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

Critical Semiotics

Instructor: Scott Simpkins

Lecture Two - Two Extensive Critiques of Semiotics.

Assigned Readings:

John Stewart, Language as Articulate Contact: Toward A Post-Semiotic Philosophy of Communication (Albany: State University Press of New York, 1995).

Robert Hodge and Gunther Kress, Social Semiotics (Ithaca: Cornell University Press, 1988).


"Disciplines, unlike cows, yield least when most contented."
Hodge and Kress, Language as Ideology

No "Basic" Concepts in Semiotics

If semiotics is considered as a wide-ranging and heterogeneous discussion, then perhaps it follows that it cannot possess basic concepts. Despite the tacit refusal of most semiotic studies to acknowledge this condition (as seen in titles like Basics of Semiotics by John Deely [see Lecture One]), this condition undeniably remains - and remains irresolvably so. Nonetheless, it can be rewarding to analyze the various attempts by semioticians to discuss what clearly are recurrent "semiotic" assumptions and terminologies (denotation, context, signifier, syntagm, interpretant, etc.) among studies associated with it. Of course, this approach does not correspondingly imply that recurrence constitutes anything like a consensus, a ground on which basic concepts could be drawn. Even a cursory perusal of any two or more studies in this area will reveal this to be the case (as do studies by the same author, as Peirce demonstrates so aptly).

The two studies addressed here - Language As Articulate Contact and Social Semiotics - offer good examples of responsive and nuanced projects designed to move beyond the imprecise but handy assumptions that often undermine semiotic studies. Stewart's back-cover blurb reveals this with the contention that "From the perspective of communication theory, this book extends some features of the postmodern critique of representationalism to develop a post-semiotic account of the nature of language as dialogic." Rather than begging the question about the homogeneity of semiotics, Stewart identifies his undertaking as one that isolates certain "features" that can be crafted into a rendition of some semiotic studies. Hodge and Kress's cover also characterizes "semiotics as an evolving theory." For them, "texts and contexts, agents and objects of meaning, social structures and forces, and their complex interrelationships together constitute the irreducible object of semiotic analysis."

However, these approaches to semiotics also are vulnerable to critiques similar to the ones they offer themselves. Again, though, identifying the presumptions behind these two studies will hardly negate whatever value they offer to future semiotic discussions. To the contrary, this examination will point to areas that are ripe for explorations to come.

(For those interested in more extensive critical descriptions of these books, please consult my reviews of Stewart - "The Semiotics of Post-Semiotics," The Semiotic Review of Books 7.1, January, 1996, 8-10 - and Hodge and Kress - "Reading the Social Text," The American Journal of Semiotics 7, 1-2, 1990, 145-151.)

Stop Signs: Stewart

John Stewart condemns semiotics for relying upon the symbol model for its basic paradigm, announces its consequent obsolescence, and asserts that the only possible hope hinges on a "post-" manifestation of semiotics. (Stewart's use of "the symbol model" refers to an eclectic assortment of paradigms sharing a common feature configured as a sign.) Stewart reflects this sentiment when he identifies the main shortcoming of semiotic studies of language as the assumption that language is "fundamentally a system of signs or symbols" (3). Consequently, "semiotic accounts of the nature of language are crippled," he asserts, "by . . . serious problems of plausibility, coherence, and applicability" (xii). These "problems" result from a handful of what Stewart refers to as "theoretical commitments" that, he argues, hamper a productive and accountable analysis of the nature of linguistic communication.

The commitment that most significantly hinders semiotic analysis for Stewart is the two-world "problem." This results from the belief in "a fundamental distinction between two realms or worlds, the world of the sign and the signifier, symbol and symbolized, name and named, word and thought" (6-7). Such a distinction does not "coherently" account for a comprehendable interface between two realms that cannot co-exist simultaneously, Stewart argues, and an ontological impasse results. Since these worlds donÕt intersect, this position must rely upon the assumption of an unnecessary plane that impedes the intelligibility of a semiotic account of language. "Distinguish[ing] . . . between two worlds alters the historical sense of the term world as the single coherent sphere that humans inhabit" (105). We don't live in the world of conceptual signs, this position holds, and as a result, we can't conceive of such a sphere. Nor can it be used to adequately explain a pragmatic sense of language usage that we all draw upon every day in "conversation" consisting of "two-person dialogue in real time" (xiii).

The other commitments Stewart aligns with semiotics extend from this initial assertion. Stewart attacks the practice of partitioning, identifying small (or the smallest) units of a given aspect of language as a means of breaking it down to what are assumed its essential elements. This practice, he insists, has spurred a false assumption of primary or foundational segments in language. It also has an accompanying skewed view of a synergistic process that is insufficiently represented if its disparate constituent elements are not considered as a whole.

