Discourse and Power: A Comprehensive Review

In Discourse and Power, translated into Arabic by Ghaidaa Al-Ali and published by the National Center for Translation, Teun van Dijk presents his socio-cognitive approach to discourse analysis. The book spans around 571 pages divided into ten chapters/articles, in addition to references and a glossary of terms. Van Dijk’s approach stands out from other critical discourse analysts by incorporating a mediating element between society and discourse: cognition. He examines the reciprocal relations within what he terms the “discourse triangle”; discourse, cognition, and society, placing particular emphasis on analyzing micro-structures (such as texts, talk, and cognitive processes) in order to address domination and control operating at the macro-level of society. Due to these features, his theoretical framework is characterized by flexibility and adaptability depending on the problem under study. Van Dijk also draws extensively on insights from other disciplines overlapping with his field, as well as on existing findings and concepts, but, unlike Fairclough, for instance, he does not delve deeply into overarching theoretical debates, focusing instead on practical cognitive and linguistic micro-concepts. His discussions are further enriched with numerous applied examples from various domains of discourse, especially media and political discourse.

What is Critical Discourse Analysis/Studies?

Critical Discourse Analysis can be seen as one of the significant efforts within the field of linguistics to move from the micro-structures of language (words and sentences themselves) to its macro-structures (discourses) and to examine their intersections with society and social theory. It can also be understood as part of the tradition opposing the liberal, structuralist, and “non-critical” approaches that dominated the social sciences in the 1960s and 1970s.

Critical Discourse Analysis (CDA) emerged in the late 1980s and early 1990s through the work of several scholars, including Norman Fairclough, Teun van Dijk, Ruth Wodak, and others. Fairclough’s book Language and Power is often regarded as marking the beginning of critical discourse analysis. Initially, the field was referred to as Critical Discourse Analysis rather than Critical Discourse Studies; however, several thinkers, including van Dijk, have come to prefer the latter term due to the growing diversity of methods, disciplines, and concepts employed in discourse analysis. Some view this diversity positively, as connecting discourse to institutions and non-discursive practices requires considerable effort and collaboration across multiple disciplines to produce results that meet the standards of all fields, such as ethnography, conversation analysis, sociology, and even architectural analysis. Others, however, see it as a negative development, as it has led to methodological and conceptual ambiguity, with fundamental concepts and overarching frameworks being redefined by each analyst. Fowler, for instance, considered this so extensive that “virtually anything could be regarded as critical discourse analysis.”¹

Critical discourse analysis studies real-world, often extended, instances of social interaction that partially take linguistic form. In this context, it is important to distinguish between critical discourse analysis and ordinary non-critical discourse studies. Scholars of critical discourse analysis set themselves a goal that goes beyond that of ordinary discourse studies from the outset: to uncover forms of social domination and inequality that are reproduced, legitimized, and expressed in and through discourse, and to advocate for direct intervention and change in the micro-structures of discourse to eliminate these forms of domination and inequality. In other words, a critical discourse analyst assumes a social stance and mission, rejecting the notion of value-free science, and pursues specific social goals throughout their research. At the same time, they maintain the necessary level of scientific objectivity. From the beginning, critical discourse analysis has aimed to link discourse and language with social theory, particularly schools of thought and thinkers associated with the political left, such as Gramsci, Foucault, Althusser, Habermas, Marxism, Neo-Marxism, the Frankfurt School, structuralism, and post-structuralism. Discourse, above all, is a social phenomenon that cannot be studied linguistically without a coherent social foundation. As Fairclough said:

“One of the key features of economic, social, and cultural changes in late modernity is that they exist both as discourses and as processes occurring outside discourse, and that the processes occurring outside discourse are fundamentally shaped by these discourses.”²

Critical discourse studies are regarded as balanced in terms of the influence of different theoretical schools; they do not favor one school of thought over another. For instance, they reject the absolute relativism of post-structuralism and the prioritization of discourse over all other forms of social interaction. Social interaction, as Fairclough notes, is partially shaped through discourse, but this does not deny the existence of absolutes or truth, nor does it downplay the importance of material reality in shaping social interaction. This is evident, for example, in the traditional Marxist heritage present in most approaches to critical discourse analysis, which we will discuss later.³

