Go to Cyber Semiotic Institute home page
Go to Course Outline
Go to Lecture: One or Two or Three or Four or Six or Seven or Eight

Critical Semiotics

Instructor: Scott Simpkins

Lecture Five: The limits of "system" and the authority of the encoder.

Assigned Readings:

Umberto Eco,The Limits of Interpretation (Bloomington: Indiana University Press, 1990).

Michel Foucault, "What is an Author?", Modern Criticism and Theory, Ed. David Lodge (London: Longman, 1988): 197-210.

Roland Barthes, "Textual Analysis: Poe's 'Valdemar'," Modern Criticism and Theory: 172-295 and "From Work to Text," Image-Music-Text, Trans. Stephen Heath (New York: Hill and Wang, 1977):155-164.


"We may assume that any social person speaking in his own personality will behave systematically, since experienced language is universally systemic. Therefore, we may study his speech and ask the question, 'What is systemic?'"

J. R. Firth ("Personality" 187)

Restraining Orders

I'd like to pursue Firth's question. For, indeed, what constitutes the systemic? Or system? Or systemicity?

G. R. Kress observes that "For Firth a system is an enumerated set of choices in a specific context" (xiii). This is consistent with the discussion of semiotics, which frequently posits the notion of a "system" of signs. Without the grounding backdrop of systemic order, it seems impossible to imagine signs functioning at all. But, give those signs a field delimited by a constraining logic of some kind, and a system is born.

This system also needs an agent to operate it. A figure granted a directive status: a god, or a force, or - in this case - an encoder. Still, this figure's actions are constrained by the system itself. This is not to say that the agent cannot act independently of the system. It is to say, though, that when the agent does so she is no longer operating according to the system. As long as she moves chess pieces in keeping with the system of chess, for example, she could be playing the game as specified by that system. As soon as she makes a move that chess does not allow, something else is taking place. Erving Goffman distinguishes between two kinds of "guided doing" along these lines (22). Moving a chess piece on the board is simply a physical gesture; but "making a move" in the course of pushing a chess piece on the board entails engaging chess's system. The chess analogy is further illuminating in this respect because the players are able to play a "unique" game in a sense, but they are nevertheless limited to playing out their unique sequence according to the constraints of the game's system. They can encode a significant move, and in this way act powerfully, yet again the system curtails their agency to a significant extent.

Combine these two elements - the system and a cooperative encoder - and a regulated field results. Add the decoder to this model, and even more regulation is assumed. After all, the decoder cannot fulfill his function in relation to the system and the encoder if he does not cooperate with the constraints they presuppose (in the case of the system) or adhere to (in the case of the encoder). Like the aberrant encoder whose "move" violates the system, the decoder who fails to cooperate is no longer acting within the confines of the system. Systemicists talk about "condition of entry" that signals one's participation in a system. (M.A.K. Halliday says: "A system is a set of options with an entry condition...a set of things of which one must be chosen, together with a statement of the conditions under which the choice is available" ["Brief" 3].) Leaving a system could be said to constitute a corresponding "condition of exit." The system, in this sense, possesses the capacity to control an underlying semiosis. It possesses "limits" in two senses, both prescriptive and proscriptive. In one sense, the system prescribes a fixed perimeter within which signification takes place under the dictates of certain specific controls which "limit" behavior (of the encoders and decoders) in accordance with itself. (The "speed limit," for example.) And, in the other sense, it proscribes the activity of participants who may endeavor to exceed its "limits" which cannot be transgressed if the agent is to operate within the system. (The traffic cop who enforces the speed limit.) In other words, the system establishes and enforces agent integration.

Following the pioneering work by Firth and others, systemicists have substantially theorized the system and what constitutes adhering to its "limits". In the case of language, "the crucial factor in the designation of any feature as present in the grammar would...be its assignment to a place in the systemic network," Halliday observes. "A putative feature which could not be shown to contrast independently with one or more others at some point would not be a distinct feature" ("Deep" 96). This can be extended to account for "each feature" which, once "recognized", becomes "a term in a system, which system is located in hierarchical and simultaneous relation to other systems" (96-97). (Take note of this phenomenon later in the discussion on Barthes's analysis of the word "extraordinary" from an Edgar Allan Poe short story.) To Halliday, "the location," moreover, is "'polysystemic': the recognition of a system, and the assignment of a feature to it, depends on the potentiality of contrast in the stated environment" (97)

"Environment" is a crucial component of systemics for both Firth and Halliday, who consider it as a means of identifying a contextual sense of "situation". From this standpoint, Halliday asserts that "it is necessary...to specify the syntagmatic environment, in order to define the point of origin of a system network" ("Deep" 97). While origin is a problematic notion, merely positing it as a provisional starting point is extremely important for systemics. Once a beginning place is designated, additional elaboration can be erected around it. Halliday proposes identifying levels as a means of further refining this systemic model. "One way of defining a point of origin for a system network," he observes, would involve "a rank-type constituent structure" (98). In this conception, "each system, like each structure, would be assigned to a given rank as its most generalized functional environment." This designation also helps to establish distinctions between multisystemic relations. "Rank" functions importantly as the "initial identification and labelling of certain stages in a constituent hierarchy in...general terms." As a consequence, Halliday adds, it "makes possible the assignment of a system to a place determined solely by constitutent status (e.g. all clauses) and allows further specification of the environment to be in terms of features" (97).

The "system" obviously offers a great deal of potential organizational framing, as someone like Umberto Eco demonstrates. Since he often promotes a conservative semiotics based on sign restraint, the "system" provides him with an ideal conceptual grid. He maintains that "texts are the human way to reduce the world to a manageable format, open to an intersubjective interpretive discourse" (Limits 21). Of course, underlying this statement is the text-system that makes textualization possible to begin with. He views semiosis, by extension, as constrained "according to a certain ground" which is not unlike the notion of "environment" (28). This ground operates as a means of, again, imposing a type of "open" limitation which enables Eco to posit the relative autonomy of the sign user while, at the same time, enforcing a margin of error, so to speak. Literary symbols, for example, are "paradigmatically open to infinite meanings" in this respect, he claims, "but syntagmatically, that is, textually, open only to the indefinite, but by no means infinite, interpretations allowed by the context" (21). Indefinition, for Eco, evidently means a wide array of valencies, but not so wide, or "infinite", as to have no systemic perimeters whatsoever.

The System of Systemics

By identifying "the system of systems" (Structural 119), A. J. Greimas highlights an often overlooked component of systemics. An overview of representative position statements regarding it may help to identify some of the salient elements that constitute its system. While usually described as framing entities consisting of highly developed organizational components, systems can obviously vary a greatly in nature and degree of complexity and order. Greimas also provides one of the more flexible depictions of systems as merely "ensembles of signification" (118) that can be described through a process he depicts as "systematics" (119). This description, he contends, can contribute to a better understanding of what he calls elsewhere "the semiotic grammar system" ("Interaction" 62).

Systemic methodologies have shared close alliances with the discussion of semiotics. So much so, in fact, that Robin Fawcett and others have come to refer to it as "systemic semiotics" or "systemiotics" (xviii). Eco, who is not usually identified among systemicists, nevertheless displays implicit alliances with them. While this can be seen in numerous examples, it is clear when he opines that "It is possible to hypothesize that for every text there is a system which organizes the possible inferences of that text, and this system can be represented in an encyclopedic format" (Limits 260).

Because Eco spends so much time discussing systemic limitations in his collection of essays published in English as Two Modelsof Interpretation, I will draw upon it extensively here to illustrate a systemic approach to literary semiotics. The book is described on its back cover as opening with "four theoretical essays dealing with various aspects of interpretive theory." The remaining essays "apply" this theory and highlight his emphasis throughout on the need for the decoder to exercise "interpretive prudence" (162). In keeping with this lecture's exploration of the implications of systemics, I would like to treat Eco's application of his "theory" as synonymous with the employment of a "system". In these four essays ("Two Models of a priori," "Unlimited Semiosis and Drift: Pragmaticism vs. 'Pragmatism'," "Intentio Lectoris : The State of the Art," and "Small Worlds") Eco once again consistently establishes a regimental operation for semiotics, as the examples cited in this lecture demonstrate.

From this viewpoint, Eco argues that through a systemic analysis of a literary text, it is possible to discern "the way it works"(Limits 57).. This claim is made, in part, to use systemics to establish a regimen of literary semiotics that is empowered by order and voluntary subservience to it by the decoder - and even the encoder, for that matter. This is dramatized in Eco's discussion of Richard Rorty's commentary on using a text as opposed to merely interpreting it. (Rorty believes that there is no immanent meaning of a literary text, so therefore what a decoder can do with it seems to be the only defendable position to adopt regarding decoding practices.) Eco feels otherwise, however. "To critically interpret a text means to read it in order to discover, along with our reactions to it, something about its nature," he suggests. "To use a text means to start from it in order to get something else, even accepting the risk of misinterpreting it from the semantic point of view". A decoder, within Eco's dynamic of condition of entry, is "in" the system when "interpreting" a text according to the system's rules. To recall the earlier discussion of this point, someone who goes outside the system is merely "using" the text (in the negative sense of "use" as in bad-faith manipulation). This formula of inclusion/exclusion is a central component of some systemics since the notion of system necessitates establishing its boundaries in order for it to acquire the status of system to begin with. No inside/outside distinction, no system, in other words. This perspective is not unlike Eija Ventola's explanation of systemics derived from James R. Martin's work: "As a text unfolds from a generic element to another, the FIELD orientations in elements are hypothesized to be realized on the discourse stratum by lexical structures generated by the choices from the LEXICAL COHESION system network" (131).

As will be seen, systemics relies on a mechanism that requires both decoder and encoder subordination. This positioning is realized through the contention that these two kinds of agents can only make choices while operating within the confines of the system. Halliday positions the concept of choice as constituent of systemicity: "Whenever we can show that, at a given place in structure, the language allows for a choice among a small fixed set of possibilities, we have a system" (Linguistic 30). "The system...formalizes the notion of choice in language," Halliday maintains ("Brief" 3). "The underlying notion in the grammar is that of choice, and this is represented through the concept of a system, which is a set of options together with a condition of entry" (6). This can also apply to the nature of options available as "range of choice," which Halliday describes as "the number of terms in a system, the number of contrastive possibilities among which a choice is made at a particular place in structure" ("Typology" 180).

Eco's version of this is "conventional rules" that as sign users we "share" (Limits 2), which again reveals his inclination toward placing the decoder in a lesser position of power in relation to system. (This would be akin to Noam Chomsky's concept of "complex rule systems" [5].) Halliday's stress is on the options available to the language user, whereas Eco uses this same model component to restrict choice to social agreement at the expense of the individual. This latter endeavor is characteristic of many of the potential abuses of systemics.