Representationalism similarly extends the independent-unit fallacy. It posits a two-world combination that resists our tangible conception of only one existent world, the one we inhabit consciously. The particular shortcoming of this commitment arises, for Stewart, in the case of represented concepts (like negations) that would exist only in the world of representation without an accompanying real-world correlative. In other words, this situation "ultimately keeps a wedge driven between the two worlds . . . because one entity of a given ontological status cannot coherently be said to 'represent' another entity of the same ontological status" (103). Stewart stresses analyzing "living language" (104) to restrict his analyses to the realm of a practical, and dialogic, discourse, as opposed to proposing a language system separate from its demonstrable use.

His final commitment attacks the instrumental view of language as a semiotic system. Instrumentalism, he says, "hypostatizes what is lived as event and imports the subject-object distinction into language scholarship" (29). Language is rendered even further distant from its social facet if it is conceived as not only a system, but a system with a pedestrian use-function. It should be considered instead, he maintains, as a decidedly human practice characterized by common-sense competence gained through interpersonal communication.

"Articulate Contact"

A great deal of Stewart's discussion is undeniably attractive. Indeed, he constructs an alternative to semiotics that would seem to answer many of the charges levelled against it in recent years. (In part, this is due to its apparent linkage with structuralism - a contention Daniel Chandler makes, for instance, when he avers that "semiotics is difficult to disentangle from structuralism.") Stewart emphasizes demonstrability in the analysis of communication and posits that it is impossible to link the real world and the conceptual one. These positions, along with his preference for studying "actual" language use over a sterile, lifeless system, are clearly compelling. Yet, at the same time, to propose that this form of analysis has to exist beyond (or after) semiotics clearly ignores the possibility of "thickening" the discussion of semiotics (something that has taken place, at least in limited - and limiting - ways, since the 1960s). In fact, as Stewart proceeds to narrate a tradition of symbol-model articulations of semiotics throughout history, it becomes increasingly apparent that he can sustain his critique of semiotic's commitments only by a very selective "thinning" of the accounts he draws upon. And the further he goes the thinner his history becomes. Again, this is virtually a given in any analysis that stakes out a territorial claim and then works it vigorously to generate sufficient yield. Yet, once more, a possible contribution to semiotics may be found by producing a different "yield" by retaining Stewart's leeriness of under-scrutinized conceptualizations that so often litter semiotic analyses.

First, however, Stewart's modus operandi for creating his "credible alternative" (ix) to semiotics: an analysis of a post-semiotic "articulate contact" that consists of the decidedly human practice of interpersonal communication. Stewart stresses the interpersonal nature of such exchanges, the give-and-take of dialogue that typically is unscripted and even chaotic in structure. His approach also is centered on generic social tasks of some kind (the job interview, the apology, etc.). But, in the process of touting his own perspective, Stewart reveals the investments underlying his endeavor in the ways he praises it. It's "credible" (ix). Its foundation is "coherent." It results in a rendition of language usage that constitutes "a plausible whole." And, it's grounded on viewpoints that enjoy "a significant contemporary consensus among philosophers and communication theorists" (x). While these value-rich descriptions sound reasonable (and even desirable), there is little of substance underlying them. This is a major weakness of Stewart's project: a conceptual sand base on which he tries to construct a vast, decidedly rigid scaffolding.

Stewart's proposed communicative model is essentially the opposite of the commitments he outlines. "There is only one human world and it is linguistic" (30), he says, which effectively negates the conceptual aspect of the sign model that appears frequently in semiotic discussions. Stewart also stresses analysis focusing exclusively on "events of speech communicating" (30) for explanatory models of the nature of language. He proposes analyzing texts derived from instances of human communication, texts that would presumably consist of "naturally-occurring interchanges" (17). By studying "language as it is lived" (19), he endeavors to conceptualize language with an interactive basis. He views language use as a form of community instead of a lifeless system. To Stewart, conceiving language as a system leaves its components microscopically (and "unnaturally")taxonimied in accordance with artificially mechanical, determined laws, not "actual" social practice. Furthermore, "language as living event can best be understood," he asserts, "by recognizing that its first business is contact." It is a decidedly human (and humanistic) enterprise, and by no means something that is primarily instrumental by design.

