publisher colophon
  • Deconstruction: A Misprision of Saussure and Charles Sanders Peirce

Derrida’s creative misreading of Saussure and Peirce underpins the radical scepticism of deconstruction, an ideological movement that arose from the disillusionment of French Marxist intellectuals with Soviet communism in the late 1950s. They moved (via Saussurean linguistics) from Sartre’s and Merleau-Ponty’s frontal attack on liberal capitalism to an outflanking movement, belatedly targeting scientific positivism seen as the philosophical underpinning of liberal capitalism; “belatedly” because the language philosophy of Wittgenstein, Austin, and Searle had already displaced Frege and Russell. A parallel disillusionment in America, prompted by the morass of Vietnam, created a welcoming literary culture for the anti-essentialist posture of structuralism and its heir, deconstruction.

Poetic influence—when it involves two strong, authentic poets—always proceeds by a misreading of the prior poet, an act of creative correction that is actually, and necessarily, a misinterpretation. The history of fruitful poetic influence, which is to say the main tradition of Western poetry since the Renaissance, is a history of anxiety, and self-saving caricature, of distortion, of perverse, wilful revisionism without which modern poetry as such could not exist.1

Jacques Derrida is a philosopher, not a poet, but his co-optation of the semiotic theories of Ferdinand de Saussure and Charles Sanders Peirce is a clear case of Bloomian misprision, as I hope to demonstrate in the following pages. I shall argue that Derridean deconstruction rests on a double misprision in that he “creatively” misunderstands [End Page 411] both Saussurean linguistic theory and Peircean semiotics, erecting a pseudo-epistemological theory on that misreading. I contrast Peircean pragmatic semiotics to Derrida’s sceptical deconstruction, as well as the similar skeptical relativism of Stanley Fish and Richard Rorty that together constitute literary postmodernism. I argue that Derrida’s view that language isolates us from the world (the “outside of language”), as opposed to “carving nature at her joints,” as Plato thought, places him in a school of linguistic thought culminating in the Whorf-Sapir hypothesis—of which he appears to be ignorant. Finally, I cite Derrida himself on the ideological motivation of deconstruction.


Let me begin by summarizing Derrida’s “story” of how the letter, the graphé (grapheme), liberated mankind from the tyranny of the spoken word. No one would dispute that the invention of alphabetic writing was a crucial cognitive leap in human history, but Derrida’s focus on the arbitrariness of the letter, as an instantiation of semiology as a science, hardly bears scrutiny. After all, phonemes are equally arbitrary. Indeed, like the letter, the phoneme is itself meaningless. Meaning resides only in the word, the lexeme, as embedded in a lexicon and in a sentence, that is, in a grammatical nexus. Of course, the phoneme is more pragmatically engaged with the speaker than the letter is with the writer—of which more later—and Derrida’s view, that the word was indivisible before writing, is plausible. Linguists distinguish between isolating languages, like Mandarin, which has no polysyllabic words and no declensions, and agglutinative or polysynthetic languages, which can have sentence-long words. Alphabetic writing plausibly would tend to move a language away from an agglutinative or synthetic form, toward an isolating form. Mandarin is a counterexample to this hypothesis. Although it has not been alphabetized, it is an extreme example of an isolating language in that it has no polysyllabic words.

Nonetheless, Derrida’s view that the phoneme (the meaningless component of the spoken word) may have been inconceivable without alphabetic writing is plausible. Other forms of writing, such as Egyptian hieroglyphics or Chinese ideographs, do not encode language so much as the conceptual framework that spoken language also encodes—a property, incidentally, that undermines Derrida’s grammatological thesis that there is no “outside of language.” A phonetic component is present in both hieroglyphic and ideographic writing systems, but the phonetic [End Page 412] component is peripheral compared to the ideographic component. Egyptian hieroglyphics retain an unmistakable iconic component, almost completely obliterated in modern Chinese characters. Of known writing systems, only the phonetic alphabet encodes language—as opposed to concepts. Grammatology must obscure the fact that languages encode concepts for Derrida’s argument to go through.2 He deploys Saussure’s semiology to achieve that bit of prestidigitation.

Derrida’s contention that only after the advent of alphabetic writing did the word (the verbum or logos) become analyzable into its meaningless components (its phonetic parts) is persuasive. For Derrida, the moment of discovery or invention of the graphemethat is, the meaningless graphic mark of pure discrimination—was the moment of discovery of the sign proper. For him the spoken word, the verbum, is not a true sign. Of course, the spoken word, like the phoneme, is not attached to its referent by any resemblance or causal nexus, any more than is the written word. The key difference for Derrida between the spoken and the written word is that the verbum is intimately attached to the speaker, while the letter is weakly attached to the writer. He deploys the notion of the “instituted trace” to characterize the relation between writer and written word, and between written word and sense: “Even before it is linked to incision, engraving, drawing, or the letter, to a signifier referring in general to a signifier signified by it, the concept of the graphie [sic] implies the framework of the instituted trace, as the possibility common to all system of signification.”3

Tellingly, Derrida cautions his reader that his notion of the trace “is not theological, as one might believe somewhat hastily. The ‘theological’ is a determined moment in the total movement of the trace.” His introduction of theology here—only to dismiss it—is rather odd. It seems he is anxious to avoid an idealistic or spiritual reading of his earlier remark: “The trace must be thought before the entity” (OG, p. 47). It is difficult not to read Derrida’s remarks here as an attempt to mystify the straightforward relationship of representation, which John Searle defines perspicuously: “Every institutional fact is underlain by a (system of) rule(s) of the form ‘x counts as y in context c.’”4 Derrida’s mystifying “framework of the instituted trace” becomes in Searle a system of rules “of the form ‘x counts as y in context c.’” The last phrase (“in context c”) allows for the role of grammar (the rules of combination), which is tellingly absent from Derrida’s understanding of language. And, of course, it also allows for “pragmatics,” that is, the relevance of the context of the utterance as a speech act—of which, more later. [End Page 413]

By introducing the notion of trace, Derrida surreptitiously denigrates spoken language. A spoken word cannot be thought of as a trace of anything. It is a troubling of the atmosphere, an evanescent will-of-the-wisp, the evidence of an act (of speech), and a guarantee of presence. All these features of speech are to be denigrated, obscured, or obliterated by the doctrines of grammatology, which treat linguistic phenomena as natural occurrences rather than intentional objects.

Derrida’s impoverished story of language function depends on the Saussure-Jakobson definition of the sign as purely an arbitrary and dyadic discriminate, and the tacit assumption that individual words are themselves meaningless, just as phonemes are. Meaning on this view arises only from the meaningless word being embedded in a lexicon of discriminates, i.e., words. On Derrida’s view a lexical entry—say “mother”—has sense only as a discriminate of an indeterminate list of other lexical entries, which would include “father,” daughter,” and “virgin”; as well as “crocodile,” “volcano,” and “automobile.” On this view, there is no need to appeal to an “outside of language” to determine the sense of a given linguistic string. To derive the sense of a sentence, and ultimately of a discourse, the reader or hearer need only survey his or her entire personal lexicon to select the sense of the string of words before her. The role played by grammar is left unexamined—not to speak of pragmatic components, such as whether the locution is a command, request, inquiry, or reference. Put in this way, it is obvious that no sense could ever be derived from even a rather simple sentence such as “The boy caught a ball.” (For a simple example of the necessity of grammar, one need only consider the rule in English that modifiers precede substantives. Without that rule, one could not discriminate between “race horse” and “horse race.” No Jakobsonian survey of discriminates of “race” and of “horse” would yield the clearly distinct meanings that the simple grammatical rule yields.)

By oversimplifying language in this way, Derrida evades the Frege-Russell distinction between sense and meaning. Bertrand Russell’s well-known exemplification of the distinction is this sentence: “The present King of France is bald.” Any English speaker will immediately understand the sense of the sentence, but it is meaningless when applied to post-revolutionary France, which has no king, because there is no correct reference to an “outside of language” in the sentence. It is no exaggeration to say that Derrida’s grammatology is a (mostly tacit) assault on the Frege-Russell referential theory of language meaning—at a time when logical positivism was already passé! [End Page 414]

Of course, no one disputes Derrida’s insistence that the word—whether voiced or traced—is arbitrary. But the notion that it is purely a dyadic discriminate is revolutionary and disputable. Saussure tells us that the sign is a dyad made up of a sensible “signifier” (either a grapheme or a phoneme) and a “signified” (a lexical item, aka “word,” rather than a thing, quality, or event, that is, the “signified” is neither a reference nor a Fregean meaning). As a linguist, Saussure was not much concerned with any link between the lexical item (the word) and the world of things, qualities, and events. So far as language is concerned, the sense (Saussure preferred the term “value”) of a lexical item arises from its difference from all other lexical items. Saussure tells us that “linguistic signals are not in essence phonetic. They are not physical in any way. They are constituted solely by differences which distinguish one such sound pattern from another.”5 In other words, languages are not notational systems providing taxonomies of concepts, things, qualities, and phenomena but are dances of discriminates hermetically sealed off from what Derrida likes to call the “transcendental signified.” (However, Derrida must ignore Saussure’s contention that linguistic signs “are not physical in any way,” since his grammè is physical.)