While ordinary discourse analysis studies the structure of discourse linguistically and how discourse conveys meaning to things, it does so without adding any political or social dimension, without interdisciplinary integration to examine discourse as a more complex social phenomenon, and, most importantly, without a specific social goal. In other words, critical discourse analysis, unlike ordinary discourse studies, are not concerned with language per se, but with the broader social phenomenon, which requires methodological and disciplinary diversity to study it and to challenge manifestations of social domination. Furthermore, Fairclough distinguishes between critical and non-critical discourse studies, identifying the dividing line as the “social orientation of discourse,”⁴ while simultaneously acknowledging the scientific contributions of non-critical methodologies or schools, which he drew upon in formulating his new methodology of critical discourse analysis (CDA). It is important to note that the critical dimension, which is the fundamental difference between CDA and other forms of discourse analysis, does not imply a negative aspect; rather, it emphasizes the essential feature of these studies: problematizing the research subject by moving taken-for-granted assumptions into the realm of doubt, in order to achieve a clearer understanding of the phenomenon under investigation. This understanding, in turn, remains open to questioning in the future. As the term “critical” suggests, and as Horkheimer expressed, it reflects a desire to move social theory from the domain of interpretation and understanding toward the transformation of society as a whole, a principle that critical discourse studies have largely embraced.⁵

Within the broader context in which this emerging approach developed, there were several attempts to link language with social theory. In his book Discourse and Social Change, Fairclough highlights the most prominent of these attempts, which, in his view, did not create a balance between social theory and language studies. The efforts of critical linguists in Britain during the 1970s, such as Roger Fowler and Michael Halliday, produced linguistically sophisticated analyses but were comparatively weaker in terms of social analysis, whereas Michel Pêcheux’s efforts in France exhibited the opposite pattern.

Theoretical and Methodological Foundations

Van Dijk views critical discourse analysis as a social movement rather than as a set of methodologies, considering them instead as “ways” of analyzing discourse. As noted earlier, approaches within critical discourse analysis may focus on very different sub-issues, domains, and fields of knowledge. For example, Jäger’s approach draws, to some extent, on structural analysis,⁶ while several other approaches draw on methods and fields such as conversation analysis, ethnography, and others. Van Dijk thus sees critical discourse studies as united by a broad framework: the investigation of domination and related social phenomena, including gender and racial discrimination, alongside a research process aimed at producing clear and practical solutions to change discourse practices associated with domination. In other words, what distinguishes a critical discourse analyst from other discourse analysts is their social role, political commitment, and fundamental alignment with those who are subject to domination, which leads to a focus on more specific issues within the broader range of discourses. Van Dijk considers analysis that does not contribute to social change to be poor analysis, highlighting the practical and applied orientation of critical discourse analysis.

Van Dijk’s theoretical framework is characterized by practical and disciplinary flexibility, with a focus on the micro-structures of discourse (texts and conversations), cognition, and society. The framework adapts according to the topic of study, and new concepts or foundations may be incorporated depending on the subject. Van Dijk draws extensively on previous empirical studies in building his theoretical framework, without engaging in detailed discussions of overarching social structures or concepts, at least compared with the attention given to micro-level discursive and cognitive structures and strategies. This may give readers the impression, when examining certain empirical examples, that there is no real connection between micro-structures and macro-structures, which is, in fact, Van Dijk’s primary aim in analyzing micro-structures: to link them to the macro-structures of society. All of this reflects the practical ambition we mentioned earlier in critical discourse approaches, related to changing habitual daily practices that express domination. This is also evident in his moving beyond debates on the relativity of standards and cultural differences, emphasizing that his work is based on “principles of justice and social fairness in our time, or according to international human rights,” without engaging in a prolonged discussion of these abstract concepts.