Like the charges of antihumanism levelled against structuralists, similar accusations could be made about the sacrifices involved in aligning oneself with what Firth called "systematology" ("Use" 34). Louis Hjelmslev identifies a parallel reluctance, as when he observes: "In certain fields a tendency to systematize may be observed, but history and, along with it, the humanities as a whole still [circa 1943] seem to be far from willing to recognize the legitimacy and possibility of any such systematization" (9). Yet the costs of this paradigm acceptance come, perhaps, as a result of narrow conception of systemics. For, it can just as easily empower the individual sign user in a way that could not be accomplished independently of systemics. On the obligation to pursue an enlightened systemicity, Hjelmslev remarks:

The individual act of speech obliges the investigator to encatalyze a system cohesive with it[;] the individual physiognomy is a totality which it is incumbent on the linguist to know through analysis and synthesis - but not a closed totality. It is a totality with outward cohesions which oblige us to encatalyze other linguistic schemata and usages, from which alone it is possible to throw light on the individual peculiarity of the physiognomy; and it is a totality with inward cohesions with a connotative purport that explains the totality in its unity and in its variety.(126)

A reluctance to abandon autonomy for the greater advantage of assuming a role in a system is certainly understandable, and may explain some of the resistance to systemicity. Hjelmslev's description of systemic implications reveals this when he depicts system as "an organized totality with linguistic structure as the dominating principle" (8). This structure, he adds, is characterized by its "aggregating and integrating constancy." There seems to be little potential for exercising individual agency within the collective assimilation into such a system.

But there are myriad advantages to the ordering capacity of sysemicity that need to be considered. Hjelmslev observes (again in 1943) that, in the course of "reject[ing] the idea of system," linguistics "has failed to carry analysis through to the end, to make its premisses clear, or to strive for a uniform principle of analysis, and it has therefore remained vague and subjective, metaphysical and aestheticizing, to say nothing of those many occasions when it has entrenched itself in a completely anecdotal form of presentation" (10). "The linguistic theoretician" who, according to a systemic methodology, "sets up a calculation of all the conceivable possibilities within certain frames" (17) may, in fact, succeed in producing a "general calculus" of a language system (18). As John Stewart notes, this relational schema is central to contemporary semiotics as it is based on the presumption of systemic intelligibility. "One reason Saussure focused on language as a system," Stewart concludes, "is that he wanted to emphasize how each linguistic unit is meaningful only in relation to the other units making up its system" (19).

Hjelmslev's hypothesis is related to that developed by Firth, who saw the systemic potential behind assessing elements in a system as "terms". "The phonetic analysis of a language does not consist in merely 'collecting' the sounds, and placing them in universal descriptive phonetic pigeon-holes with a specially appropriate letter attached to them," he asserts ("Use" 34). Instead, it identifies them as terms which he describes as "integral parts of the whole phonological system of the given language." "Distinctions" could accordingly be made to speculate on identifying "a place [for each sound] in the whole phonetic structure or system" (35). In Firth's view, systemics offered a significant explanatory force. By "apply[ing] systematic categories to the statement of the facts," systemicists would not necessarily discover "one closed system," he says ("Personality" 187). But, they can "separate from the mush of general goings-on those features of repeated events which appear to be parts of a patterned process, and handle them systematically by stating them by the spectrum of linguistic techniques."

The seeming inclination toward systemic reification is possibly its greatest shortcoming. Yet many systemicists acknowledge this tendency and include a conscious tentativity and flexibility to guard against it in their analyses. Halliday depicts language as "a 'metastable' system" whose very tenacity results from it being "constantly in flux" ("Systemic" 7). "This does not mean that we cannot characterize a particular language," he notes, "but that our characterization of it has to incorporate this feature." "Systemic theory is more like language itself - a system whose stability lies in its variation."

This openness can be said to apply to choice options in this same vein. John Wilson asserts "systemic choices from within language as an independent system allow us to create our world, not as a given, but...as something we constitute or create through our talk in interaction" (280). These choices, then, are indeed potentially autonomous actions and not just a limited ensemble hermetically separated from the larger universe of signification. Systemicity as a whole can be view in this light as "a systemic set of options that function in certain structural positions to indicate a set of meaning potentials," Wilson observes. Accordingly, "meaning is not imposed" in this schema, "but negotiated relative to a particular choice from a set of choices" (280-281). Wilson thus considers systemics as analyzing "rule potential." And, these potentials themselves are always open to change. "A certain [chess] piece is only a queen if defined by a freedom of movement not available to other pieces," he says. In effect, "this basic and core principle of systemic views of language does not tie analysis to objects in the world, or representations in the mind, but rather to abstract rule formulations that gain/give meaning in their being worked out" (281).

Systemizing the Encoder

Michel Foucault's "What is an Author?" mirrors Eco's view by stressing economic caution in the engagement of semiosis, as if limitless signification would ultimately become unsustainable otherwise. The system would thus necessarily police its reserve which, from this viewpoint, is finite and cannot afford an immoderate exchange of semiosis. One way to do this, Foucault suggests, is to presume the effective agency of what he calls an "author-function" (202).

Using analysis of literary sign systems as an example, Foucault contends that "it is a very familiar thesis that the task of criticism is not to bring out the work's relationships with the author, nor to reconstruct through the text a thought or experience" (198). "But rather," he charges, "to analyze the work through its structure, its architecture, its intrinsic form, and the play of its internal relationships." Instead of actively promoting this position, however, Foucault merely suggests that the author (or sender/encoder/enunciator, in the case of the present discussion) is a potentially useful facet that has been needlessly neglected. The author is potentially significant not only as the initiator of a given message, but also - and more importantly for Foucault - as a means of curtailing promiscuous decodings of it. The author thereby helps to impose the "limits" of the system.

The author's name can therefore serve as "the equivalent of a description," Foucault contends (200). And it is this descriptive function that enables a potential restraint over the free-floating elements within the overall process of signification by "assuring a classificatory function." The author's name not only "permits one to group together a certain number of texts, define them, [and] differentiate them from and contrast them to others" (201). The author serves as a locus for organizing a system. One such component of organization - perhaps the primary one for an author - would be consideration of a collective body of literary works. For, the author-function also "establishes a relationship among the texts," Foucault asserts. Furthermore, it "serves to characterize a certain mode of being of discourse" as "a speech that must be received in a certain mode and that, in a given culture, must receive a certain status." It "manifests the appearance of a certain discursive set and indicates the status of this discourse within a society and a culture." Within the author-function, the author's name "seems always to be present, marking off the edges of the text, revealing, or at least characterizing, its mode of being."

Along these lines, Foucault contends, the author can be viewed as "a field of conceptual or theoretical coherence" (204); a system, in other words. The author can thereby be utilized to establish a "principle of a certain unity of writing." In effect, this figure supplies "the basis for explaining not only the presence of certain events in a work, but also their transformations, distortions, and diverse modifications." This could be accomplished by drawing upon the author's "biography, the determination of his individual perspective, the analysis of his social position, and the revelation of his basic design," along with other potentially relevant components of the overall literary system. Most important for literary semiotics, however, is that the systemic author is just that. Suddenly, an amorphous and evidently uncontrollable aspect of semiosis is tamed, rendered into a network of structural connections with a rich and somewhat definable nexus of origins at its center. No longer an unaccountable, unknowable entity, the author-system, as I will call it, assumes the status of a knowable quantity.

Ultimately, Foucault reveals several motivations for granting the author-function the status outlined above. One is that it could help to construct "a typology of discourse" (208), or yet another manifestation of what I have been calling "system". I would argue, though, that for Foucault, the most important advantage is that the author-system could "reduce the great peril, the great danger with which fiction threatens our world" (209). Oddly enough, it appears that, at least to a certain degree, Foucault is chastising those who rely on systemic control as he elaborates on the cultural anxieties behind the use of the author-function. "The author," he concludes, "is therefore the ideological figure by which one marks the manner in which we fear the proliferation of meaning." "It would be pure romanticism," he declares, "to imagine a culture in which the fictive would operate in an absolutely free state." Within this chaotic vision, "fiction would be put at the disposal of everyone and would develop without passing through something like a necessary or constraining figure." As Ronald Schleifer points out, in Vladimir Propp's narrative typology, "the sender (the king, the elders, and others) usually represents the social order" ("Introduction," Structural xliv). The encoder assumes a similar status as social corroborator within the sign system. For, the individual who initiates a message can be said to establish and engage a condition of entry with an accompanying invitation to others also inclined toward semiotic sociability.

Consideration of the author-function alleviates the threat of "the cancerous and dangerous proliferation of significations" as it "allows a limitation" on semiosis, Foucault opines. This prudence is a necessity "within a world where one is thrifty not only with one's resources and riches, but also with one's discourses and their significations," he argues (209). To return to issues of systemic economy, in this respect, "the author is the principle of thrift in the proliferation of meaning." The author provides a function in this schema "by which one impedes the free circulation, the free manipulation, the free composition, decomposition, and recomposition of fiction." To Foucault, the author serves as a surrogate parental function in a given sign system, maintaining semiosis to keep it in check. Hjelmslev also had earlier identified a "principle of economy" that reflects this desire for regulating semiosis (60-1) This can be seen, too, in Eco's later observation about the regulative faculty of "rules which allow a contextual disambiguation of the exaggerated fecundity of symbols" (Limits 21).

It may be, however, as a result of an emphasis on structural analysis over consideration of the author, that this systemic component could eventually be superseded, but only by something with a comparable restraining order. While Foucault observes that in modern times, "the author has played the role of the regulator of the fictive," he speculates on the eventual passing of the author-function. This could transpire "in such a manner that fiction and its polysemic texts will once again function according to another mode, but still with a system of constraint." This "system", he suggests, "will no longer be the author," but it "will have to be determined or, perhaps, experienced" instead (210). Note, however, that Foucault cannot allow semiosis to occur outside of a systemic boundary of some kind. This new, ineffable replacement of the author-function would thus remain aligned with systemicity, but in a new way.

