Paper Congress on Immigration, Almería April, 21-22, 2005
Reproducing Racism: The Role of the Press.
Teun A. van Dijk
Universitat Pompeu Fabra,
First version. April 17, 2005
In this paper I provide a theoretical analysis of the role of the press in the reproduction of racism. Many empirical studies in many countries have shown that the media play an important role in expressing and spreading ethnic prejudice. And it is generally assumed that prejudice is one of the conditions of racist practices that define racism as the social system of ethnic power abuse. However, the relationships between media discourse and racism are generally assumed to exist, based on correlational evidence, without much detailed theoretical analysis of the precise linguistic, cognitive and social nature of these relationships. A multidisciplinary approach to the study of the press as well as of racism is able to elucidate some of these relationships in a more explicit way.
Racism is here defined as a
system of ethnic or ‘racial’ dominance, that is, of systematic power abuse of a
dominant (European, ‘white’) group against various kinds of non-European groups
-- such as ethnic minorities, immigrants and refugees -- in Europe, the
The system of ethnic domination has two major dimensions, namely a social and cognitive dimension. The first, social dimension consist of the everyday social practices of discrimination against ethnically different groups, e.g., through the exclusion from, or unequal distribution of social resources or human rights. The second, cognitive dimension consists of the ethnic beliefs, stereotypes, prejudices and ideologies that function as the motivation and legitimation of such discriminatory practices. In very general terms, this means that outgroups are represented negatively (as different, deviant or a threat) compared to the European ingroup, which represents itself as superior in all relevant attributes.
It should be emphasized from the start that the concept of ‘racism’ is not understood here to apply only to radical, extreme, violent or blatant forms of racism. On the contrary, most forms of racism today – and those that interests me – are the kind of indirect, subtle, or ‘modern’ racisms in everyday life characterizing the many daily encounters by members of majority and minority groups.
One of the most crucial social practices is discourse, that is, socially situated text and talk. Within the system of racism, this means that just like other discriminatory practices, discourse may be used to problematize, marginalize, exclude or otherwise limit the human rights of ethnic outgroups. Such may be the case either by direct discriminatory discourse in interaction with ethnic Others, or indirectly by writing or speaking negatively about the Others (see, e.g., (for detail, see Blommaert & Verschueren, 1998; Jäger, 1992, 1998; Reisigl & Wodak, 2000, 2001; Van Dijk, 1984, 1987a, 1987b, 1991, 1993; Wetherell & Potter, 1992; Wodak & Van Dijk, 2000).
One of the main roles of discourse is the reproduction of social representations, such as knowledge, attitudes, ideologies, norms and values. This means that discourse is the main interface between the social and cognitive dimensions of racism. On the one hand it may itself be a discriminatory social practice, and on the other hand it expresses and helps reproduce the negative social representations (prejudices, etc.) that are the socially shared mental basis of such social practices.
Elite discourse and racism
Not all types of discourse are equally influential in the reproduction of society and of systems of domination such as racism. Obviously, public discourses are more influential throughout society than private discourses such as everyday conversations in the family, among neighbors or friends. Those groups who are in control of most influential public discourses, that is symbolic elites such as politicians, journalists, scholars, teachers and writers, thus play a special role in the reproduction of dominant knowledge and ideologies in society. Since prejudices are not innate, but socially acquired, and since such acquisition is predominantly discursive, the public discourses of the symbolic elites are the primary source of shared ethnic prejudices and ideologies (Van Dijk, 1993).
Popular racism, and its practices and discourse, are often based on, exacerbated or legitimated by such elite discourse and racism. Although popular racism in principle may be reproduced by shared personal experiences with ethnic Others, as well as by interpersonal conversations, it is unlikely that it may have the fast and widespread influence of public discourses such as parliamentary debates, news, TV programs, novels, movies or textbooks. Even when the media may “give voice” to popular racism, it is still the media elites who are responsible for this publication and reproduction in the public sphere. That is, the elites at least preformulate, legitimate or condone popular racism.
Obviously, the same arguments hold for the reproduction of antiracist practices and ideologies in society. However, the antiracist elites in all domains of society – politics, media, education, research, etc. – have much less influence, and are themselves often problematized and marginalized.
We see that by this logic of the role the symbolic elites and of public discourse in the reproduction of racist social representations, the mass media play a prominent role. Even if politicians sometimes have the first word on ethnic issues, for instance in parliamentary debates, their discourses and opinions become influential only through media accounts. Scholars and writers may publish books and articles, but the main results of these studies become part of the public domain only when reported and popularized in the news media. Textbooks are -- obligatorily -- used by, and influence millions of children worldwide, but these kids hardly have much influence on the general racist attitudes and ideologies in society. In sum, the mass media are currently the most influential source of racist bias, prejudice and racism. We therefore need to examine this role in much more detail (see also Campbell, 1995; Chávez, 2001; Cottle, 2000; Entman & Rojecki, 2000; Gandy, 1998; Hartmann & Husband, 1974; Jacobs, 2000; Jäger & Link, 1993; Kellstedt, 2003; Prieto Ramos, 2004; Rome, 2004; Ruhrmann, 1995; Smitherman-Donaldson & Van Dijk, 1987); Ter Wal, 2002; Van Dijk, 1991, 1997; Wilson, 2005; Wilson & Gutiérrez, 1985).
