Gehart Saenger’s book, The Social Psychology of Prejudice, was one of the first systematic attempts by a social psychologist to consolidate the early literature on prejudice. Its publication in 1953 was heralded by no less a figure than Gordon Allport as setting ‘forth truths that if applied would certainly diminish the ravages of bigotry in our society’. The book opens on a parable. An anonymous narrator deliberates upon the potential admission of a Jewish man, Sidney Levy, to an exclusive social club. Mr Levy seems to be ‘a nice person’, the narrator notes, and ‘Personally I would not mind if he joined the club’. However, he is ‘definitely Jewish’, and ‘you never know with them. Sooner or later their true nature will show through. Moreover, once we take him, he may invite his Jewish friends and before long the whole club will be overrun by these kikes’ (Saenger, 1953, p. 3).
Published the following year, the introduction to Allport’s own canonical text, The Nature of Prejudice (1954, pp. 13–14), featured a similarly vivid example. In this case, the reader is invited to overhear an imaginary conversation between a Mr X and a Mr Y, who are debating ‘the trouble with Jews’.
In both books, then, stark parables of anti-Semitism introduced a concept that was to dominate the social psychology of intergroup relations in ensuing decades. In the readiness of Saenger’s and Allport’s protagonists to make hostile generalizations about members of another group and in the detachment of their attitudes from the facts of social reality, we find personified the elementary features of the concept of prejudice. Saenger (1953, p. 3) went on to define prejudice formally as a process whereby we ‘judge a specific person on the basis of preconceived notions, without bothering to verify our beliefs or examine the merits of our judgements’. Allport’s definition was more succinct. Prejudice, he famously observed, is ‘an antipathy based upon a faulty and inflexible generalization’ (1954, p. 9): it involves ‘thinking ill of others without sufficient warrant’ (1954, p. 6).
Mr X: The trouble with Jews is that they only take care of their own group.
Mr Y: But the record of the Community Chest campaign shows that they give more generously, in proportion to their numbers, to the general charities of the community than do non-Jews.
Mr X: That shows they are always trying to buy favour and intrude into Christian affairs. They think of nothing but money; that is why there are so many Jewish bankers.
Mr Y: But a recent study shows that the percentage of Jews in the banking business is negligible, far smaller than the percentage of non-Jews.
Mr X: That’s just it; they don’t go for respectable businesses; they are only in the movie business or run nightclubs.
More than fifty years later, it is not an exaggeration to claim that the concept of prejudice is fundamental to the social psychology of intergroup relations. In fact, a strong case could be made that it is the defining concept of the field. In the period following the end of the Second World War, impelled by the work of scholars such as Allport (1954) and Adorno et al. (1950), as well as that of earlier researchers such as Bogardus (1925), Katz and Braly (1933) and Dollard et al. (1939), the project of understanding prejudice became viewed by many social psychologists as central to solving a host of social problems, including problems of racial discrimination, ideological extremism and genocide. The concept subsequently came to underpin an astonishing profusion of theoretical, empirical and applied work in the discipline. Nowadays the term ‘prejudice’ is ubiquitous in our journals, monographs and conference proceedings, and rarely is it far from the lips of those of us who teach courses on intergroup relations. As Reynolds, Haslam and Turner (Chapter 2, this volume) observe, the concept of prejudice has consumed ‘many minds and research budgets’.
Given its historical significance in social psychology, we believe that a broad reappraisal of what the concept of prejudice can and cannot tell us is both timely and important. This edited collection of essays provides such a reappraisal. The book’s overall aim is not primarily to review specific traditions of research on prejudice (e.g. Brown, 1995) or to explore how competing theories might be integrated (e.g. Duckitt, 1992). Rather, we wish to trace the deeper implications of what Margaret Wetherell (Chapter 8) has called the ‘problematic’ of prejudice for how social psychologists have framed the entire problem of investigating, understanding and changing intergroup relations. What has this problematic contributed to our knowledge of relations between groups and what has it obscured or traduced? How has it expanded or narrowed the horizons of psychological inquiry? How effective or ineffective has it been in guiding our attempts to transform social relations and institutions? We also wish to discuss some emerging perspectives that have attempted, in various and sometimes contradictory ways, to move beyond the Allportian framework of prejudice research that has loomed so large over the social psychology of intergroup relations for over fifty years.
This introduction sets the scene. The first section describes the emergence of prejudice research in social psychology and situates its contribution in historical context. The second section outlines some foundational assumptions of the prejudice problematic, notably its individualistic orientation and its assumptions about the role played by cognitive irrationality and affective negativity within intergroup relations. We argue that these assumptions inform the model of social change espoused by prejudice researchers, which is based around the cognitive and emotional rehabilitation of majority group bigots. The third section presents a chapter by chapter outline.
