Semiotics and Literary Studies
Floyd Merrell

Saussurean semiology came into its own during the 1950s and 1960s, and in the 1970s it began giving ground to the exceedingly more inclusive semiotic concept of the sign developed by Charles S. Peirce. While in literary studies the Saussurean view has generally held rein, during the past two decades attention has turned increasingly toward Peirce. Much work remains for the enterprising scholar, however.

Key Words: Semiology, Semiotics, Intertextuality, Interdependence, Interrelatedness

PRELIMINARIES

Before I proceed, I must distinguish between: (1) "semiology," based primarily on the linguistics of Swiss scholar Ferdinand de Saussure, that during the 1950s and 1960s found popularity chiefly in the continental European tradition and in language and literature departments in the United States, and (2) "semiotics," more recently emerging from the work of North American philosopher, scientist, logician, and mathematician, Charles Sanders Peirce. Unfortunately, a distinction between the two terms has often been blurred. Students of the late A. J. Greimas prefer to call themselves "semioticians," though they fall within the continental tradition. Numerous other investigators working within the "semiological" framework do the same. Occasionally, the continental concept of the sign involves a somewhat forced wedding between Saussure and Peirce–Umberto Eco (1976) is a case in point–with the best man appearing in the guise of French linguist Emile Benveniste, Russian linguist Roman Jakobson, or Danish linguist Louis Hjelmslev. Be that as it may, not an insignificant number of scholars continue to use the term "semiotics" interchangeably with "semiology," "structuralism," and sometimes even "poststructuralism," which brings on more confusion than illumination.

For example, Terence Hawkes (1977:24) proclaims in Structuralism and Semiotics: "The terms semiology and semiotics are both used to refer to [the ‘science of signs’], the only difference between them being that semiology is preferred by Europeans, out of deference to Saussure’s coinage of the term, and semiotics tends to be preferred by English speakers, out of deference to the American Peirce." Shortly thereafter, Hawkes contends that the boundaries of the "field of semiotics," if indeed there be any, "are coterminous with those of structuralism: the interests of the two spheres are not fundamentally separate and, in the long run, both ought properly to be included within the province of a third, embracing discipline called, simply, communication. In such a context, structuralism itself would probably emerge as a method of analysis linking the fields of linguistics, anthropology and semiotics" (Hawkes 1977:24). Regarding Hawkes’s sweeping assertions, Thomas Sebeok (1986:80) judiciously warns: "Nothing could be a more deluded misconstrual of the facts of the matter, but the speciousness of this and associated historical deformations are due to our own inertia in having hitherto neglected the serious exploration of our true lineage."

I harbor no pretentions of being able to lay this "semiology/semiotics" conundrum to rest for all time. Rather, I intend to elucidate the problem, and let the chips fall where they may.

THE DISTINCTION

I will focus on the boundary between "semiology" and "semiotics" in this essay by pitting Peirce against Saussure, while revealing my bias in favor of the former over the latter. Given limited time and space, I must present Peirce as economically as possible. This requires my ignoring some of Peirce’s terms that would otherwise become overloaded with undesirable conceptual baggage, paring down parts of his sign theory in order to streamline it–but hopefully without doing it irreparable damage–and bringing "semiotics" to bear on our contemporary multicultural scene. In other words, I will attempt very briefly to present Peirce’s concept of the sign, while highlighting those aspects of it that remain relevant to our contemporary world.

My translating Peirce into our own culture-world is a necessary step, I believe, for our culture-world is what we made it by means of the signs we have fashioned. Each morning we awaken to William James’s "blooming, buzzing confusion," and, as Marcel Proust so aptly describes it throughout his Remembrance of Things Past, we gradually become aware of the signs of our culture-world as our consciousness unfolds, opening ourselves to its environment. And we reinitiate our navigation along the stream of semiosis, following the current as it meanders along, twisting slightly when entering its gentle eddies, bucking with the whitewater during its less benign periods, steering between bounders and fallen trees, and all the while producing and processing an untold profusion of signs. But it is not simply a matter of us and our signs. No sign is a full-blown sign without all signs, for they are all interdependent, and they incessantly engage in interrelated interaction with one another. Moreover, what we take to be "our" signs is virtually nothing outside the entire community of signs producers and processors to which we belong. All signs and all sign makers and takers compose a virtually seamless fabric: it is not a matter of signs and things but of thought-signs in the mind and sign-events "out there." It is a matter of signs perpetually becoming something other than what they are.

