The famous preface to the first edition of Immanuel Kant’s Critique of Pure Reason includes a story full of violence and wit concerning the battle for the right to legislate over the relationship between experience and interpretation. The battlefield, Kant says, is called metaphysics.
In the beginning, under the administration of the dogmatists, her rule was despotic. Yet because her legislation still retained traces of ancient barbarism, this rule gradually degenerated through internal wars into complete anarchy; and the skeptics, a breed of nomads [eine Art Nomaden] who abhor all permanent cultivation of the soil, shattered civil unity from time to time. But since there were only a few of them, they could not prevent the dogmatists from continuously attempting to rebuild, though never according to a plan unanimously accepted among themselves. (CPR 99-100)
There is much at stake for Kant in this hyperbolic and cartoonish account of the history of philosophy. The antagonism and the lack of legal authority for the state of knowledge suggests both a crisis and the promise of its resolution. The story translates the following famous philosophical dogma. Neither the rationalist project of Christian Wolff nor the empiricist project of John Locke had been able to defend metaphysics from the nomadic terror of David Hume’s skepticism. The critical philosophy of Immanuel Kant thus steps in to clear the domain. There are responses to a certain deconstruction that echo the terms of Kant’s fable. The nomadic terror of the post- disturbs, destroys even, the safety of a secure identity, a place, a domain for knowledge. The term deconstruction seems particularly suited to supporting this of course and for this reason one often finds the term skepticism used to describe it. But perhaps deconstruction remains closer to Kant’s reason than the term skepticism implies.
It is because deconstruction interferes with solid structures, “material” institutions, and not only with discourses or signifying representations, that it is always distinct from an analysis or a “critique.” (TP 19)
According to this brief citation from Jacques Derrida deconstruction cannot be compared with the academic practices that are normally grasped as belonging to academic institutions like universities. This would be because deconstruction interferes with the very structure of the institution. What is this thing called deconstruction that interferes with our structures, that disturbs the stays, the posts and beams, of the institutions within which, if we can say this, analysis and critique goes on? What I am writing about is this deconstruction, if it exists. But how can I? Concerning deconstruction (if it exists) I must first of all say that there is no field of experience or domain of thought, nor any discipline, from which it would be possible to decide whether it exists or not. And already, in case the it of my last sentence should be saying too much, I have introduced the supplement of the parenthetical (if there is any) or (if it exists) and now add a further supplement in perhaps not. There are many concepts, like God, fortune and fate, or justice, which can bear this supplement (that perhaps it does not exist) rendering any decision about it impossible. This is not controversial. But what if deconstruction names a field of operation that perhaps does not exist yet whose undecidability is grounded? My little supplements, which I’ll be collecting up gradually over the course of the paper, would constitute the ground I’m talking about. Deconstruction (if it exists) would then be the only ground available if you want to avoid what I’ll refer to as transcendental contraband in the arts and sciences. This contraband is not an insignificant matter because institutions and the practices that go on in their name are built upon it. These are the institutions that deconstruction, if it exists, interferes with.
You may have noticed that my title, “Rubrics and Lemmata or the Post in Theory,” is full of titles, or titles of titles. It is one of those titles that give a choice. You can take it as indicating that the paper will be about, will treat of, either rubrics and lemmata or the post in theory. The or, of course, indicates an equivalence. Or repeats in a different way what comes before it by what comes after it. Rubrics and lemmata are in some way the equivalent of the post in theory. That could mean that the post comes under the rubric of rubrics or of lemmata or that under the rubric post we will find a systematic treatment of rubrics and lemmata. The rubric covers the topic in essential outline to be filled in, post and pan, by the treatment itself.
In this case I have a difficulty. It is not just mine I’m pleased to say. It is a difficulty that reaches into every domain where interpretations are made, or theories applied. We can start by attempting to define and delimit the field of our inquiry. We thus begin with a rubric, a heading, the headword of my title, a red head. In that delimitation we make a distinction. On the one hand there is the field to be interpreted, the basic components of a present and practical experience--the rubrics are multiplying--the contents of a strictly empirical knowledge. On the other hand there are the headings, rubrics and lemmata, concepts, ideas, notions that are from the beginning already setting up the expectations into which the field of my empirical experience cannot but fall. Expectation is like a horizon into which the future and all its surprises just fall.
Because I’m something of a Kantian I’ll have some recourse to the term transcendental shortly as a way of shoring up my empirical knowledge against contraband ideas. When I say I am a Kantian that does not, if anybody is wondering, mean that I take a philosophical perspective that defers to Kant’s authority. Rather, as a literary critic with a post in theory I would be a Kantian perhaps in the way that a reader of Shakespeare is a Shakespearean. I read Kant. Kant, too, in my empirical field, is to be interpreted.
I’m alluding, among other things, to a reading of Kant for two reasons. The first is as a response to the lemma of the workshop “Experience and Interpretation.” A lemma in the sense I mean in this instance is something taken. There are two etymologies. In the first the Latin “lemma” comes from Greek lemma meaning “thing taken” or “assumption,” from lambanein “to take.” There are three current meanings:
First, a lemma is an auxiliary proposition used in the demonstration of another proposition. This is most commonly used in mathematics in which one can find all kinds of stunning and complex lemmata. I should say that all my propositions are auxiliary so my lemmata support only lemmata.
Secondly a lemma is an argument or theme of a composition prefixed as a title or introduction, as well as the heading or theme of a comment or note on a text. In this sense the lemma “Workshop on Experience and Interpretation Across Domains” contains the rubric “Experience and Interpretation.” It is a title under which my own title is to be found. And it is one of the two reasons why I turn to Kant.
Thirdly, a lemma is a glossed word or phrase. In that case the last few sentences compose a lemma on lemmata. I can now also say a little bit about this paper, which is entirely constructed of lemmata, and is but an excerpt from what is in principle a (perhaps) infinitely extendible piece. I can add a lemma anywhere--I might even do this today.