In order to appreciate the historical significance of the prejudice problematic, we must consider the scientific paradigms it challenged and ultimately supplanted in social psychology. During the eighteenth, nineteenth and early twentieth centuries, scientific thinking about racial relations was informed by beliefs about racial difference and hierarchy (e.g. see Haller, 1971; Goldberg, 1993). Often employing conceptual frameworks based on the concept of biological inheritance, such thinking portrayed some groups as innately inferior and backward, rooting the causes of racial hostility in the supposed characteristics of its targets.
When the fledgling discipline of social psychology was consolidated in the early years of the twentieth century, it was mired in this way of framing the ‘problem’ of race relations (see Richards, 1997; Samuelson, 1978). During this period – aptly labelled ‘the nadir’ by Pettigrew (2008) – psychologists were preoccupied with the study of racial differences, particularly differences in intellectual abilities. Journal articles published at the time indicate that empirical research did not focus exclusively on comparisons between blacks and whites. As prominent were investigations of the so-called Nordic thesis, which proposed that Northern Europeans are genetically and culturally advantaged relative to other groups (including ‘Mediterranean’ and ‘Alpine’ Europeans).
Both implicitly and explicitly, this research refracted wider political arguments about the nature, course and governance of race relations, and formed part of the ideological project that is now known as scientific racism (see Pehrson and Leach, Chapter 6, this volume, for further discussion). On the one hand, it quietly perpetuated the traditional doctrine of the ‘well-deserved reputation’ (Zadwadzki, 1948), treating racial conflict as an inevitable outcome when a biologically superior group encounters the deficiencies of less developed groups. On the other hand, some social psychologists attempted to use psychological research as a platform from which to influence social policy. William McDougall, for example, openly supported eugenic policies of immigration control as a means of limiting contact between certain groups, proclaiming that ‘some blends of human sub-races are eugenically admirable and others disastrous’ (McDougall, 1918, cited in Richards, 1997, p. 197). It is important to acknowledge, too, that many commentators were sceptical about the concept of innate racial differences and questioned both its underlying assumptions and its evidence base. Some also recognized its political consequences – one of which was to confer legitimacy on institutions such as colonialism, slavery and segregation – and thus sought to ‘puncture the biological myths’ of race (Fairchild and Gurin, 1978).
According to Samelson (1978), between the 1920s and 1940s an ‘abrupt reversal’ occurred within psychological research on ethnic and racial relations. Emphasis shifted away from the project of measuring, explaining and debating the nature of group differences, and psychologists became increasingly concerned with the problem of intergroup prejudice. Thus, by 1950, Allport could assert that research on prejudice had ‘spread like a flood both in social psychology and in adjacent social sciences. Publications are cascading from the presses. The outpouring within the past decade surely exceeds the output in all previous human history’ (p. 4). Samelson (1978) astutely cautions against interpreting this shift as a simple victory of empirical science over politics. To the contrary, the rise of the problematic of prejudice was itself the complex outcome of a range of political processes unfolding both globally and within the US. For one thing, fascism acquired a decidedly bad name in the post-war era, giving global impetus to a search for the causes of irrational hatred towards minorities. In addition, the passing of the Immigration Restriction Law of 1924 in the US reduced political momentum for finding scientific justifications for excluding ‘undesirable’ immigrant groups, while the problem of maintaining peaceful coexistence among different groups of Americans grew on the national agenda. Moreover, psychology itself became somewhat less ‘lily white’ after the 1920s (Samelson, 1978, p. 271) and the discipline’s increased social diversity probably heightened psychologists’ awareness of problems of racial discrimination.
Whatever its causes, the shift in focus from race differences to race prejudice had a profound impact on social psychology. Perhaps most important, it reversed the social target of psychological research. No longer were the causes of social disharmony attributed mainly to the psychological and cultural deficiencies of minorities; instead, they were attributed to the racial prejudices of majority group members. As Montagu (1949) emphasized, ‘It is the discriminators, not the discriminated, the prejudiced, not those against whom prejudice is exhibited, who are the problem’ (p. 176).