I shall do so primarily for the written press, even when in many ways television has more influence, because of the role of the press in the formation of elite opinion. More detailed news, editorials and opinion articles in the press are crucially involved in the formation of ethnic attitudes and ideologies.
A detailed study of the role of the press in the reproduction of racism in society presupposes the general assumptions and findings summarized above, e.g., about the role of public discourse in the reproduction of beliefs, the elite control over public discourses, the primary role of the media among elite discourses, and the role of racist discourse in the reproduction of racism.
A study of the role of the press in the reproduction of racism requires a multidisciplinary framework that systematically accounts for the contexts, the production, the structures and the effects of newspaper discourses such as news reports, editorials, and opinion articles. That is, the often observed biases of press accounts of ethnic events may be largely due to biased sources that are uncritically reproduced, to opinion leaders freely admitted to write in the newspaper, or due to the biased attitudes of reporters or editors themselves – or a combination of these factors. Let us examine these different stages or dimensions of press discourse production, discourse and discourse effects in some more detail.
So far, I have used the general
label ‘the press’ to refer to newspapers. Obviously, however, the press in
different countries, and of different types (popular or elite) and different
sociopolitical orientation and ideologies is rather varied. As to their role in
the reproduction of racism, for instance, the Guardian and The Sun in
Ideally, however, we would need to provide detailed context analysis for each newspaper, in terms of the education, social position, status, and ethnic beliefs of owners, editors and reporters, as well as their readers, on the one hand, and the specific production processes on the other hand. If I generalize below, this means that such generalizations hold for the (vast) majority of the press. If not, I shall indicate contextually based variation.
News and opinion articles in the press largely depend on outside sources, and in this sense press discourse is intimately linked to the public discourses of the other elites, mostly politicians, scholars, lawyers, writers, etc. Whereas opinion articles and columns are written by writers who may be more or less closely related to the newspaper, as well as by independent outside writers, news discourse is organized, produced and written by journalists, even when they use various sources. However, in all cases, except when owners control editors, journalists decide and are responsible for what is published in the newspaper (Gans, 1979; Tuchman, 1978).
News production in the written media is a very complex set of social and discursive interactions, controlled by editors, involving editorial meetings and negotiations, story assignments, reporters’ beats with a limited number or institutions and organizations, such as parliament, the police, the courts, the universities and (big) businesses, extensive interaction with necessary or probable sources of ‘information’ and ‘opinion’, writing up of news stories by reporters, and final editing by editors. The microsociology, social psychology and discourse analysis of all these processes and strategies of newsmaking is much more detailed, and at present not yet fully explored; indeed, editorial meetings, like many other elite encounters, are seldom accessible to scholars – and even less to critical scholars, and totally without access for critical scholars writing about racism and the press.
Selection of sources and source texts
Yet, we know that much of this process consist of various forms of ‘text processing’, in the sense that different kinds of ‘source texts’ are sought for and acquired by reporters, for instance through document research, consulting other media, interviews, press conferences, telephone conversations, and so on (Van Dijk, 1988). Many such sources texts reach the newspaper on the initiative of institutions, organizations and individual citizens, e.g., in the form of agency messages, press releases, phone calls, e-mails, faxes, letters, and so on.
Since the vast majority of all these source texts cannot possibly be published in the newspaper, reporters and editors apply a rigorous system of selection and information reduction, e.g., in the form of summarizing. That is, source texts may have influence on the knowledge and opinions of journalists, but need not as such appear in the newspaper. Also, if they appear in the newspaper they generally tend to be modified in so many ways, that the resulting news report should be seen as the collective textual product of the newspaper and its journalists.
It is important to insist on
this point when we examine the causes and responsibilities of the role of the
press in the reproduction of racism. Journalists, who are hardly used to
published criticism (indeed, which newspaper publishes critical reports about
the press, let alone about racism in the press?), tend to defend themselves
against critique also by relaying the blame to their sources, a well-known
strategy of denying racism. True, journalists are not responsible for the
(racist) talk of politicians or other elites they write about. Publishing about
such talk may even have a critical function if properly formulated, for
instance as a form of accusation. This is however generally the case only for
the (critical) reporting about the extreme right, e.g. about the Front National
In sum, on any account journalist are ultimately responsible, either for uncritically publishing racist talk of other elites, or for writing their own biased discourses.
