Phenomenology and Cartesianism

Husserl and Peirce are the modern philosophers of post-rationalism. Husserl developed phenomenology—and also wrote a semiotic. Peirce developed semiotics—and also wrote a phenomenology (phaeneroscopy). In inverse proportion each made a link between semiotics and phenomenology at the outset, and they did so in the context of their adversary position vis-à-vis Descartes.

For Descartes, signs were either natural (ultimately God-given) or conventional (man-made; like mathematical symbols).³ Husserl criticized the natural standpoint; Peirce the conventional (he made a critique of the framework of assumptions of a science). Though the type of sign each philosopher had in mind differed from the other’s, they were in basic agreement as to the falseness of the natural/conventional split. Their signs—Husserl’s Zeichen (Ausdruck and Anzeichen) and Peirce’s triad—concur in the central concern with meaning-, and meaning for an interpreter. Like their Romantic precursor, Kant, they are interested in the Understanding as the important supplement to Reason. Husserl named this understanding Consciousness; Peirce, the Interprétant. In neither is understanding located exclusively in an individual person or subjective process; it is a post-rational structure. (We may here remind the reader that Locke, too, wrote a semiotic in tandem with his Cartesian critique.)

Husserl’s stunning Cartesian Meditations set up the criticism. Assumptions about the simple origin of understanding—the thinking subject—had to be reviewed. Thinking, thought, is replaced by Husserl with consciousness, and a consciousness that it is impossible to divorce from its contents: it intends them. And the subject, in all its apparent unity and simplicity, is at least double: transcendental and empirical, imbued with otherness in its essence. Consciousness is a composition or synthesis that can be temporally or spatially arrayed and analyzed; the subject can be viewed at different levels of existence. This double analysis of consciousness as synthetic and of phenomena as consciousness is the phenomenological method, a method opposed to Descartes’s, and it avoids subject/object, mind/body, inner/outer and self/other, origin/aim divisions. It is a radical critique of the scientific attitude.

So far so good, but the scientific method was not about to capitulate at once to the phenomenological one. Still a fresh start, a renewal of the Kantian style of criticism had been made.

But phenomenology has had the peculiar fate, in the European history of ideas, of having each of the advances made in it appear as a radical réévaluation of it. The same has not been the case for Peirce, who hardly had time to name a new aspect of his theory, before it was used by his friends like Will James and became in altered form⁴ the uncriticized basis for much research. The Peirce revival has so far taken the form of an archival admiration,⁵ not a philosophical rereading. Peirce with irrepressible self-irony and an evident distaste for discipleship managed the one-man show of self-criticism; Husserl seemed fated to suffer Oedipalization. Phenomenology barely had been named as such when it underwent the existential revolt led by Heidegger and then Sartre and Merleau-Ponty, and that was only the beginning. Existentialism attacked phenomenology; structuralism attacked existentialism; semiotics attacked structuralism ... The seemingly endless conflicts masked the general post-rational alliance of these “adversaries.”

The growth of insight into structure is not a simple process, and even the “Oedipal” history we have sketched here is a fiction. Husserl’s work did not flower into immediate fruit—there was Hitler, the war, the friends become enemies, the religious questions yet to be resolved, and overall the continuing expansion of rational science and its technologies. Let us now trace the drama by which semiotics reclaimed for itself a cardinal position in post-rational thought, rediscovering its dual philosophical source in both Husserl and Peirce, but also developing its own complex style independent of philosophical debates. Even more than its genealogy it is its critical varieties that constitute the “history” of semiotics.

The Drama of Semiotics: Existentialism, Structuralism, and the Hermeneutic Method

The existentialism of Martin Heidegger and that of Husserl’s students who also followed Heidegger continued phenomenology in the same directions, essentially the same two directions proposed by the master: consciousness and phenomena studied together. The fundamental effort was to advance the phenomenological method into a nonphilosophical setting; to show how consciousness and phenomena are twin manifestations not of the same essence, but the same existence. The undoing of rationalistic oppositions continued: Merleau-Ponty sought to work over the mind/body split; Sartre the self/other dichotomy; Heidegger the past/present opposition. Two processes were involved: unifying consciousness and its contents in concrete form and advancing the assumption that all of experience is potentially meaningful: but only to the Understanding—‘meanings’ do not exist without Interpretation.

It was Martin Heidegger’s Sein und Zeit (1927) that led the advance of phenomenology into the arena of everyday life. Although Heidegger criticized Descartes’s ontology, he cut short the debate with scientific rationalism over the subject. His work began firmly within the neo-Kantian post-rational philosophy, as a critique of Wilhelm Dilthey’s hermeneutic method of interpretation. The disagreement has its roots in eighteenth-century European notions of “pity” (identification with the other; sympathy) and “self-love”⁶ as epitomized by the difference, say, between Rousseau and Adam Smith. The Romantic conclusion that the imagination is the locus of the capacity to sympathize (cf. Kant, Hazlitt, et al.) eventually entailed the revival of the hermeneutic method for Understanding the Other. (Hermeneutics is the science of textual interpretation; some place its origins in Hesiod; some in the work of Philo the Jew of Alexandria and his reconciliation via allegorical methods of the Old and New Testaments. Some modern anthropologists such as Clifford Geertz also use the method.) Heidegger begins his chef d’oeuvre with an attack on the way the neo-Kantians applied hermeneutics, attempting to reconstruct a lost totality, lost because ‘past’. Heidegger questioned the assumption of the definitive split between past and present, and also desacralized the ‘texts’ to which hermeneutic methods could be applied. Heidegger’s is perhaps the first post-medieval though non-religious return to the notion that experience itself could be significant, i.e., interpretable as meaningful to someone. Heidegger’s improvement upon the hermeneutic method involved the introduction of non-chronological time into the process of understanding, with knowledge always being prefigured, or having the structure of foreknowledge. Knowing was like a circle, a return, at a higher level, to something already in some way ‘known’. Imagination, the figurative and prefigurative, language in its most poetical form, bear in this system great powers of insight; they are closest to the structured circle of Understanding that constitutes our most profound existence.

