by the question, “What argument do you have for not believing in God?” Derrida is continually [End Page 108] assailed by the question, “What argument do you have for saying that we should not refer the text to something which is not a text?” Neither has any interesting arguments, because both are not working by the same rules as their opponents. They are trying to make up some new rules. Lack of seriousness, in the sense in which I just attributed it to Derrida, is simply this refusal to take the standard rules seriously, conjoined with the refusal to give a clear answer to the question, “Is it the old game played differently, or rather a new game?”

There is another sense, however, in which Derrida is very serious indeed—as serious as the prophets of secularization. He is serious about the need to change ourselves, serious about what he calls “deconstruction.” Thus he warns us against taking “grammatology” as the name of a new research program, as an attempt at doing something constructive and progressive, when he speaks of “the systematic crossing-out of the archē [origin] and the transformation of general semiology into a grammatology, the latter performing a critical work upon everything within semiology—right down to its matrical concept of signs—that retains any metaphysical presuppositions incompatible with the theme of differance.” (4) One can easily conclude from such passages as this that Derrida conceives of his work as purely negative—deconstructing the metaphysics of presence in order to leave the texts bare, unburdened by the need to represent. Such a view is also suggested by his excusing his high-handed treatment of Saussure in this way: “My justification would be as follows: This [a text of Saussure's] and some other indices (in a general way the treatment of the concept of writing) already give us the assured means of broaching the de-construction of the greatest totality—the concept of the episteme and logocentric metaphysics—from which are produced, without ever posing the radical question of writing, all the Western methods of analysis, explication, reading, or interpretation.” (5) This passage conforms to the picture of Derrida I have offered so far—one in which he wants to do better than Heidegger the job of “overcoming the tradition of Western metaphysics” which Heidegger attempted. This picture may, however, be too charitable. For there is a side of Derrida which looks unfortunately constructive, a side which makes it look as if he in the end succumbs to nostalgia, to the lure of philosophical system-building, and specifically that of constructing yet another transcendental idealism. I turn now, therefore, to a discussion of the luminous, constructive, bad side of Derrida’s work, as opposed to the shadowy, deconstructive, good side which I have been discussing so far. [End Page 109]

III

To explain where and why Derrida seems to get constructive, I need to go back to the point I was making earlier about his attitude towards “the philosophy of language.” One can see Derrida’s attempt to “deconstruct the greatest totality” as an attempt to get rid of the notion that language is an attempt to represent something nonlinguistic. He is taking the Wittgensteinian doctrine which Sellars calls “psychological nominalism”—the doctrine that “all awareness is a linguistic affair”—to its extreme. (6) But he sees the recent attention to language (as a general subject matter of inquiry, comparable in scope to God, nature, history, or man) as a kind of pseudonominalism. (7) It is as if the Kantians had been forced, by attacks on the notions of “thought” and “the mind,” to see that there is no way to cut beneath language to the thought which language expresses, no way, as Wittgenstein said, to “get between language and its object.” But instead of concluding that we should stop viewing language as representing something, the Kantian response has been to say something like, “Now that we see that language is not the expression of thought, but since we know that language does represent the world, we can now be properly serious about language, can pay language the attention it deserves, by exploring direct word-world connections.” What looks to modern philosophers of language like a new-found respect for language is, for Derrida, simply a disguised attempt to put language in its place, to insist that language has responsibilities to something outside itself, that it must be “adequate” to do its representative job. Derrida thinks that the proper moral to draw is that language is not a tool, but that in which we live and move. So to ask “how does language manage to do its job?” betrays psychological nominalism. If all awareness is a linguistic affair, then we are never going to be aware of a word on the one hand and a thing-denuded-of-words on the other and see that the first is adequate to the second. But the very notions of “sign” and “representation” and “language” convey the notion that we can do something like that. The notion of philosophy of language as the successor-subject to epistemology suggests that we have now found out how to study representation properly, and thus to do properly the job which Kant saw needed to be done.