As suggested, we as yet barely have insight into the details of these news production processes. From participant observation and ethnographic reports, we know some about editorial meetings and about the everyday routines of news gathering with several organizations and institutions (Gans, 1979; Tuchman, 1978). We have some insight into the kind of source texts produced by these processes, and how these are selected or discarded, and used to write news reports. We have no idea, however, how all these discourse types, also the discarded ones, manage the minds of the journalists: what knowledge and opinions result from reading or skimming (parts of) these source texts? How exactly are they read, understood and stored in memory as mental models of ethnic events, which in later news report writing may again be used.
What we do know from some empirical research, however, is that source texts of all groups who are considered to be less important, less powerful or less interesting, tend to be ignored, overlooked or discarded (Van Dijk, 1991). This is specifically also true for those of ethnic minority groups, organizations or persons. Their press releases tend to wind up in the wastepaper basket, and only the largest organizations and in special circumstances may be explicitly sought after or their press releases used in news production. Ethnic minorities, their leaders or spokespersons, are not usually considered experts about ethnic events, even about those events that involve themselves. Rather, they are typically considered biased sources, whereas (white) politicians, police officers, lawyers, scholars or organizations tend to be seen as ‘independent’ or ‘expert’ and hence as reliable sources, also on ethnic events. ‘Our’ white group and its members are never seen as being ‘ethnic’ in the first place.
Discrimination of minority journalists
The social and cognitive processes involved in the production of news, such as gathering, selection, discarding and summarization of source texts and their resulting mental models (subjective interpretations) in the minds of journalists, show how news and bias are produced. Journalists prefer white sources and their texts, and judge these to be more reliable and objective.
This is also because most
journalists of most newspapers in Europe and the
Since minority journalists are
virtually absent from most newsrooms, especially in Europe, this not only means
that news production in general will be biased by a white perspective, but also
that fundamental knowledge and expertise on ethnic communities and experiences
are usually lacking in reporting. It has often been observed that most white
journalists have few daily contacts with ethnic communities, and reporting
ethnic events is thus not only biased but often also ignorant. Few journalism
The conclusion is that not only the news production process, but the very social structure of hiring and the composition of the newsroom do not favor another perspective on news gathering and news production. Critical, antiracist sources and their texts and opinions are not routinely sought for, nor will these be easily admitted, and the same is true for antiracist journalists.
News production is thus generally ‘white news’ production, even when the news is about ethnic affairs, such as immigration. Part of this bias can be explained in general terms – also many other social minorities are thus discriminated in news production, and news and news sources generally are those of the powerful elites in society, and these happen to be largely white males in European societies.
But another part of this bias is the result of racist prejudices, e.g., when white sources are by definition found to be better and more reliable, when ethnic groups and organizations are found less important or newsworthy, or when problems and issues, e.g. racism itself, of minority groups are not found to be very relevant. And conversely, that disproportionate attention is paid to the differences, deviance or alleged threats of ethnic minority groups and their members, also shows that the very expectations and criteria of journalists are ethnically biased by negative social representations about the Others.
In sum, news production is racist because of all these factors of routine news gathering, selection, preference for white sources and their texts, discrimination of minority journalists, as well as the biased interest in specific negative topics associated with minorities.
All this holds for the press generally. More liberal newspapers do not have more minority journalists and a different news gathering system from conservative newspapers, and the same is true for elite and popular newspapers. In the Americas some more minority journalists may be employed – variable in different countries – but both in North America, as well as in most countries of Central and South America, newspaper owners, chief editors, editors and most reporters are predominantly white(r). As far as I know, no newspaper has an explicit code of non-racist news gathering and news production – and only some have guidelines for non-biased news writing.
It is not surprising that given this social context of news production, news reports on ethnic events tend to be biased in many ways. Since these are the discourses that reach millions of readers every day, their systematic and critical analysis is crucial. Probably no other public discourse influences the formation of ethnic prejudice and stereotypes as much as news, also because most white people do not have daily experiences with members of ethnic minority groups or with recent immigrants. This is true for news in the press as well as for news and other programs on television, but the elites are more generally influenced by news in the press, and the same is true for those elites who produce television programs. Moreover, television news is only a fragment of the news in the newspaper.
So let us summarize some of the findings of the many studies on racism in the news. Since news reports, like all discourse, have hundreds of relevant structural categories, we obviously can only focus on a few typical ones.
One of the first questions one may ask in a critical analysis of discourse in general, and of news about ethnic affairs in particular, is about the topics of text and talk: What do people write or speak about when they refer to ethnic minorities, immigrants, refugees or in general ethnic Others? Theoretically, this means that we ask about what have been called the ‘semantic macrostructures’ of discourse, that is global meanings that organize the local meaning of words and sentences at higher levels of paragraphs and whole discourses (Van Dijk, 1980). Such topics are not only important because they provide global coherence to discourse, but also because they are the information that is best remembered and that in turn organizes how we represent ethnic events in our personal, episodic memories of everyday experiences.