Everyday life’s “talk,” paintings, jazz, lyric poems, clothing styles, behaviors, politics, can now be viewed as texts, open to hermeneutic understanding and not subject to heavily moral distinctions between sacred and profane, authentic and inauthentic. Representations themselves no longer bear the burden of being ‘mere’ representations, but are instead conceived of as the very “stuff” of our existence. Indeed literary criticism was one of the first great beneficiaries of this existentialism,⁷ for it lent the linguistic arts greater prestige, perhaps, than they had ever before enjoyed. Their representative was no longer simply an unreal image, a superfluity; it participated in the very structure of all existence.

Social scientists did not long ignore the potential Heidegger’s existentialism held as an alternative to the rationalist/behaviorist theories of behavior. But since existentialism remained at first primarily in the hands of philosophic and religio-ethical thinkers—Sartre, Merleau-Ponty, Maritain, Jaspers, et al.—empirical social investigators found frustration. True, Sartre was enamored with the political engagement and applicability of the analysis of consciousness to experience, and Merleau-Ponty wrote on the structure of behavior. But one had to admit that they seemed to do so mainly for the sake of the mode of consciousness they analyzed—Sartre took the imagination, Merleau-Ponty the perception—rather than for the sake of the intentional objects or contents of the consciousness. (It remained for Emmanuel Lévinas and his student Jacques Derrida⁸ to work out the modality of memory—but that is another story and its irony is made clear below.) And as philosophers, Sartre and Merleau-Ponty could remain methodologically open, but the empirical researcher often needed something more ad hoc, concrete, and operational than a reflective method or a hermeneutics. For even though Heidegger worked out the justification for envisioning all of life as a text or network of associations, the anthropologist or psychoanalyst needed to demonstrate the particular textuality of their subject/matter before hermeneutic methods could be applied: a cumbersome process, perhaps even inappropriate for studying unlettered peoples or peoples with different versions of being and time.⁹

It is in this context that one should view the work of Jacques Lacan, Claude Lévi-Strauss, and Emile Benveniste: psychoanalyst, anthropologist, and linguist. They seemed at first startlingly different in approach from their contemporaries Sartre and Merleau-Ponty, although all were friends and associates, and all were loyal to existential tenets. The great division between the “existentialism” and the “structuralism” of these researchers is that the social scientists literally honored their subject/matter as bearing the burden of intentional consciousness, subordinating interpretation to it, rather than vice versa. Following Heidegger, structuralists made the strange familiar and the familiar strange. They uncovered the byways of what Heidegger called Dasein—the unthought yet continual structuring of the world by the prefigurative consciousness, best exemplified by language, but also evident in class consciousness and class conflict, etc.

But structuralism foresook the hermeneutic method that was so much a part of the Heideggerian programme. And it took Husserl’s “transcendental” ego not as a classical autonomous self but as this expanding structuring and forestructuring, à la Kant. In seeing this structuring as being “like a language” the structuralists adopted a method of language analysis more applicable than hermeneutics to mundane, ‘profane’ matters, more systematic and researchable: Saussure’s semiotics.

A radical move, and important both for the future and the past—for in reaching back behind Kant to Rousseau and to his later Swiss compatriot, structuralists opted for a method that faced head on the process of meaning-making (semiosis) without their having the security of a speculative position from which to view the process and without being able to deal with already “worked up” textualized materials. Both Rousseau and Saussure accepted the sheer arbitrariness and traditionalism of all our social institutions, of which speech appears to be the prime example. But even speech is only a special case of semiosis, the production of meanings (or Heidegger’s play of representations), the archetypal set of meanings that we always feel need to be understood. Semiosis produces meaningful entities entirely in the absence of any “reasonable basis” for their existence.

Lévi-Strauss, Lacan, and Benveniste were quick to seize upon the sign and its meaning-producing mechanism as the most effective means of organizing and analyzing the structure of Dasein. Moreover, the Dasein to be scrutinized is the unconscious manifestation of the general consciousness: it is, that is, not self-conscious, not worked up as an art or science; yet even though it seems ‘natural’, it is no less structured as a set of signs or network of representations than any art or science may be. In turning up evidence for this un/conscious structure at every juncture structuralism implicitly resuscitated the transcendental self—that which is originary in “giving the world”—even though one first perceived structuralism as a shock because it had no center, no “subjects.”¹⁰ As in the early Heidegger, questions of authenticity and profundity could be put aside and the elaboration of crystallized (traditional) institutions derived from purely arbitrary beginnings could be undertaken. For Lévi-Strauss, simple oppositions analogous to the arbitrary phonemic ones of language generated other sets of oppositions, which “read” or interpreted each other without the fiction of an authoritative mind creating the link between them: oppositions (the syntagmatic relationship in Saussure’s terminology) seemed always to generate a mediating term or opposition to the opposition: thus the division raw/cooked, which encoded the nature/culture division, produced a new sign, the “rotten,” which embodied the fundamental characteristics of each side—the natural yet processed food. Lest there be any mistake that this is an abstract formalism, Lévi-Strauss insists that these patterns must have a concrete content for them to be structures. Structures are not forms, but the interplay between formal oppositions and material oppositions.