Given this situation, Derrida looks about for a way to say something about language which will not convey the idea of “sign” or “representation” or “supplement.” His solution consists in such notions as trace, which have, recently, become something very much like a new “subject matter” for his followers. But in developing this alternative, Derrida comes perilously close to giving us a philosophy of language, and thereby perilously close to slipping back into what he and Heidegger call “the tradition of onto-theology.” That tradition is kept going by the following [End Page 110] dialectical movement: first one notices that something all-encompassing and unconditioned is being treated as if it were just one more limited and conditioned thing. Then one explains that this thing is so distinctive that it requires an entirely different vocabulary for its description, and proceeds to create one. Finally, one’s disciples become so bemused by one’s new vocabulary that they think one has invented a new field of inquiry, and the whole sequence starts up once again. This happened to “God” when the Platonism of the Church Fathers lifted the divine out of space and time and insisted on His consequent ineffability. God thus became a pigeon for Doctors of the Church who had read Aristotle; they explained how the ineffable could be effed after all, but only analogically. It happened to “Mind” when Kant explained (in the Paralogisms) that the subject was not a substance, thus permitting Fichte and the nineteenth century to explain that really there was a lot to say about the Subject, but only transcendentally. In both cases, somebody (Augustine, Kant) warns against us trying to describe the unconditioned, and somebody else (Aquinas, Fichte) dreams up a special technique designed especially for the purpose. If I am right in my suspicions about Derrida, we are in some danger of seeing this same pattern repeated by Heidegger and Derrida. We may find ourselves thinking that what Heidegger thought could not be effed really can be, if only grammatologically.

Heidegger spent his life explaining that all his predecessors had ignored the “ontological difference” between beings and Being, and finally wound up suggesting that one had better write the word being only X-ed out. (8) Heidegger kept trying to fend off disciples who said, “Now that we have the ontological difference clearly in mind, tell us something about Being.” Finally he said that the attempt to say that the tradition of metaphysics, of onto-theology, had confused beings with Being was itself a misleadingly metaphysical attempt. He ends his “Time and Being” by saying:

To think Being without beings means: to think Being without regard to metaphysics. Yet a regard for metaphysics still prevails even in the intention to overcome metaphysics. Therefore, our task is to cease all overcoming, and leave metaphysics to oneself.

If overcoming remains necessary, it concerns that thinking that explicitly enters Appropriation in order to say It in terms of It about It.

Our task is unceasingly to overcome the obstacles that tend to render such saying inadequate.

The saying of Appropriation in the form of a lecture remains itself an obstacle of this kind. The lecture has spoken merely in propositional statements.(9)

But, of course, Appropriation (Ereignis) looks like one more name for the goal of our inquiries. This movement of Heidegger’s thought back from one ineffable to another (e.g., from “Being” to “Appropriation”) [End Page 111] just as soon as people begin to eff the first ineffable, can be viewed as an attempt to find something which cannot be the subject of a commentary, something which cannot be the subject of an inquiry into, for example, “Heidegger’s doctrine of Ereignis.” Derrida thinks, or at least thought when he began De la Grammatologie, that the only way to solve Heidegger’s problem was to get away from terminology borrowed from the visual and aural imagery of earlier authors and invent a new way which had to do only with writing. One can see this impulse in the following passages:

The reassuring evidence within which Western tradition had to organize itself and must continue to live would therefore be as follows: the order of the signified is never contemporary, is at best the subtly discrepant inverse or parallel—discrepant by the time of a breath—from the order of the signifier. And the sign must be the unity of a heterogeneity, since the signified (sense or thing, noeme or reality) is not in itself a signifier, a trace: in any case is not constituted in its sense by its relationship with a possible trace. The formal essence of the signified is presence, and the privilege of its proximity to the logos as phonè is the privilege of presence. This is the inevitable response as soon as one asks: “what is the sign?,” this is to say, when one submits the sign to the question of essence, to the “ti esti.” The “formal essence” of the sign can only be determined in terms of presence. One cannot get around that response, except by challenging the very form of the question and beginning to think that the sign is that ill-named thing, the only one, that escapes the instituting question of philosophy: “what is . . . ?” (10)