In news reports, such topics are typically expressed in headlines and, in somewhat more detail, in the lead, which also form the first, most prominent part of news reports, and the part that is mostly read, sometimes exclusively (Van Dijk, 1988). The formulation of headlines and leads reflects the way the newspaper (reporter, editor, etc.) frames the topics and how these organize the meaning of the whole text.
Topics in the news may be about virtually any subject that is of public concern, especially of the elites. Topics in news on ethnic minorities, however, do not have such a broad variation. On the contrary, much research has shown that ‘ethnic’ news focuses on only a few main topics, which on the whole may be summarized as problem-news.
Initially, when non-European
groups are immigrating, such immigration tends to be represented as a major
problem, if not as an invasion and hence as a threat, for instance of our
welfare, job market or culture. Stories abound, as is currently still the case
Once the newcomers are settled in the country, the next major class of topics focuses again on problems, namely on problems of reception, housing, employment and integration. This may be an opportunity to tell many stories about the major problems immigrants have to build up their lives in the new country. Some newspapers sometimes tell such (background) stories, typically in special supplements, but on the whole, and especially in conservative tabloid newspapers, such stories focus on the problem they mean for us: they take away our jobs and create unemployment, they take away our houses, and they do not want to adapt: they have strange habits, do not want to learn our language, and so on. Most virulent in this case, for instance in the British tabloid press, are constant accusations of abuse of welfare and benefits, temporary housing in expensive hotels, and other sub-topics that may arouse indignation of the “taxpayers”. Whereas dealing with cultural differences may originally still have an exotic flavor, these soon are topicalized as unacceptable differences, that is as deviance, and finally as a threat to our welfare state and culture.
Thirdly, one of the most frequent topics associated with minorities and immigrants is crime and violence. This may begin with the emphasis on ‘illegal’ immigration and residence and the activities of mafias (even if immigrants are victims of such mafias!), but soon is a major topic describing their activities in the country. Crime reporting more generally thus becomes ‘ethnicized’, and specific forms of delinquency typically and selectively attributed to ‘foreigners’, even when the majority of such crimes are committed by nationals: drugs and hold-ups.
These are the three main topic classes that characterize most news on minorities and immigrants in most countries most of the time. The overall message is that the newcomers or minorities are at least a problem if not a threat for us, and the general perspective of such stories is white – not the problems they have with us.
Depending on the country and the current state of immigration and settlement, further topics may be the political debate on immigration, e.g. on new immigration laws, legalization and in general the reactions of politicians and the authorities to recent immigration or minority groups. Again, also under pressure of racist parties, such debate tends to focus on the problems allegedly created by immigration, residence and integration (Wodak & Van Dijk, 2000). The negative definition of the situation by the politicians thus becomes a prominent topic by itself – and a powerful legitimation of popular racism.
It is not surprising that
immigration restrictions have been the dominant answer to increased immigration
The general interest of newspapers in the political elites not only promotes interest in immigration policies and debates, but also the interest in racist parties and leaders, as was and is the case for such figures as Le Pen in France, Haider in Austria, and to a lesser extent Fini in Italy, Dewinter in Belgium and Fortuyn in Holland. As leaders of the racist right, or as vociferous opponents of immigration, they represent the “official” racism in some of these countries, and hence may be portrayed negatively. This implies that similar anti-immigration ideas and policies of the mainstream parties are not represented as racist, as is the case for recent immigration restrictions by Tony Blair’s New Labour government today.
Such special interest in the racist
right may spawn more critical stories, but as suggested this does not mean a
general anti-racist stand of newspapers. No leading newspaper in
Stories on elite racism are rare. Stories on racism in the press simply non-existent, neither in the popular tabloids, nor in left-wing quality newspapers. Racism as an everyday problem for immigrants and minorities seldom gets topicalized, even when they define racism as the major problem of living in their new home country, besides getting decent housing and a job.
This also suggests that the global meanings of news – and the same is true for editorials and opinion articles – are also characterized by the topics that are absent: The problem the Others have entering the country, finding housing and work, integration in a new social and cultural environment, or everyday experiences with more or less blatant racism, such as exclusion, marginalization and problematization. On the contrary, the every lives of the Others are seldom portrayed in the press, even on topics that would be covered for European people. The same is true for the topics that negatively reflect back on us, such as racism, prejudice, discrimination, intolerance, lacking hospitality, nationalism, and so on. If treated at all, these are portrayed as exceptional, of people of the extreme right, of a few ‘racist’ individuals, and never of the elites, never as part of the system, and never as the general rule.
The same logic of positive
self-presentation and negative other presentation explains why positive stories
about them are similarly rare.