The second reason I turn to Kant (if not just for reason itself of course) emerges when we consider an apparently tragic and as yet unfulfilled love affair. In the “Architectonic of pure reason,” which is the Third Chapter of the Transcendental Doctrine of Method, Kant outlines what an architectonic would be. Philosophy should become something like an encyclopedic system, a system that would not be a technical or instrumental aggregate of knowledge (i.e., deriving from empirical criteria arising from scientific discoveries), but would act as the lawgiver of reason, governing not simply all that is (being in general) but governing morals too. Such an architectonic does not yet exist except as one of reason’s ideas.
Such a science . . . prevents the devastations that a lawless speculative reason would inevitably perpetrate in both morality and religion. We can therefore be sure that however obstinate or disdainful they may be who know how to judge a science not in accord with its nature, but only from its contingent effects, we will always return to metaphysics as to a beloved from whom we have been estranged, since reason, because essential ends are at issue here, must work without respite either for sound insight or against the destruction of good insights that are already to hand. (CPR 700)
This estranged lover who provides the architectonic law, who restrains, who guides, who protects and who shields us from the obstinacy of the empirical scientist remains unfortunately for us an absent lover. Nonetheless, given that so much of this must be couched in terms of law, which also constitutes the nature of the crisis called postmodernism, we can now explore the problem of the kind of knowledge lurking behind these enigmatic posts of theory in terms of its law. As a reader the kind of law I’d be interested in to begin with would be the one that governed the texts to be read. But then that puts me in a peculiar situation, as the contexts multiply, because as Derrida reminds us in the opening paragraph from Plato’s Pharmacy, a text must hide its law.
A text is not a text unless it hides from its first comer, from the first glance, the law of its composition and the rules of its game. A text remains, moreover, forever imperceptible. Its law and its rules are not, however harbored in the inaccessibility of a secret; it is simply that they can never be booked, in the present, into anything that could rigorously be called a perception. (D 63)
Is it simply a structure that is hidden, as the structuralists might have said? I’d like to try to take seriously the terms law and rules in the context of the logic of the argument: “A text is not a text unless . . .” suggests that a claim is made regarding the a priori necessary conditions of possibility of a text. What must a text be in order for it to be a text? It is only a text if “it hides from the first comer, from the first glance, the law of its composition and the rules of its game.” What kind of game do you play when it is a game for which you don’t know or cannot see the rules? A text can only be a text if it fails to prescribe the rules for reading it. Thus a reading might be an attempt to supply rules. There’s no surprise there given the persistent attempts of literary theorists, philosophers and hermeneutists to do just that. Some texts, of course, do attempt to prescribe the rules of their own game, in the face of which there are many examples in critical theory of readings that describe and affirm their failure. Why should a text hide its law? If a text is something that can be read it must be readable in some future context whose conditions remain permanently and contingently unheard of. There is no law for them for they concern future contingencies. Reading is therefore also a kind of writing because it provides the necessary law—contingently and always belatedly—without which there just would be no game. But that law then also remains hidden from its first comer, the diachronically next in line—also unheard of. Is the law not then determined by the absolute alterity of its singular addressee? By a strange circularity this singularity, also always possibly absent, is that of the addresser too. Arche and telos always potentially missing, the text rests on that necessity (the necessity of this singular possible absence) and as a matter of fact hides them anyway as the singularity (doubled) of its law. Hidden possibilities possibly absent. The citation goes on: “A text remains, moreover, forever imperceptible. Its law and its rules are not, however, harboured in the inaccessibility of a secret; it is simply that they can never be booked, in the present, into anything that could rigorously be called a perception.” This kind of law could never be the object of a perception because it is determined by a future that is always to come and a past that has always passed. Otherwise a text would not be what it is. It is rather overdetermined by a play of unconscious forces, which would involve something like the resources that the unconscious uses in order to effect its displacements and condensations. This concerns what happens when one cannot reduce the effects of chance any further (Freud attempts to reduce chance to the unconscious in The Psychopathology of Everyday Life to the extent that, absurdly, there’d be no accident).
Overdetermination describes Freud’s unconscious as a “thought factory” on the analogy of an inexhaustibly productive team of weavers, the shuttle flying over here and over there, which I’ll outline below.
Freud was by no means the first neurologist to refer to the fact that a symptom may have multiple causations. He does seem to be one of the few in the late 19th century to be making claims such that multiple causation is the rule rather than the interesting exception. In Studies on Hysteria he points out that:
There is in principle no difference between the symptom’s appearing in a temporary way after its first provoking cause and its being latent from the first. Indeed the great majority of instances we find that a first trauma has left no symptom behind, while a later trauma of the same kind produces a symptom, and yet the latter could not have come into existence without the co-operation of the earlier provoking cause; nor can it be cleared up without taking all the provoking causes into account.
Overdetermination refers to all the provoking causes of an hysterical symptom. There is a hint here already of that Nachtraglichkeit--the activated-after-the-event-ness of the provocation--that Derrida picks up on in “Freud and the Scene of Writing” and which seems rather profoundly to suggest a notion of time not subordinated to the present.
The pattern is as follows: a trauma may have little or no effect at first yet a later trauma of a similar kind provokes a symptom by triggering off the provocation of the earlier trauma as well--a process which is continued repeatedly. It is also the pattern of the repetition compulsion (and is thus indicated by the function of the letter in Lacan’s reading of Poe’s “The Purloined Letter”).