Stewart's alternative to semiotics is based on the assumption that "understanding is a mode of being manifested in concrete events of conversing and that ultimately these events are what the term language labels" (112). Studying language from a systemic standpoint, on the other hand, leads to the sterile segmentation mentioned above. This misrepresents the interactive gestalt of language use as a social practice, as opposed to "languaging" which addresses "understanding in events of speech communicating" (123). "Efforts to analyze syntactic or semantic aspects of . . . the 'system' of language need to be broadened to acknowledge both the indivisible interrelationships between the verbal and the nonverbal and the inherently relational nature of events of articulate contact," he contends. Only by framing language as an entity constituted by human interaction can an anlysis reflect its existence as a fluid, heterogeneous human undertaking. "The anchor for understanding languaging should be the contact event as its participants live it," Stewart declares (125). "Little purpose is served by focusing one's explicative energy exclusively on reducing language to its atoms."

Signs in the Real World

Reflecting a partial alignment with post-structuralism, Stewart opines that instead of mediating between two worlds, "language is constitutive." Language "does not represent world," he contends, "but builds or develops it" (31). In effect, Stewart bases his entire schema of post-semiotics on this hypothesis. "The study of reported speech can provide insights into the basic processes of understanding and communication, conceived of as social, dialogical processes" (188), he suggests. Accordingly, his means of information gathering centers on the study of transcribed interactive speech.

Stewart goes so far as to inject an ethical aspect of his account of a post-semiotic orientation, suggesting that the analysis of language as a system discounts, or even entirely neglects, the human side of semiosis. "Semiotic accounts of the nature of language permit discourse to be disconnected from its ethical and ontological consequences," he argues. But, "this post-semiotic account permits no such disconnection; it points toward the intimate connection between human speech communicating and human being" (130).

By grounding post-semiotics in real-time human interaction, Stewart attempts to bring communication analysis to the realm of the actual, or single-world, perspective. This view can be experienced directly as opposed to the conceptual world that he considers as immaterial in the two-world view. This approach yields a tangible "facticity," an "acknowledgement or affirmation" that this world's existence takes place "separate from the viewer" (117). Such a world consists of a reality affirmed through interactive language use by actual beings separate from a conceptual plane or from the constraints of a "system" that exists only conceptually as well. "Humans participate in the constituting of the coherent spheres we inhabit," Stewart says, "by engaging both proactively and responsibly in the play of language events" (119).

Stewart's proposal also endeavors to focus on language study as an undertaking that, from his perspective, should always remain in the realm of the tangible, again, as opposed to the intangibly conceptual. Language, he says, should be considered as "constitutive or productive of (necessarily partial, tentative, and changing) ways of understanding rather than reproductive of cognitive states, things, or other units of language" (125). Humans cannot prove the existence of these states, palpably experience abstract things (like negations), or detect segmentation when speaking in a string of unpremeditated units. These things essentially cannot exist for us in a "real" way. "Speech communicating is a principal not a surrogational dynamic," he suggests.

Actually, a lot can be said for Stewart's position here. Many complaints have been raised regarding the neglect of human subjectivity that seems to result from systemic analyses. One of the best illustrations of an arguably parallel instance can be found in Michael RiffaterreÕs attack on Claude Lévi-Strauss and Roman Jakobson's structural analysis of a poem. In "Describing Poetic Structures: Two Approaches to Baudelaire's 'Les Chats'" (1966), Riffaterre allows for the "reasonable assumption" that "there is a causal relationship between the presence of [specifically poetic] features in the text and our empirical feeling that we have before us a poem" (26). But, he suggests, this can be taken much too far beyond the realm of human detectability. Thus a microscopic study of linguistic features of a poem's system can detect effects that are essentially beyond human detection. In this sense, such features are, in a significant way, not there - even though they appear to function as part of the microstructure of the text's signifying system on different linguistic planes. Riffaterre also admits the possibility that a "poem may contain certain structures that play no part in its function and effect [on the reader] as a literary work of art." He adds that "there may be no way for structural linguistics to distinguish between these unmarked structures and those that are literarily active." And, "conversely, there may well be strictly poetic structures that cannot be recognized as such by an analysis not geared to the specificity of poetic language" (28). Riffaterre eschews linguistic elements that are, in his view, "inaccessible to the normal reader" and argues that even the identification of elements that are accessible "do[es] not explain what establishes contact between poetry and reader." "No grammatical analysis of a poem," he concludes, "can give us more than the grammar of the poem" (36). Stewart takes this approach when he tries to account for the human use of language which may exist and function separately from what an emphasis on language as a system is capable of revealing.

One of the obvious benefits of Stewart's assertion here is that he shifts semiotics (or a post-semiotics) toward a felt enterprise. This experience may indeed seem more relevant because it's familiar to us while a sub-atomic anatomization of language usage from a systemic standpoint may come across as alien. It analyzes language usage in slow-time, as opposed to the real-time blur of languaging that usually occurs during human semiotic interaction. However, it is undeniable that any such social interaction ultimately takes place as a form of system, one that is rule-bound (to whatever extent of formal regimentation of these rules). Moreover, this interaction is - to draw upon models of the sign like those generated from disparate commentary by Peirce and Saussure - implicitly conceptual in nature. Denying these aspects of language usage will not make semiotics go away.

