[In the essay below, which was originally published in 1983, Stewart analyzes how Bakhtin's linguistic theories reject the abstract conception of language in favor of a purely social, "practical" one.]

During the period of the New Economic Policy, as Lenin sought, rather abashedly, to approach communism via a new form of "state capitalism," and as the concrete mode of peasant existence was being transformed into the abstractions of industrial labor, the contradictions between synchrony and diachrony, between "sincerity" and "irony," between insistences simultaneously upon meaning and "multivocality" were in full flower. The work of the Bakhtin school may be located within this milieu of contradiction. It is clear that Mikhail Bakhtin's project was not a linguistics but, to use his word, a "metalinguistics," an attempt to avoid an essentialist view of language and to see, within a social and historical frame, the creation and uses of both language and the term "language." [In an endnote, Stewart directs the reader to Gary Saul Morson's essay, "The Heresiarch of Meta," PTL 3, (October 1978): 407-27, for more information on this.] As Bakhtin wrote in Problems of Dostoevsky's Poetics:

The point is not the mere presence of several linguistic styles, social dialects, etc., a presence which is measured by purely linguistic criteria; the point is the dialogical angle at which they (the styles, dialects, etc.) are juxtaposed and counterposed in the work. But that dialogic angle cannot be measured by means of purely linguistic criteria, because dialogic relationships, although they belong to the province of the word, fall outside the province of its purely linguistic study.

Dialogical relationships (including the dialogical relationships of the speaker to his own word) are a matter for metalinguistics. [In an endnote Stewart explains that she follows "the current practice of attributing the works of the Bakhtin school to Bakhtin himself."]

In short, only through metalinguistics could one account for the history and social life of language.

It is important to remember, however, that Bakhtin's meta-position is not so much a move toward transcendence as it is a battle stance, a polemical insistence upon situating theories of language within the constraints of their particular social and historical periods. M. A. K. Halliday has noted that some forms of speech, such as thieves' jargon and tinkers' argot, are shaped in direct opposition to the speech of the dominant class of their times; he calls these forms of speech "anti-languages" ["Anti-Languages," American Anthropologist 78 (September 1976): 170-83]. Analogously, Bakhtin's linguistics is an anti-linguistics, a systematic questioning and inverting of the basic premises and arguments of traditional linguistic theory. It follows that recent attempts to "appropriate" Bakhtin's theories of language into the tradition he rejects are largely misguided. Erasing not only Bakhtin's sense of the radically unsystematic nature of the linguistic world but also the conflicting, anarchic nature of his very texts, semioticians and structuralists have let him speak only by silencing him.

Nowhere does this problem of appropriation emerge more clearly than in examining Bakhtin's critique of language. Indeed, even using the term "language" skews the position that Bakhtin took toward verbal behavior. In Marxism and the Philosophy of Language, his critique of "abstract objectivist" theories is directed to the following points: that such theories stabilize language at the expense of its real mutability and the creativity of its users; that such theories assume language to be outside of contextualization and consequently outside of history; and that such theories tend to hypostatize their own categories. [In an endnote, Stewart directs the reader to Marxism and the Philosophy of Language, p. 77, and Freudianism: A Marxist Critique for more on this.] Ferdinand de Saussure had written in the Cours de linguistique générale:

In separating language from speaking we are at the same time separating: (1) what is social from what is individual; and (2) what is essential from what is accessory and more or less accidental.

Language is not a function of the speaker; it is a product that is passively assimilated by the individual. It never requires premeditation, and reflection enters in only for the purpose of classification….

Speaking, on the contrary, is an individual act. It is willful and intellectual.

No position could be more the antithesis of Bakhtin's. Saussure is interested in language as an abstract and ready-made system; Bakhtin is interested only in the dynamics of living speech. Where Saussure sees passive assimilation, Bakhtin sees a process of struggle and contradiction. And whereas Saussure dichotomizes the individual and the social, Bakhtin assumes that the individual is constituted by the social, that consciousness is a matter of dialogue and juxtaposition with a social Other.

In "Discourse in the Novel" Bakhtin writes: "A passive understanding of linguistic meaning is no understanding at all, it is only the abstract aspect of meaning." For Bakhtin such an abstraction from the concrete utterance would be a dead end, reifying its own categories of the linguistic norm and producing a model with no capability of discussing linguistic—and thereby, for Bakhtin, social—change. If Bakhtin has until recently lacked his true inheritors, Saussure has not, and the major heirlooms of Saussurian linguistics—langue vs. parole, the arbitrary nature of the sign, and, more indirectly, the distinction between poetic and ordinary language—reappear in transformational grammar, in the old (and the new) stylistics, and even, surprisingly, in quasi-Marxist theories of language such as Julia Kristeva's.