For Lacan, too, the initial syntagmatic relationship—the oppositiongenerates a structure, the psyche. The ego is to the id as the self is to the other, and langage is the other within the self, etc. Benveniste’s linguistics proceeds also by means of primary binary oppositions that generate a third term: the person “I” opposes itself to “you” and together they oppose the “they” or “he,” the third person.

Structuralism was not of course welcomed at first by existentialism, nor did structuralism always agree with existential theses; but the link is there, they are in the same general enterprise. Lévi-Strauss dialogues with Sartre at the end of The Savage Mind, but he dedicates the book to Merleau-Ponty. Both existentialism and structuralism had opted to venture into the post-rational world.

Post-structuralism: The Semiotic Critique of Semiotics

Structuralism re-emphasized the dual levels of consciousness that existential theory had found too idealistic and problematic for the politics of experience. In their attention to the un/thought or un/conscious^* (the ‘foreknowledge’ of Heidegger’s circle of understanding) as a concrete set of structures—empirical, analyzable, determinable—structuralists redressed a certain imbalance and out-existentialled the existentialists by researching these structures and not leaving them to speculation. Mythic foundations, psychical history, natural language itself, were the “un/conscious”: and they were accessible because they were essentially the other modality of consciousness, its alter ego.

And though the risk of reintroducing idealism into the post-rational project was grave, it was clear that the structuralists had something there. Linguistic structure, the everyday forms of speech do “found” our ways of being in the world, our subjective sense of our own present and presence as Benveniste demonstrated.¹¹ “Myths were good to think with,” constituting the substance of our understanding in an unreflected way, as Lévi-Strauss showed for Amerindians and Barthes, ironically, for the French bourgeoisie.¹² Our primary experiences prefigure our secondary ones, shaping our persons, just as they had been prefigured by symbolic structures, as Lacan showed. And the methods are not difficult. Restricting semiotics to the level of the signifier—the speech event, the acoustic image, the material signifier—seems a fair guarantee in effect against idealizing or ideologizing the sign’s ‘meaning’ apart from its lived context. Meaning could not be idealized if it were only the ‘meaning-effect’ of the relationship between mere signifiers.

The structural methods, too, are marvelous devices and they work, they give us impressive insight into the ways experience is organized. For example, after our students had studied The Savage Mind, we assigned landscaping as a topic for analysis. The students were quick to ‘read’ landscaping not as nature vs. culture, but as nature and culture vs. agriculture (aesthetic vs. utilitarian). The great power of these methods may lie in their being semiotic, that is, they link meaning to a concrete experiential, not an abstract, sign.

Yet the tendency toward idealism is there, at both the operational level and the thematic level: operationally the insensitive investigator will tend toward automatism (but then so do signs, in their cultural development). The tendency is perhaps more disturbing, however, when it seems the necessary result of the spirit of the structuralist enterprise, or further yet, of the post-rational one: the overcoming of divisions. For it is one thing to point out the arbitrariness and traditionalism of oppositions; it is still another to devote oneself to the overcoming of these oppositions at the expense of developing others.

It had to happen. Restricting analysis to the syntagmatic (and, by extension, paradigmatic) relationship (oppositional)¹³ one seems always to generate a third or mediating term and then further oppositions and more new signs, until a saturation point is reached in which there is to be no more room for differences. Look at the case thematically, from the ethnic and personal myths studied by Lévi-Strauss and Lacan. Lévi-Strauss’s work in the context of the European history of ideas is radical indeed, for in a deep sense it overcomes the great division made since the Romantics (Mme. de Staël, Chateaubriand, Matthew Arnold) between the Judaeo-Christian and the Greek foundations of our culture. (Stendhal, Nietzsche, and Freud each tried to overcome these distinctions too.) Faced with a subject matter (Amerindians) clearly foreign to these twin sources, Lévi-Strauss resorts to both the Oedipus and the Holy Grail myths for models.¹⁴ How radical a move this is, and how much it is a practical sublation of the Romantic division between naive and complex cultures, becomes apparent when Lévi-Strauss’s work is set beside analyses of Western culture that are written “from within,” most of which, including even contemporary work like Auerbach’s Mimesis, are based on this initial and all-pervading division between Hebraism and Hellenism. It might also be said that Lacan operates this reconciliation, integrating the Greek with the Christian myths into the constitution of the self: the symbolic plus imaginary produces the real. “Jewgreek is Greekjew”—Derrida cites Joyce, and senses that with the structuralists closure is at hand.

Only to slip away. For no sooner had the apparent saturation point been reached than the next rewriting of the phenomenology began.

Jacques Derrida honored the importance of the semiotization that structuralism brought to existential analysis. So much so that he undertook to reestablish it on different grounds. His own radical revision takes him back to Husserl (to a different Husserl from the one we saw through existential eyes) and to the sign. He reformulates first the genealogy, then the methodology: like Sartre and Merleau-Ponty, Derrida continues the “faculty” analysis of consciousness by studying the memory; like Nietzsche and Heidegger, he finds forgetting a crucial aspect of memory;¹⁵ and like his phenomenological predecessors, he finds the sign to be of the essence. But instead of going to Heidegger for his version of the un/conscious—he goes to Freud—to an unconscious that is not a modality of the consciousness as is Heidegger’s unthought. And rather than going to Saussure for a model of the sign, he goes to Peirce:

(Derrida’s warrant for this move in Husserl’s phenomenology is discussed below.) And on the question of meaning he does not join the existential revolt against the signified in the name of the signifier so much as he differentiates and pluralizes meanings: the “sign” is a bounded multiplicity, like Nietzsche’s, having the relationship between signifier and signified determined by the history of the repression of one meaning by another.

