Discourse Analysis

Discourse analysis is the study of language as it is actually used—in stretches longer than a single sentence, by real people, for real purposes, in real settings.

9 frameworks, from 1962 to today · Linguistics · Humanities

still a working view ended · each leads to its entry below

Overview

Discourse analysis is the study of language as it is actually used—in stretches longer than a single sentence, by real people, for real purposes, in real settings1. Where much of twentieth-century linguistics asked how sentences are built from words and rules, discourse analysts ask how people use those sentences to do things: to tell stories, to argue, to take turns in conversation, to build a courtroom case, to signal who they are, to persuade, to exclude1. The field's central conviction is that meaning is not fully contained in the sentence. It depends on what came before, on who is speaking to whom, on what the participants take for granted, and on the social world the exchange lives in. What is at stake is nothing less than how language makes social life possible—and how social life shapes language.

Before "Discourse": Speech Acts and the Limits of the Sentence

The intellectual ground was prepared by philosophers of language. In the 1950s J. L. Austin, lecturing at Oxford, noticed that many utterances do not describe anything and cannot be judged true or false. When someone says "I promise" or "I name this ship," the words do not report an action—they perform it2. Austin called these performatives, and his posthumously published How to Do Things with Words (1962) developed the idea that saying is a kind of doing. His student John Searle systematized this in Speech Acts (1969), proposing that utterances carry an illocutionary force—a request, a warning, a promise, an assertion—that is distinct from their literal content. The classic illustration: "Can you pass the salt?" is grammatically a question about ability but is normally understood as a request. Speech act theory showed that the gap between what a sentence means and what a speaker does with it is not a defect but the normal condition of language. It gave discourse analysis one of its founding questions: how do hearers recognize what speakers are doing?

The 1960s–70s: Language in Its Social Setting

Alongside the philosophers, anthropologists and sociolinguists were studying talk in the field. Dell Hymes, working on language in communities, argued in the 1960s and 1970s that the competence a speaker needs is not just grammatical but communicative—knowing when to speak, to whom, in what style, about what. His ethnography of communication proposed that any speech event could be described by its setting, participants, ends, act sequence, key, instrumentalities, norms, and genre (the mnemonic SPEAKING). The point was that communities differ systematically in these conventions, and that a grammar alone cannot explain why a way of speaking is appropriate here and rude there.

A parallel strand came from the sociologist Erving Goffman, whose Forms of Talk (1981) and earlier work on the "presentation of self" treated conversation as a delicate social performance in which participants manage face—their public self-image—and repair the small damages that interaction constantly inflicts. John Gumperz's interactional sociolinguistics, developed from the 1970s, added the observation that much of what speakers convey is carried by contextualization cues: prosody, code-switching, formulaic phrases that signal "this is a joke," "I'm quoting," "I'm changing footing." Gumperz's studies of cross-cultural miscommunication—for instance, differences in how speakers from different backgrounds signal emphasis or disagreement—showed that misunderstanding can arise not from bad grammar but from mismatched cueing conventions.

Conversation Analysis: Order in the Ordinary

The most radical departure came from sociology. Harvey Sacks, working with Emanuel Schegloff and Gail Jefferson at UCLA and later Berkeley in the 1960s and 1970s, began from the premise that ordinary conversation is not chaotic but highly organized, and that its organization is visible in the fine grain of talk. Using recordings and detailed transcripts, Sacks discovered the turn-taking system: speakers coordinate who talks next through a set of orderly procedures, and silences, overlaps, and interruptions are meaningful against that baseline. He described adjacency pairs—greeting/greeting, question/answer, invitation/acceptance-or-refusal—as paired actions where the first sets up an expectation the second fulfills or disappoints. He showed that repair (fixing trouble in talk) has its own preference structure, with speakers preferring self-correction to correction by others.

The famous demonstration is the "doctor's office" or "call to emergency services" type of analysis, where the opening seconds of a call reveal how participants establish identities and obligations. Conversation analysis (CA) insisted on emic analysis—describing what participants themselves orient to—and on using naturally occurring recordings rather than invented examples. Its influence spread from sociology into linguistics, where it became the standard for studying spoken interaction. Jefferson's transcription system, with its symbols for pauses, stress, and overlap, became a shared instrument.

The 1970s–80s: Text Linguistics and the European Traditions

In continental Europe, a different tradition grew from text grammar. Scholars in the 1960s and 1970s asked whether the sentence grammar of Chomskyan linguistics could be extended to whole texts. The German Textlinguistik of the 1970s—associated with scholars such as Wolfgang Dressler and Robert de Beaugrande, whose Introduction to Text Linguistics (1981) was widely read—proposed that texts have their own constitutive principles: cohesion (how sentences are tied together by pronouns, conjunctions, repetition), coherence (how a text makes sense as a whole), intentionality, acceptability, and situatedness. This was a more formal, text-centered counterpart to the interactional traditions.

In France, analyse du discours developed from the 1960s onward with a distinctly political and ideological bent. Michel Pêcheux and his associates combined Marxist theory with linguistics to study how ideology is carried in language, and the journal Langages (founded 1966) became a venue for work on political discourse, institutional talk, and the relationship between text and social formation. French discourse analysis treated discourse as a site where power and ideology are enacted, not merely a neutral vehicle of communication.

