For many years, discourse analysis was less an explicit “theory” than a practical and empirical approach for supporting field work on relatively little-recorded languages and cultures (see, e.g., Grimes, Longacre, Malinowski, Pike). One domain of early work that attracted notice in general and humanistic circles was the cross-cultural study of stories and narratives (e.g., Claude Lévi-Strauss, Structural Anthropology). Major concerns later on included the discourse of schooling and education (Sinclair and Coulthard, Stubbs, Widdowson) and, with a sociological turn, the organization of conversation (Sacks, Schegloff, and Jefferson).
These practical and empirical emphases fostered some variance with the “theoretical linguistics” postulating a dichotomy between language and discourse (e.g., langue versus parole for Ferdinand de Saussure, “competence” versus “performance” for Noam Chomsky). The project of abstracting “language” away from the cultural and social contexts in which it appears as a human phenomenon seemed attractive on theoretical grounds, especially for an emergent science like linguistics, but the consensus today is that this project is unrealistic. The rising pressure upon theory and method to resituate language in these contexts accounts for the explosive interest in discourse analysis, a field that from its very beginnings has implicitly or explicitly maintained the unity of language as both structure and event, both knowledge and action, both system and process, both potential and actual (Firth, Halliday, Hartmann, Pike).
In the 1970s, discourse analysis became a convergence point for a number of trends: “text linguistics” on the European continent; “functional” or “systemic linguistics” in Czechoslovakia, Britain, and Australia; “cognitive linguistics,” “critical linguistics,” “ethnography of communication,” ethnomethodology, and the structuralism, poststructuralism, deconstruction, and feminism emanating from France; along with semiotics and cognitive science, both convergence points in their own right. This drift has made it possible, indeed essential, to contemplate discourse from multiple viewpoints: linguistic, philosophical, cognitive, social, anthropological, literary, historical, political, and ideological. Admittedly, essaying to do so makes us keenly aware of how multifarious and complex discourse transactions can be. Our best guarantee that we can ultimately make sense of all this is that they generally succeed in social practice. The task of discourse analysis is to describe the systematic organization and intersubjectivity that enable the success.
Accordingly, theories and models are being developed on numerous fronts: for the syntactical contours and the large-scale (“global”) coherence of discourse; for the interactive performance of discourse actions, or “speech acts”; for the plans, goals, and strategies of discourse participants; for the interface of meaning or significance with culture, ideology, personality, gender, and emotion; for the roles and relations of power or solidarity among participants or institutions in discourse.
The notion of “discourse” itself has been commensurately expanded. Besides being the standard designation for a recorded sequence of utterances (Longacre, Pike) or of “texts” (Beaugrande and Dressier), “discourse” may designate elaborate complexes all the way up to a definite order of concepts (Hindess and Hirst) or the entire practice and communication within a social institution (e.g., Michel Foucault, The Archeology of Knowledge; Language, Counter-Memory, Practice). Such is the diversity that one can find two “introductions” to discourse analysis with no overlap at all (see Coulthard; Macdonnell).
Still, discourse analysis does manifest some general and consistent principles, which might be formulated as follows:
1. A “discourse” is not merely a linguistic unit, but a unit of human action, interaction, communication, and cognition. The habit of identifying the “discourse” with its recorded (usually written) language trace, though deeply entrenched, must be transcended.
2. The source of data should be naturally occurring discourses rather than isolated brief examples invented by investigators. Having established the importance of context, we must discard the convenient fiction of “context-free” words or sentences. Such items are merely transposed by our citation into a different context, and we should inquire how we may be changing their significance, for example, concealing constraints or mystifying institutional commitments.
3. Discourse analysis should balance analytic with synthetic viewpoints. The traditional methods of discovering “linguistic units” and “constituents” by segmenting discourse should be more evenly correlated with methods that focus on how discourses are assembled and how the various units or aspects contribute to a constellation of mutual relevance (Beaugrande).
4. A discourse is not a static, idealized, or totalized unity of words and significances, but a dynamic field of interests, engagements, tensions, conflicts, and contradictions. This field in turn reflects the organization of society and its institutions and the roles and power structures inherent therein (Fowler et al., Wodak et al.).
5. A discourse or discourse domain should not be isolated from others but be seen in its mutual relevance to them. To appreciate the nature and problems of a domain such as “technical language,” we should not reduce it to its incidental features, such as lists of special terms or tables of formulas. We must inquire how it functions within the general acquisition of knowledge through discourse and how it could function more effectively for wider participation.
6. Discourse analysis should continually reflect upon its own procedures. Given the unmanageably large and diverse range of data, each project must be selective and focused and so should declare and justify its motives in terms of epistemological interests. The discourse of science itself should be examined (Gilbert and Mulkay), as should that of specific fields such as anthropology (Geertz).
7. Discourse analysis obliges the investigator to engage and reengage with discourse. The idealized separation of subject from object, or investigator from data, is not feasible here. Since one’s involvement in the data and one’s commitments and priorities cannot be eliminated, they can be profitably made a further object of reflection: on how the discourse being analyzed correlates with the discourse of the analysis.