The Death of the Encoder

The author-function essentially turns the encoder into an integral part of a system, resulting in the author-system. In this manner, the encoder effectively loses the ability to function as an independent agent. In fact, systemics incorporates the encoder to such a powerful extent that it is incapable of exerting anything like genuine autonomy. This alternative means to systemizing the author entails positioning the author as one who is merely making choices dictated by, and within the confines of, the system. Ironically, this is precisely what systemics yearns for, since it shows the same inclination for a type of structural "objectivity" that is reflected in Saussure's preference for langue as an object of study. Systemicists prefer to treat the author as an object in the author-system because then it has a capacity to be regulated as a term - a capacity that the quirky, aberrant human subject is thought to lack. The encoder's presence in effect is neutralized into a systemic absence as a result. Eco demonstrates this preference in a discussion on Jacques Derrida and "the deconstructive framework," something Eco casts as remotely systemic (Limits 39). "To affirm that a sign suffers the absence of its author and of its referent," Eco declares, "does not necessarily mean that it has no objective or literal linguistic meaning" (33). Even without an author, the system thrives on. And, with the problematic status of the author, contra Foucault, this contention carries even greater attraction.

Eco projects the implications of this view by conjecturing on the consequences of removing the encoder intention altogether:

Once the text has been deprived of a subjective intention behind it, its readers no longer have the duty, or the possibility, to remain faithful to such an absent intention. It is thus possible to conclude that language is caught in a play of multiple signifying games; that a text cannot incorporate an absolute univocal meaning; that there is no transcendental signified; that the signifier is never co-present with a signified which is continually deferred and delayed; and that every signifier is related to another signifier so that there is nothing outside the significant chain, which goes on ad infinitum. (Limits 33)

Eco's point here is that the encoder leaves behind residual effects on a text's signifying system. While Foucault emphasizes the author in the author-function, Eco gives prominence to the function side of the pairing. (This will be discussed later in relation to Eco's commentary on the "intention of the text.") Eco suggests, for instance, that "a text can be interpreted independently of the intention of its utterer" (39). But he adds that "we cannot deny that any text is uttered by somebody according to his/her actual intention, and this original intention was motivated by a Dynamic Object (or was itself the Dynamic Object)." Since this latter consideration is usually unrecoverable, or recoverable only in a problematic fashion, Eco opts for the alternative of analyzing the systemic components of the text that reside within its signifying complex.

One of the best illustrations of the encoder's incorporation into the system appears in Roland Barthes's "From Work to Text." (Lecture Three noted that Barthes actually dramatized this disappearance of the author in S/Z by declining to consider the author-system in his analysis of codes in Balzac's Sarrazine.) In "From Work to Text," Barthes argues that the death of the author - forecasted in his earlier essay by the same name - had taken place by 1971. "Over against the traditional notion of the work...there is now the requirement of a new object, obtained by the sliding or overturning of former categories" (156). This object - the Text - has a status distinct from the earlier humanistic concept of the Work, which reflected not only "a fragment of substance" (156), but also the kind of origin that both Foucault and Eco, albeit for different ends, describe. To the contrary, Barthes views the Text as "a methodological field" (not unlike his notion of the "writerly" also developed in S/Z) that can be "experienced only in an activity of production" (157). "The [Text] is a processs of demonstration" and "speaks according to certain rules (or against certain rules)." And, the Text "reads without the inscription of the Father" (161) which Barthes depicts as "the myth of filiation" (160). Nevertheless, "it is not that the Author may not 'come back' in the Text, in his text," he suggests, "but he then does so as a 'guest'" and is to be treated "like one of his characters." This contention signifies the dependent, inanimate status the encoder is granted upon systemicization. In the same way that a character has no real agency within the arena of decoding, the encoder is treated as a token to be manipulated without fear of contradiction.

The system of text reception also functions to disempower the encoder in other ways as well. Eco, for instance, claims that authorial intention can have no bearing on the decoder's handling of the text. Rather, decoders will employ a decoding system that is aligned with literary practices of a specific time. "When a text is produced not for a single addressee but for a community of readers...the author knows that he or she will be interpreted not according to his or her intentions," Eco contends (Interpretation 67). The informed encoder will expect to be decoded "according to a complex strategy of interactions which also involves the readers, along with their competence in language as a social treasury." And, given the prudent maintenance of economy over this "treasury", it is not something the community will condone to be "spent" carelessly. It will be parcelled out with moderation and care by the system and its cooperative agents.

This can be said of the entire process of literary decoding, Eco argues. The system is so powerful as a controlling entity that, "even when separated from its utterer, its arguable referent, and its circumstances of production," a "message" (Limits 4) nevertheless still retains a certain amount of undeniable and fixed "referential power" (5). But, the decoder is also granted a correspondingly weak position in the overall system. While Eco is famous for emphasizing the contribution made by the decoder in the overall process of semiosis, he consistently grants the other facets involved an equal - or, in the case of the system, a greater - status. He imagines a situation in which a reasonable response by an encoder to an anticipated proposed decoding might be a reluctance based on semiotic thrift. Under these circumstances, the encoder may justifiably say: "'Independently of the fact that I did not mean this, I think a reasonable reader should not accept such an interpretation, because it sounds uneconomical'" (Interpretation 73). Eco espouses an at least partially restrictive capacity of systemicity when he concludes that "the interpreted text imposes some constraint upon its interpreters" (Limits 6). "The limits of interpretation coincide with the rights of the text (which does not mean with the rights of the author)" (6-7). Considering that the author is one of the primary decisive agents in the process of literary semiosis, however, it's not surprising that elsewhere Eco will assert a systemic power accorded to him or her over the one ostensive outsider to the overall system: the decoder. Although he proposes that "any act of interpretation is a dialectic between openness and form, initiative on the part of the interpreter and contextual pressure," it is clear that the upperhand in this dialectic is always held - for Eco - by the system (21).

Systemizing The Decoder

It may be useful to revisit Eco's contention on this issue cited in Lecture Four: "the interpreter" is not "entitled to say that the message can mean everything" (Limits 5). To Eco, within such a systemics, the decoder's ability to participate is clearly circumscribed in a number of ways. One systemic component Eco employs to propose enforceable control over the decoder is that of denotation based on dictionary entry order. (This would be an example of Halliday's polysystemicity, in which one system intersects with another.) Eco remarks that "within the boundaries of a given language, there is a literal meaning of lexical items and that it is the one listed first by dictionaries as well as the one that Everyman would first define when requested to say what a given word means" (Limits 5). Eco goes so far as to link this cooperation with the extent of the decoder's semiotic well-being:

in order to explore all the possibilities of a text, even those that its author did not conceive of, the interpreter must first of all take for granted a zero-degree meaning, the one authorized by the dullest and the simplest of the existing dictionaries, the one authorized by the state of a given language in a given historical moment, the one that every member of a community of healthy native speakers cannot deny.(36)

Eco resorts to the concept of a restraining "literal meaning," positing that "no reader-oriented theory can avoid such a constraint" (Limits 6). He allows, though, that "any act of freedom on the part of the reader can come after, not before, the acceptance of that constraint." Like John Stewart's emphasis on systemic coherence (discussed in Lecture Two), Eco also relies on an internalized logic which the decoder is required to acknowledge and, more importantly, respect. "Internal textual coherence controls the otherwise uncontrollable drives of the reader," he suggests (59). Eco additionally hazards the "rule" that "the internal coherence of a text must be taken as the parameter for its interpretations" (60). This perspective thereby situates "the functioning of a text" as something that "foresees and directs...interpretive cooperation" by the addressee (45).

As was seen in other systemicists' reliance upon context derived from surveying of reader-response theory, he similarly concludes that from this perspective, "even the meaning of the most univocal message uttered in the course of the most normal communicative intercourse depends on the response of its addressee, and this response is in some way context-sensitive" (Limits 45). In this light, semiosis seems contingent on a form of context even under the most transparent of semiotic circumstances. Eco systemicizes the reader in a manner that parallels the treatment of the encoder discussed earlier here. "Interpretive cooperation" - he cites his notion of the Model Reader as an example - conceives of the encoder's "textual strategy as a system of instructions" for the decoder (52). This guidance is designed to elicit "a possible reader whose profile is designed by and within the text, can be extrapolated from it and described independently of and even before any empirical reading." This engenders "intention of the text" which is nonetheless "only...the result of a conjecture on the part of the reader" (58).

Admittedly, Eco does avoid a total commitment to a self-contained system whose immanent signified can be apprehended by the decoder. Yet, he calls this conjecture something which, itself, is ordered as "a system of expectations" (Limits 63). In addition, he suggests that "the initiative of the Model Reader consists in figuring out a Model Author that is not the empirical one and that, at the end, coincides with the intention of the text" (59). Given this systemic armature, the decoder is limited to "economical" decodings that are consistent with the valencies established by the system and its apparent "intention". This caution is particularly important for Eco, who repeatedly laments that interpretation is vulnerable to "the ascription of pertinence to the wrong element" by overzealous - or worse - overly creative decoders (Interpretation 49). (The later discussion in this lecture of Eco's analysis of an act of interpretation by Geoffrey Hartman illustrates this well.)

Like those reader-response theorists (Roman Ingarden and Wolfgang Iser, for example) who have explored the reader's significant contribution to the finalization (or "concretization") of the text, Eco likewise appears to emphasize a creative faculty by the decoder. But Eco employs this concept to argue, ironically, that the reader's participation is largely mechanical and responsive to the text's "formal devices" or "hermeneutic mechanism" (Limits 52), rather than creative and active. In his system, the decoder's completion of the "text intention" is only a matter of relatively acquiescent cooperation rather than independent creation. "Every text is a complex inferential mechanism...which has to be actualized in its implicit content by the reader," he argues (260).

In order to make sense of a text, that is, to understand it, the reader has to 'fill' the text with a number of textual inferences, connected to a large set of presuppositions defined by a given context (knowledge basis, background assumptions, construction of schemata, links between schemata and text, system of values, construction of point of view, and so on).

By situating the decoder's practice in harmony with the other aspects of systemics discussed here, the decoder is placed in a systemic field. This "decoder-system" essentially replicates the actions of the real - and thus unpredictable, unsystematic - decoders but in a fashion that is predictable and systematic. Eco gives this endeavor a wide-ranging epistemological consequence, but only to foreground a later localization of it specifically into the realm of literary semiotics. "The thought or opinion that defines reality must...belong to a community of knowers," Eco announces, "and this community must be structured and disciplined in accordance with supra-individual principles" (Limits 40).

Part of this systemicity created by the community entails directing semiosis. The decoder's engagement with the literary text is, in Eco's view, "implemented, encouraged, prescribed, or permitted by the textual linear manifestation" (Limits 44-45). Decoders who refuse to abide by this social semiotic contract risk having their activity framed as outside of the system. Eco admits that while such renegade decodings might be "interesting", he stresses repeatedly that their lack of voluntary semiosic economy threatens the status quo of the system (Interpretation 76).