Although immigrants in all European countries have significantly contributed to
the economic welfare of these countries, such contributions seldom make
headlines in the press. Not to speak of their contributions to culture and the
arts, social diversity, multiculturalism, multilingualism, international
relations, cuisine, fashion, and not to forget sports– although the latter
topics may occasionally enter the marginal human interest and sports pages. The
famous black footballer, however, will in that case typically be defined as English,
Dutch or French, and not as
Since topics are normatively expressed in headlines, the predominance of negative topics in the news also shows in the headlines, as the dominant and defining structural category of news reports. Similarly, the more negative a story about them, the more it will appear prominently and at length in the paper, that is, on the front page, on top, across several columns, with big letter types. The opposite will be the case for stories, if any, on our racism.
We see that the general ideological position of ‘ethnic’ reporting not only influence topic selection and construction, but also prominence, size, lay-out, placement, and typography – all geared towards emphasizing their bad things, and de-emphasizing our bad things.
If topics as semantic macrostructures control local meanings, negative topics organize negative local meanings, from words to complex descriptions of situations, events, actions and people. There are many ways such local semantics may be biased, and I shall summarize only a few, according to the general, macrostrategy of positive self-presentation and negative other-presentation dictated by the well-known ideological polarization between ingroup and outgroup representation in racist discourse.
Implicit vs. Explicit meanings. Most meanings of discourse are implicit, because discourse presupposes that readers have vast amounts of world knowledge that allows them to fill in or derive the information that is implicit in text and talk. This is also true for news reports and for discourse on ethnic affairs. This means that many of the negative meanings that are controlled by negative overall topics (e.g., illegal immigration, violence, drugs, cultural deviance, etc.) will typically be expressed explicitly, and thus emphasized, whereas ‘our’ prejudices and racism will remain implicit.
Presupposition. More specifically, information may be presupposed in news that is not at all a shared ‘fact’. Thus, negative attributes of minorities may simply be indirectly stated through presupposition, e.g., by such expressions as ‘delinquency among immigrants’, which presupposes that immigrants are delinquents.
Detailed vs. general descriptions. One of the ways to emphasize the negative acts of the Others, is to provide very detailed descriptions, e.g. of their alleged cultural deviance, violence or crimes. No such detailed, but at most very general, descriptions are given about the bad situation in which many immigrants live, or the ways we treat them in everyday life. The same is true for more or less precise vs. vague descriptions.
Disclaimers. Well-known are the local semantic moves that combine positive self-presentation and negative other presentation in one formula: “I have nothing against blacks (Arabs, etc.), but…”. This apparent denial, meant to avoid negative impression formation by the recipients, tends to be followed by a largely negative characterization of the Others. Disclaimers come in many formats, such as transfers (“I have no problem with them, but my clients…”), apparent concession (“Some of them are smart, but in general…”), apparent empathy (“Yes, they have problems in their country, but…”), and so on. These disclaimers typically appear in argumentative discourse in the newspaper, such as in Letters to the editor, editorials and opinion articles.
Prevalent especially in argumentative opinion discourse in the press is the ample use of rhetorical figures, which have the well-known persuasive function of emphasizing and de-emphasizing meanings, most typically represented by hyperboles and euphemisms. Thus, our racism will typically be described, if at all, in many euphemistic ways, e.g., as discrimination, bias, or even as ‘popular discontent’ – a phrase that even suggests democratic values. On the other hand, their negative actions or attributes will tend to be described with hyperboles.
Metaphors are the best known semantic and rhetorical means to make complex or abstract meanings more concrete and understandable. This is also characteristically the case in the rhetoric of ethnic event reporting, starting with the now standard description of immigration as an ‘invasion’ by ‘waves’ of immigrants or refugees. In many languages and countries, immigrants thus typically are represented as a threat, especially in terms of threatening masses of water – in which ‘we’ may drown. The description of a threat to our life can hardly get more concrete. Immigration restriction thus becomes a matter of life and death.
Given the contextual properties of news production described above, in which white sources and source texts are preferred and found more reliable or interesting, we may expect that quotation patterns reflect such biases in newsgathering. This is indeed the case. Even in the accounts of ethnic events, with mostly ethnic minority actors, those who are quoted, and hence may define the situation and give their opinions, are usually the white elites – such as the government, politicians, the police, lawyers, NGOs or professors. If the Others and their leaders are cited at all, this nearly always happens in company of white speakers. That is, the Others are seldom the ones who are allowed to define the ethnic event or situation alone.
We now have only mentioned a few examples of typical structures of news that reflect underlying mental models and social representations of journalists and writers who have access to the newspaper. We have seen that at all discourse levels of news and other press genres, such bias is expressed by an overall strategy of positive self-presentation and negative other-presentation. Immigrants and ethnic minorities thus tend to be represented as a problem, as deviant or as a threat, from the moment they enter the country in ‘waves’ of ‘illegal’ immigrants, to the many situations of daily interaction in which they are perceived as different, as strange if not as a threat to our safety, welfare or culture. This is not only true for the important selection of news topics and the headlines, but also for more or less explicit, precise or detailed local meanings, the manipulation of presuppositions, disclaimers, the choice of hyperboles and euphemisms, and metaphors, among many other structures of discourse that allow the writers to emphasize their bad things and our good things. This is also true for lay-out, placement, size, pictures, and a host of structures not mentioned above, such as arguments (and fallacies), the order and format of news reports, person and group descriptions, and so on.