First and foremost, the idea Saussure discarded but some investigators continue to hold dear is that of signs stand for something else, as surrogates of some sort of secondary status replacing genuine articles. Admittedly, Peirce used the term stand for–as well as represent and refer to–with regards to the relation between signs and their respective objects. Quite simply, they were the terms of his day. But he used them as he saw fit, a use that diverged, at times quite sharply, from their customary nineteenth-century and early twentieth-century usage.

I choose entirely to eschew the use of stand for (and correspond and represent and refer to) in this essay for two reasons. First, the stand for idea breeds a tendency to conceive a sign as a sort of static proxy standing in for an equally static thing, the sign. The sign as proxy cannot properly carry out its role of incessantly becoming other signs, along the flow of semiosis. With each new instantiation a sign has invariably become a difference; it has become a new sign, not merely the same sign standing for the same object or event. At the same time, with each new instantiation, although the sign is now something other than what it was, it nevertheless contains itself (as a trace) within itself. So the sign is, but from another vantage it is not, what it was. Second, the stand for idea tends to generate an implication of the sign function as immediate, rather than mediate. During his lifetime Peirce worked at developing the idea of the sign’s mediary role, for he believed there is no immediacy of the sign process of which we can be conscious here and now–in this respect Peirce was in line with Jacques Derrida’s (1973) argument against the "myth of presence." The concept of mediation denies our making and taking signs and their respective objects as they are in the here and now. We do not perceive and conceive our culture-world exactly as it is, but as it was in a moment now past in the river of time, and by way of mediating signs.

For example, the word "bachelor" does not stand for, refer to, or represent the collection of all unmarried men, past, present and future. The French flag does not stand for, refer to, or represent a political entity that happens to go by the name of France. And smoke does not stand for, refer to, or represent fire. Rather, "bachelor," the French flag, and a cloud of smoke are signs that, upon their being interpreted by semiotic agents–human in this particular case, but any living organism will do–interact and interrelate with other signs "out there" (as sign-events) and in the minds (as thought-signs) of those agents. In fact, those agents, upon interacting and interrelating with signs and other agents, become, themselves, nothing more than signs among signs.

Consequently, in contrast to signs standing for, referring to, or representing things, more properly speaking they interdependently interrelate and interact with them. Signs interdependently interrelate and interact with other signs in the same way they interdependently interrelate and interact with their meanings and with their makers and takers. There are no intransigent priorities here, no hierarchy of values, but rather, a rather democratic process of signs becoming other signs, of signs taking their place among signs, of signs–ourselves included–lifting themselves up by their own bootstraps.

Although I admittedly depart from the "letter" of Peirce’s terminology, I do not abandon the "spirit" of his sign theory. Granted, Peirce endowed the sign with its most general definition as something that stands for, refers to, and represents something to someone (a human or other semiotic agent) in some respect or capacity. However, in order to avoid conjuring up images and thoughts reminiscent of bygone glories and vanished dreams of language as a "picture" of the world, the mind as a "mirror of nature," and of a faithful one-to-one "correspondence" between signs and things, I do away with stand for, refer, and represent entirely. The watchwords are interdependency, interrelatedness, and interaction between signs.

Let’s get on with the Peirce/Saussure problem.

BINARISM VERSUS TRIADISM

Ferdinand de Saussure (1857-1913) has been hailed as "the father of modern linguistics," the man who reorganized the systematic study of language "in such a way as to make possible the achievements of twentieth-century linguistics." He has promoted "semiology, the general science of signs," and "structuralism, which has been an important trend in contemporary anthropology and literary criticism as well as in linguistics." He gives us "a clear expression of what we might call the formal strategies of Modernist thought: the ways in which scientists, philosophers, artists, and writers working in the early part of this century tried to come to terms with a complex and chaotic universe." His theory of language "focuses on problems which are central to new ways of thinking about man, and especially about the intimate relation between language and the human mind. This is an impressive track record indeed!