He further derides this activity by demonizing the concept of semiotic play. While Romantic aesthetics emphasized "play" as the highest form of aesthetic achievement, an inclination that was revived repeatedly by various subsequent Aestheticist outbreaks culminating in post-structuralism, play has become in the 1990s a once-again denigrated activity. Even when decoders are "using a text as a playground for implementing unlimited semiosis," Eco argues, "they can agree that at certain moments the 'play of musement' can transitorily stop by producing a consensual judgment" (Limits 41-42). "Indeed, symbols grow," he insists, "but do not remain empty" (42). Eco's reference to Peircean play here is used only as something to rise above through an act of economical maturity and communal utility. "There is something for Peirce that transcends the individual intention of the interpreter, and it is the transcendental idea of a community, or the idea of a community as a transcendental principle," Eco suggests (40). If a literary semiotics "assumes that texts are open to multiple readings," it "must also assume that it is possible to reach an agreement, if not about the meanings that a text encourages, at least about those that a text discourages" (45). The "public agreement" of readings can therefore be employed to displace the decoder who has exited the system (28).

A certain degree of historicity can also contribute to the construction of a systemic order of decoding, providing a traditional overlay that actually diminishes the individual decoder's activity range. "If the sign does not reveal the thing itself," Eco contends, "the process of semiosis produces in the long run a socially shared notion of the thing that the community is engaged to take as if it were in itself true" (Limits 41). This can be accomplished in advance, Eco argues, when the encoder engages a system that has already incorporated a specific systemic response. "The addressee should rely on certain preestablished conventional interpretations" that the encoder and subsequent text rely upon when employing a given semiotic system (5).

Like Foucault, Eco stresses saving semiosis from spendthrift decoders. While in the course of decoding a given sign vehicle, Eco says, "the addressee could make various conflicting hypotheses, but I strongly believe that there are certain 'economical' criteria on the grounds of which certain hypotheses will be more interesting than others" (Limits 5). "To validate his or her hypothesis," Eco adds, "the addressee probably ought to first make certain conjectures about the possible sender and the historical period in which the text was produced." While this "has nothing to do with researching the intentions of the sender," he contends somewhat contradictorily, he suggests that "it certainly has to do with researching the cultural framework of the original message" (5).

This raises the issue of merely "interesting", extra-systemic decoding, which Eco asserts is indeed outside of the system. "It is reasonable that the reader has to the right to enjoy" various wide-ranging readings "that the text qua text provides him or her" (Interpretation 71). "But at this point," he remarks, "the act of reading becomes a terrain vague where interpretation and use inextricably merge together." And, of course, "the criterion of economy becomes rather weak."

Rorty attempts to develop a positive alternative to this position in his own privileging of decoder "use" of a text. In response to Eco, he proposes that "the coherence of the text...is no more than the fact that somebody has found something interesting to say about a group of marks or noises - some way of describing those marks and noises which relates them to some of the other things we are interested in talking about" (Interpretation 95). This would certainly contrast Eco's touted "internal textual coherence" by locating the agency of this discernment within the decoder as opposed to the text's system. And, obviously, Rorty is using "interesting" in a much less "economical" way than systemicists like Eco prefer.

Jonathan Culler complicates this discussion by contending that systemic components are recognizable as such only after the system's material has been rearticulated by the decoder. Also in response to Eco, he proposes that "describable semiotic mechanisms function in recursive ways, the limits of which cannot be identified in advance" (Interpretation 121). And Ruqaiya Hasan takes a specifically genre-oriented approach to make essentially the same point. "The controls upon the structural make-up of a text are not linguistic in origin," Hasan observes, "in that language as a formal system does not enable one to predict what generalized structural formula could be associated with which genre" (229). Firth also had earlier supported this position:

in emphasizing the systemic nature of language, I do not propose an a priori system of general categories by means of which the facts of all languages may be stated. Various systems are to be found in speech activity and when stated must adequately account for such activity. Science should not impose systems on languages, it should look for systems in speech activity, and, having found them, state the facts in a suitable language.("Semantics" 144)

Eco, on the other hand, expresses considerable alarm over this threat to the neatly limiting implications of systemicity. His response is to call this type of decoding "hermetic drift," which he depicts as an "interpretive habit...based on principles of universal analogy and sympathy, according to which every item of the furniture of the world is linked to every other element (or to many) of this sublunar world and to every element (or to many) of the superior world by means of similitudes or resemblances" (Limits 24). A system like this would hardly provide the organizing harmony usually posited by systemicists. To the contrary, the "main feature" of this sense of drift, he insists, "seems to be the uncontrolled ability to shift from meaning to meaning, from similarity to similarity, from a connection to another" (26-27).

Not unexpectedly, Eco portrays this as semiotic anarchy. "This world perfused with signatures, ruled, as it pretends, by the principle of universal significance, results in producing a perennial shift and deferral of any possible meaning" (Limits 27). Eco can barely contain his contempt for the non-systemicity of a system of unrestricted semiosis and "pretend" rules. Not only does "Hermetic semiosis [transform] the whole world into a mere linguistic phenomenon," he concludes with alarm, it also "devoids language of any communicative power" (27). And, finally, "in the most extreme cases of Hermetic drift, no contextual stricture holds any longer" (30). "Connotations proliferate like a cancer and at every step the previous sign is forgotten, obliterated, since the pleasure of the drift is given by the shifting from sign to sign and there is no purpose outside the enjoyment of travel through the labyrinth of signs or of things" (31). This can even taint the system itself since "aspects" of texts can be "polluted and obscured" by "too many uncontrolled intentions of the readers" (62). A debacle of this nature is particularly likely in the case of the systemic nihilism of Eco's favorite villain: deconstruction. "The most radical practices of deconstruction privilege the initiative of the reader," Eco declares (52). As a consequence, these practices "reduce the text to an ambiguous bunch of still unshaped possibilities, thus transforming texts into mere stimuli for the interpretive drift." A carcinogenic mélange of cheap pleasure, aimless drift, pollution and obscuration is evidently the only foreseeable outcome for Eco should the security of a specific type of system disappear.

Structure or System?

But, systemicists like Eco aren't conflating structure with system in this debate - at least on the surface, anyway. They seem to be avoiding the vulnerability that hampers structuralist methodologies and thus are establishing a model that could potentially reconfigure the paradigm for semiotics as long as it resists becoming mired into reduction. Eco demonstrates the apparent division between structure and system when he observes:

Semiotics studies both the abstract structure of signification systems (such as verbal language, card games, road signals, iconological codes, and so on) and the processes in the course of which the users practically apply the rules of these systems in order to communicate, that is, to designate states of possible worlds or to criticize and modify the structure of the systems themselves.(Limits 207)

Careful scrutiny reveals that this conception of process is really merely a semantic displacement of "structure" for "system" and "system" for "process", however. For, Eco cannot go beyond systemic control in a manner that is also comfortably systematic - even in his most "open" apprehension of the term. This substitution by Eco may become clearer in comparison with Halliday's distinction between structure and system:

What distinguishes systemic theory is that its basic form of synoptic representation is not syntagmatic but paradigmatic; the organizing concept is not structure, but system (hence the name). Since language is a semiotic potential, the description of a language is a description of choice. The various levels, or strata, of the semiotic 'code' are interrelated networks of options. The constituent structure is the realization of these options, and hence plays a derivative role in the overall interpretation. ("Systemic" 8)

Accordingly, Halliday pairs systemic description with paradigmatic relations and structural description with syntagmatic relations ("Deep" 93). To return to Culler's assertion regarding systemic recursivity, Halliday remarks that there are "some possible consequences of regarding systemic description as the underlying form of representation, if it turned out that the structural description could be shown to be derivable from it" (93-94). If this is found to be an accurate assumption, he adds, "structure would be fully predictable." This "relatedness", in other words, would suggest that structure is "the realization of complexes of systemic features" (94).

System or Process?

While in one regard system seems synonymous with structure, in another it appears remarkably similar to process. (Greimas's concept of system as an "ensemble" could be said to apply in this instance.) But Hjelmslev provides what is arguably the most theoretically sophisticated distinction between system and process that advances clarification of their conceptual differences. As James R. Martin notes, Hjelmslev proposes system as an organized "potential", whereas process is taken as "the realization of this potential as process" (248). A system "lies behind" a process, Hjelmslev suggests, and subsequently is "ordered to" it (39). He contends, however, that while the system is necessary for the existence of the process, the system itself does not depend on a corresponding process for its own existence:

the existence of a system is a necessary premiss for the existence of a process: the process comes into existence by virtue of a system's being present behind it, a system which governs and determines it in its possible development. A process is unimaginable - because it would be in an absolute and irrevocable sense inexplicable - without a system lying behind it. On the other hand, a system is not unimaginable without a process; the eixstence of a system does not presuppose the existence of a process. The system does not come into existence by virtue of a process's being found.

It could be argued that this conceptualization pinpoints a common assumption of systemics that could have prompted much of the criticism about it, for this independence of system over process appears to validate its objective, and separate, status. That this assumption has to be developed carefully is indicated by Hjelmslev's own caution that linguistics (circa 1943) needs to "test" the "theory that a process has an underlying system - a fluctuation an underlying constancy" (10).

The most substantial means of differentiating between these two elements in Hjelmslev's model is to consider them as different "hierarchies" (29). Without going into Hjelmslev's complex development of this distinction, it is sufficient to point out that he identifies system with a "correlational hierarchy" and process with a "relational hierarchy" (39). (Which actually is quite similar to the distinction between paradigmatic and syntagmatic relations, respectively.)

To develop this system/process distinction, Firth entertains a perspective that reverses Hjelmslev's prioritization of system over process, giving the latter the initiating force for subsequent analysis. Firth suggests in 1948 that linguistics will inevitably "find it necessary to postulate the maintenance of linguistic patterns and systems (including adaptation and change) within which there is order, structure, and function" ("Semantics" 143). Significantly, he proposes that "such systems are maintained by activity, and in activity they are to be studied." "It is on these grounds," he concludes, "that linguistic s must be systemic."

One of Stewart's primary charges against semiotics is that "semiotic characterizations of language picture it as a system rather than as a process, event, or mode of human being" (19). As was mentioned in Lecture Two, he is among those currently proposing a process-oriented semiotics (or, a "post-semiotics", in Stewart's case) designed to correct this imbalance. "The inclination to treat language as a system has consistently hypostatized the process," Stewart argues, "frequently under the rationale that this is the only way to treat it systematically, objectively, or 'scientifically.'" Clearly, adding process as a significant component of semiotics may effectively reorient systemics and make it more responsive to what seem to be the actual conditions of semiosis. It is important to recall Halliday's admonition that "the name 'systemic' is not the same thing as 'systematic'" ("Brief" 3), for often the two are treated as one.