Later in Studies on Hysteria it is Joseph Breuer who first writes the actual word--although he does attribute it to Freud: “Such symptoms are invariably ‘overdetermined,’ to use Freud’s expression.” The word is überdeterminiert. When Freud employs a similar term at around this time it is überbestimmt (emphasizing the multiple causation as provocation). In the Dreambook the notion is pretty much taken for granted--a parenthesis explains to the reader why it is possible to have more than one interpretation of a dream: “The two interpretations are not mutually contradictory, but cover the same ground; they are a good instance of the fact that dreams, like all other psychological structures, regularly have more than one meaning.” The notion of meaning here should be referred to the notion of “provoking cause.” But later he defines it in the famous statement derived from Goethe’s Faust. Analyzing a dream (his own) in which “botanical” is a nodal point (of condensations) he says: “Here we find ourselves in a factory of thought where, as in the Weaver’s masterpiece:
‘ . . . a thousand threads one treadle throws,
Where fly the shuttles hither and thither,
[translation note: herüber hinüber over here, over there]
Unseen the threads are knit together,
And an infinite combination grows.’ (Trans. B. Taylor).”
The factory of thought or the textile is explained thus: “The explanation of this fundamental fact can also be put another way: each of the elements of the dream’s content turns out to have been ‘overdetermined’--to have been represented in the dream-thoughts many times over.” In other words a plural and busy production team overdetermines the textile unconscious, by actively producing, causing and provoking symptoms (like dreams and puns and jokes) ad infinitum.
In Derrida the determination that escapes all determinations is of course what is referred to by the term différance--and it is this that gives all those other determinations their chance. Derrida’s reading of Freud (vigilant against his concepts) finds a language describing the psyche in terms of forces and resistances, and which consistently uses metaphors of retentive writing machines (the mystic writing pad for instance as memory) with inexhaustible receptivity. In other words the phenomenon of overdetermination is an effect of the unconscious as reserve of repetition and a function therefore of the inexistent repeatability of the trace--overdetermining all provocations.
In terms of deconstruction this has interesting implications. Like Freud faced with a multiplicity of dream thoughts, the reader generally is faced with the question of where to begin (the beginning of Derrida’s Glas poses the problem with underestimated clarity). So Geoffrey Bennington, for instance, in “Derridabase”: “The somewhere where you always start is overdetermined (surdetermine) by historical, political, philosophical, and phantasmatic structures that in principle can never be fully controlled or made explicit.” And as if in impossible exemplification, Derrida, at the bottom of the same page:
Consign them here, but why I wonder, confide to the bottom of this book what were my mother’s last more or less intelligible sentences, still alive at the moment I am writing this, but already incapable of memory, in any case of the memory of my name, a name become for her at the very least unpronounceable . . .
But when chance cannot be reduced any further then all the signifying effects of unconscious forces must be seen to presuppose chance, in which case the possibility of accidents is an a priori necessity. Metaphysics must maintain a distinction between necessity and chance in order to ground its laws. No wonder, then, that Derrida’s readings employ the myriad puns, the tangled etymological chains, and the bizarre translation effects that all texts seem to allow. These texts would not be texts unless they did allow them—these accidents of language, as it were, are absolutely necessary possibilities for the text to function as a text at all. The etymological connection between textiles and books would be just a chance connection except that “just chance” is their possibility. And it is on that level that we have no choice but to take our chances with the text.
Each reading commits an unavoidable violence to the text. This violence is curious because if providing the law in order to play a game of which the rules are hidden is violence then it is a violence that replaces the one that the text (violently) hides from its first comer. That is, the law is always violent but you cannot do without it. The law then takes the structure of a supplement. The supplement, a term emerging from Derrida’s famous reading of Jean-Jacques Rousseau in Of Grammatology, replaces what is missing in a text (the law of its composition and the rules of its game) with another structure (another text) which thus also hides the law of its composition and the rules of its game. But a reading that fails to provide rules is no reading at all. That is why Derrida says that unless a decision is made in the face of the undecidable it is not a decision at all. Follow already written rules or rubrics and your decision was made for you in advance.
Some of the problems that people have with Derrida come down to the fact that he fails to provide a definitive authority for their reading practices (what’s his theory? what kind of a method is deconstruction?). The problem has been anticipated by Derrida himself in a section from Of Grammatology called “The exorbitant question of method” where what might be called, after Jorge Luis Borges, the “Babelian reduction” is explicitly thematised (though not in those terms, of course). The problem concerns the relationship between the finite (a text) and the infinite (the structure to which it belongs by virtue of the fact that it is writing). I’m drawn to Borges’ “Library of Babel,” which carefully and beautifully describes an infinite library, for which the “fundamental law” discovered by a “librarian of genius” is as follows: “all the books no matter how diverse they might be, are made up of the same elements: the space, the period, the comma, the twenty two letters of the alphabet. He also alleged the fact which travellers have confirmed: in the vast library there are no two identical books.” Every paragraph in that story seems to illuminate some aspect of our problem. At issue are the function of, and, crucially, the conditions of possibility for, written marks. The mark (say in the form of a word) functions in a strictly contextual way. There is nothing positive about words. There is not even any certainty about their number. The number of words in English is necessarily inexact. Having taken into account the number of words in the OED and its supplements do we count all those words that have fallen out of use or all those words of recent coining that have not yet been collected in any supplement—and how do we count technical vocabularies—do we discount them as other languages or include them? If we cannot count words how do we account for them? One answer would be to consider the median word rank of English texts (the number it takes for the most frequent words of the language to reach a probability of one half). For standard media output the median word rank is 100. For very literate writers it is more like 500. It is worth making this kind of calculation because the difference in active vocabulary is much bigger (10,000 for a newspaper and 100,000 for a canonical writer). Even for a “great” writer 24% of the text will be made up of the top ten ranking words. An interesting counter-example might be a poet like Mina Loy who goes out of her way to use the precision that archaic and technical terms add to the language. Unsurprisingly there is a mathematical law that governs things like median word rank. Take the number of different words in your language and you can calculate the boundary of approximate entropy that governs relative word frequency in any text. Benoit Mandelbrot (the inventor of Fractals) in 1961 showed that a monkey typing at random would produce a language that obeys the same laws. English has an average word length of 5.5 characters (including the space between words) and obeys the same laws as a language with, say, a nine letter alphabet whose words are made up of the digits between 0 and 1 in which the digit 0 never occurs. The median word rank for such a language is a huge 1,895,761 (compared to the 100 - 500 of standard English texts). Mandelbrot shows that