Discourse, Power and Symbolic Elites

Discourse, Power and Symbolic Elites


Critical discourse studies are especially interested in the abuse of power, that is, in the illegal ways discourse is controlled and how the public mind is discursively managed by the symbolic elites. The beliefs, knowledge and attitudes of citizens are controlled against their best interests, and in the best interest of those in power. This does not necessarily involve horrible threats or manipulation. Rather, the most influential forms of discursive mind control are subtle and indirect, and therefore hard to identify, to resist or to criticize. This is why we need sophisticated discourse analysis in order to critically examine these forms of language use and communication.

Power is usually associated with people who control more scarce social, political or economic resources than others. Powerful people have more money and property, are more famous or make more decisions about more people. Especially influential in modern societies is the symbolic power of those who control information, communication and knowledge.

One of these symbolic resources is the access to, and control over public discourses, such as the discourses of politics, the mass media, education, science, literature or the bureaucracy. It is therefore a major task of critical discourse studies to closely examine how the symbolic elites exercise, and sometimes abuse of, their discursive and communicative power.

Fundamental is the role of discourse in the formation and transformation of knowledge, beliefs, emotions, opinions, attitudes and ideologies. In other words, those who control public discourse indirectly control the public mind. It is this triangulation between discourse, cognition and society that is one of the main tasks of critical discourse studies.

The exercise of power in society may be more or less authentic. In critical discourse studies we are especially interested in the abuse of power. In this way, the beliefs, the knowledge and the attitudes of citizens are controlled against their best interests, and in the best interest of those in power. This is why we need sophisticated discourse analysis in order to critically examine these forms of language use and communication.

Control of Context

If the social power of the symbolic elites is defined first of all by their privileged access to public discourse, we first need to examine these forms of access.

The context of discourse is the way the participants define the relevant social dimensions of the communicative situation, such as the setting (time, place), participants (their social identities, roles and relationships), the ongoing social acts accomplished by the discourse, as well as the intentions and the knowledge of the participants. Such contexts influence discourse in such a way that it is appropriate in the communicative situation. For instance, people who write news reports do so as the journalists, in a specific location, at a specific time and before a specific deadline, for a specific public with a specific knowledge, and they do so with the goal to inform the readers about recent events, and from the perspective of a specific professional and social ideology.

The symbolic elites control the context of discourse, and hence indirectly discourse itself, if they decide who may speak or write, as what, to whom, when and where, and with what intentions and goals. It is a professor and not a student who decides about the setting of a class or an exam. It is the judge who decides who may or must take the floor in court, and whether as counsel, defendant or witness. It is the editor who decides who will be interviewed and cited in the press, and these will be mostly the other symbolic elites, that is, mostly white, middle-class white men in superior positions, and not women, elderly people, children, black or indigenous people, working-class people, or immigrants. In sum, who controls the context largely controls the text. Those elites are most powerful who control most contexts of most influential public discourse.

Controlling the text

Topics: The topics of discourse feature the most important information of the text. They are what the text is globally about. Such topics typically are expressed in the headlines and the leads, and are best memorized by the media users. Who controls the topics of public discourse controls what people think about and talk about, often even independently of the experiences on the interests of the citizens, as we now from the prominence in political and media discourse of topics related to Iraq, terrorism or immigration. Thus, there is a correlation between the prominence of topics in the mass media and what citizens in polls tell they are worried about.

Local meanings: Global meanings control local meanings of discourse. So, in a news report on a terrorist attack, we may expect local meanings about armed aggression, bombs, victims, terrorists (typically Arabs or Muslims, because armed aggression of our people are not called terrorism). Besides these topic-controlled local meanings, there are many other aspects of the meaning of sentences that are crucial for our understanding and that may be controlled by the symbolic elites. Among many other aspects of discourse semantics, metaphors are particularly powerful to express and shape the way we think, as is typically the case in the description of the arrival of immigrants in terms of waves, in other words, as menacing large amounts of water in which we may drown.