Consistent with the negative side of the often presumed "objectivity" of systemics is its concomitant supposition that the system exists separate from human agency. Stewart contends that "system commitment distorts language study" because "it attempts to separate the analyst from the phenomenon being analyzed - language - even though the only way to analyze language (or any other topic) is linguistically, discursively, communicatively, 'in' language" (20). If the system is considered as something separate from the realm of its agents, then it cannot account for the constitutive component of semiosis performed by those agents. That the system can be apprehended in an extra-human condition also begs a nagging question about the epistemological ground of that belief. As Stewart observes, "the so-called system of language cannot coherently be conceived as existing separate from, and in an object-to-subject relationship with, humans communicating" (50). To Stewart, systemicity too often implies that "an object of study...is unproblematically accessible to investigation by human subjects" (20).

Ventola posits a humanized alternative to Stewart's charges that seems to hold substantial potential for making systemics more attuned to what John Deely refers to as "the human use of signs." "Social encounters are systems where social processes, which realize the social activity, unfold in stages and," she asserts, "in doing so, achieve a certain goal or purpose" (1). From a textlinguistic standpoint, Ventola proposes constructing a literary genre system (not unlike Hjelmslev's) that considers "texts at the same time as products and as processes" (67). This system uses a "genre network" to represent a "synoptic system" which is portrayed as "static-potential". This "text as a product" view situates the text as "actual-static". A "flowchart" schema, on the other hand, can be used to posit a "dynamic system" which is "active-potential". The "text as a process" is accordingly "actual-active" (67).

Hjelmslev suggests that a combination of "partitioning" (associated with process) and "articulation" (associated with system) can produce a similar, dynamic sense of systemic order. He concludes that through this mutual cataloging of elements, "the inventories established by an analysis following...traditional lines would lead us to recognize a sign system behind the sign process" (44). Peter H. Fries's analysis of Lillian More's children's story, "Freddie Miller; Scientist," offers a good illustration of this approach by employing "three simple working assumptions": "important ideas tend to be repeated"; "important ideas tend to be placed in positions of prominence"; and "meaning is conveyed only where there is choice...of items within a context where certain goals are to be attained" (316). This categorization demonstrates a viable systemic method of analysis since, as Hjelmslev and others note, regularity (and even regular irregularities) are usually evidence that a given entity is ordered to some extent. While emphasizing the component of "choice" in systemics, Fries notes that "to describe this story, or any story, we must link the choices made in telling this story to the choices that are available in the system of the language being used." By "linking the choices made in telling one story to the choices available in the language system as a whole," Fries says, the story under consideration can be assessed in terms of potential locations in the literary system (316). This also can apply to the larger social semiotic system in the respect that "this story relates to the potential of stories in the language." By proceeding in this way, Fries asserts that "we can look at [the story] as an example of a type, which allows us to compare different texts." And, we can also "pay attention to those features of the text which mark it as unique and distinguish it from all other texts."

Greimas on Maupassant

A much more thorough and well-developed demonstration of a related form of systemic analysis can be found in Greimas's Maupassant - The Semiotics of Text, which is frequently considered a response to Barthes's systemics (or a-systemics, as the case might be) in S/Z . Greimas's study, along with several assessments of it by others, is a revealing instance of what systemics has to offer to semiotics. Moreover, it dramatizes some of the shortcomings discussed above that stem from systemicist presuppositions.

The operating procedure in Maupassant is theoretically foregrounded, I would argue, by an earlier essay Greimas wrote in collaboration with François Rastier, "The Interaction of Semiotic Constraints." Greimas argues that "the elementary structure of signification forms the semantic universes taken as a whole into systems" (50). In a manner similar to other geometric relational paradigms, Greimas proposes as a "constitutional model" ("Interaction" 49) of this structure what he later refers to as the "semiotic square" (Maupassant 43). (This is actually an extension of Greimas's even earlier commentary on "homologation" in Structural Semantics.)

The square is a "hypothetical model" which might reveal an entity's "figurative organization" and its "axiological values," Greimas suggests (Maupassant 43). "The semiotic square is a logical mapping out of structural possibilities," Ronald Schleifer observes ("Introduction" xxxiii). "For any content which can be understood as itself analyzable into binary oppositions (S vs. non S), the square, repeated and superposed, will exhaust the logical structural relations between its minimal elements." Schleifer suggests that in his analysis of the "modal aspect of...functions," Greimas's "functional analysis links the events of narrative to the elementary structure of signification; it creates a passage from process to system" (Greimas 124).

Nancy Armstrong constructs a concise account of the semiotic square's system that is especially useful. "Once any unit of meaning [e.g., 'life'] is conceived," she proposes, "we automatically conceive of the absence of that meaning ['non-life'], as well as an opposing system of meaning ['death'] that correspondingly implies its own absence ['non-death']" (54). (The examples in brackets - life, death, non-death, and non-life - are Greimas's from Maupassant [5].)

Before proceeding to Greimas's account and application of his system, it is fruitful to pause and assess the implications of the contentions offered by Armstrong and Schleifer. Armstrong reveals, as some critics of systemics would argue, a confident ordering feature of semiosis that appears to exhaust the semiosic relations of a given element ("life", for instance). Schleifer actually uses the term "exhaust" and projects this totalizing capacity of systemics much beyond a careful tentativity. From these two examples alone it is easy to see how systemic orientation can readily lend itself to reductively programmatic decodings.

Greimas himself engages in this assumption as he posits the systemic principle that "each term of a semiotic structure is defined by relations of conjunction and disjunction" ("Interaction" 61). Additionally, "the manifestation of a system" is not "defined solely by the relations it permits" (59). "By definition," he contends, "every system has a set of rules; they may be defined positively, but they can also be defined negatively by what they are not" (52). Additionally, "a system's rules of injunction describe compatibilities and incompatibilities (a system without incompatibilities would not be an ordered system)." Here, Greimas substantially broadens systemicity to include the factors outside of a specified system. "Nothing permits us to assert that a semiotic manifestation is dependent on only one system at a time" (60), he asserts. "And so far as it is dependent on several, its closure can be attributed to the interaction of the different systems that produce it." While this seems to resist the homogenization that frequently accompanies systemicity, it can be incorporated in such a way as to contribute to that outcome just as easily.

Greimas provides a subtle distinction of inter-systemic levels he refers to as "epistemy", or "the hierarchy of the systems" ("Interaction" 61). These levels "order the combinations that can appear, and thus not only the closure of the manifestation (negative definition of usage by nonmanifestations), but the nature of realized manifestations (positive definition of usage)." Like Halliday, Greimas employs a hierarchical model which hinges on selection by the individual agent involved in order to refine the non-differentiated concept of system. "The term choice can be used to designate the processes that produce the realized manifestations and define usage positively," he suggests. "Constraints," to the contrary, "designate the processes that cause the nonmanifestations and define usage negatively (the constraints determine asemanticity, or incompatibility of the interacting terms of the systems)."

Both choice and constraints are matters of another type of choice at another level of this system for Greimas. "An author, a producer of any semiotic object, operates within an epistemy, which is the result of his individuality and the society in which he is inscribed," he contends ("Interaction" 61). "Within this society it is possible for him to make a limited number of choices, which have as an initial result the investment of organized contents, that is, contents endowed with valencies (possibilities of relations)." Agent participation helps to integrate Greimas's sense of system with that of process, although this element continually returns to the extra-human position in the system that Stewart identifies. While Greimas allows that "epistemy accounts for the historicity of the manifestations," he supplements this with a - perhaps unintentionally - humorous elaboration. The "social component" of semiosis within his schema, he notes, "appears as common sense, implicit or not, which is an axiological and dialectical system immanent in all the semiotic structures of the society under consideration."

The first section of Maupassant adequately reveals Greimas's use of a systemics based on semiotic square dynamics. Gremias's text for analysis is Guy de Maupassant's "Two Friends" and he begins with the opening lines of the story:

Paris was blockaded, famished, a death rattle in her throat. The sparrows rarely appeared on the roofs, and even the sewers were being emptied of their regular tenants. People were eating - no matter what. (1)

Significantly, this leads Greimas to emphasize what R. W. Bailey identifies as a defining characteristic of systemic linguistics: "the idea of 'levels' as a root metaphor for language behavior," with an accompanying belief that "a description should consist of an account of patterning at a given level and a set of realization rules that link the levels" (1). Greimas says that "all discourse - and especially narrative discourse - has a multiplanar organization" consisting of "undeniable delimitations, which are sometimes situated at one or other levels of discourse" (1). Accordingly, Greimas proposes to begin with "spatio-temporal criteria of segmentation" of the text. "Pragmatic discourse," he adds, details a "series of 'events' or 'things' which are necessarily inscribed in a system of spatio-temporal co-ordinates."

A "temporal infrastructure" (Maupassant 2) is being established in Maupassant's opening paragraph, Greimas argues, proposing the opposition:

"before the war"    vs    (during the war)

This can be figured additionally as "a specifically temporal category":

/before/  vs  /during/  vs  /after/

as well as "a denominative category":

/war/  vs  /peace/

Similarly, a "spatial setting" is established through the Paris reference, spurring an axiological relation in accordance with Greimas's schema:

/enclosed/  vs  /enclosing/
__________      ___________ 

 "Paris"        (non-Paris)

Without pursuing this excessively, it should be apparent where Greimas is going with this analysis. He suggests, for instance, that the opening sequence is "composed of four co-ordinated propositions, each having a different subject: 'Paris', the 'sparrows', 'the sewers' and the 'People'" (3). Or, the opening is portrayed as a projection of the "homologation of the zoomorphic beings 'sparrows' and 'rats' with their respective spaces" (15):

"roofs"     "sparrows"     /high/      /aerial being/
_______  :  __________  :  ______  :  ________________

 "rats"      "sewers"      /low/      /chtonian being/

This overview is an extremely sketchy rendering of an immensely complicated and mature systemic account of Maupassant's story, yet it does show some of the strategies typical of this approach that may suggest further directions for the ongoing development and refinement of systemic models and analysis.