Power and Ideology

Van Dijk draws on a number of foundational social concepts from previous studies, particularly those derived from the critical school, Neo-Marxism, and the New Left in Britain, to build his theoretical framework. Within this framework, the concept of modern power is based on the idea of indirect symbolic control rather than direct material control: dominant groups exert cognitive control over the desires, aspirations, and emotions, and consequently the actions, of subordinate groups through discourse and the organization of discourse. This indirect view of power emphasizes the relationship between these groups as a fundamental condition for the persistence of power itself, which in turn opens space for resistance by subordinate groups. At the same time, this concept of power and symbolic control does not negate the importance of material power, particularly in the economic and political realms, as it is material power that primarily grants access to discourse, while subordinate groups often remain in the position of discourse recipients or are called upon to speak (for example, in courts, police stations, bureaucratic institutions, and the media).

Van Dijk then moves on to discuss the concept of ideology, drawing on several approaches from scholars in cultural studies, anthropology, philosophy of science, and Neo-Marxism. Ideology, as the “consciousness” or cognitive framework of a particular group, forms the basis of that group’s practices. When such a group is dominant, it normalizes and disseminates this ideology throughout society and its formal and informal institutions, presenting it as an official standard that determines, for example, the selection of discourse topics and the design of curricula in education and media. Other groups internalize this ideology, making it an integral part of what might be called the “order of things” for these groups, serving as a precondition for all their actions and discourses. Van Dijk’s aim is to uncover the dominant ideology in micro-level structures, such as parliamentary speeches and media discourse, to interrupt the internalization of this ideology at the macro-social level.

Van Dijk illustrates how ideology adapts to different fields of discourse (various contexts) by analyzing the discursive strategies employed in speech and texts. These strategies are typically similar across different discourse domains, differing only in the intensity of the actions used. For example, in racist discourse within public conversations and parliamentary debates. Both follow the same discursive strategies, whose racism is revealed through critical analysis. Racist discourse is one of the primary areas of Van Dijk’s analytical focus. He shows that it is based on a central, overarching strategy present in all fields of discourse, namely the ideological square (or polarization strategy), which involves distinguishing between “us” and “them,” representing “us” positively and minimizing our shortcomings, while representing “them” negatively and emphasizing their faults. A number of subsidiary discursive strategies contribute to reinforcing the ideological square, and these are shared across different domains of racist discourse.

Access and Approaches to Discourse

Van Dijk’s access/approach models, explained briefly, refer to the role of participants in discourse and the ways in which they participate or are represented within it. Their importance lies in the fact that those with greater (material) influence exert the most impact on symbolic elites (such as journalists, academics, experts, and others), thereby indirectly controlling the micro-level organization of discourse. An example of discourse organization is courtroom discourse, where there is precise hierarchical control over every detail of speaking and writing within the court and its conditions. This control governs several aspects, including the type of language used, the structure of the conversation, patterns of communication, and, ultimately, the content of the discourse itself.

To illustrate the effect of access to discourse, Van Dijk explains it through the concepts of context and discourse structures, through which cognitive control is exercised. Context refers to the mental structures of the social situation related to the production or comprehension of discourse and includes elements such as the definition of the overall situation, time, place, field of discourse, and participants, as well as mental representations of these elements, such as goals and ideologies. These elements are often defined and normalized by dominant groups. Thus, it becomes clear how dominant groups control both context and discourse structures, that is, they indirectly and cognitively control minds.

This is directly related to the access of different groups to discourse; several studies have shown that the access of minorities to media discourse, for example, is usually negative or directed toward less socially significant matters, such as arts and religion, in contrast to officials and members of dominant groups.

The main issue in Van Dijk’s discussion of practical examples from different discourse domains regarding access/approach models lies in his focus on quantitative, variable particulars, which reflects the social commitment of critical discourse analysis and its desire for immediate and direct change, such as the number of minority employees in the media. He does not provide a detailed mechanism for how this racist ideology persists and is reproduced in discourse despite the increasing number of minority participants across different discourse fields, nor how minorities with access internalize these ideologies and subsequently contribute to their reproduction, albeit in different forms. Although Van Dijk has previously alluded to this, he does not analyze its mechanisms in depth.