Biased news production and news reports would be pretty harmless if they would not have a tremendous influence on the readers. Although traditional effect-research in mass communication often has found that the media are not that influential at all, and that most people make up their own minds, more or less independently of the media (Bryant and Zillmann, 1994), this is certainly not the case for the role of the press in the reproduction of ethnic prejudice.
There is ample evidence that
what people know about immigration and minorities in most countries is learned
from the mass media, and from everyday conversations with people who have their
information from the mass media (Van Dijk, 1987a). Indeed, new immigrants can
only be known through the mass media – since much daily interaction with them
has not yet taken place. In later stages of immigration and settlement, and
with a relatively low percentage of immigrants, and given the usual social
segregation, daily encounters with immigrants is limited for most white people,
and to few neighborhoods. That is, most of what most white people know about
the Others is not based on direct or indirect personal experiences, but
necessarily on the mass media. Only in countries with large minority groups or
long-year immigration, interethnic encounters are much more common, although,
again, often in unequal situations, for instance of middle or upper class
whites with non-white maids, servants, waiters, receptionists, secretaries, and
so on, and much less as in close and daily encounters with colleagues, family
members or friends. This is typically the situation in e.g. Latin America where
indigenous people (e.g. in
In other words, in most European dominated countries, knowledge and other beliefs about the Others is exclusively or largely based on the mass media. This is especially true for the events and ‘facts’ that go beyond trivial everyday events and encounters, such as the arrival of new immigrants, government policies, the general situation of the labor market, population statistics, party programs, legislation and parliamentary debates, crime statistics, and so on.
Other sources of public
discourse about ethnic events are limited to specific segments of the public,
as is the case for textbooks and children books for children, textbooks and
scholarly publications for students, and novels for those interested in
literature. These discourse genres seldom deal with current ethnic events, and
hence largely tend to provide information or opinions on historical events or a
previous ethnic situation. Occasionally, interethnic encounters may be a topic
in soap operas in the USA or in telenovelas in Latin America – but at least in
the latter case, the appearance of black actors is known to be scandalously
scarce (D’Adesky, 2001; Van Dijk, 2003) and of indigenous actors nearly
non-existed. In the
In sum, the main source of ethnic information are the mass media. For the elites this is mainly the press. If the elites, as suggested, play a major role in the reproduction of racism, this means that the press is crucial for their ethnic opinion formation and confirmation. Since prejudices are not innate and only marginally learned from other sources, they must be based on, or directly derived from, articles in the press. Except from professional and other limited circulation communication, elites also know about each other through the press. Scholars know about politicians through the newspaper, and vice versa. And all elites know journalists only through the press. Public opinion formation and debate, especially among the elites, is largely carried out through the press, and occasionally through television, and increasingly through the internet.
In sum, a general argument may be defended that the press, and hence the news, editorials and opinion articles in the press play a prominent role in the formation of ethnic representations shared among the elites. However, this does not yet prove that ethnic reporting, even when largely negative or otherwise biased, is the main ‘cause’ of ethnic prejudices among the elites. Audience research has often suggested that the public often makes up its mind independently of the mass media. This is certainly true for those topics that are associated with the everyday lives and experiences of people, such as work, (un) employment, housing, shopping, family relations, and so on. However, for the reasons mentioned above, this is much less the case for ethnic events and experiences. If these exists, this happens vicariously through media stories, movies, soaps, etc. And since ethnic events, from immigration to integration, barely touches the every lives of most white people, also their opinions, attitudes and ideologies are not usually shaped by regular interethnic encounters. That means, quite crucially, that the generally negative representation of ethnic events and minorities in the press is seldom rejected on the basis of different personal experiences.
On the contrary, many daily experiences may themselves be shaped by the mental models learned from the media, for instance when white people are afraid of minority crime. Especially the women among senior citizens often report in interviews about ethnically mixed neighborhoods that they have become quite afraid of mugging (Van Dijk 1987a). This is not primarily the case because they or their neighbors or friends have been frequent victims, but because of what they read in the newspaper. Thus a few mugging stories in the press in which the perpetrators openly of more subtly are identified as minorities may make thousands of readers afraid of everyday life experiences that are quite exceptional, and not necessarily perpetrated more by ethnic minorities.