The transformational grammarian's devoted outlining of abstract syntactical structures and the stylistician's almost magical rendering of phonetic and morphological structures into thematic structures stand in direct contrast to Bakhtin's object of study. [In an endnote, Stewart writes: "For an attempt to link syntactic to larger social transformations, see Rosalind Coward and John Ellis, Language and Materialism: Developments in Semiology and the Theory of the Subject (London, 1977), pp. 127-30. This attempt seems at best metaphorical and at worst strangely skewed: without a corresponding theory of language use in context, the linkage involves a confusion of levels of analysis."] When Bakhtin discusses "problems of syntax," he has in mind the utterance as it occurs in context, in lived social time. Hence that object of study has rather fluid, generically determined boundaries, ranging from utterances consisting of a single word to utterances consisting of the entire text of a literary work. In The Formal Method in Literary Scholarship, he writes:

It is the whole utterance as speech performance that is directed at the theme, not the separate word, sentence, or period. It is the whole utterance and its forms, which cannot be reduced to any linguistic forms, which control the theme. The theme of the work is the theme of the whole utterance as a definite sociohistorical act. Consequently, it is inseparable from the total situation of the utterance to the same extent that it is inseparable from linguistic elements.

The critique of abstraction in Bakhtin's work is a profound and relentless one. At every point he proclaims that the model of pure linguistic form arose from neoclassical philosophies and from the study of dead languages, that only in its living reality, shaped and articulated by social evaluation, does the word exist. He insists upon contextualizing even the notion of abstraction itself, suggesting that the tradition of normative linguistics from Aristotle and Saint Augustine through the Indo-Europeanists served the needs of sociopolitical and cultural centralization. These "centripetal forces," he contends, can be perceived only against the backdrop of the very "heteroglossia" they sought to deny. We see a rejection in his work not only of a distinction between "language" and "speech" but also of a distinction between synchrony and diachrony. Bakhtin traces these dichotomies to Cartesian rationalism and Leibniz's conception of a universal grammar. Because it denies the actual creativity of language use, Bakhtin rejects the systematizing impulse of such linguistic thought: "Formal, systematic thought about language is incompatible with living, historical understanding of language. From the system's point of view, history always seems merely a series of accidential [sic] transgressions." [In an endnote, Stewart writes: "It is thus puzzling when Krystyna Pomorska, in her foreword to Bakhtin's Rabelais and His World, insists repeatedly that Bakhtin is a structuralist. Although Roman Jakobson and Jurij Tynjanov wrote in their 'Problems in the Study of Literature and Language' that 'every system necessarily exists as an evolution while, on the other hand, evolution is inescapably of a systematic nature' (quoted in Titunik, 'The Formal Method and the Sociological Method [M. M. Baxtin, P. N. Medvedev, V. N. Vološinov] in Russian Theory and Study of Literature,' appendix 2, Marxism and the Philosophy of Language, p. 187), this is hardly a mirror of Bakhtin's rejection of the distinction between synchrony and diachrony. In a later essay, 'Mixail Baxtin [Mikhail Bakhtin] and His Verbal Universe,' Pomorska writes that Bakhtin is a 'real semiotician,' and she goes on to explain that, in his rejection of an autonomous function for literature, he prefigures the Tartu school's semiotics. She also writes that, in addition to the Einsteinian revolution and Husserlian philosophy, 'the other source, more obvious for Baxtin than, say, for Jakobson or Tynjanov, is classical Marxist dialectics' (PTL 3 [April 1978]: 381, 384-85)."]

Not only does such systematization lead to a denial of history—it also results in a vision of speech as a series of "accidental transgressions," a vision we find most prevalent in Noam Chomsky's distinction between competence and performance. There is perhaps no clearer description of linguistic alienation than Chomsky's position on this point: "Any interesting generative grammar will be dealing, for the most part, with mental processes that are far beyond the level of actual or even potential consciousness; furthermore, it is quite apparent that a speaker's reports and viewpoints about his behavior and competence may be in error" [Aspects of the Theory of Syntax, 1965]. The social and political consequences of such an abstract linguistics are brought out in Bakhtin's own observations on the concept of error: "Only in abnormal and special cases do we apply the criterion of correctness to an utterance (for instance, in language instruction). Normally, the criterion of linguistic correctness is submerged by a purely ideological [i.e., thematic] criterion." Rather than assume a transcendent grammar to which actual speech performance can be only imperfectly compared, Bakhtin looks at the social articulation and uses of diversity. The consequences of the Cartesian position become clearer when we look at its current application in state policy. In such domains as the exclusion of bi- (and multi-) lingual education, language requirements attached to immigration restrictions, tensions between nonstandard and standard "dialects" (these terms themselves the necessary fictions by which a transcendent "standard" is created), and the language of state apparatuses in general, the Cartesian position functions to reinforce state institutions and to trivialize change and everyday linguistic creativity. To silence the diversity of the powerful "unsaids" of actual speech in favor of an opaque and universal form of language is to strip language of its ideological significance—a stripping that is itself strongly and univocally ideological. [In an endnote, Stewart writes: "For a sociolinguistic critique of transformational grammar, see Dell Hymes, Foundations in Sociolinguistics: An Ethnographic Approach (Philadelphia, 1974), particularly pp. 119-24, where Herderian and Cartesian linguistics are contrasted; and William Labov, Sociolinguistic Patterns (Philadelphia, 1972), p. 200. In a suggestive essay, Henri Gobard has traced some of the sociological functions of abstract languages, particularly the effect of a multinational English upon vernacular French: see Gobard, L'Alienation linguistique: Analyse tetraglossique (Paris, 1976)."]