In any case, Saussure does not rest with this posture. He qualifies it on the next page to include at least some version of a “transcendental signified”: “A linguistic system is a series of phonetic differences matched with a series of conceptual differences. But this matching of a certain number of auditory signals and a similar number of items carved out from the mass of thought gives rise to a system of values. It is this system which provides the operative bond between phonic and mental elements” (CGL, p. 118; emphasis added). I have not seen noted anywhere that this modification changes the dyad of signifier (phoneme or grapheme) and signified (word/lexical element) into a triad similar to that on which Peircean semiotics rests. Saussure’s assertion that “the system of values”—the language—“provides the operative bond between phonic and mental elements” renders language a triadic notational system consisting of signifier, signified, and concept (the mental element). Saussure himself seems not to have been aware of this shift in his theory, for he immediately returns to a dyadic model of linguistic value: “For the essential function of a language as an institution is precisely to maintain these series of differences in parallel” (p. 119). However, since what we have as the Course in General Linguistics is cobbled together from students’ notes and some “old jottings” that his editors found “in his [End Page 415] desk drawers” (p. xvii), it would not be fair to accuse Saussure of not fully understanding his own theory.

Saussure’s full view of language as the triad—signifier/signified/concept—does not support Derrida’s grammatology insofar as the latter is founded on the assertion that “there is no outside of language” (“il n’y a pas de hors-texte”) (OG, p. 158). Clearly, concepts are “outside of language.” The claim that language is purely dyadic breaches the link between sign and meaning (as opposed to significance or sense) that all other theories of language maintain. Derrida’s target, as noted, is the Frege-Russell positive theory of language meaning as “definite reference.” Nor does Saussure support Derrida’s grammatology insofar as the latter rests on the physical trace of the written word.

Derrida’s polemic loses some of its bite if one recognizes that the positivist views of Russell and Frege had ceased to dominate theory of language more than a decade before Of grammatology’s publication in 1967. More than a decade earlier, Ludwig Wittgenstein’s Philosophical Investigations (1953) and J. L. Austin’s lecture “How to Do Things with Words” (1955) had called into question the Frege-Russell view that language was purely propositional. Both Wittgenstein and Austin drew attention to the fact that language use predominantly consists of “speech acts,” that is, acts of communication, as opposed to pure acts of reference. In Philosophical Investigations Wittgenstein stressed that we play many “language games,” reference being only one of them.6 Philosophers play a different game than carpenters, poets play a different game than jurists, and so forth. Austin independently developed a taxonomy of three broad varieties of language use (unrelated to occupations, unlike Wittgenstein’s taxonomy): standard “locutions” (propositions), “illocutions” (commands, queries, etc.), and “perlocutions” (persuasion, convincing, etc.). His emphasis on ordinary language use—predominantly speech—runs diametrically counter to Derrida’s privileging of the letter over the phoneme, the bedrock of his grammatology. After Wittgenstein and Austin7 no one could suppose that language meaning was exclusively concerned with references and propositions, which, after all, make up a tiny portion of ordinary language behavior, most of which is verbal. Derrida’s attack on logical positivism in Of grammatology, then, was a day too late as well as a dollar short. [End Page 416]


Intellectuals in postwar France, as exemplified by Jean-Paul Sartre, Maurice Merleau-Ponty, Michel Foucault, Jacques Ellul, and Derrida, among others, were frozen in a posture of opposition to liberal capitalism (and its successful exploitation of empirical science), and to liberal capitalism’s handmaid, philosophical positivism. They seized on Saussure in the 1950s as a refuge from Marxist dialectical materialism, which by then had been discredited by the excesses and failures of Soviet Communism. Derrida himself refers to the disillusionment of French intellectuals with Marxism as the cradle of deconstruction: “It was . . . what we had known . . . concerning totalitarian terror in all the Eastern countries, all the socio-economic disasters of Soviet bureaucracy, the Stalinism of the past and the neo-Stalinism in process (roughly speaking, from the Moscow trials to the repression in Hungary). Such was no doubt the element in which what is called deconstruction developed—and one can understand nothing of this period of deconstruction, notably in France, unless one takes this historical entanglement into account.”8

The adoption of Saussurean structuralism permitted French intellectuals to shift their polemical target from liberal capitalism to empirical science, understood as the philosophical underpinning of liberal capitalism.9 The gist of structuralism was to move away from a concern with the properties of things and events, to a concern with their relationships, the structures in which they were embedded. These anti-essentialist moves permitted French Marxist intellectuals to evade the ugliness of Soviet Communism as well as the failure of Marxism to achieve any significant political power in Western Europe. Stuck in an intellectual economy of opposition to domestic capitalism, they found that debunking empiricism’s claims to truth could serve as a more palatable alternative to the promotion of the dictatorship of the proletariat.

Their German Marxist contemporaries—Max Horkheimer, Theodor Adorno, and Walter Benjamin—had long before adopted a strategy called “critical theory.” Writing in 1937 (with Hitler, Mussolini, and Franco in mind), they found that “the whole weight of the existing state of affairs is pushing mankind towards the surrender of all culture and relapse into darkest barbarism.” Horkheimer and Adorno articulated a “critical” theory to replace the “positive” theories represented by the positivistic, “bourgeois” physical and social sciences (as well as by traditional Marxism). They declared, “In a historical period like the present true theory is more critical than affirmative, just as the society that [End Page 417] corresponds to it cannot be called ‘productive.’ The future of humanity depends on the existence today of the critical attitude.”10

Like deconstruction, critical theory represents a move to an aporetic or negative discourse (calling into question) in lieu of a positive theory (like traditional Marxism) that is capable of challenging what they called “traditional theory”—that is, both the “bourgeois” physical and social sciences and traditional Marxism. Although French Marxists in the fifties and sixties did not face an existential threat as dire as the one that Horkheimer and Adorno had faced in 1937, they did confront a subtle and difficult double challenge: the revelation of the brutal nature of Soviet Communism on the one hand, and the peace, freedom, and prosperity of postwar liberal and capitalist Western Europe on the other. Of course, persistent imperialistic adventures in Algeria and Vietnam still drew left-wing denunciation, but increasingly on non-Marxist, anti-essentialist grounds.

Saussure’s linguistics and Claude Lévi-Strauss’s anthropological structuralism11 permitted European Marxists to turn from a critique of the actual state of affairs to its alleged cognitive or ideological underpinnings. Of course, such a “turn” maintains the foundational Marxist false consciousness principle that allegedly autonomous thought is determined by supra-personal factors. In traditional Marxism that supra-personal factor is ideology—whether religion, liberalism, or capitalism. In critical theory and deconstruction, it is still ideology, but now the maligned ideology is more fundamental than religion, liberalism, or capitalism. The target ideology is the entire tradition of rational thought descending from the Greeks that Derrida dubbed “logocentrism, the metaphysics of phonetic writing,” which he believed—in a Marxist turn—to be “nothing but the most original and powerful ethnocentrism, in the process of imposing itself upon the world” (OG, p. 3). Derrida has articulated the intimate, but difficult, relation of deconstruction to Marxism in Specters of Marx: “Deconstruction has never had any sense or interest, in my view at least, except as a radicalization, which is to say also in the tradition of a certain Marxism, in a certain spirit of Marxism” (SM, p. 115).

But there is an important asymmetry between deconstruction and Marxism. Traditional Marxism is a positive philosophy, Derrida’s is irretrievably negative (or aporetic), as he points out in Specters of Marx: “To critique, to call for interminable self-critique is still to distinguish between everything and almost everything. Now, if there is a spirit of Marxism which I will never be ready to renounce, it is not only the critical idea or the questioning stance (a consistent deconstruction must insist on them even as it also [End Page 418] learns that this is not the last or first word). It is even more a certain emancipatory and messianic affirmation, a certain experience of the promise that one can try to liberate from any dogmatics and even from any metaphysico-religious determination, from any messianism” (SM, p. 111; emphasis added).