Charles Sanders Peirce (1839-1914), whose life spans that of Saussure, is a latecomer to the humanities and human sciences in this century. Peirce was truly a polymath. Trained in chemistry, he also studied, logic, mathematics, and philosophy, and to a lesser degree he became versed in the entire range of disciplines that existed during his day. He is the father of "pragmatic philosophy," considered by many to be the only legitimate American philosophical movement. As a scientist and mathematician who acquired an international reputation during his day, he also produced many of the advances in logic (which he equated with semiotics) and scientific methodology, that have made possible a number of further developments ranging from computer science to the history and philosophy of science. Over sixty years ago philosopher Hans Reichenbach (1939) wrote that Peirce anticipated his own groundbreaking work on inductive logic. More recently, Hilary Putnam (1992) was surprised to discover how much that is now quite familiar in modern logic "actually became known to the logical world through the efforts of Peirce." And W. V. O. Quine (1953) places the beginning of modern logic in the work of Gottlob Frege and Peirce. Now these are heady credentials also!

Unfortunately, most introductions to "semiology" or "semiotics" pay homage to their respective founder. Then they reverently follow in the footsteps of the master, be he Saussure, Peirce, or whomever. Quite frequently, the authors of such introductions offer a recapitulation of some sign theory or other–many times reductionistic and equivocal–the exposition and rhetoric of which is often alien to current practices in anthropology, linguistics, literary theory, philosophy, and sociology. Yet the ideas are usually presented in rather programmatic fashion, as if handed down from the gods.

In an attempt to improve on this formula of exposition, I must point out that one of the chief distinction between Peirce and Saussure lies in the scope of their theories. Peirce’s semiotics encompasses the range of all possible signs and their human and nonhuman makers and takers alike, regarding both inorganic and organic, and living and nonliving domains–in addition to what is construed by dualists to be the realm of mind. This all-inclusive semiotic sphere exists in stark contrast to Saussure’s call for a "science of signs," which according to the proper conception was destined to become basically a "linguistic science," thus limited to distinctively human communication. But actually, Saussure was not quite as limited as many of his disciples have made him out to be. His idea had it that since linguistics "would be only part of the general science of semiology," the laws discovered by semiology, circumscribing "a well-defined area within the mass of anthropological facts," would be "specifically applicable to language" (Saussure 1966:16). Semiology according to this broader definition would incorporate all modes of communication found in human societies, including both linguistic expressions and nonverbal devices such as gestures and signals along nonlinguistic channels.

PROBLEMS EMERGE

Visionary pronouncements and the train of future events, however, are often incompatible bedfellows. Not only did Saussure’s dream in the wider sense go largely unfulfilled, but, despite his initial subordination of linguistics to the more general "science" of semiology, throughout the Course he repeatedly contradicted his initial premise. It gradually became apparent that for Saussure, language–that is, what he set up as a language-speech (langue-parole) dichotomy–to the exclusion of writing, occupies a suffocatingly privileged and unique position among all semiological systems. This dubious move, which was later effectively deconstructed by Derrida (1974:1-93), prompted certain semiologists of the 1960s to thrust language to ever greater prominence. The giant step was taken when Roland Barthes (1968:11) declared that "linguistics is not part of the general science of signs, even a privileged part, it is semiology which is a part of linguistics."

In contrast to this "linguicentric" concept of the study of signs, Sebeok (1986:80-81) effectively encapsulates the breadth of the semiotic perspective thus:

Molecular biochemistry, immunology, ethology are only some of the branches of biology which pose approachable puzzles that are ultimately solvable only in semiotic terms. The essence of neurobiology lies in the ineluctable fact that in mind is a system of signs (i.e. tokens), that is, a representative (i.e. typical model of what is commonly called "the world" (Umwelt)). This model (an icon of certain pertinent space/time relations assembled, ever shifted, and reassembled in the organism’s Innenwelt) is then mapped onto the physical fabric of the brain by a hitherto unknown recoding process. In all likelihood, brain will turn out to be yet another system of signs.