Derrida’s double substitution of Freud for Heidegger and Peirce for Saussure separates post-structuralism from structuralism, and has farreaching implications. Not only does the type of analysis differ, but its overall emphasis on recovering difference rather than overcoming oppositions implies a much more active total structure than that finally envisaged by Lévi-Strauss: Derrida undertakes his reading of Lévi-Strauss’s Tristes Tropiques in Of Grammatoiogy, part n, “Nature, Culture, Writing.” He does so again in a later essay (1970:255-59). Derrida does not draw attention to the different “signs” used by himself and Lévi-Strauss, but one should note that Derrida uses “sign” at every point where Lévi-Strauss uses “signifier” to characterize the notion of “freeplay.” (Freeplay is “a field of infinite substitutions within the closure of a finite ensemble,” a finite ensemble being, for Lévi-Strauss, one of the chief aspects of a mythic as opposed to a scientific system [1966:16-19].) Within this system, magic’s all-embracing determinism operates by means of a “superabundance of signifiers” (tantamount to primitive “mana”: like the zero phoneme in phonetics, and the phallus in Lacan’s theory, these are signifiers that lack a signified. See Lévi-Strauss, 1950). Derrida uses the term “sign” rather than “signifier” to describe this supplementary structure (1970:260). With wonderful irony Derrida masterfully links even the notion of a mythic closed totality to the totality of Western metaphysics. A further irony is that an ‘enclosure’ is one of two founding signs—the female (vaginal) sign—as distinguished from the conceptual, scientific, male, detachable part-object (phallic) one.

Structuralists have not often responded directly to this new development in post-rationalism, but one has heard oblique objections. For example, Benveniste makes it a point in the essay that founds the international journal Semiotica to decry the world implicit in a Peircian semiotic:

Benveniste opts instead for a linguistic Saussure—for a version of the Saussurian sign in which the phonological, acoustical signifier is the unit of analysis, predominating over and determining the signified. But the drama of the transition from structuralism to post-structuralism is the drama of the return of the sign—the signifier and the signifieds and the interpreter—such as Peirce and (as we shall argue) Saussure himself intended it. (The reader might recall our discussion of the difficulties involved in a semiology that subordinates itself to linguistics in the first chapter.)

For ultimately it is the sign itself (its history of pervading the world with meanings) that requires the development of semiotics—even if only to preserve our capacity of signing off from time to time.

SEMIOSIS AND THE DISCRIMINATIONS OF SIGNS

At the same time that there has been a global expansion of semiotic theory and practice there has been an inflation of the sign as well in all aspects of culture. Even before Nietzsche mobilized his “army of metaphors and metonymies” for a vision of semiosis uncontrolled and unwilled by a subject, other indications that the birth of signs was proceeding at an alarming pace appeared. Romantic literature offered both the chief offense and the chief counterattack: one need only review the luxuriating in metaphoric powers exhibited and expressed as such by Chateaubriand, for example, in Atala and the immediate critique of this style by Stendhal (in Racine et Shakespeare); Poe’s ionization of Coleridgian language is another example. It is as if a mechanism had been discovered for producing texts—one could associate any concept with any image and create a sign, a meaning-effect. Tear-jerking, thrills and chills, suspicion, the supernatural could all be automated, so to speak, or at least engineered by setting the signmechanism in motion, not merely in drama proper but, as Goffman has taught us, in the drama of everyday social life. And the deus ex machina could always be exposed, via Romantic irony, for yet another dimension. The popularity of the detective novel, of stories about confidence men, of ambiguous states of mind and soul, Poe’s tales of terror and his treatise on the mechanics of poetry, are exemplary evidence of a growing mastery of the sign-mechanism. And these are only the good examples: the sign-mechanism could make any signifier mean any signified and let the critics be damned. Modern literature has attempted to break the hold over art of total representation by signs (Mallarmé against Wagner) by a restriction to the signifier and an attack on the signified. “Meaning” as anything above or beyond signifiers could simply be disavowed: and literature need only “mean” itself. Joycean laughter rings, as he assembles and reassembles signifiers to demystify the automatism of the signifier/signified relationship.

In science and in philosophy signs have competed with nature as both mother and legislator: “Symbols grow,” wrote Peirce, “Omnesymbolum de symbo/o” (1955:114-15), and they legislate the birth and type of other symbols. Barthes (1972:141-42) writes pithily of the transformation of culture from anti-nature to pseudo-nature that the Peirceian dictum implies:

Science no longer takes ‘nature’ as its opponent to be dominated, but creates an ideologized neo-nature, a nature constructed of signs.¹⁶

Economically and politically too the sign mechanism appears to prevail. Marx’s commodity—the equivalences of values that are not only unequal but incommensurable—is technically a sign.¹⁷

Nietzsche once called this metaphor the process of equating unlikes, and it is the representative sign-character of the commodity that so amazes Marx as he opens his Capital with the picture of the wealth of capitalist nations displaying itself as an “immense accumulation of commodities.” Marx is indignant that, as with Peirce’s symbol, these representations could have a “life” of their own and enter into social relations with each other. (1965:71-83)

Today semioticians are beginning to observe the “oversaturation of semiosis in mid-twentieth-century culture” (Lotman 1977:338), as they review the plethora of primary, secondary, and even tertiary modelling systems. Entropy threatens as systematization reaches this point—the endpoint of structuralism sensed by Derrida. Where it bursts the bounds of fictitious hierarchizations and levels we need to deal with semiosis per se, meanings produced at the distributional, disseminational, and uncontrollable level.