Contrast Marx’s justification for destroying the illusions represented by religion in A Contribution to the Critique of Hegel’s “Philosophy of Right”:

Religious suffering is, at one and the same time, the expression of real suffering and a protest against real suffering. Religion is the sigh of the oppressed creature, the heart of a heartless world, and the soul of soulless conditions. It is the opium of the people.

The abolition of religion as the illusory happiness of the people is the demand for their real happiness. To call on them to give up their illusions about their condition is to call on them to give up a condition that requires illusions. The criticism of religion is, therefore, in embryo, the criticism of that vale of tears of which religion is the halo.12

Surrendering the illusions of religion would, Marx believed, enable men to act in their own self-interest. Nothing similar follows if we give up our empiricist and positivistic (logocentric) illusion that correct knowledge is possible, and rise (fall?) into honest Pyrrhonic skepticism.

Of course, many people have made this observation. One of the earliest and most trenchant (and one from a Marxist perspective) is Fredric Jameson in The Prison-House of Language. He observed that Derrida’s “repudiation of the pretenses of the spoken word . . . implies that a text can have no ultimate meaning, and that the process of interpretation, of unfolding the successive layers of the signified, each of which is then in its own turn transformed into a new signifier or signifying system in its own right, is properly an infinite one.”13 In the following pages, I will survey alternative theories of language behavior that call into question Derrida’s deconstruction of logocentrism, based as it is on a misreading of Saussure’s linguistics, plus a tacit acceptance of the Frege-Russell referential theory of language as his antagonist.


The principal exponent of speech act theory is John Searle, with whom Derrida engaged in an irritable debate in 1977.14 If we accept either Searle’s or Peirce’s definition of the sign, the discovery of the [End Page 419] grapheme loses much of the epochal significance Derrida assigns to it. For Searle, if he were to indulge in historical fabrication, the fundamental discovery would be of the “institutional fact,” that is, of the conventional agreement that “x counts as y in context c.” When the cave man shouts “tiger” (the x) in the jungle (context c) his interlocutors recognize that he is referring to a specific present danger (the y). For Peirce, the epochal discovery would be of the “symbol,” his general term that is essentially equivalent to Searle’s institutional fact. For Peirce and Searle, speech, not writing, inaugurates semiotics15 (that is, the general principle of something standing for something else), and hence both human communication and thought.

Early in Of Grammatology Derrida asserts that logocentrism is based on the assumption that “the voice, producer of the first symbols, has a relationship of essential and immediate proximity with the mind” (OG, p. 11; emphasis in the original). Further he notes that logocentrism holds that the “written signifier is always technical and representative. It has no constitutive meaning” (p. 11). That is so, he says, because writing lacks precisely that “relationship of essential and immediate proximity with the mind,” which the voice allegedly possesses. This is the crux of his calling into question the validity of “phono-centrism,” the assignment of logical and historical precedence to speech over writing. But Derrida drops the term “phono-centrism” in favor of “logocentrism,” for polemical purposes that are highlighted a few pages later where he refers to the “metaphysical-theological roots” of logocentrism: “As the face of pure intelligibility, it [the signifier] refers to an absolute logos to which it is immediately united. This absolute logos was an infinite creative subjectivity in medieval theology: the intelligible face of the sign remains turned toward the word and the face of God” (p. 13).

It is a bold leap from the assertion of an intimate connection of speech with speaker, of voice with breath, to the assertion that privileging voice over mark entails a belief in an incarnate deity. It is, of course, true that “Christian creationism and infinitism” held such a view, but it does not follow that phono-centrism entails it. Derrida cites Roman Jakobson’s articulation of Saussure’s dyadic semiology as an example of the phono-centric error on the same page. As a secular Russian Jew, Jakobson is an odd choice as an exponent of “Christian creationism.” In fact, the privileging of the voice over the mark, on Derrida’s own account, goes back at least to Aristotle, who surely cannot be accused of participating in “Christian creationism”! One may readily concede that the Greek innovation of the full alphabet (by adding vowels) was, [End Page 420] though irretrievably lost in the mists of time, an epochal moment in human civilization—as both Marshall McLuhan and Eric Havelock argued in the same decade as Derrida16—without denying that speech itself was an even greater watershed in human culture.

Derrida regards the discovery of the arbitrary dyadic discriminate—the discovery of the grammè (grapheme)—that the Greek alphabet represents, not only as a cognitive leap but also as a political one: as a liberation from the tyranny of the uttered word. A liberation that, in Derrida’s fanciful history, had to wait two-and-a-half millennia; for within Christian “ontotheology” (to use Martin Heidegger’s term, as Derrida does), writing “deferred” to speech; the letter (grapheme) “deferred” to the spirit or breath (phoneme). That deference he sees as the core of logocentrism, a doctrine that allegedly dominated Western thought from Plato until Derrida liberated it by his inauguration of “grammatology,” the science of the graphic mark or grapheme.17

Invoking Plato’s denigration of writing in the Phaedrus as “the dis-simulation of the natural, primary, and immediate presence of sense to the soul within the logos,” Derrida tells us that “deconstructing18 this tradition” will consist “of showing why the violence of writing does not befall an innocent language. There is an originary violence of writing because language is first, in a sense, I shall gradually reveal, writing. ‘Usurpation’ has already begun” (OG, p. 37). One cannot miss the political, quasi-Hegelian cast of these remarks: “dissimulation,” “violence” (twice), “reveal,” “usurpation”—all echoing the Hegelian Aufhebung or “overthrow;” and at the same time, appealing to Marxist revolution, a violent overturning of one dispensation and the inauguration of a new one. This passage is the third or fourth within pages 34 to 37 devoted to castigating Saussure for failing to recognize the priority of writing—a failure that Derrida construes as ethical or political rather than cognitive: “Saussure is faithful to the tradition that has always associated writing with the fatal violence of the political institution” (p. 36).

By representing grammatology as an overthrow of the tyranny of “logocentrism”—understood as the notion that speech expresses the soul (spirit, breath)—Derrida politicizes what might have been a dry philosophical issue. The claim that writing precedes speech is a proxy for the rejection of the Platonic exaltation of the word, the logos, an exaltation that found its apotheosis in the Christian doctrine of incarnation.

The Christian credo begins: “In the beginning was the Word, and the Word was made flesh.” Christ is the Logos, the Word—the living, spoken word, in contrast to the written word of God found in the Hebrew [End Page 421] Scriptures. Playing on the Christian identification of Christ with the logos, Derrida asserts early in Of grammatology: “The sign and divinity have the same place and time of birth. The age of the sign is essentially theological.” His constant denigration of “presence” is a double entendre: a proxy for the Christian doctrine of incarnation, as well as for speech as opposed to writing: “We are disturbed by that which, in the concept of the sign—which has never existed or functioned outside the history of (the) philosophy (of presence)—remains systematically and genealogically determined by that history” (OG, p. 14; parentheses in the original). What is “the philosophy of presence” if not the Christian doctrine of incarnation?

Identifying a motivation for Derrida’s argument, does not, of course, invalidate it. I am not propounding an ad hominem dismissal of deconstruction.19 Nonetheless, to have the motivating ideological issue clearly before us can be a helpful caution for those who might otherwise succumb to Derrida’s “dog-whistle” allusions to Christian belief. Part of his appeal, one might surmise, is that much of American academe shares his antipathy toward Christian beliefs—especially Catholic and Orthodox sacramentalism, the belief that religious rituals have efficacy; for example, that baptism cleanses the recipient of original sin.20


Let me return to the issue of the presumption that the written word (the grapheme) defers to the spoken word (the phoneme) within logo-centrism (aka Christianity). In order to overturn that “ontotheology,” Derrida argues that the ground and core of symbolization is pure difference—as is the case with the alphabet. Of course, that the signifier is different from the signified is not relevant to semiotic expression. Obviously, the vocable “horse” or its written equivalent is different from any imaginable existent horse, not to mention the abstract concept HORSE. But that difference is inoperative in language. The salient difference is not “vertical”; not between signifiers and signifieds (whether concepts or simply lexical entries). The difference is “horizontal”—between signs: between “horse” and “pony,” or “cow” or “sunset,” and so forth. One may argue (implausibly) that “horse,” “pony, “cow,” and “sunset” are pure discriminates, like the letters p and d. But it is difficult to get from there to Derrida’s notion that the difference between linguistic discriminates amounts to deference. [End Page 422]

The notion that word meanings (signifieds) arise purely from their status as discriminates—or what Derrida calls “différance”—is not Sassurean but Jakobsonian. Jakobson articulated a phonological theory in which the value (signified) of a phoneme arises simply from its being a discriminate of all other phonemes occurring in a given language—a view I readily endorse. He further pointed out that phonological systems can be organized in terms of dyadic discriminates, that is, pairs of similar phonemes, such as the English plosive p and voiced f. Jakobson further argued that these pairs were asymmetrical, with one or the other being dominant or “privileged” in a given language. For example, the Latin pater becomes père in French, but father in English and vater in German because the voiced f and v are “privileged” in the Germanic languages over the plosive p. This phonological theory underpins Derrida’s importation of “deference” into his understanding of how language works: the Latin plosive p “defers” to Germanic voiced v and f. It hardly needs saying that there is no compelling reason to equate a somewhat obscure aspect of language drift as the source or cause of social stratification and/or deference to authority or doctrine.