What should give one pause in this passage is the phrase “the only one.” It is as if Derrida thought he had done the one thing Heidegger failed to do—find the one word which cannot be the subject of a commentary, a PhD thesis on “Derrida’s doctrine of the sign,” the one expression of the unconditioned which nobody will ever be able to treat as if it were the name of one more conditioned. In the following passage also one can see such a notion: “The movement of the effacement of the trace has been, from Plato to Rousseau to Hegel, imposed upon writing in the narrow sense; the necessity of such a displacement may now be apparent. Writing is one of the representatives of the trace in general, it is not the trace itself. The trace itself does not exist. (To exist is to be, to be an entity, a being-present, to on.)” (11) One can comment cynically on this passage that, if you want to know what notion takes the place of God for a writer in the onto-theological tradition, always look for the one which he says does not exist. That will be the name of the Ineffable, of what can be shown but not said, believed but not known, presupposed but not mentioned, that in which we live and move and have our being. It is the need to express the unconditioned while realizing that it is inexpressible which brings us to the point described by Wittgenstein: “Sometimes, in doing philosophy, one just wants to utter an inarticulate [End Page 112] sound.” (12) But that will not prevent somebody writing a thesis on whatever sound one makes.

Fortunately, however, Derrida was the first to warn us against the temptation I have just described—the temptation to divinize the trace, and to treat writing as “one of the representatives of the trace in general, but not the trace itself” (a passage which seems to make trace one of the invisibilia Dei, per ea quae factae sunt cognoscuntur). In “Differance,” published just after On Grammatology, he identifies the difference he hoped to find between “the sign” as the only thing that escapes the instituting question of philosophy and all the other failed candidates for this role with Heidegger’s “ontological difference.” He turns himself, in this essay, from something dangerously like a philosopher of language, into a philosopher of philosophy, where philosophy is just the self-consciousness of the play of a certain kind of writing. Différance, unlike trace, has no more to do with signs than it does with things or gods or minds or any of the other things for which Kantian philosophy has sought the unconditioned conditions. Différance is a name of the situation which the dialectical philosopher starts from—the wish to revolt against the eternalization and cosmologization of the present vocabulary by creating a new vocabulary which will not permit the old questions to be asked; it is the “make it new” which Pound thought expressed “modernism.” In “Differance,” Derrida has a passage which forms a splendid rebuke both to Heidegger and to his previous self:

For us, differance remains a metaphysical name; and all the names that it receives from our language are still, so far as they are names, metaphysical. . . .

“Older” than Being itself, our language has no name for such a differance. But we “already know” that if it is unnamable, this is not simply provisional; it is not because our language has still not found or received this name, or because we would have to look for it in another language, outside the finite system of our language. It is because there is no name for this—not even essence or Being—not even the name “differance,” which is not a name, which is not a pure nominal unity, and continually breaks up in a chain of different substitutions. . . .

There will be no unique name, not even the name of Being. It must be conceived without nostalgia; that is, it must be conceived outside the myth of the purely maternal or paternal language belonging to the lost fatherland of thought. On the contrary we must affirm it—in the sense that Nietzsche brings affirmation into play—with a certain laughter and with a certain dance. (13)

IV

Let me now turn to what may seem the chief question raised by what I’ve said so far: granted that Derrida is the latest and largest flower on [End Page 113] the dialectical kudzu vine of which the Phenomenology of Spirit was the first tendril, does that not merely show the need to uproot this creeping menace? Can we not now see all the better the need to strip the suckers of this parasitic climber from the still unfinished walls and roofs of the great Kantian edifice which it covers and conceals? Granted that if all this nonsense about language not being a system of representations were true, Derrida would have drawn some interesting consequences from it, cannot we now return to sanity and say that it is false, and that philosophy would do well to return to the slow and patient work of understanding how representation is accomplished?