Compare this Peircean ecumenism with Saussure’s call for a "science of signs," semiology, since linguistics "would be only a part of the general science of semiology," the laws discovered by semiology, circumscribing "a well-defined area within the mass of anthropological facts," would be specifically applicable to language (Saussure 1966:16).

Actually, in its most basic form, Saussure’s Course consists of hardly more than tentative notes on a method for studying phonetics, and at most morphology, with very little in the way of syntax or semantics, to say nothing of pragmatics. Yet Saussure’s suggestions were propagated by his follows as a veritable doctrine intended to encompass the entire universe of signification. As a consequence, the Course has become lost in a plethora of glosses, commentaries, explanations, offshoots, and outgrowths such that one can hardly separate the wheat from the chaff or see what is touted to be a forest for all the scrubs trying to pass themselves off as legitimate trees. This is quite baffling. Saussure’s original strategy was relatively unambiguous: to project a monolithic, undifferentiated, field, language, gradually divide it into sharp distinctions, and then virtually eliminate one of each set of those distinctions. The result was a set of boundaries and a successive narrowing of the corpus to be analyzed. Obviously, Saussure’s less disciplined followers did not heed the suggestions of their leader: they took the narrowest of parameters and expended them inordinately.

Language, in Saussure’s view, floats in an ethereal zone above the physical world. It is arbitrarily contrived and chiefly autonomous. It creates its own "world," despite the individual language user’s whims or wishes to the contrary. Individual words are not, as they were for philosopher John Locke and many philosophers and linguists since his time, mere markers, linguistic window dressing conveying notions about a "world" whose structure is available to the mind through perception of that "world." The Saussurean "world" is what language says it is, which implies that insofar as language is structured in a particular manner, so also its "world." Thus language consists of a repertoire of signs and the possibility for their use by the speakers of a given community, while thought is a structureless haze lying in wait for language’s cutting it up and organizing it into some sort of order. And both thought and language collaborate and contrive to create a "world," the "world" common to the members of the speech community. But this is merely the first, and quite vague, step toward a grasp of what Saussure is all about. His interest rested almost exclusively on language.

But enough comparison and contrast between Saussure and Peirce. Let’s move directly to a consideration of literature

SEMIOTICS AND LITERATURE

Umberto Eco’s production is overwhelming, even when considering solely his publications in semiotics. Of importance to this essay, Eco’s opus cuts across the two strains of literary semiotics: continental semiology–including French structuralism–and Anglo-American semiotics, following Peirce. Though some scholars claim semiology and semiotics to be coterminous, obviously in light of the above sections, they are not. Yet in literary studies the two terms have often, and unfortunately, been conflated. But allow me briefly to recapitulate the historical development of literary semiotics. Then I will proceed from there.

In the 1930s Nicolai Trubetzkoy, of the Prague School of linguistics, developed a relational method of phonological analysis which aided his colleague Roman Jakobson in formulating his theory of poetics, based on the ideas of Saussure. Of particular interest were Jakobson’s distinctive features and his study of aphasia. In the latter, Jakobson and Halle (1956) focus on two major disorders, those of similarity and those of contiguity. They noted that these disorders were remarkably close to the rhetorical functions of metaphor and metonymy, which entailed the Saussurean paradigm/syntagm opposition. Relating this formulation to the nature of literature, Jakobson (1960) later argued that signs were selected along the "vertical axis" from the available repertoire of possibilities–the code–and then combined along the "horizontal axis" into constituent parts: phonomes, words, sentences, paragraphs, and texts.

During the 1960s, French semiologists–drawing on the Russian formalists, Saussure, Émile Benveniste, and Prague School linguistics–developed this Saussure/Jakobson binary logic of differences further, to give it a privileged place in literature and folklore studies (A. J. Greimas’s "logic" of narrative constraints, Claude Bremond’s "logic" of narrative events, Tzvetan Todorov’s "logic" of narrative actions, and in a broader sense, Claude Lévi-Strauss’s culinary "logic" and Roland Barthes’s "logic" of dress codes). The linguistic model reached a screaming pitch when Barthes (1968:11) inverted Saussure’s notion that linguistics should be subordinate to the general science of signs; he proclaimed that linguistics "is not part of the general science of signs, even a privileged part; it is semiology which is a part of linguistics."