Syntax: Even the formal structure of sentences may play a role in the way the symbolic elites represent events or actions in public discourse. Thus, we typically may encounter a nominalization such as discrimination without reading who actually does the discriminating. In the same way, a headline with a passive sentence such as Demonstrators killed may not tell us that it was the police who did so, thus playing down the role of ‘our’ police. Again we see that the structure of discourse may emphasize the bad actions of others, and deemphasize the bad actions of our own people.

Global formats: Not only the local forms of syntax, but also the global formats of discourse genres may be relevant in the way meaning or information is conveyed to the citizens. One way is to tell about events, and another way is to argue in favour or against an opinion. Indeed, such arguments may feature fallacies that may manipulate the way citizens make up their mind about a public issue.

Rhetoric:  It is typically with these rhetorical means that we may exaggerate or mitigate how we speak about people, as we know from the use of hyperboles and euphemisms. And again, the negative things of ‘Others’ are typically exaggerated by hyperboles, whereas ‘Our’ negative things tend to be formulated as euphemisms.

We see that there are many ways the symbolic elites may shape discourse in a way that reflects their view of public events and their interests. At all levels of public discourse we find more or less subtle ways the ideologies, goals and interests of the elites are implicated according to the general strategy of emphasizing our good things and their bad things, and deemphasizing our bad things and their good things.

Cognition: The control of public discourse by the elites would hardly be a problem if such a discourse wouldn’t have any influence on the citizens. And it is true that people are not automatically and directly influenced by what they read, hear or see in the mass media. They already have their own ideologies, attitudes, and opinions about many public issues, and these will decisively influence the way they read, understand and accept or reject the facts or opinions they are confronted with in public discourse. In other words, the autonomy of more or less independent, well-informed, intelligent citizens is crucial in the critical way they may evaluate and if necessary resist persuasive public discourse.

Context models: Context models play a crucial role in communication. The context models of discourse producers control how discourse is adapted to the communicative situation. The same is true for the context models of the readers or viewers of public discourse. They tend to adapt what they read, understand and memorize on the one hand to what they think are the intentions of the speakers or writers, but on the other hand they adapt what they read and understand to their own current goals are, knowledge, opinion, attitudes, and ideologies of interests. It is also for this reason that discourse does not automatically cause specific interpretations.

In other words, even before starting to read or hear the first words of discourse, recipients already have construed a partial context model that will influence and guide the way they are going to read, listen to, and understand the discourse in which they now participate. During the interpretation of discourse, they will then further elaborate the context model.

Mental models are assumed to be stored in so-called episodic memory, which is the part of long-term memory in which people represent their autobiographical, personal experiences. If we recall something of our past, we search and activate a mental model of our episodic memory. These mental models not only feature what happened or what we did, but also our opinions and emotions about what happened. So, they are not objective representations of what happened but essentially personal and subjective, although they have a social basis because also the interpretation of situations, events or actions is of course related to general knowledge we share with other people of our culture or community.

Event models

Context models are not the only models involved in the production and comprehension of discourse. Language users also construe mental models of the situations, events or actions the discourse refers to, what it is about. Indeed, to understand and to interpret a discourse is to construe a mental model of the event situation referred to. Whereas context models may be called pragmatic, events models may be called semantic.

Also event models are subjective, as is any interpretation of discourse. We may read about the economic crisis in the newspaper, and not only interpret what the news report is about, but also may associate these events with our personal opinion or even with an emotion.

One of the major aims of all communication is to make sure that recipients understand what speakers or writers mean. In other words, this involves that they form the mental model the speaker or the writer prefers.

We here are at the heart of what we described as mind control, namely the discursive control of the mental models of the recipients of public discourse. As soon as such discourse is able to manage the mental models of citizens about public events, the symbolic elites have reached one of their main communicative goals.

In order to influence both the context models and the event models of the recipients, the producers of public discourse must make sure to organize their discourse in a way that most likely shapes the mental models of citizens in the way the symbolic elites prefer. As we have seen above, they will typically do so by dealing with the most influential topics, the most persuasive arguments, or the most impressive rhetoric at all levels of discourse.