Semiosis and the Sign: Varieties

In this book we have tried to correct for the tendency to blur the differences between structural and post-structural thought. And, at the same time, we want to correct for an associated error of an opposite kind: the tendency to overstress the differences between the Peircian and Saussurian models of the sign. They overlap as much as they diverge. The essential definition of their basic unit is that it is not a ‘unit’ but at least double or triple in its structure—the sign is always a “sign of.”¹⁸ Saussure’s sign is a relationship between a signifier and a signified. A derivative example would be New Critical versions of metaphor wherein the vehicle would be the signifier and the tenor signified. Peirce’s formulations seem slightly more modern in that they include a third part to the sign: the act of interpretation or reading, which is built right into the structure of the sign. The Peircian model of the sign can be diagrammed as follows:

[signifier/interpretant/signified]

sign

Interestingly, this model was the ostensible one appended by B. Malinowski to Ogden and Richard’s (1946[1922]:324) The Meaning of Meaning, but their working definition of metaphor was closer to the Saussurian, rather than the Peircian, sign.¹⁹ (A very good comparison, relevant both for literary and anthropological theory, to be made is that between the model of the sign used in the Ogden and Richards texts and that used by Lévi-Strauss.)

It is the Saussurian sign that has operated in the course of the development of continental European semiotics to unleash simultaneously a wealth of studies of signs and sign-systems and a certain hostility to the sign—the same sort of hostility evident in respect to metaphors that imply the equivalence of one item for another. Thus it is that we find an astonishing depth to the types of semiological linguistic analyses inspired by Saussurian formulations.²⁰ Yet these semiological studies, systematic though they have been, have been biased toward the material signifier, the sound-image, and the speech act, of “parole,” aspect of language itself.²¹

This bias is partly due to the existential appeal of the speech act. In his definitions, Saussure names all the elements that will strike a particularly responsive chord in existentialists and structuralists:

The signifier in effect is paradigmatic of existence, while the signified is basically ‘metaphysical’ or unavailable for direct inspection. Thus the phonological semiologists concentrated their attention on signifiers and made structural gains by discovering relationships of binary oppositions and their mediations (e.g., Jakobson’s phonemic triangles) as forms that required actual “concrete” existents to fill them and make meaningful structures.

A more dramatic radicalization of this existential reading of Saussure developed by way of Saussure’s “French connection”—in the work of Jacques Lacan and Roland Barthes.^* Here the primacy of the signifier is announced with fanfare; the instance of the “event” or speech act gains a centrality that is framed ethically and politically. The unity of the sign as designated by Saussure, the absolute mutual determination of the sound-image and the concept is for these French writers tantamount to metaphor, and a metaphor of a particularly pernicious sort. For Lacan the sign-as-metaphor is the capital symptom; for Barthes the sign-as-metaphor is a pseudo-unification of opposites, the epitome of false consciousness. The sign is hysterical for Lacan; bourgeois for Barthes. According to their formulations, the sign must be “corrected”: analyzed, the false associations or links dissolved, decomposed, decoded.

There has actually been a considerable unreflected agreement as to the means for decomposing the Saussurian sign/metaphor into analyzable and controllable parts: it is clear that what is lacking is the “subject” of the speech act, the purveyor of the sign, the addresser and the addressee, who are implicated in it. Jakobson makes this move, without the French fanfare, in his essay “Closing Statements” in which he suggests the receiver/sender/context and code are all necessary for semioanalysis.²² It is equally so for Lacan and Barthes. Lacan insists we ask “Who is speaking and to whom?” as Freud did in his analysis of the phrase “A child is being beaten.” And Barthes never seems to tire of dissecting the anonymous voice—like Heidegger’s “on dit”—that addresses us, assaults us at every turn in our daily lives. Barthes achieves some of his finest effects by his uncovering the “representation” that parades as a mere “presentation”—the famous picture of the colonized Negro conscript saluting the tricolor—and then discovering who has perpetrated this representation (1972:121-26). For Barthes the source, the addresser, is the bourgeoisie itself, and so is the addressee: the aim is its own mythification.

The need for supplementing Saussure’s sign with a subject, conscious or unconscious, neutral or innocent, motivated or unmotivated, derives from the attention to the signifier. As distinct from Peirce’s formulation of the sign noted above, it is the absence of the term (interpretant) corresponding to a subject that liberated semiology from subjectivity and left semiology open to freeplay. In the absence of an identifiable subject, the potential for uncontrolled semiosis (pathological associations, the deployment of unconscious representations) is enormous, and the linguists, psychoanalysts, and critics we have been discussing here supply a subject to Saussure’s sign. This subject takes the form of an analytical intellect, the individual as interpreter or theory itself.

Thus we have, for example, Barthes’s development of the structural study of narrative in which literature is removed to the arena of discourse (parole) and a programme for its analysis is laid out (1975:237-71). Carrying a semiology that is centered on the signifier out to its fullest limits, Barthes shows how one can master the meaning of semiotic systems on the model of linguistics by a movement between levels of analysis. It is the intellect that takes what is at each distributional level of semiosis a meaningless signifier and integrates it into a higher system level. In this construction, linguistics yields its grammar to the wider rules of Discourse. These rules are, of course, rhetoric.²³ Doing the poetics of prose-as-speech-act is indeed one of the only options open to literary analysis within the limits of the signifier and speech event. Derrida recognizes this, when at the end of his lengthy critique of phonological semiolinguistics he praises Barthes for resubordinating semiology to language, a move he terms “fecund and indispensable” (1976:52). Practioners such as Greimas, Todorov, and Benveniste are evidence of the fecundity Derrida cites, as are the critics around the journal Poétique, such as Gérard Genette, Paul de Man, and Philippe Lejeune.