This "scientism"–envisioned by Saussure and, according to dream semiologists, brought to fruition in linguistics by the work of Hjelmslev (1961)–took a sharp turn during the late 1960s and early 1970s. In certain semiological circles, consensus had it that, given the sign’s arbitrary nature, reference could be discarded, to leave a shifting mass of indefinitely permutable signifiers. This transition can be seen in the evolution of Barthes’s work, from Elements of Semiology (1968) and Mythologies (1972) to S/Z (1974). In the last volume, Barthes proposed that art consisted of disembodied and referenceless, but self-reflecting, signs. The text was redefined as indeterminately multivalent, multi-subjective, and multi-disciplinary. The shift was to a relativistic, non-systematic perspective of the sign, the subject, and the text, with renewed focus on the reader. As de Man (1971) put it, the study of the literary sign became divorced from semantics.

Intertextuality later became the buzz word, and reading itself became a signifying practice. Rather than the classical bourgeois readerly (lisible) text, which reduces the reader to a passive consumer, the writerly (scriptible) text is liberating. It forces active participation in the production of literature itself. The reader always remains linguistically (i.e. semiologically) constituted within the broader domain of intertextuality and with a cultural context. Consequently, reading, though indefinitely variable, is never free of presuppositions, but remains caught in the web of meaning, an unanalyzable whole. Textual undecidability becomes the norm–notwithstanding Eco’s (1990) "limits of interpretation."

Kristeva’s later concerns, unlike those of Barthes, project outward to envision social revolution as a direct result of intertextual productivity. Kristeva (1984) draws from structural linguistics, generative-transformational grammar, Freud, and Marx to account for the process of textual productivity from within a structured matrix. What she terms the "geno-text" lies at the surface: a residual crust masking the textual engenderment at the deeper unconscious level of desires that spontaneously erupt into language in rebellion against the constraints imposed by established modes of discourse. Such notions of intertextuality may prove to be one of the more fruitful insights afforded by semiotics. The interconnectedness of all writing reflects neither physical "reality" nor any domain reducible to the logic of possible-world semantics, but a realm of imaginary dimensions. This is the world of textuality that Derrida (1974) extrapolates to its extreme form in his critique of traditional notions of the sign.

The idea of intertextuality also emerges from other directions: Juri Lotman and Mikhail Bakhtin. Lotman (1977) takes a cue from cybernetic theory in developing his concept of the text as a secondary modeling system; its building blocks consist of language, the primary system (but see Sebeok 1991b). The multiply coded aesthetic sign enters into relationship with other signs on various levels: phonological, syntactic, semantic, and contextual. The broad spectrum of culturally contextualized sign systems forms the extratextual background of a work of art. This particular form of sign is privileged, however; following the Russian formalist break between poetic language and everyday language, Lotman argues that "literariness" is overcoding by way of language’s deviation from the norm. Lotman gives Bakhtin a favorable nod as the most reliable spokesman of the semiotic interrelatedness of the artistic text–though the latter remains alienated from the formalist opposition, conceiving of the difference between poetic and practical language as relation rather than absolute.

MORE TROUBLE WITH BINARISM

Indeed, Bakhtin (1973, 1981, Volishinov 1973) launches a scathing critique of language and the sign as defined in the Saussurean and Russian formalist tradition. Against the Saussurean theory of the sign as part of a dead, abstract system, Bakhtin argues that the material life of the sign does not emerge from the physical world, but from actual practice in everyday social life. Bakhtin indefatigably rebels against the very idea of system. Highlighting dynamic historical processes of tension and conflict, he discards univocality, naively assumed to reside in Saussure’s langue, and replaces it with polyvalence, appropriate to parole. Speech, he maintains implies the "otherness" of dialogic interaction. This communicative aspect of language is emphasized by Wilhelm von Humboldt, as well as by certain of Bakhtin’s contemporaries, Karl Bühler and Émile Benveniste. In Bakhtin, however, dialogism becomes supreme. Seeking a minimum of semiotic homogenization, he stresses diversity; the terms "heteroglossia" and "polyphony" bear witness to his obsession for describing the indescribable. Bakhtin seeks to subvert the claim of a monological author’s producing a timeless, univocal, autonomous text. The text reveals itself to be temporal, polyvocal, and historically situated. This view entails constant interaction between self and other–which, semiotically, are composed of the same signifying media, chief of which is language (Holquist 1990).