Similarly, language use appears based on a predominantly unconscious internalization of these rules and paradigms. This occurs to such an extent that humans seldom consider speaking as using language from the instrumental perspective Stewart decries. "Humans cannot live in the subject-object relationship with language that the tool analogy requires," Stewart declares. "Insofar as world is linguistic, we inhabit or live in our language; we do not simply use it as a tool" (126). Yet, studies such as Sigmund Freud's The Psychopathology of Everyday Life, Kenneth Burke's A Grammar of Motives, or Erving Goffman's The Presentation of Self in Everyday Life amply demonstrate that systematic analysis of "lived" experience can offer a great deal of insight into supposedly unscripted or unconscious social behavior. To deny the instrumentality or systemic facet of this behavior requires a refusal to acknowledge this facet of semiotics, as Stewart demonstrates here. Ultimately, he stresses a "connectionist" view of language analysis that avoids what he sees as the pitfalls of representationalism. Stewart accomplishes this by emphasizing language in its forms of human usage as opposed, again, to a systemic view devoid of agency.

Forget Pluralism

It might appear logical that a "post-semiotics" could provide a pluralistic improvement over semiotics, a use of its so-called key concepts that seeks to employ them more responsively. Yet Stewart adamantly opposes blending "a coherent nonrepresentational account of linguistic or discursive reference" (238) with a position grounded on symbol models and their concomitant implications. Stewart depicts dire consequences should "semiotic and post-semiotic views collide" (198) because semiotics always taints its post- manifestation. A good illustration of this collision is found in his analysis of the shortcomings of Kenneth Burke's writings on language. Burke's attempt to meld a conception of language as both "dynamic and processual" and yet somehow composed of "identifiable, discrete units" generates a "pervasive tension." Stewart charges that tension of this nature forestalls any possible progress in semiotic analysis.

Stewart insists that a pluralistic enterprise geared toward illuminating "the basic nature of language itself" would fail to yield "coherent and useful" results (113). "Language cannot be coherently treated as simultaneously a world-constituting, characteristically human way of being, and as a system that is instrumentally employed by already -constituted humans to represent aspects of their worlds and accomplish other goals" (113). For Stewart, semiotics remains uninterested in concerns such as "the relationship between the individual and the social, the dynamics of narrative collaboration, the discursive development of subject matter, or the conversational achievement of intimacy." Consequently, the two disciplines can never meet.

This, of course, is absurd. Numerous semiotic discussions focus on the same concerns that Stewart relegates exclusively to the domain of post-semiotics. It becomes evident while proceeding through Language as Articulate Contact that Stewart has to construct narrow views of the vast array of semiotic studies in order to characterize it in this fashion. For instance, he claims that the two-world disjunction has to be discarded. "No contemporary scholar would seriously contend that one can specify any sort of one-to-one correspondence between specific signifier and specific signified" (21). However, it's unlikely that he could find someone writing on semiotics to support this claim. Even semioticians who argue that the decoder's practice can be controlled or limited would not assert that varifiable correspondences of the kind Stewart identifies are possible.

A Post-Semiotic Demonstration

A revealing illustration of Stewart's assessment of "articulate contact" appears in his commentary on a transcription of "naturally-occurring interchanges" analyzed by Douglas Maynard (17). This exchange is pedestrian and extremely "unplanned" which, to Stewart, proves the unsystematic nature of language as it is used in real life. "These transcripts capture something much closer to language as it actually occurs than the examples commonly used by philosophers, linguists, and semioticians to support their claims about the nature of language" (19), Stewart maintains. Since these are examples of "language as it is lived," he continues, "these examples are surely more paradigmatic than the hypotheticals typically discussed." Such conversations are "relatively 'spontaneous' and 'natural.'" And, they reveal the inadequacy of a "description of the nature of language offered by those who characterize it as a system of signs or symbols functioning representationally and instrumentally."

By emphasizing "articulate contact," Stewart reveals components "that would not be apparent if [one] were to treat this language simply as the systematic use of symbols" (127). Stewart employs Maynard's notion of "perspective-display sequence" to guide this analysis of conversation "operating syntactically, semantically, and pragmatically" (129). Maynard suggests, in Stewart's words, that "conversation partners use this strategy . . . to adapt a personal opinion to their listener's frame of reference." This is employed by "first soliciting the other's opinion and then producing one's own report in a way that takes the other's into account."