In other words, Van Dijk does not provide a method for analyzing the process by which minorities are incorporated within power structures and their ideologies. Additionally, his analysis of media discourse focuses explicitly on right-wing and racist newspapers, without discussing in detail how liberal or “neutral” discourse concerning minorities may also contribute to the reproduction of racist ideology, albeit in forms different from those in right-wing papers. For example, treating immigration as a general problem requiring our assistance and understanding, or emphasizing cultural differences among minorities. Van Dijk refers to these issues but does not offer an extensive discussion or analysis of them.

Primacy of Political Discourse

Van Dijk shows that political discourse is often responsible for defining the broad outlines of racial issues such as immigration and crime, through institutions and official authorities. The media then adopts these sources and the official narrative in its news and reports, which ultimately influences everyday conversation, as people largely rely on major media institutions for their beliefs and opinions. Van Dijk’s empirical analyses of thousands of news reports and several personal interviews reveal a balance between topics in daily conversation and those covered by the media.

From this, we observe the primacy of political (elite) discourse over other types of discourse, and its stronger influence compared to the opposite. Consequently, Van Dijk, and critical discourse analysts in general, attempt to enact change from the top of the hierarchy. Many of their proposals and recommendations regarding discourse therefore target institutions and official discourses, such as educational curricula. This focus on material power as hierarchical and unidirectional largely reflects the traditional, rigid Marxist conception, somewhat removed from the influences of postmodern and structuralist currents. This, I argue, contradicts the resistant ambition of critical discourse analysis: if power is strictly hierarchical, with dominant groups fully effective and subordinate groups fully submissive, how could subordinate groups possibly resist and change the practices of dominant groups?

Racism and the Denial of Racism

Van Dijk discusses strategies for denying racism within dominant racist discourse and the social, individual, and political functions these denials serve. As is typical throughout his book, he examines these strategies across various discourse domains and types, presenting them as part of the overarching polarization strategy, or ideological square, in which denial primarily aims to defend “us.” The significance of denying racism lies in the fact that racism directly contradicts the values of democracy and civilizational progress that the Western world continuously attributes to itself. Acknowledging racism as a real problem would undermine much of the history and cultural achievements they pride themselves on. Moreover, racism is not merely an accusation about an act or lapse; it concerns the identity and most stable characteristics of a person, which makes it more serious and dangerous. In Europe, in particular, racism is considered a grave accusation due to its association in the public mind with Nazi racism and institutionalized ideological fascism.

Therefore, many racist practices are mitigated through labeling them as “prejudice,” “xenophobia,” “discrimination,” or similar terms. The strategy of mitigation or softening is among the most common denial strategies in Europe. Other strategies for denying racism include: denying racist intent, redirecting the accusation toward others (such as neighbors or other members of “our group”), blaming the victim (a form of mitigation, since it does not completely exonerate the perpetrator), reversing the accusation (where the “other” becomes the accused, opposite to the previous strategy), and disavowal and qualification (portraying oneself positively followed by a “but” and then portraying the other negatively), among others.

The objectives of denying racism operate on two levels: an individual micro-level, which involves deflecting the accusation away from the individual, and a broader social macro-level, which encompasses both institutions and individuals. Even patterns of individual denial often rely on portraying the individual and their group as part of the “good citizens.” In general, institutional denial aims to present racism as an incidental or isolated problem, or as an issue limited to a few right-wing extremists, a strategy frequently employed by liberal newspapers in their coverage of racism. In cases of accusation reversal, the left or immigrants are portrayed as the racists, a strategy commonly used by far-right and tabloid newspapers to divert blame away from their own institutions.

Treating racism as an incidental rather than an institutional problem makes it a flexible issue; that is, its prevalence depends on who has the greatest influence over the public, which, according to Van Dijk, is naturally in the hands of dominant groups. This allows societal attitudes to shift easily whenever dominant groups see fit, such as the anti-Muslim conservative mood that swept the United States after the 9/11 attacks. Consequently, combating racism becomes more difficult for subordinate groups. This illustrates the political function of denying racism. Van Dijk examines liberal newspapers more closely in Chapter Five, showing how their discourse contributes to reinforcing racism, alongside, of course, right-wing newspapers.