This is a sophisticated twist on the interconnectedness idea. It also calls up Peirce, whose triadic sign consists of a representamen, and object, and an interpretant. All signs are interpreted in terms of other sign, and those in terms of others, ad infinitum. Likewise, the object in interrelation with the sign–a semiotic rather than "real" object–becomes itself another sign interrelation, which gives rise to an equally endless series of interrelations, each interrelating with the one behind it. Moreover, as with Bakhtin, all thinking is necessarily in signs. No thought can be considered self-sufficient or self-confirmatory; all thoughts necessarily require others to make them clear, in endless series.

Somewhat in this vein, Michel Riffaterre (1983) has conceived the literary text as auto-representational (the text as text) and intertextually relational. The reader, by recognizing the intertext in the text, identifies the qualities which make it a work of art. In other words, and according to Riffaterre’s appropriation of Peirce’s terminology regarding his concept of the sign, the literary sign or representamen is the segment of the text perceived as a portion of the web of intertextuality. The meaning which a text conveys upon being read depends on its capacity for actualizing relations between itself and other texts–potentially, the range of all texts. But Riffaterre’s model does not entail random association. The structure of the text imposes constraints on the intertext; both are variations on the same structural theme. Thus Riffaterre’s decidable text is a far cry from the labyrinthine, undecidable infinitude of relations suggested by Barthes, Kristeva, Derrida, and deconstructionist theories.

So we have completed a cycle. For the early French semiologists the text is a closed, self-contained set of signs "out there" apparently for the taking; but it only reluctantly reveals its hidden secrets. The later Barthes conceives reading to be a narcissistic revelry–an extraction, from within the unlimited domain of textual possibilities, of whatever suits the reader’s fancy. Kristeva’s text draws the reader in, then viciously attacks his or her flank, forcing altered perspectives and modes of behavior. Bakhtin’s dialogic text promises a verbal sparring match between author and reader within a carnivalesque atmosphere. And Riffaterre’s text gently guides the reader through the forest of signs to a relatively determinate end. Peirce, who has generally remained in the background during this transition, offers an intricate juncture of formal constructs, the concept of subject situatedness, and dialogism.

As a result of this shift, the formalist cut between poetic language and everyday language has been met with skepticism during the 1980s. Saussurean theories of the sign as arbitrary, critiqued in the 1950s by Benveniste, are rejected outright by Kristeva and most of the later semiologists: the sign is viewed as motivated by desires, drives, and social conventions. The autonomous, self-sufficient domain of literary signs envisioned by formalism and early semiology is culturally contextualized by, among others, Bakhtin and Riffaterre. And Greimas’s formal linguistic model, still popular in a diminishing number of circles, has generally lost ground. Nevertheless, Saussure’’ basic relational concepts of langue and parole, paradigmatic and syntagmatic, signifier and signified, have become a major influence upon literary theory.

STRICTLY PEIRCEAN SEMIOTICS

Relatively few scholars have written extensively on the relevance of Peirce’s semiotics to literary studies. Notable exceptions, in addition to the work of Umberto Eco (1979, 1984, 1990), are studies by Ora Avni (1990), Jørgen Dines Johansen (1993), Wojciech Kalaga (1986, 1997), Michel Riffaterre (1983), Horst Ruthrof (1992, 1997, 1999), Robert Scholes (1982), John Sheriff (1989, 1994), Kaja Silverman (1983). The Peircean approach sees the text as itself a sign (representamen) evoking other signs (semiotic objects) from the reader’s experience, thus engendering intertextual interrelations that serve to construct an imaginary world of signs (interpretants). This signs in turn become originary signs to be (re)interpreted as the reading proceeds–as well as during successive readings, which then serve retrospectively to reinterpret previous readings.