the monkey language is, in the terminology of fractal geometry, self-similar and grows on infinite trees (any branch of the tree will be identical to the tree itself), thus needing an infinite dictionary. English, on the other hand, is a massively geared down system that economises on entropy in a number of ways, e.g., the interdependence—or redundancy—of words that seems necessary in order for a text to be “meaningful.” Most letter combinations (an uncountable set) in English are nonwords. There are numerous instances, on the other hand, of incognates where the same letter combination turns up in more than one language with no connection between the two—e.g., “stricken” which means “to knit” in German (to stay with our metaphor) and the German “gift” which is poison (to return us again to the pharmakon text). The famous language poem “Men in Aida” reproduces Homer’s Greek when read out but is written entirely in English words. There are numerous examples. My favourite is “Mots d’heure gousses rames” which looks like an obscure old French poem but when read out provides Mother Goose’s Rhymes in English but in a thickly Gallic accent. Derrida’s notorious différance exploits a similar phenomenon (the French word différer means both to differ and to defer depending on context; Derrida coins the term différance to perform the simultaneous differing and deferring of the term itself). The implication is this: meaning economises on chance (produces it and limits it at the same time). More elegant than mathematics (I spare you the equations) is Derrida’s point in a text called “Mes Chances: A rendezvous with Some Epicurean Stereophonies”:
Language . . . is only one among those systems of “marks” that claim this curious tendency as their property: they “simultaneously” incline toward increasing the reserves of random indetermination as well as the capacity for coding and overcoding or, in other words, for control and self-regulation. Such competition between randomness and the code disrupts the very systematicity of the system while it also, however, regulates the restless, unstable interplay of the system. Whatever its singularity in this respect, the linguistic system of these traces or marks would merely be, it seems to me, just a particular example of the law of destabilisation.
This citation—as so often in Derrida—is drawing attention to the performing aspects of the “law” it describes, as it describes a law that it (t)here now performs. But, given all that, where does this leave the text? One would need to at least be able to imagine an infinite weave—something maybe like the delicate chaos of fractal processes—whereby a motion was somehow to feedback on itself producing a difference in the iteration, which in turn was to feedback on itself producing a difference in the iteration ad infinitum—a stochastic process (determined chaos—a pattern yet unpredictable). Then one would have to consider the general economy as it becomes restricted while the restricted becomes general, ad infinitum. Is it possible?
However if the “Babelian reduction” is used as a rubric
Heading in red letters of part of a book, an authoritative rule, something under which a thing is classed, an explanatory or introductory commentary, an editorial interpolation, a heading of a part of a book or manuscript done or underlined in a colour (as red) different from the rest, an established rule, tradition, or custom]
it is not what it claims to be after all (i.e., infinite) but an “artifice of eternity”--a fiction about the infinite--or as the Borges story suggests, a dream of a mirrored surface that represents and promises the infinite. “Men usually infer from this mirror that the library is not infinite . . . I prefer to dream that its polished surfaces represent and promise the infinite.” It’s not just an opinion or a belief that the infinite as such can only ever be a perhaps or a promise. No one has ever been to or seen the infinite as such. The reduction would be rather just another arbitrary context masquerading as an authoritative rule.
It would thus be what Kant describes as a transcendental hypothesis (about the soul or about the divine author of all things). He says, “A transcendental hypothesis would be no explanation at all, since that which one does not adequately understand on the basis of known empirical principles would be explained by means of something about which one understands nothing at all” (A772/B800). Kant disallows transcendental hypotheses of the speculative use of reason (which misuse the freedom that reason has to make good the lack of physical grounds). This kind of speculation, however, seems to be unavoidable as the example of context shows. What is the context of a text? Arguments either for what is called the context of production or for what is more recently called the context of reception each fail to adequately address the notion of contextuality generally which thus plays an interesting function as transcendental contraband, in keeping with the figure of the hidden law that continues to tie my lemmata together. As we shall shortly go on to explore, writing must be able to be repeated, and therefore cited, as it were, out of context (or it would not be writing), but at the same time there is no writing, no statement and no sign, outside any context.
At this stage of the reading it is notable that the apparently accidental contexts that are being addressed, invoked or suggested actually point to something that might build into a response to the initial demand (that I speak about deconstruction). I want to respect the promise of the infinite that upsets so many readings. What this means is that I must explore every aspect of a text that is not unique to it. This would mean locating every structure that the text belongs in. In so far as all writing inhabits countless contexts, including uncountable future contexts, the task of locating the text’s singularity would be infinite. But we cannot sit down forever with the complete and infinite library before being allowed to say anything. On one hand we might feel justified in reading the complete works of an author, all his or her influences, but even then we have hardly begun. Derrida reminds us of what he calls “an immense series of structures, of historical totalities of all orders” which would need to be accounted for. But on the other hand, we are already in the text. Theory cannot separate itself out from the practice of reading, just as the practice of reading cannot go on innocently, free of theory--free of the rubrics and lemmata that are necessary at the beginning if we are to begin reading (that’s why there’s no outside and no infinite for us).
Is it still necessary to point out that Derrida seldom picks up on just any old oppositions that happen to be unfolding as if collocation and antinomy per se were the secret to existence and the downfall of western metaphysics? Rather, the oppositions that emerge tend to fall around the type of closure according to which the empirical (and the finite plus the object and/or material world) comes to be opposed either to some transcendental beyond, a dream of some unseen, unimagined, infinite, intelligible, eternal “place” that absently explains the empirical one (in which we will all have died before arriving at the infinite) or to some mathematical matrix or structure that can be discovered ordering things from its intelligible hiding place. The point he makes for instance in “The Exorbitant Question of Method” is that because the distinction between empirical and transcendental is itself a historical institution one could not explain it either through empiricist means or through a transcendental conceptual order that separates itself off as if it was a rubric. Thus his attempt (under the term deconstruction) is “exorbitant” because it both exceeds and delimits the closure between transcendental/empirical. It is as exorbitant in this special sense that he justifies his singular route (trail, trace, path, cut) through the margins of Rousseau who--according to Derrida’s response--explains in writing what writing is. Because writing has an apparently empirical--and sensible--element and an unseen, hidden, unstable, often randomly determined intelligible aspect (which no context can exhaust for that reason), the exorbitant is the only justifiable route. The texts perform this argument rather than just assert it (such an assertion would be no more than another dream of the surface that represents or promises the infinite).