The dialectical response to this should, I think, be twofold. First, one can reply that the question of whether language is a system of representations is not the sort of question anybody (Kantian or non-Kantian) knows how to answer, and so whatever is at issue, that cannot be it. The question is not whether “language is a system of representations” is a correct representation of how things are. Second, one can reply that of course language can usefully, for many purposes, be viewed as a system of representations, just as physical theory can usefully be seen as an approximation to what we would see if we could get down there among the quanta, moral philosophy as an approximation to the Moral Law, and philosophy as a quest for a purer and better way to answer traditional questions. All that one has to do to make any of these approaches useful and productive is to take the vocabulary of the present historical period (or class or society or academy) for granted and to work within it. Once one is safely ensconced within this language-game, questions about what correctly represents what, how we know that it does, and how it manages to do so will make admirable sense and will get useful answers. There is nothing done within the Kantian tradition which the dialectical tradition cannot treat as the description of the practices of a certain historical moment—the sort of description one gets when one blinkers one’s historical consciousness temporarily for the sake of getting a clear view of what is currently going on. The traditions come into real conflict only when the Kantian tradition cosmologizes and eternalizes its current view of physics, or right or wrong, or philosophy, or language. Thus, for example, if we freeze physics at a period of what Kuhn calls “normal science,” we can describe the practice of justifying theories in terms of a determinate observation-language, meaning-rules, and canons of theory-choice. If we try to bring this heuristic apparatus to bear on all the things that might count as explaining nature in various periods and cultures, however, we either become viciously anachronistic or fall into pointless puzzlement about, e.g., “criteria for change of reference of theoretical terms.” Analogously, if we take as data a range of assertions running the gamut from “the cat is on the mat” to “the particle went through the left-hand slit,” [End Page 114] we may be able to construct an account of the contribution of the parts of the expressions to the wholes, and of the conditions under which a language-user would be justified in employing them. We go wrong only when we invoke this account to be condescending about, or be baffled by, such assertions as “caloric fluid is just a lot of moving molecules,” “language speaks man,” or “God’s essence is his existence.” If we then try to be systematically invidious or reductive by talking about “literal vs. metaphorical” or “non-statement-making uses of declarative sentences,” or the like, then philosophy of language will begin to seem relevant to epistemology, controversial, and essential to our self-understanding. It will also seem to come into conflict with the sort of thing that Heidegger and Derrida are telling us. Worse yet, the sort of thing Heidegger and Derrida are saying may come to look like competing attempts to do what, e.g., Frege and Carnap and Putnam do.

No such competition exists. There is no topic—and in particular not that of the relation between sign and signified, language and the world—on which Derrida holds a different view than that of any of the philosophers of language I have mentioned. Nor does he have any insights which complement theirs. He is not, to repeat, a philosopher of language. The closest Derrida comes to the philosophy of language is his interest in the historical question of why a view about the relation between sign and signified, the nature of representation, could ever have been thought to have been essential to our self-understanding, the starting point of the love of wisdom, first philosophy. He is interested in the connection between the “Kantian” view of philosophy and the “Kantian” view of language—in why the latest Kantian effort to cosmologize or eternalize the present should have centered on language. Here he does have something to say-—but it is something about philosophy, not about language.

Kantian philosophy, on Derrida’s view, is a kind of writing which would like not to be a kind of writing. It is a genre which would like to be a gesture, a clap of thunder, an epiphany. That is where God and man, thought and its object, words and the world meet, we want speechlessly to say; let no further words come between the happy pair. Kantian philosophers would like not to write, but just to show. They would like the words they use to be so simple as to be presuppositionless. Some of them like to think that physics, too, is not a kind of writing. So they cherish the thought that, at least in some countries, philosophy has no literary pretensions because it has attained the secure path of a science. Just as, on the Kantian view of physics, physics has no need of a historical self-understanding to enable it to point straight to the heart of matter, so, on the Kantian view of philosophy, philosophers need not be concerned with their own Kantian motives in order to point straight to the heart [End Page 115] of spirit—the relation of representation itself. Derrida’s reply is that anybody can get along without literary pretensions—without writing—if he is content simply to demonstrate how something falls into place in a previously established context. In normal physics, normal philosophy, normal moralizing or preaching, one hopes for the normal thrill of just the right piece fitting into just the right slot, with a shuddering resonance which makes verbal commentary superfluous and inappropriate. Writing, as Derrida says in commenting on Rousseau, is to this kind of simple “getting it right” as masturbation is to standard, solid, reassuring sex. This is why writers are thought effete in comparison with scientists—the “men of action” of our latter days. The important thing to notice is that the difference between the two forms of activity is not subject matter—not, for instance, a matter of the difference between the flinty particles of the hard sciences and the flexible behavior of the soft ones—but rather is determined by normality or abnormality. Normality, in this sense, is accepting without question the stage-setting in the language which gives demonstration (scientific or ostensive) its legitimacy. Revolutionary scientists need to write, as normal scientists do not. Revolutionary politicians need to write, as parliamentary politicians do not. Dialectical philosophers like Derrida need to write, as Kantian philosophers do not.