8. Discourse analysis is rich and expansive rather than formalized and reductive. Discourse cannot be adequately analyzed with a fixed algorithm for reifying it into a configuration of formal symbols. Instead, the analysis should pursue the relevance of a discourse in any direction and to any degree needed in order to grasp its status within social practices, such as in news reporting (Dijk) or psychotherapy (Labov and Fanshel, Wodak).
9. To master its issues and problems, discourse analysis must adopt an encompassing interdisciplinary perspective. In the past, interdisciplinarity has too often been restricted to programmatic statements of intent; we are now filling in the content of such programs with a creditable body of results. Hence, discourse analysis should be, not one more Kuhnian battleground for warring “paradigms,” but a domain for cooperation and integration among alternative paradigms (see Thomas S. Kuhn ).
10. Discourse analysis should interact with institutions and groups both inside and outside the academy to pursue urgent issues and problems. We cannot assume that our current methods address all the most pressing issues. Instead, we should periodically take stock of and adapt our methods to more issues, such as the discourse of politicians about the nuclear arms race (Chilton et al.) or the discourse of judges and defendants in courtrooms (Atkinson and Drew, Leodolter).
11. The highest goals of discourse analysis are to support the freedom of access to knowledge through discourse and to help in revealing and rebalancing communicative power structures. Following the lead of “critical linguistics” (Dijk, Fowler et al., Mey), this thesis has now been widely acknowledged. Special attention has been devoted to geopolitical problems such as public policy, colonialism, racism, and sexism, which, though restricted by laws and statutes, persist at deeper levels in discourse, not merely through lexical choices, but through background assumptions, hierarchical structuring, rights of turntaking, and so on.
12. The demanding tasks facing us today call for an explicit, coherent research plan. Past trends have been unduly dependent on personal or institutional commitments and decisions. Now that a global dispersion of discourse study is under way, larger projects seem feasible, provided that scholars can interact over long distances and shorter intervals.
The future of discourse analysis will depend to no small degree on whether principles such as these can be fully implemented and suitable frameworks and resources provided for research. The prospects seem especially favorable for interaction between discourse analysis and literary studies, a field in which the notion of discourse is being generally recognized as a foundational problem. The principles just enumerated readily invoke some ongoing trends as well as some future desiderata:
1. The traditional philological, formalistic, or New Critical focus on the literary text as language has been complemented by a concern for literary action, interaction, communication, and cognition, though so far (inspired by French scholars such as Foucault) more from a philosophical than a sociological or psychological orientation.
2. The literary texts taken as objects of study are almost never invented by the investigator. Yet their “natural occurrence” requires specific conditions and conventions that need to be more clearly formulated and understood (Schmidt). Further groundwork is now being supplied by literary journals with an empirical outlook, such as Poetics and Empirical Studies in the Arts.
3. Recent trends show a more even balance between the analytic tactics of “close reading” or “text exegesis” and synthetic models of literary “production” and “reception” (Jauss) (see Reception theory).
4. The traditional harmonizing, or “totalizing,” tendencies of literary criticism have been offset by widening probes of literary discourse as a field of interests, engagements, and conflicts, including the estrangement from the putative “real world” of the reader (Iser) (see Reader-response theory and criticism).
5. Scholars reveal a renewed willingness to resituate literature, long isolated as a privileged preserve set above other discourse or even in opposition to it, among the plurality of social and ideological discourses of its own time and ours (Fowler; Hayden White, Metahistory: The Historical Imagination in Nineteenth-Century Europe, Tropics of Discourse; Fredric Jameson, The Political Unconscious).
6. The enterprise of reflecting upon procedures is at the very heart of the prestigious “literary theory” movement (see Beaugrande, Critical Discourse), though the theorizing is sometimes obscure about its goals.
7. The fastidious reaching for ultimate, tidy closure of the “meaning” of the literary work has been yielding to an open-ended readiness to engage and reengage the work, notably in J. Hillis Miller’s appropriation of “deconstruction” (“Deconstructing the Deconstructors” in Theory Then and Now). The individual work itself is viewed as an “intertextual weaving” of other discourses (Geoffrey H. Hartman, Saving the Text).
8. The expectation that analysis should be rich and expansive was only rarely suppressed in literary studies by the kind of strict “scientism” we have seen in some schools of linguistics. The brief “structuralist” turn to narrow linguistic method has long since swerved toward the wide-ranging “poststructuralist” revision (both trends documented by Harari).
9. The value of an encompassing interdisciplinary perspective on literature is no longer seriously contested today, and joint projects are commonplace, for example, between the Psychological Institute of the Hungarian Academy of Science and the American Council of Learned Societies (results edited by Martindale).
10. Shifts of focus outside the academy are still regrettably rare, but promising signs can be seen in some recent detailed investigations of the reading public and the literary publishing industry, as presented at the first symposium of the International Society for the Empirical Study of Literature in 1987 (Schmidt, ed.).