Even though his own grammatology aims at meaning which is at the distributional level, Derrida does not object to Barthes’s remaining within the limits of Discourse because Barthes’s efforts have been in many ways salutary for a discipline (criticism) that, like psychology, is always in danger of unconsciously censoring or restricting semiosis and preparing rigid codes of “meanings” (“The poem means this,” “the tower means that,” etc.). It is then “meanings” that must be attacked (in the name of the signifier alone) by resorting to the hypothetical possibility of a literature that refuses to signify (or a psyche, in the case of Lacan). The notion that literature refers to “nothing,”²⁴ so popular with the group around Barthes, is basically a rhetorical definition of literature, in which only a controlled intellect, capable of an essentially allegoric process of reading signs, can produce meaning through its powers of integration. We always have the exciting prospect of discovering at least two levels of significance in the Barthesian type of analysis: Todorov can reveal the “grammar” of the Decameron; Greimas can produce a meaning-effect by using the syntactical subject/object relationship as a way of naming divisions in novels; Benveniste can talk about grammatical persons and tenses so as to make one feel that a revelation about the subject and temporality is at hand. Barthes and his group have achieved a way back to literature from linguistics, not only to save an outmoded institution, but because literature is the best ally one can recruit to attack the bourgeois and synthetic unity of the sign, to retrieve semiotic (communicative) language from mythic metalanguage. Literature, that is, has always been enemy to the hypostatized entity of “language,” especially in the form of grammar. It is no accident that one of the signs Barthes dissects is that of the Latin grammar exercise, the sentence the schoolboy is supposed to learn (1972:117-21). Like literature itself, Barthes constantly rebels against imposed meanings, and he sees good literature as that which can construct artificial myths, as Flaubert does in Bouvard and Pécuchet or Baudelaire in Les Fleurs du mal.

And yet, when the possibility of a non-intellectual liberation from imposed meanings appears—as when Benveniste faces Peirce and a world where everything is a sign (subject/object/and interpreter)—there is hesitation, a stopping on the edge of an abyss. Benveniste (1969b), in a move that really reverses the trend toward the primacy of the signifier (although it does not announce itself as such), demands of any semiotic system that it have a meaningful unit of analysis. Language is the master of semiology because its smallest units are meaningful signs (11-12), whereas in music, for example, the note is but a meaningless mark, except in relation to other notes. Like Barthes’s, Benveniste’s overall aim is to save the Understanding; if not the Reason.

The abyss of semiosis here is not so much that of the endless possibilities for meaning-generation, but the loss of levels, or hierarchies, to establish meanings, provisionally or legislatively. At stake is the masterful consciousness, the capacity to control meaning by intellectual integration. Even though control is to be accomplished in a sophisticated and enlighted fashion, even though it is understood that meanings are neither essential nor eternal, but volatile and open to change, Benveniste leads us away from semiotics and in the direction of philology, or evolutionary linguistics.

It is precisely this Benvenistian desire for clearly understanding meanings that is threatened by the semiotic revolution. For the sign presents itself first as some sort of a ‘meaning’, but a ‘meaning’ that is not understood, misinterpreted. To see the sign as ultimately ‘meaningless’, as Barthes does, is already an interpretation in defense of Understanding. Post-structural semiotics’ romance with Freud is the first systematic attempt to deal with neither Reason nor Understanding, but Misunderstanding.^*

Julia Kristeva sees the desire for synthetic integrative understanding as a weakness, an appeal, in effect, to transcendence (1975a: 19). Her work is still well within the limits of a semioanalysis that regards the signifier as primary and the signified as a mystification; but her position dethrones the importance of the intellect, or the critic, as the means of decomposing the sign. For her, the sign is subject to dissolution through a process of its own. As with the archetypal human composition—ego identity—the sign is prey to loss of composure; it can decompose under the pressure of eruption of expressions from an unconscious level. “La sémiotique” (1975c:17), whose most notable example is poetic language, disrupts the symbolic order, that falsifying set of social contracts or shared meanings that has served to disguise desires and drives. Desires derived from the drives, that is, are totally heterogeneous to the sign: signs are, in effect, the sign of desire’s exclusion from expression. Here Kristeva expands upon Lacan’s hypothesis that it is the acquisition of language, the entrance into the symbolic order during the “mirror stage,” that cuts the subject in two and relegates the drives to the unconscious or the inexpressible. Yet these same drives continue to operate to reconstruct the subject and to cross over the line: they appear in the form of psychopathologies or partially expressed feelings the totality of which Kristeva terms the semiotic (1975c:15-17). The apparent lack of motivation of the sign in de Saussure (it is arbitrary, but so is all of language—recall the remark, “It has no reasonable basis”) is radically rejected in the Kristevan analysis because for her the semiotic is essentially the space of unconscious “reasons,” motivations by desires and drives viewed from the existential point of view of the signifier.²⁵ The difference between desires (and drives) and rational reasons is that the desires can never appear in the pseudo-totality that is the sign. In a sense, Kristeva’s work attempts to correct the incipient transcendentalism we have witnessed in Benveniste and the Barthes of the sixties. It might be added that Barthes himself had begun to acknowledge the importance of unconscious desires in literary expression (“logothesis”) and that he had turned to the search for the basic signifier of this desire in literature—what Steven Ungar has called an erotheme to correspond to the phoneme (1977:64).