The Freudian distinction between the normal and the abnormal, drawn with the concreteness which is given by Derrida’s exhibition of the sexual overtones of most metaphilosophical debate, seems to me just what is needed to be properly playful about the difference between the Kantians and the Hegelians. If one thinks of this difference as that between the partisans of Eternity and of Time, or those of Theory and of Practice, Nature and History, Permanence and Change, Intellect and Intuition, the Sciences and the Arts, it all looks too momentous, too much as if there were a serious and debatable issue around. The issue between Kantian and non-Kantian philosophy is, I think, about as serious as the issue between normal and deviant sexual practices.

It is, to be sure, an issue upon which men may well feel their identity and their integrity depend. (“Men,” rather than “people,” since taking how and what one does in bed as definitive of one’s being seems a specifically masculine trait.) So it is not unserious in the sense of unimportant. But it is not a serious issue in the sense of a debatable one, on which there is much to be said on both sides. It is not an issue which we ought all to pitch in and try to resolve (in some more discursive way than massacring the opposition). Indeed, we had better not. For if the issue were ever resolved, there might not be any more philosophy. (Or any more interesting writing at all. Philosophy is, after all, dominatrix disciplinarum if no longer regina scientiarum; nobody does any really “written” writing without timidly hoping that what he writes may have “philosophical [End Page 116] implications.”) Similarly, if the difference between normal and deviant sex ever got settled—not by massacre, but by rational demonstration of the moral superiority of one side or the other, or of their being morally equivalent—then it is not clear that sex would matter nearly as much as it does now. When Freud told us that we had sexual repression to thank for the hang-ups of the neurotics who created European culture, he meant exactly what he said. If Derrida is on the right track in his post-Grammatology treatment of philosophical texts, we can be a bit more specific about just how this culture was fed by sublimated sexuality. The Kantian versus non-Kantian contrast now appears as that between the man who wants to take (and see) things as they are, and thus make sure that the right pieces go in the right holes, and the man who wants to change the vocabulary presently used for isolating pieces and holes. This helps us see why the dialectic of the conditioned and the unconditioned, the effable and the ineffable, has the peculiar thrill that it does. Unspeakable possibilities, unmentionable acts are those which are spoken and mentioned in the new, revolutionary, Hegelian, abnormal vocabulary. Sartre’s account of the attempt of the philosopher to become God by recreating himself as a pour-soi-en-soi joins up with Freud’s to suggest that the Kantian tradition plays the role in recent European culture of the normal man, the man whose respect for the law is such that he would wish the natural and the moral law to be as one.

This Freudian twist also helps us see why, even given the compatibility of, e.g., everything which Derrida says and everything Quine says, we cannot relax and split the difference. We cannot just let Kantians have their (self-eliminating) kind of writing and the Hegelians their (selfextending, kudzu-like) alternative kind. Being conciliatory in this way would obscure the fact that these traditions live each other’s death, die each other’s life—the same relation which holds between normal and abnormal sex. The dialectician will always win if he waits long enough, for the Kantian norm will in time become tedious, full of anomie and anomaly. The Kantian, on the other hand, escapes triviality, and achieves self-identity and self-conscious pride, only by the contrast between his mighty deeds and the mere words of the dialectician. He is no effete parasite, but one who does his share in the mighty time-binding work of building the edifice of human knowledge, human society, the City of Real Men. The non-Kantian knows that the edifice will itself one day be deconstructed, and the great deeds reinterpreted, and reinterpreted again, and again. But of course the non-Kantian is a parasite—flowers could not sprout from the dialectical vine unless there were an edifice into whose chinks it could insert its tendrils. No constructors, no deconstructors. No norms, no perversions. Derrida (like Heidegger) would have no writing to do unless there were a “metaphysics of presence” to overcome. Without the fun [End Page 117] of stamping out parasites, on the other hand, no Kantian would bother to continue building. Normal philosophers need to think, for example, that in forging the powerful tools of modern analytic philosophy, they are developing weapons to ensure victory in the coming final struggle with the decadent dialecticians. Everybody needs everybody else.