The following short citation is a well-known
passage from the paper I have already cited (lemma 8) that
Derrida was asked to present for a forum on psychology and the humanities,
under the rubric “Literature and Psychoanalysis.” Derrida decided to focus on
the theme of chance in his contribution, now published as the text, “Mes Chances.” Instead of speaking of the said relation, he spoke about
chance—not, supposedly, a relation.
Here is a section on what he describes as three elements, the proper
name, the number, and the letter, which are linked in so far as they are the elements
that Freud’s texts revolve around whenever they deal with the question of
chance. “One could initially ask oneself
what these texts have in common—these stoikheia that include the letter or the trait, the number,
and the proper name—such that they are to be found thus associated in the same
series and such that their relation to chance would be analogous.” He then points out that what they have in
common is their “insignifiance marquante”—something
like a marked insignificance—the mark of insignificance—but only in so far as
the mark is marked by its own insignificance.
What he means is a mark that marks the fact that marks have no natural
or necessary reference or meaning in themselves. Then the example of the proper name
The French name
Symptoms, of course, are what Freud reads accidents as—like “slips” of the tongue—but even the unconscious (if there is any) must capitalise on the chances that make “slips” possible, i.e., there must be the possibility of an actual “pure” accident in order for the unconscious to express itself in that way.
So what puts the multiplicity of
This will be equally evident in the relation between a numeral and a number and also in that between a number and a thing numbered. Between the meaning of the number 7 and the numerals 7 (Arabic or Roman numerals, the words sept, seven, seiben) there is no natural, necessary, or intrinsic affiliation. No natural bond, to use Saussurian terminology, between the signified and the signifier. Nor is there a natural bond between the signified (the general meaning of 7, the number 7) and all things (stones, horses, apples, stars or souls, men or women, for instance) that could be found together in groups of 7.
Here the function of the mark in each case occurs on the possibility of the pure accident—this is one reason why Derrida puns. The pun is itself the mark of a necessity—how irritated philosophers and literary critics get with this! So it is easy to demonstrate—Derrida is a permanent demonstration of it—that this possibility (the necessary and a priori possibility of the purely accidental mark) is necessary for all marks—and it is here that the function of iterability appears. A mark is utterly accidental but powerfully constrained by its iterability—its capability of being identified—re-marked—from one context to another in a (perhaps) infinite series. So the complexity in Derrida can be grasped in terms of this relationship between the necessarily accidental “insignificance” of the mark and its iterability. Even the proper (whether in philosophy or literary criticism—to mention two idioms that sometimes treasure the notion) has its place only thanks to the necessary randomness of the process. The thing about the “Mes Chances” essay is that it simultaneously demonstrates what it says—so the explanation is as complex as the process.
Earlier on during his paper Derrida makes the following point (about a context long since passed away):
Even here, among us, the effects of chance are at once multiplied and limited (that is, relatively tempered or neutralized) by the multiplicity of languages and codes that, while they are engaged in intense translational activity, overlap at each instant. Such activity transforms not only words, a lexicon or a syntax (for example, from French into English) but also non-linguistic marks, mobilizing thus the near totality of the present context and even that which might already exceed it. The text that I am now reading should be publishable; I was aware of this when writing this summer. It is destined in advance to addressees (destinaires) who are not easily determinable or who, as far as any possible calculation is concerned, in any case command a great reserve of indetermination. And this in turn involves, as I shall try to show, the most general structure of the mark. Running my chances over your heads, I therefore address myself to addressees unknown by you or me. But while waiting and in passing, this falls, as the French saying goes, upon you. (Mes Chances)
If we were to really take seriously the possibility that Derrida might be right about “the great reserve of indetermination” that addressees who are “destined” by a text command, then his own discourse serves as an example of the point he is making. What part of the argument does the point support? It supports an argument about a system of marks that have as their property a principle that disrupts the systematicity of the system. Instability and stability are both functions--simultaneously--of such marks. This property is as follows: “they simultaneously incline towards increasing the reserves of random indetermination as well as the capacity for coding and overcoding or, in other words, for control and self-regulation.” What makes it possible for a text written 2000 years ago by somebody, to be read today by someone the writer did not know and for it to be read in 200 years time by another addressee who has also not been imagined and in a language that does not yet exist? Derrida calls it the law of destabilization. If one was to incline towards the regularity of context and code there would be a certain point at which the text in question becomes so tightly bounded that its possibilities for signifying again elsewhere or at other times would be seriously impeded. If on the other hand one inclined in the other direction, towards indetermination--towards the randomness represented by series of coin-tosses (or any other yet more sophisticated attempt to arrive at random sequences)--the text would simply fall like Epicurean raindrops into the abyss (that is, it would be nonsense). But, against this dialectic, Derrida does not seem to be suggesting that we need a little of both--a little indetermination as well as a little code (as if we’d have the choice). Rather the system of this kind of mark (whether it is language or not) involves the simultaneous increase of the reserves of random indetermination and the increase of the capacity for coding and overcoding. The type of system that he is talking about involves simultaneous randomness and code. The effects of chance are not just economized on, limited, but they are at the same time and according to the same property multiplied. The possibility of the situation I mentioned above (an arbitrary 2200 years of textual transmission) is also the possibility of two people face to face exchanging a few meaningful words. People have asserted that rather it is the opposite way around, that face-to-face conversation was probably how language began and is thus (although this would anyway by no means follow) the very principle of linguistic exchange, with writing something that people developed afterwards to “extend” the principle of linguistic communication into spatial and temporal distances. The randomness/code argument puts an absolute distance in place (paradoxically) to begin with as the very possibility of the experience of presence that face-to-face linguistic exchange (and all other face-to-face experiences) seem to be.
