Critical Discourse Analysis

Critical discourse analysis (CDA) is a qualitative research approach that examines how language is used in texts, speech, and other forms of communication to construct, reproduce, and challenge power relations, social identities, and ideological positions. Drawing on linguistics, sociology, and critical theory, CDA treats language not as a neutral vehicle for conveying information but as a social practice that shapes and is shaped by broader structural inequalities.^[1] Key figures in the development of CDA include Norman Fairclough, Teun van Dijk, and Ruth Wodak, each of whom has articulated a distinct analytical framework while sharing the commitment to analyzing the relationship between language, power, and society.^[3]

CDA reveals the often invisible ways in which language reinforces or challenges social hierarchies. By analyzing media reports, policy documents, political speeches, institutional communications, and everyday conversations, CDA uncovers how dominant groups maintain their positions through discourse and how marginalized groups resist or are silenced. This understanding is critical for fields such as public health (analyzing how health policies frame certain populations), education (examining how curricula represent different groups), law (studying how courtroom language constructs credibility), and media studies (investigating how news coverage shapes public perception of social issues). CDA is explicitly political in its orientation, aiming not just to describe discourse but to critique and ultimately contribute to social change.^[2]

CDA involves multiple levels of analysis. Fairclough's three dimensional framework, one of the most widely used, examines discourse at three interconnected levels: the text itself (vocabulary, grammar, syntax, rhetorical devices), discursive practices (how the text is produced, distributed, and consumed), and social practices (the broader social, political, and institutional context in which the discourse operates).^[1] Analysts examine features such as nominalization (converting actions into nouns, which can obscure agency), passive voice (which, when the agent is left out, can conceal who performs an action), presupposition (embedding contested claims as taken for granted facts), and intertextuality (how texts reference and build on other texts). Van Dijk's sociocognitive approach adds a focus on how mental models and ideologies mediate between discourse and social structure.^[2][3] Wodak's discourse historical approach emphasizes the role of historical context and interdiscursive relationships.^[3]

CDA is appropriate when the research question concerns the relationship between language and power: how policy language constructs target populations, how media coverage frames social issues, how institutional communication reproduces inequality, or how political rhetoric mobilizes support. It is widely used in communication studies, political science, sociology, education, public health, and gender studies. CDA is particularly valuable when the researcher's goal extends beyond description to social critique and when the research is informed by a commitment to equity and justice.

A public health researcher uses Fairclough's three dimensional framework to analyze government policy documents on obesity. At the textual level, she notes that policy language consistently uses words such as "personal responsibility," "lifestyle choices," and "behavioral change," framing obesity as an individual failing rather than a consequence of structural factors such as food deserts, economic inequality, and agricultural subsidies. At the discursive practice level, she traces how these documents were produced by committees with no representation from communities most affected by obesity and how they were disseminated through channels that reached health professionals but not the public. At the social practice level, she connects the individualistic framing to neoliberal governance ideologies that shift responsibility from the state to the citizen. Her analysis suggests that the policy discourse obscures structural determinants of health and limits the range of interventions considered acceptable.