Barthes on Poe

Despite the a-systemic manifesto that Barthes develops as a disclaimer in his "Textual Analysis: Poe's 'Valdemar'," his actual analysis is not all that different from Greimas's. Admittedly, though, Barthes is on the outer limits of systemics, while Greimas inhabits a comfortable central position within it. Barthes limits his procedure very specifically: "in analysing the 'signifiance' of a text, we shall abstain voluntarily from dealing with certain problems" (174). Among these exclusions, Barthes says, are consideration of the author, the larger literary historical context within which Poe can be situated, and the impact of reading a text in translation. "We shall take the text as it is, as we read it," Barthes announces. (Barthes elsewhere offers a working definition of Julia Kristeva's conception of 'signifiance' as "a process in the course of which the 'subject' of the text, escaping the logic and the ego-cogito and engaging in other logics [of the signifier, of contradiction], struggles with meaning and is deconstructed ['lost']" [cited in Heath, "Translator's Note" 10].) "Textual Analysis: Poe's 'Valdemar'" demonstrates Barthes's system well, as his analysis of the first two sentences of Poe's story suggests. (His use of a sectional method and code application is a variation of his technique in S/Z.) Poe's story, "The Facts in the Case of M. Valdemar," begins:

Of course I shall not pretend to consider it any matter for wonder, that the extraordinary case of M. Valdemar has excited discussion. It would have been a miracle had it not - especially under the circumstances. (177)

These sentences "are apparently meaningless," Barthes remarks. Still, they have as an "obvious function" the role of "exciting the reader's expectation." This prompts the reader to desire "the solution of the enigma posed in the title (the 'truth'), but even the exposition of this engima is held back." Barthes assesses the function of this aspect as: "delay in posing the enigma." This delay is "a matter of whetting the reader's appetite" - an operation of the "narrative code" (177). Moreover, there is an additional value of this "patter" as it signals the text's semiotic status as a "commodity" (176). This creates an "appetiser" which stands as "a term of the narrative code (rhetoric of narration)."

Barthes identifies a site of particular semantic weight in Poe's "ambiguous" use of "extraordinary" which is marked as "meaningful" precisely because of its ambiguity (176). The striking semantic selection draws attention to the word as it signifies a condition or event that "departs from the norm but not necessarily from nature (if the case remains 'medical')." "But it can also refer to what is supernatural, what has moved into transgression," he observes "'This is the 'fantastic' element of the stories...that Poe tells'" (177).

Poe's vague diction suggests that the "story will be a horrible one (outside the limits of nature) which is yet covered by the scientific alibi (here connoted by the 'discussion', which is a scientist's word)." This produces what Barthes calls a "bonding" that is "cultural" in nature. Barthes departs somewhat from his procedure when he notes that Poe engages in the characteristically nineteenth-century fascination for "the mixture of the strange and the scientific." "There was great enthusiasm for observing the supernatural scientifically (magnetism, spiritism, telepathy, etc.)," he adds. "The supernatural adopts a scientific, rationalist alibi; the cry from the heart of that positivist age runs thus: if only one could believe scientifically in immortality!" Barthes identifies this as a subset (the "scientific code") within the larger "cultural code" which "will be of great importance throughout the narrative."

Eco on Hartman on Wordsworth

While Fries, Greimas, and Barthes demonstrate various strategies for engaging in systemic analysis, Eco provides an example of a systemicist's rejection of another systemicist's practice. (As mentioned above, Maupassant stands in a similar relation to S/Z, but far less overtly so.) In Interpretation and Overinterpretation, Eco discusses an analysis of a William Wordsworth poem by Geoffrey Hartman. Contextualizing this example will be helpful because Eco establishes an elaborate ideological position on systemics both in his preceding work on semiotics and in his present discussion up to the point that he introduces Hartman's reading.

Eco is well-known proponent of endeavors to "protect the reading...rather than open[ing] it too much" (Limits 37). While he was talking about the interpretation of Peirce's work in this statement, it applies to his sentiments about all texts, as his body of semiotic writings attests. "I feel sympathetic with the project of opening readings," Eco adds, and a surface reading of his books on The Role of the Reader and The Open Work might seem to corroborate that. He continues, however, that "I also feel the fundamental duty of protecting [readings] in order to open them, since I consider it risky to open a text before having duly protected it" (54).

To return to his distinction between use and interpretation: "I can certainly use [a] text for parody, for showing how a text can be read in relation to different cultural frameworks, or for strictly personal ends (I can read a text to get inspiration for my own musing)" (Interpretation 69). "But," he concludes, "if I want to interpret" a specific text "I must respect [the encoder's] cultural and linguistic background." "To defend the rights of interpretation against the mere use of a text does not mean that texts must never be used," he argues elsewhere. "It is only important to distinguish between use and interpretation" (Limits 62).

By aligning himself with the system and the encoder over the decoder, Eco maintains that "internal textual coherence controls the otherwise uncontrollable drives of the reader" (Interpretation 65). Eco attempts to turn the decoder's willingness to cooperate with the system into an act of voluntarily accepting systemic restraints. "A sensitive and responsible reader," he insists, "has the duty to take into account the state of the lexical system at the time" when the text was constructed (68).

This stance on systemics is borne out in Eco's reponse to Hartman's own systemics as seen in "The Interpreter's Freud" in which he analyzes "A Slumber Did My Spirit Seal" in "a Freudian context" (138). Hartman contends that Freud "created a new hermeneutics by charting compulsive and forced connections which 'regarded nothing as sacred'" (154), and this is the systemic grid he applies in his reading. For Freud, he argues, the text (the dream-text, in this instance) "becomes less of an object and more of a series of linguistic relays that could lead anywhere - depending on the system of rails and who is doing the switching" (142). (This last observation, by the way, blends the two foci of this lecture - system and encoder, a significant development for Hartman's decoding of the poem and Eco's response to it.)

Not unlike Fries's analytical practice, Hartman's analysis focuses on "three highly charged themes: incompleteness, mourning, and memory" (145). Hartman's approach consists of exploring a system of "subliminal punning" that develops these themes (149). With this observation, Hartman is essentially adhering to Eco's systemic technique of inventorying elements that make up what could be called the thematic or imagistic system of the text. "What makes Wordsworth's poetry so difficult to psychoanalyze," Hartman suggests, "is its underlying and resistant euphemism, coterminous with ordinary language, and distinguished from the courtly and affected diction of the time" (148). Similarly, Hartman limits himself to a consideration of the historical context of linguistic and literary conventions of Wordsworth's period, so in this respect as well he appears to conform to Eco's regimen. This is the systemic framework he establishes for reading these lines:

A slumber did my spirit seal;
     I had no human fears:
She seemed a thing that could not feel
     The touch of earthly years.

No motion has she now, no force;
     She neither hears nor sees,
Rolled round in earth's diurnal course
     With rocks and stones and trees.

In Hartman's reading, "'diurnal' (line 7) divides into 'die' and 'urn,' and 'course' may recall the older pronunciation of 'corpse'" (149). This type of associative analysis would be in keeping with Eco's systemic "respect" of the text in that it restricts itself to seemingly text-based interpretation. Hartman goes too far in Eco's eyes, however, when he suggests that other, far-ranging readings of these lines are equally defendable. For example, he contends that there is a "euphemistic displacement of the word grave by an image of gravitation ('Rolled round in earth's diurnal course')" (149). And even though he relies upon communal input regarding interpretation of the poem, he additionally oversteps the boundaries of propriety (for Eco) by asserting:

though there is no agreement on the tone of this stanza, it is clear that a subvocal word is uttered without being written out. It is a word that rhymes with "fears" and "years" and "hears," but which is closed off by the very last syllable of the poem: "trees." Read "tears," and the animating, cosmic metaphor comes alive, the poet's lament echoes through nature as in pastoral elegy. "Tears," however, must give way to what is written, to a dull yet definitive sound, the anagram "trees." (149-150)

In his by-now familiar formula, Eco calls Hartman's analysis "a case where the rightness of the interpretation is undecidable, but where it is assuredly difficult to assert that it is wrong" (Interpretation 60). Since he associates Hartman with deconstruction (he is "one of the leaders of the Yale deconstructionists," Eco says) which Eco often correspondingly associates with disrespectful semiotic promiscuity, Eco does not pass up an opportunity to chide Hartman for excessive, irresponsible decoding. Even "in the shrewdest representatives of this school the hermeneutic game does not exclude interpretive rules," systemic rules which apparently Hartman has grossly violated. Eco explains that he agrees with Hartman's fairly restrained assertion that "'die', 'urn', 'corpse', and 'tears' can be in some way suggested by other terms that appear in the text (namely, 'diurnal', 'course', 'fears', and 'hears')." Hartman's reading of "grave" is, however, "suggested by a 'gravitation' which does not appear in the text but is produced by a paraphrastic decision of the reader" (61). Note the procedure Eco employs to make this distinction. Systemically, certain kinds of associations are acceptable, possibly because of their "economy" in terms of sign generation. But, taking this process to the next level of suggestivity where "grave" is associated with "gravitational" is a form of unacceptable semiotic extravagance. Yet another incidence of this lavishness is identified in Hartman's proposed tears/trees connection. "'Tears' is not the anagram of 'trees'," Eco objects. "If we can to prove that a visible text A is the anagram of a hidden text B, we must show that all the letters of A, duly reorganized, produce B" (61). "If we start to discard some letters," on the other hand, "the game is no longer valid."

"In theory, one can always invent a system that renders otherwise unconnected clues plausible," Eco continues (Interpretation 62). But, there should be "at least a proof" of textual support "depending on the isolation of the relevant semantic isotopy." "The funereal interpretation of Hartman has the advantage of betting on a constant isotopy," he concludes (63). "Bets on the isotopy are certainly a good interpretive criterion, but only as long as the isotopies are not too generic." Eco goes on to point out that comparisons between figures such as Achilles and a lion could be relevant in terms of beings who are "courageous and fierce," yet comparing him with a duck because "both are bipeds" would go too far. "An analogy between Achilles and a clock based on the fact that both are physical objects," he adds, "is of no interest whatsoever."

Eco's response is puzzling in part because it's hard to believe that Hartman was using "anagram" in its "true" sense. Evidently, he was suggesting a type of scrambled resonance that operates according to a logic not unlike that of the anagram. Perhaps his sense of associative link was acoustic rather than alphabetic, which works along the same lines as an anagram. I would bet that Eco realizes this, too, but since the dictionary definition is something that he relies upon for systemic regularity (recall his comment on this above) it appears that he seized upon this deviation to build up a systemic case against Hartman's reading. Ironically, if he had respectfully adhered to and protected Hartman's reading in the spirit of his comments on the decoder's responsibility to act in this manner, Eco would have been able to incorporate Hartman's observation and retain the spirit of the point he was making instead of employing the letter of the system, so to speak, to challenge Hartman. (I asked Hartman about his use of "anagram" here despite his obvious knowledge that it was not strictly a "true" anagram, and he responded: "I guess I could have called it a 'false anagram'? But what is THAT? I couldn't find a better term, though perhaps one does exist in rhetoric or poetics.")

