SOURCE: Culler, Jonathan. “Semiotics and Deconstruction.”¹Poetics Today 1, no. 1-2 (autumn 1979): 137-41.

[In the following essay, Culler examines the interplay between deconstruction methodology and semiotics, noting that semiotics can benefit from “the most rigorous pursuit of logic” in the text that is the hallmark of deconstruction.]

The moment when semiotics is becoming well-established in America—a subject of conferences, a topic of university courses, and even a domain to which people in various traditional disciplines are beginning to relate their own work—is also, as is perhaps only appropriate, a moment when semiotics finds itself under attack, criticized as a version of precisely the scientific positivism which is itself very prone to reject semiotics. In many cases, of course, the attack on semiotics comes from a traditional humanism, affronted that a discipline with scientific pretensions should claim to treat products of the human spirit. These arguments can be countered in various ways which I shan't be discussing here. I'm interested in a more radical critique which also focuses on the scientific pretension of semiotics—a critique which compels our attention precisely because it isn't another version of traditional humanism. One could cite various examples of this position. I offer as not untypical, but among the better informed, J. Hillis Miller's argument that among literary critics who have been influenced by European developments

a clear distinction can be drawn […] between what might be called to conflate two terminologies, Socratic, theoretical, or canny critics on the one hand, and Apollonian/Dionysian, tragic, or uncanny critics on the other. Socratic critics are those who are lulled by the promise of a rational ordering of literary study on the basis of solid advances in scientific knowledge about language. They are likely to speak of themselves as “scientists” and to group their collective enterprise under some term like “the human sciences.” […] Such an enterprise is represented by the discipline called semiotics […]. For the most part these critics share the Socratic penchant, what Nietzsche defined as “the unshakeable faith that thought, using the thread of logic, can penetrate the deepest abysses of being […].” The inheritors of the Socratic faith would believe in the possibility of a structuralist-inspired criticism as a rational and rationalizable activity, with agreed upon rules of procedures, given facts, and measurable results. This would be a discipline bringing literature out into the sunlight in a “happy positivism.”

(1976: 335)

The “uncanny critics,” on the other hand, have no such faith in the possibility of general and systematic theories, because they have discovered that a careful working through of individual texts, whether literary or philosophical, leads to unmasterable paradoxes, aporias, which seem constitutive of the domain of signification itself.

I cite this argument as an example of the attitude which has come to be called “post-structuralist” or “deconstructionist,” but I am less interested in Miller's own account of the situation, which seems primarily based on psychological categories—are the proponents of semiotics “lulled by the promise of rational ordering”? do we have “unshakeable faith” in thought?—than in the powerful interpretive account of linguistics and semiotics on which statements such as Miller's are ultimately dependent: Jacques Derrida's reading of Saussure in De la grammatologie. I want to focus on this reading partly because I think Derrida is right and that his argument should be better understood among semioticians but also because I am interested in the implications for semiotics of this deconstructive reading.

But first a word about Derrida's general project. Derrida's writings, his readings of theoretical texts, are explorations of what he calls the “logocentrism” of Western culture—the “metaphysics of presence” which these texts can be shown simultaneously to affirm and to undermine. The metaphysics of presence—our metaphysics—determines being as presence, granting ontological primacy, for example, to what is deemed to be present to consciousness. The Cartesian cogito, in which the I is deemed to lie beyond doubt because it is present to itself in the act of thinking, is one instance. Another is our notion that there is a present instant which we can truly grasp in that it is self-identical and present to us, and that the reality of everything depends on its relation to this presence of the present: the past a former present and the future an anticipated present. Finally, there is the notion of meaning as something present to the consciousness of the speaker at the moment of utterance: what the speaker “has in mind” as he speaks.

Derrida is interested in the way in which this logocentrism is “deconstructed” in texts that affirm it. To deconstruct logocentrism is to show that what was taken to be the truth of the world or the ground of an enquiry is in fact a construct that has been imposed and which is contradicted by certain results of the enquiry it founds. This will become clearer as we follow Derrida's reading of Saussure. The Cours de linguistique generale and semiotics generally provide a striking case for Derrida, for it can be shown to contain a radical critique of logocentrism and the metaphysics of presence on the one hand, yet an affirmation of logocentrism and an inextricable involvement with it on the other. Let us take these movements in turn.

• 1. The critique of logocentrism. Since Saussure defines language as a system of signs, the central question becomes that of the nature and identity of signs and their constituents. Saussure is led by his investigation to argue that linguistic units do not have essences but are defined solely by relations. His famous claim that “dans la langue il n'y a que des différences, sans termes positifs,” is a principle wholly at odds with logocentrism. It maintains, on the one hand, that no terms of the system are ever simply and wholly present, for there are no positive terms and a difference can never be present as such. The reality of linguistic units cannot depend, therefore, on their presence as such. And, on the other hand, in defining identity in terms of common absences rather than in terms of common presences, the linguistic and semiotic principle puts in question the principle which is the very cornerstone of the metaphysics of presence.

• 2. The affirmation of logocentrism. Though Saussure has specified that sound itself does not belong to the linguistic system, and though he has used the example of writing to illustrate the nature of linguistic units, he adamantly denies that writing is an object of linguistic enquiry (“spoken forms alone constitute its object”) and treats writing as a parasitic form, the representation of a representation. This may seem like a relatively innocent move, but in fact a great deal depends on it. And the energy and moral indignation which Saussure displays in relating how linguists can “fall into the trap” of attending to written forms and how writing can “usurp the role” of speech and even affect pronunciation—this suggests that more is at stake than meets the eye. Ever since Plato condemned writing in the Phaedrus, discussion of language has ascribed some characteristics of language to writing and then relegated it to a position of dependence so as to confront a purified speech. To put it very schematically, if writing is set aside as dependent and derivative, accounts of language can take as the norm the experience of hearing oneself speak, where form and meaning seem given simultaneously and in an event of incarnation, rather than, say, the act of deciphering an anonymous inscription. The repression of writing minimizes the differential or diacritical nature of language in favor of supposed presences.