As such, this is no problem, since we all communicate with the intention that people understand what we say and what we mean. This is also true for public discourse. However, as soon as these mental models reflect the way the symbolic elites want us to interpret what happens in the world, such models may of course be biased. And if they are biased in the interest of the symbolic elites, and against the interests of the citizens, information or persuasion may turn into manipulation. This is typically the case, for instance, in news reports about immigration, in which the arrival of a group of immigrants may be represented as a threat, or in which culturally different behaviour of immigrants is represented as deviant and hence as problematic.


Mental models are subjective representations of unique events. Much communication and discourse, however, is not limited to inform or manipulate the recipients about single events. As is the case in educational discourse, also many other discourses aim to influence our knowledge more generally. This is what we call learning. Reading a newspaper, watching television, or reading a novel also involves learning about the world in general, and not just about single events. It is in this way that discourse is also crucial in the formation of our general knowledge. One of the ways this happens is by the generalization of mental models. By reading or listening to news stories about Iraq, people more generally learn more about war, bombs, terrorism and counterterrorism, about the Middle East, about United States, and so on. And vice versa, we only are able to understand discourse, and hence to construe mental models of events in the world, if we already have vast amounts of knowledge. This also allows discourse to be implicit and incomplete, because language users know that recipients can infer missing information from their world knowledge. Discourses are like icebergs, of which typically only new information tends to be expressed, whereas most presupposed or implied information remains invisible in the text — although it is of course present in the mental model of the text.

Education, information, persuasion and manipulation, among other forms of discursive communication, thus involve the formation and transformation of general, socially shared representations such as knowledge, for instance through the generalization of mental models. Again, this may typically involve the kind of knowledge as preferred by the symbolic elites, and this will generally be knowledge that is not inconsistent with its interests.

Attitudes and Ideologies

Public discourse often is not limited to the communication of preferred knowledge. We not only have knowledge about the world, but crucially also have personal opinions about what happens in the world, as we have seen in the formation of mental models. If these opinions are shared by many people, and if they are about important issues in social life, we speak about attitudes, for instance about abortion, euthanasia, the economic crisis, or immigration, among a host of other attitudes.

If the symbolic elites are able to form the attitudes of the citizens about crucial social issues, they do not need to influence each and every mental model about each event, because the citizens will form each new mental model about an issue according to their more general, and more fundamental, attitudes.

Finally, even these attitudes are not the ultimate and basic goals of discourse and communication. It would be even more efficient if we not only influence the attitudes of people, but rather their more general ideologies. So, if we are able to gradually influence citizens to construe, for instance, a neo-liberal ideology, which in turn organizes many attitudes about the market, jobs, the relations between the state and business enterprises, we efficiently control a vast amount of mental models about specific events in society.

In other words, ideological control is the essential goal of many forms of public discourse. Once we are able to form and confirm the ideologies of people, we indirectly control the formation of new attitudes, and these attitudes will in turn influence the formation of mental models. These mental models are crucial in the production and understanding of discourse, and more in general for the management of every day interaction.

We now have come full circle of the process of discursive power, beginning with the power of the symbolic elites to control the context of discourse production, which controls the forms and content of discourse, which controls the mental models of interpretation, which finally control and are controlled by underlying, socially shared knowledge, attitudes, and ideologies, which again influence people’s conduct and hence their discourse.

Discourse, power and legitimacy

It must be repeated that ideological control is not necessarily bad. Most of us educate our children in such a way that they have democratic ideologies. We have all been influenced by many messages that have re-educated us to become ecologically more conscious citizens. We may want to contribute to the formation of a feminist view of gender relationships, or maybe of a more socially conscious ideology about class. We educate our students with our ideological views about scientific research. And so on.

For the same reasons, not all actions, discourse and ideological control of the symbolic elites should be seen as essentially negative. The problem, rather, is when the discursive power of the symbolic elites is abused of against the best interests of the citizens, and in the best interests of the elites themselves. We here touch upon the fundamental question of legitimacy and enter into the area specifically studied by critical discourse analysts.