Of course, the notion of deference is little more than a grace note in Derrida’s project to elevate the written letter (grapheme) above the voiced phoneme. He spares no effort to embed this bogus precedence in a reconfigured cultural history that reveals the original sin of the suppression of the grapheme: “the alleged derivativeness of writing, however real and massive, was possible only on one condition: that the ‘original,’ ‘natural,’ etc. language had never existed, never been intact and untouched by writing, that it had itself always been a writing. An arché-writing whose necessity and new concept I wish to indicate and outline here; and which I continue to call writing only because it essentially communicates with the vulgar concept of writing. The latter could not have imposed itself historically except by the dissimulation of the arché-writing” (OG, p. 56). Candidates for the “arché-writing” on archeological grounds would be the pictograph, the hieroglyph, and the ideogram21—though none of these can plausibly be thought to have preceded speech. But to point to such pre-alphabetic modes of writing, as I have just done, is to succumb to what Derrida dismisses as “the vulgar concept of writing.” The vulgarity is to suppose that his abstract entity, “arché-writing,” must be instantiated in some concrete, historical form.

Not much is achieved by dubbing the proper, philosophical understanding of writing “arché-writing.” But he doesn’t stop there. He now [End Page 423] de-essentializes the grapheme as the “trace.” A trace is not a thing but a relationship. The linguistic trace is a unique relationship, one he calls “différance,” a portmanteau word combining “differ” and “defer”:

The (pure) trace is différance. It does not depend on any sensible plenitude, audible or visible, phonic or graphic. It is, on the contrary, the condition of such a plenitude. Although it does not exist, . . . it permits the articulation of speech and writing—in the colloquial sense—as it founds the metaphysical opposition between the sensible and the intelligible, then between signifier and signified, expression and content, etc. . . . Différance is therefore the formation of form. But it is on the other hand the being-imprinted of the imprint.

(OG, pp. 62–63; emphasis in the original)

This remarkable passage contains echoes of Peirce’s definition of the word: “The word itself has no existence.” (Derrida had invoked Peirce earlier in Of Grammatology [pp. 48–50].) However, Peirce is applying his token/type logic, which has nothing to do with metaphysics. The vocable “word” is the token of the type, WORD. The type, WORD, is an abstract entity, and hence has no existence. The token “word” is existent, but is indefinitely variable—in print as word, word, word, etc.; and in various pronunciations by individuals and in dialects.22 On Peirce’s authority, then, Derrida illegitimately declares that his compound relationship, “différance,” “founds the metaphysical opposition between the sensible and the intelligible.”

Peirce, cofounder with William James of pragmatism, sees no such “metaphysical opposition.” For Peirce the word (as an abstract type) has “real being consisting in the fact that existents [tokens] will conform to it” (CP 2, p. 292). Peirce’s token/type distinction is similar to Saussure’s signifier/signified distinction if one thinks across languages. Each language has a distinct signifier/token for the signified/type/concept, MAN. But in Peircean semiotics, the “types” are not just aspects of sign systems, as signifieds often are for Saussure and always are for Derrida. For Peirce, types are the labels of phenomena, entities, or properties found en dehors du langage, that is to say, in the experienced world.

Saussure, remember, was not articulating an epistemology—still less an ontology—but a linguistics. He was concerned only with the “value,” that is, the grammatical or syntactical role of the word, as opposed to its semantic or referential force. Hence, he was quite properly concerned only with relations between the elements of languages, and a primary component of that relation was discriminability, that is, différance.23 [End Page 424]


Structural linguistics’ turn away from philology’s concern with the history of word meanings24 was seized upon by Derrida and his admirers as a license to draw the (illegitimate) corollary that word meanings themselves arise from the relative position of words to other words in a sentence; that they arise from differences within linguistic strings rather than from putative reference or denotation. As Jonathan Culler put it in Structuralist Poetics: “If meaning is a function of differences between terms and every term is but a node of differential relations, then each term refers us to other terms from which it differs and to which it is in some kind of relation. These relations are infinite and all have the potential of producing meaning” (SP, p. 245). If the relations of any given word are indeed infinite, then no determinate meaning could be assigned to any string (i.e., sentence) of words—including Culler’s sentences and mine. Such a consequence is not only intolerable; it is also absurd. However, if we were to invoke the Frege-Russell distinction between sense and meaning, and substitute “sense” for “meaning” in Culler’s sentence, it is no longer absurd. “Sense” is, indeed, intralinguistic, permitting the expression of counterfactuals, gibberish, and fiction. But languages also convey meaning—whether in the Frege-Russell sense of definite reference, or in the Austin-Searle sense of speech acts, that is, propositions, assertions, queries, commands, promises, etc.

According to Derrida, logocentrism commits the error of assigning priority to a “plenitude,” which he construes as an immediacy, and later calls “presence.” One instance of this supposedly false plenitude is the supposition that the spoken word expresses the thoughts of the speaker. In a sense the phoneme (the noise) is said to defer to the thought of the speaker, and the grapheme (the “trace”) is said to defer to the spoken word. Still another deference may be of the thought to its objectthe concept (for Saussure) or the thing, event, or quality (for positivism). But for Derrida, what is essential is not so much the deference as the difference on which he believes all significance depends.

Articulate speech requires two cognitive capabilities involving the recognition of difference. One is recognition of the use/mention distinction. Domestic dogs readily learn verbal commands and references, such as “sit” and “walk,” but are incapable of distinguishing between a use and a mention, as any dog owner knows. The other is the sign/signified distinction—that “sit” signifies a command to behave in a certain way. Dogs can manage that difference. In both cases communication [End Page 425] depends upon a recognition that the utterance is symbolic, that is, one thing (the noise) standing for another (the command or reference). It is not unreasonable to label such symbolic relations as involving the recognition of differencethough the labeling is idiosyncratic.


Derrida’s project is nothing if not grandiose. He claims that his argument “gives us an assured method of broaching the de-construction of the greatest totality—the concept of the epistemé and logocentric metaphysics—within which are produced, without ever posing the radical question of writing, all the Western methods of analysis, explication, reading, or interpretation” (OG, p. 46; emphasis in the original). In other words, like Friedrich Nietzsche and Heidegger before him, Derrida sees himself as announcing an epochal moment in Western intellectual history: the conclusion of an intellectual culture that descends from a dialogue between Platonic idealism and Aristotlean empiricism.

His label, “logocentrism,” for that intellectual culture, is a bit of a pun. Although the “logos” part literally means “word,” it inevitably invokes the notion of logic, the rules of combination governing thought. Logocentric culture, then, privileges not only the notion that words convey meaning (the epistemé) but also that human thought can proceed rigorously from premise to conclusion. That whole edifice of Western thought comes tumbling down if the word is merely a mark, a grammè, a “trace.” In Derrida’s schema, the word is stripped not only of its reference but also of its sense or significance; the word is only the possibility of meaning, itself meaningless.