11. The freedom of access to the unique experiences literature affords is still not a firmly established goal, due to the elitist disdain for naive readers. But discourse analysis has shown the processing of quite ordinary discourse to be enormously sophisticated and the supposed naïveté of nonelite readers to be an illusion.
12. Literary theory has been replete with calls for an explicit, coherent research plan. So far, progress has been slowed by the idiosyncratic and self-indulgent communicative strategies of some conspicuous theorists, who seem less concerned with any such plan than with the enhancement of their personal prestige. Here, the paradigm of discourse analysis, which addresses issues of such complexity that unplanned research would remain ineffectual, could act as a model.
The problems facing both discourse analysis and literary studies in the coming years are obviously enormous, but a concerted interaction between the two would surely improve the prospects for significant advances on both sides.
John Atkinson and Paul Drew, Order in Court: The Organization of Verbal Interaction in Judicial Settings (1979); Robert de Beaugrande, Critical Discourse: A Survey of Contemporary Literary Theorists (1988), Text, Discourse, and Process (1980), Text Production (1984); Robert de Beaugrande and Wolfgang Dressier, Introduction to Text Linguistics (1981), “A New Introduction to the Study of Text and Discourse” (forthcoming); Paul Chilton, ed., Language and the Nuclear Arms Debate: Nukespeak Today (1985); Malcolm Coulthard, An Introduction to Discourse Analysis (1985); Teun van Dijk, News as Discourse (1988); Teun van Dijk, ed., Discourse Analysis: Psychological Aspects (1986), Handbook of Discourse Analysis (1985); John Rupert Firth, Papers in Linguistics, 1934-1951 (1957); Roger Fowler, Literature as Social Discourse: The Practice of Linguistic Criticism (1981); Roger Fowler, Robert Hodge, Gunther Kress, and Tony Trew, Language and Control (1979); Clifford Geertz, Works and Lives: The Anthropologist as Author (1988); Nigel Gilbert and Michael Mulkay, Opening Pandora’s Box: A Sociological Analysis of Scientists’ Discourse (1984); Joseph Grimes, The Thread of Discourse (1975); Michael Halliday, Introduction to Functional Grammar (1985); Josué V. Harari, ed., Structuralists and Structuralism: A Selected Bibliography of French Contemporary Thought (1971), Textual Strategies: Perspectives in Post-Structuralist Criticism (1979); Peter Hartmann, Theorie der Sprachwissenschaft (1963); Barry Hindess and Paul Hirst, Modes of Production and Social Formation (1977); Wolfgang Iser, Der Akt des Lesens: Theorie ästhetischer Wirkung (1976, The Act of Reading: A Theory of Aesthetic Response, trans. Iser, 1978), Der implizite Leser: Kommunikationsformen des Romans von Bunyan bis Beckett (1972, The Implied Reader: Patterns of Communication in Prose Fiction from Bunyan to Beckett, trans. Iser, 1974); Hans Robert Jauss, Ästhetische Erfahrung und literarische Hermeneutik (1982, Aesthetic Experience and Literary Hermeneutics, trans. Michael Shaw, 1982), Toward an Aesthetic of Reception (1982); William Labov and David Fanshel, Therapeutic Discourse (1977); Ruth Leodolter, Das Sprachverhalten von Angeklagten bei Gericht (1975); Robert Longacre, An Anatomy of Speech Notions (1976), Grammar of Discourse (1983); Diane Macdonnell, Theories of Discourse: An Introduction (1986); Bronislaw Malinowski, “The Problem of Meaning in Primitive Languages,” The Meaning of Meaning (by C. K. Ogden and I. A. Richards, 1923); Colin Martindale, ed., Psychological Approaches to the Study of Literary Narratives (1988); Jacob L. Mey, Whose Language? A Study in Linguistic Pragmatics (1985); Kenneth Lee Pike, Language in Relation to a Unified Theory of the Structure of Human Behavior (1967); Harvey Sacks, Emmanuel Schegloff, and Gail Jefferson, “A Simplest Systematics for the Organization of Turntaking for Conversation,” Language 50 (1974); Siegfried J. Schmidt, Foundations for the Empirical Study of Literature: Components of a Basic Theory (1982); Siegfried J. Schmidt, ed., Aspects of the Empirical Study of Art and Media, special issue, Poetics 18 (1989); John McHardy Sinclair and Malcolm Coulthard, Toward an Analysis of Discourse (1975); Michael Stubbs, Discourse Analysis (1983); Henry Widdowson, Explorations in Applied Linguistics (1979); Ruth Wodak, Language Behavior in Therapy Groups (1986); Ruth Wodak et al., Language, Power, and Ideology (1989).
Source: Groden, Michael, and Martin Kreiswirth. The Johns Hopkins Guide to Literary Theory and Criticism. Baltimore: Johns Hopkins University Press, 1994.