Critical Discourse Analysis: Language and Power

The 1980s and 1990s saw the rise of critical discourse analysis (CDA), which made the relationship between language and power its explicit object. Norman Fairclough's Language and Power (1989) and Discourse and Social Change (1992) argued that discourse both reflects and constitutes social structures—that changing how people talk about something (say, "welfare" or "terrorism") can change what is politically possible. Teun van Dijk developed a socio-cognitive approach, arguing that discourse works through shared mental models and ideologies, and applied it to the study of racism in the press. Ruth Wodak's discourse-historical approach analyzed how political and media discourse draws on historical narratives and collective memories, notably in her studies of Austrian and European political rhetoric.

CDA's method is typically to take a corpus of texts—news reports, political speeches, policy documents—and analyze lexical choice, transitivity (who does what to whom), nominalization (turning actions into things, as when "the government decided" becomes "the decision"), and presupposition. The aim is to make visible what a text takes for granted and whose interests that serves. Critics have long charged CDA with selecting texts to fit a pre-formed political conclusion and with insufficient attention to how ordinary readers actually interpret texts; defenders reply that its explicit critical stance is a strength, not a bias, and that it has been refined through corpus methods and reception studies.

Corpus-Assisted Discourse Studies

From the 1990s, the availability of large electronic corpora transformed the field's evidence base. Corpus-assisted discourse studies (sometimes called CADS) uses computational tools—frequency lists, concordances, collocation analysis—to identify patterns across thousands of texts that a close reading of a handful could miss. A researcher can ask, for instance, which adjectives most frequently modify "immigrant" in a national newspaper corpus, or how the phrase "tax burden" clusters with other terms. The method does not replace close reading; it directs attention to patterns worth reading closely. Work by scholars such as Paul Baker and Tony McEnery demonstrated that corpus methods could test and sometimes complicate the claims of critical discourse analysis, showing, for example, that a newspaper's apparent bias might be concentrated in specific genres or periods rather than uniform.

Multimodality: Beyond Words

A further expansion came with the recognition that discourse is rarely verbal alone. Multimodality, developed from the 1990s by scholars including Gunther Kress and Theo van Leeuwen (Reading Images, 1996), studies how meaning is made through the combined resources of image, layout, color, gesture, gaze, and sound. A magazine advertisement, a museum exhibit, a classroom, or a social media post communicates through the arrangement of visual elements as much as through words. Multimodal analysis borrows from systemic functional linguistics the idea that visual composition, like grammar, offers choices—what is placed at the center versus the margin, what is shown close up versus at a distance, what is depicted as given versus new—and that these choices carry meaning. The approach has become central to the study of digital communication, where text, image, and interface design are inseparable.

The Present: Plural Methods, Shared Questions

Discourse analysis today is not a single method but a family of approaches sharing a conviction that language in use is the proper object of study and that context is not optional. Conversation analysis remains the gold standard for interactional data, with its insistence on recorded, naturally occurring talk and participant-relevant analysis. Critical discourse analysis continues to interrogate power and ideology, increasingly in digital and multimodal settings. Corpus methods provide scale and replicability. Ethnographic and interactional approaches keep attention on the local, situated production of meaning3. French and German traditions remain vital, and discourse analysis has spread into fields well beyond linguistics—education, health communication, political science, media studies, and law.

Open questions persist. How do analysts reconcile the close, qualitative reading that discourse analysis demands with the scale of digital text? How should multimodal meaning be transcribed and analyzed with the rigor that Jefferson's system brought to speech? And how far can a method developed largely on English and European languages claim universal applicability? These are live debates, and they keep the field moving.

References

  1. Discourse analysis, Wikipedia
  2. Speech act, Wikipedia
  3. Interactional sociolinguistics, Wikipedia

The frameworks of Discourse Analysis

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1962-1995

    Ethnography of Communication

    What does a speaker need to know to speak appropriately?

    Fly to its tower in Nousfarer

  2. 1962-1990

    Speech act

    What are people doing when they speak?

    Fly to its tower in Nousfarer

  3. 1965-Present

    Conversation Analysis

    How do people organize turn-taking in ordinary talk?

    Fly to its tower in Nousfarer

  4. 1966-1995

    French Discourse Analysis

    How does language carry ideology and power?

    Fly to its tower in Nousfarer

  5. 1970-Present

    Interactional Sociolinguistics

    How do subtle cues cause cross-cultural misunderstanding?

    Fly to its tower in Nousfarer

  6. 1970-1995

    Text Linguistics

    What makes a stretch of sentences a coherent text?

    Fly to its tower in Nousfarer

  7. 1989-Present

    Critical Discourse Analysis

    How does discourse reproduce power and inequality?

    Fly to its tower in Nousfarer

  8. 1990-Present

    Corpus-Assisted Discourse Studies

    What patterns emerge across thousands of real texts?

    Fly to its tower in Nousfarer

  9. 1996-Present

    Multimodality

    How do images and layout make meaning with words?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Ethnography of CommunicationSpeech actConversation AnalysisFrench Discourse AnalysisInteractional SociolinguisticsText LinguisticsCritical Discourse AnalysisCorpus-Assisted Discourse StudiesMultimodality

Beyond the field

Where Discourse Analysis's ideas came from in other fields, and what they led to there: the strongest 12 of 32 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Discourse Analysis is a planet in Nousfarer

Its city has a tower for each of these 9 frameworks. Land, walk in, and climb one concept at a time.

Land on Discourse Analysis

Related fields

Fields its ideas are linked to

More in Linguistics

See where Discourse Analysis sits on the Map of Human Knowledge