His careful attention to actual social behavior also prohibits Bakhtin from accepting any facile distinction between ordinary and poetic language. Such distinctions tend to trivialize both everyday speech acts—by making them automatic or indistinguishable—and "poetic" utterances—by making them parasitic. Most important, theories of poetic language trivialize the activities of speakers by assuming an essentialist, rather than a social, definition of genre. [In an endnote, the critic writes: "See Stanley E. Fish, 'How Ordinary Is Ordinary Language?', New Literary History, Vol. 5, Autumn, 1973, pp. 41-54; and Mary Louise Pratt, 'The Poetic Language Fallacy,' Toward a Speech Act Theory of Literary Discourse, pp. 3-37."] A critique of the concept of poetic language forms a major part of The Formal Method: "If the poetic construction had been placed in a complex, many-sided relationship with science, with rhetoric, with the fields of real practical life, instead of being declared the bare converse of a fabricated practical language, then formalism as we know it would not have existed." In this study, Bakhtin not only objects to the formalist concept of the autonomy of poetic language but also, in a characteristic move, attempts to show the sources and purposes of this formalist position in futurist poetics.

Although we find in Bakhtin an early critic of the linguistics of abstract objectivism, we do not find a neat precursor of contemporary social theories of language. Whereas such studies as William Labov's on the social implications of sound-change vindicate Bakhtin's rejection of a purely "linguistic" conception of phonology, the majority of sociolinguistic studies tends, no less problematically, to emphasize context in a highly abstract way—that is, without a corresponding discussion of the location of the utterance in history and social life. [In an endnote, the critic adds: "See Labov, Sociolinguistic Patterns and Language in the Inner City: Studies in the Black English Vernacular."] In other words, as the abstract objectivists tend to hypostatize grammar, the sociolinguists often tend to hypostatize rules for speech behavior. They do not seem to realize that such rules are not simply located behind the historical processes of social life but are also emergent in them. Hence there is a tendency to want to name the situation, to close off its boundaries, particularly in speech-act theory. Consider, for instance, John Searle's original formulation of his philosophy of speech acts [in Speech Acts: An Essay in the Philosophy of Language]: "The form this hypothesis will take is that the semantic structure of a language may be regarded as a conventional realization of a series of sets of underlying constitutive rules, and that speech acts are acts characteristically performed by uttering expressions in accordance with these sets of constitutive rules."

Similarly, although the sociolinguistic study of styles of speaking goes beyond such a static concept of situation by contrasting the "referential" and "social" aspects of discourse, it does not present a model of the historical transformation of social values and ideologies. The emphasis on rule-governed behavior in current social studies of language again tends toward a focus on "form" at the expense of ideological strategy and a focus on "system" at the expense of social creativity. As a result, such studies search for a grammar of situation; and so, their Romantic humanism notwithstanding, they recapitulate many of the methodological pitfalls of the abstract objectivists.

Bakhtin's positions on verbal interaction thus overlap—but also go beyond—the aims of both sociolinguistics and speech-act theory. The difference is that his primary concern is not so much with how things work as with how things change. In contrast to Searle's atemporality, Bakhtin presents a theory of the sequential order of change: "This is the order that the actual generative process of language follows: social intercourse is generated (stemming from the basis); in it verbal communication and interaction are generated; and in the latter, forms of speech performances are generated; finally, this generative process is reflected in the change of language forms." Like Searle and John Austin, Bakhtin is concerned with identifying what he calls the "little behavioral genres" of speech situations—question, exclamation, command, request, the light and casual causerie of the drawing room. But Bakhtin is even more interested in the relationships between those genres and their contexts in the rest of social life: "The behavioral genre fits everywhere into the channel of social intercourse assigned to it and functions as an ideological reflection of its type, structure, goal, and social composition," particularly as history changes the ideological functions of such contexts. Searle specifies the "happiness conditions" of successful speech acts; Bakhtin is the master of what we might call "unhappiness conditions," those circumstances in which the utterance stands in tension or conflict with the utterances of others. For utterances are always preceded by alien utterances which face them in the form of an addressee or social Other and which surround them with an always significant silence. Whereas linguistic theory must be grateful to sociolinguistics for specifying the profound uses of silence, it must be grateful to Bakhtin for articulating the powerful force of the silenced in language use. [In an endnote, the critic adds: "See, for example, K. H. Basso, 'To Give Up on Words': Silence in Western Apache Culture, in Language and Social Context: Selected Readings, pp. 67-86."]