The possibility of absence (e.g., the absence of the addresser, the addressee, the referent) is a necessary one for there to be writing and it is this condition of possibility that I want, a little later, to bring to the very specific context of deconstruction and, thus, the deconstruction of context. Let’s get this straight. The absence need not and very often does not apply. There are often addressees, addressers and referents all at the same time (I ask a student to move a chair). However it is the possibility that none of the above be present that is at issue. The claim would be this: it is a necessary possibility. The necessary and permanent possibility of this absence can be contextualised with the problem of alterity. That is, it is the problem of the other.
One more quotation:
To be a mark and to mark its marking effect, a mark must be capable of being identified, recognized as the same, being precisely re-markable from one context to another. It must be capable of being repeated, re-marked in its essential trait as the same. This accounts for the apparent solidity of its structure, of its type, its stereotypy. […] But more precisely it is not simple since the identity of a mark is also its difference and its differential relation, varying each time according to context, to the network of other marks. The ideal iterability that forms the structure of all marks is that which undoubtedly allows them to be released from any context, to be freed from all determined bonds to its origin, its meaning, or to its referent, to emigrate in order to play elsewhere, in whole or in part, another role.
On this explanation, which in an important sense is not itself deconstructible, and as writing it performs the necessary conditions for its own existence, it is possible to show that there is nothing beyond writing as such. For this reason, Derrida’s replacement of all transcendental contraband with the term writing provides a kind of ground. It is not just a replacement. It is also a repetition. Transcendental contraband, e.g. in the form of a thought or a concept that explains things but that cannot itself be an object of thought, is also always writing. But it is never decidable whether it is anything more than just writing, once writing is thought of in terms of its conditions of possibility (the repeatability and differentiality of the written mark). In other words deconstruction can show that the conditions of possibility for knowledge render the grounds of knowledge (at least) fragile. All systems of knowledge, all structures of belief, all legislative institutions are grounded on contraband and are thus deconstructible. However deconstruction is not itself deconstructible. If deconstruction is equivalent in some sense (which it is) to the repeatability of the written mark we cannot say that we have access to that repeatability but we can say that its repeatability is grounded in repetition. In principle a mark must be repeatable infinitely. But the only access we have to it is in its finite repetition. There is no context that can exhaust it yet we cannot escape the contexts into which it always falls. At least one of those contexts is the futural “to come” that never comes. That is, all experience contains an absence that we refer to as the future. This context remains, for powerful structural reasons that I have just outlined, absolutely incalculable, unpredictable. There is an absolute surprise inherent in the futurality of the written mark as there is for all experience whatsoever. That does not mean that calculations and predictions are not possible. On the contrary the incalculable is the very ground of all calculation, given that the repeatability of the written mark from the very beginning cushions the surprise of an event. That is why experience feels like a mixture of surprise and regulation. Nothing is absolutely regulated (except in the ideal and thus contraband representations of abstract science) and nothing is absolutely surprising. Repeatability is both at one and the same time the absolute alterity of the other (in its futural mode) and the acknowledgment in advance of the other’s identity (in the mode of its historicity). The infinite deferral of absolute identity is brought down to earth in its finite repetition. This repeatability is what makes prediction and calculation possible but absolute calculability for that very reason is impossible.
If writing were subordinated to the function of representing the sign then the transcendental contraband in that representation would be something like the signified (meaning). But what if there isn’t any signified? If we provisionally agree on the existence of something called a sign (we don’t have to but bear with me) and follow Ferdinand de Saussure’s division of it into signifier/signified we have a sensible part and a non-sensible part, conforming to the most traditional notions but focusing the distinction between empirical and transcendental on one observable unit--the sign. One thing that Saussure’s linguistics insists on is generality so we cannot use this notion of the sign, in the sense that it sometimes has, to mean the sign of a thing (some signs don’t refer to things). But the sign is always, definitively, the sign of something other even if the other is just another sign. (A parody of a deconstructive argument might follow from the question what is the sign “sign” a sign of? The sign is a sign of a sign of a sign . . .). So Saussure finds it convenient to isolate a signified that is not reducible to a referent as such. Saussure’s teaching is consistent too, then, with the sense and reference arguments of Gottlob Frege and Bertrand Russell--especially in the latter’s deliberate bracketing out of the actual world of objects, not to say that there is no “real world,” but on the contrary to say that language is an entirely separate system from the real world. In this respect Russell’s attempt to develop a logical language was the attempt to find a language adequate to “the actual” (his word). Anyway, Saussure’s system of differences with no positive terms, which is a dogma that many people impute to deconstruction, finds no support in his exposition until he turns--in a remarkable about-turn--to writing. Now writing (in the logic that we have provisionally accepted) is a kind of a sign whose job it is to represent spoken language. It is a sign of a sign of a signified (if there is any). Given that Derrida tends to like his seconds first, writing is going to be extra special--a kind of a third (which we know will go not first but before the first--but let’s not jump ahead). Saussure had banned it from his linguistics until then (along with the world of referents). Speech works in the way that writing works--it is the sign of a sign (in the logic we have provisionally accepted but which is now looking overworked and in need of a rest--but we won’t give it one!). Well that’s easy then, we know how writing works--it is, after all, everywhere. I could put it like this. There are certain conditions that must necessarily be able to obtain for there to be writing as follows:
1) A written mark must be able to be repeated in (perhaps) infinitely many contexts (copies will always do; print and even handwriting differences are unlikely to have any effect on the identity of the writing itself). It must be repeatable--able to be repeated--that’s what gives us the distancing effect that is always associated with writing.
2) Following from the above writing implies the possible absence, in principle and often in fact, of addresser, addressee, and referent.