Another way that Eco accomplishes this is by speculating about the intention of Wordsworth's text to build a systemic consensus against Hartman. This intention "was certainly - it would be difficult to doubt it - to suggest by the use of the rhyme a strong relationship between 'fears' and 'years', 'force' and 'course'," he conjectures (Interpretation 70). He oddly shifts this argument by turning to "Mr Wordsworth in person." "Are we sure" that the poet "wanted to evoke the association, introduced by the reader Hartman, between 'trees' and 'tears', and between an absent 'gravitation' and an absent 'grave'?" This, of course, runs contrary to Eco's frequent protestations that the real author has no control over the intention of a literary text. Something like a Model Author, instead, is what Eco promotes to provide the service that Foucault ascribes to the author-function.

At first, it appears that Eco is actually going to side with Hartman in considering Wordsworth-the-person as someone capable of accepting the tears/trees association. "If a normal English-speaking human being is seduced by the semantic relationships between words in prasentia and words in absentia," Eco says, "why should not one suspect that even Wordsworth was unconsciously seduced by these possible echo-effects?" (Interpretation 70). Eco's response to his own question falls back on his elevation of system over all of its (real) agents, including the encoder in this instance. Eco cites an idea from one of his students, Mauro Ferraresi, of an author existing in a third position between the Model Author (which is "an explicit textual strategy" [69]) and the empirical author. This figure would be the "Author on the Threshold" or the "Liminal Author," something positioned "between the intention of a given human being and the linguistic intention displayed by a textual strategy" (69). In light of Hartman's analysis, Eco suggests that "on the threshold situation where Mr Wordsworth was no longer an empirical person and not yet a mere text, he obliged the words (or the words obliged him) to set up a possible series of associations" (70).

By employing the encoder/system distinction, Eco grounds this situation between "finding in a text either what its author intended to say, or what the text said independently of the intentions of the author" (Interpretation 63-64). In the second case, he notes, there is a question whether "what is found is what the text says by virtue of its textual coherence and of an original underlying signification system, or what the addressees found in it by virtue of their own systems of expectations" (64). Again, this would unveil the "intention of the text"; something which is "not displayed on the textual surface"; rather, "one has to decide to 'see' it." This process is "the result of a conjecture on the part of the reader." Furthermore, "the initiative of the model reader consists in figuring out a model author that is not the empirical one and that, in the end, coincides with the intention of the text." As a consequence, "the intention of the text is basically to produce a model reader able to make conjectures about it." Eco concludes that the text, as "an object that the interpretation builds up in the course of the circular effort of validating itself on the basis of what it makes up as its result," is accordingly "more than a parameter to use in order to validate the interpretation."

Thus, while Hartman's analysis "sounds, if not fully convincing, at least charming," it is safe to say that, due to its excess, it is not "interesting" to Eco (Interpretation 61). Eco confidently asserts that Hartman is "certainly not suggesting here that Wordsworth actually wished to produce these associations - such searching after the author's intentions would not fit Hartman's critical principles" (61-62). (Evidently, because he's a "deconstructionist".) Hartman is suggesting, he says, "that it is legitimate for a sensitive reader to find what he finds in the text," only as a result of the "associations" arising "at least potentially" because they were "evoked by the text" (62). This finds similar possible corroboration in that "the poet might (perhaps unconsciously) have created some 'harmonics' to the main theme."

Curiously, again, Eco seems to soften toward Hartman as he sifts through the implications of his reading. "As far as Wordsworth is concerned," Eco argues, "though on the one hand nothing proves that the text suggests neither tomb nor tears, on the other hand nothing excludes it" (Interpretation 62). In fact, what he is suggesting by this evident equivocation is that Hartman has clearly constructed a systemic foundation for his reading, it's just that Eco doesn't approve of its semiosic license. In the final outcome, "one may judge [Hartman's] interpretation too generous, but not economically absurd," he allows. "The evidence may be weak, but it does fit in."

Eco evidently is using Hartman's demonstration as a means of dramatizing degrees of success for systemic analysis. To Eco, Hartman is a semiotic spendthrift. Yet, in the course of his critique of Hartman, Eco is simply refusing to adhere to Hartman's proclaimed system, which is why he finds questionable validity in Hartman's account. He's not using the same system that Hartman says he's using, in other words.

Perhaps Eco is falling prey to what Michael Reddy refers to as "the conduit metaphor," the belief that words (for instance) are receptacles that can somehow contain and transport meaning without significant alteration by the decoder. Reddy cites as an example one category of expressions "implying that human language functions like a conduit enabling the transfer of repertoire members from one individual to another" (311). (An illustration of this is seen in the statement: "It's very hard to get that idea across in a hostile atmosphere.") This metaphor gives the encoder a certain amount of power in relation to system, but this power also bears with it a certain vulnerable responsibility.

It is easier, when speaking and thinking in terms of the conduit metaphor, to blame the speaker for failures. After all, receiving and unwrapping a package is so passive and so simple - what can go wrong? A package can be difficult or impossible to open. But, if it is undamaged, and successfully opened, who can fail to find the right things in it? (288-289)

According to this metaphorical stance, the reader is then "reading things into" something, "having surreptitiously made use of his power to insert thoughts into words when he should have restricted himself purely to extraction," Reddy says (289). The reader "sneaked those thoughts into the words himself, and then turned around and pretended that he found them there."

By recalling Hartman's explanation of his systemics, it can be pointed out that he announced his intention to engage in a semiotic prodigality consistent with Freud's analysis of the dream-text. Systems that proceed according to a quasi-anagrammatic principle - that would not adhere to the "sacred" definition of "anagram" - would certainly be appropriate models to employ for such an analysis. And that is exactly Hartman said he was going to do, and did.

The Politics of Systemics

The point behind comparing these analyses is that in different ways they reveal the inherently ideological nature of systemics. Whatever theoretical positioning one chooses to employ will substantially influence the outcome of a systemic analysis. This could be the largest shortcoming to systemics insofar as it is used by individuals like Eco to attempt to conceal their own motives through appeals to a system that ostensibly transcends individual agency. (The liminal author - or even the Model Author - over the real author, for instance.) An emphasis on process-oriented, transformational systemics may make this conceptual modeling more effective, even though, as Halliday observes in 1985, "dynamic models of semiotic systems are not yet very well developed..." ("Systemic"10).

To a certain extent, that may be desirable. Maybe a dynamic model should, by necessity, remain underdeveloped and then adapted to given decoding circumstances, as Firth suggests. After all, this type of modeling could also yield results that reflect "everyday" experience, which is seldom systemic in a logical or predictable way. "The 'real world' context may always override all the expectations set out by the conventional and conceptual systems involved," Ventola remarks (29). This applies equally to systemic analysis. "The problem of the differences existing between the expected global structures of text types and the actual global structures manifested in text realizations of texts is not solved in the procedural approach, which ultimately appeals to the immediate context for explanations," Ventola asserts. As an alternative, Ventola focuses on "irregularities and modifications" instead of "yielding to regularities and expectedness" in systemics. Halliday notes in a related vein:

Systemic theory is explicitly constructed both for thinking with and for acting with. Hence - like language, again - it is rather elastic and rather extravagant. To be an effective tool for these purposes, a theory of language may have to share these properties with language itself: to be non-rigid, so that it can be stretched and squeezed into various shapes as required, and to be non-parsimonious, so that it has more power at its disposal than is actually needed in any one context. ("Systemic"11)

Culler applies this systemic approach to Eco's plea for economy by arguing that "an excessive propensity to treat as significant elements which might be simply fortuitous" may actually be "the best source of the insights into language and literature that we seek, a quality to be cultivated rather than shunned" (Interpretation 122). "It would be sad indeed," he observes, "if fear of 'overinterpretation' should lead us to avoid or repress the state of wonder at the play of texts and interpretation" (123).

To be completely fair to Eco, he does acknowledge that certain elements of semiosis "cannot be foreseen by an [coded] system of signification" (Limits 212). (He cites as some examples: "textual co-reference, topic, text coherence, reference to a set of knowledge idiolectally posited by a text as referring to a fictional world, [and] conversational implicature.") Yet he continues to cling to the belief that certain facets of signification possess systemic fixity. (Eco's examples: "presupposition, prediction of ordinary contexts, rules for felicity conditions, and so on.") Finally, he declares it "epistemological fanaticism" (24) to embrace either pole of the open/closed debate, yet decidedly inclines toward the closed end in depicting his compromise position. While those on other end also may be understandably classed among the fanatics, there seems to be a middle ground between the two that is actually configured much like the end opposite of systemicists like Eco.

Up From Systemics: Semiosystemics

In theory as opposed to application, Barthes's essay, "Textual Analysis: Poe's 'Valdemar'," outlines a means for comprehending a "paradoxical" view of system that is parallel in many respects to the proposal outlined above for a dynamic form of systemics that could be called a "semiosystemics". Barthes does this by contrasting two "tendencies" in contemporary structuralist analysis of narratives. One "seeks to establish a narrative model - which is evidently formal - , a structure or grammar of narrative" (172). Once this paradigm is in place, it supposedly can function as a "narrative model" according to which "each particular narrative will be analysed in terms of divergences" from "all the narratives in the world." But, within the other orientation, narratives are categorized "under the notion of 'text', space, process of meanings at work." The narrative is not considered "a finished, closed product"; it is an intertextual "production in progress...and thereby articulated with society and history in ways which are not determinist but citational." (This latter form, which Barthes calls "Textual Analysis," will be discussed in greater detail in Lecture Seven.) Briefly, Barthes proposes this approach as endeavoring not to "describe [or "record"] the structure of a work." His goal will be that of "producing a mobile structuration of the text" - one that is "displaced from reader to reader throughout history." "Textual analysis does not try to find out what it is that determines the text," Barthes concludes, "but rather how the text explodes and disperses."

Barthes's claims for a free-form literary semiotics do rely on consequential elements of systemicity, but they are incorporated flexibly. His method attempts to "locate and classify without rigour, not all the meanings of the text...but the forms and codes to which meanings are possible" (172-3). "The avenues of meaning," in other words (173). For Barthes, this designation establishes an "operating field" parallel to the systemic concept of the delimited arena and its constituent logic. Moreover, Barthes's system is opposed to the static sense of structure which so often spurs protests from process-oriented systemicists.

