Abstract
Research on intergroup prejudice has generally adopted a model of social change that is based around the psychological rehabilitation of members of advantaged groups in order to foster intergroup harmony. Recent studies of prejudice-reduction interventions among members of disadvantaged groups, however, have complicated psychologists’ understanding of the consequences of inducing harmonious relations in historically unequal societies. Interventions encouraging disadvantaged-group members to like advantaged-group members may also prompt the disadvantaged to underestimate the injustice suffered by their group and to become less motivated to support action to challenge social inequality. Thus, psychologists’ tendency to equate intergroup harmony with “good relations” and conflict with “bad relations” is limited.
Research on reducing ethnic and racial prejudice has had three defining features. First, it has traditionally focused on the reactions of historically advantaged groups toward historically disadvantaged groups. Second, it has focused on antipathy in those reactions, working on the assumption that such antipathy underlies the problems of discrimination, inequality, and conflict that damage the lives of so many people in so many societies. Third, and by implication, it has adopted a model of social change that is based around the psychological rehabilitation of advantaged-group members in order to foster intergroup harmony.
Few would deny the importance of this model of social change. Research on racial prejudice, for example, has shown that the emotional biases of historically advantaged individuals may influence their decision making in contexts such as the workplace, the classroom, and the courtroom. It has thus inspired a variety of interventions designed to combat resulting forms of discrimination (see Dovidio, Glick, & Rudman, 2005). Recent research on the impact of prejudice reduction on the attitudes of members of historically disadvantaged groups, however, has complicated psychologists’ understanding of intergroup harmony. This research suggests that standard prejudice-reduction interventions often have different effects on advantaged- and disadvantaged-group members. Yet even when such interventions succeed in inducing disadvantaged-group members to like advantaged-group members, this transformation may often come at a cost: It may lead the disadvantaged to underestimate the injustice and discrimination suffered by their group, diminishing their support for action to challenge inequality.
The present article discusses these paradoxes, reviewing work that has contemporaneously investigated relations in South Africa, India, Israel, and the United States. The first section discusses evidence on the effects of prejudice-reduction techniques on the attitudes of disadvantaged groups, focusing on the role of intergroup contact—a classic technique of prejudice reduction. The second section considers some emerging explanations of the paradoxical consequences of harmony. Building on Wright and Lubensky’s (2008) framework, the article’s conclusion sketches some deeper tensions between two models of social change in social psychology: a prejudice-reduction model and a collective-action model. In light of these tensions, we argue, the widespread tendency for psychologists to equate harmony with “good relations” and conflict with “bad relations” is limited.
The Contact Hypothesis
Research on the contact hypothesis represents the most important tradition of social psychological work on prejudice reduction (see Pettigrew & Tropp, 2006). The hypothesis states that positive interaction (contact) between members of different groups improves intergroup relations, especially when such interaction occurs under favorable conditions (e.g., equality of status between groups) such as those outlined by Allport (1954). In large part, intergroup contact produces this effect because it decreases our negative reactions to others (e.g., anxiety) and increases our positive reactions to others (e.g., empathy; Pettigrew & Tropp, 2008). Although early research on the contact hypothesis was beset by mixed findings and skepticism about its practicality in real-world circumstances, in recent years a mood of optimism has swept the field. Hewstone, Cairns, Voci, Hamberger, and Niens (2006) capture this mood when noting that contact research “has contributed greatly to the fact that psychology is now in its best position ever to make a contribution to the advancement of world peace by actively promoting intergroup tolerance” (p. 100).
As is true of most research on prejudice, however, research on the contact hypothesis has concentrated mainly on the attitudes of historically advantaged groups. Moreover, it has emphasized the reduction of antipathy between groups as its hallmark of a “successful” outcome. The assumption here is that if the advantaged can be persuaded to hold more favorable attitudes about the disadvantaged, then the foundations for a more peaceful and just society may be laid.
We do not dispute that this perspective on social change has the potential to offer many benefits for society. Even so, our analysis of research on contact—and similar techniques of prejudice reduction—indicates that the consequences of promoting intergroup harmony may not be so simple (see also Dovidio, Gaertner & Saguy, 2009).