The conversation considered involves the problem of dangerous bicyclists and two university students' attempts to cope with campus overcrowding. From this standpoint, the opening utterance - So - and the remainder of the first three lines of the discussion entail an initial "perspective-display invitation" (127). This is followed by a reply which then elicits a statement of opinion by the first speaker:

1. John: So what do you think about the bicycles on campus?
2. Judy: I think they're terrible.
3. John: Sure is about a million of 'em.

(Note: I have not maintained the transcription conventions employed in Maynard's rendering of this conversation.) In his examination of Maynard's analysis, Stewart asserts that this approach uncovers human practices of communication the symbol model would neglect. Maynard, for example, "finds evidence about the relationship displayed between the individual and the social, the dynamics of narrative collaboration, the discursive development of subject matter ('bicycles'), and the achievement of intimacy" (128). And, he does so "without getting caught up in any effort to analyze 'signifiers' and 'signifieds'." This is crucial to Stewart's condemnation of the symbol model approach. Maynard, he insists, "notices what he does in this discourse because he recognizes that these interlocutors are coconstructing the world they share in the aural-oral contact." Moreover, they "both produce stories that are 'wrapped in' the other's parallel story" to constitute what Stewart calls "collaborative construction." The conversational stresses that mark their exchanges "[reveal] another level of the interlocutors' intimacy, one embedded in the aural-oral dimensions of intonation, emphasis, and facial expression" (129). Finally, "each speaker displays a world open to the other's participation, and both positively affirm the other's involvement in their worlds."

Although the depiction of Stewart's account of an "articulate contact" is necessarily truncated here, it should suffice to reveal his approach. Once again, he has to engage in considerable truncation of "semiotics" himself in order to make post-semiotics significantly different from, as well as superior to, semiotics. In effect, he employs a form of semiotic analysis (if one could precisely determine what that involves) without using the terminology and concepts frequently associated with semiotics (no "signifiers" or "signifieds," etc.). Through several fuzzy distinctions that set up ontological roadblocks to derail symbol-model based inquiry, he constructs in its place essentially the same approach with different terminological distinctions.

Thus, Stewart is simply calling for the type of semiotics that already focuses on the very issues he claims are beyond its scope. And one of the best examples of this is Hodge and Kress's Social Semiotics. In fact, take away the frequently myopic critique of semiotics from Stewart's book and he makes many of the same arguments found in Hodge and Kress's study.

The Social System: Hodge and Kress

Hodge and Kress's earlier study, Language as Ideology (1979), outlined a "critical linguistics" that stressed "the primacy of the social dimension in understanding language structures and processes." Their stated goal was to yield "a theory of language whose aim was to provide an illuminating account of verbal language as a social phenomenon." They particularly wanted to assist "critical theorists in a range of disciplines . . . who wanted to explore social and political forces and processes as they act through and on texts and forms of discourse" (vii). (Although this sounds like just what Stewart has in mind, significantly, he doesn't refer to either study by Hodge and Kress.) Hodge and Kress recount feeling unnecessarily constrained by focusing on verbal language alone (a similar problem with Stewart's study from a semiotic standpoint). They subsequently broadened their focus to consider "all sign systems" in Social Semiotics. "Meaning resides so strongly and pervasively in other systems of meaning, in a multiplicity of visual, aural, behavioural and other codes, that a concentration on words alone is not enough." As a result, they attempt to produce "a general theory of the social processes through which meaning is constituted and has its effects" (viii). Hodge and Kress, again like Stewart, also endeavor to redirect semiotics by emphasizing social interaction over system (although ultimately what they offer is the system of social interaction).

In many respects, Hodge and Kress engage in an undertaking not unlike the notion of "critical semiotics" developed here. They suggest, for example, that their goal is to demonstrate that it is "not only . . . possible but . . . necessary to attempt a reconstitution of semiotics" (2). "Equally important," they continue,

a practical semiotics should have some account of the relationship of semiosis and "reality", that is, the material world that provides the objects of semiosis and semiotic activity. Unless semiotics confronts this relationship, it can have no relevance to the world of practical affairs with its confident assumptions about "reality", and it cannot account for the role of semiotic systems in that world. (23)

Ideology and Semiosis

This emphasis on the "social" is exactly aligned with Stewart's concerns, yet Hodge and Kress go far beyond what they consider a needless limitation of focusing only on verbal language. Instead, they draw upon a diverse selection of "texts" ranging from billboards to comic strips, from paintings to transcriptions of actual speech. Much of the selection criteria Hodge and Kress use for their data is evidently based (not surprisingly) on power struggles. It is this emphasis that both enhances and detracts from the value of their contribution. Their declaration of support for this endeavor is revealing:

Decoding classic texts is as crucial an enterprise [to social semiotics] as the elite culture claims after all. However, to do it properly requires systematic study of many kinds of non-classic text, which the same elite excludes from its own definition of culture. (203)

This approach is not unlike the classic Marxist study by Ariel Dorman and Armand Mattelart of the signifying system of Walt Disney's comics, How to Read Donald Duck: Imperialist Ideology in the Disney Comic. Still, Hodge and Kress's emphasis on one facet of social semiotics leads them to examples that prove the point they want to make while neglecting the much larger social spheres also involved. Traffic lights, for instance, are not explored as means for enforcing desirable vehicular flow and pedestrian safety. They are - evidently in equal amounts - "ultimately social and ideological" (37). Because these lights are typically approached from a purely communicational stance, their ideological facet has become obscured, Hodge and Kress contend. Their analysis reflects the argument that "the traditional illustration of traffic lights should be stripped of its implicit ideology of the communication process" (39). "The traffic signals transmit an ideological message as well as particular instructions," they add. "They present a version of society, an image of impersonal rationality operating impartially on behalf of all." The lights thereby signify "power understood as control by one social agent of the behaviour of others."

It is ironic that Hodge and Kress grant such hegemonic control in social semiotics. For, one of their agendas is to persuasively demonstrate how an individual sign user, contra Saussure, can indeed significantly alter a communal sign system (by changing the message of a billboard sponsored by a large, powerful tobacco company, for example). But in the instance of the traffic lights, they conclude: "The behaviour of the participants is constrained by logonomic systems which operate through messages about their identity and relationship, signifying status, power and solidarity" (40). (Hodge and Kress define a "logonomic system" as "a set of rules prescribing the conditions for production and reception of meanings" [4].) They extend this observation further by identifying regimental constructs they call "ideological complexes," loci for disseminating ideologically driven means of social control. In keeping with this position, they posit the communal order as grounded by "characteristic structures of domination" that sign users have to contend with in the course of engaging in social semiosis (3).

A Social-Semiotic Demonstration

One of the best examples, because it's so extensive, of Hodge and Kress's approach is found in their examination of a text whose overall "sign" is changed significantly by a group of decoders. This illustration is especially revealing because it shows how they trace potential subversive power plays at the disposal of the seemingly disempowered receiver of a given sign. This sign use, they contend, is usually conditioned by mutual influences of "reception regimes (rules constraining reception)" and "production regimes (rules constraining production)" (4). Of course, these dynamics would be obviously attractive to Hodge and Kress considering their own ideological concerns.

Their text is a Marlboro cigarette advertisement displayed on a billboard in a "public space" (9) in an undisclosed location. The ad, which shows the "Marlboro Man" smoking a cigarette on a horse, reads: "New. Mild. And Marlboro." Hodge and Kress address various elements that constitute the signifying field surrounding and including the billboard such as the symbolic systems employed, its public setting, and the linguistic conventions involved in an advertisement.

For instance, regarding the billboard as a generic field, they observe:

The original advertisement is a text on a large scale, displayed on a billboard, which is itself mounted on a brick wall in a public space. This indicates one set of logonomic rules immediately: the right to erect a billboard of this size is explicitly controlled by local government laws, and there are agencies which control the appearance of messages in a "public" space such as this.(9)

They extend this approach by exploring the "different kinds of institutional legitimation" entailed by the semiotic conventions of this form of advertisement.

The advertisement's linguistic text is approached in a similar fashion:

The text itself is of a scale and kind which implies the use of significant material resources. The availability of such resources is understood by a reader to be a precondition of the production of such a text and that gives the text a particular status, and places readers in a particular position. (9)

Hodge and Kress proceed in this way to reveal how a powerful private corporation uses advertisements to subtly convey its power within the larger social realm.

Following this fairly conventional approach to a text of this nature, they reconsider the entire text following its alteration by an urban guerilla group, Billboard Using Graffitists Against Unhealthy Promotions, which had significantly changed it. BUGAUP painted a grave headstone on the western landscape, a thick cloud of smoke rising from the cigarette, a sunset and a dollar sign. They also added a label on the cigarette package ("CANCER Sticks") and dialogue in which the smoker's horse remarks "POO THIS MACHO STINKS" and the smoker coughs. The group also changed the ad's slogan from "New. Mild. And Marlboro." to "New. Vile. And a bore." The BUGAUP alteration of the billboard substantially shifts the power wielded by an institutionalized "ideological complex" to control public dissemination of images about its products, Hodge and Kress maintain.