Derrida invokes Peirce to support this position, claiming that what Peirce calls “symbol” is “playing here a role analogous to that of the sign which Saussure opposes precisely to the symbol” (OG, p. 48). He then cites Peirce on the iterability and self-enclosed nature of the symbol, as if that were Peirce’s view of semiotics in general—which is not the case. Indeed, Derrida is indulging in a bit of equivocation here—whether deliberately or accidentally, I do not know. “Symbol” is Peirce’s term for the arbitrary sign, whereas Saussure uses “symbol” in the standard literary sense as a “motivated sign.” “Mother,” for example, is a natural symbol of nurturing; “lamb” a conventional symbol of Christ, and so forth. They are “motivated” in the sense that the primary designates (MOTHER or LAMB) have properties that “motivate” or prompt a derived symbolic sense. [End Page 426]

Peirce’s general term is “sign,” which he analyzes into three types: symbol (arbitrary), icon (“resemblant”), and index (cause/effect). The latter two would count as motivated signs in Saussure’s schema. Indexical signs are such relationships as smoke for fire, thunder for lightning, etc.; icons are resemblants—pictures, but also graphs, models, and anything from which one can derive properties of its designate by direct inspection or manipulation of the icon—the quadratic equation, for example.

A crucial difference between Peirce’s semiotics and Saussure’s semiology, ignored by Derrida, is the former’s triadic nature: 1) object (thing, quality, or event); 2) sign (itself a triad); and 3) interpretant (mind). As already noted, Saussure’s semiology is dyadic: sign (arbitrary mark), signified (lexical entry). Peirce’s most-often-cited definition of the sign gives the construer of the sign (a mind) prominence: “A sign, or representamen, is something which stands to somebody for something in some respect or capacity” (CP 2, p. 228; emphasis added). As already noted, for the most part Saussure leaves the mind out of the account—an omission that underpins Derrida’s assertion that the meaning of a sign is another sign.

After citing Peirce on the three branches of the science of semiotics—which turn out to be the traditional ones: grammar, logic, and rhetoric, all qualified by the epithet “pure”—Derrida co-opts Peirce as a precursor:

Peirce goes very far in the direction that I have called the de-construction of the transcendental signified, which, at one time or another, would place a reassuring end to the reference from sign to sign. I have identified logocentrism and the metaphysics of presence as the exigent, powerful, systematic, and irrepressible desire for such a signified. Now Peirce considers the indefiniteness of reference as the criterion that allows us to recognize that we are indeed dealing with a system of signs. What broaches the movement of signification is what makes its interruption impossible. The thing itself is a sign.

(OG, p. 49, emphasis in the original)

Derrida goes much too far in the last claim. Certainly, for Peirce anything may be taken as a sign, and all signs are also things. By the phrase “thing in itself,” Derrida no doubt means to invoke either the Kantian Ding an sich or Edmund Husserl’s sich selbst (he mentions Husserl on the page quoted above). But both sorts of idealism are quite antipathetic to the pragmatist, Peirce. The following paragraph is as lucid a statement of a “pragmatic” argument that one is likely to find: [End Page 427]

And what do we mean by the real? It is a conception which we must first have had when we discovered that there was an unreal, an illusion; that is, when we first corrected ourselves. Now the distinction, for which alone this fact logically called, was between an ens [entity] relative to private inward determinations, to the negations belonging to idiosyncrasy, and an ens such as would stand in the long run. The real, then, is that which, sooner or later, information and reasoning would finally result in, and which is therefore independent of the vagaries of me and you. Thus, the very origin of the conception of reality shows that this conception essentially involves the notion of a COMMUNITY, without definite limits, and capable of a definite increase of knowledge. And so, those two series of cognition—the real and the unreal—consist of those which, at a time sufficiently [in the] future, the community will always continue to re-affirm; and of those which, under the same conditions, will ever after be denied.

(CP 5, p. 331; emphasis added)

Peirce’s pragmatic understanding that (in the absence of any direct access to the ens) the real is determined by a community that progressively discards error, flies in the face of the cognitive relativism that underpins deconstruction. It allows for the “transcendental signified” outside of language even though it is not incorrigibly accessible. Unlike Fish’s “interpretive communities,”25 Peirce’s community is singular—the consensus of the entire scientific/scholarly community. Nor is Peirce’s posture equivalent to Rorty’s radical relativism, which Rorty calls “contingency”: “when we no longer worship anything, where we treat nothing as a quasi-divinity, where we treat everything—our language, our conscience, our community—as a product of time and chance.”26 Rorty’s substitute for justified true belief (to put Peirce’s posture in more contemporary terms) is a notion of social cohesion, which he calls “solidarity”: “that a belief can still regulate action, can still be thought worth dying for, among people who are quite aware that this belief is caused by nothing deeper than contingent historical circumstances” (CIS, p. 189). Rorty chooses “solidarity” because it is a feature of societies, not of individual knowledge. His cognitive relativism rises to a Pyrrhonic level, that is, the assertion that human cognition is so restricted that there is no such a thing as truth or falsehood.27

Despite Derrida, Peirce is in no sense a precursor of such radical skepticism, still less of Pyrrhonism.28 Peirce’s realism, and his triadic model of semiosis, are incompatible with such postmodern predilections. His triadic semiosis has no social component: “A Sign, or Representamen, is a First which stands in such a genuine triadic relation to a Second, called [End Page 428] its Object, as to be capable of determining a Third, called its Interpretant.” In less abstract language, the sign is a sensible thing that refers to, resembles, or indicates something else for a sentient being. He calls that sentient being an “interpretant.” Later in the same paragraph, he makes clear that the “interpretant” is a mind: “A Sign is a Representamen with a mental Interpretant. Possibly there may be Representamens that are not Signs. . . . But thought is the chief, if not the only, mode of representation” (CP 2, p. 274; emphasis added).29


We must, then, reject Derrida’s derivation of his semiological solipsism—expressed in the axiom that the meaning of a sign is always another sign—from Peirce. Of course, we cannot communicate meanings except through signs (in Peirce’s general sense, not structuralism’s restricted sense of arbitrary signs) and that therefore the decoding, translation, or interpretation of a sign or string of signs is always expressed in more signs. However, Derrida’s claim, cited above, that “Peirce goes very far in the direction that I have called the de-construction of the transcendental signified, which, at one time or another, would place a reassuring end to the reference from sign to sign” is not supported by the passage he cites.

The Peircean remark that Derrida keys on is “the laws by which in every scientific intelligence one sign gives birth to another” (CP 2, S1, item 229). It must be read in the context of the previous sentence, in which Peirce endorses “Kant’s fashion of preserving old associations of words in finding nomenclature for new conceptions” (emphasis added). In short, Peirce is speaking within the prevailing view that signs (words in this case) are arbitrarily assigned meanings (“concepts”), and endorsing a species of linguistic conservation in which old labels are assigned new meanings. For Peirce—as for J. S. Mill and Saussure—signs do not refer to other signs, but to concepts (as well as things, events, and qualities), and hence one can mint new meanings from old words. Perhaps the best-known Kantian instance of such a practice is his use of “transcendent.” As it happens, that is a recoinage that Derrida exploits.

Derrida’s oft-quoted slogan “il n’y a pas de hors-texte” is sometimes construed ontologically to mean that there is no reality transcendent of language. Such a position is difficult to categorize. The only precedent I can think of for such a view is the form of Kabbalism that holds that the Hebrew Bible contains all of the cosmos, in the sense that the [End Page 429] nature of the cosmos can be teased out of it by means of one or another interpretive strategy. In his survey of the literature on Derrida’s affinity with Kabbalism, Sanford L. Drob cites a Hasidic master, Rabbi Zadoq ha-Kohen of Lublin, to that effect: “Thus I have received that the world in its entirety is a book that God, blessed be He, made, and the Torah is the commentary that he composed on that book.”30 Drob plausibly characterizes that remark as equivalent to Derrida’s “Il n’y a pas de hors-texte.” Derrida has denied the influence of Kabbalism on his thought, and of any affinity between his ideas and Kabbalism (Drob, p. 3). However, as I argue below, “Il n’y a pas de hors-texte” need not be construed as “there is nothing outside of the text,” which, expressed unequivocally, would be “Il n’y a pas dehors du texte.”

It would take us far afield to pursue that topic further. For my purposes, it is sufficient to suppose that Derrida means the remark to be construed in a constrained epistemological, rather than an ontological, manner: to mean that we cannot cognitively stand outside of semiological systems. On this reading, not only are we unable to communicate outside of, or beyond, semiological (or semiotic) systems but we cannot think outside of them. On that reading Derrida belongs in the company of a strain of American linguistics quite distinct from Saussurean structuralism.