Early work quickly established the scale and severity of this problem of dominant group prejudice. For example, in their classic study – later replicated by Rosenblith (1949) – Allport and Kramer (1946) reported that racial prejudice and anti-Semitism were widespread among white Americans and associated with factors such as childhood experiences, acceptance of parental authority and segregation. Their study also painted an unflattering early picture of the prejudiced person, highlighting their lack of insight, conventionality and suspiciousness, as well as the ‘dull’, ‘unaware’ and ‘stencilled’ quality of their thinking (p. 35). At the same time, Allport and Kramer (1946, p. 9) proclaimed the existence of ‘almost unanimous agreement’ among their contemporaries that prejudice should be regarded not as an inborn, fixed state, but as amenable to reduction. They predicted that research on prejudice would ultimately yield interventions to promote social change and proposed intergroup contact as one such intervention, anticipating a rich tradition of later research (see Pettigrew and Tropp, 2006; Brown and Hewstone, 2005). Although the relationship between theories of prejudice and processes of social change would turn out to be complex and, at times, fraught, early work successfully established the field’s applied emphasis as another defining feature. Relaxing their customary posture of scientific detachment, many prejudice researchers became passionate advocates of a more tolerant society and sought to understand how scientific knowledge might promote social transformation. The challenge was clear. ‘Prejudice,’ as Adorno et al. (1950) wrote in the preface to The Authoritarian Personality, ‘is one of the problems of our times for which everyone has a theory but no one has an answer.’
Overviews of the subsequent development of prejudice research have rightly highlighted the field’s diversity (e.g. Brown, 1995; Dovidio, 2001; Dovidio et al., 2005; Duckitt, 1992; Nelson, 2009). In his useful periodization, Dovidio (2001) has identified three distinct phases in its evolution. Early work treated prejudice primarily as the product of abnormal personality development; in an intermediate phase, prejudice was viewed mainly as the outcome of ordinary, if imperfect, forms of information processing; and nowadays researchers increasingly emphasize unconscious, automatic and ‘aversive’ prejudices. Each of these phases has been marked not only by the development of new theoretical perspectives, but also by the emergence of new methodological techniques for measuring intergroup attitudes and stereotypes. The concept of prejudice has been decomposed into an array of forms, captured by emerging distinctions between old-fashioned and symbolic prejudice (Kinder and Sears, 1981), implicit and explicit prejudice (Greenwald and Banaji, 1995), subtle and blatant prejudice (Pettigrew and Meertens, 1995), and Jim Crow and laissez faire prejudice (Bobo et al., 1997), among others.
While respecting its historical diversity, the contributors to this book hold that prejudice research is also unified by a number of deep-seated assumptions about the nature of the social psychological processes that underlie intergroup relations. These assumptions have become institutionalized within the conceptual frameworks and research practices of social psychologists, lending coherence to an otherwise disparate array of perspectives on the nature of intergroup relations. In this section, we outline some foundations of the prejudice problematic and trace their implications for how psychologists have framed questions of social change. As the book unfolds, several of its other core features will become apparent.
With some notable exceptions (e.g. Sherif, 1967), prejudice researchers have given causal priority to processes of cognition, emotion and personality lying within the individual, while acknowledging – with varying degrees of enthusiasm and theoretical sophistication – that such processes are also shaped by the contexts in which they unfold. The individualism of psychological research is evidenced most clearly in work focused on individual differences in prejudice, which stretches from early research on personality characteristics such as authoritarianism and dogmatism through to more recent work on topics such as social dominance and aversive racism (e.g. Pratto et al., 1994). The overarching goal of such work has been to explain how and why people vary in their propensity to express prejudice towards others, based on the assumption that ‘the cognitive processes of prejudiced people are in general different from the cognitive processes of tolerant people’ (Allport, 1954, p. 170). Other work has proceeded from the opposite assumption, viewing prejudice as the result of psychological processes that are universal and part of ordinary cognition (see Fiske, 2005). Again, however, the individual has served as the primary locus of causality. Prejudice has been conceived as a process that arises – unaided and sometimes contrary to our conscious intentions – from the inner workings of our minds, even if its precise form and expression is modulated by environmental factors.
The theoretical individualism of psychological research on prejudice is complemented by its methodological individualism. Indeed, as Reicher (2007, p. 825) observes, the design of psychological research often fabricates ‘a monadic world of isolates and a world of silence, which we might try to create within our experimental studies but which exists in few places outside’. In such a world, the isolated individual is the main methodological and analytic unit of analysis, a fact that has profound, if often unacknowledged, consequences for the kinds of ‘data’ prejudice researchers collect and the forms of knowledge they generate. Durrheim (Chapter 9) and Condor and Figgou (Chapter 10) explore this point in some depth.