Thus, Bakhtin presents us with a "generative" linguistics, but that linguistics is accounted for in a social sense. The "rules" it seeks are conventions of genre, conventions of voice, character, idea, temporality, and closure which will be modified by the ongoing transformations of social life. Because it emphasizes the social, it is directly opposed to those contemporary theories of language, such as Chomsky's, that ultimately locate transformation in biological evolutionary processes. And although Bakhtin presents an investigation of utterances in context, his concern with dialogue, with conflict, and, especially, with the cumulative forces of history acting upon each speech situation distinguishes his work from contemporary sociolinguistic theories. Finally, although many careful comparisons have been made between Bakhtin's semiotics and contemporary semiotic theory, Bakhtin's position on the sign differs from traditional semiotics in several crucial ways. [In an endnote, the critic adds: "See Matejka, 'On the First Russian Prolegomena to Semiotics,' appendix 1, MPL, pp. 161-74; Viach Vs. Ivanox, 'The Significance of M. M. Bakhtin's Ideas on Sign, Utterance, and Dialogue for Modern Semiotics (1),' Soviet Studies in Literature 11 Spring-Summer, 1975, pp. 186-243; and Tzvetan Todorov, Mikhail Bakhtine et la théorie de l'histoire littéraire."]

The powerful critique of "language" offered in Marxism and the Philosophy of Language is supplemented throughout the rest of his works by an equally powerful critique of the concept of the sign. The Saussurian theory of the referent naively assumes a univocality of meaning, and Charles Sanders Peirce's theory of iconicity assumes an actual physical referent to which the sign-vehicle corresponds. In contrast, Bakhtin clearly distinguishes between the mechanistic and pragmatic functions of signals and the cultural and "polyvocal" functions of signs: "The process of understanding is on no account to be confused with the process of recognition. These are thoroughly different processes. Only a sign can be understood; what is recognized is a signal." For Bakhtin, the material life of the sign does not arise out of the world of physical objects; rather, it arises out of the actual material practices of everyday social life. And, unlike Saussure, Bakhtin does not see the sign as a part of an abstract system resulting from the structure of psychological perception. Instead, he looks for the ontology of the sign in the "practical business of living speech." Here again we see Bakhtin's rebellion against system. The semiotic character of culture is the result of concrete and dynamic historical processes, processes of tension and conflict inseparable from the basis of social and economic life.

Bakhtin's critique of the univocal sign is perhaps most fully developed in his study of Dostoevsky and in his early essay, "Discourse in Life and Discourse in Art." In the works of Dostoevsky, Bakhtin both found and created the aesthetic correspondent to his theories of thought and language:

The idea, as seen by Dostoevsky the artist, is not a subjective individual-psychological formulation with a "permanent residence" in a person's head; no, the idea is interindividual and intersubjective. The sphere of its existence is not the individual consciousness, but the dialogical intercourse between consciousnesses. The idea is a living event which is played out in the point where two or more consciousnesses meet dialogically. In this respect the idea resembles the word, with which it forms a dialogical unity. Like the word, the idea wants to be heard, understood, and "answered" by other voices from other positions. Like the word, the idea is by nature dialogical, the monolog being merely the conventional form of its expression which arose from the soil of the ideological monologism of modern times.

The idea and the word are here conceptualized as "arenas of conflict," and this conflict arises not simply, as sociolinguistics suggests, out of the tension between the referent and the physical context of utterances but rather from bringing all past experience with the word to bear upon the present situation. For example, Bakhtin sees the works of Dostoevsky as integrating the aphoristic thinking of the Enlightenment and Romanticism into locations of contrast and conflict. And those locations invite the social value judgments of readers who are themselves implicated in the text.

In "Discourse in Life and Discourse in Art," these aspects of polyvalence are worked into a sociological theory of literature. Bakhtin outlines the ways in which the relations between the author's, hero's (character's), and reader's voices intersect within the constraints of genre. "The interrelationship of author and hero, never, after all, actually is an intimate relationship of two; all the while form makes provision for the third participant—the listener—who exerts crucial influence on all the other factors of the work." It is out of these contrasting and collaborating positions that satire, parody, and irony arise as forms depicting conflicting social value judgments.

