TEUN VAN DIJK
Few notions in the humanities and social sciences are as important and ubiquitous as those of ‘discourse’ and ‘power’. It therefore makes sense to investigate the relations between them, if only because there are people who exercise their power through talk and text, as is the case for the ‘symbolic elites’, such as journalists, teachers and politicians. Before examining the relations between these two notions and the social phenomena they refer to, let us briefly define what we understand by them.
As is the case for all fundamental notions, also the notion of ‘discourse’ can be defined in many ways, as happens in the multidisciplinary field of Discourse Studies. First of all, both written and spoken discourse, text or talk, are forms of language use, and therefore studied in linguistics. Secondly, discourse is a form of social interaction studied in the social sciences, as is the case for a conversation, a parliamentary debate, a telenovela or a WhatsApp message. Thirdly, with discourse we express and communicate mental states such as knowledge, opinions and emotions, which requires further analysis in cognitive and social psychology. Similarly, parliaments debate laws, that is, genres of discourse that are of special interest for political science. Newspapers, television and the internet bring news and other media messages that are studied as forms of mass communication. Books, newspapers, television and the internet offer text and talk for sale, as cultural goods that can be studied in economics. And finally, most of the sources of historians are forms of discourse. And so on. In sum, discourse has many facets; it is ubiquitous in society; and it is studied in many disciplines. This is not surprising because discourse is uniquely human, and hence relevant in all studies that are about human activities and societies.
As forms of language use and social interaction, discourse is typically analyzed in terms of constituent units at several levels, such as sounds or letters, words, phrases, clauses, sentences, paragraphs or turns at talk, as questions and answers, as introductions and conclusions, as headlines and leads, and so on. Some of these units are studied in sentence or discourse grammars, others in conversation analysis, the study of narrative or argumentation or more generally in genre studies.
These numerous different forms of text and talk have many structural and functional properties, such as more or less strict rules that define their sequencing: Articles in many languages precede nouns, headlines and titles are on top of texts, questions come before answers, and so on. Similarly, units may combine and larger units, a fundamental property of discourse construction.
Each of the units of discourse not only consists of sounds, letters or images that can be expressed and perceived, but also express local or global meanings. Thus discourse not only obeys laws of structural ordering, such as grammaticality, but also should be meaningful so as to be able to function as forms of communication. And by expressing ordered and meaningful text and talk in specific communicative and social contexts, people engage in many kinds of social interaction. By thus discursively engage in social interaction, people may exercise power.
The notion of power is hardly less complex than that of discourse. As such, power is abstract: it cannot be seen or touched. It is a property of, or a relation between people. Here we are specifically interested in social power, as a relation between social groups, such as between men and women, blacks and whites, rich and poor, gay and straight, masters and slaves, old and young or bosses and employees.
Of the many ways to define social power, we shall use the notion of control. A group of people has power over other people if they control the other group and its members. This control takes two forms: control of minds and control of actions. Usually to control the actions of other people we first need to control their minds, so that these decide to act in accordance with the wishes of the powerful.
To control the minds – knowledge, opinions, emotions, wishes, etc. – of other people we typically use discourse. The actions of people may be controlled by force, as is the case for military or police force, or the abuse of force by men over women or adults over children. But most forms of mind control in everyday life are discursive, as is the case for laws and regulations, commands and recommendations, instructions, information, education and manipulation, for instance as exercised by the symbolic elites: politicians, journalists and teachers. Some of these forms power are legitimate, others are illegitimate, as is the case for abuse of power or domination.
The power of powerful groups has a specific power basis, a special material or symbolic resource one group has more than another groups, as is the case for force, money, land, status, fame, position or knowledge. Characteristic of the power of the symbolic elites is that they have control over, or preferential access to, the most influential forms of public discourse. And because minds are largely controlled by discourse, the symbolic elites also have more (indirect) control of the public mind.
We thus discover a crucial link between discourse and power: Many if not most forms of legitimate social power are not exercised by direct (bodily) force, but indirectly by text and talk. Thus, those social groups who have exclusive or preferential control over, or access to, to most influential discourse, usually also have more power, as is precisely the case for the symbolic elites.
Yet, as suggested, the relation between discourse and power is indirect: It is mediated by the mind. Discourse is interpreted by social members as language users, and may thus influence their knowledge, opinions, emotions as well as the intentions based on them. We may thus be forced or manipulated to do as the powerful want, but we may also ignore or resist the discursively communicated wishes of the powerful. Some forms of discourse are more ‘powerful’ than others – if they are more successful in influencing the minds of the recipients, and as intended by powerful speakers or authors.
It is at this point that theory and research of the relations between discourse and power becomes most interesting but also most complex. Contemporary discourse studies has become very sophisticated in its analysis of the many complex structures of text and talk. We also know much about the properties and kinds of social power exercised by social groups, organizations and institutions, for instance about sexism, racism or the manipulation or misinformation by politicians or journalists.
We know much less about the details of how the mind, such as our knowledge and especially our opinions and intentions, are controlled by discourse, and therefore about how people’s actions, based on these mental states, are influenced by discourse. Indeed, there are many examples of public discourses, in politics or the media, which did not have the intended influence on the minds and actions of the public. In other words, the relations between discourse and social power are mediated by the mind.
Discourse, Cognition and Society
Theoretically, these complex relations between discourse and power can only be fully understood in multidisciplinary research that examines how structures of discourse are linked with mental structures and processes, and how these are in turn related to forms of interaction and more generally to societal structures. So let us briefly examine this cognitive interface in somewhat more detail.