The privileging of speech is not only a weighty matter, it is also very nearly inescapable. Why is this so? Because linguistic analysis, and by extension semiotic analysis, depends upon the possibility of identifying signs. To identify signs one must be able to identify signifieds, since a sequence is a signifier only if it is correlated with a concept or signified. We know bet and pet are different signifiers because each has associated with it a different signified. And if we ask how we know this, at what moment or place this association is given, the answer requires some form of presence and will ultimately refer to the moment of speech: the moment of utterance when signifier seems to deliver the signified which is present in it or which it expresses at that moment. The moment of utterance seems to present positive terms, given in and of themselves, and thus to provide a point of departure for analysis of a system said to consist only of differences.

Since the possibility of grasping or identifying signifieds is necessary to the semiotic project, it is no accident that semiotic theory should find itself implicated in phonocentrism and logocentrism. It is neither an accident nor, I want to insist, an error. Let me quote a passage from the Grammatology:

The privilege of the phonè does not depend on a choice that could have been avoided. It responds to a moment of economy (let us say of the “life” of “history” or of “being-as-self-relationship”). The system of s'entendre parler through the phonic substance—which presents itself as the non-exterior, non-mundane, and therefore non-empirical or contingent signifier—has necessarily dominated the history of the world during an entire epoch, and has even produced the idea of the world, the idea of world-origin, that arises from the difference between the worldly and the non-worldly, the outside and the inside, ideality and non-ideality, universal and non-universal, transcendental and empirical, etc.

(1976: 7-8)

These are large claims. They may become more comprehensible if one notes that oppositions such as outside/inside, transcendental/empirical, etc., depend on a point of differentiation, a line of division which distinguishes the two terms and commands the opposition. The claim is that the moment of speech, where signifier and signified seem given together, where inner and outer or physical and mental are for an instant perfectly fused, serves as the point of reference in relation to which all these distinctions are posited.

Since the privileging of speech is essential to our metaphysics, Derrida does not argue that Saussure was mistaken in asserting the primacy of voice and founding linguistic analysis on the necessarily logocentric notion of the sign. On the contrary, Derrida's analyses of the ubiquity of logocentrism—even Georges Bataille can be shown ultimately to be a Kantian—show that analysis is necessarily logocentric: even the most rigorous critiques of logocentrism cannot escape it since the concepts they must use are part of the system being deconstructed. There are, of course, various ways of playing with or resisting the system that one cannot escape, but it would be an error to suggest that Derrida and deconstruction have provided us with an alternative to semiotics and logocentrism. Grammatology, Derrida has said, is not a new discipline which could replace a logocentric semiology; it is the name of a question (1972: 22). Indeed, Derrida's own writing involves a series of strategic manoeuvres and displacements in which he modifies his terms, producing a chain of related but non-identical operators—différance, supplément, trace, hymen, espacement, greffe, pharmakon, parergon—to prevent any of his terms from becoming “concepts” of a new science.

Derrida's reading of Saussure is an exploration of the self-deconstruction of semiotics. Indeed, in the interview in Positions entitled “Semiologie et grammatologie” he identified his double science or double reading not with a mode of discourse that would lie outside or beyond semiotics but with a special practice within semiotics.

One can say a priori that in every semiotic proposition or system of research metaphysical presuppositions will cohabit with critical motifs by virtue of the fact that up to a certain point they inhabit the same language, or rather the same system of language. Grammatology would doubtless be less another science, a new discipline charged with a new content or a new and well-delimited domain than the vigilant practice or exercise of this textual division (la pratique vigilante de ce partage textuel).

(1972: 49-50)

Three points by way of conclusion. Deconstruction isn't, at least in the work of Derrida and its other most skillful practitioners, some kind of “new irrationalism,” as is occasionally suggested. Though it reveals “irrationalities” in our systems and theories, it is the most rigorous pursuit of the logic of the text, be it a theoretical or a literary text.

The aporias deconstruction reveals are contradictions, paradoxes, which semiotics cannot escape: in this case that the theory of the sign leads to theoretical principles which must be repressed if analysis is to take place. Semiotics cannot help supposing positive terms which its theory must disallow. Further discussion would lead us to the unmasterable oppositions between langue and parole, system and event, synchronic and diachronic, which can never be held together in a single self-consistent system. Generally, semiotics is not the self-consistent discourse of a science but a text.

What deconstruction advises is not a change of direction—the correction of some error which would make it a “true science”—or a change of purpose—a shift from semiotics to a new discipline of grammatology. There is no escape from textuality; one can only engage it with critical vigilance: “la pratique vigilante de ce partage textuel.”

Note

1. This paper was delivered at the International Conference on the Semiotics of Art, sponsored by the University of Michigan, May 1978, and will appear in the proceedings of that conference. For a more extensive development of this discussion, see my “Structuralism and Grammatology,” boundary 2 (forthcoming).

References

Derrida, Jacques, 1976. Of Grammatology, Gayatri Spivak, trans. (Baltimore: Johns Hopkins).

1972 Positions (Paris: Minuit).

Miller, J. Hillis, 1976. “Stevens's Rock and Criticism as Cure,” The Georgia Review 30:2, 330-405.