We might contrast this approach to literature with modern speech-act theories that assume, in Mary Louise Pratt's term, that literary works are "verbal displays." Such theories often neglect the specific effects that the literary work uses to create distancing and irony. [In an endnote, the critic adds: "Pratt presents a convincing argument regarding the ways in which narrative literary works should not be separated formally from narratives of everyday life. But her assumptions of the linearity and univocality of both types of narrative preclude consideration of the ways in which face-to-face narratives most often are constructed collaboratively and, hence, reveal conflicting social value judgments just as literary works do."] In other words, speech-act theories of literature often assume the same systematic and transparent univocality we find in speech-act theories of language. But Bakhtin's literary theory assumes that the problems of dialogue and multivocality that are found in face-to-face communication will be compounded by the specific effects used within the structure of the literary genre. Because the literary work relies on a common ideological purview of both author and reader and, at the same time, cannot rely upon an apparent "extraverbal" context, the work is a complex presentation of display and concealment, of the over- and under-articulated. This presentation is further complicated by the history of generic conventions. Bakhtin writes: "We might say that a poetic work is a powerful condenser of unarticulated social evaluations—each word is saturated with them. It is these social evaluations that organize form as their direct expression."

Bakhtin's concerted opening up of the word may be characterized as having distinguishable, if interwoven, formal and semantic levels. We have seen how his position takes a stance against both abstract objectivism and Romanticism, but we also might consider the influence of Bakhtin's historical work on the development of his theory. In the introduction to Rabelais and His World, Bakhtin writes that the function of the "carnival-grotesque" is "to consecrate inventive freedom, to permit the combination of a variety of different elements and their rapprochement, to liberate from the prevailing point of view of the world, from conventions and established truths, from clichés, from all that is humdrum and universally accepted." It is characteristic of Bakhtin that, in several ways, the carnival-grotesque serves as a model of the normal. For carnival contains both the conventional and the unexpected, the established and the creative. He also focuses on the transitional linguistic forms of the Hellenic period, the late Middle Ages, and the Renaissance—that is, on loci of change. Here and in his discussion of the prehistory of the novel, he is interested in how traditional forms are parodied, or "carnivalized," as part of the complex interaction of social forces within particular historical periods of upheaval and transformation. He consequently paid great attention to the creolized language of the marketplace and street while he formed an image of language as mediating between conventionality and creativity.

For Bakhtin, language is mutable, reversible, antihierarchical, contaminable, and powerfully regenerative. It is always meeting—has always been meeting—what is strange, foreign, other:

Linguistics, itself the product of [the] foreign word, is far from any proper understanding of the role played by the foreign word in the history of language and linguistic consciousness. On the contrary, Indo-European studies have fashioned categories of understanding for the history of language of a kind that preclude proper evaluation of the role of [the] alien word. Meanwhile, that role, to all appearances, is enormous.

At language's point of origin, Bakhtin assumes ambivalence, multivocality, conflict, incorporation, and transformation. Not the least profound implication of this position might be that the model which linguistics has assumed, whereby stable languages eventually become creolized, has been moving backward; instead, we might assume creolization at the point of origin and view stabilization of the linguistic system not as the normal but as the restricted case.

Semantically, Bakhtin's dialogic conception of the word can be seen not only as a contribution to linguistic theory but also as a contribution to the theory of ideology. For although Bakhtin continually erases the abstract concept of language, he just as continually reformulates the concept of ideology. Indeed, it might be more appropriate to place Bakhtin among theorists of ideology rather than among theorists of linguistics and semiotics. To understand the radicalism of Bakhtin's theory of ideology, we must first turn to his own outline of the subject.

Alongside the rejection of transcendence implicit in Bakhtin's critique of abstraction and system is a corresponding rejection of "individual" consciousness. His critique of Freud suffers from the naiveté of his rather knee-jerk reaction to Freud's early published writings; yet that critique substantially predicts Jacques Lacan's reformulation of Freudianism in light of linguistic theory, particularly the translation of the unconscious into a form of language. In place of the concept of the unconscious, which Bakhtin viewed as unanalyzable so long as it remained neither physiological nor verbal, Bakhtin advances the concept of inner speech: through inner speech, all consciousness is social in its formulation. Accordingly, Bakhtin sees inner speech and outer, articulated speech as having ideological status. Inner speech is no less subjected to social evaluation than outer speech, because of the intrinsically social history and nature of the word:

The complex apparatus of verbal reactions functions in all its fundamental aspects also when the subject says nothing about his experiences but only undergoes them "in himself," since, if he is conscious of them, a process of inner ("covert") speech occurs (we do, after all, think and feel and desire with the help of words; without inner speech we would not become conscious of anything in ourselves). This process of inner speech is just as material as is outward speech.