While the original advertisement "positions the pliable, acquiescent reader in a passive role in the act of communication," they argue,

the BUGAUP reading attacks the [logonomic] system in a radical fashion. An individual could deface this advertisement in exactly the same way in a magazine; as a private act it would cause no ripple. By "defacing" a billboard the BUGAUP readers/authors are inserting themselves into a forbidden semiotic role, as communicators of subversive meanings presented publicly, in a public space. (11)

The motives underlying Hodge and Kress's readings of social semiotics are amply demonstrated by this example. Throughout Social Semiotics to varying degrees, they contend that power can be wielded by both the encoder and the decoder in any given instance of semiosis. Hodge and Kress's depiction of ongoing transformations following the release of a sign-vehicle also dramatizes their open approach to social semiotics in which they view the exchange of signs as ceaseless and always undergoing change. "The BUGAUP additions constitute a specifically dialogic text, in which one reading of the original text is reclaimed and incorporated into the text itself," they argue. "However, even after this interaction the flow of discourses will still continue, situating the new text in relation to other agents of discourse and their interests" (12).

The Expense of Ideology

Clearly, the solidity of their approach to social semiotics yields a consistent and homogeneous view of semiosis within a given cultural site. (While, admittedly, this seamlessness comes at the expense of a heterogeneous consideration of other facets that affect and constitute semiosis.)

Furthermore, they frame this notion of the social in another way beyond Stewart's presumption of a "natural" order. They do so by considering the ideologically based power struggles that often subtend semiotic exchanges (a position not unlike that held by the Roland Barthes of his Mythologies period). "Society is typically constituted by structures and relations of power, exercised or resisted," they suggest.

it is characterized by conflict as well as cohesion, so that the structures of meaning at all levels, from dominant ideological forms to local acts of meaning will show traces of contradiction, ambiguity, polysemy in various proportions, by various means. (viii)

While this tight focus on power may be a shortcoming (by imposing a narrow stricture), it nonetheless enables them to develop substantial readings in an arguably limited way. In other words, their contribution to the discussion of semiotics can be said to rest on their emphasis on power relations that may well exist as an integral component of social semiosis. Not "integral" in an essentialistic sense, but rather, as in an element that appears consistently regardless of whether it is intrinsically necessary.

While Stewart would presumably agree with the real-time social emphasis that Hodge and Kress embrace, he would no doubt identify as a flaw their emphasis on the two-world model and their distinction between the semiosic and mimetic planes. Or maybe not, since their model offers a one-world/two-plane conceptual grid, a division that perhaps would not seem divisive to him. They conceptualize the semiosic field as the site on which "the social process by which meaning is constructed" takes place (5). It is the site of "semiotic event(s), linking producers and receivers and signifiers and signified into a significant relationship" (262). The mimetic plane, on the other hand, constitutes "some version(s) of reality as a possible referent." It is, in other words, "the plane in which representation occurs," given that "The message ['the smallest semiotic form that has concrete existence'] is about something, which supposedly exists outside itself. It is connected to a world to which it refers in some way, and its meaning derives from this representative or mimetic function it performs" (5).

An Accretive Semiotics

As I contended in Lecture One, Hodge and Kress offer a model of semiotic discussion that benefits considerably from tentativity. They view Social Semiotics as a "stage" in a developing project. (Their earlier proposal of a "critical linguistics" remains similarly under construction. A second edition [1993] recently appeared featuring a new chapter on developments in this area following the initial publication of their study.) This restraint - one shared to a far lesser extent by Stewart in his proposed contours of a post-semiotics - allows Hodge and Kress to propose narrow, programmatic readings of social semiotic relations. But, they do so in a profitable way by virtue of their hedge on constructing a semblance of totality. They do this in response to the illusory reification of an extremely diffuse and divergent discussion often concretized under the rubric "semiotics" (as also discussed in Lecture One). As a result, they repeatedly endeavor to broaden the viability perimeters of an ongoing and developing enterprise. Semiotics, they contend, can become "more comprehensible for being offered as provisional, a stage in a continuing debate, a continuing struggle for clarification" (36). They add, further, in a fashion that mirrors Julia Kristeva's commentary on "the ethics of linguistics," that "The most scrupulous reading of signs must always be complemented by a scepticism based on an awareness of the inherent slipperiness of meaning in use" (110). To a significant extent, Hodge and Kress adhere closely to this scruple. However, the ideological suspicion behind their hermeneutic often comes across as a little too suspicious.