Intergroup Contact and Attitudes of the Disadvantaged
Our interest in this topic originated from parallel lines of research on intergroup contact among disadvantaged and advantaged groups. Meta-analytic data from hundreds of studies reveal that the prejudice-reducing effects of contact are typically weaker for members of disadvantaged racial groups than they are for members of advantaged racial groups (Tropp & Pettigrew, 2005). In part, these trends emerge because the disadvantaged tend be more pessimistic in their views of racial equality and hence more distrustful in their relations with the dominant racial group. U.S. national survey data show, however, that close contact (i.e., cross-group friendship) can yield positive attitude change for Black Americans and render perceptions of racial discrimination less important in predicting their racial attitudes toward Whites (Tropp, 2007).
Concurrently, our national survey of attitudes in post-apartheid South Africa attempted to identify the factors that predict support for, or opposition to, a range of policies being implemented by the South African government to overcome the country’s legacy of racial inequality. Among other factors, we explored the role of interracial contact in shaping respondents’ attitudes toward policies such as affirmative action, educational desegregation, and land restitution, and found an interesting disjunction between the results for White and Black respondents (see Dixon, Durrheim, & Tredoux, 2007). White South Africans who reported having positive contact with Blacks tended to be more supportive of government efforts to achieve change. However, Black South Africans who reported having positive contact with Whites tended to be less supportive of such efforts. A follow-up study (Dixon et al., in press) showed that positive contact was associated with more tolerant racial attitudes for both groups, at the same time as it identified a similar disjunction for Blacks’ and Whites’ perceptions of racial justice and discrimination (see Table 1 ). Together, these studies suggested to us that positive contact may increase disadvantaged-group members’ trust in members of advantaged groups while decreasing their perceptions of racial inequality and support for the implementation of social change.
Correlations Between Contact, Racial Tolerance, and Perceptions of Racial Injustice and Discrimination Among Black and White South Africans
Note. Correlations for White respondents are presented above the diagonal, and correlations for Black respondents are presented below the diagonal. Correlations are measures of agreement that can vary between -1.00 and +1.00; a value of 1 indicates complete agreement and a value of 0.00 indicates that scores are unrelated. The shaded cells highlight that the correlations between experiences of positive, cross-race contact and perceptions of racial injustice and discrimination were statistically significant for both Black and White respondents in this survey (adapted from Dixon et al., in press) but that they operated in opposite directions. For White respondents, more positive contact was associated with higher perceptions of racial injustice and discrimination; for Black respondents, more positive contact was associated with lower perceptions of racial injustice and discrimination.
Research conducted elsewhere suggests that this “paradox” is by no means unique to our South African studies. Tausch, Saguy, and Singh (2009) found that having Hindu friends improved Muslims’ attitudes toward Hindus but also diminished their awareness of group inequalities and, in so doing, diminished their intentions to engage in actions to improve the situation of Muslims in India. Saguy, Tausch, Dovidio, and Pratto (2009, Study 2) similarly found that cross-group friendship was positively associated with Israeli Arabs’ perceptions of Israeli Jews as “fair” and negatively associated with their perceptions of the material inequality suffered by their own group. Via its impacts on these variables, contact was again related to a decrease in respondents’ support for social change to improve the situation of Arabs in Israel. Finally, in research on African American and Latino students in the United States, Wright and Lubensky (2008) found that contact with Whites diminished support for collective action to achieve racial equality, even though, once more, it improved their racial attitudes toward Whites.
The results of these surveys are open to different interpretations, of course. We shall suggest that how they are interpreted depends on the model of social change one adopts. On the one hand, a decline in disadvantaged-group concerns about injustice and an improvement in their attitudes toward the advantaged group can be read as positive signs that intergroup relations are improving. On the other hand, in societies characterized by ongoing histories of ethnic and racial inequality—South Africa, India, Israel, and the United States all qualify—such psychological shifts can be read more negatively—that is, as encouraging disadvantaged group members to accept their disadvantage and acquiesce rather than challenge their group’s treatment.