Why retain the notion of the sign? The great work has been the liberation of the signifier from any link to the mystification of the signified, to the pretentious and “scientific” concept. Even though this liberation has tended to hypostatize the signifier and to inflate, ever so slightly, the role of the critical intellect, surely Lacan’s and Kristeva’s correctives are beneficial, and retaining the sign appears to be a regression. In some instances it is. However, two major figures who retain the sign will allow us to see the very different directions in which the sign (as distinct from the signifier) can lead semiotics.

The Meaningful Sign and the Indicative Sign

One conception of the sign is that it is an essence, of a Husserlian type—a monad, basically immaterial and spiritual—that its manifestation is clearest in the work of art, and that its semiotic system is the totality of artistic achievement.

Gilles Deleuze’s book Proust and Signs comes as close as any work one can think of to epitomizing the idealization (some would say the ideologization) of the sign:

Deleuze, of course, implies that this is a rendering of Proust’s own theses; we shall not take time to argue with this debatable hypothesis. No matter who is promulgating the postulates here, Proust or Deleuze, the voice that addresses us aims at anonymity, at transcending personality. As such, it is legitimate to treat the work as an essay on signs, not on Proust, and to consider the semiotic it proffers.

Here is a semiotic in which the sign remains intact: at least nominally. But any sign that seems to have a worldly element in it is rated inferior to the spiritual sign: “worldly signs,” the “signs of love,” and the “sensuous signs” are a waste of time (20). Yet we know that in order to be a sign there must be a manifestation: this is the work of art in the first citation. In order for the work of art to be a superior sign, it must be spiritual:

Delineations of spiritual ascent to the truth, the meaning of signs, appear in Deleuze’s text (pp. 85-87). He gives the spiritual disproportionate attention, even though many of Proust’s citations belie this emphasis, and even though Deleuze admits that the body involuntarily betrays material truth (the body is for him another “kind” of intelligence). The sign appears to be the essence of truth, true meaning, and it also appears as a kind of monad, unity, existing by its distinction from the degraded forms of other signs. Deleuze/Proust recognizes, however, that to be a sign it must address someone, it must communicate, but not in the degenerate form of interpersonal communication. The work of art is a “formal structure” and self-sufficient, inventing a code of its own (149; Deleuze relies on Umberto Eco’s terms here) and yet it could remain a mere structure with no formal unity between its parts. The capacity of one part of a structure to communicate (and form a “unit”) with another Deleuze calls “transversality” (a term borrowed from the psychoanalyst Félix Guattari): the structure can be unified and totalized by the intercommunication of its parts. This intercommunication has two other parallels: the transversality that articulates a single work of art to other works by the same artist; and the transversality that links the work to the public and to all other works of art (150). At all times the notions of address, communication, and codes are given a nonempirical interpretation—texts can influence each other in the absence of empirical or demonstrable activity. (We use the word “influence” advisedly here, because of the suggestion of spirituality it has always carried.)

Kristeva calls Deleuze and Guattari’s analyses “lacanisme de droite” (1975a:33), and if there is any correctness at all to this characterization, it stems from the hypostatizing of the sign that is effected in their analyses. The materiality of the sign is either left out or totally swallowed up, so that the body itself becomes only its essence—desire. In later texts Deleuze and his colleagues will extend this essentialism and make of desire the motive force for all meanings and systems: identities, groups, institutions, economies.²⁶ Once the essence has been perceived, like the Cusan God, as being everywhere and nowhere, one can redescend to the worldly and discover traces of the essence in its signs (Deleuze and Guattari 1977a:88).

The spirituality of this version of the sign, the sign as embodiment of an essence or spiritual meaning, has a curious outcome politically. Jean-Marie Benoist (1970:15-17) accuses the Rebels of 1968 of having been mesmerized by signs, and signs of a very Deleuzian sort—signs imbued with essence, transsubstantiated signs, in which the God has become the demon, Descartes’s malin génie.²⁷ This type of spiritual reading of the sign is only possible if one makes the essentializing move; taking the sign to be meaning ful and its meaning available only to those prepared to hear it speak. And although we must disagree with Benoist’s interpretation of the events of May 1968, it is clear that the politics of the sign have been taken by many in exactly the way he has defined them: even the book of graffiti on the events is entitled “Les murs ont eu la parole” (Besançon 1968). The Deleuzian sign is, in fact, hardly a sign in semiotic terms: the weight is all on the signified. Gone is the material aspect of the sign, Saussure’s insistence on the mutual determination of image and concept (both are chaotic outside their relationship to each other [Saussure 1966:112]); gone is the limited and finite aspect of the sign that distinguishes it from the abstract concept; which is capable of infinite transformations, like a god, the contrast between sign and concept pictured so vividly by Lévi-Strauss in The Savage Mind (1966:18). The impulse toward meaning that does not depend radically upon its sign character (signifier and signified)—and a meaning at that, one meaning, desire—is in every way counter to the semiotic revolution.²⁸

But there is another approach to the sign, the one made by Derrida that releases its revolutionary potential. In his early critique of Husserl’s theory of signs Derrida was able to discover the alternative to the meaningful sign in and alongside of the “central” hypothesis of meaningful signs. In an aside, in the margins of his discourse, Husserl provides a theory that signs, though signs, do not mean by being filled with meaning. The sign as mere mark (Anzeichen), whose essence is not meaning but only potential readability, avoids both the impasse for the critics of the signifier (its tendency toward transcendentalism) and the Deleuzian hypostasis of meaning described above. Marks, notations (even the musical notes expelled from semiology by Benveniste), have all the attributes Saussure gave to the sign: they have value (1966:114) by virtue of their differential position vis-à-vis other such signs in the finite system in which they occur; and they acquire meaning by a different sort of difference, that of deferred interpretation after the fact of their presentation. On the margins of Husserl, Derrida discovers this theory of signs that are not-yet-full of meaning; and he goes on to draw out the latent radicalism of Saussure’s sign on the basis of his discovery.