The supposition that our thought is both formed and constrained by the language in which it is expressed is known as the Whorf-Sapir hypothesis. It has been widely rejected by the dominant strain of Anglo-American academic philosophy. Oddly, its close similarity to Derrida’s deconstructive hypothesis is seldom noted. One of the few mentions I have found is in John Holcombe’s web essay, “Jacques Derrida.” Holcombe dismisses Derrida’s arguments as old hat and indefensible, remarking: “At best, reality can only be partially circumscribed by words, and what we know of brain functioning would make it highly unlikely that anything as complicated as consciousness could be governed by the small areas responsible for linguistic skills.” He goes on to observe: “Only the weak form of the Whorf-Sapir hypothesis is generally accepted; that is, that language directs and influences but does not entirely constrain perception.”31

Here is Benjamin Whorf’s expression of the hypothesis in a nutshell, published posthumously in The Theosophist: [End Page 430]

Actually, thinking is most mysterious, and by far the greatest light upon it that we have is thrown by the study of language. This study shows that the forms of a person’s thoughts are controlled by inexorable laws of patterns of which he is unconscious. These patterns are the unperceived intricate systematizations of his own language—shown readily enough by a candid comparison and contrast with other languages, especially those of a different linguistic family. His thinking itself is in a language—in English, in Sanskrit, in Chinese. And every language is a vast pattern-system, different from others, in which are culturally ordained the forms and categories by which the personality not only communicates, but also analyses nature, notices or neglects types of relationships and phenomena, channels his reasoning, and builds the house of his consciousness.32

I have italicized the salient claim of the hypothesis. Whorf was a chemical engineer, but presented his linguistic ideas at several academic conferences, eventually ecoming a professional linguist under the tutelage of Edward Sapir.

Sapir’s formulation of the hypothesis is essentially equivalent to Whorf’s, quoted above, except that Sapir omits the claim that language also determines our understanding of natural phenomena: “It is quite an illusion to imagine that one adjusts to reality essentially without the use of language and that language is merely an incidental means of solving specific problems of communication or reflection. The fact of the matter is that the ‘real world’ is, to a large extent, unconsciously built upon the language habits of the group. No two languages are ever sufficiently similar to be considered as representing the same social reality. The worlds in which different societies live are distinct worlds, not merely the same worlds with different labels attached.”33

If we reduce Derrida’s deconstruction to the weak claim that languages form and modulate our thought in some unspecified way, the revolutionary implications of his posture are largely defanged. We can situate him in a long tradition in linguistics that began with Alexander von Humboldt, and is continued by Wittgenstein, Austin, and Searle. Both hermeneutic and Marxist thought postulate some degree of reflexivity—that is, that we project our understanding onto the world, rather than passively receiving impressions. As long ago as 1971, M. H. Abrams surveyed these contrary views as reflected in literary culture in The Mirror and the Lamp. Abrams argued that classical culture regarded the arts as essentially mimetic—as reflecting the world; whereas Romantic culture thought of the arts as casting a revelatory light on the world, inevitably illuminating some aspects and leaving others in darkness. Although [End Page 431] Abrams was no Marxist, Marxist false consciousness theory makes the same claim. More recently, Marshall McLuhan argued in Gutenberg Galaxy that the advent of print altered the psyches of Europeans, and that electronic communications would do so again.34


However, Derrida stands apart from the Humboldtian tradition in his dismissal of the “transcendental signified,” which must be “hors du texte.” He first introduced the term in his discussion of Heidegger: “Heideggerian thought would reinstate rather than destroy the instance of the logos and of the truth of being as ‘primum signatum’: the ‘transcendental’ signified.” Derrida qualifies this assertion in a parenthetical remark: “‘transcendental’ in a certain sense, as in the Middle Ages the transcendental—ens,35 unum, verum, bonum—was said to be the ‘primum cognitum’” (OG, p. 20). These remarks clearly associate the “transcendental signified” with the Christian and post-Christian (Heideggerean) assumption that the spiritual or mental is transcendent of the physical.

It is a connection that the American Transcendentalists also made—in their case invoking Immanuel Kant. Here is Emerson on transcendentalism: “It is well known to most of my audience, that the Idealism of the present day acquired the name of Transcendental, from the use of that term by Immanuel Kant, of Konigsberg, who . . . showing that there was a very important class of ideas, or imperative forms, which did not come by experience . . . that these were intuitions of the mind itself; and he denominated them Transcendental forms.”36 The “transcendental signified,” on this reading, is found in a spiritual/mental realm, and may even be thought to be divinity itself.

Of course, Derrida is not interested in debunking American Transcendentalism, or even Husserl’s or Heidegger’s phenomenology. His target is Russellian and Fregean positivism, which grounds linguistic meaning in references to things, events, and properties that lie outside of language. Typically, Derrida approaches that target obliquely; its identity becomes clear only later in a much-cited passage where the equivalence of referent and transcendental signified is (inadvertently?) asserted: “the methodological considerations that we risk applying here to an example are closely dependent on general propositions that we have elaborated above, as regards the absence of the referent or the transcendental signified. There is nothing outside of the text” (OG, p. 158; emphasis in the original). Aware of the audacity of this remark, the translator, Gayatri [End Page 432] Spivak, gives an alternate translation plus the French sentence: “[there is no outside-text; il n’y a pas de hors-texte.].” Clearly the first, italicized translation is not warranted, and the alternate translation should have been preferred, as noted above. It seems that Spivak considered the alternate translation as equivalent to the first. But it is not. Spivak’s preferred translation asserts an ontological state of nonexistence, with strong echoes of the Kabbalistic belief that the Pentateuch contains the material universe.

The alternate translation is susceptible to a Kantian reading, asserting the impossibility of transcending language, of accessing the “transcendental signified,” which we can construe as Kant’s Ding an sich, “the thing in itself.” On such a reading, Derrida is saying nothing new or startling. However, his assertion that our blindness to ultimate reality is a result of our dependence on language is hardly sustainable. If it were truly impossible to transcend language, translation would be impossible—an undesirable consequence for Derrida, whose fame in America rests primarily on English translations of his French-language texts. And is it appropriate to include mathematics in the category of language? Contemporary physics, chemistry, and even biology are largely expressed in mathematical rather than linguistic terms and formulae. Of course, Derrida’s followers (and even Rorty) do not shy from including the physical sciences in their polemic against logocentrism. Derrida himself is more cautious.37

From a Kantian perspective, language can refer only to ideas and phenomena, and cannot pierce behind phenomena to ultimate reality. However, language can and does refer to ideas and phenomena that clearly are “transcendent,” in the ordinary use of the term, of language. They are “outside” of language in the uncontroversial sense that one can translate Kant’s Ding an sich as “thing in itself.” For that to be possible, there must be a conceptual realm “transcendent” of German, English, etc. to which translators have access. And one can eat an apple, pomme, Apfel, melo in any language.

From another angle, Derrida’s “grammatology” can be seen as derivative of Nietzsche’s aphoristic anticipation of the Whorf-Sapir hypothesis in The Will to Power—that we “have to cease to think if we refuse to do it in the prison house of language.”38 Jameson used Nietzsche’s remark for the title of his 1972 critique of structuralism (and of Derrida), The Prison-House of Language: A Critical Account of Structuralism and Russian Formalism. Of course, Derrida goes much further—or at least appears to. Although he does not deny that we can think within language, for him [End Page 433] thinking becomes a mere game, a moving about of “traces” that themselves do not exist but are merely the gaps, spaces, or intervals between signifiers and signifieds that he calls “différance” (OG, pp. 62–63).

The Peircean arbitrary symbol (the word), by contrast, “would,” he wrote, “lose the character which renders it a sign if there were no interpretant”—that is, no mind to construe it. The symbol is that sign (the general term for Peirce) “which signifies what it does only by virtue of being understood to have that signification” (CP 2, p. 304). Symbols “come into being by development out of other signs, particularly from icons, or from mixed signs partaking of the nature of icons and symbols. We think only in signs. These signs are of a mixed nature; the symbol parts of them are called concepts. If a man makes a new symbol, it is by thoughts involving concepts” (p. 302; emphasis added).

Peirce’s belief that “we think only in signs” appears to be compatible with Nietzsche’s notion of the prison house of language, but it is not. Peirce’s theory of signs is triadic: the arbitrary symbol (the word) is supplemented by the index and the icon, which are clearly transcendent of language. A Peircean index “points at” its cause: a bullet hole indicates a bullet; lightning indicates thunder; red spots on one’s skin indicate measles, and so forth. The significance of the icon is founded on some affinity with its referent. It is not limited to pictures or other representations based on resemblance, such as imitative sounds. The model for this class, however, is the picture: “Every picture (however conventional its method) is essentially [an icon].” But so, too, is an algebraic formula, “though it may seem at first glance that it . . . might as well, or better, be regarded as a compound conventional sign.” For Peirce, the “great distinguishing property of the icon is that by the direct observation of it, other truths concerning its object can be discovered than those which suffice to determine its construction” (CP 2, p. 279). In short, the icon, like all of Peirce’s signs, is not defined simply by any properties it may possess but also by the manner in which it is interpreted. This feature of the sign is essential to Peircean triadic semiotics consisting of sign -> object -> interpretant.