Of course, the argument that prejudice research prioritizes the role of individual psychology has a long pedigree and complex historical roots that need not be elaborated here (e.g. see Gordon, 2010). Suffice to say that the theme of individualism has been the subject of considerable discussion within social psychology (and without); and the problem of understanding the so-called ‘interaction’ between individual and contextual determinants of prejudice remains a live debate. Historically, sociologists and other social scientists have lined up to criticize psychologists for individualizing the historical, structural and political roots of intergroup conflict (e.g. Blumer, 1958; Rose, 1956). Psychologists, in their turn, have defended the value of research focused on the ‘intrapsychic’ level of analysis. We suspect that most social psychologists (ourselves included) would nowadays accept the ecumenical view expressed by Allport (1962, p. 134) in his paper titled: ‘Prejudice: is it societal or personal?’, who argued that ‘There are no good reasons for professional rivalry and backbiting among social scientists preferring one approach or the other. They can and should be blended in our outlook.’
We also believe, however, that the history of psychological work on prejudice shows that such general statements hide as much as they reveal. What do we mean by ‘societal’ and ‘personal’ factors? How can or should they be ‘blended’? What kinds of methodological and analytic frameworks might allow us to clarify best their complex interrelations? Although acknowledging that there are a variety of positions within this debate, and that valuable attempts to forge integrative models have been made (e.g. Duckitt, 1992), several contributors to the present volume argue that the conceptual frameworks and methodological practices that inform psychological research on prejudice continue to perpetuate an individualistic perspective on intergroup relations (e.g. see Chapters 1, 3, 8, 9 and 10). One reason for this, we would argue, is psychologists’ abiding emphasis on the role of individual irrationality and error as the main source of prejudice.
The modern roots of the term ‘prejudice’ lie in the Enlightenment liberalism of the eighteenth century, which distinguished opinions based on religious authority and tradition from opinions based on reason and scientific rationality. As Billig notes in Chapter 7, the Enlightenment injunction to live one’s life in the ‘light of reason’ found its antithesis in the concept of prejudice, which became synonymous with unreasoning faith. As its usage became established in the social sciences during the early years of the twentieth century, the term acquired a more restrictive set of meanings. Prejudice came to designate negative opinions about members of certain categories of person, particularly the categories of ethnicity, race and nation. The semantic links to irrationality remained foundational, however (Newman, 1979). Whereas Enlightenment philosophers had berated the blind faith of the clerics, early psychologists berated the blind hatred of Jews, blacks, immigrants and other historically disadvantaged groups.
Subsequent generations of prejudice researchers have faced the challenge of designing research that clarifies how, when and why such irrationality infects our reactions to other people. This challenge has inspired numerous, often highly innovative, methodological paradigms, which demonstrate that prejudice produces judgements whose outcomes seem biased, distorted, misdirected, error prone and sometimes plain wrong. Some of the most famous studies in the history of prejudice research fall into this category (e.g. Allport and Postman, 1946; Duncan, 1976; Hamilton and Gifford, 1976). On a more general level, researchers have employed methods that demonstrate how the prejudiced mind departs from ideal models of rational thought processes. Individuals who score high in prejudice are more likely, for example, to persevere with inefficient problem-solving strategies (Rokeach, 1948), to produce simplistic memories of physical objects (Fisher, 1951) and to display a variety of other forms of cognitive rigidity, including inflexibility, concretization, overgeneralization and intolerance of ambiguity (Kutner and Gordon, 1964). In the ‘third wave’ of prejudice research (Dovidio, 2001), the irrationality of the prejudiced mind has been further exposed by new techniques for measuring implicit associations (see Durrheim, Chapter 9 of this volume, for a detailed review). A disturbing implication of this research is that prejudice operates not only beneath the threshold of conscious awareness, but also in ways that may run contrary to what individuals experience as their conscious rationality (e.g. see Devine, 1989; Dovidio and Gaertner, 2004). If earlier studies suggested that the prejudiced are poor judges of the extent to which others share their social attitudes (e.g. Koenig and King, 1964), this emerging work suggests that they are also often poor judges of their own implicit attitudes.