While Barthes's depiction of this system is fairly vague, he does at least repeatedly characterize it as unhindered by structuralist boundaries. "What founds the text is not an internal, closed, accountable structure," he contends, but instead, "the outlet of the text on to other texts, other signs; what makes the text is the intertextual" (174). Barthes proposes "the conjunction of two ideas which for a long time were thought incompatible: the idea of structure and the idea of combinational infinity." "The conciliation of these two postulations is forced upon us now because language, which we are getting to know better, is at once infinite and structured." Later in the essay, Barthes reiterates this stance by imagining the linking of two sets of seemingly mutually exclusive oppositions: "narrative...in the process of self-construction," he says, "implies at once structure and movement, system and infinity" (191). This position effectively identifies what also could be called an "infinite systemicity," although it might seem that an infinity of that nature essentially dissolves the notion of systemic confinement.

Barthes constructs this intersection of infinity and system through a code-based analysis not unlike that discussed in Lecture Three. In this instance, Barthes does not assess "codes" "in the rigorous, scientific, sense of the term" (191). He imagines codes as "simply associative fields, a supra-textual organization of notations which impose a certain idea of structure." Through this assumption, Barthes's decoding "is not a question of delivering the 'structure' of Poe's story, and even less that of all narratives." He will, to the contrary, engage in a process of "simply... returning more freely, and with less attachment to the progressive unfolding of the text, to the principal codes that we have located."

Barthes's treatment of the Text as an open system provides a viable area of pursuit for a critical semiotics that is consonant with the proposed conceptual blending of system and process into a semiosystemics. While obviously far less systemic than Hjelmslev would have proposed, Barthes's model does have the advantage of not falling prey to the allure of static systemicity that is so common. Like his decidedly unsystematic development and deployment of code analysis in S/Z (also discussed in Lecture Three), Barthes's systemics is chaotic, poorly theorized, and zestily idiosyncratic. And it is for exactly these reasons that it provides so much potential for future study of sign systems not unlike the semiosystemics anticipated by Firth, Halliday, Culler, Ventola and others.

By positing the separability of the signifier and the signified, Barthes moors the Work to the signified, while the Text perpetually floats unfettered by virtue of only associating with the signifier. While "the work closes on a signified," the "dilatory" Text "practices the infinite deferment of the signified" ("Work" 158). It remains "an irreducible (and not merely an acceptable) plural" (159). Yet, while he imagines a sense of the Text that appears radically polysemic, Barthes's modeling perimeters operate within the system's comfortable confines (a familiar stance among the discussion of semiotics). This can be seen in Barthes's use of the notion of the "field". The Text's "field is that of the signifier and the signifier must not be conceived of as 'the first stage of meaning', its material vestibule, but, in complete opposition to this, as its deferred action" (158). Moreover,

the infinity of the signifier refers not to some idea of the ineffable (the unnameable signified) but to that of playing; the generation of the perpetual signifier (after the fashion of a perpetual calendar) in the field of the text (better, of which the text is the field) is realized not according to an organic progress of maturation or a hermeneutic course of deepening investigation, but, rather, according to a serial movement of disconnections, overlappings, variations.

In this way, the Text is "like language," in that "it is structured but off-centred, without closure" (159). This leads to what Barthes calls "a paradoxical idea of structure: a system with neither close nor centre." "The metaphor of the Text is that of the network;" he argues, "if the Text extends itself, it is as a result of a combinatory systematic" (161).

What is it about the "network" that evokes greater growth potential and accountability than "system"? The metaphorical connection of generation might serve as one explanation. After all, networks are associated with organic adaptation, restraint without absolute containment, asymmetry as well as symmetry, and perhaps most importantly, with individual constructedness. As Halliday notes, from a conceptual standpoint "the network is open-ended" (3). Systems, however, suggest rigidity, control, stasis, containment, and disempowerment of the individual. Halliday's distinction between system and set is helpful here: "the range of possibilities in a closed choice is called technically a SYSTEM, that in an open choice a SET" (Linguistic 22). "As a reminder of this distinction," he notes, "we often talk of 'closed system' and 'open set'." Halliday also characterizes system as "an abstract representation of a paradigm," and it is this abstraction, along with its connotations of closure, that may account for its seeming distance from the realm of actual experience ("Language" 55). Furthermore, it seems to evoke a yearning for total encapsulization of semiosis, a realization of the ideal into something akin to Chomsky's notion of a "perfect system" (1).

Perhaps a nomenclature that is even less forbidding could be employed to avoid the stigma associated with systemicity. Hasan offers a term - "texture" - that might be more suitable. "A random string of sentences differs from a set of sentences representing a (part of a) text, precisely in that the latter possesses the property of texture," she remarks (228). In this example, she specifies texture as a condition in which "the lexicogrammatical units representing a text hang together" and create "linguistic cohesion within the passage." Texture, she contends, "is what makes the sentences of a text cohere" (241).

By including the constituent force of semiosic activity into an ever-changing grid of process-oriented systemicity, semiotics may well move to the next level into a systemics that could be founded under the rubric of a dynamic model - perhaps a semiosystemics. It would be a different type of networking indeed.


Armstrong, Nancy. "Inside Greimas's Square: Literary Characters and Cultural Restraint," The Sign in Music and Literature, Ed. Wendy Steiner (Austin: University of Texas Press, 1981).

Bailey, R. W. "Negotiations and Meaning: Revisiting the 'Context of Situation,'" Systemic Perspectives on Discourse, Vol. 2, Selected Theoretical Papers from the 9th International Systemic Workshop, Ed. James D. Benson and William S. Greaves (Norwood, NJ: Ablex Publishing Corp., 1985): 1-17.

Chomsky, Noam. The Minimalist Program (Cambridge: The MIT Press, 1995).

Deely, John. The Human Use of Signs, Or, Elements of Anthroposemiosis (Lanham, MD: Rowman and Littlefield Publishers, 1994).

Eco, Umberto, with Richard Rorty, Jonathan Culler, and Christine Brooke-Rose. Interpretation and Overinterpretation, Ed. Stefan Collini (Cambridge: Cambridge University Press, 1991).

Fawcett, Robin. "Foreword," Eija Ventola, The Structure of Social Interaction: xvii-xix.

Firth, J. R. "Personality and Language in Society," Papers in Linguistics 1934-1951 (London: Oxford University Press, 1957): 177-189..

---. "The Semantics of Linguistics cience," Papers: 139-147.

---. "The Use and Distribution of Certain English Sounds," Papers: 34-46.

Fries, Peter H. "How Does a Story Mean What it Does? A Partial Answer," Systemic Perspectives on Discourse, Vol. 1, Selected Theoretical Papers from the 9th International Systemic Workshop, Ed. James D. Benson and William S. Greaves (Norwood, NJ: Ablex Publishing Corp., 1985): 295-321.

Goffman, Erving. Frame Analysis: An Essay on the Organization of Everyday Experience (New York: Harper and Row, 1974).

Greimas, A. J. "The Interaction of Semiotic Constraints," On Meaning: Selected Writings in Semiotic Theory, Trans. Paul Perron and Frank Collins (Minneapolis: University of Minnesota Press, 1987): 48-62.

---. Maupassant - The Semiotics of Text, Trans. Paul Perron (Amsterdam: John Benjamins Publishing Co, 1988).

---. Structural Semantics: An Attempt at a Method, Trans. Daniele McDowell, Ronald Schleifer, and Alan Velie (Lincoln: University of Nebraska Press, 1983).

Halliday, M.A.K. "A Brief Sketch of Systemic Grammar," Halliday: System and Function in Language, Ed. G. R. Kress (London: Oxford University Press, 1976): 3-6.

---. "Deep Grammar: System as Semantic Choice," Halliday: 88-98.

---. "Language in A Social Perspective," Explorations in the Functions of Language (London: Edward Arnold, 1973): 48-71.

---."Systemic Background," Systemic Perspectives on Discourse, Vol. 1: 1-15.

---. "Typology and the Exotic," Patterns of Language: Papers in General, Descriptive and Applied Linguistics (London: Longman, 1966).

Halliday, M.A.K. with A. McIntosh and P. Strevens. The Linguistic Sciences and Language Teaching (London: Longman, 1964).

Hasan, Ruqaiya. "Meaning, Context and Text: Fifty Years After Malinowski," Systemic Perspectives on Discourse, Vol. 1: 16-49.

Hartman, Geoffrey. 'The Interpreter's Freud," Easy Pieces (New York: Columbia University Press, 1985): 137-154.

Heath, Stephen. "Translator's Note," Roland Barthes, Image-Music-Text (New York: Hill and Wang, 1977): 7-11.

Hjelmslev, Louis. Prolegomena to a Theory of Language, Trans. Francis J. Whitfield (Madison: University of Wisconsin Press, 1961).

Kress, G. R. "Introduction," Halliday: vii-xxi.

Martin, James R. "Process and Text: Two Aspects of Human Semiosis," Systemic Perspectives on Discourse, Vol. 1: 248-274.

Reddy, Michael. "The Conduit Metaphor - A Case of Frame Conflict in Our Language about Language," Metaphor and Thought, Ed. Andrew Ortony (Cambridge: Cambridge University Press, 1979): 284-324.

Schleifer, Ronald. A. J. Greimas and the Nature of Meaning: Linguistics, Semiotics and Discourse Theory (Lincoln: University of Nebraska Press, 1987).

---. "Introduction," A. J. Greimas, Structural Semantics: i-lvi.

Stewart, John. "The Symbol Model vs. Language as Constitutive Articulate Contact," Beyond the Symbol Model: Reflections on the Representational Nature of Language, Ed. John Stewart (Albany: State University of New York Press, 1996): 9-68.

Ventola, Eija. The Structure of Social Interaction: A Systemic Approach to the Semiotics of Service Encounters (London: Frances Pinter Publishers, 1987).

Wilson, John. "Discourse Worlds and Representation," Beyond the Symbol Model: 279-302.

Next Lecture:
Finite Infinite Semiosis


Scott Simpkins, "Reeling in the Signs: Unlimited Semiosis and the Agenda of Literary Semiotics," Versus: Quaderni di studi semiotici 55/56, 2 (Gennaio-Agosto 1990), 153-173;
Umberto Eco, The Open Work, trans. Anna Cancogni (Cambridge: Harvard University Presss, 1989).

copyright Scott Simpkins 1997
Send comment or question to Scott Simpkins: scotts@unt.edu
Go to Cyber Semiotic Institute home page
Go to Course Outline
Go to Lecture: One or Two or Three or Four or Six or Seven or Eight