This triadic nature of Peircean semiotics distinguishes it sharply from the structuralists’ dyadic semiology—signifier/signified—on which Derrida’s argument depends. If interpretation were constrained by such a dyadic model, it would, indeed, be impossible to transcend language. But we know that every sentient being transcends language every moment of its existence. Not only tigers, eagles, and fish but human hunters, automobile drivers, and lovers “read” icons and indices that surround [End Page 434] them. The hunter does not need to verbalize, “I see a deer ten degrees to my right, seventy-five yards away, running to my left” before aiming and firing; a driver does not need to label the deer in the headlights as “a deer in the headlights” before applying the brakes; and a lover does not need to enumerate the virtues of the beloved to become smitten. The dyadic view, as Saussurean semiology is understood by deconstructors, omits the most important element of signification: the mind that construes signs.

That Saussure omitted the “interpretant” from his semiology is not a fault for linguistics. But by grounding his epistemology on Saussure’s dyadic semiology, Derrida predetermines his conclusion that “il n’y a pas de hors-texte.” And, of course, it is not just the material world that lies outside the text, but also the interpretants—Derrida himself and all of his readers. Without transcendent interpretants (minds), there is no semiology, no language, no significance, no meaning. The “play” of meaning ends when the interpretant (you or I) construes a sign as meaning this or that. True, the sign may have many different endings, but not as many as readers or hearers, otherwise communication would be impossible.

Why deconstruction took important segments of the American academic world by storm is a difficult question to answer. I think the affinity of such a skeptical philosophy with Marxist false-consciousness theory is an important factor in that spectacular success. However, it is more difficult to account for what amounts to a Marxist conquest of American literary culture in the years of the Vietnam adventure—an adventure motivated by knee-jerk anti-Communism. Jeffrey Mehlman, a disillusioned early adopter of deconstruction, tried to account for that conquest in “Writing and Deference: The Politics of Literary Adulation.” Somewhat enigmatically, he compares the work of Gerhard Heller, Hitler`s cultural attaché in Paris during the German occupation, to the American academics’ welcoming of Derrida: “His [Heller’s] work, that is, a cultural impresario on behalf of a dominant power offers an odd prototype for the activities of many a French department chairman in the United States. That suggestion may seem scandalous . . .” His clarification leaves one a little perplexed: “The similarity lies in the hegemony—be it financial, political, or military—enjoyed by the national community of adulators in relation to the France of the adulated—and the inevitability of that superiority’s figuring in the relation.”39 Whatever that means exactly, it clearly indicates that Mehlman agrees with Derrida that there [End Page 435] is a political/ideological dimension to the vogue of deconstruction in the United States, as there was for its earlier rise in France.

Leon Surette
Western University


1. Harold Bloom, “Clinamen or Poetic Misprision,” New Literary History 3, no. 2, On Interpretation: I (Winter 1972): 381.

2. In an apparently careless—and characteristically ambivalent—remark in a 1989 interview, Derrida seems to have conceded that language is not self-contained: “No doubt all language refers to something other than itself or to language as something other. One must not play around with this difficulty.” (“This Strange Institution Called Literature: An Interview with Jacques Derrida” in Acts of Literature, ed. Derek Attridge [New York: Routledge, 1992], p. 48.)

3. Jacques Derrida, Of Grammatology, trans. G. G. Spivak (1967; repr., Baltimore: Johns Hopkins University Press, 1976), p. 46 (emphasis in the original); hereafter abbreviated OG.

4. John R. Searle, Speech Acts (Cambridge: Cambridge University Press, 1969), pp. 51–52; hereafter abbreviated SA.

5. Ferdinand de Saussure, Course in General Linguistics, trans. Roy Harris (La Salle: Open Court, 1983), p. 117, emphasis added; hereafter abbreviated CGL.

6. Ludwig Wittgenstein, Philosophical Investigations, ed. G. E. M. Anscombe and R. Rhees (Oxford: Blackwell, 1953).

7. For an assessment of Derrida’s grasp of speech act theory, see Raoul Moati, Derrida/Searle: Deconstruction and Ordinary Language, trans. Timothy Attanucci and Maureen Chun (New York: Columbia University Press, 2014). He claims that Derrida had “no familiarity with Grice” (p. 30) and adds:

Nevertheless, a set of theoretical misunderstandings interferes with this discussion [of Austin and Searle’s language theory] and causes, on Derrida’s side, some questionable interpretations. First questionable interpretation: Searle’s reworking of Austin’s theory of speech acts in light of intentionality confirms, for Derrida, his own intentionalist reading of Austin. Second problematic interpretation, which should not be underestimated: by “Husserlizing” all forms of intentionality, Derrida, who knew neither “Speech Acts” nor “Expression and Meaning” before working on his own response to Searle, has been led to caricature Searle’s true position. The debate with Searle quickly transforms into a reiteration of the debate with phenomenology, which shows yet again the strong anchor attaching deconstruction to the work of Husserl, here through the instrumentalization of Austin’s and Searle’s theses.

(p. 116; emphasis added)

8. Jacques Derrida, Specters of Marx, trans. Peggy Kamuf (New York: Routledge Classics 1994), p. 16; hereafter abbreviated SM.

9. The first scholar to co-opt Saussure was a philosopher, Maurice Merleau-Ponty, not a literary critic. In “Le phénomène du langage,” a 1951 paper read at an International Colloquium of Phenomenology in Brussels (printed in Éloge de la philosophie, 1960) he brought Saussure out of obscurity—so far as philosophy and literary criticism are concerned. Merleau-Ponty was cofounder, and coeditor with Jean-Paul Sartre, of the Marxist cultural journal Les Temps Moderne. He is said to have converted Sartre to Marxism, but his disillusionment with Marxism around the time of the Korean War led him to Saussure’s structuralist linguistics.

For an account of Merleau-Ponty’s “misprision” of Saussure, see D. R. Koukal, “Merleau-Ponty’s Reform of Saussure: Linguistic Innovation and the Practice of Phenomenology,” presented at the Society for Phenomenology and Existential Philosophy (October 1995), available at

10. Max Horkheimer, “Traditional and Critical Theory” in Critical Theory: Selected Essays, trans. Matthew J. O’Connell (New York: Continuum, 1975), p. 142.

11. Claude Lévi-Strauss’s 1952 paper “Linguistics and Anthropology” does not mention Saussure at all. Instead, he invokes the Russian linguist Roman Jakobson (whom he had recently met in New York) and the controversial American linguist Benjamin Lee Whorf. Nor did Saussure receive any attention in “Structural Analysis in Linguistics and Anthropology” published seven years earlier (now chapter 2 of Claude Lévi-Strauss, Structural Anthropology [1958; repr., New York: Penguin Press, 1968], pp. 31–80). Culler cites that 1945 paper at the beginning of Structuralist Poetics as one of the foundational texts in the turn to linguistics, despite the absence of any reference to Saussure in it (Jonathan Culler, Structuralist Poetics [London: Routledge and Kegan Paul, 1975], p. 4; hereafter abbreviated SP).

12. Karl Marx, Works of Karl Marx, vol. 3, ed. Frederick Engels (London: Lawrence and Wishart, 1974).

13. Fredric Jameson, The Prison-House of Language (Princeton: Princeton University Press, 1972) p. 176.

14. For Derrida’s incomprehension of Searle’s view of language, see Jacques Derrida, Limited Inc., ed. Gerald Graff (Evanston: Northwestern University Press, 1988), which includes his response to Searle’s “Reiterating the Differences” (Glyph 2 [1977]), a critique of Derrida’s “Signature, Event, Context” (Glyph 1 [1977]).

15. Peirce’s term for the theory of reference as a branch of logic is “semiotics,” as opposed to “semiology,” Saussure’s term for his general theory of language. Obviously, Peirce’s icons and indices are “hors du langage.”

16. Marshall McLuhan, Gutenberg Galaxy (Toronto: University of Toronto Press, 1962); Eric A. Havelock, Preface to Plato (Cambridge: Harvard University Press, 1963). McLuhan focused on the epochal nature of the invention of moveable type in the fourteenth century as completing the rupture between thought and utterance that writing had begun—as Plato had lamented in the Phaedrus. Havelock’s focus was on alphabetic writing and the Greek addition of vowels, thus disambiguating the Semitic alphabet, which lacked vowels.