Cognition and Mental Models

Van Dijk approaches discourse primarily as an expression of three levels of cognition: individual/personal cognition, group/social cognition, and collective/global cognition. A single discourse or text can simultaneously reflect all three forms. To analyze cognition more precisely, Van Dijk introduces the concept of mental models: the representations formed by the mind regarding events and (political) contexts. Their primary function is to link personal information, ideas, and experiences on one hand, and social knowledge on the other. Speakers base their discourse on these mental models, and listeners understand the speaker’s discourse through them as well. Consequently, mental models connect macro-political structures and events with micro-political structures, such as actions and political statements.

Van Dijk builds his cognitive framework on previous studies in cognitive psychology and related concepts, such as long- and short-term memory (where incoming information in short-term memory is processed and stored in long-term memory), episodic and semantic memory, and their different functions. Mental models serve as organizers for classifying and distributing information in long-term memory. Therefore, discourse not only expresses the speaker’s mental models but also influences and activates the audience’s mental models of events and contexts. The audience’s mental model, in turn, determines whether they accept the discourse presented regarding a given situation or construct an alternative model of the event. This means that mental models of context and events are not fixed; they change through social and discursive interaction.

Van Dijk distinguishes the function of event models as responsible for content (what is said, and to some extent how it is said) while context models more strongly govern how this content is formulated (for example, the choice of actions or phrasing). As we have seen, discourse formulations vary from parliamentary speeches to media discourse and everyday conversation depending on context models, even though the strategies and content of discourse remain largely similar. It is important to note, however, that a single discourse does not fully express the speaker’s personal model; it reflects only the small portion related to the current context and event.

From the concept of mental models, it is clear that they rely heavily on personal information, experiences, and orientations, which makes each text or speech unique in some way, rather than merely another expression of a given social tendency. Thus, the triangle of discourse, cognition, and society emerges as the core of Van Dijk’s approach, highlighting the centrality of the cognitive dimension in his framework.

Van Dijk also explains how discourse influences the audience’s mental models through manipulation, which he analyzes using a “triangulation” mechanism (a three-part cognitive strategy used in manipulation). He argues, first, that manipulation involves the misuse of power: it is carried out by dominant groups to convince others of something that benefits only the former. Socially, this occurs through the mechanisms discussed earlier: access, power, and various symbolic and material resources, combined with the subordinate groups’ lack of alternative knowledge regarding the topic under discussion.

Cognitively, manipulation occurs by affecting social representations and information, which as mentioned, constitute the second component of mental models. These are processed in episodic memory and stored in long-term memory, becoming stable and later invoked by event mental models when similar events occur. Dominant groups manipulate through a variety of linguistic and cognitive strategies, some of which resemble legitimate persuasion, such as generalization, convincing others that the interests of subordinate groups align with those of the dominant groups, controlling knowledge and information, polarization, and all the derivative strategies associated with these, among others.

Conclusion: Evaluation and Overview

In this book, Van Dijk makes a significant contribution to critical discourse studies, primarily by incorporating the cognitive dimension as an intermediary between language and society. This aspect had often been overlooked in previous studies of critical discourse analysis. Van Dijk assigns it a central role, based on his view that language and linguistic structures themselves are stable and do not change regardless of whether the discourse is legitimate or not. In other words, even legitimate discourse uses the same linguistic structures as illegitimate discourse. This implies that the fundamental difference lies in the context and context models (cognition), without which language does not acquire any true independent meaning.