To be sure, Bakhtin recognizes that "the formation of verbal connections (the establishment of connections among visual, motor, and other kinds of reactions over the course of interindividual communication, upon which the formation of verbal reactions depends) proceeds with special difficulty and delay in certain areas of life (for example, the sexual)," but he nevertheless does not explore in any depth the tensions between the unarticulated and articulated in those cases. In his work, little distinction is made between the nature of inner and outer speech, and he sometimes describes inner speech as a mere practice ground for what will or may later be articulated. This theoretical lack might be attributed to Bakhtin's apparent adherence in Freudianism: A Marxist Critique to a rather mechanistic behavioral psychology. He does write, however, that both inner and outer speech form a type of behavioral ideology that "is in certain respects more sensitive, more responsive, more excitable and livelier than an ideology that has undergone formulation and become 'official'."

Thus we begin to receive an outline of constraints that could distinguish between the qualities of inner and outer speech. His sensitivity to those varying constraints was most likely responsible for a major contribution in his discussion of Freud: Bakhtin stresses the shaping power of the specific dialogic situation of the psychoanalytic interview. Going beyond Freud's own individual-centered notions of transference, Bakhtin explains that the interview situation is a highly complex one and must be understood in light of the social dynamic between doctor and patient, and not—or not only—in terms of the patient's individual psyche. [In an endnote, the critic writes: "See Vološinov, Fr, pp. 78-79; Cf. Gregory Bateson, 'The Message "This Is Play,"' in Group Processes: Transactions of the Second Congress of the Josiah Macy, Jr., Foundation, pp. 145-242; and Ray L. Birdwhistell, 'Contribution of Linguistic-Kinesic Studies to the Understanding of Schizophrenia,' in Schizophrenia: An Integrated Approach, pp. 99-123."]

Bakhtin's insistence upon the primary place of the social, the "already said," in the formation of consciousness is at the heart of his struggle against the "bourgeois ideology" of individualism. In Marxism and the Philosophy of Language, we see a critique of Herderian linguistics, with all its Romantic assumptions about the individual soul. In the work of Wilhelm von Humboldt, Karl Vossler, and their followers, Bakhtin points out that a reformulation of this bourgeois philosophy appears in the theory that laws of linguistic creativity are laws of individual psychology. Bakhtin argues that such a position strips language of its ideological content and neglects the intrinsically social nature of linguistic change. In his book Freudianism, he pursues his attack on the abstract concept of the individual by criticizing Freud's asocial and tautological notion of self-consciousness. Here Bakhtin writes:

In becoming aware of myself, I attempt to look at myself, as it were, through the eyes of another person, another representative of my social group, my class. Thus, self-consciousness, in the final analysis, always leads us to class consciousness, the reflection and specification of which it is in all its fundamental and essential respects. Here we have the objective roots of even the most personal and intimate reactions.

According to Bakhtin, all social, antisocial, and warring impulses within consciousness are reflections of social, antisocial, and warring impulses within the mutually experienced world of lived reality. When a class is in decline, we may expect to see manifestations of its decay in the behavior of its individual members.

In Freudianism and the essay on discourse in life and art, the critique of individualism is mirrored within an aesthetic theory. Bakhtin criticizes theories of art that place the significance of the artwork within the psyche of either the creator or the contemplator: "We might say that such a thing is similar to the attempt to analyze the individual psyche of a proletarian in order thereby to disclose the objective production relations that determine his position in society." Instead, Bakhtin places the work in the interaction between these two positions and concludes that artistic value arises only in the dynamics of such social communication. Similarly, in Problems of Dostoevsky's Poetics, Bakhtin situates the subject within sociality and argues that identity is produced by speech, particularly through the contradictions of narrative. Using Dostoevsky as an example, Bakhtin observes that each of that novelist's heroes is the man of an idea, an idea that is itself a construct of contradiction and dialogue:

Only the unfinalizable and inexhaustible "man in man" can become the man of an idea, whose image is combined with the image of a full-valued idea. This is the first condition of the representation of the idea in Dostoevsky.

But this condition contains, as it were, its inverse as well. We can say that in Dostoevsky's works man overcomes his "thingness" (veshchnost') and becomes "man in man" only by entering the pure and unfinalized sphere of the idea, i. e., only by becoming the selfless man of an idea. Such are all of Dostoevsky's leading characters, i. e. all of the participants in the great dialog.

Here we find a radical departure from traditional Marxist aesthetics, in fact, the inverse of Marx's position in The German Ideology where he writes: "We do not set out from what men say, imagine, conceive, nor from men as narrated, thought of, imagined, conceived, in order to arrive at men in the flesh. We set out from real, active men, and on the basis of their real life-process we demonstrate the development of the ideological reflexes and echoes of this life-process." Rather than assume the "real" man at the point of beginning for ideology, Bakhtin would say that it is precisely within narrative, and within ideological structures, that the concept of the individual subject, of the "real" man is born. And the conclusion to The Formal Method makes clear that such ideological structures are themselves constituted by and through speech.