Very often, biased media stories may be in the general interest of the white audience, e.g., when ‘our’ welfare, culture, safety, and so on are at stake and allegedly threatened by the newcomers. In other words, the generally negative reporting in the press needs a very strong counter-reading, based on an antiracist ideology, in order for negative news stories and opinion articles to be rejected as biased. Since few white readers have the experiences (e.g., from ethnic minorities among their friends or in their family) or the ideological socialization that may contribute to the formation of explicit antiracist ideologies and to dissident reading practices, only a small part of the audience will be relatively immune to negative ethnic reporting. This small group will often also use minority sources, alternative media, internet or critical studies as other sources for the formation of their ethnic knowledge and opinions. The majority of the white public, and the majority of the elites, however, will be rather passive readers of ethnically biased information, and given their previous mediated ideology formation, most likely will generally agree with the negative slant of ethnic news and opinion. This will in turn confirm for the journalists that they write what the audience wants to read, thus completing the cycle of influence and reproduction they have initiated in the first place.
Finally, press discourse is not limited to news, but also features editorials, background articles, features and opinion articles, the first written by journalists and the last by journalists or external elite writers, such as scholars, politicians or social leaders. Most newspapers allow some ideological variation, especially in debates where powerful groups in society disagree – e.g. among different parties. However, such ideological bandwidth is limited. Thus, very critical antiracist positions (nor very racist articles) do not usually make it into the press. Indeed, both ‘radical’ positions tend to be criticized or ridiculed – and antiracist positions sometimes seen as a bigger threat to the country than racist ideologies and practices. Rather, as we have seen before, the majority of the elites, also in the press, have the tendency to reject only extremist racism, while denying or downplaying the more subtle, and especially the elite forms of racism.
Such ideological variation in ethnic affairs, as represented in the press, also shapes and is again shaped by similar opinions among the audience. However, the personal opinions of a scholar or politician or other elite members are usually limited to his or her own professional domain, and become part of the public sphere only when published in (or reported about) in the newspaper. The selection of such articles is made by the editors of the newspaper. So again journalists are responsible for which ethnic views are legitimate part of the public sphere and debate. Thus, together with the elites who are their sources and their selected writers, the journalists are mainly responsible for shaping the contents and the ideological bandwidth of ethnic opinions. They may not influence the details of all the opinions of all the elites, but they most certainly influence the topics, the general tendencies, the overall definitions of the situation, and especially the issues not talked about in the media and hence in the public sphere. It is true for nearly the whole white press that thorough an and critical articles on elite, mainstream racism are not usually published in the press. That is, even the newspapers that are usually more critical, and that tend to publish less racist articles, still are part of the problem of racism by denying its prevalence. This is true for virtually all white elites and their discourses, also in politics, scholarship, literature, and so on.
The detailed social and cognitive processes of newspaper selection, news and opinion article reading, comprehension, storage, memory, recall, uses, and so on, are beyond the scope of this paper (Graber, 1984; Van Dijk & Kintsch, 1983; Van Oostendorp & Goldman, 1999). Such processes not only need to explain the role of the press in the daily reproduction of racism, but also the reproduction of dissent and antiracism, even when these are also shaped by other discourse sources and experiences than those of the mass media.
Yet, all the evidence we currently have about racism, discrimination, prejudice, and the role of the elites in general, and the mass media in particular, strongly suggests that the dominant negative reporting of ethnic affairs in most of the press, is at the same time an expression of dominant elite racism as well as the main cause and source of other elite racisms and their discourses. Much more detailed theoretical and empirical work will be necessary on many aspects of this complex process of ideological reproduction and the role of the mass media, such as the details of newsgathering, the role of sources and source texts, editorial meetings and decision making, hiring and promotion of journalists, daily practices of news writing, the mutual influence of the mass media, as well as the ways different kinds of readers read and process the news and opinion articles on ethnic affairs, with what previous beliefs, and with what cognitive and social consequences. We now have some general insight into the role of the press in the reproduction of racism, but the really interesting aspects of this process may be hidden in details we still ignore.
B. (1998). Black journalists, white media. Stoke on
L., & Solomos, J. (Eds.). (2000). Theories of Race and Racism. A reader.
J., & Verschueren, J. (1998). Debating diversity: Analysing the discourse
B. R. (Ed.). (2001). Race and racism.
J., & Zillmann, D. (Eds.). (1994). Media effects: Advances in theory and
M., & Solomos, J. (Eds.). (1999). Racism.
M., & Solomos, J. (2004). Researching race and racism.
Buezas, T. (1997). Racismo y solidaridad de espanoles, portugueses y
latinoamericanos: Los jóvenes ante otros pueblos y culturas.
C. P. (1995). Race, myth and the news.
E. (2003). Encyclopedia of race and ethnic studies.
L.R. (2001). Covering immigration. Popular images and the politics of the
IOE. (2001). No quieren ser menos! Exploración sobre la discriminación laboral
de los inmigrantes en España.