The structures and processes of the mind define what we call cognition and take place in memory as it is implemented in the billions of neurons of the brain. Although current neuroscience has made much progress in the study of the brain, we still have little insight in how exactly the typical structures of the mind, such as knowledge and opinions, are implemented in the ‘hardware’ of the brain. Yet, even without such insight about their neurological basis, cognitive psychology has made many useful theoretical distinctions that explain human thought and action.
Thus, we distinguish between different types and functions of memory, such as between Short Term Memory (STM) and Long Term Memory (LTM). When we understand text or talk, we first sequentially process and interpret the words and sentences in STM and then store the result of this process of understanding in LTM, from where we may later (partially) recall what we have read or heard. It is also in this way how we acquire knowledge from discourse about specific events, or about the world in general.
Another useful theoretical distinction made in cognitive psychology is that between Episodic Memory and Semantic Memory. Episodic Memory (EM) records all our individual experiences and hence has a more personal, subjective and autobiographical nature. Most of the more trivial personal memories are only relevant ad hoc and no longer retrievable after longer periods of time. Few of us recall a year from now what we bought in the supermarket today. We do however better recall emotionally relevant events, such as accidents, or larger events, such as holidays or when and where we have studied.
Today it is often assumed that the representation of specific events in EM takes the form of hierarchically and schematically organized mental models, consisting of a spatiotemporal Setting, an Event or Action, and the Participants (and their identities, roles and relations). When we understand discourse, we not only interpret its meaning, but at the same time construe a mental representation of the events or action a discourse is about, that is, a mental model. Conversely, if we tell a story about a personal experience, as it is represented in a subjective mental model in EM, we transform contextually relevant parts of the mental model into meanings expressed in talk. Mental models are not only personal but also subjective and may also feature opinions and emotions of the events we experience. They are typically multimodal: they may have visual, auditory, sensorimotor and other information derived from our senses and processed in different parts of the brain.
Social Cognition: Knowledge, Attitudes and Ideologies
The mental representations of Semantic Memory, on the other hand, are more generic and socially shared with other members of social groups and communities, as is the case for the knowledge of our language, and our generic knowledge of the world. This knowledge is partly acquired through our personal experiences, for instance by generalization and abstraction of mental models. We know about supermarkets, traffic, houses, and holidays, among hundreds of thousands of things of social life, by generalizing from our personal experiences of them. We may also acquire knowledge more directly for instance by generic discourse in the mass media or textbooks, for instance about abstract notions or about aspects of the world that are not experienced in everyday life, for instance knowledge about politics, geography, biology or genetics.
The knowledge shared by the members of an (epistemic) community not only serves to form new personal mental models (we need to have generic knowledge about cars and accidents to understand a story or news report about car accidents), but also for all forms of human interaction and communication, and hence for discourse.
Similarly, based on this knowledge, members of social groups may gradually form socially shared attitudes, for instance about immigration of abortion, or ideologies such as those of sexism or antisexism, racism and antiracism, neoliberalism and socialism. Such forms of social cognition are typically acquired by discourse, such as ideological text and talk in the mass media or the internet, or from conversations with other members of an ideological group.
The sociocognitive and discursive reproduction of power and domination
It is at the crucial point where the link is established between discourse and power. Members of dominant groups not only have preferential access to public discourse, but also share forms of social cognition, such as knowledge, attitudes and ideologies that influence their discourse. Such discourse may in turn influence the mental models of the recipients (as we do when we interpret news stories), which in turn may influence the attitudes (for instance about immigrants) and even the ideologies (such as racism) of the recipients. Complex processes of persuasion and manipulation may thus result in social attitudes and ideologies that are in the interest of dominant groups.
We see how through complex structures of discourse and cognition social power and power abuse can be reproduced in society, for instance when dominated groups accept and internalize the attitudes and ideologies of dominant groups – a process often called hegemony.
The same processes also explain counterpower, namely when members of dominated groups have access to public discourse, for instance via social media, and are able to communicate alternative discourses, expressing alternative attitudes and ideologies and criticizing the dominant ones. This recently happened for instance during the Arabic Spring, as well as in the many social movements that have arisen during the economic crisis.
An example: “Waves” of immigrants
To illustrate this complex relation between discourse, cognition and power, consider the routinely used metaphors in the coverage of immigration: The arrival of immigrants is often described in the news in terms of waves or invasions.
These are not just innocent metaphors used to write about the arrival of large numbers of immigrants. Recipients of news construe mental models of such news items in which the metaphors construe multimodal experiences related to waves and invasions, such as the sensation to drown in such masses of immigrants or that there neighborhood, city or country is occupied by a alien force. These sensations may typically trigger emotions of fear or anger, which then are associated with the mental model of the news report.
If such news reports, metaphors and their corresponding mental models are repeated, they may become generalized and socially shared as negative attitudes about immigration. These may finally be organized in racist or xenophobic ideologies when other negative attitudes about minorities or immigrants are thus created.
These processes account for the relation between discourse and social cognition. When such attitudes and ideologies are in the interest of symbolic elites, such as politicians, their discourses may be used to enhance and reproduce their power, for instance when many members of the public at large vote for their political party. If such a political party thus acquires more power and even enters government, they may enact laws further restricting immigration. This is exactly what has happened in the last decades in Europe.
We thus see how metaphors (as well as other negative aspects of text and talk: topics, group descriptions, etc.) may influence mental models and social cognition, which in turn may contribute to the power and domination of social groups, resulting in social inequality.
Teun A. van Dijk es profesor en la Universitat Pompeu Fabra. Excatedrático de estudios del discurso en la Universidad de Ámsterdam www.discursos.org
Publicado en Beerderberg
Leer en español
Descargar en pdf
Ver el resto de artículos del número 5