Bakhtin's concept of ideology differs significantly from the early reflectionist theories of Marx. In Capital, Marx writes:

The religious world is but the reflex of the real world. And for a society based upon the production of commodities, in which the producers in general enter into social relations with one another by treating their products as commodities and values, whereby they reduce their individual private labour to the standard of homogeneous human labour—for such a society, Christianity with its cultus of abstract man, more especially in its bourgeois developments, Protestantism, Deism, etc., is the most fitting form of religion.

Marx's theory is deferred and utopian in its outlook: ideology represents the false consciousness produced under class society; after the revolution of the proletariat, the slippage between this false consciousness and the real world will be healed and an actual relation to reality lived. This notion of ideology as false consciousness also lies behind Louis Althusser's distinction between ideology and science and Georg Lukács' attempt in History and Class Consciousness to identify truth with proximity to proletarian consciousness. Similarly, in Lucien Goldmann's distinction between a limited ideology and an embracing world view, we see a kind of "monologic" vision of ideology.

By contrast, Bakhtin asserts that ideology is manifested and created in the practical material activity of speech behavior: hence his notion of "behavioral ideology." As in his theory of linguistics, Bakhtin rejects the abstract concept in favor of the material and dynamic relation. The dialogic nature of the sign, its inner and outward form, allows the intersection of sign with sign, idea with idea, at the same time that it ensures continual upheaval and change in signifying practices as they occur in concrete historical contexts. Thus, ideology is not only the product of social life but is also both productive and reproductive of lived social relations. Furthermore, although in Bakhtin's theory of ideology the socioeconomic base is seen as determining, it is not a base that locks ideology into a static and transcendent form. Rather, ideology is seen as an arena of conflict: one's speech both reveals and produces one's position in class society, in such a way, moreover, as to set into dialogue the relations among classes. Consider this passage from Bakhtin's Marxism:

Existence reflected in sign is not merely reflected but refracted. How is this refraction of existence in the ideological sign determined? By an intersecting of differently oriented social interests within one and the same sign community, i.e.,… with the community, which is the totality of users of the same set of signs for ideological communication. Thus various different classes will use one and the same language. As a result, differently oriented accents intersect in every ideological sign. Sign becomes an arena of class struggle.

Once he moves the materiality of language away from essence into the domain of practice, Bakhtin can present the cacophony of voices present in any utterance, can reject a notion of speech community based on phonology in favor of a much more useful one based on interest and on what might be termed "positionality," the place of the subject within the social structure, a place where subject and structure are mutually articulated. [In an endnote, the critic notes that in The Formal Method, Bakhtin writes: "We think and conceptualize in utterances, complexes complete in themselves. As we know, the utterance cannot be understood as a linguistic whole, and its forms are not syntactic forms. These integral, materially expressed inner acts of man's orientation in reality and the forms of these acts are very important. One might say that human consciousness possesses a series of inner genres for seeing and conceptualizing reality. A given consciousness is richer or poorer in genres, depending on its ideological environment."] What Bakhtin's theory of ideology offers is a model of ideological production. In this model, ideology is not assumed to be either a foggy lens or a mirroring cloud. It is, rather, assumed to be an ongoing product and producer of social practices. The semantic transition from reflection to refraction marks a movement from repetition to production.

Most radically, Bakhtin is unwilling to limit the place of ideology to a particular or narrow sphere of social life. Instead, he concludes that "all of these things [ideological phenomena] in their totality comprise the ideological environment, which forms a solid ring around man. And man's consciousness lives and develops in this environment. Human consciousness does not come into contact with existence directly, but through the medium of the surrounding ideological world." This is a considerable departure from traditional Marxist positions, which either locate the real in the supposedly direct purview of science or in the deferred idealism of revolution. [In an endnote, the critic adds: "In the recent work of Terry Eagleton we find a similar position, one which sees science, no less than ideology, as the product of concrete social practices." See Eagleton, "Ideology, Fiction, Narrative," Social Text, Vol. 1, Winter, 1979.] In contrast, Bakhtin concludes that both science and ideology are products whose absolute "reality"—if it existed at all—could be apprehended only by a transcendent consciousness, a consciousness that would itself be, ironically, an ideological construct. Bakhtin's movement away from a reflectionist theory can perhaps be traced to his familiarity with the carnival mode, where refraction and inversion considerably complicate a traditional functionalist model.

To understand how ideological practice is performed, we cannot begin with a model of the utterance as the spontaneous production of an individual consciousness. Rather, the utterance must be seen as bearing within itself a complex and contradictory set of historical elements. In this sense, Bakhtin observes, all speech is reported speech, for all speech carries with it a history of use and interpretation by which it achieves both identity and difference. It is within this rather remarkable capacity for making present the past that speech acquires its social meaning. Hence for Bakhtin the proper study of ideology would begin with an examination of ideological form, with the study of genre, and not in any autonomous or transcendent sense of genre or form but in the sense that form presents a location of tension between the past and the present. Bakhtin begins by distinguishing ideological objects both from instruments of production (which are consumed by their function) and from consumer goods (which, in existing for individual use, are not available for social evaluation). The specificity of ideological objects lies in their "concrete material reality" and "social meaning."