A good demonstration of this approach is their observation that semiotic studies often focus on synchronic models of static signifying systems, a "structure" that an entity is presumably somehow built upon. To supplement this emphasis, they suggest adding other plane models that will help to diversify such an undertaking. "A diachronic account of a tradition frees the reader from the oppressive sense that it is monolithic, unchanging, without inconsistencies" (36). Once more, to return to a point raised in Lecture One, they propose a concept such as "mainstream semiotics" not to posit a concrete discipline or science (which, they agree, would be an "oversimplification" [13]). Instead, they do so as a means of constructing a temporary ground for their own discussion. At times, though, they do give in to the admittedly alluring temptation of convenient generalizations that so often hobble semiotic discussions, as when they argue that

Traditional semiotics likes to assume that the relevant meanings are frozen and fixed in the text itself, to be extracted and decoded by the analyst by reference to a coding system that is impersonal and neutral, and universal for users of the code. Social semiotics cannot assume that texts produce exactly the meanings and ef- fects that their authors hope for: it is precisely the struggles and their uncertain outcomes that must be studied at the level of social action, and their effects in the production of meaning. (12)

This assertion carries a lot of rhetorical suasiveness until its presumption about a totalized "semiotics" is challenged. For it certainly in no way accurately depicts a consensus (if it could be said that there were one) among the discussants of semiotics. Likewise:

"Mainstream semiotics" emphasizes structures and codes, at the expense of functions and social uses of semiotic systems, the complex interrelations of semiotic systems in social practice, all of the factors which provide their motivation, their origins and destinations, their form and substance. It stresses system and product, rather than speakers and writers or other participants in semiotic activity as connected and interacting in a variety of ways in concrete social contexts. (1)

Hodge and Kress also allow themselves to fall into the easy assumption of the "basics" of semiotics, which (as mentioned earlier) is a convenience that carries with it considerable imprecision. For example, in a foundational chapter titled "The Founding Fathers Revisited," they "give a critical reading of some of the founding fathers and founding concepts of modern semiotics from [the social semiotic] point of view" (13). Still, by substantially framing this review of basic principles as selective and subject-positioned, they do present their conception as much more subjective than others in the semiotics discussion who propose the existence of a "basic" conceptual ensemble. "We want to contest one particular version of history which underpins a specific and limiting conception of what semiotics was and is, and should be and do," they observe. "We will refer to this as the dominant tradition without implying either that it is all of a piece or continuous with itself as it reached back to claim its past" (13).

Ultimately, Hodge and Kress provide one of the most flexible and fluid accounts of something akin to a "critical semiotics," despite some potential problems. While their readings often mire in ideologically based reductions, they still offer illuminating models for a progressive semiotic discussion based on the very inclinations that John Stewart claims semiotics cannot fruitfully accommodate.


Barthes, Roland. Mythologies, Trans. Annette Lavers. New York: Hill and Wang, 1972.

Burke, Kenneth. A Grammar of Motives. Berkeley: University of California Press, 1969.

Dorman, Ariel and Armand Mattelart, How to Read Donald Duck: Imperialist Ideology in the Disney Comic, Trans. David Kunzle (New York International General, 1975).

Freud, Sigmund. The Psychopathology of Everyday Life. Standard Edition of the Complete Psychological Works of Sigmund Freud. Trans. under the general editorship of James Strachey in collaboration with Anna Freud, assisted by Alix Strachey and Alan Tyson. 24 vol.s. London: Hogarth Press, 1953-74.

Goffman, Erving. The Presentation of Self in Everyday Life. New York: Doubleday, 1959.

Hodge and Kress, Language as Ideology, 2nd ed. London: Routledge, 1993.

Jakobson, Roman and Claude Lévi-Strauss, "Les Chats de Charles Baudelaire," L'Homme 2 (1962): 5-21.

Kristeva, Julia. "The Ethics of Linguistics." Desire in Language: A Semiotic Approach to Literature and Art, Ed. Leon S. Roudiez, Trans. Thomas Gora, Alice Jardine, and Leon S. Roudiez. New York: Columbia University Press, 1980: 23-35.

Maynard, Douglas. "Perspective-Display Sequences in Conversation," Western Journal of Speech Communication 53 (1989): 91-113.

Riffaterre, Michael. "Describing Poetic Structures: Two Approaches to Baudelaire's 'Les Chats'," Reader-Response Criticism: From Formalism to Post-Structuralism, Ed. Jane Tompkins. Baltimore: The Johns Hopkins University Press, 1980: 26-40.

Lecture Three:
The Implications of Codes


Roland Barthes, S/Z: An Essay, Trans. Richard Miller (New York: Hill and Wang, 1974).
copyright Scott Simpkins 1996
Send comments or questions to Scott Simpkins: scotts@unt.edu
Go to Cyber Semiotic Institute home page
Go to Course Outline
Go to Lecture: One or Three or Four or Five or Six or Seven or Eight