17. Derrida introduces the term “grammatology” as a replacement for Saussure’s “semiology” for a general theory of signs. As “the theory of writing,” grammatology is given “the scope needed to counter logocentric repression [that is, the Frege-Russell positive theory of linguistic meaning] and the subordination to linguistics” (OG, p. 51; emphasis added).

18. Incidentally, this is his first, and rather innocent, use of the term “deconstruction.”

19. Deconstructors are, however, prone to succumb to ad hominem arguments. See Derrida’s response to Searle’s critique of his views in Limited Inc. (note 14 above). See also the response of the editors of Lingua Franca to Alan Sokal’s hoax, detailed in Alan Sokal and Jean Bricmont, Fashionable Nonsense: Postmodern Intellectuals’ Abuse of Science (New York: Picador, 1998).

20. Compare Hillary Kaell, Walking Where Jesus Walked: American Christians and Holy Land Pilgrimage (New York: New York University Press, 2014): “While the Catholic Church has become more critical of many devotions, especially since the Second Vatican Council, there is still a rich theological tradition that connects the sacred to the material world. In contrast, the protest of the Reformation was largely a rejection of Catholic sacramentalism. Early Protestant theologians believed that the Church encouraged the collapse of the sacred and the secular, which muddled God’s relationship to man and distracted from grace, faith and scripture” (p. 17).

21. For a survey of critical commentary on Derrida’s use/neglect of Chinese ideographic writing, see Sean Meighoo, “Derrida’s Chinese Prejudice,” Cultural Critique 68 (Winter 2008): 163–209.

22. “A Symbol is a Representamen whose Representative character consists precisely in its being a rule that will determine its Interpretant. All words, sentences, books and other conventional signs are Symbols. We speak of writing or pronouncing the word ‘man’; but it is only a replica, or embodiment of the word, that is pronounced or written. The word itself has no existence although it has a real being, consisting in the fact that existents will conform to it. It is a general mode of succession of three sounds [“m,” “a,” and “n”] or representamens of sounds which becomes a sign only in the fact that a habit, or acquired law, will cause replicas of it to be interpreted as meaning a man or men” (C. S. Peirce, Elements of Logic, vol. 2 of Collected Papers of Charles Sanders Peirce, ed. Charles Hartshorne and Paul Weiss [Cambridge: Harvard University Press, 1932], p. 292, emphasis in the original; hereafter abbreviated CP and cited by volume and page number).

Derrida read Peirce during his sojourn in New York. See Floyd Merrell, Deconstruction Reframed (West Lafayette: Purdue University Press, 1985), and E. San Juan Jr., “Charles Sanders Peirce’s Theory of Signs, Meaning, and Literary Interpretation,” St. John’s University Humanities Review 2, no. 2 (May 2004). Both point out Derrida’s misprision of Peircean semiotics.

23. Jeffrey Barnouw’s article, “Peirce and Derrida: ‘Natural Signs’ Empiricism versus ‘Originary Trace’ Deconstruction” (Poetics Today 7, no. 1 [1986]): 73–94, magisterially demonstrates Derrida’s misprision of both Saussure and Peirce: “Precisely in his refutation of the idea of intuition as immediate non-inferential grasp, or what Derrida calls ‘presence,’ Peirce elaborates the conception of the sign that will prove unassimilable both to Derrida’s Saussurean conception and to his quasi-Husserlian critique” (p. 78). Also: “In fact, however, Derrida follows Husserl in his conception of phenomenology, rather than Peirce, just as he follows Saussure, rather than Peirce, in his conception of the sign” (p. 79). His article seems to have been ignored by deconstructors, and is rather too technical to be incorporated into my argument, although it supports it.

24. James Turner describes philological practice as follows: “The word philology in the nineteenth century covered three distinct modes of research: (1) textual philology (including classical and biblical studies, ‘oriental’ literatures such as those in Sanskrit and Arabic, and medieval and modern European writings); (2) theories of the origin and nature of language; and (3) comparative study of the structures and historical evolution of languages and of language families” (James Turner, Philology: The Forgotten Origins of the Modern Humanities [Princeton: Princeton University Press, 2014], p. x).

25. See Stanley Fish, Is There a Text in This Class? (Cambridge: Harvard University Press, 1980).

26. Richard Rorty, Contingency, Irony, and Solidarity (Cambridge: Cambridge University Press, 1989) p. 22, emphasis in the original; hereafter abbreviated CIS.

27. For Rorty’s Pyrrhonism see Robert B. Brandom, ed., Rorty and His Critics (Oxford: Blackwell, 2000). Rorty himself invokes Pyrrhonic skepticism in Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1979), pp. 46, 94n8, 113, 139–40.

28. For a defense of Derrida against the charge of Pyrrhonism, see Bob Plant, “Perhaps . . . Derrida and Pyrrhonian Scepticism,” Angelaki Journal of the Theoretical Humanities 3 (December 2006): 132–57. He concedes: “Given the Pyrrhonist’s radical devaluing of truth, preference for rhetoric over reason, and ethical-political indifference, some might view Derrida’s work as a sort of neo-Pyrrhonism.” But he argues: “There is a story to be told about Derrida and Pyrrhonism, but it does not offer any straightforward correlations. Rather, where Derrida’s work looks most Pyrrhonian is where their paths diverge” (p. 141). It turns out that Plant believes Derrida to be even more pessimistic than Pyrrhon: “Contrary to Pyrrhonism’s teleology of liberation, Derrida—following Levinas—promises nothing in this regard” (p. 142).

29. No doubt for Peirce the interpretant was the human mind. But there is no reason to restrict his semiotics to human interpretants. Clearly sniffing dogs construe scents as indices, hunting eagles construe flying pigeons as icons of prey, and trained dogs construe voiced commands as symbols.

30. Sanford L. Drob, “Jacques Derrida and the Kabbalah” (2006),, p. 7, emphasis added; hereafter abbreviated Drob.

31. John Ellis also points out the precedence of the Whorf-Sapir hypothesis in Against Deconstruction (Princeton: Princeton University Press, 1989), p. 38. However, he does not explore the similarity.

32. Benjamin Lee Whorf, “Language, Mind, and Reality” Theosophist (1942), repr. in Language, Thought, and Reality: Selected Writings of Benjamin Lee Whorf, ed. John B. Carroll (New York: John Wiley and Sons / MIT Press, 1956), p. 252; emphasis added.

33. Edward Sapir, Selected Writings of Edward Sapir, ed. David G. Mandelbaum (Berkeley: University of California Press, 1958), p. 162; emphasis added.

34. A search of the Derridex site for “Marshall McLuhan” produces no results—much to my surprise.

35. Notice that Peirce uses the “medieval term” ens in the passage cited above.

36. Ralph Waldo Emerson, “The Transcendentalist,” lecture read at the Masonic Temple, Boston (January1842),

37. In Fashionable Nonsense, Sokal and Bricmont explain the absence of commentary on Derrida: “Since there is no systematic misuse of (or indeed attention to) science in Derrida’s work, there is no chapter on Derrida in this book” (p. 8). Sokal is the author of the infamous hoax article in Social Text 46, no. 47 (Spring/Summer 1996): 217–52, and the subsequent Beyond the Hoax: Science, Philosophy and Culture (Oxford: Oxford University Press, 2008).

38. Friedrich Nietzsche, The Will to Power, trans. Walter Kaufmann and R. J. Hollingdale (New York: Vintage Books, 1968), p. 283 (aphorism 522).

It is worth noting that, according to the website Derridex, Derrida has never mentioned Benjamin Lee Whorf or Edward Sapir in his writing. Neither are Bertrand Russell, Alfred North Whitehead, Gottlob Frege, or Derrida’s American contemporary Nelson Goodman mentioned. Derrida did consider J. L. Austin’s theories in “Signature, événement, contexte” in Lecture, Communication: Congrès International des Sociétés de Philosophie de Langue Française (Montreal, August 1971). It can be found as “Signature, Event, Context” in Jacques Derrida, Margins of Philosophy, trans. Alan Bass (Chicago: University of Chicago Press, 1985). That lecture prompted a defense of Austin by Searle (see note 14 above).

39. Jeffrey Mehlman, “Writing and Deference: The Politics of Literary Adulation,” Representations 15 (Summer 1986): 1–14.

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.