At a theoretical level too, work on prejudice has been dominated by assumptions about its irrational nature. Early work focused on the general psychodynamic mechanisms through which prejudiced individuals may project unacceptable aspects of themselves onto others or displace environmentally induced aggression onto ‘scapegoat’ groups (e.g. Dollard et al., 1939). Subsequent waves of research on the ‘prejudiced personality’ conceived prejudice primarily as the result of aberrant personality development, which creates a maladjusted view of social reality. In research on the authoritarian personality, for instance, the irrationality of prejudice was viewed as distorting not only individuals’ relationships with others, but also their relationships to broader social and political structures, rendering them susceptible to ideological extremism (Adorno et al., 1950). Anticipated in social psychology by the work of Allport (1954; see also Chapter 2) and Tajfel (1969), the cognitive revolution entrenched the notion that negative reactions towards others are grounded in misperceptions of social reality (Fiske and Taylor, 1984). Broader streams of theoretical work on the fallibility of human judgement (e.g. Tversky and Kahneman, 1973) entered the discipline primarily via research on stereotyping and also via research on related phenomena such as illusory correlations, errors of attribution, confirmation biases and false consensus effects. In contrast with earlier theories, such work emphasized the all too human nature of prejudiced cognitions, which were treated as a regrettable byproduct of otherwise adaptive mechanisms for processing information (see Fiske, 2005; McCrae and Bodenhausen, 2000 for reviews of work in this tradition). This remains overwhelmingly the dominant perspective on the relationship between cognition and prejudice. It is a perspective, however, that is critically interrogated by several contributors to the present volume (e.g. see Chapters 1, 2, 7, 8, 9, 10, 14 and the Conclusion).
Yet prejudice has seldom been treated purely as a matter of cold cognition. Allport (1954, p. 22) argued that our emotional responses to others sometimes take precedence over cognitive judgements. Emotions, he noted, operate ‘like sponges. Ideas, engulfed by an overpowering emotion, are more likely to conform to the emotion than to objective evidence.’ He argued further that emotions help to explain why prejudice is sometimes impervious to rational counter-arguments and disconfirming evidence, memorably noting that even when it is ‘defeated intellectually’ prejudice often ‘lingers emotionally’ (p. 328).
Theories of the nature of such emotional reactions have varied extensively. In instinctive accounts of human aggression – delightfully labelled ‘blood and guts’ theories by Tajfel (1969) – the human propensity for intergroup fear and hatred has long been construed as part of our biological and evolutionary inheritance, an idea that is witnessing a revival (e.g. see Schaller et al., 2003; Neuberg and Cottrell, 2006). Classic motivational theories (e.g. the frustration–aggression hypothesis) have used psychoanalytic concepts such as ambivalence, catharsis and displacement to explain the intensity of aggressive feelings that minorities receive under certain social conditions. Such work has cashed out Freud’s (1930/1975, p. 51) bleak dictum that: ‘It is always possible . . . to bind together a considerable number of people in love, so long as there are other people left over to receive the manifestations of their aggressiveness.’ Until fairly recently, social cognition research somewhat neglected the affective dimension of prejudice. Certainly, it disregarded the more extreme manifestations of intergroup emotions that engaged social psychologists in the post-war period. However, even in this tradition, the affective dimensions of prejudice have always been tacit (e.g. within motivational concepts such as ‘ingroup bias’), and they have become increasingly overt, with work on stereotypic beliefs about others being integrated with work on feelings towards them (Bodenhausen et al., 2001; Mackie and Hamilton, 1993).
Yet what exactly is the nature of the emotional responses that define prejudice? Some early researchers argued that they should be treated as differentiated rather than unitary in character. Notably, Kramer (1949) advocated disaggregating prejudiced emotions into sub-categories such as fear, disgust, contempt, envy and anger. He insisted that both the cognitive and emotional components of prejudice vary markedly depending on the nature of outgroup being appraised and are associated with quite different predispositions to act (e.g. to engage in inclusion/exclusion, withdrawal/non-withdrawal). His essay prefigured the kind of work that has enriched the study of intergroup emotions in recent years (e.g. see Mackie and Smith, 2002; Mackie et al., 2008). For much of the history of prejudice research, however, the emotional specificity and complexity of prejudice has not been a central concern. To the contrary, prejudice has been defined as a kind of generic affective response towards members of other groups, which varies in intensity from low to high and is in practice defined by its negative valence (even if textbook discussions dutifully remind readers that prejudice can, in principle, involve warm as well as hostile feelings). Theories of its antecedents have varied widely, sometimes achieving considerable sophistication, but prejudiced emotion itself has been conceived in remarkably simple terms in most research. In a nutshell, prejudice occurs when ‘we’ dislike ‘them’ and don’t have a sensible reason for doing so. As we shall see, however, several chapters in the present volume complicate this seemingly self-evident claim, revealing how prejudice typically involves a complex blend of positive and negative feelings (see especially Chapters 1, 3, 4, 11, 12 and the Conclusion). They also indicate that the relationship between getting people to like one another and changing intergroup relations may be more complex than the prejudice problematic suggests.