Kantian and non-Kantian metaphilosophers, when this point in the development of their self-consciousness is reached, like to explain that their opponents really want to do what they themselves are doing. The Kantian thinks of the non-Kantian as somebody who would like to have a proper, disciplined, philosophical view about, e.g., words and the world, but can’t quite manage to get it together into a coherent, rigorous form. The Hegelian likes to think that there is not really a contrast between the vine and the edifice it covers—rather, the so-called edifice is just accumulated dead wood, parts of the Great Vine itself, which once were fresh and flower-laden but now have come to lie in positions which suggest the outlines of a building. So the normal man sees the abnormal as not quite up to it—more to be pitied than censured. The abnormal sees the normal as someone who never had the courage to come out, and so died inside while his body lived on—more to be helped than despised.

This kind of crosstalk can continue indefinitely. Derrida’s point, I take it, is that that crosstalk is all that we are going to get, and that no gimmick like “the new science of grammatology” is going to end or aufheben it. Once one thinks of philosophy as a kind of writing, one should not be surprised at this result. For to think this is to stop trying to have a philosophy of language which is “first philosophy,” a view of all possible views, an epistēmē epistēmēs (knowledge of knowledge), a bootstrap self-elevation to a point from which all past and future writing can be seen as contained within a permanent framework. Only one who had levitated to such a point would have the right to look down on writing, to view it as a second-best (like Plato) or as an abnormal activity to which sin has condemned him (like Rousseau), or as something which a discipline can dispense with on reaching the secure path of a science. Derrida’s polemic against the notion that speech is prior to writing should be seen as a polemic against what Sartre calls “bad faith”—the attempt to divinize oneself by seeing in advance the terms in which all possible problems are to be set, and the criteria for their resolution. If the “logocentric,” Platonic notion of speech as prior to writing were right, there might be a last Word. Derrida’s point is that no one can make sense of the notion of a last commentary, a last discussion note, a good piece of writing which is not an occasion for a better piece. [End Page 118]

Notes

This essay was originally published in “Literary Hermeneutics,” New Literary History 10, no. 1 (1978).

1 On Derrida’s relation to contemporary philosophy of language, and especially to Wittgenstein, see Newton Garver, preface to Jacques Derrida, Speech and Phenomenon, and Other Essays on Husserl’s Theory of Signs, trans. David B. Allison (Evanston, IL: Northwestern Univ. Press, 1972), and “Derrida on Rousseau on Writing,” Journal of Philosophy 74 (1977): 663–73; Marjorie Grene, “Life, Death and Language: Some Thoughts on Wittgenstein and Derrida,” in Philosophy In and Out of Europe (Berkeley and Los Angeles: Univ. of California Press, 1976), 142–54; John Searle’s exchange with Derrida in the first two volumes of Glyph; Richard Rorty, “Derrida on Language, Being, and Abnormal Philosophy,” Journal of Philosophy 74 (1977): 673–81.

2 Derrida, Of Grammatology, trans. Gayatri Chakravorty Spivak (Baltimore: Johns Hopkins Univ. Press, 1976), 17–18.

3 Derrida, Of Grammatology, 158.

4 From “Differance,” in Derrida, Speech and Phenomenon, 146.

5 Derrida, Of Grammatology, 46.

6 See Wilfrid Sellars, Science, Perception and Reality (London: Humanities Press, 1963), 160 ff.

7 Derrida, Of Grammatology, 6.

8 Cf. Martin Heidegger, The Question of Being, trans. William Kluback and Jean T. Wilde (New York: Twayne Publishers, 1958).

9 Heidegger, On Time and Being, trans. Joan Stambaugh (New York: Harper and Row, 1972), 24.

10 Derrida, Of Grammatology, 18–19.

11 Derrida, Of Grammatology, 167.

12 Ludwig Wittgenstein, Philosophical Investigations (New York: Macmillan, 1953), pt. 1, sec. 261.

13 Derrida, Speech and Phenomenon, 158–59. [End Page 119]