Saguy et al. (2009, Study 1) provide an experimental analogue that supports the latter interpretation. They manipulated power relations by giving members of one of two randomly created groups control over the allocation of valued resources—in this case, university course credits. Members of both groups participated in one of two forms of intergroup contact: (a) a positive, “commonality” focused contact, or (b) a less positive, “group differences” focused contact. Members of the advantaged group were then asked to distribute course credits between the groups, while members of the disadvantaged group estimated the nature of the resulting distribution. The results showed that disadvantaged-group members in the commonality-focused-contact condition tended to overestimate the extent to which advantaged-group members would distribute credits equitably (see Fig. 1 ). Harmonious contact, in other words, engendered false expectations of fair treatment. Why?
The number of experimental credits expected by low-power group members and the actual number received from high-power members under conditions of differences-focused and commonality-focused contact. In the commonality-focused condition, members of the two groups discussed “things the groups had in common”; in the differences-focused condition, they discussed their different task roles and statuses in the experiment. Adapted from Saguy et al., 2009.
Explaining the Ironic Effects of Contact
Numerous theories of intergroup prejudice and group dominance might be invoked to account for these ironic effects of contact on disadvantaged groups (see Dovidio et al., 2005). In the present analysis, we wish to highlight three possible explanations.
Perceived personal and collective discrimination
An obvious explanation might highlight the congruence between an individual’s personal experiences of fair treatment and their estimates of the extent of discrimination at a collective level. For instance, if a Black person’s everyday interactions with Whites are generally cooperative, warm, and friendly, then this may decrease her sense of being a personal target of racial discrimination. Such reductions in perceived personal discrimination may, in turn, decrease her judgments of the severity of discrimination at a macro level (Dixon et al., in press). In other words, the effects of contact on perceived collective injustice may be mediated by its effects on perceived personal discrimination, with disadvantaged-group members generalizing on the basis of their own experiences when evaluating the broader structure of intergroup relations.
Changes in outgroup attitudes
Relatedly, the relationship between contact and perceptions of inequality may be mediated by changes in attitudes toward the advantaged, an idea supported by several of the studies reviewed above. The role of contact in altering our feelings toward members of other groups is well documented. For example, in over 96% of the studies reviewed by Pettigrew and Tropp (2006), contact was found to relate to a reduction in intergroup prejudice—a pattern that demonstrates the efficacy of the contact hypothesis as a technique of psychological change. Ironically, however, when members of disadvantaged groups are encouraged to like members of advantaged groups and to form more positive beliefs about them (e.g., as “fair” or “trustworthy”), the disadvantaged may also become less ready to view the advantaged as the beneficiaries, and perhaps the agents, of discrimination.
Common group identification
A darker side of fostering positive feelings and cognitions among members of historically disadvantaged groups is perhaps illustrated most starkly by a third explanation. This explanation focuses on the role of intergroup contact in promoting a sense of common identity—that is, an identity based around an inclusive sense of “we.” The potential psychological benefits of this form of recategorization are well-established (Gaertner & Dovidio, 2000), as are its shortcomings in relation to disadvantaged-group concerns (see Hornsey & Hogg, 2000). When members of two groups come to see themselves as part of a common ingroup, intergroup prejudice tends to decrease and their relations become more cooperative, forgiving, and helpful. Under certain circumstances, however, this blurring of intergroup boundaries could reduce disadvantaged-group members’ tendency to make intergroup comparisons around issues of status and power. As such, common group identification may deflect attention away from the inequalities that characterize hierarchical societies and increase expectations of “equal treatment” (see Dovidio, Gaertner, et al., 2009). Thus, as shown in Saguy et al.’s (2009, Study 1) experiment, disadvantaged-group members who have contact emphasizing commonality with the advantaged group may be primed to form unduly optimistic conceptions of distributive justice (see Fig. 1).
Competing Pathways to Social Change?
McConahay (1978) once cautioned that “amicable relations among racial and ethnic groups can exist alongside grossly unjust inequalities of opportunities and outcomes. Ceteris paribus, harmonious race relations and unprejudiced attitudes might be worthy goals—but only if other things are equal, or nearly so” (p. 77). His point was that harmonious relations do not necessarily translate into material justice. Taking this argument a step further, the research outlined in this article suggests that such relations are sometimes implicated in the reproduction of injustice. Not only may harmony exist alongside social inequality, but it may also inculcate psychological responses that make it more difficult for the disadvantaged to perceive and challenge such inequality.