3) Writing always might function as part of a system (like Saussure’s La Langue) but no single system can constrain it because of point (1). This brings an essentially arbitrary and undetermined element into the written mark--which is thus always open to future significations, which have not yet been imagined. And there is an anxiety-inducing unpredictability associated with it. Derrida compares it with a postal system in which letters always might not arrive when and where they are supposed or desired to. (Sorry Jacques Lacan).
No wonder the tradition doesn’t always like writing.
Archaeologists go to enormous lengths, requiring the most admirable patience, in unearthing from buried depths the charred remains of ancient pieces of papyrus. It seems that only burned writing survives the centuries--the burning somehow “preserving” as a faint and ghostly pallor the inscriptions which, having been left in a more pristine state, would almost certainly have been dissolved into dust. Only after weeks of careful reconstruction can they tell what the text is. It often turns out to be a copy of something they already have, that someone somewhere or some institution thought was worth preserving while someone else carelessly or deliberately allowed to be consigned to the flames. If they haven’t, on the other hand, got a copy of it already, then it often turns out to be relatively uninteresting to them (no lost sections of Aristotle’s Poetics forthcoming yet), and its incineration thus seems a little more explicable.
However, when it comes to discussion of speech it turns out that the conditions that make writing possible must also always be in place for speech to be possible too (alas!). Speech is associated by the tradition with signified meaning, intention (independent of language) and the presence of the world (independent of language). But speech too is only possible on the conditions it shares with writing: distance, absence and the possibilities of mistakes, accidents and misuse. But here is the best bit: there is then no signified. No wonder it’s always been so damnably difficult to define. With no signified there is no signifier (it’s out of job I’m afraid--unless you’re Jacques Lacan, for whom the signifier just goes around being its own signified). What this implies is that without writing and its conditions of possibility there would be no possible experience of the things that signs are generally used to refer to. Writing is the basis of experience.
Just as an aside, and while we’re on the subject, other phenomena that are possible because of writing are: metaphor, translation, analogy, allegory (and we may find) institutions, laws, legislations, economics, exchange, money, and that phenomenon we still call literature.
If we were to use the Kantian meaning of the word transcendental then transcendental knowledge “is occupied with the mode of our knowledge of objects in so far as this mode of knowledge is to be possible a priori.” In that case a lot of what we’ve been reading Derrida doing with the mark can be classed as developing a transcendental knowledge in the Kantian sense. However, Kant develops the meaning of the word transcendental by consistently contrasting it with a very interesting series of contraries. Empirical knowledge first is knowledge of objects in so far as the intuition of such objects “contain sensation” and thus presuppose the presence of said objects to the subject’s gaze (or whatever sensation). There can be no empirical knowledge of a pure concept. Psychology is also contrasted to transcendental knowledge--and would be empirical as far as Kant is concerned. We should also note Kant’s distinction between the sensible and the intelligible. The distinction from the point of view of the transcendental would recognize sensible objects and the sensible world and the connecting of appearances by efficient causes as all belonging to an empirical knowledge governed by understanding yet subject to a transcendental “law of causality” that is strictly transcendental in so far is it can never be an object of thought. There are no intelligible objects and there is no intelligible world for Kant (as against a certain Platonism of course) but a transcendental use of the concept “intelligible” is necessary to account for a ground that can never be the object of any kind of knowledge. The logical distinction between intelligible and sensible is also a contrary of the transcendental because logic is only concerned with whether the form of the distinction is clear or confused. Finally, metaphysical knowledge too, whose concepts are dogmatic, is contrary to the transcendental.
But Derrida could not quite be a transcendental philosopher, by virtue of the iterable mark, which is not a transcendental concept in the precise Kantian sense but is not reducible to empirical, psychological, logical or metaphysical knowledge either. We would have to refute everything we have pretty much established regarding the mark in order to establish the insistence that Derrida is a transcendental philosopher. And so deconstruction, if it is a form of knowledge, must be considered as quasi-transcendental.
The quasi-transcendental adds a supplement to transcendental claims. If the transcendental indicates that which lies beyond experience but which is nonetheless necessary for experience then Thomas Aquinas’ doctrine on analogy is pertinent in relation to what we have so far established concerning the written mark. He says that because a univocal sign for God is impossible (mainly because there are too many pagans who use the same sign) the Catholic must rescue it from the equivocality that infects all finite language, by using it analogically. Analogy lies somewhere between univocity and equivocality. A univocal sign, for instance, would mean what it meant, mean it once and then fall out of use. It would just mean this one thing and that is not how signs work (as both Frege and Russell eloquently show). An equivocal one, on the other hand, means just anything depending on the context and the users. Such signs would obviously not be much use. Derrida’s working out of the repeatability and differentiality of the written mark, under the term iterability (from the Latin “iter” = likewise) implies both a repetition (an identity) and a difference (a non-identity). The condition of lying somewhere between, as writing and différance among so many other things in specific contexts (including deconstruction and justice), informs a great deal of what goes on, in very different ways and contexts, under the head-word (rubric) post-. The name of God names an absence that is never in fact present to experience. The tradition has always taken absence as a prolongation of the absence of some thing, some state of affairs, some entity or existent that was once fully present to perception. In other words absence is subordinated to the idea of presence.
The subtitle to Rene Descartes’ third Meditation reads as follows: “On the Essence of Material Things; And Likewise [iterum, derechef] of God, That He Exists.” Derrida’s subtitle to the third part of his own reading of J.L. Austin in “Signature, Event, Context,” reads as an ironic repetition: “Parasites. Iter, Of Writing: That It Perhaps Does Not Exist.” Parasites, in this context, refer to the kinds of speech act that J.L. Austin considers “infelicitous.” Speech acts like naming a ship and performing a marriage ceremony can be acted out (e.g., in a theatrical performance) and that makes them parasitic on the proper occasions. But it is the necessary repeatability of the graphematic mark that makes all such speech acts possible (whether felicitous or infelicitous) so the parasite tag is revealed to be inappropriate unless all utterances are regarded as parasitic (on their general repeatability).