Derrida takes Saussure literally when the Swiss writes of the absolute implication of the concept in the signifier and vice versa. In a move quite distinct from the Barthesian/Kristevan et al. move, Derrida does not decompose the sign into signifier and signified and eliminate the latter from semioanalysis. The splitting of the sign—salutary when the signifier retains primacy, suspect when the signified does—limits one to Discourse on the one hand (no matter how broadly conceived by Barthes, Kristeva, Foucault) and to ineffable spiritual meaning on the other.

Conclusion

We have seen how the splitting of the sign has been in part occasioned by an incapacity to control the concept or signified, which has, according to Lévi-Strauss, an unlimited capacity for transformations. The vision of meaning running unchecked or uncontrolled has tended to lend prestige to the signifier (as material image) as counterweight, ballast. But this resort to the imagination as opposition to the rational “concept” remains within the limits of Romantic Idealism’s appeal to the imagination, falling short of Saussure’s overcoming this particular concept/image opposition in his definitions of the sign. For Derrida, such a move is unnecessary if the sign is taken literally enough: the concept is enmeshed in its signifier, there is no concept without its signifier. But more than this, the concept is not single or unique. For Derrida, every signifier in a sense is overdetermined. Its signified is multiple, though not infinite. Meaning does not run uncontrolled because of this multiplicity in Derrida’s disseminational view: those meanings (signifieds) that are most manifest are always potentially eroded by the latent meanings their manifestation has provisionally repressed. For Derrida signs are dis-unities, as in Peirce and Saussure; they are ambiguities (a term, oddly enough, with which Barthes will almost never deal). The latent meanings repressed by the most manifest one are prior readings or interpretations of the sign. Moreover, it is when a sign appears to be insignificant that its latency can be uncovered. It is for this reason that Derrida, a la Freud, tracks the margins, the “asides,” the signatures.

We can illustrate the revolutionary implications of Derrida’s model of the sign, by again taking up an issue that has produced a great deal of confusion within semiotics: namely, Derrida’s alleged failure to comment on Marx or to relate his grammatology to Marxism. Derrida’s replies to critical inquiries on this point have always rung with ironic ambiguity.²⁹ We think this is because Derrida knows his Grammatology to be something more than a footnote to Marx. It is a radical re-reading that pulls out Marx’s semiotic. Recall that Marx, like Nietzsche, discovered the sign in modern culture: the original equivalency character (metaphor) of the commodity and its secondary capacity for endless substitution (metonymy). Once the reading of Marx is reframed in this way, his analysis of money appears to deviate from the overall plan of Capital. Money, according to Marx, is a double abstraction: the representation of a representation,³⁰ or the sign of a value which is itself already a sign of equivalences. In short, Marx failed to do a Marxist analysis of money, that is, money as a commodity produced under material conditions.

A Marxist analysis of money would lead to the discovery of the merely representational character of the stock exchange, against which Marxists have directed so much criticism to no avail. The stock exchange is a dramatization, a spectacle, a diversion. The real work of making money at the mint controls the course of economic life. Engraving, printing, writing the money, increasing the “money supply,” are the moves that give the central government its legitimation, creating “authority” out of “authorship.”

To our knowledge, no Marxist has studied the making of money itself—indeed, such a study would make no sense outside a semiotic rereading of Marx, even though its salutary implications for revolutionary praxis are easy enough to see, once stated. Derrida is clearly trying to pull us in this direction. He has titled the first chapter of Grammatology “exergue,” a term that means “hors d’oeuvre,” ex-ergos, outside the main work. It also means the mark on a coin or engraved medal that gives the signature or sign of the engraver and/or the date and place of its having been struck Derrida’s first chapter is framed as an implicitly Marxist semiotic wherein he treats the “circulation” of signs, “values,” and the “inflation” of the sign, fully integrating these with an advanced concern for the exploitation of man by man. (Althusser’s works conflate Lacan and Marxian semiotic terms.³¹ For example, see the chapter in For Marx (1977:89-127) on “Contradiction and Overdetermination,” which is in effect a reading of the Russian revolution as a mixed historical sign of the Derridean sort.)

Close attention to the bounded multiplicity of meanings that a signifier can have and has already acquired prevents the radical sign from participating in the a-historical tendencies in the bourgeois hypostasis of the sign. Ironically, it is the mixed and thoroughly historical sign that is rejected by the “New French Philosophers,” the darlings of Time magazine—Glucksmann, Lévy, and Lardreau³²—even though the leading academic followers of Derrida in America have claimed them to be Derridists (Spivak and Ryan 1978). This new non-philosophy is little more than a series of melodramatic pronouncements aimed at obfuscating the discrimination of signs: “Marx is dead”; “Russian bombs and U.S. B-52’s are all the same”; “the camps, Gulag, the camps!” The most overworked term in the writings of this group is: ‘pure and simple’. If there is any single feasible characterization of a semiotic, it is that it is neither of these.