Wright and Lubensky (2008) have discussed this conundrum in relation to some deeper tensions between two models of change in social psychology. The prejudice-reduction model has focused on transforming the hostile emotions and stereotypic beliefs of members of historically advantaged groups in order to reduce ethnic and racial discrimination. The collective-action model has focused on transforming the political orientations of members of historically disadvantaged groups in order to motivate them to challenge the status quo.
It would be comforting to assume that the models depicted in Table 2 simply offer different but complementary routes toward a better society. The problem, as Wright and Lubensky (2008) highlight, is that those models entail psychological processes that are potentially antithetical. Prejudice-reduction interventions are designed to weaken our tendency to view the world in “us versus them” terms, to nurture positive emotional responses such as empathy and trust, and to diminish the potential for intergroup conflict. Collective-action interventions, by contrast, are designed to encourage members of disadvantaged groups to form a strong sense of group identity, to recognize injustice, to become angry about it, and to take collective action with the goal of challenging the status quo. The latter course of action almost inevitably involves direct confrontation between representatives of disadvantaged and advantaged groups, for the beneficiaries of social hierarchies rarely concede their privileges without a struggle.
Agents, Interventions, Psychological Processes, and Behavioral Outcomes of Two Models of Social Change in Historically Unequal Societies
Conclusions
The challenge of reconciling, integrating, or prioritizing these two models of social change now preoccupies the research teams whose work has been outlined in this article. Divergent points of view are emerging, and indeed ours accommodates a variety of viewpoints about the best way forward. Some of us feel that encouraging people to like one another has eclipsed more important issues concerning the social psychology of (re)distributive justice in inequitable societies. Others feel that the different models of social change represented in Table 2 may be reconciled in practice by considering the historical trajectory of intergroup relations within such societies. For example, conflict may initially serve as a necessary motor of (re)distributive justice in societies characterized by gross material inequalities; however, the reduction of prejudice may subsequently prove vital to longer-term stability and acceptance of diversity. And yet others continue to be optimistic that greater contact and dialogue across group boundaries can lead advantaged-group members to become both more aware of social inequalities and more motivated to combat them, such that they work together with disadvantaged-group members to promote social change.
One point on which all of the participants in this conversation agree, however, is that psychologists’ tendency to treat intergroup conflict as the problem and intergroup harmony as the solution is at best simplistic—an idea that resonates with wider developments in social psychology (e.g. see Nadler, 2002; Dovidio, Saguy & Shnabel, 2009). As the “fire under the boiler of social change” (Cooley, 1918), short-term conflict sometimes lays the foundations for longer-term justice in historically unequal societies. What is more, harmony is not always what it seems. It may entail either unequal relations in which each group “knows their place” or equal relations between groups in which social change and justice have truly been achieved.
Recommended Reading
Dovidio, J.F., Gaertner, S.L., & Saguy, T. (2009). (See References). A comprehensive overview of the common-identity model of prejudice reduction, including discussion of that model’s limitations.
Glick, P., & Fiske, S.T. (2001). An ambivalent alliance: Hostile and benevolent sexism as complementary justifications for gender inequality. American Psychologist, 56, 109–118. An article that demonstrates the complexity of discriminatory gender attitudes.
Jackman, M.R. (1994). The velvet glove: Paternalism and conflict in gender, class, and race relations. Berkeley: University of California Press. A seminal analysis of the limits of conflict-based theoretical models of intergroup relations in hierarchical societies.
Van Zommeren, M., Postmes, T., & Spears, R. (2008). Towards an integrative social identity model of collective action: A quantitative research synthesis of three socio-psychological perspectives. Psychological Bulletin, 134, 504–535. A synthesis of work on the social psychology of collective action.
Wright, S.C., & Lubensky, M. (2008). (See References). An excellent conceptual analysis of the tensions between prejudice-reduction and collective-action models of social change.
Footnotes
Some of the research reported in this article was funded by grants awarded to the first author by the Economic and Social Research Council in the United Kingdom (RES-000-22-1750) and to the third author by the National Research Foundation in South Africa.