However, this focus on the macro- and micro-structures of cognition has led to a genuine neglect at the level of society and social theory. Van Dijk’s theoretical framework, regarding the social level, relies on a set of general previous concepts and findings in the field of social theory, without providing an applied explanation of how discourse is connected to society at the macro level, except for some glimpses regarding patterns of access, a conception based on the traditional Marxist hierarchical division of society, which assumes the existence of dominant and subordinate groups. This, in turn, conflicts with Van Dijk’s theoretical discussion of power as allowing space for resistance and agency. This inconsistency may be partly due to the fact that Van Dijk does not show coherence between theory and practice at the macro-social level. It also stems partially from his reliance on previous findings in social theory, which may combine multiple intellectual traditions, such as neo-Marxism, post-structuralism, and even Robert Dahl’s pluralism in political science, that may not always be compatible, without Van Dijk providing detailed discussion of these findings and concepts, or clearly linking them in practice to discourse and cognition, except for the general assertion that the discursive and cognitive structures under study contribute to the continuation of domination and the misuse of power at the macro-social level.

This critique is echoed by several critical discourse analysts and may stem from a shared view expressed by Wodak, and even indirectly by Van Dijk, that social theory is a tool, like linguistic tools, which a critical discourse analyst can use and modify in their analysis. This leads to a form of inconsistency when social theories are treated merely as tools.⁷ Furthermore, Van Dijk’s conception (at the micro-social level) of patterns of access no longer shows continuity sufficient for reliance today. In a pluralistic society such as the United States, minority members may have genuine access to discourse but may internalize elite ideologies to the point of reproducing those same ideologies themselves, a dynamic that Van Dijk does not examine. Foucault’s concept of power, with its assimilative nature, could potentially complement Van Dijk’s framework in this regard.⁸

The central concept of access today arguably requires reevaluation to account for its new and varied forms. In the current era, with digital media and social networking sites, the role of traditional media in shaping public discourse has become less influential. One could argue that this has enhanced the agency and capacity for resistance of subordinate groups, and revealed new modes of access to public discourse that merit study in order to understand how power is reproduced through discourse in the present context. This is particularly relevant considering the significant investments by certain states and governments in these emerging media to monitor citizens, restrict “inappropriate” content, and disseminate fabricated content to shape public opinion, In other words, to exert greater access to discourse, though the mechanisms today differ entirely from those of the era of television and print media.

1. Roger Fowler. “On Critical Linguistics.” In Texts and Practices: Readings in Critical Discourse Analysis, edited by Carmen Rosa Caldas-Coulthard and Malcolm Coulthard, 3–14. London and New York: Routledge, 1996. ↩︎
2. Chouliaraki, Lilie, and Norman Fairclough. Discourse in Late Modernity: Rethinking Critical Discourse Analysis. Edinburgh: Edinburgh University Press, 1999. ↩︎
3. For a detailed discussion on situating critical discourse analysis within other traditions and approaches in discourse analysis, see: Marianne Jørgensen and Louise Phillips. Discourse Analysis: Theory and Method. Translated by Shawqi Bouanani. Manama: Bahrain Authority for Culture and Antiquities, Knowledge Transfer Project, 2019. ↩︎
4. Fairclough, Norman. Discourse and Social Change. Cambridge: Polity Press, 1992. ↩︎
5. Wodak, Ruth. Methods of Critical Discourse Studies. 3rd ed. London: Sage, 2015. ↩︎
6. For more, review: Jäger, Siegfried, and Florentine Maier. “Theoretical and Methodological Aspects of Foucauldian Critical Discourse Analysis and Dispositive Analysis.” In Methods of Critical Discourse Studies, edited by Ruth Wodak and Michael Meyer, 34–61. London: Sage, 2009.
   Caborn, Joannah. “On the Methodology of Dispositive Analysis.” Critical Approaches to Discourse Analysis Across Disciplines 1, no. 1 (2007): 112–123. ↩︎
7. Breeze, Ruth. “Critical Discourse Analysis and Its Critics.” Pragmatics 21, no. 4 (2011): 493–525. https://doi.org/10.1075/prag.21.4.01bre ↩︎
8. For more: Foucault, Michel. The History of Sexuality: The Will to Knowledge. Translated by Salman Harfouch. Cairo: Dar al-Tanweer, 2017 ↩︎