Reference, construed in the Fregean tradition, becomes of little consequence. The word fox in a text conveys a message from writer to reader in terms of its conjunctive association with other signs with which it possesses something in common. The same can be said of the word phoenix. Although the phoenix does not exist, phoenix descriptions do; this endows the word with the capacity to function semiotically with a force equal to that of fox. In this sense, semiotically "real" objects, like literature and the arts, are fashioned rather than found (Goodman 1979). It is not merely a matter of what there is, but what might be; and this evokes a multiplicity of interpretants–or what would be, if certain conditions inhered. The early semiologists, inspired by Saussure, looked at literature and found structure. Peirce’s conception of the sign entails formal as well as experiential categories, which renders it more comprehensive, though rather unwieldy.

Peirce also shares with Bakhtin the idea that, since all thought is in signs, it is dialogic in nature (Ponzio 1990). Even at its most private and silent, thought is nonetheless a dialogue between the self of one moment and the oncoming self of the next moment. The self is a fabricator (by way of semiosis) which interacts dialogically with itself and others. In this context, Peirce uses the word "sign" in the widest possible sense, as a medium for extending the interconnected fabric of signification. As such, the sign is determined by its object, and determines its interpretant. But in order that the fabric be extended, the sign must have already been embodied in a subject (which implies its own object). This being the case, the fabric remains independent of the sign; yet the sign, through its object, represents the fabric in the way that the latter represents itself to be.

RESERVATIONS THAT FAVORED TEXTUALISM AND LITERARINESS

Within the humanities and social sciences, disturbing problems were revealed in light of work during the 1970s and 1980s by, among others, philosophers Derrida and Richard Rorty (1979, 1982), historians Michel Foucault (1965, 1977), Thomas Kuhn (1970), and Hayden White (1978, 1987), ethnologists Clifford Geertz (1973, 1983), James Clifford (1986, 1988), Marcus (1992), and Stephen Tyler (1987), and philosopher of science Paul Feyerabend (1975, 1987).

What, in the final analysis, distinguishes the literary text, or fiction–and philosophy and science for that matter–from "reality"? Either fiction or "reality," or both, or neither, it seems, rest in the eye of the beholder. One might claim that the literary text enjoys no "real" referent, but offers the mere illusion of a referent; or one might go halfway by proposing that the text embodies a dialectic between fiction and "reality."

Yet, following Peirce, a distinction must be made between the inwardly "real," the outwardly "real," and the fictional (see Dozoretz 1979). Irrational numbers are inwardly "real," though in Western thought they were swept under the rug for centuries. Yet the fact remains that they are governed by a rather strict set of rules. Fictions, in contrast, are relatively free-wheeling: they consist largely of whatever one thinks or imagines them to be. Since, as Peirce believes, there can be no absolute guarantee that a given thought is identical to something outwardly "real"–a sort of counterpart of Williard V. O. Quine’s (1953) "indeterminacy of reference"–at any given moment we cannot with certainty draw a line between a fiction and the world "out there" (Merrell 1995, 1997).

This initially appears to be a concession to the Peircean notion of cultural conventions. Peirce argues repeatedly that sign interpretation is dependent upon public agreement, but that total and unwavering opinion is impossible outside some chimerical community of infinite capacity. In such a case, the semiotic enterprise cannot be interminable. For semiosis, the constant shifting of signs back to other signs, and the pushing forward of signs toward new signs, is a never-ending process. This process favors the notion of the literary sign and linguistic signs of everyday cultural practices, and indeed, all modes of extralinguistic communication, in addition to traditionally conceived signs of formal language and logic. Perhaps quite unfortunately, Peirce scholarship has focused, at times obsessively, on language, logic, scientific method and scientific discourse, and mathematics. Attention to Peircean semiotics with a keen eye turned to culture in general and literature and the arts in particular could prove fruitful.