S. (Ed.). (2000). Ethnic Minorities and the Media.
W. (1997). Black journalists. The NABJ story.
W., & Bonilla-Silva, E. (Eds.). (2003). White out. The continuing
significance of racism.
D'Adesky, J. (2001). Racismos e
anti-racismos no Brasil.
R. M., & Rojecki, A. (2000). The black image in the white mind: Media and
P. (1991). Understanding everyday racism: An interdisciplinary theory.
P., & Goldberg, D.T. (Ed.). (2002). Race critical theories text and
J. R. (2000). Racist
J. R., Vera, H., & Batur, P. (2001). White racism the basics.
O. H. (1998). Communication and race: A structural perspective.
H. (1979). Deciding what's news.
D. T. (1997). Racial subjects: Writing on race in
D. T. (2002). The racial state.
D. T., & Solomos, J. (Ed.). (2002). A Companion to racial and ethnic
D. A. (1984). Processing the news: How people tame the information tide.
P., & Husband, C. (1974). Racism and the mass media.
R. N. (2000). Race, media and the crisis of civil society. From
S. (1992). BrandSätze. Rassismus im Alltag. ('Brandsätze' --Inflammatory
Sentences / Firebombs. Racism in everyday life). DISS-Studien.
S. (1998). Der Spuk ist nicht vorbei völkisch-nationalistische Ideologeme im
öffentlichen Diskurs der Gegenwart.
& Link, J. (1993). Die vierte Gewalt. Rassismus und die Medien. (The Fourth
Power. Racism and the Media).
P. M. (2003). The mass media and the dynamics of American racial attitudes.
G. (1988). Power and prejudice: The politics and diplomacy of racial
Bilbao, C. (1999). El grito
M. (2002). The great wells of democracy the meaning of race in American life.
Rojo, L., Gómez Esteban, C., Arranz Lozano, F., & Gabilondo Pujol, (1994).
Hablar y dejar hablar sobre racismo y xenofobia.
Ramos, F. (2004). Media & Migrants. A critical analysis of Spanish and
Irish discourses on immigration.
Reisigl, M., & Wodak, R. (Eds.). (2000). The semiotics of racism. Approaches in critical discourse analysis. Wien: Passagen.
M., & Wodak, R. (2001). Discourse and discrimination rhetorics of racism
Ruhrmann, G. (Ed.). (1995). Das Bild der Ausländer in der Öffentlichkeit. Eine theoretische und empirische Analyse zur Fremdenfeindlichkeit. (The image of foreigners in the public sphere. A theoretical and empirical analysis of xenophobia). Opladen: Leske
O., Sidanius, J., & Bobo, L. (Eds.). (2000). Racialized politics: The
debate about racism in
G., & Van Dijk, T. A. (Eds. ). (1987). Discourse and discrimination.
J., & Back, L. (1996). Racism and society.
Racismo (2000). Informe anual 2000 sobre el racismo en el Estado español.
Racismo (2001). Informe anual 2001 sobre el racismo en el Estado español.
Racismo (2002). Informe anual 2002 sobre el racismo en el Estado español.
Racismo (2003). Informe anual 2003 sobre el racismo en el Estado español.
J. (Ed.). (2002). Racism and cultural diversity in the mass media. An overview
of research and examples of good practice in the EU Member States, 1995-2000.
G. (1978). Making news: A study in the construction of reality.
Dijk, T. A. (1980). Macrostructures: An interdisciplinary study of global
structures in discourse, interaction, and cognition.
Dijk, T. A. (1984). Prejudice in discourse. An analysis of ethnic prejudice in
cognition and conversation.
Dijk, T. A. (1987a). Communicating racism: Ethnic prejudice in thought and
Dijk, T. A. (1987b). Schoolvoorbeelden van racisme. De reproduktie van racisme
in maatschappijleerboeken (Textbook examples of racism. The reproduction of
racism in social science textbooks).
Dijk, T. A. (1988). News as discourse.
Dijk, T. (1991). Racism and the Press.
Dijk, T. A. (1993). Elite discourse and racism.
T. A. (1997). Racismo y análisis crítico de los medios. (Racism and the
critical analysis of the media).
Dijk, T. A. (2003). Dominación étnica y racismo discursivo en España y América
Dijk, T. A., & Kintsch, W. (1983). Strategies of discourse comprehension.
M., & Potter, J. (1992). Mapping the language of racism: Discourse and the
legitimation of exploitation.
M. (Ed.). (1994). Racisme et xénophobie en
M. (1998). Le racisme: Une introduction.
D. (2005). Inventing black on black violence. Discourse, space and representation.
C. C., & Gutiérrez, F. (1985). Minorities and media: Diversity and the end
of mass communication.
& Van Dijk, T. A. (Eds.). (2000). Racism at the Top. Parliamentary
Discourses on Ethnic Issues in Six European States.
J., & Solomos, J. (Eds.). (1993). Racism and migration in