It is not easy to distinguish between legitimate and illegitimate uses of discursive power. When exactly is the influence of a professor on her students beneficial to the students and when does manipulation start? The same can be asked about politicians holding speeches in Parliament, journalists informing the public, or business companies advertising their services or other products. It is probably not too cynical to assume that most discourse and communication is primarily self-serving. Most of us, in most situations, will try to persuade others to think and do as we like. The same is obviously true for the symbolic elites, and hence the difference between legitimate and illegitimate control of context, text and mind is probably quite fuzzy.

Let me give a concrete example from my own research on racism and discourse. Newspapers daily inform about the arrival of new immigrants. We know that these news reports have quite generally led to negative attitudes about immigrants and immigration, and even to more fundamental racist ideologies among many white Europeans. Journalists most likely will deny that their coverage is to blame for this development. They will claim that they only report the facts, or they blame politicians who take advantage of immigration to propagate racist attitudes. So, when is reporting about new immigrants a legitimate form of information, and when is it an illegitimate, even racist practice of manipulating the attitudes of the citizens?

A critical discourse analysis of such media coverage will point out that such news reports are seldom neutral, let alone objective, descriptions of the facts. Headlines, metaphors, adjectives and many other ways in which immigrants or their arrival are described express and convey, sometimes quite subtly, negative opinions and attitudes about the Others. Immigration typically is represented as a problem (that is, a problem for Us and not primarily for Them), metaphorically couched in terms such as waves and avalanches, and rhetorically emphasized by a daily number game of specifying how many new people have arrived. For comparison, think of the way the arrival of tourists is usually covered, namely rather as an economic boon for the country. Although most economic analyses of immigration has shown that the contributions of immigrants to our economy are fundamental, it is striking that as soon as poor immigrants especially from non-European countries arrive, they tend to be represented in negative terms.

And once immigrants are ‘here,’ negative coverage doesn’t stop. Because then we read news reports and opinion articles about problems of integration, differences of language or religion, different cultural habits, and other aspects that somehow seem to bother us. A crucial example is the media treatment of the use of the hijab (head scarf) by some Muslim women. Suddenly many of our symbolic elites, also and even on the left, are suddenly seriously worried about this major threat to our culture or about the male domination of women, sometimes even without realizing the continued domination of women in our own societies.

On the other hand, the real problems of immigrants themselves are hardly dealt with in the press, such as the many forms of every day prejudice, discrimination and racism, the endless harassment by the police and bureaucrats, the red tape of many papers that have to be shown, the serious problems of learning another language, trying to find a decent job, having to accept a lower wage than autochthonous workers, being treated with suspicion in shops, having to raise children in such an unwelcome environment, and so on, and so on. This daily life of immigrants and minorities is not what we daily read about. These are facts. And these facts we do not read about.      So, the least we can say is that the coverage is biased, self-centred, incomplete and often negative. If the same is true in much political discourse about immigration, and if such political discourse reaches the citizens, often again through mass media that are not very critical of such political discourse, we are entitled to conclude that this dominant discourse conditions the formation of ethnic prejudices and racist ideologies among the population at large. And since such attitudes and ideologies demonstrably are in our best interests, namely to maintain white European control over our societies, and against the best interests of the newcomers, we must conclude that such discourse is illegitimate.

Many other examples of illegitimate discursive domination may be given, for example in the way textbooks express and communicate dominant ideologies, how poor people in the world are ignored in virtually all dominant discourse, how women continue to be stereotyped in many forms of text and image, and how the mass media give preferential access to the symbolic elites are not to many others.

Critical discourse studies aim to examine these illegitimate discursive practices as forms of social and political domination. It is our goal to provide more insight into the subtle ways power may be discursively reproduced in society, with the hope that such critical analyses will contribute to the resistance against such forms of discursive domination.

The main tenets of CDA

1. CDA addresses social problems

2. Power relations are discursive

3. Discourse constitutes society and culture

4. Discourse does ideological work

5. Discourse is historical

6. The link between text and society is mediated

7. Discourse analysis is interpretative and explanatory

8. Discourse is a form of social action.