In The Formal Method, Bakhtin insists that literary form is unique in that it refracts the generating socioeconomic reality in ways particular to its own history and at the same time "reflects and refracts the reflections and refractions of other ideological spheres." Thus, literature serves as a type of super superstructure, in part because of the levels of representation involved in literary production. Here we might contrast the various possibilities of slippage offered by this model to the currently fashionable notion, offered by newer Marxist critics, of the absences in ideological discourse. [In an endnote, the critic adds: see Coward and Ellis, Language and Materialism; Pierre Macherey, A Theory of Literary Production; and Eagleton, Criticism and Ideology: A Study in Marxist Literary Theory.] Bakhtin offers a much more positivist outlook than this deconstructionist one, for he believes that the utterance will carry within it a set of articulate silences and that the common ideological purview of author and reader will work toward the discernment of patterns in the unsaid. Thus, his theory is not necessarily burdened with a nostalgia for full presence: here the contradictions, ambivalences, and silences of the text are seen as part of its essentially dialogic nature.

According to Bakhtin, the reason that literature is the most ideological of all ideological spheres may be discovered in the structure of genre. He criticizes the formalists for ending their theory with a consideration of genre; genre, he observes, should be the first topic of poetics. The importance of genre lies in its two major capacities: conceptualization and "finalization." A genre's conceptualization has both inward and outward focus: the artist does not merely represent reality; he or she must use existing means of representation in tension with the subject at hand. This process is analogous to the dual nature of the utterance, its orientation simultaneously toward its past contexts and its present context. "A particular aspect of reality can only be understood in connection with the particular means of representing it." Genre's production of perception is not simply a matter of physical orientation; it is also a matter of ideology: "Every significant genre is a complex system of means and methods for the conscious control and finalization of reality." According to Bakhtin, nonideological domains are "open work;" not subject to an ultimate closure; but one goal of works of art is precisely to offer closure, a "finalization" that accounts for their ideological power and their capacity to produce consciousness. In the particular finalization of genre, we see a continual tension between tradition and situation. [In an endnote, the critic adds: "For a discussion of the tension between genre and performance, and between tradition and situation, in folkloric performances, see Hymes, 'Folklore's Nature and the Sun's Myth,' Journal of American Folklore Vol. 88 October-December, 1975, pp. 345-69."] As Terry Eagleton suggests in Criticism and Ideology, "A power-loom, for one thing, is not altered by its products … in the way that a literary convention is transformed by what it textually works." Analogously, Bakhtin writes that "the goal of the artistic structure of every historical genre is to merge the distances of space and time with the contemporary by the force of all-penetrating social evaluation." It is perhaps because of this purported goal that Bakhtin himself seemed to prefer the novel, which he viewed as a meta-genre incorporating at once all domains of ideology and all other literary genres. Finally, we must emphasize that Bakhtin's model of genre rests upon his insistence that literary evolution is not the result of device reacting against device, as Viktor Shklovsky believed, but rather of ideological, and ultimately socioeconomic, changes.

We see, then, that Bakhtin's work, in its radical rejection of abstraction, system, and the ideology of bourgeois individualism, forms an arena for a powerful struggle between linguistics and speech, theory and history. His theories' capacities for negation and critique are apparent whether we contrast them to the linguistic theories of his time or of ours. Moreover, this capacity for dialogue, contradiction, and complexity also exists in his work's inner speech—in its allusions to, or silences in the presence of, its own social context. In The Formal Method, at the culmination of Bakhtin's presentation of multivalence, we find as a dominant motif an insistence not only on meaning but also on meaningfulness. Bakhtin cannot accept the futurist model of perpetual and content-less motion: he continually rejects the futurists, and their influence on formalism, as nihilistic, even hedonistic, perhaps reminding us of Trotsky's position in Literature and Revolution. In Rabelais and His World, Bakhtin describes a historical period in which the language of the body was transformed by the rise of capitalism into an alien form of discourse. Bakhtin himself lived in a period when a similar drama was being enacted in the transformations of the peasantry by the industrial state. And yet we never find in his work a discussion of the effects of industrial practice or mechanical reproduction on ideological thought. If we look for a pattern of absences in these texts, we may gradually limn the image of the futurist machine and its totalitarian capacity for the negation of dialogue.

Susan Stewart, "Shouts on the Street: Bakhtin's Anti-Linguistics," in Bakhtin: Essays and Dialogues on His Work, edited by Gary Saul Morson, The University of Chicago Press, 1986, pp. 41-57.