Derrida distinguishes between what seems to be an absent present (for the tradition) and an absent absence (in his reinscription). It can be reduced to the formula (for example): The absence of God is represented by virtue of the iterability that writing is. The absence of writing is represented by the iterability that writing is. In both cases the addition of the supplemental “perhaps not” renders the question of existence or nonexistence undecidable. But only in the second case is that undecidability grounded. There is a very complex analogy that could be schematically outlined. It concerns the existence/essence, material things/God problematic in Descartes’ Meditations on the one hand and the iterum (likeness and repetition)/writing problematic on the other. In so far as the “Iterum” already plays a leading part (as analogy) in the former then the analogical form of its repetition in the rubrics of Derrida’s subtitle is interesting.
The traditional opposition between presence and absence relates absence in the form of a continuity, continuous absence, to an origin at a simple present. Writing would thus be that which represents. The absence derived from Derrida’s response to the tradition would precede the presencing of any present and would thus not be derivative of any present. This kind of absence would be an anomaly in the teaching of the tradition. With regard to time, as we know, it suggests a past--a sense of pastness--that would never have been present and, a commonplace notion, a future that will never be present. These absences can be quite systematically added to the chain that includes the nonexistent, the insignificant and something else that I would like to make more distinct. Iterability would seem to be something of a divine name. It would be worth examining Derrida’s claim that any distinction between presence and absence--even one that would abolish presence completely--presupposes the authority of a discourse capable of distinguishing a simple presence and a simple absence, and capable of distinguishing between existence and non-existence, against a possible writing that lies outside the province of that authority (Derrida’s claim for graphematics). The iterable mark, as Derrida claims, is neither present nor absent. What does he mean?
The name, within ontotheological discourse, for what cannot be known but which makes all things possible or impossible, is God. Hence Derrida’s (analogical) subtitle, “Parasites. Iter, Of Writing: That It Perhaps Does Not Exist,” which he discusses at length in “Limited Inc a b c . . .” Here he makes clear that his argument cannot be distinguished--separated out--from the effects of the graphematic that he simultaneously mobilizes; and these are quite precise (quite). Can we make it more explicit? The Cartesian “Iterum” (“et derechef,” and likewise) relates “the essence of matter” to God in Descartes’ own subtitle in the Meditations: “On the Essence of Material Things; And Likewise [iterum, derechef] of God, That He Exists.” So the subtitle--Derrida is dead serious about this--is itself an iteration as well as a rubric--an iteration of a rubric concerning the proof of God--a proof that is itself an iteration of a previous proof. What kind of iteration would this be (in the typology of forms of iteration that Derrida suggests)? “More or less cryptic (‘perhaps’ cryptic), more or less parodic, ironic, altered, lateral and literal, but at the same time very serious (as serious as the question concerning the proof of God).” The iter (likewise) is a citation.
The absence: The rubric “Iterum de Deo, quod existat” is repeated (likewise) regarding writing (governed by a graphematics of iterable marks summed up in the shorthand of the “iterum”) with the “ironic” supplement of the “perhaps not” which was always anyway the fear or the suspicion (see Browning’s “Fears and Scruples”--a favorite of J.L. Borges) regarding the kind of presence presupposed by the iterability of its representations and (proper) name. The precise sense, then, of absolute absence is already a predicate of the absence of the tradition. Derrida, writing of “Signature, Event, Context,” states: “sec names writing in this place where the iterability of the proof (of God’s existence) produces writing, drawing the name of God (of the infinite Being) into a graphematic derive (a drifting) that excludes (for instance) any decision as to whether God is any more than the name of God.” “Perhaps not” is undecidable. That doesn’t mean that writing does not exist. It means that the question of the existence (or non-existence) of writing is rendered undecidable by virtue of iterability (graphematics). The absolute absence that is proper to writing is not, then, the simple absence of writing. Where God’s presence is grounded only in the iterability of the proper name (signaling only his absence), writing too--which just is iterability--is grounded only in the iterability which writing already just is.
Iterability, or repeatability and differentiality combined, implies an unavoidable dislocation at the very core of all experience. Repeatability at the origin, originary repetition, is a paradox. This “internal dislocation” is described by Jacques Derrida in a short section of his seminal Of Grammatology, which outlines the character and ground of the Western fantasy of the East. Derrida links his analysis of the ethnocentrism that has always controlled the concepts of writing and of science in the history of metaphysics to the temporal dimension of the future, which, he says, “can only be anticipated in the form of an absolute danger” (5). It is what is intrinsically incalculable about the future that throws the whole structure of temporality (in metaphysics time is always oddly structural) out of “synch” (as it were). It renders experience (and all knowledge and theory) intrinsically incomplete yet open to infinite and random undetermined contingencies called, for convenience, “the future.” This disorienting temporal dimension opens all writing, all knowledge, science and technics to the monstrosity of its other in absolute alterity. That is, the future is not simply the dimension in which knowledge might ultimately be complete (the utopian aspect of modernity). Rather it is, as an integral component to experience per se, that which cannot ever be known. Knowledge is thus essentially incomplete. Modernity tends to respond to the experience of incompleteness with a characteristic ethnocentrism. For instance, Descartes, Leibniz and others saw Chinese writing as the answer to the problem of a universal language by which philosophy might complete itself. “The concept of a Chinese writing,” writes Derrida, “functions as a sort of European hallucination,” an occultation that, “far from proceeding from ethnocentric scorn, takes the form of an hyperbolical admiration” (80). The concept of the sign and the problem of writing are what bring to light the central aporia of western thought, the inability to overcome the difference between what is empirical (therefore lacking) and what is transcendental (a mysterious beyond). The attempt to reduce this difference is the effort to maintain and fulfil an interior, the sense of a complete identity, and can thus be understood as an attempt to domesticate the alterity of the other, whose appearance is strictly and necessarily always to come in a monstrous, incalculable future.